Jakobiec 039 S Principles Amp Practice of Ophthalmology Volume One

Albert: Albert & Jakobiec's Principles &
Practice of Ophthalmology
THIRD EDITION
Daniel M. Albert, MD MS
Chair Emeritus, F. A. Davis Professor and Lorenz F. Zimmerman Professor, Department of
Ophthalmology and Visual Sciences, Retina Research Foundation Emmett A. Humble Distinguished
Director, of the Alice R. McPherson, MD, Eye Research Institute, University of Wisconsin
Medical School, Madison, Wisconsin, USA
Joan W. Miller, MD
Henry Willard Williams Professor of Ophthalmology, Chief and Chair, Department of
Ophthalmology, Massachusetts Eye and Ear Infirmary, Harvard Medical School, Boston,
Massachusetts, USA
Associate Editors:
Dimitri T. Azar, MD
B.A. Field Chair of Ophthalmologic Research, Professor and Head, Department of Ophthalmology
and Visual Sciences, University of Illinois Eye and Ear Infirmary, Chicago, Illinois, USA
Barbara A. Blodi, MD
Associate Professor, Department of Ophthalmology and Visual Sciences, University of Wisconsin
Medical School, Madison, Wisconsin, USA
Managing Editors:
Janet E. Cohan
Administrative Manager, Department of Ophthalmology, Massachusetts Eye and Ear Infirmary,
Harvard Medical School, Boston, Massachusetts, USA
Tracy Perkins, MPH
Administrative Director, Alice R. McPherson, MD Eye Research Institute, University of
Wisconsin Medical School, Madison, Wisconsin, US
DEDICATION
To CLAES H. DOHLMAN
Superb surgeon, mentor, teacher, innovator and friend.
D.M.A & J.W.M
SAUNDERS ELSEVIER
SAUNDERS is an imprint of Elsevier Inc.
? 2000, 1994 by W.B Saunders Company
? 2008, Elsevier Inc. All rights reserved.
First published 2008
First edition 1994
Second edition 2000
Third edition 2008
No part of this publication may be reproduced, stored in a
retrieval system, or transmitted in any form or by any means,
electronic, mechanical, photocopying, recording or otherwise,
without the prior permission of the Publishers. Permissions may
be sought directly from Elsevier's Health Sciences Rights
Department, 1600 John F. Kennedy Boulevard, Suite 1800,
Philadelphia, PA 19103-2899, USA: phone: (+1) 215 239 3804;
fax: (+1) 215 239 3805; or, e-mail:
healthpermissions@elsevier.com. You may also complete your
request on-line via the Elsevier homepage (http://
www.elsevier.com), by selecting Support and contact and then
Copyright and Permission.
ISBN: 978-1-4160-0016-7
Notice
Medical knowledge is constantly changing. Standard safety
precautions must be followed, but as new research and clinical
experience broaden our knowledge, changes in treatment and drug
therapy may become necessary or appropriate. Readers are
advised to check the most current product information provided
by the manufacturer of each drug to be administered to verify
the recommended dose, the method and duration of
administration, and contraindications. It is the responsibility
of the practitioner, relying on experience and knowledge of the
patient, to determine dosages and the best treatment for each
individual patient. Neither the Publisher nor the author assume
any liability for any injury and/or damage to persons or
property arising from this publication.
The Publisher
Preface to the 3rd Edition

Do clinicians and trainees really need textbooks anymore? In an
era of ever-expanding connectivity and immediate access to
published articles, why would anyone consult a textbook, which
by its very nature is incomplete before it is even published? No
doubt these are strange questions coming from the editors of
the third edition of the most popular multi-volume ophthalmic
textbook, but they must be asked and answered. Our answer
is an unequivocal yes! Books like this serve an extremely
important function that of a repository for expert reviews of
our current understanding of ophthalmic health and disease.
The chapters and sections in Albert and Jakobiec are an
important resource for the clinician and student, providing a
comprehensive information base on an extensive list of topics.
Of course journal articles continue to be the most useful source
of information about new developments in the field but they do
not replace books. Constraints on the length of journal articles,
inattention to the provenance of the ideas they contain, and an
understandable tendency to self-promote the authors thesis,
limit the value of many original contributions. Readers
of journal articles forearmed with information found in an
encyclopedic text can place these articles into perspective.
Thus, the two sources are complimentary. In a very real sense
this textbook serves as a springboard to the constantly
expanding universe of published scientific literature.
What is new in the third edition? The second edition (2002)
was a reworking of the very successful first edition (1996) of
Albert and Jakobiecs Principles and Practice of Ophthalmology.
For the third edition we undertook a critical evaluation of each
section and chapter to ensure that topics were well-covered with
minimal redundancy, that new areas of practice and research
were adequately described, and that topics that were overrepresented could be substantially shortened or deleted. This
evaluation involved all of the editors (Dan Albert, Joan Miller,
Barbara Blodi and Dimitri Azar) as well as new and returning
section editors. As an example, under the direction of Dimitri
Azar, we incorporated a new section on refractive surgery that
provides the principles of refractive surgery as well as useful
descriptions of evaluation techniques and procedures. The
Oncology section was substantially expanded and revised under
the section editorship of Evangelos Gragoudas and Joan
OBrien. Pediatrics was also extensively revised by David
Hunter and Monte Mills, and the Pharmacology and Toxicology
sections were combined and revised under the direction of Mark
Abelson. Barbara Blodi and Joan Miller reworked the extensive
retina section, to include current techniques, new diagnostic
modalities (including OCT), and new drug therapies. The
human genome project and modern genetics are revolutionizing
medicine, and genetics information has been incorporated into

all sections. Finally, the last section of the textbook headed by
Kathy Colby and Nancy Holekamp is a section on Ethics and
Professionalism topics that are increasingly important to
practicing clinicians, and an ACGME requirement for resident
training. A concerted effort was made throughout the third
edition to complement the text with diagrams, line drawings
and color figures. In addition, each chapter contains a key
points section. Overall, the third edition has exceeded the
expectations of all of the editors. We were pleased by the
enthusiasm of new and returning authors, more than 600 in
total, as well as new and returning section editors, and were
excited by the teamwork and cooperation shown in upgrading
and improving this important project. The result is a definitive
textbook in ophthalmology, available in hardcover and by web
access.
The editorial team has been a wonderful collaboration and
the senior editors are very grateful for the prodigious efforts of
Drs. Dimitri Azar and Barbara Blodi. We were saddened that Dr.
Frederick Jakobiec, a co-founder of this project and co-editor on
editions 1 and 2, was unable to participate as an editor in the
third edition, although still contributing as a co-author. We look
forward to his return to the ophthalmology community, and we
can report that Dr. Jakobiec is pleased and supportive of the
upcoming 3rd edition of the textbook named for him and Dr.
Albert. All of the editorial team is most appreciative of the
unstinting and generous support of Elsevier Publishing; in
particular the leadership of the senior editor, Russell Gabbedy,
and the hard work and diligence of Zak Knowles, contributing
editor, whose efforts in collecting and coordinating chapters, as
well as initial editing of chapters were unsurpassed. The
managing editors, Tracy Perkins and Janet Cohan, provided
important coordination between the authors, section editors,
editors and publisher, and handled all of their responsibilities
with aplomb. Above all, the contributing authors who wrote the
chapters and the section editors who delineated the section
content and edited the component chapters deserve the greatest
credit for the superb quality of the textbook.
We sincerely hope that the third edition of Albert and
Jakobiecs Principles and Practice in Ophthalmology provides
ophthalmologists and trainees with a gateway into the
wonderful science and art of ophthalmology in order to provide
the best care for our patients, and to continually advance
our field.
Daniel M. Albert and Joan W. Miller
xvii
Preface to the 1st Edition

INCIPIT. The medieval scribe would write this Latin word,
meaning so it begins, to signal the start of the book he was
transcribing. It was a dramatic word that conveyed promise of
instruction and delight. In more modern times INCIPIT has
been replaced by the PREFACE. It may be the rst thing the
reader sees, but it is, in fact, the last thing the author writes
before the book goes to press. I appreciate the opportunity to
make some personal comments regarding Principles and
Practice of Ophthalmology.
One of the most exciting things about writing and editing a
book in a learned eld is that it puts the authors and editors in
touch with those who have gone before. Each author shares
with those who have labored in past years and in past centuries
the tasks of assessing the knowledge that exists in his or her
eld, of determining what is important, and of trying to convey
it to his or her peers. In the course of the work the author
experiences the same anticipation, angst, and ennui of those
who have gone before. He or she can well envision the various
moments of triumph and despair that all authors and editors
must feel as they organize, review, and revise the accumulating
manuscripts and reassure, cajole, and make demands of their
fellow editors, authors, and publisher.
This feeling of solidarity with early writers becomes even
more profound when one is a collector and reviewer of books,
and conversant with the history of ones eld. In Ecclesiastes
it is stated, of the making of books, there is no end (12:12).
Indeed, there are more books than any other human artifact on
earth. There is, however, a beginning to the making of books
in any given eld. The rst ophthalmology book to be published
was Benvenuto Grassis De Oculis in Florence in 1474. Firmin
Didot in his famous Bibliographical Encyclopedia wrote that
Grassus, an Italian physician of the School of Solerno, lived in
the 12th century and was the author of two books, the Ferrara
Quarto (1474) and the Venetian Folio (1497). Eye care in the
15th century was in the hands of itinerant barber surgeons and
quacks, and a treatise by a learned physician was a remarkable
occurrence. The next book on the eye to appear was an anonymous pamphlet written for the layperson in 1538 and entitled
Ein Newes Hochnutzliches Bchlin von Erkantnus der
Kranckheyten der Augen. Like Principles and Practice of
Ophthalmology, the Bchlin stated its intention to provide
highly useful knowledge of eye diseases, the anatomy of the eye,
and various remedies. It was illustrated with a fullpage woodcut
of the anatomy of the eye (Fig. 1). At the conclusion of the book,
the publisher, Vogtherr, promised to bring more and better
information to light shortly, and indeed, the next year he
published a small book by Leonhart Fuchs (15011566) entitled
Alle Kranckheyt der Augen.
Fuchs, a fervent Hippocratist, was Professor rst of Philosophy and then of Medicine at Ingolstadt, Physician of the
Margrave Georg of Brandenburg, and nally Professor at
Tbingen for 31 years. Like the earlier Bchlin, his work begins
with an anatomic woodcut (Fig. 2) and then lists in tabular

form various eye conditions, including strabismus, paralysis,
amblyopia, and nictalops. The work uses a distinctly Greco
Roman terminology, presenting information on the parts of the
eye and their affections, including conjunctivitis, ophthalmia,
carcinoma, and glaucoma. The book concludes with a remedy
collection similar to that found in the Bchlin. Most signicant
in the association of Leonhart Fuchs with this book is the fact
that a properly trained and well recognized physican addressed
the subject of ophthalmology.
Julius Hirschberg, the ophthalmic historian, noted that
Fuchs Alle Kranckheyt, along with the anonymous Bchlin,
apparently influenced Georg Bartisch in his writing of Das Ist
Augendienst. This latter work, published in 1583, marked the
founding of modern ophthalmology. Bartisch (15351606) was
an itinerant barber surgeon but nonetheless a thoughtful and
skillful surgeon, whose many innovations included the rst
procedure for extirpation of the globe for ocular cancer. Bartisch
proposed standards for the individual who practices eye surgery,
noting that rigorous training and concentration of effort were
needed to practice this specialty successfully.
By the late 16th century, eye surgery and the treatment of eye
disease began to move into the realm of the more formally
trained and respected surgeon. This is evidenced by Jacques
Guillemeaus Trait des Maladies de LOeil, published in 1585.
Guillemeau (15501612) was a pupil of the surgical giant
Ambroise Par, and his book was an epitome of the existing
knowledge on the subject.
The transition from couching of cataracts to the modern
method of treating cataracts by extraction of the lens, as
introduced by Jacques Daviel in 1753, further dened the skill
and training necessary for the care of the eyes. The initiation
of ophthalmology as a separate specialty within the realm of
medicine and surgery was signaled by the publication of George
Joseph Beers two volume Lehre von den Augenkrankheiten in
18131817. Beer (17631821) founded the rst eye hospital in
1786 in Vienna, and his students became famous ophthalmic
surgeons and professors throughout Europe.
In England, it was not only the demands of cataract surgery
but also the great pandemic of trachoma following the Napoleonic wars that led to the establishment of ophthalmology as a
recognized specialty. Benjamin Travers (17831858) published
the earliest treatise in English on diseases of the eye, A Synopsis
of the Diseases of the Eye, in 1820. In the United States,
acceptance of ophthalmology as a specialty had to await the
description of the ophthalmoscope by Helmholtz in 1851,
and the additional special skills that using the early primitive
Augenspiegel required.
As the complexity of ophthalmology increased and as subspecialization began to develop in the 19th century, multiauthored books began to appear. This culminated in the
appearance in 1874 of the rst volume of the GraefeSaemisch
xix
FIGURE 1.
xx
Handbuch. The nal volume of this great collective work, of

which Alfred Carl Graefe (18301899) and Edwin Theodor
Saemisch (18331909) were editors, appeared in 1880. The
denitive second edition, which for more than a quarter of a
century remained the most comprehensive and authoritative
work in the eld, appeared in 15 volumes between 1899 and
1918. The great French counterpart to the Graefe Saemisch
Handbuch was the Encyclopdie Franaise dOphtalmologie,
which appeared in nine volumes (19031910), edited by Octave
Doin, and lled a similar role for the French speaking
ophthalmologist.
In 1896, the rst of four volumes of Norris and Olivers
System of Diseases of the Eye was published in the United
States. The senior editor, Dr. William Fisher Norris
(18391901), was the rst Clinical Professor of Diseases of
the Eye at the University of Pennsylvania. Charles A. Oliver
(18531911) was his student. Norris considered the System
to be his monumental work. For each section he chose an
outstanding authority in the eld, having in the end more than
60 American, British, Dutch, French, and German ophthalmologists as contributors. Almost 6 years of combined labor on
the part of the editors was needed for completion of the work.
In 1913, Casey A. Wood (18561942) introduced the rst of
his 18 volumes of the American Encyclopedia and Dictionary
of Ophthalmology. The nal volume appeared in 1921. Drawn
largely from the Graef Saemisch Handbuch and the Encyclopdie Franaise dOphtalmologie, Woods Encyclopedia
provided information on the whole of ophthalmology through a
strictly alphabetic sequence of subject headings.
The book from which the present work draws inspiration
is Duke Elders Textbook of Ophthalmology (7 volumes; 1932)
and particularly the second edition of this work entitled System
of Ophthalmology (15 volumes, published between 1958 and
1976). The System of Ophthalmology was written by Sir
Stewart Duke Elder (18981978) in conjunction with his
colleagues at the Institute of Ophthalmology in London. In
1976, when the last of his 15 volumes appeared, Duke Elder
wrote in the Preface:
FIGURE 2.
The writing of these two series, the Textbook and the System,
has occupied all my available time for half a century. I cannot
deny that its completion brings me relief on the recovery of my
freedom, but at the same time it has left some sadness for I have
enjoyed writing it. As Edward Gibbon said on having written
the last line of The Decline and Fall of the Roman Empire:
A sober melancholy has spread over my mind by the idea
that I have taken everlasting leave of an old and agreeable
companion.
Duke Elder adds a nal line that I hope will be more propos
to the present editors and contributors. At the same time
the prayer of Sir Francis Drake on the eve of the attack of the
Spanish Armada is apposite: Give us to know that it is not the
beginning but the continuing of the same until it is entirely
nished which yieldeth the true glory. The void that developed
as the Duke Elder series became outdated has been partially
lled by many ne books, notably Thomas Duanes excellent 5
volume Clinical Ophthalmology.
Inspiration to undertake a major work such as this is derived
not only from the past books but also from teachers and role
models. For me, this includes Francis Heed Adler, Harold
G. Scheie, William C. Frayer, David G. Cogan, Ludwig von
Sallmann, Alan S. Rabson, Lorenz E. Zimmerman, Frederick C.
Blodi, Claes H. Dohlman, and Matthew D. Davis.
Whereas the inspiration for the present text was derived from
Duke Elders Textbook and System and from teachers and role
models, learning how to write and organize a book came for
me from Adlers Textbook of Ophthalmology, published by
W.B. Saunders. This popular textbook for medical students and
general practitioners was rst produced by Dr. Sanford Gifford
(18921945) in 1938. Francis Heed Adler (18951987), after
writing the 6th edition, published in 1962, invited Harold G.
Scheie (19091989), his successor as Chairman of Ophthalmology at the University of Pennsylvania, and myself to take
over authorship. We completely rewrote this book and noted
in the Preface to the 8th edition, published in 1969: This
book aims to provide the medical student and the practicing
physician with a concise and profusely illustrated current text,

organized in a convenient and useable manner, on the eye and
its disorders. It is hoped that the beginning, or even practicing,
ophthalmologist may nd it of value.
In 1969 it was apparent that even for the intended audience,
contributions by individuals expert in the subspecialties of
ophthalmology were required. The book was published in
Spanish and Chinese editions and was popular enough to
warrant an updated 9th edition, which appeared in 1977. One
of the high points of this work was interacting with John
Dusseau, the Editor in Chief for the W.B. Saunders Company.
As a 10th edition was contemplated, I became increasingly
convinced that what was needed in current ophthalmology was
a new, comprehensive, well illustrated set of texts intended
for the practicing ophthalmologist and written by outstanding
authorities in the eld. I envisioned a work that in one series of
volumes would provide all of the basic clinical and scientic
information required by practicing ophthalmologists in their
everyday work. For more detailed or specialized information,
this work should direct the practitioner to the pertinent journal
articles or more specialized publications. As time progressed, a
plan for this work took shape and received support from the
W.B. Saunders Company.
Memories of the formative stages of the Principles and
Practice of Ophthalmology remain vivid: Proposing the project
to Frederick Jakobiec in the cafeteria of the Massachusetts
Eye and Ear Inrmary in early 1989. Having dinner with Lewis
Reines, President and Chief Executive Ofcer, and Richard
Zorab, Senior Medical Editor, at the Four Seasons Hotel in
May 1989, where we agreed upon the scope of the work. My
excitement as I walked across the Public Garden and down
Charles Street back to the Inrmary, contemplating the work
we were to undertake. Finalizing the outline for the book in
Henry Allens well stocked faculty lounge in a dormitory at
Colby College during the Lancaster Course. Meeting with
members of the Harvard Faculty in the somber setting of the
rare book room to recruit the Section Editors. Persuading Nancy
Robinson, my able assistant since 1969, to take on the job of
Managing Editor. The receipt of our rst manuscript from Dr.
David Cogan.
We considered making this work a departmental undertaking, utilizing the faculty and alumni of various Harvard
programs. However, the broad scope of the series required
recruitment of outstanding authors from many institutions.
Once the Section Editors were in place, there was never any
doubt in my mind that this work would succeed. The Section
Editors proved a hardworking and dedicated group, and their
choice of authors reflects their good judgment and persuasive
abilities. I believe that you will appreciate the scope of
knowledge and the erudition.
The editorship of this book provided me not only with an
insight into the knowledge and thinking of some of the nest
minds in ophthalmology but also with an insight into their
lives. What an overwhelmingly busy group of people! Work was
completed not through intimidation with deadlines but by
virtue of their love of ophthalmology and their desire to share
their knowledge and experience. The talent, commitment,
persistence, and good humor of the authors are truly what made
this book a reality.
It was our intent to present a work that was at once scholarly
and pragmatic, that dealt effectively with the complexities
and subtleties of modern ophthalmology, but that did not
overwhelm the reader. We have worked toward a series of
volumes that contained the relevant basic science information
to sustain and complement the clinical facts. We wanted a
well illustrated set that went beyond the illustrations in any
textbook or system previously published, in terms of quantity

and quality and usefulnesss of the pictures.
In specic terms, in editing the book we tried to identify
and eliminate errors in accuracy. We worked to provide as
uniform a literary style as is possible in light of the numerous
contributors. We attempted to make as consistent as possible
the level of detail presented in the many sections and chapters.
Related to this, we sought to maintain the length according to
our agreed upon plan. We tried, as far as possible, to eliminate
repetition and at the same time to prevent gaps in information.
We worked to direct the location of information into a logical
and convenient arrangement. We attempted to separate the
basic science chapters to the major extent into the separate
Basic Sciences volume, but at the same time to integrate basic
science information with clinical detail in other sections as
needed. These tasks were made challenging by the size of the
work, the number of authors, and the limited options for
change as material was received close to publishing deadlines.
We believe that these efforts have succeeded in providing
ophthalmologists and visual scientists with a useful resource in
their practices. We shall know in succeeding years the level of
this success and hope to have the opportunity to improve all
these aspects as the book is updated and published in future
editions. Bacon wrote: Reading maketh a full man, conference
a ready man, and writing an exact man. He should have added:
Editing maketh a humble man.
I am personally grateful to a number of individuals for
making this book a reality. Nancy Robinson leads the list. Her
intelligent, gracious, and unceasing effort as Managing Editor
was essential to its successful completion. Mr. Lewis Reines,
President of the W.B. Saunders Company, has a profound
knowledge of publishing and books that makes him a worthy
successor to John Dusseau. Richard Zorab, Senior Medical
Editor, and Hazel N. Hacker, Developmental Editor, are
thoroughly professional and supportive individuals with whom
it was a pleasure to work. Many of the black and white
illustrations were drawn by Laurel Cook Lhowe and Marcia
Williams; Kit Johnson provided many of the anterior segment
photographs. Archival materials were retrieved with the aid
of Richard Wolfe, Curator of Rare Books at the Francis A.
Countway Library of Medicine, and Chris Nims and Kathleen
Kennedy of the Howe Library at the Massachusetts Eye and Ear
Inrmary.
The most exciting aspect of writing and editing a work of
this type is that it puts one in touch with the present day
ophthalmologists and visual scientists as well as physicians
training to be ophthalmologists in the future. We hope that this
book will establish its own tradition of excellence and usefulness and that it will win it a place in the lives of ophthalmologists today and in the future.
EXPLICIT, scribes wrote at the end of every book.
EXPLICIT means it has been unfolded. Olmert notes in The
Smithsonian Book of Books, the unrolling or unfolding of
knowledge is a powerful act because it shifts responsibility from
writer to reader.... Great books endure because they help us
interpret our lives. Its a personal quest, this grappling with the
world and ourselves, and we need all the help we can get. We
hope that this work will provide such help to the professional
lives of ophthalmologists and visual scientists.
DANIEL M. ALBERT, M.D., M.S.
MADISON, WISCONSIN
xxi
List of Contributors
Juan-Carlos Abad MD
Clinica Oftalmologica de Medellin
Medellin
Colombia
Mark B Abelson MD CM FRCS
Associate Clinical Professor of
Ophthalmology
Massachusetts Eye and Ear Infirmary
Harvard Medical School
Clinical Senior Scientist
Schepens Eye Research Institute
Boston MA
USA
David H Abramson MD
Chief
Ophthalmic Oncology Service
Department of Surgery
Memorial Sloane Kettering Cancer Center
New York NY
USA
Martin A Acquadro MD
Perioperative Medical Doctor
Director
Department of Anesthesiology and Pain
Caritas Carney Hospital
Dorchester MA
USA
Anthony P Adamis MD
Chief Scientic Ofcer
Executive Vice President, Research &
Development
(OSI) Eyetech Pharmaceuticals
New York NY
USA
Wesley H Adams MD
Ophthalmology Resident
Department of Ophthalmology
Wake Forest University Eye Center
Winston-Salem NC
USA
Natalie A Afshari MD
Associate Professor of Ophthalmology
Duke University Eye Center
Durham NC
USA
Lloyd P Aiello MD PhD

Director of Beetham Eye Institute
Section Head of Eye Research
Joslin Diabetes Center
Beetham Eye Institute
Boston MA
USA
Levent Akduman MD
Assistant Professor of Ophthalmology
St Louis University School of Medicine
St Louis MO
USA
Marissa L Albano MD
c/o Robert P Murphy
The Retina Group of Washington
Fairfax VA
USA
Daniel M. Albert MD MS
Chair Emeritus, F. A. Davis Professor and
Lorenz F. Zimmerman Professor
Department of Ophthalmology and Visual
Sciences
Retina Research Foundation Emmett A.
Humble Distinguished Director
Alice R. McPherson, MD, Eye Research
Institute
University of Wisconsin
Madison WI
USA
Terry J Alexandrou MD
Chief Resident
Science
University of Chicago
Chicago IL
USA
Eduardo C Alfonso MD
Professor, Edward W D Norton Chair in
Ophthalmology
Medical Director
Ocular Microbiology Laboratory
Bascom Palmer Eye Institute
University of Miami
Miami FL
USA
Everett Ai MD
Director
Retina Unit
California Pacic Medical Center
San Francisco CA
USA
Jorge L Ali MD PhD

Professor and Chairman of Ophthalmology,
Miguel Hernandez University
Medical Director, VISSUM
Instituto Oftalmolgico de Alicante
Alicante
Spain
Lloyd M Aiello MD
Clinical Professor of Medicine
Joslin Diabetes Center Beetham Eye
Institute
Boston MA
USA
Hassan Alizadeh PhD

University of Texas Southwestern Medical
Center
Dallas TX
USA
Ibrahim A Al Jadaan MD
Chief
Glaucoma Division
King Khaled Eye Specialist Hospital
Riyadh
Kingdom of Saudi Arabia
Sabah Al-Jastaneiah MD
Consultant Ophthalmologist
Anterior Segment and Refractive Surgery
Division
Riyadh
Calliope E Allen MD
Fellow
Eye Plastics, Orbital and Cosmetic Surgery
Massachusetts Eye & Ear Inrmary
Boston MA
USA
David Allen BSc FRCS FRCOphth
Consultant Ophthamologist
Sunderland Eye Inrmary
Sunderland
United Kingdom
Robert C Allen MD (deceased)
Formerly Professor of Ophthalmology and
Pharmacology
Formerly Chairman, Department of
Ophthalmology
Virginia Commonwealth University
Richmond VA
USA
Albert Alm MD PhD
Professor
Department of Neuroscience, Ophthalmology
University Hospital
Uppsala
Sweden
Samar Al-Swailem MD
Anterior Segment Division
Riyadh
Abigail K Alt BA
c/o Thaddeus P Dryja MD
Massachusetts Eye and Ear Inrmary
Boston MA
USA
Michael M Altaweel MD FRCS(C)
Assistant Professor & Co-Director, Fundus
Photograph Reading Center
Science
Madison WI
USA
xxiii
Russell Anderson BA
Medical Writer
Dry Eye Department
Ophthalmic Research Associates
North Andover MA
USA
Christopher M Andreoli MD
Ophthalmologist
Boston MA
USA
Soa Androudi MD
First Department of Ophthalmology
Aristotle University of Thessaloniki
Thessaloniki
Greece
Leonard P K Ang MD MMed(Ophth) FRCS(Ed)
MRCOphth
Consultant
Department of Cataract and Comprehensive
Ophthalmology
Singapore National Eye Centre
Singapore
Fahd Anzaar MD
Research Coordinator
Massachusetts Eye Research and Surgery
Institute
Cambridge MA
USA
David J Apple MD
Professor of Ophthalmology and Pathology
Director of Research Pawek-Vallotton
University of South Carolina
Charleston SC
USA
Claudia A Arrigg MD MEd
Senior Surgeon
Lawrence General Hospital
Lawrence MA
USA
Pablo Artal PhD
Professor of Optics
Centro de Investigacion en Optica y
Nanosica (CiOyN)
Universidad de Murcia
Murcia
Spain
Penny Asbell MD
Professor of Ophthalmology
Mount Sinai Medical Center
New York NY
USA
George K Asdourian MD
Chief, Division of Ophthalmology
University of Massachusetts Memorial
Medical Center
Worcester MA
USA
xxiv
Isabelle Audo MD PhD

Ophthalmologist
Laboratory of Cellular Physiopathology and
Retinal Molecules
Faculty of Medicine
INSERM
Universit Pierre et Marie Curie
Hpital St Antoine
Paris
France
Gerd U Auffarth Priv-Doz Dr med
Research Group Leader
Heidelberg IOL & Refractive Surgery
Research Group
University of Heidelberg
Heidelberg
Germany
Robin K Avery MD
Section Head, Transplant Infectious Disease
Department of Infectious Diseases
Cleveland Clinic Foundation
Cleveland OH
USA
Dimitri T Azar MD
B A Field Chair of Ophthalmologic Research
Professor and Head, Department of
Ophthalmology and Visual Sciences
University of Illinois Eye and Ear Inrmary
Chicago IL
USA
Ann S Baker MD (deceased)
Formerly Director of the Infectious Disease
Service
Formerly Associate Professor of
Ophthalmology
Boston MA
USA
Mark Balles MD
Retina Center of Maine
South Portland ME
USA
Scott D Barnes MD
Fellow, Cornea Service, Massachusetts Eye
and Ear Inrmary and Harvard Medical
School
Chief, Ophthalmology and Refractive Surgery
Womack Army Medical Center
Fort Bragg NC
USA
Donald M Barnett MD
Assistant Clinical Professor of Medicine
Boston MA
USA
Neal Atebara MD
Ophthalmologist
Retina Center of Hawaii
Honolulu HI
USA
Neal P Barney MD
Sciences
University of Wisconsin School of Medicine
Madison WI
USA
Pelin Atmaca-Sonmez
Research Fellow
University of Michigan
Ann Arbor MI
USA
Fina C Barouch MD
Eye Institute
Lahey Clinic Medical Center
Peabody MA
USA
George B Bartley MD
Mayo Medical School
Chief Executive Officer
Mayo Clinic
Jacksonville FL
USA
Jason J S Barton MD PhD FRCPC
Director of Neuro-Ophthalmology
Professor and Canada Research Chair
Neuro-Ophthalmology
VGH Eye Care Center
Vancouver BC
Canada
Irmgard Behlau MD
Instructor In Medicine, Harvard Medical
School
Boston MA
USA
Jose I Belda MD PhD EBO
Chairman
Hospital de Torrevieja
Alicante
Spain
Jeffrey L Bennett MD PhD
Associate Professor of Neurology &
Ophthalmology
Department of Neurology
University of Colorado Health Sciences
Center
Denver CO
USA
Timothy J Bennett CRA FOPS
Ophthalmic Photographer
Penn State Milton S Hershey Medical Center
Hershey PA
USA
Gregg J Berdy MD FACS
Assistant Professor of Clinical Ophthalmology
& Visual Science
Science
Washington University School of Medicine
St Louis MO
USA
Carlo Roberto Bernardino MD FACS
Yale University School of Medicine
New Haven CT
USA
Vitaliano Bernardino MD
Ophthalmologist
Private Practice
Langhorne PA
USA
Eliot L Berson MD
Director, Electroretinography Service
Massachusetts Ear and Eye Inrmary
William F Chatlos Professor of
Ophthalmology
Boston MA
USA
Luigi Borrillo MD
Retina-Vitreous Associates Inc
El Camino Hospital
Mountain View, CA
USA
Alfred Brini MD
Emeritus Professor of Ophthalmology
Louis Pasteur University
Strasbourg
France
Robert Bhisitkul MD PhD

UCSF Beckman Vision Center
San Francisco CA
USA
Gary E Borodic MD
Ophthalmologist
Boston MA
USA
Donald L Budenz MD MPH

Associate Professor
Epidemiology and Public Health
Miami FL
USA
Ravinder D Bhui BApSc in Elec Eng

Senior Medical Student
Schulich School of Medicine and Dentistry
The University of Western Ontario
London ON
Canada
S Arthur Boruchoff MD
Professor (Retired)
Boston University School of Medicine
Boston MA
USA
Jurij Bilyk MD
Attending Surgeon
Oculoplastic and Orbital Surgery Service
Wills Eye Institute
Philadelphia PA
USA
Swaraj Bose MD
Associate Professor
University of California, Irvine
Irvine CA
USA
Valrie Biousse MD
Associate Professor of Ophthalmology and
Neurology
Emory Eye Center
Emory University School of Medicine
Atlanta GA
USA
Michael E Boulton PhD

Director of AMD Center
Sciences
University of Texas Medical Branch
Galveston TX
USA
Alan C Bird MD FRCS FRCOphth

Professor
Department of Clinical Ophthalmology
Moorelds Eye Hospital
London
United Kingdom
R W Bowman MD
Professor
Center
Dallas TX
USA
Amitabh Bharadwaj MD
Ophthalmologist
Wills Eye Hospital
Philadelphia PA
USA
Norman Paul Blair MD

Professor of Ophthalmology, Director of
Vitreoretinal Service
Sciences
University of Illinois
Chicago IL
USA
Barbara A Blodi MD
Associate Professor, Specialist in Retinal
Disease
Department of Ophthalmology & Visual
Sciences
University of Wisconsin-Madison
Madison WI
USA
Mark S Blumenkranz MD
Professor and Chairman
Stanford University School of Medicine
Stanford CA
USA
H Culver Boldt MD
University of Iowa
Iowa City IA
USA
Mark S Borchert MD
Childrens Hospital Los Angeles
Los Angeles CA
USA
Elizabeth A Bradley MD
Mayo Clinic
Rochester MN
USA
Periklis D Brazitikos MD
Department of Ophthamology
Aristotle University of Thessaloniki
Thessaloniki
Greece
Robert Breeze MD
Professor and Vice Chair
Deparment of Neurosurgery
Center
Aurora CO
USA
Neil M Bressler MD
The Wilmer Eye Institute
The Johns Hopkins University School of
Medicine
Baltimore MD
USA
Susan B Bressler MD
John Hopkins Hospital
Baltimore MD
USA
Angela N Buffenn MD MPH

Childrens Hospital Los Angeles
Los Angeles CA
USA
Scott E Burk MD PhD
Ophthalmologist
Cincinnati Eye Institute
Cincinnati OH
USA
Salim Butrus MD
Associate Clinical Professor
George Washington University
Washington DC
USA
David Callanan MD
Vitreoretinal Specialist
Texas Retina Associates
Arlington TX
USA
J Douglas Cameron MD
Clinical Ophthalmology
Scheie Eye Institute
University of Pennsylvania
Philadelphia PA
USA
Louis B Cantor MD
Indiana University School of Medicine
Indianapolis IN
USA
William A Cantore MD
Neurology
Penn State University College of Medicine
Hershey PA
USA
Jorge Cantu-Dibildox MD
Centro de Oftalmologia San Jose, S C
Fundacin de Ojos Vidaurri, A C
Monterrey NL
Mexico
Victoria Casas MD
Research Fellow
Ocular Surface Research & Education
Foundation
Miami FL
USA
Miriam Casper MD
c/o David J Apple MD
University of South Carolina
Charleston SC
USA
xxv
Robin J Casten PhD
Assistant Professor
Department of Psychiatry and Human
Behaviour
Thomas Jefferson University
Philadelphia PA
USA
Yara P Catoira MD
Indianapolis IN
USA
Jerry Cavallerano OD PhD
Assistant to the Director
Boston MA
USA
Samantha J Chai MD
Medical Resident
Cullen Eye Institute
Baylor College of Medicine
Houston TX
USA
Maria R Chalita MD PhD
Director of Cornea and Refractive Surgery
Federal University of Brazil
Sao Paulo
Brazil
Sherman M Chamberlain MD FACP FACG
Assistant Professor of Medicine
Gastroenterology and Hepatology
Medical College of Georgia
Augusta GA
USA
Audrey S Chan MD
Cornea and Refractive Surgery Fellow
Boston MA
USA
Chi-Chao Chan MD
Head, Immunopathology Section
National Eye Institute
National Institutes of Health
Bethesda MD
USA
Paul Chan MD
New York Presbyterian
Wiell Medical College of Cornell University
New York NY
USA
Matthew J Chapin MD
Ophthalmic Research Associates, Inc
North Andover MA
USA
Karen L Chapman MD
University of South Florida
Sarasota Memorial Hospital
Sarasota FL
USA
Eric Chen MD
Retina Research Center
Austin TX
USA
xxvi
Joe Chen MD
c/o Keith L Lane MD
ORA Clinical Research and Development
North Andover MA
USA
Antonio P Ciardella MD
Chief, Department of Ophthalmology
Denver Health Medical Center
Denver CO
USA
Julie A Chen MD
c/o Joan M OBrien MD
Division of Ophthalmology
University of California San Francisco
Medical Center
San Francisco CA
USA
Mortimer Civan MD
Professor of Physiology
Department of Physiology
University of Pennsylvania Health System
Philadelphia PA
USA
Teresa C Chen MD
Glaucoma Service
Assistant Professor of Medicine, Harvard
Medical School
Boston MA
USA
Zhou Chen PhD
Senior Pharmacologist and Toxicologist
Center for Drug Evaluation and Research
Food & Drug Administration
Silver Spring MD
USA
Patricia Chvez-Barrios MD
Clinical Assistant Professor
Departments of Ophthalmology & Pathology
Baylor College of Medicine and the Texas
Childrens Cancer Center
Houston TX
USA
Emily Y Chew MD
Medical Ofcer, Division of Biometry and
Epidemiology
Bethseda MD
USA
Mark Chiang MBBS
Birmingham and Midland Eye Centre
Birmingham
United Kingdom
James Chodosh MD
University of Oklahoma Health Sciences
Center
Oklahoma City OK
USA
Eva-Marie Chong MBBS
Physician
Arizona Medical Center
Peoria AZ
USA
Denise Chun BS
Doctoral Candidate in Genetics, Harvard
Medical School
Department of Molecular Biology
Massachusetts General Hospital
Boston MA
USA
Leo T Chylack Jr MD
Director of Research
Center for Ophthalmic Research
Brigham & Womens Hospital
Boston MA
USA
Liane Clamen MD
Boston MA
USA
John I Clark PhD
Professor, Biological Structure
School of Medicine
University of Washington
Seattle WA
USA
Glenn Cockerham MD
Clinical Associate Professor
Stanford University
Stanford CA
USA
Andre Cohen MD
Ophthalmologist
Marietta Eye Consultants
Marietta GA
USA
Elisabeth J Cohen MD
Director Cornea Service, Attending Surgeon,
Wills Eye Hospital
Professor, Department of Ophthalmology
Jefferson Medical College of Thomas
Jefferson University
Philadelphia PA
USA
Kathryn A Colby MD PhD
Director, Joint Clinical Research Center
Attending Surgeon, Cornea Service
Assistant Professor of Ophthalmology,
Boston MA
USA
Anne L Coleman MD PhD
Professor of Ophthalmology and
Epidemiology
Departments of Ophthalmology and
Epidemiology
Jules Stein Eye Institute
Los Angeles CA
USA
Hanna R Coleman MD
Staff Clinician
New York Presbyterian Hospital
Columbia University Medical Center
New York NY
USA
Joseph Colin MD
C H U Morvan
Brest
France
J Michael Collier PhD
Instructor, Harvard Medical School
Senior Medical Physicist
Department of Radiation Oncology
Boston MA
USA
Donald J DAmico MD
Ophthalmologist-in-Chief
Weill Cornell Medical College
New York NY
USA
Grant M Comer MD
Assistant Professor
Kellogg Eye Center
Ann Arbor MI
USA
Reza Dana MD MSc MPH

Director, Cornea and Refractive Surgery
Services
Professor, Harvard Medical School
Senior Scientist & W Clement Stone Scholar
Boston MA
USA
M Ronan Conlon MD FRCSC

Eye Physician and Surgeon
Midwest Eye Care Institute
Saskatoon SK
Canada
Kim E Cooper MD
Associate Professor
Southwest College of Naturopathic Medicine
Tempe AR
USA
James J Corbett MD
McCarty Professor and Chairman for
Neurology
University of Mississippi Medical Center
Jackson MS
USA
Miguel C Coma MD FEBOphth
Institute
Cambridge MA USA
Hospital de Len, Len, Spain
Aude Danan-Husson MD
Service dophtalmologie
Centre Hospitalier National dOphtalmologie
des Quinze-vingts
Paris
France
Adam G de la Garza MD
Chief Resident, Wake Forest University Eye
Center
Wake Forest University School of Medicine
Winston-Salem NC
USA
Margaret M DeAngelis PhD
Instructor of Ophthalmology
Boston MA
USA
Sheri L DeMartelaere MD
Director of Orbital and Ocular Trauma
Ophthalmology Service
Brooke Army Medical Center
Fort Sam Houston TX
USA
FRANZCO
Joseph L Demer MD PhD

Leonard Apt Professor of Ophthalmology
Neurology
Los Angeles CA
USA

University of Auckland Medical School
Auckland
New Zealand
Avninder Dhaliwal MD
University of Minnesota Medical School
Minneapolis MN
USA
Helen B Danesh-Meyer MBChB MD
Ronald P Danis MD
Professor of Ophthalmology and Visual
Science
Director, Fundus Photograph Reading Center
Science
Madison WI
USA
J Paul Dieckert MD
Center Director, Division of Ophthalmology
Scott and White Memorial Hospital
Temple TX
USA
Jason K Darlington MD
University of California at Davis
Sacramento CA
USA
Diana V Do MD
Fellow in Advanced Speciality Training in
Medical and Surgical Diseases of the Retina
The Johns Hopkins University School of
Medicine
Baltimore MD
USA
Linda R Dagi MD
Director of Adult Strabismus, Instructor in
Ophthalmology
Childrens Hospital
Boston MA
USA
Stefanie L Davidson MD
Assistant Clinical Professor, University of
Pennsylvania
Childrens Hospital of Philadelphia
Philadelphia PA
USA
Marshall G Doane PhD

Emeritus Senior Scientist
Boston MA
USA
Matthew A Dahlgren MD
Fellow, Cornea and Anterior Segement,
Minneapolis MN
USA
Janet L Davis MD MA
University of Miami
Miami FL
USA
Christopher Dodds MBBS MRCGP FRCA

Professor of Anaesthesia
Academic Anaesthetic Department
James Cook University Hospital
Middlesbrough
United Kingdom
Timothy J Daley BS
University of Wisconsin Hospital and Clinics
Madison WI
USA
Elizabeth A Davis MD FACS

Adjunct Clinical Assistant Professor,
University of Minnesota
Director, Minnesota Eye, Laser and Surgery
Center
Bloomington MN
USA
Claes H Dohlman MD PhD

Professor of Ophthalmology, Harvard Medical
School, Chief Emeritus
Cornea Service
Boston MA
USA
Marshall N Cyrlin MD
Clinical Professor of Biomedical Sciences
Eye Research Institute
Oakland University
Rochester MN
USA
Andrea P Da Mata MD
Ocular Immunology and Uveitis Foundation
Institute
Cambridge MA
USA
Bertil Damato MD PhD FRCOphth
Ocular Oncology Service
Royal Liverpool University Hospital
Liverpool
United Kingdom
Jose J de la Cruz MD
Cornea Fellow, Department of
Ophthalmology and Visual Science
University of Illinois at Chicago
Chicago IL
USA
Guy Donati MD
Chare DEnsign
Department of Pathology
University of Geneva
Geneva
Switzerland
xxvii
Eric D Donnenfeld MD FACS
Co-director, Cornea Division
Ophthalmic Consultants of Long Island
New York NY
USA
Arlene Drack MD
Chief of Ophthalmology, Childrens Hospital
Associate Professor
Center
Aurora CO
USA
Thaddeus P Dryja MD
Director, David C Cogan Eye Pathology
Laboratory
Boston MA
USA
David Dueker MD
The Eye Institute
Medical College of Wisconsin
Milwaukee WI
USA
Jay S Duker MD
Director New England Eye Center
Chairman and Professor of Ophthalmology
Tufts University School of Medicine
Tufts New England Medical Center
Boston MA
USA
Jennifer A Dunbar MD
Director of Pediatric Ophthalmology
Loma Linda University
Loma Linda CA
USA
James P Dunn MD
John Hopkins School of Medicine
Baltimore MD
USA
William J Dupps Jr, MD PhD
Associate Staff, Ophthalmology and
Biomedical Engineering
Cole Eye Institute
Cleveland Clinic and Lerner Research
Institute
Cleveland OH
USA
Marlene L Durand MD
Director of Infectious Diseases,
Assistant Professor of Medicine, Harvard
Medical School;
Infectious Diseases Unit
Boston MA
USA
Jonathan J Dutton MD PhD
Professor and Vice Chair
University of North Carolina
Chapel Hill NC
USA
xxviii
Chiara M Eandi MD
The LuEsther T Mertz Retinal Research
Fellow
Manhattan Eye, Ear and Throat Hospital
New York NY
USA
Deepak P Edward MD
Department of Ophthalmology - Suma Health
Systems
Northeastern Ohio University School of
Medicine
Akron OH
USA
Robert A Egan MD
Assistant Professor of Ophthalmology and
Neurology
Neurology
Casey Eye Institute
Portland OR
USA
David A Eichenbaum MD
Associate
Retina-Vitrous Associates of Florida
St Petersburg FL
USA
Susan E Eklund BA
Assistant, Department of Ophthalmology
Childrens Hospital
Boston MA
USA
Elizabeth C Engle MD
Associate Professor of Neurology
Department of Neurology, Program in
Genomics, Childrens Hospital
Boston MA
USA
Kristine Erickson OD PhD
Senior Director Clinical Affairs
Unigene Corporation
Boonton NJ
USA
Bita Esmaeli MD FACS
Associate Professor of Ophthalmology;
Director of Ophthalmic Plastic and
Reconstructive and Orbital Surgery
Fellowship
Department of Medicine
The University of Texas
Houston TX
USA
Aaron Fay MD
Interim Director, Ophthalmic Plastic Surgery
Assistant Clinical Professor of
Ophthalmology, Department of
Ophthalmology, Harvard Medical School
Boston MA
USA
Leonard Feiner MD PhD
Ophthalmology Department
Monteore Medical Center
Lawrence NY
USA
Sharon Fekrat MD
Assistant Professor
Vitreoretinal Surgery
Duke Eye Center
Durham NC
USA
Frederick L Ferris III MD

Director, Division of Epidemiology and
Clinical Research
Bethesda MD
USA
Howard F Fine MD MHSc
Vitreoretinal surgical fellow
Vitreous Retina Macula New York
New York NY
USA
Donald C Fletcher MD
Medical Director
Frank Stein & Paul May Center for Low Vision
Rehabilitation
Scientist, Smith-Kettlewell Eye Research
Institute
San Francisco CA
USA
Paul Flikier MD
Farmacia Alvarez, Heredia
Director, Centro Medico de la Vision
San Jose
Costa Rica
Richard P Floyd MD
Clinical Instructor
Boston MA
USA
Harry W Flynn Jr, MD
Professor, The J Donald M Gass
Distinguished Chair of Ophthalmology
The University of Miami Miller School of
Medicine
Miami FL
USA
Donald S Fong MD MPH
Director, Cinical Trials Research
Kaiser Permanente Southern California
Pasadena CA
USA
Ramon L Font MD
Professor of Pathology and Ophthalmology
The Sarah Campbell Blaffer Chair of
Ophthalmology
The Neurosensory Center
Houston TX
USA
Brian J R Forbes MD PhD
The Childrens Hospital of Philadelphia
Wallingford PA
USA
Rod Foroozan MD
Houston TX
USA
Bradley S Foster MD
Assistant Clinical Professor of Ophthalmology
New England Retina Consultants
West Springeld MA
USA
C Stephen Foster MD FACS
Founder and President
The Massachusetts Eye Research Institute
Clinical Professor of Ophthalmology
Cambridge MA
USA
David Friedman MD
Assistant Professor
Wilmer Eye Institute
John Hopkins University School of Medicine
Baltimore MD
USA
Jill A Foster MD
Assistant Clinical Professor
The William H Havener Eye Institute
The Ohio State University
Columbus OH
USA
Deborah I Friedman MD FAAN

Neurology
Neurology
University of Rochester School of Medicine
and Dentistry
Rochester NY
USA
Gary N Foulks MD FACS

Arthur & Virginia Keeney Professor of
Ophthalmology
University of Louisville School of Medicine
Louisville KY
USA
Tamara R Fountain MD
Associate Professor
Rush University in Chicago
Northbrook IL
USA
Gregory M Fox MD
Clinical Instructor of Ophthalmology
Allegheny University
Wilmington DE
USA
Thomas F Freddo OD PhD FAAO
Professor and Director
School of Optometry
University of Waterloo
Waterloo ON
Canada
Sharon F Freedman MD
Associate Professor of Pediatrics
Department of Pediatric Ophthalmology &
Strabismus
Durham NC
USA
K Bailey Freund MD
Retina Specialist
Vitreous-Retina-Macula Consultants of
New York
New York NY
USA
Thomas R Friberg MD
Professor of Ophthalmology, Professor of
Bioengineering
Director of the Retina Service
Bioengineering
UPMC Eye Center
Pittsburgh PA
USA
Alan H Friedman MD
Mount Sinai School of Medicine
New York NY
USA
Ephraim Friedman MD
Former Chief, Department of Ophthalmology
Harvard Medical School Retina Service
Boston MA
USA
Arthur D Fu MD
Ophthalmologist
Pacic Vision Foundation
San Francisco CA
USA
Anne B Fulton MD
Senior Associate in Ophthalmology
Childrens Hospital
Boston MA
USA
Ahmed Galal MD PhD
Department of Refractive Surgery
Vissum/Instituto Oftalmologico de Alicante
Alicante
Spain
Steven Galetta MD
Director, Neuro-Ophthalmology Services
Hospital of the University of Pennsylvania
Philadelphia PA
USA
Mark Gallardo MD
Resident Physician
Ofce of Border Health
Texas Tech University Health Sciences Center
El Paso TX
USA
Brenda Gallie MD FRCS(C)
Departments of Medical Biophysics and
Molecular and Medical Genetics
University of Toronto
Head of Cancer Informatics
University Health Network
Ontario Cancer Institute
Princess Margaret Hospital
Toronto ON
Canada
Alec Garner MD
Head of Department
Institute of Ophthalmology
London
United Kingdom
James A Garrity MD
Mayo Clinic
Rochester MN
USA
Damien Gatinel MD
Assistant Professor
Fondation Ophtalmologique A de Rothschild
Paris
France
Steven J Gedde MD
Professor of Ophthalmology and Residency
Program Director
Miami FL
USA
Craig E Geist MD FACS
Chairman, Department of Ophthalmology
Associate Professor, Ophthalmology,
Neurology, Neurological Surgery
Director, Oculoplastics, Orbit, Lacrimal
Director, Neuro-Ophthalmology
The George Washington University
Washington DC
USA
Steve Gerber MD
Chairman
Memorial Hospital
South Bend IN
USA
Ramon C Ghanem MD
Sadalla Amin Ghanem
Hospital de Olhos
Batista
Joinville - SC
Brazil
Jon P Gieser MD
Wheaton Eye Clinic
Wheaton IL
USA
Michael S Gilmore PhD
Charles L Schepens Professor of
Ophthalmology
President and Ankeny Director of Research
The Schepens Eye Research Institute
Boston MA
USA
Howard V Gimbel MD MPH FRCSC FACS
Chair and Professor of The Department of
Ophthalmology
Refractive Surgery, Department of
Ophthalmology
Loma Linda University
Loma Linda CA
USA
Ilene K Gipson PhD
Senior Scientist and Professor of
Ophthalmology
Boston MA
USA
xxix
Tyrone Glover MD
Clinical Professor, Ophthalmology
Oculoplastic Surgery
Kaiser Permanente
Sacramento CA
USA
Robert A Goldberg MD FACS
Chief, Division of Orbital and Ophthalmic
Plastic Surgery
Los Angeles CA
USA
Mordechai Goldenfeld MD
Senior Attending Ophthalmologist
The Sam Rothberg Glaucoma Centre
Goldschleger Eye Institute
Sheba Medical Center
Tel-Hashomer
Israel
Scott M Goldstein MD
Clinical Associate
Tricounty Eye Physicians & Surgeons
Southampton PA
USA
Vamsi K Gullapalli MD PhD

Resident
Science
Institute of Ophthalmology and Visual
Science
University of Medicine and Dentistry of
New Jersey
Newark NJ
USA
Michael J Greaney
Senior Clinical Lecturer, Department of
Ophthalmology, University of Bristol
Senior Consultant
Bristol Eye Hospital
Bristol
United Kingdom
Padma Gulur MD
Instructor in Anaesthesia, Harvard Medical
School
Pain Center Department of Anesthesia and
Critical Care
Boston MA
USA
Daniel G Green PhD

Professor Emeritus, Ophthalmology and
Visual Sciences
Professor, Biomedical Engineering
The University of Michigan Kellogg Eye
Center
Ann Arbor MI
USA
Cintia F Gomi MD
Hamilton Glaucoma Center
University of California, San Diego,
La Jolla CA
USA
Franz Grehn Dr h.c.

University of Wrzburg
Wrzburg
Germany
Haiyan Gong MD MS PhD

Research Assistant Professor
Boston MA
USA
Jack V Greiner DO PhD

Boston MA
USA
John A Gonzales MD
Physician
Immunopathology Section
Laboratory of Immunology
Besthesda MD
USA
Craig M Greven MD FACS

Director, Professor and Chairman
Wake Forest University School of Medicine
Winston-Salem NC
USA
John Goosey MD
Director
Houston Eye Associates
Houston TX
USA
Justin L Gottlieb MD
Associate Professor
Sciences
Madison WI
USA
Joshua Gould DO
Physician
Eye Care Center of New Jersey
Bloomingeld NJ
USA
Evangelos S Gragoudas MD
Director, Retina Service
Professor of Ophthalmology, Harvard Medical
School
Boston MA
USA
xxx
David B Granet MD FACS FAAP FAAO

Anne F Ratner Professor of Ophthalmology &
Pediatrics
Director, Pediatric Ophthalmology & Adult
Re-Alignment Services
Anne F & Abraham Ratner Childrens Eye Center
Shiley Eye Center
University of California, San Diego
La Jolla CA
USA
Gregory J Griepentrog MD
Chief Resident Associate
Mayo Clinic
Rochester MN
USA
Carl Groenewald MD
Consultant Vitreoretinal Surgeon
St Pauls Eye Unit
Royal Liverpool University Hospital
Liverpool
United Kingdom
Cynthia L Grosskreutz MD PhD
Co-Director, Glaucoma Service
Boston MA
USA
Lori Latowski Grover OD
Lions Vision Research and Rehabilitation
Center
Baltimore MD
USA
Jonathan Gunther MD
Sciences
University of Wisconsin Medical School
Madison WI
USA
Manish Gupta DNB FRCS(Glasg), MRCS(Ed)
NHS Greater Glasgow and Clyde
Stobhill and Gartnevel Hospital
Glasgow
United Kingdom
Mayank Gupta
c/o Deepak P Edward MD
Northeastern Ohio University School of
Medicine
Akron OH
USA
David R Guyer MD
Clinical Professor
NYU Medical Center
New York NY
USA
Darin R Haivala MD
University of Oklahoma
Dean A McGee Eye Institute
Oklahoma City OK
USA
Julia A Haller MD
Robert Bond Welch Professor of
Ophthamology
Wilmer Ophthalmological Institute
Johns Hopkins Medical Institutions
Baltimore MD
USA
G M Halmagyi MD BSc FACS DCH
Professor of Neurology
Royal Prince Albert Hospital
Sydney NSW
Australia
Lawrence S Halperin MD FACS
Physician
Retina Vitreous Consultants of South Florida
Fort Lauderdale FL
USA
Islam M Hamdi FRCS MD
Magrabi Center
Jeddah
Steven R Hamilton MD
Clinical Associate Professor of
Ophthalmology and Neurology
Department of Ophthalmology and Neurology
Neuro-Ophthalmic Associates Northwest
Seattle WA
USA
Kristin M Hammersmith MD
Assistant Surgeon, Cornea Service, Wills Eye
Hospital
Instructor, Thomas Jefferson Medical College
Wills Eye Institute
Philadelphia PA
USA
Dennis P Han MD
Jack A and Elaine D Klieger Professor of
Ophthalmology, Vitreoretinal Section Head
Milwaukee WI
USA
Ronald M Hansen PhD
Instructor
Childrens Hospital and Harvard Medical
School
Boston MA
USA
J William Harbour MD
Distinguished Professor of Ophthalmology,
Cell Biology, and Medicine (Molecular
Oncology)
Director, Ocular Oncology Service
St Louis MO
USA
Seenu M Hariprasad MD
Assistant Professor and Director of Clinical
Research
Chief, Vitreoretinal Service
Science
Chicago IL
USA
Mona Harissi-Dagher MD
Assistant in Ophthalmology
Boston MA
USA
Shirin E. Hassan PhD
c/o David Friedman
Assistant Professor
Baltimore MD
USA
Mark P Hatton MD
Clinical Instructor, Harvard Medical School
Adjunct Clinical Scientist
Ophthalmic Consultants of Boston
Boston MA
USA
Pamela Hawley MS
Genetic Counseling Center
Childrens Hospital
Harvard
Boston MA
USA
Yasutaka Hayashida MD PhD

Research Fellow
Ocular Surface Research and Education
Foundation
Miami FL
USA
John R Heckenlively MD FRCOpath
Paul R Lichter Professor of Ophthalmic
Genetics
Science
Kellogg Eye Centre
Ann Arbor MI
USA
Thomas R Hedges III, MD
Director, Neuro-Ophthalmology Service
Co-Director, Electrophysiology Service
Director, Neuro-Ophthalmology Fellowship
Program
New England Eye Center
Boston MA
USA
Alfred D Heggie MD
Professor Emeritus of Pediatrics
Departments of Pediatrics, Preventive
Medicine, and Obstetrics and Gynecology
Case Western Reserve University School of
Medicine
Cleveland OH
USA
Katrinka L Heher MD
Director, Aesthetic Eyelid & Facial Surgery
Director, Oculoplastic & Orbital Surgery
Service
Director, Ophthalmic Plastics and
Reconstructive Surgery Fellowship Program
Boston MA
USA
Jeffrey S Heier MD
Vitreoretinal Specialist
Boston MA
USA
Ahmed A Hidayat MD
Chief, Ophthalmic Pathology
Armed Forces Institute of Pathology
Rockville MD
USA
Eva Juliet Higginbotham MD
Professor of Ophthalmology and Chair
University of Maryland Medicine
Baltimore MD
USA
Tatsuo Hirose MD
Schepens Retina Associates
Boston MA
USA
Allen C Ho MD
Retina Service
Philadelphia PA
USA
ThucAnh T Ho MD
Vitreoretinal Fellow
Illinois Retina Associates
Rush University Medical Center
Chicago IL
USA
R Nick Hogan MD PhD
University of Texas South Western Medical
Center
Dallas TX
USA
David E Holck MD
Director, Oculoplastic, Reconstructive, Orbit,
and Ocular Oncology Service
Wilford Hall Medical Center
Assistant Professor of Surgery, USUHS
University of Texas
San Antonio TX
USA
J Fielding Hejtmancik MD PhD

Medical Ofcer
Ophthalmic Genetics and Visual Function
Branch
Bethesda MD
USA
Nancy M Holekamp MD
Associate Professor of Clinical
Ophthalmology
Science
Barnes Retina Institute
St Louis MO
USA
Bonnie A Henderson MD FACS

Boston MA
USA
Peter G Hovland MD PhD

Physician
Colorado Retina Associates
Denver CO
USA
Peter S Hersh MD FACS

Director, Cornea and Laser Eye Institute Hersh Vision Group
Chief, Cornea and Refractive Surgery
University of Medicine and Dentistry New
Jersey
Teaneck NJ
USA
Thomas C Hsu MD
Boston MA
USA
William C Hsu MD
Boston MA
USA
xxxi
Andrew J W Huang MD MPH
Director of Cornea and Refractive Surgery
Minneapolis MN
USA
Fei Ji PhD
Research Associate
Laboratory of Statistical Genetics
Rockefeller University
New York NY
USA
Mark S Hughes MD
Adjunct Assistant Clinical Scientist
The Schepens Eye Research Institute
Boston MA
USA
David L Johnson MD
Clinical Instructor/Vitreoretinal Fellow
Sciences
University Of Wisconsin Medical School
Madison WI
USA
Jennifer Hui MD
Ophthalmology Resident
Miami FL
USA
David G Hunter MD PhD
Associate Professor of Ophthalmology,
Ophthalmologist-in-Chief
Richard Robb Chair in Ophthalmology
Childrens Hospital Boston
Boston MA
USA
Laryssa A Huryn MD
Miami FL
USA
Deeba Husain MD
Retina Service - Dept of Ophthalmology
Boston MA
USA
Robert A Hyndiuk MD
The Eye Institute
Milwaukee WI
USA
Michael Ip MD
Sciences
Fundus Photograph Reading Center
Madison WI
USA
Brian J Jacobs MD
Chicago IL
USA
Frederick A Jakobiec MD DSc(Med)
Former Henry Willard Williams Professor and
Former Chief of Ophthalmology
Departments of Pathology and
Ophthalmology
Boston MA
USA
Lee M Jampol MD
Louis Feinberg Professor and Chairman
Northwestern University Medical School
Chicago IL
USA
xxxii
Harold G Jensen PhD

Clinical Project Manager
Allergan, Inc
Irvine CA
USA
Douglas H Johnson MD (deceased)

Formerly Professor of Ophthalmology
Mayo Clinic
Rochester MN
USA
Mark W Johnson MD
Professor
Kellogg Eye Center
Ann Arbor MI
USA
R Paul Johnson MD
Associate Professor of Medicine
Infectious Diseases Unit
Charlestown MA
USA
Robert N Johnson MD
University of California
West Coast Retina Medical Group
San Francisco CA
USA
Karen M Joos MD PhD
Associate Professor
Sciences
Vanderbilt University
Nashville TN
USA
Nancy C Joyce PhD
Senior Scientist Associate Professor, Harvard
Medical School
Boston MA
USA
J Michael Jumper MD
Director, Retina Service
San Francisco CA
USA
Ula V. Jurkunas MD
Instructor in Ophthalmology
Clinical Scientist
Boston MA
USA
Alon Kahana MD PhD

Assistant Professor; Eye Plastics, Orbit and
Facial Cosmetic Surgery
Sciences
Kellogg Eye Center
Ann Arbor MI
USA
Malik Y Kahook MD
Assistant Professor and Director of Clinical
Research
Rocky Mountain Lions Eye Institute
University of Colorado at Denver Health
Sciences Center
Aurora CO
USA
Elliott Kanner MD PhD
Hamilton Eye Institute
University of Tennessee Health Science
Center
Memphis TN
USA
Kevin Kalwerisky MD
Department of Otolaryngology, Head & Neck
Surgery
The New York Presbyterian Hospital
Weill Medical College of Cornell University
New York NY
USA
Henry J Kaplan MD
Sciences
University of Louisville
Louisville KY
USA
Ekaterini C Karatza MD
Staff Ophthalmologist
Cincinnati Eye Institute
Cincinnati OH
USA
Randy Kardon MD PhD
Director of Neuro-ophthalmology
The University of Iowa Hospitals and Clinics
Iowa City IA
USA
James A Katowitz MD
Attending Surgeon
Ophthalmology
Philadelphia PA
USA
William R Katowitz MD
University of Rochester School of Medicine
and Dentistry
Rochester NY
USA
Melanie Kazlas MD
Acting Director; Instructor
Pediatric Ophthalmology & Strabismus
Boston MA
USA
Kelly S Keefe CAPT MC USN
Staff Ophthalmic Pathologist
Comprehensive Ophthalmologist
Naval Medical Center
San Diego CA
USA
Lara Kelley MD
Assistant Professor, Dermatology
Beth Israel Deaconess Medical Center
Boston MA
USA
Charles J Kent MD
Fellowship Training in Ocuplastics and Ocular
Pathology
Everett & Hurite Ophthalmology Associates
Pittsburgh PA
USA
Kenneth R Kenyon MD FACS
Harvard Medical School;
Eye Health Vision Centers
North Dartmouth MA
USA
Bilal F Khan MD
Assistant in Ophthalmology
Boston MA
USA
Jemshed A Khan MD
Kansas University
Kansas City MO
USA
Naheed W Khan PhD
Electrophysiologist
Sciences
W K Kellogg Eye Center
Ann Arbor MI
USA
Peng Tee Khaw PhD FRCP FRCS FRCOphth
FIBiol FRCPath FMedSci
Professor of Glaucoma and Ocular Healing
and Consultant Ophthalmic Surgeon
Biomedical Research Centre (Ophthalmology)
UCL Institute of Ophthalmology and
London
United Kingdom
Femida Kherani MD FRCSC
Ophthalmic Cosmetic Surgeon
Heights Laser Centre
Burnaby BC
Canada
Eva C Kim MD
Fellow in Ocular Inflammation/Uveitis
The Proctor Foundation
San Francisco CA
USA
Hee Joon Kim MD
Resident
Science
University of Texas Health Science Center at
Houston
Houston TX
USA
Ivana K Kim MD
Retina Service
Boston MA
USA
Jonathan W Kim MD
Physician
Memorial Sloan-Kettering Cancer Center
New York NY
USA
Rosa Y Kim MD
Physician
Vitreoretinal Consultants
Houston TX
USA
Stella K Kim MD
Section of Ophthalmology
MD Anderson Cancer Center
Houston TX
USA
Thomas Kohnen MD
Deputy Chairman
Klinik fur Augenheilkunde
Johann Wolfgang Goethe University
Frankfurt
Germany
Takeshi Kojima MD PhD
Research Group for Environmental
Conservation Processing
Department of Material Development
Takasaki Radiation Chemistry Research
Establishment
Japan Atomic Energy Research Institute
Takasaki-shi
Japan
Tae-Im Kim MD PhD

Yonsei University Health System
Seoul
South Korea
Tobias Koller MD
Refractive Surgeon
Institute of Ophthalmic and Refractive
Surgery
Zurich
Switzerland
Christina M Klais MD
Retina Fellow
LuEsther T Mertz Retinal Research Center
New York NY
USA
David A Kostick MD
Mayo Clinic College of Medicine
Jacksonville FL
USA
Stephen R Klapper MD FACS

Ophthalmologist
Klapper Eyelid & Facial Plastic Surgery
Carmel IN
USA
Joel A Kraut MD
Medical Director
Vision Rehabilitation Service
Boston MA
USA
Barbara E K Klein MD MPH

Professor of Medicine
Sciences
University of Wisconsin Medical School
Madison WI
USA
Guy Kleinmann MD
Adjunct Assistant Professor of
Ophthalmology
Herman Eye Center
Houston TX
USA
Thomas Klink DrMed
Scientic Assistant
University of Wrzburg
Wrzburg
Germany
Dino D Klisovic MD
Nationwide Childrens Hospital
Midwest Retina Inc
Columbus OH
USA
Stephen D Klyce
Executive Editor
Louisiana State University Eye Institute
New Orleans LA
USA
Tolga Kocaturk MD
Adnan Menderes University Medical School
Aydin
Turkey
Chandrasekharan Krishnan MD
Glaucoma and Cataract Service
Boston MA
USA
Ronald R Krueger MD MSE
Director of Refractive Surgery, Cleveland
Clinic Foundation, Cleveland, OH, USA
Saint Louis University Eye Institute
Saint Louis University School of Medicine
St Louis MO
USA
Joseph H Krug Jr, MD
Assistant Director of Glaucoma Consultation
Service
Boston MA
USA
Sara Krupsky MD
Tel Hashomer
Israel
Rachel W Kuchtey MD PhD
Clinical Ophthalmologist, Glaucoma
Vanderbilt University of Ophthalmology &
Visual Sciences
Nashville TN
USA
xxxiii
Ramsay S Kurban MD
Department of Dermatology
Penn State University
Milton S Hershey Medical Center
Hershey PA
USA
Paul A Kurz MD
Casey Eye Institute
Oregon Health & Science University
Portland OR
USA
J R Kuszak PhD
Pathology
Chicago IL
USA
Young H Kwon MD PhD
University of Iowa
Iowa City IA
USA
Thad A Labbe MD
Glaucoma Specialist
Ophthalmologist
Eye Associates of Central Texas
Austin TX
USA
Deborah L Lam MD
Pacic Northwest Eye Associates
Tacoma WA
USA
Jeffrey C Lamkin MD
Akron City Hospital
The Retina Group of NE Ohio Inc
Akron OH
USA
Kathleen A Lamping MD
Case Western Reserve University
South Euclid OH
USA
Anne Marie Lane MPH
Clinical Research Manager, Retina Service
Instructor in Ophthalmology, Harvard Medical
School
Boston MA
USA
Katherine A Lane MD
Resident, Oculoplastic and Orbital Surgery
Service
Wills Eye Hospital
Philadelphia PA
USA
Keith J Lane MD
Senior Manager, Research and Development
/Preclinical
ORA Clinical Research and Development
North Andover MA
USA
xxxiv
Jonathan H Lass MD
Charles I Thomas Professor and Chairman
CWRU Department of Ophthalmology and
Visual Sciences Chairman
Sciences
University Hospitals Case Medical Center
Cleveland OH
USA
Mary G Lawrence MD MPH
Associate Professor, Glaucoma, Cataract and
Visual Rehabilitation
Glaucoma Service
Minneapolis MN
USA
Andrew G Lee MD
Professor of Ophthalmology, Neurology and
Neurosurgery
Departments of Ophthalmology, Neurology
and Neurosurgery
University of Iowa Hospitals
Iowa City IA
USA
Carol M Lee MD
Clinical Professor, Department of
Ophthalmology
NYU Medical Center
New York NY
USA
Michael S Lee MD
Associate Professor
Departments of Ophthalmology, Neurology,
and Neurosurgery
Minneapolis MN
USA
Paul P Lee MD JD
Durham NC
USA
William B Lee MD
Eye Consultant
Eye Consultants of Atlanta
Piedmont Hospital
Atlanta GA
USA
Igal Leibovitch MD
Oculoplastic and Orbital Division
Tel-Aviv Medical Center
Tel-Aviv
Israel
Bradley N Lemke MD FACS
Clinical Professor of Oculofacial Surgery
Sciences
University of Wisconsin - Madison
Madison WI
USA
Craig A Lemley MD
The Eye Institute
Milwaukee WI
USA
Andrea Leonardi MD
Assistant Professor in Ophthalmology
Department of Neuroscience, Ophthalmology
Unit
University of Padua
Padua
Italy
Simmons Lessell MD
Director, Neuro-Ophthalmology Service
Professor, Harvard Medical School
Boston MA
USA
Leonard A Levin MD PhD
Sciences, Neurology, and Neurological
Surgery
University of Wisconsin School of Medicine
and Public Health
Madison WI
USA
Canada Research Chair of Ophthalmology
and Visual Sciences
University of Montreal
Montreal QC
Canada
Grace A Levy-Clarke MD
Fellowship Program Director
Uveitis and Ocular Immunology
Bethesda MD
USA
Julie C Lew MD
Suny Downstate Medical Centre
Brooklyn NY
USA
Craig Lewis MD
Cole Eye Institute
Cleveland Clinic
Cleveland OH
USA
Wei Li MD PhD
Research Fellow
Ocular Surface Center
Miami FL
USA
Laurence S Lim MBBS
Principal Investigator
Singapore
Lyndell L Lim MBBS FRANZCO
Mankiewicz-Zelkin Crock Fellow
Centre for Eye Research Australia
University of Melbourne
East Melbourne VIC
Australia
Wee-Kiak Lim FRCOphth FRCS(Ed) MMED
Associate Consultant
Ocular Inflammation and Immunology
Singapore
Grant T Liu MD
Neuro-ophthalmologist
Philadelphia PA
USA
John I Loewenstein MD
Retina Service
Boston MA
USA
McGregor N Lott MD
Augusta GA
USA
Jonathan C Lowry MD
Ophthalmologist
Morganton Eye Physicians
Morganton NC
USA
David B Lyon MD FACS
Associate Professor
University of Missouri-Kansas City School of
Medicine
Prairie Village KS
USA
Robert E Lytle MD
Ophthalmologist
Boston MA
USA
Mathew MacCumber MD PhD
Associate Professor
Associate Chairman of Research
Chicago IL
USA
Bonnie T Mackool MD MSPH
Director of Dermatology
Consultation Service
Boston MA
USA
Nalini A Madiwale MD
Physician
Albany-Troy Cataract & Laser Associates
Troy NY
USA
Francis Mah MD
University of Pittsburgh Medical Center
Pittsburgh PA
USA
Martin A Mainster PhD MD FRCOphth
Fry Endowed Professor and Vice Chairman of
Ophthalmology
University of Kansas School of Medicine
Kansas City MO
USA
Michael H Manning Jr
c/o Sherman M Chamberlain MD FACP
FACG
Augusta GA
USA
Steven L Mansberger MD MPH
Associate Scientist
Devers Eye Institute
Portland OR
USA
Robert E Marc PhD

John A Moran Eye Center
Salt Lake City UT
USA
Mellone Marchong
Department of Applied Molecular Oncology
Ontario Cancer Institute - University Health
Network
Princess Margaret Hospital
Toronto ON
Canada
Dennis M Marcus MD
Professor of Clinical Ophthalmology
Department of Clinical Ophthalmology
Southeast Retina Center
Augusta GA
USA
Julie A Mares PhD
Professor
Sciences
WARF
Madison WI
USA
Brian P Marr MD
Oncology Service
Wills Eye Institute
Philadelphia PA
USA
Carlos E Martinez MS MD
Eye Physicians of Long Beach
Long Beach CA
USA
Robert W Massof PhD
Professor of Ophthalmology, Professor of
Neuroscience
Director, Lions Vision Research and
Rehabilitation Center
Wilmer Ophthalmological Institute
Johns Hopkins University School of Medicine
Baltimore MD
USA
Yukihiro Matsumoto
Research Fellow
Ocular Surface Research and Education
Foundation
Miami FL
USA
Cynthia Mattox MD
Ophthalmology - New England Eye Center
Tufts-New England Medical Center
Boston MA
USA
Marlon Maus MD
DrPH Candidate
University of California at Berkeley
Berkeley CA
USA
Cathleen M McCabe MD
Indiana LASIK Center
Fort Wayne IN
USA
Steven A McCormick MD
Director of Pathology and Laboratory
Medicine
The New York Eye and Ear Inrmary
New York NY
USA
Michael McCrakken
Clinical Instructor
Center
Denver CO
USA
James P McCulley MD
Professor & Chairman of Ophthalmology
Center
Dallas TX
USA
John A McDermott
New York Eye and Ear Inrmary
New York NY
USA
H Richard McDonald MD
Director, San Francisco Retina Foundation
Co-Director, Vitreoretinal Fellowship
California Pacic Retina Center
San Francisco CA
USA
Marguerite B McDonald MD FACS
Lynbrook NY
USA
Peter J McDonnell MD
William Holland Wilmer Professor of
Ophthalmology
Director, Wilmer Ophthalmological Institute
Baltimore MD
USA
Robert McGillivray BSEE CLVT
Director
Low Vision Services
The Carroll Center for the Blind
Low Vision Engineering Consultant
Massachusetts Commission for the Blind
Newton MA
USA
Craig A McKeown MD
Ophthalmology
Miller School of Medicine
University of Miami
Miami FL
USA
James McLaughlin MD
Medical Writer
Ophthalmic Research Associates, Inc
North Andover MA
USA
W Wynn McMullen MD
Vitereoretinal Consultant
Coastal Eye Associates
Houston TX
USA
Shlomo Melamed MD
The Sam Rothberg Glaucoma Centre
Tel-Hashomer
Israel
xxxv
George Meligonis FRCPath
Corneoplastic Unit
Queen Victoria Hospital
East Grinstead
East Sussex
United Kingdom
Efstratios Mendrinos MD
Ophthalmic Fellow
Ophthalmic Service
Geneva University
Geneva
Switzerland
Dale R Meyer MD
Director, Ophthalmic Plastic Surgery
Lions Eye Institute
Albany Medical Center
Albany NY
USA
Catherine B Meyerle MD
Retinal Physician
Bethesda MD
USA
William F Mieler MD
Science
Chicago IL
USA
Michael Migliori MD
The Warren Alpert Medical School
Brown University
Providence RI
USA
Martin C Mihm Jr, MD
Clinical Professor of Pathology
Senior Dermatopathologist
The Pigmented Lesion Clinic
Boston MA
USA
Darlene Miller DHSc MPH SM CIC
Research Assistant Professor
Scientific Director
Abrams Ocular Microbiology Laboratory
Anne Bates Leach Eye Hospital
Miller School of Medicine
University of Miami
Miami FL
USA
David Miller MD
Associate Clinical Professor of Ophthamology
Jamaica Plain MA
USA
Joan W Miller MD
Henry Willard Williams Professor of
Ophthalmology
Chief and Chair, Department of
Ophthalmology
Boston MA
USA
xxxvi
Neil R Miller MD
Professor of Ophthalmology, Neurology and
Neuro-Ophthalmology
Departments of Ophthalmology, Neurology
and Neuro-Ophthalmology
Johns Hopkins Hospital
Baltimore MD
USA
David M Mills MD
Oculofacial Plastic, Reconstructive, and
Cosmetic Surgeon
Nicolitz Eye Consultants
Jacksonville FL
USA
Monte D Mills MD
Chief, Division of Ophthalmology
Philadelphia PA
USA
Tatyana Milman MD
Co-director, Ophthalmic Pathology Division
Science
UMDNJ-New Jersey Medical School
Newark NJ
USA
A Linn Murphree MD
Director
The Retinoblastoma Centre
Childrens Hospital of Los Angeles
Los Angeles CA
USA
Robert P Murphy MD
The Retina Group of Washington
Fairfax VA
USA
Timothy G Murray MD MBA FACS
Miami FL
USA
Philip I Murray PhD FRCP FRCS FRCOphth
Academic Unit of Ophthalmology
City Hospital NHS Trust
Birmingham
United Kingdom
Karina Nagao MD
Boston MA
USA
Lylas Mogk MD
Director
Visual Rehabilitation and Research Center
Henry Ford Health System
Livonia MI
USA
Jay Neitz PhD

R D and Linda Peters Professor
Milwaukee WI
USA
Marja Mogk PhD

Assistant Professor of English
California Lutheran University
Los Angeles CA
USA
Maureen Neitz PhD

Richard O Schultz-Ruth A
Works-Ophthalmology Research Professor
The Eye Institute
Milwaukee WI
USA
Jordi Mons MD
Institut de la Macula i de la Retina
Barcelona
Spain
Robert Montes-Mic OD MPhil
Optica
Facultat de Fisica
Universidad de Valencia
Valencia
Spain
Christie L Morse MD
Concord Eye Care
Concord NH
USA
Asa D Morton MD
Eye Care of San Diego/CA Laser Vision, Inc
Escondido CA
USA
Anne Moskowitz OD PhD
Research Associate in Ophthalmology
Childrens Hospital, Boston
Boston MA
USA
Shizuo Mukai MD
Boston MA
USA
Peter A Netland MD PhD

Siegal Professor of Ophthalmology, Director
of Glaucoma, Academic Vice-Chair
Hamilton Eye Institute
University of Tennessee Health Science
Center
Memphis TN
USA
Arthur H Neufeld PhD
Forsythe Laboratory for the Investigation of
Aging Retina
Northwestern University Fienberg School of
Medicine
Chicago IL
USA
Nancy J Newman MD
Professor of Ophtalmology and Neurology
Neuro-Ophthalmology Unit
Emory Eye Center
Atlanta GA
USA
Eugene W M Ng MD
Eyetech Pharmaceuticals, Inc
New York NY
USA
Quan Dong Nguyen MD MSc
Diseases of the Retina and Vitreous, and
Uveitis
Johns Hopkins Hospital
Baltimore MD
USA
Jerry Y Niederkorn PhD
Center
Dallas TX
USA
Robert J Noecker MD
Vice Chair, Clinical Affairs
Eye and Ear Institute
Associate Professor
University of Pittsburgh School of Medicine
Pittsburgh PA
USA
Robert B Nussenblatt MD MPH
Scientic Director and Chief, Laboratory of
Immunology, Intramural Program
Section Head, Clinical Immunology Section
Bethesda MD
USA
Joan M OBrien MD
Professor of Ophthalmology and Pediatrics
Director of Ocular Oncology
Medical Center
San Francisco CA
USA
Paul D OBrien FRCSI MRCOphth MMedSci
Specialist Registrar in Ophthalmology
Royal Victoria Eye and Ear Hospital
Dublin
Ireland
Terrence P OBrien MD
Charlotte Breyer Rodgers Distinguished Chair
in Ophthalmology
Director of the Refractive Surgery Service
Palm Beach FL
USA
Denis ODay MD FACS
Vanderbilt Eye Institute
Nashville TN
USA
R Joseph Olk MD
Bond Eye Associates
Peoria IL
USA
Karl R Olsen MD
Clinical Assistant Professor of Ophthalmology
Retina Vitreous Consultants
Pittsbrugh PA
USA
Sumru Onal MD
Marmara University School of Medicine
Istanbul
Turkey
Yen Hoong Ooi MD

c/o Douglas Rhee MD
Boston MA
USA
E Mitchel Opremcak MD
Ohio State University College of Medicine
Columbus OH
USA
George Ousler BS
Director
Dry Eye Department
North Andover MA
USA
Randall R Ozment MD
Physician
Dublin Eye Associates
Dublin GA
USA
Samuel Packer MD
Professor of Clinical Ophthalmology,
New York University School of Medicine
Chair, Department of Ophthalmology
North Shore Long Island Jewish Health
System
New York NY
USA
Millicent L Palmer MD
Associate Professor, Department of Surgery
Creighton University Medical School
Creighton University Medical Center
Omaha NE
USA
George N Papaliodis MD
Instructor in Ophthalmology and Internal
Medicine
Boston MA
USA
D J John Park MD
Resident
Department of Plastics and Reconstructive
Surgery
Irvine CA
USA
David W Parke II MD
Edward L Gaylord Professor and Chairman
President and CEO
The Dean A McGee Eye Institute
Oklahoma City OK
USA
Cameron F Parsa MD
Krieger Childrens Eye Center
Baltimore MD
USA
M Andrew Parsons FRCPath
Honorary Consultant in Ophthalmic
Pathology
Academic Unit of Pathology
Royal Hallamshire Hospital
Shefeld
United Kingdom
Louis R Pasquale MD
Co-Director, Glaucoma Service
Boston MA
USA
Neha N Patel MD
Resident
Department of Ophthalmic and Visual
Science
Chicago IL
USA
Sayjal J Patel MD
Baltimore MD
USA
Thomas D Patrianakos DO
Attending Physician
John H Stroger Hospital of Cook County
Chicago IL
USA
James R Patrinely MD FACS
Plastic Eye Surgery Associates PLLC
Houston TX
USA
Deborah Pavan-Langston MD FACS
Surgeon and Director of Clinical Virology
Harvard School of Medicine
Boston MA
USA
Eli Peli MSc OD
Moakley Scholar in Aging Eye Research
Boston MA
USA
Susan M Pepin MD
Assistant Professor of Surgery
Section of Ophthalmology
Dartmouth Hitchcock Medical Center
Lebanon NH
USA
Victor L Perez MD
Assistant Professor
University of Miami School of Medicine
Miami FL
USA
Juan J Prez-Santonja MD PhD
Instituto Oftalmolgico de Alicante
Alicante
Spain
John R Perfect MD
Director, Duke University Mycology Research
Unit (DUMRU)
Division of Infectious Diseases
Duke University
Winston-Salem NC
USA
xxxvii
Henry D Perry MD FACS
Founding Partner
Director: Cornea Division
Rockville Center NY
USA
Joram Piatigorsky PhD
Chief
Laboratory of Molecular and Developmental
Biology
National Eye Institute - National Institute of
Health
Bethesda MD
USA
Dante Pieramici MD
Co-Director
California Retina Consultants
Santa Barbara CA
USA
Eric A Pierce MD PhD
F.M. Kirby Center for Molecular
Ophthalmology
Scheie Eye Institute
University of Pennsylvania School of Medicine
Philadelphia PA
USA
Roberto Pineda II MD
Assistant Professor
Chief of Ophthalmology, Brigham & Womens
Hospital, Boston
Assistant Professor, Department of
Boston MA
USA
Misha L Pless MD
Director, Division of General Neurology
Boston MA
USA
Howard D Pomeranz MD PhD
System
Great Neck NY
USA
Constantin J Pournaras MD
Geneva University Hospitals
Geneva
Switzerland
William Power MBBCH FRCS FRCOphth
Consultant Ophthalmic Surgeon
Blackrock Clinic
Blackrock
Co Dublin
Ireland
Manvi Prakash MD
Postdoctoral Fellow
Boston MA
USA
xxxviii
Anita G Prasad MD
Sciences
Washington University Medical School
St Louis MO
USA
Valerie Purvin MD
Clinical Professor of Ophthalmology &
Neurology
Neurology
Indiana Medical Center
Indianapolis IN
USA
David A Quillen MD
George and Barbara Blankenship Professor
and Chair
Penn State College of Medicine
Hershey PA
USA
Graham E Quinn MD
Attending Surgeon, Research Fellow
The Childrens Hospital of Philadelphia
Philadelphia PA
USA
Melvin D Rabena BSc
California Retina Consultants
Santa Barbara CA
USA
James L Rae PhD
Professor of Ophthalmology and Physiology
Physiology and Biomedical Engineering
Mayo Clinic
Rochester MN
USA
Michael B Raizman MD
Ophthalmic Consultant
Ophthalmic Consultants Of Boston
Tafts University School of Medicine
Boston MA
USA
Alessandro Randazzo MD
Istituto Clinico Humanitas Rozzano
Milano University
Milan
Italy
Narsing A Rao MD
Doheny Eye Institute
Los Angeles CA
USA
Christopher J Rapuano MD
Co-Director Cornea Service
Co-Director Professor of Ophthalmology,
Jefferson Medical College
Co-Director, Cornea Service
Refractive Surgery Department
Wills Eye Hospital
Philadelphia PA
USA
Sherman W Reeves MD MPH
Cornea, External Disease and Retractive
Surgery
Minnesota Eye Consultants
Minneapolis MN
USA
Carl D Regillo MD FACS
Wills Eye Hospital
Philadelphia PA
USA
Elias Reichel MD
Vitreoretinal Diseases
Boston MA
USA
Martin H Reinke MD
Private Practice
Southlake TX
USA
Douglas Rhee MD
Boston MA
USA
Claudia U Richter MD
Boston MA
USA
Joseph F Rizzo lll MD
Boston MA
USA
Richard M Robb MD
Childrens Hospital Boston
Boston MA
USA
Anja C Roden MD
c/o Diva R Salomao MD
Mayo Clinic
Rochester MN
USA
I Rand Rodgers MD
Assistant Clinical Professor, Mount Sinai
Medical Center
Director of Ophthalmic Facial and Plastic
Surgery
North Shore University Hospital NYU
Private Practice
New York NY
USA
Merlyn M Rodrigues MD PhD
c/o Kelly S Keefe MD
Naval Medical Center
San Diego CA
USA
Yonina Ron MD
Rabin Medical Center
Beilinson Campus
Petah Tiqva
Israel
Geoffrey E Rose DSC MS MRCP FRCS
FRCOphth

Adnexal Department
London
United Kingdom
Emanuel S Rosen MD FRCS FRCOphth
Manchester Central Health Care Authority
Manchester
United Kingdom
James T Rosenbaum MD
Professor of Medicine, Ophthalmology and
Cell Biology
Chief, Division of Arthritis and Rheumatic
Diseases
Director, Uveitis Clinic Casey Eye Institute
Oregon Health and Science University
Portland OR
USA
Perry Rosenthal MD
Boston Foundation for Sight
Boston MA
USA
Strutha C Rouse II MD
Horizon Eye Care
Charlotte NC
USA
Mark S Ruttum MD
Head, Pediatric Ophthalmology and Adult
Strabismus Section
Milwaukee WI
USA
Michael A Sandberg PhD

Berman-Gund Laboratory
Boston MA
USA
Allan R Rutzen MD FACS

University of Maryland
Baltimore MD
USA
Virender S Sangwan MD
Head, Cornea and Anterior Segment Services
L V Prasad Eye Institute
Hyderabad
India
Edward T Ryan MD
Director, Tropical & Geographic Medicine
Center
Associate Professor of Medicine
Assistant Professor
Dept of Immunology and Infectious Diseases
Harvard School of Public Health
Boston MA
USA
Barry W Rovner MD
Professor & Medical Director
Department of Psychiatry and Human
Behavior
Philadelphia PA
USA
Alfredo A Sadun MD PhD

Thornton Professor of Ophthalmology and
Neurosurgery
Doheny Eye Institute
Kech School of Medicine
Los Angeles CA
USA
Malgorzata Rozanowska PhD

Lecturer
School of Optometry and Vision Sciences
Cardiff University
Cardiff
United Kingdom
Jos-Alain Sahel MD
Head, Laboratory of Retinal Pathobiology
University Louis Pasteur
Strasbourg
France
Michael P Rubin MD
Fellow in Vitreoretinal Diseases and Surgery
Massachusetts Eye and Ear Inrmary,
Boston MA
USA
Leorey Saligan MD
Nurse Practitioner
Bethesda MD
USA
Peter A D Rubin MD FACS

Eye Plastics Consultant
Brookline MA
USA
Sarwat Salim MD FACS

Yale Eye Center
New Haven CT
USA
Shimon Rumelt MD
Attending Physician
Western Galilee - Nahariya Medical Center
Nahariya
Israel
John F Salmon MD FRCS FRCOphth

The Radcliffe Inrmary
Oxford Eye Hospital
Oxford
United Kingdom
Anil K Rustgi MD
Professor of Medicine and Genetics
Chief of Gastroenterology
University of Pennsylvania Medical Center
Philadelphia PA
USA
Diva R Salomo MD
Associate Professor of Pathology
Mayo Clinic
Rochester MN
USA
Tina Rutar MD
Resident
San Francisco CA
USA
David Sami MD
Division Chief for PSF Ophthalmology
CHOC Childrens Hospital
Orange CA
USA
Maria A Saornil MD
Ocular Pathology Unit
Hospital Clinico Universitario
Valladolid
Spain
Joseph W Sassani MD
Pennsylvania State University
Hershey Medical Center
Hershey PA
USA
Rony R Sayegh MD
Research Fellow
Cornea and Refractive Surgery Service
Boston MA
USA
Andrew P Schachat MD
Vice Chairman for Clinical Affairs
Cole Eye Institute
Cleveland Clinic Foundation
Cleveland OH
USA
Wiley A Schell MD
Director, Medical Mycology Research Center
Assistant Professor of Medicine Department
of Medicine
Division of Infectious Diseases and
International Health
Duke University Medical Center
Durham NC
USA
Amy C Schefler MD
Resident in Ophthalmology
Miami FL
USA
Tina Scheufele MD
Vitreoretinal Surgeon
Boston MA
USA
Vivian Schiedler MD
Oculoplastic and Orbital Surgeon,
Charlottesville, VA
Ophthalmic Plastic & Reconstructive Surgery
Fellow
Seattle WA
USA
Gretchen Schneider MD
Adjunct Assistant Professor in the Genetic
Counseling program
Genetic Counseling Faculty
Brandeis University
Waltham MA
USA
xxxix
Alison Schroeder BA
Laboratory Manager
Boston MA
USA
Irina Serbanescu BA
Research
Division of neurology
The Hospital for Sick Children
Toronto ON
Canada
Bradford J Shingleton MD
Assistant Clinical Professor of
Boston MA
USA
Ronald A Schuchard PhD

Director of Rehabilitation Research and
Development Center
Associate Professor
Emory University School of Medicine
Atlanta GA
USA
Briar Sexton MD FRCSC

Fellow in Neuro-Ophthalmology
VGH Eye Care Center
Vancouver BC
Canada
John W Shore MD FACS

Texas Oculoplastics Consultants
Austin TX
USA
Joel S Schuman MD
Eye and Ear Foundation Professor and
Chairman
Pittsburgh PA
USA
Ivan R Schwab MD FACS
University of California at Davis
Sacramento CA
USA
Adrienne Scott MD
Clinical Associate
Vitreoretinal Surgery
Durham NC
USA
Ingrid U Scott MD MPH
Professor of Ophthalmology and Health
Evaluation Sciences
Penn State College of Medicine
Hershey PA
USA
Marvin L Sears MD
Professor and Chairman Emeritus
Science
New Haven CT
USA
Johanna M Seddon MD ScD
Director, Ophthalmic Epidemiology and
Genetics Service
Boston MA
USA
Theo Seiler MD PhD
Professor
Institut fr Refractive und Ophthalmochirurgie
(IROC)
Zrich
Switzerland
Robert P Selkin MD
Private Practice
Plano TX
USA
xl
Richard D Semba MD MA MPH

W Richard Green Professor of
Ophthalmology
Baltimore MD
USA
Tarek M Shaarawy MD
Chef
Clinique dophtalmologie
Secteur du Glaucome
Hpitaux Universitaires de Gnve
Gnve
Switzerland
Peter Shah BSc (Hons) MBChB FRCOphth
Consultant
City Hospital
Birmingham
United Kingdom
Aron Shapiro BS
Director
Anti-inflammatory/Anti-infectives Department
North Andover MA
USA
Savitri Sharma MD MAMS
Associate Director, Laboratory Services
L V Prasad Eye Institute
Bhubaneswar, Orissa
India
Jean Shein MD
Attending Physician
Crane Eye Care Hana Kukui Center
Lihue HI
USA
Debra J Shetlar MD
Staff Physician
Michael E DeBakey V A Medical Center
Houston TX
USA
M Bruce Shields MD
Science
Yale Eye Center
New Haven CT
USA
Carol L Shields MD
Professor of Ophthalmology, Thomas
Attending Surgeon and Associate Director
Wills Eye Hospital
Philadelphia PA
USA
Jerry A Shields MD
Professor of Ophthalmology, Thomas
Jefferson University
Director
Oncology Services
Wills Eye Hospital
Philadelphia PA
USA
Lesya M Shuba MD PhD

Assistant Professor
Sciences
Dalhousie University
Halifax NS
Canada
Guy J Ben Simon MD
Tel Hashomer
Israel
Richard J Simmons MD
Emeritus Ophthalmic Surgeon
Boston MA
USA
Michael Simpson
c/o David Miller MD
Jamaica Plain MA
USA
Arun D Singh MD
Director
Department of Ophthalmic Oncology
Cole Eye Institute and Taussing Cancer Center
Cleveland OH
USA
Omah S Singh MD
Director
Beverley MA
USA
Karen Sisley BSc PhD
Non-Clinical Lecturer Ocular Oncology
Academic Unit of Ophthalmology and
Orthoptics
University of Shefeld
Shefeld
United Kingdom
Arthur J Sit MD
Mayo Clinic
Rochester MN
USA
David Smerdon FRCSEd FRCOphth
James Cook University Hospital
Middlesbrough
United Kingdom
William E Smiddy MD
Miami FL
USA
Ronald E Smith MD
Professor and Chair
Keck School of Medicine of USC
Los Angeles CA
USA
Terry J Smith MD
Professor and Head
Division of Molecular Medicine
David Geffen School of Medicine
Harbor-UCLA Medical Center
Torrance CA
USA
Neal G Snebold MD
Ophthalmologist
Boston MA
USA
Lucia Sobrin MD
Retina and Uvetis Services
Boston MA
USA
John A Sorenson MD
Attenting Surgeon
Vitreoretinal Service
Manhattan Eye, Ear, and Throat Hospital
New York NY
USA
Sarkis H Soukiasian MD
Director: Cornea and External Disease
Director: Ocular Inflammation and Uveitis
Lahey Clinic
Burlington MA
USA
George L Spaeth MD FRCO FACS
Louis Esposito Research Professor of
Ophthalmology
Director of the William & Anna Goldberg
Glaucoma Service
Wills Eye Institute
Philadelphia PA
USA
Richard F Spaide MD
Associate Clinical Professor of Ophthalmology
Manhattan Eye, Ear, and Throat Hospital
New York NY
USA
Monika Srivastava MD
Department of Dermatology
New York University
New York NY
USA
Sunil K Srivastava MD
Section of Vitreoretinal Surgery & Disease
Emory Eye Center
Atlanta GA
USA
Alexandros N Stangos MD
Department of Clinical Neurosciences
University Hospitals of Geneva
Geneva
Switzerland
Tomy Starck MD
Director
UltraVision Center
San Antonio TX
USA
Walter J Stark MD
Director of the Stark-Mosher Center
The John Hopkins Hospital, Wilmer Eye
Institute
Baltimore MD
USA
Joshua D Stein MD MS
Assistant Professor
Sciences
Kellogg Eye Center
Ann Arbor MI
USA
Roger F Steinert MD
Professor of Ophthalmology and Biomedical
Engineering
Director of Cornea, Refractive and Cataract
Surgery
Vice Chair of Clinical Ophthalmology
University of California Irvine
Irvine CA
USA
Leon Strauss MD
Instructor
Baltimore MD
USA
Barbara W Streeten MD
State University of New York
Upstate Medical University
Syracuse NY
USA
J Wayne Streilein MD (deceased)
Formerly Senior Scientist, President, Charles
L Schepens Professor of Ophthalmology,
Professor of Dermatology
Formerly Vice Chair for Research,
Boston MA
USA
James D Strong CRA
Senior Ophthalmic Imager
Penn State Milton S Hershey Medical Center
Hershey PA
USA
Ilene K Sugino MS
Director, Ocular Cell Transplantation
Laboratory
Institute of Ophthalmology and Visual Science
New Jersey Medical School
Newark NJ
USA
Eric B Suhler MD MPH
Chief of Ophthalmology
Portland VA Medical Center
Assistant Professor of Ophthalmology and
Co-director
Casey Eye Institute
Portland OR
USA
Timothy J Sullivan FRANZCO FRACS

Eyelid, Lacrimal and Orbital Clinic
Royal Brisbane Hospital
Herston QLD
Australia
Jennifer K Sun MD
Lecturer
Boston MA
USA
Janet S Sunness MD
Medical Director
Richard E Hoover Rehabilitation Services for
Low Vision and Blindness
Greater Baltimore Medical Center
Baltimore MD
USA
Francis C Sutula MD
Milford Eye Care
Milford MA
USA
Nasreen A Syed MD
Assistant Professor, Ophthalmology and
Pathology
Sciences
University of Iowa
Iowa City IA
USA
Christopher N Ta MD
Stanford University
Palo Alto CA
USA
Hidehiro Takei MD
Staff Pathologist
The Methodist Hospital
Houston TX
USA
Jonathan H Talamo MD
Ophthalmology
Waltham MA
USA
Richard R Tamesis MD
Loma Linda University Medical Center
Loma Linda CA
USA
Madhura Tamhankar MD
Associate Professor
University of Pennsylvania Medical School
Philadelphia PA
USA
Kristen J Tarbet MD SACS
Private Practice
Bellevue WA
USA
xli
Michelle Tarver-Carr MD PhD
Assistant, Ocular Immunology
Departments of Medicine and Epidemiology
Baltimore MD
USA
Mark A Terry MD
Director, Corneal Services
Clinical Professor, Department of
Ophthalmology
Devers Eye Institute
Oregon Health Sciences University
Portland OR
USA
Joseph M Thomas MD
Case Western Reserve University School of
Medicine
Cleveland OH
USA
Vance Thompson MD
University of South Dakota School of
Medicine
Director of Refractive Surgery
Sioux Valley Clinic
Vance Thompson Vision
Sioux Falls SD
USA
Jennifer E Thorne MD PhD
Division of Ocular Immunology
Baltimore MD
USA
Matthew J Thurtell BSc(Med) MBBS MScMed
Neuro-Ophthalmology Fellow
Royal Prince Albert Hospital
Sydney NSW
Australia
David P Tingey MD FRCSC
Associate Professor
Ivey Eye Institute
London Health Sciences Center
London ON
Canada
King W To MD
Brown University School of Medicine
Barrington RI
USA
Faisal M Tobaigy MD
Massachusetts Eye and Ear Inrmary and the
Boston MA
USA
Michael J Tolentino MD
Director of Research, Center for Retina and
Macular Disease
Center for Retina and Macular Disease
Winter Haven FL
USA
xlii
Melissa G Tong BSc

Philadelphia PA
USA
Gail Torkildsen MD
Physician
Andover Eye Associates
Andover MA
USA
Cynthia A Toth MD
Biomedical Engineering
Duke Eye Center
Durham NC
USA
Elias I Traboulsi MD
The Cole Eye Institute
Cleveland OH
USA
Michele Trucksis PhD MD
Associate Director Clinical Pharmacology
Merck & Co. Inc
Boston MA
USA
James C Tsai MD
Robert R Young Professor and Chairman
Science
New Haven CT
USA
Julie H Tsai MD
Assistant Professor
University of South Carolina School of
Medicine
Columbia SC
USA
David T Tse MD FACS
Miami FL
USA
Scheffer C G Tseng MD PhD
Research Director
Ocular Surface Center
Miami FL
USA
Elmer Y Tu MD
Ophthalmology
Director of the Cornea and External Disease
Service
Chicago IL
USA
Ira J Udell MD
Albert Einstein College of Medicine
New York NY
USA
Alejandra A Valenzuela MD
Assistant Professor
Sciences
Dalhousie University
Halifax NS
Canada
Russell N Van Gelder MD PhD

Visual Sciences
Sciences
St Louis MO
USA
Gregory P Van Stavern MD
Assitant Professor of Ophthalmology,
Neurology and Nerosurgery
Kresge Eye Institute
Wayne State University
Detroit MI
USA
Deborah K Vander Veen MD
Assistant Professor
Childrens Hospital and Harvard Medical
School
Boston MA
USA
Demetrios Vavvas MD PhD
Instructor in Ophthalmology
Retina Service Department of Ophthalmology
Boston MA
USA
David H Verity MA FRC Ophth
Adnexal Departments
London
United Kingdom
Paolo Vinciguerra MD
Medical Director
Studio Oculistico Vincieye SRL
Milan
Italy
Paul F Vinger MD
Clinical Professor
Ophthalmology
New England Medical Center
Boston MA
USA
Nicholas J Volpe MD
Professor of Ophthalmology and Neurology
Vice Chair and Residency Program Director
PENN Eye Care
Philadelphia PA
USA
Werner Wackernagel MD
Physician
Medical University Graz
Graz
Austria
Sonal Desai Wadhwa MD
University of Maryland
Baltimore MD
USA
Michael D Wagoner MD
Sciences
University of Iowa Hospitals and Clinics
Iowa City IA
USA
Nadia K Waheed MD
Fellow
Immunology and Uveitis Service
Boston MA
USA
David S Walton MD
Boston MA
USA
Martin Wand MD
University of Connecticut School of Medicine
Farmington CT
USA
Jie Jin Wang MMed PhD
Associate Professor of Epidemiology
Westmead Millennium Institute
University of Sydney
Sydney NSW
Australia
Scott M Warden MD
Retina Service
Boston MA
USA
Lennox Webb FRCOphth FRCS(Ed)
Royal Alexandra Hospital
Paisley
United Kingdom
David Weber MD
Assistant Professor
Department of Physical Medicine &
Rehabilitation
Mayo Clinic College of Medicine
Rochester MN
USA
Daniel Wee MD
The Palmetto Health/ University of South
Carolina School of Medicine
Columbia SC
USA
Corey B Westerfeld MD
Research Fellow
Boston MA
USA
Christopher T Westfall MD
Jones Eye Institute & Arkansas Childrens
Hospital
University of Arkansas for Medical Sciences
Little Rock AR
USA
Scott M Whitcup MD
Executive Vice President
Head of Research and Development
Allegran Inc
Irvine CA
USA
Valerie A White MD FRCPC
Professor
Department of Pathology & Laboratory
Medicine,
University of British Columbia
Vancouver General Hospital
Vancouver BC
Canada
William L White MD
The Eye Foundation
University of Missouri-Kansas City
Kansas City MO
USA
Jason Wickens MD
Barnes Retina Institute
St Louis MO
USA
Janey L Wiggs MD PhD
Boston MA
USA
Jacob T Wilensky MD
Director, Glaucoma Service
University of Illinois College of Medicine
Chicago IL
USA
Charles P Wilkinson MD
Greater Baltimore Medical Center
Professor, Department of Ophthalmology
John Hopkins University
Baltimore MD
USA
Patrick D Williams MD
Vitreo Retinal Specialist
Texas Retina Associates
Arlington TX
USA
David J Wilson MD
Associate Professor
Department of Ophthalmology;
Director, Christensen Eye Pathology
Laboratory
Casey Eye Institute
Oregon Health Sciences University
Portland OR
USA
M Roy Wilson MD MS
Chancellor
University of Colorado and Health Sciences
Center
Denver CO
USA
Steven E Wilson MD
Director of Corneal Research and Professor
of Ophthalmology
The Cleveland Clinic Foundation
Cole Eye Institute
Cleveland OH
USA
Jules Winokur MD
System
New York NY
USA
William J Wirostko MD
The Eye Institute
Medical College of Milwaukee
Milwaukee WI
USA
Gadi Wollstein MD
Assistant Professor and Director
Ophthalmic Imaging Research Laboratories
The Eye & Ear Institute
Dept of Ophthalmology
UPMC Eye Center
Pittsburgh PA
USA
Albert Chak Ming Wong FCOph(HK)
FHKAM(Ophth)
Associate Consultant
Caritas Medical Center
Shamshuipo, Kowloon
Hong King
China
Tien Y Wong MBBS MMED (Ophth) FRCSE
FRANZCO FAFPHM MPH PhD
Department of Ophthalmology & Centre for
Eye Research Australia
University of Melbourne
East Melbourne VIC
Australia
John J Woog MD FACS
Associate Professor of Ophthalmology,
Ophthalmic Plastic and Reconstructive
Surgery
Mayo Clinic
Rochester MN
USA
Michael Wride PhD
Lecturer
School of Optemetry and Vision Sciences
Cardiff University
Cardiff
United Kingdom
Carolyn S Wu MD
Boston MA
USA
Darrell WuDunn MD PhD
Indianapolis IN
USA
Jean Yang MD
North Shore-Long Island Jewish Medical
Center
Great Neck NY
USA
Lawrence A Yannuzzi MD
Vice-Chairman, Department of
Ophthalmology
Director of Retinal Services
New York NY
USA
xliii
Michael J Yaremchuk MD
Clinical Professor of Surgery
Boston MA
USA
R Patrick Yeatts MD FACS
Professor and Vice Chairman
Winston-Salem NC
USA
Richard W Yee MD
Medical Director LADARVISION Center
Hermann Eye Center
Memorial Hermann Hospital
Houston TX
USA
Steven Yeh MD
Clinical Fellow
Uveitis and Ocular Immunology
National Institute of Health
Bethesda MD
USA
Lucy H Y Young MD PhD FACS
Associate Professor
Boston MA
USA
xliv
Jenny Y Yu MD
Consulting Physician
UPMC Childrens Hospital of Pittsburgh
Pittsburgh PA
USA
Beatrice Y J T Yue PhD
Thanis A Field Professor of Ophthamology
Sciences
Chicago IL
USA
Charles M Zacks MD
Corneal Specialist
Maine Eye Center
Portland ME
USA
Bruce M Zagelbaum MD FACS
Ophthalmology
New York University School of Medicine
New York NY
USA
Maryam Zamani MD
Oculoplastic Fellow
London
United Kingdom
Marco Zarbin MD PhD FACS

Professor of Ophthalmology and
Neuroscience
Science
University of Medicine and Dentistry,
New Jersey
Newark NJ
USA
Leonidas Zografos MD
Jules Gonin Eye Hospital
Lausanne
Switzerland
Christopher I Zoumalan MD
Resident in Ophthalmology
Stanford University Medical Center
Stanford CA
USA
SECTION 1
GENETICS
Edited by Janey L. Wiggs and Thaddeus P. Dryja
CHAPTER
Fundamentals of Genetics
Thaddeus P. Dryja
A GENE IS DEFINED BY A PHENOTYPE

Genes are the fundamental units used in the study of inherited
traits or diseases. A gene is classically dened by the phenotype
that is associated with it. For example, the gene causing
choroideremia is the choroideremia gene, and the gene causing
retinoblastoma is the retinoblastoma gene. However, in more
recent years, many genes have been dened on the basis of the
encoded protein product, irrespective of any phenotypes known
to be associated with variations or mutations. For instance, a
gene on chromosome 3 is named the rhodopsin gene because
it encodes rhodopsin. Years after the isolation and characterization of the rhodopsin gene, it was discovered that mutations at
this gene can cause retinitis pigmentosa or stationary night
blindness. Rather than renaming the locus as the retinitis
pigmentosa gene or otherwise, this gene retains its name as the
rhodopsin gene.
The term gene is actually somewhat ambiguous, because it
can refer to the position on a chromosome (a locus) that governs
a heritable trait or to a form of the DNA sequence at the locus
(an allele) that is associated with a particular phenotype. Therefore, in common usage, one might state that a variation in iris
color is due to a gene, and it is also correct to state that a
brown-eyed person has the gene for a brown iris. In the rst
case, one is stating that a genetic locus has alleles that specify
iris color, and in the second case, one is referring to a particular
allele at the iris color locus. To be more specic and unambiguous, one should state that a genetic locus controls iris color
and that an individual with brown eyes carries a brown allele
at that locus. The distinction is important, especially when one
counsels a family with a hereditary disease such as retinoblastoma. The family may speak of the affected child as having
the retinoblastoma gene. They will be surprised to learn from
the ophthalmologist that all family members have the retinoblastoma gene, but that some relatives have normal versions of
the gene that do not predispose to the cancer. Only those relatives with a mutant version have a high risk of being affected.
Despite the ambiguities, the different uses for the word gene
are so ingrained that any attempt to change them is futile.
LINEAR POLYMERS OF DNA ARE THE

CHEMICAL BASES FOR GENES
The chemical material that contains genetic information is
DNA. This is a linear polymer with two complementary
strands. Each strand is made up of a linear array of purine bases,
guanine (G) and adenine (A), and pyrimidine bases, cytosine (C)
and thymine (T). Each base is linked covalently to a pentose;
the combination is called a nucleoside. A single strand of
DNA has a series of the four bases coupled through these

carbohydrate moieties by phosphate bonds. The genetic
information is contained in the specic sequence of the four
bases in the 5 to 3 direction, where the 5 and 3 designations
refer to the sites on the pentose moieties where phosphate
bonds are linked. This strand is called the sense strand. The
complementary strand, or antisense strand, runs in the opposite direction and invariably has nucleotides complementary to
those in the sense strand as illustrated in Figure 1.1.
DNARNAPROTEIN
A gene is determined by the particular order of bases within a
specied region (locus) in a molecule of DNA. Each gene codes
for one protein. RNA is the chemical intermediate that conveys the base sequence in DNA to the protein-synthesizing
machinery (ribosomes) in the cytoplasm of a cell. RNA is composed of the same purine and pyrimidine bases as DNA, except
that the pyrimidine base thymine (T) present in DNA is instead
uracil (U) in RNA. Another difference is that the pentose linked
to each base is ribose rather than deoxyribose. The RNA molecules that transmit the DNA base sequence to the cytoplasm
of a cell are called messenger RNA molecules, or mRNA. The
synthesis of mRNA molecules from a DNA template is called
transcription. The synthesis of strands of amino acids based on
the sequence of bases in mRNA is called translation.
ORGANIZATION OF A EUKARYOTIC GENE

Eukaryotic genes, including human genes, are transcriptional
units; that is, each gene is organized for the synthesis of a
distinct mRNA sequence that codes for a distinct protein.
Transcriptional units are organized in the following manner
(Fig. 1.2). At the 5 end is a region extending a few hundred
bases called the promoter region. This region has sequences
recognized by factors (typically proteins) that control the
expression of the gene, as well as one or more binding sites for
RNA polymerase. Besides the promoter region, other regions
within a gene or at some distance from it can also have roles in
determining the proper tissue-specic expression of a gene at
the proper time during the life of the organism.1
Downstream of the promoter region is the transcription
start site, which is a specic base at which the enzyme RNA
polymerase initiates the synthesis of an RNA copy of the
DNA sequence. The sequence of bases in the transcribed RNA
molecule will be identical to the sequence in the sense strand of
DNA, except that the base uracil (U) will be used instead of
thymine (T), as noted earlier. Next comes the 5 untranslated
region, or the region of sequence that is included in the RNA
SECTION 1
GENETICS
FIGURE 1.1. Chemical structure of DNA.

(a) Two hydrogen bonds (dotted lines) couple
the bases thymine and adenine, and three
hydrogen bonds couple guanine and cytosine.
(b) The double-helical structure of the linear
DNA strands.
CHAPTER 1
FIGURE 1.2. Functional organization of a transcriptional unit. The organization of the human blue cone opsin gene, which consists of ~4000 bp
of DNA within human chromosome 7, is shown.47 Top, Schematic representation of the position of each of the ve exons. The letters (a) through
(d) indicate the four regions illustrated in more detail below, where the DNA sequence (sense strand only) at each of the four positions is shown.
(a) The 5 end of the gene. The TATA box is the sequence TATAA, which is an important recognition sequence for the binding of a factor that
allows RNA polymerase to initiate transcription. The transcription start site is the point at which an RNA copy of the DNA sequence is begun.
The RNA sequence differs from the DNA sequence only in that a U (uridine) is used instead of a T (thymine). The rst segment of transcribed
DNA is the 5 untranslated region. Translation begins with the sequence AUG, which is called the initiation codon or the start codon. It species
methionine, which will be at the amino terminus (N) of the resultant amino acid sequence. (b) Intron 1. The rst intron begins with the
dinucleotide sequence GT and ends with the sequence AG. These dinucleotide sequences are almost invariably present at the ends of introns
and are called the splice donor and splice acceptor sites, respectively. Notice that a codon is split by the intron. This is neither the rule nor the
exception. (c) Termination of translation. In the last exon (exon 5) a stop codon occurs in this case the sequence TGA. Although transcription
of RNA continues beyond this codon, the remaining RNA sequence is not translated into an amino acid sequence and therefore is called the 3
untranslated region. (d) Polyadenylation. The polyadenylation signal sequence, ATTAAA, is recognized by factors that cause the termination of
transcription 20 bases downstream. At the end of the RNA sequence, a large string of As is added. The nal RNA transcript, after the excision of
the four introns and the addition of the poly-A sequence, is called a messenger RNA, or mRNA. It is transported to the cytoplasm for translation
by the ribosomes.
SECTION 1
GENETICS
transcript but is not used to code for a protein. The coding
region begins with the initiation codon, which is always the
triplet of bases ATG coding for methionine. The succeeding
sequence of bases is called the coding region and is organized
into codons or triplets of bases that specify the amino acids
of the encoded protein. The coding region ends with a stop
codon (either TGA, TAG, or TAA), which is followed by the 3
untranslated region. Finally, a polyadenylation signal sequence
registers the end of transcription by RNA polymerase.
A noteworthy feature of eukaryotic genes, but not prokaryotic
genes, is that the coding region in genomic DNA is generally
interrupted by one or more introns. After an RNA transcript is
produced from a gene, these intron sequences are excised. This
is one of the steps necessary to make mature messenger RNA
or mRNA. The term cDNA is given to any DNA fragment with
a sequence identical to that found in an mRNA molecule (i.e.,

a DNA sequence lacking intron sequences). cDNA molecules are
not normally produced in living cells; instead, they are produced
in research laboratories and are used as reagents helpful in
studying genes.
GENETIC CODE
The DNA sequence that species the sequence of amino acids
of a protein is in the form of a genetic code. In the cytoplasm
of cells, ribosomes translate the code (Fig. 1.3). Each set of three
consecutive nucleotides, called a codon, in the coding region of
an mRNA molecule species one amino acid.
Figure 1.4 shows the amino acid specied by each codon. The
codon ATG, which species the amino acid methionine, is
the only codon used by the ribosome to initiate translation.
Hence, all proteins are rst synthesized with the amino acid
methionine at their amino terminus. (This amino acid may be
subsequently removed as a posttranslational modication of
the protein.) Ribosomes recognize the correct ATG sequence
present near the 5 end of the mRNA for initiating translation;
other ATG codons nearby are customarily ignored through mechanisms that remain unclear. Downstream from the initiating
codon, every three bases specify one amino acid. There is no
skipping or overlapping of codons. This process continues until
one of the codons TAG, TGA, or TAA is encountered in the
same frame as the initiating codon. These three codons are
called stop or termination codons, because any one of them
serves to terminate the translation of an mRNA molecule.
HOW GENES ARE ORGANIZED IN HUMAN

CELLS
FIGURE 1.3. Translation of mRNA. A ribosome is depicted

schematically in the process of synthesizing a molecule of blue cone
opsin.
DNA molecules that carry genetic information are packaged

into chromosomes. A chromosome is thought to be composed
of a single long DNA molecule and numerous associated
proteins and perhaps other substances. The complex of DNA
and associated materials in chromosomes is called chromatin.
FIGURE 1.4. The genetic code. This wheel

gives the amino acid specied by any threebase codon. The codon is read from the center
to the periphery of the wheel. Amino acids are
abbreviated using the standard three-letter
code. At the bottom of the gure is the oneletter code, the three-letter code, and the full
name of each amino acid.
Adapted from Ausubel FM, Brent R, Kingston RE, et
al: Current protocols in molecular biology. New York:
Wiley; 1991.
FIGURE 1.5. A normal human karyotype.

Below the 22 pairs of autosomes are the sex
chromosomes. Since both X and Y
chromosomes are present, this karyotype is
from a male.
CHAPTER 1
Courtesy of Cynthia Morton, PhD.
HUMAN CHROMOSOMES
Each nucleus of a human cell has 23 pairs of chromosomes
(Fig. 1.5), corresponding to 46 molecules of DNA. The two
chromosomes in each pair typically have an identical appearance and have the same complement of genetic loci in the same
order. They are distinguished because they can carry different
alleles at each locus. Each member of a pair of chromosomes is
derived from a different parent. Of the 23 pairs of chromosomes, 22 are called autosomes; the remaining pair embodies
the sex chromosomes. The 22 autosomes are numbered
according to their size, with chromosome 1 being the largest
chromosome, chromosome 2 the next in size, and so forth. The
only exception to this rule involves chromosomes 21 and 22,
because chromosome 21, not 22, is the smallest. The sex
chromosomes are not named by numbers but instead are called
the X and Y chromosomes.
Each chromosome has a centromere that divides it into two
arms, the short arm and the long arm (Fig. 1.6). The short arm
and long arm are called the p arm and the q arm, respectively.
The proximal portion of a chromosome arm is the region close
to the centromere; the distal portion is far from the centromere.
A chromosome with a very small short arm is called an
acrocentric chromosome. Acrocentric human chromosomes are
numbers 13, 14, 15, 21, and 22. The short arms of acrocentric
chromosomes contain multiple copies of the genes coding for
ribosomal RNA rather than for proteins.
Until the early 1970s, chromosomes could only be distinguished on the basis of their overall size and the relative size
of their short and long arms. Because of this, many human
chromosomes could not be uniquely distinguished, and
chromosomes of similar morphology were lumped into groups
(e.g., the A group, B group). As an example, the D group
included chromosomes 13, 14, and 15; all of these are acro-
centric chromosomes of approximately the same size. A patient

with a deletion of any of those three chromosomes was diagnosed as having a D-deletion. A few cases of retinoblastoma
with a deletion of a D group chromosome were reported in
the 1960s, and this association was called D-deletion
retinoblastoma.2,3
Improved chromosome banding techniques, using dyes such
as quinacrine or Giemsa, became widely used by the early
1970s. A pattern of staining that is unique to each chromosome
arm allowed the recognition of every human chromosome.
There is now a standardized nomenclature for the set of darkly
and lightly staining bands characteristic of each human chromosome arm. To continue the example of D-deletion retinoblastoma, after the new karyotyping techniques were developed,
it was discovered that in all cases of D-deletion retinoblastoma, the deleted chromosome was always chromosome
13, hence the name of the association was changed to 13deletion retinoblastoma. Furthermore, in every case, the deletion
included the band 14 on the long arm of the chromosome,4 so
that the term 13q14 deletion or 13q14 is more precise.
Another important deletion associated with ophthalmologic and
systemic abnormalities involves chromosome 11p13; deletions
of this chromosomal segment cause a syndrome including
aniridia and elevated predisposition to Wilms tumor.5
SIZE OF THE HUMAN GENOME

A set consisting of one of each autosome as well as both sex
chromosomes is called a human genome. It includes one copy
of every human locus. The chromosomal molecules of DNA
from one human genome, if tandemly arranged end to end,
contain a sequence of ~3.2 billion bp. The amount of information contained within 3.2 billion bp can be instructively related
SECTION 1
GENETICS
of a chromosome. In such a situation, the abnormality is called
trisomy for the chromosome involved. For example, patients
with Downs syndrome have three copies of chromosome 21,
also referred to as trisomy 21. Much the same phenotype can
also result from trisomy of only the long arm of chromosome
21, or trisomy 21q.
If one copy of a pair of chromosomes is absent, the defect
is called haploidy or deletion. Haploidy for an entire human
chromosome is probably lethal, but individuals do exist who
have a deletion of a segment of a chromosome.
TRANSLOCATIONS
Occasionally, a hybrid chromosome will be observed in the
karyotype of an individual, with a mixture of material derived
from two separate chromosomes. As a hypothetical example, a
part of chromosome 1q might be fused to 3p. Depending on the
number of normal chromosomes 1 and 3, an individual who
carries a translocation (1q;3p) could be trisomic or monosomic
for these chromosome arms. A translocation is balanced if
there is a diploid amount of each chromosome band.
SISTER CHROMATIDS
FIGURE 1.6. Anatomy of a chromosome, in this case human

chromosome 7.
to the quantity of information stored on modern desktop computers. At each position in DNA there is one of four possible
bases (A, T, G, and C), which is equivalent to two bits of
computer code. Since there are eight bits in a byte of computer
memory, each byte could store the equivalent of four bases of
DNA sequence. The DNA sequence of the human genome
would occupy ~800 MB. The sequence could be stored on a
1-GB hard drive (small by todays standards) with plenty of
room to spare. Obtaining the complete sequence of the human
genome within the rst decade of the twenty-rst century was
one of the initial goals of the Human Genome Project. The rst
draft of the complete human genome sequence was obtained
in 2001.6
In terms of the physical size of the human genome, the corresponding DNA would be 1 m long but only 2 nm in diameter.
The total volume of a human genome, assuming the DNA is a
cylinder, is about one hundred millionth of a microliter. Current
estimates are that there are 60 000100 000 genes embedded in
this DNA sequence. On an average, there is one gene about
every 30 000 bp.
HAPLOIDY, DIPLOIDY, TRIPLOIDY
A set consisting of one of each autosome as well as an X or a Y

chromosome is called a haploid set of chromosomes. The
normal complement of two copies of each gene (or two copies of
each chromosome) is called diploidy. In unusual circumstances,
a cell or organism may have three copies of each chromosome;
this is called triploidy. A triploid human is not viable; however,
some patients have an extra chromosome or an extra segment
Just before a cell divides, each chromosome arm is duplicated,

so that chromosomes have two identical short arms and two
identical long arms (see Fig. 1.6). At this point, there are four
copies of each gene in a cell. Each chromosome has two short
arms and two long arms, and each arm is called a chromatid. A
pair of similar arms from the same chromosome is called a pair
of sister chromatids. When one examines the karyotype of a
cell, the chromosomes are observed just before the cell divides.
Consequently, each chromosome has two sister chromatids corresponding to the short arm and two sister chromatids corresponding to the long arm. Sister chromatids always share the
same alleles, whereas the two chromosome homologs in a human
cell (one derived from each parent), can have different alleles at
any locus.
ALLELES ARE VARIATIONS IN THE

NUCLEOTIDE SEQUENCE
An allele is a specic nucleotide sequence at a locus that is
associated with an observable phenotype. The most common
allele at a locus is called the wild-type allele, often abbreviated
+ or wt. An allele that is different from the wild type is
customarily given an abbreviated name that is somehow related
to the phenotype or the nucleotide sequence. For example, an
allele in the rhodopsin gene causing autosomal dominant retinitis
pigmentosa could be labeled RhoPro23His or rhodopsin, Pro23His,
where Pro23His indicates that codon 23, which species proline
in the wild-type allele, species histidine in the mutant allele.7
Although a genetic locus usually corresponds with a transcriptional unit, the boundaries of a locus in a DNA sequence
are often not very precise. One reason for this is that DNA
sequences many thousands of bases from the transcriptional
unit can be important for the proper expression of a gene at the
correct time during the development of a specic cell type.1 It
is conceivable that a mutation in such distant sequences can
change the expression of a transcriptional unit and produce a
phenotype associated with the locus. Hence, it is a simplication to state that alleles are the result of variations in the
nucleotide sequence inside a transcriptional unit. In practice,
however, this is usually the case.
If an allele has a frequency of 12% or higher and is not
associated with a disease, it is called a polymorphism. Since
humans have two alleles at each locus, the arbitrary criterion of
a 1% allele frequency corresponds with a polymorphism for

which ~2% of unrelated individuals are carriers. An example is
the still unidentied locus on chromosome 19, where a polymorphism species the presence or absence of green iris color.8
If an allele occurs with a frequency less than 1%, it is a rare
variant. If an allele causes disease, it is customarily called a
mutation. Most mutations are rare variants. However, at least
one is at a frequency high enough to be considered a polymorphism: ~2% of whites carry the Phe508del mutation that
causes cystic brosis.9
Genetic diseases are dened clinically before the underlying
causative gene defects are known. Most clinically dened
hereditary diseases turn out to be genetically heterogeneous.
Allelic heterogeneity is the term used when different mutant
alleles at the same locus can produce the same disease. For
example, numerous mutations in the Rab escort protein gene
have been found to produce choroideremia.10 Nonallelic heterogeneity refers to the situation when mutations in different genes
can produce the same clinically dened disease. An example of
nonallelic heterogeneity is retinitis pigmentosa, which can be
produced as a result of defects in any of dozens of different
genes.11 Gene sharing occurs if different mutations in the same
gene can produce different phenotypes. For instance, defects in
the Norrie disease gene can produce either Norrie disease,
exudative vitreoretinopathy, or predisposition to retinopathy
of prematurity.1214 Another example of two diseases sharing
the same genes is retinitis pigmentosa and congenital
stationary night blindness. Different defects in the rhodopsin
gene can produce these two diseases;7,15,16 so too can different
defects in the gene encoding the b subunit of rod cGMPphosphodiesterase.17,18
Key Features: Fundamentals of Genetics
Genes are dened by phenotypes and are chemically

composed of DNA.
In cells DNA is packaged into chromosomes, and a genome is
a complete set of chromosomes. The human genome contains
two copies each of 22 autosomes and two sex chromosomes.
Alleles are variations in DNA sequence at genetic loci.
Human disorders can be inherited as dominant, recessive,
X-linked, mitochondrial (also called maternal), digenic, and
polygenic traits.
DNA sequence variations among human populations have
made it possible to develop a map of the human genome.
Mutations are changes in DNA sequence that have biological
consequences.
daughter cell receives one member of each homologous pair.

The daughter cells are therefore haploid. They, nevertheless,
have two of each chromatid. The chromosomes separate during
the second meiotic division to produce haploid germ cells with
only one of each chromatid.
CHAPTER 1
RECOMBINATION
In somatic cells, it is the general rule that each chromosome
homolog has a set of alleles derived from one parent. After
meiosis, a germ cell is haploid; that is, it has only one member
of each pair of chromosomes. Hence, a germ cell could have the
maternally derived chromosomes 1, 2, 4, 7, and so on, and the
paternally derived chromosome 3, 5, 6, 8, and so forth.
This mixing of chromosomes is one source of the diversity
that is provided by sexual reproduction. However, it is only half
of the story. During the rst meiotic division, chromatids from
homologous chromosomes can recombine or crossover (Fig.
1.7). During this process, the chromatids exchange linear sets
of alleles so that the daughter chromosomes have a mixture of
maternal and paternal alleles. This is the second major source
for new combinations of genes. The resultant germ cells receive
a random mixture of these hybrid chromosomes.
Roughly 30 crossovers (also called recombination events)
occur during each meiosis. Crossovers can take place anywhere
along the length of a chromosome arm, although there appear
to be regions that are especially susceptible to it (called recombination hot spots). Also, there is a relatively greater likelihood
of a crossover happening in the distal portion of a chromosome
arm compared with the proximal portion. The rate of recombination occurring at any particular region of a chromosome
can be different in males and females.
During oogenesis, the two X chromosomes carried by a
female can recombine anywhere along their length just as with
autosomes. In contrast, the X and Y chromosomes of a male
usually do not recombine, and if they do, crossovers occur only
within the distal short arms.
Considering that during meiosis an average of 30 crossovers
occur among the 23 pairs of human chromosomes, most
chromosomes in germ cells are recombinant. Furthermore,
because there is also a random assortment of chromosomes
during meiosis, there is the potential for a huge number of
possible combinations of alleles. In effect, each gamete has a
unique, haploid set of alleles. An individual conceived as the
union between two such gametes is likewise unique.
HOMOZYGOTES AND HETEROZYGOTES

HEREDITARY TRANSMISSION OF GENETIC
INFORMATION
SOMATIC CELLS VERSUS GERM CELLS
Most of the cells in the human body are somatic cells. Somatic
cells have a diploid set of chromosomes (i.e., two copies of
each autosome, one derived from each parent) and two sex
chromosomes (either XX or XY). Somatic cells are produced as
a consequence of mitosis or cell division (Fig. 1.7). Before a cell
divides into two daughter cells, the entire complement of
chromosomes duplicates so that the cell has four copies of every
autosomal gene. Each daughter cell receives a complete, diploid
set of chromosomes with solitary short and long arms.
The second category of human cells involves those in the
germ line; that is, cells whose descendants are germ cells
(sperm and ova). Germ cells are haploid. The process that
creates germ cells is called meiosis. Meiosis encompasses two
cell divisions (Fig. 1.7). In the rst meiotic division, each
Since an individual has two copies of each autosome, he or she

will have two copies of each autosomal locus. One copy is
derived from the mother and one from the father. How similar
are these two copies? Between any two chromosomes in a pair,
the nucleotide sequence of the DNA is very similar: more than
99 of 100 bp are identical. Most of the variations result in no
observable phenotype and are therefore silent polymorphisms
or rare variants. The less frequent variations in DNA sequence
that correspond with a phenotype are the fundamental chemical
basis for alleles.
The two copies of a given locus in an individual can by
chance be identical, in which case the individual is homozygous
for that particular allele. On the other hand, an individual can
have two different alleles, one derived from each parent, and
the individual is then heterozygous. An individual who is
heterozygous for two different alleles, neither of which is wildtype, is called a compound heterozygote.
Uniparental disomy or isodisomy is the term given for the
rare occasions when a locus is homozygous, but both identical
SECTION 1
GENETICS
FIGURE 1.7. Steps involved in mitosis and meiosis. In both processes, the rst step involves the replication of DNA so that each chromosome
arm is duplicated, producing chromosomes with sister chromatids. In mitosis, the chromosomes divide so that each daughter cell receives a
short and long chromatid from each chromosome in the pair. In meiosis, there is often recombination between chromatids from homologous
chromosomes. After this, there is the rst meiotic division, which segregates the chromosome pairs, followed by the second meiotic division
which produces gametes with one set of chromatids from only one member of each pair of chromosomes.
alleles are derived from the same parent. As an illustration,

some patients with cystic brosis have been found who are
homozygous for a mutant allele that is present in only one
parent.19 A patient with rod monochromatism has been
reported with isodisomy for chromosome 14q; this case possibly
indicates that a recessive gene for the disease is on that
chromosome.20 Isodisomy has also been implicated in Ushers
syndrome21 and retinal dystrophy associated with mutations in
RPE65 and MERTK genes.22
PATTERNS OF HUMAN INHERITANCE

The major types of inheritance of human disease are: dominant,
recessive, X-linked, mitochondrial (also called maternal),
digenic, and polygenic. Of these, the rst four are the most
commonly considered in ophthalmologic practice and will be
discussed in most detail. For reference, Figure 1.8 provides
schematic pedigrees illustrating each of these four inheritance
patterns.
DOMINANT (ALSO CALLED AUTOSOMAL

DOMINANT)
If a mutation is present in one of the two gene copies at an
autosomal locus, and if this heterozygous mutation produces a
disease, the mutation is called dominant. For example, a patient
with dominant retinitis pigmentosa will have a defect in one

copy of one retinitis pigmentosa gene inherited from one parent
who, in most cases, is also affected with retinitis pigmentosa.
The other copy of that gene, the one inherited from the
unaffected parent, is normal (wild type). The term dominant
comes from the fact that the defective copy dominates over the
wild-type gene copy to cause disease.
1. Nature of a dominant gene defect. Most dominant
mutations cause disease through one of the following three
general mechanisms.
a. Novel function. The mutant allele produces a protein
that has a new function not present in the wild-type
protein. The mutant protein might have a novel
enzyme activity, or it might be toxic.
b. Dominant-negative effect. The mutant protein forms a
complex with the wild-type protein encoded by the
homologous wild-type allele and thus inactivates the
wild-type protein. The phenotype is then a
consequence of little or no functional protein
remaining.
c. Haplo-insufciency. The mutation produces no
functional protein. The homologous wild-type allele
produces functional protein, but because this is the
only functional allele, the target tissues have only 50%
of the normal level of the protein. This reduced level of
functional protein results in disease.
FIGURE 1.8. Factitious pedigrees illustrating various hereditary patterns. Circles represent females; squares represent males. Filled-in circles or
squares represent individuals exhibiting a hypothetical hereditary trait.
CHAPTER 1
SECTION 1
10
GENETICS
2. Note on the classical denition of a dominant allele. It is
customary in human genetics to view a dominant
mutation as one that confers a disease or some other
phenotype when present heterozygously. However, in the
classic, mendelian lexicon, a dominant allele is one that
produces its designated phenotype whether it is present
homozygously or heterozygously. Proven examples of
classically dened, dominant alleles in humans are
uncommon. The Val30Met mutation in the transthyretin
gene is a true dominant, because patients who are
heterozygous for this allele have vitreous amyloidosis and
polyneuropathy comparable in severity to those who are
homozygous.23 In contrast, most dominant human alleles
are loosely categorized as such if they are known to
produce phenotypes when present heterozygously,
regardless of the phenotype produced in a homozygote or
compound heterozygote. This denition is necessary
because individuals who are homozygotes or compound
heterozygotes for dominant alleles causing disease may be
nonexistent. The disease alleles might be so rare that the
likelihood that two affected heterozygous carriers mating, a
precondition for the production of a homozygous offspring,
is exceedingly low. Occasionally, the disease produced by a
dominant mutation is so severe that affected
heterozygotes do not reproduce at all; again, there would be
little possibility for a homozygous individual to be
conceived and the corresponding phenotype to be
displayed. In some exceptional circumstances individuals
who are homozygotes or compound heterozygotes for
purportedly dominant ophthalmic disease alleles have been
identied. They are sometimes found to have a phenotype
that is markedly different from that found in
heterozygotes. For example, a newborn with mutations of
both copies of the aniridia gene had anophthalmia and
severe developmental defects of the central nervous system
that led to death soon after birth.24 If a homozygote for a
dominant allele has a more severe form of the same
recognizable phenotype, the mutant allele is more
appropriately called semidominant. Alleles in the PAX3
gene, causing Waardenburgs syndrome, are semidominant,
exemplied by the report of a family in which a homozygote
had very severe disease (very exaggerated dystopic
canthorum and severely malformed upper limbs) compared
with the heterozygote relatives with more typical disease.25
3. Transmission of a dominant gene defect. A patient with a
dominant mutation at a disease locus can transmit the
normal copy or the defective copy to a child. Each copy has
an equal chance of being passed on, so that each child will
have a 50/50 chance of getting the defective gene copy.
Male and female children are equally likely to inherit the
defective copy. A dominant disease can be inherited from a
father or a mother. Unaffected individuals in a family do
not carry the defective gene copy and therefore cannot pass
a defective copy to their children.
4. Features of a family with a dominant disease. One can be
fairly condent that a disease is dominant in a family if
the following criteria are met:
a. The disease is found in three consecutive generations,
such as grandparents, parents, and children.
b. Every affected member has an affected parent.
c. There is at least one instance of transmission from an
affected father to an affected son.
Many families with a dominant disease do not meet all three
criteria. One will still be able to presume that a dominant mode
of inheritance is likely if some of the criteria are met. For exam-
ple, if there is transmission of the disease directly from a parent

to a child, it is likely that the gene defect is a dominant one.
There are two common sources of error in cataloguing a dominant gene. First, in a family with two generations of affected
individuals, there is the possibility that the allele under study is
actually recessive, that the affected parent is homozygous for
the allele, and that the unaffected parent carries the allele
heterozygously. In this situation, offspring would invariably
inherit the recessive, disease-inducing allele from the affected
parent and would have a 50% chance of inheriting the recessive
allele from the unaffected parent. This situation is called
pseudodominance and is covered later. Pseudodominance is
very unlikely if a family exhibits three consecutive generations
of affected family members.
A second problem occurs when an X-linked allele is incorrectly designated as an autosomal dominant allele. Through a
process called lyonization (discussed later), it is possible for
females heterozygous for an X-linked recessive mutation to
exhibit the corresponding phenotype. If such a female had two
affected sons among four or ve children in all, the pedigree
would mimic that found for autosomal dominant retinitis
pigmentosa. Suspicion of this type of mistake should be high
whenever all affected children of an affected mother are male.
This mistake is eliminated if one stipulates that a pedigree must
show father-to-son transmission of a trait before autosomal
dominant inheritance is diagnosed conclusively.
RECESSIVE (ALSO CALLED AUTOSOMAL

RECESSIVE)
A recessive disease arises if it is necessary for defects to be
present in both gene copies at an autosomal locus. One wildtype allele together with one recessively defective allele does not
cause disease. Hence a wild-type allele always dominates over a
recessive one. The same recessive defect might affect both gene
copies, in which case the patient is said to be a homozygote.
Different recessive defects might affect the two gene copies, in
which case the patient is a compound heterozygote.
1. Nature of a recessive gene defect. Most recessive
mutations that have been functionally characterized result
in null alleles, which are dened as alleles that produce no
functional protein. It is the lack of the proteins activity
that causes disease. For example, patients with gyrate
atrophy have recessive mutations in both copies of the
locus normally encoding the enzyme ornithine
aminotransferase. The disease is produced as a
consequence of the lack of functional enzyme.26
2. Note on the classical denition of a recessive allele.
Classically dened recessive mutations are frequently
encountered in human genetics. The heterozygote parents
of an affected child (who is either a homozygote or a
compound heterozygote) have a wild-type phenotype. In
certain cases, however, recessive mutations are loosely
dened. Consider alleles at the hemoglobin locus, where
the sickle-cell allele is called recessive. However, an
individual homozygous for a wild-type allele is not
phenotypically equivalent to the heterozygote that carries
one wild-type and one sickle allele. The latter individual,
who has the sickle trait, can become symptomatic if he or
she visits an environment with low oxygen pressure such
as the upper atmosphere.
3. Transmission of a recessive gene defect. In a family with
recessive disease, both parents are unaffected carriers, each
having one wild-type allele and one mutant allele. Each
parent has a 50% chance of transmitting the defective
allele to a child. Since a child must receive a defective
allele from both parents to be affected, each child has a

25% chance of being affected (50% 50% = 25%).
4. Features of a family with a recessive disease. The following
features make it likely that a family has a recessive disease.
a. The parents are unaffected, and there is no previous
family history of the disease. If the parents are blood
relatives (e.g., cousins), the disease in the offspring is
even more likely to be recessive.
b. Male and female children are affected equally severely.
On an average, one in four offspring of two carrier parents will
be a homozygote and affected. Consanguineous mates tend to
be carriers of the same rare alleles, so that children with recessive disease are often the product of such marriages. If a sibship
with a presumed recessive disease has only affected males, the
possibility of X-linked inheritance should be considered.
X-LINKED (ALSO CALLED X-LINKED

RECESSIVE)
Mutations of the X chromosome produce distinctive inheritance patterns, because males have only one copy of the
X chromosome whereas females have two. Almost all X-linked
gene defects are of the X-linked recessive category. Carrier
females are unaffected because they have one normal copy of
the gene in question and one defective copy. Carrier males will
be affected because their only copy is defective; that is, there is
no normal copy to compensate for the recessive defect.
1. Nature of an X-linked recessive defect. Like recessive
mutations involving autosomal loci, most recessive
mutations of the X chromosome result in null alleles that
produce no functional protein.
2. Transmission of an X-linked recessive gene defect. First
consider the situation of a male affected with an X-linked
disease. He has only one copy of any X-linked gene, thus
he will transmit his defective X-linked gene to every
daughter. All his daughters will be carriers. All his sons
will be unaffected and will not be carriers, because fathers
do not pass any X-linked genes to sons. Note that neither
the daughters nor the sons of a male affected with an
X-linked disease will be affected.
Next consider the situation of a carrier female who
carries one defective allele at an X-chromosome locus.
Each child of the carrier female has a 50% chance of
inheriting the defective allele. If a son inherits the defective
copy, he will be affected. If a daughter inherits the defective
copy, she will be a carrier like her mother. If either a
daughter or a son inherits the mothers normal gene copy,
the child will be unaffected and will not be a carrier.
Ordinarily, no carrier females will be affected. However,
for some X-linked diseases, female carriers can exhibit a
phenotype that is usually less severe than that found in the
affected male relatives. This could be due to the process of
lyonization. In order for males (with one X chromosome)
and females (with two X chromosomes) to have equal
levels of expression of X-linked genes, female cells express
genes from only one of the two X chromosomes that they
have. The decision as to which X chromosome is expressed
is made early in embryogenesis, and the line of cells
descending from each decision-making progenitrix cell
faithfully adheres to the choice of the active X chromosome
of the progenitrix. Hence, females are mosaics with some
of the cells in each tissue expressing the maternally derived
set of X-linked alleles and the remainder expressing the
paternally derived X-linked alleles. The proportion of cells
that express the mutant versus the wild-type alleles in
each tissue can vary. By chance a susceptible tissue might

have a preponderance of cells expressing the mutant
X chromosome, in which case the corresponding disease
would become manifest. An example of this is offered by
some female carriers of X-linked retinitis pigmentosa who
develop symptoms, fundus signs, and electroretinographic
abnormalities of the disease. Most females affected with
X-linked retinitis pigmentosa because of lyonization have
milder disease than that found in their male relatives.27
Another explanation for a female affected with an
X-linked disease involves the unusual situation in which
the father is affected and the mother is a carrier. The
father invariably will transmit his defective copy to every
daughter. If the mother happens to transmit the defective
copy to a daughter, the daughter will be a homozygote or
compound heterozygote at the disease locus. This is the
usual explanation for females who show protan or deutan
color vision abnormalities due to defects in the genes
encoding red and green cone opsins on the X chromosome.
About 6% of X chromosomes in whites have defects in the
red and green cone opsin genes, so ~6% 6% = 0.36% of
females, or ~1 in 280, would be homozygotes or
compound heterozygotes. For most ophthalmic diseases,
however, the proportion of female carriers is very low. For
example, for X-linked retinitis pigmentosa, only ~1 in
every 7000 women is a carrier. In view of this low
proportion of carriers, it is very unlikely for an affected
father to marry by chance a female carrier of X-linked
retinitis pigmentosa. Hence, very few females with retinitis
pigmentosa will be homozygotes or compound
heterozygotes for mutations in an X-linked retinitis
pigmentosa gene; most will have autosomal recessive or
autosomal dominant retinitis pigmentosa instead.
3. Features of a family with an X-linked recessive disease. The
following features of a family point to an X-linked recessive
disease gene:
a. The disease is found only in males. (In unusual
circumstances, females may be affected; see the
discussion earlier.)
b. There is no instance of an affected male having an
affected child.
c. If the disease is present in more than one generation,
the affected males are related through a carrier female.
For example, an affected male might have an affected
maternal uncle or an affected maternal grandfather, but
he would not have affected relatives on his fathers side.
CHAPTER 1
LESS COMMON INHERITANCE PATTERNS

1. Maternal or mitochondrial inheritance. The 23 pairs of
human chromosomes described earlier are located in the
nucleus of each cell. In addition, there is a small amount
of DNA in the cytoplasm. This DNA is from the
mitochondrial chromosome, a relatively tiny chromosome
with only 16 569 bp of DNA. Thirteen mitochondrial
proteins, 2 ribosomal RNAs, and 22 tRNAs are encoded by
this chromosome. It is a clinically important chromosome
because mutations are known to cause human disease
(examples relevant to ophthalmology are Leber hereditary
optic atrophy28,29 and KearnsSayre syndrome30). A
noteworthy feature of these mutations is that they are
maternally inherited, because almost all the mitochondria
of a one-cell embryo are derived from the ovum. A father
does not transmit mitochondria to his offspring.
Mitochondrially inherited diseases are inherited invariably
through the maternal lineage.
11
SECTION 1
GENETICS
One other peculiar feature of alleles in the
mitochondrial genome is that an individual is neither
homozygous nor heterozygous for them but rather is
heteroplasmic. A typical cell has numerous mitochondria,
each with ~210 copies of the mitochondrial genome. The
proportion of mutant mitochondrial genomes in each
mitochondrion, and the proportion of mutant
mitochondria in a cell, can vary from one cell to another in
an individual. Differences in the relative proportions of
mutant mitochondria can partly explain the observed
variable severity of mitochondrial diseases. In addition, the
proportion of mutant mitochondria can change during the
lifetime of a patient, which helps to explain the variable
age of onset of mitochondrial diseases.
Upon analysis of a pedigree with a mitochondrially
inherited disease, one may note examples of mother-to-son
and mother-to-daughter transmission, but one should
never observe father-to-child transmission. In a particular
family, the severity of disease can vary tremendously
because of heteroplasmy and perhaps other factors, and
one must be aware of possible asymptomatic carriers when
scrutinizing a pedigree. In the case of Leber optic atrophy, a
mitochondrially inherited disease, individuals with the
same mutation may have signicant variations in disease
progression for unknown reasons.31
2. Pseudodominance. This is the term given to an apparent
dominant inheritance pattern due to recessive defects in a
disease gene. Consider the situation in which an affected
parent has recessive disease due to defects in both copies of
a disease gene and the spouse happens to be a carrier with
one normal gene copy and one copy that has a recessive
defect. Children from this couple will always inherit a
defective gene copy from the affected parent and will have
a 50% chance of inheriting the defective gene copy from the
unaffected carrier parent. On average, half of the children
will inherit two defective gene copies and will be affected.
The pedigree would mimic a dominant pedigree (Fig. 1.9)
because of an apparent direct transmission of the disease
from the affected parent to affected children and because
~50% of the children will be affected. Pseudodominant
transmission is uncommon, because few people are
asymptomatic carriers for any particular recessive gene.
3. Autosomal dominant with reduced penetrance. In some

pedigrees with an autosomal dominant disease, some
individuals who carry the defective gene do not get disease.
This would cause skipped generations; that is, cases
where an unaffected offspring of an affected individual
would have children with the disease. This phenomenon is
typically locus-specic. For example, many families with
dominant retinitis pigmentosa with reduced penetrance
have a defective gene on chromosome 19q13;32 those with
dominant retinitis pigmentosa with full penetrance have
mutations at other loci.
4. X-linked dominant inheritance. A few families with
retinitis pigmentosa appear to have this distinctive
inheritance pattern.33 The inheritance pattern is similar to
X-linked recessive inheritance, but all carrier females are
affected rather than unaffected. All carrier males are
affected as well. Other diseases with ophthalmic
manifestations that are loosely considered to have X-linked
dominant inheritance are Aicardi syndrome (frequent
features are agenesis of the corpus-callosum and patches
of absent retinal pigment epithelium) and incontinentia
pigmenti (irregularly pigmented atrophic scars on the
trunk and the extremities, congenital avascularity in the
peripheral retina with secondary retinal neovascularization).
Both Aicardi syndrome and incontinentia pigmenti occur
almost exclusively in females; it is likely that the X
chromosome gene defects causing these diseases are
embryonic lethals when present hemizygously in males.34,35
5. Digenic inheritance. This is another rare form of
inheritance, which till now has been found only in a few
families with retinitis pigmentosa or ocular albinism.36,37
Digenic inheritance occurs when a patient has
heterozygous defects in two different genes, and the
combination of the two gene defects causes disease.
Individuals who are heterozygous for a mutation only at
one or the other locus are wild-type. Digenic inheritance is
different from recessive inheritance, because the two
mutations involve different gene loci. Affected individuals
are called double heterozygotes rather than compound
heterozygotes. Triallelic inheritance (three mutations
required for disease) has recently been reported in patients
with BardetBiedl syndrome.38
6. Polygenic and multifactorial inheritance. If the expression
of a heritable trait or predisposition is influenced by the
combination of alleles at multiple loci, it is polygenic. The
contributing loci may be quantitative trait loci reflecting
the mathematical formulations used to calculate their
relative impacts on the phenotype or the predisposition.
If environmental factors contribute to a polygenic trait or
disease, the term multifactorial is used. Examples of
phenotypes in ophthalmology likely to be multifactorial
are myopia,39 age-related macular degeneration,40 and
adult-onset open-angle glaucoma.41
PEDIGREE ANALYSIS TO CATEGORIZE

ALLELES
12
FIGURE 1.9. An example of pseudodominance. Beneath each

schematic family member are the alleles of the disease locus under
scrutiny. A is the dominant, wild-type allele; a is the recessive allele
that causes the hypothetical disease. The parent-to-child transmission
of the disease occurs because the unaffected parent is actually a
carrier of the recessive allele.
The classication of a genetic disease or trait can often be made

by examining the relationships between the affected individuals
in a pedigree. The following are general guidelines for using this
method. It should be noted that in many circumstances, it is
not possible to be certain of the mode of inheritance in a
particular family because of the small size of the family or
because of uncertainties in the diagnosis of key family members
who might be too young, unavailable, or deceased.
Pedigree analysis is sometimes not necessary to determine
the inheritance pattern in a family, because for some conditions
there is only one known inheritance pattern. In those cases, the

diagnosis will immediately provide the inheritance pattern. For
example, currently, all known cases of choroideremia have an
X-linked pattern of inheritance. For other diseases, such as
hereditary cataract or hereditary retinal degeneration, many
different inheritance patterns have been observed. In those
cases, pedigree analysis can often be helpful. One constructs a
family tree indicating which members in the family have the
disease in question. It is important to make sure that the information on the pedigree is as complete and correct as possible.
For example, if a distant relative is reported to have had poor
eyesight, one must know whether that report reflects the
ophthalmic disease in question or simply the relatives need for
eyeglasses. Examination of the pedigree rarely proves the type
of inheritance beyond any doubt, but it can allow one to infer
the most likely inheritance pattern.
DISEASE IS PRESENT IN ONLY ONE FAMILY

MEMBER
Isolate or simplex cases of disease refer to families in which
two parents with no previous family history of the disease in
question have one affected child. In some cases, a simplex case
might not have a hereditary disease at all. For example,
~8090% of unilateral, simplex cases of retinoblastoma are not
hereditary. Alternatively, simplex cases might represent
autosomal recessive disease, with both parents being carriers
and the affected child having inherited a defective gene copy
from each parent. If the affected simplex case is a male, it is
possible that he has X-linked disease, with the mother possibly
being a carrier. For some diseases such as retinitis pigmentosa,
a careful ophthalmologic evaluation including an electroretinogram of the mother might give clues as to her status in
this regard. Another possibility is that the simplex case has a
new gene defect not present in either parent. This is thought to
be infrequent, because so few genes become mutant from one
generation to the next.
DISEASE PRESENT IN TWO OR MORE

INDIVIDUALS IN THE SAME GENERATION
An example of this situation would be a family with two or
more siblings with a disease and no previous family history of
the disease. In such families, the inheritance pattern is usually
autosomal recessive. However, if the affected children are all
males, the possibility of X-linked disease should be considered.
Other unusual inheritance patterns, such as maternal, digenic,
or multifactorial are possible.
DISEASE PRESENT IN TWO CONSECUTIVE

GENERATIONS
The disease is most likely to be autosomal dominant. If there is
direct transmission from a father to a son, an autosomal
dominant gene is inferred with even more certainty. Uncommon exceptions include pseudodominance or digenic inheritance.
If there is direct transmission from a mother to a child, an
autosomal dominant gene is still very likely, but maternal and
X-linked inheritance should be considered as well.
DISEASE PRESENT IN TWO GENERATIONS

SEPARATED BY AN UNAFFECTED
GENERATION
If the unaffected individual connecting the affected generations
is a female and if all affected individuals are male, X-linked
inheritance is likely. Alternatively, this could represent
autosomal dominant inheritance with reduced penetrance. This

type of inheritance pattern may also result from imprinting,
where the disease is expressed only when inherited from the
mother (for some disease) or the father (for other disease).42
CHAPTER 1
DISEASE PRESENT IN THREE OR MORE

CONSECUTIVE GENERATIONS
Dominant inheritance is most likely, although digenic and
X-linked dominant inheritance are also possibilities.
MAP OF THE HUMAN GENOME

LINKAGE
Because of the mixing of genes caused by meiotic crossovers and
the random assortment of chromosomes, alleles at two distinct
loci are usually inherited together ~50% of the time. In the less
common circumstance when alleles at two loci are inherited
together more than 50% of the time, the two loci are linked.
Linked loci are physically close to each other on the same
chromosome.
The distance between two linked loci can be measured two
ways: by the number of base pairs of DNA separating the loci
(physical distance) or by the frequency of meiotic crossovers
occurring between the two loci (genetic distance or recombination distance). How are the two measures related? A
haploid human genome contains ~3.2 billion bp of DNA. Since
30 crossovers occur in a typical meiosis, there is an average of
one crossover per 100 million bp per meiosis. Between two loci
physically separated by a distance of 1 million bp, there would
be approximately one crossover per 100 meioses, or a 1% crossover rate. This distance is called 1 centimorgan (cM) and is one
of the basic units in genetics for measuring the separation
between two loci. The conversion of 1 cM/million bp is an
overall average for the human genome, since the frequency of
crossovers is not equal throughout the length of each
chromosome. The actual gure for a segment of a chromosome
can be more than 10 times greater or less. Furthermore, it can
be different in germ cells from males compared with females.
One of the major contemporary goals in the study of human
genetics is the construction of a map of the physical position of
every human gene and the correlation of that map with the
recombination distances between linked loci. This was one goal
of the human genome project which was a formidable task,
because the human genome is so large. The physical map that
was the rst step of this endeavor was started by physically
assigning many human genes to their specic locations on
chromosomes.43 These and other landmarks within the human
genome sequence led to the nal determination of the DNA
sequence for each chromosome of the human genome.6
DNA POLYMORPHISMS
A major step in the human genome project was the construction of a linkage map of the human genome. This involves the
determination of which human loci are linked and the recombination distances between them. This work is based on sites
in the human genome where there is variation in the DNA
sequence, called DNA polymorphisms. Most DNA polymorphisms are unrelated to clinically evident phenotypes,
however single nucleotide polymorphisms (SNPs), may change
the amino acid sequence of a protein causing an abnormal
function and disease phenotype.
Three major categories of DNA polymorphisms were used for
linkage maps of the human genome: RFLPs (for restriction fragment length polymorphisms), VNTRs (for a variable number of
13
SECTION 1
14
GENETICS
tandem repeats), and microsatellites. RFLPs are the result of
occasional variations that typically affect a single base pair in
the DNA sequence. They are detectable with enzymes, called
restriction endonucleases, that are puried from bacteria. A
restriction endonuclease cleaves DNA at specic locations,
usually specied by a particular stretch of 46 bp called the
recognition sequence. If even a single base pair is altered at a
recognition site, a restriction endonuclease will not cleave
DNA at that site. For example, the restriction endonuclease
EcoRI cleaves DNA at the sequence GAATTC (its recognition
sequence) but would not cleave the sequence GAAGTC or
GATTTC. Restriction endonucleases allow one to trace relatively
easily the inheritance of a single-base polymorphism if a recognition sequence is created or destroyed by the variation.
VNTRs are sites in the human genome where there is a
tandem repetition of a DNA sequence. The repeat unit is
~1560 bp in length and typically has a core sequence that is
common to all VNTRs.44 The number of repeat units at a
VNTR varies from a few to dozens, and this variation is the
basis for the alleles specied by these polymorphisms. In some
cases, VNTR variation may contribute to regulation of gene
expression.45 Microsatellites are like VNTRs in that they are
tandemly repeated DNA sequences, but the repeated unit is
much smaller, typically 24 bp. The most frequently used
microsatellites are repeats of the dinucleotide sequence CA;
these microsatellites are also known as CA repeats. VNTRs
and microsatellites were preferred for the linkage studies that
dened the human genome because they are multiallelic. A
higher proportion of individuals are heterozygous for polymorphisms with numerous alleles, and therefore VNTRs and
microsatellites provide more linkage data than RFLPs, which
are biallelic.
By following the inheritance of distinct DNA polymorphisms
in human pedigrees, one can learn which are linked with each
other and at what recombination distances. To date, linkage
maps of each human chromosome are available with highly
informative polymorphic markers distributed roughly every
13 cM or less.46
With such a linkage map, it is possible to determine the
location of a gene causing a human disease once one has a set
of families with the disease available for study. DNA samples
from family members are rst obtained. Leukocyte DNA is
typically used; DNA from 10 mL of venous blood is sufcient to
assay hundreds of DNA polymorphisms distributed throughout
the genome. The polymorphic site that most often correlates
with the disease is the one that is closest to the disease gene
(Fig. 1.10). By knowing the chromosomal location of that DNA
polymorphism, one has the approximate chromosomal location
for the disease gene. The strategies embodied in the term
positional cloning allow one to proceed from the approximate
chromosomal location of a disease gene, based on the data from
the DNA polymorphisms, to the actual isolation of the gene.
Positional cloning approaches are typically very labor-intensive,
but they have been successful in identifying a number of genes
causing ophthalmologic disease. Examples are the retinoblastoma
gene (on chromosome 13), X-linked genes for choroideremia
and one form of retinitis pigmentosa (RPGR), the aniridia gene
(chromosome 11), and a gene for Usher syndrome type I
(chromosome 11).
SNPs are single-letter variations in a DNA base sequence,
and are the most common source of genetic variation in the
human genome.47 Over 10 million SNPs are present in the
human genome with a density of one SNP approximately every
100 bases. In addition to their abundance, SNPs are useful
genetic markers because the high quality of the data makes the
automation of the analysis possible. Some SNPs (nonsynonomous
SNPs) change the amino acid composition of the protein and
FIGURE 1.10. An example of a linkage study using RFLPs or other

DNA markers. In this hypothetical example, a large pedigree with
autosomal dominant retinitis pigmentosa is illustrated. Filled circles
and squares indicated affected individuals. The numbers beneath
each symbol are the alleles at marker loci that have been studied. This
gure only shows the results of informative markers, i.e., for markers
where the affected members of generations I and II are heterozygotes
(1,2) and the unaffected spouses were homozygotes (2,2). (Note that
any markers that are not heterozygous in the affected members of
generations I and II would provide little useful information for this
analysis.) Beneath the symbols for the members of the generation III
are the alleles at the informative markers, as well as the chromosomal
location of each marker. At each of the marker loci, the 1 allele is
dened as the allele that was transmitted from the affected male in
generation I to the affected male in generation II. (This way of naming
the 1 allele is done for pedagogic purposes for this gure.) If a
marker locus is close to the disease gene, then the affected members
of generation III should usually have marker 1 allele and the
unaffected members should not. The markers G and S most closely t
this prediction. For both of these markers, nine out of the 10 members
of generation III t the expected pattern for close linkage; the two
members who do not probably are examples of meiotic recombination
between the marker loci and the disease locus. Since both these
markers come from the long arm of chromosome 3 (bands 3q21 and
3q24, respectively), these data indicate that the locus for the disease
gene in this family is probably within or near this region. Data of this
sort led to the search for mutations of the rhodopsin gene in patients
with autosomal dominant retinitis pigmentosa, since the rhodopsin
gene was known to lie in the region 3q21q24.
can be associated with disease. For example, the amino acid

change in the complement factor H gene recently shown to be
a risk factor for macular degeneration is a nonsynonomous
SNP.48 Although SNPs are biallelic (RFLPs are a subset of SNPs)
whole genome association studies using automated technologies are currently possible, allowing a large number of SNPs
to be evaluated in a genetic study. Screening many SNPs and
creating haplotypes, which are groups of SNPs that are inherited
together, compensates for the low information content of the
polymorphism. Another recent advance of the Human Genome
Project is the HapMap which denes haplotype blocks for four
ethnic populations to be used for disease gene identication
studies.49
MUTATIONS
CHAPTER 1
CATEGORIES OF MUTATIONS
A new alteration in the DNA sequence of a gene is called a
mutation. The word mutant can refer to the specic sequence
abnormality (i.e., a mutant base pair), to the defective allele
(mutant gene or mutant allele), to the gene product (mutant
protein), or to the organism that is affected by the mutation
(mutant mouse). There are various ways that mutations can be
organized for didactic purposes. Mutations can be grouped
according to whether they cause a dominant or a recessive
phenotype, or no phenotype at all (silent mutations). Recessive
mutations are often loss-of-function, or null mutations because
they often interfere in some way with the production of an
active protein product. Dominant alleles can be loss-offunction, but typically represent gain-of-function mutations.
TYPES OF LESIONS IN DNA

Another way to classify mutations is according to the type of
lesion affecting the DNA sequence. A point mutation is the
change of a single base for another. If a purine changes to
another purine, or if a pyrimidine changes to another
pyrimidine, the point mutation is called a transition. If a purine
changes to a pyrimidine or vice versa, the mutation is a transversion. Although there are 12 possible transversions and four
possible transitions (Fig. 1.11), transitions outnumber transversions at most human loci where naturally occurring mutations have been characterized. Among the transitions, the change
from a C to a T is the most frequent and most commonly
occurs if the C is part of the dinucleotide sequence CG.
A point mutation can change a codon so that it species a
different amino acid. This is called a missense mutation. For
example, a C-to-A transversion in codon 23 of the human
rhodopsin gene, a cause of autosomal dominant retinitis
pigmentosa, changes that codon from one that species proline
(CCT) to one specifying histidine (CAT).7
A nonsense mutation, also called a premature stop codon, is
one that changes a codon that normally species an amino acid
into a termination codon. For example, a C-to-T transition in
codon 446 of the retinoblastoma gene, found to be the cause of
hereditary retinoblastoma in one pedigree, changes the codon
from CGA (arginine) to TGA (stop). During translation of the
resultant mRNA, the encoded protein will have only the rst
445 amino acid residues, whereas the normal protein product
has 928 residues. The truncated, nonfunctional, mutant protein
will not be able to prevent retinoblastoma.
A point mutation or other alteration affecting either of the
ends of an intron will interfere with the proper splicing of the
transcribed RNA. The 5 end of an intron absolutely requires
the dinucleotide sequence GT (called the splice donor sequence),
and the 3 end must have the dinucleotide sequence AG (the
splice acceptor sequence). If a mutation changes either the splice
acceptor or splice donor sequences, it is called a splice site
mutation. The mRNA transcript will either improperly include
sequence from the intron or will eliminate part or all of an exon.
In either case, one expects a major alteration of the translated
protein product.
Other areas of a transcriptional unit may be exquisitely sensitive to single base changes. For example, the promoter region
upstream of a transcribed sequence has binding sites for factors
necessary for the proper expression of a gene. A change in the
sequence of these binding sites can bring about underexpression
or overexpression of the protein product. Additional sequences
that modulate the expression of a gene can be located in diverse
regions of a transcriptional unit, such as within introns or
FIGURE 1.11. Transitions and transversions. The black arrows

indicate base changes that would be termed transitions, because they
involve an interchange of two bases of the same type (e.g., both
purines). Transversions (gray arrows) involve the interchange of a
purine and a pyrimidine.
within the 5 or 3 untranslated regions, or even many thousands

of bases away from the cluster of exons and introns. Mutations
in these regions can also affect the expression of a gene and
cause an observable phenotype.
A frameshift mutation occurs when one or more bases are
inserted into or deleted from the coding region of a gene. A
frameshift mutation changes the reading frame of the encoded
message. Since the genetic code uses consecutive, nonoverlapping triplets of DNA sequence, the number of bases that are
inserted or deleted to cause a frameshift cannot be a multiple of
three. Downstream of a frameshift mutation there is a drastic
alteration of the amino acid sequence, often with a premature
termination codon so that the encoded protein is truncated as
well. If the number of base pairs removed or inserted in the
coding region is a multiple of 3, the mutation is called an
in-frame deletion or insertion. Only the amino acids encoded by
the deleted or inserted codons will be affected.
Large deletions might remove a large portion of a transcriptional unit (an internal deletion), or the 5 or 3 end of a
gene, or an entire transcriptional unit. Very large deletions
might remove a number of closely linked genes. To be
observable in a karyotype (i.e., to be detectable cytogenetically),
a deletion must remove at least a few million base pairs of
DNA. Since the density of genes in the human genome is ~1
per 30 00050 000 bp, a cytogenetically detectable deletion
usually affects dozens of genes. Like deletions, insertions can
interfere with a gene if they interrupt a coding region or if they
occur in a region that is important for proper RNA splicing or
the proper expression of a gene.
15
SECTION 1
GENETICS
This general categorization of mutations is not always applicable to naturally occurring defects in human DNA. Occasionally a single mutational event causes many single-base
substitutions in a gene. Some deletions are complex, causing a
foreign segment of DNA to be inserted where the normal
sequence was deleted. More complex rearrangements have been
documented, such as inversions where a segment of DNA is
flipped backwards and relocated to a different region of the
gene or to another gene. Such complex mutations represent a
minority of the lesions that cause a disease.
Finally, because of our limited understanding of the molecular
control of the regulation of transcription, splicing, and translation, the precise effect of a mutation sometimes cannot be
deduced with certainty from inspection of the DNA sequence
alone. The arrangement of bases in the coding region of a gene
not only species the amino acid sequence of the protein
product but also has some role in the recognition of splice sites
and in maintaining the nuclear and cytoplasmic stability of the
nal mRNA product. Consequently, a point mutation labeled
as a missense mutation, since it changes the amino acid
specicity of a codon, might actually interfere with the splicing
of an RNA transcript so that a very different protein product is
produced. In some cases, considerable effort in a research laboratory is necessary to establish the exact biochemical consequences of a mutant allele of a known DNA sequence.
ORIGIN OF MUTATIONS
Germline mutations either arise de novo in an individual or are
inherited from a carrier parent. Actually, all mutations arise
de novo in some individuals. Sometimes that individual is a
distant ancestor who is called the founder or progenitor of the
mutation.
VARIABILITY IN THE RATE OF NEW GERMLINE

MUTATIONS
For any given genetic disease, the proportion of patients who
have a new germline mutation (as opposed to those who have
inherited a mutation) is dependent on the mutation rate and the
ability of those who carry the mutation to survive and reproduce.
In practice, the quantication of both of these factors is difcult. Mutation rates at human loci extend over many orders of
magnitude. New mutations at some loci, such as the Duchenne
muscular dystrophy locus or the retinoblastoma locus, occur in
more than one in 50 000 live births. For other diseases, such as
tritanopia (due to a defect in the gene for blue cone opsin), the
mutation rate is thought to be well below one in 10 million live
births. The explanation for the wide range of mutation rates at
different human loci is obscure. Possibilities include the size of
the transcriptional unit (the Duchenne locus and the retinoblastoma locus are both large, encompassing 2 million and 180
thousand bp, respectively), limitations on the types of mutations
that can cause a disease (almost all mutations of the rhodopsin
gene causing dominant retinitis pigmentosa are missense mutations), or inherent variation in the mutability of loci based on
their DNA sequences or their positions in the genome.
MUTATION SPECTRUM OF A GENE
16
An examination of mutations might provide clues to the

mechanisms that are responsible for them. A mutation spectrum is a compilation of the frequency of each type of mutation
at a specied locus; that is, the percentage of deletions, insertions, point mutations (broken down into transitions and transversions, or the specic nucleotide changes), frameshifts, and so
forth. Tabulating the types of mutations causing a disease can
give clues as to the functional domains of the encoded protein.

Laboratory studies suggest that each class of mutagens causes
certain types of mutations. For example, approximately half of
the mutations resulting from gamma radiation are deletions
and only ~20% are transitions. Ultraviolet light, on the other
hand, induces deletions very infrequently but appears to facilitate transitions (~50% of the resultant mutations). Thus, knowledge of the mutation spectrum can provide evidence implicating
specic environmental mutagens as the cause of a disease.
Indeed, ultraviolet light has been implicated by such evidence in
the genesis of squamous cell carcinoma in sun-exposed skin.50
Unfortunately, the mutation spectrum of only a few genes is
known with any accuracy. The available data do not implicate
any specic environmental mutagen as the cause of most naturally occurring mutations in humans.
PARENTAL ORIGIN OF NEW MUTATIONS

An individual with a new germline mutation carries that mutation on the gene copy derived from either the mother or the
father (except for males with a new mutation on the X chromosome, a chromosome necessarily derived from a sons mother).
The parental origin of an autosomal allele with a new mutation
can be determined in some situations. At many human loci, the
general rule is that new germline mutations preferentially arise
on a paternally derived allele. For example, ~8090% of new
germline mutations at the retinoblastoma locus51 or the von
Recklinghausen neurobromatosis locus52 affect the paternally
derived allele. One attractive explanation for this bias relates to
the fact that more than 300 cell divisions separate a one-cell
male embryo from his resultant sperm (produced decades later)
compared with ~20 cell divisions separating a one-cell female
embryo from her resultant ova (produced while the female is
still in utero).53 The excess of mutant sperm may pertain to the
fact that mutations chiefly arise during DNA replication.
EPIGENETIC MUTATIONS
Defects that do not alter the sequence of DNA are called
epigenetic. How such defects are transmitted through the
germline, if at all, is open to speculation. One possible basis for
epigenetic defects is that some bases of DNA are modied by
the addition of methyl groups. The classic example of this
involves the dinucleotide sequence CG. The cytosine in a CG
dinucleotide sequence is customarily methylated in human
DNA. However, in the vicinity of the promoter region at the 5
end of a gene, cytosines are unmethylated in cells that express
the gene.54 If this region of a gene is aberrantly methylated, the
gene will not be expressed. Despite no change in the DNA
sequence, the allele will be inactive and thus equivalent to one
with a null mutation. There is evidence that epigenetic defects
in the retinoblastoma gene are one cause of retinoblastoma.5557
IMPRINTING
Human cells have the capacity to distinguish the maternally
derived allele from the paternally derived allele at some loci.
This may be due to differences in the pattern of methylation of
the two alleles or to differences in the conguration of DNAbinding factors that are present in chromatin. This imprinting
of DNA has clinical importance because it explains peculiar
patterns seen for some genetic diseases. For example, a deletion
of q11q13 of human chromosome 15 causes PraderWilli
syndrome if it affects the paternally derived chromosome 15,
but Angelman syndrome if it affects the maternally derived
chromosome homolog.58 Angelman syndrome can be associated
with oculocutaneous albinism.59
REFERENCES
1. Sipos L, Gyurkovics H: Long-distance
interactions between enhancers and
promoters. FEBS J 2005; 272:32533259.
2. Wilson MG, Towner JW, Fujimoto A:
Retinoblastoma and D-chromosome
deletions. Am J Hum Genet 1973; 25:57.
3. Lele KP, Penrose LS, Stallard HB:
Chromosome deletion in a case of
retinoblastoma. Ann Hum Genet 1963;
27:171.
4. Sparkes RS, Sparkes MC, Wilson MG, et al:
Regional assignment of genes for human
esterase D and retinoblastoma to
chromosome band 13q14. Science 1980;
208:10421044.
5. Crolla JA, van Heyningen V: Frequent
chromosome aberrations revealed by
molecular cytogenetic studies in patients
with aniridia. Am J Hum Genet 2002;
71:11381149.
6. Lander ES, Linton LM, Birren B, et al: Initial
sequencing and analysis of the human
genome. Nature 2001; 409:860921.
7. Dryja TP, McGee TL, Reichel E, et al: A
point mutation of the rhodopsin gene in
one form of retinitis pigmentosa. Nature
1990; 343:364.
8. Eiberg H, Mohr J: Major genes of eye color
and hair color linked to LU and SE. Clin
Genet 1987; 31:186191.
9. Bobadilla JL, Macek M Jr, Fine JP,
Farrell PM: Cystic brosis: a worldwide
analysis of CFTR mutations correlation
with incidence data and application to
screening. Hum Mutat 2000; 16:143156.
10. van den Hurk JA, Schwartz M, van
Bokhoven H, et al: Molecular basis of
choroideremia (CHM): mutations involving
the Rab escort protein-1 (REP-1) gene.
Hum Mut 1997; 9:110117.
11. Rivolta C, Sharon D, DeAngelis MM,
Dryja TP: Retinitis pigmentosa and allied
diseases: numerous diseases, genes, and
inheritance patterns. Hum Mol Genet 2002;
11:12191227.
12. Berger W, van de Pol D, Warburg M, et al:
Mutations in the candidate gene for Norrie
disease. Hum Mol Genet 1992; 1:461467.
13. Chen ZY, Battinelli EM, Fielder A, et al: A
mutation in the Norrie disease gene (NDP)
associated with X-linked familial exudative
vitreoretinopathy. Nature Genet 1993;
5:180183.
14. Hutcheson KA, Paluru PC, Bernstein SL,
et al: Norrie disease gene sequence
variants in an ethnically diverse population
with retinopathy of prematurity. Mol Vis
2005; 11:501508.
15. Dryja TP, Berson EL, Rao VR, et al:
Heterozygous missense mutation in the
rhodopsin gene as a cause of congenital
stationary night blindness. Nature Genet
1993; 4:280283.
16. Neidhardt J, Barthelmes D, Farahmand F,
et al: Different amino acid substitutions at
the same position in rhodopsin lead to
distinct phenotypes. Invest Ophthalmol Vis
Sci 2006; 47:16301635.
17. McLaughlin ME, Sandberg MA, Berson EL,
et al: Recessive mutations in the gene
encoding the b-subunit of rod
phosphodiesterase in patients with retinitis
pigmentosa. Nature Genet 1993; 4:130134.
18. Dryja TP: Molecular genetics of Oguchi
disease, fundus albipunctatus, and other
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
forms of stationary night blindness: LVII

Edward Jackson Memorial Lecture. Am J
Ophthalmol 2000; 130:547563.
Fares F, David M, Lerner A, et al: Paternal
isodisomy of chromosome 7 with cystic
brosis and overgrowth. Am J Med Genet
A 2006; 140:17851788.
Pentao L, Lewis RA, Ledbetter DH, et al:
Maternal uniparental isodisomy of
chromosome 14: Association with
autosomal recessive rod monchromacy. Am
J Hum Genet 1992; 50:690699.
Rivolta C, Berson EL, Dryja TP: Paternal
uniparental heterodisomy with partial
isodisomy of chromosome 1 in a patient
with retinitis pigmentosa without hearing
loss and a missense mutation in the Usher
syndrome type II gene USH2A. Arch
Ophthalmol 2002; 120:15661571.
Thompson DA, McHenry CL, Li Y, et al:
Retinal dystrophy due to paternal isodisomy
for chromosome 1 or chromosome 2, with
homoallelism for mutations in RPE65 or
MERTK, respectively. Am J Hum Genet
2002; 70:224229.
Sandgren O, Holmgren G, Lundgren E:
Vitreous amyloidosis associated with
homozygosity for the transthyretin
methionine-30 gene. Arch Ophthalmol
1990; 108:15841586.
Glaser T, Jepeal L, Edwards JG, et al: PAX6
gene dosage effect in a family with
congenital cataracts, aniridia,
anophthalmia, and central nervous system
defects. Nature Genet 1994; 7:463471.
Zlotogora J, Lerer I, Bar-David S, et al:
Homozygosity for Waardenburg syndrome.
Am J Hum Genet 1995; 56:11731178.
Peltola KE, Nanto-Salonen K, Heinonen OJ,
et al: Ophthalmologic heterogeneity in
subjects with gyrate atrophy of choroid and
retina harboring the L402P mutation of
ornithine aminotransferase. Ophthalmology
2001; 108:721729.
Lorenz B, Andrassi M, Kretschmann U:
Phenotype in two families with RP3
associated with RPGR mutations.
Ophthalmic Genet 2003; 24:89101.
Wallace DC, Singh G, Lott MT, et al:
Mitochondrial DNA mutation associated
with Lebers hereditary optic neuropathy.
Science 1988; 242:1427.
Howell N: LHON and other optic nerve
atrophies: the mitochondrial connection.
Dev Ophthalmol 2003; 37:94108.
DiMauro S: Lessons from mitochondrial
DNA mutations. Semin Cell Dev Biol 2001;
12:397405.
Spruijt L, Kolbach DN, de Coo RF, et al:
Influence of mutation type on clinical
expression of Leber hereditary optic
neuropathy. Am J Ophthalmol 2006;
141:676682.
Rivolta C, McGee TL, Rio Frio T, et al:
Variation in retinitis pigmentosa-11
(PRPF31 or RP11) gene expression
between symptomatic and asymptomatic
patients with dominant RP11 mutations.
Hum Mutat 2006; 27:644653.
Sullivan LS, Bowne SJ, Birch DG, et al:
Prevalence of disease-causing mutations in
families with autosomal dominant retinitis
pigmentosa: a screen of known genes in
200 families. Invest Ophthalmol Vis Sci
2006; 47:30523064.
34. Rosser T: Aicardi syndrome. Arch Neurol

2003; 60:14711473.
35. Berlin AL, Paller AS, Chan LS: Incontinentia
pigmenti: a review and update on the
molecular basis of pathophysiology. J Am
Acad Dermatol 2002; 47:169187.
36. Kajiwara K, Berson EL, Dryja TP: Digenic
retinitis pigmentosa due to mutations at the
unlinked peripherin/RDS and ROM1 loci.
Science 1994; 264:16041608.
37. Morell R, Spritz RA, Ho L, et al: Apparent
digenic inheritance of Waardenburg
syndrome type 2 (WS2) and autosomal
recessive ocular albinism (AROA). Hum Mol
Genet 1997; 6:659664.
38. Katsanis N, Ansley SJ, Badano JL, et al:
Triallelic inheritance in BardetBiedl
syndrome, a Mendelian recessive disorder.
Science 2001; 293:22562259.
39. Klein AP, Duggal P, Lee KE, et al: Support
for polygenic influences on ocular refractive
error. Invest Ophthalmol Vis Sci 2005;
46:442446.
40. Haddad S, Chen CA, Santangelo SL,
Seddon JM: The genetics of age-related
macular degeneration: a review of progress
to date. Surv Ophthalmol 2006;
51:316363.
41. Hewitt AW, Craig JE, Mackey DA: Complex
genetics of complex traits: the case of
primary open-angle glaucoma. Clin Exp
Ophthalmol 2006; 34:472484.
42. Van Heyningen V, Yeyati PL: Mechanisms
of non-Mendelian inheritance in genetic
disease. Hum Mol Genet 2004;
13:R225R233.
43. Olivier M, Aggarwal A, Allen J, et al: A highresolution radiation hybrid map of the
human genome draft sequence. Science
2001; 291:12981302.
44. Nakamura Y, Leppert M, OConnell P, et al:
Variable number of tandem repeat (VNTR)
markers for human gene mapping. Science
1987; 235:16161622.
45. Knight JC: Regulatory polymorphisms
underlying complex disease traits. J Mol
Med 2005; 83:97109.
46. Nievergelt CM, Smith DW, Kohlenberg JB,
Schork NJ: Large-scale integration of
human genetic and physical maps.
Genome Res 2004; 14:11991205.
47. Feuk L, Marshall CR, Wintle RF, Scherer SW:
Structural variants: changing the
landscape of chromosomes and design of
disease studies. Hum Mol Genet 2006;
15:R57R66.
48. Wiggs JL: Complement factor H and
macular degeneration: the genome yields
an important clue. Arch Ophthalmol 2006;
124:577578.
49. International HapMap Consortium: A
haplotype map of the human genome.
Nature 2005; 437:12991320.
50. Hussein MR: Ultraviolet radiation and skin
cancer: molecular mechanisms. J Cutan
Pathol 2005; 32:191205.
51. Dryja TP, Morrow JF, Rapaport JM:
Quantication of the paternal allele bias for
new germline mutations in the
retinoblastoma gene. Hum Genet 1997;
100:446449.
52. Jadayel D, Fain P, Upadhyaya M, et al:
Paternal origin of new mutations in von
Recklinghausen neurobromatosis. Nature
1990; 343:558559.
CHAPTER 1
17
SECTION 1
18
GENETICS
53. Vogel F, Rathenberg R: Spontaneous
mutation in man. Adv Hum Genet 1975;
5:223.
54. Callinan PA, Feinberg AP: The emerging
science of epigenomics. Hum Mol Genet
2006; 15:R95R101.
55. Mertens D, Wolf S, Tschuch C, et al:
Allelic silencing at the tumor-suppressor
locus 13q14.3 suggests an epigenetic
tumor-suppressor mechanism.
Proc Natl Acad Sci USA 2006;
103:77417746.
56. Ferres-Marco D, Gutierrez-Garcia I,

Vallejo DM, et al: Epigenetic silencers and
Notch collaborate to promote malignant
tumours by Rb silencing. Nature 2006;
439:430436.
57. de Andrade AF, da Hora Barbosa R,
Vargas FR, et al: A molecular study of rst
and second RB1 mutational hits in
retinoblastoma patients. Cancer Genet
Cytogenet 2006; 167:4346.
58. Kantor B, Shemer R, Razin A: The
PraderWilli/Angelman imprinted domain
and its control center. Cytogenet Genome

Res 2006; 113:300305.
59. Fridman C, Hosomi N, Varela MC, et al:
Angelman syndrome associated with
oculocutaneous albinism due to an
intragenic deletion of the P gene. Am J
Med Genet A 2003; 119:180183.
CHAPTER
Molecular Mechanisms of Inherited Disease

Janey L. Wiggs
DNA mutations occurring in genes may result in the formation

of a defective gene product. If the normal protein product of a
mutated gene is necessary for a critical biologic function, then
an alteration of the normal phenotype may occur. Many changes
in phenotype are considered normal variations among humans,
for example, brown hair instead of blond hair. However, some
changes produce phenotypes that seriously affect health; these
are the major focus of study in clinical genetics laboratories.
The inheritance pattern of a disease is determined by the function of the normal and abnormal protein products of the gene as
well as the type of mutation causing the disease. For example,
mutations that create an abnormal protein that is detrimental
to cells are typically dominant, because only one mutant gene is
required to disrupt the normal functions of the cell. Mutations that
result in proteins with reduced biologic activity (loss of function)
may be inherited as dominant or recessive conditions depending
on the number of copies of normal genes (and the amount of normal protein) required. Disorders caused by mutations in mitochondrial DNA have a characteristic inheritance pattern, and
mutations in genes carried on the X chromosome also result in
typical inheritance patterns. Examples of the types of mutations
responsible for different inheritance patterns are described in
the following sections.
One of the goals of disease gene discovery is the development
of novel therapeutics. Disease treatment, including gene therapy, cannot be developed without knowledge about the underlying molecular mechanisms. Diseases that are caused by a loss
of protein function could be treated by protein and gene replacement therapies, while disease caused by a gain of function or
dominant negative effect would require inactivation of the
abnormal gene.
Aniridia-PAX6
Mutations in the PAX6 gene cause disease through haploinsufciency. Most of the mutations responsible for these disorders
alter the paired-box sequence within the protein product, which
is in the homeobox family of transcription factors (Fig. 2.1).1
The paired box is an important region of the protein that participates in the regulation of expression of other genes.2 PAX6 plays
a critical role in ocular development, presumably by regulating
the expression of a set of genes that are essential for this process.3
A reduction in the amount of active PAX6 gene product changes
the level at which these other genes operate.
There is extensive variation in the range of phenotypes exhibited by patients with PAX6 mutations. Patients typically have
various anterior segment abnormalities, such as aniridia,4 Peters
anomaly,5 or autosomal dominant keratitis.6,7 This spectrum of
phenotypic abnormalities resulting from mutations in one gene
is termed variable expressivity and is a common feature of disorders that result from haploinsufciency. The variability of the
mutant phenotype possibly results from the random activation
of downstream genes that occurs when only half the required
gene product is available.
Other examples of ocular disorders caused by haploinsufciency are: PITX2 causing AxenfeldRieger syndrome,8 LMX1B
causing nail patella syndrome and glaucoma,9 FOXC1 causing
anterior segment dysgenesis syndromes,10 SALL4 defects causing
Duane-radial ray syndrome,11 OPA1 causing autosomal dominant
(Kjers) optic atrophy,12 CRX causing conerod dystrophy,13 and
Waardenbergs syndrome caused by defects in another homeobox gene, PAX3.14 Of interest, the majority of these genes are
regulatory proteins involved in ocular developmental processes,
suggesting that gene dosage of regulatory proteins is an important
factor in eye development.
AUTOSOMAL DOMINANT DISORDERS
LOSS OF FUNCTION
Disorders inherited as autosomal dominant traits result from

mutations that occur in only one copy of a gene (i.e., in heterozygous individuals). Usually the parental origin of the mutation
does not matter. However, if the gene is subject to imprinting
(see further ahead), then mutations in the maternal or paternal
copy of the gene may give rise to different phenotypes.
Autosomal dominant traits may result from mutations in one

copy of a gene that increase the likelihood, but are not sufcient
to cause the disease. For the disease to become manifest, a second
hit that affects the remaining copy of the gene must occur. If the
second hit is a common event, the inheritance of one mutant
copy of the gene almost always results in the disease and the
trait appears to be inherited in a dominant fashion. However, at
the cellular level, the mutations appear recessive since cells
must be homozygotes or compound heterozygotes to display the
mutant phenotype.
HAPLOINSUFFICIENCY
Some cellular processes require a level of protein production that
can only be furnished if both copies of a particular gene are active.
Such proteins may be involved in a variety of biologic processes.
If one copy of a gene is mutant and the protein level is reduced
by half, a disorder may result.
Retinoblastoma
Tumor suppressor genes such as the retinoblastoma gene provide
good examples of loss-of-function dominant mutations. A gene
19
SECTION 1
GENETICS
FIGURE 2.1. Schematic diagram of the PAX6 gene.
responsible for retinoblastoma was identied in 1986 on chromosome 13q14.15 The gene product is involved in regulating the
cell cycle.16 An absence of this protein in a sensitive embryonic
retinal cell results in uncontrolled cell growth that eventually
produces a tumor. Susceptibility to hereditary retinoblastoma is
inherited as an autosomal dominant trait. Mutations in the
retinoblastoma gene result in underproduction of the protein
product or in production of an inactive protein product.17 A
retinal cell with only one mutant copy of the retinoblastoma gene
will not become a tumor. However, inactivation of the remaining
normal copy of the retinoblastoma gene is very likely in at least
one retinal cell out of the millions present in each retina. Most
individuals who inherit a mutant copy of the gene sustain a
second hit to the remaining normal copy of the gene and
develop the disease (Fig. 2.2).18
GAIN-OF-FUNCTION DOMINANT NEGATIVE

EFFECT
Autosomal dominant disorders can be caused by mutant proteins
that have a detrimental effect on the native tissue. Under this
scenario, mutations in one copy of a gene produce a mutant
protein that may interfere with normal cellular processes or may
accumulate as a toxic product, or both. This toxicity is a function not present in the wild-type protein; hence the mutation is
termed a gain-of-function mutant. If the mutant protein interferes with the function of the wild-type protein expressed by the
remaining normal copy of the gene, the mutation is described
as dominant negative.19 It is possible to have gain-of-function
mutations which can also be dominant negative because the new
function of the protein also interferes with the function of the

remaining normal copy of the gene.
Corneal Dystrophies
The autosomal dominant corneal dystrophies are excellent examples of gain-of-function mutations that result in the formation
of an aberrant protein. The four most common autosomal dominant corneal stromal dystrophies are: Groenouws (granular)
type 1,20 lattice type 1,21 Avellinos (combined granular lattice),22,23
and ReisBcklers.24 Although all four corneal dystrophies
affect the anterior stroma, the clinical and pathologic features
differ. The granular dystrophies typically form discrete white
localized deposits that progressively obscure vision. Histopathologically, these deposits stain bright red with Massons
trichrome and have been termed hyalin. In lattice dystrophy,
branching amyloid deposits gradually opacify the cornea. These
deposits exhibit a characteristic birefringence and dichroism
under polarized light after staining with Congo red. Avellinos
dystrophy has features of both granular and lattice dystrophies.
ReisBcklers primarily involves Bowmans layer and the
supercial stroma.24 All four dystrophies have been genetically
mapped to a common interval on chromosome 5q31.2528
Mutations in a single gene, TGFB1/BIGH3, have been identied in a number of affected families.29 An abnormal protein
product of this gene, keratoepithelin, accumulates in patients
carrying mutations. The normal protein product is probably an
extracellular matrix protein that modulates cell adhesion. Four
different missense mutations occurring at two arginine codons
in the gene have been found (Fig. 2.3). Interestingly, different
mutations at the same arginine codon cause lattice dystrophy
type I or Avellinos dystrophy, the two dystrophies characterized
by amyloid deposits. The mutations that cause Avellinos and
lattice dystrophies abolish a putative phosphorylation site that
is probably required for the normal structure of keratoepithelin.
Destruction of this aspect of the protein structure leads to the
formation of the amyloid deposits that cause opacication of
the cornea. As a result, the mutant protein is destructive to the
normal tissue. Mutations at the other arginine codon appear to
result in either granular dystrophy or ReisBcklers dystrophy.
The mutation analysis of this gene demonstrates that different
mutations within a single gene can result in different phenotypes.
Of interest, pathologic deposits caused by keratoepithelin accumulation have only been observed in the cornea and not in other
tissues or organs.30 Because the TGFB1/BIGH3 gene is expressed
in other tissues, these results suggest a cornea-specic mechanism causing the accumulation of mutant keratoepithelin.
Retinitis Pigmentosa Rhodopsin
20
FIGURE 2.2. Inheritance of retinoblastoma. Individuals inheriting a

mutation in the retinoblastoma gene are heterozygous for the mutation
in all cells of their bodies. The second hit to the remaining normal
copy of the gene occurs in a developing retinal cell and leads to
tumor formation.
Examples of gain-of-function mutations causing retinal degenerative disorders include: rhodopsin causing retinitis pigmentosa, transthyretin mutations causing vitreous amyloidosis,31
and possibly TIMP3 mutations causing Sorsbys dystrophy.32
Mutations in rhodopsin demonstrate how a gain-of-function
mechanism can cause a retinal degeneration. Mutations in the
gene for rhodopsin can cause retinitis pigmentosa.33 To explore the
pathogenic mechanisms relating to these mutations, transgenic
CHAPTER 2
FIGURE 2.3. Schematic diagram of the keratoepithelin gene. D1 to D4, homologous domains. Arrows point to the location of the reported
mutations.
mice were created that carried mutant copies of the gene.34 Histopathologic studies of these mice showed an accumulation of
vesicles containing rhodopsin at the junction between the inner
and the outer segments of the photoreceptors. The vesicles
probably interfere with the normal regeneration of the photoreceptors, causing photoreceptor degeneration.
Osteogenesis Imperfecta
Osteogenesis imperfecta is an example of a dominant negativetype mutation. Osteogenesis imperfecta is a group of inherited
disorders of type I collagen that predispose a patient to easy fracturing of bones, and skeletal deformity. Ocular ndings include
thinned sclera. The type I procollagen molecule is formed from
two proalpha-1 chains and one proalpha-2 chain. To create a
collagen molecule, the three chains form an a-helix beginning
at the carboxyl terminus. Mutations that affect the amino acid
sequence of an individual procollagen molecule disrupt the
formation of the helix, and this results in the disease.35
ANTICIPATION TRINUCLEOTIDE REPEATS

A new class of mutations responsible for autosomal dominant
inheritance was discovered with the identication of the gene
responsible for Huntingtons disease.36 Huntingtons disease is
a neurodegenerative disorder that results in motor, cognitive,
and emotional disturbance. Huntingtons disease demonstrates
anticipation, which means that subsequent generations of affected
individuals are more severely affected and are affected at an
earlier age than their predecessors.37 The gene defect responsible
for this disease is an expanded and unstable trinucleotide repeat in
the open-reading frame of the Huntington disease gene located on
chromosome 4. The repeated DNA sequence causes the encoded
protein to have a long span of the same amino acid residue
repeated many times. A critical observation was made when the
repeat lengths were correlated with the severity and the age of
onset of the disease. Longer repeat lengths result in more severe
disease at an earlier age of onset. The number of repeats within
the gene expands with each subsequent generation and is likely to
be the cause of the increased severity of the disease (Fig. 2.4).38
Since the discovery of the Huntington gene, a number of other
disorders caused by unstable trinucleotide repeats have been recognized, including myotonic dystrophy,39 spinocerebellar ataxia,40
Friedreichs ataxia,41 and fragile X syndrome.42 Although the
specic mechanisms responsible for trinucleotide repeat disease
are not completely understood, the autosomal dominant inheritance suggests that only one mutant copy of the gene is required
and that the repeat in some way has a detrimental effect on the
cell. This molecular mechanism should be considered whenever
FIGURE 2.4. Pedigree illustrating anticipation associated with

expansion of a trinucleotide repeat. Affected individuals are shown as
solid circles or squares. The age of onset of the disease is shown
beneath the pedigree symbol for each affected individual. The number
of trinucleotide repeats within the disease gene (e.g., the gene
responsible for Huntingtons disease) is schematically represented
beneath each affected individual. Successive generations have an
earlier age of onset and a higher number of repeats (compare
individual one with individual six).
pedigree analysis shows increased disease severity with each

new generation.
IMPRINTING
Some mutations give rise to autosomal dominant traits that are
transmitted by parents of either sex, but they are expressed only
when inherited from a parent of one particular sex. In families
affected with these disorders they would appear to be transmitted
in an autosomal dominant pattern from one parent (either the
mother or the father) would not be transmitted from the other
parent. Figure 2.5 provides an example of a trait that is expressed
only when transmitted from the father. Occasionally the same
mutation gives rise to a different disorder, depending on the sex
21
SECTION 1
GENETICS
Paternal imprinting
Autosomal dominant
1
II
IV
II
M
III
10
11
10
III
IV
12
14
15
10
13
16
FIGURE 2.5. Pedigree illustrating paternal imprinting compared with segregation of an autosomal dominant trait that is not imprinted. Affected
individuals are shown as solid circles or squares. Those individuals carrying a mutation are indicated by the M beneath the pedigree gure.
Notice that in the pedigree transmitting the mutation as an autosomal dominant trait, all individuals carrying the mutation are affected, while in
the paternally imprinted pedigree, only individuals who have inherited the mutation from their father are affected. Individuals can inherit the
mutation from the mother, but in that case it is not expressed and they are phenotypically normal. These mutation carriers can, however, transmit
the mutation to their offspring, and the offspring who inherit the mutation from male mutation carriers will be affected.
of the parent transmitting the trait. These parental sex effects

are evidence of a phenomenon called imprinting. Although the
molecular mechanisms responsible for imprinting are not completely understood, it appears to be associated with DNA methylation patterns that can mark certain genes with their parental
origin.43 PraderWilli syndrome and Angelman syndromes are
examples of imprinted conditions.44 Diseases caused by mutations in imprinted genes can give rise to unusual inheritance
patterns (Fig. 2.5).
AUTOSOMAL RECESSIVE DISORDERS

Autosomal recessive disorders result from mutations present on
both the maternal and the paternal copies of a gene. Mutations
responsible for recessive disease typically cause a loss of biologic
activity, either because they create a defective protein product that
has little or no biologic activity or because they interfere with
the normal expression of the gene (regulatory mutations). Most
individuals heterozygous for autosomal recessive disorders are
clinically normal.
LOSS OF FUNCTION (Albinism)
22
Autosomal recessive diseases often result from defects in enzymatic proteins. Albinism is the result of a series of defects in the
synthesis of melanin pigment.45 Melanin is synthesized from
the amino acid tyrosine, which is rst converted to dihydroxyphenylalanine through the action of the copper-containing
enzyme tyrosinase. An absence of tyrosinase results in one form
of albinism. Mutations in the gene coding for tyrosinase are
responsible for this disease cluster in the binding sites for copper,
disrupting the metal ionprotein interaction necessary for enzyme
function.46 Both copies of the gene for tyrosinase must be mutated
before a signicant interruption of melanin production occurs.
Heterozygous individuals do not have a clinically apparent pheno-
type, suggesting that one functional copy of the gene produces

sufcient active enzymes that the melanin level is phenotypically
normal.
X-LINKED RECESSIVE DISORDERS

X-linked recessive disorders, like autosomal recessive disorders,
result from a mutant gene that causes a loss of a critical biologic
activity. Because males have only one X chromosome, one mutant
copy of a gene responsible for an X-linked trait results in the
disease. Usually females are heterozygous carriers of recessive
X-linked traits. In somatic cells of females, only one X chromosome is active; the second X chromosome is inactivated and
becomes a Barr body. X inactivation has been associated with the
geneticist Mary Lyon, and has been called Lyonization. Inactivation of either the maternal or the paternal X chromosome
occurs early in embryonic life. In any one cell, the inactive X may
be maternal or paternal, and once the X is inactivated, it remains
inactive. Because females inherit two copies of the X chromosome, they can be homozygous for a disease allele at a given
locus, heterozygous, or homozygous for the normal allele at the
locus. Since only one X chromosome is active in any given
somatic cell, about half the cells of a heterozygous female express
the disease allele, and about half express the normal allele. Like
autosomal recessive traits, the female heterozygote expresses
~50% of the normal level of the protein product. For recessive
conditions, this is sufcient for a normal phenotype.
Retinoschisis
Retinoschisis is a maculopathy that is caused by intraretinal splitting. The defect most likely involves retinal Mllers cells.47
Retinoschisis is inherited as an X-linked recessive trait.48 Female
carriers with one normal and one abnormal copy of the gene do
not demonstrate any clinical abnormalities. Fifty percent of the
male offspring of female carriers are affected by the disease.
Mutations in a gene located in the retinoschisis interval and

expressed in the retina have been found in a protein that is implicated in cellcell interaction and may be active in cell adhesion
processes during retinal development. Mutational analysis of
the retinoschisis gene (XLRS1) in affected individuals from nine
unrelated families showed one nonsense, one frame shift, one
splice acceptor, and six missense mutations.49 Presumably these
mutations all result in an inactive protein product.
X-LINKED DOMINANT DISORDERS

X-linked dominant mutations are less common than X-linked
recessive mutations. Clinically, X-linked dominant inheritance
is difcult to recognize because of the random inactivation of the
X chromosome in females (Lyons hypothesis).50 The random
inactivation of the X chromosome produces females who are X
chromosome mosaics, with ~50% of the cells expressing genes
from the paternally derived X and 50% of the cells expressing
genes from the maternally derived X. If one of the X chromosomes
has a mutant gene, these cells may display the phenotype; however, 50% of the female cells are normal, even for a dominant
mutation. As a result, for recessive and dominant X-linked traits,
the disease phenotype may not be evident in females carrying the
mutation. X-linked dominant mutations could produce a protein that has a detrimental effect on normal biologic processes
(gain-of-function or dominant negative effect). Mutations that
result in haploinsufciency of the X chromosome could also be
X-linked dominant. X-linked dominant disorders include incontinentia pigmenti and X-linked hypophosphatemia rickets. A
family with X-linked dominant retinitis pigmentosa has also
been described.51
DIGENIC INHERITANCE
Digenic inheritance describes a pattern of inheritance that is
similar to recessive inheritance, except that the trait only develops
when mutations are found in one copy of each of the two independent genes simultaneously. In recessive disorders the mutations are found in both copies of one gene. Digenic inheritance
is an example of the complex interactions that occur between multiple gene products in polygenic inheritance (see further ahead).
BBS genes.55 In some BBS pedigrees, affected individuals carry

three mutations in one or two BBS genes. In these pedigrees
unaffected individuals only had two abnormal alleles.56 In some
families it has been proposed that BBS may not be a single-gene
recessive disease but a complex trait requiring at least three
mutant alleles to manifest the phenotype. This would be an
example of triallelic inheritance.57
CHAPTER 2
MITOCHONDRIAL DISORDERS
Mutations in mitochondrial DNA can also result in human disease. The characteristic segregation and assortment of Mendelian
disorders depends on the meiotic division of chromosomes found
in the nucleus of cells. There are several hundred mitochondria
in a cell, and each mitochondrion contains several copies of the
mitochondrial genome. Mitochondria divide in the cellular
cytoplasm by simple ssion. Not all mitochondria present in a
disease tissue carry DNA mutations. During cell division, mitochondria and other cytoplasmic organelles are arbitrarily distributed to the daughter cells. Because each cell contains a
population of mitochondrial DNA molecules, a single cell can
contain DNA molecules that are normal as well as DNA molecules that are mutant (Fig. 2.6). This heterogeneity of DNA
composition, called heteroplasmy, is an important cause of
variable expression in mitochondrial diseases. As the diseased
mitochondria are distributed to developing tissues, some tissues
accumulate more abnormal mitochondria than others.
Disorders that result from mutations in mitochondrial DNA
demonstrate a maternal inheritance pattern (see also Chapter 1).
Maternal inheritance differs from Mendelian inheritance in that
only affected females transmit the disease to their offspring.
Unlike nuclear DNA that is equally contributed to the embryo
by the mother and the father, mitochondria and mitochondrial
DNA are derived solely from the maternal egg. A mutation
occurring in mitochondrial DNA is present in cells containing
mitochondria, including the female gametes. Sperm have few
mitochondria, and they are not transmitted to the egg. A male
Retinitis Pigmentosa Peripherin and ROM1

At least one form of retinitis pigmentosa is inherited as a digenic
trait.52 In pedigrees demonstrating digenic inheritance there is
direct parent-to-child transmission of the disease; however, affected
families have unusual features for a dominantly inherited disease:
the disease originates in the offspring of an ancestral mating
between two unaffected individuals, and the affected individuals
transmitted the disease to less than 50% of their offspring
(~25% rather than 50%). In some retinitis pigmentosa families,
mutation analysis of the peripherin gene and the ROM1 gene
showed that the affected individuals had specic mutations in
both genes. Individuals who had a mutation in one copy of either
gene were unaffected by the disease. Mutant copies of ROM1
and peripherin can also cause autosomal dominant forms of
retinitis pigmentosa.53,54 These results suggest that some mutant
forms of peripherin and ROM1 cause retinitis pigmentosa in a
digenic pattern, whereas other mutations can independently
cause autosomal dominant forms of the disease.
BardetBiedl Syndrome
BardetBiedl syndrome (BBS) is a genetically heterogeneous disorder characterized by multiple clinical features that include
pigmentary retinal dystrophy, polydactyly, obesity, developmental
delay, and renal defects. BBS is considered an autosomal recessive
disorder, and positional cloning efforts have identied eleven
FIGURE 2.6. Heteroplasmy in mitochondria. Daughter cells resulting

from the division of a cell containing mitochondria with mutant DNA
may contain unequal numbers of mutant mitochondria. Subsequent
divisions lead to a population of cells with varied numbers of normal
and abnormal mitochondria.
23
SECTION 1
GENETICS
carrying a mitochondrial DNA mutation will not transmit the
disease to his offspring.
Lebers Hereditary Optic Neuropathy

Lebers hereditary optic neuropathy was one of the rst diseases
to be recognized as a mitochondrial DNA disorder.58 For some time
clinicians had observed maternal inheritance of this condition
in affected families, but it wasnt until mutations in mitochondrial DNA of affected individuals were demonstrated that
the cause of the inheritance pattern was understood. In familial
cases of the disease, all affected individuals are related through
the maternal lineage, consistent with the inheritance of human
mitochondrial DNA.
Patients affected by Lebers hereditary optic neuropathy
typically present with acute or subacute, painless, central vision
loss leading to a permanent central scotoma and loss of sight.
The manifestation of the disease can vary tremendously especially
with respect to the onset of loss of vision and severity of the
outcome.59 The eyes can be affected simultaneously or sequentially. The vision may be lost rapidly over a period of weeks to
months, or slowly over several years. Within a family the disease
may also vary among affected family members. Several factors
contribute to the variable phenotype of this condition. Certain
mutations are associated with more severe disease. For example,
the most severely affected patients with the 11 778-bp mutation
may have no light perception,60 whereas the most severely
affected patients with the 3460-bp mutation may retain light
perception.61 Another important factor that affects the severity
of the disease in affected persons is the heteroplasmic distribution of mutant and normal mitochondria. This partially explains
why some patients develop a more severe optic neuropathy. Other
genetic or environmental factors are likely to play a role as well.
POLYGENIC INHERITANCE
susceptible to a disease, and other genes or environmental conditions may influence the full expression of the phenotype.
Secondary genes responsible for the modulation of the expression of a specic genetic mutation are called modier genes;
modier genes may be inherited completely independently from
the gene directly responsible for the disease trait. For example,
recent evidence suggests that WDR36, a gene associated with
glaucoma but not sufcient to cause glaucoma, is a modier
gene that contributes to the severity of the glaucoma phenotype
in individuals carrying a WDR36 variant in addition to another
glaucoma gene.62 Not every individual who inherits a mutation
partly responsible for a complex trait also inherits the set of
modier genes that is required for full expression of the disease.
The digenic inheritance of retinitis pigmentosa seen by certain
mutant alleles of peripherin and ROM1 is an example of the
simplest form of polygenic inheritance (see previous discussion).
Certain conditions may require multiple genes or a combination
of different genes and environmental conditions to be manifest.
In addition to adult-onset primary open-angle glaucoma, examples of ocular disorders that are multifactorial are age-related
macular degeneration, and myopia.63
Key Features
Human phenotypes inherited as polygenic or complex traits do

not follow the typical patterns of Mendelian inheritance. Generally, complex traits are commonly found in the human population. Multiple genes are likely to contribute to the expression
of the disease phenotype. Some genes may render an individual
Disease treatment, including gene therapy, cannot be

developed without knowledge about the underlying molecular
mechanisms that are responsible for the disease.
Autosomal dominant disorders result from one abnormal copy
of a gene; the defect may cause a loss of protein function, or a
gain of a novel detrimental function.
Autosomal recessive disorders are caused by abnormalities in
both copies of a gene. The defective gene copies usually result
in loss of protein function.
Digenic inheritance describes a pattern of inheritance that is
similar to recessive inheritance except that the trait only
develops when mutations are found in one copy of each of the
two independent genes simultaneously.
Disorders that result from mutations in mitochondrial DNA
demonstrate a maternal inheritance pattern.
REFERENCES
24
1. Read AP: Pax genes-Paired feet in three

camps. Nature Genet 1995; 9:333334.
2. Ton CCT, Hirvonen H, Mira H, et al:
Positional cloning and characterization of a
paired box- and homeobox-containing
gene from the aniridia region. Cell 1991;
67:10591074.
3. Richardson J, Cvekl A, Wistow G: Pax-6 is
essential for lens-specic expression of
zeta-crystallin. Proc Natl Acad Sci USA
1995; 92:46764680.
4. Glaser T, Jepeal L, Edwards JG, et al: PAX6
gene dosage effect in a family with
congenital cataracts, aniridia, anophthalmia
and central nervous system defects. Nature
Genet 1994; 7:463471.
5. Hanson IM, Fletcher JM, Jordon T, et al:
Mutations at the PAX6 locus are found in
heterogeneous anterior segment
malformations including Peters anomaly.
Nature Genet 1994; 6:168173.
6. Mirzayans F, Pearce WG, MacDonald IM, et
al: Mutation of the PAX6 gene in patients
with autosomal dominant keratitis. Am J
Hum Genet 1995; 57:539548.
7. Davis A, Cowell JK: Mutations in the PAX6
gene in patients with hereditary aniridia.
Hum Mol Genet 1993; 2:20932097.
8. Hjalt TA, Semina EV: Current molecular

understanding of AxenfeldRieger syndrome.
Expert Rev Mol Med 2005; 7:117.
9. Sato U, Kitanaka S, Sekine T, et al:
Functional characterization of LMX1B
mutations associated with nail-patella
syndrome. Pediatr Res 2005; 57:783788.
10. Zhang HZ, Li P, Wang D, et al: FOXC1 gene
deletion is associated with eye anomalies
in ring chromosome 6. Am J Med Genet A
2004; 124:280287.
11. Kohlhase J, Chitayat D, Kotzot D, et al:
SALL4 mutations in Okihiro syndrome
(Duane-radial ray syndrome), acro-renalocular syndrome, and related disorders.
Hum Mutat 2005; 26:176183.
12. Ferre M, Amati-Bonneau P, Tourmen Y, et
al: eOPA1: an online database for OPA1
mutations. Hum Mutat 2005; 25:423428.
13. Freund CL, Gregory-Evans CY, Furukawa T,
et al: Cone-rod dystrophy due to mutations
in a novel photoreceptor-specic
homeobox gene (CRX) essential for
maintenance of the photoreceptor. Cell
1997; 91:543553.
14. Tassabehji M, Read AP, Newton VE, et al:
Waardenburgs syndrome patients have
mutations in the human homologue of the
15.
16.
17.
18.
19.
20.
21.
22.
Pax-3 paired box gene. Nature 1992;

355:635636.
Friend SH, Dryja TP, Weinberg RA:
Oncogenes and tumor-suppressing genes.
N Engl J Med 1988; 318:618622.
Weinberg RA: The retinoblastoma protein
and cell cycle control. Cell 1995; 81:323330.
Dryja TP, Cavenee W, White R, et al:
Homozygosity of chromosome 13 in
retinoblastoma. N Engl J Med 1984;
310:550553.
Knudson AG Jr: Genetics of human cancer.
Annu Rev Genet 1986; 20:231251.
Herskowitz I: Functional inactivation of
genes by dominant negative mutations
[review]. Nature 1987; 329:219322.
Moller HU: Inter-familial variability and
intra-familial similarities of granular corneal
dystrophy Groenouw type I with respect to
biomicroscopical appearance and
symptomatology. Act Ophthalmol 1989;
67:669677.
Klintworth GK: Lattice corneal dystrophy:
an inherited variety of amyloidosis
restricted to the cornea. Am J Pathol 1967;
50:371399.
Folberg R, Alfonso E, Croxatto JO, et al:
Clinically atypical granular corneal
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
dystrophy with pathologic features of

lattice-like amyloid deposits.
Ophthalmology 1988; 95:4651.
Rosenwasser GO, Sucheski BM, Rosa N,
et al: Phenotypic variation in combined
granular-lattice (Avellino) corneal dystrophy.
Arch Ophthalmol 1993; 111:15461552.
Kuchle M, Green WR, Volcker HE, et al:
Reevaluation of corneal dystrophies of
Bowmans layer and the anterior stroma
(ReisBcklers and ThielBehnke types): a
light and electron microscopic study of
eight corneas and a review of the literature.
Cornea 1995; 14:333354.
Eiberg H, Moller HU, Berendt I, et al:
Assignment of granular corneal dystrophy
Groenouw type I locus to within a 2 cM
interval. Eur J Hum Genet 1994; 2:132138.
Stone EM, Mathers WD, Rosenwasser GO,
et al: Three autosomal dominant corneal
dystrophies map to chromosome 5q.
Nature Genet 1994; 6:4751.
Gregory CY, Evans K, Bhattacharya SS:
Genetic renement of the chromosome 5q
lattice corneal dystrophy to within a 2 cM
interval. J Med Genet 1995; 32:224226.
Small KW, Mullen L, Barletta J, et al:
Mapping of ReisBcklers corneal
dystrophy to chromosome 5q. Am J
Ophthalmol 1996; 121:384390.
Munier FL, Korvatska E, Djemai A, et al:
Kerato-epithelin mutations in four 5q31linked corneal dystrophies. Nature Genet
1997; 15:247251.
El Kochairi I, Letovanec I, Uffer S, et al:
Systemic investigation of keratoepithelin
deposits in TGFBI/BIGH3-related corneal
dystrophy. Mol Vis 2006; 12:461466.
Sandgren O, Holmgren G, Lundgren E:
Vitreous amyloidosis associated with
homozygosity for the transthyretin
methionine-30 gene. Arch Ophthalmol
1990; 108:15841586.
Felbor U, Suvanto EA, Forsius HR, et al:
Autosomal recessive Sorsby fundus
dystrophy revisited: molecular evidence for
dominant inheritance. Am J Hum Genet
1997; 60:5762.
Dryja TP, McGee TL, Reichel E, et al: A
point mutation of the rhodopsin gene in
one form of retinitis pigmentosa. Nature
1990; 343:364366.
Li T, Snyder WK, Olsson JE, et al:
Transgenic mice carrying the dominant
rhodopsin mutation P347S: evidence for
defective vectorial transport of rhodopsin
to the outer segments. Proc Natl Acad Sci
USA 1996; 93:1417614181.
Stacey A, Bateman J, Choi T, et al:
Perinatal lethal osteogenesis imperfecta in
transgenic mice bearing an engineered
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
mutant pro-alpha 1(I) collagen gene. Nature

1988; 332:131136.
Richards RI, Sutherland GR: Dynamic
mutations: a new class of mutations
causing human disease. Cell 1992;
70:709712.
Myers RH, Madden JJ, Teague JL, et al:
Factors related to onset age of Huntington
disease. Am J Hum Genet 1982;
34:481488.
Ranen NG, Stine OC, Abbott MH, et al:
Anticipation and instability of IT-15 (CAG)n
repeats in parent-offspring pairs with
Huntington disease. Am J Hum Genet
1995; 57:593602.
Mooers BH, Logue JS, Berglund JA: The
structural basis of myotonic dystrophy from
the crystal structure of CUG repeats. Proc
Natl Acad Sci USA 2005; 102:1662616631.
Abou-Sleymane G, Chalmel F, Helmlinger
D, et al: Polyglutamine expansion causes
neurodegeneration by altering the neuronal
differentiation program. Hum Mol Genet
2006; 15:691703.
Clark RM, Bhaskar SS, Miyahara M, et al:
Expansion of GAA trinucleotide repeats in
mammals. Genomics 2006; 87:5767.
Ranum LP, Day JW: Pathogenic RNA
repeats: an expanding role in genetic
disease. Trends Genet 2004; 20:506512.
Lewis A, Reik W: How imprinting centres
work. Cytogenet Genome Res 2006;
113:8189.
Kantor B, Shemer R, Razin A: The
PraderWilli/Angelman imprinted domain
and its control center. Cytogenet Genome
Res 2006; 113:300305.
Spritz RA: Molecular genetics of
oculocutaneous albinism. Hum Mol Genet
1994; 3:14691475.
Spritz RA, Strunk K, Giebel LB, et al:
Detection of mutations in the tyrosinase
gene in a patient with type IA
oculocutaneous albinism. N Engl J Med
1990; 322:17241728.
Yanoff M, Kertesz Rahn E, Zimmerman LE:
Histopathology of juvenile retinoschisis.
Pawar H, Bingham EL, Lunetta KL, et al:
Rened genetic mapping of juvenile Xlinked retinoschisis. Hum Hered 1995;
45:206210.
Sauer CG, Gehrig A, Warneke-Wittstock R,
et al: Positional cloning of the gene
associated with X-linked juvenile
retinoschisis. Nature Genet 1997;
17:164170.
Krill AE: X-chromosome-linked disease
affecting the eye: status of the
heterozygote female. Trans Am Ophthalmol
Soc 1969; 67:535608.
51. McGuire RE, Sullivan LS, Blanton SH, et al:

X-linked dominate cone-rod degeneration:
linkage mapping of a new locus for retinitis
pigmentosa (RP15) to Xp22.13p22.11. Am
J Hum Genet 1995; 57:8794.
52. Kajiwara K, Berson EL, Dryja TP: Digenic
retinitis pigmentosa due to mutations at the
unlinked peripherin/RDS and ROM1 loci.
Science 1994; 264:16041608.
53. Kajiwara K, Hahn LB, Mukai S, et al:
Mutations in the human retinal
degeneration slow gene in autosomal
dominant retinitis pigmentosa. Nature 1991;
354:480483.
54. Bascom RA, Schappert K, NcInnes RR:
Cloning of the human and murine ROM1
genes: genomic organization and sequence
conservation. Hum Mol Genet 1993;
2:385391.
55. Chiang AP, Beck JS, Yen HJ, et al:
Homozygosity mapping with SNP arrays
identies TRIM32, an E3 ubiquitin ligase,
as a Bardet-Biedl syndrome gene
(BBS11). Proc Natl Acad Sci USA 2006;
103:62876292.
56. Beales PL, Badano JL, Ross AJ, et al:
Genetic interaction of BBS1 mutations with
alleles at other BBS loci can result in nonMendelian BardetBiedl syndrome. Am J
Hum Genet 2003; 72:11871199.
57. Eichers ER, Lewis RA, Katsanis N, Lupski
JR: Triallelic inheritance: a bridge between
Mendelian and multifactorial traits. Ann
Med 2004; 36:262272.
58. Wallace DC, Singh G, Lott MT, et al:
Mitochondrial DNA mutation associated
with Lebers hereditary optic neuropathy.
Science 1988; 242:14271430.
59. Brown MD, Voljavec AS, Lott MT, et al:
Lebers hereditary optic neuropathy; a
model for mitochondrial neurodegenerative
diseases. FASEB J 1992; 6:27912799.
60. Johns DR, Smith KH, Savino PJ, et al:
Lebers hereditary optic neuropathy.
Clinical manifestations of the 15257
mutation. Arch Ophthalmol 1993;
110:981986.
61. Johns DR, Smith KH, Miller NR: Lebers
hereditary optic neuropathy. Clinical
manifestations of the 3460 mutation. Arch
Ophthalmol 1992; 110:15771581.
62. Hauser MA, Allingham RR, Linkroum K, et
al: Distribution of WDR36 DNA sequence
variants in patients with primary openangle glaucoma. Invest Ophthalmol Vis Sci
2006; 47:25422546.
63. Wiggs JL: Complex disorders in
ophthalmology. Sem Ophthalmol 1995;
10:323330.
CHAPTER 2
25
CHAPTER
Genetic Testing
Janey L. Wiggs
The identication of genes responsible for inherited ocular disorders makes it possible to perform genetic testing for diseaseassociated mutations that can help determine the clinical
diagnosis and prognosis. For some diseases, genetic testing can
serve as a screening tool to identify individuals at risk before the
clinical symptoms of the disease are manifest. The amount of
information provided by a genetic test and the methods used for
testing depend on what information is known about the gene(s)
involved. If the gene has been identied then direct genetic testing
can be performed, if only the location of the gene is known then
an indirect testing approach is used.
Direct testing uses a biological sample from the patient to
prepare DNA, RNA, or protein, to test for a specic alteration.
Typically, DNA or RNA is evaluated to determine if a specic
sequence change, or genotype, causing the disease is present in
the patients sample. Protein samples can be analyzed for specic
amino acid changes. Direct testing requires only a biologic sample
from the patient; however, detailed knowledge about the gene,
including the gene structure and the normal gene sequence, is
required.
Indirect testing uses family analysis to detect copies of the chromosome that contains the mutant form of a disease-causing gene.
In this approach, DNA samples from all family members (affected
and unaffected) are analyzed for genetic markers that are known
to be located near the disease gene. The advantage of this approach
is that specic knowledge of the disease gene is not required. The
disadvantage is that multiple family members need to be tested.
Figure 3.1 shows a flow diagram outlining the protocol for
clinical genetic testing. The evaluation begins with a patient with
phenotypic characteristics of a disease (clinical ndings, imaging
studies, laboratory studies) who presents to a physician. The clinical evaluation may suggest a diagnosis that could be conrmed
by genetic testing. The rst step is to determine if there is a family
history of the disease that would support a Mendelian inheritance
pattern (autosomal dominant, autosomal recessive, X-linked dominant, X-linked recessive). If Mendelian inheritance is supported
by the family history, the next step is to determine if a gene has
been genetically mapped to a chromosomal region (genetic locus),
and if the gene has been identied within the locus. If the gene
has been mapped but not identied, indirect testing can be
performed using all available family members. If the gene has
been identied, and the gene sequence is known, the gene can
be screened using direct testing for disease-causing mutations.
If Mendelian inheritance is not supported by the family history,
the next step is to determine if there is a maternal inheritance
pattern that would support a diagnosis of a mitochondrial DNA
disorder. If the disease affects both male and female offspring and
is only transmitted by an affected mother, then mitochondrial
DNA screening should be considered. In the absence of Mendelian
inheritance or maternal inheritance, a diagnosis of a complex

genetic trait remains a possibility and screening of genetic risk
factors associated with the trait should be considered. If genetic
risk factors have not been identied, then genetic counseling
focused on risk avoidance (environmental exposures) and risk of
familial recurrences can be provided. In all cases, genetic counseling can help the physician and patient understand the genetic
risks associated with the disease.
DIRECT TESTING APPROACHES AND

METHODS
The optimal, though not always practical or possible, method of
laboratory genetic diagnosis is to test a persons gene or gene
product directly to determine if the sequence is normal or mutant.
Direct genetic testing can only be performed if the gene(s) responsible for a disease have been identied and the normal sequence
is known. Most of the methods used for direct testing are
dependent on the polymerase chain reaction (PCR) (Fig. 3.2).
This enzymatic procedure makes many copies of the DNA (or
RNA) that will be used for genetic analysis.1 For PCR short
oligonucleotide segments (usually 2030 bp in length) are synthesized to match the normal DNA sequence that flanks the
DNA region of interest (usually an exon of a gene). The patient
DNA is denatured into two single strands and the synthetic
oligonucleotides are allowed to hybridize. A thermoresistant
version of DNA polymerase is added to the reaction which adds
a new DNA strand from the end of each of the two oligonucleotides
flanking the region of interest, thus making two copies of the
DNA segment to be tested. The process is repeated 3050 times
resulting in an exponential expansion of the number of copies
of the desired DNA segment. The copied DNA segment can be
puried and used for additional tests to detect mutations. Typically, screening an entire gene is done by selectively amplifying
each gene exon followed by further analysis after purication of
the amplication products. Since PCR is the fundamental step
for direct genetic testing, PCR artifacts or reaction failures can
result in testing errors (see further ahead).
For direct testing, a biological sample needs to be obtained
from the patient. Family members may be included, but are not
necessary for the testing. Blood samples are the most widely used
source of DNA from adults, and yield more DNA than other
sources. For children or individuals not comfortable with blood
drawing, mouthwash samples or buccal swabs can be used.2 These
samples yield sufcient DNA for initial screening of a typical
gene. If more tests are required, or if patient resampling is difcult, then whole genome amplication can be used to make many
copies of the patient DNA sample before selectively amplifying
regions of the DNA for testing.3 Occasionally, direct testing is
27
SECTION 1
GENETICS
FIGURE 3.1. Decision flow diagram for genetic
testing.
Clinical evaluation
Laboratory tests
Imaging studies
Phenotype
Mendelian Inheritance
Maternal
Inheritance
Genetic
Risk Factor
NO
Risk
Avoidance
NO
YES
NO
YES
YES
Genetic Loci
Identified
NO
Screen
Mitochondrial
DNA
YES
Recurrence
Risks
Gene
Identified
NO
Indirect
Testing
Genetic
Test
YES
Direct
Testing
FIGURE 3.2. PCR. A DNA sample is heated to

produce single-stranded DNA which is then
allowed to be hybridized with an excess of
short oligonucleotide primers. Taq DNA
Singlestranded DNA polymerase is added and DNA synthesis
proceeds elongating the primers to full-length
strands. The newly synthesized doublestranded DNA is heated again, and the cycle
repeats. At the end of the second cycle, four
double-stranded copies have been formed.
Cycles are repeated 3050 times to generate
sufficient DNA for further studies.
Double-stranded DNA
Heat and denature
Singlestranded DNA
+ oligonucleotides
Taq DNA polymerase

DNA synthesis
Heat and denature
+ oligonucleotides
Taq DNA polymerase

DNA synthesis
28
performed after a patient is deceased on material obtained from

archived pathology specimens4 or from hair recovered from a
hair brush.5
Genetic testing can be performed using DNA, RNA, or protein.
Of these, DNA is the easiest to purify and analyze. RNA in the
form of an RT-PCR product may be preferable for a large gene
such as retinoblastoma (Fig. 3.3).6 However, RNA is less stable
than DNA, and samples must be processed rapidly and under
specic conditions to avoid degradation. RNA expression in
accessible tissues may be a problem, and the mutant form of the
RNA may not be stable in vivo and may not be recovered in the
sample to be assayed. Protein assays can determine if a mutation
exists and if the mutation interferes with the protein function.
Ideally, the protein function information would be useful for all
genetic tests; however, proteins are far more difcult to purify and
assay for activity than DNA. For example, mutations in myocilin,
a gene responsible for some forms of early onset glaucoma, cause
the protein to be insoluble in an in vitro assay.7 However, to
perform this assay on patients would require access to disease
tissue and purication of the protein product. Information about

the gene mutation can be gained from purifying the DNA from
a blood or mouthwash sample and identifying the mutation
responsible for the abnormal protein. If the mutation can be
linked to abnormal protein function (using other information
such as this in vitro assay), then the same information has been
gained. If protein function information is not available for a
specic mutation, then it would be necessary to validate the mutation in other ways. Despite the attraction of accessing the protein abnormality directly, for routine screening it is more efcient
to purify a DNA sample from the patient and identify the DNA
sequence change that causes the mutation, than to purify and
characterize the abnormal protein product.
For some diseases, affected individuals may carry the same
mutation, or one of a small number of mutations associated with
the disease. For example, most patients affected by Lebers hereditary optic neuropathy have one of three mutations.8 Hence,
for patients who are suspected of having a maternally inherited
optic neuropathy, testing would focus rst on these three
Gene
mRNA
E x on 1
E x on 2
Intron 1
Exon 1
Exon 2
Add reverse transcriptase

and oligonucleotide
Reverse transcriptase
extends oligonucleotide to
make a DNA copy of the mRNA
FIGURE 3.3. RT-PCR. mRNA is purified from

the patient, and is the starting material for PCR
rather than DNA. The first step is to make a
DNA copy of the mRNA using reverse
transcriptase and an oligonucleotide primer that
matches the target sequence. After
synthesizing a DNA copy, RNase H is used to
remove the original mRNA. Next DNA
polymerase and another oligo marking the
other end of the target sequence are used to
make a double-stranded DNA. The target
double-stranded DNA can then be used for
PCR as described in Figure 3.2.
CHAPTER 3
Genetic Testing
RNase H
RNase H removes mRNA

leaving singlestranded DNA
DNA polymerase
Add DNA polymerase
and oligonucleotide
DNA polymerase
DNA polymerase synthesizes

second strand DNA
Amplify using PCR
mutations. Such mutation redundancy among a population of

affected individuals may be the result of a hot spot in the gene
for mutations, a dependency of the disease on a specic type of
abnormality in the protein product caused by only a few mutations, or a founder effect caused by a limited number of original
mutations. For some CYP1B1 gene mutations causing congenital
glaucoma, founder chromosomes have been identied,9 and the
mutations located on these chromosomes have been found in
multiple populations indicating the ancient distributions of the
original mutations. Approximately 50% of mutations in the
BIGH3 gene responsible for dominant corneal dystrophies involve
two sites in the gene, identifying these regions as mutation hot
spots.10 Generally, for disorders that are caused by a limited
number of mutations, those mutations are tested for initially,
and if the common mutations are not found then the entire
causative gene is screened.
METHODS FOR DIRECT MUTATION

TESTING
TESTING FOR A KNOWN MUTATION
Testing for a specic mutation can be done when there are a limited
number of mutations that have been associated with a disease,
or for diagnosis within a family when screening the entire causative gene has already dened a mutation in one family member.
Methods to test for a known mutation are simpler and less time
consuming than methods used to screen the entire gene. All of
the methods used to test for known mutations depend on PCR
amplication of a portion of the causative gene, followed by a
DNA sequence-dependent assay. There are many such assays
including: testing for the presence or absence of a restriction
enzyme site,11 allele-specic oligonucleotide hybridization,12 allelespecic PCR amplication,13 oligonucleotide ligation assay,14 and
more recently quantitative PCR approaches using TaqMan or
related fluorescer-quencher methods.15 The general principles of
these direct methods are illustrated by a more detailed description of the TaqMan assay (Fig. 3.4).
The TaqMan assay uses quantitative PCR to identify alleles at

a selected single nucleotide variation. Single nucleotide changes
can be missense or nonsense mutations or may be polymorphisms called single-nucleotide polymorphisms (SNPs).16 In the
TaqMan assay, a specic probe of 2030 bp is designed to
hybridize specically with the DNA sequence of interest. The
TaqMan probe is labeled with both a fluorescent reporter dye and
a fluorescent quencher dye and is also altered so that it cannot
be used as a primer for extension. Two additional unlabeled primers that flank the sequence of interest including the TaqMan
probe are used for PCR after hybridization of the TaqMan probe.
During PCR, the 5 exonuclease activity of the Taq DNA polymerase degrades the TaqMan probe from the 5 end, thus
releasing the reporter dye that is now able to fluoresce because
the quencher dye is no longer in proximity. As the PCR reaction
continues the fluorescence intensity of the reporter dye increases.
To detect a specic DNA sequence variation, two TaqMan probes
are developed, one for each allele, with reporters that fluoresce
as different colors. An advantage of this approach is that it is a
closed system without the need for purication or electrophoresis
of the amplication products, thus reducing the opportunity for
sample mix-up and contamination. The procedure also allows
for relatively high throughput as 96 samples can be analyzed in
a single assay and two to three assays can be run each day. Only
a very small amount (50 ng) of template DNA is required.
The protein truncation test (PTT) is a specic test for frame
shifts, splice site, or nonsense mutations that truncate a protein
product.17 Since the PTT only detects certain classes of mutations,
it is only useful for diseases that are predominantly caused by
mutations that cause a truncated protein product. An advantage
of this method is that it only detects pathogenic mutations.
SCREENING THE ENTIRE CAUSATIVE GENE

For most diseases many different mutations can be responsible
for the disorder, and genetic testing requires a search for mutations anywhere within or near the relevant gene. To comprehensively screen a gene for mutations, PCR amplication of gene
29
SECTION 1
GENETICS
R
TaqMan probe hybridized
with reporter (R) and
quencher (Q) tags, as well
as unlabeled primers and
DNA polymerase
FIGURE 3.4. TaqMan Assay. A specific

TaqMan oligonucleotide of 2030 bp is
designed to hybridize specifically with the DNA
sequence to be tested. The TaqMan probe is
labeled with both a fluorescent reporter dye
and a fluorescent quencher dye and is also
altered so that it cannot be used as a primer for
extension. Two additional unlabeled primers
that flank the sequence of interest including the
TaqMan probe are used for PCR after
hybridization of the TaqMan probe. During
PCR, the 5 exonuclease activity of the Taq
DNA polymerase degrades the TaqMan probe
from the 5 end, thus releasing the reporter dye
that is now able to fluoresce because the
quencher dye is no longer in proximity. As the
PCR reaction continues the fluorescence
intensity of the reporter dye increases. To
detect a specific DNA sequence variation two
TaqMan probes are developed, one for each
allele, with reporters that fluoresce as different
colors.
R
Q
Primer extension with

initial degradation by DNA
polymerase exonuclease
activity
R
Q
Further extension and
degradation releasing the
reporter and allowing fluorescence
Completion of extension and

release of the reporter, cycle
ready to repeat
segments (typically exons) followed by direct sequencing is usually

the method of choice. It is possible to screen gene segments without sequencing using techniques such as SSCP (single-strand
conformation polymorphism)18 or DGGE (denaturing gradient
gel electrophoresis);19 however, these methods are laborious and
can miss some mutations. Mutations identied by the screening
methods are typically conrmed by sequencing. Direct sequencing is costly; however, it provides the most reliable and reproducible results.
DNA microarrays or chips have been adapted for DNA
sequence detection.2022 Hybridization chips contain oligonucleotides matching all wild-type and single-nucleotide substitution
sequences in a gene. The patient DNA to be tested is amplied
using PCR, fluorescently labeled and hybridized to the array.
Minisequencing chips use arrayed oligonucleotide primers with
a free end that will be used for extension by DNA polymerase if
the free end matches the patient DNA. If the oligonucleotide
primer is allowed to extend the sequence of the new DNA strand
can be determined. The arrays are made with primers specic
for the normal sequence as well as for all possible mutations.
MUTATION VALIDATION
30
Direct mutation testing frequently reveals novel DNA sequence

changes that have not been previously associated with a disease
phenotype. Such sequence variants may be causative mutations
or they may be benign polymorphisms. Before the sequence change
can be recognized as disease-causing, it is important that the association of the putative mutation with the disease is supported
by additional studies. Ideally it would be best to demonstrate
that the mutant protein has an abnormal function, but this is
not always practical or feasible. Creating a transgenic animal
that carries the mutation and inspecting for signs of the disease
is another approach, but this can be extremely laborious and
time consuming and could not be done for every new mutation
discovered. It is important to determine if the sequence change
affects a region of the gene coding for a portion of the protein
that is critically important for its function. It is also possible to

determine if the DNA sequence change is in a part of the
protein that is evolutionarily conserved which is an indication
that the changed sequence is in a region of the protein that is
functionally important. A control group of individuals without
evidence of the disease should be screened for the mutation. To
be reasonably certain that the DNA sequence change is not a rare
polymorphism, at least 100 control patients (200 chromosomes)
should be analyzed. If the patient carrying the putative mutation
has family members (both affected and unaffected) then segregation of the sequence change in the family with the disease can
be evaluated. The characteristics of a disease-causing mutation
would include location in an evolutionarily conserved region of
the protein that may have critical function, not present in at least
100 controls and evidence of segregation in affected families.
Studies that will advance the knowledge of disease gene (and
protein product) functions and development of disease-specic
mutation databases will help make this task easier in the future.
INDIRECT TESTING AND METHODS

If the causative gene is not known, but the chromosome location
of the gene is known, then it may be possible to use genetic
markers located in the same region as the gene to identify family
members at risk for the disease. This method can only be used
if the disease is inherited as a Mendelian trait, and if the chromosome location of the causative gene has been previously determined using genetic linkage studies. In addition, the individual
to be tested must have affected family members and also a
sufciently large family that the parental chromosomes and the
chromosome carrying the abnormal copy of the gene can be
identied (Fig. 3.4). Identifying the chromosome carrying the
disease gene (determining phase) is enhanced by genetic markers
that are informative in the parents, i.e., that they carry different
alleles at the marker (heterozygous) so that both copies of their
chromosomes can be identied. Microsatellite repeat markers
are highly informative because they have on average six to eight
alleles. With the completion of the human genome, over 10 000

microsatellite markers have been mapped across the human
genome, making it almost always possible to nd an informative marker that maps close to the disease locus.23
Because indirect testing is looking for a DNA marker located
near the gene and not the gene itself, there is a risk that a recombination event will occur between the marker and the gene which
can cause the disease chromosome to be inaccurately identied.
The closer the marker is to the true location of the gene, the less
the risk of a recombination event occurring between the marker
and the disease gene. Thus, indirect testing is most accurate for
disease genes that have been tightly linked to a small chromosome region, and with multiple highly polymorphic markers
located on opposite sides of the disease locus so that recombination events can be visualized. The actual genetic risk can be
calculated using several methods including Bayesian calculations
and linkage programs.2426
POPULATION SCREENING
Screening a population for a disease-related risk factor may identify a group of individuals who are at high risk for the disease.
If this knowledge enables actions that can modify the risk, then
the screening test has merit. For example, patients with higher
than normal intraocular pressure are at increased risk for optic
nerve disease related to glaucoma. Knowing that their pressure
is high, patients can initiate treatment to reduce their pressure
and lower their risk.27 A genetic risk factor could identify a population of individuals at increased risk for developing a disease,
and if the knowledge of this increased risk makes it possible to
pursue treatment or behavior modication to reduce the risk
then the genetic testing is useful. Ideally the useful outcome is
treatment, but for many diseases this is not currently possible.
Other outcomes that may be useful are to avoid environmental
exposures that increase the risk and increase disease surveillance. Emerging evidence may suggest that screening macular
degeneration patients for the complement factor H risk allele
and the LOC387715 risk allele may help identify groups of
patients that should avoid smoking.2830
SPECIFICITY AND SENSITIVITY OF

GENETIC TESTING
An ideal test should be both specic and sensitive. Specicity is
the number of unaffected individuals that are negative for the
test compared with the total number of unaffected individuals
tested (including those that tested positive for the test). Sensitivity is the number of affected individuals that are positive for
a test compared with the total number of affected individuals
(including those that tested negative for the test) (Fig. 3.5). In
general, false positives (individuals without the disease who test
positively) and false negatives (individuals with the disease who
test negatively) are serious failures of a diagnostic test. For genetic tests, false positives are rare. The most likely causes of false
positives in DNA testing are laboratory or clerical errors. Falsenegative tests are much more common in DNA testing. Falsenegative tests can arise for a number of reasons including:
genetic heterogeneity (more than one gene is responsible for the
condition), PCR artifacts caused by primer binding site polymorphisms and deletions/insertions of the PCR primer sites,
deletion/insertion of an entire exon or the entire gene that interferes with PCR amplication, preferential amplication of the
smaller allele in a large insertion, and tissue mosaicism. Because
a negative result cannot completely eliminate the possibility that
a person carries a mutation in a causative gene, genetic counseling
and patient and physician education are important components
of genetic testing.
Specificity and sensitivity

Affected
Unaffected
individuals
individuals
Individuals
CHAPTER 3
Genetic Testing
positive for test

Individuals
negative for test
Sensitivity
A
A+C
Specificity
D
B+D
FIGURE 3.5. Definition of sensitivity and specificity for a laboratory

test. Sensitivity is defined as the number of affected individuals
positive for the test (A) divided by the total number of affected
individuals tested (A + C). Specificity is defined as the number of
unaffected individuals negative for the test (D) divided by the total
number of unaffected individuals tested (B + D).
CLIA LABORATORIES
Laboratories offering genetic testing must comply with regulations under the Clinical Laboratory Improvement Amendments
of 1988 (CLIA). CLIA, administered by the Centers for Medicare
and Medicaid Services, requires that laboratories meet certain
standards related to personnel qualications, quality control
procedures, and prociency testing programs in order to receive
certication. This regulatory system was put in place to encourage
safe, accurate, and accessible genetic tests. In addition to ensuring
that consumers have access to genetic tests that are safe, accurate, and informative, these policies encourage the development
of genetic tests, genetic technologies, and the industry that produces these products. A number of CLIA-certied laboratories
performing genetic testing for eye diseases exist in the United
States. For a list of CLIA-certied laboratories participating in
the National Eye Institute sponsored eyeGENE network, see the
NEI website at: http://www.nei.nih.gov.
Key Features
Genetic testing uses information about the gene(s) responsible

for a disease to identify individuals who carry abnormal forms
of a gene that may increase their risk of disease, alter the
progression of a disease, or identify them as carriers of a
disease.
The type of genetic testing depends on the available
information about the genetic disease. If the disease gene is
known then direct testing can be performed, if only the
chromosomal location of the gene is known then indirect
testing is performed.
Direct testing evaluates the DNA or RNA from a patient for a
specic sequence change, or genotype that causes the
disease. In some cases, protein samples can be analyzed for
specic amino acid changes.
Indirect testing uses family analysis to detect copies of the
chromosome that contains the mutant form of a diseasecausing gene.
Laboratories offering genetic testing must comply with
regulations under the CLIA of 1988, and genetic counseling
and patient and physician education are important
components of genetic testing.
31
SECTION 1
32
GENETICS
REFERENCES
1. Saiki RK, Bugawan TL, Horn GT, et al:
Analysis of enzymatically amplied betaglobin and HLA-DQ alpha DNA with allelespecic oligonucleotide probes. Nature
1986; 324:163166.
2. Mulot C, Stucker I, Clavel J, et al:
Collection of human genomic DNA from
buccal cells for genetics studies:
comparison between cytobrush,
mouthwash, and treated card. J Biomed
Biotechnol 2005; 2005:291296.
3. Barker DL, Hansen MS, Faruqi AF, et al:
Two methods of whole-genome
amplication enable accurate genotyping
across a 2320-SNP linkage panel. Genome
Res 2004; 14:901907.
4. Onadim Z, Cowell JK: Application of PCR
amplication of DNA from parafn
embedded tissue sections to linkage
analysis in familial retinoblastoma. J Med
Genet 1991; 28:312316.
5. Suenaga E, Nakamura H: Evaluation of
three methods for effective extraction of
DNA from human hair. J Chromatogr B
Analyt Technol Biomed Life Sci 2005;
820:137141.
6. Chuang EY, Chen X, Tsai MH, et al:
Abnormal gene expression proles in
unaffected parents of patients with
hereditary-type retinoblastoma. Cancer Res
2006; 66:34283433.
7. Zhou Z, Vollrath D: A cellular assay
distinguishes normal and mutant
TIGR/myocilin protein. Hum Mol Genet
1999; 8:22212228.
8. Spruijt L, Kolbach DN, de Coo RF, et al:
Influence of mutation type on clinical
expression of Leber hereditary optic
neuropathy. Am J Ophthalmol 2006;
141:676682.
9. Sena DF, Finzi S, Rodgers K, et al: Founder
mutations of CYP1B1 gene in patients with
congenital glaucoma from the United
States and Brazil. J Med Genet 2004; 41:e6.
10. Munier FL, Frueh BE, Othenin-Girard P, et
al: BIGH3 mutation spectrum in corneal
dystrophies. Invest Ophthalmol Vis Sci
2002; 43:949954.
11. Sieving PA, Bingham EL, Kemp J, et al:

Juvenile X-linked retinoschisis from XLRS1
Arg213Trp mutation with preservation of
the electroretinogram scotopic b-wave. Am
J Ophthalmol 1999; 128:179184.
12. Ali M, Venkatesh C, Ragunath A, Kumar A:
Mutation analysis of the KIF21A gene in an
Indian family with CFEOM1: implication of
CpG methylation for most frequent
mutations. Ophthalmic Genet 2004;
25:247255.
13. Kuo NW, Lympany PA, Menezo V, et al:
TNF-857T, a genetic risk marker for acute
anterior uveitis. Invest Ophthalmol Vis Sci
2005; 46:15651571.
14. Li J, Chu X, Liu Y, et al: A colorimetric
method for point mutation detection using
high-delity DNA ligase. Nucleic Acids Res
2005; 33:e168.
15. Hantash FM, Olson SC, Anderson B, et al:
Rapid one-step carrier detection assay of
mucolipidosis IV mutations in the
Ashkenazi Jewish population. J Mol Diagn
2006; 8:282287.
16. Ranade K, Chang MS, Ting CT, et al: Highthroughput genotyping with single
nucleotide polymorphisms. Genome Res
2001; 11:12621268.
17. Tsai T, Fulton L, Smith BJ, et al: Rapid
identication of germline mutations in
retinoblastoma by protein truncation
testing. Arch Ophthalmol 2004;
122:239248.
18. Vincent A, Billingsley G, Priston M, et al:
Further support of the role of CYP1B1 in
patients with Peters anomaly. Mol Vis 2006;
12:506510.
19. Mashima Y, Shiono T, Inana G: Rapid and
efcient molecular analysis of gyrate
atrophy using denaturing gradient gel
electrophoresis. Invest Ophthalmol Vis Sci
1994; 35:10651070.
20. Mandal MN, Heckenlively JR, Burch T, et
al: Sequencing arrays for screening
multiple genes associated with early-onset
human retinal degenerations on a highthroughput platform. Invest Ophthalmol Vis
Sci 2005; 46:33553362.
21. Yzer S, Leroy BP, De Baere E, et al:

Microarray-based mutation detection and
phenotypic characterization of patients with
Leber congenital amaurosis. Invest
Ophthalmol Vis Sci 2006; 47:11671176.
22. Zernant J, Kulm M, Dharmaraj S, et al:
Genotyping microarray (disease chip) for
Leber congenital amaurosis: detection of
modier alleles. Invest Ophthalmol Vis Sci
2005; 46:30523059.
23. Kong X, Murphy K, Raj T, et al: A combined
linkage-physical map of the human
genome. Am J Hum Genet 2004;
75:11431148.
24. Kuno S, Furihata S, Itou T, et al: Unied
method for Bayesian calculation of genetic
risk. J Hum Genet 2006; 51:387390.
25. Wiggs J, Nordenskjold M, Yandell D, et al:
Prediction of the risk of hereditary
retinoblastoma, using DNA polymorphisms
within the retinoblastoma gene. N Engl J
Med 1988; 318:151157.
26. Wiggs JL, Dryja TP: Predicting the risk of
hereditary retinoblastoma. Am J
Ophthalmol 1988; 106:346351.
27. Kass MA, Heuer DK, Higginbotham EJ, et
al: The ocular hypertension treatment
study: a randomized trail determines that
topical ocular hypotensive medication
delays or prevents the onset of primary
open-angle glaucoma. Arch Ophthalmol
2002; 120:701713.
28. Wiggs JL: Complement factor H and
macular degeneration: the genome yields
an important clue. Arch Ophthalmol 2006;
124:577578.
29. Schmidt S, Hauser MA, Scott WK, et al:
Cigarette smoking strongly modies the
association of LOC387715 and age-related
macular degeneration. Am J Hum Genet
2006; 78:852864.
30. Sepp T, Khan JC, Thurlby DA, et al:
Complement factor H variant Y402H is a
major risk determinant for geographic
atrophy and choroidal neovascularization in
smokers and nonsmokers. Invest
CHAPTER
Principles of Genetic Counseling

Gretchen Schneider and Pamela Hawley
The rapid advance in knowledge about genetic diseases and the

genetic contribution to common disorders, the improvements
in diagnostic testing, and the availability of some therapeutic
options have greatly enhanced the usefulness of genetic counseling to families. The principles of genetic counseling can be
readily appreciated from the denition recommended by an ad
hoc committee of the American Society of Human Genetics.1
This denes genetic counseling as a communication process
aimed at helping families or individuals understand the implications of a denitive diagnosis or a risk for a disease, and the
hereditary implications for the patient, parents, and, when indicated, other family members. Properly trained professionals
must be prepared to help the individual and the family comprehend available options for dealing with risk and to appropriately guide and support them in choosing the best course
of action.
Although the committee published this denition in 1974, these
goals of genetic counseling still remain widely accepted and
disseminated.2 What are changing rapidly are the diagnostic
tools available to meet these goals as well as the use of the
principles of genetic counseling as they apply to an increasingly
broadened scope of clinical scenarios. Because accurate genetic
counseling is predicated on a precise risk or accurate diagnosis,
knowledge of these new diagnostic tools and a consistent
approach to clinical evaluation are essential to the process.
WHY REFER PATIENTS FOR A GENETIC

EVALUATION
WHO PROVIDES GENETIC COUNSELING
INDICATIONS FOR REFERRAL TO A

GENETICS SPECIALIST
The providers of genetic counseling have changed greatly in the

past few decades. In the 1970s, when genetic counseling was
growing in recognition, many counselors were MDs and PhDs
who had no formal training. Physicians, nurses, and social workers
have continued to provide genetic counseling, mostly by learning
from experience. As genetic counseling became better dened,
the need was recognized for persons trained specically to deal
with this process and its integration with medical science and
psychology.
Masters level genetic counseling programs are designed to
train medical professionals, called genetic counselors, who provide such a service. These 2-year programs have combined
molecular and clinical genetics with counseling psychology in
settings that emphasize clinical rotations to gain experience.
More than 1500 genetic counselors have been trained at over 30
2-year programs. Genetic counselors often work with other health
professionals, including board-certied geneticists, obstetricians,
genetic fellows, nurses, social workers, and laboratory personnel.
This team approach allows comprehensive genetic services in
prenatal, pediatric, adult, cancer, specialty clinic, and commercial
settings.
Accurate genetic counseling starts with a thorough genetic

evaluation. It is important for both families and physicians to
realize what is involved in the process and its value to the
patient and immediate relatives. The genetic evaluation is
important in a number of major ways:
1. It may help in understanding a patients problems by
providing a unifying diagnosis. When the diagnosis is a
well-described entity, it can sometimes provide prognostic
information. It may also change the clinical management
of a patient.
2. It may establish an increased risk of developing a disease
based on genetic markers, for example, breast or colon
cancer. This, too, can provide insight into options for
increased surveillance, or changes in management based
on this risk.
3. A specic diagnosis or the presence of a genetic risk factor
may have implications for other family members. Relatives
may also be at risk or become similarly affected. In many
instances, these relatives should be encouraged to receive
genetic counseling. Future children in the family may be at
risk. This risk is called the recurrence risk, and it
sometimes can be mathematically quantied.
Although the need for a genetic evaluation or genetic counseling

often is obvious, this is not always the case. A child born with multiple anomalies may have no clearly identiable diagnosis until
pedigree analysis reveals a pattern diagnostic of a genetic syndrome. This is particularly important whenever parents are
planning additional children and are justiably concerned about
those children having similar problems. Even when a clinical
diagnosis and the relevant genetic counseling may seem straightforward, unanticipated benecial information might be gained
from a visit to a genetics specialist.
ESTABLISHED GENETIC CONDITION

For a child or adult with an established diagnosis, the focus of a
genetics visit might be to understand the hereditary implications
of the diagnosis and the recurrence risks. For example, in a child
with retinoblastoma and a positive family history, the diagnosis
is clear. These families may be referred for genetic counseling to
review recurrence risks in a setting separate from the ophthalmologists ofce. An ophthalmologist may not feel well versed
33
SECTION 1
GENETICS
in the details of molecular testing and its use in testing other
family members and in prenatal diagnosis. A genetics specialist
can also discuss alternative reproductive options for those who
may not want prenatal testing.
Genetic evaluation sometimes suggests a clinical diagnosis of
a disorder that displays genetic heterogeneity. An example is oculocutaneous albinism. There are several types of albinism due to
various mutations in any of several genes. A genetic evaluation
might uncover relatives who clearly have albinism; this information might allow diagnosis with a mildly affected index patient.
Conrmation of that diagnosis might require biochemical or
molecular tests.
EYE FINDINGS WITH OTHER CONGENITAL

ANOMALIES
A child is sometimes born with a number of malformations
including ophthalmologic abnormalities. Some cases obviously
t a particular syndrome, but others do not. For example, a child
might have microphthalmia, congenital heart disease, and delays
in development, with no syndrome diagnosis immediately recognizable. Yet these multiple medical problems suggest a unifying
explanation for these ndings. This constellation of ndings could
be the syndrome of coloboma, heart defects, choanal atresia,
retarded growth and development, genital hypoplasia in males,
and ear anomalies the CHARGE syndrome or it could be
caused by a chromosome anomaly such as 13q. In these situations, the experience of a geneticist in recognizing malformation
patterns and understanding the variability of genetic conditions
can aid in diagnosis. A genetics professional is also more likely
to be aware of the latest testing available, which may also be an
important component of the evaluation and diagnostic process.
If an underlying cause is identied, relatives can then undergo
genetic counseling.
EYE FINDINGS WITH OTHER MINOR

ANOMALIES
Some patients referred to the ophthalmology clinic may have no
obvious extraocular medical problems. During their visit, however, one may observe dysmorphic features or other seemingly
unrelated minor medical signs or symptoms. For example, retinitis pigmentosa is a feature of a number of syndromes whose
other signs and symptoms may be subtle. A child with retinitis
pigmentosa, obesity, and polydactyly may have BardetBiedl
syndrome, whereas one with prominent central incisors and
slender hands and feet may have Cohens syndrome. Similarly,
a child referred for myopia who has micrognathia could have
Sticklers syndrome. One with ectopia lentis due to Marfans
syndrome might be tall and lanky. Physical features that may
not be classied as medical problems, when combined with eye
ndings, may lead to a syndrome diagnosis which is more easily
recognizable by a genetics professional.
SPECIFIC EYE DISEASES
34
A genetic evaluation may be important for patients with a purely

ocular disease for a number of reasons. A family history might
reveal similar eye disease or other ndings that, when compared,
may lead to a genetic diagnosis in the family. A comprehensive
pedigree analysis sometimes reveals a genetic basis for such
diseases. Many frequently encountered ophthalmologic diseases,
such as cataracts or glaucoma, have a well-documented Mendelian
inheritance pattern. Others may not be purely Mendelian, but
the presence of multiple affected family members would indicate
increased risk for other relatives. Identifying the inheritance
pattern might lead to the identication of affected relatives who

could be diagnosed and treated early in the course of disease.
This is especially important in families with such conditions as
dominantly inherited juvenile glaucoma.
INCIDENTAL EYE FINDINGS

Eye ndings with important genetic implications are sometimes
observed incidentally during ophthalmologic evaluation. For
example, a child may undergo ophthalmologic evaluation because
of a failed eye test at school but be found to have Lisch nodules,
which suggests neurobromatosis type 1. Another child might
have the stellate iris pattern of Williams syndrome. Heterochromia irides indicate an examination for the possibility of
Waardenburgs syndrome. Although such ndings may not have
any clinical implications, in some patients their strong association
with genetic conditions is an indication for a genetic evaluation.
Despite the numerous situations in which it is important to
explore the possibility of a genetic etiology, an identiable genetic condition is often not found. This does not exclude the
possibility of an underlying genetic cause for the individuals
problems. Family members need to be aware of the possibility
of recurrence risk even if no specic diagnosis is made.
WHAT IS INVOLVED IN A GENETIC

EVALUATION
A genetic counselor begins a visit by ascertaining the clients
understanding of the reason for the referral. The components of
a genetics evaluation are described and, when appropriate, the
client is cautioned that the evaluation does not always result in
a denite diagnosis or establish a specic genetic etiology.
FAMILY HISTORY
A detailed pregnancy, medical, and developmental history is
obtained, as is a three-generation family health history that
includes the ethnic origins of the ancestors. The possibility of consanguinity should be explored. The family history is obtained
not only to establish a hereditary pattern for the referring diagnosis but also to identify other conditions that could have hereditary implications. For example, if the parents of the patient
are of Eastern European Jewish ancestry, their children are at
increased risk for TaySachs disease, a recessive neurodegenerative condition for which carrier testing is available. If the
family history reveals developmental delay in a pattern
suggestive of fragile X syndrome, carrier testing could be offered.
Several modes of inquiry ascertain whether families could be at
risk for certain conditions unrelated to the referring diagnosis
(Table 4.1).
PHYSICAL EXAMINATION
A complete physical examination is performed with attention to
growth parameters, developmental milestones and subtle physical ndings that can be important for establishing a syndrome
diagnosis. Careful anthropometric measurements (e.g., inner
canthal, outer canthal, and interpupillary distances; midnger/total hand length; and upper body to lower body ratios) may
be obtained. Photographs also can be used to record nonmeasurable dysmorphic features.
Examination of other family members may be indicated to
determine if a particular nding is hereditary. Sometimes this is
incidental to the reason for referral. Findings such as fth-nger
clinodactyly, although a part of many syndromes, may also be
an isolated hereditary trait without other medical implications.
TABLE 41. Family History Considerations Regardless of Reason for Referral

Family History Positive for:
CHAPTER 4
Consider:
Ancestry
Eastern European Jewish*
TaySachs disease carrier testing

Canavans disease carrier testing
Cystic brosis carrier testing
Fanconi anemia type C
Gaucher disease
NiemannPick type A
French Canadian
TaySachs disease carrier testing

Caucasian
African American
Sickle cell anemia carrier testing
Mediterranean
b-Thalassemia carrier testing
Southeast Asian
a and b-thalassemia carrier testing
More than two miscarriages
Parental chromosome studies to rule out translocation
Birth defects in near relatives
Chromosome studies in parent
Developmental delay
Fragile X testing if family history indicates pattern

Because of the possibility of asymptomatic transmitting males and
affected females, the inheritance is not the typical X-linked
recessive pattern
Maternal age over 35
Prenatal chromosome studies
Neonatal/childhood deaths in
rst-degree relative
Review of records, particularly autopsy
Known genetic disease
Possible carrier testing (i.e., cystic brosis, Duchennes muscular

dystrophy)
* The extent of screening for individuals of Ashkenazi descent varies by institution and laboratory and may include fewer, or more, tests than those listed.
COMPUTER-ASSISTED DIAGNOSTICS
ASSESSMENT
Many databases can be accessed as part of the genetics evaluation (Table 4.2). Pregnancy exposures may be assessed through
REPROTOX, a computerized database of potential teratogens
(available at many institutions free of charge through
MicroMedex). Standard computer literature searches are performed. If ndings are multiple and the patients history and
clinical ndings do not suggest an obvious syndrome, the patients
information may be entered into genetic syndrome databases
such as POSSUM or London Dymorphology (these are available
by purchase) in an effort to diagnose a syndrome. If a specic
syndrome is being considered or an isolated nding has been
established, On-Line Mendelian Inheritance in Man (OMIM) is
often useful. OMIM is a frequently updated catalog of more
than 8400 human genetic conditions that is available to the
public through the NIH. It contains a historical summary of the
condition, current information regarding available diagnostic and
treatment options, details of genetic etiology, and references.
GENETESTS is another database of up-to-date clinical and
research diagnostic testing for specic conditions as well as a
library of comprehensive reviews written by genetic experts on
many genetic diseases. When circumstances and time permit,
computer searches such as these are conducted prior to or
during the initial visit. While there are many additional sources of
information on the Internet, it is advisable to select well-known
databases or websites with accurate and up to date information
when using it for patient assessment.
The initial assessment of an individual may include recommending testing or specialty consultations based on the history,
examination, or computer searches. Ophthalmologic examinations for relatives may be indicated to detect relevant eye ndings. These examinations can be helpful in establishing familial
patterns when autosomal dominant or X-linked conditions are
being considered. For example, Bests disease is an autosomal
dominant form of macular degeneration that causes a distinctive macular lesion in its early stage. Scarring at the site of the
lesion can lead to decreased central vision. Macular lesions are
not present in all affected patients, but all affected patients have
abnormal electrooculogram ndings. Ophthalmologic examinations of the parents of an affected child can help provide
them with a recurrence risk assessment as well as identify
which side of the family may have affected relatives. Another
example is Lowe syndrome, an X-linked condition with ndings
that include congenital cataracts, neurologic impairment, and
renal tubular dysfunction. Female carriers typically show no
neurologic or renal defects as detected by physical examination
or laboratory testing. However, slit-lamp examination reveals
specic lenticular changes in up to 94% of carriers.3 Although
molecular diagnostic testing is clinically available, careful
ophthalmologic examination is also valuable in assessing the
carrier status and therefore the recurrence risk for this condition, particularly in families in which diagnostic testing was
negative.
35
SECTION 1
GENETICS
TABLE 42. Computer-Assisted Diagnostics

Program
Database
REPROTOX
Teratogens
Reprotoxicology Center
Columbia Hospital For Women, Washington, DC
London Dysmorphology
Syndrome identication
Oxford University Press

POSSUM
Syndrome identication
Murdoch Institute for Research into Birth Defects

Royal Childrens Hospital, Melbourne, Australia
OMIM
Human genetic conditions
http://www3.ncbi.nlm.nih.gov/omim/
GENETESTS
http://www.genetests.org/
Availability of clinical and research

diagnostic testing
Expert Written Disease Reviews
PubMed
http://www.ncbi.nlm.nih.gov/sites/entrez?db=pubmed
Literature search
It may be necessary to obtain documentation of previous

testing including chromosome analysis, DNA testing or other
types of diagnostic tests and to review the studies (such as a
karyotype) to conrm the adequacy of the study. Obtaining records
to document a condition reported in a family member may also
be indicated. Because of these numerous steps involved in the
assessment process, review of the nal assessment sometimes
requires a follow-up visit.
At the completion of the genetic evaluation of a patient
referred with a specic ocular nding, assessments can fall into
one of three general areas:
1. Isolated ocular disease or anomaly.
2. Nonocular ndings with a pattern that ts no
recognized genetic syndrome.
3. Nonocular ndings with a pattern that ts a
recognizable syndrome or association.
In the latter two situations, the ophthalmologist may not recognize other clinical implications and the family may benet from
discussion of these with a genetics professional. In any of these
three situations, a genetic component may be at work that
influences the risk of disease in the patients offspring, parents,
and other family members.
EXPLANATION OF CONCLUSIONS
Genetic counseling involves explaining the assessment process
and its conclusions to the family, including what is known about
the genetics of the patients condition and any possible medical
and developmental implications.
MEDICAL AND DEVELOPMENTAL

IMPLICATIONS
36
A genetic evaluation that results in a specic diagnosis may

provide information regarding previously obscure medical or
developmental implications. It is important to discuss clinical
variability in syndromes and to note that individuals do not
usually develop all the ndings associated with a given condition.

Even if genetic testing has conrmed a diagnosis, it seldom
provides information regarding the likelihood or severity of specic features of a genetic disease. However, for some syndromes,
empirical data exist regarding the probability of the associated
ndings. A genetic specialist can explain the indications for
medical monitoring or evaluations and can make appropriate
referrals. The importance of age-appropriate developmental
assessment and intervention programs in helping patients reach
their maximum potential is also emphasized. An established
diagnosis may have no additional medical or developmental implications, or no denitive diagnosis may be reached. In these
cases, the focus is primarily on the genetic implications of the
diagnosis.
GENETIC IMPLICATIONS
PRECISION
The extent to which the genetic component of a disorder is
understood can vary a great deal. This understanding affects the
precision of risk assessment and the options available for
modifying the risk. Some diseases have a denite inheritance
pattern that permits risks to be calculated according to the laws
of Mendelian genetics. For example, in a patient with Marfans
syndrome, an autosomal dominant condition, there is high
condence in declaring a risk of 50% for offspring. Similarly, in
a family with a child with an autosomal recessive disease such
as BardetBiedl syndrome, the risk of recurrence in siblings is
one in four.
In contrast, in other diseases there is genetic heterogeneity, and
various inheritance patterns are possible. This can complicate
the prediction of risk. Instructive examples are nonsyndromic
retinitis pigmentosa or congenital cataracts. The inheritance
pattern can be autosomal recessive, autosomal dominant, or
X-linked recessive. For an isolated male case of retinitis pigmentosa, empirical data suggest that his offspring have a 12%
risk of recurrence.4 In fact, the recurrence risk ranges from less
than 1%, if it can be established that the patient has recessive

retinitis pigmentosa, and up to 50% if he has dominant retinitis
pigmentosa. In other scenarios, the risk differs from case to case.
One example is when a syndrome whose genetic etiology is not
well dened has been diagnosed in a child, but a recurrence risk
of 2% has been reported. Another is when a child has a constellation of ndings that has not previously been recognized. The
actual recurrence in siblings could be negligible if the etiology is
nongenetic, 25% if it is autosomal recessive, or ~50% if a
parent carries the mutant gene but does not express it clinically
(i.e., nonpenetrant).
Counselors must be cautious in providing recurrence risk in
a family with a child who has a well-established dominant syndrome if neither parent shows evidence of the disease. At rst
glance, we might assume that the affected child represents a new
dominant mutation, in which case the parents are genetically
normal and the recurrence risk for siblings is vanishingly small.
However, two possibilities by which recurrence risk could be much
higher need to be considered. One, nonpenetrance, is dened as
the absence of phenotypic features in a person who has the
mutant genotype. If one of the parents is a nonpenetrant carrier,
the recurrence risk for subsequent children approaches 50%.
Another possibility is gonadal mosaicism, in which the mutation
has occurred during the growth and development in a parent, so
that it is present in a proportion of that parents germ cells.
Although genetic testing or empirical data may be available to
determine if a parent is a nonpenetrant carrier, testing is often
not available to evaluate gonadal mosaicism, and empirical data
on the frequency of gonadal mosaicism for specic conditions
are rare.
PATIENTS UNDERSTANDING OF RISKS AND

OPTIONS
It is important to explain inheritance patterns and risks in ways
that patients will understand. A patients understanding of the
risks can be aided by presenting the risk estimates in more than
one way. Risk can be given as a fraction and as a percentage, and
risks can be given for both affected and unaffected offspring. For
example, one might explain that there is a 25%, or one in four,
chance that a disease would occur in the next child and a 75%,
or three in four, chance that it will not. The risk of recurrence
can also be put into context by providing the general population
risk for the particular condition, when available, as well as the
general population risk for a newborn child to have a serious
birth defect (34%).
A persons interpretation of a recurrence risk is affected by a
number of factors, including personality (e.g., risk-taker versus
risk-averse), family goals and beliefs, and perceived physical,
emotional, and nancial consequences of having a child with a
particular condition. In addition, a patients actual experience with
the condition in question can signicantly affect the perception
of risk. The woman at risk for sons with Lowe syndrome might
feel differently about this condition if her uncle experienced the
renal failure associated with this disorder and died before she
was born than if her yet mildly affected son had been recently
diagnosed.
It is not surprising, therefore, that a risk considered high by
some will be viewed as low by others. Reviewing how these different factors affect interpretation of information and the choices
that are made can help clients. The counselor also needs to be
aware of his or her own perceptions of risk and burden. To the
greatest extent possible, the information provided to a patient
should emphasize the objective nature of risk gures and avoid
the subjective nature of how people perceive risk and the consequences of a disease. There is no cutoff as to whether a given
risk gure is high or low or whether a specic disease-given consequence is severe or minor. Clients also need to hear whether
a specic disease is severe or mild. Patients should be told that
decisions regarding having (more) children, seeking prenatal
testing, or considering alternative ways to have families are
their own decisions and are not based on perceived orders of
their doctor or genetic counselor. Patients choose their future based
on their own goals, beliefs, and values.
CHAPTER 4
RISK MODIFICATION FOR FUTURE

OFFSPRING
PRENATAL DIAGNOSIS
One means of risk modication for future children is prenatal
diagnosis. For conditions in which a diagnosis can be conrmed
with chromosome, biochemical, or molecular studies, three
procedures can usually be offered:
1. Routine amniocentesis at 1516 weeks gestation.
2. Early amniocentesis at 12 weeks gestation.
3. Chorionic villus sampling at 1012 weeks gestation.
If diagnostic testing is not available for a condition that includes
major congenital malformations, serial ultrasound examinations
may be performed as a means of prenatal diagnosis. The examinations need to be performed by an ultrasonographer expert at
detecting fetal malformations; even then, the rate of detection
is not 100%.
If prenatal diagnosis is an option, a separate session should
be arranged to discuss the information more thoroughly. The risks,
benets, and limitations of the procedures can be reviewed in
detail. Couples need to be reminded that many conditions cannot
be detected prenatally and that normal results from prenatal
diagnostic evaluation do not guarantee a healthy child. All
couples, regardless of their ages or family history, have a 34%
risk of having a child with a birth defect. Also, many inherited
conditions display considerable clinical variability. Couples need
to be aware that prenatal diagnosis usually does not predict the
severity of a condition.
In counseling for prenatal diagnosis, it is important to stress
to parents that they are not committed in advance to any particular course of action in the event of an abnormal nding.
Although termination of an affected pregnancy is available, this
is clearly not an acceptable alternative for all couples. Some may
wish to know in advance if the baby will be affected because this
may affect delivery site and neonatal management. For others,
early knowledge can help their families prepare and adjust for the
baby. Many couples consider prenatal testing for the reassurance
associated with the more likely event that the results are normal.
Thus, prenatal diagnosis should not be summarily dismissed
for those couples who indicate that they will not consider elective pregnancy termination.
A relatively recent option for some conditions is preimplantation diagnosis with in vitro fertilization (IVF). Following IVF,
typically at the 816-cell blastomere stage, genetic material from
single cells is analyzed for DNA or chromosomal abnormalities.
Only embryos with a normal genetic complement (for the disorder testing for) are then implanted into the mothers uterus.
Though this procedure is very accurate, follow-up prenatal diagnosis is recommended to conrm the ndings discovered by PGD.
The procedure is also expensive and may not be covered by
insurance. Although some states require third-party payers to
cover IVF, this is usually mandated for infertile couples, and those
seeking preimplantation diagnosis are not infertile. Finally,
because relatively few facilities offer the procedure, logistics can
preclude its availability.
37
SECTION 1
GENETICS
ASSISTED REPRODUCTIVE TECHNOLOGIES

AND ADOPTION
Some risk revision options do not involve prenatal testing. Assisted
reproductive technologies offer a means for reducing risk,
particularly for Mendelian disorders or familial chromosome
changes. IVF with donor egg when the mother has an autosomal
dominant condition or is a carrier for an X-linked condition
reduces the risk to the level of population incidence. Risk is
similarly reduced with articial insemination by donor if the
father has an autosomal dominant condition. With recessive
conditions, articial insemination by donor usually reduces the
risk to less than 1%. Adoption can be an alternative for couples
who perceive the recurrence risk or consequences to be too high
but whose personal goals include a (larger) family. Therefore, for
families faced with risks to future children, alternative options
should be discussed as well.
CARRIER TESTING
For some conditions, carrier testing is available to revise risk. If
the concern is regarding future children, this means that prenatal diagnosis is available as well. However, assessment of
carrier status sometimes helps a couple decide if they wish to
pursue another pregnancy even if prenatal diagnosis is available.
It could also have implications for other family members.
For example, if a child has microphthalmia and other congenital anomalies related to a translocation trisomy 13 and both
parents have normal chromosomes, the risk of recurrence for their
offspring and those born to other relatives is extremely low. In
contrast, if one parent carries a balanced arrangement involving
chromosome 13, the empirical recurrence risk data would be
known for both parents and any sibling of the parent who
carries the rearrangement. Fabrys disease is another example in
which carrier testing is useful. This is an X-linked condition in
which affected patients accumulate glycolipid as a result of an
a-galactosidase deciency. Onset is typically in childhood or adolescence and includes episodes of severe extremity pain, angiokeratomas, and characteristic corneal and lenticular opacities.
Cardiac, renal, and cerebrovascular complications can occur later
in life. Carrier assessment includes ophthalmologic examination. Corneal opacities detectable only by slit-lamp examination
are present in ~80% of carriers.5 Assaying a-galactosidase levels
is another carrier testing option for this disorder. Both eye
examination results and enzyme level can be normal in carriers,
however, because of X-chromosome inactivation. Therefore, molecular testing may offer more denitive results to identify
females in a family who are at risk of having affected sons.
MOLECULAR TESTING: DISTINCTIONS AND

LIMITATIONS
38
Molecular testing often is used for prenatal testing and carrier

detection. When newly developed technology is being considered,
it is important that families be aware of whether the testing is
provided on a clinical or research basis. Clinical testing implies
well-established protocols with quality control measures and
available data regarding sensitivity and specicity. The time
required for testing is predictable, and a charge is often involved.
Research testing is performed in an unpredictable time frame,
and usually there is no charge.
Progress toward understanding the genetic basis of disease can
be expected to affect diagnostic capabilities rst. Treatment or
management of a genetic disease generally lags behind considerably, although considerable progress has been made for inborn
errors of metabolism such as Gaucher disease and Fabry disease
for which enzyme replacement therapy now exists. Although

gene therapy receives a great deal of media attention, clinical
application is so far limited. Genetic counselors must explain
this distinction between diagnostic and therapeutic interventions.
If a gene is mapped and DNA markers linked to the gene are
available, linkage analysis in some families may be used to
predict affected status in at-risk individuals. This, however, may
not always be informative because of the limited size of the
family. Studies should be performed on several family members
before it is known whether linkage studies will be useful for carrier
or prenatal assessment in that family. For those families in
which study results are informative, the studies will provide a
revised risk rather than a denitive answer, because with linkage studies, recombination is always possible. The degree of risk
revision varies from family to family, depending on which markers
are used. Accuracy is highest for families with informative
flanking markers. Another limitation of linkage studies is the
possibility that an altered gene at a location unlinked to the
markers could cause a similar clinical condition. If the gene
mutation or product is not testable, this potential heterogeneity
remains a concern.
When direct analysis of an actual gene mutation or gene product is possible, issues need to be discussed with families to help
them understand how the information is useful to them. For
example, if all possible mutations causing a condition cannot be
identied, testing will not be denitive in all cases. Although
blood is an easily accessible source of genetic material and useful for linkage and mutation analysis, it may not be a good source
for gene product testing. In this situation, additional tissue may
be necessary, and the appropriateness of a more invasive test
needs to be discussed with family members. The invasiveness
of a test should be weighed against the additional information
that will likely be obtained.
DOCUMENTATION AND FOLLOW-UP

Clients who are counseled should receive a detailed written summary of the evaluation. Although writing clear and informative
summaries can be extremely time consuming, it is necessary for
several reasons. It is unlikely that all the verbal information
provided during the visit will be remembered, and what is
remembered may be difcult for an individual to explain to
others. A summary serves as an extension of the communication
process that allows for review by the recipient.
Genetic counselors are available to clients on an ongoing basis
to reexamine and clarify the issues covered during the visit(s)
and in the written summary. They provide reassurance that the
clients responses to a situation are expected and appropriate;
this can be reinforced by providing families with information about
support organizations. In addition, families need to be informed
that genetics is a rapidly advancing area of medicine. Even if an
evaluation has failed to identify a specic diagnosis, families who
have received genetic counseling are encouraged to reestablish
contact whenever planning a pregnancy to take advantage of
any pertinent new developments.
ETHICAL CONSIDERATIONS IN GENETIC

COUNSELING
The increased understanding of genetic disease and the genetic
components of common disease as well as the availability of testing bring many challenges to genetic assessment and counseling
and raise a number of ethical issues. Although most genetic
counseling situations do not give rise to these dilemmas, it is
important for health professionals to be aware of these
possibilities.
CONFIDENTIALITY
Issues of genetic privacy are much discussed in the genetics community and society as a whole. There is debate over who should
have access to genetic information and how it can be used. Of
particular concern is the potential for discrimination by insurance
companies or employers. There is fear that insurance companies
may use test results to deny coverage, claiming that a genetic
disease is a preexisting condition. Alternatively, they may consider an affected individual to be an insurance risk if his or her
condition could cause medical problems in the future. Others
are concerned that employers may try to use genetic information to make hiring decisions, basing their assessment on risk for
medical complications or disability. Currently, numerous states
have genetic privacy legislation which protects patients from
discrimination, and national legislation has been presented but
is currently under review.
Regardless of protections in place, these issues often lead families or individuals to be wary of genetic testing. Some decide to
decline testing even if a positive test result could alter medical
management. Others choose to pay for testing themselves to
prevent the insurance company from having access to this information. Still others request that test results not be put in their
medical record. Families may desire to have total control over
the information to help minimize the risk of the information
being used against them.
Genetic professionals support the patients right to privacy
with regard to results of genetic testing. Those arranging testing
should discuss the issues of condentiality prior to the initiation of testing so there is consensus on how results are reported,
who receives results, and where the information is documented.
CONTROVERSIAL USES OF GENETIC

TESTING
A number of situations may arise where patients want to use
genetic testing for less traditional purposes. Because many patients
have access to different types of genetic testing, particularly if
they pay for it themselves, genetic counselors may be asked to
arrange testing for reasons with which they do not necessarily
agree. It is important for medical professionals to be aware of
these scenarios, recognize their own opinions, and be able to
refer patients to others if they do not feel that they can support
such patients wishes. Some specic examples of these situations are discussed below.
SEX SELECTION
A couple might wish to choose the sex of their child by testing
during a pregnancy, or through PGD with implantation of only
embryos of the desired sex. Having a child of a particular gender
has strong roots in some cultures, justifying these measures for
some couples. Other couples may simply wish to ensure that they
have children of both sexes in their family, a concept known as
family balancing. Although this is not illegal, and is offered at
some institutions, it can make those providing the testing
uncomfortable.
PRESYMPTOMATIC TESTING OF CHILDREN

Because testing is available for a number of disorders with later
onset, such as Huntington disease or breast cancer, it is possible
to test children or even fetuses for conditions that may not affect
their lives for many years. Although parents may feel that this
is in the best interest of their children, some fear it may cause
stigmatization. Others argue that undergoing testing should be
the decision of the individual, once he or she reaches adulthood,

particularly if it would not affect medical management. Current
recommendations discourage testing in children for disorders for
which the results would not warrant a change in their immediate
medical management. However, parents with strong desires to
pursue such testing may be able to nd someone willing to do it.
CHAPTER 4
TESTING FOR SELECTION OF AFFECTED

PERSONS
Patients with certain conditions or physical limitations may desire
to have similarly affected children. Patients with achondroplasia,
for example, have wanted to have children with achondroplasia
because this is what they have come to consider normal. This
could lead them to choose prenatal diagnosis to rule in achondroplasia, possibly resulting in the termination of an unaffected
pregnancy. The same might be true of a couple in which both are
deaf. Such testing is theoretically available, if a genetics center
feels comfortable performing it.
DISCOVERY OF UNANTICIPATED OR
HARMFUL INFORMATION
Because genetic testing can involve looking for a broad array of
abnormalities (e.g., when looking at chromosomes) or studying
a number of persons in a family (via linkage analysis), it can sometimes uncover information that family members did not anticipate or do not want to know. Prior to the initiation of testing,
it is important to discuss not only the possible benets of
genetic testing but also the potential for unanticipated results.
NONPATERNITY
Genetic testing can lead to the discovery of nonpaternity. Raising
this as a possible outcome prior to testing may help to avoid an
awkward situation when test results become available.
DISCLOSURE OF DISEASE STATUS

In large families studied by linkage analysis, a number of persons
may learn a family members disease status. Some such persons
may have no relationship with the physician or genetic counselor
that organized the testing. If possible, these persons should be
referred to a qualied physician or local genetics center where
they can learn about their disease status and discuss the implications of their test results. It is also best to determine which
family members do not want to know their results before testing
begins. Care must be taken to avoid divulging their status to other
family members. Those not requesting information should have
the option of obtaining it later, should they change their minds.
NONDISCLOSING PRENATAL DIAGNOSIS

A special situation surrounding genetic testing involves prenatal
diagnosis for an autosomal dominant condition in which a parent
is at risk but does not want to know his or her disease status.
Prenatal diagnosis using linkage analysis is most accurate in
families with affected individuals in more than one generation.
In this scenario, if a fetus is found to be unaffected, the parents
status would not need to be conveyed (Fig. 4.1). However, the
diagnosis of an affected fetus would indicate that the parent is
also affected. This would necessarily prompt a couple to come to
terms with the diagnosis in the parent.
Alternatively, testing to determine which grand-parental allele
a fetus received without establishing linkage of the disease gene
to a particular allele can be used when only one affected family
39
SECTION 1
GENETICS
affected. Another situation might be during prenatal diagnosis
for advanced maternal age, where the couple is told the fetus is
being tested for Down syndrome, but turns out to have a different chromosomal abnormality. In situations where genetic
testing is performed, all possible testing outcomes should be
discussed prior to testing.
DUTY TO RECONTACT
FIGURE 4.1. Linkage analysis with letters (AE) represents specic

RFLPs (see Chapter 1). The fetus is unaffected in both scenarios. The
fathers disease state is determined but need not be disclosed. (a) The
fetus and father both have the nondisease allele of the affected
grandmother. (b) The fetus receives the allele of the unaffected
grandfather, but the father has the disease allele from the affected
grandmother.
In the era of rapid scientic discovery, particularly in molecular

diagnostics, the question arises as to how to keep families informed
of new information. Parents of a child with albinism seen years
ago might now benet from molecular testing. Carriers of a fragile
X pre-mutation who had been told in the past that this has no
medical implications may need to be informed of the nowrecognized risk of premature ovarian failure or tremor-ataxia
syndrome. Therefore, what a family is told at a genetic counseling
session could eventually become outdated. At the same time, it
is not generally possible for medical professionals to contact
previous patients when new knowledge or testing becomes
available.
As discussed previously (see section on Documentation and
Follow-Up), genetic counselors must remain available to families.
In addition, the importance of genetic counseling for affected
children when they reach child-bearing age should be stressed.
This allows for a review of the genetic implications as well as an
update on the possibilities for diagnostic testing. Finally,
periodic follow-up visits may be suggested to help families keep
up-to-date on both clinical and molecular developments.
CONCLUSION
FIGURE 4.2. The risk of the fetus being affected is 50%. The fathers
risk remains unchanged. The fetus receives the grand-maternal allele,
but testing cannot determine whether it is the disease allele.
member is available for testing and when parents want to

guarantee that their status is not determined by testing. This
could exclude (within the limits of recombination) a fetus being
affected if it received an allele from the unaffected grandparent.
If the fetus received the allele of the affected grandparent, this
would not prove that the fetus is affected but would increase the
risk from 25% to 50% (Fig. 4.2).
DIAGNOSIS OF AN UNANTICIPATED
DISORDER
At times, a genetic test may provide unanticipated results.
Examples include performing hemochromatosis or CF carrier
testing on an individual only to determine they are actually
Genetic counseling involves the transfer of technical and conceptual information that is complex and sometimes different
from information the family may have previously encountered.
This information is often conveyed to persons who are feeling
anxious, guilty, depressed, or overwhelmed. By recognizing and
exploring the psychological impact of genetic counseling issues,
counselors can better integrate medical and genetic information
so that families feel competent in making informed decisions.
Such autonomy can reestablish their sense of control and aid in
their psychological adjustment.
Key Features
Genetic counselors often work with other health professionals,

including board-certied geneticists, obstetricians, genetic
fellows, nurses, social workers, and laboratory personnel to
provide genetic counseling.
A genetic evaluation includes family history, physical
examination, and assessment of laboratory and ancillary
testing.
Genetic counseling involves explaining the assessment
process and its conclusions to the family, including what is
known about the genetics of the patients condition, any
possible medical and developmental implications, and risk of
recurrence to other family members.
REFERENCES
40
1. Fraser FC: Genetic counseling. Am J Hum

Genet 1974; 26:636659.
2. Marks JH: 2003 ASHG award for
excellence in human genetics education.
The importance of genetic counseling. Am
J Hum Genet 2004; 74:395396.
3. Nussbaum RL, Suchy SF: The

oculocerebrorenal syndrome of Lowe (Lowe
syndrome). In: Scriver CR, Beaudet AL,
Sly WS, Valle D, eds. The metabolic and
molecular bases of inherited disease. 8th edn.
New York: McGraw Hill; 2001:62576266.
4. Hartong DT, Berson EL, Dryja TP: Retinitis

pigmentosa. Lancet 2006; 368:17951809.
5. Metabolic disorders. In: Gorlin RJ, Cohen
MM Jr, Hennekam RCM, eds. Syndromes
of the head and neck. 4th edn. New York:
Oxford University; 2001.
6. Baker DL, Schuette JL, Uhlmann WR eds:

A guide to genetic counseling. New York:
Wiley-Liss; 1998.
7. Bernhardt BA: Empirical evidence that
genetic counseling is directive: where do
we go from here? Am J Hum Genet 1997;
60:1720.
8. Furu T, Kaarianinen H, Sankilla EM, et al:
Attitudes towards prenatal diagnosis and
selective abortion among patients with

retinitis pigmentosa or choroideremia as
well as among their relatives. Clin Genet
1993; 43:160165.
9. Harper PS: Practical genetic counseling.
6th edn. Oxford: Butterworth-Heinemann;
2004.
10. Raz AE, Atar M: Nondirectiveness and its
lay interpretations: the effect of counseling
style, ethnicity and culture on attitudes

towards genetic counseling among Jewish
and Bedouin respondents in Israel. J Genet
Couns 2003; 12:313332.
11. Weil J: Psychosocial genetic counseling in
the post-nondirective era: a point of view.
J Genet Couns 2003; 12:199211.
12. Weil J: Psychosocial genetic counseling.
Oxford: Oxford University Press; 2000.
CHAPTER 4
41
SECTION 2
IMMUNOLOGY
Edited by C. Stephen Foster and M. Reza Dana
CHAPTER
Immunology An Overview
Reza Dana and C. Stephen Foster
All organisms live under the threat of attack from other living
organisms that express foreign, potentially immunogenic,
antigens. Additionally, a wide array of non-pathologic cellular
exposures (ultraviolet radiation from sun exposure, injury, etc.)
and responses (e.g., cell death, stress, and oxidation) can lead to
activation of immune responses to autoantigens. Among primitive single-celled eukaryotes, defense depends on physicochemical barriers at the cell surface and the capacity to engulf,
phagocytize, and digest the attacking pathogen. As multicellular
organisms evolved, and individual cells assumed differentiated
functions important to the well-being of the host, defense
against invading pathogens and complex immunoregulatory
pathways that ensure a measured response to immunogenic
insults, became the responsibility of specialized cells and molecules. The multifaceted array of sophisticated cells and molecules of the mammalian immune system is the evolutionary
descendant of these early forms of defense mechanisms.
The immune system found in mammals and higher
vertebrates is divided into two functionally distinct, but also
overlapping and interregulated, components termed innate
and adaptive immunity. Innate immunity is evolutionarily
more ancient and provides the host organism with an immediate protective response that does not require gene arrangement and is not antigen-specic. Adaptive immunity, by
contrast, provides protection that takes time to develop, is
antigen-specic, but is remembered through time (involves
memory), thereby allowing for efcient responses to be
generated in case of chronic or recurrent challenge by the
inciting antigen. Whereas innate immunity has the capacity to
recognize and respond to invading pathogens, the capacity to
accurately distinguish between self-molecules and molecules
of the pathogen (non-self) is much more highly developed in
the adaptive immune system (Table 5.1).
INNATE IMMUNITY
Innate, or natural, immunity consists of physicochemical
barriers, erected at interfaces between the host and the
environment, and a distinctive array of cells and molecules.13
Intact body surfaces, such as the skin and mucous membranes
with tight junctions among adjacent epithelial lining cells,
provide physical barriers to the entry of pathogens. In the case
of the eye, mechanical phenomena such as the wiping action of
eyelids, coverage of much of the epithelia with mucinous
glycoproteins, and the bulk flow of tears across the ocular
surface, all provide natural protection against pathogens. The
chemical components of body fluids (such as the tears)
including fatty acids, lysozyme, and complement components,
also make essential contributions to innate immunity. Finally,
cellular effectors of innate immunity include bone marrow-
TABLE 5.1. Characteristics of Innate vs Adaptive Immunity

Innate Immunity
Adaptive Immunity
Specificity
Not antigen-specific
Antigen-specific
Efficiency
Rapid
Primary responses
slower
Memory
Absent
Present
Chief Effectors
Neutrophils, macrophages,
natural-killer (NK) T cells
Lymphocytes
derived cells, including neutrophils, macrophages, and natural

killer cells, that are mobilized in the natural defenses against
invading pathogens.
Innate immunity is activated, for example, when an invading
bacterium, perhaps by releasing endotoxins or other bacterial
products, elicits a stereotypic inflammatory response by interacting with toll-like receptors on host cells, inducing microvascular dilatation, leukocyte inltration, and participation of
serum complement proteins. Innate immunity is also revealed
when a virus penetrates through the skin and evokes within the
draining lymph node an accumulation of natural killer cells
with the capacity to lyse virus-infected cells directly. In both of
these examples, the cells and molecules responsible for innate
immunity recognize and respond to the pathogen, but in neither
case is the recognition specic for the particular organism.
Moreover, if and when the attacker has been eliminated, the host
is not protected against a second invasion from the same agent
any more than it was the rst time, since there is no memory.
ADAPTIVE IMMUNITY
Adaptive, or acquired, immunity depends on a highly developed, sophisticated set of lymphoid organs (thymus, spleen,
lymph nodes, bone marrow, mucosa-associated lymphoid
tissues), cells (T and B lymphocytes, antigen-presenting cells
including dendritic and Langerhans cells, and macrophages),
and molecules (antibodies, cytokines, growth factors, and celladhesion molecules).1 The interactions between and among
these elements allow the adaptive immune system to meet four
important challenges as listed in Table 5.2.
FEATURES OF ADAPTIVE IMMUNITY

Certain features of the adaptive immune response set it apart
from all other ways in which an organism can respond to its
environment:
43
IMMUNOLOGY
TABLE 5.2. Critical Functions of the Immune System
SECTION 2
1. To create a repertoire of recognition structures (antibodies by

B cells, T cell receptors for antigen) that recognize biologically
important molecules in our universe
2. To eliminate or suppress lymphocytes whose recognition
structures bind to self-molecules and therefore threaten
autoimmunity and autoimmune disease
3. To create a diversity of effector mechanisms designed to
counter the diverse virulence strategies used by the many
different potential pathogens
4. To fashion immune responses in individual organs and tissues
such that protection is provided without interfering with the
tissues differentiated function
1. Adaptive immunity is acquired. Exposure of an adult

individual to a foreign antigen for the rst time leads to an
immune response that is rst detected (e.g., as antibody in
the blood) within 57 days. During the silent interval
after initial exposure, the adaptive immune system is
learning about the presence of the antigen. Thus, adaptive
immunity is acquired.
2. The immune response is specic for the eliciting antigen.
The antibodies that form within 57 days react with the
eliciting antigen alone and not with any other molecule
(unless there are shared structural residues between the
antigen that elicited the response and another antigen to
which the immune response is reacting). Exposure of the
same individual to a second (different) antigen elicits
another antibody response that is equally specic for the
second antigen and nonreactive with the rst antigen.
Thus, adaptive immunity is molecularly specic.
3. Reexposure of an individual to an antigen for a second
time elicits a response that is accelerated in onset and
exaggerated in amount. This means that what was
learned by the immune system during its rst exposure to
an antigen is remembered through time, and the
secondary response is the manifestation of that memory.
Thus, adaptive immunity is remembered.
4. Adaptive immunity can be transferred from an individual
who has it to another individual, thus conferring an
identical immunity on the recipient. Both antibodies and
specically sensitized lymphocytes are capable of
transferring adaptive immunity. Thus, adaptive immunity
is transferable.
5. Adaptive immunity can be specically prevented by
administering antigen under highly specialized, often
experimental, conditions. Individuals treated with antigen
in this manner may be rendered unable subsequently to
acquire immunity to the same antigen if administered in a
conventional fashion. Individuals rendered specically
unable to respond to a particular antigen are said to be
immunologically tolerant. Thus, tolerance is a
manifestation of adaptive immunity.
BENEFITS OF IMMUNITY
44
In mature mammals and higher vertebrates, both innate and

adaptive immune systems exist. Virtually every immune
response represents the summation of both innate and adaptive
responses, and the two systems are inextricably entwined.4 To
describe briefly the interplay between innate and adaptive
immunity, the following examples are given. Infection of the
lung with Streptococcus pneumoniae is prevented from
proceeding to consolidating pneumonia primarily by the innate
immune response. Neutrophils and, to a lesser extent, macrophages form the primary defense system, aided by acute-phase
reactants (e.g., C reactive protein) and members of the
complement cascade of proteins. The innate response in this
setting is activated to phagocytose and neutralize the invading
pathogen before large numbers of cells are infected. Adaptive
immunity, in which S. pneumoniae-specic antibodies are
produced, comes into play well after the primary infection has
already been contained, providing additional protection for that
infection, but most importantly also for protection once the
host gets reexposed to S. pneumoniae. In influenza virus infections of the lung, natural killer cells act early to limit virus
spread, but the infection appears to be terminated by virusspecic cytotoxic T cells that eliminate all virus-infected cells.
In parasitic infections, where clearance and elimination of the
organism may never be achieved, adaptive immunity plays the
key role in containing the organism in situ.
While the importance of immunity in infectious disease is
obvious, immunity is also believed to play a key role in the
control of neoplasms.5,6 Because tumors arise from host tissues,
the antigenic differences between tumors (non-self ) and self
tissues are necessarily narrower. On the one hand, this makes
it more difcult for the immune system to detect neoplastic
cells, and, on the other hand, raises the possibility that
immunity directed at antigens on tumor cells may spill over
onto normal tissues because of shared antigenic moieties. Still,
the immunity generated against neoplastic tissues is important,
manifested by the enhanced propensity of chronically
immunosuppressed individuals to a variety of malignancies.
HAZARDS OF IMMUNITY
There are two important ways in which immunity can harm
the host. First, most (if not all) immune responses that lead to
elimination of a pathogen require the participation of nonspecic host defense (innate immune) mechanisms. Because
they lack the high specicity of antibodies, T lymphocytes,
neutrophils, macrophages, and natural killer cells are incapable
of conning their destructive forces to pathogenic organisms.
Similarly, activated proteases of the complement system are
indiscriminate in their choice of substrates at the site of
infection. Thus, host tissues adjacent to an infection are usually
damaged, sometimes irreparably, by the intense inflammation
taking place in their midst. This penchant for innate immunity
to cause unwanted tissue damage is further enhanced by cells
and molecules of the adaptive immune system.1 For example,
the T cells that mediate delayed hypersensitivity responses
secrete cytokines that can serve as powerful attractants and
stimulants of macrophages and other leukocytes. As a
consequence, tissue injury and death is almost an invariant
outcome of delayed hypersensitivity responses directed at
infecting pathogens. Similarly, complement-xing antibodies
recruit and amplify the participation of neutrophils and
macrophages at the site where they bind target pathogens, leading to exaggerated inflammation and necrosis. Thus, immunity
can inadvertently produce injury to otherwise healthy host
tissues, and immunopathogenic mechanisms are important
causes of disease in many different organs and tissues.
Second, the adaptive immune response must meet the
challenge of eliminating or suppressing T and B cells with
recognition structures (e.g., T cell receptors) specic for selfantigens, so-called autoreactive lymphocytes.7 This is one of
the central tenets of central tolerance that allows the thymus
to delete such autoreactive cells from circulation. When this
challenge is not met, autoimmunity can arise. In truth, not all
autoimmunity is deleterious. For example, there is evidence
suggesting that immunity against certain self-components may
be a necessary part of the healing response to injury and

infection. However, certain types of autoimmunity are destructive, and these can give rise to tissue-restricted inflammatory
diseases. Examples abound, including rheumatoid arthritis,
Sjgrens syndrome, uveitis, inflammatory bowel disease, and
others. A hierarchy of self-antigens exists, dictated by the extent
to which the antigens are accessible to lymphocytes of the
systemic immune apparatus. For instance, circulating plasma
proteins have an extremely low potential for evoking an autoimmune response. By contrast, proteins expressed on cells found
only in the eye (e.g., photoreceptors) or testis (spermatozoa)
have a high potential for eliciting an autoimmune response. In
addition, tissue-restricted factors (e.g., bloodtissue barriers)
influence whether a response that is autoimmune becomes
immunopathogenic and therefore causes disease.
SPECIAL CASE OF THE EYE: IMMUNE

PRIVILEGE
Most organs of the body can sustain substantial amounts of
permanent damage from immune and inflammatory reactions
without losing appreciable function. For example, inflammation
in the skin, heart, liver, kidney, and bone can be associated
with the typical consequences of inflammation-damage to the
normal cells of the organ and scarring from the compensatory
reparative processes associated with injury. These organs, however, are very forgiving, in that they can each sustain substantial
amounts of inflammation (provided that it is temporary) and
still retain sufcient viability after the reparative processes to
carry on the normal functions required for normal living
activities. The same is not true for the eye.
Inflammation that in other tissues would be trivial is not
tolerated well by the eye and visual system. The vulnerability of
the eye to even small amounts of inflammation derives from
the need to preserve the anatomic integrity of the visual axis.
Very slight alterations in components of the visual axis prevent
light images from landing precisely on the retina, causing signicant visual impairment. Thus, innocent bystander damage
to ocular tissues during the course of inflammation can be
associated with a profound loss of function (i.e., blindness or
substantial impairment of useful vision). For example, even
slight temporary inflammation in the central part of the cornea
can have substantial, long-term effects on functional visual
acuity after resolution of the inflammation, simply because the
reparative processes result in disorganization of the normally
ordered arrangement of collagen brils within the corneal
stroma, an organization that is critical to continued clarity in
the cornea. Similarly, inflammation involving the retina
(especially the macula), the vitreous, and the uveal tract can
also produce signicant loss in visual function.
Thus, the eye is confronted with a dilemma. On the one
hand, the eye is covered by a mucosal surface that leaves it
largely exposed to the myriad noxious stimuli of the environ-
ment, including microbial pathogens, and needs the protection

afforded by the immune system. And yet, on the other hand,
immunity is necessarily mediated in part by nonspecic host
defense mechanisms that carry the threat of innocent bystander
injury. To resolve this dilemma evolutionarily, the eye and the
immune system have arranged a compromise in which certain
forms of immunity are permitted, whereas others are suppressed. This compromise is expressed experimentally in the
phenomenon of immune privilege.8 It has been known for
more than a century that foreign tissues implanted in the
anterior chamber of the eye enjoyed prolonged survival compared with the fate of foreign tissues implanted at conventional
body sites. In the 1950s, Medawar correctly inferred that the
ability of foreign grafts to survive in the eye was due to a failure
of immunologic rejection.9 At the time, Medawar proposed that
immune privilege resulted from sequestration of intraocular
antigenic material from the systemic immune apparatus. The
term immunologic ignorance has been used to identify this
situation. However, in recent years, it has become clear that
ocular immune privilege is a state that is actively maintained by
a variety of immunoregulatory mechanisms, rather than simply
antigenic sequestration afforded by physical and tight junction
barriers.
Immune privilege is an actively acquired and maintained
state in which ocular factors, acting on cells of the immune
system, suppress both the induction and expression of
immunity within the eye, and alter the induction of systemic
immunity to ocular antigens, leading to a stereotypic systemic
immune response called anterior chamber associated immune
deviation (ACAID).10 As a consequence, systemic immune
responses to eye-derived antigens are decient in T cells that
mediate delayed hypersensitivity and in antibodies that activate
complement components. Thus, systemic immunity engendered by eye-derived antigens lacks the two effector modalities
most closely linked to intense inflammation and innocent
bystander injury-delayed hypersensitivity and complementxing antibodies.
It is important to emphasize that immune privilege in the eye
is not simply the consequence of a failed immune response;
rather, it results from modications in the immune response
that afford immune protection for the eye that carries a minimal threat to nonspecic injury. The importance of this
understanding lies in the implications that it holds for the
diagnosis and treatment of ocular inflammatory and infectious
disorders. The sections and chapters that follow are designed to
provide more specic information to ophthalmologists and
vision scientists about the cells and molecules that affect and
regulate inflammation and immunity in the eye.
CHAPTER 5
Immunology An Overview
ACKNOWLEDGMENT
The authors would like to acknowledge the signicant material contribution
of Dr J Wayne Streilein to the previous edition of this chapter.
REFERENCES
1. Janeway CA Jr, Travers P, eds:
Immunobiology. 6th edn. New York:
Garland Publishing Inc; 2004.
2. Akira S, Uematsu S, Takeuchi O: Pathogen
recognition and innate immunity. Cell 2006;
124:783801.
3. Koehn B, Gangappa S, Miller JD, et al:
Patients, pathogens, and protective
immunity: the relevance of
virus-induced alloreactivity in
transplantation. J Immunol 2006;
176:26912696.
4. Pulendran B, Ahmed R: Translating innate

immunity into immunological memory:
implications for vaccine development.
Cell 2006; 124:849863.
5. Moller G: Tumor immunology. Immunol Rev
1995; 145:112.
6. Karin M, Lawrence T, Nizet V: Innate
immunity gone awry: linking microbial
infections to chronic inflammation and
cancer. Cell 2006; 124:823835.
7. Hogquist KA, Baldwin TA, Jameson SC:
Central tolerance: learning self-control in
the thymus. Nat Rev Immunol 2005;

5:772782.
8. Streilein JW: Perspective: unraveling
immune privilege. Science 1995; 270:1158.
9. Medawar PB: Immunity to homologous
grafted skin. III. The fate of skin homografts
transplanted to the brain, to subcutaneous
tissue, and to the anterior chamber of the
eye. Br J Exp Pathol 1948; 29:58.
10. Streilein JW: Ocular immune privilege and
the Faustian dilemma. Invest Ophthalmol
Vis Sci 1996; 37:19401950.
45
CHAPTER
A Cast of Thousands: The Cells of the Immune

System
C. Stephen Foster
The cellular components of the immune system include lymphocytes, macrophages, Langerhans cells, neutrophils, eosinophils, basophils, and mast cells. Many of these cell types can
be further subdivided by subtypes and subsets. For example,
lymphocytes include T lymphocytes, B lymphocytes, and nonT, non-B (null) lymphocytes. Each type can be further subcategorized, both by functional differences and by differences in
cell-surface glycoprotein specializations and uniqueness. The
latter differentiating aspect of cell types and cell-type subsets
has been made possible through the development of hybridomamonoclonal antibody technology. This phenomenon of cellsurface glycoprotein specialization and uniqueness among cell
types, and the technology for identifying those unique differences among cell types, are so important that a synopsis of
the evolution and current understanding of this phenomenon
follows.
Jeorges Kohler and Cesar Milstein, at Cambridge University,
succeeded in immortalizing antibody-producing cells in 1975
by fusing them with myeloma tumor cells using a myeloma cell
line with a selective deciency of hypoxanthine phosphoribosyltransferase.1 These researchers developed a technique for
successfully recovering only the cells that had successfully fused
to the myeloma cells (i.e., the hybridomas). Only the hybridoma
cells survived in a tissue culture medium containing hypoxanthine, aminopterin, and thymidine, because the antibodyforming cell component of the hybridoma contributed enough
hypoxanthine phosphoribosyltransferase to ensure survival of
the hybrid. Selecting individual hybrids that produce the desired
antibody against a particular immunogen (antigen or antigenicdeterminant or epitope) and then allowing that hybrid cell
(hybridoma) to proliferate generated an immortal monoclonal
cell population (i.e., a hybrid cell population derived from a
single original cell) and thus produced a never-ending supply of
a highly specic antibody (monoclonal antibody) directed
against the original immunogen of interest. For this innovative
and important work, these researchers were awarded the Nobel
Prize for Medicine in 1984.
Reinherz and Schlossman2 exploited the monoclonal
antibody technology in the late 1970s, rst taking advantage of
the fact that T lymphocytes possess well-known, unique cellsurface determinants (e.g., a binding receptor for sheep
erythrocytes), which made it possible to separate T lymphocytes
into pure preparations from peripheral blood lymphocytes.
Immunization of mice with such a puried preparation of T
cells, with subsequent preparation of hybridomas from spleen
cell populations harvested from those immunized mice, was
followed by screening and selection of hybridomas that
synthesize antibodies that would stick to the cell surface of T
cells and by cloning of these hybridomas. This same strategy or
similar strategies based on functional assays (e.g., beginning
with cells that were efcient at helping an immune response to

develop or beginning with cells that efciently suppressed an
immune response) resulted in the additional development of
monoclonal antibody reagents that were specic for and
identied the two major T lymphocyte subsets, helper-inducer
T cells, and suppressor-cytotoxic T cells.
Because the original work was performed in collaboration
with Ortho Pharmaceuticals, the original designation of the
cell-surface determinants for T cells was OKT 3, the designation for helper-inducer T cells was OKT 4, and that for
suppressor-cytotoxic T cells OKT 8. As additional companies
began to develop their reagents using the same technology,
additional naming schemes developed, and the name game for
cell-surface determinants became extremely complicated.
Investigator workshops have now generated a universal nomenclature system for cell-surface glycoproteins, or antigens,
and this system is based on the so-called clusters of
differentiation designation. Hence, the proper designation for
the cell-surface glycoprotein unique to T cells is now CD3, and
the designation for the cell-surface glycoprotein unique to
helper/inducer T cells is CD4. Table 6.1 presents a partial list of
current clusters of differentiation designations and the cell
types that express these CD antigens.
LYMPHOCYTES
Lymphocytes are mononuclear cells, round, 78 mm in
diameter, found in lymphoid tissue (lymph node, spleen,
thymus, gut-associated lymphoid tissue, mammary-associated
lymphoid tissue, and conjunctiva-associated lymphoid tissue)
and in blood. They ordinarily constitute ~30% of the total peripheral white blood cell count. The lymphocyte is the premier
character in the immune drama; it is the primary recognition
unit for foreign material, the principal specic effector cell type
in immune reactions, and the cell exclusively responsible for
immune memory.
T lymphocytes, or thymus-derived cells, compose 6580% of
the peripheral blood lymphocyte population, 3050% of the
splenocyte population, and 7085% of the lymph node cell
population. B lymphocytes compose 515% of peripheral blood
lymphocytes, 2030% of splenocytes, and 1020% of lymph
node cells.
T cells possess cell-surface receptors for sheep erythrocytes
and for the plant-derived mitogens concanavalin A and phytohemagglutinin. They do not possess surface immunoglobulin or
surface membrane receptors for the Fc portion of antibody-two
notable cell-surface differences from B lymphocytes, which do
possess these two entities. B cells also exhibit cell-surface
receptors for the third component of complement, for the
EpsteinBarr virus and the plant mitogen known as pokeweed
47
IMMUNOLOGY
SECTION 2
TABLE 6.1. Clusters of Differentiation (CD) Designations

Clusters
Cell Specificity
Function
CD1a b c d
Thymocytes, Langerhans cells dendritic cells,

B cells (CD1c), intestinal epithelium, smooth
muscle, blood vessels (CD1d)
MHC class I-like molecule, associated with b

2-microglobulin. Role in presentation of lipid antigens
CD2
T cells, NK subset
Receptor/sheep erythrocyte receptor; adhesion molecule

binds to LFA-3 (CD58), binds Lck intracellularly and
activates T cells
CD3
T cells
T-cell antigen-complex receptor
CD4
Helperinducer T cells, TH1 and TH2T cells
MHC class II immune recognition; HIV receptor (HIV-1 and

HIV-2 gp120)
CD5
T cells, B-cell subset
Scavenger receptor
CD6
T cell, subset, B cells in chronic lymphatic leukemia
Binds CD166 (scavenger receptor)
CD7
T cells, NK cells, platelets
Binds PI 3-kinase. Marker for T cell acute lymphatic

leukemia and pluripotential stem cell leukemias
CD8
Cytotoxic suppressor T cells
MHC class I immune recognition, binds Lck kinase
CD9
Pre-B cells, monocytes, eosinophils, basophils,

platelets, activated T cells, brain and peripheral
nerves, vascular smooth muscle
Mediates platelet aggregation and activation via FcgRIIa,

may play a role in cell migration
CD10
Common acute
lymphocytic
leukemia
antigen (CALLA)
Pre-B cells, neutrophils
Neutrophil endopeptidase.
Zinc metalloproteinase, marker for pre-B acute lymphatic
leukemia (ALL)
CD11a
Leukocytes
Adhesion molecule (LFA-1) binds to CD54 (ICAM-1), CD102

(ICAM-2), and CD50 (ICAM-3)
CD11b ( Mac-1)
Monocytes, granulocytes, NK cells
a-Chain of complement receptor CR3;, binds CD54,

complement component iC3b, and extracellular matrix
proteins
CD11c
Monocytes, granulocytes, NK cells
Adhesion (aX subunit of integrin CR4 (associated with

CD18), binds fibrinogen)
CD11d
Leukocytes
aD subunits of integrin; associated with CD18; binds to

CD50
CDw12
Monocytes, granulocytes, platelets
Unknown
CD13
Monocytes, granulocytes,
Aminopeptidase N (Zinc metalloproteinase)
CD14
Macrophages
Lipopolysaccharide receptor
CD15
Neutrophils, activated T cells, eosinophils
Terminal trisaccharide expressed on glycolipids and many

cell-surface glycoproteins
CD15s
Leukocytes, endothelium
Ligand for CD62E, P
CD15u
Sulphated CD15
Terminal trisaccharide expressed on glycolipids and many

cell-surface glycoproteins
CD16
Granulocytes, macrophages, NK cells
Fc receptor IgG (Fc-g RIII); activation of NK cells
CDw17
Neutrophils, monocytes, platelets
Lactosyl ceramide, a cell-surface glycosphingolipid
CD18
Leukocytes
Intergrin b2 subunit; associates with CD11a, b, c, and d
CD19
B cells
B-cell activation (binds tyrosine kinases and PI 3-kinase)
CD20
B cells
B-cell activation (oligomers from a calcium channel)
CD21
B cells
Complement receptor CR2 (C3d) EpsteinBarr virus

receptor
CD22
B cells
Adhesion; B-cell activation
CD23
Activated B cells, macrophages, activated

macrophages, eosinophils, follicular dendritic
cells, platelets
Low-affinity Fc-e receptor, induced by IL-4
CD24
B cells, granulocytes
Unknown
CD25
Activated T cells, B cells
IL-2 receptor
Continued
48
A Cast of Thousands: The Cells of the Immune System
Clusters
Cell Specificity
Function
CD26
Activated B and T cells, macrophages
Exopeptidase, cleaves N terminal X-Pro or X-Ala dipeptides

from polypeptides
CD27
Medullary thymocytes, T cells, NK cells, some B cells
TNF receptor, Binds CD70; can function as a co-stimulator

for T and B cells
T cells
Receptor for co-stimulator molecules B7.1 (CD80) and B7.2

(CD86)
CD28
CD29
Leukocytes
Integrin b1 subunit, associates with CD49a in VLA-1 integrin
CD30
Activated B and T cells
Binds CD30L (CD153); cross-linking CD30 enhances

proliferation of B and T cells
CD31
Platelets, monocytes, and B cells
Role in leukocyteendothelial adhesion (PECAM-1 mediated

leukocyte-endothelial and endothelial-endothelial
interactions)
CD32
B lymphocytes, granulocytes, macrophages,

eosinophils
Fc receptor IgG (Fc-gRIII) ADCC
CD33
Myeloid progenitor cells, monocytes
Binds sialoconjugates
CD34
Hematopoietic precursors, capillary endothelium
Ligand for CD62L (L-selectin)
CD35
B cells, erythrocytes, neutrophils, mononuclear cells
Complement receptor CR1 (binds C3b and C4b, mediates

phagocytosis)
CD36
Platelets, monocytes, endothelial cells
Platelet adhesion molecule, (GPIV, GPIIIb) involved in

recognition and phagocytosis of apoptosed cells
CD37
B cells
Unknown, may be involved in signal transduction
CD38
Activated T and plasma cells, early B and T cells
NAD glycohydrolase, augments B cell proliferation
CD39
Activated B cells, activated NK cells, macrophages,

dendritic cells
Unknown, may mediate adhesion of B cells
CD40
B cells
Co-stimulatory molecule for B-cell activation by T-cell

contact binds CD154 (CD40L), promotes growth,
differentiation, and isotype switching of B cells
CD41
Megakaryocytes, platelets
Associates with CD61 to form GPIIb; binds fibrinogen,

fibronectin, von Willebrand factor, and thrombospondin;
Fn receptor,
CD42 a,b,c,d
Megakaryocytes, platelets
GpIb platelet adhesion; binds von Willebrand factor,

thrombin
CD43
Leukocytes
T-cell activation
CD44
Leukocytes
Pgp1 (Hermes) receptor; homing receptor for matrix

components (e.g., hyaluronate)
CD45
All leukocytes
Leukocyte common antigen signal transduction

(tyrosine phosphatase)
CD45RA
Naive cells
CD45RO
Activated/memory T cells
CD45RB
B cells, T-cell subsets, monocytes, macrophages,

granulocytes
CD46
Hematopoietic and nonhematopoietic nucleated cells
Membrane co-factor protein; binds to C3b and C4b to

permit their degradation by Factor I
CD47
All cells
Adhesion molecule; thrombospondin receptor
CD48
Leuckocytes
Putative ligand for CD244
CD49a (VLA-1)
Activated T cells, monocytes, neuronal cells, smooth

muscle
a1 integrin; associates with CD29; binds collagen, laminin-1
CD49b (VLA-2)
B cells, monocytes, platelets, megakaryocytes,

neuronal, epithelial and endothelial cells, osteoclasts
a2 integrin; associates with CD29; binds collagen, laminin
CD49c (VLA-3)
B cells, many adherent cells
a3 integrin; associates with CD29; binds laminin-5,

fibronectin, collagen, entactin, invasin
CHAPTER 6
TABLE 6.1. Clusters of Differentiation (CD) DesignationsContd
Continued
49
IMMUNOLOGY
SECTION 2

Clusters
Cell Specificity
Function
CD49d (VLA-4)
Broad distribution includes B cells, thymocytes,

monocytes, granulocytes, dendritic cells
a4 integrin; associates with CD29; binds fibronectin,

MAdCAM-1, VCAM-1
CD49e (VLA-5)
Broad distribution includes memory T cells, monocytes,

platelets
a5 integrin; associates with CD29; binds fibronectin,

invasin
CD49f (VLA-6)
T lymphocytes, monocytes, platelets, megakaryocytes,

trophoblasts
a6 integrin; associates with CD29; binds laminin, invasin,

merosine
CD50 (ICAM3)
Thymocytes, T cells, B cells, monocytes, granulocytes
Binds integrin CD11a/CD18
CD51
Platelets, megakaryocytes
aV integrin; associates with CD61; binds vitronectin, von

Willebrand factor, fibrinogen, and thrombospondin; may
be receptor for apoptotic cells
CD52
(CAMPATH 1)
Thymocytes, T cells, B cells (not plasma cells),

monocytes, granulocytes, spermatozoa
Unknown
CD53
Leukocytes
Unknown
CD54 (ICAM-1)
Activated cells
Adhesion to LFA-1 (CD11a/CD18 integrin) and MAC

1(CD11b/CD18); rhinovirus receptor
CD55
Hematopoietic and nonhematopoietic cells
Decay accelerating factor (DAF); binds C3b; disassembles

C3/C5 convertase
CD56
NK
NCAM (neural cell adhesion molecule) adhesion
CD 57
NK cells, subsets of T cells, B cells, and monocytes
Oligosaccharide, found on many cell-surface glycoproteins
CD58 (LFA-3)
B cells, antigen-presenting cells
Binds to CD2
CD59
Binds complement components C8 and C9; blocks

assembly of membrane attack complex
CD61
Platelets, megakaryocytes, macrophages
Intergrin b3 subunit; associates with CD41 (GPIIb/IIIa) or

CD51 (vitronectin receptor)
CD62E (E-selectin,
ELAM-1)
Endothelial cells
Adhesion (binds CD34, GlyCAM, mediates

rolling interactions with endothelium)
CD62L
(L-selectin,
LAM-1)
T cells, B cells
Adhesion (binds CD34, GlyCAM, mediates rolling

interactions with endothelium)
CD62P (P-selectin)
PADGEM
Platelets, endothelial cells, megakaryocytes
Adhesion (binds CD162 (PSGL-1), mediates interaction

of platelets with endothelial cells, monocytes, and rolling
leukocytes on endothelium)
CD63
Activated platelets, monocytes, macrophages
Unknown
CD64
Monocytes, macrophages
Adhesion, FC-g receptor; antibody-dependent, cell

mediated cytotoxicity
CD65
Myeloid cells
Oligosaccharide component of a ceramide

dodecasaccharide
CD66a
Neutrophils
Unknown
CD66b
Granulocytes
Unknown
CD66c
Neutrophils
Unknown
CD66d
Neutrophils
Unknown
CD66e
Adult colon epithelium, colon carcinoma
Unknown
CD60a,b,c
CD66f
50
Unknown
CD68
Monocytes, macrophages, neutrophils, basophils,

large lymphocytes
Unknown
CD69
Activated lymphocytes
Unknown
CD70
Activated T and B cells, and macrophages
Ligand for CD27
CD71
Proliferating cells
Transferrin receptor
CD72
B cells
Ligand for CD5; B cell T cell interactions
CD73
B and T cells
Ecto-5-nucleotidase; dephosphorylates nucleotides to

allow nucleoside uptake
Continued
Clusters
Cell Specificity
Function
CD74
B cells, macrophages, monocytes, MHC class II

positive cells
MHC class II-associated invariant chain
CD75
Mature B cells, T-cells subsets
Lactosamines; ligand for CD22; mediates B-cell-B-cell

adhesion
CD75s
Mature B cells, T-cells
subsets a-2,6-sialylated lactosamines
CD77
Germinal center B cells
Neutral glycosphingolipid; binds Shiga toxin; cross-linking

induces apoptosis
CD79
B cells
Components of B-cell antigen receptor analogous to CD3;

required for cell-surface expression and signal transduction
CD80 (B7-1)
B cells, dendritic cells, macrophages
Ligand for CD28 and CTLA4; co-stimulator for T-cell

activation
CD81
Lymphocytes
Associates with CD19, CD21 to form B cell co-receptor
CD82
Leukocytes
Unknown
CD83
Leukocytes
Unknown
CDw84
Monocytes, platelets, circulating B cells
Unknown
CD85
Dendritic cells
ILT/LIR family
CD86
Monocytes, activated B cells, dendritic cells
Ligand for CD28 and CTLA4
CD87
Granulocytes, monocytes, macrophages, T cells,

NK cells, wide variety of nonhematopoietic cell types
Receptor for urokinase plasminogen activator
CD88
Polymorphonuclear leukocytes, macrophages, mast cells Receptor for complement component C5a
CD89
(Fc-a receptor)
Neutrophils, monocytes
CD90
CD34 + prothymocytes (human), thymocytes
Unknown
CD91
Monocytes, many nonhematopoietic cells
a2-macroglobulin receptor
CD92
Neutrophils, monocytes, platelets, endothelium
Unknown
CD93
Neutrophils, monocytes, endothelium
Unknown
CD94
T-cell subsets, NK cells
Unknown
CD95 (Fas)
Multiple cell types
Role in programmed cell death (Bbinds TNF-like Fas ligand)
CD96
Activated T cells, NK cells
Unknown
CD97
Activated B and T cells, monocytes, granulocytes
Binds CD55
CD98
T cells, B cells, natural killer cells, granulocytes, all

human cell lines
Unknown
CD99
Peripheral blood lymphocytes, thymocytes
Unknown
CD100
Hematopoietic cells
Unknown
CD101
Monocytes, granulocytes, dendritic cells, activated

T cells
Unknown
CD102 (ICAM-2)
Endothelial cells, monocytes
Ligand for LFA-1 integrin (CD11a/CD18)
CD103 (HML-1)
T cells
Role in T-cell homing to mucosae
CD104
CD4 CD8 thymocytes, neuronal, epithelial, and some

endothelial cells, Schwann cells, trophoblasts
Integrin b4 associates with CD49f;, binds laminins
CD105
Endothelial cells, activated monocytes and

macrophages, bone marrow cell subsets
Binds TGF-b
CD106 (VCAM-1)
Endothelial cells, macrophages
Receptor for VLA-4 integrin; adhesion
CD107a,b
Activated platelets, activated T cells, activated

neutrophils, activated endothelium
Unknown
CD108
Erythrocytes, circulating lymphocytes, lymphoblasts
Unknown
CD109
Activated T cells, activated platelets, vascular

endothelium
Unknown
CD110
Platelets
MPL, TPO R
CHAPTER 6
IgA-dependent cytotoxicity
Continued
51
IMMUNOLOGY
SECTION 2

Clusters
Cell Specificity
Function
CD111
Myeloid cells
PPR1/Nectin1
CD112
Myeloid cells
PPR2
CD114
Granulocytes, monocytes
Granulocytes colony-stimulating factor (G-CSF) receptor
CD115
Macrophage colony-stimulating factor (M-CSF) receptor
CD116
Monocytes, neutrophils, eosinophils, endothelium
Granulocyte-macrophage colony-stimulating factor

(GMCSF) receptor a chain
CD117
Hematopoietic progenitors
Stem-cell factor (SCF) receptor
CD118
Many cell types
Interferon-a, b receptor
CD119
Macrophages, monocytes, B cells, endothelium
Interferon-g receptor
CD120a,b
TNF receptor; binds both TNF-a and TNF-b
CD121a
Thymocytes, T cells
Type I interleukin-1 receptor; binds IL-1a and IL-b
CDw121b
B cells, macrophages, monocytes
Type II interleukin-1 receptor; binds IL-1a and IL-1b
CD122
NK cells, resting T-cell subsets, some B-cell lines
IL-2 receptor b chain
CD123
Bone marrow stem cells, granulocytes, monocytes,

megakaryocytes
IL-3 receptor a chain
CD124
Mature B and T cells, hematopoietic precursor cells
IL-4 receptor
CD125
Eosinophils, basophils, activated B cells
IL-5 receptor
CD126
Activated B cells and plasma cells (strong), most

leukocytes (weak)
IL-6 receptor a subunit
CD127
Bone marrow lymphoid precursors, pro-B cells, mature

T cells, monocytes
IL-7 receptor
CDw128
Neutrophils, basophils, T-cell subsets
IL-8 receptor
CD129
52
Unknown
CD130
Most cell types, especially activated B cells and plasma

cells
Common subunit of IL-6, IL-11, oncostain-M (OSM)

and leukemia inhibitory factor (LIF) receptors
CDw131
Myeloid progenitors, granulocytes
Common b subunit of IL-3, IL-5, and GM-CSF receptors
CD132
B cells, T cells, NK cells, mast cells, neutrophils
IL-2 receptor g chain; common subunit of IL-2, IL-4, IL-7,

IL-9, and IL-15 receptors
CD133
Stem/progenitor cells
AC133
CD134
Activated T cells
May acts as adhesion molecule co-stimulator
CD135
Multipotential precursors, myelomonocytic and B-cell

progenitors
Growth factor receptor
CDw136
Monocytes, epithelial cells, central and peripheral

nervous system
Chemotaxis, phagocytosis, cell growth, and differentiation
CDw137
T and B lymphocytes, monocytes, some epithelial cells
Co-stimulator of T-cell proliferation
CD138
B cells
Heparan sulphate proteoglycan binds collagen type I
CD139
B cells
Unknown
CD140a.b
Stromal cells, some endothelial cells
Platelet-derived growth factor (PDGF) receptor a and b

chains
CD141
Vascular endothelial cells
Anticoagulant; binds thrombin, the complex then activates

protein C
CD142
Epidermal keratinocytes, various epithelial cells,

astrocytes, Schwann cells
Inducible by inflammatory mediators Binds Factor VIIa;

this complex activates Factors VII, IX, and X in blood
clotting
CD143
Endothelial cells, except large blood vessels and kidney,

epithelial cells of brush borders of kidney and small
intestine, neuronal cells, activated macrophages and
some T cells.
Soluble form in plasma Zn 2+ metallopeptidase dipeptidyl

peptidase; cleaves angiotensin I and bradykinin from
precursor forms
CD144
Endothelial cells
Organizes adherens junction in endothelial cells (cadherin)
CD145
Endothelial cells, some stromal cells
Unknown
CD146
Endothelium
Potential adhesion molecule, localized at cell-cell junctions

Continued
Clusters
Cell Specificity
Function
CD147
Leukocytes, red blood cells, platelets, endothelial cells
Potential adhesion molecule
CD148
Granulocytes, monocytes, dendritic cells, T cells,

fibroblasts, nerve cells
Contact inhibition of cell growth
CD150
Thymocytes, activated lymphocytes
Unknown
CD151
Platelets, megakaryocytes, epithelial cells, endothelial

cells
Associates with b integrins
CD152 (CTLA 4)
Activated T cells
Receptor for B7.1 (CD80), B7.2 (CD86); negative regulator

of T-cell activation
CD153
Activated T cells, activated macrophages, neutrophils,

B cells
Ligand for CD30, may co-stimulate T cells
CD154
Activated CD4 T cells
Ligand for CD40; inducer of B-cell proliferation and

activation
CD155
Monocytes, macrophages, thymocytes, CNS neurons
Normal function unknown; receptor for polio virus
CD156a.b
Neutrophils, monocytes
Unknown
CD157
Granulocytes, monocytes, bone marrow stromal cells,

vascular endothelial cells, follicular dendritic cells
ADP-ribosyl cyclase; cyclic

ADP-ribose hydrolase
CD158a,b
NK cells
Inhibits NK cell cytotoxicity
CD159a
NK cells
Binds CD94 to form NK receptor; inhibits NK cell

cytotoxicity on binding MHC class I molecules
CD160
T cells
Unknown
CD161
NK cells, T cells
Regulates NK cytotoxicity
CD162
Neutrophils, lymphocytes, monocytes
Ligand for CD62P
CD162R
NK cells
Unknown
CD163
Unknown
CD164
Epithelial cells, monocytes, bone marrow stromal cells
Unknown
CD165
Thymocytes, thymic epithelial cells, CNS neurons,

pancreatic islets, Bowmans capsule
Adhesion between thymocytes and thymic epithelium
CD166
Activated T cells, thymic epithelium, fibroblasts, neurons
Ligand for CD6; involved integrin neurite extension
CD167a
Normal and transformed epithelial cells
Binds collagen
CD168
Breast cancer cells
Adhesion molecule.
CD169
Some macrophages
Adhesion molecule.
CD170
Neutrophils
Adhesion molecule
CD171
Neurons, Schwann cells, lymphoid and myelomonocytic

cells, B cells, CD4 T cells
Adhesion molecule, binds CD9, CD24, CD56
CD172a
Unknown
Adhesion molecule; is a substrate of activated receptor

tyrosine kinases and binds to SH2 domains
CD173
All cells
Blood group H type 2; carbohydrate moiety
CD174
All cells
Lewis y blood group; carbohydrate moiety
CD175
All cells
Tn blood group; carbohydrate moiety
CD175s
All cells
Sialyl-Tn blood group; carbohydrate moiety
CD176
All cells
TF blood group; carbohydrate moiety
CD177
Myeloid cells
Unknown
CD178
Activated T cells
Fas ligand; binds to Fas to induce apoptosis
CD179a
Early B cells
Associates noncovalently with immunoglobulin l-like

polypeptide 1 to form a surrogate light chain that is
selectively expressed at the early stages of B-cell
development. Mutations in the CD179b gene have been
shown to result in impairment of B-cell development and
agammaglobulinemia in humans
CD179b
Associates noncovalently with immunoglobulin iota chain to

form a surrogate light chain (a component of the pre-Bcell receptor which plays a critical role in early B-cell
differentiation)
Continued
CHAPTER 6
53
IMMUNOLOGY
SECTION 2
54
Clusters
Cell Specificity
Function
CD180
B cells
Membrane protein consisting of extracellular leucine-rich

repeats
CD183
Malignant B cells from chronic lymphoproliferative

disorders
CXC chemokine receptor involved in chemotaxis of

malignant B lymphocytes
CD184
Immature CD34 + haematopoietic stem cells 1
Binding to SDF-1 (LESTR/fusin); acts as a co-factor for

fusion and entry of T-cell line; trophic strains of HIV-
CD195
Promyelocytic cells
Receptor for a CC type chemokine; binds to MIP-1a,

MIP-1b and RANTES; may play a role in the control of
granulocytic lineage proliferation or differentiation; acts as
co-receptor with CD4 for HIV-1
CDw197
Activated B and T lymphocytes, strongly upregulated in

B cells infected with EBV and T cells infected with
HHV6 or 7
Receptor for the MIP-3b chemokine; probable mediator

of EBV effects on B lymphocytes or of normal lymphocyte
functions
CD200
Normal brain and B-cell lines
Unknown
CD201
Endothelial cells
Endothelial cell-surface receptor that binds with high-affinity

to protein C and activated protein C; downregulated by
exposure of endothelium to tumor necrosis factor
CD202b
Endothelial cells
Receptor tyrosine kinase, binds angiopoietin-1; important in

angiogenesis, particularly for vascular network formation
in endothelial cells; defects in TEK are associated with
inherited venous malformations; the TEK signaling
pathway appears to be critical for endothelial cell-smooth
muscle cell communication in venous morphogenesis
CD203c
Myeloid cells
Ectoenzymes that are involved in hydrolysis of extracellular

nucleotides. They catalyze the cleavage of phosphodiester
and phosphosulfate bonds of a variety of molecules,
including deoxynucleotides, NAD, and nucleotide sugars
CD204
Myeloid cells
Mediate the binding, internalization, and processing of a

wide range of negatively charged macromolecules;.
Iimplicated in the pathologic deposition of cholesterol in
arterial walls during atherogenesis
CD205
Dendritic cells
Lymphocyte antigen 75; putative antigen-uptake receptor on

dendritic cells
CD206
Macrophages, endothelial cells
Type I membrane glycoprotein; only known example of a

C-type lectin that contains multiple C-type CRDs
(carbohydrate-recognition domains); it binds highmannose structures on the surface of potentially
pathogenic viruses, bacteria, and fungi
CD207
Langerhans cells
Type II transmembrane protein; Langerhans cell specific

C-type lectin; potent inducer of membrane superimposition
and zippering leading to BG (Birbeck granules) formation
CD208
Interdigitating dendritic cells in lymphoid organs
Homologous to CD68, DC-LAMP is a lysosomal protein

involved in remodeling of specialized antigen-processing
compartments and in MHC class II-restricted antigen
presentation; upregulated in mature DCs induced by
CD40L, TNF-a and LPS.
CD209
Dendritic cells
C-type lectin; binds ICAM3 and HIV-1 envelope glycoprotein

gp120 enables T-cell receptor engagement by
stabilization of the DC/T-cell contact zone, promotes
efficient infection in trans cells that express CD4 and
chemokine receptors; type II transmembrane protein
CDw210
B cells, T-helper cells
Interleukin 10 receptor a and b
CD212
Activated CD4, CD8, and NK cells
IL-12 receptor b chain; a type I transmembrane protein

involved in IL-12 signal transduction.
CD213a1
B cells, monocytes, fibroblasts, endothelial cells
Receptor which binds IL-13 (low affinity); together with IL

4Ra can form a functional receptor for IL-13, also serves
as an alternate accessory protein to the common
cytokine receptor gamma chain for IL-4 signaling
CD213a2
B cells, monocytes, fibroblasts, endothelial cells
IL-13 receptor which binds as a monomer to interleukin-13

(high affinity), but not to IL-4; human cells expressing
IL-13RA2 show specific IL-13 binding with high affinity
Continued
Clusters
Cell Specificity
Function
CDw217
Activated memory T cells
Interleukin 17 receptor homodimer
CD220
Nonlineage molecules
Insulin receptor; integral transmembrane glycoprotein

comprised of two a and two b subunits; this receptor
binds insulin and has a tyrosine- protein kinase activity
autophosphorylation activates the kinase activity
CD221
Insulin-like growth factor I receptor binds insulin-like growth

factor with a high affinity. It has tyrosine kinase activity
and plays a critical role in transformation events.
Cleavage of the precursor generates a and b subunits
CD222
Transmembrane protein. Its main functions include

internalization of IGF-II, internalization or sorting of
lysosomal enzymes, and other M6P-containing proteins
CD223
Activated T and NK cells
Involved in lymphocyte activation; binds to HLA class-II

antigens; role in downregulating antigen-specific response
CD224
Predominantly a membrane-bound enzyme; plays a key role

in the g-glutamyl cycle, a pathway for the synthesis and
degradation of glutathione. This enzyme consists of two
polypeptide chains, which are synthesized in precursor
form from a single polypeptide
CD225
Leukocytes and endothelial cells
Interferon-induced transmembrane protein 1 is implicated

in the control of cell growth.
CD226
NK cells, platelets, monocytes, and a subset of T cells
Adhesion glycoprotein; mediates cellular adhesion to other

cells bearing an unidentified ligand and cross-linking
CD226 with antibodies causes cellular activation
CD227
Human epithelial tumors, such as breast cancer
Epithelial mucin containing a variable number of repeats

with a length of twenty amino acids, resulting in many
different alleles. Direct or indirect interaction with actin
cytoskeleton
CD228
Predominantly in human melanomas
Tumor-associated antigen (melanoma) identified by

monoclonal antibodies 133.2 and 96.5; involved in cellular
iron uptake.
CD229
Lymphocytes
May participate in adhesion reactions between T

lymphocytes and accessory cells by homophilic interaction
CD230
Expressed both in normal and infected cells
Unknown
CD231
T-cell acute lymphoblastic leukemia, neuroblastoma

cells, and normal brain neuron
Unknown
CD232
Receptor for an immunologically active semaphorin (virus

encoded semaphorin protein receptor)
CD233
Erythroid cells
Band 3 is the major integral glycoprotein of the erythrocyte

membrane. It has two functional domains. Its integral
domain mediates a 1:1 exchange of inorganic anions
across the membrane, whereas its cytoplasmic domain
provides binding sites for cytoskeletal proteins, glycolytic
enzymes, and hemoglobin. Multifunctional transport protein
CD234
Erythroid cells and nonerythroid cells
Fy-glycoprotein; Duffy blood group antigen; nonspecific

receptor for many chemokines such as IL-8, GRO,
RANTES, MCP-1, and TARC. It is also the receptor for the
human malaria parasites Plasmodium vivax and
Plasmodium knowlesi
CD235a
Erythroid cells
Major carbohydrate-rich sialoglycoprotein of human

erythrocyte membrane which bears the antigenic
determinants for the MN and Ss blood groups. Also binds
influenza virus
CD235b
Erythroid cells
This protein is a minor sialoglycoprotein in human

erythrocyte membranes. Along with GYPA, GYPB is
responsible for the MNS blood group system.
CD236
Erythroid cells
Glycophorin C (GPC) and glycophorin D (GPD) are closely

related sialoglycoproteins in the human red blood cell
membrane. GPD is a ubiquitous shortened isoform of
GPC, produced by alternative splicing of the same gene.
The Webb and Duch antigens, also known as glycophorin D,
result from single point mutations of the glycophorin C gene
Continued
CHAPTER 6
55
IMMUNOLOGY
SECTION 2

Clusters
Cell Specificity
Function
CD236R
Erythroid cells
Glycophorin C (GPC) is associated with the Gerbich (Ge)

blood group deficiency. It plays an important role in
regulating the mechanical stability of red cells and is a
putative receptor for the merozoites of Plasmodium
falciparum
CD238
Erythroid cells
KELL blood group antigen; homology to a family of zinc

metalloglycoproteins with neutral endopeptidase activity,
type II transmembrane glycoprotein
CD239
Erythroid cells
A type I membrane protein.The human F8/G253 antigen,

B-CAM, is a cell-surface glycoprotein that is expressed
with restricted distribution pattern in normal fetal and
adult tissues, and is upregulated following malignant
transformation in some cell types.
CD240CE
Erythroid cells
Rhesus blood group, CcEe antigens.
CD240D
Erythroid cells
Rhesus blood group, D antigen. May be part of an

oligomeric complex which is likely to have a transport or
channel function in the erythrocyte membrane.
CD241
Erythroid cells
Rhesus blood group-associated glycoprotein RH50,

component of the RH antigen multisubunit complex;
required for transport and assembly of the Rh membrane
complex to the red blood cell surface. Defects in RhAg are
a cause of a form of chronic hemolytic anemia associated
with stomatocytosis, and spherocytosis, reduced osmotic
fragility, and increased cation permeability
CD242
Erythroid cells
Intercellular adhesion molecule 4, LandsteinerWiener

blood group. LW molecules may contribute to the
vasoocclusive events associated with episodes of acute
pain in sickle cell disease
CD243
Stem/progenitor cells
Multidrug resistance protein 1 (P-glycoprotein). P-gp has

been shown to utilizese ATP to pump hydrophobic drugs
out of cells, thus increasing their intracellular concentration
and hence their toxicity.
CD244
NK cells
2B4 is a cell-surface glycoprotein related to CD2 and

implicated in the regulation of natural killer and T
lymphocyte function.
CD245
T cells
Cyclin E/Cdk2 interacting protein p220. NPAT is involved in

a key S phase event and links cyclical cyclin E/Cdk2 kinase
activity to replication-dependent histone gene transcription
CD246
Expressed in the small intestine, testis, and brain but

not in normal lymphoid cells
Anaplastic (CD30+ large cell) lymphoma kinase; plays an

important role in brain development; involved in anaplastic
nodal non-Hodgkins lymphoma or Hodgkins disease
with translocation t(2;5)(p23;q35) or inv2(23;q35).
CD247
T cells, NK cells
T-cell receptor z; has a probable role in assembly and

expression of the TCR complex as well as signal
transduction upon antigen triggering. TCR z together with
TCRa:b and g:b heterodimers and CD3-g, -d, and -e, forms
the TCR-CD3 complex. The z chain plays an important
role in coupling antigen recognition to several intracellular
signal-transduction pathways. Low expression of
the antigen results in impaired immune response
(Adapted in part from Janeway CA, Travers P, Walport M, Shlomchik M: Immunobiology 6: the immune system in health and disease, 6th Edition, New York, Garland
Science 2004.)
ELAM, endothelial leukocyte adhesion molecule; LAM, leukocyte adhesion molecule; MAC, macrophage; HIV, human immunodeficiency virus; ICAM, intercellular
adhesion molecule; IL, interleukin; LPS, lipopolysaccharide; NCAM, neutrophil cellular adhesion molecule; NK, natural killer; MHC, major histocompatibility complex; LFA,
a2b2-integrins; VCAM, vascular cellular adhesion molecule; VLA, a2b1-integrins.
56
mitogen, as well as for the puried protein derivative of Mycobacterium tuberculosis and lipopolysaccharide.
Null cells are lymphocytes that possess none of the aforementioned cell-surface antigens characteristic of T cells or B
cells. This cell population is heterogeneous, and some authorities include natural killer (NK) cells among the null cell
population even though the origin of NK cells appears to be
in monocyte/macrophage precursor lines rather than the
lymphocyte lineage. Nonetheless, the morphologic characteristics and behaviors of NK cells, along with the ambiguity of
their origin, allow one license to include them under the null
cell rubric. NK cells are nonadherent (unlike macrophages, they
do not stick to the surface of plastic tissue culture dishes)
mononuclear cells present in peripheral blood, spleen, and
lymph node. The most notable function of these cells is killing
of transformed (malignant) cells and virus-infected cells.
Because they do this without prior sensitization, they are an

important component of the early natural response in the
immune system. The cytotoxicity of NK cells is not major
histocompatibility complex (MHC)-restricted, a dramatic
contrast with cytotoxic T cells. (More about the MHC and the
products of those gene loci later.) The large granules present in
NK cells (the cells are sometimes called large granular lymphocytes) contain perforin and perhaps other cell membranelysing enzymes, and it is the enzymes in these granules that are
responsible for the lethal-hit cytolysis for which NK cells are
famous.
Killer cells or LAK cells (lymphocyte-activated killer cells) are
the other notable null cell subpopulation. These cells do have
receptors for the Fc portion of immunoglobulin G (IgG) and
thus can attach themselves to the Fc portion of IgG molecules.
Through this receptor, they are a primary cell responsible for
the cytolysis in the so-called antibody-dependent, cell-mediated
cytotoxicity reaction. These cells probably participate in type II
Gell and Coombs hypersensitivity reactions and are involved in
immune removal of cellular antigens when the target cell is too
large to be phagocytosed.
It is clear that both B cells and T cells can be further divided
into specialized subsets. B cells, for example, are subdivided
into the B cells that synthesize the ve separate classes of immunoglobulin (IgG, IgA, IgM, IgD, and IgE). All B cells initially
produce IgM specic for an antigenic determinant (epitope) to
which it has responded, but some subsequently switch from
synthesis of IgM to synthesis of other immunoglobulin classes.
The details of the control of antibody synthesis and classswitching are covered in Chapter 8. Less known is the fact that
functionally distinct subsets of B cells exist, in addition to the
different B cells in terms of antibody class synthesis. The eld
of B-cell diversity analysis is embryonic, but it is clear that the
exploitation of monoclonal antibody technology will distinguish, with increasingly ne specicity, differences in B-cell
subpopulations. It is clear, for example, that a subpopulation of
B lymphocytes possesses the CD5 glycoprotein on the cell
surface plasma membrane (a CD glycoprotein not ordinarily
present on B lymphocytes but rather on the cell surface of T
cells).3 These cells appear to be associated with autoantibody
production.4
It is also clear now that B cells are functionally important as
antigen-presenting cells (APCs), a fact that startles most
physicians who studied immunology before 1991. T-cell
receptors (TCRs) cannot react with native antigen; rather they
respond to processed antigenic determinants of that antigen.
APCs phagocytose the antigen, process it, and display
denatured, limited peptide sequences of the native antigen on
the cell surface of the APC in association with cell surface class
II MHC glycoproteins. B cells, as well as classic APCs, such as
macrophages and Langerhans cells, can perform this function.
The antigen is endocytosed by the B cell and processed in the Bcell endosome (possibly through involvement of cathepsin D) to
generate short, denatured peptide fragments, which are then
transported to the B-cell surface bound to class II glycoprotein
peptides, where the antigenic peptides are presented to CD4
helper T lymphocytes, along with the delivery of a costimulatory signal via its B71 and 2 molecules (CD80 and
CD 86) interaction with T-cell stimulatory molecules, CD 28
and CTLA 4.
Finally, regarding B-cell heterogeneity, it is becoming
apparent that some B lymphocytes also have suppressor or
regulatory activity. The emerging data on B-cell functional and
cell surface heterogeneity will be exciting to follow in the
coming years.
Much more widely recognized, of course, is that subsets of T
lymphocytes exist. Helper (CD4) T cells help in the induction
of an immune response, in the generation of an antibody

response, and in the generation of other, more specialized
components of the immune response. Cytotoxic (CD8) T cells,
as the name implies, are involved in cell killing or cytotoxic
reactions. Delayed-type hypersensitivity (CD4) T cells are
the classic participants in the chronic inflammatory responses
characteristic of certain antigens such as mycobacteria. Regulatory T cells, Treg, are responsible for modulating immune
responses, preventing uncontrolled, host-damaging inflammatory responses. There are at least 2 subsets of Treg, cells:
CD4+ CD25+ and CD8+ CD25+ cells. It is even likely that
there are sub-subsets of these T cells. Excellent evidence exists,
for example, that there are at least three subsets of regulatory T
cells and at least two subsets of helper T cells.
Mosmann and Coffman5 described two types of helper (CD4)
T cells with differential cytokine production proles. TH1 cells
secrete interleukin-2 (IL-2) and interferon-g (IFN-g) but do not
secrete IL-4 or IL-5, whereas TH2 cells secrete IL-4, IL-5, IL-10,
and IL-13, but not IL-2 or IFN-g. Furthermore, TH1 cells can be
cytolytic and can assist B cells with IgG, IgM, and IgA synthesis
but not IgE synthesis. TH2 cells are not cytolytic but can help B
cells with IgE synthesis as well as with IgG, IgM, and IgA
production.6 It is becoming clear that CD4 TH1 or CD4 TH2
cells are selected in infection and autoimmune diseases. Thus,
TH1 cells accumulate in the thyroid of patients with autoimmune thyroiditis,7 whereas TH2 cells accumulate in the
conjunctiva of patients with vernal conjunctivitis.8 The T cells
that respond to M. tuberculosis protein are primarily TH1 cells,
whereas those that respond to Toxocara canis antigens are TH2
cells. Romagnani has proposed that TH1 cells are preferentially
selected as participants in inflammatory reactions associated
with delayed-type hypersensitivity reactions and low antibody
production (as in contact dermatitis or tuberculosis), and TH2
cells are preferentially selected in inflammatory reactions
associated with persistent antibody production, including
allergic responses in which IgE production is prominent.9
Further, it is now clear that these two major CD4 T-lymphocyte
subsets regulate each other through their cytokines. Thus, TH2
CD4 lymphocyte cytokines (notably IL-10) inhibit TH1 CD4
lymphocyte proliferation and cytokine secretion, and TH1 CD4
lymphocyte cytokines (notably IFN-g) inhibit TH2 CD4 lymphocyte proliferation and cytokine production.
CHAPTER 6
MACROPHAGES
The macrophage ( large eater) is the preeminent professional
APC. These cells are 1215 mm in diameter, the largest of the
lymphoid cells. They possess a high density of class II MHC
glycoproteins on their cell surface, along with receptors for
complement components, the Fc portion of Ig molecules, receptors for bronectin, interferons-a, -b, and -g, IL-1, tumor
necrosis factor, and macrophage colony-stimulating factor.
These cells are widely distributed throughout various tissues
(when found in tissue, they are called histiocytes), and the
microenvironment of the tissue profoundly influences the
extent of expression of the various cell surface glycoproteins as
well as the intracellular metabolic characteristics. It is clear that
further compartmentalization of macrophage subtypes occurs
in the spleen. Macrophages that express a high density of class
II MHC glycoproteins are present in red pulp, and macrophages
with signicantly less surface expression are in the marginal
zone, where intimate contact with B cells exists. It is likely that
just as in the murine system,10 in humans one subclass
of macrophage preferentially presents antigen to one particular
subset of helper T cell responsible for induction of regulatory
T-cell activation, whereas a different subset of macrophage
preferentially presents antigen to a different helper T-cell subset
57
SECTION 2
IMMUNOLOGY
responsible for cytotoxic or delayed-type hypersensitivity
effector functions.
Macrophages also participate more generally in inflammatory
reactions. They are members of the natural (early defense)
immune system and are incredibly potent in their capacity to
synthesize and secrete a variety of powerful biologic molecules,
including proteases, collagenase, angiotensin-converting
enzyme, lysozyme, IFN-a, IFN-b, IL-6, tumor necrosis factor-a,
bronectin, transforming growth factor-b, platelet-derived
growth factor, macrophage colony-stimulating factor,
granulocyte-stimulating factor, granulocyte-macrophage colonystimulating factor, platelet-activating factor, arachidonic acid
derivatives (prostaglandins and leukotrienes), and oxygen
metabolites (oxygen free radicals, peroxide anion, and hydrogen
peroxide). These cells are extremely important, even pivotal,
participants in inflammatory reactions and are especially
important in chronic inflammation. The epithelioid cell typical
of so-called granulomatous inflammatory reactions evolves
from the tissue histiocyte, and multinucleated giant cells form
through fusion of many epithelioid cells.
Specialized macrophages exist in certain tissues and organs,
including the Kupffer cell of the liver, dendritic histiocytes in
lymphoid organs, interdigitating reticulum cells in lymphoid
organs, and Langerhans cells in skin, lymph nodes, conjunctiva, and cornea.
Langerhans cells are particularly important to the ophthalmologist. They probably are the premier APC for the external
eye. Derived from bone marrow macrophage precursors, like
macrophages, their function is basically identical to that of the
macrophage in antigen presentation. They are rich in cellsurface class II MHC glycoproteins and have cell-surface
receptors for the third component of complement and the Fc
portion of IgG. Langerhans cells are abundant in the mucosal
epithelium of the mouth, esophagus, vagina, and conjunctiva.
They are also abundant at the corneoscleral limbus, less so in
the peripheral cornea; they are normally absent from the central
third of the cornea.11 If the center of the cornea is provoked
through trauma or infection, the peripheral cornea Langerhans
cells quickly stream into the center of the cornea.12 These
CD1-positive dendritic cells possess a characteristic racketshaped granule on ultrastructural analysis, the Birbeck granule.
Birbeck granules are subdomains of the endosomal recycling
compartment that are rich in Langerin (CD 207), a protein
specic to Langerhans cells, and a type II membrane-associated
C-type lectin which recognizes mannose residues and may
serve with CD 1 to present lipid antigens by Langerhans cells
after endocytosis and processing.13,14
POLYMORPHONUCLEAR LEUKOCYTES
Polymorphonuclear leukocytes (PMNs) are part of the natural
immune system. They are central to host defense through
phagocytosis, but if they accumulate in excessive numbers,
persist, and are activated in an uncontrolled manner, the result
may be deleterious to host tissues. As the name suggests, they
contain a multilobed nucleus and many granules. PMNs are
subcategorized as neutrophils, basophils, or eosinophils, depending on the differential staining of their granules.
NEUTROPHILS
58
Neutrophils account for more than 90% of the circulating

granulocytes. They possess surface receptors for the Fc portion
of IgG (CD16) and for complement components, including C5a
(important in chemotaxis), CR1 (CD35), and CR3 (CD11b)
(important in adhesion and phagocytosis). When appropriately
stimulated by chemotactic agents (complement components,
TABLE 6.2. Neutrophil Granules and Their Contents

Azurophil Granules
Specific Granules
Other Granules
Myeloperoxidase
Alkaline
phosphatase
Acid
phosphatase
Acid phosphatase
Histaminase
Heparinase
5-Nucleotidase
Collagenase
b-Glucosaminidase
Lysozyme
Lysozyme
a-Mannosidase
Elastase
Vitamin B12-binding
proteins
Acid proteinase
Cathepsins B, D, G
Plasminogen
activator
Lactoferrin
Elastase gelatinase
Proteinase 3
Glycosaminoglycans
b-Glycerophosphatase
b-Glucuronidase
N-acetylb-glucosaminidase
Cytochrome
a-Mannosidase
Arylsulfatase
a-Fucosidase
Esterase
Histonase
Cationic proteins
Defensins
Bactericidal
permeabilityincreasing
protein (BPI)
Glycosaminoglycans
brinolytic and kinin system components, and products from

other leukocytes, platelets, and certain bacteria), neutrophils
move from blood to tissues through margination (adhesion
to receptors or adhesion molecules on vascular endothelial cells)
and diapedesis (movement through the capillary wall).
Neutrophils release the contents of their primary (azurophilic)
granules (lysosomes) and secondary (specic) granules (Table
6.2) into an endocytic vacuole, resulting in:1 phagocytosis of a
microorganism or tissue injury;2 type II antibody-dependent,
cell-mediated cytotoxicity; or3 type III hypersensitivity reactions
(immune complex-mediated disease). Secondary granules
release collagenase, which mediates collagen degradation. Aside
from the products secreted by the granules, neutrophils produce
arachidonic acid metabolites (prostaglandins and leukotrienes)
as well as oxygen free radical derivatives.
EOSINOPHILS
Eosinophils constitute 35% of the circulating PMNs. They
possess surface receptors for the Fc portion of IgE (low afnity)
and IgG (CD16) and for complement components, including
C5a, CR1 (CD35), and CR3 (CD11b). Eosinophils play a
special role in allergic conditions and parasitoses. They also participate in type III hypersensitivity reactions or immune
complex-mediated disease following attraction to the inflammatory area by products from mast cells (eosinophil chemotactic
factor of anaphylaxis), complement, and other cytokines from
TABLE 6.3. Granular Contents of Eosinophils
TABLE 6.4. Mast Cell Types and Characteristics
Lysosomal hydrolases
Characteristic
Mucosal Mast Cell

Mast (MC-T, MMC)
Connective Tissue
Mast Cell (MC-TC,
CTMC)
Size
Small, pleomorphic
Large, uniform
Nucleus
Unilobed or bi-lobed
Unilobed
Granules
Few
Many
Location
Histochemistry
Gut
Peritoneum, skin
Collagenase
Cathepsin
Protease
Tryptase
Tryptase and
chymase
Proteoglycans
Chondroitin sulfate
Heparin
Histamine
<1 pg/cell
15 pg/cell
IgE
Surface and
cytoplasmic
Surface
Eosinophil peroxidases
Formalin sensitive
Yes
No
Phospholipases
In Vitro Effect of:
Lysophospholipases
Compound 48/80
Proliferation
Degranulation
Polymyxin
Proliferation
Degranulation
Life Span
40 days
>40 days
Proliferation
Thymus-dependent
Thymus-independent
Antigen
Yes
Yes
Anti-IgE
Yes
Yes
Compound 48/80
No
Yes
Arylsulfatase
Acid phosphatase
b-Glycerophosphatase
Ribonuclease
Proteinases
Morphology
Histaminase
Peroxisomes
Major basic proteins
Eosinophil cationic protein
other inflammatory cells. Eosinophils release the contents of

their granules to the outside of the cell after fusion of the
intracellular granules with the plasma membrane (degranulation).
Table 6.3 shows the known secretory products of eosinophils;
the role these products of inflammation play, even in nonallergic
diseases (such as Wegeners granulomatosis), is underappreciated.
Secretagogues
BASOPHILS
Bee venom
No
Yes
Basophils account for less than 0.2% of the circulating

granulocytes. They possess surface receptors for the Fc portion
of IgE (high afnity) and IgG (CD16) and for complement
components, including C5a, CR1 (CD35), and CR3 (CD11b).
Their role, other than perhaps as tissue mast cells, is unclear.
Con A
Yes
Yes
Alcian blue
Yes
Yes
Safranin
No
Yes
Berberine sulfate
No
Yes
Compounds
No
Yes
Cromoglycate
No
Yes
Theophylline
Yes
Yes
Enhancement of
Secretion
No
Yes
Phosphatidyl serine
Yes
Yes
Prostaglandin D2
Leukotrienes B4,
C4, D4
MAST CELLS
The mast cell is indistinguishable from the basophil in many
respects, particularly its contents. There are at least two classes
of mast cells based on their neutral protease composition,
T-lymphocyte dependence, ultrastructural characteristics, and
predominant arachidonic acid metabolites (Table 6.4). Mucosaassociated mast cells (MMC or MC-T) contain primarily
tryptase as the major protease (hence, some authors designate
these MC-T, or mast cells-tryptase) and prostaglandin D2 as
the primary product of arachidonic acid metabolism. MMCs are
T-cell-dependent for growth and development (specically IL-3dependent), and are located predominantly in mucosal stroma
(e.g., gut). MMCs are small and short-lived (< 40 days). They
contain chondroitin sulfate but not heparin, and their
histamine content is modest (Table 6.5). MMCs degranulate in
response to antigen-IgE triggering but not to exposure to
compound 48/80, and are not stabilized by disodium cromoglycate. They are formalin-sensitive, so formalin-xation of
tissue eliminates or greatly reduces our ability to nd these cells
by staining technique. With special xation techniques, MMC
granules stain with alcian blue but not with safranin.
Connective tissue mast cells (CTMCs) contain both tryptase
and chymase (so some authors designate them MC-TC), as well
as leukotrienes B4, C4, and D4, as the primary products of
CHAPTER 6
b-Glucuronidase
Staining
Antiallergic
Doxantrile
Adenosine
Predominant
Arachidonic Acid
Metabolite
Ultrastructural
Lattice
Features of Granules
Scroll
arachidonic acid metabolism. CTMCs are T-cell-independent.

They are larger than MMCs and are located principally in skin
and at mucosal interfaces with the environment. They contain
heparin and large amounts of histamine, and degranulate in
response to compound 48/80 in addition to antigen-IgE
59
SECTION 2
IMMUNOLOGY
interactions. CTMCs are stabilized by disodium cromoglycate.
They stain with alkaline Giemsa, toluidine blue, alcian blue,
safranin, and berberine sulfate.
The ultrastructural characteristics of MMCs and CTMCs are
also different. Electron microscopy shows that the granules of
MMCs contain lattice-like structures; the granules of CTMCs
contain scroll-like structures. Mast cells play a special role in
allergic reactions they are the preeminent cell in the allergy
drama. They also can participate in type II, III, and IV hypersensitivity reactions, however. Their role in these reactions,
aside from notable vascular effects, is not well understood. NonIgE-mediated mechanisms (e.g., C5a) can trigger mast cells to
release histamine, platelet-activating factor, and other biologic
molecules when antigen binds to two adjacent IgE molecules
on the mast cell surface. Histamine and other vasoactive
amines cause increased vascular permeability, allowing
immune complexes to become trapped in the vessel wall.
PLATELETS
Blood platelets, cells well adapted for blood clotting, also are
involved in the immune response to injury, a reflection of their
evolutionary heritage as myeloid (inflammatory) cells. They
possess surface receptors for the Fc portion of IgG (CD16) and
IgE (low afnity), for class I histocompatibility glycoproteins
(human leukocyte antigen-A, -B, or -C), and for factor VIII.
They also carry molecules such as GpII b/ III a (CDw41), which
bind brinogen, and Gp1b (CDw42), which binds von
Willebrands factor.
After endothelial injury, platelets adhere to and aggregate at
the endothelial surface, releasing permeability-increasing
molecules from their granules (Table 6.6). Endothelial injury
may be caused by type III hypersensitivity. Platelet-activating
factor released by mast cells after antigen-IgE antibody complex
formation induces platelets to aggregate and release their
vasoactive amines. These amines separate endothelial cell tight
junctions and allow the immune complexes to enter the vessel
wall. Once the immune complexes are deposited, they initiate
an inflammatory reaction through activation of complement
components and neutrophil lysosomal enzyme release.
TABLE 6.5. Mast Cell Contents

Histamine
Serotonin
Rat mast-cell protease I and II
Heparin
Chondroitin sulfate
b-Hexosaminidase
b-Glucuronidase
b-4DGalactosidase
Arylsulfatase
Eosinophil chemotactic factor for anaphylaxis (ECF-A)
Slow reactive substance of anaphylaxis (SRS-A)
High molecular weight neutrophil chemotactic factor
Arachidonic acid derivatives
Platelet-activating factor
TABLE 6.6. Platelet Granules and Their Contents

a-Granules
Fibronectin
Fibrinogen
Plasminogen
Thrombospondin
von Willebrand factor
a2-Plasmin inhibitor
Platelet-derived growth factor (PDGF)
Platelet factor 4 (PF4)
Transforming growth factor (TGF) a and b
Thrombospondin
ONTOGENY OF THE IMMUNE SYSTEM
60
Cells of the hematologic system are derived from primordial

stem cell precursors of the bone marrow. Embryonically, these
cells originate in the blood islands of the yolk sac.13 These cells
populate embryonic liver and bone marrow.14 All the blood
elements are derived from these primordial stem cells:
erythrocytes, platelets, PMNs, monocytes, and lymphocytes.
These primordial stem cells are pluripotential, and the exact
details of the influences that are responsible for a particular
pluripotential primordial stem cells evolving along one
differentiation pathway (e.g., into a monocyte) as opposed to
some other differentiation pathway (e.g., into a lymphocyte) are
incompletely understood. It appears, however, that special
characteristics of the microenvironment in the bone marrow,
particularly with respect to the association with other resident
cells in the bone marrow, contribute to or are responsible for the
different pathways of maturation and differentiation. For
example, specic cells in the bone marrow in the endosteal
region promote the differentiation of hematopoietic stem cells
into B lymphocytes.1521 In birds, primordial pluripotential stem
cells that migrate to a gland near the cloaca of the chicken
known as the bursa of Fabricius (for reasons of probable stimuli
in the bone marrow as yet not understood) are influenced by the
epithelial cells in that gland to terminally differentiate into
B lymphocytes.22,23 Interestingly, various candidates for the
b-Lysin
Permeability factor
Factors D and H
Decay-accelerating factor
Dense granules
Serotonin
Adenosine diphosphate (ADP)
Others
Arachidonic acid derivatives
so-called bursal equivalent that is responsible for B-cell

differentiation in humans were proposed for many years before
the role of the bone marrow itself for this function became
evident. Extra-bone marrow tissues that had been proposed as
bursal equivalent candidates included the appendix, tonsils,
liver, and Peyers patch.
T-cell development results from pluripotential hematopoietic
stem cell migration (stimulus unknown) from the bone marrow
to the thymus. Thymic hormones (at least 20 have been
preliminarily described) produced by the thymic epithelium
initiate the complex series of events that result not only in
Hormone
No. of Amino Acids
Thymosin
28
Thymopoietin
49
Thymic humoral factor
31
Facteur thymique serique
differentiation of the hematopoietic stem cells into T

lymphocytes but also in subdifferentiation of T lymphocytes
into their various functional subsets; helper function, killer
function, and suppressor function are acquired while the T cells
are still in the thymus. Table 6.7 lists the four thymic hormones
most rigorously studied to date. Note that all are involved in Tcell differentiation and in the development of helper T-cell
function and that three of the four can be involved or are
involved in the acquisition of suppressor T-cell activity. Clearly,
the story is considerably more complex than the part we
currently understand, and additional factors are undoubtedly
responsible for the nal differentiation of T lymphocytes into
their functionally distinct subsets.
These various hormones are also undoubtedly responsible for
the induction of cell surface glycoprotein expression on the
surface of T cells. The cell-surface expression of the various
glycoproteins changes during T-cell maturation in the thymus.
For example, the CD2 glycoprotein is the rst that can be
identied on the differentiating T cell, but this is eventually
joined by CD5; these are both eventually replaced (CD2
completely and CD5 partially) by CD1 glycoprotein, which in
turn is lost and replaced by the mature CD3 marker. CD4 and
CD8 glycoproteins are acquired prior to emigration from the
thymus of helper and cytotoxic-regulatory T cells, respectively.
Monocytes, NK cells, and killer cells evolve from pluripotential hematopoietic stem cells through influences that are
incompletely understood. All three types of cells do arise from a
common monocyte precursor and later subdifferentiate under
unknown influences.
The primary or central lymphoid organs are the bone marrow,

thymus, and liver. The peripheral lymphoid organs include
lymph nodes, spleen, gut-associated lymphoid tissue, bronchusassociated lymphoid tissue, and conjunctiva-associated
lymphoid tissue. The anatomic characteristics of the thymus,
lymph node, and spleen are described briefly.
The thymus consists of a medulla, containing thymic epithelial tissue and lymphocytes, and a surrounding cortex densely
packed with small, proliferating T lymphocytes (Fig. 6.1).
The cells in the cortex emigrate from the thymus: The cell
population turns over completely every 3 days. Only ~1% of the
cells produced in the thymus, however, actually emigrate from
it; 99% are destroyed locally, probably in a process designed to
prevent autoreactive T lymphocytes from gaining access to the
extrathymic regions of the organism. Thymic nurse cells, epithelial cells in the cortical region, may be responsible in part for
some of the later events in T-lymphocyte differentiation (e.g.,
into helper and regulatory T cells).
Lymph nodes (Fig. 6.2) are also composed of medulla and
cortex. The medulla, rich in the arterial and venous components of the lymph node, contains reticular cells that drain into
the efferent lymphatic vessels. The cortex contains the primary
lymphoid follicles, containing mature, resting B cells, secondary
lymphoid follicles with their germinal centers (full of antigenstimulated B cells and dendritic cells) and mantle, and lymphocytes. The paracortical region close to the medulla is rich in
T cells, particularly CD4+ T cells.
The arrangement of the spleen is similar to that of the
thymus and lymph node, though lymph node-type follicles are
not so clearly distinguished (Fig. 6.3). The lymphoid follicles
and surrounding lymphocytes are called the white pulp of the
spleen. The red pulp of the spleen is composed of the sinusoidal
channels that typically contain a relatively large number of red
blood cells. Popiernik has described the white pulp as being organized as a lumpy cylindrical sheath surrounding central
arterioles. The arterioles curve back on the white pulp to
develop it as the marginal sinus, which separates the white pulp
from the red.24 B cells predominate in the marginal zone, but
CHAPTER 6
PRIMARY (CENTRAL) LYMPHOID ORGANS
TABLE 6.7. Thymic Hormones
FIGURE 6.1. (a) and (b) Human thymus. Note

the organization into individual lobules
separated by connective tissue trabeculae, with
dense collections of tightly packed, deeply
stained immature thymocytes in the cortex
and more mature lymphocytes in the medulla.
(c) Hassalls corpuscles, probably composed
of degenerated epithelial cells, are found
scattered throughout the medulla.
61
SECTION 2
IMMUNOLOGY
FIGURE 6.2. (a) Human lymph node. Note the

organization, in some respects similar to that of
the thymus, into two predominant areas the
cortex and the medulla. The cortex is rich in B
cells; the medulla contains cords of lymphoid
tissue that contain both B and T cells; and an
intermediate zone called the paracortex is rich
in T cells. The paracortex, in addition to being
rich in T cells, contains APCs. (b) The medulla
contains macrophages and plasma cells as well
as B and T cells. The cortex contains the
primary and secondary follicles, the distinction
between the two being the germinal center (site
of actively proliferating B cells) in the secondary
follicles.
FIGURE 6.3. (a) Human spleen. Note the red

pulp, primarily involved in destruction of old red
blood cells and red blood cells containing
immune complexes, and white pulp, organized
primarily around central arterioles and hence
forming a follicle or a periarteriolar lymphoid
sheath (PALS). (b) T cells are particularly rich
around the central arteriole of the PALS And B
cells in the periphery of the PALS. The far
periphery of the PALS, adjoining the red pulp,
contains macrophages as well as B cells.
a
CD4+ T cells are present as well. T cells are clustered tightly

around the central arteriole, where ~70% of the T cells are
CD4+. B cells also predominate in the lumpy eccentric follicle
of white pulp. Table 6.8 outlines some of the characteristics of
these three lymphoid organs and their organization. The spleen
is the primary site of immune responses to intravenous and
anterior chamber-introduced antigens.
LYMPHOID TRAFFIC
Lymphatic vessels and blood vessels connect these lymphatic
organs to each other and the other organs of the body. Lymphatic vessels drain every organ except the nonconjunctival
parts of the eye, internal ear, bone marrow, spleen, cartilage, and
some parts of the central nervous system. The interstitial fluid
and cells entering this system are propelled (predominantly by
skeletal muscle contraction) to regional lymph nodes. Efferent
lymphatics draining these regional nodes converge to form large
lymph vessels that culminate in the thoracic duct and in the
right lymphatic duct. The thoracic duct empties into the left
subclavian vein, carrying approximately three-quarters of the
lymph, whereas the right lymphatic duct empties into the right
subclavian vein.
The subject of lymphocyte trafc, like so many areas of
immunology, has undergone intensive reexamination since the
1980s; since then, discoveries relating to homing receptors,
addressins, and other adhesion molecules have revolutionized
TABLE 6.8. Lymphoid Organs

Primary
Secondary
Thymus
Lymph nodes
Bone marrow
62
Spleen
Mucosa-associated lymphoid tissue
our understanding of how lymphoid cells migrate into and out

of specic areas. For example, it is clear that one or more
homing receptors is present on the surface of all lymphoid cells.
These receptors can be regulated, induced, and suppressed.
Furthermore, induction and suppression of other cell-surface
moieties that may regulate lymphoid cell exit from one location
or another occurs. For example, cortical thymocytes rich in
peanut agglutinin on their surface have a paucity of homing
receptors, a fact that might ordinarily allow them to migrate out
of the thymus to some other location. Butcher and Weissman
have hypothesized that terminal sialidation could release
formerly peanut agglutinin-positive thymocytes from hypothetical peanut agglutinin-like lectins in the thymus, providing
exit visas for their release from the thymus.25 In any event,
one thing is clear: mature T cells emerging from the thymus
cortex toward the medulla are rich in cell surface or plasma
membrane-homing receptors, or adhesion molecules or
adhesomes, which are ligands for various addressins or
adhesion molecules at other, remote loci. In the mouse, homing
receptors on the surface of mature T cells have been identied
for the lymph node (MEL-14 or L-selectin (LFA-1)) and for
Peyer s patch (LPAM-1 a4b7 integrin, CD44). Equivalent
homing receptors exist in humans.26 The Hermes glycoprotein
on the surface of T and B lymphocytes has been shown to be
identical to the CD44 molecule.27 Antibodies to this
glycoprotein prevent binding of lymphocytes to mucosal lymph
node high endothelial venules.28 Other cell-surface homing and
adhesion molecules, along with their homing receptor ligands,
are shown in Table 6.9.
IMMUNE RESPONSE
Professional APCs phagocytose foreign material (antigens),
process it through protease endosomal-lysosomal degradation,
package it with MHC molecules, and transport the peptideMHC complex to the cell surface. B cells and dendritic cells
(including Langerhans cells) perform this function too, but
TABLE 6.9. Adhesion Molecules

LFA-1a
(CD11a)
MAC-1
(CD11b)
GP150,95
(CD11c)
LFA-1b
(CD18)
Integrin a4
(CD49d)
TCRab
TCRg/d
LFA-2
(CD2)
CD 22
NCAM
(CD56)
ICAM-1
(CD54)
LFA-3
(CD58)
LECAM-1
CD5
HCAM
(CD44)
HPCA-2
(CD34)
CD28
88-1
sites that bind proteins (enzymes), like phosphatidylinositol

phospholipase C-g1 (PI-PLC-g1) with SH2 binding domain. PIPLC-g1 in turn is phosphorylated (and thereby activated), and it
catalyzes hydrolysis of plasma membrane phosphatidylinositol
4,5 bisphosphate into inositol 1,4,5 triphosphate (ID3) and
diacylglycerol. IP3 then provokes the release of calcium from
its endoplasmic reticulum storage sites. The increased
intracellular calcium concentration that results from the release
from storage in turn results in increased binding of calcium to
calmodulin; this then activates the phosphatase, calcineurin.
Calcineurin catalyzes the conversion of phosphorylated nuclear
factor of activated T cells, cytoplasmic component (NFATc), to
free NFATc. This protein (and probably others) then enters the
cell nucleus, where gene transcription of cellular protooncogenes/transcription factor genes, cytokine receptor genes,
and cytokine genes is then activated and regulated by it (or
them). For example, NFATc translocates to the nucleus, where
it combines with AP-1 proteins; this complex then binds to the
NFATc-binding site of the IL-2 promoter. This, coupled with
NFkB binding by proteins possibly induced by the events
stimulated by CD28-CD80 signal transduction, results in IL-2
gene transcription typical of T-cell activation (see Fig. 6.2).
Thus, this activation phase of the acquired immune response is
characterized by lymphocyte proliferation and cytokine
production.
CHAPTER 6
EXPRESSION OF IMMUNITY
PECAM
(CD31)
GMP140
(CD62)
HNK-1
(CD57)
differences in protease types and class II MHC molecules among

these APCs may influence the type of T cell activated by an
antigen. It is this unit of antigenic peptide determinant and selfMHC glycoproteins, along with the aid of adhesion molecules
(ICAM-1([CD54) and LFA-3 (CD58)) and co-stimulatory
molecules (B7 (CD80)), that forms the recognition unit for the
TCRs specic for the antigenic epitope of the foreign material.
The TCR is composed of recognition units for the epitope and
for the autologous MHC glycoprotein. Endogenous antigens,
such as endogenously manufactured viral protein, typically
result in cytoplasm, associate with class I MHC molecules, and
are transported to the surface of the APC, where the class I
MHC-peptide complex preferentially associates with the TCR
of CD8+ cells. Exogenous antigens that are phagocytized
typically associate, as described earlier, in the endosomal,
endoxytic, exocytic pathways with class II MHC molecules, and
this type of complex preferentially associates with CD4+ TCRs.
The ab heterodimer of the TCR is associated with CD3 and
zh proteins and (for CD4 cells) the CD4 molecule, forming the
TCR complex. Antigen presentation can then occur as the TCR
complex interacts with the antigenic determinant/MHC
complex on the macrophage, with simultaneous CD28-CD80
interaction. Macrophage secretion of IL-1 during this cognitive
presentation phase of the acquired immune response to CD4
T cells completes the requirements for successful antigen
presentation to the helper T cell (see Fig. 6.1).
The CD3 and zh proteins are the signal-transducing
components of the TCR complex; transmembrane signaling via
this pathway results in activation of several phosphotyrosine
kinases, including those of the tyk/jak family and other signal
transduction and activation of transcription molecules and
phosphorylation of tyrosine residues in the cytoplasmic tails of
the CD3 and zh proteins, resulting in the creation of multiple
The emigration of hematopoietic cells from the vascular system

typically occurs at the region of postcapillary high endothelial
venule cells. These cells are rich in the constitutive expression
of so-called addressins, which are tissue- or organ-specic
endothelial cell molecules involved in lymphocyte homing.
These adhesion molecules are lymphocyte-binding molecules
for the homing receptors on lymphocytes. Thus, the mucosal
addressin27 specically binds to the Hermes 90-kDa
glycoprotein. In the murine system, a 90-kDa glycoprotein
(designated MECA-79) is a peripheral lymph-node addressin
specically expressed by high endothelial venules.30 In
peripheral lymph nodes.29 MECA-367 and MECA-89 are
additional addressin glycoproteins in the murine system that
are specic for mucosal vascular high endothelial venules. In
addition to the constitutive expression of addressins or adhesion
molecules, expression of additional adhesion molecules is induced
by a panoply of proinflammatory cytokines. It is this directed
trafcking of inflammatory cells via adhesion molecules that
gives the expression of an immune response its focus, its
specically directed, targeted expression.
Lymphocytes, monocytes, and neutrophils preferentially
migrate or home to sites of inflammation because of this upregulation of cytokines and the induction of adhesion molecules
they promote. Thus, L-selectin (CD62L) on the neutrophil cellsurface membrane does not adhere to normal vascular
endothelium, but intercellular adhesion molecule (ICAM) and
endothelial leukocyte adhesion molecule (ELAM) (CD62E)
expression on the vascular endothelial cell surface induced by
IFN-a, IFN-g, IL-1, IL-17, or a combination thereof results in
low-afnity binding of CD62L, with resultant slowing of
neutrophil transit through the vessel, neutrophil rolling on the
endothelial surface, and (with complement split product and IL8-driven chemotaxis of increasing numbers of neutrophils)
neutrophil margination in the vessels of inflamed tissue.31
Neutrophil LFA-1 (CD11a, CD18) activated expression
(stimulated by IL-6 and IL-8) then results in stronger adhesion
of the neutrophil to endothelial cell ICAM molecules, with
resultant neutrophil spreading and diapedesis into the subendothelial spaces and the surrounding tissue.
63
SECTION 2
IMMUNOLOGY
IMMUNOLOGIC MEMORY
TABLE 6.10. Cytokines and Target Cells
The anamnestic capacity of the acquired immune response

system is one of its most extraordinary properties. Indeed, it is
this remarkable property that was the rst to be recognized by
the Chinese ancients and (later) by Jenner. We take it as
axiomatic that our immunization in childhood with killed or
attenuated smallpox and polio virus provoked not only a
primary immune response but also the development of longlived memory cells that immediately produce a rapid, vigorous secondary immune response whenever we might
encounter smallpox or polio virus, thereby resulting in specic
antibody and lymphocyte-mediated killing of the microbe and
defending us from the harm the virus would otherwise have
done. But just what do we know about the cells responsible for
this phenomenon? What special characteristics enable memory
cells to live for prolonged periods in the absence of continued or
repeated antigen exposure?
Neils Jerne rst hypothesized a clonal selection theory to
explain at once the specicity and diversity of the acquired
immune response, and Macfarlene Burnet expanded on Jernes
original hypothesis, clearly predicting the necessary features
that would prove the theory; many subsequent studies have
done so. Clones are derived from the development of antigenspecic clones of lymphocytes arising from single precursors
prior to and independent from exposure to antigen. Approximately 109 such clones have been estimated to exist in an
individual, allowing him or her to respond to all currently
known or future antigens. Antigen contact results in preferential activation of the preexisting clone with the cell-surface
receptors specic for it, with resultant proliferation of the clone
and differentiation into effector and memory cells. The
secondary or anamnestic immune response is greater and more
rapid in onset than is the primary immune response because of
the large number of lymphocytes derived from the original clone
of cells stimulated by the primary contact with antigen, as well
as the long-lived nature of many of the cells (memory cells).
The memory cells can survive for very long periods, even
decades. They express certain cell-surface proteins not expressed by nonmemory cells (CD45RO). In memory cells, the
level of cell-surface expression of peripheral lymph node
homing receptors is low compared with the population of such
receptors on the surface of nonmemory cells; in contrast, the
population of other adhesion molecules on the surface of
memory cells is much greater than that of the surface of
nonmemory cells. These adhesion molecules include CD11a,
CD18 (LFA-1), CD44, and VLA molecules. Because of the
constitutive expression of the cell-surface adhesion molecules,
memory T cells rapidly home to sites of inflammation, looking
for antigen to which they might respond.
Cytokine
Source
Target Cell
IL-1
Mj, TH, FB, NK,

B, Nj, EC
Pluripotent stem cells, or not

TCTH, B, Mj, FB, Nj
IL-2
TH1
TCTH, B, NK
IL-3
BM, TH, MC
TCTH, B, MC, stem cells
IL-4
TH2, MC
TH1, B, Mj, MC, TH2, NK,

FC
IL-5
TH2, MC, Ej
TCTH, B, Ej
IL-6
BM, Mj, MC, EC, Pluripotent stem

B, TH2, FB
cells, or not TCTH, B, FB, Nj
IL-7
FB, BM
Subcapsular and thymocytes,

TCTH, F, FB
IL-8
BM, FB, EC, Mj,

Nj, Ej
TCTH, Mj, Nj
IL-9
TH2

TCTH, MC
IL-10
TH2, B, Mj
TCD2, TC, TH1, MC, Mj
IL-11
BM

TCTH, B
IL-12
Mj, Nj, B
NK, THTH1
IL-13
TH2
TH1, Mj, B
IL-14
IL-15
Mj, FB, BM
T, NK, B
IL-16
T, Ej, MC
T, Ej
IL-17
TH
FB, T
IL-18
Mj
T, NK
TNF-a
Mj, NK,T
TCTH, B, Mj,
FB
TNF-b
TC, TH1, B
EC, Nj
GMCSF
TH, Mj, MC
SUMMARY
64
The evolutionary advantage of the immune system is obvious.

The complexity of the system that has evolved to protect us,
however, is extraordinary, and our understanding of the
immune system is far from complete. The major cell types of
the system are well known, but subtypes and sub-subtypes are
still being identied. The primary products of one of the major
cell types, the B lymphocytes, have been well characterized
(antibody), but additional cellular products or cytokines from
these cells, which in the 1980s were believed to secrete only
immunoglobulins in their mature (plasma cell) state, are being
discovered. Thus, the 18 interleukins and other cytokines listed
in Table 6.10 will be an incomplete list of the known cytokines
of the immune system by the time this edition is published.
The seemingly never-ending story of immunologic discovery is
Null cells, FB
TCTH, Ej, Nj
GCSF
BM, Mj, FB
MCSF
BM, Mj, FB
LIF
BM, fibroblasts
Myeloid progenitor
SCF
BM
Myeloid progenitor
Cortical thymocytes
IFN-g
NK, TH1
NK, TC, TH2, B, FB, MC
IFN-a
Mf
TCTH, B
IFN-b
FB
TCTH
TGF-b
Mf,T,
chondrocytes
TCTH, B, Mf, FB
TCTH, FB, Nj
B, B cell; BM, bone marrow; CSF, colony-stimulating factor; Ej, eosinophil; EC,
endothelial cell; FB, fibroblast; GM, granulocyte, macrophage; IFN, interferon;
IL, interleukin; LIF, leukocyte inhibitory factor; Mj, macrophage; MC, mast cell;
Nj, neutrophil; NK, natural killer cell; SCF, stem cell factor; TC, cytotoxic T cell;
TGF, transforming growth factor; TH, helper T cell; TNF, tumor necrosis factor.
at once as fascinating as any Shakespeare play and as frustrating

as attempting to understand the universe and the meaning of
life. Each year, a chapter brings new knowledge and new questions, and the wise physician will realize that schooling never
ends in immunology as in so many other biologic sciences. Stay
tuned.
1. Kohler J, Milstein C: Continuous cultures of

fused cells secreting antibody of
predened specicity. Nature 1975;
256:495.
2. Reinherz EL, Schlossman SF: The
differentiation and function of human T
lymphocytes. Cell 1980; 19:821.
3. Hardy RR, Hayakawa K, Parks DR,
Herzenberg LA: Murine B cell differentiation
lineages. J Exp Med 1984; 1959:1169.
4. Hardy RR, Hayakawa K, Schimizu M, et al:
Rheumatoid factor secretion from human
Leu-1 B cells. Science 1987; 236:81.
5. Mosmann TR, Coffman R: Two types of
mouse helper T cell clones: implications
from immune regulation. Immunol Today
1987; 8:233.
6. Coffman R, OHara J, Bond MW, et al: B
cell stimulatory factor-1 enhances the IgE
response of lipopolysaccharide-activated B
cell. J Immunol 1986; 136:4538.
7. Mariotti S, del Prete GF, Mastromauro C, et
al: The autoimmune inltrates of Basedows
disease: analysis of clonal level and
comparison with Hashimotos thyroiditis.
Exp Clin Endocrinol 1991; 97:139.
8. Maggi E, Biswas P, del Prete GF, et al:
Accumulation of TH2-like helper T cells in
the conjunctiva of patients with vernal
conjunctivitis. J Immunol 1991; 146:1169.
9. Romagnani S: Human TH1 and TH2
subsets: doubt no more. Immunol Today
1991; 12:256.
10. Murphy DB, Mamauchi K, Habu S, et al:
T cells in a suppressor circuit and non-T:
non-B cells bear different I-J determinants.
Immunogenetics 1981; 13:205.
11. Gillette TE, Chandler JW, Greiner JV:
Langerhans cells of the ocular surface.
12. Tagawa Y, Takeuchi T, Saga T, et al:
Langerhans cells: role in ocular surface
immunopathology. In: OConnor GR,
Chandler JW, eds. Advances in
immunology and immunopathology of the
eye. New York: Masson; 1985:203207.
13. Mc Dermott R, Ziylan U, Spehner D, et al:

Birbeck granules are subdomains of
endosomal recycling compartment in
human epidermal Langerhans cells, which
form where Langerin accumulates. Mol Biol
Cell 2002; 13:317335.
14. Mizumoto N, Takashima A: CD1a and
langerin: acting as more than Langerhans
cell markers. J Clin Invest 2004;
113:658660.
15. Le Douarin NM: Ontogeny of hematopoietic
organ studies in avian embryo interspecic
chimeras. Cold spring harbor meeting on
differentiation of normal and neoplastic
hematopoietic cells. In: Clarkson D, Marks
PA, Till JE, eds. Cold Spring Laboratory,
NY: Cold Spring Harbor Laboratory;
1978:532.
16. Metcalf D, Moore MAS: Hematopoietic
cells. In: Neuberger A, Tatum EL, eds.
Frontiers of biology. Amsterdam: Elsevier
North-Holland; 1971.
17. Hermans MJA, Hartsuiker H, Opstaelten D:
An insight to study of B lymphocytopoiesis
in rat bone marrow: topographical
arrangement of terminal yatsi nucleotidal
transferase positive cells and pre-B cells.
J Immunol 1989; 44:67.
18. Muller-Sieburg CL, Whitlock CA, Weissman
YL: Isolation of two early B lymphocyte
progenitors from mouse marrow: a
committed pre-B cell and a clonogenic 51
hematopoietic stem cell. Cell 1986; 44:653.
19. Whitlock CA, Witte ON: Longterm culture of
B lymphocytes and their precursors from
murine bone marrow. Proc Natl Acad Sci
USA 1982; 79:3608.
20. Whitlock CA, Tidmarsh TS, Mueller C, et al:
Bone marrow stromal cells with lymphoid
activity express high levels of pre-B
neoplasia-associated molecule. Cell 1987;
48:1009.
21. Hunt T, Robertson D, Weiss D, et al: A single
bone marrow-derived stromal cell type
supports the in vitro growth of early lymphoid
and myeloid cells. Cell 1987; 48:997.
22. Dorshkind K, Johnson A, Collins A, et al:

Generation of bone marrow stromal cultures
that support lymphoid and myelocyte
precursors. Immunol Methods 1986; 89:37.
23. Smith L, Weissman IL, Heimfeld S:
Metapoietic stem cells give rise to pre-B
cells. In: Paul W, ed. Fundamental
immunology. 2nd edn. New York: Raven;
1989:4167.
24. Szengerg A, Warner ML: Association of
immunologic responsiveness in fowls with
a hormonally arrested development of
lymphoid material. Nature 1962; 194:146.
25. Cooper MD, Peterson RD, South MA, Good
RA: The functions of the thymus system
and the bursa system in the chicken. J Exp
Med 1966; 123:75.
26. Popiernik M: Lymphoid organs. In: Bach JF,
ed. Immunology. 2nd edn. New York: Wiley;
1982:1537.
27. Butcher EC, Weissman IL: Lymphoid
tissues and organs. In: Paul W, ed.
Fundamental immunology. 2nd edn.
New York: Raven; 1989:117137.
28. Berg EL, Goldstein LA, Jutila MA, et al:
Homing receptors and vascular addressins:
cell adhesion molecules that direct
lymphocyte trafc. Immunol Rev 1989;
108:5.
29. Picker LJ, de los Toyos J, Tellen MJ, et al:
Monoclonal antibodies against the CD 44
and Pgp-1 antigens in man recognize the
Hermes class of lymphocyte homing
receptors. J Immunol 1989; 142:2046.
30. Holzmann B, McIntyre BW, Weissman IC:
Identication of a murine Peyers patchspecic lymphocyte homing receptor as an
integrin molecule with an a chain
homologous to human VLA-4a. Cell 1989;
56:37.
31. Streeter PR, Rause ET, Butcher EC:
Immunohistologic and functional
characterization of a vascular addressin
involved in lymphocyte homing into
peripheral lymph nodes. J Cell Biol 1988;
107:1853.
CHAPTER 6
REFERENCES
65
CHAPTER
T-Lymphocyte Responses
Reza Dana and J. W. Streilein
T lymphocytes, or T cells, stand at the center of the adaptive

immune response (see Chapter 5 for discussion of innate vs
adaptive immunity).1 T cells are absolutely critical for antigenspecic cell-mediated immunity, as well as for tolerance. In the
absence of T cells, only primitive antibody responses and no
cell-mediated immune responses can be made; even there,
the repertoire of antibodies generated suffers in the absence
of T cell help since CD4+ T cells play an important role in
supporting B-cell responses. The majority of T cells undergo
differentiation in the thymus gland and, upon reaching
maturity, disseminate via the blood to populate secondary
lymphoid organs and to circulate among virtually all tissues of
the body. A second population of T cells undergoes differentiation extra-thymically and has a somewhat different set of
functional properties. T cells are exquisitely antigen-specic, a
property conferred on them by unique surface receptors that
recognize antigenic material in a highly distinctive manner.
Once activated, T cells initiate or participate in the various
forms of cell-mediated immunity, humoral (antibody-mediated)
immunity, and tolerance.
T-LYMPHOCYTE DEVELOPMENT
The ontogeny of the various lymphocyte populations is complex
and incompletely understood. In essence, it is believed that a
pluripotent hematopoietic stem cell leads to a lineage of
cells that becomes the oligopotent lymphocyte progenitor.2
During fetal life, this lineage of cells is observed rst in the
liver, but as the fetus matures, the lymphocyte progenitors
shift to the bone marrow. According to developmental signals
not completely understood, lymphocyte progenitors in the
marrow differentiate into (at least) three distinct lineages of
committed precursor cells: pre-thymocytes, pre-B lymphocytes,
and pre-natural killer (NK) lymphocytes. Pre-thymocytes,
which give rise eventually to T lymphocytes, escape from
the bone marrow (or fetal liver) and migrate via the blood
primarily to the thymus, where cell-adhesion molecules on
microvascular endothelial cells direct them into the cortex.
The differentiation process that thymocytes experience within
the thymus accomplishes several critical goals: (1) each cell
acquires a unique surface receptor for an antigen; (2) cells with
receptors that recognize non-self antigenic molecules in the
context of self class I or class II molecules (encoded by genes
within the major histocompatibility complex (MHC)) are
positively selected;3 (3) cells with receptors that recognize
self-antigenic molecules in the context of self-MHC molecules
are negatively selected (i.e., deleted);4 and (4) each mature cell
acquires unique effector functions the capacity to respond to
antigen by secreting cytokines or by delivering a lethal hit to a
target cell.
DIFFERENTIATION IN THE THYMIC

CORTEX
Within the thymus cortex, pre-thymocytes receive differentiation signals from resident thymic epithelial cells and thus
initiate the process of maturation.2 A unique set of genes is
activated, including: (1) genes that commit the cells to proliferation, (2) genes that encode the T-cell receptors for antigen,
and (3) genes that code accessory molecules that developing and
mature T cells use for antigen recognition and signal
transduction. The genes that make it possible for T cells to
create surface receptors for antigen are the structural genes that
encode the four distinct polypeptide chains (a, b, g, d) from
which the T-cell receptor (Tcr) for antigen is composed, as well
as the genes that create genetic rearrangements that confer an
extremely high degree of diversity on Tcr molecules. The
portion of the Tcr that is involved in antigen recognition resides
at the ends of the peptide chains distal to the cell surface and is
called the combining site. It is thus that within the thymus
cortex, individual pre-thymocytes proliferate, come to express a
unique Tcr for an antigen, and simultaneously express CD3,
CD4, and CD8 on the cell surface. Each day, a very large
number of thymocytes is generated and, therefore, an enormous
diversity of Tcr is generated. Conservative estimates place
the number of novel Tcr produced each day in excess of 109, or
one billion!
NATURE OF ANTIGEN RECOGNITION BY

T CELLS
Understanding the nature of the antigenic determinants
detected by individual T-cell receptors for antigen is central to
understanding the differentiation process that occurs among
thymocytes in the thymus gland. Thymocytes acquire one of
two types of T-cell receptors: ab-Tcr are heterodimers composed
of polypeptides encoded by the Tcr-a and Tcr-b chain genes;
gd-Tcr are heterodimers composed of polypeptides encoded by
the Tcr-g and Tcr-d chain genes.5 Because much is known about
ab-Tcr, whereas much remains to be learned about gd-Tcr, this
discussion is limited to the former.
The ab-T-cell receptor for antigen does not recognize a
protein antigen in its native conguration. Rather, the Tcr
recognizes peptides (ranging in size from 7 to 22 amino acids in
length) derived from limited proteolysis of the antigen, and it
recognizes these peptides when they are bound noncovalently
to highly specialized regions of antigen-presenting molecules.6
Two types of antigen-presenting molecules exist, and both are
encoded within the MHC.7 Class I molecules are transmembrane proteins expressed on antigen-presenting cells
(APC). These molecules possess on their most distal domains a
67
SECTION 2
IMMUNOLOGY
groove that accommodates peptides (generated by regulated
proteolysis of antigenic proteins) ranging from seven to nine
amino acids in length. Class II molecules are also transmembrane proteins expressed on APC, and the platforms on
their distal domains contain similar grooves that accept
peptides of 1522 amino acids in length. Thus, the conditions
that must be met for successful recognition of antigen by Tcr
are: (1) a class I or class II molecule must be available on an
APC, and (2) a peptide must occupy the groove of the presenting
MHC molecule.
Within the thymus cortex, epithelial cells express class I and
class II molecules encoded by the individuals own MHC genes.2
When Tcr-bearing thymocytes are generated in the cortex, cells
with Tcr that recognize peptide-containing self-class I or selfclass II molecules are induced to undergo successive rounds of
proliferation, leading to clonal expansion. By contrast, Tcrbearing thymocytes that fail to recognize peptide-containing
self-class I or self-class II molecules are not activated within the
cortex. In the absence of this cognate signal, all such cells enter
a default pathway, which ends inevitably in cell death (apoptosis).
This process is called positive selection, because thymocytes
with Tcr that have an afnity for self-MHC molecules (plus
peptide) are being selected for further clonal expansion.
Unselected cells simply die by apoptosis. At the completion of
their sojourn in the thymus cortex, large numbers of positively
selected Tcr+, CD3+, CD4+, and CD8+ thymocytes migrate into
the thymus medulla.
DIFFERENTIATION IN THE THYMIC

MEDULLA AND MATURATION OF T CELLS
68
In addition to epithelial cells, the thymic medulla contains a

unique population of bone marrow-derived cells called dendritic
cells.8,9 These cells express large amounts of class I and class II
molecules and actively endocytose proteins in their environment. Peptides derived from these proteins by proteolysis are
loaded onto the grooves of MHC-encoded antigen presentation
platforms. Within the thymic medulla, the vast majority of such
endocytosed proteins are self proteins. As thymocytes enter the
medulla from the cortex, a subpopulation expresses Tcr that
recognize peptides of self proteins expressed on self-class I or
self-class II molecules. When these cells engage self-derived
peptides plus MHC molecules on the medullary dendritic cells,
a death (apoptotic) signal is generated to the T cells, and all
such cells undergo apoptosis. This process is called negative
selection because thymocytes with Tcr that have an afnity
for self-peptides in self-MHC molecules are being eliminated so
as to prevent these autoreactive cells from reacting to self
antigens a process that could lead to autoimmune disease.
Many other thymocytes that enter the medulla express Tcr
that are unable to engage self-class I or self-class II molecules on
dendritic cells, because the relevant peptide does not occupy the
antigen-presenting groove. T cells of this type proceed to
downregulate expression of either CD4 or CD8 and acquire the
properties of mature T cells. The mature T cells that are ready
at this point to leave the thymus are Tcr+, CD3+, and either
CD4+ or CD8+ (but not both). Moreover, they are in G0 of the
cell cycle, and hence resting. The number of such cells exported
from the thymus per day is very large; in humans, it is
estimated that more than 108 new mature T cells are produced
daily. These cells are fully immunocompetent and are prepared
to recognize and respond to a large diversity of foreign antigens,
but because they are antigen-inexperienced, they are called
naive. It is estimated that the number of different antigenic
specicities that can be recognized by mature T cells (i.e., the
T cell repertoire for antigens) exceeds 109, that is, far more than
the number of proteins expressed by the genome.
PROPERTIES AND FUNCTIONS OF

MATURE T LYMPHOCYTES
Mature, resting T cells migrate from the thymus to all tissues
of the body, but there are vascular specializations (postcapillary
venules) in secondary lymphoid organs (lymph nodes, Peyers
patches, tonsils) that promote the selective entry of T cells into
these tissues. More than 99% of T cells in blood that traverse a
lymph node are extracted into the parafollicular region of the
cortex. This region of the nodal cortex is designed to encourage
the interaction of T cells with APC, since this region is also the
preferential site where a majority of antigen-bearing APCs that
drain from peripheral tissues, also home. Because the encounter
of any single, antigen-specic T cell with its antigen of interest
on an APC is a relatively rare event, most T cells that enter a
secondary lymphoid organ fail to nd their antigen of interest
that is, the antigen for which they express the specic Tcr. In
this case, the T cells migrate into the effluent of the node,
passing through lymph ducts back into the general blood circulation. An individual unstimulated T cell may make journeys
such as this numerous times during a single day, and countless
journeys are accomplished during its lifetime. Remarkably, this
monotonous behavior changes dramatically if and when a
mature T cell encounters its specic antigen loaded on an APC
in a secondary lymphoid organ. It is this critical encounter that
initiates T cell-dependent antigen-specic immune responses.
T-CELL ACTIVATION BY ANTIGEN

There is a general rule regarding the requirements for activation
of lymphocytes, including T cells, which are normally in a
resting state: two different surface signals received simultaneously are required to arouse the cell out of G0.8 One signal
(referred to as signal 1) is triggered by successful engagement
of the Tcr with its peptide in association with an MHC
molecule. The other signal (referred to as signal 2) is delivered
through numerous cell surface molecules other than the Tcr.
Signals of this type are also referred to as co-stimulatory signals
and are the result of receptor/ligand interactions in which the
receptor is on the T cell and the ligand is expressed on the APC.
For example, B7.1 (CD80) and B7.2 (CD86) are surface
molecules expressed on APC; these molecules engage the
receptor CD28 on T cells, thus delivering an activation signal
to the recipient cells that also promotes their survival through
upregulation of signals that oppose apoptosis.10 Similarly,
CD40 ligand on T cells and CD40 on APC function in a costimulatory manner. When both conditions are met signal 1
(Tcr binds to peptide plus MHC molecule) and signal 2
(e.g., B7.1 binds to CD28) the T cell receives coordinated
signals across the plasma membrane, and these signals initiate
a cascade of intracytoplasmic events that lead to dramatic
changes in the genetic and functional programs of the T cells.
ANTIGEN-ACTIVATED T-CELL RESPONSES

When a T cell encounters its antigen of interest along with a
satisfactory signal 2, it escapes from G0. Under these circumstances, the genetic program of the cell shifts in a direction that
makes it possible for the cell to proliferate and to undergo
further differentiation. Proliferation results in emergence of a
clone of cells, all of the identical phenotype, including the Tcr.
This process is called clonal expansion, and results from the
elaboration of growth factors (e.g., IL-2), and represents a hallmark of the process of immunization or sensitization, that is,
the process by which the lymphocytes that are specic to an
antigen expand. The signal that triggers proliferation arises rst
from the APC, but sustained T-cell proliferation takes place
T-Lymphocyte Responses
T-CELL ANERGY
On occasion, T cells may encounter their antigen of interest
(in association with an MHC molecule) under circumstances
where an appropriate signal 2 does not exist. In this case,
delivery of signal 1 alone fails to activate the T cells. However,
if these same T cells are re-exposed subsequently to the same
antigen/MHC signal 1 on viable APC capable of delivering a
functional signal 2, activation of the T cells still fails. The
inability of T cells rst activated by signal 1 in the absence of
signal 2 to respond subsequently to functional signal 1 and
signal 2 is referred to as anergy (discussed in more detail in
Chapter 10).
T-CELL HETEROGENEITY AND

REGULATORY T CELLS
The adaptive immune response is separable into a cell-mediated
immune arm and an antibody or humoral immune arm (see
Chapter 5). T cells initiate and mediate cell-mediated immunity,

and also play a critical role in promoting antibody-mediated
responses.
CELL-MEDIATED IMMUNITY
Cell-mediated immunity arises when effector T cells are
generated within secondary lymphoid organs in response to
antigen-induced activation. Effector cells can be broadly divided
into two types: (1) for the most part CD4+ T cells that elicit
delayed-type hypersensitivity (DTH), and (2) CD8+ T cells that
are cytotoxic for antigen-bearing target cells. T cells that elicit
DTH recognize their antigen of interest on cells in peripheral
tissues and upon activation secrete proinflammatory cytokines
such as IFN-g and TNF-a, and thereby can cause signicant
bystander damage to neighboring cells. These cytokines act on
microvascular endothelium, promoting edema formation and
recruitment of monocytes, neutrophils, and other leukocytes to
the site. In addition, monocytes and tissue macrophages exposed
to these cytokines are activated to acquire phagocytic and
cytotoxic functions. Since it takes hours for these inflammatory
reactions to emerge, they are called delayed. It is generally
believed that the T cells that elicit delayed hypersensitivity
reactions are CD4+ and recognize antigen of interest in association with class II MHC molecules. However, ample evidence
exists to also implicate CD8+ T cells in this process (especially
in reactions within the central nervous system). Although the
elicitation of delayed hypersensitivity reactions is antigenspecic, the inflammation that attends the response is itself
nonspecic since there the cytokines secreted by DTH effector
T cells have profound paracrine effects on other nearby cells. In
contrast, effector responses elicited by cytotoxic T cells possess
much less nonspecic inflammation. Cytotoxic T cells interact
directly with antigen-bearing target cells and deliver a lethal
hit that is clean and highly cell-specic; there is virtually no
innocent bystander injury in this response.
CHAPTER 7
because the responding T cell activates its own IL-2 and IL-2
receptor genes.11,12 IL-2 is a potent growth factor for T cells, and
T cells expressing the IL-2R respond to IL-2 by undergoing repetitive rounds of replication. IL-2 is not the only growth factor for
T cells; another important growth factor is IL-4, which is also
made by T cells. Thus, once activated, T cells have the capacity
to autocrine stimulate their own proliferation, so long as their
Tcr remains engaged with the antigen (plus MHC) of interest.
In addition to proliferation, antigen-activated T cells proceed
down pathways of further differentiation. This is an important
concept, since not all antigen-specic T cells, even when activated, share the same functional properties. For example, CD4+
T cells can differentiate down distinct paths that allow them to
contribute differentially to the type of immune response
(T helper-1 vs T helper-2 type) generated.13 Additionally, CD8+
T cells can acquire the capacity for cytotoxicity, that is the
ability to lyse target cells.14 These functional properties are
often called the functional phenotype of the T-cell response,
and are largely determined by the pattern of cytokines produced
by the T cell(s). The list of lymphokines that an activated
mature T cell can make is long: IL-2, IL-3, IL-4, GM-CSF, IL-5,
IL-6, IL-10, interferon-gamma, etc. Similarly, the range of
biologic activities attributable to these cytokines is extremely
broad, and no single T cell produces all of these factors simultaneously, but in general, the specic immune response generated to an antigen (e.g., microbial, transplant, allergen, etc.) is
dominated by a specic T-cell response phenotype.
The ability of cytotoxic T cells to lyse antigen-bearing target
cells is embodied in specializations of the cells cytoplasm and
cell surface, including possession of granules that contain a
molecule, perforin, that can polymerize and insert into the
plasma membrane of a target cell, creating large pores. The
granules also contain a series of lytic enzymes (granzymes) that
enter the target cell, perhaps through the perforin-created pores,
and trigger apoptosis. There is a second mechanism by which T
cells can cause death of neighboring cells. Activated T cells
express high levels of Fas (also known as CD95), a cell-surface
glycoprotein that binds Fas ligand (CD95 ligand). It is a
member of the TNF receptor superfamily and its cytoplasmic
tail contains a death domain. After sustained activation,
T cells also express Fas ligand; when Fas interacts with Fas
ligand, the cell bearing Fas undergoes programmed cell death.
Thus, Fas ligand+ T cells can trigger apoptotic death in adjacent
cells that are Fas+, including other T cells. In fact, the ability of
antigen-activated T cells to elicit apoptosis among neighboring,
similarly activated, T cells serves as an important mechanism
for downregulating the immune response.
HUMORAL IMMUNITY
Humoral immunity arises when B cells produce antibodies in
response to antigenic challenge. Although antigen alone may be
sufcient to activate B cells to produce IgM antibodies, this
response is amplied in the presence of helper CD4+ T cells.
Signicant research since the 1990s has focused on how the
patterns of cytokines secreted by T cells can regulate B-cell
responses and the type of immunity generated.13 For example,
one polar form of helper T cell called Th1 responds to antigen
stimulation by producing IL-2, IFN-g, and TNF-a. In turn,
these cytokines influence B-cell differentiation in the direction
of producing complement-xing IgG antibodies. Th1 cells are
also responsible for generating DTH (as discussed earlier), and
hence are relevant to both humoral and cell-mediated
immunity. By contrast, Th2 cells (the other polar form of helper
T cell) respond to antigen stimulation by producing IL-4, IL-5,
IL-6, and IL-10. In turn, these cytokines influence B-cell
differentiation in the directions of producing non-complementxing IgG antibodies or IgA and IgE antibodies. The discovery
of these two polar forms of helper T cells (as well as numerous
intermediate forms) has had a profound impact on our understanding of the immune response and its regulation.
REGULATORY T CELLS
It is important to appreciate that the default setting of the
immune system is unresponsiveness, or more precisely having
a measured response. Were it not for this feature of immunity,
unchecked clonal expansion of lymphocytes would result in
69
SECTION 2
IMMUNOLOGY
lymphomatous growths, and unregulated inflammatory
responses in peripheral tissues in response to antigenic
challenges would lead to relentless tissue destruction. Since
these responses are not compatible with normal organ/tissue
function, or indeed life in some cases, the immune system has
generated protean mechanisms for tightly regulating how it
responds to challenges and how quickly these responses are
quenched (see Chapter 10 for details). Immune regulation at the
level of T cells is effected by numerous mechanisms: anergy,
clonal deletion, tolerance, regulation of APC maturity and
migration to lymphoid compartments, and cell death. These
mechanisms are tightly controlled and work in concert to
regulate both the induction and expression of immunity. Of
critical importance are natural T-regulatory cells that actively
promote immunologic quiescence in an antigen-dependent
fashion.15,16 In this way, immunity generated to foreign
(e.g., transplant) or self-antigens can be quenched in a timely
manner; thus, reexposure to the antigen will lead to a measured
response. Signicant research is currently underway to use
these T-regulatory cells in a manner that provides therapeutic
potential in autoimmune diseases.
T-CELL-DEPENDENT INFLAMMATION
Primarily by virtue of the lymphokines they produce, T cells
can cause immunogenic inflammation if they encounter their
antigen of interest in a peripheral tissue (see Chapter 9). As
noted above, CD4+ T cells are particularly capable of causing
tissue injury. In the case of Th1 type CD4+ T cells, these cells
produce IFN-g and other proinflammatory molecules. IFN-g
is a potent activator of microvascular endothelial cells and
macrophages. Activated endothelial cells become leaky,
permitting edema fluid and plasma proteins to accumulate at
the site. Activated endothelial cells also promote the immigration of blood-borne leukocytes, including monocytes, into
the site, and it is the activated macrophages that provide
much of the toxicity at the inflammatory site. These cells
respond to IFN-g by upregulating the genes responsible for nitric
oxide (NO) synthesis. NO, together with newly generated
reactive oxygen intermediates, creates much of the local necrosis associated with immunogenic inflammation. Because Th2
cells do not make IFN-g in response to antigenic stimulation,
one might expect that Th2 cells would not promote inflammatory injury, but this does not appear to be the case.13 Th2
cells have been directly implicated in immune inflammation,
including that found in the eye. One responsible Th2 cytokine
in this setting known to be capable of causing inflammation
is IL-4.
T CELLS IN DISEASE: INFECTIOUS,

IMMUNOPATHOGENIC, AUTOIMMUNE
It is generally believed that T cells were developed in response
to evolutionary pressure to respond to microbial, in particular
intracellular, pathogens, a belief based on the ability of T cells
to detect peptides derived from degradation of intracellular
or phagocytosed pathogens. This property is most obviously
revealed in viral infections where CD8+ T cells detect peptides
on virus-infected cells derived from viral proteins in association
with self class I molecules (so called altered self recognition).
Once recognition has occurred, a lethal hit is delivered to the
target cell, and lysis aborts the viral infection. T-cell immunity
is also conferred when CD4+ T cells detect peptides derived
from bacteria (or other pathogens) phagocytosed by macrophages or other antigen-presenting cells. Recognition in this
case does not result in delivery of a lethal hit; instead,
proinflammatory cytokines released by the activated T cells
cause the macrophages to acquire phagocytic and cytotoxic
functions that lead to the death of the offending pathogen.
To a limited extent with CD8+ cells, but to a greater extent
with CD4+ cells, the inflammation associated with the immune
attack on the invading pathogen can lead to injury of surrounding tissues. If the extent of this injury is of sufcient magnitude,
disease may result from the inflammation itself, quite apart
from the toxicity of the pathogen. This is the basis of the
concept of T-cell-dependent immunopathogenic disease. As
previously mentioned (see chapters on Overview of Immunology and Immune regulation), certain organs and tissues, especially the eye, are particularly vulnerable to immunopathogenic
injury. In tissues of this type, the immune response may prove
to be more problematic than the triggering infection itself!
In some pathologic circumstances, T cells mistakenly
identify self molecules as foreign, thus mediating an autoimmune response that can eventuate in disease. Although this
idea is conceptually sound, it is often difcult to identify the
offending self-antigen. Because of this difculty, it is frequently
impossible to determine whether a particular inflammatory
condition, initiated by T cells, is immunopathogenic in origin
(and, therefore, triggered by an unidentied pathogen) or autoimmune in origin. This is a particularly common problem in
the eye. To make matters more complicated, the increasing
appreciation for regulatory T cells makes it clear that not all
T lymphocytes are pathogenic, and that certain populations of
these cells may actually aid in terminating or attenuating
the immunoinflammatory response, providing yet one more
untoward complication of nonspecic immunosuppressive
medicines, in particular those that cause lymphopenia.
REFERENCES
70
1. Janeway CA Jr, Travers P, eds.

Immunobiology. 6th edn. New York:
Garland Publishing Inc; 2004.
2. Wu L: T lineage progenitors: the earliest
steps en route to T lymphocytes. Curr Opin
Immunol 2006; 18:121126.
3. Ladi E, Yin X, Chtanova T, Robey EA:
Thymic microenvironments for T cell
differentiation and selection. Nat Immunol
2006; 7:338343.
4. Siggs OM, Makaroff LE, Liston A: The why
and how of thymocyte negative selection.
Curr Opin Immunol 2006; 18:175183.
5. Krogsgaard M, Davis MM: How T cells
see antigen. Nat Immunol 2005;
6:239245.
6. Germain RN: MHC-dependent antigen
processing and peptide presentation:
7.
8.
9.
10.
11.
providing ligands for T lymphocyte

activation. Cell 1994; 76:287.
Germain RN, Jenkins MK: In vivo antigen
presentation. Curr Opin Immunol 2004;
16:120125.
Janeway CA, Bottomly K: Signals and
signs for lymphocyte responses. Cell 1994;
76:275.
Sprent J, Webb SR: Intrathymic and
extrathymic clonal deletion of
T cells. Curr Opin Immunol 1995; 7:196.
Kroczek RA, Mages HW, Hutloff A:
Emerging paradigms of T-cell
co-stimulation. Curr Opin Immunol 2004;
16:321327.
Jain J, Loh C, Rao A: Transcription
regulation of the IL-2 gene. Curr Opin
Immunol 1995; 7:333.
12. Minami Y, Kono T, Miyazaki T, Taniguchi T:

The IL-2 receptor complex: its structure,
function, and target genes. Annu Rev
Immunol 1993; 11:245.
13. Gor DO, Rose NR, Greenspan NS: Th1Th2: a procrustean paradigm. Nat Immunol
2003; 4:503505.
14. Catalfamo M, Henkart PA: Perforin and the
granule exocytosis cytotoxicity pathway.
Curr Opin Immunol 2003; 15:522527.
15. Randolph DA, Fathman CG: CD4+CD25+
regulatory T cells and their therapeutic
potential. Annu Rev Med 2006;
57:381402.
16. Picca CC, Caton AJ: The role of selfpeptides in the development of
CD4+CD25+ regulatory T cells. Curr Opin
Immunol 2005; 17:131136.
CHAPTER
B-Lymphocyte Responses
C. Stephen Foster and Fahd Anzaar
B-lymphocyte development from pluripotential bone marrow

stem cells influenced by endosteal region bone marrow
interstitial cells is introduced in Chapter 6. The rst stage to
develop in the bone marrow is designated the pro-B lymphocyte, which represents the earliest committed B-cell precursor. CD 19 expression is rst seen in this cell type, and
continues to be expressed in all subsequent (downstream)
B-cell lineages (including plasma cells) earning its designation
as the pan-B cell marker. However, it does not express CD 20,
whose expression is rst seen in the next stage of development,
the pre-B lymphocyte. Pro-B cells express the recombination
activating genes (RAG1 and 2), terminal deoxynucleotidyl
transferase (TdT) as well as genes that encode the surrogate
light chains, and the pro-B cell receptor, which has an unknown
function. Expression of the pre-B-cell receptor allows development and further maturation of the pre-B cells, which
contain cytoplasmic, but not membrane, immunoglobulin M
(IgM) heavy chains that associate with surrogate light chains
devoid of variable regions. These primitive immunoglobulin
molecules in pre-B cells, composed of complete, mature heavy
chains and surrogate light chains, are critical to the further
development of the B cell into the immature B lymphocyte
containing complete k or l light chains with suitable variable
regions. IgM is then expressed on the immature B-cell surface.
Interleukin-7, BAFF (B-cell activating factor of the TNF family)
and APRIL (a proliferation-inducing ligand) are important in the
process of B-cell development (acting by phosphorylating and
thus activating STAT5)1 as is tyrosine kinase in bone marrow
stromal cells and stem cells. Several B-cell transcription factors
(e.g., the E box proteins (E2A, HEB, E22) and early B-cell factor
(EBF)) are involved in this process, activating the B-cell
commitment factor Pax5, which in turn activates B-cell specic
genes (such as CD 19 and BLNK) and simultaneously represses
genes for other cell lines (through a TLE4 Groucho protein).2
Inhibition of Pax 5 is so detrimental to the development of B
cells that it has been shown to reprogram them to become
macrophages.3 When an antigen encounters cell-surface IgM
that has binding specicities for the antigen (e.g., self-antigens),
tolerance to the antigen is the typical result if such an encounter
precedes emigration of the B cell from the bone marrow.
Once the immature B cell has acquired its exit visa (complete surface IgM), it leaves the bone marrow, residing primarily
in the peripheral lymphoid organs (and blood), where it further
matures to express both IgM and IgD on its cell surface. It is
now a mature B cell, responsive to antigen with proliferation
and antibody synthesis.
CD 20 expression is limited to pre-B, immature, and mature
B cells. It is not seen in plasma cells or memory cells. This
forms the basis of therapy with Rituxan (rituximab), a chimeric
monoclonal antibody against CD 20, which induces lympho-
cyte death by activating apoptotic pathways (phospholipase Cg,

c-myc, bax, STAT3). The United States Food and Drug
Administration has approved it for treating B-cell nonHodgkins lymphomas, but is has also been used successfully
for treating autoimmune thrombocytopenia, systemic lupus
erythematosus, and rheumatoid arthritis.4 A major advantage
of Rituxan is that it does not affect stem cells or plasma cells,
and so has no effect on immunoglobulin levels, and does
not subject patients to the risk of developing opportunistic
infections. Conversely, the presence of long-lived plasma cells
may lead to continued production of pathogenic autoantibodies,
necessitating indenite treatment. CD 19 monoclonal antibodies have been tested in animal models of autoimmune
disease, and show a more durable depletion of B cells than does
anti-CD 20 therapy, affecting pre-B and immature B cells
(present, for example, in early lymphoblastic leukemias
unresponsive to Rituxan), eliminating them before antigenreceptor selection (and production of other pathogenic antibodies responsible for other disease states) occurs.5
The hallmark of the vertebrate immune system is its ability
to mount a highly specic response against virtually any foreign
antigen, even those never before encountered. The ability to
generate a diverse immune response depends on the assembly
of discontinuous genes that encode the antigen-binding sites of
immunoglobulin and T-cell receptors during lymphocyte
development. Diversity is generated through the recombination
of various germline gene segments, imprecise joining of
segments with insertion of additional nucleotides at the junctions, and somatic mutations occurring within the recombining
gene segments. Other factors, such as chromosomal position of
the recombining gene segments and the number of homologous
gene segments, may play a role in determining the specicities
of the antigen-recognizing proteins produced by a maturing
lymphocyte.
ANTIBODY DIVERSITY
The paradox of an individual possessing a limited number of
genes but the capability to generate an almost innite number
of different antibodies remained an enigma to immunologists
for a considerable time. The discovery of distinct variable (V)
and constant (C) regions in the light and heavy chains of
immunoglobulin molecules (Fig. 8.1) raised the possibility that
immunoglobulin genes possess an unusual architecture. In
1965, Dreyer and Bennett proposed that the V and C regions of
an immunoglobulin chain are encoded by two separate genes in
embryonic (germline) cells (germline gene diversity).6 According
to this model, one of several V genes becomes joined to the C
gene during lymphocyte development. In 1976, Hozumi and
Tonegawa discovered that variable and constant regions are
71
IMMUNOLOGY
SECTION 2
several J (joining)-segment genes, which encodes part of the last

hypervariable segment (Fig. 8.4).911 Additional diversity is
generated by allowing V and J genes to become spliced in
different joining frames (junctional diversity) (Fig. 8.5).10 There
are at least three frames for the joining of V and J. Two forms
FIGURE 8.1. Structure of IgG showing the regions of similar

sequence (domains).
encoded by separate, multiple genes far apart in germline DNA

that join to form a complete immunoglobulin gene active in B
lymphocytes.7 Immunoglobulin genes are thus translocated
during the differentiation of antibody-producing cells (somatic
recombination) (Fig. 8.2).
FIGURE 8.3. Hypervariable or CDRs on the antigen-binding site of

the variable regions of IgG.
STRUCTURE AND ORGANIZATION OF

IMMUNOGLOBULIN GENES
The V regions of immunoglobulins contain three hypervariable
segments that determine antibody specicity (Fig. 8.3).8 Hypervariable segments of both the light (L) and heavy (H) chains
form the antigen-binding site. Hypervariable regions are also
called complementarity-determining regions (CDRs). The V
regions of L and H chains have several hundred gene segments
in germline DNA; the exact number of segments is still being
debated but is estimated to range between 250 and 1000
segments.
FIGURE 8.4. A V gene is translocated near a J gene in forming a

light-chain V region gene.
LIGHT-CHAIN GENES
A complete gene for the V region of a light chain is formed by
the splicing of an incomplete V-segment gene with one of
72
FIGURE 8.2. Translocation of a V-segment gene to a C gene in the

differentiation of an antibody-producing B cell.
FIGURE 8.5. Imprecision in the site of splicing of a V gene to a J gene

(junctional diversity).
of light chains exist: (k) and (l). For kl chains, assume that
there are ~250 V-segment genes and four J-segment genes.
Therefore, a total of 250 4 3 (for junctional diversity), or
3000, kinds of complete VK genes can be formed by combinations of V and J.
which are flanked on both 5 and 3 ends by recognition

sequences of the 12-bp type.
HEAVY-CHAIN GENES
For 250 VH, 15 DH, and 4 JH gene segments that can be joined
in three frames, at least 45 000 complete VH genes can be
formed. Therefore, more than 108 different specicities can be
generated by combining different V, D, and J gene segments and
by combining more than 3000 L and 45 000 H chains. If the
effects of N-region addition are included, more than 1011
different combinations can be formed. This number is large
enough to account for the immense range of antibodies that can
be synthesized by an individual.
Far fewer V genes than VK genes encode light chains.
However, many more V amino-acid sequences are known.1517
It is therefore likely that mutations introduced somatically give
rise to much of the diversity of l light chains (somatic hypermutation).10 Likewise, somatic hypermutation further amplies
the diversity of heavy chains. To summarize, four sources of
diversity are used to form the almost limitless array of antibodies that protect a host from foreign invasion: germline gene
diversity, somatic recombination, junctional diversity, and
somatic hypermutation.
FIGURE 8.6. The variable region of the heavy chain is encoded by

V-, D-, and J-segment genes.
CHAPTER 8
Heavy-chain V-region genes are formed by the somatic

recombination of V, an additional segment called D (diversity),
and J-segment genes (Fig. 8.6). The third CDR of the heavy
chain is encoded mainly by a D segment. Approximately 15 D
segments lie between hundreds of VH and at least four JH gene
segments. A D segment joins a JH segment; a VH segment then
becomes joined to the DJH to form the complete VH gene. The
D to J rearrangements occur in pro-B cells, when the recombination activating genes (RAG 1 and 2) introduce a singlestranded nick on either side of the segments, assisted by
DNA-bending high mobility group proteins (HMGB1 and 2).
The V to DJ joining occurs in pre-B cells, and a pre-B-cell
receptor is expressed. The light chain gene rearragements now
take place, forming an immature B cell with a complete immunoglobulin molecule that is then expressed on the cells surface.
To further diversify the third CDR of the heavy chain, extra
nucleotides are inserted between V and D and between D and J
(N-region addition) by the action of terminal deoxyribonucleotidyl transferase.12 Introns, which are noncoding intervening
sequences, are removed from the primary RNA transcript.
The site-specic recombination of V, D, and J genes is
mediated by enzymes (immunoglobulin recombinase) that
recognize conserved nonamer and palindromic heptamer
sequences flanking these gene segments.13,14 The nonamer and
heptamer sequences are separated by either 12-base pair (bp) or
23-bp spacers (Fig. 8.7). Recombination can occur only between
the 12- and 23-bp types but not between two 12-bp types or two
23-bp types (called the 12/23 rule of V-gene-segment
recombination). For example, VH segments and JH segments are
flanked by 23-bp types on both their 5 and 3 ends.
Consequently, they cannot recombine with each other or
among themselves. Instead, they recombine with D segments,
SOURCES OF IMMUNOGLOBULIN GENE

DIVERSITY
REGULATION OF IMMUNOGLOBULIN
GENE EXPRESSION
Immunoglobin gene rearrangements are separated in time
(as discussed earlier) and also restricted to one locus. An incomplete V gene becomes paired to a J gene on only one of a pair
of homologous chromosomes. Successful rearrangement of one
heavy-chain V region prevents the process from occurring on
FIGURE 8.7. Recognition sites for the recombination of V-, D-, and
J-segment genes. V and J genes are flanked by sites containing
23-bp spacers, whereas D-segment genes possess 12-bp spacers.
Recombination can occur only between sites with different classes of
spacers.
73
SECTION 2
IMMUNOLOGY
the other heavy-chain allele. Only the properly recombined
immunoglobulin gene is expressed. Therefore, all of the V
regions of immunoglobulins produced by a single lymphocyte
are the same. This is called allelic exclusion.18,19
There are ve classes of immunoglobulins. An antibodyproducing cell rst synthesizes IgM and then IgG, IgA, IgE, or
IgD of the same specicity. Different classes of antibodies are
formed by the translocation of a complete VH (VHDH) gene from
the CH gene of one class to that of another.20 Only the constant
region of the heavy chain changes; the variable region of the
heavy chain remains the same (Fig. 8.8). The light chain
remains the same in this switch. This step in the differentiation
of an antibody-producing cell is called class switching and is
mediated by another DNA rearrangement called SS recombination (Fig. 8.9).21 This process is regulated by cytokines
produced by helper T cells, and also by BAFF10,22 For example,
switching to IgE class immunoglobulin production is provoked
by the CD4 TH2 cytokine, IL-4. Repetitive DNA sequences
called switch regions are located upstream of each CH gene;
double-stranded breaks in these regions precede the development of stem-and-loop structures, and a CSR recombinase
enzyme (aided by AID (activation-induced cytidine deaminase))
then combines the new variable and heavy chain segments.
New evidence indicates that in addition to the cytokine milieu,
the type of antibody produced is also biased towards those heavy
chain gene segments that are in closest proximity to the preswitch heavy chain gene.22 The number of cells that have
undergone class switching depends on the number of divisions
the cell has performed rather than on the time since
stimulation by cytokines.22
DETERMINATION OF B-CELL REPERTOIRE

V-segment genes can be grouped into families based on their
DNA sequence homologies. In general, variable genes sharing
greater than 80% nucleotide similarity are dened as a family.23
There are 11 VH gene families currently known in the
mouse2326 and 6 in humans.2730 At least 29 families are known
for the V of murine light-chain genes.31,32 In fetal pre-B cells,
74
FIGURE 8.8. The VH region is rst associated with Cm and then with
another C region to form an H chain of a different class in the
synthesis of different classes of immunoglobulins.
chromosomal position is a major determinant of VH rearrangement frequency, resulting in a nonrandom repertoire that is
biased toward use of VH families closest to the JH segments.3336
In contrast, random use of VH families based on the number of
members in each family occurs in mature B cells without bias
toward JH proximal families.3739 The preferential VH gene rearrangement frequency seen in pre-B cells presumably becomes
normalized when contact of the organism with a foreign antigen
selects for the expression of the entire VH gene repertoire. One
can speculate that members of VH families preferentially used in
the pre-B cell encode antibody specicities that are needed in
the early development of the immune system.40
Immunoglobulins are serum proteins that migrate with the
globulin fractions by electrophoresis.7 Although they are glycoproteins, the molecules primary functions are determined by
their polypeptide sequence.8 At one end of the immunoglobulin,
the amino terminus, is a region that binds a site (epitope) on an
antigen with great specicity. At the other end, the carboxyl
terminus, is a non-antigen-binding region responsible for
various functions, including complement xation and cellular
stimulation via binding to cell-surface Ig receptors. The generalized structure of immunoglobulin is best understood initially by
examining its most common class, IgG (see Fig. 8.1).
IgG is composed of four polypeptide chains: two identical
heavy chains and two identical light chains. Heavy chains
weigh about twice as much as light chains. The identical heavy
chains are covalently linked by two disulde bonds. One light
chain is associated with each of the heavy chains by a disulde
bond and noncovalent forces. The two light chains are not
linked. Asparagine residues on the heavy chains contain carbohydrate groups. The amino terminals of one light chain and its
linked heavy chain compose the region for specic epitopebinding. The carboxyl termini of the two heavy chains constitute the non-antigen-binding region.
Each polypeptide chain, whether light or heavy, is composed
of regions that are called constant (C) or variable (V). A variable
FIGURE 8.9. The VHDJH gene moves from its position near Cm to one
near Cg1 by SS recombination.
region on a light chain is called VL, the constant region of a

heavy chain is called CH, and so forth. If the amino acid
sequence of multiple light or heavy chains is compared, the
constant regions will vary little, whereas the variable regions
differ greatly. The light chains are divided approximately equally
into a constant (CL) and variable (VL) region at the carboxyl and
amino terminals, respectively. The heavy chains also contain a
similar length of variable region (VH) at the amino terminals,
but the constant region (CH) is three times the length of the
variable region (VH). The variable regions are responsible for
antigen-binding, and it is this variability that accounts for the
ability to bind to millions of potential and real epitopes.9 Because each antibody molecule has two antigen-binding sites
with variable regions, cross-linking of two identical antigens
may be performed by one antibody. The constant regions carry
out effector functions that are common to all antibodies of a
given class (e.g., IgG) without the requirement of unique
binding sites.
The function of various regions of the immunoglobulin
molecule was determined in part by the use of proteolytic
enzymes that digest these molecules at specic locations. These
enzymes have also been exploited for the development of
laboratory reagents. The enzyme papain splits the molecule on
the amino terminal side of the disulde bonds that link the
heavy chains, resulting in three fragments: two identical Fab
fragments (each composed of the one entire light chain and a
portion of the associated heavy chain) and one Fc fragment
composed of the linked carboxyl terminal ends of the two
heavy chains. In contrast, treatment with the enzyme pepsin
results in one molecule composed of two linked Fab fragments
known as F(ab).7 The Fc fragment is degraded by pepsin
treatment.
Within some classes of immunoglobulins, whole molecules
may combine with other molecules of the same class to form
polymers with additional functional capabilities. J chains
facilitate the association of two or more immunoglobulins
(Fig. 8.10), most notably IgA and IgM. Secretory component is
a polypeptide synthesized by nonmotile epithelium found near
mucosal surfaces. This polypeptide may bind noncovalently to
IgA molecules, allowing their transport across mucosal surfaces
to be elaborated in secretions.
Five immunoglobulin classes are recognized in humans: IgG,
IgM, IgA, IgE, and IgD (Table 8.1). Some classes are composed
of subclasses as well. The class or subclass is determined by the
structure of the heavy-chain constant region (CH).10 The heavy
chains g, m, a, e, and d are found in IgG, IgM, IgA, IgE, and IgD,
respectively. Four subclasses of IgG and two subclasses of both
IgA and IgM exist (Table 8.2). The two light chains on any
immunoglobulin are identical and, depending on the structure
of their constant regions, may be designated k or l. Kappa
chains tend to predominate in human immunoglobulins
regardless of the heavy chain-determined class. Whether an
immunoglobulin is composed of two k or two l chains does not
determine its functional capabilities. Heavy chain-determined
class does dictate important capacities.11
CHAPTER 8
c
FIGURE 8.10. Schematic diagram of polymeric human
immunoglobulins. (a) IgM. (b) Secretory IgA. (c) Serum IgA.
xation. IgG is the only immunoglobulin class to cross the

placenta, an important aspect in fetal defense. Via their Fc
portion, IgG molecules bind Fc receptors found on a host of
inflammatory cells. Such binding activates cells such as macrophages and natural killer cells, enhancing cytotoxic activities
important in the immune response.
IMMUNOGLOBULIN G
The most abundant of the human classes in serum, immunoglobulin G (IgG) constitutes about three-quarters of the total
serum immunoglobulins. Respectively, IgG1 and IgG2 make up
~60% and 20% of the total IgG. IgG3 and IgG4 are relatively
minor components. IgGs are the primary immunoglobulin
providing immune protection in the extravascular compartments of the body. IgG is able to x complement in the serum,
an important function in inducing inflammation and controlling infection. IgG3 and IgG1 are most adept at complement
IMMUNOGLOBULIN M
Less abundant in the serum than IgG, IgM typically exists as a
pentameric form, stabilized by J chains, theoretically allowing
the binding of 10 epitopes. (In vivo, this is usually limited by
steric considerations.) IgM appears early in the immune
response to antigen and is especially efcient at initiating agglutination, complement xation, and cytolysis. IgM probably
preceded IgG in the evolution of the immune response and is
the most important antibody class in defending the circulation.
75
IMMUNOLOGY
IMMUNOGLOBULIN INTRACLASS
DIFFERENCES
SECTION 2
IMMUNOGLOBULIN A
Immunoglobulin A (IgA) is found in secretions of mucosal
surfaces as well as in the serum. In secretions, it exists as a
dimer coupled by J chains and stabilized by secretory component. IgA protects mucosal surfaces from infections but may
also be responsible for immunologic surveillance at the site of
rst contact with antigen. IgA in secretion is hardy, able to
withstand the ravages of proteolytic degradation.
Differences among the immunoglobulin classes are known as

isotypes, because all of the normal individuals in a species
possess all of the classes. Allotype refers to antigenic structures
on immunoglobulins that may differ from one individual to
another within a species. Idiotype refers to differences among
individual antibodies and is determined by the variable domain.
Just as the variable domain allows for antibodies to recognize
many antigens (epitopes), these differences also allow individual
antibodies to be recognized on the basis of their idiotype. In
fact, antibodies directed against antibodies exist and are called
anti-idiotypic antibodies. They are crucial to the regulation of
the antibody response and constitute the basis for Jernes
idiotype network.
IMMUNOGLOBULIN D
Immunoglobulin D (IgD) is present in minute amounts in the
serum and is the least stable of the immunoglobulins. Its
function is not known, but it probably serves as a differentiation
marker. IgD is found on the surface of B lymphocytes (along
with IgM) and may have a role in class switching and tolerance.
COMPLEMENT
IMMUNOGLOBULIN E
The complement system functions in the immune response by

allowing animals to recognize foreign substances and defend
themselves against infection.29 The pathways of complement
activation are complex (Fig. 8.11).30 Activation begins with
the formation of antigen-antibody complexes and the ensuing
generation of peptides that lead to a cascade of proteolytic
events. The particle that activates the system accumulates a
Immunoglobulin E (IgE) is notable for its ability to bind to mast

cells; when cross-linked by antigen, it causes a variety of
changes in the mast cell, including release of granule contents
and membrane-derived mediators. Although recognized as a
component of the allergic response, the role of IgE in protective
immunity is speculative.
TABLE 8.1. Diversity in TCR and Immunoglobulin Genes

Immunoglobulin
H
250
100
25
10
0
4
0
50
2
12
0
3
2
2
Germline
Variable (V)
2501000
Segments
Diversity (D)
Joining (J)
15
4
Variable region combinations
TCR
62 500250 000
2500
50
Use of different D and

J segments
Yes
Yes
Yes
Yes
Yes
Junctional
Variability in 3
Rarely
Rarely
Yes
No
Yes
Yes
Diversity
Joining of V and J
D joining in all three
reading frames
Rarely
N-region diversity
V-D, D-J
None
V-J
Often
V-D, D-J
Often
V-J
V-D, D1-D2
Junctional combinations
108
1015
1018
Total repertoire
1011
1017
1019
The numbers of the V, D, and J gene segments in the murine genome are shown. Total repertoire produced by the various mechanisms for generating diversity was
estimated.
TABLE 8.2. Human Immunoglobulin Subclasses
76
Immunoglobulin
Subclasses
Predominant Subclass
Unique Characteristics
IgG
1, 2, 3, and 4
1 (65%) and 2 (25%)
IgG2 crosses placenta poorly

IgG3 aggregates spontaneously
IgG4 blocks IgE binding; poor classic complement fixation
IgA
1 and 2
IgM
1 and 2
similar to that of cell-mediated cytotoxicity (as with natural

killer cells). Membrane lesions result from insertion of tubular
complexes into the membranes, leading to uptake of water with
ion-exchange disruption and eventual osmotic lysis.
The complement system interfaces with a variety of immune
responses, as outlined earlier, and with the intrinsic coagulation
pathways.36 Complement activity is usually measured by
assessing the ability of serum to lyse sensitized sheep red blood
cells.37 Values are expressed as 50% hemolytic complement
units per millimeter. The function of an individual component
may be studied by supplying excess quantities of all the other
components in a sheep red blood cell lysis assay.38 Components
are quantitated by radial diffusion or immunoassay. Complement may be demonstrated in tissue sections by immunofluorescence or enzymatic techniques.
Complement plays a role in a number of human diseases.
Complement-mediated cell lysis is the nal common pathologic
event in type III hypersensitivity reactions. Deciencies of
complement exist in the following human disorders: systemic
lupus erythematosus, glomerulonephritis, Raynauds phenomenon, recurrent gonococcal and meningococcal infections,
hereditary angioedema, rheumatoid disease, and others.33
CHAPTER 8
B-CELL RESPONSE TO ANTIGEN

PRIMARY RESPONSE
FIGURE 8.11. Simplied schematic of steps in classic and alternate

complement cascades.
protein complex on its surface that often leads to cellular

destruction via disruption of membranes.
Two independent pathways of complement activation are
known. The classic pathway is initiated by IgG- and IgMcontaining immune complexes.31 The alternative pathway is
activated by aggragated IgA or complex polysaccharides from
microbial cell walls.32 One component, C3, is crucial to both
pathways and in its proactive form can be found circulating in
plasma in large concentrations. Deciency or absence of C3
results in increased susceptibility to infection.33 Cleavage of C3
may result in at least seven products (lettered a through g), each
with biologic properties related to cellular activation and
immune and nonimmune responses.34 C3a, for instance,
causes the release of histamine from mast cells, neutrophil
enzyme release, smooth muscle contraction, suppressor T-cell
induction, and secretion of macrophage IL-1, prostaglandin,
and leukotriene.35 C3e enhances vascular permeability. C3b
binds to target cell surfaces and allows opsonization of biologic
particles.
The alternative pathway probably is a rst line of defense,
because unlike the classic pathway, it may neutralize foreign
material in the absence of antibody. The initiating enzyme of
this pathway, factor D, circulates in an active form and may
protect bystander cells from inadvertent destruction following
activation of the pathway.
The nal step of both pathways is membrane damage leading
to cytolysis. Both pathways require the assembly of ve
precursor proteins to effect this damage: C5, C6, C7, C8, and
C9. The mechanism of complement-mediated cell lysis is
Naive B cells respond to protein antigen in much the same way

that T cells do, through the help of antigen-presenting cells and
helper T cells. An antigen-presenting cell (usually a macrophage or dendritic cell) processes the antigen and presents it to
an antigen-specic helper (CD4) T cell, generally in the T-cellrich zones of the required lymph node. The T cell is thus
activated, expresses the membrane protein gp39, secretes
cytokines (e.g., IL-2 and IL-6), and binds to similarly activated
antigen-specic B cells (activated by the binding cross-linking of
antigen to surface IgM- and IgD-binding sites). The T-cell/B-cell
proliferation and a cascade of intracellular protein phosphorylation events, together with T-cell cytokine signals, result in
production of transcription factors that induce transcription of
various B-cell genes, including those responsible for production
of IgM light and heavy chains with paratopes specic to the
antigen epitopes that initiated this primary B-cell response. The
proliferating B cells form germinal centers in the lymph node
follicles, and somatic hypermutation of the IgV genes in some
of these cells results in the evolution of a collection of B cells in
the germinal center with surface IgM of even higher antigenbinding afnity. This phenomenon is called afnity maturation
of the primary antibody response. Those cells with the greatest
antigen-binding afnity survive as this primary B-cell response
subsides, persisting as long-lived memory cells responsible for
the classic distinguishing characteristics of the secondary
humoral immune response.
SECONDARY RESPONSE
The development of the secondary humoral immune response
is markedly accelerated compared with the primary response,
and it is greatly amplied in terms of magnitude of antibody
production (Fig. 8.12). The secondary response differs from the
primary one in the isotype or isotypes of antibody produced, as
well as in the avidity of the paratopes for the epitopes on the
elicited antigen. IgG, IgA, and IgE isotypes may now be seen
in the effector phase of this secondary humoral immune
response, and the binding afnities of these antibodies are
usually greater than that of the IgM elicited in the primary
response.
77
SECTION 2
IMMUNOLOGY
FIGURE 8.12. Relative synthesis of IgG and IgM following initial and
subsequent antigen injection.
The cellular and molecular events of the secondary B-cell

response are considerably different from those of the primary
response. Memory B cells themselves become the preeminent
antigen-binding, processing, and presenting cells, presenting
peptide fragments (antigenic determinants) to CD4 helper T
cells in the typical major histocompatibility complex-restricted
fashion, with processed peptide/human leukocyte antigen/DR
motifs interacting with the appropriate elements of the T-cell
receptor for antigen at the same time that B-cell CD40 and Tcell gp39 signaling occurs. Additionally, various T-cell cytokines
induce the memory B cells to divide, proliferate, produce
antibody, and switch the class of antibody being produced,
depending on the sum-total message being received by the B
cell: the nature of the antigenic stimulus, the amount and the
site of stimulation, and the site of the cells involved in the
cognitive and activation phases of the secondary response.
Memory cells of each immunoglobulin isotype involved in the
secondary response will, of course, persist after devolution of
the response.
REFERENCES
78
1. Johnson SE, Shah N, PanoskaltsisMortari A, LeBien TW: Murine and human

IL-7 activate STAT5 and induce proliferation
of normal human pro-B cells. J Immunol
2005; 175:73257331.
2. Milili M, Gauthier L, Veran J, et al: A new
Groucho TLE4 protein may regulate the
repressive activity of Pax5 in human
B lymphocytes. Immunology 2002;
106:447455.
3. Xie H, Ye M, Feng R, Graf T: Stepwise
reprogramming of B cells into
macrophages. Cell 2004; 117:663676.
4. Silverman GJ, Weisman S: Rituximab
therapy and autoimmune disorders.
Arthritis Rheum 2003; 48:14841492.
5. Yazawa N, Hamaguchi Y, Poe JC, Tedder TF:
Immunotherapy using unconjugated CD19
monoclonal antibodies in animal models for
B lymphocyte malignancies and
autoimmune disease. Proc Natl Acad Sci
USA 2005; 102:1517815183.
6. Dreyer WJ, Bennett JC: The molecular
basis of antibody formation: a paradox.
Proc Natl Acad Sci USA 1965; 54:864.
7. Hozumi N, Tonegawa S: Evidence for
somatic rearrangement of immunoglobulin
genes coding for variable and constant
regions. Proc Natl Acad Sci USA 1976;
73:3628.
8. Wu TT, Kabat EA: An analysis of the
sequences of the variable regions of Bence
Jones proteins and myeloma light chains
and their implications for antibody
complementarity. J Exp Med 1970;
132:211.
9. Leder P: The genetics of antibody diversity.
Sci Am 1982; 246:102.
10. Tonegawa S: Somatic generation of
antibody diversity. Nature 1983; 302:575.
11. Honjo T, Habu S: Origin of immune
diversity: genetic variation and selection.
Annu Rev Biochem 1985; 54:803.
12. Alt FW, Baltimore D: Joining of
immunoglobulin heavy chain gene
segments: implications from a
chromosome with evidence of three D-JH
fusions. Proc Natl Acad Sci USA 1982;
79:4118.
13. Early P, Huang H, Davis M, et al: An
immunoglobulin heavy chain variable region
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
gene is generated from three segments of

DNA: VH, D and JH. Cell 1980; 12:981.
Sakano H, Huppi K, Heinrich G,
Tonegawa S: Sequences at the somatic
recombination sites of immunoglobulin
light-chain genes. Nature 1979; 280:288.
Weigert MG, Cesari IM, Yondovich SJ,
Cohn M: Variability in the lambda light
chain sequences of mouse antibody.
Nature 1970; 228:1045.
Brack C, Hirama M, Lenhard-Schuller R,
Tonegawa S: A complete immunoglobulin
gene is created by somatic recombination.
Cell 1978; 15:1.
Bernard O, Hozumi N, Tonegawa S:
Sequences of mouse immunoglobulin light
chain genes before and after somatic
changes. Cell 1978; 15:1133.
Pernis BG, Chiappino G, Kelus AS,
Gell PGH: Cellular localization of
immunoglobulins with different allotypic
specicities in rabbit lymphoid tissues.
J Exp Med 1965; 122:853.
Cebra J, Colberg JE, Dray S: Rabbit
lymphoid cells differentiated with respect to
alpha-, gamma-, and mu-heavy
polypeptide chains and to allotypic markers
for Aa1 and Aa2. J Exp Med 1966;
123:547.
Kataoka T, Kawakami T, Takahasi N,
Honjo T: Rearrangement of immunoglobulin
g1-chain gene and mechanism for heavychain class switch. Proc Natl Acad Sci USA
1980; 77:919.
Gritzmacher CA: Molecular aspects of
heavy-chain class switching. Cri Rev
Immunol 1989; 9:173.
Yaish B, Mehr R. Models for the
dynamics and order of immunoglobulin
isotype switching. Bull Math Biol 2005;
67:1532.
Brodeur PH, Riblet R: The immunoglobulin
heavy chain variable region (Igh-V) locus in
the mouse I. One hundred Igh-V genes
comprise seven families of homologous
genes. Eur J Immunol 1984; 14:922.
Winter EA, Radbruch A, Krawinkel U:
Members of novel VH gene families are
found in VDJ regions of polyclonally
activated B lymphocytes. EMBO J 1985;
4:2861.
25. Kofler R: A new murine Ig VH family.

J Immunol 1988; 140:4031.
26. Reininger L, Kaushik A, Jaton JC: A
member of a new VH gene family encodes
anti-bromelinised mouse red blood cell
autoantibodies. Eur J Immunol 1988;
18:1521.
27. Rechavi G, Bienz B, Ram D, et al:
Organization and evolution of
immunoglobulin VH gene subgroups. Proc
Natl Acad Sci USA 1982; 79:4405.
28. Rechavi G, Ram D, Glazer R, et al:
Evolutionary aspects of immunoglobulin
heavy chain variable region (VH) gene
subgroups. Proc Natl Acad Sci USA 1983;
80:855.
29. Matthyssens G, Rabbitts TH: Structure and
multiplicity of genes for the human
immunoglobulin heavy chain variable
region. Proc Natl Acad Sci USA 1980;
77:6561.
30. Berman JE, Mellis SJ, Pollock R, et al:
Content and organization of the human Ig
VH locus: Denition of three new VH
families and linkage to the Ig CH locus.
EMBO J 1988; 7:727.
31. Potter M, Newell JB, Rudikoff S, Haber E:
Classication of mouse VK groups based
on the partial amino acid sequence to the
rst invariant tryptophan: impact of 14 new
sequences from IgG myeloma proteins.
Mol Immunol 1982; 12:1619.
32. DJoostelaere LA, Huppi K, Mock B, et al:
The immunoglobulin kappa light chain
allelic groups among the Igk haplotypes
and Igk crossover populations suggest a
gene order. J Immunol 1988; 141:652.
33. Yancopoulos GD, Desiderio SV, Pasking M,
et al: Preferential utilization of the most
JH-proximal VH gene segments in pre-B
cell lines. Nature 1984; 311:727.
34. Perlmutter RM, Kearney JF, Chang SP,
Hood LE: Developmentally controlled
expression of immunoglobulin VH genes.
Science 1985; 227:1597.
35. Reth M, Jackson N, Alt FW: VHDJH
formation and DJH replacement during
pre-B differentiation: non-random usage of
gene segments. EMBO J 1986; 5:2131.
36. Lawler AM, Lin PS, Gearhart PJ: Adult
B-cell repertoire is biased toward two
38. Dildrop R, Krawinkel U, Winter E,
Rajewsky K: VH-gene expression in murine
lipopolysaccharide blasts distributes over
the nine known VH-gene groups and may
be random. Eur J Immunol 1985; 15:1154.
39. Schulze DH, Kelsoe G: Genotypic analysis
of B cell colonies by in situ hybridization.
Stoichiometric expression of the three VH

families in adult C57BL/6 and BALB/c
mice. J Exp Med 1987; 166:163.
40. Krawinkel U, Cristoph T, Blankenstein T:
Organization of the Ig VH locus in mice
and humans. Immunol Today 1989;
10:339.
CHAPTER 8
heavy-chain variable region genes that

rearrange frequently in fetal pre-B cells.
Proc Natl Acad Sci USA 1987; 84:2454.
37. Yancopoulos GD, Malynn B, Alt FW:
Developmentally regulated and strainspecic expression of murine VH gene
families. J Exp Med 1988; 168:417.
79
CHAPTER
Immune-Mediated Tissue Injury

C. Stephen Foster, Miguel C. Coma, and J. Wayne Streilein
The immune response of an organism to an antigen may be

either helpful or harmful. If the response is excessive or
inappropriate, the host may incur tissue damage. The term
hypersensitivity reactions has been applied to such excessive
or inappropriate immune responses. Four major types of
hypersensitivity reactions are described, and all can occur in the
eye (Table 9.1). The necessary constituents for these reactions
are already present in or can be readily recruited into ocular
tissues. Immunoglobulins, complement components, inflammatory cells, and inflammatory mediators can, under certain
circumstances, be found in ocular fluids (i.e., tears, aqueous
humor, vitreous) and in the ocular tissues, adnexa, and orbit.
Unfortunately, these tissues (especially the ocular tissues) can
be rapidly damaged by inflammatory reactions that produce
irreversible alterations in structure and function. Some authors
have described a fifth type of hypersensitivity reaction, but this
adds little to our real understanding of disease mechanisms and
is unimportant to us as ophthalmologists in the study and care
of patients with destructive ocular inflammatory diseases. For
this reason, this discussion is confined to the classic four types
of hypersensitivity reactions that were originally proposed by
Gell, Coombs, and Lackmann.
Multiple theories about the etiology of these autoimmune
diseases have been postulated. Infections play a crucial role in
the induction and exacerbation, but sometimes also in
inhibition of these entities. The protection, induced by
infection, against some autoimmune and atopic disorders could
be related to immunoregulation that normally keeps the

immune system balanced, generated by production of
immunosuppressive cytokines, such as IL-10 or TGF-.1 On
the other hand, there is also good evidence supporting infection
as a possible cause of exacerbation or even generation of
autoimmune and allergic disorders (as in postinfectious
encephalitis disseminata or rheumatic fever).2
INJURY MEDIATED BY ANTIBODY

TYPE I HYPERSENSITIVITY
The antigens typically responsible for type I (immediate)
hypersensitivity reactions are ubiquitous environmental
allergens such as dust, pollen, dander, microbes, and drugs.
Under ordinary circumstances, exposure of an individual to
such materials is associated with no harmful inflammatory
response. The occurrence of such a response is considered,
therefore, out of place (Greek, a topos) or inappropriate, and it
is for this reason that Cocoa and Cooke coined the word atopy
in 1923 to describe individuals who develop such inappropriate
inflammatory or immune responses to ubiquitous environmental
agents.3 The antibodies responsible for type I hypersensitivity
reactions are homocytotropic antibodies, principally immunoglobulin E (IgE) but sometimes IgG4 as well. The mediators of
the clinical manifestations of type I reactions include
histamine, serotonin, leukotrienes (including slow-reacting
TABLE 9.1. Gell, Coombs, and Lackmann Hypersensitivity Reactions

Type
Participating Elements
Systemic Examples
Ocular Examples
Type I
Allergen, IgE, mast cells
Allergic rhintis, allergic asthma,

anaphylaxis
Seasonal allergic conjunctivitis, vernal

keratoconjunctivitis, atopic keratoconjunctivitis,
giant papillary conjunctivitis
Type II
Antigen, IgG, IgG3, or

IgM, complement,
neutrophils (enzymes),
macrophages (enzymes)
Goodpastures syndrome, myasthenia

gravis
Ocular cicatricial pemphigoid, pemphigus vulgaris

dermatitis herpetiformis
Type III
Antigen, IgG, IgG3, or IgM,

complement-immune
complex, neutrophils
(enzymes), macrophages
(enzymes)
StevensJohnson syndrome,
rheumatoid arthritis, systemic
lupus erythematosus, polyarteritis
nodosa, Behets disease,
relapsing polychondritis
Ocular manifestations of diseases are Systemic

Examples
Type IV
Antigen, T cells, neutrophils,

macrophages
Transplant rejection, tuberculosis,

sarcoidosis,Wegeners
granulomatosis
Contact hypersensitivity (drug allergy), herpes

disciform keratitis, phlyctenulosis, corneal
transplant rejection, tuberculosis, sarcoidosis,
Wegeners granulomatosis, uveitis, herpes
simplex virus, stromal keratitis, river blindness
81
IMMUNOLOGY
SECTION 2
TABLE 9.2. Mast Cell Mediators

Preformed in Granules
Newly Synthesized
Histamine
LTB4
Heparin
LTC4
Tryptase
LTD4
Chymase
Prostaglandins
Kinins
Thromboxanes
Eosinophil chemotactic factor
Platelet-activating factor
Neutrophil chemotactic factor

a
Serotonin
Chondroitin sulfate
Arylsulfatase
FIGURE 9.1. Type I hypersensitivity reaction mechanism. (a) Mast cell

Fc receptors have antigen-specific IgE affixed to them by virtue of the
patients being exposed to the antigen and mounting an inappropriate
(atopic) immune response to that antigen, with resultant production of
large amounts of antigen-specific IgE antibodies. The antibodies have
found their way to the MMC, have bound to the mast cells, but have
not provoked allergic symptoms because the patient is no longer
exposed to the antigen. (b) Second (or subsequent) exposure to the
sensitizing antigen or allergen results in a bridging binding reaction of
antigen to two adjacent IgE antibodies affixed to the mast cell plasma
membrane. (c) The antigenantibody bridging reaction shown in (b)
results in profound changes in the mast cell membrane, with
alterations in membrane-bound adenyl cyclase, calcium influx, tubulin
aggregation into microtubules, and the beginning of the degranulation
of the preformed mast cell mediators from their storage granules.
(d) The degranulation reaction proceeds, and newly synthesized
mediators, particularly those generated by the catabolism of
membrane-associated arachidonic acid, begin to work. The array of
liberated and synthesized proinflammatory and inflammatory
mediators is impressive.
82
substance of anaphylaxis (SRS-A)), kinins, and other vasoactive

amines. Examples of type I hypersensitivity reactions include
anaphylactic reactions to insect bites or to penicillin injections,
allergic asthma, hay fever, and seasonal allergic conjunctivitis.
It should be emphasized that in real life the four types of
hypersensitivity reactions are rarely observed in pure form, in
isolation from each other, and it is typical for hypersensitivity
reactions to have more than one of the classic Gell and
Coombs responses as participants in the inflammatory
problem. For example, eczema, atopic blepharokeratoconjunctivitis, and vernal keratoconjunctivitis have hypersensitivity
reaction mechanisms of both type I and type IV. The atopic
individuals who develop such abnormal reactions to environmental materials are genetically predisposed to such responses.
The details of the events responsible for allergy (a term coined
in 1906 by von Pirquet, in Vienna, meaning changed reactivity)

are clearer now than they were even a decade ago.4
Genetically predisposed allergic individuals have defects in
the population of suppressor T lymphocytes responsible for
modulating IgE responses to antigens. After the initial contact
of an allergen with the mucosa of such an individual, abnormal
amounts of allergen-specific IgE antibody are produced at the
mucosal surface and at the regional lymph nodes. This IgE has
high avidity, through its Fc portion, to Fc receptors on the
surface of mast cells in the mucosa. The antigen-specific IgE
antibodies, therefore, stick to the receptors on the surface of the
tissue mast cells and remain there for unusually long periods.
Excess locally produced IgE enters the circulation and binds to
mast cells at other tissue locations as well as to circulating
basophils. A subsequent encounter of the allergic individual
with the antigen to which he or she has become sensitized
results in antigen-binding by the antigen-specific IgE molecules
affixed to the surface of the tissue mast cells.
The simultaneous binding of the antigen to adjacent IgE
molecules on the mast cell surface results in a change in the
mast cell membrane and particularly in membrane-bound
adenyl cyclase (Fig. 9.1). The feature common to all known
mechanisms that trigger mast cell degranulation (including
degranulation stimulated by pharmacologic agents or
anaphylatoxins like C3a and C5a and antigen-specific IgEmediated degranulation) is calcium influx with subsequent
aggregation of tubulin into microtubules, which then
participate in the degranulation of vasoactive amines (see Fig.
9.1). In addition to the degranulation of the preformed
mediators such as histamine, induction of synthesis of newly
formed mediators from arachidonic acid also occurs with
triggering of mast cell degranulation (Table 9.2). The preformed
and newly synthesized mediators then produce the classic
clinical signs of a type I hypersensitivity reaction: wheal
(edema), flare (erythema), itch, and in many cases the subsequent, delayed appearance of the so-called late-phase reaction
characterized by subacute signs of inflammation.
Type I hypersensitivity has been postulated as a strategy to
avoid and remove multicellular parasite infections affecting the
respiratory and gastrointestinal systems.5 The consequence of
mast-cell degranulation is not only vasodilatation and increase
in production and release of exudative fluid, but also goblet cell
hyperplasia, synthesis of mucin of augmented viscosity and
increased peristaltic movement, which are demonstrated
successful mechanisms to eliminate parasitic nematodes.
Indeed one of the main symptoms in asthmatic patients, the
viscous and obstructive mucus secreted by the respiratory

these factors control the production of IgE-potentiating factor
and IgE-suppressor factor by the central helper T cell and, thus,
ultimately control the amount of IgE produced (see Fig. 9.2).
They probably do so through regulation of IgE B lymphocyte
proliferation and synthesis of IgE by these cells.
It has become increasingly clear that at least two subpopulations of mast cells exist. Connective tissue mast cells
(CTMCs) contain heparin as the major proteoglycan, produce
large amounts of prostaglandin D2 in response to stimulation,
and are independent of T cell-derived interleukins for their
maturation, development, and function. These cells stain
brilliantly with toluidine blue in formalin-fixed tissue sections.
Mucosal mast cells (MMCs) do not stain well with toluidine
blue. They are found primarily in the subepithelial mucosa in
gut and lung, contain chondroitin sulfate as the major
proteoglycan, manufacture leukotriene C4 as the predominant
arachidonic acid metabolite after stimulation, and are
dependent on IL-3 (and IL-4) for their maturation and proliferation. Interestingly, MMCs placed in culture with
fibroblasts rather than T cells transform to cells with the
characteristics of CTMCs. Disodium cromoglycate inhibits
histamine release from CTMCs but not from MMCs. Steroids
suppress the proliferation of MMCs, probably through
inhibition of IL-3 production.
CHAPTER 9
Mast Cell Subpopulations
Atopy Genetics and Immunology the Role of the

Environment
FIGURE 9.2. Diagrammatic display of IgE synthesis. Glycosylationenhancing factor, glycosylation-inhibiting factor, IgE-promoting factor,
IgE suppressor factor, and the helper and suppressor T lymphocytes
specific for regulation of IgE synthesis are shown.
epithelium, is thought to play a protective role in parasitic

infections (the parasite, because of the mucus, cannot
effectively penetrate the epithelial cells, which is essential to its
development). Type I hypersenstivity reactions would be hostdestructive only when they occur more intensely, improperly, or
as a result of a mistake in the perception of the existence of an
intruder, even though there is no true threat.
Control of IgE Synthesis

The Th2 subset of helper T cells bearing Fce receptors produce,
in addition to interleukin-4 (IL-4), IgE-binding factors after
stimulation by interleukins produced by antigen-specific helper
T cells activated by antigen-presenting cells and antigen. The
two known types of IgE-binding factor that can be produced are
IgE-potentiating factor and IgE-suppressor factor; both are
encoded by the same codon, and the functional differences are
created by posttranslational glycosylation. The glycosylation is
either enhanced or suppressed by cytokines derived from other
T cells. For example, glycosylation-inhibiting factor (identical to
migration inhibitory factor) is produced by antigen-specific
suppressor T cells. Glycosylation-enhancing factor is produced
by an Fc receptor helper T cell (Fig. 9.2). The relative levels of
Both genetic and environmental components are clearly

involved in the allergic response. Offspring of marriages in
which one parent is allergic have ~30% risk of being allergic,
and if both parents are allergic the risk to each child is greater
than 50%. At least three genetically linked mechanisms govern
the development of atopy1: general hyperresponsiveness,2
regulation of serum IgE levels,3 and sensitivity to specific
antigens. General hyperresponsiveness, defined as positive skin
reactions to a broad range of environmental allergens, is
associated HLA-B8/HLA-DW3 phenotype, and this general
hyperresponsiveness appears not to be IgE class specific. Total
serum IgE levels are also controlled genetically, and family
studies indicate that total IgE production is under genetic
control. Finally, experimental studies using low molecular
weight allergenic determinants disclose a strong association
between IgE responsiveness to such allergens and HLADR/DW2 type, whereas for at least some larger molecular
weight allergens, responsiveness is linked to HLA-DR/DW3. In
mice at least, gene regulation of IgE production occurs at several
levels, including regulation of antigen-specific,1 IgE-specific
suppressor T cells,2 manufacture of glycosylation-inhibiting
factor or of glycosylation-enhancing factor by helper T cells,3 at
the level of IL-4 regulation of class switching to IgE synthesis,
and at the level of IgE-binding factors such as IgE-potentiating
factor and IgE-suppressor factor.4
It is likely that the genetic architecture of the clinical
conditions of asthma or atopic keratoconjunctivitis differs.
However there are many common genes and pathways which
contribute to the onset, course, or severity of these related
entities. Certainly, well-known phenotypes associated with
them, such as bronchial hyperresponsiveness or the amount of
total and specific IgE, are influenced by the same genes. In 1996
the first genome-wide search for asthma and atopy
susceptibility loci was completed, and there have been multiple
publications on the genetic basis of these complex phenotypes.6
The most frequent loci reported as associated with asthma or
atopy phenotypes are the following genes: IL4, IL13, ADRB2,
TNF, HLA-DRB1, FCER1B, IL4RA, CD14, HLA-DQB1, and
83
SECTION 2
IMMUNOLOGY
ADAM33.7 However no one gene will be the atopy gene in all
populations, which reflects the tremendous complexity of these
pathologies in terms of genetic predisposition and the modest
effects of these genes on risk.
The environment plays a major role in whether or not a
genetically predisposed individual expresses major clinical
manifestations of atopy. The dose of allergens to which the
individual is exposed is a critical determinant of whether or not
clinical expression of an allergic response develops. Less well
recognized, however, is the fact that the general overall quality
of the air in an individuals environment plays a major role in
whether clinical expression of allergic responses to allergens to
which the individual is sensitive does or does not develop. It has
become unmistakably clear that as the general quality of the air
in urban environments has deteriorated and as the air has
become more polluted, the prevalence in the population of overt
atopic clinical manifestations has increased dramatically. On a
global level, the immediate environment in which an individual
finds himself much of the time, the home, plays an important
part in the expression of allergic disease. Allergically predisposed persons, at least one member of whose household smokes
cigarettes, have enhanced sensitivity to allergens such as house
dust, mites, and molds, among others. It is probably also true
that the overall health and nutritional status of an individual
influence the likelihood of that person developing a clinically
obvious allergy.
Evidence linking stress to the expression of conditions such
as atopy is still growing. The reported influence of stress on
neuroimmunoregulation and oxidative stress pathways may
interact with the hypersensitivity to environmental conditions
as previously described, playing a crucial role in the genesis
of the characteristic clinical manifestations.8 Both roles,
genegene and geneenviroment interactions, are important in
determining susceptibility. Further studies to determine risk for
specific patients will have to consider the influence of the genes
under a certain environmental context, as much as possible, to
clarify the degree of responsibility of each factor.
Diagnosis of Type I Reactions

The definite diagnosis of type I hypersensitivity reactions
requires the passive transfer of the reaction via a method known
as the PrausnitzKustner reaction. Intradermal injection of the
serum of a patient suspected of having a type I hypersensitivitymediated problem into the skin of a volunteer is followed by
injection of varying dilutions of the presumed offending antigen
at the same intradermal sites as the patients serum injection.
A positive PrausnitzKustner reaction occurs when local flare
and wheal formation follows the injection of the antigen. This
method for proving type I reactions is not used clinically;
therefore, diagnosis of type I mechanisms contributing to a
patients inflammatory disorder is always based on a collection
of circumstantial evidence that strongly supports the hypothesis
of a type I reaction. A typical history (e.g., of a family history of
allergy or personal history of eczema, hay fever, asthma, or
urticaria) elicitation of allergic symptoms following exposure to
suspected allergens involves itching as a prominent symptom,
elevated IgE levels in serum or other body fluids, and blood
or tissue eosinophilia. Chapter 11 covers these points in
general, as well as the importance of the histopathologic
characteristics of conjunctival biopsy tissue, in particular in the
evaluation of patients with chronic cicatrizing conjunctivitis.
Therapy for Type I Reactions
84
Therapy for type I reactions must include scrupulous avoidance

of the offending antigen. This is not easy, and it is a component
of proper treatment that is often neglected by the patient and
the physician alike. It is crucial, however, for a patient with an
incurable disease such as atopy to recognize that throughout a

lifetime he or she will slowly sustain cumulative permanent
damage to structures affected by atopic responses (e.g., lung,
eye) if he or she is subjected to repetitive triggering of the
allergic response. Pharmacologic approaches to this disorder can
never truly succeed for careless patients who neglect their
responsibility to avoid allergens. A careful environmental
history is, therefore, a critical ingredient in history-taking, and
convincing education of the patient and family alike is an
essential and central ingredient in the care plan.
A careful environmental history and meticulous attention to
environmental details can make the difference between relative
stability and progressive inflammatory attacks that ultimately
produce blindness. Elimination of pets, carpeting, feather pillows,
quilts, and wool blankets and installation of air-conditioning
and air-filtering systems are therapeutic strategies that should
not be overlooked.9
One of the most important advances in the care of patients
with type I disease during the past two decades has been the
development of mast cell-stabilizing agents. Disodium
cromoglycate, sodium nedocromil, and lodoxamide are three
such agents. Topical administration is both safe and effective in
the care of patients with allergic eye disease.10,11 This
therapeutic approach is to be strongly recommended and is very
much favored over the use of competitive H1 antihistamines.
Clearly, if the mast cells can be prevented from degranulating,
the therapeutic effect of such degranulation-inhibiting agents
would be expected to be vastly superior to that of antihistamines simply by virtue of preventing liberation of an entire
panoply of mediators from the mast cell rather than competitive
inhibition of one such mediator, histamine.
Histamine action-inhibition by H1 antihistamines can be
effective in patients with ocular allergy provided the drugs are
administered systemically. The efficacy of such agents when
given topically is marginal at best, and long-term use can
result in the development of sensitivity to ingredients in the
preparations. The consistent use of systemic antihistamines,
however, can contribute significantly to long-term stability,
particularly of the newer noncompetitive antihistamines
such as astemizole. Additionally, slow escalation of the
amount of hydroxyzine used in the care of atopic patients
can help to interrupt the itchscratchitch psychoneurotic
component that often accompanies eczema and atopic
blepharokeratoconjunctivitis.
Generalized suppression of inflammation, through use of
topical corticosteroids, is commonly used for treatment of type
I ocular hypersensitivity reactions, and this is appropriate for
acute breakthrough attacks of inflammation. It is, however,
completely inappropriate for long-term care. Corticosteroids
have a direct effect on all inflammatory cells, including eosinophils, mast cells, and basophils. They are extremely effective,
but the risks of chronic topical steroid use are considerable and
unavoidable, thus chronic use is discouraged.
Although desensitization immunotherapy can be an important additional component to the therapeutic plan for a patient
with type I hypersensitivity, it is difficult to perform properly.
The first task, of course, is to document to which allergens the
patient is sensitive. The second task is to construct a serum
containing ideal proportions of the allergens that induce the
production of IgG-blocking antibody and stimulate the
generation of antigen-specific suppressor T cells. For reasons
that are not clear, the initial concentration of allergens in such
a preparation for use in a patient with ocular manifestations of
atopy must often be considerably lower than the initial
concentrations usually used when caring for a person with
extraocular allergic problems. If the typical starting concentrations for nonocular allergies are employed frequently, a
TABLE 9.3. Therapy of the Atopic Patient

Environmental control
Mast cell stabilizers
Systemic antihistamines
Desensitization immunotherapy
Plasmapheresis
Intravenous gamma globulin
Cyclosporine (systemic and topical)
CHAPTER 9
Topical steroids (for acute intervention only)
Psychiatric intervention for the patient and family
dramatic exacerbation of ocular inflammation immediately

follows the first injection of the desensitizing preparation.
Plasmapheresis is an adjunctive therapeutic maneuver that
can make a substantial difference in the care of patients
with atopy, high levels of serum IgE, and documented
Staphlyococcus-binding antibodies.9 This therapeutic technique
is expensive, is not curative, and must be performed at highly
specialized centers, approximately three times each week,
indefinitely. It is also clear, from our experience, that the
aggressiveness of the plasmapheresis must be greater than that
typically employed by many pheresis centers. Three to four
plasma exchanges per pheresis session typically are required to
achieve therapeutic effect for an atopic person.
Intravenous or intramuscular gamma globulin injections
may also benefit selected atopic patients. It has been recognized
that, through mechanisms that are not yet clear, gamma
globulin therapy involves much more than simple passive
immunization through adoptive transfer of antibody
molecules. In fact, immunoglobulin therapy has a pronounced
immunomodulatory effect, and it is because of this action that
such therapy is now recognized and approved as effective
therapy for idiopathic thrombocytopenic purpura.12 The use of
gamma globulin therapy is also being explored for other
autoimmune diseases, including systemic lupus erythematosus
and atopic disease.
Cyclosporine is being tested in patients with certain atopic
diseases. Preliminary evidence suggests that topical cyclosporine can have some beneficial effect on patients with atopic
keratoconjunctivitis and vernal keratoconjunctivitis.13
Furthermore, in selected desperate cases of blinding atopic
keratoconjunctivitis, we have demonstrated that systemic
cyclosporine can be a pivotal component of the multimodality
approach to the care of these complex problems.9
The calcineurin-inhibitors, such as Pimecrolimus or
Tacrolimus, were introduced in the early 2000s as alternative
topical treatments, acting more selectively and providing
certain advantages over corticosteroids. These agents have
demonstrated efficacy in the management of patients with
atopy-related diseases, such as atopic dermatitis or severe atopic
keratoconjunctivitis.14,15 These agents appear to offer the
potency of a corticosteroid without its adverse side effects.
Tacrolimus, also known as FK506, is a potent immunosuppressive agent (close relative of cyclosporine in terms of
action mechanism, but up to 100 times more potent) that
has been used orally since 1994 to prevent allograft rejection in
liver and kidney transplant recipients. Its systemic use may
also be considered in selected patients with severe atopic
keratoconjunctivitis.
Finally, appropriate psychiatric care may be (and usually is)
indicated in patients with severe atopy (and family members). It
FIGURE 9-3. Type II hypersensitivity. (a) A sensitized cell with two

antibodies specific for antigenic determinants on the cell surface has
attached to the target cell. C1q, C1r, and C1s complement
components have begun the sequence that will result in the classical
cascade of complement-factor binding. (b) The complement cascade
has progressed to the point of C5 binding. Note that two
anaphylatoxin and chemotactic split products, C3a and C5a, have
been generated, and a neutrophil is being attracted to the site by
virtue of the generation of these two chemotactic moieties. (c) The
complement cascade is complete, with the result that a pore has been
opened in the target cell membrane, and osmotic lysis is the nearly
instantaneous result. (d) A variant of the type II hypersensitivity
reaction is the antibody-dependent cellular cytotoxicity (ADCC)
reaction. Target-specific antibody has attached to the target cell
membrane, and the Fc receptor on a neutrophil, a macrophage, or a
killer (K) cell is attaching to that membrane-affixed antibody. The result
will be lysis of the target cell.
is not hyperbole to state that in most cases, patients with severe

atopic disease and the family members with whom they live
demonstrate substantial psychopathology and destructive
patterns of interpersonal behavior. The degree to which these
families exhibit self-destructive, passiveaggressive, and
sabotaging behaviors is often astonishing. Productive engagement in psychiatric care is often difficult to achieve, but it
can be extremely rewarding when accomplished successfully.
Table 9.3 summarizes the components of a multifactorial
approach to the care of atopic patients.
TYPE II HYPERSENSITIVITY REACTIONS

Type II reactions require the participation of complement-fixing
antibodies (IgG1, IgG3, or IgM) and complement. The
antibodies are directed against antigens on the surface of
specific cells (i.e., endogenous antigens). The damage caused by
type II hypersensitivity reactions, therefore, is localized to the
particular target cell or tissue. The mediators of the tissue
damage in type II reactions include complement as well as
recruited macrophages and other leukocytes that liberate their
enzymes. The mechanism of tissue damage involves antibodybinding to the cell membrane with resultant cell membrane
lysis or facilitation of phagocytosis, macrophage and neutrophil
cell-mediated damage (Fig. 9.3ac), and killer cell damage to
target tissue through antibody-dependent cell-mediated
cytotoxicity (ADCC) reaction (see Fig. 9.3d). It is important to
remember (particularly in the case of type II hypersensitivity
reactions that do not result in specific target cell lysis through
the complement cascade with eventual osmotic lysis) that
neutrophils are prominent effectors of target cell damage.
85
SECTION 2
IMMUNOLOGY
Neutrophil adherence, oxygen metabolism, lysosomal enzyme
release, and phagocytosis are tremendously upregulated by
IgGC3 complexes and by the activated split product of C5a. As
mentioned in the description of type I hypersensitivity
reactions, mast cells also participate in nonallergic inflammatory reactions, and type II hypersensitivity reactions provide
an excellent example of this. The complement split products
C3a and C5a both produce mast cell activation and
degranulation. The result is the liberation of preformed
vasoactive amines and upregulation of membrane synthesis of
leukotriene B4, the most potent (and also other cytokines (e.g.,
TNF-a)) known chemoattractant for neutrophils, even more
potent than IL-8/RANTES, eosinophil chemotactic factor,
and other arachidonic acid metabolites. Neutrophils and
macrophages attracted to this site of complement-fixing IgG or
IgM in a type II hypersensitivity reaction cannot phagocytose
entire cells and target tissues, and thus liberate their proteolytic
and collagenolytic enzymes and cytokines in frustrated
phagocytosis. It is through this liberation of tissue-digestive
enzymes that the target tissue is damaged. Direct target cell
damage (as opposed to innocent bystander damage caused by
liberation of neutrophil and macrophage enzymes) in type II
hypersensitivity reactions may be mediated by killer (K) cells
through the antibody-dependent cytotoxicity reaction. In fact,
definitive diagnosis of type II reactions requires the demonstration of fixed antitissue antibodies at the disease site as well
as a demonstration of killer cell activity in vitro against the
tissue. No ocular disease has been definitively proved to
represent a type II reaction, but several candidates, including
ocular cicatricial pemphigoid, exist.
This type II hypersensitivity reaction has been postulated as
a tactic to deal with small extracellular organisms.5 The first
step, the interaction between antibodies and antigens, results in
opsonization of extracellular microbes resistant to phagocytosis.
The second step, the liberation of neutrophil chemoattractants,
is designed to be a magnet for PMNs to the site of the
inflammation. As in the type I reaction, this would be
pathologic only if it occurs in other circumstances different
from a response to such kind of infections.
The classic human autoimmune type II hypersensitivity
disease is Goodpastures syndrome. Many believe ocular
cicatricial pemphigoid is analogous (in mechanism at least) to
Goodpastures syndrome, in which complement-fixing antibody
directed against a glycoprotein of the glomerular basement
membrane fixes to the glomerular basement membrane. This
action causes subsequent damage to the membrane by
proteolytic and collagenolytic enzymes liberated by phagocytic
cells, including macrophages and neutrophils.
Therapy for Type II Reactions

Therapy for type II reactions is extremely difficult, and
immunosuppressive chemotherapy has, in general, been the
mainstay of treatment. Experience with ocular cicatricial
pemphigoid has been especially gratifying in this regard.1618
Progressive cicatricial pemphigoid affecting the conjunctiva
was, eventually, almost universally blinding before the advent of
systemic immunosuppressive chemotherapy for this condition.
With such therapy now, however, 90% of cases of the disease are
arrested and vision is preserved.19
TYPE III HYPERSENSITIVITY REACTIONS
86
Type III reactions, or immune complex diseases, require, like

type II hypersensitivity reactions, participation of complementfixing antibodies (IgG1, IgG3, or IgM). The antigens
participating in such reactions may be soluble diffusible
antigens, microbes, drugs, or autologous antigens. Microbes
FIGURE 9.4. Type III hypersensitivity reaction. Circulating immune

complexes (shown here as triangle-shaped moieties in the vascular
lumen) percolate between vascular endothelial cells but become
trapped at the vascular endothelial basement membrane. Neutrophils
and other phagocytic cells are attracted to this site of immune
complex deposition. These phagocytic cells liberate their proteolytic
and collagenolytic enzymes and damage not only the vessel but also
the surrounding tissue.
that cause such diseases are usually those that cause persistent
infections in which not only the infected organ but also the
kidneys are affected by the immune complex-stimulated
inflammation. Autoimmuneimmune complex diseases are the
best known of these hypersensitivity reactions: the classic
collagen vascular diseases and StevensJohnson syndrome.
Kidney, skin, joints, arteries, and eyes are frequently affected in
these disorders. Mediators of the tissue damage include
antigenantibodycomplement complexes and the proteolytic
and collagenolytic enzymes from phagocytes such as
macrophages and neutrophils. As with type II reactions, the
C3a and C5a split products of complement exert potent
chemotactic activity for the phagocytes and also activate mast
cells, which through degranulation of their vasoactive amines,
TNF-a increase vascular permeability and enhance emigration
of such phagocytic cells. It is again through frustrated
phagocytosis that the neutrophils and macrophages liberate
their tissue-damaging enzymes (Fig. 9.4).
Arthus reaction, a special form of type III hypersensitivity, is
mentioned for completeness. Antigen injected into the skin of
an animal or individual previously sensitized with the same
antigen, and with circulating antibodies against that antibody,
results in an edematous, hemorrhagic, and eventually necrotic
lesion of the skin. A passive Arthus reaction can also be created
if intravenous injection of antibody into a normal host recipient
is followed by intradermal injection of the antigen. An
accumulation of neutrophils develops in the capillaries and
venule walls after deposition of antigen, antibody, and
complement in the vessel walls.
Immune complexes form in all of us as a normal consequence of our immunologic housekeeping. Usually, however,
Therapy for Type III Reactions

Therapy for type III reactions consists predominantly of large
doses of corticosteroids, of immunosuppressive chemotherapeutic agents, or both. Cytotoxic immunosuppressive
chemotherapy may or may not be necessary to save both the
sight and the life of a patient with Behets disease, but it is
TABLE 9.4. Types of Delayed Hypersensitivity Reactions

Reaction Type
Example
Peak Reaction
Tuberculin contact
Tuberculin skin test
4872 h
Contact
Drug contact
hypersensitivity
4872 h
Granulomatous
Leprosy
14 days
JonesMote
Cutaneous basophil
hypersensitivity
24 h
CHAPTER 9
these immune complexes are continually removed from the

circulation. In humans, the preeminent immune complexscavenging system is the red blood cells, which have a receptor
(CR1) for the C3b and C4b components of complement. This
receptor binds immune complexes that contain complement,
and the membrane-bound complexes are removed by fixedtissue macrophages and Kupffer cells as the red blood cells pass
through the liver. Other components of the reticuloendothelial
system, including the spleen and the lung, also remove
circulating immune complexes. Small immune complexes may
escape binding and removal, and not surprisingly, smaller
immune complexes are principally responsible for immune
complex-mediated hypersensitivity reactions. It is also true that
IgA complexes (as opposed to IgG or IgM complexes) do not
bind well to red blood cells. They are found in the lung, brain,
and kidney rather than in the reticuloendothelial system.
The factors that govern whether or not immune complexes
are deposited into tissue (and if so, where) are complex and
rather incompletely understood. It is clear that the size of the
immune complex plays a role in tissue deposition. It is also
clear that increased vascular permeability at a site of immune
system activity or inflammation is a major governor of whether
or not immune complexes are deposited in that tissue.
Additionally, it is clear that immune complex deposition is
more likely to occur at sites of vascular trauma; this includes
trauma associated with the normal hemodynamics of a
particular site, such as the relatively high pressure inside
capillaries and kidneys, the turbulence associated with
bifurcations of vessels, and obviously at sites of artificial trauma
as well. Excellent examples of the latter include the areas of
trauma in the fingers, toes, and elbows of patients with
rheumatoid arthritis, where subsequently vasculitic lesions and
rheumatoid nodules form, and in the surgically traumatized
eyes of patients with rheumatoid arthritis or Wegener s
granulomatosis, where immune complexes are deposited
subsequently and necrotizing scleritis develops.20 It is likely
that addressing or other attachment factors in a local tissue play
a role in the homing of a particular immune complex.
Antibody class and immune complex size are also important
determinants of immune complex localization at a particular
site, as is the type of the basement membrane itself.
Type III hypersensitivity reactions have been postulated as a
strategy to prevent further injury in the viremic phase of viral
infections.5 The potential harmful effect of this reaction would
be the one described by Gell and Coombs. But under more
physiological conditions, the results are probably beneficial to
the host. In fact, the binding of excess complement to
preformed antigenantibody complexes seems to result in their
disaggregation into smaller entities that no longer bind more
complement. Furthermore, these complexes do not trigger the
lytic components of complement and do not liberate
anaphylotoxins, and can be ingested and later eliminated by the
reticuloendothelial system. This reaction may have a hostprotective response and is possibly the best one to eliminate
circulating viral particles. However, when C3 falls under critical
levels, this mechanism fails, obstructing this degradation of
antigenantibody complexes into smaller and soluble fragments
which then deposit in certain areas of the host: this is why, for
example, renal disease in systemic lupus erythematosus (SLE) is
inversely related to complement levels.
categorically required to save the life of a patient with either

polyarteritis nodosa21 or Wegeners granulomatosis.22 In the
case of rheumatoid arthritis-associated vasculitis affecting the
eye, it is likely that systemic immunosuppression will also be
required if death from a lethal extraarticular, extraocular,
vasculitic event is to be prevented.23
INJURY MEDIATED BY CELLS

TYPE IV HYPERSENSITIVITY REACTIONS:
IMMUNE-MEDIATED INJURY DUE TO
EFFECTOR T CELLS
The original classification of immunopathogenic mechanisms
arose in an era when considerably more was known about
antibody molecules and serology than about T cells and cellular
immunity. Out of this lack of knowledge, T cell-mediated
mechanisms were relegated to the type IV category, and all
manner of responses were unwittingly grouped together (Table
9.4).24 We now know that T cells capable of causing immunebased injury exist in at least three functionally distinct
phenotypes: cytotoxic T cells (typically CD8+) and two
populations of helper T cells (typically CD4+) (Fig. 9.5). Since
cytotoxic T lymphocytes (CTLs) were discovered well after the
original Gell and Coombs classification, they were, therefore,
never anticipated in that classification system. As mentioned
previously, CD4+ T cells can adopt one of two polar positions
with regard to their lymphokine secretions (IL-12 induces Th1
cells, and IL-10 induces Th2 cells).25 Th1 cells secrete IL-2,
IFN-g, and lymphotoxin, whereas Th2 cells were identified in
the 1940s and 1950s as the initiators of delayed hypersensitivity reaction by secretion of cytokines such as IL-4, IL-5,
and IL-6. The latter cells, in addition to providing helper factors
that promote IgE production, also mediate tissue inflammation,
albeit of a somewhat different type than Th1 cells.
Immunopathogenic T Cells
CTLs exhibit exquisite antigen specificity in their recognition of
target cells, and the extent of injury that CTLs effect is usually
limited to target cells bearing the relevant instigating antigens.
Therefore, if a CTL causes tissue injury, it is because host cells
express an antigen encoded by an invading pathogen, an antigen
for which the Tcr on the CTL is highly specific. Delivery of a
cytolytic signal eliminates hapless host cells, and in so doing
aborts the intracellular infection. Assuming that the infected
host cell is one of many and can thus be spared (e.g., epidermal
keratinocytes), there may be little or no physiologic consequence of this CTL-mediated loss of host cells. However, if the
infected cell is strategic, limited in number, or cannot be replaced
by regeneration (e.g., neurons, corneal endothelial cells), then
the immunopathogenic consequences may be severe.
CD4+ effector cells also exhibit exquisite specificity in
recognition of target antigens. However, the extent of injury
87
SECTION 2
IMMUNOLOGY
FIGURE 9.5. Type IV hypersensitivity reaction. DTH (CD4) T

lymphocytes and cytotoxic (CD8 and CD4) T lymphocytes directly
attack the target cell or the organism that is the target of the type IV
hypersensitivity reaction. Surrogate effector cells are also recruited
through the liberation of cytokines. The most notable surrogate or
additional effector cell is the macrophage or tissue histiocyte. If the
reaction becomes chronic, certain cytokines or signals from
mononuclear cells result in the typical transformation of some
histiocytes into epithelioid cells, and the fusion of multiple epithelioid
cells produces the classic multinucleated giant cell.
that these cells can effect is diffuse and is not limited to cells
bearing the target antigen. CD4+ effector cells secrete cytokines
that possess no antigen specificity in their own right. Instead,
these molecules indiscriminately recruit and activate macrophages, natural killer cells, eosinophils, and other mobile cells
that form the nonspecific host defense network. It is this
defense mechanism that leads to eradication and elimination of
the offending pathogen. In other words, CD4+ effector cells
protect by identifying the pathogen antigenically, but they cause
the elimination of the pathogen by enlisting the aid of other
cells. The ability of CD4+ effector cells to orchestrate this
multicellular response rests with the capacity of these cells to
secrete proinflammatory cytokines to arm inflammatory cells
with the ability to kill. Once armed, these mindless assassins
mediate inflammation in a nonspecific manner that leads often,
if not inevitably, to innocent bystander injury to surrounding
tissues. For an organ that can scarcely tolerate inflammation of
even the lowest amount, such as the eye, innocent bystander
injury is a formidable threat to vision.
Autoimmune T Cells
88
The foregoing discussion addresses immunopathogenic injury

due to T cells that develops among host tissues invaded by
pathogenic organisms. However, there is another dimension to
immunopathology. T cells can sometimes make a mistake and
mount an immune attack on host tissues simply because those
tissue cells express self molecules (i.e., autoantigens). Although
an enormous amount of experimental and clinical literature is
devoted to autoimmunity and autoimmune diseases, very little
is known in a factual sense that enables us to understand this
curious phenomenon. What seems clear is that T cells with
receptors that recognize self antigens, as well as B cells bearing
surface antibody receptors that recognize self antigens, exist
under normal conditions.24 Moreover, there are examples of T
and B cells with self -recognizing receptors that become
activated in putatively normal individuals. Thus, immunologists have learned to distinguish an autoimmune response
(not necessarily pathologic) from an autoimmune disease.
Whereas all autoimmune diseases arise in a setting where an
autoimmune response has been initiated, we understand little
about what causes the latter to evolve into the former. Whatever
the pathogenesis, autoimmune disease results when effector T
cells (or antibodies) recognize autoantigens in a fashion that
triggers a destructive immune response.26,27
The pathogenesis of autoimmunity is probably related to a
complex phenomenon called cripticity.28 This is directly
connected with the hierarchy of antigenic determinants within
self-antigens and is a product of the extent of proper presentation of the antigen and the affinity of the T-cell receptor. The
well-processed and -presented determinants constitute a
dominant self , whereas the inadequately processed and/or
presented determinants will be invisible to T cells and comprise
a cryptic self , which plays a crucial role in the genesis of
autoimmunity. A similar hierarchy is established in the thymus
with both positive and negative selections. This would explain
why experimental model systems show that T cells against
dominant self-determinants get positive tolerance, whereas
those potentially directed against cryptic epitopes escape
tolerance induction. Under normal physiological conditions,
the cryptic epitopes of a native antigen are unproductively
processed and presented and there is no threat of initiation of
an anti-self immune response by such epitopes. However, under
inflammatory and other specific conditions, upregulation of
antigen-processing events can lead to improved presentation of
the previously cryptic epitopes by the antigen-presenting cells,
that can lead to priming cryptic-epitope specific T cells.
The eye consists of unique cells bearing unique molecules.
Moreover, the internal compartments of the eye exist behind a
bloodtissue barrier. The very uniqueness of ocular molecules,
and their presumed sequestration from the systemic immune
system, has provoked immunologists to speculate that ocular
autoimmunity arises when, via trauma or infection, eye-specific
antigens are revealed to the immune system. Sympathetic
ophthalmia is a disease that almost fits this scenario perfectly.
Trauma to one eye, with attendant disruption of the
bloodocular barrier and spillage of ocular tissues and
molecules, leads to a systemic immune response that is specific
to the eye. This response is directed not only at the traumatized
eye but also at its putatively normal fellow eye. However, even
in sympathetic ophthalmia, not every case of ocular trauma
leads to this outcome; in fact, only in a few cases does this type
of injury produce inflammation in the undamaged eye.
Suspicion is high that polymorphic genetic factors may be
responsible for determining who will, and who will not, develop
sympathetic ophthalmia following ocular injury. However,
environmental factors may also participate.
Range of Hypersensitivity Reactions Mediated by

T Cells
Because a wealth of new information about T cell-mediated
immunopathology has accrued within the past decade, our
ideas about the range of hypersensitivity reactions that can be
mediated by T cells have expanded. But, as yet, any attempt to
classify these reactions must necessarily be incomplete. In the
past, four types of delayed hypersensitivity reactions were
described:1 tuberculin,2 contact hypersensitivity,3 granulomatous, and JonesMote.4 Delayed hypersensitivity reactions of
these types were believed to be caused by IFN-g-producing CD4+
T cells and to participate in numerous ocular inflammatory
disorders, ranging from allergic keratoconjunctivitis, through
Wegeners granulomatosis, to drug contact hypersensitivity.
Based on recent knowledge concerning other types of effector T
cells, this list must be expanded to include cytotoxic T cells,

and proinflammatory, but not IFN-g-secreting, Th2 type cells,
such as the cells that are believed to cause corneal clouding in
river blindness.29
Additionally, graft versus host disease is a result of cellular
immunity and is an example of a delayed T-helper cell response.
A rejected allograft has a similar histological appearance to a
tuberculin reaction, and rejection is mediated by T cells with an
important role for the NK cells.30 The histopathological
findings are mononuclear cell infiltration and tissue
destruction. The CD8+ T cells are the primary cells inducing
the lesions, although a minor role for CD4+ has been described.
As in the other hypersensitivity reactions, this one is a clear
example of an anomaly in a well-organized cellular response to
pathogens. T cells represent the best choice against intracellular
infections, usually viral, in order to prevent further damage and
offspring of the infective agent.5 There appears to be a
connection between antecedent viral infection, susceptible
MHC class II alleles, and the inception of certain diseases
included in this range. The protective mechanisms to the host
(such as control of cell proliferation by cytokines or induction of
apoptosis of target cells by different ways) are the same as those
which cause injurious effects to the host.
Herpes Simplex Keratitis as an Example of T CellMediated Ocular Inflammatory Disease

Infections of the eye with herpes simplex virus are significant
causes of morbidity and vision loss in developed countries.
Although direct viral toxicity is damaging to the eye, the
majority of intractable herpes infections appear to be
immunopathogenic in origin. That is, the immune response to
antigens expressed during a herpes infection leads to tissue
injury and decompensation, even though the virus itself is
responsible for little pathology directly. Herpes stromal keratitis
(HSK) is representative of this type of disorder.31
Numerous experimental model systems have been developed
in an effort to understand the pathogenesis of HSK. Perhaps the
most informative studies have been conducted in laboratory
mice. Evidence from these model systems indicates that T cells
are central to the corneal pathology observed in HSK.31 At least
four different pathogenic mechanisms have been discovered,
each of which alone can generate stromal keratitis. Genetic
factors of the host seem to play a crucial role in dictating which
mechanism will predominate. First, HSV-specific cytotoxic T
cells can cause HSK and do so in several strains of mice.
Second, HSV-specific T cells of the Th1 type, which secrete
IFN-g and mediate delayed hypersensitivity, also cause HSK,
but in genetically different strains of mice. Third, HSV-specific
T cells of the Th2 type, that secrete IL-4 and IL-10, correlate
with HSK in a yet different strain of mice. Fourth, T cells have

been found in association with HSK that recognize an antigen
uniquely expressed in the cornea. The evidence suggests that
this corneal antigen is unmasked during a corneal infection
with HSV, and an autoimmune response is evoked in which the
cornea becomes the target of the attack.
Only time will tell whether similar immunopathogenic
mechanisms will prove to be responsible for HSK in humans,
but the likelihood is very great that this will be the case.
Furthermore, it is instructive to emphasize that quite different
pathologic T cells can be involved in ocular pathology, which
implies that it will be necessary to devise different therapies in
order to meet the challenge of preventing immunopathogenic
injury from proceeding to blindness.
CHAPTER 9
SUMMARY
Faced with a patient who is experiencing extraocular or
intraocular inflammation, the thoughtful ophthalmologist will
try, to the best of his or her ability, to diagnose the specific cause
of the inflammation, or at the very least to investigate the
problem so that the mechanisms responsible for the
inflammation are understood as completely as possible. Armed
with this knowledge, the ophthalmologist is then prepared to
formulate an appropriate therapeutic plan rather than to
indiscriminately prescribe corticosteroids. It is clear as we move
into the twenty-first century that the past four decades of
relative neglect of ocular immunology by mainstream
ophthalmic practitioners is coming to an end. Most
ophthalmologists are no longer satisfied to cultivate practices
devoted exclusively to the tissue carpentry of cataract surgery
or even to a broad-based ophthalmic practice that includes
medical ophthalmology but is restricted to problems related
exclusively to the eye (e.g., glaucoma) and divorced from the eye
as an organ in which systemic disease is often manifested. More
ophthalmologists than ever before are demanding the
continuing education they need to satisfy intellectual curiosity
and to prepare for modern care of the total patient when a
patient presents with an ocular manifestation of a systemic
disease. It is to these doctors that this chapter is directed.
The eye can be affected by any of the immune hypersensitivity
reactions, and understanding the mechanism of a particular
patients inflammatory problem lays the ground-work for
correct treatment. In the course of the average ophthalmologists working life, the diagnostic pursuit of mechanistic
understanding will also result in a substantial number of
instances when the ophthalmologist has been responsible for
diagnosing a disease that, if left undiagnosed, would have
been fatal.
REFERENCES
1. Kamradt T, Gggel R, Erb KJ: Induction,
exacerbation and inhibition of allergic and
autoimmune diseases by infection. Trends
Immunol 2005; 26:260.
2. Bach JF: The effect of infections on
susceptibility to autoimmune and allergic
diseases. N Engl J Med 2002; 347:911.
3. Cocoa AF, Cooke RA: On the classification
of the phenomena of hypersensitiveness.
J Immunol 1923; 8:163.
4. von Pirquet C: Allergie. Munch Med
Wochenschr 1906; 53:1457.
5. Rajan TV: The Gell-Coombs classification
of hypersensitivity reactions, a
reinterpretation. Trends Immunol 2003;
24:376.
6. Daniel SE, Bhattacharrya S, James A, et al:

A genome-wide search for quantitative trait
loci underlying asthma. Nature 1996;
383:247.
7. Ober C, Hoffjan S: Asthma genetics 2006,
the long and winding road to gene
discovery. Genes Immun 2006; 7:95.
8. Wright RJ: Stress and atopic disorders.
J Allergy Clin Immunol 2005; 116:1301.
9. Foster CS, Calonge M: Atopic
keratoconjunctivitis. Ophthalmology 1990;
97:992.
10. Foster CS, Duncan J: Randomized clinical
trial of disodium cromoglycate therapy in
vernal keratoconjunctivitis. Am J
Ophthalmol 1980; 90:175.
11. Foster CS: Evaluation of topical cromolyn

sodium in the treatment of vernal
95:194.
12. Bussel JB, Kimberly RP, Inamen RD, et al:
Intravenous gamma globulin treatment of
chronic idiopathic cytopenic purpura.
Blood 1983; 62:480.
13. Bleik JH, Tabbara KS: Topical cyclosporine
in vernal keratoconjunctivitis.
14. Hultsch T, Kapp A, Spergel J:
Immunomodulation and safety of topical
calcineurin inhibitors for the treatment
of atopic dermatitis. Dermatology 2005;
211:174.
89
SECTION 2
IMMUNOLOGY
90
15. Mark J, Kaufman HE, Insler M: Topical

tacrolimus for the treatment of refractory
anterior segment inflammatory disorders.
Cornea 2005; 24:417.
16. Foster CS: Cicatricial pemphigoid. Thesis
of the American Ophthalmological Society.
Trans Am Ophthalmol Soc 1986; 84:527.
17. Foster CS, Wilson LA, Ekins MB:
Immunosuppressive therapy for
progressive ocular cicatricial pemphigoid.
18. Tauber J, Sainz de la Maza M, Foster CS:
Systemic chemotherapy for ocular
cicatricial pemphigoid. Cornea 1991;
10:185.
19. Neumann R, Tauber J, Foster CS:
Remission and recurrence after withdrawal
of therapy for ocular cicatricial pemphigoid.
20. Sainz de la Maza M, Foster CS: Necrotizing
scleritis after ocular surgery: a clinical
pathologic study. Ophthalmology 1991;
98:1720.
21. Leib ES, Restivo C, Paulus AT:

Immunosuppressive and corticosteroid
therapy of polyarteritis nodosa. Am J Med
1979; 67:941.
22. Wolf SM, Fauci AS, Horn RG, Dale DC:
Wegeners granulomatosis. Ann Intern Med
1974; 81:513.
23. Foster CS, Forstot SL, Wilson LA: Mortality
rate in rheumatoid arthritis patients
developing necrotizing scleritis or
peripheral ulcerative keratitis.
24. Janeway CA Jr, Travers P, eds.
Immunobiology: the immune system in
health and disease. 3rd edn. New York:
Current Biology/Garland Publishing; 1997.
25. Mosmann TR, Coffman RL: TH1 and TH2
cells: different patterns of lymphokine
secretion lead to different functional
properties. Ann Rev Immunol 1989; 7:145.
26. Steinman L: Escape from horror
autotoxicus: pathogenesis and treatment
of autoimmune disease. Cell 1995; 80:7.
27. Tan EM: Autoantibodies in pathology and

cell biology. Cell 1991; 67:841.
28. Mougdil KD, Sercarz EE: Understanding
cripticity is the key to revealing the
pathogenesis of autoimmunity. Trends
Immunol 2005; 26:355.
29. Pearlman E, Lass HJ, Bardenstein DS,
et al: Interleukin 4 and T helper type 2 cells
are required for development of
experimental onchocercal keratitis (river
blindness). J Exp Med 1995; 182:931.
30. Black CA: Delayed hypersensitivity, current
theories with an historic perspective.
Dermatol Online J 2005; 5:7.
31. Streilein JW, Dana MR, Ksander BR:
Immunity causing blindness: five different
paths to herpes stromal keratitis. Immunol
Today 1997; 18:443.
CHAPTER
10
Regulation of Immune Responses

Reza Dana and C. Stephen Foster
Immunization with an antigen leads, under normal circumstances, to a robust immune response in which effector T cells
and antibodies are produced with specicity for the initiating
antigen (see Chapter 5). The purpose of these effectors is to
recognize and combine with antigen (e.g., on an invading
pathogen) in such a manner that the antigen (pathogen) and/or
infected cell are eliminated. Once the antigen has been eliminated, there is little need for the persistence of high levels of
effector cells and antibodies, and what is regularly observed is
that levels of these effectors in blood and peripheral tissues fall
dramatically. Only the T cells and B cells that embody antigenspecic memory are retained.
The ability of the immune system to respond to an antigenic
challenge in a sufcient, and yet measured, manner is a dramatic
expression of the ability of the system to regulate itself. If it
were not for this capacity, uncontrolled expansion of immune
cells against an antigen would wreak havoc in the host and
cause signicant morbidity, or even lymphomatous spread of
these cells. It is therefore critical to have an understanding of
how immunity regulates itself so that its response is checked
tightly in both time and space. Table 10.1 lists several of the
key methods by which immunity is regulated locally and
systemically.
REGULATION BY ANTIGEN
Antigen itself is a critical factor in regulating an immune
response. When nonreplicating (e.g., nonviral) antigens have
been studied, it has been found that the high concentration of
antigen required for initial sensitization begins to fall through
time. In part, this occurs because antibodies produced by
immunization interact with the antigen and cause its elimination. As the antigen concentration falls, the efciency with
which specic T and B cells are stimulated to proliferate and
TABLE 10.1. Levels of Immune Regulation

Regulation by antigen
Phenotype of the T-cell response (T-helper (Th)-1 and Th-2)
Suppressor/regulatory T cells
Induction of tolerance
Anergy
Clonal deletion
Suppression
Immune deviation
differentiate also falls, and eventually, when antigen concentration slips below a critical threshold, further activation of
specic lymphocytes stops. The use of anti-Rh antibodies
(RhoGAM) to prevent sensitization of Rh-negative women
bearing Rh-positive fetuses is a clear, clinical example of the
ability of antibodies to terminate (and in this particular case,
even prevent) a specic (unwanted) immune response.
REGULATION BY TH1 AND TH2 CELLS

More than 20 years ago, experimentalists discovered that certain
antigen-specic T lymphocytes are capable of suppressing
immune responses,1 and the mechanism of suppression was
found to be unrelated to the simple act of clearing the antigen
from the system. It is now understood that CD4+ T cells primed
against an antigen can differentiate into two largely distinct
phenotypes, called T-helper 1 (Th1) and Th2, based on the
cytokine products they secrete, which in turn have a signicant
effect in the character of the secondary responses generated
against that antigen.
Helper T cells are so called since they facilitate other lymphocytes to differentiate into effector and antibody-producing cells.
Any particular immunizing event does not necessarily lead to
the production of the entire array of effector modalities, and one
of the reasons for this is that helper T cells tend to polarize into
one or other of two largely distinct phenotypes.2 Th1 cells
provide a type of help that leads to the generation of T-cell
effectors that mediate delayed hypersensitivity, and B cells that
secrete complement-xing antibodies, and they perform this
function chiefly through expression of specic cytokines such as
interferon (INF)-g, tumor necrosis factor (TNF)-a, and interleukin (IL)-2. By contrast, Th2 cells provide a type of help that
leads to the generation of B cells that secrete non-complementxing IgG antibodies, as well as IgA and IgE, critical in many
humoral antibody-mediated responses in many conditions
including allergy and immunity against parasitic infection. In
turn, the ability of Th2 cells to promote these types of antibody
responses rests with their capacity to secrete a different set of
cytokines-IL-4, IL-5, IL-6, and IL-10.
As it turns out, Th1 and Th2 cells can crossregulate each
other. Thus, Th1 cells with specicity for a particular antigen
secrete IFN-g, and in the presence of this cytokine, Th2 cells
with specicity for the same antigen fail to become activated.
Similarly, if Th2 cells respond to a particular antigen by
secreting their unique set of cytokines (e.g., IL-10), Th1 cells in
the same microenvironment are prevented from responding to
the same antigen. Thus, precocious activation of Th1 cells to an
antigen, such as ragweed pollen, may prevent the activation
of ragweed-specic Th2 cells and therefore prevent the production of ragweed-specic IgE antibodies. Alternatively, precocious
91
SECTION 2
IMMUNOLOGY
activation of Th2 cells to an antigen (e.g., urushiol, the agent
responsible for poison ivy dermatitis) may prevent the
activation of urushiol-specic Th1 cells and thus eliminate the
threat of dermatitis when the skin is exposed to the leaf of the
poison ivy plant.
As it turns out, there is more to regulation and differentiation
of T cells than the neat dichotomy afforded by the Th1/Th2
paradigm. Nevertheless, the discovery of Th1 and Th2 cell
diversity has led to a profound rethinking of immune regulation.
However, it is still too early to know precisely the extent to
which the ability to influence an immune response toward the
Th1 or Th2 phenotypes will have therapeutic value in humans.
REGULATION BY SUPPRESSOR
(REGULATORY) T CELLS
Suppressor T cells are dened operationally as cells that
suppress an antigen-specic immune response.3 Cells of this
functional property were actually described before the discovery
of Th1 and Th2 cells. While it is now apparent that some of the
phenomena attributed to suppressor T cells initially are actually
explained by the crossregulating abilities of Th1 and Th2 cells,
there are distinct examples of immune suppression that cannot
be explained by either Th1 or Th2 cells.
The designation suppressor T cell has evolved over the past
decade in favor of regulatory T cells. Various experimental
maneuvers have been described that lead to the generation of
these T cells. These include1 injection of soluble protein antigen
intravenously,2 application of a hapten to skin previously exposed
to ultraviolet B radiation,4 ingestion of antigen by mouth,3
injection of allogeneic hematopoietic cells into neonatal mice,5
injection of antigen-pulsed antigen-presenting cells (APCs) that
have been treated in vitro with transforming growth factor
(TGF)-b or with fluids replete with immunosuppressive
cytokines (e.g., aqueous humor, cerebrospinal fluid, or amniotic
fluid),6 and engraftment of a solid tissue (e.g., heart, kidney)
under cover of immunosuppressive agents. In each of these
examples, T cells harvested from the lymphoid organs of these
experimentally manipulated animals induce antigen-specic
unresponsiveness when injected into immunologically competent but naive (antigen-inexperienced) animals.5
What is key, however, is that the suppressor function of
regulatory T cells is now understood not to be simply a consequence of experimental manipulation of laboratory animals,
but also an important part of normal physiology that is critical
in preventing autoimmunity.6,7 Whether experimentally induced,
or normally present, the cast of regulatory T cells that induce
unresponsiveness to self or foreign antigens is highly heterogeneous; these cells can be CD4+ or CD8+ or even natural killer
(NK) T cells.8 Some of the CD8 cells (the classically dened
suppressor T cells) inhibit the activation of CD4+ helper or
CD8+ cytotoxic T cells, whereas others interfere with B-cell
function. There are even suppressor cells that inhibit the
activation and effector functions of macrophages and other
APCs. The mechanisms that mediate the suppressor function
of regulatory T cells are the subject of intense current investigation. Certain T cells secrete immunosuppressive cytokines,
such as TGF-b or IL-10, whereas other regulatory cells inhibit
the function of other cells only when they make direct cellsurface contact with target cells; example of the latter include
CD4+CD25+ cells.6,9
TOLERANCE AS AN EXPRESSION OF
IMMUNE REGULATION
92
Classic immunologic tolerance is dened as the state in which

immunization with a specic antigen fails to lead to a detec-
table immune response. In this sense, tolerance represents the

ultimate expression of the effectiveness of immune regulation
since the unresponsiveness is maintained. Originally described
experimentally in the 1950s,10 but accurately predicted by
Ehrlich and other immunologists at the end of the nineteenth
century, immunologic tolerance has been the subject of
considerable study during the past 50 years. It has been learned
that several distinct mechanisms contribute singly, or in
unison, to creation of the state of tolerance. These mechanisms
include clonal deletion, clonal anergy, suppression, and
immune deviation.
MECHANISMS INVOLVED IN TOLERANCE

The term clonal refers to a group of lymphocytes all of which
have identical receptors for a particular antigen. During regular
immunization, a clone of antigen-specic lymphocytes responds
by proliferating and undergoing differentiation. Clonal deletion
refers to an aberration of this process in which a clone of
antigen-specic lymphocytes responds to antigen exposure by
undergoing apoptosis (programmed cell death). Deletion of a
clone of cells in this manner eliminates the ability of the
immune system to respond to that antigen, hence rendering the
immune system tolerant to that antigen. Subsequent exposures
to the same antigen fail to produce the expected immune
response (sensitized T cells and antibodies) because the relevant
antigen-specic T and B cells are missing.
Clonal anergy resembles clonal deletion in that a particular
clone of antigen-specic lymphocytes fails to respond to antigen
exposure by proliferating and undergoing differentiation.
However, in clonal anergy, the lymphocytes within the clone
are not triggered to undergo apoptosis. Rather, due to inadequate co-stimulation of the T cells by specic molecules, they
fail to become adequately activated to expand, but rather enter
an altered state in which their ability to respond is suspended,
even though these cells survive this encounter with antigen.
Still, subsequent encounters continue to fail to cause their
expected activation, rendering the immune system tolerant of
that antigen.
Antigen-specic immune suppression or regulation, as
described earlier, is another mechanism that has been shown to
cause immunologic tolerance. As in clonal deletion and anergy,
immune suppression creates a situation in which subsequent
encounters with the antigen in question fail to lead to signs of
sensitization. However, in suppression, the failure to respond is
actively maintained.
Immune deviation is a special form of immune suppression.11
Originally described in the 1960s, immune deviation refers to
the situation where administration of an antigen in a particular
manner leads to a response, but fails to elicit the expected
response. In the rst such experiments, soluble antigens
injected intravenously into naive experimental animals failed to
induce delayed hypersensitivity responses. With respect to
delayed hypersensitivity, one could say that the animals were
tolerant. However, the sera of these animals contained unexpectedly large amounts of antibody to the same antigen,
indicating that the so-called tolerance was not global, but rather
deviant. In other words, the immune response is deviated from
the expected pattern.
FACTORS THAT PROMOTE TOLERANCE

RATHER THAN IMMUNITY
Experimentalists have dened various factors that influence
or promote the development of immunologic tolerance. The
earliest description of tolerance occurred when antigenic material was injected into newborn (and therefore developmentally
immature) mice. This indicates that exposure of the developing

immune system to antigens before the system has reached
maturity leads to antigen-specic unresponsiveness. However,
tolerance can also be induced when the immune system is
developmentally mature. The factors that are known to
promote tolerance under these conditions include the physical
structure of the antigen, the dose of antigen, and the route of
antigen administration. More specically, soluble antigens are
more readily able to induce tolerance than particulate or
insoluble antigens. Very large doses as well as extremely small
quantities of antigens are also likely to induce tolerance. This
indicates that the immune system is disposed normally to
respond to antigens within a relatively broad, but dened,
range of concentrations or amounts. Injection of antigen
intravenously, or its ingestion,12 favors tolerance induction,
whereas injection of antigen cutaneously favors conventional
sensitization.
Additional factors influencing whether tolerance is induced
concern the status of the immune system itself. For example,
antigen X may readily induce tolerance when injected intravenously into a normal, immunologically naive individual.
However, if the same antigen is injected into an individual
previously immunized to antigen X, then tolerance will not
occur. Thus, a prior state of sensitization mitigates against
tolerance induction. Alternatively, if a mature immune system
has been assaulted by immunosuppressive drugs, either by debilitating systemic diseases, or by particular types of pathogens
(the human immunodeciency virus is a good example), it may
display increased susceptibility to tolerance. Thus, when an
antigen is introduced into an individual with a compromised
immune response, tolerance may develop and be maintained,
even if the immune system recovers.
REGIONAL IMMUNITY AND THE EYE

In the Overview of Immunology chapter, we discussed how
evolution had to meet the challenge of designing an immune
system that is at once capable of responding to pathogenic
antigens with a response that is effective in eliminating the
threat, while at the same time not damaging the tissue itself.
Because pathogens with different virulence strategies threaten
different types of tissues, the immune system consists of a
diversity of immune effectors. The diversity includes different
populations (e.g., CD4, CD8) of effector T cells and different
types of antibody molecules (IgM, IgG1, IgG2, IgG3, IgG4, IgA,
and IgE). Thus, different tissues and organs display markedly
different susceptibilities to immune-mediated tissue injury. The
regional specicity of an immune response is nowhere better
manifested than in the eye.13 Because integrity of the microanatomy of the visual axis is absolutely required for accurate
vision, the eye can tolerate inflammation to only a very limited
degree. Vigorous immunogenic inflammation, such as that found
in a typical delayed hypersensitivity reaction in the skin, wreaks
havoc with vision, and it has been argued that the threat of
blindness has dictated an evolutionary adaptation in the eye
that limits the expression of inflammation. Therefore, certain
aspects of immunity in the eye are considered deviant or
privileged, a description of which follows.
OCULAR SURFACE IMMUNITY

CONJUNCTIVA, LACRIMAL GLAND, TEAR
FILM, CORNEA, AND SCLERA
The human conjunctiva is an active participant in immune
defense of the ocular surface against invasion by exogenous substances. The presence of blood vessels and lymphatic channels
fosters transit of immune cells that can participate in the
afferent and efferent arms of the immune response. The

marginal and peripheral palpebral arteries and anterior ciliary
arteries are the main blood suppliers of the conjunctiva.
Lymphatics of the palpebral conjunctiva on the lateral side
drain into the preauricular and parotid lymph nodes, whereas
the lymphatics draining the palpebral conjunctiva on the medial
side drain into the submandibular lymph nodes. Major immune
cells found in normal human conjunctiva are dendritic cells,
T and B lymphocytes, mast cells, and neutrophils. Dendritic
cells, Langerhans cells, and macrophages have been detected
in different regions of the conjunctiva and cornea, but the
normal cornea is devoid of T cells.14,15 Dendritic cells act as
APCs to stimulate antigen-specic T lymphocyte responses.15
T lymphocytes, the predominant lymphocyte subpopulation in
conjunctiva, are represented in the epithelium and the substantia propria. T lymphocytes are the main effector cells in
immune reactions such as delayed hypersensitivity or cytotoxic
responses. B lymphocytes are fewer, and mostly scattered in the
substantia propria of the fornices. Plasma cells are detected only
in the conjunctival accessory lacrimal glands of Krause or minor
lacrimal glands.16 Plasma cells from major and minor lacrimal
glands synthesize Igs, mainly IgA.17,18 IgA is a dimer that is
transported across the mucosal epithelium bound to a receptor
complex. IgA dimers are released to the luminal surface of the
ducts associated with a secretory component after cleavage of
the receptor and are excreted with the tear lm. Secretory IgA is
a protectant of mucosal surfaces. Although secretory IgA does
not seem to be bacteriostatic or bactericidal, it may modulate
the normal flora of the ocular surface.19 Foreign substances can
be processed locally by the mucosal immune defense system.
After exposure to antigen, specic IgA helper T lymphocytes
stimulate B lymphocytes to differentiate into IgA-secreting
plasma cells. Dispersed T and B lymphocytes and IgA-secreting
plasma cells of the conjunctiva and lacrimal gland are referred
to as the conjunctival and lacrimal gland-associated lymphoid
tissue (CALT).17 CALT is considered part of a widespread
mucosa-associated lymphoid tissue (MALT) system, which
includes the oral mucosa and salivary gland-associated
lymphoid tissue, the gut-associated lymphoid tissue (GALT),
and the bronchus-associated lymphoid tissue (BALT). CALT
drains to the regional lymph nodes in an afferent arc; effector
cells may in turn return to the eye via an efferent arc comprised
of blood vessels; in this the lymph and blood vessels contribute
to different aspects (induction and expression, respectively) of
the immune system on the ocular surface.20
Mast cells are located mainly perilimbally, although they can
also be found in bulbar conjunctiva. Their degranulation in
response to an allergen or an injury results in the release of
vasoactive substances such as histamine, heparin, plateletactivating factor, and leukotrienes, which can cause blood vessel
dilatation and increased vascular permeability. The tears contain
several substances known to have antimicrobial properties.
Lysozyme, immunoglobulins, and lactoferrin may be synthesized by the lacrimal gland. Lysozyme is an enzyme capable of
lysing bacteria cell walls of certain Gram-positive organisms.
Lysozyme may also facilitate secretory IgA bacteriolysis in the
presence of complement. The tear IgG has been shown to
neutralize virus, lyse bacteria, and form immune complexes
that bind complement and enhance bacterial opsonization and
chemotaxis of phagocytes. Lactoferrin, an iron-binding protein,
has both bacteriostatic and bactericidal properties.21 Lactoferrin
may also interact with a specic antibody to produce an
antibacterial effect more powerful than that of either lactoferrin
or antibody alone.22
The unique anatomic and physiologic characteristics of the
human cornea explain, on the one hand, its predilection for
involvement in various immune disorders and, on the other
CHAPTER 10
93
SECTION 2
IMMUNOLOGY
hand, its ability to express immune privilege.23 The peripheral
cornea differs from the central cornea in several ways. The
former is closer to the vascularized and lymphatic-rich conjunctiva, rendering the peripheral cornea much more immunoreactive. The limbal vasculature allows diffusion of some
molecules, such as immunoglobulins and complement components, into the cornea; moreover, it signicantly facilitates
the recruitment of a wide variety of leukocyte populations into
the peripheral corneal matrix since the intravascular compartment is the chief source of these bone marrow-derived cells.24,25
IgG and IgA are found in similar concentrations in the peripheral and central cornea; however, more IgM is found in the
periphery, probably because its high molecular weight restricts
diffusion into the central area.25,26 Both classic and alternative
pathway components of complement and its inhibitors have
been demonstrated in normal human corneas. However
although most of the complement components have a
peripheral-to-central cornea ratio of >1, C1 is denser in the
periphery by a factor of ve. The higher concentration of
antibodies, complement components, APCs, and inflammatory
leukocytes in the corneal periphery and perilimbal area make
the peripheral cornea far more susceptible to involvement in a
wide variety of autoimmune and hypersensitivity disorders,
such as Moorens ulcer and collagen vascular diseases.27
The sclera consists almost entirely of collagen and proteoglycans. It is traversed by the anterior and posterior ciliary
vessels but retains a scanty vascular supply for its own use. Its
nutrition is derived from the overlying episclera and underlying
choroid;28 similarly, both classic and alternative pathway
components of complement are derived from these sources.29
Normal human sclera has few, if any, lymphocytes, macrophages, Langerhans cells, or neutrophils. In response to an
inflammatory stimulus in the sclera, the cells pass readily from
blood vessels of the episclera and choroid. Because of the
collagenous nature of the sclera, many systemic autoimmune
disorders, such as the collagen vascular diseases, may affect it.30
INTRAOCULAR IMMUNITY AND OCULAR

IMMUNE PRIVILEGE
94
For more than 100 years, it has been known that foreign tissue
grafts placed within the anterior chamber of an animals eye can
be accepted indenitely.31 The designation of this phenomenon
as immune privilege had to await the seminal work of Medawar
and colleagues, who discovered the principles of transplantation
immunology in the 1940s and 1950s. These investigators
studied immune privileged sites the anterior chamber of the
eye, the brain as a method of exploring the possible ways to
thwart immune rejection of solid tissue allografts.3234 It had
been learned that transplantation antigens on grafts were
carried to the immune system via regional lymphatic vessels
and that immunization leading to graft rejection took place
within draining lymph nodes. Because the eye and brain were
regarded at the time as having no lymphatic drainage (a concept
that has since been shown to be fallacious), and because both
tissues resided behind a bloodtissue barrier, Medawar and
associates postulated that immune privilege resulted from
immunologic ignorance. What these investigators meant was
that foreign tissues placed in immune-privileged sites were
isolated by physical vascular barriers ( antigenic sequestration)
from the immune system and that they never alerted the
immune system to their existence. During the past quarter
century and more, immunologists who have studied immune
privilege at various sites in the body have learned that this
original postulate is basically untrue.3539 First, some privileged
sites possess robust lymphatic drainage pathways the testis is
a good example. Second, antigens placed in privileged sites,
including the cornea,40 are known to escape and drain to distant

sites, including lymphoid organs such as the lymph nodes and
spleen. Third, antigens in privileged sites evoke antigenspecic, systemic immune responses, albeit of a unique nature.
Thus, the modern view of immune privilege states that privilege
is an actively acquired, dynamic state in which the immune
system conspires with the privileged tissue or site in generating
a response that is protective, rather than destructive.
IMMUNE-PRIVILEGED TISSUES AND SITES

Immune-privileged sites (Table 10.2) are regions of the body
where allografts survive for extended, even indenite, periods of
time, compared with nonprivileged, or conventional sites where
these same allografts are readily rejected. The eye contains
examples of both privileged tissues and sites, of which the beststudied site is the anterior chamber, and the best-studied tissue
is the cornea. Much has been learned about the phenomenon of
immune privilege since the 1990s. The forces that confer
immune privilege have been shown to act during both induction
and expression of the immune response to antigens placed
within, or expressed on, privileged sites and tissues.
The forces that shape immune-privileged sites and tissues
include an ever-expanding list of microanatomic, biochemical,
and immunoregulatory features. A short list of privilegepromoting features is displayed in Table 10.3. The eye expresses
virtually every one of these features. Although passive physical
features such as the bloodocular barrier, lack of lymphatics,
and low expression of major histocompatibility complex (MHC)
class I and II molecules are important, experimental attention
has focused on immunomodulatory molecules expressed on
ocular tissues and present in ocular fluids.
REGULATION OF IMMUNE EXPRESSION IN

THE EYE
There are many levels at which immune privilege is maintained
in the eye, covering virtually every step of the induction and
expression of immunogenic inflammation.14,23,34,38 APCs are
kept at an immature state, rendering them highly capable of
picking up foreign antigen but poor in stimulating T cells; lack
of lymphatics in the cornea reduces the efciency with which
antigen-laden APCs can gain access to lymphoid tissues; lack of
TABLE 10.2. Immune Privileged Sites

Eye
Cornea
Vitreous cavity
Subretinal space
Lens
Brain
Cartilage
Placenta/fetus
Testis
Ovary
Adrenal cortex
Liver
Hair follicles
Tumors
TABLE 10.3. Features of Immune Privileged Sites

Passive
Bloodtissue barriers
Decient efferent lymphatics
Tissue fluid that drains into blood vasculature
Reduced expression of major histocompatibility complex class I
and II molecules
Active
Constitutive expression of inhibitory cell surface molecules: Fas
ligand, DAF, CD59, CD46
Immunosuppressive microenvironment: TGF-b, a-MSH, VIP, CGRP,
MIF, free cortisol
MIF, melanocyte-inhibiting factor; MSH, melanocyte-stimulating hormone; VIP,
vasoinhibitory peptide; CGRP, calcitonin gene-related peptide.
blood vessels and maintenance of the bloodocular barrier

reduces the efciency by which effector T cells can gain access
to ocular tissues; and immunosuppressive and proapoptotic
signals in the eye actively suppress or delete lymphocytes that
have gained access to ocular compartments.38 Herein, we shall
focus on a few of the mechanisms that regulated T-cell activation in the eye.
It is know that activated T cells upregulate expression of the
death receptor, Fas (CD95), on their surface, and by doing so
become vulnerable to programmed cell death if they encounter
other cells that express Fas ligand (CD95L).41 Constitutive
expression of Fas ligand on cells that surround the anterior
chamber has been shown to induce apoptosis among T cells and
other Fas+ leukocytes exposed to this anterior chamber.42 More
important, Fas ligand expressed by cells of the cornea play a key
role in rendering the cornea resistant to immune attack and
rejection.43 Similarly, constitutive expression on corneal endothelial cells, as well as iris and ciliary body epithelium, of
several membrane-bound inhibitors of complement activation
are strategically located to prevent complement-dependent intraocular inflammation and injury.44 More recently, another factor,
which is a member of the B7 costimulatory superfamily, known
as programmed death ligand-1 (PD-L1) has been shown to be
constitutively expressed at very high levels by the cornea, implicating this factor in the active deletion of PD-1+ T cells from the
anterior segment.
Cells that are not deleted/killed in this microenvironment are
rendered less hostile by a highly immunosuppressive milieu.
For example, transforming growth factor-beta 2 (TGF-b2), a
normal constituent of aqueous humor,45 is a powerful immunosuppressant that inhibits various aspects of T cell and macrophage activation. Other relevant factors in the aqueous humor
include alpha-melanocyte-stimulating hormone,46 vasoactive
intestinal peptide,47 calcitonin gene-related peptide,48 and
macrophage migration inhibitory factor,49 among others. It is
important to emphasize, however, that aqueous humor does not
inhibit all immune reactivity. For example, antibody neutralization of virus infection of target cells is not prevented in the
presence of aqueous humor.50
REGULATION OF INDUCTION OF IMMUNITY

TO EYE-DERIVED ANTIGENS
Another dimension to immune privilege is the ability of the eye
to regulate the nature of the systemic immune response to
antigens placed within it, an issue of paramount importance
as it is the systemic immune response that plays a critical role

in sustaining immunity in peripheral tissues including the
eye. It has been known from the 1980s that injection of alloantigenic cells into the anterior chamber of rodent eyes evokes
a distinctive type of immune deviation, now called anterior
chamber-associated immune deviation (ACAID).51,52 In
ACAID, eye-derived antigens elicit an immune response that is
selectively decient in T cells that mediate delayed hypersensitivity, and B cells that secrete complement-xing antibodies. There is not, however, a global lack of response, because
animals with ACAID display a high level of antigen-specic
serum antibodies of the non-complement-xing varieties and
primed cytotoxic T cells.37 In ACAID, regulatory T cells are also
generated which, in an antigen-specic manner, suppress both
the induction and expression of delayed hypersensitivity to the
antigen in question.53,54 ACAID can be elicited by diverse types
of antigens, ranging from soluble protein to histocompatibility
to virus-encoded antigens.
Induction of ACAID by intraocular injection of antigen
begins within the eye itself.5557 After injection of antigen into
the eye, local APCs capture the antigen, migrate across the
trabecular meshwork into the canal of Schlemm, and then
trafc via the blood to the spleen. In the splenic white pulp,
the antigen is presented in a unique manner to T and
B lymphocytes, resulting in the spectrum of functionally
distinct antigen-specic T cells and antibodies found in
ACAID. The ocular microenvironment sets the stage for this
sequence of events by virtue of the immunoregulatory
properties of the aqueous humor described earlier. This ocular
fluid, or more precisely, TGF-b2, confers upon conventional
APCs the capacity to induce ACAID. Thus, the ocular microenvironment not only regulates the expression of immunity
within the eye, but also the functions of eye-derived APCs and
thus promotes a systemic immune response that is decient in
those immune effector modalities most capable of inducing
immunogenic inflammation-delayed hypersensitivity T cells
and complement-xing antibodies.
CHAPTER 10
IMMUNE PRIVILEGE AND INTRAOCULAR

INFLAMMATORY DISEASES
Ocular immune privilege has been implicated in1 the extraordinary success of corneal allografts,5862 progressive growth of
intraocular tumors,63 resistance to herpes stromal keratitis,64
and4 suppression of autoimmune uveoretinitis.65,66 When
immune privilege prevails within the eye, corneal allografts
succeed; trauma to the eye heals without incident; and ocular
infections are cleared without inflammation. However, the price
of this compromise is that ocular tumors may then grow relentlessly, and uveal tract infections may persist and recur.34,37 In
contrast, the consequences of failed immune privilege are
protean. For example, ocular trauma may result in sympathetic
ophthalmia, ocular infections may produce sight-threatening
inflammation, and corneal allografts may undergo irreversible
rejection.
CORNEAL TRANSPLANTATION
IMMUNOLOGY
Our objective here is not to provide a thorough review of the
immunobiology of corneal transplantation, which has been
extensively reviewed elsewhere.6167 Rather, we shall focus on
the mechanisms of ocular immune privilege as they affect the
fate of corneal allografts, and demonstrate how abrogation of
such privilege can lead to immunogenic graft failure.
The cornea is an immune privileged tissue and, in part, this
accounts for the extraordinary success of corneal transplants in
95
SECTION 2
IMMUNOLOGY
both experimental animals and humans. However, despite the
many advances that have been made in corneal tissue preservation and surgical techniques, a signicant proportion of
grafts eventually fail,68 and this is nowhere as signicant a
problem as when grafts are placed onto inflamed and
neovascularized host beds. Regardless of host bed parameters,
or the indication for transplantation, the main cause of corneal
graft failure is immune-mediated graft rejection, the rate
ranging from as low as 10% in grafts performed for keratoconus
and bullous keratopathy, to well over 50% in grafts performed
for corneal burns and other conditions associated with surface
disease and stromal vascularization.61 Corneal vascularization,
either preoperative from recipient herpetic, interstitial, or
traumatic keratitis, or stimulated by silk or loose sutures,
contact lenses, infections, persistent epithelial defects, and
other disorders associated with inflammation, has been widely
recognized as a clear risk factor for decreased graft survival.
Other factors that increase the risk of allograft rejection include
a history of previous graft loss, eccentric and large grafts, and
glaucoma.6971
TRANSPLANT ANTIGENS ON CORNEAL

TISSUE
96
In outbred species, such as humans, where genotypic variation

is high, transplants of solid tissue grafts usually fail unless the
recipient is immunosuppressed. The reason for this is
development of an immune response directed at so-called
transplantation antigens displayed on cells of the graft.
Immunologists have separated transplantation antigens into
two categories: major and minor, primarily because of purely
empirical evidence that major antigens induce more vigorous
alloimmunity than do minor antigens. The genes that encode
the major transplantation antigens in humans are located
within the MHC, called human leukocyte antigen (HLA).
Minor histocompatibility antigens are encoded at numerous
loci spread throughout the genome. The HLA complex, which
is a large genetic region, is situated on the short arm of the sixth
human chromosome. HLA genes that encode class I and class
II antigens are extremely polymorphic. Similarly, minor
histocompatibility loci contain highly polymorphic genes. In
the aggregate, polymorphisms at the major and minor
histocompatibility loci account for the observation that solid
tissue grafts exchanged between any two individuals selected at
random within a species are acutely rejected.
The expression of HLA antigens on corneal cells is somewhat
atypical.7274 Class I MHC antigens are expressed strongly on
the epithelial cells of the cornea, comparable in intensity to
the expression of epidermal cells of skin. Keratocytes express
less class I than conventional broblasts, and corneal
endothelial cells express small amounts of class I antigens
under normal circumstances. Additionally, class II MHC
(e.g., HLA D/DR) antigen expression is essentially absent in
the normal corneal tissue. However, corneal cells respond to
specic cytokines, such as INF-g, by upregulating MHC antigen
expression.
If the normal cornea exhibits little MHC expression, but can
acquire high-level expression when inflamed, what is the
benet of tissue matching? The evidence for HLA tissue typing
in corneal transplantation is conflicting.7581 There seems to
be little controversy regarding the influence of tissue typing
on grafts placed in eyes of low-risk patients. In the low-risk
situation, with a few exceptions,81 virtually no studies suggest a
positive typing effect. Most likely, the rate of graft success is
so high in low-risk transplants under cover of topical steroids
that there is little opportunity for a matching effect to be seen.
However, in high-risk situations, the literature contains many
disparate reports with conflicting conclusions regarding the

utility of HLA matching. On balance, however, notwithstanding
the results of the Collaborative Corneal Transplantation
Studies (CCTS), a multicenter study completed in the United
States in the early 1990s that failed to demonstrate any
protection from HLA matching,79 the majority of large
studies have supported the concept of antigen-matching for
corneal transplants conducted in hosts at high risk for graft
rejection.
One of the unexpected outcomes of the CCTS was the
nding that ABO blood type matching was signicantly
protective of corneal transplants.79 This was difcult to explain
in the early 1990s, until studies on corneal transplantation
performed in rodents reported that minor transplantation
antigens offer a signicant barrier to graft success.82,83 Minor
antigens are thus called since in conventional solid tissue
(e.g., skin) grafts, they are not as determining of graft success
as compared to MHC antigens. However, as described earlier,
there is signicantly reduced expression of MHC antigens by
corneal grafts. Hence, in the cornea, minor transplantation
antigens are potentially quantitatively more numerous than
MHC antigens, and ABO antigens may well represent possible
minor antigens.
CORNEAL TRANSPLANT SURVIVAL AN

EXAMPLE OF THE SUCCESS OF IMMUNE
PRIVILEGE
The normal cornea is an immune-privileged tissue, and several
features are known to contribute to the privileged status. First,
as mentioned earlier, the expression of MHC class I and class II
molecules is reduced and impaired, especially on the corneal
endothelium. The net antigenic load of corneal tissue is thus
reduced compared with other tissues, which has a mitigating
effect on both the induction and expression of alloimmunity.
Second, the cornea lacks blood and lymph vessels. The absence
of these vascular structures provides relative isolation for
corneal antigens in a manner that reduces, though does not
prevent, antigenic information from escaping from the tissue
while at the same time suppressing immune effectors from
gaining access to the tissue. Third, the cornea is decient in
activated APCs that exhibit high levels of MHC class II and
requisite co-stimulatory molecules (e.g., CD40, CD80, CD86)
for priming T cells. Indeed, the bone marrow-derived cells of the
cornea are of a highly immature phenotype and uniformly
MHC class II-negative.38 Fourth, as detailed above, there is
considerable expression of a variety of immunosuppressive
factors by various tissues in the anterior segment of the eye that
impair induction and expression of conventional immunity.4350
These immunosuppressive molecules have powerful immunomodulatory effects on APCs, T cells, B cells, NK cells, and
macrophages, and can suppress many forms of immunity
including alloreactive responses. Fifth, cells of the cornea
constitutively express surface molecules, including DAF, CD59,
CD46, PD-L1, and others that can inhibit numerous complement and T cell effector functions.
The dramatic expression of immune privilege is mirrored by
the success of keratoplasties performed in low-risk situations in
humans. Modest amounts of topical steroids in the early
postoperative period, even followed by cessation of all therapy
later, is still associated, in the vast majority of cases, with
indenite survival of most corneal transplants. However, not all
grafts are successful. In high-risk transplantation, performed
in inflamed host beds, the prognosis is worse than many forms
of solid organ transplants. What are the mechanisms that
lead to graft rejection, and how does immune privilege fail in
some circumstances?
CORNEAL TRANSPLANT REJECTION THE

EROSION OF IMMUNE PRIVILEGE
The immunopathogenic mechanisms that lead to corneal
transplant rejection have been reviewed elsewhere.6167 Basic
investigations into the mechanisms responsible for alloimmunity in the high-risk setting have shown how the principal modalities that dictate immune privilege in the healthy/
physiologic setting can erode after sustained inflammation,
setting the stage for transplant rejection.
It is instructive to place these events in the context of
immune privilege reviewed in the earlier section. First, surgery
itself leads to expression of MHC molecules by the cornea.40
Second, inflammation leads to induction of angiogenic processes, prompting growth of both blood and lymph vessels into
the corneal matrix, thereby affecting the relative sequestration
and protection of the cornea from the immune system.84,85
Third, profound changes occur in relation to corneal APCs; the
rst is that there is massive mobilization of these cells into
the graft;38,61 the second is that under conditions of intense
inflammation the APCs change their phenotype and mature
(become activated) by acquisition of MHC class II and costimulatory molecules that render them highly capable of
sensitizing host T cells.86 These changes are reflected in the fact
that in both animal models and the clinical setting, high-risk
graft rejection occurs at an accelerated rate, reflecting the
efciency by which the host has become sensitized to graft
antigens. For example, sensitization develops in recipient
animals with surprising rapidity when grafts are placed in highrisk eyes. Within 7 days of engraftment, immune donor-specic
T cells can be detected in lymphoid tissues. Similar grafts placed
in low-risk mouse eyes do not achieve T-cell sensitization until
at least 3 weeks after engraftment. It is very likely that the
vulnerability to rejection of grafts placed in high-risk eyes is
dictated by the efciency with which APCs are mobilized in the
graft, and migrate to regional lymph nodes where recipient

T cells are initially activated.40 Fourth, the signicant overexpression of proinflammatory cytokines generated in inflamed
eyes in the postoperative period can effectively counteract the
function of many of the immunosuppressive cytokines that
normally downmodulate immunity in the healthy eye under
the physiologic state. Hence, under conditions of intense
inflammation, as may occur after transplant surgery, and particularly in the high-risk host, the inherent immune privileged
status of the graft is clearly insufcient to overcome the
fact that the graft site can no longer act as an immuneprivileged site.
SUMMARY AND CONCLUSION
CHAPTER 10
The eye is defended against pathogens, just as is every other

part of the body. Components of both the natural and the
acquired immune systems respond to pathogens in the eye, but
the responses are different from those following antigen
encounter in most other places in the body, perhaps as a result
of evolutionary pressures resulting in the survival of those
species and species members in which a blinding, exuberant
inflammatory response was prevented by regulation of the
response. In any event, we are left for the moment with an
organ (the eye) in which special immunologic responsiveness
allows us to enjoy a degree of privilege tolerance to transplanted tissue not experienced by other organs. It is clear now
that this tolerance is an active process, not simply a passive one
derived from the invisibility of the transplant from the
recipients immune system.
ACKNOWLEDGMENT
The authors would like to acknowledge the signicant material contribution
of Dr J Wayne Streilein to the previous edition of this chapter.
REFERENCES
1. Qin S, Cobbold SP, Pope H, et al:
Infectious transplantation tolerance.
Science 1993; 259:974977.
2. Mosmann TR, Coffman RL: Th1 and Th2
cells: different patterns of lymphokine
secretion lead to different functional
properties. Ann Rev Immunol 1989;
7:145173.
3. Sigal LH: Basic science for the clinician:
T-cells with regulatory function. J Clin
Rheumatol 2005; 11:286289.
4. Gor DO, Rose NR, Greenspan NS:
Th1Th2: a procrustean paradigm. Nat
Immunol 2003; 4:503505.
5. Taylor A, Verhagen J, Blaser K, et al:
Mechanisms of immune suppression by
interleukin-10 and transforming growth
factor-beta: the role of T regulatory cells.
Immunology 2006; 117:433442.
6. Dejaco C, Duftner C, Grubeck-Loebenstein
B, Schirmer M: Imbalance of regulatory
T cells in human autoimmune diseases.
Immunology 2006; 117:289300.
7. Siatskas C, Chan J, Field J, et al: Gene
therapy strategies towards immune
tolerance to treat the autoimmune
diseases. Curr Gene Ther 2006;
6:4558.
8. Cardell SL: The natural killer T lymphocyte:
a player in the complex regulation of
autoimmune diabetes in non-obese
diabetic mice. Clin Exp Immunol 2006;
143:194202.
9. Randolph DA, Fathman CG: CD4+CD25+

regulatory T cells and their therapeutic
potential. Annu Rev Med 2006; 57:381402.
10. Billingham RE, Brent L, Medawar PB:
Actively acquired tolerance of foreign cells.
Nature 1953; 172:603.
11. Asherson GL, Stone SH: Selective and
specic inhibition of 24-hour skin reactions
in the guinea-pig. I. Immune deviation:
description of the phenomenon and the
effect of splenectomy. Immunology 1965;
9:205211.
12. Khoury SJ, Hancock WW, Weiner HL: Oral
tolerance to myelin basic protein and
natural recovery from experimental
autoimmune encephalomyelitis are
associated with downregulation of
inflammatory cytokines and differential
upregulation of transforming growth factor
b, interleukin 4, and prostaglandin E
expression in the brain. J Exp Med 1992;
176:13551364.
13. Streilein JW: Regional immunology of the
eye. In: Pepose JW, Holland GN, Wilhemus
KR, eds. Ocular infection and immunity.
Philadelphia: Mosby-Year Book;
1996:1933.
14. Hamrah P, Huq SO, Gulati A, Dana MR:
Mechanisms of the ocular surface immune
response. In: Pflugfelder S, Beuerman R,
Stern M, eds. Dry eye and ocular surface
disorders. Chapter 6. New York: Marcel
Dekker; 2004:111142.
15. Dana MR, Hamrah P: Corneal antigenpresenting cells: function and identity. In:
Zierhut M, ed. Immunology of the lacrimal
gland, tear lm, and ocular surface. London:
Taylor and Francis Group; 2005:141149.
16. Sacks E, Wieczorek R, Jakobiec FA, et al:
Lymphocytic sub-populations in the normal
human conjunctiva. Ophthalmology 1986;
93:12761283.
17. Franklin RM, Remus LE: Conjunctivalassociated lymphoid tissue: evidence
for a role in the secretory immune
system. Invest Ophthalmol Vis Sci 1984;
25:181187.
18. Wieczorek R, Jakobiec FA, Sacks E, et al:
The immunoarchitecture of the normal
human lacrimal gland. Ophthalmology
1988; 95:100109.
19. Gibbons RJ: Bacterial adherence to the
mucosal surfaces and its inhibition by
secretory antibodies. Adv Exp Med Biol
1974; 45:315325.
20. Dana MR, Dana MR: Angiogenesis and
lymphangiogenesis: novel implications for
corneal immunity. Sem Ophthalmol 2006;
21:1922.
21. Arnold RR, Cole MF, McGhee JR: A
Bactericidal effect for human lactoferrin.
Science 1977; 197:263.
22. Badgy GC: Interaction of lactoferrin
monocytes and lymphocyte subsets in the
regulation of steady-state granulopoiesis in
vitro. J Clin Invest 1981; 68:56.
97
SECTION 2
IMMUNOLOGY
98
23. Dana MR, Qian Y, Hamrah P: Twenty-ve

year panorama of corneal immunology:
emerging concepts in the
immunopathogenesis of microbial keratitis,
peripheral ulcerative keratitis, and corneal
transplant rejection. Cornea 2000;
19:625643.
24. Zhu S, Dana MR: Expression of cell
adhesion molecules on limbal and
neovascular endothelium in corneal
inflammatory neovascularization. Invest
25. Allansmith MR, McClellan BH:
Immunoglobulins in the human cornea. Am
J Ophthalmol 1975; 80:123132.
26. Mondino BJ, Brady KJ: Distribution of
hemolytic complement in the normal
cornea. Arch Ophthalmol 1981;
99:14301433.
27. Mondino BJ: Inflammatory diseases of the
peripheral cornea. Ophthalmology 1988;
95:463472.
28. Watson PG, Hazleman BL: The sclera and
systemic disorders. Philadelphia, PA: WB
Saunders; 1976.
29. Brawman-Mintzer O, Mondino BJ, Mayer
FJ: Distribution of complement in the
sclera. Invest Ophthalmol Vis Sci 1989;
30:22402244.
30. Fong LP, Sainz de la Maza M, Rice BA,
et al: Immunopathology of scleritis.
31. van Dooremall JC: Die Entwicklung der in
fremden Grund versetzten lebenden
gewebe. Graefes Arch Clin Exp Ophthalmol
1873; 19:358373.
32. Medawar P: Immunity to homologous
transplanted to the brain, to subcutaneous
tissue and to the anterior chamber of the
33. Barker CF, Billingham RE: Immunologically
privileged sites. Adv Immunol 1977;
25:154.
34. Streilein JW: Perspective: unraveling
immune privilege. Science 1995;
270:11581159.
35. Streilein, JW: Immune regulation and the
eye: a dangerous compromise. FASEB J
1987; 1:199208.
36. Niederkorn JY: Immune privilege and
immune regulation in the eye. Adv Immunol
1990; 48:191226.
37. Streilein JW: Ocular immune privilege and
the Faustian dilemma. Invest Ophthalmol
Vis Sci 1996; 37:19401950.
38. Dana MR: Corneal antigen-presenting cells:
diversity, plasticity, and disguise: the
Cogan lecture. Invest Ophthalmol Vis Sci
2004; 45:722727, 721.
39. Streilein JW, Ksander BR, Taylor AW:
Commentary: immune privilege, deviation
and regulation in the eye. J Immunol 1997;
158:35573560.
40. Liu Y, Hamrah P, Zhang Q, et al: Draining
lymph nodes of corneal transplant hosts
exhibit evidence for donor MHC class
II-positive dendritic cells derived from MHC
class II-negative grafts. J Exp Medicine
2002; 195:259268.
41. Nagata S, Golstein P: The Fas death factor.
Science 1995; 267:14491456.
42. Grifth TS, Brunner T, Fletcher SM, et al:
Fas ligand-induced apoptosis as a
mechanism of immune privilege. Science
1995; 270:11891192.
43. Yamagami S, Kawashima H, Tsuru T, et al:
Role of Fas/Fas ligand interactions in the
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
immunorejection of allogeneic mouse

corneal transplantation. Transplantation
1997; 64:11071111.
Bora NS, Gobleman CL, Atkinson JP, et al:
Differential expression of the complement
regulatory proteins in the human eye.
Invest Ophthal Vis Sci 1993; 34:35793584.
Cousins SW, McCabe MM, Danielpour D,
et al: Identication of transforming growth
factor-beta as an immunosuppressive
factor in aqueous humor. Invest
Taylor AW, Streilein JW, Cousins SW:
Identication of alpha-melanocyte
stimulating hormone as a potential
immuno-suppressive factor in aqueous
humor. Curr Eye Res 1992; 11:11991206.
Taylor AW, Streilein JW, Cousins SW:
Vasoactive intestinal peptide (VIP)
contributes to the immunosuppressive
activity of normal aqueous humor.
J Immunol 1994; 153:10801086.
Wahlestedt C, Beding N, Ekman R:
Calcitonin gene-related peptide in the eye:
release by sensory nerve stimulation and
effects associated with neurogenic
inflammation. Regul Pept 1986; 16:107115.
Apte RS, Niederkorn JY: MIF: A novel
inhibitor of NK cell activity in the anterior
chamber (AC) of the eye. J Allergy Clin
Immunol 1997; 99:S467.
Kaiser CJ, Ksander BR, Streilein JW:
Inhibition of lymphocyte proliferation by
aqueous humor. Reg Immunol 1989;
2:4249.
Kaplan HJ, Streilein JW: Immune response
to immunization via the anterior chamber of
the eye. I. F1 lymphocyte-induced immune
deviation. J Immunol 1977; 118:809814.
Kaplan HJ, Streilein JW: Immune response
to immunization via the anterior chamber of
the eye. II. An analysis of F1 lymphocyte
induced immune deviation. J Immunol
1978; 120:689693.
Streilein JW, Niederkorn JY:
Characterization of the suppressor cell(s)
responsible for anterior chamber
associated immune deviation (ACAID)
induced in BALB/c mice by P815 cells.
J Immunol 1985; 134:13811387.
Ferguson TA, Kaplan HJ: The immune
response and the eye. II. The nature of
T suppressor cell induction of anterior
chamber-associated immune deviation
(ACADI). J Immunol 1987; 139:352357.
Wilbanks GA, Streilein JW: Characterization
of suppressor cells in anterior
chamber-associated immune deviation
(ACAID) induced by soluble antigen:
evidence of two functionally and
phenotypically distinct T-suppressor cell
populations. Immunology 1990;
71:383389.
Wilbanks GA, Streilein JW: Studies on the
induction of anterior chamber associated
immune deviation (ACAID). I. Evidence that
an antigen-specic, ACAID-inducing, cellassociated signal exists in the peripheral
blood. J Immunol 1991; 146:26102617.
Wilbanks GA, Mammolenti MM, Streilein
JW: Studies on the induction of anterior
chamber associated immune deviation
(ACAID). II. Eye-derived cells participate in
generating blood borne signals that induce
ACAID. J Immunol 1991; 146:30183024.
Maumanee AE: The influence of donorrecipient sensitization on corneal grafts.
Am J Ophthalmol 1951; 34:142152.
59. Sonoda Y, Streilein JW: Orthotopic corneal

transplantation in mice: evidence that the
immunogenetic rules of rejection do not
apply. Transplantation 1992; 54:694703.
60. Qian Y, Boisgerault F, Benichou G, Dana
MR: Blockade of CD40CD40L
costimulatory pathway promotes survival of
allogeneic corneal transplants. Invest
61. Qian Y, Dana MR: Molecular mechanisms
of immunity in corneal allotransplantation
and xenotransplantation. Expert Rev Mol
Med 2001; 3:121.
62. Boisgerault F, Liu Y, Anosova N, et al: Role
of CD4+ and CD8+ T cells in
allorecognition. Lessons from corneal
transplantation. J Immunol 2001;
167:18911899.
63. Niederkorn J, Streilein JW, Shadduck JA:
Deviant immune responses to allogeneic
tumors injected intracamerally and
subcutaneously in mice. Invest Ophthalmol
Vis Sci 1980; 20:355363.
64. McLeish W, Rubsamen P, Atherton SS, et
al: Immunobiology of Langerhans cells on
the ocular surface. II. Role of central
corneal Langerhans cells in stromal keratitis
following experimental HSV-1 infection in
mice. Reg Immunol 1989; 2:236243.
65. Mizuno K, Clark AF, Streilein JW: Induction
of anterior chamber associated immune
deviation in rats receiving intracameral
injections of retinal S antigen. Curr Eye Res
1988; 7:627632.
66. Hara Y, Caspi RR, Wiggert B, et al:
Suppression of experimental autoimmune
uveitis in mice by induction of anterior
chamber associated immune deviation with
interphotoreceptor retinoid binding protein.
J Immunol 1992; 148:16851692.
67. Niederkorn JY: The immune privilege of
corneal grafts. J Leukoc Biol 2003;
74:167171.
68. Wilson SE, Kaufman HE: Graft failure after
penetrating keratoplasty. Surv Ophthalmol
1990; 34:325356.
69. Vlker-Dieben HJM, DAmaro J, Kok-van
Alphen CC: Hierarchy of prognostic factors
for corneal allograft survival. Aust N Z J
70. Boisjoly HM, Bernard P-M, Dube I, et al:
Effect of factors unrelated to tissue etching
on corneal transplant endothelial rejection.
Am J Ophthalmol 1989; 107:647654.
71. Dana MR, Streilein JW: Loss and
restoration of immune privilege in eyes with
corneal neovascularization. Invest
72. Fujikawa LS, Colvin RB, Bhan AK, et al:
Expression of HLA-A/B/C and -DR locus
antigens on epithelial, stromal and
endothelial cells of the human cornea.
Cornea 1982; 1:213.
73. Whitsett CF, Stulting RD: The distribution
of HLA antigens on human corneal
tissue. Invest Ophthalmol Vis Sci 1984;
25:519524.
74. Treseler PA, Foulks GN, Sanlippo F: The
expression of HLA antigens by cells in the
human cornea. Am J Ophthalmol 1984;
98:763772.
75. Batchelor JR, Casey TA, Gibbs DC, et al:
HLA matching and corneal grafting. Lancet
1976; 1:551554.
76. Foulks GN, Sanlippo FP, Locascio JA,
et al: Histocompatibility testing for
keratoplasty in high-risk patients.
77. Sanlippo F, MacQueen JM, Vaughn WK,

et al: Reduced graft rejection with good
HLA-A and -B matching in high-risk
corneal transplantation. N Engl J Med
1986; 315:2935.
78. Boisjoly HM, Bernard P-M, et al:
Association between corneal allograft
reactions and HLA compatibility.
79. Stark W, Stulting D, Maguire M, et al: The
collaborative corneal transplantation
studies (CCTS): effectiveness of
histocompatibility matching of donors and
recipients in high risk corneal
transplantation. Arch Ophthalmol 1992;
110:13921403.
80. Gore SM, Vail A, Bradley BA, et al:

HLA-DR matching in corneal
transplantation. Transplantation 1995;
60:10331039.
81. Sundmacher R: A clinicians outlook on
HLA matching for keratoplasty. Dev
82. Sonoda Y, Streilein JW: Impaired cell
mediated immunity in mice bearing healthy
orthotopic corneal allografts. J Immunol
1993; 150:17271734.
83. Sonoda Y, Sano Y, Ksander B, et al:
Characterization of cell mediated immune
responses elicited by orthotopic corneal
allografts in mice. Invest Ophthalmol Vis
Sci 1995; 36:427434.
84. Dana MR: Angiogenesis and

lymphangiogenesis: novel implications for
corneal immunity. Sem Ophthalmol 2006;
21:1922.
85. Yamagami S, Hamrah P, Zhang Q, et al:
Early ocular chemokine gene expression
and leukocyte inltration after high-risk
corneal transplantation. Mol Vision 2005:
11:632640.
86. Huq S, Liu Y, Benichou G, Dana MR:
Relevance of the direct pathway of
sensitization in corneal transplantation is
dictated by the graft bed
microenvironment. J Immunol 2004:
173:44644469.
CHAPTER 10
99
SECTION 3
MICROBIOLOGY
Edited by Michael S. Gilmore
CHAPTER
11
Ocular Bacteriology
Christopher N. Ta, Robert W. Bowman, and James P. McCulley
Overview
Bacteria are ubiquitous in the environment and are part of the
normal flora of humans. The balance between the virulence of
the bacteria and the strength of the immune system plays a role
in whether or not an infection will occur. In order to initiate an
infection, bacteria must be able to adhere to the surface,
multiply, colonize, and evade the immune system, and finally,
invade the tissue. In contrast, the host defense system includes
mechanical removal of bacteria, such as the tear film and
blinking reflexes. The immune system, both humoral and cellular
response, is important in preventing and eliminating a bacterial
infection. Once an infection has occurred, the treating physician
must attempt to identify the etiology of the infection. The most
common classification of bacteria is based on the Gram stain
characteristics. The available tests include the traditional stains
and culture media, and more recently, the tests such as
polymerase chain reaction. The mainstay for treatment of
bacterial infections are antibiotics, although recent evidence
suggests that resistance to many commonly prescribed
antibiotics is on the rise. Finally, it is critical to consider
prophylaxis against infections in patients undergoing ocular
surgery with the use of antiseptic agents, most commonly with
povidone-iodine.
Bacterial infections comprise a complex and constantly
changing group of ocular diseases. Various bacteriologic
processes involve the eyes and periocular structures, from
something as simple as colonization of the skin and lashes alone
without invasive disease to necrotizing bacterial keratitis. The site
of infection may be the periocular skin or lid or an anaerobic
environment such as the canalicular system or the capsular bag.
The source of bacteria may be local (i.e., from the lids and
lashes), or it may be from a remote site (as in metastatic
endophthalmitis) or from the nasopharynx or sinuses. In recent
years, significant advances in our understanding of the
mechanisms of bacterial diseases have been made. Bacterial
antibiotic resistance has been on the increase, and newer
antibiotics that are more specific in their coverage have become
available. We are constantly understanding more and more
about the hostbacterial interaction, its effect on bacterial
virulence and pathogenicity, and the resultant therapeutic
implications. The methods of identifying bacteria are gradually
shifting away from traditional staining and culture techniques to
newer automated or rapid-identification techniques. More
recently, atypical bacteria have been found to be associated with
infectious keratitis following refractive surgery. The role of slime
is increasingly recognized to play an important role in the
pathogenesis of ocular infections, particularly with regard to
contact lens and intraocular lens related infections. The ability to
diagnose and treat infections correctly is critical. One might
ask, What should I know that will help me in the management of
my patient with a bacterial infection? In this chapter, we
attempt to give the reader the basis for understanding this
ever-changing field.
ANATOMY, PHYSIOLOGY, AND LIFE CYCLE

Bacteria belong to the kingdom Protista, which encompasses
fungi, protozoa, and algae as well. The more complex eukaryotic organisms are the fungi, protozoa, and algae; the simpler
prokaryotic organisms are the bacteria. The taxonomy of the
bacteria is extensive, having undergone frequent revisions in the
past but now requiring the approval of an ofcial international
body.1 With newer techniques such as deoxyribonucleic acid
(DNA) typing and sequencing, the heterogeneity of bacteria
within their various groups becomes more apparent. The
determination of DNA composition by identifying the G + C
(the amino acids guanine, G, and cytosine, C) content of DNA
has shown that the whole phylum of vertebrates ranges only
from 36% to 44% G + C, whereas bacteria range from 25% to
75%. For example, in the genera Staphylococcus and Micrococcus, which are in the family Micrococcaceae, the former has
3040% G + C, whereas the latter has 6575% G + C.2 Such a
variation in DNA sequencing among bacteria is now being used
clinically to develop rapid diagnostic systems.
The most practical method of classifying bacteria still
depends on their Gram-staining properties and their cell morphology. Also important, however, are their fermentation
products, their ability to metabolize various substrates, their
sensitivity to different antibiotics, and their colonial morphology. Bacteria lack any nuclear or mitotic apparatus; their
DNA is organized into a single, naked, circular chromosome
that is ~1 mm in length. Some bacteria, such as Borrelia
burgdorferi, which causes Lyme disease, have a linear
chromosome. Smaller molecules of DNA known as plasmids
are signicant, because they may carry information for drug
resistance or they may code for toxins that can affect human
cellular functions. The structure of bacterial cells is termed
prokaryotic; whereas those with a membrane-bounded nucleus
are called eukaryotic. Owing to their small size, there is a limit
to the number of molecules that can be present in the cell at
any given time. Prokaryotic cells have come to regulate their
synthesis by induction, regression, or end product inhibition
to produce only what is required for metabolism or growth in a
particular environment.3
Phospholipids and proteins make up the bacterial cell membrane, and in contrast to eukaryotic cells, bacterial cell walls
(except for those of mycoplasmas) do not contain sterols.
Because prokaryotic cells lack both mitochondria and an endoplasmic reticulum, electron transport systems are located in
the cell wall itself.
The cell wall or cell envelope plays an important role in
many bacterial cell functions. Besides containing the electron
transport systems, the envelope also serves as an osmotic
101
SECTION 3
MICROBIOLOGY
102
barrier and regulates the transport of solutes. Thus, the cell wall
protects the cell against rupture from the high internal osmotic
pressure. In hypertonic environments, bacteria may survive as
spheroplasts, or L forms, without their rigid cell wall, but as
a result they may lose their pathogenicity. A macromolecule
unique to the cell wall of many bacteria is the peptidoglycan
(PG). This component of the cell wall is responsible for shape
denition and maintenance.3
The cell wall is the site of many antigenic determinants of
the various bacteria. Moreover, when endotoxin is present, it
is located in the cell wall. The cell envelope of Gram-positive
bacteria has only a thick (1580 nm) PG layer surrounded by a
polysaccharide capsule. PG is a cross-linked heteropolymer
of amino acids and amino sugars that constitute ~50% of
the cell wall by weight.4 Teichoic acid (TA) is a negatively
charged ribitol-phosphate polymer that attaches to PG by
covalent bonds, accounting for 40% of the cell wall.5 The cell
envelope of Gram-negative bacteria is more complex than that
of Gram-positive bacteria. Although the PG layer is thinner
(only 12 nm), there is a phospholipid outer membrane that
forms a protective barrier, making Gram-negative bacteria more
resistant to hydrolytic enzymes and toxic substances. Membrane proteins that are present in the outer membrane serve to
regulate transport through transmembrane prexing, or porins,
allowing the passive diffusion of low-molecular-weight compounds such as sugars and amino acids. Antibiotics are much
larger molecules and therefore have difculty penetrating the
outer membrane and in part are responsible for Gram-negative
bacterial antibiotic resistance. For example, Pseudomonas
aeruginosa are highly resistant to antibiotics due to the outer
membrane. The number and diameter of the porin channels
vary among different Gram-negative species, which helps explain
some of their intrinsic differences in antibiotic susceptibility.6
Gram-negative bacteria possess a periplasma between the inner
and outer walls of the cell membrane. The periplasma contains
at least 50 different properties. Important among these may be
b-lactamase and aminoglycoside phosphorylase that function
to inactivate certain antibiotics.3 Also found in the outer
membrane of Gram-negative bacteria is endotoxin, composed of
lipopolysaccharide (LPS). It is endotoxin that confers virulence
and species specicity. Variability of this surface polysaccharide
allows serologic differentiation of bacterial isolates. The lipid A
portion is mainly responsible for toxicity.7 Mycoplasmas lack a
rigid cell wall, and agents such as Treponema, Borrelia, and
Leptospira have flexible thin walls.
The outer capsule that encloses many bacteria can be well
organized, as in Streptococcus pneumoniae, or it can consist
of a diffuse layer known as glycocalyx, or slime layer, as in
Staphylococcus epidermidis. This outer capsule can prevent
phagocytosis and the glycocalyx aids in the adherence of
bacteria to tissues and to articial devices such as prostheses,
catheters,8 and intraocular lenses. The capsules of
N. meningitidis group B and the capsule of Escherichia coli
are the two best known examples. Biolm is an accumulation
of bacteria encased in an exopolysaccharide matrix, allowing
the bacteria to adhere to each other or to a solid surface. This
biolm is potentially important in ophthalmology, because it
prevents skin antisepsis.9 Biolm may also play a role in
staphylococcal adherence to plastic polymers such as intraocular and contact lenses.10 Streptococci appear to use biolms
to strengthen their adherence to mucosal surfaces.11 First the
bacteria attach to the surface and initiate cellular division to
colonize the surface. Once a threshold is reached, specic genes
are turned on to secrete an extracellular polysaccharide. The
bacteria within the extracellular polysaccharide matrix are
protected from the host immune system as well as antibiotics.
This may explain the high resistance of bacteria to antibiotic
treatment in diseases such as endocarditis or infection of

prosthetic devices.
Bacterial flagella allow bacteria to swim through liquid and
move over solid surfaces (aprocytophaga exhibits gliding motility
that may contribute to its potential to produce infections in
immunocompromised patients). Flagella are complex machinery
allowing bacteria to migrate toward specic nutrients, or away
from toxins, a process called chemotaxis. The bacteria are able
to detect a difference in the concentration of specic molecules
over a period of time. Fimbriae also aid in bacterial adherence
to tissues.12 Shorter and more hairlike than the longer flagella
that provide bacteria mobility, the mbriae function as adhesins,
mediating adhesion to specic surfaces. This is important in
pathogenesis, especially for gonococcus and E. coli. In Neisseria
gonorrhoeae, at least two surface components have been identied aiding in attachment to genitourinary cells. These components are protein II and type-specic pili. Piliated strains
attach much better than nonpiliated strains. E. coli type 1
mbriae potentiate the uptake of nutrients from and the
delivery of toxins to eukaryotic cells.13 Bacteria can shift rapidly
between a form that possesses mbriae and one that does not.
Although the mbriae help bacteria initially to establish colonization in a host, they also increase the bacterial susceptibility
to phagocytosis. Loss of the mbriae after adherence may
therefore aid in tissue invasion. Different types of mbriae vary
in specicity for the host glycoprotein receptor to which they
attach. S. pyogenes also possess a nonmbrial adhesin, protein
F, which mediates attachment of the bacteria to bronectin.
Most adhesins are lectins and have a high afnity for binding to
specic carbohydrates.
Bacteria reproduce by an asexual process called binary ssion.
Cell division begins with an ingrowth of the cytoplasmic membrane, called septal mesosomes, which eventually produces a
complete cross-wall. Bacteria lack mitotic spindle. The chromosomes are replicated and attached to the cell membrane during
cellular division. Differences in cross-wall formation and
cleavage account for the bacterial shape and arrangement.
Incomplete cleavage results in bacterial chains. Streptococci
form long chains by producing parallel cross-walls, whereas
staphylococci form clumps by beginning each new septum
perpendicular to the preceding one.14
Although much remains to be discovered about the growth of
the individual bacterial organisms, we do know that bacterial
growth depends on DNA synthesis controlled by RNA and that
it depends on messenger RNA. Under unbalanced or adverse
conditions which are frequently present in the body, DNA
synthesis can occur in the absence of RNA once the growth
cycle has already begun. Typically, at least in the laboratory, the
bacterial growth cycle has four phases: the lag phase, the
logarithmic growth phase, the stationary growth phase, and a
decline phase. Bacteria vary in their temperature requirements
for growth and can be divided into three categories according to
the temperature at which their growth or generation time is
optimal. Psychrophiles grow best at a temperature of 020.5C;
mesophiles thrive from 2040C; and thermophiles multiply
best at higher temperatures of 4090C. Most bacteria are
mesophiles; some important mesophiles can grow at temperatures below their normal range. Staphylococci grow slowly at
5C and may contaminate donor corneas in preservative media
or nonpreserved drops stored in the refrigerator. Because antibiotics may not inhibit their growth at these low temperatures,
it is recommended that corneal tissue and its storage media be
allowed to come to room temperature before transplantation.
Streptococci and Proteus vulgaris also possess the ability for
psychrophilic growth.15
Iron is an essential nutrient for bacteria. In the human body,
transferrin in the blood and lactoferrin in external secretions
Ocular Bacteriology
CLASSIFICATION OF COMMON OCULAR

BACTERIA
Identication of bacteria is a time-consuming and laborious
task and not without controversy and debate. After a pure
bacterial culture has been isolated and undergone a Gram
stain, the bacterium is further identied as to genus and species
by the results of various physiologic and biochemical tests
(Table 11.1). Commercially available kits are being used
frequently, especially in nonreference laboratories for the rapid
identication of bacteria; there are however, some who question
the accuracy and cost of such methods. Bergeys Manual is the
denite taxonomy source. Recent developments have seen a
shift from conventional phenotypic identication methods to
modern molecular techniques.25
Conventional dehydration methods utilize morphology,
cultured appearances, requirements for growth, metabolism and
biochemical activities, and susceptibility to physical and
chemical agents.
GRAM-POSITIVE COCCI
Staphylococci
Staphylococci belong to the family Micrococcaceae, which
encompasses two genera: Staphylococcus and Micrococcus.
The species in the genus Staphylococcus are divided into those
that are coagulase-positive and those that are coagulasenegative. Coagulase-positive staphylococci include S. aureus,
S. intermedius, and S. hyicus. At least 17 species of coagulasenegative staphylococci (CNS) have been identied. The bestknown member of this family and the most common bacterium
cultured from the eyelids and conjunctiva is S. epidermidis.26
The absence of coagulase should not be equated with lack of
virulence, because members of this group (e.g., S. haemolyticus)
can have pathogenic potential.27
Both coagulase-positive and -negative staphylococci are
responsible for various ocular diseases. That staphylococci are
the organisms responsible for infection in some conditions such
as dacryocystitis, keratitis, and endophthalmitis is obvious, but
their role in blepharitis, marginal keratitis, and phlyctenulosis
TABLE 11.1. Bacteria Commonly Associated with Ocular

Infections
Gram-Positive
Cocci
Micrococcaceae
Staphylococci
Coagulase-positive (e.g., Staphylococcus aureus)
Coagulase-negative (e.g., Staphylococcus
epidermidis)
Streptococcaceae
Streptococci (e.g., Streptococcus pneumonia,
Streptococcus viridans)
Bacilli
Bacillus cereus
Propionibacterium acnes
Listeria
Actinomyces
Nocardia
Gram-Negative
Cocci
Neisseriaceae
Neisseria
Branhamella
Moraxella
Kingella
Acinetobacter
Bacilli
Enterobacteriaceae
Escherichia
Shigella
Salmonella
Klebsiella
Enterobacter
Serratia
Proteus
Yersinia
Vibrionaceae
Pseudomonadaceae
Pseudomonas
Pastereurellaceae
Haemophilus
Actinobacillus
Pasteurella
CHAPTER 11
bind most of the iron.16 Lactoferrin is able to bind iron even

under the more acidic conditions that are present at sites of
infection.17 Organisms unable to obtain iron in vivo will not
proliferate, but it is clear that pathogens can circumvent this
problem. For example, the Neisseria species possess a major
iron-regulated protein (MIRP) to help the pathogen in iron
acquisition and subsequent pathogenicity.18 Other organisms
such as Branhamella catarrhalis possess iron-acquisition
proteins that aid in virulence.19 Iron availability may influence
the nature of the disease and whether it stays in one place or
disseminates; it may also determine whether the disease is
extracellular or intracellular and the site of pathogenicity.
Owing to its avascularity, the eye is iron decient, and this may
aid in its resistance to bacteria.20 Bacteria undergo phenotypic
changes in metabolism and outer membrane proteins that
enable them to acquire iron. N. meningitidis becomes more
virulent after growth in iron-restricted conditions at low pH.21
Under conditions of iron-restricted growth, pathogenic bacteria
appear to produce exotoxins.22 These exotoxins include toxin A,
elastase, alkaline phosphatase, protease, and hemagglutinin
from P. aeruginosa, a-toxin from Clostridium perfringens, and
b-toxin from Serratia marcescens.23 Bacteria can break down
almost any organic compound into usable components. For
example, some Pseudomonas species can grow on camphor and
naphthalene, and this may explain the propensity of Pseudomonas for growing in make-up.24
is more complex. McCulley and Dougherty have shown that

blepharitis can be divided into several distinct clinical forms
and that CNS, as well as S. aureus, are important in the production of staphylococcal blepharitis and seborrheic blepharitis
with a staphylococcal component.26,2831 Meibomian gland
secretions from patients with meibomian gland involvement
have an abnormality in the free fatty acid component that may
be mediated by the normal ocular flora. Assays of the most
common bacterial lid flora in normal subjects and patients with
chronic blepharitis have shown that strains of CNS isolated
from patients with a meibomian gland abnormality more frequently produced both a fatty wax esterase and a cholesterol
esterase.32,33 Tetracycline and minocycline have been shown to
decrease or eliminate bacterial flora, resulting in an improvement of blepharitis.3436 These ndings point out the important
relationship among indigenous flora, environmental factors
(e.g., temperature and pH), bacterial virulence factors, and
exoenzyme production.
Streptococci
The genus belongs to the family Streptococcaceae. Species are
classied according to the presence of certain surface antigenic
and physiologic characteristics.37 Important ophthalmic
pathogens in this group include S. pneumoniae (formerly
diplococcus), which is part of the respiratory flora, b-hemolytic
103
MICROBIOLOGY
SECTION 3
streptococci, and group D enterococci, which are part of the

enteric flora. Streptococci can be classied based on the type
of hemolysis produced on blood agar. S. pneumoniae is an
a-hemolytic streptococcus. Viridans streptococci is optochinresistant and insoluble in bile. Differentiation of the species and
the sensitivity to various antibiotics have become crucial as astreptococci have been found to be resistant to aminoglycoside
and polymyxin B and they are becoming increasingly so to
penicillin and fluoroquinolones.38 A type of nutritionally
decient streptococci has recently been described. They require
pyridoxine for growth and as a result will not grow on blood agar
or in broth without the addition of pyridoxine. Nutritionally
decient streptococci are a known cause of endocarditis and
can invade the eye as well, producing infectious crystalline
keratitis.39 Crystalline keratitis is most commonly caused by
streptococci but also occurs with other bacteria such as
nontuberculosis mycobacterium.40
GRAM-NEGATIVE COCCI
Neisseriaceae
The family Neisseriaceae includes the genera Neisseria,
Branhamella, Moraxella, Kingella, and Acinetobacter, all of
which are potential ocular pathogens. The organisms are either
diplococci or short bacilli. Their laboratory diagnosis is based
on sugar fermentation reactions or serologic techniques.41 All
members of the Neisseriaceae are oxidase- and catalase-positive
(except for Acinetobacter, which is oxidase-negative). Neisseria
species and ~50% of Acinetobacter species ferment glucose.
The differentiation of Neisseria from Branhamella can be
difcult. Branhamella will typically grow on blood agar but not
on ThayerMartin medium, and it does not ferment glucose,
dextrose, maltose, or lactose. N. gonorrhoeae are commonly
resistant to penicillin, as well as tetracycline, with increasing
resistance to fluoroquinolones.42 Gonococcal ophthalmia neonatorum is prevented by the application of topical erythromycin
ophthalmic ointment immediately after birth in newborns.
Acinetobacter species are commensal organisms of the upper
respiratory tract, skin, and genitourinary tract that can be confused with Neisseria. They are Gram-negative aerobic bacteria
that appear coccobacillary or coccal in shape. However, a negative
oxidase test result will readily differentiate Acinetobacter from
Neisseria.
Moraxella species are either bacillary or coccobacilli, forming
either pairs or short chains of pairs in smears. Presumptive
identication in smears can usually be made owing to the large
size and end-to-end conguration of Moraxella organisms,
although they may appear to be Gram-positive on thick smears.
Moraxella species grow on MacConkey agar and do not ferment
carbohydrates. Most species are susceptible to penicillin.
Moraxella are part of the normal flora of the upper respiratory
tract, but can cause conjunctivitis, keratitis, and endophthalmitis. Kingella species were formerly classied as Moraxella
and, like Moraxella, are nonmotile Gram-negative rod, coccobacillary, or diplococcal in shape and oxidase-positive. Kingella
can rarely cause endophthalmitis.43
GRAM-NEGATIVE BACILLI
Enterobacteriaceae
104
catalase-positive and oxidase-negative. They also lack cytochrome oxidase activity. Important genera include Escherichia,
Shigella, Salmonella, Klebsiella, Enterobacter, Serratia, tribe
Proteae (Proteus, Morganella, and Providencia), and Yersinia.
Escherichia coli has rarely caused endogenous endophthalmitis following septicemia.45 However, E. coli can acquire
and transmit multiple antibiotic-resistant plasmids. Serratia
was once considered to include a nonpathogen and was used to
study air currents by being released from air balloons and blown
through hospital ventilation systems.15 Today, we know that
Serratia causes infectious keratitis and endophthalmitis.
Members of the tribe Proteae, especially Proteus mirabilis,
can produce ocular disease and are typically resistant to polymyxins and tetracycline.46 On blood agar, P. mirabilis produces
gray, swarming colonies that are oxidase- and indole-negative.
Yersinia pestis causes bubonic plague, which had a devastating
effect on Western civilization in the fourteenth century.
Although now it is not commonly associated with ocular
disease, Yersinia species have been cultured from patients with
Parinauds oculoglandular syndrome.47
The family Enterobacteriaceae comprises at least 27 genera

and seven enteric groups, with more than 110 species.44 Members of this family are either motile with peritrichous flagella or
nonmotile, and they do not form spores. All members grow
both aerobically and facultatively anaerobically. The Enterobacteriaceae ferment glucose, reduce nitrates to nitrites, and are
Vibrionaceae
Members of the family Vibrionaceae are non-spore-forming
Gram-negative bacilli that are oxidase-positive. They move by
means of a polar flagellum and are capable of aerobic or anaerobic growth. Although they are rarely found to be the cause of
ocular disease, three genera, Vibrio, Aeromonas, and Plesiomonas, do sometimes cause keratitis and endophthalmitis.4850
Pseudomonadaceae
The genus Pseudomonas comprises ubiquitous Gram-negative
bacilli. The presence of cytochrome oxidase distinguishes them
from the Enterobacteriaceae. A polar flagella may be present.
The growth requirements of Pseudomonas are simple: They can
use a variety of compounds for nutrition, and some strains can
even grow in distilled water. This may explain the incidence of
Pseudomonas infections associated with homemade saline
solution and soft contact lenses and inadequately sterilized
intraocular lenses. The most common organism causing cornea
ulcers associated with contact lens wear are Pseudomonas and
Serratia. Pseudomonas can cause rapid destruction of the
cornea resulting in poor visual outcome.51
Pasteurellaceae
The bacteria of the family Pasteurellaceae are small non-sporeforming, Gram-negative bacilli. They are nonmotile and either
aerobic or facultative anaerobic. Most are fastidious, requiring
enriched media in the laboratory. The family has three genera:
Haemophilus, Actinobacillus, and Pasteurella. Haemophilus
species are the most common pathogens. They require hemin
(X factor) and nicotinamide-adenine dinucleotide (NAD). The
cell wall of Haemophilus is typical for a Gram-negative bacterium showing endotoxic activity. Many H. influenzae possess
a polysaccharide capsule and can be divided into serotypes
based on the capsular reaction.
Many other species of the Pasteurellaceae can produce
ocular disease, and they can be differentiated on the basis of
their individual requirements for hemin and NAD. A variety
of tests including indole production, urease activity, ornithine
decarboxylase reactivity, and carbohydrate fermentation of
glucose, sucrose, and lactose can also be used.52 Many Haemophilus influenzae produce b-lactamase. Effective treatment
includes new generations of cephalosporins, aminoglycosides,
and fluoroquinolones.
Actinobacillus species require carbon dioxide for growth.
The only known pathogen of the genus is A. actinomycetem-
Ocular Bacteriology
MISCELLANEOUS GRAM-NEGATIVE
BACTERIA
E. corrodens is a normal inhabitant of the human mouth and
upper respiratory tract. It can cause infection following a
human bite, and it can be the culprit in an opportunistic disease. Eikenella species are non-spore-forming, facultatively
anaerobic, moderately sized, Gram-negative bacilli. These
bacteria grow slowly on common media with CO2, and about
half of the isolates form distinctive pits on the agar. Certain
strains are mobile on moist surfaces and produce an endotoxin. E. corrodens is susceptible to ampicillin, newer penicillins and cephalosporins but resistant to aminoglycosides and
clindamycin. E. corrodens have been reported to cause keratitis
and endophthalmitis.55 Another common member of the oral
flora, Capnocytophaga, has been documented as the cause of
keratitis and endophthalmitis.5658
Although Debre rst recognized cat-scratch disease in 1931,
his ndings were not reported until 1950. Ocular involvement
typically takes the form of Parinauds oculoglandular syndrome
with a conjunctival granuloma at the inoculation site.59 Catscratch bacilli have been identied in conjunctival granulomas.
The differential diagnosis of Parinauds oculoglandular syndrome is quite long, including a number of bacterial and viral
infections. Bartonella henselae has been found to be the principal cause of cat-scratch disease.60 It is a small, pleomorphic,
Gram-negative rod.61 Treatment of cat-scratch disease is
usually supportive with spontaneous resolution over
24 months. Oral ciprofloxacin may speed resolution of the
disease.62
the eyelid and the conjunctiva.29 P. acnes is an important cause

of chronic endophthalmitis.66
Anaerobic, Gram-positive bacilli that are spore-forming
belong to the genus Clostridium. They can cause several serious
diseases, including botulism and tetanus. In addition,
C. difcile causes pseudomembranous colitis.
Listeria species are short, Gram-positive, facultatively
anaerobic (but not strictly) bacilli and they exhibit characteristic tumbling motility in suspension or in a hanging drop.
L. monocytogenes, the most common species, is catalasepositive and Voges-Proskauer-positive; it hydrolyzes esculin
but does not produce hydrogen sulde or reduce nitrite. Listeria
species are known ocular pathogens. Zaidman and co-workers
developed a rabbit model of L. monocytogenes infection and
concluded that the best treatment is a combination of penicillin
and gentamycin.67 Listeria can also cause endogenous
endophthalmitis.68
Actinomyces and Nocardia

Actinomyces species are facultatively anaerobic or strictly
anaerobic Gram-positive bacilli that are usually arranged in
hyphae but can fragment into short bacilli. A. israelii, the most
common opportunistic species, grows on blood agar enriched
with vitamin K. The organisms can cause a chronic canaliculitis.69 Penicillin remains the most effective treatment.
Similar in appearance to Actinomyces and almost indistinguishable on Grams stains is the genus Nocardia. Nocardia
species are strict aerobic bacilli that are Gram-positive, yet
they may appear to be Gram-negative with intracellular
Gram-positive beads. They have a cell wall similar to that of
mycobacteria and are acid-fast with weak acids, which helps to
distinguish them from Actinomyces species. Members of the
Nocardia are catalase-positive and grow on nonselective media.
Norcardia is a known cause of kerititis and the treatment of
choice is amikacin.70 Endophthalmitis caused by Norcardia has
poor prognosis.71
CHAPTER 11
comitans, which can cause endophthalmitis.53 Pasteurella

infections, which are usually transmitted through contact with
animals that are carrying the bacilli, can cause conjunctivitis,
corneal ulceration, and endophthalmitis.54
ANAEROBIC GRAM-NEGATIVE BACILLI

Anaerobic Gram-negative bacilli are a group of non-sporeforming bacteria that comprises part of the normal anaerobic
oral and intestinal flora. Bacteroides fragilis is the most commonly isolated organism. Unlike most anaerobes, B. fragilis is
resistant to many antibiotics, including penicillin. Cuchural
reviewed the antibiotic sensitivities of a number of strains of
B. fragilis.63 Resistance rates to imipenem and ticarcillinclavulanic acid were 0.2% and 1.7%, respectively. No isolates
were resistant to either metronidazole or chloramphenicol.
The rate of resistance to clindamycin was 5% and to cefoxitin
11%. B. fragilis rarely cause ocular infection, with one case of
endophthalmitis reported.64
GRAM-POSITIVE BACILLI
Gram-positive bacilli are comparatively large spore-forming
bacilli that grow on nonselective media producing nonhemolytic
rapidly growing colonies. They are ubiquitous and have been
known to cause a severe endophthalmitis after trauma has
occurred.65 Bacillus cereus is the most common pathogen.
Vancomycin, clindamycin, and the aminoglycosides are usually
the drugs of choice.65
The most important of the non-spore-forming Gram-positive
bacilli are the genera Corynebacterium and Propionibacterium.
The organisms are small, nonmotile, and catalase-positive, and
they ferment carbohydrates producing lactic acid (Corynebacterium) or propionic acid (Propionibacterium). Propionibacterium species are anaerobic and are a common isolate from
MYCOBACTERIA
Mycobacterium tuberculosis and M. leprae remain two of the
most prevalent and serious causes of infections worldwide.
They are acid-fast, although M. leprae is more sensitive to
decoloration. The growth of these nonmotile slender rods is
slow, with some species taking 26 weeks, although growth of
fast-growing species can occur in 35 days. Runyon classied
mycobacteria into four groups based on their rate of growth and
chromogenicity. In ophthalmology, it is probably more practical
to divide mycobacteria into two groups: M. tuberculosis and
atypical mycobacteria. Atypical mycobacteria (especially
M. fortuitum and M. chelonei) are emerging as a frequent cause
of keratitis following refractive surgery. These bacteria are
sometimes difcult to diagnose and treat, with potentially poor
visual outcome.72,73 Topical amikacin has been effective in the
treatment of corneal ulcers. Newer generations of fluoroquinolones, such as gatifloxacin, have been shown to be effective
against M. chelonae in a rabbit model.74
MOLLICUTES
Mollicutes are a class of microorganism bounded only by a
membrane. The two most important genera are Mycoplasma
and Ureaplasma. Three pathogen strains have been identied:
M. pneumoniae, M. hominis, and Ureaplasma urealyticum.
They can be differentiated by their ability to metabolize glucose
(M. pneumoniae), arginine (M. hominis), or urea (U. urealyticum).
M. pneumoniae causes pneumonia. M. hominis causes post-
105
MICROBIOLOGY
partum fever.75 U. urealyticum is associated with urethritis in
men and lung diseases in premature infants.76 Mycoplasmas
resemble chlamydiae, rickettsiae, and viruses in passing through
450-nm lters but, like bacteria, they are Gram-negative,
able to grow on articial media, and capable of dividing by
binary ssion. Erythromycin and tetracycline are usually
effective, although some M. hominis are resistant to erythromycin and some ureaplasmas are resistant to tetracycline.75,77
Mollicute-like organisms (MLO) are found in chronic uveitis,
especially gastrointestinal tract-associated disease.78
SECTION 3
INFECTION OF THE HOST
106
Bacteria produce a variety of ocular diseases. Bacterial conjunctivitis and bacterial keratitis are commonly seen. Endophthalmitis presents a challenging clinical problem. Blepharitis
in its various forms may constitute an imbalance in the
normal relationship between bacteria and the skin of the eyelid.
The exact roles of CNS and their toxin production, and of
Propionibacterium acnes in meibomian gland dysfunction
continue to be studied and dened. Infections of the periocular
tissue include canaliculitis, dacryocystitis, and preseptal
and orbital cellulitis. Bacteria also can have remote effects
such as syphilitic interstitial keratitis and mycobacterial
phlyctenulosis.
The virulence of a pathogenic organism depends on its
potential to produce disease. One important factor is its ability
to adhere to mucosal surfaces and to enter epithelial cells.
Invasive properties are carried in various ways in plasmids,
bacterial phage, and DNA segments in the bacterial chromosome. These properties can be exchanged between bacteria,
rendering noninvasive bacteria invasive. Characteristics of
bacteria important in ocular infections include: virulence of
the organism, the invasiveness of the organism, the number
of organisms entering the host, and their site of entry. Certain
extracellular enzymes may be important in the establishment of
infection and in its spread through tissues. These include
collagenase (C. perfringens), coagulase (staphylococci), hyaluronidases (staphylococci, streptococci, clostridia, pneumococci),
streptokinase or brinolysis (hemolytic streptococci), hemolysins and leukocidins (streptococci, staphylococci, clostridia,
Gram-negative rods), and proteases (neisseriae, streptococci)
that can hydrolyze immunoglobulins, such as secretory IgA.75
In blepharitis, staphylococci and P. acnes produce lipases and
esterases.
The host determines the effect of many virulence factors.
That is, certain characteristics of the host can influence the
development of disease. For example, the hosts age, use of
drugs, and sexual habits can all determine the effect of virulence
factors. The use of contact lens or surgical trauma increase
the risk of ophthalmic disease. Blepharitis, dry eye states, canaliculitis, chronic nasolacrimal duct obstruction, and previous
ocular disease also increase the risk. Damaged epithelium in
the cornea is particularly susceptible to bacterial adherence;
bacteria adhere to the epithelial edge rather than the bare
stroma.79 Tissue injury results from: the direct action of the
bacteria, from microbial toxins, from indirect injury, from
inflammation, or from immunopathologic processes. In response
to an injury, polymorphonuclear cells, as well as macrophages
and lymphocytes, enter the site. Tissue fluids provide plasma
proteins, including immunoglobulins such as IgG, complement, and properdin. The primary mediators of inflammation
include histamine, tumor necrosis factor, cytokines, leukotrienes and prostaglandins. The phagocytic cells play a key role
in the interaction with the microorganism, ingesting and killing
bacteria. The inflammatory process releases chemokines which
attract additional inflammatory cells.
ADHERENCE, COLONIZATION, AND INVASION

Cellular microbiology is a rapidly developing eld that deals
with the interaction of bacteria and their host cells. Epithelial
cells with their tight cellular junctions act as a barrier to
bacterial adherence, penetration, and the entry of soluble
toxins. Epithelial cells may respond to bacterial adherence by
secreting cytokines, causing a major cytoskeletal rearrangement
and playing an important role in the mucosal immune
response. However, the relationship between the host and the
potential pathogen is complex and still incompletely understood regarding why some bacteria are invasive and others
colonize the cell surface. Some produce exotoxins that destroy
host cell functions, whereas others utilize the host cell to
advance their pathogenic potential.80
Microbial adhesion to host tissue is a primary event in
colonization and an important stage in microbial pathogenesis.
Adhesive ligands in bacteria range from rod-like structures (pili
or mbriae) to outer membrane proteins and polysaccharides.
Individual bacteria may possess multiple adhesins that target
distinct host cell molecules and deliver diverse signals resulting
in extracellular location or internalization. Both the nature
and the density of the target receptor on the host cell may be
determining factors in the outcome of the bacteriahost
interaction.80
The invasion of mucosal surfaces and ocular tissues by
bacteria occurs in several steps. First, bacteria must establish
themselves in close proximity to the ocular surfaces, such as
the lids and lashes. This, by the way, is why the cleansing and
isolation of these surfaces is so critical in ocular surgery.
Second, the bacteria must avoid being swept away, which is one
of several reasons why patients with severely dry eyes are at
increased risk of infection. Next, bacteria must acquire essential nutrients for growth, especially iron, and be able to replicate
at a rate sufcient to maintain or expand their population.
Finally, the bacteria must resist local host defenses.
Association, that is localization of bacteria on a surface, must
take place before adherence can occur. Most bacteria and host
tissue carry negative charges. In order to overcome the repelling
forces, many mechanisms are utilized by the bacteria to adhere
to the host surface. This may be as simple as possessing
hydrophobic forces which help adhere to host tissue. Motility of
bacteria may enhance association. Bacteria may associate with
mucus or exudates, forming noncovalent bonds. Chemotaxis
may help bacteria to penetrate the mucous barrier, thus
enhancing contact with receptors on the epithelial surface.80
Bacterial attachment is essential in order for colonization to
occur in environments with a surface exposed to a fluid flow.
Adhesion of bacteria to the epithelial surface depends upon
adhesins, the complex polymers on the bacterial surface. The
presence of pili, hair-like appendages that extend from the
surface of the cell, aid in the adhesion of bacteria to host cells.81
For example, E. coli have pili that allow the bacteria to adhere
to the epithelial cells in the intestinal wall.82 The presence of
mbriae assist in bacterial adhesiveness.83 These are frequently
present on Gram-negative organisms. A variety of bacteria
produce adhesins that tend to be outer membrane proteins.
Outer membrane proteins, as well as mbriae, aid in adhesion
of N. gonorrhoeae to epithelial cells. Staphylococci and
streptococci can adhere to epithelial cells and thus colonize
skin and mucous membranes.110 The important components
of mbriae consist of lipoteichoic acid (LTA), protein F, and
M protein.83 Lipteichoic acid and protein F adhesion to
epithelial cells are mediated by bronectin. The M protein
prevents phagocytosis.81 S. aureus produces a surface protein
with specic afnity for bronectin.84 A variety of streptococci
and staphylococci species can bind bronectin, probably
through afnity with their cell wall LTA. The presence of
bronectin on the cell surface appears to enhance bacterial

adhesion as well.85 LTA can interfere with the killing or
phagocytosis by polymorphonuclear leukocytes.86 Some isolates
of S. epidermidis can inhibit the bacterial phagocytic activity of
neutrophils, independent of adherence. This inhibition of
neutrophils may represent another virulence factor.87
Adherence of P. aeruginosa to the corneal epithelium may be
the rst step in the pathogenesis of infection.8890 Pseudomonas
adheres to the basal epithelial cells through the interaction of
a specic adhesion-receptor. In order for bacterial adherence to
occur, several steps must take place. First, van der Waals forces
produced by surface molecules overcome the normal repulsive
forces of two similarly charged cells.91 Then, once the cells
become close enough, hydrophobic binding holds the bacteria to
the surface, and strong bonds form between the exopolysaccharides of the bacteria and the substrate glycoprotein of
the target cell. The signicant differential adherence between
basal and nonbasal corneal epithelial cells is probably the reason why supercial trauma or epithelial cell damage allows
Pseudomonas infections to develop.92 This may play a signicant role in contact lens-associated Pseudomonas keratitis.
Using a rabbit model, Koch and associates showed that a
bacterial suspension of P. aeruginosa alone caused no inflammation but that corneal infection developed in 11 of 14 eyes
wearing new or worn contaminated soft contact lenses.93
Trancassini and associates demonstrated that strains of
P. aeruginosa that produce alkaline protease and elastase adhere
better.94 Bacterial adherence may also depend on nonbacterial
factors. Deighton and Balkau investigated the adherence of
strains of S. epidermidis to glass and plastic material.95 They
found that the degree of adherence depended mainly on the
growth media; adherence was enhanced by the addition of
glucose or oleic acid and it was inhibited by serum. After
attachment takes place, penetration of the epithelial cells must
occur. LPS core with an exposed terminal glucose residue
expressed in P. aeruginosa has been shown to highly correlate
with the level of adhesion to epithelial cells.96 In the case of
E. coli, this is a process similar to phagocytosis.97
When they are present, bacterial cell wall capsules are
important virulence factors.98 While cell wall capsules are more
commonly seen in Gram-negative bacteria, encapsulated
staphylococci may be seen in vivo.99 The primary virulence
factor of H. influenzae surface antigen, the type b capsular
polysaccharide, is polyribosylribitol phosphate (PRP).100 Some
bacteria, such as Bacteroides species, become encapsulated
during an inflammatory process, further increasing their
pathogenicity as a result.101 The capsule thus formed inhibits
phagocytosis by covering and thus making the recognition sites
of opsonins (C3b and IgG) inaccessible to phagocytic cells.102
M-protein inhibits opsonization and impairs complement
activation and binding of C3b to the bacterial cell wall.103,104
Surface sialylation of the bacterial capsule also helps microorganisms to resist host defenses.105 In a mouse model of
Campylobacter infections, Pei and Blaser demonstrated that
virulence was enhanced when S-protein was present on the
bacterial cell surface as a capsule.106
Bacterial glycocalyx also may aid in colonization and infectivity by protecting the bacteria from antibiotics and from the
hosts immune system and phagocytic cells.107 Glycocalyx
production is important in the adhesion of certain P. aeruginosa
strains to respiratory tissues.108 For staphylococcal strains, protein A and clumping factor may be important mediators of
adherence.109 Protein A interferes with opsonic activity of antibodies, because it binds to the Fc portion of IgG (except IgG3),
and to a lesser extent, IgM and IgA2.110 Streptococci also carry
an Fc binding protein on the cell wall and therefore evade the
natural host defense mechanisms.111
The ability of specic bacteria to adhere to the sites at which

they produce clinical disease has been shown in various situations, including S. pneumoniae to human pharyngeal
epithelial cells, S. pyogenes to pharyngeal epithelial cells, and
E. coli to bladder epithelium. S. aureus, P. aeruginosa,
H. influenzae, and S. pneumoniae adhere to mucus in the
respiratory tract. S. aureus, S. pneumoniae, and P. aeruginosa,
three of the most common causes of corneal ulceration, exhibit
markedly greater adherence to human corneal epithelial cells
than do other bacterial species.112 S. aureus produces a number
of cell surface proteins that bind to host protein. These include
bronectin, brinogen, vitronectin, bone sialoprotein, thrombospondin, collagen, IgA, elastin, prothrombin, plasminogen,
laminin, and mucin.113 Protein A binds IgG in such a way that
F1-receptors on phagocytic cells cannot bind to the F1 protein
of the immunoglobulin. After establishing adhesion, some
bacterial pathogens enter epithelial cells by endocytosis.
Intracellular invasion provides a new source of nutrients and
affords protection from some host defenses; however, the
bacteria must survive inside an endocytic vacuole, and, while
exposed to products such as lysozyme, they must multiply and
spread to other cells.114 Many pathogenic microbes may invade
the host by inducing their own endocytosis. This phenomenon
has been designated as parasite-directed endocytosis. Although
still poorly understood for most pathogens, it is thought that in
the case of most bacteria, this represents biologic mimicry, with
the bacteria producing a molecule that resembles a natural host
ligand for which there is a host cell receptor.115 Organisms such
as Mycobacterium, Actinomyces, Corynebacterium, Listeria,
and Francisella species contain large quantities of structural
lipid that protects them from digestion by the lysosomes of
phagocytes, probably because of their ability to scavenge oxygen
radicals.116
The virulence of bacteria also depends on their ability to
produce enzymes that are directed at host defenses. Coagulase
produced by staphylococci forms a brin clot from brinogen,
thus protecting the bacteria from phagocytosis. Streptococci
can produce a streptokinase that dissolves brin clots and
allows further spread of the bacteria. Streptokinase activation of
plasminogen produces brinogen degradation products.117
Whitnack and co-workers showed that the binding of brinogen
and brinogen degradation products to M-protein enhances its
antiopsonic property.118 S. pneumoniae pneumolysin inhibits
polymorphonuclear leukocyte chemotaxis and the ability to kill
opsonized pneumococcus.119,120 Neuraminidase may also be an
important virulence factor of S. pneumoniae. Neuraminidase
might alter glycoproteins on the ocular surface, thus enhancing
bacterial attachment. Pneumococci can adhere to corneal
epithelial cells in vitro.112 Hyaluronidase digests hyaluronic
acid, which is an important tissue cement and aids in the
spread of some streptococci and staphylococci. Leukocidin,
produced by some staphylococci and streptococci and some
bacilli, disintegrates neutrophils and tissue macrophages.
Catalase destroys the hydrogen peroxide present in lysosomes.
N. gonorrhoeae, N. meningitides, H influenzae, and
S. pneumoniae produce an IgA protease that destroys
immunoglobulin IgA1.81 Other bacteria produce cytolysins,
such as hemolysins that kill red blood cells or leukocidins,
that lyse leukocytes.81 Streptococci group A produce
streptolysin O and S, which lyse red blood cells and are lethal
for mice.121 Endotoxin activity is an important aspect of Gramnegative virulence. P. aeruginosa produces an elastase, alkaline
protease, exotoxin A, and LPS endotoxin. The P. aeruginosa
exotoxin A has a cytopathic effect, and alkaline protease is
active against collagen.122128 Gram-positive bacteria, although
they do not contain LPS, do have PG that can lead to vascular
dilation and hypotension similar to LPS but not as severe.
CHAPTER 11
Ocular Bacteriology
107
MICROBIOLOGY
Burns and associates have shown that a metalloproteinase
inhibitor (HSCH2) inhibits P. aeruginosa elastase and that, in a
rabbit model, delayed the onset of corneal melting and
perforation.129
SECTION 3
HOST DEFENSES
108
Several defense systems are important in the prevention of

microbial infection. The rst barrier consists of the skin and its
indigenous flora that help to create a milieu inhospitable to
most pathogens. Lactic acid and fatty acids in sweat and
sebaceous glands serve to lower the pH to a point at which
most pathogenic bacteria will not survive. The mechanical
flushing action of the lids and tears, in addition to antibody,
lactoferrin, b-lysin, and lysozyme present in tears, serve as the
next major barrier to infection. The conjunctiva and mucous
membranes are important in preventing bacterial adherence
and in allowing natural antibodies such as IgM, humoral
immunity, and cell-mediated immunity (CMI) access to the
ocular surface.
NONSPECIFIC DEFENSES
The normal conjunctiva contains all immunologic components
and high levels of inflammatory cells (~300 000 per mm2).130
Although immunoglobulins and complement system are the
most important factors in the hosts defense against bacteria,
other factors include bronectin, C-reactive protein, lysozyme,
and transferrin play a signicant role. Immunoglobulins G
and M (IgG and IgM) have the greatest bactericidal activity,
whereas IgA is very effective in restricting bacterial adhesion
on mucosal surfaces.131,132 These components contribute to
specic as well as nonspecic defense mechanisms. Tears
usually contain IgA, IgE, IgG, and complements. Secretory IgA,
usually in conjunction with complement activated by the alternate pathway, can be bacteriolytic.133,134 IgA has an important
role in preventing infections as evidenced by an increased
incidence of staphylococcal infections observed in atopic disease with its associated defects in IgA and CMI.135
The complement system is also very important in defending
against bacterial infections. The main outcomes of complement
activations are: (1) lysis of bacteria, (2) production of inflammatory mediators, (3) opsonization of organisms for phagocytosis,
and (4) facilitate antibody-mediated immune responses.81
Complement assists phagocytic cells by depositing an opsonic
protein (C3b) on the bacterial surface that then interacts with
specic receptors on the phagocytic cell surface. It is clear that
phagocytic killing by leukocytes is an important defense
mechanism against bacterial infection, because patients with
abnormalities of polymorphonuclear leukocyte function are
susceptible to recurrent or persistent infections.136 Pneumolysin
can activate the classic complement pathway, whereas the
alternate pathway may be activated by the PG of group A
streptococci or the TA of S. pneumoniae.4,137,138 In Gramnegative infections, complement can be directly bactericidal
through the assembly of a membrane attack complex (C5b-9)
that can lyse susceptible Gram-negative bacteria. Complements
are also chemotactic, drawing leukocytes into the cornea. Typically, an antigenantibody complex activates the complement
reaction, but interaction of bacteria directly with C1q can also
activate complement.139,140 Bacterial cell wall components such
as LPS can activate the alternate complement pathway.141
Through its interaction with specic antibody, LPS can activate
complement via both the classic and alternate pathways; LPS
alone activates the alternate pathway.142 Deposition of
LPSantibody complexes may cause ring inltrates in Gramnegative corneal infections.143
Neutrophils are the primary cells found at the site of bacterial corneal infections.144 During phagocytosis they release
prostaglandins, which increase vascular permeability and
induce degranulation of mast cells and basophils. Mast cells
in turn release histamine, eosinophil chemotactic factor,
prostaglandins, and SRS-A. Neutrophil lysosomal products
include cationic proteins, acid proteases, and neutral proteases.
The cationic proteins increase vascular permeability and are
chemotactic for mononuclear phagocytes. The acid proteases
degrade basement membrane, and neutral proteases degrade
brin, elastin, and collagen. Neutrophils also contain lysozyme,
hydrolytic enzymes, collagenase, lactoferrin, and toxic nitrogen
oxides.145 Antimicrobial neutrophil peptides (defensins) have been
isolated in the tear lm.146 Cullor and associates have demonstrated that neutrophil defensins possess both bacteriostatic
and bactericidal activity against various ocular pathogens.147
Lysozyme is an enzyme that can lyse certain bacteria by
acting as a muramidase to cleave the glycosidic bond of the
N-acetylmuramic acid residues in the bacterial cell wall.148
Lysozyme makes up 40% of the tear protein, with levels in
normal adults ranging from 1.3 to 1.4 0.6 mg/mL.149,150 The
lysozyme content in tears decreases with age and decreases in
several eye diseases, including keratoconjunctivitis sicca, chronic conjunctivitis, and nutritional deciency with xerosis.151153
Lysozyme is primarily effective against saprophytic Grampositive bacteria such as micrococci. Some coagulase-positive
staphylococcal strains can produce lysozyme, which may help
them overcome any inhibitory effect of the indigenous flora.154
Lysozyme may also interact with a recently described substance
called lysostaphin. Certain staphylococcal strains produce
lysostaphin. In contrast to lysozyme, lysostaphin inhibits many
strains of staphylococci including S. aureus, but it does not
inhibit micrococci.155 Lysozyme appears to increase the
antistaphylococcal activity of lysostaphin from 16- to 200fold.156 In Gram-negative bacteria, lysozyme aids the action of
complement on the cells cytoplasmic membrane.157
HUMORAL IMMUNITY
Normal tears contain antibodies against bacteria. Local antibody synthesis takes place in the lacrimal gland, but some
antibodies originate from lymphocyte sensitization in the
mucosal immune system.158 In P. aeruginosa infections, Berk
and associates showed that mice develop IgM and IgG antibodies corresponding to their ability to recover from corneal
infection.159 Antibodies attach to the outer membrane proteins
(porin protein F) and protect the cornea.160 IgA at the ocular
surface can prevent bacterial attachment to epithelial cells.150
However, not all antibody responses are benecial to the host.
Grifss and associates have reported that serum IgA directed
against N. meningitidis blocks the lytic activity of IgG and IgM
for this organism.161
Complement and opsins, discussed earlier, are necessary for
the adherence of bacteria to polymorphonuclear leukocytes.
Complement can destroy bacteria directly or by causing
chemotaxis of neutrophils. Antibody-coated bacteria may be
unable to adhere to corneal epithelium. Antibodies can also
neutralize the exotoxins released by some bacteria.
CELL-MEDIATED IMMUNITY
CMI contributes to the defense against microorganisms. When
a T lymphocyte becomes sensitized to a bacterial antigen, it
releases a soluble factor (lymphokine) that can help to activate
the macrophage and localize it at the site of an infection.
The sensitized lymphocyte can also release chemotactic factors
for macrophages, neutrophils, basophils, and eosinophils.
Cytokines are released by inflammatory cells and have multiple

effects, such as activation and differentiation of other inflammatory cells, chemotaxis, and cytotoxic in bacteria. Upon entry
of the invading bacteria, the antigen is engulfed by macrophages. The antigen is processed and presented on the cell
surface to the T lymphocytes. Once recognized by the
T lymphocytes, the lymphocytes are activated and start to
proliferate. PG, TA, and other cell wall components may be
polyclonal activators of both B and T cells. Polyclonal activation
of human lymphocytes may be useful to the host as a mechanism of resistance to infectious diseases; however, the process
could also have adverse effects by triggering or perpetuating
chronic inflammatory disease. Studies in animals indicate
that immunization with the capsular polysaccharide provides
a T-cell-dependent immunity to abscess development when
challenged with Bacteroides fragilis. Also, it appears that the
killing of B. fragilis is T-cell dependent.162 Group A streptococcal cell membranes appear to enhance certain T-cell
functions.163
this is the minimum number of samples that should be taken.

Whenever there is a large, fulminating ulcer or sufcient
material is available, separate scrapings of the ulcer should be
done for each plate. In our laboratory, we have had more success
using separate plates for each site cultured. Although it requires
more plates and labeling, this technique facilitates the isolation
and identication of individual pathogens, particularly in
polymicrobial infections.
In cases of endophthalmitis, both aqueous and vitreous
should be cultured.165 Compared to aqueous fluid or fluid from
the vitrectomy cassettes, undiluted vitreous provide the highest yield of positive cultures. If there is sufcient material,
smears should also be performed for Gram-stain for bacteria or
KOH stain for fungus. Although smears may not always be consistent with culture results, they may nevertheless be invaluable
in conrming a bacterial process in cases of culture-negative
endophthalmitis. A positive Gram-stain is useful information;
whereas a negative Gram-stain result had little correlation with
culture results.165
DIAGNOSTIC TESTS
MEDIA
The diversity of infectious processes that involve the eye makes

it necessary for the ophthalmologist to be aware of a variety of
basic microbiologic techniques. Jones and associates have
written what still remains the most comprehensive approach to
ocular laboratory diagnosis.164 Both the ophthalmologist and
laboratory must be knowledgeable in determining which
bacteria are considered pathogens in ocular disease versus
contaminants or normal flora. Frequently, the material obtained
from cultures is small and must be inoculated onto media
immediately. The specic technique to be used and the cultures
taken will depend on the clinical diagnosis and setting; it is
useful to have protocols written out beforehand in order to
avoid needless errors. It is also helpful to maintain a culture
tray that is readily available. Routine culture media can be
stored in a refrigerator, but only fresh plates of media should
be used. Media that appear dry or that have pulled back from
the edges of the Petri dish should be replaced. Plates should be
brought to room temperature before inoculating them with
clinical material.
The method used to collect a specimen depends upon the site
and etiology of the infection. Cultures of the cornea, conjunctiva, and eyelids can be done either with the Kimura platinum
spatula or with swabs. For eyelid cultures, our procedure is to
use a moistened calcium alginate swab. The use of a moistened
swab helps to prevent drying of the material and to create a
capillary attraction may enhance bacterial pickup. Furthermore,
the moistened swab allows release of the material over several
plates and avoids cutting into the media surface, which can
make recognition and isolation of colonies more difcult. If the
blepharitis is ulcerative, the platinum spatula may be used to
remove the brin scale, and this material may be cultured as
well. In cases of conjunctivitis, we will again use the swab
moistened in sterile saline or nutrient broth, reserving the
spatula to obtain specimens for cytology.
In cases of suspected microbial keratitis, a four-step approach
to the culture is taken. First, a moistened swab is used to
culture the ulcer base. Next the ulcer is scraped, usually with
a platinum spatula, but in some cases a Bard-Parker No. 15
blade or a Beaver blade may be required to obtain sufcient
material. The material obtained should then be immediately
inoculated onto culture media transferred to a moistened swab
and streaked onto appropriate media. The spatula is used to
obtain material for smears and slides, and nally a moistened
swab is again applied to the ulcer in order to pick up any bacteria brought to the ulcer surface. It should be emphasized that
Media can be divided into two broad types: broad-spectrum and

selective. All of the media used in ophthalmology are enriched
and nonselective, because selective media contain chemical
substances or antibiotics to inhibit the growth of all but the
desired organism. The basic media used for culture and
identication of most ocular bacterial pathogens are listed in
Table 11.2.
CHAPTER 11
Ocular Bacteriology
BLOOD AGAR
Blood agar consists of a Brucella agar base with a peptic digest
of animal tissue, dextrose, and yeast extract. Most aerobic bacteria (and fungi) will grow on it except for the more fastidious
pathogens, especially Neisseria, Haemophilus, Moraxella, and
atypical mycobacteria. When incubated under anaerobic
conditions, most anaerobic organisms will grow on blood agar
as well but it must be supplemented with hemin, vitamin K,
and sometimes cysteine. It also has the advantage of revealing
the hemolytic reaction of the organism. This is the best single
general purpose culture medium for the diagnosis of ocular
pathogens.
CHOCOLATE AGAR
Chocolate agar is prepared by using GC agar base and bovine
hemoglobin. Growth factors, hemin (X factor), and nicotinamide adenine dinucleotide (V factor) are added to the
TABLE 11.2. Bacterial Culture Media

Routine
Blood agar
Chocolate agar
Enriched thioglycolate broth
Sabouraud dextrose agar (for fungi)
Optional (Depends on Availability and the Clinical Situation)
Brain heart infusion broth
Lowenstein Jensen medium
Middlebrook agar
109
MICROBIOLOGY
agar.166 These nutrients are essential for the growth of
Haemophilus, N. gonorrhoeae, N. meningitidis, and Moraxella.
When one suspects N. gonorrhoeae, ThayerMartin medium
should also be used. ThayerMartin medium contains 3 mg of
vancomycin, 7.5 mg of colistin, and 12.5 U of nystatin per
milliliter of agar to inhibit other bacteria or yeasts that could
inhibit the growth of gonococcus. However, ThayerMartin
medium is only a supplement to and not a replacement for
chocolate agar, because potentially nongonococcal strains of
Neisseria may be inhibited by the added antibiotics. Incubation
of ThayerMartin plates should be done in an atmosphere
containing 310% CO2.
SECTION 3
BRAINHEART INFUSION BROTH
110
A highly nutritious and buffered liquid is a useful adjunct to

solid media for several reasons. Material picked up by the swab
but not released onto the solid agar thus has an opportunity to
grow. Any antibiotics or other inhibitors of bacterial growth
will be diluted and, therefore, have less effect. Inoculation of
broth also allows the use of antimicrobial removal devices, such
as those developed by Osato. However, they do not permit one
to conrm that growth is occurring along the inoculum streak
nor do they allow one to quantify the amount of growth.
Other useful selective media include eosin methylene blue
(EMB) agar and MacConkey agar. These media are primarily
useful for the isolation of Gram-negative bacteria. Methylene
blue agar inhibits Gram-positive bacteria and has carbohydrates
that can be fermented by Escherichia coli and other Gramnegative bacteria. MacConkey agar contains the carbohydrate
lactose, a fermentable carbohydride, as well as bile salts, which
inhibit the growth of Gram-positive bacteria.
Anaerobic cultures are routinely done in thioglycollate broth
without indicator. The broth is supplemented with hemin and
vitamin K. At times, aerobes also grow in thioglycollate, usually
near the surface; anaerobes, on the other hand, grow below the
surface. A disadvantage is that an anaerobic pathogen can be
overgrown by other anaerobic bacteria or by aerobic bacteria.167
In cases in which anaerobic cultures are especially important,
such as a possible P. acnes endophthalmitis or chronic canaliculitis, other anaerobic media should be used. Prereduced
anaerobically sterilized media (PRAS), anaerobic blood agar, or
chocolate agar can be used.168 In cases in which one obtains a
fluid sample, such as in endophthalmitis, the sample can
be injected through the rubber stopper into a chopped meat
glucose medium. Aerobic and anaerobic blood culture bottles
can also be used.
LowensteinJensen medium is used for the isolation of
mycobacteria. It contains ribonucleic acid adequate for microbacterial growth, along with penicillin and nalidixic acid, which
inhibit contaminating organisms. Nocardia species will also
grow on this medium.169 Middlebrook agar are used for the
detection of mycobacteria, and may be more sensitive than
LowensteinJensen medium.170 These two media are especially
important in patients diagnosed with an infectious keratitis
following refractive surgery given that nontuberculous mycobacteria are common causes of the infection.73 Many of the
Mycobacterium chelonaeMycobacterium abscessus complex
will also grow on blood agar media.
Proper conditions during incubation are essential. Aerobic
and anaerobic cultures should be kept at 35C. Blood and
chocolate agar should be incubated under higher carbon dioxide
tension (310%). Routine cultures should be kept for 1 week,
but anaerobic cultures should be incubated for 2 weeks. Fungal,
actinomycete, and mycobacterial cultures should be held for
8 weeks. Mycobacteria grow best under a carbon dioxide tension
of 510%.
STAINS
While the results of smears may not always be consistent
with the nal cultured organisms, smears are an important
component of bacterial diagnosis. Although one could base
initial therapy on Gram stain ndings, given the incongruity
between smear and culture results, it would seem most prudent to use the smear results to add to therapy rather than
delete from the standard initial treatment. Smears are also useful in identifying polymicrobial processes in which one type of
bacteria may inhibit or delay the identication of other bacterial pathogens. Furthermore, smears may identify the presence of organisms that do not appear on culture for days or
even weeks. Smears are invaluable whenever cultures prove to
be negative, especially in patients who have previously received
antibiotics. In the laboratory, stains are essential in order to
identify cultured bacteria.
The proper preparation and examination of smears requires
both experience and patience. Smears are prepared by spreading
a thin lm of the specimen over a dened area of the slide.
Smears that are too thick can obscure many important
details. Smears spread out over an entire slide increase the
length of time required to completely examine the slide and
increase the possibility of overlooking pathogens. The slide
should be free of lint and ngerprints, air-dried, and gently
heat-xed. One must look at a large number of slides in order
to be able to distinguish between the occasional bacteria of
the normal flora and an actual pathogen. In repertory results,
microbiologists should report only cell morphology and a Gram
reaction, not whether they think they see pathogens or
normal flora.
One of the oldest and most commonly used stains is the
Gram stain. As we have discussed earlier, this is a differential
stain in that bacteria are either Gram-positive (blue-purple) or
Gram-negative (orange-red). There are several theories to explain
why bacteria respond differently to a Gram stain. One theory
suggests that crystal violet and iodine form a chemical complex
in the bacterial cytoplasm. Alcohol in the staining process may
dissolve lipid, allowing the crystal violetiodine complex to leak
out of the cytoplasm. Gram-negative bacteria with their high
lipid content in the cell wall would therefore lose more stain
than would Gram-positive bacteria. The cell walls of Grampositive bacteria are less permeable to small molecules than are
those of Gram-negative organisms. PG in the cell wall of Grampositive bacteria may trap the crystal violetiodine complex.
Because Gram-negative bacteria have less PG, they would trap
considerably less stain.171 In any case, knowing whether an
organism is Gram-positive or Gram-negative continues to have
important diagnostic and therapeutic implications. Variable
Gram staining may occur with excessive decolorizing, with
smears that are too thick, or with older cultures. Gram-positive
organisms may appear Gram-negative if there has been previous
antibiotic treatment, leukocytic destruction, or excessive
heating of the slide.169 The safranin counterstain can replace
crystal violet, thus the slide should not be counterstained for
a prolonged time. Giemsa staining is not as important in
bacterial infections, because it has no differential value, but its
ability to delineate cellular types and detect inclusion bodies or
multinucleated giant cells make it an important investigative
tool in ocular diagnosis. Bacteria generally stain blue. The
BrownHopps stain is a Gram stain modied for tissues.
Aniline can be added to the Gram stain to improve identication of actinomycetes.
Acridine orange (AO) stains all DNA and RNA regardless
of organism. AO has recently received renewed interest owing
to its ability to stain Acanthamoeba species. The AO stain is
very good for bacteria too and is more sensitive than a Gram
Ocular Bacteriology
HIGH-TECHNOLOGY DIAGNOSTIC
METHODS
Newer diagnostic methods may be used increasingly in
bacteriologic diagnosis. Antigen detection tests have been developed utilizing a variety of techniques, including counterimmunoelectrophoresis (CIE), coagglutination (CoA), latex
agglutination (LA), enzyme immunoassay (EIA), enzyme-linked
immunosorbent assays (ELISA), radioimmunoassay (RIA),
solid-phase immunofluorescence and fluorescence polarization
immunoassay (FPIA), and immunoblotting (Western blot).
These tests have tremendous potential and to date have been
useful in detecting cerebral spinal fluid pathogens, especially if
there has been pretreatment with antibiotics.81 Las and Western
blot have been used for the detection of Lyme disease and
Chlamydia trachomatis, respectively. In ophthalmology, these
tests are used most commonly for the detection of Chlamydia,
viruses, fungi, and ocular protozoal disease.
DNA probes are particularly useful when looking for a particular organism such as a mycobacterium. These probes are
also helpful for the detection of organisms that are present
in small numbers or are fastidious and difcult to cultivate.
Radiolabeled DNA probes are more sensitive and more specic,
but results take several days. Nonradioactive probes are generally less sensitive but faster. Various kits based upon the use
of specic nucleic acid probes are now available commercially
for identifying specic bacteria in a sample. They combine
high specicity with speed.175 These procedures do not distinguish between viable and nonviable bacteria, which may be
an advantage, especially when prior antibiotic treatment has
been used. The problem of sample size can be overcome by
nucleic acid amplication. The most widely accepted method is
the polymerase chain reaction (PCR). These methods rely on
the hybridization of a specic nuclei acid probe to a specic
DNA sequence of the organism. Despite the need for specic
primers, the main problem with the use of PCR is its exquisite
sensitivity, making contamination a real possibility. The 16S
rRNA is a highly conserved portion of bacteria RNA with many
copies present in each organism. This allows for rapid and
specic identication of the microorganisms. These tests are
available for many bacteria such as mycobacterium species,
C. trachomatis and N. gonorrhoeae. Commercially available
systems of ligase chain reaction (LCR) are available for
C. trachomatis and N. gonorrhoeae. PCR can also be performed
for the detection of RNA targets called reverse transcriptase
PCR. Other systems of RNA amplication include transcription-mediated amplication (TMA) and the nucleic acid
sequence-based amplication (NASBA).81
Gas-liquid chromatography (GLC) and high-pressure liquid

chromatography (HPLC) have been useful in the clinical microbiology laboratory, especially in the identication of quinones
and in carbohydrate analysis for taxonomic classication.176
Also, analysis of cell wall phospholipid fatty acid has shown
that each genus has a unique lipid ngerprint. Several automated bacteria identication systems are currently marketed.
ANTIBIOTIC SUSCEPTIBILITY AND

SENSITIVITY
Susceptibility tests help to determine the most effective
therapeutic agent available. These tests are somewhat articial, because they do not consider the hosts defenses and
immune status, the number and accessibility of the organisms,
and whether the bacteria are intra- or extracellular, all of which
may influence antibiotic selection. In serious ocular infections,
bactericidal rather than bacteriostatic antibiotics should be
utilized whenever possible. In bacterial keratitis, sensitivity
testing does not take into account the antibiotic levels
obtainable through the use of fortied drops. Antibiotic drug
levels can be much higher on the ocular surface than in serum,
where the cut-off susceptibility is determined. Therefore, even
if the bacteria are reported to be resistant to a specic antibiotic, the organisms may still be killed by topical antibiotic due
to the high drug level achieved with frequent topical applications. Clinical response is the most important parameter in
evaluating patients with infectious keratitis. Just as it is
important for the clinical microbiology laboratory to report
and identify all bacteria present in ocular cultures, it is vital to
make sure that the clinical laboratory performing the sensitivity
testing is aware of the specic agents available for ophthalmic
use so that these antibiotics can be routinely tested. Antibiotics
such as polymyxin B, bacitracin, and neomycin are no longer
included in most clinical laboratories sensitivity panel, but they
remain important ocular therapeutic agents.
Susceptibility testing using either disk diffusion or dilutional
tests should be performed on all potential pathogens. In order
to accelerate the selection of appropriate antibiotics, direct
susceptibility testing has been advocated.177 A pure culture is
required for the test to be reliable and several factors, including
the density of the inoculum and the presence of other microorganisms, can make the results misleading. It is probably
better to prescribe broad-spectrum antibiotics and then, once
the microorganism has been identied, modify therapy, if
necessary, based on clinical response and antibiotic sensitivities
of the organism. Disk diffusion tests are the most commonly
used technique.178 Antimicrobial-containing disks are placed
on the agar surface inoculated with a pure culture of the
organism. A zone of inhibition occurs around the disk. The
extent of this inhibition determines whether the bacteria are
sensitive to the particular antibiotic. The signicant zone of
inhibition is different for each antibiotic owing to differences
in diffusion rates between antibiotics. Disk diffusion techniques
do have some limitations. They depend upon rapidly growing
organisms. The disk does not measure bactericidal activity,
and combinations of agents cannot be assayed. The disks only
reflect the usually obtainable serum concentrations and not
the higher levels obtainable within the tear lm or cornea or
intraocularly. Therefore, organisms reported as resistant may be
susceptible in the ophthalmic setting. The most common
clinical setting in which this occurs is in the patient in the ICU
or burn unit who is infected with multiple aminoglycosideresistant Pseudomonas organisms and may respond to fortied
aminoglycosides, especially when they are combined with carbenicillin or ticarcillin.179,180 A recently introduced BIOGRAM
(Giles Scientic, New York, NY) translates disk diffusion zone
CHAPTER 11
stain, requiring fewer organisms to yield a positive result.172

Bacteria can stain red, orange, or green depending on relative
amounts of DNA versus RNA, whereas nonbacterial cells such
as squamous cells and polymorphonuclear leukocytes stain
green-yellow.173 If bacteria are detected, then a Gram stain
can be performed on the same slide without decolorization.
The major disadvantage is that the AO stain requires a
fluorescent microscope. Acid-fast staining is useful to detect
Mycobacterium species. The brilliant green counterstain allows
for improved contrast between acid-fast organisms and the
background. These include the Carbol-fuschsin or Ziehl
Neelsen stains for acid-fast organisms. If Nocardia is suspected,
then an aqueous solution of 1% sulfuric acid rather than 3%
hydrochloric acid in 95% ethanol must be used as the
decolorizing agent. Fluorescein-conjugated lectins have been
used to identify microorganisms, primarily fungi, but do not
offer any advantages over existing stains in bacteriologic
diagnosis.174
111
SECTION 3
MICROBIOLOGY
112
sizes into minimal inhibitory concentrations (MICs), using

regression line analysis. A printed report is produced that
includes calculated MICs, KirbyBauer interpretations, and
inhibitory quotients that are based on achievable serum, urine,
bile, and cerebrospinal fluid concentrations.181 Potential
advantages include the ability to select from 34 antibiotics, the
ability to read results for many organisms in just 56 h, and
9095% correlation with reference laboratory results.182
Another approach for determining antibiotic susceptibility
is an elution method. The antimicrobial elutes from paper disks
into broth or agar, thus providing a desired concentration of the
antimicrobial agent in the medium. This approach is used in
some automated systems for susceptibility testing of aerobic
and facultatively anaerobic bacteria as well as in susceptibility
testing of anaerobic bacteria and mycobacteria.183 Paper
diffusion methods are superior for the detection of methicillinresistant strains, provided that either a medium with a high
sodium chloride content is used or plates are incubated at 30C
for at least 24 h.184
Dilutional tests have several advantages over disk diffusion
testing. Besides determining the MIC, the minimal lethal concentration (MLC), or minimal bactericidal concentration (MBC)
can also be determined. Microdilution methods that place the
antimicrobial agents in microtiter tray wells are more practical
and lend themselves more to automation, because the trays can
then be read photometrically. The small sample size may make
detection of resistant subpopulations less likely, especially as
incubation times are reduced. Clinically, this is important in
detecting third-generation cephalosporin resistance because of
depressed b-lactamase production in Enterobacter, Serratia, and
P. aeruginosa.185 In order to consider the organism susceptible,
the peak obtainable concentration should be two to four times
higher than the MIC. The MBC level assumes greater
importance in clinical situations in which the cure of an
infection depends entirely on the antibiotic and bactericidal
activity. This is important for immune-decient patients and
for those with CNS infections, but it also may be an important
consideration in endophthalmitis. Serum bactericidal activity
can be measured by the Schlichter test. Although not entirely
standardized, this test considers other factors that influence
antibiotic activity (especially serum protein binding) and has
been used primarily in the treatment of endocarditis and
osteomyelitis.186 Interpretation of MIC data is confusing to
many clinicians; one should encourage the laboratory to include
interpretative data with the report. Other pharmacodynamic
factors in bacterial infections of importance are the rate and
extent of bactericidal action, postantibiotic effect, minimal
antibiotic concentration, and postantibiotic leukocytic effect.187
Bacteria have shown great ability to develop resistance to
antibodies usually by the transfer of DNA between bacteria of
the same or different species. Much of the antibiotic resistance
encoded by genes is carried on plasmids. The production of
b-lactamase by H. influenzae, N. gonorrhoeae, and staphylococci correlates well with resistance to penicillin. Tests such as
the nitrocen test can provide results in a matter of minutes
rather than overnight.188 This is increasingly important as antibiotic resistance is seen more frequently in clinical situations,
for example, in coagulase-negative staphylococcal endophthalmitis.189 Pericellular resistance has now been found in
S. pneumoniae not due to b-lactamase production but due to
changes to the genes encoding the target enzymes.190 There has
also been an increasing number of bacteria resistant to fluoroquinolones, a commonly prescribed ophthalmic antibiotic.191
ANTISEPTICS AND DISINFECTION

Sterilization and disinfection are important concepts that are
taken for granted every day. Sterilization implies destruction of
all forms of life, including spores, and generally requires a
physical agent such as pressurized steam or ethylene oxide.
Disinfection refers to the destruction of pathogens and frequently involves the use of a chemical agent. Antimicrobial
agents are used daily in ophthalmic practice to preserve
medicines, sterilize instruments, and prepare the operative eld
for surgery. There are numerous factors to be considered in the
selection of an appropriate antiseptic. The chemical must be
bactericidal and nontoxic to the host. The length of exposure,
pH, and temperature are also taken into account. Some
methicillin-resistant strains of S. aureus (MRSA) containing
plasmids encoding gentamicin resistance (MGRSA) also have
increased MIC values toward biocides such as GACs, chlorhexidine, acridines, and propamidine isethionate.107,192 Gramnegative bacteria such as Pseudomonas are usually less sensitive
to chemical biocides (antiseptics, disinfectants, preservatives,
and sterilants) than are Gram-positive cocci. The main reason
is due to the great complexity of the outer cell membrane.193
Recent reports suggest that there is an increase in the resistance
of organisms to biocides, with increasing pressure for selecting
out antibiotic-resistant organisms.194
Key Features
Most common ocular surface bacteria flora are Gram-positive

cocci, mainly CNS.
The most common causes of ocular infections such as
infectious keratitis and endophthalmitis, are due to Grampositive cocci, such as staphylococci and streptococci.
Pseudomonas are frequent causes of infectious keratitis in
contact lens associated infections.
Minimizing the risk of postoperative infections is achieved by
eliminating bacteria from the ocular surface with the use of
antiseptic and antibiotics in the perioperative period.
Skin asepsis is important in ophthalmic surgery, because, as

noted earlier, most cases of endophthalmitis arise from the
patients own flora.195,196 Hendley and Ashe evaluated the effectiveness of various antimicrobial agents in eradicating CNS
from the surface and stratum corneum of the skin.197 They
evaluated ve antiseptic solutions and four antimicrobial
ointments. The skin surface was effectively sterilized by eight
of the nine agents tested. A soap-and-water wash was ineffective, but solutions of povidone-iodine, chlorhexidine-ethanol,
and 2% tincture of iodine eliminated surface bacteria. However,
sterilization of the stratum corneum was much more difcult to
accomplish. The rates of eradication of CNS from the stratum
corneum after surface treatment with chlorhexidine-ethanol
and povidone-iodine were not different from the control sites.
Only triple antibiotic ointment (neomycin, polymyxin B sulfate,
and bacitracin) was effective initially and inhibited overnight
repopulation from occurring. Only povidone-iodine has been
demonstrated to decrease the risk of endophthalmitis following
intraocular surgery.198,199 However, multiple studies have
demonstrated the effectiveness of povidone-iodine and
antibiotics in eliminating bacteria from the ocular surface at the
time of ocular surgery.200202
Ocular Bacteriology
1. Krieg NR, Holt JG, eds: Bergeys manual of

systematic bacteriology. In: Bergeys
Manual of systematic bacteriology.
Baltimore: Williams & Wilkins; 1984; 1:13.
2. Muto A, Osawa S: The guanine and
cytosine content of genomic DNA and
bacterial evolution. Proc Natl Acad Sci USA
1987; 84:166169.
3. Holt SC, Leadbetter ER: Structurefunction
relationships in prokaryotic cells. In: Balows
A, Duerden BI, eds. Topley & Wilsons
microbiology and microbial infections.
Oxford: Oxford University Press 9th ed;
1998:1144.
4. Wilkinson BJ, Kim Y, Peterson PK, et al:
Activation of complement by cell surface
components of Staphylococcus aureus.
Infect Immun 1978; 20:388392.
5. Meyer TE, Cusanovich MA, Kamen MD:
Evidence against use of bacterial amino
acid sequence data for construction of
all-inclusive phylogenetic trees. Proc Natl
Acad Sci USA 1986; 83:217220.
6. Hancock RE: Role of porins in outer
membrane permeability. J Bacteriol 1987;
169:929933.
7. Mims CA: The pathogenesis of infectious
disease. London: Academic Press/Grune &
Stratton; 1987.
8. Christensen GD, Simpson WA, Bisno AC,
Beachy E: Adherence of slime-producing
strains of Staphylococcus epidermidis to
smooth surfaces. Infect Immun 1982;
37:318326.
9. Adams D, Quayum M, Worthington T:
Evaluation of a 2% chlorhexidine gluconate
in 70% isopropyl alcohol skin disinfectant.
J Hosp Infect 2005; 61:287290.
10. Zegans ME, Becker HI, Budzik J, OToole G:
The role of bacterial biolms in ocular
infections. DNA Cell Biol 2002; 21:415420.
11. Jefferson KK: What drives bacteria to
produce a biolm? FEMS Microbiol Lett
2004; 236:163173.
12. Hoffman S, Sorkin BC, White PC: Chemical
characterization of a neural cell adhesion
molecule puried from embryonic brain
membranes. J Biol Chem 1982;
257:77207729.
13. Eisenstein BI: Type 1 mbriae of
Escherichia coli: genetic regulation,
morphogenesis, and role in pathogenesis.
Rev Infect Dis 1988; 10(Suppl 2):
S341S344.
14. Davis BD: Bacterial architecture. In:
Microbiology, Davis BD, ed. Philadelphia:
JB Lippincott: 1990:3839.
15. Alcamo IE: Fundamentals of Microbiology.
Alcamo IE, ed. Menlo Park: Benjamin
Cummings; 1991:119.
16. Bexkorovainy A: Iron proteins. In: Bullen JJ,
Grifths E, eds. Iron and infection:
molecular, physiological and clinical
aspects. England: Wiley; 1987:2767.
17. Abdallah FB, El Hage Chahine JM:
Transferrins: iron release from lactoferrin.
J Mol Biol 2000; 303:255266.
18. Morse SA, Chen CY, LeFaou A, Mietzner TA:
A potential role for the major iron-regulated
protein expressed by pathogenic Neisseria
species. Rev Infect Dis 1988; 10(Suppl
2):S306S310.
19. Catlin BW: Branhamella catarrhalis: an
organism gaining respect as a pathogen.
Clin Microbiol Rev 1990; 3:293320.
20. Tauber F, Krause A: The role of iron,

copper, zinc, and manganese in the
metabolism of the ocular tissues, with
special reference to the lens. Am J
Ophthalmol 1943; 26.
21. Brener D, DeVoe IW, Holbein BE: Increased
virulence of Neisseria meningitidis after in
vitro iron-limited growth at low pH. Infect
Immun 1981; 33:5966.
22. Grifths E: The iron-uptake systems of
pathogenic bacteria. In: Bullen JJ,
Grifths E, eds. Iron and infection:
molecular, physiological and clinical
aspects. Chichester, UK: Wiley;
1987:69137.
23. Brown V: Iron supply as a virulence factor.
In: Jackson GG, Thomas H, eds. The
pathogenesis of bacterial infections.
Berling: Springer; 1985:168.
24. World Health Organization: Methods of
assessment of avoidable blindness.
Geneva: WHO; 1980.
25. Gurtler V, Mayall BC: Genomic approaches
to typing, taxonomy and evolution of
bacterial isolates. Int J Syst Evol Microbiol
2001; 51(Pt 1):316.
26. McCulley JP, Dougherty JM, Deneau DG:
Classication of chronic blepharitis.
27. Packer AJ, Koontz FP: Ocular
staphylococcal infections. Am J
28. Dougherty JM, McCulley JP: Comparative
bacteriology of chronic blepharitis. Br J
Ophthalmol 1984; 68:524528.
29. McCulley JP, Dougherty JM: Bacterial
aspects of chronic blepharitis. Trans
Ophthalmol Soc UK 1986; 105(Pt
3):314318.
30. McCulley JP, Shine WE: Changing concepts
in the diagnosis and management of
blepharitis. Cornea 2000; 19:650658.
31. McCulley JP, Shine WE: Eyelid disorders:
the meibomian gland, blepharitis, and
contact lenses. Eye Contact Lens 2003;
29(1 Suppl):S93S95; discussion
S115S118, S192S194.
32. Dougherty JM, McCulley JP: Bacterial
lipases and chronic blepharitis. Invest
33. Shine WE, McCulley JP: Meibomianitis:
polar lipid abnormalities. Cornea 2004;
23:781783.
34. Dougherty JM, McCulley JP, Silvany RE,
Meyer DR: The role of tetracycline in
chronic blepharitis. Inhibition of lipase
production in staphylococci. Invest
35. Shine WE, McCulley JP, Pandya AG:
Minocycline effect on meibomian gland
lipids in meibomianitis patients. Exp Eye
Res 2003; 76:417420.
36. Ta CN, McCulley JP, Sylvany R, et al:
Effects of minocycline on the ocular flora of
patients with acne rosacea or seborrheic
blepharitis. Cornea 2003; 22:545548.
37. Ruoff K, Whiley R, Beighton D:
Streptococcus. In: Murray P, et al, eds.
Manual of clinical microbiology 8th ed.
American Society Microbiology;
Washington D.C.: ASM Press; 2003.
38. Miller D, Alfonso EC: Comparative in vitro
activity of levofloxacin, ofloxacin, and
ciprofloxacin against ocular streptococcal
isolates. Cornea 2004; 23:289293.
39. Ormerod LD, Ruoff KL, Meisler DM, et al:

Infectious crystalline keratopathy. Role of
nutritionally variant streptococci and other
bacterial factors. Ophthalmology 1991;
98:159169.
40. Umapathy T, Singh R, Dua HS, Donald F:
Non-tuberculous mycobacteria related
infectious crystalline keratopathy. Br J
Ophthalmol 2005; 89:13741375.
41. Janda W, Knapp J: Neisseria and
Branhamella catarrhalis. In: Murray P, et al,
eds. Manual of clinical microbiology.
8th ed., Washington, D.C.: American
Society Microbiology; 2003.
42. Ison CA: Antimicrobial agents and
gonorrhoea: therapeutic choice, resistance
and susceptibility testing. Genitourin Med
1996; 72:253257.
43. Carden SM, Colville DJ, Gonis G, Gilbert GL:
Kingella kingae endophthalmitis in an
infant. Aust N Z J Ophthalmol 1991;
19:217220.
44. Farmer JI: Enterocacteriaceae: introduction
and identication. In: Murray P, et al, eds.
Manual of clinical microbiology. 8th ed.,
Washington D.C.: American Society
Microbiology Press; 2003.
45. Park SB, et al: Endogenous
endophthalmitis caused by Escherichia
coli. Ann Ophthalmol 1993; 25:9599.
46. Brinser JH: Ocular bacteriology. In:
Tabbara KF, Hyndiuk RA, eds. Infections of
the eye. Boston: Little Brown; 1986:139.
47. Chin GN, Noble RC: Ocular involvement in
Yersinia enterocolitica infection presenting
as Parinauds oculoglandular syndrome.
48. Penland RL, Boniuk M, Wilhelmus KR:
Vibrio ocular infections on the US Gulf
Coast. Cornea 2000; 19:2629.
49. Altwegg M, Geiss HK: Aeromonas as a
human pathogen. Crit Rev Microbiol 1989;
16:253286.
50. Marshman WE, Lyons CJ: Congenital
endophthalmitis following maternal shellsh
ingestion. Aust N Z J Ophthalmol 1998;
26:161163.
51. Cheng KH, Leung SL, Hookman HW, et al:
Incidence of contact-lens-associated
microbial keratitis and its related morbidity.
Lancet 1999; 354:181185.
52. Killian M: Haemophilus. In: Murray P, et al,
eds. Manual of clinical microbiology.
American Society Microbiology; 2003.
53. Binder MI, Chua J, Kaiser PK, et al:
Actinobacillus actinomycetemcomitans
endogenous endophthalmitis: report of two
cases and review of the literature. Scand J
Infect Dis 2003; 35:133136.
54. Purcell JJ Jr, Krachmer JH: Corneal ulcer
caused by Pasteurella multocida. Am J
Ophthalmol 1977; 83:540542.
55. Klein B, Couch J, Thompson J: Ocular
infections associated with Eikenella
corrodens. Am J Ophthalmol 1990;
109:127131.
56. Alexandrakis G, Palma LA, Miller M,
Alfonso EC: Capnocytophaga keratitis.
Ophthalmology 2000; 107:15031506.
57. Chodosh J: Cats tooth keratitis: human
corneal infection with Capnocytophaga
canimorsus. Cornea 2001; 20:661663.
58. Phipps SE, Tamblyn DM, Badenoch PR:
Capnocytophaga canimorsus
endophthalmitis following cataract surgery.
CHAPTER 11
REFERENCES
113
MICROBIOLOGY
59.
60.
61.
62.
SECTION 3
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
114
Clin Experiment Ophthalmol 2002;

30:375377.
Wear DJ, Malaty RH, Zimmerman LE, et al:
Cat scratch disease bacilli in the
conjunctiva of patients with Parinauds
oculoglandular syndrome. Ophthalmology
1985; 92:12821287.
Ormerod LD, Dailey JP: Ocular
manifestations of cat-scratch disease.
Curr Opin Ophthalmol 1999; 10:209216.
Brooks G, Butel J, Morse S: Medical
microbilogy. 23rd edn. Lange Medical
Books; New York: McGraw Hill 2004.
Jackson MA, Tyson M, Burry FV, et al:
Antimicrobial therapy for Parinauds
oculoglandular syndrome. Pediatr Infect Dis
J 1992; 11:130132.
Cuchural GJ Jr, Tally FP, Jacobus NV, et al:
Comparative activities of newer betalactam agents against members of the
Bacteroides fragilis group. Antimicrob
Agents Chemother 1990; 34:479480.
Kelly LD, Gaynon M: Bacteroides fragilis
endophthalmitis: a case report. Can J
Ophthalmol 1990; 25:208209.
Das T, Choudhury K, Sharma S, et al:
Clinical prole and outcome in Bacillus
endophthalmitis. Ophthalmology 2001;
108:18191825.
Aldave AJ, Stein JD, Deramo VA, et al:
Treatment strategies for postoperative
Propionibacterium acnes endophthalmitis.
Ophthalmology 1999; 106:23952401.
Zaidman GW, Coudron P, Piros J: Listeria
monocytogenes keratitis. Am J Ophthalmol
1990; 109:334339.
Deramo VA, Shah GK, Garden M, Magiure JI:
Good visual outcome after Listeria
monocytogenes endogenous
endophthalmitis. Retina 1999; 19:566568.
Anand S, Hollingworth K, Kumar V,
Sandramouli S: Canaliculitis: the incidence
of long-term epiphora following
canaliculotomy. Orbit 2004; 23:1926.
Sridhar MS, Sharma S, Garg P, Rao GN:
Treatment and outcome of nocardia
keratitis. Cornea 2001; 20:458462.
Haripriya A, Lalitha P, Mathen H, et al:
Nocardia endophthalmitis after cataract
surgery: clinicomicrobiological study. Am J
Ophthalmol 2005; 139:837846.
Solomon A, Karp CL, Miller D, et al:
Mycobacterium interface keratitis after
laser in situ keratomileusis. Ophthalmology
2001; 108:22012208.
Giaconi J, Pham R, Ta CN: Bilateral
Mycobacterium abscessus keratitis after
laser in situ keratomileusis. J Cataract
Refract Surg 2002; 28:887890.
Sarayba MA, Shamie N, Reiser BJ, et al:
Fluoroquinolone therapy in Mycobacterium
chelonae keratitis after lamellar
keratectomy. J Cataract Refract Surg 2005;
31:13961402.
Jawetz E, Melnick JL, Adelberg EA: Review
of medical microbiology 17th ed. Los Altos:
Lange; 1987:162.
Sanchez PJ, Regan JA: Vertical
transmission of Ureaplasma urealyticum
from mothers to preterm infants. Pediatr
Infect Dis J 1990; 9:398401.
Murray PR, Drew WL: Medical microbiology.
St Louis: CV Mosby; 1990:254.
Wirostko E, Johnson L, Wirostko B:
Ulcerative colitis associated chronic uveitis.
Parasitization of intraocular leucocytes by
mollicute-like organisms. J Submicrosc
Cytol Pathol 1990; 22:231239.
79. Aly R, Shinefield HI, Strauss WG, Maiback HI:

Bacterial adherence to nasal mucosal cells.
Infect Immun 1977; 17:546549.
80. Salyers A: Baterial pathogenesis. 2nd edn.
American Society for Microbiology;
2002.
81. Brooks G, Butel J, Morse S: Medical
microbiology. 23rd edn. New York: Lange
medical Books/McGraw Hill, 2004.
82. Le Bouguenec C: Adhesins and invasins of
pathogenic Escherichia coli. Int J Med
Microbiol 2005; 295:471478.
83. Jonson AB, Normark S, Rhen M: Fimbriae,
pili, flagella and bacterial virulence. Contrib
Microbiol 2005; 12:6789.
84. Espersen F, Clemmensen I: Isolation of a
bronectin-binding protein from
Staphylococcus aureus. Infect Immun
1982; 37:526531.
85. Abraham SN, Beachey EH, Simpson WA:
Adherence of Streptococcus pyogenes,
Escherichia coli, and Pseudomonas
aeruginosa to bronectin-coated and
uncoated epithelial cells. Infect Immun
1983; 41:12611268.
86. Raynor RH, Scott DF, Best GK:
Lipoteichoic acid inhibition of phagocytosis
of Staphylococcus aureus by human
polymorphonuclear leukocytes. Clin
Immunol Immunopathol 1981; 19:181189.
87. Noble MA, Grant SK, Hajen E:
Characterization of a neutrophil-inhibitory
factor from clinically signicant
Staphylococcus epidermidis. J Infect Dis
1990; 162:909913.
88. Badenoch PR, Coster DJ: Antibiotics and
corticosteroids: functions and interaction
in ocular disease. In: Cavanagh HD, ed.
The cornea: transactions of the World
Congress on the cornea III. New York:
Raven; 1998:475.
89. Ramphal R, McNiece MT, Polack FM:
Adherence of Pseudomonas aeruginosa
to the injured cornea: a step in the
pathogenesis of corneal infections.
Ann Ophthalmol 1981; 13:421425.
90. Stern GA, Weitzenkorn D, Valenti J:
Adherence of Pseudomonas aeruginosa to
the mouse cornea. Epithelial v stromal
adherence. Arch Ophthalmol 1982;
100:19561958.
91. Gristina AG, Oga M, Webb LX, Hobgood CD:
Adherent bacterial colonization in the
pathogenesis of osteomyelitis. Science
1985; 228:990993.
92. Stern GA, Lubniewski A, Allen C:
The interaction between Pseudomonas
aeruginosa and the corneal epithelium.
An electron microscopic study. Arch
Ophthalmol 1985; 103:12211225.
93. Koch JM, Refojo MF, Hanninen LA, et al:
Experimental. Pseudomonas aeruginosa
keratitis from extended wear of soft contact
lenses. Arch Ophthalmol 1990;
108:14531459.
94. Trancassini M, Magni A, Ghezzi MC, et al:
Role of alkaline protease and elastase in
the adherence of Pseudomonas aeruginosa
to WEHI cells. Microbiologica 1989;
12:257261.
95. Deighton MA, Balkau B: Adherence
measured by microtiter assay as a
virulence marker for Staphylococcus
epidermidis infections. J Clin Microbiol
1990; 28:24422447.
96. Zaidi TS, Fleiszig SM, Preston MJ, et al:
Lipopolysaccharide outer core is a ligand
for corneal cell binding and ingestion of
97.
98.
99.
100.
101.
102.
103.
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
Pseudomonas aeruginosa. Invest

Staley TE, Jones EW, Corley LD:
Attachment and penetration of Escherichia
coli into intestinal epithelium of the ileum in
newborn pigs. Am J Pathol 1969;
56:371392.
Melly MA, Duke LJ, Liau DF, Hash JH:
Biological properties of the encapsulated
Staphylococcus aureus M. Infect Immun
1974; 10:389397.
Wiley BB, Maverakis NH: Capsule
production and virulence among strains of
Staphylococcus aureus. Ann N Y Acad Sci
1974; 236:221232.
Munson RS Jr: Haemophilus influenzae:
surface antigens and aspects of virulence.
Can J Vet Res 1990; 54(Suppl):S63S67.
Brook I: Pathogenicity of the Bacteroides
fragilis group. Ann Clin Lab Sci 1989;
19:360376.
Wilkinson BJ, Sisson SP, Kim Y, Peterson PK:
Localization of the third component of
complement on the cell wall of
encapsulated Staphylococcus aureus M:
implications for the mechanism of
resistance to phagocytosis. Infect Immun
1979; 26:11591163.
Gemmell CG, Peterson PK, Schmeling D,
et al: Potentiation of opsonization and
phagocytosis of Streptococcus pyogenes
following growth in the presence of
clindamycin. J Clin Invest 1981;
67:12491256.
Peterson PK, Schmeling D, Cleary PP, et al:
Inhibition of alternative complement
pathway opsonization by group A
streptococcal M protein. J Infect Dis 1979;
139:575585.
Wessels MR, Rubens CE, Benedi VJ,
Kaspar DL: Denition of a bacterial
virulence factor: sialylation of the group B
streptococcal capsule. Proc Natl Acad Sci
USA 1989; 86:89838987.
Pei Z, Blaser MJ: Pathogenesis of
Campylobacter fetus infections. Role of
surface array proteins in virulence in a
mouse model. J Clin Invest 1990;
85:10361043.
Costerton JW, Lam J, Lam K, Chan R: The
role of the microcolony mode of growth in
the pathogenesis of Pseudomonas
aeruginosa infections. Rev Infect Dis 1983;
5(Suppl 5):S867S873.
Marcus H, Baker NR: Quantitation of
adherence of mucoid and nonmucoid
Pseudomonas aeruginosa to hamster
tracheal epithelium. Infect Immun 1985;
47:723729.
Austin RM, Daniels CA: The role of protein
A in the attachment of staphylococci to
influenza-infected cells. Lab Invest 1978;
39:128132.
Goding JW: Use of staphylococcal protein
A as an immunological reagent. J Immunol
Methods 1978; 20:241253.
Woof JM: The human IgA-Fc alpha
receptor interaction and its blockade by
streptococcal IgA-binding proteins.
Biochem Soc Trans 2002; 30:491494.
Reichert R, Stern G: Quantitative
adherence of bacteria to human corneal
epithelial cells. Arch Ophthalmol 1984;
102:13941395.
Foster TJ, Hartford O, ODonnell D:
Hostpathogen proteinprotein interactions
in staphylococcus. In: McCrae MA,
Saunders JR, Smyth CJ, Stow ND, eds.
114.
115.
116.
117.
118.
119.
120.
121.
122.
123.
124.
125.
126.
127.
128.
129.
130.
131.
Molecular aspects of hostpathogen

interactions. Cambridge, UK: Cambridge
University Press; 1997:6794.
Moulder JW: Comparative biology of
intracellular parasitism. Microbiol Rev 1985;
49:298337.
Gorby GL, Robinson EN, Jr., Barley LR,
et al: Microbial invasion: a covert activity?
Can J Microbiol 1988; 34:507512.
Chan J, Fujiwara T, Brennan P, et al:
Microbial glycolipids: possible virulence
factors that scavenge oxygen radicals. Proc
Natl Acad Sci USA 1989; 86:24532457.
Whitnack E, Beachey EH: Inhibition of
complement-mediated opsonization and
phagocytosis of Streptococcus pyogenes
by D fragments of brinogen and brin
bound to cell surface M protein. J Exp Med
1985; 162:19831997.
Whitnack E, Beachey EH: Biochemical and
biological properties of the binding of
human brinogen to M protein in group A
streptococci. J Bacteriol 1985;
164:350358.
Johnson MK, Boese-Marrazzo D,
Pierce WA Jr: Effects of pneumolysin on
human polymorphonuclear leukocytes and
platelets. Infect Immun 1981; 34:171176.
Paton JC, Ferrante A: Inhibition of human
polymorphonuclear leukocyte respiratory
burst, bactericidal activity, and migration by
pneumolysin. Infect Immun 1983;
41:12121216.
Ginsburg I: Is streptolysin S of group A
streptococci a virulence factor? Apmis
1999; 107:10511059.
Berk RS, Brown D, Coutinho I, Meyers D:
In vivo studies with two phospholipase C
fractions from Pseudomonas aeruginosa.
Infect Immun 1987; 55:17281730.
Heck LW, Morihara K, Abrahamson DR:
Degradation of soluble laminin and
depletion of tissue-associated basement
membrane laminin by Pseudomonas
aeruginosa elastase and alkaline protease.
Infect Immun 1986; 54:149153.
Howe TR, Iglewski BH: Isolation and
characterization of alkaline proteasedecient mutants of Pseudomonas
aeruginosa in vitro and in a mouse eye
model. Infect Immun 1984; 43:10581063.
Johnson MK, Allen JH: The role of
hemolysin in corneal infections with
Pseudomonas aeruginosa. Invest
Liu PV: Extracellular toxins of
Pseudomonas aeruginosa. J Infect Dis
1974; 130(Suppl(0)):S94S99.
Nicas TI, Iglewski BH: The contribution of
exoproducts to virulence of Pseudomonas
aeruginosa. Can J Microbiol 1985;
31:387392.
Ohman DE, Burns RP, Iglewski BH: Corneal
infections in mice with toxin A and elastase
mutants of Pseudomonas aeruginosa.
J Infect Dis 1980; 142:547555.
Burns FR, Paterson CA, Gray RD, Wells JT:
Inhibition of Pseudomonas aeruginosa
elastase and Pseudomonas keratitis using
a thiol-based peptide. Antimicrob Agents
Chemother 1990; 34:20652069.
Baum JL: Current concepts in
ophthalmology. Ocular infections. N Engl J
Med 1978; 299:2831.
Gibbons RJ: Bacterial adherence to
mucosal surfaces and its inhibition by
secretory antibodies. Adv Exp Med Biol
1974; 45:315325.
132. Williams RC; Gibbons RJ: Inhibition of

bacterial adherence by secretory
immunoglobulin A: a mechanism of antigen
disposal. Science 1972; 177:697699.
133. Burdon DW: The bactericidal action of
immunoglobulin A. J Med Microbiol 1973;
6:131139.
134. Smolin G: Immunology of ocular infections.
In: Duane TD, Jaeger EA, eds. Biomedical
foundations of ophthalmology. Philadelphia:
Harper & Row; 1985.
135. Luckasen JR, Sobad A, Goltz RW, et al:
T and B lymphocytes in atopic eczema.
Arch Dermatol 1974; 110:375377.
136. Palestine AG, Meyern SM, Fauci AS,
Gallin JI: Ocular ndings in patients with
neutrophil dysfunction. Am J Ophthalmol
1983; 95:598604.
137. Greenblatt J, Boackle RJ, Schwab JH:
Activation of the alternate complement
pathway by peptidoglycan from
streptococcal cell wall. Infect Immun 1978;
19:296303.
138. Paton JC, Rowan-Kelly B, Ferrante A:
Activation of human complement by the
pneumococcal toxin pneumolysin. Infect
Immun 1984; 43:10851087.
139. Baker CJ, Edwards MS, Webb BJ,
Kaspar DL: Antibody-independent classical
pathway-mediated opsonophagocytosis of
type Ia, group B streptococcus. J Clin
Invest 1982; 69:394404.
140. Leist-Welsh P, Bjornson AB:
Immunoglobulin-independent utilization of
the classical complement pathway in
opsonophagocytosis of Escherichia coli by
human peripheral leukocytes. J Immunol
1982; 128:26432651.
141. Fearon DT, Austen KF: Current concepts in
immunology: the alternative pathway of
complement a system for host resistance
to microbial infection. N Engl J Med 1980;
303:259263.
142. Joiner KA, BrownEJ, Frank MM:
Complement and bacteria: chemistry and
biology in host defense. Annu Rev Immunol
1984; 2:461491.
143. Mondino BJ, Rabin BS, Kessleer E, et al:
Corneal rings with gram-negative bacteria.
144. Badenoch PR, Finlay-Jones JJ, Coster DJ:
Enzymatic disaggregation of the infected
rat cornea. Invest Ophthalmol Vis Sci 1983;
24:253257.
145. Kao WW, Ebert J, Kao CWC, et al:
Development of monoclonal antibodies
recognizing collagenase from rabbit PMN;
the presence of this enzyme in ulcerating
corneas. Curr Eye Res 1986; 5:801815.
146. McIntosh RS, Cade JE, Al-Abed M, et al:
The spectrum of antimicrobial peptide
expression at the ocular surface. Invest
147. Cullor JS, Mannis MJ, Murphy CJ, et al: In
vitro antimicrobial activity of defensins
against ocular pathogens. Arch Ophthalmol
1990; 108:861864.
148. Chipman DM, Sharon N: Mechanism of
lysozyme action. Science 1969;
165:454465.
149. Velos P, Cherry PM, Miller D: An improved
method for measuring human tear
lysozyme concentration. Arch Ophthalmol
1985; 103:3133.
150. McClellan KA: Mucosal defense of the outer
eye. Surv Ophthalmol 1997; 42:233246.
151. Bonavida B, Sapse AT: Human tear
lysozyme. II. Quantitative determination
152.
153.
154.
155.
156.
157.
158.
159.
160.
161.
162.
163.
164.
165.
166.
167.
168.
with standard Schirmer strips. Am J

Mackie IA, Seal DV: Quantitative tear
lysozyme assay in units of activity per
microlitre. Br J Ophthalmol 1976;
60:7074.
Pietsch RL, Pearlman ME: Human tear
lysozyme variables. Arch Ophthalmol 1973;
90:9496.
Jay JM: Production of lysozyme by
staphylococci and its correlation with three
other extracellular substances. J Bacteriol
1966; 91:18041810.
Schindler CA, Schuhardt VT: Lysostaphin:
a new bacteriolytic agent for the
Staphylococcus. Proc Natl Acad Sci USA
1964; 51:414421.
Cisani G, Varaldo PG, Grazi G, Soro O:
High-level potentiation of lysostaphin antistaphylococcal activity by lysozyme.
Antimicrob Agents Chemother 1982;
21:531535.
Martinez RJ, Carroll SF: Sequential
metabolic expressions of the lethal process
in human serum-treated Escherichia coli:
role of lysozyme. Infect Immun 1980;
28:735745.
Friedman MG: Antibodies in human tears
during and after infection. Surv Ophthalmol
1990; 35:151157.
Berk RS, Montgomery IN, Hazlett LD:
Serum antibody and ocular responses to
murine corneal infection caused by
Pseudomonas aeruginosa. Infect Immun
1988; 56:30763080.
Moon MM, Hazlett LD, Hancock RD, et al:
Monoclonal antibodies provide protection
against ocular Pseudomonas aeruginosa
infection. Invest Ophthalmol Vis Sci 1988;
29:12771284.
Grifss JM, Bertram MA:
Immunoepidemiology of meningococcal
disease in military recruits. II. Blocking of
serum bactericidal activity by circulating
IgA early in the course of invasive disease.
J Infect Dis 1977; 136:733739.
Onderdonk AB, Cisneros RL, Finberg R,
et al: Animal model system for studying
virulence of and host response to
Bacteroides fragilis. Rev Infect Dis 1990;
12(Suppl 2):S169S177.
Toffaletti DL, Schwab JH: Modulation of
lymphocyte functions by group A
streptococcal membrane. Cell Immunol
1979; 42:317.
Jones DB, Liesegang TJ, Robinson NM:
Laboratory diagnosis of ocular infections.
In: Washington JAI, ed. Cumitech 13,
cumulative techniques and procedures in
clinical microbiology. Washington, DC:
American Society for Microbiology;1981:10.
Barza M, Pavan PR, Doft BH, et al:
Evaluation of microbiological diagnostic
techniques in postoperative
endophthalmitis in the Endophthalmitis
Vitrectomy Study. Arch Ophthalmol 1997;
115:11421150.
Thayer JD, Martin JE Jr: Improved medium
selective for cultivation of N. gonorrhoeae
and N. meningitidis. Public Health Rep
1966; 81:559562.
Perry LD, Brinser JH, Kolodner H:
Anaerobic corneal ulcers. Ophthalmology
1982; 89:636642.
Brinser JH, Burd EM: Principles of
diagnostic ocular microbiology. In: Tabbara
KF, Hyndiuk RA, eds. Infections of the eye.
Boston: Little Brown; 1977:77.
CHAPTER 11
Ocular Bacteriology
115
SECTION 3
MICROBIOLOGY
116
169. Grayson M: In: Diseases of the cornea,

2nd ed St Louis: CV Mosby; 1983:53.
170. Idigoras P, P-Terez-Trallero, Alcorta M,
Gutierrez C, Munoz-Baroja I: Rapid
detection of tuberculosis and nontuberculosis mycobacteria by microscopic
observation of growth on Middlebrook
7H11 agar. Eur J Clin Microbiol Infect Dis
1995; 14:610.
171. Alkamo IE: Fundamentals of microbiology,
3rd ed. Menlo Park: Benjamin-Cummings;
1991:85.
172. Groden LR, Rodnite J, Brisner JH,
Genvert GI: Acridine orange and Gram
stains in infectious keratitis. Cornea 1990;
9:122124.
173. Kronvall G, Myhre E: Differential staining of
bacteria in clinical specimens using
acridine orange buffered at low pH. Acta
Pathol Microbiol Scand B 1977;
85:249254.
174. Robin JB, Arffa RC, Auni I, Rao NA: Rapid
visualization of three common fungi using
fluorescein-conjugated lectins. Invest
175. Duerden BI, Towner KJ, Megee JT:
Isolation, description and identication of
bacteria. In: Balows A, Duerden BI, eds.
Topley & Wilsons microbiology and
microbial infections. Oxford, UK: Oxford
176. Martin R, Schneider WA: Chromatography
for the identication of microorganisms.
In: Wentworth BB, ed. Diagnostic
procedures for bacterial infections.
Washington, DC: American Public Health
Association; 1987:703.
177. Mino de Kaspar H, Newbauer AS, Molnar A,
et al: Rapid direct antibiotic susceptibility
testing in endophthalmitis. Ophthalmology
2002; 109:687693.
178. Bauer AW, Kirby WMM, Sherris JC, Turck M:
Antibiotic susceptibility testing by a
standardized single disk method. Am J Clin
Pathol 1966; 45:493496.
179. Gelender H, Rettich C: Gentamicinresistant Pseudomonas aeruginosa corneal
ulcers. Cornea 1984; 3:2126.
180. Ormerod LD, Heseltine PN, Alfonso G,
et al: Gentamicin-resistant pseudomonal
infection. Rationale for a redenition of
181.
182.
183.
184.
185.
186.
187.
188.
189.
190.
191.
192.
ophthalmic antimicrobial sensitivities.

Cornea 1989; 8:195199.
Ellner PD, Neu HC: The inhibitory quotient.
A method for interpreting minimum
inhibitory concentration data. JAMA 1981;
246:15751578.
DAmato RF, Jochstein L, Vernaleo JR,
Cleri JF: Evaluation of the BIOGRAM
antimicrobial susceptibility test system. J
Clin Microbiol 1985; 22:793798.
Washington JA: In vitro testing of
antimicrobial agents. Infect Dis Clin North
Am 1989; 3:375387.
Brumtt W, Hamilton-Miller J: Methicillinresistant Staphylococcus aureus. N Engl J
Med 1989; 320:11881196.
Sanders CC, Sanders WE Jr: Microbial
resistance to newer generation beta-lactam
antibiotics: clinical and laboratory
implications. J Infect Dis 1985; 151:399406.
Jordan GW, Kawachi MM: Analysis of
serum bactericidal activity in endocarditis,
osteomyelitis, and other bacterial
infections. Medicine (Baltimore) 1981;
60:4961.
Levison ME, Bush LM: Pharmacodynamics
of antimicrobial agents. Bactericidal and
postantibiotic effects. Infect Dis Clin North
Am 1989; 3:415421.
OCallaghan CH, Morris A, Kirby SM,
Shingler AH: Novel method for detection of
beta-lactamases by using a chromogenic
cephalosporin substrate. Antimicrob
Davis JL, Koidou-tsiligianni A, Pfligelder SC,
et al: Coagulase-negative staphylococcal
endophthalmitis. Increase in antimicrobial
resistance. Ophthalmology 1988;
95:14041410.
Spratt BG: Resistance to antibiotics
mediated by target alterations. Science
1994; 264:388393.
Goldstein MH, Kowalski RP, Gordon YJ:
Emerging fluoroquinolone resistance
in bacterial keratitis: a 5-year
review. Ophthalmology 1999;
106:13131318.
Lyon BR, Skurray R: Antimicrobial
resistance of Staphylococcus aureus:
genetic basis. Microbiol Rev 1987;
51:88134.
193. Russell AD: Microbial susceptibility and

resistance to chemical and physical agents.
In: Balows A, Duerden BI, eds. Topley &
Wilsonss microbiology and microbial
infections. Oxford, UK: Oxford University
Press; 1998:149184.
194. Sheldon AT Jr: Antiseptic resistance: real
or perceived threat? Clin Infect Dis 2005;
40:16501656.
195. Bannerman TL, Rhoden DL, McAllister SK,
et al: The source of coagulase-negative
staphylococci in the Endophthalmitis
Vitrectomy Study. A comparison of eyelid
and intraocular isolates using pulsed-eld
gel electrophoresis. Arch Ophthalmol 1997;
115:357361.
196. Speaker MG, Milch FA, Shah MK, et al:
Role of external bacterial flora in the
pathogenesis of acute postoperative
98:639649; discussion 650.
197. Hendley JO, Ashe KM: Effect of topical
antimicrobial treatment on aerobic bacteria
in the stratum corneum of human skin.
35:627631.
198. Speaker MG, Menikoff JA: Prophylaxis of
endophthalmitis with topical povidoneiodine. Ophthalmology 1991;
98:17691775.
199. Schmitz S, Dick HB, Krummenauer F,
Pfeiffer N: Endophthalmitis in cataract
surgery: results of a German survey.
Ophthalmology 1999; 106:18691877.
200. Isenberg SJ, Apt L, Yoshimori R, Khwang S
et al: Chemical preparation of the eye in
ophthalmic surgery. IV. Comparison of
povidone-iodine on the conjunctiva with a
prophylactic antibiotic. Arch Ophthalmol
1985; 103:13401342.
201. Mino de Kaspar H, Chang RT, Singh K,
et al: Prospective randomized comparison
of 2 different methods of 5% povidoneiodine applications for anterior segment
intraocular surgery. Arch Ophthalmol 2005;
123:161165.
202. Ta CN, Egbert PR, Singh K, et al:
Prospective randomized comparison of
3-day versus 1-hour preoperative ofloxacin
prophylaxis for cataract surgery.
Ophthalmology 2002; 109:20362041.
CHAPTER
12
Chlamydial Disease
Irmgard Behlau
ANATOMY, PHYSIOLOGY, AND LIFE CYCLE

OF THE MICROORGANISM
TAXONOMY
Historically based on different phenotypic properties,1 all
chlamydiae were classied into the order Chlamydiales, one
family Chlamydiaceae, and one genus Chlamydia, which
was composed of four species, C. trachomatis, C. psittaci,
C. pneumonia, and C. pecorum.1 Only the rst three species are
associated with human disease. Humans are the natural hosts
of C. trachomatis and C. pneumoniae. These species have no
animal reservoirs, and transmission is from human to human.
Birds and some mammals are the natural hosts of C. psittaci
(Table 12.1).16 Based on recent analysis of 16S and 23S ribonucleic acid (rRNA) analysis, a new taxonomic classcation7,8
proposes that the order Chlamydiales be divided into four
families, with the family Chlamydiaceae divided into two
genera, Chlamydia and Chlamydophila. The genus Chlamydia
is composed of three species: C. trachomatis, C. muridarum,
and C. suis; the genus Chlamydophila is composed of six
species: C. pneumoniae, C. psittaci, C. pecorum, C. abortus,
C. caviae, and C. felis.7,8
MICROBIAL CHARACTERISTICS
For many years Chlamydia organisms were considered viruses,
due to their obligate intracellular replication and small size
(diameter 2001500 nm). They contain both DNA and RNA,
replicate by binary ssion, possess a cell wall, and are inhibited
by antimicrobial agents. These bacteria are nonmotile and have
morphologic similarities to Gram-negative organisms with a
trilaminar outer membrane, but lack classic peptidoglycan.9,10
At this time, seven chlamydial genomes have been sequenced;
the molecular mass of the chlamydial genome is 660 106
Da which is smaller than any other prokaryote except for
Mycoplasma sp.11 Certain metabolic pathways are missing
including amino acid and purinepyrimidine biosynthesis,
anaerobic fermentation, and transformation competence
proteins.12 Chlamydiae possess the metabolic pathways to
synthesize adenosine triphosphate (ATP), but are unable to
synthesize ATP or to produce metabolic energy;13 therefore,
these use the ATP produced by the host cell for their own energy
requirements. Although chlamydiaes dependence on the host
cell may appear restrictive, chlamydiae are proving to be highly
evolved pathogens. They are capable of infecting warm- and
cold-blooded animals and a variety of cell types from soil
TABLE 12.1. Characteristics of Human Chlamydial Species

C. trachomatis
C. pneumoniae
C. psittaci
Genus
Chlamydia
Chlamydophila
Chlamydophila
Natural hosts
Humans
Humans
Animals and birds
Serovars
18
Mode of transmission
Person to person, mother to infant
Airborne person to person
Airborne bird excreta to humans
Human diseases and

associated serovar
Trachoma: A, B, Ba, C
Genital infections: D, Da, E, F, G,
H, I, Ia, J, K
LGV: L1, L2, L2a, L3
Upper and lower respiratory

tract disease; coronary artery
disease; ?age-related macular
degeneration
Pneumonia (psittacosis), fever of

unexplained origin
Elementary body
Coccoid
Pear shaped
Coccoid
Morphology of Inclusion body
Single, round-vacuolar
Multiple, uniform-dense
Multiple, variable-sized dense

inclusions
Synthesize folate
Sulfa sensitivity
Iodine-staining glycogen in
inclusions
117
MICROBIOLOGY
protists to brain microglial cells.1 Their high prevalence rate
of infection in humans and birds suggests that adaptation of
Chlamydia to obligate intracellular parasitism offers some
evolutionary advantage. Most recently, they appear to be able to
enter an alternative nonreplicative and persistent life-cycle,14,15
allowing them an optimal survival mechanism, thereby
allowing recurrent, relapsing, and persistent infections.
SECTION 3
MORPHOLOGY AND LIFE CYCLE

The evolutionarily distinct, intracellular biphasic life cycle
shared by all Chlamydiae has been well characterized under
favorable environmental conditions; it consists of inactive
infectious elementary bodies (EBs) and metabolically active but
noninfectious reticulate bodies (RBs). All EBs are of similar size
(300 nm); C. trachomatis and C. psittaci are spherical particles,
while the EB of C. pneumoniae is pear-shaped (Fig. 12.1a). The
chlamydial life cycle (Fig 12.2) begins when infectious,
metabolically inert EBs attach to cells of a susceptible host
epithelial cell via uncertain mechanisms.16 EBs stimulate
uptake and entry into the cell by receptor-mediated endocytosis
via postulated clarithin-coated pits1618 but pinocytosis via
noncoated pits and use of heparin-like bridging molecules are
also speculated. Ingestion by the host cell results with the
internalized EB within a host-derived vacuole termed inclusion.
Through an unknown process requiring bacterial protein synthesis, inclusions are stable, not maturing into late endosomes
or fusing with lysosomes. Phagolysosomal fusion does not occur
and the organism is protected from digestion by lysozymes.19
The chlamydial phagosome, or inclusion body, is transported to
a juxtanuclear position that corresponds to the peri-Golgi
region. The inclusion body then intercepts cellular metabolites
being transported from the Golgi apparatus to the cell membrane via the trans-Golgi exocytic pathway.20,21 Approximately
8 h after entering the cell, the EB reorganizes into a reticulate
body (RB), so-called because of the dispersed brillar pattern of
its nucleic acids (Fig. 12.1b).22,24 The RB is the replicative phase
in the life cycle of chlamydiae. Transition of EB to RB is
associated with: loss of infectivity, an increase in diameter to
8001000 nm, and an increase in ratio of DNA to RNA from
1:1 in the EB to 3:1 in the RB,23 an increase in the rate of
metabolic activity compared to a metabolically inert EB, and a
change in the cell wall from rigid and impermeable in the EB to
flexible and permeable in the RB. These cell wall changes are
thought to result from reduction of cross-linked disulde bonds
in the outer membrane proteins by the intraphagosomal reducing conditions to which the EB is exposed after endocytosis.2428
The increased permeability of the RB cell wall permits uptake
of ATP and nutrients from the host cell. RBs typically line the
inner margin of the inclusion body membrane which contrasts
with the EBs that are distributed randomly throughout the
inclusion.4,29 The RBs initiate RNA and DNA synthesis and
multiply by binary ssion until the original phagosome
becomes distended by its content of several hundred to more
than 1000 chlamydial cells. After 812 rounds of multiplication, the RB asynchronously differentiate to EBs.29 As RB replication proceeds, the reducing power of the microenvironment
probably decreases, and free sulfhydryl groups are oxidized,
forming disuldes. This restores the rigidity and impermeability of the cell wall and produces a decrease in the rate
of metabolism, coincident with reorganization of RB into EB.24
At 4884 h postinfection (depending on the infecting species),
the host cell and its intracytoplasmic inclusions rupture, and
the newly formed EB progeny are released into the extracellular
milieu, infecting other cells or a new host to begin a new
cycle.29a,29b,29c
The recognition that chlamydiae may cause persistent
infections in their hosts dates back to 1933 with latent
psittacosis in birds.30 There is increasing evidence in vitro and
in vivo that chlamydiae persist in an altered form during
chronic disease.31 Under adverse conditions, such as glucose or
amino acid limitation, elevated temperatures, or sublethal
antibiotic concentrations, chlamydiae are capable of conversion
to a noncultivable growth stage with nonreplicating persistent
bodies (PBs) which appear aberrant and display altered gene
expression.14 The different in vitro persistence systems share
altered growth and ultrastructural characteristics with enlarged,
pleomorphic RBs that are inhibited in binary ssion, but accumulate chromosomes and do not differentiate to EBs. These
changes are reversible27,38,39 once either the factor that inhibits
growth is removed (antibiotics,28,32,33 cytokine-induction,
particularly interferon gamma [IFN-g],3436 or infection with
phage),37 or replacement of a missing nutrient.27,29,39 In contrast
to other persistence model systems, chlamydiae become
spontaneously persistent following infection of monocytes40,41
and when maintained under continuous culture conditions.42,43
Supportive in vivo observations for chlamydial persistence
include epidemiologic reports of recurrences which are most
likely due to reactivation of persistent infections rather than
reinfections15 (active trachoma decades after the initial
infection,44 altered morphological forms in vivo (recent electron
microscopic visualization of C. pneumoniae aberrant RBs
(resembling those seen in vitro) within macrophages in patients
with degenerative aortic valve stenosis),45 detection of
chlamydial macromolecules in diseased hosts in the absence
of cultivability (Chlamydia pneumoniae in human choroidal
FIGURE 12.1. Electron micrographs of

Chlamydia trachomatis showing (a) an EB with
cell wall and electron-dense core of nucleic
acids and (b) an RB with DNA and ribosomes
distributed in a brillar pattern. RB is
characteristically larger than EB (bars =
0.1 mm).
From Clark RB, Nachamkin I, Schatzki PF, et al:
Localization of distinct surface antigens on Chlamydia
trachomatis HAR-13 by immune electron microscopy
with monoclonal antibodies. Infect Immun 1982;
38:1273.
118
Chlamydial Disease
Proposed
Persistence
Phase
C
D
A
J
Life Cycle
of
Chlamydia
I
K
H
strains for vaccine development.33 There are other antigens

associated with species and serotype specicity which are
incorporated into the cell wall of C. trachomatis. In addition,
soluble antigens that are released into the supernatant fluids of
cell cultures infected by C. trachomatis have been described,71,72
including a new family of polymorphic outer membrane proteins
(POMPs). IncA is the prototype of exported proteins which
localize in the cytoplasmic surface of the inclusion membrane,73
inject the host cell by a type III secretion mechanism,7476 and
may provoke immunopathogenic responses in the host. Typespecic antigens have not been characterized for C. pneumoniae,
therefore isolates from this species are serologically identical.
F
G
Elementary Body (EB)
SYSTEMIC INFECTION OF THE HOST
Aberrant RB
Nucleus
Golgi Apparatus
FIGURE 12.2. Life cycle of Chlamydia organisms.
neovascular membranes due to age-related macular degeneration),46 and clinical antibiotic resistance.15 Further evidence to
discount that the in vivo evidence may represent enhancement
of an inapparent low-grade infections, are the demonstrated
similarities in chlamydial gene or protein expression between
persistent cell culture systems and tissue samples from sites of
chronic disease.4753 The mechanism by which chlamydiae
enter and exit the persistent phase is yet to be dened, but the
important survival advantage of a persistent phase warrants
the addition of this phase to the well-accepted biphasic life cycle
of chlamydiae (Fig 12.2).
NATURAL HISTORY OF CHLAMYDIAL

INFECTIONS
Spectrum of Chlamydia trachomatis Infections
Since C. trachomatis can infect columnar or transitional
epithelium at any anatomic site, multiple-organ involvement
is possible. The most frequently infected sites are those most
accessible to infected mucosal secretions such as the external
genital tract, conjunctivae, and upper respiratory tract. From
these external sites, infection can spread within an organ
system and result in infection of structures (e.g., salpingitis,
epididymitis, pneumonitis, perihepatitis) that are protected
against primary contact.7780 Infection can also spread from one
infected external site to another (e.g., urethra, cervix, rectum,
conjunctivae) by natural drainage of infected secretions or poor
personal hygiene. The oculogenital serovars of C. trachomatis
(A through K) can infect any squamocolumnar epithelial
mucosa. LGV serovars are more invasive and can infect lymph
nodes and associated structures.
CHAPTER 12
Reticulate Body (RB)
ANTIGENS
Chlamydiae contain both common antigens and speciesspecic antigens that play a role in pathogenesis and diagnosis
of infection. All chlamydiae share the genus-specic common
antigen which is a glycoprotein that is similar to the
lipopolysaccharide (LPS) found in the outer membranes of
Gram-negative bacteria.54 It contains a ketodeoxyoctanoic acidreactive moiety55 and is present in the outer membranes of both
EBs and RBs. Type-specic antigens have been characterized in
C. trachomatis and C. psittaci. The microimmunofluorescence
(MIF) test has identied 15 serovars of C. trachomatis:56,57
serovars A, B, Ba, and C are usually isolated in areas of endemic
trachoma,58 serovars D through K are the most prevalent
sexually transmitted5961 and ocular infection with these
serovars results in inclusion conjunctivitis, and serovars L1, L2,
and L3 are the agents of lymphogranuloma venereum58 (Table
12.1). Three additional serovars (Ba, Da, Ia, and L2a) of
C. trachomatis have more recently been identied.62 Speciesspecic and type-specic antigens of C. trachomatis are located
in the major outer membrane protein (MOMP),6365 encoded by
the ompA gene of C. trachomatis constitutes ~60% of its outer
membrane, has a molecular mass of 3842 kDa,6668 and has
four surface-exposed variable domains which confer serotypespecic epitopes, and are immunodominant.63,65,69 Part of
the reason that C. trachomatis evades the hosts immunologic
defenses is MOMP antigenic variation resulting from allelic
polymophism at the omp1 locus70 Molecular evaluation of the
major outer membrane protein (MOMP) gene (omp1) offers a
more precise method of characterizing C. trachomatis than does
immunotyping by MIF.70 Determination of omp1 genotypes will
be useful in epidemiologic studies to identify reservoirs and
transmission patterns of C. trachomatis and to select candidate
INFECTION AND INFLAMMATORY RESPONSE

Natural infection with C. trachomatis appears to confer little
protection against reinfection. Multiple or persistent infections
are essential characteristics in the pathogenesis of ocular
trachoma. Chlamydial infections elicit an inflammatory
response that is characterized by PMN predominance with a
shift to lymphocyte predominance and the formation of
lymphoid follicles on infected mucosal surfaces as the infection
progresses. PMNs have been shown to phagocytose chlamydial
EBs8183 during initial exposure of the host, and impede spread
of infection by EBs released into the extracellular milieu during
subsequent chlamydial growth cycles. The role of lymphocytes
is incompletely understood, but intact lymphocyte function is
apparently important, because duration of infection and
infection-related mortality rates from the mouse pneumonitis
strain of C. trachomatis were greater in athymic nude mice than
in immunocompetent animals.8488 Similarly, guinea pigs
treated with antithymocyte serum to suppress cell-mediated
immune function were unable to eliminate genital infection by
the guinea pig inclusion conjunctivitis strain of C. psittaci.89
Lymphoid follicle formation is characteristic of human ocular
and genital chlamydial infections.9096 There is thinning or loss
of epithelium overlying the follicles and they may become
necrotic as the disease progresses with resultant brosis and
scarring.
ANTIBODY RESPONSE
Our understanding of the role of antibody in natural infection
is incomplete. C. trachomatis infections cause immunoglobulin
M (IgM) and IgG antibodies to appear in the serum and IgG
119
SECTION 3
MICROBIOLOGY
and IgA antibodies to appear in mucosal secretions.9799 These
antibodies are directed against several chlamydial antigens,
including MOMP, as well as 60-kDa and 75-kDa proteins.100102
In vitro, EBs that have been exposed to antibodies fail to
replicate in cell culture, although they attach to the cells and
induce endocytosis.103105 In the mouse, high levels of serum
antibodies protect against the mouse pneumonitis strain of
C. trachomatis.87 In contrast, preexisting serum antibodies in
humans do not appear to protect against infection, but may
be important for containment and resolution of chlamydial
infections. Most persons in groups at high risk for sexually
transmitted infections have serum antibodies but are subject to
repeated infections from both previously unencountered
chlamydial serovar or genotype and reinfection with preexisting
serovar-specic antibody.100 Consistent with these ndings is
the observation that infants become infected with maternal
serovars of C. trachomatis even if they acquired maternal IgG
antibody transplacentally.106 In guinea pig inclusion conjunctivitis (GPIC), produced by a strain of C. psittaci, disease was
more prolonged, severe, and invasive when the humoral
antibody response was suppressed.107,108 In a study of women
with cervical C. trachomatis infection who underwent elective
abortion without prior antichlamydial treatment, ascending
infection and salpingitis occurred less frequently in patients
who had higher titers of serum antibodies.101 Although infection occurs at birth in infants with congenital C. trachomatis
infection, the incidence of pneumonia is highest during the
second and third months of life, a period that coincides with the
decline in titer of transplacentally acquired antibodies.109
CELL-MEDIATED IMMUNE RESPONSE

Cell-mediated immune responses (CMIs) to chlamydial infections, as detected by antigen-directed lymphocyte proliferation
assays, have been demonstrated in both humans and
animals.109,110 CMIs in animals have also been demonstrated
by induction of footpad swelling in response to local antigen
injection in the mouse pneumonitis model of chlamydial
infection.111 CMI appears to contribute to control and resolution
of infection. For example, transfer of T cells from mice with
normal immune function confers protection against the
prolonged infection and high mortality otherwise observed in
athymic mice infected with the mouse pneumonitis agent.86
The same serovar of C. trachomatis also produces nonresolving
genital infections in athymic mice but not in mice with an
intact CMI.88 Induction of cytotoxic T lymphocytes is another
CMI mechanism that may be important in the resolution of
chlamydial infections.112115 Although cytotoxicity was directed
principally against Chlamydia-infected cells mediated by the
cytokine IFN-g,116 nonspecic cytotoxicity against uninfected
cells was also noted (mediated by tumor necrosis factor alpha
(TNF-a)).117 Further studies are needed to delineate the role of
CMI in chlamydial infections.
SUSCEPTIBILITY TO ANTIMICROBIAL
DRUGS
120
The macrolide (erythromycin, azithromycin, and clarithromycin)

and the tetracycline (tetracycline, doxycycline, and minocycline)
antibiotics are structurally unrelated, but block chlamydial
protein synthesis by inhibition of the 50S and 30S ribosomal
subunits, respectively.118 Although their action is bacteriostatic,
they are the most effective therapeutic agents in the treatment
of chlamydial infections.119 Azithromycin given as a single dose
has become the treatment of choice for uncomplicated lower
genital infections with C. trachomatis120 and trachoma.121
Community wide treatment with azithromycin is part of efforts
to control trachoma.121,122 Fluoroquinolones may also be effective but are second- or third-line agents. Due to rapid development of resistance, rifampin cannot be recommended despite
good in vitro activity. Since chlamydial cell walls do not contain
peptidoglycan, it is not surprising that b-lactam antibiotics
remain ineffective against chlamydial infections.119 Aminoglycosides and cephalosporins are also not active against Chlamydia.
Key Features: Recommended Treatment for

Lymphogranuloma Venereum225
Recommended Regimen
Doxycycline 100 mg orally twice a day for 21 days
Alternative Regimen
Erythromycin base 500 mg orally four times a day for 21 days
Azithromycin 1.0 g orally once weekly for 3 weeks is probably
effective, although clinical data are lacking
Key Features: Recommended Treatment of Chlamydial

Urethritis/Cervicitis in Adults and Adolescents225
Recommended Regimens
Azithromycin 1 g orally in a single dose OR
Alternative Regimens
Erythromycin base 500 mg orally four times a day for 7 days OR
Erythromycin ethylsuccinate 800 mg orally four times a day for
7 days OR
Ofloxacin 300 mg orally twice a day for 7 days OR
Levofloxacin 500 mg orally once daily for 7 days
Key Features: Recommended Treatment Regimens for

Chlamydial Infections in Pregnancy225
Recommended Regimens
Amoxicillin 500 mg orally three times a day for 7 days
Alternative Regimens
7 days OR
14 days
Erythromycin estolate is contraindicated during pregnancy
because of drug-related hepatotoxicity. The lower dose 14-day
erythromycin regimens may be considered if gastrointestinal
tolerance is a concern
Key Features: Recommended Treatment Regimens for

Chlamydial Infections in Children225
Recommended Regimens for Children Who Weigh < 45 kg
Erythromycin base or ethylsuccinate 50 mg kg1 day1 orally
divided into 4 doses daily for 14 days
Recommended Regimen for Children Who Weigh >45 kg but
Who Are Aged <8 Years
Azithromycin 1 g orally in a single dose
Recommended Regimens for Children Aged >8 years
Sexual assault or sexual abuse of children must be considered.
Follow-up cultures are necessary to ensure that treatment has
been effective
Chlamydial Disease
Ophthalmia Neonatorum Caused by C. trachomatis

Considered for all infants aged <30 days who have conjunctivitis,
especially if the mother has a history of untreated Chlamydia
infection.
Diagnostic Considerations
Sensitive and specic methods used to diagnose chlamydial
ophthalmia in the neonate include both tissue culture and
nonculture tests (e.g., DFA tests, EIA, and NAAT).
The majority of nonculture tests are not FDA-cleared for the
detection of Chlamydia from conjunctival swabs, and clinical
laboratories must verify the procedure according to CLIA
regulations. Specimens must contain conjunctival cells, not
exudate alone. Specimens for culture isolation and nonculture
tests should be obtained from the everted eyelid using a Dacron
tipped swab or the swab specied by the manufacturers test kit.
A specic diagnosis of C. trachomatis infection conrms the need
for treatment not only for the neonate but also for the mother and
her sex partner(s). Ocular exudate from infants being evaluated for
chlamydial conjunctivitis also should be tested for N. gonorrhoeae.
Recommended Regimen
divided into 4 doses daily for 14 days.
Topical antibiotic therapy alone is inadequate for treatment of
chlamydial infection and is unnecessary when systemic treatment
is administered.
The efcacy of erythromycin treatment is ~80%; a second
course of therapy might be required and follow-up is necessary.
Infant Pneumonia Caused by C. trachomatis
Diagnostic Considerations
Specimens for chlamydial testing should be collected from the
nasopharynx. Tissue culture is the denitive standard for
chlamydial pneumonia. Nonculture tests (e.g., EIA, DFA, and NAAT)
can be used, although nonculture tests of nasopharyngeal
specimens have a lower sensitivity and specicity than nonculture
tests of ocular specimens. DFA is the only FDA cleared test for the
detection of C. trachomatis from nasopharyngeal specimens.
Tracheal aspirates and lung biopsy specimens, if collected, should
be tested for C. trachomatis. Because of the delay in obtaining test
results for Chlamydia, the decision to provide treatment for
C. trachomatis pneumonia must frequently be based on clinical
and radiologic ndings.
The results of tests for chlamydial infection assist in the
management of an infants illness and determine the need for
treating the mother and her sex partner(s).
Recommended Regimen
divided into 4 doses daily for 14 days.
The effectiveness of erythromycin in treating pneumonia caused
by C. trachomatis is ~80%; a second course of therapy might be
required. Follow-up of infants is recommended.
HOSTMICROBE INTERACTION IN
THE EYE
NATURAL HISTORY OF TRACHOMA
Blinding trachoma, the end-stage of a chronic process caused by
repeated infections with C. trachomatis, occurs in impoverished
populations living under conditions of poor hygiene.123125 The
disease is particularly prevalent in the Middle East and parts
of southeast Asia. In hyperendemic areas, infection is acquired
during infancy, and most children are infected by 2 years of
age.126 Primary infection induces purulent follicular conjunctivitis (except during the neonatal period). The follicles consist
of lymphoid germinal centers.127 Because lymphoid tissue is
absent from the conjunctivae of neonates, lymphoid follicles

do not form. Infection at this age produces acute purulent
conjunctivitis, but the tissue reaction is one of papillary hypertrophy.128 Primary infection resolves spontaneously and induces
transient protective immunity; in endemic areas, however,
reinfection is inevitable. The same serovar of C. trachomatis is
often transmitted reciprocally among members of a household.129
With repeated infections, healing is associated with central
degeneration and necrosis of lymphoid follicles, thinning of
the overlying conjunctival epithelium, and proliferation of
broblasts, resulting in broses and scarring.130 Uninterrupted
progression of this process eventually converts the normally
smooth and lubricating conjunctival epithelium into one that
is xerotic and cicatrized. Extensive brosis produces entropion
and trichiasis. End-stage blindness is the result of corneal
drying, ulceration, and scarring.
PATHOGENESIS OF TRACHOMA
Studies in Humans
The observation that repeated chlamydial infections are
characteristic of the course of blinding trachoma has led to the
concept that the disease constitutes an immunopathologic
response of the host to C. trachomatis infections.129,131 Initial
infection presumably induces immune sensitization of the host
but only transient or incomplete protective immunity.
Reinfections or relapses result in intensied inflammatory
reactions, brosis, scarring, and pannus formation. In vaccine
studies using inactivated EB as antigen, recipients immunized
with an antigen dose that proved to be inadequate to induce
immunity against infection developed more severe disease
with subsequent infections than did unvaccinated controls.132
Reinfection also frequently results in exacerbation of
trachoma.129131 Consistent with this observation is a report
that trachoma did not progress further in persons who moved
from an endemic to a nonendemic area where they were no
longer exposed to the pathogen.133 Immunopathogenesis is
further evidenced by the nding that in trachoma-endemic
areas, proliferative responses of peripheral blood lymphocytes to
stimulation by chlamydial antigens, a marker of CMI, are more
common in patients with trachoma than in controls without
disease.134
The apparent genetic susceptibility to trachoma further
supports this concept. In a study in Gambia, the frequencies of
the human leukocyte antigen (HLA) complex class I antigen,
HLA-A28, and the A6806 allele were signicantly greater in
patients with trachoma than in age-, sex-, and location-matched
controls.135 Immunopathology may be associated with HLAA6802-restricted T-lymphocyte responses. In Chlamydiaassociated involuntary tubal infertility, another disease of
suspected immunopathogenic origin, antibodies to the 60-kDa
C. trachomatis heat shock protein, a putative immunopathogenic
antigen, are more common in affected individuals than in
controls.136139 Heat shock or stress proteins are produced by
all prokaryotic and eukaryotic cells in response to damaging
stimuli such as elevated environmental temperature.125 They
are major antigens of many pathogens and appear to be important
to the immune response, including immune surveillance and
autoimmunity.140,141 In mice, the immune response to the
60-kDa heat shock protein of C. trachomatis is genetically
controlled.142 This observation adds support to the concept that
the outcome of chlamydial infections in humans may also have
a genetic component.
CHAPTER 12
Key Features: Recommended Management of

Chlamydial Infections in Infants225
Studies in Animals
Animal experiments support the hypothesis that trachoma is
an immunopathologic process induced by repeated ocular
121
SECTION 3
MICROBIOLOGY
infections with C. trachomatis.143 In primate studies, progressive
conjunctival and limbal scarring and pannus formation
occurred only in animals that had received more than one
chlamydial inoculation or that had previously been immunized
with an experimental trachoma vaccine.143149 Similar results
were seen regardless of the serovar involved (serovar A or
serovar E),150 which suggests that repeated ocular infection
induces trachoma. The inflammatory reaction decreased in
severity with repeated inoculations of both serovars, and
Chlamydia could not be reisolated from the eyes after six to
eight weekly inoculations, despite continuation of the inoculations. This is consistent with the fact that C. trachomatis
can seldom be isolated from the eyes of humans with advanced
trachoma. This progression of disease in the absence of
detectable Chlamydia organisms suggests that the immune
response is partially protective, but continued antigenic
stimulation elicits a pathologic immune response. Repeated
inoculation with live organisms was essential to development
of chronic disease.151,152
Taylor and co-workers, by infecting cynomolgus monkeys
with C. trachomatis,153,154 determined that internal antigens
(isolated by a soluble triton extract) rather than surface antigens
(MOMP, LPS) are the stimuli involved in the pathogenesis of
trachoma, not surface antigens.152,155 Ocular delayed hypersensitivity was similarly demonstrated in guinea pigs156 and the
ability of a triton extract of GPIC EBs to produce an inflammatory response in the eyes of monkeys previously infected
with C. trachomatis, suggests that the sensitizing antigen is
genus-specic rather than species-specic.153 Lymphocytes
in the inflammatory response were antigen-specic for
Chlamydia.149 In guinea pigs, infection of the conjunctivae,
vagina, or intestine, but not intramuscular injection of live
GPIC EBs, resulted in ocular sensitization and a delayed hypersensitivity reaction on subsequent conjunctival challenge with
triton-extracted antigen.157 This suggests that ocular delayed
hypersensitivity can be induced by prior infection of mucosal
surfaces, not only of the eye but other anatomic sites.158,159
Cytokines elaborated by the host in response to chlamydial
infections may also be important to the progression of trachoma.
In animal studies, chlamydial infections induce host production
of both IFN-g and TNF-a.160,161 TNF-a stimulates collagenase,
prostaglandin E2, and hyaluronic acid production by human
broblasts.162,163 IFN-g also stimulates hyaluronic acid
production.163,164
INCLUSION CONJUNCTIVITIS
Inclusion Conjunctivitis in Adults

Studies in western Europe and in the United States identied
C. trachomatis via culture as the pathogen in as many as 9% of
cases of acute conjunctivitis and 19% of cases of chronic
conjunctivitis.175181 In one study that limited patients to 2025
years of age, the isolation rate was as high as 23%.177 Adults
with chlamydial conjunctivitis frequently have a concurrent
genital infection. Presumably, poor personal hygiene results in
contamination of the conjunctivae by infected genital secretions.
Because repeated ocular infections are rare, corneal scarring,
although reported, appears to be unusual.
LABORATORY DIAGNOSIS
Key Features: Diagnostic Tests for Chlamydia Infections
Cell Culture
Clinical specimen cultured on cell monolayer (McCoy or HeLa)
Sensitivity is 7580% by expert laboratories; specicity is
~100%
Advantages are highly specic and all Chlamydia species can
be cultivated
Disadvantages are expense, high level of technical expertise,
stringent cold-chain transportation, and time until results
(37 days) have limited its use
Direct Fluorescent Assay
An antigen in the membrane of Chlamydia trachomatis (usually
MOMP) is detected directly by an antibody labeled with a
fluorochrome, examined under ultraviolet light
Sensitivity is 8090%; specicity is 95% compared to culture
Advantages are direct assessment of specimen adequacy,
cost-effective, rapid results (30 min), and no special
transportation
Disadvantages are highly trained personnel, performance
variability due to xation technique, number of EBs present,
serotype and antibody used
Nucleic Acid Amplication Test
Important advance in diagnosis of Chlamydia infection; uses
species-specic primers to amplify Chlamydia DNA
Highest sensitivity 90%; highest specicity for nonculture test
99100%
Advantages are not dependent on the viability of the organism
and able to detect to as low as 10 copies of Chlamydia DNA
Disadvantages are the inhibition by substances (problem
overcome by Amplicor, Roche); stringent lab conditions to
avoid carry-over lab contamination
Neonatal Inclusion Conjunctivitis
122
C. trachomatis is the most frequent cause of neonatal

conjunctivitis.165,166 When a pregnant woman has culturepositive cervical infection with C. trachomatis at the time of
labor and delivery, the infant born per vaginal birth has an
1850% chance of becoming clinically infected.167 The seroconversion rate of infection may be as high as 70%.106 The conjunctivae of the infant delivered via an infected birth canal
appear to be the usual site of initial infection; subsequently,
infection spreads to the nasopharynx.168 If untreated, the
infection may involve the lower respiratory tract and cause
pneumonia.167,169 The rectum and vagina may also become
colonized.169,171 Almost all infants with conjunctival infection
develop conjunctivitis within the rst 3 weeks of life, which,
even if it is not treated, is usually self-limited.168 In industrialized nations, infants seldom become reinfected, and progression to trachoma does not occur. In cases of persistent or
untreated infection, however, corneal micropannus and palpebral
conjunctival scarring occur occasionally.172174
Detection by Cell Culture

Cell cultures have been considered the gold standard for
detection of C. trachomatis, but the denition of gold standard
has been now dened by a combination of tests (culture, DFA,
PCR). The principal disadvantages of cell culture are that (1) it
may give false-negative results if the organism is inactivated by
improper collection, transport, or storage; (2) it requires special
laboratory facilities and experienced personnel; (3) it takes
several days to perform the test and obtain results; and (4) it is
expensive. Chlamydiae are relatively labile organisms and
viability is enhanced by keeping specimens cold and minimizing
transport time to the laboratory. Because Chlamydia organisms
are present in infected epithelial cells and not in the exudate
produced by infection, the specimen should contain as many
epithelial cells as possible. To collect conjunctival specimens,
one should cleanse the eye of exudate and swab the conjunctival
surface with pressure sufcient to exfoliate cells. Swabs with
Chlamydial Disease
CHAPTER 12
metal or plastic shafts rather than wood shafts are preferred,

because toxic products from wood may be leached into the
collection medium and have toxic effects on the cell culture into
which it is inoculated. Sucrose phosphate buffer is frequently
used as a collection medium.182 Antibiotics (usually aminoglycosides) and fungicides to which C. trachomatis organisms are
resistant are usually incorporated into the collection medium to
inactivate contaminating bacteria and yeast that otherwise
would grow in and destroy inoculated cell cultures. After
collection, specimens may be stored at 4C if they are to be
cultured within 2448 h. Specimens that cannot be cultured
within that time frame should be stored at 70C to retard
inactivation. Isolation rates are highest when specimens are
cultured promptly after collection.
Since C. trachomatis is an obligate intracellular parasite, it
replicates only in living cells. Although the organism was rst
successfully cultivated in 1957 in the yolk sacs of embryonated
eggs, this method is labor-intensive and less sensitive than the
cell culture technique that was developed later.183,184 The yolk
sac method is only used to prepare antigens for the MIF test
discussed below. Most laboratories use cell culture for isolation
and demonstration of intracytoplasmic inclusion by various
staining procedures. The cell types most frequently used for
cultivation and detection of C. trachomatis are McCoy cells
and HeLa 229 cells185,186 A nutrient-rich cell culture medium
is employed, and the cultures are treated with metabolic
inhibitors such as cycloheximide or cytochalasin B to prevent
the cells from competing with the parasite for nutrients.187,188
Despite this favorable microenvironment, C. trachomatis,
except for serovars L1, L2, and L3, does not readily infect cell
cultures. Infection requires enhancement by centrifugation of
inoculated cultures at 25003000 g for 60 min.189194 After
inoculation, cultures are usually incubated for 72 h at 35C
and then stained and examined for chlamydial cytoplasmic
inclusion bodies. Giemsas or iodine stains can be used to stain
the inclusions; however, the sensitivity of the method is
increased by staining with fluorescein-conjugated monoclonal
antibody prepared against C. trachomatis.195,196 Chlamydial
inclusions fluoresce with a bright apple-green color. Figure
12.3a shows an example of inclusions in an infected McCoy cell
culture stained with fluorescein-conjugated monoclonal
antibody.
C. psittaci can be isolated from respiratory tract secretions,
blood, and tissue biopsy specimens (spleen, liver) from patients
with ornithosis (psittacosis). The organism can be isolated by
inoculation of the yolk sac of embryonated eggs or of cell
cultures of L cells or McCoy cells. C. psittaci inclusion bodies
are detected by Giemsa staining of infected cell culture monolayers or impression smears of infected yolk sac membranes.
For isolation of C. pneumoniae, throat swabs or specimens
of respiratory tract secretions are obtained and placed in the
same transport medium that is used for C. trachomatis.
C. pneumoniae was originally isolated in HeLa 229 cells, but
HL, HEp-2, and H292 cell cultures have been reported to be
more sensitive.197202 Inclusions in infected cells can be
specically identied by staining with fluorescein-conjugated
monoclonal antibodies.
C. trachomatis is a biosafety level 2 (BL2) agent and is not
considered a dangerous pathogen in the laboratory. Occasional
reports of laboratory associated follicular conjunctivitis have
been reported. The LGV biovar is more invasive and after
aerosolization by sonication or centrifugation, pneumonia and
lymphadenitis has been reported. C. psittaci is a biosafety level
3 organism and needs to be handled in laboratories with BL 3
containment. C. pneumonia infections in the laboratory have
occurred, but these are mild.2
b
FIGURE 12.3. Diagnosis of Chlamydia trachomatis infections by
immunofluorescence test with monoclonal antibodies. (a) Fluoresceinconjugated antibody was reacted with McCoy cell culture 48 h after
infection with C. trachomatis. Fluorescing structures are
intracytoplasmic chlamydial inclusions (400). (b) A direct cervical
specimen from a patient with culture-conrmed chlamydial infection.
Fluorescing material consists of single or clumped chlamydial EBs or
RBs from infected and disrupted cervical mucosal cells (630).
From Tam MR, Stamm WE, Handseld HH, et al: Culture-independent diagnosis
of Chlamydia trachomatis using monoclonal antibodies. N Engl J Med 1984;
310:1146.
Direct Cytological Examination

C. trachomatis was discovered in 1907 by cytologic examination of conjunctival cells from patients with trachoma.203 In
patients with ocular trachoma or acute chlamydial inclusion
conjunctivitis, the juxtanuclear cytoplasmic inclusions of
C. trachomatis can often be detected in Giemsa-stained smears
of conjunctival cell scrapings.204 In inclusion conjunctivitis,
stained scrapings are positive in up to 90% of infants, but only
in 50% of adults.205207 In mild active ocular trachoma, it is
relatively insensitive with inclusion-bearing cells found in only
1030% of scrapings. In a study of genital infections, the
Giemsa method detected only 15% of infections of the male
urethra and 41% of cervical infections.208 Papanicolaou-stained
cervical smears are also insensitive and nonspecic for
detection of cervical infections.209,210
Antigen Detection
Direct staining of specimens by fluoresceinconjugated monoclonal antibody (DFA)
In this test, smears of cells obtained by swabbing infected
mucous membranes are stained with fluorescein-conjugated
123
SECTION 3
MICROBIOLOGY
monoclonal antibodies prepared against C. trachomatis. When
examined under a fluorescent microscope, intact inclusion
bodies or scattered EBs from ruptured cells fluoresce a bright
apple-green. The technique was rst used to detect urethral and
cervical infections, but it is equally useful for detection of
conjunctival infections.211216 Figure 12.3b shows a positive
cervical smear. The test can also be used for rectal specimens,
but the typically high concentrations of other bacteria in such
specimens sometimes produce false-positive results from crossreactive staining.217 Compared with cell culture, the sensitivity
of DFA testing in various reports has ranged from 70% to 100%,
and specicity appears to be greater than 95%.218 A study of
neonatal conjunctivitis reported sensitivity of 100% and
specicity of 94%.165 DFA testing has the following advantages:
(1) Unlike cell culture, DFA detects both viable and nonviable
Chlamydia organisms, therefore, the rigorous transport and
storage conditions that are essential for prevention of
inactivation are not as necessary; (2) The test is more rapid and
results are available in hours; (3) The cost of a DFA test is
approximately a fourth that of culture; (4) The adequacy of the
specimen can be assessed during the procedure by noting the
presence or absence of columnar or cuboidal epithelial cells.
Absence or paucity of these cells indicates an inadequate
specimen. The technique also has certain disadvantages: (1) It
requires a fluorescent microscope and an experienced microscopist who can distinguish between fluorescing chlamydial
particles and nonspecic fluorescence. (2) Cross-reactive
staining sometimes occurs in specimens that contain large
numbers of other bacteria. This is most common with rectal
specimens and is seldom a problem in conjunctival specimens.
Several DFA assays are commercially available. The antiMOMP monoclonal antibodies (Syva Microtak; Trinity Biotech)
are species-specic for C. trachomatis, and will not stain
C. psittaci or C. pneumoniae. Since MOMP is distributed
evenly on the surface of chlamydiae, the quality of fluorescence
is good and it takes only 30 min to perform. Monoclonal
antibodies to LPS (Pathnder; Kallestad) will stain all
chlamydiae and are distributed unevenly.
Enzyme immunoassay
124
In enzyme immunoassay (EIA), C. trachomatis antigen is

detected by a colorimetric signal generated by antigenantibody
reactions. A number of EIAs are commercially available and
they use either monoclonal or polyclonal antibodies to detect
chlamydial LPS, which is more soluble than MOMP. Like DFA,
EIA is quicker and less expensive than culture, and the viability
of C. trachomatis organisms in the specimen is irrelevant to the
validity of the test. Most EIAs take several hours to perform and
are suitable for batch processing.219 The test has an objective
end-point (photometric measurement of color intensity), in
contrast with the subjective interpretation required by
microscopic examination in DFA. However, the adequacy of the
specimen (presence of epithelial cells) cannot be assessed by
EIA.220 Like DFA, EIA is less sensitive and specic than
isolation of the organism in cell culture by an experienced
laboratory.221 When large numbers of specimens are processed,
however, EIA requires less technologist time per specimen than
DFA does because the objective (photometric) end-point of EIA
makes the test much less labor-intensive than the microscopic
examination required by DFA. The performance of commercial
EIAs for C. trachomatis varies considerably, but increases in
sensitivity have been achieved by using cycling enzymes to
amplify the signal component in the IDEA PCE test (DAKO
Ltd, Ely, UK).222224 These tests have a specicity of only 97%
which makes them not amenable to screen low prevalence
populations due to a low predictive value. With the use of
conrmatory tests, the specicity approaches 99.5%. Two types

of conrmatory tests are used. In one assay, all positive results
are repeated in the presence of a monoclonal antibody directed
against the type-specic epitope on the LPS.225 Another
approach is to use a second test by a different method such as a
DFA test based on MOMP detection to conrm an LPS-based
EIA.226
Nucleic acid tests

Nucleic acid hybridization (NAH) tests for C. trachomatis are
used in parts of the world as extensively as EIAs. One utilizes
DNARNA hybridization (PACE 2, Gen-Probe, San Diego, CA)
to enhance sensitivity to detect chlamydial RNA. It is about
as sensitive as the better antigen detection and cell culture
methods and is relatively specic.227228 Another NAH test uses
signal amplication to increase the sensitivity up to 90% of
the nucleic acid amplication (NAA) tests. Five NAA methods
are currently licensed for detection of C. trachomatis. They
are based on detection of chlamydial DNA or RNA using
amplication procedures such as polymerase chain reaction
(PCR), ligase chain reaction (LCR), chlamydial ribosomal RNA
using transcription-mediated amplication or strand displacement amplication. The PCR, LCR, and strand displacement
amplication assays amplify nucleotide sequences of the cryptic
plasmid present in each C. trachomatis EB. The transcriptionmediated amplication is directed against rRNA. Both the
cryptic plasmid of EB and rRNA are present in multiple copies,
so theoretically they should be able to detect less than one EB.
Sampling and specimen variability cause the actual sensitivity
to be lower.229 All assays are highly specic if crosscontamination is kept minimal. The NAA tests are more
sensitive than culture and other nonculture techniques. The
NAA methods are becoming the tests of choice in routine
clinical laboratories, especially for urogenital chlamydial
infections. However when organisms are needed for further
study, isolation in cell culture will continue to be used.
Serologic diagnosis
Chlamydial antibodies can be detected by complement xation
(CF), MIF testing, and enzyme-linked immunosorbent assay
(EIA),218 using group or species-specic antigens or a combination of these to measure immunoglobulin G (IgG), IgA,
IgM, or total classes of antibodies to individual or multiple
chlamydial serovars. The CF test is rarely performed today, is
based on the group-specic chlamydial LPS, which is relatively
insensitive, and was used for LGV. The genus-specic CF test
can be used for serologic diagnosis of psittacosis (C. psittacosis).
MIF, in contrast, is a sensitive and specic test that detects both
IgG- and IgM-class antibodies in serum, tears, and genital
secretions.230 The MIF test is most useful in epidemiologic
studies; it has limited diagnostic application in C. trachomatis
infections due to many high-risk patients having already experienced a primary infection and it often requires retrospective
pairing of sera.230 Two exceptions are: (1) chlamydial pneumonia
with detection of IgM-class antibodies (primary infection)
especially in infants and up to 70% sensitivity in adults231,232
and (2) C. trachomatis ocular infections where the presence of
IgG or IgA chlamydial antibodies in tears appears to correlate
with disease activity.230,233235 Several recombinant EIA tests are
commercially available for detection of chlamydial antigens by
either monoclonal or polyclonal antibodies to detect a
chlamydiae-specic recombinant fragment of LPS, 3-deoxy-Dmanno-2-octulopyranosonic acid. This reduces cross-reactivity
from other Gram-negative bacteria containing LPS. Comparisons of these recombinant immunoassays with traditional
CF or the gold standard MIF test has shown a slightly lower
sensitivity and specicity for these serum antibodies to peptides
Chlamydial Disease
Key Features: Antimicrobial Susceptibility
Beta-lactam antibiotics are ineffective

Mechanism of action is inhibition of 50S and 30S ribosomal
subunits
SUMMARY
Despite the long recognition of chlamydial infections, our
knowledge of its pathogenesis and immunology, detection,
treatment, and most importantly prevention, continues to lag.
C. trachomatis and C. pneumoniae are pathogens of humans
and have no animal reservoirs. C. psittaci is principally a
pathogen of birds that causes pneumonia and systemic
infection when transmitted to humans. C. trachomatis is the

most prevalent sexually transmitted pathogen in Western
societies and an important cause of acute and chronic
conjunctivitis, including trachoma. Protective immunity is
incomplete; repeated infections often cause brosis and scarring
of affected tissues, believed as a result of an immunopathologic
process. The most recent advances described here are a new
taxonomic classication, an additional pathway of persistence
and latency to its previously described biphasic life-cycle, and
newer molecular diagnostics for detection.239 The treatment
choices (tetracyclines, macrolides) remain essentially unchanged,240 and, an effective vaccine continues to be elusive by
our incomplete understanding of the immunology and
pathogenesis of chlamydial infections.
ACKNOWLEDGEMENT
The author of this chapter acknowledges Joseph M. Thomas, Alfred D.
Heggie, and Jonathan H. Lass for their contributions from Albert &
Jakobiecs Principles and Practice of Ophthalmology, Second edition.
REFERENCES
1. Moulder JW: Order Chlamydiales and
family Chlamydiaceae. In: Krieg NR, Holt
JG, eds. Bergeys manual of systemic
bacteriology. Baltimore, MD: The Williams
and Wilkins ; 1984:729739.
2. Mahony JB, Coombes BK, Chernesky MA:
Chlamydia and Chlamydophila. In: Murray
PR, Baron EJ, Jorgensen JH, et al, ed.
Manual of clinical microbiology. 8th edn.
Washington, DC: ASM Press;
0000:9911004.
3. Gordon FB, Quan AL: Occurrence of
glycogen in inclusions of the psittacosislymphogranuloma venereum-trachoma
agents. J Infect Dis 1965; 115:186196.
4. Matsumoto A: Structural characteristics of
chlamydial bodies. In: Barron AL, ed.
Microbiology of Chlamydia. Boca Raton,
FL: CRC Press; 1988:2145.
5. Kuo C-C, Chen H-H, Wang S-P, et al:
Identication of a new group of Chlamydia
psittaci strains called TWAR. J Clin
Microbiol 1986; 24:10341037.
6. Rice CE: Carbohydrate matrix of the
epithelial cell inclusion in trachoma. Proc
Soc Exp Biol Med 1935; 33:317.
7. Pudjiatmoko, Fukushi H, Ochiai Y, et al:
Phylogenetic analysis of the genus
Chlamydia based on 16S rRNA gene
sequences. Int Syst Bacteriol 47:425431.
8. Everett KD, Bush RM, Andersen AA:
Emended description of the order
Chlamydiales, proposal of
Parachlamydiaceae fam. nov. and
Simkaniaceae fam. nov., each containing
one monotypic genus, revised taxonomy of
the family Chlamydiaceae, including a new
genus and ve new species, and standards
for the identication of organisms. Int J
Syst Bacteriol 1999; 49(Pt 2):415440.
9. Fox A, Rogers JC, Gilbart J, et al: Muramic
acid is not detectable in Chlamydia psittaci
or Chlamydia trachomatis by gas
chromatography-mass spectrometry. Infect
Immun 1990; 58:835837.
10. Garret AJ, Harrison MJ, Manire GP: A
search for the bacterial mucopeptide
component, muramic acid, in Chlamydiae.
J Gen Microbiol 1974; 80:315.
11. Stephens RS, Kalman S, Lammel C, et al:

Genome sequence of an obligate
intracellular pathogen of humans: Chlamydia
trachomatis. Science 1998; 282:754759.
12. Stamm WE, Jones RB, Batteiger BE:
Introduction to Chlamydial diseases. In:
Mandell GL, Bennett JE, Dolin R, eds.
Principles and practice of infectious
diseases. Philadelphia, PA: Churchill
Livingstone; 2005.
13. Weiss E, Wilson NN: Role of exogenous
adenosine triphosphate in catabolic and
synthetic activities of Chlamydia psittaci.
J Bacteriol 1969; 97:719.
14. Mathews S, George C, Flegg C, et al:
Differential expression of ompA, ompB,
pyk, nlpD and Cpn0585 genes between
normal and interferon-gamma treated
cultures of Chlamydia pneumoniae. Microb
Pathog 2001; 30:337345.
15. Hogan RJ, Mathews SA, Mukhopahhyay S,
et al: Chlamydial persistence: beyond the
biphasic paradigm. Infect Immun 2004;
72:18431855.
16. Rockey DD: Chlamydial interactions with
host cells. In: Schaecter J, Christiansen G,
Clarke IN, et al, eds. Chlamydial Infections:
proceedings of the tenth international
symposium on human chlamydial infection.
San Francisco: International Chlamydial
Symposium. 2002:3545.
17. Zhang JP, Stephens RS: Mechanism of C.
trachomatis attachment to eukaryotic cells.
Cell 1992; 69:861869.
18. Wuppermann FN, Hegemann JH,
Jantos CA: Heparan sulfate-like
glycosaminoglycan is a cellular receptor for
Chlamydia pneumoniae. J Infect Dis 2001;
184:181187.
19. Friis RR: Interaction of L cells and
Chlamydia psittaci: entry of the parasite
and host response to its development.
J Bacteriol 1972; 110:706.
20. Hackstadt T, Scidmore MA, Rocky DD:
Lipid metabolism in Chlamydia
trachomatis-infected cells: directed
trafcking of Golgi-derived sphingolipids to
the chlamydial inclusion. Proc Natl Acad
Sci USA 1995; 92:4877.
21. Hackstadt T, Rockey DD, Heinzen RA,

Scidmore MA: Chlamydia trachomatis
interrupts an exocytic pathway to acquire
endogenously synthesized sphingomyelin
in transit from the Golgi apparatus to the
plasma membrane. EMBO J 1996; 15:964.
22. Tamura A, Matsumoto A, Higashi N:
Purication and chemical composition of
reticulate bodies of the
meningopneumonitis organisms. J Bacteriol
1967; 93:2003.
23. Tamura A, Matsumoto A, Manire GP, et al:
Electron microscopic observations on the
structure of the envelopes of mature
elementary bodies and developmental
reticulate forms of Chlamydia psittaci.
J Bacteriol 1971; 105:355.
24. Bavoil P, Olin A, Schachter J: Role of
disulde bonding in outer membrane
structure and permeability in Chlamydia
trachomatis. Infect Immun 1984; 44:479.
25. Wilbert J, Newhall V: Biosynthesis and
disulde cross-linking of outer membrane
components during the growth cycle of
Chlamydia trachomatis. Infect Immun
1987; 55:162.
26. Matsumoto A: Electron microscopic
observations of surface projections and
related intracellular structures of Chlamydia
organisms. J Electron Microsc 1981;
30:315.
27. Hatch TP, Allan I, Pearce JH: Structural and
polypeptide differences between envelopes
of infective and reproductive life cycle
forms of Chlamydia species. J Bacteriol
1984; 157:13.
28. Hatch TP, Miceli M, Sublett JE: Synthesis
of disulde-bonded outer membrane
proteins during the developmental cycle of
Chlamydia psittaci and Chlamydia
trachomatis. J Bacteriol 1986; 165:379.
29. Todd WJ, Caldwell HD: The interaction of
Chlamydia trachomatis with host cells:
ultrastructural studies of the mechanism of
release of a biovar II strain from HeLa 229
cells. J Infect Dis 1985; 151:1037.
29a. Moulder JW: Interaction of chlamydiae and
host cells in vitro. Microbiol Rev 1991;
55:143190.
CHAPTER 12
or recombinant antigens than obtained using whole EBs as the

antigen.236238 It has been postulated that these results reflect
individual variability in humoral response in a population to
single chlamydial antigens.
125
SECTION 3
MICROBIOLOGY
126
29b. Wolf K, Fischer E, Hackstadt T:

Ultrastructural analysis of developmental
events in Chlamydia pneumonia-infected
cells. Infect Immun 2000; 68:23792385.
29c. Ward ME: The chlamydial developmental
cycle. In: Barron AL, ed. Microbiology of
Chlamydia. Boca Raton, FL: CRC Press;
1988:7195.
30. Meyer KF, and Eddie B: Latent psittacosis
infections in shell parakeets. Proc Soc Exp
Biol Med 1933; 30:484488.
31. Beatty WL, Morrison RP, Byrne GI:
Persistent chlamydiae: from cell culture to
a paradigm for chlamydial pathogenesis.
Microbiol Re 1994; 58:686699.
32. Clark RB, Schatzki PF, Dalton HP:
Ultrastructural analysis of the effects of
erythromycin on the morphology and
developmental cycle of Chlamydia
trachomatis HAR-13. Arch Microbiol 1982;
133:278282.
33. Matsumoto A, Manire GP: Electron
microscopic observations on the effects of
penicillin on the morphology of Chlamydia
psittaci. J Bacteriol 1970; 101:278285.
34. Pantoja LG, Miller RD, Ramirez JA, et al:
Characterization of Chlamydia pneumoniae
persistence in Hep-2 cells treated with
gamma interferon. Infect Immun 2001;
69:79277932.
35. Beatty WL, Byrne GI, Morrison RP:
Morphologic and antigenic characterization
of interferon-gamma mediated persistent
Chlamydia trachomatis infection in vitro.
90:39984002.
36. Beatty WL, Morrison RP, Byrne GI:
Reactivation of persistent Chlamydia
trachomatis infection in cell culture. Infect
Immun 1995; 63:199205.
37. Hsia R-C, Ohayon H, Gounon P, et al:
Phage infection of the obligate intracellular
bacterium, Chlamydia psittaci strain Guinea
Pig Inclusion Conjunctivitis. Microbes Infect
2000; 2:761772.
38. Harper A, Pogson CI, Jones ML, Pearce
JH: Chlamydial development is adversely
affected by minor changes in amino acid
supply, blood plasma amino acid levels,
and glucose deprivation. Infect Immun
2000; 68:14571464.
39. Al-Younes HM, Rudel T, Brinkmann V, et al:
Low iron availability modulates the course
of Chlamydia pneumoniae infection. Cell
Microbiol 2001; 3:427437.
40. Airenne S, Surcel H-M, Alakarppa H, et al:
Chlamydia pneumoniae infection in human
monocytes. Infect Immun 1999;
67:14451449.
41. Koehler L, Nettelnbreker E, Hudson AP, et al:
Ultrastructural and molecular analyses of the
persistence of Chlamydia trachomatis
(serovar K) in human monocytes. Microb
Pathog 1997; 22:133142.
42. Ofcer JE, Brown A: Serial changes in virus
and cells in cultures chronically infected
with psittacosis virus. Virology 1961;
14:8899.
43. Lee CK: Factors affecting the rate at which
a trachoma strain of Chlamydia trachomatis
establishes persistent infections in mouse
broblasts (McCoy cells). Infect Immun
1981; 33:954957.
44. Thygeson P: Epidemiologic observations
on trachoma in the United States. Invest
Ophthalmol 1963; 2:482489.
45. Skowasch D, Yeghiazaryan K, Schrempf S,
et al: Persistence of Chlamydia
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
pneumoniae in degenerative aortic valve

stenosis indicated by heat shock protein60 homologues. J Heart Valve Dis 2003;
12:6875.
Kalayoglu MV, Bula D, Arroyo J, et al:
Identication of Chlamydia pneumoniae
within human choroidal neovascular
membranes secondary to age-related
macular degeneration. Graefes Arch Clin
Exp Ophthalmol 2005; 243:10801090.
Beatty WL, Morrison RP, Byrne GI:
Immunoelectron-microscopic quantitation
of different levels of chlamydial proteins in
a cell culture model of persistent
Chlamydia trachomatis infection. Infect
Immun 1994; 62:40594062.
Kinnunen J, Paavonen J, Surcel H-M: Heat
shock protein 60 specic T-cell response in
chlamydial infections. Scand J Immunol
2001; 54:7681.
Byrne GI, Ouellette SP, Wang Z, et al:
Chlamydia pneumoniae expresses genes
required for DNA replication but not
cytokinesis during persistent infection of
HEp-2 cells. Infect Immun 2001;
69:54235429.
Caldwell HD, Wood H, Crane D, et al:
Polymorphisms in Chlamydia trachomatis
tryptophan synthase genes differentiate
between genital and ocular isolates. J Clin
Investig 2003; 11:17571769.
Belland RJ, Nelson DE, Virok D, et al:
Transcription analysis of chlamydial growth
during IFN-gamma-mediated persistence
and reactivation. Proc Natl Acad Sci USA
2003; 100:1597115976.
Fehlner-Gardiner C, Roshick C, Carlson JH,
et al: Molecular basis dening human
Chlamydia trachomatis tissue tropism:
a possible role for tryptophan synthase.
J Biol Chem 2002; 277:2689326903.
Gerard HC, Freise J, Wang Z, et al:
Chlamydia trachomatis genes whose
products are related to energy metabolism
are expressed differentially in active vs
persistent infection. Microbes Infect 2002;
4:1322.
Nurminen M, Leinonen M, Saikku P, et al:
The genus-specic antigen of Chlamydia:
resemblance to the lipopolysaccharide of
enteric bacteria. Science 1983; 220:1279.
Maaheimo H, Kosma P, Brade L, et al:
Mapping the binding of synthetic
disaccharides representing epitopes of
chlamydial lipopolysaccharide to antibodies
with NMR. Biochemistry 2000;
39:1277812788.
Wang S-P, Kuo C-C, Grayston JT: A
simplied method for immunological typing
of trachoma-inclusion conjunctivitislymphogranuloma venereum organisms.
Infect Immun 1973; 7:356.
Newhall WJ, Terho P, Wilde CE, et al:
Serovar determination of Chlamydia
trachomatis isolates by using type-specic
monoclonal antibodies. J Clin Microbiol
1986; 23:333.
Grayson JT, Wang S-P: New knowledge of
chlamydiae and the diseases they cause.
J Infect Dis 1975; 132:87.
Thompson JE, Washington AE:
Epidemiology of sexually transmitted
Chlamydia trachomatis infections.
Epidemiol Rev 1983; 5:96.
Kuo C-C, Wang S-P, Holmes KK, et al:
Immunotypes of Chlamydia trachomatis
isolates in Seattle, Washington. Infect
Immun 1983; 41:865.
61. Barnes RC, Wang S-P, Kuo C-C, et al:

Immunotyping of Chlamydia trachomatis
with monoclonal antibodies in a solidphase enzyme immunoassay. J Clin
Microbiol 1985; 22:609.
62. Wang S-P, Grayston JT: Three new
serovars of Chlamydia trachomatis: Da, Ia,
and L2a. J Infect Dis 1991; 163:403.
63. Stephens RS, Sanchez-Pescador R, Wagar
EA, et al: Diversity of Chlamydia
trachomatis major outer protein genes.
J Bacteriol 1987; 169:3879.
64. Stephens RS, Wagar EA, Schoolnik GK:
High-resolution mapping of serovar-specic
and common antigenic determinants of the
major outer membrane protein of Chlamydia
trachomatis. J Exp Med 1988; 167:817.
65. Baehr W, Zhang YX, Joseph T, et al:
Mapping antigenic domains expressed by
Chlamydia trachomatis major outer
membrane protein genes. Proc Natl Acad
Sci USA 1988; 85:4000.
66. Caldwell HD, Kromhout J, Schachter J:
Purication and partial characterization of
the major outer membrane protein of
Chlamydia trachomatis. Infect Immun 1981;
31:1161.
67. Caldwell HD, Schachter J: Antigenic
analysis of the major outer membrane
protein of Chlamydia species. Infect Immun
1982; 35:1024.
68. Batteiger BE, Newhall WJ, Terho P, et al:
Antigenic analysis of the major outer
membrane protein of Chlamydia
trachomatis with murine monoclonal
antibodies. Infect Immun 1986; 53:530.
69. Zhang YX, Stewart S, Joseph T, et al:
Protective monoclonal antibodies recognize
epitopes located on the major outer
membrane protein of Chlamydia
trachomatis. J Immunol 1987; 138:575.
70. Dean D, Schachter J, Dawson CR,
Stephens RS: Comparison of the major
outer membrane protein variant sequence
regions of B/Ba isolates: a molecular
epidemiologic approach to Chlamydia
trachomatis infections. J Infect Dis 1992;
166:383.
71. Richmond SJ, Stirling P: Localization of
chlamydial group antigen in McCoy cell
monolayers infected with Chlamydia
trachomatis or Chlamydia psittaci. Infect
Immun 1981; 34:561.
72. Stuart ES, MacDonald AB: Genus
glycolipid exoantigen from Chlamydia
trachomatis: component preparation,
isolation and analyses. In: Oriel D, Ridgway
G, Schachter J, et al, eds. Chlamydial
infections. Proceedings of the 6th
International Conference on Human
Chlamydial Infections. Cambridge:
Cambridge University Press;
1986:122125.
73. Rockey DD, Grosenbach D, Hruby DE,
et al: Chlamydia psittaci IncA is
phosphorylated by the host cell and is
exposed on the cytoplasmic face of the
developing inclusion. Mol Microbiol 1997;
24:217228.
74. Stephens RS, Kalman S, Lammel C, et al:
Genome sequence of an obligate
intracellular pathogen of humans:
Chlamydia trachomatis. Science 1998;
282:754759.
75. Hsia R-C, Pannekoek Y, Ingerowski E,
Bavoil PM: Type III secretion genes identify
a putative virulence locus of Chlamydia.
Mol Microbiol 1997; 25:351359.
76. Bavoil PB, Hsia R-C: Type III secretion in

Chlamydiae: a case of dj vu? Mol
Microbiol 1998; 28:860862.
77. Berger RE, Alexander ER, Monda GD, et al:
Chlamydia trachomatis as a cause of acute
idiopathic epididymitis. N Engl J Med
1978; 298:301.
78. Thompson SE, Dretler RH: Epidemiology
and treatment of chlamydial infections in
pregnant women and infants. Rev Infect
Dis 1982; 4:747.
79. Wolner-Hanssen P, Westrom L, Mardh P-A:
Perihepatitis and chlamydial salpingitis.
Lancet 1986; 1:901.
80. Wang S-P, Eschenbach DA, Holmes KK,
et al: Chlamydia trachomatis infection in
Fitz-Hugh-Curtis syndrome. Am J Obstet
Gynecol 1980; 139:1034.
81. Zvillich M, Sarov I: Interaction between
human polymorphonuclear leukocytes and
Chlamydia trachomatis elementary bodies:
electron microscopy and chemiluminescent
response. J Gen Microbiol 1985; 131:2627.
82. Yong EC, Chi EY, Chen W-J, et al:
Degradation of Chlamydia trachomatis in
human polymorphonuclear leukocytes: an
ultrastructural study of peroxidase positive
phagolysosomes. Infect Immun 1986;
53:427.
83. Register KB, Davis CH, Wyrick PB, et al:
Non-oxidative antimicrobial effects of
human polymorphonuclear leukocyte
granule proteins on Chlamydia species in
vitro. Infect Immun 1987; 55:2420.
84. Williams DM, Schachter J, Drutz DJ et al:
Pneumonia due to Chlamydia trachomatis
in the immunocompromised (nude) mouse.
J Infect Dis 1981; 143:238.
85. Williams DM, Schachter J, Grubbs B, et al:
The role of antibody in host defense
against the agent of mouse pneumonitis.
J Infect Dis 1982; 145:200.
86. Williams DM, Schachter J, Coalson JE,
et al: Cellular immunity to the mouse
pneumonitis agent. J Infect Dis 1984;
149:630.
87. Williams DM, Schachter J, Weiner MH et al:
Antibody in host defense against mouse
pneumonitis agent (murine Chlamydia
trachomatis). Infect Immun 1984; 45:674.
88. Rank RG, Soderberg LSF, Barron AL:
Chronic chlamydial genital infection in
congenitally athymic nude mice. Infect
Immun 1985; 48:847.
89. Rank RG, Barron AL: Effect of
antithymocyte serum on the course of
chlamydial genital infection in female
guinea pigs. Infect Immun 1983; 41:876.
90. Kuo C-C, Chen W-J: A mouse model of
Chlamydia trachomatis pneumonitis.
J Infect Dis 1980; 141:198.
91. Taylor HR, Prendergast RA, Dawson CR,
et al: An animal model for cicatrizing
trachoma. Invest Ophthalmol Vis Sci 1981;
21:422.
92. Patton DL, Kuo C-C, Wang S-P, et al:
Chlamydial salpingitis in subcutaneous
mbrial transplants in monkeys. In: Oriel D,
Ridgway G, Schachter J, et al, eds.
Chlamydial infections. Proceedings of the
6th International Symposium on Human
Cambridge University Press; 1986:367370.
93. Quinn TC, Kappus EW, James SP: The
immunopathogenesis of lymphogranuloma
venereum rectal infection in primates. In:
Oriel D, Ridgway G, Schachter J, et al, eds.
94.
95.
96.
97.
98.
99.
100.
101.
102.
103.
104.
105.
106.
107.

1986:404407.
Dunlop EMC, Hare MJ, Darougar S, et al:
Chlamydial infection of the urethra in men
presenting because of nonspecic
urethritis. In: Nichols RL, ed. Trachoma and
related disorders. Amsterdam: Excerpta
Medica; 1971:494500.
Schachter J, Dawson CR: Human
Chlamydial infections. Littleton, MA: PSG;
1978:63109.
Dunlop EMC, Garner A, Darougar S, et al:
Colposcopy, biopsy and cytology results in
women with chlamydial cervicitis.
Genitourin Med 1989; 65:22.
Wang S-P, Grayston JT:
Microimmunofluorescence antibody
responses in Chlamydia trachomatis
infection: a review. In: Mardh
P-A, Holmes KK, Oriel JD, et al, eds.
5th International Conference on Human
Chlamydial Infections. Amsterdam: Elsevier
Biomedical; 1982:301316.
Williams DW: Stimulator of the immune
response. In: Barron AL, ed. Microbiology
of Chlamydia. Boca Raton, FL: CRC Press;
1988:209216.
Wager EA, Schachter J, Bavoil P, Stephens,
RS: Differential human serologic response
to two 60,000 molecular weight Chlamydia
trachomatis antigens. J Infect Dis 1990;
162:922.
Jones RB, Batteiger BE: Human immune
responses to Chlamydia trachomatis
infections. In: Oriel D, Ridgway G, Schachter
J, et al, eds. Chlamydial infections.
Proceedings of the 6th International
Symposium on Human Chlamydial
Infections. Cambridge: Cambridge
Brunham RC, Maclean I, MacDowell J,
et al: Chlamydia trachomatis antigen
specic serum antibodies among women
who did and did not develop acute
salpingitis following therapeutic abortion.
In: Oriel D, Ridgway G, Schachter J, et al,
eds. Chlamydial infections. Proceedings of
the 6th International Symposium on Human
1986:221224.
Maclean IW, Peeling RW, Brunham RC:
Characterization of Chlamydia trachomatis
antigens with monoclonal and polyclonal
antibodies. Can J Microbiol 1988; 34:141.
Howard LV: Neutralization of Chlamydia
trachomatis in cell culture. Infect Immun
1975; 11:698.
Caldwell HD, Perry LJ: Neutralization of
Chlamydia trachomatis infectivity with
antibodies to the major outer membrane
protein. Infect Immun 1982; 38:745.
Peeling R, Maclean IW, Brunham RC: In
vitro neutralization of Chlamydia
trachomatis with monoclonal antibody to
an epitope on the major outer membrane
protein. Infect Immun 1984; 46:484.
Schachter J, Grossman M, Holt J, et al:
Prospective study of chlamydia infection in
neonates. Lancet 1979; 2:377.
Rank RG, White HJ, Barron AL: Humoral
immunity in the resolution of genital
infection in female guinea pigs infected
with the agent of guinea pig inclusion
conjunctivitis. Infect Immun 1979; 26:573.
108. White HJ, Rank RG, Soloff BL, et al:

Experimental chlamydial salpingitis in
immunosuppressed guinea pigs infected in
the genital tract with the agent of guinea
pig inclusion conjunctivitis. Infect Immun
1979; 26:728.
109. Kunimoto E, Brunham RC: Human immune
response and Chlamydia trachomatis
infection. Rev Infect Dis 1985; 7:665.
110. Hanna L, Schmidt L, Sharp M, et al:
Human cell-mediated immune response to
chlamydial antigens. Infect Immun 1979;
23:412.
111. Barron AL, Rank RG, Moses EB: Immune
response in mice infected in the genital
tract with mouse pneumonitis agent
(Chlamydia trachomatis serovar). Infect
Immun 1984; 44:82.
112. Lammert JK: Cytotoxic cells induced after
Chlamydia psittaci infection in mice. Infect
Immun 1982; 35:1011.
113. Byrne GI, Krueger DA: In vitro expression
of factor-mediated cytotoxic activity
generated during the immune response to
Chlamydia in the mouse. J Immunol 1985;
134:4189.
114. Pavia CS, Schachter J: Failure to detect
cell-mediated cytotoxicity against
Chlamydia trachomatis-infected cells. Infect
Immun 1983; 39:1271.
115. Qvigstad E, Hirschberg H: Lack of cellmediated cytotoxicity towards Chlamydia
trachomatis infected target cells in humans.
Acta Pathol Microbiol Immunol Scand
[Copenh] 1984; 92:153.
116. Byrne GI, Grubbs B, Marshall TJ, et al:
Gamma-interferon-mediated cytotoxicity
related to murine Chlamydia trachomatis
infection. Infect Immun 1988; 56:2023.
117. Williams DM, Bonewald LF, Roodman GD,
et al: Tumor necrosis factor alpha is a
cytotoxin induced by murine Chlamydia
trachomatis infection. Infect Immun 1989;
57:1351.
118. Edwards DI: Antimicrobial drug action.
Baltimore, MD: Univeristy Park;
1980:200216.
119. Gerding DN, Peterson LR, Hughes CE,
et al: Extravascular antimicrobial
distribution in man. In: Lorian V, ed.
Antibiotics in laboratory medicine.
Baltimore, MD: Williams & Wilkins;
1986:938994.
120. Martin DH, Mroszkowski TF, Dalu ZA, et al:
A controlled trial of a single dose of
azithromycin for the treatment of
chlamydial urethritis and cervitis. N Engl J
Med 1992; 327:921.
121. Bailey RL, Arullendran P, Whittle HC,
Mabey DCW: Randomised controlled trial
of single-dose azithromycin in treatment of
trachoma. Lancet 1993; 342:453.
122. Schachter J, West, S, Mabey D, et al:
Azithromycin in control of trachoma. Lancet
1999; 354:630635.
123. Assaad FA, Sundaresan T, Maxwell-Lyons
F: The household pattern of trachoma in
Taiwan. Bull World Health Organ 1971;
44:605.
124. Jones BR: Prevention of blindness from
trachoma. Trans Ophthalmol Soc UK 1975;
95:16.
125. Dawson CR, Jones BR, Tarizzo ML: Guide
to trachoma control. Geneva: World Health
Organization; 1981:1224.
126. Dawson CR, Daghfous M, Messodi M,
et al: Severe endemic trachoma in Tunisia.
Br J Ophthalmol 1976; 60:245.
CHAPTER 12
Chlamydial Disease
127
SECTION 3
MICROBIOLOGY
128
127. Schachter J, Dawson CR: Human

Chlamydial Infections. Littleton, MA: PSG;
1978:65.
128. Thygeson PL: The etiology of inclusion
blenorrhea. Am J Ophthalmol 1934;
17:1019.
129. Grayston JT, Wang S-P, Yeh L-J, et al:
Importance of reinfection in the
pathogenesis of trachoma. Rev Infect Dis
1985; 7:717.
130. Kuo C-C: Host response. In: Barron AL, ed.
Microbiology of Chlamydia. Boca Raton,
FL: CRC Press; 1988:193208.
131. Grayston JT, Yeh L-H, Wang S-P, et al:
Pathogenesis of ocular Chlamydia
trachomatis infections in humans. In:
Hobson D, Holmes KK, eds.
Nongonoccocal urethritis and related
infections. Washington, DC: American
Society for Microbiology; 1977:113125.
132. Grayston JT, Wang S-P, Lin HM, et al:
Trachoma vaccine studies in volunteer
students of the National Defense Medical
Center. II. Response to challenge eye
inoculation of egg grown trachoma virus.
Chin Med J 1961; 8:312.
133. Detels R, Alexander ER, Dhir SP: Trachoma
in Punjabi Indians in British Columbia: a
prevalence study with comparisons to
India. Am J Epidemiol 1966; 84:81.
134. Mabey DCW, Holland MJ, Bailey RL, et al:
In vitro studies of cell mediated immunity to
chlamydial antigens in trachoma. In:
Bowie WR, Caldwell HD, Jones RP, et al,
eds. Chlamydial infections. Cambridge:
1990:283286.
135. Conway DJ, Holland MJ, Campbell AE,
et al: HLA class I and II polymorphisms and
trachomatous scarring in a Chlamydia
trachomatis-endemic population. J Infect
Dis 1996; 174:643.
136. Brunham RC, Peeling R, Maclean I, et al:
Chlamydia trachomatis-associated ectopic
pregnancy: serologic and histologic
correlates. J Infect Dis 1992; 165:1076.
137. Toye B, Laferriere C, Claman P, et al:
Association between antibody to the
chlamydial heat-shock protein and tubal
infertility. J Infect Dis 1993; 168:1236.
138. Arno JN, Yuan Y, Cleary RE, Morrison RP:
Serologic responses of infertile women to
the 60-kd chlamydial heat shock protein
(hsp60). Fertil Steril 1995; 64:730.
139. Kimani J, Maclean IW, Bwayo JJ, et al:
Risk factors for Chlamydia trachomatis
pelvic inflammatory disease among sex
workers in Nairobi, Kenya. J Infect Dis
1996; 173:1437.
140. Young RA, Elliot TJ: Stress proteins,
infection, and immune surveillance. Cell
1989; 59:5.
141. Kaufmann SHE: Heat shock proteins and
the immune response. Immunol Today
1990; 11:129.
142. Zhong G, Brunham RC: Antibody
responses to the chlamydial heat shock
proteins hsp 60 and hsp 70 are H-2 linked.
143. Taylor HR, Prendergast RA, Dawson CR,
et al: An animal model for cicatrizing
trachoma. Invest Ophthalmol Vis Sci 1981;
21:422.
144. Wang S-P, Grayston JT: Trachoma in the
Taiwan monkey, Macaca cyclopis. Ann N Y
Acad Sci 1962; 98:177.
145. Wang S-P, Grayston JT: Pannus with
experimental trachoma and inclusion
146.
147.
148.
149.
150.
151.
152.
153.
154.
155.
156.
157.
158.
159.
conjunctivitis agent infection of Taiwan

monkeys. Am J Ophthalmol 1967; 63:1133.
Wang S-P, Grayston JT, Alexander ER:
Trachoma vaccine studies in monkeys. Am
J Ophthalmol 1967; 63:1615.
Monnickendam MA, Darougar S: An animal
model for hyperendemic trachoma: a study
of immunity and hypersensitivity to
Chlamydia. In: Silverstein AM, OConnor
GR, eds. Immunology and
immunopathology of the eye. New York:
Masson; 1979:375380.
Whittum-Hudson JA, Taylor HR, Farazdaghi
M, et al: Immunohistochemical study of the
local inflammatory response to chlamydial
ocular infection. Invest Ophthalmol Vis Sci
1986; 27:64.
Whittum-Hudson JA, Taylor HR:
Antichlamydial specicity of conjunctival
lymphocytes during experimental ocular
Taylor HR, Johnson SL, Prendergast RA,
et al: An animal model of trachoma. II. The
importance of repeated infection. Invest
Taylor HR, Prendergast RA, Dawson CR,
et al: Animal model of trachoma. III. The
necessity of repeated exposure to live
chlamydia. In: Mardh PA, Holmes KK, Oriel
JD, et al, eds. Chlamydial infections.
Symposium on Human Chlamydial
Infections. Amsterdam: Elsevier
Taylor HR, Kolarczyk RA, Johnson SL, et al:
Effect of bacterial secondary infection in an
animal model of trachoma. Infect Immun
1984; 44:614.
Taylor HR, Johnson SL, Schachter J, et al:
Pathogenesis of trachoma: the stimulus for
inflammation. J Immunol 1987; 138:3023.
Taylor HR, Schachter J, Caldwell HD: The
stimulus for conjunctival inflammation in
trachoma. In: Oriel D, Ridgway G,
Schachter J, et al, eds. Chlamydial
infections. Proceedings of the 6th
International Symposium on Human
Cambridge University Press; 1986:167170.
Wang S-P, Grayston JT: Immunologic
relationship between genital TRIC,
lymphogranuloma venereum, and related
organisms in a new microtiter indirect
immunofluorescence test. Am J
Watkins NG, Caldwell HD: Delayed
hypersensitivity as a pathogenic
mechanism in chlamydial disease. In: Oriel
D, Ridgway G, Schachter J, et al, eds.
1986:408411.
Watkins NG, Hadlow WJ, Moos AB, et al:
Ocular delayed hypersensitivity: a
pathogenetic mechanism of chlamydial
conjunctivitis in guinea pigs. Proc Natl
Acad Sci USA 1986; 83:7480.
Morrison RP, Lyng K, Caldwell HD:
Chlamydial disease pathogenesis. Ocular
hypersensitivity elicited by a genus-specic
57-kD protein. J Exp Med 1989; 169:663.
Morrison RP, Belland RJ, Lyng K, et al:
Chlamydial disease pathogenesis. The
57-kD chlamydial hypersensitivity antigen is
a stress response protein. J Exp Med 1989;
170:1271.
160. Byrne GI, Grubbs B, Marshall TJ, et al:

Famma-interferon-mediated cytotoxicity
related to murine Chlamydia trachomatis
161. Williams DM, Bonewald LF, Roodman GD,
et al: Tumor necrosis factor-alpha is a
cytotoxin induced by murine Chlamydia
trachomatis infection. Infect Immun 1989;
57:1351.
162. Dayer J-M, Beutler B, Cerami A: Cachectin/
tumor necrosis factor stimulates collagenase
and prostaglandin E2 production by human
synovial cells and dermal broblasts. J Exp
Med 1985; 162:2163.
163. Elias JA, Krol RC, Freundlich B, et al:
Regulation of human lung broblast
glycosaminoglycan production by
recombinant interferons, tumor necrosis
factor, and lymphotoxin. J Clin Invest 1988;
81:325.
164. Peeling RW, Brunham RC: Chlamydiae as
pathogens: new species and new issues.
Emerg Infect Dis 1996; 2:307.
165. Rapoza PA, Quinn TC, Kiessling LA, et al:
Epidemiology of neonatal conjunctivitis.
166. Fisher MC: Conjunctivitis in children.
Pediatr Clin North Am 1987; 34:1447.
167. Alexander ER, Harrison HR: Role of
Chlamydia trachomatis in perinatal
infection. Rev Infect Dis 1983; 5:713.
168. Heggie AD, Lumicao GG, Stuart LA, et al:
Chlamydia trachomatis infection in mothers
and infants. Am J Dis Child 1981; 135:507.
169. Beem MO, Saxon EM: Respiratory-tract
colonization and a distinctive pneumonia
syndrome in infants infected with
Chlamydia trachomatis. N Engl J Med
1977; 296:306.
170. Schachter J, Dawson CR: Is trachoma an
ocular component of a more generalized
chlamydial infection? Lancet 1979; 1:702.
171. Schachter J, Grossman M, Sweet RL, et al:
Prospective study of perinatal transmission
of Chlamydia trachomatis. JAMA 1986;
225:3374.
172. Forster RK, Dawson CR, Schachter J: Late
follow-up of patients with neonatal
conjunctivitis. Am J Ophthalmol 1970;
69:467.
173. Goscienski PJ, Sexton RR: Follow-up
studies in neonatal inclusion conjunctivitis.
Am J Dis Child 1972; 124:180.
174. Mordhorst CH: Clinical epidemiology of
oculogenital chlamydia infection. In:
Hobson D, Holmes KK, eds.
Nongonococcal urethritis and related
infections. Washington, DC: American
Society for Microbiology; 1977:126134.
175. Ronnerstam R, Personn K: Chlamydial
conjunctivitis in a Swedish population. In:
Mardh PA, Holmes KK, Oriel JD, et al, eds.
Chlamydia Infections. Proceedings of the
176. Wishart PK, James C, Wishart MS, et al:
Prevalence of acute conjunctivitis caused
by chlamydia, adenovirus, and herpes
simplex virus in an ophthalmic casualty
department. Br J Ophthalmol 1984; 68:653.
177. Bialasiewicz AA, Jahn GJ: Epidemiology of
chlamydial eye diseases in a mixed
rural/urban population of West Germany.
178. Potts MJ, Paul ID, Roome APCH, et al:
Rapid diagnosis of Chlamydia trachomatis
infection in patients attending an
179.
180.
181.
182.
183.
184.
185.
186.
187.
188.
189.
190.
191.
192.
193.
194.
195.
196.
ophthalmic casualty department. Br J

Fitch CP, Rapoza PA, Owens S, et al:
Epidemiology and diagnosis of acute
conjunctivitis at an inner-city hospital.
Heggie AD: Incidence and etiology of
conjunctivitis in Navy recruits. Milit Med
1990; 155:1.
Rapoza PA, Quinn TC, Terry AC, et al:
A systematic approach to the diagnosis
and treatment of chronic conjunctivitis.
Smith TF, Wentworth BB: Chlamydial
infections. In: Wentworth BB, Judson FN,
eds. Laboratory manual for the diagnosis of
sexually transmitted diseases. Washington,
DC: American Public Health Association;
1984:81104.
Tang FF, Chang HL, Huang YT, et al:
Studies on the etiology of trachoma with
special reference to isolation of the virus in
chicken embryo. Chin Med J 1957; 75:429.
Gordon FB, Harper IA, Quan AL, et al:
Detection of Chlamydia (Bedsonia) in
certain infections of man. I. Laboratory
procedures: comparison of yolk sac and
cell culture for detection and isolation.
J Infect Dis 1969; 120:451.
Gordon FB, Quan AL: Isolation of the
trachoma agent in cell culture. Proc Soc
Exp Biol Med 1965; 118:354.
Kuo C-C, Wang S-P, Wentworth BB, et al:
Primary isolation of TRIC organisms in
HeLa 229 cells treated with DEAE-dextran.
J Infect Dis 1972; 125:665.
Ripa KT, Mardh P-A: Cultivation of
Chlamydia trachomatis in cycloheximidetreated McCoy cells. J Clin Microbiol 1977;
6:328.
Sompolinsky D, Richmond S: Growth of
Chlamydia trachomatis in McCoy cells
treated with cytochalasin B. Appl Microbiol
1974; 28:912.
Quinn TC, Gupta PK, Burkman RT, et al:
Detection of Chlamydia trachomatis
cervical infection: a comparison of
Papanicolaou and immunofluorescent
staining with cell culture. Am J Obstet
Gynecol 1987; 157:394.
Darougar S, Cubitt S, Jones BR: Effect of
high-speed centrifugation on the sensitivity
of irradiated McCoy cell culture for the
isolation of Chlamydia. Br J Vener Dis
1974; 50:308.
Reeve P, Owen J, Oriel JD: Laboratory
procedures for the isolation of Chlamydia
trachomatis from the human genital tract.
J Clin Pathol 1975; 28:910.
Allan I, Pearce JH: Modulation by
centrifugation of cell susceptibility to
chlamydial infection. J Gen Microbiol 1979;
111:87.
Kuo C-C, Grayston JT: Interaction of
Chlamydia trachomatis organisms and
HeLa 229 cells. Infect Immun 1976;
13:1103.
Rota TR, Nichols RL: Infection of cell culture
by trachoma agent. Enhancement by DEAEdextran. J Infect Dis 1971; 124:419.
Stevens RS, Kuo C-C, Tam MR: Sensitivity
of immunofluorescence with monoclonal
antibodies for detection of Chlamydia
trachomatis inclusions in cell culture. J Clin
Microbiol 1982; 16:4.
Stamm WE, Tam M, Koester M, et al:
inclusions in McCoy cell cultures with
197.
198.
199.
200.
201.
202.
203.
204.
205.
206.
207.
208.
209.
210.
211.
212.
213.
fluorescein-conjugated monoclonal
antibodies. J Clin Microbiol 1983; 17:666.
Grayston JT, Kuo C-C, Wang S-P, et al: A
new Chlamydia psitacci strain, TWAR,
isolated in acute respiratory tract
infections. N Engl J Med 1986; 315:161.
Cles LD, Stamm WE: Use of HL cells for
improved isolation and passage of
Chlamydia pneumoniae. J Clin Microbiol
1990; 28:938.
Kuo C-C, Grayston JT: A sensitive cell line,
HL cells, for isolation and propagation of
Chlamydia pneumoniae. J Infect Dis 1990;
162:755.
Theunissen JJH, van Heijst BYM,
Wagenvoort JHT, et al: Factors influencing
the infectivity of Chlamydia pneumoniae
elementary bodies on HL cells. J Clin
Microbiol 1992; 30:1388.
Wong KH, Skelton SK, Chan YK: Efcient
culture of Chlamydia pneumoniae with cell
lines derived from the human respiratory
tract. J Clin Microbiol 1992; 30:1625.
Grayston JT: Infections caused by
Chlamydia pneumoniae strain TWAR. Clin
Infect Dis 1992; 15:757.
Halberstaedter L, von Prowazek S: Uber
Zelleinschlusse parasitarer Natur beim
Trachom. Arb Kaiserlichen
Gesundheitsamte 1907; 26:44.
World Health Organization: Guide to the
laboratory diagnosis of trachoma. Geneva:
World Health Organization; 1975.
Schachter J, Dawson CR: Comparative
efcacy of various diagnostic methods for
chlamydial infection. In: Hobson D, Homes
KK, eds. Nongonococcal urethritis and
related infections. Washington, DC:
1977:337341.
Darougar S, Woodland RM, Jones BR,
et al: Comparative sensitivity of fluorescent
antibody staining of the conjunctival
scrapings and irradiated McCoy cell culture
for the diagnosis of hyperendemic
trachoma. Br J Ophthalmol 1980; 64:276.
Sandstrom KI, Bell TA, Chandler JW, et al:
Microbial causes of neonatal conjunctivitis.
J Pediatr 1984; 105:706.
Smith TF, Wentworth BB: Chlamydial
infections. In: Wentworth BB, Judson FN,
eds. Laboratory manual for the diagnosis of
sexually transmitted diseases. Washington,
1984:81104.
Dorman SA, Danos LM, Wilson DJ, et al:
Detection of chlamydial cervicitis by
Papanicolaou stained smears and culture.
Am J Clin Pathol 1983; 79:421.
Quinn TC, Gupta PK, Burkman RT, et al:
cervical infection: a comparison of
Papanicolaou and immunofluorescent
staining with cell culture. Am J Obstet
Gynecol 1987; 157:394.
Tam RT, Stamm WE, Handseld HH, et al:
Culture-independent diagnosis of
Chlamydia trachomatis using monoclonal
antibodies. N Engl J Med 1984;
310:1146.
Stamm WE, Harrison HR, Alexander ER,
et al: Diagnosis of Chlamydia trachomatis
infections by direct immunofluorescence
staining of genital secretions A multicenter
trial. Ann Intern Med 1984; 101:638.
Bell TA, Kuo C-C, Stamm WE, et al: Direct
fluorescent monoclonal antibody stain for
rapid detection of infant Chlamydia
214.
215.
216.
217.
218.
219.
220.
221.
222.
223.
224.
225.
226.
227.
228.
trachomatis infections. Pediatrics 1984;

74:224.
Taylor HR, Rapoza PA, Kiessling A, et al:
Rapid detection of Chlamydia trachomatis
with monoclonal antibodies. Lancet 1984;
2:38.
Mabey DCW, Booth-Mason S: The
detection of Chlamydia trachomatis by
direct immunofluorescence in conjunctival
smears from patients with trachoma and
patients with ophthalmia neonatorum using
a conjugated monoclonal antibody. J Hyg
1986; 96:83.
Rapoza PA, Quinn TC, Kiessling LA, et al:
Assessment of neonatal conjunctivitis with
a direct fluorescent monoclonal antibody
stain for Chlamydia. JAMA 1986; 255:3369.
Rompalo AM, Suchland RJ, Price CB, et al:
Rapid diagnosis of Chlamydia trachomatis
rectal infection by direct fluorescence
staining. J Infect Dis 1987; 155:1075.
Barnes RC: Laboratory diagnosis of human
chlamydial infections. Clin Microbiol Rev
1989; 2:119.
Baselski VS, McNeeley SG, Ryan G, et al:
A comparison of nonculture-dependent
methods for detection of Chlamydia
trachomatis infections in pregnant women.
Obstet Gynecol 1987; 70:47.
Mabey DCW, Robertson JN, Ward ME:
Detection of Chlamydia trachomatis by
enzyme immunoassay in patients with
Hipp SS, Yangsook H, Murphy D:
Assessment of enzyme immunoassay and
immunofluorescence test for detection of
Chlamydia trachomatis. J Clin Microbiol
1987; 25:1938.
Chernesky M, Jang D, Dopes D, et al:
Comparison of a polymer conjugateenhanced enzyme immunoassay to a ligase
chain reaction for diagnosis of Chlamydia
trachomatis in endocervical swabs. J Clin
Microbiol 2001; 39:23062307.
Tanaka M, Nakayama H, Yoshida H, et al:
Detection of Chlamydia trachomatis in
vaginal specimens from female commercial
sex workers using a new improved enzyme
immunoassay. Sex Transm Infect 1998;
74:435438.
Tanaka M, Nakayama H, Sagiyama K, et al:
Evaluation of a new amplied enzyme
immunoassay (EIA) for the detection of
Chlamydia trachomatis in male urine,
female endocervical swab, and patient
obtained vaginal swab specimens. J Clin
Pathol 53:350354.
Moncada J, Schachter J, Bolan G, et al:
Conrmatory assay increases specicity of
the Chlamydiazyme test for Chlamydia
trachomatis in urogenitial specimens. Diagn
Microbiol Infect Dis 1990; 15:663668.
Hammerschlag MR, Roblin PM, Cummings
C, et al: Comparison of enzyme
immunoassay and culture for diagnosis of
chlamydial conjunctivitis and respiratory
infections in infants. J Clin Microbiol 1987;
25:2306.
Clarke LM, Sierra MF, Daidone BJ, et al:
Comparison of the Syva MicroTrak enzyme
immunoassay and Gen-Probe PACE 2 with
cell culture for diagnosis of cervical
Chlamydia trachomatis infection in a highprevalence female population. J Clin
Microbiol 1993; 31:968971.
Lauderdale TL, Landers L, Thorneycroft I,
Chapin K: Comparison of the PACE 2
assay, two amplication assays, and
CHAPTER 12
Chlamydial Disease
129
SECTION 3
MICROBIOLOGY
130
Clearview enzyme immunoassay for

detection of Chlamydia trachomatis in
female endocervical and urine specimens.
J Clin Microbiol 1999; 37:22232229.
229. Berg E, Anestad G, Moi H, et al: Falsenegative results of a ligase chain reaction
assay to detect Chlamydia trachomatis due
to inhibitors in urine. Eur J Clin Microbiol
Infect Dis 1997; 16:727731.
230. Wang S-P, Grayston JT: Microimmunofluorescence antibody responses in
Chlamydia trachomatis infection: a review.
In: Mardh P-A, Holmes KK, Oriel JD, et al,
eds. Chlamydial infections. Proceedings of
the 5th International Symposium on Human
231. Wang S-P, Grayston JT:
Microimmunofluorescence serological
studies with the TWAR organism. In: Oriel
JD, Ridgway G, Schachter J, et al, eds.
232.
233.
234.
235.
236.

1986:329332.
Campbell LA, Kuo C-C, Wang S-P, et al:
Serological response to Chlamydia
pneumoniae infection. J Clin Microbiol
1990; 28:1261.
Schachter J, Grossman M, Azimi PH:
Serology of Chlamydia trachomatis in
infants. J Infect Dis 1982; 146:530.
Darougar S, Treharne JD, Minassian D,
et al: Rapid serologic test for diagnosis of
chlamydial ocular infections. Br J
Treharne JD, Dwyer RS, Darougar S, et al:
Antichlamydial antibody in tears and sera.
Bas S, Muzzin P, Ninet B, et al: Chlamydial
serology: comparative diagnostic value of
237.
238.
239.
240.
immunoblotting, microimmunofluorescence
tests, and immunoassays using different
recombinant proteins as antigens. J Clin
Microbiol 2001; 39:13691377.
Persson K, Boman J: Comparison of ve
serologic tests for diagnosis of acute
infections by Chlamydia pneumoniae. Clin
Diagn Lab Immunol 2000; 7:739744.
Schumacher A, Lerkerod AB, Seljeflot I,
et al: Chlamydia pneumoniae serology:
importance of methodology in patients with
coronary heart disease and healthy
individuals. J Clin Microbiol 2001;
39:18591864.
Batteiger BE, Jones RB: Chlamydial
infections. Infect Dis Clin North Am 1987;
1:55.
Centers for Disease Control and
Prevention: Sexually transmitted diseases
treatment guidelines, 2006. MMWR 2006;
55(No. RR-11): 2142.
CHAPTER
13
The Spirochetes
Marlene L. Durand
INTRODUCTION
Spirochetes are mobile, corkscrew-shaped bacteria that represent
a phylogenetically ancient bacterial group.1 They are 10 times
longer and thinner than most pathogenic bacteria (Table 13.1),
and nearly all are invisible on Gram stain. With rare exception,
they cannot be cultured in clinical microbiology laboratories. As
a consequence, the diagnosis of most spirochetal diseases relies
on serologic tests or microscopy using special techniques (e.g.,
silver stain, dark-eld microscopy).
Some spirochetes, such as nonpathogenic treponemes, are
members of the normal human oral or gastrointestinal flora,
while others are pathogenic. Pathogenic spirochetes include
Treponema, Borrelia, Leptospira, and Spirillum (Table 13.2).
Those that can cause disease of the central nervous system
(CNS) can also cause ocular disease, as would be expected. Systemic spirochetal infections that may involve the eye include
syphilis, Lyme disease, relapsing fever, and leptospirosis.
TREPONEMES
NONPATHOGENIC TREPONEMES
The treponemes include both nonpathogenic and pathogenic
species. Nonpathogenic oral treponemes, such as Treponema
denticola, Treponema vincentii, and Treponema medium, are
normal colonizers of the mouth. They play important roles in
subgingival plaque and periodontal disease.2 Oral treponemes
differ from the pathogenic treponemes in many ways. Most oral
treponemes may be readily cultured anaerobically if selected
media are used,3 while pathogenic treponemes such as the
syphilis treponeme, Treponema pallidum, cannot be propogated.
The genome of T. denticola was recently sequenced and was found
to be much larger than that of T. pallidum, with little DNA
sequence homology.4 The pathogenic treponemes include the
nonsexually transmitted endemic treponemes as well as the
agent of syphilis.
SYPHILIS
urban population were infected, and 25% of patients progressed

to a chronic illness.5
Epidemiology
Syphilis is found worldwide, and there are more than 12 million
cases. In the US, the incidence has declined dramatically since the
advent of penicillin in the 1940s. It is primarily a sexually transmitted disease, although it can be acquired transplacentally
(congenital syphilis), by kissing or other close contact with an
active skin or mucous membrane lesion, and by blood transfusion. Transfusion-related transmission is now very rare both
because blood donors with positive serologic tests are excluded,
and because the organism cannot survive more than 2448 h
under conditions of blood bank storage.
Microbiology and Pathogenesis

Syphilis is caused by T. pallidum subspecies pallidum. The spirochete has outer and cytoplasmic membranes, a thin peptidoglycan
layer, and flagella that lie in the periplasmic space. It contains a
circular chromosome of ~1000 kbp, making it one of the smallest
bacterial genomes. The mechanism of T. pallidum pathogenesis is
poorly understood, and no known virulence factors have been
identied. The outer membrane is mostly lipid with few surface
proteins. This has led to the hypothesis that this is a stealth
organism that minimizes the number of surface membranebound targets in order to evade the host immune system.6
The number of organisms required to establish infection varies
between patients, but an inoculum of only four bacteria can
establish infection in rabbits. The dividing time is 30 h, and clinical lesions appear when there are 107 organisms per milligram
of tissue.7 A larger inoculum will therefore lead to a clinically
apparent lesion sooner than a small inoculum.
Organisms gain entrance to the body through mucous membranes or abraded skin, and a lesion appears at the site on inoculation an average of 3 weeks later. However, T. pallidum has
already spread throughout the body by this time, since there is
a spirochetemia within hours to days of the initial inoculation.
Any organ may be infected, although the CNS is especially
targeted. Evidence of organ infection may not become clinically
apparent until decades later, however.
History
Syphilis is a disease of great historical signicance. It was rst
reported in Europe in the late fteenth century and coincided with
the return of Columbuss ships from the New World. Syphilis
quickly reached epidemic proportions in Europe, and spread across
the world with the explorations of the sixteenth century. By the
turn of the twentieth century, syphilis was the leading cause of
neurologic and cardiovascular disease among middle-aged people.5
With the advent of the Wasserman test in 1906, the prevalence
of the disease was appreciated; between 8% and 14% of the
Stages
Syphilis has long been divided into stages (Table 13.3), and
clinical manifestations, serologic results, and treatment depend
on the stage of disease. Although untreated syphilis is a life-long
infection, it is only contagious during the early stages (up to
4 years after initial infection).
Primary syphilis includes the development of a chancre at
the inoculation site, usually the external genitalia. A chancre is
a painless, ulcerated lesion with a smooth base. There is no
131
MICROBIOLOGY
TABLE 13.1. Characteristics of Spirochetes in Comparison with Common Pathogenic Bacteria.

Organism
Size ( m)
Cultivable?
Usual Diagnostic Method
Treponemes (pathogenic)
0.15 515
No
Serology, microscopy
Borrelia
0.2 2030
Difcult
Serology
Spirochetes:
Lyme relapsing fever
0.2 830
Difcult
Microscopy
Leptospira
0.1 620
Yes
Culture, serology
Spirillum
0.2 35
No
Microscopy
Staphylococcus
1 (sphere)
Yes
Culture
Pseudomonas
0.5 2
Yes
Culture
SECTION 3
Common Bacteria:
TABLE 13.2. Overview of Spirochetes

Organism
Disease
Transmission
Locale
Eye Disease
Syphilis
Sexual contact, congenital,

transfusion
Worldwide
Yes
Treponemes
T. pallidum*
T. pertenue*
Yaws
Direct contact
Tropical, worldwide
No
T. endemicum*
Bejel
Direct contact, fomite
Arid, North Africa,

Arabian peninsula
No
T. carateum
Pinta
Direct contact
Amazon
No
B. burgdorferi
Lyme
Tick
Europe, North America
Yes
Borrelia species
Relapsing fever
Tick louse
Worldwide
Central/East Africa,
Andes
No
L. interrogans
Leptospirosis
Zoonosis
Worldwide India,
Hawaii
Yes
Spirillum minus
Rat-bite fever
Rat bite
Asia
No
Borrelia
Leptospira
* Syphilis, yaws, and bejel are all caused by the same genus and species, T. pallidum, but by different subspecies. Therefore the correct names for these spirochetes are
T. pallidum subsp pallidum, T. pallidum subsp pertenue, T. pallidum subsp endemicum. Treponema carateum is a separate species, rather than a subspecies of
T. pallidum.
Yaws, bejel, and pinta are endemic treponematoses that are transmitted by direct contact with skin lesions, rather than by sexual contact. In bejel, transmission may
also be by mucous membrane contact or fomites (sharing drinking cups).
132
exudate, and the chancre does not bleed when scraped. In some
cases no chancre develops, and in others only a small papule
occurs. Multiple chancres may occur, especially in HIV-infected
patients. The chancre heals spontaneously in 36 weeks. Serologic
tests may be negative, since these tests cannot detect antibodies
until 13 weeks after the development of the chancre.8 Diagnosis
is usually made by nding the treponemes in chancre scrapings
using either dark-eld microscopy or immunostaining with
fluorescent antibodies (DFA-TP).
Secondary syphilis begins 28 weeks after the chancre appears
and is the phase most associated with constitutional symptoms.
A rash develops in the majority of patients and usually involves
the palms and soles. Painless moist plaques called condoloma
lata may develop in intertriginous areas; these are highly contagious. Constitutional symptoms such as fever, sore throat,
arthralgias, and malaise develop in 70% of patients. The CNS is
involved in 40% of patients, although fewer are symptomatic.
This is called acute neurosyphilis to distinguish it from tertiary

neurosyphilis. An aseptic meningitis is seen in 12% of
patients. Ocular involvement, usually uveitis, may occur. The
RPR is reactive, usually at high titer, in virtually all patients
with secondary syphilis. The symptoms of secondary syphilis
may resolve and then relapse; relapses are usually milder.
Latent syphilis is, by denition, that stage when the patient is
asymptomatic and there are no signs of the disease (other than
positive serology). This stage is divided into early latent and late
latent. Early latent usually comprises the rst 4 years of infection, during which a relapse may occur and the patient may still
be contagious.7 However, a recent publication by the Centers for
Disease Control and Prevention (CDC) considers early latent
syphilis as infection acquired within the preceding 1 year.9 If the
date of onset of syphilis cannot be determined, as is usually the
case, patients are treated as late latent syphilis. Late latent
syphilis may last decades. Although the specic treponemal
The Spirochetes
TABLE 13.3. The Stages of Syphilis and Their Treatment

Stage
Symptoms/Signs*
RPR
FTA-abs
Treatment
Primary
Painless chancre
+ or
+ or
IM benzathine PCN||
2.4 MU 1 dose
Secondary
Rash, flu-like symptoms;

may have aseptic meningitis
IM benzathine PCN 1;
IV PCN if neurosyphilis
or ocular syphilis
Early latent
None
IM benzathine PCN 1
Late latent
None
+ or
IM benzathine PCN
weekly 3 weeks#
Tertiary
Cardiovascular, neurologic,
ocular, otosyphilis
+ or
* Symptoms and signs listed are those typical for the stage; exceptions except.
RPR or VDRL = nontreponemal tests.
FTA-abs or TPPA = specic treponemal tests.
For details, including treatment in special hosts (e.g., pregnant patients, penicillin-allergic patients, children, etc.) see Workowski KA, Berman SM for the Centers for
Disease Control and Prevention. Sexually transmitted treatment guidelines, 2006. Morbid Mortal Weekly Report 2006;55 (RR 11):194.
IM = intramuscular; IV = intravenous; PCN = penicillin; MU = million units.
#
HIV-infected patients who have late latent syphilis, or latent syphilis of unknown duration, should have a lumbar puncture to determine if asymptomatic neurosyphilis is
also present. If the cerebrospinal fluid is abnormal, they should be treated for neurosyphilis with IV penicillin.
tests (e.g., FTA-abs, TPPA) are positive during this stage, the
nonspecic tests (e.g., RPR, VDRL) may wane with time, so that
many patients with late latent syphilis have a nonreactive RPR.
Tertiary syphilis, also called late syphilis, is primarily manifested by cardiovascular or CNS symptoms. In the preantibiotic
era, up to 25% of patients progressed to tertiary syphilis. Tertiary
syphilis is seen even in the antibiotic era, and often represents
unrecognized infection acquired decades earlier. It also may
represent failure of benzathine penicillin therapy given for the
early stages of syphilis. Benzathine penicillin, the standard
treatment for primary, secondary, and latent syphilis, does not
cross the bloodbrain barrier. As a consequence, a patient may
develop late neurosyphilis despite having been treated for
syphilis years earlier. Such failures are known to occur in one
patient per 3331000 treated patients.7
Cardiovascular syphilis will occur in 10% of untreated patients
with syphilis. It is mainly an aortitis, and the classic nding is
a fusiform aortic aneurysm of the ascending aorta. Concurrent
late neurosyphilis is common.
Late neurosyphilis, as distinguished from the acute neurosyphilis that may be seen during secondary syphilis, is a chronic
meningitis involving all parts of the CNS. Asymptomatic neurosyphilis is the most common form of late neurosyphilis and is
diagnosed by an abnormal cerebrospinal fluid (CSF). The CSF
VDRL is positive in only half of the cases of neurosyphilis, so
other abnormalities (e.g., pleocytosis, elevated CSF protein) are
signicant. Symptomatic late neurosyphilis includes ndings of
meningovascular or parenchymatous involvement. There may
be personality changes, memory loss, slurred speech, and psychiatric manifestations such as megalomania. The patient may
be misdiagnosed with Alzheimers disease. There may be
demyelination of the posterior columns of the spinal cord, leading
to an ataxic gait, loss of bladder or bowel function, shooting
pains, and peripheral neuropathy.
Ocular syphilis or otosyphilis may occur as part of tertiary
syphilis and are often considered subsets of neurosyphilis. This
may lead to confusion, since ocular or otosyphilis may occur
without involvement of the brain or meninges. A normal CSF
formula does not exclude ocular or otosyphilis.
Nonspecic tests for syphilis (RPR or VDRL) may be negative
in up to 50% of patients with tertiary syphilis, because these
reactions wane with time. Specic treponemal tests (FTA-abs,
TPPA) usually remain positive for life, however.
CHAPTER 13
IV PCN 1014 days

(usual dose 4 MU q4h)
Ocular Syphilis
Ocular syphilis may occur either during secondary or tertiary
syphilis. The ndings of ocular syphilis are protean, and are
discussed in detail in other chapters (see Chapters 345 and 351).
General recommendations for serologic diagnosis and treatment
in ocular syphilis are listed in Table 13.4. The details of treatment
in various groups (HIV, penicillin-allergic, children, pregnant
patients, etc.) are given by the CDC in their 2006 guideline.9
All patients with ocular syphilis should be screened for asymptomatic neurosyphilis. If CSF abnormalities exist, treatment
with IV penicillin is the same, but a follow-up lumbar puncture
is required at 6 months to determine adequacy of therapy. If the
CSF is still abnormal at that point, the patient should be
retreated. All patients with ocular syphilis should be screened
for HIV, as there is a higher incidence of ocular syphilis in HIVinfected patients than in the non-HIV-infected patients.10 A
recent study of 320 HIV-positive patients receiving highly active
antiretroviral therapy at a Washington, DC, infectious disease
clinic and screened for syphilis found that 7.5% had syphilis,
TABLE 13.4. Ocular Syphilis: Recommendations for Serologic
Diagnosis and Treatment
1. Screen with both RPR and FTA-abs. A nonreactive RPR does
not exclude ocular syphilis.
2. Conrm a reactive FTA-abs with a TPPA (to exclude falsepositive FTA-abs).
3. A patient who has eye ndings consistent with ocular syphilis as
well as a reactive TPPA should be treated for presumed ocular
syphilis. A history of prior treatment for syphilis with IM benzathine
penicillin does not exclude this diagnosis.
4. Test for HIV, as there is a higher incidence of ocular syphilis in HIV.
5. Perform a lumbar puncture (LP) to exclude concomitant
neurosyphilis. A normal CSF does not exclude ocular syphilis, but
an abnormal CSF will require a follow-up LP 6 months after
treatment to ensure adequacy of therapy for neurosyphilis.
6. Treat ocular syphilis the same as for neurosyphilis, with IV
penicillin 4 million units every 4 hours for 1014 days in adults with
normal renal function. Patients with penicillin allergy may require
desensitization with the help of an allergist. At the end of IV
therapy, some experts also prescribe IM benzathine penicillin
2.4 million units once weekly for 3 weeks.
133
MICROBIOLOGY
and 13% of these patients with syphilis had ocular syphilis.11
All patients in this study with ocular syphilis also had an
abnormal CSF, consistent with coexisting neurosyphilis.
Syphilis in HIV-Infected Patients

In general, syphilis in patients with HIV is more severe and
protracted. These patients are especially likely to develop neurosyphilis and ocular syphilis, and relapses with these manifestations despite standard benzathine penicillin are well described.
Therefore, a more vigorous or protracted treatment regimen is
recommended for HIV-coinfected patients with syphilis. The
CDC recommends that HIV-positive patients with late latent
syphilis or syphilis of unknown duration have a lumbar puncture.9
Patients with abnormal CSF should be treated for neurosyphilis.
SECTION 3
Syphilis Serology
Syphilis is diagnosed primarily by serologic tests. Nonspecic tests
for syphilis include rapid plasma regain (RPR) and Venereal
Disease Research Laboratory (VDRL). These tests vary with the
stage of disease and response to treatment. A VDRL or RPR
should become nonreactive within 1 year of treatment for primary syphilis and 2 years for secondary syphilis. The RPR or
VDRL may be negative in primary syphilis, but ~100% of
patients have a reactive test in secondary syphilis, usually at
high titer. The highest titers occur during untreated secondary
and early latent syphilis and decline thereafter, usually to less
than 1:4. Between 25% and 50% of patients with late latent or
neurosyphilis have negative RPR or VDRL test results. All positive RPR or VDRL results should be conrmed by a specic treponemal test, as false-positive results occur. Specic treponemal
tests measure antibodies against specic treponemal antigens.
The most commonly used tests are FTA-abs (fluorescent treponemal antibody absorbed) and TPPA (T. pallidum particle agglutination). The FTA-abs is an older test but has occasional
false-positive results, so the TPPA is preferred but may not be
as readily available. The specic treponemal tests usually become
positive during early syphilis and usually remain positive for
life, even after successful treatment.
False-positive tests for RPR or VDRL are more common than for
FTA-abs, but occur in both. Other spirochetal diseases can cause
false-positive results. The endemic treponematoses cause identical
serologic results as syphilis. Lyme disease is a well-known cause
of a false-positive FTA-abs, although the RPR is usually negative.
Rheumatologic conditions frequently cause false-positive RPR or
VDRL reactions, and may also produce false-positive FTA-abs.
A second specic test, such as the TPPA or syphilis Western
blotting method, should be used to conrm a positive FTA-abs,
especially in patients with rheumatologic diseases. A study
using the Western blot as the gold standard in 107 patients with
rheumatologic disease found that the FTA-abs had a specicity
for syphilis of only 68%, with 32 false-positive results.12
ENDEMIC TREPONEMATOSES
134
Endemic treponematoses include yaws, bejel, and pinta. These

are non-sexually transmitted infections with skin lesions as their
early manifestation. Yaws and bejel are caused by treponemes
that are very closely related to syphilis: T. pallidum subspecies
pertenue and endemicum, respectively. Pinta is caused by a
separate species, Treponema carateum. The spirochetes of
endemic treponematoses are indistinguishable morphologically
and serologically from syphilis.
Prior to mass treatment programs of 3050 years ago, endemic
treponematoses were prevalent especially in impoverished and
rural communities. Yaws was a worldwide disease of the tropics
and subtropics, including the Caribbean islands such as Haiti.
Bejel was seen in arid regions in North Africa, the Middle East, and
the Arabian peninsula. Pinta was found only in the Caribbean

and South America. In 1950, there were an estimated 50 million
cases of yaws worldwide, and from 1952 through 1969, procaine
penicillin G was administered in mass treatment campaigns
conducted by the World Health Organization (WHO) and the
United Nations Childrens Fund. These campaigns resulted in
a marked decrease in this disease and other endemic treponematoses, although 2.5 million people are still affected. Today,
yaws-endemic foci persist in west and central Africa, Southeast
Asia, on some Pacic Islands such as Papua New Guinea, and
Central America. Foci of bejel exist in the Middle East and the
Sahel region of Africa. Pinta is found only in some Indian tribes
in the Amazon region.13
Yaws and bejel are seen mainly in children under age 15, while
pinta may affect young adults. Transmission in all three is by
direct contact with infected skin or mucous membrane lesions.
In yaws, skin lesions begin as a papule, usually on the legs, and
slowly enlarge into a raspberry-like mass. Lesions spontaneously
regress, followed by the appearance of secondary skin lesions.
Secondary lesions also usually resolve, but in 10% of patients,
late disease occurs characterized by destructive bony or cartilaginous lesions. Bejel has similar manifestations, although initial lesions are most often painless patches on oral mucosa. Late
disease also involves chronic destructive lesions involving cartilage or bone. Pinta only involves the skin and does not have late
destructive lesions.
Treatment of the endemic treponematoses is with penicillin.
In mass campaigns, IM penicillin was used, but the need for refrigeration of the medication makes this difcult in many areas.
Recently, a trial using oral penicillin in Guyana was found to be
effective.14
False-Positive Syphilis Tests

Endemic treponematoses are not sexually transmitted yet produce serologic results (RPR, FTA-abs, TPPA) identical to those
of syphilis. For patients who grew up in a yaws-endemic area such
as Haiti prior to the mass treatment programs of the 1950s and
1960s, for example, a positive syphilis serology may reflect this
early exposure to yaws rather than infection with syphilis. However, the patient should always be treated for the possibility of
syphilis given the serious sequelae of untreated disease. Unlike
syphilis, the endemic treponematoses do not involve the CNS at
any stage of disease.13 As a consequence, it seems unlikely that
late yaws, bejel, or pinta would cause ocular disease. Patients in
yaws- or bejel-endemic areas with ndings consistent with ocular
syphilis and positive syphilis serologies most likely have ocular
syphilis. However, some reports have attributed these eye ndings
to late yaws or bejel even though syphilis cannot be excluded.15
BORRELIA
Summary: Treponemes
Nonpathogenic treponemes are part of the normal oral flora

and play a role in dental plaque and periodontal disease.
Syphilis, caused by T. pallidum, silently invades the CNS in
many patients soon after infection.
Clinical signs of early syphilis may be missed by the patient,
so late syphilis may occur decades later in a patient with no
known history of syphilis.
Patients with clinical ndings consistent with late syphilis (e.g.,
ocular syphilis, neurosyphilis, cardiovascular) should be tested
by both TPPA and RPR, as the RPR titer may be negative.
Yaws, bejel, and pinta are primarily childhood skin diseases
that are not sexually transmitted but lead to identical serologic
test results (RPR, FTA-abs) as syphilis.
The spirochete is transmitted by the Ixodes tick, which has a larval,

nymphal, and adult stage. These ticks require a blood meal for
each stage. Nymphs are the size of the head of a pin and are
responsible for most disease transmission to humans. The tick
bite is painless and the tick may go unrecognized. The peak
months of human disease mirror the peak months of nymphal
feeding, May through July. The main foci of US disease are in
the Northeast from Massachusetts to Maryland, Wisconsin and
Minnesota, and northern California. Deer and white-footed
mice are the major mammalian hosts for the tick.
Other manifestations include encephalitis, motor or sensory

radiculoneuritis, mononeuritis multiplex, cerebellar ataxia, and
myelitis. Untreated, these symptoms may last weeks to months.
The most common cardiac manifestation is heart block. This
may be rst degree, Wenckebach, or complete heart block, and
usually resolves in a few days so a permanent pacemaker is not
indicated. Ocular disease other than conjunctivitis is rare, may
occur during stage 2, and may include interstitial keratitis,
iritis, or choroiditis.
Stage 3 represents the chronic stage of disease, and occurs
months after infection. It is characterized by either arthritis or
chronic neurologic abnormalities. Approximately 60% of
untreated patients will develop arthritis, usually involving the knee
or other large joints. Recurrent attacks, separated by periods of
remission, are typical, but eventually these resolve in most
patients. Joint fluid shows a neutrophil-predominant, inflammatory inltrate. Arthritis resolves with antibiotic treatment in
90% of patients, but in 10% a chronic arthritis develops. This
may be a postinfectious immune response, as testing of joint
fluid or synovial tissue for Borrelia DNA is often negative.
Chronic neuroborreliosis may occur years after the primary infection, often following an asymptomatic latency period. In the
US, the characteristic symptom is a subtle cognitive disturbance,
often with a mild memory loss. There are usually no abnormalities
in the CSF, although intrathecal antibody tests may be positive.
Microbiology
Diagnosis
There are three different groups of B. burgdorferi. The strain

found in North America is B. burgdorferi (sensu strictu).
Although all three groups have been found in Europe, Borrelia
garinii and Borrelia afzelii cause most disease there, and these
are the only two groups found in Asia. Clinical manifestations
of Lyme disease vary somewhat in these different regions of the
world and may be due to this strain variability.
The complete genome for B. burdorferi has been sequenced.16
It contains a linear chromosome of 950 kbp plus nine circular
and 12 linear plasmids. The organism uses plasmid-encoded outer
surface proteins (Osp) A through F to adapt to different environments.17 The spirochete expresses OspA in the tick midgut but
OspC when in the mammalian host. Another surface lipoprotein
(VIsE) undergoes signicant antigenic variation during dissemination in the host. The spirochete depends on the host for most
of its nutritional requirements.
The organism may be cultured in special BarbourStoennerKelly
media, though such cultures are not available in most clinical
labs. Organisms are usually cultivable only from patients with
early disease, usually from the initial rash of erythema migrans,
and occasionally from plasma or CSF.
The diagnosis is made primarily by serology. Serologic testing is

performed in two stages, with a screening ELISA (enzyme-linked
immunosorbant assay) followed by a Western blot conrmation
of any positive ELISA results. The screening test has many
false positives, so only those conrmed by Western blot are considered true positives. Serologic tests are often negative during
the rst 12 weeks of primary infection, and IgM antibodies
appear subsequently. IgM antibodies may persist for years, and
are not recommended for diagnosis of chronic infections. Most
patients develop IgG antibodies within 1 month of infection,
and these also may remain positive for years despite treatment.
LYME DISEASE
Lyme disease is caused by Borrelia burgdorferi, an organism that
is the longest and thinnest of the spirochetes. The disease is transmitted to humans by ticks, and it is now the most common vectorborne disease in the US and Europe. The disease was recognized
as a new entity in the US in 1976, when a cluster of children in
Lyme, Connecticut appeared to have juvenile rheumatoid arthritis.
Diseases with similar characteristics had been previously recognized in Europe as Bannworths syndrome, erythema chronicum
migrans, and acrodermatitis chronica atrophicans. The recovery
of the organism from ticks and infected humans established the
link between these diseases.
Epidemiology
CHAPTER 13
The Spirochetes
Treatment
The treatment of early Lyme disease is with oral doxycycline or
oral amoxicillin for 23 weeks. Doxycycline is preferred because
it will also treat other tick-borne diseases (e.g., babesiosis,
ehrlichiosis) that may have been simultaneously introduced by
the tick bite. Patients with arthritis should be treated with these
agents for 12 months, or with IV ceftriaxone 2 g once daily for
24 weeks. Neurologic disease, either during early or late stages
of Lyme disease, is treated with IV ceftriaxone 2 g once daily for
24 weeks; most experts treat late neuroborreliosis for 4 weeks.
Clinical Manifestations
Lyme disease resembles syphilis in that it has three stages.
Stage 1 occurs 3 days to 1 month after the tick bite, and is characterized by a local erythema migrans skin lesion at the site of
the bite. Over half of the patients are unaware of the bite. The
skin lesion is initially homogeneously red, then the center may
become intensely red, indurated vesicular, or necrotic. Often the
circular lesion expands leaving a lighter center, giving a bullseye appearance.
Stage 2 occurs days to weeks after stage 1. Multiple secondary
annular lesions may develop, and they are usually smaller than
the initial lesion. The patient may have flu-like symptoms with
fatigue, headache, fever, myalgias, and lymphadenopathy. After
several weeks, 15% of untreated patients in the US develop
neurologic signs and 5% develop cardiac abnormalities. The
neurologic manifestations most often include an aseptic
meningitis with lymphocytic pleocytosis (~100 cell/mm3) and
an associated facial palsy. The facial palsy may be bilateral.
RELAPSING FEVER
Relapsing fever is an infection characterized by recurrent fevers
and flu-like symptoms interspersed with periods of apparent
health. It is caused by Borrelia species, and there are two types
of disease, tick-borne and louse-borne.
Louse-Borne Relapsing Fever

Louse-borne relapsing fever (LBRF) is caused by B. recurrentis
and usually occurs in epidemics during wartime, famine, or
other upheavals. The last large epidemic occurred during World
War II when 50 000 people died of this disease. The disease still
occurs in northeastern and central Africa, especially Ethiopia,
Sudan, and Somalia. The disease is transmitted by the human
body louse, which ingests the organism during a blood meal from
an infected person, then releases Borrelia to another person when
the louse is crushed. The Borrelia can then penetrate intact skin
135
MICROBIOLOGY
and cause disease after an incubation period of ~1 week. Onset
of symptoms is usually abrupt, and symptoms include high
fever, myalgias, headache, hepatomegaly, splenomegaly, and cough.
Hemoptysis, hematemesis, or hematuria may also present.
Neurologic involvement occurs in 30% cases. After an average
of 5 days, the patient becomes asymptomatic for ~9 days and
then suffers a relapse. Diagnosis is by clinical suspicion and
demonstration of the spirochetes on peripheral blood smears of
febrile patients. Serodiagnosis by detecting antibodies against a
surface protein of the spirochete has recently been proposed.18
Treatment is with tetracycline or penicillin, but this often induces
a dangerous JarischHerxheimer reaction.19 The latter may be
prevented by pretreatment with antibodies against tumor necrosis
factor alpha.20 Untreated, up to 40% of patients may die.
SECTION 3
Tick-Borne Relapsing Fever

Tick-borne relapsing fever (TBRF) is seen sporadically and in occasional outbreaks. It has been reported worldwide. In the US,
most cases have occurred after patients have stayed in mountain
cabins in the Western US. The illness is caused by at least 15
Borrelia species, with B. hermsii most commonly reported. All
species are transmitted by soft ticks of the genus Ornithodoros.
These ticks require blood meals but can survive without a meal
for up to 15 years. Animal reservoirs for the ticks include mice,
rats, squirrels, rabbits, owls, and lizards. The tick cannot travel
more than 50 yards except on an animal host, so most cases of
infection occur near a particular locale. The same location may
be a source of subsequent cases. An outbreak occurred in 62 campers staying in log cabins on the north rim of the Grand Canyon
in 1973,21 and another cluster of 15 cases occurred in the same
location in 1990.22 The tick feeds at night and its painless bite
transmits the Borrelia to humans. Symptoms of disease are
similar to those of LBRF, although more relapses usually occur
in TBRF and case fatality rates are lower (25%).
Uveitis
Iritis and iridocyclitis may occur during the acute illness of LBRF.
Uveitis may also occur in TBRF. A case of anterior and intermediate uveitis recently occurred in a 12-year-old boy in Oregon
who had developed TBRF several weeks earlier.23
LEPTOSPIRA
LEPTOSPIROSIS
Leptospirosis is a worldwide zoonosis most common in tropical
regions. In the US, it is most common in Hawaii. It is caused by
various Leptospira species, most often Leptospira interrogans.
Leptospires are motile, tightly coiled spirochetes with pointed
ends. They are best seen by dark-eld microscopy and can be
cultured on polysorbatealbumin media. The leptospires are maintained in nature by chronic renal infection of carrier animals,
such as rodents and cattle, and human infection usually occurs
after exposure to contaminated water or damp soil. Clinical
disease is manifested either as a self-limited flu-like illness or as
a severe illness characterized by renal and liver failure as well as
a hemorrhagic pneumonia (Weils disease). Leptospirosis is a
biphasic illness in 50% of cases, with an asymptomatic period
between the two acutely symptomatic phases. Uveitis may occur
and may be anterior, posterior, or panuveitis.24 Retinal vasculitis
is seen in 550% of cases.
SPIRILLUM MINUS (RAT-BITE FEVER)

Spirillum minus is one of two causes of a relapsing, febrile illness
that follows a rat bite (the other being due to a Gram-negative rod,
Streptobacillus moniliformis). Spirillum minus is a short thick
spirochete (Table 13.1), and is carried by 25% of rats. Rat-bite
fever is rare. Most cases in the US are caused by Streptobacillus
moniliformis, while cases in Asia are caused by Spirillum minus.
In Japan, the illness is called sodoku (so = rat, doku = poison).
The illness occurs 14 weeks following a rat bite. The site of the
bite becomes swollen and purple, and subsequently ulcerates
and develops an eschar. There is regional lymphadenopathy and
a flu-like febrile illness, often accompanied by a maculopapular
rash. Fevers follow a relapsing course, with febrile episodes lasting
34 days and interspersed with afebrile periods lasting 39 days.
The organism cannot be cultured, and diagnosis is made by
microscopic visualization of the organism in blood, exudate, or
lymph node samples. Treatment is with penicillin. Eye disease
has not been reported.
Summary: Borrelia
Summary: Leptospirosis and Rat-Bite Fever
Lyme disease, caused by B. burgdorferi, is transmitted by a

tick and is endemic in Massachusetts to Maryland, Wisconsin
and Minnesota, and northern California.
There are three different groups of B. burgdorferi; the two
found in Europe and Asia cause a slightly different
manifestation of disease than the group (sensu strictu) found in
the US.
ELISA screening tests for Lyme have many false-positive
results and must be conrmed by a Western blot.
Relapsing fever is caused by Borrelia species and may be
either louse-borne or tick-borne. Uveitis has been described
in both forms. The tick-borne form is seen in the US, primarily
in patients who have camped in mountain cabins in the West.
Leptospirosis is a zoonosis seen most often in tropical or

subtropical regions.
In the US, leptospirosis has been most often seen in Hawaii.
Leptospirosis is biphasic in half of patients, with an initial flulike illness, recovery, then a late immune phase.
Uveitis may occur during the immune phase, weeks to months
following the initial illness.
Rat-bite fever is rare, and in Asia it is caused mainly by a
spirochete, Spirillum minus. There are no reports of eye
disease.
REFERENCES
136
1. Fraser CM, Norris SJ, Weinstock GM, et al:

Complete genome sequence of Treponema
pallidum, the syphilis spirochete. Science
1998; 281:375389.
2. Asai Y, Jinno T, Igarashi H, et al: Detection
and quantication of oral treponemes in
subgingival plaque by real-time PCR. J Clin
Microbiol 2002; 40:33343340.
3. Koseki T, Benno Y, Zhang-Koseki YJ, et al:

Detection frequencies and the colonyforming unti recovery of oral treponemes
by different cultivation methods. Oral
Microbiol Immunol 1996; 11:203208.
4. Seshadri R, Myers GS, Tettelin H, et al:
Comparison of the oral pathogen
Treponema denticola with other spirochete
genomes. Proc Natl Acad Sci USA 2004;

101:56465651.
5. Tramont EC: The impact of syphilis on
humankind. Infect Dis Clin N Am 2004;
18:101110.
6. Radolf JD: Role of outer membrane
architecture in immune evasion by
Treponema pallidum and Borrelia
7.
8.
9.
10.
11.
burgdorferi. Trends Microbiol 1994;

2:307311.
Tramont EC: Treponema pallidum (Syphilis).
In: Mandell GL, Bennett JE, Dolin R, eds.
Mandell, Douglas, and Bennetts principles
and practice of infectious diseases. 6th edn.
Philadelphia, PA: Elsevier Churchill
Livingstone; 2005:2770.
Liu H, Rodes B, Chen C-Y, et al: New tests
for syphilis: rational design of a PCR
method for detection of Treponema
pallidum in clinical specimens using unique
regions of the DNA polymerase I gene.
J Clin Microbiology 2001; 39:19411946.
Workowski KA, Berman SM (for the Center
for Disease Control and Prevention):
Sexually transmitted diseases treatment
guidelines, 2006. Morb Mortal Wkly Rep
2006; 55(RR11):194.
Thami GP, Kaur S, Gupta R, et al: Syphilitic
panuveitis and asymptomatic neurosyphilis:
a marker of HIV infection. Int J STD AIDS
2001; 12:754756.
Balba GP, Kumar PN, James AN, et al:
Ocular syphilis in HIV-positive patients
receiving highly active antiretroviral therapy.
Am J Med 2006; 119:448.e21448.e25.
12. Murphy FT, George R, Kubota K, et al: The

use of Western blotting as the conrmatory
test for syphilis in patients with rheumatic
disease. J Rheumatol 1999; 26:24485243.
13. Antal GM, Lukehart SA, Meheus AZ: The
endemic treponematoses. Microbes Infect
2002; 4:8394.
14. Scolnik D, Aronson L, Lovinsky R, et al:
Efcacy of a targeted, oral penicillin-based
yaws control program among children living
in rural South America. Clin Infect Dis 2003;
36:12321238.
15. Tabbara KF, Al Kaff AS, Fadel T: Ocular
manifestations of endemic syphilis (bejel).
16. Fraser CM, Casjens S, Huang WM, et al:
Genomic sequence of a Lyme disease
spirochete, Borrelia burgdorferi. Nature
1997; 390:580.
17. de Silva AM, Fikrig E: Arthropod- and hostspecic gene expression of Borrelia
burgdorferi. J Clin Invest 1997; 99:377.
18. Porcella SF, Raffel SJ, Schrumpf ME, et al:
Serodiagnosis of louse-borne relapsing
fever with glycerophosphodiester
phosphodiesterase (GlpQ) from Borrelia
recurrentis. J Clin Micro 2000; 38:35613571.
19. Seboxa T, Rahlenbeck SI: Treatment of

louse-borne relapsing fever with low dose
penicillin or tetracycline: a clinical trial.
Scand J Infect Dis 1995; 27:2931.
20. Fekade D, Knox K, Hussein K, et al:
Prevention of JarischHerxheimer reactions
by treatment with antibodies against tumor
necrosis factor alpha. N Engl J Med 1996;
335:311.
21. Centers for Disease Control and
Prevention. Relapsing fever. Morb Mortal
Wkly Rep 1973; 22:242246.
22. Paul WS, Maupin G, Scott-Wright AO, et al:
Outbreak of tick-borne relapsing fever at
the north rim of the Grand Canyon:
evidence for effectiveness of preventive
measures. Am J Trop Med Hyg 2002;
66:7175.
23. Lim LL, Rosenbaum JT: Borrelia hermsii
relapsing fever and uveitis. Am J
Ophthalmol 2006; 142:348349.
24. Rathinam SR: Ocular manifestations of
leptospirosis. J Postgrad Med 2005;
51:189194.
CHAPTER 13
The Spirochetes
137
CHAPTER
14
Parasitic and Rickettsial Ocular Infections

Michael S. Gilmore and Juan-Carlos Abad
PARASITIC INFECTIONS
INTRODUCTION TO PARASITOLOGY
Terminology
Parasitology is the study of different species from the animal
kingdom that live together or in close association (on or in the
body of another).1 A parasite living on the surface of its host is
an ectoparasite; an internal parasite is an endoparasite. Infestation is associated with ectoparasitism and infection with
endoparasitism. Parasites are either obligate (they exist only as
parasites) or facultative (they may also exist in a free-living
state). Parasites can be permanent (complete life cycle within
the host) or temporary.
Parasite Classication
Morphology, life cycle, genetics, reproduction, and aspects of
parasite growth and development are used to classify and categorize parasitic species. Serology, biochemistry, electron microscopy, isoenzyme electrophoresis, DNA, RNA, and protein
analysis techniques may be required to differentiate members of
a species that are otherwise indistinguishable.
Key Features: Parasitic Infection
Protozoa
Acanthamoeba, Trypanosoma, Leishmania, Giardia,
Toxoplasma, and Plasmodium
Metazoa
Platyhelminthes
Taenia and Schistosoma
Nematoda
Trichinella, Ascaris, Toxocara, and Onchocerca
Arthropoda
Sarcoptes and Demodex
The single-celled Protozoa, long considered to be one phylum,

have recently been divided into a number of groups assigned
phylum rank.2 These phyla are: Sarcomastigophora, Labyrinthomorphorpha, Apicomplexa, Microspora, Acestospora,
Myxozoa, and Ciliophora. Examples of human parasitic protozoans are Acanthamoeba, Trypanosoma, Leishmania, Giardia,
Toxoplasma, and Plasmodium.
The phylum Platyhelminthes are worms characterized by
bilateral symmetry with rudimentary development of sensory
and motor nerve elements. Platyhelminthes are divided into
four classes: Turbellaria, Monogenea, Cestoidea, and Trematoda.
Adult cestodes, commonly called tapeworms, have a head
(scolex) and a segmented body (strobila) and live within the
digestive tract of their host. Examples of Cestoidea are Taenia,

Echinococcus, and Spirometra. Adult trematodes in the subclass Digenea are commonly called ukes, and their development occurs in at least two hosts. Examples of Trematoda are
Schistosoma and Paragonimus.
The phylum Nematoda comprises a large number of organisms commonly known as roundworms. Nematodes are divided
into two classes, Phasmidia and Aphasmidia, based on the presence or absence of cuticle-lined organs (phasmids). Examples
of nematodes are Trichinella, Ascaris, Toxocara, Dracunculus,
Loa, and Onchocerca.3
The phylum Arthropoda includes organisms from the classes
Arachnida, Insecta, and Crustacea; all have a hard cuticle
exoskeleton. Examples of Arthropoda are Sarcoptes, Demodex,
Phthirus, Oestrus, Dermatobia, and Hypoderma.
Table 14.1 is a summary of parasites that cause major ocular
diseases.
HOSTPARASITE INTERACTIONS
Interactions between the host and the parasite are crucial for
maintenance and continued transmission of parasitic infections. Parasitic adaptations that limit the host response include:
(1) life-cycle stages (eggs, larvae, adult organisms, cysts) that
evoke different host immune responses; (2) parasite surface
composition variation;4 and (3) tissue location (i.e., intracellular
versus extracellular). Host factors that render humans
particularly susceptible to infection include: (1) nutritional
status/malnourishment, (2) genetic susceptibility (a relative
resistance to Plasmodium vivax occurs in African-Americans,
and it has been attributed to the Duffy-negative phenotype
present in this population5), and (3) endogenous or exogenous
immunosuppression.
PROTOZOA
ACANTHAMOEBA
Several genera of free-living amebae cause disease in humans.
Acanthamoeba infections are the most important among ocular
pathogens. They cause keratitis in healthy persons. In immunosuppressed patients, Acanthamoeba infections may result in
granulomatous amebic encephalitis (GAE) and disseminated
infection. Vahlkampa and Hartmannella have also been implicated as a cause of infectious keratitis.6
Distribution
Acanthamoeba species are widespread in nature. They are
found in fresh, sea, tap, bottled, and brackish water,7 as well as
in dust, sewage, sludge, swimming pools (especially in warmer
139
MICROBIOLOGY
TABLE 14.1. Ocular Parasitic Diseases in Humans

Parasite
Ocular Lesions
Geographic
Distribution
Laboratory Tests
Therapy
SECTION 3
Protozoa
Acanthamoeba
Indolent, painful corneal

ulcer and infiltrates,
iridocyclitis
Worldwide
Calcoflour white stain,

culture on Escherichia coli
Polyhexamethylene
biguanide or
chlorhexidine;
propamidine or
hexamidine;
itraconazole
American trypanosomiasis
(Tripanosoma cruzi)
Bipalpebral edema,
unilateral conjunctivitis,
Romaas sign
Central and South

America
Blood smears
Nifurtimox
Giardia lamblia
Retinal vasculitis
Worldwide
Cysts and trophozoites in stool
Metronidazole
Leishmania tropica,
braziliensis (Oriental
sore, espundia)
Lid ulcer
Middle East, Asia

Minor, Central
and South
America
Scrapings of skin lesions
Antimony sodium
gluconate,
allopurinol, or
ketoconazole
Malaria (Plasmodium
species)
Retinal hemorrhages, papillederma, retinal edema
Equatorial region
Blood smear
Chloroquine,
primaquine
Microsporidiosis
(Encephalitozoon
species in immunosuppressed patients)
(Nosema species in
immunocompetent
patients)
Superficial epithelial
keratopathy
Worldwide
Corneal scrapings
Debridement, topical
fumagillin,
itraconazole
Stromal keratitis
Worldwide
Corneal scrapings and biopsy
Trimethoprimsulfamethoxazole
Pneumocystis carinii
Choroidal granulomas
Worldwide
Bronchial washings, sputum

cultures, tissue biopsy
Pentamidine
isothionate,
trimethroprimsulfamethoxazole
Toxiplasma gondii
Retinochoroiditis, papillitis,
retinal vasculitis, uveitis,
secondary glaucoma
Worldwide
Serum ELISA, aqueous or

vitreous PCR
Pyrimethamine,
trisulfapyrimidine
or sulfadiazine,
clindamycin;
steroids, laser,
cryotherapy
Rare intraocular worm,

vitamin A deficiency
Worldwide
Eggs in stool, complement

fixation larva in ocular
granuloma or histopathology
Mebendazole,
piperazine
Baylisascaris procyonis
Diffuse unilateral
subacute retinitis
Southeastern
United States
and Caribbean
Direct observation
Laser
photocoagulation;
thiabendazole or
ivermectin
Dracunculus medinensis
Filariasis
Eyelid and orbital mass
Africa and India
Examination of the worm
Surgical excision
1. Dirofilaria species
Periobital or intraocular
worm
Worldwide
ELISA
Surgical excision
2. Lymphatic filariasis
(Wuchereria
bancrofti, Brugia
malayi, Brugia
timor)
Elephantiasis, anterior
chamber or subretinal
microfilaria (rare)
Tropical areas,
Far East
Peripheral blood
Diethylcarbamazine
3. Loa loa
Subcutaneous nodule,
subconjunctival worm,
periobital swelling and
pain
Central Africa
Blood smear, tissue biopsy
Diethylcarbamazine
4. Onchocerca
volvulus
Skin and eye nodules,

keratitis, uveitis,
chorioretinitis, optic
atrophy
Africa, Central and

South America
Skin snip, nodule biopsy
Ivermectin
Intestinal Nematodes
Ascaris lumbricoides
Extraintestinal Nematodes
140
Continued
TABLE 14.1. Ocular Parasitic Diseases in HumansContd

Geographic
Distribution
Laboratory Tests
Therapy
Conjunctivitis, extraocular
muscle paresis, orbital
granuloma
Central America
Biopsy lesion for worm
Surgical excision
Toxocara canis, cati
Posterior and peripheral

retinal granuloma,
panuveitis
Worldwide
ELISA on serum, aqueous

or vitreous; CT
Thiabendazole,
mebendazole
Trichinella spiralis
Lid and periorbital edema,

extraocular muscle
paresis and pain
Worldwide
Serology, skin biopsy
Thiabendazole and
steroids
Paragonimus westermani
Periocular cyst
Far East, India,

Africa, Central
and South
America
Eggs in feces or sputum,

serum ELISA
Praziquantel
Schistosoma haematobium
and japonicum (bilharzia,
schistosomiasis)
Dacryoadenitis, conjunctival
and orbital granulomas
Africa, Middle
East, Far East
Eggs in urine, lesion biopsy,

CT
Praziquantel,
niridazole
Coenuriasis (Multiceps
multiceps, Taenia
brauneri)
Lids and intraocular cyst
Sheep-raising areas
(New Zealand),
Argentina,
California
Casonis intradermal test
Surgical excision
Echinococcus granulosus
Orbital cyst (common),

intraocular cyst (rare)
Sheep-raising
areas (Africa)
Skin test, indirect

hemagglutination or
immunofluorescent serology,
radiography, CT
Praziquantel
Sparganum proliferum
Orbit or anterior chamber

cyst
Far East
DIrect observation
Surgical excision
Cysticercus cellulosae
Intraocular granuloma
Worldwide
Skin test, radiograph for

calcified cysts
Praziquantel,
niridazole
Chronic blepharitis
Worldwide
Direct observation
Lid hygiene
1. Ophthalmomyasis
externa
(Dematobia
hominis,
Chrysomia
bezziana)
Lid furuncule and cellulitis,

orbital invasion
Central and South

America,
Old World
Direct observation
Mechanical removal
2. Ophthalmomyasis
interna
(Hypoderma
lineaturm)
Subretinal tracks,
intravitreal invasion
Tropical areas
Direct observation,
parasite recovery
Laser
photocoagulation,
removal of the
parasite
Ophthalmia nodosa
(caterpillar hairs)
Conjunctival nodule
Worldwide
Histopathology
Surgical excision
Phthirus pubis
Chronic blepharitis
Worldwide
Direct observation
Lid hygiene, antibiotic

or eserine ointment
Parasite
Ocular Lesions
Thelazia callineda or
californiensis
Tapeworms
CHAPTER 14
Trematodes (Flukes)
Arthropods
Demodex folliculorum
Myasis
CT, computed tomography; ELISA, enzyme-linked immunosorbent assay; PCR, polymerase chain reaction.
months), hot tubs, air conditioning ducts, dialysis units, human

and animal feces, human oral cavities, and contact lenses
and associated paraphernalia. Acanthamoeba cysts are stable
and still infective after being stored in water at 4C for
24 years.8
Acanthamoeba keratitis has been associated with corneal
trauma, exposure to contaminated water and dust,9 and contact
lens wear. The use of homemade saline solutions, improper
contact lens care, and eye exposure to contaminated water
while wearing lenses are responsible for the association of
Acanthamoeba with contact lens use.10 Males and females are

affected equally. Since the rst documented case of Acanthamoeba keratitis was reported in 1973,11,12 the number of cases
has increased steadily.3,13 A recent series using a confocal microscope as a diagnostic aid suggests that Acanthamoeba keratitis
may be more common than previously thought.14
GAE remains infrequent.15 Several cases of disseminated
Acanthamoeba infection in patients with acquired immunodeciency syndrome (AIDS) with mainly cutaneous manifestations have been reported.16
141
MICROBIOLOGY
SECTION 3
Morphology, Biology, and Life Cycle

Acanthamoeba exists in two stages: trophozoite and cyst.
Trophozoites are the proliferative, active forms; and size
depends on species (2040 mm).7 They have irregular shape and
pseudopodia with characteristic spine-like processes (Fig. 14.1).
The cytoplasm contains a single nucleus with a large, dense,
central nucleolus surrounded by a clear zone called the zona
pellucida. Cytoplasmic organelles are evident, as is a characteristic large contractile vacuole. Trophozoites move by gliding in
straight lines and feed on Escherichia coli and other enteric
Gram-negative bacilli. The trophozoite, when exposed to
unfavorable conditions (desiccation, lack of food, contact
with toxic substances or solutions), undergoes immediate
encystment. Acanthamoeba proliferate by binary ssion.
Acanthamoeba cysts are the resistant, dormant stage of this
parasite. Cysts are characterized by a double-walled envelope.
The outer wall, the exocyst, is wrinkled, and the inner wall,
the endocyst, is smooth. There is a space between the two
walls except at the ostiole, where the exocyst is joined to the
endocyst. Cyst morphology and size are species-specic
(12.519.2 mm), and encystment states can be differentiated by
shape (e.g., spherical, polygonal).7 The cytoplasm of the cyst
contains a single nucleus located centrally, several lipid
droplets, mitochondria, and other cytoplasmic organelles but
lacks a functioning contractile vacuole. Excystment occurs

when favorable environmental conditions return.
Infection of the Host

The mechanism for development of Acanthamoeba keratitis
may be related to epithelial trauma, strain virulence, the number
of organisms present, and favorable amebacornea contact
conditions.17 The proliferation and binding of Acanthamoeba to
contact lenses is enhanced by co-contamination of the contact
lens care system with Gram-negative bacteria.18 Acanthamoeba
infection causes destruction of the corneal epithelium and
stroma, with subsequent inltration of inammatory cells,
descemetocele formation, and corneal perforation.19 The
cellular reaction around necrotic organisms may be more
intense.20 Acanthamoeba castellani has been shown to produce
a plasminogen activator21 and nonspecic collagenases,22 which
might be related to its pathogenicity.
Diagnosis
In cases of Acanthamoeba keratitis, smears and culture isolation are the initial diagnostic steps. Generally, deep corneal
scrapes are necessary to detect Acanthamoeba. The confocal
microscope has been used for in vivo diagnosis of Acanthamoeba
keratitis.23,24 If these diagnostic measures are unrewarding and
clinical suspicion is high, corneal biopsy is recommended.25
Corneal Smears
142
In Giemsa-stained or Gram-stained samples, Acanthamoeba

may resemble leukocytes, macrophages, and other mononuclear
cells (Fig. 14.2). Gomori-methenamine silver (stains the cyst
wall black) as well as periodic acid-Schiff (stains the cyst wall
red) may help in identifying the organisms. Calcouor white, a
chemouorescent dye, has proved useful in detecting Acanthamoeba cysts.26 Smear preparations can be xed in methyl alcohol
and processed using an aqueous solution of 0.1% calcouor
white with Evans blue counterstain. The slides are examined by
uorescent microscopy. The cyst wall appears bright apple-green;
trophozoites and other cells appear red-brown. Fluorescent
antibody staining of corneal scrapes can also provide a rapid
diagnosis of Acanthamoeba keratitis with the added advantage
of species differentiation.27 Slides can be xed in 10% buffered
formaldehyde, incubated with diluted rabbit anti-Acanthamoeba
serum, followed by second-labeled antirabbit serum. Cysts and
trophozoites uoresce brightly. More recently, isoenzyme
proles28 and restriction fragment length polymorphisms of
mitochondrial DNA29 have been used in differentiating
Acanthamoeba.
FIGURE 14.1. Acanthamoeba trophozoites; unstained culture, fresh

wet preparation, phase contrast (400). Acanthamoeba species (a),
A. polyphaga (b), A. culbertsonii (c), A. astronyxis (d), and
A. castellani (e).
FIGURE 14.2. Corneal scraping from a patient with Acanthamoeba

keratitis shows double-walled polygonal cysts. Giemsa stain 400.
Acanthamoeba grows at 2535C. For corneal culture recovery,

nonnutrient agar overlaid with E. coli is a common culture
medium. The scraped specimen is placed on the agar surface
without streaking or cutting the agar. The plates are sealed with
adhesive tape to prevent dehydration and observed for a minimum of 2 weeks. If culture plates are not available, transport
solutions can be used. Pages saline solution (a low-osmolarity
solution) allows trophozoites to survive transportation at
ambient temperature for up to 48 h.30
Corneal Biopsy
If corneal smears and cultures from the corneal scrapings are
negative, corneal biopsy is the next viable diagnostic approach.
A 34-mm dermatologic punch is used to make a half-thickness
corneal trephination straddling the lesion and normal cornea.
The specimen is split in half. One part is xed in glutaraldehyde
for light and electron microscopy studies. The other half is
hand-carried to the microbiology laboratory for bacteria, fungi,
and Acanthamoeba culture. The same diagnostic stains and
culture media used in the scrapings are used in addition to
uorescent antibody stains. Electron microscopy may be used
as well.31
Prevention
Acanthamoeba keratitis, because of its association with contact
lenses, may be prevented by meticulous lens care and
sterilization precautions. Thermal disinfection solutions are
effective against Acanthamoeba.32 For lenses requiring chemical
disinfection, solutions containing chlorhexidine killed Acanthamoeba in 30 min, benzalkonium chloride systems required
at least 1 h, and hydrogen peroxide systems required up to 2 h.33
Solutions containing sorbate, polyaminopropylbiguanide, or
polyquaternium-1 may not be effective in killing Acanthamoeba
organisms.33 Contact lenses should not be worn during activities
that may increase exposure to potentially contaminated water.
Treatment
Cationic antiseptics such as polyhexamethylene biguanide
(Baquacil)34 and chlorhexidine35 kill Acanthamoeba cysts and
trophozoites by disrupting the parasites plasmalemma.
Aromatic diamidines, such as propamidine isethionate (Brolene)
and hexamidine (Desomedine),36 inhibit the parasites DNA
synthesis and can be used in combination. Aminoglycosides
(neomycin, paromomycin) and the antifungal imidazoles (miconazole, clotrimazole37) have some efcacy as topical agents.
Oral itraconazole has been used by some authors,38 and higher
doses of antimicrobials may provide additional value in
treatment.39 Early animal work suggested that corticosteroids
block the conversion of trophozoites to cysts, hence enhancing
the effect of the amebicidal medications, although this remains
controversial.40 Steroids suppress the hosts immune response
and decrease inammatory signs, making the patient more
comfortable,41 but they may be associated with a poor outcome.42 A subconjunctival vaccine composed of Acanthamoeba
antigens was successfully evaluated in a pig model.43
AMERICAN TRYPANOSOMIASIS
American trypanosomiasis (Chagas disease) is caused by the
protozoan Trypanosoma cruzi. South and Central America are
endemic areas of Chagas disease.

In Chagas disease, triatomid insects are infected with the
parasite during a blood meal from a contaminated human.
They are also called besadores (kissing bugs) because of their
tropism to bite in the head region. During the next blood meal,
the insect defecates near the bite wound; the host experiences a
mild itching sensation and rubs the feces contaminated with
trypomastigotes into the insect bite. If the insect bites near the
eye or mouth, the parasites can penetrate directly into the host
via mucosal membranes. Trypomastigotes enter a wide variety
of cells (cardiac, striated muscle bers, and cells of the reticuloendothelial system), where they transform into amastigotes
(1.55 mm in length; aagellated). Intracellularly, the amastigotic forms replicate by binary ssion and destroy the cell.
Amastigotic forms released in the peripheral blood rapidly
transform into trypomastigotes and infect other cells or are
ingested by triatomid insects. American trypanosomiasis can be
transmitted congenitally and in blood transfusion.

In Chagas disease, acute-phase reactions depend on the route
of entry of the parasite. When the trypanosomes enter via the
conjunctiva, Romaas sign (unilateral bipalpebral edema with
conjunctivitis and lymphadenopathy) may be observed.44 If
trypanosomes enter through the skin, a hypersensitivity
reaction, called chagoma (furuncle-like lesion with swelling of
the regional lymph nodes), may be present. There is a mild
febrile illness that usually goes unnoticed. In the chronic phase,
cardiomyopathy and motility alterations of the digestive tract
(megaesophagus and megacolon) are common complications.
CHAPTER 14
Acanthamoeba Culture
Diagnosis
During the acute stage of Chagas disease, direct examination of
peripheral blood smears can conrm the diagnosis of trypanosomiasis. Fresh anticoagulated blood may demonstrate motile
trypomastigotes, or the parasite may be identied on Giemsastained blood smears. During chronic disease, the parasite is
rarely found in the peripheral blood. Xenodiagnosis (the feeding
of uninfected triatomids on an infected patient and subsequent
demonstration of parasites in the insect), hemoculture, or
animal inoculation are limited by the time lag until they
become positive.45 Serologic examinations are affected by crossreactivity with antileishmaniasis antibodies.46 Clinical ndings
of cardiac arrhythmias, right bundle branch block, and heart
failure in conjunction with megaesophagus and megacolon in a
patient from an endemic area suggest trypanosomiasis.
Prevention
For Chagas disease, elimination of triatomid insects in endemic
areas is useful. Chemoprophylaxis is controversial. The use of
insect repellents and appropriate clothing decreases the chances
of acquiring the infection.
Treatment
Nifurtimox and benznidazole can be used in the treatment of
acute trypanosomiasis.47 They have no proven effect on the
chronic manifestations of the disease.
LEISHMANIASIS
Leishmaniasis is a cutaneous, mucocutaneous, or visceral
infection caused by protozoa of the genus Leishmania (family
Trypanosomatidae).
Distribution
Four major clinical syndromes are caused by several species of
leishmania: cutaneous leishmaniasis of the Old (L. tropica) and
New (L. mexicana and L. braziliensis) Worlds; mucocutaneous
leishmaniasis or espundia (L. braziliensis braziliensis); diffuse
cutaneous leishmaniasis in patients with decreased immunity;
and visceral leishmaniasis, or kala-azar (L. donovani).
143
MICROBIOLOGY
Leishmania organisms are found in two stages: promastigote

(agellated) and amastigote (nonagellated). The life cycle alternates between the vector sandy Phlebotomus (Old World) or
Lutzomyia (New World) and a mammal host. The female y
acquires the parasite during a blood meal from an infected host.
The promastigotic form (infectious stage for humans) proliferates extracellularly in the intestine of the sandy and is
introduced into the mammalian host by the y bite. Promastigotes in the host enter macrophages and transform into
obligate intracellular amastigotes ((25.5) (12 mm)). Disease
spread occurs through infection of new macrophages, following
lysis of parasite-infected cells.
The parasites are transmitted through the bite of the infected

female anopheline mosquito, the denitive host for all
Plasmodium species. The mosquito becomes infected when it
ingests the macrogametocytic and microgametocytic forms of
the parasite in the peripheral blood of an infected human, the
intermediate host. After fusion of the gametocytes (sexual
cycle), a zygote develops into an ookinete, forms an oocyst, and
then differentiates into sporozoites. The sporozoites (23 mm),
the infectious form of the parasite, remain in the mosquitos
salivary glands and are inoculated into humans along with the
salivary secretions during blood feeding. The sporozoites, once
in the human circulatory system, rapidly enter the hepatic
parenchymal cells, differentiate into merozoites (1.5 mm),
replicate, rupture the cells, and are released back into the
circulatory system. Alternatively, in infections by P. vivax and
P. ovale, hepatic merozoites can differentiate into hypnozoites,
a dormant form that can cause disease relapse many years later.
Merozoites released into the circulatory system cannot enter
new parenchymal cells but enter red blood cells instead,
initiating the erythrocytic cycle. In red blood cells, merozoites
transform into trophozoites, which enlarge and then give rise to
multiple merozoites (schizogony) that rupture the red blood
cells and are released into the circulatory system to enter new
red blood cells. Trophozoites can also differentiate into macrogametocytes (female presexual stage, 10 mm) or microgametocytes (male presexual stage). The macrogametocytes and
microgametocytes are ingested by the anopheles mosquito
during the blood feeding and reinitiate the sexual life cycle.
SECTION 3

The human cutaneous infection, in the early form of the
disease, is a single nodule at the site of the bite. The nodule can
progress centrifugally, ulcerate, and scar. Mucocutaneous leishmaniasis is characterized by lesions involving the lower
extremities, followed by lesions of mucous membranes and
cartilage of the oral cavity, nasal septum, and larynx. Ocular
infection may result in eyelid edema, ulceration, and scarring.
Conjunctival granuloma and interstitial keratitis have been
reported.46,48
Diagnosis
Denitive diagnosis of leishmaniasis is by direct identication
of the parasite. Stained smears (Wrights or Giemsa stain) or
biopsy (H&E or Wilders reticulin stain) may demonstrate
amastigotic or intracellular forms. Needle aspiration culture
from the lesion edge or inoculation of a tissue biopsy specimen
in appropriate culture media may demonstrate the promastigotic form. Serologic tests provide only indirect evidence of
Leishmania infection. The leishmanin skin test (Montenegro
test) is a delayed hypersensitivity reaction to dead promastigotes injected intradermally. Negative hypersensitivity results
occur in cases of diffuse cutaneous leishmaniasis, and strongly
positive results occur in leishmaniasis recidivans. In visceral
leishmaniasis, the leishmanin skin test result is negative during
active disease and positive in most patients several months to
1 year after recovery.

Sudden attacks of headaches, spiking fever, perspiration, and
shaking chills, interspersed with asymptomatic normal periods,
are clinical symptoms of acute-phase malaria. Subacute,
chronic, and recurrent forms of the disease also can occur.
Ocular manifestations of malaria include blotchy preretinal and
retinal hemorrhages believed to be caused by cytoaggregation of
the parasitized erythrocytes.53,54 In children with cerebral
malaria, papilledema or retinal edema beyond the arcades are
markers of a poor prognosis.55
Diagnosis
Prevention
Insect repellents, appropriate clothing, and y netting may
provide protection.
Treatment
The drugs of choice for all forms of the disease are pentavalent
antimonials: sodium stibogluconate or meglumine antimoniate
(Glucantime). Alternatives for cutaneous leishmaniasis include
allopurinol49,50 or ketoconazole.51 Amphotericin B and pentamidine can be used in severe cases.47
MALARIA
Malaria is an infection caused by the protozoan Plasmodium.
Four species have been identied as human pathogens:
P. falciparum, P. vivax, P. ovale, and P. malariae. P. vivax, the
species most commonly infecting humans, causes benign tertian
malaria. P. falciparum is the most dangerous species, causing
malignant tertian malaria.
Distribution
144
Malaria is endemic in hot and humid (tropical or subtropical)

regions of Africa, Asia, and Central and South America,
affecting an estimated 200 million people and causing over
1 million deaths every year, especially among children.52
Malaria is diagnosed by detection of the trophozoite or gametocyte in blood smears. Several smears should be collected at
hourly intervals and stained with Giemsa or Grams stain.
Two smears should be prepared at each time interval, one
thick, for parasite detection, and another thin, for morphologic
analysis. Diagnostic serologic techniques are not routinely
available.56
Prevention
Prevention of malaria is achieved by personal protection from
mosquito exposure and by the use of insecticides. Chemoprophylaxis can also be used in endemic areas. Blood banks
should follow the American Association of Blood Banks regulations in screening donors for preexisting malarial infection.56 A
malaria vaccine against the merozoite has shown variable
results.57,58
Treatment
Chloroquine is the drug of choice for the erythrocytic phase of
the infection. In cases of chloroquine-resistant P. falciparum,
quinine or the antiarrhythmic quinidine could be used. Alternatives include meoquine and pyrimethamine/sulfadoxine
(Fansidar). Primaquine is used to eradicate the hypnozoites in
cases of infections by P. vivax or P. ovale. Caution should be
taken in patients with glucose-6-phosphate deciency. A
MICROSPORIDIOSIS
Microsporidia is the nontaxonomic term given to a group of
eukaryotic, obligate intracellular protozoan parasites. They
infect a wide variety of life forms, ranging from protozoa to
humans. Only two genera of Microsporidia, Encephalitozoon
and Nosema, cause infection that affects the ocular tissues.

Microsporidia are endemic in the tropics,59 but it seems that
not all healthy people are susceptible to this disease. Recognition of this disease has increased because of the AIDS
pandemic. Horizontal transmission is believed to take place in
animals and possibly in humans. Infection with Microsporidia
is believed to occur after ingestion or inhalation of spores from
fecal or urine contamination. The spores that infect humans
usually measure 12 mm by 24 mm.60 Organisms usually infect
the epithelial cells in the intestinal or respiratory tracts, and
from there they could disseminate to other organ systems.61
The most common presentation of Microsporidia in humans is
chronic diarrhea in AIDS patients.61 Two forms of keratitis are
recognized. The rst type is caused by Nosema, which affects
immunocompetent people and produces stromal keratitis.62,63
Only four cases have been reported. The second type is caused
by Encephalitozoon, and it affects the corneal epithelium in the
form of punctate epithelial keratitis in AIDS and immunosuppressed patients.64
the second and third decades of life.72 In contrast with intracranial disease, toxoplasmic retinochoroiditis appears to be
uncommon in patients with AIDS.73

T. gondii exists in three forms. Trophozoites (tachyzoites) are
the propagative form of the parasite. Tissue cysts (bradyzoites)
occur in the chronic stage of the disease. Oocysts are shed in the
cats feces after sexual reproduction of the parasite (Fig. 14.3).
Intestinal phase
When cats are infected by ingestion of bradyzoite cysts from an
infected intermediate host, such as rodents and birds, bradyzoites rapidly transform into tachyzoites, penetrate the cats
intestinal mucosa, and undergo an enteroepithelial cycle of
sexual proliferation, resulting in the development of oocysts.
Oocysts detach from the intestinal epithelium and are shed in
the feces. Each oocyst ((1114) (911) mm) contains four
sporozoites. In the external environment, the oocyst undergoes
sporulation within 13 days and then becomes infectious. Cats
can shed 3100 million oocysts after primary infection.
Tissue phase
Intermediate hosts (as well as cats) can be infected by: (1)
ingesting bradyzoites or tachyzoites from uncooked meat,
unpasteurized milk, or contaminated water from an intermediate host; (2) ingesting or inhaling oocysts shed in the cats
feces; and (3) congenital transmission of tachyzoites (see
Fig. 14.3). After exposure, the host immune defenses are
initiated, and the proliferative stage of the infection is curtailed.
Organisms encyst and remain viable in the cell tissues, where
they can reactivate at a later date.
CHAPTER 14
number of antibiotics, including the tetracyclines, rifampin,

clindamycin, trimethoprim, sulfonamides, and doxycycline,
have some effect.47
Diagnosis
In corneal scrapes, the acid-fast and Gomori-methenamine silver
stains demonstrate the organism well.63 Electron microscopy
might be required for the diagnosis. Histopathologic features of
keratoplasty specimens in patients with corneal nosematosis
demonstrate invasion of the stroma by multiple organisms,
areas of necrosis, and multinucleated giant cells. In cases of
AIDS, the parasites seem to be conned to the corneal epithelium with absent inammation.65

Toxic products from Toxoplasma and hypersensitivity reactions
are responsible for the tissue damage. Inammatory reactions
are not usually observed around the bradyzoite cysts, owing
possibly to incorporation of host elements into the cyst walls,
masking the parasite antigens.74 The infection recurs when a
cyst ruptures, releasing parasites that proliferate and invade
neighboring cells. Bradyzoite cysts can be located in many
Treatment
In cases of Encephalitozoon keratitis, local debridement65 could
be combined with topical fumagillin.66 Oral itraconazole or
albendazole67 has been used as an adjuvant.
TOXOPLASMOSIS
Toxoplasmosis is an infection caused by the protozoan Toxoplasma gondii. Cats are the only known denitive host of the
parasite, but intermediate hosts, including humans, are at risk
of infection.
Distribution
Both animals and humans demonstrate serologic evidence of
Toxoplasma infection worldwide. Toxoplasmosis can be
congenital or acquired. In the United States, 3060% of adults
have positive serology results for Toxoplasma.68 In developing
countries, acquired toxoplasmosis occurs at a younger age with
a higher prevalence in the adult population.68 In congenital
toxoplasmosis, 45% of untreated women that develop primary
toxoplasmosis during gestation give birth to infected infants;
8% of these infants are severely affected.69 Estimates of fetal
infection in the United States range from 4200 to 16 800 cases
per year.70 T. gondii is one of the most frequent causes of
retinochoroiditis and posterior uveitis,71 occurring mainly in
FIGURE 14.3. Toxoplasmosis. Life cycle of Toxoplasma gondii. The

human as an intermediate host could get infected by ingesting
oocysts shed in the cats feces, by eating meat contaminated with
tissue cysts, or by transplacental (congenital) infection.
145
MICROBIOLOGY
tissues and are most numerous in the brain, skeletal muscle,
myocardium, and retina.68
Infection in immunocompetent patients

The acute infection in healthy persons leads to a mononucleosislike clinical picture with fever, malaise, headache, arthralgia,
hepatosplenomegaly, and lymphadenopathy. It is transient and
usually of no consequence, except in cases of placental
transmission or delayed retinochoroiditis.
ocular production of antibodies, thus aiding in the diagnosis of

difcult ocular toxoplasmosis cases.80,81 Other serologic tests,
such as complement xation, hemagglutination, latex agglutination, and immunouorescent antibody, have been largely
replaced by the ELISA test. The PCR may be useful in detecting
Toxoplasma parasite DNA when cysts cannot be visualized.82 In
cases of retinochoroiditis, the diagnostic yield of PCR is higher
in the vitreous than in the aqueous.83
Histologic identication
SECTION 3
Infection in immunocompromised patients

Toxoplasmosis in the immunocompromised host is most
probably reactivation of a previous latent infection,75 although
in certain circumstances (leukemia and organ transplantation),
infection can be acquired from blood transfusions and contaminated donor tissue. The cell-mediated immune response is
an important mechanism for resistance to T. gondii infection.
Chronic immunosuppression can reactivate latent infection.
Retinochoroiditis
The most common form of retinal involvement is necrotizing
retinochoroiditis, although cases of neuroretinitis75 and progressive panophthalmitis76 have been reported. Elderly patients
seem to be prone to a particularly severe form of Toxoplasma
retinochoroiditis.77 Ocular disease in healthy persons is mainly
the result of reactivation of encysted organisms after congenital
infection,72 although several cases of acquired retinochoroiditis
have been reported from endemic areas.78 Ruptured retinal cells
sensitize lymphocytes and initiate the production of autoantibodies that may contribute to the retinitis.79
The parasite is identied by routine microscopic examination of

H&E-stained or Giemsa-stained tissue sections. Identication
of tachyzoites indicates an active infection; detection of cysts
indicates a chronic stage of the disease (except for identication
of cysts in placental or fetal tissues). Fluorescent antibodies84
and peroxidaseantiperoxidase techniques85 are reliable methods
for Toxoplasma detection.
Prevention
Oocyst contamination
Toxoplasma oocysts can be destroyed by exposure to heat in
excess of 60C; chemical disinfectants are usually ineffective.
Hand washing is indicated after contact with soil contaminated
by cat feces and when changing cat litter boxes.
Bradyzoite contamination
Bradyzoite cysts in tissues may remain viable in meat for several
days at room and refrigerator temperatures. All bradyzoites are
destroyed by cooking meat to 70C. Hands should be washed
after handling raw meat. Soap, alcohol, and chemical disinfectants inactivate bradyzoites on the skin.
Congenital infection
Congenital transmission of toxoplasmosis occurs when a
Toxoplasma infection is acquired during pregnancy or 6 months
earlier. The neonate of a woman with previous antibodies to
Toxoplasma will not have congenital toxoplasmosis.72 The
disease is usually more severe in the fetus than in the mother.
Transplacental transmission of Toxoplasma increases when the
infection is acquired in the second and third trimesters of
pregnancy. Severe fetal disease, however, is more prevalent when
the infection is acquired in the rst trimester of pregnancy.69
Congenital toxoplasmosis
Pregnant women should be cautioned about exposure to Toxoplasma. Seronegative pregnant women in high-incidence areas
may be tested repeatedly; if seroconversion is detected, prompt
therapy should be initiated with nonmutagenic drugs. To facilitate early diagnosis and treatment, pregnant women in highincidence areas should be familiarized with the clinical
symptoms of acquired toxoplasmosis.
Treatment
Diagnosis
Laboratory diagnosis of T. gondii infection includes serologic
analysis and its histologic identication.
Serologic tests
146
The high prevalence and persistence of Toxoplasma antibodies

in the general population makes interpretation of serologic test
results difcult. Diagnosis of acquired infection requires
demonstration of seroconversion and a rise in antibody titer in
samples taken 46 weeks apart. The presence of Toxoplasmaspecic IgM indicates a recently acquired infection. Because
IgM does not cross the placenta, an increase in IgM titers in the
neonatal period is an indicator of congenital toxoplasmosis.
Recurrent Toxoplasma chorioretinitis may not increase IgG
levels, and IgM antibody is not detected. When ocular lesions
suggest toxoplasmosis, serum antibodies are considered to be
signicant at any level of detection, although a positive serologic
test result is not conclusive proof of toxoplasmosis. A negative
serologic test result in an undiluted sample should exclude the
diagnosis of toxoplasmosis, although exceptions have occurred,
especially in patients with AIDS.76 No association between
serologic Toxoplasma antibody titers and eye disease severity
has been reported. ELISA is used to identify and quantify IgM
and IgG antibodies individually. Toxoplasma antibodies can be
detected in ocular uids, and the ELISA can demonstrate local
Although Toxoplasma eye disease is self-limiting, some cases

may require treatment. The combination of sulfadiazine and
pyrimethamine86 (given concomitantly with folinic acid) is
usually the rst line of treatment in cases of toxoplasmic retinochoroiditis. Clindamycin,87 spiramycin, and trimethoprim
sulfamethoxazole are alternative drugs. Steroids can be added to
the antimicrobial therapy if the ocular lesions threaten the
macula or the optic nerve. Cryotherapy and laser photocoagulation may be indicated in special cases.
METAZOA
INTESTINAL NEMATODES
Ascariasis
Ascariasis is a nematode infection caused by Ascaris
lumbricoides.
Distribution
Ascariasis occurs worldwide, more frequently where hygiene
and sanitary conditions are inadequate.
Morphology, biology, and life cycle

Ascaris infection occurs when fertilized eggs (4570 mm
3550 mm) are ingested from contaminated soil or vegetables.

Ingested eggs hatch in the host intestine after the outer coating
is dissolved by gastric acid. The larvae penetrate the intestinal
mucosa and are disseminated via the lymphatic and circulatory
systems. The larvae become trapped in the lungs circulation,
penetrate the alveolar wall, migrate to the trachea and
esophagus, and are swallowed. In the small intestine, the larvae
mature and mate. Adult A. lumbricoides are large parasites
(female, 2040 cm 36 mm; male, 1530 cm 24 mm). The
female passes an average of 200 000 eggs a day.
Filariasis
Infection of the host
Dirolariasis
The adult parasite inhabits the small intestine, where it can

cause symptoms that range from vague abdominal pain to complete intestinal obstruction. Single worms can migrate to the
biliary tree, pancreatic duct, or appendix, causing obstruction.
In cases of massive ascaris infection, vitamin A absorption may
be decreased, which in turn causes xerophthalmia.88 Systemic
manifestations can occur during the larval migration stage,
including fever, pneumonitis, and even invasion of the intraocular or periocular tissues.
Dirolaria immitis is the heartworm of dogs; D. repens is found

in cats and dogs in Asia, Europe, and South America; and
D. tenuis infects raccoons in North America. They are accidentally transmitted to humans by the same vectors that infect the
animal hosts, Aedes and Culex mosquitos. The parasite is
unable to produce microlariae in the human host. Subcutaneous nodules and cardiopulmonary coin lesions have been
reported. Ophthalmic dirolarial infections are more common
in the eyelids and periorbital tissues,100 conjunctiva,101 orbit,102
vitreous,103 and anterior chamber, in that order. The most
common clinical presentation is a well-encapsulated nonviable
parasite, although an occasional viable parasite has been detected.
Diagnosis is serologic using a highly specic ELISA test.104
Surgical removal is the mainstay of therapy.
The diagnosis of ascariasis is made by identication of eggs in

feces or, more rarely, larvae in sputum. Occasionally, adult
worms are expelled from the mouth or rectum. Abdominal
radiographs may demonstrate parasites as worm outlines; chest
radiographs may show eeting inltrates (Lfers pneumonia)
owing to migrating larvae. The ELISA test can also be used.89
Prevention
Adequate hygienic and sanitary conditions contribute to prevention of ascariasis. Water should be boiled and uncooked
vegetables avoided in endemic areas.
Treatment
Mebendazole and albendazole inhibit glucose uptake by the
parasite.90,91 Mebendazole is slowly and only slightly absorbed
from the gastrointestinal tract.92 Mebendazole is teratogenic in
rats and should not be given to pregnant women.92 In cases of
massive parasite load, these drugs should be used with caution
because they might promote parasite migration (i.e., biliary
duct or appendix obstruction). Pyrantel pamoate is effective
against Ascaris. It produces spastic paralysis and could lead to
intestinal obstruction in cases of massive infection. In these
cases, piperazine citrate, which produces accid paralysis of the
parasite, should be used. Most of the anthelmintics kill the
adult parasite, not the larvae, so a second course of treatment
is often given 2 weeks after the rst to allow time for the larvae
to complete the pulmonary cycle and mature into adult
parasites.89
EXTRAINTESTINAL NEMATODES
Diffuse Unilateral Subacute Neuroretinitis
Diffuse unilateral subacute neuroretinitis93 is a syndrome caused
by the subretinal migration of the larval or adult form of a
parasite of the class Nematoda. Most reported cases have been
from the southeastern United Stated and the Caribbean. Several
nematodes have been implicated, including Toxocara species
and Ancylostoma caninum. Recent reports have implicated the
raccoon and skunk roundworm Baylisascaris procyonis.9496 The
migration of the parasite causes unilateral damage to the retina,
pigment epithelium, and optic nerve along with vitreal inammation. There is usually severe loss of visual acuity. If the parasite
is seen, photocoagulation is an effective means of treatment.96,97
If no parasite is seen and clinical suspicion is high, thiabendazole98
or ivermectin9 can be used but their value is controversial.99
CHAPTER 14
Diagnosis
Human larial parasites infect an estimated 200 million people

and cause a range of disease manifestations. Adult larial
worms are threadlike, live in the subcutaneous tissues and
lymphatics, and reproduce sexually to produce microlariae, the
rst larval stage. Microlariae are ingested by hematophagous
arthropods, in which they develop into infective larvae that
molt in the vertebrate host and mature into male or female
worms.
Lymphatic Filariasis
Wuchereria bancrofti, Brugia malayi, and Brugia timori are
larial nematodes with a propensity for lymphatic invasion.
W. bancrofti is distributed throughout Africa, Asia, the
Caribbean, Latin America, and Western and South Pacic
Islands. B. malayi and B. timori are found in the Far East.
Infection of the mosquito vector occurs when the insect takes a
blood meal of an infected host. Ocular lariasis by these
organisms is rare. Adult B. malayi worms have been found in
the conjunctiva and probably result from direct inoculation to
the eye rather than migration. Elephantiasis of the eyelid has
been reported. One case of a subretinal worm,105 and a second
of an immature W. bancrofti in the iris,106 represent rare intraocular cases. The nding of living adult worms in lymphatic
vessels is suggestive. A single dose of 100 mg of diethylcarbamazine (DEC) provokes the emergence of microlariae
into the peripheral circulationblood should be drawn 1 h after
the administration of DEC. Treatment consists of a 21-day
regimen of DEC, although infection may recur. Topical 1%
atropine solution has been described as an agent capable of
killing microlariae in the anterior chamber.106
Loiasis
Loiasis is a nematode infection caused by the laria Loa loa.
Distribution
Endemic areas of loiasis are the rain forests of West and Central
Africa.

The vectors, female ies of the genus Chrysops (family
Tabanidae), are infected by ingesting human blood contaminated with the parasitic microlariae. The larvae become
infectious in the arthropod and penetrate the host skin during
the next blood meal. Larvae develop into adult roundworms
(male, 47 cm in length; female, 23 cm) in the subcutaneous
tissues of the host. After mating, gravid females release microlariae, which enter the circulatory system and, after transmission to another y, initiate a new life cycle. The microlariae
exhibit diurnal activity, appearing in the peripheral blood only
from dawn to dusk.
147
MICROBIOLOGY
The disease is often asymptomatic, although transient pruritic

or painful subcutaneous swellings (known as Calabar swellings)
are a classic manifestation of the disease. Adult worms can
sometimes be observed beneath the skin or conjunctiva
(Fig. 14.4).107
Black ies, members of the family Simuliidae (order Diptera),

are the only known vectors for O. volvulus. The ies are found
mostly near fast-owing rivers in tropical and subtropical
regions. Female black ies are blood feeders, and it is during
the blood meal that the y can transmit or receive the infection
from humans. When a black y (15 mm long; black, gray, or
tan) bites an infected person, microlariae in the circulatory
system are ingested along with the blood meal. In the insect
vector, microlariae (300360 59 mm, unsheathed) develop
into infectious larvae and are retransferred to human skin
during the next blood meal. They enter humans via the y
bite wound and develop into adult nematodes within 23
months. Adult worms (females, 2550 cm 0.250.50 mm;
males, 1.94.2 0.130.15 mm) are white or cream-colored,
threadlike roundworms, living in the subcutaneous tissues,
deep fasciae, or joints, commonly in clusters; they may be
encapsulated (onchocercoma) by a host immune response.
The worms reproduce sexually, and new microlariae appear
within a year after primary infection. The adult female can
produce millions of microlariae during her lifetime (15 years).
O. volvulus can be transmitted congenitally from severely infected
mothers, but this is rare. Parasitic nodules are usually concentrated in the area of the original black y bites. African black
ies more frequently bite on the hips and legs; Central and
South American black ies usually bite the head area.
Diagnosis
Denitive diagnosis is made by identication of either microlariae in the blood or adult worms in subcutaneous tissues or
conjunctiva. Blood should be drawn during daylight because of
the diurnal periodicity of microlaremia. Serologic testing for
specic IgG immunoglobulin may be useful in the diagnosis of
L. loa in amicrolaremic cases.108
SECTION 3
Prevention
Loiasis is prevented by protection against y bites (appropriate
clothing, insect repellents).
Treatment
Diethylcarbamazine citrate is the drug of choice in the treatment of loiasis. Adult worms should be surgically removed from
the subconjunctiva.109
Onchocerciasis
Onchocerciasis, or river blindness, is a chronic larial disease
caused by the nematode Onchocerca volvulus. It is one of the
major causes of infectious blindness worldwide.
Distribution
Onchocerciasis is an endemic disease with over 18 million
infected persons worldwide, of whom ~2 million have some
form of visual impairment and ~400 000 suffer from blindness.110 Endemic areas include Equatorial Africa and several
foci in Central America, South America, and the Arabian
peninsula. All age groups are affected. The intensity of infection
increases with host age and reaches a plateau during the second
decade of life. In hyperendemic areas in West Africa, approximately one-third of people over the age of 15 years have microlariae in the anterior chamber of the eye, and half of those
over the age of 40 become blind from the disease.111 Men are
more commonly affected than women because of occupational
exposure.110

Living Onchocerca microlariae cause little adverse reaction in
humans and appear to be undetected by the host immune
system. Damage caused by onchocerciasis is due to dead or
dying microlariae. The pathogenicity varies with the species of
Onchocerca.112 If a large number of microlariae die at the same
time (e.g., after DEC treatment in heavily infected persons), an
inammatory/immune response called the Mazzotti reaction
may result.113 The reaction causes a localized or generalized
skin pruritic rash, fever, lymph node inammation, headache,
nausea, joint and muscle pain, tachycardia, respiratory distress,
and hypotension. Deaths caused by the Mazzotti reaction have
been reported.
In an important new advance, it has been shown that much
of the inammation that occurs upon death of microlariae is
attributable to the release of lipopolysaccharide from the cell
wall of an endosymbiont of the bacterial genus Wolbachia.114
Wolbachia belong to the order of Rickettsiales and are essential
for reproduction of the iaria. Antibiotics that kill the endosymbionts stop embryogenesis in female worms. Tetracyclines,
rifampicin, and chloramphenicol are active against Wolbachia.
Doxycycline (100 mg/day) for 6 weeks blocked embryogenesis
of worms over a period of 18 months, and higher doses
(200 mg/day) were effective for 24 months without severe side
effects. Combined with ivermectin treatment, more than 90%
of the patients were free of microlaridermia for 18 months.114
In cases where Wolbachia elimination is not attempted, the
anterior segment manifestations of ocular onchocerciasis, such
as sclerosing keratitis and iritis, as well as the presence of optic
neuritis and atrophy, are sometimes reversible after ivermectin
therapy.115
Diagnosis
FIGURE 14.4. Loa loa. Note the adult worm in the subconjunctival
space.
Courtesy of Roberto Pineda II, MD, and Susannah Rowe, MD.
Photo by Kit Johnson.
148
Clinical diagnosis Detection of typical subcutaneous nodules

suggests the diagnosis of onchocerciasis, which must be conrmed by histologic examination.116 Detection of intraocular
O. volvulus microlariae is diagnostic for onchocerciasis.111
Serologic tests are nonspecic; blood analysis usually reveals
moderate eosinophilia.
Skin biopsy Skin biopsy is used not only for diagnosis, but also
to assess the intensity of infection (number of microlariae per
milligram of skin).89 Usually, 1 mg of healthy skin is sliced to a
depth of 0.5 mm from several sites (shoulders, buttocks). The
skin snips are placed immediately into 0.5 mL of saline
solution, where they are held for 3 h to allow the microlariae
sufcient time to migrate from the tissue. Detection of a single
microlaria is a denitive diagnosis; a moderately infected
patient has 20100 microlariae per milligram of skin.
sporadic but widespread. Demographic factors, such as socioeconomic status, hygiene practices, and association with dogs,
inuence infection rates.121 Seroprevalence rates of toxocariasis
in children (111 years) in different geographic regions of the
United States range between 4.6% and 7.3% and are higher in
warmer climates.121 The frequency of seropositive titers declines
markedly with increasing age; peak infection occurs at 15 years.
Children with geophagic behavior and who are exposed to dogs
are most likely to develop OLM.122
Prevention
Areas of black y infestation should be avoided because no

prophylactic drug is effective against the infectious larvae.
Personal protection, such as appropriate clothing and insect
repellents, should be used. The Onchocerciasis Control
Program established by the World Health Organization has
been effective in reducing transmission of onchocerciasis in a
700 000-km2 area involving seven countries in Central and
West Africa.110
Dogs and other canines (denitive hosts) are infected by several

routes: ingestion of infectious eggs, ingestion of late-stage larvae
or immature adult worms (during maternal grooming of the
litter), ingestion of larvae in tissues of paratenic hosts (e.g.,
mice), and transplacental or transmammary transmission.
Infection in cats is similar to that in dogs, although there is no
evidence of transplacental infection. The life cycle in puppies
initiates with ingestion of Toxocara eggs (7585 mm, spherical
with a thick shell) that hatch in the stomach or small intestine
of the denitive host and release infectious larvae (20
400 mm).123 The larvae burrow into the intestinal mucosa, enter
the lymphatic and circulatory systems, and migrate to the lung
capillary bed within 35 days. In the lungs, the larvae enter the
bronchioles, trachea, and pharynx and are swallowed to develop
into adult worms (T. canis, 418 cm; T. cati, 312 cm) in the
intestine. Adult worms produce eggs (200 000/day)124 that are
shed in the feces 45 weeks after infection. Eggs are noninfective when shed and require appropriate soil conditions for
development of the infectious larvae.
Transmission to humans may occur by ingestion of eggs from
the soil, contaminated hands, and fomites, or less frequently by
ingestion of the larval stage from undercooked meat. If the host
is large enough (adult dogs and humans), larvae pass through
the pulmonary capillaries and are distributed to somatic tissues
instead of being trapped in the alveoli. Humans are paratenic
hosts, with larvae migrating aimlessly in the tissues for varying
time periods. The larvae reach the eye via the choroidal blood
vessels, where they migrate into the subretinal space or vitreous
cavity.125
Treatment
Additional clinical trials to determine optimum antibiotic
activity for eliminating Wolbachia from the worms and
rendering them sterile, are currently underway. Previously,
ivermectin has been the drug of choice.115,117 It causes a spastic
paralysis of microlariae, thus reducing the side effects of
treatment related to migration of the parasites. It does not affect
adult worms.111 The drugs formerly used in the treatment of
onchocerciasis, suramin and DEC, can cause severe reactions
related directly to the patient load of microlariae and are
not currently recommended. Nodulectomy may be useful to
decrease the adult worm load.
Thelaziasis
Nematode members of the family Spiruroidea, genus Thelazia,
are parasites of birds and mammals and are usually located in
the conjunctiva and lacrimal gland ducts. Adult worms are
cream-colored and measure 0.75 17 mm. Some species
(T. callipaeda, Asia, China, and Korea; T. californiensis, North
America) have been reported in humans. Flies of the genera
Musca and Fannia are the intermediate hosts for this parasite.
Denitive hosts include dogs, cats, horses, sheep, bears, and
deer. In humans, the worms invade the conjunctiva, causing
pain and watery conjunctivitis.118 They can be seen as creamy
white, threadworm masses coiled in the conjunctival sac or
migrating over the cornea. Eyelids and extraocular muscles
can also be compromised. Intraocular penetration does not
occur. Therapy for ocular thelaziasis is surgical removal of
the parasite.
CHAPTER 14

The tissue damage observed in toxocariasis results from
larva migration (mechanical) and immune reaction. Clinical
manifestation of the disease depends on the organ and the
number of invading larvae. Several larvae in the liver may
cause no disease, whereas a single larva in the eye can cause
blindness.
Diagnosis
Toxocariasis
Dogs and cats are the denite hosts for Toxocara canis and
Toxocara cati, which are members of the nematode family
Ascarididae. Toxocariasis in humans (an intermediate host)
is caused predominantly by T. canis, and it is manifested clinically as either visceral larva migrans (VLM) or ocular larva
migrans (OLM).
Distribution
T. canis has a worldwide distribution in dogs and is uniformly
prevalent in North America.119 Pregnant and lactating dogs are
the most important factors in Toxocara infection. In puppies,
intestinal infection rates can reach 100%; in adult dogs, the rate
falls to less than 20%.120 T. cati infection also appears to occur
worldwide in cats, with a prevalence in North America varying
between 24% and 67%.119 Toxocara infection in humans is
Serology ELISA is used for serodiagnosis. Titers may be equal126

but are usually lower127 in patients with ocular infections compared with patients with systemic disease. ELISA titers of 1:32
are indicative of VLM (78% sensitivity, 92% specicity),128 and
titers of 1:8 are indicative of OLM (90% sensitivity, 91% specicity).129 ELISA can also be used on intraocular uids.130 High
titers can be detected in the aqueous humor and the vitreous
when concomitant serum titers are low or absent, suggesting
localized antibody production.131,132 Aqueous humor (especially
when cells are observed at the clinical examination) and vitreous
cytology can demonstrate eosinophils, suggesting a parasitic
infection.133
Blood analysis Patients with VLM may have leukocytosis,
hypereosinophilia, and hypergammaglobulinemia (IgG, IgM, or
IgE); blood ndings are usually normal in patients with OLM.
149
MICROBIOLOGY
SECTION 3
Ocular imaging studies Detection of intraocular calcications

by computed tomography may provide a differential diagnosis
with retinoblastoma, although small retinoblastomas can
remain uncalcied, and cases of toxocariasis with calcium
deposits have been reported.134 Echographic ndings such as a
solid, highly reective peripheral mass; a vitreous band or
membranes extending between the posterior pole and the mass;
and a traction retinal detachment or fold from the posterior pole
to the mass suggest ocular toxocariasis.135
Histopathology In tissue sections, circumscribed granulomatous reactions with neutrophil and eosinophil inltrates
are seen, occasionally with the larvae located in the center of
the reaction (Fig. 14.5). Fibrinoid necrosis may occur in the
central area of recent lesions, whereas older lesions may reveal
brous encapsulation. Giant cells, epithelioid cells, macrophages, and lymphocytes are usually present around degenerating larvae.136
Prevention
Newborn litters and lactating dogs and cats should be
dewormed at regular intervals. Because deworming medication
does not eradicate all somatic larvae, pregnant dogs require
repetitive prophylaxis and deworming with each new litter.
Treatment
Thiabendazole is controversial because the death of the parasite
entices an intense inammatory response.137 Steroids are frequently used to decrease it. Photocoagulation, cryopexy, and
vitrectomy have been employed.

There are no intermediate hosts, and both the adult and larval
stages develop in the same animal. After ingestion of contaminated meat, encysted Trichinella larvae (0.4 0.26 mm) are
freed by gastric digestion of the cyst wall. The larvae develop
into adult worms (females, 23.6 mm 7590 mm, are approximately twice the length of males) in the small intestinal
mucosa. Following copulation, the male dies, and within a week
the viviparous female releases larvae (100160 mm 67 mm),
which enter the mucosal vascular channels and are distributed
throughout the body. Larviposition continues for ~46 weeks.
Only larvae that encyst in skeletal muscles mature and become
infectious. The muscles of the diaphragm, tongue, and eye are
mostly affected. Calcication of cysts begins in 618 months.
The cycle is repeated when the host is eaten by another
carnivore.

Disease severity is directly related to the numbers of larvae
ingested, varying from completely asymptomatic to severe with
neurologic, pulmonary, and cardiovascular complications. In
the intestine, the adult worms cause inammation and mucus
production. Muscle invasion by the larvae can cause myalgia
and weakness. Encysted larvae, localized in extraocular muscles,
cause periorbital inammation with conjunctivitis, hemorrhage,
edema, pain, and photophobia.3 Eosinophilia is frequent.
Diagnosis
Denitive diagnosis is made by direct observation of encysted,
coiled larvae in tissue biopsy specimens. Serologic test results
are positive after the third week of infection.
Trichinosis
Trichinosis is a nematode infection caused by the roundworm
Trichinella spiralis. Humans are infected by eating raw or
improperly cooked meat, especially pork.
Prevention
Trichinosis is prevented by proper cooking of pork.
Treatment
Distribution
Trichinosis is endemic where pork is consumed, especially in
the Western hemisphere and Western Europe. Several cases
have been attributed to the consumption of wild carnivores,
such as bear and wild boar. Between 1982 and 1986, the annual
average number of cases in the United States was 57.138
Mebendazole and thiabendazole are available for the treatment

of trichinosis. Thiabendazole therapy has been associated more
frequently with side effects, such as dizziness, mental changes,
rash, nausea, and StevensJohnson syndrome in children.139
The administration of corticosteroids is indicated for the treatment of the allergic reaction to dead parasites.
Schistosomiasis
Schistosomiasis is an infection caused by three species of
Schistosoma: S. mansoni, S. japonicum, and S. haematobium.
Distribution
S. mansoni is prevalent in Africa, the Middle East, and South
and Central America; S. japonicum in the Far East; and
S. haematobium in the Middle East and Africa.
FIGURE 14.5. Intraocular toxocariasis. Fibrotic mass with many

eosinophils. Toxocara canis larvae within the brotic proliferation.
Massons trichrome stain 250.
Courtesy of Miguel Burnier Jr, MD.
150
The intermediate host of Schistosoma is the snail (Biomphalaria species). Humans are the only denite host and only
signicant disease reservoir. Schistosoma eggs in fresh water
release miracidium larvae that enter the snail and differentiate
into cercariae (nal larval stage). Cercariae pass from the snail
to the water and penetrate the human skin. After penetration,
the cercariae migrate to the lungs and then to the liver as
worms, where they mature and mate. Females of S. mansoni
and S. japonicum lay their eggs in the smallest venules of the
intestinal wall, and the eggs are shed with the feces. Females of
S. haematobium lay their eggs in the smallest vessels of the
vesical plexuses, and the eggs are shed in the urine. The eggs
reach fresh water, and the cycle is repeated again.
The prepatent period in humans (from cercaria penetration

until appearance of eggs in the feces or urine) is ~50 days.140
Local dermatitis after contact with infested water is common
(swimmers itch). In cases of S. mansoni or S. japonicum infection, the acute phase may include abdominal pain, chills, fever,
cough, diarrhea, and eosinophilia; during chronic phases,
hepatosplenomegaly, ascites, and esophageal varices, with
recurrent episodes of hematemesis, can occur. In cases of
S. haematobium infection, dysuria, hematuria, and suprapubic
pain, as well as obstructive uropathy, may occur. Infection of the
eye includes granulomatous choroiditis,141,142 dacryoadenitis,143
and conjunctivitis,144 and lid masses145 in endemic areas.
Diagnosis
Denitive diagnosis is made by detecting the eggs in feces or
urine. Biopsy of the rectal or urinary bladder mucosa is rarely
indicated.
Prevention
Prevention can be accomplished by improving sanitation and
reducing egg contamination in fresh water. Snail control with
molluscicides may be useful in endemic areas.
Treatment
Praziquantel, oxamniquine, metrifonate, and niridazole are
available for the specic treatment of schistosomiasis.47
Tapeworms
Taeniasis and cysticercosis
Tapeworms of the genus Taenia can cause two different human
diseases: taeniasis and cysticercosis. Taeniasis is an intestinal
infection caused by the adult T. solium and T. saginata.
Cysticercosis is a tissue infection caused by the larval form of
T. solium (Cysticercus cellulosae).
Distribution Taeniasis and cysticercosis occur where sanitary
conditions are poor and where raw or undercooked contaminated pork and beef are routinely consumed. Endemic foci of
the disease are South and Central America and Africa.
Morphology, biology, and life cycle Taeniasis is acquired by
ingestion of raw or poorly cooked meat contaminated with the
larval form of the parasite (cysticerci). Taenia larvae attach to
the host intestinal mucosa and develop into adult worms
(39 m) in the intestinal lumen. Terminal gravid segments
of the worm, called proglottids (T. saginata, 20 57 mm;
T. solium, 12 5 mm), are shed in feces and contain
50 000100 000 viable eggs. Eggs (3040 mm) in proglottids are
infectious immediately after shedding. Ingestion of eggs by
intermediate hosts (pigs, cattle, or humans) results in hatching
of the eggs into larvae (5 10 mm, with a scolex) and
penetration through the intestinal wall. The larvae are
transmitted through the lymphatic and circulatory systems,
where they invade various organs and develop into cysticerci
(infectious form). Humans develop cysticercosis via ingestion of
T. solium eggs, either from exogenous sources or from their own
stools. Only larvae of T. solium penetrate the human intestine;
T. saginata does not cause human cysticercosis because the
larvae cannot penetrate the intestinal wall.
Infection of the host Patients with taeniasis are usually
asymptomatic. Patients with cysticercosis may also be asymptomatic, although clinical manifestations of neurocysticercosis
(epilepsy, intracranial hypertension, and mental disturbances),
ophthalmocysticercosis (loss of vision, periorbital pain, scotoma,
and photopsia),146 and subcutaneous and muscular cysticercosis

(subcutaneous nodules) may be noted. In the eye, the cysticercus cyst may be localized in the orbit,147 the subconjunctival
space, or intraocularly in the anterior or posterior chamber.
Larvae can be identied in the subretinal space, where they
cause hemorrhage and edema.148
Diagnosis Taeniasis is diagnosed by isolation and identication of the proglottids in feces. If T. solium proglottids are
identied, additional evaluation for potential cysticercosis is
warranted. Clinical ndings, such as brain calcications, cystic
lesions in the CNS, and demonstration of larvae with scoleces
within the eye, are diagnostic of cysticercosis. Ocular ultrasonography may be an alternative to computed tomography and
magnetic resonance imaging in the evaluation of patients of
suspected intraocular or orbital cysticercosis.149 Indirect hemagglutination and ELISA may be helpful, although false-positive
results can occur.150
Prevention Appropriate sanitation and personal hygiene are
important in the control of fecal contamination of water and
food. Raw or improperly cooked pork should be avoided,
especially in endemic areas.
CHAPTER 14
Treatment Anthelminthic drugs used in the treatment of

taeniasis and cysticercosis include praziquantel (drug of choice),
niclosamide, and paromomycin. Mebendazole and albendazole
are effective against Taenia but not against Cysticercus. In cases
of ocular cysticercosis, surgical removal of cysts is often
necessary.151
ARTHROPODS
DEMODICOSIS
Demodex folliculorum and D. brevis are two species of follicle
mites causing demodicosis in humans. D. folliculorum lives on
hair follicles in the facial region, and D. brevis inhabits sebaceous glands. The disease is extremely common, with infestation rates reaching 97% in endemic areas.152 Demodicosis is
usually a benign infestation, although follicle mites have been
associated with blepharitis.152
MYIASIS
Ophthalmomyiasis refers to the involvement of the ocular
tissues by larvae from ies of the order Diptera.
Distribution
Myiasis is a worldwide disease, occurring more frequently in
warm climates. The prevalence of the different species of
ies varies according to the locale. Dermatobia hominis is
endemic in transequatorial coffee-growing areas of South
America. Chrysomyia bezziana is primarily a cattle parasite in
the Old World. Calliphora vomitoria organisms are present in
decaying animal or vegetable matter worldwide. Ophthalmomyiasis is the infestation that occurs in the ocular or periocular
tissues.

Larvae from several y species can cause ophthalmomyiasis.
These larvae are usually obligatory parasites, requiring host
tissue for completion of their larval stages. Eggs or larvae may
be transported to the eye by the adult y, by a secondary vector
such as a tick or mosquito, or by the patients hands. D. hominis,
C. vomitoria, and Chrysomyia bezziana infection occurs via
151
MICROBIOLOGY
oviposition on periocular tissue. Hypoderma lineatum larvae, a
cattle parasite, penetrate the skin and migrate aimlessly,
causing painful abscesses.

Ocular disease may be external or internal. In external ophthalmomyiasis, lid edema,153 furuncular lesions,154 orbital involvement,155 and even loss of the eye156 can occur (Fig. 14.6).
Internal ophthalmomyiasis is caused predominantly by larvae
of H. lineatum. Subretinal tracks (trails of depigmentation in
the retinal pigment epithelium) are the result of maggot migration in the subretinal spaces and are pathognomonic of internal
ophthalmomyiasis.157 The larvae could migrate into the vitreal
cavity. Visual compromise varies from nonexistent158 to severe
visual loss.159
SECTION 3
Diagnosis and Treatment

Myiasis is diagnosed on the basis of recovery or visualization of
the larvae. In cases of ophthalmomyiasis externa, covering of
the skin lesion with bland medicinal oil or petroleum jelly
forces the larvae to the skin surface, facilitating removal with a
forceps. In cases of ophthalmomyiasis interna, laser photocoagulation of the subretinal larvae160 or extraction by vitrectomy
of the intravitreal larvae has been attempted.
OPHTHALMIA NODOSA
Ophthalmia nodosa is a condition caused by an immune reaction
to caterpillar hairs or other insect matter. Caterpillar hairs are
acquired by direct contact or via airborne transmission. The
hairs induce a granulomatous inammatory response with pain
and foreign body sensation. The most commonly affected tissue
is the conjunctiva, where nodules have been occasionally
reported.161 The caterpillar hairs may penetrate into the deeper
ocular tissues, causing keratitis, iridocyclitis, and even endophthalmitis.162 Ophthalmia nodosa is treated by surgically
removing the caterpillar hair and by topical steroids.
PHTHIRIASIS
Phthiriasis is a lice infestation caused by the arthropod Phthirus
pubis.
152
Distribution
Lice infestation is cosmopolitan; transmission occurs by direct
physical contact with infected persons. The 1540-year-old age
group is more commonly affected. In children, infestation with
P. pubis results from contamination from an adult.163

Phthiriasis is considered a venereal disease. The source of lice
is generally the hair in the pubic area of an affected person. The
lashes become infected by either direct contact or by contact
with contaminated bedding and clothes. Other species of lice,
such as P. humanus capitis (head louse) and P. humanus
humanus (body louse), do not affect the eyelashes. The reason
the lashes are affected by P. pubis seems to be related to the
parasites arm span. There is itching and erythema of the lid
margin. Chronic follicular conjunctivitis is common. The oval
and transparent parasites eggs or nits are glued to the
eyelashes. The adult louse is frequently overlooked because of
its transparency.
Diagnosis
The diagnosis of lice infestation is based on the demonstration
of nits and adult lice in the lashes. Wood-light illumination can
be used to demonstrate the uorescence of the nits.164
Treatment
Physostigmine (Eserine) ointment can be used to suffocate the
parasite.165 Lindane should be used in the pubic area.
RICKETTSIAL INFECTIONS
Rickettsial infection is an acute disease caused by the bacterialike microorganisms of the family Rickettsiaceae. In addition to
Wolbachia mentioned above, three genera are involved:
Rickettsia, Rochalimaea, and Coxiella, with human infections
caused primarily by Rickettsia prowazekii, Rickettsia typhi,
Rickettsia rickettsii, Rickettsia tsutsugamushi, Coxiella burnetii,
and Rochalimaea quintana. Rickettsia can infect a wide number
of hosts, from invertebrates to vertebrates. Rickettsial diseases
in humans can be divided clinically into the typhus group
(epidemic typhus, murine typhus), the spotted fever group (Rocky
Mountain spotted fever, boutonneuse fever, rickettsialpox), and
FIGURE 14.6. Ophthalmomyasis externa. A 94-year-old woman from Cundinamarca (Colombia) with altered mental status found with massive
orbital infestation by Dermatobia hominis. Note the marked lid edema and distorted anterior segment (a). The larvae had destroyed all the
intraocular contents (b).
Courtesy of Pedro I Navarro, MD.
DISTRIBUTION
Key Features: Rickettsial Infection
Small, Gram-negative coccobacillary bacteria

Replicate intracellularly
Use host ATP
Athropod vectors
Rickettsial infections occur worldwide. Improved treatment and

prevention methods have decreased the incidence of rickettsioses, but they have not been completely eliminated.
MORPHOLOGY, BIOLOGY, AND LIFE CYCLE

Rickettsia are pleomorphic, Gram-negative organisms (0.20.5 mm
0.82 mm) that resemble bacteria in their structural and
chemical characteristics but are distinct organisms, because
several species have an obligate intracellular nature. They
multiply by binary ssion in the cytoplasm of infected cells or,
as with the spotted fever group organisms, replication can also
occur in the cell nucleus. R. prowazekii replicates until the cell
lyses, whereas R. rickettsii does not cause cell lysis and leaves
the host cell early in the course of infection to infect other cells.
Disease transmission is via arthropods.166
Lice (Pediculus humanus) are the vectors of the epidemic
typhus caused by R. prowazekii. The organisms invade the
louses intestinal epithelial cells and multiply, causing cell lysis.
The louse does not survive more than 10 days after the primary
infection, and during this period it sheds rickettsiae in its feces.
Contaminated louse feces are deposited on the skin during
insect blood meals, and the rickettsia gains entrance into the
body via wounded or scratched skin. Humans are an important
reservoir host for epidemic typhus.
Ticks (several Dermacentor species) are the vectors of the
Rocky Mountain spotted fever caused by R. rickettsii. The
vector is contaminated by feeding on infected animals (e.g.,
rodents), with rickettsiae remaining in the arthropod salivary
glands. Humans are only accidentally infected. R. rickettsii are
not pathogenic for the ticks; infection is maintained among
ticks by transovarial transmission.
Several species of Leptotrombidium (mites) are the vectors of
the scrub, or chigger-borne, typhus, caused by R. tsutsugamushi.
Adult mites and larvae (chiggers) are infected by feeding on
contaminated vertebrates (e.g., mice). Rickettsiae are located in
the arthropod salivary glands and are inoculated into the host
during the blood meal. R. tsutsugamushi is not harmful to the
mites; infection is maintained among mites by transovarial
passage. The mites function as both reservoirs and vectors of
the disease. Because R. tsutsugamushi has strain variations,
some patients may experience a second attack of scrub typhus.
Lice (P. humanus) are the vectors of the trench fever caused
by R. quintana. The body louse acquires and passes the
infection by feeding on a rickettsemic human. Organisms grow
extracellularly in the louse intestinal lumen; humans are
contaminated through louse feces deposited on the skin.
Humans are reservoirs for trench fever. Transovarial transmission of R. quintana among lice has not been observed.
Fleas (Xenopsylla cheopis) are the vectors of the murine
typhus caused by R. typhi. Humans are accidentally infected.
Organisms proliferate in the ea intestinal cells, and the disease
is transmitted by contaminated ea feces deposited on the skin.
Fleas do not transmit R. typhi to offspring transovarially.167
Mites (Allodermanyssus sanguineus) are the vectors of

the rickettsialpox caused by R. akari. Humans are only accidentally infected. The mite also transmits the infection
transovarially.
Q fever is caused by C. burnetii. Ticks transmit the infection
to domestic animals that shed the rickettsia in milk, urine, feces,
and placental products. C. burnetii is highly resistant to extremes
of temperature and desiccation. Humans and other animals are
infected by inhalation or mucosal contact with dust containing
the rickettsiae. In ticks, infection with one species may prevent
subsequent infection with other rickettsial species.168
INFECTION OF THE HOST

In humans, rickettsiae multiply in endothelial cells of small
blood vessels, causing endothelial proliferation and perivascular
inltration, subsequent extravasation of uid with edema, and
hypotension. If untreated, the disease can progress to gangrene
and disseminated intravascular coagulation. Formation of a
typhus nodule or glial nodule (a perivascular aggregation of
mononuclear cells such as lymphocytes and macrophages) in
the CNS is characteristic of the disease.166,169 Skin and several
other organ tissues (kidney, heart, lung) can be involved,
causing skin rash, encephalitis, and renal and liver failure, and
may lead ultimately to death of the host. Rickettsial infection
may induce resistance to reinfection or, in contrast, persistent
lymphoid tissue disease as in Q fever and recrudescent epidemic
typhus. Table 14.2 summarizes the epidemiology and clinical
ndings of some human rickettsial diseases.
CHAPTER 14
other rickettsial diseases (scrub typhus or chigger-borne typhus,

Q fever, trench fever).
CLINICAL FINDINGS
The clinical spectrum of rickettsial disease varies widely
according to the organism involved and the host response.
Fever, rash, and history of arthropod exposure suggest the
disease, although these signs are not always present.170 Other
signs, including prostration, nausea, vomiting, abdominal and
back pain, myalgia, arthralgia, cough, photophobia, and
conjunctivitis, may be present. A primary cutaneous lesion
(eschar) may be observed at the site of the insect bite or
attachment. In epidemic typhus, a recrudescent mild form of
the disease, called BrillZinsser disease, can occur. Classic Q
fever presents as atypical pneumonia or with inuenza-like
symptoms. Ocular ndings in all rickettsial diseases may
include sore, red eyes with conjunctival papillae, chemosis,
and petechiae; iritis, retinitis (edema, hemorrhage, exudate);
venous engorgement; arteriole occlusion; and optic nerve
edema.171
DIAGNOSIS
Demonstration of rising antibody titers to rickettsial antigens
using paired acute and convalescent sera is the most widely
used method of clinical diagnosis of rickettsial infection. A
fourfold or higher rise in titer suggests acute disease. Serologic
methods include indirect immunouorescent antibody, complement xation, indirect hemagglutination, and ELISA. The
Weil-Felix reaction is an agglutination test using Proteus
mirabilis strains OX19, OX2, or OXk with antigens similar to
those of Rickettsia. The Weil-Felix reaction is not completely
reliable, and rickettsialpox and Q fever are not associated with
Weil-Felix antibody rises.
Rickettsiae stain poorly with Grams stain but can be
visualized using Giemsa or Macchiavellos stain. Culture using
enriched blood-agar media can be used for recovery of
R. quintana. All other rickettsiae require living cells (embryonated eggs or other tissue culture systems) for culture.
153
MICROBIOLOGY
SECTION 3
TABLE 14.2. Epidemiology and Clinical Characteristics of Rickettsial Diseases

Mammalian
Geographic
Incubation
Organism
Transmission
Reservoir
Distribution
Disease
(Days)
Clinical Signs*
Rickettsia
prowazekii
Louse feces
Humans
North and South

America, Africa,
Asia
Epidemic typhus
523
Generalized
maculopapular
rash; central
nervous system
involvement,
myocarditis, renal
insufficiency; no
eschar; may be
recrudescent
Rickettsia
typhi
Flea feces
Rodents
Worldwide
Murine typhus
415
Generalized
maculopapular
rash; no eschar
Rickettsia
rickettsii
Tick bite, dogs
Rodents
Western hemisphere
Rocky Mountain
spotted fever
214
Maculopapular
(petechial) rash
on extremities
and later on
trunk; eschar
Rickettsia
tsutsugamushi
Mite bite
Rodents
Asia
Scrub typhus
812
Maculopapular rash
on trunk
spreading to
palms and
soles; eschar
Coxiella burnetii
Inhalation, goats
Cattle, sheep
Worldwide
Q fever
839
Interstitial
pneumonia; no
eschar; rare rash;
chronic form:
hepatitis and
endocarditis
Rickettsia akari
Mite bite
Mice
USA, former USSR,

Korea
Rickettsialpox
1024
Mild condition;
vesicular lesions
on initial papular
rash; eschar
Rochalimaea
quintana
Louse feces
Humans
Europe and Africa
Trench fever
830
Splenomegaly;
macular rash
*All patients usually present with high fever and headache that may be accompanied by prostration, myalgia, arthralgia, and conjunctivitis.
PREVENTION
Personal protection against vector contact (protective clothing)
and use of insect repellents in endemic areas are preventive
measures. Lice infestation can be avoided by frequent changes
of clothing or by application of insecticides. Forceps and hand
protection while removing ticks are recommended because both
tissues and uids from crushed ticks are contaminated. Vector
and reservoir control may be indicated in endemic areas.
Milkborne transmission, observed in Q fever, can be prevented
by pasteurization. Chemoprophylaxis is not recommended.170
Effective vaccines for the major rickettsial infections (e.g., Rocky

Mountain spotted fever) have been developed but are not used
frequently168 because rickettsial diseases, if promptly recognized
and treated, are no longer lethal.169
TREATMENT
Tetracyclines are preferred drugs in the treatment of rickettsiosis. Chloramphenicol is also effective.171
REFERENCES
154
1. Noble ER, Noble GA: Parasitology. The

biology of animal parasites. Philadelphia,
PA: Lea & Febiger; 1982.
2. Markell EK, Voge M, John DT: Medical
parasitology. 7th edn. Philadelphia, PA: WB
Saunders; 1992.
3. Kean BH, Sun T, Ellsworth RM: Color
atlas/text of ophthalmic parasitology. New
York: Igaku-Shoin; 1991.
4. Borst P, Cross GAM: Molecular basis for
trypanosome antigenic variation. Cell 1982;
29:291303.
5. Wyler DJ: Malaria Resurgence,

resistance, and research (second of two
parts). N Engl J Med 1983; 308:934940.
6. Aitken D, Hay J, Kinnear FB: Amebic
keratitis in a wearer of disposable contact
lenses due to a mixed Vahlkampa and
Hartmannella infection. Ophthalmology
1996; 103:485494.
7. Khan NA: Acanthamoeba: biology and
increasing importance in human health.
FEMS Microbiol Rev 2006;
30:564595.
8. Mazur T, Hadas E, Iwanicka I: The

duration of cyst stage and the viability
and virulence of Acanthamoeba
isolates. Trop Med Parasitol 1996;
46:106108.
9. Chynn EW, Lopez MA, Pavan-Langston D,
Talamo JH: Acanthamoeba keratitis:
contact lens and noncontact lens
characteristics. Ophthalmology 1995;
102:13691373.
10. Talamo JH, Larkin DS: Bilateral
Acanthamoeba keratitis and gas-permeable
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
contact lenses [letter]. Am J Ophthalmol

1993; 116:651652.
Jones DB, Visvesvara GS, Robinson NM:
Acanthamoeba polyphaga keratitis and
Acanthamoeba uveitis associated with fatal
meningoencephalitis. Trans Ophthal Soc
UK 1975; 95:221232.
Naginton J, Watson PG, Playfair TJ:
Amoebic infection of the eye. Lancet 1974;
2:15371540.
Stehr-Green JK, Bailey TM, Visvesvara G:
The epidemiology of Acanthamoeba
keratitis in the United States. Am J
Ophthalmol 1989; 107:331336.
Mathers WD, Sutphin JE, Folberg R:
Outbreak of keratitis presumed to be
caused by Acanthamoeba. Am J
Ophthalmol 1996; 121:129142.
Ma P, Visvesvara GS, Martinez AJ:
Naegleria and Acanthamoeba infection:
review. Rev Infect Dis 1990; 12:490513.
Murakawa GJ, McCalmont T, Altman J:
Disseminated acanthamoebiasis in patients
with AIDS: a report of ve cases and a
review of the literature. Arch Dermatol
1995; 131:12911296.
Moore MB, McCulley JP, Newton C:
Acanthamoeba keratitis: a growing problem
in soft and hard contact lens wearers.
Gorlin AI, Gabriel MM, Wilson LA,
Ahearn DG: Effect of adhered bacteria
on the binding of Acanthamoeba to
hydrogel lenses. Arch Ophthalmol 1996;
114:576580.
Garner A: Pathogenesis of acanthamoebic
keratitis: hypothesis based on a histological
analysis of 30 cases. Br J Ophthalmol
1993; 77:366370.
Blackman HJ, Rao NA, Lemp MA:
Acanthamoeba keratitis successfully
treated with penetrating keratoplasty:
suggested immunogenic mechanisms of
action. Cornea 1984; 3:125130.
Mitra MM, Alizadeh H, Gerard RD,
Niederkorn JY: Characterization of a
plasminogen activator produced by
Acanthamoeba castellani. Mol Biochem
Parasitol 1995; 73:157164.
Mitro K, Bhagavathiammai A, Zhou OM:
Partial characterization of the proteolytic
secretions of Acanthamoeba polyphaga.
Exp Parasitol 1994; 78:377385.
Cavanagh HD, Petroll WM, Alizadeh H:
Clinical and diagnostic use of in vivo
confocal microscopy in patients with
corneal disease. Ophthalmology 1993;
100:14441454.
Cavanagh HD, McCulley JP: In vivo
confocal microscopy and Acanthamoeba
keratitis [editorial]. Am J Ophthalmol 1996;
121:207208.
Tay-Kearney ML, McGhee CN, Crawford
GJ, Trown K: Acanthamoeba keratitis: a
masquerade of presentation in six cases.
Aust N Z J Ophthalmol 1993; 21:237245.
Wilhelmus KR, Osato MS, Font RL: Rapid
diagnosis of Acanthamoeba keratitis using
calcouor white. Arch Ophthalmol 1986;
104:13091312.
Epstein RJ, Wilson LA, Visvesvara GS:
Rapid diagnosis of Acanthamoeba keratitis
from corneal scrapings using indirect
uorescent antibody staining. Arch
Ophthalmol 1986; 104:13181321.
Matias R, Schottelius J, Raddatz CF,
Michel R: Species identication and
characterization of an Acanthamoeba strain
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
from human cornea. Parasitol Res 1991;

77:469474.
Kilvington S, Beeching JR, White DG:
Differentiation of Acanthamoeba strains
from infected corneas and the environment
by using restriction endonuclease digestion
of whole-cell DNA. J Clin Microbiol 1991;
29:310314.
Gradus MS, Koenig SB, Hyndiuk RA:
Filter-culture technique using amoeba
saline transport medium for the noninvasive
diagnosis of Acanthamoeba keratitis. Am J
Clin Pathol 1989; 92:682685.
Mathers WD, Stevens GJ, Rodriguez M:
Immunopathology and electron microscopy
of Acanthamoeba keratitis. Am J
Ophthalmol 1987; 103:626635.
Ludwig IH, Meisler DM, Rutherford I:
Susceptibility of Acanthamoeba to soft
contact lens disinfection systems. Invest
Silvany KE, Dougherty JM, McCulley JP:
The effect of currently available contact
lens disinfection systems on
Acanthamoeba castellani and
Acanthamoeba polyphaga. Ophthalmology
1990; 97:286290.
Larkin DFP, Kilvington S, Dart JKG:
Treatment of Acanthamoeba keratitis with
polyhexamethylene biguanide.
Hay J, Kirkness CM, Seal DV, Wright P:
Drug resistance and Acanthamoeba
keratitis: the quest for alternative
antiprotozoal chemotherapy. Eye 1994;
8:555563.
Brasseur G, Favennec L, Perrine D:
Successful treatment of Acanthamoeba
keratitis by hexamidine. Cornea 1994;
13:459462.
Driebe WT, Stern GA, Epstein RJ:
Acanthamoeba keratitis: potential role for
topical clotrimazole in combination
chemotherappy. Arch Ophthalmol 1988;
106:11961201.
Ishibashi Y, Matsumoto Y, Dabata T: Oral
itraconazole and topical miconazole with
debridement for Acanthamoeba keratitis.
Am J Ophthalmol 1990; 109:121126.
Mathers W: Use of higher medication
concentrations in the treatment of
Acanthamoeba keratitis. Arch Ophthalmol
2006; 124:923.
John T, Lin J, Sahm D, Rockey JH: Effects
of corticosteroids in experimental
Acanthamoeba keratitis [abstract]. Rev
Infect Dis 1991; 13(Suppl):440442.
Horsburgh B, Hirst LW, Carey T: Steroid
sensitive Acanthamoeba keratitis. Aust N Z
J Ophthalmol 1991; 19:349350.
Ravinovitch T, Weissman SS, Ostter HB:
Acanthamoeba keratitis: clinical signs and
analysis of outcome [abstract]. Rev Infect
Dis 1991; 13(Suppl):427.
Alizadeh H, He Y, McCulley JP: Successful
immunization against Acanthamoeba
keratitis in a pig model. Cornea 1995;
14:180183.
Lam S: Keratitis caused by leishmaniasis or
trypanosomiasis. Ophthalmol Clin North
Am 1995; 7:635.
Schmidt GD, Roberts LS: Foundations of
parasitology. 3rd edn. St Louis: Times
Mirror/Mosby College Publishing; 1985.
Malchiodi EL, Chiaramonth MG, Taranto N:
Cross-reactivity studies and differential
serodiagnosis of human infections caused
by Tripanosoma cruzi and Leishmania spp:
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
use of immunoblotting and ELISA with a

puried antigen (Ag163B6). Clin Exp
Immunol 1994; 97:417423.
Pavan-Langston D, Dunkel EC: Handbook
of ocular drug therapy and ocular side
effects of systemic drugs. Boston: Little,
Brown; 1991.
Roizenblatt J: Interstitial keratitis caused by
American (mucocutaneous) leishmaniasis.
Martinez S, Marr JJ: Allopurinol in the
treatment of American cutaneous
leishmaniasis. N Engl J Med 1992;
326:741744.
Saenz RE: Treatment of American
cutaneous leishmaniasis with orally
administered allopurinol riboside. J Infect
Dis 1989; 160:153158.
Saenz RE, Paz H, Berman JD: Efcacy of
ketoconazole against Leishmania
braziliensis panamensis cutaneous
leishmaniasis. Am J Med 1990; 89:147155.
Powell RD: Malaria and babesiosis. In:
Goldsmith R, Heyneman D, eds. Tropical
medicine and parasitology. Norwalk, CT:
Appleton & Lange; 1989:303.
Hidayat AA, Nalbandian RM, Sammons
DW: The diagnostic histopathologic
features of ocular malaria. Ophthalmology
1993; 100:11831186.
Biswas J, Fogla R, Srinivasan P, et al:
Ocular malaria: a clinical and
histopathologic study. Ophthalmology
1996; 103:14711475.
Lewallen S, Taylor TE, Molyneux ME:
Ocular fundus ndings in Malawian children
with malaria. Ophthalmology 1993;
100:857861.
Randall G, Seidel JS: Malaria. Pediatr Clin
North Am 1985; 32:893916.
Valero MV, Amador LR, Galindo C:
Vaccination with SPf66, a chemically
synthesized vaccine, against Plasmodium
falciparum malaria in Colombia. Lancet
1993; 341:705710.
Nosten F, Luxemburger C, Kyle DE:
Randomized double-blind placebocontrolled trial of SPf66 malaria vaccine in
children in northwestern Thailand. Shoklo
SPf66 Malaria Vaccine Trial Group. Lancet
1996; 348:701707.
WHO Parasitic Diseases Surveillance:
Antibody to Encephalitozoon cuniculi in
man. WHO Weekly Epidem Rec 58:3032,
1983.
Davis RM, Font RL, Keisler MS, Shadduck
JA: Corneal microsporidiosis: a case report
including ultrastructural observations.
Weiss LM: And now microsporidiosis. Ann
Intern Med 1995; 123:954956.
Ashton N, Wirasinha PA:
Encephalitozoonosis (nosematosis) of the
cornea. Br J Ophthalmol 1973; 57:669674.
Pinnolis M, Egbert PR, Font RL:
Nosematosis of the cornea: case report,
including electron microscopy studies. Arch
Ophthalmol 1981; 99:10441047.
Friedberg DN, Stenson SM, Orenstein JM,
et al: Microsporidial keratoconjunctivitis in
acquired immunodeciency syndrome.
Yee RW, Tio FO, Martinez JA: Resolution of
microsporidial epithelial keratopathy in a
patient with AIDS. Ophthalmology 1991;
98:196201.
Diesenhouses MC, Wilson LA, Corrent GF:
Treatment of microsporidial
CHAPTER 14
155
MICROBIOLOGY
67.
68.
69.
SECTION 3
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
156
keratoconjunctivitis with topical fumagillin.

Am J Ophthalmol 1993; 115:293298.
Molina J, Oksenhendler E, Beauvaia B,
et al: Disseminated microsporidiosis due to
Septata intestinalis in patients with AIDS:
clinical features and response to
albendazole therapy. J Infect Dis 1994;
171:245249.
Frenkel JK: Transmission of toxoplasmosis
and the role of immunity in limiting
transmission and illness. J Am Vet Med
Assoc 1990; 196:233240.
Daffos F, Forestier F, Capella-Pavlovsky M:
Prenatal management of 746 pregnancies
at risk for congenital toxoplasmosis. N Engl
J Med 1988; 318:271275.
Roberts T, Frenkel JK: Estimating income
losses and other preventable costs caused
by congenital toxoplasmosis in people in
the United States. J Am Vet Med Assoc
1990; 196:249256.
Henderly DE, Genstler AJ, Smith RE:
Changing patterns of uveitis. Am J
Ophthalmol 1987; 103:131136.
Perkins ES: Ocular toxoplasmosis. Br J
Holland GN, Engstrom REJ, Glasgow BJ:
Ocular toxoplasmosis in patients with the
acquired immunodeciency syndrome. Am
J Ophthalmol 1988; 106:653667.
Dutton GN, McMenamin PG, Flay J: The
ultrastructural pathology of congenital
murine toxoplasmic retinochoroiditis. Part
II. The morphology of the inammatory
changes. Exp Eye Res 1986; 43:545560.
Fish RH, Hoskins JC, Kline LB: Toxoplasma
neuroretinitis. Ophthalmology 1993;
100:11771182.
Moorthy RS, Smith RE, Rao NA: Progressive
ocular toxoplasmosis in patients with
acquired immunodeciency syndrome. Am J
Ophthalmol 1993; 115:742747.
Johnson MW, Greven CM, Jaffe GJ, et al:
Atypical, severe toxoplasmic
retinochoroiditis in elderly patients.
Silveira C, Belfort R Jr, Burnier M Jr:
Acquired toxoplasmic infection as the
cause of toxoplasmic retinochoroiditis in
families. Am J Ophthalmol 1988;
106:362364.
Buzoni-Gatel D, Schulthess J, Menard LC,
Kasper LH: Mucosal defences against
orally acquired protozoan parasites,
emphasis on Toxoplasma gondii infections.
Cell Microbiol 2006; 8:535-544.
Desmonts G: Denitive serological
diagnosis of ocular toxoplasmosis. Arch
Ophthalmol 1966; 76:839851.
Rollins DF, Tabbara KF, OConnor GR:
Detection of toxoplasmal antigen and
antibody in ocular uids in experimental
ocular toxoplasmosis. Arch Ophthalmol
1983; 101:455457.
Norose K, Tokushima T, Yano A:
Quantitative polymerase chain reaction in
diagnosing ocular toxoplasmosis. Am J
Ophthalmol 1996; 121:441442.
Garweg J, Boehnke M, Koerner F:
Restricted applicability of the polymerase
chain reaction for the diagnosis of ocular
toxoplasmosis. Ger J Ophthalmol 1996;
5:104108.
Rao NA, Font RL: Toxoplasmic
retinochoroiditis. Arch Ophthalmol 1977;
95:273277.
Dutton GN, Hay J, Flair DM:
Clinicopathological features of a congenital
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98.
99.
100.
101.
102.
103.
104.
105.
106.
murine model of ocular toxoplasmosis.

Graefes Arch Clin Exp Ophthalmol 1986;
224:256264.
Colebunders R, Mathis R: Ocular
toxoplasmosis treated with pyrimetamine.
Araujo FG, Remington JS: Effect of
clindamycin on acute and chronic
toxoplasmosis in mice. Antimicrob Agents
Chemother 1974; 5:647651.
Curtale F, Pokhrel RP, Tilden RL, Higashi G:
Intestinal helminths and xerophthalmia in
Nepal: a case-control study. J Trop Pediatr
1995; 41:334337.
Markell EK: Intestinal nematode infections.
Pediatr Clin North Am 1985; 32:971986.
Keystone JS, Murdoch JK: Mebendazole.
Ann Intern Med 1979; 91:582586.
Bethony J, Brooker S, Albonico M, et al:
Soil-transmitted helminth infections:
ascariasis, trichuriasis, and hookworm.
Lancet 2006; 367:1521-1532.
Burd EM: Antiparasitic agents. In: Tabbara
KF, Hyndiuk RA, eds. Infections of the Eye.
2nd edn. Boston: Little, Brown; 1996:281.
Gass JDM, Gilbert WRJ, Guerry RK, Scelfo
R: Diffuse unilateral subacute neuroretinitis.
Kazacos KR, Vestre WA, Kazacos EA,
Raymond LA: Diffuse unilateral subacute
neuroretinitis syndrome: probable cause.
Kazacos KR, Raymond LA, Kazacos EA,
Vestre WA: The raccoon ascarid: a probable
cause of human ocular larva migrans.
Goldberg MA, Kazacos KR, Boyce WM,
et al: Diffuse unilateral subacute
neuroretinitis: morphometric, serologic, and
epidemiologic support for Baylisascaris as
a causative agent. Ophthalmology 1993;
100:16951701.
Raymond LA, Gutierrez Y, Strong LE, et al:
Living retinal nematode (larial-like)
destroyed with photocoagulation.
Gass JDM, Callanan DG, Bowman CB: Oral
therapy in diffuse unilateral subacute
neuroretinitis. Arch Ophthalmol 1992;
110:675680.
Callanan D, Davis JL, Cohen SM, et al: The
use of ivermectin in diffuse unilateral
subacute neuroretinitis. Ophthalmology
1993; 100(Suppl):114.
Font RL, Neae RC, Perry HD:
Subcutaneous dirolariasis of the eyelid
and ocular adnexa: report of six cases.
Orsoni JG, Coggiola G, Minazzi P: Filaria
conjunctivae. Ophthalmologica 1985;
190:243246.
Brumback GF, Marrison HM, Weatherly NF:
Orbital infection with Dirolaria. South Med
J 1968; 61:188192.
Moorhouse DE: Dirolaria immitis: a cause
of human intraocular infection. Infection
1978; 6:192193.
Sun S, Sugane K: Immunodiagnosis of
human dirolariasis by enzyme-linked
immunosorbent assay using recombinant
DNA-derived fusion protein. J Helminthol
1992; 66:220226.
Gupta A, Agarwal A, Dogra MR: Retinal
involvement in Wuchereria bancrofti lariasis.
Acta Ophthalmol 1992; 70:832835.
Tham MH, Hall IB: Impacted microlaria in
the lens capsule. Br J Ophthalmol 1971;
55:484486.
107. Lee BYP, McMillan R: Loa loa: ocular

lariasis in an African student in Missouri.
108. Akue JP, Hommel M, Devaney E: Markers
of Loa loa infection in permanent residents
of a loiasis endemic area of Gabon. Trans R
Soc Trop Med Hyg 1996; 90:115118.
109. Gendelman D, Blumberg R, Sadun A:
Ocular Loa loa with cryoprobe extraction of
subconjunctival worm. Ophthalmology
1984; 91:300303.
110. Ament CS, Young LH: Ocular
manifestations of helminthic infections:
onchocersiasis, cysticercosis, toxocariasis,
and diffuse unilateral subacute
neuroretinitis. Int Ophthalmol Clin 2006;
46:110.
111. Taylor HR: Onchocerciasis. Int Ophthalmol
1990; 14:189194.
112. Zimmerman P, Dadzie K, DeSole G:
Onchocerca volvulus DNA probe
classication correlates with epidemiologic
patterns of blindness. J Infect Dis 1992;
165:964968.
113. Gleich GJ, Ottesen EA, Leiferman KM:
Eosinophils and human disease. Int Arch
Allergy Appl Immunol 1989; 88:5962.
114. Hoerauf A, Bttner DW, Adjei O, Pearlman
E: Science, medicine, and the future:
onchocerciasis. BMJ 2003; 326:207210.
115. Mabey D, Whitworth JA, Eckstein M: The
effects of multiple doses of ivermectin on
onchocerciasis: a six-year follow up.
Ophthalmology 1996; 103:10011008.
116. Albiez EJ, Buttner DW, Duke BOL:
Diagnosis and extirpation of nodules in
human onchocerciasis. Trop Med Parasitol
1988; 39(Suppl 4):331346.
117. Aziz MA, Diallo S, Diop IM: Efcacy and
tolerance of ivermectin in human
onchocerciasis. Lancet 1982; 2:171173.
118. Kirschner BI, Dunn JP, Ostler HB:
Conjunctivitis caused by Thelazia
californiensis. Am J Ophthalmol 1990;
110:573574.
119. Walker MD, Zunt JR: Neuroparasitic
infections: nematodes. Semin Neurol 2005;
25:252-261.
120. Glickman LT, Schantz PM, Cypess RH:
Canine and human toxocariasis: review of
transmission, pathogenesis, and clinical
disease. J Am Vet Med Assoc 1979;
175:12651269.
121. Herrmann N, Glickman LT, Schantz PM:
Seroprevalence of zoonotic toxocariasis in
the United States: 1971-1973. Am J
Epidemiol 1985; 122:890896.
122. Schantz PM, Weis PE, Pollard ZF: Risk
factors for toxocaral ocular larva migrans: a
case-control study. Am J Public Health
1980; 70:12691272.
123. Scothorn MW, Koutz FR, Groves HF:
Prenatal Toxocara canis infection in pups. J
Am Vet Med Assoc 1965; 146:4548.
124. Schantz PM, Glickman LT: Toxocaral
visceral larva migrans. N Engl J Med 1978;
298:436439.
125. Kielar RA: Toxocara canis endophthalmitis
with low ELISA titer. Ann Ophthalmol 1983;
15:447449.
126. Schantz PM, Meyer D, Glickman LT:
Clinical, serologic, and epidemiologic
characteristics of ocular toxocariasis. Am J
Trop Med Hyg 1979; 28:2428.
127. Pollard ZF: Long-term follow-up in patients
with ocular toxocariasis as measured by
ELISA titers. Ann Ophthalmol 1987;
19:167169.
128. Glickman LT, Schantz P, Dombrsoke R:

Evaluation of serodiagnostic tests for
visceral larva migrans. Am J Trop Med Hyg
1978; 27:492498.
129. Pollard ZF, Jarret WH, Hagler WS: ELISA
for diagnosis of ocular toxocariasis.
130. Benitez del Castillo JM, Herreros G, et al:
Bilateral ocular toxocariasis demonstrated
by aqueous humor enzyme-linked
immunosorbent assay. Am J Ophthalmol
1995; 119:514516.
131. Biglan AW, Glickman LT, Lobes LAJ: Serum
and vitreous Toxocara antibody in
nematode endophthalmitis. Am J
Ophthalmol 1979; 88:898901.
132. Despommier D: Toxocariasis: clinical
aspects, epidemiology, medical ecology,
and molecular aspects. Clin Microbiol Rev
2003; 16:265272.
133. Shields JA, Lerner HA, Felberb NT:
Aqueous cytology and enzymes in
nematode endophthalmitis. Am J
Ophthalmol 1977; 84:319322.
134. Howard GM, Ellsworth RM: Differential
diagnosis of retinoblastoma: a statistical
survey of 500 children. 1. Relative
frequency of the lesions which simulate
retinoblastoma. Am J Ophthalmol 1965;
60:610618.
135. Wan WL, Cano MR, Pince KJ: Echographic
characteristics of ocular toxocariasis.
136. Dent JH, Nichols RL, Beaver PC: Visceral
larva migrans: with a case report. Am J
Pathol 1956; 32:777803.
137. Sturchler D, Schubarth P, Gualzata M,
et al: Thiabendazole vs albendazole in
treatment of toxocariasis: a clinical trial.
Ann Trop Med Parasitol 1989; 83:473478.
138. Bailey TM, Schantz PM: Trends in the
incidence and transmission patterns of
trichinosis in humans in the United States:
comparisons of the periods 19751981 and
19821986. Rev Infect Dis 1990; 12:511.
139. Frierson JG: Trichinosis. In: Goldsmith R,
Heyneman D, eds. Tropical medicine and
parasitology. Norwalk, CT: Appleton &
Lange; 1989:423.
140. Cline BL: Schistosomiasis mansoni. In:
Goldsmith R, Heyneman D, eds. Tropical
141.
142.
143.
144.
145.
146.
147.
148.
149.
150.
151.
152.
153.
154.
155.
medicine and parasitology. Norwalk, CT:

Appleton & Lange; 1989:434.
Orece F, Simal CJ, Pittella JEH:
Schistosomotic choroiditis. I. Fundoscopic
changes and differential diagnosis.
Pittella JEH, Orece F: Schistosomotic
choroiditis. II. Report of rst case.
Jakobiec FA, Gess L, Zimmerman LE:
Granulomatous dacryoadenitis caused by
Schistosoma haematobium. Arch
Ophthalmol 1977; 95:278280.
Welsh NH: Bilharzial conjunctivitis. Am J
Ophthalmol 1968; 66:933938.
Kabo AM, Warter A: A propos of 1 case of
ophthalmologic manifestations of
bilharziasis. Bull Soc Pathol Exot 1993;
86:174175.
Topilow HW, Yimoyines DJ, Freeman HM:
Bilateral multifocal intraocular
cysticercosis. Ophthalmology 1981;
88:11661172.
Stewart CR, Salmon JF, Murray AD, Sperry
C: Cysticercosis as a cause of severe
medial rectus muscle myositis. Am J
Ophthalmol 1993; 116:510511.
Kruger-Leite E, Jalkh AE, Quiroz H:
Intraocular cysticercosis. Am J Ophthalmol
1985; 99:252257.
Atul K, Kumar TH, Mallika G, Sandip M:
Socio-demographic trends in ocular
cysticercosis. Acta Ophthalmol Scand
1995; 73:438441.
Brown WJ, Voge M: Cysticercosis: a
modern day plague. Pediatr Clin North Am
1985; 32:953969.
Lim WK, Chee SP: Nonsurgical
management of subretinal cysticercosis.
Retina 2004; 24:469471.
English FP, Nutting WB: Demodicosis of
ophthalmic concern. Am J Ophthalmol
1981; 91:362372.
Wilhelmus KR: Myiasis palpebrarum [letter].
Am J Ophthalmol 1986; 101:496498.
Savino DF, Margo CE, McCoy ED, Friedl
FE: Dermal myiasis of the eyelid.
Kersten RC, Shoukrey NM, Tabbara KF:
Orbital myiasis. Ophthalmology 1986;
93:12281232.
156. Navarro P, Vera Cristo L: Miasis

multilarvaria orbitaria interna. Rev Soc Col
Oftalmol 1990; 23:28.
157. Gass JDM, Lewis RA: Subretinal tracks in
ophthalmomyiasis. Arch Ophthalmol 1976;
94:15001505.
158. Slusher MM, Holland WD, Weaver RG, Tyler
ME: Ophthalmomyiasis interna posterior.
Subretinal tracks and intraocular larvae.
159. Edwards KM, Meredith TA, Hagler WS,
Healy GR: Ophthalmomyiasis interna
causing visual loss. Am J Ophthalmol 1984;
97:605610.
160. Fitzgerald C, Rubin M: Intraocular parasite
destroyed by photocoagulation. Arch
Ophthalmol 1974; 91:162164.
161. Lertchavanakul A, Pearce WG, Nigam S:
Ophthalmia nodosa. Can J Ophthalmol
1975; 10:8689.
162. Haluska FG: Experimental gypsy moth
(Lymantria dispar) ophthalmia nodosa. Arch
Ophthalmol 1983; 101:799801.
163. Gurevitch AW: Scabies and lice. Pediatr
Clin North Am 1985; 32:9781018.
164. Couch JM, Green WR, Hirst LW, De La
Cruz ZC: Diagnosing and treating Phthirus
pubis palpebrarum. Surv Ophthalmol 1982;
26:219225.
165. Mathew M, DSouza P, Mehta DKA: A new
treatment for phthiriasis palpebrarum. Ann
Ophthalmol 1982; 14:439441.
166. Wu JJ, Huang DB, Pang KR, Tyring SK:
Rickettsial infections around the world,
part 1: pathophysiology and the spotted
fever group. J Cutan Med Surg 2005;
9:5462.
167. Brezina R, Murray ES, Tarizzo ML:
Rickettsiae and rickettsial diseases. Bull
WHO 1973; 49:433442.
168. Weiss E: The biology of rickettsiae. Annu
Rev Microbiol 1982; 36:345370.
169. Walker DH: Diagnosis of rickettsial
diseases. Pathol Annu 1988; 23(Pt 2):6996.
170. WHO Working Group on Rickettsial
Disease: Rickettsioses: a continuing
disease problem. Bull WHO 1982;
60:157164.
171. Duffey RJ, Hammer ME: The ocular
manifestations of Rocky Mountain spotted
fever. Ann Ophthalmol 1987; 19:301306.
CHAPTER 14
157
CHAPTER
15
Fungal Infections of the Eye

Wiley A. Schell, Gary N. Foulks, and John R. Perfect
The rst reported case of fungal infection of the cornea dates

back to 1879, involving a farmer who was struck in the eye by
oat chaff with resultant keratomycosis caused by Aspergillus
glaucus.1 Physicians and microbiologists subsequently have
realized the unique relationship between fungi and human
ocular disease. The frequent association of fungal ocular
infection with occupational trauma and exposure to vegetable
material is well documented.210 Increasing recognition of
fungal ocular infection in the 1950s and 1960s concurrent with
the increased use of topical antibiotics and corticosteroids on
the eye led to more than 148 case reports by 1962 and rmly
established the association of fungal infection with impaired
host defenses or physical trauma.11 Subsequent work has
conrmed the importance of impaired host defenses or broken
anatomic barriers; and, has examined fungal growth characteristics as they relate to expression of clinical disease, providing
insight into improved therapy against these infrequent but
extremely tenacious invaders.
Many fungal species have been identied in human ocular
diseases.1214 Chorioretinal or orbital diseases are most often a
result of systemic mycoses contracted through respiratory tract
exposure (Histoplasma, Cryptococcus, Blastomyces, Coccidioides)
or dissemination from the gastrointestinal tract or an intravascular catheter (Candida).1524 In contrast, the fungal species
associated with lacrimal, corneal, or traumatic intraocular
infections are found in soil and vegetable matter and can be
cultured from 2.5% to 52% of normal eyes, depending on
climate and occupation. Fungi are not part of the normal flora
of the lids or conjunctiva of normal eyes but are only transient
colonizers. When specimens are taken from the conjunctiva or
lids, the same fungus is rarely isolated sequentially in an
individual, and most cultures grow only one or two fungal
colonies, suggesting a very low burden of organisms.25,26 Almost
half of the reported cases of ocular surface infections are
attributed to environmentally common species of the genera
Aspergillus, Penicillium, Fusarium, and to Candida albicans, a
commensal of humans. This nding correlates with epidemiologic studies in which these fungi have been transiently isolated
from normal eyes.
Typically, environmental fungi cause keratitis after penetrating into the cornea through trauma. Also, topical therapy
with antibiotics and corticosteroids generally increases fungal
colonization of the eyelids and conjunctiva and is thus a major
predisposing factor for oculomycosis through superinfection.27,28
Isolation of fungal species in eyes with known underlying
abnormalities such as dacryocystitis has increased. An association of seborrheic blepharitis with Malassezia furfur colonization or infection has been suggested. Finally, an increase in
colonization of eye structures may result from exogenous
factors, including the use of mascara contaminated by fungi

such as Candida parapsilosis.
HOSTFUNGI INTERACTIONS IN THE EYE

Key Features: Fungal Infections of the Eye
Exogenous
Keratomycosis
Chorioretinitis
Endophthalmitis
Endogenous
Chorioretinitis
Endophthalmitis
Orbital
Ocular defenses to fungal infection are numerous, and

oculomycosis is common only when anatomic structures are
breached. Normal flora of the eyelids, the conjunctival sac with
normal lacrimation, and the mechanical movements of the
eyelids create an unfavorable environment for the growth of
most opportunistic fungi, such as Aspergillus and Candida
species. Alteration of the normal flora with systemic or topical
antibacterial agents or corticosteroids, however, can decrease
this barrier and allow colonization and growth of fungi. Because
many fungi do not grow at elevated temperatures, normal body
temperature is high enough to prevent many environmental
fungi from becoming pathogenic. The lower temperature of the
cornea relative to the rest of the body and eye, and its exposure
to potential trauma, may partially explain why keratomycosis is
the most common ocular fungal infection. The intact corneal
epithelium is generally resistant to fungal penetration and
infection; this affords great protection. Breach of the epithelial
barrier is often a prerequisite for keratomycosis, which explains
its association with trauma through occupational, recreational,
or surgical exposures. First, direct inoculation by trauma may
occur when the fungus is carried on a projectile. Second,
colonizing fungi may invade the wound after trauma; such
invasion is particularly enhanced by the use of antibacterials,
corticosteroids, or both. Third, surgical procedures such as
keratoplasty, corneal transplantation, or radial keratotomy are
occasionally associated with introducing fungi into the eye via
transplant or contaminated irrigating solutions.2935 Several
well-described outbreaks of ocular fungal infections with
C. parapsilosis and Paecilomyces lilacinus have been associated
with lens implants and contaminated irrigation solutions.3639
Finally, soft contact lenses can act as a nidus for fungal invasion
159
SECTION 3
MICROBIOLOGY
160
into the cornea if they are not properly cleaned and disinfected
(Fig. 15.1).8,4041 Corneal infection allows extension to the
sclera or intraocular space because there are few subsequent
tissue barriers. The role of local antibodies and complement
in protection against fungal infections of the eye is uncertain.
The polysaccharide nature of the fungal cell wall can activate
complement, and secretory immunoglobulin A (IgA) can protect
against mucosal infection with Candida species, but the
importance of such local immunity protection in the eye is not
well understood. On the other hand, clinical experience demonstrates that topical and systemic corticosteroids enhance the
risk of ocular fungal infections and clearly suggests that local
immunity factors are important in protecting the eye from
fungal invasion.
The second avenue for fungal invasion is through the blood
stream (endogenous rather than exogenous). This oculomycosis
generally occurs when there is some systemic host immune
depression. The most common example is white blood cell
defects, particularly chemotherapy-induced neutropenia.
During neutropenia, invasion of the eye is particularly difcult
to diagnose because general hallmarks of infection, such as an
inflammatory response in the chorioretina or vitreous body, are
not always visible.42 Candida and Aspergillus species, however,
can reach the retina in the presence of a normal granulocyte
count if the systemic inoculum is high, as occurs in certain
human infections and in experimental animal models. For
example, fungal ocular infections have occurred during hyperalimentation, post partum, during prolonged antibiotic therapy,
in the neonatal period, and with intravenous drug use.4361 The
cell-mediated immune system is a well-characterized protective
system against fungal infection and obviously is important in
preventing and ghting established ocular fungal infections.
Debilitating diseases or generalized impairment of the immune
system are predilecting factors for fungal infection, both
systemically and ophthalmically. Rhino orbital zygomycosis in
the diabetic or cancer patient represents invasion of blood
vessels within the orbit secondary to an underlying immune
depression. C. neoformans invasion of the orbit or chorioretinal
area has become more common in severely immunocompromised hosts with cell-mediated defects, such as patients
with acquired immunodeciency syndrome (AIDS) and those
on high-dose corticosteroids.
Although ocular involvement with C. neoformans has
increased during the AIDS epidemic, infection with this fungus
was frequently reported in prior years. One study found ocular
FIGURE 15.1. Fusarium solani growing from soft contact lens.
signs and symptoms in 45% of all patients with meningitis.17

Manifestations range from ocular palsies to involvement of the
choroidretina.17,62 In one-fourth of cases, eye involvement is
diagnosed before meningitis.63 Simultaneous infections with
C. neoformans and other pathogens such as human immunodeciency virus and cytomegalovirus can occur in severely
immunosuppressed patients.64,65 Although most cases of ocular
cryptococcosis arise from bloodstream dissemination, the eye
has been the direct portal of entry in such cases as donor
transmission through a corneal transplant30 and cryptococcal
keratitis after keratoplasty procedures.32,33 Thus, some cases of
disseminated cryptococcosis might originate in the eye rather
than the lung.
Ocular cryptococcosis can lead to visual loss. In fact, most
cases of cryptococcal endophthalmitis lead to severe visual loss;
successful management is rare.66,67 The AIDS epidemic has
given rise to reports of catastrophic loss of vision in patients
with cryptococcosis without evidence of endophthalmitis.68,69
The funduscopic examination yields either normal results or
evidence of papilledema. The clinical manifestations suggest
two pathogenic processes. First, some patients experience rapid
visual loss within 12 h to a few days. This clinical syndrome
suggests optic neuritis in which the optic nerve and its vessels
are inltrated by large numbers of yeast cells. No successful
therapeutic strategies are known for this form of visual loss.
Other patients can present with slow visual loss that generally
begins later during antifungal therapy and gradually progresses
over weeks to months. Symptoms may be related to increased
intracranial pressure in these patients, and treatment with
central nervous system shunts or optic nerve fenestrations may
halt the progression of visual loss.69
In contrast with infections, in which ocular defenses clearly
fail to prevent fungal inoculum from replicating, there is a
syndrome called presumed ocular histoplasmosis, which is
characterized by chorioretinal scars, hemorrhages, and neovascularization. It has been suggested that these host reactions
are due to the presence of the yeast cells or antigens of
H. capsulatum, but only rarely have viable organisms been
documented for this syndrome.70 The thrust of treatment has
been corticosteroids or laser therapy to stop the lesions
advancement;71 antifungal therapy rarely has been helpful.70
When oculomycosis occurs, the fungus tends to invade
directly into tissue planes. This is particularly apparent in
keratomycosis, as demonstrated in Figure 15.2, a case of
C. parapsilosis infection in a keratoplasty patient on long-term
topical steroid therapy. The host response to the organism can
be acute suppurative inflammation, chronic inflammation, or
granulomatous inflammatory reaction, depending on the fungal
species and tissue location (Fig. 15.3). The organism can
actively damage host tissue by stimulating the host to elaborate
inflammatory mediators such as oxidative products. The fungus
may also secrete products that injure the eye. For example, a
potential virulence factor for C. albicans is its production of
extracellular acid proteases and phospholipases, which may
further aid in tissue destruction.7274 Aspergillus species can
produce elastase, which most likely facilitate hyphal invasion
into blood vessels and may further contribute to damage of eye
tissue.75 Certain fungi produce mycotoxins under specic
conditions, but to date no such products have been detected in
or shown to contribute to destruction of ocular tissue.
Fungi possess poorly understood factors that allow a certain
tropism for eye structures during bloodstream invasion. For
instance, during fungemia with C. albicans in the rabbit model
of candidiasis, yeast cells consistently localize in the eyes and
kidneys while other tissues are spared. In humans, the propensity for ocular invasion during candidemia is high.46,4960 This
may be related to the unique vascular arrangements of the eye,
FIGURE 15.2. (a) Clinical picture of stromal keratitis in corneal graft (Candida parapsilosis). (b) Histologic section demonstrating deep lamellar
inltration of yeast (C. parapsilosis) with acute and chronic inflammatory cellular inltrate. Methenamine silver stain, 33.
a1
a2
FIGURE 15.3. Histopathology of mycotic

ocular infections. (a1) Stromal keratitis due to
Candida parapsilosis with acute and chronic
inflammatory inltrate. (a2) Methenamine silver
stain, 132. Keratitis with inltration by
Cryptococcus neoformans showing
granulomatous reaction. Papanicolaou stain,
600. (b1) Endophthalmitis due to a
zygomycetous fungus. H&E, 132.
(b2) Chorioretinitis due to Aspergillus species.
PAS, 132.
CHAPTER 15
(b1 and b2) Reprinted from Perry HD, Donnenfeld ED:

Cryptococcal keratitis after keratoplasty. Am J
b1
b2
but specic fungal factors for this localization also are likely.
Findings suggest that early pseudohyphal formation plays a role
in establishing an endogenous ocular infection. This propensity
for C. albicans ocular infections has been corroborated in
human infections, of which the vast majority are associated
with this Candida species. However, other Candida species
occasionally cause endogenous eye infection, particularly when
the inoculum is as large as can occur with C. parapsilosis
infection during hyperalimentation. Spores from Aspergillus
species, which are found on fomites such as drug paraphernalia,
can reach ocular structures and establish infection when
inoculated intravenously.53,7678
DIAGNOSTIC TESTING
The diagnosis of fungal etiology in ocular infection can be
difcult. Certain clinical characteristics may be helpful to
ophthalmologists, including duration and features of the ocular
lesions. These are reviewed elsewhere in this book. However, it
must be emphasized that there remains no substitute for the
proper collection of specimens for histologic and cultural identication (Fig. 15.4). Infections are diagnosed in the laboratory
by culture or microscopy performed on clinical specimens.
Recently, it also has been shown that polymerase chain reaction
(PCR) amplication of fungal DNA can be used to detect and

identify the infecting fungus.7984
Microscopy of clinical specimens can be performed by
various methods. Calcofluor white/KOH is one extremely
sensitive technique. It is rapid and easy to perform but is not a
permanent preparation. Giemsa stain, periodic acid-Schiff, and
methenamine silver stain are sensitive and permanent
preparations. Grams stain detects yeasts such as Candida
species but is not reliable for other fungi such as molds and
should not be relied on for detecting mycotic infection. Gramstained slides can be decolorized and reused with one of the
preferred reagents. Microscopy may reveal yeast or hyphae of
the infecting organism, but specic identication of the species
of fungus requires culture. Fluorescein-conjugated lectins or
fluorescent antibody conjugates have been used to allow
differentiation among species such as Candida, Aspergillus, and
Fusarium, but these stains are not commercially available.85
Certain molds, particularly Paecilomyces lilacinus, sometimes
form spores within the infected tissue, and this can be a useful
differential characteristic.86 These spore forms can be mistaken
for Candida species. Supercial infections can be identied by
scraping surface lesions, with organisms identied by culture
and often corroborated by microscopy of stained smears prepared
from the scrapings. Denitive diagnosis of deep keratitis or
161
MICROBIOLOGY
SECTION 3
THERAPEUTIC CONCEPTS
Key Features: Categories of Primary Therapeutic
Agents
e
FIGURE 15.4. Colony and microscopic appearance of pathogenic

fungi. (a) Creamy, round colony growth of Candida albicans.
(b) Grams stain of yeast cells, Candida species. (c) Filamentous
colony of Aspergillus fumigatus, with characteristic blue-green color
due to sporulation. (d) Cotton blue stain of hyphae and sporulating
structures of A. fumigatus. (e) Filamentous white colonies of Fusarium
solani. (f) Microscopic view of spores of F. solani.
162
laboratory personnel to determine whether growth is on or off

the inoculation streaks and thus differentiate possible pathogens from airborne contaminants. Specimens should be
collected from external ocular surface infections or lacrimal
infections with a moist applicator and inoculated by streaking
directly onto culture media.88 Scrapings from corneal ulcers in
cases of keratomycosis or aspirates from the anterior chamber
and the vitreous cavity in cases of endophthalmitis should be
directly inoculated onto both Sabourauds (or similar) agar
media and brain-heart infusion broth medium as well as blood
agar plates.86,88 Incubation should be at 2430C (30C is
preferred) and should be prolonged. Although species of some
genera, such as Candida, Fusarium, Paecilomyces, Curvularia,
and Alternaria, normally are visible within 3 days, as many as
25% of fungal isolates may require up to 2 weeks of incubation.88
When Histoplasma or Blastomyces is suspected, cultures should
be incubated for at least 4 weeks.
intraocular infection often requires culture of an aspirate,

because direct smears do not always reliably correlate with
culture-proved infection.9,10,8688 Biopsy of deep corneal lesions
may be required to demonstrate the organism by special
histologic stains. In deep infections of the cornea, supercial
scraping may not yield enough organisms to identify or culture.
Corneal biopsy with histopathologic examination may be
required. In such situations, the use of periodic acid-Schiff,
methenamine silver, or calcofluor white stains is helpful in
demonstrating the organism; detection with fluoresceinconjugated lectins or fluorescent antibody conjugates also is
possible.89
Although microscopy can be the most rapid laboratory test
for establishing a diagnosis, culture is required in order to
perform in vitro antifungal susceptibility testing. Also, because
of a high error rate for named sequences in public databases,
it is strongly recommended that PCR results be correlated with
results obtained by traditional culture methods. Because ocular
infections are often caused by common saprobes in the
environment and access to tissue or other diagnostic specimens
is limited, special techniques for specimen evaluation must be
used to diagnose fungal infections of the eye. The clinicians and
laboratory personnel must communicate effectively, to agree on:
protocols for using media that do not inhibit fungal growth,
inoculation techniques that help differentiate infective
organisms from contaminants, and prolonged incubation times
and optimal temperature to allow for the slow growth of some
fungal species.
Media for culture should not include cycloheximide, which
inhibits fungal growth, but inclusion of gentamicin or chloramphenicol may be needed to suppress bacterial overgrowth. A
streak inoculation technique on media with specimens obtained
from the examining room or operative suite should allow
Polyenes (amphotericin B, pimaricin)

Pyrimidines (5-fluorocytosine)
Imidazoles/triazoles (fluconazole, itraconazole, ketoconazole,
posaconazole, voriconazole)
Echinocandins (anidulafungi, caspofungin, micafungin)
Biguanides (polyhexamethyl biguanide)
Therapy of fungal infections can be difcult and prolonged. The

difculty in treatment is due to a combination of the growth
characteristics of fungi, the limited availability of effective
antifungal agents, and the poor tissue penetration of these
agents. Until 1950, safe and reliable treatment for deep fungal
infections did not exist, and treatment of supercial infections
depended on empirical topical preparations. Nystatin was rst
introduced in the mid-1950s, and amphotericin B came to
dominate treatment of deep mycoses in the 1960s. In the 1970s,
5-fluorocytosine was introduced as treatment for candidiasis
and cryptococcosis, but drug resistance became a problem.
Since the mid-1980s, several N-substituted imidazole or
triazole compounds have been introduced and modied with
signicant improvement in activity and pharmacokinetics.
The most useful antifungal agents are of two groups: those
affecting cell walls or membranes, and those interrupting nucleic
acid or protein synthesis. The polyene macrolide antibiotics
interact with the sterols in the fungal cell membranes to impair
their barrier function and thus produce leaking of cellular
substances with subsequent metabolic disturbance and resulting cell death. The toxicity of amphotericin B, however, is
related to similar interactions with sterols in host cells.
Resistance of fungi to amphotericin B is rare and probably
occurs by alterations in the sterol composition of the cell
membrane.90 The azole antifungals share an imidazole or
triazole ring with N-carbon substitution that allows interaction
with primary target sites within the fungal cell. At low concentrations, these compounds inhibit cytochrome P-450 enzymes,
which leads to the accumulation of 14-g-methylsterols and
reduced biosynthesis of ergosterol. At higher concentrations,
some azoles can cause direct cell membrane damage. The
fluorinated pyrimidine, 5-fluorocytosine, is deaminated once
inside the susceptible yeast cell: A cytosine deaminase converts
it to 5-fluorouracil for incorporation into fungal RNA and thus
disruption of protein synthesis. The echinocandin class of
antifungals is the most recent of antifungal systemic agents and

it targets a cell wall synthesis enzyme, 1,3-b-glucan synthetase.
This enzyme inhibition results in fungicidal activity against
many Candida species and inhibition of hyphal tip growth in
Aspergillus species.
In vitro testing of antifungal susceptibility and its correlation
with in vivo response historically have been difcult because
minimum inhibitory concentrations vary greatly under different
test conditions. However, progress has been made and standardized protocols now are in effect for susceptibility testing of
yeasts and molds.9193 Standardization of in vitro antifungal
testing involves comparing direct antifungal activity, pharmacokinetics of the agents, and prior clinical experience on
treatment of certain fungal infections. Efforts are ongoing to
establish standardized testing for new antifungal drugs effective
against a wider range of fungi. Despite the concerns about
clinical validation, our opinion is that yeasts and possibly molds
from serious oculomycoses should be evaluated comparatively
by in vitro susceptibility testing with available antifungal
agents. This can allow detection of possibly drug-resistant fungi
and can provide the grounds for clinical judgment of the best
antifungal regimen.
Response to therapy depends on several factors. Host factors
include the integrity of the immune defense mechanisms
(especially cell-mediated functions) and the location and extent
of infection. Pharmacokinetic factors include penetration and
tissue distribution of the antifungal agent as well as predilection
for tissue binding. Antimicrobial factors include the observable
effect on the fungal organisms and the response in growth
characteristics of the fungus in the presence of the antifungal
agent. A further clinical problem in treatment is that when the
organism encounters adverse conditions (elevated temperature,
anaerobiosis, chemotherapeutic agents), it may revert to a
dormant or slow growth state that is more difcult to eradicate
with cell wall- or cell membrane-active antifungals and thus
requires longer treatments. Finally, clinical experience both
that of the attending clinician and that gleaned from references
in the literature can be a helpful guiding factor. The following
discussion summarizes specic therapeutic concepts in
management of oculomycoses.
The single most important factor in the success of treatment
for oculomycosis is early diagnosis and treatment. Fungal infections can have an indolent course, and the longer these infections remain untreated, the more difcult they are to eradicate.
For this discussion, infections are divided into three categories:
(1) keratomycosis, (2) endophthalmitis, and (3) orbital infection.
KERATOMYCOSIS
Fungal keratitis is usually caused by environmentally widespread molds such as Aspergillus species, Fusarium species,
Paecilomyces species, and Curvularia species, but other fungi,
such as Candida species and Cryptococcus neoformans, also
can cause keratitis in susceptible hosts. Identication of the
fungus and comparative in vitro susceptibility testing to
available antifungal drugs usually are important. For fungal
corneal ulcers, pimaricin remains the most reliable topical
antifungal agent in a 5% suspension or as a 1% ointment for
treatment of supercial ocular injuries or prophylaxis with
high-risk injuries for oculomycosis. It also is not as irritating
to the eye as the other polyenes, such as amphotericin B.
Unfortunately, pimaricin therapy has two drawbacks. First,
although it has broad-spectrum antifungal activity across many
species, isolates may be relatively resistant to its antifungal
activity, with only half of studied strains being inhibited by
3 mg/mL or less.94 Second, it has limited ability to penetrate the
cornea. Third, the lm formed on the cornea by pimaricin is
not conducive to simultaneous use of multiple antifungal

eyedrops. Nystatin is less active in vitro than the other polyenes
but is reasonably well tolerated in a 3% ointment. Amphotericin
B can be irritating to the eye, and in high concentrations (5%)
can lead to punctate epithelial erosions. Even so, it is frequently
used in a topical solution for serious infections. Topical antifungals are likely to be most successful early in the infection,
before it has extended into deeper layers of the cornea. Novel
topical disinfectants, notably polyhexamethylene biguanide
(Bacquacil) have shown efcacy in experimental testing and in
limited clinical use.95,96 Although evidence-based studies are
lacking, use of multiple antifungal eyedrops to achieve potential
synergistic or additive activity might be tried in particularly
refractive cases. It is emphasized that proper cultures for
isolation and identication of the fungus should be taken before
beginning therapy.
The second approach to therapy of keratomycosis is the use
of systemic antifungal agents. For supercial fungal ulcers, this
second line of therapy may not be necessary, but deeper corneal
infections may require it. The azole compounds have become
attractive candidates for systemic administration. They are safe
and relatively broad-spectrum. The ocular pharmacology of
these azole compounds (miconazole, ketoconazole, fluconazole,
and itraconazole) has been examined in both humans and
animals.97,98 The rank of penetration into eye structures such as
vitreous body and aqueous humor, from highest to lowest, is
fluconazole, ketoconazole, miconazole, and itraconazole. The
azoles penetration into the eye appears to be improved by
inflammation, as is the case with other drugs. Azoles have been
shown to penetrate into corneal tissue of rabbits and can be
found in corneal tissue even when the eyes are not inflamed.98,99
Therefore, it is reasonable to anticipate that future reports will
show the success of these agents in the management of fungal
keratitis and scleritis. Flucytosine is another agent with excellent
penetration into eye structures and has shown some success in
Candida keratitis. Its major limitation is its narrow spectrum
of activity. It inhibits only a portion of Candida species,
C. neoformans, and some dematiaceous molds. For corneal
infections, systemic amphotericin B therapy has not been
widely used. It has been used, however, as a topical preparation
and in fungal scleral infections as a subconjunctival injection.
The success of subconjunctival injection of amphotericin B
remains unclear, and it can be extremely painful and sometimes
produces tissue necrosis and nodules.100,101 Its limited eye
penetration and toxicity have reduced enthusiasm for its use in
infection at this site except for the most difcult cases.
Voriconazole has been successfully used as a topical and
systemic drug to treat Fusarium keratitis that was refractory to
amphotericin B and itraconazole.102 Voriconazole concentration
in the anterior chamber of the eye in this case was 160% of that
measured in the plasma. Another study showed voriconazole
concentrations in the vitreous and aqueous that were 3853%
that of the plasma in patients who received two 400 mg doses
12 h apart.103 Other reports of clinical efcacy for voriconazole
in keratomycosis include three cases caused by Scedosporium
apiospermum.104106
Econazole (2%), in a randomized trial of 112 patients, was
shown to be as effective as 5% natamycin in the treatment of
keratomycosis caused by molds, mainly A. flavus and Fusarium
(species not specied).107
Caspofungin experience in keratomycosis is scant, but comparison of 0.5% caspofungin with 0.15% amphotericin B in a
rabbit model of C. albicans keratomycosis showed equal
efcacy.108
In addition to antifungal therapy, some eyes require excisional
keratoplasty, particularly in cases of impending perforation.
Even in these cases, however, aggressive antifungal chemotherapy
CHAPTER 15
163
MICROBIOLOGY
before and after surgery may improve the nal level of visual
acuity.
There is no strong evidence that topical or systemic steroids
help in the management of fungal eye infections. In fact, they
are often the major risk factor for these infections and their
progression. Prevention of inflammation and resultant tissue
destruction and the preservation of visual acuity are vital
objectives, but there are no guidelines to balance the positive
effects of steroids on inflammation versus the negative effects of
stimulating fungal growth. Therefore, adjunctive corticosteroid
therapy should not routinely be used in fungal eye infections.
SECTION 3
ENDOPHTHALMITIS
164
There are two types of fungal endophthalmitis. Exogenous

endophthalmitis is associated with trauma or surgery in which
the organism is introduced directly into the ocular structures.
Endogenous endophthalmitis is generally produced by Candida
species or Aspergillus species from a chorioretinal lesion, and
extension into the vitreous body accompanies systemic dissemination of the fungus. It may also occur with the endemic
mycoses, such as blastomycosis, after the initial pulmonary
infection. The need to manage these infections has signicantly
intensied over the last decade because of expanding immunocompromised populations, complex surgical procedures, and
increasing use of antibiotics and intravenous catheters. The
most important therapeutic principle in endophthalmitis is
early diagnosis and correct identication of the fungus.109 For
instance, in patients with candidemia who are not neutropenic,
a prospective evaluation of the eye may identify an early ocular
infection in a third of the patients.45 Early treatment is more
likely to yield a better visual outcome. Animal models of endogenous C. albicans endophthalmitis suggest that early treatment with either azoles or amphotericin B is more successful
than delaying treatment for a week despite similar numbers of
yeasts at each time period.98,110 Correct identication of the
organism by blood or ocular fluid cultures and determination of
in vitro susceptibility to various antifungal agents helps identify
the most promising antifungal agents for successful treatment.
Candida remains the most common invasive ocular pathogen
for endophthalmitis. Because there are no comparative studies
on therapeutic regimens, it remains reasonable to select the
antifungal agent with the most successful experience, amphotericin B. Systemic amphotericin B in doses of 0.51 mg kg1 day1
has been used to control Candida endophthalmitis but at the
higher doses toxicity is substantial. Amphotericin B has very
low levels as measured in the vitreous body and aqueous humor,
but these measurements do not account for drug that is bound
to tissue.111,112 Because the penetration of amphotericin B is
poor, however, intraocular therapy combined with vitrectomy
frequently has been used. In a primate model, up to 3 mg of
intravitreal amphotericin B was tolerated without permanent
retinal toxicity, and a human took 50 mg of amphotericin B over
a 6-month period without serious retinal toxicity.113 A slowly
given 15 mg intravitreal injection is probably not toxic to the
retina. Now that liposomal amphotericin B at 35 mg kg1 day1
is available, it may be possible to deliver even more drug to
this site of infection safely.114,115 The value of intravitreal
amphotericin B is not proved and toxicity questions do remain,
but it may be of particular benet when the vitreous body is
signicantly involved, as in cases requiring vitrectomy and in
Aspergillus infections extending into the vitreous body.115
Flucytosine remains a possible agent for ocular Candida
infections with its high penetration into the vitreous body and
aqueous humor.116 There has been little experience with its use
alone, and concern over primary resistance in a portion of
Candida isolates remains.117 An attractive regimen for Candida
endophthalmitis would be combination chemotherapy with

amphotericin B and flucytosine.118 This combination regimen
has been successful in prospective studies in the treatment
of cryptococcal meningitis, and its in vitro synergy against
Candida by virtue of different mechanisms of antifungal action
theoretically could eradicate the fungus more rapidly and
improve visual outcome. However, no prospective studies have
proved this hypothesis.
With the advent of the azole compounds, clinicians have
another treatment avenue. The early azoles (ketoconazole and
miconazole) had some successes and failures. The newer azole
(fluconazole) has excellent ocular pharmacokinetics and may be
helpful in managing ocular fungal infections. The only comparative data regarding the efcacy of these compounds are
from animals.98,110 These models suggest that amphotericin B
may still be more potent in eradicating Candida from the eye
than the azole compounds are. There have also been case reports
of Candida and Coccidioides infections in which miconazole
was not effective but patients improved after receiving
amphotericin B therapy.119 Such results, however, should not
necessarily dissuade clinicians from carefully using these newer
azole compounds in fungal endophthalmitis, because more
clinical experience with these compounds in ocular infections is
needed. For example, one report on ocular candidiasis in drug
addicts cited an excellent response to ketoconazole treatment.47
Voriconazole reports thus far suggest this new azole can
be effective in treating some cases of endophthalimitis caused
by Aspergillus species,120,121 Scedosporium apiospermum,
Paecilomyces lilacinus,122 Scytalidium dimidiatum,123 Fusarium
verticilloides (as F. moniliforme),124 and Candida sp.125 In addition to high intraocular levels being measured in cases of
systemic administration, an animal model indicates that
intraocular injections of voriconazole are well tolerated.126
Although amphotericin B and flucytosine remain the most
attractive combination regimen for Candida, a polyeneazole
combination might be useful in certain eye infections,
particularly if both antifungals have in vitro activity against the
fungus. The concern about polyeneazole antagonism in vitro
has not been proved in vivo. Another combination regimen that
may be considered is fluconazole plus flucytosine. These two
oral agents reach high drug levels within ocular tissue. Finally,
the regimen of amphotericin B plus rifampin has been used
successfully both in animals and in humans.127,127a,128 The
point of this discussion is that combination antifungal chemotherapy can be considered rational treatment if proper
identication and comparative in vitro susceptibility testing on
the fungus are performed.
The newest class of antifungal compounds targets the
synthesis of 1,3-b glucan within the fungal cell wall. One of
these, caspofungin, is now available but clinical experience is
very limited at this time. One prospective study reported
success in all seven Candida endophthalmitis patients using
caspofungin.129 Another reported successful use of caspofungin
in treatment of C. glabrata endophthalimitis130 A. fumigatus
was successfully treated by adding caspofungin to a regimen
of voriconazole.124 A retrospective review of endophthalmitis
suggested that combination voriconazolecaspofungin can be
effective in the treatment of Candida endophthalmitis as
well.131 In contrast, however, treatment failure of C. albicans
endophthalmitis accompanied by poor ocular penetration has
been reported.132 Two additional echinocandins, mycafungin
and anidulafungin, have been approved for use with esophageal
and invasive candidiasis but experience with thse drugs has not
yet been reported for fungal eye infections. Therapeutic
vitrectomy may be helpful in certain patients and likely clears
the eye of inflammatory debris.48,49,58,61,78,133143 For this treatment, our current understanding makes it reasonable to select
ORBITAL INFECTION
Fungal infections in the orbit that do not initially invade ocular
structures are generally caused by a member of Zygomycetes
such as Rhizopus species or by Aspergillus species.16,21,144 The
rhinocerebral form of zygomycosis is a characteristic acute
progression of infection into the orbit, causing orbital swelling
and eventual paralysis of orbital structures.21,145 Generally
caused by R. arrhizus, this infection primarily affects diabetics,
particularly if acidosis has occurred; cancer patients; or patients
receiving chelation or steroid therapy. The infection starts in
the nasal or sinus cavities and invades the regional arterial
vessels by direct extension, causing thrombosis and leading to
ischemic necrosis. Extension through the orbital apex into the
brain occurs as infection progresses. A black eschar in the nasal
area or drainage of black pus from the eye suggests this
diagnosis.21 Identifying the patient at risk and performing an
early examination of the nasal and sinus areas for signs of
disease often leads to diagnosis before the orbit becomes involved.
Aspergillus infections of the sinus have eroded through bone or
invaded local vessels and entered the orbit, producing proptosis.
Therefore, evaluation of recent proptosis of ocular structures
should include a careful examination of the sinuses.
Early debridement of infarcted tissue is essential to a successful outcome and may obviate the need for subsequent orbital
exenteration. The goal of treatment remains the prevention
of extension into the brain. The immediate control of the
underlying disease, such as acidosis, is also important; nally,
amphotericin B at 0.71 mg kg1 day1 or a lipid formulation of
amphotericin B at 5 mg kg1 day1 is usually given. The lipid
formulation of amphotericin B offers reduced toxicity compared
to the non-lipid formulation. Posaconazole, a new triazole, is
gaining positive experience in treatment of zygomycosis and
may become part of the management strategy.146 The length of
therapy should be tailored to the patients response and extent
of infection.
CONCLUSION
Fungal infection in the eye is most often of exogenous origin in
an immunocompetent host whose local tissue defenses have
been damaged. The growth characteristics of the fungus can
result in supercial infection or invasion into deep tissues,
where it may alter its growth pattern in response to the local
milieu. Effective therapy of such infections must be selected
from the small number of antifungal agents and requires
recognition of the limitations of susceptibility testing, the
importance of tissue penetration and absorption, and the need
for protracted treatment. Because of these limitations, success
of therapy primarily depends on early diagnosis of the fungal
infection and correct identication of the particular fungus.
CHAPTER 15
patients with extensive vitreous involvement and likely visual

impairment from scarring, with progressive inflammation
despite antifungal agents, and patients with extensive vitreal
involvement but an unclear underlying pathogen.
REFERENCES
1. Leber T: Keratomycosis aspergillina als
Ursache von hypopyonkeratitis. Graefes
Arch Clin Exp Ophthalmol 19; 25:285301.
2. Chick EW, Conant NF: Mycotic ulcerative
keratitis: a review of 148 cases from the
literature. Invest Ophthalmol Vis Sci 1962;
1:419420.
3. Blake J: Ocular hazards in agriculture.
Ophthalmologica 1969; 158:125.
4. Laspina F, Samudio M, Cibils D, et al:
Epidemiological characteristics of
microbiological results on patients with
infectious corneal ulcers: a 13-year survey
in Paraguay. Graefes Arch Clin Exp
Ophthalmol 2004; 242:204209.
5. Kunimoto DY, Sharma S, Reddy MK, et al:
Microbial keratitis in children.
6. YinWong T, SiewFong K, Tan DTH: Clinical
and microbial spectrum of fungal keratitis
in Singapore: a 5-year retrospective study.
Int Ophthalmol 1997, 21:127130.
7. Garg P, Gopinathan U, Choudhary K, Rao
G N: Keratomycosis: clinical and microbiologic experience with dematiaceous
fungi. Ophthalmology 2000; 107: 574580.
8. Tanure MAG, Cohen EJ, Sudha S, et al:
Spectrum of fungal keratitis at Wills Eye
Hospital, Philadelphia, Pennsylvania.
Cornea 2000; 19:307312.
9. Chowdhary A, Singh K: Spectrum of fungal
keratitis in North India. Cornea 2005;
24:815.
10. Gopinathan U, Garg P, Fernandes M, et al:
The epidemiological features and
laboratory results of fungal keratitis: a
10-year review at a referral eye care center
in South India. Cornea 2002; 21:555559.
11. Devoe AG, Silva-Hutner M: Fungal
infections of the eye. In: Locatcher-Khorazo
D, Seegal BC, eds. Microbiology of the
eye. St Louis: CV Mosby; 1972:208239.
12. Jones DB: Fungal keratitis. In: Duane T, ed.

Clinical ophthalmology. Philadelphia, PA:
Harper & Row; 1986.
13. Schell WA: Oculomycosis caused by
dematiaceous fungi. Proceedings of the
VIth International Conference on the
Mycoses. Pan American Health
Organization 1986; 879:105109.
14. Thomas PA, Geraldine P: Oculomycosis.
In: William GM, Roderick JH, eds.
Topley & Wilsons microbiology &
microbial infections. 10th edn. Medical
Mycology. London: Hodder Arnold;
2005:273344.
15. Scholz R, Green WR, Kutys R, et al:
Histoplasma capsulatum in the eye.
16. Agarwal LP, Malik SRK, Mohan M,
Mahopatra LN: Orbital aspergillosis.
17. Okun E, Butler WT: Ophthalmologic
complications of cryptococcal meningitis.
18. Font RL, Spaulding AB, Green WR:
Endogenous mycotic panophthalmitis
caused by Blastomyces dermatitidis. Arch
19. Blumenkranz MS, Stevens DS:
Endogenous coccidioidal endophthalmitis.
20. Cassady JR, Foerster HC: Sporotrichum
schenckii endophthalmitis. Arch
21. Rinaldi MG: Zygomycosis. Infect Dis Clin
North Am 1989; 3:1941.
22. Edwards JE, Foos RY, Montgomerie JZ:
Ocular manifestation of Candida
septicemia. Review of 76 cases of
hematogenous Candida endophthalmitis.
Medicine 1974; 53:4775.
23. Grifn JR, Pettit TH, Fishman LS: Bloodborne Candida endophthalmitis: a clinical
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
and pathologic study of 21 cases. Arch

Ophthalmol 1973; 89:450456.
Rao NA, Hidayat AA: Endogenous mycotic
endophthalmitis: variations in clinical and
histopathologic changes in candidiasis
compared with aspergillosis. Annual
Meeting of the American Ophthalmologic
Society, Pebble Beach 2000.
Ando N, Takator K: Fungal flora of the
conjunctival sac. Am J Ophthalmol 1982;
94:67.
Wilson LA, Ahearn DG, Jones DB, Sexton
RR: Fungi from the outer eye. Am J
Mitsui Y, Hanabusa J: Corneal infections
after cortisone therapy. Br J Opthalmol
1955; 39:244.
Nema HV, Ahuja OP, Bal A, Mohapatra LN:
Mycotic flora of the conjunctiva. Am J
Kobayashi GS: Fungi. In: Davis BD, ed.
Microbiology. Philadelphia, PA: JB
Lippincott; 1990.
Beyt BE, Waltman SR: Cryptococcal
endophthalmitis after corneal
transplantation. N Engl J Med 1978;
298:825826.
Gordon MA, Norton SW: Corneal transplant
infection by Paecilomyces lilacinus.
Sabouraudia 1985; 23:295301.
Perry HD, Donnenfeld ED: Cryptococcal
keratitis after keratoplasty. Am J
Ophthalmol 1990; 110:320321.
Castro LEF de, Al-Sarraf O, Lally JM,
et al: Cryptococcus albidus keratitis after
corneal transplantation. Cornea 2005;
24:882883.
Sutphin JE, Pfaller MA, Hollis RJ, Wagoner
MD. Donor-to-host transmission of
Candida albicans after corneal
transplantation. Am J Ophthalmol 2002;
134:120121.
165
SECTION 3
MICROBIOLOGY
166
35. Merchant A, Zacks CM, Wilhelmus K, et al:

Candidal endophthalmitis after
keratoplasty. Cornea 2001; 20:226229.
36. McCray E, Rampell N, Solomon SL, et al:
Outbreak of Candida parapsilosis
endophthalmitis after cataract extraction
and intraocular lens implantation. J Clin
Microbiol 1986; 24:625628.
37. Miller GR, Rebell G, Magoon RC, et al:
Intravitreal antimycotic therapy and the
cure of mycotic endophthalmitis caused by
a Paecilomyces lilacinus contaminated
pseudophakos. Ophthalmic Surg 1978;
9:5463.
38. ODay DM: Fungal endophthalmitis caused
by Paecilomyces lilacinus after intraocular
lens implantation. Am J Ophthalmol 1977;
83:130131.
39. Pettit TH, Olson RJ, Foss RY, Martin WJ:
Fungal endophthalmitis following
intraocular lens implantation: a surgical
epidemic. Arch Ophthalmol 1980;
98:10251039.
40. Wilson LA, Ahearn DG: Association of fungi
with extended-wear soft contact lenses.
Am J Ophthalmol 1986; 101:434436.
41. Goldblum D, Frueh BE, Zimmerli S, Bhnke
M: Treatment of postkeratitis Fusarium
endophthalmitis with amphotericin B lipid
complex. Cornea 2000; 19:853856.
42. Henderson DK, Hockey LJ, Vukakic LJ,
Edwards JE Jr: Effect of
immunosuppression on the development of
experimental hematogenous Candida
endophthalmitis. Infect Immun 1980;
27:628631.
43. Freeman JB, Davis PL, Maclean LD:
Candida endophthalmitis associated with
intravenous hyperalimentation. Arch Surg
1974; 108:237240.
44. Cantrill HS, Rodman WP, Ramsey RC,
Knobloch WH: Post-partum Candida
endophthalmitis. JAMA 1980;
243:11631165.
45. Brooks RG: Prospective study of Candida
endophthalmitis in hospitalized patients
with candidemia. Arch Intern Med 1989;
149:22262228.
46. Edwards JE Jr: Candida endophthalmitis.
In: Bodey GP, Fainstein V, eds. Candidiasis.
New York: Raven Press; 1985:211227.
47. Dupont B, Drouhet E: Cutaneous, ocular
and osteoarticular candidiasis in heroin
addicts. New clinical and therapeutic
aspects in 38 patients. J Infect Dis 1985;
152:577591.
48. Martinez-Vazquez C, Fernandez-Ulloa J,
Bordon J, et al: Candida albicans
endophthalmitis in brown heroin addicts:
response to early vitrectomy preceded and
followed by antifungal therapy. Clin Infect
Dis 1998; 27:11301133.
49. Aboltins CA, Daffy JR, Allen P: Fungal
endophthalmitis in intravenous drug users
injecting buprenorphine contaminated with
oral Candida species. Med J Aust 2005;
182:427.
50. Thien L le, Fajnkuchen F, Chaine G:
Choriortinite candidosique chez les
toxicomanes. A propos de 2 cas. (Candidal
chorioretinitis in drug addicts: report of 2
cases.) Journal Franais dOphtalmologie
1998; 21:387392.
51. McNeil MM, Lasker BA, Lott TJ, Jarvis WR:
Postsurgical Candida albicans infections
associated with an extrinsically
contaminated intravenous anesthetic agent.
52. Gupta A, Gupta V, Dogra MR, et al: Fungal

endophthalmitis after a single intravenous
administration of presumably contaminated
dextrose infusion fluid. Retina (Philadelphia)
2000; 20:262268.
53. Weishaar PD, Flynn HW Jr, Murray TG,
et al: Endogenous Aspergillus
endophthalmitis. Clinical features and
treatment outcomes. Ophthalmology 1998;
105:5765.
54. Essman TF, Flynn HW Jr, Smiddy WE, et al:
Treatment outcomes in a 10-year study of
endogenous fungal endophthalmitis.
Ophthalmic Surg Lasers 1997; 28:18594.
55. Fahey D K, Fenton S, Cahill M, Acheson
RW: Candida endophthalmitis: a diagnostic
dilemma. Eye (London) 1999; 13:596598.
56. Ohnishi Y, Tawara A, Murata T, et al:
Postmortem ndings two weeks after oral
treatment for metastatic Candida
endophthalmitis with fluconazole.
57. Klein A, Haller-Schober EM, Faulborn J:
Candida-Endophthalmitis bei
Drogenabusus: eine
Patientendemonstration. (Candida
endophthalmitis in drug abuse: a case
report). Ophthalmologe 2000; 97:619622.
58. Vose M, Beatty S, Charles SJ: Candida
endophthalmitis: an unusual complication
of prolonged intravenous access. Postgrad
Med J 2001; 77:119120.
59. Donahue SP, Hein E, Sinatra RB: Ocular
involvement in children with candidemia.
Am J Ophthalmol 2003; 135:886887.
60. Baley JE, Ellis FJ: Neonatal candidiasis:
ophthalmologic infection. Semin Perinatol
2003; 27:401405.
61. Bagnoud M, Baglivo E, Hengstler J, et al:
Endophtalmies fongiques endognes:
rsultats aprs traitement anti-mycosique
associ ou non une vitrectomie.
(Endogenous fungal endophthalmitis:
results after anti-fungal treatment with or
without vitrectomy). Klinische Monatsbltter
fr Augenheilkunde 2001; 218:398400.
62. Crump JR, Elner SG, Elner VM, Kauffman
CA: Cryptococcal endophthalmitis: case
report and review. Clin Infect Dis 1992;
14:10691073.
63. Blachie JD, Danta G, Sorrell T, Collignon P:
Ophthalmological complications of
cryptococcal meningitis. Clin Exp Neurol
1985; 21:263270.
64. Doft BH, Curtin VT: Combined ocular
infection with cytomegalovirus and
cryptococcosis. Arch Ophthalmol 1982;
100:18001803.
65. Denning DW, Armstrong RW, Fishman M,
Stevens DA: Endophthalmitis in a patient
with disseminated cryptococcosis and
AIDS who was treated with itraconazole.
Rev Infect Dis 1991; 13:11261130.
66. Sheu SJ, Chen YC, Kuo NW, et al:
Endogenous cryptococcal endophthalmitis.
67. Johnston SR, Corbett EL, Foster O, et al:
Raised intracranial pressure and visual
complications in AIDS patients with
cryptococcal meningitis. J Infect 1992;
24:185189.
68. Rex JH, Larsen RA, Dismukes WE, et al:
Catastrophic visceral loss due to
Cryptococcus neoformans meningitis.
Medicine 1993; 72:207224.
69. Gonzales CA, Scott IU, Chaudhry NA, et al:
Endogenous endophthalmitis caused by
Histoplasma capsulatum var. capsulatum. A
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
case report and literature review.

Macular Photocoagulation Study Group:
Argon laser photocoagulation for ocular
histoplasmosis. Results of a randomized
trial. Arch Ophthalmol 1983; 101:1347.
Sanglard D, Hube B, Monod M, et al: A
triple deletion of the secreted aspartyl
proteinase genes SAP4, SAP5, and SAP6
of Candida albicans causes attenuated
virulence. Infect Immun 1997; 65:3539.
Ghannoum MA: Potential role of
phospholipases in virulence and fungal
pathogensis. Clin Microbiol Rev 2000;
13:122143.
Naglik JR, Challacombe SJ, Hube B:
Candida albicans secreted aspartyl
proteinases in virulence and pathogenesis.
Microbiol Molec Biol Rev 2003; 76:400.
Kathary MH, Chase T Jr, MacMillan JD:
Correlation of elastase production by some
strains of Aspergillus fumigatus with ability
to cause pulmonary invasive Aspergillus in
mice. Infect Immun 1984; 43:320325.
Edwards JE, Montgomerie JZ, Ishida K,
et al: Experimental hematogenous
endophthalmitis due to Candida: species
variation in ocular pathogenicity. J Infect
Dis 1977; 135:294297.
Lance SE, Friberg TR, Kowalski RP:
Aspergillus flavus endophthalmitis and
retinitis in an intravenous drug abuser. A
therapeutic success. Ophthalmology 1988;
95:947949.
Roney P, Barr CC, Chun CH, Raff MJ:
Endogenous aspergillus endophthalmitis.
Rev Infect Dis 1986; 8:955958.
Riddell J IV,McNeil SA, Johnson TM, et al:
Endogenous Aspergillus endophthalmitis:
report of 3 cases and review of the
literature. Medicine (Baltimore) 2002;
81:311320.
Gaudio PA, Gopinathan U, Sangwan V,
Hughes TE: Polymerase chain reaction
based detection of fungi in infected
corneas. Br J Ophthalmol 2002;
86:755760.
Anand AR, Madhavan HN, Neelam V, Lily
TK: Use of polymerase chain reaction in
the diagnosis of fungal endophthalmitis.
Ferrer C, Colom F, Frass S, et al:
Detection and identication of fungal
pathogens by PCR and by ITS2 and 5.8S
ribosomal DNA typing in ocular infections.
J C Microbiol 2001; 39:28732879.
Hidalgo JA, Alangaden GJ, Eliott D, et al:
Fungal endophthalmitis diagnosis by
detection of Candida albicans DNA in
intraocular fluid by use of a speciesspecic polymerase chain reaction assay.
J Infect Dis 2000; 181:11981201.
Jaeger EEM, Carroll NM, Choudhury S,
et al: Rapid detection and identication of
Candida, Aspergillus, and Fusarium
species in ocular samples using nested
PCR. J C Microbiol 2000; 38:29022908.
Robin JB, Arffa RC, Avni I, Rao NA: Rapid
visualization of three common fungi using
fluorescein-conjugated lectins. Invest
Liu K, Howell DN, Perfect JR, Schell WA:
Morphologic criteria for the preliminary
identication of Fusarium, Paecilomyces,
and Acremonium species by histopathology.
Am J Clin Pathol 1998; 109:4554.
ODay DM, Akrabawi PL, Head WS, Ratner
HB: Laboratory isolation techniques in
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98.
99.
100.
101.
102.
103.
human and experimental fungal infections.

Wilson LA, Sexton RR: Laboratory
diagnosis in fungal keratitis. Am J
Sharma S, Kunimoto DY, Gopinathan U,
et al: Evaluation of corneal scraping smear
examination methods in the diagnosis of
bacterial and fungal keratitis: a survey of
eight years of laboratory experience.
Cornea 2002; 21:643647.
Robin JB, Chan R, Rao NA, et al:
Fluorescein-conjugated lectin visualization
of fungi and acanthamoeba in infectious
keratitis. Ophthalmology 1989; 96:1198.
Pierce AM, Pierce HD, Unrau AM,
Oehlschlager AC: Lipid composition and
polyene antibiotic resistance of Candida
albicans mutants. Biochem Cell Biol 1978;
56:135.
National Committee for Clinical Laboratory
Standards: Reference method for Broth
dilution antifungal susceptibility testing of
yeasts; approved standard M27-A2.
Wayne, Pa: National Committee for Clinical
Laboratory Standards; 2002(a).
Standards: Reference method for Broth
dilution antifungal susceptibility testing of
lamentous fungi; approved standard
M38-A. Wayne, Pa: National Committee for
Clinical Laboratory Standards; 2002(b).
Standards: Method for antifungal disk
diffusion susceptibility testing of yeasts;
approved guideline M44-A. Wayne, Pa:
Standards; 2004.
Jones BR: Principles in the management of
oculomycosis. Am J Ophthalmol 1975;
79:719751.
Panda A, Ahuja R, Biswas NR, et al: Role
of 0.02% polyhexamethylene biguanide
and 1% povidone iodine in experimental
Aspergillus keratitis. Cornea 2003;
22:138141.
Steinbach WJ, Schell WA, Miller JL, Perfect
JR: Scedosporium prolicans osteomyelitis
in an immunocompetent child treated with
voriconazole and caspofungin, as well as
locally applied polyhexamethylene
biguanide. J Clin Microbiol 2003;
41:39813985.
Foster CS, Stefanygzyn M: Intraocular
penetration of miconazole in rabbits. Arch
Ophthalmol 1979; 97:17031706.
Savani DV, Perfect JR, Cobo LM, Durack
DT: Penetration of new azole compounds
into the eye and efcacy in experimental
Candida endophthalmitis. Antimicrob
Ishibashi Y, Matsumoto T: Oral
ketoconazole therapy for experimental
Candida albicans keratitis in rabbits.
Sabouraudia 1984; 22:323330.
Foster CS: Ocular toxicity of topical
antifungal agents. Arch Ophthalmol 1980;
99:10811084.
Bell RN, Ritchey JP: Subconjunctival
nodules after amphotericin B injection.
Klont RR, Eggink CA, Rijs AJMM, et al:
Successful treatment of Fusarium keratitis
with cornea transplantation and topical and
systemic voriconazole. Clin Infect Dis 2005;
40, 12, 1869, e110e112.
Hariprasad SM, Mieler WF, Holz ER, et al:
Determination of vitreous, aqueous, and
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
115.
116.
117.
118.
119.
120.
121.
plasma concentration of orally administered

voriconazole in humans. Arch Ophthalmol
2004; 122:4247.
Shah KB, Wu TG, Wilhelmus KR, Jones
DB: Activity of voriconazole against corneal
isolates of Scedosporium apiospermum.
Cornea 2003; 22:3336.
Nulens E, Eggink C, Rijs AJMM, et al:
Keratitis caused by Scedosporium
apiospermum successfully treated with a
cornea transplant and voriconazole. J Clin
Microbiol 2003; 41:22612264.
Hernndez Prats C, Llinares Tello F, Burgos
San Jos A, et al: Voriconazole in fungal
keratitis caused by Scedosporium
apiospermum. Ann Pharmacother 2004;
38:414417.
Prajna NV, John RK, Nirmalan PK, et al:
A randomised clinical trial comparing 2%
econazole and 5% natamycin for the
treatment of fungal keratitis. Br J
Ophthalmol 2003; 87:12351237.
Goldblum D, Frueh BE, Sarra GM, et al:
Topical caspofungin for treatment of
keratitis caused by Candida albicans in a
rabbit model. Antimicrob Agents
Chemother 2005; 49:13591363.
Jones DB: Therapy of postsurgical fungal
85:357373.
Jones DB, Green MT, Osato MS, et al:
Endogenous Candida albicans
endophthalmitis in the rabbit. Arch
Ophthalmol 1981; 99:21822187.
Fisher JF: Penetration of amphotericin B
into the human eye. J Infect Dis 1983;
147:164165.
Denning DW, Stevens DA: Antifungal and
surgical treatment of invasive aspergillosis:
review of 2,121 published cases. Rev Infect
Dis 1990; 12:11471201.
Perraut LE Jr, Perraut LE, Bleiman B, Lyons
J: Successful treatment of Candida
albicans endophthalmitis with intravitreal
amphotericin B. Arch Ophthalmol 1981;
99:15651567.
Stern GA, Fetkenhour CL, OGrady RB:
Intravitreal amphotericin B treatment of
Candida endophthalmitis. Arch Ophthalmol
1977; 95:8993.
Axelrod AJ, Peyman GA, Apple DJ: Toxicity
of intravitreal injection of amphotericin B.
Walsh JA, Halft MH, Miller MH, et al:
Ocular penetration of 5-flucytosine. Invest
Robertson DM, Riley FC, Hermans PE:
Endogenous Candida oculomycosis: report
of two patients treated with flucytosine.
Medoff G, Comfort M, Kobayashi GS:
Synergistic action of amphotericin B and
5-flurocytosine against yeast-like
organisms (35943). Proc Soc Exp Biol Med
1971; 138:571574.
Blumenkranz MS, Stevens DA: Therapy of
endogenous fungal endophthalmitis:
miconazole or amphotericin B for
coccidioidal and candidal infection. Arch
Ophthalmol 1980; 98:12161220.
Maalouf T, Schmitt Cl, Crance J, et al:
Endophtalmie endogne Aspergillus:
propos dun cas. (Endogenous Aspergillus
endophthalmitis: a case report). Journal
Franais dOphtalmologie 2000;
23:170173.
Schelenz S, Goldsmith DJA: Aspergillus
endophthalmitis: an unusual complication
of disseminated infection in renal transplant

patients. J Infect 2003; 47:336343.
122. Garbino J, Ondrusova A, Baligvo E, et al:
Successful treatment of Paecilomyces
lilacinus endophthalmitis with voriconazole.
Scand J Infect Dis 2002; 34:701703.
123. Blzquez R, Losada M, Menasalvas A,
et al: Recurrent post-traumatic Scytalidium
dimidiatum fungal endophthalmitis. Clin
Microbiol Newsletter 2004; 26:4446.
124. Durand ML, Kim IK, DAmico DJ, et al:
Successful treatment of Fusarium
endophthalmitis with voriconazole and
Aspergillus endophthalmitis with
voriconazole plus caspofungin. Am J
Ophthalmol 2005; 40:552554.
125. Varma D, Thaker HR, Moss PJ, et al: Use
of voriconazole in Candida retinitis. Eye
2005; 19:485487.
126. Gao H, Pennesi ME, Shah K, et al:
Intravitreal voriconazole: an electroretinographic and histopathologic study. Arch
Ophthalmol 2004; 122:16871692.
127. Stern GA, Okumoto M, Smolin G:
Combined amphotericin B and rifampin
treatment of experimental C. albicans
keratitis. Arch Ophthalmol 1979;
79:721722.
127a. Lou P, Kazdan J, Bannatyne RM, Cheung
R: Successful treatment of Candida
endophthalmitis with a synergistic
combination of amphotericin B and
rifampin. Am J Ophthalmol 1977; 83:1215.
128. Lou P, Kazdan J, Bannatyne RM, Cheung
129. Mora-Duarte J, Betts R, Rotstein C, et al:
Comparison of caspofungin and
amphotericin B for invasive candidiasis.
N Engl J Med 2002; 347:20202029.
130. Sarria JC, Bradley JC, Habash R, et al:
Candida glabrata endophthalmitis treated
successfully with caspofungin. Clin Infect
Dis 2005; 40:5, e46e48.
131. Breit SM, Hariprasad SM, Mieler WF, et al:
Management of endogenous fungal
caspofungin. Am J Ophthalmol 2005;
139:135140.
132. Gauthier GM, Nork TM, Prince R, Andes D:
Subtherapeutic ocular penetration of
caspofungin and associated treatment
failure in Candida albicans endophthalmitis.
Clin Infect Dis 2005; 41:3, e27e28.
133. Snip RC, Michels RG: Pars plana
vitrectomy in the management of
endogenous Candida endophthalmitis.
134. Huang K, Peyman GA, McGetrick J:
Vitrectomy in experimental endophthalmitis.
Part I. Fungal infection. Ophthalmic Surg
1979; 10:84.
135. Economou-Stamatelopoulou C,
Roussopoulos GP, Prouskas JC,
Apostolopoulos M: Antifungal activity of
intraocularly used liquids against
Aspergillus. Ophthalmologica 2004;
218:323327.
136. Spirn MJ, Roth DB, Yarian DL, Green SN:
Postoperative fungal endophthalmitis
caused by Trichosporon beigelii resistant to
amphotericin B. Retina (Philadelphia) 2003;
23:404405.
137. Mootha VV, Schluter ML, Das A: Intraocular
hemorrhages due to warfarinfluconazole
drug interaction in a patient with presumed
CHAPTER 15
167
MICROBIOLOGY
SECTION 3
Candida endophthalmitis. Arch Ophthalmol

2002; 120:9495.
138. Leck AK, Thomas PA, Hagan M, et al:
Aetiology of suppurative corneal ulcers in
Ghana and south India, and epidemiology
of fungal keratitis. Br J Ophthalmol 2002;
86:12111215.
139. Narang S, Gupta A, Gupta V, et al: Fungal
endophthalmitis following cataract surgery:
clinical presentation, microbiological
spectrum, and outcome. Am J Ophthalmol
2001; 132:609617.
140. Smith JR, Chee Soon Phaik: Endogenous
Aspergillus endophthalmitis occurring in a
168
child with normal immune function. Eye

(London) 2000; 14:670671.
141. Tsai CC, Chen SJ, Chung YM, et al:
Postpartum endogenous Candida
endophthalmitis. J Formos Med Assoc
2002; 101:432436.
142. Sato Y, Miyasaka S, Shimada H: Prognosis
of endogenous fungal endophthalmitis and
utility of Ishibashis classication. Jpn J
Ophthalmol 2001; 45:181186.
143. Weissgold DJ, Orlin SE, Sulewski ME, et al:
Delayed-onset fungal keratitis after
105:258262.
144. Sacho H, Stead KJ, Klugman KP, Lawrence

A: Infection of the human orbit by
Aspergillus stromatoides. Mycopathologia
1987; 97:9799.
145. Gass JDM: Ocular manifestations of acute
mucormycosis. Arch Ophthalmol 1961;
65:226.
146. Van Burik JA, Hare RS, Holomon HF, et al:
Posaconazole is effective as salvage
therapy in zygomycosis: a retrospective
summary of 91 cases. Clin Infect Dis 2006;
42:61.
CHAPTER
16
Ocular Virology
James Chodosh
INTRODUCTION
Viruses are obligate intracellular pathogens without the capacity
to replicate outside the host. They may cause clinically evident
infection, establish latent infections with or without signicant
disease, or less commonly, induce encephalitis and other lifethreatening illnesses. This chapter focuses on the general
description of viruses and elucidates common mechanisms
relevant to ocular viral pathogenesis.
DESCRIPTION AND CLASSIFICATION

Viruses are small (10400 nm in diameter) infectious units
each consisting of a nucleic acid genome and a protein capsid
shell. Some virus families also express an external lipid envelope. Viruses lack any independent means of energy metabolism, molecular biosynthesis, or replication. Viral genes are
transcribed and viral progeny produced only inside a permissive
host cell. The existence of viruses as distinct infectious
organisms was rst suggested by early experiments in which
specic infections were transmitted experimentally by a ltrate
of secretions from an infected animal using lter pore sizes
small enough to exclude bacteria.1 In the absence of detailed
knowledge of viruses beyond their associated clinical syndromes, initial schemes of viral categorization grouped human
viruses by the affected organ or other clinical criteria. Thus, all
viruses associated with hepatitis were considered together. We
now know that the hepatitis viruses are diverse.
In 1966, the International Committee on Nomenclature of
Viruses (ICNV), later to become the International Committee
on Taxonomy of Viruses (ICTV), began to classify the myriad
of different viruses into groups. In generating a taxonomy of
viruses, the ICTV considers virus morphology, physical properties, nucleic acid type and strandedness, physical state of the
genome, proteins expressed, antigenic properties, and serologic
cross-reactivity, as well as biologic effects of infection.2 Viruses
are then classied broadly by the type of nucleic acid, its strandedness, and if single-stranded, whether positive- or negativesense, and by the presence or absence of an external lipid bilayer
envelope (Table 16.1). In most cases, classication by ultrastructural appearance correlates well. For example, the eight
human herpes viruses so far identied all have an identical
electron microscopic appearance and a high degree of genomic
homology.3 Viral culture with negative staining transmission
electron microscopy to directly examine virus morphology and
size, and thin section electron microscopy of infected tissues to
directly observe viruses during viral infection remain timehonored means of identifying previously unknown viruses.
By agreement, virus orders are designated by the sufx
-virales, families by -viridae, subfamilies by -virinae, and genera
TABLE 16.1. Classication of Virus Families by Nucleic Acid

Type and Strandedness, and Presence of an Envelope
Examples
RNA Viruses
Single-stranded, positive sense,
nonenveloped
Astroviridae, Caliciviridae,
Picornaviridae
Single-stranded, positive sense,

enveloped
Coronaviridae, Flaviviridae,
Retroviridae, Togaviridae
Single-stranded, negative sense,

enveloped
Arenaviridae, Bunyaviridae,
Filoviridae, Orthomyxoviridae,
Paramyxoviridae, Rhabdoviridae
Double-stranded, positive sense,

nonenveloped
Birnaviridae, Reoviridae
DNA Viruses
Single-stranded, nonenveloped
Circoviridae, Parvoviridae
Double-stranded, nonenveloped
Adenoviridae, Papovaviridae
Double-stranded, enveloped
Herpesviridae, Iridoviridae,
Poxviridae
Single/double-stranded,
enveloped
Hepadnaviridae
(Reproduced with permission from reference 38, with permission from Lippincott,
Williams & Wilkins.) Reproduced from Chodosh J, Stroop WG: Introduction to
viruses in ocular disease. In: Tasman W, Jaeger EA, eds. Duanes foundations of
clinical ophthalmology. Philadelphia: Lippincott Williams & Wilkins; 1998:110.
by -virus. Family members share a characteristic morphology,

replicate in a similar fashion, and have relatively conserved
nucleic acid sequences. The recently published VIII report of
the ICTV classied a total of three orders, 73 families, nine
subfamilies, 287 genera and more than 6000 viruses.4 Genbank
contains the genomes of an additional 3142 viruses which have
not yet been classied.5 As more viruses are sequenced, more
viruses will be classied, others possibly reclassied, and phylogenetic relationships claried. In the near future, rapid
identication of viruses by viral microarray analysis,6,7 followed
by sequence analysis may reduce or render unnecessary less
direct and more time-consuming methods of identication and
classication.
VIRAL COMPONENTS
A virion is a single viral infectious unit including nucleic acid,
capsid, and if present, an external envelope. Viral nucleic acids
consist of either RNA or DNA. Viral RNA genome may be
either single- or double-stranded, and in the case of single-
169
SECTION 3
MICROBIOLOGY
stranded viruses, either positive-sense, with the same polarity
as the viral messenger RNA (mRNA), or negative-sense, with
opposite polarity to the viral mRNA. Furthermore, RNA viral
genomes are either segmented, with discrete nucleic acid molecules, or nonsegmented, with all of the genetic information on
a single nucleic acid molecule. Finally, DNA and RNA genomes
exist in either linear or circular (episomal) form. These
characteristics of nucleic acid structure determine much of the
specic mechanics of viral replication.
The viral capsid is a protein shell that surrounds the viral
nucleic acid. The capsid interacts internally with the genome to
stabilize it, protects the genome from the external environment,
and in the case of nonenveloped viruses, expresses on its surface
the necessary ligand(s) for virushost cell binding.8 The viral
capsid proteins also assist in delivery of the viral genome to the
intracellular site of viral replication. Thus, viral capsid structure
is integrally related to key viral functions, in particular, transmission, attachment, and entry into host target cells, but also
virion assembly and egress. The capsid and nucleic acid
together are referred to as the nucleocapsid. Occasionally, as
with herpesviruses, the nucleocapsid is surrounded by an
additional protein layer, the tegument.
Capsid structure is specied by the viral genome, and the
economy of genome size frequently dictates a capsid of repeating protein subunits. Simplicity further dictates that subunits
interact in symmetrical forms with conserved subunit interactions.9 Common capsid structural motifs include the icosahedron with its 20 plane surfaces and the helix.10 Electron
microscopy and X-ray diffraction crystallography, in conjunction with nucleic acid and protein sequencing, can delineate the
components of capsid structure at the molecular level.
A viral envelope surrounds the capsid of some virus families.
The envelope consists of viral genome-encoded glycoproteins
and a few host cell proteins embedded in a host cell-derived
lipid bilayer.11 Viral glycoproteins act as ligands for receptors on
host targets, as well as antigens for neutralizing antibodies
directed against the virus. In the initial stages of infection,
envelope glycoproteins mediate attachment of the virus to its
receptor on the host cell surface and fusion of the viral envelope
with the host cell membrane. During viral replication, viralencoded glycoproteins are targeted on a molecular level to
specic membranes in the host cell in order to serve as sites of
interaction between the viral nucleocapsid and the host cell
membrane prior to budding. Cell membranes used by enveloped
viruses include the nuclear envelope, endoplasmic reticulum,
Golgi apparatus, and plasma membrane. Polarized epithelial
cells, such as those found at mucosal surfaces, maintain tight
intercellular junctions, and possess biochemically and
morphologically distinct apical and basolateral cell membranes.
Due to differential targeting of viral glycoproteins to apical
versus basolateral membranes, polarized cells typically release
enveloped viruses from either the apical or basolateral cell
surface. Virus shed apically into mucosal secretions such as the
tear lm creates the potential for transmission. Virus shed
basolaterally may infect deeper tissues and/or disseminate.12
Because of the lipid component of their viral envelopes, viruses
such as herpes simplex virus or human immunodeciency virus
are intrinsically vulnerable to damage by ultraviolet light,
detergents, alcohols, and general-use antiseptics. Nonenveloped
viruses such as adenoviruses may be quite resistant to
degradation even under relatively harsh conditions.13
VIRAL RECEPTORS AND VIRAL TROPISM

170
Viral tropisms for specic cell types and tissues require a ligand
on the viral capsid surface (nonenveloped viruses) or envelope
(enveloped viruses) that can bind to a receptor on the target cell.
TABLE 16.2. Selected Ocular Viruses and Their Possible

Receptors
Virus
Host Cell Receptor
Adenovirus type 37
CD4625
EpsteinBarr virus
CD2115,16
Herpes simplex virus
Heparan sulfate26,27
Human cytomegalovirus
Heparan sulfate28
Human papillomavirus
Integrin a622
Influenza virus
Sialic acid29,30
Rhinovirus
ICAM-11719
Vaccinia virus
EGF receptor31
Reproduced from Chodosh J, Stroop WG: Introduction to viruses in ocular

disease. In: Tasman W, Jaeger EA, eds. Duanes foundations of clinical
ophthalmology. Philadelphia: Lippincott Williams & Wilkins; 1998:110.
Viral ligands are typically glycoproteins. Host cell virus

receptors are diverse and may be protein, glycoprotein, lipid, or
carbohydrate.14 Although viral ligandhost cell receptor interaction is essential for adsorption of the virus to the cell surface,
the ligand receptor complex also often mediates subsequent
internalization of the virus and uncoating of the capsid. The
polarized location of the virus receptor on epithelial tissues with
distinct apical and basolateral cell surfaces, and the changes in
receptor expression during cell differentiation largely determine
tissue susceptibility to infection in vivo. For example, virus
receptor expression only on the basolateral surfaces of less
differentiated epithelial cells would permit infection by virus
presented across an underlying basement membrane, but not by
virus present in mucosal fluids or on undamaged skin.12
Viruses presumably evolved the capacity to bind constitutive
host cell membrane components with essential primary cellular
functions (Table 16.2). Therefore, binding of virus to a cell
surface component subverts the natural function of that cellular
molecule. For example, the B lymphocyte receptor for
EpsteinBarr virus is the C3d complement receptor, CD21.15,16
Rhinoviruses bind to intercellular adhesion molecule-1 (ICAM1),1719 present on nasopharyngeal20 and conjunctival21
epithelial cells. Human papillomavirus (HPV) appears to bind
the a6 component of the a6b4 integrin complex.22 Adenovirus
type 2 utilizes the CAR protein for attachment23 and integrins
for internalization,24 while the corneal pathogen adenovirus
type 37 binds the C3b complement binding protein CD46.25
In classic lytic viral infections, virus replication diverts
cellular protein production machinery for the synthesis of viral
proteins. However, before shutdown of host macromolecular
synthesis, the cell may respond to viral infection by upregulation of specic genes. For instance, binding of cytomegalovirus
to cells in vitro stimulates production of protooncogenes.32
Adenovirus binding stimulates the rapid induction of host cellderived proinflammatory cytokines by an intracellular signal
transduction pathway.3335 The cellular function of each host
cell virus receptor likely influences initial molecular responses
to infection.
VIRAL INFECTION AND REPLICATION

Viruses may infect the human host via the placenta and birth
canal, ingestion of breast milk, inhalation of airborne secretions, contaminated food, by insect bite, inadvertent intravascular injections, or intimate and/or sexual contact. Ocular
infection by viruses most often follows direct contact with virus
Ocular Virology
viruses infecting the human eye and adnexa is presented in
Table 16.3.
The ultimate objective of infection for a virus, whether latent
or not, is the generation of viral progeny. The synthesis of viralencoded proteins is essential to the ability of the virus to
replicate and be transmitted, and largely determines the specic
effects of viral infection on the cell. Although differences exist
between enveloped and nonenveloped viruses in the mechanics
of infection, the replicative cycle of all viruses can be divided
into six stages: (1) attachment, (2) penetration, (3) uncoating,
(4) replication, (5) assembly, and (6) release (Figure. 16.1).
externally, either from infected secretions in the birth canal

(herpes simplex virus, HPV), on fomites (adenovirus), or
airborne particles (rhinovirus), or is acquired during viremia
(human cytomegalovirus, measles virus). Other mechanisms
of ocular viral infection include extension from contiguous
adnexal disease (herpes simplex virus), neuronal spread down
trigeminal sensory nerve bers (herpes simplex virus),36 spread
from the upper respiratory tract via the nasolacrimal duct
(rhinovirus), and transplacental passage of infectious virus
(rubella virus). Rarely, ocular infection may disseminate
elsewhere (enterovirus 70).37 A summary and classication of
Virus
Family
Subfamily/Genus
Nucleic Acid
Envelope
Capsid
Ocular Target
Herpes simplex
virus, type 1
(HHV1)
Herpesviridae
Alphaherpesvirinae/
Simplexvirus
dsDNA
Icosahedral
Eyelid
Conjunctiva
Cornea
Trabecular meshwork
Uvea
Retina
Herpes simplex
virus, type 2
(HHV2)
Herpesviridae
Alphaherpesvirinae/
Simplexvirus
dsDNA
Icosahedral
Eyelid
Conjunctiva
Cornea
Trabecular meshwork
Uvea
Retina
Varicella zoster
virus (HHV3)
Herpesviridae
Alphaherpesvirinae/
Varicellovirus
dsDNA
Icosahedral
Eyelid
Conjunctiva
Cornea
Trabecular meshwork
Uvea
Retina
Optic nerve
EpsteinBarr
virus (HHV-4)
Herpesviridae
Gammaherpesvirinae/
Lymphocryptovirus
dsDNA
Icosahedral
Lacrimal gland
Conjunctiva
Cornea
Uvea
Retina
Optic nerve
Human
cytomegalovirus
(HHV5)
Herpesviridae
Betaherpesvirinae/
Cytomegalovirus
dsDNA
Icosahedral
Retina
Optic nerve
Human herpes
virus 6 (HHV6)
Herpesviridae
Betaherpesvirinae/
Roseolovirus
dsDNA
Icosahedral
Retina
Human herpes
virus 7 (HHV7)
Herpesviridae
Betaherpesvirinae/
Roseolovirus
dsDNA
Icosahedral
Human herpes
virus 8 (HHV8)
Herpesviridae
Gammaherpesvirinae
dsDNA
Icosahedral
Conjunctiva
(Kaposis sarcoma)
Adenovirus
Adenoviridae
Mastadenovirus
dsDNA
Icosahedral
Conjunctiva
Cornea
HPV
Papovaviridae
Papillomavirus
dsDNA
Icosahedral
Eyelid
Conjunctiva
Cornea
Smallpox (variola)
virus
Poxviridae
Orthopoxvirus
dsDNA
Complex
Eyelid
Conjunctiva
Cornea
Uvea
Optic nerve
Vaccinia virus
Poxviridae
Orthopoxvirus
dsDNA
Complex
Eyelid
Conjunctiva
Cornea
Molluscum
contagiosum virus
Poxviridae
Molluscipoxvirus
dsDNA
Complex
Eyelid
Conjunctiva
Cornea
Continued
CHAPTER 16
TABLE 16.3. Classication Table of Viruses Affecting the Human Eye
171
MICROBIOLOGY
SECTION 3
TABLE 16.3. Classication Table of Viruses Affecting the Human EyeContinued

Virus
Family
Subfamily/Genus
Nucleic Acid
Envelope
Capsid
Ocular Target
Orf virus
Enterovirus(es)
(includes Poliovirus,
Coxsackievirus,
Echovirus,
Enterovirus)
Poxviridae
Picornaviridae
Parapoxvirus
Enterovirus
dsDNA
ssRNA
Complex
Icosahedral
Eyelid
Conjunctiva
Cornea
Rhinovirus
Picornaviridae
Rhinovirus
ssRNA
(+)
Icosahedral
Conjunctiva
Rubella virus
Togaviridae
Rubrivirus
ssRNA
(+)
Icosahedral
Cornea
Uvea
Lens
Trabecular meshwork
Retina
Globe
Alphavirus/Flavivirus
(encephalitis, encephalomyelitis, yellow
fever, dengue viruses)
Togaviridae
Rubrivirus
ssRNA
(+)
Icosahedral
Conjunctiva
Influenza virus
Orthomyxoviridae
Influenzavirus
(A, B, C)
ssRNA
()
Helical
Lacrimal gland
Conjunctiva
Episclera
Cornea
Uvea
Retina
Optic nerve
Cranial nerves
Human coronavirus
Coronaviridae
Coronavirus
ssRNA
(+)
Helical
Conjunctiva
Newcastle disease
virus
Paramyxoviridae
Paramyxovirus
ssRNA
()
Helical
Conjunctiva
Cornea
Parainfluenzavirus(es)
Paramyxoviridae
Paramyxovirus
ssRNA
()
Helical
Conjunctiva
Respiratory syncitial
virus
Paramyxoviridae
Pneumovirus
ssRNA
()
Helical
Conjunctiva
Mumps virus
Paramyxoviridae
Paramyxovirus
ssRNA
()
Helical
Lacrimal gland
Conjunctiva
Sclera
Cornea
Uvea
Optic nerve
Cranial nerves
Measles (rubeola)
virus
Paramyxoviridae
Morbillivirus
ssRNA
()
Helical
Conjunctiva
Cornea
Uvea
Retina
Optic nerve
Cranial nerves
Rift Valley fever

virus
Bunyaviridae
Bunyavirus
ssRNA
()
Coiled
Retina
Colorado tick fever

virus
Reoviridae
Coltivirus
dsRNA
(+/)
Icosahedral
(?: reported to cause

photophobia,
retroocular pain)
Rabies virus
Rhabdoviridae
Lyssavirus
ssRNA
()
Helical
(Transmission via
corneal button)
Human immunodeciency virus
Retroviridae
Lentivirus
ssRNA
(+)
Coiled
Lacrimal gland
Retina
+, enveloped; , Nonenveloped; (+), Ppositive sense RNA genome; (), Nnegative sense RNA genome.
Reproduced from Chodosh J, Stroop WG: Introduction to viruses in ocular disease. In: Tasman W, Jaeger EA, eds. Duanes foundations of clinical ophthalmology.
Philadelphia: Lippincott Williams & Wilkins; 1998:110.
172
Ocular Virology
RNA viruses
Attachment
Attachment
Penetration
Penetration
Uncoating
Uncoating
Early
transcription
Translation
Replication
Late
transcription
Adapted from Chodosh J, Stroop WG: Introduction to

viruses in ocular disease. In: Tasman W, Jaeger EA,
eds. Duanes foundations of clinical ophthalmology.
Philadelphia: Williams & Wilkins; 1998:110.
() Sense
(+) Sense
Transcription
Translation
Translation
Proteins
CHAPTER 16
DNA viruses
FIGURE 16.1 Representations of the stages of

RNA and DNA virus replication. Attachment of
virus to susceptible cells is followed by
penetration and uncoating. Most DNA viruses
undergo transcription, replication, and
assembly in the nucleus. The DNA virus shown
is released from the cell by lysis. Most RNA
viruses replicate in the cytoplasm. The dashed
line illustrates the transcription pathway of the
negative-sense viruses, and the solid line
indicates the pathway taken by the positivesense RNA viruses. The RNA virus shown is
released from the cell by budding through the
plasma membrane.
Proteins
Replication
Translation
Capsid
proteins
Translation
Assembly
Capsid
proteins
Release
Following adsorption to the host cell receptor, penetration

occurs by endocytosis or translocation, or in the case of
enveloped viruses, fusion of the envelope with the host plasma
membrane. Virus capsid components play an active role in
transport of the virus into the cell. Uncoating, or shedding of
capsid components, typically occurs in the cell cytoplasm.
Replication takes place in the nucleus for most DNA viruses
and in the cytoplasm for most RNA viruses. Mechanisms of
viral replication are summarized in Figure 16.2. Assembly of
the virus, the process by which capsid is added to newly
replicated genome, typically occurs in the cytoplasm. Release of
virus from the cell occurs by budding or cell lysis.
Transcription of viral nucleic acid to produce the enzymes
and structural proteins necessary for replication varies with
the type of viral genome. With the exception of the positivesense single-stranded picornaviruses, alphaviruses, and flaviviruses, it is necessary to rst transcribe an mRNA. DNA
Assembly
Release
viruses that replicate in the cell nucleus utilize cell-derived

polymerases. Otherwise, generation of a viral-encoded RNA
polymerase is required. Lastly, because eukaryotic host cells
do not recognize internal initiation sites within mRNA
molecules, posttranslational modications of viral proteins by
cellular or viral enzymes are often used to produce the individual proteins necessary for replication and maturation of
the virion.
Assembly of infectious virus and subsequent release of virus
from the cell are tightly linked and largely determine the outcome of infection. The assembly of nonenveloped viruses in the
cell nucleus or cytoplasm typically exposes the cell to capsid
components that may inhibit cell function and cause cell death.
To acquire envelopes, viruses encode proteins for insertion into
host cell membranes that then act as binding targets for
immature virions. Egress of the virus via budding may itself
lead to cell lysis, as with herpesviruses.
173
MICROBIOLOGY
dsDNA
Positive-sense RNA
Early transcription
+ Strand RNA
Negative-sense RNA
Retrovirus
Early transcription
Early transcription
Strand RNA
tRNA primer
+ Strand RNA
Early mRNAs
Translated by
Reverse
host ribosomes
Early proteins
(
) initiate
SECTION 3
genome synthesis
Replication
Proteins needed
transcription
+ Strand RNA
for replication
and encapsidation
Transcription
Translated by
RNA: DNA
host ribosomes
duplex
+ Strand RNA
Proteins needed
for replication
Replication
Strand RNA
and encapsidation
Replication
+ Strand RNA
dsDNA
Late transcription
provirus
Integration into host DNA

Replication
mRNAs
Progeny + strand
genome
Capsid proteins
FIGURE 16.2 Representations of viral transcription and replication strategies. dsDNA viruses: dsDNA virus early mRNAs are transcribed from
separate promoters (two such transcripts are shown). The mRNA is translated in the cytoplasm and the proteins are returned to the nucleus.
Replication involves binding of early-produced transcriptases to the genome; new DNA strands are synthesized by semi-conservative strand
displacement (as illustrated) or discontinuous mechanisms. Late transcription follows DNA replication and involves transcription of mRNAs
encoding structural proteins. Positive-sense RNA viruses: the RNA genome is directly translated by host ribosomes, producing the proteins
needed for replication. Transcription of the nascent positive-sense RNA by genome-encoded RNA-dependent RNA transcriptase produces a
negative-sense RNA, which serves as a template for synthesis of new genomes. Negative-sense RNA viruses: Negative-sense RNA viruses carry
RNA-dependent RNA polymerase in the virus particle, which transcribes the negative-sense genome into positive-sense molecules. These are
translated into the proteins needed for replication and encapsidation. The positive-sense molecules also serve as templates for generation of
new negative-stranded genomes. Retroviruses: Retroviruses carry reverse transcriptase, which converts the single-stranded RNA genome into a
circular, double-stranded DNA proviral molecule. Transcription of the rst strand of DNA is initiated at the tRNA primer; circularization of the RNA
allows transcription to proceed along the length of the RNA strand. The genomic RNA is degraded by the RNAse property of reverse
transcriptase, and the second DNA strand is synthesized using the rst DNA strand as a template. The fully dsDNA circular molecule integrates
into host chromosomal DNA; host DNA flanking sequences are indicated by the broken lines. Replication involves transcription of mRNAs
encoding viral proteins and transcription of full-length, positive-sense RNA from the integrated provirus.
174
Adapted from Chodosh J, Stroop WG: Introduction to viruses in ocular disease. In: Tasman W, Jaeger EA, eds. Duanes foundations of clinical ophthalmology.
Philadelphia: Williams & Wilkins; 1998:110.
Clinical illness is inadequate as a criterion to assess viral

infection, because viral infection may be subclinical or essentially asymptomatic. Viruses cause disease by a variety of mechanisms, including altered cellular metabolism due to viral gene
products, altered host gene expression mediated by interactions
between viral proteins and the host genome, and host immune
response to viral infection of the cell. The end results of viral
infection range from frank destruction of host tissues, disrupted
function on cellular, tissue, organ, and/or systemic levels,
recurrent disease due to intermittent viral expression over time
from latently infected cells, neoplastic transformation, and
immunologically mediated disease.
VIRAL DIAGNOSTICS
Ocular viral infections are often diagnosed on clinical criteria.
However, when atypical, particularly severe, or when a correct
diagnosis will alter subsequent treatment, laboratory investigation may be indicated. Multiple approaches are available
to achieve laboratory conrmation of a specic viral entity, but
conrmation of ocular viral infections depends on the clinician
to obtain specimens at appropriate times during the course
of infection and on the proper specimen handling after
collection.3840 Commonly used techniques to identify viral
pathogens include viral culture, microscopy, antigen detection,
nucleic acid detection, and serology. Communication between
the physician and laboratory staff regarding the differential
diagnosis generally improve the likelihood of identifying a viral
pathogen.
Viral culture followed by direct or indirect immunoflourescent antigen detection remains the gold standard of
virus detection against which all other methods are compared,
although in cases of latent viral infection with intermittent
virus shedding, isolation of a virus may be misleading with
regards to causality of disease.4143 Skin, conjunctival, or
corneal scrapings, or intraocular fluids (in exceptional cases) are
obtained during the acute phase of infection. In the laboratory,
the inoculum is transferred onto the appropriate cell line for
growth of the virus. The choice of cell type depends on the virus
one wishes to cultivate. Any given cell line is generally capable
of supporting the growth of only a limited range of viruses.
Clinical laboratories typically grow viruses in primary, diploid,
or heteroploid continuous cell lines derived from human
cancers or animals. When the virus is inoculated onto a
susceptible cell line, it produces a characteristic change in the
host cell, termed cytopathic effect (CPE). The specic appearance of CPE varies between virus families and may allow a
presumptive identication of the virus. Although rapidly
growing viruses such as herpes simplex virus can produce a
detectable CPE within a day or two, others, such as CMV,
rubella, and some adenoviruses, can take 14 weeks. Once CPE
is evident, the virus is usually identied with direct or indirect
immunofluorescence techniques. When the identity of the virus
is unknown or the CPE is uncharacteristic, morphologic
examination with electron microscopy may be helpful. A
relatively recent innovation in viral cell culture, the shell vial
technique, allows the rapid identication of viruses. Cultures
are centrifuged at low speed for 13 days and stained by direct
immunofluorescence for viral antigen, prior to development
of CPE.
Other methods of virus identication include microscopic
examination of scrapings or tissue samples with Giemsa stain,
electron microscopy, or with antigen detection systems using
immunofluorescence or immunoperoxidase. Cytology may in
some circumstances allow the initial, early recognition of viral
infection. Scrapings from clinical lesions (skin or ocular tissues)
streaked onto a glass microscope slide, xed, and subsequently

stained with hematoxylin and eosin, Tzanck, Giemsa, or
Papanicolaou stain and examined by light microscopy, may
show distinctive inclusions that represent abnormal accumulations of host cellular material caused by the virus-induced
disruption of host cell metabolic activity. Cytology may be
helpful in herpes simplex virus, varicella-zoster virus, CMV,
measles, and rabies infections. Multinucleated giant cells and
ballooning cytoplasm may be observed in herpes simplex,
varicella-zoster, and cytomegaloviral infections and are
characteristic of these human herpes virus infections.
Antigen detection requires the technician to know what virus
is suspected. The immunoperoxidase technique is useful in
laboratories without access to a fluorescent microscope.44
Agglutination tests to detect viral antigens are based on visible
agglutination of particles, such as latex, red blood cells, or
polystyrene, to which virus-specic antibody has been adsorbed.
Agglutination methods are easily performed, but few commercial kits have been shown to detect ocular viral pathogens.
Electron microscopy is limited by the need for large quantities
of virus in the specimen, and is relatively insensitive for clinical
specimens. Solid phase immunoassays are rapid, available,
quantiable, and relatively inexpensive, but have not been
widely adopted for ocular infections. Nucleic acid detection
by hybridization or polymerase chain reaction (PCR) are
promising, in particular those PCR techniques that are able to
detect and differentiate several different viruses in one
experimental run.45 The high sensitivity of PCR is also a
detriment as the technique does not differentiate bystander
viruses, for example herpes simplex shed into the tear lm,
from viral pathogens. In situ hybridization on tissue sections for
viral gene expression within pathologic tissue cells then
becomes the gold standard to prove that an abnormal cell is
actually infected with the virus. The same conclusion may be
obtained by immunohistochemistry for viral antigen.
Serologic tests for virus-specic IgG antibody require patient
sera during both the acute and convalescent periods 24
weeks after the onset of clinical disease at which time the
information may no longer be clinically useful. Increases in
virus-specic IgG may be sufciently robust to assist in
diagnosis of primary infection and in re-infection, but not
in viral reactivation. On the other hand, serology for virusspecic IgM is very useful in primary infections. Therefore,
serology for herpes simplex virus might be useful in children
with recent onset of suspicious but not classic skin and/or
corneal lesions, but herpes simplex virus serology is rarely
informative in older adults in whom the prevalence of herpes
simplex virus-specic IgG antibody is high. Unlike IgG, IgM
does not cross the placental barrier. Therefore, in a neonate, the
nding of IgM antibodies indicates infection of the child.46
Intraocular antibody titers can be useful in intraocular
infections when the serology is suggestive of past infection. The
GoldmannWitmer coefcient compares the ratios of pathogenspecic antibody to albumin between intraocular fluid and
serum. A ratio of antibody (eye)/albumin (eye) to antibody
(serum)/albumin (serum) of greater than three indicates
intraocular infection with the specic pathogen.47
CHAPTER 16
Ocular Virology
VIRAL IMMUNOPATHOGENESIS
Virus infections may be suppressed by neutrophils, natural
killer cells, B lymphocyte-derived antibodies, and effector T
lymphocytes.48 Unlike B and T lymphocytes, natural killer cells
can act without antigen specicity or immunologic memory.49
Interferon-stimulated natural killer cells limit the extent of viral
infection early on, before the machinery of acquired antigen-
175
SECTION 3
MICROBIOLOGY
176
specic immunity has fully engaged. Additionally, activated

natural killer cells attack cells with reduced MHC class 1
expression to counter viral evasion of MHC class 1 presentation
(see further ahead). In the eye, chemokine expression by virusinfected host cells induces rapid migration of leukocytes into
virus-infected tissues, but may serve to increase local tissue
damage and lead to reduced vision.50
In patients with secondary immunity, virus-specic antibodies can neutralize free virus in blood or mucosal secretions.
They also mediate cell death of infected cells through
complement-mediated killing and by antibody-dependent cellmediated cytotoxicity. Viral neutralization by antibody depends
on recognition of viral epitopes present on virus surfaces such
as envelope glycoproteins, or, in the case of nonenveloped virus,
surface capsid proteins. High quantities of virusantibody
immune complexes in the blood can induce immune complexmediated immunopathology at distant tissue sites.
CD8+ cytotoxic T lymphocytes (CTLs) typically recognize
viral epitopes in the context of MHC class 1 molecules
expressed on the surface of virus-infected cells, and are critical
to the elimination of virus-infected cells. Nucleated cells
express class 1 molecules, and so any virus-infected cell may be
a target for CTLs. Killing occurs through a directional release
of perforin and granzymes. However, stimulation of T cell
immunity may be accompanied by production of tumor
necrosis factor and other cytokines that have deleterious effects
at local and systemic levels.51
Some viruses possess the means to evade the host immune
system. A herpes simplex virus-encoded protein, ICP47,
successfully competes with antigenic viral peptides for transport
into the endoplasmic reticulum where peptides are loaded
onto the MHC complex.52 Thus, herpes simplex virus-infected
cells can be resistant to CTL lysis. Similarly human
cytomegalovirus-encoded US11 dislocates MHC molecules into
the cytosol where they are degraded by cell proteases.53 These
and other examples of the means by which ocular viruses may
evade the immune system are presented in Table 16.4.
Certain viruses produce homologs of human proteins that
can influence host immunity. EpsteinBarr virus encodes a
homolog of human IL-10.61 Expression of viral IL-10 by infected
cells inhibits interferon-g production and T cell immunity, and
results in enhanced survival of virus-infected cells. Human
herpesvirus 8 encodes a structural homolog of IL-6, suggested
to influence the pathogenesis of Kaposis sarcoma.62 Human
cytomegalovirus encodes a homolog for a human chemokine
receptor, providing the capacity to divert host cell-derived
chemokines and thereby prevent inflammation and viral
clearance.63,64
Pathogenic roles for viruses in a variety of autoimmune
diseases have been suggested.6569 Some viruses stimulate
polyclonal B cell activation and lead to excessive deposition of
immune complexes in sensitive tissues. The altered cytokine
milieu associated with viral infection can stimulate autoreactive
T cells, resulting in inadvertent damage to normal tissues.
Inflammation of immunologically sequestered tissues such as
those present in the central nervous system or the eye in the
mature adult could expose previously hidden epitopes and lead
to local hyperimmunologic responsiveness with devastating
functional consequences.70 Finally, shared antigenic determinants between virus and host can lead in genetically susceptible
individuals to immunologic recognition of self epitopes
(molecular mimicry),71 with immune-mediated damage at
distant and ostensibly normal sites.
Viruses that infect the eye or its adnexa tend to produce
stereotypic pathologic changes in target tissues. Infection of the
eyelid skin by viruses typically induces the formation of vesicles
and ulceration. Infection of the conjunctiva results in increased
TABLE 16.4. Ocular Viruses and Molecular Means by Which

They Evade Host Immunity
Virus
Examples of Immune Escape

Mechanisms
Virus-encoded ICP47 blocks peptide

translocation to MHC class 152
Virus-encoded proteins bind and
neutralize complement components54
Latency in sensory neurons
Human cytomegalovirus
Virus-encoded US11 causes cytosolic

degradation of MHC class 1 heavy
chains53
Viral MHC class I homolog inhibits
NK cell attack55
Latency in glandular tissue
EpsteinBarr virus
Exclusive EBNA-1 expression in type 1

latency reduces recognition by
cytotoxic T lymphocytes56
Adenovirus type 2
Prevents MHC class I transport to cell

surface57
Protection from TNF-mediated
cytolysis58
Influenzavirus
Inhibits cytolysis by interferon59

Antigenic shift and drift
Vaccinia virus
Blocks antiviral effect of interferon60,61
Reproduced from Chodosh J, Stroop WG: Introduction to viruses in ocular

disease. In: Tasman W, Jaeger EA, eds. Duanes foundations of clinical
ophthalmology. Philadelphia: Lippincott Williams & Wilkins; 1998:110.
numbers and size of conjunctival lymphoid follicles along with

the enlargement of corresponding draining lymph nodes. Viral
infection of the corneal epithelium invariably causes punctate
epithelial cytopathic effect evident biomicroscopically as
isolated swollen epithelial cells (punctate epithelial keratitis)
and loss of individual epithelial cells (punctate epithelial
erosions). When extensive, the punctate erosions may coalesce
to form confluent epithelial ulcers (dendritic, dendritiform, and
geographic ulcers).72 Corneal stromal infection results in white
blood cell recruitment to the site of infection,73 with resultant
stromal inltration. Retinal infection leads to retinal necrosis.
Viral encephalitis, encephalomyelitis, and meningitis may lead
to cranial nerve inflammation and dysfunction of vision and/or
extraocular motility.
LATENCY, CARCINOGENESIS, LOSS OF

FUNCTION
Although some virus species may cause self-limited infections
with complete clearance of the virus, others can persist
indenitely in the host.74,75 For example, adenoviruses persist
within nasopharyngeal lymphoid tissue, EpsteinBarr virus in
nasopharyngeal epithelial cells and B lymphocytes, herpes
simplex and varicella zoster viruses in sensory ganglia, and
HPV in skin and mucosal epithelia. An infection is said to be
latent when persistent but not currently productive of infectious
virus. In latent infections, only limited viral gene expression
occurs, and the immune systems response to the few gene
products of the latent virus is absent or altered.76 Interestingly,
latent infection frequently occurs in cell types poorly permissive
for lytic infection by the virus, such as with herpes simplex
virus and sensory neurons, EpsteinBarr virus and B lymphocytes, and HPV and basal skin epithelial cells. Latent
infections also tend to occur in slow-cycling cells. Persistent
infections may consistently or intermittently produce infectious
virus in tears or at skin and mucosal surfaces. When virus is

produced in what had been a latent infection, the infection is
said to be reactivated.
Persistent viral infection of susceptible cells can lead to
malignant transformation. Viral proteins, whether directly
through interaction with the host genome or by interaction
with cellular proteins, can induce transformation of the cell and
loss of senescence. HPV-induced squamous cell carcinoma is an
elegant example of tumor induction by viruses.7779 HPV
tropisms for skin and mucosa derive in part from tissue-specic
gene expression.80 HPV types 6 and 11 are maintained in a
latent state within basal epithelial cells as circular episomes
with very limited viral gene transcription and low copy number.
Early viral gene products stimulate cell growth and lead to a
skin wart or a conjunctival papilloma. As HPV-containing basal
epithelial cells mature and differentiate into supercial
epithelial cells, they become permissive for complete viral gene
expression and produce infectious virus. Carcinomatous
transformation due to HPV-6 or HPV-11 is very rare. In
contrast, HPV-16 and HPV-18 stereotypically integrate their
viral genome into host chromosomal DNA, and this in turn has
been strongly associated with malignant transformation and
squamous cell carcinoma. In the episomal state, transcription
of HPV protooncogenes E6 and E7 is effectively repressed by the
HPV E2 gene product. When HPV genome integrates into host
cell chromosomal DNA, the circular (episomal) viral DNA
molecule breaks at a recombination site within the E2 open
reading frame, resulting in a truncated E2 protein, and
disinhibition of E6 and E7 transcription. The E6 protein binds
to and initiates the degradation of the cellular p53 tumor
suppressor gene product. The E7 protein displaces cellular pRB
from its complex with cellular E2F transcription factor. E2F

then activates transcription of genes that initiate the cell cycle.
Hence, increased cellular levels of E6 and E7 proteins contribute
to the malignant phenotype of HPV-16- and HPV-18-infected
squamous epithelium.
The pathologic consequences of viral infection depend on a
complicated array of factors. The presence of viral receptors
on host cells at a surface exposed to infectious virus, the
permissiveness of the cell to viral gene expression, the capacity
of the host to eliminate the virus as balanced by the damage to
host tissue due to the immune response, and nally the ne
function of the host cell and its tissue, all determine the
functional and anatomic derangements associated with viral
infection. For a virus like herpes simplex, tropic for almost all
ocular tissues, the morbidity of ocular infection varies with the
tissue infected. Herpes simplex virus infection of the
conjunctiva is self-limited and leaves no visual decit, while
infection of the corneal stroma may result in varying degrees of
vision loss, and infection of the retina may result in complete
loss of useful vision. In contrast, HPV ocular tropism is limited
to the conjunctiva, limbus, and eyelid skin. Blinding sequelae of
HPV infection occur with malignant transformation of infected
tissues. As classication of viruses proceeds on a molecular
genetic level, the mechanisms by which viruses infect ocular
cells, destroy critical ocular structures, evade the immune
system, and induce cancer may be better understood.
CHAPTER 16
Ocular Virology
ACKNOWLEDGMENT
The author wishes to acknowledge Thomas J Liesegang, MD for his work
as author of this chapter in the previous (second) edition of this book.
REFERENCES
1. McIntosh K: Diagnostic virology. In: Fields
BN, Knipe DM, Howley PM, et al, eds.
Fields virology. 3rd edn. Philadelphia, PA:
Raven; 1996:401430.
2. Murphy FA: Virus taxonomy. In: Fields BN,
Knipe DM, Howley PM, et al, eds. Fields
virology. 3rd edn. Philadelphia, PA: Raven;
1996:1557.
3. Johnson JE, Chiu W: Structures of virus
and virus-like particles. Curr Opin Struct
Biol 2000; 10:229235.
4. Fauquet CM, Mayo MA, Maniloff J, et al,
eds: Virus taxonomy, VIII Report of the
ICTV. London: Elsevier/Academic Press;
2005.
5. Fauquet CM, Fargette D: International
committee on taxonomy of viruses and the
3,142 unassigned species. Virol J 2005;
2:6473.
6. Striebel HM, Birch-Hirschfeld E, Egerer R,
Foldes-Papp Z: Virus diagnostics on
microarrays. Curr Pharm Biotechnol 2003;
4:401415.
7. Wang D, Urisman A, Liu YT, et al: Viral
discovery and sequence recovery using
DNA microarrays. PLoS Biol 2003; 1:E2.
8. Nicklin SA, Wu E, Nemerow GR, Baker AH:
The influence of adenovirus ber structure
and function on vector development for
gene therapy. Mol Ther 2005; 12:384393.
9. Zlotnick A: Theoretical aspects of virus
capsid assembly. J Mol Recognit 2005;
18:479490.
10. Harrison SC, Shehel JJ, Wiley DC: Virus
structure. In: Fields BN, Knipe DM, Howley
PM, et al, eds. Fields virology. 3rd edn.
Philadelphia, PA: Raven; 1996:5999.
11. Cantin R, Methot S, Tremblay MJ: Plunder

and stowaways: incorporation of cellular
proteins by enveloped viruses. J Virol 2005;
79:65776587.
12. Gan YJ, Chodosh J, Morgan A, Sixbey JW:
Epithelial cell polarization is a determinant
in the infectious outcome of
immunoglobulin A-mediated entry by
EpsteinBarr virus. J Virol 1997;
71:519526.
13. Gordon YJ, Gordon RY, Romanowski E,
Araullo-Cruz TP: Prolonged recovery of
desiccated adenoviral serotypes 5, 8, and
19 from plastic and metal surfaces in vitro.
Ophthalmology 1993; 100:18351839.
14. Knipe DM: Virushost cell interactions. In:
Fields BN, Knipe DM, Howley PM, et al,
eds. Fields virology. 3rd edn. Philadelphia,
PA: Raven; 1996:273199.
15. Frade R, Barel M, Ehlin-Henriksson B,
Klein G: gp140, the C3d receptor of human
lymphocytes, is also the EpsteinBarr virus
receptor. Proc Nat Acad Sci USA 1985;
82:14901493.
16. Nemerow GR, Wolfert R, McNaughton ME,
Cooper NR: Identication and
characterization of the EpsteinBarr virus
receptor on human B lymphocytes and its
relationship to the C3d complement
receptor (CR2). J Virol 1985; 55:347351.
17. Greve JM, Davis G, Meyer AM, et al: The
major human rhinovirus receptor is ICAM-1.
Cell 1989; 56:839847.
18. Staunton DE, Merluzzi VJ, Rothlein R, et al:
A cell adhesion molecule, ICAM-1, is the
major surface receptor for rhinoviruses. Cell
1989; 56:849853.
19. Tomassini JE, Graham D, Dewitt CM, et al:

cDNA cloning reveals that the major group
rhinovirus receptor on HeLa cells is
intercellular adhesion molecule 1. Proc Natl
Acad Sci USA 1989; 86:49074911.
20. Winther B, Greve JM, Gwaltney JM, et al:
Surface expression of intercellular adhesion
molecule 1 on epithelial cells in the human
adenoid. J Infect Dis 1997; 176:523525.
21. Bouchard CS; Lasky JB; Cundiff JE;
Smith BS: Ocular surface upregulation of
intercellular adhesive molecule-1 (ICAM-1)
by local interferon-gamma (IFN-gamma) in
the rat. Curr Eye Res 1996; 15:203208.
22. Evander M, Frazer IH, Payne E, et al:
Identication of the a6 integrin as a
candidate receptor for papillomaviruses.
J Virol 1997; 71:24492456.
23. Bergelson JM, Cunningham JA, Droguett G,
et al: Isolation of a common receptor for
coxsackie B viruses and adenoviruses 2
and 5. Science 1997; 275:13201323.
24. Wickham TJ, Mathias P, Cheresh DA,
Nemerow GR: Integrins alphav beta3 and
alphav beta5 promote adenovirus
internalization but not virus attachment.
Cell 1993; 73:309319.
25. Wu E, Trauger SA, Pache L, et al:
Membrane cofactor protein is a receptor
for adenoviruses associated with epidemic
keratoconjunctivitis. J Virol 2004;
78:38973905.
26. Shieh MT, WuDunn D, Montgomery RI,
et al: Cell surface receptors for herpes
simplex virus are heparin sulfate
proteoglycans. J Cell Biol 1992;
116:12731281.
177
MICROBIOLOGY
27. WuDunn DW, Spear PG: Initial interaction
of herpes simplex virus with cells is
binding to heparan sulfate. J Virol 1989;
63:5258.
28. Comptom T, Nowlin DM, Cooper NR:
Initiation of human cytomegalovirus
infection requires initial interaction with cell
surface heparan sulfate. Virology 1993;
193:834841.
29. Higa HH, Rogers GN, Paulson JC:
Influenza virus hemagglutinins differentiate
between receptor determinants bearing
N-acetyl-N-glycollyl- and N-Odiacetylneuraminic acid groups. Virology
1985; 144:279282.
30. Weis W, Brown J, Cusack S, et al:
Structure of the influenza virus
hemagglutinin complexed with its receptor,
sialic acid. Nature 1988; 333:426431.
31. Eppstein DA, Marsh YV, Schreiber AB,
et al: Epidermal growth factor receptor
occupancy inhibits vaccinia virus infection.
Nature 1985; 318:663665.
32. Albrecht T, Boldogh I, Fons MP: Receptorinitiated activation of cells and their
oncogenes by herpes-family viruses.
J Invest Dermatol 1992; 98:S29S35.
33. Bruder JT, Kovesdi I: Adenovirus infection
stimulates the Raf/MAPK signaling pathway
and induces interleukin-8 expression.
J Virol 1997; 71:398404.
34. Natarajan K, Rajala MS, Chodosh J:
Corneal IL-8 expression following
adenovirus infection is mediated by c-Src
activation in human corneal broblasts.
J Immunol 2003; 170:62346243.
35. Xiao J, Chodosh J: JNK regulates MCP-1
expression in adenovirus type 19-infected
human corneal broblasts. Invest
36. Smith GA, Gross SP, Enquist LW:
Herpesviruses use bidirectional fast-axonal
transport to spread in sensory neurons.
98:34663470.
37. Higgins PG: Enteroviral conjunctivitis and
its neurological complications. Arch Virol
1982; 73:91101.
38. Wilson ML: General principles of specimen
collection and transport. Clin Infect Dis
1996; 22:766777.
39. Lennette DA: Collection and preparation of
specimens for virological examination. In:
Baron EJ, Pfaller MA, Tenover FC,
Yolken RH, eds. Manual of clinical
microbiology. 6th edn. Washington, DC:
ASM Press; 1995:868875.
40. Schwab IR: Diagnostic techniques for
ocular viral infections. In: Tasman W,
Jaeger EA, eds. Duanes foundations of
clinical ophthalmology. Philadelphia, PA:
Raven; 1990:110.
41. Collum LM, Mullaney J, Hillery M, et al:
Two laboratory methods for diagnosis of
herpes simplex keratitis. Br J Ophthalmol
1987; 71:742745.
42. Crosby M: Immunologic tests for ocular
herpes simplex virus. Ophthalmology 1990;
97:694695.
43. Shibata D, Fu YS, Gupta JW, et al:
Detection of human papillomavirus in
normal and dysplastic tissue by the
polymerase chain reaction. Lab Invest
1988; 59:555559.
44. Catalano RA, Webb RM, Smith RS, et al:
A modied immunoperoxidase method for
178
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
rapid diagnosis of herpes simplex I

keratitis. Am J Clin Pathol 1986;
86:102104.
Dabil H, Boley ML, Schmitz TM, Van
Gelder RN: Validation of a diagnostic
multiplex polymerase chain reaction assay
for infectious posterior uveitis. Arch
Ophthalmol 2001; 119:13151322.
Grifths PD, Stagno S, Pass RF, et al:
Congenital cytomegalovirus infection:
diagnostic and prognostic signicance of
the detection of specic immunoglobulin M
antibodies in cord serum. Pediatrics 1982;
69:544549.
Dussaix E, Cerqueti PM, Pontet F, BlochMichel E: New approaches to the detection
of locally produced antiviral antibodies in
the aqueous of patients with endogenous
uveitis. Ophthalmologica 1987;
194:145149.
Doherty PC, Turner SJ: The virus-immunity
ecosystem. Arch Virol Suppl 2005;
19:1732.
Hamerman JA, Ogasawara K, Lanier LL:
NK cells in innate immunity. Curr Opin
Immunol 2005; 17:2935.
Kumaraguru U, Davis I, Rouse BT:
Chemokines and ocular pathology caused
by corneal infection with herpes simplex
virus. J Neurovirol 1999; 5:4247.
Watts TH: TNF/TNFR family members in
costimulation of T cell responses. Annu
Rev Immunol 2005; 23:2368.
York IA, Roop C, Andrews DW, et al:
A cytosolic herpes simplex virus protein
inhibits antigen presentation to CD8+
T lymphocytes. Cell 1994; 77:525535.
Wiertz EJ, Jones TR, Sun L, et al: The
human cytomegalovirus US11 gene
product dislocates MHC class I heavy
chains from the endoplasmic reticulum to
the cytosol. Cell 1996; 84:769779.
Friedman HM, Cohen GH, Eisenberg RJ,
et al: Glycoprotein C of herpes simplex
virus 1 is an inhibitor of the complement
cascade. Nature 1984; 309:633635.
Reyburn HT, Mandelbolm O, Vaels-Gomez M,
et al: The class I MHC homologue of
human cytomegalovirus inhibits attack by
natural killer cells. Nature 1997;
386:514517.
Levitskaya J, Coram M, Levitsky V, et al:
Inhibition of antigen presentation by the
internal repeat region of the EpsteinBarr
virus nuclear antigen-1. Nature 1995;
375:685688.
Burgert H-G, Kvist S: An adenovirus type 2
glycoprotein blocks cell surface expression
of human histocompatibility class 1
antigens. Cell 1985; 41:987997.
Gooding LR, Elmore LW, Tollefson AE, et al:
A 14,700 mw protein from the E3 region of
adenovirus inhibits cytolysis by tumor
necrosis factor. Cell 1988; 53:341346.
Katze MG, Krug RM: Translational control
in influenza virus-infected cells. Enzyme
1990; 44:265277.
Beattie E, Tartaglia J, Paoletti E: Vaccinia
virus encoded elF-2 alpha homolog
abrogates the antiviral effect of interferon.
Virology 1991; 183:419422.
Moore KW, Viera P, Fiorentino DF, et al:
Homology of cytokine synthesis inhibitory
factor (IL-10) to EpsteinBarr virus
gene BCRF1. Science 1990;
248:12301234.
62. Neipel F, Albrecht J-C, Ensser A, et al:

Human herpesvirus 8 encodes a
homolog of interleukin-6. J Virol 1997;
71:839842.
63. Chee MS, Satchwell SC, Preddie E, et al:
Human cytomegalovirus encodes three G
protein-coupled receptor homologues.
Nature 1990; 344:774777.
64. Boomker JM, van Luyn MJ, The TH, et al:
US28 actions in HCMV infection: lessons
from a versatile hijacker. Rev Med Virol
2005; 15:269282.
65. Barnett LA, Fujinami RS: Molecular
mimicry: a mechanism for autoimmune
injury. FASEB J 1992; 6:840844.
66. Kim B, Kaistha SD, Rouse BT: Viruses and
autoimmunity. Autoimmunity 2006; 39:7177.
67. Olson JK, Ercolini AM, Miller SD: A virusinduced molecular mimicry model of
multiple sclerosis. Curr Top Microbiol
Immunol. 2005; 296:3953.
68. Sawada S, Takei M: EpsteinBarr virus
etiology in rheumatoid synovitis.
Autoimmun Rev 2005; 4:106110.
69. Whitton JL, Feuer R: Myocarditis, microbes
and autoimmunity. Autoimmunity 2004;
37:375386.
70. Wildner G, Diedrichs-Mohring M:
Autoimmune uveitis and antigenic mimicry
of environmental antigens. Autoimmun Rev
2004; 3:383387.
71. Fujinami RS, Oldstone MBA: Amino acid
homology between the encephalitogenic
site of myelin basic protein and virus:
mechanism for autoimmunity. Science
1985; 230:10431045.
72. Chodosh J, Miller D, Stroop WG,
Pflugfelder SC: Adenoviral epithelial
73. Oakes JE, Monteiro CA, Cubitt CL,
Lausch RN: Induction of interleukin-8 gene
expression is associated with herpes
simplex virus infection of human corneal
keratocytes but not human corneal
epithelial cells. J Virol 1993; 67:47774784.
74. Rapp F, Cory JM: Mechanisms of
persistence in human virus infections.
Microb Pathog 1988; 4:8592.
75. Hirschhorn K, Gabelman N, Price PM,
Waxman S: Evolutionary signicance of
persistence of latent oncogenic virus
information in vertebrates. Lancet 1973;
1:11581159.
76. Khanna KM, Lepisto AJ, Decman V,
Hendricks RL: Immune control of herpes
simplex virus during latency. Curr Opin
Immunol 2004; 16:463469.
77. Ferenczy A, Franco E: Persistent human
papillomavirus infection and cervical
neoplasia. Lancet Oncol 2002; 3:1116.
78. Munger K, Howley PM: Human
papillomavirus immortalization and
transformation functions. Virus Res 2002;
89:213228.
79. Stoler MH: The virology of cervical
neoplasia: an HPV-associated malignancy.
Cancer J 2003; 9:360367.
80. Steinberg BM, Auborn KJ, Brandsma JL,
Taichman LB: Tissue site-specic enhancer
function of the upstream regulatory region
of human papillomavirus type 11 in
cultured keratinocytes. J Virol 1989;
63:957960.
SECTION 4
PHARMACOLOGY AND TOXICOLOGY

Edited by Mark B. Abelson
CHAPTER
17
Ocular Pharmacokinetics
Denise K. Chun, Aron Shapiro, and Mark B. Abelson
INTRODUCTION
DOSING FREQUENCY
Getting a particular drug to a receptor often requires that it be

administered at a site that is remote from the target, such as by
injection into the blood stream or via oral administration. Ophthalmic delivery for diseases of the frontal eye area is unique in
that medication in the form of an eye drop can be delivered
directly to the diseased tissue. However, diseases of the posterior areas of the eye remain challenging therapeutic targets.
The administered drug must diffuse across several tissues
(absorption), distribute into a variety of tissues and fluids (distribution), be subject to a wide array of metabolizing enzymes
(metabolism), and then be eliminated from the area (elimination).
The study of these processes makes up the eld of pharmacokinetics, which is essential for choosing the appropriate design,
delivery system, and dosing regimen for any therapeutic agent.
Pharmacokinetics describes the quantitative relationship between
the administered dose and dosing regimen and the observed
plasma and/or tissue concentration of the drug as a function of
time. Pharmacodynamics can be dened as the quantitative
relationship between the observed plasma and/or tissue concentration, of the active form of the drug and the pharmacologic
effect or biologic response.1,2 These terms are more loosely
described as what the body does to the drug (pharmacokinetics)
and what the drug does to the body (pharmacodynamics).3
There are unique limitations to classic pharmacokinetic
approaches when designing and evaluating ocular drug therapies.
Numerous factors that affect the bioavailability (i.e., how much
drug was instilled or injected versus how much actually got in) of
ocular drugs are based on constraints imposed by the anatomy of
the eye, as well as by formulations of the drug itself. Nevertheless,
there has been signicant progress in novel topical and vitreoretinal ophthalmic pharmaceuticals during the last couple of
decades.410 Many products have been or are being developed in
this area, including suspensions, ointments, gels, newly formulated
solutions, intravitreal, subconjunctival, and intravenous injectables, liposomes, micro- and nanoparticles, iontophoretic systems,
mucoadhesives, and erodible and nonerodible inserts. One of the
most important tools for developing and assessing these products
is an accurate pharmacokinetic model. The primary objective of a
pharmacokinetic model must be to enhance the accuracy of estimates of the dynamic state of drug behavior in an actual clinical
situation.11 Many pharmacokinetic models have been reported
in the literature and represent varied levels of sophistication.
Several excellent reviews on this subject are available.1217
The frequency with which drugs are administered is typically

governed by how rapidly the drugs are removed by metabolism
or clearance of the unmetabolized drug. The loss of a drug is
described by the half-life of the drug, i.e., the time for the tissue
concentration to fall to one-half of its value. Dosing frequency
is commonly every half-life so that a series of peaks and valleys
is established (Fig. 17.1). One factor that affects a drugs halflife, and thus the dosing frequency, is the afnity the drug has
to its receptor. Drug receptor afnity can be taken as the inverse
of the dissociation rate constant, and the half-life of a drug can
be easily calculated from the dissociation rate constant.
Timoptic (timolol maleate ophthalmic solution), used in the
treatment of ocular hypertension or open-angle glaucoma, is an
example of a topically administered drug in which dosing frequency can be decreased after the start of treatment from twice
a day (bid) to once a day (qd), if intraocular pressure (IOP) is shown
to be maintained at satisfactory levels. Similarly, azithromycin,
a topical antibiotic in development for the treatment of bacterial
conjunctivitis, is under evaluation for dosing regimen of bid on
days 12 and qd days 35.
TIME TO REACH STEADY STATE

Figure 17.1 shows that the tissue drug level rises because each
subsequent dose adds to the quantity of drug left from the earlier
dose. Steady state is achieved at a specific dosing rate when the
tissue concentration equals the rate of elimination. The half-life
CLINICAL UTILITY
Following is a brief explanation of the applications of pharmacokinetics to clinical practice.
FIGURE 17.1. Hypothetical tissue drug concentration versus time for

multiple doses of the same drug at set time intervals.
179
SECTION 4
FIGURE 17.2. Plot of hypothetical tissue drug concentration versus

time for multiple doses of the same drug given at time intervals either
before or after the half-life of the drug. - - - - before the half-life;
after the half-life.
of a drug determines the time to reach steady state. In most

cases, the time it takes to reach steady state is about four to ve
times the half-life of a drug.
STEADY-STATE MAXIMUMS AND STEADYSTATE MINIMUMS

Suppose the dosing interval is much earlier, or later, than the
half-life of the drug. Figure 17.2 gives a few hypothetical
examples of the impact of dosing interval. It is easy to see that
dosing too soon can push the drug into the toxic range, while
dosing too late can give periods of time when the levels are
subtherapeutic. The importance of the dosing interval cannot
be underestimated in achieving therapeutic effectiveness and
minimal toxicity.
FIGURE 17.3. Typical prole of drug concentration versus time in an

ocular tissue. Cmax, maximal level of drug in tissue; Tmax, time to reach
maximal level of drug in tissue.
Area under the curve (AUC) The AUC yields the total
amount of drug absorbed from an applied dose. The
bioavailability of a drug is computed from the AUC,
which is most important in determining therapy for
chronic medication.
PHARMACOKINETIC METHODS
Key Features
In the eye, it is difcult to clearly dene classic

pharmacokinetics parameters such as clearance and volume
of distribution, as well as rate and extent of absorption
Human pharmacokinetics studies are limited to noninvasive
means of data collection
Pharmacodynamic, instead of pharmacokinetic measurements,
can be taken with the caveat of patient variability with such
responses
The use of animal models, especially rabbits, is integral for
testing ophthalmic drugs
METHODS OF DRUG APPLICATION

The eye is an extraordinarily protected organ that excludes foreign
chemicals, such as drugs, through a variety of mechanisms.
Understanding the various loss pathways of a topically applied
drug can ensure that therapy is maximized and both local and
systemic toxicity minimized. These loss pathways and potential
remedies are discussed in detail later in this chapter.
PHARMACOKINETIC PARAMETERS
180
A prole of drug concentration in ocular tissue can be dissected

to provide important information. Figure 17.3 shows a typical
prole.
Cmax The maximal concentration of drug in the tissue is
Cmax. The level that is reached dictates therapeutic and
toxic responses and is directly related to the applied
drug concentration and the absorption and elimination
rate constants.
Tmax The time to reach a maximal level of drug in the
tissue is Tmax. This parameter is a function of only the
absorption and elimination rate constants and is
independent of the applied concentration.
HUMANS
Unlike systemic pharmacokinetic studies where the absorption,
distribution, metabolism, and elimination rate constants can
easily be calculated from measurements of drug concentration
in blood samples, human ocular pharmacokinetic studies are
limited to noninvasive observation of fluorescence or gammascintigraphic probes, as well as determination of drug concentrations from the aqueous humor during cataract surgery, or
from explanted tissues, conjunctival biopsies, or enucleation.
Attempts are being made to nd new methods for evaluating
ocular pharmacokinetics with the hopes of eventual application
to humans. For example, magnetic resonance imaging (MRI)
has been used to study the real-time release of a drug surrogate
(Gd-DTPA) from a polymer-based intravitreal implant in
rabbits.18 Also, cell culture models of ocular tissues are being
developed in order to test drug transport into the tissues, and
for potential toxicological screening of compounds.19
As an alternative to direct pharmacokinetic measurements,
ophthalmic pharmacodynamic responses such as miosis and
mydriasis,20,21 light reflex inhibition,2227 and IOP have been
ANIMAL MODELS
Rabbit Model
Because many anatomic and physiologic factors of the rabbit
and human eye are similar (Table 17.1) and because the animal
is relatively inexpensive, easy to handle, and has a larger eye
compared to other animals making it easier to perform clinical
studies, the rabbit is the animal model of choice in most ocular
experiments. In order to determine starting doses of ophthalmic
drugs for human preclinical studies, topical therapeutics should
be normalized to concentration (e.g., mg/area of application) or
amount of drug (mg) at the application site. Intraocular therapeutics should be normalized between rabbits and humans
according to the compartmental volumes and concentrations of
the drug, as opposed to normalization between species based on
body surface area (mg/m2) as typically done with systemically
administered drugs.27a
As useful as the rabbit model is, there are some differences

between the rabbit and human eye that can affect drug kinetics.
For example, the blink rate in humans (615 times/min) is higher
than in rabbits (45 times/h), which could allow the penetration
of drug through the cornea of the rabbits more than that of
humans because of a high drug concentration at the corneal
surface28,29 and low drug solution drainage (e.g., in the New
Zealand albino rabbit eye).30 Rabbits have a nictitating membrane that humans do not possess, which may absorb many
substances and act as a depot, affecting pharmacokinetic measurements. Although the albino has been used for most studies,
the absence of pigment will lead to differences in the
pharmacokinetics compared to the human eye. Consideration
should be given to using pigmented rabbits, especially for
drugs that work inside the eye. Moreover, rabbits appear to be
less sensitive than humans to moderate increases of vehicle
viscosity. For example, a suspension-type parafn ointment
gives better results in humans than rabbits, probably because
shear effects facilitate drug release.20 Therefore, clinical trials in
humans must always be used to conrm data from rabbits.
Other Animals
Other animals besides rabbits are also used in ocular pharmacokinetic studies, but to a much lesser degree due to various
reasons. The eyes of rats and mice are too small for testing of
different delivery systems. Dogs, cats, tree shrews, and monkeys
are also used, but for ethical reasons, should only be used for
invasive ocular pharmacokinetic studies when necessary, and are
thus in general practice limited to noninvasive kinetic measurements and pharmacodynamics.
CHAPTER 17
used as parameters for investigating the effectiveness of ocular

drug administration. One caveat on using pharmacodynamic
measurements in designing ophthalmic drugs is that the same
dose often produces a different magnitude of effect in various
individuals. Some of the factors that can contribute to this variation include eye pigmentation, whether or not the individual
wears contact lenses, patient compliance, the clinical state of
the eye (i.e., age of the individual and disease status), and such
physiological factors as the volume and turnover rate for both
tears and the aqueous humor.
The limitations of performing human pharmacokinetic
studies have led to the widespread use of animal models for
ocular studies.
TABLE 17.1. Comparison of Pharmacokinetic Factors between Rabbit and Human Eye
Pharmacokinetic Factors
Rabbit
Human
Tear volume (mL)
510
730*
Tear turnover rate (mL/min)
0.50.8
0.52.2
45 times/h
615 times/min
Spontaneous blinking rate
Lacrimal punctum or puncta
Nictitating membrane
Present
Absent
pH of lacrimal fluids
7.37.7
7.37.7
Turnover rate of lacrimal fluids (%/min)
16
Buffering capacity of lacrimal fluids
Poor
Poor
Milliosmolarity of tear (mOsm/L)
305
305
Initial drainage rate constant (/min)
0.55
1.6
Corneal thickness (mm)
0.350.45
0.520.54
15
1112
Corneal surface area (cm )
1.52.0
1.04
pH of aqueous humor
8.2
7.17.3
Aqueous humor volume (mL)
0.250.3
0.10.25
Aqueous humor turnover rate (mL/min)
34.7
23
Protein content of tears (%)
0.5
0.7
Protein content of aqueous humor (mg/mL)
0.55
30
Ratio of conjunctival surface to corneal surface
17
Corneal diameter (mm)

2
*Range depends on blinking rate and conjunctival sac volume.
Occurs during normal waking hours without apparent external stimuli.
Signicance of nictitating membrane from precorneal area is small relative to overall loss rate.
Data from references 4, 11, 1315, 17, 20, 28, 30, 54, 55, and 136141.
181
PHARMACOKINETICS MODELS
kabs
kelim
Key Features
SECTION 4
The pharmacokinetics of topically applied ophthalmic drugs is

usually analyzed using compartmental modeling
In multicompartmental modeling, the eye is divided into
kinetically homologous compartments divided by barriers that
do not necessarily correlate with anatomical compartments
The pharmacokinetics of a drug following intravitreal injection
has been modeled based on Ficks second law of diffusion and
assuming three major pathways for elimination
Various models have been made in order to describe the
kinetics of drugs delivered by different types of systems,
including controlled-release devices and nanoparticle
preparations
A typical example of an aqueous humor drug concentration prole for a topically applied drug such as pilocarpine is shown in
Figure 17.4. There are several important characteristics of this
gure. First, the drug disappears from the aqueous humor in
discrete steps and, in fact, the disappearance is triphasic. This
probably represents the distribution of the drug into various
anterior segment tissues that become reservoirs of the drug.
Over time, the loss of drug occurs with successively smaller
elimination rate constants. Second, the drug achieves Cmax in a
relatively short period of time 2040 min is typical giving
the impression that the drug is rapidly absorbed across the
cornea. In fact, the drug is typically not rapidly absorbed across
the cornea, and the early peak drug level is due to an unusual
constraint imposed by the kinetics of drug loss from the
precorneal pocket.
ONE- AND TWO-COMPARTMENT MODELS

The simplest pharmacokinetic model is to consider the eye as
one compartment (Fig. 17.5a).31,32 The equation describing drug
concentration in this model is dependent on absorption and
elimination rate constants. For systemically administered
drugs, absorption is generally the faster process, but for most
kabs
kelim
kloss
FIGURE 17.5. Schematic of two-compartment model without (a) and
with (b) the precorneal loss constant. kloss, precorneal loss constant;
kabs, absorption rate constant; kelim, elimination rate constant.
ophthalmic drugs, the true rate constant for absorption into the
eye is much smaller than the elimination rate constant, resulting in a scheme known as the flip-flop pharmacokinetic model.
However, this is an oversimplication, and a scheme known as
a parallel elimination pathway more accurately describes the
pharmacokinetics of ocular therapeutics. In actuality there are
many factors that contribute to precorneal kloss. Thus, all rate
constants describing loss of the instilled dose from the tear lm
are added together and the sum of these constants yields an
overall loss rate constant (kloss), and produces an apparent absorption rate constant (kabs) that is larger than the elimination rate
constant. In this model, the apparent absorption rate constant
kabs is described as:
Apparent kabs = kloss + true kabs
The magnitude of kloss is typically in the range of 0.5 min1,
whereas true kabs is two to three magnitudes smaller. Therefore,
most topically applied drugs show an early peak drug level, and
the time of this peak level is essentially independent of properties of drug. Figure 17.5b shows the nature of the model taking
into account precorneal loss, and represents a two-compartment
model. These data suggest that to signicantly improve ocular
drug bioavailability, it is necessary to make the kloss term smaller
by one to two orders of magnitude (for example, by using gels or
inserts to decrease loss through drainage) or to increase the true
kabs by one to two orders of magnitude (for example, by the
addition of penetration enhancers).
MULTICOMPARTMENT MODELS
182
FIGURE 17.4. Aqueous humor concentration of pilocarpine versus

time prole after institution of 25 L of 1 102 M solution.
A much more complicated model is needed to adequately describe

the pathway from precorneal application through the cornea and
into the aqueous humor followed by distribution into the surrounding tissues. The four-compartment model shown in Figure 17.6
was used to t the drug concentration data for both cornea and
aqueous humor obtained after topical administration of pilocarpine to the albino rabbit eye. A mathematical derivation of this
pharmacokinetic model was also reported.32 However, the model
treated the cornea as a simple semipermeable membrane. In fact,
the cornea consists of an epithelium, stroma, and endothelium.
The results from pharmacokinetic studies demonstrated that
the lipophilic epithelium acts as a barrier to drug penetration by
hydrophilic drugs such as pilocarpine. Movement of watersoluble drugs through the hydrophilic stroma is usually rapid.
Therefore, the corneal stroma and endothelium are kinetically
homogeneous with the aqueous humor. A four-compartment
model that treats the cornea as three separate tissues corrects
this deciency (Fig. 17.6b and Table 17.2).31,32 In order to more
kloss
kCabs
kahabs
kcdist
kRdist
kahdist
ktarget
TABLE 17.2. Parameters of Models Described in Figure 17.6

Parameter
Coefcient Associated With
kloss
Elimination rate constant from the precorneal area
abs,
dist,
ah
Apparent absorption rate constants into the cornea and aqueous humor, respectively
abs
r
k dist, k
ah
dist
Distribution rate constants into the cornea, reservoir, and aqueous humor, respectively
ktarget
Absorption rate constant into the target area
Pp
Transfer of drug between precorneal area and corneal epithelium
Pn
Nonproductive loss
kd
Drainage
QT(t)
Tear flow
Pa
1. Transfer of drug between corneal epithelium and corneal stromaendotheliumaqueous humor

2. Transfer of drug between corneal epithelium and corneal stromaendothelium
Pm
Drug loss via metabolism in or lateral diffusion from corneal epithelium
Pao
Drug elimination from aqueous humor
Pr
1. Transfer of drug between corneal stromaendotheliumaqueous humor and reservoir

2. Transfer of drug between aqueous humor and reservoir
Pro
Drug elimination from reservoir
Ps
Transfer of drug between corneal stromaendothelium and aqueous humor
Pso
Drug elimination from corneal stromaendothelium
accurately model the pharmacokinetics of lipophilic drugs, such

as fluorometholone, the corneal stromaendothelium and aqueous
humor are logically separated33 in a ve-compartment model
(Fig. 17.6c and Table 17.2).
INTRAVITREAL INJECTION
Drugs that are introduced into the vitreous humor by intravitreal
injection spread through the vitreous humor and into the anterior chamber at the same rate that they diffuse in free solution.13
Two pathways of exit from the vitreous chamber have been
predicted: (1) through the anterior hyaloid membrane into the

posterior chamber and out of the eye with aqueous drainage and
(2) directly across the retinal surface.
In one study, computer simulation was used to evaluate the
in vivo and in vitro pharmacokinetic correlation of dexamethasone sodium after intravitreal injection of m-sulfobenzoate in
rabbits.34 The mathematical model was developed based on
Ficks second law of diffusion by assuming that the vitreous
body is a cylinder with three major pathways for elimination:
the posterior aqueous chamber, the retinalchoroidscleral
(RCS) membrane, and the lens (Fig. 17.7). Results showed that
CHAPTER 17
FIGURE 17.6. (a) Schematic of simplied four-compartment model. (b) Schematic of four-compartment model that considers corneal
stromaendothelium and aqueous humor as one compartment. (c) Schematic of ve-compartment model that considers the corneal
stromaendothelium and aqueous humor as separate compartments. For description of parameters see Table 17.2.
183
SECTION 4
FIGURE 17.7. Cylindrical model of the vitreous body of rabbits for

analyzing the pharmacokinetics of intravitreal drug delivery; the
surface of the vitreous body is divided into three areas of elimination
pathways: the posterior chamber, the RCS membrane, and the lens.
RCS, retina/choroid/sclera; H, effective height of vitreous body; R, b0,
effective radius of vitreous body and lens, respectively; x, y, horizontal
and vertical axes, respectively.
the major route of elimination of the drug was through the posterior aqueous humor because of an absence of barrier membrane between the boundaries. By using the ratio of the product
of the diffusion coefcient and the effective area of the posterior
chamber, the RCS membrane, and the lens (50:4:0.1), the
authors concluded that after intravitreal injection, most hydrophilic drugs are eliminated by the annular gap between the lens
and the ciliary body (i.e., from the posterior chamber and flow
into the anterior chamber), and the RCS membrane may act as
a major route of elimination of lipophilic drugs.
A recent ocular model, also based upon Ficks second law of
diffusion, assumes a spherical, modied cylindrical eye, and can
predict the time course of the local tissue concentration in the
eye following a variety of ocular drug delivery systems including
topical, systemic, and transdermal administration as well as
vitreous injection and implantable delivery.35
by plane sheet barriers of known physical thickness with constant surface area. In this model, four routes of drug loss
lacrimal drainage, conjunctival absorption, aqueous drainage,
and irisciliary body absorption were included. The model was
validated by using the experimental in vivo data compared with
predicted aqueous humor drug concentrations from the model.
The results showed an excellent correlation, and it was also
possible to predict the amount of drug lost through each of the
four elimination pathways. This model was modied by adding
the compartments for the conjunctiva and the irisciliary body
in order to compare pharmacokinetic differences between ocular
inserts placed under the eyelid in the conjunctival fornix and
eye drops of timolol.37 The investigators observed increased
absorption into the irisciliary body and aqueous humor for
ocular inserts, and this is thought to partially be the result of
increased drug penetration across the conjunctiva and sclera.
Grass and Lee38 described and developed methods for constructing a pharmacokinetic model that can be used to predict
the effect of increasing drug retention in the conjunctival sac,
and varying the rate of release of the drug from a controlled drug
delivery device, on the ratio of drug concentration in aqueous
humor and plasma after topical dosing in rabbits. In addition, a
computer model was recently developed to describe the threedimensional convectivediffusive transport of drug released
from an intravitreal controlled release source.39
A multicompartment model was constructed to describe ophthalmic drug delivery with nanoparticle preparations.40 This
model was constructed from data that suggested that nanoparticle preparations might be able to create a precorneal depot,41
thus enhancing drug penetration directly to its site of action,
the trabecular meshwork,40 through the scleral or noncorneal
pathway.42,43
FACTORS INFLUENCING BIOAVAILABILITY

Key Features
MODELS DERIVED FROM DRUG DELIVERY

The ve-compartment model (Fig. 17.8) was developed to study
the mechanism involved in transcorneal permeation of drugs
from delivery devices.36 The model consists of the tear lm,
epithelium, stroma, endothelium, and aqueous humor, which
were assumed to be perfectly mixed and adequately represented
Due to a number of anatomical and biological factors that exist

to protect the eye, the intraocular bioavailability of topically
administered medications is typically only 110%
Smaller may be better: A smaller instilled eye drop may result
in decreased blinking, increased retention time, and greater
absorption
A large portion of a topically instilled drop results in
nasolacrimal drainage and systemic absorption, which may
lead to adverse side effects
The cornea is a potent barrier to drug absorption due to its
small surface area and its low permeability to both lipophilic
and hydrophilic drugs
Some aspects of drug formulation that affect bioavailability
include hydrophilicity/lipophilicity, concentration, osmoticity,
pH, and viscosity
TOPICAL DELIVERY
184
FIGURE 17.8. Schematic of ve-compartment model that was

developed for drug delivery devices.
There are several possible absorption pathways of a topically

delivered ophthalmic drug (Fig. 17.9). The primary ocular
absorption pathway for small lipophilic drugs is from the tear
lm to ocular tissues, via the cornea and the aqueous humor.
After absorption into ocular tissues and systemic circulation,
the drug is eventually eliminated from the body. A substantial
portion of topically applied drug is lost due to drainage, and the
reduced amount of drug reaching systemic circulation because
of drainage is an important consideration in the dosage and
delivery of ocular drugs, as it is a major contributor of adverse
effects.
bioavailability. The two main approaches have been to alter or
supplement ophthalmic drug formulations in order to increase
absorption, or to improve upon or design new delivery systems.
The following is by no means a complete list of the recent
research in this eld, but it will serve to highlight some of the
major areas of focus.
FIGURE 17.9. Typical prole of the fate of a topically applied drug.
Another important absorption pathway is through the conjunctiva, the vascularized thin mucous membrane lining the
inside of the eyelids and anterior sclera. The conjunctiva has a
much larger surface area and greater permeability to watersoluble compounds than the cornea. In fact, the conjunctiva
competes so effectively with the cornea for drug absorption that
it has been calculated that the conjunctiva is as important as
solution drainage loss in reducing the fraction of pilocarpine
available for corneal absorption.44
There is evidence that the absorption of ophthalmic drugs
into the sclera represents a signicant pathway for large, hydrophilic drugs. Ahmed and associates45 tested the scleral absorption of the lipophilic drugs propranolol, timolol, nadolol, and
penbutolol, and the hydrophilic compounds sucrose and inulin.
The results showed that resistance to penetration for all compounds tested in the outer layer of the sclera is much less than
that of the corneal epithelium. The cornea offered substantially
more resistance to inulin (a hydrophilic drug) than did the sclera.46
However, the cornea and conjunctiva offered comparable resistance against timolol (a lipophilic drug).45 In addition, Schoenwald
and co-workers47 have shown that the conjunctivalscleral
route of entry produced higher irisciliary body concentrations
of methazolamide analogs and 6-carboxyfluorescein, but not of
rhodamine B (a lipophilic dye). The explanation of this phenomenon is that a hydrophilic drug is absorbed into the ciliary
body through vessel uptake into the sclera and deposits within
the ciliary body, whereas a lipophilic drug penetrates across the
cornea and diffuses through the pupil against aqueous flow to
enter the posterior chamber.48
Most ophthalmic medications are administered topically, a
route of delivery that has major advantages including localized
drug effects, avoidance of hepatic rst pass metabolism, and convenience. In fact, it has been shown that topically administered
allergy drugs have greater efcacy in relieving symptoms as
compared to systemically or nasally administered drugs.49,50 On
the other hand, topically administered ocular drugs have the
disadvantage of low bioavailability to intraocular tissues due to
a number of anatomical and biological factors that exist to protect
the eye, and by consequence, the entry of ocular therapeutics. In
fact, it is estimated that the intraocular bioavailability of topically administered medications is typically only 110%.6,51 A
great deal of research has been dedicated to improving ocular
One factor that influences ocular bioavailability after topical

delivery is the retention of the therapeutic in the preocular area.
The volume of liquid that the conjunctival sac can contain is
~2030 mL52 and the volume of the tear lm is 7 2 mL.53 Due
to physical limitations of eye drop size when delivered from a
standard dropper, however, most bottles deliver 3050 mL
instead of the ideal drop size of 1020 mL.52 The delivery of this
larger volume causes reflex blinking, which increases the
drainage rate to the nasolacrimal canal, spilling on the cheeks
and splashing the excess of the solution to the eyelashes.54,55
This results in both wasted amounts of medication and possible
negative side effects due to high systemic absorption. It has been
found that a 50 mL drop has the same pharmacological activity
as a 20 mL drop of pilocarpine,56 and in fact, it has been proposed that reducing the volume of the instilled drop of a drug
with low corneal permeability increases its bioavailability by
four times.57 Thus it appears as though a smaller instilled drop
may result in decreased blinking, increased retention time, and
thus greater absorption. The physical blockage of the lacrimal
drainage system by punctal occlusion has also been studied as a
means for increasing ocular drug retention, but currently it is
unclear as to whether silicone punctal plugs provide any additional therapeutic benet for topical antiglaucoma medications.58,59 A less invasive method of punctal occlusion is to press
down with a nger over the tear duct after administering eye
drops. This technique has been shown to improve efcacy and
results in safer usage of several antiglaucoma medications.60 A
pathological obstruction of the nasolacrimal duct may also
similarly alter bioavailability.
Tear lm breakup also improves the absorption of topically
administered ophthalmic drugs. The tear lm is a complex fluid
that covers the ocular surface, and functions to protect and
maintain the surface of the eye. The tear lm consists of three
layers, a lipid, aqueous, and mucous layer. The tear lm structure remains intact for a certain period of time before it begins
to break apart or rupture, exposing the ocular surface, at which
point, blinking is necessary to replenish this complex fluid. The
barrier function of the tear lm makes it difcult for an ophthalmic agent to be effective by restricting the products interaction
with target receptors of the ocular surface. Additionally, the tear
lm composition is responsible for diluting ophthalmic agents,
resulting in further reduced efcacy. Abelson and associates
have found that by having patients refrain from blinking, and
thus allowing tear lm breakup to occur, for 6 s before drop
instillation, the efcacy of pilocarpine (1%) and tropicamide
(1%) in constricting and dilating the pupil, respectively, is
signicantly improved.61
CHAPTER 17
PREOCULAR RETENTION
SYSTEMIC ABSORPTION
It has been calculated that, of an eye drop ~50 mL in volume,
~20 mL or 40% does not touch the cornea but goes directly to
the highly vascular drainage apparatus.16 The excess volume
from the standard eye drop is rapidly pumped by the lacrimal
puncta, passes into the lacrimal canaliculi, then successively
travels into the lacrimal sac, the nasolacrimal duct and nally
the nasal cavity. In the nasal cavity, the active ingredients are
absorbed by mucosal vessels and distributed into the general
185

circulation. There are several examples of severe adverse systemic effects correlated with topically administered ophthalmic
drugs. Used in the treatment of glaucoma, b-adrenoceptor antagonists, or b-blockers, are the most frequent cause of systemic
adverse effects due to ophthalmic treatments. These topical
antiglaucoma treatments are known to be associated with pulmonary, cardiac, and central nervous system effects. A body of
research is devoted to determining if cardioselective b-blockers
(such as betaxolol) have fewer or less severe systemic effects
than nonselective b-blockers.62 There have also been reports of
cardiac irregularities such as palpitations and tachycardia occurring after topical ocular epinephrine administration, which is
used in the treatment of glaucoma and ocular hypertension.
Mydriatic drops, such as phenylephrine, applied either as a treatment or as a diagnostic tool can also have potential adverse effects,
including severe cardiovascular and neurological disorders.63,64
SECTION 4
CORNEAL ABSORPTION
186
The cornea is a potent barrier to drug absorption due to its small

surface area and low permeability, attributable to its anatomical
structure. The cornea is comprised of three layers: the lipophilic
outer epithelium; the hydrophilic stroma, which constitutes 90%
of the thickness of the cornea; and the inner endothelium, consisting of a single layer of flattened epithelial-like cells. Because
the cornea has both hydrophilic and lipophilic structures, it acts
as an effective barrier against both lipophilic and hydrophilic
drugs. Thus, unique approaches have been taken in designing
ocular drugs with increased corneal absorption, such as
prodrugs.
ADDITIVES
Various compounds can be added to topically administered ophthalmic drugs in order to increase corneal absorption, and most
fall into one of two categories: compounds that increase corneal
residence time, and compounds that increase corneal penetration.
Included in the category of increased corneal residence time are
those compounds that increase the viscosity of the therapeutic
or have mucoadhesive properties. These compounds will be
discussed later.
Preservatives such as benzalkonium chloride (BAK) and cetylpyridinium (CPC), which act as surfactants, have been hypothesized to increase absorption of ocular drugs across the cornea.
Other preservatives, such as thimerosal, chlorobutanol, chlorohexinide digluconate (CHD or CDG), hydrogen peroxide, sorbic
acid, sodium bisulte, and EDTA, have also been shown to
increase corneal absorption. BAK has been accused of increasing
drug penetration as a result of its toxicity to the ocular surface.
However, these studies utilized exaggerated dosing regimens or
drug concentrations that contain unrealistically high levels of
BAK and were conducted in animal or in vitro cell models that
do not translate into relevant clinical information.6569 Studies
using clinically applicable concentrations and dosing of BAK have
not shown toxicity to corneal epithelial cells and no signicant
effects were observed with regard to corneal healing and epithelial migration rates when BAK 0.01% was instilled qid.7072
Recently, research has focused on the use of polymers as
penetration enhancers. The cationic polymer compound chitosan
hydrochloride (Ch-HCl) was shown to signicantly enhance intraocular drug penetration, thought to be due to increased corneal
permeability.73 Also, basic amino acid polymers such as poly-Larginine (PLA) appear to enhance the permeation of hydrophilic
compounds through the cornea, conjunctiva, and conjunctiva/
sclera composite, and thus may be used as permeation enhancers
for ocular drug delivery via both the corneal and noncorneal
pathways.74
SUSPENSIONS
Because the outer epithelium is such an effective barrier against
hydrophilic compounds, absorption of moderately lipophilic
compounds is favored. There are solubility issues associated
with formulating lipophilic drugs as ophthalmic eye drops,
therefore many compounds must be formulated as suspensions.
Topical ophthalmic suspensions have a few limitations
including particle distribution (they need to be shaken before
use) and sterility.6
As an alternative to conventional suspensions, water-soluble
derivatives have been used to enable drug formulation as aqueous
solutions.75,76 Though these formulations have decreased intraocular absorption due to the lipophilic corneal epithelium, this
is offset by increased driving concentration resulting from the
increased aqueous solubility. Cyclodextrins and b-cyclodextrins
are being explored as a different formulation strategy with the
goal of increasing the aqueous solubility of compounds with low
water solubility. Cyclodextrins are cyclic sugars that have a hydrophilic outer surface and a central lipophilic cavity. The lipophilic
cavity enables cyclodextrins to hide a variety of compounds
with low water solubility in aqueous solution, while the hydrophilic outer surface allows these complexes to remain water
soluble.77,78 A recent example of the use of cyclodextrins in
topical ophthalmic formulations is of methylated b-cyclodextrin
added to dorzolamide, a carbonic anhydrase inhibitor, used in
the treatment of glaucoma.79
It is important to keep in mind a caveat about suspensions:
the driving force for drug absorption is drug concentration, so a
10% solution is absorbed at a rate that is 10 times that of a 1%
solution. This is not so with a suspension. A 10% and a 1%
suspension have exactly the same amount of drug in solution,
and all additional drug is insoluble. Although suspension drug
particles that remain in the cul de sac can act as depot, the residence time of the solid in the precorneal pocket is only 2 min.
Undissolved drug is lost from the front of the eye and does not
contribute to ocular tissue drug levels. Thus, a 10% suspension
is rarely 10 times more bioavailable than a 1% suspension.
METABOLISM
Drug metabolism can affect bioavailability in a positive way by
utilizing endogenous enzymes in the corneal area. Some ocular
therapeutics are prodrugs, which can be chemically or enzymatically converted to the active parent drug, either within the cornea
or after the corneal penetration. A classic example is dipivefrin,
an epinephrine-derived ester that is 600 times more lipophilic
than the native form of epinephrine. After passing through the
corneal epithelium, dipivefrin is hydrolyzed by esterases to yield
active epinephrine. The nal bioavailability of dipivefrin is
17 times greater than that of an eye drop that contains native
epinephrine, and thus produces similar therapeutic properties in
the eye with fewer adverse effects.80,81 This is due to the addition
of pivaloyl groups to epinephrine to make dipivefrin, enhancing
its lipophilicity and thus its penetration to the anterior chamber.
The role of metabolism in ocular tissues continues to be an
important area of research, with numerous advances in ocular
prodrug therapeutics.82,83
DRUG FORMULATION
HYDROPHILICITY VERSUS LIPOPHILICITY
One inherent characteristic of an ocular therapeutic that can be
an important factor in determining the extent of its absorption
into ocular tissue is the drug molecules hydrophilicity or lipophilicity. For example, it has been shown that moxifloxacin, a
SOLUTION OSMOTICITY
form a cross-linked lattice that helps the mucin layer to adhere

to the eye in desiccated areas. Systane was shown to protect the
cornea from desiccation in an in vivo rabbit model, and signicantly relieve dry eye symptoms in patients in comparison with
control solutions.8890
NOVEL DRUG DELIVERY DEVICES

Key Features
Tears are slightly hypertonic (~330 mOsm). Hypertonic solutions

above 400 mOsm may be unpleasant to the eye and may induce
lacrimation, which in turn causes greater precorneal drainage loss.
In contrast, hypotonic solutions as low as 100 mOsm are still
comfortable in the eye and may actually lead to an increase in
the bioavailability of water-soluble drugs, presumably through a
solvent drag effect. For comparison, the osmolarity of the ocean
is ~1000 mOsm.
SOLUTION pH
For stability reasons, many eye drops are formulated at pH values

other than pH 7.4. The comfort zone of an ocular solution is
rather narrow and typically in the pH 68 range. Outside this
range, the solution can be uncomfortable and induce lacrimation, resulting in drug loss. The pH boundary outside of which
actual tissue damage may occur is below pH 3 and above pH 10.
The ability of the eye to restore physiologic pH is very good, and
this occurs within a short time because of lacrimation, the high
turnover rate of tears (which is ~16%/min) and the tears buffer
system.53,54 With an ionizable drug, it is sometimes tempting to
adjust the pH either above or below the comfort range pH of 68
to convert the drug to a more favorable form for absorption (i.e.,
the undissociated form of the drug). When the pH is adjusted in
this way, any gain in drug bioavailability is generally negated by
precorneal loss owing to discomfort and lacrimation.
VISCOSITY
It is clear that the factor contributing the most to precorneal
drug loss is drainage. An increase in the viscosity of the solution
might appear to remedy this problem. It has been shown that
increased viscosity increases dwell time on the ocular surface,
but for formulations above 70 cp, there is an increased likelihood
of unwanted effects such as lid caking and blurring, and general
discomfort in the eye. Solutions of ~70 cp are able to maximize
residence time without these side effects. Of viscosity-enhancing
polymers, poly(vinyl alcohol) and poly(vinylpyrrolidone) are considered ideal because of their spreading characteristics the thickness of the applied medication layer over the precornea area.86
Another approach to the precorneal residence time problem
is to employ a phase-change polymer. These systems are liquid
in the bottle, but when placed in the eye the polymers solidify
because of differences in temperature, pH, or mono/divalent ion
concentrations. Such phase-change solutions are typically better
accepted by the patient. A couple of examples of these gel-forming
systems are Timoptic-XE, used in the treatment of glaucoma and
ocular hypertension and Systane, which is used to treat dry eye.
Timoptic-XE contains timolol maleate, a nonselective b-blocker,
and Gelrite a puried gellan gum that forms a gel upon contact
with the precorneal tear lm. Timoptic-XE was shown to reduce
systemic absorption as compared to a timolol maleate ophthalmic solution.87 Systane contains two demulcents (propylene
glycol, PEG-400) and a gelling agent, Hydroxypropyl (HP)-Guar.
Guar has a neutral pH of 7.0, and when it interacts with slightly
alkaline (pH ~7.5) human tears, it bonds with borate ions to
An ideal delivery system should be effective and consistent,

inexpensive, and comfortable for the patient
Mucoadhesive polymers increase ocular drug bioavailability by
prolonging contact with the corneal layer
Sustained release devices are able to reach near zero-order
kinetics, in which the level of administered drug remains
constant throughout the delivery period
Liposomes are vesicles composed of phospholipid bilayers
and are able to accommodate both lipophilic and hydrophilic
drug molecules
Particulate polymeric drug delivery systems such as microand nanoparticles are stable, relatively comfortable in the eye,
and prolong ophthalmic absorption times
There are advantages and disadvantages to each of the topical

ocular delivery systems available. Improvements continue to be
made on these systems, and novel drug delivery devices are being
researched. An ideal delivery system should be effective and
consistent, inexpensive, and comfortable for the patient.
METERED DELIVERY SYSTEMS

There are several new delivery systems being developed, including UniDoser, Eye Instill, and Visine Pure Tears, that improve
upon the standard eyedropper by addressing drop size and
allowing for the delivery of a multidose medication without the
need for preservatives.
A new type of ocular drug delivery utilizes a system that
delivers a mist containing medication. For example, Kahn et al
are developing a small volume nebulizer system91 and Optimyst
Systems are working on a small, handheld device that uses
ultrasonic vibrations to create a ne mist.
CHAPTER 17
novel fourth-generation fluoroquinolone, has higher maximum

concentrations in ocular tissues in comparison to other fluoroquinolones. This is thought to be due to the unique structure of
moxifloxacin that combines high lipophilicity for enhanced corneal penetration with high aqueous solubility at physiological
pH. The latter property creates a high concentration gradient at
the tear lm/corneal epithelial interface, providing a driving
force for better ocular penetration for moxifloxacin.84,85
MUCOADHESIVES
Poly(acrylic acid) and hyaluronic acid are two examples of mucoadhesive polymers that can interact with the mucosal layer on
the cornea and sclera, increasing retention of the drug.92,93
Mucoadhesives can aid in the localized delivery of topical
ophthalmic drugs and increase bioavailability due to prolonged
contact with the corneal layer. After the mucoadhesive polymer
is administered, and after contact with water and subsequent
swelling, the polymer and mucin become physically entangled.
Un-ionized carboxylic acid residues on the polymer then form
hydrogen bonds with the mucin molecule.94
SUSTAINED RELEASE DEVICES

Zero-Order Kinetics
Ideal delivery of drugs would follow zero-order kinetics, in
which the level of administered drug would remain constant
throughout the delivery period. Zero-order release kinetics can
be calculated from data obtained from in vitro drug release studies:
the slope(s) of the log(percent drug release) versus log(time)
plots can be calculated from tted linear regression lines. A
slope of 1.0 represents zero-order kinetics.
187

There is interest in the development of devices with a controlled and sustained release of ophthalmic drugs in order to
circumvent the inconvenience of frequent dosing. If dosages
could be sustained for extended periods of time, therapeutic
levels could be maintained for weeks or longer. These devices
represent the best hope of the drug delivery systems available
for reaching zero-order kinetics.
Delivery of most drugs follows what is known as rst-order
kinetics, in which initially high levels of the drugs are attained
followed by an exponential decrease in concentration. There is
some difculty in designing an optimal dosing regimen for drugs
that follow rst-order kinetics. Drug concentrations in target
tissues need to be maintained above the minimum concentrations for therapeutic effectiveness, but below toxic levels, and
staying in this range is tricky due to the rapid rise and fall in
drug concentrations.
SECTION 4
INSERTS
Sustained release devices or inserts fall into two different categories: those that are insoluble or nonerodible, and those that are
soluble or erodible. An obvious advantage of erodible systems is
the fact that the delivery system does not have to be removed from
ocular tissues after the drug has been released. A disadvantage of
the erodible systems is that they are more likely to show patientto-patient variability in release kinetics due to different rates of
tear production/turnover and concentration of metabolic enzymes
in the tear lm.
Unfortunately, all of the insoluble ocular inserts on the
market have yet to gain wide acceptance. One example is
Ocusert Pilo, a pilocarpine-loaded, insoluble device placed in
either the upper or lower cul-de-sac for the treatment of glaucoma. This device failed to achieve widespread use because of
its costliness and poor patient compliance due to ejection from
patients eyes. Other examples of insoluble sustained-release
devices in development include: presoaked hydrophilic contact
lenses;95 OphthaCoil, which consists of a drug-loaded adherent
hydrogel coating on a thin coiled metallic wire inserted into the
conjunctival sac;96 and a one-side-coated insert that releases
drug from only the uncoated side.97
A few examples of soluble or erodible ocular inserts include:
corneal collagen shields;98,99 and gel-forming erodible inserts for
the delivery of ofloxacin.100
IONTOPHORESIS
Iontophoresis, the process in which an electrical current drives
ions into cells or tissues, is not a new mode of drug delivery.
However, the method has been the focus of innovation, including the use of hydrogel-containing probes.101,102 Recently, an
iontophoresis device has been shown to be safe and well tolerated
in a clinical setting for the management of active corneal graft
rejection,103 shows enhanced ocular absorption of small cationic
compounds such as carboplatin,104 and is being tested for its efciency in anterior delivery of Clonidine, an ocular hypotensive.
LIPOSOMES
188
Liposomes are microscopic vesicles composed of one or more

phospholipid bilayers. Due to the biphasic nature of liposomes,
both lipophilic and hydrophilic drug molecules are accommodated, thus almost any type of drug can be encapsulated. Acting
as a drug carrier, liposomes bind to the cellular membrane and
facilitate the transport of drug across the membrane. The rst
use of liposomes in ocular therapy was reported by Smolin et al,
who tested the effectiveness of liposome-associated idoxuridine
as compared to solution for the treatment of herpes simplex
keratitis in the rabbit eye.105 Recent studies on liposome ocular

therapeutics include: the encapsulation of acyclovir;106,107 disulram, a potential anticataract agent;108 the mydriatic tropicamide;109
cyclosporine A;110 and antisense oligonucleotides, which could
potentially be used in the treatment of ocular viral infections.111
The efcacy of liposome-encapsulated O-palmitoyl prodrug
of tilisolol, a nonselective b-blocker, after both topical and intravitreal injection has been studied,112 as well as liposomes carrying
plasmid DNA. Masuda et al showed that the delivery of these
liposomes resulted in efcient and stable transfer of the functional gene to the cornea, iris, ciliary body, and retina of rats.113
MICROSPHERES AND NANOPARTICLES

While liposomes represent a promising avenue of ocular drug
delivery, they are less stable than particulate polymeric drug
delivery systems such as micro- and nanoparticles. The drugs
can be incorporated into or absorbed by the particles. This mode
of ocular drug delivery has been shown to increase drug absorption in the eye as compared to ophthalmic solutions, due to a
much slower elimination rate of the particles.114 Particles in the
micrometer size range (>1 mm) are called microparticles or microspheres, whereas those in the nanometer size range (<1 mm) are
called nanoparticles. The upper size threshold for microparticles
for ophthalmic administration is ~510 mm. Patients experience
discomfort after application of particles above this size, and
generally, the smaller the particles, the better the patient toleration of the drug. Some recent examples of research into microand nanoparticles in topical ocular delivery systems include:
5-fluorouracil microspheres;115 sodium ibuprofen-loaded polymeric
nanoparticle suspensions (Eudragit RS100);116 biodegradable
calcium phosphate nanoparticles (CAP) containing 7-hydroxy2-dipropyl-aminotetralin (7-OH-DPAT), an IOP-lowering agent;117
solid lipid nanoparticles (SLN) used in the delivery of tobramycin;118 and chitosan (CS) nanoparticles for the delivery of
cyclosporine A.119 These particulates drain through the lacrimal
duct, but a certain percentage remains in the eyes for several
hours or even longer, thus an important characteristic of these
delivery systems is that the particulates are made of biodegradable polymers.
VITREORETINAL DRUG DELIVERY

Key Features
There is poor penetration of ocular therapeutics to the

posterior tissues of the eye due to the bloodocular barrier and
tight junctional complexes in the retinal epithelium
Microspheres, iontophoresis, and sustained release implants
are some of the delivery systems being developed to treat
vitreoretinal diseases
Some treatments for AMD are being tested that target proteins
involved in angiogenesis, such as VEGF or PEDF
Lately there has been increased attention on the development of

ocular therapeutics for vitreoretinal diseases. A great deal of
research is focused on treatments for age-related macular degeneration (AMD), macular edema, diabetic retinopathy, and diabetic
macular edema. There is poor penetration of systemically administered ocular therapeutics to the posterior tissues of the eye due
to the bloodretinal barrier. The limited permeability of the
bloodretinal barrier results from the network of tight junctional complexes (zonulae occludens) present in the retinal pigment epithelium and the endothelial membrane of the retinal
vessels. Intravitreal injection is a relatively safe and easy method
of drug delivery to the posterior tissues but due to its invasive
nature, there is the possibility of complications such as vitreous
hemorrhage, retinal detachment, and infection such as endophthalmitis. In order to improve upon the delivery of drugs to the
posterior segments of the eye, alternative delivery routes, such as
subconjunctival injection, are being investigated, and less invasive delivery systems, as well as advancements on injectable therapeutics, are in development for the treatment of vitreoretinal
diseases.
Subconjunctival injection refers to the injection of up to 0.5 mL

of a drug solution underneath the thin membrane lining the
eye, known as the conjunctiva. There are currently only a few
studies that have examined the pharmacokinetics of ocular
drugs delivered via this route, but it appears as though there is
greater absorption of drugs delivered by subconjunctival absorption than by systemic or topical administration. For example,
Weijtens et al120 measured the concentration of dexamethasone
in aqueous, vitreous, and serum in phakic patients following
subconjunctival injection of dexamethasone disodium phosphate
and compared the results to those following peribulbar and oral
administrations. It was found that the subconjunctival injection
was more effective than peribulbar and oral administration,120
as well as topical instillation.121 Some effects of subconjunctival
injection include backward drainage of solution along the
needle track, or diffusion across the conjunctiva,122,123 as well as
considerable systemic absorption.120,124
NOVEL VITREORETINAL DELIVERY SYSTEMS

The use of drugs encapsulated in microspheres in the treatment
of vitreoretinal diseases has been studied by several groups. For
example, Moritera et al investigated the intravitreal injection of
poly(lactic acid) microspheres containing 5-fluorouracil,125 and
Saishin et al tested the periocular injection of microspheres
containing PKC412, a kinase inhibitor that has been shown to
inhibit ocular neovascularization in mice, as a potential treatment
for AMD.126
Other novel vitreoretinal delivery systems are being explored.
For example, it has been shown that transscleral Coulombcontrolled iontophoresis (CCI) following intravitreal injection
enhances penetration of corticosteroids.127
An interesting potential delivery system is the Encapsulated Cell
Technology (NT-501) which uses encapsulated retinal pigment
cells, genetically modied to secrete ciliary neurotrophic factor,
for the treatment of glaucoma, retinitis pigmentosa, and AMD.128
INJECTABLE THERAPEUTICS
There are a variety of implants that are being tested for their use
in treating vitreoretinal diseases. Vitrasert is one of the initial
drug delivery devices for vitreoretinal disease. It is used in the
treatment of AIDS-related cytomegalovirus retinitis and is a device
that is surgically implanted into the vitreous, where it releases
the antiviral drug ganciclovir remaining active for approximately
seven and a half months.129 Retisert, which uses a similar technology as Vitrasert, is an intravitreal device approved to release a
constant amount of the steroid, fluocinolone acetonide, over a
treatment period of 30 months, with the potential to treat posterior uveitis, diabetic macular edema, and AMD.130 Another
ocular insert with the potential to treat AMD is the I-vation
implant, developed for site specic delivery of the steroid triamcinolone acetonide (TA) into the posterior chamber of the eye
over a time period of 6 months to 2 years.131 Anecortave acetate,
marketed as Retaane 15 mg for the treatment of AMD, is a
synthetic analog of cortisol with angiostatic but not glucocor-
FIGURE 17.10. Anecortave acetate is administered outside the globe

with a curved, blunt-tipped cannula. The cannula is inserted between
Tenons capsule and the sclera and the drug forms a depot directly
behind the macula where it is slowly released over 6 months. A
counter-pressure device (CPD; shown with a notch covering the
cannula) is used to prevent reflux of the suspension.
CHAPTER 17
SUBCONJUNCTIVAL INJECTION
ticoid receptor mediated activity. Anecortave acetate is delivered

as a posterior juxtoscleral depot (PJD) onto bare sclera near the
macula (see Fig. 17.10).
Evidence is growing that targeting proteins involved in the
process of angiogenesis, such as VEGF or PEDF, might provide
specic and effective treatment of AMD. Abnormal regulation
of angiogenesis, or formation of new blood vessels from preexisting ones, has been implicated in the pathogenesis of several
disorders, including AMD. Vascular endothelial growth factor
type A (VEGF-A) is a stimulator of angiogenesis; its binding to
VEGF receptors has been shown to promote endothelial cell
migration and proliferation, two key features required for the
development of new blood vessels. Pegaptanib sodium injection
(Macugen) is a selective VEGF antagonist that requires repeated
injections into the vitreous cavity. Ranibizumab (rhuFab V2;
Lucentis), a humanized monoclonal antibody fragment against
VEGF, is also delivered by intravitreal injection.132 Budesonide
is capable of inhibiting VEGF expression through glucocorticoid
receptor activity. Kompella and colleagues showed that subconjunctivally administered budesonide-PLA nano- and microparticles sustain retinal drug delivery as compared with the
budesonide solution-treated group at the end of day seven of
their study.133 It is now believed that angiogenesis is regulated
by a balance between VEGF and PEDF (pigment endothelium
derived factor), as evidence is emerging that PEDF may inhibit
new blood vessel growth.134 AdPEDF.11 is an adenovector carrying a progene for human PEDF, and is currently in clinical trials
as a genetic therapy via intravitreous injection for wet AMD.135
CONCLUSIONS
Quantitative understanding of the time course of drugs in the
eye through pharmacokinetic analysis provides mechanistic
189

insight into the fate of drug disposition in this organ. It also aids
in the design of new, and improvement upon existing, ophthalmic
therapies either by enhancing efcacy or reducing toxicity, as
well as in the development of clinical strategies for how to opti-
mize drug use in treatment. As we gain more information about

the pharmacokinetics of ophthalmic drugs in the years to come,
this knowledge will translate into more clinically efcacious,
safer, and more comfortable ocular therapies in the near future.
SECTION 4
REFERENCES
190
1. Fleishaker JC, Ferry JJ: Pharmacokineticpharmacodynamic modeling in drug

development: concepts and applications.
In: Derendorf H, Hochhaus G, eds.
Handbook of pharmacokinetic/
pharmacodynamic correlation. Boca Raton,
FL: CRC Press; 1995:5778.
2. Lima JJ: Pharmacokinetics and
pharmacodynamics. In: Swarbrick J,
Boylan JC, eds. Encyclopedia of
pharmaceutical technology. New York:
Dekker; 1995:2952.
3. Holford NH, Scheiner LB: Kinetics of
pharmacologic response. In: Rowland M,
Tucker G, eds. Pharmacokinetics: theory
and methodology. Elmsworth, NY:
Pergamon; 1986:189212.
4. Lee VH-L, Robinson JR: Review: topical
ocular drug delivery: recent developments
and future challenges. J Ocul Pharmacol
1986; 2:67108.
5. Bourlais CL, Acar L, Zia H, et al: Ophthalmic
drug delivery systems recent advances.
Prog Retin Eye Res 1998; 17:3358.
6. Davies NM: Biopharmaceutical
considerations in topical ocular drug
delivery. Clin Exp Pharmacol Physiol 2000;
27:558562.
7. Kaur IP, Kanwar M: Ocular preparations:
the formulation approach. Drug Dev Ind
Pharm 2002; 28:473493.
8. Kurz D, Ciulla TA: Novel approaches for
retinal drug delivery. Ophthalmol Clin North
Am 2002; 15:405410.
9. Duvvuri S, Majumdar S, Mitra AK: Drug
delivery to the retina: challenges and
opportunities. Expert Opin Biol Ther 2003;
3:4556.
10. Yasukawa T, Ogura Y, Tabata Y, et al: Drug
delivery systems for vitreoretinal diseases.
11. Reddy IK, Ganesan MG: Ocular therapeutics
and drug delivery: an overview. In: Reddy
IK, ed. Ocular therapeutics and drug
delivery: a multi-disciplinary approach.
Lancaster, PA: Technomic; 1996:329.
12. Ghate D, Edelhauser HF: Ocular drug
delivery. Expert Opin Drug Deliv 2006;
3:275287.
13. Maurice DM, Mashima S: Ocular
pharmacokinetics. In: Sear ML, ed.
Handbook of experimental pharmacology:
pharmacology of the eye. New York:
Springer; 1984:19116.
14. Schoenwald RD: Ocular pharmacokinetic/
pharmacodynamic. In: Mitra AK, ed.
Ophthalmic drug delivery system. New
York: Dekker; 1993:83110.
15. Schoenwald RD: Pharmacokinetics in
ocular drug delivery. In: Edman P, ed.
Biopharmaceutics of ocular drug delivery.
Boca Raton, FL: CRC Press;
1993:159191.
16. Shell JW: Pharmacokinetics of topically
applied ophthalmic drugs. Surv Ophthalmol
1982; 26:207218.
17. Urtti A, Salminen L: Animal
pharmacokinetic studies. In: Mitra AK, ed.
Ophthalmic drug delivery system. New
York: Dekker; 1993:121136.
18. Kim H, Lizak M J, Tansey G, et al: Study of

ocular transport of drugs released from an
intravitreal implant using magnetic
resonance imaging. Ann Biomed Eng 2005;
33:150164.
19. Hornof M, Toropainen E, Urtti A: Cell
culture models of the ocular barriers. Eur J
Pharm Biopharm 2005; 60:207225.
20. Saettone MF, Giannaccini B, Barattini F, et al:
The validity of rabbits for investigations on
ophthalmic vehicles: a comparison of four
different vehicles containing tropicamide in
humans and rabbits. Pharm Acta Helv
1982; 57:4755.
21. Chien D-S, Schoenwald RD: Ocular
pharmacokinetics and pharmacodynamics
of phenylephrine and phenylephrine
oxazolidine in rabbit eyes. Pharm Res
1990; 7:476483.
22. Kelly JA, Molyneux PD, Smith SA, et al:
Relative bioavailability of pilocarpine from a
novel ophthalmic delivery system and
conventional eyedrop formulations. Br J
Ophthalmol 1989; 73:360362.
23. Putnam ML, Schoenwald RD, Duffel MW,
et al: Ocular disposition of aminozolamide
in the rabbit eye. Invest Ophthalmol Vis Sci
1987; 28:13731382.
24. Lee VH-L, Luo AM, Li S, et al:
Pharmacokinetic basis for nonadditivity of
intraocular pressure lowering in timolol
combinations. Invest Ophthalmol Vis Sci
1991; 32:29482957.
25. Urtti A, Rouhiainen H, Kaila T, et al:
Controlled ocular timolol delivery: systemic
absorption and intraocular pressure effects
in humans. Pharm Res 1994;
11:12781282.
26. Brechue WF, Maren TH: pH and drug
ionization affects ocular pressure lowering
of topical carbonic anhydrase inhibitors.
Invest Ophthalmol Vis Sci 1993;
34:25812587.
27. Sharir M, Pierce WM Jr, Chen D, et al:
Pharmacokinetics, acidbase balance and
intraocular pressure effects of
ethyloxaloylazolamide: a novel topically
active carbonic anhydrase inhibitor. Exp
Eye Res 1994; 58:107116.
27a. Estimating the Maximum Safe Starting Dose
in Initial Clinical Trials for Therapeutics in
Adult Healthy Volunteers. FDA, 21 Jul 2005.
Available: http://www.fda.gov/Cder/
guidance/index.htm 25 May 2006.
28. Maurice DM: Prolonged-action drops. Int
Ophthal Clin 1993; 33:8191.
29. Maurice DM: The effect of the low blink
rate in rabbits on topical drug penetration.
J Ocul Pharmacol Therapeutics 1995;
11:297304.
30. Urtti A, Salminen L: Minimizing systemic
absorption of topically administered
ophthalmic drugs. Surg Ophthalmol 1993;
37:435456.
31. Lee VH-L, Robinson JR: Mechanistic and
quantitative evaluation of precorneal
pilocarpine disposition in albino rabbits.
J Pharm Sci 1979; 68:673684.
32. Makoid MC, Robinson JR:
Pharmacokinetics of topically applied
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
pilocarpine in the albino rabbit eye.

J Pharm Sci 1979; 68:435443.
Sieg JW, Robinson JR: Mechanistic studies
on transcorneal penetration of
fluorometholone. J Pharm Sci 1981;
70:10261029
Ohtori A, Tojo K: In vivo/in vitro correlation
of intravitreal delivery of drugs with the help
of computer simulation. Biol Pharm Bull
1994; 17:283290.
Tojo K: A pharmacokinetic model for ocular
drug delivery. Chem Pharm Bull (Tokyo)
2004; 52:12901294.
Friedrich SW, Cheng Y-L, Saville BA:
Theoretical corneal permeation model for
ionizable drugs. J Ocul Pharmacol 1999;
9:229249.
Friedrich S, Saville BA, Cheng Y-L, et al:
Pharmacokinetic differences between
ocular inserts and eyedrops. J Ocul
Pharmacol 1996; 12:518.
Grass GM, Lee VH-L: A model to predict
aqueous humor and plasma
pharmacokinetics of ocular applied drugs.
34:22512259.
Stay MS, Xu J, Randolph TW, Barocas VH:
Computer simulation of convective and
diffusive transport of controlled-release
drugs in the vitreous humor. Pharm Res
2003; 20:96102.
Zimmer A: Pharmacokinetics and
pharmacodynamics of pilocarpine-loaded
PBCA nanoparticles in the glaucomatous
rabbit eye. Doctoral thesis. John W. Goethe,
University of Frankfurt, Germany, 1993.
Zimmer A, Kreuter J, Robinson JR: Studies
on transport pathway of PBCA nanoparticles
in ocular tissues. J Microencapsul 1991;
8:497504.
Ahmed I, Patton TF: Disposition of timolol
and inulin in the rabbit eye following
corneal versus non-corneal absorption.
Int J Pharm 1987; 38:921.
Ahmed I, Patton TF: Importance of the
noncorneal absorption route in topical
ophthalmic drug delivery. Invest
Lee VH, Robinson JR: Mechanistic and
quantitative evaluation of precorneal
pilocarpine disposition in albino rabbits.
J Pharm Sci 1979; 68:673684.
Ahmed I, Gokhale RD, Shah MV, et al:
Physicochemical determinants of drug
diffusion across the conjunctiva, sclera, and
cornea. J Pharm Sci 1987; 76:583586.
Ahmed I, Patton TF: Disposition of timolol
and inulin in the rabbit eye following
corneal versus non-corneal absorption.
Int J Pharm 1987; 38:921.
Schoenwald RD, Deshpande G,
Rethwisch DG, et al: Penetration into the
anterior chamber via the conjunctival/scleral
pathway. J Ocul Pharmacol Therapeutics
1997; 13:4159.
Schoenwald RD, Deshpande GS,
Rethwisch DG, Barfknecht CF: Penetration
into the anterior chamber via the
conjunctival/scleral pathway. J Ocul
Pharmacol Ther 1997; 13:4159.
49. Spangler DL, Abelson MB, Ober A,

Gotnes PJ: Randomized, double-masked
comparison of olopatadine ophthalmic
solution, mometasone furoate monohydrate
nasal spray, and fexofenadine
hydrochloride tablets using the conjunctival
and nasal allergen challenge models.
Clin Ther 2003; 25:22452267.
50. Weiner JM, Abramson MJ, Puy RM:
Intranasal corticosteroids versus oral H1
receptor antagonists in allergic rhinitis:
systematic review of randomised controlled
trials. BMJ 1998; 317:16241629.
51. Lee VHL: Precorneal, corneal, and
postcorneal factors. In: Mitra AK, ed.
Ophthalmic drug delivery system. New York:
Dekker; 1993:5981.
52. Van Santvliet L, Ludwig A: The influence of
penetration enhancers on the volume
instilled of eye drops. Eur J Pharm Biopharm
1998; 45:189198.
53. Mishima S, Gasset A, Klyce SD Jr, Baum
JL: Determination of tear volume and tear
flow. Invest Ophthalmol 1966; 5:264276.
54. Chrai SS, Patton TF, Mehta A, et al:
Lacrimal and instilled fluid dynamics in
rabbit eye. J Pharm Sci 1973;
62:11121121.
55. Zaki I, Fitzgerald P, Hardy JG, et al: A
comparison of the effect of viscosity on the
precorneal residence of solutions in rabbit
and man. J Pharm Pharmacol 1986;
38:463466.
56. File RR, Patton TF: Topically applied
pilocarpine. Human pupillary response as a
function of drop size. Arch Ophthalmol
1980; 98:112115.
57. Keister JC, Cooper ER, Missel PJ, et al:
Limits on optimizing ocular drug delivery.
J Pharm Sci 1991; 80:5053.
58. Bartlett JD, Boan K, Corliss D, Gaddie IB:
Efcacy of silicone punctal plugs as
adjuncts to topical pharmacotherapy of
glaucoma a pilot study. Punctal Plugs in
Glaucoma Study Group. J Am Optom
Assoc 1996; 67:664668.
59. Ariturk N, Oge I, Erkan D, et al: The effects
of nasolacrimal canal blockage on topical
medications for glaucoma. Acta
Ophthalmol Scand 1996; 74:411413.
60. Sharir M, Zimmerman TJ: Nasolacrimal
occlusion improves the therapeutic index of
antiglaucoma medications. J Assoc Acad
Minor Phys 1994; 5:6267.
61. Abelson MB, Ousler GW: Instillation of
ophthalmic agents after tear lm break-up
time to enhance treatment effect. Invest
Ophth Vis Sci 2001; 42:E-Abstract 944.
62. Rait JL: Systemic effects of topical
ophthalmic beta-adrenoceptor antagonists.
Aust N Z J Ophthalmol 1999; 27:5764.
63. Rosales T, Isenberg S, Leake R, Everett S:
Systemic effects of mydriatics in low
weight infants. J Pediatr Ophthalmol
Strabismus 1981; 18:4244.
64. Merli GJ, Weitz H, Martin JH, et al: Cardiac
dysrhythmias associated with ophthalmic
atropine. Arch Intern Med 1986; 146:4547.
65. De Saint Jean M, Brignole F, Bringuier AF,
et al: Effects of benzalkonium chloride on
growth and survival of Chang conjunctival
cells. Invest Ophthalmol Vis Sci 1999;
40:619630.
66. Burstein NL: Preservative cytotoxic
threshold for benzalkonium chloride and
chlorhexidine digluconate in cat and rabbit
corneas. Invest Ophthalmol Vis Sci 1980;
19:308313.
67. Burgalassi S, Chetoni P, Monti D,

Saettone MF: Cytotoxicity of potential
ocular permeation enhancers evaluated on
rabbit and human corneal epithelial cell
lines. Toxicol Lett 2001; 122:18.
68. Noecker R: Effects of common ophthalmic
preservatives on ocular health. Adv Ther
2001; 18:205215.
69. De Saint Jean M, Debbasch C, Brignole F,
et al: Toxicity of preserved and unpreserved
antiglaucoma topical drugs in an in vitro
model of conjunctival cells. Curr Eye Res
2000; 20:8594.
70. Berdy GJ, Abelson MB, Smith LM,
George MA: Preservative-free articial tear
preparations. Assessment of corneal
epithelial toxic effects. Arch Ophthalmol
1992; 110:528532.
71. Schaefer K, George MA, Abelson MB,
Garofalo C: A scanning electron
micrographic comparison of the effects of
two preservative-free articial tear solutions
on the corneal epithelium as compared to a
phosphate buffered saline and a 0.02%
benzalkonium chloride control. Adv Exp
Med Biol 1994; 350:459464.
72. Collin HB, Grabsch BE: The effect of
ophthalmic preservatives on the healing
rate of the rabbit corneal epithelium after
keratectomy. Am J Optom Physiol Opt
1982; 59:215222.
73. Di Colo G, Zambito Y, Burgalassi S, et al:
Effect of chitosan and of
N-carboxymethylchitosan on intraocular
penetration of topically applied ofloxacin.
Int J Pharm 2004; 273:3744.
74. Nemoto E, Takahashi H, Kobayashi D, et al:
Effects of poly-L-arginine on the permeation
of hydrophilic compounds through surface
ocular tissues. Biol Pharm Bull 2006;
29:155160.
75. Musson DG, Bidgood AM, Olejnik O:
Comparative corneal penetration of
prednisolone sodium phosphate and
prednisolone acetate in NZW rabbits.
J Ocul Pharmacol 1991; 7:175182.
76. Olejnik O, Weisbecker CA: Ocular
bioavailability of topical prednisolone
preparations. Clin Ther 1990; 12:211.
77. Loftsson T, Stefansson E: Cyclodextrins in
eye drop formulations: enhanced topical
delivery of corticosteroids to the eye. Acta
78. Kaur IP, Garg A, Singla AK, Aggarwal D:
Vesicular systems in ocular drug delivery:
an overview. Int J Pharm 2004; 269:114.
79. Sigurdsson HH, Stefansson E,
Gudmundsdottir E, et al: Cyclodextrin
formulation of dorzolamide and its
distribution in the eye after topical
administration. J Control Release 2005;
102:255262.
80. Wei CP, Anderson JA, Leopold I: Ocular
absorption and metabolism of topically
applied epinephrine and a dipivalyl ester of
epinephrine. Invest Ophthalmol Vis Sci
1978; 17:315321.
81. Mandell AI, Stentz F, Kitabchi AE: Dipivalyl
epinephrine: a new pro-drug in the
treatment of glaucoma. Ophthalmology
1978; 85:268275.
82. Duvvuri S, Majumdar S, Mitra AK: Role of
metabolism in ocular drug delivery. Curr
Drug Metab 2004; 5:507515.
83. Sasaki H, Yamamura K, Mukai T, et al:
Enhancement of ocular drug penetration.
Crit Rev Ther Drug Carrier Syst 1999;
16:85146.
84. Robertson SM, Curtis MA, Schlech BA,

et al: Ocular pharmacokinetics of
moxifloxacin after topical treatment of
animals and humans. Surv Ophthalmol
2005; 50(Suppl 1):S32S45.
85. Wagner RS, Abelson MB, Shapiro A,
Torkildsen G: Evaluation of moxifloxacin,
ciprofloxacin, gatifloxacin, ofloxacin, and
levofloxacin concentrations in human
conjunctival tissue. Arch Ophthalmol 2005;
123:12821283.
86. Saettone MF, Giannaccini B, Teneggi A,
et al: Vehicle effects on ophthalmic
bioavailability: the influence of different
polymers on the activity of pilocarpine in
rabbit and man. J Pharm Pharmacol 1982;
34:464466.
87. Shedden AH, Laurence J, Barrish A,
Olah TV: Plasma timolol concentrations of
timolol maleate: timolol gel-forming solution
(TIMOPTIC-XE) once daily versus timolol
maleate ophthalmic solution twice daily.
Doc Ophthalmol 2001; 103:7379.
88. Ubels JL, Clousing DP, Van Haitsma TA,
et al: Pre-clinical investigation of the
efcacy of an articial tear solution
containing hydroxypropyl-guar as a
gelling agent. Curr Eye Res 2004;
28:437444.
89. Hartstein I, Khwarg S, Przydryga J: An
open-label evaluation of HP-Guar gellable
lubricant eye drops for the improvement of
dry eye signs and symptoms in a moderate
dry eye adult population. Curr Med Res
Opin 2005; 21:255260.
90. Christensen MT, Cohen S, Rinehart J,
et al: Clinical evaluation of an HP-guar
gellable lubricant eye drop for the relief of
dryness of the eye. Curr Eye Res 2004;
28:5562.
91. Kahn M: Bioavailability of vitamin B using a
small-volume nebulizer ophthalmic drug
delivery system. Clin Experiment
Ophthalmol 2005; 33:402407.
92. Davies NM, Farr SJ, Hadgraft J, Kellaway IW:
Evaluation of mucoadhesive polymers in
ocular drug delivery. I. Viscous solutions.
Pharm Res 1991; 8:10391043.
93. Camber O, Edman P, Gurny R: Influence of
sodium hyaluronate on the meiotic effect of
pilocarpine in rabbits. Curr Eye Res 1987;
6:779784.
94. Krishnamoorthy R, Mitra AK: Mucoadhesive
polymers in ocular drug delivery. In:
Mitra AK, ed. Ophthalmic drug delivery
system. New York: Dekker; 1993:199221.
95. Tian X, Iwatsu M, Kanai A: Disposable
1-day Acuvue contact lenses for the
delivery of lomefloxacin to rabbits eyes.
Clao J 2001; 27:212215.
96. Pijls RT, Sonderkamp T, Daube GW, et al:
Studies on a new device for drug delivery
to the eye. Eur J Pharm Biopharm 2005;
59:283288.
97. Sasaki H, Nagano T, Sakanaka K, et al:
One-side-coated insert as a unique
ophthalmic drug delivery system. J Control
Release 2003; 92:241247.
98. Hariprasad SM, Shah GK, Chi J, Prince RA:
Determination of aqueous and vitreous
concentration of moxifloxacin 0.5% after
delivery via a dissolvable corneal collagen
shield device. J Cataract Refract Surg
2005; 31:21422146.
99. Kuster P, Taravella M, Gelinas M, Stepp P:
Delivery of trifluridine to human cornea and
aqueous using collagen shields. Clao J
1998; 24:122124.
CHAPTER 17
191
SECTION 4
192
100. Di Colo G, Burgalassi S, Chetoni P, et al:

Gel-forming erodible inserts for ocular
controlled delivery of ofloxacin. Int J Pharm
2001; 215:101111.
101. Eljarrat-Binstock E, Raiskup F, Stepensky D,
et al: Delivery of gentamicin to the rabbit
eye by drug-loaded hydrogel iontophoresis.
45:25432548.
102. Frucht-Pery J, Mechoulam H, Siganos CS,
et al: Iontophoresis-gentamicin delivery into
the rabbit cornea, using a hydrogel delivery
probe. Exp Eye Res 2004; 78:745749.
103. Halhal M, Renard G, Courtois Y, et al:
Iontophoresis: from the lab to the bed side.
Exp Eye Res 2004; 78:751757.
104. Hayden BC, Jockovich ME, Murray TG, et al:
Pharmacokinetics of systemic versus focal
Carboplatin chemotherapy in the rabbit
eye: possible implication in the treatment of
retinoblastoma. Invest Ophthalmol Vis Sci
2004; 45:36443649.
105. Smolin G, Okumoto M, Feiler S, Condon D:
Idoxuridine-liposome therapy for herpes
simplex keratitis. Am J Ophthalmol 1981;
91:220225.
106. Chetoni P, Rossi S, Burgalassi S, et al:
Comparison of liposome-encapsulated
acyclovir with acyclovir ointment: ocular
pharmacokinetics in rabbits. J Ocul
107. Law SL, Huang KJ, Chiang CH: Acyclovircontaining liposomes for potential ocular
delivery. Corneal penetration and
absorption. J Control Release 2000;
63:135140.
108. Ito Y, Cai H, Koizumi Y, et al: Effects of lipid
composition on the transcorneal
penetration of liposomes containing
disulram, a potential anti-cataract agent,
in the rabbit. Biol Pharm Bull 2000;
23:327333.
109. Nagarsenker MS, Londhe VY, Nadkarni GD:
Preparation and evaluation of liposomal
formulations of tropicamide for ocular
delivery. Int J Pharm 1999; 190:6371.
110. Pleyer U, Elkins B, Ruckert D, et al: Ocular
absorption of cyclosporine A from
liposomes incorporated into collagen
shields. Curr Eye Res 1994; 13:177181.
111. Bochot A, Mashhour B, Puisieux F, et al:
Comparison of the ocular distribution of a
model oligonucleotide after topical
instillation in rabbits of conventional and
new dosage forms. J Drug Target 1998;
6:309313.
112. Kawakami S, Yamamura K, Mukai T, et al:
Sustained ocular delivery of tilisolol to
rabbits after topical administration or
intravitreal injection of lipophilic prodrug
incorporated in liposomes. J Pharm
Pharmacol 2001; 53:11571161.
113. Masuda I, Matsuo T, Yasuda T, Matsuo N:
Gene transfer with liposomes to the
intraocular tissues by different routes of
administration. Invest Ophthalmol Vis Sci
1996; 37:19141920.
114. Diepold R, Kreuter J, Himber J, et al:

Comparison of different models for the
testing of pilocarpine eyedrops using
conventional eyedrops and a novel depot
formulation (nanoparticles). Graefes Arch
Clin Exp Ophthalmol 1989; 227:188193.
115. Chiang CH, Tung SM, Lu DW, Yeh MK: In
vitro and in vivo evaluation of an ocular
delivery system of 5-fluorouracil
microspheres. J Ocul Pharmacol Ther
2001; 17:545553.
116. Pignatello R, Bucolo C, Ferrara P, et al:
Eudragit RS100 nanosuspensions for the
ophthalmic controlled delivery of ibuprofen.
Eur J Pharm Sci 2002; 16:5361.
117. Chu TC, He Q, Potter DE: Biodegradable
calcium phosphate nanoparticles as a new
vehicle for delivery of a potential ocular
hypotensive agent. J Ocul Pharmacol Ther
2002; 18:507514.
118. Cavalli R, Gasco MR, Chetoni P, et al: Solid
lipid nanoparticles (SLN) as ocular delivery
system for tobramycin. Int J Pharm 2002;
238:241245.
119. de Campos AM, Diebold Y, Carvalho EL,
et al: Chitosan nanoparticles as new ocular
drug delivery systems: in vitro stability, in
vivo fate, and cellular toxicity. Pharm Res
2004; 21:803810.
120. Weijjtens O, Feron EJ, SchoemakerRC, et al:
High concentration of dexamethasone in
aqueous and vitreous after subconjunctival
injection. Am J Ophthalmol 1999;
128:192197.
121. Weijtens O, Schoemaker RC, Romijn FP,
et al: Intraocular penetration and systemic
absorption after topical application of
dexamethasone disodium phosphate.
Ophthalmology 2002; 109:18871891.
122. Maurice DM, Ota Y: The kinetics of
subconjunctival injections. Jpn J Ophthalmol
1978; 22:95100.
123. Wine NA, Gornall AG, Basu PK: The ocular
uptake of subconjunctivally injected C14
hydrocortisone. Time and major route of
penetration in a normal eye. Am J
Ophthalmol 1964; 58(Pt 1):362366.
124. Lee TW, Robinson JR: Drug delivery to the
posterior segment of the eye III: the effect of
parallel elimination pathway on the vitreous
drug level after subconjunctival injection.
J Ocul Pharmacol Ther 2004; 20:5564.
125. Moritera T, Ogura Y, Honda Y, et al:
Microspheres of biodegradable polymers as
a drug-delivery system in the vitreous. Invest
126. Saishin Y, Silva RL, Callahan K, et al:
Periocular injection of microspheres
containing PKC412 inhibits choroidal
neovascularization in a porcine model.
44:49894993.
127. Behar-Cohen FF, El Aouni A, Gautier S, et al:
Transscleral Coulomb-controlled
iontophoresis of methylprednisolone into
the rabbit eye: influence of duration of
treatment, current intensity and drug
128.
129.
130.
131.
132.
133.
134.
135.
136.
137.
138.
139.
140.
141.
concentration on ocular tissue and fluid

levels. Exp Eye Res 2002; 74:5159.
Thanos CG, Bell WJ, ORourke P, et al:
Sustained secretion of ciliary neurotrophic
factor to the vitreous, using the
encapsulated cell therapy-based NT-501
intraocular device. Tissue Eng 2004;
10:16171622.
Dhillon B, Kamal A, Leen C: Intravitreal
sustained-release ganciclovir implantation
to control cytomegalovirus retinitis in AIDS.
Int J STD AIDS 1998; 9:227230.
Lim LL, Smith JR, Rosenbaum JT: Retisert
(Bausch & Lomb/control delivery systems).
Curr Opin Investig Drugs 2005; 6:11591167.
Beeley NR, Stewart JM, Tano R, et al:
Development, implantation, in vivo elution,
and retrieval of a biocompatible, sustained
release subretinal drug delivery system.
J Biomed Mater Res A 2006; 76:690698.
Gaudreault J, Fei D, Rusit J, et al:
Preclinical pharmacokinetics of
Ranibizumab (rhuFabV2) after a single
intravitreal administration. Invest
Kompella UB, Bandi N, Ayalasomayajula
SP: Subconjunctival nano- and
microparticles sustain retinal delivery of
budesonide, a corticosteroid capable of
inhibiting VEGF expression. Invest
Schlingemann RO: Role of growth factors
and the wound healing response in agerelated macular degeneration. Graefes Arch
Campochiaro PA, Nguyen QD, Shah SM,
et al: Adenoviral vector-delivered pigment
epithelium-derived factor for neovascular
age-related macular degeneration: results
of a phase I clinical trial. Hum Gene Ther
2006; 17:167176.
Carney LG, Hill RM: Human tear buffering
capacity. Arch Ophthalmol 1979;
97:951952.
Olejnik O: Conventional systems in
ophthalmic drug delivery. In: Mitra AK, ed.
Ophthalmic drug delivery system. New York:
Dekker; 1993:177198.
Stjernschantz J, Astin M: Anatomy and
physiology of the eye, physiological
aspects of ocular drug therapy. In: Edman
P, ed. Biopharmaceutics of ocular drug
delivery. Boca Raton, FL: CRC Press;
1993:125.
Watsky MA, Jablonski M, Edelhauser HF:
Comparison of conjunctival and corneal
surface areas in rabbit and human. Curr
Eye Res 1988; 7:483486.
Conrad JM, Robinson JR: Aqueous
chamber distribution volume measurement
in rabbits. J Pharm Sci 1977; 66:219224.
Hutak CM, Jacaruso RB: Evaluation of
primary ocular irritation: alternatives to the
Draize test. In: Reddy IK, ed. Ocular
therapeutics and drug delivery: a
multidisciplinary approach. Lancaster, PA:
Technomic; 1996:489525.
CHAPTER
18
Anesthetics
Padma Gulur, David Weber, and Martin A. Acquadro
Anesthesia, over the years, has become very safe. The risk from
anesthesia when compared to the risk due to patient and
surgical factors is relatively low.1 This can be attributed in part
to better agents, medications, and advances in monitoring.
The American Society of Anesthesiologists has stratied the
risk of patients undergoing anesthesia (Table 18.1). This has
become a useful universal nomenclature. Modern practice of
this specialty strives for anesthesia, analgesia, amnesia,
areflexia, and autonomic stability. General anesthesia usually
involves premedication, induction, maintenance, and recovery.
PREMEDICATION
Premedication for alleviation of anxiety is not a substitute for
adequate preoperative discussion with the patient. A study
comparing various techniques including, no preoperative visit
or drug, preoperative discussion alone, premedications alone,
and preoperative discussion with premedication, demonstrated
interesting results. The patients who displayed the most anxiety
were those who were premedicated without preoperative
discussion or consultation. The patients with the least anxiety
were those who had both preoperative discussion and preoperative medication. The patients who had only preoperative
discussion, without any premedication, were not much more
anxious, as a percentage of the population studied, than those
who received both preoperative discussion and premedication.1
TABLE 18.1. American Society of Anesthesiologists

Classication of Preoperative Risk
ASA Class
Systemic Disturbance
Mortality
1.
Healty patient with no disease outside

of the surgical process
<0.03%
2.
Mild to moderate systemic disease

caused by the surgical condition or
by other pathological processes,
medically well-controlled
0.2%
3.
Severe disease process which limits

activity but is not incapacitating
1.2%
4.
Severe incapacitating disease

process that is a constant threat to life
8%
5.
Moribund patient not expected to

survive 24 h with or without an
operation
34%
E.
Sufx to indicate emergency surgery

for any class
Increased
Adapted from Cohen MM, Ducan PG, Tate RB, JAMA 1988; 260:2859.
The goals of preanesthetic medication include decreased

anxiety, analgesia if preoperative pain is evident, and, if necessary, diminished airway secretions and diminished gastric
acidity and volume. All these goals should be accomplished
without excessive sedation, which could compromise the
cardiopulmonary system.24
BENZODIAZEPINES
Anxiolytics are usually used as preoperative medications.
Benzodiazepines (Table 18.2) are the most common of the
anxiolytic agents. When given in the usual doses, they produce
the greatest relief of anxiety with the least cardiopulmonary
depression. These drugs are rarely implicated as a cause of
nausea and vomiting. They can raise the threshold for central
nervous system (CNS) toxicity of local anesthetics5 and are not
analgesic, but they compound the anxiolytic effects of some
analgesics in small to moderate doses. Diazepam is usually
given orally. The solvent used in parenteral preparations can
result in pain and phlebitis. Lorazepam can be given orally or
parenterally and often produces amnesia. It can also result in
prolonged sedation.4
Midazolam has become popular because of its water solubility,
rapid onset and short duration of action, and reliability. It can
be given intramuscularly or intravenously and often produces
amnesia with few side effects. Mental function returns to
normal within 4 h, making midazolam a popular choice for
ambulatory surgery and regional anesthesia. Diazepam is more
likely to produce cumulative effects than lorazepam or
midazolam.4
NARCOTICS
If the patient experiences preoperative pain, morphine is an
effective preoperative analgesic (Table 18.3). The choice of
narcotic is usually governed by the desired duration of activity.
Morphines clinical effects persist 46 h; fentanyls action lasts
~12 h. Urinary retention, wheezing, constipation, nausea,
and vomiting are not uncommon with opioid analgesics. The
respiratory depressant action of morphine may cause hypoventilation and increased carbon dioxide tension with resultant
increased intracranial pressure. Advantages and disadvantages
need to be considered in the decision to use opioids in preanesthetic medication.2,4
Meperidine is used commonly as an intramuscular medication. There is some concern that the metabolite of meperidine,
normeperidine, may result in confusion, agitation, and seizures,
particularly in the elderly, in patients with renal failure, and in
children. This is more often a problem with long-term repeated
dosing.
193
TABLE 18.2. Premedicants Anxiolytics and Hypnotics4,7,10

Agent
Dosage
Metabolism
IV: 0.030.07 mg/kg
tE: 14 h
M: Liver
E: Kidney
Effects
Benzodiazepines
Midazolam
Diazepam
IV: 0.030.1 mg/kg

PO: 0.050.15 mg/kg
tE: 710 h (28 days

for active metabolites)
M: Liver
E: Kidney
Lorazepam
IV: 0.05 mg/kg

PO: 110 mg/kg
tE: 14 h
M: Liver
E: Kidney
IV: 1 mg/kg up to
500 mg
IM: 100200 mg
PO: 100200 mg
tE: 2050 h
M: Liver
E: Kidney (mostly),
liver
CNS depression,
amnesia, seizure
threshold, BP and
respiratory
depression; may
cause paradoxical
CNS excitement
Barbiturates
Pentobarbital
CNS depression; may

cause depression of
BP, hiccoughs,
laryngospasm,
respiratory
depression,
exacerbation of
porphyria; agents
cross placenta; may
antagonize oral
anticoagulants
SECTION 4
BP, blood pressure; CNS, central nervous system; E, route of excretion; IM, intramuscularly; IV, intravenously;
M, site of metabolism; PO, per os; tE, elimination halflife.
TABLE 18.3. Premedicants Analgesics4,7,10

Agent
Dosage
Metabolism
Opioid Effects
Meperidine
IV: 0.51.0 mg/kg

IM/SC: 0.51.0 mg/kg
PO: 1 mg/kg q 24 h
tE: 1.54 h
M: Liver
E: Kidney
Morphine
IV: 0.1 mg/kg

IM: 0.1 mg/kg
t E: 24 h
M: Liver
E: Kidney
Fentanyl
IV: 10100 mg
IM: 50100 mg/kg
t1/2: a: 12 min
t E: 4 h
M: Liver
E: Kidney
Analgesic, CNS depression,

euphoria, respiratory depression,
bronchospasm (rare), blood
pressure, nausea, vomiting,
dysphoria, biliary pressure,
gastrointestinal/genitourinary
motility. Agents cross placenta.
Greater incidence of skeletal
rigidity with fentanyl (accumulation
with frequent dosing). Narcotics,
particularly Demerol, should be
avoided in patients taking MAO
inhibitors.
E, route of excretion; IM, intramuscularly; IV, intravenously; M, site of metabolism; PO, per os; SC,
subcutaneously; tE, elimination halflife; t1/2, halflife.
Remifentanil is an ultra-short-acting opioid, unique among

the other opioids, secondary to rapid metabolism rather than
redistribution. Rapid metabolism occurs from hydrolysis of a
methyl ester side chain by blood and tissue esterases. Because
of its short duration of action, it is generally administered by
continuous intravenous infusion.
ANTIEMETICS
194
Ondansetron hydrochloride is a commonly used selective

blocking agent of the serotonin 5-HT3 receptor, administered
orally or intravenously. These serotonin 5-HT3 receptors are
found centrally in the area-postrema-chemoreceptor trigger
zone, and peripherally on vagal nerve terminals. Although it is
not certain whether ondansetrons effectiveness comes from
antagonism of central, peripheral, or both receptor sites, the

drug is effective against perioperative and chemotherapy-induced
emesis. The incidence of side effects is low when the drug is
given in normal doses to normal patients.6
BUTYROPHENONES
The most common butyrophenone is droperidol (Table 18.4). In
adults it is an antiemetic in very small doses, so that
cardiopulmonary stability is maintained. It should be noted
that droperidol does have a1-adrenergic-blocking activity and
must be given with caution if hypotension is already evident.
Restlessness and extrapyramidal dyskinesia may be noted.
Atropine is an effective antidote. A patient may exhibit
catatonia and appear outwardly calm though he or she is in fact
Anesthetics
TABLE 18.4. Premedicants Antiemetics4,6,7,10

Agent
Dosage
Metabolism
Effects
Droperidol
IV: 0.6252.5 mg
IM: 2.510 mg
tE: ?
M: Liver
E: Kidney, liver
Antiemetic, antipsychotic. May

cause dysphoria, extrapyramidal
effects, hypotension secondary to
a-blockade. Black box warning in
effect from FDA for QT
prolongation
Hydroxyzine
IM: 25100 mg
PO: 25100 mg
tE: 3 h
M: Liver
E: Liver, kidney
CNS depression, antiemetic

effects, antagonism of histamine
action on H1 receptors. May cause
dry mouth
Ondansetron
IV: 4 mg slow
PO: 816 mg,
1 h before induction
M: Liver
E: Kidney, liver
Antiemetic chemotherapy and

postanesthesia N/V. Hypotension,
bradycardia, tachycardia, angina,
second degree heart block,
bronchospasm, extrapyramidal
effects, seizures
E, route of excretion; IM, intramuscularly; IV, intravascularly; M, site of metabolism; PO, per os; tE, elimination
half-life.
Agent
Dosage
Metabolism
Effects
Cimetidine
IV/IM/PO: 300 mg q 68 h
tE: 2 h
M: Liver
E: Kidney
May increase blood levels

of propranolol or
benzodiazepines and
potentiate oral
anticoagulants; may cause
confusion
Ranitidine
IV/IM: 50 mg q 68 h
PO: 150 mg q 12 h
tE: 23 h
M: Liver
E: Kidney
Antagonizes histamine
action on H2 receptors with
decreased gastric acid
secretion
Metoclopramide
IV/IM: 10 mg
PO: 1015 mg
tE: 26 h
M: Liver
E: Kidney
Gastrointestinal motility
and esophageal sphincter
tone; extrapyramidal
symptoms (rare)
CHAPTER 18
TABLE 18.5. Premedicants Antagonists and Gastrokinetic Agents4,7,10
E, route of excretion; IM, intramuscularly; IV, intravenously; M, site of metabolism; PO, per os; t E, elimination
half-life.
experiencing panic secondary to dysphoria produced by the

action of droperidol.4
Dexmedetomidine is a more specic alpha-2 agonist that is

also gaining popularity as a premedication. Bradycardia and dry
mouth are possible side effects with this class of medications.
ANTIHISTAMINES
Antihistamines are occasionally used as premedication because
these act as anxiolytics as well as H1 histamine receptor blockers.
Hydroxyzine and diphenhydramine are common agents.4
ALPHA-2 AGONISTS
Clonidine is a centrally acting alpha-2 agonist used as a
premedication for its sedative properties and for attenuation
of autonomic reflexes such as hypertension, tachycardia and
cathecholamine release associated with preoperative anxiety
and surgical stimulus. Caution is advised in patients on clonidine for long periods due to risk of rebound hypertension with
its withdrawal. Dose is usually 2 mg/kg orally.
ANTICHOLINERGICS
This class of drugs is not routinely used as a premedication
in present practice. When used, these are chosen for their
antisialogogand sedative effects as well as for prevention of
reflex bradycardia with the latter being the most common
reason for their use as a premedication. This is especially true
of atropine and glycopyrrolate. Glycopyrrolate does not cross
the bloodbrain barrier and therefore has the least of the undesirable side effects of these drugs (central anticholinergic
syndrome). Physostigmine (1560 mg/kg) is a specic treatment
for this syndrome due to atropine or scopolamine (Table 18.5).
Scopolamine in the form of a patch is gaining popularity as
an effective antiemetic agent. It is applied preoperatively and is
195

worn for 72 h. Side effects of this class of drugs include central
anticholinergic syndrome, tachycardia, lower esophageal
sphincter relaxation, body temperature increase, drying of airway secretions and an increase in physiologic dead space.
H2 HISTAMINE RECEPTOR ANTAGONISTS

Cimetidine and ranitidine block H2 receptors and decrease
gastric acid secretion. Ranitidine has become more popular,
because it appears to cause fewer cardiovascular and CNS side
effects than cimetidine.4 Routine use of these medications is
not recommended (Table 18.6). Instead, these are usually
reserved for patients at high risk for aspiration.
ANTACIDS
Particulate and nonparticulate antacids effectively raise gastric
acid pH. If aspiration is a concern, a nonparticulate antacid is
preferred, because particulate antacids may cause more lung
damage. Sodium citrate is a commonly used nonparticulate
antacid.4
GENERAL ANESTHETICS
Summary Box
Premedications
The goals of preanesthetic medication include decreased
anxiety, analgesia if preoperative pain is evident, and, if
necessary, diminished airway secretions and diminished
gastric acidity and volume
Decreased anxiety is the most common reason for
premedication and benzodiazepines (midazolam in particular)
are the commonly used class of medications. Midazolam is
ideal based on its short duration of action and favorable
pharmacokinetics
Anticholinergics are used if reducing airway secretions is the
goal
For diminished gastric acidity and volume, H2 receptor
antagonists, antacids (especially Bicitra which is a clear nonparticulate antacid), and gastrointestinal motility agents such
as metaclopromide are used
SECTION 4
GASTROINTESTINAL MOTILITY AGENTS

Metoclopramide, a dopaminergic antagonist, increases gastrointestinal motility and pyloric relaxation, thereby increasing the
speed of gastric emptying. Sodium citrate or anticholinergic
agents may interfere with the action of metoclopramide.4
Cisapride is another medication in this class of drugs. The
antibiotic erythromycin is being touted for use as a gastric
emptying agent to decrease risk prior to emergency anesthesia.
PHARMACOKINETICS
Among benzodiazepines, diazepam is metabolized by the liver,
with one-third of the metabolites being oxazepam. The active
metabolites are principally excreted by the kidneys. In general,
the benzodiazepines, barbiturates, and antihistamines are
metabolized by the liver and excreted by the kidneys, though
the amount of drug eliminated by the kidneys and liver varies
somewhat.4,7 Ondansetron and the butyrophenones are also
metabolized by the liver and excreted by the kidneys.6 Ten
percent of droperidol is excreted unchanged.4,7 Morphine is
metabolized by the liver and excreted by the kidneys, as are
the other opioids. Tables 18.2 to 18.6 list many of the drugs
commonly used.4,7
A patient under general anesthesia has no perception of any

sensation. This state, which allows surgical procedures to be
performed, can be induced with a wide variety of drugs, usually
used in combination. The objectives of a general anesthetic
include analgesia, unconsciousness, and absence of movement
and autonomic stability.811
General anesthetics are commonly administered intravenously
or inhalationally. These routes are preferred over the intramuscular or oral route because of greater drug predictability and
reliability. Common inhalational and intravenous agents are
reviewed in this section.10
INHALATIONAL AGENTS
The common inhalational general anesthetic agents include
nitrous oxide and the halogenated agents like halothane,
enflurane, isoflurane, desflurane, and sevoflurane. Enflurane is
no longer commonly used due to risk of seizures. To compare
various inhalational agents and the concentrations in the
alveoli during steady state that produce equivalent levels of
anesthesia, the concept and denition of minimum alveolar
concentration (MAC) are necessary. The MAC of anesthetic at
TABLE 18.6. Premedicants Anticholinergics4,7,10
196
Agent
Dosage
Metabolism
Effects
Atropine
IV/IM: 0.41.2 mg
t1/2 a: 1 min
tE: 2 h
M: Minimal
E: Kidney
(some by liver)
Tachydysrhythmias, dry mouth,

urinary retention; crosses
bloodbrain barrier
and placenta
Scopolamine
IV/IM: 0.30.6 mg
tE: 3 h
E: Kidney
Crosses bloodbrain barrier

and placenta; may cause
excitement or delirium; superior
antisialogog
Glycopyrrolate
IV/IM/SC: 0.10.2 mg
PO: 12 mg
E: Kidney
Does not cross bloodbrain

barrier or placenta; otherwise
similar to atropine
E, route of excretion; IM, intramuscularly; IV, intravenously; M, site of metabolism; PO, per os; tE, elimination
half-life.
Anesthetics
excitation during ventilatory induction when compared with
enflurane, isoflurane, or desflurane.6,10 The principal advantages
of sevoflurane and desflurane over halothane, enflurane, and
isoflurane are their low solubility in blood, which produces
rapid induction of anesthesia, and low tissue solubility, which
results in rapid elimination and awakening.6
The depth of anesthesia with the halogenated agents can be
judged by observing blood pressure, because they produce dosedependent reductions of arterial blood pressure principally
through peripheral vasodilation.6,10,17 There should be little
change of pulse rate or blood pressure and no body movement
in response to surgical stimulation. Following induction of
anesthesia with the halogenated agents halothane or sevoflurane, or with hypnotic intravenous agents such as thiopental or
propofol, the clinician should start with a high-inspired
concentration of the inhalational agent. As maintenance of
anesthesia proceeds, the inspired concentration of anesthetic is
lowered, because the alveolar concentration increases during
maintenance.17 As a steady state is approached, based on
patient response to surgical stimulation, further appropriate
concentration adjustments of the inhalational agents can be
made rapidly.11,17
Summary for Inhalational Agents
Inhalational anesthetics form the mainstay of maintenance

when general anesthesia is administered. They may also be
used for induction in patients without intravenous access. This
practice is more common in children
Commonly used Inhalational anesthetics today are isoflurane,
sevoflurane, and desflurane. Halothane has fallen out of favor
for its dysrhythmogenic potential and association with
halothane hepatitis. Enflurane is rarely used secondary to its
epileptogenic potential
Sevoflurane is the most commonly used inhalational agent for
induction of anesthesia as it does not irritate the airways.
Nitrous oxide is usually used in conjunction with narcotics for
the maintenance of anesthesia
one atmosphere that produces immobility in 50% of patients or

animals exposed to a noxious stimulus is a useful measure of
potency of inhalational agents.5,10,12,13
Anesthetic potency is correlated with lipophilia. The more
potent the general anesthetic, the more lipophilic it is.10
Researchers cannot agree on one specic mechanism of action;
many believe that general anesthetics work at many different
levels and by a variety of mechanisms. This may explain why
diverse inorganic and organic compounds can bring on the state
of general anesthesia. The various theories of the mechanism of
action of general anesthetics are reviewed in references 1416.
Although general inhalational anesthesia can start with
administration of oxygen, nitrous oxide, and an inhalation
agent, the more common technique is to administer a hypnotic,
such as propofol or thiopental sodium (Pentothal), intravenously.17 General inhalational anesthesia is often maintained
with oxygen, nitrous oxide, and a halogenated agent.10
Additional agents may include opiates or muscle relaxants.
The decisions to administer inhalational agents by mask or
endotracheal intubation, and to allow the patient to breathe
spontaneously or to control ventilation, are based on surgical
and anesthetic requirements.
Halothane, Enflurane, Isoflurane, Desflurane, and

Sevoflurane
Cardiovascular system
With all ve agents, blood pressure decreases by peripheral
vasodilation as the depth of anesthesia increases. Cardiac output with halothane decreases 2050% from the baseline value.
The decrease in cardiac output is less with enflurane, desflurane, and sevoflurane. Cardiac output is well maintained with
isoflurane. Heart rate decreases most with halothane, less
with enflurane, desflurane, and sevoflurane, and may increase
with isoflurane. This may explain why cardiac output is
maintained by use of isoflurane. All ve agents diminish
baroreceptor reflex responses (tachycardia) to hypotension and
vasomotor reflex responses (increased peripheral resistance) to
hypovolemia, and they produce little change in the sympathoadrenal response and levels of catecholamines in the plasma.6,10
Inotropy and contractility diminish with all ve agents, most
notably with halothane. Negative inotropy is less obvious and
similar with equipotent concentrations of isoflurane and
sevoflurane. Desflurane produces the least negative inotropy.
All ve agents diminish sympathetic activity and increase vagal
predominance, particularly halothane. This is most common
when halothane is given to a child, especially in association
with manipulation of the airway.6,10
Like isoflurane, desflurane in typical clinical settings does not
sensitize the heart to catecholamines; in one study, however,
Pharmacodynamics
Figure 18.1 shows the chemical structures of the general
inhalational halogenated anesthetic agents in common use.10
Enflurane and isoflurane are ethers with a difluoromethyl group
bonding to the one carbon via an ether bond. The newer
halogenated agents, desflurane and sevoflurane, are also ethers.
Desflurane is a fluorinated methyl ethyl ether, and sevoflurane
is a fluorinated isopropyl ether. For children and adults,
halothane and sevoflurane are far less irritating to breathe and
have a lower incidence of coughing, laryngospasm, and
CI
Br
CI
Halothane
F
O
CI
Enflurane
F
O
Desflurane
C
F
FIGURE 18.1. Chemical structure of ve

commonly used inhalational agents.6,10
F
O
CHAPTER 18
Isoflurane
F2C
H
OCH2F
F2C
Sevoflurane
197
SECTION 4
the ventricular arrhythmogenic threshold of sevoflurane was

between that of enflurane and isoflurane with submucosal
injection of epinephrine.6 Dysrhythmias are most common
with halothane. Reentrant tachycardia is common, because the
normal conduction pathway is slowed and the refractory period
of the conductive tissue is increased. Increased automaticity
also occurs with halothane, which is augmented by adrenergic
agonists. Exogenous epinephrine should be limited in local
anesthetics to a concentration of 1:100 000. No more than
0.1 mg of epinephrine in 10 min or 0.3 mg of epinephrine in 1 h
should be administered when halothane is used.9 With enflurane, isoflurane, desflurane, or sevoflurane, three times this
amount may be permissible. Unlike isoflurane, the other halogenated agents halothane, enflurane, desflurane, and probably
sevoflurane do not cause coronary artery vasodilation that
may lead to coronary artery steal syndrome. With the exception
of isoflurane, the coronary circulation generally remains
responsive to myocardial demands for oxygen. With isoflurane,
coronary blood vessels are maximally dilated at ~1.5 MAC.
Blood flow is maintained despite decreased myocardial oxygen
demand. Some patients with ischemic heart disease have
narrowed blood vessels in some regions of myocardium. These
regions depend on collateral vessels for their blood supply.
Dilation of normal coronary vessels by isoflurane may result in
a steal of blood from the collateral vessels that exacerbates
ischemia.6,18
Pulmonary system
The halogenated agents all cause increasing respiratory
depression as the concentration of the agent is increased. They
all cause a moderate (~20%) increase in PaCO2 that reflects an
increase in the rate of breathing, though, insufcient to offset a
decrease in tidal volume. Minute volume is reduced with all ve
agents. Depression of ventilation reflects a direct depressant
effect on the medullary ventilatory center and perhaps
peripheral effects on intercostal muscle function. Bronchial
smooth muscle relaxation may be produced by a direct effect or
indirectly by reductions in afferent nerve trafc or central
medullary depression of bronchoconstriction reflexes.6 With
all ve agents, respiratory depression is more evident when
opioids are used; assisted or controlled ventilation is usually
administered to avoid excessive hypercarbia. Hypercarbia in
relation to dysrhythmia potential can be more problematic with
halothane than with the other halogenated agents. With all ve
inhalational agents, pulmonary exchange of oxygen becomes
less efcient, and an inspired oxygen concentration of 35% or
more is indicated. All produce blunting of hypoxic pulmonary
vasoconstriction, which can result in increased pulmonary
shunt flow of blood.
All ve agents produce increase in secretions, coughing,
and laryngospasm, though halothane and sevoflurane are least
often problematic. This is why sevoflurane and the less costly
halothane are often employed in spontaneously ventilated
children and adult patients for induction of anesthesia. For
patients who tolerate an intravenous line at the start of
anesthesia, and no anticipated problems with endotracheal
intubation, intravenous induction is generally the method of
choice.6,10,19
Nervous system
198
Of the ve agents, enflurane is associated with a higher

incidence of seizure activity. The seizures are short-lived and
self-limited and generally can be prevented by avoiding deep
anesthesia or hyperventilation. Interestingly, the drug does not
appear to aggravate seizures in epileptic patients, but avoidance
of enflurane is recommended for these patients. The halogenated agents have similar effects on the CNS. With the
halogenated agents, cerebral oxygen consumption is decreased.

There is also basal vasodilation, and cerebral blood flow is
increased whereas perfusion pressure remains constant. As a
result, intracranial pressure is increased. All effects are most
marked with halothane. The cerebrovascular system remains
responsive to carbon dioxide tension; with hyperventilation,
cerebral blood flow, metabolism, and intracranial pressure are
reduced.10,19
Muscular system
All ve halogenated agents reduce the response of skeletal
muscle to nerve stimulation and enhance the neuromuscular
blocking effects of depolarizing and nondepolarizing muscle
agents. All ve agents produce uterine vasodilation and a dosedependent decrease in uterine blood flow. The halogenated
agents have a direct muscle relaxing effect and appear to act
centrally as well as peripherally at the neuromuscular junction.
The halogenated agents potentiate muscle relaxants, and less
neuromuscular blocking agent is required. The least potentiation occurs with halothane and nitrous oxide. Potentiation of
neuromuscular blocking drugs may involve desensitization of
the postjunctional membrane. Any of the three halogenated
agents can trigger malignant hyperthermia.6,10
Renal system
All ve agents cause a dose-dependent reduction in renal blood
flow and glomerular ltration rate. The effects can be somewhat
attenuated by preoperative hydration and prevention of
hypotension. The changes in renal function are rapidly reversed
on conclusion of anesthesia and during recovery. The quantity
of fluoride released by metabolism is least with desflurane,
followed by isoflurane, and these agents are most frequently
used for patients with renal disease. Sevoflurane undergoes
oxidative metabolism in the liver with a serum fluoride concentration of ~22 mmol/L after a 1-MAC-hour exposure. The
magnitude of sevoflurane metabolism resembles that of enflurane (peak plasma fluoride concentrations after a 2.5-MAChour exposure to enflurane are ~20 mmol/L).6 When enflurane
is used in the presence of renal failure, concentrations of
fluoride ion decline rapidly after the anesthetic is discontinued.
It is postulated that much of the fluoride enters bone. It is
therefore probable that anesthesia with enflurane or sevoflurane
is safe for patients with renal disease.6,10,19
Gastrointestinal system
With halothane, enflurane, and isoflurane, and probably with
desflurane and sevoflurane, blood flow decreases with increasing
depth of anesthesia as systemic arterial pressure declines. There
is no evidence of direct ischemia. Hepatic necrosis has been
reported with repeated administration of enflurane. Hepatic
failure has not been reported with isoflurane. Isoflurane is less
metabolized by the liver when compared with enflurane and
halothane; this could be the reason why isoflurane is not linked
to hepatic failure. Halothane has been studied most extensively.
The diagnosis of halothane hepatitis is one of exclusion. The
pathologic appearance of hepatitis is similar whether the cause
is sensitivity to halothane, damage by some other hepatotoxic
drug, or transmission of hepatitis virus. The National Halothane
Study of 1966, a retrospective analysis of more than 850 000
administrations of anesthetics, suggested a small incidence of
hepatic necrosis in which there was no damage by some other
hepatotoxic drug, no transfusion of blood, and no evidence of
transmission of hepatitis virus or involvement of the liver by
some other disease process.
The incidence of halothane hepatitis appears to be low,
approximately one in 10 000 administrations for adults, and far
less for children. It often occurs after repeated administrations
of halothane over a short period. The unpredictable occurrence

of this syndrome may be the principal reason that halothane
use in adults has declined. More recent thinking indicates that
the inherent risks of the surgery involved, along with such factors
as major blood loss, major volume shifts, intraabdominal and
intrathoracic operations, and periods in which prolonged
hypotension may occur, may contribute to hepatic damage.
Furthermore, if hepatitis is caused by a halogenated agent, that
agent does not necessarily have to be halothane (enflurane and
isoflurane may also be involved). It is postulated that the
oxidative, and particularly the reductive, metabolites of these
inhalational agents are responsible for the hepatitis. A
chemically reactive or immunogenic product may result. This
excess of toxic product or metabolite may be capable of inducing
an immune response, which may be the main factor that leads
to hepatitis.6,10,1922
Pharmacokinetics
Some 6080% of halothane is exhaled in the rst 24 h after it
is administered. Smaller amounts continue to be exhaled for
several days to weeks. Of the portion not exhaled, ~50% undergoes biotransformation. The remainder is eliminated unchanged
via other routes. The cytochrome P-450 system of the endoplasmic reticulum of hepatocytes is responsible for the
biotransformation. Little fluorine is removed, but chlorine, and
to a lesser extent bromine, is removed. Analysis of the urine
shows the fluorine-containing compounds in the form of
trifluoroacetic acid.10
Approximately 80% of enflurane can be recovered unchanged
in expired gas. Of the remaining enflurane, 210% is metabolized by the liver.23 A number of factors make enflurane, an
ether, different from halothane. The ether bond increases molecular stability. The carboflurane bond is a higher-energy bond
than that between carbon and bromine or carbon and chlorine.
With the absence of bromine, and the presence of chlorine and
fluorine, the incorporation of the ether bond results in less
biotransformation of enflurane. Furthermore, because it is less
soluble than halothane in fatty tissue, enflurane leaves the fatty
tissue more rapidly in the postoperative period. This allows less
time for degradation of enflurane.10
Desflurane undergoes the least biotransformation, followed
by isoflurane, with 0.2% being metabolized.24 There is far less
liver metabolism than that for halothane, and less liver metabolism than for enflurane and sevoflurane. The magnitude of
sevoflurane metabolism resembles that of enflurane. With less
biotransformation by liver metabolism, smaller quantities of
fluorine and trifluoroacetic acid are generated. This accounts for
hepatic and renal toxicity being lowest with desflurane and
isoflurane when compared with enflurane, possibly sevoflurane,
or halothane.6,10
Tables 18.7 and 18.8 summarize the advantages and disadvantages of the pharmacodynamic and pharmacokinetic
properties of the three halogenated agents.6,10
Nitrous Oxide
Nitrous oxide is a colorless and odorless gas with very low
solubility in blood. Nitrous oxide alone can predictably cause
surgical anesthesia only when given under hyperbaric conditions. The MAC value is 105%, but variability among patients
is considerable. Analgesia can be induced with 20% nitrous
oxide; some patients lose consciousness when breathing 30%
nitrous oxide, and the majority do so with 80%. Using nitrous
oxide as a single agent at 80% concentration has risk of hypoxia.
Patients also often recall intraoperative events when nitrous
oxide is used alone. Even with nitrous oxide plus a narcotic,
intraoperative recall is not uncommon. If a combination of
narcotic, nitrous oxide, and muscle relaxant is used, the patient
is immobilized and unable to communicate, but unconsciousness cannot be ensured. Because this can be unsettling to the
patient, frequently the clinician adds a potent inhalational
agent or intravenous drug such as a hypnotic or anxiolytic. The
main advantage of nitrous oxide is to reduce the needed
concentration of inhalational anesthetic. Smaller doses of
halogenated agents combined with nitrous oxide produce less
circulatory and respiratory depression and more rapid recovery.
The uptake of nitrous oxide is rapid, which has two benecial
effects during the administration of anesthesiac: the concentration effect and the second-gas effect.
When a very high concentration of an anesthetic is inhaled,
the partial pressure of the anesthetic in arterial blood increases
faster than if a smaller concentration of the anesthetic were
administered. As the anesthetic is rapidly taken up by the blood,
the gas administered by the anesthesia machine is rapidly drawn
into the alveoli, which continue to lose gas rapidly to the
passing blood. This is the advantage of using a high percentage
of nitrous oxide in the initial stage of anesthesia, and it makes
use of the concentration effect. The second-gas effect occurs
when a potent inhalational agent is combined with nitrous
oxide. As nitrous oxide is rapidly taken up by the blood from the
alveoli, and nitrous oxide in the alveoli is rapidly being replaced
by the anesthesia machine, the rate of delivery of halogenated
agent to the alveoli increases. Thus, the rise in arterial tension
of halogenated agents is more rapid.
To summarize, the concentration effect results from the
capacity of a rapidly absorbed gas to facilitate its own uptake.
In the second-gas effect, a rapidly absorbed gas increases the
rate of uptake of the second anesthetic gas.19,25 During emergence from anesthesia, the process is reversed. The possibility
of diffusional hypoxia is a concern because it can cause postoperative hypoxemia, particularly if this is accompanied by
respiratory depression. As nitrous oxide rapidly comes out of
blood into the alveoli, oxygen concentration can be diluted. If
room air is used, nitrous oxide lling the alveoli from the blood
can bring the 21% oxygen concentration of room air down to
much lower levels, and hypoxia can result. This is why 100%
oxygen is administered during the emergence phase.10,19
In general, nitrous oxide has a sympathomimetic effect when
added to halogenated agents.10,26,27 The combined use of nitrous
oxide and halogenated anesthetic results in decreased amounts
of halogenated agents required and less hypotension.10 With
nitrous oxide combined with enflurane, activation of the sympathetic nervous system is less marked than when nitrous oxide
is combined with halothane.27 When nitrous oxide is used alone
with narcotics, it does not displace sympathomimetic activity
but rather causes further cardiovascular depression. Nitrous
oxide has little effect on respiration when used alone, but it
further depresses respiration when combined with other
inhalational agents.10 Nitrous oxide has little effect on the CNS,
but response to hypoxia is diminished. Little, if any, skeletal
muscle relaxation occurs when nitrous oxide is used alone.10
There is no evidence that nitrous oxide triggers malignant
hyperthermia. The gastrointestinal, renal, and hepatic systems
show no effect from administration of nitrous oxide.10
Methionine synthetase, a vitamin B12-dependent enzyme, is
inactivated following prolonged administration of nitrous oxide,
which results in interference with DNA synthesis. This can
cause diminished bone marrow production of red and white
blood cells. Also, oxidation of the cobalt atom in vitamin B12 by
nitrous oxide can result in megaloblastic changes in the bone
marrow, with neuropathy. These changes do not normally occur
during clinical anesthesia for surgery.10
Nitrous oxide is excreted by the lungs, and there is little, if
any, biotransformation. Table 18.9 summarizes the advantages
and disadvantages of nitrous oxide.10
CHAPTER 18
Anesthetics
199
TABLE 18.7. Pharmacodynamics of Inhalational General Anesthetics6,10

Organ System Effects
Halothane
Isoflurane
Desflurane
Sevoflurane
Peripheral vasodilation
++
Blood pressure
Inotropy
Heart rate
++
=+
Cardiac output
Propensity for dysrhythmias
++
Catecholamines
Sympathoadrenal activity
Bronchodilation
Response to hypoxia
End tidal CO2
Shunt (Q/S)
Hypoxic pulmonary
vasoconstriction
Airway irritation
++
++
Seizure activity
Cerebral blood flow
+++
Cerebrospinal fluid pressure
++
Intracranial pressure
++
Cerebral metabolic rate
Cardiovascular
SECTION 4
Pulmonary
CNS
Muscle
Relaxation
Synergism with relaxants
Malignant hyperthermia
trigger
Renal blood flow
Glomerular ltration rate
Fluoride ion
+ (minimal)
+ (minimal)
+ (minimal)
Splanchnic blood flow
Hepatic cell function
Trifluoroacetic acid
++
Renal
Hepatic/Gastrointestinal
=, no change; +, increase; decrease.
TABLE 18.8. Pharmacokinetics of Inhalational General Anesthetics

Halothane
Metabolism
200
20% Liver
Isoflurane
Desflurane
Sevoflurane
0.2% Liver
<0.02%
5%
Ion concentration
CL > Br > F
F (minimal)
F (minimal)
Elimination
6080% via lung

in rst 24 h
99%+ via lung
99%+ via lung
95%
Anesthetics
Agent
Advantages
Disadvantages
Nitrous oxide
Nonirritating, colorless, odorless

Very rapid onset and recovery
Little or no toxicity with ordinary use
Excellent supplement with halogenated or opioid agents
(smaller doses of all agents and fewer complications)
No muscle relaxant activity

If used alone to achieve adequate anesthesia, can result in
hypoxia
Transient postanesthetic hypoxia may occur as large volume
is exhaled
Air pockets in closed spaces may expand in skull, chest,
abdomen
Halothane
Causes laryngospasm but is least irritating to airway

Bronchospasm uncommon
Controlled hypotension decreases blood loss
For proper analgesia, nitrous oxide or opioids usually must be

added
Relaxant drugs added for enhanced muscle relaxation
Visceral reflexes blunted with atropine
Transient dysrhythmias
Incidence of hepatic necrosis
Isoflurane
More rapid adjustment of anesthesia depth compared

with halothane
Cardiac output well maintained
Dysrhythmias less likely when used with epinephrine
compared with halothane
Potentiates muscle relaxants (lower concentration sufces)
More pungent odor than halothane

Increasing depression of cardiopulmonary function with
increasing depths of anesthesia
Desflurane
More rapid induction and emergence than isoflurane,

enflurane, or halothane
Minimal liver metabolism
No change in serum fluoride concentration
No coronary steal
Otherwise similar to isoflurane
Coughing and excitement

Otherwise similar to isoflurane
Sevoflurane
Less of an airway irritant; good for mask induction

More rapid induction and emergence than isoflurane,
enflurane, or halothane
No coronary steal
Serum fluoride concentration is similar to that of enflurane
INTRAVENOUS AGENTS
Summary for Intravenous Agents
Intravenous agents commonly used for induction of anesthesia

and on occasion may be used for maintenance of anesthesia
(total intravenous anesthesia)
Commonly used agents are thiopental, propofol, etomidate,
and ketamine. Thiopental and Propofol are the most commonly
used agents for their ease of titration and favorable
pharmacokinetic prole
Etomidate is favored for use in situations of hemodynamic
instability for its relatively low impact on hemodynamics. Its
prolonged use has been implicated in adrenocortical
suppression
Ketamine has the advantage of maintaining spontaneous
respirations and is associated with hypertension
Benzodiazepines are primarily amnestics and anxiolytics and
opioids are used primarily for analgesia in a balanced
anesthetic technique
Hypnotics
Barbiturates are not analgesics and may even increase sensitivity to pain.2,10 Their main uses are induction of anesthesia
and induction of amnesia. The respiratory and cardiovascular
systems are depressed, and excessive doses may cause marked
hypotension and apnea. When barbiturates are used alone
without analgesia, it is not unusual to see tachycardia and other
sympathetic responses, including dilated pupils, tears, sweating,
tachypnea, and even movement or vocalization in response to
surgical stimulation (Table 18.10).10
When a barbiturate is administered for induction of general
anesthesia, coughing, laryngospasm, and bronchospasm can occur
CHAPTER 18
TABLE 18.9. Advantages and Disadvantages of Inhalational Agents6,10
upon mask ventilation or early attempts at laryngoscopy without

muscle paralysis. Saliva, insertion of an airway, obstruction by
soft tissues, and airway manipulation may trigger these responses.10
Thiopental is the most common induction agent used, followed by methohexital sodium. Both cause a decrease in arterial
blood pressure and reduction of cardiac output. The clinician
must be careful when administering these agents in the
presence of hypovolemia, sepsis, or any kind of cardiovascular
instability, because a normal induction dose may result in
cardiac arrest.10
Extravascular injection may result in severe pain and tissue
necrosis. With intraarterial injection, the endothelium and deeper
layers of the arterial blood vessels can be immediately damaged
and endarteritis can follow. Associated thrombosis and arterial
spasm is common, which can result in vascular ischemia and
gangrene.10 Propofol is chemically unrelated to the barbiturates.
It is a propylphenol. The principal indication is amnesia and
unconsciousness, and the emergence from anesthesia is more
rapid with propofol than with thiopental. Emergence is
characterized by minimal postoperative confusion.10
Propofol can cause a 30% decrease in systemic arterial pressure
predominantly due to peripheral vasodilation. This can be of some
concern in the elderly, and one must be careful when administering propofol in conjunction with opioids.10 There is some
pain at the site of injection, but phlebitis or thrombosis is rare.10
Benzodiazepines
Benzodiazepines can be used for induction, but these mainly
function as anxiolytics and amnestics. Larger doses of benzodiazepines can induce hypnosis and unconsciousness. Use of a
benzodiazepine as a sole agent is helpful when no analgesia is
required. The principal advantage of benzodiazepines is the
201
SECTION 4
TABLE 18.10. Intravenous General Anesthetics4,6,7,10

Agent
Induction Dose
Half-Life
Organ Systems Effects
Metabolism/Elimination
Thiopental
14 mg/kg
t1/2 a: 3 min
tE: 510 h
CNS
CBF
ICP
BP
HR
RR
Bronchospasm
Liver/kidneys
Methohexital
12 mg/kg
tE: 12 h
CNS
RR
CV
Liver/kidneys
Midazolam
0.250.35 mg/kg
tE: 14 h
CNS
Amnesia
Seizure threshold
Liver/kidneys
Diazepam
0.10.5 mg/kg
tE: 710 h; (active

metabolite: 28 days)
CNS
Amnesia
Seizure threshold
Liver/kidneys
Morphine
13 mg/kg
tE: 24 h
Analgesia
CNS
Euphoria
Respiratory depression
Liver/kidneys
Fentanyl
50100 mg/kg
T a: 12 min
tE: 4 h
Similar to morphine, but chest wall

rigidity more common with fentanyl
Liver/kidneys
Ketamine
IV loading dose
(LD): 13 mg/kg
t1/2 a: 1018 min

tE: 2.5 h
Maintenance dose:
1/31/2 LD
Poor visceral analgesia; good somatic

analgesia
Airway reflexes
HTN
IOP
CBF
Cerebral metabolic rate
Liver/kidneys
Propofol
IV induction
2.02.5 mg/kg
IV maintenance
100200 mg kg1 min1
t1/2: 510 min

tE: 13 days
CNS
RR
BP
Liver
BP, blood pressure; CBF, cerebral blood flow; CNS, central nervous system; HR, heart rate; HTN, hypertension; ICP, intracranial pressure; IOP, intraocular pressure; RR,
respiratory rate.
minimal depression of the cardiovascular system. Very large

doses, however, can cause a 20% decline in systemic arterial
blood pressure and vascular resistance. The stability of the
cardiovascular system with smaller doses has made these drugs
particularly attractive for use in monitored anesthetic care and
general anesthesia. One must be prepared for apnea, and
ventilatory support should be readily available. Benzodiazepines
generally have little effect on renal, hepatic, and gastrointestinal
systems. These do not produce neuromuscular paralysis, but
can be used to induce relaxation of spastic muscles.
CNS depression can be antagonized by physostigmine.
Physostigmine inhibits acetylcholinesterase. It crosses the blood
brain barrier more easily than other acetylcholinesterase agents.
It is wise to consider administering atropine or glycopyrrolate
with physostigmine to prevent excessive salivation, abdominal
cramps, nausea and vomiting, and bradydysrhythmia.10
Opioids
202
Opioids are principally used for analgesia. In larger doses, opioids

can induce unconsciousness, but the common technique of
combining nitrous oxide and narcotic alone can result in
insufcient amnesia in some patients. Some patients become
hypertensive during surgical stimulation and may recall
intraoperative events. Table 18.9 reviews narcotic agents.7,10,28,29
TOTAL INTRAVENOUS ANESTHESIA

With the advent of newer agents such as propofol, remifentanil,
and alfentanil, all of which have the desirable properties of
quick onset and short duration of action, traditional methods of
maintenance of anesthesia with inhalational agents has given
way to total intravenous anesthesia.
While the present cost of these drugs can be inhibitory, the
future is promising. Shorter stay in recovery and quicker
mobilization offset the initial costs. Also vaporizers and special
equipment for the delivery of these agents are not needed which
makes delivery of anesthesia outside of the OR more feasible.
LOCAL ANESTHETICS
Local anesthetics are a class of similar compounds that reversibly
block conduction in peripheral and central nervous tissue when
applied in appropriate concentrations. Local anesthetics cause
both sensory and motor paralysis in the innervated area by
blocking the generation and propagation of electrical impulses.
Nitrous oxide is usually used in conjunction with another
halogenated agent or in combination with narcotics for the
maintenance of anesthesia.
The era of local anesthesia commenced in 1864, when Koller
described the local anesthetic effect of cocaine and introduced it
Anesthetics
Increasing the size of the alkyl substitution produces compounds that are more hydrophobic, thus increasing the duration
and potency of the agent.
MECHANISM OF ACTION
Local anesthetics block the generation and conduction of nerve
impulses. All excitable cells have ionic disequilibria across
semipermeable membranes, providing the potential energy for
impulse conduction. The Na+, K+-ATPase, the membranebound enzyme, maintains the ionic disequilibrium in nerve
cells, pumping out three sodium (Na+) ions for every two of
potassium (K+) that are absorbed. During an action potential,
Na+ channels open briefly, allowing a small quantity of Na+ to
flow into the cell, causing depolarization. Local anesthetics block
impulses by inhibiting individual Na+ channels, thereby reducing the aggregate Na+ current, which may be modied by inhibition of the recently discovered K+ channels.3032 The interplay
FIGURE 18.2. Chemical structure of lidocaine

and procaine.
CHAPTER 18
for use in ophthalmology. Because cocaine, the alkaloid isolated

in 1860 from the leaves of an Andean mountain shrub,
Erythroxylon coca, has serious CNS toxicity and causes sloughing
of the corneal epithelium, its use in ophthalmology is limited.
This prompted the German chemical industry to seek less toxic
synthetic substitutes and resulted in the discovery in 1905 of
procaine, which became the prototype for current local anesthetics. The most widely used agents in ophthalmology today
are lidocaine, ropivacaine, and mepivacaine. Mepivacaine 2%
results in a good motor blockade, and can be used alone,
avoiding the toxic potential of the traditional mixture of lidocaine
and bupivicaine.
All clinically useful agents are either aminoesters or
aminoamides (Fig. 18.2). The amide or ester link contributes to
the anesthetic potency. The typical local anesthetic molecule,
exemplied by procaine and lidocaine, consists of a lipophilic
(hydrophobic) aromatic ring group joined to a more hydrophilic
base, the tertiary amine, by an intermediate band (Fig. 18.3).
FIGURE 18.3. Commonly used local

anesthetics in ophthalmology.
203
SECTION 4

between these competing channels determines the relative
potency of the various local anesthetics, whose pharmacologic
effects also depend on the temperature and pH of the medium.
Biochemical analysis of Na+ channels shows the presence of
one major glycoprotein with a molecular mass of ~200 000 Da,
with differing numbers of subunits of 40 000 Da, depending on
the tissue of origin. The Na+ channel is oriented with its
glycosylated groups of the glycoprotein on the outside surface of
the cell membrane.
Similarly, voltage-gated K+ channels make up a large molecular family of membrane proteins involved in the generation
of nerve impulses. Like the proteins gating the Na+ channels,
these proteins span the cell membranes, forming K+-selective
pores that are rapidly switched open or closed, depending on the
membrane voltage. Recent cloning of the rst K+ channel has
resulted in recombinant DNA manipulation of the K+-channel
genes, leading to a molecular understanding of K+-channel
behavior, especially toward elucidation of functional domains
responsible for channel gating and ionic selectivity. Local anesthetics act by several different mechanisms on ionic channels.
They may decrease the fraction of active channels by interfering
directly with activation; they may inhibit or alter the conformational steps whereby channels change from an open form; or
they may reduce the ionic currents flowing through open
channels. In spite of various methods of detecting currents
through single-ion channels, the lack of general approaches for
crystallizing membrane proteins has prevented a direct view of
the structural complexities of their mechanisms.
Recent work by Franks and Lieb33 suggests a more precise
theory of both local and general anesthetic action. Challenging
the well-entrenched lipid hypothesis, these authors suggest
that anesthetics operate not indiscriminately on membrane
lipids but precisely on certain sensitive membrane proteins
regulating ionic channels that govern the responses of nerve
cells. If the nerve cells anesthetic-sensitive proteins are isolated,
designer anesthetics could be synthesized to lock onto the sites
specically in order to enhance an anesthetics sensitivity and
minimize its toxicity.
The chronology of local anesthetic action can be summarized
as follows:34
1. When local anesthetic molecules are deposited near the nerve,
partial removal of the molecules occurs by circulation, tissue
binding, and local hydrolysis of aminoester anesthetics. The
remaining molecules penetrate the nerve sheath.
2. After equilibrium is achieved inside the nerve axons
membranes, depending on the lipophilia of base and cation
species, Na+ channels are prevented from opening by
inhibition of conformational changes that occur with
channel activation.
3. The rates and onset of recovery from block are governed by
the slow diffusion of local anesthetic molecules in and out
of the nerve, not by the much faster binding and
dissociation from ionic channels.
CLINICAL PHARMACOLOGY
204
Successful ophthalmic anesthesia depends on knowledge of

the pharmacologic properties of commonly used local agents.
Aminoesters such as procaine are hydrolyzed in the plasma by
cholinesterase enzymes. The aminoamides, lidocaine and ropivacaine, are extremely stable and undergo biotransformation
and enzymatic degradation in the liver. Allergic reactions to
aminoamides are extremely rare compared with reactions to
aminoesters.
For a local anesthetic to be successfully and safely used in
ophthalmic anesthesia, it must have potency, rapid onset of
action, long duration of sensory and motor block, and minimal
systemic toxicity. The individual prole of an agent is

determined mainly by its physicochemical characteristics.
In addition to the physicochemical properties, latency also
depends on the concentration. Lidocaine has a more rapid onset
of action than ropivacaine, and 0.75% ropivacaine causes a more
rapid anesthetic effect than 0.25% ropivacaine. Procaine has a
short duration of action, lidocaine an intermediate duration,
and ropivacaine the longest duration. Mixtures of local anesthetics, such as lidocaine and bupivacaine, have been popular
for ophthalmic anesthesia, because they combine the advantages
of rapid onset but short duration of action of lidocaine, and
slow onset but long duration of action of bupivacaine. For
example, a 2% solution of lidocaine mixed with equal parts of a
0.75% solution of bupivacaine produces anesthesia within 5 min
that lasts 34 h. At a concentration of 1:200 000, vasoconstrictors such as epinephrine, mixed into the local anesthetic,
decrease the rate of vascular absorption and subsequent biotransformation. This allows more anesthetic agent to reach the
membrane receptors and prolongs the depth and duration of
anesthesia. With a judicious combination of lidocaine and bupivacaine and a dilute vasoconstrictor such as epinephrine, the
duration of sensory and motor blockade is considerably
enhanced; this permits the ophthalmologist to perform complicated intraocular procedures and minimize postoperative pain
and discomfort.
TOXICITY
The effectiveness and safety of local anesthetics depend on
proper dosage, correct administration, and preparedness for
emergencies. Systemic side effects, such as neurologic and
cardiac crises, are avoided by using the smallest effective anesthetic dose for a given procedure, thereby avoiding high plasma
levels and their associated effects. Unintentional intravascular
injection of local anesthetics can cause convulsions and respiratory depression, and possibly arrest. Cardiovascular stimulation or depression and cardiac arrest also may occur. Thus,
clinicians must be well versed in basic life support techniques
in order to manage toxic reactions due to local anesthetics. Ready
availability of oxygen and of cardiopulmonary resuscitative
drugs administered by a skilled anesthesiologist promotes rapid
and successful recovery.
Anesthetic solutions that contain epinephrine should be used
with extreme caution in patients with cardiovascular disease
such as hypertension, arteriosclerotic or cerebrovascular disease,
diabetes, heart block, or thyrotoxicosis. Patients taking medication
for systemic hypertension may also be more susceptible to
alterations in blood pressure.
DRUG INTERACTIONS
Cardiovascular arrhythmia may occur when local anesthetic
agents with epinephrine are used during general anesthesia with
halothane. Patients receiving monoamine oxidase inhibitors or
tricyclic antidepressants may experience severe and prolonged
hypertension with local anesthetics containing epinephrine,
thus vasoconstrictors are best avoided. CNS toxicity may occur
when local anesthetics are used in conjunction with narcotic
analgesics and phenothiazine-type compounds. In patients taking
echothiophate for control of glaucoma, inhibition of plasma
cholinesterases may result in increased plasma levels of local anesthetics and possibly cardiovascular and neurologic complications.
NERVE BLOCKS FOR OPHTHALMIC SURGERY

Retrobulbar block involves insertion of short-beveled needle
into the junction of the lateral and middle thirds of the inferior
Anesthetics
orbital rim behind the globe. Complications include vasovagal
reactions from fear and anxiety, ocular-cardiac reflex, retrobulbar hematoma (most common), and direct trauma to the globe
or optic nerve. Direct local anesthetic toxicity from intraarterial
injection via the ophthalmic artery produces seizures.
Epidural/intrathecal injection via the optic sheath, produces a
wide range of CNS side effects ranging from shivering,
dysphagia, tachycardia, HTN, dilation of the contralateral
pupil, loss of consciousness, and respiratory/cardiac arrest.35
Peribulbar block technique is performed by injection of
local anesthetic above and below the orbit. Complications are
reduced as compared to the retrobulbar block, however, onset
time is slower, the incidence of incomplete anesthesia and
akinesia is greater, and globe perforation can still occur.
Summary for Local Anesthetics
Local anesthetics cause sensory and motor paralysis in the

innervated area by blocking the generation and propagation of
electrical impulses
For a local anesthetic to be successful and safe when used in
ophthalmic anesthesia, it must have potency, rapid onset of
action, long duration of sensory and motor block, and minimal
systemic toxicity
The most widely used agents in ophthalmology are lidocaine,
ropivacaine, and mepivacaine
1. Egbert LD, Battit GE, Turndorf H, et al: The

value of the preoperative visit by an
anesthetist. JAMA 1963; 185:553.
2. Dripps RD, Eckenhoff JE, Vandam LD:
Premedication, transport to the operating
room, and preparation for anesthesia. In:
Dripps RD, Eckenhoff JE, Vandam LD, eds.
Introduction to anesthesia: the principles of
safe practice. 6th edn. Philadelphia, PA:
WB Saunders; 1982:3444.
3. Firestone LL: General preanesthesic
evaluation. In: Firestone LL, et al, eds.
Clinical anesthesia procedures of the
Massachusetts General Hospital. 3rd edn.
Boston, MA: Little, Brown; 1988:314.
4. Kennedy SK, Longnecker DE: History and
principles of anesthesiology. In: Gilman AG,
et al, eds. The pharmacological basis of
therapeutics. 8th edn. New York:
Pergamon; 1990:269284.
5. de Jong RH, Hearmer JE: Diazepam- and
lidocaine-induced cardiovascular changes.
Anesthesiology 1973; 39:633.
6. Omoigui S: The anesthesia drugs
handbook. 2nd edn. St Louis, MO: CV
Mosby; 1995:256, 296, 359391.
7. Kofke WA, Firestone LL: Commonly used
drugs. In: Firestone LL, et al, eds. Clinical
anesthesia procedures of the
8. Nunn JF, Utting JE, Brown BR Jr:
Introduction. In: Nunn JF, Utting JE, Brown
BR Jr, eds. General anesthesia. 5th edn.
London: Butterworths; 1989:16.
9. Calverley RK: Anesthesia as a specialty:
past, present, and future. In: Barash PG,
Cullen BF, Stoelting RK, eds. Clinical
anesthesia. Philadelphia, PA: JB Lippincott;
1989:334.
10. Marshall BE, Longnecker DE: General
anesthetics. In: Gilman AG, et al, eds. The
pharmacological basis of therapeutics. 8th
edn. New York: Pergamon; 1990:285310.
11. Hickel RS: Administration of general
anesthesia. In: Firestone LL, et al, eds.
Clinical anesthesia procedures of the
12. Eger EI, Saidman LJ, Brandstater B:

Minimum alveolar anesthetic concentration,
a standard of anesthetic potency.
13. Eger EI: Anesthetic uptake and action.
Baltimore, MD: Williams & Wilkins; 1974.
14. Richter JJ: Mechanisms of general
anesthesia. In: Barash PG, Cullen BF,
Stoelting RK, eds. Clinical anesthesia.
Philadelphia, PA: JB Lippincott;
1989:281292.
15. Koblin DD: Mechanisms of action. In: Miller
RD, ed. Anesthesia. New York: Churchill
16. Halsey MJ: Molecular mechanisms of
anaesthesia. In: Nunn JF, Utting JE, Brown
BR Jr, eds. General anesthesia. 5th edn.
London: Butterworths; 1989:1929.
Fundamentals of inhalational anesthesia. In:
Dripps RD, Eckenhoff JE, Vandam LD, eds.
18. Bufngton CW, Davis KB, Gillispie S,
Pettinger M: The prevalence of steal-prone
coronary anatomy in patients with coronary
artery disease: an analysis of the Coronary
Artery Surgery Study Registry.
Inhalational anesthetics. In: Dripps RD,
Eckenhoff JE, Vandam LD, eds.
20. Stock JG, Strunin L: Unexplained hepatitis
following halothane. Anesthesiology 1985;
63:424.
21. Boden JM, Rice SA: Metabolism and toxicity.
In: Miller RD, ed. Anesthesia. 3rd edn.
New York: Churchill Livingstone; 1990.
22. Berman LM, Holaday DA: Inhalation
anesthetic metabolism and toxicity. In:
Barash PG, Cullen BF, Stoelting RK, eds.
Clinical anesthesia. Philadelphia, PA: JB
Lippincott; 1989.
23. Carpenter RL, Eger EI II, Johnson BH, et al:
The extent of metabolism of inhaled
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
anesthetics in humans. Anesthesiology

1986; 65:201.
Holaday DA, Fiserova-Bergerova V, Latto IP,
Zumbiel MA: Resistance of isoflurane to
biotransformation in man. Anesthesiology
1975; 43:325.
Epstein RM, Rackow H, Salanitre E, Wolf
G: Influence of the concentration effect on
the uptake of anesthetic mixtures: the
second gas effect. Anesthesiology 1964;
25:364.
Hornbein TF, Martin WE, Bonica JJ, et al:
Nitrous oxide effects on the circulatory and
ventilatory responses to halothane.
Smith NT, Caverly RK, Prys-Roberts C,
et al: Impact of nitrous oxide on the
circulation during enflurane anesthesia in
man. Anesthesiology 1978; 48:345.
Philbin DM, Rosow CE, Schneider RC,
et al: Fentanyl and sufentanil anesthesia
revisited: how much is enough?
Hug CC Jr: Does opioid anesthesia exist?
Strichartz GR, Ritchie JM: The action of
local anesthetics on ion channels of
excitable tissues. In: Strichartz GR, ed.
Handbook of experimental pharmacology.
Berlin: Springer; 1987:2152.
Miller C: 1990: Annus mirabilis of potassium
channels. Science 1991; 252:1092.
Butterworth JF, Strichartz GR: Molecular
mechanisms of local anesthesia: a review.
Franks NP, Lieb WR: Stereospecic effects
of inhalational general anesthetic optic
isomers on nerve ion channels. Science
1991; 254:427.
Strichartz GR, Covino BG: Local
anesthetics. In: Miller RD, ed. Anesthesia.
3rd edn. New York: Churchill Livingstone;
1990.
Ripart J, Lefrant JY, de La Coussaye JE,
et al: Peribulbar versus retrobulbar
anesthesia for ophthalmic surgery.
CHAPTER 18
REFERENCES
205
CHAPTER
19
Antibacterials
Harold G. Jensen, Henry D. Perry, and Eric D. Donnenfeld
FLUOROQUINOLONES
OVERVIEW AND MECHANISM OF ACTION
The fluoroquinolones are the newest class of antibacterials
available in the ght against microbes (Table 19.1). Fluoroquinolones are bactericidal agents that act by inhibiting DNA
replication. They have dual targets, topoisomerase II (DNA
gyrase) and topoisomerase IV, which are related but distinct
enzymes involved in DNA synthesis.1,2 By inhibiting bacterial
DNA gyrase and topoisomerase IV, DNA replication and transcription are halted. Because DNA gyrase exists only in plant
and bacterial cells, fluoroquinolones have low toxicity in humans
relative to other antibacterial agents; the drugs will not affect
normal cell replication. The available topical agents include ciprofloxacin, ofloxacin, levofloxacin, gatifloxacin, and moxifloxacin.
Ciprofloxacin and ofloxacin have been used in the treatment
of ocular infections for over 10 years. They are active against
most Gram-negative bacteria and some Gram-positive bacteria.
Levofloxacin is the L-isomer of ofloxacin and has demonstrated
increased activity against Gram-positive bacteria, but less potent
activity against Pseudomonas aeruginosa and certain Enterobacteriaceae. Gatifloxacin and moxifloxacin are the most recently
approved quinolones for ophthalmic infections. Earlier generations
of fluoroquinolones targeted only DNA gyrase, whereas the
newer fourth-generation agents, gatifloxacin and moxifloxacin,
target both DNA gyrase and topoisomerase IV.3
SPECTRUM OF ACTIVITY
Although all the quinolone agents have broad-spectrum activity,
gatifloxacin and moxifloxacin have demonstrated enhanced activity against Gram-positive organisms, especially Streptococcus
pneumoniae. Increased activity toward some of the atypical and
anaerobic organisms has also been shown with the newer fluoroquinolones. All the quinolones are generally active against enteric
Gram-negative rods such as P. aeruginosa, Haemophilus influenzae, and Neisseria gonorrhoeae. Gatifloxacin and moxifloxacin
have increased activity against most Staphylococcus aureus and
Staphylococcus epidermidis strains, but ciprofloxacin may have
slightly better activity against Pseudomonas aeruginosa. The betahemolytic streptococcal and enterococcal sensitivities vary among
the older quinolones, but increased sensitivities are observed
with the newer agents.4 Bacterial resistance does not commonly
develop during treatment with quinolones for ocular infections;
however, it is still a possibility. Because quinolones target two
enzymes, bacterial resistance develops much slower with gatifloxacin and moxifloxacin than with the older quinolones. Many
ciprofloxacin- and ofloxacin-resistant organisms are susceptible
to the two newer quinolones. Gatifloxacin and moxifloxacin are

more active than ciprofloxacin and ofloxacin against the atypical
Mycobacteria, including Mycobacterium avium-intracellulare,
Mycobacterium marinum, and Mycobacterium chelonei. Ofloxacin and gatifloxacin have also shown activity against Chlamydia
trachomatis. In general, gatifloxacin and moxifloxacin have a
very comparable spectrum of activity against Gram-positive and
Gram-negative organisms. Gatifloxacin has minimally better
activity against Gram-negative bacteria, whereas moxifloxacin has
a minimally better spectrum of activity against Gram-positive
infections.
PHARMACOLOGY
The quinolones are well absorbed after oral or intravenous
administration and have variable pathways of metabolism and
excretion. The oral quinolones achieve systemic levels comparable to those of intravenous antibiotics because of their high
absorption and intrinsic solubility. After oral administration,
concentrations in serum peak after 12 h. The half-lives of
fluoroquinolones range from 3.5 h in ciprofloxacin to 20 h in
sparfloxacin, which allows for once- or twice-daily dosing. The
quinolones easily penetrate into phagocytes, thereby producing
concentrations within neutrophils and macrophages up to fourteen times their concentration in the serum.5 This accounts for
their excellent in vivo activity against such intracellular pathogens as Listeria spp., Salmonella spp., and Mycobacterium spp.
Ofloxacin exhibits little or no in vivo metabolism, and is excreted
mainly (90%) via the kidneys. The other quinolones are cleared
by both hepatic and renal routes in various proportions, with
elimination occurring via the kidneys. Small amounts of these
drugs are also excreted in the bile.
OPHTHALMIC INDICATIONS
Ciprofloxacin, ofloxacin, and gatifloxacin are available in a 0.3%
commercial solution. Levofloxacin is available in both a 0.5%
and 1.5% solution, while moxifloxacin is available only as a
0.5% solution for ophthalmic use. Moxifloxacin is the only topical
solution without a preservative. All the ophthalmic quinolones
have labeled indications and have been shown to be effective for
the treatment of bacterial conjunctivitis.610 Ciprofloxacin,
ofloxacin, and levofloxacin have labeled indications for the treatment of corneal ulcers and are particularly active against enteric
Gram-negative bacilli and quinolone-sensitive Pseudomonas spp.
In double-masked control clinical trials, ciprofloxacin and ofloxacin
were shown to be equivalent to fortied tobramycin and cefazolin in the treatment of bacterial keratitis.11,12 Gatifloxacin has
shown excellent activity in both a human corneal ulcer study and
207
SECTION 4
TABLE 19.1 Available Ocular Antibacterials
208
Antibiotic Class
Generic Name
Trade Name(s)
Effective Against
Fluoroquinolone
Ciprofloxacin
Ofloxacin
Levofloxacin
Gatifloxacin
Moxifloxacin
Cipro, Ciproxin
Floxin
Levaquin
Tequin
Avelox
Broad spectrum. All effective against Listeria spp.,

Salmonella spp., and Mycobacterium spp.
Tetracycline
Chlortetracycline
Oxytetracycline
Doxycycline
Minocycline
Aureomycin
Terramycin
Ocular uses include Periostat,
Vibramycin, Doryx
Minomycin, Minocin,
Arestin, Akamin, Aknemin,
Solodyn, and Dynacin
Certain Enterobacteriaceae, Vibrio spp.,

Rickettsia spp., Mycobacterium marinum, and some
protozoans (e.g., Plasmodium spp., Entamoeba
histolytica)
Aminoglycoside
Neomycin
Gentamicin
Tobramycin
Amikacin
Maxitrol
Garamycin
Tobrex
Amikin
Staphyloccus aureus, Enterobacteriacae,

P. aeruginosa, Acinetobacter spp.
Glycopeptide
Vancomycin
Teicoplanin
None patent expired

Targocid
Methicillin-susceptible and resistant staphylococci,

enterococci, Corynebacterium spp., Bacillus spp.,
Listeria monocytogenes, Clostridium spp.
Macrolides
Erythromycin
None patent expired
Gram + cocci, gram + bacilli, Neisseria spp.,

mycoplasmas, treponemes, rickettsiae, and
chlamydiae, Haemophilus influenzae,
Bartonella spp., Bacillus fragilis, Prevotella spp.,
Porphyromonas spp., Propionibacterium acnes,
Clostridium spp., M. Avium-intracellulare complex,
M. scrofulaceum, M. kansasii, M. chelonae
Chloramphenicol
Chloramphenicol
Many; ocular uses include

Chloroptic and Chloromycetin
Chlamydia, mycoplasmas, rickettsia. Neisseria

meningitides, Haemophilus influenzae, most
Enterobacteriaceae. Many anaerobia bacteria are
inhibited at concentrations <10 g/mL
Sulfonamides and
trimethoprim
Sulfonamide and
trimethoprim
Bactrim
Active in vitro, but increasing resistance has limited

their efcacy. The combination of the two drugs
enhances their activity
Bacitracin and
gramicidin
Bacitracin and
gramicidin
None patent expired
Active against staphylococci and group A

beta-hemolytic streptococci. Some spirochetes,
Entamoeba histolytica, Actinomyces, and
Fusobacterium
Polymyxins
Polymyxin
Polysporin, Neosporin
Pseudomonas spp., Serratia spp., Proteus spp.,

Providencia spp.
rabbit corneal ulcer studies against Pseudomonas aeruginosa and

Staphylococcus aureus. Moxifloxacin has also demonstrated
activity against these two organisms in a rabbit ulcer model.1316
associated with crystal deposits in the cornea.17,18 This has not

been seen with the other topical quinolones, presumably due to
their higher solubility.
ADVERSE EVENTS
TETRACYCLINES
Toxicity, fever, rash, and nausea occur in ~4% of patients given

oral quinolone therapy. On occasion, patients develop elevated
levels of liver enzyme. The drugs can crystallize in the urine,
especially in patients who are dehydrated. Interstitial nephritis
has been reported after high doses of ciprofloxacin. Insomnia
and restlessness have occurred in elderly patients taking fluoroquinolones. Studies with animals have shown quinolones to
cause irreversible cartilage erosions and skeletal abnormalities.
Therefore, although such effects have not yet been observed in
humans, quinolone use should be avoided in young children
until further research has been completed. There is no evidence
in humans for ocular toxicity with the new fluoroquinolones,
despite the fact that cataracts occurred in cats after months of
perfloxacin therapy, and macular bulla formation occurred in
patients with renal failure on flumequine (a quinolone used in
Europe). The topical administration of ciprofloxacin has been

Tetracyclines are broad-spectrum antibiotics that inhibit bacterial
protein synthesis by binding to the 30-S ribosomal subunit of the
bacteria. This prevents bacterial polypeptide synthesis. They are
bacteriostatic for most organisms. Various forms of tetracycline
are available, including chlortetracycline (topical), oxytetracycline,
doxycycline, and minocycline (see Table 19.1). Tetracyclines also
inhibit collagenase and polymorphonuclear leukocyte migration.
They also have an antilipase action, fostering the production of
long-chain fatty acids.19
Tetracyclines are active against most Gram-positive organisms, certain Enterobacteriaceae, Vibrio spp. Rickettsia spp. Mycobacterium
Antibacterials
PHARMACOLOGY
Of the available tetracyclines, doxycycline has the best penetration into the eye. Ocular penetration of oxytetracycline and
chlortetracycline is hindered by the corneal epithelium and
therefore improved by the presence of a corneal defect. The
more lipophilic derivatives of tetracycline, such as minocycline,
appear to have better ocular penetration when administered
systemically than do derivatives such as chlortetracycline. These
drugs should be avoided in patients with renal failure, as they
are antianabolic and can speed the decline of renal function in
persons with chronic renal failure. Doxycycline is highly protein
bound, with a long half-life, so that it can be dosed once
daily. Doxycycline and minocycline can also be administered
intravenously.
Topical tetracycline is indicated for the treatment of ocular trachoma and is used prophylactically for gonococcal ophthalmia
neonatorum. Oral tetracyclines are effective against several diseases caused by chlamydia, including conjunctivitis, urethritis,
cervicitis, and pneumonitis. They may also be effective for Lyme
disease and nocardial infections. Minocycline has been used to
treat M. marinum infections. Tetracyclines have been shown to
be active in treating noninfectious corneal ulceration and acne
rosacea.20 Because of increased resistance to many of the common
Gram-positive and Gram-negative ocular pathogens, tetracycline
is not a common rst-line antibiotic for most ocular bacterial
infections.
ADVERSE EVENTS
Tetracyclines have irritative effects on the upper gastrointestinal
tract, producing esophageal ulcerations, nausea, vomiting, and
epigastric distress. Hypersensitivity reactions are unusual, generally manifesting themselves as urticaria, xed drug eruptions,
morbilliform rashes, and anaphylaxis. These drugs may cause
depression of bone growth, permanent discoloration of the teeth,
and enamel hypoplasia when given during tooth and skeletal
development.21 Therefore, these drugs are usually avoided in
children <8 years old and in women during pregnancy.
AMINOGLYCOSIDES
The aminoglycosides used in ophthalmology are generally limited to neomycin, gentamicin, tobramycin, and amikacin (see
Table 19.1). Aminoglycosides inhibit bacterial protein synthesis
by binding irreversibly to the bacterial 30S ribosomal subunit.
The aminoglycoside-bound bacterial ribosomes then become
unavailable for translation of mRNA during protein synthesis,
thereby leading to cell death. The aminoglycosides have a well
characterized postantibiotic effect, which means there is continued suppression of bacterial growth despite the decline of
antimicrobial concentration.
Aminoglycoside antibiotics are active primarily against aerobic
Gram-negative bacilli and Staphylococcus aureus. As a group,
they are particularly potent against the Enterobacteriaceae,
P. aeruginosa, and Acinetobacter spp. Although gentamicin and
tobramycin have similar antibacterial proles, gentamicin is
more active in vitro against Serratia spp., whereas tobramycin is
more active against P. aeruginosa. Amikacin is often used as the
aminoglycoside of choice when resistance with gentamicin and
tobramycin is prevalent. In addition, amikacin is active against
many Mycobacterium spp. These agents are only moderately
active against Haemophilus spp. and Neisseria spp. They are
not active against anaerobes.
PHARMACOLOGY
Gastrointestinal absorption is low with oral aminoglycosides.
After intravenous administration they are freely distributed in
extracellular spaces, but do not penetrate well into the cerebrospinal fluid (CSF), vitreous, and biliary tract, even in the presence
of inflammation. In adults with normal renal function, the
aminoglycosides have half-lives in serum of ~23 h. The aminoglycosides are excreted primarily by glomerular ltration. In
patients with renal failure, aminoglycosides accumulate and dosage
reductions are necessary.
Historically, the aminoglycosides have been a mainstay in the
treatment of ocular infections. However, increasing resistance
has limited their use in recent years. Gentamicin and tobramycin
are available as 0.3% topical solutions and ointments. Neomycin
is available only as a topical ointment and amikacin is not
available in a topical formulation. Either gentamicin or tobramycin is often used as a fortied solution usually in addition to
one of the cephalosporins for the treatment of severe corneal
ulcers, especially those caused by Pseudomonas spp. The aminoglycosides have shown a synergistic effect with the penicillins
and cephalosporins; however, penicillins may inactivate the
aminoglycosides if mixed together for topical application. Each
solution should be administered separately.
CHAPTER 19
marinum, and some protozoans such as Plasmodium spp. and

Entamoeba histolytica. They are not usually effective against
P. aeruginosa, Bacteroides species, or group B streptococci.
Organisms commonly acquire resistance to tetracycline via
plasmids, S. aureus resistance has climbed to ~40% in the
United States since the early 1990s. Due to increased resistance
rates, in vitro susceptibility testing is necessary to conrm the
activity of tetracycline against most organisms.
ADVERSE EVENTS
Nephrotoxicity and auditory or vestibular toxicity are the most
serious adverse events and are characteristic of all the aminoglycosides. Neomycin is too toxic for parenteral administration,
and its use is limited to topical applications. Tobramycin and
amikacin are less ototoxic than gentamicin. Ototoxicity is a
result of selective destruction of the hair cells in the cochlea.
Approximately 2% of patients receiving systemic aminoglycosides develop ototoxicity and half of these cases are irreversible.
Gentamicin and amikacin are more likely to be nephrotoxic than
tobramycin. Nephrotoxicity, which results from a high concentration of aminoglycosides in proximal renal tubules, may present as mild proteinuria to severe azotemia. As many as 26% of
patients receiving prolonged treatment with systemic aminoglycosides develop evidence of mild renal impairment.22 The
likelihood of nephrotoxicity increases when cephalosporins or
other nephrotoxic drugs are coadministered with aminoglycosides.
Risk factors for nephrotoxicity and ototoxicity include long
duration of treatment, high aminoglycoside levels in the serum,
renal insufciency and previous treatment with other ototoxic or
nephrotoxic drugs. Frequent dosing of fortied aminoglycoside
preparations used to treat bacterial keratitis can result in severe
209

corneal epithelial toxicity. Occurrence of pseudomembranous
conjunctivitis is common with fortied topical gentamicin and
occasionally results from treatment with topical fortied
tobramycin.
GLYCOPEPTIDES
Vancomycin and teicoplanin are similar bactericidal antibiotics
which inhibit peptidoglycan synthesis in the bacterial cell wall
by complexing with cell wall precursors (see Table 19.1). Only
vancomycin is available in the United States, but teicoplanin is
available in many countries outside the United States.
SECTION 4
These are narrow spectrum antibiotics that are active primarily
against aerobic and anaerobic Gram-positive organisms, including methicillin-susceptible and -resistant staphylococci, streptococci, enterococci, Corynebacterium spp., Bacillus spp., Listeria
monocytogenes, Clostridium spp., and Actinomyces spp. Teicoplanin is two- to fourfold more active than vancomycin against
most Gram-positive cocci.23 Increasing resistance of vancomycin
has been observed among clinical isolates of Enterococcus faecalis,
E. faecium,24 and coagulase-negative staphylococci.25 Crossresistance with teicoplanin is variable with these strains. Neither
vancomycin nor teicoplanin are active against Gram-negative
organisms or mycobacteria.
PHARMACOLOGY
Although the glycopeptides can be administered orally or parenterally, the drugs are poorly absorbed after oral administration.
Because intramuscular administration is extremely painful, parenteral use is limited to intravenous administration. In patients
with healthy renal function, the glycopeptides are eliminated
from the body by glomerular ltration. The half-life of vancomycin in serum is 6 h and ~45 h for teicoplanin. In patients
with severe renal insufciency, their excretion may be prolonged
to ~9 days.
Although there are no ocular formulations for vancomycin, it is
used for topical, subconjunctival, and intravitreal administration.
Vancomycin is frequently used for the treatment of infectious
corneal ulcers or endophthalmitis when Gram-positive organisms are suspected. Initial treatment of serious ocular infections
with other antiinfectives is often changed to vancomycin when
culture results are proven to be a methicillin-resistant staphylococcus or an Enterococcus spp. Intravenous, subconjunctival
and topical administrations do not result in detectable vitreous
levels. Intravitreal doses of 2 mg or less have been shown to be
nontoxic to the rabbit retina.26 To protect against increased bacterial resistance, use of vancomycin should be limited to sightthreatening infections caused by Gram-positive organisms
resistant to other antibiotics.
ADVERSE EVENTS
210
The most frequent side effects of vancomycin are fever, chills,

and phlebitis at the site of infusion. Rapid infusion causes tingling and flushing of the face, neck, and thorax, known as the red
man syndrome, as a result of histamine release by basophils
and mast cells.27 Vancomycin is also ototoxic and nephrotoxic
when given systemically. Ototoxicity is associated with high

serum levels of the drug and may result in permanent deafness.
The risk of ototoxicity and nephrotoxicity may be increased
when vancomycin is used in combination with aminoglycosides.
Hypersensitivity reactions, including fever, eosinophilia, urticaria,
and anaphylaxis may also occur.28 Subconjunctival injections
may cause conjunctival necrosis and sloughing. Topical administration has also been shown to retard epithelial wound healing
in rabbits.29
MACROLIDES
Until recently, erythromycin was the only macrolide formulated
for ophthalmic infections (see Table 19.1). However, clarithromycin and azithromycin are derivatives that offer signicant
advantages over erythromycin because of expanded antmicrobial spectra, improved pharmacokinetic parameters, and less
frequent and severe side effects. An ophthalmic formulation of
azithromycin 1.0% is currently being investigated for treating
bacterial conjunctivitis. Macrolides are generally bacteriostatic
agents that inhibit bacterial RNA-dependent protein synthesis.
They may be bactericidal in high concentration. The macrolides
bind reversibly to the 23S tRNA of the 50S ribosomal subunits
of susceptible organisms blocking peptide chain elongation.
Macrolide antibiotics have a broad range of activity which includes
Gram-positive cocci, Gram-positive bacilli, Neisseria spp., mycoplasmas, treponemes, rickettsiae, and chlamydiae. Clarithromycin
is more active against sensitive streptococci and staphylococci,
but cross-resistance does occur. Azithromycin is less active against
staphylococci and streptococci, and none of the three are active
against methicillin-resistant staphylococci. All are moderately
active against Neisseria gonorrhoeae; azithromycin is the more
active against Haemophilus influenzae. Both azithromycin and
clarithromycin are more active than erythromycin against Chlamydia spp. and are frequently used for systemic treatment of
the disease. The macrolides are among the most potent agents
against Bartonella spp., and have good activity against anaerobic
bacteria such as the Bacillus fragilis group, Prevotella spp.,
Porphyromonas spp., Propionibacterium acnes, and Clostridium
spp. Several of the atypical mycobacteria including M. aviumintracellulare complex, M. scrofulaceum, M. kansasii, and
M. chelonae have also shown sensitivity to the macrolides.30,31
PHARMACOLOGY
Erythromycin is available in various topical, parenteral (lactobionate and gluceptate), and oral (base stearate, ethylsuccinate,
and estolate) preparations. Erythromycin is rapidly inactivated
when administered orally, whereas the newer macrolides are stable
against acid degradation. Tissue distribution of macrolides is
excellent, with concentrations in various tissues 10- to 100-fold
higher than in serum. Erythromycin and clarithromycin are
metabolized by the liver and excreted primarily in the bile.
Azithromycin is excreted largely unchanged in the bile. Because
both azithromycin and clarithromycin have extended half-lives,
once a day dosing has been shown effective. Topical preparations
of erythromycin do not penetrate the cornea well, but are useful
for the treatment of conjunctivitis and blepharitis caused by
susceptible organisms. A topical ocular formulation of azithromycin is being investigated within a delivery system (DuraSite,
Insite Vision, Alameda, CA) that remains in the eye for up to
Antibacterials
several hours, which allows for sustained ocular penetration
and reduced dosing.
ing CSF, where levels are generally 3050% of the concentrations in serum. Inactivated in the liver, active and inactive drug
are excreted rapidly in the urine.
The topical ocular formulation of erythromycin is used for conjunctivitis and staphylococcal blepharitis. Because of its activity
against N. gonorrhoeae, erythromycin is used in many parts of
the world for prophylaxis of ophthalmia neonatorum. A primary
indication for oral erythromycin is for the treatment of Chlamydia
trachomatis infections in children. It is as effective as the tetracyclines for chlamydial infections and is safer for pregnant women
and children under the age of eight. The topical ocular formulation of azithromycin has been shown to be as safe and effective
as tobramycin based on clinical resolution and bacterial eradication in both pediatrics and adults with bacterial conjunctivitis.
In this study, azithromycin was dosed bid on days 1 and 2 and
qd on days 35, whereas tobramycin was dosed qid for 5 days.
Azithromycin in a single 1-g dose orally, or doxycycline at a dosage
of 100 mg orally twice per day for 7 days is recommended for
urogenital infections caused by chlamydia.32 Once-daily dosing
with azithromycin has also shown promising results in children
with ocular chlamydial infections in randomly selected
Ethiopian villages.33
The use of chloramphenicol in the US has declined over the
past decades because of its potential for inducing severe systemic
adverse reactions, and the availability of newer antibiotics.
Outside the US, however, chloramphenicol drops continue to be
a commonly used and effective antibiotic for bacterial conjunctivitis.35 Because of the risk of fatal idiosyncratic aplastic
anemia after topical administration,36,37 there should be careful
patient follow-up.
ADVERSE EVENTS
Bone marrow toxicity is the major complication of chloramphenicol use. This may occur as either dose-related bone marrow
suppression or idiosyncratic aplastic anemia. Chloramphenicol
occasionally causes hypersensitivity reactions, including skin
rashes, drug fevers, and anaphylaxis. It should not be used with
other drugs known to produce hematologic side effects.
Erythromycin is one of the safest antibiotics used. Side effects

are dose-related with gastrointestinal irritation including abdominal cramps, nausea, vomiting, and diarrhea which occur primarily with oral administration, but may also occur when given
intravenously. Side effects are similar with azithromycin and
clarithromycin; however, less nausea has been reported with the
newer drugs. Ototoxicity and reversible hearing loss may occur
with the use of large doses and high concentrations of the
macrolides.

Sulfonamides competitively inhibit the bacterial modication of
p-aminobenzoic acid into dihydrofolate, and trimethoprim inhibits
bacterial dihydrofolate reductase. The sequential inhibition of
folate metabolism ultimately prevents the synthesis of bacterial
DNA. Because mammalian cells do not synthesize folic acid,
human purine synthesis is not affected by these agents. These
compounds act synergistically to enhance their spectrum of
activity.
CHLORAMPHENICOL
Sulfonamides are active in vitro against a variety of Gram-positive

and Gram-negative bacteria, actinomycetes, and chlamydia.
However, increasing resistance has limited their efcacy against
many of these organisms. Serratia marcescens, Pseudomonas
aeruginosa, enterococci, and anaerobes are usually resistant.
Trimethoprim is also active in vitro against many Gram-positive
cocci and Gram-negative bacilli. However, P. aeruginosa, most
anaerobes, Mycoplasma pneumoniae, Neisseria spp., Moraxella
catarrhalis, and mycobacteria are resistant. The combination of
these two drugs produces a synergistic effect greatly enhancing
the efcacy of either drug alone (see Table 19.1). Combinations of
trimethoprim with other agents such as rifampin, polymyxins,
and aminoglycosides have also demonstrated in vitro synergistic
antibacterial activity against many Gram-negative bacilli.
Chloramphenicol is a unique antibiotic molecule that contains

a nitrobenzene ring and was originally derived from Streptomyces
venezuelae (see Table 19.1). The drug is a bacteriostatic agent that
inhibits protein synthesis by binding reversibly to the peptidyltransferase component of the 50S ribosomal subunit and prevents the transpeptidation process of peptide chain elongation.
Chloramphenicol is active against many Gram-positive and
Gram-negative bacteria, chlamydia, mycoplasmas, and rickettsia;
however, the drug is often inactive against methicillin-resistant
Staphylococci spp. and is variably active against enterococci.
Neisseria meningitides, Haemophilus influenzae and most
Enterobacteriaceae are susceptible. Activity against Serratia and
Enterobacter isolates is variable and Pseudomonas spp. are usually resistant. Many anaerobic bacteria, including Bacillus fragilis
are inhibited at concentrations of <10 g/mL.34
PHARMACOLOGY
Chloramphenicol is not absorbed in any signicant amount
when applied topically, but it is rapidly and completely absorbed
from the gastrointestinal tract and peak serum levels are reached
in 2 h. It diffuses well into many tissues and body fluids, includ-
CHAPTER 19
SULFONAMIDES AND TRIMETHOPRIM

ADVERSE EVENTS
PHARMACOLOGY
Orally administered sulfonamides are absorbed quickly and completely from the gastrointestinal tract. They are metabolized in
the liver and excreted by the kidney. The sulfonamides are well
distributed with levels in the CSF and synovial, pleural, and
peritoneal fluids of ~80% of the levels found in serum. During
pregnancy, they cross the placenta and enter into the fetal
circulation. The antibacterial action of the sulfonamides can be
inhibited by blood, pus, and tissue enzymes, because the bacterial breakdown requirements of folic acid decrease in media
that contain purines and thymidine. Therefore, they should not
211

be used for infections with marked purulent exudation. Orally
administered trimethoprim is absorbed almost completely in
the gastrointestinal tract. Peak levels are reached in the serum
in 14 h and are distributed widely in various body tissues,
including the kidneys, lungs, and prostrate. Its half-life in
serum is ~10 h in healthy subjects. About 80% is excreted in
the urine; the remaining 20% is excreted as inactive metabolites
by the kidneys or bile.
SECTION 4
A relatively high incidence of bacterial resistance has occurred
for individual use of the sulfonamides and trimethoprim, however a synergistic activity to each other and other antiinfective
compounds allow these combinations to be useful for ophthalmic infections. Sulfonamides are available in a topical formulation as sulfacetamide. The formulation may also include a steroid
such as prednisolone for an antiinflammatory effect. Sulsoxazole
can be used to treat chlamydial urethritis, and sulfacetamide
ophthalmic solution has shown efcacy for trachoma and inclusion conjunctivitis. The sulfonamides are active against Nocardia asteroides and show moderate activity against several of the
atypical mycobacteria, especially in combination with trimethoprim.38 Trimethoprim in combination with polymyxin B is used
as a broad-spectrum topical solution for adult and pediatric
bacterial conjunctivitis.24,39 The addition of polymyxin B to
trimethoprim makes the combination more effective against
Gram-negative organisms, especially Pseudomonas spp.
ADVERSE EVENTS
The sulfonamides can produce a wide variety of side effects which
are common to the group. Crystalluria, hematuria, and anuria
have been associated as complications with systemic use of
these drugs. Anorexia, nausea, vomiting, and diarrhea are also
common side effects with systemic therapy. All routes of administration including topical may show hypersensitivity reactions,
including urticaria and rashes which may be accompanied by
pruritus and fever. Contact dermatitis is common with topical
application and have caused such problems as erythema nodosum,
erythema multiforme (StevenJohnson syndrome), and exfoliative dermatitis. Transient myopia has been known to occur with
systemic use.40 Myopia is usually bilateral, but the refractive
state usually returns to normal when the serum drug level
decreases. The most frequent adverse event from patients using
trimethoprimpolymyxin solution is local irritation, with transient burning or stinging, and itching following instillation.
Less than 2% of patients experience a hypersensitivity reaction
with this combination and there are no cross-allergic reactions
between trimethoprim/polymyxin B and the sulfonamides.
BACITRACIN AND GRAMICIDIN

Bacitracin and gramidicin are bactericidal peptide antibiotics
with similar activities against most Gram-positive organisms
(see Table 19.1). Bacitracin disrupts bacterial cell-wall synthesis
by inhibiting the dephosphorylation of a lipid pyrophosphate,
while gramicidin interferes more with cell membrane permeability. Bacitracin may also act as a chelating agent.
212
These drugs are primarily active against staphylococci and

group-A beta-hemolytic streptococci. Group C and G streptococci are less susceptible and group B is usually resistant.41
Neisseria spp. and Haemophilus influenzae may be susceptible

to bacitracin, but other Gram-negative organisms are resistant.
Some spirochetes, Entamoeba histolytica, Actinomyces, and
Fusobacterium have shown susceptibility to bacitracin.
PHARMACOLOGY
Bacitracin is limited to topical preparations for cosmetics, ophthalmic and cutaneous ointments, and solutions for wound
irrigation. Signicant amounts of bacitracin are not absorbed
systemically when used as a topical preparation. There is poor
penetration through the cornea, which may be enhanced in the
presence of an epithelial defect. Large doses of bacitracin used
for wound irrigation may be associated with systemic toxicity.
Bacitracin is generally combined with polymyxin B and zinc or
neomycin to provide a broad-spectrum antibiotic ointment for
ophthalmic infections. These ointments provide coverage for a
wide range of organisms implicated in conjunctivitis and staphylococcal blepharoconjunctivitis. Although bacitracin is unstable
in solution, gramicidin is not and can be combined with polymyxin B and neomycin to have a similar broad-spectrum product
in solution form.
ADVERSE EVENTS
Systemic administration of bacitracin results in signicant nephrotoxicity. Gramicidin is a potent hemolytic agent. Side effects
are rare when the drug is applied topically. It is generally nonirritating to skin and mucous membranes, however recent
reports concerning allergic contact dermatitis and anaphylaxis
may limit its use as a dermatological antibiotic.42,43
POLYMYXINS
Polymyxins are a group of related cyclic basic polypeptides
originally derived from Bacillus polymyxa (see Table 19.1). Polymyxin E (colistin) was used to treat Pseudomonas spp. infections
prior to the advent of newer antibiotics, but now polymyxin B
is primarily used in formulations for ocular infections. These
bactericidal agents interact with the phospholipids of the bacterial cell membrane, which increases the cell permeability and
disrupts osmotic integrity. This process results in leakage of
intracellular constituents, leading to cell death.
The polymyxins are active against most gram-negative bacilli,
especially Pseudomonas spp., Serratia spp., Proteus spp., and
Providencia spp. Gram-negative cocci, including Neisseria spp.,
are generally resistant, as are Gram-positive organisms. Crossresistance with other antibiotics has not been observed.
PHARMACOLOGY
The polymyxins may be administered parenterally, orally, or topically. They are not absorbed well when given orally or topically,
and intramuscular injections are usually painful. When used topically for ophthalmic infections they are poorly absorbed through
skin and mucous membrane surfaces. Polymyxin E is much less
irritating, but is about one-fourth as potent as polymyxin B. If
polymyxin B is used for irrigation of wound cavities or used in
subconjunctival injections, toxicity and necrosis may occur.
Antibacterials
ADVERSE EVENTS
Polymyxin B is generally used in combination products to provide the necessary Gram-negative coverage. Combinations with
trimethoprim, bacitracin, or neomycin are available commercially
for ophthalmic infections in either an ointment or solution
formulation. Combinations with an added antiinflammatory
agent are also available for more persistent ocular infections
such as staphylococcal blepharitis.
Systemic use of the polymyxins is limited due to possible severe

neural and renal toxicity. Dose-related renal dysfunction occurs
in ~20% of patients on appropriate therapy. Allergic reactions
including fever and skin rashes are rare, but may occur after
rapid intravenous infusion. Topical administration of polymyxin
B may cause hypersensitivity reactions, and chronic use may
lead to toxic conjunctivitis.
1. Hooper D, Wolfson J: Fluoroquinolone

antimicrobial agents. N Engl J Med 1991;
324:384394.
2. Chen FJ, Lo HJ: Molecular mechanisms of
fluoroquinolone resistance. Microbiol
Immunol Infect 2003; 36:19.
3. Blondeau JM: Expanded activity and utility
of the new fluoroquinolones: a review. Clin
Ther 1999; 21:340.
4. Gatifloxacin and moxifloxacin: two new
fluoroquinolones. Med Lett Drugs Ther
2003; 42:1517.
5. Van der Auwera P, Matsumoto T, Husson M:
Intraphagocytic penetration of antibiotics.
J Antimicrob Chemother 1988; 22:185192.
6. Gwon A: Ofloxacin vs tobramycin for the
treatment of external ocular infections.
Ofloxacin Study Group II. Arch Ophthalmol
1992; 110:12341237.
7. Power WJ, Collum LM, Easty DL, et al:
Evaluation of efcacy and safety of
ciprofloxacin ophthalmic solution versus
chloramphenicol. Eur J Ophthalmol 1993;
3:7782.
8. Hwang DG, Schanzlin DJ, Rotberg MH,
et al: A phase III, placebo controlled clinical
trial of 0.5% levofloxacin ophthalmic
solution for the treatment of bacterial
conjunctivitis. Br J Ophthalmol 2003;
87:10041009.
9. Yee RW, Tepidino M, Bernstein P, et al: A
randomized, investigator-masked clinical
trial comparing the efcacy and safety of
gatifloxacin 0.3% administered bid versus
qid for the treatment of acute bacterial
conjunctivitis. Curr Med Res Opin 2005;
21:425431.
10. Silver LH, Woodside AM, Montgomery DB:
Clinical safety of moxifloxacin ophthalmic
solution 0.5% (Vigamox) in pediatric and
nonpediatric patients with bacterial
conjunctivitis. Surv Ophthalmol 2005;
50(Suppl):S55S63.
11. Hyndiuk RA, Eiferman RA, Delmar RC, et al:
Comparison of ciprofloxacin ophthalmic
solution 0.3% (Ciloxan) to fortied
tobramycin/cefazolin in the treatment of
bacterial corneal ulcers. Ophthalmology
1996; 103:18541863.
12. OBrien TP, Maguire MG, Fink NE, et al:
Efcacy of ofloxacin versus cefazolin and
tobramycin in the therapy for bacterial
113:12571265.
13. Romanowski EG, Mah FS, Yates KA, et al:
The successful treatment of gatifloxacinresistant Staphylococcus aureus keratitis
with Zymar (gatifloxacin 0.3%) in a NZW
rabbit model. AJO 2005; 139:867877.
14. Prajna V, Vajpayee R, Trocme S, et al:
Safety and efcacy of gatifloxacin 0.3% as
compared with ciprofloxacin 0.3% for the
treatment of acute bacterial, corneal ulcers.
ARVO 2006; B277 (poster #1916).
15. Jensen HG, Zerousala C, Carrier M, et al:

Comparison of ophthalmic gatifloxacin
0.3% and ciprofloxacin 0.3% in healing of
corneal ulcers associated with
Pseudomonas aeruginosa induced
ulcerative keratitis in rabbits. J Ocular
Pharm Ther 2005; 21:3643.
16. Aliprandis E, Ciralsky J, Lai H, et al:
Comparative efcacy of topical
moxifloxacin versus ciprofloxacin and
vancomycin in the treatment of P.
aeruginosa and ciprofloxacin-resistant
MRSA keratitis in rabbits. Cornea 2005;
24:201205.
17. Leibowitz HM: Clinical evaluation of
ciprofloxacin 0.3% ophthalmic solution for
treatment of bacterial keratitis. Am J
Ophthalmol 1991; 112:34S47S.
18. Bower KS, Kowalski RP, Gordon YJ:
Fluoroquinolones in the treatment of
bacterial keratitis. Am J Ophthalmol 1996;
121:712715.
Meyer DR: The role of tetracycline in
chronic blepharitis inhibition of lipase
production in staphylococci. Invest
20. Perry HD, Hodes LW, Seedor JA, et al:
Effects of doxycycline hyclate on corneal
epithelial wound healing in the rabbit alkali
burn model. Cornea 1993; 12:379382.
21. Grossman ER, Walcheck A, Freedman:
Tetracycline and permanent teeth: the
relationship between doses and tooth
color. Pediatrics 1971; 47:567570.
22. Smith CR, Lipshy JJ, Laskin OL, et al:
Double-blind comparison of the
nephrotoxicity and auditory toxicity of
gentamicin and tobramycin. N Engl J Med
1980; 302:11061109.
23. Gorzyski EA, Amsterdam D, Beam TR Jr,
Rotstein C: Comparative in vitro activities
of teicoplanin, vancomycin, oxacillin, and
other antimicrobial agents against
bacteremic isolates of gram-positive cocci.
33:20192022.
24. Eliopoulos GM: Vancomycin-resistant
enterococci: mechanism and clinical
relevance. Infect Dis Clin North Am 1997;
11:851865.
25. Schwalbe RS, Stappleton JT, Gilligan PH:
Emergence of vancomycin resistance in
coagulase negative staphylococci. N Engl
J Med 1987; 316:927931.
26. Plugfelder SC, Hernandez E, Fliesler SJ,
et al: Intravitreal vancomycin. Retinal
toxicity, clearance and interaction with
gentamicin. Arch Ophthalmol 1987;
105:831837.
27. Polk RE, Healy DP, Schwartz LB, et al:
Vancomycin and the red-man syndrome:
pharmacodynamics of histamine release.
J Infect Dis 1988; 157:502507.
28. Kucers A: A good antimicrobial prescribing:

chloramphenicol, erythromycin,
vancomycin, tetracyclines. Lancet 1982;
2:425429.
29. Petroutos G, Guimarres R, Pouliquen Y:
The effect of concentrated antibiotics on
the rabbits corneal epithelium. Int
30. Barry AL, Jones RN, Thornsberry C: In vitro
activities of azithromycin (CP 62,993),
clarithromycin (A-56268, TE-031),
erythromycin, roxithromycin and
clindamycin. Antimicrob Agents Chemother
1988; 32:752754.
31. Wallace RJ Jr, Swenson JM, Silcox VA,
Bullen MG: Treatment of non-pulmonary
infections due to Mycobacterium fortuitum
and Mycobacterium chelonei on the basis
of in vivo susceptibilities. J Inf Dis 1985;
152:500514.
31A. Protzko EE, Abelson MB, Shapiro AM; the
AzaSite Clinical Study Group: A randomized
trial assessing the safety and tolerability of
1.0% azithromycin ophthalmic solution vs
tobramycin in pediatric and adult subjects
with bacterial conjunctivitis. Presented at:
The association for research in vision and
ophthalmology annual meeting,
Ft. Lauderdale, FL, 30 Apr 30 4 May 4 2006.
31B. Abelson MB, Protzko EE, Shapiro AM; the
AzaSite Clinical Study Group. A randomized
trial assessing microbial eradication and
clinical efcacy of 1.0% azithromycin
ophthalmic solution vs tobramycin in
pediatric and adult subjects with bacterial
conjunctivitis. Presented at: The association
for research in vision and ophthalmology
annual meeting. Ft Lauderdale, FL, 30 Apr
to 4 May 2006.
32. Miller KE: Diagnosis and treatment of
Chlamydia trachomatis infection. Am Fam
Physician 2006; 73:14111416.
33. Chidambaram JD, Alemayehu W,
Melese M, et al: Effect of a single mass
antibiotic distribution on the prevalence of
infectious trachoma. JAMA 2006;
295:11421146.
34. Musial CE, Rosenblatt JE: Antimicrobial
susceptibilities of anaerobic bacteria
isolated at the Mayo Clinic during 1982
through 1987: comparison with results from
1977 through 1981. Mayo Clin Proc 1989;
64:392399.
35. The TrimethoprimPolymyxin B Sulfate
Ophthalmic Ointment Study Group:
Trimethoprimpolymyxin B sulfate
ophthalmic ointment versus
chloramphenicol ophthalmic ointment in
the treatment of bacterial conjunctivitis-a
review of four clinical studies. J Antimicrob
Chemother 1989; 23:261266.
36. Fraunfelder FT, Bagby GC Jr, Kelly DJ:
Fatal aplastic anemia following topical
administration of ophthalmic
CHAPTER 19
REFERENCES
213
SECTION 4
chloramphenicol. Am J Ophthalmol 1982;

93:356360.
37. Fraunfelder FT, Bagby GC Jr: Letter to the
editor. Ocular chloramphenicol and aplastic
anemia. N Engl J Med 1983; 308:1536.
38. Rodloff AC: In-vitro susceptibility test of
non tuberuculous mycobacteria to
sulphamethoxazole, trimethoprim, and
combinations of both. J Antimicrob
Chemother 1982; 9:195199.
214
39. Wagner RS: Results of a survey of children

with acute bacterial conjunctivitis treated
with trimethoprimpolymyxin B ophthalmic
solution. Clin Ther 1995; 17:875881.
40. Rittenhouse EA: Myopia after use of
sulfanilamide. Arch Ophthalmol 1940;
24:11391143.
41. Finland M, Garner C, Wilcos C, Sabath LD:
Susceptibility of beta-hemolytic
streptococci to 65 antibacterial agents.

9:1119.
42. Jacob SE, William JD: From road rash to
top allergan in a flash. Bacitracin Derm
Surg 2004; 30:521524.
43. Dyck ED, Vadas P: Anaphylaxis to topical
bacitracin. Allergy 1997; 52:870871.
CHAPTER
20
Antivirals
Deborah Pavan-Langston and Thomas John
Viruses are obligate intracellular parasites that use the

metabolic processes of the invaded host cell. Therefore, a major
challenge in antiviral therapy has been formulating antiviral
drugs that do not interfere with the normal host-cell
metabolism by causing toxic side effects in the uninfected host
cells. Theoretically, antiviral drugs may be effective by
interacting directly with the virus, a virus-encoded enzyme or
protein, or a cellular receptor or factor required for viral replication or pathogenesis.1 To date, the most effective molecular
targets of antiviral treatment have been the viral enzymes and
proteins that play a role in the assembly of the virus. The
continuing search for new antiviral agents may result in the
development of drugs that are effective at one or more stages of
viral infection of the host cell, particularly the initial adherence
or adsorption of the virus to the host cells by electrostatic
interaction and receptors; viral penetration into the host cell
(e.g., by pinocytosis); release of viral nucleic acid by uncoating;
and replication, transcription, and translation of viral genome
within the infected host cell. The development of antiviral
drugs that are licensed currently for clinical use is the result of
an increased understanding of the molecular biology of viral
structures, enzymes, and replicative mechanisms and

virushost-cell interactions. Although newer antiviral agents
are being introduced into the marketplace, continued research
in this eld is required to provide better and safer antiviral drugs
in the future.
CLASSIFICATION OF VIRUSES
Viruses are made up of a nucleic acid core that contains either
ribonucleic acid (RNA) or deoxyribonucleic acid (DNA) and is
surrounded by a protein-containing outer coat. The classication of a virus is based on the type of nucleic acid core (RNA
or DNA). Viruses can also be subdivided based on their
morphology, the site of viral multiplication (in the nucleus or in
the cytoplasm of the host cell), and serologic type.
The eye and adnexal structures may be directly infected by
RNA and DNA viruses or involved secondarily as part of a
systemic viral infection. Viral infections of ocular importance
are described in detail under Viral Infections of the Cornea and
Anterior Segment and under Retinitis. The viral families are
outlined in Table 20.1.
TABLE 20.1. Virus Classications

Classication
Family
Examples
RNA viruses
Togaviridae
Rubella virus (rubella, German measles)
Paramyxoviridae
Measles virus (rubeola, measles); mumps virus (mumps, epidemic parotitis);

Newcastle virus (Newcastle disease)
Orthomyxoviridae
Influenza virus (influenza)
Picornaviridae
Enterovirus type 70 (acute hemorrhagic conjunctivitis, picornaviral hemorrhagic

conjunctivitis); coxsackie A24 virus (acute hemorrhagic conjunctivitis, picornaviral
hemorrhagic conjunctivitis)
Rhabdoviridae
Rabies virus (rabies, hydrophobia)
Retroviridae
Human immunodeciency virus types 1 and 2 (HIV-1, HIV-2) (AIDS)
Herpesviridae
Herpes simplex virus (HSV) types 1 and 2 (herpes simplex infection, cold sores,
keratitis, genital infections, encephalitis); varicella-zoster virus (VZV) herpes virus 3
(chickenpox and shingles); EpsteinBarr virus (EBV) or herpesvirus 4 (infectious
mononucleosis, association with Burkitts lymphoma); cytomegalovirus (CMV) or
herpesvirus 5 (CMV disease, cytomegalic inclusion disease)
Adenoviridae
Adenovirus types 3 and 7 (pharyngoconjunctival fever, acute follicular conjunctivitis);

adenovirus types 8, 19, and 37 (epidemic keratoconjunctivitis)
Poxviridae
Molluscum contagiosum virus (molluscum contagiosum); vaccinia virus

(ocular vaccinia); variola virus (smallpox)
DNA viruses
215
SECTION 4
UNITED STATES FOOD AND DRUG

ADMINISTRATION APPROVED ANTIVIRAL
DRUGS
Twenty antiviral drugs are currently FDA approved for clinical
use. Half are for the treatment of HIV infections. The others are
for herpesvirus (e.g., herpes simplex virus (HSV), varicella
zoster virus and cytomegalovirus), hepatitis B virus, hepatitis C
virus, and influenza virus infections. Recent studies have
focused on antiviral therapies for viral infections that appear
amenable to antiviral drug treatment, as well as for viral
infections for which, to date, no antiviral drugs have been
approved (e.g., adenoviruses, human herpesvirus type 6, poxviruses, coronavirus, severe acute respiratory syndrome, and
hemorrhagic fever viruses).2 Gardasil (Merck), a vaccine against
human papilloma virus (HPV) was approved by the FDA in June
of 2006, and Glaxo-Smith-Kline is expected to see approval for
its HPV vaccine, Cervarix, in early 2007. Vaccines have also
been approved for human papilloma viruses related to cervical
carcinoma as well as for varicella (Varivax, Merck) and for
herpes zoster (Zostavax, Merck). The various types of antiviral
drugs are outlined in Table 20.2.3,4
This chapter reviews idoxuridine, trifluridine, vidarabine,
acyclovir, valacyclovir, and famciclovir, and discusses glanciclovir, foscarnet, cidofovir, and bromovinyldeoxyuridine where
pertinent.5,6 The numerous anti-HIV agents include nonnucleoside reverse transcriptase inhibitors, nucleoside/nucleotide
reverse transcriptase inhibitors, and protease inhibitors.
Together, they are used in various combinations to make up
Key Features: Virus Types and Antiviral Targets
Viruses are classied based on their nucleic acid core: either

RNA or DNA. The core is surrounded by a protein outer coat
and sometimes a triple membrane
The most effective molecular targets of antiviral treatment are
the viral enzymes and proteins that play a role in viral assembly
Twenty antiviral drugs are currently FDA approved for clinical
use, nine with proven efcacy in ocular viral disease:
idoxuridine (IDU, Herplex), vidarabine (Ara-A, Vira A), trifluridine
(TFT, F3T, Viroptic), acyclovir (ACV, Zovirax), famciclovir (FCV,
Famvir), and valacyclovir (VCV, Valtrex), and
bromovinyldeoxyuridine (BVDU, Brivudine). Ganciclovir
(DHPG*, Cytovene), foscarnet (PFA, Foscavir), and HPMPC
(Cidovir). All but BVDU are FDA approved
HAART, which stands for highly active antiretroviral therapy.

HAART is extremely useful in boosting the bodys immune
response, thereby enhancing the efcacy of the antiviral defenses,
often without additional alternative antiviral therapy.
IDOXURIDINE (STOXIL, HERPLEX)

Idoxuridine (5-iodo-2-deoxyuridine, IDU, Herplex), a
nucleoside analog of thymidine, was the rst clinically effective
antiviral drug used as a topical ophthalmic preparation.710
Thymidine, a nucleoside found in DNA, has a methyl group at
the 5 position of the pyrimidine ring. In IDU the methyl group
is replaced by a single iodide substituent (Fig. 20.1). This
chemical substitution provides IDU with its antiviral property.
It replaces thymidine in the enzymatic step of viral replication.
Thus, IDU irreversibly inhibits the incorporation of thymidine
into viral DNA. This incorporation of the thymidine analog,
namely, IDU, into viral DNA renders the newly formed viral
particles noninfective.11,12 However, newer drugs (particularly
trifluridine) are more efcient and have higher efcacies;
therefore, idoxuridine is no longer commercially available.
TRIFLURIDINE (VIROPTIC)
Trifluridine (5-trifluoromethyl-2-deoxyuridine, trifluorothymidine, F3T, Viroptic) is a fluorinated nucleoside analog of
thymidine. The methyl group at the 5 position of the
pyrimidine ring of thymidine (see Fig. 20.1) is changed in F3T
such that each hydrogen of the methyl group is replaced by a
fluoride substituent (Fig. 20.1). This chemical change provides
F3T with its antiviral properties. TFT is a potent inhibitor of
thymidylate synthetase and therefore inhibits DNA synthesis.
Trifluridine is incorporated into viral DNA directly, rendering
the viral particle noninfectious.13 However, its antiviral
mechanism of action is not fully known. In addition, F3T is also
incorporated into mammalian cells. It has exerted mutagenic,
DNA-damaging, and cell-transforming activities in various
standard in vitro test systems. From a clinical standpoint, the
signicance of these test results has yet to be fully understood.
Trifluridine is active against HSV types 1 and 2 and vaccinia
virus, both in vitro and in vivo. It also has an in vitro inhibitory
effect against some strains of adenovirus.
Trifluridine in a 1% solution is twice as potent and 10 times
more soluble than IDU.1417 It is also lipid-soluble. The drugs
biphasic solubility enhances corneal penetration by simple
TABLE 20.2. Antiviral Drugs
216
Antiviral Drug
Abbrev.
Brand Name
Proven Efcacy in
Ocular Viral Disease
Specialized Roles
FDA Approved in
One or More Forms3,4
Idoxuridine*
IDU
Herplex
Vidarabine*
ara-A
Vira A
Trifluridine
TFT, F3T
Viroptic
Acyclovir
ACV
Zovirax
Famciclovir
FCV
Famvir
Valacyclovir
VCV
Valtrex
Bromovinyldeoxy
uridine
BVDU
Brivudine
Ganciclovir
DHPG
Cytovene
Foscarnet
PFA
Foscavir
Cidofovir
HPMPC
Vistide
*No longer commercially available because other drugs offer greater convenience, or because of overlapping efciencies.
Antivirals
O
O
CH3
HOCH2
N
O
HOCH2
N
O
HOCH2
H2N
N
N
OH
O
C
H2N
HOCH2
OH
HOCH2
H2N
HOCH2
HOCH2
OH H
ACYCLOVIR
NH2
DEOXYGUANOSINE
N
O
CH3
HO
P
HO
O NM
CH3
CH2OCH2CH2OC C CH2(CH2)2
VALACICLOVIR
H2N
GANCICLOVIR
FOSCARNET
H2N
OH
HN
H2N
HO
PENCICLOVIR OH
OH H
OH H
VIDARABINE
IDOXURIDINE TRIFLURIDINE
O
* - racemic phosphorothiate
N
HOCH2
5-d-[G*C*G*T*T*T*G*C*T*C*T*T*C*T*T*C*T*T*G*C*G]-3
sodium salt
O
HO
From Pavan-Langston D: Ocular pharmacology of

antiviral drugs. In: Tasman W, Jaeger E, Wilhelmus K,
eds. Duanes foundations of clinical ophthalmology.
Philadelphia, PA: JB Lippincott; 2004:124.
HO
OH
THYMIDINE
FIGURE 20.1. Structures of thymidine and

antivirals.
NH3
HN
CF
CHAPTER 20
O
I
OH
H
O
CIDOFOVIR
FAMCICLOVIR
NH3
O H
Br
HN
Br
H
MN
O
O
HO P CH2
OH
O
O
CH
CH2 OH
HPMPC
CH2
HO-H2C
HO-H2C
HO
OH
HO
BV Ara U
diffusion.18 Trifluridine penetrates the intact cornea into the
aqueous humor, and corneal penetration is further enhanced by
epithelial disruption. Experimentally, F3T is partly metabolized
to 5-carboxy-2 deoxyuridine as the drug passes through the
cornea, as evidenced by the presence of both F3T and 5-carboxy2-deoxyuridine on the endothelial side. In a rabbit model of
herpetic uveitis, topical F3T was shown to be effective because
of its penetration into the anterior chamber.19 In another study
of rabbit herpes simplex keratouveitis, 1% F3T and 0.1% IDU
had almost identical control of uveitis, keratitis, and
conjunctivitis.20 The efcacy of topical 1% F3T was also
demonstrated in rabbits with herpes simplex keratitis (HSK)
and may also be due to its intracorneal penetration property.21
As in the experimental studies, intraocular penetration of
topical F3T has been shown to occur in humans.22 This
penetration of F3T into the aqueous humor may be enhanced in
the presence of compromised corneal integrity and corneal
stromal or uveal inflammation. However, unlike the in vitro
results of ocular penetration of F3T, 5-carboxy-2-deoxyuridine
was not found in detectable concentrations within the aqueous
humor at the time of penetrating keratoplasty in patients who
BVDU
received F3T preoperatively.22 The passage of F3T through the
human cornea without undergoing any signicant metabolic
degradation has not been found to be therapeutically helpful in
the treatment of herpes keratouveitis.
Systemic absorption of F3T following therapeutic dosing
appears to be negligible. The half-life of F3T in serum is only
12 min; therefore, it is ineffective as a systemic antiviral agent.
The drug should not be used during pregnancy unless the
potential benets outweigh the potential hazards to the fetus.
Although it is unlikely that F3T is excreted in human milk after
ophthalmic use, it should not be prescribed for nursing mothers
unless the potential benets outweigh the potential risks.
TFT is used as an effective topical therapy for HSV
keratitis.23,24 TFT not only interferes with the replication of
HSV-1 and HSV-2, but also has an effect on vaccinia and certain
adenoviruses.25 TFT (0.21.7 mg/mL) inhibits the cytopathic
effects of HSV-1 by 50% in plaque reduction assays.26 Plaque
formation was reduced by over 98% when HSV-1 grown into
Vero cells was treated with 17 mg/mL TFT.27 TFT activity in
vitro is comparable to IDU, and TFT is considerably more
active on a weight-for-weight basis than is vidarabine. As
217
SECTION 4

observed for both IDU and vidarabine, the strain of HSV-1
appears to be of major importance in determining the relative
antiviral efcacy. TFT was shown to inhibit ve strains of
HSV-1 within a narrow range; however, the susceptibility of ve
HSV-2 strains was variable, with two strains being insensitive
at the maximum nontoxic concentration.28
When TFT and IDU were compared with respect to their
ability to eradicate viruses from the preocular tear lm, no virus
was recovered on days 2 and 4 of the 7-day treatment with TFT.
However, HSV-1 was present in IDU treated eyes throughout
the treatment regimen.29,30 Two days following discontinuation
of therapy, rebound virus shedding had occurred in both TFT
and IDU groups, with virus titers higher than those observed in
control, placebo-treated animals. These results indicate that a
critical time period exists in an acute herpetic infection during
which time continued presence of the antiviral is necessary to
control rebound virus shedding, even though infectious viruses
cannot be detected in the tear lm.
Clinical studies comparing topical 1% TFT to 0.1% IDU
drops, to 3.0% vidarabine, and to acyclovir ointments have
shown that overall, the latter two drugs and TFT have efcacy
rates between 90% and 95%, even if steroids are in use.3135 IDU
efcacy was only ~76%. There are several hypotheses for this
reduced efcacy: possible steroid use in some patients, perhaps
the agent has been in clinical use for so long that certain
organisms have become resistant, or patients may have become
allergic to the agent. While TFT had a slight edge over all other
drugs in the face of concomitant steroid therapy, no statistically
signicant difference could be shown.
The usual recommended dosage for infectious HSV dendritic
or dendrogeographic ulcers is nine times daily for 5 days and, if
the keratitis is improving, ve times daily for a total of 23
weeks.3,5,6,24,36 This should produce a therapeutic response
within 24 days, and complete healing of more than 90% of
uncomplicated cases in 12 weeks. Atopic or immunosuppressed patients may take somewhat longer and need
combined oral and topical therapy. Patients with a history of
IDU treatment failure are usually responsive to TFT (87%).24,30
One study, however, demonstrated that IDU-resistant HSV is
cross-resistant to acyclovir, of intermediate resistance to TFT,
and fully sensitive to vidarabine and ganciclovir.
If healing has not occurred by 3 weeks, the possibility of a
toxic or trophic epithelial defect should be considered and
management should be changed. TFT should be stopped, and
lubricant antibiotic ointment, such as bacitracin or polymyxinbacitracin initiated TID. TFTs toxic side effects may mimic
infectious disease; these side effects include follicular conjunctivitis, supercial punctate keratitis (SPK), toxic epithelial ulceration, lacrimal punctal occlusion, anterior segment ischemia,
interference with wound healing, and true allergic blepharodermatitis.3335,37,38 Severe reversible ocular anterior segment
ischemia following topical F3T treatment for herpes simplex
keratouveitis has also been reported. Table 20.3 gives a more
complete listing of toxic reactions from topical antivirals. These
effects are rarely seen when the drugs are used for 2 weeks or
less and are reversible on cessation of their administration.3941
Trifluridine is supplied as a 1% sterile ophthalmic solution
that should be refrigerated (28C; 3646F). The preservative
in F3T 1% solution is thimerosal 0.001%.
VIDARABINE (VIRA-A)
218
Vidarabine (9-b-D-arabinofuranosyladenine) is a substituted

purine nucleoside previously known as adenine arabinoside
(Ara-A, Vira) (Fig. 20.1). Once widely available commercially for
topical and intravenous use, it is now available only through
compounding pharmacists for patients unable to use alternative
Key Features: Trifluorothymidine (TFT)
TFT eyedrops not only interfere with the replication of HSV-1

and HSV-2 but also have an effect on vaccinia and certain
adenoviruses (DNA viruses)
Clinical studies comparing topical TFT drops, vidarabine and
acyclovir ointments in HSV dendrogeographic keratitis have
shown that all three have efcacy rates between 90% and
95% regardless of whether steroids are in use
If healing has not occurred by 3 weeks, the possibility of a toxic
or trophic epithelial defect should be considered. TFT should be
stopped and lubricant antibiotic ointment TID initiated
TFT toxic side effects may mimic infectious disease; these include
follicular conjunctivitis, SPK, toxic epithelial ulceration, lacrimal
punctal occlusion, anterior segment ischemia, interference with
wound healing, and true allergic blepharodermatitis
TABLE 20.3. Topical Antiviral Ocular Toxicity

Site
Toxicity
Cornea
Fine punctate epithelial keratopathy

Filamentary keratitis
Indolent corneal ulceration
Perilimbal edema
Late supercial vascularization
Supercial stromal opacication
Conjunctiva
Punctate staining with rose bengal or

fluorescein
Follicular conjunctivitis
Chemosis, congestion
Perilimbal edema
Conjunctival scarring
Lid margins
Edema of meibomian gland orices

Punctal edema and occlusion
Lids
Ptosis
Allergic contact blepharodermatitis
Other
Preauricular lymphadenopathy
antivirals. Along with TFT, it is recommended for therapy of

vaccinia blepharokeratoconjunctivitis.42
Vidarabine was the second antiviral drug developed for
human use.43 Researchers rst synthesized the compound in
the early 1960s as a potential anticancer agent.44,45 It has
subsequently been obtained from fermentation cultures of
Streptomyces antibiotics.46 The mechanism of action of
vidarabine, although not fully established, appears to interfere
with the early steps of viral DNA synthesis and arrests the
growth of the viral deoxynucleotide chain. It is known that it is
not a completely selective antiviral agent. Although vidarabine
can affect normal cells, it is thought to be sufciently safe for
systemic use. Vidarabine is rapidly deaminated to hypoxanthine
arabinoside (Ara-Hx). The principal metabolite, Ara-Hx,
possesses antiviral activity that is less potent than the parent
drug, vidarabine. Vidarabine is effective against herpes simplex,
varicella-zoster, and vaccinia (DNA viruses).4749 It has a limited
range of activity against RNA viruses and no antiviral action
against adenovirus keratoconjunctivitis.50 Subepithelial corneal
inltrates developed in both vidarabine-treated patients and
controls.
Because vidarabine is relatively insoluble, it is formulated as
a 3% ophthalmic ointment. The recommended dosage is ve
times a day at 3-h intervals. Clinicians should consider other
forms of treatment if there is no clinical improvement after
1 week, or if complete corneal reepithelialization fails to occur
within 3 weeks. Following reepithelialization, an additional
week of treatment at a reduced dosage of twice daily should be
Antivirals
Key Features: Vidarabine 3% Ointment
Vidarabine is effective against herpes simplex, varicella zoster,

and vaccinia (DNA viruses)
It is now available only through compounding pharmacists for
patients unable to use alternative antivirals
In clinical trials no signicant difference was noted between
vidarabine, trifluridine, or 3% acyclovir ointment in the
treatment of dendritic or geographic HSK
Vidarabine side effects are generally mild but may include
corneal epithelial punctate keratopathy, punctal occlusion,
conjunctival hyperemia, irritation, photophobia, and lacrimation
shown vidarabine to be effective in many patients intolerant of

or resistant to IDU.62
ACYCLOVIR (ZOVIRAX)
Acyclovir (9-2-hydroxyethoxymethyl guanine, ACV, Zovirax), a
second-generation antiviral drug, is a synthetic purine
nucleoside analog derived from guanine. It differs from guanine
by the presence of an acyclic side chain.3,63,64 Acyclovir is used
against HSV and VZV in pill or liquid form, intravenously (IV),
and as a dermal ointment. It is also available as a 3%
ophthalmic ointment for HSV infections in Canada and Europe
and through compounding pharmacists in the United States.
There are multiple clinical uses of ACV. This important drug
is indicated or has been effective in the following conditions:
(1) primary genital HSV (PO or IV), (2) recurrent genital HSV in
immunocompetent patients (PO), (3) mucocutaneous HSV in
immunocompromised patients (PO or IV), (4) HSV encephalitis
(IV), (5) neonatal HSV (IV), (6) varicella in immunocompetent
(PO) or immunocompromised patients (PO or IV), (7) herpes
zoster in immunocompetent (PO) or immunocompromised
patients (IV or PO), and (8) possibly in EBV infections (PO). It
also has antiviral activity against EBV, herpes simiae (B virus),
and CMV but is infrequently used to treat these infections.
When used to treat HSV and VZV, acyclovir interferes with
DNA synthesis, thus inhibiting virus replication. In
herpesvirus-infected cells in vitro, the antiviral activity of
acyclovir appears to be dependent primarily on the intracellular
conversion of acyclovir to acyclovir triphosphate. The
conversion of acyclovir to acyclovir monophosphate occurs
mainly via virus-coded thymidine kinase (TK). Acyclovir
monophosphate is phosphorylated to the diphosphate via
cellular guanylate kinase and to the triphosphate via other
cellular enzymes (e.g., phosphoglycerate kinase, pyruvate
kinase, phosphoenolpyruvate carboxykinase). In contrast,
acyclovir is only minimally phosphorylated by host cell
enzymes in uninfected cells in vitro. Because acyclovir has
antiviral activity against viruses that seem not to code for viral
TK (e.g., EBV and CMV), acyclovir is apparently converted to
acyclovir triphosphate by other mechanisms. However, research
suggests that acyclovir triphosphate is at least partially
produced within herpesvirus EBV- and CMV-infected cells; the
responsible cellular phosphorylating enzymes have not yet been
identied. The exact mechanisms of action against other
susceptible viruses are not fully understood.6569
Acyclovir takes advantage of the subtle differences between
viral and cellular enzyme function in DNA synthesis. A slight
difference exists between the viral and cellular TK. Because
acyclovir is a nucleoside analog, it can function as a substrate
for viral TK but not for cellular TK. Therefore, acyclovir can
enter the sequence of DNA formation primarily in virusinfected cells. The viral DNA polymerase more effectively
utilizes the acyclovir triphosphate than does the cellular DNA
polymerase. The viral DNA polymerase has a 10- to 30-fold
greater afnity in vitro for the acyclovir triphosphate than the
cellular a-DNA polymerase. When the acyclovir analog enters
the DNA chain, DNA synthesis is terminated. Thus, viral
DNA growth is more susceptible to acyclovir than the DNA of
uninfected host cells.7073 Because of its poor uptake into these
cells, acyclovir has minimal pharmacologic effects in vitro on
the uninfected host cells; phosphorylation and intracellular
conversion to acyclovir triphosphate are minimal, and the
cellular a-DNA polymerase has a low afnity for acyclovir
triphosphate.
Acyclovir has been detected in the brain, kidney, saliva, lung,
liver, muscle, spleen, uterus, vaginal mucosa and secretions,
semen, cerebrospinal fluid, and herpetic vesicular fluid.
CHAPTER 20
continued to prevent recurrence of infection. Vidarabine

treatment should not be continued for more than 3 weeks.
Vidarabine penetrates the aqueous humor better than IDU.
Two hours after topical application of 3% vidarabine in petrolatum to rabbit eyes, aqueous levels of 6 mg/mL of the drug were
detected; 0.5% IDU failed to produce any detectable aqueous
levels.51 This is compatible with the clinical impression that
vidarabine treatment may be useful in herpetic uveitis.
Although vidarabine has been used intravenously in humans for
herpetic uveitis, this is not a popular mode of treatment.52
Vidarabine was also the rst drug shown to be effective systemically in the treatment of herpetic encephalitis.53
Like other antiviral agents, vidarabine is not free from side
effects; a common one is corneal epithelial punctate keratopathy.54,55 Other possible adverse reactions include foreign body
sensation, lacrimation, conjunctival hyperemia, burning,
irritation, pain, photophobia, sensitivity, and punctal occlusion.
Signicant systemic absorption of vidarabine is not expected to
occur after topical ocular use. Animal trials have shown that
vidarabine is rapidly deaminated to its principal metabolite,
Ara-Hx, in the gastrointestinal tract. Although the chance of
fetal damage with ocular use of vidarabine during pregnancy is
remote, it is best avoided unless the potential benet of therapy
justies any potential risk to the fetus.
No signicant difference was noted between vidarabine and
trifluridine in the treatment of herpes simplex dendritic corneal
ulcers.56,57 However, trifluridine was slightly more effective than
vidarabine in the treatment of herpes simplex geographic
corneal ulcers.56 A multicenter study involving 66 patients
compared the overall efcacy of 3% vidarabine ointment with
3% acyclovir ointment in the treatment of dendritic or geographic herpetic keratitis. No statistically signicant difference
existed between the two medications with regards to healing rate,
the nal visual acuity, the frequency of selected complications
such as punctate epithelial keratitis, or the development of
stromal keratitis.58 This is contrary to the earlier in vitro and
animal experiments, the results of which suggested that acyclovir
might be a more effective antiviral agent than vidarabine.59,60
Experimentally, vidarabine was compared with IDU to
evaluate which drug was less toxic to the corneal epithelium.61
The rate of rabbit corneal epithelial wound closure of 5- and
10-mm epithelial defects was not signicantly different among
the eyes treated with 3% vidarabine, 0.5% IDU, and placebo
antibiotics, indicating that neither 3% vidarabine nor 0.5% IDU
retarded corneal epithelial wound healing.40,47,61 The quality of
the regenerated corneal epithelium as evaluated by slit lamp
was signicantly better with vidarabine than with IDU.
However, 3% vidarabine and 0.5% IDU, and 1% TFT all
interfere with stromal healing to the same degree.
Vidarabine therapy may be useful in cases of IDU resistance.
In one study in which vidarabine 3% ointment was used to treat
56 cases of IDU-resistant HSK, 80% of epithelial herpes
keratitis cases and 52% of herpes stromal keratitis cases healed
within 2 weeks of treatment initiation.55 Others have also
219
SECTION 4
220
Acyclovir diffuses into cerebrospinal fluid and crosses the placenta. There is evidence that the drug is distributed into milk
via an active transport mechanism. Acyclovir is metabolized to
9-carboxymethoxymethylguanine (CMMG) and 8-hydroxy-9(2-hydroxyethoxymethyl) guanine. In in vitro herpesvirusinfected cells, acyclovir is metabolized to acyclovir mono-, di-,
and triphosphate. The drug is excreted mainly in urine, via
glomerular ltration and tubular secretion.
Acyclovir is a crystalline white powder with a solubility of
1.3 mg/mL in water at 25C. Commercially available acyclovir
sodium is a sterile, white, crystalline, lyophilized powder. At a
pH of 7.4 and 37C, it is almost completely unionized and has
a maximum solubility of 2.5 mg/mL. Acyclovir capsules, pediatric suspension, and the commercially available acyclovir sodium
sterile powder should be stored in tight, light-resistant containers
at 1525C. Reconstituted acyclovir sodium solution (50 mg
acyclovir/mL) is stable for 12 h at 1530C. Upon refrigeration,
a precipitate may form which will redissolve at room temperature. This precipitation and subsequent redissolution do not
appear to affect drug potency. Bacteriostatic water that contains
parabens should not be used for injection because this diluent
is incompatible with the drug and may cause precipitation.
Poirier and colleagues evaluated the intraocular penetration
of 3% acyclovir ointment, vidarabine monophosphate, and 1%
F3T drops following their administration to patients with normal
corneas before cataract extraction.74 The authors detected substantial levels of acyclovir in the aqueous humor, although only
meager levels of vidarabine monophosphate. In addition, no
F3T was detected. Hence, 3% acyclovir may be superior to other
antiviral agents with regards to corneal penetration and in the
treatment of deep herpetic keratitis and uveitis. However,
acyclovir topical treatment did not signicantly reduce the incidence of stromal keratitis that developed with herpes simplex
epithelial keratitis.75
Three groups of rabbits with herpes simplex corneal infections were treated ve times a day with 0.5% IDU, 3% vidarabine, or 3% acyclovir ointment. There was 50% less incidence
of severe iritis, epithelial loss, and conjunctivitis in the acyclovir
group compared with the other groups.76 Also, recoverable virus
levels on day 6 were much less in the acyclovir-treated rabbit eyes
compared with the other two groups. Acyclovir does not interfere with corneal epithelial or stromal healing in rabbit eyes.
Pavan-Langston and associates compared the efcacy of acyclovir 3% ointment with vidarabine 3% ointment in the treatment of patients with dendritic or geographic herpes keratitis.73
Within 2 weeks, more than 90% of the patients healed, with no
signicant difference between the two drugs. However, herpes
dendritic corneal ulcers healed more rapidly when 3% acyclovir
was combined with debridement when compared with 3%
acyclovir alone (2 and 5 days, respectively).77 In its antiherpetic
effect, acyclovir is comparable to topical F3T.78
The use of oral acyclovir has had a revolutionary effect on the
treatment and prognosis of herpetic disease in every parameter
of the disease, in both immunocompetent and immunocompromised patients: genital herpes simplex infections, herpes
simplex encephalitis, acute herpes zoster (shingles), VZV
(chickenpox), and in mucosal or cutaneous herpes simplex
(HSV-1 and HSV-2) infections. In a study comparing oral acyclovir (400 mg ve times daily) to 3% acyclovir ointment (ve
times daily) in the treatment of herpes simplex dendritic corneal
ulceration, the authors found that healing occurred within
5 days in 89% of patients on oral acyclovir and in 97% of patients
on topical acyclovir ointment.79 Thus, oral acyclovir may be an
alternative to topical acyclovir ointment for the treatment of
herpes simplex dendritic lesions. In a controlled trial of oral
acyclovir versus placebo for 7 days with minimal wiping,
debridement in herpes simplex dendritic corneal ulcers was
carried out in 31 patients.80 At the end of treatment, the corneal

lesions had healed in 67% of patients receiving acyclovir and in
43% of patients given a placebo. Although there was no
signicant difference in the proportion of corneal lesions that
healed in the two groups at 7 days, the rate of healing was
signicantly faster in the acyclovir group.80,81 Jensen and
colleagues found that 3% topical acyclovir ointment was useful
both in epithelial and stromal herpes simplex corneal
infections. However, they also found that acyclovir ointment
was equally effective in herpetic keratitis in patients either
receiving debridement or no debridement.82
For ocular HSV, oral ACV 400 mg 5id is equivalent to topical
ACV in treating epithelial keratitis, with 90% of patients ulcers
healing in a mean of 5 days.75,79,83,84 Two hundred mg PO 5id
healed 18 of 19 patients with combined HSV epithelial and
stromal keratitis in 521 days.84 Other studies have conrmed
these ndings.7786 The therapeutic pediatric dosage is 20
40 mg kg1 day1 for 714 days as a pediatric elixer (200 mg/tsp).83
A series of federally funded multicenter studies on the
efcacy of oral ACV and/or topical steroids on various forms of
ocular HSV were reported between 1994 and 2001.8693 The
results may be briefly summarized as follows: (1) One year of
ACV 400 mg bid signicantly reduced recurrence of herpetic
disease after resolution of any form of ocular HSV and without
rebound up to 6 months after ACV was stopped. (2) There was
no statistically signicant benet of ACV 400 mg 5/day for
10 weeks in treating active HSV stromal keratitis in patients
already on steroids and TFT. (3) Steroids were signicantly
more effective than the placebo in resolving active stromal
keratitis, and postponing steroids slowed resolution, but had no
difference in outcome by 6 months. (4) Treatment of iritis with
ACV 400 mg 5/day for 10 weeks in patients already on steroids
and TFT may have some benecial effect. (5) A 3-week course
of ACV 400 mg 5/day for epithelial keratitis in patients already
on TFT did not alter the subsequent incidence of stromal
keratitis or iritis. (6) ACV 400 mg bid for 1 year signicantly
reduced the recurrence of HSV stromal or epithelial keratitis,
with greatest benet in the stromal group. (7) Previous stromal
keratitis markedly increased the risk of the disease recurring.
Previous epithelial keratitis did not correlate with increased
recurrence. (8) Psychological stress, sun exposure, contact lens
wear, and systemic illness could not be shown to be triggers for
HSV reactivation. (9) The number of past episodes of either
epithelial or stromal keratitis was strongly associated with the
likelihood of a recurrence. (10) Long-term oral ACV signicantly lowers recurrence of either form of HSV keratitis, but is
more effective in preventing recurrence in stromal disease than
epithelial. It should be noted that prophylactic use of oral
antivirals is legitimate but dened as off-label use.
In a study of 105 HSK patients, Wu and Chen reported an
even more positive effect of prophylaxis on preventing recurrent
epithelial HSV in nonkeratoplasty patients than did the HEDS
study.94 Using low-dose ACV at 300 mg/day for 1 year, they
found a statistically signicant difference between treated and
control groups: ve recurrences of epithelial and one case of
stromal keratitis in the ACV group, and 14 cases of epithelial and
four cases of stromal keratitis in the untreated control group.
In their study on long-term oral acyclovir therapy in reducing
recurrences of dendritic or geographic HSK, Simon and PavanLangston evaluated 13 patients with a history of frequently
recurring HSK (mean 27 months) during long-term oral
acyclovir.95 Eight were followed after the acyclovir was discontinued. Treatment was given for an average of 34 months. During
treatment, the number of recurrences per month decreased from
0.15 to 0.03, and the average duration of relapses decreased
from 12.6 to 7.8 days (p < 0.01). Recurrences correlated with
daily doses of oral acyclovir (usually less than, but no more than
800 mg), intraocular surgery within 6 weeks of initiating

treatment, and discontinuation of therapy against medical
advice. This small study demonstrates that therapeutic doses of
oral acyclovir reduce both the rate and duration of recurrences
of infectious herpetic keratitis.
Additional indications for PO acyclovir in patients with
herpetic keratitis include use as an adjunct to topical antivirals
in patients with atopic disease or in immunosuppressed patients,
especially in AIDS patients. In AIDS patients, PO or IV therapy
is determined on the basis of severity of immunosuppression
(CD4+ helper lymphocytes less than 200 cells/mL, viral burden
> 1057 plasma HIV RNA copies/mL).5,6,9699 Dosage in atopy of
400 mg PO 5id for 23 weeks is generally quite effective.
Another indication for IV therapy is for patients who are unable
or unwilling to take topical antiviral agents for epithelial keratitis, such as those with crippling arthritis; children or uncooperative adults; those whose occupation makes topical agents
difcult to use; and those with ocular toxic medicamentosa
from local antivirals.
Dosage, although not FDA approved for ocular use, in
nonatopic patients weighing over 50 pounds is 400 mg PO
tid5id. For children who weigh less than 50 pounds, dosage is
2040 mg/kg in divided dose for 710 days. Prophylaxis of HSV
epithelial recurrences in post-HSV keratoplasty patients with
PO ACV is effective and indicated.95,99,100 It is discussed in
further detail in the chapter on Viral Infections of the Cornea
and Anterior Segment under HSV Surgical Factors and
Management.
ACV-resistant HSV strains have been isolated with greater
frequency from patients with AIDS. These strains do not
produce TK for drug activation.5,63,101,102 Alternative drugs for
treatment of ACV-resistant strains include vidarabine, which is
phosphorylated by cellular (rather than viral) TK, and foscarnet,
which requires no phosphorylation.103,104 Clinical experience is
limited with these alternatives, although some clinical success
has been reported.105 Sonkin and associates reported isolation of
an ACV-sensitive HSV with altered TK activity from a keratoplasty patient.106 However, the clinical course deteriorated on
treatment, and ACV-resistant HSV with decient TK activity
was isolated. Despite foscarnet sensitivity, the graft eventually
failed. Foscarnet therapy of resistant HSV is discussed further in
the section titled Foscarnet.107
In a more recent study in the Netherlands, where ACV
is available over the counter, there were 542 isolates from 496
patients (410 HSV-1 and 132 HSV-2).108 Thirteen isolates
(8 HSV-1 and 5 HSV-2) from 10 patients (2%) were found
resistant to ACV. A single ACV-resistant strain was identied
among isolates from 368 immunocompetent patients (0.27%).
Resistant HSV was identied in nine isolates from 128 immunocompromised patients (p < 0.05). None of the isolates were
resistant to foscarnet. This study indicates that the prevalence
of ACV resistance is low in immunocompetent patients
(0.27%), but relatively frequent in immunocompromised
patients (7%; p < 0.0001). Also, drug susceptibility monitoring
of HSV infections is essential in immunocompromised patients
with persisting infections, even with antiviral therapy.83 The
management approach to therapy of ACV-resistant HSV is
reviewed in the clinical section on Ocular HSV in Immunocompromised Patients.
The role of ACV in herpes zoster ophthalmicus (HZO) is well
established for systemic use; however, because this drug is now
a second-line agent, famciclovir and valaciclovir are currently
the drugs of choice. Data are mixed for topical therapy, with
some studies reporting 5% acyclovir ointment 5id was highly
effective in resolving zoster epithelial keratitis and in
signicantly reducing the incidence of recurrent disease. Other
studies showed topical steroids were useful in the management
of the inflammatory aspects of zoster ophthalmicus, but

showed no clear benet of topical acyclovir ophthalmic
ointment when used alone.109 Zaal et al have reported that at
3 months post onset of HZO, patients who received 3% topical
ACV had longer duration of periocular lesions and signicantly
more visual loss compared to the group receiving oral ACV, and
that all patients on combined topical ACV and dexamethasone
drops developed chronic disease.110
Because of the high complication rate in HZO, several
studies have been conducted to compare ACV to placebo or
other antiviral therapy. ACV 800 mg PO 5id for 710 days
induces prompt resolution of rash and pain, induces more rapid
healing, reduces the duration of viral shedding and reduces the
duration of new vesicle formation. There is also a signicant
reduction in the incidence and severity of acute dendritiform
keratopathy, scleritis, episcleritis, and iritis. The incidence (but
not the severity) of corneal stromal immune keratitis, and the
incidence of late-onset ocular inflammatory disease (e.g.,
episcleritis, scleritis, iritis) was also reduced.111115 The effect on
post herpetic neuralgia (PHN) was variable, with some reports
showing no efcacy, and others showing a notable decrease in
severity and incidence.111,114117 As noted below, both
famciclovir and valaciclovir are superior in their effect on PHN.
Gastrointestinal upset, particularly diarrhea, is a distressing
side effect of ACV. This appears to be due to lactose intolerance,
which is relatively common in North American adults: ~75%
of Native Americans and blacks, 51% of hispanics, and 21% of
whites are lactose intolerant.118 ACV tablets contain lactose;
intolerance to lactose is a common cause of intolerance to milk
and milk products in people who lack the intestinal enzyme
lactase. Manka has reported reversal of this oral ACV side-effect
by administration of oral lactase in the form of one Lactaid
caplet ve times daily PO. Other systemic side effects of oral
acyclovir include nausea, vomiting, and headache. Less
common adverse reactions include dizziness, anorexia, fatigue,
edema, skin rash, leg pain, inguinal adenopathy, taste
perversion, and sore throat.
Acyclovir has also been used intravitreally experimentally
and clinically.119121 Two patients with acute retinal necrosis
were treated with intravitreal infusion of acyclovir, vitrectomy,
and prophylactic scleral buckles; both patients had an
uneventful postoperative course and recovered visual acuity.
CHAPTER 20
Antivirals
Key Features: Acyclovir (ACV, Zovirax)
Oral ACV 400 mg 5id resolves 90% of infectious HSV

dendrogeographic corneal ulcers in a mean of 5 days.
Recommended effective dosages range from 400 mg PO tid to
5id with higher doses usually for immune-altered patients such
as excema or HIV+ patients
The HEDS study key ndings include: (1) 1 year of ACV 400 mg
bid signicantly reduced recurrence of herpetic disease after
resolution of epithelial or stromal keratitis; (2) no effect on
active stromal keratitis and equivocal effect on iritis;
(3) previous stromal, but not epithelial, keratitis markedly
increased the risk of recurrent similar disease in the future; and
(4) the number of past episodes of either epithelial or stromal
keratitis was strongly associated with the likelihood of a
recurrence
For zoster ophthalmicus ACV 800 mg PO 5id for 710 days,
induces signicant resolution of rash, pain, new vesicles and
viral shedding, lower incidence and severity of acute and late
dendritiform keratopathy, scleritis, episcleritis, iritis, and the
incidence but not severity of stromal keratitis
Poor effect on postherpetic neuralgia makes this drug now a
second choice for zoster infections, famciclovir or valaciclovir
being the drugs of choice
221
SECTION 4
VALACICLOVIR (VCV, VALTREX)

Valaciclovir (L-valine 2-(guanin-9-ylmethoxy)ethyl ester, VCV,
Valtrex) is a prodrug of acyclovir synthesized to enhance GI
uptake of ACV. It is hydrolyzed back to ACV, increasing the
bioavailability to ve times that of ACV.122,123 Studies on the
ocular penetration of acyclovir and its prodrugs following systemic administration revealed that the plasma bioavailability
for acyclovir, valacyclovir, and val-valacyclovir were similar, but
that anterior segment area under curve values were 53.70
(35.58), 139.85 (9.43), and 291.05 (88.13) min mmol/L,
respectively. The mean residence time values were 46.47
(24.94), 76.30 (7.24), and 188.39 (80.73) min, respectively.
This indicates that the valine and valine-valine ester prodrugs
of ACV penetrated the anterior segment of the eye signicantly
better than acyclovir alone, probably via a carrier mediated
transport mechanism. Vitreous levels of the prodrugs were not
measurable.124 Viral susceptibility is similar to that of ACV.
Clinical studies comparing valaciclovir 1.0 g PO tid with
acyclovir 800 mg PO 5 day for 7 or 14 days in 1141
immunocompetent zoster patients (35 with HZO) revealed
drug-equivalency in acceleration of dermal healing and
reduction of duration of viral shedding. VCV was signicantly
better in acute pain resolution and reduced duration of PHN
through 1 year of follow up.125,126 Data from 14 days of
treatment did not differ from that of 7 days. Studies on PHN
revealed that the median time to pain resolution was 38 days
with VCV and 51 days with acyclovir (p < 0.03). Other studies
support the high efcacy of valaciclovir in herpes zoster,
particularly if started within 72 h of rash onset.127131
In a study by Sozen et al of 30 eyes in 28 ocular HSV
patients, patients were randomized to receive either topical
ACV ointment or oral VCV. The corneal lesions healed
signicantly faster in the oral VCV-treated eyes than in the
topical ACV-treated eyes. Symptoms were also lower in the
VCV group.132
Only one ocular study compared ACV with VCV. In 121
immunocompetent patients with acute HZO; an incidence of
keratitis, uveitis, and episcleritis was reported that was similar
in both groups.133 Neither group had any incidence of
neurotrophic keratitis or scleritis, and acute pain was noted in
~66% of each group. It was concluded that VCV was a valid
alternative to ACV in treatment of HZO, but as with
famciclovir (see further ahead), was superior in acute and longterm pain inhibition and in patient compliance with only tid
dosing. The absence of neurotrophic keratopathy is in marked
contrast to this authors experience, however.
For ocular HSV, VCV dosages have been adopted from genital
HSV data.3 Recommendations for acute genital HSV are VCV
1 g PO bid for 710 days. For recurrent genital HSV, VCV therapy
is 500 mg PO bid for 5 days and for suppression of recurrent
Key Features: Valaciclovir (VCV, Valtrex)
222
VCV is a prodrug of acyclovir with enhanced GI uptake. It is

hydrolyzed back to ACV resulting in ve times the
bioavailability of the latter drug. VCV is very effective clinically
in HSV-1, HSV-2, and VZV infections
For ocular HSV VCV dosages have been adapted from genital
HSV data. Author-suggested doses for acute infections are
somewhat higher
Clinical studies comparing valciclovir with acyclovir in
treatment of zoster ophthalmicus revealed drug-equivalency
but also that VCV was signicantly better in acute pain
resolution and reduced duration of postherpetic neuralgia, thus
making it one of the two drugs of choice for zoster
episodes of HSV 500 mg to 1 g PO qd up to 1 year. The author

recommends 250 mg PO bid tid for acute or recurrent episodes to add therapeutic leeway. Tolerance to valaciclovir, like
its active metabolite acyclovir, is generally good. Central neurological toxicity may be observed with high doses, but regresses
on withdrawal.134 Some severely immunocompromised HIV
patients have developed thrombocytopenic purpura/hemolytic
uremic syndrome, with a few deaths. As a result, this drug is
not FDA approved for use in immunocompromised patients,
but is approved for therapy of varicella zoster and genital HSV.
FAMCICLOVIR (FCV, FAMVIR)

Famciclovir (2-(acetyloxymethyl)-4-(2-aminopurin-9-yl)-butyl
acetate, FCV, Famvir), a third-generation nucleoside, is the
orally bioavailable diacetyl ester of the active antiviral,
penciclovir. It is similar to ACV in mechanism of action and
antiviral activity against HSV-1 and HSV-2, VZV, and
EpsteinBarr virus but superior in GI absorption: 77% versus
only 30% for ACV.135139 FCV is metabolized to penciclovir
intracellularly, where it is active 1020 times as long as ACV.
It inhibits viral DNA polymerase-mediated chain elongation.
The drug is FDA approved for treatment of herpes zoster
infection at doses of 500 mg tid for 7 days. It is preferable to
start treatment within 72 h of onset of rash. Clinical studies
indicate that famciclovir accelerates healing time as well as
ACV, but with less frequent dosing.140144
For therapy of ocular HSV, VCV dosages have been adopted
from genital HSV data.3 For acute rst episodes, it is 250 mg PO
tid for 710 days. For recurrent episodes, 125 mg bid for 5 days
is recommended. For the eye, however, 710 days at 250 PO
bidtid is recommended by the author.
The efcacy and safety of long-term randomized, doubleblind, placebo-controlled famciclovir for suppression of
recurrent genital HSV (> 6 recurrences per year) revealed that
with doses of 250 mg PO BID for 52 weeks, a signicantly
greater proportion of famciclovir-treated patients (151/191,
79%) were free from HSV recurrences at 6 months than placebo
recipients (48/184, 26%) (p < 0.001); efcacy was maintained at
12 months. Furthermore, the median time for the rst clinically
conrmed lesional episode was signicantly prolonged for the
famciclovir group (more than 1 year) compared with
the placebo group (59 days; p < 0.0001). Famciclovir was well
tolerated, with an adverse-experience prole comparable to
placebo.145
Prophylaxis of herpesvirus infections using any of the oral
antivirals usually involves preventing the recurrence of HSV,
but also on rare occasions to prevent complications of herpes
zoster in immunocompetent patients. In immunocompromised
patients, prophylaxis is used to prevent opportunistic virus reactivation by HSV or VZV. The effectiveness of acyclovir 400
mg twice daily in preventing the recurrence of HSV eye disease
in immunocompetent patients has been well demonstrated in
HEDS. The issue of treatment duration for patients with highly
recurrent ocular herpes remains unresolved.146 In the authors
personal experience ACV, VCV, and FCV are all comparable in
their efcacy as prophylactic antivirals.
FCV is a very effective drug in acute zoster. As with VCV, prevention of PHN has occurred with antiviral therapy (famciclovir
500 mg PO tid or valaciclovir 1 g PO tid 7 days), started
within 72 h of onset of the rash, and with analgesic treatment.
However, the best adjunct for minimizing or even preventing both
acute zoster and PHN is a zoster vaccine (Zostavax, Merck)
recently approved by the FDA. In a double-masked study of
more than 37 000 patients, there was signicantly lower incidence (>55%) and severity (>60%) of both zoster and PHN.147
Key Features: Famciclovir (FCV, Famvir)
FCV, the diacetyl ester of an ACV relative, penciclovir, is similar

to ACV in mechanism of action and antiviral activity against
HSV-1 + HSV-2, and VZV but superior in GI absorption and
intracellular half-life
For ocular HSV FCV dosages have been adopted from genital
HSV data. Author-suggested doses for acute infections are
somewhat higher
Progress in prevention of PHN has been made with FCV
antiviral therapy started within 72 h of onset of the rash, and
analgesic treatment
GANCICLOVIR (DHPG, GCV, CYTOVENE)

Ganciclovir
(9-(1,3-dihydroxy-2-propoxy(methylguanine)),
DHPG, GCV, Cytovene), a synthetic purine nucleoside analog
of guanine, is structurally and pharmacologically related to
acyclovir. It differs from acyclovir only by a second terminal
hydroxymethyl group at C-2 of the acyclic side chain on the
ribose ring.148 This structural difference contributes to the
substantially increased antiviral activity of ganciclovir against
CMV and in less selectivity for viral DNA. Although ganciclovir
has antiviral activity both in vitro and in vivo against various
herpesviridae (herpes simplex types 1 and 2, human herpesvirus
type 6, EBV, and VZV), its main clinical use has been against
human CMV.
The exact mechanism of action of ganciclovir is not fully
known. It appears to exert its antiviral effect on human CMV
and other human herpesviruses by interfering with DNA
synthesis via competition with deoxyguanosine for incorporation into viral DNA, and by incorporation into growing
viral DNA chains.149151 The formation of ganciclovir
monophosphate appears to be the rate-limiting step in the
formation of ganciclovir triphosphate. In contrast to acyclovir,
which is only minimally phosphorylated by cellular (host cell)
enzymes, ganciclovir seems to be more susceptible to phosphorylation by enzymes in uninfected cells, especially in rapidly
dividing cells (e.g., bone marrow). This phosphorylation in
uninfected cells can range from less than 10% to being equal to
that in virus-infected cells. Unfortunately, this also makes the
drug more toxic to the bone marrow, causing a signicant
neutropenia in more than 50% of patients treated. Other less
frequent side effects include nausea, neurotoxicity, hepatic
dysfunction, fever, and local rash or phlebitis (DHPG = pH 11).
DHPG is also carcinogenic, teratogenic, and induces azoospermia.
The phosphorylated form of ganciclovir that is active can
competitively inhibit viral DNA polymerase and can also be
incorporated into growing DNA chains as a false nucleotide.
This results in the termination of DNA synthesis and in the
formation of a mutant DNA chain, and thus inhibition of viral
replication. Although the drug inhibits cellular a-DNA polymerase, it requires a higher concentration than that required to
inhibit viral DNA polymerase. The increased antiviral effect of
ganciclovir against CMV compared with acyclovir has been
attributed to slower catabolism of ganciclovir triphosphate by
intracellular phosphatases. The drug does not code for TK and
is, therefore, of use in TK-resistant HSV and VZV strains.
As with all other antivirals, ganciclovir is virostatic rather
than virucidal.5 Because it is only virostatic, continuous therapy
with the IV drug is necessary to prevent viral breakthrough in
the immunosuppressed patient. However, despite careful
management, ~40% of patients ultimately experience reactivation of disease. Experimentally, when the drug is removed from
culture medium in vitro, previously inhibited viral DNA
synthesis resumes with restored viral replication. Additional

data supporting ganciclovir as virustatic come from histopathologic studies of enucleated globes from patients who died
while receiving ganciclovir therapy.152,153 These studies showed
that ganciclovir does not eliminate CMV from the retina, nor
does it suppress expression of all viral genes.
Because ganciclovir is poorly absorbed from the gastrointestinal tract, intravenous administration is preferred.
Ganciclovir is 12% bound to plasma proteins. Although the
tissue distribution of ganciclovir is not fully known, autopsy
studies on patients who received intravenous ganciclovir
suggest that the drug concentrates mainly in the kidneys with
lower concentrations in the liver, lung, brain, and testes.154 The
drug appears to have good ocular distribution following
intravenous administration; concentrations in the aqueous and
vitreous humors 2.5 h after intravenous administration were,
respectively, 0.4 and 0.6 times the simultaneous plasma
concentration of the drug.155 Ganciclovir crosses the blood
brain barrier. It is unknown whether ganciclovir is distributed
into human milk; however, no drug is present in the milk of lab
animals. It also crosses the placenta in lab animals. The
primary route of excretion is in urine, and it appears to be
mainly via glomerular ltration. Except for intracellular phosphorylation of the drug, it is not signicantly metabolized in
humans and is mainly excreted unchanged in the urine.
As noted, the primary clinical use of ganciclovir is in the
treatment of CMV retinitis in immunocompromised patients,
especially those with AIDS. The safety and efcacy of the drug
have not been established for congenital or neonatal CMV
disease, for the treatment of other cytomegaloviral infections,
such as pneumonitis or colitis, or for use in nonimmunocompromised individuals. The intravenous route of ganciclovir
therapy has been shown to be effective in the treatment of
cytomegaloviral retinitis in immunocompromised patients.156162
However, because ganciclovir is only suppressive against CMV
it does not result in increased immunocompetence the
retinitis will recur or progress following cessation. After induction therapy with ganciclovir for CMV retinitis and discontinuation of the drug, relapse of CMV usually occurs within
4 weeks in immunosuppressed patients. Hence, for the duration of the patients immunosuppression, long-term maintenance therapy and intermittent induction therapy seem to be
necessary. The advent of HAART for the treatment of AIDS
itself, however, has greatly reduced the number of cases of CMV
retinitis in the past few years.163
The most common dose-limiting adverse effect of ganciclovir
is neutropenia (absolute neutrophil count < 1000/mm3), which
is potentially fatal. Usually, interruption of ganciclovir therapy
or a decrease in dosage results in increased neutrophil counts.
Thrombocytopenia (platelet count < 50 000/mm3) can also
result from a direct, dose-dependent effect of the drug. Less
commonly, anemia and eosinophilia can occur. Ocular side
effects include rhegmatogenous retinal detachment as a result
of ganciclovir-induced resolution of retinitis. As a result,
ganciclovir has also been administered intravitreally in patients
with CMV retinitis.164169 It was found to be effective and safe
both as an alternative to intravenous ganciclovir therapy in
myelosuppressed patients and as a supplement to intravenous
therapy in uncontrolled CMV retinitis.170
Ganciclovir may also have a topical therapeutic role. Two
randomized HSV clinical trials have been carried out in Africa
and Europe comparing ganciclovir 1.5% gel with 3% ACV
ointment in treating herpetic keratitis in 107 patients.171 There
was no statistically signicant difference between the treatment
groups, although the group receiving 0.15% ganciclovir gel had
healing rates of 85% compared with 72% in the group receiving
CHAPTER 20
Antivirals
223
SECTION 4

acyclovir ointment. Local tolerance was superior with the gel
formulation of ganciclovir with fewer complaints of discomfort
(stinging, burning) or blurred vision after application. Systemic
absorption of the drug was low and no hematologic changes
were detected.
The drug should be stored at room temperature and should
not be exposed to temperatures greater than 40C. Reconstituted ganciclovir sodium solution with sterile water for
injection (ganciclovir 50 mg/mL) is stable for 12 h at 1530C
and should not be refrigerated, as a precipitate may form. To
avoid precipitation, bacteriostatic water for injection containing
parabens should not be used to reconstitute ganciclovir sodium.
Oral ganciclovir and ganciclovir implants are effective
alternative routes of drug administration. Oral ganciclovir is
valganciclovir (Valcyte), which has a much higher GI absorption
than its prodrug form. As a result, it may be given in
therapeutically effective doses for treatment of CMV retinitis.
Dosage is 900 mg PO bid for 3 weeks, then 900 mg PO qd.
Myelosuppression and CNS or liver toxicity are potential side
effects.172
The ganciclovir implant (Vitrasert) reflects an alternative
approach to treating CMV retinitis by providing local
concentrated therapy to the infected retina without the risks of
systemic toxicity associated with other routes of administration.173 Additionally, the sustained intravitreal release of
ganciclovir negates the need for repeated injections. The
implant is placed surgically in the vitreous cavity, and can
provide therapeutic levels of up to 8 months depending on the
rate of drug release.174,175 Although the ganciclovir implant has
been shown to be effective in treating CMV retinitis, there was
the increased risk of CMV retinitis developing in the fellow eye
and of systemic involvement in the patients who received
implants compared with patients who received the drug
intravenously. To decrease this risk, these patients may be given
oral ganciclovir.176 On the whole, intravitreal therapy has been
well tolerated, and local reactions (such as foreign-body
sensation, small conjunctival or vitreous hemorrhage, conjunctival scarring, and scleral induration) have been noted only
occasionally in patients receiving multiple intravitreal
injections (see Table 20.3). Because of the high pH of the
ganciclovir infusion solution, inflammation, phlebitis, and pain
at the site of intravenous infusion can occur.
Key Features: Ganciclovir (DHPG, Cytovene)
224
Ganciclovir, a synthetic purine nucleoside analog of guanine, is

structurally and pharmacologically related to acyclovir. It is
poorly absorbed from the gastrointestinal tract, necessitating
IV or intravitreal administration
In contrast to acyclovir, which is only minimally phosphorylated
by cellular (host cell) enzymes, ganciclovir seems to be more
susceptible to phosphorylation by enzymes in uninfected cells
thus making it more toxic
DHPG has good antiviral activity against HSV-1 and HSV-2,
VZV, EBV, and HHV-6, its clinical use is in CMV retinitis
primarily in immunocompromised patients. Because ganciclovir
is only suppressive against CMV; without improvement in
immunocompetence, the retinitis will recur or progress
following cessation of drug
Oral ganciclovir is valganciclovir (Valcyte) which has a much
higher GI absorption than its prodrug form. As a result, it may
be given in therapeutically effective doses for treatment of
CMV retinitis
The ganciclovir implant (Vitrasert) provides local concentrated
therapy to the infected retina without the risks of systemic
toxicity
FOSCARNET (PFA, FOSCAVIR)

Foscarnet (phosphonomethanoic acid, phosphonoformic acid
trisodium, PFA, Foscavir), an organic analog of inorganic
pyrophosphate, is structurally unrelated to other available antiviral drugs. Following intravenous administration of foscarnet,
it is not metabolized to any signicant extent, and therefore
does not cause any major interference with the host cellular
processes.177 The drug is excreted renally. It is active against
herpesviruses (CMV, HSV, EBV, VZV), and the retrovirus HIV. It
inhibits herpesvirus DNA polymerases and HIV-1 reverse
transcriptase. Foscarnet directly affects the pyrophosphate
binding site of DNA polymerase and, therefore, does not require
phosphorylation to activate. Because it does not need
phosphorylation by TK to be activated, it is of use (and superior
to vidarabine) in treatment of ACV-resistant (and presumably
famciclovir or valaciclovir-resistant) HSV and VZV, which is
most commonly seen in AIDS patients. It is FDA approved for
treatment of CMV retinitis. In a rapid screen test for
susceptibility to acyclovir and foscarnet in 320 clinical HSV
isolates (16% type 1, 84% type 2), 60% were resistant to ACV
and only 5% were resistant to foscarnet. This correlated closely
with clinical response.178181
Like other antivirals, foscarnet is virustatic. It may be
administered intravenously or intravitreally (Table 20.4) to treat
CMV retinitis. Foscarnet has poor oral absorption, and
gastrointestinal side effects are common; therefore, it is not
used orally. Foscarnet should also not be administered by rapid
or bolus intravenous injection because the toxicity may be
increased by excessive drug levels in the plasma. An infusion
pump must be used.
Foscarnet, like ganciclovir, is considered a drug of choice to
treat CMV retinitis in patients with AIDS. It is especially useful
in those patients who are intolerant to (or unresponsive to)
ganciclovir therapy. Because foscarnet does not cause myelosuppression, it can be used in conjunction with zidovudine and
other antiretroviral agents. Foscarnet can be administered
intravenously in combination with ganciclovir in patients with
CMV retinitis that is resistant to one drug. This combination
therapy reduces the dosage of the individual drug, appears to be
fairly well tolerated, and has prolonged sight in patients with
CMV retinitis.182 In the initial treatment of CMV retinitis in
patients with AIDS, foscarnet seems to be as effective as
ganciclovir.183,184 However, to prevent recurrent CMV retinitis,
chronic maintenance therapy is required with foscarnet, as with
ganciclovir.185 Foscarnet is more effective than ganciclovir in
prolonging the lives of AIDS patients, which may be the result
of its anti-HIV effect, and because it can be used with
zidovudine.186
Foscarnet is not tolerated as well as ganciclovir by patients.
Side effects include fever and gastrointestinal upset, including
nausea, vomiting, diarrhea, anorexia, and abdominal pain. The
most signicant side effect with foscarnet is renal impairment.
It is necessary to monitor the serum creatinine levels and adjust
TABLE 20.4. Intravitreal Antivirals

Drug
Dosage
Ganciclovir (Cytovene)
200400 mg/0.1 mL
Foscarnet (Foscavir)
1200 mg/0.05 mL
Cidofovir (Vistide)
20 mg/0.1 mL
Prusoff WH, Bakhle YS, McCrea JF: Incorporation of 5-iodo-2-deoxyuridine into

the deoxyribonucleic acid of vaccinia virus. Nature 1963; 199:1310.
Antivirals
Key Features: Foscarnet (PFA, Phosphonoformate,

Foscavir)
Foscarnet, an organic analog of inorganic pyrophosphate, is

structurally unrelated to other available antiviral drugs but
effective against CMV, HSV, VZV, and EBV but used primarily
for CMV retinitis or ACV-resistant HSV. It may be administered
intravenously or intravitreally
As with ganciclovir, to prevent recurrent CMV retinitis, chronic
maintenance therapy is required with foscarnet
Intravitreal foscarnet has been used to treat CMV retinitis in
patients with AIDS. This route is useful for patients in whom
ganciclovir is contraindicated as a result of acyclovir allergy, and
intravenous foscarnet is contraindicated because of renal failure
CIDOFOVIR (HPMC, VISTIDE)

Cidofovir ((1-(4-amino-2-oxo-pyrimidin-1-yl)-3-hydroxy-propan-2-yl) oxymethylphosphonic acid, HPMPC, Vistide,
Forvade), another derivative of phosphonoformic acid, does not
require activation by TK. It works by DNA polymerase inhibition and resists degradation, thus persisting intracellularly
up to 65 h.194,195 It is effective against HSV-1 and HSV-2, VZV,
EBV, DHPG-sensitive and -resistant CMV, as well as several
adenoviruses. The drug is FDA approved for IV treatment of
CMV retinitis but has signicant toxic ocular side effects.196 It
has been used intravitreally to treat CMV retinitis in patients
with AIDS (see Table 20.4). Ocular side effects include
decreased intraocular pressure and mild uveitis.197
Cidofovir is also of interest as a broad-spectrum anterior
segment antiviral. In preclinical trials, it has been shown to be
therapeutically effective as a topical 0.2% drop against
adenovirus 5 and to be as effective as TFT against HSV-1.198,199
In a clinical case report of HSV-1 and HSV-2 infection in an
AIDS patient, topical HPMPC on the skin was therapeutically
effective when foscarnet and ACV had failed.200
For immunocompromised patients of any age, restoring

immunity inhibits or prevents herpesvirus disease, as demonstrated for cytomegalovirus (CMV) in AIDS patients receiving
HAART (highly active antiretroviral therapy).201 Specic
antiviral therapy during the initial period after transplantation
could prevent reactivation of HSV or CMV in seropositive
recipients. Whether preemptive therapy or prophylaxis with
ganciclovir is the optimal approach against CMV remains
controversial, and the relative merits and limitations of each
approach may guide the choice. In stem cell transplantation,
preemptive therapy with foscarnet avoids the neutropenia and
related complications associated with ganciclovir. In renal
transplant recipients, universal prophylaxis of CMV infection
with valaciclovir has the same efcacy as ganciclovir. Although
it is relatively toxic, cidofovir should be further evaluated
because of its in vitro activity against most DNA viruses.202
Key Features: Cidofovir (HPMPC, VISTIDE)
HPMPC, another derivative of phosphonoformic acid, does not

require activation by TK and persists intracellularly up to 65 h
It is effective against HSV-1 + HSV-2, VZV, EBV, DHPGsensitive and -resistant CMV as well as several adenoviruses
but used clinically as IV therapy for CMV retinitis
It has been used intravitreally to treat CMV retinitis in patients
with AIDS
BROMOVINYLDEOXYURIDINE (BVDU,
BRIVUDINE)
This antimetabolite ((E)-5-(2-bromovinyl)-2-deoxyuridine,
BVDU, Zostex, Zerpex, Zonavir, Brivudine) is a highly potent
and selective inhibitor of HSV-1 and VZV infections. The high
selectivity of BVDU, like ACV, VCV, and FCV, depends
primarily on a specic phosphorylation of BVDU by the virusencoded TK. It is a highly effective topical treatment of herpetic
keratitis, of recurrent herpes labialis, and of the systemic (oral)
treatment of herpes zoster.203 In studies on its efcacy in acute
zoster, there was equivalent efcacy of brivudin and famciclovir
regarding the prevention of chronic pain and the resolution of
symptoms. Compared with famciclovir, brivudin provides
equivalent efcacy and safety at a more convenient once-daily
dose schedule of 125 mg qd.204 Compared to ACV, BVDU was
signicantly better in resolution of PHN.205 The drug is
available throughout Europe, but has not yet been reviewed for
approval in the United States.206
CHAPTER 20
the drug dosage accordingly.187 Because foscarnet can alter

plasma electrolyte levels and cause seizures, patients treated
with foscarnet should be monitored.181,188192
The current foscarnet induction dose recommendations are
either 60 mg/kg three times a day or 90 mg/kg twice a day for
a 23-week period. Subsequent maintenance therapy is required
with foscarnet, and the dosage range suggested is 90
120 mg kg1 day1. Some doctors recommend the higher dosage
of 120 mg kg1 day1 to obtain a better response when treating
CMV retinitis without signicantly increasing toxicity.
Intravitreal foscarnet has been used to treat CMV retinitis in
patients with AIDS. This route is especially useful for patients
in whom ganciclovir is contraindicated as a result of acyclovir
allergy, and in whom intravenous foscarnet is contraindicated
because of renal failure. Foscarnet is passed through a 0.22-mm
lter, and 1200 mg (0.05 mL) is injected intravitreally.193 The
recommended dose is two injections of foscarnet as induction
therapy once per week for 3 weeks, followed by a maintenance
dose of one injection per week (see Table 20.4).193
Key Features: (E)-5-(2-Bromovinyl)-2-Deoxyuridine

(BVDU, Brivudin)
This antimetabolite, activated by virus-encoded TK, is a highly

potent and selective inhibitor of HSV-1 and VZV infections
It is highly effective topical treatment of herpetic keratitis and
recurrent herpes labialis and the systemic (oral) treatment of
herpes zoster
REFERENCES
1. Crumpacker CS: Molecular targets of
antiviral therapy. N Engl J Med 1989;
321:163168.
2. DeClercq E: Recent highlights in the
development of new antiviral drugs. Curr
Opin Microbiol 2005; 8:552560.
3. Abramowicz M, ed. Handbook of
antimicrobial therapy. 17th edn. New
Rochelle: The Medical Letter on Drugs and
Therapeutics; 2005:177179.
4. Hayden F: Antiviral drugs. In: Mandell G,

Bennett J, Dolin R, eds. Principles and
Practice of Infectious Diseases 6th ed.
Philadelphia, PA: Elsevier; 2005:515542.
5. Pavan-Langston D: Ocular pharmacology
of antiviral drugs. In: Jaeger E, Tasman W,
Wilhelmus K, eds. Phildelphia, PA: Williams
& Wilkins; 2004:124.
6. Pavan-Langston D: Viral disease of the
ocular anterior segment: basic science and
clinical disease. In: Foster S, Azar D,

Dohlman C, eds. Philadelphia, PA: Williams
& Wilkins; 2005:297397.
7. Robins RK: Nucleosides and nucleotides:
past, present, and future. Ann N Y Acad
Sci 1975; 255:597610.
8. Prusoff WH, Chen MS, Fischer PG, et al:
Role of nucleosides in virus and cancer
chemotherapy. Adv Ophthalmol 1979;
38:316.
225
SECTION 4
226
9. Kaufman HE: Clinical cure of herpes

simplex keratitis by 5-iodo-2-deoxyuridine.
Proc Soc Exp Biol Med 1962; 109:251252.
10. Kaufman HE, Martola EL, Dohlman CH:
Use of 5-iodo-2-deoxyuridine (IDU) in the
treatment of herpes simplex keratitis. Arch
Ophthalmol 1962; 68:235239.
11. Prusoff WH, Bakhle YS, McCrea JF:
Incorporation of 5-iodo-2-deoxyuridine into
the deoxyribonucleic acid of vaccinia virus.
Nature 1963; 199:13101311.
12. Jones BR: Prospects in treating viral
diseases of the eye. Trans Ophthalmol Soc
UK 1967; 87:537.
13. Prusoff WH, Goz B: Potential mechanisms
of action of antiviral agents. Fed Proc 1973;
32:1679.
14. Heidelberger C, Parsons DG, Remy DC:
Synthesis of 5-trifluoromethyluracil and
5-trifluoromethyl-2-deoxyuridine. J Med
Chem 1964; 7:18.
15. Kaufman HE, Heidelberger C: Therapeutic
antiviral action of 5-trifluoromethyl-2deoxyuridine in herpes simplex keratitis.
Science 1964; 145:585.
16. McGill JI, Holt-Wilson AD, McKinnon JR,
et al: Some aspects of the clinical use of
trifluorothymidine in the treatment of
herpetic ulceration of the cornea. Trans
Ophthalmol Soc UK 1974; 94:342.
17. McGill J, Fraunfelder FT, Jones BR: Current
and proposed management of ocular
herpes simplex. Surv Ophthalmol 1976;
20:358.
18. OBrien WJ, Edelhauser HF: The corneal
penetration of trifluorothymidine, adenine
arabinoside, and idoxuridine: a
comparative study. Invest Ophthalmol Vis
Sci 1977; 16:1093.
19. Sugar J, Varnell E, Centifanto Y,
Kaufman HE: Trifluorothymidine treatment
of herpetic iritis in rabbits and ocular
penetration. Invest Ophthalmol 1973; 12:532.
20. Pavan-Langston D, Lass J, Campbell R:
Antiviral drops: comparative therapy of
experimental herpes simplex keratouveitis.
21. McNeill JI, Kaufman HE: Local antivirals in
a herpes simplex stromal keratitis model.
22. Pavan-Langston D, Nelson DJ: Intraocular
penetration of trifluridine. Am J Ophthalmol
1979; 87:814.
23. OBrien W: Antiviral agents. Tabbara K,
Hyndiuk R, eds. Boston, MA: Little Brown;
1996:269280.
24. Pavan-Langston D, CS Foster:
Trifluorothymidine and idoxuridine therapy
of ocular herpes. Am J Ophthalmol 1977;
818825.
25. Hyndiuk R, Seideman S, Leibsohn J:
Treatment of vaccinial keratitis with
trifluorthymidine. Arch Ophthalmol 1976;
94:1976.
26. Lin T, Chai C, Prusoff W: Synthesis and
biological activities of 5-trifluoromethyl-5azido-2,5 dideoxyuridine and 5trifluoromethyl-5-amino-2,5-dideoxyuridin
e. J Med Chem 1976; 19:915.
27. Collins P, Bauer D: Comparison of activity
of herpes virus inhibitors. J Antimicrob
Chemother 1977; 3(Suppl A):73.
28. Reyes P, Heidelberger C: Fluorinated
pyrimidines: XXVI. Mammalian thymidylate
synthetase. Mol Pharmacol 1965; 1:14.
29. Dohlman C, Zucker B: Long-term treatment
with IDU and steroids. Arch Ophthalmol
1965; 74:172175.
30. Coster D, McKinnon J, McGill J, et al:

Clinical evaluation of adenine arabinoside
and trifluorothymidine in the treatment of
corneal ulcers caused by herpes simplex
virus. J Infect Dis 1976; 133:173177.
31. Margolis TP, Ostler HB, Wilhelmus K:
Treatment of ocular disease in eczema
herpeticum. Am J Ophthalmol 1990; 110:
274279.
32. Fardeau C, Langlois M, Nugier F, et al:
Cross-resistances to antiviral drugs of
IUdR-resistant HSV-1 in rabbit keratitis and
in vitro. Cornea 1993; 12:1924.
33. La Lau C, Oosterhuis J, Versteeg J, et al:
Acyclovir and thrifluorothymidine in
herpetic keratitis a multicenter trial. Br J
Ophthalmol 1982; 66:506508.
34. Hyndiuk R, Raimundo E, Charlin T, et al:
Trifluridine in resistant human herpetic
keratitis. Arch Ophthalmol 1978; 96:1839.
35. Wilhelmus K: The treatment of herpes
simplex virus epithelial keratitis. Trans Am
Ophthalmol Soc 2000; 98:505532.
36. Laibson PR, Arentsen JJ, Mazzanti WD,
Eiferman RA: Double controlled
comparison of IDU and trifluorothymidine in
thirty-three patients with supercial
herpetic keratitis. Trans Am Ophthalmol
Soc 1977; 75:316324.
37. Coster DJ, Jones BR, McGill JI: Treatment
of amoeboid herpetic ulcers with adenine
arabinoside or trifluorothymidine. Br J
38. Shearer D, Bourne W: Severe ocular
anterior segment ischemia after long-term
trifluridine treatment for presumed herpetic
keratitis. Am J Ophthalmol 1990;
109:346347.
39. Jayamanne DG, Vize C, Ellerton CR, et al:
Severe reversible ocular anterior segment
ischaemia following topical
trifluorothymidine (F3T) treatment for
herpes simplex keratouveitis. Eye 1997;
11(Pt 5):757759.
40. Foster C, Pavan-Langston D: Corneal
wound healing and antiviral medication.
41. Lass J, Langston R, Foster C, et al:
Antiviral medications and corneal wound
healing. Antiviral Res 1984; 4:143157.
42. Pepose J, Margolis T, Pavan-Langston D:
Ocular vaccinia: complications and
treatment, Am J Ophthalmol 2003;
136:343347.
43. Pavan-Langston D, Dohlman CH: A
double-blind clinical study of adenosine
arabinoside therapy of viral
keratoconjunctivitis. Am J Ophthalmol
1972; 74:8188.
44. Lee WW, Benitez A, Goodman L, Baker BR:
Potential anticancer agents. XL. Synthesis
of the b-anomer of 9-(D-arabinofuranosyl)adenine. J Am Chem Soc 1960; 82:2648.
45. Reist EJ, Benitez A, Goodman L, et al:
Potential anticancer agents. LXXVI.
Synthesis of purine nucleosides of b-Darabinofuranose. Organic Chem 1962;
27:3274.
46. Parke, Davis, and Company: British Patent
1,159,290 (1969). Chem Abstr 1969;
71:797572.
47. Pavan-Langston D: Use of vidarabine in
ophthalmology: a review. Ann Ophthalmol
1977; 9:835.
48. Schabel FM Jr: The antiviral activity of
9-beta-D-arabinofuranosyladenine
(ARA-A). Chemotherapy 1968; 13:321338.
49. Keeney RE, Buchanan RA: Clinical
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
application of adenine arabinoside. Ann N

Y Acad Sci 1975; 255:185.
Waring GO III, Laibson PR, Satz JE, Joseph
NH: Use of vidarabine in epidemic
keratoconjunctivitis due to adenovirus
types 3, 7, 8 and 19. Am J Ophthalmol
1976; 82:781785.
Pavan-Langston D, Dohlman CH, Geary P,
Sulzewski D: Intraocular penetration of Ara
A and IDU therapeutic implications in
clinical herpetic uveitis. Trans Am Acad
Ophthalmol Otolaryngol 1973;
77:op455466.
Abel R Jr, Kaufman HE, Sugar J:
Intravenous adenine arabinoside against
herpes simplex keratouveitis in humans.
Whitley RJ, Soong SJ, Dolin R, et al:
Adenine arabinoside therapy of biopsyproved herpes simplex encephalitis.
National Institute of Allergy and Infectious
Diseases collaborative antiviral study.
N Engl J Med 1977; 297:289294.
Jones BR: Rational regimen of
administration of antivirals. Trans Am Acad
Ophthalmol Otolaryngol 1975; 79:104.
Pavan-Langston D, Buchanan RA:
Vidarabine therapy of simple and IDUcomplicated herpetic keratitis. Trans Am
Acad Ophthalmol Otolaryngol 1976;
81:op813825.
Van Bijsterveld OP, Post H:
Trifluorothymidine versus adenine
arabinoside in the treatment of herpes
simplex keratitis. Br J Ophthalmol 1980;
64:3339.
Romanowski E, Bartels S, Gordon Y:
Comparative antiviral efcacies of cidofovir,
trifluridine and acyclovir in the HSV-1 rabbit
keratitis model. Invest Ophthalmol Vis Sci
1999; 40:378384.
Jackson WB, Breslin CW, Lorenzetti DW, et
al: Treatment of herpes simplex keratitis:
comparison of acyclovir and vidarabine.
Can J Ophthalmol 1984; 19:107111.
Pavan-Langston D, Campbell R, Lass J:
Acyclic antimetabolite therapy of
experimental herpes simplex keratitis.
Bauer DJ, Collins P, Tucker WE Jr, Macklin
AW: Treatment of experimental herpes
simplex keratitis with acycloguanosine. Br
J Ophthalmol 1979; 63:429436.
Langston RH, Pavan-Langston D, Dohlman
CH: Antiviral medication and corneal
wound healing. Arch Ophthalmol 1974;
92:509513.
ODay DM, Poirier RH, Jones DB,
Elliott JH: Vidarabine therapy of
complicated herpes simplex keratitis. Am J
Ophthalmol 1976; 81:642649.
US Pharmcopeia Acyclovir. Greenwood
City, Co: Thomson Micromedex Inc;
2005:2432.
Morgan KS, Wander AH, Kaufman HE,
et al: Toxicity and tolerance of 9-(2hydroxyethoxymethyl) guanine.
Chemotherapy 1980; 26:405.
DeMiranda P, Blum M: Pharmacokinetics of
acyclovir after intravenous and oral
administration. J Antimicrob Chemother
1983; 12:2937.
Prusoff WH, Ward DC: Nucleoside analogs
with antiviral activity. Biochem Pharmacol
1976; 25:12331239.
Elion G: Mechanism of action and
selectivity of acyclovir. Am J Med 1982;
73(1A):713.
68. Elion GB, Furman PA, Fyfe JA, et al: The

selectivity of action of an antiherpetic
agent, 9-(2-hydroxyethoxymethyl) guanine.
Reproduced from Proc Natl Acad Sci USA
1977; 74:57165720, Rev Med Virol 1999;
9:147152; discussion 152143.
69. Schaeffer H, Beauchamp L, DeMiranda P,
et al: 9-(2-Hydroxyethoxymethyl) guanine
activity against viruses of the herpes group.
Nature 1978; 272:583589.
70. Miller WH, Miller RL: Phosphorylation of
acyclovir (acycloguanosine)
monophosphate by GMP kinase. J Biol
Ophthalmol 1974; 92:509.Chem 1980;
255:72047210.
71. Derse D, Cheng YC, Furman PA, et al:
Inhibition of puried human and herpes
simplex virus-induced DNA polymerase by
9-(2-hydroxyethoxymethyl) guanine
triphosphate. J Biol Chem 1981;
265:1144711451.
72. Furman PA, St Clair MH, Fyfe JA, et al:
Inhibition of herpes simplex virus-induced
DNA polymerase activity and viral DNA
replication by 9-(2-hydroxyethoxymethyl)
guanine and its triphosphate. J Virol 1979;
32:7281.
73. Pavan-Langston D, Lass J, Hettinger M,
Udell I: Acyclovir and vidarabine in the
treatment of ulcerative herpes simplex
92:829835.
74. Poirier RH, Kingham JD, de Miranda P,
Annel M: Intraocular antiviral penetration.
75. Collum LM, Logan P, McAuliffe-Curtin D,
et al: Randomised double-blind trial of
acyclovir (Zovirax) and adenine arabinoside
in herpes simplex amoeboid corneal
ulceration. Br J Ophthalmol 1985; 69:847.
76. Lass JH, Pavan-Langston D, Park NH:
Aciclovir and corneal wound healing. Am J
Ophthalmol 1979; 88:102108.
77. Wilhelmus KR, Coster DJ, Jones BR:
Acyclovir and dbridement in the treatment
of ulcerative herpetic keratitis. Am J
Ophthalmol 1981; 91:323327.
78. La Lau C, Oosterhuis JA, Versteeg J,
et al: Acyclovir and trifluorothymidine in
herpetic keratitis. Preliminary report of a
multicenter trial. Doc Ophthalmol 1981;
50:287.
79. Collum LMT, McGettrick P, Akhtar J, et al:
Oral acyclovir (Zovirax) in herpes simplex
dendritic corneal ulceration. Br J
Ophthalmol 1986; 70:435438.
80. Hung SO, Patterson A, Clark DI, Rees PJ:
Oral acyclovir in the management of
dendritic herpetic corneal ulcerations.
81. Young B, Patterson A, Ravenscroft T:
Double-blind clinical trial of acyclovir and
adenine arabinoside in herpetic corneal
ulceration. Am J Med 1982; 73:311312.
82. Jensen KB, Nissen SH, Jessen F: Acyclovir
in the treatment of herpetic keratitis. Acta
83. Schwartz G, Holland G: Oral acyclovir for
the management of herpes simplex
keratitis in children. Ophthalmol 2000;
107:278282.
84. Schwab I: Oral acyclovir in the
management of herpes simplex ocular
infections. Ophthalmol 1988; 95:423432.
85. de Koning EWJ, van Bijsterveld OP, Cantell
K: Combination therapy for dendritic
keratitis with acyclovir and alpha-interferon.
86. HEDS: A controlled trial of oral acyclovir for

the prevention of stromal keratitis or iritis in
patients with herpes simplex virus epithelial
115:703712.
87. HEDS: Acyclovir for the prevention of
recurrent herpes simplex virus eye disease.
New Engl J Med 1998; 339:300306.
88. Barron B, Shimeld C, Claoue C, et al:
The herpetic eye disease study: a
controlled trial of acyclovir in steroidtreated herpes simplex stromal keratitis.
1994:18711882.
89. Wilhelmus K, Gee L, Hauck W, et al:
Herpetic eye disease study: a controlled
trial of topical cortcosteroids for herpes
simplex stromal keratitis. Ophthalmology
1994; 101:18831896.
90. HEDS: Predictors of recurrent herpes
simplex virus keratitis. Cornea 2001;
20:123128.
91. HEDS: Oral acyclovir for herpes simplex
virus eye disease: effect on prevention of
epithelial keratitis and stromal keratitis.
Herpetic Eye Disease Study Group. Arch
Ophthalmol 2000; 118:10301036.
92. HEDS: A controlled trial of oral acyclovir for
iridocyclitis caused by herpes simplex
virus. Arch Ophthalmol 1996;
114:10651072.
93. Sudesh S, Laibson PR: The impact of the
herpetic eye disease studies on the
management of herpes simplex virus ocular
infections. Curr Opin Ophthalmol 1999;
10:230233.
94. Wu X, Chen X: Acyclovir for the treatment
and prevention of recurrent infectious
herpes simplex keratitis. Chin Med J
(English) 2002; 115:15691572.
95. Simon AL, Pavan-Langston D: Long-term
oral acyclovir therapy. Effect on recurrent
infectious herpes simplex keratitis in
patients with and without grafts.
Ophthalmology 1996; 103:13991404;
discussion 14041405.
96. Cunningham EJ, Margolis T: Ocular
manifestations of HIV infection. N Engl J
Med 2000; 339:236244.
97. Herbort CP, Buchi ER, Piguet B, et al: Highdose oral acyclovir in acute herpes zoster
ophthalmicus: the end of the corticosteroid
era. Curr Eye Res 1991;
10(Suppl):171175.
98. Dunn J, Holland G: Human
immunodeciency virus infection and AIDS.
Tabbara K, Hyndiuk R, eds. Boston, MA:
Little Brown Co.; 1996:625644.
99. Tambasco F, Cohen E, Nguyen L, et al: Oral
acyclovir after penetrating keratoplasty for
herpes simplex keratitis. Arch Ophthalmol
1999; 117:445449.
100. Barney N, Foster C: A prospective
randomized trial of oral acyclovir following
penetrating keratoplasty for herpes. Cornea
1994; 13:232236.
101. Balfour HJ: Antiviral drugs. New Eng J Med
1999; 340:12551268.
102. Teich SA, Castle J, Friedman AH, et al:
Active cytomegalovirus particles in the
eyes of an AIDS patient being treated with
9-[2-hydroxy-1-(hydroxymethyl)
ethoxymethyl] guanine (ganciclovir). Br J
Ophthalmol 1988; 72:293298.
103. Hirsch MS, Schooley RT: Resistance to
antiviral drugs: the end of innocence.
N Engl J Med 1989; 320:313314.
104. Jacobson MA, Berger TG, Fikrig S, et al:
Acyclovir-resistant varicella zoster virus
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
115.
116.
117.
118.
119.
120.
121.
122.
infection after chronic oral acyclovir therapy

in patients with the acquired
immunodeciency syndrome (AIDS).
Ann Int Med 1990; 112:187191.
Erlich K, Mills J, Chatis P, et al: Acyclovirresistant herpes simplex virus infections in
patients with the acquired
immunodeciency syndrome. New Engl J
Med 1989; 320:293296.
Sonkin P, Baratz K, Frothingham R,
Cobo L: Acyclovir-resistant herpes simplex
virus keratouveitis after penetrating
keratoplasty. Ophthalmology 1992;
99:18051808.
Coen DM, Schaffer PA: Antiherpesvirus
drugs: a promising spectrum of new drugs
and drug targets. Nat Rev Drug Discov
2003; 2:278288.
Stranska R, Schuurman R, Nienhuis E,
et al: Survey of acyclovir-resistant herpes
simplex virus in the Netherlands:
prevalence and characterization. J Clin
Virol 2005; 32:718.
Marsh R, Cooper M: Double masked trial of
topical acyclovir and steroids in the
treatment of herpes zoster ocular
inflammation. Br J Ophthalmol 1991;
75:542546.
Zaal MJ, Maudgal PC, Rietveld E, Suir EP:
Chronic ocular zoster. Curr Eye Res 1991;
10(Suppl):125130.
Cobo L, Foulks G, Liesegang T, et al: Oral
acyclovir in the treatment of acute herpes
zoster ophthalmicus. Ophthalmol 1986;
93:763.
acyclovir in the therapy of acute herpes
zoster ophthalmicus: an interim report.
Ophthalmol 1985; 92:15741583.
Borruat F, Buechi E, Piguet F, et al:
Prevention of ocular complications of
herpes zoster ophthalmicus by adequate
treatment with acyclovir. Klin Monatsbl
Augenheilkd 1991; 198:358360.
Harding S, Porter S: Oral acyclovir in
herpes zoster ophthalmicus. Curr Eye Res
1991; 10:177182.
Hoang-Xuan T, Buchi ER, Herbort CP,
et al: Oral acyclovir for herpes zoster
ophthalmicus. Ophthalmology 1992;
99:10621071.
Aylward G, Claoue C, Marsh R,
Yasseem N: Influence of oral acyclovir on
ocular complications of herpes zoster
ophthalmicus. Eye 1994; 8:7074.
Pepose J: The potential impact of varicella
vaccine and new antivirals on ocular
disease related to varicella-zoster virus. Am
J Ophthalmol 1997; 123:243249.
Manka R: Exogenous lactase in the
treatment of oral acyclovir intolerance. Am
J Ophthalmol 1989; 108:733739.
Pulido JS, Palacio M, Peyman GA, et al:
Toxicity of intravitreal antiviral drugs.
Ophthalmic Surg 1984; 15:666669.
Small GH, Peyman GA, Srinivasan A,
et al: Retinal toxicity of combination
antiviral drugs in an animal model. Can J
Peyman GA, Goldberg MF, Uninsky E,
et al: Vitrectomy and intravitreal antiviral
drug therapy in acute retinal necrosis
syndrome: report of two cases. Arch
Ophthalmol 1984; 102:16181621.
Burnette TC, de Miranda P: Metabolic
disposition of the acyclovir prodrug
valaciclovir in the rat. Drug Metab Dispos
1994; 22:6064.
CHAPTER 20
Antivirals
227
SECTION 4
228
123. Weller S, Blum M, Smiley M: Phase I

pharmacokinetics of the acyclovir prodrug,
valacyclovir. Antivir Res 1993; 20(Suppl
1):144148.
124. Dias C, Nashed Y, Atluri H, Mitra A: Ocular
penetration of acyclovir and its peptide
prodrugs valacyclovir and val-valacyclovir
following systemic administration in
rabbits: an evaluation using ocular
microdialysis and LC-MS. Curr Eye Res
2002; 25:243252.
125. Colin J, Prisant O, Cochener B, et al:
A double blind randomized trial to compare
the efcacy and safety of valaciclovir and
acyclovir for treatment of herpes zoster
107:15071511.
126. Beutner K, Friedman D, Forszpaniak C,
et al: Valaciclovir compared with acyclovir
for improved therapy for herpes zoster
immunocompetent adults. Antimicrob
127. Fiddian A: Antiviral drugs in development
for herpes zoster. Scand J Infect Dis Suppl
1996; 100:5154.
128. Herne K, Cirelli R, PL, et al: Antiviral
therapy of acute herpes zoster in older
patients. Drugs Aging 1996, 8:97112.
129. Tyring S, Beutner K, Tucker B, et al:
Antiviral therapy for herpes zoster:
randomized controlled clinical trial of
valacyclovir and famciclovir therapy in
immunocompetent patients 50 years and
older. Arch Fam Med 2000; 9:863869.
130. Liesegang T: Varicella-zoster virus eye
disease. Review. Cornea 1999; 18:511531.
131. Gnann JW Jr, Whitley RJ: Clinical practice.
Herpes zoster. N Engl J Med 2002;
347:340346.
132. Sozen E, Avunduk AM, Akyol N:
Comparison of efcacy of oral valacyclovir
and topical acyclovir in the treatment of
herpes simplex keratitis: a randomized
clinical trial. Chemotherapy 2006;
52:2931.
Comparison of the efcacy and safety of
valaciclovir and acyclovir for the treatment
of herpes zoster ophthalmicus.
Ophthalmology 2000; 107:15071511.
134. Lebrun-Vignes B: Valaciclovir. Ann
Dermatol Venereol 2002; 129:708715.
135. US Pharmcopeia Famvir. Greenwood
Village, CO: Thomson Micromedex Inc;
2005:13901392.
136. Boyd M, Safrin S, Kern E: Penciclovir: a
review of spectrum of activity, selectivity,
and cross-resistance pattern. Antiviral
Chem Chemother 1993; 4(Suppl 1):2536.
137. Vere Hodge R, Cheng Y-C: The mode of
action of penciclovir. Antiviral Agents
Chemother 1993; 4(Suppl 1):1324.
138. Perry CM, Wagstaff AJ: Famciclovir. A
review of its pharmacological properties
and therapeutic efcacy in herpesvirus
infections. Drugs 1995; 50:396415.
139. Goldani LZ: Treatment of severe infectious
mononucleosis with famciclovir. J Infect
2002; 44:9293.
140. deGreef H: Famciclovir, a new oral
antiherpes drug: results of the rst
controlled clincial study demontrating its
efcacy and safety in the treatment of
uncomplicated herpes zoster in
immunocompetent patients. Int J
Antimicrob Agents 1994; 4:241246.
141. deGreef H: Famciclovir, a new oral antiviral
drug: its efcacy and safety in the
142.
143.
144.
145.
146.
147.
148.
149.
150.
151.
152.
153.
154.
155.
156.
157.
treatment of uncomplicated herpes zoster

in immunocompetent patients. Int J
Gheeraert P, Group, TFHZCS: Efcacy and
safety of famciclovir in the treatment of
uncomplicated herpes zoster. Program and
abstracts. 32nd ICAAC. Washington DC:
American Society for Microbiology; 1992:
abstract 1108.
Saltzman R, Jurewicz R, Boon R: Safety of
famciclovir in patients with herpes zoster
and genital herpes. Antimicrob Agents
Chemother 1994; 24542457.
Tyring S, Engst R, Corriveau C, et al:
Famciclovir for ophthalmic zoster: a
randomised aciclovir controlled study.
Tyring, SK, Diaz-Mitoma, F, Shafran, SD,
et al.: Oral famciclovir for the suppression
of recurrent genital herpes: the combined
data from two randomized controlled trials.
J Cutan Med Surg 2003; 7:449454.
Fillet AM: Prophylaxis of herpesvirus
infections in immunocompetent and
immunocompromised older patients. Drugs
Aging 2002; 19:343354.
Oxman MN, Levin MJ, Johnson GR, et al:
A vaccine to prevent herpes zoster and
postherpetic neuralgia in older adults.
N Engl J Med 2005; 352:22712284.
Masur H, Lane H, Palestine A, et al: Effect
of 9-(1,3-dihydroxy-2-propoxymethyl)
guanine on serious cytomegalovirus
disease in eight immunosuppressed
homosexual men. Ann Int Med 1986;
104:4150.
Matthews T, Boehme, R: Antiviral activity
and mechanism of action of ganciclovir.
Rev Infect Dis 1988; 10:S490S494.
Tocci MJ, Livelli TJ, Perry HC, et al: Effects
of the nucleoside analog 2-nor-2deoxyguanosine on human
cytomegalovirus replication. Antimicrob
US Pharmcopeia Ganciclovir. Greenwood
2005:15091515.
Pepose JS, Newman C, Bach MC, et al:
Pathologic features of cytomegalovirus
retinopathy after treatment with the antiviral
agent ganciclovir. Ophthalmology 1987;
94:414424.
Teich SA, Castle J, Friedman AH, et al:
Active cytomegalovirus particles in the
eyes of an AIDS patient being treated with
9-[2-hydroxy-1-(hydroxymethyl)
ethoxymethyl] guanine (ganciclovir).
Shepp DH, Dandliker PS, de Miranda P, et
al: Activity of 9-[2-hydroxy-1(hydroxymethyl) ethoxymethyl] guanine in
the treatment of cytomegalovirus
pneumonia. Ann Intern Med 1985;
103:368373.
MacArthur RB: Ganciclovir: Approved and
investigational uses for the treatment of
cytomegalovirus retinitis. Mt Sinai J Med
1990; 57:378381.
Felsenstein D, DAmico DJ, Hirsch MS,
et al: Treatment of cytomegalovirus retinitis
with 9-[2-hydroxy-1-(hydroxymethyl)ethoxymethyl] guanine. Ann Intern Med
1985; 103:377380.
DAmico DJ, Talamo JH, Felsenstein D,
et al: Ophthalmoscopic and histologic
ndings in cytomegalovirus retinitis treated
with BW-B759U. Arch Ophthalmol 1986;
104:17881793.
158. MacDonald EA: Treatment of

cytomegalovirus retinitis in a patient with
AIDS with 9-(1,3-dihydroxy-2propoxymethyl) guanine. Can J Ophthalmol
1987; 22:4852.
159. Holland GN, Sakamoto MJ, Hardy D, et al:
Treatment of cytomegalovirus retinopathy
in patients with acquired immunodeciency
syndrome. Arch Ophthalmol 1986; 104:1794.
160. Holland GN, Sidikaro Y, Kreiger AE, et al:
Treatment of cytomegalovirus retinopathy
with ganciclovir. Ophthalmology 1987;
94:815823.
161. Jabs DA, Newman C, Bustros SD, Polk BF:
Treatment of cytomegalovirus retinitis with
ganciclovir. Ophthalmology 1987;
94:824833.
162. Orellana J, Teich SA, Winterkorn JS, et al:
ganciclovir (9-[2-hydroxy-1-(hydroxymethyl)
ethoxymethyl) guanine (BW B759U)]. Br J
Ophthalmol 1988; 72:525529.
163. Kahraman G, Krepler K, Franz C, et al:
Seven years of HAART impact on
ophthalmic management of HIV-infected
patients. Ocul Immunol Inflamm 2005;
13:213218
164. Henry K, Cantrill H, Fletcher C, et al:
Use of intravitreal ganciclovir (dihydroxy
propoxymethyl guanine) for
cytomegalovirus retinitis in patients with
AIDS. Am J Ophthalmol 1987; 103:1722.
165. Ussery FM, Gibson SR, Conklin RH, et al:
Intravitreal ganciclovir in the treatment of
AIDS-associated cytomegalovirus retinitis.
166. Heery S, Hollows F: High-dose intravitreal
ganciclovir for cytomegaloviral (CMV)
retinitis. Aust N Z J Ophthalmol 1989;
17:405408.
167. Harris ML, Mathalone MBR: Intravitreal
ganciclovir in CMV retinitis: case report.
168. Cantrill HL, Henry K, Melroe NH, et al:
intravitreal ganciclovir: long-term results.
169. Heinemann MH: Long-term intravitreal
ganciclovir therapy for cytomegalovirus
retinopathy. Arch Ophthalmol 1989;
107:17671772.
170. Palestine AG: Intraocular therapy for
cytomegalovirus retinitis. J Int Assoc
Physicians AIDS Care 1996; 2:2528.
171. Colin J, Bing Hoh H, Easty D, et al:
Ganciclovir ophthalmic gel (Virgan; 0.15%)
in the treatment of herpes simplex keratitis.
Cornea 1997; 16:393399.
172. Valganciclovir: new preparation. CMV
retinitis: a simpler, oral treatment. Prescribe
Int 2003; 12:133135.
173. Chang M, Dunn JP: Ganciclovir implant in
the treatment of cytomegalovirus retinitis.
Expert Rev Med Devices 2005; 2:421427.
174. Martin DF, Parks DJ, Mellow SD, et al:
an intraocular sustained-release ganciclovir
implant. Arch Ophthalmol 1994; 112:1531.
175. Sanborn GE, Anand R, Torti R, et al:
Sustained release ganciclovir therapy for
treatment of cytomegalovirus retinitis: use
of an intravitreal device. Arch Ophthalmol
1992; 110:188.
176. Schwartz DM: New therapies for
cytomegalovirus retinitis. In: Smolin G, ed.
New drugs in ophthalmology. Philadelphia,
PA: Lippincott, Williams, Wilkins; Int
Ophthalmol Clin 1996; 36:19.
177. US Pharmcopeia Foscarnet. Greenwood

2005:14921496.
178. Safrin S, Assaykeen T, Follansbee S, et al:
Foscarnet therapy for acyclovir-resistant
mucocutaneous herpes simplex virus
infection in 26 AIDS patients. J Infect Dis
1990; 161:10781084.
179. Safrin S, Berger TG, Gilson I, et al:
Foscarnet therapy in ve patients with
AIDS and acyclovir-resistant varicellazoster virus infection. Ann Intern Med 1991;
115:1921.
180. Youle MM, Hawkins DA, Collins P, et al:
Acyclovir-resistant herpes in AIDS treated
with foscarnet. Lancet 1988; 2:341342.
181. Rickman LS, Freeman WR: Medical and
virological aspects of ocular HIV infection
for the ophthalmologist. Semin Ophthalmol
1995; 10:91.
182. Weinberg DV, Murphy R, Naughton K:
Combined daily therapy with intravenous
ganciclovir and foscarnet for patients with
recurrent cytomegalovirus retinitis. Am J
183. Palestine AG, Polis MA, DeSmet MD, et al:
A randomized controlled trial of foscarnet
in the treatment of cytomegalovirus retinitis
in patients with AIDS. Ann Intern Med
1991; 115:665.
184. Studies of Ocular Complications of AIDS
Research Group in Collaboration with the
AIDS Clinical Trial Group: Foscarnetganciclovir cytomegalovirus retinitis trial: 4.
Visual outcomes. Ophthalmology 1994;
101:1250.
185. Jacobson MA, ODonnell JJ, Mills J:
Foscarnet treatment of cytomegalovirus
retinitis in patients with acquired
immunodeciency syndrome. Antimicrob
186. Studies of Ocular Complications of AIDS
Research Group, in Collaboration with
the AIDS Clinical Trial Group: Mortality in
patients with the acquired
immunodeciency syndrome treated with
either foscarnet or ganciclovir for
cytomegalovirus retinitis. N Engl J Med
1992; 326:213.
187. LeHoang P, Girard B, Robinet M, et al:

Foscarnet in the treatment of
cytomegalovirus retinitis in acquired
immune deciency syndrome.
188. Colucciello M: Phosphonoformate
(foscarnet) for CMV retinitis in AIDS. Am J
189. Friedberg DN: Cytomegaloving retinitis. In:
Stenson SM, Friedberg DN, eds. Aids and
the eye. New Orleans: Contact Lens
Association of Ophthalmologists;
1995:6583.
190. Ahmed I, Everett A: Medical management
of cytomegalovirus retinitis. In: Everett A,
Ahmed I, eds. AIDS and ophthalmology:
new solutions. Ophthalmol Clin North Am
1997; 10:15.
191. Holland GN, Levinson RD, Jacobson MA,
et al: Dose-related differences in
progression rates of cytomegalovirus
retinopathy during foscarnet maintenance
therapy. Am J Ophthalmol 1995; 119:576.
192. Jacobsen MA, Causey D, Polsky B, et al: A
dose-ranging study for daily maintenance
intravenous foscarnet therapy for
cytomegalovirus retinitis in AIDS. J Infect
Dis 1993; 168:444.
193. Diaz-Llopis M, Chipont E, Sanchez S, et al:
Intravitreal foscarnet for cytomegalovirus
retinitis in a patient with acquired
immunodeciency syndrome. Am J
194. US Pharmacopeia Cidofovir. Greenwood
City, Co: Thomson Micromedex Inc;
2004:818820.
195. De Clercq E: Cidofovir in the therapy and
short-term prophylaxis of poxvirus
infections. Trends Pharmacol Sci 2002;
23:456458.
196. Hackethal U, Holzapfel C, Gerding H,
Husstedt IW: Ocular complications of the
virostatic drug cidofovir. Klin Monatsbl
Augenheilkd 2003; 220:391395.
197. Kirsch LS, Arevalo JF, Chavez de la Paz E,
et al: Intravitreal cidofovir (HPMPC)
treatment cytomegalovirus retinitis in
patients with acquired immune deciency
syndrome. Ophthalmology 1995; 102:533.
198. Gordon Y, Romanowski E, Araullo-Cruz T:

HPMPC, a broad-spectrum topical antiviral
agent, inhibits herpes simplex virus type 1
replication and promotes healing of
dendritic keratitis in the rabbit ocular
model. Cornea 1994; 13:516520.
199. Gordon RY, Romanowski E, Araullo-Cruz
TP: Prolonged recovery of desiccated
adenoviral serotypes 5, 8, and 19 from
plastic and metal surfaces in vitro.
Ophthalmology 1993; 100:18351839.
200. Snoeck R, Andrei G, Gerard M, et al:
Successful treatment of progressive
mucocutaneous infection due to acyclovirand foscarnet-resistant herpes simplex
virus with (S)-1-(3-hydroxy-2phosphonylmehtoxypropyl)cytosine
(HPMPC). Clin Infect Dis 1994; 18:570578.
201. Hsu WM, Chiou SH, Chen SS, et al: The
HIV RNA Levels of Plasma and Ocular
Fluids in AIDS Patients with Ophthalmic
Infections. Ophthalmologica 2004;
218:328332.
202. Fillet AM: Prophylaxis of herpesvirus
infections in immunocompetent and
immunocompromised older patients. Drugs
Aging 2002; 19:343354.
203. DeClercq E: (E)-5-(2-bromovinyl)-2deoxyuridine (BVDU). Med Res Rev 2005;
25:120.
204. Wassilew S: Brivudin compared with
famciclovir in the treatment of herpes
zoster: effects in acute disease and chronic
pain in immunocompetent patients. A
randomized, double-blind, multinational
study. J Eur Acad Dermatol Venereol 2005;
19:4755.
205. Wassilew SW, Wutzler P: Oral brivudin in
comparison with acyclovir for herpes
zoster: a survey study on postherpetic
neuralgia. Antiviral Res 2003; 59:5760.
206. Gross G, Schofer H, Wassilew S, et al:
Herpes zoster guideline of the German
Dermatology Society (DDG). J Clin Virol
2003; 26:277289; discussion 291273.
CHAPTER 20
Antivirals
229
CHAPTER
21
Antifungal Agents
Eduardo C. Alfonso, Jorge Cantu-Dibildox, Terrence OBrien, and Darlene Miller
The choice of an antifungal agent in ophthalmology depends

on several variables, including the primary site of infection, the
route of administration, the organism involved, and the sensitivity data available.15 The major classes of antifungals used in
ophthalmology are polyenes, imidazoles, and pyrimidines
(Table 21.1).6 Other compounds have been tried as antifungals,
but the clinical experience is very limited.7,8 These include rose
bengal, salicylic acid, benzoic acid, thimerosal, gentian violet,
silver nitrate, zinc, copper sulfate, boric acid, potassium, iodide,
and iodine. A great number of experimental compounds are
described in the literature.912 For most of these, sufcient data
on the treatment of human mycoses are lacking.1316
esterols in the fungal cell wall, forming blisters and causing

lysis of the cell. This action is not concentration dependent.
The larger molecules, such as amphotericin, work by creating
pores in the cell wall, allowing small ions such as potassium
to leak out and causing imbalances in the osmotic gradient and
eventual cell lysis. This mechanism of action is concentration
dependent and may be altered by changes in the osmotic environment.22 Other factors have been implicated in the interaction of the polyenes with cell membranes.23 The most widely
used of the polyenes are amphotericin B and natamycin.24
POLYENE ANTIBIOTICS
Amphotericin B is most commonly used in ophthalmology as

a topical preparation for keratitis and scleritis, intraocularly for
endophthalmitis, and systemically for these conditions and
for scleritis, dacryocystitis, and cellulitis.2529 The spectrum of
organisms and in vitro sensitivities identied in the published
literature and in our laboratory is presented in Tables 21.2 and
21.3, respectively.3033 Dosages for antifungal agents are given
in Table 21.4.
For the treatment of keratitis and scleritis, a topical concentration of 2.510 mg/mL given every 3060 min for the rst
4872 h appears to deliver the optimal dose.34,35 Higher concentrations may cause surface toxicity.36,37 This concentration is
achieved by mixing the powdered amphotericin with sterile
Polyene antibiotics are produced from a Streptomyces species.17,18

Their chemical conguration gives them their basic classication based on the number of double bonds as well as the number
of carbon atoms (group I <30 atoms; group II >30 atoms).19
They interact with cell membrane sterols, primarily ergosterol,
which causes increased permeability that leads to cell lysis.20 It
is the binding to mammalian cell membrane cholesterol that
accounts for their toxicity. Two mechanisms of action of the
polyene antibiotics are known and depend on the size of the
antifungal molecule.21 Small molecules such as natamycin
work by an all-or-none mechanism of action. They bind to the
AMPHOTERICIN B
TABLE 21.1. Classification of Antifungals

Polyenes
Imidazoles
Triazoles
Pyrimidines
*
Pradicimicins||
Amphotericin B
Clotrimazole
Fluconazole
Amphotericin B methyl ester
Miconazole
Itraconazole*
Cispentacin||
Natamycin
Econazole
Terconazole
Jasplakinolide||
Flucytosine
Others
Ketoconazole
Vibunazole
Terbinafine||
Thiabendazole*
Alteconazole*
Nystatin
Bifonazole
Voriconazole*
Caspofungin
Butoconazole
Croconazole
Posaconazole
Ravuconazole*
Fenticonazole
*Oral.
Intravenous.
Ocular.
Dermatologic.
||
Not available.
231
TABLE 21.2. Antimicrobial Activity of Antifungal Agents Based on Published Reports

Antifungal
Agent
Alternaria
Aspergillus
Candida
Cephalosporium
Cladosporium
Curvularia
Fusarium
Paecilomyces
Penicillium
Polyenes
Amphotericin
Nystatin
Natamycin
S
S
I
I
Imidazoles
Clotrimazole
Miconazole
Econazole
Ketoconazole
Triazoles
Itraconazole
Fluconazole
S
S
Pyrimidines
Flucytosine
SECTION 4
Abbreviations: S, susceptible; I, variable susceptibility; R, resistant.
TABLE 21.3. Ten-Year Summary of Sensitivity Testing of

Clinical Isolates at the Microbiology Laboratory of the Bascom
Palmer Eye Institute*
Antifungal
Fusarium
(n = 40)
Candida
(n = 10)
Aspergillus Curvularia
(n = 15)
(n = 6)
Range
0.0785.0
0.085.0
0.012.5
0.040.31
Mean
1.2 (S)
2.7 (S)
1 (S)
0.16 (S)
Range
0.155.0
0.315.0
0.6225.0
0.622.50
Mean
1.5 (S)
2.5 (S)
2 (S)
1.4 (S)
Range
0.7850.0
0.101.6
0.78250
0.2012.50
Mean
10.9 (I)
0.71 (S)
4 (S)
2.7 (S)
Range
0.7850.0
0.7862.0
0.203.10
0.053.1
Mean
14.21 (I)
2 (S)
1.2 (S)
1.3 (S)
Range
0.05100.0
0.053.10
25100
Mean
921 (R)
1.2 (S)
68 (R)
Amphotericin
Natamycin
Ketoconazole
Miconazole
Flucytosine
Abbreviations: S, susceptible; I, variable susceptibility; R, resistant.

*Ranges and means in micrograms per milliliter.
232
water.38 The mixture should be stored in a dark bottle and

refrigerated to maintain drug stability. Subconjunctival injection
of amphotericin is not recommended because of severe
toxicity.7,39
For endophthalmitis, intravitreal injection of 5 g of
amphotericin in 0.1 mL appears to be safe and effective in
humans.1,4044 Concurrent surgical management of the vitreous

is often necessary to control the infection.45,46
For intravenous use, a test dose of 1 mg of amphotericin in
150 mL of 5% dextrose in water is given.47,48 Once this test
dose is tolerated, 15 mg is given over 46 h. The dose is
increased by 5 mg daily until the desired dose of 0.51 mg kg1
day1 is reached. If chills, fever, nausea, or hypertension develops
with the test dose, the patient may require concomitant use
of 2530 mg of hydrocortisone intravenously.49 Also, aspirin,
diphenhydramine, or prochlorperazine may be required. Other
potential side effects are a decrease in the glomerular ltration
rate to 2060% of normal, which may be restored to normal after
cessation of therapy for ~5 days.50 Hypokalemia may require
potassium supplements. A drop in the platelet count and hematocrit may also be observed during therapy. Hepatic damage
occurs rarely. The water-soluble semisynthetic methyl ester
derivative of amphotericin B has been shown in animal models
to carry fewer side effects than the parent compound.5153
NATAMYCIN
Natamycin (pimaricin) is a small semisynthetic tetraene and is
considered the least toxic, the least irritating, and the most
stable of the polyenes.23 It has been available for topical use
as a 5% suspension since its approval by the US Food and
Drug Administration in the late 1970s.54,55 It has a broad
spectrum of sensitivities, especially to Fusarium species, as
shown in Table 21.3.56,57 It has decreased penetration through
an intact epithelium, and surface debridement may be desirable during therapy,58,59 although experiments have shown that
its penetration on intact epithelium is greater than amphotericin B.60 Since natamycin is used as a suspension, it can
dry on the ocular surface and cause irritation.56 Lavage with a
saline solution of the lid margins is often necessary. Natamycin
can be toxic to the corneal and conjunctival epithelium, causing
hyperemia and epithelial defects.59 As with amphotericin,
topical therapy is given every 3060 min for the rst 4872 h,
and treatment is usually continued on a tapering fashion for
36 weeks depending on the activity of the keratitis.60
Antifungal Agents
TABLE 21.4. Antifungal Dosages

Antifungal Agent
Topical
Subconjunctival
Intravitreal
Intravenous
Oral
Amphotericin B
2.510.9 mg/mL
750 mg/mL every

other day
510 mg
Maintenance dose
1 mg kg1 day1
refrigerated
Clotrimazole
1% Suspension
1% Solution
510 mg
(0.51 mL)
Econazole
1% Suspension
1% Ointment
Fluconazole
2% Suspension
1% Solution
400 mg/day initial dose

200 mg/day maintenance dose
Itraconazole
2% Suspension
200 mg/day
Ketoconazole
1% Suspension
200400 mg/day
Miconazole
1% Suspension
60150 mg kg1 day1

(adults)
30 mg kg1 day1
510 mg
(0.51 mL)
0.25 mg
200 mg t.i.d.
6003600 mg/day
divided into three doses
1% Solution
(10 mg/mL)
1% Solution (not
available commercially)
Natamycin
50 mg/mL
Nystatin
Ointment 100 000 U/g
Thiabendazole
4% Suspension
Flucytosine
10 mg/mL
Caspofungin
Subconjunctival and intravitreal administration are not

recommended because of signicant toxicity.61,62 Systemic
intravenous use of natamycin does not render signicant levels
in the eye, and oral preparations are not well absorbed.63,64
However, natamycin is considered to be the mainstay of topical
therapy for most fungal keratitis.64a
NYSTATIN
Nystatin has been studied experimentally in ophthalmology,
and cases have been reported in which it has been used in
external ocular infections caused by Candida.40,65 It has been
used as a dermatologic ointment, which has a concentration
of 100 000 U/g, and at a frequency of application every 46 h.
Subconjunctival injections show marked toxicity, and experimental intravitreal injection of 0.1 mL of a concentration of
2000 U/mL did not cause a signicant reaction and cured an
experimental case of Aspergillus endophthalmitis.18,66
AZOLES
IMIDAZOLES
The imidazoles possess a broad spectrum of antifungal activity,
but in contrast to the polyenes, they are relatively resistant to
light, hydrolysis, and pH changes and are soluble in organic
substances.67 A number of compounds are available as approved
preparations for systemic use.
The imidazoles have a combination of mechanisms for
antimycotic activity.6870 At low concentrations, miconazole,
econazole, and ketoconazole affect the formation of ergosterol
needed by the cell membranes.71 At high concentrations, clotri-
3 mg kg1 h1,
over 12 h IV
200 mg/12 h oral
25 mg kg1 day1
1
50150 mg kg day
at 6-h intervals
70 mg/day 1, followed
by 50 mg/day IV
CHAPTER 21
Voriconazole
mazole and miconazole can disrupt lysosomes, causing direct

cell membrane damage. In addition, most imidazoles inhibit
catalase and cytochrome C peroxidase intracellulary, causing
accumulation of hydrogen peroxide and leading to cell death.
There also appears to be a triggering mechanism of host defense
cells by the imidazoles. When ketoconazole is added in vitro to
polymorphonuclear leukocytes and macrophages, it has the
ability to eradicate both the yeast and the mycelial forms of
Candida, in the absence of polymorphonuclear leukocytes and
macrophages.72 One can see that because of these combined
mechanisms of action, most of the imidazoles can be fungistatic
and fungicidal.73,74
CLOTRIMAZOLE
Clotrimazole has a wide spectrum of activity against numerous
fungi, but poor results have been obtained with Fusarium.
Most strains are inhibited at concentrations of 24 mg/mL,
which can be readily achieved with topical and oral administration (see Table 21.3).75,76 It is poorly absorbed parenterally.77
The topical preparation of clotrimazole is made by dilution
in arachis oil to a 1% solution. It has been applied hourly for
23 days, then tapered over 812 weeks.78 Oral administration
in a dosage range of 60150 mg kg1 day1 can be given with an
achievable serum concentration of 0.45.5 mg/mL. No commercial oral dosage forms are available in the United States.
Clotrimazole has been recommended by several authors as
the drug of choice for Aspergillus infections of the eye.7880 Side
effects of the systemic administration of clotrimazole may
include anorexia, nausea, hallucinations, confusion, and
epigastric pain. It should not be given in the rst 3 months of
pregnancy or to patients with a history of hypersensitivity,
233

adrenal, or liver problems. Liver enzyme level elevations are
normal with the use of clotrimazole, and these tend to return to
normal once the drug is withdrawn.81
SECTION 4
MICONAZOLE
Miconazole is a phenethylimidazole that is very stable in
solution.82 Its mechanism of action is similar to that of the
other imidazoles.70 It has a broad spectrum of activity against
Cryptococcus, Aspergillus, Curvularia, Candida, Microsporum,
Paecilomyces, and Trichophyton (see Table 21.3).8386
Miconazole may be given intravenously in dosages ranging
from 200 to 3600 mg/day in three divided doses. In children,
a dose of 15 mg/kg per infusion should not be exceeded.82 It
may also be used as a topical, subconjunctival, or intravitreal
preparation.87 For topical use, a 1% solution in arachis oil or
a 10 mg/mL commercial solution (Monistat IV) is well tolerated. It is also available as a 2% dermatologic ointment, but
this may cause some irritation to the eye.88 For subconjunctival
injections, 10 mg/day may be used. For intravitreal injections,
0.250.50 mg may be used.86,89,90
After intravenous administration of miconazole, reported
side effects may be a rash with pruritis, chills, nausea, and
vomiting. These side effects may be minimized by the concomitant administration of antihistamines and antiemetics.91,92
Reports also mention a possible decrease in sodium levels and
the hematocrit, with aggregation of erythrocytes and thrombocytosis.85 Topical use of miconazole may cause surface toxicity
after prolonged use.90,93,94
KETOCONAZOLE
Ketoconazole is a synthetic acetylchichlorophenyl imidazole.
It dissolves in water with a resultant pH of ~3.95 Its mechanism
of action is similar to that of the other imidazoles.68,96 This
drug has a broad spectrum of activity in vitro (see Table 21.3).97
Ketoconazole is available for oral administration. It is well
absorbed from the gastrointestinal tract and bound to albumin,
and high therapeutic blood levels are maintained.68 Ninety
percent of the drug is excreted by the liver and the remainder
by the kidneys.95 Ketoconazole is available in 200-mg tablets
with a recommended daily dose of 200400 mg. A topical preparation may be formulated in a 15% concentration by dissolving in arachis oil.98,99 Ketoconazole may also be dissolved in
polyethooxylated castor oil67 or in 4.5% boric acid.7,100
Systemic side effects associated with the use of ketoconazole
have been minor and usually reversible. Pruritus, nausea,
vomiting, diarrhea, cramps, gynecomastia,101 and elevations in
liver enzyme levels have been reported after oral administration.101 Topical use of ketoconazole shows minimal reversible
toxicity in animals.102 Ketoconazole can affect the efcacy and
concentration of cyclosporine, warfarin, phenytoin, and
theophylline.103
In ophthalmology, topical ketoconazole has been used clinically and experimentally for the treatment of keratitis.99,104,105
Oral ketoconazole has been used in both experimental35 and
human keratitis.106 In experimental endophthalmitis, ketoconazole was effective if started 24 h after injection.107 It has
been suggested that oral ketoconazole may augment topical
natamycin therapy.25,108
THIABENDAZOLE
234
Thiabendazole is a thiazolyl benzimidazole. Its primary clinical

use for many years has been in the treatment of roundworm
infections.108 Its mechanism of action is similar to that of the
other imidazoles.68 It has been shown to be active against ocular
isolates of fungi, but poor results have been obtained against

Candida and Aspergillus species (see Table 21.3).85,99
Oral thiabendazole may be given at a dose of 25 mg/kg two
times per day with a maximal daily dose of 3 g. Its peak serum
concentration is in 12 h, and 90% is excreted in the urine.68
Topical application of a 4% thiabendazole suspension has been
reported in the treatment of Aspergillus avus keratitis.109 Side
effects have been few, the major ocular side effects being surface
irritation and dryness and mild reversible hepatic disease.18
Clinical experience with thiabendazole in ophthalmology is
limited, and this drug has been reserved for cases unresponsive
to conventional treatment.110
ECONAZOLE
Econazole is a deschlorophenethylimidazole.23 Its mechanism
of action is similar to that of the other imidazoles.68 The
spectrum of activity is similar to that of the other imidazoles,
with increased activity against Aspergillus, Fusarium, and
Penicillium. It has less activity against Candida.111
Econazole is available as a dermatologic ointment. For topical
use, a 1% suspension may be prepared in arachis oil.112 For oral
use, 200 mg of econazole three times a day may be used. For
intravenous use, 30 mg kg1 day1 is recommended.112 The systemic preparation is not commercially available in the United
States.
The clinical use of econazole in ophthalmology is very
limited,112 although some studies suggest that it could be as
effective as natamycin for a broad spectrum of fungal
keratitis.113 However, there appears to be no synergism between
concurrent use of econazole and natamycin as topical treatments for fungal keratitis.114
TRIAZOLES
The triazoles uconazole, itraconazole, terconazole, and
others (see Table 21.1) were developed in order to increase the
spectrum of activity and reduce the side effects of their
predecessors, the imidazoles.
FLUCONAZOLE
Fluconazole is perhaps the most widely used member of the
triazoles because of in vitro studies that have shown a very
wide spectrum of activity against many pathogens.115 The
in vivo activity has not followed its laboratory spectrum of activity. It has been used for the treatment of Candida species.116
Unlike amphotericin B, uconazole is capable of penetrating
intact corneal epithelium, due to its lower molecular weight.117
It has also been used for the treatment of experimental endophthalmitis in its oral form16 and in the treatment of experimental
Candida albicans keratitis in a topical solution.16,118 Animal
studies suggest efcacy in both topical and oral form against
Aspergillus fumigatus.119
Oral uconazole can be given in a dose of 5040 mg/day,
with the usual adult dose being 200 mg/day. A topical 1%
solution in sterile water can be made. The 2 mg/L aqueous
solution for intravenous use can also be applied topically.120
Human studies on the subconjunctival use of uconazole
have given promising results in the treatment of severe nonresponding fungal keratitis121, and some animal studies demonstrate peak concentrations in the central cornea at 2 h after
subconjunctival injection.122 Further studies need to be done on
this alternative to evaluate safety, dosage, and efcacy.
Systemic side effects of uconazole include gastrointestinal
upset, headaches, rash, hepatotoxicity, anaphylaxis, Stevens
Johnson syndrome, and thrombocytopenia. Fluconazole can
Antifungal Agents
ITRACONAZOLE
Itraconazole also has, like uconazole, a wider spectrum of
activity than the imidazoles. Its spectrum of activity includes
excellent in vitro activity against Aspergillus. Its broad spectrum
of antifungal activity includes Candida species, Paecilomyces,
Paracoccidioides, and Coccidioides.124 It has not been very
effective against Fusarium.125
It has had a very limited use in clinical ophthalmology. In
an experimental model of Candida endophthalmitis, it was
shown to be as effective as uconazole and ketoconazole.16
There is a published report of successful treatment of
Aspergillus scleritis with oral itraconazole after cataract
surgery.126 The oral administration of itraconazole appears to
have less penetration than other triazoles into the cornea,
aqueous, and vitreous.16
Itraconazole has been used in its oral preparation as an
adult dose of 200 mg/day. Side effects include gastrointestinal
upset, hypertriglyceridemia, and hypokalemia.127 Although
natamycin continues to be the treatment of choice for lamentous fungal keratitis, in its absence topical itraconazole
therapy should be considered, specially if the infection is due to
Aspergillus.128
PYRIMIDINES
The pyrimidines are a group of antimetabolites with known
antifungal activity. The main drug in this group is
ucytosine.129
FLUCYTOSINE
Flucytosine (5-FC) is a uorinated pyrimidine that is soluble in
water and alcohol. Several mechanisms of action have been
described.130 It may alter fungal RNA and DNA synthesis. It
enters the cytoplasm by the action of cytosine permease and is
then deaminated by cytosine deaminase into 5-uorouracil. It
is then phosphorylated and incorporated into RNA. In the
nucleus, 5-FC forms 5-uoro-2-deoxyuridylic acid (FdUMP),
which inhibits thymidilate synthetase and thus DNA
synthesis.131
Flucytosine has a limited spectrum of activity, and resistance
may be acquired at low doses (see Table 21.2).48,132 The limited
activity and resistance of 5-FC are due to the fungal cells
inability to transport the drug into its cytoplasm and incorporate it into its RNA or insufcient FdUMP synthesis to
inhibit DNA formation.130 The spectrum of activity may be
enhanced and the emergence of resistance may be reduced by
concomitant administration of amphotericin B.2,32,132,133
Both topical and oral preparations of 5-FC may be used.134 It
is available for oral administration in 250- and 500-mg
capsules. It is water soluble and rapidly absorbed from the
gastrointestinal tract. The recommended dose of 5-FC is
50150 mg kg1 day1 at 6-h intervals. The drug is excreted
unchanged in the urine, and thus the dosage should be adjusted
according to the creatinine clearance.135
A topical preparation of 1% 5-FC may be formulated; it has
limited penetration and thus is primarily effective for surface
infections (conjunctivitis, blepharitis, and canaliculitis) and
anterior stromal keratitis.136
Most side effects reported with 5-FC have been minimal and
reversible.133 Reversible elevations in levels of liver enzymes,
aspartate aminotransferase, and alkaline phosphatase may be

seen. Anemia, leukopenia, and thrombocytopenia have been
reported in patients with other severe underlying disorders who
are taking 5-FC. Two patients with intestinal perforations have
been reported.
In ophthalmology, 5-FC has been used to treat primarily
surface infections such as blepharitis, conjunctivitis, canaliculitis, and anterior keratitis.108 The topical preparation of
5-FC is preferred, since subconjunctival injections offer little
enhancement of penetration and are associated with toxicity
and discomfort.134 Its primary use has been in cases of Candida
keratitis that have not responded clinically to amphotericin B,
in which 5-FC is added to the topical regimen.137
New Agents
Voriconazole
Voriconazole is a new triazole antifungal agent derived from
uconazole with activity against various fungi resistant to uconazole. It can be used orally and intravenously. Its bioavailability is 96%, and reaches peak plasma concentration 23 h
after oral dosing. Its intraocular penetration in oral dosage has
been found to be 1.13 0.57 mg/mL and 0.81 0.31 mg/mL in
aqueous and vitreous respectively.138 Animal studies have
demonstrated that up to 25 mg/mL of intravitreal injection of
voriconazole causes no ERG changes or histologic abnormalities in the retina.139 The most common side effect is photopsia,
followed by skin rashes. As with other azole agents, hepatic
enzyme elevations can occur. In vitro studies from nonocular
isolates have shown voriconazole to have broad spectrum of
fungistatic action against most yeast and many lamentous
fungi. It has been approved for treatment of invasive
aspergillosis, and infections from P. boydii, S. apiospermeen,
and Fusarium infections in patients intolerant or with
refractory infections to other agents. Its role in ocular infections
needs to be studied further.140
Under current development in this drug group are new agents
such as posaconazole (a second-generation triazole), with fungicidal activity against Aspergillus, and ravuconazole, a fungicidal
with a long half-life (100 h), structurally similar to voriconazole.
Although some studies suggest high effectiveness of these
agents, further studies are awaited to determine safety and any
possible ophthalmologic application.141
CHAPTER 21
increase cyclosporines serum concentration and decrease the

metabolism of warfarin. Rifampin can increase the metabolism
of uconazole.123
Caspofungin
Caspofungin acetate is a parenteral antifungal for the treatment
of invasive aspergillosis in patients intolerant or refractive to
other antifungal agents. It is a member of a new class of
echinocandins, whose mechanism of action is distinct from
other antifungals, in that it inhibits synthesis of B(1,3)-Dglucan, a component of fungal cell wall. It has demonstrated in
vitro antifungal activity against Aspergillus, Candida albicans,
C. glabrata, C. parapsilosis, and other Candida species. Some
intermediate activity has been found against Histoplasma
capsulatum and Blastomyces dermatitides. Cryptococcus
neoformans and Fusarium spp. have demonstrated resistance to
caspofungin in vitro. The dosage in patients with normal
hepatic function is 70 mg intravenously on day 1, followed by
50 mg daily. Adverse effects include fever, phlebitis, and
headaches.142
Studies in animal models suggest that topical caspofungin
0.5% can be as effective as amphotericin B 0.15% for the
treatment of Candida keratitis.143 There is also evidence of
possible clinical efcacy of intravenous use of caspofungin for
treatment of endophthalmitis by Candida glabrata.144 Further
studies are necessary to determine its clinical usefulness in
ophthalmology.
235
SECTION 4
REFERENCES
236
1. Axelrod AJ, Peyman GA, Apple DJ: Toxicity

2. Beggs WH: Mechanisms of synergistic
interactions between amphotericin B and
ucytosine. J Antimicrob Chemother 1986;
17:402404.
3. Harris DJ Jr, Stulting RD, Waring GO III,
Wilson LA: Late bacterial and fungal
keratitis after corneal transplantation.
Spectrum of pathogens, graft survival and
visual prognosis. Ophthalmology 1988;
95:14501457.
4. Jones DB, Sexton R, Rebell G: Mycotic
keratitis in south Florida: a review of 39
cases. Trans Ophthalmol Soc UK 1969;
89:781797.
5. ODay DM: Selection of appropriate
antifungal therapy. Cornea 1987;
6:238245.
6. Cohen J: Antifungal chemotherapy. Lancet
1982; 2:532537.
7. Duane TD, ed: Clinical ophthalmology.
Hagerstown, MD: Harper & Row; 1990.
8. Jones DB: Fungal keratitis. In: Duane T,
ed. Clinical ophthalmology. Hagerstown,
MD: Harper & Row; 1985.
9. Cruciani M, Di Perri G, Concia E, et al:
Fluconazole and fungal ocular infection
[letter]. J Antimicrob Chemother 1990;
25:718720.
10. Davey PG: New antiviral and antifungal
drugs. BMJ 1990; 300:793. Erratum BMJ
1990; 300:1378.
11. Delescluse J: Itraconazole in tinea
versicolor: a review. J Am Acad Dermatol
1990; 23:551554.
12. Odds FC, Cheesman SL, Abbott AB:
Antifungal effects of uconazole (UK
49858), a new triazole antifungal, in vitro.
13. Garcia-de Lomas J, Fons MA, Nogueira
JM, et al: Chemotherapy of Aspergillus
fumigatus keratitis: an experimental study.
Mycopathologia 1985; 89:135138.
14. Dupont B, Drouhet E: Early experience with
itraconazole in vitro and in patients:
pharmacokinetic studies and clinical
results. Rev Infect Dis 1987; 9:571576.
15. Ringel SM: New antifungal agents for the
systemic mycoses. Mycopathologia 1990;
109:7587.
16. Savani DV, Perfect JR, Cobo LM, Durack
DT: Penetration of new azole compounds
into the eye and efcacy in experimental
Candida endophthalmitis. Antimicrob
17. Donovick R, Gold W, Pagano JF, Stoud HA:
Amphotericins A and B, antifungal
antibiotics produced by a streptomycete.
I. In vitro studies. Antibiotics Annu
1995; 3:579586.
18. Havener WH: Ocular pharmacology. 5th
edn. St Louis, MO: CV Mosby; 1983.
19. Kotler-Brajtburg J, Medoff G, Kobayashi
GS, et al: Classication of polyene
antibiotics according to chemical structure
and biological effects. Antimicrob Agents
Chemother 1979; 15:716722.
20. ODay DM, Ray WA, Robinson RD, et al: In
vitro and in vivo susceptibility of Candida
keratitis to topical polyenes. Invest
21. Kuroda S, Uno J, Arai T: Target substances
of some antifungal agents in the cell
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
membrane. Antimicrob Agents Chemother

1978; 13:454459.
Hamilton-Miller JM: Chemistry and biology
of the polyene macrolide antibiotics.
Bacteriol Rev 1973; 37:166196.
Lorian V: Antibiotics in laboratory medicine.
4th edn. Baltimore, MD: Williams & Wilkins;
1996:202.
Goodman LS, Gilman AG, Gilman A: The
pharmacological basis of therapeutics.
New York: Macmillan; 1990.
Grayson M: Diseases of the cornea. 2nd
ed. St Louis, MO: CV Mosby; 1983.
Jones DB, Green MT, Osato MS, et al:
Endogenous Candida albicans
endophthalmitis in the rabbit.
Chemotherapy for systemic effect. Arch
Ophthalmol 1981; 99:21822187.
85:357373.
Stern GA, Fetkenhour CL: Intravitreal
amphotericin B treatment of Candida
endophthalmitis. Arch Ophthalmol 1977;
95:8993.
Stern GA, Okumoto M, Smolin G:
Combined amphotericin B and rifampin
treatment of experimental Candida albicans
97:721722.
Brajtburg J, Elberg S, Medoff G, Kobayashi
GS: Increase in colony-forming units of C.
albicans after treatment with polyene
antibiotics. Antimicrob Agents Chemother
1981; 19:199200.
Brajtburg J, Kobayashi D, Medoff G,
Kobayashi GS: Antifungal action of
amphotericin B in combination with other
polyene and imidazole antibiotics. J Infect
Dis 1982; 146:138146.
Edwards JE Jr., Morrison J, Henderson DK,
Montgomerie JZ: Combined effect of
amphotericin B and rifampin on Candida
sp. Antimicrob Agents Chemother 1980;
17:484487.
Eilard T, Beskow D, Norrby R, et al:
Combined treatment with amphotericin B
and ucytosine in severe fungal infections.
Chin GN, Hyndiuk RA, Kwasny GP, Schultz
RO: Keratomycosis in Wisconsin. Am J
Ophthalmol 1975; 79:121125.
Green WR, Bennett JE, Goos RD: Ocular
penetration of amphotericin B. Arch
Ophthalmol 1965; 73:769775.
Foster JB, Almeda E, Littman ML: Some
intraocular and conjunctival effects of
amphotericin B in man and in the rabbit.
Wood TO, Williford W: Treatment of
keratomycosis with amphotericin B 0.15%.
ODay DM, Head WS, Robinson RD,
Clanton JA: Bioavailability and penetration
of topical amphotericin B in the anterior
segment of the rabbit eye. J Ocul
Pharmacol 1986; 2:371378.
Bell R, Ritchey JP: Subconjunctival nodules
after amphotericin B injection: medical
therapy for Aspergillus corneal ulcer. Arch
Ophthalmol 1973; 90:402404.
Axelrod AJ, Peyman GA: Intravitreal
amphotericin B treatment of experimental
fungal endophthalmitis. Am J Ophthalmol
1973; 76:584588.
41. Axelrod AJ, Peyman GA, Apple DJ: Toxicity

42. Fine BS, Zimmerman LE: Therapy of
experimental intraocular Aspergillus
infection. Arch Ophthalmol 1960;
64:849861.
43. Perraut LE Jr, Perraut LE: Successful
treatment of Candida albicans
endophthalmitis with intravitreal
amphotericin B. Arch Ophthalmol 1981;
99:15651567.
44. Lou P, Kazdan J, Bannatyne RM, Cheung
45. Brod RD, Flynn HW Jr, Clarkson JG, et al:
Endogenous Candida endophthalmitis.
Management without intravenous
amphotericin B. Ophthalmology 1990;
97:662672.
46. Huang K, Peyman GA, McGetrick J:
Vitrectomy in experimental endophthalmitis.
1. Fungal infection. Ophthalmic Surg 1979;
10:8486.
47. Medoff G, Dismukes EE, Meade RH III,
Moses JM: A new therapeutic approach to
Candida infections. Arch Intern Med 1972;
130:241245.
48. Medoff G, Kobayashi GS: Medical
progress. Strategies in the treatment of
systemic fungal infections. N Engl J Med
1980; 302:145155.
49. Bennett JE, Dismukes WE, Duma RJ, et al:
A comparison of amphotericin B alone and
combined with ucytosine in the treatment
of cryptococcal meningitis. N Engl J Med
1979; 301:126131.
50. Butler WT, Bennett JE, Alling DW, et al:
Nephrotoxicity of amphotericin B: early and
late effects in 81 patients. Ann Intern Med
1964; 61:175187.
51. Bannatyne RM, Cheung R: Comparative
susceptibility of Candida albicans to
amphotericin B and amphotericin B methyl
ester. Antimicrob Agents Chemother 1977;
12:449450.
52. McGetrick JJ, Peyman GA, Nyberg MA:
Amphotericin B methyl ester: evaluation for
intravitreous use in experimental fungal
endophthalmitis. Ophthalmic Surg 1979;
10:2529.
53. ODay DM, Ray WA, Head WS, Robinson
RD: Efcacy of antifungal agents in the
cornea. IV: amphotericin B methyl ester.
Invest Ophthalmol Vis Sci 1984; 25:851854.
54. Jones DB: Decision-making in the
management of microbial keratitis.
55. Natamycin for keratomycosis. Med Lett
Drugs Ther 1979; 21:7980.
56. Jones DB, Forster RK, Rebell G: Fusarium
solani keratitis treated with natamycin
(pimaricin). Arch Ophthalmol 1972;
88:147154.
57. ODay DM, Ray WA, Robinson RD, Head
WS: Correlation of in vitro and in vivo
susceptibility of Candida albicans to
amphotericin B and natamycin. Invest
58. Newmark E, Ellison AC, Kaufman HE:
Pimaricin therapy of Cephalosporium and
Fusarium keratitis. Am J Ophthalmol 1970;
69:458466.
59. Newmark E, Kaufman HE, Polack RM,

Ellison AC: Clinical experience with
pimaricin therapy in fungal keratitis. South
Med J 1971; 64:935941.
60. ODay DM, Head WS, Robinson RD,
Clanton JA: Corneal penetration of topical
amphotericin B and natamycin. Curr Eye
Res 1986; 5:877882.
61. Ellison AC: Intravitreal effects of pimaricin
in experimental fungal endophthalmitis. Am
J Ophthalmol 1976; 81:157161.
62. Ellison AC, Newmark E: Intraocular effects
of pimaricin. Ann Ophthalmol 1976;
8:987995.
63. Ellison AC: Intravenous effects of pimaricin
on mycotic endophthalmitis. Ann
Ophthalmol 1979; 11:157164.
64. Ellison AC, Newmark E, Kaufman HE:
Chemotherapy of experimental
keratomycosis. Am J Ophthalmol 1969;
68:812819.
64a. Rosa RH Jr, Miller D, Alfonso EC: The
changing spectrum of fungal keratitis in
south Florida. Opthalmology 1994;
101:10051013.
65. Mangiaracine AB, Liebman SD: Fungus
keratitis (Aspergillus fumigatus). Treatment
with nystatin. Arch Ophthalmol 1957;
58:695698.
66. Tabbara KF, Hyndiuk RA: Infections of the
eye. Boston, MA: Little, Brown; 1986.
67. Plempel M: Pharmacokinetics of imidazole
antimycotics. Postgrad Med J 1979;
55:662666.
68. Borgers M: Mechanism of action of
antifungal drugs, with special reference to
the imidazole derivatives. Rev Infect Dis
1980; 2:520534.
69. Iwata K, Kanda Y, Yamaguchi H, Osumi M:
Electron microscopic studies on the
mechanism of action of clotrimazole on
Candida albicans. Sabouraudia 1973;
11:205209.
70. Sud IJ, Feingold DS: Heterogeneity of
action mechanisms among antimycotic
imidazoles. Antimicrob Agents Chemother
1981; 20:7174.
71. De Nollin S, Borgers M: The ultrastructure
of Candida albicans after in vitro treatment
with miconazole. Sabouraudia 1974;
12:341351.
72. Stern GA: In vitro antibiotic synergism
against ocular fungal isolates. Am J
Ophthalmol 1978; 86:359367.
73. Moody MR, Young VM, Morris MJ,
Schimpff SC: In vitro activities of
miconazole, miconazole nitrate, and
ketoconazole alone and combined with
rifampin against Candida sp. and
Torulopsis glabrata recovered from cancer
patient. Antimicrob Agents Chemother
1980; 17:871875.
74. Schacter LP, Owellen RJ, Rathbun HK,
Buchanan B: Antagonism between
miconazole and amphotericin B. Lancet
1976; 2:318.
75. Beggs WH, Sarosi GA, Steele NM:
Inhibition of potentially pathogenic
yeast-like fungi by clotrimazole in
combination with 5-uorocytosine or
amphotericin B. Antimicrob Agents
Chemother 1976; 9:863865.
76. Plempel M, Buchel KH, Bartmann K, Regel
E: Antimycotic properties of clotrimazole.
Postgrad Med J 1974; 50(Suppl 1):1112.
77. Duhm B, Medenwald H, Puetter J, et al:
The pharmacokinetics of clotrimazole 14C.
Postgrad Med J 1974; 50:(Suppl):1316.
78. Jones BR: Principles in the management of

oculomycosis. Trans Am Acad Ophthalmol
Otolaryngol 1975; 79:719751.
79. Jones BR: Principles in the management of
ocular mycoses. Am J Ophthalmol 1975;
79:719.
80. Jones BR, Richards AB: Clotrimazole in the
treatment of ocular infection by Aspergillus
fumigatus. Postgrad Med J 1974; 50(Suppl
1):3945.
81. Tettenborn D: Toxicity of clotrimazole.
Postgrad Med J 1974; 50:(Suppl):1720.
82. Van Cutsem JM, Thienpont D: Miconazole,
a broad-spectrum antimycotic agent with
antibacterial activity. Chemotherapy 1972;
17:392404.
83. Corrado ML, Kramer M, Cummings M, Eng
RH: Susceptibility of dematiaceous fungi to
amphotericin B, miconazole, ketoconazole,
ucytosine and rifampin alone and in
combination. Sabouraudia 1982; 20:109113.
84. Cosgrove RF, Beezer AE, Miles RJ: In vitro
studies of amphotericin B in combination
with the imidazole antifungal compounds
clotrimazole and miconazole. J Infect Dis
1978; 138:681685.
85. Dixon D, Shadomy S, Shadomy HJ, et al:
Comparison of the in vitro antifungal
activities of miconazole and a new
imidazole, R41,400. J Infect Dis 1978;
138:245248.
86. Fitzsimons RB, Nicholls MD, Billson FA,
et al: Fungal retinitis: a case of Torulopsis
glabrata infection treated with miconazole.
87. Foster CS: Miconazole therapy for
91:622629.
88. Foster CS, Lass JH, Moran-Wallace K,
Giovanoni R: Ocular toxicity of topical
99:10811084.
89. Foster CS, Stefanyszyn M: Intraocular
Ophthalmol 1979; 97:17031706.
90. Fowler BJ: Treatment of fungal
endophthalmitis with vitrectomy and
intraocular injection of miconazole. J Ocul
Ther Surg 1984; 3:431706.
91. Fitsimons R, Peters AL: Miconazole and
ketoconazole as a satisfactory rst-line
treatment for keratomycosis. Am J
Ophthalmol 1986; 101:605608.
92. Ishibashi Y, Matsumoto Y, Takei K: The
effects of intravenous miconazole on fungal
keratitis. Am J Ophthalmol 1984; 98:433437.
93. Gallo J, Grunstein H, Clifton-Bligh P, et al:
Miconazole in fungal endophthalmitis.
Lancet 1982; 1:53.
94. Jaben SL, Forster RK: Intraocular
miconazole therapy in fungal
endophthalmitis. Invest Ophthalmol Vis Sci
1981; 20(Suppl):109.
95. Bisschop MP, Merkus JM, Scheygrond H,
et al: Treatment of vaginal candidiasis with
ketoconazole, a new, orally active
antimycotic. Eur J Obstet Gynaecol Reprod
Biol 1979; 9:253259.
96. Van Den Bossche H, Willemsens G, Cools
W, Cornelissen F: Inhibition of ergosterol
synthesis in Candida albicans by
ketoconazole. Arch Int Physiol Biochim
1979; 87:849851.
97. Borelli D, Fuentes J, Leiderman E, et al:
Ketoconazole, an oral antifungal:
Laboratory and clinical assessment of
imidazole drugs. Postgrad Med J 1979;
55:657661.
98. Oji EO: Ketoconazole: a new imidazole

antifungal agent has both prophylactic
potential and therapeutic efcacy in
keratomycosis of rabbits. Int Ophthalmol
1982; 5:163167.
99. Oji EO: Study of ketoconazole toxicity in
rabbit cornea and conjunctiva. Int
Ophthalmol 1982; 5:169174.
100. Torres MA, Mohamed J, Cavazos-Adame
H, Martinez LA: Topical keratoconazole for
fungal keratitis. Am J Ophthalmol 1985;
100:293298.
101. DeFelice R, Johnson DG, Galgiani JN:
Gynecomastia with ketoconazole.
19:10731074.
102. Komadina TG, Wilkes TDI, Shock JP, et al:
Treatment of Aspergillus fumigatus keratitis
in rabbits with oral and topical
ketoconazole. Am J Ophthalmol 1985;
99:476479.
103. Bodey GP: Azole antifungal agents. Clin
Infect Dis 1992; 14(Suppl 1):161169.
104. Maichuk IUF, Karimov MK, Lapshina NA:
Ketoconazole in the treatment of ocular
mycoses. Vestn Oftalmol 1990; 106:4446.
105. Rajasekaran J, Thomas PA, Srinivasan R:
Ketoconazole in keratomycosis. In: Blodi F,
Brancato R, Cristini G, et al, eds. Acta XXV
Concilium Ophthalmologicum. Amsterdam
Netherlands: Kugler Ghedin, 1988: 24622467.
106. Ishibashi Y: Oral ketoconazole therapy for
95:342345.
107. Hendy KK, Chu W, Foster CS: Intraocular
penetration of ketoconazole in rabbits.
Cornea 1992; 11:329333.
108. Smolin G, Thoft RA: The cornea. Boston,
MA: Little, Brown; 1987.
109. Upadhyay MP, West EP, Sharma AP:
Keratitis due to Aspergillus avus
successfully treated with thiabendazole.
110. Smolin G, Okumoto M, eds: Antimicrobial
agents in ophthalmology. New York:
Masson; 1983.
111. Rysselaere M: The effect of econazole in
experimental oculomycosis in rabbits.
Mykosen 1981; 24:238240.
112. Oji EO, Clayton YM: The role of econazole
in the management of oculomycosis. Int
Ophthalmol 1982; 4:137142.
113. Prajna NV, John RK, Nirmalan PK, et al: A
randomized clinical trial comparing 2%
econazole and 5% corneal natamycin for
the treatment of fungal keratitis. Br J
Ophthalmol 2003; 87:12351237.
114. Prajna NV, Nirmalan PK, Mahalakshmi R,
et al: Concurrent use of 5% nathamycin
and 2% econazole for the management
of fungal keratitis. Cornea 2004;
23:793796.
115. Richardson K, Cooper K, Marriott MS,
et al: Design and evaluation of a
systemically active agent, uconazole.
Ann N Y Acad Sci 1988; 544:411.
116. Isulka B, Stambridge T: Fluconazole in the
treatment of candidal prosthetic valve
endocarditis. BMJ 1988; 297:178179.
117. Manzouri B, Vadis G, Wyse R:
Pharmacotherapy of fungal eye infections.
Expert Opin Pharmacother 2001;
2:18491857.
118. Brooks JH, OBrien TP, Wilhelmus KR,
et al: Comparative topical triazole therapy
of experimental Candida albicans keratitis.
31(Suppl):2793.
CHAPTER 21
Antifungal Agents
237
SECTION 4
238
119. Avunduk AM, Beuerman RW, Warnel ED,

et al: Comparison of efcacy of topical and
oral uconazole treatment in esperimental
Aspergillus keratitis. Curr Eye Res 2003;
26:113117.
120. Brammer KW, Farrow PR, Faulkner JK:
Pharmacokinetics and tissue penetration of
uconazole in humans. Rev Infect Dis
1990; 12(Suppl 3):S318326.
121. Yilmaz S, Maden A: Severe fungal keratitis
treated with subconjunctival uconazole.
Am J of Opthalmol 2005; 140:454458.
122. Klippenstein K, ODay DM, Robinson RD,
et al: The qualitative evaluation of the
pharmacokinetics of subconjunctivally
injected antifungal agents in rabbits.
Cornea 1993; 12:512516.
123. Rhee P, OBrien TP: Pharmacotherapy of
fungus infections of the eye. In:
Zimmerman TJ, ed. Textbook of ocular
pharmacology. Philadelphia, PA: LippincottRaven; 1997:587607.
124. Sugar AM: Fluconazole and itraconazole:
current status and prospects for antifungal
therapy. Curr Clin Top Infect Dis 1998;
13:7498.
125. Bloom PA, Laidlaw DA, Easty DL,
Warnoch DW: Treatment failure in a case
of fungal keratitis caused by
Pseudallescheria boydii. Br J Ophthalmol
1992; 76:367368.
126. Carlson AN, Foulks J, Perfect J, Kim J:
Fungal scleritis after cataract surgery.
Cornea 1992; 11:151154.
127. Heykants J, Van Peer A, Lavrijsen K, et al:
Pharmacokinetics of oral antifungals and
128.
129.
130.
131.
132.
133.
134.
135.
136.
137.
their clinical implications. Br J Clin Pract

1990; 71(Suppl):5056.
Kalavathy CM, Parmar P, Kaliamurthy J,
et al: Comparison of topical itraconazole
1% with topical natamycin 5% for the
treatment of lamentous fungal keratitis.
Cornea 2005; 24:449452.
Shadomy S, Kirchoff CB, Ingroff AE: In vitro
activity of 5-uorocytosine against Candida
and Torulopsis species. Antimicrob Agents
Chemother 1973; 3:914.
Wagner GE, Shadomy S: Studies on the
mode of action of 5-uorocytosine in
Aspergillus species. Chemotherapy 1979;
25:6169.
Diasio RB, Bennett JE, Myers CE: Mode of
action of 5-uorocytosine. Biochem
Pharmacol 1978; 27:703707.
Firkin FC: Therapy of deep-seated fungal
infections with 5-uorocytosine. Aust N Z J
Med 1974; 4:462467.
Harder EJ, Hermans PE: Treatment of
fungal infections with ucytosine. Arch
Intern Med 1975; 135:231237.
Walsh JA, Haft DA, Miller MM HG, et al:
Ocular penetration of 5-uorocytosine.
Invest Ophthalmol 1978; 17:691694.
Polak A: Pharmacokinetics of amphotericin
B and ucytosine. Postgrad Med J 1979;
55:667670.
Romano A, Segal E, Eyelan E, Stein R:
Treatment of external ocular Candida
infections with 5-uorocytosine.
Montgomerie JZ, Edwards JE Jr, Guze LB:
Synergism of amphotericin B and
138.
139.
140.
141.
142.
143.
144.
5-uorocytosine for Candida species.

J Infect Dis 1975; 132:8286.
Breit SM, Hariprasad SM, Mieler WF, et al:
Management of endogenous fungal
caspofungin. Am J Ofthalmol 2005;
139:135140.
Gao H, Pennesi M, Shah K, et al: Safety of
intravitreal voriconazole;
electroretinographic and histopathologic
studies. Trans am Opthalmol Soc 2003;
101:183189.
Marangon FB, Miller D, Giaconi JA, Alfonso
EC: In vitro investigation of voriconazole
susceptibility for keratitis and
endophthalmitis fungal pathogens. Am J
Ophthalmol 2004; 137:820825.
Steinbach WJ, Benjamin DK: New
antifungal agents under development in
children and neonates. Curr Opin Infect Dis
2005; 18:484489.
Rybowicz J, Gurk-Turner C: Caspofungin:
the rst agent available in the echinocandin
antifungals. Proc Bayl Univ Med Cent 2002;
15:9799.
rabbit model. Antimicrob Agents
Chemother 2005; 49:13591363.
Sarria JC, Bradley JC, Habash R, et al:
Candida glabrata endophthalmitis treated
successfully with caspofungin. Clin Infect
Dis 2005; 40:4648.
CHAPTER
22
Antiparasitics
Savitri Sharma, Virender S. Sangwan, and Nalini A. Madiwale
PARASITES AFFECTING THE EYE
INTRODUCTION
Parasitology as a science has grown dramatically in the recent
years, particularly owing to the parasitic diseases that have
found prominence in patients with compromised immune
systems. The rapidity of modern international travel has only
added to the spread of parasitic diseases from endemic to
nonendemic areas.
Parasitic infections of the eye may be a manifestation of
generalized systemic disease or a localized phenomenon.
Considerable ocular morbidity and blindness can be caused by
parasites, some of them conned to geographical areas and
some of them widespread. While onchocerciasis is common in
Africa, Acanthamoeba infections have been reported from all
over the world.
The control and eradication of parasitic infections require a
multifaceted approach that includes vector control, health education, and improved sanitation. Nevertheless, chemotherapy
remains the most efcient and effective means of control of
parasitic diseases. Chemotherapy with antiparasitics is required
to deal with these infections that may be important causes of
morbidity and mortality.
This chapter describes chemotherapeutic agents, currently
used for the treatment of ocular parasitic infections, along
with brief description of the parasite and accompanying ocular
manifestations.
Key Features
Classication
Nomenclature
Relationship with intermediate host
Whittaker in 1969 proposed ve kingdoms for all living organisms: Monera, Protista, Fungi, Plantae and Animalia. Protozoa
are eukaryotic unicellular organisms belonging to the kingdom
Protista and helminths are eukaryotic multicellular organisms
and are placed in the kingdom Animalia.1 Table 22.1 lists the
parasites that have been reported to affect the eye.
Parasitic infections may originate from a large number of
sources, contaminated water and soil being the commonest.
Other sources include fresh water shes, crabs, undercooked/
raw beef or pork, blood sucking insects, housefly, pet animals,
etc. In most cases, the denitive host is the mammalian host in
which either the most developed form of the parasite occurs or
the sexual reproduction of the parasite takes place. Table 22.2
outlines the relationship of some of the common parasites to
the intermediate host, which harbors the larval or sexual stage
of the parasite. The modalities of chemotherapy often depend
on the stage of the parasite occurring in the human host.
TABLE 22.1. Classication of Parasitic Eye Infections Caused by Protozoa, Helminths and Arthropods
Protozoa
Helminths
Arthropods
Nematodes
Cestodes
Trematodes
Toxoplasmosis
Toxocariasis
Cysticercosis
Schistosomiasis
Acanthamoebiasis
Ascariasis
Echinococcosis
Paragonimiasis
Entamoebiasis
Onchocerciasis
Coenurosis
Malaria
Loiasis
Sparganosis
Giardiasis
Dirolariasis
Leishmaniasis
Filariasis
Trypanosomiasis
Dracunculiasis
Pneumocystosis
Thelaziasis
Microsporidiosis
Ophthalmomyiasis
Gnathostomiasis
Angiostrongyliasis
Trichinosis
239
TABLE 22.2. Relationship of Common Parasites to Their Intermediate Host

No Intermediate Host
Helminths
Parasite
Intermediate
Host
Acanthamoeba
Trichuris trichiura
Taenia solium
Pig
Microsporidia
Ascaris lumbricoides
Taenia saginata
Cow
Giardia
Ancylostoma duodenale
Echinococcus granulosus
Man
Entamoeba
Necator americanus
Plasmodium
Man
Trypanosoma cruzi
Reduviid bug
Wuchereria bancrofti
Mosquito
Brugia malayi
Mosquito
Schistosoma
Snail
Leishmania
Sandfly
Trypanosoma
Tsetse fly
Loa loa
Chrysops
Onchocerca volvulus
Simulium fly
SECTION 4
ANTIPARASITICS FOR PROTOZOAL

INFECTIONS
240
TOXOPLASMOSIS
Key Features
One Intermediate Host
Protozoa
Geographical distribution
Life cycle
Ocular manifestations
Treatment
Toxoplasmosis is a common parasitic infection in humans. It is

estimated to infect at least 10% of adults in northern temperate
countries and more than half of adults in Mediterranean and
tropical countries. Toxoplasmosis is caused by Toxoplasma
gondii, an obligate intracellular protozoan of cosmopolitan
distribution. The domestic cat is the denitive host. Oocysts
excreted in cat feces have been shown to survive in soil for long
periods of time. Human infection can occur after ingestion of
either tissue cysts (bradyzoites) or oocysts (sporozoites).
Transmission occurs by contact with contaminated feces,
ingestion or handling of infected meat, or drinking of contaminated water. Transplacental spread causes a congenital
infection. On entry into the host, the cyst wall is disrupted,
releasing actively replicating, invasive tachyzoites. The hosts
immune response then transforms the tachyzoites into slowly
dividing bradyzoites in tissue cysts. The life cycle is completed
only when the cat ingests infected uncooked meat.
Acute focal retinochoroiditis, papillitis, papilledema, vitritis,
and recurrent retinitis are commonly seen ocular manifestations. A granulomatous anterior uveitis is sometimes seen.
In the immunocompetent host, toxoplasmosis is a self-limiting
disease. In the immunocompromised host the retinochoroiditis
takes on a severe necrotizing form and may occur in conjunction with life-threatening systemic infection.
The goal of medical therapy is to prevent damage to the
retina and optic nerve, thereby preventing permanent vision
loss. The management of ocular toxoplasmosis in immunocompetent adults must consider various factors such as: selflimiting nature of the active phase of the disease, retinal
necrosis due to proliferation of organisms, damage to the intraocular tissues due to immune response to the organisms, and
inability of the current drugs to eliminate tissue cysts and
prevent recurrence.
In 1991 Engstrom and associates conducted a survey of all
the physician members of the American Uveitis Society to
determine the current practices in the management of ocular
toxoplasmosis.2 Among the respondents, only 6% treated all the
active lesions, regardless of ocular ndings. The majority of
respondents felt that the lesions should be observed without
treatment if the visual acuity remained 20/20 in the affected eye
and lesions were located in the far periphery of the retina.
Majority of the respondents agreed that the following factors
were indications for medical therapy: any decrease in visual
acuity, macular or peripapillary lesions, lesions greater than one
disk diameter in size, lesions associated with moderate to severe
vitiritis, presence of multiple active lesions, persistence of active
lesions for more than a month, and any ocular lesions associated with recently acquired infection. Various drugs used for
treatment of ocular toxoplasmosis are listed in Table 22.3.
Systemic corticosteroids should be used either concomitant
with antimicrobials or after 2448 h of antimicrobial therapy.
The combination of pyrimethamine and sulfadiazine is
probably most effective against toxoplasmosis and therefore
recommended as the treatment of choice for sight-threatening
ocular toxoplasmosis.3 Quadruple therapy, consisting of
clindamycin, pyrimethamine, sulfonamides, and prednisone,
has been claimed to represent an even more effective alternative, but no comparison between the triple and quadruple
therapy is available. Some of the newer antimicrobial agents,
including atovaquone and azithromycin, reduce the number
of tissue cysts in animal models.4 Rothova and associates found
a relationship between treatment with pyrimethamine/
sulfadiazine and reduction of lesion size.5
ACANTHAMOEBIASIS
Acanthamoeba is an important cause of microbial keratitis. It is
a free-living ubiquitous protozoa and is an opportunistic
pathogen. It exists in nature as a dormant cyst, which under
favorable conditions turns into active trophozoite. First
described in 1973, the reported incidence of Acanthamoeba
Antiparasitics
Drug
Dosage
Pyrimethamine
Adults: 100 mg loading dose, followed by 25 mg/day for 3060 days

Children: 4 mg/kg loading dose followed by 1 mg/kg divided dose
Newborns should be treated daily for rst 6 months and then 3 times a week for rest of life
Dosage: 1 mg/day divided into 2 doses
Sulfadiazine
Adults: 2 g loading dose followed by 1 g every 6 h for 3060 days

Children: 100 mg kg1 day1 divided every 6 h
Newborns should be treated daily for their rst year of life. Dosage: 100 mg kg1 day1 divided into
two doses
Folinic acid
520 mg/day during pyrimethamine therapy, depending on neutrophil and platelet count
Azithromycin
5001000 mg/day for 3 weeks
Trimethoprim/Sulfamethoxazole
160/800 mg (one tablet) twice-daily for 3040 days
Atovaquone
750 mg every 6 h 46 weeks
Clindamycin
300 mg every 6 h for 3040 days

Children: 16-20 mg kg1 day1 divided every 6 h
Spiramycin
Adults: 500750 mg every 6 hour for 3040 days

Children: 100 mg kg1 day1 divided every 6 h
Tetracycline
500 mg every 6 h loading dose, followed by 250 mg every 6 h for 3040 days
Minocycline
100200/day for 3040 days
Clarithromycin
1 g every 12 h loading dose followed by 500 mg every 12 h for 4 week
Prednisone
Adults: 40100 mg/day

Children: 12 mg kg1 day1
keratitis increased in 1980s in association with the rising

popularity of contact lens wear in UK 6 and many other
countries in Europe and USA. The disease has been reported
from almost all parts of the world.7,8
Before the development of potent amoebicidal drugs in the
mid 1980s and early 1990s, the prognosis of Acanthamoeba
keratitis was generally poor. Successful use of topical propamidine isethionate (Brolene) and neomycinpolymyxin
bacitracin (Neosporin) has been reported by many.9,10 An
extensive review on in vitro efcacy of a large number of drugs
against Acanthamoeba was reported by Wright et al in 1985.11
Remarkable clinical and visual improvement were reported by
Larkin et al by using topical (0.02%) polyhexamethylene
biguanide (PHMB) in six cases of Acanthamoeba keratitis
refractory to multiple antiamoebic agents.12 Chlorhexidine and
PHMB are potent cysticidal drugs and at 0.02% concentration
are safe to the ocular surface. Although their mode of action
is similar (cationic antiseptic) they have been shown to be
synergistic in vitro.13 Commercial eye drops of these medications are not available and they need to be made in local
pharmacy. Currently, a combination of topical propamidine
isethionate with PHMB or chlorhexidine is considered efcacious in the treatment of Acanthamoba keratitis.14,15 Combination therapy with PHMB and chlorhexidine has also shown to
be efcacious.7
The dormant Acanthamoeba cysts in the cornea may contribute to chronic disease with propensity to recur. Over 25% of
patients were shown to have at least one recurrence in a review
of 20 patients reported recently.16 All patients had received
topical PHMB with propamidine isethionate and some had in
addition received chlorhexidine or neosporin. A wide range of
treatment duration (572 months) was seen in this study.
Surgical intervention (penetrating keratoplasty) was required in
30% of the cases, however, no patient lost the eye. The role of
topical steroid therapy has been debated inconclusively.17 The
pathogenicity of Acanthamoeba cysts and trophozoites has
been shown to increase with dexamethesone both in vitro

and in vivo.18
CHAPTER 22
TABLE 22.3. Drugs Used in the Treatment of Ocular Toxoplasmosis
ENTAMOEBIASIS
Early reports of ocular amoebiasis associated with Entamoeba
histolytica are based on circumstantial evidence, i.e., eye lesions
were present along with intestinal amoebiasis responding to
antiamoebic therapy but the organism was not isolated from
ocular samples. Although amoebic choroidosis was described
with excellent documentation19 E. histolytica is believed to
rarely affect the eye. Case reports of cutaneous amoebiasis
affecting the eyelid are available.20 The treatment of amoebiasis
depends on the stage of the disease and general health of
the patient. Symptomatic intestinal amoebiasis is treated with
a combination of metronidazole and diiodohydroxyquin,
750 mg three times per day for 10 days of the former and
650 mg three times per day for 20 days of the latter. For liver
abscess treatment, a combination of metronidazole and
dehydroemetine or emetine is preferred.
MALARIA
Ocular manifestations in malaria include retinal hemorrhage or
exudates, usually in cerebral malaria and indicate a poor
prognosis.21 Retinopathy after chloroquine treatment has
also been reported.22 Other rare ndings in malaria include
malarial amaurosis, optic neuritis, oculomotor paralysis, and
cortical blindness. Oral therapy of malaria consists of
chloroquine phosphate and in cases with chloroquine - resistant
Plasmodium falciparum (CRPF) infections, quinine sulfate
with pyrimethamine and sulfadiazine. For patients sensitive to
pyrimethamine or sulfadiazine, the preferred drug is quinine
sulfate with tetracycline for the treatment of CRPF. In
emergencies, intravenous use of quinine dihydrochloride or
quinine gluconate is recommended. Chloroquine phosphate,
241

500 mg once a week, beginning 1 week before travel to an
endemic area and continuing until 6 weeks after return, is
recommended by CDC for chemoprophylaxis of malaria.
Mefloquine is the drug of choice for travelers at risk of infection
with CRPF.
GIARDIASIS
Giardiasis is a waterborne infection caused by Giardia lamblia,
a binuclear flagellate protozoan that affects the upper part of the
gastrointestinal (GI) tract. Water supply contaminated with
cysts is the usual source of infection. An increased prevalence
among homosexual males has been documented. Iridocyclitis,
choroiditis, and a hemorrhagic retinopathy can coexist with
both latent and overt systemic infections. The basis of the
ocular involvement is thought to be immunologic.23
Quinacrine hydrochloride, 100 mg three times a day for
5 days, and metronidazole, 250 mg three times a day, for 5 days
are equally effective. Concurrent ocular steroids are needed to
control the exacerbation of inflammation that occurs after
initiation of treatment.
SECTION 4
LEISHMANIASIS
Mucocutaneous leishmaniasis is caused by Leishmania
braziliensis. About 1020% of the cases show ocular involvement. The extracellular flagellate, and promastigote forms are
injected into the skin through the bite of the phlebotomus
mosquito. The parasites proliferate as aflagellate amastigotes
within macrophages and endothelial cells of capillaries. Lysin
of the amastigotes by host macrophages and lymphocytes
causes an open ulcer. During a mosquito bite, the amastigotes
enter the vector and transform into promastigotes that are
transmitted to the next human through the saliva of the
infected vector.24
Ocular manifestations include granular or nodular conjunctivitis, interstitial keratitis, nodular keratitis with heavy pannus
formation, and ulcerative keratitis.25 Cutaneous leishmaniasis
generally involves eyelids, most often on the external corner.26
Eyelid lesions are usually ulcerative, with occasional spread to
conjunctiva and lacrimal ducts.
Sodium stibogluconate (Pentostam) is the drug of choice
for the treatment of leishmaniasis. A single course consists
of 10 mg/kg to a maximum of 600 mg intramuscularly or
intravenously for 610 days. A maximum of three courses of
treatment can be repeated at 10-days intervals. However,
amphotericin B, 0.51.0 mg kg1 day1 intravenously for up
to 8 weeks is used when antimonials are ineffective or
contraindicated.
TRYPANOSOMIASIS
242
Sleeping sickness or African trypanosomiasis is caused by

Trypanosoma brucei gambiense and Trypanosoma brucei
rhodesiense and the vector is tsetse fly. The ocular manifestations of this disease are generally mild and may be
associated with congestion of the eyes, edema of the lids, diffuse
corneal opacication or interstitial keratitis.27 Unilateral
anterior uveitis with or without corneal involvement may be
present. In the terminal stage, papilledema, ophthalmoplegia,
ptosis, papillitis and optic neuritis may be present, especially
with rhodesiense infections.
Treatment depends on the stage of the disease. During the
early stages suramin is given intravenously. At rst a test dose
of 100200 mg is given followed by one gram intravenously on
days 1,3,7,14, and 21. Pentamidine isethionate may be given
intramuscularly in the dosage of 4 mg kg1 day1 for 10 days.
In CNS involvement the standard drug is melarsoprol

(Mel B), a trivalent arsenic compound that may cause severe
reactive arsenic encephalopathy.28 An alternative and safer
drug is eflornithine (a-difluoromethylornithine) in the dosage
of 400 mg kg1 day1 intravenously for 14 days followed by
300 mg kg1 day1 orally for 2128 days.29 This is effective and
safer than melarsoprol.
American trypanosomiasis or Chagas disease is caused by
Trypanosoma cruzi and is transmitted by reduviid bugs.
The most important ocular manifestation is unilateral
palpebral edema which is a pathognomonic feature of Chagas
disease. Granulomatous uveitis, with the presence of T. cruzi
in the inltrate has been reported.27 The drug of choice for
Chagas disease is nifurtimox given 810 mg kg1 day1 orally
in four divided doses for 120 days. Alternatively, benznidazole,
57 mg kg1 day1 may be given for 30120 days.
PNEUMOCYSTOSIS
Pneumocystis carinii is considered a protozoan although one
study indicated it to be closer to fungi than protozoa.30 The
organism has three development stages; precyst, cyst and
trophozoite. Pneumocystosis was originally described as an
epidemic form of interstitial plasma cell pneumonitis in
children following the second world war in Europe. Since 1979,
P. carinii pneumonia (PCP) is being reported in patients with
acquired immunodeciency syndrome, which is probably a
reactivation of latent subclinical infections.
Manifestations in the eye probably occur when there is
disseminated infection. P. carinii choroidopathy has been
documented.31,32 The drug of choice is a combination of trimethoprim and sulfamethoxazole with the dosage of the former
being 20 mg kg1 day1 and the latter 100 mg kg1 day1, either
oral or intravenous in four divided doses, for 14 days. Alternative therapy with pentamidine isethionate has been reported.
MICROSPORIDIOSIS
Microsporidia are obligate intracellular parasites belonging to
the phylum Microspora. Multiple genera are involved in a wide
range of clinical diseases. The most common infection involves
the GI tract and others include encephalitis, sinusitis, myositis,
ocular infections and disseminated infection. Two forms of
microsporidial infection of the cornea have been described,
stromal or interstitial keratitis in immunocompetent33 and
supercial keratoconjunctivits seen in immunosuppressed 34 or
immunocompetent individuals.35 Various therapeutic agents
have been used, however, there are no dened guidelines for the
optimal treatment of microsporidial infections. Costa and
Weiss have described antimicrosporidial drugs in an extensive
review recently.36 Table 22.4 describes the drugs that have been
used for the treatment of ocular microsporidiosis.
TABLE 22.4. Drugs Used in the Treatment of Ocular

Microsporidiosis36
Drug
Microsporidial Species
Albendazole
Encephalitozoon cuniculi
Encephalitozoon hellem
Encephalitozoon intestinalis
Fumagillin
Encephalitozoon hellem
Encephalitozoon intestinalis
Itraconozole
Antiparasitics
ANTIPARASITICS FOR HELMINTHIC

INFECTIONS
Key Features
Geographical distribution
Life cycle
Treatment
suitability for mass therapy, and its superiority over DEC.41

Community-based treatment with ivermectin has been shown
to reduce the transmission of onchocerciasis. Ivermectin is
usually given in a single, annual, oral dose of 150 mg/kg. This
dosage seems to be adequate for all except the intensely infested
patients with severe ocular involvement in hyperendemic areas.
LOIASIS
Loiasis is caused by Loa loa and is transmitted by mango flies
of genus Chrysops. It is endemic in Central and West Africa.
The clinical disease mainly results from the migration of the
adult worms in the subcutaneous tissues called Calabar or
fugitive swelling. The worms may migrate across the bulbar
conjunctiva. Loa loa-induced retinopathy, uveitis, and
migration of the worm in the eyelid, the vitreous and the
anterior chamber have been documented.42 The drug of
choice for treatment of loiasis is DEC in a complex dosage
schedule.43
DIROFILARIASIS
Toxocariasis is caused by dog ascarid, Toxocara canis and less

frequently by Toxocara cati, the cat ascarid. Infection of man by
these organisms leads to persistent larval migration in various
viscera (visceral larva migrans) including the eye (ocular larva
migrans). The latter is usually seen in older children and young
adults and may manifest as unilateral, solitary painless lesion
located posteriorly close to optic nerve and disk. Diethylcarbamazine (DEC), thiabendazole and mebendazole are useful
in the treatment of systemic toxocariasis.
Dirolariasis is caused by Dirolaria immitis, D. tenuis,

D. repens, D. ursi or D. subdermata. Primarily seen in dogs, the
disease has been reported in humans from almost all parts
of the world. It is transmitted by mosquitoes of genera Aedes,
Anopheles and Culex. Ocular form of dirolariasis is less
common than pulmonary and subcutaneous forms. Eyelids are
commonly involved followed by orbit, subconjunctival tissue
and intraocular tissues.44 The larvae inoculated by mosquitoes
migrate and mature in the subcutaneous tissues. Most
infections consist of single worm and its surgical removal
achieves complete cure. DEC, in dosage similar to loaiasis, is
commonly used.
ONCHOCERCIASIS
FILARIASIS
Onchocerciasis is caused by the nematode Onchocerca

volvulus. It is widely distributed across the African continent
and South America.
Humans are the only known reservoir of onchocerciasis. The
female Simulium fly is the intermediate host and vector that
ingests the microlariae on biting an infected person during a
blood meal. The larvae then transform into infective forms that
may enter a new host when the simuliid takes another blood
meal. The larvae migrate in the body for ~1 year before they
settle in a nodule, which is most frequently subcutaneous.
Here, the male and female mate and produce numerous
microlariae that migrate to various parts of the body.
Ocular manifestations of onchocerciasis include punctate
keratitis surrounding dead microlariae, sclerosing keratitis,
anterior uveitis with secondary cataract and glaucoma,
chorioretinitis, and papillitis with severe constriction of the
visual elds.40
For several years DEC and suramin were the only two drugs
available for the treatment of onchocerciasis. DEC is effective
against microlariae but causes an initial aggravation of the
ocular disease and has several troublesome side effects. Suramin
is active against adult worms but has a very high intrinsic
toxicity. These two drugs were at best suboptimal for mass
treatment regimen and consisted of decreasing doses of DEC
over 18 days followed by suramin intravenously, 1 g/week for
5 weeks.
In recent years, ivermectin has revolutionized the treatment
of onchocerciasis and has largely replaced DEC and suramin.
Numerous double-blind placebo-controlled studies have
demonstrated the efcacy and safety of ivermectin, its
Bancroftian lariasis is caused by Wuchereria bancrofti and

brugian lariasis by either Brugia malayi or Brugia timori. The
adult worms live in lymphatic systems and the infection is
transmitted by mosquitoes. The ocular manifestations may be
caused by either the adult worms or microlariae. The
treatment of choice has been DEC given orally for 21 days. Oral
ivermectin can be used alternatively. Recurrence has been
reported following therapy with either drug.
TOXOCARIASIS
CHAPTER 22
The supercial corneal lesions in microsporidial keratoconjunctivitis have been reported to have resolved following
dbridement and oral itraconazole.37 Administration of albendazole (400 mg twice-daily for 24 weeks) has led to resolution
of symptoms in patients with AIDS and symptomatic Encephalitozoon intestinalis infection. Fumagillin, an antiangiogenic
agent derived from Aspergillus fumigatus, inhibits replication of
E. cuniculi in vitro and has been used topically to treat ocular
infections due to E. hellem and E. intestinalis.38,39 Fumidil B, a
puried fumagillin, has been used as topical drops in the
treatment of microsporidial keratoconjunctivitis.
DRACUNCULIASIS
Dracunculus medinensis, also known as guinea worm, causes
dracunculiasis or dracunculosis. Man acquires infection by
drinking contaminated water containing infected cyclops.
The disease is endemic in Africa and Asia. After primary
infection, the gravid female worm forms swellings in the lower
extremity and releases larvae when in contact with water.
Orbital involvement is described in early literature and the only
case describing a swelling of 4 mm diameter on the bulbar
conjunctiva is from India.45 Mechanical removal of the worm
accompanied with medical treatment with metronidazole or
thiabendazole is the usual mode of therapy.
TRICHINOSIS
Trichinosis is caused by larvae of Trichinella spiralis. Eating
infected pork is the commonest mode of infection. The larvae
parasitize skeletal muscles where they encyst. Ocular signs and
symptoms may be the rst in early phase of muscle invasion.
The earliest sign may be bilateral palpebral edema which is
243

due to invasion of extraocular muscles and concomitant
systemic allergy due to the parasite. Patient may have subconjunctival hemorrhage, photophobia, diplopia, visual hallucinations, etc. Fundus examination may reveal hyperemia,
papillary edema, retinal hemorrhages, optic neuritis or
neuroretinitis. The effective drug of choice is thiabendazole,
25 mg/kg twice-daily for 5 days during the intestinal phase.
During the muscular invasion phase mebendazole should be
used. Albendazole may also be effective in tissue phase.
Anthelminthic therapy is usually combined with topical
corticosteroids for relief of pain and swelling.46
SECTION 4
CYSTICERCOSIS
Cysticercosis is caused by larvae of tape worms Taenia solium
or Taenia saginata, the larvae of the former being called
Cysticercus cellulosae and that of latter Cysticercus bovis. In
taeniasis, man is the denitive host, the adult tape worms
residing in the intestine. In cysticercosis, man acts as the
intermediate host. Most commonly the infection is contracted
by ingesting eggs in contaminated food or water. It can occur in
patients with taeniasis, either by fecaloral auto infection or by
reverse peristalsis of proglottids into the stomach.
Ocular involvement is very common in cysticercosis
(1346%) and it is the most common helminthic ocular
infection in man.47 Posterior segment of the eye is involved in
more than 70% of reported ocular cases. In subcutaneous
cysticercosis, the lesions are numerous, rm, elastic, round,
painless nodules or papules which may become caseated or
calcied. Cysticercosis of the extraocular muscles is not
uncommon.48
The recommended treatment for neurocysticercosis includes
praziquantel therapy, however, effect of this drug in ocular
cysticercosis is not known.49,50 Metrifonate, 75 mg/kg daily for
5 days, repeated six times at 2-week intervals, is reported to be
successful in the treatment of ocular as well as cerebral and
subcutaneous cysticercosis. Treatment with a combination of
oral albendazole and prednisolone was reported to be effective
in a series of 26 cases of ocular myocysticercosis from southern
India.48 Similar combination therapy was found effective by
these authors in a series of orbital cysticercosis.50
SCHISTOSOMIASIS AND PARAGONIMIASIS
244
Schistosomiasis or bilharziasis is caused by fluke species

Schistosoma japonicum, Schistosoma mansoni and
Schistosoma haemotobium. Man gets infected through skin on
contact with water contaminated with schistosomal cercariae.
Adult worms grow in liver veins and migrate to mesenteric or
vesical veins and the damage to liver or urinany bladder is
caused by the eggs deposited in the vessels. Damage to the eye
is caused in a similar manner. Egg granulomas may be located
in the conjunctiva, lacrimal gland or in the choroid. Adult
S. mansoni worm bas been reported from the anterior chamber
and superior ophthalmic vein.51
Praziquantel is the drug of choice, the dosage for S. mansoni
being 40 mg/kg in two doses for 1 day. The other recommended
drug is oxamniquine in single dose of 15 mg/kg. Metrifonate
has been used for the treatment of S. haemetobium infection.
Paragonimiasis is caused by a lung fluke; Paragonimus
westermani. Man gets infected by eating infected crustacean
hosts such as crabs or craysh. Ocular manifestations of uveitis
is mainly due to migration of the immature worm in the ocular
tissues. There may be associated retinal hemorrhage, vitreous
hemorrhage, exudative inflammation and secondary glaucoma.
The parasite is susceptible to praziquantel at a dosage of
25 mg/kg body weight three times daily for 2 days.
PHARMACOLOGY OF ANTIPARASITIC
AGENTS
Key Features
Systemic agents
Topical agents
Dosage
Efcacy
Toxicity and side effects
SYSTEMIC AGENTS
DEC
DEC was discovered in 1947 as a result of an intensive search
for antilarials. It is a piperazine derivative with the following
structural formula:
Diethylcarbamazine
O
H3CN
NCN
C2H5
C2H5
It is used as a citrate salt that is highly soluble in water.

The mechanism of action of DEC is twofold, consisting of;
(1) decrease in the muscular activity of the microlariae and
their immobilization, probably by virtue of the hyperpolarizing
effect of the piperazine moiety; and (2) change in the surface
membranes of the microlariae, rendering them more susceptible to the host defense mechanisms. DEC is effective
against adult worms and microlariae of Loa loa and only
microlariae of O. volvulus.
DEC is rapidly absorbed from the GI tract. Peak plasma levels
of 1.6 mg/mL are achieved 12 h after a single oral dose of
200 mg. The minimum effective blood level appears to be
0.81.0 mg/mL. It rapidly equilibrates with all tissues except
fat and does not have a cumulative effect. Over 50% of the drug
is excreted unchanged in acidic urine.
DEC is a drug with low intrinsic toxicity. Anorexia, nausea,
headache, and less frequently vomiting and skin rash occur and
subside in a few days despite the continuation of treatment.
The drug appears to be safe in pregnancy. The major adverse
effects of DEC are a direct or indirect result of the death of
the microlariae and adult worms. A severe encephalitis may
be induced in Loa loa-infected patients. Patients with onchocerciasis typically manifest the Mazotti reaction, which occurs
in a few hours after the rst dose and lasts 37 days. It consists
of itching, skin rash, painful lymphadenopathy, fever, tachycardia, arthralgia, and headache. Higher doses can be tolerated
after this reaction subsides. In the eye, it produces migration of
microlariae into the cornea, straightening and immobility of
the microlariae, reaction around dead microlariae, globular
limbal inltrates of uncertain (probably immunologic) origin,
and worsening of eye lesions in heavily infected patients.
Retinal pigment epithelial changes also are known to occur. The
benecial effects of DEC, namely, a decrease in skin and corneal
microlariae, are short lived, making it an unsuitable agent for
the prophylaxis or mass treatment. DEC is well absorbed on
topical ocular and skin application but neither preparation has
any added advantage over oral administration.
Antiparasitics
Itraconazole is an investigational triazole antifungal agent.

Its mode of action against Acanthamoeba remains to be
elucidated. It has been used in the treatment of microsporidial
keratoconjunctivitis.37
Itraconazole is closely related to ketoconazole. Its absorption
from the GI tract is enhanced when given with food. The mean
plasma level of a single dose of 100 mg is 13267 ng/mL. The
plasma levels rise in the rst 13 days, with a half-life of 36 h
after 15 days of dosing. Active drug is not detectable in the urine
or cerebrospinal fluid (CSF).
Itraconazole is well tolerated. Ten to 15% of patients complain of nausea and vomiting. Rash, pruritus, dizziness, vertigo,
pedal edema, paresthesia, decreased libido, and impotence have
been reported occasionally.
Ivermectin
Ivermectin is a member of a new class of semisynthetic macrocyclic lactones called avermectins. It has a broad spectrum of
antiparasitic activity. It is now the drug of choice for onchocerciasis. It is absorbed through the GI tract and is mainly
concentrated in the liver and adipose tissue. Peak plasma levels
are achieved in 4 h after oral administration. Its half-life is
~10 h. Animal studies indicate nearly all ivermectin is excreted
in the feces unchanged. Extremely low levels of the drug are
found in the brain. Not much is known about the pharmocokinetics of ivermectin in the eye. It can be speculated that
because the drug is a macrocyclic lactone, it has poor ocular
penetration and therefore does not achieve microlaricidal
concentrations in the eye. This would cause microlarial movement out of the eye along a concentration gradient.
The exact mode of action of ivermectin is unknown. It
modies the release of the neurotransmitter g-aminobutyric
acid (GABA) but the relationship of this property to the
microlaricidal activity is unclear. The microlaricidal action
of ivermectin is slow, unlike that of DEC, and hence there is
no exacerbation of ocular inflammation. Ivermectin is neither
macrolaricidal nor embryotoxic. It causes an initial increase
followed by a decrease in embryogenesis. There is a sequestration of normally developed embryonic forms in the uterus of
the adult female worms. The failure of microlariae to be
released explains the lack of build-up of microlariae after
single-dose treatment of ivermectin is continued for the life
span of the adult worm (1015 years). It can interrupt transmission and provide clinical prophylaxis and treatment of
ocular onchocerciasis.
Systemic side effects of ivermectin are mild and transient,
consisting of headache, and painful glands lasting a few hours;
skin rash lasting a few days, an asymptomatic and intermittent
increase in the pulse rate, a decrease in the blood pressure,
an increase in temperature, and electrocardiographic (ECG)
changes.
Ivermectin therapy is not associated with exacerbation of
ocular inflammation and this is an overwhelming advantage
over medications previously used in the treatment of onchocerciasis. The hematologic changes associated with the administration of ivermectin are a transient decrease in hemoglobin,
neutrophil leukocytosis, and lymphocytopenia and an initial fall
followed by a steady rise in the eosinophil count.52
Metronidazole
Metronidazole is a nitroimidazole with a broad spectrum of
antiprotozoal and antimicrobial activity. It has the following
structural formula:
Metronidazole is directly effective against trophozoites of
Giardia lamblia at concentrations of 150 mg/mL in vitro.
Mechanism of action is linked to the ability of the nitro group
Metronidazole
H
NH COOCH3
C
O
N
N
to trap electrons from electron transport proteins and divert

them from normal energy-yielding pathways. Studies with
mammalian DNA reveal that reduced metronidazole can cause
the loss of helical structure and strand breakage of DNA.
Metronidazole is completely and promptly absorbed from the
GI tract and therapeutic plasma levels are observed 1 h after
oral administration of a single dose of 500 mg. The half-life of
the drug is 8 h. Ten percent of the drug is bound to plasma
proteins. It shows good penetration into body tissues and fluids.
Metronidazole crosses the bloodbrain barrier. Greater than
50% of the systemic clearance occurs in the liver. Phase I
biotransformation by oxidation yields active metabolites.
Conjugation with glucuronides also occurs.
The most common side effects associated with metronidazole are headache, nausea, dry mouth, and a metallic taste.
Occasionally, vomiting, diarrhea, and abdominal pain occur.
Neurotoxicity in the form of dizziness, ataxia, convulsions,
encephalopathy, and sensory neuropathies occur. These necessitate prompt withdrawal of the drug. Temporary and reversible
leukopenia can occur. Metronidazole has a well-documented
disulram-like effect. Patients should therefore be cautioned
against alcohol. Active CNS disease is a contraindication and
severe hepatic or renal dysfunction necessitate reduction in
dosage. Metronidazole and its metabolites have mutagenic activity
and hence should not be used in the rst trimester of pregnancy.
CHAPTER 22
Itraconazole
Pentosam (Sodium Stibogluconate)

Pentosam is a pentavalent antimonial that interferes with the
glycolysis and oxidation of fatty acids in the organelles called
glycosomes within the amastigotes of Leishmania brasiliensis.
Nonspecic binding of antimony to the sulfhydryl groups in the
amastigote protein may be another mechanism of action.
Pentosam is rapidly absorbed when given intramuscularly or
intravenously and is eliminated in two phases: The rst rapid
phase has a half-life of 2 h and a second slow phase has half-life
of 3376 h.
Pain at the site of intramuscular injection, GI disturbance,
muscle pain, joint stiffness, and a reversible increase in hepatic
transaminases are relatively mild side effects of pentosam
administration. However, reversible T-wave flattening and
increase in QT interval may precede serious arrhythmias.
Pyrimethamine
Pyrimethamine is a diaminopyrimidine with the following
structural formula.
It is a competitive antagonist of folic acid by virtue of its preferential inhibition of dihydrofolate reductase of the parasites.
This prevents the reduction of dihydrofolate to tetrahydrofolate
that is necessary for synthesis of purines and pyrimidines. Pyrimethamine is synergistic to sulfas by virtue of this sequential
inhibition and hence is almost always used with sulfonamide.
It is only active against actively proliferating Toxoplasma organisms. Pyrimethamine is slowly and completely absorbed after
oral administration. It accumulates in the kidney, lung, liver,
and spleen. Elimination is slow, with a half-life of 8095 h.
Occasional skin rash and decreased hematopoiesis are associated with the use of pyrimethamine. Large doses of pyrimethamine
245

Pyrimethamine
NH
CH3 CH2
associated with decient hydration and acidic or neutral urinary

pH. Hemolytic anemia, especially in patients with a glucose-6
phosphate dehydrogenase (G-6PD) deciency; reversible agranulocytosis; and an irreversible aplastic anemia are rarely seen.
The StevensJohnson syndrome, exfoliative dermatitis, serum
sickness, and sometimes, a fatal acute necrosis of the liver can
occur on the basis of hypersensitivity to the sulfonamides.
Suramin
NH2
C
for a long period of time can cause a megaloblastic anemia that
is readily reversible by discontinuing the drug or administration
of folinic acid. A severe reversible thrombocytopenia as a result
of hematologic depression is an important side effect of pyrimethamine therapy and necessitates discontinuation of the drug.
SECTION 4
Quinacrine
Quinacrine is an acridine derivative previously used as an
antimalarial but currently being used only for the treatment of
giardiasis. It is readily absorbed from the GI tract and is slowly
eliminated. Quinacrine has a cumulative effect. Its metabolism
and its mode of antiparasitic action are not well understood.
Headache, dizziness, and vomiting are frequent side effects
associated with quinacrine use. Blood dyscrasias, urticaria,
exfoliative dermatitis, yellow pigmentation of the skin, and blue
or black pigmentation of the nails may also occur. Occasionally,
ocular toxicity resembling that of chloroquine occurs.
Quinacrine should be administered with caution in patients
with psoriasis, because it can cause a severe exacerbation.
Sulfonamides
Sulfonamides are structural analogs and competitive antagonists of para-aminobenzoic acid (PABA). They act by the
inhibition of dihydropteroate synthetase, which is the enzyme
responsible for the incorporation of PABA into dihydropteroic
acid, the immediate precursor of folic acid. Sulfonamides are
synergistic to other antifolates such as pyrimethamine and
trimethoprim. The structural formula of sulfadiazine is as
follows:
Sulfadiazine
N
H2N
SO2HN
N
246
Sulfadiazine in combination with pyrimethamine is the

treatment of choice for toxoplasmosis.
Sulfonamides are rapidly absorbed from the GI tract. After a
single dose, peak plasma levels are reached in 36 h and therapeutic concentrations occur in the CSF in 4 h. They readily
cross the placental barrier. Sulfonamides are metabolized in the
liver and excreted mainly by the kidneys in the acetylated and
the free form. The excretion of both forms is accelerated by the
administration of alkali, which decreases tubular reabsorption.
The acetylated form of sulfonamides loses the antimicrobial
activity while retaining the toxicity of the parent compound.
The most common side effects associated with the use of
sulfonamides are fever, urticaria, and GI disturbances. Urinary
tract disturbances such as crystalluria and hematuria are
Suramin is the only drug effective against adult Onchocerca

volvulus. It is mirolaricidal to a lesser extent. Suramin is an
organic urea compound with high intrinsic toxicity and hence
needs to be administered under close supervision. The exact
mechanism of action of suramin is not clear. Its interference
with DNA and RNA metabolism may be the basis of its
antiparasitic action. Suramin acts mainly on female worm,
causing its death and degeneration in 5 weeks.
Suramin can only be administered intravenously. It binds
rmly to plasma proteins. After intravenous administration,
the plasma concentration of suramin drops rapidly in the rst
few hours and then stabilizes in a few days. It has a half-life of
48 h. Suramin is a large polar anion that does not enter cells
readily. It does not cross the bloodbrain barrier. Suramin is not
metabolized to any extent and is excreted unchanged, mainly by
the kidney.
Suramin therapy is usually associated with signicant
morbidity due to systemic side effects such as malaise, nausea,
nervous fatigue, fever, arthralgia, myalgia, peripheral neuropathy, and the worsening of ocular signs and symptoms that
occur in the initial phases of treatment. Optic atrophy has also
been reported. Rarely, circulatory shock and coma can occur as
an immediate reaction to suramin. Other serious reactions such
as agranulocytosis, renal shutdown, hemolytic anemia, and
jaundice are fortunately rare. Fatal reaction to suramin therapy
has been reported. Suramin has largely been replaced by
ivermectin in the treatment of onchocerciasis.
TOPICAL AGENTS
Dibromopropamide Isethionate and Propamidine
Isethionate
Dibromopropamidine isethionate and propamidine isethionate
are both aromatic diamidines with a broad spectrum of antibacterial and antifungal activity. They are marketed in England
as Brolene ointment (0.15%) and drops (0.1%). They are not
available in the United States. Intensive use of the ointment
causes local irritation and similar use of drops causes increased
conjunctival injection, chemosis, follicular conjunctivitis,
punctate corneal lesions which are reversible and do not
necessitate discontinuation of medication.11,53
Miconazole
Miconazole is an imidazole antifungal agent that also has
antiamoebic activity. All imidazoles can be made into a 1%
suspension in arachis oil or a 10-mg/mL solution for topical
use. Foster et al have shown that in rabbits miconazole reaches
high levels in the cornea and aqueous humor after topical
or subconjunctival administration.54 It was also shown to
readily penetrate the bloodaqueous barrier after intravenous
administration. Ocular side effects include supercial punctate
keratitis and stinging.
Cationic Antiseptics
Chlorhexidine and PHMB are two important cationic antiseptics that are topically used in the treatment of Acanthamoeba keratitis. While chlorhexidine is a biguanide, PHMB is a
polymeric biguanide. Both act by compromising the integrity of
Antiparasitics
the mucopolysaccharide plug that seals the ostiole of the
Acanthamoeba cyst. Irreversible loss of essential cellular components through the damaged plasmalemma results in cell
death. Corneal epithelial toxicity (clinically) is minimal for
chlorhexidine and PHMB at a concentration of 0.02%.55 Both
chlorhexidine and PHMB have amoebicidal and cysticidal
activity.56 PHMB is manufactured principally as an industrial
grade sterilant. It is used in cosmetics and soaps as preservatives, as an algastatic compound in swimming pools and a
constituent of contact lens disinfecting fluids. In early 1990,

PHMB was found to be highly effective in killing both cysts and
trophozoites in in vitro studies.57 Larkin et al reported its
successful clinical use at a concentration of 0.02%.12 Lam et al
reported that topical PHMB monotherapy leads to persistence
of infection and hence suggested use of combination therapy in
treatment of Acanthamoeba keratitis.58 PHMB has advantages
over propamidine in having high consistent cysticidal activity
and no toxicity.
1. Arora DR, Arora B: General parasitology.

Chapter 1. In: Medical parasitology.
2nd edn. New Delhi: CBS publishers and
Distributors; 2005:11.
2. Engstrom R, Holland GN, Nussenblatt RB:
Current practices in the management of
ocular toxoplasmosis. Am J Ophthalmol
1991; 111:601-610.
3. Bosch-Driessen LH, Verbraak FD,
Suttorp-Schulten MS, et al: A prospective,
randomized trial of pyrimethamine and
azithromycin vs pyrimethamine and
sulfadiazine for the treatment of ocular
toxoplasmosis. Am J Ophthalmol 2002;
134:3440.
4. Holland GN: LX Edward Jackson
Memorial Lecture. Ocular toxoplasmosis:
a global reassessment. Part II. Disease
manifestations and management. Am J
5. Rothova A, Meenken C, Buitenhuis HJ,
et al: Therapy for ocular toxoplasmosis.
Am J Ophthalmol 1993; 115:517523.
6. Radford CF, Minassion DC, Dark JK:
Acanthamoeba keratitis in England and
Wales: incidence, outcome, and risk
factors. Br J Ophthalmol 2002;
86:536542.
7. Sharma S, Garg P, Rao GN: Patient
characteristics, diagnosis and treatment
of non-contact lens related Acanthamoeba
keratitis. Br J Ophthalmol 2000;
84:11031108.
8. Sun X, Zhang Y, Li R, et al: Acanthamoeba
keratitis: clinical characteristics and
management. Ophthalmology 2006;
113:412416.
9. Moore MB, McCulley JP: Acanthamoeba
keratitis associated with contact lenses: six
consecutive cases of successful
management. Br J Ophthalmol 1989;
73:271275.
10. Sharma S, Srinivasan M, George C:
Acanthamoeba keratitis in non-contact lens
wearers. Arch Ophthalmol 1990;
108:676678.
11. Wright P, Warhurst D, Jones BR:
Acanthamoeba keratitis successfully
treated medically. Br J Ophthalmol 1985;
69:778782.
12. Larkin DFP, Kilvington S, Dart KG:
Treatment of Acanthamoeba keratitis with
13. Angel JT, Gabriel MM, Wilson LA, et al:
Effect of polyhexamethylene biguanide
and chlorhexidine on four species of
Acanthamoeba in vitro. Curr Eye Res
1996; 15:225228.
14. Seal DV, Hay J, Kirkness C, et al:
Successful medical therapy of
Acanthamoeba keratitis with topical
chlorhexidine and propamidine. Eye 1996;
10: 413421.
15. Duguid IG, Dart JK, Morlet N, et al:

Outcome of Acanthamoeba keratitis treated
with polyhexamethylene biguanide and
propamidine. Ophthalmology 1997;
104:15871592.
16. Butler TKH, Males JJ, Robinson LP, et al:
Six-year review of Acanthamoeba keratitis
in New South Wales, Australia: 19972002.
17. Park DH, Palay DA, Days SM, et al: The
role of topical corticosteroids in the
management of Acanthamoeba keratitis.
Cornea 1997; 16: 277283.
18. McClellan K, Howard K, Niederkorn JY,
et al: Effect of steroids on Acanthamoeba
cysts and trophozoites. Invest Ophthalmol
Vis Sci 2001; 42:28852893.
19. Barely AE, Hamilton HE: Central serous
choroidosis associated with amebiasis. A
record of 9 cases. Arch Ophthalmol 1957;
58:119.
20. Baez MJ, Ramirez BEJ: Cutaneous
amebiasis of the face. A case report. Am J
Trop Med Hyg 1986; 35:6971.
21. Looareesuwan S, Warrell DA, White NJ,
et al: Retinal hemorrhage, a common sign
of prognostic signicance in cerebral
malaria. Am I Trop Med Hyg 1983;
32:911915.
22. Sassani JW, Brucker AJ, Cobbs W, et al:
Progressive chloroquine retinopathy. Ann
23. Anderson ML, Grifth DG: Intestinal
giardiasis associated with ocular
inflammation. J Clin Gastroenterol 1985;
7:169172.
24. Markell EK, Voge M, John DT: Medical
parasitology. 6th edn. Philadelphia: WB
Saunders; 1986.
25. Duke Elder S: System of ophthalmology.
XV. Summary of systemic ophthalmology.
St Louis: CV Mosby; 1976.
26. Sodafy M, Aminlari A, Resaei H:
Ophthalmic leishmaniasis. Clin Exp
Dermatol 1981; 6:485488.
27. Rodger FC: Eye disease in the tropics.
Edinburgh: Churchill Livingstone; 1981:
8384.
28. Haller L, Adams H, Merouze F, et al:
Clinical and pathological aspects of human
African Med Hyg trypanosomiasis
(T.b. gambiense) with particular reference
to reactive arsenical encephalopathy. Am J
Trop Med Hyg 1986; 35:9499.
29. Doua F, Boa FY, Schechter PJ, et al:
Treatment of human late stage gambiense
trypanosomiasis with adifluoromethylornithine (eflornithine): efcacy
and tolerance in 14 cases in Cote dIvoire.
Am J Trop Med Hyg 1987; 37:525533.
30. Edman JC, Kovacs JA, Masur H, et al:
Ribosomal RNA sequence shows
Pneumocystis carinii to be a member of the
fungi. Nature 1988; 334:519522.
31. Freeman WR, Gross JG, Labelle J, et al:

Pneumocystis carinii choroidopathy: a new
clinical entity. Arch Opthalmol 1989;
107:863867.
32. Rao NA, Zimmerman PL, Boyer D, et al:
A clinical histopathologic, and electron
microscopic study of Pneumocystis carinii
choroiditis. Am J Ophthalmol 1989;
107:218228.
33. Vemuganti GK, Garg P, Sharma S:
Is microsporidial keratitis an emerging
cause of stromal keratitis? a case series
study. BMC Ophthalmol 2005; 5:19.
http://www.biomedcentral.com/
14712415/5/19.
34. Yee RW, Tio FO, Martinez, et al: Resolution
of microsporidial epithelial keratopathy in a
patient with AIDS. Ophthalmol 1991;
98:196201.
35. Joseph J, Sridhar MS, Murthy S, et al:
Clinical and microbiological prole of
microsporidial keratoconjunctivitis in southern
India. Ophthalmol 2006; 113:531537.
36. Costa SF, Weiss LM: Drug treatment of
microsporidiosis. Drug Resist Updat 2000;
3:116.
37. Sridhar MS, Sharma S: Microsporidial
keratoconjunctivitis in a HIV seronegative
patient treated with debridement and oral
itraconazole. Am J Ophthalmol 2003;
136:745746.
38. Diesenhouse MC, Wilson LA, Corrent CF,
et al: Treatment of microsporidial
keratoconjunctivitis with topical fumagillin.
Am J Ophthalmol 1993; 115:293298.
39. Roserger DF, Serdaravic ON, Evlandson RA,
et al: Successful treatment of microsporidal
keratoconjunctivits with topical fumagillin in
a patient with AIDS. Cornea 1993;
112:261265.
40. Thylefors B: Onchocerciasis on review. Int
41. Greene BM, Taylor HR, Cupp EW, et al:
Comparison of ivermectin and
diethylcarbamazine in the treatment of
onchocerciasis. N Engl J Med 1985;
313:133138.
42. Gendelman D, Blumberg R, Sadun A:
Ocular Loa loa with cryoprobe extraction of
subconjunctival worm. Ophthalmol 1984;
91:300303.
43. Drugs for parasitic infections. Med Lett
1990; 32:23.
44. Barraquer-Somers E, Green WR, Miller NR:
Orbital infection by dirolaria. MD State
Med J 1982; 31:5862.
45. Verma AK: Ocular dracontiasis. Int Surg
1968; 50:508509.
46. Kean BH, Sun T, Ellsworth RM, Eds:
Color atlas/text of ophthalmic parasitology.
New York: Igaku-Shoin Medical Publishers,
Inc; 1991:166.
47. Guillory SL, Zinn KM: Intravitreal
Cysticercus cellulosae: ultrasonographic
CHAPTER 22
REFERENCES
247
48.
49.
50.
SECTION 4
51.
248
and fluorescein angiographic features.

Bull NY Acad Med 1980; 56:655661.
Sekhar GC, Lemke BN: Myocysticercosis:
experience with imaging and therapy.
Ophthalmol 1999; 106:23362340.
Santos R, Chavarria M, Aquirre AE: Failure
of medical treatment in two cases of
intraocular cysticercosis. Am J Ophthalmol
1984; 97:249.
Sekhar GC, Lemke BN: Orbital
cysticercosis. Ophthalmol 1997; 104:1599.
Badir G: Schistosomiasis of the
conjunctiva. Br J Ophthalmol 1946; 30:215.
52. Awadzi K, Dadzie KY, Schulz-Key H, et al:

The chemotherapy of onchocerciasis X. An
assessment of four single dose treatment
regimes of MK-933 (Ivermectin) in human
onchocerciasis. Ann Trop Med Parasitol
1985; 79:63.
53. Johns KJ, Head WS, O Day DM: Corneal
toxicity of propamidine. Arch Ophthalmol
1988; 106:68.
54. Foster CS, Stefayzn M: Intraocular
55. Lindquist TD: Treatment of Acanthamoeba

56. Hay J, Kirkiness CM, Seal DV, et al: Drug
resistance and Acanthamoeba keratitis: the
quest for alternative antiprotozoal
chemotherapy. Eye 1994; 8:555.
57. Illingworth CD, Cook SD: Acanthamoeba
keratitis. Surv Ophthalmol 1998; 42:493.
58. Lam DS, Lyon D, Poon AS, et al:
Polyhexamethylene biguanide (0.02%)
alone is not adequate for treating chronic
Acanthamoeba keratitis. Eye 2000; 14:678.
CHAPTER
23
Corticosteroids in Ophthalmic Practice

Mark B. Abelson and Salim Butrus
Key Features
Corticosteroids are 21-carbon structures synthesized naturally

or synthetically through adenocorticotropic hormone-controlled
conversion of cholesterol
Although their mechanism of acitgon is still enigmatic, it is
known that corticosteroids work at both molecular and cellular
levels
The effectiveness of a corticosteroid is largely determined by
its ability to penetrate the cornea
In ophthalmic practice, corticosteroids are most frequently
used to control post-surgical inflammation. They are also used
to treat symptoms of immune hyperreactivity and to treat
diseases with immune and infectious processses.
The practioner must be vigilant for the onset of ocular side
effects, which can occur with prolonged steroid use.Side
effects most frequently involve the anterior segment, and can
include glaucoma, cataracts, and enhanced bacterial infection.
They can also inhibit corneal epithelial and stromal healing
Corticosteroids (glucocorticoids and mineralocorticoids) are

21-carbon structures that are synthesized by adrenocorticotropic
hormone (ACTH)-controlled conversion of cholesterol in the
adrenal cortex. They can take the form of cortisol, cortisone,
corticosterone, or aldosterone. They can also exist in synthetic
forms such as prednisone, methylprednisolone, dexamethasone,
triamcinolone, betamethasone, medrysone, fluorometholone
(FML), and others.
In 1930, Swingle, Pffner, Hartman, and co-workers prepared
adrenocortical extracts that had a reasonable degree of activity.
In 1935, Kendall rst isolated and characterized cortisone in
the laboratory. In 1942, Reichstein and Shoppee identied the
chemical and crystalline structure of steroids.1 The rst
advantageous clinical result of steroids was reported by Hench
and co-workers in 1949.2 They observed the dramatic effects of
cortisone and ACTH in the treatment of rheumatoid arthritis
and subsequently provoked the interest of many investigators
with remarkable therapeutic applications that extended to other
diseases.
In 1954, Stone and Hechter established that ACTH actually
controls the enzymatic conversion of cholesterol to steroids in
the adrenal cortex through cleavage of the side chain of the cholesterol molecule.3 Later, Haynes took a further step by demonstrating that this conversion is mediated by adenosine 3,5-cyclic
monophosphate (cAMP).4,5
Corticosteroids and ACTH were rst introduced into ocular
therapy by Gordon and McLean in 1950. It was not until 1951,
with the introduction of topical and systemic use of cortisone,
that cortisone acetate was prepared in eye drop, ointment, and
subconjunctival, retrobulbar, and anterior chamber-injection
formulations. In 1952, ocular penetration studies of steroids

started to surface. By that time, modication of chemical structures of cortisone and hydrocortisone led to a series of compounds with better penetration and bioavailability and more
potent antiinflammatory effects. In 1959, 0.1% Decadron eye
drops were introduced for treating ocular inflammation.6 In 1956,
it became clear that inflammation in anterior ocular structures
is best treated with steroid drops and posterior uveitis by oral
therapy. It was quickly recognized that topical therapy minimized systemic side effects, but its ocular side effects began to
be appreciated.
CHEMICAL PROPERTIES AND

STRUCTUREACTIVITY RELATIONSHIPS
Cortisone, the rst steroid used therapeutically for antiinflammatory effect, is a 21-carbon four-ringed structure (Fig. 23.1).
Modication of this structure at different sites changes its biologic potency, transcorneal penetration, and, thus, effectiveness
and side effects.5
Different sites of alterations (Fig. 23.1) result in different
antiinflammatory potency and duration of action of these different compounds (Table 23.1). These modications and alterations can be summarized as follows:
1. Prednisone and prednisolone have, in addition to the basic
nucleus, a 1,2 double bond in ring A (Fig. 23.1b). This
FIGURE 23.1. The cortisol nucleus (a). Note the sites where different
chemical groups are added to form compounds with different
antiinflammatory potency. Prednisolone (b); dexamethasone (c);
triamcinolone (d).
249
TABLE 23.1 Classication of Glucocorticoids

Biologic
Half-Life (h)
Antiinflammatory
Effect (h)
Cortisol
812
Cortisone
812
0.8
Corticosterone
812
0.3
Natural Steroids
SECTION 4
Synthetic Steroids
Prednisone
1236
Prednisolone
1236
6-Methylprednisolone
1236
Triamcinolone
1236
9-Fluorocortisol
1236
10
Paramethasone
3672
10
Betamethasone
3672
25
Dexamethasone
3672
25
2.
3.
4.
5.
6.
7.
modication increases their carbohydrate-regulating

potency and prolongs their metabolism compared with
cortisol.
Methylation of carbon 6 in ring B leads to 6a-methyl
prednisolone. This compound has slightly greater
antiinflammatory effect than prednisolone.
Fluorination at a 9a-position in ring B, as in
fluorocortisone (9a-fluorocortisol) enhances its
antiinflammatory property.
11-Desoxycortisol has an oxygen function at the C-11 site
of ring C, augmenting its antiinflammatory activity.
Methylation or hydroxylation at site 16 in ring D
eliminates the sodium-retaining effects and has only a
slight effect on the antiinflammatory potency.
In ring D, 17a-hydroxylation is present in most of the
antiinflammatory steroids.
Most of the active synthetic analogs and all natural
corticosteroids have the hydroxyl group attached to carbon
21 in ring D.
MECHANISM OF ACTION, SITE OF

ACTIVITY, AND OPHTHALMIC INDICATIONS
250
Corticosteroids have numerous effects on many stages of inflammation and arms of the immune response. Despite widespread
use, their precise mechanism of action is not well understood.
There is consensus that they work at two levels: molecular and
cellular. At the molecular level corticosteroids freely penetrate cell
membranes and bind to a specic steroid-binding protein receptor
in the cytoplasm, forming a steroidreceptor complex.718 This
complex then moves into the nucleus and binds to chromatin,
signaling the production of messenger RNA and coding for
enzymes and proteins that determine the response of that particular cell to the hormone (Fig. 23.2).5,19
The cytoplasmic steroid-binding receptor has binding sites
that exhibit high afnity for glucocorticoids (e.g., the naturally
occurring cortisol and corticosterone) and synthetic corticosteroids
(e.g., prednisolone, dexamethasone, and triamcinolone).20 In contrast, these receptors have a low afnity for estrogens, androgens,
cortisone, and prednisone. Hence, cortisone and prednisone are
inactive compounds that are activated when transformed to
FIGURE 23.2. Binding of corticosteroid to a receptor and subsequent

entry into the cell cytoplasm and nucleus. This leads to the synthesis
of specic proteins and specic target cell responses.
cortisol and prednisolone. Glucocorticoid receptors have been

identied in the iris, ciliary body, cornea, sclera, trabecular meshwork, and Schlemms canal.2123
These molecular and cellular changes result in steroid-induced
inhibition of all the cardinal signs of inflammation, such as pain,
heat, redness, and edema.13,24 This is achieved through inhibition
of: (1) leukocyte chemotaxis, (2) production of potent chemical
mediators, and (3) function of immunocompetent cells. Corticosteroids have the dual characteristics of being both antiinflammatory and immunosuppressant.25 They accomplish their
antiinflammatory activity through the following mechanisms:
1. Constriction of blood vessels and reduction of vascular
permeability induced by acute inflammation. This
minimizes leakage into the target site of fluid, proteins,
and inflammatory cells.26
2. Stabilization of intracellular lysosomal membranes and
inhibition of the expression of various damaging enzymes;
polymorphonuclear (PMN)-cell degranulation is also
signicantly inhibited.
3. Stabilization of mast cell and basophil membranes is
important in inhibiting the process of degranulation and
subsequent release of histamine (vasoactive amines),
bradykinin, platelet-activating factor (PAF), proteases, and
eosinophilic chemotactic factors (ECFs).
4. Mobilization of PMNs from the bone marrow results in
neutrophilic leukocytosis (Fig. 23.3).27 Corticosteroids
simultaneously prevent adherence of PMNs to the
vascular endothelium, making them less mobile and less
accessible to the site of inflammation.28
5. Suppression of lymphocyte proliferation and lymphopenia.
In small- to moderate-sized doses, corticosteroids more
signicantly affect T lymphocytes. In larger doses,
B lymphocytes are affected as well. Corticosteroids do
not destroy T lymphocytes but rather affect their
redistribution into circulation, concentrating them in the
bone marrow (Fig. 23.4).2931
6. Reduction of circulating eosinophils and monocytes.
7. Inhibition of macrophage recruitment and migration.32,33
Steroids also interfere with the ability of macrophages to
process antigens.
CHAPTER 23
FIGURE 23.3. Schematic effects of corticosteroids on bone marrow

and circulating neutrophils.
Adapted from Nussenblatt RB, Palestine AG: Uveitis: fundamentals and clinical
practice. Chicago, IL: Year Book; 1989.
8. Suppression of broplasia.34
9. Depression of the bactericidal activity of monocytes and
macrophages.
10. Steroids inhibit phospholipase A2, via a protein called
macrocortin, resulting in inhibition of arachidonic acid
degradation and subsequent synthesis of prostaglandins
and leukotrienes by cyclooxygenase and lipoxygenase
pathways (Fig. 23.5).3539
ABSORPTION RATE AND EXCRETION

AFTER OPHTHALMIC DELIVERY
Corticosteroids are readily absorbed by the cornea, conjunctiva,
and sclera. Corneal penetration is a limiting factor for their antiinflammatory effect. The penetration of corticosteroids through
the normal cornea is a complex process in which multiple factors
determine the rate of penetration. In general, these factors are
similar to those governing penetration (i.e., relative solubility in
water and lipid).40,41 Other factors include viscosity, concentration, hydrogen ion concentration (pH), tonicity, condition of the
FIGURE 23.4. Schematic effects of corticosteroids on lymphocytes.

Adapted from Nussenblatt RB, Palestine AG: Uveitis: fundamentals and clinical
practice. Chicago, IL: Year Book; 1989.
corneal epithelium, size of particles in suspension, and addition

of other compounds or vehicles, such as preservatives or methylcellulose. Part of the topically applied corticosteroid can go
through the upper and lower puncti and then through the nasal
mucosal blood vessels into the circulation, where it binds to
globulin and albumin. Eighty percent of circulating cortisol is
bound to a-globulin as transcortin (corticosteroid-binding globulin), an inactive transport complex. A smaller portion is bound
to albumin, and this portion can diffuse into the extravascular
fluid and bathe tissue cells. Synthetic analogs of cortisol do not
compete with it for binding to transcortin. In addition, synthetic
analogs are less bound to albumin, enabling them to diffuse more
completely into the extravascular tissue than cortisol.25
Tritiated dexamethasone applied topically to rabbit eyes was
traced and found in plasma, kidneys, urine, and liver. Systemic
absorption of topical dexamethasone phosphate is considerable:
as much as 2035% of the drug was found systemically in rabbits
24 h after instillation.42,43 Reduction of the double bond in the
1,5-position in the liver and kidney renders the corticosteroid
251
SECTION 4
FIGURE 23.5. Corticosteroids prevent formation of prostaglandins

and leukotrienes through inhibition of phospholipase A2 and release of
arachidonic acid.
inactive. All synthetic analogs of cortisol are metabolized more

slowly by the liver, owing to chemical modications of the
steroid molecule (Fig. 23.1) and the rapid equilibration in blood
and peripheral tissues.
PHARMACOKINETICS
252
Four factors regarding ophthalmic corticosteroids must be considered:20 (1) ocular penetration of the corticosteroid through
the cornea; (2) antiinflammatory potency, topically and once in
the aqueous humor; (3) duration of action; and (4) side effects.
Different routes by which corticosteroids are delivered into the
eye include topical, periocular, oral, parenteral, and intravitreal.
The penetration of corticosteroids is dependent on the cornea
and on the physical and chemical properties of the corticosteroid. The ideal steroid should be biphasic in polarity, because
the cornea contains both hydrophobic and hydrophilic layers.44
Removal of the corneal epithelium reduces the hydrophobic
properties and allows greater penetration by hydrophilic preparations. Particle size may also affect the bioavailability of corticosteroids.45,46 Results suggest that ophthalmic dexamethasone
suspensions can be optimized for bioavailability by using suspensions with the smallest particle possible. Particle size for
prednisolone acetate (<5 mm and 510 mm in diameter), however,
did not affect the degree of corneal penetration. Both fractions
of prednisolone acetate achieved comparable levels in the
aqueous humor.
Topical preparations can take the form of solutions, suspensions, or ointments. Phosphate and hydrochloride preparations
are relatively hydrophilic and thus are water soluble. Acetate
and alcohol derivatives are hydrophobic and fat soluble. Alcohol
preparations possess intermediate hydrophobicity between phosphates and acetates.47 Owing to the respective polarities, phosphates are generally formulated as solutions, whereas acetates are
generally formulated as suspensions and ointments. Acetates,
owing to their hydrophobic nature, appear to penetrate the cornea
to a greater extent than do phosphates.4851
Corticosteroids can also be released from a drug depot placed
on the ocular surface or by iontophoresis. Examples of drug
depots are cotton pledgets52 and collagen shields.53 One advantage of drug depots is the steady, sustained, and slow release of
the corticosteroid over the ocular surface.
Dexamethasone phosphate penetrates into the cornea and
aqueous humor within 10 min. It reaches a peak within 3060 min
and remains inside the eye from several hours to 24 h.42 The
corneal tissue concentration of tritiated dexamethasone alcohol
(Maxidex) reaches 14.79 mg/g of cornea 7.5 min after instillation,
then declines to 1.86 mg/g at 4 h.54
One percent prednisolone phosphate (Inflamase) is a highly
soluble compound with limited lipid solubility. Thus, it traditionally was thought that this compound had limited solubility
through an intact cornea. Its corneal level, however, reaches
10 mg/g, while the aqueous humor concentration reaches 0.5 mg/g,
30 min after instillation. When the corneal epithelium is removed,
the corneal concentration reaches 235 mg/g and in aqueous
humor 17 mg/g.55
It has been shown that 1.1% tritiated dexamethasone
phosphate instilled into rabbit eyes reaches the aqueous humor.
Its major metabolite in the anterior chamber is 9a-fluoro-11bhydroxy-16a-methyl-1,4-androstadiene-3,17dione. Ocular penetration of corticosteroids is better when they are injected
subconjunctivally than when they are instilled. Hydroxycortisone is found in the anterior chamber almost immediately after
subconjunctival injection. Its degree of penetration is not related
to external factors such as lid movements or tear volume. It is
usually injected near the site of inflammation to obtain maximal
antiinflammatory benets.
ANTIINFLAMMATORY EFFECTS OF
TOPICAL OPHTHALMIC
CORTICOSTEROIDS
Any attempt to compare the inflammatory potency of different
ophthalmic corticosteroids should take into account the following considerations: (1) type of corticosteroid, (2) formulation,
(3) concentration, and (4) what model of inflammation is used.
Models of ocular inflammation in animals and humans are difcult to design and standardize, and some do not reflect clinical
action in humans. Furthermore, some current data on corticosteroids have been extrapolated from previous studies conducted on organ systems other than the eye.56
Studies by Leibowitz, Kupferman, and Cox involved measuring
decreased radioactivity of radiolabeled neutrophils in a rabbit
keratitis model induced by injection of clove oil.5767 This research
focused on the comparison of the sodium phosphate, alcohol,
and acetate derivatives of dexamethasone and prednisolone
(Table 23.2). Data indicated that after a given period, corneal
drug concentration administered with the corneal epithelium
intact was highest with prednisolone acetate, followed by prednisolone sodium phosphate and dexamethasone alcohol suspension; no dexamethasone sodium phosphate was absorbed. With
a denuded epithelium the highest concentration was achieved
with the prednisolone sodium phosphate solution, followed by
the dexamethasone sodium phosphate solution, prednisolone
acetate, and last, the dexamethasone alcohol suspension. With
intact and denuded epithelium, the drug concentrations in the
aqueous humor followed the same pattern. The results with
denuded epithelium may more accurately represent the clinical
situation in keratitis.5760
With an intact epithelium but in the presence of intraocular
inflammation (i.e., experimentally induced anterior or posterior
uveitis), prednisolone acetate concentration in the cornea was
highest with the sodium phosphate solutions of prednisolone and
dexamethasone equivalent, and was least with the dexamethasone acetate and alcohol. Concentrations of prednisolone acetate
suspension and the sodium phosphate solution were equivalent
TABLE 23.2 Comparison of Different Topical Corticosteroids in Suppressing Rabbit Corneal Inflammation
% Decrease
Prednisolone acetate 1%
51
Prednisolone acetate 1%
53
Dexamethasone alcohol 0.1%
40
Dexamethasone alcohol 0.1%
42
Prednisolone sodium phosphate 1%
28
Prednisolone sodium phosphate 1%
47
Fluoromethalone alcohol 0.1%
31
Fluorometholone alcohol 0.1%
37
Dexamethasone sodium phosphate 0.1%
19
Dexamethasone sodium phosphate 0.1%
22
Dexamethasone sodium phosphate ointment 0.05%
13
in the aqueous humor in the eye inflamed with uveitis, followed

by the dexamethasone solution, and last, the dexamethasone
alcohol suspension. Thus, with intraocular inflammation, for
which the highest concentration of drug is most desirable in the
aqueous humor, it is interesting that there was no difference
between the prednisolone acetate suspension and the sodium
phosphate solution.64
Leibowitz and Kupferman also evaluated these steroid derivatives for antiinflammatory potency in a model of corneal inflammation. A signicant increase in antiinflammatory effect was
noted with prednisolone acetate compared with the sodium phosphate solution when evaluated with the corneal epithelium intact.
When the corneal epithelium was absent there was no signicant
difference between the two64,66 or dexamethasone alcohol. Thus,
when a break in the corneal epithelium is associated with corneal
inflammation, the greater absorption of the sodium phosphate
solution equilibrates their relative potency. The dexamethasone
sodium phosphate solution was clearly signicantly inferior with
the epithelium intact or absent.64,66
Changing the concentration and dosing frequency of a particular steroid obviously changes its antiinflammatory potency.
Increasing the concentration of prednisolone acetate from 0.125%
to 1% produces a signicant increase in its corneal concentration58
and antiinflammatory effectiveness.63,64
The concentrations in the cornea and aqueous humor of the
corticosteroid, and thus their antiinflammatory potency, depend
to a large extent on the frequency of instillation. For example,
hourly instillation of 1% prednisolone acetate produces much
more effective suppression of corneal inflammation than does
instillation every 4 h (Table 23.3).65
The ocular bioavailability of topical prednisolone preparations
has been further investigated. One criticism of the clove oil model
used by Leibowitz and others is that the oil alters the absorption
of water-soluble drugs in favor of water-insoluble drugs because
TABLE 23.3 Dosage Schedules and Antiinflammatory

Effectiveness of Topical Prednisolone Acetate 1%
Regimen
Total Doses
Delivered (No.)
Decrease in Corneal
Inflammation (%)
1 drop q 4 h
11
1 drop q 2 h
10
30
1 drop q 1 h
18
51
1 drop q 30 min
34
61
1 drop q 15 min
66
68
1 drop each eye

for 5 min every h
90
72
Corneal Epithelium Absent
% Decrease
of the oil barrier in the stroma after injection. A pharmacokinetic model of absorption of water-insoluble drugs, such as
prednisolone acetate, and water-soluble drugs, such as prednisolone phosphate, was used to compare the drug elimination
rate in the precornea and anterior chamber, the rate of drug dissolution, the rate of drug penetration in the cornea, and the rate
of drug transport into the aqueous humor. In this mathematical
model, the two forms of prednisolone had similar absorption
capacity.55 Similar bioavailability was also found in a rabbit eye
model in vivo when prednisolone phosphate, acetate, and their
metabolite, prednisolone, were directly quantitated in aqueous
humor by reverse-phase high-performance liquid chromatography
(HPLC).68,69
In light of the fact that the acetate and phosphate forms may
actually be equivalent under optimal conditions of dissolution,
the drawbacks of using a suspension in clinical practice may be
the deciding factor in determining which is superior. Suspensions
need to be shaken, and if particles are not evenly distributed,
incorrect doses may be removed from the bottle. Patient compliance for shaking suspension eye drops has been reported to
be poor.70 The risks of incorrect dosing and sudden cessation of
steroid administration are well known.71,72 The difculty of predicting a steroid concentration in suspension drops suggests that
the consistent dosing provided by solutions may be superior.
Two weaker topical corticosteroids are also available for ocular
use. FML 0.1% and 0.25% suspensions have much less corneal
penetration73 than prednisolone but do have moderate antiinflammatory effects.74 Surprisingly, the lower concentration of
0.1% FML acetate has a therapeutic effect comparable to 1% prednisolone in alleviating corneal (but not intraocular) inflammation.
Lower ocular levels are required to produce a substantial therapeutic effect in the cornea. FML has mildly hydrophilic properties, concentrating in the corneal epithelial layer and reaching
saturation levels before passing on through the hydrophilic layers
of the stroma. This may explain why FML penetrates the cornea
in comparatively low concentrations, yet produces moderate but
effective suppression of corneal inflammation.74
Medrysone (HMS) is another relatively weak corticosteroid.
It comes in a 1% suspension and, owing to its weak effect on the
cornea, is used only for minor conjunctival inflammation.
Loteprednol etabonate is a novel soft steroid that was designed
to improve the benet/risk ratio of topical corticosteroid therapy.
Its molecular structure is a modication of prednisolone (see
Fig. 23.1b), where a labile ester function occupies the 17-position
and a stable carbonate group occupies the 17-position. The soft
drug undergoes rapid hydrolysis in the anterior chamber to the
inactive 17-carboxylic acid derivative after it penetrates the
cornea.75 In animals it was shown to retain its antiinflammatory
effects in the cornea,76 and in one study in humans, it was shown
to be useful in treating giant papillary conjunctivitis.77 In recent
years, loteprednol ophthalmic solution has been investigated for
use in treating inflammation due to keratoconjunctivitis sicca
CHAPTER 23
Corneal Epithelium Intact
253

in patients with delayed tear clearance78 and has been approved
for treating seasonal and perennial allergic conjunctivitis.79 It is
also used to treat inflammation after cataract surgery.80
Rimexolone is another soft steroid with decreased propensity
to raise IOP.81 The corticosteroid is indicated for the treatment
of postoperative inflammation following cataract surgery and
for treatment of anterior uveitis, and is commercially available
as a 1% ophthalmic suspension (Vexol). In a study consisting of
197 patients who had undergone cataract extraction, rimexolone
1% was signicantly more effective than placebo in reducing
postoperative inflammation.82 The degree of improvement with
rimexolone was comparable to that of bethamethasone.83
Corticosteroids are also available as ointments. Although ointments increase contact time between the drug and the ocular
surface, it has been shown that dexamethasone phosphate ointment allows less drug absorption in the cornea and anterior
chamber than the solution form. This may be because the ointment forms a barrier, preventing rapid release of the drug into
the tears.66 In the case of FML, it was shown that FML crystals
suspended in water or ointment both produced similar concentrations of drug in the aqueous humor, possibly because the tear
lm is oversaturated by microcrystals of the dissolved drug.84
Corticosteroids may also be injected into parts of the eye.
Supratarsal injection of corticosteroids has been investigated to

treat refractory vernal keratoconjunctivitis.85 All patients experienced dramatic symptomatic relief within 15 days, regardless
of type of corticosteroid injected.
OPHTHALMIC INDICATIONS FOR

CORTICOSTEROID THERAPY86
Since corticosteroids were rst reported effective in the treatment
of rheumatoid arthritis more than 50 years ago, they have become
the most widely used antiinflammatory and immunosuppressant
agents in medicine and ophthalmology. It is estimated that more
than 5 million patients are treated with corticosteroids yearly.
The antiinflammatory and antiallergic activities of corticosteroids
are the most important reason for their clinical use in ocular
disease. Table 23.4 lists the ophthalmic indications for corticosteroid treatment as primary or adjunctive therapy. Some of
these indications are isolated inflammatory conditions and some
are part of a multisystem process. It must be remembered that
the antiinflammatory and immunosuppressive qualities of corticosteroids are nonspecic, palliative, and never curative.
The use of steroids in clinical ophthalmic practice may be
divided into three classes of therapy: (1) posttraumatic control
TABLE 23.4 Some Indications for the Use of Corticosteroids in Ocular Disease
SECTION 4
Conjunctivitis
Allergic (hay fever, vernal, atopic GPC)
Iridocyclitis
Viral (EKC, herpes zoster)
Posterior uveitis
Chemical burns
Sympathetic ophthalmia
Cicatricial pemphigoid
VogtKoyanagiHarada syndrome
Mucocutaneous inflammation
Pars planitis
(StevensJohnson, graft vs host disease,

toxic epidermal necrosis)
Endophthalmitis
Keratitis
Retina
Herpes zoster
Vasculitides
Disciform herpes simplex
Choroiditis
Interstitial keratitis (syphilis, herpes simplex)
Retinitis
Immune inltrates (Staphylococcus, herpes,

varicella, contact lens, EKC, leukemia)
Cystoid macular edema
Peripheral ulcerative (connective tissue

disease, e.g., Wegeners granulomatosis,
polyarteritis nodosa)
Moorens ulcer
Reiters, Lyme disease, sarcoid
Corneal graft rejection
Post-refractive surgery (DLK)
Dry eyes
Trauma and Postsurgery
Juvenile xanthogranuloma
Hemangioma
Lids
Blepharitis
Atopic dermatitis
Discoid lupus
254
Uvea
Chalazion
Acute retinal necrosisOptic

Nerve
Optic neuritis
Temporal arteritis
Orbit
Graves orbitopathy
Pseudotumor
Extraocular Muscles
Myositis
Myasthenia gravis
Sclera
Epscleritis
Scleritis
SIDE EFFECTS OF TOPICAL

CORTICOSTEROID THERAPY
Corticosteroid-induced side effects are either systemic or ocular,
or both. Systemic side effects are most often associated with
oral or parenteral corticosteroid therapy. It has been shown that
6 weeks of treatment with topical 0.1% dexamethasone sodium
phosphate caused suppression of the adrenal cortex, reflected in
a decrease in serum cortisol levels. Systemic absorption of steroids
after topical treatment is actually considerable, and, if given to
a patient with hay fever, it may improve systemic symptoms and
decrease the blood eosinophil count. Potential systemic complications of corticosteroid therapy88 are included in Table 23.5.
Since topical corticosteroids are the most widely used drug in
the treatment of many ocular conditions, their ocular toxicity
and side effects should always be recognized. The patient must
be aware of these side effects, particularly if corticosteroids are
to be used for an extended period. Ocular side effects involve
mainly the anterior segment, including the cornea, conjunctiva,
trabecular meshwork, anterior chamber, and iris (Table 23.6).
Topical corticosteroids may cause glaucoma or cataracts,
enhance secondary herpetic or bacterial infections of the ocular
surface, or inhibit corneal epithelial and stromal healing,
resulting in further corneal melting and perforation. All of these
potential ocular complications of prolonged corticosteroid
therapy can be devastating and threaten vision.
TABLE 23.5 Systemic Complications of Corticosteroid Therapy

Musculoskeletal
Myopathy
Osteoporosis, vertebral compression fractures
Aseptic necrosis of bone
Gastrointestinal
Peptic ulcer (often gastric)
Gastric hemorrhage
Intestinal perforation
Pancreatitis
Central Nervous System
Psychiatric disorders
Pseudotumor cerebri
Ophthalmic
Glaucoma
Posterior subcapsular cataracts
Cardiovascular and Renal
Hypertension
Sodium and water retention edema
Hypokalemic alkalosis
Metabolic
Precipitation of clinical manifestations, including ketoacidosis,
diabetes mellitus
Hyperosmolar nonketotic coma
CHAPTER 23
of inflammation after surgery; (2) abnormalities of excessive

immunoreactivity; and (3) for diseases that have combined
immune and infectious processes. Control of postoperative
inflammation is certainly the indication for which steroids are
used most.
The second group of conditions for which steroids are used is
disorders of immune hyperreactivity. The immune system can
cause damage with overzealous defense mechanisms which can
lead to permanent tissue impairment. These disorders include
iritis, posterior uveitis, immune inltrates, allergic disorders, such
as allergic conjunctivitis, atopic and vernal keratoconjunctivitis,
and graft rejection.
The third class of disorders treated with steroids may originate
with an infectious process. Disorders such as disciform herpes
and bacterial corneal ulcers are treated very cautiously and judiciously with steroids, whereas the infection is treated or controlled with antibiotics. It must be recognized that even in the
absence of an infectious agent, whenever complete immunosuppression is established by the use of steroids, prophylactic
antimicrobial therapy should be considered. The sensitivity of
treating such serious problems with steroids must be emphasized, because often only certain phases of these diseases respond
to steroids, and in other phases steroids may be contraindicated.
For a complete discussion of medical treatments the reader
should refer to specic diseases.
In general, steroids are at rst administered in medium- or
large-size doses to adequately suppress inflammation. The dose
is then tapered gradually to prevent rebound inflammation. Often
the physician can gain insight into the amount and severity of
inflammation by observing the patients response to steroids.
The potential usefulness of prophylactic therapy with
steroids or of a loading, pretreatment period needs to be
established. These are commonly recommended courses of
treatment with systemic steroids. We have shown in the
allergen challenge model that a 48-h loading period was needed
to achieve efcacy in inhibiting the signs and symptoms of
ocular allergy.87 Loading periods are considered the standard for
nonsteroidal antiinflammatory agents, yet steroids are not
commonly used like this in the preoperative period. Further
investigation is needed to clarify this issue.
Hyperlipidemia
Centripetal obesity
Endocrine
Growth failure
Secondary amenorrhea
Suppression of hypothalamicpituitaryadrenal system
Inhibition of Fibroplasia
Impaired wound healing
Subcutaneous tissue atrophy
Suppression of the Immune Response
Superimposition of a variety of bacterial, fungus, and viral
infections in steroid-treated patients
TABLE 23.6 Ocular Side Effects of Corticosteroid Therapy

Cataracts
Glaucoma
Secondary infection
Retardation of wound healing
Uveitis
Mydriasis
Ptosis
Exophthalmos
Pseudotumor cerebri
255

The generalized effect of steroids on the delay of wound healing is important to consider, both in postoperative therapy and
in association with epithelial and stromal defects. The steroids
effect on the broblast results in delayed collagen synthesis,
which can cause or exacerbate corneal melting.34,72
SECTION 4
CATARACT INDUCTION BY
CORTICOSTEROIDS
Several years after corticosteroids became widely used for rheumatoid arthritis, Black and co-workers89,90 reported the development of cataracts in patients receiving long-term systemic therapy.
The dosage and duration of steroid therapy correlated with the
incidence of posterior subcapsular cataract (PSC) formation.
Seventy-ve percent of patients who receive more than 16 mg/day
of prednisone develop cataracts. If the dose is decreased to
10 mg/day for 1 year, the chance of PSC formation is minimal.
Individuals who have undergone prolonged topical corticosteroid
therapy, such as for vernal or atopic keratoconjunctivitis or
those who received corneal transplantation for keratoconus, are
under the threat of developing PSCs. Donshik and co-workers
have shown that 28 eyes of 86 transplanted for keratoconus
developed PSCs after 1 year of 0.1% dexamethasone therapy.91
It seems that PSC formation is signicantly related to the total
cumulative steroid dose and the total time that steroids were
administered. Once PSCs have developed, cessation of corticosteroids does not resolve the opacity. It is also important to
consider the overall status of the patient, because factors such
as diabetes appear to increase susceptibility to these complications of topical steroid administration. The pathogenesis of
corticosteroid-induced cataract formation has not been fully
explained. One theory holds that corticosteroids enter the lens
and bind to its bers, leading to biochemical changes and protein
aggregation in the cells.
STEROIDS AND GLAUCOMA

Corticosteroids have been shown to produce increased intraocular pressure when applied topically to the eye9299 or given
systemically.100,101 This elevation in intraocular pressure is usually reversible but can lead to optic nerve damage and visual
eld changes similar to those seen in patients with chronic
open-angle glaucoma. The genetic basis for this predisposition
is probably a recessive homozygous gene. Although the exact
mechanism of corticosteroid-induced glaucoma is not clear, there
is evidence of mucopolysaccharide deposition in the trabecular
meshwork.102 Identifying the effects of topical application of
0.1% dexamethasone has no predictive value.103
Steroids such as FML, which has limited intraocular bioavailability, have been shown to have less tendency toward induction
of ocular hypertension.104108 In a corticosteroid provocative
test, Akingbehin found that 15 of 24 eyes treated with 0.1%
dexamethasone showed a rise in intraocular pressure of more
than 5 mmHg, whereas only two of the 24 eyes treated with 0.1%
FML showed such an increase.104 In a study of 14 steroid
responders to 0.1% dexamethasone, 13 were not affected by subsequent treatment with 0.1% FML.105 Also, the time to an evoked
ocular hypertension in known steroid responders was signicantly longer (4 weeks) for 0.1% FML actetate than for 0.1%
dexamethasone sodium phosphate.106 Cantrill and associates
showed that 0.1% dexamethasone had more than three times
the ocular hypertensive effect of 0.1% in corticosteroid
responders.107 Mean intraocular pressure increases were also
signicantly lower with twice the concentration of FML (0.25%)
compared with 0.1% dexamethasone sodium phosphate-treated
eyes in known steroid responders who took the drugs four times
daily for as long as 6 weeks.104
There has been much investigation into the development of
steroids that do not elicit ocular hypertension and glaucoma.
Lodoprednenolol, a steroid developed using the soft drug concept, is an inactive compound that is activated locally in the eye
and is degraded in the bloodstream, thus limiting systemic
activity.76,77 It has been proposed that the side chain responsible
for the steroid ocular hypertensive response is absent from this
compound; however, most research into the structureactivity
relationships of steroids has shown that a steroids antiinflammatory activity is closely related to its ocular hypertensive
activity.109
INFECTIONS ENHANCED BY
CORTICOSTEROIDS
For bacterial, viral, and protozoal ocular infections, use of corticosteroids should always be given careful consideration. Corticosteroids substantially suppress the activation and migration
of leukocytes, which is a major part of the cellular host defense
against invading microorganisms and infection. Secondary infections caused by corticosteroids can take the form of bacterial
conjunctivitis and keratitis, viral keratitis, or more serious visionthreatening infections, such as fungal keratitis, fungal endophthalmitis, and toxoplasmic chorioretinitis. Management of these
complications involves tapering, and eventually stopping, the
corticosteroid and initiating therapy with appropriate antiinfective agents. Prophylactic coverage with appropriate antiviral or
antibacterial agents should be considered.110
REFERENCES
256
1. Reichstein T, Shoppee CW: The hormone

of the adrenal cortex. Vitam Horm 1943;
1:346.
2. Hench PS, Kendall EC, Slocumb CH, et al:
The effect of a hormone of the adrenal cortex
(17-hydroxy-11-dehydrocorticosterone;
compound E) and of pituitary
adrenocorticotropic hormone on
rheumatoid arthritis. Proc Staff Meet Mayo
Clin 1949; 24:181197.
3. Stone D, Hechter O: Studies on ACTH
action in perfused bovine adrenals: the site
of action of ACTH in corticosteroidogenesis.
Arch Biochem Biophys 1954; 51:457469.
4. Haynes RC Jr, Koritz SB, Peron FG:
Influence of adenosine 3,5-monophosphate
on corticoid production by rat adrenal
glands. J Biol Chem 1959; 234:14211423.
5. Haynes RC Jr, Murad F: Adrenocorticotropic

hormone: adrenocortical steroids and their
synthetic analogs. Inhibitors of
adrenocortical steroid biosynthesis. In:
Gilman AG, Goodman LS, Rall TW, Murad F,
eds. Goodman and Gilmans the
New York: MacMillan; 1970.
6. Gordon DM: Use of dexamethasone in eye
disease. JAMA 1960; 172:311312.
7. Ballard PL, Baxter JD, Higgins SJ, et al:
General presence of glucocorticoid
receptors in mammalian tissues.
Endocrinology 1974; 94:9981002.
8. Baxter JD, Rousseau GG, eds.
Glucocorticosteroid hormone action. Berlin:
Springer; 1979.
9. Cake MH, Litwack G: The glucocorticoid
10.
11.
12.
13.
receptor in biochemical action of hormones.

In: Litwack G, ed. Biochemical actions of
hormones. New York: Academic Press; 1975.
Feldman D: The role of hormone receptors
in the action of adrenal steroids. Annu Rev
Med 1975; 26:8390.
Higgins SJ, Gehring U: Molecular
mechanisms of steroid hormone action.
Adv Cancer Res 1978; 28:313397.
Wicks WD: The mode of action of
glucocorticoids. In: Rickenberg HV, ed.
Biochemistry of hormones. Baltimore, MD:
University Park Press; 1974.
Fahey JV, Guyre PM, Munck A:
Mechanisms of anti-inflammatory actions
of glucocorticoids. In: Weissmann G.
Advances in inflammation research.
New York: Raven; 1981.
14. Parrillo JE, Fauci AS: Mechanisms of

glucocorticoid action on immune
processes. Annu Rev Pharmacol Toxicol
1979; 19:179201.
15. Fauci AS: Mechanisms of the
immunosuppressive and antiinflammatory
effects of glucorticosteroids.
Immunopharmacology 1978; 1:1.
16. OMalley BW: Mechanisms of action of
steroid hormones. N Engl J Med 1971;
284(7):370377.
17. Baxter JD, Forshaim PH: Tissue effects
of glucocorticoids. Am J Med 1972;
53:573589.
18. Thompson EB, Lippman ME: Mechanism of
action of glucocorticoids. Metabolism
1974; 23:159202.
19. Hallahan C, Young DA, Munck A: Time
course of early events in the action of
glucocorticoids on rat thymus cells in vitro.
J Biol Chem 1973; 248:2922.
20. Leopold IH, Gaster RN: Ocular inflammation
and anti-inflammatory drugs. In:
Kaufman HE, Barron BA, McDonald MB,
Waltman SR, eds. The cornea. New York:
Churchill Livingstone; 1988.
21. Hernandez MR, Wenk EJ, Weinstein BI,
et al: Glucocorticoid target cells in human
outflow pathway: autopsy and surgical
specimens. Invest Ophthalmol Vis Sci
1983; 24:16121616.
22. Southren AL, Dominguez MO, Gordon GG,
et al: Nuclear translocation of cytoplasm
glucocorticoid receptor in the iris-ciliary
body and adjacent corneoscleral tissue of
the rabbit following topical administration
of various glucocorticoids. Invest
23. Mondino BJ, Aizuss DH, Farley MK:
Steroids. In: Lamberts DW, Potter DE, eds.
Clinical ophthalmic pharmacology. Boston,
24. Jasami MK: Anti-inflammatory steroids:
mode of action in rheumatoid arthritis and
homograft rejection. In: Vane JR, Ferriera
SH, eds. Antiinflammatory drugs. Berlin:
Springer; 1979.
25. Melby JC: Systemic corticosteroid therapy:
pharmacology and endocrinologic
considerations. Ann Intern Med 1974;
8:505512.
26. Greeson TP, Levan NE, Freedman RI, et al:
Corticosteroid-induced vasoconstriction
studied by xenon clearance. Invest
Dermatol 1973; 61:242244.
27. Bishop CR, Athens JW, Boggs DR, et al:
Leukokinetic studies. XIII. A non steadystate kinetic evaluation of the mechanism
of cortisone-induced granulocytosis. J Clin
Invest 1967; 47:249260.
28. Friedlaender MH: Corticosteroid therapy of
ocular inflammation. Int Ophthalmol Clin
1983; 23:175182.
29. Cohen JJ: Thymus-derived lymphocytes
sequestered in the bone marrow of
hydrocortisone treated mice. J Immunol
1972; 108:841844.
30. Smolin G: Immunology. In: Smolin G,
Thoft RA, eds. The cornea. Scientic
foundations and clinical practice. Boston,
31. Claman HN: Corticosteroids and
lymphoid cells. N Engl J Med 1972;
287:388397.
32. Werb Z: Biochemical actions of
glucocorticoids on macrophages in culture.
J Exp Med 1978; 147:16951712.
33. Dannenberg AM Jr: The antiinflammatory

effects of corticosteroids. Inflammation
1979; 3:329343.
34. Ashton N, Cook C: Effect of cortisone on
healing of corneal wounds. Br J
Ophthalmol 1951; 35:708717.
35. Vane J, Botting R: Inflammation and the
mechanism of action of antiinflammatory
drugs. Endocr Rev 1984; 5:8996.
36. Floman N, Zor U: Mechanism of steroid
action in ocular inflammation. Invest
37. Kanrowitz F, Robinson DR, McGuire MB,
et al: Corticosteroids inhibit prostaglandin
production by rheumatoid synovia. Nature
1975; 258:737739.
38. Gryglewski RJ: Steroid hormones,
antiinflammatory steroids and prostaglandins.
Pharmacol Res Commun 1976; 8:337340.
39. Blackwell GJ, Carnuccio R, DiRosa M, et al:
Macrocortin in polypeptide causing the
antiphospholipase effect of glucocorticoids.
Nature 1980; 287:147149.
40. Gardner SK: Ocular drug penetration and
pharmacokinetic principles. In: Lamberts DW,
Potter DE, eds. Clinical ophthalmic
pharmacology. Boston, MA: Little, Brown;
1987.
41. Maurice DM: Factors influencing the
penetration of topically applied drugs. In:
Holly FJ, ed. Clinical pharmacology of the
anterior segment. Boston, MA: Little,
Brown; 1980.
42. Rosenblum C, Dengler RE, Geoffroy RF:
Ocular absorption of dexamethasone
phosphate disodium by the rabbit. Arch
Ophthalmol 1967; 77:234237.
43. Roters S, Aspacher F, Diestelhorst M: The
influence of dexamethasone 0.1% eye
drops on plasma cortisol and ACTH
concentrations after cataract surgery.
44. Kupferman A, Pratt MV, Camon EJ:
Topically applied steroids in corneal
disease. III. The role of drug derivative in
stromal absorption of dexamethasone.
45. Bisrat M, Glazer M: Effect of particle size on
ocular permeability of prednisolone acetate
in rabbits. Acta Pharm Nord 1992; 4:58.
46. Schoenwald RD, Stewart P: Effect of
particle size on ophthalmic bioavailability
of dexamethasone suspensions in rabbits.
J Pharm Sci 1980; 69:391394.
47. Gardner S: Comparison of topical
ophthalmic corticosteroid drops. Ocular
Ther Man 1992; 3:114.
48. Leibowitz HM, Kupferman A: Bioavailability
and therapeutic effectiveness of topically
administered corticosteroids. Trans Am
79:7888.
49. Leibowitz HM, Kupferman A:
Antiinflammatory effectiveness in the
cornea of topically administered
prednisolone. Invest Ophthalmol 1974;
13:757766.
50. Leibowitz HM, Kupferman A: Use of
corticosterioids in the treatment of corneal
inflammation. In: Leibowitz HM, ed.
Corneal disorders: clinical diagnosis and
management. Philadelphia, PA: WB
Saunders; 1984.
51. McGhee CNS, Watson DG, Midgeley JM,
et al: Penetration of synthetic corticosteroids
in human aqueous humor. Eye 1990;
4:526530.
52. Katz IM, Blackman WM: A soluble

sustained-release articial ophthalmic
delivery unit. Am J Ophthalmol 1977;
83:728734.
53. Hwang DG, Stern WH, Hwang PH, et al:
Collagen shield enhancement of topical
dexamethasone penetration. Arch
Ophthalmol 1989; 107:13751800.
54. Short C, Keates RH, Donovan EF, et al:
Ocular penetration studies. I. Topical
administration of dexamethasone. Arch
Ophthalmol 1966; 75:689692.
55. Olejnick O, Weisbecker CA: Ocular
bioavailability of topical prednisolone
preparations. Clin Ther 1990;
12:211.
56. Havener WH: Corticosteroid therapy in
ocular pharmacology. 5th edn. St Louis,
MO: CV Mosby; 1983.
57. Cox WV, Kupferman A, Leibowitz HM:
Topically applied steroids in corneal
disease. I. The role of inflammation in
stromal absorption of dexamethasone.
58. Kupferman A, Leibowitz HM: Topically
applied steroids in corneal disease. IV.
The role of drug concentration in stromal
absorption of prednisolone acetate. Arch
Ophthalmol 1974; 91:377380.
applied steroids in corneal disease. V.
Dexamethasone alcohol. Arch Ophthalmol
1974; 92:329330.
applied steroids in corneal disease. VI.
Kinetics of prednisolone phosphate. Arch
Ophthalmol 1974; 92:331334.
61. Leibowitz HM, Less JH, Kupferman A:
Quantitation of inflammation in the cornea.
62. Leibowitz HM, Stewart RH, Kupferman A,
et al: Evaluation of dexamethasone acetate
as a topical ophthalmic formulation. Am J
Ophthalmol 1978; 86:418423.
63. Leibowitz HM, Kupferman A:
Antiinflammatory effectiveness in the cornea
of topically administered prednisolone.
64. Kupferman A, Leibowitz HM:
Antiinflammatory effectiveness of topically
administered corticosteroid in the cornea
without epithelium. Invest Ophthalmol
1975; 14:252258.
65. Leibowitz HM: Management of
inflammation in the cornea and conjunctiva.
Acad Ophthalmol 1975; 79:7887.
67. Leibowitz HM, Kupferman A: Kinetics of
topically administered prednisolone acetate
optimal concentration for treatment of
inflammatory keratitis. Arch Ophthalmol
1976; 94:13871389.
68. Musson DG, Bidgood AM, Olejnick O:
Assay methodology for prednisolone,
prednisolone acetate and prednisolone
sodium phosphate in rabbit aqueous
humor and ocular physiological solutions.
J Chromatogr 1991; 565:89102.
69. Musson DG, Bidgood AM, Olejnick O: An in
vitro comparison of the permeability of
prednisolone, prednisolone sodium
phosphate, and prednisolone acetate
across the NZW rabbit cornea. J Ocul
Pharmacol 1992; 8:139150.
CHAPTER 23
257
SECTION 4
258
70. Apt L, Henrick A, Silverman LM: Patient

compliance with the use of topical
ophthalmic corticosteroid suspensions.
71. Burch PG, Migeon CJ: Systemic absorption
of topical steroids. Arch Ophthalmol 1968;
79:174176.
72. Aronson SB, Moore TE Jr: Corticosteroid
therapy in central stromal keratitis. Am J
Ophthalmol 1969; 67:873896.
73. Leibowitz HM, Kupferman A: Penetration of
fluorometholone into the cornea and
aqueous humor. Arch Ophthalmol 1975;
93:425427.
74. Kupferman A, Leibowitz HM: Therapeutic
effectiveness of fluorometholone in
inflammatory keratitis. Arch Ophthalmol
1975; 93:10111014.
75. Druzgala PD, Wu WM, Winwood D, et al:
Ocular absorption and distribution of
loteprednol etabonate: a soft steroid.
32(Suppl):735.
76. Leibowitz HM, Kupferman A, Ryan WJ,
et al: Corneal anti-inflammatory efcacy of
loteprednol etabonate: A new steroidal soft
drug. Invest Ophthalmol Vis Sci 1991;
32(Suppl):735.
77. Laibovitz RA, Ghormley NR, Insler MS,
et al: Treatment of giant papillary
conjunctivitis with loteprednol etabonate:
a novel corticosteroid. Invest Ophthalmol
Vis Sci 1991; 32(Suppl):734.
78. Pflugfelder SC, Maskin SL, Anderson B,
et al: A randomized, double-masked,
placebo-controlled, multicenter comparison
of loteprednol etabonate ophthalmic
suspension, 0.5%, and placebo for
treatment of keratoconjunctivitis sicca in
patients with delayed tear clearance. Am J
Ophthalmol 2004; 138:444457.
79. Ilyas H, Slonim CB, Braswell GR, et al:
Long-term safety of loteprednol etabonate
0.2% in the treatment of seasonal and
perennial allergic conjunctivitis. Eye
Contact Lens 2004; 30:1013.
80. Holzer MP, Solomon KD, Sandoval HP,
Vroman DT: Comparison of ketorolac
tromethamine 0.5% and loteprednol
etabonate 0.5% for inflammation after
phacoemulsication: prospective
randomized double-masked study.
J Cataract Refract Surg 2002; 28:9399.
81. Leibowitz HM, Bartlett JD, Rich R, et al:
Intraocular pressure-raising potential of
0.1% rimexolone in patients responding to
corticosteroids. Arch Ophthalmol 1996;
114(8):933937.
82. Lehmann R, Sil K, Stewart R, et al:
Comparison of rimexolone 1% ophthalmic
suspension to placebo in control of
83.
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
postcataract surgery inflammation.

36(Suppl):S793.
Corboy JM: Corticosteroid therapy for the
reduction of postoperative inflammation
after cataract extraction. Am J Ophthalmol
1976; 82:923927.
Holsclaw DS, Whitcher JP, Wong IG, et al:
Supratarsal injection of corticosteroid in the
treatment of refractory vernal
1996; 121:243249.
Sieg J, Robinson J: Vehicle effects on
ocular drug bioavailability. I. Evaluation of
fluorometholone. J Pharm Sci 1975;
64:931936.
Jaanus SD: Anti-inflammatory drugs. In:
Bartett JD, Jaanus SD, eds. Clinical ocular
pharmacology. 2nd edn. Boston, MA:
Butterworths; 1989.
George MA, Smith LM, Abelson MB:
Efcacy of 1.0% prednisolone sodium
phosphate in alleviating the signs and
symptoms of allergic conjunctivitis induced
by allergen challenge. Invest Ophthalmol
Vis Sci 1991; 32(Suppl):736.
Melby JC: Systemic corticosteroid therapy:
pharmacology and endocrinologic
considerations. Ann Intern Med 1974;
81:505512.
Black RL, Oglesby RB, Von Sallman L,
et al: Posterior subcapsular cataracts
induced by corticosteroids in patients with
rheumatoid arthritis. JAMA 1960;
174:166171.
Spaeth GL, Von Sallmann L:
Corticosteroids and cataracts. Int
Donshik PC, Cavanaugh HD, Boruchoff DA,
et al: Posterior subcapsular cataracts
induced by topical corticosteroids following
keratoplasty for keratoconus. Ann
Francois J: Cortisone et tension oculaire.
Ann Ocul 1954; 187:805816.
Stern JJ: Acute glaucoma during cortisone
therapy. Am J Ophthalmol 1953;
36:389390.
Armaly M: Effect on intraocular pressure
corticosteroids and fluid dynamics. I. The
effect of dexamethasone in the normal eye.
Armaly M: Effect of corticosteroids on
intraocular pressure and fluid dynamics.
II. The effect of dexamethasone in the
glaucomatous eye. Arch Ophthalmol 1963;
70:492499.
Armaly M: Statistical attributes of the
steroid hypertensive response in the
clinically normal eye. Invest Ophthalmol Vis
Sci 1965; 4:187198.
97. Armaly M: Heritable nature of

dexamethasone-induced ocular
hypertension. Arch Ophthalmol 1966;
75:3235.
98. Becker B, Mill SW: Corticosteroids and
intraocular pressure. Arch Ophthalmol
1963; 70:500507.
99. Becker B, Hahn KA: Topical corticosteroids
and heredity in primary open-angle
glaucoma. Am J Ophthalmol 1964;
57:543551.
100. Bernstein HN, Schwartz B: Effects of long
term systemic steroids on ocular pressure
and tonographic values. Arch Ophthalmol
1962; 68:742753.
101. Covell LL: Glaucoma induced by systemic
steroid therapy. Am J Ophthalmol 1954;
45:108109.
102. Hodapp EA, Kass MA: Corticosteroidinduced glaucoma. In: Ritch R, Shields MB,
eds. The secondary glaucomas. St Louis,
MO: CV Mosby; 1982.
103. Johnson DH, Bradley JV, Acott IS: The effect
of dexamethasone on glycosaminoglycans
of human trabecular meshwork in perfusion
organ culture. Invest Ophthalmol Vis Sci
1990; 31:25682571.
104. Akingbehin AO: Comparative study of the
intraocular pressure effects of
fluorometholone 0.1% versus
dexamethasone 0.1%. Br J Ophthalmol
1983; 67:661663.
105. Morrison E, Archer DB: Effect of
fluorometholone (FML) on the intraocular
pressure of corticosteroid responders. Br J
Ophthalmol 1984; 68:581584.
106. Stewart RH, Smith JP, Rosenthal AL:
Ocular response to fluorometholone
acetate and dexamethasone sodium
phosphate. Curr Eye Res 1984; 3:835839.
107. Cantrill HL, Palmberg PF, Zink HA, et al:
Comparison of in vitro potency of
corticosteroids with ability to raise
intraocular pressure. Am J Ophthalmol
1975; 79:10121017.
108. Kass M, Cheetham J, Duzman E, et al: The
ocular hypertensive effect of 0.25%
fluorometholone in corticosteroid responders.
Am J Ophthalmol 1986; 102:159163.
109. McLean JM: Discussion of Woods AC:
clinical and experimental observation on
the use of ACTH and cortisone in ocular
inflammatory disease. Trans Am
Ophthalmol 1959; 48:293296.
110. Stern GA, Buttross M: Use of
corticosteroids in combination with
antimicrobial drugs in the treatment of
infectious corneal disease. Ophthalmology
1991; 98:847853.
CHAPTER
24
Nonsteroidal Antiinammatory Drugs

King W. To, Mark B. Abelson, and Arthur H. Neufeld
Key Features
NSAIDs affect the cyclooxygenase pathway of the arachidonic

acid cascade, and offer varying degrees of anti-inflammatory
and analgesic effects through inhibition of prostaglandins.
Ophthalmic NSAIDs tend to be associated with fewer adverse
events than systemic NSAIDs. The mechanisms through which
the undesired effects appear are uncertain, and may be linked
to concurrent conditions, rather than directly to the NSAID.
Topical NSAIDs can reduce intraoperative miosis during ocular
surgery, thereby increasing the surgeons visualization and
decreasing the risk of complications. Pre-operative use is key
to achieving the NSAIDs full effect.
Given the more favorable side effect profile of NSAIDs, they
are being increasingly used over corticosteroids to control
inflammation after cataract surgery.
NSAIDs are gaining off-label attention for their ability to
prevent and treat cystoid macular edema, which can arise as a
complication of cataract surgery.
Prior to the development of corticosteroids, aspirin was used to

treat intraocular inammation.1 Salicylic acid (orthohydroxybenzoic acid) or aspirin (acetylsalicylic acid) was introduced over a century ago as an antipyretic and for the treatment
of rheumatic fever. Aspirin reduces inammation primarily by
inhibiting the cyclo-oxygenase involved in the production of
prostaglandins2,3 although additional antiinammatory actions
are probably involved. Prostaglandins (PGs) are 20-carbon,
unsaturated fatty-acid derivatives with a cyclopentane ring; these
biologically active lipids have a diverse spectrum of actions,
including the control of the inammatory response, pain, body
temperature, intraocular pressure, blood coagulation, lipid and
carbohydrate metabolism, and cardiovascular, respiratory, and
renal physiology. The PGs are eicosanoids, which are a family
of molecules derived from arachidonic acid. The mechanism of
TABLE 24.1. Ocular Effects of PGs

Prostaglandin
Effect
Stimulates vasodilation and chemosis
E1, E2
Increase inammation
Increase intraocular pressure
Increase capillary permeability
Stimulate vasodilation
Stimulate miosis
F2
Reduces intraocular pressure

Has minimal effect on inammation
Has minimal effect on miosis
action of PGs is not well understood. Some PGs act antagonistically with one another, whereas individual PGs can have
different effects on different tissues. In addition, responses to
certain PGs can vary signicantly in different animal models
and human studies. The ocular effects of PGs that have been
isolated from the eye are summarized in Table 24.1.
In the past 20 years, research has led to the development of
useful aspirin-like, nonsteroidal antiinammatory drugs
(NSAIDs). NSAIDs are among the most commonly prescribed
drugs. Their most useful application is in the management of
inammation in diseases such as osteoarthritis, rheumatoid
arthritis, and ankylosing spondylitis. This chapter provides an
overview of NSAIDs and their ophthalmic applications.
CHEMICAL PROPERTIES
The NSAIDs, a heterogeneous group of compounds, all have
some degree of antiinammatory, antipyretic, and analgesic
properties; however, their therapeutic properties differ signicantly. Because PGs have such a diverse range of actions,
NSAIDs, which inhibit the production of PGs, also possess a
broad range of pharmacologic properties. Systemic NSAIDs at
therapeutic doses can produce adverse changes in the gastrointestinal, respiratory, hepatic, endocrine, coagulation, and renal
systems.4 The NSAIDs can be divided into the following groups:
salicylates, fenamates, and derivatives of indole, pyrazolone,
propionic acid, phenylacetic acid, and oxicam (Table 24.2). Only
the derivatives of indole, propionic acid, and phenylacetic acid
are commercially available as topical ophthalmic agents.
Indocid, a commercial form of ophthalmic indomethacin
solution, currently is not yet available in the United States. Six
Food and Drug Administration (FDA)-approved NSAID topical
ophthalmic agents are currently available (Table 24.3).
MECHANISMS OF ACTION
Arachidonic acid is the primary precursor of PGs, leukotrienes
(LTs), and related compounds (Fig. 24.1). Arachidonic acid may
be ingested or derived from dietary linoleic acid. Arachidonic
acid is bound to phospholipids in the plasma membrane; its
release by phospholipases is closely regulated by a wide variety
of chemical, physical, and hormonal factors. The blockage of
PG biosynthesis by NSAIDs is primarily due to the inhibitory
effects of NSAIDs on cyclo-oxygenase, which is responsible for
the conversion of arachidonic acid to endoperoxides (PG G2,
PG H2) in ocular and nonocular tissues.5 Endoperoxides are
precursors of all other PGs. The inhibitory activity of NSAIDs
on cyclo-oxygenase demonstrably correlates with its antiinammatory activity.3 Experimental studies have shown that certain
PGs are potent mediators of ocular inammation.6,7 Topical
259
TABLE 24.2. Classes of NSAIDs Available in the United States

Generic Name
Trade Name
Salicylates
Aspirin
Multiple names and manufacturers
Fenamates
Mefenamate
Meclofenamate
Ponstel
Meclomen
Indole Derivatives
Indomethacin
Ketorolac
Sulindac
Tolmetin
Indocin
Toradol, Acular*
Clinoril
Tolectin
Pyrazolone Derivatives
Phenylbutazone
FIGURE 24.1. Structure of arachidonic acid cascade; synthesis of

prostaglandins and related compounds.
Butazolidin
Propionic Acid Derivatives

Fenoprofen
Flurbiprofen
Ibuprofen
Ketoprofen
Naproxen
Suprofen
Nalfon
Ansaid, Ocufen*
Advil, CoAdvil, IBU-TAB Medipren, Motrin,
Nuprin, Childrens Motrin, Rufen
Orudis
Naprosyn
Profenal*
SECTION 4
Phenylacetic Acid Derivatives

Diclofenac
Voltaren, Voltaren Ophthalmic*
Oxicam Derivatives
Piroxicam
Feldene
*Ophthalmic topical agents.
TABLE 24.3. Topical Ophthalmic Suspension NSAIDs Available

in the United States
260
Generic Name and

Solution Concentration
Trade Name
(Manufacturer)
Indication(s) for Use

Approved by the FDA
Ketorolac 0.4%
Acular, 0.5%
Acular
LS (Allergan)
1. Seasonal allergic
conjunctivitis
2. Intraocular
inammation after
cataract surgery
3. Reduction of pain
after corneal
refractive surgery
Flurbiprofen 0.03%
Suprofen 1%
Ocufen (Allergan) 1. Minimizing

Profenal (Alcon)
intraoperative
miosis during
cataract surgery
Diclofenac 0.1%
Voltaren
(CibaVision)
1. Intraocular inammation following

cataract surgery
and photophobia
after cataract
surgery
Nepafenac 0.1%
Nevanac (Alcon)
and inammation
associated with
cataract surgery
Bromfenac 0.09%
Xibrom (Ista)
1. Treatment of
postoperative
inammation
following cataract
surgery
application of arachidonic acid or certain PGs produces dilation

of conjunctival vessels with chemosis, changes in intraocular
pressure, and miosis.8 PG levels are elevated in the aqueous
humor following argon laser iridectomy,9 cataract surgery,10 and
trauma.11 By inhibiting cyclo-oxygenase, NSAIDs have been
shown to reduce the de novo synthesis of PGs.1113 Unlike
NSAIDs, corticosteroids affect both the cyclo-oxygenase and
lipoxygenase pathways by preventing the release of arachidonic
acid.14,15 However, NSAIDs do not inhibit lipoxygenase and
may lead to an increase in the production of LTs by increasing
the amount of arachidonic acid available to be metabolized by
lipoxygenase. The additional inhibition of leukotriene formation
may be partially responsible for the greater antiinammatory
activity of corticosteroids. Other sources provide detailed
discussion on the broad spectrum of actions of the PGs
systemically16 and in the eye.17
PHARMACOKINETICS
In general, orally ingested NSAIDs are rapidly absorbed and
distributed throughout most body tissues. The NSAIDs are
bound extensively to plasma proteins, and concentrations peak
in blood 12 h after administration. Biotransformation occurs
primarily in the hepatic endoplasmic reticulum and mitochondria. The unchanged NSAID and its metabolic products
are then eliminated in the urine. Therefore, patients with
underlying liver or kidney dysfunction are at signicant risk for
the development of a wide range of toxic effects from normal
doses of systemic NSAIDs.
COMPLICATIONS
Oral NSAID therapy is associated with a variety of complications. Only the most common and clinically signicant adverse
effects are addressed here. The most common undesirable effect
is gastrointestinal irritation, which can lead to nausea, vomiting,
cramps, and gastric or intestinal ulceration.18,19 Gastrointestinal ulceration can lead to signicant blood loss and anemia. In
addition to the local irritative effects of the NSAIDs on the
gastrointestinal mucosa, inhibition of certain key gastric PGs
(E2, I2) that normally protect against erosion may contribute to
this side effect. The NSAIDs also increase the bleeding time by
inhibiting platelet production of thromboxane A2, a potent
aggregating agent.20 Although NSAIDs do not signicantly
affect renal function in healthy young patients, these aspirinlike drugs can produce acute renal failure in patients with
chronic renal disease, congestive heart failure, cirrhosis with
ascites, volume depletion secondary to diuretics, and hypotension secondary to hemorrhage. PGs protect the kidneys in
disease states when renal perfusion is compromised by stimulating vasodilation and maintaining renal perfusion. NSAIDs
block this PG-mediated compensatory response.21 Therefore, it
is not surprising that NSAIDs may produce renal compromise
in the elderly,22 which is important because the prevalence of
rheumatic disease, in which the treatment of choice is NSAIDs,
increases with age. StevensJohnson syndrome has been
reported in association with rofecoxib (Vioxx).23
Topical NSAIDs generally appear to be signicantly safer than
oral NSAIDs. Application of these topical agents sometimes
causes a stinging sensation. The benets of greater comfort
cannot be overemphasized, because comfort is clearly an
important factor in a patients adherence to a therapeutic
regimen. Topical NSAIDs should be avoided in patients with a
history of aspirin or NSAID sensitivity. Bronchospastic exacerbation was caused by topical ketorolac in a patient with asthma
and nasal polyps.24 Rare corneal complications such as corneal
melting after topical NSAID use have been reported.2529 Topical
diclofenac, ketorolac, and bromfenac have all been associated
with corneal ulceration. The exact mechanisms remain unclear,
but this rare side effect reminds us to carefully observe all our
patients on topical NSAIDS.
Some have suggested that the increased bleeding of ocular
tissues (including hyphemas) in the setting of surgery and
impairment of wound healing is associated with topical NSAID
use.30 In our clinical experience, the potential for increased
bleeding and impairment of wound healing with topical NSAID
use does not seem to be a problem. Whether topical NSAIDs
may be used safely in the presence of fungal, bacterial, or viral
infections remains unclear.
PREVENTION OF INTRAOPERATIVE MIOSIS

Miosis is a well-known complication of surgical trauma. In an
effort to identify the agent responsible for stimulating miosis as
part of the ocular response to trauma, researchers isolated a
substance called irin more than 40 years ago.31,32 Irin, which
was isolated from extracts of iris tissue, was found to produce
miosis when introduced into the anterior chamber of animal
eyes. PGs were later identied in these iris extracts. Although
the mechanism of the PG-mediated miotic response, as well as
what other compounds in irin may be involved remains to be
determined, topical application of cyclooxygenase blockers
appears to help minimize the amount of intraoperative miosis.
For many years, topical urbiprofen 0.03% (Ocufen) has been
used in preventing intraoperative miosis. Miosis during eye
surgery, a common occurrence, can severely limit the surgeons
visualization and potentially increase the complication rate of
the procedure. Surgical trauma that stimulates the production
of PGs appears to play an integral role in the development of
intraoperative miosis. PGs have been observed in the aqueous
humor of traumatized eyes and appear to induce miosis
independent of cholinergic mechanisms.33 By inhibiting PG
synthesis by blocking the cyclooxygenase pathway,34 0.03%
urbiprofen, when administered every 30 min beginning 2 h
preoperatively, has limited intraoperative miosis during anterior
segment surgery in animal35,36 and human eyes.37 Preoperative
treatment is the key, because once the PGs are released, topical
urbiprofen does not block the PGs effect on the iris.
Some cataract surgeons have suggested that urbiprofen may
retard the reversal of the mydriasis by agents such as intracameral acetylcholine and carbachol, which potentially increase
the chances of such complications as intraocular lens pupillary
capture. Theoretically, urbiprofen should have no effect on
intracameral acetylcholine or carbachol; there is no known
pharmacologic basis for any such interaction.38 A possible

explanation may be that some surgeons tend to rub the end of
the cannula on the iris as the intraocular solution of acetylcholine or carbachol is injected to hasten the development of
the miosis. Such a maneuver in eyes not dosed wtih
flurbiprofen would likely stimulate the iris to produce PGs and
induce miosis; eyes previously treated with urbiprofen would
not show similar effects.38
Topical urbiprofen, however, does not appear to be as
effective in minimizing miosis during vitreoretinal surgery.39,40
Whether this is because surgical manipulation is generally
greater with vitreoretinal surgery than with anterior segment
surgery and, therefore, more PGs are released, leading to miosis,
remains to be determined. Another topical NSAID, suprofen,
has been demonstrated to also be effective in reducing pupillary
constriction during cataract surgery.41 The relative efcacy of
urbiprofen and suprofen remains to be determined. Additionally, although topical diclofenac is only approved by the
FDA for treatment of uveitis following cataract surgery, this
drug can also minimize intraoperative miosis.42
The mechanisms involved in surgical miosis are complex.
Although certain PGs have been associated with producing
miosis, no single PG possesses a miotic effect in all species or
is potent enough of a miotic to completely account for surgical
miosis.43,44 The specic mechanism of action of cyclooxygenase
blockers such as urbiprofen may well have a variety of biologic
effects that cannot be satisfactorily explained by inhibition of
PG synthesis alone.
POSTSURGICAL INFLAMMATION AND

DISCOMFORT
A number of topical NSAIDs have been tested as potential
substitutes for topical corticosteroids for the treatment of
postoperative inammation. Because steroid use after cataract
surgery may be associated with increased intraocular pressure
and glaucoma, increased risk of infection, and inhibition of
wound healing, a topical NSAID has been sought for the treatment of postsurgical inammation. Because intraocular inammation is associated with the breakdown of the bloodaqueous
barrier, investigators have used the leakage of uorescein into
the anterior chamber after systemic administration to indirectly
gauge the amount of inammation.45,46 It has been suggested
that a reduction in the leakage of uorescein with NSAID
treatment is an indication of a reduction in inammation. The
breakdown of the bloodaqueous barrier, assessed by uorophotometry or slit-lamp examinations after cataract surgery,
appears to be reduced by several topical NSAIDs, including ketorolac tromethamine, diclofenac sodium, and urbiprofen.4752
Randomized, controlled studies to compare the antiinammatory actions of 0.5% ketorolac tromethamine versus 0.1%
dexamethasone47 and 0.01%, 0.05%, or 0.1% diclofenac sodium
versus 1% prednisolone sodium phosphate49 demonstrated that
topical NSAIDs were superior to the topical steroids in reducing
breakdown of the bloodaqueous barrier as measured by uorophotometry. These preliminary studies suggest that topical
NSAIDs are a useful substitute for topical corticosteroids in the
management of postoperative inammation.
The only topical NSAIDs currently approved by the FDA
for the treatment of inammation following cataract surgery
are diclofenac sodium 0.1% (Voltaren), ketorolac 0.4% (Acular
LS), bromfenac 0.9% (Xibrom), and nepafenac 0.1%
(Nevanac).5254 In addition to reducing trauma-induced inammation, topical nepafenac, an amide analog of the NSAID
amfenac, has also been shown to inhibit inammationmediated retinal edema and ocular neovascularization in
animal models.54,55
CHAPTER 24
261

Topical diclofenac and ketorolac are approved for use after
refractive surgery. The frequent occurrence of pain after
refractive surgery, such as photorefractive keratectomy or radial
keratotomy, has been reduced with the administration of topical
ketorolac or diclofenac.5662 Topical diclofenac has also been
shown to be a suitable replacement for topical steroids in
managing postoperative inammation following strabismus
surgery.63
SECTION 4
OCULAR INFLAMMATORY DISORDERS

Few areas in ophthalmology have received more attention than
cystoid macular edema (CME). Although CME still remains
poorly understood, most researchers would agree that inammation is important to its pathogenesis. While there is no FDAapproved treatment of CME following cataract surgery, preliminary studies involving topical or systemic NSAIDs have been
encouraging.54,6469 These studies suggested that NSAIDs may
be useful in the prophylaxis and treatment of CME following
cataract surgery. In our clinical practice, we initially start our
CME patients on intensive topical steroids (eight times a day)
and topical NSAIDs (four times a day) for 24 weeks. If there is
no response or if the CME worsens, topical NSAIDs are
discontinued and the use of systemic NSAIDs is considered.
The NSAIDs have also been evaluated in the treatment of
inammatory diseases of the sclera. When taken orally, urbiprofen may be effective in treating scleritis and episcleritis;70
however, the topical form does not appear to be useful in the
management of episcleritis.71 Oral NSAIDs also may be useful
as an adjunct in the management of chronic iridocyclitis in
childhood.72 When children with idiopathic iridocyclitis or
iridocyclitis in association with juvenile rheumatoid arthritis
were treated with oral NSAIDs, both inammation in the
anterior chamber and the need for topical and systemic steroids
were reduced.72
Another potentially useful application of NSAIDs is in suppressing the inammatory response associated with ocular
infections. It is well known that topical steroid use can
exacerbate viral, bacterial, and fungal infections of the eye. The
effect of topical NSAIDs on corneal epithelial herpes simplex
viral infections remains controversial; two experimental studies
have found that topical NSAIDs did not worsen herpes simplex
viral infections of the cornea,73,74 whereas an earlier study
suggested that the exacerbation of ocular herpes simplex viral
infections by topical urbiprofen is similar to that of topical
dexamethasone.75 Preliminary studies have found that topical
NSAIDs have no adverse effect on either bacteria76 or fungal77
ocular infections.
Traditional NSAIDs were once thought to have some promise

in the management of allergic disorders of the eye. Topical
urbiprofen (0.03%) and suprofen (1%) have been used in the
treatment of allergic conjunctivitis78 and vernal conjunctivitis79
respectively. However, topical ketorolac remains the only FDAapproved topical NSAID for seasonal allergic conjunctivitis. Its
approval was based on two fairly small studies,80 and other
agents, such as antihistamine/mast cell stabilizers, are generally
preferable.80,81 NSAIDs have little, if any, place in the treatment
of ocular allergy.80,81
While bromfenac 0.09% (Xibrom) is only FDA approved for
treatment of inammation after cataract surgery, like other
NSAIDS, bromfenac may prove useful in managing other forms
of ocular inammation. Topical bromfenac is unique due to its
twice-a-day dosing and may improve patient compliance.
Oral aspirin has been shown to be useful as both primary82
and adjunctive83 therapy with steroids in the relief of
conjunctival and episcleral redness and in the resolution of
keratitis and limbal inltrates in vernal keratoconjunctivitis
(VKC). Patients with VKC have shown improvement after
treatment with up to 1 g of oral aspirin daily for 6 weeks.
Because of the relatively high dose, the clinician should closely
monitor any patient during an aspirin therapy regimen, and
should be aware of all contraindications to aspirin use. Aspirin
has many other properties we have yet to dene, and clinically,
aspirin is of interest as a potential ocular therapeutic agent,
particularly if the barriers to developing a safe method of topical
ocular aspirin delivery can be overcome.
PREVENTION OF CATARACT FORMATION

Although corticosteroid use is associated with cataract formation, aspirin84,85 and other NSAIDs86,87 may protect against
cataracts. The mechanism for this apparent protective effect
remains nebulous; however, it may be related to aspirins
acetylation of the lens proteins, which protects these proteins
from a variety of chemical insults.88,89 In addition, the lowering
of blood glucose levels in diabetics and nondiabetics associated
with NSAIDs may play a role in preventing cataracts.87 Nearly
half of all patients with cataracts have been estimated to
have abnormal glucose tolerance.90 Because diabetes is clearly
associated with cataracts, perhaps the glucose-lowering effect
of NSAIDs serves to favorably affect these patients with
chronic elevation of glucose levels. Other observational
studies9193 and a randomized study94 did not nd that aspirin
lowered the incidence of cataracts. It seems that aspirin or
aspirin-like agents neither prevent nor slow cataract formation,
although a small benet cannot be ruled out.94
REFERENCES
262
1. Gifford H: On the treatment of sympathetic

ophthalmia by large doses of salicylate of
sodium aspirin or other salicylic
compounds. Ophthalmoscope 1910;
8:257259.
2. Vane JR: Inhibition of prostaglandin synthesis
as a mechanism of action for aspirin-like
drugs. Nature 1971; 231:232235.
3. Vane JR, Botting R: Inammation and
mechanism of action of anti-inammatory
drugs. FASEB J 1987; 1:8996.
4. Rainsford KO: Inammation mechanisms
and actions of traditional drugs,
anti-inammatory and anti-rheumatic
drugs. Boca Raton, FL: CRC Press; 1985.
5. Bhattacherjee P: The role of arachidonate
metabolites in ocular inammation. Prog
Clin Biol Res 1989; 312:211227.
6. Eakins KE: Prostaglandin and nonprostaglandin-mediated breakdown of the

bloodaqueous barrier: in the ocular and
cerebrospinal uid. Exp Eye Res 1977;
25:483498.
7. Bhattacherjee P: Prostaglandin and
inammatory reactions in the eye. Methods
Find Exp Clin Pharmacol 1980; 2:1731.
8. Abelson MB, Butrus SI, Kliman GH, et al:
Topical arachidonic acid: a model for
screening anti-inammatory agents. J Ocul
Pharmacol 1987; 3:6375.
9. Unger WG, Bass MS: Prostaglandin and
nerve-mediated response of the rabbit eye
to argon laser irradiation of the iris.
10. Miyake K, Sugiyama S, Norimatsu L, et al:
Prevention of cystoid macular edema after
lens extraction by topical indomethacin

(111): radioimmunoassay measurement
of prostaglandins in the aqueous during
and after lens extraction procedures.
Graefes Arch Klin Exp Ophthalmol 1978;
209:8388.
11. Eakins KE: Prostaglandin and nonprostaglandin mediated break-down of the
bloodaqueous barrier. Exp Eye Res 1977;
25:483498.
12. Conquet P, Plazonnet B, LeDouarec J:
Arachidonic acid-induced elevation of
intraocular pressure and anti-inammatory
agents. Invest Ophthalmol 1975;
14:772775.
13. Podos SM: Prostaglandin, nonsteroidal antiinammatory agents and eye disease. Trans
Am Ophthalmol Soc 1976; 74:637660.
14. Gryglewski RJ, Panczenko B, Korbut R,

et al: Corticosteroids inhibit prostaglandin
release from perfused lungs of sensitized
guinea pig. Prostaglandins 1975; 10:343355.
15. Hong SL, Levine L: Inhibition of
arachidonic acid release from cells as the
biochemical action of anti-inammatory
corticosteroids. Proc Natl Acad Sci USA
1976; 73:17301734.
16. Campbell WB, Halushka PV: Lipid-derived
autocoids: eicosanoids and plateletactivating factor. In: Hardman JG,
Limbird LE, Molinoff PB, et al, eds.
Goodman and Gilmans the pharmacological
basis of therapeutics. 9th edn. New York:
McGraw-Hill; 1996:601616.
17. Bito LZ: Ocular effects of prostaglandins
and other eicosanoids. Surv Ophthalmol
1997; 41(Suppl 2):1144.
18. Langman MJS: Peptic ulcer complications
and the use of nonaspirin nonsteroidal
anti-inammatory drugs. Adverse Drug
React Bull 1986; 120:448.
19. Paulus HE: Arthritis advisory committee
meeting. Risks of agranulocytosis aplastic
anemia, ank pain and adverse
gastrointestinal effects with the use of
nonsteroidal anti-inammatory drugs.
Arthritis Rheum 1987; 30:593.
20. Hamberg M, Svensson J, Samuelsson B:
Thromboxane: a new group of biologically
active compounds derived from
prostaglandin endoperoxides. Proc Natl
Acad Sci USA 1975; 72:29942998.
21. Clive DM, Stoff JS: Renal syndromes
associated with nonsteroidal antiinammatory drugs. N Engl J Med 1984;
310:563572.
22. Gurwitz JH, Avorn J, Ross-Degnan D, et al:
Nonsteroidal anti-inammatory
drug-associated azotemia in the very old.
JAMA 1990; 264:471475.
23. Goldberg D, Panigrahi D, Barazi M, et al:
A case of Rofecoxib-associated
Stevens-Johnson syndrome with corneal
and conjunctival changes. Cornea 2004;
23:736737.
24. Sitenga GL, Ing EB, Van Dellen RG, et al:
Asthma caused by topical application of
ketorolac. Ophthalmology 1996;
103:890892.
25. Lin JC, Papuano CJ, Laibson PR, et al:
Corneal melting associated with use of
topical NSAIDs after ocular surgery. Arch
Ophthalmol 2000; 118:11291132.
26. Congdon NG, Schein OD, von Kulajta P,
et al: Corneal complications associated
with topical ophthalmic use of NSAIDS.
27. Guidera AC, Luchs JI, Udeil IJ: Keratitis,
ulceration, and perforations associated with
topical NSAIDs. Ophthalmology 2001;
108:936944.
28. Asai T, Nakagami TO, Mochizuki M, et al:
Three cases of corneal melting after
instillation of a new NSAID. Cornea 2006;
25:224227.
29. Mian SZ, Gupta A, Pineda R: Corneal
ulceration and perforation with Ketorolac
Tromethamine use after PRK. Cornea 2006;
25:232234.
30. Miller D, Gruenberg P, Miller R, et al:
Topical urbiprofen or prednisolone: effect
on corneal wound healing in rabbits. Arch
Ophthalmol 1981; 99:681682.
31. Ambache N: Irin, a smooth-muscle
contracting substance present in rabbit iris.
J Physiol 1955; 129:6569.
32. Ambache N: Properties of irin, a

physiological constituent of the rabbit iris.
J Physiol 1957; 135:114132.
33. Cole DF, Unger WG: Prostaglandins as
mediators for the responses of the eye due
to trauma. Exp Eye Res 1973; 17:357368.
34. Podos SM, Becker B: Comparison of
ocular prostaglandin synthesis inhibitors.
35. Anderson JA, Chen CC, Vita JB, et al:
Disposition of topical urbiprofen in normal
and aphakic rabbit eyes. Arch Ophthalmol
1982; 100:642645.
36. Dufn RM, Camras CB, Gardner SK, et al:
Inhibitors of surgically-induced miosis.
37. Keates RH, McGowan KA: Clinical trial of
urbiprofen to maintain pupillary dilation
during cataract surgery. Ann Ophthalmol
1984; 16:919921.
38. Holmes JM, Jay WM: The effect of
preoperative urbiprofen on miosis produced
by acetylcholine during cataract surgery. Am
J Ophthalmol 1991; 111:735738.
39. Vander JF, Greven CM, Maguire JI, et al:
Flurbiprofen sodium to prevent intraoperative
miosis during vitreoretinal surgery. Am J
Ophthalmol 1989; 108:288291.
40. Smiddy WE, Glaser BM, Michels RG, et al:
Miosis during vitreoretinal surgery. Retina
1990; 10:4246.
41. Stark WJ, Fagadau WR, Stewart RH, et al:
Reduction of pupillary constriction during
cataract surgery using suprofen. Arch
Ophthalmol 1986; 104:364366.
42. Roberts CW: A comparison of diclofenac
sodium to urbiprofen for maintaining
intraoperative mydriasis. Invest Ophthalmol
Vis Sci 1993; 35:1967.
43. Camras CB, Miranda OC: The putative role
of prostaglandins in surgical miosis. In:
Bito LZ, Stjernschantz J, eds. The ocular
effects of prostaglandins and other
eicosanoids. New York: Alan R Liss;
1989:197210.
44. Miranda OC, Bito LZ: The putative and
demonstrated miotic effects of
prostaglandins in mammals. In: Bito LZ,
Stjernschantz J, eds. The ocular effects of
prostaglandins and other eicosanoids. New
York: Alan R. Liss; 1989:171195.
45. Sanders DR, Kraff MC, Lieberman HL,
et al: Breakdown and reestablishment of
bloodaqueous barrier with implant surgery.
46. Sanders DR, Kraff MC: Steroidal and
nonsteroidal anti-inammatory agents. Arch
Ophthalmol 1984; 102:14531456.
47. Flach AJ, Kraff MC, Sanders DR, et al: The
quantitative effect of 0.5% ketorolac
tromethamine solution and 0.1%
dexamethasone sodium phosphate solution
on postsurgical blood aqueous barrier. Arch
Ophthalmol 1988; 106:480483.
48. Araie M, Sawa M, Takase M: Topical
urbiprofen and diclofenac suppress
bloodaqueous barrier breakdown in
cataract surgery: a uorophotometric study.
Jpn J Ophthalmol 1983; 27:535542.
49. Kraff MC, Sanders DR, McGuigan L, et al:
Inhibition of the bloodaqueous barrier
breakdown with diclofenac. Arch
Ophthalmol 1990; 108:380383.
50. Flach AJ, Graham J, Kruger LP, et al:
Quantitative assessment of postsurgical
breakdown of bloodaqueous barrier
following administration of 0.5% ketorolac
tromethamine solution: a double-masked,
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
paired comparison with vehicle-placebo

solution study. Arch Ophthalmol 1988;
106:344347.
Flach AJ, Lavelle CJ, Olander KW, et al:
The effect of ketorolac tromethamine
solution 0.5% in reducing postoperative
inammation after cataract extraction and
intraocular lens implantation.
Vickers FF, McGuigan LJB, Ford C, et al:
The effect of diclofenac sodium on the
treatment of postoperative inammation.
Invest Ophthalmol Vis Sci 1991; 32(ARVO
Suppl):793.
Ket L, Graff G, Spellman JM, et al:
Nepatenac, a unique nonsteroidal pro drug
with potential utility in the treatment of
trauma-induced ocular inammation.
Inammation 2000; 24:371384.
Kapin MA, Yanni JM, Brady MT, et al:
Inammation-mediated retinal edema in the
rabbit is inhibited by topical nepafenac.
Inammation 2003; 27:281291.
Takahashi K, Saishin Y, Saishin Y, et al:
Topical nepafenac inhibits ocular
neovascularization. Invest Ophthalmol Vis
Sci 2003; 44:409415.
Eiferman RA, Hoffman RS, Sher NA:
Topical diclofenac reduced pain following
photorefractive keratectomy. Arch
Ophthalmol 1993; 111:1022.
Sher NA, Frantz JM, Talley A, et al: Topical
diclofenac in the treatment of ocular pain
after excimer photorefractive keratectomy.
Refract Corneal Surg 1993; 9:425436.
Arshinoff EA: Use of topical nonsteroidal
anti-inammatory drugs in excimer laser
photorefractive keratectomy. J Cataract
Refract Surg 1994; 20:216222.
Szerenyi K, Sorken K, Garbus JJ, et al:
Decrease in normal human corneal
sensitivity with topical diclofenac sodium.
Am J Ophthalmol 1994; 118:312315.
Epstein RL, Laurence EP: Relative
effectiveness of topical ketorolac and
topical diclofenac on discomfort after radial
keratotomy. J Cataract Refract Surg 1995;
21:156159.
Seitz B, Sorken K, LaBree LD, et al:
Corneal sensitivity and burning sensation:
comparing topical ketorolac and
diclofenac. Arch Ophthalmol 1996;
114:921924.
Tomas-Barberan S, Trngren L, Lundberg K,
et al: Effect of diclofenac on prostaglandin
liberation in the rabbit after photorefractive
keratectomy. J Refract Surg 1997;
13:154157.
Wright M, Butt Z, McIllwaine G, et al:
Comparison of the efcacy of diclofenac
and betamethasone following strabismus
surgery. Br J Ophthalmol 1997;
81:299301.
Kraff MC, Sanders DR, Jampol LM, et al:
Prophylaxis of pseudophakic cystoid
macular edema with topical indomethacin.
Abelson MB, Smith LK, Ormcrod LD:
Prospective, randomized trial of oral
piroxicam in the prophylaxis of
postoperative cystoid macular edema.
Flach AJ, Dolan BJ, Irvine AR:
Effectiveness of ketorolac tromethamine
0.5% ophthalmic solution for chronic
aphakic and pseudophakic cystoid
macular edema. Am J Ophthalmol 1987;
103:479486.
CHAPTER 24
263
SECTION 4
264
67. Flach AJ, Stegman RC, Graham J, et al:

Prophylaxis of aphakic cystoid macular
edema without corticosteroids.
68. Solomon LD: Flurbiprofen-CME Study
Group I. Efcacy of topical urbiprofen and
indomethacin in preventing pseudophakic
cystoid macular edema. J Cataract Refract
Surg 1995; 21:7381.
69. Rosseti L, Chaudhuri J, Dickersin K:
Medical prophylaxis and treatment of
cystoid macular edema after cataract
surgery. Ophthalmology 1998;
105:397405.
70. Watson PG: Doyne memorial lecture. Trans
71. Lyons CJ, Hakin KN, Watson PG: Topical
urbiprofen: an effective treatment for
episcleritis? Eye 1990; 4:521525.
72. Olson NY, Lindsley CB, Godfrey WA:
Nonsteroidal anti-inammatory drug
therapy in chronic childhood iridocyclitis.
Am J Dis Child 1988; 142:12891292.
73. Fraser-Smith EB, Mathews TR: Effect of
ketorolac on herpes simplex virus type one
ocular infection in rabbits. J Ocul
Pharmacol 1988; 4:321326.
74. Colin J, Bodin C, Malet F, et al: La keratite
herpetique experimentale du lapin. J Fr
Ophtalmol 1989; 12:255259.
75. Trousdale MD, Dunkel EC, Nesburn AB:
Effect of urbiprofen on herpes simplex
keratitis in rabbits. Invest Ophthalmol Vis
Sci 1980; 19:267270.
ketorolac on Pseudomonas aeruginosa
ocular infection in rabbits. J Ocul
Pharmacol 1988; 4:101109.
ketorolac on Candida albicans ocular
78.
79.
80.
81.
82.
83.
84.
85.
86.
infection in rabbits. Arch Ophthalmol 1987;

105:264267.
Bishop K, Abelson M, Cheetharn J, et al:
Evaluation of urbiprofen in the treatment
of antigen-induced allergic conjunctivitis.
Suppl):487.
Buckley DC, Caldwell DR, Reaves TA:
Treatment of vernal conjunctivitis with
suprofen, a topical nonsteroidal antiinammatory agent. Invest Ophthalmol Vis
Sci 1986; 27(ARVO Suppl):29.
Tinkelman DG, Rupp G, Kaufman H, et al:
Double-masked, paired-comparison clinical
study of ketorolac tromethamine 0.5%
ophthalmic solution compared with
placebo eyedrops in the treatment of
seasonal allergic conjunctivitis. Surv
Ophthalmol 1993; 38:141148.
Laibovitz RA, Koester J, Schaich L, et al:
Safety and efcacy of diclofenac sodium
0.1% ophthalmic solution in acute seasonal
allergic conjunctivitis. J Ocul Pharmacol
1995; 11:361368.
Meyer E, Kraus E, Zonis S: Efcacy of
antiprostaglandin therapy in vernal
71:497499.
Abelson MB, Butrus SI, Weston JH: Aspirin
therapy in vernal conjunctivitis. Am J
Ophthalmol 1983; 95:502505.
Cotlier E: Aspirin and senile cataract in
rheumatoid arthritis. Lancet 1981;
1:338339.
Cotlier E, Sharma YG, Niven T, et al:
Distribution of salicylate in lens and
intraocular uids and its effects on cataract
formation. Am J Med 1983; 74:8390.
van Heyningen R, Harding JJ: Do
aspirin-like analgesics protect against
87.
88.
89.
90.
91.
92.
93.
94.
cataract? A case-control study. Lancet

1986; 1:11111113.
Harding JJ, van Heyningen R: Drugs,
including alcohol, that act as risk factors
for cataract, and possible protection
against cataract by aspirin-like analgesics
and cyclopenthiazide. Br J Ophthalmol
1988; 72:809814.
Crompton M, Rixon KC, Harding JJ: Aspirin
prevents carbamylation of soluble lens
proteins and prevents cyanate-induced
phase separation opacities in vitro: a
possible mechanism by which aspirin could
prevent cataract. Exp Eye Res 1985;
40:297311.
Rao GN, Lardis MP, Cotlier E: Acetylation
of lens crystallins: a possible mechanism
by which aspirin could prevent cataract
formation. Biochem Biophys Res Commun
1985; 128:11251132.
Dugmore WN, Tun K: Glucose tolerance
tests in 200 patients with senile cataract.
Siegel D, Sperduto RD, Ferris F: Aspirin
and cataracts. Ophthalmology 1982;
89:47A.
Klein BK, Klein R, Moss S: Is aspirin use
associated with lower rates of cataracts in
diabetic individuals? Diabetes Care 1987;
10:495499.
West SK, Munoz BE, Newland HS, et al:
Lack of evidence for aspirin use and
prevention of cataracts. Arch Ophthalmol
1987; 105:12291231.
Seddon JM, Christen WG, Manson JE,
et al: Low-dose aspirin and risks of
cataract in a randomized trial of U.S.
physicians. Arch Ophthalmol 1991;
109:252255.
CHAPTER
25
Antihistamines and Mast Cell Stabilizers in Allergic

Ocular Disease
Gregg J. Berdy, Andrea Leonardi, and Mark B. Abelson
Ophthalmologists frequently see allergic diseases of the eye.

They may be the most common clinical problems involving the
external ocular adnexa. Approximately 20% of the US population (~60 million people) is affected with these disorders.
Although allergic ocular diseases may affect the skin and
subcutaneous tissues of the eyelids, it is the conjunctiva, the
mucous membrane of the eye, which is more commonly and
severely affected. In certain cases, the eye may be the only organ
system involved. In most of these patients, however, the ocular
tissues participate as part of a systemic allergic response to
exogenous or intrinsic antigens.
Allergic conjunctivitis is observed more frequently in industrialized countries as a consequence of the deviation of the
immune system toward a T helper cell lymphocyte (Th2-type)
immune response favored by a reduction in infection, air
pollution, and modern lifestyles. This disease ranges in severity
from mild forms, which still interfere signicantly with quality
of life, to severe cases characterized by potential impairment of
visual function.
Ocular allergy encompasses a spectrum of diseases characterized by the IgE- and Th2-mediated hypersensitivity responses.
The most common ocular allergies are seasonal and perennial
allergic conjunctivitis (SAC and PAC), the ocular counterpart of
allergic rhinitis. Exposure to environmental allergens such as
pollens, animal dander, and dust causes the symptoms and
signs of ocular hay fever in sensitized persons. An acute attack
is characterized by conjunctival injection, chemosis, tearing,
eyelid swelling, burning, and ocular and periocular itching.
The chronic allergic ocular diseases, vernal keratoconjunctivitis (VKC), atopic keratoconjunctivitis (AKC), and giant papillary
conjunctivitis (GPC) are relatively rare but clinically well characterized. Mast cells, T-cell lymphocytes, eosinophils, and their
mediators all play major roles in the clinical manifestation of
these diseases. Typical Th2-type cytokines, IL-4, IL-5, and IL-13,
as well as other proinammatory cytokines, chemokines, growth
factors, and enzymes are overexpressed in the conjunctiva of
patients with chronic allergic diseases. Each of these diseases
has specic clinical features in terms of diagnosis and treatment.
MAST CELL AND PATHOPHYSIOLOGY

Knowledge of the pathogenesis of ocular allergic disease is
critical to understanding the role of therapeutic medications
used in the treatment of these diseases. SAC is the prototype of
this group of diseases and begins as an antigenIgE antibody
interaction on the surface of conjunctival mast cells.1 Exposure
of sensitized IgE-coated mast cells to specic allergen causes the
cross-linking of membrane-bound IgE receptors (FCeRI), the
activation of mast cells, and the release of preformed and newly
formed mediators.
Mast cell populations in humans demonstrate heterogeneity

in different organ systems, and the neutral protease content of
the mast cell cytoplasmic granules has provided one basis for
subclassication. Mast cells containing tryptase alone (mucosal
type mast cells or MCT) are found most frequently at mucosal
sites. Those mast cells containing tryptase and chymase (connective type mast cells or MCCT) are more characteristic of
connective tissue sites.2 Immunohistochemical phenotyping of
mast cells in the normal human conjunctiva has demonstrated
that the MCCT phenotype is predominant, similar to the ndings
in human skin.3 In addition to chymase, MCCT also contain
cathepsin-G and carboxypeptidase-A, both of which are absent
from MCT.4
The exogenous allergen binds to two separate IgE molecules,
creating a dimer formation that initiates a chain of reactions in
the mast cell plasma membrane.5,6 It is thought that the bridging
of mast cell IgE molecules (cross-linking) induces activation
of membrane-associated enzymes, leading to an increase in
the uptake of calcium.7 Enzymes identied with intracellular
calcium mobilization and initiation of the biochemical process
of histamine release are membrane-associated proteolytic
enzymes,8 methyltransferases,9 and adenylate cyclase.10,11 In
addition, the cross-linking of membrane-bound IgE molecules
induces the activation of phospholipase A2 with subsequent
release and metabolism of arachidonic acid.12 This 20-carbon,
unsaturated fatty acid serves as a precursor for newly synthesized substances, such as prostaglandins, leukotrienes,13 and
platelet-activating factor,14,15 that have been implicated as
important mediators of clinical allergic disease.16
The FceRI on mast cells consists of an a-chain, a b-chain,
and two g-chains. The a-chain is responsible for IgE binding,
while the b-chain promotes stability and enhances the signaling
capacity. Monomeric IgE binding to the a-chain does not result
in conformational changes, but enhances mast cell survival and
growth. The dimer of the g-chain is shared by other Fc receptor
complexes and carries two immunoreceptor tyrosine-based
activation motifs (ITAMs) for downstream signaling. The signal
for mast cell degranulation is aggregation of FceRI and the
minimal signal only requires dimerization. Maximal degranulation of mast cells and basophils is associated with distinct
aggregation of both b- and g-chains, but g-chain aggregation
alone can result in suboptimal stimulation. The process of FceRI
cross-linking (mediated through interaction of antigen with
receptor-bound IgE) results in phosphorylation of the ITAMs (of
the b and g subunits) by the Src family tyrosine kinase lyn
(probably under regulation of the phosphatase, CD45) and
recruitment of the protein tyrosine kinase Syk. Syk amplies
the signal as it targets multiple proteins for activation (including phospholipase Cg (PLCg, the guanine nucleotide exchange
factor Vav1, and adaptor molecules SH2 domain-containing
265
SECTION 4
266
leukocyte protein of 76 kDa (SLP-76), and linker for activation

of T cells (LAT)). This promotes activation of various kinase
cascades, which overlap and share signaling components. The
phospholipase C (PLC) inositol pathway is primarily involved
in degranulation, which is the result of Ca2+ mobilization and
cytoskelatal changes and culminates in immediate release of
stored mediators.17
At the ultrastructural level, it has been demonstrated that
human lung mast cells, once stimulated, show swelling of individual granules, with the subsequent fusion and formation of
interconnected chains of altered granules. These intracellular
cytoplasmic channels eventually fuse with the plasma membrane
of the mast cell, thereby releasing their contents into the extracellular space.1820 These secretory granules contain several
preformed mediators, including biogenic amines (histamine),
neutral proteases (chymase, tryptase), proteoglycans (heparin),
and acid hydrolases, that initiate and promulgate the allergic
response. The downstream signaling stimulates a number of
transcription factors, leading to activation of genes regulating
the release of newly formed mediators such as prostaglandin
(PG)D2, leukotriene (LT)C4, and cytokines (e.g., TNFa). These
transcription factors include nuclear factor of activated T cells
(NF-AT), nuclear factor kappa B (NFkB), signal transducer and
activator of transcription (STAT)-6, activator protein-1 (AP-1),
c-fos, and c-jun.17
In combination with the other mast cell serine protease,
chymase, tryptase may be implicated in the activation of other
proteases, such as collagenase (MMP-1), gelatinases A and B
(MMP-2, MMP-9), and stromelysin (MMP-3), which are all
involved in extracellular matrix degradation and inammatory
cell inltration.
Conjunctival mast cells have been shown to be a source of
several cytokines and growth factors. Interleukin (IL)-4, IL-5,
IL-6, tumor necrosis factor alpha (TNFa), transforming growth
factor beta (TGF-b)-1, or (FGF) and stem cell factor were localized to mast cells in normal and allergic conjunctiva.21,22 The
pattern of cytokine expression in the two mast cell subtypes
showed that IL-4 and IL-13 were preferentially associated with
the MCCT subset, whereas IL-5 and IL-6 were associated to the
MCT subset, suggesting that differences in protease phenotype
may also reect functional differences manifested by different
patterns of cytokine distribution.23 These cytokines appeared to
be stored within the cytoplasmic secretory granules, suggesting
that they may be rapidly released upon IgE- and non-IgEmediated mast cell activation.
Mast cell degranulation releases proinammatory mediators
and cytokines which induces the activation of epithelial cells
and vascular endothelial cells leading to the expression of
chemokines (e.g., RANTES, MCP-1, IL-8, eotaxin) and adhesion
molecules (e.g., ICAM-1, VCAM, and p-selectin).24 These
factors initiate the recruitment phase of inammatory cells in
the conjunctival mucosa.
Several cytokines can be found in tears of allergic and nonallergic subjects, however, the cellular source of these cytokines
is difcult to determine. Altered ratios of proinammatory
cytokines could reect differences in the patterns of TH2 versus
TH1 and proinammatory versus antiinammatory cytokines
between nonallergic and allergic tears.25
Great advances in conjunctival mast cell biology and function
were gained from a series of studies using in vitro cultured mast
cells derived from human cadaveric conjunctival tissues.
Stimulated conjunctival mast cells have been shown not only to
express mRNA for TNFa but also to release TNFa protein,
consequently upregulating ICAM-1 expression on conjunctival
epithelial cells.26 A subsequent paper demonstrated that
mast cells express functional receptors such as ICAM-1, c-kit,
and FceRI have surface bound IgE.27 The expression of these
surface markers was modied by stimulus with TNFa and

IL-4, showing that cytokines may modify mast cell functions.
In vitro mast cell behavior may also be considered a model for
studying the effects of antiallergic drugs.
ROLE OF HISTAMINE
The sentinel role of histamine in the acute allergic response has
been well established. Histamine was rst synthesized in 1907
and discovered to be an imidazolylethylamine.28 In 1910, the
biologic activity of this amine was discovered when it was
detected as a uterine stimulant in extracts of ergot. Later that
year, Dale and Laidlaw29 observed bronchospastic and vasodilator activity in animals with the intravenous administration
of histamine. In 1919, these authors observed that histamine
applied locally produced redness, swelling, and edema. In
addition, they noted that large doses of intravenous histamine
produced a symptom complex that was identical to that of a
systemic anaphylactic reaction.30 Eight years later, investigators
deduced that histamine was a humoral mediator involved in
acute allergic reactions.31
In 1953, the presence of histamine was noted in mast cells
taken from human skin.32 This discovery spurred the interest of
many researchers, leading to the elucidation of histamines
synthesis, secretion, metabolism, and biologic activity.33,34 It is
the biologic activity of histamine that creates the signs and
symptoms of the acute allergic reaction in ocular hay fever.
The physiologic and pharmacologic effects of histamine are
mediated by specic receptor subtypes present on effector cell
surfaces. Four distinct histamine receptors have been characterized to date and it is generally accepted that the H1 receptor
plays the greatest role in allergic disease. In 1966, Ash and
Schild35 identied specic receptors that were blocked by the
antihistamines known at that time and labeled them H1
receptors. These authors discovered that only certain responses
to histamine were blocked by the histamine antagonist mepyramine, and these responses were dened as being mediated by
H1 receptors. Six years later, Black et al36 identied a second
histamine receptor subtype, H2, by using specic antagonists
that blocked only the H2 receptors. They demonstrated that
histamine-induced hypotension that was only partially relieved
by mepyramine was totally blocked by the addition of the
H2-receptor antagonist burimamide. H2 and H3 receptors play
critical roles in a variety of tissues including the central and
peripheral nervous systems, gastrointestinal tract, and heart.
The H4 receptor is the most recently discovered of the histamine receptors.37,38 This novel receptor is highly expressed in
peripheral blood leukocytes and to the greatest extent on
eosinophils. Organ specicity of expression demonstrated high
levels of mRNA in several organs that are critical to immune
regulation such as bone marrow, spleen, and thymus. It is
speculated that the H4 receptor may become an important
future therapeutic target for regulation of immune function,
particularly with respect to allergy and asthma.
Histamine receptors belong to the large family of seven transmembrane G-protein coupled receptors (GPCR). G-proteins
derive their name from a high afnity for guanine nucleotides.
The binding of a ligand molecule to a GPCR in the plasma
membrane stimulates the trimeric G-protein resulting in
initiation of the PLC-inositol pathway. Generation of diacylglycerol (DAG) and inositol 1,4,5-triphosphate (IP3) results in
activation of a Ca2+-dependent protein kinase (PKC) and Ca2+
mobilization from the endoplasmic reticulum, respectively.
Ca2+ functions as a ubiquitous intracellular messenger. When
activated, PKC phosphorylates specic serine or threonine
residues on target proteins, such as MAPK and IkB-NFkB, leading
to increased transcription of specic genes. Furthermore,
Antihistamines and Mast Cell Stabilizers in Allergic Ocular Disease

can be induced in human conjunctival broblasts. In both types
of cells, blockage of H1 receptor activation using selective H1
receptor antagonists, antagonizes these events. The effects of
histamine are not signicantly blocked by the H2 and H3 antagonists cimetidine and thioperamide. Furthermore, histaminemediated activation of epithelial cells and broblasts increases
the permeability of the epithelium, the expression of adhesion
molecules and cytokines resulting in increased permeability of
macromolecules, such as allergens, and increased recruitment
and survival of inammatory cells that is observed in both acute
and chronic ocular allergic inammation.
THERAPEUTIC OPTIONS
Treatment of allergic ocular diseases, specically allergic conjunctivitis, may be approached in the same manner as one
would treat allergic rhinitis. Ideally, removing the offending
allergen or modifying the patients environment would be most
effective. However, this is not always practical. Systemic
medications such as oral antihistamines may be employed, but
these agents do not reliably relieve ocular symptoms, and their
soporic effects may mitigate their use. In most cases,
treatment with topical medications in the form of eye drops has
provided symptomatic relief without systemic side effects.
Topical corticosteroid preparations, such as uorometholone,
prednisolone 0.125%, and loteprednol etabonate 0.2% are
extremely effective in providing relief of itching, chemosis, and
mucous discharge. These drugs should be used only in cases
that do not respond to other forms of therapy, because they have
been associated with the development of elevated intraocular
pressure, cataract formation, and secondary bacterial, fungal,
and viral infections.57
Mast cell stabilizer preparations have been purported to
stabilize the mast cell plasma membrane, thereby preventing
subsequent degranulation and release of inammatory
mediators. The ophthalmic literature has debated the therapeutic value of disodium cromoglycate in allergic conjunctivitis.
Several studies have demonstrated a salutary effect,58,59 whereas
others have shown no effect.60,61 A second-generation preparation, lodoxamide 0.1%, has shown salutary effects in patients
with VKC.62,63
Mast cell modulation is thus a fundamental target for antiallergic treatment. In fact, most of the ocular antiallergic drugs
have been designed as mast cell stabilizers. A decrease of calcium
inux into the cytoplasm is reported to be the mechanism of
the most widely used ocular mast cell stabilizers: sodium cromoglycate, lodoxamide, nedocromil and pemirolast. An advancement in the treatment of ocular allergy comes from newly
designed ocular antihistamine compounds, such as olopatadine,
ketotifen, azelastine, and epinastine.64 These drugs have a dual
activity as antihistamines and mast cell stabilizers, probably
due to their effect on calcium mobilization or on phospholipid
cellular membrane. In fact, mast cell stabilizers inhibit
degranulation by interrupting the normal chain of intracellular
signals resulting from the cross-linking and activation of the
high-afnity IgE receptor (FceRI) by allergen.17 This promotes
activation of various kinase cascades, resulting in Ca2+ mobilization, cytoskeletal changes and culminating in immediate
release of stored mediators.
The drugs most commonly used to treat ocular hay fever are
topical antihistamines. Their mechanism of action is competitive inhibition with histamine for the histamine receptors
on effector cells. Currently, the only antihistamine preparations
available are H1-receptor antagonists. These agents reliably
relieve the symptoms of itching found in allergic conjunctivitis;
however, several preparations have little effect on chemosis and
redness.65 As such, these drugs are manufactured in combination
CHAPTER 25
DAG can be cleaved to release arachidonic acid, which acts

as a messenger as well as a substrate in the synthesis of
eicosanoids. It should be noted that there is a great deal of
overlap between the target proteins involved in FceRI and H1
receptor mediated activation.17 Therefore, while mast cell
mediators, particularly histamine, released upon ocular exposure to allergen can initiate all of the symptoms associated with
ocular allergy, it is the subsequent effect of histamine on other
ocular surface cells that is thought to perpetuate the inammatory response.
Identication of the H1 and H2 receptors has permitted
investigators to better understand histamines role in human
allergic disease. Owen and co-workers39 concluded that the
vasodilator response to histamine was mediated by both H1 and
H2 receptors; however, the increase in vascular permeability was
mediated solely by H1 receptors. When injected intradermally,
histamine causes a localized triple response. The initial component is the development of erythema immediately surrounding
the injection site as the result of vasodilation mediated by both
H1 and H2 receptors.40,41 A second component is the cutaneous
are that occurs as an indirect response to stimulation of
histamine receptors on afferent nonmyelinated nerve endings.
Antidromic nerve conduction initiates a reex arc that
culminates in the release of various neuropeptides, including
substance P and calcitonin gene-related peptide, which directly
affect arteriolar vasodilation.42 The wheal results from exudation of plasma through gaps between vascular endothelium of
postcapillary venules and is mediated by H1 receptors.43 Additionally, intradermal injection of histamine causes a sensory
response that is manifested as the sensation of itching.
Allergic conjunctivitis can be characterized as ocular
anaphylaxis occurring when a sensitized person is exposed to a
specic aeroallergen. Abelson et al44 demonstrated the presence
of mast cell-derived mediators in subconjunctival tissues and
precorneal tear lm in patients with ocular atopic diseases. This
is not unexpected, because the human conjunctiva contains
large numbers of mast cells subjacent to the epithelium.4547
Previously, Abelson and co-workers48 demonstrated the
presence of histamine in the tear lm of normal humans at
concentrations of 510 ng/mL whereas tear samples of patients
with active VKC contained signicantly higher levels of
histamine. It has been calculated that a single conjunctival
mast cell contains 4.6 pg of histamine,49 signifying that the
total potential amount of histamine that can be released with
massive mast cell degranulation is 23 ng/mm3.50 The increased
levels of histamine in tears of VKC patients may be related not
only to a massive mast cell degranulation but also to a reduced
activity of histaminase enzymes.51 Inducing an acute reaction
by challenging allergic patients with specic allergen, tear
histamine levels are signicantly increased compared with baseline. These levels are even higher when histaminase enzymes
are inactivated.52 Using this procedure, increased histamine
tear levels have also been found during the late phase reaction.53
The topical instillation of histamine produced the itching
and redness associated with allergic conjunctivitis in a dosedependent fashion.54 Identication of specic histamine
receptors on the ocular surface has made it possible to selectively identify the pathologic effects of histamine. Stimulation
of H1 receptors with the highly selective H1-receptor agonist
2-(2-aminoethyl) thiazoledihydrochloride elicited symptoms of
ocular itching.55 On the other hand, selective stimulation of H2
receptors by dimethylaminopropylisothiourea, a highly selective
H2-receptor agonist, produced vasodilation of conjunctival
vessels without itching.56
Histamine stimulates PI turnover and Ca2+ mobilization in
the human conjunctival epithelium inducing the release of
cytokines (IL-6, IL-8, GM-CSF) in these cells. Similar effects
267

with a vasoconstrictor agent that helps to relieve ocular
injection. Recently, several H1-selective receptor antagonists
have been introduced that relieve both the itching and the
redness associated with allergic conjunctivitis.66,67
SECTION 4
ANTIHISTAMINES
In 1927, Lewis68 described the wheal-and-are response seen in
human skin and suggested that histamine could be released
from intracellular stores by local injury. Armed with this
information, investigators began the search to develop
pharmacologic methods to blunt histamines profound effects.
In 1937, Bovet and Staub69 fortuitously noted that a compound
that they had been screening for adrenergic-blocking activity
also possessed some antihistaminic activity. This compound,
2-isopropyl-5-methyl-phenoxyethyldiethylamine, when administered to guinea pigs protected them from lethal doses of
histamine, antagonized histamine-induced smooth muscle
contraction, and diminished the systemic symptoms of
anaphylaxis. Unfortunately, this substance was too toxic for
clinical use, but it led to the discovery of phenbenzamine
(Antergan), a dimethylamine derivative that was the rst antihistaminic compound to be used in humans.70 In 1944, Bovet
and co-workers71 discovered another clinically effective
compound, pyrilamine maleate (Neo-Antergan), which is still
used today.
The rst description of topical antihistamine use in the eye
was published in 1946 by Bourquin.72 He observed satisfactory
results with the use of antazoline (Antistine) in patients with
vernal catarrh, phlyctenular conjunctivitis, conjunctivitis associated with hay fever, and scleritis. Two years later, in the
American literature, Hurwitz73 reported favorable results with
the same drug. Since the discovery that topical antihistamines
could alleviate symptoms of allergic conjunctivitis, several
authors have published results demonstrating that topical H1
antihistamines were clinically effective.74,75
Topical antihistamines are the rst line in the treatment of
ocular allergy.76 These drugs are H1 receptor competitive
antagonists of varying specicity, potency, and duration of
action. The rst-generation antihistamines, pheniramine and
antazoline, have a long safety record, but are known for their
burn upon instillation, their rapid onset and disappearance
of their effects, and their limited potency. These are still
available in over-the-counter products, particularly in association with vasoconstrictors. The newer antihistamines are
still H1 antagonists, but have a longer duration of action
(46 h), and are better tolerated then their predecessors. These
include levocabastine hydrochloride (Livostin, 0.5%) and
emedastine difumarate (Emadine, 0.05%). Both drugs are
effective and well tolerated also in pediatric subjects with
allergic conjunctivitis.
The clinical efcacy of ophthalmic levocabastine was shown
in numerous studies,66,77 while the newer emedastine appears
to be stronger and more selective. In fact, in a direct comparison
with levocabastine, emedastine proved signicantly more
effective in alleviating chemosis and lid swelling.78 In two in
vitro studies, emedastine, and to a much lesser degree levocabastine, blocked histamine-stimulated proinammatory
cytokines (IL-8 and IL-6) release from conjunctival epithelial
cells and broblasts.79,80
MAST CELL STABILIZERS
268
Several mast cell stabilizers are available for use in the eye:
cromolyn sodium 4%, nedocromil sodium 2%, lodoxamide
tromethamine ophthalmic solution 0.1%, spaglumic acid 4%,
and pemirolast potassium ophthalmic solution 0.1%. All these
drugs inhibit mast cell degranulation, the release of histamine

and the other preformed mediators and the arachidonic acid
cascade.
Cromolyn sodium (DSCG), a derivative of khellin, a
chromone found in Ammi visnaga, an eastern Mediterranean
plant, was rst synthesized in 1965. The drug is thought to act
on the mast cell plasma membrane via control of transmembrane calcium ux. The effect of DSCG is to stabilize the
membrane, thereby preventing degranulation and release of
inammatory mediators.81,82 Thus, DSCG must exert its effect
prior to allergen binding or, at least, before the mast cell
membrane is altered with subsequent mediator release.
Since its discovery, investigators have shown DSCG to have
salutary effects in patients with allergic asthma and other IgEmediated diseases.8390 In 1984, the US Food and Drug Administration (FDA) granted approval of DSCG for ocular use in
patients with VKC on the basis that the drug alleviated
symptoms and signs of the disease and allowed a reduction in
the frequency of steroid use in these patients.91,92 However,
the ability of DSCG to suppress ocular allergic symptoms in
environmental studies has yielded conicting results.9397 To
date, results of studies evaluating DSCG in allergic conjunctivitis have been encouraging, but the effectiveness of the drug
in this condition remains controversial. However, both 4%
DSCG and 0.1% lodoxamide have been shown to be effective in
controlling the signs and symptoms of VKC.62,91,98
Nedocromil appears to be more potent than cromolyn, and is
approved for two times daily dosing.99,100 Nedocromil was
shown to stabilize both connective tissue and mucosal mast
cells, as opposed to cromolyn, and appears to inhibit by a
common pathway mast cells, eosinophils, epithelial cells, and
sensory nerves. It has been shown to be superior to placebo and
cromolyn in trials of seasonal and PAC, and other ocular allergic
disorders.101
Lodoxamide has been available since 1993 in the United
States and Europe for the treatment of VKC; however, it has
also been shown effective against allergic conjunctivitis. Its
mechanism of action is thought to be similar to that of
cromolyn, since it was shown to prevent histamine release.
Inhibition of eosinophil activation and degranulation is the
proposed mechanism for its efcacy against corneal signs such
as keratitis and shield ulcers in severe allergic disease.102
Lodoxamide was shown superior to placebo,103 cromolyn104
and N-acetyl aspartyl glutamic acid105 for treatment of
VKC, and equal or superior to cromolyn for the treatment
of allergic conjunctivitis.106 Its recommended dosing is four
times daily.
Pemirolast is another mast cell stabilizer that has been shown
to alleviate the signs of allergic conjunctivitis.107 Previous in
vitro and in vivo studies have demonstrated the efcacy of
pemirolast in inhibiting the antigen-induced release of inammatory mediators (e.g., histamine, leukotriene C4, D4, E4)
from human mast cells and subsequently in preventing signs
and symptoms associated with allergic conjunctivitis. Pemirolast is currently approved for a four times daily (QID) dosing
regimen.108
Dipeptide N-acetyl-aspartyl glutamic acid (NAAGA) 6% has
been widely used in Europe as topical eye drops in the treatment
of allergic conjunctivitis, VKC, and GPC.105,109 NAAGA is
known to inhibit leukotriene synthesis, histamine release by
mast cells, and complement-derived anaphylatoxin production.
This antiallergic compound was also shown to directly inhibit
leukocyte adhesion to endothelial cells induced by proinammatory stimuli, and abrogates TNFa-induced expression of
adhesion molecules on granulocytes and endothelial cells.110
These pharmacological properties confer a potential antiinammatory activity to NAAGA.
This new category of drugs with dual mechanism of action

includes molecules that inhibit both mediator release from
mast cells (mast cell stabilizing effect) and, competitively,
histamine binding to H1 receptors (antihistaminic effect). In
this class are included olopatadine, ketotifen, azelastine, and
epinastine. The advantage offered by these molecules is the
rapidity of symptomatic relief given by immediate histamine
receptor antagonism, which alleviates itching and redness,
coupled with the long-term disease-modifying benet of mast
cell stabilization.
Olopatadine has been shown in numerous in vitro, in vivo,111
and clinical studies112 to effectively and potently inhibit conjunctival mast cells in allergic patients with seasonal and PAC
and allergic symptoms associated with contact lens wear.112
Olopatadine 0.1% was shown to reduce the levels of histamine,
the cellular inltrate, and ICAM expression compared with
placebo after conjunctival allergen challenge, indicating that it
reduced the release of mast cell-derived mediators in humans.113
Olopatadine was demonstrated more effective and comfortable
than ketotifen in seasonal studies114116 and more effective than
the mast cell stabilizer, nedocromil,117 and the antiinammatory agent, ketorolac, in the conjunctival allergen challenge
model in allergic subjects.118
Ketotifen has been shown to inhibit the release of inammatory mediators from mast cells, basophils, and neutrophils,
to inhibit the production and release of LTC4 and LTB4, platelet
activating factor (PAF) production by normal human neutrophils
and eosinophils, and eosinophil chemotaxis.119 Clinically, it has
been shown effective in the allergen challenge model, superior
to both placebo120 and cromolyn,121 and a safe treatment option
for children with allergic conjunctivitis.122
Azelastine, available in the past for rhinitis, has been approved
for ocular itching associated with allergic conjunctivitis.
Azelastine was shown to reduce ICAM-1 expression on conjunctival epithelium, and inammatory cell inltration during
both early and late phase allergic reactions.123 In placebocontrolled environmental and antigen challenge clinical trials,
azelastine was demonstrated to be signicantly effective in
adults and children of at least 4 years of age, and to be at least
as effective as levocabastine.124127 The duration was shown to
be at least 8 h. The most signicant side effect with azelastine
is an unpleasant taste following instillation.
Epinastine is a new generation histamine H1-receptor antagonist with mast cell stabilizing activity and no effect on
muscarinic receptors.128 Epinastine was shown to suppress
allergic inammation not only through its strong antihistamine
and antimediator effects, but also by inhibiting eosinophilic
chemotaxis and the expression of adhesion molecules involved
in chemotaxis.129,130 Its safety and efcacy have been investigated in the clinical conjunctival allergen challenge model, and
in patients with active seasonal allergy, where it was shown to
rapidly and signicantly inhibit hyperemia, chemosis and lid
swelling for at least 8 h.131,132
HISTAMINE H1-RECEPTOR ANTAGONISTS

CHEMISTRY
Histamine receptors were dened pharmacologically by the
actions of their agonists and antagonists. Histamine H1receptor antagonists pharmacologically compete with histamine
at the H1-receptor site on effector cells and have been classied
by their chemical structures into six groups: ethylenediamines,
ethanolamines, alkylamines, phenothiazines, piperazines, and
piperadines (Table 25.1). The H1-receptor antagonist compounds can be described by the general structure shown in
Figure 25.1. These compounds are composed of one or two
aromatic (heterocyclic) rings connected via a nitrogen, carbon,
or oxygen atom (X) to the ethylamine group. The nitrogen atom
of the ethylamine group is tertiary that is, it has two substituents. The H1-receptor antagonists are structurally similar
to histamine in that they both contain an ethylamine group.
However, histamine consists of a single heterocyclic ring, in this
case imidazole, which is connected directly to the ethylamine
group. Unlike that of the H1-receptor antagonists, the nitrogen
atom of the ethylamine group is primary or unsubstituted.
STRUCTUREACTIVITY RELATIONSHIP
The H1-receptor antagonists possess two chemical moieties
that determine the pharmacokinetic properties of this group of
drugs and thereby confer pharmacologic activity (Table 25.2).
The H1 antihistamines contain multiple aromatic rings, which
make these compounds very lipophilic and contribute to
receptor site binding via hydrophobic forces.133 The second
functional moiety is the positively charged side chain, which is
usually an ammonium group. Both histamine and the H1receptor antagonists share an amino group that is believed
to be important for H1-receptor recognition.134 Table 25.2
demonstrates the chemical structural similarities and
differences between histamine and the H1- and H2-receptor
antagonists.
MECHANISM OF ACTION
The H1-receptor antihistamines act by occupying H1 receptor
on effector cells. Binding of antagonists to the receptor site does
not initiate a response in the effector cell; rather, it prohibits
histamine from binding. Therefore, histamine is unable to
cause an effector cell response. The binding of the H1-receptor
antagonist is a reversible, competitive equilibrium reaction and
is determined by the relative concentrations of histamine and
H1-receptor antagonist in the area of the receptor site. To ensure
effective blockade of the H1-receptor, the antihistamine concentration should be sufciently high to compete with tissue
histamine levels created by local mast cell degranulation.
CHAPTER 25
DUAL ACTION ANTIHISTAMINE/MAST CELL

STABILIZERS
PHARMACOKINETICS: ABSORPTION,
DISTRIBUTION, BIOTRANSFORMATION, AND
ELIMINATION
The majority of H1-receptor antagonists are chemically stable
and do not contain labile ester or amide moieties. The equilibrium constant of the base and its conjugate acid of the antihistamine compounds is greater than 8.0. Thus, at physiologic
pH, all of the compounds would be at least 90% protonated
and water-soluble. As a result of their basic properties, the
H1-receptor antihistamines may be administered orally.
Following oral administration, the drugs are rapidly absorbed
and render symptomatic relief beginning within 1530 min.
The duration of action usually is 36 h. The H1 receptorblocking agents are widely distributed in body tissues and
cross the bloodbrain barrier. The compounds are metabolized
in the liver and excreted in the urine within 24 h of an oral
dose.135,136
Little information is available on the pharmacokinetics of
topically applied ocular H1 antihistamines. These drugs are
administered to the ocular surface via application of watersoluble salts; maleate salts and phosphoric acid are most
commonly used in ocular preparations. Currently, only three H1
antihistamines are approved for use in the eye; these include
269
SECTION 4
TABLE 25.1 The Six Major Groups of Classic H1 Antihistamines

Linkage Atom
General Class
Other Members
General Comments
Ethylenediamines
Antazoline
Methapyrilene
Tripelennamine
Relatively weak CNS effects, but drowsiness may occur

in some patients; gastrointestinal side effects
common
Ethanolamines
(aminoalkyl ethers)
Bromodiphenhydramine
Carbinoxamine
Clemastine
Dimenhydrinate
Diphenylpyraline
Doxylamine
Phenytoloxamine
Significant antimuscarinic activity; CNS depression

common in about half of the patients using
members of this group; relatively low incidence
of gastrointestinal side effects
Alkylamines
(propylamine derivatives)
Brompheniramine
Dexbrompheniramine
Dexchlorpheniramine
Dimethindene
Pheniramine
Pyrrobutamine
Triprolidine
Cause less CNS depression than members of other

groups; some CNS stimulation possible; best classic
group of antihistamines for daytime use
N (in phenothiazine
ring)
Phenothiazines
Methdilazine
Trimeprazine
Sedative effects very prominent with this class; most

have pronounced antimuscarinic activity; usually used
primarily as antiemetics
N (in piperazine
ring)
Piperazines
Buclizine
Chlorcyclizine
Hydroxyzine
Meclizine
Degree of sedation and antimuscariniceffects produced

by this class is relatively mild; buclizine, cyclizine,
and meclizine are used for treating motion sickness;
hydroxyzine is used as sedative, tranquilizer, and
antiemetic
N (in piperidine
ring)
Piperidines
Azatadine
Phenindamine
Sedative potential is comparable to that of the

ethylenediamine class; drowsiness is most common
side effect
From Trzeciakowski JP, Mendelsohn N, Levi R: Antihistamines. In: Middleton E, Reed CE, Ellis EF, et al, eds. Allergy principles and practice. 3rd edn. St Louis, MO: CV
Mosby; 1988.
TABLE 25.2 Chemical Differentiation between Histamine and

Its Respective Receptor Antagonists
H2 Antagonist
Histamine
H1 Antagonist
Imidazole for
analogous ring
Imidazole
Aryl rings
Hydrophilic
Hydrophilic
Lipophilic
Thiourea or guanidine
Ammonium
Ammonium (or similar

group)
Preferably uncharged
Charged
Charged
From Ganellin CR: Chemistry and structureactivity relationship of H2-receptor

antagonists. In: Rocha e Silva M, ed. Handbook of experimental pharmacology.
Histamine II and antihistaminics: chemistry, metabolism, physiological, and
pharmacological actions. New York: Springer; 1978.
PHARMACOLOGIC PROPERTIES
FIGURE 25.1. A comparison of the chemical structure of histamine
(top) and of H1-receptor antagonists (bottom).
270
pheniramine maleate, antazoline phosphate, and pyrilamine

maleate. These preparations are well distributed in the preocular
tear lm and seem to have excellent penetration into the conjunctival epithelium and substantia propria. Systemic absorption occurs via drainage through the nasal lacrimal duct with
subsequent absorption by the nasopharyngeal and oropharyngeal mucosal surfaces.
The pharmacologic actions of the H1-receptor antagonist

subclasses are similar: They block the effects of histamine
mediated by the H1 receptors on effector cells. The effects of
histamine on the vascular system are mediated by both H1 and
H2 receptors.137 Stimulation of H1 receptors causes systemic
vasodilation as well as localized cutaneous erythema due to
capillary dilation.41 However, when H1 receptor-blocking agents
are administered alone, the systemic hypotension caused by
histamine-induced vasodilation is only partially blocked. When
H1- and H2-receptor blockers are given concurrently prior to
histamine challenge, the fall in blood pressure is negated. Cutaneous capillary permeability is increased after local injection of
histamine, resulting in the formation of edema.39 H1-receptor
antihistamines antagonize this action of histamine and inhibit

the egress of plasma through capillary walls.
Histamine has a direct constrictor action on smooth muscle.
In humans, histamine-induced bronchoconstriction of respiratory smooth muscle can be blocked with prophylactic administration of H2-receptor antagonists.138 In animal species, in vivo
experiments have demonstrated histamine-induced contraction
of gastrointestinal smooth muscle. The guinea pig ileum model
had been used to provide early evidence for the effects of
histamine and to document the presence of specic histaminereceptor subtypes. In addition, this animal model had been used
to test various types of H1-receptor antihistamines as these
agents were developed.
In the eye, topical application of histamine induces ocular
itching and conjunctival vasodilation. It has been demonstrated
that the H1 receptors mediate the symptoms of itching, whereas
conjunctival vasodilation is mediated by both H1 and H2
receptors.55,56,65 Pretreatment with topical H1 antihistamines
blocks the histamine-induced itching and decreases the amount
of conjunctival hyperemia.
Many of the H1 antihistamines possess pharmacologic properties unrelated to H1-receptor blockade. These agents possess
varying degrees of anticholinergic activity that is dose dependent
and varies among the subclasses. The anticholinergic action has
been used in treating several diseases, including motion
sickness, vertigo resulting from vestibular disorders, and rigidity
associated with Parkinsons disease.
Several H1-receptor antagonist compounds have been demonstrated to possess local anesthetic action.139 However, this effect
occurs only with concentrations several orders of magnitude
greater than the pharmacologic dosages employed to block the
H1 receptor. In eyes pretreated with antazoline phosphate,
itching was blocked after topical histamine challenge, whereas
corneal sensation was shown not to be decreased by
anesthesiometry.65
ADVERSE EFFECTS
SYSTEMIC ADMINISTRATION
Therapeutic doses of oral H1 antihistamines may be associated
with mild systemic side effects; however, occasionally the untoward responses may necessitate drug withdrawal. The most
common adverse effect observed with H1-receptor antagonists is
sedation, which varies between the drug subclasses and individual patient response.140 Although sedation may not be
problematic when medication is administered upon retiring
for the night, this soporic effect may lead to potentially
life-threatening accidents in patients who drive or operate
heavy automated machinery. Other central nervous system
(CNS) side effects include disturbed coordination, dizziness,
fatigue, and difculty in concentration, which result from a
generalized depression of the CNS. Paradoxically, patients
may also experience euphoria, nervousness, insomnia, and
tremors.
Gastrointestinal adverse effects occur less frequently and
include loss of appetite, nausea, vomiting, epigastric distress,
and constipation or diarrhea. Occasionally, these symptoms are
diminished by administering oral H1 antihistamines with meals.
Several less-frequent side effects of the H1-receptor antagonists
are attributable to their anticholinergic properties. Patients may
note dryness of the mucous membranes of the oropharynx and
the appearance of dry eye symptoms that may lead to contact
lens intolerance or frank keratoconjunctivitis sicca. Other
atropine-like effects include mydriasis that could precipitate an
attack of acute angle-closure glaucoma in untreated, predisposed
persons. Ciliary muscle paresis with an associated decrease in
accommodation may account for visual difculties experienced

by some patients.
Systemic H1 antihistamines should be used judiciously in
young children; acute poisoning may result from an inability to
metabolize the drugs rapidly and may produce dangerously high
blood concentrations. The CNS effects of the H1-receptor antagonists constitute the greatest danger to children, and the constellation of signs and symptoms are related to anticholinergic
activity as evidenced by excitement, nervousness, irritability,
incoordination, insomnia, and tremors.141 Other signs assoiated with cholinergic blockage are xed and dilated pupils,
facial ushing, and elevated body temperature.
Safety in pregnancy for humans has not been established for
systemic H1 antihistamines.142 However, the piperazine compounds may have teratogenic effects.
The use of systemic antihistamines for the treatment of
ocular allergy is controversial. Ocular allergy is a topical disease
with typical anatomical and pharmacological conditions for
convenient local delivery.143 Topical antiallergy eyedrops provide
faster relief of ocular symptoms compared with oral agents,
because the former are delivered directly onto the target tissue
at a higher concentration. In contrast, allergic rhinitis is an
equally frequent condition generally treated with systemic antihistamines, which have been proven effective in relieving nasal
signs and symptoms. First generation oral H1-receptor antagonists may provide some relief of ocular itching, but are
sedating and possess anticholinergic effects such as dry mouth,
dry eye, blurred vision, and urinary retention. The second
generation oral H1 antihistamines offer the same efcacy as
their predecessors, but with a low-sedating prole and lack of
anticholinergic activity. These drugs attentuate the early phase
and some of the features of the late phase ocular response,
including swelling and redness. Second generation antihistamines include acrivastine, cetirizine, ebastine, fexofenadine,
loratadine, and mizolastine. Desloratadine and levocetirizine
are considered a further evolution of the second generation
agents. Nevertheless, in a recent study, the most successful
systemic antihistamine, loratadine, was shown to be inferior to
local, topical antiallergy therapy in alleviating the signs and
symptoms of allergic conjunctivitis.144
CHAPTER 25
TOPICAL OCULAR ADMINISTRATION

Topical administration of H1-receptor antagonists in the eye has
been associated with a low incidence of systemic adverse effects.
However, these agents are available only in combination with
sympathomimetic decongestant agents that have been associated with systemic side effects. Ocular medications gain
access to the systemic circulation via absorption through the
nasal and oropharyngeal mucosae. Therefore, combination drugs
should be used with caution in patients with poorly controlled
hypertension, cardiovascular disease with arrhythmias, and
poorly controlled diabetes mellitus. Additionally, patients using
monoamine oxidase inhibitors for hypertensive disease may
suffer a hypertensive crisis if administered a topical sympathomimetic decongestant agent.145,146
Pupillary mydriasis may be induced by either component of
an H1-receptor antagonist/decongestant combination and may
trigger an attack of acute angle-closure glaucoma. The combination drugs have not been evaluated for safety during pregnancy.
PREPARATIONS AND DOSAGES

Currently, both prescription and over-the-counter H1-receptor
antagonist antihistamine agents are available to treat disease.
Three over-the-counter H1-receptor antagonist antihistamines
are available for topical ocular administration and are produced
271
TABLE 25.3 AntihistamineDecongestant Combinations

Generic Name
Commercial
Preparations
Recommended
Dosage
Antazoline PO4 (0.5%)

Naphazoline
HCl (0.05%)
Albalon A
Vasocon A
12 drops/eye q34h
or less to relieve
symptoms
Pheniramine maleate
(0.3%)
AK-Con A
12 drops/eye q34h
or less to relieve
symptoms
Naphazoline HCl
(0.025%)
Opcon A
Naphcon A
Pyrilamine maleate
(0.1%)
Prefrin-A
Prefrin-A
12 drops/eye q34h
or less to relieve
symptoms
Phenylephrine (0.12%)
From Pavan-Langston D, Dunkel EC: Handbook of ocular drug therapy and
ocular side effects of systemic drugs. Boston, MA: Little, Brown; 1991.
TABLE 25.4 Ocular Decongestants, DecongestantAstringents,

and DecongestantAntibacterials
Generic Name
Commercial
Preparations (drops)
Recommended
Dosage (714 Days)
Naphazoline
(0.12% [Rx];
0.012% [OTC])
Albalon (OTC)
Clear Eyes (OTC)
Degest-2 (OTC)
Opcon (OTC)
Naphcon (OTC)
Naphcon Forte (Rx)
Vasoclear (OTC)
Vasocon Regular
(OTC)
1 drop/eye q34h or
less to relieve
symptoms
Phenylephrine
(0.12%)
AK-nephrine (OTC)
Prefrin (OTC)
Relief (OTC)
1 drop/eye q34h or
less to relieve
symptoms
Tetrahydrozoline
HCl (0.05%)
Collyrium (OTC)
Murine PLUS (OTC)
Visine (OTC)
1 drop/eye q34h or
less to relieve
symptoms
Decongestants
SECTION 4
DecongestantAstringents
only as combination antihistamine/decongestant preparations.

The H1 antihistamines are 0.5% antazoline phosphate, 0.3%
pheniramine maleate, and 0.1% pyrilamine maleate and are
found in combination with either 0.0250.05% naphazoline
hydrochloride or 0.012% phenylephrine (Table 25.3). Each of
the three H1 antihistaminic agents is efcacious in reducing the
chemosis and itching associated with allergic conjunctivitis.
The decongestant agents are included for their vasoconstrictor
properties and are efcacious in relieving conjunctival injection.
The recommended dosage is one to two drops instilled in the
eye up to four times daily as needed to control symptoms.
New H1-selective receptor antagonist agents have been
introduced that block both the itching and redness associated with
allergic conjunctivitis. Levocarbastine (Livostin) is a potent new
topical ocular H1-receptor antagonist that has been demonstrated
to effectively control the symptoms of allergic conjunctivitis.147 A
topical preparation of 0.05% levocarbastine hydrochloride
administered prior to conjunctival histamine challenge effectively
prevented itching, conjunctival injection, and chemosis.148 In
conjunctival antigen challenge (CAC) studies, 0.05%
levocarbastine hydrochloride has been shown to be more effective
than placebo and 4% DSCG in inhibiting itching, hyperemia,
eyelid swelling, chemosis, and tearing after allergen challenge.66,149
A second H1-selective receptor antagonist agent, 0.1% olopatadine (Patanol), has been added to the armamentarium to treat
allergic conjunctivitis. In CAC studies, 0.1% olopatadine has been
shown to be more effective than placebo in inhibiting itching and
redness after antigen challenge.150 Additionally, the recommended
dosing schedule of 0.1% olopatadine is twice daily, and it has
been approved for use in children at least 3 years of age.
In addition to the antihistamine/decongestant preparations,
several over-the-counter and prescription decongestant preparations and decongestant/astringent combinations are available
(Table 25.4). These agents may be used in circumstances of
mild ocular irritation or allergic conditions and are effective in
reducing conjunctival injection and clearing mucus from the
ocular surface.
HISTAMINE H2-RECEPTOR ANTAGONISTS

CHEMISTRY
272
The H2-receptor antagonists were born of the idea to develop

compounds that would block those responses induced by
Naphazoline
(0.02%)
Zinc SO4 (0.25%)
Vasoclear A (OTC)
12 drops/eye up to
4 times daily
Phenylephrine HCl
(0.12%)
Zinc SO4 (0.25%)
Visine AC (OTC)
12 drops/eye up to
4 times daily
Tetrahydrozoline
HCl (0.05%)
Zinc SO4 (0.25%)
Zincfrin (Rx)
12 drops/eye up to
4 times daily
DecongestantAntibacterials
Phenylephrine HCl
(0.12%)
Sulfacetamide
Na (15%)
Vasosulf (Rx)
12 drops/eye q4h
12 drops/eye q4h
Modified from Pavan-Langston D, Dunkel EC: Handbook of ocular drug therapy

and ocular side effects of systemic drugs. Boston, MA: Little, Brown; 1991.
histamine that could not be blocked by the currently available

H1-receptor antagonists. The H2-receptor antagonists were
synthesized by a series of modications of the histamine
molecule and therefore have a structural relationship to histamine (Table 25.5). The rst selective H2-receptor antagonist,
burimamide, was synthesized in 1969 by substituting bulkier,
uncharged side chains to the imidazole ring.36 Subsequently,
two imidazole ring congeners metiamide,151 a thione-containing
compound, and cimetidine,152 a cyanimino compound were
developed. More recently, ranitidine, a furan derivative, has
become available.153 Each of these compounds contains a polar
heterocyclic ring in its side chain.
The H2-receptor antagonists bear a closer structural relationship to histamine than do the H1-receptor antagonists.
Burimamide, metiamide, and cimetidine have an imidazole ring
and are polar, hydrophilic compounds similar to histamine (see
Table 25.2). It appears that the imidazole or another heterocyclic, side chain ring is critical for H2-receptor site recognition
and plays a role in determining drug activity.133 The H2-receptor
antagonist compounds have similar equilibrium constants (pKa
values of ~14). These drugs are weak bases and highly watersoluble; thus, they exist in the uncharged form in aqueous
solutions under physiologic conditions (pH of 7.4).133
Structure
and Name
Ring Type
Relative Antagonist
Potency
Imidazole
0.001
Durant et al
Durant et al
Ganellin
Imidazole
0.1
Black et al
Durant et al
Ganellin
Imidazole
~1
Black et al
Durant et al
Forrest et al
Ganellin
Reference
Imidazole
Brimblecombe et al
Durant et al
Ganellin
Furan
35
Brittain and Daly
Imidazole
14
Blakemore et al
Mills et al
From Trzeciakowski JP, Mendelsohn N, Levi R: Antihistamines. In: Middleton E,

Reed CE, Ellis EF, et al, eds. Allergy principles and practice. 3rd edn. St Louis,
MO: CV Mosby; 1985.
MECHANISM OF ACTION
The H2-receptor antagonists work in a manner similar to that
of the H1-receptor antihistamines. These agents bind reversibly
and competitively to the histamine H2 receptors on effector
cells. When bound to the receptor site, the H2-receptor antagonist agents do not elicit a tissue response and block the effect
of histamine.
ELIMINATION
Cimetidine, the prototype of the H2-receptor antagonist drugs, is
well absorbed after oral administration. After an oral dose, peak
blood concentrations are reached in ~6090 min with good tissue
distribution throughout the body.136 The one exception is the
CNS; cimetidine penetrates the CNS poorly because the
compound is poorly lipophilic. The drug has been found to
cross the placental barrier and is excreted in breast milk.136
The majority of an oral dose of cimetidine is excreted in
the urine, with a minor portion handled in the bile and by hepatic
microsomal biotransformation. In patients with normal renal
function, the plasma half-life (t1/2) is ~2 h. However, the t1/2
increases in patients with impaired hepatic or renal function.154
Cimetidine and the other H2-receptor antagonist antihistamines are selective in their action and block the effects of
histamine mediated through the H2-receptor. The most noteworthy systemic effect is the ability of these agents to inhibit
gastric secretion induced by histamine, gastrin, or pentagastrin
in humans.155157 Cimetidine inhibits all phases of physiologic
secretion of gastric acid. In humans, a single 300-mg dose
decreases the fasting secretion of gastric acid and decreases the
amount of acid induced by food or via vagal stimulation.136
When given intravenously in high doses, cimetidine may
cause bradycardia and hypotension. However, when given to
normal volunteer subjects, the cardiovascular changes were
minor.155 As previously mentioned, systemic administration of

histamine caused vasodilation and severe hypotension that
were completely blocked only by the concurrent use of both H1and H2-receptor antihistamines.
In the eye, stimulation of H2-receptors with a selective H2
agonist produced diffuse conjunctival vasodilation.56 Cimetidine has been the only H2-receptor antagonist to be formulated
into an ophthalmic preparation. Studies have shown that the
addition of an H2-receptor antagonist to a classic H1 antihistamine reduced the amount of conjunctival vasodilation in
response to histamine challenge.75
ADVERSE EFFECTS
The H2-receptor antagonists are generally well tolerated when
taken systematically. The side effects of cimetidine are minor,
seldom posing a serious problem, and include headaches, fatigue,
myalgias, constipation, and skin rashes. The CNS-depressive
effects seen with the H1 antihistamines are not seen with the H2receptor blockers, because these compounds are hydrophilic and
penetrate the bloodbrain barrier poorly. However, cimetidine has
been associated with confusion, delirium, and convulsions, usually
occurring in patients with concurrent liver or kidney disease.
Cimetidine possesses weak antiandrogenic effects and has
been responsible for reports of gynecomastia in men and
galactorrhea in women. These effects have occurred in patients
treated for an extended length of time. Cimetidine has been
demonstrated to release prolactin when given in large intravenous doses.158,159 There have been sporadic reports in the
literature of bone marrow suppression associated with cimetidine therapy. Patients have experienced leukopenia, thrombocytopenia, and hemolytic anemia, which seems to be an
idiosyncratic reaction.160
Cimetidine is metabolized partially by the hepatic microsomal
enzyme system and therefore may impair the elimination of
drugs that are catabolized in this manner. These drugs include
oral anticoagulants,161 theophylline,162 benzodiazepines,163 and
propranolol.164 Additionally, the pharmacokinetics of calcium
channel blockers are altered by cimetidine.165
CHAPTER 25
TABLE 25.5 Representative Histamine H2-Antagonists

Compared with Histamine

Currently, no H2-receptor antihistamines are approved for ocular
use. However, it is conceivable that combination drops consisting of H1 and H2 antagonists have a place in the treatment
of ocular allergic disorders. Studies have shown that combination drops have a synergistic effect in reducing conjunctival
vasodilation and chemosis when compared with the individual
agents alone. Topical epinastine, a dual acting molecule with
mast cell stabilizing and antihistaminic effect, has also an H2
antagonism.

Studies evaluating cimetidine as a topical ocular preparation
have found concentrations of 0.1%, 0.5%, and 1.0% to be well
tolerated and efcacious in reducing ocular symptoms induced
by histamine challenge.
MAST CELL STABILIZERS

CHEMISTRY
The rst mast cell-stabilizing compound was developed in the
late 1960s from khellin, a chromone (benzopyrene) derived from
273
FIGURE 25.2. The chemical structure of disodium cromoglycate

(cromolyn sodium) (1,3-bis(2-carboxychromon-5-yloxy)-2hydroxypropane).
SECTION 4
Ammi visnaga, an eastern Mediterranean plant.166 Successive

modications in structure yielded several bis-chromone compounds, one of which was DSCG. DSCG is the disodium salt of
1,3-bis(2-carboxychromon-5-yloxy)-2-hydroxypropane (Fig. 25.2).
The compound is composed of two chromone rings joined by a
exible carbon chain, with each ring possessing a polar carboxyl
group. The compound is an odorless, white, dehydrated crystalline powder that is moderately soluble in water but practically
insoluble in alcohol.167 The drug was rst discovered to have
antiasthma properties when Altounyan demonstrated on himself that cromolyn could afford protection against an asthmatic
attack induced by bronchial provocation with pollen antigens.83
Disodium cromoglycate forms complexes with divalent cations,
including magnesium (Mg2+), calcium (Ca2+), strontium (Sr2+),
barium (Ba2+), zinc (Zn2+), and manganese (Mn2+) when placed
in organic solvents. These complexes are formed by an electrostatic interaction between the two carboxyl groups of DSCG
and the divalent cations with a 1:1 stoichiometry.168 Although
cromolyn has been associated with reduced calcium ux across
the mast cell membrane, chelation of calcium by cromolyn does
not fully account for the drugs ability to inhibit mast cell
degranulation. It has been demonstrated that DSCG interacts
with a membrane-bound cromolyn receptor, which is a calciumtransporting protein necessary for the secretion of histamine.
This interaction requires the presence of calcium ions in order
to proceed.169
MECHANISM OF ACTION
274
It had been thought that DSCG possessed membranestabilizing features in that the drug somehow modied the mast
cell membrane to prevent histamine release in the presence of
IgE antibody. When cromolyn was discovered, little was known
of its mechanism of action. However, since the 1980s, evidence
from research has shed light on the interaction between this
drug and the mast cell. In 1980, Mazurek and co-workers
identied a binding site on mast cells and basophils for
DSCG.169 The authors identied the cromolyn receptor as a
membrane-binding protein that required the presence of
calcium ions for the interaction to proceed. The evidence from
the experiments suggests that the cromolyn-binding protein is a
calcium-transporting protein that is necessary for the secretion
of histamine after stimulation by an IgE antibodyantigen interaction.170 It is theorized that the membrane-bound cromolynbinding protein interacts with the Fc receptors for IgE in such a
way that cross-linking of the Fc receptors does not occur upon
antigen binding to the IgE molecule.171
Also in 1980, other researchers examined the association
between DSCG and protein phosphorylation in the activation
and regulation of histamine secretion in mast cells. Theoharides
and associates demonstrated that cromolyn induced phosphorylation of a 78 000-Da mast cell protein.172 These authors
presented compelling data suggesting that DSCG and phosphorylation of the membrane-bound protein are intricately involved
in the regulation of histamine secretion. The concentration
range over which DSCG induced phosphorylation of proteins
was similar to that for cromolyn-induced inhibition of histamine release stimulated by compound 48/80. Additionally, both
activation of phosphorylation and inhibition of secretion by
DSCG demonstrated tachyphylaxis that is, a second exposure
to cromolyn failed to induce phosphorylation in mast cells that
were pretreated with the drug. Lastly, dephosphorylation after
cromolyn-induced phosphorylation of the 78 000-Da protein had
a time course identical to that of the loss of sensitivity of mast
cells to the inhibition of histamine release caused by cromolyn.
The mechanism of protein phosphorylation has not been
elucidated; however, it has been shown that cyclic guanosine
monophosphate can phosphorylate the same 78 000-Da mast
cell protein as cromolyn does. Thus, it has been theorized that
cromolyn may act via a cyclic guanosine monophosphatedependent protein kinase.173 This is not surprising in that
DSCG has been identied as an inhibitor of cyclic nucleotide
phosphodiesterase.174
ELIMINATION
DSCG is poorly absorbed from the gastrointestinal tract after
oral administration. Therefore, it is available as an inhalant
that can be administered via the nasal or respiratory tract.
When given as an inhaled dose, ~8% is absorbed systemically
through the bronchial tree.175 The half-life (t1/2) of the compound
is ~80 min, with more than 98% being eliminated within
24 h.136 Cromolyn is not metabolized and is excreted unchanged
in the urine and bile.
Little information is available on the pharmacokinetics of
topically applied ocular disodium cromoglycate. Cromolyn is
administered to the ocular surface via application of a watersoluble solution.
Currently, two mast cell-stabilizing drugs are approved for
use in the eye: 4% DSCG (Crolom) and 0.1% lodoxamide
tromethamine (Alomide). Both preparations are well distributed
in the preocular tear lm and seem to adequately penetrate the
conjunctival epithelium and substantia propria. When
administered to normal volunteer subjects, ~0.03% of DSCG
was absorbed following an ocular dose.
The pharmacologic actions of mast cell stabilizers result from
the ability of the drug to bind to membrane-bound protein
receptors on mast cells. This interaction inhibits histamine
release when IgE-primed mast cells are challenged with antigen.
Mast cell stabilizers do not interfere with the binding of IgE to
the Fc receptors on mast cells or the interaction between mast
cell-bound IgE and antigen. Mast cell stabilizers have no bronchodilator, antiinammatory, or anticholinergic activity; rather,
they suppress the mast cell secretory response to antigen. Thus,
the drug is effective only when given prophylactically prior to an
antigenIgE antibody interaction.
Inhaled DSCG is recognized as an effective prophylactic drug
for the treatment of asthma.8486 Cromolyn has also been
demonstrated to have salutary effects in patients with food
allergy,87 systemic mastocytosis,88 and seasonal allergic
rhinitis.89,90 However, it should be noted that the effectiveness
of the drug in these conditions remains controversial.
ADVERSE EFFECTS
Therapeutic doses of DSCG are well tolerated by patients. Most
adverse reactions are mild and are associated with a direct irritant
effect of the powder on the bronchial tree, including bronchospasm,
wheezing, cough, sneezing, nasal congestion, and pharyngeal
irritation.136 Other adverse effects have been documented in
case reports and consist of dermatitis, gastroenteritis, myositis,
urethral burning, and pulmonary allergic granulomatosis.176179
DSCG has no known effect on pregnancy in laboratory
animals; however, safety for human use during pregnancy has
not been established, and no controlled human studies have
been performed.136 It is not known whether the drug is excreted
in human breast milk, and the safety and efcacy of cromolyn
have not been established in children younger than 4 years.

Topical administration of mast cell stabilizers in the eye has
been associated with a low incidence of systemic adverse effects.
Ocular side effects are common but usually mild and selflimited. Ocular administration of cromolyn has been associated
with transient stinging and conjunctival injection. Other local

adverse reactions include chemosis and ocular and periocular
itching and irritation.

Currently, the only mast cell-stabilizing drugs formulated for
topical ocular use are cromolyn sodium 4%, nedocromil sodium
2%, lodoxamide tromethamine ophthalmic solution 0.1%, spaglumic acid 4%, and pemirolast potassium ophthalmic solution
0.1%. Four percent disodium cromoglycate contains 40 mg of
DSCG in puried water with a preservative and is a clear, colorless, sterile solution with a pH of 4.07.0. The recommended
dosage is one to two drops instilled in the eye four times daily.
One drop of the solution contains ~1.6 mg of DSCG. The 0.1%
lodoxamide tromethamine contains 1.78 mg of lodoxamide
tromethamine in puried water with EDTA, benzalkonium
chloride, and other inactive ingredients. This preparation has
been shown to be 2500 times more potent than DSCG180 and
has demonstrated satisfactory results in patients with AKC and
GPC.181 The recommended dosage is one drop applied four times
daily, although patients have been able to use 0.1% lodoxamide
tromethamine twice daily and still remain asymptomatic.
Because therapy with mast cell stabilizers is prophylactic, it
is advisable to initiate treatment before the onset of allergic
symptoms. It is not unexpected that symptomatic response to
treatment may take up to 2 weeks with DSCG and up to 4 days
with 0.1% lodoxamide tromethamine. Once therapy has
commenced, it should be continuous and maintained even after
symptomatic improvement.
NEW THERAPY FOR OCULAR ALLERGY

Since the 1980s, researchers have explored different methods to
block the allergic response in type I hypersensitivity reactions.
It was discovered that Fc fragments from IgE antibody could
competitively inhibit IgE binding to effector cells and block the
PrausnitzKstner reaction when preinjected into skin.182
HEPP (pentigetide), a synthetic pentapeptide derived from the
Fc region of human IgE, was developed and consisted of an
amino acid sequence of aspartyl-seryl-aspartyl-prolyl-arginine.
In tests on atopic persons, HEPP blocked the PrausnitzKstner
reaction;183 however, its mechanism of action remains unknown.
In a double-blind, randomized, parallel study, 0.5% pentigetide
(Pentyde) ophthalmic solution was compared with 4% DSCG in
patients with allergic conjunctivitis.184 After a 2-week comparison, patients treated with 0.5% pentigetide experienced
signicant improvement in conjunctival hyperemia, chemosis,
tearing, and itching. With further study, this drug may prove to
be a useful adjunct in the treatment of allergic conjunctivitis.
As mentioned previously, other mediators of inammation
contribute to and help perpetuate the ocular allergic response.
Several classes of pharmacologic agents have demonstrated
efcacy in blocking the effects of these mediators of inammation, and hence possess antiallergic properties when used as
ocular preparations. In CAC studies, topical nonsteroidal antiinammatory drugs such as 0.5% ketorolac tromethamine
(Acular), 0.03% urbiprofen sodium (Ocufen), and 0.1% diclofenac sodium (Voltaren) and topical corticosteroid agents such
as 0.5% loteprednol etabonate (Lotemax) and 1% rimexolone
(Vexol) have demonstrated effectiveness in controlling the signs
and symptoms of allergic conjunctivitis. Researchers are actively
investigating compounds that blunt the response to or inhibit
the action of these inammatory mediators. In the future, we
expect to have available topical medications such as antiplatelet
activating factor and leukotriene inhibitors to add to our list of
antiallergic drugs.
CHAPTER 25
In the eye, 0.1% lodoxamide tromethamine62,63 and DSCG91,92

have shown effectiveness in relieving the signs and symptoms
of VKC. The latter helped to reduce the frequency of steroid use
in patients with VKC. When these two drugs were compared in
a multicenter, double-masked, parallel-group clinical study,
0.1% lodoxamide was found to be statistically superior to 4%
cromolyn in alleviating itching, tearing, foreign body sensation,
and discomfort in patients with VKC.62
Likewise, clinical studies have demonstrated encouraging
results with DSCG in acute allergic conjunctivitis. In general,
investigators have reported satisfactory results with DSCG in
acute allergic conjunctivitis. Greenbaum and associates93 conducted the rst environmental study evaluating 4% DSCG in a
double-blind, placebo-controlled fashion and reported that eye
symptom scores for patients receiving DSCG were signicantly
lower when compared with the previous years ragweed season.
In a double-masked, placebo-controlled, parallel-group prospective environmental study, Friday et al94 demonstrated
that 4% DSCG was a safe and effective method of controlling
the symptoms of ragweed conjunctivitis in patients with
serum IgE levels less than 100 ng/mL. Patients with serum IgE
levels greater than 100 ng/mL did not experience a signicant
improvement in symptoms. Leino and Tuovinen95 evaluated
DSCG in 33 patients with VKC, allergic conjunctivitis, or
chronic conjunctivitis in a prospective uncontrolled study.
The authors reported a benecial effect associated with the
use of DSCG; however, regression of the signs and symptoms
varied widely.
Two well-designed, double-blind, placebo-controlled, comparative environmental studies reported that DSCG suppressed
allergic eye symptoms in specic groups of patients identied by
serum IgE antibody levels. However, the results were contradictory. Welsh et al96 showed that 4% DSCG caused a signicant reduction in eye itching and irritation in subjects whose
preseasonal IgE ragweed antibody level was less than 99 ng/mL;
patients with IgE levels exceeding 100 ng/mL did not experience
the same benet. Kray et al97 stratied their subjects by radioallergosorbent (RAST) scores to IgE antibodies, including
ragweed. These investigators noted a signicant suppression of
eye symptoms in subjects with class 3 or 4 RAST scores (higher
antibody level). Subjects with classes 0, 1, and 2 RAST scores
noted no signicant difference between DSCG and placebo.
275
SECTION 4
REFERENCES
276
1. Allansmith MR, Abelson MB: Ocular

allergies. In: Smolin G, Thoft RA, eds. The
cornea. Boston, MA: Little Brown;
1983:231243.
2. Irani AM, Butrus SI, Tabbara KF,
Schwartz LB: Human conjunctival mast
cells: distribution of MCT and MCTC in
vernal conjunctivitis and giant papillary
conjunctivitis. J Allergy Clin Immunol 1990;
86:3440.
3. Morgan SJ, Williams JH, Walls AF, et al:
Mast cell numbers and staining
characteristics in the normal and allergic
human conjunctiva. J Allergy Clin Immunol
1991; 87:111116.
4. Baddeley SM, Bacon AS, McGill JI, et al:
Mast cell distribution and neutral protease
expression in acute and chronic allergic
conjunctivitis. Clin Exp Allergy 1995;
25:4150.
5. Ishizaka T, Ishizaka K: Biology of
immunoglobulin E: molecular basis of
reaginic hypersensitivity. Prog Allergy 1975;
19:60121.
6. Siraganian RP, Hook WA, Levine BB: Specic
in vitro histamine release from basophils by
bivalent haptens: evidence for activation by
simple bridging of membrane bound
antibody. Immunochemistry 1975; 12:149.
7. Foreman FC, Hallett MB, Mangar J: The
relationship between histamine secretion
and 45calcium uptake by mast cells.
J Physiol (Lond) 1977; 271:193214.
8. Austen KF, Brocklehurst WE: Anaphylaxis in
chopped guinea pig lung. I. Effect of
peptidase substrate and inhibitors. J Exp
Med 1960; 133:521.
9. Hirata F, Axelrod J: Enzymatic synthesis
and rapid translocation of
phosphatidylcholine by two
methyltransferases in erythrocyte
membranes. Proc Natl Acad Sci USA 1978;
75:23482352.
10. Ishizaka T, Hirata F, Sterk AR, et al:
Bridging of IgE receptors activates
phospholipid methylation and adenylate
cyclase in mast cell plasma membranes.
78:68126816.
11. Lewis RA, Holgate ST, Roberts LJ II, et al:
Effects of indomethacin on cyclic
nucleotide levels and histamine release
from rat serosal mast cells. J Immunol
1979; 123:16631668.
12. Flowers RJ, Blackwell GJ: The importance
of phospholipase A2 in prostaglandin
biosynthesis. Biochem Pharmacol 1976;
25:285291.
13. Warner JA, Peters SP, Lichtenstein LM,
MacGlashan DW Jr: 3H arachidonic acid
incorporation and metabolism in puried
human basophils [abstract]. Fed Proc 1986;
45:735.
14. Hanahan DJ, Demopoulos CA, Liehr J,
Pinckard RN: Identication of naturally
occurring platelet activating factor as
acetyl-glyceryl-ether-phosphorylcholine
(AGEPC). J Biol Chem 1980;
255:55145516.
15. Chilton FH, Ellis JM, Olson SC, Wykle RL:
1-O-alkyl-2 arachidonoyl-sn-glycero-3phosphocholine: a common source of
platelet activating factor and arachidonate
in human polymorphonuclear leukocytes.
J Biol Chem 1984; 259:1201412019.
16. Barnes PJ: New concepts in the

pathogenesis of bronchial
hyperresponsiveness and asthma. J Allergy
Clin Immunol 1989; 83:1013.
17. Cook EB, Stahl JL, Barney NP, Graziano FM:
Mechanisms of antihistamines and mast
cell stabilizers in ocular allergic
inammation. Curr Drug Targets Inamm
Allergy 2002; 1:167180.
18. Dvorak AM, Galli SJ, Schulman ES, et al:
Basophil and mast cell degranulation:
ultrastructural analysis of mechanisms of
mediator release. Fed Proc 1983; 42:2510.
19. Dvorak AM, Hammel I, Schulman ES, et al:
Differences in behavior of cytoplasmic
granules and lipid bodies during human
lung mast cell degranulation. J Cell Biol
1984; 99:16781687.
20. Dvorak AM, Schulman ES, Peters SP, et al:
Immunoglobulin E-mediated degranulation
of isolated human lung mast cells. Lab
Invest 1985; 53:4556.
21. Macleod JD, Anderson DF, Baddeley SM,
et al: Immunolocalization of cytokines to
mast cells in normal and allergic
conjunctiva. Clin Exp Allergy 1997;
27:13281334.
22. Zhang S, Anderson DF, Bradding P, et al:
Human mast cells express stem cell factor.
J Pathol 1998; 186:5966.
23. Anderson DF, Zhang S, Bradding P, et al:
The relative contribution of mast cell
subsets to conjunctival T(H)2-like
cytokines. Invest Ophthalmol Vis Sci 2001;
42:9951001.
24. Bacon AS, Ahluwalia P, Irani AM, et al: Tear
and conjunctival changes during the
allergen-induced early- and late-phase
responses. J Allergy Clin Immunol 2000;
106:948954.
25. Cook EB, Stahl JL, Lowe L, et al:
Simultaneous measurement of multiple
cytokines in a single sample of human
tears using microparticle-based ow
cytometry. J Immunol Methods 2001;
254:109118.
26. Cook EB, Stahl JL, Miller ST, et al: Isolation
of human conjunctival mast cells and
epithelial cells: tumor necrosis factor-alpha
from mast cells affects intercellular
adhesion molecule 1 expression on
epithelial cells. Invest Ophthalmol Vis Sci
1998; 39:336343.
27. Stahl JL, Cook EB, Graziano FM, Barney NP:
Human conjunctival mast cells: expression of
Fc epsilonRI, c-kit, ICAM-1, and IgE. Arch
Ophthalmol 1999; 117:493497.
28. Windaus A, Vogt W: Syntse des
imidazolylathylamins. Ber Dtsch Chem Ges
1907; 3:3691.
29. Dale HH, Laidlaw PP: The physiologic
action of beta-imidazolethylamine. J
Physiol (Lond) 1910; 41:318344.
30. Dale HH, Laidlaw PP: Histamine shock.
J Physiol (Lond) 1919; 52:355390.
31. Best CH, Dale HH, Dudley HW, Thorpe WV:
The nature of vasodilator constituents of
certain tissue extracts. J Physiol (Lond)
1927; 62:397417.
32. Riley JF, West GB: The presence of
histamine in tissue mast cells. J Physiol
(Lond) 1953; 120:528537.
33. Schwartz LB, Austen KF: Structure and
function of the chemical mediators of mast
cells. Prog Allergy 1984; 34:271321.
34. Snyder SH, Axelrod J: Tissue metabolism

of histamine-14C in vivo. Fed Proc 1965;
24:774776.
35. Ash ASF, Schild HO: Receptors mediating
some actions of histamine. Br J Pharmacol
Chemother 1966; 27:427.
36. Black JW, Duncan WAM, Durant GJ, et al:
Denition and antagonism of histamine
H2-receptors. Nature 1972; 236:385390.
37. Oda T, Morikawa J, Saito Y, et al: Molecular
cloning and characterization of a novel type
of histamine receptor preferentially
expressed in leukocytes. J Biol chem 2000;
225:36783786.
38. Zhang M, Thurmond RL, Dunford PJ:
The histamine H(4) receptor: a novel
modulator of inflammatory and immune
disorders. Pharmacol Ther 2007;
113:594606.
39. Owen DAA, Poy E, Woodward DF:
Evaluation of the role of histamine H1- and
H2-receptors in cutaneous inammation in
the guinea-pig produced by histamine and
mast cell degranulation. Br J Pharmacol
1980; 69:615623.
40. Robertson I, Greaves MW: Responses of
human skin blood vessels to synthetic
histamine analogues. Br J Clin Pharmacol
1978; 5:319.
41. Harvey RP, Schocket AL: The effect of H1
and H2 blockade on cutaneous histamine
response in man. J Allergy Clin Immunol
1980; 65:136139.
42. McCusker MT, Chung KF, Roberts NM,
Barnes PJ: Effects of topical capsaicin on
the cutaneous responses to inammatory
mediators and to antigen in man. J Allergy
Clin Immunol 1989; 83:11181124.
43. Smith JA, Manseld LE, deShazo R,
Nelson HS: An evaluation of the
pharmacologic inhibition of the immediate
and late cutaneous effects to allergen.
J Allergy Clin Immunol 1980;
65:118121.
44. Abelson MB, Baird RS, Allansmith MR:
Tear histamine levels in vernal conjunctivitis
and other ocular inammations.
45. Levene RZ: Mast cells and amines in
normal ocular tissues. Invest Ophthalmol
1962; 1:531543.
46. Smelser GK, Silver S: The distribution of
mast cells in the normal eye: a method
of study. Exp Eye Res 1963; 2:134140.
47. Allansmith MR, Greiner JV, Baird RS:
Number of inammatory cells in the normal
conjunctiva. Am J Ophthalmol 1978;
86:250259.
48. Abelson MB, Soter NA, Simon MA, et al:
Histamine in human tears. Am J
Ophthalmol 1977; 83:417418.
49. Miller S, Cook E, Graziano F, et al: Human
conjunctival mast cell responses in vitro to
various secretagogues. Ocular Immunol
Inammation 1996; 4:3949.
50. Leonardi A: Role of histamine in allergic
conjunctivitis. Acta Ophthalmol Scan 2000;
78(Suppl 230):1821.
51. Abelson MB, Leonardi A, Fregona IA, et al:
Histaminase activity in patients with vernal
102:19581963.
52. Berdy GJ, Levene RB, Bateman ST, et al:
Identication of histaminase activity in
human tears after conjunctival antigen
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
challenge. Invest Ophthalmol Vis Sci 1990;

31(suppl):65.
Leonardi A, Smith LM, Fregona IA,
Secchi AG: Tear histamine and histaminase
during the early (EPR) and late (LPR)
phases of the allergic reaction and the
effect of lodoxamide. Eur J Ophthalmol
1996; 6:106112.
Abelson MB, Allansmith MR: Histamine and
the eye. In: Silverstein AM, OConnor GR,
eds. Immunology and immunopathology of
the eye. New York: Masson; 1979:362364.
Weston JH, Udell IJ, Abelson MB: H1
receptors in the human ocular surface.
20(Suppl):32.
Abelson MB, Udell IJ: H2-receptors in the
human ocular surface. Arch Ophthalmol
1981; 99:302304.
Friedlaender MH: Corticosteroid therapy of
ocular inammation. Int Ophthalmol Clin
1983; 23:175182.
Greenbaum J, Cockcroft D, Hargreave FE,
Dolovich J: Sodium cromoglycate in
ragweed-allergic conjunctivitis. J Allergy
Clin Immunol 1977; 59:437439.
Friday GA, Biglan AW, Hiles DA, et al:
Treatment of ragweed allergic conjunctivitis
with cromolyn sodium 4% ophthalmic
solution. Am J Ophthalmol 1983;
95:169174.
Welsh PW, Yunginger JW, Tani DG, et al:
Topical ocular administration of cromolyn
sodium for treatment in seasonal ragweed
64:209215.
Tani DG, Welsh PW, Bourne WM, et al:
Cromolyn sodium treatment of seasonal
ragweed conjunctivitis. Invest Ophthalmol
Vis Sci 1978; 17(Suppl):227.
Caldwell DR, Verin P, Hartwich-Young R,
et al: Efcacy and safety of lodoxamide
0.1% vs cromolyn sodium 4% in patients
with vernal keratoconjunctivitis. Am J
Ophthalmol 1992; 113:632637.
Santos CI, Huang AJ, Abelson MB, et al:
Efcacy of lodoxamide 0.1% ophthalmic
solution in resolving corneal epitheliopathy
associated with vernal keratoconjunctivitis.
Am J Ophthalmol 1994; 117:488497.
Yanni JM, Miller ST, Gamache DA, et al:
Comparative effects of topical ocular
anti-allergy drugs on human conjunctival
mast cells. Ann Allergy Asthma Immunol
1997; 79:541545.
Abelson MB, Allansmith MR,
Friedlaender MH: Effects of topically
applied ocular decongestant and
antihistamine. Am J Ophthalmol 1980;
90:254.
Abelson MB, George MA, Schaefer K,
Smith LM: Evaluation of the new
ophthalmic antihistamine, 0.05%
levocabastine, in the clinical allergen
challenge model of allergic conjunctivitis.
Sharif NA, Xu SX, Miller ST, et al:
Characterization of the ocular antiallergic
and antihistaminic effects of olopatadine
(AL-4943A), a novel drug for treating ocular
allergic diseases. J Pharm Exp Ther 1996;
278:12521261.
Lewis T: The blood vessels of the human
skin and their responses. London: Shaw;
1927.
Staub AM, Bovet D: Action de la
thymoxyethyl-diethylamine (929F) et des
ethers phenoliques sur le choc
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
anaphylactique du cobaye. CR Soc Biol

1937; 125:818.
Halpern BN: Les antihistaminiques de
syntse: essais de chimotherapie destats
alleriques. Arch Int Pharmacodyn Ther
1942; 68:339.
Bovet D, Horclois R, Walthort F: Propriets
antihistaminiques de la N-pmethoxybenzyl-N-dimethylaminoethyl
amino-pyridine. CR Soc Biol 1944; 138:99.
Bourquin TB: A new synthetic
antihistaminic substance (Antistine) and its
use in ophthalmology. Schweiz Med
Hurwitz P: Antistine in ocular allergy. Am J
Miller J, Wolf EH: Antazoline phosphate
and naphazoline hydrochloride, singly and
in combination for the treatment of allergic
conjunctivitis A controlled, double-blind
clinical trial. Ann Allergy 1975; 35:8186.
Leon J, Charap A, Duzman E, Shen C:
Efcacy of cimetidine/pyrilamine eyedrops,
a dose response study with histamine
challenge. Ophthalmology 1986;
93:120123.
Abelson MB, Schaefer K, Wun PJ:
Antihistamines and
antihistamine/vasoconstrictor
combinations. In: Abelson MB, ed. Allergic
diseases of the eye. Philadelphia, PA WB
Saunders Co; 2001:206214.
Dechant KL, Goa KL: Levocabastine: a
and therapeutic potential as a topical
antihistamine in allergic rhinitis and
conjunctivitis. Drugs 1991; 41:202224.
Secchi A, Leonardi A, Discepola M, et al:
An efcacy and tolerance comparison of
emedastine difumarate 0.5% and
levocabastine hydrochloride 0.05%:
reducing chemosis and eyelid swelling in
subjects with seasonal allergic
conjunctivitis. Acta Ophthalmol Scand
2000; 78(Suppl 230):4851.
Yanni JM, Sharif NA, Gamache DA, et al:
A current appreciation of sites for
pharmacological intervention in allergic
conjunctivitis: effects of new topical ocular
drugs. Acta Ophthalmol 1999;
77(228):3337.
Leonardi A, DeFranchis G, DePaoli M, et al:
Histamine-induced cytokine production and
ICAM-1 expression in human conjunctival
broblasts. Curr Eye Res 2002;
25:189196.
Mazurek N, Berger G, Pecht I: A binding
site on mast cells and basophils for the
anti-allergic drug cromolyn. Nature 1980;
286:722723.
Theoharides TC, Sieghart W, Greengard P,
Douglas WW: Antiallergic drug cromolyn
may inhibit histamine secretion by
regulating phosphorylation of a mast cell
protein. Science 1980; 207:8082.
Altounyan REC: Inhibition of experimental
asthma by a new compound, disodium
cromoglycate, INTAL. Acta Allergol 1967;
22:487.
Bernstein IL, Siegel SC, Brandon ML, et al:
A controlled study of cromolyn sodium
sponsored by the Drug Committee of the
American Academy of Allergy. J Allergy Clin
Immunol 1972; 50:235245.
Howarth PH, Durham SR, Lee TH, et al:
Inuence of albuterol, cromolyn sodium
and ipratropium bromide on the airway and
circulating mediator responses to allergen
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98.
99.
100.
101.
102.
103.
bronchial provocation in asthma. Am Rev

Respir Dis 1985; 132:98692.
Hughes D, Mindorff C, Levison H: The
immediate effect of sodium cromoglycate
on the airway. Ann Allergy 1982; 48:68.
Businco L, Cantani A, Benincori N, et al:
Effectiveness of oral sodium cromoglycate
(SCG) in preventing food allergy in children.
Ann Allergy 1983; 51:4750.
Dolovich J, Punthakee ND, MacMillan AB,
Osbaldeston GJ: Systemic mastocytosis:
control of lifelong diarrhea by ingested
disodium cromoglycate. Can Med Assoc J
1974; 111:684685.
Sorri M, Jokinen K, Palva A: Disodium
chromoglycate (sic) therapy in perennial
rhinitis. Acta Otolaryngol 1979;
360(Suppl):3032.
Pelikan Z: The effects of disodium
cromoglycate and beclomethasone
diproprionate on the late nasal mucosa
response to allergen challenge. Ann Allergy
1982; 49:200212.
Foster CS, Duncan J: Randomized clinical
trial of topically administered cromolyn
sodium for vernal keratoconjunctivitis.
Easty DL, Rice NSC, Jones BR: Clinical
trial of topical disodium cromoglycate in
vernal kerato-conjunctivitis. Clin Allergy
1972; 2:99104.
Greenbaum J, Cockcroft D, Hargreave FE,
Clin Immunol 1977; 59:437439.
Friday GA, Biglan AW, Hiles DA, et al:
Treatment of ragweed allergic conjunctivitis
with cromolyn sodium 4% ophthalmic
solution. Am J Ophthalmol 1983; 95:169.
Leino M, Tuovinen E: Clinical trial of the
topical use of disodium cromoglycate in
vernal, allergic and chronic conjunctivitis.
Welsh PW, Yunginger JW, Tani DG, et al:
Topical administration of cromolyn sodium
for treatment in seasonal ragweed
64:209215.
Kray KT, Squire EN, Tipton WR, et al:
Cromolyn sodium in seasonal allergic
76:623627.
Foster CS, Duncan J: Randomized clinical
trial of topically administered cromolyn
sodium for vernal keratoconjunctivitis.
Calonge M, Montero JA, Herreras JM,
et al: Efcacy of nedocromil sodium and
cromolyn sodium in an experimental model
of ocular allergy. Ann Allergy Asthma
Immunol 1996; 77:124130.
Blumenthal M, Casale T, Dockhorn R, et al:
Efcacy and safety of nedocromil sodium
ophthalmic solution in the treatment of
seasonal allergic conjunctivitis. Am J
Ophthalmol 1992; 113:5663.
Kjellman MN-I, Stevens MT: Clinical
experience with Tilavist: an overview of
efcacy and safety. Allergy 1995;
50(Suppl 21):1422.
Leonardi A, Borghesan F, Avarello A, et al:
Effect of lodoxamide and disodium
cromoglycate on tear eosinophil cationic
protein in vernal keratoconjunctivitis.
Santos CI, Huang AJ, Abelson MB, et al:
Efcacy of lodoxamide 0.1% ophthalmic
solution in resolving corneal epitheliopathy
CHAPTER 25
277
104.
105.
106.
107.
SECTION 4
108.
109.
110.
111.
112.
113.
114.
115.
116.
278
associated with vernal keratoconjunctivitis.

Am J Ophthamol 1994; 117:488497.
Caldwell DR, Verin P, Hartwich-Young R,
et al: Efcacy and safety of lodoxamide
0.1% vs. cromolyn sodium 4% in patients
with vernal keratoconjunctivitis. Am J
Ophthalmol 1992; 113:632637.
Gunduz K, Ucakhan O, Budak K, et al:
Efcacy of lodoxamide 0.1% versus
N-acetyl aspartyl glutamic acid 6%
ophthalmic solutions in patients with vernal
keratoconjunctivitis. Ophthalmic Res 1996;
28:8087.
Cerqueti PM, Ricca V, Tosca MA, et al:
Lodoxamide treatment of allergic
conjunctivitis. Int Arch Allergy Immunol
1994; 105:185189.
Abelson MB, Berdy GJ, Mundorf T, et al:
Pemirolast study group. Pemirolast
potassium 0.1% ophthalmic solution is an
effective treatment for allergic
conjunctivitis: a pooled analysis of two
prospective, randomized, double-masked,
placebo-controlled, phase III studies.
J Ocul Pharmacol Ther 2002;
18:475488.
Gous P, Ropo A: A comparative trial of the
safety and efcacy of 0.1 percent
pemirolast potassium ophthalmic solution
dosed twice or four times a day in patients
with seasonal allergic conjunctivitis. J Ocul
Denis D, Bloch-Michel E, Verin P, et al:
Treatment of common ocular allergic
disorders; a comparison of lodoxamide and
NAAGA. Br J Ophthalmol 1998;
82:11351138.
Bouhlal H, Blondin C, Haeffner-Cavaillon N,
Goldschmidt P: N acetyl-aspartyl glutamic
acid (NAAGA) inhibits the adhesion of
leukocytes to activated endothelial cells
and down-modulates the cytokine-induced
expression of adhesion molecules. J Fr
Ophtalmol 2002; 25:9931000.
Yanni JM, Stephens DJ, Miller ST, et al: The
in vitro and in vivo ocular pharmacology of
olopatadine (AL-4943A) and effective
anti-allergic/anti-histaminic agent. J Ocular
Pharmacol 1996; 12:389400.
Abelson MB: A review of olopatadine for
the treatment of ocular allergy. Expert Opin
Pharmacother 2004; 5:19791994.
Leonardi A, Abelson MB: Double-masked,
randomized, placebo-controlled clinical
study of the mast cell-stabilizing effects of
treatment with olopatadine in the
conjunctival allergen challenge model in
humans. Clin Ther 2003; 25:25392552.
Aguilar AJ: Comparative study of clinical
efcacy and tolerance in seasonal allergic
conjunctivitis management with 0.1%
olopatadine hydrochloride versus 0.05%
ketotifen fumarate. Acta Ophthalmol Scand
2000; 78(Suppl. 230):5255.
Berdy GJ, Spangler DL, Bensch G, et al:
A comparison of the relative efcacy and
clinical performance of olopatadine
hydrochloride 0.1% ophthalmic solution
and ketotifen fumarate 0.025% ophthalmic
solution in the conjunctival antigen
challenge model. Clin Therapeutics 2000;
22:826833.
Leonardi A, Zarakis P: Efcacy and
comfort of olopatadine versus ketotifen
ophthalmic solutions: a double-masked,
environmental study of patient
preference. Curr Med Res Opin 2004;
20:11671173.
117. Butrus S, Greiner JV, Discepola M,

Finegold I: Comparison of the clinical
efcacy and comfort of olopatadine
hydrochloride 0.1% ophthalmic solution
and nedocromil sodium 2% ophthalmic
solution in the human conjunctival allergen
challenge model. Clin Ther 2000;
22:14621472.
118. Deschenes J, Discepola M, Abelson MB:
Comparative evaluation of olopatadine
ophthalmic solution (0.1%) versus
ketorolac ophthalmic solution (0.5%) using
the provocative antigen challenge model.
Acta Ophthmol Scand 1999; 77(Suppl
228):4752.
119. Grant SM, Goa KL, Fitton A, Sorkin EM:
Ketotifen: a review of its pharmacodynamic
and pharmakinetic properties, and
therapeutic use in asthma and allergic
disorders. Drugs 1990; 40:412448.
120. Crampton HJ: A comparison of the relative
clinical efcacy of a single dose of ketotifen
fumarate 0.025% ophthalmic solution
versus placebo in inhibiting the signs and
symptoms of allergic rhinoconjunctivitis as
induced by the conjunctival allergen
challenge model. Clin Ther 2002;
24:18001808.
121. Greiner JV, Michaelson C, McWhirter CL,
Shams NB: Single dose of ketotifen
fumarate .025% vs 2 weeks of cromolyn
sodium 4% for allergic conjunctivitis. Adv
Ther 2002; 19:185193.
122. Abelson MB, Ferzola NJ, McWhirter CL,
Crampton HJ: Efcacy and safety of
single- and multiple-dose ketotifen
fumarate 0.025% ophthalmic solution in a
pediatric population. Pediatr Allergy
Immunol 2004;15:551557.
123. Ciprandi G, Buscaglia S, Catrullo A, et al:
Azelastine eye drops reduce and prevent
allergic conjunctival reaction and exert
anti-allergic activity. Clin Exp Allergy 1997;
27:182191.
124. Friedlander MH, Harris J, LaVallee N, et al:
Evaluation of the onset and duration of
effect of Azelastine Eye Drops (0.05%)
versus placebo in patients with allergic
conjunctivitis using an allergen challenge
model. Ophthalmology 2000;
107:21522157.
125. Lenhard G, Mivsek-Music E, PerrinFayolle M, et al: Double-blind, randomised,
placebo-controlled study of two
concentrations of azelastine eye drops in
seasonal allergic conjunctivitis or
rhinoconjunctivitis. Curr Med Res Opin
1997; 14:2128.
126. Sabbah A, Marzetto M: Azelastine eye
drops in the treatment of seasonal allergic
conjunctivitis or rhinoconjunctivitis in young
children. Curr Med Res Opin 1998;
14:161170.
127. Giede C, Metzenauer P, Petzold U,
Ellers-Lenz B: Comparison of azelastine
eye drops with levocabastine eye drops in
the treatment of seasonal allergic
conjunctivitis. Curr Med Res Opin 2000;
16:153163.
128. Fraunfelder FW: Epinastine hydrochloride
for atopic disease. Drugs Today (Barc)
2004; 40:677683.
129. Tasaka K: Epinastine: an update of its
pharmacology, metabolism, clinical efcacy
and tolerability in the treatment of allergic
diseases. Drugs Today 2000; 36:735757.
130. Matsukura M, Yajima A, Yamazaki F, et al:
Epinastine inhibits eosinophil chemotaxis
131.
132.
133.
134.
135.
136.
137.
138.
139.
140.
141.
142.
143.
144.
and adhesion molecules in atopic

dermatitis. Skin Pharmacol Appl Skin
Physiol 2003; 16:405410.
Abelson MB, Gomes P, Crampton HJ, et al:
Efcacy and tolerability of ophthalmic
epinastine assessed using the conjunctival
antigen challenge model in patients with a
history of allergic conjunctivitis. Clin Ther
2004; 26:3547.
Whitcup SM, Bradford R, Lue J, et al:
Efcacy and tolerability of ophthalmic
epinastine: a randomized, double-masked,
parallel-group, active- and
vehicle-controlled environmental trial in
patients with seasonal allergic
conjunctivitis. Clin Ther 2004; 26:2934.
Ganellin CR: Chemistry and
structureactivity relationships of
H2-receptor antagonists. In: Rocha e
Silva M, ed. Histamine II and antihistaminics: chemistry, metabolism, and
physiological and pharmacological actions.
Handbook of experimental pharmacology.
New York: Springer; 1978:251294.
Ariens EJ, Simonis AM: Autonomic drugs
and their receptors. Arch Int Pharmacodyn
1960; 127:479.
Rocha e Silva M: Kinetics of antagonist
action. In: Rocha e Silva M, ed. Histamine II
and anti-histaminics: chemistry,
metabolism, and physiological and
pharmacological actions. Handbook of
experimental pharmacology. New York:
Springer; 1978:295332.
Douglas W: Histamine and serotonin and
their antagonists. In: Goodman L, Gilman A,
Rall T, Murad F, eds. The pharmacologic
basis of therapeutics. 7th edn. New York:
Macmillan; 1985:605638.
Witiak DT, Lewis NJ: Absorption,
distribution, metabolism and elimination of
antihistamines. In: Rocha e Silva M, ed.
Histamine II and anti-histaminics:
chemistry, metabolism, and physiological
and pharmacological actions. Handbook of
experimental pharmacology. New York:
Springer; 1978:513560.
Nathan RA, Segall N, Schocket AL:
A comparison of the actions of H1 and
H2 antihistamines on histamine-induced
bronchoconstriction and cutaneous wheal
response in asthmatic patients. J Allergy
Clin Immunol 1981; 67:171177.
Landau SW, Nelson NA, Gay LN:
Antihistamine properties of local
anesthetics and anesthetic properties of
antihistaminic compounds. J Allergy 1951;
22:19.
Carruthers SG, Shoeman DW, Hignite CE,
Azarnoff DL: Correlation between plasma
diphenhydramine level and sedative and
antihistamine effects. Clin Pharmacol Ther
1978; 23:375382.
Crandall DC, Leopold IH: The inuence of
systemic drugs on tear constituents.
Lofer BH, Lemp MA: The effect of an
antihistamine (chlorpheniramine maleate)
on tear production in humans. Ann
Bielory L, Lien KW, Bigelsen S: Efcacy
and tolerability of newer antihistamines in
the treatment of allergic conjunctivitis.
Drugs 2005; 65:215228.
Abelson MB, Welch DL: An evaluation of
onset and duration of action of Patanol
(olopatadine hydrochloride ophthalmic
solution 0.1%) compared to Claritin
145.
146.
147.
148.
149.
150.
151.
152.
153.
154.
155.
156.
(loratadine 10 mg) tablets in acute allergic

conjunctivitis in the conjunctival allergen
challenge model. Acta Ophthalmol Scand
2000; 78(230):6063.
Wyngaarten JB, Seevers MH: The toxic
effect of antihistaminic drugs. JAMA 1951;
145:277.
Sadusk JF, Palmisano PA: Teratogenic
effect of meclizine, cyclizine and
chlorcyclizine. JAMA 1965; 194:987.
Pecoud A, Zuber P, Kolly M: Effect of a
new selective H1 receptor antagonist
(levocabastine) in a nasal and conjunctival
provocation test. Int Arch Allergy Appl
Immunol 1987; 82:541.
Abelson MB, Smith LM: Levocabastine
evaluation in the histamine and compound
48/80 models of ocular allergy in humans.
Abelson MB, George MA, Smith LM:
Evaluation of 0.05% levocabastine versus
4% sodium cromolyn in the allergen
challenge model. Ophthalmology 1995;
102:310.
Spitalny L, Abelson M: Olopatadine
ophthalmic solution decreases itching and
redness associated with allergic
conjunctivitis. Invest Ophthalmol Vis Sci
1996; 37(Suppl):593.
Black JW, Duncan WAM, Emmett JC, et al:
Metiamide: an orally active histamine
H2-receptor antagonist. Agents Actions
1973; 3:133.
Colin-Jones DG, Langman MJS, Lawson
DH, Vessey MP: Post-marketing
surveillance of the safety of cimetidine:
mortality during second, third, and fourth
years of follow up. Br Med J 1985;
291:1085.
Zeldis JB, Friedman LS, Isselbacher KJ:
Ranitidine: a new H2-receptor antagonist.
N Engl J Med 1983; 309:1368.
Schentag JJ, Cerra FB, Calleri GM, et al:
Age, disease and cimetidine disposition in
healthy subjects and chronically ill patients.
Clin Pharmacol Ther 1981; 29:737.
Bodemar G, Norlander B, Walan A:
Cimetidine in the treatment of active peptic
ulcer disease. In: Burland WL, Simkins MA,
eds. Cimetidine: second international
symposium on histamine H2-receptor
antagonists. Amsterdam: Excerpta
Medica; 1977.
Brimblecombe RW, Duncan WAM: The
relevance to man of pre-clinical data for
cimetidine. In: Burland WL, Simkins MA,
eds. Cimetidine: second international
symposium on histamine H2-receptor
antagonists. Amsterdam: Excerpta
Medica; 1977.
157. Burland WL, Duncan WAM, Hessello T,

et al: Pharmacological evaluation of
cimetidine, a new histamine H2-receptor
antagonist, in healthy man. Br J Clin
Pharmacol 1975; 2:481.
158. Lee PK, Lai LL, Lok ASF, et al:
Haemodynamic responses to intravenous
cimetidine in subjects with normal lung
function and in subjects with chronic
airway obstruction. Br J Clin Pharmacol
1981; 11:339.
159. Carlson HE, Ippoliti AF, Swerdloff RS:
Endocrine effects of acute and chronic
cimetidine administration. Dig Dis Sci 1981;
26:428.
160. MuGuigan JE: A consideration of the
adverse effects of cimetidine.
Gastroenterology 1981; 80:181.
161. Serlin MJ, Sibeon RG, Mossman S, et al:
Cimetidine interaction with oral
anticoagulants in man. Lancet 1979; 2:317.
162. Jackson JE, Powell JR, Wandell M, et al:
Cimetidine decreases theophylline clearance.
Am Rev Respir Dis 1981; 123:615.
163. Klotz U, Reiman I: Delayed clearance of
diazepam due to cimetidine. N Engl J Med
1980; 302:1012.
164. Feely J, Wilkinson GR, Wood AJJ:
Reduction of liver blood ow and
propranolol metabolism by cimetidine.
N Engl J Med 1981; 304:692.
165. Langman MJS: Gastrointestinal drugs.
Side Eff Drugs Annu 1986; 10:323.
166. Cox JSG, Beach JE, Blair AMJN, et al:
Disodium cromoglycate (Intal). Adv Drug
Res 1970; 5:115.
167. Cox JSG: Disodium cromoglycate (FPL 67)
(Intal): a specic inhibitor of reaginic
antibody-antigen mechanisms. Nature
1967; 216:1328.
168. Mazurek N, Geller Bernstein C, Pecht I:
Afnity of calcium ions to the antiallergic
drug cromoglycate. FEBS Lett 1980;
111:194.
169. Mazurek N, Berger G, Pecht I: A binding
site on mast cells and basophils for the
antiallergic drug disodium cromoglycate.
Nature 1980; 286:722.
170. Mazurek N, Bashkin P, Loyter A, Pecht I:
Restoration of Ca2+ inux and
degranulation capacity of variant RBL-2H3
cells upon implantation of isolated
cromolyn binding protein. Proc Natl Acad
Sci USA 1983; 80:6014.
171. Mazurek N, Dulic V, Pecht I, et al: The role
of Fc receptors in calcium channel opening
in rat basophilic leukemic cells. Immunol
Lett 1986; 12:31.
172. Theoharides TC, Sieghart W, Greengard P,
Douglas WW: Antiallergic drug cromolyn
173.
174.
175.
176.
177.
178.
179.
180.
181.
182.
183.
184.
may inhibit histamine secretion by

regulating phosphorylation of a mast cell
protein. Science 1980; 207:80.
Wells E, Mann J: Phosphorylation of a
mast cell protein in response to treatment
with antiallergic compounds: implication
for the mode of action of sodium
cromoglycate. Biochem Pharmacol 1983;
32:837.
Bergstrand H, Lundquist B, Schurmann A:
Rat mast cell high afnity cyclic nucleotide
phosphodiesterases: Separation and
inhibitory effects of two antiallergic agents.
Mol Pharmacol 1978; 14:848.
Moss GF, Jones KM, Ritchie JT, et al:
Plasma levels and urinary excretion of
disodium cromoglycate after inhalation by
human volunteers. Toxicol Appl Pharmacol
1971; 20:147.
Lobel H, Machtey J, Eldror MY: Pulmonary
inltrates with eosinophilia in an asthmatic
patient treated with disodium
cromoglycate. Lancet 1972; 2:1032.
Burgher LW, Kass I, Schenken JR:
Pulmonary allergic granulomatosis: a
possible drug reaction in a patient
receiving cromolyn sodium. Chest 1974;
66:84.
Sheffer AL, Rocklin RE, Goetzl EJ:
Immunologic components of
hypersensitivity reactions to cromolyn
sodium. N Engl J Med 1975; 293:1220.
Settipane GA, Klein DE, Boyd GK, et al:
Adverse reactions to cromolyn. JAMA
1979; 241:811.
Johnson HG, VanHout CA, Wright JB:
Inhibition of allergic reactions by
cromoglycate and by a new anti-allergy
drug U-42,585E. I: activity in rats. Int Arch
Allergy Appl Immunol 1978; 56:416.
Verstappen AA, Smith J, Rosenthal A:
A double-masked efcacy and safety
evaluation of lodoxamide 0.1% ophthalmic
solution versus opticrom 2%: a multicenter
study in patients with allergic eye
disorders. Alcon Report No.
008:34350:1287, 1988.
Geha RS, Helm B, Gould H: Inhibition of
the PrausnitzKstner reaction by an
immunoglobulin E-chain fragment
synthesized in E. coli. Nature 1985;
315:577.
Hamburger RN: Peptide inhibition of the
PrausnitzKstner reaction. Science 1975;
189:389.
Kalpaxis JG, Thayer TO: Double-blind trial
of pentigetide ophthalmic solution, 0.5%,
compared with cromolyn sodium, 4%,
ophthalmic solution for allergic
conjunctivitis. Ann Allergy 1991; 66:393.
CHAPTER 25
279
CHAPTER
26
Tear Film and Blink Dynamics

Mark B. Abelson, Marshall G. Doane, and George Ousler
DYNAMICS OF THE BLINKING PROCESS

Blinking is usually performed as a nonconscious closing of the
eyelids and serves to carry secreted tear uid from the superior
and inferior marginal menisci over the anterior surface of the
eye, continuously reestablishing the tear lm over the cornea.
Also, the blinking action wipes debris and particulate matter
from the surface of the cornea and sclera into the inferior
marginal tear meniscus. As we shall see, once in the meniscus,
such debris is effectively directed toward the medial canthal
region by subsequent blinks and usually drained with used tear
uid via the punctal openings. This constant drainage via the
puncta is necessary to allow for the removal of used tear uid,
but it also removes instilled tear substitutes from the menisci,
therefore limiting their effective residence time and requiring
frequent reinstillation of such products.
Although the normal blink rate is often given as 1215 per
minute, this is, at best, an average of a greatly variable parameter, and it is strongly inuenced by external events. A loud
noise or bright ash of light, of course, immediately elicits a
blink by reex action, but more subtle events such as a visually
intensive task (e.g., reading, watching a computer monitor)
reduces the blink rate, thus increasing the length of interblink
periods and minimizing how often vision is blocked by the
closing of the eyelids. A persons tear lm stability also can
inuence the blink rate; because discomfort is usually associated with the breakup and drying of the tear lm on the
cornea, this can stimulate blinking. Thus, many dry-eyed
patients tend to have shortened blink intervals (i.e., high blink
rates) as a result of the decreased tear lm breakup time during
interblink periods. A major goal of any tear substitute is to
increase the stability of the tear lm layer, usually by incorporating surfactants and viscosity agents, as described in the
following section.
The details of the actual motion of the blinking eyelids occur
too rapidly to see. With a high-speed motion picture or video
camera, the recorded images can be replayed at a slower speed
and the details of the motion accurately determined. Truly
nonconscious blinks can only be recorded if the subject is not
aware that such blinks are being measured or indeed that blinking is a subject of interest. Self-conscious blinks are invariably
forced, and such blinks differ considerably in their dynamics
and time course from the ordinary, nonconscious blinking that
occurs thousands of times each day.
Studies using a high-speed camera and long telephoto lens
placed behind a one-way mirror have recorded the normal, nonconscious blinks of an unknowing subject.1 With a lm-recording
rate of up to 500 pictures per second, the resultant images were
subsequently analyzed frame by frame for the details of motion
of each lid, including their instantaneous velocities.
MOTION OF THE UPPER LID

The upper lid is responsible for wiping the anterior surface of
the globe and restoring a clean, new tear lm with each blink.
From its open, resting position, the upper lid rapidly accelerates
downward until reaching the center of the cornea. It then
decelerates, often slowing to a stop and reversing its motion
before actually contacting the lower lid. Even when such contact
does occur, it is seldom forceful except during strong, voluntary
blinks. Figure 26.1 shows typical time and velocity proles of
the upper lid for four consecutive nonconscious blinks in four
subjects. The point of zero velocity is the instant of reversal of
motion between the closing and opening phase of the blink.
Note that the opening phase consumes about twice the time of
the closing phase and is particularly slow during the last few
millimeters of lid opening. The reversal of lid motion is rapid,
occurring in less than 2 ms; in voluntary blinking, the lid
remains stationary for much longer, a consistent feature of such
blinks. The maximum downward excursion of the upper lid is,
of course, limited to the width of the palpebral ssure at the
open, resting position of the lids. Many, if not most, blinks are
less than complete, with the amount of lid excursion less than
the maximum possible for a given individual. For the examples
shown, the maximum excursion of the upper lid ranges from
5 to 13 mm and peak velocities from 80 to 300 mm/s.
As indicated by the lid velocity proles shown in Figure 29.1,
blink velocities vary considerably among individuals and even
between consecutive blinks in the same person. Nevertheless,
by averaging the information for many nonconscious blinks,
the data for a standard, nonforced blink have been obtained
(Table 26.1).
This blinking action of the upper lid efciently spreads tear
uid from the marginal menisci over the entire anterior surface
of the globe. This easily can be demonstrated by instilling a small
drop of uorescein solution from a micropipette into the tear
meniscus along the inferior lid margin. A single blink uniformly
distributes the uorescein over the cornea. Thus, the instillation
of a small quantity of a miscible tear substitute into the inferior
tear meniscus can be reasonably expected to mix with the natural
tear uid in the meniscus and be spread over the entire anterior
surface of the globe by the next few blinks of the lids.
MOTION OF THE LOWER LID

The lower lid undergoes little vertical movement, its major
motion being a horizontal translation directed toward the medial
canthus during the closing of the upper lid. This motion
reverses its direction in synchrony with the beginning of the
opening phase of the upper lid. Total translation of the lower lid
is proportional to the extent of movement of the upper lid,
usually in the range of 24 mm.
281

FIGURE 26.1. Plots of blink motion dynamics
in four subjects. The upper curve of each pair
represents the time course of the upper lid
displacement during its closing and opening
phases. The lower curve in each case is the
time course of the instantaneous velocity of the
upper lid, which is zero at the point where the
lid reverses direction. Note the variation
between individuals. The subject for plot D had
a narrow palpebral ssure; consequently the lid
excursions and velocities were less than
normal.
SECTION 4
From Doane MG: Interaction of eyelids and tears in

corneal wetting and the dynamics of the normal
human eyeblink. Am J Ophthalmol 1980; 89:507516.
TABLE 26.1. Dynamics of Upper Eyelid Motion During a Blink*

Factor
Value
Duration of closing phase
82.1 2.1 ms
Duration of opening phase
175.8 11.0 ms
Total blink duration
257.9 11.3 ms
Maximum closing velocity
18.7 1.7 cm/s
Maximum opening velocity
9.7 0.7 cm/s
*Each value given is an average of 40 blinks, standard error of the mean, in 10

different subjects.
From Doane MG: Interaction of eyelids and tears in corneal wetting and the
dynamics of the normal human eyeblink. Am J Ophthalmol 1980; 89:507516.
Published with permission from the American Journal of Ophthalmology.
Copyright by the Ophthalmic Publishing Company.
TEAR MIXING, TURNOVER, AND DRAINAGE

TEAR VOLUME AND MIXING
282
The tear lm is a quasilayered structure with numerous

components secreted from many spatially separated sources. A
meshwork of small molecules and membrane-bound mucins,

known as the glycocalyx, is generated by healthy ocular
epithelial cells and serves to anchor the tear lm to the eye.
Above this lies a large phase accounting for well over 90% of the
tear volume in which mucins are suspended in an aqueous
solution. Current models suggest that although mucins arae
scattered throughout the tear film, they are more concentrated
closer to the ocular surface. The outermost layer of the tear lm
is composed of lipids. While the mucins serve to attach the tear
volume to the ocular surface, and also have lubricating
properties, the lipid layer both protects against external insult
and helps delay tear evaporation.
The primary and accessory lacrimal glands secrete the major
portion of the tear uid volume the accessory glands of Krause
and Wolfring aid in natural tear production, while it is thought
that the large, primary lacrimal gland is in fact responsible for
reex tearing. The contributions of the goblet cells in the
conjunctiva (mucin glycoproteins) and the meibomian glands in
the lids (lipids) are no less important than the lacrimal glands
in maintaining a functional tear layer. Thus, in addition to the
tear lm resurfacing action of the blinking lids, the blink also
serves the important function of combining and mixing these
tear uid components into a quasistable mixture. Included in
this mixture would be any added tear substitute, suggesting the
importance of the compatibility of the additive and the natural

tear uid.
The total tear volume on the anterior segment of the eye is
usually between 6 and 7 mL,2 and there is a certain amount of
preferential compartmentalization for this uid. When available
uid is decient, the fornices ll rst and can contain about
5.9 mL of uid.2a Blinking carries uid upward over the corneal
surface, establishing the precorneal tear lm, requiring another
1 mL of uid. The fornices cannot hold more uid than their
enclosed, dened space allows, nor can the tear lm substantially increase its uid volume. Once these compartments are
lled, any excess uid goes to ll the marginal tear menisci,
which can hold 23 mL of tear uid. Freshly spread tear film is
also drawn out of the meniscus, but in the opposite direction.
Fluid much greater than this amount raises the level of the
inferior meniscus above that of the punctal opening and, as
described in following paragraphs, it is soon drained away via
the canaliculi into the lacrimal sac. Thus, it is of little benet
to overll the menisci with an instilled uid, because any excess
uid that does not actually overow onto the lids is quickly
drained away by the blink-driven drainage system.
Within their sustainable volume range, the marginal menisci
act as a variable reservoir. Often, the relative amount of tear
volume in an eye can be somewhat quantitatively assessed by
noting the height of the marginal tear menisci.3 The height of
the inferior tear meniscus often is reduced in patients with
keratoconjunctivitis sicca, although individual variations in lid
apposition, tightness, and thickness can also affect meniscus
height.
For diagnostic purposes, Schirmer strips may be placed in the
eyes for a short period of time to estimate tear volume. Nonanesthetized Schirmer tests can stimulate reex, rather than
natural, tear production, which makes for inaccurate tear
volume values. For more exact readings, a uorophotometer
may be used. The uorophotometer rst measures autouorescence in the tear lm. This reading is then subtracted from a
uorescence value measured following the instillation of a
microdrop of uorescein in a patients eye. The result generated
is an approximale tear volume value, and the procedure can be
repeated to yield a series of results.4
Figure 26.2 is a schematic representation of the abovementioned compartmentalization of the tear volume.
tend to extend and elevate themselves from the leading edge of

the lids during the blink. Being located near the medial juncture
of the lid structure, this region of the superior and inferior lid
margins meet, often forcefully, by the time overall lid closure is
only one-third to one-half complete. From this point to the
completion of the lid motion associated with the blink, the
punctal openings are largely occluded.
The primary effect of the second half of lid closure is to
squeeze the elastic walls of the canaliculi, forcing any tear uid
within them onward into the lacrimal sac. Fluorescein experiments indicate minimal regurgitation of uid out of the punctal
openings, with the rm apposition of the lid margins minimizing this retrograde ow. Detailed high-speed, close-up
photography shows that the region of the lid margins containing the punctal openings remains in tight contact until the
lids are near the end of their opening phase. Then, the region of
the lid margins containing the punctal openings suddenly pops
apart when the force of the separating lids nally overcomes the
suction force holding them together.5 This suction is generated
by the elastic walls of the canaliculi (and to some extent the
lacrimal sac) trying to expand to contain their normal volume
once the pressure of the closing lids is released.
Once the puncta are separated, a rapid, pulsatile ow of tear
uid is drawn into the puncta from the marginal menisci owing
to the suction force generated within the canaliculi. When tear
volume is normal, this ow typically lasts 1 or 2 s, as long as the
height of the uid in the meniscus reservoir is sufcient to
maintain contact with the punctal openings. Any instillation of
a tear substitute that temporarily increases the volume of uid
to a level above the punctal opening prolongs this exit ow, and
the excess volume is quickly removed from the meniscus. Once
the meniscus height falls below the slightly elevated position
of the punctal openings, further drainage stops.6 Because of
differences in blink strength, degree of completion, and uid
volume in the marginal menisci, not all blinks result in equally
CHAPTER 26
MECHANISM OF TEAR FLUID DRAINAGE

The single punctal opening in each of the lid margins is located
at the apex of the lacrimal papillae, in the medial canthal region
of the lids. Each punctal opening leads to a single tubular
conduit, or canaliculus, which makes a right-angle bend
~2 mm from the edge of the lid and then parallels the lid margin
for most of its length. The superior and inferior canaliculi usually
join into a common pathway just before entering the lacrimal tear
sac just posterior and superior to the center of its lateral wall.
There is evidence for a one-way restriction, or valve, in this
common canaliculus, allowing uid to ow from the canaliculus
into the lacrimal sac but restraining ow in the reverse direction.
A duct, the nasolacrimal canal, descends from the inferior portion
of the sac, opening into the nasal meatus.
The passage of tear uid through the punctal openings, into
the canaliculi, and onward into the lacrimal sac, is driven by the
squeezing actions and muscular contractions associated with
the blink action of the lids. This process involves a denitive,
rapid sequence of events.5 As the blinking action of the lids
commences, the upper lid begins its downward sweep over the
anterior portion of the globe, with the lower lid starting its
movement medially, carrying with it the uid in the marginal
meniscus. The lacrimal papillae containing the punctal openings
FIGURE 26.2. Schematic representation of tear-uid

compartmentalization and outow. Nearly all the effective tear volume
is secreted by the main and accessory lacrimal glands, with an added
contribution from the conjunctival goblet cells (mucin) and the
meibomian glands in the lids (lipid). The tear uid is rst used to ll the
volume between the globe and lids (superior and inferior fornices) and
the tear lm over the exposed globe. Any excess uid then goes into
the reservoir of the marginal menisci, from which drainage via the
punctal openings occurs. Smaller amounts of uid are lost by
evaporation and absorption by the conjunctiva.
283

action. Recall that drainage from the inferior meniscus occurs
from its highest, uppermost portion, which is drawn into the
elevated punctal openings. If a viscous tear substitute is
carefully instilled into the inferior cul-de-sac, it often acts as a
longer-lasting depot of uid that is not readily drained and is
slowly mixed into the tear lm by subsequent blinks. In fact, in
monitoring the concentration of applied agents by interferometry, a few strong blinks often elicit a sudden increase in the
amount of viscous agent in the tear lm by forcing the depot
out of the inferior cul-de-sac many minutes after initial instillation of the articial tear solution. However, applied agents that
are too viscous are detrimental to retention time, because they
blur vision, elicit foreign body sensations that stimulate
blinking, and are not well liked by users because of stickiness
and the tendency to collect in the eyelashes.
WETTING AND DRYING OF THE CORNEAL

SURFACE
NATURE OF THE WETTING PROCESS
We can dene wetting as the spreading of a uid over a solid
surface, a complex process from a molecular, surface-chemical
viewpoint. The degree of spreading depends on the relative
forces of cohesion between the like molecules of the uid and
SECTION 4
efcient ow patterns. Because the puncta occlusion by the

opposing lid margin occurs even with half blinks, with
associated squeezing of the canaliculi, some tear uid is often
drawn into them even after incomplete blinks, although the
amount of uid drainage is reduced. Figure 26.3 is a schematic
representation of this cycle.
From the time of initial instillation, any applied uid is
decreasing in its overall tear uid concentration as time goes by.
The time of contact between the ocular surface and the applied
uid is directly limited by the rate of drainage from the marginal
menisci; in addition to drainage, the concentration of therapeutic agents that does remain is continuously diminished by
newly secreted tear uid. Thus, any means of increasing the
retention time of instilled solutions at therapeutic levels is of
crucial interest. Clinically, the rate of drainage, or tear ow,
can be estimated (similarly to tear volume) using a uorophotometer. By recording the amount of uorescent dye present
in the tear lm at various time points after instillation, the
uorophotometer can determine the rate at which the dye is
being washed out due to tear turnover.4
The stimulation of a faster blink rate (such as by the administration of a solution that stings or is otherwise uncomfortable)
is undesirable, because this causes a more rapid drainage of tear
uid from the marginal menisci. An increase in the viscosity of
a tear substitute may result in an effectively prolonged time of
FIGURE 26.3. Mechanism of lacrimal drainage. Clockwise from the top: (1) At the start of the blink, the lacrimal drainage passages already
contain tear uid that has entered following the previous blink. (2) As the upper lid descends, the papillae containing the punctal openings
elevate from the medial lid margin. By the time the upper lid has descended halfway, the papillae meet the opposing lid margin, occluding the
puncta and resisting uid regurgitation. (3) The remaining portion of the lid closure acts to squeeze the canaliculi and sac through the action of
the orbicularis oculi, forcing out contained uid that has not been absorbed by the mucosa of the sac and nasolacrimal duct. (4) At complete lid
closure, the system is compressed and devoid of uid. (5) During the start of the opening phase of the blink, the puncta are still occluded and
valving action at the distal end of the canaliculi (and perhaps in the nasolacrimal duct) acts to prevent reentry of uid or air. As lids open,
compressive action ends and the elastic walls of the canaliculi attempt to expand to their normal shape. This elastic force causes a partial
vacuum, or suction, to form within the canaliculi and sac. (6) The suction force holding the punctal region of the lid margins together is released
when lid separation is sufcient, at about two-thirds of the fully open position. The punctal openings are now accessible for uid entry from the
marginal tear menisci, and tear uid is drawn into the canaliculi during the rst few seconds following the blink.
From Doane MG: Blinking and the mechanics of the lacrimal drainage system. Ophthalmology 1981; 88:844.
284
the forces of adhesion between the unlike molecules of the solid

surface and those of the uid. Thus, when a uid rests on a
solid surface, the relative strength of these two forces determines the degree of uid spreading. The stronger the relative
cohesive forces attracting the uid molecules together, the less
the uid increases its surface area to spread out on the solid
surface. Thus, in order to spread and wet a surface, the
uidsolid adhesion forces must be greater than (or at least
comparable to) the uiduid cohesive forces.
However, wettability is more complex than this simple
explanation; it also depends on the degree of polarity and type
of charge of the molecular groups exposed on the surface of the
solid. For instance, exposed polar groups tend to have an attraction for the polar molecules of water. Materials with nonpolar
surfaces (Teon, oils) have a low attraction for polar groups
such as those in water and thus are inherently hydrophobic, or
water-repelling. Surface-active agents, or surfactants, can greatly
increase the wettability of a surface by acting as a bridge
between polar and nonpolar molecules. Typically, such agents
have molecules with some exposed moieties that are hydrophobic (such as alkyl groups) and others on the same molecule
that are hydrophilic (such as carboxyl groups). Mucin glycoproteins are thought to act as a wetting agent in tear uid.
Virtually all articial tear preparations contain one or more
chemical surfactants that enhance their wetting of the
cornea.
Corneal epithelial cells secrete glycocalyx, which has a
similar chemical composition and characteristics of other
mucins; therefore, the surface is intrinsically wettable. In
addition, the surface of the cornea is covered by an adsorbed
layer of mucin, perhaps 1 mm thick, allowing the tear uid to
spread easily over this surface.
BREAKUP OF THE PRECORNEAL TEAR

FILM
The surface of the cornea is rewet with a fresh layer of uid,
forming the precorneal tear lm, by each blink of the eyelids.
This periodic action is necessary owing to the deterioration of
this thin uid layer between blinks. Immediately after a new
tear lm is formed, it undergoes a progressive overall thinning
owing to evaporation and, more importantly, begins to develop
localized areas that thin even more rapidly than the tear layer
as a whole. It is these localized regions, usually small in area,
that result in the rst micelles, or dry spots, observed after a
blink. These spots appear as dark, nonuorescent areas when
uorescein is in the tear uid, because where there is no uid,
there is no uorescein and hence no uorescence. The time
from the completion of a blink to the rst appearance of these
dark spots is the tear lm breakup time (TFBUT). TFBUTs vary
from blink to blink and person to person an individual average
TFBUT under 10 s has traditionally been considered indicative
of dry eye syndrome. However, the innovative diagnostic combination of a microdrop of sodium uorescein (5 mL), a yellow
Wratten lter, and a digital video capture system with an
onscreen timer, has yielded more precise TFBUT values. These
measurements have indicated that 5 s is a more appropriate cutoff point for association of TFBUT with dry eye symptoms.7 Of
course, persistently short blink intervals can somewhat
moderate the symptomatic effects of shortened TFBUTs.
A new diagnostic calculation, called the ocular protection
index (OPI), quanties the relationship between blink frequency,
TFBUT, and ocular surface protection. The concept is simple:
an interblink interval (IBI) greater than the average TFBUT
value in a given eye indicates that there is a period of time
(namely, the difference between IBI and TFBUT) during which
the ocular epithelium is wholly or in part unprotected by the

tear lm. An IBI less than the average TFBUT suggests the eye
is blinking frequently enough to replenish the tear lm before it
breaks to expose the ocular surface. To quantify this relation in
a single value, one can simply divide IBI by TFBUT by IBI any
value greater than or equal to 1 indicates a generally protected
ocular surface, while any value less than 1 suggests at the very
least occasional exposure of the ocular epithelium.8
Examination of TFBUT video images has not only yielded
visual proof of the dynamic nature of tear lm breakup, but has
shown that the tear lm may break up in any of a number of
unique patterns. The tear lm breakup patterns (TFBUP)
identied so far include: spotting, amorphous blob, linear,
fractured, and wispy. Each pattern is generally reproducible by
patient, by eye. It is evident that certain patterns are
considerably more prevalent in certain dry eye populations,
leading to the belief that the pattern present upon breakup is at
least partially indicative of the advancement or nature of the
syndrome. TFBUPs appear to be modiable by alteration of tear
lm composition for instance, by treatment with a tear
substitute, meibomian gland expression, or stimulation of
reex tearing. Ultimately, tear lm breakup patterns show that
there is more to the dispersion of the tear lm than how quickly
it occurs, and these patterns could prove to be a useful visual
diagnostic in the future.9 The causes of tear lm breakup are
not, however, as easily explicable as the methods of quantifying
and recording it.
Although evaporative effects progressively thin the tear lm
and promote eventual drying and breakup, theoretical calculations indicate that the times required to thin the tear layer to
dryness should be much longer than those actually observed
for measured TFBUTs.10 Also, long before overall drying of the
anterior surface of the eye occurs, the small, localized areas of
drying discussed earlier are seen. Clearly, evaporation is not the
sole (or even primary) cause of tear lm breakup.
The supposed non-wettability of the corneal surface is
thought to be an artefact;10 rinsing the surface with
acetylcysteine returns the surface to its wettable state.11,12 The
presence of glycocalyx makes the surface wettable; however, it
has been shown that newly exposed epithelial cell surfaces,
revealed when overlying cells desquamate, have not yet
developed or fully-expressed their glycocalyx.13 The surface will
hence be relatively non-wettable. Theoretically, even a single
non-wettable cell can initiate a dry spot;14 it may be that the
pattern of dry spots may indicate where surface cells have
recently desquamated. Ultimately, the mechanisms of tear film
breakup are not clearly understood.
It is surmised that over time (i.e., within seconds after a
blink), the mucin layer on the epithelium becomes contaminated by nonpolar components of the tear lm, primarily lipid
from the supercial layer. This oily layer is, of course, only ~5
or 6 mm above the surface of the mucin layer under the best of
conditions. Microscopic ow patterns, either of thermal origin
or due to the turbulence of the blink action, can bring this oating lipid into contact with the mucin on the corneal surface.
Although a small amount of lipid contamination can be masked
by the mucin molecules, sufciently large areas eventually
become contaminated whereby the mucin can no longer act as
an effective surfactant. Then, nonwetting areas develop, with
spontaneous thinning of the tear layer immediately above them,
with eventual rupture of the tear lm. When these localized
nonwetting areas resist the formation of a new, clean mucin
layer during blinking, persistent tear lm breakup over the same
area follows within a few seconds of each blink. A series of
strong, forced blinks often reestablish tear lm continuity over
some of these areas by covering them with mucin.
CHAPTER 26
285

Disabling the drainage mechanism (through punctal plugs or
surgery) or continuously instilling tear substitutes can temporarily boost tear volume. Ideally, the application of a tear substitute should aid in wetting the corneal surface and prolong
the TFBUT of the tear lm for the entire time between instillations. Although the initial instillation of a drop of solution
increases tear volume and usually tear lm thickness as well,
this effect is largely transitory, and the tear volume quickly
reverts to its prior value as the applied solution is drained away.
Thus, it is unrealistic to expect retention of applied uid
volume per se to provide long-term benet unless the solution
is frequently applied to the eye. Various compounds are in
development that approach tear lm deciency from different,
more therapeutic, angles some seek to upregulate mucin
production so that the tear adheres longer to the ocular surface,
some have antievaporative qualities, and some, like cyclosporine A, are antiinammatory agents. In the future, neurotransmitters and hormonal treatments may also be explored,
assuming the causes of many forms of dry eye to lie in the
neuronal feedback loop. Eventually, it should be possible to
permanently regulate tear volume and related factors such as
blink frequency, through treatment.
Key Features
Blink Dynamics
1215 blinks per minute is considered normal
Typically nonconscious, though can be performed consciously
Upper lid undergoes considerable vertical movement
(513 mm) at varying velocities (80300 mm/s) and spreads
tear lm over eye surface
Lower lid barely moves at all vertically, but does undergo fairly
substantial horizontal motion (24 mm)
Tear Mixing, Turnover, Drainage
The tear lm is anchored to the eye by the glycocalyx and
consists of mucins suspended in aqueous at varying
concentrations, covered by an outer lipid layer
Assessment of the height of the marginal tear menisci and the
geometry of the punctal opening can help approximate tear
volume
Tear volume can be measured using Schirmer strips or
uorophotometry
Fluid drains from the eye through the puncta, proceeds
through the canaliculi, and ends in the lacrimal sac
Tear ow or drainage is stimulated by blinking and can be
measured using uorophotometry
SECTION 4
Wetting and Drying of the Corneal Surface

The tear lm spreads across the ocular surface postblink, and
its outer lipid layer reduces surface tension
Tear lm breakup is marked by the appearance of dry spots on
the ocular surface
In healthy eyes, a blink should replenish the tear lm prior to
tear lm breakup
Measurements of the rapidity and pattern of tear lm breakup
can be used as diagnostic tests for dry eye
REFERENCES
1. Doane MG: Interaction of eyelids and tears
in corneal wetting and the dynamics of the
normal human eyeblink. Am J Ophthalmol
1980; 89:507516.
2. Mishima S, Gasset A, Klyce SD, Baum JL:
Determination of tear volume and tear ow.
2a. Yokoi N, Bron AJ, Tiffany JM, et al:
Relationship between tear volume and tear
meniscus curvature. Arch Ophthalmol
2004; 122:12651269.
3. Scherz W, Doane MG, Dohlman CH: Tear
volume in normal eyes and
keratoconjunctivitis sicca. Graefes Arch
Klin Exp Ophthalmol 1974; 192:141150.
4. Gbbels M, Goebels G, Breitbach R,
Spitznas M: Tear secretion in dry eyes as
assessed by objective uorophotometry.
Ger J Ophthalmol 1992; 1:350353.
5. Doane MG: Blinking and the mechanics of
286
6.
7.
8.
9.
the lacrimal drainage system.

Doane MG: Blinking and tear drainage. In:
Bosniak SB, ed. Advances in plastic and
reconstructive surgery. The lacrimal
system. New York: Pergamon; 1984:3952.
Abelson MB, Ousler GW, et al: Alternative
reference values for tear lm break-up in
normal and dry eye populations. In:
Sullivan D, et al, eds. Lacrimal gland, tear
lm, and dry eye syndromes 3. Kluwer
Academic/Plenum; 2002.
Ousler GW, Emory TB, Welch D, Abelson MB:
Factors that inuence the inter-blink
interval (IBI) as measured by the ocular
protection index (OPI). (Poster presentation)
The Association of Research in Vision and
Ophthalmology (ARVO); 2002.
Ousler GW, Lemp MA, Schindelar MR,
et al: Tear film break up patterns (TFBUP
10.
11.
12.
13.
14.
a novel method of evaluating tear film

stability (abstract). The Ocular Surface
2005; 3 suppl: S100.
Cope C, Dilly PN, Kaura R, Tiffany JM.
Wettability of the corneal surface: A
reappraisal. Curr Eye Res. 1986; 5:77785.
Tiffany JM. Measurement of wettability of
the corneal epithelium. I. Particle
attachment method. Acta Ophthalmol
(Copenh). 1990; 68:17581.
Tiffany JM. Measurement of wettability of
the corneal epithelim. II. Contact angle
method. Acta Ophthalmol (Copehn). 1990;
68:1827.
Gipson IK, Inatomi T. Mucin genes
expressed by the ocular surface epithelium.
Progr Retinal Eye Res 1997; 16:8198.
Sharma A, Coles WH. Physico-chemical
factors in tear film breakup. IOVS ARVO
abstract 1990; 31:552.
CHAPTER
27
Tear Substitutes
Mark B. Abelson, George Ousler, and Russell Anderson
Estimates of the prevalence of dry eye in the general population

range from 11% to 45%,13 and this prevalence only appears to
be increasing as factors such as prolonged computer use and
contact lens wear become more and more widespread. Although
treatment options such as punctal plugs, steroids, or prescription
therapy are available to patients, these methods tend to only be
efcacious on a patient-to-patient basis, and the majority of sufferers still primarily manage their dry eye with over-the-counter
tear substitutes. Estimates suggest that at least 710 million
Americans use articial tears.4
Key Features
Most patients manage dry eye with tear substitutes, which are
over-the-counter eyedrops intended to temporarily help relieve
the clinical signs and symptoms associated with dry eye.
Demulcents are lubricating compounds contained in articial
tears that help soothe the ocular surface, while viscosity
agents make for thicker drops that can augment the tear lm
and may require less frequent use.
Although some modern tear substitutes are preservative-free,
many still contain ophthalmic preservatives, which are useful
for their antimicrobial action, but can be potentially irritating to
dry-eye patients with sensitive ocular surfaces.
In formulating an articial tear, the other formulary aspects to
consider are electrolytes (these can make for a healthier tear
lm), osmolarity/osmolality (a tear should be slightly
hypoosmotic), and pH (drops should be somewhat alkaline).
Topical cyclosporine A is currently the only approved
prescription therapy for dry eye, but various classes of agents,
including secretagogs, antievaporatives, antiinflammatories,
mucomimetics, and even neuronal feedback loop regulators
are in development as potential dry-eye treatments.
A tear substitute is generally used to supplement a tear lm

that is inferior in either quality or quantity in a patient with some
dysfunction of the ocular surface or tear secretory system. As
described in other chapters, this system includes the main and
accessory lacrimal glands, the meibomian glands, the glands of
Zeis and Moll, and the goblet cells. Identication of the
patients specic dysfunction, whether it is an aqueous, lipid, or
mucous layer deciency, helps to determine optimal therapy.5
Whether dry eye is primary or secondary to lid conditions, such
as blepharitis or ocular rosacea, must be determined in order to
initiate the proper concomitant therapy, such as lid hygiene or
oral tetracycline.6,7 Disorders of the ocular surface such as ocular
cicatricial pemphigoid or StevensJohnson syndrome must be
identied and treated as described in other chapters.
Tear substitutes may come in various forms from somewhat
viscous liquids that are dispensed from bottles to thick ointments
squeezed from tubes and their components are restricted

within the connes of the US Food and Drug Administration
(FDA) monograph on over-the-counter products. This compendium lists all acceptable ingredients, both active and inactive, as
well as acceptable concentration ranges allowed by the FDA in
ophthalmic over-the-counter formulations. Ingredients that have
been historically and traditionally used in ophthalmic products
are included in this list based on the safety prole established
through their numerous years of use. The various ingredients
allowed by the FDA are classied as demulcents, emulsiers,
surfactants, viscosity agents, and preservatives.
Generally, tear substitutes are hypotonic or isotonic buffered
solutions containing demulcents, viscosity agents, electrolytes
and other components. Historically, all formulations were preserved, multidose preparations. However, unit-dose, preservativefree systems are now common. The type and concentration of
demulcent or viscosity agent, preservative system, and electrolyte
composition are the primary variables in ophthalmic lubricant
formulations. Various diagnostic measures may be used to help
determine the efcacy of an articial tear, from tear lm breakup
time (TFBUT) to corneal and conjunctival vital dye staining or
symptom questionnaires.8,9 Novel diagnostic tools such as ocular
protection index (OPI) and tear lm breakup patterns (TFBUP)
will most likely be used to evaluate future tear substitutes.10 In
addition to improving signs and symptoms of dry eye, a tear
substitute should ideally have a long duration of action to minimize the necessary frequency of instillation11 this is especially
important in the case of formulations containing potentially
irritating preservatives. The primary objectives of the physician
caring for patients with dry eye are to improve subjective comfort and to minimize ocular surface desiccation and cell death.
Symptoms can often be reduced but rarely are eliminated.
DEMULCENTS AND VISCOSITY AGENTS

Demulcents are compounds intended to protect mucous membranes and soothe epithelial surfaces. Simultaneously, their
mucilaginous consistency can provide lubricity for the ocular
surface, which can help minimize the abrasive action of the
upper lid on already desiccated epithelial cells. The FDA recognizes six categories of ophthalmic demulcents, with each category
containing one or more compounds: cellulose derivatives, dextran
70, gelatin, liquid polyols, polyvinyl alcohol, and povidone
(Table 27.1).12 The demulcents covered by the monograph are
allowed in over-the-counter preparations if they fall within
certain concentration ranges. An ophthalmic preparation may
contain up to three demulcents of any type, and in some cases,
as with dextran 70, combination with another demulcent is
required. Up to three demulcents may also be combined with
either a single ophthalmic vasoconstrictor or a vasoconstrictor/
287
TABLE 271 Marketed Tear Substitutes*

Concentration
(When Available)
Demulcent(s)
Carboxymethylcellulose
0.5%
Trade Name
Preservative
a
Refresh Tears
SOC
Refresh Plus
1.0%
1.0%
w/Glycerin
Glycerin
w/ Polysorbate 80
Hydroxypropyl cellulose
Hydroxypropyl methylcellulose
None
a
Refresh Liquigel
SOC
0.25%
TheraTears
0.5%
Optivea
0.3%
None
Celluvisca
None
SOC
c
Moisture Eyes
BAK
c
1.0%
Computer Eye Drops
1.0% (both)
Refresh Enduraa
5 mg/insert
0.3%
BAK, EDTA
None
d
Lacrisert (biodegradable insert)

e
GenTeal
Sodium perborate
GenTeal PFe
None
e
Genteal Gel Drops

0.2%
Sodium perborate
GenTeal Mild
Sodium perborate
Bion Tearsf
w/ Dextran 70
None
g
SECTION 4
Tears Renewed
BAK, EDTA
c
0.8%, 0.1%
Moisture Eyes Liquid Gel
BAK
0.8%, 0.1%
Moisture Eyes Liquid Gel PFc
None
w/ Glycerin
Clear Eyes CLR
Sorbic acid, EDTA

f
w/ Glycerin & Dextran 70
Tears Naturale Forte
Polyquaternium-1
Tears Naturale Freef
None
w/ Glycerin & PEG-400
Visine Tears
BAK
i
Methylcellulose
1.0%
Mineral Oil
4.5%
w/ Light Mineral Oil
1.0%
Visine Tears PF
None
Murocelc
Methylparabens,
Propylparabens
Polyhexamethylene biguanide
Soothe
Aquasitee
Polycarbophil, PEG-400, Dextran 70
EDTA
Aquasite Multi-Dosee
Polyvinyl Alcohol
1.4%
w/ PEG-400, Dextrose
1.0%
AKWA Tears
1.4%
Propylene Glycol, PEG-400
0.3%, 0.4%
Hypotears
BAK, EDTA
EDTA
Murine Tears
BAK, EDTA
Systane
Polyquaternium
Systane PFf
a
astringent combination in order to provide an articial tear with

additional redness- or discomfort-reducing properties.
Cellulose derivatives are the demulcents most commonly contained in modern tear substitutes and are allowed in concentrations between 0.2% and 2.5%. Hydroxypropyl methylcellulose
(HPMC) and carboxymethylcellulose (CMC) are the two most
None
Allergan Pharmaceuticals; Advanced Vision Research; Bausch and Lomb Pharmaceuticals; Merck & Co.; Novartis Pharmaceuticals; fAlcon Laboratories; gAKORN
Pharmaceuticals; hMedtech; iPzer, Inc.; jAlimera Biosciences; kRoss Laboratories.
*Concentrations of the listed components are identied when possible. Ethylenediaminetetraacetic acid (EDTA) is listed in the preservative column for some products,
though it is not technically a preservative. BAK is benzalkonium chloride, and SOC is stabilized oxychloro complex, both preservatives. Most information is available in
Physicians Desk Reference for Ophthalmology. 34th edn. Oradell, NJ: Medical Economics; 2006.
Preservative-free, unit-dose vials.
288
EDTA, sorbic acid

BAK, EDTA
Hypotears PFe
w/ Povidone
None
notable cellulose derivatives. Drops containing HPMC can be

formulated as oil-in-water emulsions, and the mucoadhesive properties of the polymer in combination with an oil can help supplement both the mucin and lipid components of the tear lm.13
Cellulose derivatives double as viscosity agents in the sense that
increasing their concentration can increase the viscosity of an
Tear Substitutes
TABLE 272 Dry Eye Ointments*

Trade Name
Primary Components and Concentration

(When Available)
AKWA tears
Ointmenta
White petrolatum, liquid lanolin, mineral oil
GenTeal PMb
85% White petrolatum, 15% mineral oil
Hypotears
Ointmentb
Tears Reneweda
White petrolatum, light mineral oil
Lacrilube S.O.P.c
56.8% White petrolatum, 42.5% mineral

oil, chlorobutanol, lanolin alcohols
Refresh P.M.c
42.5% Mineral oil, 57.3% white

petrolatum, lanolin alcohols
AKORN Pharmaceuticals; bNovartis Pharmaceuticals; cAllergan Pharmaceuticals.

*Concentrations of the listed components are identied when possible. Most
information is available in Physicians Desk Reference for Ophthalmology. 34th edn.
Oradell, NJ: Medical Economics; 2006.
(HP)-Guar is a gelling agent derived from guar gum that is used

in an articial tear with the demulcents PG and PEG 400 its
gelling action is apparently triggered by contact with the tear
lm.20 In even more viscous formulations dry-eye ointments
dispensed from tubes the primary components are often petrolatum or mineral oil (Table 27.2). These tear substitutes have
very long retention times and often provide signicant symptomatic relief for patients, but due to the excessive blurring
their high viscosities induce, they can usually only be instilled
in the evening prior to sleep.
PRESERVATIVES
The advent of preservative-free single-dose tear substitutes was
an important advancement in the management of dry eye. The
antimicrobial properties of preservatives are almost always
accompanied by mild toxicity, and in dry-eye patients, who often
have sensitive or damaged ocular surfaces to begin with and may
use a tear substitute numerous times daily, this can be particularly detrimental. Preservatives that have stronger antimicrobial
action are also often more toxic to the surface of the eye, so
unpreserved formulations that can be immediately discarded
after use are preferable for many patients.
However, there are drawbacks to the use of single unit-dose
tear substitutes: the expense and the inconvenience of carrying
many vials. For these reasons, some patients may attempt to
reuse their unit-dose vials by recapping them or standing them
upright until it is time for the next dose this greatly increases
the likelihood of contamination. The ideal articial tear would be
either a preservative-free unit-dose tear substitute that retains antimicrobial efcacy for a 24-h span after rst use, or a preservativefree multidose formulation that can maintain sterility even with
frequent use. Both possibilities are being explored, while other
formulations seek to incorporate effective preservatives that are
minimally toxic to the ocular epithelium.
FDA-required components of all multidose ophthalmic preparations since 1953, preservatives must pass specied efcacy
tests to gauge their antibacterial and antifungal action prior to
inclusion in an eye-drop. Traditional ophthalmic preservatives
that were found to have harsh effects on ocular surface cells and
cause discomfort in patients include thimerosal, chlorobutanol,
and sorbate.2124 These are primarily chemical preservatives, or
detergents, and typically exhibit excellent preservative efcacy.
Another chemical preservative, the antiseptic benzalkonium
chloride (BAK), is probably the most common preservative in
CHAPTER 27
articial tear and prolong its retention time in the tear lm.
This is evident in marketed formulations containing HPMC,
which include either 0.2% or 0.3% concentrations depending on
the severity of the dry eye they are intended to manage. CMC
varies more substantially, with different formulations containing
either 0.5% or 1.0% CMC the latter tear substitute is a thick,
gel-like liquid. However, when articial tear viscosity is increased
by means of higher demulcent concentrations, the resultant tear
is often prone to causing blurring of the vision and in some cases
will leave residue on the lid margins as it dries.14 At the same
time, the greater retention time yielded by the gel-like consistency is more likely to improve dry-eye signs and symptoms
than a less viscous tear would.
Liquid polyols (polyhydric alcohols) are also demulcents that
are commonly found in modern articial tears and are allowable
in concentrations of 0.21.0%. While polyols typically do not
double as viscosity agents like cellulose derivatives can, they are
often more effective lubricants.15 A comparative study demonstrated that a tear substitute containing propylene glycol 0.3%
(PG) and polyethylene glycol 0.4% (PEG 400) provided better
lubricity than a tear containing HPMC, which in turn was more
effective at creating lubricity between two moving surfaces than
a product containing CMC.16 Glycerin and polysorbate are additional liquid polyols that are often used in varying concentrations as combinative agents in oil emulsion systems that target
the lipid layer of the tear lm. Both are also found as inactive
agents in the cyclosporine A formulation that represents the
only FDA-approved prescription dry-eye therapy.
The remaining demulcents covered by the FDA monograph
include gelatin (allowable in 0.01% concentration, but seldom
used), povidone, and dextran 70 (which can only be incorporated
in conjunction with another demulcent) all three can serve as
viscosity agents as well as lubricants in articial tears. Polyvinyl
alcohol (PVA) was one of the original demulcents incorporated
into articial tears and can be included in concentrations from
0.1% to 4.0%. PVA is still used in some drops and can act alone
or in combination with another demulcent such as povidone.
The stability of the tear lm depends on the chemicalphysical
characteristics among the three layers. Classically, the mucin
layer was thought to act as a wetting agent by lowering the
surface tension of the relatively hydrophobic ocular surface,
rendering the corneal and conjunctival cells wettable.17
Evidence has shown, however, that the mucin layer is much
thicker than previously thought, and in fact extends into the
aqueous phase.18 Its role may be similar to that of mucin in
the stomach, where a mucin gel protects the epithelium from
a harsh surrounding environment.19 This may explain why
water-containing lubricants are only partially effective in
restoring the health of the ocular surface. The function of the
tear lms mucin component is more than that of a wetting
agent. The effect of most available lubricants is probably to help
hydrate available mucin and wash away irritating or toxic
substances in the tear lm. While some patients with dry eye
have a deciency in the aqueous layer, a primary or secondary
mucin deciency may also be present. The demulcents and
viscosity agents added to articial tears lubricate and can work
to fortify the mucin layer or the thin outer lipid layer, which
prevents tear lm evaporation. The addition of a viscosity agent
to increase residence time can play a role in active drug formulations by prolonging ocular surface contact, thereby increasing
the drugs duration of action and comfort.
While demulcents such as cellulose derivatives are often used
in high concentrations to increase the viscosity of articial tears,
other compounds, for instance gels and gelling agents, may also
be incorporated into formulations. Carbopol 980, a gel composed
of linked carboxylic acid polymers, is used in some marketed
tear substitutes to create a highly viscous drop. Hydroxypropyl
289
SECTION 4

modern ophthalmic formulations. However, BAK is not ideal
for inclusion in formulations intended for frequent dosing by
dry-eye patients due to its moderate toxicity to the ocular surface
in an in vitro study, human corneal epithelial cells exposed to
BAK underwent cell retraction as normal cytokinesis, cell movement, and mitotic activity were disrupted causing cell degeneration within 2 h.24 When it is present in a tear substitute, BAK is
usually incorporated in a 0.01% concentration, and is tempered
with ethylenediaminetetraacetic acid (EDTA) even in this low
concentration, it retains a preservative efcacy superior to many
other preservatives commonly found in articial tears.25
Belonging to the same class of compound as BAK, polyquaternium is a polymeric biocide that has not shown the same
propensity to irritate or disrupt ocular surface cells and, as a
result, is incorporated into several currently marketed articial
tears. Polyquaternium-1, a quaternary ammonium compound,
exhibited no negative effects on cell movement or mitotic activity
in human corneal surface cells when tested in vitro, nor did it
cause any cell death in the cultures.24 Although it is highly
effective against ophthalmic bacteria, some tests suggest polyquaternium to have a more limited ability to eradicate fungi in
certain product formulations.26
In addition to chemical preservatives, a second class known
as oxidative preservatives have become common tear substitute
components. The two primary oxidative preservatives used in
dry-eye formulations are stabilized oxychloro complex (SOC)
and sodium perborate. SOC is a mixture of compounds with
sodium chlorite (NaClO2) as the main component. It is light sensitive, so that articial tears containing SOC must be dispensed
from opaque bottles. Although it is considered quite mild and not
prone to irritation of the ocular surface, SOC also has exhibited
weak anti-bacterial action in some preservative efcacy trials.25
Sodium perborate is a bleaching agent with antiseptic properties
that is thought to break down upon contact with the tear lm.
However, hydrogen peroxide (H2O2), which sodium perborate
breaks down into, has been shown to have toxic effects on the
cornea in one study it brought about epithelial cell death in
human corneal cell cultures 1224 h after exposure.24,27 As an
oxidative preservative, sodium perborate is generally considered
milder on the ocular surface than most chemical preservatives,
and is found in current marketed articial tears.
Any tear substitute that contains a preservative has the potential to irritate the ocular surface of a dry-eye patient, especially
a patient who has severe epithelial damage to begin with. Harmful
effects may be minimized by less frequent dosing or limiting
drop volume upon instillation, but ultimately a preservativefree formulation is the best way to avoid the ocular irritation
ophthalmic preservatives sometimes cause.28
OTHER FORMULARY ASPECTS OF TEAR

SUBSTITUTES
ELECTROLYTE COMPOSITION
290
Electrolytes occur naturally in physiological fluids and they help

regulate metabolic processes in the tear lm. Tear substitutes
with electrolyte compositions have been shown to help improve
and restore damaged corneal surfaces.2933 One of two ions
can be found in most articial tears containing electrolytes:
bicarbonate and potassium. Bicarbonate can help in the
recovery of damaged corneal epithelial cells and in the
maintenance of ocular surface health.30 Previous studies suggest
that naturally occurring bicarbonate in the human body is
responsible for aiding in the maintenance of the mucin gel that
lines and protects the stomach.19 Similarly, it is believed that
bicarbonate as an ingredient in tear substitutes may help
maintain the mucin layer of the tear lm.30 This theory was
supported by a rabbit model study in which treatment with an

electrolyte solution yielded an increase in goblet cell density
while simultaneously decreasing conjunctival staining and tear
osmolarity.33 Potassium can also be a useful element in
ophthalmic solutions due to its ability to retain corneal
thickness.34
Some currently marketed products attempt to mimic the electrolyte composition of human tears. One of these formulations
has demonstrated an ability to increase goblet cell density in
LASIK-induced dry eye.35 Because bicarbonate reacts with air to
produce carbon dioxide, it is often necessary for these electrolytebalanced drops to be dispensed from single-dose plastic vials
and packaged in foil.
OSMOLARITY/OSMOLALITY
Analyses of dry-eye patients have demonstrated that dry eye is
consistent with increased tear lm osmolarity (crystalloid osmolarity), most likely owing to increased evaporation in patients
with lipid layer deciencies.36,37 Hyperosmolarity may be toxic
to the corneal epithelium, compounding already-present surface
damage.38 For this reason, some articial tears aim to lower the
osmolarity of the tear lm via hypoosmotic formulation.
Colloidal osmolality, mainly dependent on macromolecule
content, is another factor that varies in articial tear formulations. Colloidal osmolarity, or oncotic pressure, is important for
the control of water transport in tissues it is dened as osmotic
pressure due to the presence of colloids in a solution. Differences
in osmolality affect the net water flow across membranes. This
water flow is eliminated by applying hydrostatic pressure to the
downside of the water flow. The magnitude of osmotic pressure
is determined by respective osmolalities on the two sides of the
membrane. Damaged epithelial cells swell due either to breaks
in the cell membranes or pumping mechanism dysfunctions. If
a fluid with a high colloidal osmolality is added to this damaged
and swollen cell surface, the oncotic pressure exerted causes cell
deturgescence and a return to normal cell physiology. Thus, an
articial tear formulation with a high colloidal osmolality may
be of value.
VISCOSITY
Viscosity, or fluid thickness, is still considered one of the most
important properties of an articial tear. Viscosity is typically
measured in centipoise, and may be enhanced in a formulation
through the inclusion of any of the previously mentioned
viscosity agents (i.e., gelling agents, demulcents, etc.). The more
viscous a tear substitute is, the longer it is expected to remain
in the tear lm, and the greater benet to signs and symptoms
of dry eye it is expected to yield. Of course, as a result of its thicker
constitution, a highly viscous articial tear or ointment is prone
to blurring the vision after instillation, and in some cases may
leave a residue on the lashes or lids as it dries. Articial tears
seek to achieve a viscosity that maximizes both clinical efcacy
and visual clarity.
ALKALINITY
Research has suggested that patients with ocular surface disease
or tear lm deciency tend to have higher than average tear lm
pH values.39,40 Ocular rosacea, in particular, has been associated
with increased tear alkalinity and it is believed that dry-eye
patients have similarly alkaline tears.39 It is difcult to standardize pH measurement in tears, because multiple factors such
as goblet cell secretions, tear flow rate, conjunctival metabolism
and carbon dioxide escape can influence pH levels.41 Some estimates place the mean pH in healthy eyes ~7.57.6.40,42 It has
also been demonstrated that when non-dry-eye patients hold

their eyes open for 60 s, the tear-lm pH can increase to greater
than 9.40 This effect occurs when bicarbonate in the tear-lm
alkalizes in an effort to achieve an equilibrium with the partial
pressure of carbon dioxide in the surrounding air.40 A marketed
articial tear that contains the gelling agent HP-Guar utilizes
this pH adjustment in the tear lm to trigger its gelling action.20
Blinking and tear production appear to reverse increasing alkalinity in tears and lower pH levels it is presumed that articial
tear instillation has a similar effect.
Several clinical studies have looked at the relationship between
the pH value of articial tears and patient drop preference and
tolerability.43 On the whole, it has been shown that patients
prefer to dose with slightly alkaline isotonic tear substitutes.44
When articial tears that vary in pH are tested, patient tolerability is greater for drops that are more alkaline.43,44
DRY-EYE THERAPIES AND

INVESTIGATIONAL COMPOUNDS
Tear substitutes, though capable of managing dry eye, do not
generally treat it to a degree constituent of therapy. Currently
only one active compound cyclosporine A, a partial immunomodulator believed to limit the proliferation of T lymphocytes
is approved and indicated for treatment of dry-eye syndrome
(Table 27.3).45 In the US cyclosporine has only shown clinical
efcacy in patients with aqueous decient dry eye, although lab
tests have suggested it may increase goblet cell density.46,47 In
its current marketed eyedrop formulation, however, cyclosporine A
is reported to cause discomfort in 17% of patients upon
instillation.46 Some tear substitutes have found further potential use as concomitant treatment in patients using topical
cyclosporine clinical trials suggest that this dosing combination
can yield improvements in patient symptoms and clinical signs.48
Many compounds are currently under investigation as potential dry-eye therapies. Secretagogs, such as diquafosol tetraso-
dium, ecabet sodium, gefarnate, rebamipide, and 15(S)-HETE,

are being studied in the hope they may upregulate production of
one or more of the tear lm layers (primarily the mucin
layer).4952 Antievaporative agents such as sodium hyaluronate
that are intended to increase tear dwell time by boosting lipid
layer production are also in development. The exploratory compounds known as mucomimetic agents incorporate synthetic
mucins, ideally to enhance both the mucin and aqueous layers
and promote adhesion of tears to the ocular surface. Antiinflammatory compounds both steroids and nonsteroidal antiinflammatory drugs (NSAIDs) are being tested to see if they
can improve signs and symptoms of dry eye.53 Some clinicians
currently prescribe corticosteroid drops off-label on a pulse
regimen for dry eye, and it is believed that this treatment may
prove quite effective against the signs and symptoms of dry eye.
With any ocular steroid treatment, however, there is always
concern about the possibility of elevated intraocular pressure
(IOP) levels.
Oral supplements, particularly Omega 3 fatty acids, have been
studied for their effects on the lacrimal system and are often
marketed or recommended by physicians for dry-eye management or prevention purposes.54,55 In the foreseeable future, dry-eye
treatment may attempt to go directly to the source of the condition by utilizing neuronal feedback loop regulators to counteract
the neurological changes that are believed to catalyze the onset
of dry eye.56 Hormonal treatments (e.g., androgen) are being
considered and tested as potential therapies due to hormonal
deciencies associated with some dry-eye etiologies.57,58 Neurotransmitters could soon be investigated for their potential effects
on dry-eye syndrome as well.
Ultimately, even if dry-eye therapy does move quickly forward,
tear substitutes will not disappear. They are useful for as-needed
dosing, and could prove effective as concomitant medications
with treatments. Also, tear substitutes are becoming increasingly
efcacious at individually managing the signs and symptoms of
dry-eye syndrome.
CHAPTER 27
Tear Substitutes
TABLE 273 Prescription Therapy for Dry Eye

Active Components
a
Cyclosporine
Concentration
Trade Name
Preservative
0.05%
Restasis*
None
Allergan Pharmaceuticals.
* Preservative-free, unit-dose vials. All tabulated information is available in Physicians Desk Reference for
Ophthalmology. 34th edn. Oradell, NJ: Medical Economics, 2006.
REFERENCES
1. Schaumberg DA, Sullivan DA, Dana MR:
Epidemiology of dry eye syndrome. In:
Sullivan DA, Stern ME, Tsubota K, et al,
eds. Lacrimal gland, tear lm, and dry eye
syndromes 3, Part B. New York: Plenum;
2002:989998.
2. Lemp MA: Epidemiology and classication
of dry eye. In: Sullivan DA, Dartt DA,
Meneray MA, eds. Lacrimal gland, tear lm
and dry eye syndromes 2. New York:
Plenum; 1998:791803.
3. Caffery BE, Richter D, Simpson T, et al:
The Canadian dry eye epidemiology study.
In: Sullivan DA, Dartt DA, Meneray MA,
eds. Lacrimal gland, tear lm and dry eye
syndromes 2. New York: Plenum;
1998:805806.
4. Kaswan R: Cyclosporine drops: a potential
breakthrough for dry eyes. In: Research to
prevent blindness science writers seminar.
5.
6.
7.
8.
9.
New York: Research to Prevent Blindness;

1989:1820.
Abelson MB, Lamberts DW (eds): Dry eye
update. Princeton, NJ Excerpta Medica; 1983.
Smith RE, Flowers CW Jr. Chronic
blepharitis: a review. CLAO J 1995;
21:200207.
Quarterman MJ, Johnson DW, Abele DC,
Lesher JL Jr, et al: Ocular rosacea: signs,
symptoms, and tear studies before and
after treatment with doxycycline. Arch
Dermatol 1997; 133:4954.
Abelson MB, Ousler GW III, Nally LA, et al:
Alternative reference values for tear lm
break up time in normal and dry eye
populations. Adv Exp Med Biol 2002;
506(Pt B):11211125.
Lemp MA: Report of the National Eye
Institute/Industry workshop on Clinical Trials
in Dry Eyes. CLAO J 1995; 21:221232.
10. Ousler GW III, Gomes PJ, Welch D,

Abelson MB: Methodologies for the study
of ocular surface disease. Ocul Surf 2005;
3:143154.
11. Abelson MB, Knight E: Dry eye therapy:
evaluation of current directions and clinical
trials. Adv Exp Med Biol 1994;
350:431436.
12. Code of Federal Regulations: Title 21, Sec.
349.12, 349.3, 349.6.
13. Rieger G: Lipid-containing eye drops: a
step closer to natural tears.
14. Ridder WH III, Lamotte JO, Ngo L,
Fermin J: Short-term effects of articial
tears on visual performance in normal
subjects. Optom Vis Sci 2005;
82:370377.
15. Meyer AE, Baier RE, Chen H, Chowhan M:
Differential tissue-on-tissue lubrication by
291
16.
17.
18.
19.
20.
21.
22.
SECTION 4
23.
24.
25.
26.
27.
28.
29.
30.
292
ophthalmic formulations. J Biomed Mater

Res B Appl Biomater 2007; 82: 7488.
Ketelson HA, Asgharian B, Chowhan M,
Meadows D. Characterization of physical
property attributes for polymer systems
used in articial tear products. Invest
Ophthalmol Vis Sci 2004; 45:E-Abstract 70.
Holly F, Lemp M: Wettability and wetting of
corneal epithelium. Exp Eye Res 1971;
11:239250.
Prydal JI, Artal P, Woon H, Campbell FW:
Study of human precorneal tear lm
thickness and structure using laser
interferometry. Invest Ophthalmol Vis Sci
1992; 33:20062011.
Slomiany BL, Slomiany A: Role of mucus in
gastric mucosal protection. J Physiol
Pharmacol 1991; 42:147161.
Christensen MT, Cohen S, Rinehart J, et al:
Clinical evaluation of an HP-guar gellable
lubricant eye drop for the relief of dryness
of the eye. Curr Eye Res 2004; 28:5562.
Simmons PA, Clough SR, Teagle RH,
Jaanus SD: Toxic effects of ophthalmic
preservatives on cultured rabbit corneal
epithelium. Am J Optom Physiol Opt 1988;
65:867873.
Fassihi AR, Naidoo NT: Irritation associated
with tear-replacement ophthalmic drops: a
pharmaceutical and subjective investigation.
S Afr Med J 1989; 75:233235.
Josephson JE, Caffery B: Sorbic acid
revisited. J Am Optom Assoc 1986;
57:188189.
Tripathi BJ, Tripathi RC, Kolli SP:
Cytotoxicity of ophthalmic preservatives on
human corneal epithelium. Lens Eye Toxic
Res 1992; 9:361375.
Charnock C: Are multidose over-the-counter
articial tears adequately preserved?
Cornea 2006; 25:432437.
Codling CE, Maillard JY, Russell AD:
Aspects of the antimicrobial mechanisms
of action of a polyquaternium and an
amidoamine. J Antimicrob Chemother
2003; 51:11531158.
Hayden BJ, Zhu L, Sens D, Tapert MJ,
Crouch RK: Cytolysis of corneal epithelial
cells by hydrogen peroxide. Exp Eye Res
1990; 50:1116.
Burstein NL: The effects of topical drugs
and preservatives on the tears and corneal
epithelium in dry eye. Trans Ophthalmol
Soc UK 1985; 104(Pt 4):402409.
Gilbard J, Rossi S, Heyda KG: Ophthalmic
solutions, the ocular surface, and a unique
therapeutic articial tear formulation. Am J
Ubels J, McCartney M, Lantz W, et al:
Effects of preservative-free articial tear
solutions on corneal epithelial structure and
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
function. Arch Ophthalmol 1995;

113:371378.
Bernal DL, Ubels JL: Articial tear
composition and promotion of recovery of
the damaged corneal epithelium. Cornea
1993; 12:115120.
Nelson J, Drake M, Brewer J, Tuley M:
Evaluation of physiologic tear substitute in
patients with keratoconjunctivitis sicca.
Adv Exp Med Biol 1994; 350:453457.
Gilbard JP, Rossi SR: An electrolyte-based
solution that increases corneal glycogen
and conjunctival goblet-cell density in a
rabbit model for keratoconjunctivitis sicca.
Green K, MacKeen DL, Slagle T, Cheeks L:
Tear potassium contributes to maintenance
of corneal thickness. Ophthalmic Res 1992;
24:99102.
Lenton LM, Albietz JM: Effect of
carmellose-based articial tears on the
ocular surface in eyes after laser in situ
keratomileusis. J Refract Surg 1999;
15(2 Suppl):S227S231.
Gilbard JP: Tear lm osmolarity and
keratoconjunctivitis sicca. CLAO J 1985;
11:243250.
Rolando M, Refojo M, Kenyon K: Increased
tear lm evaporation in eyes with
keratoconjunctivitis sicca. Arch Ophthalmol
1983; 101:557.
Gilbard J, Carter J, Sang D, et al:
Morphologic effect of hyperosmolarity on
rabbit corneal epithelium. Ophthalmology
1984; 91:1205.
Abelson MB, Sadun AA, Udell IJ, Weston JH:
Alkaline tear pH in ocular rosacea. Am J
Ophthalmol 1980; 90:866869.
Fischer FH, Wiederholt M: Human
precorneal tear lm pH measured by
microelectrodes. Graefes Arch Clin Exp
Ophthalmol 1982; 218:168170.
Janszky I, Vamosi P, Orszagh I, Berta A:
Demonstration of increasing pH calue of
lacrimal fluid with increase of flow rate.
Acta Ophthalmol Scand 2001; 79:180183.
Yamada M, Mochizuki H, Kawai M, et al:
Fluorophotometric measurement of pH of
human tears in vivo. Curr Eye Res 1997;
16:482486.
Motolko M, Breslin CW: The effect of pH
and osmolarity on the ability to tolerate
articial tears. Am J Ophthalmol 1981;
91:781784.
Raber I, Breslin CW: Toleration of articial
tears the effect of pH. Can J Ophthalmol
1978; 13:247249.
Tatlipinar S, Akpek EK: Topical ciclosporin
in the treatment of ocular surface
disorders. Br J Ophthalmol 2005;
89:13631367.
46. Sall K, Stevenson OD, Mundorf TK, et al:

Two multicenter, randomized studies of the
efcacy and safety of cyclosporine
ophthalmic emulsion in moderate to severe
dry eye disease. Ophthalmology 2000;
107:631639.
47. Kunert KS, Tisdale AS, Gipson IK: Goblet
cell numbers and epithelial proliferation in
the conjunctiva of patients with dry eye
syndrome treated with cyclosporine.
48. Sall KN, Cohen SM, Christensen MT,
Stein JM: An evaluation of the efcacy of a
cyclosporine-based dry eye therapy when
used with marketed articial tears as
supportive therapy in dry eye. Eye Contact
Lens 2006; 32:2126.
49. Gamache DA, Wei ZY, Weimer LK:
Spellman JM, Yanni JM: Preservation of
corneal integrity by the mucin secretagogue
15(S)-HETE in a rabbit model of
desiccation-induced dry eye. Adv Exp Med
Biol 2002; 506(Pt A):335340.
50. Hamano T: Dry eye treatment with eye
drops that stimulate mucin production.
Adv Exp Med Biol 1998; 438:965968.
51. Urashima H, Okamoto T, Takeji Y, et al:
Rebamipide increases the amount of
mucin-like substances on the conjunctiva
and cornea in the N-acetylcysteine-treated
in vivo model. Cornea 2004; 23:613619.
52. Fujihara T, Murakami T, Fujita H, et al:
Improvement of corneal barrier function by
the P2Y(2) agonist INS365 in a rat dry eye
model. Invest Ophthalmol Vis Sci 2001;
42:96100.
53. Abelson MB, Sloan J: Nonsteroidal antiinflammatory drugs: current ophthalmic
therapy. J Fla Med Assoc 1994;
81:261263.
54. Miljanovic B, Trivedi KA, Dana MR, et al:
Relation between dietary n-3 and n-6 fatty
acids and clinically diagnosed dry eye
syndrome in women. Am J Clin Nutr 2005;
82:887893.
55. Brown NA, Bron AJ, Harding JJ, Dewar
HM: Nutrition supplements and the eye.
Eye 1998; 12(Pt 1):127133.
56. OBrien PD, Collum LM: Dry eye: diagnosis
and current treatment strategies. Curr
Allergy Asthma Rep 2004; 4:314319.
57. Sullivan DA: Androgen deciency and dry
eye syndromes. Arch Soc Esp Oftalmol
2004; 79:4950.
58. Sullivan DA, Edwards JA: Androgen
stimulation of lacrimal gland function in
mouse models of Sjogrens syndrome.
J Steroid Biochem Mol Biol 1997;
60:237245.
CHAPTER
28
Viscoelastics
Jack V. Greiner
Viscoelastic substances have viscous and elastic properties. In

ophthalmology, viscoelastics are used in articial tears and surgical formulations. Viscoelastic articial tears and contact lens
rewetting solutions are used to treat the ocular surface and relieve
ocular discomfort. Viscoelastics are also used in anterior surgical procedures like cataract extraction, intraocular lens (IOL)
implantation and exchange, keratoplasty, anterior chamber
reconstruction, and in some vitreoretinal procedures.
PHYSICAL AND CHEMICAL PROPERTIES

OF VISCOELASTIC MATERIALS
The ideal viscosurgical material would be noninflammatory, nonpyogenic, nontoxic, nonantigenic, and highly viscous. It should
be able to pass through small channels such as ne cannulas,
30-gauge needles, or pores of the trabecular meshwork. Elastic
qualities should enable viscoelastic substances to rebound after
mechanical stress or compression. A number of natural and synthetic viscoelastic polymers are used in viscosurgery (Table 28.1).
Healon was the rst viscoelastic sodium hyaluronate solution
to be marketed for ophthalmic use. Sodium hyaluronates combined viscous, elastic, and pseudoplastic properties make it well
suited for anterior segment surgical applications. Healon is derived
from rooster combs and is available as a 1% sterile solution
sealed in 0.51 mL glass syringes (10 mg/mL). Blue-tinted Healon
is also available to facilitate intraocular visualization of the
polymer.1 Provisc is comparable to Healon in terms of sodium
hyaluronate concentration, viscosity, and cohesive properties.
Among Healon formulations, Healon5 has the highest viscosity
and elasticity when exposed to shearing and is most likely to
improve maintenance of anterior chamber depth.2 Healon5
exhibits a dynamic frequency dependence in the presence of
turbulence and phacoemulsication power (continuous high
shear rates). Healon5 has been reported to fragment dispersively
and form a cavity with an outer retentive shell during phacoemulsication.2 The cohesive and dispersive properties of
Healon5 are better than Healon and Healon GV for all stages of
phacoemulsication.2
Like Healon, Amvisc is a sodium hyaluronate solution puried from rooster combs. Unlike Healon, Amvisc is formulated
to a consistent viscosity with a concentration varying between
1% and 1.4% hyaluronate (Healon is prepared to a specic concentration of 1% with a variable viscosity). Amvisc Plus has a
higher concentration of sodium hyaluronate (1.6%), and is 30%
more viscous than Amvisc. Healon GV has a 1.4% sodium
hyaluronate concentration and is considerably more viscous
than other ophthalmic viscoelastic preparations.
AMO Vitrax contains 3% sodium hyaluronate, the highest
concentration currently available. Viscoat combines 3% sodium
hyaluronate with 4% chondroitin sulfate. Chondroitin sulfate is
structurally similar to HA. Chondroitin sulfate has a double

negatively charged sulfate group per repeating disaccharide subunit. HA has one negative charged sulfate group per subunit.
Chondroitin sulfate (CS) is not a pseudoplastic fluid; it maintains a consistent viscosity at various shear rates. One randomized clinical comparison of Healon GV and Viscoat found both
formulations effectively protected the corneal endothelium during
endocapsular phacoemulsication and IOL implantation.3 The
median thicknesses of Amvisc Plus, Healon GV, and Viscoat
remaining adherent to the endothelium after phacoemulsication,
however, were found to be different.4 Viscoat provided the endothelium with the greatest protection, according to the study.4
Different viscoelastic materials are suitable for different types
of surgical procedures (see Table 28.1). Cohesive (or dispersive)
viscoelastics tend to adhere to ocular surfaces, protecting them
without excessive leakage during irrigation. Low cohesive properties are generally advantageous during iris plane and anterior
chamber phacoemulsication, particularly when endothelial protection is critical, for example, in Fuchs endothelial dystrophy.
Dispersive viscoelastics are difcult to remove. Cohesive viscoelastics are more easily aspirated from the eye. More cohesive
viscoelastics are desirable for anterior chamber maintenance,
when tissue manipulation and easy removal are the principal
goals. With high positive vitreous pressure, cohesive viscoelastics have the ability to create and maintain a deep anterior
chamber. Cohesive viscoelastics are effective during capsulorrhexis and IOL implantation, particularly when very ne foldable lenses are used.
DuoVisc is a viscoelastic system, containing Viscoat and
Provisc in separate syringes, which allows the surgeon to choose
the appropriate viscoelastic material. Viscoats tissue protective
properties are preferable in the initial stages of an anterior segment
procedure such as extracapsular cataract extraction by phacoemulsication. Proviscs cohesive properties would be more appropriate for later phases of the procedure, such as expansion of the
capsular opening, maintaining space, and IOL implantation.
Hydroxypropyl methylcellulose (2% solution) also has been used
successfully as an adjunct to anterior segment surgery. Hydroxypropyl and methyl groups make this linear polymer of glucose
more hydrophilic than its parent molecule, cellulose. Occucoat
and Cellugel are available in the same concentration but different
viscosities (see Table 28.1). Both are less viscous than Healon.
INDICATION FOR USE

The United States Food and Drug Administration (FDA) has
classied visocelastics as devices (not drugs). Viscoelastics are
indispensable in certain procedures in which the maintenance
of anatomic spaces and traumatic tissue manipulation are
required. They can also be used to lubricate and protect the eye.
293
294
AMO
AMO
AMO
Healon GV
Healon5
CoEase
Viscosity measured at 1 second at 25C.
-1
Cytosol
AMO
Healon
Shellgel
Alcon
Cellugel
AMO
Alcon
DisCoVisc
STAAR Surgical
Alcon
Duovisc
STAARVisc II
Alcon
Provisc
Vitrax
Bausch & Lomb
Alcon
Bausch & Lomb
Amvisc
Viscoat
Bausch & Lomb
Amvisc Plus
Occucoat
Company
Product
1.2% HA
1.2% HA
3.0% HA
1.2% HA
2.3% HA
1.4% HA
1.0% HA
2.0% HPMC
1.7% HA
4.0 CDS
Small (0.35 mL Viscoat/0.4 mL

Provisc)
Large (0.5 mL Viscoat/0.55 mL
Provisc)
1.0% HA
3.0% HA
4.0% CDS
2% HPMC
1.2% HA
1.6% HA
Composition
(in Saline)
TABLE 28.1 Comparison of Viscoelastics
>2.0 million
>2.0 million
0.5 million
>2.0 million
4.0 million
5.0 million
4.0 million
0.3 million
1.7 million
2.5 million
NaHA>500 000
CDS 22 500
>0.08 million
>2.0 million
1.6 million
Molecular Weight
(Daltons)
40 000
40 000
40 000
40 000
300 000
2 000 000
200 000
20 000
40 000
110 000
30 000
40 000
50 000
4000
40 000
55 000
Viscosity
(cP)
SECTION 4
High
High
Low
High
High
High
High
None
Medium
High
Low
None
High
Medium
Cohesion
320
320
310
320
320
302
302
315
29832
310
325
285
320
340
Osmolarity
(mOsM)
6.87.6
6.87.6
7.07.5
6.87.6
6.87.6
7.07.5
7.07.5
7.2
6.87.6
7.2
7.2
7.2
5.57.0
5.57.0
pH
28
28
Room temperature
28
28
28
28
Room temperature
28
28
28
Room temperature
28
28
Storage
Conditions (C)
Viscoelastics
CATARACT EXTRACTION
When injected into the cleavage plane between the lens nucleus
and cortex, viscoelastics can greatly facilitate phacoemulsication
of the nucleus during cataract extraction.5
Such viscodissection is especially useful when the cataract has
a soft nucleus, negotiating the phacoemulsication tip beneath
the nucleus is difcult, and zonular tears or posterior lens capsular ruptures could occur.6 Viscoelastic materials can also
maintain hydration of the ocular surface for extended periods
during surgery. Nuclear viscoexpression has been recommended
after capsulorrhexis during extracapsular cataract extraction.711
The superviscous properties of Healon5 appear to lead to a
higher completion rate of continuous curvilinear capsulorrhexis
in pediatric cataract surgery.12 Differences in osmolarity among
viscoelastic substances (Table 28.1), may explain the differences
in corneal thickness following cataract surgery. Viscoelastic substances with osmolalities of 305 mOsmol/kg or slightly higher
may be preferable, especially in patients with compromised
corneal endothelial cells.13
VISCOANESTHETICS
Mixtures of viscoelastics and anesthetics such as hydroxypropyl
methylcellulose 2.25% and licocaine 1%14 or sodium hyaluronate
1.5% and lidocaine 1%15 may minimize patients pain and discomfort during cataract operations.
RECOVERY OF SUBLUXATED LENS

Sodium hyaluronate has been successfully used for severely subluxated lens removal.16 Injections of SH can elevate the lens,
prevent total luxation, and simplify lensectomy. Viscoelastic dissection has been used for relocation of off-axis IOL implants.17
ENDOTHELIUM
Viscoelastics are able to protect the corneal endothelium from
mechanical trauma in anterior chamber surgery especially during
IOL insertion. Metallic instruments can cause cataracts with even
a slight touch to the crystalline lens. Viscoelastics can minimize
such operative complications.
Glasser and colleagues1 compared Healon, Amvisc, and Viscoat
and found that all three viscoelastics provided complete corneal
endothelium protection during contact with an IOL in vitro.
However, a more recent study by Glasser et al18 discovered that
Viscoat was better than Healon at preventing endothelial cell
loss in vivo during phacoemulsication with IOL implantation.
The authors hypothesize that chondroitin sulfate in Viscoat
makes the viscoelastic more adherent to the corneal endothelium,
and therefore, more protective. Viscoat also effectively protects
the endothelium from air-bubble damage.19 Physical trauma to
the endothelium can be prevented by coating the IOL with a
viscoelastic polymer before implantation.
PUPILS
Eyes receiving hydroxypropyl methylcellulose may develop noncreative semidilated pupils more readily than eyes receiving
sodium hyaluronate, according to one study (Healonid).20 How-
ever, a later study reported no statistical difference in pupil size

or reactivity after the use of Occucoat or Healonid in the course
of cataract surgery.21
INTRAOCULAR PRESSURE
IOP may increase postoperatively following the use of viscoelastics.22 This transient rise in IOP characteristically occurs
624 h after surgery and usually resolves spontaneously within
72 h.23 Berson et al24 have suggested that viscosurgery-associated
IOP elevations may be due to mechanical obstruction of aqueous
outflow by the viscoagent. They recommend thoroughly irrigating and aspirating the eye with a balanced salt solution to
remove the viscoagent. In some instances, it may be necessary
to treat the elevated IOP with antiglaucoma medications.
PROTECTIVE EFFECT ON THE CORNEAL

SURFACE
In corneal surgery, viscoelastics are primarily used to protect
corneal endothelial cells. However, viscoagents can also be applied
to the corneal surface during anterior segment procedures to
prevent the trauma and desiccation of the corneal epithelium.
Corneal surfaces coated with viscoagents prior to cataract extraction do not need to be repeatedly rehydrated with a balanced salt
solution during surgery. The use of a topical viscoagent during
corneal surgery signicantly improved corneal epithelial integrity
1 week after keratoplasty,25 according to one study.
REATTACHMENT OF DESCEMETS
MEMBRANE
One complication of sodium hyaluronate injection, and IOL or
surgical instrument insertion through the corneoscleral or
corneal wound is Descemet membrane detachment.2628
Sodium hyaluronate29,30 can be used to move Descemets membrane back to its normal anatomic position, and avoid further
detachment.
CHAPTER 28
In ophthalmology, viscoelastics are most commonly used in

articial tears, and rewetting solutions. In ocular surgery, viscoelastics are most commonly used during cataract extraction.
Comparative studies have demonstrated that all viscoelastics
(Table 28.1) effectively maintain the intraocular space and control
posterior pressure while intraocular tissues are manipulated.
GLAUCOMA FILTRATION SURGERY

Viscoelastic materials can be used in glaucoma ltration procedures. Viscoelastics have been shown to prevent the collapse
of the anterior chamber and stabilize early postoperative pressure.3133 One study found glaucoma ltering procedures with
Healon resulted in permanent blebs, more open clefts, less
scarring, less peripheral anterior synechia formation, and signicantly lower long-term IOP.34 Viscoelastics can also be used
to dilate Schlemms canal in viscocanalostomy.35
VITREOUS INCARCERATION
Sodium hyaluronate has been used to treat vitreous incarceration
in patients with corneal decomposition.36,37 Filling the anterior
chamber with Healon may reduce postoperative corneal complications during neodymium: yttriumaluminum-garnet treatment
for vitreolysis.
INTRAOCULAR HEMORRHAGE
Viscoagents can be used to control intraocular hemorrhage. Viscoelastic materials trap clotted blood in the anterior chamber,
however, so viscoelastics should be used cautiously if blood is
present. Ten percent sodium hyaluronate can be used to manage
suprachoroid hemorrhages postoperatively.38,39 Sodium
hyaluronate allows for good visualization of instruments in the
eye and avoids image minication and distortion from the
295

airfluid interface. Although balanced salt solution can be used,
sodium hyaluronate viscoelastic is less likely to egress through
rents in the posterior lens capsule or between zonular bers and
therefore provides a more effective and durable expansion of
intraocular volume.38 Viscoelastics can also prolong the maintenance of the IOP after ltering surgery.40 Sustaining the IOP
would help facilitate drainage of suprachoroid hemorrhage while
avoiding choroid effusion and hemorrhage incurred by ocular
hypotonia. Using a generous amount of Healon and flattening
the retinochoroid elevations of a suprachoroid hemorrhage promotes expression of blood from the suprachoroid spaces.41
RETINAL DETACHMENT SURGERY

Viscoelastics can be used for retinal detachment repair. For
example, suprachoroid implantation of a viscoelastic substance
can temporarily induce a choroid elevation for closing, retinal
tears.4246 Sodium hyaluronate has even been used to repair giant
retinal tears.47,48
SECTION 4
VITRECTOMY SURGERY
Procoagulate effects of HA after diabetic vitrectomy have been
reported,49 and sodium hyaluronate has been used to perform
delamination at the vitreoretinal juncture in diabetic eye disease. Such viscodelamination can separate the vitreous cortex
from the brovascular epiretinal membranes.50 Viscodelamination is especially valuable in eyes with combined traction and
rhegmatogenous retinal detachment. The viscodelamination
technique has a signicant risk of retinal breaks, however. The
risk is particularly high when adherent brovascular epiretinal
membranes are elevated excessively.50 Healon has been used to
elevate epiretinal membranes from the retina.51
LACRIMAL SURGERY
Sodium hyaluronate, injected into the lacrimal sac, is useful for
identifying the extent of the sac lumen.52,53 Sodium hyaluronate
has been reported to facilitate the passage of lacrimal probes
during lacerated canaliculi repair.54 For such a procedure,
hyaluronate should be applied to the lacrimal sac through the

intact lacrimal canaliculus and probes for bicanaliculonasal
intubation should be inserted. Hyaluronate is thought to coat
and distend the lumen of the lacrimal passage, allowing the
probe tip to nd its way to the injured canaliculus.54 Sodium
hyaluronate can help the surgeon nd a cut medial canaliculi
when it is injected into the lacrimal sac.55,56
STRABISMUS SURGERY
Sodium hyaluronate has been used in strabismus surgery with
adjustable sutures to minimize tissue drag in the conjunctiva,
Tenons capsule, and muscle.56 Healon has been reported to reduce
postoperative muscle adhesions57 and to increase the period of
suture adjustability in operated muscles.58
DRY-EYE TREATMENT
Sodium hyluronate can both subjectively and objectively improve
dry-eye symptoms.5966 Patients with severe keratoconjunctivitis
sicca respond particularly well to sodium hyluronate.67 Dry-eye
symptoms can also be relieved with topical chondroitin sulfate
solution and viscoelastic articial tears.66 Viscoelastic contact
lens rewetting solutions are also available.68
CONCLUSIONS
Viscoelastic polymers are valuable surgical adjuncts. They maintain anatomic space, manipulate intraocular tissues, and prevent
mechanical trauma to fragile cells such as the corneal endothelium. Viscosurgery may temporarily elevate IOP if the anterior
chamber is not properly irrigated at the end of the procedure.
Comparative data suggest there are no major differences
between the commercially available viscosurgical agents. All
viscoelastics have similar optical clarity, protect tissues, raise
IOP postoperatively and maintain space.28,69,70 When cost is a
concern, methylcellulose preparations should be considered.
Although a number of viscoelastic solutions are available to the
ophthalmic surgeon, no single formulation appears signicantly
more efcacious.
REFERENCES
296
1. Glasser DB, Matsuda M, Edelhauser HF:

A comparison of the efcacy and toxicity of
and intraocular pressure response to
viscous solutions in the anterior chamber.
2. Dick HB, Krummenauer F, Augustin AJ,
et al: Healon5 viscoadaptive formulation:
comparison to Healon and Healon GV.
3. Miller KM, Colvard DM: Randomized
clinical comparison of Healon GV and
Viscoat. J Cataract Refract Surg 1999;
25:1630.
4. McDermott ML, Hazlett LD, Barret RP, et al:
Viscoelastic adherence to corneal
endothelium following phacoemulsication.
5. Blaydes JE, Fritz KJ, Fogle JA: New
techniques of viscosurgery with
phacoemulsication. Am Intraocul Implant
Soc J 1985; 11:395.
6. DeLuise VP: Viscodissection as an adjunct
to phacoemulsication. Ophthalmic Surg
1988; 19:682.
7. Thim K, Krag S, Corydon L: Hydroexpression
and viscoexpression of the nucleus through
a continuous circular capsulorrhexis.
8. Burton RL, Pickering S: Extracapsular

cataract surgery using capsulorrhexis
with viscoexpression via a limbal section.
J Cataract Refract Surg 1995;
21:297.
9. Bellucci R, Morselli S, Pucci V, Bonomi L:
Nucleus viscoexpression compared with
the other techniques of nucleus removal in
extracapsular cataract extraction with
capsulorrhexis. Ophthalmic Surg 1994;
25:432.
10. Schirmer K: Nuclear expression using
viscoelastic versus small incision surgery.
11. Korynta J: Viscoexpression of the lens
nucleus in extracapsular cataract
extraction. Cesk Oftalmol 1996; 52:176.
12. Jeng BH, Hoyt CS, McLeod SD:
Completion rate of continuous curvilinear
capsulorrhexis in pediatric cataract surgery
using different viscoelastic materials.
13. Dick HB, Augustin AJ, Pfeiffer N: Osmolality
of various viscoelastic substances:
comparative study. J Cataract Refract Surg
2000; 26:1242.
14. Hosny M, Eldin SG, Hosny H: Combined
lidocaine 1% and hydroxypropyl
15.
16.
17.
18.
19.
20.
methylcellulose 2.25% as a single

anesthetic/viscoelastic agent in
phacoemulsication. J Cataract Refract
Surg 2002; 28:834.
Bournas P, Condilis N, Lioumi D, et al:
The use of a new viscoelastic substance
combined with anaesthetic in cataract
surgery by phacoemulsication. Ann Ital
Chir 2005; 76:383.
Toczolowski JR: The use of sodium
hyaluronate (Hyalcon) for the removal of
severely subluxated lenses. Ophthalmic
Surg 1987; 18:214.
Mandelcorn M: Viscoelastic dissection for
relocation of off-axis intraocular lens
implant: A new technique. Can J
Glasser DB, Katz HR, Boyd JE, et al:
Protective effects of viscous solutions in
phacoemulsication and traumatic lens
implantation. Arch Ophthalmol 1989;
107:104107.
Kim EK, Cristol SM, Kang SJ, et al:
Viscoelastic protection from endothelial
damage by air bubbles. J Cataract Refract
Surg 2002; 28:1047.
Tan AKK, Humphrey RC: The xed dilated
pupil after cataract surgery is it related to
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
intraocular use of hydromellose? Br J

Easen J, Seward HC: Pupil size and
reactivity following hydroxypropyl
methylcellulose and sodium hyaluronate.
Kusman B, Jaffe NS, Clayman HM,
Jaffe MS: Sodium hyaluronate (Healon)
and intraocular pressure. In: Miller D,
Stegmann R, eds. Healon (sodium
hyaluronate): a guide to its use in ophthalmic
surgery. New York: Wiley; 1983:195.
Larson RS, Lindstrom RL, Skelnik DL:
Viscoelastic agents. CLAO J 1989; 15:151.
Berson FG, Patterson MM, Epstein DL:
Obstruction of aqueous outflow by sodium
hyaluronate in enucleated human eyes.
Reed DB, Mannis MJ, Hills JF, Johnson CA:
Corneal epithelial healing after penetration
keratoplasty using topical Healon versus
balanced salt solution. Ophthalmic Surg
1987; 18:525.
Graether JM: Detachment of Descemets
membrane by injection of sodium
hyaluronate (Healon). J Ocul Ther Surg
1984; 3:178.
Hoover DL, Giangiacomo J, Benson RL:
Descemets membrane detachment by
sodium hyaluronate. Arch Ophthalmol
1985; 103:805.
Ostberg A, Trnquist G: Management of
detachment of Descemets membrane
caused by injection of hyaluronic acid.
McAuliffe KM: Sodium hyaluronate in the
treatment of Descemets membrane
detachment. J Ocul Ther Surg 1982; 1:58.
Donzis PB, Karcioglu ZA, Insler MS:
Sodium hyaluronate (Healon) in the surgical
repair of Descemets membrane
detachment. Ophthalmic Surg 1986; 17:735.
Raitta C, Setl K: Trabeculectomy with the
use of sodium hyaluronate. One-year
follow-up. Acta Ophthalmol 1987; 65:709.
Raitta C, Setl K: Trabeculectomy with the
use of sodium hyaluronate: a prospective
study. Acta Ophthalmol 1987; 64:09.
Merriam JC, Wahlig JB, Konrad H, Zaider M:
Extracapsular cataract extraction and
posterior-lip sclerectomy with viscoelastic.
Alpar JJ: Sodium hyaluronate (Healon) in
glaucoma ltering procedures. Ophthalmic
Surg 1986; 17:724.
Wild GJ, Kent AR, Pengh Q: Dilation of
Schlemms canal in viscocanalostomy: a
comparison of 2 viscoelastic substances.
J Cartaract Refract Surg 2001; 27:1294.
Alpar JJ: The role of 1 percent sodium
hyaluronate in treating vitreous
incarceration with the neodymium: UAG
laser in patients with corneal
decompensation. J Cataract Refract Surg
1986; 12:502.
Alpar JJ: The role of 1 percent sodium
hyaluronate in anterior capsulotomy with
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
the neodymium: YAG laser in patients with

diseased cornea. J Cataract Refract Surg
1986; 12:658.
Baldwin LB, Smith TJ, Hollins JL,
Pearson PA: The use of viscoelastic
substances in the drainage of
postoperative suprachoroidal hemorrhage.
Shin DH, Frenkel RE: The use of
viscoelastic substances in the drainage of
postoperative suprachoroidal hemorrhage.
Gressel MG, Parrish RK, Heuer DK: Delayed
nonexpulsive suprachoroidal hemorrhage.
Frenkel REP, Shin DH: Prevention and
management of delayed suprachoroidal
hemorrhage after ltration surgery. Arch
Ophthalmol 1986; 104:1459.
Pruett RC, Schepens CL, Swan DA:
Hyaluronic acid vitreous substitute: A sixyear clinical evaluation. Arch Ophthalmol
1979; 97:2325.
Stenkula S, Ivert L, Gislason I, et al: The
use of sodium-hyaluronate (Healon) in the
treatment of retinal detachment.
Poole TA, Sudarsky RD: Suprachoroidal
implantation for the treatment of retinal
detachment. Ophthalmology 1986;
93:1408.
Mittl RN, Tiwari R: Suprachoroidal injection
of sodium hyaluronate as an internal
buckling procedure. Ophthalmic Res 1987;
19:255.
Lavin MJ, Leaver PK: Sodium hyaluronate
and giant retinal tears. Arch Ophthalmol
1990; 108:480.
Meredith TA: Giant retinal tears. Arch
Brown GC, Benson WE: Use of sodium
hyaluronate for the repair of giant retinal
tears. Arch Ophthalmol 1989; 107:1246.
Packer AJ, McCuen BW II, Hutton WL,
Ramsay RC: Procoagulant effects of
intraocular sodium hyaluronate (Healon)
after phakic diabetic vitrectomy. A
prospective randomized study.
McLeod D, James CR: Viscodelamination
at the vitreoretinal juncture in severe
diabetic eye disease. Br J Ophthalmol
1988; 72:413.
Stenkula S, Tornquist R: Use of Healon in
vitrectomy and difcult retinal detachments.
In: Miller D, Stegmann R, eds. Healon
(sodium hyaluronate): a guide to its use in
ophthalmic surgery. New York: Wiley;
1983:207.
Hurwitz JJ, Nik N: Lacrimal sac
identication for dacryocystorhinostomy:
The role of sodium hyaluronate. Can J
Lerner HA, Boynton JR: Sodium
hyaluronate (Healon) as an adjunct to
lacrimal surgery. Am J Ophthalmol 1985;
99:365.
54. Vila-Coro AA, Vila-Coro AA: Hyaluronate

facilitates passage of lacrimal probes for
repair of lacerated canaliculi. Arch
55. Seiff SR, Ahn JC: Locating cut medial
canaliculi by direct injection of sodium
hyaluronate into the lacrimal sac.
56. Clorefeine GS, Parker WT: Use of Healon in
eye muscle surgery with adjustable sutures.
57. Searl SS, Metz HS, Lindahl KJ: The use of
sodium hyaluronate as a biologic sleeve in
strabismus surgery. Ann Ophthalmol 1987;
19:259.
58. Manjoney D, Mathias S, Morris W, et al:
Effect of Healon on adjustable suture
strabismus surgery. Invest Ophthalmol Vis
Sci (ARVO Suppl) 1985; 26:80.
59. Polack FM, McNiece MT: The treatment of
dry eyes with Na hyaluronate (Healon).
Cornea 1982; 1:133.
60. DeLuise VP: Viscodissection as an adjunct
to phacoemulsication. Ophthalmic Surg
1988; 19:682.
61. Stuart JC, Linn JG: Dilute sodium
hyaluronate (Healon) in the treatment of
ocular surface disorders. Ann Ophthalmol
1985; 17:190.
62. Nelson JD, Farris RL: Sodium hyaluronate
and polyvinyl alcohol articial tar
preparations. A comparison in patients with
1988; 106:484.
63. Laflamme MY, Swieca R: A comparative
study of two preservative-free tear
substitutes in the management of severe
dry eye. Can J Ophthalmol 1988; 23:174.
64. Bohm E, Rama P, Tallandini L, et al: Low
molecular weight sodium hyaluronate in the
treatment of tear lm changes and of dry
eye. Ophthalmologie 1988; 2:353.
65. Orsoni JG, Chiari M, Guazzi A, et al:
Efcacy of hyaluronic acid eye drops in the
treatment of dry eye. Cytologic study using
an optical microscope and computerized
microscope. Ophthalmologie 1988; 2:355.
66. Limberg MB, McCaa C, Kissling GE, et al:
Topical application of hyaluronic acid and
chondroitin sulfate in the treatment of dry
eyes. Am J Ophthalmol 1987; 103:194.
67. Sand BB, Marner K, Norn MS: Sodium
hyaluronate in the treatment of
keratoconjunctivitis sicca. A double
masked clinical trial. Acta Ophthalmol
1989; 67:181.
68. Physicians Desk Reference
(Ophthalmology) 2005, 59th ed. Thomson
Healthcare, New Jersey.
69. Genstler DE, Keates RH: Amvisc in
extracapsular cataract extraction. Am IntraOcular Implant Soc J 1983; 9:317.
70. McKnight SJ. Giangiacomo J, Adelstein E:
Inflammatory response to viscoelastic
materials. Ophthalmic Surg 1987; 18:804.
CHAPTER 28
Viscoelastics
297
CHAPTER
29
Pharmacologic Agents with Osmotic Effects

Gail Torkildsen, Ula V. Jurkunas, and Tolga Kocaturk
Ocular osmotic forces cause prominent signs and symptoms in

many disease states. The mechanism by which water and electrolytes are linked and transferred between epithelial and endothelial cells remains an unsolved problem. Dysfunction of these
layers is thought to gure prominently in many diverse diseases
such as corneal edema, cataract, glaucoma and some vitreous
and retinal detachments and central serous choroidopathy.
Pharmacologic agents exert osmotic effects within the eye and
should be considered in both treatment of disease as well as
avoiding unintended effects.
OSMOSIS, OSMOLARITY, AND TONICITY

Key Features
Osmotic forces affect the flow of water into or out of tissues,

tear lm, aqueous, and vitreous
Hyperosmotic tear lm can induce inflammatory cytokines
Systemic osmotic agents such as mannitol increase plasma
oncotic pressure and draw water out of the vitreous to lower
intraocular pressure
Intracameral and intravitreal injections must be carefully
prepared due to the powerful osmotic forces they can exert
Osmotic forces may be a factor in cataract development
Diffusion is a constant motion of molecules among each other

which results in a solute or solvent moving from a higher concentration to a region of lower concentration. If there is a membrane between the two regions and that membrane is permeable
to water only, a net movement of water occurs through the membrane. Water will diffuse from an area of high water concentration into the area of low water concentration. Diffusion of
water across the membrane is called osmosis, and it is driven by
the internal energy of water molecules. The net flow of water can
be prevented by the application of an opposing force, osmotic
pressure. This pressure is directly proportional to the concentration of nondiffusible molecules on the opposite side. As a
result, the direction of water flow is determined by the solute
concentration and not by the molecular weight of the solute.
Facilitated diffusion refers to the interaction of a solute interacting with a carrier protein in a cell membrane to aid the solutes
passage. Osmolarity (osmoles per liter of water) is a total concentration of solutes in a solution and is independent of whether
the solutes can cross the membrane. Osmolality (osmoles per
kilogram) is a total number of osmotically active particles in a
solution. Tonicity is the effective osmolality or concentration of
the solutes that have the capacity to exert the osmotic force
across the membrane (Fig. 29.1).
A cell membrane, which is a highly complex semipermeable
membrane, has both outer and inner lipid layers and a middle
H2O
1M
Urea
A. Isotonic
Not Iso osmotic
1M
1 M Sucrose
Urea
B. Not Isotonic
Iso osmotic
H2O
1M
Sucrose
C. Not Isotonic
Not Iso osmotic
FIGURE 29.1. Schematic representation of tonicity and osmolarity.
aqueous layer. A lipid-soluble substance passes through the

lipid-soluble layers with greater ease than a water-soluble substance, whereas the latter transgresses only the middle layer with
comparative ease. Carrier proteins imbedded in a cell membrane
aid the passage of a given substance and facilitate diffusion. The
layers have different permeability to the different substances.
The process of dialysis results when protein is on one side of the
semipermeable membrane. Water moves toward the protein,
and salt flows away from the protein. The nal distribution of
salt and protein is described by the GibbsDonnan equilibrium
in which the product of cations and anions is the same on both
sides of the membrane and the number of cations on the protein
side equals the sum of anions and proteins on the other side.1
Filtration is the process of forcing fluid across a membrane
exerting pressure. Ultraltration results when a hydrostatic force,
such as blood pressure, acts on the solutions that contain protein.
OSMOTIC PHENOMENA IN THE EYE

TEARS
Cellular integrity of air-exposed cells of both the cornea and the
conjunctiva is maintained by the presence of a complex barrier
of isotonic fluid, the tear lm. In response to the external stimuli
and the internal requirements of the cornea and conjunctiva the
regulatory mechanisms of the tear lm alter its composition,
volume, and structure. The classical view of the tear lm as
a three-layered structure, composed of an outer lipid layer, a
middle aqueous layer and an inner mucous layer structure
has been modied. Some authors state that mucous and the
aqueous layers are not distinct, and some suggest that a lipid
layer is a monolayer with polar and nonpolar phases.2 The
regulation of the osmotic flow of fluids between the corneal and
conjunctival epithelial cells and the tear lm is mainly achieved
by aqueous phase electrolytes such as Na+, K+, and Cl that
buffer tear pH and control cell membrane permeability.
The osmolarity of the normal human tear lm is 302 6
(SD) mOsm/L, which is higher than serum osmolarity of
299
SECTION 4

290 mOsm/L.3 Tear osmolarity is the lowest in the morning after
prolonged lid closure, and increases as the day progresses.4,5
The concentrations of chloride and potassium are higher in the
tears than in the serum, and glucose concentration is lower
than tears.2 The importance of tear lm osmolarity and the
stable balance of essential electrolytes on the epithelial surface
maintenance have been shown in animal models and in vivo.58
Elevated tear lm osmolarity of greater than 310 mOsm/L is
often seen in patients with dry-eye syndrome.3,6 The hyperosmolarity of the tear lm indicates an imbalance between the
rate of tear secretion and the rate of evaporation, as seen in
aqueous tear deciency and evaporative tear loss, respectively.3,6
In aqueous tear deciency states, lacrimal gland secretion
rate declines and tear osmolarity increases independent of evaporative loss.6,7 In keratoconjunctivitis sicca, tear replacement
from the lacrimal gland is decreased, often with striking morphologic changes in both the conjunctival and the corneal epithelium. When the tear osmolarity increases, there is an abnormal
discharge of mucin glycoprotein granules and a decrease in goblet
cell density that contribute to the ocular surface pathology of
dry eyes.8
Hyperosmotic conditions can potentiate cytokine synthesis
by nonimmune resident ocular surface cells, including corneal
and conjunctival epithelial and stromal cells. Expression and
production of matrix metalloproteinases (MMP): MMP-9
(gelatinase), MMP-1 and MMP-13 (collagenases) and MMP-3
(stromelysin) are increased with increasing osmolarity. This
increase in mediated in part by the c-Jun N-terminal kinases (JNK)
which is a stress activated protein kinase (SAPK).9 Effectors of
cytokine synthesis in dry eye include mitogen-activated kinases
(MAP kinase, p38 kinase), JNK, and I-k kinase (IKK).
Hypersomotic conditions can induce inflammatory processes
which upregulate several gene products. One of these products
is the nuclear transcription factor, NF-kB. In its quiescent state
NF-kB exists as a heterodimer with the protein Ik-Ba. This
masks the nuclear localization signals and DNA binding
domain of the former protein. Under inflammatory conditions
Ik-Ba is phosphorylated, causing a conformational change
which results in its tagging with multiple copies of the ubiquitin
protein. Ubiquinated Ik-Ba is recognized and degraded by the
proteasome, which liberates NF-kB. The free protein is translocated to the nucleus, where it binds to the appropriate DNA
sequence and upregulates the production of several inflammatory mediators, such as COX-2, iNOS, IL-1, and TNF-a.
CORNEA
300
The sodium concentration gradient is thought to be the predominant force acting on the corneal endothelium. This creates
a net osmotic force drawing water out of the stroma via osmosis
while other ions oppose it. In disease states, the ionic gradients
across the endothelium cannot be maintained resulting in corneal
edema and swelling (Fig. 29.2). Corneal edema is characterized
by a marked increase in corneal thickness, and intercellular and
extracellular edema of the basal epithelial cell layer of the
epithelium. In addition, corneal edema is associated with loss
of stromal proteoglycans and hydropic degeneration or cell lysis
of keratocytes.
Water movement within the epithelium is slowed by the presence of lipid membranes. Zonula occludens or tight junctions
encircle the cells just below the apical surface and constitute an
additional barrier to the passive movement of water, electrolytes
and macromolecules.10 In contrast, water moves rapidly within
the stroma because of the abundance of collagen brils, which
are separated by proteoglycans and water. Although endothelial
cells have junctional complexes, they are much more leaky to
water than epithelium; the result is relative freedom of water
Lipid & aqueous soluble

H2O
Lipid soluble
Isotonic Tear Film
Epithelium
Zonulae
Occludentes
Blocked by lipid
membranes in
epithelium
Cornea
Proteoglycans + H2O
Endothelium
Junctional
Complexes
Aqueous Humor
Active Secretion
Na HCO3-
Na+ Ion Pump
Lens
FIGURE 29.2. Osmotic forces in the anterior chamber.
movement.11,12 Thus, only lipid-soluble substances cross freely

the epithelial and endothelial membranes, and water-soluble
substances pass with equal freedom through the stromal layer.
Substances soluble in both lipid and aqueous penetrate the cornea
more easily. Surfacants like benzalkonium chloride (BAK) may
improve the ocular penetration of a drug in a transscleral drug
delivery system without producing toxic reactions by acting on
tight junctions.13
Corneal transparency is directly related to the corneal hydration. Fluid traverses the endothelium transcellularly in response
to the osmotic gradient created by electrolyte transport and
utilizing the osmotic permeability of aquaporins. Electroosmosis,
whereby a recirculating current causes fluid movement via
paracellular shifts, may be the prominent mechanism of fluid
transport. Trans-endothelial fluid transport can be rapidly
modulated to control stromal hydration in response to small
NaCl osmotic stresses in a way that cushions the shock and
reduces the change in corneal thickness.14
When the endothelial cell density decreases below a critical
level (200400 cells/mm2), the leak rate of fluid into the stroma
becomes greater than the pump rate of fluid out of the stroma,
producing corneal edema and clouding.15 The blurred vision
from the epithelial edema in the mornings is due to the lack of
tear evaporation under the closed lids. After opening the eyelids,
the evaporation causes transient hypertonicity of the tear lm
which extracts the water from the epithelial cells and aids in
clearing the vision.15
OSMOTIC AGENTS
TOPICAL
Pathologic changes in dry eyes produce hyperosmolar tear lm
that draws the water from the corneal epithelial cells, reduces
microplicae, disrupts cell membranes and decreases cell vitality.3,8,16 The main aims of dry-eye treatment with topical agents
are tear supplementation and conservation. To counterbalance
the hyperosmotic environment of dry-eye conditions, tear substitutes have been developed that dilute and decrease the osmo-

re-equilibrates and can lead to rebound increases in IOP. Osmotic
agents cause a total body diuresis and should not be used in
cardiac and renal patients. Side effects can include headache,
backache, diabetic ketoacidosis, congestive heart failure, and
myocardial infarction due to increased preload on heart. Central
nervous system effects can include confusion and subdural and
subarachnoid hemorrhages.
Osmotic agents include mannitol, glycerin, urea and isosorbide.
IV Mannitol is the most commonly used systemic drug in this
class. Mannitol is not metabolized and the dosage is 1.52 g/kg
body weight over 3045 min. Glycerin can be given orally but is
rapidly metabolized to glucose and should be used cautiously in
diabetics. Dosage is 11.5 g/kg body weight. Isosorbide is available in a 45% oral preparation, and is physiologically similar to
glycerin. It is essentially not metabolized and is excreted by the
kidney. Dosage is 1.5 g/kg body weight.
INTRAOCULAR IRRIGANTS
Irrigating solutions with the corneal endothelium, lens,
trabecular meshwork, vitreous and retina may have important
consequences for cellular survivability and function. An
irrigating solution must maintain both physiologic and anatomic
integrity. An ideal irrigating solution is isoosmotic with intraocular fluids and contains the nutrients necessary for cellular
viability. Currently available intraocular irrigants have osmolarity
of 277305 mOsm.22 The major ions present in the solutions
are sodium, potassium, magnesium, calcium, and bicarbonate.
Some solutions contain dextrose and reduced glutathione (GSH)
and/or oxidized glutathione (GSSG). Addition of GSH and
GSSG to the irrigating solutions showed a benecial effect in
preventing corneal swelling by maintaining intracellular levels of
GSH in corneal endothelium.23 GSH is a powerful antioxidant
effective in detoxifying the free radicals released during
intraocular surgery.24 In particular, GSSG an ingredient of BSS
plus (Alcon Laboratories, Fort Worth, TX, USA), was shown to
be benecial on the maintenance of the barrier function of
corneal endothelium, retinal pigment epithelium, and the
bloodaqueous barrier.25,26
The pH and osmotic tolerance range of the human corneal
endothelium are important considerations when combining
intraocular medications and ophthalmic solutions. The corneal
endothelium has a pH tolerance between 6.8 and 8.2, similar to
the natural aqueous humor bicarbonate buffer system.27 During
phacoemulsication, the osmolality of the anterior chamber can
vary due to medications, viscoelastics, and solutions. Hyperosmolarity or hypoosmolarity can cause the endothelial cells to
swell, degenerate, become apoptotic, or necrotic. The corneal
endothelial cells have been shown to tolerate a wide range of
osmolalities from 250 to 350 mOsmoles.28 Therefore, both the
pH and osmolality of the intraocular solution are critical in
maintaining the corneal endothelium.29
SYSTEMIC AGENTS FOR THE REDUCTION

OF INTRAOCULAR PRESSURE
OSMOTIC FORCES ON THE LENS
Acute treatment of ocular hypertension and preparation of the

eye for intraocular surgery are the two prominent therapeutic
indications for systemic delivery of osmotic agents. Osmotic
agents cause rapid reductions in intraocular pressure by increasing blood osmolality which draws fluid from vitreous to blood
thus decreasing vitreous volume and decreasing IOP. In angle
closure glaucoma, the decreased IOP reverses iris ischemia and
improves its responsiveness to pilocarpine and other drugs. If
the blood aqueous barrier is disrupted, osmotic agents can enter
eye and are less effective at decreasing IOP. These medications
may be more effective during inflammation. Osmotic agents
can not be used long term as the osmotic gradient quickly
Human lens has a requirement for the maintenance of an elaborate antioxidant system, failure of which has been associated
with cataract formation. A constant supply of glucose from
aqueous humor serves as a main source of energy for the anaerobic glycolysis in the lens.30 In diabetic patients, posterior
subcapsular cataract formation has been associated with prolonged irrigation during intraocular surgery. Some surgeons
advocate adding supplemental glucose to the intraocular irrigants
to prevent the cataract formation in the diabetic patients undergoing vitrectomy. The addition of glucose raises the osmolority
from 305 to 320 mOsm, a level consistent with the diabetic
patients aqueous humor osmolarity.
CHAPTER 29
larity of the tear lm and restore normal tear physiology. Most

tear substitutes are isotonic with natural tear lm, and some are
hypotonic.
The reports on the utility of hypotonic versus isotonic tear
substitutes in treatment of dry eyes have been contradictory.
Initial studies by Gilbard et al noted that electrolyte solution with
osmolarity of 175 mOsm/L (TheraTears) effectively decreased
tear osmolarity, increased goblet cell density and improved dryeye symptoms.17 Other studies have shown that both isotonic
and hypotonic solutions were equally effective in the relief of
dry-eye symptoms.18,19 The authors postulate that the effect of
hypotonic tear substitutes on the corneal surface is of short
duration of action, and is achieved by isotonic preparations just
as well.19 Even though increased tear osmolarity is present in
dry-eye patients, the focus of tear substitute design should not
be on the tonicity, but rather on tear replacement retention,
mucomimetic action, secondary effects of preservatives,
lubricating properties, and nally actual comfort.
Various polymers are added to the tear substitutes to enhance
tear retention by increasing the viscosity, decreasing surface
tension and enhancing tear lm stability. Increasing viscosity
with the addition of polymeric ingredients causes a longer interval of contact with the eye. Sodium hyaluronate, a constituent
of extracellular matrix has been shown to have clear benet in
promoting corneal epithelial healing and relief of dry-eye symptoms.19 A therapeutic soft contact lens, with frequent instillation of saline or another tear substitute, also prolongs contact of
the tear solution. Ointments are useful when frequent instillation is not possible.
In contrast to dry-eye treatment, conditions that cause corneal
edema are treated by hyperosmotic agents. They transiently
increase the tonicity of the tear lm and enhance water movement from the cornea, especially the epithelial cell layer. Most
frequently used agents in a clinical setting are sodium chloride
2% and 5% solution and ointment (Muro-128) and glycerin
(50100% preparations). Sodium chloride is most commonly
used in cases of corneal edema due to endothelial dysfunction,
post-LASIK corneal flap edema, and to acute corneal hydrops in
keratoconus.20,21 Sodium chloride drops are particularly benecial in reducing epithelial edema upon awakening. Hypertonic
sodium chloride ointment at bedtime reduces the amount of
corneal hydration while the eyelids are closed during sleep. Intact
epithelium provides a barrier to solute movement and enhances
the osmotic effect of the hypertonic solutions. Ocular irritation
is a common side effect of hypertonic saline eyedrops.
Glycerin is a fast acting osmotic agent when in contact with
the corneal surface. The effects of glycerin are transient as the
mixture with water decreases the solutions effective osmolarity.
The main clinical use of glycerin is in corneal edema due to
acute angle glaucoma, or endothethelial dysfunction. In the former,
the application of glycerin aids in gonioscopic examination.
301
SECTION 4

Osmotic stress due to the accumulation of sorbitol in the lens
is most likely the cause of diabetic cataract. Sorbitol accumulates
in the lenses of diabetic animals and the administration of an
inhibitor to aldose reductase (AR), the enzyme that converts
glucose to sorbitol, prevents the formation of diabetic cataracts.
Sorbital along with myo-inositol (MI) and taurine are the major
osmolytes in the lens. For lens epithelial cells, an increase in
extracellular osmotic pressure induces the expression of a Na+dependent MI transporter (SMIT), AR, and taurine transporter.
Consequently, intracellular levels of MI, sorbitol, and taurine
are increased to balance the increased osmotic pressure.
Overexpression of SMIT in the lens causes congenital cataract.28
Transporter proteins in the cell wall play a role in how ions
move among cells. One of these is the potassium chloride
cotransporter (KCC) which is involved in the regulation of lens
volume and transparency. Under normal isotonic conditions,
a constitutively active flux of Cl ions exists in the lens that
regulates ber cell volume. Under certain conditions, KCC
activity can be increased, not only through dephosphorylation
of the protein, but also by increasing the number of transporters
in the plasma membrane.35
Electrical current flow around the lens may play a role in lens
transparency. A recirculating sodium gradient may drive fluid
into the lens anteriorly and fluid may exit posteriorly. Taking
into account the known presence of membrane channels, transporters, and an aquaporin in lens epithelium, there may exist a
classical epithelial fluid transport mechanism in this layer
which may be of great importance for lens homeostasis.31
OSMOTIC FORCES IN AQUEOUS

PRODUCTION
Aqueous humor formation depends on hydrostatic pressure and
the oncotic pressure gradient across the ciliary epithelium.
Numerous ion channels and ports have been characterized in
the ciliary epithelium contributing to aqueous formation. Sodium,
choride, and potassium are actively transported from plasma in
the ciliary body stroma into the pigmented ciliary epithelial
cells by a Na+/K+/2Cl exchanger (symport).32 The pigmented
and nonpigmented epithelial cells are united by electric and
metabolic coupling. From the nonpigmented epithelial cells,

Na+, Cl, and bicarbonate ions are pumped into the clefts
between nonpigmented cells creating an osmotic gradient
which draws water into the clefts. Tight junctions at the apical
side direct fluid into the posterior chamber.33
POSTERIOR POLE
During vitrectomy the irrigating solutions keep the globe
inflated and serve as a vitreous substitute. Studies have shown
that bicarbonate and glucose are especially important in maintaining normal retinal cell metabolic activity.22 Most additives,
such as antibiotics and epinephrine may decrease the pH of the
solution and cause retinal toxicity.22 The recommendations are
ones of caution when requesting the additions to the intraocular
irrigants, as their efcacy and safety have not been fully
established.
Osmotic forces probably play a role in neuronal degeneration
in the detached retina. Retinal detachment causes a decrease of
the plasma membrane K+ conductance of Mller cells. The
decrease of the K+ currents is associated with a decrease in the
gene and protein expression for the main K+ channel subtype of
Mller cells, Kir4.1. Downregulation of the Kir4.1 protein may
cause an altered current pattern in Mller cells. Impaired spatial
buffering of K+ ions (normally performed by Mller cells by
means of their Kir channels) may contribute to neuronal degeneration in the detached retina, by favoring neuronal hyperexcitation and glutamate toxicity. In the postischemic retina of
the rat, it has been shown that the decrease in K+ currents is
associated with altered osmotic swelling characteristics of Mller
cells, which may contribute to edema development in the retina.
By formation of glial scars and cellular hypertrophy, reactive
Mller glial cells may inhibit regular neuroregeneration in the
detached and reattached retina.34
Investigation of the osmotic phenomenon within the eye
remains an active area of research. Many important ophthalmic
disease states involve imbalances of osmotic forces. Medications may exert osmotic effects, impacting disease states and
understanding osmotic principles may allow more targeted
therapy.
REFERENCES
302
1. Gilbard JP, Rossi SR, Gray Heyda K:

Ophthalmic solutions, the ocular surface,
and a unique therapeutic articial tear
formulation. Am J Ophthalmol 1989;
107:348355.
2. Hodges RR, Darlene D: Keratoconjunctivitis
sicca: physiology and biochemistry of the
tear lm. In: Foster CS, Azar DT, Dohlman
CH, eds. Smolin and Thofts the cornea.
Philadelphia, PA: Lippincott Williams &
Wilkins; 2005:577602.
3. Gilbard JP, Farris RL, Santamaria J II:
Osmolarity of tear microvolumes in
1978; 96:677681.
4. Terry JE, Hill RM: Human osmotic tear
pressure: Diurnal variations and the closed
eye. Arch Ophthalmol 1978; 96:120.
5. Farris RL, Stuchell RN, Mandel ID: Tear
osmolarity variation in the dry eye: Trans
6. Gilbard JP: Human tear lm electrolyte
concentrations in health and dry-eye
disease. Int Ophthalmol Clin
1994; 34:27.
7. Gilbard JP, Dartt DA: Changes in rabbit

lacrimal gland fluid osmolarity with flow rate.
8. Huang AJ, Belldegrun R, Hanninen N, et al:
Effect of hypertonic solutions on
conjunctival epithelium and mucinlike
glycoprotein discharge. Cornea 1989; 8:15.
9. De-Quan Li, Zhuo Chen, Xiu Jun Song, et al:
Stimulation of matrix metalloproteinases by
hyperosmolarity via a JNK pathway in
human corneal epithelial cells. Invest
10. Sugrue S, Zieske J: ZO1 in corneal
epithelium: association to the zonula
occludens and adherens junctions Exp Eye
Res 1997; 64:1120.
11. Cogan DG, Kinsey VE: Hydration properties
of the whole cornea. Arch Ophthalmol
1942; 28:449.
12. Maurice DM, Giardini AA: Swelling of the
cornea in vivo after the destruction of its
limiting layers. Br J Ophthalmol 1951;
35:791.
13. Okabe K, Kimura H, Okabe J, et al: Effect
of benzalkonium chloride on transscleral
14.
15.
16.
17.
18.
19.
drug delivery. Invest Ophthalmol Vis Sci

2005; 46:703708.
Ruberti JW, Klyce SD: NaCl osmotic
perturbation can modulate hydration control
in rabbit cornea. Exp Eye Res 2003;
76:349359.
Klyce SD: Corneal physiology. In: Foster
CS, Azar DT, Dohlman CH, eds. Smolin and
Thofts the cornea. Philadelphia, PA:
Lippincott Williams & Wilkins; 2005:3858.
Katsuyama I, Arakawa T: A convenient
rabbit model of ocular epithelium damage
induced by osmotic dehydration. J Ocul
Gilbard JP, Rossi SR: An electrolyte-based
solution that increases corneal glycogen
and conjuntival goblet-cell density in a
rabbit model for keratoconjunctivitis sicca.
Wright P, Cooper M, Gilvarry AM: Effect of
osmolarity of articial tear drops on relief of
dry eye symptoms: BJ6 and beyond. Br J
Ophthalmol 1987; 71:161164.
Papa V, Aragona P, Russo S, et al:
Comparison of hypotonic and isotonic
21.
22.
23.
24.
25. Araie M, Shirasawa E, Ohashi T: Intraocular

irrigating solutions and permeability of the
blood-aqueous barrier. Arch Ophthalmol
1990; 8:882885.
26. Araie M, Kimura M: Intraocular irrigating
solutions and barrier function of retinal
pigment epithelium. Br J Ophthalmol 1997;
81:150153.
27. Gonnering R, Edelhauser HF, Van Horn DL,
et al: The pH tolerance of the rabbit and
human corneal endothelium. Invest
Ophthalmol Vis Sci 1979; 18:373390
28. Edelhauser HF, Hanneken AM, Pederson
HJ, Van Horn DL: Osmotic tolerance of
rabbit and human corneal endothelium.
Arch Ophthalmol 1981; 99:12811287
29. Edelhauser HF: The Balance between
Corneal Transparency and Edema The
Proctor Lecture. Invest Ophthalmol Vis Sci
2006; 47:17551767.
30. Christiansen JM, Kollarits CR, Fukui H,
et al: Intraocular irrigating solutions
and lens clarity. Am J Ophthalmol 1976;
82:594.
31. Jorge Fischbarg, Friedrich PJD, Kunyan

Kuang, et al: Transport of fluid by lens
epithelium. Am J Physiol Cell Physiol 1999;
276:C548C557.
32. Edelman JL, Sachs G, Adorante JS: Ion
transport asymmetry and functional coupling
in bovine pigmented and nonpigmented
ciliary epithelial cells. Am J Physiol 1994;
266:C1210.
33. Green K, Bountra C, Georgiou C, House R:
An electrophysiological study of rabbit
ciliary epithelium. Invest Ophthalmol Vis Sci
1985; 26:371.
34. Iandiev I, Uckermann O, Pannicke T, et al:
Glial cell reactivity in a porcine model of
retinal detachment. Invest Ophthalmol Vis
Sci 2006; 47:21612171
35. Kaa-Sandra NC, Joerg K, Paul JD: Roles
for KCC transporters in the maintenance of
lens transparency. Invest Ophthalmol Vis
Sci 2006; 47:673682.
36. Dohlman CH, Hedbys BO, Mishima S: The
swelling pressure of the corneal stroma.
CHAPTER 29
20.
solutions containing sodium hyaluronate on

the symptomatic treatment of dry eye
patients. Ophthalmologica 2001;
215:124127.
Tufts SJ, Gregory WM, Buckely RJ: Acute
corneal hydrops in keratoconus.
Ophthalmology 1994; 101:17381744.
Loh RS, Hardten DR: Noninflammatory flap
edema after laser in situ keratomileusis
associated with asymmetrical preoperative
corneal pachymetry. J Cataract Refract
Surg 2005; 31:922929.
Edelhauser HF, Amass R, Lampert R:
Intraocular irrigating solutions. In:
Zimmerman et al, eds. Textbook of ocular
pharmacology. Philadelphia, PA: LippincottRaven; 1997:635654.
Araie M, Shirasawa E, Hikita M: Effect of
oxidized glutathione on the barrier function
of the corneal endothelium. Invest
Ganea E, Harding JJ: Glutathione-related
enzymes and the eye. Curr Eye Res 2006;
31:111.
303
CHAPTER
30
Pharmacologic Treatment of Immune Disorders

and Specically of Immune Ocular Inflammatory
Disease
C. Stephen Foster
Overview
Immunomodulatory therapy (IMT) is playing an increasingly

important role in the care of patients with ocular inflammatory
disease (OID)
Outcomes analysis in evidence-based medicine reviews
supports the notion that IMT represents the standard of care
for certain specific disorders and for most if not all cases of
steroid-dependent chronic OID
Ophthalmologists would be well advised to acquaint
themselves with these studies and with the
recommendations of the International Uveitis Study Group
and of the American Uveitis Society
TABLE 30.1. Agents Commonly Used to Treat Autoimmune

Inflammatory Conditions
Class
Type of Agent
Nonproprietary Names
Alkylating agents
Nitrogen mustards
Cyclophosphamide
Chlorambucil
Antimetabolites
Natural products
Folic acid analogs
Methotrexate
Pyrimidine analogs
5-Fluorouracil
Purine analogs
Azathioprine
Antibiotics
Cyclosporine
Dapsone
Tacrolimus
In its broadest scope, the rubric immune disorders would include

all disorders in which the immune system is abnormal. A
treatise on the pharmacologic treatment of such immune disorders would necessarily include material devoted to the treatment
of immunodeciency diseases, including acquired immunodeciency syndrome (AIDS) caused by the human immunodeciency virus (HIV), as well as material on immunoregulatory
disorders that result in autoimmunity or an overaggressive
immune response. The authors charge for this chapter is to
address the latter group of disorders. Because inflammation is
the paradigm for the expression of autoimmune disease, a discussion of all therapies for inflammation might be appropriate
here, but the pharmacology and use of the steroidal and nonsteroidal antiinflammatory drugs are dealt with in Chapters 25
and 27. This chapter therefore limits its discussion to the properties and uses of the immunosuppressive chemotherapeutic agents
in the treatment of immune inflammatory or autoimmune
diseases.
Although the use of immunosuppressive and biologic agents
to inhibit immune reactions dates back at least half a century,1
the mechanisms of action of most of the immunosuppressive
agents are incompletely understood. Often we do not even
know whether a particular agent is in fact suppressing immune
responses or suppressing the inflammatory expression of these
responses. By denition, immunosuppressive agents suppress
the development of at least one type of immune reaction: They
modify the specic immune sensitization of lymphoid cells.2
Table 30.1 lists chemotherapeutic agents useful in the treatment of neoplastic disease, many of which are also commonly
used to treat autoimmune inflammatory diseases. Usually only
one, or at most two, agents from a given class of these chemotherapeutic agents has been used extensively enough as an immunosuppressive agent in the treatment of immune disorders to
allow us to make wise choices about using such agents to treat
autoimmune inflammatory disease. This is why only one or
Mitomycin
Antibodies
Antilymphocyte serum
Anti-T-cell antibody
Gamma globulin
two agents are usually chosen to represent each class of chemotherapeutic agent in the following sections.
One feature common to many of the immunosuppressive
agents is their ability to interfere with synthesis of nucleic acid,
protein, or both. This interference commonly is assumed to be
the immunosuppressive mechanism, because lymphoid cells
stimulated by antigen to proliferate and produce lymphokines are
exquisitely sensitive to interference with nucleic acid or protein
synthesis. Bach2 and others have emphasized, however, that the
effect of immunosuppressive agents cannot be explained solely
by this simple notion. Considering the extraordinary complexity
of the idiotypicantiidiotypic immunoregulatory network of
T-lymphocyte subsets, B-lymphocyte subsets, and antigenpresenting cells and macrophage subsets, it is remarkable that the
rst physicians to explore the possible use of immunosuppressive chemotherapeutic agents in the treatment of autoimmune
inflammatory disorders discovered dosages that produced enough
differential effect on subsets of helper and cytotoxic cells to cause
immunosuppression.
ALKYLATING AGENTS
CHEMICAL PROPERTIES AND MECHANISM
OF ACTION
Nitrogen mustards, ethylenimines and methylmelamines, alkylsulfonates, nitrosoureas, and triazenes all act in similar ways,
305

almost completely metabolized. It is the slowest-acting nitrogen
mustard in clinical use, and its cytotoxic effects on bone marrow,
lymphoid organs, and epithelial tissues are similar to those of
the other nitrogen mustards.
SECTION 4
FIGURE 30.1. Chemical structure of cyclophosphamide.
306
through nucleophilic substitution reactions. Of these agents, only

members of the nitrogen mustard family are commonly used as
immunosuppressive chemotherapeutic agents in the treatment of
autoimmune inflammatory disease; of the nitrogen mustards,
only cyclophosphamide and chlorambucil have been used enough
to warrant discussion here.
Cyclophosphamide (Cytoxan), the most potent of the therapeutic alkylating agents, is used extensively throughout the
world to treat a variety of conditions (Fig. 30.1). All alkylating
agents act through nucleophilic substitution reactions, and such
reactions with DNA probably account for their predominant
immunosuppressive activity (Fig. 30.2). Breaks occur in singlestranded DNA. When these breaks are repaired, phosphodiester
bonds form and result in defective cell function. Cross-linking
reactions occur between DNA strands, between DNA and RNA,
and between these molecules and cell proteins, generally resulting in death of the affected cell.
Like most other immunosuppressive agents, cyclophosphamide is not immunosuppressive in its native state. After
oral or intravenous administration, it is activated by the liver
P-450 microsome system. Phosphoamidase, which is present in
especially high concentrations in liver microsomes, catalyzes
the conversion of the drug into its active principles, aldophosphamide and 4-hydroxycyclophosphamide. In clinical doses,
alkylating drugs are very cytotoxic for lymphoid cells. The effect
on B and T cells appears to be nearly equal, except that large
doses enhance the effect on B cells. Cyclophosphamide has a
potent effect on antibody responses when given with, or even up
to 4 days after, antigen encounter. It suppresses secondary antibody responses in previously primed animals and patients.
Cyclophosphamide effectively inhibits delayed hypersensitivity
reactions and is as effective as azathioprine in liver, cardiac,
bone marrow, skin, and pulmonary allograft rejection reactions.
It is the only immunosuppressive agent that can induce immune
tolerance to particulate antigen. The pharmacokinetics and
kinetics of the development of such tolerance are complex. The
drug must be given 2448 h after antigen priming. Tolerance is
probably mediated, at least predominantly, by regulatory T
lymphocytes that develop after antigen priming. On the other
hand, at least in the murine experimental model, low-dose cyclophosphamide therapy can eliminate regulatory T lymphocytes
that actively mediate tolerance, resulting in release from tolerance and in expression of immunoreactivity in the form of a
delayed hypersensitivity reaction to the relevant antigen. The dose
and timing of administration of cyclophosphamide apparently
are critical to its effect on lymphocyte subsets. This, of course,
makes judgments about clinical use of the drug in new applications difcult. Cyclophosphamide inhibits monocyte precursor
development but has little effect on fully developed macrophages.
It is spectacularly effective in preventing the development of autoimmune disease in the NZB/NZW F1 mouse model of systemic
lupus erythematosus. Cyclophosphamide is readily absorbed after
oral administration. The standard initial daily dose is 12 mg/kg.
The serum half-life is 7 h, and allopurinol prolongs that half-life.
Chlorambucil (Leukeran) (Fig. 30.3) is also readily absorbed
after oral administration. The standard initial daily dose is
0.10.2 mg/kg. The half-life in plasma is ~1 h, and the drug is
NONOPHTHALMIC USES AND POTENTIAL

SIDE EFFECTS
Cyclophosphamide is used extensively to treat Wegeners granulomatosis, polyarteritis nodosa, and other forms of systemic
vasculitis. It is still sometimes used to treat human allograft
recipients and often to treat bullous pemphigoid. It is sometimes used when severe rheumatoid arthritis is refractory to
more conventional therapy, and it is a common drug of choice
for nephrotic syndrome in children. It is also still sometimes
employed in the polydrug approach to malignancies, including
multiple myeloma; chronic lymphocytic leukemia; lung, breast,
cervical, and ovarian carcinoma; neuroblastoma; retinoblastoma;
and some other neoplasms of childhood.
Potential complications of cyclophosphamide therapy include
severe bone marrow depression with resultant anemia, leukopenia, thrombocytopenia, and secondary infection; anorexia,
nausea, vomiting, hemorrhagic colitis, and oral mucosal ulceration; jaundice; hemorrhagic cystitis; gonadal suppression;
alopecia; and interstitial pulmonary brosis. Sterile hemorrhagic cystitis occurs in 510% of patients; this has been attributed to chemical irritation of the lining of the bladder
produced by reactive metabolites of cyclophosphamide, particularly acrolein. This potentially devastating complication, which
can lead to bladder carcinoma, can usually be avoided with
correct administration (i.e., restricting consumption of cyclophosphamide to the early hours of the day and forcing fluid
intake during the remainder of the day). Acetylcysteine or mesna
(sodium 2-mercaptoethanesulfonate) can prevent or reverse
cyclophosphamide-induced hemorrhagic cystitis. If a patient
taking cyclophosphamide develops dysuria or microscopic hematuria, the physician should conrm that the patient is taking
the drug correctly and is adequately hydrated and should perform emergency cystoscopy to conrm that the source of the
blood is the lining of the bladder rather than the kidney. If, for
example, a patient being treated for Wegeners granulomatosis
develops microscopic hematuria, cessation of cyclophosphamide
would be inappropriate if the red blood cells are coming from
Wegeners inflammatory activity in the kidney rather than from
cyclophosphamide-induced cystitis.
Chlorambucil is still the treatment of choice for chronic lymphocytic leukemia and primary (Waldenstrms) macroglobulinemia. It is also sometimes used to treat Hodgkins disease and
other lymphomas as well as vasculitis associated with rheumatoid arthritis and autoimmune hemolytic anemia with cold
agglutinins.
Potential complications of chlorambucil therapy include bone
marrow suppression, gastrointestinal discomfort, azoospermia,
amenorrhea, pulmonary brosis, seizures, dermatitis, and
hepatotoxicity. A marked increase in the incidence of leukemia,
lymphoma, and other neoplasms has been reported among
patients receiving long-term adjuvant chemotherapy for breast
cancer and patients being treated for polycythemia vera.
Any patient who requires systemic immunosuppressive chemotherapeutic agents for an ocular inflammatory disease (OID)
must be managed by an experienced chemotherapist who is, by
virtue of formal training and experience, an expert in the use of
immunosuppressive drugs and in the recognition and treatment
of drug-induced side effects and potentially serious complications.
CHAPTER 30
Pharmacologic Treatment of Immune Disorders and Specically of Immune Ocular Inflammatory Disease
FIGURE 30.2. Diagrammatic representation of the mechanism of action of alkylating agents.
FIGURE 30.3. Structural formula of chlorambucil.
The present authors experience suggests that, in general, the

chemotherapy experts with whom ophthalmologists can most
consistently and effectively collaborate are oncologists or hematologists. The chemotherapist is completely responsible for the
chemotherapeutic aspects of the patients care. He or she personally sees the patient regularly; monitoring blood counts and
blood chemistry without seeing the patient is inappropriate
management. The ophthalmologist apprises the chemotherapist
307
SECTION 4
308
regularly of the status of the ophthalmic inflammatory condition.

If the problem is not sufciently controlled, it is the chemotherapist who decides, for instance, whether or not it is safe and
appropriate to increase the patients immunosuppressive medications, to add a second medication with or without stopping the
initial one, or to supplement medications with systemic steroids.
Foster and associates published guidelines suggest initial doses
of various agents and one routine for careful hematologic monitoring, avoiding depressing the white count below 3500 cells/mm3
and the neutrophil count below 1500 cells/mm3.3 Foster also
suggests avoiding thrombocytopenia below 75 000 platelets/mm3,
including urinalysis every 2 weeks during the initial treatment
period, and then once a month when the patient is on a steady
maintenance drug program.
Cyclophosphamide is the treatment of choice for any patient
with ocular manifestations of Wegeners granulomatosis or
polyarteritis nodosa. It is also unquestionably the most effective
treatment for patients with highly destructive forms of inflammation in association with rheumatoid arthritis. Few other drugs
have allowed us to intervene successfully in the progression of
rheumatoid arthritis-associated necrotizing scleritis with associated peripheral ulcerative keratitis. Interestingly, Watson and
Hazleman4 nd that the necrotizing scleritis and peripheral
ulcerative keratitis in some patients with relapsing polychondritis may be more refractory to therapy than that associated
with Wegeners granulomatosis, polyarteritis nodosa, or rheumatoid arthritis. Although dapsone is commonly effective in the
extraocular manifestations of this disease, the author has rarely
found it effective in abrogating ocular inflammation in this disorder. Cyclophosphamide, with or without oral steroid and nonsteroidal antiinflammatory drug therapy, is often required to
treat necrotizing scleritis associated with relapsing polychondritis.
Either cyclophosphamide or chlorambucil is an appropriate
choice for effective treatment of other OID, including
posterior uveitis or retinal vasculitis manifestations of
AdamantiadesBehets disease. Chlorambucil may be the more
effective of the two, but cyclophosphamide, particularly when
given as intravenous pulse therapy, is highly effective. Baer and
Foster,5 and others6 nd both drugs to be superior to cyclosporine
(cyclosporin A, CsA) in the care of patients with posterior segment manifestations of AdamantiadesBehets disease.
Cicatricial pemphigoid affecting the conjunctiva usually
responds to cyclophosphamide therapy. If the patient with cicatricial pemphigoid has very active disease that is progressive,
cyclophosphamide is the drug of rst choice. Therapy typically
lasts at least 1 year. The relapse rate after discontinuation of
cyclophosphamide is ~20%.7
The use of cyclophosphamide or chlorambucil in the treatment of patients with other OID is slightly more problematic.
There is little question that each can be effective in the care of
youngsters with juvenile idiopathic arthritis (JIA)-associated
iridocyclitis that does not respond to steroids and other conventional treatments, and that in this role these drugs can be sightsaving. This is a complex area, however, given the age of the
patients and the potential risks for delayed malignancy or sterility
associated with the treatment. The relative risks and benets
must be explored individually with patient and parents alike.
The author hopes that longitudinal comparative trials in this
patient group will help clarify the issue of relative risks and
benets of systemic immunosuppressive chemotherapeutic treatment early in the course of chronic iritis associated with JIA.
Other forms of uveitis that do not respond to conventional
treatment or are associated with intolerable steroid-induced side
effects may also respond to cyclophosphamide or chlorambucil
therapy. The guidelines for such an approach vary from clinic to
clinic around the world, but ample precedents exist for this
alternative in patients with slowly blinding uveitis.812 Whether
the patient has pars planitis or uveitis associated with Reiters

syndrome, with ankylosing spondylitis, with inflammatory bowel
disease, or even with idiopathic uveitis, the author employs a
stepladder approach to the treatment of that patients ocular
inflammation, always using steroids rst, and aggressively, via
all potential routes of administration (topical, periocular injection, intraocular, systemic) and in the largest doses tolerated. It
is typical to obtain informed consent and dispense printed
handouts that describe the potential risks of topical, periocular,
and systemic steroids. If, in spite of this approach, the patients
disease is chronic or relapses each time steroids are tapered or
discontinued, the author adds oral nonsteroidal antiinflammatory drugs to the treatment plan (with the patients consent).
If this combination does not achieve the goal of total quiescence
of all inflammation off all steroids, or if treatment-induced side
effects appear that are unacceptable to patient or doctor, the
patient is offered the alternative of immunomodulatory therapy
(IMT) with a systemic immunosuppressive chemotherapeutic
drug. The choice of that drug depends on the individual patient,
the particular disease, the patients age, and the patients sex.
Some of the entities the author has treated successfully with
systemic immunosuppressive chemotherapeutic agents, including cyclophosphamide and chlorambucil, are as follows: sympathetic ophthalmia; Vogt-Koyanagi-Harada syndrome; birdshot
retinochoroidopathy; multifocal choroiditis with panuveitis;
retinal vasculitis associated with systemic lupus erythematosus;
multifocal choroiditis associated with progressive systemic sclerosis; retinal vasculitis associated with sarcoidosis; pars planitis
associated with multiple sclerosis; severe uveitis associated with
ankylosing spondylitis, with Reiters syndrome, or with inflammatory bowel disease; idiopathic uveitis; and bilateral Moorens
ulcer,13 cicatricial pemphigoid; scleritis associated with relapsing
polychondritis with polyarteritis, with Wegener granulomatosis
and with rheumatoid arthritis. One series reported recently was
comprised of 28 patients with uveitis, 10 of them with JIAassociated uveitis who had failed lesser immunomodulatory
strategies. Sixty-eight percent of the patients were able to
discontinue corticosteroid therapy with uveitis relapse, and 50%
had induction of drug-free durable remission.14
PURINE ANALOGS
OF ACTION
Thiopurines, such as mercaptopurine and azathioprine (Imuran)
(Fig. 30.4), interfere with purine metabolism and, so, with
synthesis of DNA, RNA, and protein. Purine analogs interfere
with the synthesis of purine bases. They inhibit purine nucleotide
interconversion reactions and the formation and function of
coenzymes (such as coenzyme A), thereby inhibiting RNA and
DNA synthesis. These agents or their metabolites are incorporated into DNA and RNA, but that probably is not the locus of
their suppressive effect. These drugs must be converted to active
FIGURE 30.4. Structural formula of azathioprine.

SIDE EFFECTS
Purine analogs, most notably azathioprine, are used extensively
in human heart, kidney, and lung allograft recipients. They have
also been used to treat blistering dermatoses (pemphigus vulgaris
and bullous pemphigoid), rheumatoid arthritis, and regional
ileitis (Crohns disease).
The author has suggested an initial dose of 23 mg kg1 day1;
dose adjustments are based on clinical response and drug
tolerance. Allopurinol inhibits xanthine oxidase and so inhibits
the conversion of azathioprine to its inactive metabolites; the
dose must be reduced accordingly.
Potential drug-induced complications of azathioprine therapy
include hepatotoxicity, severe bone marrow depression with
resultant anemia, leukopenia, thrombocytopenia, secondary infection, anorexia, nausea, vomiting, gastrointestinal distress,
diarrhea, rash, fever, and arthralgia. The most notable potential
adverse effect of mycophenolate mofetil is secondary infection.
Azathioprine can be effective in patients with ocular inflammatory manifestations of AdamantiadesBehets syndrome.15 The
present author, however, has not found it to be the most effective drug for this purpose. Still, it can be effective and should be
included in every doctors therapeutic armamentarium for this
potentially devastating, frequently blinding disease. Andrasch and
co-workers9 rigorously studied azathioprine in the treatment of
uveitis of various causes. It was judged effective in 12 patients
and ineffective in 10, either because of drug-induced side effects
or because of inadequate response to treatment. Moore16 stopped
the inflammation associated with sympathetic ophthalmia, and
Hemady and associates17 have noted azathioprines effectiveness in patients with JIA-associated uveitis that does not
respond to conventional steroid therapy. It also can be effective
in the treatment of cicatricial pemphigoid18 and in the care of
relapsing polychondritis-associated scleritis.19 The author has
also used it as a steroid-sparing drug for patients with multi-
focal choroiditis with panuveitis, sympathetic ophthalmia,

VogtKoyanagiHarada syndrome, sarcoidosis, pars planitis, and
Reiters syndrome-associated iridocyclitis.
Mycophenolate mofetil has been shown to be effective in the
care of patients with ocular cicatricial pemphigoid,20,21 scleritis,22
uveitis,23,24 and orbital pseudotumor.25 Control of inflammation
with mycophenolate mofetil as monotherapy occurred in 65% of
a series studied by the author, with 18% of the patients requiring
discontinuation of the drug because of adverse events.23
FOLIC ACID ANALOGS

OF ACTION
Methotrexate (Fig. 30.5), a folic acid analog also known as
amethopterin, binds to folic reductase, thus blocking the conversion of dihydrofolic acid to tetrahydrofolic acid. This interferes with thymidine synthesis and, so, with DNA synthesis
and cell division. Methotrexate has little effect on resting cells
but pronounced effects on rapidly proliferating cells. It affects
both B and T lymphocytes and can inhibit humoral and cellular
responses when administered during antigenic encounter. The
drug is excreted unchanged in the urine. Folinic acid can reverse
the metabolic block produced by methotrexate, thus rescuing
viable cells.
Methotrexate is absorbed after oral administration, but the
drug can also be given by intramuscular or intravenous routes.
It is excreted unchanged in the urine within 48 h. Renal compromise delays excretion and causes undesirable side effects.
Consumption of sulfa drugs, salicylates, phenytoin, chloramphenicol, or tetracycline also increases the risk of methotrexateinduced complications through displacement of methotrexate
from plasma proteins. The drug does not require metabolic
conversion to active principles. The concurrent use of drugs that
affect the kidney, such as nonsteroidal antiinflammatory agents,
can delay drug excretion and lead to severe myelosuppression.
Leucovorin rescue may help reverse some methotrexateinduced toxic effects.
5-Fluorouracil (5-FU) (Fig. 30.6) mimics uracil after intracellular conversion to nucleotide and subsequent incorporation
into both DNA and RNA. The drug is especially toxic to rapidly
dividing cells.
CHAPTER 30
principles, predominantly in the liver. One such metabolically

active product is thioinosinic acid.
At clinical nontoxic doses of 23 mg kg1 day1, azathioprine
has little effect on humoral immunity. Immunoglobulin levels
and specic antibody responses are relatively unaffected. In experimental systems, large doses of thiopurine given within 48 h of
antigen priming can suppress the antibody response and can
induce temporary tolerance to the antigen when given in conjunction with large doses of the antigen.
Thiopurines appear to exert a relatively selective effect on
T lymphocytes: they prolong renal, skin, lung, and cardiac allografts; suppress mixed lymphocyte reaction in vitro; depress
recirculating T lymphocytes that are in the process of homing;
suppress development of monocyte precursor cells; inhibit participation of K cells (which arise from monocyte precursors) in
antibody-dependent cytotoxicity reactions; and inhibit delayed
type hypersensitivity reactions. On the other hand, they do not
affect the onset or progression of the lupus-like autoimmune
disease in NZB/NZW F1 mice, and their immunosuppression
of renal transplant patients, for example, is partial because such
patients consistently show lymphocyte responsiveness in vitro
(proliferation, lymphokine production, cytotoxicity, cytotoxic
antibody) to donor antigen.
Mycophenolate mofetil (Cellcept), converted to mycophenolic
acid, inhibits inosine monophosphate dehydrogenase, which is
critical to de novo purine synthesis. It is administered orally at
13 g day 1.

SIDE EFFECTS
Methotrexate is used to treat certain types of cancer, acute lymphoblastic leukemia, psoriasis, rheumatoid arthritis refractory
to conventional therapy, JIA, and sarcoidosis. Potential complications include severe bone marrow depression with resultant
anemia, leukopenia, and thrombocytopenia; cirrhosis and hepatic
atrophy; ulcerative stomatitis, nausea, vomiting, and diarrhea;
interstitial pneumonitis; malaise, fatigue, and secondary infection; rash; cystitis; nephritis; headache, blurred vision, and
drowsiness; and sterility. The hepatic brosis and cirrhosis
associated with methotrexate therapy are related to dose and
treatment duration, as well as to alcohol consumption. The risk
FIGURE 30.5. Structural formula of methotrexate.
309
SECTION 4
FIGURE 30.6.
Structural formula of
5-fluorouracil.
310
of this potentially devastating complication can be minimized

by administering it only once a week, insisting on total abstinence from alcohol, avoiding other drugs that may enhance the
effects of methotrexate, and monitoring the liver carefully and
regularly. 5-FU is used intravenously to treat metastatic breast,
liver, pancreatic, colon, ovarian, prostatic, and bladder cancer.
Topical 5-FU is used to treat basal cell carcinomas.
Idiopathic cyclitis,12 sympathetic ophthalmia,26 ocular manifestations of rheumatoid arthritis,27 and the uveitis of JIA are
particularly well suited for once-a-week therapy with oral methotrexate. Other varieties of OID, including uveitis including that
associated with Reiters syndrome, ankylosing spondylitis,
inflammatory bowel disease, or psoriasis, may also respond to
methotrexate. This drug may be sufcient to control scleritis
associated with the collagen diseases such as Reiters syndrome
and rheumatoid arthritis; the author has found it effective in
selected persons with progressive cicatricial pemphigoid. The
suggested regimen is 2.57.5 mg once a week, with gradual
escalation of the dose, as indicated by the clinical response, to a
maximum of 50 mg/week.
Regrettably, despite abundant published evidence to the contrary, most ophthalmologists consider methotrexate dangerous.
They undoubtedly remember the complications associated with
high-dose or daily methotrexate therapy in the care of patients
with a malignancy or with psoriasis. Liver toxicity and bone
marrow suppression were indeed prevalent in such patients.
Although the potential risk for such problems in patients treated
with a weekly low dose of methotrexate is not zero, the likelihood of such a problem is clearly low, provided the patient is
managed and monitored correctly.2833 Proper monitoring is
important; this obviously requires the involvement of an additional specialist and regular laboratory testing in these patients,
but the alternative of slow degeneration in visual function is
considerably more costly in both human and economic terms.
At the time of this writing, the sole ophthalmic application of
5-FU is subconjunctival injection after glaucoma ltering surgery
in an effort to prevent subconjunctival brosis and bleb failure.34
The primary toxic effect of subconjunctival 5-FU consists of
supercial punctate keratopathy and persistent corneal epithelial
defect.
SIGNAL TRANSDUCTION INHIBITORS

OF ACTION
Cyclosporin A (CsA, Sandimmune, Neoral) (Fig. 30.7) is a fungal
metabolite originally isolated from cultures of Tolypocladium
inflatum Gams and Cylindrocarpon lucidum Booth by Sandoz
Laboratories as part of a screening program of fungal products
with antifungal activity. This undecacyclic peptide is also
produced by C. lucidum. Borel35 found that it had potent immunosuppressive properties. Subsequent work in experimental
models showed the drug to be truly immunosuppressive and
capable of suppressing allograft reactions to heterotopic heart
allografts in rats. CsA also prolonged the viability of renal
allografts in dogs, heart allografts in pigs, and kidney allografts
in rabbits.
Tacrolimus (Prograf) is another fungus-derived immunosuppressant, isolated from Streptomyces tsukubaensis. It is structurally similar to rapamycin (Fig. 30.8) and is ~100 times more
potent than CsA in preventing allograft rejection in animals.
Rapamycin (sirolimus, Rapamune) is a macrolide isolated from
an actinomycete.
The mechanism of action of CsAs and tacrolimus immunosuppressive properties is incompletely understood, but the best
available evidence suggests that these drugs interfere with
receptors on the surface membranes of certain T lymphocytes
(particularly helper T cells) that recognize DR antigens on other
cells, most notably antigen-presenting cells like macrophages. A
17 kDa protein, cyclophilin, which is a cytosolic protein, binds
CsA and concentrates it intracellularly. Tacrolimus is similarly
bound by another family of immunophilins, FKBP or FK-506binding protein. These binding proteins are peptidylprolyl cistrans isomerases; at least 26 have been identied to date. DR
antigens participate in the production of interleukin-2 (IL-2) by
helper T lymphocytes by rendering the IL-2-producing T cells
sensitive to IL-1. CsA and tacrolimus interfere with helper T-cell
response to IL-1 and block IL-2 production or IL-2 release from
helper T cells. It appears that a complex composed of calcineurin
A, CsA, or tacrolimus, and the relevant immunophilin, inhibits
calmodulin binding, with resultant inhibition of a phosphatase
activity and consequent inhibition of transport of cytoplasmic
NF-AT and NFK6 into the nucleus; the result is inhibition of
IL-2 mRNA transcription. CsA and tacrolimus also may inhibit
IL-1 release from antigen-presenting cells such as macrophages.
Both inhibit expression of IL-3, IL-4, IL-5, and interferon-g.
FIGURE 30.7. Structural formula of cyclosporine.
Rapamycin, unlike cyclosporine and tacrolimus, acts not

through calcineurin but rather through engagement of FKBP12,
creating complexes that bind the target of rapamycin (TOR),
inhibition of which prevents/blocks signal transduction of
cytokine receptors (e.g., IL-2 and IL-4)
CsA and tacrolimus have a fairly selective suppressive effect
on T lymphocytes, which occurs early in the phase of T cellsubset interactions. The drugs profoundly decrease antibody
production to T cell-dependent antigens, inhibit cytotoxic activity
generated in mixed leukocyte reaction, and prolong the life of
skin, kidney, and heart allografts in experimental animals and
humans. They also may prevent or mitigate graft-versus-host
disease and may prolong the life of other organ transplants, such
as pancreas and cornea.

SIDE EFFECTS
CsA is used extensively for prevention of human allograft rejection and for the treatment of a variety of other diseases, including psoriasis. Tacrolimus has been approved by the Food and
Drug Administration for prevention of human liver allograph
rejection. Potential side effects associated with systemic use of
CsA include an apparent increase in the incidence of B-cell
lymphomas, interstitial pneumonitis, and opportunistic infections, particularly from herpes simplex virus and Candida and
Pneumocystis organisms, as well as renal tubular necrosis with
compromise of kidney function.
CsA may be particularly useful in the treatment of various
forms of posterior uveitis, especially when both retina and
choroid are involved in the inflammatory process and especially
if used as part of a multidrug IMT receipe. Thus, sympathetic
ophthalmia, VogtKoyanagiHarada syndrome, multifocal
choroiditis with panuveitis, and posterior uveitis associated with
AdamantiadesBehets syndrome may lend themselves to effective treatment with CsA. The author has been disappointed, however, with the effectiveness of CsA monotherapy compared with
cytotoxic immunosuppressive drugs in treating posterior uveitis
associated with AdamantiadesBehets syndrome when the dose
of cyclosporine is in the acceptable range (57 mg kg1 day1)
from the standpoint of risk for kidney damage. Early enthusi-
DAPSONE
Dapsone (4,4-diaminodiphenylsulfone, Fig. 30.9) is a sulfone
used for the antibiotic treatment of leprosy. In addition to its
antibacterial activity, it is a myeloperoxidase inhibitor and stabilizes lysosomal membranes. Its antiinflammatory and immunosuppressive effects are most dramatic in dermatitis herpetiformis
and cicatricial pemphigoid. It is in the latter disease that ophthalmologists nd it most useful. The author found that, provided the cicatrizing conjunctivitis of cicatricial pemphigoid is not
highly inflamed or rapidly progressive, dapsone halts progression of brosis in 70% of cases.18 And although dapsone may
help patients with relapsing polychondritis, Hoang-Xuan and
co-workers found that treating the scleritis of this disease with
dapsone was disappointing.19
Dapsone may produce profound hemolysis in patients decient
in glucose-6-phosphate dehydrogenase, so any patient considered for dapsone therapy must rst be evaluated for glucose-6phosphate dehydrogenase level. The author begins therapy with
25 mg twice daily; monitor the hemogram, reticulocyte count,
and methemoglobin level biweekly; and increase to as much as
150 mg/day if needed and if tolerated. Additional potential toxic
effects of dapsone include nausea, vomiting, hepatitis, peripheral neuropathy, blurred vision, psychosis, and a nephrotic-like
syndrome.
CHAPTER 30
FIGURE 30.8. Structural formula of FK-506.
astic reports of the effectiveness of CsA in the therapy of

AdamantiadesBehets syndrome were based on dosing schedules of 10 mg kg1 day1.29 Unfortunately, it was subsequently discovered that all patients who consumed this dose of CsA long
enough to achieve the desired therapeutic effect in Behets disease
developed renal damage from the drug. In the authors experience, the lower, less toxic dose of 57 mg kg1 day1, is distinctly
inferior to azathioprine, chlorambucil, and cyclophosphamide
in the care of patients with ocular AdamantiadesBehets disease.
Others report similar disappointment.30 In contrast, it is highly
effective in the care of patients with birdshot retinochoroidopathy, even at low doses.36 However, durable drug-free remissions
are much more likely to result from combination CsAmycophenolate mofetil IMT. CsA can be of enormous benet in
the care of patients with severe eczema, especially those with
signicant atopic keratoconjunctivitis. Topical CsA was investigated for the treatment of corneal graft rejection and the results
were disappointing. It is, however, effective for keratoconjunctivitis sicca. Two other antibiotics with immunosuppressive
properties that have ophthalmic indications are dapsone and
mitomycin C.
MITOMYCIN C
Isolated from Streptococcus calspitosus in 1958, mitomycin
(Fig. 30.10) reacts with DNA in ways similar to alkylating
agents. It cross-links DNA and inhibits its synthesis. It is a
highly effective antimitotic agent. It is used intravenously to
treat carcinoma of the stomach and colon and sometimes as
adjunctive therapy for cancer of the pancreas, breast, bladder, or
lung. The major systemic side effect is myelosuppression.
FIGURE 30.9. Structural formula of dapsone.
311
SECTION 4
FIGURE 30.10. Structural formula of mitomycin C.
The ocular indications for mitomycin C are recurrent pterygium and glaucoma ltering surgery. Kunitoma and Mori37 and
later Choon and Fong38 reported favorably on the efcacy of
mitomycin C eye drops in preventing pterygium recurrence
after resection of pterygium that had recurred many times. Singh
and Foster conrmed these observations,39 and also studied giving
smaller doses of the drug than had been previously employed in
an effort to avoid toxicity, and they compared the efcacy of
topical mitomycin C with that of conjunctival transplantation for
treatment of recurrent pterygium.40 It is clear that topical mitomycin C is effective in this role. It is clearly simpler and cheaper
than either conjunctival transplantation or b-irradiation. The
smallest effective dose and shortest duration of therapy are not
yet clear, however. Foster currently uses a single application of
0.02% at the end of surgery.
The efcacy of mitomycin C as an adjunctive component to
glaucoma ltering surgery is now well established, although, as
in pterygium surgery, in glaucoma surgery the best concentration of the drug and best technique and duration of application
of the drug are not yet dened. The author applies it to the
scleral bed of the guarded trabeculectomy site, 0.4 mg/mL in
saturated cellulose sponges, with conjunctiva draped over the
sponges for 4 min, and then vigorously irrigate the area with
45 mL of balanced salt solution after removal of the sponges.
Potential complications of topical mitomycin C ocular therapy appear to be limited to instances of abuse and negligence, to
drug dosage error, and to use of the drug in patients with ocular
surface disorders, such as sicca syndrome and ocular rosacea. The
author is aware of four cases of scleral or corneal ulceration after
such abuse. Applications were continued for 36 weeks after
surgery rather than the prescribed 1 week.
BIOLOGIC RESPONSE MODIFIERS

OF ACTION
312
Heterologous antisera to leukocytes relevant to immune reactions

have been used experimentally for immunosuppression since
1956 and clinically in humans since the late 1970s. The most
extensively studied and widely used agent is antiserum prepared
against human lymphocytes. Various antilymphocyte serum (ALS)
preparations have been used; the most potent usually are obtained
after immunization of horses with human thymus or thoracic
duct cells. The greatest immunosuppressive activity usually
appears in the immunoglobulin G (IgG) fraction of the immunized horse 24 weeks after immunization begins.
The effects of such antiserums after intravenous administration include leukopenia (highly immunosuppressive preparations
of ALS sharply reduce the number of T lymphocytes); depletion

of thymus-dependent areas in spleen and other lymphoid tissue;
inhibition of delayed hypersensitivity reactions; prolonged
viability of skin, renal, cardiac, liver, and lung allografts; and
suppression of primary and secondary antibody responses if the
antisera are given before antigen priming. Toxic effects of ALS
include anaphylaxis and possible tumorigenesis.
Monoclonal antibodies directed against T lymphocytes (antiOKT3 antibodies) have primarily the same effect as ALS, but their
effect is more limited, being aimed only at T lymphocytes rather
than all lymphocytes. Treatment with intravenous OKT3 antibodies (Orthoclone) can reverse renal allograft rejection reactions.
Complications of anti-OKT antibody therapy include increased
risk of malignancy, fever, malaise, severe nausea, and vomiting.
Pooled human immunoglobulin (gammamune) is used not
only for passive immunization to modify hepatitis A, prevent or
modify measles, and provide replacement therapy for patients with
agammaglobulinemia, but also, in its immunomodulatory role,
to treat idiopathic thrombocytopenic purpura, and an expanding
array of other autoimmune diseases. It must be administered
intravenously or intramuscularly and must be given repeatedly
to achieve an immunomodulatory effect. Adverse reactions include
malaise, nausea, vomiting, fever, chills, headache, arthralgia,
and abdominal pain.

SIDE EFFECTS
ALS has been used in humans predominantly for organ transplantation, in conjunction with corticosteroid and cytotoxic
drug therapy (usually azathioprine). As mentioned earlier, antiOKT antibodies have been used exclusively in humans for
attempted reversal of kidney transplant allograft rejection.
Human immunoglobulin has been used principally as replacement therapy for patients who are hypogammaglobulinemic or
agammaglobulinemic and in treating hepatitis A infections,
herpes zoster infections, and measles infections. Human immunoglobulin has also been used as an immunomodulatory agent
for idiopathic thrombocytopenic purpura and in the experimental treatment of systemic lupus erythematosus and severe
atopic dermatitis. Its toxic effects include malaise, fever, chills,
headache, nausea, vomiting, shortness of breath, and back or hip
pain. Patients with prior allergic responses to immunoglobulin
may experience true anaphylactic reactions.
To the present authors knowledge, anti-OKT3 antibody therapy
has been used only once for an ophthalmic indication. The
author treated a woman with bilateral keratoconus whose body
was rejecting her fourth human leukocyte antigen-matched
corneal graft, in the right eye, in spite of aggressive topical,
regional injection, oral and intravenous pulse steroids, and
topical and systemic CsA therapy with seven days intravenous
OKT3 monoclonal antibody therapy. Her graft was saved, but
this expensive in-hospital effort was an exercise in heroics that
the author suspects will nd little use in ophthalmology. Intravenous gamma globulin therapy has been used extensively in
the care of patients with severe eczema, and the author has used
this treatment modality in several patients whose severe atopic
keratoconjunctivitis did not respond adequately to strict environmental controls and systemic antihistamine therapy. The drug
must be given each week, and the author prefers the intravenous route over the intramuscular one.
We have used IVIg to great effect in our care of patients with
ocular cicatricial pemphigoid which was inadequately responsive to more conventional immunomodulatory agents.14
Daclizumab (Zenapax) is a humanized monoclonal antibody

directed against the alpha chain of the CD-25 glycoprotein,
which is expressed on the surface of activated T lymphocytes. It
is approved and marketed for the treatment of solid allograft
rejections. We41 and others42 have shown that it can be remarkably safe and effective in the care of patients with otherwise
treatment-resistant ocular inflammation, particularly uveitis,
but also scleritis, atopic disease and cicatricial pemphigoid. The
author employs it at a dose of 1 mg kg 1, intravenous, every
2 weeks initially, infused over ~1 h.
24 weeks) and it has been associated with development of

malignancies in some instances46,47 and with increased susceptibility to infection and to reactivation of latent tuberculosis.
The present authors experience suggests that, while treatment
failures are not rare, sufcient evidence for efcacy in sufcient
numbers of cases exists to encourage performance of a placebocontrolled trial. (Sobrin L, Kim E, Christen WG, Papadaki T,
Letko E, Foster CS. Infliximab for the Treatment of Refractory
Ocular Inflammatory Disease, under review, Archives of Opthalmology). The same may be said for adalimumab (Humira) but
not for etanercept (Enbrel).48
TNF-a Inhibitors
Infliximab (Remicade) is a mousehuman monoclonal antibody
which neutralizes TNF-a. It is remarkably effective for the
arthritis associated with rheumatoid arthritis, for the dermatitis
associated with psoriasis, and for the colitis associated with
Crohns disease and with ulcerative colitis. Multiple authors
have reported small series, unmasked and uncontrolled, attesting to its efcacy in treating various forms of uveitis.4345 The
drug must be administered intravenously (510 mg kg 1 every
Summary
IMT is the standard of care for many patients with OID

Ophthalmologists should partner with an ocular immunologist
or with a chemotherapist in order to provide their patients who
have OID with such standard of care
The appropriate goal is durable remission of the OID: no
inflammation OFF all steroids
1. Hektoen L, Corper JH: Effect of mustard

gas on antibody formation. J Infect Dis
1921; 28:279.
2. Bach JF: The mode of action of
immunosuppressive agents. Amsterdam:
Elsevier/North-Holland; 1975.
3. Foster CS, Wilson LA, Ekins MB:
Immunosuppressive therapy for
progressive ocular cicatricial pemphigoid.
4. Watson PG, Hazleman BL: The sclera and
systemic disorders. Philadelphia, PA: WB
Saunders; 1976:90154.
5. Baer JC, Foster CS: Ocular Behets
disease in the United States: Clinical
presentation and visual outcome in 29
patients. In: Usui M, Ohno S, Aoki K, eds.
Symposium on the Immunology and
Immunopathology of the Eye. Tokyo, 1315
March, 1990. Int Cong Ser 918. New York:
Excerpta Medica; 1990:383386.
6. Fain O, Du B, Wechsler I, et al: Intravenous
cyclophosphamide therapy in Behets
disease. In: ODuffy JS, Kokinen E, eds.
Behets disease: basic and clinical
aspects. New York: Dekker; 1989:569.
8. Godfrey WA, Epstein WV, OConnor GR,
et al: The use of chlorambucil in intractable
idiopathic uveitis. Am J Ophthalmol 1974;
78:415.
9. Andrasch RH, Pirofsky B, Burns RP:
Immunosuppressive therapy for severe
chronic uveitis. Arch Ophthalmol 1978;
96:247.
10. Brubaker R, Font RL, Shephard EM:
Granulomatous sclerouveitis. Regression of
ocular lesions with cyclophosphamide and
prednisone. Arch Ophthalmol 1971; 86:517.
11. Buckley CE III, Gills JP: Cyclophosphamide
therapy of peripheral uveitis. Arch Intern
Med 1969; 124:29.
12. Lazar M, Weiner MJ, Leopold IH: Treatment
of uveitis with methotrexate. Am J
13. Foster CS: Immunosuppressive therapy for

external ocular inflammatory disease.
14. Miserocchi E, Baltatzis S, Ekong A, et al:
Efcacy and safety of chlorambucil in
intractable noninfectious uveitis.
15. Yazici H, Pazarli H, Barnes C, et al: A
controlled trial of azathioprine in Behets
syndrome. N Engl J Med 1990; 332:281.
16. Moore D: Sympathetic ophthalmia
treated with azathioprine. Br J Pathol 1968;
52:688.
17. Hemady R, Baer JC, Foster CS:
Immunosuppressive drugs in the
management of progressive, corticosteroidresistant uveitis associated with juvenile
rheumatoid arthritis. Int Ophthalmol Clin
1992; 32:241.
18. Foster CS: Cicatricial pemphigoid. Trans
19. Hoang-Xuan T, Foster CS, Rice BA: Scleritis
in relapsing polychondritis. Ophthalmology
1990; 97:892.
20. Zurdel J, Aboalchamat B, Zierhut M, et al:
Early clinical results with mycophenolate
mofetil in immunosuppressive therapy of
ocular pemphigoid. Klin Monatsbl
Augenheilkd 2001; 218:222228.
21. Choudhary A, Harding SP, Bucknall RC,
Pearce IA: Mycophenolate mofetil as an
immunosuppressive agent in refractory
inflammatory eye disease. J Ocul
22. Siepmann K, Huber M, Stubiger N, et al:
Mycophenolate mofetil is a highly effective
and safe immunosuppressive agent for the
treatment of uveitis: a retrospective
analysis of 106 patients. Graefes Arch Clin
Exp Ophthalmol 2006; 244:788794.
23. Baltatzis S, Tufail F, Yu EN, et al:
Mycophenolate mofetil as an
immunomodulatory agent in the treatment
of chronic ocular inflammatory disorders.
Ophthalmology 2003; 110:10611065.
24. Thorne JE, Jabs DA, Qazi FA, et al:
Mycophenolate mofetil therapy for
inflammatory eye disease. Ophthalmology
2005; 112:14721477.
25. Hatton MP, Rubin PA, Foster CS:

Successful treatment of idiopathic orbital
inflammation with mycophenolate mofetil.
Am J Ophthalmol 2005; 140:916918.
26. Wong VG, Hersh EM, McMaster PRB:
Treatment of a presumed case of
sympathetic ophthalmia with methotrexate.
developing necrotizing scleritis or
peripheral ulcerative keratitis: Effects of
systemic immunosuppression.
28. Graham LD, Myones BL, Rivas-Chacon RF:
Methotrexate associated with long-term
methotrexate therapy in juvenile
rheumatoid arthritis. Pediatr Pharmacol
Ther 1992; 120:468.
29. Giannini EH, Brewer EJ, Kuzmina N, et al:
Methotrexate in resistant juvenile
rheumatoid arthritis. N Engl J Med 1992;
326:1043.
30. Tagwell P, Bennett K, Bell M, et al:
Methotrexate in rheumatoid arthritis. Ann
Intern Med 1989; 110:581.
31. Lehman TJA: Aggressive therapy for
childhood rheumatic diseases. Arthritis
Rheum 1993; 36:71.
32. Wallace CA, Sherry DD: Preliminary report
of higher dose methotrexate treatment in
juvenile rheumatoid arthritis. J Rheumatol
1992; 19:1064.
33. Rose CD, Singsen BH, Eicheneld AH:
Safety and efcacy of methotrexate
therapy for juvenile rheumatoid arthritis.
J Pediatr 1990; 117:655.
34. Fluorocil Filtering Study Group: Fluorocil
ltering surgery study: One-year follow-up.
35. Borel JF: Comparative study of in vitro and
in vivo drug effects on cell-mediated
cytotoxicity. Immunology 1976; 31:631.
36. Vitale AT, Rodriguez A, Foster CS: Lowdose cyclosporin therapy in the treatment
of birdshot retinochoroidopathy.
37. Kunitoma N, Mori S: Studies on pterygium.
IV. Treatment of pterygium by mitomycin C
CHAPTER 30
REFERENCES
313
38.
39.
40.
SECTION 4
41.
314
instillation. Acta Soc Ophthalmol Jpn 1953;

67:601.
Choon LK, Fong CY: The pterygium and
mitomycin C therapy. Med J Malaysia
1976; 31:69.
Singh G, Wilson MR, Foster CS: Mitomycin
eye drops as treatment for pterygium.
Singh G, Wilson MR, Foster CS: Long-term
follow-up study of mitomycin eye drops as
adjunct treatment for pterygium and its
comparison with conjunctival autograft
transplantation. Cornea 1990; 9:331.
Papaliodis GN, Chu D, Foster CS:
Treatment of ocular inflammatory disorders
with daclizumab. Ophthalmology 2003;
110:786789.
42. Nussenblatt RB, Peterson JS, Foster CS:

et al. Initial evaluation of subcutaneous
daclizumab treatments for noninfectious
uveitis: a multicenter noncomparative
interventional case series. Ophthalmology
2005; 112:764770.
43. Kahn P, Weiss M, Imundo LF, Levy DM:
Favorable response to high-dose infliximab
for refractory childhood uveitis.
44. Rajaraman RT, Kimura Y, Li S, et al:
Retrospective case review of pediatric
patients with uveitis treated with infliximab.
45. Suhler EB, Smith JR, Wertheim MS, et al: A
prospective trial of infliximab therapy for
refractory uveitis: preliminary safety and
efcacy outcomes. Arch Ophthalmol 2005;

123:903912.
46. Bongartz T, Sutton AJ, Sweeting MJ, et al:
Anti-TNF antibody therapy in rheumatoid
arthritis and the risk of serious infections
and malignancies: systematic review and
meta-analysis of rare harmful effects in
randomized controlled trials. JAMA 2006;
295:22752285.
47. Bucher C, Degen L, Dirnhofer S, et al:
Biologics in inflammatory disease:
infliximab associated risk of lymphoma
development. Gut 2005; 54:732733.
48. Foster CS, Tufail F, Waheed NK, et al:
Efcacy of etanercept in preventing relapse
of uveitis controlled by methotrexate. Arch
Ophthalmol 2003; 121:437440.
CHAPTER
31
Angiogenic Factors and Inhibitors

Michael J. Tolentino, Anthony P. Adamis, and Joan W. Miller
INTRODUCTION
New blood vessel formation can occur either through angiogenesis or vasculogenesis.1 Vasculogenesis is the formation of
new vessels from the differentiation of angioblasts that subsequently form primitive blood vessels. Formation of new
blood vessels from preexisting microvasculature is called
angiogenesis. Angiogenesis can occur both physiologically and
pathologically. Physiologic angiogenesis occurs mainly in
females during menstruation, ovulation, and the development
of the placenta. Pathologic angiogenesis, on the other hand, can
occur in both sexes. In the fully developed adult, ocular angiogenesis in most cases is pathologic and is a major component of
several blinding conditions. These conditions include agerelated macular degeneration (AMD), diabetic retinopathy
(DR), neovascular glaucoma, corneal neovascularization,
retinopathy of prematurity, and intraocular tumors and
represent some of the most common causes of blindness in the
United States. Understanding the cascade of events that results
in angiogenesis can hopefully elucidate ways to inhibit this
blinding process. In this chapter we discuss the steps involved
in new vessel formation, research techniques to study angiogenesis, angiogenic factors involved in ocular neovascularization, and newly discovered angiogenesis inhibitors.
4. Differentiation 3. Proliferation
2. Migration
Angiogenic
stimuli
1. Dissolution of basement
membrane
and interstitial matrix
FIGURE 31.1. The cascade of angiogenesis begins with an

angiogenic stimulus that leads to the dissolution of basement
membrane and extracellular matrix. This allows the endothelial cell to
migrate and proliferate. After proliferation, the endothelial cell can
differentiate and recruit smooth muscle cells and pericytes, thus
signaling the end of neovascularization.
STEPS IN ANGIOGENESIS
There are two types of angiogenesis: sprouting and nonsprouting (intussusception).2 The events leading to sprouting
angiogenesis begin with dissolution of vessel basement membrane and interstitial matrix. Angiogenesis occurs in response
to angiogenic factors that stimulate the migration and proliferation of vascular endothelial cells. Canalization is followed by
the formation of branches and loops of confluent sprouts that
eventually support blood flow. New vessels can then begin the
process of maturation and differentiation by the recruitment of
pericytes and the deposition of basement membrane signaling
the end of the neovascular cascade (Fig. 31.1). Nonsprouting
angiogenesis involves the proliferation of endothelial cells that
form a lumen within a preexisting vessel. Interstitial tissue
columns in the lumen of preexisting vessels grow, stabilize, and
partition the vessel lumen, resulting in new blood vessel
formation. Nonsprouting angiogenesis has been described more
in the embryonic lung and in tumor models; however, sprouting
and nonsprouting angiogenesis can occur concurrently.2
Intervention at each step of angiogenesis can be used to
inhibit or stimulate new vessel formation. A balance between
endogenous stimulators and inhibitors leads to the maintenance
of mature vessels and the control of physiologic neovascularization. An imbalance results in pathologic neovascularization.
In ocular neovascularization, overexpression of a stimulator of

angiogenesis has been postulated since the late 1940s.3 It was
apparent then that hypoxia and ischemia result in a release of a
factor X that results in the formation of new blood vessel
growth.4,5 It is the identication of this factor X and the hope of
inhibiting its effect that have spurred interest in angiogenesis
research in ophthalmology.
ANGIOGENESIS RESEARCH
METHODOLOGY
The process of new blood vessel growth can be studied by
several in vitro and in vivo bioassays. Bioassays are required to
dene the angiogenic properties of stimulators and inhibitors of
angiogenesis. In vitro endothelial cell chemotaxis, proliferation,
and lumen formation can be used to dene angiogenic or
angiostatic activity. In vivo, there are many bioassays of angiogenesis. The chick chorioallantoic membrane (CAM) assay is one
of the rst in vivo assays used. The corneal neovascularization
micropocket model is probably the most widely used. Others
include chemical- or cautery-induced corneal neovascularization
models; the oxygen-cycling model of retinopathy of prematurity;
and retinal vein occlusion and laser-induced subretinal
315

neovascularization models. A murine transgenic model of
retinal vascular endothelial growth factor (VEGF) upregulation
has been developed. Although these models do not fully mimic
true ocular disease, they can be used to test the in vivo effects
of angiogenic factors and inhibitors in the different vessel beds
of the eye.
SECTION 4
IN VITRO ASSAYS
Capillary endothelial cell cultures were an important step to
study the angiogenic activity of various factors.6 This technique
allows the angiogenic process to be dissected into several steps.
With endothelial cell cultures, angiogenesis does not have to be
measured as an all-or-nothing event; three separate steps are
measurable: proliferation, motility, and capillary tube
formation.
Endothelial cell proliferation is measured by determining cell
counts, thymidine uptake, and other markers of cellular
proliferation and can be used to determine the endothelial cell
mitogenic activity of various compounds. In the presence of a
known angiogenic compound, cellular proliferation can be used
to screen for angiostatic compounds. Endothelial cell migration
can also be measured using the Boyden chamber assay.7 This
measures the chemotactic activity of various factors.
Capillary tube formation is measured in several ways. In
most cases it requires the growth of endothelial cells into a
three-dimensional collagen matrix to form tube-like structures
and lumens.8 A fragment of human placental blood vessel
embedded in a brin gel can give rise to a complex network of
microvessels during a period of 721 days in culture.9 Similar
tube-formation models have been used to assay angiogenic
factors and to screen for angiogenic inhibitors.1012 Fibrinolytic
activity of cell types may also be predictive of the successful
formation of capillary-like structures.13 The mechanism
underlying capillary formation in these in vitro assays is
dependent on the matrix the cells are grown on. Plating human
umbilical vein endothelial cells on Matrigel results in a
posttranslational-dependent capillary-like formation whereas
plating them on brin involves gene transcription and translation.14 These ndings may be helpful in further dissecting the
angiogenic process.
IN VIVO MODELS
Chick CAM Assay
The CAM assay most commonly involves removing a fertilized
chicken egg from its shell and growing it in a culture dish.15
Potential angiogenic substances can be placed on the CAM to
assay their ability to induce angiogenesis. To quantify
angiogenesis, a collagen gel impregnated with an angiogenic
factor is situated between nylon mesh and placed on the CAM
surface. By counting the squares containing new vessels, one
can quantify angiogenesis.16,17 This assay has been used to
identify angiogenic factors and to test angiogenic inhibitors.18
Corneal Neovascularization Models
316
One of the most widely used angiogenesis assays involves the

implantation of an angiogenic stimulant into a corneal micropocket, which induces vessel growth from the limbus toward
the stimulant (Fig. 31.2). Various models have been described
in mice, rats, and rabbits using endotoxin, basic broblast
growth factor, VEGF, and other angiogenic compounds
contained within sustained-release polymers.1923 The rabbit
models offer the advantage of size, but the mouse models offer
the capability for genetic manipulation. A corneal micropocket
model in knockout or transgenic mice can be a useful assay to
FIGURE 31.2. Corneal neovascularization induced in a mouse cornea

by a Hydron pellet impregnated with basic broblast growth factor.
determine if a targeted endogenous factor can inhibit or accentuate neovascularization. For these assays to be effective, the
bottom of the pocket has to be within a critical distance from
the limbus. Chemical cautery, epithelial scraping, and xenograft
corneal transplants have been used to develop injury-induced
models of corneal neovascularization.24,25
Branch Retinal Vein Occlusion Models

Retinal vein occlusion models in rabbits, pigs, cats, and
monkeys have been developed using diathermy and photocoagulation.2629 Various degrees of retinal and iris neovascularization have developed in these models. In a pig model,
photodynamic, laser-induced, branch vein occlusion develops
preretinal and optic nerve head neovascularization.30 A miniature pig model of laser-induced branch retinal vein occlusion
develops only preretinal neovascularization.29
In monkeys, branch vein occlusions produce intraretinal
without preretinal neovascularization.31 When two temporal
retinal veins were occluded, iris neovascularization and disk
neovascularization developed in four of six monkeys. Occluding
three retinal veins and performing vitrectomy-lensectomy
resulted in 100% of monkeys developing iris neovascularization, and two of 12 monkeys developed neovascular
glaucoma.32 The use of dye yellow laser produced iris
neovascularization in 7095% of monkeys without the need
for vitrectomylensectomy (Fig. 31.3).33 A grading system
using standardized fluorescein iris angiograms and masked
readers provides semiquantitative analysis, allowing this
monkey model to be used in the evaluation of angiogenic
inhibitors.27,34,35
A laser-induced venous thrombosis rat model of preretinal
neovascularization has been described.28 With an argon
blue-green laser, 70% of the eyes developed retinal neovascularization and traction retinal detachment. Retinal neovascularization included optic disk neovascularization and
neovascularization elsewhere.
Retinopathy of Prematurity Models

Although constant high oxygen exposure was originally thought
to be the cause of retinopathy of prematurity (ROP), oxygen
fluctuations are a more likely cause. In retinopathy of
prematurity animal models, the developing retinal vasculature
is exposed to different cycles of relative hyperoxia and hypoxia.
The hyperoxia produces vasoconstriction of the immature
FIGURE 31.3. Iris neovascularization and laser-induced branch retinal vein occlusions. (a) Laser-photocoagulated retinal veins in a monkey
retina. (b) Subsequent iris neovascularization. (c) Leakage of fluorescein into the anterior chamber, demonstrating florid iris neovascularization.
Laser-Induced Choroidal Neovascularization

A monkey model of choroidal neovascularization (CNV) was
rst developed using laser-induced retinal vein occlusion and
disruption of Bruchs membrane. The model was inconsistent,
and 30% of the monkeys developed retinal neovascularization,
with 33% developing vitreous hemorrhage.42 Argon laser burns
to the macular area without retinal vein occlusion produced a
higher percentage of monkeys with CNV (Fig. 31.4).43 Unlike
AMD, this model is injury induced, but the development of
CNV bears many similarities to that of AMD. The model
produces a membrane that leaks fluorescein into the subretinal
space.44 VEGF and, aVb3 integrin which have been implicated
in CNV, are also expressed in this model.45,46
Transgenic VEGF-Dependent Mouse Model

A transgenic mouse overexpressing VEGF in the retina has been
created. To produce a VEGF-induced transgenic model of retinal
and subretinal neovascularization, a bovine rhodopsin promoter
was linked to VEGF complementary DNA. This transgenic
mouse produced upregulation of VEGF in the photoreceptors

and very limited systemic expression of the transgene. Three
transgenic founders were described, and one resulted in
intraretinal neovascularization that grew into the subretinal
space.47 Although this pattern of retinal neovascularization is
not seen in disease, this model can be a useful means of
studying VEGF overexpression and its inhibitors in the eye.
Diabetes Models
Many models of diabetes have been developed using mice, rats,
monkeys, and dogs.4852 Both bred rats49 and streptozotocintreated rats51 have produced consistent models of diabetes.
Galactose-fed dogs can produce retinopathy similar to that seen
in diabetes.53 The Koletsky spontaneous hypertensive, noninsulin-dependent rat was observed to have microangiopathic
retinopathy with progressive retinal capillary dropout, and
elevated vascular tortuosity with fluorescein leakage.52 The
Koletsky rat52 and galactose-fed dogs53 are the only two models
of diabetes that develop proliferative retinopathy.
CHAPTER 31
retinal vessels whereas hypoxia produces vasoproliferation

characteristic of retinopathy of prematurity.
Several species have been used, including rat, cat, mouse, and
dog.3641 The models use alteration from high to low oxygen
levels in newborn animals to produce preretinal neovascularization. In a rat model, alternating the oxygen levels from 40% to
80% for several days followed by room air produced histologically conrmed preretinal neovascularization in two-thirds of
the animals.37 In a newborn mouse model, 100% of the animals
developed histologically determined preretinal neovascularization
when placed in 75% oxygen for 5 days followed by room air.40
ANGIOGENIC FACTORS IN OCULAR

NEOVASCULAR DISEASE
The discovery of specic factors that are operative in
angiogenesis has facilitated the accelerated pace of angiogenesis
research. Many angiogenic factors have been discovered to date
(Table 31.1), but the contributions of the majority to ocular
neovascular diseases have not been established. The remaining
discussion will focus on four factors for which evidence
supports this role: VEGF, angiopoietins (Ang), ephrins, and
platelet-derived growth factor-B (PDGF-B).
FIGURE 31.4. Laser-induced CNV. (a) Day 1 after laser treatment. (b) Four weeks after laser treatment, demonstrating subretinal
neovascularization. (c) Angiographically demonstrated CNV 4 weeks after laser treatment.
317
TABLE 31.1. Pro-angiogenic and Anti-angiogenic Factors
VASCULAR ENDOTHELIAL GROWTH

FACTOR
Pro-Angiogenic Factors
SECTION 4
Angiogenin
Angiopoietin-1
Del-1
Fibroblast growth factors: acidic (aFGF) and basic (bFGF)
Follistatin
Granulocyte colony-stimulating factor (G-CSF)
Hepatocyte growth factor (HGF)/scatter factor (SF)
Interleukin-8 (IL-8)
Leptin
Midkine
Pigment epithelium derived growth factor
Placental growth factor
Platelet-derived endothelial cell growth factor (PD-ECGF)
Platelet-derived growth factor-BB (PDGF-BB)
Pleiotrophin (PTN)
Progranulin
Proliferin
Transforming growth factor-alpha (TGF-a)
Transforming growth factor-beta (TGF-b)
Tumor necrosis factor-alpha (TNF-a)
Vascular endothelial growth factor (VEGF)
Anti-Angiogenic Factors
Angioarrestin
Angiostatin (plasminogen fragment)
Antiangiogenic antithrombin III
Cartilage-derived inhibitor (CDI)
CD59 complement fragment
Endostatin (collagen XVIII fragment)
Fibronectin fragment
Gro-b
Heparinases
Heparin hexasaccharide fragment
Human chorionic gonadotropin (hCG)
Interferon a/b/g
Interferon inducible protein (IP-10)
Interleukin-12
Kringle 5 (plasminogen fragment)
Metalloproteinase inhibitors (TIMPs)
2-Methoxyestradiol
Pigment epithelium derived growth factor
Placental ribonuclease inhibitor
Plasminogen activator inhibitor
Platelet factor-4 (PF4)
Prolactin 16 kDa fragment
Proliferin-related protein (PRP)
Retinoids
Tetrahydrocortisol-S
Thrombospondin-1 (TSP-1)
(TGF-b)
Vasculostatin
Vasostatin (calreticulin fragment)
Reproduced, with minor adaptations, with permission from the Angiogenesis
Foundation. From: List of known angiogenic growth factors. In: Understanding
angiogenesis. Available:
http://www.angio.org/understanding/content_understanding.html; accessed 9
October 9, 2006.
318
Key Features
VEGF is the master regulator of physiological and pathological

angiogenesis
Ocular neovascularization is an inflammatory process
promoted by elevated levels of VEGF
VEGF also promotes the inflammation-mediated vascular
damage characteristic of DR and DME
The VEGF165 isoform is especially potent in mediating these
inflammation-related processes
VEGF-targeted therapies (pegaptanib and ranibizumab) are
approved for treating AMD; pegaptanib has also shown
efcacy in treating DME
VEGF AS THE KEY REGULATOR OF

ANGIOGENESIS
Among the numerous factors that contribute to the control of
angiogenesis, only VEGF has proved essential for this process in
the clinic.54 VEGF (also known as VEGF-A) was isolated on two
separate occasions in the 1980s, rst as a tumor-derived factor
that increased vascular permeability55 and subsequently as a
mitogen which showed high specicity for endothelial cells;
molecular cloning revealed that these substances were
identical.56,57 Genetic knockout of only a single VEGF allele
leads to embryonic lethality, demonstrating a critical contribution for VEGF in embryonic vasculogenesis.58,59 Interestingly,
modest overexpression of VEGF (varying from 75% to an
approximate doubling depending on the tissue being examined)
also proved lethal to the embryo. Thus VEGF levels must be
closely regulated for development to proceed normally.60
VEGF acts through binding to two receptor tyrosine kinases,
VEGFR-1 and VEGFR-2, which respond in typical fashion to
ligand binding by activation of signal transduction cascades.61
VEGFR-2 is principally responsible for mediating the effects of
VEGF on angiogenesis and vascular permeability.62 VEGFR-1
has been implicated in mediating monocyte chemotaxis to
VEGF,63,64 a process that may contribute to pathologic angiogenesis,6567 induction of matrix metalloproteinase-9,68 and
release of hepatic paracrine factors.69 Its functions may also
include negative regulation of VEGF by sequestering it, thereby
making it less available to VEGFR-2.61
VEGF is a member of the VEGF-PDGF family (reviewed by
Robinson and Stringer70 and by Ferrara61). The VEGF gene is
composed of eight exons and seven introns, with alternative
splicing resulting in six principal isoforms, containing 121, 145,
165, 183, 189, and 206 amino acids. VEGF165, the isoform that
has been most intensively studied, is a heparin-binding, homodimeric, 45 kDa glycoprotein; a signicant fraction of VEGF165
is bound to the cell surface and to the extracellular matrix.71
Both VEGF189 and VEGF208 are basic, demonstrate strong binding
to heparin, and are largely sequestered in the extracellular
matrix while VEGF121 is acidic, does not bind to heparin, and is
freely diffusible.61
Recently, it has been reported that alternative 3 splicing of the
VEGF gene leads to an alternate family of VEGF isoforms, varying only in the last six amino acids at the carboxyl terminus. These
isoforms can bind to VEGFR-2 but cannot activate it.72,73 This
isoform family, termed VEGFxxxb, appears to constitute a group
of physiological inhibitors of angiogenesis and may contribute to
regulation of angiogenesis since downregulation of these isoforms has been reported in several cancers72,73 as well as in DR.74
VEGF IN OCULAR NEOVASCULAR DISEASES

A major research effort has established a causative role for
VEGF in pathologic ocular neovascularization. Clinical studies
have correlated elevations of VEGF in a variety of ocular
diseases while studies in preclinical model systems have helped
to elucidate the cellular and molecular mechanisms contributing to VEGF-mediated pathogenesis of these conditions.
This review will focus primarily on two major areas of investigation, namely, the role of VEGF in promoting CNV as
well as the importance of VEGF in the etiology of DR and its
associated condition DME. Clinical studies have also demonstrated elevated levels of VEGF in iridal neovascularization,112
retinal vein occlusion,112 neovascular glaucoma,113 and
retinopathy of prematurity.114
Elevation of Vitreous Levels of VEGF in Ocular

Neovascular Diseases
VEGF is produced by many cell types in the retina,90,115,116 and
a series of studies has conrmed that VEGF is elevated in the
ocular fluid in the majority of patients suffering from ocular
neovascularization but only rarely in those where neovascularization was absent.112,117 Studies of eyes removed at autopsy
demonstrated elevated levels of VEGF in both the retinal pigment epithelium and in choroidal blood vessels of maculae with
AMD when compared to control maculae.118 Several groups
have reported that VEGF was overexpressed in retinal pigment
epithelial cells of surgically excised CNV membranes.119,120
These early studies included patients with DR, with the
proliferative form being associated with higher ocular levels of
VEGF than the nonproliferative form.112,117 There have since
been additional studies conrming these initial reports,121123
although it was recently reported that VEGF levels were higher
in eyes with nonproliferative DR as compared to the proliferative form.124 In patients with DR, elevations in VEGF also
have been found in association with increased levels of other
growth factors, including interleukin-6,122 stromal-derived
factor-1,123 angiopoietin 2,125 and erythropoietin.126 In DME,
similar correlations have been established between vitreous
levels of VEGF and angiotensin II,127 interleukin-6,128 stromalderived factor-1,123 and ICAM-1.129 In some cases, these correlations may reflect the interdependence of VEGF and other
cellular constituents; for example, VEGF induces the expression
of ICAM-1,95,130 while VEGF expression is itself upregulated by
angiotensin II131,132 and stromal-derived factor-1.133 Interestingly, in several studies elevated VEGF levels in diabetic
eyes were found to be accompanied by reduced levels of pigment
epithelium derived-factor,124,134,135 which has been reported
to downregulate VEGF expression.136 In contrast to VEGF,
expression of pigment epithelium-derived factor is downregulated by hypoxia and upregulated by hyperoxia.137 Finally, a
recent study has provided evidence that expression of the
VEGFxxxb isoform family may be relevant to the etiology of
DR.74 VEGFxxxb constituted 64 7% of the total vitreous VEGF
in 18 control patients compared to only 12.5 3.6% in 13
diabetic patients (p < 0.001), suggesting that development of
DR is accompanied by a switch in splicing from predominantly
nonangiogenic VEGF isoforms to the angiogenic isoforms.74
CHAPTER 31
A variety of different molecular pathways are involved in

VEGF-mediated vasculogenesis and angiogenesis. First, as mentioned, VEGF acts as a potent mitogen, with endothelial cells
being the primary targets, although mitogenic effects have been
found in other cell types, including pancreatic duct cells,75
Schwann cells,76 and the retinal pigment epithelium.77 VEGF
can mobilize endothelial cell precursors from the bone marrow
during vasculogenesis78 as well as in pathologic conditions such
as tumor angiogenesis and CNV.79,80 It also promotes the survival of retinal endothelial cells through the inhibition of apoptosis81 and induces them to express and secrete plasminogen
activator82 and matrix metalloproteinases.68,83 These actions
facilitate the growth of blood vessels through the surrounding
stroma and may contribute to an amplication of local VEGF
concentrations, since plasmin can release VEGF from the matrix,71
and matrix metalloproteinases can cleave matrix-bound VEGF
to release active amino-terminal fragments.84 In addition,
VEGF upregulates endothelial nitric oxide synthase, increasing
the production and release of nitric oxide; nitric oxide not
only stimulates angiogenesis but can induce greater synthesis
of VEGF.8587
Two properties of VEGF that are particularly relevant in
the context of ocular neovascular disease are its actions on
vascular permeability and its regulation by hypoxia. First,
hypoxia is a key positive regulator of VEGF mRNA expression,8891 which is mediated through upregulation of hypoxiaactivated transcription factor-161 and may be important in
promoting ocular neovascularization in such conditions as
retinopathy of prematurity and DR. Secondly, VEGF is the most
potent known enhancer of vascular permeability, some 50 000
times more effective than histamine,92 which contributes
signicantly to the macular edema and the attending vision loss
in such conditions as AMD and diabetic macular edema
(DME). Both indirect and direct effects contribute to VEGFmediated vascular permeability. Its direct effects include the
induction of fenestrations in the plasma membrane of
endothelial cells93 and the dissolution of tight junctions
between cells.94 Indirect mechanisms involve the VEGFmediated upregulation of endothelial cell expression of
adhesion molecules such as intercellular adhesion molecule-1
(ICAM-1), which promotes the adhesion of leukocytes that act
to damage the endothelium.95,96
Finally, it should be noted that while the main body of VEGF
research has been premised on its potential as a target in
promoting pathologic angiogenesis, it is becoming increasingly
clear that VEGF is a pleuripotent growth factor, acting in a
variety of contexts, some related to its role in promoting angiogenesis, and others quite independent of it. Recent work has
demonstrated that VEGF is required for trophic maintenance
of capillaries,97 and regression of the normal vasculature has
been observed in response to nonselective VEGF inhibition.98
In addition, VEGF is known to be important for processes such
as bone growth,99,100 female reproductive cycling,99,101 wound
healing,102,103 vasorelaxation,104 kidney development and
function,105,106 skeletal muscle regeneration107 and protection
of hepatic cells against hepatotoxins.69 Surprisingly, VEGF
has been found to play a key role in neural survival and
may offer a therapeutic strategy against diseases such as
amyotrophic lateral sclerosis.108 This neuroprotective action
may be important in maintaining the health of retinal neurons,
since VEGF has been shown to promote their survival in
conditions of ischemia.109 Finally, conditional gene knockout
experiments have established that VEGF is essential for
development of the choroicapillaris in mice110 while its
secretion by the retinal pigment epithelium provides trophic
support for this tissue.111
Preclinical Models Demonstrate the Importance of

VEGF in Ocular Neovascularization
Both AMD and DR are diseases with a complex pathophysiology resulting from changes over time in the physiochemical structures of the eye and ultimately resulting in
neovascularization. In AMD, the aberrant blood vessels
originate in the choroid while in DR they proliferate from
319
SECTION 4

retinal blood vessels.138,139 In neither case is there an animal
model that adequately replicates the clinical course of these
diseases, but important insights have nonetheless been gleaned
from experimental systems in which ocular neovascularization
can be induced. In addition to neovascularization, an important
component of the pathology of DR involves damage to the
existing retinal vasculature resulting in excess permeability and
leakage. Both the neovascularization and the vascular damage
appear to be mediated by inflammatory processes in which
VEGF plays a key role, with the VEGF165 isoform behaving as
an especially potent inflammatory cytokine.
Numerous experimental studies using a variety of approaches
have established that elevating VEGF in the eye results in
ocular neovascularization, while inactivating VEGF inhibits its
development. In an early study involving experimental iris
neovascularization induced by laser occlusion of retinal veins
in monkeys, VEGF levels increased in direct proportion to the
degree of induced neovascularization.27 Direct injection of
VEGF into monkey eyes resulted in iris, intraretinal, and
preretinal neovascularization.140142 The induced blood vessels
were aberrant,142 showing evidence of endothelial cell hyperplasia and the excessive tortuosity and leakiness that are
characteristic of CNV.143 Similar neovascular responses have
been induced by overexpression of VEGF from transfected
recombinant DNA in rodents144,145 and in transgenic mice
engineered to overexpress VEGF in the retina.146,147
Several strategies have been employed to demonstrate that
inactivation of VEGF results in inhibition of ocular neovascularization whether in the iris,35,148 retina,65,149151 choroid,152
or cornea.25 Agents employed have included VEGF receptor
fusion proteins149 or transfected DNA constructs expressing the
same,150 antibodies to VEGF,25,35 an antibody fragment binding
to VEGF,152 an anti-VEGF165 aptamer,65 and an antisense oligonucleotide against the VEGF coding sequence.148 In an interesting recent nding, studies in a murine model of retinopathy
of prematurity determined that intravitreous injection of
VEGF165b, one of the family of inhibitory VEGF isoforms,
resulted in a signicant reduction of the pathologic neovascularization that is normally induced after exposure to an
elevated oxygen environment.153
Finally, it was reported that intravitreous injection of
VEGF164, usually considered to be exclusively proangiogenic,
can be inhibitory to the development of CNV caused by laser
injury in mice.154 In these studies, the effect of VEGF was
proangiogenic when the injection was performed prior to injury
and inhibitory when the injection followed the wounding. The
inhibitory effect involved a complex interaction between VEGF,
VEGFR-1, and VEGFR-2 and was modulated by the activity of
SPARC (secreted protein, acidic, rich in cysteine).154 It remains
to be established whether these ndings are specic to the laser
wounding model of CNV or whether they also have relevance to
neovascularization in the clinical setting.
The Role of Inflammation in the Pathogenesis of

AMD and DR
320
One major theme that has emerged from these studies is the
inflammatory nature of both AMD155157 and DR.158 Supporting
evidence comes from studies demonstrating that macrophages,
important mediators of inflammation, are present in surgically
excised CNV membranes120,159 and that induction of
experimental CNV was suppressed in the absence of
macrophages.65,66,160 In this context, VEGF165 has been found to
act as a potent inflammatory cytokine.65
Other evidence that inflammation contributes to ocular
neovascular diseases derives from studies showing that certain
haplotypes of factor H, a regulatory component of the complement cascade, are associated with an increased risk of
developing AMD157 and laser-induced experimental CNV is

dependent on factor C3, another component of the complement
system;161 this dependence may reflect the importance of C3 in
upregulating VEGF expression in this model.162 AMD also has
been associated with elevated systemic levels of C-reactive protein, a marker of inflammation,163 as well as ocular Chlamydia
infection.164 Finally, some patients suffering from AMD165 and
DME166 have experienced regression of their lesions when
treated with intravenous infliximab, an antibody against tumor
necrosis factor-a, a major inflammatory cytokine.
The pathophysiology of DR is associated with the
accumulation of polyols and advanced glycation end products,
oxidative damage, and activation of protein kinase C.139,167 This
leads to alterations in the retinal vasculature characterized
by the death of pericytes, thickening of the basement membrane, and adhesion of leukocytes to the endothelium that
contribute to blockages and capillary dropout resulting in local
hypoxia.139,168 In turn, hypoxia is believed to contribute to local
upregulation of VEGF.169 In addition, reactive oxygen intermediates,170 advanced glycation end products,171 and insulinlike growth factor172 are believed to directly stimulate the
expression of VEGF.
There is now a substantial body of evidence linking elevations in ocular VEGF levels with damage to the existing retinal
vasculature. This process appears to be mechanistically related
to the pruning of the retinal vasculature in normal
development, a process whereby local adhesion of leukocytes
induces endothelial cell apoptosis.173 Much of our information
has come from a rodent model of diabetes, which is induced by
intraperitoneal injection of streptozotocin, with a key
mechanism being the VEGF-mediated upregulation of ICAM-1.
In common with clinical ndings in patients with DR,174 the
onset of diabetes in the rodent model is accompanied by
increased expression of ICAM-1 together with increased retinal
leukostasis; capillary blockage by the leukocytes then leads to
local nonperfusion and leakage, phenomena that can be
prevented by the administration of an antibody directed against
ICAM-1.175 This treatment also reduces the leukostasis-related
injury and death of endothelial cells.96
In the diabetic model, retinal VEGF levels are increased by
3.2-fold after 1 week; this increase is accompanied by increased
vascular permeability and breakdown of the bloodretinal
barrier.176 These effects, as well as the increases in ICAM-1
and retinal leukocyte adhesion, can be signicantly reduced by
the inactivation of VEGF through the administration of a
soluble VEGFRFc fusion protein.176,177 Reductions in
leukostasis, endothelial cell injury, and the number of acellular
capillaries have been seen in transgenic mice that lack either
ICAM-1 or its ligand on leukocytes, CD18.178 Taken together,
these experiments support a mechanism in which the increased
expression of VEGF in turn leads to increased ICAM-1
synthesis by the endothelial cells followed by increased
leukocyte adhesion and the resultant vascular injury.
The nal step in the inflammatory damage is believed to
involve Fas/Fas ligand-mediated apoptosis. During the development of streptozotocin-induced diabetes, FasL expression
was found to be upregulated in neutrophils while Fas expression
was upregulated in the retinal vasculature.179 Systemic
administration of an anti-FasL antibody signicantly inhibited
endothelial cell apoptosis as well as the breakdown of the
bloodretinal barrier.179
VEGF165 as Key Mediator in Pathologic Ocular

Neovascularization
Detailed studies of neovascularization in rodent models have
provided new insights into the molecular and cellular events
underlying the response to retinal ischemia and have implicated

inactivation. Injection of VEGF164 into the eyes of nondiabetic
rats was approximately twice as potent as the administration
of VEGF120 in inducing upregulation of ICAM-1 and leukocyte
adhesion, as well as in promoting bloodretinal barrier
breakdown.183 In parallel experiments with diabetic rats, the
injection of pegaptanib, which specically targets VEGF165/164,
signicantly inhibited leukostasis and bloodretinal barrier
breakdown both in early and in late diabetes.183 Taken together
with the nding that inactivation of VEGF165/164 is especially
potent in mediating ischemia-related neovascularization,65
these ndings provided support for subsequent trials
investigating pegaptanib for the treatment of AMD and DME.
VEGF INHIBITION IN THE TREATMENT OF

OCULAR NEOVASCULAR DISEASES
The strategy of targeting VEGF for the treatment of ocular
neovascular diseases is based on the premise that inactivation
of a major regulator of angiogenesis should offer therapeutic
benets for patients with such conditions. The strategy has
proved successful, yielding two therapies, pegaptanib184,185 and
ranibizumab186,187 both of which are administered by intravitreous injection for the treatment of neovascular AMD.
Pegaptanib has shown excellent long-term safety.188 In a rst for
an AMD therapy, ranibizumab was shown to improve the mean
visual acuity of patients. In contrast to the laser-ablative
approaches, both pegaptanib and ranibizumab are indicated for
all angiographic subtypes of AMD, effectively obviating the
need for angiographic classication of patients prior to
determining their suitability for treatment.
In addition, in a phase 2 trial involving 172 patients with
DME,189 those receiving intravitreous pegaptanib had better
mean visual acuity than those receiving sham injections as well
as a greater likelihood of reduced central thickness and a lesser
need for photocoagulation therapy; furthermore, many of those
patients who had retinal neovascularization experienced
regression of neovascularization in response to pegaptanib
treatment.190 Similar results were also obtained in a recent
phase 2 trial involving 98 patients testing pegaptanib as a
treatment for macular edema secondary to central retinal vein
occlusion.191
Physiologic Revascularization
Pathologic neovascularization
30
1.0
P<.01
Control
Control
Area (mm2)
P<.01
0.4
VEGF164-selective blockade
VEGF164-selective blockade
Nonselective VEGF blockade
Nonselective VEGF blockade
Area (mm2)
0.8
0.6
CHAPTER 31
one VEGF isoform, VEGF165, as being especially important in

mediating pathologic neovascularization through its
acceleration of inflammatory processes. Ishida et al65 used a rat
retinopathy of prematurity model to compare pathologic retinal
neovascularization to the physiologic neovascularization that
normally occurs in postnatal rats. Compared to physiologic
neovascularization, retinal VEGF expression was dramatically
enhanced during pathologic neovascularization; moreover the
ratio of VEGF164/VEGF120, which was 2.2 1.1 in physiologic
neovascularization, increased to 25.3 8.7 in the pathologic
form (VEGF164 and VEGF120 are the respective rodent versions
of the human VEGF165 and VEGF121 isoforms).65
Studies also revealed that VEGF164 was approximately twice
as potent as VEGF120 in promoting monocyte chemotaxis.130
This nding is notable in that pathologic neovascularization
has been shown to be accompanied by an influx of adherent
leukocytes and was inhibited by inactivation of monocyte
lineage cells;65 these cells may contribute to pathologic neovascularization by secreting VEGF,180,181 thus promoting local
amplication of inflammation. Other experiments involving
laser-induced CNV models demonstrated that the development
of CNV was inhibited when macrophages were depleted with
clodronate liposomes66,160 or in knockout mice lacking
chemokine receptor CCR2, the receptor for monocyte chemoattractant protein-1.182 Other evidence for the importance of
leukocytes in experimentally induced CNV comes from studies
demonstrating a reduction in the severity of CNV in mice with
targeted inactivation of either ICAM-1 or the leukocyte
adhesion molecule CD18.67
Another key nding is that intravitreous injection of
pegaptanib, which binds to VEGF164 but not to VEGF121,
inhibited leukocyte adhesion and pathologic ocular neovascularization while leaving physiologic neovascularization
unaffected. In contrast, injection of a VEGFR-Fc fusion protein,
which binds to all isoforms of VEGF, inhibited both physiologic
and pathologic ocular neovascularization (Fig. 31.5).65 It is of
particular interest that VEGF120/188 mice, lacking VEGF164
entirely, develop normal retinal vasculature.65
Studies in the diabetic rat model provided further important
evidence of the specic inflammatory nature of VEGF165 and of
the potential therapeutic value that could result from its
20
10
0.2
a 0.0
FIGURE 31.5. The role of VEGF164 in pathologic and physiologic retinal neovascularization. (a) Both nonselective blockade with a VEGFRFc
fusion protein and blockade of VEGF164 with a pegylated anti-VEGF aptamer signicantly inhibited pathologic retinal neovascularization; (b) The
VEGFRFc fusion signicantly inhibited physiologic retinal neovascularization, but it was not impaired by blocking VEGF164.
Adapted from Ishida S, Usui T, Yamashiro K, et al: VEGF164-mediated inflammation is required for pathological, but not physiological, ischemia-induced retinal
neovascularization. J Exp Med 2003; 198:483489.
321
CONCLUSIONS
VEGF is a key mediator of angiogenesis and contributes to
ocular neovascular disease through its effects as an endothelial
cell mitogen, vascular permeability factor, and inducer of
inflammation. Clinical trials demonstrating the efcacy of
VEGF blockade in diseases such as AMD and DME conrm the
important role of VEGF in ocular neovascular diseases and
validate the strategy of attacking the pathogenesis of ocular
neovascular diseases with molecularly targeted agents. Future
therapeutic approaches combining anti-VEGF therapies with
agents targeting other molecular components involved in
angiogenesis may provide even better efcacy.
PLATELET-DERIVED GROWTH FACTOR-B

Key Features
SECTION 4
PDGF-B is a growth factor structurally related to VEGF

The contributions of PDGF-B to angiogenesis are mediated
largely through its effects on mural cells such as pericytes and
vascular smooth muscle cells
Mural cells are recruited to the endothelium primarily in
response to endothelial cell-secreted PDGF-B, which activates
PDBF-b receptors on the mural cells and stimulates their
migration and proliferation
Vasculature that is stably covered with mural cells is largely
resistant to the effects of VEGF withdrawal, making it less
susceptible to therapeutic intervention with VEGF blocking
agents
Blocking both VEGF and PDGF may provide improved efficacy
in the treatment of established ocular neovascular lesions
INTRODUCTION
322
Members of the PDGF family, so named because the rst PDGF

was isolated from platelets,192 are dimeric proteins composed
of four different polypeptide chains, PDGF-A-D (reviewed
by Fredriksson et al193). Most PDGFs are homodimeric (i.e.,
PDGF-AA and PDGF-BB), although heterodimers can form
between A and B chains, forming PDGF-AB. All PDGF chains
are structurally related to VEGF and share a highly conserved
growth factor domain. PDGF is produced by a wide range of
different cell types, including broblasts, vascular smooth
muscle cells, endothelial cells, retinal pigment epithelial cells,
and macrophages (reviewed by Helden and Westermark194).
Although most of these cell types make both A and B chains,
their expression is differentially regulated.195
The receptors for PDGF are two related tyrosine kinases:
PDGF receptor (PDGFR)-a and PDGFR-b; due to their dimeric
nature, PDGFs can interact with two PDGF receptors simultaneously, promoting PDGF receptor dimerization and
autophosphorylation.196 PDGF-A chains can bind to only
PDGFR-a, whereas PDGF-B chains can interact with both
PDGFR-a and PDGFR-b; therefore, homodimeric PDGF-BB can
induce dimerizeration of three combinations, PDGF-aa,
PDGF-ab, and PDGF-bb.194,197 PDGF-b has been detected in
many cells that play a role in angiogenesis, including vascular
smooth muscle cells, capillary endothelial cells, pericytes,
retinal pigment epithelial cells, myeloid hematopoietic cells,
and macrophages; the expression of PDGFR-a appears to be
more restricted though it is notable that platelets express only
PDGFR-a.194
PDGFs have wide-ranging functions, including roles in
embryonic vascular and central nervous system development,
wound healing, atherosclerosis, and kidney brosis.194 PDGF-B
(used hereafter to describe the homodimeric form PDGF-BB)

has been found to be important in angiogenesis, particularly in
respect to its effects on the recruitment of vascular smooth
muscle cells and pericytes to areas of neovascularization.198
This section will focus on the role of PDGF-B as an angiogenic
factor, with emphasis on its contribution to pathologic ocular
neovascularization.
PDGF-B IN ANGIOGENESIS
In normal embryos, PDGF-B is expressed by vascular endothelial cells and megakaryocytes while expression in neurons
and macrophages occurs postnatally.199 Knockout mice decient
for PDGF-B die perinatally, with abnormal renal glomerular
development, hemorrhages, thrombocytopenia, and anemia.200
PDGFR-b expression in embryonic mice occurs primarily in
pericytes, with larger arteries being surrounded by several layers
of PDGFR-b-positive mesenchymal cells.201 The phenotype for
PDGFR-b knockout mice is very similar to that of PDGF-B
mice,202 suggesting that PDGF-B mediates its effects largely
through its interactions with PDGFR-b.
Early studies evaluating the potential angiogenic effects of
PDGF-B reported that cultured rat microvessel endothelial cells
expressed both PDGF receptor a and b chains; PDGF-B was
mitogenic for these cells while PDGF-A was not; PDGF-AB was
mitogenic, but the effects were not as great as with PDGF-B.203
Other studies reported that PDGF-B stimulated proliferation of
human microvascular endothelial cells204 and was involved in
formation of capillary-like tubes in cultured bovine aortic
endothelial cells.205
Subsequent work has claried the role of PDGF in angiogenesis to be particularly important in recruitment of mural
cells expressing PDGFR-b to developing vasculature. Mural cells
include pericytes and vascular smooth muscle cells. Pericytes
are solitary cells associated with small vessels such as arterioles,
capillaries, and venules and share their basement membrane
with the endothelium while vascular smooth muscle cells form
concentric layers around larger vessels such as veins and
arteries.198 Characterization of PDGF-B knockout mice
identied a specic defect in which pericyte loss resulted in
capillary microaneurysms due to instability of the capillary
walls.201 It was further demonstrated that embryos lacking
PDGF-B or PDGFR-b expression had reduced proliferation of
mural cell progenitors, which normally proliferate at sites of
endothelial PDGF-B expression.198 These ndings are
consistent with a model in which PDGF-B that is released by
the endothelium drives vascular smooth muscle recruitment
and migration, resulting in abnormal vasculature when PDGFB is not present (Fig. 31.6).198
The localization of PDGF in the pericellular space may be
important in mediating its effects. In studies involving an
endothelial cell line that overexpresses PDGF-B, the majority of
newly synthesized PDGF-B was found to be associated with the
cellular matrix through an interaction with heparin sulfate
proteoglycans and was released in response to a-thrombin.206
The carboxyterminal heparin-binding motif, which is highly
analogous to a similar motif found in certain isoforms of VEGF,
was found to be important in maintaining normal growth and
fertility in knockout mice.207 Mice lacking this domain had a
reduction in pericyte density associated with partial
dissociation of pericytes from the vasculature, consistent with a
model in which PDGF-B retention is required for the formation
of depots or gradients that conne pericyte migration to the
abluminal surfaces of microvessels.207
There is also evidence that VEGF and PDGF-B may work
together in promoting angiogenesis. For example, PDGF has
been found to induce VEGF expression in endothelial cells208
PDGF-B driven
vSMC proliferation
and migration
Wild type
PDGF-B
PCGF-B
PDGF-B
vSMC induction
PDGF-B
Reduced vSMC
proliferation
and migration
PDGF-B or
PDGFA-b
knock-out
FIGURE 31.6. The role of PDGF-B in the development of vessel

walls. Undifferentiated mesenchymal cells (gray) surrounding the
newly formed endothelial tube (yellow) are induced to become
vascular smooth muscle cells (vSMC) and to assemble into a vascular
wall (red). During vessel growth and sprouting, PDGF is released by
the endothelium to drive vSMC proliferation and migration. In mice
lacking PDGF-B or PDGFR-b, there is reduced vSMC proliferation and
migration, which results in vSMC hypoplasia of larger vessels and
pericyte deciency in capillaries.
Adapted from Hellstrom M, Kalen M, Lindahl P, et al: Role of PDGF-B and
PDGFR-beta in recruitment of vascular smooth muscle cells and pericytes during
embryonic blood vessel formation in the mouse. Development 1999;
126:30473055.
and to enhance angiogenesis in gliomas by stimulating VEGF

expression and pericyte recruitment.209 In an assay in which
angiogenesis-promoting gel plugs were implanted subcutaneously into mice, VEGF-A and broblast growth factor-2
synergistically promoted neovascularization by enhancing
PDGF-B signaling; the mechanism was believed to involve
upregulation of endothelial cell PDGF-B expression by VEGF
and upregulation of PDGFR-b expression by broblast growth
factor-2.210
Pericyte recruitment to developing vasculature has been
found to lag behind the formation of the endothelial cell plexus,
providing a plasticity window during which endothelial cells
are highly sensitive to VEGF withdrawal.211,212 Intraocular injection of PDGF-B caused detachment of pericytes from newly
formed retinal vessels and abnormal vascular modeling in a rat
retinopathy of prematurity model, presumably by competing
with endogenous signaling; VEGF accelerated pericyte coverage
of developing vasculature.211 Once the pericyte coating was
complete, the vasculature was stabilized and resistant to VEGF
withdrawal.212
PDGF-B IN OCULAR NEOVASCULAR DISEASE

PDGF has been demonstrated to be an autocrine growth
factor213 and a chemotactic factor214 for retinal pigment epithelial cells. These effects appear to be predominantly mediated
through PDGFR-b as shown by studies demonstrating that
rabbit retinal endothelial cells migrated in response to either
PDGF-B or PDGF-AB, but not to PDGF-A.215 Similarly, PDGFB, but not PDGF-A, stimulated migration of rabbit corneal
broblasts and epithelial cells.216 However, the contribution by
PDGFR-a, which also is activated by PDGF-B, is not clear.
Studies showed that truncated PDGFR-a lacking the intracellular domain was able to block the effects of wild-type
PDGFR-a when coexpressed in rabbit conjunctival broblasts

and that it was able to reduce the experimental proliferative
retinopathy when these cells were injected into the vitreous of
rabbits.217 In a subsequent study, retinal detachment resulting
from intravitreous injection of rabbit conjunctival broblasts
was completely inhibited by coinjecting a retrovirus that
expressed the truncated PDGFR-a.218 Yet, the relative
contribution of PDGFR-a to ocular neovascularization is not
well established.
A variety of models have been used in an attempt to dene a
role for PDGF-B in ocular neovascular diseases. To gain a better
understanding of the role of PDGF-B in DR in mice, which
would otherwise be impossible due to the embryonic lethality of
PDGF-B knockouts, mice were engineered with selective
inactivation of PDGF-B in endothelial cells.219 These mice had
vascular aberrations of retinal capillary formation corresponding to pericyte deciency resulting in areas of proliferative
retinopathy when pericyte density was less than 50% of
normal.219 Similar ndings were reported with the use of a
kinase inhibitor that blocks PDGFR signaling.220 In another
model, transgenic mice engineered for photoreceptor-specic
expression of PDGF-B had traction retinal detachment
characterized by proliferation of astrocytes, pericytes, and
endothelial cells.221 These effects were largely blocked by
administration at postnatal day seven of a single intravitreous
injection of an aptamer that binds PDGF-B, conrming the role
of PDGF-B in the pathogenesis of these lesions.222 In a model of
corneal neovascularization, mice treated systemically with a
PDGF inhibitor had a loss of pericytes and reduced vessel
density in the cornea; in this model the inhibitor appeared to be
effective in reducing pericyte coverage in existing vessels as well
as in growing vessels.223
Recent studies involving three different murine models have
further elucidated the respective contributions of the VEGF and
PDGF-B signaling pathways in ocular neovascular disease.224
PDGF-B signaling was inhibited by systemic administration of
an anti-PGDFR-b antibody and VEGF-A signaling was inhibited
by systemic administration of the anti-VEGF aptamer
pegaptanib. One model evaluated the effects of VEGF and
PDGFR-b inhibition on the physiologic development of retinal
vasculature in neonates. The anti-PDGF-b antibody, but not
pegaptanib alone, signicantly inhibited retinal blood vessel
growth at postnatal day three; further reductions occurred when
both agents were administered simultaneously. However, in a
CNV model, the anti-PDGFR-b antibody had little effect on
developing or established CNV, while signicant reduction
occurred with pegaptanib; addition of the anti-PDGFR-b
antibody provided greater reduction than achieved with
pegaptanib alone.224
A corneal neovascularization model was used to investigate
the effects of the PDGFR-b blockade on regression of
established vasculature. In this model, neovascularization
occurs primarily in the rst 7 days postinjury and vessels do not
naturally regress even through 28 days, making it suitable for
assessing the effects of pharmacologic intervention on vessel
regression. When mice were treated with anti-PDGFR-b antibody between 10 and 20 days postinjury, mural cells appeared
to detach from corneal neovessels (Fig. 31.7a). When mice were
treated daily immediately after corneal injury, there was
signicant reduction in the area of neovascularization with
pegaptanib, but not a signicant reduction with the antiPDGFR-b antibody; when both agents were combined, there
was a signicantly greater reduction than with pegaptanib alone
(Fig. 31.7b).
It is interesting that inhibition of PDGFR-b signaling in these
models resulted in pericyte depletion of established vessels as
well as in developing vasculature but not in quiescent limbal
CHAPTER 31
323
PBS control
Anti-PDGFR- antibody
PBS
control
Anti-VEGF
aptamer
Anti-VEGF aptamer +
anti-PDGFR- antibody
SECTION 4
Anti-PDGFRantibody
b
FIGURE 31.7. The effects of PDGF-B blockade on mural cells and vascular growth in a corneal neovascularization model. (a) Mice were injected
with anti-PDGFR-b antibody or phosphate-buffered saline (PBS) every day starting at 10 days postinjury and sacriced at 20 days postinjury.224
Neovessels from mice treated with the anti-PDGF-b antibody had reduced mural cell coverage when compared with PBS-treated mice. Scale
bar = 20 m. (b) Immediately after corneal injury, mice were treated daily with one of the following: PBS, a pegylated anti-VEGF aptamer, an antiPDGFR-b or a combination of the anti-VEGF aptamer and the anti-PDGFR-b antibody. Neovessels are delineated in green. Scale bar = 100 m.
Quantitative analysis demonstrated that the anti-VEGF aptamer signicantly reduced neovascularization when compared with PBS or the antiPEGFR-b antibody (p < 0.01), while the combination signicantly reduced neovascularization when compared with the aptamer alone (p < 0.05).
Adapted from Jo N, Mailhos C, Ju M, et al: Inhibition of platelet-derived growth factor B signaling enhances the efcacy of anti-vascular endothelial growth factor
therapy in multiple models of ocular neovascularization. Am J Pathol 2006; 168:20362053.
vessels.224 These ndings are consistent with the notion of a

plasticity window for both PDGFR-b signaling inhibition and
VEGF dependency and provide support for a combination
therapeutic approach using inhibitors of both VEGF and
PDGF-B for treatment of ocular neovascular disease.
324
ANGIOPOIETINS
Key Features
CONCLUSIONS
PDGF-B has been shown to provide an important contribution

to angiogenesis, mediated primarily through its effects on mural
cells such as pericytes and vascular smooth muscle cells. The
recruitment of mural cells to the endothelium occurs primarily
in response to endothelial cell-secreted PDGF-B, which
activates PDBF-b receptors on the mural cells and stimulates
their migration and proliferation. Mature vasculature, characterized by a stable coverage of mural cells, is largely resistant to
the effects of VEGF withdrawal and may be less susceptible to
therapeutic intervention. Combined therapies in which both
VEGF and PDGF are blocked may provide improved regression
of established ocular neovascular lesions.
Angiopoietins-14 (Ang14) form a family of growth factors

involved in angiogenesis; only Ang1 and Ang2 currently are
known to have roles in ocular neovascular disease
Ang1 and Ang2 are ligands for the receptor tyrosine kinase
Tie2, which is expressed primarily on endothelial cells
Ang1 is produced by cultured vascular smooth muscle cells
and is believed to be involved in maintaining the integrity of the
quiescent endothelium through its interaction with Tie2;
experimental elevation of Ang1 within the context of ocular
neovascularization is inhibitory to the development of
pathological angiogenesis
Ang2 is produced by endothelial cells and serves as a naturally
occurring antagonist to Ang1/Tie2 angiogenesis; in ocular
neovascularization, when VEGF concentrations are high
elevation of Ang2 promotes neovascularization and when
VEGF concentrations are low elevation of Ang2 is inhibitory
The angiopoietins (Ang1Ang4) form a family of growth factors.

Ang1 was rst identied as a secreted glycoprotein capable of
both binding and inducing the phosphorylation of Tie2, a
receptor tyrosine kinase,225229 whereas Ang2 is an antagonist
of Ang1 and Tie2 (reviewed by Eklund and Olsen230 and
Maisonpierre et al228). Ang3 and Ang4 represent murine and
human counterparts, respectively, of what appears to be the
same gene product despite their highly divergent structure.229
To date, little is know about the biology of Ang3 and Ang4.
The Ang receptor Tie2, expressed primarily on endothelial
cells,225229 is part of a family that includes Tie1. Tie2 knockout
mice die by embryonic day 10.5 due to defective formation of
microvessels; in adult mice, Tie2 is expressed both during
angiogenesis and also in quiescent vasculature in many
tissues.231 Tie1 is essential for structural integrity of the
vascular endothelium and hence survival as its gene knockout
in mice leads to death shortly after birth.232 The function of
Tie1 has been much less studied than that of Tie2, whose
actions are the focus of the remaining discussion.
The relative contributions of Ang1 and Ang2 to angiogenesis
and ocular neovascularization are complex. The preclinical and
clinical studies that will be summarized in this section will
show that although Ang1 is essential for the development of the
normal vasculature, experimental elevation of Ang1 in the
context of ocular neovascularization is inhibitory to the development of pathologic angiogenesis. Studies involving Ang2, in
contrast, have shown that depending on the local concentration
of VEGF the experimental elevation of Ang2 may either
promote or inhibit ocular neovascularization; when VEGF concentrations are high elevation of Ang2 promotes neovascularization and when VEGF concentrations are low elevation
of Ang2 is inhibitory.
ANGIOPOEITIN-1
Role in Angiogenesis
Knockout mice lacking expression of Ang1 suffer embryonic
lethality characterized by failure to remodel the primary
capillary plexus, a phenotype similar to that seen in Tie2
knockout mice.233 Ang1 has been found to be expressed in close
proximity to developing embryonic vasculature226,234 and is
secreted by cultured vascular smooth muscle cells but not by
endothelial cells.235 These ndings are consistent with studies
demonstrating that granular deposits of Ang1 are found in the
extracellular matrix of an Ang1-expressing cultured carcinoma
cell line.236 Adherence of endothelial cells to these cultures
stimulated the release of the granules from the matrix and
resulted in phosphorylation of Tie2 on the endothelial cells.236
In the adult, it is believed that constitutive expression of
Ang1 induces continuous activation of Tie2 and contributes to
maintaining the integrity of the quiescent endothelium,237,238 as
depicted in Figure 31.8.239 Ang1 protects endothelial cells
against apoptosis240 and has been found by some investigators
to induce proliferation of cultured endothelial cells241 while
others have reported contrasting results.242 Treatment with
Ang1 has also been shown to promote proliferation of
endothelial cells in vivo, leading to vessel diameter enlargement
in the venous circulation,243 an effect that is restricted to a
dened and brief postnatal period. In contrast, in adult mice,
the use of collagen oligomeric matrix protein, Ang1, a fusion
protein that is an especially potent activator of Tie2, led to
vessel enlargement in many tissues.244 Overexpression of Ang1
in tumors is associated with tumor vessel maturation and
reduced permeability, together with lowered tumor
growth.245,246
FIGURE 31.8. AngTie functions in the regulation of quiescent and

activated vasculature. The quiescent, resting endothelium (upper) has
an antithrombotic and antiadhesive luminal cell surface. Ang1 (shown
as multimeric (white)), is secreted by periendothelial cells at a
constitutive low level. By acting on the endothelium to maintain low
level Tie2 phosphorylation, Ang1contributes to maintaining the
vascular endothelium in the resting state. Ang2 (dimeric (gray)) is
stored in endothelial cell WPB of the quiescent vasculature.
Endothelial cell activation (lower) involves the release of the
endothelial cell WPBs, and concomitant liberation of a variety of
stored factors, including Ang2. The resultant Ang1/Ang2 ratio is now
biased more in favor of Ang2, leading to endothelial destabilization,
and making the endothelial cell layer more responsive to other stimuli,
including proinflammatory cytokines.
CHAPTER 31
INTRODUCTION
Adapted from Ptaff D, Fiedler U, Augustin HG: Emerging roles of the

Angiopoietin-Tie and the ephrin-Eph systems as regulators of cell trafcking.
J Leukoc Biol 2006.
Like VEGF, Ang1 possesses a range of functions important in

angiogenesis, including the ability to promote endothelial cell
chemotaxis242 and to stimulate secretion of plasmin and matrix
metalloproteinases247 as well as to depress secretion of tissue
inhibitors of metalloproteinases.247 These functions are
believed to underlie the vascular remodeling effects of Ang1 in
vivo, as well as the induction of tubule formation in model
systems.238
In some instances, Ang1 has been found to modify or
contribute to the effects of VEGF. For example, Ang1 inhibits
vascular leakage induced by VEGF243,248 and has been found to
antagonize the proinflammatory effect of VEGF through
inhibition of VEGF-mediated upregulation of ICAM-1 and vascular cell adhesion molecule-1249 and to inhibit the induction of
tissue factor expression by VEGF and tumor necrosis
factor-a.250 Ang1 also has been found to induce dose-dependent
capillary sprouting in endothelial cell monolayers; suboptimal
concentrations of Ang1 and VEGF acted synergistically in this
assay.251 Other work using a human broblast/endothelial cell
coculture assay showed that Ang1-induced sprouting could be
inhibited by blocking VEGF signaling.252
325

In turn, VEGF has been found to impact on the expression of
Ang1. In cultured human retinal pigment epithelium cells,
VEGF upregulated Ang1 mRNA expression in a dose-dependent
manner.253 Similar ndings were reported for cultured bovine
retinal pericytes in which both VEGF treatment and hypoxia
signicantly increased Ang1 mRNA expression; contrasting
ndings were reported for bovine aortic endothelial cells in
which Ang2, but not Ang1, expression was upregulated by these
two stimuli.254
SECTION 4
Role in Pathologic Ocular Neovascularization

The effects of Ang1 on endothelial homeostasis have led
investigators to evaluate the role of Ang1 in ocular neovascular
diseases. In studies using a micropocket assay to promote
corneal neovascularization, Ang1 did not affect neovascularization on its own but increased perfusion of the microvasculature when administered in conjunction with VEGF; the
effect was blocked with the addition of excess soluble Tie2.255 In
another model, systemically administered soluble Tie2 also
reduced neovascularization in laser-induced CNV and
ischemia-induced retinopathy in mice, supporting a role for
Ang/Tie2 signaling in pathologic ocular neovascular diseases.256
Evidence that Ang1 plays a protective role against pathologic
ocular neovascularization was provided in studies using rodent
models of DR. Intravitreous injection of Ang1 in early diabetes
normalized expression of VEGF and ICAM-1, reducing
leukocyte adhesion to the retinal vasculature and damage to
endothelial cells and bloodretinal barrier breakdown.257
Systemically administered Ang1 produced similar effects in
animals with established diabetes.257
Further evidence of the potential utility of Ang1 in ocular
neovascular disease comes from studies with transgenic mice in
which Ang1 expression could be specically induced in the
retina.258 In this model, elevated retinal levels of Ang1 suppressed the development of CNV following laser wounding and
inhibited the development of retinal neovascularization
following ischemic retinopathy. Moreover, induced elevation of
Ang1 also inhibited the bloodretinal barrier breakdown
following intravitreous injection of VEGF.258 Subsequent studies
in transgenic mice in which retinal expression of both Ang1 and
VEGF could be induced demonstrated that simultaneous
induction of both factors suppressed VEGF-induced CNV and
prevented retinal detachment.259
ANGIOPOIETIN-2
326
Ang2 was rst identied through its homology to Ang1.228

Studies in Ang2 knockout mice have demonstrated that Ang2 is
not required for embryonic vascular development, but it is
essential for subsequent angiogenic remodeling and proper
lymphatic vessel development; most Ang2-decient mice die
within a few weeks of birth.260 However, transgenic overexpression of Ang2 disrupted embryonic mouse blood vessel
formation resulting in an embryonic lethal phenotype
resembling the loss of either Ang1 or Tie2.228 Genetic rescue
with Ang1 was able to correct the lymphatic but not the
angiogenesis, suggesting that Ang2 serves as an agonist for Tie2
in establishing the lymphatic vasculature but is antagonistic for
Tie2 in angiogenesis.260 Other studies showed that Ang2 bound
Tie2 on endothelial cells with comparable binding afnity
to that of Ang1 but did not induce Tie2 phosphorylation.261
Together these ndings support the hypothesis that Ang2 serves
as a naturally occurring antagonist to Ang1/Tie2 angiogenesis.228 This model is not supported in all contexts, however.
In cultured endothelial cells, Ang2 was able to activate Tie2
following long exposures262 and to induce tube formation.262,263

Like Ang1, Ang2 can stimulate matrix metalloproteinase
expression by cultured retinal endothelial cells.264
Endothelial cells are a primary source of Ang2 production260,265 where it is stored in WeibelPalade bodies (WPB)
from which it can be released by a variety of stimuli,239 shown
in Figure 31.8. Expression of Ang2 is upregulated by hypoxia
and VEGF.254,266,267 In contrast to the quiescent maintenance
function exerted by Ang1, expression of Ang2 occurs
prominently at sites of vascular remodeling, where it serves to
destabilize the endothelial layer.230 This was demonstrated in
studies showing that cultured endothelial cells rapidly detach
following exposure to Ang2 and that these effects can be rescued
by Ang1, soluble Tie-2, or VEGF.268 Similarly, local upregulation
of Ang2 in tumors was associated with vascular regression in
the absence of VEGF whereas angiogenesis occurred when
VEGF was present.269 Thus, dependent on the local availability
of molecules such as VEGF, the destabilizing action of Ang2 can
either enhance or decrease local blood vessel formation.
Role in Ocular Neovascular Disease

While both Ang1 and Ang2 have been found in association with
VEGF in proliferative membranes from patient eyes with ocular
neovascular diseases,253,270 Ang2 was particularly localized in
highly vascularized areas of the membranes.270 In patient eyes
with proliferative DR, vitreous levels of Ang2 were signicantly
higher than in nondiabetic patients.125 In contrast, while high
levels of Ang2 were detected in eyes with nonproliferative DR
with macular edema during pars plana vitrectomy, Ang2 was
undetectable in eyes with proliferative disease.271 The reason for
these discrepant ndings is unclear but may be related to the
possibility that those undergoing vitrectomy were more likely to
have well-established lesions.
The effects of Ang2 in the eye have been found to be dependent on VEGF in some instances. For example, in a micropocket assay of corneal neovascularization, Ang2 administered
in conjunction with VEGF led to longer vessels together with
enhanced sprouting.255 Furthermore, fusion peptides that
inhibited the interaction of Ang2 and Tie2 prevented VEGFstimulated corneal neovascularization.272 An aptamer specic
for Ang2 also inhibited broblast growth factor-induced neovascularization in a similar model.273
More detailed studies of the relationship of Ang2 expression
have provided evidence for complex interactions with VEGF in
ocular neovascularization, with implications for possible therapeutic approaches.274,275 In transgenic mice with inducible
expression of Ang2 and VEGF, induction of Ang2 expression in
the rst 2 weeks after birth led to an increased density of retinal
capillaries that had normalized by postnatal day 18, suggesting
that Ang2 expression does not affect mature retinal vessels.274
In mice in which ischemia was induced by transient exposure
to high oxygen between postnatal days 7 and 12, induced
expression of Ang2 had divergent impacts depending on the
time of onset. Between postnatal days 12 and 17, when VEGF
levels were high, induction of Ang2 dramatically increased
retinal neovascularization; if this induced expression was
delayed until P20, when retinal ischemia was less intense and
VEGF levels were concomitantly lower, regression of the
neovascularization was intensied. Finally, in mice with sustained, low-level VEGF expression in photoreceptors, laserinduced CNV was suppressed by elevated expression of Ang2.
The investigators concluded that while more mature vessels are
not affected by Ang2 expression, its elevation may lead nascent
vessels either to proliferate, or to regress, depending on the ratio
of VEGF and Ang2 concentrations.274 From these ndings it
was proposed that elevation of Ang2, in conjunction with

inactivation of VEGF, could be a useful therapeutic approach for
treating ocular neovascularization.275
CONCLUSIONS
PDZ
P
Ephrin-A
Ephrin-B
The complex roles of Ang1 and Ang2 and their receptor Tie2
have only recently been appreciated in ocular neovascular
diseases such as AMD and DR. Evidence from preclinical and
clinical studies suggests that Ang1 is largely inhibitory to the
development of neovascularization while Ang2 may promote or
inhibit it depending on the local concentration of VEGF. There
is still much to be learned about the mechanisms for these
effects and how they may best be applied to the treatment of
pathologic ocular neovascularization.
EPHRINS
P
Key Features
Ephrins are ligands for the Eph class of receptors; both are
membrane-bound proteins
Ephrins are divided into two subclasses, ephrinA and ephrinB,
as are the receptors EphA and EphB; there are multiple
members in each of these subclasses
Ephrin/Eph interactions have a wide range of functions in
morphogenesis and development of neural networks,
embryonic vascular development and postnatal angiogenesis;
these interactions are also are a primary determinant of
venous/arterial identity
Recent evidence supports a role for eprhin/Eph interactions in
pathological forms of ocular neovascularization, suggesting that
modulating these interactions may offer therapeutic options
INTRODUCTION
Among the proteins that play an active role in angiogenesis,
VEGFs stimulate endothelial cell proliferation and migration,
Angs mediate blood vessel plasticity and maturation, and
ephrins are involved in vessel patterning.276 Ephrins are ligands
for the Eph class of receptors, whose name originally derived
from the identication and molecular cloning of a novel kinase
receptor gene from an erythropoietin-producing human hepatocellular carcinoma line.277 Ephrins are divided into two
subclasses, ephrinA and ephrinB, as are the receptors EphA and
EphB; there are multiple members in each of these subclasses
(reviewed by Zhang and Hughes278). In general, interactions
between A and B subclasses are limited, while multiple ligands
are capable of binding to multiple receptors within a subclass.278
At least one interaction between an EphB receptor and an
ephrinA ligand has been reported, however.279
Ephrins are membrane-bound proteins; ephrinAs are tethered
to the cellular membrane and lack any cytoplasmic portion
while ephrinBs are transmembrane proteins that possess a
cytoplasmic signaling domain (Fig. 31.9) (reviewed by Doudelet
and Pasquale280). Thus, ephrinB/EphB interactions may produce
both forward or reverse signaling, depending on the direction
(reviewed by Davy and Soriano281).
Ephs and ephrins have a wide range of functions in morphogenesis, in the development of neural networks, and in
embryonic and postnatal angiogenesis (reviewed by Palmer and
Klein282) and also are involved in controlling trafcking of
circulating cells within the vascular system (reviewed by Pfaff et
al239). In this section, the involvement of Eph/ephrin interactions in promotion of angiogenesis, particularly pathologic
forms of ocular neovascularization, will be discussed.
EphA
EphB
P
PDZ
Ligand binding
Kinase
Cysteine-rich
SAM
Fibronectin type III
CHAPTER 31
SH2
FIGURE 31.9. The structure of ephrins and Eph receptors. Both

ephrins and Eph receptors are membrane-bound proteins. Whereas
ephrinAs are tethered to the cellular membrane, ephrinBs have
transmembrane and cytoplasmic signaling domains. Binding of
ephrins to Eph receptors leads to receptor clustering and subsequent
autophosphorylation of multiple tyrosine residues, providing docking
sites for src-homology domain-containing downstream effectors. The
carboxyl terminus of both Eph receptors contains a sterile alpha motif
(SAM) and a PDZ domain (shown here for EphA, but these also apply
to EphB), which promote receptor clustering after ligand binding.
Adapted from Dodelet VC, Pasquale EB: Eph receptors and ephrin ligands:
embryogenesis to tumorigenesis. Oncogene 2000; 19:56145619.
EPHRINA
Although ephrinA1 is widely expressed in embryonic vasculature,283 a clear role for ephrinA/EphA interactions in
embryonic angiogenesis has not been established. In contrast,
ephrinA/EphA interactions have been found to play an active
role in regulating postnatal angiogenesis, as shown by in vitro
and in vivo studies. Pulmonary microvascular endothelial cells
from adult EphA2-decient mice had normal proliferation and
survival but failed to migrate and form capillary-like structures
in response to ephrinA1 stimulation; assembly into structures
in these cells was restored by overexpressing EphA2.284 EphA2
receptor phosphorylation was shown to be critical for migration
of these cells. Furthermore, EphA2-decient mice had impaired
angiogenesis in response to implanted ephrinA1-impreganted
sponges, demonstrating that EphA2 is a regulator of
angiogenesis in adult endothelial cells.284
327
SECTION 4
EphrinA/EphA Interactions Modulate the

Angiogenic Effects of VEGF
EphrinA/EphA interactions are involved in modulating the
induction of angiogenesis by VEGF. Stimulation of EphA
receptor signaling in bovine retinal endothelial cells by
ephrinA1-Fc inhibited VEGF-induced phosphorylation of
VEGFR-2 and subsequent endothelial cell migration and tubule
formation.285 In other work, a soluble EphA2-Fc fusion protein
(which reportedly blocks the interaction of EphA2 with its
ligands) was found to inhibit VEGF-dependent human
endothelial cell migration, sprouting, and survival in vitro but
did not affect endothelial cell proliferation.286 VEGF was found
to induce expression of ephrinA1 on endothelial cells; since the
levels of ephrinA1 expression directly correlated with the level
of phosphorylation of EphA2, it is likely that EphA2 activation
is involved in these effects.286 Hypoxia also was reported to
upregulate the expression of EphA2 and ephrinA1 in a murine
dermal model.287
In subsequent work using bovine retinal endothelial cells,
soluble EphA2-Fc inhibited both ephrinA1- and VEGF-induced
migration and tubule formation.288 VEGF-induced endothelial
cell tubule formation was impaired using cells from EphA2decient mice, providing evidence that EphA2 stimulation is
necessary for maximal induction of neovascularization by
VEGF.288 The mechanism by which soluble EphA2-Fc regulates
VEGF-induced angiogenesis is not clear; an indirect effect on
VEGF signaling is unlikely in that no effects on expression or
phosphorylation of VEGFR-2 in retinal endothelial cells were
found nor does EphA2-Fc affect VEGF-induced endothelial cell
proliferation. An alternate possibility is that VEGF and Eph
may regulate two distinct pathways.288
EphA/ephrinA in Ocular Neovascularization

EphA/ephrinA interactions have been demonstrated to be involved in pathologic ocular neovascularization as demonstrated
in rodent models. EphA2-Fc-inhibited VEGF-induced angiogenesis (but not that induced by basic broblast growth factor)
in a model in which pellets impregnated with ephrinA2, VEGF,
or basic broblast growth factor were implanted into mouse
corneas to provoke neovascularization.286 Intravitreal administration of EphA2-Fc also reduced the severity of pathologic
neovascularization in a rat model of retinopathy of prematurity
while having no effect on normal retinal vascular development.288
Finally, other rodent studies demonstrated that intravitreal
injection of ephrinA1-Fc inhibited both VEGF-induced
neovascularization and vascular permeability.285
EPHRINB
Essential for Development of Embryonic
Vasculature
328
The importance of ephrinB/EphB interactions in embryonic

vascular development was demonstrated by the early embryonic
lethality of knockout mice lacking ephrinB2 or EphB4.289,290
Since these two knockouts yield identical lethal phenotypes, it
suggests that EphB4 is the major receptor for ephrinB2 in
cardiovascular development.290 The EphB4/ephrinB2 pair is
believed to be a primary determinant of venous/arterial identity
in that EphB4 is preferentially expressed on veins290 and
ephrinB2 on arteries.289 Targeted disruption of ephrinB2
prevents proper remodeling of veins into branched structures
and disrupts the remodeling of arteries, providing evidence that
reciprocal interactions between arterial and venous endothelial
cells may be required for angiogenesis.289 In addition to EphB4,
ephrinB1 and EphB3 were found to be expressed on embryonic
veins while arteries expressed ephrinB1 in addition to ephrinB2;
aortic arches expressed ephrinB1, ephrinB2, and EphB3.291 The
relative contribution of these alternate receptors to embryonic

angiogenesis is not well characterized.
Reverse Signaling is Required for Angiogenesis

Mice that express ephrinB2 lacking the cytoplasmic domain
also demonstrated early embryo lethality due to vascular defects
although the cytoplasmic domain was not found to be required
for activation of EphB4, suggesting that reverse signaling
through ephrinB2 is required for proper embryonic vascular
development.292 An alternate possibility is that ephrinB2 ligand
clustering, which is dependent on the cytoplasmic domain, may
be required for its full activation as has been demonstrated for
ephrinB1.293 EphrinB ligands have a conserved carboxyterminal
sequence that serves as a binding site for PDZ domains,294
named for the rst three proteins found to contain this motif
(PSD95, DLG, and ZO-1) and which are important in
clustering and anchoring transmembrane proteins.295
EphB/ephrinB Interactions in Endothelial Cell

Migration and Capillary Formation
The effects of the EphB4/ephrinB2 pair on endothelial cells have
been the most fully characterized of the B subclass, with
forward signaling generally resulting in decreased proliferation
and migration of EphB4-expressing cells and reverse signaling
promoting increased proliferation and migration of ephrinB2expressing cells.296 Consistent with this generalization, coating
adhesive culture dishes with ephrinB2-Fc, which stimulates
EphB4, completely blocked adhesion of human umbilical vein
endothelial cells to culture dishes.297 In addition, soluble
ephrinB2-Fc inhibited cell migration, VEGF-driven chemotaxis,
capillary-like network formation, and sprouting angiogenesis.
In turn, soluble EphB4-Fc, which would be expected to stimulate ephrinB2 signaling, was proadhesive and stimulated endothelial cell migration and angiogenesis.297 Co-mingling between
endothelial cells expressing either EphB4 or ephrinB2 showed
that forward signaling through EphB4 restricts intermingling of
cells; this separation of EphB4 expressing cells from ephrinB2
cells may be important in segregating arteries from veins.297
The stimulating effects of ephrinB2-Fc on endothelial cell
migration were demonstrated in both human umbilical vein
endothelial cells298 and in cloned human mesenteric microvascular endothelial cells.299 However, contrasting ndings were
reported for proliferation, which was not stimulated in human
umbilical vein endothelial cells298 but was stimulated
moderately in microvascular endothelial cells.299 EphB4 and
ephrinB2 were shown to be upregulated by hypoxia in a murine
dermal model, which may contribute to amplication of their
angiogenic effects under conditions of ischemia.287
The contributions of other members of the EphB class to
angiogenesis are not well studied. EphB1-Fc has been shown to
stimulate ephinB1 phosphorylation on human microvascular
endothelial cells and to promote their migration and integrinmediated attachment.300 Also, multimeric ephrinB1-Fc promoted attachment and tubule formation in human renal
microvascular endothelial cells and resulted in activation of
EphB1 receptors.293 Thus, in these systems, EphB1/ephrinB1
interactions appear similar to those of EphB2/ephrinB2.
EphB/ephrin B Interactions in Ocular

Neovascularization
Both ephrinB2-Fc298 and ephrinB1-Fc300 have been demonstrated to promote corneal angiogenesis following implantation
of slow-release pellets into corneal micropockets. In other
experiments using a similar model but involving transgenic
mice that are heterozygous for ephrinB2 and the reporter gene
b-galactosidase, the administration of either ephrinB2 or VEGF
stimulated corneal neovascularization.301 Furthermore, EphB4
CONCLUSIONS
Ephrin/Eph interactions are critical for embryonic vascular
development and contribute to postnatal angiogenesis as well.
They have been found to be important in modulating
angiogenic responses to VEGF through both forward and reverse
signaling. Recent evidence documents that ephrin/Eph
interactions also are involved in pathologic forms of ocular
neovascularization in adults, suggesting that modulating these
interactions may offer therapeutic options. Since ephrins/Ephs
are involved in a myriad of other functions, including retinal
axon development304 and the trafcking of immune cells,239 the
consequences of such interventions cannot easily be predicted.
NOTCH
Key Features
The Notch signaling pathway is essential for cell fate

determination and pattern formation in a wide variety of tissues
In mammals, the Notch 1 Delta-like ligand 4 (Dll4) is especially
important for arterial patterning in the embryo, while its
expression in the adult is elevated in endothelial cells during
physiological and pathological angiogenesis
Preclinical ndings suggest that interference with Dll4/Notch
signaling may prove to be a useful strategy for inhibiting
pathological angiogenesis
INTRODUCTION
Notch is a 300 kDa transmembrane receptor305,306 for which a
mutated phenotype involving notches on margins of the wing
was rst described in Drosophila in 1917.307 Notch has since
been found in all animal phyla examined, and is expressed in a
wide variety of tissues where it acts to determine cell fates and
to regulate pattern formation (reviewed by Lai et al308). Ligand
activation of Notch leads to its being cleaved in two sequential
proteolytic steps, releasing an intracellular domain which
translocates to the nucleus and leads to activation of Notchtargeted genes.308 There are four Notch genes in mammals
(Notch 1 through 4) and groups of ligands for Notch have been
identied in different animal phyla. Following the Drosophila
nomenclature, for which the Delta gene is a major ligand,
mammalian versions include the Delta-like ligand (DLL) series
as well as Jagged1 and Jagged2.308 In keeping with its wide range
of tissue expression, the Notch signaling pathway is important
in numerous processes including liver and kidney development,
somatogenesis, cardiovascular development, neurogenesis and
T-cell differentiation;308,309 in addition, aberrant Notch signaling has been implicated in development of various cancers.310
NOTCH/DELTA SIGNALING IN ANGIOGENESIS

Studies of Notch signaling in mammalian vasculature development have focused on the ligand Delta-like 4 (Dll4). Like VEGF,
gene knockout of only one Dll4 gene leads to embryonic
lethality311,312 although expression of the heterozygous lethality
is dependent on genetic background.313 During embryogenesis,
expression of Dll4 is restricted to the arterial endothelium314
where it plays an essential role in arterial patterning311313 while
expression in the adult is seen in many adult tissues with a high
proportion of endothelial cell types.314 Expression is elevated in
capillaries during the physiological angiogenesis that
accompanies ovarian cycling;314 in addition, Dll4 is expressed
in the endothelium of cancerous tumors.314 Like VEGF, its
expression in cultured endothelial cells is upregulated by
hypoxia.314
Research into the mechanism of Dll4-Notch signaling in
vascular development is still in its early stages, but already has
revealed interactions with other key regulators of angiogenesis;
for example, endothelial cell expression of Dll4 in vitro is
upregulated by VEGF.315,316 Several lines of evidence suggest
that interference with Notch signaling may be a useful
antiangiogenic strategy. Overexpression of an inhibitor of
Notch signaling can partially inhibit vascular development in
collagen gels,315 while overexpression of Dll4 in endothelial
cells can lead to reduced expression of VEGFR-2, as well as to
reduced migratory and proliferative responses to VEGF.317 In
addition, inhibition of venous endothelial cell migration and
differentiation can be effected either by pharmacological
inhibitors or by excess soluble Dll4;318 this latter nding may
reflect the need for cellcell contact in Notch-mediated
signaling. While the relationship of these model systems to the
processes that mediate angiogenesis in physiological contexts
remains to be established, these studies suggest that several
possible strategies are potentially available for inhibiting
Dll4-Notch signaling as a means of preventing pathological
angiogenesis.
CHAPTER 31
mRNA expression was signicantly upregulated in the corneal

tissues after 3 days. Although the total length of the neovasculature produced by these agents was similar, the extent of the
arterial vessels induced by ephrinB2 was signicantly less than
that induced by VEGF. These ndings suggest that ephrinB2
induces venous rather than arterial angiogenesis and is consistent with a mechanism of forward signaling through EphB4.301
EphB4/ephrinB2 interactions also have been investigated in
retinal models of ocular neovascularization. In a rat model,
soluble EphB4 reduced the extent of laser-induced CNV and
fluorescein leakage.302 These results suggest that the mechanism
for these effects may involve blocking of ephrinB2 activation by
soluble EphB4 or reverse inhibition of EphB4 signaling and
downstream pathways.302 In a mouse model of oxygen-induced
retinopathy, intravitreal injection of ephrinB2-Fc or EphB4-Fc
decreased formation of pathologic retinal vascular tufts and of
neovascular nuclei anterior to the inner limiting membrane;
supercial and deep vascular beds were not affected.296
Finally, studies in which proliferative membranes were
isolated from the eyes of patients with proliferative DR or
retinopathy of prematurity demonstrated that ephrinB2, EphB2,
and EphB3 were expressed on broproliferative membranes but
not EphB4.303 These molecules were detected on both vascular
endothelial cells and stromal mesenchymal cells. The detection
of ephrinB2 on these membranes may indicate a difference
between embryonic and pathologic angiogenesis, in that its
expression is restricted to arteries during embryonic
angiogenesis. The nding that EphB4 was not detected in these
tissues suggests that the lack of expression of EphB4 may be a
contributing factor to the disorganization of neovasculature in
proliferative membranes.303
329
SECTION 4
REFERENCES
330
1. Folkman J, Shing Y: Angiogenesis. J Biol

Chem 1992; 267:1093110934.
2. Risau W: Mechanisms of angiogenesis.
Nature 1997; 386:671674.
3. Michaelson I: The mode of development of
the vascular system of the retina, with
some observations on its signicance for
certain retinal disease. Trans Ophthalmol
Soc UK 1948; 68:137180.
4. Ashton N: Retinal vascularization in health
and disease: Proctor Award lecture of the
association for research in ophthalmology.
5. Wise GN: Retinal neovascularization. Trans
Am Acad Ophthalmol Otolaryngol 1956;
54:729826.
6. Folkman J, Haudenschild C: Angiogenesis
by capillary endothelial cells in culture.
Trans Ophthalmol Soc UK 1980;
100:346353.
7. Boyden S: The chemotactic effect of
mixtures of antibody and antigen on
polymorphonuclear leucocytes. J Exp Med
1962; 115:453466.
8. Montesano R, Orci L: Tumor-promoting
phorbol esters induce angiogenesis in vitro.
Cell 1985; 42:469477.
9. Brown KJ, Maynes SF, Bezos A, et al: A
novel in vitro assay for human angiogenesis.
Lab Invest 1996; 75:539555.
10. Montesano R, Vassalli JD, Baird A, et al:
Basic broblast growth factor induces
angiogenesis in vitro. Proc Natl Acad Sci
USA 1986; 83:72977301.
11. Pepper MS, Ferrara N, Orci L,
Montesano R: Leukemia inhibitory factor
(LIF) inhibits angiogenesis in vitro. J Cell
Sci 1995; 108(Pt 1):7383.
12. Pepper MS, Montesano R, Vassalli JD, Orci
L: Chondrocytes inhibit endothelial sprout
formation in vitro: evidence for involvement
of a transforming growth factor-beta. J Cell
Physiol 1991; 146:170179.
13. Vailhe B, Ronot X, Lecomte M, et al:
Description of an in vitro angiogenesis
model designed to test antiangiogenic
molecules. Cell Biol Toxicol 1996;
12:341344.
14. Zimrin AB, Villeponteau B, Maciag T:
Models of in vitro angiogenesis: endothelial
cell differentiation on brin but not matrigel
is transcriptionally dependent. Biochem
Biophys Res Commun 1995; 213:630638.
15. Auerbach R, Kubai L, Knighton D, Folkman
J: A simple procedure for the long-term
cultivation of chicken embryos. Dev Biol
1974; 41:391394.
16. Nguyen M, Shing Y, Folkman J:
Quantitation of angiogenesis and
antiangiogenesis in the chick embryo
chorioallantoic membrane. Microvasc Res
1994; 47:3140.
17. Ribatti D, Vacca A, Roncali L,
Dammacco F: The chick embryo
chorioallantoic membrane as a model for in
vivo research on angiogenesis. Int J Dev
Biol 1996; 40:11891197.
18. Moses MA, Sudhalter J, Langer R:
Identication of an inhibitor of
neovascularization from cartilage. Science
1990; 248:14081410.
19. Kenyon BM, Voest EE, Chen CC, et al:
A model of angiogenesis in the mouse
cornea. Invest Ophthalmol Vis Sci 1996;
37:16251632.
20. Li WW, Grayson G, Folkman J,

DAmore PA: Sustained-release endotoxin.
A model for inducing corneal
Sci 1991; 32:29062911.
21. Loughman MS, Chatzistefanou K, Gonzalez
EM, et al: Experimental corneal
neovascularisation using sucralfate and
basic broblast growth factor. ANZ
J Ophthalmol 1996; 24:289295.
22. Fournier GA, Lutty GA, Watt S, et al:
A corneal micropocket assay for
angiogenesis in the rat eye. Invest
23. Muthukkaruppan V, Auerbach R:
Angiogenesis in the mouse cornea.
Science 1979; 205:14161418.
24. Benelli U, Ross JR, Nardi M, Klintworth GK:
Corneal neovascularization induced by
xenografts or chemical cautery. Inhibition
by cyclosporin A. Invest Ophthalmol Vis Sci
1997; 38:274282.
25. Amano S, Rohan R, Kuroki M, et al:
Requirement for vascular endothelial
growth factor in wound- and inflammationrelated corneal neovascularization. Invest
26. Danis RP, Yang Y, Massicotte SJ, Boldt HC:
Preretinal and optic nerve head
neovascularization induced by
photodynamic venous thrombosis in
domestic pigs. Arch Ophthalmol 1993;
111:539543.
27. Miller JW, Adamis AP, Shima DT, et al:
Vascular endothelial growth factor/vascular
permeability factor is temporally and
spatially correlated with ocular
angiogenesis in a primate model. Am J
Pathol 1994; 145:574584.
28. Saito Y, Park L, Skolik SA, et al:
Experimental preretinal neovascularization
by laser-induced venous thrombosis in rats.
Curr Eye Res 1997; 16:2633.
29. Pournaras CJ, Tsacopoulos M,
Strommer K, et al: Experimental retinal
branch vein occlusion in miniature pigs
induces local tissue hypoxia and
vasoproliferative microangiopathy.
30. Danis RP, Bingaman DP, Yang Y, Ladd B:
Inhibition of preretinal and optic nerve head
neovascularization in pigs by intravitreal
triamcinolone acetonide. Ophthalmology
1996; 103:20992104.
31. Hamilton AM, Marshall J, Kohner EM,
Bowbyes JA: Retinal new vessel formation
following experimental vein occlusion. Exp
Eye Res 1975; 20:493497.
32. Packer AJ, Gu XQ, Servais EG, Hayreh SS:
Primate model of neovascular glaucoma.
Int Ophthalmol 1986; 9:121127.
33. Miller JW: Vascular endothelial growth
factor and ocular neovascularization. Am J
Pathol 1997; 151:1323.
34. Miller JW, Stinson WG, Folkman J:
Regression of experimental iris
neovascularization with systemic alphainterferon. Ophthalmology 1993; 100:914.
35. Adamis AP, Shima DT, Tolentino MJ, et al:
Inhibition of vascular endothelial growth
factor prevents retinal ischemia-associated
iris neovascularization in a nonhuman
primate. Arch Ophthalmol 1996; 114:6671.
36. Reynaud X, Dorey CK: Extraretinal
neovascularization induced by hypoxic
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
episodes in the neonatal rat. Invest

Penn JS, Tolman BL, Lowery LA:
Variable oxygen exposure causes preretinal
neovascularization in the newborn rat.
34:576585.
McLeod DS, Crone SN, Lutty GA:
Vasoproliferation in the neonatal dog model
of oxygen-induced retinopathy. Invest
Chan-Ling T, Gock B, Stone J: The effect
of oxygen on vasoformative cell division.
Evidence that physiological hypoxia is the
stimulus for normal retinal vasculogenesis.
36:12011214.
Smith LE, Wesolowski E, McLellan A, et al:
Oxygen-induced retinopathy in the mouse.
Patz A: Clinical and experimental studies
on retinal neovascularization. XXXIX
Edward Jackson Memorial Lecture. Am J
Ophthalmol 1982; 94:715743.
Archer DB, Gardiner TA: Morphologic
fluorescein angiographic, and light
microscopic features of experimental
choroidal neovascularization. Am J
Ophthalmol 1981; 91:297311.
Ryan S: Subretinal neovascularization after
argon laser photocoagulation. Graefes Arch
Klin Ophthalmol 1980; 215:2942.
Ryan SJ: Subretinal neovascularization.
Natural history of an experimental model.
Husain D, Ryan A, Cuthbertson R, et al:
Vascular endothelial growth factor (VEGF)
expression is correlated with choroidal
neovascularization in a monkey model.
Invest Ophthalmol Vis Sci 1997; 38:s501.
Corjay M, Husain D, Stoltenborg J, et al:
Alpha V beta 3, alpha V beta 5, and
ostopontin immunostaining in experimental
choroidal neovascularization in the monkey.
Invest Ophthalmol Vis Sci 1997; 38:s965.
Okamoto N TT, Hackett SF, Ozaki H, et al:
Animal model. Transgenic mice with
increased expression of vascular
endothelial growth factor in the retina. a
new model of intraretinal and subretinal
neovascularization. Am J Pathol 1997;
151:281291.
Danis RP, Yang Y: Microvascular
retinopathy in the Zucker diabetic fatty rat.
34:23672371.
Kern TS, Engerman RL: A mouse model of
diabetic retinopathy. Arch Ophthalmol
1996; 114:986990.
Buchi ER, Kurosawa A, Tso MO:
Retinopathy in diabetic hypertensive
monkeys: a pathologic study. Graefes Arch
Papachristodoulou D, Heath H, Kang SS:
The development of retinopathy in sucrosefed and streptozotocin-diabetic rats.
Diabetologia 1976; 12:367374.
Huang S, Khosrof S, Koletsky R, et al:
Characterization of retinal vascular
abnormalities in lean and obese
spontaneously hypertensive rats. Clin
Exp Pharmacol Physiol 1995; 22(Suppl
l):S129S131.
Kador PF, Takahashi Y, Wyman M,
Ferris F 3rd: Diabetes-like proliferative
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
retinal changes in galactose-fed dogs. Arch

Ophthalmol 1995; 113:352354.
Ferrara N, Davis-Smyth T: The biology of
vascular endothelial growth factor. Endocr
Rev 1997; 18:425.
Senger DR, Connolly DT, Van de Water L,
et al: Purication and NH2-terminal amino
acid sequence of guinea pig tumorsecreted vascular permeability factor.
Cancer Res 1990; 50:17741778.
Keck PJ, Hauser SD, Krivi G, et al: Vascular
permeability factor, an endothelial cell
mitogen related to PDGF. Science 1989;
246:13091312.
Leung DW, Cachianes G, Kuang WJ, et al:
Vascular endothelial growth factor is a
secreted angiogenic mitogen. Science
1989; 246:13061309.
Carmeliet P, Ferreira V, Breier G, et al:
Abnormal blood vessel development and
lethality in embryos lacking a single VEGF
allele. Nature 1996; 380:435439.
Ferrara N, Carver-Moore K, Chen H, et al:
Heterozygous embryonic lethality induced
by targeted inactivation of the VEGF gene.
Nature 1996; 380:439442.
Miquerol L, Langille BL, Nagy A: Embryonic
development is disrupted by modest
increases in vascular endothelial growth
factor gene expression. Development 2000;
127:39413946.
Ferrara N: Vascular endothelial growth
factor: basic science and clinical progress.
Endocr Rev 2004; 25:581611.
Matsumoto T, Claesson-Welsh L: VEGF
receptor signal transduction. Sci STKE
2001; 2001:RE21.
Barleon B, Sozzani S, Zhou D, et al:
Migration of human monocytes in response
to vascular endothelial growth factor
(VEGF) is mediated via the VEGF receptor
flt-1. Blood 1996; 87:33363343.
Clauss M, Gerlach M, Gerlach H, et al:
Vascular permeability factor: a tumorderived polypeptide that induces
endothelial cell and monocyte procoagulant
activity, and promotes monocyte migration.
J Exp Med 1990; 172:15351545.
Ishida S, Usui T, Yamashiro K, et al:
VEGF164-mediated inflammation is
required for pathological, but not
physiological, ischemia-induced retinal
neovascularization. J Exp Med 2003;
198:483489.
Sakurai E, Anand A, Ambati BK, et al:
Macrophage depletion inhibits experimental
choroidal neovascularization. Invest
Sakurai E, Taguchi H, Anand A, et al:
Targeted disruption of the CD18 or ICAM-1
gene inhibits choroidal neovascularization.
44:27432749.
Hiratsuka S, Nakamura K, Iwai S, et al:
MMP9 induction by vascular endothelial
growth factor receptor-1 is involved in lungspecic metastasis. Cancer Cell 2002;
2:289300.
LeCouter J, Moritz DR, Li B, et al:
Angiogenesis-independent endothelial
protection of liver: role of VEGFR-1.
Science 2003; 299:890893.
Robinson CJ, Stringer SE: The splice
variants of vascular endothelial growth
factor (VEGF) and their receptors. J Cell Sci
2001; 114:853865.
Park JE, Keller GA, Ferrara N: The vascular
endothelial growth factor (VEGF) isoforms:
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
differential deposition into the subepithelial

extracellular matrix and bioactivity of
extracellular matrix-bound VEGF. Mol Biol
Cell 1993; 4:13171326.
Bates DO, Cui TG, Doughty JM, et al:
VEGF165b, an inhibitory splice variant of
vascular endothelial growth factor, is downregulated in renal cell carcinoma. Cancer
Res 2002; 62:41234131.
Woolard J, Wang WY, Bevan HS, et al:
VEGF165b, an inhibitory vascular
endothelial growth factor splice variant:
mechanism of action, in vivo effect on
angiogenesis and endogenous protein
expression. Cancer Res 2004;
64:78227835.
Perrin RM, Konopatskaya O, Qiu Y, et al:
Diabetic retinopathy is associated with a
switch in splicing from anti- to proangiogenic isoforms of vascular endothelial
growth factor. Diabetologia 2005;
48:24222427.
Oberg-Welsh C, Sandler S, Andersson A,
Welsh M: Effects of vascular endothelial
growth factor on pancreatic duct cell
replication and the insulin production of
fetal islet-like cell clusters in vitro. Mol Cell
Endocrinol 1997; 126:125132.
Sondell M, Lundborg G, Kanje M: Vascular
endothelial growth factor has neurotrophic
activity and stimulates axonal outgrowth,
enhancing cell survival and Schwann cell
proliferation in the peripheral nervous
system. J Neurosci 1999; 19:57315740.
Guerrin M, Moukadiri H, Chollet P, et al:
Vasculotropin/vascular endothelial growth
factor is an autocrine growth factor for
human retinal pigment epithelial cells
cultured in vitro. J Cell Physiol 1995;
164:385394.
Asahara T, Takahashi T, Masuda H, et al:
VEGF contributes to postnatal
neovascularization by mobilizing bone
marrow-derived endothelial progenitor
cells. EMBO J 1999; 18:39643972.
Lyden D, Hattori K, Dias S, et al: Impaired
recruitment of bone-marrow-derived
endothelial and hematopoietic precursor
cells blocks tumor angiogenesis and
growth. Nat Med 2001; 7:11941201.
Csaky KG, Baf JZ, Byrnes GA, et al:
Recruitment of marrow-derived endothelial
cells to experimental choroidal
neovascularization by local expression of
vascular endothelial growth factor. Exp Eye
Res 2004; 78:11071116.
Alon T, Hemo I, Itin A, et al: Vascular
endothelial growth factor acts as a survival
factor for newly formed retinal vessels and
has implications for retinopathy of
prematurity. Nat Med 1995; 1:10241028.
Pepper MS, Ferrara N, Orci L,
Montesano R: Vascular endothelial growth
factor (VEGF) induces plasminogen
activators and plasminogen activator
inhibitor-1 in microvascular endothelial
cells. Biochem Biophys Res Commun
1991; 181:902906.
Lamoreaux WJ, Fitzgerald ME, Reiner A,
et al: Vascular endothelial growth factor
increases release of gelatinase A and
decreases release of tissue inhibitor of
metalloproteinases by microvascular
endothelial cells in vitro. Microvasc Res
1998; 55:2942.
Lee S, Jilani SM, Nikolova GV, et al:
Processing of VEGF-A by matrix
metalloproteinases regulates bioavailability
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98.
99.
and vascular patterning in tumors. J Cell

Biol 2005; 169:681691.
Papapetropoulos A, Garcia-Cardena G,
Madri JA, Sessa WC: Nitric oxide
production contributes to the angiogenic
properties of vascular endothelial growth
factor in human endothelial cells. J Clin
Invest 1997; 100:31313139.
Uhlmann S, Friedrichs U, Eichler W, et al:
Direct measurement of VEGF-induced nitric
oxide production by choroidal endothelial
cells. Microvasc Res 2001; 62:179189.
Dulak J, Jozkowicz A, Dembinska-Kiec A,
et al: Nitric oxide induces the synthesis of
vascular endothelial growth factor by rat
vascular smooth muscle cells. Arterioscler
Thromb Vasc Biol 2000; 20:659666.
Plate KH, Breier G, Weich HA, Risau W:
potential tumour angiogenesis factor in
human gliomas in vivo. Nature 1992;
359:845848.
Shweiki D, Itin A, Soffer D, Keshet E:
Vascular endothelial growth factor induced
by hypoxia may mediate hypoxia-initiated
angiogenesis. Nature 1992; 359:843845.
Aiello LP, Northrup JM, Keyt BA, et al:
Hypoxic regulation of vascular endothelial
growth factor in retinal cells. Arch
Ophthalmol 1995; 113:15381544.
Shima D, Adamis A, Ferrara N, et al:
Hypoxic induction of endothelial cell growth
factors in retinal cells: identication and
characterization of vascular endothelial
growth factor (VEGF) as the mitogen. Mol
Med 1995; 1:182193.
Senger DR, Galli SJ, Dvorak AM, et al:
Tumor cells secrete a vascular permeability
factor that promotes accumulation of
ascites fluid. Science 1983; 219:983985.
Roberts WG, Palade GE: Increased
microvascular permeability and endothelial
fenestration induced by vascular
endothelial growth factor. J Cell Sci 1995;
108(Pt 6):23692379.
Antonetti DA, Barber AJ, Hollinger LA,
induces rapid phosphorylation of tight
junction proteins occludin and zonula
occluden 1. A potential mechanism for
vascular permeability in diabetic
retinopathy and tumors. J Biol Chem 1999;
274:2346323467.
Miyamoto K, Khosrof S, Bursell SE, et al:
Vascular endothelial growth factor (VEGF)induced retinal vascular permeability is
mediated by intercellular adhesion
molecule-1 (ICAM-1). Am J Pathol 2000;
156:17331739.
Joussen AM, Murata T, Tsujikawa A, et al:
Leukocyte-mediated endothelial cell injury
and death in the diabetic retina. Am J
Pathol 2001; 158:147152.
Kamba T, Tam BY, Hashizume H, et al:
VEGF-dependent plasticity of fenestrated
capillaries in the normal adult
microvasculature. Am J Physiol Heart Circ
Physiol 2006; 290:H560H576.
Inai T, Mancuso M, Hashizume H, et al:
Inhibition of vascular endothelial growth
factor (VEGF) signaling in cancer causes
loss of endothelial fenestrations, regression
of tumor vessels, and appearance of
basement membrane ghosts. Am J Pathol
2004; 165:3552.
Ryan AM, Eppler DB, Hagler KE, et al:
Preclinical safety evaluation of
rhuMAbVEGF, an antiangiogenic humanized
CHAPTER 31
331
100.
101.
102.
103.
104.
SECTION 4
105.
106.
107.
108.
109.
110.
111.
112.
113.
332
monoclonal antibody. Toxicol Pathol 1999;

27:7886.
Gerber HP, Vu TH, Ryan AM, et al: VEGF
couples hypertrophic cartilage remodeling,
ossication and angiogenesis during
endochondral bone formation. Nat Med
1999; 5:623628.
Fraser HM, Wilson H, Rudge JS,
Wiegand SJ: Single injections of vascular
endothelial growth factor trap block
ovulation in the macaque and produce a
prolonged, dose-related suppression of
ovarian function. J Clin Endocrinol Metab
2005; 90:11141122.
Nissen NN, Polverini PJ, Koch AE, et al:
Vascular endothelial growth factor mediates
angiogenic activity during the proliferative
phase of wound healing. Am J Pathol 1998;
152:14451452.
Deodato B, Arsic N, Zentilin L, et al:
Recombinant AAV vector encoding human
VEGF165 enhances wound healing. Gene
Ther 2002; 9:777785.
Liu MH, Jin H, Floten HS, et al: Vascular
endothelial growth factor-mediated,
endothelium-dependent relaxation in
human internal mammary artery. Ann
Thorac Surg 2002; 73:819824.
Kitamoto Y, Tokunaga H, Tomita K:
Vascular endothelial growth factor is an
essential molecule for mouse kidney
development: glomerulogenesis and
nephrogenesis. J Clin Invest 1997;
99:23512357.
Eremina V, Sood M, Haigh J, et al:
Glomerular-specic alterations of VEGF-A
expression lead to distinct congenital and
acquired renal diseases. J Clin Invest 2003;
111:707716.
Arsic N, Zacchigna S, Zentilin L, et al:
Vascular endothelial growth factor
stimulates skeletal muscle regeneration in
vivo. Mol Ther 2004; 10:844854.
Storkebaum E, Lambrechts D, Carmeliet P:
VEGF: once regarded as a specic
angiogenic factor, now implicated in
neuroprotection. Bioessays 2004;
26:943954.
Nishijima K, Ng Y-S, Zhong L, et al:
Vascular endothelial growth factor-A is a
survival factor for retinal neurons and a
critical neuroprotectant during adaptive
response to ischemic injury. Am J Pathol
2007; 171:5367.
Marneros AG, Fan J, Yokoyama Y, et al:
expression in the retinal pigment epithelium
is essential for choriocapillaris development
and visual function. Am J Pathol 2005;
167:14511459.
Blaauwgeers HG, Holtkamp GM, Rutten H,
et al: Polarized vascular endothelial growth
factor secretion by human retinal pigment
epithelium and localization of vascular
endothelial growth factor receptors on the
inner choriocapillaris. Evidence for a
trophic paracrine relation. Am J Pathol
1999; 155:421428.
Aiello LP, Avery RL, Arrigg PG, et al:
Vascular endothelial growth factor in ocular
fluid of patients with diabetic retinopathy
and other retinal disorders. N Engl J Med
1994; 331:14801487.
Tripathi RC, Li J, Tripathi BJ, et al:
Increased level of vascular endothelial
growth factor in aqueous humor of patients
with neovascular glaucoma.
114. Lashkari K, Hirose T, Yazdany J, et al:

Vascular endothelial growth factor and
hepatocyte growth factor levels are
differentially elevated in patients with
advanced retinopathy of prematurity.
Am J Pathol 2000; 156:13371344.
115. Adamis AP, Shima DT, Yeo KT, et al:
Synthesis and secretion of vascular
permeability factor/vascular endothelial
growth factor by human retinal pigment
epithelial cells. Biochem Biophys Res
Commun 1993; 193:631638.
116. Famiglietti EV, Stopa EG, McGookin ED,
et al: Immunocytochemical localization of
vascular endothelial growth factor in
neurons and glial cells of human retina.
Brain Res 2003; 969:195204.
117. Adamis AP, Miller JW, Bernal MT, et al:
Increased vascular endothelial growth
factor levels in the vitreous of eyes with
proliferative diabetic retinopathy. Am J
Ophthalmol 1994; 118:445450.
118. Kliffen M, Sharma HS, Mooy CM, et al:
Increased expression of angiogenic growth
factors in age-related maculopathy. Br J
Ophthalmol 1997; 81:154162.
119. Lopez PF, Sippy BD, Lambert HM, et al:
Transdifferentiated retinal pigment epithelial
cells are immunoreactive for vascular
endothelial growth factor in surgically
excised age-related macular degenerationrelated choroidal neovascular membranes.
120. Grossniklaus HE, Ling JX, Wallace TM,
et al: Macrophage and retinal pigment
epithelium expression of angiogenic
cytokines in choroidal neovascularization.
Mol Vis 2002; 8:119126.
121. Boulton M, Foreman D, Williams G,
McLeod D: VEGF localisation in diabetic
retinopathy. Br J Ophthalmol 1998;
82:561568.
122. Funatsu H, Yamashita H, Shimizu E, et al:
Relationship between vascular endothelial
growth factor and interleukin-6 in diabetic
retinopathy. Retina 2001; 21:469477.
123. Brooks HL Jr, Caballero S Jr, Newell CK,
et al: Vitreous levels of vascular endothelial
growth factor and stromal-derived factor 1
in patients with diabetic retinopathy and
cystoid macular edema before and after
intraocular injection of triamcinolone. Arch
Ophthalmol 2004; 122:18011807.
124. Patel JI, Tombran-Tink J, Hykin PG, et al:
Vitreous and aqueous concentrations of
proangiogenic, antiangiogenic factors and
other cytokines in diabetic retinopathy
patients with macular edema: implications
for structural differences in macular
proles. Exp Eye Res 2006; 82:798806.
125. Watanabe D, Suzuma K, Suzuma I, et al:
Vitreous levels of angiopoietin 2 and
vascular endothelial growth factor in
patients with proliferative diabetic
retinopathy. Am J Ophthalmol 2005;
139:476481.
126. Watanabe D, Suzuma K, Matsui S, et al:
Erythropoietin as a retinal angiogenic factor
in proliferative diabetic retinopathy. N Engl
J Med 2005; 353:782792.
127. Funatsu H, Yamashita H, Ikeda T, et al:
Angiotensin II and vascular endothelial
growth factor in the vitreous fluid of
patients with diabetic macular edema and
other retinal disorders. Am J Ophthalmol
2002; 133:537543.
128. Funatsu H, Yamashita H, Ikeda T, et al:
Vitreous levels of interleukin-6 and vascular
129.
130.
131.
132.
133.
134.
135.
136.
137.
138.
139.
140.
141.
142.
endothelial growth factor are related to

diabetic macular edema. Ophthalmology
2003; 110:16901696.
Funatsu H, Yamashita H, Sakata K, et al:
Vitreous levels of vascular endothelial
growth factor and intercellular adhesion
molecule 1 are related to diabetic macular
edema. Ophthalmology 2005; 112:806816.
Usui T, Ishida S, Yamashiro K, et al:
VEGF164(165) as the pathological isoform:
differential leukocyte and endothelial
responses through VEGFR1 and VEGFR2.
45:368374.
Williams B, Baker AQ, Gallacher B,
Lodwick D: Angiotensin II increases
vascular permeability factor gene
expression by human vascular smooth
muscle cells. Hypertension 1995;
25:913917.
Chua CC, Hamdy RC, Chua BH:
Upregulation of vascular endothelial growth
factor by angiotensin II in rat heart
endothelial cells. Biochim Biophys Acta
1998; 1401:187194.
Kijowski J, Baj-Krzyworzeka M, Majka M,
et al: The SDF-1-CXCR4 axis stimulates
VEGF secretion and activates integrins but
does not affect proliferation and survival in
lymphohematopoietic cells. Stem Cells
2001; 19:453466.
Ogata N, Wada M, Otsuji T, et al:
Expression of pigment epithelium-derived
factor in normal adult rat eye and
experimental choroidal neovascularization.
43:11681175.
Funatsu H, Yamashita H, Nakamura S,
et al: Vitreous levels of pigment epitheliumderived factor and vascular endothelial
growth factor are related to diabetic
macular edema. Ophthalmology 2006;
113:294301.
Zhang SX, Wang JJ, Gao G, et al: Pigment
epithelium-derived factor downregulates
vascular endothelial growth factor (VEGF)
expression and inhibits VEGF-VEGF
receptor 2 binding in diabetic retinopathy.
J Mol Endocrinol 2006; 37:112.
Dawson DW, Volpert OV, Gillis P, et al:
Pigment epithelium-derived factor: a potent
inhibitor of angiogenesis. Science 1999;
285:245248.
Ambati J, Ambati BK, Yoo SH, et al: Agerelated macular degeneration: etiology,
pathogenesis, and therapeutic strategies.
Surv Ophthalmol 2003; 48:257293.
Caldwell RB, Bartoli M, Behzadian MA,
and diabetic retinopathy:
pathophysiological mechanisms and
treatment perspectives. Diabetes Metab
Res Rev 2003; 19:442455.
Tolentino MJ, Miller JW, Gragoudas ES,
et al: Vascular endothelial growth factor is
sufcient to produce iris neovascularization
and neovascular glaucoma in a nonhuman
primate. Arch Ophthalmol 1996;
114:964970.
Tolentino MJ, Miller JW, Gragoudas ES,
et al: Intravitreous injections of vascular
endothelial growth factor produce retinal
ischemia and microangiopathy in an adult
primate. Ophthalmology 1996;
103:18201828.
Tolentino MJ, McLeod DS, Taomoto M,
et al: Pathologic features of vascular
endothelial growth factor-induced
143.
144.
145.
146.
147.
148.
149.
150.
151.
152.
153.
154.
155.
156.
retinopathy in the nonhuman primate.

Am J Ophthalmol 2002; 133:373385.
van Wijngaarden P, Coster DJ, Williams KA:
Inhibitors of ocular neovascularization:
promises and potential problems. JAMA
2005; 293:15091513.
Baf J, Byrnes G, Chan CC, Csaky KG:
Choroidal neovascularization in the rat
induced by adenovirus mediated expression
of vascular endothelial growth factor. Invest
Spilsbury K, Garrett KL, Shen WY, et al:
Overexpression of vascular endothelial
growth factor (VEGF) in the retinal pigment
epithelium leads to the development of
choroidal neovascularization. Am J Pathol
2000; 157:135144.
Schwesinger C, Yee C, Rohan RM, et al:
Intrachoroidal neovascularization in
transgenic mice overexpressing vascular
endothelial growth factor in the retinal
pigment epithelium. Am J Pathol 2001;
158:11611172.
Ohno-Matsui K, Hirose A, Yamamoto S,
et al: Inducible expression of vascular
endothelial growth factor in adult mice
causes severe proliferative retinopathy and
retinal detachment. Am J Pathol 2002;
160:711719.
Bhisitkul RB, Robinson GS, Moulton RS,
et al: An antisense oligodeoxynucleotide
against vascular endothelial growth factor
in a nonhuman primate model of iris
neovascularization. Arch Ophthalmol 2005;
123:214219.
Aiello LP, Pierce EA, Foley ED, et al:
Suppression of retinal neovascularization in
vivo by inhibition of vascular endothelial
growth factor (VEGF) using soluble VEGFreceptor chimeric proteins. Proc Natl Acad
Sci USA 1995; 92:1045710461.
Honda M, Sakamoto T, Ishibashi T, et al:
Experimental subretinal neovascularization
is inhibited by adenovirus-mediated soluble
VEGF/flt-1 receptor gene transfection: a
role of VEGF and possible treatment for
SRN in age-related macular degeneration.
Gene Ther 2000; 7:978985.
Agostini H, Boden K, Unsold A, et al:
A single local injection of recombinant
VEGF receptor 2 but not of Tie2 inhibits
retinal neovascularization in the mouse.
Curr Eye Res 2005; 30:249257.
Krzystolik MG, Afshari MA, Adamis AP,
et al: Prevention of experimental choroidal
neovascularization with intravitreal antivascular endothelial growth factor antibody
fragment. Arch Ophthalmol 2002;
120:338346.
Konopatskaya O, Churchill AJ, Harper SJ,
et al: VEGF165b, an endogenous
C-terminal splice variant of VEGF, inhibits
retinal neovascularization in mice. Mol Vis
2006; 12:626632.
Nozaki M, Sakurai E, Raisler BJ, et al: Loss
of SPARC-mediated VEGFR-1 suppression
after injury reveals a novel antiangiogenic
activity of VEGF-A. J Clin Invest 2006;
116:422429.
Penfold PL, Madigan MC, Gillies MC,
Provis JM: Immunological and aetiological
aspects of macular degeneration. Prog
Retin Eye Res 2001; 20:385414.
Kijlstra A, La Heij E, Hendrikse F:
Immunological factors in the pathogenesis
and treatment of age-related macular
degeneration. Ocul Immunol Inflamm 2005;
13:311.
157. Donoso LA, Kim D, Frost A, et al: The role

of inflammation in the pathogenesis of agerelated macular degeneration. Surv
Ophthalmol 2006; 51:137152.
158. Gardner TW, Antonetti DA, Barber AJ,
et al: Diabetic retinopathy: more than
meets the eye. Surv Ophthalmol 2002;
47(Suppl 2):S253S262.
159. Grossniklaus HE, Cingle KA, Yoon YD, et al:
Correlation of histologic 2-dimensional
reconstruction and confocal scanning laser
microscopic imaging of choroidal
neovascularization in eyes with age-related
maculopathy. Arch Ophthalmol 2000;
118:625629.
160. Espinosa-Heidmann DG, Suner IJ,
Hernandez EP, et al: Macrophage depletion
diminishes lesion size and severity in
experimental choroidal neovascularization.
44:35863592.
161. Bora PS, Sohn JH, Cruz JM, et al: Role of
complement and complement membrane
attack complex in laser-induced choroidal
neovascularization. J Immunol 2005;
174:491497.
162. Nozaki M, Raisler BJ, Sakurai E, et al:
Drusen complement components C3a and
C5a promote choroidal neovascularization.
103:23282333.
163. Seddon JM, Gensler G, Milton RC, et al:
Association between C-reactive protein and
age-related macular degeneration. JAMA
2004; 291:704710.
164. Kalayoglu MV, Bula D, Arroyo J, et al:
Identication of Chlamydia pneumoniae
within human choroidal neovascular
membranes secondary to age-related
macular degeneration. Graefes Arch Clin
Exp Ophthalmol 2005; 243:10801090.
165. Markomichelakis NN, Theodossiadis PG,
Skakis PP: Regression of neovascular
age-related macular degeneration following
infliximab therapy. Am J Ophthalmol 2005;
139:537540.
166. Skakis PP, Markomichelakis N,
Theodossiadis GP, et al: Regression of
sight-threatening macular edema in type 2
diabetes following treatment with the antitumor necrosis factor monoclonal antibody
infliximab. Diabetes Care 2005;
28:445447.
167. Fong DS, Aiello LP, Ferris FL 3rd, Klein R:
Diabetic retinopathy. Diabetes Care 2004;
27:25402553.
168. Miyamoto K, Hiroshiba N, Tsujikawa A,
Ogura Y: In vivo demonstration of
increased leukocyte entrapment in retinal
microcirculation of diabetic rats. Invest
169. Comer GM, Ciulla TA: Pharmacotherapy for
diabetic retinopathy. Curr Opin Ophthalmol
2004; 15:508518.
170. Kuroki M, Voest EE, Amano S, et al:
Reactive oxygen intermediates increase
vascular endothelial growth factor
expression in vitro and in vivo. J Clin Invest
1996; 98:16671675.
171. Lu M, Kuroki M, Amano S, et al: Advanced
glycation end products increase retinal
vascular endothelial growth factor
expression. J Clin Invest 1998;
101:12191224.
172. Punglia RS, Lu M, Hsu J, et al: Regulation
of vascular endothelial growth factor
expression by insulin-like growth factor I.
Diabetes 1997; 46:16191626.
173. Ishida S, Yamashiro K, Usui T, et al:

Leukocytes mediate retinal vascular
remodeling during development and vasoobliteration in disease. Nat Med 2003;
9:781788.
174. McLeod DS, Lefer DJ, Merges C, Lutty GA:
Enhanced expression of intracellular
adhesion molecule-1 and P-selectin in the
diabetic human retina and choroid. Am J
Pathol 1995; 147:642653.
175. Miyamoto K, Khosrof S, Bursell SE, et al:
Prevention of leukostasis and vascular
leakage in streptozotocin-induced diabetic
retinopathy via intercellular adhesion
molecule-1 inhibition. Proc Natl Acad Sci
USA 1999; 96:1083610841.
176. Qaum T, Xu Q, Joussen AM, et al: VEGFinitiated bloodretinal barrier breakdown in
early diabetes. Invest Ophthalmol Vis Sci
2001; 42:24082413.
177. Joussen AM, Poulaki V, Qin W, et al: Retinal
vascular endothelial growth factor induces
intercellular adhesion molecule-1 and
endothelial nitric oxide synthase expression
and initiates early diabetic retinal leukocyte
adhesion in vivo. Am J Pathol 2002;
160:501509.
178. Joussen AM, Poulaki V, Le ML, et al:
A central role for inflammation in the
pathogenesis of diabetic retinopathy.
FASEB J 2004; 18:14501452.
179. Joussen AM, Poulaki V, Mitsiades N, et al:
Suppression of Fas-FasL-induced
endothelial cell apoptosis prevents diabetic
bloodretinal barrier breakdown in a model
of streptozotocin-induced diabetes. FASEB
J 2003; 17:7678.
180. Harmey JH, Dimitriadis E, Kay E, et al:
Regulation of macrophage production of
vascular endothelial growth factor (VEGF)
by hypoxia and transforming growth
factor beta-1. Ann Surg Oncol 1998;
5:271278.
181. Barbera-Guillem E, Nyhus JK, Wolford CC,
secretion by tumor-inltrating macrophages
essentially supports tumor angiogenesis,
and IgG immune complexes potentiate the
process. Cancer Res 2002; 62:70427049.
182. Tsutsumi C, Sonoda KH, Egashira K, et al:
The critical role of ocular-inltrating
macrophages in the development of
choroidal neovascularization. J Leukoc Biol
2003; 74:2532.
183. Ishida S, Usui T, Yamashiro K, et al:
VEGF164 is proinflammatory in the diabetic
retina. Invest Ophthalmol Vis Sci 2003;
44:21552162.
184. Gragoudas ES, Adamis AP, Cunningham ET
Jr, et al: Pegaptanib for neovascular agerelated macular degeneration. N Engl J
Med 2004; 351:28052816.
185. Chakravarthy U, Adamis AP,
Cunningham ET, et al: Year 2 efcacy
results of two randomized, controlled
clinical trials of pegaptanib, an anti-VEGF
aptamer, for neovasxular age-related
macular degeneration. Ophthalmology
2006; 113:1508.
186. Rosenfeld PJ, Brown DM, Heier JS, et al:
Ranibizumab for neovascular age-related
macular degeneration. N Engl J Med 2006;
355:14191431.
187. Brown DM, Kaiser PK, Michels M, et al:
Ranibizumab versus verteporn for
neovascular age-related macular
degeneration. N Engl J Med 2006;
355:14321444.
CHAPTER 31
333
SECTION 4
334
188. DAmico DJ, Patel M, Adamis AP, et al:

Pegaptanib sodium for neovascular
age-related macular degeneration:
two-year safety results of the two
prospective, multicenter, controlled clinical
trials. Ophthalmology 2006; 113:9921001.
189. Cunningham ET Jr, Adamis AP, Altaweel M,
et al: A phase II randomized doublemasked trial of pegaptanib, an antivascular endothelial growth factor aptamer,
for diabetic macular edema.
Ophthalmology 2005; 112:17471757.
190. Adamis AP, Altaweel M, Bressler NM, et al:
Changes in retinal neovascularization after
pegaptanib (Macugen) therapy in diabetic
individuals. Ophthalmology 2006;
113:2328.
191. Wells JA III, Pegaptanib in Central Retinal
Vein Occlusion Study Group: Pegaptanib
sodium for treatment of macular edema
secondary to central retinal vein occlusion
(CRVO). Invest Ophthalmol Vis Sci 2006;
47:E-Abstract 4279.
192. Antoniades HN, Scher CD, Stiles CD:
Purication of human platelet-derived
growth factor. Proc Natl Acad Sci USA
1979; 76:18091813.
193. Fredriksson L, Li H, Eriksson U: The PDGF
family: four gene products form ve dimeric
isoforms. Cytokine Growth Factor Rev
2004; 15:197204.
194. Heldin CH, Westermark B: Mechanism of
action and in vivo role of platelet-derived
growth factor. Physiol Rev 1999;
79:12831316.
195. Dirks RP, Bloemers HP: Signals controlling
the expression of PDGF. Mol Biol Rep
1995; 22:124.
196. Heldin CH, Ernlund A, Rorsman C,
Ronnstrand L: Dimerization of B-type
platelet-derived growth factor receptors
occurs after ligand binding and is closely
associated with receptor kinase activation.
J Biol Chem 1989; 264:89058912.
197. Seifert RA, Hart CE, Phillips PE, et al: Two
different subunits associate to create
isoform-specic platelet-derived growth
factor receptors. J Biol Chem 1989;
264:87718778.
198. Hellstrom M, Kalen M, Lindahl P, et al: Role
of PDGF-B and PDGFR-beta in recruitment
of vascular smooth muscle cells and
pericytes during embryonic blood vessel
formation in the mouse. Development
1999; 126:30473055.
199. Kaminski WE, Lindahl P, Lin NL, et al: Basis
of hematopoietic defects in platelet-derived
growth factor (PDGF)-B and PDGF betareceptor null mice. Blood 2001;
97:19901998.
200. Leveen P, Pekny M, Gebre-Medhin S, et al:
Mice decient for PDGF B show renal,
cardiovascular, and hematological
abnormalities. Genes Dev 1994;
8:18751887.
201. Lindahl P, Johansson BR, Leveen P,
Betsholtz C: Pericyte loss and
microaneurysm formation in PDGF-Bdecient mice. Science 1997; 277:242245.
202. Betsholtz C: Role of platelet-derived
growth factors in mouse development.
Int J Dev Biol 1995; 39:817825.
203. Marx M, Perlmutter RA, Madri JA:
Modulation of platelet-derived growth
factor receptor expression in microvascular
endothelial cells during in vitro
angiogenesis. J Clin Invest 1994;
93:131139.
204. Beitz JG, Kim IS, Calabresi P,

Frackelton AR Jr: Human microvascular
endothelial cells express receptors for
platelet-derived growth factor. Proc Natl
Acad Sci USA 1991; 88:20212025.
205. Battegay EJ, Rupp J, Iruela-Arispe L, et al:
PDGF-BB modulates endothelial
proliferation and angiogenesis in vitro via
PDGF beta-receptors. J Cell Biol 1994;
125:917928.
206. Soyombo AA, DiCorleto PE: Stable
expression of human platelet-derived
growth factor B chain by bovine aortic
endothelial cells. Matrix association and
selective proteolytic cleavage by thrombin.
J Biol Chem 1994; 269:1773417740.
207. Lindblom P, Gerhardt H, Liebner S, et al:
Endothelial PDGF-B retention is required
for proper investment of pericytes in the
microvessel wall. Genes Dev 2003;
17:18351840.
208. Wang D, Huang HJ, Kazlauskas A,
Cavenee WK: Induction of vascular
endothelial growth factor expression in
endothelial cells by platelet-derived growth
factor through the activation of
phosphatidylinositol 3-kinase. Cancer Res
1999; 59:14641472.
209. Guo P, Hu B, Gu W, et al: Platelet-derived
growth factor-B enhances glioma
angiogenesis by stimulating vascular
endothelial growth factor expression in
tumor endothelia and by promoting
pericyte recruitment. Am J Pathol 2003;
162:10831093.
210. Kano MR, Morishita Y, Iwata C, et al:
VEGF-A and FGF-2 synergistically promote
neoangiogenesis through enhancement of
endogenous PDGF-B-PDGFRbeta
signaling. J Cell Sci 2005; 118:37593768.
211. Benjamin LE, Hemo I, Keshet E: A plasticity
window for blood vessel remodelling is
dened by pericyte coverage of the
preformed endothelial network and is
regulated by PDGF-B and VEGF.
Development 1998; 125:15911598.
212. Benjamin LE, Golijanin D, Itin A, et al:
Selective ablation of immature blood
vessels in established human tumors
follows vascular endothelial growth factor
withdrawal. J Clin Invest 1999;
103:159165.
213. Campochiaro PA, Hackett SF, Vinores SA,
et al: Platelet-derived growth factor is an
autocrine growth stimulator in retinal
pigmented epithelial cells. J Cell Sci 1994;
107(Pt 9):24592469.
214. Campochiaro PA, Glaser BM: Plateletderived growth factor is chemotactic for
human retinal pigment epithelial cells. Arch
Ophthalmol 1985; 103:576579.
215. Koyama N, Watanabe S, Tezuka M, et al:
Migratory and proliferative effect of
platelet-derived growth factor in rabbit
retinal endothelial cells: evidence of an
autocrine pathway of platelet-derived
growth factor. J Cell Physiol 1994; 158:16.
216. Kamiyama K, Iguchi I, Wang X, Imanishi J:
Effects of PDGF on the migration of rabbit
corneal broblasts and epithelial cells.
Cornea 1998; 17:315325.
217. Ikuno Y, Leong FL, Kazlauskas A:
Attenuation of experimental proliferative
vitreoretinopathy by inhibiting the plateletderived growth factor receptor. Invest
218. Ikuno Y, Kazlauskas A: An in vivo gene
therapy approach for experimental
219.
220.
221.
222.
223.
224.
225.
226.
227.
228.
229.
230.
231.
232.
233.
234.
proliferative vitreoretinopathy using the

truncated platelet-derived growth factor
alpha receptor. Invest Ophthalmol Vis Sci
2002; 43:24062411.
Enge M, Bjarnegard M, Gerhardt H, et al:
Endothelium-specic platelet-derived
growth factor-B ablation mimics diabetic
retinopathy. EMBO J 2002; 21:43074316.
Wilkinson-Berka JL, Babic S, De Gooyer T,
et al: Inhibition of platelet-derived growth
factor promotes pericyte loss and
angiogenesis in ischemic retinopathy.
Am J Pathol 2004; 164:12631273.
Seo MS, Okamoto N, Vinores MA, et al:
Photoreceptor-specic expression of
platelet-derived growth factor-B results in
traction retinal detachment. Am J Pathol
2000; 157:9951005.
Akiyama H, Kachi S, Silva RL, et al:
Intraocular injection of an aptamer that
binds PDGF-B: a potential treatment for
proliferative retinopathies. J Cell Physiol
2006; 207:407412.
Dell S, Peters S, Muther P, et al: The role of
PDGF receptor inhibitors and PI3-kinase
signaling in the pathogenesis of corneal
Sci 2006; 47:19281937.
Jo N, Mailhos C, Ju M, et al: Inhibition of
platelet-derived growth factor B signaling
enhances the efcacy of anti-vascular
endothelial growth factor therapy in
multiple models of ocular
neovascularization. Am J Pathol 2006;
168:20362053.
Dumont DJ, Yamaguchi TP, Conlon RA, et
al: Tek, a novel tyrosine kinase gene located
on mouse chromosome 4, is expressed in
endothelial cells and their presumptive
precursors. Oncogene 1992; 7:14711480.
Davis S, Aldrich TH, Jones PF, et al:
Isolation of angiopoietin-1, a ligand for the
TIE2 receptor, by secretion-trap expression
cloning. Cell 1996; 87:11611169.
Partanen J, Armstrong E, Makela TP, et al:
A novel endothelial cell surface receptor
tyrosine kinase with extracellular epidermal
growth factor homology domains. Mol Cell
Biol 1992; 12:16981707.
Maisonpierre PC, Suri C, Jones PF, et al:
Angiopoietin-2, a natural antagonist for
Tie2 that disrupts in vivo angiogenesis.
Science 1997; 277:5560.
Valenzuela DM, Grifths JA, Rojas J, et al:
Angiopoietins 3 and 4: diverging gene
counterparts in mice and humans. Proc
Natl Acad Sci USA 1999; 96:19041909.
Eklund L, Olsen BR: Tie receptors and their
angiopoietin ligands are context-dependent
regulators of vascular remodeling. Exp Cell
Res 2006; 312:630641.
Wong AL, Haroon ZA, Werner S, et al: Tie2
expression and phosphorylation in
angiogenic and quiescent adult tissues.
Circ Res 1997; 81:567574.
Sato TN, Tozawa Y, Deutsch U, et al:
Distinct roles of the receptor tyrosine
kinases Tie-1 and Tie-2 in blood vessel
formation. Nature 1995; 376:7074.
Suri C, Jones PF, Patan S, et al: Requisite
role of angiopoietin-1, a ligand for the TIE2
receptor, during embryonic angiogenesis.
Cell 1996; 87:11711180.
Moyon D, Pardanaud L, Yuan L, et al:
Selective expression of angiopoietin 1 and
2 in mesenchymal cells surrounding veins
and arteries of the avian embryo. Mech
Dev 2001; 106:133136.
235. Kim I, Kim HG, So JN, et al: Angiopoietin-1

regulates endothelial cell survival through
the phosphatidylinositol 3-Kinase/Akt
signal transduction pathway. Circ Res
2000; 86:2429.
236. Xu Y, Yu Q: Angiopoietin-1, unlike
angiopoietin-2, is incorporated into the
extracellular matrix via its linker peptide
region. J Biol Chem 2001;
276:3499034998.
237. Morisada T, Kubota Y, Urano T, et al:
Angiopoietins and angiopoietin-like
proteins in angiogenesis. Endothelium
2006; 13:7179.
238. Brindle NP, Saharinen P, Alitalo K: Signaling
and functions of angiopoietin-1 in vascular
protection. Circ Res 2006; 98:10141023.
239. Pfaff D, Fiedler U, Augustin HG: Emerging
roles of the Angiopoietin-Tie and the
ephrin-Eph systems as regulators of cell
trafcking. J Leukoc Biol 2006; 80:719726.
240. Kwak HJ, Lee SJ, Lee YH, et al:
Angiopoietin-1 inhibits irradiation- and
mannitol-induced apoptosis in endothelial
cells. Circulation 2000; 101:23172324.
241. Kanda S, Miyata Y, Mochizuki Y, et al:
Angiopoietin 1 is mitogenic for cultured
endothelial cells. Cancer Res 2005;
65:68206827.
242. Witzenbichler B, Maisonpierre PC, Jones P,
et al: Chemotactic properties of
angiopoietin-1 and -2, ligands for the
endothelial-specic receptor tyrosine
kinase Tie2. J Biol Chem 1998;
273:1851418521.
243. Thurston G, Rudge JS, Ioffe E, et al:
Angiopoietin-1 protects the adult
vasculature against plasma leakage. Nat
Med 2000; 6:460463.
244. Cho CH, Kim KE, Byun J, et al: Long-term
and sustained COMP-Ang1 induces longlasting vascular enlargement and enhanced
blood flow. Circ Res 2005; 97:8694.
245. Hawighorst T, Skobe M, Streit M, et al:
Activation of the tie2 receptor by
angiopoietin-1 enhances tumor vessel
maturation and impairs squamous cell
carcinoma growth. Am J Pathol 2002;
160:13811392.
246. Stoeltzing O, Ahmad SA, Liu W, et al:
Angiopoietin-1 inhibits vascular
permeability, angiogenesis, and growth of
hepatic colon cancer tumors. Cancer Res
2003; 63:33703377.
247. Kim I, Kim HG, Moon SO, et al:
Angiopoietin-1 induces endothelial cell
sprouting through the activation of focal
adhesion kinase and plasmin secretion.
Circ Res 2000; 86:952959.
248. Thurston G, Suri C, Smith K, et al:
Leakage-resistant blood vessels in mice
transgenically overexpressing angiopoietin1. Science 1999; 286:25112514.
249. Kim I, Moon SO, Park SK, et al:
Angiopoietin-1 reduces VEGF-stimulated
leukocyte adhesion to endothelial cells by
reducing ICAM-1, VCAM-1, and E-selectin
expression. Circ Res 2001; 89:477479.
250. Kim I, Oh JL, Ryu YS, et al: Angiopoietin-1
negatively regulates expression and activity
of tissue factor in endothelial cells. FASEB
Journal 2002; 16:136138.
251. Koblizek TI, Weiss C, Yancopoulos GD,
et al: Angiopoietin-1 induces sprouting
angiogenesis in vitro. Curr Biol 1998;
8:529532.
252. Saito M, Hamasaki M, Shibuya M:
Induction of tube formation by
253.
254.
255.
256.
257.
258.
259.
260.
261.
262.
263.
264.
265.
266.
267.
angiopoietin-1 in endothelial cell/broblast

co-culture is dependent on endogenous
VEGF. Cancer Sci 2003; 94:782790.
Hangai M, Murata T, Miyawaki N, et al:
Angiopoietin-1 upregulation by vascular
endothelial growth factor in human retinal
pigment epithelial cells. Invest Ophthalmol
Vis Sci 2001; 42:16171625.
Park YS, Kim NH, Jo I: Hypoxia and
vascular endothelial growth factor acutely
up-regulate angiopoietin-1 and Tie2 mRNA
in bovine retinal pericytes. Microvasc Res
2003; 65:125131.
Asahara T, Chen D, Takahashi T, et al: Tie2
receptor ligands, angiopoietin-1 and
angiopoietin-2, modulate VEGF-induced
postnatal neovascularization. Circ Res
1998; 83:233240.
Hangai M, Moon YS, Kitaya N, et al:
Systemically expressed soluble Tie2
inhibits intraocular neovascularization. Hum
Gene Ther 2001; 12:13111321.
Joussen AM, Poulaki V, Tsujikawa A, et al:
Suppression of diabetic retinopathy with
angiopoietin-1. Am J Pathol 2002;
160:16831693.
Nambu H, Nambu R, Oshima Y, et al:
Angiopoietin 1 inhibits ocular
neovascularization and breakdown of the
bloodretinal barrier. Gene Ther 2004;
11:865873.
Nambu H, Umeda N, Kachi S, et al:
Angiopoietin 1 prevents retinal detachment
in an aggressive model of proliferative
retinopathy, but has no effect on
established neovascularization. J Cell
Physiol 2005; 204:227235.
Gale NW, Thurston G, Hackett SF, et al:
Angiopoietin-2 is required for postnatal
angiogenesis and lymphatic patterning, and
only the latter role is rescued by
Angiopoietin-1. Dev Cell 2002; 3:411423.
Fiedler U, Krissl T, Koidl S, et al:
Angiopoietin-1 and angiopoietin-2 share
the same binding domains in the Tie-2
receptor involving the rst Ig-like loop and
the epidermal growth factor-like repeats.
J Biol Chem 2003; 278:17211727.
Teichert-Kuliszewska K, Maisonpierre PC,
Jones N, et al: Biological action of
angiopoietin-2 in a brin matrix model of
angiogenesis is associated with activation
of Tie2. Cardiovasc Res 2001; 49:659670.
Mochizuki Y, Nakamura T, Kanetake H,
Kanda S: Angiopoietin 2 stimulates
migration and tube-like structure formation
of murine brain capillary endothelial cells
through c-Fes and c-Fyn. J Cell Sci 2002;
115:175183.
Das A, Fanslow W, Cerretti D, et al:
Angiopoietin/Tek interactions regulate
mmp-9 expression and retinal
neovascularization. Lab Invest 2003;
83:16371645.
Stratmann A, Risau W, Plate KH: Cell typespecic expression of angiopoietin-1 and
angiopoietin-2 suggests a role in
glioblastoma angiogenesis. Am J Pathol
1998; 153:14591466.
Oh H, Takagi H, Suzuma K, et al: Hypoxia
and vascular endothelial growth factor
selectively up-regulate angiopoietin-2 in
bovine microvascular endothelial cells.
J Biol Chem 1999; 274:1573215739.
Mandriota SJ, Pepper MS: Regulation of
angiopoietin-2 mRNA levels in bovine
microvascular endothelial cells by cytokines
and hypoxia. Circ Res 1998; 83:852859.
268. Scharpfenecker M, Fiedler U, Reiss Y,

Augustin HG: The Tie-2 ligand
angiopoietin-2 destabilizes quiescent
endothelium through an internal autocrine
loop mechanism. J Cell Sci 2005;
118:771780.
269. Holash J, Maisonpierre PC, Compton D, et
al: Vessel cooption, regression, and growth
in tumors mediated by angiopoietins and
VEGF. Science 1999; 284:19941998.
270. Otani A, Takagi H, Oh H, et al: Expressions
of angiopoietins and Tie2 in human
choroidal neovascular membranes. Invest
271. Patel JI, Hykin PG, Gregor ZJ, et al:
Angiopoietin concentrations in diabetic
retinopathy. Br J Ophthalmol 2005;
89:480483.
272. Oliner J, Min H, Leal J, et al: Suppression
of angiogenesis and tumor growth by
selective inhibition of angiopoietin-2.
Cancer Cell 2004; 6:507516.
273. White RR, Shan S, Rusconi CP, et al:
Inhibition of rat corneal angiogenesis by a
nuclease-resistant RNA aptamer specic
for angiopoietin-2. Proc Natl Acad Sci USA
2003; 100:50285033.
274. Oshima Y, Oshima S, Nambu H, et al:
Different effects of angiopoietin-2 in
different vascular beds: new vessels are
most sensitive. FASEB J 2005; 19:963965.
275. Campochiaro PA: Ocular versus extraocular
neovascularization: mirror images or vague
resemblances. Invest Ophthalmol Vis Sci
2006; 47:462474.
276. Erber R, Eichelsbacher U, Powajbo V, et al:
EphB4 controls blood vascular
morphogenesis during postnatal
angiogenesis. EMBO J 2006; 25:628641.
277. Hirai H, Maru Y, Hagiwara K, et al: A novel
putative tyrosine kinase receptor encoded
by the eph gene. Science 1987;
238:17171720.
278. Zhang J, Hughes S: Role of the ephrin and
Eph receptor tyrosine kinase families in
angiogenesis and development of the
cardiovascular system. J Pathol 2006;
208:453461.
279. Himanen JP, Chumley MJ, Lackmann M, et
al: Repelling class discrimination: ephrin-A5
binds to and activates EphB2 receptor
signaling. Nat Neurosci 2004; 7:501509.
280. Dodelet VC, Pasquale EB: Eph receptors
and ephrin ligands: embryogenesis to
tumorigenesis. Oncogene 2000;
19:56145619.
281. Davy A, Soriano P: Ephrin signaling in vivo:
look both ways. Dev Dyn 2005; 232:110.
282. Palmer A, Klein R: Multiple roles of ephrins
in morphogenesis, neuronal networking,
and brain function. Genes Dev 2003;
17:14291450.
283. McBride JL, Ruiz JC: Ephrin-A1 is
expressed at sites of vascular development
in the mouse. Mech Dev 1998; 77:201204.
284. Brantley-Sieders DM, Caughron J, Hicks D,
et al: EphA2 receptor tyrosine kinase
regulates endothelial cell migration and
vascular assembly through
phosphoinositide 3-kinase-mediated Rac1
GTPase activation. J Cell Sci 2004;
117:20372049.
285. Ojima T, Takagi H, Suzuma K, et al:
EphrinA1 inhibits vascular endothelial
growth factor-induced intracellular signaling
and suppresses retinal neovascularization
and bloodretinal barrier breakdown. Am J
Pathol 2006; 168:331339.
CHAPTER 31
335
SECTION 4
336
286. Cheng N, Brantley DM, Liu H, et al:

Blockade of EphA receptor tyrosine kinase
activation inhibits vascular endothelial cell
growth factor-induced angiogenesis. Mol
Cancer Res 2002; 1:211.
287. Vihanto MM, Plock J, Erni D, et al: Hypoxia
up-regulates expression of Eph receptors
and ephrins in mouse skin. FASEB J 2005;
19:16891691.
288. Chen J, Hicks D, Brantley-Sieders D, et al:
Inhibition of retinal neovascularization by
soluble EphA2 receptor. Exp Eye Res 2006;
82:664673.
289. Wang HU, Chen ZF, Anderson DJ:
Molecular distinction and angiogenic
interaction between embryonic arteries and
veins revealed by ephrin-B2 and its
receptor Eph-B4. Cell 1998; 93:741753.
290. Gerety SS, Wang HU, Chen ZF,
Anderson DJ: Symmetrical mutant
phenotypes of the receptor EphB4 and its
specic transmembrane ligand ephrin-B2 in
cardiovascular development. Mol Cell
1999; 4:403414.
291. Adams RH, Wilkinson GA, Weiss C, et al:
Roles of ephrinB ligands and EphB
receptors in cardiovascular development:
demarcation of arterial/venous domains,
vascular morphogenesis, and sprouting
angiogenesis. Genes Dev 1999;
13:295306.
292. Adams RH, Diella F, Hennig S, et al: The
cytoplasmic domain of the ligand ephrinB2
is required for vascular morphogenesis but
not cranial neural crest migration. Cell
2001; 104:5769.
293. Stein E, Lane AA, Cerretti DP, et al:
Eph receptors discriminate specic ligand
oligomers to determine alternative signaling
complexes, attachment, and assembly
responses. Genes Dev 1998; 12:667678.
294. Lin D, Gish GD, Songyang Z, Pawson T:
The carboxyl terminus of B class ephrins
constitutes a PDZ domain binding motif.
J Biol Chem 1999; 274:37263733.
295. Fanning AS, Anderson JM: PDZ domains:
fundamental building blocks in the
organization of protein complexes at the
plasma membrane. J Clin Invest 1999;
103:767772.
296. Zamora DO, Davies MH, Planck SR,
et al: Soluble forms of EphrinB2 and
EphB4 reduce retinal neovascularization
in a model of proliferative retinopathy.
297.
298.
299.
300.
301.
302.
303.
304.
305.
306.

46:21752182.
Fuller T, Korff T, Kilian A, et al: Forward
EphB4 signaling in endothelial cells
controls cellular repulsion and segregation
from ephrinB2 positive cells. J Cell Sci
2003; 116:24612470.
Maekawa H, Oike Y, Kanda S, et al: EphrinB2 induces migration of endothelial cells
through the phosphatidylinositol-3 kinase
pathway and promotes angiogenesis in
adult vasculature. Arterioscler Thromb Vasc
Biol 2003; 23:20082014.
Steinle JJ, Meininger CJ, Forough R, et al:
Eph B4 receptor signaling mediates
endothelial cell migration and proliferation
via the phosphatidylinositol 3-kinase
pathway. J Biol Chem 2002;
277:4383043835.
Huynh-Do U, Vindis C, Liu H, et al: EphrinB1 transduces signals to activate integrinmediated migration, attachment and
angiogenesis. J Cell Sci 2002;
115:30733081.
Hayashi S, Asahara T, Masuda H, et al:
Functional ephrin-B2 expression for
promotive interaction between arterial and
venous vessels in postnatal
neovascularization. Circulation 2005;
111:22102218.
He S, Ding Y, Zhou J, et al: Soluble EphB4
regulates choroidal endothelial cell function
and inhibits laser-induced choroidal
Sci 2005; 46:47724779.
Umeda N, Ozaki H, Hayashi H, Oshima K:
Expression of ephrinB2 and its receptors
on broproliferative membranes in ocular
angiogenic diseases. Am J Ophthalmol
2004; 138:270279.
Mann F, Harris WA, Holt CE: New views on
retinal axon development: a navigation
guide. Int J Dev Biol 2004; 48:957964.
Wharton KA, Johansen KM, Xu T,
Artavanis-Tsakonas S: Nucleotide
sequence from the neurogenic locus notch
implies a gene product that shares
homology with proteins containing EGF-like
repeats. Cell 1985; 43:567581.
Kidd S, Kelley MR, Young MW: Sequence
of the notch locus of Drosophila
melanogaster: relationship of the encoded
protein to mammalian clotting and growth
factors. Mol Cell Biol 1986; 6:30943108.
307. Morgan T: The theory of the gene. Am Nat

1917; 51:513544.
308. Lai EC: Notch signaling: control of cell
communication and cell fate. Development
2004; 131:965973.
309. Weinmaster G, Kopan R: A garden of
Notch-ly delights. Development 2006;
133:32773282.
310. Sjolund J, Manetopoulos C,
Stockhausen MT, Axelson H: The Notch
pathway in cancer: differentiation gone
awry. Eur J Cancer 2005; 41:26202629.
311. Gale NW, Dominguez MG, Noguera I, et al:
Haploinsufciency of delta-like 4 ligand
results in embryonic lethality due to major
defects in arterial and vascular
development. Proc Natl Acad Sci USA
2004; 101:1594915954.
312. Krebs LT, Shutter JR, Tanigaki K, et al:
Haploinsufcient lethality and formation of
arteriovenous malformations in Notch
pathway mutants. Genes Dev 2004;
18:24692473.
313. Duarte A, Hirashima M, Benedito R, et al:
Dosage-sensitive requirement for mouse
Dll4 in artery development. Genes Dev
2004; 18:24742478.
314. Mailhos C, Modlich U, Lewis J, et al:
Delta4, an endothelial specic notch ligand
expressed at sites of physiological and
tumor angiogenesis. Differentiation 2001;
69:135144.
315. Liu ZJ, Shirakawa T, Li Y, et al: Regulation
of Notch1 and Dll4 by vascular endothelial
growth factor in arterial endothelial cells:
implications for modulating arteriogenesis
and angiogenesis. Mol Cell Biol 2003;
23:1425.
316. Patel NS, Li JL, Generali D, et al:
Up-regulation of delta-like 4 ligand in
human tumor vasculature and the role of
basal expression in endothelial cell
function. Cancer Res 2005; 65:86908697.
317. Williams CK, Li JL, Murga M, et al: Upregulation of the Notch ligand Delta-like 4
inhibits VEGF-induced endothelial cell
function. Blood 2006; 107:931939.
318. Hainaud P, Contreres JO, Villemain A, et al:
The role of the vascular endothelial growth
factor-delta-like 4 ligand/Notch4-Ephrin B2
cascade in tumor vessel remodeling and
endothelial cell functions. Cancer Res
2006; 66:85018510.
CHAPTER
32
Principles of Toxicology of the Eye

Keith J. Lane, Zhou Chen, and Matthew J. Chapin
Key Features
Toxicity is a result of harmful byproduct production that occurs

during drug metabolism, and consequently toxicity is generally
dose dependent. The ease with which the body can convert
xenobiotic substances (drugs) to hydrophilic byproducts and
excrete these byproducts is the single largest factor
determining tissue toxicity. If lengthy and extensive metabolism
is required leading to excessive generation of free radicals then
the substance is more likely to buildup in the tissue and cause
toxic effects.
Ocular irritation and ocular toxicity are independent processes
that are often confused. Irritation involves direct action on pain
receptors rather than secondary toxic activity related to
xenobiotic metabolism.
The Draize test is still considered a standard evaluation for
assessing ophthalmic irritation potential, however the original
Draize design has been improved to better qualify the potential
toxicity of long-term repeated dosing with topical ophthalmic
drugs. Improvements to the original Draize test include
changes to the scales, better standardization of dosing and
timepoints, improved imaging technology to better assess
tissue damage, and the addition of systemic and organ health
assessments to better account for any systemic effects
following repeated dosing.
Regulatory requirements for approval of an ophthalmic agent
vary depending on intended route of administration, intended
dosing regimen, and historical knowledge of drug side effect
prole. The FDA requires a series of genotoxicity studies,
acute in vivo toxicity studies in multiple species (featuring dosing
via the intended route of administration and via systemic
administration), chronic dosing in vivo toxicity studies, and
reproductive toxicity studies to establish a margin of safety
prior to approving a drug for human use. Close communication
with the FDA is recommended to clarify appropriate toxicology
study designs and to ensure that a toxicology program fully
supports the intended clinical use of a drug.
INTRODUCTION PRINCIPLES OF
TOXICOLOGY OF THE EYE
Historically, ophthalmic toxicology has been presented as a list
of classes of agents and their record of ophthalmic toxicities.
Unfortunately, the traditional approach to qualifying ophthalmic toxicology is decient, failing to provide the clinician with
an encompassing view of the underlying science of toxic reactions in the eye, the research methods by which toxicity is
categorized, and the requirements for toxicology testing put in
place by regulatory authorities. This chapter will supplement

the traditional view of ophthalmic toxicity in several ways. We
will identify the cellular processes by which ophthalmic agents
exert their toxic affects. We will review common models of
ophthalmic toxicology beginning with the traditional Draize
test originally published in 1944 and followed by a review of
renements to the model. Regulatory requirements for approval
of new topical ophthalmic agents will be reviewed, including
descriptions of GLP study designs currently recommended by
the FDA prior to commencing clinical trials and for new drug
application (NDA) submission. Finally, we will discuss common ndings noted during GLP toxicology studies and how
these ndings are interpreted.
MOLECULAR MECHANISMS OF
OPHTHALMIC TOXICITY
At the molecular level, toxicity is a byproduct of a series of
reactions by which the body converts hydrophobic, xenobiotic
substances into hydrophilic compounds that can be easily
excreted. In response to exposure to foreign substances, a series
of metabolic reactions occur at a cellular level including oxidation, hydroxylation, and reduction followed by conjugation
reactions including glucuronidation, sulfation, and acetylation.1
These metabolic responses are mediated by a variety of different
enzymes, including those associated with the cytochrome
P-450 system. Free radicals, including nitric oxide (NO), and
hydrogen peroxide are generated as byproducts during this
metabolism and tissue damage that occurs with exposure to
these volatile molecules is what immediately translates into a
toxic clinical affect.2
The cytochrome P-450 enzymes are critical in the metabolism of a number of foreign substances and are involved early
in the process of xenobiotic metabolism. Most P-450 enzymes
have been identied in the endoplasmic reticulum of the cell
and are typically membrane bound. The P-450 system is most
famous for its involvement in steroid biosynthesis, however its
role in detoxication is also well known, with a specic role in
xenobiotic metabolism in the ocular tissues well established.3
Cytochrome P-450 enzyme systems have been detected in the
ciliary epithelium, conjunctiva, retina, and the corneal
epithelium.4,5
With xenobiotic metabolism, the P-450 system is activated
immediately upon exposure to the drug, or other foreign substance. P-450 enzymes then catalyze a variety of reactions
including oxidation, reduction, and hydrolation. These enzymes
are known to activate oxygen and are classied as monoxygenases.6 During these early reduction reactions, the P-450
enzyme typically donates an oxygen molecule to the substrate
337
SECTION 4

to which it binds with the second oxygen molecule reduced as
water. Oxidation, reduction, and hydrolation reactions can
result in free radical production, which in turn contribute
to local toxicity, inflammation, and in some instances
neovascularization.7
Xenobiotic substances may be further metabolized via glucuronidation. Glucuronidation involves the interaction of the
xenobiotic with UDP-glucuronate and the enzyme UDPglucuronyltransferase and typically transforms xenobiotics into
hydrophilic products which can more easily be removed from
circulation. Glucuronidation is used by the body to recycle
certain molecules such as heme, which is produced during the
metabolism of hemoglobin. Glucuronidation is responsible for
the conversion of heme product bilirubin into bilirubin
diglucuronide, a water-soluble compound that can be easily
removed from the body.
Sulfation is another metabolic method by which the body
eliminates xenobiotic substances. A sulfate molecule is donated
to a substrate by 3-phosphoadenosine-5-phosphosulfate in the
presence of sulfotransferase. Phenols, alcohols, and aromatic
amines are frequently metabolized via sulfation. Free radicals
can be generated during sulfation reactions.
Acetylation provides still another mechanism by which
xenobiotics are further metabolized and prepared for excretion.
N-Acetylation of arylamine is a common route of metabolism
for many drugs in which an acetyl group is transferred to
the arylamine substrate in the presence of arylamine
N-acetyltransferase.8
Biooxidation reactions occur frequently during xenobiotic
metabolism and the byproducts of biooxidation (free radicals,
super oxides, and hydrogen peroxide) can cause toxicity via a
number of different mechanisms. If free radicals are present at
sufciently high levels, than lipid peroxidation may occur. Lipid
peroxidation occurs when free radicals incorporate themselves
into polyunsaturated fatty acids, which will eventually cause
the break down of biological membranes.9 Breakdown of the cell
membrane will cause cell death and spilling of cellular contents,
which in turn exacerbates inflammation, causes influx of
inflammatory cells and may lead to further tissue damage and
swelling. Both the lens and the tissue of the retina, because they
are exposed to high levels of light, are especially susceptible to
biooxidation reactions. In the retina, biooxidation and lipid
peroxidation have been related to AMD while in the lens
biooxidation has been shown to contribute to the development
of cataracts.
The ease with which the body can convert xenobiotic substances (drugs) to hydrophilic byproducts and excrete these
byproducts is the single largest factor determining tissue toxicity. If lengthy and extensive metabolism is required leading to
excessive generation of free radicals then the substance is more
likely to buildup in the tissue and cause toxic effects. On the
other hand, if a drug easily metabolized and excreted without
signicant generation of free radicals, super oxides, hydrogen
peroxides, and otherwise volatile molecules, it can be assumed
that the agent will have a less toxic prole.
IRRITATION VERSUS TOXICITY:

UNDERSTANDING THE DIFFERENCES
338
Recognizing the difference between toxicity and irritation is

critical to understanding toxic responses in the eye. Toxicity
generally occurs in a dose-dependent fashion and is reproducible. There is, in general, a consistent relationship between the
level of toxicity exhibited and the amount of an agent given.
Because toxicity is dependent on metabolism and buildup of the
specic compound administered, toxic effects are not always
immediate. Irritation is not entirely related to reactive (and tissue-
destructive) oxygen byproducts that occur with compound

metabolism. Instead irritation is dependent on the physical
characteristics of the compound itself (pH, osmolality, concentration, etc.) and is mediated primarily by immediate activation
of sensory neurons in the eye, rapid prostaglandin release, and
minor tissue damage. Irritation in its most severe forms can be
considered trauma, with severe alkali and acid burns causing
immediate tissue necrosis.
Irritation occurs when a chemical substance applied to the
surface of the eye elicits antidromic sensory activation, alterations in membrane permeability, and/or release of arachidonic
acid metabolites, causing immediate pain, discomfort and
inflammation.10 The exact involvement of the various factors
that contribute to irritation differs depending on the extent and
nature of the ocular irritation and the particular irritating agent.
Antidromic sensory activation quickly differentiates irritation
from toxicity, as toxic effects are apparent following drug metabolism rather than preceding it. Here, the irritant binds directly
at temperature and chemical sensitive receptors found on
sensory nerve endings causing immediate sensation of pain or
irritation. A variety of nerve ending receptors may play a part in
the immediate irritation response. Receptors of the transient
receptor potential (TRP) family are associated with ocular irritation.11 The vanilloid receptor is a member of the TRP family
that is activated by capsaicin, hot temperatures and protons
(H+).12 It is therefore likely that the vanilloid receptor is at least
in part responsible for the irritation that occurs following
administration of an acidic eyedrop.
Others receptors involved in ocular irritation include the
menthol receptor, a member of the TRP receptor family activated
by cold stilumi and neurokinin-1 receptor which binds substance
P in the traditional pain response pathway.13 Additionally,
members of the TRP receptor family are bound by bradykinin,
a well-known proinflammatory peptide released following tissue
damage and involved in inducing pain.
RESEARCH METHODS FOR QUALIFYING

TOXIC EFFECTS
Ocular toxicity and irritation testing is required by all manner
of regulatory authorities for approval of ophthalmic products
and nonophthalmic products, including new raw materials,
additives, or any substance which could at any point be exposed
to the eye. Generally speaking, there is not an in vitro test
available that can effectively predict the human response to
ocular exposure to a specic agent. As a result, animal models
remain standard for evaluating ocular toxicity. The Draize test,
an in vivo model which was originally published in 1944, is still
frequently relied upon to qualify ophthalmic toxicity of both
ophthalmic and nonophthalmic agents, despite the fact that the
Draize test was initially designed to test the ocular irritation
potential of nonophthalmic products. Over time it has become
apparent that the Draize test has some deciencies. The test
was designed to assess acute irritation only and many agents are
known to be toxic only after long-term, repeated exposures.
While certainly of some value to researchers, new research
methods are being employed which seek to improve on the
Draize model. In this section, we will discuss the original
Draize test and the modications to the Draize test that aid
with identication and qualication of the toxic potential of
ophthalmic drugs.
THE DRAIZE TEST FOR OCULAR TOXICITY

The Draize test, which was developed some 60 years ago, is still
a standard test for assessing acute toxic potential of ophthalmic
agents.14 The primary deciency of the Draize test is that it was
designed to assess acute irritation potential (which is frequently

related to pH, osmolality, etc.) rather than long-term toxicity
that may occur after repeated dose administration. Ocular toxicity does of course involve reactions that occur with extended
exposure and repeated metabolism of the foreign agent. As per
the original Draize test methodology, only alterations to the
anterior portion of the eye, the conjunctiva, cornea, and iris were
formally assessed. Any changes that occur at the retina cannot
be evaluated using original Draize methodology. Nonetheless,
the Draize test still provides a good idea of the irritation or
acute toxicity potential of a compound and can be helpful for
determining how to label chemical products and other agents
for severity of irritation.
The original Draize design involves topical dosing with
100 mL of an agent to the conjunctiva of the New Zealand White
rabbit. Following dosing, a series of clinical endpoints are evaluated at predetermined timepoints. These clinical endpoints
include conjunctival congestion, chemosis, discharge, iritis, corneal opacity (size and degree of) and are graded in accordance
with an incremented scaling system. These scales are combined
to produce a cumulative Draize score which provides an indication of how irritating or toxic a compound will be to the
human eye. Historically, six rabbits have been used per test
compound when conducting the Draize test.
To a certain extent, the Draize test is predictable. An extremely acidic or basic agent (below pH 2.5 or above pH 11.5) will
cause ocular irritation. Furthermore, dermal tests can be
performed on guinea pigs, rats, mice, and rabbits that can often
determine whether or not ocular dosing will cause irritation. In
many instances, it can be determined whether or not an agent
will cause ocular irritation prior to putting an agent in the
rabbits eye. If the Draize test alone is not predictive of longterm toxicity and is not necessary for identication of acute
irritation, than it may be of questionable utility. Our experience
has been that the Draize test alone is a decent indicator of acute
irritation potential and we typically use the Draize test over
other procedures that involve application via a dermal route;
however we will not test an ophthalmic agent that falls outside
of preestablished comfort matrixes (pH ranges, etc.).
Because it is an acute design, the Draize test also fails to take
into account the possible sensitizing effect of an ophthalmic
agent. A minimum of two doses must be given to allow for
buildup of IgE in the system (sensitization in response to
immunoavailability of dose 1) and subsequent hypersensitivity
response (cross-linking of IgE in response to dose 2).
Other criticisms of the Draize test include the fact that
graders do read responses differently, despite the fact that the
Draize test is based on a standard scale. Additionally, the Draize
test does not mandate specic assessment timepoints and different investigators may score irritation at different timepoints,
leading to inconsistencies in scoring. Finally and perhaps most
importantly, the Draize test does not adequately dene a
method for instilling drug. Some investigators may pipette an
agent directly onto the surface of the eye while different
investigators administer the test agent into the conjunctival
sac. These different methods of administration can cause a net
difference in the amount of dose given, as drug may fall out of
the eye more readily with one method than the other, and can
cause a disparity in the amount of drug that comes into contact
with certain ocular tissues. These variations can make it
difcult to compare Draize test results conducted by different
investigators.
IMPROVEMENTS TO THE DRAIZE TEST

While the Draize test is an acceptable means for identifying the
acute irritation potential of a compound that comes into con-
tact with the ocular surface, repeated exposure and assessment

of long-term toxicity mandates different and improved methodology. The Draize test is an old test, and since its development
a variety of novel techniques have been developed that can
assess the toxic potential of agents in a variety of different
ocular tissues beyond the anterior region of the eye. Confocal
microscopy, pachymetry, specular microscopy, fluorophotometry,
fluorescein staining, tonometry, and histological assessments
provide a more sensitive means for assessing alterations to the
ocular tissues that may occur with toxicity. Furthermore, the
collection and histological evaluation of all ocular tissues
provides an encompassing picture of the way an agent impacts
the eye, rather than a narrow view of the conjunctiva, cornea,
and anterior chamber.
The McDonaldShadduck scoring system for ocular lesions15
improved upon the limited Draize scale and is widely used as an
alternative test for scoring clinical signs of ocular toxicity. The
McDonaldShadduck scoring system provides standardized
scales for assessment of conjunctival congestion, conjunctival
swelling, conjunctival discharge, aqueous flare, iris involvement, corneal opacity (area and severity), pannus, and fluorescein staining, expanding on the limited Draize parameters.
That is not to say that the McDonaldShadduck scale is sufcient; however, when combined with dilated fundus exams to
assess the retina and optic nerve, tonometry to assess IOP,
and histology on all ocular tissues following sacrice, the
McDonaldShadduck scoring system provides an acceptable
methodology for evaluating toxic effects related to repeat dosing.
The importance of new technologic developments in imaging
to supplement the standard gross clinical endpoints described
using the McDonaldShadduck scoring system cannot be
overemphasized. The use of pachymetry can also be employed
to measure subtle changes in corneal thickness that might not
be readily visible under slit-lamp exam.16 Confocal microscopy
can be used to assess pathophysiological events as they occur
in vivo.17 For this reason, the confocal technique holds signicant promise for use in understanding toxic responses.
Additionally, only a small dose needs to be given to observe
molecular activity at the surface of the eye using the confocal
technique, which is useful as dosing with small quantities of
drug limits potential for irritation. Specular microscopy can be
used to assess changes in endothelial cell permeability following
drug treatment at the ocular surface when combined with
fluorescein administration.18 Changes in the structure of the
corneal epithelium leading to increased penetration of agents
could result in increased toxicity that occurs with increased
exposure. Electroretinography (ERG) is quickly becoming an
important technique for assessing toxicity in the retina. Using
a corneal contact lens electrode to measure the response of the
retina to light stimuli, toxic effects on photoreceptor function
can be identied. A specic pigmented rat model has been
developed, however dog, cat, and monkey models are also
frequently employed.1921
The Kligman maximization test, rst published in 1969, was
developed to account for the sensitization potential of an agent.
While this is not ocular toxicity design, it can be used to supplement an appropriate ocular toxicity study and provide information on whether an ophthalmic agent may act as a sensitizer.
The design calls for dermal application but can be used to
screen ophthalmic compounds.22 The Kligman test features
repeated dermal application of the test agent followed by
evaluation of the skin to determine whether or not a hypersensitivity reaction has occurred. Known sensitizers are used as
controls. Since it was originally published, the test has been
altered slightly.23
It is important also to remember that repeated ocular dosing
may have systemic toxic effects, usually less than oral or
CHAPTER 32
339
SECTION 4
340
intravenous administration, as drugs delivered to the eye enter

the bloodstream and are often metabolized by the liver. As a
result, a thorough evaluation of toxicity of an ophthalmic agent
should take into account any potential systemic effects by
observing clinical signs, body weight changes, food consumption changes, behavioral changes, hematology, coagulation
studies, urinalysis, clinical chemistry, toxicokinetic assessments, and changes in organ weights/appearance following
necropsy. Any ndings should be followed by full histology on
the impacted tissues.
Other basic improvements made to the Draize test include a
decrease in the dose size given and a decrease in the number of
animals used per concentration. The original Draize methodology called for a dose size of 100 mL; however, the average
eyedrop size is ~3040 mL. One hundred mL is far too high a
volume to deliver effectively to the eye, the eye cannot hold
such a high volume of fluid and such a volume is not clinically
relevant. (Remember, the human eye holds ~2030 mL.) Most
toxicity studies employing topical dosing have decreased drop
size to clinically relevant volumes (3040 mL). Furthermore, the
number of animals used per concentration of test agent was
dened as nine under original Draize methodology. Under the
Code of Federal Regulations set forth in 1979, this gure has been
reduced to six animals, and further reductions are possible.
TOXICOLOGY REQUIREMENTS FOR IND

SUBMISSION
DRUG APPROVAL PROCESS:

REQUIREMENTS FOR APPROVAL OF A
TOPICAL OPHTHALMIC AGENT
GENETOX STUDIES
The requirements put in place by regulatory authorities for

approval of topical ophthalmic agents are designed to ensure
that the agent is safe, nonirritating, and nontoxic, with an ample
margin of error based on the amount of drug administered. This
simple paradigm can serve as the overall strategy for designing
dose selection and dosing regimen for good laboratory practice
(GLP) ocular toxicity studies which will be submitted to regulatory authorities as a component of an investigational new
drug application (IND) or NDA. In general doses selected and
dosing regimen should approximate or exceed clinical expectations (preferably exceed). For studies submitted as components
of the IND, the duration of GLP ocular toxicity studies should
exceed or at least equal the duration of proposed clinical studies
to again ensure an appropriate safety margin.
The extent of toxicology testing required for a new ophthalmic drug depends on several factors; the amount of toxicology
data already available for that particular drug/class of agents, the
intended dosing regimen/indication, and the route of administration. For a new molecular entity, toxicity studies using a
systemic route (e.g., oral, subcutaneous, or intravenous
depending on bioavailability of the compound) should also be
performed to assess potential toxicity with systemic absorption.
For agents that are administered daily for extended periods (e.g.,
ocular hypotensives), toxicology requirements will be greater
than for agents that are administered as needed or irregularly
(antibiotics) or regularly for a specic but relatively short period
of time (e.g., NSAIDs).
This section will describe those standard studies that
regulatory authorities require for approval of topical ophthalmic
agents. While additional studies may be required for certain
classes of agents or chronic indications, or for ophthalmic
agents administered via a nontopical route, the following studies represent the core toxicology studies that in our experience
should be completed for development and approval of all ocular
ophthalmic products. We will discuss toxicology requirements
of the Food and Drug Administration (FDA) for both IND and
NDA submissions, based on experience working with the FDA
as consultants on 26 NDA approvals.
Included in the IND submission should be a standard battery of

genetic toxicity tests (Ames reverse mutation assay, in vitro test
with cytogenetic evaluation of chromosomal damage with
mammalian cells or an in vitro mouse lymphoma TK assay, and
an in vivo test for chromosomal damage using rodent hematopoietic cells), a non-GLP melanin binding assay, and GLP
ocular toxicity studies in two different species for a duration
that will exceed that of the initial proof of concept clinical trial.
The genetic toxicity studies are standard designs that assay
genotoxic potential of the compound itself and do not use the
ophthalmic solution, rather the active pharmaceutical ingredient (API). The melanin binding study is desired to gain a
better understanding of the pharmacokinetic prole of the test
agent, and is also necessary to determine whether or not pigmented animals are needed in follow-up GLP toxicology studies
as an agent which binds melanin will have a different pharmacokinetic in an albino species. The GLP ocular toxicity
studies involve dosing via the proposed clinical route of administration (ocular topical, subconjunctival, intravitreal, etc.).
Clinical signs, gross necropsy and histopathology are all monitored to ensure that the test agent does not cause a toxic effect.
The Ames reverse mutation assay is a standard design used to

assess the ability of the test substance to induce reverse
mutations in the histidine and tryptophan genes of Salmonella
typhimurium and Escherichia coli respectively. Histidine and
tryptophan are required for the growth of Salmonella
typhimurium and Escherichia coli respectively. Strains of S.
typhimurium and E. coli with a mutation in these genes are
grown in the presence of exogenous histidine or tryptophan.
These organisms cannot grow without addition of these
proteins unless they are able to reverse the mutation in these
specic genes and produce the protein endogenously.
A range nding assay is performed as an initial step to determine the concentration at which the test article is cytotoxic.
The test bacterial strains are then exposed to a series of standard concentrations of the test agent below the established level
of toxicity to determine the ability of the agent to reverse
mutations in the aforementioned genes, with and without a
metabolic activator (Aroclor-induced rat liver S9). Colonies are
plated immediately after exposure and revertant colonies (any
colonies that are able to survive and grow once plated) are
counted. A conrmatory assay is performed with a preincubation period to verify results of the initial assay. Positive
and negative controls are employed to ensure a valid and sensitive
test system.
The chromosomal aberration assay is also a standard design
which tests the ability of the test agent to induce chromosomal
aberrations in Chinese hampster ovary (CHO) cells in the
presence and absence of a metabolic activating system. The
initial step in the chromosomal aberration study is a rangending assay which identies the concentration of active drug
that is cytotoxic. From this concentration, a standard series of
dilutions are performed to create concentrations which will be
used for the actually assay. Drug concentrations are incubated
with CHO cells for a period of 3 h with or without a metabolic
activator (Aroclor-induced rat liver S9). Following this incubation period, cells are assayed for chromosomal structural
aberrations or polyploidy to determine whether or not the test
substance alters these parameters. Typically, a conrmatory
assay with increased duration of exposure (up to 21 h) is also
employed without metabolic activation. Positive controls
mitomycin C and cyclophosphamide are employed to verify

sensitivity of the test system.
The in vitro mouse lymphoma assay is a third genetox assay
required under ICH S2B. While there are a number of cell lines
that can be used, the L5178Y TK+/3.7.2C mouse lymphoma
cell is standard. This assay identies mutations that are
associated specically with cancerous changes and other human
genetic illnesses.24 The in vitro mouse lymphoma assay is not
always requested by FDA prior to IND submission for an
ophthalmic compound. The Ames and chromosomal aberration
assays are almost always performed, however, the sponsor may
choose either mouse lymphoma TK assay or chromosomal
aberration assay as the second in vitro genotoxicity study.
The test agent is administered at several different dose levels
and compared to positive (mitomycin C) and negative controls.
Mitomycin C is a known mutagen that induces polychromatic
erythrocytes.
It is possible that in vivo metabolism of a particular agent
could have mutagenic effects not identiable using the in vitro
Ames and chomosomal aberration designs and therefore the
mouse micronucleus test is require by the FDA as assurance
that the compound is not mutagenic. This in vivo genetic
toxicity study should be completed prior to Phase 2 initiation.
This of course depends on the nature of the compound and
dosing regimen.
NON-GLP MELANIN BINDING

Melanin is the pigment protein found in the iris, skin, and a
variety of other tissues of the body. The pharmacokinetic prole
of a topical ophthalmic agent is greatly influenced by whether or
not that particular agent binds to melanin. If the agent does
bind melanin, then the iris can act as a sink (holding drug until
it is broken down and metabolized by the body), or it can have
a sustained-release effect, binding melanin for a brief period of
time before being released, sometimes adding to duration of
effect. In either case, in vivo GLP toxicology studies must be
performed in pigmented animals if the active agent does in fact
bind melanin. As an example, the New Zealand White rabbit,
which is the standard albino rabbit used in toxicity testing,
should be substituted out in favor of the DutchBelted pigmented rabbit (or another pigmented species) if the test agent is
found to bind melanin. Failure to do so could produce falsepositive toxic effects that would not normally occur clinically (if
the iris acts as a sink and facilitates drug removal), or false
negatives (if melanin binding causes a sustained release and
prolonged effect.) A non-GLP melanin-binding study is a quick
and easy in vitro study that can be performed using synthetic
melanin or bovine melanin. The test agent is incubated for a
specic time period with synthetic melanin and is then ltered
and assayed for determination of melanin binding. Differences in
weight of free compound versus compound bound with melanin
are used to determine whether or not melanin binding occurs.
GLP OCULAR TOXICITY STUDY IN

RABBITS
The rabbit has been the preferred standard species for GLP
toxicology studies to support development of ophthalmic
products. While there are certainly differences between rabbit
and human eyes, the size of the rabbit eye and availability
(compared with primates) make them an ideal choice. The
GLP ocular toxicity study in rabbits is usually conducted with
concentrations higher than that which is anticipated as the
clinical dose, with elevated dosing frequency, and for a duration
that exceeds that of the intended proof of concept clinical trial.
As mentioned previously, this study should be performed in a
pigmented species if a positive result is achieved during the

melanin binding study. Both male and female animals are
enrolled to determine if any toxic effects are gender specic.
The GLP ocular toxicology study involves repeated dosing via
the intended clinical route. If the agent being screened is a
topical ophthalmic agent then animals should be dosed topically. If the agent is instead intended for intravitreal or subconjunctival administration, then dose should be administered by
intravitreal or subconjunctival route, respectively. Dosing
regimen generally elevates beyond intended clinical regimen to
provide a margin of safety. For example, a topical agent that is
intended for qd dosing may be administered as bid or tid while
an intravitreal agent intended for once a month dosing may be
administered twice a month.
Standard ophthalmic and systemic endpoints are evaluated
during the GLP ocular toxicity study. These parameters include
assessments of body weight and body weight gain, ophthalmic
observations (including fundoscopy, tonometry, slit-lamp biomicroscopy with fluorescein staining, corneal opacity, iris ndings, conjunctival redness, chemosis, and discharge), clinical
observations including appearance of fur, skin, eye and mucus
membranes, behavioral changes, necropsy ndings including
weights of major organs, and histology on all ocular tissues.
Ophthalmic observations are typically performed using the
MacDonnaldShadduct scoring system for ocular lesions.
Organs weighed during necropsy typically include the following
(liver, kidneys, adrenals, testes, prostate, ovaries, uterus, thymus,
lungs, spleen, brain, pituitary gland, heart, thyroid, submandibular gland, stomach, intestines, and pancreas). Any premature
deaths that occur during the toxicity study may need to be
evaluated more thoroughly with full organ histopathology,
depending on the nature of the death. Findings are considered
in terms of observed toxic effects and are distinguished by
animal sex and treatment arm. Dose ranging is typically required
as an element of GLP ocular toxicity studies. Up to three
concentrations are typically run against a vehicle control. Initial
ocular toxicology studies should also include (either as part of
the study or as a separate study) preliminary pharmacokinetic
assessments for systemic absorption.
CHAPTER 32
GLP OCULAR TOXICITY STUDY IN DOGS,

MONKEYS, OR OTHER SPECIES
The GLP ocular toxicity study in beagle dogs or monkeys
utilizes an identical design to the GLP study in rabbits and
fullls the FDAs requirement for a second species prior to
initiating clinical testing. As with the rabbit study, dosing regimen and dose given generally exceed that which is anticipated
clinically to ensure an appropriate safety margin. Dose ranging
should be included, with three concentrations being a standard
dose-ranging approach. Parameters evaluated should include
assessments of body weight and body weight gain, ophthalmic
observations (including fundoscopy, tonometry, slit-lamp biomicroscopy with fluorescein staining, corneal opacity, iris ndings, conjunctival redness, chemosis, and discharge), clinical
observations including appearance of fur, skin, eye and mucus
membranes, behavioral changes, clinical pathology, necropsy
ndings including weights of major organs, and histology on all
ocular tissues.
REQUESTS FOR WAIVERS ON PRE-IND

TOXICOLOGY REQUIREMENTS
Historically, ophthalmic drug products have often been developed as second-line products following systemic development
for similar indications. Topical antiinflammatory agents were
developed rst for systemic inflammation, many angiogenic
341

blockers developed for AMD were developed rst as anticancer
therapies, and many antihistamines developed for allergic
conjunctivitis were developed rst as systemic agents for
treatment of rhinitis. Given sufcient existing toxicology data,
the FDA may grant a waiver on systemic toxicity studies and
one of the required GLP ocular toxicity studies. Typically waivers
are requested for ethical considerations. The Pre-IND meeting
is an important time to review the proposed toxicology plan and
ensure that FDA has the opportunity to comment on the proposed toxicology studies. While systemic studies may be waived
or shortened, ocular toxicology with the nal ophthalmic
formulation is still needed.
SECTION 4
TOXICOLOGY STUDIES REQUIRED FOR NDA

SUBMISSION
With the IND submission, FDA will require toxicology data to
support the safety of the test agent for the duration of the
intended proof of concept clinical studies. As a component of
the NDA, the FDA will require toxicology studies that support
the long-term safety of the ophthalmic drug product. At the
time of NDA submission, several clinical studies will be completed, however the duration of dosing for these studies may
vary considerably depending on indication and design. To be
included in the NDA submission are chronic ocular toxicity
studies (with toxicokinetic assessments), reproductive toxicity
studies (segments I, II, and III), absorption, distribution, metabolism and excretion (ADME) studies to identify primary route
of excretion and distribution of drug in the ocular tissues
following dosing, acute or repeat systemic toxicity as needed, a
mouse micronucleus test to assess in vivo mutagenic potential,
and if applicable, carcinogenicity studies.
The proposed indication, route of administration, and existing database of studies performed with a particular agent will
greatly influence the toxicology testing required by the FDA for
NDA submission. The following are standard studies recommended by the FDA for approval of a topical ophthalmic agent.
CHRONIC OCULAR TOXICITY STUDIES
342
The chronic ocular toxicity study is often performed in parallel

with the phase-3 development program; however, for a drug
with a chronic indication, the study should be performed ahead
of the pivotal clinical studies. Typically, the FDA requires a
minimum of two species for a period of 6 months of repeat
ocular dosing for NDA approval of a topical ophthalmic agent.
If ocular and systemic toxicity following ocular dosing does not
appear to be a concern, or other approved ophthalmic drugs in
the same class demonstrated very low systemic and ocular
toxicity potential, chronic ocular toxicity studies in one most
appropriate species may be acceptable.
The chronic ocular toxicity study is usually designed
similarly to the GLP ocular toxicity studies performed prior to
IND submission. Route of administration should be consistent
with the intended clinical route of administration and some
dose ranging should be included. Frequently a recovery group is
employed to determine whether or not any toxicity noted at the
nal sacrice resolves following discontinuation of dosing. This
recovery group may be continued for 4 weeks to several months
after discontinuation of dosing. Doses given and dosing regimen
often approximate or exceed the intended clinical dose and
dosing regimen, with appropriate vehicle controls.
As described above, endpoints evaluated during the chronic
ocular toxicity study generally include assessments of body
weight and body weight gain, ophthalmic observations
(including fundoscopy, tonometry, slit-lamp biomicroscopy with
fluorescein staining, corneal opacity, iris ndings, conjunctival
redness, chemosis, and discharge), clinical observations including

appearance of fur, skin, eye and mucus membranes, behavioral
changes, necropsy ndings including weights of major organs,
and histology on all ocular tissues. Organs weighed during
necropsy typically include the following: liver, kidneys, adrenals,
testes, prostate, ovaries, uterus, thymus, lungs, spleen, brain,
pituitary gland, heart, thyroid, and pancreas. An added element
of the chronic toxicity study includes the evaluation of
pharmacokinetic parameters, hematology, coagulation, urinalysis,
and clinical chemistry. Pharmacokinetic parameters, such as
AUC, Cmax, and Tmax are calculated at specic timepoints during
the study to ensure that there is no buildup of drug following
repeated dosing. As with earlier studies, histology is performed
on all ocular tissues with full histopathology on any animals
that die prematurely. Pharmacokinetic assessments should be
completed after initial dosing and following repeated dosing
(e.g., at least following 1-month dosing).
SEGMENT I, II, AND III REPRODUCTIVE

TOXICITY
The effects of the ophthalmic drugs on all aspects of the
reproductive/developmental process (fertility and early embryonic
development, embryofetal development, and prenatal and
postnatal development) should be assessed to ensure that the
test agent does not adversely impact viability of offspring, cause
birth defects or impact fertility. If very low systemic exposures
occur and no systemic effects occur in repeated-dose studies,
it might be acceptable to perform only embryofetal development studies. If the drug is intended for use in women with no
child-bearing potential, reproductive toxicity studies in female
animals can be waived. These are standard study designs that
many appropriate contract laboratories are experienced in
performing.
ADME STUDY
An ADME study should be performed for submission as a
component of the NDA. The purpose of the ADME study is to
provide the FDA with information on how drug is metabolized,
distribution following dosing, and route of excretion.
Typically the ADME study is performed in rabbits or another
appropriate species. The study drug can be radiolabeled and a
known quantity of radioactivity is administered via the intended clinical route. Animals are sacriced at postdose timepoints
and ocular tissues are harvested for assessment of radioactivity.
Ocular tissues assessed include aqueous humor, upper and
lower eyelids, conjunctiva, cornea, iris/ciliary body, lens, optic
nerve, retina, choroid, sclera, and vitreous. Select organs may be
taken and assessed for levels of radioactivity. Plasma is taken at
each timepoint and assessed for levels of radioactivity. Urine
and feces are collected at specic increments following dosing
for determination of route of excretion.
The typical prole for a topical ophthalmic agent is to nd
the bulk of radioactivity is located in the cornea, bulbar and
palpebral conjunctiva (eyelids), and aqueous humor at early
assessment timepoints. The majority of the dose given is
excreted within 24 h of dose administration. With topical
ophthalmic products, it is often difcult to get a full mass
balance, as a portion of the dose may be lost during instillation,
even if particular care is taken during instillation. Percent
recovery may vary considerably. For example, per their approval
documents, Optivar (Azelastine HCl 0.05%) percent recovery of
radioactivity was 84% while with Elestat (Epinastine HCl
0.05%), percentage recover was 97% (Summary basis of approval
for optivar (NDA#021127) and Elestat (NDA#021565). The
purpose of the ADME study is not to necessarily raise any
ACUTE SYSTEMIC DOSE TOXICITY STUDY

The acute systemic dose toxicity study in the rat is performed
to determine the maximum tolerated dose (MTD) and the no
observable adverse effect level (NOAEL). This study features IV
administration using escalating doses to identify the level at
which adverse drug-related effects can be seen. For an ophthalmic agent, this is relevant as there should be a considerable
safety margin between NOAEL identied in the acute systemic
dose study and the anticipated systemic availability of the drug
when given via topical ocular dose. This study should employ
higher concentrations than are intended for clinical administration as some level of toxicity is desired. Again, several
difference doses should be tested.
In this study, a single intravenous infusion is performed.
Blood is collected prior to termination for hematology, clotting
prole, and clinical chemistry evaluations. Animals are evaluated for behavioral changes/clinical observations for a period of
14 days following dosing. Necropsies are performed on all
animals. Acute systemic toxicity studies can also be performed
with other routes (oral, subcutaneous, etc.).
CARCINOGENICITY STUDY
In many cases carcinogenicity studies are necessary to support
an NDA ling for ophthalmic products. The ICH Guidance
S1A states that carcinogenicity studies may be waived for drugs
given by the ocular route unless there is cause for concern or
unless there is signicant systemic exposure. Causes for concern include a previous demonstration of carcinogenic potential
in the product class, pharmacologic activity, a structureactivity
relationship suggesting carcinogenic risk, positive genotoxicity
results, evidence of preneoplastic lesions in multiple dose
toxicity studies, and/or long-term tissue retention of parent
compound or metabolites resulting in local tissue reactions or
other pathophysiological responses. A request for a waiver
should be submitted to the FDA if the sponsor considers the
drug eligible.
INTERPRETATION OF FINDINGS
In the event that abnormal ndings are noted during toxicology
studies, it is the obligation of the researcher to determine the
nature of the ndings, evaluate the severity, and determine the
clinical implications of the identied abnormalities. In some
instances, toxic ndings are immediate and obvious, causing
animal death and/or overt sickness and immediately halting the
development program. More frequently, however, toxicity does
not manifest in animal death or morbidity. In most instances,
signs of toxicity are subtle and may include differences in organ
weights or organ appearance at necropsy, presentation of some
abnormal ophthalmic signs (such as injection, chemosis,
staining, or discharge), mild alterations in clinical signs, animal
appearance or behavior, changes in clotting, hematology or
clinical chemistry, or abnormal hisptopathological ndings.
Subtle ndings may present with little or no additional
indication that the animals long-term health is in jeopardy. An
animal may act and appear healthy, gaining weight at the same
rate as controls and presenting with no obvious changes, only to
have abnormalities noted at necropsy. It is the obligation of the
researcher to determine the implications of ndings during
toxicology studies, and clinical relevance that subtle ndings
hold.
In general, ndings during GLP ophthalmic toxicology

studies should be interpreted based on any dose-related associations and whether or not ndings fall within accepted norms
when working with animals. A brief outline of interpretation
of ndings (necropsy, ocular signs, and histopathogy) is
included.
NECROPSY
Changes in organ weights/organ appearances at gross necropsy
are among the more common ndings associated with topical
ocular toxicity studies. It is important to note that statistically
signicant differences within treatment arms do not necessarily
indicate toxicity. The toxicology lab should have a database of
organ weight ranges that are considered normal and it is possible for organ weight to fall within these accepted norms and
still be statistically different from controls.
The toxicology reviewer will evaluate the extent of ndings
and look for an indication of a dose-related response. If a
change in mean organ weight was apparent between the low
dose group and the placebo control however the high-dose group
was not statistically different, then it is less likely that a toxic
effect is occurring. If, however, there is a dose-related effect,
then that can be an indication of toxicity. If the lowest dose
exhibits abnormalities, then the FDA may request additional
toxicology work.
OCULAR SIGNS
In the interpretation of positive ndings in ocular toxicity
studies, the incidence, severity, and reversibility should be
evaluated, and toxicological signicance and clinical relevance
should be considered. For example, minimal conjunctival redness is common in untreated animals and is considered as
normal and not toxicologically signicant.
In ocular toxicity studies, the drug is usually administered to
one eye, the other eye is used as an untreated control. Positive
ndings in the drug-treated eyes should be compared with those
in untreated eyes and in vehicle-treated eyes to gure out if the
ndings are drug-related, vehicle-related, or spontaneous
ndings. In studies with intravitreal injections, injection procedure-induced inflammation is not unexpected. If the
inflammation in control and drug-treated groups was similar
regarding severity, incidence, and reversibility, it is not toxicologically signicant.
CHAPTER 32
red flags on toxicity per se, but instead to provide a better

description of drug activity, or to help correlate with known
toxic effects.
HISTOPATHOLOGY
Histopathology changes provide an indication of toxicity at the
cellular level that might not be easily identiable during clinical
exams. The histopathology evaluation can successfully identify
a wide range of conditions, including bacterial or fungal
infection, cancer, or chronic inflammation. Histopathology
performed on dead or morbid animals can provide a diagnosis
of the animals condition and cause of death and is therefore an
important tool in the interpretation of toxic ndings. Logically,
an increased incidence of bacterial infection may indicate an
immunosuppressive effect of a test agent. Identication of
transformed cells may indicate mutagenic potential.
Histopathological ndings should be interpreted in a similar
fashion as other toxic ndings. It is important to consider any
dose-related response and it is also important to consider the
historical incidence of infection or disease as it relates to
histopathological ndings. A seemingly disproportionate rate of
infection in a particular treatment group may not be toxicologically signicant if incidence within the study population
falls within accepted norms.
343
SUMMARY
A toxic response occurs as the byproduct of drug metabolism
and generally presents in a dose-related fashion over time. As a
drug is broken down via a variety of different xenobiotic metabolic processes, oxygen radicals are produced. These and other
toxic byproducts cause the tissue damage that manifests as
clinical signs of toxicity. In contrast, irritation occurs via direct
binding of sensory neuron receptors in an acute fashion and
is generally not related to metabolism of the drug. Specic
qualities of the drug product, such as pH, contribute to its
irritation potential, rather than the ease by which it is
metabolized.
The Draize test was originally published in 1944 as a means
of evaluating the single-dose, acute irritation potential of nonophthalmic agents. Unfortunately, the test was frequently used
to assess the toxicity prole of ophthalmic drug products. The
original Draize methodology was not designed for this application, nor was it adequate to accurately evaluate the sometimes
subtle effects of long-term, repeated ophthalmic administration.
Improved methodologies have been developed for qualifying the
long-term toxicity prole of ophthalmic drug products. Current
methodologies evaluate not only ophthalmic clinical signs
following repeated dosing but histopathology, body weight and
body weight gain, necropsy parameters, clinical pathology,

hematology, behavioral changes, and clinical appearance.
The FDA requires that agents are safe, nonirritating, and
nontoxic, with an ample margin of error based on the amount
of drug administered. In general, doses selected and dosing
regimen should approximate or exceed clinical expectations
(preferably exceed). Studies submitted as components of the
IND include GLP ocular toxicity studies in at least two species,
genetox studies, melanin binding studies, and acute irritation
studies. The duration of GLP ocular toxicity studies should
exceed or at least equal the duration of proposed clinical studies
to again ensure an appropriate safety margin. Studies submitted
as components of the NDA include GLP chronic ocular toxicity
studies, reproductive toxicity studies, ADME, carcinogenicity
studies, and systemic toxicity studies.
The requirements for toxicology testing referenced above are
based on our experience as a clinical regulatory group and do not
represent ofcially regulatory guidance. We believe it is important
for pharmaceutical developers to contact the FDA early in the
development process and remain in close contact with the FDA
through the preclinical and clinical development phases. The
FDA will assist with conrming the appropriateness of the toxicology plan and can help with reviewing and interpreting results
of toxicology studies and implications for drug development.
SECTION 4
REFERENCES
344
1. Song ZH: A Schroeder Molecular basis of

ophthalmic toxicology. In: Chiou GCY, ed.
Ophthalmic Toxicology, 2nd edition.
Philadephia, PA. Taylor & Francis 1999;
2741.
2. Tafazoli S, Spehar DD, OBrien PJ: Oxidative
stress mediated idiosyncratic drug toxicity.
Drug Metab Rev 2005; 37:311325.
3. Asakura T, Shichi H: Cytochrome P450mediated prostaglandin omega/omega-1
hydroxylase activities in porcine ciliary
body epithelial cells. Exp Eye Res 1992;
55:377384.
4. Matsumoto K, Kishida K, Manabe R, et al:
Induction of cytochrome P-450 in the rabbit
eye by phenobarbital, as detected
immunohistochemically. Curr Eye Res
1987; 6:847854.
5. Kulkarni PS, Srinivasan BD: Cyclooxygenase
and lipoxygenase pathways in anterior
uvea and conjunctiva. Prog Clin Biol Res
1989; 312:3952.
6. Mclean KJ, Sabri M, Marshall KR, et al:
Biodiversity of cytochrome P450 redox
systems. Biochem Soc Trans 2005; 33(Pt
4):796801.
7. Michaelis UR, Fisslthaler B,
Barbosa-Sicard E, et al: Cytochrome P450
epoxygenases 2C8 and 2C9 are implicated
in hypoxia-induced endothelial cell
migration and angiogenesis. J Cell Sci
2005; 118(Pt 23):54895498.
8. Sugamori KS, Brenneman D, Grant DM: In
vivo and in vitro metabolism of arylamine
procarcinogens in acetyltransferasedecient mice. Drug Metab Dispos 2006.
9. Konishi M, Iwasa M, Yamauchi K, et al:

Lactoferrin inhibits lipid peroxidation in
patients with chronic hepatitis C. Hepatol
Res 2006; 36:2732.
10. Unger WG: Mediation of the ocular
response to injury and irritation: peptides
versus prostaglandins. Prog Clin Biol Res
1989; 312:293328.
11. Bandell M, et al: Noxious cold ion
channel TRPA1 is activated by pungent
compounds and bradykinin. Neuron 2004;
41:849857.
12. Klionsky L, et al: A polyclonal antibody to
the pre-pore loop of TRPV1 blocks channel
activation. J Pharmacol Exp Ther 2006;
319:192198.
13. Unger WG: Mediation of the ocular
response to injury and irritation: peptides
versus prostaglandins. Prog Clin Biol Res
1989; 312:293328.
14. Draize JH, Woodard G, Calvery HO:
Methods for the study of irritation and
toxicity of substances applied topically
to the skin and mucus membranes.
J Pharmacol Exp Ther 1944;
82:377390.
15. Marzulli FN, Maibach HI: Eye irritation. In:
McDonald TO, Shadduck JA, eds.
Dermatoxicology. 1977:579582.
16. Green K, Bowman KA, et al: Dose-effect
response of the rabbit eye to
cetylpyridinium chloride. J Toxicol Cutan
Ocul Toxicol 1985; 4:1326.
17. Zhivov A, et al: In vivo confocal microscopy
of the ocular surface. Ocul Surf 2006;
4:8193.
18. Parikh C, Sippy BD, Martin DF,

Edelhauser HF: Effects of enzymatic
sterilization detergents on the corneal
endothelium. Arch Ophthalmol 2002;
120:165172.
19. Maehara S, Osawa A, Itoh N, et al:
Detection of cone dysfunction induced by
digoxin in dogs by multicolor
electroretinography. Vet Ophthalmol 2005;
8:407413.
20. Imai R, Sugimoto S, Ando T, et al: A
procedure for recording electroretinogram
and visual evoked potential in freely moving
cats. J Toxicol Sci 1990; 15:263274.
21. Blomstrand R, Ingemansson SO: Studies
on the effect of 4-methylpyrazole on
methanol poisoning using the monkey as
an animal model: with particular reference
to the ocular toxicity. Drug Alcohol Depend
1984; 13:343355.
22. Magnusson B, Kligman AM: The
identication of contact allergens by animal
assay. The guinea pig maximization test.
J Invest Dermatol 1969; 52:268276.
23. Kligman AM, Basketter DA: A critical
commentary and updating of the guinea
pig maximization test. Contact Dermatitis
1995; 32:129134.
24. Hozier J, Applegate M, Moore MM: In vitro
mammalian mutagenesis as a model for
genetic lesions in human cancer. Mutat Res
1992; 270:201209.
CHAPTER
33
Toxicology of Ophthalmic Agents by Class

Millicent L. Palmer, Robert A. Hyndiuk, Mark S. Hughes, Ann Sullivan Baker, Kristine
Erickson, Alison Schroeder, James McLaughlin, Keith Lane, Sarkis H. Soukiasian,
Michael B. Raizman, and Cynthia Mattox
In this chapter, we will outline common toxic and irritative

effects associated with frequently prescribed classes of
ophthalmic drugs. Focus will be on identifying general trends
associated with specific drug classes rather than listing side
affects recorded for individual ophthalmic drugs. Covered under
this chapter are antiinflammatory agents, antiallergics, ocular
hypotensives, antiinfectives, and antiangiogenesis agents
prescribed for age-related macular degeneration (AMD).
adverse effects caused by local (topical and periocular)

administration of corticosteroids (CSs), the most commonly
used antiinflammatory agents. We also review the adverse
ocular and systemic effects of other agents used to treat ocular
inflammations, including nonsteroidal antiinflammatory drugs
(Nonsteroidal Antiinflammatory Drugs [NSAIDs]), antihistamines and decongestants, mast cell-stabilizing agents, and
immunosuppressive drugs. This section focuses primarily on
conditions of the anterior segment.
Key Features
Corticosteroids, well known for their highly effective

antiinflammatory profile, are associated with a variety of side
effects including ocular hypertension, cataract formation, and
delayed wound healing and increased susceptibility to
infection.
Nonsteroidal antiinflammatory drugs in general have a less
adverse side affect profile compared to steroids. These agents
do impact platelet function and therefore carry a risk of
increased ocular bleeding. In addition, NSAIDs have been
shown to inhibit corneoscleral wound healing and create some
susceptibility to infection.
Topical antiinfective are associated with a wide range of side
effects, including superficial irritation, chemosis, conjunctival
necrosis, epithelial toxicity, and macular infarction (following
intravitreal administration).
Ocular hypotensive agents produce a wide range of side
effects, given that different classes of hypotensive agents act
via a variety of different mechanism. Beta-blockers are well
known for systemic side effects including bradycardia and
respiratory distress. Adrenergic agonists are known to cause
mydriasis and are associated with a high rate of drug-induced
allergy. Prostaglandins, generally considered the optimum
treatment for management of ocular hypertension, may cause
photophobia, conjunctival injection, pain, breakdown of the
blood aqueous barrier evidenced by anterior chamber cell and
flare, and changes in pigmentation and eyelash growth.
Antiangiogenesis drugs for use in treating AMD act primarily by
blocking the function of VEGF. While these agents are relatively
new to the market, systemic availability of an administered
anti-VEGF agent could in theory impact the natural function of
VEGF in the body. In addition to its role in angiogenesis, VEGF
has both vasodilative and neuroprotective effects.
TOXICOLOGY OF CORTICOSTEROIDS AND

OTHER ANTIINFLAMMATORY AGENTS
The use of antiinflammatory agents is common in ophthalmic
practice because inflammation, a nonspecific response to tissue
injury,1,2 is frequently encountered. In this chapter, we review
SIDE AFFECTS OF CORTICOSTEROIDS

Systemic Side Effects due to Local Corticosteroid
Administration
Systemic side effects due to topical administration are unusual
even with long-term therapy,3 however, measurable and physiologically significant systemic effects associated with frequent
topical use of concentrated preparations of potent CSs have
been reported. Burch and Migeon4 described a 1972% reduction of urinary excretion of 17-hydroxy CSs after bilateral
administration of 0.01% dexamethasone every 2 h for 4 days
(daily systemic dose 0.75 mg); the cortisol production rate
decreased by more than 50% during the experimental period.
Prednisone, 10 mg, or its equivalent daily for 4 weeks may
suppress normal growth in the pediatric population.3 The
administration of a single drop of 0.1% dexamethasone sodium
phosphate four times daily in each eye yields a systemic dose of
0.25 mg.3 Lowering of plasma cortisol levels after 6 weeks of
such a regimen has been observed; the hypothalamicpituitary
axis functions normally as measured by metyrapone testing.5
Prolonged orbital injections of CSs may also have systemic side
effects, such as adrenal suppression.6
Ocular Hypertension and Glaucoma

One of the most well-known side affects associated with topical
corticosteroid use is an elevation in intraocular pressure (ocular
hypertension). Ocular hypertension and glaucoma have been
well documented after both topical and systemic CS
administration.7,8 In 1950, McLean7 suggested that topical
steroid therapy might increase intraocular pressure (IOP)7;
the first case of cortisone glaucoma was reported by Francois
in 1954.8
The doseresponse relationship is particularly important in
understanding this undesirable side effect.911 Steroid-induced
ocular hypertension can reach clinically significant levels in
~36% of normal subjects on a short-term steroid regimen.10 A
differential susceptibility among individuals expressed as a
skewed distribution has been observed.9 A more pronounced
effect of increased IOP or disturbed aqueous fluid dynamics
has been noted in individuals with suspected glaucoma,1218
345
SECTION 4

in those with myopia,19 in older patients, in patients with
glaucoma,1215,17,18,2021 in relatives of glaucoma patients,17,22,2326
in patients with Krukenbergs spindle,27 and in diabetic
patients.28
A timeresponse relationship of CS-induced ocular
hypertensive response has important implications. A clinically
significant rise in IOP typically, but not always, requires greater
than 12 weeks of topical therapy.10 With systemic steroid
administration, the ocular hypertensive response may require
longer treatment.29 A dose-dependent response after systemic
CS therapy has also been observed.30 A smaller concentration of
drug reaches ocular sites by the systemic route and may explain
the difference in the IOP response time between topical and
systemic routes of administration.10 The magnitude of the
ocular hypertensive response after systemic administration has
been noted to be similar to the response following topical
therapy.29
The mechanism of the steroid ocular hypertensive response
appears to involve an initial increase in aqueous inflow, as
suggested by Linner,31 with a secondary effect on the facility of
outflow.32 Several biochemical mechanisms have been proposed
to explain the decrease in outflow facility based on CS effects on
cells of the trabecular meshwork. These may include inhibition
of prostaglandin mediators33 and alteration of glycosaminoglycan production or metabolism.32,34,3537
The CS-induced ocular hypertensive response is usually
reversible, especially with short-duration therapy (weeks)10;
however, irreversible steroid-induced glaucoma has been clearly
documented, especially in patients with myopia.34,19,3841
Discontinuation of CSs may result in normalization of the IOP,
but visual field abnormalities and optic nerve damage may be
permanent.34,38,39 Many patients respond to medical antiglaucomatous therapy. When medical treatment fails and the
continued use of CSs is required to control ocular inflammatory
disease, argon laser trabeculoplasty and glaucoma filtering procedures may be required.39
Corticosteroid-Induced Cataract Formation
346
It is generally accepted that CSs are cataractogenic, commonly

producing posterior subcapsular cataracts (PSCs). The association of PSCs with systemic CS use was first reported by Black
and associates in 1960.42 PSCs developed in patients receiving
moderate or high doses of CSs for greater than 1 years
duration. Further work by Oglesby and co-workers,43 Giles and
colleagues,44 and Crews45 indicated that patients receiving
doses of less than 10 mg/day of prednisone or its equivalent or
patients receiving CS therapy for less than 1 year were unlikely
to develop PSCs. However, cataracts have been observed after
even short-term CS therapy,46 and some authors now argue
against the concept of a safe noncataractogenic dose.47 Both
systemic and topical administration of CS may induce PSC
formation; those caused by systemic use are bilateral.4250
Although steroid-induced PSCs occurrence is dose and duration
dependent, the precise relationship of lens changes to the total
dose, the intensity of the dose, and the duration of therapy is
not fully understood.47 Some studies suggest that individual
susceptibility and perhaps even genetic determinants may be
important.46,47,51 Children46,50 and diabetic patients52,53 appear
to be more susceptible.
The pathophysiology of steroid-induced cataracts is similar
to the mechanism of cataract formation proposed for galactose
in that steroids increase the influx of cations,54 resulting in an
increase in the cellular water content, producing cellular
intumescence and disparity of the refractive index from that of
the surrounding medium.39 Glucocorticoids also bind to
specific amino acid groups of the lens cell fibers, leading to a
conformational change and exposure of buried sulfhydryl
groups.39 These moieties (i.e., the sulfhydryl groups) form

disulfide bonds and create protein aggregation and a change in
the refractive index.39
Delayed Wound Healing and Effects on Corneal

Reepithelialization
The effect of CSs on corneal wound healing has been the focus
of several investigations.5568 Corneal wound integrity has been
evaluated by determining the tensile strength,58,59,61,63,64,69
histologic appearance55,58,60 and uptake of tritiated thymidine by
keratocytes.62 Although the results of these studies are somewhat inconsistent, a doseresponse effect of CSs on corneal
wound healing has been demonstrated.63 Impaired corneal
wound healing has been less pronounced when steroids were
withheld until after the tenth postoperative day, after which
topical CS treatment did not significantly interfere with the
tensile strength of the healing wound.69
Topical and systemic cortisone derivatives have a depressant
effect on many phases of the healing process. Alterations in
fibroblast proliferation, vascularization, and deposition of extracellular matrix have been observed.56 CSs primarily affect
stromal healing to a greater extent than epithelial healing.
Effects of CSs on corneal epithelial healing have been observed
and may be related to the extent of epithelial injury. Topical CSs
do not impair epithelialization after partial corneal denudation,66,67 but impairment is observed after complete denudation
in a rabbit model.70
Investigative studies have demonstrated that the enzyme
collagenase is produced in Pseudomonas and herpes simplex
corneal ulcers, alkali burns, and ulcerations associated with
collagen vascular diseases and StevensJohnson syndrome.65
CSs may induce rapid destruction or corneal melting and even
perforation in these conditions, possibly by enhancing
collagenase activity.3,65
Corticosteroids and Infectious Keratitis

Because CSs alter the host immunologic responses to infection,
their use in the presence of an active infectious process is often
contraindicated.3,67 In addition, chronic use of CSs may alter
normal and pathogenic flora of the lids and conjunctiva.71 The
incidence of corneal thinning and perforation in severe
infectious keratitis may be increased owing to the potential
enhancement of collagenolytic enzymes or decreased collagen
synthesis and wound healing.3,72 In certain cases, judicious use
of CSs may be appropriate to limit the structural damage related
to the inflammatory process. In general, the use of CSs should
be avoided until the infectious process has been controlled by
specific antimicrobial therapy. In the following sections,
important issues regarding the use of CSs are briefly reviewed
by the category of infectious agents: viral (herpes simplex),
bacterial, and fungal.
Herpes Simplex Virus

Topical CS therapy is contraindicated in the presence of active
viral replication associated with herpes simples virus (HSV)
epithelial keratitis.73 The deleterious effects of CSs in
management of HSV infection have been clearly
documented.7476 Local CSs, however, do not appear to
reactivate latent HSV keratitis or stimulate an episode of
dendritic or stromal keratitis.77 CS therapy plays a role in
controlling the immunologically mediated inflammation of
HSV stromal disease. The results of the Herpetic Eye Disease
Study, a multicenter, randomized, double-masked clinical trial,
revealed the efficacy of topical CSs in HSV stromal keratitis.78
The initiation of CS therapy should be avoided if steroids were
never used previously. Clinically, a careful risk-benefit analysis
should be made on an individual basis.
Bacterial Keratitis
The risks associated with CS therapy in the management of
bacterial keratitis are a subject of controversy. Several reports
favor adjunctive CS therapy in bacterial keratitis. Davis and coworkers79 demonstrated that concurrent treatment with CSs
did not inhibit the effect of antibiotics in Pseudomonas
keratitis. Aronson and Moore80 claimed that CS treatment promoted the resolution of inflammation associated with
infectious keratitis in their series. A favorable visual outcome,
however, was observed only in mild cases of paracentral
keratitis. Leibowitz and Kupferman81 concluded that the
concurrent use of topical CSs with an effective bactericidal
antibiotic regimen did not enhance the replication of Staphylococcus aureus or Pseudomonas aeruginosa if the CS was not
instilled more frequently than the antibiotic.
CSs have been shown to enhance P. aeruginosa replication
within the cornea if there is inadequate antimicrobial
therapy.82,83 Animal studies have indicated that despite 5 days
of treatment with an effective antibiotic, corneas infected with
Pseudomonas were not sterilized.84 Further recurrences of
Pseudomonas keratitis have been reported in eyes treated with
CSs.85 CSs are contraindicated in eyes that have advanced
corneal thinning with the potential for perforation, owing to
possible enhancement of collagenolytic enzymes or inhibition
of collagen synthesis. A controlled prospective clinical study by
Carmichael and co-workers86 evaluated CS therapy with and
without antibiotic therapy in bacterial ulcers and found no
differences in visual outcome. Considering the risks, steroids
should not be used if there is not a significant chance of
preventing visual loss or of recovering lost vision; that is,
control of inflammation alone should not be the deciding factor.
Fungal Keratitis
The use of CSs in the early treatment of fungal keratitis is
generally contraindicated owing to an enhancement of growth
of both yeast and opportunistic fungi.87 A clinical worsening of
fungal keratitis has been demonstrated after treatment with
CSs.8891 In contrast to the number of available antibiotics,
there are relatively few antifungal agents. In general, these
agents are poorly soluble and have limited ocular penetration;
therefore, ocular bioavailability and the efficacy of antifungal
agents are less than ideal, reaching only fungistatic, as opposed
to fungicidal, corneal levels.92 CSs may negate the effects of
antifungal therapy and suppress host immune responses. Host
immune responses may be particularly critical in controlling
the inflammation of keratomycoses. The use of CSs to reduce
stromal scarring, intraocular inflammation, and corneal
neovascularization continues to be controversial.93 Adjunctive
CS therapy should be considered only in combination with an
effective antifungal agent or agents in the later stages of a
healing fungal keratitis.93
NONSTEROIDAL ANTIINFLAMMATORY DRUGS

NSAIDs have analgesic, antiinflammatory, and antipyretic
properties.94102 The antiinflammatory activity is related to the
inhibition of the enzyme cyclooxygenase; this enzyme is
responsible for the conversion of arachidonic acid to prostaglandins, which are potent inflammatory mediators.94100 These
agents are widely used in the treatment of musculoskeletal
disorders such as osteoarthritis, rheumatoid arthritis, ankylosing spondylitis, and acute gout.94 Topical NSAIDs have been
approved for pain, postoperative inflammation, inhibition of
miosis during intraocular surgery, and for photophobia.
Oral NSAIDs have been useful in the management of
patients with uveitis, particularly recurrent anterior uveitis.96
Foster96 has reported that diflunisal (Dolobid) was the safest
and most effective; naproxen (Naprosyn) and indomethacin

(Indocin-SR) were of intermediate efficacy; and piroxicam
(Feldene), sulindac (Clinoril), and ibuprofen (Motrin) have been
the least effective. Long-term maintenance therapy on oral
NSAIDs may help to control inflammation caused by anterior
uveitis without steroids and thereby reduce the steroid
requirement.96 In addition, oral NSAIDs may play a role in the
management of cystoid macular edema (CME) associated with
cataract surgery, posterior uveitis and secondary retinal
vasculitis; NSAID therapy has not been of benefit in the
management of primary retinal vasculitis.96
These agents share several important systemic side effects.94
The most common is the induction of gastric or intestinal
ulceration. In some cases, anemia from gastrointestinal blood
loss may occur. Gastrointestinal side effects are explained on
the basis of two mechanisms. First, local irritation by orally
administered agents allows back-diffusion of acid into the
gastric mucosa, resulting in tissue damage. Inhibition of the
biosynthesis of gastric prostaglandins that inhibit gastric acid
secretion and induce gastric secretion of cytoprotective mucus
in the intestine is the proposed mechanism.94
Additional untoward effects of these agents that are related to
inhibition of the synthesis of endogenous prostaglandins
include altered platelet function, impairment of renal function,
and prolongation of gestation or spontaneous labor.94 NSAIDs
prevent the formation by platelets of thromboxane A2, a potent
platelet-aggregating agent. This results in an increased bleeding
time.94 These agents have a known effect on renal
hemodynamics and fluid and electrolyte balance. In normal
patients, little effect of NSAIDs is seen because the production
of vasodilatory prostaglandins plays a minor role in the
presence of normal sodium balance.94,103 NSAIDs, however,
promote a decrease in renal blood flow and glomerular filtration
in patients with congestive heart failure, chronic renal disease,
hepatic cirrhosis with ascites, or hypovolemia of any cause.94
Salt and water retention may also be induced secondary to
the reduction of prostaglandin-induced inhibition of both
reabsorption of chloride and function of antidiuretic hormone.94
Edema may result in some patients. Hyperkalemia is also
promoted by the use of NSAIDs.94,103 Another renal side effect
is acute interstitial nephritis, with nephrotic-range proteinuria
in 73% of cases.103 Renal failure may be severe enough to
require temporary dialysis in 32% of patients.103 Propionic acid
derivatives have been most often associated with acute
interstitial nephritis.103
NSAIDs bind firmly to plasma proteins and therefore may
displace certain other drugs from binding sites.94 Thus, with
concurrent use of drugs such as warfarin, sulfonylurea
hypoglycemic agents, or methotrexate, an adjustment in the
dosage of these drugs may be required.94 This is particularly
important in patients receiving the anticoagulant warfarin, in
view of the effect of NSAIDs on platelet function.
Use of these aspirin-like agents is contraindicated in patients
with hypersensitivity to NSAIDs and in those with the
syndrome of nasal polyps, angioedema, and bronchospastic
response to aspirin.96 The use of NSAIDs in children should be
restricted to those agents extensively tested in the pediatric
population, namely aspirin, naproxen, and tolmetin.94 Owing to
the association of Reyes syndrome with aspirin treatment of
children with febrile viral illness, NSAIDs should be strictly
avoided in this clinical setting.94
NSAIDs have the potential to produce photosensitivity reactions and are a frequent cause of cutaneous reactions. Cutaneous
reactions such as vesiculobullous eruptions, serum sickness,
exfoliative erythroderma, erythema multiforme, and toxic epidermal necrolysis are well summarized in a report by Stern and
Bigby.104 Reactions to piroxicam were reported most frequently.104
CHAPTER 33
347
SECTION 4
Ocular Side Affects of Systemic NSAIDS
348
Adverse ocular effects of systemic NSAIDs have been reported;

however, in many cases these are isolated reports or data
obtained from retrospective studies in which a cause-and-effect
relationship cannot be clearly established.105 Generally,
NSAIDs are photosensitizers and have the potential for
inducing phototoxicity of the anterior and posterior segments of
the eye.105 Optic neuritis has been associated with this class of
drugs and is presumed to occur as an idiosyncratic response
that is reversible on cessation of therapy.105 In the case of
ibuprofen, a widely used NSAID, there have been enough
occasional cases in which the drug has been rechallenged that
changes in refractive error, diplopia, and diminished color vision
seem to be well documented.105107 The occurrence of a
reversible toxic amblyopia has also been described.108111
Patients taking this drug should be advised to stop if a sudden
decrease in vision occurs.
Interpretation of reports of indomethacin-induced retinal and
macular disease is complicated by almost equal numbers of
contradictory studies.105,112,113 Nevertheless, the potential for
ocular toxicity exists. There are data to support the occurrence
of superficial corneal crystalline deposits secondary to
indomethacin that resolve with discontinuation of the
drug.112,114 There have also been several reports of papilledema
associated with pseudotumor cerebri.105,115
Aspirin has been implicated in increasing the incidence of
rebleeding in traumatic hyphema.116 Therefore, this agent as well
as the aspirin-like NSAIDs should be avoided in this condition.
Adverse Affects of Topical Ophthalmic NSAIDS

The ophthalmic NSAIDs currently available include flurbiprofen sodium 0.03% (Ocufen), suprofen 1% (Profenal),
diclofenac sodium 0.1% (Voltaren), bromfenac sodium 0.09%
(Xibrom), nepafenac sodium 0.1% (Nevanac), and ketorolac
tromethamine 0.5% (Acular).97 Both flurbiprofen and suprofen
are approved for the prevention of intraoperative
miosis.97,98,116118 Ketorolac tromethamine 0.5% is indicated for
the treatment of ocular itch due to seasonal allergic
conjunctivitis.119,120 Diclofenac sodium 1%, bromfenac sodium
0.09%, nepafenac 0.1% and ketorolac tromethamine 0.5% are
indicated for the treatment of postoperative inflammation in
patients who have undergone cataract extraction.117 Nepafenac
is also effective in reducing pain following cataract extraction.
Diclofenac sodium 1%, and ketorolac tromethamine 0.5% have
also been effective in the reduction of pain and inflammation
after excimer laser photorefractive keratectomy.121,123
Several double-masked, randomized studies of the effects of
flurbiprofen on postoperative inflammation have been
published.123126 Topical administration of flurbiprofen has also
been shown to reduce the inflammation of experimental
anterior uveitis.127 These studies indicate that flurbiprofen does
have some potential as an antiinflammatory agent, but
additional well-controlled clinical trials are needed.
In general, flurbiprofen sodium 0.03% is well tolerated. The
most frequent side effect is transient burning and stinging with
instillation.117 Flurbiprofen has been shown to inhibit
corneoscleral wound healing127,128 and exacerbate epithelial
HSV keratitis,129 effects similar to those seen with topical CSs.
A more recent report by Asbell and co-workers,130 however,
demonstrated that flurbiprofen sodium did not enhance HSV
epithelial keratitis. The strain of HSV, however, was not
specified in this report, and the timing of CS intervention after
infection differed. In a review of topical antiinflammatory
agents in an experimental model of microbial keratitis, a
worsening of Pseudomonas keratitis with topical CSs was
confirmed, and a greater worsening was observed with
flurbiprofen sodium 0.03%.131 Concomitant therapy with an
effective antibiotic prevented the steroid- and flurbiprofeninduced worsening of Pseudomonas keratitis. Pneumococcal
keratitis was not worsened by the use of either CSs or
flurbiprofen in the presence of appropriate antimicrobial
therapy.131
There have been reports that flurbiprofen sodium may
promote bleeding of ocular tissues in the setting of ocular
surgery, particularly in the case of concomitant systemic
dipyridamole, an antiplatelet agent, or oral NSAIDs.117,132 The
manufacturers of all of the currently available topical NSAIDs
advise caution with the use of these agents in patients with
bleeding disorders or individuals taking systemic medications
that may prolong the bleeding time.117,133
A double-masked study evaluating the effects of topical
flurbiprofen sodium 0.03% on the IOP revealed that this agent
did not alter the IOP in known CS responders. In this study,
treatment with flurbiprofen did not prevent the steroid-induced
increase in IOP or the decrease in outflow facility.134
Suprofen may cause minor irritation, itching, redness,
allergic reaction, iritis, pain, chemosis, photophobia, and punctate keratopathy.116 The use of diclofenac may be associated
with minor symptoms of irritation. Concurrent use of
diclofenac and hydrogel contact lenses may cause burning and
redness.116 Ketorolac tromethamine 0.5% ophthalmic solution
may cause mild, transient burning and stinging on
instillation.119
A case of asthma exacerbated by topical ketorolac has been
reported.135 Caution must be exercised in prescribing topical
NSAID eye drops for patients with a history of asthma, nasal
polyps, and allergy to aspirin or NSAIDs.
Bromfenac sodium and nepafenac are newer agents with
recent approvals and consequently there is less clinical safety
data available on these agents. Initial clinical trials for both
agents demonstrated rare occurrences of a variety of side effects.
Both agents site potential for drug-induced allergy and increased
bleeding of ocular tissues due to interference with platelet
aggregation.
ANTIHISTAMINES AND DECONGESTANTS

Side affects of topical antihistamines are generally mild. Antihistamines may cause allergic responses and local irritation.71
H1 blockers also have local anesthetic properties; however, the
concentrations required for this effect are much greater than
those used therapeutically to antagonize the histamine
response.136
Adverse systemic reactions to topical decongestants (alpha-1
adrenergics) are uncommon, but headache,137139 dizziness,138
nervousness,140 hypotension,141,142 hypertension,137,138,143-145
and cardiac dysrhythmias have been reported. The most
commonly reported ocular side effect is stinging on instillation.
Blurred vision,139,145 mydriasis,146150 epithelial erosions,151,152
punctal stenosis,153,154 corneal pigment deposition,255 iris
pigment release,148,156 iritis,157 change in IOP137,146,148,149,158,159
and acute angle closure have also been described.139,149,150
A case series report identified acute and chronic conjunctivitis due to over-the-counter ophthalmic decongestants.161
Three clinical patterns in order of decreasing frequency were
observed and include: (1) a pharmacologically induced rebound
conjunctival hyperemia, (2) a toxic follicular conjunctivitis, and
(3) an allergic, eczematoid blepharoconjunctivitis. The authors
note that the longer the duration of eye drops use before
presentation, the longer the recovery period required.
Systemic Antihistamines
Oral antihistamines have a drying effect on the eye that may
worsen or induce keratoconjunctivitis sicca and cause contact
lens intolerance.160,162 Chlorpheniramine maleate, a commonly

prescribed oral antihistamine, which is also available over the
counter, has been shown to decrease tear production significantly, as measured by standard Schirmer testing, in normal
patients.163
The most common adverse effect of systemic administration
of antihistamines is drowsiness.161,136,164 This may be
hazardous to those patients who must drive or operate
machinery. These agents also enhance the action of narcotics
and sedatives.136 Astemizole, terfenadine, loratadine, and
fexofenadine have fewer sedative and anticholinergic side
effects. 165,136 Gastrointestinal side effects of oral antihistamines
such as nausea, emesis, anorexia, epigastric distress, and altered
bowel habits may be reduced by ingestion of the medication
with meals.161 Less common central nervous system effects
may include lassitude, dizziness, tinnitus, incoordination,
blurred vision, diplopia, euphoria, nervousness, tremors, and
insomnia.161 The anticholinergic action of these drugs may
induce mydriasis, triggering acute angle-closure glaucoma as
well as a reduction in accommodation by effects on ciliary
muscles.105,161 Rare side effects of visual hallucinations,
temporary blindness, and an absence of pupillary light reflexes
have been induced by overdosage.161
Serious adverse cardiovascular events, including death, cardiac
arrest, torsades de pointes, and other ventricular arrhythmias
have been reported with concomitant use of terfenadine with
erythromycin and related macrolide antibiotics, ketoconazole,
or itraconazole, and significant hepatic dysfunction. Use of
terfenadine is therefore contraindicated in these situations.
Ocular Mast Cell-Stabilizing Agents

Commonly reported side effects of disodium cromoglycate are
transient burning and stinging on instillation.117,166,167
Hyperemia and bulbar conjunctival chemosis have been
reported in 35% of patients.167 Less common adverse effects
include watery and itchy eyes, sties, puffiness, and dryness
around the eyes.168 EDTA, a solution stabilizer in DSCG, has
been implicated as the cause of conjunctival injection in
some cases.168
Clinical trials have indicated that treatment-related ocular
adverse effects of lodoxamide are mild, nonserious, and
transient. Reported adverse effects include minor discomfort,
itching, and pain.169171 Headache171 and nausea169 are
nonocular side effects that have been reported in rare instances.
Both disodium cromoglycate and lodoxamide may decrease
the steroid requirement, thus reducing the potential adverse
effects from long-term corticosteroid therapy.166,168,169173 Acute
exacerbations and severe forms of ocular allergy may require
topical steroids, however. In these cases a pulse steroid regimen
with aggressive but brief corticosteroid treatment with rapid
tapering and maintenance therapy with these mast cellstabilizing agents may have a therapeutic advantage while
minimizing risks of steroid therapy. The therapeutic effect of
mast cell-stabilizing agents is not as immediate as that of
corticosteroids, taking usually several weeks of regular use for a
desired therapeutic response. Patients should be advised of this
when using these drugs, or compliance may be a problem.
Cyclosporine
Topical Restasis (cyclosporine A) is the only prescription
therapy available for treatment of dry eye. There are few side
affects associated with topical cyclosporine application however
systemic CsA therapy is associated with a number of side
effects. Nephrotoxicity has been observed following systemic
dosing, manifested by decreased creatinine clearance, elevated
serum creatinine levels, and a disproportionate increase in
blood urea nitrogen with preserved urine output and sodium
reabsorption.174 It is important to note that the serum

creatinine level underestimates the glomerular filtration rate
and therefore should not be the sole marker of renal toxicity.175
The renal toxicity occurs at the level of the arteriole,
glomerulus, and proximal tubule.39,174 CsA-induced alteration
in renal hemodynamics has been proposed.101 Systemic
hypertension is another significant side effect, occurring in 25%
of patients following systemic CsA therapy; it tends to be more
frequent in those with impaired renal function.39 The exact
mechanism remains unknown but appears to be dose related.
Hypertension is also more common in patients receiving CsA
and steroids than in those receiving CsA alone.176 Leukopenia
is not seen with systemic CsA; however, a normochromic
normocytic anemia is observed in 25% of patients, and other
causes of anemia should be ruled out.39 An increase in the
erythrocyte sedimentation rate has been noted in 40% of
patients, but this does not correlate with the clinical course of
the underlying disease and should not be used as an index of
disease activity.39 An increased incidence of lymphoma was
once thought to be related to CsA use; however, in a large
clinical series of 5000 transplant recipients, the incidence of
lymphoma was no greater in patients receiving CsA than in
those receiving other immunosuppressive agents.177 Other side
effects of CsA include hirsutism, gingival hyperplasia, central
nervous system toxicity, and an increased incidence of viral
infections.39,94,178,179 Several important drug interactions have
been observed with the administration of CsA.39,176,179,180187
Ocular side effects reportedly due to systemic use include
decreased vision, eyelid or conjunctival erythema, nonspecific
conjunctivitis, urticaria, visual hallucinations, and conjunctival
and retinal hemorrhages related to drug-induced anemia.105
TOXICITY OF OCULAR ANTIINFECTIVES
CHAPTER 33
Topical ophthalmic antiinfectives are prescribed in response to

diagnosis of an active ocular infection and as a preventative
measure in situations where threat of ocular infection exists
(trauma, injury, etc) Common side effects associated with
topical and systemic antibiotics, antivirals and antifungal
agents will be described, with emphasis on the more frequently
prescribed agents of these classes.
ANTIBIOTICS
The antibiotic section briefly covers the ocular toxicity of a
variety of different antibiotics with emphasis placed on the
ocular toxicity of aminoglycosides, cephalosporins, fluoroquinolones, saulfanomide and vancomycin, because they are
more commonly employed. For many antibiotics, the mechanism of toxicity is ill understood.
Aminoglycosides
The aminoglycosides are bactericidal antibiotics that
irreversibly inhibit protein synthesis and ribosome function.
Amikacin, gentamicin, neomycin, and tobramycin are in
common use today. The narrow therapeutic index of the
aminoglycosides has stimulated extensive investigation into the
mechanism of aminoglycoside toxicity.
Extensive studies of the mechanism of aminoglycoside
toxicity have been performed owing to the oto- and
nephrotoxicity associated with systemic use of aminoglycoside.
Previous investigators have demonstrated selective accumulation of aminoglycoside within the lysosomes of cultured
fibroblasts.188,189 This may result from the protonation of
aminoglycoside molecules, thus trapping the drug in the lowpH lysosomal environment. The accumulation of aminoglycoside produces disturbances in phospholipid catabolism,
349
SECTION 4
350
possibly through lowered activity of sphingomyelinase and

phospholipases as demonstrated by Aubert-Tulkens and
associates in cultured rat fibroblasts.188
Libert and colleagues190 evaluated the cellular toxicity of
subconjunctival gentamicin. Electron microscopy revealed an
accumulated substance within the lysosomes that consisted of
granular material with a pleomorphic lamellar structure,
corresponding to the presence of complex lipids. Laurent and
co-workers191 demonstrated that systemic gentamicin induces a
loss of activity of lysosomal sphingomyelinase and
phospholipase A in rats. Furthermore, they found that amikacin
binds more loosely to phospholipid bilayers, induces less
inhibition of phospholipases in vitro, and is taken up less by
tubular cells in vivo. Given amikacins lesser nephro- and
retinotoxicity, lysosomal alterations may be an early step in
aminoglycoside-induced toxicity.
Intravitreal injection of aminoglycoside antibiotics is an
established mode of therapy for bacterial endophthalmitis.
DAmico and co-workers192 have evaluated the comparative
toxicity of intravitreal aminoglycoside antibiotics in a rabbit
model. The observations ranked gentamicin as the most toxic,
then tobramycin, and amikacin as the least toxic: full-thickness
retinal necrosis was induced with 800, 1600, and 3000 mg,
respectively. Toxic lesions secondary to intravitreal aminoglycoside injection consist of focal areas of lysosomal storage,
with outer retinal necrosis, whereas other areas of the retina
appear normal with the exception of mild accumulation of
complex lysosomal lipids. Owing to the focal areas of toxicity
and the potential that pigmented eyes may raise the threshold
for toxicity,193 it is particularly difficult to determine the
threshold toxic dose.
After intravitreal aminoglycoside injection, the area of
toxicity is localized to the RPEphotoreceptor outer segment
complex.194,195 The production of lamellar lysosomal inclusions
in the RPE indicates the accumulation of complex lipids. The
aminoglycoside concentrates within the lysosomes and may
interfere with one or more lysosomal enzymes, causing
accumulation of unmetabolized substrates.194 A number of
possible mechanisms have been proposed. Gentamicin-treated
cells exhibit a greater decrease in the activity of sphingomyelinase than amikacin.188,189 Alternatively, the cytoplasmic
enzyme phospholipase C correlates well with the nephrotoxicity. Finally, investigators have proposed a direct effect on
the mitochondria, with disruption of oxidative phosphorylation
with aminoglycoside toxicity. Fleisher and associates196 reported
that intraocular injections of tunicamycin produce photoreceptor-specific degeneration. The glycosylation of opsin can
be blocked by tunicamycin in vitro in conditions where polypeptide synthesis is only slightly decreased. Thus, aminoglycoside toxicity may be mediated by disturbance in
glycoprotein metabolism. Certainly, some combination of these
mechanisms could be responsible for the pathogenesis of
aminoglycoside toxicity.
Some have proposed that the toxicity of intravitreal
antibiotics can also be effected by the surgical status of
the eye.197 Talamo and co-workers198 showed that posterior capsulectomy and vitrectomy do not change the therapeutic index
(toxic dose:therapeutic dose) for intravitreal aminoglycoside
despite the dramatic reduction in vitreous half-life. Retinal
damage may be related to the peak concentration of the drug to
which the retina is exposed after intravitreal injection. No
additional protection from aminoglycoside toxicity is noted
after vitrectomy.
Macular infarction has been reported after intravitreal
aminoglycoside injection.199,200 Conway and associates201
evaluated the effect of intravitreal gentamicin in the primate
retina. Intravitreal gentamicin doses of 1000 and 3000 mg were
employed. The inner retinal layers exhibited considerable

swelling of the nerve fiber and ganglion cell layers; however,
the outer segments and RPE appeared normal by light
microscopy. Electron microscopy of the 3000-g intravitreal
gentamicin specimens revealed intracellular edema with
massive thickening of the ganglion cell axons. A prominent
inflammatory response was noted on the internal limiting
membrane. Despite these findings there was no evidence of
retinal vasculitis. In areas of the retina that had shown
nonperfusion of the capillary bed, granulocyte plugs were seen
filling the vessels. The authors hypothesize that the inflammation of the inner layers of the retina associated with the toxic
effects of gentamicin may induce granulocytic plugging with
permanent closure of the capillary bed.201,202 Granulocytes
obstruct the lumen by adhering to the endothelium; this
occlusion, combined with oxygen-free radical formation and
lysosomal enzyme activity, may cause ischemic injury.203 The
findings of Conway and co-workers provide strong evidence that
gentamicin toxicity occurs in normal retinal tissue.201 This
effect is in keeping with the known neurotoxicity of
gentamicin.204 Tabatabay and associates205 examined the
immunohistochemical localization of gentamicin in the rabbit
after a single intravitreal injection. Initially, gentamicin was
localized to the ganglion cell layer, inner plexiform and nuclear
layers, and the photoreceptors. By 24 h, gentamicin was
predominantly in the RPE and choriocapillaris. Haines and
associates evaluated the morphologic changes after intravitreal
injection of gentamicin in pig eyes. Three mg of gentamicin was
injected intravitreally to observe the toxicity-related changes
that occurred in the retina. Vacuolization of the nerve fiber layer
and perivascular swelling was seen within 6 h and subsequently
descended deeper into the retina. Vascular endothelial cells,
photoreceptors, and the RPE appeared to be spared from the
toxic effect of gentamicin. By 48 and 72 h after injection,
numerous large and small retinal vessels showed congestion
and leukocyte margination. These changes could not be
prevented by changing the pH to the gentamicin to 7.2. Thus,
the authors conclude that the gentamicin toxicity effect is not a
pH-related phenomenon but that the primary targets for
gentamicin are the neurons and the glia of the inner retina, and
as a result, retinal infarction occurs secondarily owing to
leukocyte plugging.206 In addition, Haines and colleagues
speculate that the predisposition for macular infarction is due
to the dependent position of the macula during surgery as well
as the higher density of ganglion cells in the perimacular
area.206
A survey of retinal specialists from the Retina, Macula, and
Vitreous Societies revealed 101 cases of macular infarction due
to aminoglycoside administration.207 Interestingly, 93 cases
were associated with gentamicin, five with amikacin, and three
with tobramycin. Of the 93 gentamicin cases, 21 used
intravitreal doses of 100-200 mg, a dose considered to be nontoxic. Twenty-three cases of gentamicin toxicity resulted from
prophylactic subconjunctival injections after cataract
extraction. Although dilution errors cannot be ruled out, this
reference clearly points out that the safe therapeutic window for
ocular use of aminoglycoside is sufficiently narrow to be a
significant clinical problem. The authors advocate reserving
aminoglycoside for known or highly suspicious gram-negative
infections.207 They recommend: (1) abandoning routine use of
subconjunctival aminoglycoside after ocular surgery (using
cefazolin instead), and (2) avoiding intravitreal aminoglycoside
in the prophylaxis of penetrating ocular trauma.208 The
authors207 recommend vancomycin (or clindamycin) and
ceftazidime or imipenem210 for penetrating ocular trauma.
More recently, Campochiaro and Lim reported on the results
of a survey of 13 patients who received 200400 mg of amikacin

sulfate or 100200 mg of gentamicin sulfate for prophylaxis
or treatment of endophthalmitis. Low-dose gentamicin or
amikacin can cause macular infarction, even with doses
prepared by hospital pharmacists using typewritten protocol. Of
note, several cases exhibited very discrete macular involvement,
causing the authors to speculate on the role of a localized
increase in concentration in dependent areas of the retina.210
Generally, retinal specialists recommend vancomycin, 1 mg,
and ceftazidime, 2.25 mg, for intravitreal injections. However,
the choice of antibiotic may vary based on the patients clinical
circumstances.
of both 500 and 1000 mg but not with 250 mg of ciprofloxacin.

In addition, electrophysiology revealed that the amplitude ratios
were significantly reduced after the 1000-mg dose. At the 100or 250-mg ciprofloxacin dose, histologic sections were comparable between control eyes, and ERG ratios were unchanged from
the baseline level.217
Kawasaki and associates found that 200 mg of ofloxacin did
not cause deterioration of the b-wave, c-wave, or the oscillatory
potential over a 2-month period in the rabbit model.218 Mochizuki
and colleagues studied the effects of ofloxacin on the rabbit ERG
in vivo.219 They also determined that 200 mg of ofloxacin did
not cause an alteration in the ERG in the rabbit model.219
Cephalosporins
Fluoroquinolones
The fluoroquinolones are structurally related to nalidixic acid.
These agents block enzymatic activity of bacterial DNA gyrase
and alter the structure and functioning ability of bacterial DNA.
Currently, ciprofloxacin, norfloxacin, and ofloxacin are
available in the United States, although a number of fluoroquinolones are under study, including pefloxacin.212 Topical
0.3% ciprofloxacin is tolerated well. Only mild untoward ocular
events are noted, the most frequent one being a white
crystalline precipitate, commonly located in the superficial
portion of the corneal defect. This precipitate has been identified as ciprofloxacin. Hobden and colleagues demonstrated the
efficacy of transcorneal iontophoresis of 1% ciprofloxacin for
therapy of aminoglycoside-resistant Pseudomonas keratitis.213
They found no evidence of toxicity with the 1% formulation.
Stamer and co-workers214 evaluated the effect of ciprofloxacin
on rabbit corneal endothelial viability. A concentration of 10
g/mL of ciprofloxacin had no effect on endothelial cell counts
or viability, whereas 100 mg/mL caused a 2% reduction in viable
endothelial cells. Haller-Yeo and associates215 evaluated
intravitreal ciprofloxacin; they reported no toxicity in cat eyes
with doses of 1, 10, 100, and 1000 mg when evaluated by light
microscopy and electrophysiology. Steven and co-workers216
evaluated the intraocular use of ciprofloxacin in phakic and
aphakic rabbits. Corneal decompensation occurred in aphakic
vitrectomized rabbits with intravitreal doses of 100 g of
ciprofloxacin, whereas retinal toxicity was noted on electron
microscopy with doses higher than 250 mg. At 1000 g, electron
microscopy revealed loss of the outer rod segments, followed
by atrophic changes of the inner rod segments as well as the
outer and inner nuclear cell layers. Marchese and colleagues
evaluated the toxicity and pharmacokinetics of ciprofloxacin.
They studied the pigmented rabbit model and injected doses of
100, 200, 500, and 1000 mg. An evaluation was performed by
indirect ophthalmoscopy, electrophysiology study, and
histology. Focal areas of retinitis were observed after injections
Sulfonamides
The sulfonamides interfere with bacterial utilization of paminobenzoic acid (PABA). These drugs are bacteriostatic, and
the various preparations have different chemical, pharmacologic, and antibacterial properties. Flach and associates220
demonstrated that topical 25% sulfisoxazole diolamine
ointment and exposure to ultraviolet light resulted in a phototoxic reaction. Boettner and co-workers221 found that topical
use of sulfadiazine ointment for 1 year caused formation of
multiple small white concretions in cysts of the palpebral
conjunctiva, identified by spectroscopy as sulfadiazine. Hook
and colleagues141 reported a case of transient myopia induced by
sulfonamides. A-scan measurements and cycloplegic refraction
demonstrated the primary mechanism of sulfonamide-induced
myopia to be lens thickening from ciliary body edema.
Vancomycin
Vancomycin is a bactericidal antibiotic that inhibits bacterial
cell wall synthesis through interference with glycopeptide
polymerization. Pryor and associates222 evaluated topical and
subconjunctival administration of vancomycin in rabbits. They
found no evidence of toxicity with subconjunctival doses of
12.5 or 25 mg, whereas a 5% aqueous solution given every 5 min
for 30 min revealed minimal superficial punctate keratopathy.
Fortified vancomycin in doses of 1425 mg/mL has been reported
to cause irritation, conjunctival injection, and superficial punctate keratopathy. Although subconjunctival vancomycin injections have been reported to cause conjunctival necrosis and
sloughing,223 Lindquist and co-workers demonstrated the safety
and efficacy of vancomycin in corneal storage media.224 They
found no evidence of endothelial damage at doses of 150 mg/mL
vancomycin in gentamicin-free Dex-Sol. Kattan and Pflugfelder225
evaluated the corneal toxicity of vancomycin in corneal storage
media and found no evidence of endothelial damage with concentrations of 5 mg/mL. Garcia-Ferrer and associates223 evaluated
the antimicrobial efficacy and lack of corneal endothelial
toxicity of Dex-Sol corneal storage medium supplemented
with vancomycin, 10 mg/mL. Choi and Lee226 documented an
8.8% decrease in endothelial cell counts with transcorneal
iontophoresis of vancomycin in rabbit eyes, compared with
5.4% decrease with balanced saline solution.
Intravitreal vancomycin has been evaluated extensively.
Homer and associates evaluated the toxicity, clearance, and
therapeutic effectiveness of intravitreal vancomycin in a rabbit
model of staphylococcal endophthalmitis.155 Concentrations of
vancomycin ranged from 0.25 to 500 mg/0.1 mL. At doses
higher than 5 mg/0.1 mL the vitreous exhibited a whitish
reaction, although ERG abnormalities were associated with
doses higher than 2 mg. Histologic study of doses of 25 mg
revealed toxicity localized to the retina with photoreceptor outer
segment degeneration. In contrast, Smith and co-workers228
evaluated the toxicity, clearance, and efficacy of intravitreal
vancomycin in an experimental rabbit model of methicillinresistant S. epidermidis endophthalmitis. Doses of 1, 2, and
CHAPTER 33
Cephalosporins are b-lactam antibiotics that interfere with

bacterial cell wall synthesis. The first-generation cephalosporins
include cefazolin, cephalothin, cephapirin, cephradine, cephaloxin,
and cefadroxil. The second-generation cephalosporins include
cefamandole, cefuroxime, cefonicid, cefoxitin, ceforanide, and
cefaclor. The third-generation cephalosporins include cefepime,
cefoperazone, cefotaxime, ceftizoxime, moxalactam, ceftazidime,
and ceftriaxone. Although many cephalosporins have been
evaluated for use in ocular infections, cefazolin, ceftazidime,
and cefuroxime are the most often used cephalosporins. The
intravitreal use of cefazolin has been supplanted mainly by
vancomycin.
Common side effects of these agents include superficial
irritation (conjunctival injection, chemosis, conjunctival
necrosis and lid edema.211 Some photoreceptor toxicity has been
noted in preclinical studies performed with intravitreally
administered ceftazidime.102
351
SECTION 4

5 mg were evaluated by light microscopy. There were no
discernible retinal abnormalities except one eye injected with
5 mg confirmed the absence of extensive toxicity. Smith and
co-workers believe that the histologic change noted by Homer
and co-workers155 might be the result of tissue processing.
Borhani and associates evaluated vancomycin in the vitrectomy
infusion solution. They found that concentrations of 8, 16, and
32 g/mL of vancomycin in infusion solution caused no
abnormal ERG or histologic changes. However, electrophysiologic depression and abnormal histologic changes
occurred with concentrations of 100 mg/mL of vancomycin in
the infusion solution.228
Pflugfelder and colleagues229 evaluated the retinal toxicity,
clearance, and interaction of intravitreal vancomycin with
gentamicin in phakic and aphakic vitrectomized rabbits.
Clinically, with doses higher than 2 mg there was immediate
clouding of the vitreous and within 24 h opacification of the
retina. By 2 weeks the retinal opacification had cleared, but the
RPE showed pigment clumping and atrophy. Electrophysiologic
testing revealed no evidence of toxicity up to 2 mg; however,
there was marked reduction in the a- and b-wave amplitudes in
the eye that received a 5-mg dose. Ultrastructural studies of
doses greater than 2 mg revealed a number of pathologic
changes, including: (1) hypertrophy of the RPE with abnormal
clustering of pigment granules in the cytoplasm; (2) loss of
photoreceptor outer segmentRPE interdigitation due to
retraction of apical microvilli of RPE; (3) appearance of lucent
vacuoles in the RPE basal cytoplasm beneath the plasmalemma
infoldings; (4) gross disorganization of the photoreceptor outer
segments with distention and displacement of the inner segments past the external limiting membrane; and (5) accumulation of cellular debris in the subretinal space. Pflugfelder and
colleagues229 found that lensectomy and vitrectomy increased
the intraocular clearance of vancomycin but did not alter the
threshold for retinal toxicity. Oum and associates230,231 studied
the effect of combined and repeated injections of intravitreal
vancomycin and aminoglycoside. They found increasing retinal
toxic reaction with repeated injections. The exact biochemical
mechanism of toxicity is unknown.232
ANTIFUNGAL AGENTS
Ocular fungal infections continue to challenge ophthalmologists.233 The selection of appropriate antifungal chemotherapy is limited by the paucity of effective drugs.234,235 The
only approved ophthalmic antifungal is 5% natamycin;
however, amphotericin B, flucytosine, miconazole, and
ketoconazole have a role in the management of ocular fungal
infections.236,237
Natamycin
352
Natamycin is a tetraene polyene and is the only antifungal

available in the United States in a topical form. Topical natamycin is well tolerated. Superficial punctate keratopathy has
been reported with prolonged use 238,239 Foster and coworkers240 demonstrated that natamycin did not retard the
healing of corneal epithelial defects. Ellison and Newmark158
demonstrated conjunctival necrosis after subconjunctival
injection of natamycin.
Intraocular use of natamycin is not well tolerated.241 Anterior
chamber injection of 250 mg of natamycin is tolerated in a rabbit
model; however, with a dose of 500 mg corneal decompensation
and iridocyclitis develop. Ellison and Newmark242 reported the
intravitreal effects of pimaricin: 25 mg was not toxic but was not
therapeutic either; doses higher than 50 mg destroy the retina.
Other formulations of amphotericin B have been evaluated
topically. Amphotericin B methyl ester, which is water soluble
and has reduced toxicity, has been evaluated in a 1%

formulation and has been found to be nontoxic and to penetrate
better than amphotericin B.243 Owing to leukoencephalopathy
associated with systemic administration, further work on
amphotericin B methyl ester has not been pursued.
Intravitreal injection of amphotericin B has been studied in
detail. Foster and associates244 reported a case of Volutella
fungal infection after cataract extraction that was treated with
three intravitreal injections of 35-40 mg over 1 month. The eye
was sterilized, although blind with a corneal pannus, updrawn
pupil, and total, funnel-shaped retinal detachment. Green and
co-workers245 reported successful sterilization of a postcataract
fungal infection with 20 mg of intracameral amphotericin B
combined with topical and subconjunctival amphotericin B,
although the final acuity was extremely poor. Axelrod and
associates246 evaluated the toxicity of intravitreal amphotericin
B in a rabbit model. They found that doses of 25500 mg of
intravitreal amphotericin B resulted in retinal detachment with
a proteinaceous exudate and cloudy vitreous with monocytes in
the vitreous cavity. They proposed that the amphotericin B
alters cell membranes, with resultant transudation of subretinal
fluid. Of note, they found that sodium deoxycholate is not toxic
to the retina, and intravitreal doses of 510 mg of amphotericin
B produced no abnormalities by electrophysiologic testing or
light microscopy. In addition, 25 mg of amphotericin B injected
close to the retina resulted in immediate focal retinal necrosis.
Axelrod and Peyman247 demonstrated that, in the setting of
experimental fungal endophthalmitis, 5 mg of intravitreal
amphotericin B was nontoxic (as determined by light
microscopy) in rabbits. Souri and Green248 documented that
intravitreal doses of amphotericin B as small as 1 mg resulted in
focal retinal necrosis in the rabbit when injected adjacent to the
retina.
Different formulations and delivery systems for intravitreal
amphotericin B have been evaluated. Amphotericin B methyl
ester, although it has much less antifungal activity, is a watersoluble compound with a much wider range of therapeutic
doses. McGetrick and associates185 found that amphotericin B
methyl ester showed no evidence of retinal toxicity by light
microscopy or electrophysiologic studies when intravitreal
doses were 50 mg or less. Doses of 100 mg of amphotericin B
methyl ester resulted in degeneration of the photoreceptor layer;
this was caused by the drug and not by the ascorbic acid used to
solubilize the antifungal agent. Raichand and co-workers249
evaluated the toxicity of amphotericin B methyl ester in
vitrectomy infusion fluid and found that the maximal nontoxic
dose was 75 mg/mL; at 100 mg/mL. Electrophysiologic studies
revealed a decreased response, although no toxic damage was
appreciated by light microscopy. Amphotericin B methyl ester
was found to cause leukoencephalopathy when used systemically and has not, therefore, been a candidate for
intraocular use.
Imidazoles
The imidazoles are a group of synthetic antifungals that are
fungistatic in low concentration and fungicidal in high
concentrations. They possess a broad spectrum of antifungal
activity. They inhibit ergosterol synthesis at low concentrations
and interfere with the mitochondrial oxidative and peroxidase
enzymes.
Ketoconazole
Ketoconazole is a weakly dibasic synthetic imidazole that
inhibits ergosterol synthesis. Foster and co-workers240,250
evaluated the toxicity of 1% ketoconazole with Cremophore EL
as the carrier. They found a slight delay in corneal epithelial
wound healing. Grossman and Lee251,252 evaluated transscleral
Itraconazole
Itraconazole is a triazole derivative with broad-spectrum
antifungal activity in vitro and in animal models. This
antifungal drug is lipophilic and practically insoluble in water.
Schulman and colleagues injected intravitreal itraconazole in
doses ranging from 10 to 100 mg devolved in 100% DMSO into
the eyes of New Zealand rabbits. Ocular toxicity studies
performed 5 weeks after administration showed no substantial
retinal or histologic changes in eyes injected with either 100%
DMSO or 10 mg of itraconazole. Higher doses cause focal areas
of retinal necrosis.255
Fluconazole
Fluconazole is a bis-triazole, potent antifungal with low toxicity
and excellent water solubility; it is currently available in oral
and intravenous forms. Brooks and associates256 found that
topical fluconazole, 100 mg/mL, appeared to be equivalent to,
and potentially less toxic than, amphotericin B in an experimental Candida keratitis model. Schulman and co-workers255
evaluated the toxicity of intravitreal fluconazole in the rabbit.
They found no corneal, lenticular, or retinal changes by light
microscopy and no evidence of depressed electrophysiologic
testing at doses of 100 mg. Fluconazole has excellent ocular
penetration when taken systemically; further work is required
to evaluate the efficacy and toxicity of ocular fluconazole
treatment.
ANTIVIRALS
Great strides have been made in the chemotherapy of ocular
viral diseases since the introduction of idoxuridine in 1962.
Currently, ophthalmic preparations of idoxuridine (IDU),
vidarabine, and trifluridine (TFT) are available. Acyclovir is
available for systemic use and as a dermatologic preparation;
ganciclovir is available for systemic and intraocular use.257
Idoxuridine.
The adverse ocular effects of topical IDU are common and
result from direct toxicity or allergic reactions. Local irritation,
with conjunctival injection, follicular conjunctivitis, allergic
blepharoconjunctivitis,258 and perilimbal filaments have been
reported.259 Lass and associates260 have reported IDU-induced
conjunctival cicatrization. Corneal problems such as superficial
punctate keratitis, delayed corneal wound healing, and corneal
edema have been reported.259 Punctal scarring and occlusion

have also been reported, particularly after long-term therapy.
The mechanism for the observed toxicity is believed to be the
activation of IDU in normal cells, particularly rapidly dividing
cells, resulting in disruption of normal DNA synthesis.
Vidarabine
Adenine arabinoside (vidarabine, Ara-A) is a purine analog. AraA is phosphorylated by viral thymidine kinase, then
triphosphorylated. The active form inhibits DNA polymerase
and ribonucleotide reductases, thus blocking viral DNA
synthesis. Ara-A is available in a 3% ophthalmic ointment and
an intravenous suspension (200 mg/mL).257
Similar to IDU, the adverse effects of Ara-A are due to direct
toxicity or to allergic reactions. Local ocular reactions include
conjunctival injection, follicular conjunctivitis, and punctal
scarring. With prolonged treatment, conjunctival cicatrization,
corneal scarring, or permanent punctal occlusion can result.261
Lass has reported that Ara-A has insignificant effects on corneal
epithelial wound healing, although there is significant delay in
stromal wound healing.261 Kaufman and associates262 have
found that subconjunctival injections of Ara-A can be toxic;
daily subconjunctival injection of 5% Ara-A results in
significant conjunctival inflammation; 20% injections result in
the formation of conjunctival granuloma.
Different methods of delivery of Ara-A have been evaluated.
Hill and associates263 found that iontophoresis (0.5 mA in 4
min) of Ara-AMP (vidarabine adenosine-5-phosphate) resulted
in higher corneal and intracameral levels without evidence of
toxicity. Pulido and associates264 evaluated the toxicity of
intravitreal injections and infusions of vicarabine in rabbits.
Intravitreal injections of 80 mg/0.1 mL Ara-A revealed no
abnormalities in electrophysiologic testing or light microscopy.
However, after vitrectomy/lensectomy, disorganization of the
external retina was visible by light microscopy in rabbits that
received infusions of 100 mg/mL Ara-A.
CHAPTER 33
and transcorneal iontophoresis of ketoconazole in a rabbit

model. Subconjunctival ketoconazole (50-mg) injections were
compared with iontophoresis and produced no evidence of
toxicity at the doses employed. Intravitreal ketoconazole in
dimethyl sulfoxide (DMSO) was evaluated by Yoshizumi and
Banihashemi.253 The ocular toxicity of experimental intravitreal
DMSO has been evaluated254; a single 0.1mL injection of
100% DMSO results in transient focal retinal edema and a 50%
decrease in the amplitude of the photopic, flicker fusion,
scotopic, and combined photopic and scotopic response.
Electrophysiologic response returned to normal after 1 month
and retinal edema resolved within a week. Intravitreal
injections of ketoconazole in DMSO at doses of 2240 mg
resulted in retinal edema and necrosis with marked photoreceptor outer segment loss; electron microscopy of the RPE
revealed degeneration of mitochondria and a decline in the
number of melanin granules.253 Doses of 720 mg of intravitreal
ketoconazole produced toxic vacuolizations of the inner
segments of the photoreceptors detected by electron microscopy.
The study determined that doses up to 540 mg produced no
ocular toxicity, giving a much wider therapeutic window than
miconazole.
Trifluridine
TFT is a halogenated pyrimidine with three fluorines in place of
the 5-methyl group of thymidylate. It is a potent inhibitor of
thymidine synthesis, which in turn inhibits DNA synthesis. It
is preferentially incorporated into viral DNA, thus producing
defective DNA. TFT is available as a 1% ophthalmic solution.
The adverse ocular effects of topical TFT are due to direct
toxicity or to allergic reaction. Local reactions include conjunctival injections, superficial punctate keratopathy, filamentary keratitis, and punctal occlusion with prolonged
treatment.259 Udell265 has reported conjunctival cicatrization
after topical TFT, whereas Maudgal and associates266 have
reported corneal epithelial dysplasia after TFT.
Carmine and associates267 reported no evidence of toxicity of
1% TFT in normal rabbit eyes. However, with a standard
corneal epithelial defect and 8 days of eight-times-daily 1% TFT,
they noted pathologic changes in the regenerating epithelium,
which resolved when the TFT was discontinued. They also
noted that stromal wound healing was affected with decreased
tensile strength; this was confirmed by Gassett and Katzin.263
Wellings and associates268 have demonstrated that TFT is more
effective than IDU, associated with fewer failures and less
toxicity, although the toxicity may reflect failure to control the
herpetic keratitis rather than a toxic reaction to IDU. Hyndiuk
and associates269 documented a case of reversible crystalline
epithelial keratitis with 1% TFT, which presented with
superficial punctate keratopathy and gray epithelium with fine
linear retractile crystalline intraepithelial deposits. Maudgal
and associates270 did report on conjunctival ischemia, corneal
epithelial dysplasia, filamentary keratitis, and punctal stenosis
353

with the use of 2% TFT (with 1% neomycin) in an experimental
model of herpes simplex keratouveitis. The complications
increased with prolonged use.
Peyman and associates212 have investigated the toxicity of
intravitreal TFT. Pang and associates270 found that intravitreal
injections of 200 mg/0.1 mL and vitrectomy infusion solutions
of 60 mg/mL were not toxic to rabbits. With injections of 500 mg
a mild decrease in ERG functions was noted; however, no
damage was seen by light microscopy. With injection of 1000 mg
(and infusions of 100 mg/mL) there was a moderate depression
in b-wave amplitudes, and photoreceptor clumping and
degeneration were noted by light microscopy. Liu and
associates271 evaluated liposomal delivery of intravitreal TFT.
Injections of 42.9 mg revealed no evidence of toxicity by clinical
examination, ERG, or light microscopy, and vitreal drug levels
remained for 28 days in the range of ID46 for many strains of
herpesvirus and human cytomegalovirus (CMV).
consisted of clinical examinations and light microscopy. DiaLlopis and colleagues reported no evidence of toxicity with
intravitreal injection of 1200 mg in an AIDS patient with CMV
retinitis; again toxicity was evaluated by post mortem light
microscopy.277
In order to evaluate the effect of repeated intravitreal injections of foscarnet, Turrini and co-workers evaluated the retinal
toxicity of two, four, and six intravitreal injections of 3.6 mg
of foscarnet in 16 pigmented rabbits using ophthalmoscopy,
histology, and electrophysiology.278 All rabbits revealed evidence
of yellowish punctate retinopathy in the midperiphery and
posterior pole after the first injection. After four or six
injections, there was a significant decrease in the scotopic ERG,
whereaas after six injections there was a significant decrease in
the mesopic ERG. Of note, light microscopy revealed mild
vacuolization and rarefaction in the photoreceptors and inner
nuclear layers. After six intravitreal injections, focal areas of
photoreceptor layer destruction was observed.278
SECTION 4
Acyclovir
Acyclovir, a purine analog similar to Ara-A, is activated by
virus-induced thymidine kinase to the monophosphate form
and then to the triphosphate form. It inhibits viral DNA polymerase. Acyclovir is available as a 5% dermatologic ointment, a
3% ophthalmic ointment (not available in the United States),
and in oral and intravenous formulations. The adverse ocular
effects of topical acyclovir (not approved for ophthalmic use) are
mild: local irritation with mild superficial punctate keratitis and
follicular conjunctivitis.274 One report of punctal stenosis with
the topical preparation was reported,261 but it is not known
whether stenosis was due to acyclovir or to herpes zoster
keratouveitis.
Because of the low toxicity of acyclovir, it has been investigated for intraocular injection. Pulido and associates275
investigated intravitreal injections and infusion solutions of
acyclovir in rabbits. They found that injections of 240 mg/0.1
mL revealed no evidence of abnormalities on histopathologic
examination or electrophysiologic testing. Infusion solutions
containing 400 mg/mL revealed disorganization of the external
layers of the retina after lensectomy or vitrectomy.
Ganciclovir
Ganciclovir is a synthetic nucleoside analog of 2-deoxyguanosine, similar to acyclovir. Virus-specified thymidine
kinase converts ganciclovir to the monophosphate form, which
is then converted to the di- and triphosphate form, which
competitively inhibits virus DNA polymerase, thus preventing
viral replication. Ganciclovir is available for systemic use.
The rising incidence of AIDS and the widespread use of
immunosuppressive drugs have caused an increase in CMV and
herpesvirus retinitis. In particular, ganciclovir is effective in
CMV retinitis; unfortunately, its systemic toxicity (bone
marrow suppression) has limited its use. Much research has
focused on intraocular injections and new delivery systems for
intraocular ganciclovir.
Foscarnet
354
Phosphonoformate (PFA, foscarnet) is a highly water-soluble

pyrophosphate analog that effectively inhibits in vitro
replication of HSV-1 and HSV-2, varicella zoster, and CMV
through noncompetitive binding to the exchange site of virus
DNA polymerase, thus blocking viral DNA synthesis. Foscarnet
has been used as an alternative to ganciclovir in the systemic
treatment of CMV infections. A number of investigators have
evaluated the effects of intravitreal foscarnet. She and coworkers found that single doses of intravitreal foscarnet in doses
ranging from 200 to 1000 mg/0.1 mL are nontoxic to the retina
in New Zealand albino rabbits.276 Of note, their evaluation
TOXICOLOGY OF ANTIGLAUCOMA DRUGS

Therapeutic agents designed to treat ocular hypertension reduce
intraocular pressure either by reducing aqueous humor
production or by enhancing aqueous humor outflow (through
the trabecular meshwork or via uveoscleral pathway). These
agents are commonly prescribed as ocular hypertension impacts
a high percentage of the elderly population. In this section,
ocular side effects will be broken down by agent classes.
CHOLINERGIC DRUGS
Much has been learned about the ophthalmic toxicology of
cholinergic drugs from testing in animals and humans, from
observations in accidental poisonings, and in connection with
medical uses of both direct-acting agents and anticholinesterases in the treatment of glaucoma, accommodative
strabismus, and myasthenia gravis. Most commonly known
effects include miosis (constriction of the pupil), induction of
pupillary cysts, enhancement of accommodation (i.e.,
adjustment of the lens of the eye to focus on near objects),
formation of cataracts, and reduction of IOP. A variety of other
effects are less well known or are less well established. Systemic
poisoning, which has occasionally been caused by the use of
anticholinesterase eye drops, is manifested by both muscarinic
and nicotinic symptoms, which can include paralysis of the
respiratory muscles mediated by stimulation of nicotinic
receptors. Although the ocular and systemic side effects of
anticholinesterases are generally more frequent, similar
patterns are noted with direct-acting muscarinic agents such as
pilocarpine. Therefore, direct-acting agonists and anticholinesterases are considered as a group.
ADRENERGIC AGONISTS
Nonselective adrenergic agonists, such as epinephrine,
stimulate both a- and b-adrenergic receptors. They exert their
therapeutic effect on IOP by reducing aqueous outflow
resistance. Although adrenergic agonists have been used in the
treatment of POAG for over 100 years, the site of action of the
outflow resistance-decreasing effect is unknown. The
mechanism of action is through stimulation of the b-adrenergic
receptors.279283 Adrenergic receptors are found on almost all
ocular tissues, and adrenergic agonists are known to affect a
number of ocular physiologic parameters, including smooth
muscle tone in the iris and ciliary body, aqueous humor
production, and intraorbital and extraorbital vascular tone. The
incidence of cardiovascular stimulation after topical ocular
treatment is a potential major side effect that limits the

therapeutic use of epinephrine. The prodrug formulation
dipivefrin has allowed smaller doses of epinephrine to be
administered, limiting the risk of adverse systemic effects.
In in vitro studies, epinephrine in clinically relevant doses
was toxic to trabecular cells.284 In contrast, examination of the
outflow pathway tissue of normal cynomolgus monkey eyes
after 6 months of topical treatment with epinephrine revealed
no apparent toxicity to the outflow pathway tissues.285
However, the ciliary muscle appeared to be displaced anteriorly,
narrowing the chamber angle. Also, changes in the ciliary
processes consistent with hypersecretion were noted in some
sections and hyposecretion in others.286
Experimentally, epinephrine dramatically reduces blood flow
to the ciliary processes in the cynomolgus monkey287 and the
albino rat288 but not to the ciliary muscle288 as observed by a
functional resin-casting method. Similar results were noted
using radioactive microspheres in the albino rabbit eye, in
which 2% epinephrine administered topically three times a day
over a 56-week period resulted in decreased blood flow to the
iris and ciliary processes but not to the posterior uvea or optic
nerve head.289
As reviewed by Grant,290 epinephrine-induced retinal toxicity
was not recognized until the 1960s. A reduction in visual acuity
can occur with the long-term administration of epinephrine.
Generally, the reduction in visual acuity is reversible within
several months.
Much has been written about the incidence of hypertension
and heart palpitations after the administration of topical
epinephrine. An extensive review is presented by Grant.290
Additional caution is necessary when epinephrine is employed
in combination with a local anesthetic, such as occurs in the
course of otolaryngologic procedures. Additionally, if a patient is
taking b-blockers, the possibility of serious complications
resulting from additional epinephrine results.291 What is
sometimes observed is a hypertensive crisis that is immediately
followed by cardiac slowing and possible cardiac arrest. Despite
the potential for adverse cardiovascular effects, the use of
intraocular epinephrine has become standard practice in
cataract surgery, and no untoward effects on the cardiovascular
system have been noted.292295
ADRENERGIC b-RECEPTOR-BLOCKING
DRUGS
The b-blockers timolol, betaxolol, and levobunolol are widely
used in the treatment of primary open-angle glaucoma. Both
timolol and levobunolol are nonselective b-blockers (e.g., they
bind b1- and b2-receptors with nearly equal affinity). Betaxolol is
somewhat selective for the b1-receptor. More recently, carteolol
has been introduced into the medical treatment of glaucoma.
Carteolol is a nonselective b-blocker that also has some
intrinsic sympathomimetic activity (ISA).
Much has been written about the tendency of b-blockers to
cause cardiovascular and respiratory problems.290,295,296
Theoretically, the selectivity of a b-blocker for b1- or b2-receptors
would make it a better choice for use in patients with asthma
and cardiovascular insufficiency, respectively. However, the
drugs currently in use do not have a selectivity sufficient to
prevent their binding of all b-receptors at therapeutic
concentrations.
In addition to receptor selectivity, several other pharmacologic parameters determine the profile of side effects
associated with a given b-blocker. b-Blockers with some ISA,
such as carteolol, pindolol, acebutolol, and penbutolol, are less
likely to cause cardiovascular insufficiency, bronchospasm, or
adverse changes in serum lipids.297 The degree of lipid solubility
should influence how much drug needs to be given topically to

reach therapeutic levels in the anterior chamber. Also, the
degree of plasma protein binding influences how much free drug
is available to the systemic circulation. b-Blockers also differ in
their activity as membrane-stabilizing (and anesthetic) agents.
All these factors influence the degree of local and systemic
toxicity.
A recent area of investigation has involved the development
of prodrugs of timolol and levobunolol that might allow greater
corneal permeability; therefore, the required topical dose of drug
could be reduced, minimizing possible systemic toxicity.298,299
Timoptic-XE (a formulation of timolol that forms a gel on
contact with the ocular surface) administered once a day was
shown to be equally effective in lowering IOP as the equivalent
concentration of topical timolol administered twice a day. The
safety profile is similar to that of equivalent concentrations of
timolol.300
The ocular administration of b-blockers results in rapid
systemic absorption of the drugs in sufficient quantities to
affect the heart and the lungs.301 Early clinical trials showed
timolol to be without serious systemic side effects. However, as
summarized by Nelson and colleagues,302 many of these early
studies did not include patients with underlying cardiovascular
or respiratory problems. As of 1985, the US Food and Drug
Administration and the National Registry of Drug-Induced
Ocular Side Effects have tabulated a total of 450 case reports of
serious cardiovascular or respiratory complications, 32 of which
resulted in death, after the administration of topical timolol. Of
the 212 patients for which a medical history was provided, 92%
had either cardiovascular or respiratory problems.302 Therefore,
a careful medical history is necessary before prescribing topical
b-blockers for the treatment of glaucoma in order to eliminate
the possibility of exacerbating an underlying condition.
In addition to the contraindications noted later, b-blockers
should not be used in combination with calcium channel
blockers, since sudden death has been reported after the
systemic administration of a b-blocker and verapamil.303,304
b-Blockers cause bronchial constriction as a consequence of
binding to b2-receptors in the bronchi. b-blockers that are
nonselective (such as timolol) may compromise ventilation in
patients with obstructive lung disease, asthma, or bronchospasm.
The National Registry of Drug-Induced Ocular Side Effects has
received over 200 reports of topical timolol-induced respiratory
problems. Sixteen fatal attacks of status asthmaticus have
occurred after the application of topical timolol.305
CHAPTER 33
ADRENERGIC a-RECEPTOR BLOCKING

DRUGS
Clonidine is a relatively selective a2-adrenergic agonist that is
used clinically as an antihypertensive agent. The hypotensive
effect is mediated by the activation of a2-receptors in the central
nervous system.306 Topically, clonidine reduces IOP307311 and
aqueous humor flow.312 It is thought to act by binding
a2receptors in the ciliary body that inhibit adenylate cyclase.313
Apraclonidine is a p-amino derivative of clonidine, which is
incapable of penetrating the bloodbrain barrier. Therefore, the
use of topical apraclonidine should prevent the systemic
hypotension that can occur with the use of topical clonidine.
Apraclonidine is as effective as clonidine in lowering IOP314,315
and has seen use clinically in preventing the large elevations in
IOP that occur after argon laser iridotomy,316,317 argon laser
trabeculoplasty,317,318 and Nd-YAG posterior capsulotomy.317,319
There are indications of possible usefulness in the treatment of
POAG,320,321 particularly when a patient on maximally tolerated
medical therapy is awaiting surgery. Long-term use of
apraclonidine requires frequent monitoring due to the frequent
355

occurrence of tachyphylaxis. Brimonidine, an a2-adrenergic
agonist that is 10-fold more selective than apraclonidine, also
binds to imidazoline receptors. It functions similarly to
apraclonidine, by reducing aqueous inflow and uvealscleral flow.
Brimonidine (0.5%) was developed for post-argon-laseriridotomy and (0.2%) for glaucoma treatment. The advantage of
brimonidine over apraclonidine is that there appears to be a
lower incidence of allergic reaction and tachyphylaxis does not
occur.
SECTION 4
CARBONIC ANHYDRASE INHIBITORS

Inhibition of carbonic anhydrase in the ciliary processes of the
eye reduces aqueous humor secretion, presumably by slowing
the formation of bicarbonate ions with a subsequent reduction
in sodium and fluid transport. The result is a reduction in
IOP.323
Acetazolamide (Diamox) has been used in the treatment of
glaucoma. It has been administered orally on account of the
inability of the compound or other carbonic anhydrase
inhibitors such as methazolamide, ethoxzolamide, and
dichlorphenamide to cross the cornea.324 Even though
systemically administered carbonic anhydrase inhibitors are
effective in lowering IOP, the constellation of side effects
associated with their use has limited the clinical usefulness of
carbonic anhydrase inhibitors in the treatment of glaucoma.
Recently, as summarized by Podos and Serle,325 three
derivatives of acetazolamide that are permeable to the cornea
have been introduced. They are effective in reducing the IOP
with systemic drug levels too low to produce systemic side
effects.326 Most literature concerns the effects of MK-927.327337
There is some evidence that MK-417, the enantiomer of MK927, is slightly more effective in lowering the IOP with
multiple-dose administration to patients with glaucoma.336
Finally, early results with a third derivative, MK-507, suggest
that it may be longer lasting than the other two derivatives.337
Dorzolamide has been developed as a long-awaited carbonic
anhydrase inhibitor that can be administered topically rather
than systemically. It inhibits carbonic anhydrase type II,
reduces the IOP by 21.8% (bid) to 26.2% (tid), and is used alone
or as an adjunctive therapy. Although it is administered
topically, the potential for systemic absorption exists.
Therefore, its use is contraindicated with oral carbonic
anhydrase inhibitors.323
PROSTAGLANDINS
356
Prostaglandins were discovered in the eye in the course of a

search for mediators of ocular inflammation. Prostaglandins D2,
E2, and F2a are synthesized by ocular tissues338 and are actively
transported out of the eye.339 Aside from playing a role in
intraocular inflammation, there is some evidence that
prostaglandins play an endogenous role in normal physiologic
processes.339 Some prostaglandins may actually attenuate an
inflammatory response.340 Prostaglandin F2a causes a dramatic
reduction in IOP in monkey eyes,341347and in normal348,349 and
glaucomatous365 human eyes, which is apparently mediated by
increased nonconventional outflow.342,344346,340 Latanoprost, a
prostaglandin F2a analog that has been introduced for the
treatment of glaucoma, is a prodrug and is metabolized by
corneal esterases. Latanoprost reduces the IOP ~27% when
administered once daily in the morning and, interestingly,
~35% when administered once daily in the evening.
Prostaglandin E2 also apparently reduces the IOP in human
eyes.351 However, the potential for an irritative response is
apparently greater with prostaglandin E1 and prostaglandin E2
than with prostaglandin F2a.347
The major ocular side effects that result from prostaglandin

use relate to their capacity to influence the blood-aqueous and
bloodretinal barriers. Prostaglandin E2 (0.02%) administration
to human eyes is associated with a transient mild eye ache,
photophobia, and conjunctival vasodilatation without clinical
evidence of ciliary flush or anterior chamber cells and flare.351
Intravenously administered prostaglandin E1 resulted in retinal
vasodilation in normal human adults.352 In a single-dose study,
administration of the trimethalamine salt of prostaglandin F2a
in doses ranging from 62.5 to 250 mg resulted in reddened skin
of the lower lid, ocular irritation, conjunctival hyperemia, and
headache without evidence of pupillary changes or anterior
chamber cells or aqueous flare.348
In another study, chronic administration of the more lipidsoluble isopropylester of prostaglandin F2a in doses of 1 mg once
daily or 0.5 mg twice daily for 2 weeks resulted in a significant
reduction in IOP in normal human eyes that was associated
with a dose-dependent hyperemia, foreign body sensation, pain,
and photophobia with no evidence of ocular inflammation.349
No studies in animals or humans have noted a systemic side
effect related to the topical application of prostaglandins.
The ocular side effects associated with latanoprost are a
foreign body sensation, punctate epithelial keratopathy,
stinging, conjunctival hyperemia, blurred vision, itching,
burning, and iris pigmentation. In preclinical studies,
latanoprost was found to increase pigmentation in the iris of
monkeys.353 Additionally, in a 6-month study comparing
latanoprost with timolol in open-angle glaucoma and ocular
hypertensive patients, 10% of patients developed increased iris
pigmentation. All these patients had hazel irises.354 Latanoprost
increases the amount of brown pigment in the iris by increasing
the number of melanosomes within melanocytes, rather than
melanocyte proliferation. The increase in brown pigment does
not progress after discontinuation of treatment, but the
resultant color change may be permanent.353,354
TOXICOLOGY OF AGE-RELATED MACULAR

DEGENERATION DRUGS
The progressive deterioration of central vision in exudative
(wet) age-related macular degeneration (AMD) is caused by
choroidal neovascularization (CNV). Although the majority of
AMD cases are nonexudative, most severe vision loss is attributable to wet disease, and consequently pharmaceutical
development efforts have focused on wet AMD.
Currently approved treatments for CNV associated with wet
AMD include verteprofin, a lipophilic molecule; pegaptanib
sodium, an aptamer; and ranibizumab, a monoclonal antibody.
In addition, bevacizumab, a monoclonal antibody related to
ranibizumab, is sometimes used off-label. Pegaptanib, ranibizumab, and bevacizumab inihibit the vascular endothelial
growth factor (VEGF), an angiogenic factor.
ANTIANGIOGENESIS DRUGS
In the healthy eye, the bloodretinal barrier (BRB) isolates the
eye from systemic circulation. However, neovascular disease
can compromise BRB integrity, making it semi-permeable to
intraocular drugs.355 Excess anti-VEGF in nonocular vascularization could down-regulate healthy angiogenesis systemically.
Cardiovascular and cerebrovascular events have been observed
in short-term studies of anti-VEGFs, but the risk of chronic
exposure, which is theoretically more of a concern for systemic
safety, has not yet been quantified.
VEGF has many physiologic roles which could be adversely
affected by a VEGF inhibitor. Its pivotal role in the angiogenic
cascade as a growth and permeability factor is needed for wound

Pegaptanib, ranibizumab, and bevacizumab all disrupt the
angiogenic cascade at the beginning of the neovascular process
by binding to VEGF, the most important and abundant protein
mediator type. The pharmological differences between the three
treatments explain the drugs different efficacies. Pegaptanib
selectively blocks VEGF165, one subtype of the VEGF-A splice
variant class that also includes VEGF121, VEGF189, and
VEGF206.369 Ranibizumab and bevacinzumab bind to all
VEGF-A isoforms, and therefore have broader inhibitive,
antiangiogenic effects.
Ranibizumab
Serious side effects of ranibizumab attributable to intravitreal
injection include endopthalmitis (1.3%), and intraocular
inflammation (1.7%). Adverse events potentially related to the
drugs systemic toxicity include myocardial infarction (2.1%)
and cerebral vascular events (0.9%).343,371 It is important to
note, however, that the aforementioned clinical data was
gathered in studies of limited duration, and may not reliably
predict the consequences of long-term management of
exudative AMD with ranibizumab.
Pegaptanib
Adverse events associated with pegaptanib sodium include
traumatic lens injury (0.7%), retinal detachment (0.6%),
vitreous floater (33%), endophthalmitis (1.3%) and retinal
detachment (0.6%).372 No extraocular complications were
observed, but since patients at risk of cardiovascular and
cerebrovascular events were excluded from major clinical
studies, the systemic safety of the drug cannot be assumed.
Bevacizumab
Because bevacizumab has not been FDA-approved for
ophthalmic use, there is no official data regarding its intraocular
toxicity. The small case studies that do exist claim intravitreal
bevacizumab has no systemic toxicity,373376 but since they were
not random or controlled, their conclusions can be viewed
skeptically. Also, the side effects associated with ranibizumab
are probably comparable, to a greater or lesser extent, to the side
effects of bevacizumab, due to the molecular similarity of the
two compounds.
CHAPTER 33
healing. VEGF also has vasodilative and neuroprotective

effects,356 and helps maintain vessels such as the coronary
artery.357 Therapeutic down-regulation of VEGF using wet
AMD drugs could theoretically impact any or all of these
processes. Systemic side effects of intraocular anti-VEGFs have
thus far been rare, but since no data exist on prolonged antiVEGF exposure, serious adverse reactions cannot be ruled out.
Our understanding of the adaptation of the angiogenic cascade
in the presence of anti-VEGF agents is incomplete. Because of
these gaps in our knowledge, it is impossible to be sure of the
long-term toxicity of anti-VEGF drugs.
All 3 anti-VEGF drugs used intraocularly to treat AMD
pegaptanib, ranibizumab, and bevacizumab are administered
via intravitreal injection. While the procedure delivers
therapeutic levels of medication to the posterior segment of the
eye, it has been associated a number of vision-affecting complications, including endopthalmitis, retinal detachment, vitreous
floaters, traumatic cataract, and vitreous hemorrhage.358,359
Anti-VEGF drugs are also used to inhibit abnormal
neovascularization associated with tumor growth in cancer
patients. Adverse events associated with intravenous bevacizumab used to treat colon cancer include cerebral infarctions,
myocardial infarctions, other arterial thromboembolic events,
hemorrhage and gastrointestinal perforations.360
Although ophthalmic anti-VEGF therapies sucha as ranibizumab, are similar to oncological drugs, such as bevacizumab,
their systemic side effects are less severe. The doses needed to
inhibit abnormal ocular neovascularization are far lower than
the doses needed to disrupt angiogenesis in malignant tumors.
Also, intravitreal injection limits systemic exposure, even if the
bloodretinal barrier is breached by disease.
Intravenous verteporfin used in conjunction with a
nonthermal red laser, is also used to treat wet CNV. The
combination treatment, known as photodynamic therapy
(PDT), involves a photosensitive compound which, when
activated by a low-power laser dissociates into volatile oxygen
free radicals. The unstable oxygen compounds injure the
neovascular endothelium so that it secretes procoagulant and
vasoactive factors which occlude abnormal vessels in the
macula. PDT can selectively treat diseased tissue while leaving
healthy tissue intact.
Since PDT is a combination drug and device therapy, it is
difficult to attribute specific adverse reactions to a specific stage
of treatment. That said, the following side effects have been
observed in exudative AMD patients treated with PDT:
blepharitis (1.7%), conjuctivitis (6.7%), dry eye (2.7%), ocular
itching (3.5%), retinal capillary nonperfusion (0.2%), retinal
detachment (1.0%), subretinal hemorrhage (2.2%), and vitreous
hemorrhage (1.7%).361 Because the drug is photosensitive, uncontrolled exposure to light could activate the drug, causing it to
occlude normal and abnormal vessels without distinction.362
Some cohort studies recommend high dose vitamin
supplements for nonexudative AMD projected to progress to
exudative AMD.363 Vitamins C, E, beta-carotene, and other
carotenoids365 such as zeaxanthin and lutein, have been
considered for both prophylactic and therapeutic treatment of
AMD.366 The antioxidant (vitamins C, E and beta carotene)
plus zinc formulation used in the Age-Related Eye Disease
Study (AREDS) has been shown to reduce the risk of developing
advanced AMD.381a At daily recommended levels, antioxidants
are safe and essential, but the much higher dosage required to
treat AMD has been associated with side effects. Beta-carotene
has been shown to increase the risk of lung cancer in
smokers,367 and zinc supplementation can lead to systemic
copper deficiency.368 High doses of vitamin E and C have been
associated with an increased risk of heart failure in patients
with vascular disease or diabetes.
OCULAR TOXICITY OF SYSTEMIC

MEDICATIONS
The intermittent or chronic administration of certain oral,
transdermal, and parenteral (including intrathecal and
intracarotid) medications may produce a variety of side effects
in one or more areas in the eye and visual system. It is
important to recognize the ocular effects following the systemic
application of drugs. Although adverse effects are frequently
encountered within the first 2 weeks of therapy, they may be
delayed. The ophthalmologist may not initially relate the ocular
side effects to the systemically applied drugs. This is especially
true if there is a long latent period between drug intake and the
pathologic eye changes or if the toxic effects of the drug are
persistent or even progressive after withdrawal of the drug, as
with phenothiazine.
The toxic effect of certain drugs may be cumulative and dose
dependent. For example, the retinopathy associated with
chloroquine therapy may appear years into therapy. Therefore,
the daily dose and duration need to be monitored. The toxic
effect of other drugs may be idiosyncratic and occur after a
single dose, as in StevensJohnson syndrome (in which a variety
of ingested drugs such as sulfonamides, barbiturates,
salicylates, phenylbutazone, penicillin, phenytoin, and others
have been implicated) or with ibuprofen-induced optic neuritis.
357
SECTION 4

Toxicity may depend on the solubility characteristics of the
drug and its ability to gain access through certain barriers such
as the bloodbrain barrier or the bloodocular barrier. The route
of administration becomes critically important, since such
barriers may be bypassed (as in the case of intrathecal
administration of certain chemotherapeutic agents). Massive
concentrations of a drug may be locally delivered in a fashion
that bypasses the hepatic metabolism to limit systemic toxicity,
but with significant local toxic manifestations (as in the case of
intracarotid administration of nitrosoureas, such as
bischloroethylnitrosourea (BCNU), for the treatment of primary
central nervous system tumors).
The route of drug delivery to the eye and its specific
characteristics influence the type of toxicity. Drugs that gain
access to the eye through tears may manifest ocular surface
abnormality in the form of toxic conjunctivitis or epithelial
keratitis (e.g., certain antimetabolites such as methotrexate).
Many systemic agents, including oral antihistamines, cause
ocular drying that can exacerbate keratoconjunctivitis secca and
cause extreme discomfort and ocular surface irritation. Drug
access into the eye via the aqueous humor may produce
lenticular or posterior corneal changes (e.g., the pigmentary
deposits on the anterior lens capsule and posterior cornea seen
with chronic phenothiazine use or the formation of posterior
subcapsular cataracts due to lens epithelial toxicity from
antimetabolites such as busulfan).
A variety of drugs used for a diverse range of medical
conditions may manifest a similar pattern of toxicity if they
possess similar chemicalphysical properties. A whorl-pattern
epithelial keratopathy may be produced by the group of drugs
that possess cationic amphiphilic properties (e.g., amiodarone
used for cardiac arrhythmias, chloroquine used as an
antimalarial and in collagen vascular disease, indomethacin
used as an analgesic and antiinflammatory, and suramin used
as an antiprotozoal and as a reverse transcriptase inhibitor of
human T-cell lymphotrophic virus III. By binding to polar lipids
and thus accumulating in the lysosomes of epithelial cells,
these agents produce a lysosomal disorder similar to the
lysosomal enzyme disorder seen in Fabrys disease and with a
similar clinical pattern. Drugs that have affinity for particular
chemical components often manifest their toxicity in areas
where high concentrations of these components are present.
Since the uvea has the highest melanin content of any tissue in
the body, it is not surprising that drugs with a high affinity for
melanin (e.g., chloroquine and hydroxychloroquine) induce
retinal-uveal toxicity.
Not all ocular changes due to systemic drugs require
discontinuation of the drug, since some may be inconsequential
and ultimately reversible, as with the whorl-like corneal
epithelial deposits seen with the cardiac antiarrhythmic
amiodarone or the ocular hypotensive effects of orally
administered b-blockers used for the therapy of hypertension or
angina pectoris. However, other adverse ocular reactions may be
irreversible, as with ethambutol-associated optic neuropathy.
Thus, it is critical to be aware of the nature of the toxicity and
the prognosis in order to plan the appropriate strategies for
patient monitoring and management.
Side effects caused by one member of a given chemical family
are often, but not always, caused by other members of the same
drug group. Therefore, knowledge of side effects of one drug
should alert one to monitor for side effects when drugs from a
similar family are used. Experimental trials of new agents must
include monitoring for the side effects anticipated by the
chemical family.
The ophthalmologist must be familiar with the appropriate
visual tests and monitoring requirements appropriate for
particular drug regimens. It is critical to
1. Identify the toxic agent and know its chemical family
2. Know if the effects are reversible or irreversible in order to
determine the appropriate plan of action
3. Be aware if a drug toxicity is cumulative or dose dependent
and monitor the specific parameters carefully
4. Be familiar with the appropriate diagnostic tests
Comprehensive reviews of ocular toxicity exist.377381 Table 33.1
summarizes important side effects with practical information
for recognizing and managing toxicity: The information has
been limited to effects seen in humans and has been divided
into clinically useful categories of anterior segment, posterior
segment, and clinically relevant systemic toxicities.
ACKNOWLEDGMENTS
The preparation of this manuscript was supported in part by grants from
the National Eye Institute (EYO 7321), Research to Prevent Blindness,
and the Massachusetts Lions Eye Research Fund.
TABLE 33.1. Ocular Toxicity and Side Effects of Systemic Medications

Drug
Class
Uses
Route
Side Effects
Dose
Relationship
Comments
Methyl ether
Inhalation
anesthetic
Inhalation
Crystalline retinopathy Prolonged

anesthesia
Calcium oxalate
crystals in retinal
pigment epithelium
and retina
Atenolol
(including
labetalol,
metoprolol,
nadolol, and
pindolol)
b-Adrenergicblocking
agent
Antianginal and
Oral, IV
antihypertensive
Sicca syndrome,
Yes
visual hallucinations,
myasthenic
neuromuscularblocking effect
(may worsened
myasthenia gravis)
Work-up myasthenia
if patient exhibits
extraocular muscle
paresis
Diltiazem (also
nifedipine,
verapamil)
Calcium
channel
blocker
Antianginal
Anesthetic (inhalation)
Methoxyflurane
Antianginal
358
Oral,
Rare; ocular irritation
sublingual,
with periorbital
IV
edema and blurred
Yes
Reversible
Continued
TABLE 33.1. Ocular Toxicity and Side Effects of Systemic Medicationscontd

Drug
Class
Uses
Route
Side Effects
Dose
Relationship
Comments
Duration related
Side effects reversible

in early stages, but
decreased tear
production may
persist. This drug
for general use has
been withdrawn
from the market.
Mechanism of toxicity
may be related to
production of antibodies to practolol
metabolite.
Oculomucocutaneous
findings not seen
with other
b-adrenergicblocking agents
Practolol
agent
Antianginal and
Oral, IV
antihypertensive
Keratoconjunctivitis
sicca, conjunctival
cicatrisation,
keratitis with
opacities, and
myasthenic
neuromuscularblocking effect
Propranolol
agent
Antianginal and
Oral, IV
antihypertensive
Sicca syndrome,
Yes
myasthenic neuromuscular-blockingeffect, visual
hallucinations, and
?inflammatory
orbital pseudotumor
Reversible
Antianxiety
Alprazolam
(including
clonazepam,
flurazepam,
triazolam)
Benzodiazepine Antianxiety
Oral, IV, IM
Decreased
corneal reflex,
accommodation,
and depth
perception,
abnormal
extraocular muscle
movement, allergic
conjunctivitis,
?mydriasis
precipitating
narrow-angle
glaucoma
No
Benzofuran
derivative
Oral, IV
Whorl-like (vortex
pattern) epithelial
keratopathy (98%),
resulting in photophobia (3%), halos
(2%), and blurred
vision (1%), sicca
syndrome, lens
opacities, skin
pigmentation,
papillopathy and
optic neuropathy,
pseudotumor
cerebri,
?retinopathy
(hypopigmentation)
Dosage and
Keratopathy due
duration
to cationic
related
amphophilic
(keratopathy) with
properties of drug
minimal corneal
that binds to polar
deposits with
lipids and produces
dosages <200
a lysosomal
mg/day but in
disorder, as in
nearly all patients
Fabrys disease.
with >400 mg/
Onset of
day; unclear for
keratopathy as
papillopathy
early as 6 days but
usually by 6 weeks;
usually resolves in
3 months after
discontinuation but
may have
prolonged effect
owing to long halflife. Keratopathy not
indication to
discontinue drug,
but papillopathy is a
relative indication
CHAPTER 33
vision, retinal
ischemia, and
transient blindness
Antiarrhythmic
Amiodarone
Antiarrhythmic
(ventricular)
Continued
359

Drug
Class
Uses
Route
Side Effects
Dose
Relationship
Comments
Digitalis
Digitalis
glycoside
Antiarrhythmic and Oral, IV

for congestive
heart failure
1125% side effects

with toxic doses.
Color vision
abnormalities
(yellow-blue), visual
sensations and
hallucinations,
scotomas, retinal
toxicity with
abnormal ERG
amplitude
Yes
Reversible. Toxicity
may be made
worse with
concomitant
quinidine therapy
(ERG may be
helpful). Color
testing (yellow-blue)
may be helpful in
adjusting dosage
Disopyramide
Anticholinergic Antiarrhythmic
Oral, IV
Blurry vision,
decreased accommodation and
lacrimation;
mydriasis may
precipitate narrowangle glaucoma
Yes
Side effects due to

anticholinergic
effects, which are
reversible
Procainamide
Procaine
hydrochloride
analog
Antiarrhythmic
Oral, IV
Rare; lupus-like
syndrome with
scleritis
No
Hydantoin
Anticonvulsant
Oral, IV, IM
Nystagmus, lens
opacities, benign
intracranial
hypertension,
ocular teratogenic
effects
Yes
Amitriptyline
(including
desipramine,
imipramine,
nortriptyline)
Tricyclic
antidepressant
Antidepressant
Oral
Mydriasis and
Yes
cycloplegia (may
precipitate narrowangle glaucoma),
aggravate keratoconjunctivitis sicca
owing to anticholinergic effects,
oculomotor
abnormalities
Reversible
Carbamazepine
Iminostilbene
derivative
Antidepressant,
pain associated
with trigeminal
neuralgia
Oral
Blurred vision,
Yes; side effects
extraocular muscle
with dosages
abnormalities with
>1.2 g
diplopia, downbeat
nystagmus,
sluggish pupil and
papilledema with
toxic doses, retinal
pigmentary changes
Reversible with
decrease in dosage
Doxepin
(including
amoxapine,
clomipramine)
Tricyclic
antidepressant
Antidepressant
(also for
psychoneurotic
anxiety)
Oral, IV
Blurred vision,
mydriasis,
accommodation
disturbances, and
aggravation of
keratoconjunctivitis
sicca due to anticholinergic effects.
Nystagmus and
ophthalmoplegia
with toxic states
Yes
Reversible
Methylphenidate
Piperidine
derivative
Antidepressant and Oral, IV (see Rare; mydriasis.

for hyperkinetic
Comments)
Talc retinopathy
syndrome in
(see Comments)
children
Yes, usually with

overdose
Illicit IV use of crushed

tablets is responsible
for talc and
cornstarch (used as
fillers) retinopathy
Anticonvulsant
SECTION 4
Phenytoin
Nystagmus may
persist for months
after
discontinuation.
Fine nystagmus at
therapeutic doses;
coarse nystagmus
in toxic states
Antidepressant
360
Continued

Drug
Class
Uses
Route
Side Effects
Dose
Relationship
Comments
Phenelzine
Monoamine
oxidase
inhibitor
Antidepressant
Oral
Rare; mydriasis,
miosis, anisocoria,
nystagmus,
diplopia, and
myasthenic
neuromuscular
blockade
Yes, usually with

overdose
MAO inhibitor activity

increased with
concomitant use of
other MAO
inhibitors and
tricyclic
antidepressants
Brompheniramine
(also chlorpheniramine,
dexbrompheniramine,
dimethindene,
triprolidine)
Alkylamine
See
Oral
Cyproheptadine
See Pyrilamine.
Facial dyskinesia
with chronic use
See Pyrilamine
Alkylamine has the

lowest incidence of
ocular side effects
Cyproheptadine
(Periactin) (also
azatadine)
Phenothiazine
analog
Antihistamine used Oral

in allergic or
vasomotor
rhinitis, allergic
conjunctivitis
Rare. Atropine-like
effects causing
mydriasis and
decreased
secretions
aggravating
sicca
Side effects usually

disappear even with
continued use.
May precipitate
narrow-angle
glaucoma
Diphenyhydramine Ethanolamine
(Benadryl)
See
Oral
Cyproheptadine
See Pyrilamine
Toxic doses
responsible for
visual hallucinations
and nystagmus
Pyrilamine (also
tripelennamine)
Ethylenediamine
See
Oral
Cyproheptadine
See Cyproheptadine. Visual hallucinations See Cyproheptadine

With long-term use,
with overdose
anisocoria,
decreased
accommodation,
and blurred vision.
Facial dyskinesia
(blepharospasm),
visual hallucinations
Vitamin
Antihyperlipidemic Oral
Metamorphosia,
blurring, central or
paracentral
scotoma,
maculopathy,
atypical CME with
no accumulation
of fluorescein on
angiogram
Clonidine
a-Adrenergic
agonist
Antihypertensive
Oral
Miosis and mydriasis Yes

(toxic doses),
?retinal
abnormalities
(depigmentation,
degeneration, tears)
Reversible; unclear if
retinal findings
coincidental or drug
related
Hydralazine
Phthalazine
derivative
Antihypertensive
Oral, IV
Nonspecific ocular
irritation, lupus-like
syndrome with
episcleritis, retinal
vasculitis, and
exophthalmos
Reversible
Oral
Rare. With rechallenge Optic neuritis and

refractive error
toxic amblyopia
changes, diplopia,
are idiosyncratic
photophobia, dry
eyes, decrease in
CHAPTER 33
Antihistamine
Antihyperlipidemic
Niacin (nicotinic
acid)
Yes, >1.5/day
Symptoms precede
findings. Amsler
grid may
demonstrate central
visual change.
Reversible
Antihypertensive
Transient
Antiinflammatory
Ibuprofen
(Motrin, Advil)
Nonsteroidal
Antiinflammatory,
antianalgesic,
inflammatory
antipyretic
drug that
Osteoarthritis,
inhibits
rheumatoid
Optic neuritis and

toxic amblyopia are
reversible with
visual acuity
returning to normal
Continued
361

Drug
Class
SECTION 4
cyclooxygenase
(propionic
acid)
Uses
Route
arthritis, gout,
ankylosing
spondylitis,
cystoids
macular edema,
?ocular
inflammation
Side Effects
Dose
Relationship
color vision, optic

neuritis with central
scotomas, toxic
amblyopia
Comments
in 13 months, but
color vision not
returning for up to
8 months. May be
irreversible is drug
is not discontinued
Indomethacin
(Indocin)
See Ibuprofen
(indole)
See Ibuprofen
Oral
Decreased vision,
?
color vision defects,
hypersensitivity
reactions, including
Stevens-Johnson
syndrome, corneal
deposits, including
whorl-like epithelial
deposits,
papilledema
secondary to orbital
pseudotumor
Corneal deposits is
not a indication to
discontinue the
drug
Ketoprofen
(Orudis)
See Ibuprofen
(propionic
acid)
See Ibuprofen
Oral
Nonspecific
conjunctivitis and
dermatologic
reactions,
cholinergic crisis,
and papilledema
secondary to
orbital pseudotumor
In general,
nonsteroidal
antiinflammatories
are photosensitizers
Naproxen
(Naprosyn)
See Ibuprofen
(propionic
acid)
See Ibuprofen
Oral
Whorl-like corneal
opacities, optic
neuritis
Optic neuritis is
idiosyncratic
This drug is a
photosensitizer;
?role in
maculopathy or
necrotizing
vasculitis
Piroxicam (Felden)
See Ibuprofen See Ibuprofen

(oxicam and
enolic acid)
Oral
Rare and insignificant
Idiosyncratic
Most widely
prescribed
nonsteroidal antiinflammatory
worldwide
Prednisone
Corticosteroids Antiinflammatory
Adrenocortical
insufficiency
replacement
Oral
Cataracts (PSC),
Cataracts usually
Exophthalmos may
ocular hypertension
dose related,
not completely
and glaucoma,
increased risk of
reverse. Increase,
pseudotumor cerebri
pressure elevation
then slowly taper
and papilledema
with ocular
dose in
with withdrawal,
hypertension,
pseudotumor
exophthalmos with
glaucoma, or
cerebri. Pressure
long-term use,
family history of
may rarely remain
decreased tear
glaucoma and
elevated after
lysozyme,
diabetes
discontinuation.
?decreased
Cataracts may
resistance to
rarely progress after
infection,
discontinuation;
myasthenic
may be reversible
neuromuscularin children
blocking effect
(extraocular muscle
paresis, ptosis),
delayed wound
healing
Sulindac (Clinoril)
See Ibuprofen
(indene)
See Ibuprofen
Oral
Rare and insignificant
Quinolone
Antimalarial and
antirheumatic
Rheumatoid
arthritis, lupus
erythematosus
Oral
Whorl-like corneal
Yes (cumulative
Toxicity greater with
epithelial deposits,
dose); little toxicity
chloroquine than
Hudson-Stahli line,
if 3.5 mg/kg/day,
with
accommodation,
<250 mg/day
hydroxychloroquine.
motility, subcapsular
for small patients,
Corneal changes
cataracts, central
<100 g total
reversible. Rental
Idiosyncratic
Antimalarial
Chloroquine (see
also hydroxychloroquine)
362
Continued

Class
Uses
Route
Side Effects
and paracentral
scotoma,
photophobia,
nyctalopia,
photopsia, macular
pigmentation, loss
of macular reflex,
macular edema,
bulls-eye
maculopathy, bone
spicule formation,
optic disc pallor,
vascular attenuation
(end stage), ERG and
EOG abnormalities
Dapsone
Sulfone
Antimalarial and
Oral
anti-inflammatory
Rare. Optic atrophy
Dose
Relationship
< year
Dose related
Comments
changes may be
irreversible or
progressive after
discontinuation.
Since early changes
are nonspecific
and patients with
toxicity may be
asymptomatic,
routine testing is
required. Every 6
months: vision,
history, Amsler grid,
central visual field
with red target,
color testing, ?ERG,
?EOG
With massive doses
Hydroxychloroquine Quinolone
(see also
chloroquine)
See Chloroquine
See
See Chloroquine.
Chloroquine
Safe <6.5 mg/kg/
day or 400 mg/day
for smaller patients
See Chloroquine
Quinine
Alkaloid
Antimalarial
Nocturnal leg
cramps,
myotonia
congenital,
myokymia,
attempted
abortions
Oral
Toxic amblyopia,
Dose related
sudden vision loss,
(massive);
retinal arterial
occasionally with
constriction, venous
low chronic
congestion, retinal
administration
edema, macular
pigmentary changes,
disc edema, optic
nerve hypoplasia,
myasthenic
neuromuscular
blockade
(extraocular muscle
paralysis, ptosis)
Analog of
ubiquinone
Antiparasitic and
Pneumocystis
carinii in AIDS
Oral
Vortex keratopathy
Cefazolin (including Cephalosporin

first, second,
and third
generations)
Antibacterial
Oral, IV, IM
Rare. Allergic reactions, No

including StevensJohnson syndrome,
?retinopathy
(cephaloridine)
Chloramphenicol
Dichloracetic
acid
derivative
Antibacterial
Oral, IV
Rare. Decreased vision, Dose related; total

optic neuritis, optic
>100 g or
atrophy, toxic
duration > 6
amblyopia,
weeks
retinopathy
Findings most often in

children. Most
feared side effect is
aplastic anemia,
which is
idiosyncratic
Ciprofloxacin
Fluoroquinolone Antibacterial
Oral, IM, IV
Rare. Blurred vision,

Dose related;
photophobia, altered
?duration related
color vision, toxic
optic neuropathy
Quinolone group
common to quinine
and chloroquine
may be responsible
for optic nerve
toxicity. Optic
neuropathy is
slowly reversible
Clofazimine
Phenazine
derivative
Oral
Eyelid and conjunctival Yes

pigmentation,
corneal epithelial
changes
Side effects are

reversible
Use on rise, especially

in street drugs.
Vision loss may be
acute or
progressive with
usually some return
of vision. Prenatal
maternal ingestion
may cause optic
nerve hypoplasia.
Acute therapy
unclear
CHAPTER 33
Drug
Antimicrobial
Atovaquone
Antibacterial used
for leprosy
Continued
363

Drug
Class
Doxycycline (also
tetracycline,
minocycline)
Polycyclic
Antibacterial
naphthacene
carboxamide
Tuberculostatic
Route
Side Effects
Dose
Relationship
Oral
Eyelid skin conjunctival Orbital pseudotumor Most side effects are

hyperpigmentation,
not dose related;
reversible. Orbital
hyperpigmented
pigmentation
pseudotumor
conjunctival cysts,
dose related
mostly seen with
blue-gray sclera
tetracycline and
pigmentation
minocycline ?due
(minocycline), orbital
to greater lipid
pseudotumor,
solubility. Sclera
extraocular muscle
pigmentation
paralysis,
(minocycline)
aggravation of
frequently
myasthenia gravis
associated with
pigmentary
changes elsewhere
Oral
Color vision
abnormalities, visual
field changes
(scotomas), axial
and paraxial optic
neuritis
Dose related.
Infrequent with
doses 15 mg/
kg/day
SECTION 4
Ethambutol
Uses
364
Comments
Optic neuritis
symptoms usually
noted at 36
months. Increased
toxicity with renal
disease. With
regular doses,
home visual acuity
and color vision
testing
recommended. With
higher doses,
screen patient at
2- to 4-week
intervals. Visualevoked response
helpful in detecting
subclincal toxic
effects. Visual
recovery variable.
?Treat optic nerve
toxicity with zinc
sulphate or
parenteral
hydroxycobalamin
Gentamicin
(including
tobramycin,
streptomycin)
Aminoglycoside Antibacterial
IV, IM,
intrathecal
Papilledema secondary No
to pseudotumor
cerebri, myasthenic
neuromuscular
blockade (paralysis
of extraocular
blockade and
ptosis), blindness
and optic atrophy
with intrathecal
administration
Most side effects are

reversible after
discontinuation
Isoniazid
Hydrazide of
isonicotinic
acid
Antitubercular
Oral
Rare. Optic and

No
retrobulbar neuritis
with visual field and
color vision
abnormalities

seen in
malnourished or
chronically ill
patients. Many side
effects can be
prevented by daily
administration of
pyridoxine
Nalidixic acid
Naphthyridine
Antibacterial
Oral
Visual disturbances,
Most not dose
color vision defects,
related
papilledema due to
increased
intracranial pressure,
lupoid skin changes

if dosage is
decreased or drug
discontinued.
Increased
intracranial pressure
reported in persons
younger than
age 20
Continued

Class
Uses
Route
Side Effects
Penicillin (including Penicillin

semisynthetic
penicillins)
Antibacterial
Oral, IM, IV
Rare. Allergic reactions,

including StevensJohnson syndrome.
Aggravation of
ocular signs of
myasthenia gravis
(ampicillin), including
paralysis of extraocular muscles,
diplopia, and ptosis,
pseudotumor cerebri
Rifabutin
Synthetic
rifamycin
Antitubercular and Oral

prophylaxis
against
Mycobacterium
avium complex
in AIDS
Rifampin
Hydrazone
Antitubercular and Oral
derivative of
antibacterial
rifamycin B
Sulfamethoxazole Sulfonamide
(including other
sulfa-containing
medications
such as
sulfadiazine and
sulfasalazine)
Antibacterial,
?antiinflammatory
Suramin
Antiprotozoan
Oral
used for
adjuvant therapy
in AIDS patients
(inhibitor of
reverse
transcriptase
of HTLV III)
Nonmetallic
polyanion
Oral
Dose
Relationship
Comments
Anterior uveitis,
hypopyon uveitis,
white-yellow
opacities
vitreous
May be dose related Occurs with

with increased
concomitant use of
incidence with
rifabutin with
600 mg/day, less
clarithromycin and
common with
fluconazole.
300 mg/day
Immunologically
mediated process
rather than direct
drug toxicity;
resolves with
topical
corticosteroid
therapy frequently
without
discontinuation of
rifabutin
Conjunctival
hyperaemia,
conjunctivitis (may
be exudative),
orange staining of
contact lenses
Yes
Reversible ocular side

effects in 515% of
patients and more
frequently seen with
intermittent use
Myopia due to lens

No
thickening from
ciliary body edema,
allergic reactions,
including StevensJohnson syndrome,
anterior uveitis,
optic neuritis
Side effects rare and

reversible
Vortex-like epithelial
keratopathy, ocular
irritation, optic
atrophy
Dose related

depend on
nutritional status.
Optic atrophy
secondary to
inflammatory
response to dead
microfilariae.1
Keratopathy due to
lysosomotropic
properties that
inhibit lysosomal
enzymes.
Reversibility of
keratopathy unclear
at present owing to
prolonged half life
sicca, posterior
subcapsular
cataract with
polychromatic
sheen 1030%
Yes, 26 mg/day for

months to years
CHAPTER 33
Drug
Antineoplastic or Immunosuppressive
Busulfan
Alkylating
agent
Cancer
chemotherapy:
chronic
leukemia,
polycythemia
vera,
myelofibrosis
Continued
365
SECTION 4

Drug
Class
Uses
Route
Chlorambucil
Alkylating
agent
Cancer
Oral
chemotherapy:
chronic leukemia
Immunosuppression:
vasculitis with
RA, Behets
disease,
autoimmune
hemolytic
anemia
cis-Platinum
(cisplatin)
Alkylating
agent
Cancer
IV,
Blurred vision (62%),
Yes, >600 mg/m2
chemotherapy:
intracarotid
impaired color
testicular cancer,
vision (23%), retinal
breast cancer,
toxicity (ERG) (84%),
bladder cancer,
macular
lung cancer,
pigmentation (46%),
gastrointestinal
disc edema,
cancer,
retrobulbar neuritis,
lymphoma,
cortical blindness.
osteogenic
With intracarotid
sarcoma
administration,
ipsilateral vision
loss due to retinal
and optic nerve
ischemia (1560%)
Dose
Relationship
Rare but includes

keratitis,
hemorrhagic
retinopathy, and
oculomotor
disturbances
Yes
Cyclophosphamide Alkylating
agent
Cancer
Oral, IM, IV
chemotherapy:
lymphoma,
breast cancer
Immunosuppressive:
rheumatoid
arthritis,
Wegeners
granulomatosis,
Moorens ulcer,
cicatricial
pemphigoid,
Behets
disease, Graves
disease
ophthalmopathy
Blurred vision (17%),

Yes
sicca (50%), pinpoint
pupil due to
parasympathomimetic effect
Cytosine
arabinoside
Cancer
chemotherapy:
acute leukemia,
refractory
lymphoma
Keratoconjunctivitis,
Yes
central punctate
opacities with
subepithelial granular
deposits, microcysts,
reversible superficial
punctate keratitis
(38100%)
Optic neuropathy (may
be potentiated by
cranial irradiation)
Pyrimidine
analog
IV
Intrathecal
366
Side Effects
Comments
Blurred and color

vision abnormalities
are reversible
Resolution of
symptoms in
weeks,
prednisolone
phosphate or 2deoxycytidine
prophylaxis
Doxorubicin
(Adriamycin)
Antimicrobial
Cancer
anthracycline
chemotherapy:
that binds
sarcoma,
DNA
leukemia,
lymphoma
Lacrimation (25%), red

discoloration
of tears
Fludarabine
Purine analog
Cancer
chemotherapy:
leukemia
Decreased vision due

to optic neuritis or
cortical blindness,
encephalopathy
Yes
5-Fluorouracil
Pyrimidine
analog
Cancer
chemotherapy:
breast cancer,
GI cancer,
GU cancer
Blurred vision, ocular

pain, photophobia,
lacrimation,
conjunctivitis,
blepharitis, keratitis
Most are reversible Massage, topical

614 months for
corticointubation
cicatricial changes
IV
Continued

Drug
Class
Uses
Actinic keratosis
Route
Topical
Side Effects
Dose
Relationship
Comments
Yes (IV)
Resolves off therapy;

artificial tears
(25=n38%),
cicatricial ectropion,
punctal and
canalicular stenosis,
blepharospasm,
oculomotor
disturbance,
nystagmus, optic
neuropathy
Systemic absorption
may cause similar
corneal and external
disease findings
Folic acid
analog
Cancer
Oral, IM, IV
Periorbital edema,
chemotherapy:
photophobia, ocular
leukemia, solid
pain and burning,
tumors
blepharitis,
Immunosupconjunctivitis, and
pressive:
decreased tear
rheumatoid
Intrathecal,
production (25%),
arthritis,
intracarotid
optic neuropathy,
psoriasis, uveitis
macular edema and
pigment epithelial
changes
Mitomycin C
Antimicrobial
that cross
links DNA
Cancer
chemotherapy:
solid tumors
Blurred vision
Mitotane
Antimicrobial
DDT
derivative
Cancer
chemotherapy:
adrenocortical
cancer
Neuroretinopathy, disc
edema, retinal
hemorrhages, retinal
edema, cataracts
(316%)
Nitrogen mustard
Alkylating
agent
Cancer
chemotherapy:
lymphoma,
brain tumor
Nitrosoureas
(BCNU, CCNU,
methyl CCNU)
Alkylating
agent
Cancer
Oral, IV,
Usually benign.
Yes (dose and
chemotherapy:
intracarotid
Conjunctival
rapidity of
primary CNS
hyperemia and
infusion) with
tumor,
blurred vision (4%),
intracarotid
lymphoma,
?optic neuritis,
administration
multiple
ipsilateral periorbital
myeloma, colon
edema, orbital pain
and gastric
and congestion,
cancer
conjunctivitis,
chemosis,
neuroretinal toxicity
(70%) (NFL infarcts,
intraretinal
hemorrhages, and
disc edema, with
intracarotid
administration)
Plicamycin
(mithramycin)
Antimicrobial
Cancer
Inhibits RNA
chemotherapy:
synthesis by
testicular cancer,
binding DNA
hypercalcemia
Tamoxifen
Antihormonal
estrogen
antagonist
Cancer
chemotherapy:
breast cancer
IV,
intracarotid
Necrotizing uveitis
and vasculitis
(intracarotid)
CHAPTER 33
Methotrexate
Most likely
ERG; pressure on eye

during infusion or
Honans balloon to
limit toxicity
Periorbital pallor
Oral
Whorl like epithelial

keratopathy,
maculopathy with
superficial white
refractile opacities
associated with
cystoid macular
edema, optic disc
edema, posterior
subcapsular
cataracts
Yes (120200 mg/m2 May be irreversible

for >1 year;
Toxicity unlikely
cumulative dose
with standard
of 90230 g)
doses242 but has
been reported.243,244
Presence of a
few intraretinal
crystals in absence
of macular edema
or vision loss or
presence of
Continued
367

Drug
Class
Uses
Route
Side Effects
Dose
Relationship
Comments
posterior
subcapsular
opacities does not
warrant
discontinuation of
drug
Vincristine
Yes
Vinca alkaloid
Cancer
chemotherapy:
leukemia,
lymphoma,
solid tumors
IV
Cranial nerve palsy

Yes
(50%), internuclear
ophthalmoplegia,
corneal hypesthesia,
optic neuropathy
demyelination,
night blindness,
and cortical
blindness
Increased toxicity with

hepatic dysfunction.
Resolves in
3 months
?Irreversible
Reversible in
114 days
Chlorpromazine
Phenothiazine
Antipsychotic
Oral, IM, IV
Similar to thioridazine.
Pigmentation of
skin, conjunctiva,
and cornea,
pigmentary
retinopathy (fine)
Pigmentary changes
may be reversible
Haloperidol
Buterophenone Antipsychotic
derivative
Oral, IM
Decrease or paralysis
of accommodation,
mydriasis that may
precipitate narrow
angle glaucoma1
and ?cataracts
Yes
Transient and
reversible side
effects
Lithium carbonate
Lithium salt
Antipsychotic
Oral
Ocular irritation and

photophobia,
blurred vision,
extraocular muscle
abnormalities,
exophthalmos,
papilledema due
to pseudotumor
cerebri
Yes
Reversible; toxic drug

response related
to blood levels
(>2 mEq/L);
exophthalmos may
be seen at normal
levels owing to
effect on thyroid
Thioridazine
Phenothiazine
Antipsychotic
Oral, IM, IV
Decreased vision,
Dose and duration
Symptoms improve
paralysis of
related.
after
accommodation,
Rare, <1000 mg/day
discontinuation, but
mydriasis due to
Recommended dose
fundus changes
anticholinergic
<300 mg/day;
may progress
properties, corneal
maximum
pigment deposits
800 mg/day
(epithelium and
Descemets
membrane), corneal
edema, lens
surface deposits.
Granularity of
posterior pole,
transient disc and
retinal edema,
nummular retinopathy, paracentral
and ring scotoma,
abnormal ERG and
EOG, myasthenic
neuromuscular
blockade, extraocular muscle
paralysis, diplopia,
ptosis
SECTION 4
Antipsychotic
Antiparkinsonism
Amantadine
368
Tricyclic amine Parkinsons disease Oral

Antiviral used in
prophylaxis of
Rare. Transient
Dose related
decreased vision,
superficial punctate

with discontinuation
Continued

Drug
Class
Uses
Route
influenza A
Side Effects
Dose
Relationship
Comments
keratitis, sudden
vision loss, visual
hallucinations
Benztropine
(also biperiden,
chlorphenoxamine)
Anticholinergic Parkinsons disease Oral

Control of
extrapyramidal
disorders
Decreased
accommodation;
rarely mydriasis
may precipitate
narrow angle
glaucoma,
hallucinations
Dose related
Ocular side effects

more common with
benztropine versus
biperiden
Levodopa
b Adrenergicblocking
agent
Mydriasis may
Dose related
precipitate narrow
angle glaucoma,
miosis, ptosis,
blepharospasm,
visual hallucinations,
oculogyric crisis
Parkinsons disease Oral
Allopurinol
Xanthine
oxidase
inhibitor
Chronic
hyperuricemia,
gout
Oral
?Cataract, ?macular
Unclear with
edema and
cataracts and
hemorrhage, toxic
maculopathy,
epidermal necrolysis
not dose related
(Lyells syndrome)
toxic epidermal
necrolysis
Gold
Heavy metal
Rheumatoid
arthritis, lupus
erythematosus
IM
Conjunctival and
corneal deposition,
occasionally lens
deposition, rarely
ptosis, diplopia,
nystagmus
Yes, >1 g, 1 g/day

for years for
lenticular
deposition
Cornea and lens

deposits do not
affect visual acuity
and are not an
indication for
discontinuing
therapy
Deposits reversible
after
discontinuation
Anticholinergic Antispasmodic
Oral
Rare. Decreased
vision, mydriasis
(rarely may
precipitate narrow
angle glaucoma),
decreased
accommodation,
and photophobia
Yes
Due to mild
antichoinergic
activity. Side effects
reversible and not
indication to
discontinue drug
Carotenoid
(non
provitamin
A)
Tanning agent for

vitiligo,
photosensitive
dermatitis
Oral
Metamorphopsia,
decreased vision,
yellow, refractile
inner retinal
deposits
surrounding fovea
Yes; total 3040 g,

>50%
retinopathy;
total >60 g,
>55100%
retinopathy
Increased retinopathy
with ingestion of
other carotenoids
Keratinolytic
Topical
Yes
Symptoms rarely last

for weeks after
discontinuation
Vitiliginous lesions
Oral, topical
?Cataracts
CHAPTER 33
Antirheumatic (see also Antiinflammatory and Antineoplastic or Immunosuppressive)
Antispasmodic
Dicyclomine
Dermatologic
Canthaxanthine
Chrysarobin
Methoxsalen (also
trioxsalen)
Psoralen
Used in conjunction
with ultraviolet light
for photochemotherapy (PUVA).
Patient requires
adequate UV
blocking goggles
after therapy
Continued
369

Drug
Class
Uses
Route
Side Effects
Dose
Relationship
Comments
Oral,
inhalation
Decreased vision,
nystagmus,
mydriasis, disc and
retinal edema,
central and
cecocentral
scotoma, optic
atrophy and
excavation
Variable, as low as
1 oz
Primary site of injury is

the optic nerve.
Emergency medical
therapy (respiratory
support, dialysis,
ethanol) is required.
Vision may improve,
usually in 6 days
Immunosuppressive (see Antineoplastic or Immunosuppressive)

Industrial
Methanol
Alcohol
(rubbing,
wood)
Industry
Stimulant (Gastrointestinal and Urinary Tracts)

Bethanechol
Quaternary
ammonium
parasympathomimetic
Gastrointestinal
and urinary
tract stimulant
Oral,
subcutaneous
Rare. Occular irritation

with lacrimation,
decreased
accommodation,
and miosis
Deferoxamine
mesylate
Chelating
agent
Removal of
IV, subexcess systemic
cutaneous
iron
Cataracts, visual loss,

optic neuropathy,
retinal pigmentary
degeneration
Pamidronate
Biphosphonate Inhibitor of bone

IV
resorption used
in hypercalcemia
of malignancy,
painful bone
metastases, and
Pagets disease
Side effects may

continue long after
the drug is
discontinued
SECTION 4
Miscellaneous
Mild to severe anterior

uveitis and
nonspecific
conjunctivitis
Duration related
Toxicity may be rapid

in onset and
irreversible.
Retinopathy
reported with single
subcutaneous dose
Anterior uveitis
frequently bilateral
and may require
topical therapy
Abbreviations: AIDS, acquired immunodeficiency syndrome; BCNU, carmustine; CCNU, lomustine; CME, cystoid macular edema; CNS, central nervous system; DDT,
chlorophenothane; EOG, electrooculogram; ERG, electroretinogram; GI, gastrointestinal; GU, genitourinary; HTLV, human T cell lymphotrophic virus; MAO, monoamine
oxidase; NFL, nerve fiber layer; PSC, posterior subcapsular cataract; PUVA, psoralen ultraviolet light application; RA, rheumatoid arthritis; UV, ultraviolet.
REFERENCES
370
1. Leopold IH, Gaster RN: Ocular

inflammation and anti-inflammatory drugs.
In: Kaufman HE, Barron BA, McDonald MB,
Churchill Livingstone; 1988:67.
2. Richardson KT: Pharmacology and
pathophysiology of inflammation. Arch
3. Havener WH: Anti-inflammatory agents.
In: Havener WH, ed. Ocular pharmacology.
4. Burch PG, Migeon CJ: Systemic absorption
of topical steroids. Arch Ophthalmol 1968;
79:174.
5. Krupin T, Mandell AI, Podos SM, Becker B:
Topical corticosteroid therapy and pituitaryadrenal function. Arch Ophthalmol 1976;
94:919.
6. ODay DM, McKenna TJ, Eliott JH: Ocular
corticosteroid therapy: systemic hormonal
effects. Trans Am Acad Ophthalmol
Otolaryngol 1975; 79:71.
7. McLean JM: Clinical and experimental
observation on the use of ACTH and
cortisone in ocular inflammatory disease.
Trans Am Ophthalmol Soc 1950;
48:259.
8. Francois J: Cortisone et tension oculaire.
Ann Ocul 1954; 187:805.
9. Schwartz B: The response of ocular

pressure to corticosteroids. Int Ophthalmol
Clin 1966; 6:929.
10. Armaly MF: Factors affecting the doseresponse relationship in steroid-induced
ocular hypertension. In: Kaufman HE, ed.
Symposium on ocular anti-inflammatory
therapy. Springfield, IL: Charles C Thomas;
1970:88.
11. Hovland KR, Ellis PP: Ocular changes in
renal transplant patients. Am J Ophthalmol
1967; 63:283.
12. Becker B, Mills DW: Corticosteroids and
intraocular pressure. Arch Ophthalmol
1963; 70:500.
13. Becker B: Intraocular pressure response to
topical corticosteroids. Invest Ophthalmol
Vis Sci 1965; 4:198.
14. Nicholas JP: Topical corticosteroids and
aqueous humor dynamics. Arch
15. Spaeth GL: Effects of topical
dexamethasone on intraocular pressure
and the water drinking test. Arch
16. Levene R, Wigdor A, Edelstein A, Baum J:
Topical corticosteroid in normal patients
and glaucoma suspects. Arch Ophthalmol
1967; 77:593.
17. Spiers F: Topical steroids and intraocular

pressure. I. Clinical investigation on the
reactions of 93 outpatients to monocular
steroid provocation and to subsequent
water drinking test. Acta Ophthalmol 1965;
43:735.
18. Weekers R, Grieten J, Collignon-Brach J:
Contribution ltude de lhypertension
oculaire provoque par la dexamethasone
dans le glaucoma angle ouvert.
19. Podos SM, Becker B, Morton WR: High
myopia and primary open angle glaucoma.
20. Armaly MF: Effect of corticosteroids on
intraocular pressure and fluid dynamics.
II. Effect of dexamethasone in the
glaucomatous eye. Arch Ophthalmol 1963;
70:98.
21. Becker B, Hahn KA: Topical corticosteroids
and heredity in primary open angle
glaucoma. Am J Ophthalmol 1974;
57:543.
22. Armaly MF: Inheritance of dexamethasone
hypertension and glaucoma. Arch
23. Lee PF: The influences of systemic steroid
therapy on the intraocular pressure. Am J
24. Ballin N, Becker B: Provocative testing for

primary open angle glaucoma in senior
citizens. Invest Ophthalmol Vis Sci 1967;
61:121.
25. Becker B, Ballin N: Glaucoma and
corticosteroid provocative testing. Arch
26. Becker B, Chevrette L: Topical
corticosteroid testing in glaucoma siblings.
27. Becker B, Podos SM: Krukenbergs spindle
and primary open angle glaucoma. Arch
28. Becker B: Diabetes mellitus and primary
open angle glaucoma. The XXVII Edward
Jackson Memorial Lecture Am J
29. Bernstein HN, Schwartz B: Effects of longterm systemic steroids on ocular pressure
and tonographic values. Arch Ophthalmol
1962; 68:742.
30. Bernstein HN, Mills DW, Becker B: Steroidinduced elevation of intraocular pressure.
31. Linner E: Adrenocorticosteroids and
aqueous humor dynamics. Doc Ophthalmol
1959; 13:210.
32. Polansky JR, Weinreb RN: Antiinflammatory agents-steroids as antiinflammatory agents. In: Sears ML, ed.
Pharmacology of the eye. New York:
Springer; 1984:459.
33. Weinreb RN, Mitchell M, Polansky JB:
Prostaglandin synthesis by human
trabecular cells: inhibitory effect of
dexamethasone. Invest Ophthalmol Vis Sci
1983; 24:15411545.
34. Spaeth GL, Rodriques MM, Weinreb S:
Steroid induced glaucoma. A. Persistent
elevation of intraocular pressure. B.
Histopathological aspects. Trans Am
35. Francois J, Victoria-Troncoso V:
Corticosteroid glaucoma. Ophthalmologica
1977; 174:195.
36. Knepper PA, Brun M, Weinstein HG, Black LT:
Intraocular pressure and glycosaminoglycan
distribution in the rabbit eye: effect of age
and dexamethasone. Exp Eye Res 1978;
27:567.
37. Polanksy JR, Gospodarowicz D, Weinreb
RN, Alvarado JJ: Human trabecular
meshwork cell culture and
glycosaminoglycan synthesis. Invest
Ophthalmol Vis Sci 1978; 17(Suppl):207.
38. Yoshiaki K, Horie T: The prognosis of
corticosteroid-responsive individuals. Arch
39. Rubin B, Palestine AG: Complications of
ophthalmic drugs and solutions. Int
40. Briggs HH: Glaucoma associated with the
use of topical corticosteroids. Arch
41. Goldmann H: Cortisone glaucoma. Arch
42. Black RL, Oglesby RB, von Sallman L,
Bunim JJ: Posterior subcapsular cataracts
induced by corticosteroids in patients with
rheumatoid arthritis. JAMA 1960; 174:150.
43. Oglesby RB, Black RL, von Sallman T,
Bunim JJ: Cataracts in rheumatoid arthritis
patients treated with corticosteroids. Arch
44. Giles CL, Mason GL, Duff IF, McLean JA:
The association of cataract formation and
systemic corticosteroid therapy. JAMA
1962; 182:179.
45. Crews SJ: Posterior subcapsular lens

opacities in patients on long-term
corticosteroid therapy. Br J Med 1963;
5346:1644.
46. Loredo A, Rodriquez RS, Murillo L:
Cataracts after short-term corticosteroid
treatment. N Engl J Med 1962; 286:160.
47. Skalka H, Prchal JT: Effect of
corticosteroids on cataract formation. Arch
48. Urban RC, Cotlier E: Corticosteroidinduced cataracts. Surv Ophthalmol 1986;
31:102.
49. Sundmark E: The cataract-inducing effect
of systemic corticosteroid therapy. Acta
50. Braver DA, Richards RD, Good TA:
Posterior subcapsular cataracts in steroidtreated children. Arch Ophthalmol 1967;
77:161.
51. Forman AR, Loreto JA, Tina LU:
Reversibility of corticosteroid-associated
cataracts in children with nephrotic
syndrome. Am J Ophthalmol 1977; 84:75.
52. Spaeth GL, von Sallmann L:
Corticosteroids and cataracts. Int
53. Yablonski ME, Burde RM, Kolker AE,
Becker B: Cataracts induced by topical
dexamethasone in diabetics. Arch
54. Harris JE, Gruber L: The electrolyte and
water balance of the lens. Exp Eye Res
1962; 1:372.
55. Ashton N, Cook C: Effect of cortisone on
healing of corneal wounds. Br J
56. Aquavella JV, Gasset AR, Dohlman CH:
Corticosteroids in corneal wound healing.
57. Beams R, Linabery L, Grayson M: Effect of
topical corticosteroids on corneal wound
strength. Am J Ophthalmol 1968; 66:1131.
58. Newell FW, Dixon JM: Effect of
subconjunctival cortisone upon the
immediate union of experimental corneal
grafts. Am J Ophthalmol 1952; 34:979.
59. Gasset AR, Dohlman CH: The tensile
strength of corneal wounds. Arch
60. Palmerton ES: The effect of local cortisone
on wound healing in rabbit corneas. Am J
61. Fink AI, Baras I: Effect of steroids on tensile
strength of corneal wounds. Am J
62. Polack FM, Rosen PN: Topical steroids and
tritiated thymidine uptake. Effect on corneal
healing. Arch Ophthalmol 1967; 77:400.
63. Gasset AR, Lorenzetti DWC, Ellison EM,
Kaufman HE: Quantitative corticosteroid
effect on corneal wound healing. Arch
64. McDonald TO, Borgmann AR, Roberts MD,
Fox LG: Corneal wound healing.
I. Inhibition of stromal healing by three
dexamethasone derivatives. Invest
65. Brown SI, Weller CA, Vadrich AM: Effect of
corticosteroids on corneal collagenase
of rabbits. Arch Ophthalmol 1970; 70:744.
66. Srinivasan BD, Kulkarni PS: The effect of
steroidal and nonsteroidal antiinflammatory agents on corneal
re-epithelialization. Invest Ophthalmol Vis
Sci 1981; 20:688.
67. Ho PC, Elliott JH: Kinetics of corneal
epithelial regeneration. II. Epidermal growth
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
factor and topical corticosteroids. Invest

Leopold IH, Purnell JE, Cannon EJ, et al:
Local and systemic cortisone in ocular
disease. Am J Ophthalmol 1951; 34:361.
Lorenzetti DWC: Therapeutic and toxic
dose-response effect of corticosteroids.
In: Kaufman HE, ed. Symposium on ocular
anti-inflammatory therapy. Springfield, IL:
Charles C Thomas; 1970:205.
Flach AJ, Kraff MC, Sanders DR,
Tanenbaum L: Quantitative effect of 0.5%
ketorolac tromethamine solution and 0.1%
dexamethasone sodium phosphate solution
on postsurgical blood-aqueous barrier.
Wilson FM: Adverse external ocular effects
of topical ophthalmic medications. Surv
Hyndiuk RA, Chin GN: Corticosteroid
therapy in corneal disease. Int Ophthalmol
Clin 1973; 13:103.
Kleinert RW, Palmer MP: Anti-inflammatory
agents. In: Tabbara KF, Hyndiuk RA, eds.
Infections of the eye. Boston: Little, Brown;
1996:301.
Hughes WF: Treatment of HSV keratitis:
a review. Am J Ophthalmol 1999;
67:313328.
Jones BR: Prospects in treating viral
disease of the eye. Trans Ophthalmol Soc
UK 1967; 87:537.
Laibson PR: Ocular inflammation, viruses
and the two-edged sword of corticosteroid.
Med Clin North Am 1969; 53:1145.
Kubrick S, Takahashi GH, Leibowitz HM,
Laibson PR: Local corticosteroid therapy
and reactivation of herpetic keratitis. Arch
Wilhelmus KR, Gee L, Hauck WW, et al:
Herpetic eye disease study: a controlled
trial of topical corticosteroids for herpes
1994; 101:1883.
Davis SD, Sarff LD, Hyndiuk RA:
Corticosteroid in experimentally induced
Pseudomonas keratitis: failure of
prednisolone to impair the efficacy of
tobramycin and carbenicillin therapy. Arch
Aronson SB, Moore TE: Corticosteroid
therapy in central stromal keratitis. Am J
Leibowitz HM, Kupferman A: Topically
administered corticosteroids-effect on
antibiotic-treated bacterial keratitis. Arch
Stern GA, Okumoto M, Friedlander M,
Smolin G: The effect of combined
gentamicin-corticosteroid treatment on
gentamicin-resistant Pseudomonas
keratitis. Ann Ophthalmol 1980; 12:1011.
Badenoch PR, Hay GJ, McDonald PJ,
Coster DJ: A rat model of bacterial
keratitis: effect of antibiotics and
corticosteroid. Arch Ophthalmol 1985;
103:718.
Davis SD, Sarff LD, Hyndiuk RA: Antibiotic
therapy of experimental Pseudomonas
keratitis in guinea pigs. Arch Ophthalmol
1977; 95:1638.
Harbin T: Recurrence of a corneal
Pseudomonas infection after topical steroid
therapy. Am J Ophthalmol 1974; 58:670.
Carmichael TR, Gelfand Y, Welsh NH:
Topical steroids in the treatment of central
and paracentral corneal ulcers. Br J
CHAPTER 33
371
SECTION 4
372
87. Liesegang TJ: Bacterial and fungal keratitis.

Waltman SR, eds. The Cornea. New York:
88. Kaufman HE: Use of corticosteroids in
corneal disease and external diseases
of the eye. Int Ophthalmol Clin 1966;
6:827.
89. Mitsue Y, Hanabusa J: Corneal infections
after cortisone therapy. Br J Ophthalmol
1955; 39:244.
90. Ley AP: Experimental fungal infections of the
cornea. Am J Ophthalmol 1966; 42:59.
91. Jones DB: Principles in the management of
oculomycoses. Am J Ophthalmol 1975;
79:719.
92. ODay DM, Ray WA, Robinson R, Head
WS: Efficacy of antifungal agents in the
cornea. II. Influence of corticosteroids.
93. Koenig SB: Fungal keratitis. In: Tabbara KF,
Hyndiuk RA, eds. Infections of the eye.
Boston: Little, Brown; 1986:331.
94. Nussenblatt RN, Palestine AP: Uveitis:
fundamentals and clinical practice.
Chicago: Year Book; 1989.
94a. Insel PA: Analgesic-antipyretics and antiinflammatory agents; drugs employed in
the treatment of rheumatoid arthritis and
gout. In: Gilman AG, Rall TW, Nies AS,
Taylor P, eds. Goodman and Gilmans the
8th edn. New York: Macmillan; 1990:638.
95. Masudu K: Anti-inflammatory agents.
Nonsteroidal anti-inflammatory drugs. In:
Sears ML, ed. Pharmacology of the Eye.
New York: Springer; 1984:539.
96. Foster CS: Nonsteroidal anti-inflammatory
drugs and immunosuppressives. In:
Lamberts DN, Potter DE, eds. Clinical
ophthalmic pharmacology. Boston: Little,
Brown; 1987:173.
97. Flach AJ: Cyclo-oxygenase inhibitors in
ophthalmology. Surv Ophthalmol 1992;
36:259.
98. Flach AJ: Non-steroidal anti-inflammatory
agents. In: Olin BR, ed. Drug facts and
comparisons. St Louis: Wolters Kluwer;
1991:488a.
99. Kraff MC, Sanders DR, McGulgan L,
Raanan MG: Inhibition of blood-aqueous
humor barrier breakdown with diclofenac. A
fluorophotometric study. Arch Ophthalmol
1990; 108:380.
100. Sher NA, Barak M, Daya S, et al: Excimer
laser photorefractive keratectomy in high
myopia. Arch Ophthalmol 1992; 110:935.
101. Simon LS, Mills JA: Nonsteroidal antiinflammatory drugs. Part I. N Engl J Med
1980; 302:1179.
102. Simon LS, Mills JA: Nonsteroidal antiinflammatory drugs. Part II. N Engl J Med
1980; 302:1237.
103. Physicians desk reference. Montvale, NJ:
Medical Economics; 1997.
103. Pirson Y, van Ypersele de Strihou C: Renal
side effects of nonsteroidal antiinflammatory drugs: clinical relevance. Am
J Kidney Dis 1986; 8:338.
104. Stern RS, Bigby M: An expanded profile of
cutaneous reactions to NSAIDs. Reports to a
specialty-based system for spontaneous
reporting of adverse reactions to drugs.
JAMA 1984; 252:1433.
105. Fraunfelder FT, Meyer SM, eds. Druginduced ocular side effects and drug
interactions. Philadelphia: Lea & Febiger;
1989.
106. Asherov J, Schoenberg A, Weinberger A:

Diplopia following ibuprofen administration.
JAMA 1982; 248:649.
107. Registry of possible drug-induced ocular
side effects. Ophthalmology 1980; 87:87.
108. Palmer CAL: Toxic amblyopia from
ibuprofen. Br Med J 1972; 23:765.
109. Tullio CJ: Ibuprofen-induced visual
disturbance. Am J Hosp Pharm 1981;
38:1362.
110. Callum LMT, Bowen DI: Ocular side effects
of ibuprofen. Br J Ophthalmol 1971; 55:472.
111. Hamburger HA, Beckman H, Thompson R:
Visual evoked potentials and ibuprofen
(Motrin) toxicity. Ann Ophthalmol 1984;
16:328.
112. Burns CA: Indomethacin, reduced retinal
sensitivity, and corneal deposits. Am J
113. Carr RE, Siegel IM: Retinal function in
patients treated with indomethacin. Am J
114. Tillmann W, Keitel L: Indomethacin-induced
corneal deposits. Klin Monatsbl
Augenheilkd 1977; 170:756.
115. Katzman B: Pseudotumor cerebri: An
observation on review. Ann Ophthalmol
1981; 13:887.
116. Crawford JS, Lewandowski RL, Chan W:
The effect of aspirin on rebleeding in
traumatic hyphema. Am J Ophthalmol
1978; 80:543.
117. Physicians desk reference for
ophthalmology. Montvale, NJ: Medical
Economics; 2005.
118. Keates RH, McGowan KA: Clinical trial of
flurbiprofen to maintain pupillary dilation
during cataract surgery. Ann Ophthalmol
1984; 16:919.
119. Tinkleman DG, Rupp G, Kaufman H, et al:
Double-masked, paired comparison clinical
study of ketorolac tromethamine 0.5%
ophthalmic solution. Surv Ophthalmol
1993; 38(Suppl):133.
120. Ballas Z, Blumenthal M, Tinkleman DG,
et al: Clinical evaluation of ketorolac
tromethamine 0.5% ophthalmic solution for
the treatment of seasonal allergic
conjunctivitis. Surv Ophthalmol 1993;
38(Suppl):141.
121. Sher NA, Frantz JM, Talley A, et al: Topical
diclofenac in the treatment of ocular pain
after excimer PRK. Refract Corneal Surg
1993; 9:425.
122. Arshinoff S, DAddario D, Sadler C, et al: Use
of topical NSAIDs in excimer PRK. J Cataract
Refract Surg 1994; 20(Suppl):216.
123. Hotchkiss ML, Robin AL, Pallock IP,
Quigley HA: Nonsteroidal anti-inflammatory
agents after argon laser trabeculoplasty.
A trial with flurbiprofen and indomethacin.
124. Sabiston D, Tessler H, Sumers K, et al:
Reduction of inflammation following
cataract surgery by the NSAID, flurbiprofen.
125. Weinreb RN, Robin AL, Baerveldt G, et al:
Flurbiprofen pretreatment in argon laser
trabeculoplasty for primary open angle
glaucoma. Arch Ophthalmol 1984; 102:1629.
126. Hurwitz LM, Spaeth CL, Zakhour I, et al:
Comparison of the effect of flurbiprofen,
dexamethasone, and placebo on
cyclocryotherapy induced inflammation.
127. Leopold IH, Murray D: Noncorticosteroidal
anti-inflammatory agents in ophthalmology.
128. Miller D, Gruenberg P, Miller R, Bergamini

MVW: Topical flurbiprofen or prednisolone:
effect on corneal wound healing in rabbits.
129. Trousdale MD, Dunkel EC, Nesburn AB:
Effect of flurbiprofen on herpes simplex
keratitis in rabbits. Invest Ophthalmol Vis
Sci 1980; 19:267.
130. Asbell PA, Kenamar T, Torres MA: Effect of
flurbiprofen on herpes simplex keratitis in
rabbits. Invest Ophthalmol Vis Sci 1990;
31(Suppl):221.
131. Gritz DC, Lee TY, Kwitko S, McDonnell PJ:
Topical anti-inflammatory agents in an
animal model of microbial keratitis. Arch
Ophthalmol 1990; 108:1001.
132. Feinstein NC, Rubin B: Toxicity of
flurbiprofen sodium. Case report. Arch
133. Harker LA, Slichter SJ: The bleeding time
as a screening test for evaluation of platelet
function. N Engl J Med 1972; 287:155.
134. Gieser DK, Hodapp E, Goldenberg I, et al:
Flurbiprofen and intraocular pressure. Ann
135. Sitenga GL, Ing EB, Van Dellen RG, et al:
Asthma caused by topical ketorolac.
136. Garrison JC: Histamine and 5hydroxytryptamine (serotonin) and their
antagonists. In: Gilman AG, Rall TW, Nies
AS, Taylor P, eds. Goodman and Gilmans
the pharmacological basis of therapeutics.
8th edn. New York: Macmillan; 1990:575.
137. Caputo AR, Schnitzer RE: Systemic response
to mydriatic eye drops in neonates:
mydriatics in neonates. J Pediatr Ophthalmol
138. Fraunfelder FT, Scafidi AF: Possible
adverse effects from topical ocular 10%
phenylephrine. Am J Ophthalmol 1978;
85:447.
139. Meyer SM, Fraunfelder FT: Phenylephrine
hydrochloride. Ophthalmology 1980; 87:1177.
140. Mertins PS: Excessive self-medication with
naphazoline hydrochloride (privine
hydrochloride). JAMA 1947; 134:1175.
141. Mindlin RL: Accidental poisoning from
tetrahydrozoline eyedrops. N Engl J Med
1966; 275:112.
142. Rogers GW: Safe use of eye drops.
N Engl J Med 1966; 275:447.
143. Lansche RK: Systemic reactions to topical
epinephrine and phenylephrine. Am J
144. Heath P: Neosynephrine hydrochloride:
some uses and effects in ophthalmology.
145. Rosales T, Isenberg S, Leake R, Everett S:
Systemic effects of mydriatics in lowweight infants. J Pediatr Ophthalmol
146. Hurwitz P, Thompson JM: Uses of
naphazoline (Privine) in ophthalmology.
147. Gordon DM: Ocular decongestants. Am J
148. Epstein DL, Boger WP III, Grant WM:
Phenylephrine provocative testing in the
pigmentary dispersion syndrome. Am J
149. Rumelt MB: Blindness from misuse of overthe-counter eye medications. Ann
150. Weiss DI, Shaffer RN: Mydriatic effects of
one-eight percent phenylephrine: a
potential cause of angle-closure glaucoma.
151. Hanna C, Brainard J: Allergic

dermatoconjunctivitis caused by
phenylephrine. Am J Ophthalmol 1983;
95:703.
152. Komi T, Maeda I, Uno Y, Otsuka H:
Inhibitory effect of sodium chondroitin
sulfate on epithelial keratitis induced by
naphazoline. Nippon Geka Gakkai Zasshi
1964; 68:154.
153. Lisch K: Conjunctival alterations by
sympathomimetic drugs. Klin Monatsbl
154. Saraux H, Offret H, de Rancourt de
Mimerand E: Pseudopemphigus oculaire
induit par les collyres: propos de 3
observations. Bull Soc Ophthalmol Fr 1980;
80:41.
155. Herman DC, Bartley GB: Corneal opacities
secondary to topical naphazoline and
antazoline (Albalon-A). Am J Ophthalmol
1987; 103:110.
156. Mathias CGT, Maibach HI, Irvine A, Adler
W: Allergic contact dermatitis to
echothiophate iodide and phenylephrine.
157. Aronson SB, Yamamoto EA: Ocular
hypersensitivity to epinephrine. Invest
158. Hill K: Whats the angle on mydriasis? Arch
159. Lee PF: The influence of epinephrine and
phenylephrine on intraocular pressure. Arch
160. Soparkar CN, Wilhelmus KR, Koch DD, et al:
Acute and chronic conjunctivitis due to overthe-counter ophthalmic decongestants.
161. Abelson MB, Weston JH: Antihistamines.
In: Lamberts DN, Potter DE, eds. Clinical
Brown; 1987:417.
162. Farber AS: Ocular side effects of
antihistamine-decongestant combinations.
163. Koffler BH, Lemp MA: The effect of an
antihistamine (chlorpheniramine maleate)
on tear production in humans. Ann
164. Carruthers SG, Shoeman DW, Hignite CE,
Azarnoff DL: Correlation between plasma
diphenhydramine levels and sedative and
antihistamine effects. Clin Pharmacol Ther
1978; 23:375.
165. Fraunfelder FT: Interim report: national
registry of possible drug-induced ocular
side effects. Ophthalmology 1980; 87:87.
166. Ostler HB, Martin RG, Dawson CR: The use
of disodium cromoglycate in the treatment of
atopic ocular disease. In: Leopold IH, Burns
RP, eds. Symposium on ocular therapy. New
York: Wiley; 1977:99.
167. Foster CS, The Cromolyn Sodium
Collaborative Study Group: Evaluation of
topical cromolyn sodium in the treatment of
vernal keratoconjunctivitis. Ophthalmology
1988; 95:194.
168. Butrus SI, Weston JH, Abelson MB: Ocular
mast cell stabilizing agents. In: Lamberts
DN, Potter DE, eds. Clinical ophthalmic
pharmacology. Boston: Little, Brown;
1987:483.
169. Caldwell DR, Verin P, Hartwich-Young R,
et al: Efficacy and safety of lodoxamide
0.1% versus cromolyn sodium 4% in
patients with vernal keratoconjunctivitis.
170. Fahy GT, Easty DL, Collum MT, et al:
Randomized double-masked trial of
171.
172.
173.
174.
175.
176.
177.
178.
179.
180.
181.
182.
183.
184.
185.
186.
187.
188.
lodoxamide and sodium cromoglycate in

allergic eye disease. A multicenter study.
Santos CL, Huong AJ, Abelson MB, et al:
Efficacy of lodoxamide 0.1% in resolving
corneal epitheliopathy associated with
Foster CS, Calonge M: Atopic
97:992.
Allansmith MR, Rose RN: Ocular allergy
and mast cell stabilizers. Surv Ophthalmol
1986; 30:229.
Kahan BD: Cyclosporine nephrotoxicity:
pathogenesis, prophylaxis, therapy and
prognosis. Am J Kidney Dis 1986; 8:323.
Tomlanovich S, Golbetz H, Pelrath M, et al:
Limitations of creatinine in quantifying the
severity of cyclosporine-induced chronic
nephropathy. Am J Kidney Dis 1986; 8:332.
Loughran TP, Deeg HJ, Dahlberg S, et al:
Incidence of hypertension after marrow
transplantation among 112 patients
randomized to either cyclosporine or
methotrexate as graft vs host disease
prophylaxis. Br J Haematol 1985; 59:547.
Nussenblatt RB, Palestine AG:
Cyclosporine: immunology, pharmacology
and therapeutic uses. Surv Ophthalmol
1986; 31:159.
Atkinson K, Biggs J, Darveniza P, et al:
Cyclosporine-associated CNS toxicity after
allogeneic bone marrow transplantation.
N Engl J Med 1984; 310:527.
Thompson JF, Chalmers DH, Hunnisett AG,
et al: Nephrotoxicity of trimethoprim-cotrimoxazole in renal allograft recipients
treated with cyclosporine. Transplantation
1983; 36:204.
Morales JM, Andres A, Priete C, et al:
Reversible acute renal toxicity synergistic
effect between gentamicin and
cyclosporine. Clin Nephrol 1988; 29:272.
Morganstern GR, Powles R, Robinson B,
et al: Cyclosporin interaction with
ketoconazole and melphalan. Lancet 1982;
2:1342.
Handschumacher RE: Drugs used for
immunosuppression. In: Gilman AG, Rall
TW, Nies AS, Taylor P, eds. Goodman and
Gilmans the pharmacological basis of
therapeutics. 8th edn. New York:
Macmillan; 1990:1264.
Kessler M, Louis J, Renoult E, et al:
Interaction between cyclosporin and
erythromycin in a kidney transplant patient.
Eur J Clin Pharmacol 1986; 30:633.
Gupta SK, Barkan A, Johnson RWG,
Rowland M: Erythromycin enhances the
absorption of cyclosporin A. Br J Clin
Pharmacol 1988; 25:401.
McGetrick JJ, Peyman GA, Nyberg MA:
Amphotericin B methyl ester: evaluation for
intravitreous use in experimental fungal
10:25.
Freeman DJ, Laupacis A, Keown PA, et al:
Evaluation of cyclosporin-phenytoin
interaction with observation of cyclosporin
metabolite. Br J Clin Pharmacol 1984;
18:887.
Dorian P, Strauss M, Cardella C, et al:
Digoxin-cyclosporine interactions: severe
digoxin toxicity after cyclosporine
treatment. Clin Invest Med 1988; 11:108.
Aubert-Tulkens G, Van Hoof F, Tulkens P:
Gentamicin induced phospholipidosis in
189.
190.
191.
192.
193.
194.
195.
196.
197.
198.
199.
200.
201.
202.
203.
204.
205.
206.
207.
208.
cultured rat fibroblasts. Lab Invest 1979;

40:481.
Tulkens P, Trouet A: The uptake and
intracellular accumulation of aminoglycoside
antibiotics in lysosomes of cultured rat
fibroblasts. Lab Invest 1979; 40:481.
Libert J, Ketelbart-Balasse PE, Van Hoof F,
et al: Cellular toxicity of gentamicin. Am J
Laurent G, Carlier M-B, Rollman B, et al:
Mechanism of aminoglycoside induced
liposomal phospholipidosis: in vitro and in
vivo studies with gentamicin and amikacin.
Biochem Pharmacol 1982; 31:3861.
DAmico DJ, Caspers-Velu L, Libert J, et al:
Comparative toxicity of intravitreal
aminoglycoside antibiotics. Am J
Kane A, Barza M, Baum J: Intravitreal
injection of gentamicin in rabbits. Invest
Tabatabay CA, DAmico DJ, Hanninen LA,
Kenyon KR: Experimental drusen formation
induced by intravitreal aminoglycoside
injection. Arch Ophthalmol 1987; 105:826.
Tabatabay CA, DAmico DJ, Hanninen LA,
et al: Residual bodies in the retinal pigment
epithelium induced by intravitreal netilmicin.
Fleisher SJ, Rapp LM, Hollyfield JG:
Photoreceptor-specific degeneration
caused by tunicamycin. Nature 1984;
311:575.
Cobo LM, Forster RK: The clearance of
intravitreal gentamicin. Am J Ophthalmol
1981; 92:59.
Talamo JH, DAmico DJ, Hanninen LA,
et al: The influence of aphakia and
vitrectomy on experimental retinal toxicity
of aminoglycoside antibiotics. Am J
Conway BP, Campochiaro PA: Macular
infarction after endophthalmitis treated with
vitrectomy and intravitreal gentamicin. Arch
McDonald HR, Schatz H, Allen AW, et al:
Retinal toxicity secondary to intraocular
gentamicin injection. Ophthalmology 1986;
93:871.
Conway BP, Tabatabay CA, Campochiaro
PA, et al: Gentamicin toxicity in the primate
retina. Arch Ophthalmol 1989; 107:107.
Mayrovitz HN, Tuma RF, Weidman MP:
Leukocyte adherence in arterioles following
extravascular tissue trauma. Microvasc Res
1980; 20:264.
Schmid-Schonbein GW: Capillary plugging
by granulocytes and the no-reflow
phenomenon in the microcirculation. Fed
Proc 1987; 46:2397.
Hodges GR, Watanabe I: Chemical injury of
the spinal cord of the rabbit after
intracisternal injection of gentamicin.
J Neuropathol Exp Neurol 1980; 39:452.
Tabatabay CA, young LHY, DAmico DJ,
Kenyon KR: Immunocytochemical
localization of gentamicin in the rabbit
retina following intravitreal injection. Arch
Ophthalmol 1990; 107:723726.
Haines J, Vinores SA, Campochiaro PA:
Evolution of morphologic changes after
intravitreous injection of gentamicin. Curr
Eye Res 1993; 12:521529.
Campochiaro PA, Conway BP:
Aminoglycoside toxicity: a survey of retinal
specialists. Arch Ophthalmol 1991; 109:946.
Peyman GA, Carrol CP, Raichand M:
Prevention and management of traumatic
CHAPTER 33
373
209.
210.
211.
212.
213.
214.
SECTION 4
215.
216.
217.
218.
219.
220.
221.
222.
223.
224.
225.
226.
227.
374

87:320.
Derick RJ, Paylor R, Peyman GA: Toxicity
of imipenem in vitreous replacement fluid.
Campochiaro PA, Lim JI: Aminoglycoside
toxicity in the treatment of endophthalmitis.
Barza M, Kane A, Baum JL: Intraocular
levels of cefamandole compared with
cefazolin after subconjunctival injection in
18:250.
Cochereau-Massir I, Bauchet J, Faurisson
F, et al: Ocular kinetics of pefloxacin after
intramuscular and intravitreal administration
in albino and pigmented rabbits.
35:11121115.
Hodden JA, OCallaghan RJ, Reidy JJ,
et al: Transcorneal iontophoresis of
ciprofloxacin for therapy of aminoglycoside
resistant Pseudomonas aeruginosa
keratitis. Invest Ophthalmol Vis Sci 1990;
31(Suppl 40):570.
Stamer WD, Jahnke J, McDermott ML,
Snyder RW: Effect of ciprofloxacin on
rabbit corneal endothelial viability. Invest
Haller-Yeo J, OBrien TP, Green WR, et al:
Ciprofloxacin: Oral penetration into the eye
and retinal toxicity of intravitreal injection.
Steven SX, Fouraker BD, Jensen HG:
Intraocular safety of ciprofloxacin. Arch
Ophthalmol 1991; 109:1737.
Marchese AL, Slana VS, Holmes EW, Jay
WM: Toxicity on pharmacokinetics of
ciprofloxacin. J Ocul Pharmacol 1993;
9:6976.
Kawasaki K, Mochizuki K, Torisaki M, et al:
Antibiotic toxicity. Lens Eye Toxicity Res
1990; 7:693704.
Mochizuki K, Torisaki M, Wakabayashi K:
Effects of vancomycin and ofloxacin on
rabbit ERG in vivo. Jpn J Ophthalmol 1991;
35:435445.
Flach AJ, Peterson JS, Toby Mathias CG:
Photosensitivity to topically applied
sulfisoxazole ointment. Arch Ophthalmol
1982; 100:1286.
Boettner EA, Fralick FB, Wolter JR:
Conjunctival concretions of sulfadiazine.
Pryor JG, Apt L, Leopold IH: Intraocular
penetration of vancomycin. Arch
Garcia-Ferrer FJ, Pepose IS, Murray PR, et al:
Antimicrobial efficacy and corneal endothelial
toxicity of Dex-Sol corneal storage medium
supplemented with vancomycin.
Lindquist TD, Roth BP, Fritsche TR: Safety
and efficacy of corneal storage media.
32(Suppl):1063.
Kattan HM, Pflugfelder SC: Corneal
endothelial toxicity of vancomycin in
corneal preservation media. Invest
Choi TB, Lee DA: Transscleral and
transcorneal iontophoresis of vancomycin
in rabbit eyes. J Ocul Pharmacol 1988;
4:153.
Smith MA, Sorenson JA, Lowy FD, et al:
Treatment of experimental methicillinresistant Staphylococcus epidermidis
endophthalmitis with intravitreal
228.
229.
230.
231.
232.
233.
234.
235.
236.
237.
238.
239.
240.
241.
242.
243.
244.
245.
246.
247.
vancomycin. Ophthalmology 1986;

93:1328.
Borhani H, Peyman GA, Wafapoor H: Use
of vanomycin in vitrectomy infusion
solution and the value of evaluation of
retinal toxicity. Int Ophthalmol 1993;
17:8588.
Pflugfelder SC, Hernndez E, Fleister SJ,
et al: Intravitreal vancomycin: retinal toxicity,
clearance, interaction with gentamicin. Arch
Oum BS, DAmico DJ, Wong KW:
Intravitreal antibiotic therapy with
vancomycin and aminoglycoside: an
experimental study of combinations and
repetition injections. Arch Ophthalmol
1989; 107:1055.
Oum B, Wong K, Kwak H, et al: Intravitreal
antibiotic therapy with a combination of
vancomycin and aminoglycoside:
examination of the influence of repetitive
injections after vitreous and lens removal.
31(Suppl):510.
Kattan HM, Pflugfelder SC, Hernandez E,
Ravinowitz S: Retinal toxicity of combined
intravitreal vancomycin and aminoglycoside
in the rabbits eyes. Invest Ophthalmol Vis
Sci 1990; 31(Suppl):510.
Ryan SJ, ed: Retina. St Louis: CV Mosby;
1989.
endophthalmitis. Symp Postoperative
Endophthalmitis 1978; 85:357373.
Noske W: Inaccuracy in preparation of
intravitreal solutions. Arch Ophthalmol
1986; 104:1748.
Johns KJ, ODay DM: Pharmacologic
management of keratomycoses. Surv
Jones DB: New drugs for fungal bugs.
Newmark E, Ellison AC, Kaufman HE:
Pimaricin therapy of cephalosporum and
Fusarium keratitis. Am J Ophthalmol 1970;
69:458.
ODay DM, Head WS, Robinson RD,
Clanton JA: Corneal penetration of topical
amphotericin B and natamycin. Curr Eye
Res 1986; 5:877.
Foster CS, Lass JH, Moran-Wallace K,
Giovanoni R: Ocular toxicity of topical
99:1081.
Ellison AC: Intravitreal effects of pimaricin
in experimental fungal endophthalmitis. Am
Ellison AC, Newmark E: Intraocular effects
of pimaricin. Ann Ophthalmol 1976; 8:897.
ODay DM, Ray WA, Head WS, Robinson
RD: Efficacy of antifungal agents in the
cornea. IV. Amphotericin B methyl ester.
Foster JBT, Almeda E, Littman ML, Wilson
ME: Some intraocular and conjunctival
effects of amphotericin B in man and in the
rabbit. Arch Ophthalmol 1958; 60:555.
Green WR, Bennett JE, Goos RD: Ocular
penetration of amphotericin B. Arch
1973; 76:584.
1973; 76:584.
248. Souri EN, Green WR: Intravitreal

amphotericin B toxicity. Am J Ophthalmol
1974; 78:77.
249. Raichand M, Peyman GA, West CS, et al:
Toxicity and efficacy of vitrectomy fluids:
amphotericin B methyl ester in the
treatment of experimental fungal
11:246.
250. Foster CS, Lass JH, Moran K, Giovanoni R:
Ocular toxicity of topical antifungal agents.
18(Suppl):132.
251. Grossman R, Lee DA: Transscleral and
transcorneal iontophoresis of ketoconazole
in the rabbit eye. Invest Ophthalmol Vis Sci
1989; 30(Suppl):247.
252. Grossman R, Lee DA: Transscleral and
transcorneal iontophoresis of ketoconazole
in the rabbit eye. Ophthalmology 1989;
96:724.
253. Yoshizumi MO, Banihashemi AF:
Experimental intravitreal ketoconazole in
DMSO. Retina 1988; 8:210.
254. Silverman CA, Yoshizami MO: Ocular toxicity
of experimental intravitreal DMSO. J Toxicol
Cutan Ocul Toxicol 1983; 2:193.
255. Schulman JA, Peyman GA, Didlein J,
Fiscella R: Ocular toxicity of expermental
intravitreal itraconozole. Int Ophthalmol
1991; 15:2124.
256. Brooks JG, OBrien T, Wilhelmus KR, et al:
Comparative topical triazole therapy of
experimental candida albicans keratitis.
31(Suppl):570.
257. Walsh J, Gold A, Charles H, eds: PDR for
ophthalmology. Oradell, NJ: Medical
Economics Co; 1991.
258. Pavan-Langston D: Clinical evaluation of
adenine arabinoside and idoxuridine in the
treatment of ocular herpes simplex. Am J
259 Fraunfelder FT, ed: Drug-induced ocular
side effects and drug interactions. 3rd edn.
Philadelphia: Lea & Febiger; 1989.
260. Lass JH, Thoft DA, Dohlman CH:
Idoxuridine-induced conjunctival
cicatrization. Arch Ophthalmol 1983;
101:747.
261. Lamberts DW, Potter DE: Clinical
Brown; 1987.
262. Kaufman HE, Ellison ED, Townsend WM:
The chemotherapy of herpes iritis with
adenine arabinoside and cytarabine. Arch
263. Hill JM, Park H-N, Gangarosa LP, et al:
Iontophoresis of vidarabine
monophosphate into rabbit eyes. Invest
264. Pulido JS, Palacio M, Peyman GA, et al:
Toxicity of intravitreal antiviral drugs.
265. Udell IJ: Trifluridine-associated conjunctival
cicatrization. Am J Ophthalmol 1985;
99:363.
266. Maudgal PC, Van Damme B, Missoten L:
Corneal epithelial dysplasia after trifluridine
use. Arch Clin Exp 1983; 220:6.
267. Carmine AA, Brogden RN, Heel RC, et al:
Trifluridine: a review of its antiviral activity
and therapeutic use in the topical treatment
of viral eye infections. Drugs 1982; 23:329.
268. Wellings PC, Awdry PN, Bors FH, et al:
Clinical evaluation of trifluorothymidine in
the treatment of herpes simplex corneal
ulcers. Am J Ophthalmol 1972; 73:932.
269. Hyndiuk RA, Charlin RE, Alpren TVP,

Schultz RO: Trifluridine in resistant human
herpetic keratitis. Arch Ophthalmol 1978;
96:1839.
270. Maudgal PC, Vrijghem JC, Colemans M,
Missoten L: Effect of topical acyclovir
therapy on experimental herpes simplex
keratouveitis. Arch Ophthalmol 1985;
103:1389.
271. Peyman GA, Schulman JA, Khoobehi B,
et al: Toxicity and clearance of a
combination of liposome-encapsulated
ganciclovir and trifluridine. Retina 1989;
9:232.
272. Pang MP, Peyman GA, Nikoleit J, et al:
Intravitreal trifluorothymidine and retinal
toxicity. Retina 1986; 6:260.
273. Liu K-R, Peyman GA, Khoobehi B, et al:
Intravitreal liposome-encapsulated
trifluorothymidine in a rabbit model.
274. Klauber A, Ottovay E: Acyclovir and
idoxuridine treatment of herpes simplex
keratitis: a double blind clinical study. Acta
275. Pulido JS, Peyman GA, Lesar T, Vernot J:
Intravitreal toxicity of hydroxyacyclovir
(BW-B759U): a new antiviral agent. Arch
276. She S, Peyman GA, Schulman JA: Toxicity
of intravitreal injection of foscarnet in the
rabbit eye. Int Ophthalmol 1988;
12:151154.
277. Dia-Llopis M, Chipont E, Sanchez S, et al:
Intravitreal foscarnet for cytomegalovirus
retinitis in a patient with acquired
immunodeficiency syndrome. Am J
Ophthalmol 1992; 114:742747.
278. Turrini B, Tognon MS, DeCaro E, Secoh
AG: Intravitreal use of foscarnet: retinal
toxicity of repeated injections in the
rabbit eye. Ophthalmic Res 1994;
26:110115.
279. Kaufman PL, Robinson JC: Epinephrine,
norepinephrine and timolol effects on
outflow facility in the cynomolgus monkey.
Suppl):444.
280. Kaufman PL: The effects of drugs on
the outflow of aqueous humor. In:
Drance SM, ed. Applied pharmacology
in the medical treatment of glaucomas.
New York: Grune & Stratton; 1984:429458.
281. Neufeld AH, Bartels SP: Receptor
mechanisms for epinephrine and timolol.
In: Ltjen-Drecoll E, ed. Basic aspects of
glaucoma research. Stuttgart: FK
Schattauer; 1982:113122.
282. Thomas JV, Epstein DL: Timolol and
epinephrine in primary open angle
glaucoma: transient additive effect. Arch
283. Erickson-Lamy KA, Ostovar B, Hunnicutt
EJ, et al: Epinephrine increases facility
of outflow and trabecular meshwork
cAMP content in the human eye in vitro.
33:26722678.
284. Tripathi BJ, Tripathi RC: Effect of
epinephrine in vitro on the morphology,
phagocytosis and mitotic activity of human
trabecular endothelium. Exp Eye Res 1984;
39:731.
285. Ltjen-Drecoll E, Kaufman PL: Long-term
timolol and epinephrine in monkeys. II.
Morphological alterations in trabecular
meshwork and ciliary muscle. Trans
286. Ltjen-Drecoll E, Kaufman PL, Eichhorn M:

Long-term timolol and epinephrine in
monkeys. I. Functional morphology of the
ciliary process. Trans Ophthalmol Soc UK
1986; 105:180.
287. Funk R, Rohen JW: SEM studies of the
functional morphology of the ciliary
process vasculature in the cynomolgus
monkey: reactions after application of
epinephrine. Exp Eye Res 1988; 47:653.
288. Seki R: Scanning electron microscopic
observations on vascular casts of ciliary
processes in normal and topical
epinephrine-treated rats. Jpn J Ophthalmol
1988; 32:288.
289. Green K, Hatchett TL: Regional ocular
blood flow after chronic topical glaucoma
drug treatment. Acta Ophthalmol 1987;
65:503.
290. Grant WM: Toxicology of the eye. 3rd edn.
Springfield, IL: Charles C Thomas; 1986.
291. Brummett RE: Warning to otolaryngologists
using local anesthetics containing
epinephrine: potential serious reaction
occurring in patients treated with beta
adrenergic receptor blockers. Arch
292. Fiore PM, Cinotti AA: Systemic effects of
intraocular epinephrine during cataract
surgery. Ann Ophthalmol 1988; 20:23.
293. Yamaguchi H, Matsumoto Y: Stability of
blood pressure and heart rate during
intraocular epinephrine irrigation. Ann
294. Dupeyron G, Eledjan JJ, Poupard P, et al:
Perfusion dadrenaline intra-camerulaire
dans la chirugie extra-capsulaire du
cristallin interet et limites de la methode.
Bull Soc Ophtalmol Fr 1985; 85:631.
295. Novack GD, Leopold IH: The toxicity of
topical ophthalmic beta blockers. J Toxicol
Cutan Ocul Toxicol 1987; 6:283.
296. Novack GD: Ophthalmic beta-blockers
since timolol. Surv Ophthalmol 1987;
31:307.
297. James IM: Pharmacologic effects of betablocking agents used in the management of
glaucoma. Surv Ophthalmol 1989; 33:453.
298. Chang SC, Bundgaard H, Buur A, et al:
Improved corneal penetration of timolol by
prodrugs as a means to reduce systemic
drug load. Invest Ophthalmol Vis Sci 1987;
28:487.
299. Potter DE, Shumate DJ, Bundgaard H,
et al: Ocular and cardiac b-antagonism by
timolol prodrugs timolol and levobunolol.
Curr Eye Res 1988; 7:755.
300. Timoptic-XE (timolol maleate ophthalmic
gel forming solution, 0.25% and 0.5%):
product insert. West Point, PA: Merck &
Co; 1993.
301. Kaila T, Salminen L, Huupponen R:
Systemic absorption of topically applied
ocular timolol. J Ocul Pharmacol 1985;
1:79.
302. Nelson WL, Fraunfelder FT, Sills JM, et al:
Adverse respiratory and cardiovascular
events attributed to timolol ophthalmic
solution 19781985. Am J Ophthalmol
1986; 102:606.
303. Brown JH, McGeown MG: Chronic renal
failure associated with topical application
of paraphenylenediamine. Br Med J 1987;
294:155.
304. Collignon P: Cardiovascular and pulmonary
effects of b-blocking agents: implications
for their use in ophthalmology [summary].
Surv Ophthalmol 1989; 33(Suppl):455.
305. Fraunfelder FT, Meyer SM: Ocular

toxicology update. Aust J Ophthalmol
1984; 12:391.
306. Schneeweiss A: Drug therapy in
cardiovascular diseases. Philadelphia: Lea
& Febiger; 1986:793794.
307. Harrison R, Kaufmann CS: Clonidine:
effects of a topically administered solution
on intraocular pressure and blood pressure
in open-angle glaucoma. Arch Ophthalmol
1977; 95:1368.
308. Hodapp E, Kolker AE, Kass MA: The effect of
topical clonidine on intraocular pressure. Arch
309. Kaskel D, Becker H, Rudolf H:
Fhwirkungen von Clonidin Adrenalin und
Pilocarpin auf den Augennendruck und
Episkleralvenendruck des gesunden
menschlichen Auges. Graefes Arch Clin
310. Krieglstein GK, Langham ME, Leydhecker
W: The peripheral and central neural
actions of clonidine in normal and
glaucomatous eyes. Invest Ophthalmol Vis
Sci 1978; 17:149.
311. Ralli R: Clonidine effect on the intraocular
pressure and eye circulation. Acta
312. Lee DA, Topper JE, Brubaker RF: Effect of
clonidine on aqueous humor flow in normal
human eyes. Exp Eye Res 1984; 38:239.
313. Mittag TW, Tormay A: Drug responses of
adenylate cyclase in iris-ciliary body
determined by adenine labelling. Invest
314. Abrams DA, Robin AL, Pollack IP, et al: The
safety and efficacy of topical 1% ALO 2145
(p-aminoclonidine hydrochloride) in normal
volunteers. Arch Ophthalmol 1987;
105:1205.
315 Gharagozloo NZ, Relf SJ, Brubaker RF:
Aqueous flow is reduced by a-adrenergic
agonist apraclonidine hydrochloride (ALO
2145). Ophthalmology 1988; 95:1217.
316. Robin AL, Pollack IP, de Faller JM: Effects
of topical 1% ALO 2145 (p-aminoclonidine
hydrochloride) on the acute intraocular
pressure rise after argon laser iridotomy.
317. Brown RH, Stewart RH, Lynch MG, et al:
ALO 2145 reduces the intraocular pressure
elevation after anterior segment laser
surgery. Ophthalmology 1988; 95:378.
318. Robin AL, Pollack IP, House B, et al: Effects
of ALO 2145 on intraocular pressure
following argon laser trabeculoplasty. Arch
319. Pollack IP, Brown RH, Crandall AS, et al:
Prevention of the rise in intraocular
pressure following neodymium-YAG
posterior capsulotomy using topical 1%
apraclonidine. Arch Ophthalmol 1988;
106:754.
320. Jampel HD, Robin AL, Quigley HA, et al:
Apraclonidine hydrochloride: a one-week
dose response study. Arch Ophthalmol
1988; 106:1069.
321. Morrison JC, Robin AL: Adjunctive
glaucoma therapy: a comparison of
apraclonidine and dipivefrin when
added to timolol maleate. Ophthalmology
1989; 96:3.
322. Alphagan (brimonidine tartrate ophthalmic
solution): product monograph. Irvine, CA:
Allergan, Inc; 1996:142.
323. Trusopt (dorzolamide hydrochloride
ophthalmic solution). Product insert. West
Point, PA: Merck & Co; 1994.
CHAPTER 33
375
SECTION 4
376
324. Kass MA: Topical carbonic anhydrase

inhibitors. Am J Ophthalmol 1989; 107:280.
325. Podos SM, Serle JB: Topically active
carbonic anhydrase inhibitors for
glaucoma. Arch Ophthalmol 1991; 109:38.
326. Buclin T, Lippa EA, Biollaz J, et al: Absence
of metabolic effects of the novel topically
active carbonic anhydrase inhibitor MK-927
and its S-isomer during a two-week ocular
administration. Eur J Clin Pharmacol Ther
1991; 49:665673.
327. Maren TH, Bar-llan A, Conroy CW, et al:
Chemical and pharmacological properties
of MK-927 a sulfonamide carbonic
anhydrase inhibitor that lowers intraocular
pressure by the topical route. Exp Eye Res
1990; 50:27.
328. Sugrue MF, Gautheron P, Grove J, et al:
MK-927: A topically active ocular
hypotensive carbonic anhydrase inhibitor.
329. Wang RF, Serle JB, Podos SM, et al: The
effect of MK-927, a topical carbonic
anhydrase inhibitor, on IOP in
glaucomatous monkeys. Curr Eye Res
1990; 9:163.
330. Wang RF, Serle JB, Podos SM, et al: The
ocular hypotensive effect of the topical
carbonic anhydrase inhibitor L-671 152 in
glaucomatous monkeys. Arch Ophthalmol
1990; 108:511.
331. Lippa EA, Von Denffer HA, Hofmann HM,
et al: Local tolerance and activity of MK927: a novel topical carbonic anhydrase
inhibitor. Arch Ophthalmol 1988; 106:1694.
332. Bron AM, Lippa EA, Hofmann HM, et al:
MK-927: A topically effective carbonic
anhydrase inhibitor in patients. Arch
Ophthalmol 1989; 107:1143.
333. Pfeiffer N, Hennekes R, Lippa EA, et al:
MK-927: A single-dose efficacy of a novel
topical carbonic anhydrase inhibitor. Br J
334. Higginbotham EJ, Kass MA, Lippa EA,
et al: MK-927: A topical carbonic
anhydrase inhibitor: dose response and
duration of action. Arch Ophthalmol 1990;
108:65.
335. Serle JB, Lustgarten JS, Lippa EA, et al:
MK-927: a topical carbonic anhydrase
inhibitor. Arch Ophthalmol 1990; 108:838.
336. Bron A, Lippa EA, Gunning F, et al:
Multiple-dose efficacy comparison of the
two topical carbonic anhydrase inhibitors
sezolamide and MK-927. Arch Ophthalmol
1991; 109:50.
337. Sugrue MF, Mallorga P, Schwam H, et al: A
comparison of L-671 152 and MK-927: two
topically effective ocular hypotensive
carbonic anhydrase inhibitors in
experimental animals. Curr Eye Res 1990;
108:607.
338. Goh Y: The metabolism and actions of
prostaglandins in the eye. Folia Ophthalmol
Jpn 1989; 40:2589.
339. Bito LZ: Prostaglandins and other
eicosanoids: their ocular transport
pharmacokinetics and therapeutic effects.
Trans Ophthalmol Soc 1986; 105:162.
340. Yamane A, Tokura T, Sano T, et al:
Experimental studies on the relationship of
prostaglandin to the occurrence of corneal
edema and neovascularization in anterior
segmental ischemia in the rabbit eyes.
Folia Ophthalmol Jpn 1987; 38:1579.
341. Kaufman PL: Effects of intracamerally
infused prostaglandins on outflow facility in
cynomolgus monkey eyes with intact or
342.
343.
344.
345.
346.
347.
348.
349.
350.
351.
352.
353.
354.
355.
356.
357.
retrodisplaced ciliary muscle. Exp Eye Res

1986; 43:819.
Crawford K, Kaufman PL: Pilocarpine
antagonizes prostaglandin F2ainduced
ocular hypotension in monkeys. Evidence
for enhancement of uveoscleral outflow by
prostaglandin F2a. Arch Ophthalmol 1987;
105:1112.
Kerstetter JR, Brubaker RF, Wilson SE,
et al: Prostaglandin F2a-1-isopropylester
lowers intraocular pressure without
decreasing aqueous humor flow. Am J
Crawford K, Kaufman PL, Gabelt BT:
Effects of topical PGF2a on aqueous humor
dynamics in cynomolgus monkeys. Curr
Eye Res 1987; 6:1035.
Nilsson SFE, Samuelsson M, Bill A, et al:
Increased uveoscleral outflow as a possible
mechanism of ocular hypotension caused
by prostaglandin F2a-1-isopropylester in the
cynomolgus monkey. Exp Eye Res 1989;
48:707.
Gabelt BT, Kaufman PL: Prostaglandin F2a
increases uveoscleral outflow in the
cynomolgus monkey. Exp Eye Res 1989;
49:389.
Camras CB, Friedman AH, Rodrigues MM,
et al: Multiple dosing of prostaglandin F2a
or epinephrine on cynomolgus monkey
eyes. Invest Ophthalmol Vis Sci 1988;
29:1428.
Lee PY, Shao H, Xu L, Qu CK: The effect of
prostaglandin F2a on intraocular pressure in
normotensive human subjects. Invest
Villumsen J, Alm A: Prostaglandin
F2a-isopropylester eye drops: effects in
normal human eyes. Br J Ophthalmol 1989;
73:419.
Villumsen J, Alm A, derstm M:
Prostaglandin F2a-isopropylester eye drops:
effect on intraocular pressure in open-angle
glaucoma. Br J Ophthalmol 1989; 73:975.
Flach AJ, Eliason JA: Topical prostaglandin
E2 effects on normal human intraocular
pressure. J Ocul Pharmacol 1988; 4:13.
Fujiwara H, Fukutomi T, Katayama T, et al:
Study on the pharmacological effects of
prostaglandin E1 (PGE1) on human eye.
Report 2. Analysis of effect of PGE1 on
human retinal vessels. Acta Soc
Ophthalmol Jpn 1986; 90:393.
Xalatan (latanoprost solution 0.005%). A
new direction in glaucoma therapy. Product
monograph. Kalamazoo, MI: Pharmacia &
Upjohn; 1996:111.
Watson P, Stjernschantz J: A six month,
randomized, double-masked study
comparing latanoprost with timolol on
open-angle glaucoma and ocular
hypertension. Ophthalmology 1996;
103:126.
Gaudreault J, Fei D, Resit J, et al:
Preclinical pharmacokinetics of
ranibizumab after single intravitreal
administration. Invest Opthalmol Vis Sci
2005; 46: 726733.
Liu MH, Jin H, Floten HS, et al: Vascular
endothelial growth factor-mediated,
endothelium-dependant relaxation in
human internal mammary artery. Ann
Thorac Surg 2002; 73:819824.
Harada K, Friedman M, Lopez JJ, et al:
administration in chronic myocardial
ischemia. Am J Physiol 1996;
270:H1791H802.
358. Morlet N, Young S, Strachan D, Coroneo

MT: Technique of intravitreal injection. ANZ
J Ophthalmol 1993; 21:130131.
359. DAmico DJ, Bird AC: VEGF inhabition
study in ocular neovascularization-1
(Vision-1); Safety evaluation from the
pivotal macugen (pegatanib sodium)
clinical trials. ARVO Absract 2004; 2363.
360. The United States Food and Drug
Administration. Avastin (bevacizumab).
Drug label approved October 11, 2006.
361. Azab M, Benchaboune M, Blinder KJ, et al:
Verteporfin therapy of subfoveal choroidal
neovascularization in age-related macular
degeneration. Retina 2004; 24:1.
362. Azab M, Benchaboune M, Blinder KJ, et al:
Verteporfin therapy of subfoveal choroidal
neovascularization in age-related macular
363. Age-related Eye Disease Study Research
Group: A randomized, placebo-controlled,
clinical trial of high-dose supplementation
with vitamins C and E, beta carotene, and
zinc for age-related macular degeneration
and vision loss: AREDS report no 8. Arch
Ophthalmol 2001; 119:14171436.
364. Hall NF, Gale CR: Prevention of age related
macular degeneration. BMJ 2002; 325.
365. West S, Vitale S, Hallfrisch J: Are
antioxidants or supplements protective for
age-related macular degeneration? Arch
Ophthalmol 1994; 122:99104.
366. Seddon JM, Ajani UA, Sperduto RD, et al:
Dietary carotenoids, vitamins A, C and E, and
advanced age-related macular degeneration.
JAMA 1994; 272:14131420.
367. Omenn GS, Goodman GE, Thornquist MD,
et al: Effects of a combination of beta
carotene and vitamin a on lung cancer and
cardiovascular disease. N Engl J Med
1996; 334:11501155.
368. Hoffman HN, Phyliky RL, Fleming CR:
Zinc-induced copper deficiency.
Gastroenterology 1988; 94:508512.
369. Park JE, Keller GA, Ferrara N: The vascular
endothelial growth factor (VEGF) isoforms:
differential deposition into the subepithelial
extracellular matrix and bioactivity of
extrecellular matrix-bound VEGF. Am Soc
Cell Bio 4:13171326.
370. Rosenfield PJ, Brown DM, Heier JS, et al:
Ranibizumab for neovascular age-related
macular degeneration. N Engl J Med 2006;
355:14191431.
371. Brown DM Kaiser PK, Michels M, et al:
Ranibizumab versus verteporfin for
degeneration. N Engl J Med 2006;
355:14321444.
372. Gragoudas ES, Adamis AP, Cunningham
ET, et al: Pegaptanib for neovascular agerelated macular degeneration. N Engl J
Med 2004; 351:28052816.
373. Manzano RP, Peyman GA, Khan P, Kivilcim
M: Testing intravitreal toxicity of
bevacizumab (Avastin). Retina 2006;
26:257261.
374. Maturi RK, Bleau LA, Wilson DL:
Electrophysiologic findings after intravitreal
bevacizumab (Avastin) treatment. Retina
2006; 26:270274.
375. Spaide RF, Klancnik J, Sorenson J, et al:
Intravitreal bevacinzumab for choridal
neovascularization secondary to age-related
macular degeneration. Invest Ophthalmol
Vis Sci 2006; 47:E-Abstract 2962.
376. Pieramici D, Avery R, Rabena MD, et al:
Bevacizumab (Avastin) in the treatment of

378. Grant WM: Toxicology of the eye. 3rd edn.
Springfield, IL: Charles C Thomas; 1986.
379. Pavan-Langston D, Dunkel EC: Handbook
of ocular drug therapy and ocular side
effects of the systemic drugs. Boston:
Little, Brown; 1991.
380. Physicians desk reference: Oradell, NJ:

Medical Economics; 1991.
381. Imperia PS, Lazarus HM, Lass JH: Ocular
complications of systemic cancer
chemotherapy. Surv Ophthalmol 1989;
34:209.
CHAPTER 33

degeneration. Invest Ophthalmol Vis Sci
2006; 47:E-Abstract 3540.
377. Fraunfelder FT, Meyer SM: Drug-induced
ocular side effects and their drug
interactions. 3rd edn. Philadelphia: Lea &
Febiger; 1989.
377
SECTION 5
PRINCIPLES OF EPIDEMIOLOGY
Edited by Frederick L. Ferris III and Emily Y. Chew
CHAPTER
34
Epidemiology and Clinical Research

Jie Jin Wang and Tien Y. Wong
CLINICAL RESEARCH AND THE

SCIENTIFIC HYPOTHESIS
Clinical research seeks to answer a scientic question by conducting a study in humans. This question may cover etiology,
pathogenesis or risk factors of a particular disease, its natural
history and prognosis, and possible treatment options.
Table 34.1 shows the principles typically followed in conducting clinical research. Each clinical research project should
have a sound hypothesis that the proposed study will address.
Researchers will need to: (1) logically display all the evidence
supporting the hypothesis (research background information);
(2) ask a research question that is answerable with the proposed
study (study aim); (3) design a feasible project to provide the
highest quality of evidence as possible to answer the research
question (research methods); and (4) nally conduct the study.
Common examples of clinical research questions include the
following: Does smoking (risk factor exposure) increase the risk
of age-related macular degeneration (disease outcome)? Does
the use of systemic steroids (treatment exposure) increase the
risk of multiple sclerosis following optic neuritis (prognostic
outcome)? Will a retinal photography screening program
(intervention exposure) reduce blindness from diabetic retinopathy (effectiveness outcome)?
ESTIMATES FOR FREQUENCY OF

DISEASES (RATES)
In epidemiology research, rates are preferred to absolute
numbers. A rate provides the proportion of individuals with a
particular disease or a certain characteristic which facilitates
comparison between groups or studies, while the absolute
number of cases provides very little information because the
size of the group (the denominator) can vary widely. The two
rates commonly used in clinical research and epidemiological
studies are the prevalence rate and the incidence rate.
PREVALENCE RATE
The prevalence rate refers to the frequency with which a disease
or condition is present in the study sample of a specic
population under study at a particular point in time. Prevalence
relates to a condition present at the time of examination or
assessment (at a point in time), regardless of when that
condition developed. Prevalence is calculated as follows:
Prevalence (at a point in time) = n N
where n is the number of all cases with the condition at the
point in time, and N is the total size of the study sample.
In the Blue Mountains Eye Study, there were 253 participants

who had diabetes and 82 of these participants had signs of
diabetic retinopathy.1 Thus, the prevalence of diabetic
retinopathy is 82 253 = 32.4% among diabetics, with one in
three persons presenting diabetes affected with retinopathy.
The prevalence rate is important in assessing the disease
burden in a country or community, indicating the proportion of
people who are blind at a certain point in time in a population.
In the US, a study in 2004 estimated that the prevalence of
blindness, dened according to the WHO denition as bestcorrected visual acuity of <20/400 in the better eye, was 5 per
1000 or 0.5% in persons aged 40 years and older.2 Extrapolating
that to the US population, the authors estimated that more
than 900 000 Americans 40 years and older were blind.
Prevalence rates allow comparison between studies. For
example, the prevalence of blindness reported in that US study
can be compared with the prevalence of blindness of 4.3% in
an Indian population of similar age range, showing that the rate
is eight times higher in India.3 A study in Beijing, China among
4319 persons 40 years and older reported that the prevalence
of myopia was 21.8%.4 This prevalence rate is higher than
similarly aged white persons in Australia (17%),5 but considerably lower than similarly aged Chinese adults in Singapore
(38.7%).6
The prevalence rate of a specic condition can also be
compared between different time points to assess time trends,
i.e., whether the prevalence has increased or decreased after a
period of time. Examples of such time trend comparisons in the
Blue Mountains Eye Study have been reported for diabetic
retinopathy and cataract, which were assessed initially in
199294 and then subsequently in 199798.7,8
INCIDENCE RATE
In contrast to prevalence, the incidence rate refers to the frequency with which new cases of a disease or condition develops
over a dened period of time. Incidence is also called absolute
risk (as opposed to relative risk (RR), which is described below)
and is calculated as follows:
Incidence (over a dened period) = n N
where n is the number of new cases that develop over the
period chosen, and N is the total number of study subjects who
were free of the disease at the beginning of the time period.
The incidence rate is the rate a condition develops over a
time period in a particular population. In a study on the incidence of retinopathy among people with diabetes, Klein and
colleagues observed how many new cases of retinopathy developed over a 10-year period: 75% in participants with type-1
379
SECTION 5
TABLE 34.1. Principles of Clinical Research

1. Dene overall goal of project
Dene a clear overall goal of the research project. First, is the subject important? Does the project
address etiology, pathogenesis, natural history, treatment, or impact?
2. Perform comprehensive
literature review
Familiarize with all relevant background information. Identify what is known and unknown, what has
been done, and what are the remaining gaps in the literature
3. Identify specic research

questions
Specify a focused question and the underlying hypothesis behind the question. These questions
should be directly answerable by the study
4. Design an appropriate study
Select an appropriate study design that can answer the specic questions with the highest quality
evidence, taking into account feasibility and resources to do this study
5. Select sample size
Prior knowledge and pilot studies help determine the expected strength of association and expected
difference in study groups. Decide on a sample size that has adequate power, but is sufciently cost
effective
6. Select study population
Selection of cohort for cohort studies and selection of cases and controls for case control studies, etc.
7. Identify study site
Where will the study be conducted?
8. Identify methods to measure

exposure and outcome
Both exposure and outcome should be measured using objective, standardized methods
9. Determine if masking is needed
Ideally, both study exposures and outcomes should be determined in a masked fashion
10. Write a detailed protocol
Provide clear documentation of study progress. Important reference of all procedures; and basis of
Introduction, Methods, and Discussion sections in paper
11. Standardize study forms and

procedures
Procedures should be tested before study commencement usually in pilot settings. Quality
control procedures should be in place prior to study start and examined at frequent intervals
thereafter
12. Examine for possible bias
Determine if there are major biases early in the study
13. Review information as study

progress
Review factors for nonparticipation and response rates. If there is a major loss to follow up,
evaluate the patterns that might explain why. Patients lost to follow up are important sources of
selection bias
14. Examine the primary results at

the conclusion of the study
Results should be reported starting with primary endpoints. Subgroup

(often interesting) results should only be reported after the primary results
15. Perform appropriate

statistical tests
The best studies report simple statistics as true associations are usually obvious. Avoid
report-complicated statistics
16. Compare study results with

other studies
Consistency between studies increases the likelihood that the observed results are real
17. Consider alternative

explanations
Are the observed results due to chance, confounding or bias and are they clinically meaningful
18. Discuss inferences
Explain the meaning behind the study results. What do the results show? Do the observations have a
sound biological basis? Speculate on how the study may change clinical practice, but do not stretch
the conclusion beyond what the data show
19. Consider limitations
All studies have limitations. Were all confounders appropriately controlled for? Are there selection
information and other types of biases? Is there a signicant loss to follow-up?
20. Conclusions and future research
Conclude by directly answering the research question. Consider future research. The best studies
usually lead to further questions
21. Publish the study!
380
diabetes, 70% in participants with type-2 diabetes who were on

insulin treatment, and 50% in participants who were not on
insulin treatment.10
The incidence rate is commonly used to examine the effects
of treatment, typically in the setting of randomized clinical
trials (see section on Study Designs). In this situation, the
investigator compares the incidence of an outcome in a group
who received a particular treatment with the incidence of the
same outcome in the group who did not receive this treatment.
For example, the question whether antioxidant supplementations can reduce the risk of age-related macular degeneration
(AMD) was tested in the randomized control trial (RCT), the
Age-related Eye Disease Study, which compared the incidence of
advanced AMD in a group of patients taking antioxidant supplements with another group not taking these supplements.11
The association of dietary intake and/or supplements of
antioxidants with the incidence of developing advanced AMD

can be evaluated in an observational cohort study.12
Prevalence and Incidence

The prevalence and incidence rates are closely interrelated. If a
condition is irreversible (e.g., AMD or glaucoma), and if people
with it have the same mortality rate as the rest of the
population (an assumption that is not often true), then a
condition with a high incidence will also be highly prevalent.
Many epidemiological studies have provided estimates of both
prevalence and incidence of disease in specic populations.
Prevalence rates are assessed from a cross-sectional analysis of
the baseline data and incidence rate is subsequently determined
with follow-up of the same study sample over a period of time.
For example, Leske and colleagues studied glaucoma in a
population sample of black persons in the Barbados Eye Study.
STUDY DESIGNS
Clinical studies fall into one of two large categories: controlled
and uncontrolled (Table 34.2). Anecdotal case reports and case
series are the uncontrolled studies. Controlled studies have at
least one concurrent comparison group, generally a standard
care or untreated group. In ophthalmology, as in medicine in
general, there have been many uncontrolled case series studies
reporting new and exciting ndings, many of which have
been shown to be unproven in subsequent controlled studies.
Examples of these include the use of intravitreal steroids as a
primary monotherapy treatment for neovascular AMD15 and
the effectiveness of optic nerve sheath decompression surgery to
improve vision in patients with nonarteritic anterior ischemic
optic neuropathy.16
TABLE 34.2. Study Designs

Controlled Studies
Experimental
Randomized clinical trials
Observational
Cohort studies
Case-control studies
Cross-sectional studies
Uncontrolled Studies
Case reports and case caries
Controlled studies can be further divided into experimental

and observational studies. The preferred experimental study
design in clinical research is the RCT. The major difference
between experimental and observational studies is that in
observational studies, investigators have no control over the
allocation of intervention or exposure factors, while in the RCT,
the intervention or exposure is randomly allocated, making it
likely that both known and unknown confounders will be
similar in the intervention and control groups.
One of the major issues in controlled studies is comparability
between study groups in factors other than those under study.
In clinical research, it is important to dene, for example, exposures (risk factors, treatments, interventions) and outcomes
(disease development or progression). The link between
exposure and outcome can be typically expressed by the classic
2 2 table (Table 34.3). This table can be used to understand
estimates of the associations between exposure and outcome
(see further ahead).
RANDOMIZED CONTROLLED TRIALS

An RCT compares an outcome (often incidence or progression
of a disease) among groups with and without particular exposure(s) or intervention(s). The RCT is similar to an observational cohort study except that the allocation of exposure or
study intervention to participants is not self-selected (as in
observational cohort studies) but by random chance.
TABLE 34.3. Classic Two by Two Table in Research

Yes
Exposure
Outcome
No
Total
Yes
A+C
No
B+D
Total
A+B
C+D
A+B+C+D
The incidence or risk of outcome in the exposed group is A/(A + C).

The incidence or risk of outcome in non-exposed group is B/(B + D).
The relative risk of outcome in association with exposure is
{A/(A + C)} {B/(B + D)}.
Randomization is used to assign study participants with equal

probability to the intervention group(s) or the control group.
This makes it likely that patient characteristics (both known
and unknown) will be equally distributed between groups except
for the intervention under investigation, if the sample size is
sufciently large. Because this random allocation is dictated by
chance only, with no subjective influence on the intervention
allocation from either study subjects or study investigators, the
RCT is classied as an experimental study.
The RCT provides the highest quality of evidence in clinical
research and is the ideal study design to investigate the effectiveness and safety of a new treatment. For studying etiological
causes, harmful exposures or established treatments, however,
it is often unethical and not feasible to use RCT design. For
example, to determine if smoking is associated with risk of
AMD, it is obviously not possible or ethical to randomly assign
participants to cigarette smoking.
The Diabetes Control and Complications Trial (DCCT) was
a classic RCT that assessed the effect of intensive glycemic
control on the development of diabetic retinopathy and other
vascular complications in patients with type-1 diabetes.17 The
researchers randomly assigned patients to tight glycemic
control versus standard glycemic control, and compared the
incidence of retinopathy over a 6.5-year period between the two
groups. The study found that the incidence of retinopathy was
75% lower in the group assigned to tight glycemic control, thus
concluding that glycemic control is important in preventing
diabetic retinopathy. The results of this RCT provide the
cornerstone of diabetes management.
Appropriate conduct of the RCT is paramount to the interpretation of study results. Issues such as noncompliance with
the treatment, substantial or selective loss of follow-up of study
participants, the occurrence of unexpected adverse events and
the application of the study ndings from the study population
to appropriate target populations in the community are
important considerations. Even the procedure of randomization
itself is not a guarantee that patient characteristics are evenly
distributed. In some studies, particularly those with small
sample sizes, the distribution of characteristics may not always
be comparable, and it is essential to check this prior to making
comparisons and drawing conclusions from the RCT. In an
RCT studying whether lisinopril, an ACE inhibitor, would
reduce the rate of diabetic retinopathy progression in type-1
diabetes, the European controlled trial of lisinopril in insulindependent diabetes mellitus (EUCLID) study showed that
patients randomly assigned to lisinopril treatment had a lower
risk of diabetic retinopathy progression than controls.18
However, the investigators found that at baseline patients in the
lisinopril group had lower baseline hemoglobin A1C levels than
control patients. Thus, whether the observed lower risk of
retinopathy progression in the treatment group was actually due
CHAPTER 34
They initially reported the prevalence of glaucoma at baseline

(all glaucoma cases at the time of the survey),13 and then
subsequently reported the 4-year incidence of glaucoma (new
glaucoma cases that developed between the baseline and followup examination at 4 years) in the same population.14 Although
both prevalence and incidence are estimates of frequency of a
disease, the incidence rate is more valuable in understanding
the etiology of a disease, while the prevalence rate is more
valuable to help policymakers in evaluating the impact of a
disease on demand for health services.
381
to the effect of the intervention (lisinopril) or due to the
selection of patients with good glycemic control in the treatment group is unclear.
SECTION 5
PROSPECTIVE COHORT STUDIES

A prospective or cohort study is a study that follows a group of
individuals over time with an aim to determine the rate at
which a disease outcome occurs over the time interval. It is also
used to examine the prognosis (e.g., visual loss) of the disease
over time. Prospective cohort studies provide data on the
incidence of the disease and can answer research questions pertaining to etiology or risk factors, intervention, and prognosis.
Prospective cohort studies can also be used to determine if the
efcacy observed in RCT translates into real world effectiveness in community-based populations. For both RCTs and
cohort studies, assessment of risks is as important as assessment of efcacy.
In a prospective cohort study one can assess whether the
incidence or risk of a disease (e.g., glaucoma) is related to a
particular factor (e.g., race), a group of individuals is initially
recruited from a target population. These individuals are
classied on the basis of presence or absence of exposure to the
specic risk factor in question (i.e., white and black race), and
all individuals must be initially free of the disease under
investigation (i.e., do not have glaucoma). The study sample is
then followed over time to assess the incidence rates of the
disease, which are then compared (i.e., incidence of glaucoma in
whites versus blacks). The relative incidence is expressed as the
RR or risk ratios (see section on Measure of Associations).
In ophthalmology, two classic examples of prospective cohort
studies have examined the relationship between cigarette
smoking at baseline and the 5-year incidence of AMD: the
Beaver Dam Eye Study and the Blue Mountains Eye Study.19,20
Both studies showed that cigarette smoking was associated with
a threefold higher risk of AMD, independent of age and other
risk factors. The results of these two studies, and others, have
led to an increased public health awareness of the potential
blinding effects of smoking.
Advantage and Disadvantage of RCT and

Prospective Cohort Study
382
Traditionally, it has been claimed that observational studies nd

stronger treatment effects than RCT. However, this is not
always the case. A review of 136 reports covering 19 diverse
treatments showed that the treatment effects from observational studies and RCT were similar.21
The major difference between the two study designs,
however, is that the exposure or intervention factors are selfselected in cohort studies, and therefore cohort studies are
vulnerable to selection bias and other confounding issues
(see section on Bias). Observational studies of hormone replacement therapy (HRT) in middle-aged women provide a classic
example of how confounding can lead to spurious conclusions.
More than 50 observational studies showed apparent benets of
HRT on a variety of cardiovascular and health outcomes. In
2002, however, a large RCT, the Womens Health Initiative
study, showed that long-term use of HRT was associated with
an increased risk of invasive breast cancer, heart disease, stroke,
and pulmonary embolism.22 The RCT was terminated before
study completion and discontinuation of HRT was recommended for the 16 000 participating women. Observers
concluded that the difference between data from previous
observation studies and the RCT was due to uncontrolled
confounding.23 In the observational studies, women who were
on HRT were healthier and saw doctors on a more frequent
basis than women who did not receive HRT. When HRT users
had better outcomes, it was erroneously credited to a benecial
effect of HRT. This example illustrates that in cohort studies,
factors that determine whether a person received a particular
exposure or treatment can result in a signicant difference in
the study outcomes. Often these factors are unclear, and there
are many unknown reasons why participants may have selected
a specic treatment. These factors can easily confound the
assessment of effectiveness of intervention. However, as
discussed above, while the RCT is the ideal study design for any
new treatments, it is not always feasible or ethical to conduct
an RCT. In these cases, a cohort study provides the next best
level evidence.
CASE-CONTROL STUDY
A case-control study, sometimes referred to as the retrospective
study, differs from prospective cohort studies in one major
aspect. In the cohort study, the exposure factor is dened at the
beginning of the study and the outcome at the conclusion of the
study. In contrast, in a case-control study, the outcome is rst
determined at the beginning of the study and the exposure is
retrospectively assessed. The principles of conducting a casecontrol study are as follows. rst, the investigator chooses two
patients groups with and without the outcome of interest. The
group in which the individuals have the disease or outcome
(cases) is compared to the group in which they do not have the
disease or outcome (controls) for whether they had been
exposed in the past to the study factors of interest. Instead of
prospectively following the study groups over a period of time as
in cohort studies, exposure data are collected retrospectively
from cases and controls, or examined on study participants after
recruitment in the case-control study. By comparing the
frequency of different exposures, risk factors or characteristics
between the two groups, the investigator can determine if a
specic risk factor occurs more frequently in cases than
controls. If so, it suggests that this factor may be associated
with the disease outcome. The measure of this association in
the case-control study is the odds ratio (OR) (see section on
Measure of Associations).
Acute endophthalmitis is a devastating postoperative complication after cataract surgery. The risk factors for endophthalmitis are unclear. In a case-control study, Wong and Chee
investigated the risk factors associated with endophthalmitis
following cataract surgery.24 In a review they identied 34
patients with acute endophthalmitis presenting within 6 weeks
after cataract surgery (cases), and selected another group of
cataract surgery patients who did not have endophthalmitis
(n = 102). Findings showed that endophthalmitis cases were
more likely to have silicone intraocular lens (ve out of 34 cases
or 15.1%) compared with controls (four out of 102 controls or
4%). Thus, patients implanted with silicone lens were about
four times (15.1% vs 4%) more likely to have endophthalmitis
as patients implanted with polymethylmethacrylate lens. This
association is expressed as the OR (see section on Measure of
Associations).
Selection of cases and controls requires clearly dened
diagnostic criteria, ideally including gold standard tests being
used to diagnose the disease. Case denitions should be based
not only on a patients clinical history or symptoms but also on
objective evidence from pathological or other diagnostic tests.
Controls should be representative of the referent population
from which cases are selected (i.e., comparable), and should
have the same probability of being selected as cases, if they had
the disease as cases do. Cases and controls can be recruited
from patients in the hospital (hospital-based case-control study)
Advantages and Disadvantages of the CaseControl Study and the Cohort (Prospective) Study
Cohort and case-control studies have relative advantages and
disadvantages (Table 34.4). Compared with clinical trials, these
observational approaches studies are more prone to problems
associated with bias and uncontrolled confounding. Casecontrol studies are typically much less expensive and less time
consuming than cohort studies. Case-control studies do not
provide estimates pertaining to incidence rates and absolute
risks, but only ORs for association assessment. There are also
biases with known and unknown bias directions (such as selection bias and recall bias) inherent with case-control studies.
While they can be a useful means of generating research
questions, case-control studies cannot by themselves provide
sufcient evidence for causal inferences.
TABLE 34.4. Comparison of Cohort and Case-Control Studies

Cohort Studies
Case-Control
Studies
Causal inference
Can be made
Cannot be made
Estimating incidence rates
Yes
No
Estimating relative risks
Yes
No
Cost
High
Low
Time
Long
Short
Loss to follow-up
Potential problem Not an issue
Studying rare diseases
Inefcient design
Efcient design
Studying multiple outcomes
Able to
Not able to
Studying multiple risk factors Possible
Possible
Nested Case-Control Study

Sometimes a case-control study can be nested within a larger,
population-based cross-sectional or cohort study. This type of
hybrid study incorporates the advantage of a population-based
sampling design and the cost-effectiveness of performing
investigation of specic study factors on cases and randomly
selected or matched controls (instead of the whole study
population). In a case-control study nested in a populationbased cohort study, incident cases are usually dened at followup visits, and study factors collected or assessed at baseline are
retrospectively examined; thus a recall bias on these prior
collected factors can be avoided. An example of this type of casecontrol study was conducted in the Beaver Dam Eye Study, in
which investigators examined the relationship between retinal
microvascular signs (exposure) and cardiovascular mortality
(outcome).26 Cases were study participants who had died from
cardiovascular disease since the baseline examination and three
controls per case were selected from the baseline cohort,
matched on age and gender. Retinal photographs taken at

baseline were measured for various retinal microvascular signs
and the frequency of these signs were compared in cases and
controls. The study showed that retinal microvascular signs
were more frequent in cases than controls, and concluded that
these signs may be a risk marker for cardiovascular mortality.
The advantage of this design included the fact that retinal
photographs were taken prior to the occurrence of outcome and
that measurement of retinal microvascular signs from baseline
retinal photographs was limited only to cases and controls (total
of ~1200 photographs) and not the entire study population
(~5000 photographs).
CROSS-SECTIONAL STUDY (SURVEY)

A cross-sectional study is usually conducted in a representative
sample of the target population, either within a geographically
dened community or randomly drawn from the entire population. In a cross-sectional study, the exposure and disease
outcome are determined at the same time. This study design is
useful for estimating the prevalence of diseases as well as crosssectional associations with diseases.
The National Health and Nutrition Examination Survey
(NHANES) in the US is a typical cross-sectional survey.27 The
NHANES used a multistage probability sampling design to
recruit and examine the prevalence of visual impairment in
representative populations aged 12 years and older in the US,
reporting a prevalence of visual impairment (presenting visual
acuity of 20/50 or worse) of 6.4%. This type of study is
important for public health planning purposes.
Like the case-control study, cross-sectional studies allow one
to determine possible associations between exposures or risk
factors and disease outcomes. A major limitation of this study
design, however, is that the temporal relationship between
exposures and disease outcomes is not known, and therefore,
evidence from cross-sectional studies, as with other observational study approaches, can be used to generate research
questions (hypothesis generation), but not causal inference
(hypothesis testing). For example, in a cross-sectional study of
the association of myopia and cataract, Wong and colleagues
showed that participants who had high myopia were more likely
to have nuclear cataract.28 However, it is not possible to
conclude that the observed association is due to myopia leading
to the development of nuclear cataract, or due to nuclear
cataract itself as a cause of index myopia.
CHAPTER 34
or from the community (population-based case-control study).

The latter is less likely to be subject to selection bias and thus
will provide better quality of evidence than the former.
An example of a case-control study that examined risk factors
for AMD is the Eye Disease Case-Control Study.25 In this study,
the investigators recruited 421 patients with neovascular AMD
and 615 controls and examined possible risk factors through
interviews, clinical examinations, and laboratory analyses of
blood samples. The study was one of the rst to identify that
AMD was associated with cigarette smoking and hypercholesterolemia.
MEASURE OF ASSOCIATIONS
There are two common measures of associations between
exposure factors and dichotomous outcomes: the RR (including
hazard ratio, used in prospective studies with time-dependent
outcomes) and the OR. For continuous exposure and outcome
factors, correlation and linear regression models are often used
to assess the degree of association. The latter statistics will not
be discussed here.
RELATIVE RISK
The RR provides an estimate of the difference in incidence or
risk associated with an exposure compared with the incidence
or risk associated with the absence of the exposure. The RR
indicates the risk of developing the outcome/disease in the
exposed group (people with a risk factor) relative to the risk in
those who are not exposed (people without the risk factor). RR
is often calculated in cohort studies and the RCT. The RR is
383
dened as the ratio of two absolute risk measurements and

calculated as follows:
RR = Absolute risk among exposed Absolute risk among
unexposed
The RR can be interpreted as follows:
RR = 1; the risks are the same for the exposed and
unexposed, and the exposure is therefore not related to the
disease.
RR > 1; the risk is higher for the exposed than the
unexposed, and the exposure is positively related to the
disease.
RR < 1; the risk is lower for the exposed than the
unexposed, and the exposure is protective to the disease.
SECTION 5
ODDS RATIO
The OR involves comparing odds, or likelihood, and is dened
as the ratio of the odds of being a case in the exposed group
compared to the odds of being a case in the unexposed group.
The odds is not a proportion, but the probability that an event
occurs relative to the probability that the event did not occur.
Thus, odds = p (1p), where p is probability of having an
event (or being exposed) and 1p is the probability of not
having the event (or not exposed). The OR is usually calculated
when absolute risk or incidence rates cannot be calculated and
is therefore usually used in case-control or cross-sectional
studies.
OR = Odds of disease in the exposed Odds of disease in the
unexposed
The OR can be interpreted as follows:
OR = 1; the odds of having the disease is the same for the
exposed and unexposed, and the exposure is not related to
the probability of having the disease.
OR > 1; the odds of having the disease is higher for the
exposed than the unexposed, and the exposure is
associated with an increased probability of having the
disease.
OR < 1; the odds of having the disease is lower for the
exposed than the unexposed, and the exposure is
associated with a reduction in the probability of having the
disease.
INFORMATION BIAS
BIAS
DIAGNOSTIC AND SCREENING TESTS
The validity of a study is the extent to which the observed

association (e.g., smoking and AMD) is attributed to the study
exposure (i.e., smoking) rather than other factors. Bias occurs
when the true associations are distorted due to systematic
(nonrandom) differences during sample selection (selection
bias), assessment of exposure and outcome factors (information
bias, measurement error) or other factors. There are many types
of biases, but only the major types are discussed here.
SENSITIVITY AND SPECIFICITY
SELECTION BIAS
384
clinic controls is higher than the rate of diabetes in the general

population, as persons with diabetes are more likely to see an
eye doctor for retinopathy assessment. If this were the case, the
control group taken from eye clinic patients may have a higher
prevalence of hypertension than a control group selected from
the general population. Thus, even if hypertension was truly a
risk factor for AMD, the investigator may nd that the
frequency of hypertension in AMD cases and controls in the
study samples are similar, and falsely conclude that hypertension is not associated with AMD.
To enhance comparability, investigators may select controls
matched for cases on some important characteristics, most
commonly age, gender, and race. In the above example, the
investigator may choose to match for diabetes status between
AMD cases and controls. The closer the match, the more valid
are the comparisons between cases and controls. However, it is
often not feasible to match for more than three characteristics
between cases and controls. Also, if you match on a factor you
can not assess whether it is associated with the disease.
A particular form of selection bias seen in cohort studies is
survival bias. If an investigator were to conduct a cohort study
of the association between Alzheimers disease and AMD, but
participants with Alzheimers disease and AMD were more
likely to die prior to the follow-up visits, the investigator may
not be able to detect a true association between Alzheimers
disease and AMD. In this situation, selection bias, due to differential loss to follow up, affected the true association between
Alzheimers disease and AMD.
Selection bias may occur even in an RCT if the study participants were lost to follow up either substantially or differentially after randomization.
Selection bias occurs when the study population differs in some

systematic way that influences the study results and can render
them invalid. Observational studies, particularly case-control
studies, are prone to selection bias. For example, an investigator
may be interested in studying the association of hypertension as
a potential risk factor for AMD in a case-control study. The
investigator may choose as controls a sample of patients seen in
the eye clinic for other conditions, as long as they do not have
AMD. It is possible that the rate of diabetes among these eye
Information bias occurs during the collection of study exposure

or outcome factors. Interview data are particularly prone to
information bias, particularly if the interviewer or the patient
believes that a particular question on a risk factor is related to
the study outcome. For example, in a case-control study of
smoking and AMD, patients with AMD may be more likely to
remember and report a past history of smoking than controls,
who may dismiss a short prior history of smoking. This type of
information bias, called recall bias, can lead to either an over- or
underestimation of the true association.
There are two estimates that are fundamental to evaluating

diagnostic and screening tests: sensitivity and specicity. The
sensitivity is the probability of a positive test in subjects who
TABLE 34.5. Sensitivity and Specicity of a Test Disease
Test
Total
Yes
No
Yes
True
Positive (TP)
False Positive
(FP)
TP + FP
No
False
Negative (FN)
True Negative
(TN)
FN + TN
Total
TP + FN
FP + TN
Sensitivity of test = TP/(TP + FN).

Specicity of test = TN/(FP + TN).
Positive predictive value of test = TP/(TP + FP).
Negative predictive value of test = TN/(FN + TN).
VARIABILITY AND RELIABILITY

When two physicians examine the same patient for the presence
of a disease, they often do not arrive at the same diagnosis. The
variability between the two physicians for the same disease is
called interobserver variability. Additionally, when the same
physician examines the same patient again at another time, he or
she may not arrive at the same diagnosis at the subsequent
examination. The latter is termed intraobserver variability.
Interobserver and intraobserver variability provides an
estimate of the reliability of the measurement or test by different observers and by the same observer over time. It is not a
reflection of the validity or accuracy of the test (which is dened
above by the sensitivity and specicity), as the same observer
may have good reliability but make the same error during
repeated measurement; thus, the intraobserver variability can
be low but not valid or accurate.
To minimize interobserver and intraobserver variability,
objective measures with detailed criteria (including reference
photographs where appropriate) and frequent standardization
across observers and instruments are recommended. Standardized measures, such as automated blood pressure device, or
computer assisted imaging, will help to reduce the measurement noise introduced by human errors.
STATISTICS
SAMPLE SIZE
Apart from quality of study design and quality of data collected,
an important determination of eventual study success is the
calculation of the required sample size to detect a statistically
signicant association or difference between study groups.
There are numerous studies where a true difference existed but
the difference was not statistically signicant because the sample size was insufcient to demonstrate the true assoiation.
On the other hand, a study with a very large sample size

could detect even small associations or differences between
study groups that although statistically signicant may not be
clinically meaningful (see section on Clinical Versus Statistical
Signicance).
STATISTICAL SIGNIFICANCE: p VALUE

VERSUS CONFIDENCE INTERVALS
A test of signicance is used to determine whether an observed
association is due to chance. A p value provides the probability
that an association observed in a study (e.g., smoking and
AMD) might have arisen purely by chance. A p value of <0.05
implies that the likelihood that this observation has arisen by
chance alone is less than 5%. In other words, the probability
that this association was seen, when in fact there was no real
association, is less than 5%. A value of p < 0.01 indicates that
the probability that the association was due to chance alone is
less than 1 in 100. A small p value does not prove that the
association is absolutely real but only that the probability that
the association was a chance nding is highly unlikely. The p
value is only an accurate assessment if there is no important
bias or confounding affecting the study groups.
The condence interval (CI) is another test commonly seen
in clinical research and is a summary of precision around a
point estimate. A 95% CI indicates that if the study was
repeated multiple times, the observed associations would lie
within the CI boundaries 95% of the time. The narrower the CI,
the more precise is the observed point estimate. CI is more
informative than a p value. For example, a nonsignicant p
value by itself provides no information about the power of the
study to nd a difference between groups. However, the breadth
of CI indicates how large a difference is likely to exist between
study groups, whether the results are statistically signicant or
nonsignicant.30
Both the p value and CI depend on sample size and the degree
of variability of the data, standard deviation and standard error.
Studies with small sample size and large standard deviation or
standard error provide estimates with wide CTs, and are less
likely to be able to detect a signicant difference.
CHAPTER 34
have the disease, while the specicity is the probability of a

negative test in subjects who do not have the disease. Table 34.5
shows how these estimates are calculated. An ideal (usually not
feasible) test is to have both sensitivity and specicity of 100%.
Nonmydriatic retinal photography has been suggested as a
possible means to screen for diabetic retinopathy. However, this
type of photography may miss some patients with diabetic
retinopathy (false negative) and may also falsely identify
patients without retinopathy (false positive). The question is,
what is the sensitivity and specicity if nonmydriatic photography as a screening test for diabetic retinopathy, compared
with 7-eld, mydriatic retinal photography (the current gold
standard)? The sensitivity tells us how frequently patients with
diabetic retinopathy are identied correctly from nonmydriatic
photography (true positive), and the specicity how frequently
patients without retinopathy have normal photographs (true
negatives).
To test a new diagnostic tool, the investigators usually need a
gold standard tool to compare against but each disease may not
always have a gold standard diagnostic test that is accepted by
the eld. Kuo and colleagues, for example, compared a single
eld, nonmydriatic retinal photography with a detailed ocular
examination by ophthalmologists, but it is not clear that the
latter can be considered gold standard in diagnosis of diabetic
retinopathy.29
An important characteristic of the sensitivity and specicity
analysis is that the results are independent of the prevalence of
the disease. As can be seen in Table 34.5, each statistic is
column specic: sensitivity only involves those with the disease
and specicity those without the disease.
CLINICAL VERSUS STATISTICAL

SIGNIFICANCE
A result can be statistically signicant (i.e., p value < 0.05) but
may not be clinically meaningful. A small size of the effect of an
intervention or a weak association between a risk factor and an
outcome can have still a very small p value, if the study sample
size is large enough.
Whether a result is clinically meaningful is usually a matter
of clinical judgment. Investigators should ask this question: Is
the association seen or the difference between study groups
large enough to be clinically important and worth achieving?
When investigators conclude that their study shows a highly
signicant result, in most instances the investigators mean
the results were statistically signicant. Investigators rarely
comment on whether the differences are large enough to affect
clinical practice. For example, a new drug might lower IOP by
2 mmHg more than another drug, but while the improvement
is statistically signicant and likely to be real, it may not be
clinically meaningful.
INFERENCES AND CAUSALITY

The goal of most studies is to determine whether an exposure
is associated with an outcome in the real world. However,
reported associations in studies should be described as they are,
385
TABLE 34.6. Principles of Causality

Principles of Causality
Examples
Strength of association
Strong association of cigarette smoking and risk of lung cancer34
Consistency of association between studies and coherence of

evidence
Smoking and AMD demonstrated in three population-based studies35
Specicity of association. Removal of the exposure is associated

with a reduction in risk of outcome
Control of hyperglycemia and reduction in risk of diabetic retinopathy17
Temporal relationship between exposure and outcome
Hyperglycemia and duration of diabetes link to subsequent

development of diabetic retinopathy36
Doseresponse of association
Increasing dose of inhaled corticosteroid use and increasing risk of

posterior subcapsular cataract risk37
Biologic plausibility of the association
SECTION 5
Conrmation in experiments
and no more and no less can usually be inferred than what was
actually observed. Additional studies and information are
usually needed to show causality between the exposure and
outcome. For example, if AMD rates are observed to be higher
in whites than blacks, this does not necessarily imply that
genetic factors play a role in the etiology of AMD.31 Further
studies showing evidence that whites have some genes that are
different from blacks and that these genes are responsible for
AMD development are needed. If patients with diabetes are
more likely to undergo cataract surgery than persons without
diabetes, it does not necessarily imply that diabetes is a cause of
cataract, until there is cumulated evidence from other studies.
In general observational studies can not test for causality in the
way that clinical trials can. However, there are many examples
where the totality of the evidence leads to a conclusion of
causality despite the lack of clinical trial results. At this point
there is little doubt that cigarette smoking causes lung cancer
even though there are no clinical trial data. There are ample
observational studies using different approaches demonstrating
consistently high ORs for lung cancer among smokers, as well
as biologic plausibility conrmed by animal models.
CAUSALITY
386
The demonstration of an association between an exposure and

an outcome in a study is only the rst step. To assess the
validity of the observed association, one must address whether
the observed association is a true association, or is due to
chance, confounding or bias. After determining the validity of
an association, one can consider whether it is causal.
First, one should determine if the association is likely to be a
chance observation? This question is usually determined by
examining whether the association is statistically signicant, as
indicated by the p value or the CI. However, a statistically
signicant association at the level of a p value of < 0.05 only
suggests that the probability of the association occurring by
chance is less than 5%. There is still a probability of 1 in 20 that
the observations occurred by chance alone, which is referred
to as a type I error (i.e., a signicant nding that is not true).
Thus, statistical signicance is not sufcient and other criteria
for causality are needed for correct interpretation of study
ndings.
Second, is the observed association between an exposure and
outcome due to an indirect association with a third factor, a
confounder? In other words, is the link between the exposure
and outcome explained by their common association with a

third factor (confounder)? For example, pterygium has been
observed to be more common in countries nearer the equator.32
Is residence near the equator (exposure) therefore a direct cause
of pterygium (outcome)? More likely, while this association may
be real, the geographic location itself is not a direct cause but a
surrogate for a longer duration of sunlight exposure, which is
associated with the geographic location (countries near the
equator). In this case, sunlight exposure is the confounder for
the association between residence near the equator and
pterygium.
In another study, people with diabetes who were using
insulin were more likely to have diabetic retinopathy than those
not using insulin. Does this then imply that use of insulin is a
direct risk factor for retinopathy? A more probable explanation
is that people who use insulin have more severe diabetes and
poorer glycemic control putting them at a higher risk for
diabetic retinopathy. In this example, poor glycemic control is
the confounder for the observed association between insulin use
(exposure) and diabetic retinopathy (outcome).
Third, is the observed association due to bias that may lead
to spurious inferences? An association can be due to selection
or information biases that the researchers have not taken into
consideration. Some study designs (e.g., RCT) minimize
possible biases and confounding factors better than other study
designs (e.g., case-control studies).
Finally, assuming that the association is not due to chance,
confounding, or bias, and is likely to be real, the question still
remains as to whether the association is due to a causal relationship. Criteria for causal inference were proposed by
Bradford Hill and are shown in Table 34.6.33
GENERALIZATION OF STUDY FINDINGS

Having demonstrated and precisely dened an association that
appears to be real and causal, the nal process of assessing a
study is to determine whether the ndings are widely applicable
or generalizable. The purpose of research is not conned to a
simple demonstration of an association in the study sample.
Equally important, one must assess whether the study results
can be extrapolated to the community and to clinical practice.
In other words, the aim of clinical research is to translate
research ndings to clinical practice, in order to improve health
services and health outcomes. Ideally, a random sample of the
entire relevant population should be studied. In practice, this is

Key Features
CONCLUSIONS
There are several challenges for clinical research in
ophthalmology. Investigators need to move progressively from
conducting a purely descriptive type of clinical studies
(hypothesis development) to controlled trials (hypothesis
testing). Research studies should incorporate objective,
quantitative measurements of exposures and outcomes.
The complex etiology and pathogenesis of most chronic eye
diseases poses an additional challenge for clinical researchers. It
is likely that identifying new genetic factors and examining
geneenvironmental interactions will become increasingly
important in understanding how these chronic diseases
develop. Finally, improvement in methods to measure change
and progression of disease outcomes (e.g., progression of
cataract) and standardizing and uniformly using denitions and
classication for diseases (e.g., glaucoma) will enhance the
quality of clinical research outcomes, leading to further
understanding of the nature and impact of the major blinding
ocular diseases.
Clinical research answers a scientic question by conducting

studies in humans. This question may cover etiology,
pathogenesis, risk factors, natural history, and treatment
options for this disease.
The prevalence of a disease refers to the frequency of existing
disease in a specic population at a particular point in time.
The incidence of a disease refers to the frequency with which
new cases of a disease develops over a dened period of time
(e.g., one month or one year).
Clinical studies can be divided into uncontrolled and controlled
studies. Anecdotal case reports and case series are the
uncontrolled studies. Controlled studies can be further divided
into experimental and observational studies.
The experimental study in clinical research is the randomized
control trial, in which the intervention (e.g., treatment) to study
participants is randomly allocated.
Observational studies include the cohort or prospective study,
the case-control or retrospective study or the cross-sectional
study. In all these studies, there is a control group in which
comparisons to disease or risk factors can be made.
As far as possible, clinical research studies should incorporate
objective, quantitative measurements of both exposures (risk
factor), and outcomes (disease).
REFERENCES
1. Mitchell P, Smith W, Wang JJ, Attebo K:
Prevalence of diabetic retinopathy in an
older community. The Blue Mountains Eye
Study. Ophthalmology 1998; 105:406411.
2. Congdon N, OColmain B, Klaver CC, et al:
Causes and prevalence of visual impairment
among adults in the United States. Arch
Ophthalmol 2004; 122:477485.
3. Thulasiraj RD, Nirmalan PK, Ramakrishnan
R, et al: Blindness and vision impairment in
a rural south Indian population: the Aravind
Comprehensive Eye Survey.
Ophthalmology 2003; 110:14911498.
4. Xu L, Li J, Cui T, et al: Refractive error in
urban and rural adult Chinese in Beijing.
Ophthalmology 2005; 112:16761683.
5. Wensor M, McCarty CA, Taylor HR:
Prevalence and risk factors of myopia in
Victoria, Australia. Arch Ophthalmol 1999;
117:658663.
6. Wong TY, Foster PJ, Hee J, et al:
Prevalence and risk factors for refractive
errors in adult Chinese in Singapore. Invest
7. Tan AG, Wang JJ, Rochtchina E, Mitchell P:
Comparison of age related cataract
prevalence in two population based
surveys 6 years apart. BMC Ophthalmol
2006; 6:17.
8. Cugati S, Kifley A, Mitchell P, Wang JJ:
Temporal trends in the prevalence of
diabetes and diabetic retinopathy: ndings
from two population based surveys of older
Australians. Diabetes Res Clin Pract 2006;
74:301308.
9. Seah SK, Foster PJ, Chew PT, et al:
Incidence of acute primary angle-closure
glaucoma in Singapore. An island-wide
survey. Arch Ophthalmol 1997;
115:14361440.
10. Klein R, Klein BE, Moss SE,
Cruickshanks KJ: The Wisconsin
Epidemiologic Study of diabetic
retinopathy. XIV. Ten-year incidence and
11.
12.
13.
14.
15.
16.
17.
18.
progression of diabetic retinopathy. Arch

Ophthalmol 1994; 112:12171228.
Age-Related Eye Disease Study Research
and vision loss: AREDS report no. 8. Arch
Ophthalmol 2001; 119:14171436.
van Leeuwen R, Boekhoorn S, Vingerling JR,
et al: Dietary intake of antioxidants and risk
of age-related macular degeneration. JAMA
2005; 294:31013107.
Leske MC, Connell AM, Schachat AP,
Hyman L: The Barbados Eye Study.
Prevalence of open angle glaucoma. Arch
Ophthalmol 1994; 112:821829.
Leske MC, Connell AM, Wu SY, et al:
Incidence of open-angle glaucoma: the
Barbados eye studies. The Barbados Eye
Studies Group. Arch Ophthalmol 2001;
119:8995.
Gillies MC, Simpson JM, Luo W, et al:
A randomized clinical trial of a single dose
of intravitreal triamcinolone acetonide for
degeneration: one-year results. Arch
Ophthalmol 2003; 121:667673.
The Ischemic Optic Neuropathy
Decompression Trial Research Group:
Optic nerve decompression surgery for
nonarteritic anterior ischemic optic
neuropathy (NAION) is not effective and
may be harmful. JAMA 1995; 273:625632.
The Diabetes Control and Complications
Trial Research Group: The effect of
intensive treatment of diabetes on the
development and progression of long-term
complications in insulin-dependent
diabetes mellitus. N Engl J Med 1993;
329:977986.
Chaturvedi N, Sjolie AK, Stephenson JM,
et al: Effect of lisinopril on progression of
retinopathy in normotensive people with
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
type 1 diabetes. The EUCLID Study Group.

EURODIAB Controlled Trial of Lisinopril in
Insulin-Dependent Diabetes Mellitus.
Lancet 1998; 351:2831.
Klein R, Klein BE, Moss SE: Relation of
smoking to the incidence of age-related
maculopathy. The Beaver Dam Eye Study.
Am J Epidemiol 1998; 147:103110.
Mitchell P, Wang JJ, Smith W, Leeder SR:
Smoking and the 5-year incidence of agerelated maculopathy: the Blue Mountains
Eye Study. Arch Ophthalmol 2002;
120:13571363.
Benson K, Hartz AJ: A comparison of
observational studies and randomized,
controlled trials. N Engl J Med 2000;
342:18781886.
Rossouw JE, Anderson GL, Prentice RL,
et al: Risks and benets of estrogen plus
progestin in healthy postmenopausal
women: principal results from the womens
health initiative randomized controlled trial.
JAMA 2002; 288:321333.
Enserink M: Womens health. The vanishing
promises of hormone replacement. Science
2002; 297:325326.
Wong TY, Chee SP: Risk factors of acute
endophthalmitis after cataract extraction:
a case-control study in Asian eyes. Brit J
The Eye Disease Case-Control Study
Group: Risk factors for neovascular agerelated macular degeneration. Arch
Ophthalmol 1992; 110:17011708.
Wong TY, Klein R, Nieto FJ, et al: Retinal
microvascular abnormalities and 10-year
cardiovascular mortality: a populationbased case-control study. Ophthalmology
2003; 110:933940.
Vitale S, Cotch MF, Sperduto RD:
Prevalence of visual impairment in the
United States. JAMA 2006; 295:21582163.
Wong TY, Foster PJ, Johnson GJ, Seah SK:
Refractive errors, axial ocular dimensions,
CHAPTER 34
rarely feasible and investigators have to assume that persons

with characteristics similar to those enrolled in their study will
respond in similar ways, although this is not always true.
387
SECTION 5
and age-related cataracts: the Tanjong

Pagar survey. Invest Ophthalmol Vis Sci
2003; 44:14791485.
29. Kuo HK, Hsieh HH, Liu RT: Screening for
diabetic retinopathy by one-eld, nonmydriatic, 45 degrees digital photography
is inadequate. Ophthalmologica 2005;
219:292296.
30. Rothman KJ: A show of condence. N Engl
J Med 1978; 299:13621363.
31. Klein R, Rowland ML, Harris MI:
Racial/ethnic differences in age-related
maculopathy. Third National Health and
388
Nutrition Examination Survey. Am J

Ophthalmol 1995; 102:371381.
32. Wong TY, Foster PJ, Johnson GJ, et al: The
prevalence and risk factors for pterygium in
an adult Chinese population in Singapore:
the Tanjong Pagar survey. Am J
Ophthalmol 2001; 131:176183.
33. Hill AB: Principles of medical statistics
London. Lancet 1971; 312320.
34. Doll R, Peto R, Boreham J, Sutherland I:
Mortality in relation to smoking: 50 years
observations on male British doctors. BMJ
2004; 328:1519.
35. Smith W, Assink J, Klein R, et al: Risk

factors for age-related macular
degeneration: pooled ndings from three
continents. Ophthalmology 2001;
108:697704.
36. Klein R, Klein BE, Moss SE, et al:
Glycosylated hemoglobin predicts the
incidence and progression of diabetic
retinopathy. JAMA 1988; 260:28642871.
37. Cumming RG, Mitchell P, Leeder SR: Use
of inhaled corticosteroids and the risk
of cataracts. N Engl J Med 1997;
337:814.
CHAPTER
35
Epidemiology of Age-Related Cataract

Barbara E. K. Klein
Key Features
Risk Factors for Age-Related Cataracts
Diabetes (cortical cataracts)
Smoking (nuclear cataracts)
Sunlight exposure (cortical cataracts)
Controversial Risk Factors for Cataracts
Dietary/nutritional supplements
Alcohol intake
Genetics
PUBLIC HEALTH SIGNIFICANCE

Cataract is thought to be the most common cause of blindness
worldwide1 and is the leading cause of diminished vision in the
United States.2 It is the single most common cause of blindness
in black Americans being associated with 36.8% of this important
loss of function.3 It is estimated that 20.5 million Americans
older than 40 years of age have cataracts in either eye and that
this prevalence will rise to 30.1 million by 2020.4 In developed
countries such as the United States, cataract surgery procedures
are both successful and widely available, but they carry a signicant cost. It has been estimated that if cataract formation were
delayed by 10 years, the need for cataract extraction surgery
might be reduced by 45%.5
AGE-RELATED CATARACT:
METHODOLOGIC CONSIDERATIONS
Prevalence data help dene the magnitude of the disease burden.
While prevalence surveys may not be undertaken primarily for
purposes of developing health policy, they may provide a gross estimate of the potential need for surgical and rehabilitative services.
A major advance in epidemiologic research on cataract has
been the development of photographic documentation and codied grading schemes to identify the presence and severity of
cataracts.6 Much of the data accumulated are based on clinical
examinations whose validity and reliability are difcult to assess.
The data we report, because of the relative paucity of use of lens
photography and gradings, include studies of several different
national and international populations using both grading of
standardized lens photographs and clinical gradings. We group the
cataract types for global descriptions of prevalence of (any)
cataract because many studies do not describe cataract prevalence by type of opacity. This is appropriate when visual function is the object of the analysis. However, when searching for
potentially etiologic relationships, it is necessary to investigate
specic cataract type.
PREVALENCE STUDIES
The Eye Disease Prevalence Research Group was composed of
investigators of several population-based studies of age-related
eye disease in order to estimate current prevalence of these
conditions in adult Americans and to project these ndings to
the year 2020. Data on cataract or cataract surgery were
included for 31448 adults 40 years of age or older, 1897 of
whom contributed to cataract prevalence.4 Data for ~15 000 of
these persons were derived from photographic documentation.
For all studies, roughly comparable denitions were used to
classify cataract presence. The data for estimates for white
Americans came from ve studies containing substantial numbers of persons with European background, two studies with
substantial numbers of persons of African ancestry, and one
study of Mexican-Americans (only cataract surgery data). The
pooled estimates from these data indicate increased prevalence
with increased age in blacks and whites. Both black and white
women had higher prevalence than their male counterparts.
With regard to cataract surgery, persons of Mexican ancestry
had higher prevalence than blacks or whites in each age group
for both women and men.
Studies in the Punjab (India), Tibet and Saudi Arabia, although
not recent, conrm the importance of age in the prevalence of
cataract in these populations.79 However, it is difcult to compare
prevalence between them and to compare those data estimates
from the Eye Disease Prevalence Research Group studies. It is
likely that in these countries the burden of blindness is largely
attributable to cataract, as is true in Andhra Pradesh (India)
(44% of cases of blindness)10 and Nigeria (44% of blindness)11
but may reflect the decreased ability to assess the presence of
other sight-limiting conditions.
TYPES OF AGE-RELATED CATARACTS

There are three common types of age-related cataract: nuclear,
cortical, and posterior subcapsular. There is ample reason to
believe that their causes, environmental and genetic, differ both
qualitatively and quantitatively.1218 Therefore, identifying the factors that influence the risk of a cataract severe enough to require
surgery is likely to be quite complicated although very important.
Data by cataract type are available for the Barbados Eyes Study,
the Beaver Dam Eye Study, the Blue Mountains Eye Study, the
Salisbury Eye Evaluation project (SEE), and the Melbourne Visual
Impairment Survey.19 Nuclear cataract occurred in ~0.20.4%
of whites 4049 years of age and in ~4450% in those who were
7579 years of age. The data for blacks are limited, but prevalence
appears to be lower at each age than in whites. Cortical cataract
is also relatively common, and it appears to be more frequent in
389
blacks than whites. Posterior subcapsular cataract (PSC) is the
least common, but its prevalence is also related to age. Racial/
ethnic differences are uncertain.
RISK FACTORS FOR CATARACT

FORMATION
DIABETES
The role of diabetes in the development of cataract has been
controversial. In the past, some investigators thought that
cataracts did not occur more often in diabetics but rather were
simply diagnosed more often because of the increased frequency
of visual examinations in this group. Diabetics may also be
more likely to undergo cataract extraction because of the need
to visualize the retina clearly in order to monitor the development and progression of retinopathy. Thus, it is important to
evaluate the association of diabetes and cataract in populationbased surveys and not by studies of either persons undergoing
screening or cataract extraction. The Framingham Eye Study
and the National Health and Nutrition Examination Survey
(NHANES) found a three- to fourfold excess risk of cataract
among diabetics less than 65 years of age.18 Because of the relatively strong association indicated by these data (and the biologic
plausibility of the association), it is now generally accepted that
diabetics have a higher risk of cataract, but it is not clear why.
Not all cataracts are uniformly increased in diabetics.20 Cortical
cataracts are the type most often associated with it20 but are the
least likely to lead to cataract extraction.21
SECTION 5
SMOKING
Most epidemiologic studies have noted an increased risk of cataract
among smokers. In a cross-sectional study, Klein and associates22
assessed smoking and cataract among diabetics and found a
positive association in those who were diagnosed with diabetes
after age 30. Data from many other studies of various designs
have found an increase in the risk of nuclear cataract to cigarette
smoking. Flaye and co-workers23 found that nuclear cataracts were
2.5 times as common among current smokers compared with
nonsmokers. Among ex-smokers, cataracts were more common in
those who had smoked heavily, whereas no increase in risk was
noted in past light or moderate smokers. In a case-control study,
current smokers, dened as those who had smoked at least one
cigarette per day for at least 1 year and still smoked, had a 70%
increased risk of nuclear cataract.24 Among 21 316 US male
physicians,25 current smokers of 20 or more cigarettes per day
had a twofold increase in cataract risk relative to nonsmokers.
In case-control studies conducted in India,26 Italy,27 and
Maryland,28 no association between smoking and cataract was
reported. In general, positive results are most consistent for a
relationship between smoking and nuclear cataract. An increase
in risk has not been noted for cortical cataract and, except for the
reports by Bochow et al,28 Christen et al,25 and Hankinson et al,29
most studies have had a limited ability to assess the effect of
smoking on PSC cataract. Although ndings have not been
entirely consistent, smoking does appear to be one of the best
conrmed risk factors for (nuclear) cataract.
One hypothesis to explain the risk is that smoking increases
oxidative stress in the lens, however, direct evidence to support
this hypothesis is lacking.
SUNLIGHT EXPOSURE
390
Sun exposure is known to be damaging to a number of tissues

in the eye. Both cortical and PSC cataract have been induced by
UV irradiation in animal studies.30 UV radiation could increase

the risk of cataract through disruption of the membranecation
transport system or injury to nucleic acids in the epithelial cells
of the lens.30
Studies have been conducted to assess the association between
sun exposure and cataract formation in persons in geographically
dened areas. The advantages of such studies are that they can
be performed quickly, at low cost, and may generate important
hypotheses concerning exposuredisease relationships. However,
it is not possible to know whether persons developing the
disease of interest were those exposed and, even if exposed, at
what level of exposure. In addition, data are often lacking on
other potentially important variables that might serve to alter
the relationship between exposure and disease.
In a study conducted in Australia by Hollows and Moran,31
the cataract status of 64 307 aborigines and 41 254 nonaborigines
was examined. Cataract prevalence was higher in the higher UV
light zones. Another study conducted among 30 565 lifelong residents of Nepal32 found that altitude and cataract were inversely
associated in this study; this was attributed to the blockage of
sunlight at higher altitudes by neighboring mountains. Neither
of these studies was able to control for potentially important
variables, such as smoking status and diet.
Data from the NHANES were used to assess the association
between cataract and annual sunlight exposure. The prevalence
of cataract increased with increasing exposure. In another
assessment of sun exposure using these data, UV radiation was
estimated33 at each site using data on latitude, elevation, and
cloud cover. An increase in cataract prevalence was associated
with increase in UV-B exposure. For example, controlling for
age, education, diabetes, race, and urban/rural residence, the
prevalence of cataract for those exposed to UV-B at a level
similar to that in Tucson, Arizona, was 58% higher than for
those exposed to UV-B levels similar to those in Albany,
New York.
A detailed study of sun exposure and cataract was conducted
of 838 Chesapeake Bay watermen in Maryland34 to assess sun
exposure since adolescence, the use of eyeglasses and hats,
medical history, smoking, and diet. By incorporating laboratory
data on the effectiveness of eyeglasses and hat use in blocking sun
exposure of the lens and data from UV monitors, the investigators calculated annual and cumulative sun exposure for each
individual. The risk of cortical cataract was 60% greater (risk ratio
(RR) = 1.60; 95% condence interval (CI) = 1.012.64) with a
doubling of cumulative sun exposure. Bochow and colleagues28
examined 160 persons with PSC and 160 controls, matched by
age, sex, and referral pattern. Sun exposure, both annual and
cumulative, was calculated as it was in the Watermen Study,
and as in that study, the positive association between cataract
and sun exposure was statistically signicant.
Recently, West et al modeled the risk of cortical cataract in
the US presumably due to the increase in UV radiation that is
due to stratospheric ozone depletion. Ambient UV exposures were
estimated based on extensive questionnaire data from the SEE
project participants.35 Estimates of exposure for agegenderrace
categories were based on questionnaire responses, and these were
extrapolated to various locations around the country (which are
expected to vary with ozone depletion over the coming years).
Based on these analyses, the authors calculated that there
would be 167 000830 000 additional cases of cortical cataract
by the year 2050. They posit that were these cases to result in
cataract surgery, the costs could be monumental.
The results of animal and ecologic studies, in conjunction with
the strength of the proposed biologic mechanism, all support an
association between sun exposure and cataract, although the
exact nature and strength of the association remains uncertain.
It is also not certain whether UV exposure is associated with only
DIET/SUPPLEMENTS
The possible effects of dietary or supplemental vitamins on
cataract development have been assessed in some epidemiologic
studies, although results have been inconclusive.
The Age Related Eye Disease Study (AREDS), a randomized
controlled clinical trial designed to determine whether antioxidant vitamins and/or zinc decreased the risk of progression
of age-related macular degeneration (AMD), also evaluated
incidence and progression of age-related cataracts. There was no
evidence of a benecial effect of the study preparations on any
of the three age-related cataracts.36 However, in subsequent
analyses using the propensity score, which adjusts for factors
associated with the use of multivitamin there was a benecial
protective effect such as on the development of cataracts,
especially for the nuclear type.36a In the Beaver Dam Eye Study,
past use of multivitamins was protective for severity of
prevalent nuclear sclerosis.37
A combined antioxidant nutrient score has been reported to be
inversely associated with cataract.24,38 Further data on specic
antioxidants are needed before any public health recommendations can be made.
Some studies focus on individual vitamin supplements,
although ignoring intake from food sources may underestimate exposure. Even when high in particular vitamins,
foods contain other nutrients and structural components that
influence absorption and availability of any particular vitamin
or mineral.
Levels of vitamin C are reduced in the cataractous lens39 and
levels are reportedly increased with vitamin C supplementation.40
Leske and colleagues24 assessed dietary intake of vitamin C in a
large study with 945 case-patients and 435 controls and noted
a decreased risk of nuclear cataract among those in the top 20% of
intake. The NHANES study showed no association with cataract prevalence when either vitamin C intake (calculated from a
24-h dietary recall) or usual frequency of fruit consumption was
assessed.41
Vitamin E (tocopherol), a fat-soluble antioxidant, breaks the
chain reaction of lipid peroxidase formation in cell membranes
and may help maintain the integrity of cell membranes in the
lens.42 One case-control study24 revealed a statistically signicant
decrease in cortical and mixed cataract among persons in the
highest quintile of vitamin E intake and found an inverse association between plasma vitamin E and nuclear cataract.43 An
inverse association was also reported in the Baltimore
Longitudinal Study of Aging.44
Several studies have assessed the association between either
carotene or retinol and cataract. The Baltimore Longitudinal
Study on Aging reported no substantial association with plasma
-carotene and cataract.44 Leske and co-workers24 reported that
total vitamin A was protective for cortical, nuclear, and mixed
cataracts, whereas in the large Italian case-control study,27
retinol intake was not associated with risk of cataract.
Associations with specic foods were not evaluated in any of
these studies. Incidence data from the Beaver Dam Eye Study
cohort gave little evidence of a protective effect of carotenoids in
serum and nuclear cataract.45
Riboflavin is required for the synthesis of flavin adenine
dinucleotide, a cofactor for the antioxidant enzyme glutathione
reductase.46 In several animal species, a deciency in riboflavin
results in cataract formation.47 Leske and associates noted a

40% decrease in the risk of cortical cataract among persons in
the highest quintile of dietary riboflavin intake24 and found an
inverse association of the highest levels of plasma riboflavin with
both nuclear and PSC catarae. A number of other studies26,27,48
have found no association of riboflavin with cataract. In a clinical trial conducted in China among adults with multiple chronic nutrient deciencies, a supplement combining riboflavin
and niacin resulted in a 41% lower prevalence of nuclear
cataract,49 suggesting that cataract risk is increased at or near
deciency levels.
LOW-PROTEIN OR AMINO ACIDDEFICIENT DIETS

Both low-protein diets and specic amino acid deciencies have
been proposed as risk factors for cataract based on animal models.
Two studies, both conducted in India, have examined lowprotein diets as a risk factor for cataract. In a cross-sectional
study, Chatterjee and colleagues7 found that persons with the
lowest reported intake of beans, lentils, meat, milk, eggs, and
curd had a 1.5- to 2.5-fold increased risk of cataract when
controlled for age, caste, marital status, education, and weight.
In a case-control study, Mohan et al26 collected information on
usual monthly consumption of foods containing protein, thiamine,
riboflavin, vitamin A, ascorbic acid, vitamin E, and calcium.
Unfortunately, it was not possible to discern whether the
nutrient associated with risk of cataract was protein or another
dietary constituent.
ALCOHOL INTAKE
A positive association between alcohol consumption and cataract, particularly PSC, has been reported.50,51
ASPIRIN USE
A possible relation between aspirin use and decreased risk of
cataracts has been reported.52 In another case-control study, the
odds ratio associated with aspirin use was 0.25 (95% CI =
0.100.66).53 In contrast, Klein and colleagues22 assessed aspirin
use among 1370 diabetic patients and found no association
with cataract. This was conrmed in a randomized controlled clinical trial of aspirin for patients with diabetes.53a Other studies in
persons not selected by diabetes status have failed to nd any
association of aspirin and cataracts.54,55
CHAPTER 35
cortical cataract or whether PSC is also increased with UV

exposure. Difculties in quantifying exposure (i.e., collecting data
on time spent outdoors in the sun, the level of UV radiation in
specic locales, and the use of eyeglasses and hats, and in
sorting out possible ethnic differences in susceptibility) make
such studies complex.
POSTMENOPAUSAL HORMONE USE

Two recent cross-sectional studies have noted an inverse relationship between current postmenopausal hormone use and cataract. However, the studies were somewhat inconsistent in that
the inverse association was noted for nuclear cataract in one56
and cortical cataract in the other.57
SEVERE DIARRHEA
It has been proposed that the dehydration and uremia
associated with severe diarrhea could increase the levels
of cyanate in the body and that cyanate-associated
carbamylation of lens proteins would result in cataract
formation. To date, few studies have addressed this association.
In a matched case-control study conducted in India, casepatients were four times as likely as controls to have reported at
least one severe episode of diarrhea; they were 21 times more
likely to have reported two or more bouts of diarrhea (95% CI =
8.931.0).58
391
analyses by Klein et al,62 genetic effects may be obscured or
modied by environmental factors.
GENETIC FACTORS
There have been recent reviews in the literature describing sites
in the genome related to congenital or early-onset cataract.59,60
A study of twins61 indicated greater concordance of nuclear and
cortical in monozygotic compared to dizygotic twins. In the
population-based Beaver Dam Eye Study, Heiba et al and Klein
et al described evidence of familial effects on nuclear15,62 and
cortical cataracts.16 Iyengar et al discovered multiple loci associated with cortical cataract in this population. Judging from the
CONCLUSION
Age-related cataracts are common throughout the world. There
are many factors environmental, personal, and genetic that
appear to influence their frequencies. Further research, both experimental and observational, is needed to develop interventions
that will have an impact on their prevalence.
SECTION 5
REFERENCES
392
1. Thylefors B, Negrel AD, Pararajasegaram R,

Dadzie KY: Global data on blindness. Bull
World Health Organ 1995; 73:115121.
2. Rahmani B, Tielsch JM, Katz J, et al: The
cause-specic prevalence of visual
impairment in an urban population. The
Baltimore Eye Survey. Ophthalmology
1996; 103:17211726.
3. The Eye Disease Prevalence Research
Group: Causes and prevalence of visual
impairment among adults in the United
States. Arch Ophthalmol 2004; 122:477485.
Group: Prevalence of cataract and
pseudophakia/aphakia among adults in the
United States. Arch Ophthalmol 2004;
122:487494.
5. Report of the Cataract Panel. Vision
research: a national plan, 19831987. US
Department of Health and Human Services.
NIH Publication Number 832473; 1983.
6. Klein BE, Klein R, Linton KL, et al:
Assessment of cataracts from photographs
in the Beaver Dam Eye Study.
7. Chatterjee A, Milton RC, Thyle S:
Prevalence and aetiology of cataract in
Punjab. Br J Ophthalmol 1982; 66:3542.
8. Hu TS, Zhen Q, Sperduto RD, et al:
Age-related cataract in the Tibet Eye Study.
9. Tabbara KF, Ross-Degnan D: Blindness
in Saudi Arabia. JAMA 1986;
255:33783384.
10. Dandona L, Dandona R, Srinivas M, et al:
Blindness in the Indian state of Andhra
Pradesh. Invest Ophthalmol Vis Sci 2001;
42:908916.
11. Johnson JG, Goode SV, Faal H: Barriers to
the uptake of cataract surgery. Trop Doct
1998; 28:218220.
12. Klein BE, Klein R, Linton KL, Franke T:
Cigarette smoking and lens opacities: the
Beaver Dam Eye Study. Am J Prev Med
1993; 9:2730.
13. Cruickshanks KJ, Klein BE, Klein R:
Ultraviolet light exposure and lens
opacities: the Beaver Dam Eye Study. Am J
Public Health 1992; 82:16581662.
14. Klein BE, Klein R, Jensen SC, Linton KL:
Hypertension and lens opacities from the
Beaver Dam Eye Study. Am J Ophthalmol
1995; 119:640646.
15. Heiba IM, Elston RC, Klein BE, Klein R:
Genetic etiology of nuclear cataract:
evidence for a major gene. Am J Med
Genet 1993; 47:12081214.
16. Heiba IM, Elston RC, Klein BE, Klein R:
Evidence for a major gene for cortical
cataract. Invest Ophthalmol Vis Sci 1995;
36:227235.
17. Iyengar SK, Klein BE, Klein R, et al:

Identication of a major locus for age-related
cortical cataract on chromosome 6p12q12
in the Beaver Dam Eye Study. Proc Natl
Acad Sci USA 2004; 101:1448514490.
18. Ederer F, Hiller R, Taylor HR: Senile lens
changes and diabetes in two population
studies. Am J Ophthalmol 1981; 91:381395.
19. Johnson GJ, Weale RA, Minassian DC,
West SK: The epidemiology of eye disease.
2nd edn. London: Arnold; 2003.
20. Klein BE, Klein R, Wang Q, Moss SE:
Older-onset diabetes and lens opacities.
The Beaver Dam Eye Study. Ophthalmic
Epidemiol 1995; 2:4955.
21. Klein BE, Klein R, Moss SE: Incident
cataract surgery: the Beaver Dam Eye
22. Klein BE, Klein R: Cataracts and macular
degeneration in older Americans. Arch
Ophthalmol 1982; 100:571573.
23. Flaye DE, Sullivan KN, Cullinan TR, et al:
Cataracts and cigarette smoking. The City
Eye Study. Eye 1989; 3(Pt 4):379384.
24. Leske MC, Chylack LT Jr, Wu SY: The Lens
Opacities Case-Control Study. Risk factors
for cataract. Arch Ophthalmol 1991;
109:244251.
25. Christen WG, Manson JE, Seddon JM, et
al: A prospective study of cigarette
smoking and risk of cataract in men. JAMA
1992; 268:989993.
26. Mohan M, Sperduto RD, Angra SK, et al:
India-US case-control study of age-related
cataracts. India-US Case-Control Study
Group. Arch Ophthalmol 1989;
107:670676.
27. The Italian-American Cataract Study
Group: Risk factors for age-related cortical,
nuclear, and posterior subcapsular
cataracts. Am J Epidemiol 1991;
133:541553.
28. Bochow TW, West SK, Azar A, et al:
Ultraviolet light exposure and risk of
posterior subcapsular cataracts. Arch
Ophthalmol 1989; 107:369372.
29. Hankinson SE, Willett WC, Colditz GA, et
al: A prospective study of cigarette
smoking and risk of cataract surgery in
women. JAMA 1992; 268:994998.
30. Taylor HR: Ultraviolet radiation and the eye:
an epidemiologic study. Trans Am
31. Hollows F, Moran D: Cataract the
ultraviolet risk factor. Lancet 1981;
2:12491250.
32. Brilliant LB, Grasset NC, Pokhrel RP, et al:
Associations among cataract prevalence,
sunlight hours, and altitude in the
Himalayas. Am J Epidemiol 1983;
118:250264.
33. Hiller R, Sperduto RD, Ederer F:

Epidemiologic associations with cataract in
the 19711972 National Health and
Nutrition Examination Survey. Am J
Epidemiol 1983; 118:239249.
34. Taylor HR, West SK, Rosenthal FS, et al:
Effect of ultraviolet radiation on cataract
formation. N Engl J Med 1988;
319:14291433.
35. West SK, Longstreth JD, Munoz BE, et al:
Model of risk of cortical cataract in the US
population with exposure to increased
ultraviolet radiation due to stratospheric
ozone depletion. Am J Epidemiol 2005;
162:10801088.
36. Age-Related Eye Disease Study Research
with vitamins C and E and beta carotene
for age-related cataract and vision loss:
AREDS report no. 9. Arch Ophthalmol
2001; 119:1439-1452.
36a. Clemons TE, Kurinij N, Sparduto RD,
Bressler SB: for the Age-Related Eye
Disease Study Research Group. A
randomized, placebo-controlled, clinical
trial of high-dose supplementation with
vitamins C and E and beta carotene for
age-related cataract and vision loss:
AREDS report no. 9. Arch Ophthalmol
2001; 119:14391452.
37. Mares-Perlman JA, Klein BE, Klein R, Ritter
LL: Relation between lens opacities and
vitamin and mineral supplement use.
38. Jacques PF, Chylack LT Jr, McGandy RB,
Hartz SC: Antioxidant status in persons
with and without senile cataract. Arch
Ophthalmol 1988; 106:337340.
39. Taylor A: Associations between nutrition
and cataract. Nutr Rev 1989; 47:225234.
40. Taylor A, Jacques PF, Nadler D, et al:
Relationship in humans between ascorbic
acid consumption and levels of total and
reduced ascorbic acid in lens, aqueous
humor, and plasma. Curr Eye Res 1991;
10:751759.
41. Goldberg J, Flowerdew G, Smith E, et al:
Age-related macular degeneration and
cataract: are dietary antioxidants protective?
Am J Epidemiol 1988; 128:904905.
42. Bunce GE, Hess JL: Cataract what is the
role of nutrition in lens health? Nutr Today
1988; 23:612.
43. Leske MC, Wu SY, Hyman L, et al:
Biochemical factors in the Lens Opacities
Case-Control Study Group. Arch
Ophthalmol 1995; 113:11131119.
44. Vitale S, West S, Hallfrisch J, et al: Plasma
antioxidants and risk of cortical and nuclear
cataract. Epidemiology 1993; 4:195203.
52.
53.
53a.
54.
55.
56.
ethanol (U-curve) and non-signicance of

smoking. Ophthalmic Res 1996; 28:237247.
Cotlier E, Sharma YR: Aspirin and senile
cataracts in rheumatoid arthritis. Lancet
1981; 1:338339.
Chen TT, Hockwin O, Dobbs R, et al:
Cataract and health status: a case-control
study. Ophthalmic Res 1988; 20:19.
Early Treatment Diabetic Retinopathy Study
Research Group: Effects of aspirin
treatment on diabetic retinopathy. ETDRS
report number 8. Ophthalmology 1991;
98:757765.
Walker AM, Jick H, Gorman MR, Wallach
RW: Steroids, diabetes, analgesics, and the
risk of cataract: lessons from the
epidemiology of cataract extraction. J Clin
Res Drug Dev 1988; 2:227232.
Peto R, Gray R, Collins R, et al:
Randomised trial of prophylactic daily
aspirin in British male doctors. Br Med J
(Clin Res Ed) 1988; 296:313316.
Klein BE, Klein R, Ritter LL: Is there
evidence of an estrogen effect on agerelated lens opacities? The Beaver Dam Eye
Study. Arch Ophthalmol 1994; 112:8591.
57. Cumming RG, Mitchell P: Hormone

replacement therapy, reproductive factors,
and cataract. The Blue Mountains Eye
Study. Am J Epidemiol 1997; 145:242249.
58. Minassian DC, Mehra V, Jones BR:
Dehydrational crises from severe diarrhoea
or heatstroke and risk of cataract. Lancet
1984; 1:751753.
59. Graw J, Loster J: Developmental genetics
in ophthalmology. Ophthalmic Genet 2003;
24:133.
60. Hejtmancik JF, Smaoui N: Molecular
genetics of cataract. Dev Ophthalmol 2003;
37:6782.
61. Hammond CJ, Snieder H, Spector TD,
Gilbert CE: Genetic and environmental
factors in age-related nuclear cataracts in
monozygotic and dizygotic twins. N Engl J
Med 2000; 342:17861790.
62. Klein AP, Duggal P, Lee KE, et al: Polygenic
effects and cigarette smoking account for a
portion of the familial aggregation of
nuclear sclerosis. Am J Epidemiol 2005;
161:707713.
CHAPTER 35
45. Lyle BJ, Mares-Perlman JA, Klein BE, et al:

Serum carotenoids and tocopherols and
incidence of age-related nuclear cataract.
Am J Clin Nutr 1999; 69:272277.
46. Draper HH: Nutritional modulation of
oxygen radical pathology. Adv Nutr Res
1990; 8:119145.
47. Bunce GE, Kinoshita J, Horwitz J:
Nutritional factors in cataract. Annu Rev
Nutr 1990; 10:233254.
48. Jacques PF, Hartz SC, Chylack LT Jr, et al:
Nutritional status in persons with and
without senile cataract: blood vitamin and
mineral levels. Am J Clin Nutr 1988;
48:152158.
49. Sperduto RD, Hu TS, Milton RC, et al: The
Linxian cataract studies. Two nutrition
intervention trials. Arch Ophthalmol 1993;
111:12461253.
50. Munoz B, Tajchman U, Bochow T, West S:
Alcohol use and risk of posterior
subcapsular opacities. Arch Ophthalmol
1993; 111:110112.
51. Phillips CI, Clayton RM, Cuthbert J, et al:
Human cataract risk factors: signicance of
abstention from, and high consumption of,
393
CHAPTER
36
Epidemiology of Primary Open-Angle Glaucoma

Anne L. Coleman, Steven L. Mansberger, and M. Roy Wilson
The ultimate goal in the management of a chronic disease such

as primary open-angle glaucoma (POAG) is to design intervention programs that can prevent, or at least control, the debilitating outcomes associated with the disease. The chapters on
glaucoma in this book are devoted to providing a better understanding of the pathophysiologic mechanisms of disease, as well
as treatment strategies. This chapter uses epidemiology to
describe the prevalence, incidence, risk factor, treatment, and
screening of POAG.
Epidemiology studies are directed to nding the distribution,
determinants, and frequency of disease in groups of persons in
order to improve our understanding of prevalence, incidence,
pathogenesis, and treatment. Implicit in the denition of epidemiology is the fact that disease is not randomly distributed
throughout a population; instead, the frequency differs among
subgroups. Knowledge of this uneven distribution, and of the
factors that influence this distribution, may provide valuable
clues as to what factors are important in pathogenesis and
development of glaucoma.
POAG AS A PUBLIC HEALTH PROBLEM

A recent meta-analysis estimated the overall prevalence of
POAG in the United States to be 2.2 million persons based on
2000 US census data.1 Worldwide, 48 million persons2,3 are
thought to have POAG. This high prevalence is likely to increase
in the future because of the aging population. In the year 2000,
people aged 65 and older made up 12.4% of the US population.
By 2040, this group will increase to 20.4% and comprise one
fth of the population. During the same period, the 3 million
Americans aged over 85 are expected to triple to 9.8 million.4
Overall, this change in population demographics will increase
the prevalence of glaucoma in the United States by 50% to
3.4 million persons by the year 2020.1 A similar demographic
change will occur worldwide and will increase the prevalence of
glaucoma by 100%.2
The real public health impact of POAG in the US is the visual
limitation it causes. Studies estimate that approximately 0.37%
of US adults over the age of 40 years are bilaterally blind, which
corresponds to over 400 000 (119 million 0.37%) persons. This
figure increased to almost 5 million globally.5 Glaucoma is the
second leading cause of blindness.3
The societal costs are staggering. The estimated expenditures
for treatment are at least $1.6 billion. The federal government
provides $1.05 billion per year in income support to assist
persons blind from glaucoma (e.g., Social Security disability
income, automatic Medicare and Medicaid eligibility, and
income tax credits).6 Additional costs which are difcult to
estimate, include lost earnings and requirements for care taking
services. A recent clinical trial7 evaluating the quality of life in
persons undergoing initial surgical versus medical treatment for

glaucoma reported that more than 50% of newly diagnosed
glaucoma patients were worried about going blind at the start
of the study. This percentage decreased to ~25% at 5 years
of follow-up. Females and older subjects had more functional disabilities related to activities of daily living during
the study, and subjects randomized to surgery had more local
eye symptom complaints than subjects randomized to medications. Several investigators have found that even mild visualeld loss is associated with decreased vision-related quality
of life.8,9
PREVALENCE
Prevalence of disease is one of the cornerstones to epidemiologybased knowledge and control programs. Many studies report the
prevalence of glaucoma in a variety of ethnic groups and regions;
however, the methods and denitions vary. The prevalence of a
disease would best be estimated: (1) on a well-dened population,
(2) by examining and reporting on all of the dened population
or a specied sample of the dened population, (3) if sampling
is used, sampled subjects should represent the population, with
no subgroup systematically excluded from examination, and
(4) by specifying and consistently applying the case denitions
for glaucoma. Studies based on self-selected or small nonrepresentative segments of the population are particularly susceptible
to bias. A population-based study design is preferable. Yet, even
among studies using this design, methodological shortcomings
are often present, and study results must be compared with
caution.
Major differences exist in the case denitions of glaucoma in
prevalence studies. Studies have used elevated intraocular pressure (IOP), optic nerve pathology, and/or visual-eld defects to
dene glaucoma. More recent studies have excluded a specic
level of IOP as part of the denition of glaucoma but have
uniformly required the presence of a glaucomatous-appearing
optic disk and/or visual-eld changes.
However, visual-eld testing creates a myriad of diagnostic
dilemmas when performed as part of a prevalence study, especially when the majority of participants are novice to perimetry.
For example, a common nding in participants new to perimetry is an abnormal or unreliable test result in the presence of
a normal optic disk.10,11 In such cases, should the study protocol require repeat visual-eld testing or should repeat testing be
performed only in patients with an abnormal optic disk? Some
subjects are unable to perform visual-eld testing reliably, but
have an obviously glaucomatous disk. Should these subjects be
classied denitively as having glaucoma? Partly because of
these dilemmas, researchers have developed a denition of glaucoma to be used for prevalence studies based on cup-to-disk
395
SECTION 5
396
ratio (C/D) and visual-eld results.12 This denition includes

three categories for glaucoma: category 1 includes patients with
glaucomatous optic neuropathy and conrmed visual-eld loss;
category 2 includes participants with a C/D greater than 97.5
percentile of the population, but who are unable to perform
visual-eld testing satisfactorily; and category 3 includes participants with a history of glaucoma surgery or extremely elevated
IOP, but the examiner is unable to view the optic disk. This
denition may lead to a more uniform denition of the prevalence of glaucoma in the future that will allow researchers to
more easily compare studies.
An example of the difculty in dening glaucoma in prevalence studies is highlighted in the different prevalence rates in
the surveys performed in Ferndale, Wales (0.47%),13 Dalby,
Sweden (0.86%),14 and Framingham, Massachusetts (1.6%).15
Why did these populations have an almost fourfold difference in
prevalence? One reason may be the difference in the case
denition for glaucoma. The diagnostic criteria for POAG in the
two European studies were based on abnormal disk cupping and
loosely dened visual-eld defects. Thus, these studies may
have excluded subjects with high IOPs and eld defects only.
Conversely, the Framingham study relied solely on rigorously
dened visual-eld criteria and did not require subjects who
were not suspected of having glaucoma (such as those with
normal IOP), to undergo perimetric examinations. Overall,
these methodological differences would make it difcult to
compare study results.
Over the last 30 years, investigators have performed population-based prevalence studies in most regions of the world and
in most ethnic populations, which add to our understanding
of primary open-angle glaucoma. Quigley and Broman5
summarized the worldwide prevalence of glaucoma. They used
generalized estimating equations to account for different sample
sizes and age distributions, which allowed them to collate results
from 34 different studies into separate regions that were expected
to have similar glaucoma prevalence. The eight regions
included: European, Middle East/North Africa, Latin American,
African (south of the Sahara), South East Asia, Indian group,
China group, and Japan. They included studies with random
population-based sampling, high proportion of examinations
(>50%), high proportion of visual-eld testing (>50%), optic
disk evaluation by an ophthalmologist, and denition of glaucoma based on optic disk and visual-eld criteria. The proportion of open-angle glaucoma in each region from highest to
lowest was: African (4.2%), Japan (3.3%), Latin America (3.2%),
Europe (2.0%), India (1.8%), China (1.4%), Middle East (1.3%),
and SE Asia (1.2%). The overall prevalence of POAG worldwide
was estimated to be 2.0%. These results provide valuable
information for forecasting the burden of OAG in these groups.
The authors also estimate the prevalence of angle closure glaucoma, which is the cause of a high proportion of glaucoma in
several regions,1618 and is described further in other chapters.
Some prevalence studies describe intra-regional and intraethnic variations in the prevalence of open-angle glaucoma
especially in Asia, Africa, and Latin America. For example, the
Tajimi Eye Study in Japan19,20 found 7:1 ratio of POAG
(including normal tension glaucoma) to primary angle-closure
glaucoma (PACG). However, PACG was more common in
participants of Chinese descent with a ratio of POAG:PACG of
1.6:1 in Chinese living in Singapore16; and a 1:3 ratio in
Mongolia, China.18 Similar discrepancies can be found in the
population-based studies among Africans and Latin Americans.
Data from the West Indies11,21 and from Baltimore22 suggest a
very high POAG prevalence for Africans, but almost a twofold
difference in prevalence between the two studies with a
prevalence of 8% and 4%, respectively. Recent studies suggest
similar regional differences in Ghana23 (8.5%) when compared
to Tanzania24 (3.1%) and South Africa25 (2.1%). Finally, the

Proyecto VER population-based study in Arizona26and the Los
Angeles Latino Eye Study (LALES)27 show differences in the
prevalence of POAG of 2.0% and 4.7%, respectively. These
results underscore the fact that regions and ethnic groups may
vary greatly in the prevalence of POAG. While many of these
differences may come from genetic and environmental influences,
some may be the result of different denitions of glaucoma and
varied age stratication within the sampled population.
A population-based prevalence survey among Alaskas
Northwestern Inuit found a glaucoma prevalence of 0.65%, but
most were of the angle-closure variety.28 This study suffered
from the fact that the diagnosis of open-angle glaucoma was
based on visual-eld defects, with the tangent screen in the
presence of either an elevated IOP (>21 mmHg) or a C/D of
more than 0.5. Because a substantial proportion of persons with
glaucoma present with normal IOPs and may have C/D of less
than 0.5, the reported prevalence of POAG (0.06%) was
undoubtedly underestimated. Nonetheless, it is probably safe to
conclude that the prevalence of POAG in this population is very
low. Two other studies have been reported recently in American
Indians. They show a prevalence of glaucoma of 5.6% in
Oklahoma Indians29 and a prevalence of 6.2% in Northwest
American Indians.30 The latter study is noteworthy in that 90%
of the glaucoma were normal-tension glaucoma with IOP
<21 mmHg and no cases of angle closure glaucoma were
detected. Overall, these ndings in mainland American Indians
were similar to those found in the Japanese and different from
the ndings in Alaskan Inuit.
Other interesting results from recent prevalence studies
improve our understanding of POAG. The Tajimi study19
(Japan) showed that 92% of participants with POAG had IOP
<21 mmHg and showed an age-associated decrease in IOP.
Similar age decreases in IOP were found in a study in Ireland.31
Pseudoexfoliation, which is known to be a common cause of
open-angle glaucoma in Northern Europe and Greece, has also
been found to be a signicant cause of open-angle glaucoma in
India32 and Africa.25,33 Finally, most prevalence studies27,34,35
show a high prevalence of undiagnosed glaucoma, from 50% to
as high as 93%.36 Overall, these studies underscore the importance of prevalence studies, which has generated considerable
research interest in morphologic, genetic, and environmental
influences into the causes of glaucoma.
INCIDENCE
Prevalence data are valuable, but they do not provide estimates
of disease over time, nor do they permit etiologic inference.
Incidence data, on the other hand, provide this information.
They give a direct measure of the rate at which individuals in a
given population develop disease and the probability of risk of
the disease. Despite their desirability, reliable POAG incidence
data are scarce.
One can divide studies measuring open-angle glaucoma incidence into two types: (1) those that are population-based and
(2) those that target a specic high-risk subpopulation. Because
of the relatively low incidence, large cohorts and long follow-up
periods are necessary to obtain a sufcient number of newly
diagnosed glaucoma cases to ensure valid estimates; thus only
a few such studies have been conducted.
Several population-based studies were designed to yield incidence data. A 9-year incidence of 4.4% (95% CI: 3.75.2) was
reported for a black West Indian population.21 Although issues
related to study design preclude making a direct comparison, a
glaucoma annual incidence of between 0.19% and 0.24% was
reported in the white population of Dalby, Sweden.37 In the
Melbourne Visual Impairment Project,38 there was a 1.1%
5-year incidence of OAG, while in Olmsted County, Minnesota 39

the annual age-adjusted incidence of OAG was estimated to be
0.0145% in a predominantly Caucasian population. The
incidence estimate from Olmsted County was most likely an
underestimate because it was based on a review of medical
records while the other population-based studies involved the
reexamination of study subjects. The Rotterdam study reported
a 5-year incidence in glaucoma of 0.6% in participants greater
than 55 years old.
Because of this low incidence, various investigators have
attempted to determine incidence rates in selected high-risk
subpopulations. The Collaborative Glaucoma Study,40 which
was a prospective study conducted over a 13-year period to
identify factors that influence the development of glaucomatous
visual-eld defects, enrolled relatives of patients with openangle glaucoma. This was augmented with a group of persons
with IOPs greater than 20 mmHg. Glaucomatous visual-eld
defects developed in 1.7% of the 5886 eyes included in the
analysis, over a maximal time period of 7 years. The annual
incidence rates ranged from 0.25% to 0.54%. Using life table
analyses for different levels of IOP, for the 5-year survival (free
of visual-eld defects) rate for eyes with IOPs greater than or
equal to 20 mmHg was 93.3%, whereas it was 98.5% for eyes
with lower IOPs.
Other studies followed cohorts of subjects with higher-thannormal IOP for variable time periods.4243 Unfortunately, these
studies do not permit the calculation of incidence rates and
comparison of the results of these studies is difcult because of
different inclusion criteria and diagnostic criteria. Despite these
caveats, the studies had two major ndings: (1) visual-eld
defects develop infrequently, even in subjects with higher-thannormal IOP; and (2) the higher the baseline IOP, the greater the
risk of subsequently developing visual-eld defects. In the
Ocular Hypertension Treatment Study (OHTS) the 5-year incidence of developing glaucoma was 9.5% in ocular hypertensive
subjects who were randomized to no treatment.44
The lack of good population-based incidence data has
prompted the derivation of incidence estimates from age-specic
prevalence data.45 Such estimates are regarded as gross approximations, and their use is usually restricted to certain specic
purposes such as the planning of epidemiologic studies.
RISK FACTORS FOR DISEASE

DEVELOPMENT
IOP is the most important known risk factor for glaucoma
development. Evidence clearly indicates that increased IOP can
cause glaucoma. Experimentally induced high IOP in animals
results in typical glaucomatous cupping.46,47Acute angle closure
and many cases of unilateral high IOP glaucoma support a
causeeffect relationship between high IOP and glaucomatous
damage. Even at normal IOP levels, asymmetric IOP has been
noted to correlate with asymmetric cupping and eld loss, with
greater damage occurring on the side with higher pressure.48,49
Population surveys provide additional support that there is
an increase in the prevalence of POAG with increasing levels of
IOP.13,14,48 However, many subjects with elevated IOP do not
have glaucoma, and longitudinal studies of subjects with elevated IOP have demonstrated that most patients with ocular
hypertension never develop glaucoma. Thus, elevated IOP is
frequently not sufcient and is, in fact, not a necessary condition for glaucomatous damage. Population surveys have consistently demonstrated that 3090% of subjects (depending on
ethnicity) did not have elevated IOP at the time of
diagnosis.14,15,22,31,50 Even if researchers could perform multiple
IOP measurements, they would nd a signicant proportion of
subjects with glaucoma and statistically normal IOP. The fact
that some eyes with high IOP do not develop glaucomatous

damage and some eyes with low IOP suffer denite glaucomatous damage indicates that other factors may contribute to
the pathogenesis of glaucoma.
Data regarding the possible role of myopia are conflicting.
Although a number of studies have demonstrated an association between myopia and POAG,51,52 these studies were clinicbased studies (rather than population-based), and the potential
for selection bias must be considered, because persons with
refractive errors are more likely to seek eye care and have a
higher probability of being diagnosed with glaucoma. In the
BMES, there was a threefold increased risk of POAG in individuals with myopia of 3.0 D or greater.53 This increased risk
was independent of IOP and other glaucoma risk factors.
Disks with large C/D also tend to have a larger disk size with
proportionately more neural rim tissue.54 Thus, whether larger
C/D, per se, predispose to glaucomatous damage is unclear. An
enlarged C/D, as well as asymmetric cupping, may be a sign of
early disease. From a practical standpoint, subjects with
suspicious disks must be observed closely for development of
signs of clinically signicant glaucoma.
The transient nature of disk hemorrhages makes it difcult
to assess the importance of this factor for subsequent glaucomatous damage. Fairly consistent evidence indicates a poorer
prognosis in glaucomatous eyes with disk hemorrhages compared with those without hemorrhages.5557 Although disk
hemorrhages have been shown to precede retinal nerve ber
layer defects, glaucomatous changes of the optic nerve head,
and glaucomatous visual-eld defects, it is not known how
frequently this occurs.5861 An extreme view, one that is not well
supported or accepted, proposes that disk hemorrhages precede
all cases of open-angle glaucoma.62 A recent study showed that
the presence of optic disk hemorrhages resulted in mean risk of
visual-eld deterioration over 9 years of 80% (hazard ratio
of 5.4) and 89% (hazard ratio of 3.6) in normal tension and
primary open-angle glaucoma patients, respectively.63 Another
study showed optic disk deterioration but no visual-eld
changes with a history of optic disk hemorrhage.64 Although
clinicians may consider advancing glaucoma therapy in a glaucoma patient with a recent disk hemorrhage, 70% of subjects in
the BMES who had disk hemorrhages had no other signs of
glaucoma.65
Recent studies have demonstrated the association of thin
central corneal thickness (CCT) with existing glaucoma, with
progression to glaucoma from ocular hypertension, and progression of existing glaucoma.6670 In contrast, one randomized
controlled clinical trial did not show an association of CCT
with progressive glaucoma.57 The explanation for why CCT is a
risk factor for glaucoma is not known. The simple explanation
is that it is a surrogate for the known risk of IOP as the actual
IOP is higher than the measured IOP in eyes with thin
CCT.7173 However, the OHTS study showed a strong,
independent association of CCT with development of glaucoma
suggesting that a linear or nonlinear correction of IOP by CCT
would not explain the association of CCT with glaucoma. This
indicates that CCT may correlate with other biomechanical
factors associated with glaucoma such as lamina cribrosa compliance and scleral compliance. One histological study showed
no association between CCT and lamina cribosa thickness;
however, artifacts and sectioning methods may have prevented
accurate measurements.74 Further studies are needed to explore
this relationship.
Race, age, and family history are nonocular factors related to
glaucoma risk. As indicated earlier, blacks, Hispanics, and
American Indians have a disproportionately high prevalence of
POAG when compared to Caucasians.22,26,27,30 Although racespecic incidence data are not yet available for all of these
CHAPTER 36
397
SECTION 5
398
groups. Precisely why certain ethnic groups are more likely to

develop glaucoma is not known. However, some researchers
have found higher IOP, thinner CCT,75 and larger C/D in blacks
when compared to Caucasians, and suggest these factors as
causal. However, data conflict as to whether blacks have higher
IOP than do whites,76,77 and as discussed previously, the
relevance of larger C/D is uncertain.
Nearly every population-based study has demonstrated that
the prevalence of glaucoma increases with advancing
age.11,14,15,21,22,31,50,7882 The oldest age groups have prevalence
estimates approximately three to eight times higher when compared to persons in their 40s. Further, the Collaborative Glaucoma
Study identied age as the major predictor of glaucoma
incidence.40 As with race, the exact causal mechanisms are
unknown, and underlying susceptibility factors must be investigated. The higher IOP noted with increased age in most
studies were not found among the Japanese.78 Yet, glaucoma
prevalence increased with age in the Japanese, suggesting that
the optic nerves of the elderly are more susceptible to damage
for reasons other than just higher IOP.
Familial factors are also important in the underlying susceptibility to POAG. Several ocular parameters associated with
POAG, such as IOP and C/D, are known to be influenced by
heredity.59 Relatives of glaucoma patients would thus be expected
to exhibit abnormalities of these parameters more often and to
more likely be diagnosed as having glaucoma, either from selection bias, or better case nding, or from a true increase in the
prevalence of glaucoma in relatives. Most studies investigating
risk in relatives had selection and recall bias because of incomplete ascertainment of relatives, therefore, accurate estimates of
the exact risk of POAG in relatives are lacking. Other chapters
in this book explain the genetic associations of glaucoma.
Systemic factors provide other information about POAG risk.
Diabetes,79 systemic hypertension,51 and various other vascular
abnormalities such as migraines80 have been implicated as risk
factors for glaucoma. Much of the data regarding possible associations between these factors and glaucoma are contradictory.
The role of diabetes as a risk factor for POAG is controversial.
The Blue Mountain Eye Study found an association between
diabetes and glaucoma;81 while the Baltimore Eye Survey did
not detect an association overall but did nd an association
among persons in whom glaucoma had been diagnosed prior to
the survey examination.82 Persons with diabetes are more likely
to be in the health care system and thus lead to a bias in having
glaucoma detected. Surprisingly, the Ocular Hypertension
Treatment Study found that the presence of diabetes was protective toward developing glaucoma. Overall, the relationship of
diabetes to glaucoma development is controversial.
Although the evidence that systemic hypertension is a risk
factor for glaucoma is not strong, the hypothesis that microcirculatory effects on the optic disk may lead to increased glaucoma susceptibility is biologically plausible. The Rotterdam
Study reported an association of systemic hypertension with
high-tension glaucoma but not with normal-tension glaucoma.83
The Blue Mountains Eye Study investigators reported a 1.5 times
increased risk of open-angle glaucoma in subjects with systemic
hypertension, independent of IOP and other glaucoma risk
factors.84 The Baltimore Eye Survey reported modest, positive
but not statistically signcant associations of increased systolic
and diastolic blood pressure with POAG. However, lower
perfusion pressure (blood pressure IOP) was strongly associated with an increased prevalence of POAG.85 These results
suggest that POAG may be associated with a change in factors
related to ocular blood flow.
Migraine and peripheral vasospasm may be important in the
development of some cases of glaucoma, particularly those in
which IOP is in the normal or low range.86 Investigations of
other associations of other systemic factors with POAG have

been scant, and the results have been inconclusive. One
interesting observation has been that among Japanese, IOP has
not been found to increase with age as it does in Western
populations.87 One explanation for this apparent discrepancy is
that IOP is related to body build, and that Japanese typically do not
get obese with age when compared to Americans and Europeans.
Another interesting nding has been the relationship between lean
body mass and increased prevalence of POAG among the
participants of the Barbados Eye Study.88 These results suggest that
anthromorphologic considerations may warrant further study.
It is unclear whether POAG is more frequently associated
with men or with women. A higher prevalence among women
was reported in Dalby14 and Blue Mountains,89 a higher prevalence among men in Tierp90 Framingham,15 and Barbados,21
and no difference in St Lucia,11 Wales,13 Baltimore,22 Beaver
Dam,50 Melbourne,91 Los Angeles,27and Arizona.92
A variety of risk factors may be present in an ocular hypertension or glaucoma patient, but each patient encompasses a
unique combination of risk factors that the clinician must take
into account.93 Recently, investigators have developed risk
calculators to estimate the risk of developing glaucoma from
ocular hypertension using data from the OHTS study.94,95 The
OHTS and European Glaucoma Prevention Study (EGPS)
demonstrated that age, corneal thickness, IOP, pattern standard
deviation (PSD), diabetes mellitus status, and vertical C/D were
independent predictive variables for the development of
glaucomatous optic disk or visual-eld changes. The OHTS
and EGPS combined their data set to provide more precise
estimates.
Key Features: Reliable Risk Factors for Glaucoma
Increased IOP is the most important risk factor

However, elevated IOP is frequently not sufcient and is, in
fact, not a necessary condition for glaucomatous damage
Increasing age
Racial differences, more common in African-Americans and
other ethnic groups
Family history
Thin CCT
Controversial risk factors for glaucoma include:
Myopia
Presence of disk hemorrhages
Diabetes
Hypertension
TREATMENT ISSUES
Thus far, treatments for POAG have focused exclusively on
lowering IOP. Several trials have documented the efcacy of
medications, laser, and surgery in lowering IOP. However, only
recently, have randomized clinical trials demonstrated the value
of these treatments in reducing the occurrence or progression of
visual-eld damage.
The Ocular Hypertension Treatment study44,66 and European
Glaucoma Prevention Study96 determined the efcacy of ocular
hypotensive treatment in ocular hypertension patients; the
Collaborative Initial Glaucoma Treatment Study97 and the Early
Manifest Glaucoma Treatment study57 examined treatment of
early or newly diagnosed glaucoma patients; the Collaborative
Normal-Tension Glaucoma Study98 (NTGS) examined patients
with mild to moderate glaucoma and normal IOP; and the
Advanced Glaucoma Intervention Study99 investigated surgery
or laser in patients with moderate to severe glaucoma. These
studies guide clinicians in their treatment of glaucoma patients,

loss, or optic disk deterioration in ocular hypertension and glaucoma patients. These studies provide important information
guiding the treatment of ocular hypertension and glaucoma.
SCREENING FOR GLAUCOMA

Glaucoma, as one of the leading causes of blindness, may be a
strong candidate disease for screening programs. It is asymptomatic in the early stages and treatment decreases the risk of
visual-eld loss. Unfortunately, major impediments to widespread glaucoma screening are a lack of a screening test(s) with
appropriate diagnostic precision and lack of evidence that
screening for glaucoma prevents visual impairment. The following section outlines the deciencies and strengths of current
screening tests.
Tonometry has been used as a screening test for glaucoma for
more than 40 years. However, we have yet to identify a set of
tonometric criteria that adequately classify persons in terms of
their disease status. Data from a number of studies have
demonstrated the futility of using the widely accepted cutoff of
21 mmHg for screening purposes.105 Moreover, no matter what
IOP level is chosen, the balance of sensitivity and specicity (or
diagnostic precision) is unacceptable.
Optic disk evaluation has other difculties. Ophthalmoscopy
and optic disk photography are difcult to obtain in many
participants for reasons such as ocular media abnormalities and
difculties with cooperation such as blinking. The Baltimore
Eye Study photographers had difcultly attaining optic disk
photos in over 20% of participants.100 Photography requires
technical expertise to perform and needs expert opinion to grade
the optic disk photos. Finally, experts disagree when grading
optic nerve photos.105
Standard achromatic automated perimetry (SAP) has good
diagnostic precision for glaucoma106,107 but is nonspecic
because abnormal results can occur from other conditions such
as cataracts and retinal disease. Even normal eyes can have
abnormal results from small pupil size,108110 uncorrected
refractive error,111,112 fatigue,113 and learning effects.114118
Abnormalities from uncorrected refractive error are a particular
problem as refractive error are a common source of visual
impairment in the community.119
Overall, the traditional methods of detecting participants at
risk for glaucoma in the community are fraught with difculties
that reduce their feasibility and diagnostic precision. Investigators are researching new methods of screening. Their goals
are to develop a screening program with high diagnostic precision and immediate results, as well as being able to be
performed by paraprofessionals such as ophthalmic technicians
and nurses. These include new methods of visual-eld testing
and examining the optic disk.
Studies have reported results with smaller, faster visual-eld
machines such as frequency doubling technology perimetry,120126
oculokinetic perimetry127 and laptop computer techniques.128
Despite the fact that these techniques have been available for
several years, they still require validation in population-based
screening settings.
Objective structural testing with optic imaging devices such
as confocal scanning laser ophthalmoscopy (CSLO), scanning
laser polarimetry, and ocular coherence tomography has
promise for screening for glaucoma. They are able to image the
optic disk without dilation and in patients with cataract or
other mild media abnormalities. Studies indicate that they have
similar and reasonable diagnostic precision for early glaucoma
when compared to normal subjects.129 However, the machines
are expensive and somewhat difcult to transport. Further
studies will need to determine the diagnostic precision in an
unselected screening population.
CHAPTER 36
examining the full range of glaucoma from preperimetric

glaucoma to advanced glaucoma.
Ocular hypertension is present in ~8% of adults over the age
of 40 years in the United States.100 The Ocular Hypertension
Treatment Study recently demonstrated that treatment of ocular
hypertension decreases the risk of development of visual-eld
loss.44 In contrast, the European Glaucoma Prevention Study
showed no benet of treatment with dorzolamide eye drops
compared with placebo.96 The latter study had the benet of
placebo control, but it suffered from high loss to follow-up
(30%), no target IOP for treatment, low clinical applicability
(only dorzolamide treatment) and only one baseline IOP
measurement. These methodological flaws would tend to
decrease effect of treatment. Despite these different results, most
clinicians recommend treating ocular hypertension patients at
high risk for developing glaucoma.
Recently, randomized controlled trials have demonstrated
that patients with denite glaucoma, regardless of disease stage,
should be treated. The Early Manifest Glaucoma Trial and the
Collaborative Initial Glaucoma Treatment Study evaluated the
treatment of newly diagnosed glaucoma patients. The Early
Manifest Glaucoma Trial (EMGT) randomized patients with early
glaucoma either to argon laser trabeculoplasty plus betaxolol
(n = 129) or to monitoring without immediate treatment
(n = 126).101 These were newly diagnosed glaucoma patients
found during a community glaucoma screening. The rate of
progression was 45% in the treated group versus 62% in the
untreated group. The treatment reduced IOP by ~20% and
decreased the risk of worsening glaucoma by 50%.57 The
Collaborative Initial Glaucoma Treatment Study (CIGTS)
enrolled 607 patients with newly diagnosed open-angle
glaucoma and randomized them to treatment with topical
ocular hypotensive medications or trabeculectomy surgery.102
The 5-year outcomes reported in the CIGTS demonstrated that
both medications and surgery resulted in reduced IOP, and both
groups of patients had similar low rates of visual-eld
progression.97 Only 11% of patients treated medically versus
14% of patients treated with surgery had signicant progression
during follow-up. Medically treated patients were less likely to
develop cataracts, suffer noncataract-related visual acuity loss,
or complain of ocular side effects.
The Collaborative Normal-Tension Glaucoma Study (CNTGS)
evaluated the treatment of moderate to advanced normal tension glaucoma (<20 mmHg) by randomly assigning 240 patients
to treatment versus no treatment. It required patients to have
documented progression or a specic visual-eld defect. The
treatment included medications, laser, or surgery to reduce IOP
by at least 30%.98 The rate of progressive visual-eld loss was
slower in the treated group than in a group that did not receive
treatment when the analysis was adjusted for the effect of
cataracts.
Finally the Advanced Glaucoma Intervention Study (AGIS)
evaluated the treatment of uncontrolled primary open-angle
glaucoma. The study randomly assigned 591 persons to a treatment sequence of argon laser trabeculoplasty, trabeculectomy,
and trabeculectomy (ATT sequence): or trabeculectomy, argon
laser trabeculoplasty, and trabeculectomy (TAT sequence). The
main outcome measures of the study were visual acuity and
visual eld, although the study also evaluated IOP, complications
of treatment, time to treatment failure, and need for adjunctive
medications. The study reported no difference in visual acuity
and visual-eld outcomes by treatment regimen.99 A post hoc
subanalysis found that the ATT sequence was favored for black
patients while the TAT was better for whites for a visual acuity
outcome.103
Overall, these randomized controlled clinical trials demonstrate
that lowering IOP decreases the risk of subsequent visual-eld
399
One of the most important issues is that the prevalence of
glaucoma in the unselected, general population is relatively low.
Thus, the predictive power of a positive test result will be low.
Only a small proportion of those identied as glaucomatous by
the screening test even with a highly valid and suitable test
will actually have the disease; the remainder will nonetheless
undergo costly, unproductive diagnostic work-ups. Focus has
gradually shifted from widespread population-based screening
to case-nding in high-risk individuals to obtain a high yield of
true cases.
SUMMARY
Much of our knowledge of glaucoma epidemiology has come
from population-based prevalence studies. These studies have
documented the relatively common occurrence of higher-thannormal IOP without evidence of glaucomatous damage, glaucomatous damage with normal IOP, and the influence of age and
race on disease prevalence. Research is now available on glaucoma incidence, investigating possible risk factors for disease
development, and evaluating factors that influence glaucoma
progression and outcome.
SECTION 5
REFERENCES
400
1. Friedman DS, Wolfs RC, OColmain BJ,

et al: Prevalence of open-angle glaucoma
among adults in the United States. Arch
Ophthalmol 2004; 122:532538.
2. Goldberg I: How common is glaucoma
worldwide? In: Weinreb R, Kitazawa Y,
Krieglstein G, eds. Glaucoma in the 21st
Century. London: Mosby; 2000:19.
3. Quigley HA: Number of people with
glaucoma worldwide. Br J Ophthalmol
1996; 80:389393.
4. U.S. interim projections by age, sex, race,
and Hispanic origin. Washington, DC: U.S.
Census Bureau.
5. Quigley HA, Broman AT: The number of
people with glaucoma worldwide in 2010
and 2020. Br J Ophthalmol 2006;
90:262267.
6. Tielsch JM: Therapy for glaucoma: costs
and consequences. In: Ball SF, Franklin RM,
eds. Transactions of the New Orleans
Academy of Ophthalmology. Amsterdam:
Kugler; 1993:6168.
7. Janz NK, Wren PA, Lichter PR, et al: The
Collaborative Initial Glaucoma Treatment
Study: interim quality of life ndings after
initial medical or surgical treatment of
glaucoma. Ophthalmology 2001;
108:19541965.
8. Gutierrez P, Wilson MR, Johnson C, et al:
Influence of glaucomatous visual eld loss
on health-related quality of life. Arch
Ophthalmol 1997; 115:777784.
9. Parrish RK 2nd, Gedde SJ, Scott IU, et al:
Visual function and quality of life among
patients with glaucoma. Arch Ophthalmol
1997; 115:14471455.
10. Ciof GA, Mansberger S, Spry P, et al:
Frequency doubling perimetry and the
detection of eye disease in the community.
98:195199.
11. Mason RP, Kosoko O, Wilson MR, et al:
National survey of the prevalence and
risk factors of glaucoma in St. Lucia,
West Indies. Part I. Prevalence ndings.
12. Foster PJ, Buhrmann R, Quigley HA,
Johnson GJ: The denition and classication
of glaucoma in prevalence surveys. Br J
Ophthalmol 2002; 86:238242.
13. Hollows FC, Graham PA: Intra-ocular
pressure, glaucoma, and glaucoma
suspects in a dened population. Br J
Ophthalmol 1966; 50:570586.
14. Bengtsson B: The prevalence of glaucoma.
15. Kahn HA, Milton RC: Revised Framingham
eye study prevalence of glaucoma and
diabetic retinopathy. Am J Epidemiol 1980;
111:769776.
16. Foster PJ, Oen FT, Machin D, et al: The

prevalence of glaucoma in Chinese
residents of Singapore: a cross-sectional
population survey of the Tanjong Pagar
district. Arch Ophthalmol 2000;
118:11051111.
17. Jacob A, Thomas R, Koshi SP, et al:
Prevalence of primary glaucoma in an
urban south Indian population. Indian J
18. Foster PJ, Baasanhu J, Alsbirk PH, et al:
Glaucoma in Mongolia. A population-based
survey in Hovsgol province, northern
Mongolia. Arch Ophthalmol 1996;
114:12351241.
19. Iwase A, Suzuki Y, Araie M, et al: The
prevalence of primary open-angle
glaucoma in Japanese: the Tajimi Study.
Ophthalmology 2004; 111:16411648.
20. Yamamoto T, Iwase A, Araie M, et al: The
Tajimi Study report 2: prevalence of primary
angle closure and secondary glaucoma in a
Japanese population. Ophthalmology 2005;
112:16611669.
21. Leske MC, Connell AM, Schachat AP,
Hyman L: The Barbados Eye Study.
Prevalence of open angle glaucoma.
22. Tielsch JM, Sommer A, Katz J, et al: Racial
variations in the prevalence of primary
open-angle glaucoma. The Baltimore Eye
Survey [see comments]. JAMA 1991;
266:369374.
23. Ntim-Amponsah CT, Amoaku WM, OfosuAmaah S, et al: Prevalence of glaucoma in an
African population. Eye 2004; 18:491497.
24. Buhrmann RR, Quigley HA, Barron Y, et al:
Prevalence of glaucoma in a rural East
African population. Invest Ophthalmol Vis
Sci 2000; 41:4048.
25. Rotchford AP, Kirwan JF, Muller MA, et al:
Temba glaucoma study: a populationbased cross-sectional survey in urban
South Africa. Ophthalmology 2003;
110:376382.
26. Quigley HA, West SK, Rodriguez J, et al:
The prevalence of glaucoma in a
population-based study of Hispanic
subjects: Proyecto VER. Arch Ophthalmol
2001; 119:18191826.
27. Varma R, Ying-Lai M, Francis BA, et al:
Prevalence of open-angle glaucoma and
ocular hypertension in Latinos: the Los
Angeles Latino Eye Study. Ophthalmology
2004; 111:14391448.
28. Arkell SM, Lightman DA, Sommer A, et al:
The prevalence of glaucoma among
Eskimos of northwest Alaska. Arch
Ophthalmol 1987; 105:482485.
29. Lee ET, Russell D, Morris T, et al: Visual
impairment and eye abnormalities in
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
Oklahoma Indians. Arch Ophthalmol 2005;

123:16991704.
Mansberger SL, Romero FC, Smith NH,
et al: Causes of visual impairment and
common eye problems in northwest
American Indians and Alaska natives.
Am J Public Health 2005; 95:881886.
Coffey M, Reidy A, Wormald R, et al:
Prevalence of glaucoma in the west of
Ireland. Br J Ophthalmol 1993; 77:1721.
Ramakrishnan R, Nirmalan PK,
Krishnadas R, et al: Glaucoma in a rural
population of southern India: the Aravind
comprehensive eye survey. Ophthalmology
2003; 110:14841490.
Rotchford AP, Johnson GJ: Glaucoma in
Zulus: a population-based cross-sectional
survey in a rural district in South Africa.
Bonomi L, Marchini G, Marraffa M, et al:
Prevalence of glaucoma and intraocular
pressure distribution in a dened
population. The EgnaNeumarkt Study.
Weih LM, Nanjan M, McCarty CA, Taylor HR:
Prevalence and predictors of open-angle
glaucoma: results from the visual
impairment project. Ophthalmology 2001;
108:19661972.
Dandona L, Dandona R, Srinivas M, et al:
Open-angle glaucoma in an urban
population in southern India: the Andhra
Pradesh eye disease study. Ophthalmology
2000; 107:17021709.
Bengtsson BO: Incidence of manifest
glaucoma. Br J Ophthalmol 1989;
73:483487.
Mukesh BN, McCarty CA, Rait JL, Taylor HR:
Five-year incidence of open-angle
glaucoma: the visual impairment project.
Ophthalmology 2002; 109:10471051.
Schoff EO, Hattenhauer MG, Ing HH, et al:
Estimated incidence of open-angle
glaucoma in Olmsted County, Minnesota.
Armaly MF, Krueger DE, Maunder L, et al:
Biostatistical analysis of the collaborative
glaucoma study. I. Summary report of the
risk factors for glaucomatous visual-eld
defects. Arch Ophthalmol 1980;
98:21632171.
David R, Livingston DG, Luntz MH: Ocular
hypertension a long-term follow-up of
treated and untreated patients. Br J
Ophthalmol 1977; 61:668674.
Hovding G, Aasved H: Prognostic factors
in the development of manifest open
angle glaucoma. A long-term follow-up
study of hypertensive and normotensive
eyes. Acta Ophthalmol (Copenh) 1986;
64:601608.
43. Walker WM: Ocular hypertension. Follow-up

of 109 cases from 1963 to 1974. Trans
44. Kass MA, Heuer DK, Higginbotham EJ, et al:
The Ocular Hypertension Treatment Study:
a randomized trial determines that topical
ocular hypotensive medication delays or
prevents the onset of primary open-angle
glaucoma. Arch Ophthalmol 2002;
120:701713; discussion 829830.
45. Leske MC, Ederer F, Podgor M: Estimating
incidence from age-specic prevalence in
glaucoma. Am J Epidemiol 1981;
113:606613.
46. Gaasterland D, Tanishima T, Kuwabara T:
Axoplasmic flow during chronic experimental
glaucoma. 1. Light and electron
microscopic studies of the monkey optic
nervehead during development of
glaucomatous cupping. Invest Ophthalmol
Vis Sci 1978; 17:838846.
47. Quigley HA, Addicks EM: Chronic
experimental glaucoma in primates. II.
Effect of extended intraocular pressure
elevation on optic nerve head and axonal
transport. Invest Ophthalmol Vis Sci 1980;
19:137152.
48. Cartwright MJ, Anderson DR: Correlation of
asymmetric damage with asymmetric
intraocular pressure in normal-tension
glaucoma (low-tension glaucoma). Arch
Ophthalmol 1988; 106:898900.
49. Crichton A, Drance SM, Douglas GR,
Schulzer M: Unequal intraocular pressure
and its relation to asymmetric visual eld
defects in low-tension glaucoma.
50. Klein BE, Klein R, Sponsel WE, et al:
Prevalence of glaucoma. The Beaver Dam
Eye Study. Ophthalmology 1992;
99:14991504.
51. Lotufo D, Ritch R, Szmyd L Jr, Burris JE:
Juvenile glaucoma, race, and refraction.
JAMA 1989; 261:249252.
52. Perkins ES, Phelps CD: Open angle
glaucoma, ocular hypertension, low-tension
glaucoma, and refraction. Arch Ophthalmol
1982; 100:14641467.
53. Mitchell P, Hourihan F, Sandbach J,
Wang JJ: The relationship between
glaucoma and myopia: the Blue Mountains
Eye Study. Ophthalmology 1999;
106:20102015.
54. Quigley H, Brown A, Morrison J, Drance S:
The size and shape of the optic disc in
normal human eyes. Arch Ophthalmol
1990; 108:5157.
55. Airaksinen PJ, Tuulonen A: Early glaucoma
changes in patients with and without an
optic disc haemorrhage. Acta Ophthalmol
(Copenh) 1984; 62:197202.
56. Shihab ZM, Lee PF, Hay P: The
signicance of disc hemorrhage in openangle glaucoma. Ophthalmology 1982;
89:211213.
57. Leske MC, Heijl A, Hussein M, et al:
Factors for glaucoma progression and the
effect of treatment: the early manifest
glaucoma trial. Arch Ophthalmol 2003;
121:4856.
58. Airaksinen PJ, Mustonen E, Alanko HI:
Optic disc haemorrhages precede retinal
nerve bere layer defects in ocular
hypertension. Acta Ophthalmol (Copenh)
1981; 59:627641.
59. Armaly MF: Genetic determination of
cup/disc ratio of the optic nerve. Arch
60. Drance SM, Begg IS: Sector haemorrhage

a probable acute ischaemic disc change
in chronic simple glaucoma. Can J
Ophthalmol 1970; 5:137141.
61. Sonnsjo B, Bengtsson B, Krakau CE:
Observations concerning the course of
glaucoma. Acta Ophthalmol (Copenh)
1989; 67:261264.
62. Krakau T: Disc haemorrhages and the
etiology of glaucoma. Acta Ophthalmol
Suppl 1989; 191:3133.
63. Rasker MT, van den Enden A, Bakker D,
Hoyng PF: Deterioration of visual elds in
patients with glaucoma with and without
optic disc hemorrhages. Arch Ophthalmol
1997; 115:12571262.
64. Kim SH, Park KH: The relationship between
recurrent optic disc hemorrhage and
glaucoma progression. Ophthalmology
2006; 113:598602.
65. Healey PR, Mitchell P, Smith W, Wang JJ:
Optic disc hemorrhages in a population
with and without signs of glaucoma.
66. Gordon MO, Beiser JA, Brandt JD, et al:
The Ocular Hypertension Treatment Study:
baseline factors that predict the onset of
primary open-angle glaucoma. Arch
Ophthalmol 2002; 120:714720; discussion
829830.
67. Herndon LW, Weizer JS, Stinnett SS:
Central corneal thickness as a risk factor
for advanced glaucoma damage. Arch
Ophthalmol 2004; 122:1721.
68. Medeiros FA, Sample PA, Zangwill LM,
et al: Corneal thickness as a risk factor for
visual eld loss in patients with preperimetric
glaucomatous optic neuropathy. Am J
Ophthalmol 2003; 136:805813.
69. Herndon LW, Choudhri SA, Cox T, et al:
Central corneal thickness in normal,
glaucomatous, and ocular hypertensive
eyes. Arch Ophthalmol 1997;
115:11371141.
70. Kim JW, Chen PP: Central corneal
pachymetry and visual eld progression in
patients with open-angle glaucoma.
Ophthalmology 2004; 111:21262132.
71. Whitacre MM, Stein RA, Hassanein K: The
effect of corneal thickness on applanation
tonometry. Am J Ophthalmol 1993;
115:592596.
72. Ehlers N, Hansen FK, Aasved H: Biometric
correlations of corneal thickness. Acta
Ophthalmol 1975; 53:652659.
73. Doughty MJ, Zaman ML: Human corneal
thickness and its impact on intraocular
pressure measures: a review and metaanalysis approach. Surv Ophthalmol 2000;
44:367408.
74. Jonas JB, Holbach L: Central corneal
thickness and thickness of the lamina
cribrosa in human eyes. Invest Ophthalmol
Vis Sci 2005; 46:12751279.
75. Brandt JD, Beiser JA, Kass MA, Gordon
MO: Central corneal thickness in the Ocular
Hypertension Treatment Study (OHTS).
Ophthalmology 2001; 108:17791788.
76. Coulehan JL, Helzlsouer KJ, Rogers KD,
Brown SI: Racial differences in intraocular
tension and glaucoma surgery. Am J
Epidemiol 1980; 111:759768.
77. Sommer A, Tielsch JM, Katz J, et al:
Relationship between intraocular pressure
and primary open angle glaucoma among
white and black Americans. The Baltimore
Eye Survey. Arch Ophthalmol 1991;
109:10901095.
78. Shiose Y, Kitazawa Y, Tsukahara S, et al:

Epidemiology of glaucoma in Japan a
nationwide glaucoma survey. Jpn J
Ophthalmol 1991; 35:133155.
79. Hall JA, Roter DL, Katz NR: Meta-analysis
of correlates of provider behavior in
medical encounters. Med Care 1988;
26:657675.
80. Phelps CD, Thompson HS, Ossoinig KC:
The diagnosis and prognosis of atypical
carotid-cavernous stula (red-eyed shunt
syndrome). Am J Ophthalmol 1982;
93:423436.
81. Mitchell P, Smith W, Chey T, Healey PR:
Open-angle glaucoma and diabetes: the
Blue Mountains eye study, Australia.
82. Tielsch JM, Katz J, Quigley HA, et al:
Diabetes, intraocular pressure, and primary
open-angle glaucoma in the Baltimore
Eye Survey. Ophthalmology 1995;
102:4853.
83. Dielemans I, Vingerling JR, Wolfs RC, et al:
The prevalence of primary open-angle
glaucoma in a population-based study in
The Netherlands. The Rotterdam Study.
Ophthalmology 1994; 101:18511855.
84. Mitchell P, Lee AJ, Rochtchina E, Wang JJ:
Open-angle glaucoma and systemic
hypertension: the blue mountains eye
study. J Glaucoma 2004; 13:319326.
85. Tielsch JM, Katz J, Sommer A, et al:
Hypertension, perfusion pressure, and
primary open-angle glaucoma. A
population-based assessment. Arch
Ophthalmol 1995; 113:216221.
86. Corbett JJ, Phelps CD, Eslinger P,
Montague PR: The neurologic evaluation of
patients with low-tension glaucoma. Invest
87. Shiose Y: Intraocular pressure: new
perspectives. Surv Ophthalmol 1990;
34:413435.
88. Leske MC, Connell AM, Wu SY, et al: Risk
factors for open-angle glaucoma. The
Barbados Eye Study. Arch Ophthalmol
1995; 113:918924.
89. Mitchell P, Smith W, Attebo K, Healey PR:
Prevalence of open-angle glaucoma in
Australia. The Blue Mountains Eye Study.
Ophthalmology 1996; 103:16611669.
90. Ekstrom C: Prevalence of open-angle
glaucoma in central Sweden. The Tierp
Glaucoma Survey. Acta Ophthalmol Scand
1996; 74:107112.
91. Wensor MD, McCarty CA, Stanislavsky YL,
et al: The prevalence of glaucoma in the
Melbourne Visual Impairment Project.
92. Quigley HA, West SK, Rodriguez J, et al:
The prevalence of glaucoma in a
population-based study of Hispanic
subjects: Proyecto VER. Arch Ophthalmol
2001; 119:18191826.
93. Ciof GA, Liebmann JM: Translating
the OHTS results into clinical practice.
J Glaucoma 2002; 11:375377.
94. Mansberger SL: A risk calculator to
determine the probability of glaucoma.
J Glaucoma 2004; 13:345347.
95. Medeiros FA, Weinreb RN, Sample PA,
et al: Validation of a predictive model to
estimate the risk of conversion from ocular
hypertension to glaucoma. Arch
Ophthalmol 2005; 123:13511360.
96. Miglior S, Zeyen T, Pfeiffer N, et al: Results
of the European Glaucoma Prevention
CHAPTER 36
401
SECTION 5
402
97. Lichter PR, Musch DC, Gillespie BW, et al:

Interim clinical outcomes in the
Collaborative Initial Glaucoma Treatment
Study comparing initial treatment
randomized to medications or surgery.
Ophthalmology 2001; 108:19431953.
98. Drance S, Anderson DR, Schulzer M: Risk
factors for progression of visual eld
abnormalities in normal-tension glaucoma.
Am J Ophthalmol 2001; 131:699708.
99. The Advanced Glaucoma Intervention
Study (AGIS): 12. Baseline risk factors for
sustained loss of visual eld and visual
acuity in patients with advanced glaucoma.
Am J Ophthalmol 2002; 134:499512.
100. Tielsch JM, Katz J, Singh K, et al: A
population-based evaluation of glaucoma
screening: the Baltimore Eye Survey. Am J
Epidemiol 1991; 134:11021110.
101. Leske MC, Heijl A, Hyman L, Bengtsson B:
Early Manifest Glaucoma Trial: design and
baseline data. Ophthalmology 1999;
106:21442153.
102. Musch DC, Lichter PR, Guire KE,
Standardi CL: The Collaborative Initial
Glaucoma Treatment Study: study design,
methods, and baseline characteristics of
enrolled patients. Ophthalmology 1999;
106:653662.
103. The Advanced Glaucoma Intervention
Study (AGIS): 4. Comparison of treatment
outcomes within race. Seven-year results.
Ophthalmology 1998; 105:11461164.
104. Tielsch J, Katz J, Singh K: Populationbased evaluation of glaucoma screening:
The Baltimore Eye Survey. Am J
Epidemiology 1991; 134:11021110.
105. Lichter P: Variability of expert observers in
evaluating the optic disc. Trans Am
106. Harper RA, Reeves BC: Glaucoma
screening: the importance of combining
test data. Optom Vis Sci 1999; 76:537543.
107. Sponsel WE, Ritch R, Stamper R, et al:
Prevent blindness America visual eld
screening study. The prevent blindness
America glaucoma advisory committee.
Am J Ophthalmol 1995; 120:699708.
108. Stewart WC, Hunt HH: Threshold variation
in automated perimetry. Surv Ophthalmol
1993; 37:353361.
109. Herse PR: Factors influencing normal

perimetric thresholds obtained using the
Humphrey Field Analyzer. Invest
110. Wood JM, Wild JM, Bullimore MA,
Gilmartin B: Factors affecting the normal
perimetric prole derived by automated
static threshold LED perimetry. I. Pupil size.
Ophthalmic Physiol Opt 1988; 8:2631.
111. Aung T, Foster PJ, Seah SK, et al:
Automated static perimetry: the influence
of myopia and its method of correction.
112. Weinreb RN, Perlman JP: The effect of
refractive correction on automated
perimetric thresholds. Am J Ophthalmol
1986; 101:706709.
113. Wild JM, Searle AE, Dengler-Harles M,
ONeill EC: Long-term follow-up of baseline
learning and fatigue effects in the
automated perimetry of glaucoma and
ocular hypertensive patients. Acta
Ophthalmol (Copenh) 1991; 69:210216.
114. Heijl A, Lindgren G, Olsson J: The effect of
perimetric experience in normal subjects.
115. Werner EB, Krupin T, Adelson A, Feitl ME:
Effect of patient experience on the results
of automated perimetry in glaucoma
suspect patients [see comments].
116. Wild JM, Dengler-Harles M, Searle AE, et al:
The influence of the learning effect on
automated perimetry in patients with
suspected glaucoma. Acta Ophthalmol
(Copenh) 1989; 67:537545.
117. Autzen T, Work K: The effect of learning
and age on short-term fluctuation and
mean sensitivity of automated static
perimetry. Acta Ophthalmol (Copenh) 1990;
68:327330.
118. Kulze JC, Stewart WC, Sutherland SE:
Factors associated with a learning effect in
glaucoma patients using automated
perimetry. Acta Ophthalmol (Copenh) 1990;
68:681686.
119. Tielsch JM, Sommer A, Witt K, et al:
Blindness and visual impairment in an
American urban population. The Baltimore
Eye Survey. Arch Ophthalmol 1990;
108:286290.
120. Khong JJ, Dimitrov PN, Rait J, McCarty CA:

Can the specicity of the FDT for glaucoma
be improved by conrming abnormal
results? J Glaucoma 2001; 10:199202.
121. Patel SC, Friedman DS, Varadkar P,
Robin AL: Algorithm for interpreting the
results of frequency doubling perimetry [In
Process Citation]. Am J Ophthalmol 2000;
129:323327.
122. Quigley HA: Identication of glaucomarelated visual eld abnormality with the
screening protocol of frequency doubling
technology. Am J Ophthalmol 1998;
125:819829.
123. Paczka JA, Friedman DS, Quigley HA, et al:
Diagnostic capabilities of frequencydoubling technology, scanning laser
polarimetry, and nerve ber layer
photographs to distinguish glaucomatous
damage. Am J Ophthalmol 2001;
131:188197.
124. Trible JR, Schultz RO, Robinson JC,
Rothe TL: Accuracy of glaucoma detection
with frequency-doubling perimetry. Am J
Ophthalmol 2000; 129:740745.
125. Yamada N, Chen PP, Mills RP, et al:
Screening for glaucoma with frequencydoubling technology and Damato
campimetry. Arch Ophthalmol 1999;
117:14791484.
126. Burnstein Y, Ellish NJ, Magbalon M,
Higginbotham EJ: Comparison of
frequency doubling perimetry with
humphrey visual eld analysis in a
glaucoma practice [In Process Citation].
Am J Ophthalmol 2000; 129:328333.
127. Damato BE: Oculokinetic perimetry:
a simple visual eld test for use in the
community. Br J Ophthalmol 1985;
69:927931.
128. Quigley HA, West SK, Munoz B, et al:
Examination methods for glaucoma
prevalence surveys. Arch Ophthalmol 1993;
111:14091415.
129. Medeiros FA, Zangwill LM, Bowd C, et al:
Evaluation of retinal nerve ber layer, optic
nerve head, and macular thickness
measurements for glaucoma detection
using optical coherence tomography.
CHAPTER
37
Epidemiology of Diabetic Retinopathy

Hanna R. Coleman and Donald S. Fong
INTRODUCTION
Diabetic retinopathy is the most common microvascular
complication of diabetes and is a leading cause of visual
impairment among Americans.1 Blindness is 25 times more
common in diabetics.2,3 Approximately 3.4% of the population
is estimated to be affected with diabetic retinopathy and this
translates to ~4.1 million persons with diabetic retinopathy
with 1 in 12 having vision threatening diabetic retinopathy.4
Future projections of the rates of development of diabetic
retinopathy are signicant increases of public health importance as the population will live longer with a high prevalence
of developing diabetes in the future. Diabetic retinopathy can
develop over a long period of time. Knowledge of the demographic distribution of this disease and understanding of the
clinical risk factors is crucial in patient management. This
chapter presents descriptive and analytic epidemiologic data for
diabetic retinopathy in the US.
DEFINITION AND DIAGNOSIS

There are two common types of diabetes mellitus: insulindependent, known as type-1 and non-insulin-dependent, type-2
diabetes mellitus. The latter is more common, comprising
9095% of all cases in the US. These two types differ in their
clinical characteristics, etiologies, and pathophysiologic basis.
A main difference is their propensity to develop diabetic
ketoacidosis in the basal metabolic state. Insulin is required
in type 1 to prevent ketoacidosis, whereas in type 2, ketoacidosis is unlikely even with poor glycemic control. Typically,
type 1 presents acutely with polyphagia, polydipsia, and polyuria. Type 2 is insidious and may be present for years before
being diagnosed clinically.
From the ophthalmic standpoint, patients with type-1
diabetes have a higher risk of developing severe proliferative
diabetic retinopathy. However, a greater percentage of cases of
severe proliferative diabetic retinopathy (PDR) are caused by
type 2 because of the higher prevalence in the general population.
There is evidence that retinopathy begins to develop at least
7 years before the clinical diagnosis of type-2 diabetes is made.5
Diabetic retinopathy has been traditionally divided into
nonproliferative and proliferative categories. Diagnostic correlation among various specialists has varied from 50% to 85%.6,7
The recognized gold standard for grading the severity of diabetic retinopathy in clinical trials is the Early Treatment
Diabetic Retinopathy Study (ETDRS) severity scale. The
ETDRS was a randomized clinical trial of photocoagulation
versus deferral of photocoagulation that allowed the observation of the natural history of diabetic retinopathy. It identied three retinal lesions that are predictive of progression
according to their severity: (1) intraretinal microvascular abnormality, (2) hemorrhages and microaneurysms, and (3) venous
abnormalities. The ETDRS also developed a severity scale that
considers the natural history of diabetic retinopathy as well
as the orderly progression of risk of severe visual loss.8 It was
based on the modied Airlie House classication of diabetic
retinopathy but has not proven to be easy or practical.9,10
A simpler diabetic retinopathy severity scale was developed
by the Global Diabetic Retinopathy Group.11 It consists of
ve levels with increasing risks of retinopathy and is described
in Table 37.1.
The ETDRS also showed that clinically signicant macular
edema (CSME) can lead to moderate visual loss if not treated
by focal photocoagulation. Macular edema is dened as thickening in the macula as seen by biomicroscopy or fundus photography. The diabetic macular edema (DME) disease severity
scale separates eyes with apparent DME from those with no
apparent thickening or lipid in the macula and is summarized
in Table 37.2.
These two clinical disease severity scales are intended to be
a practical method of grading severity of diabetic retinopathy
and DME that will allow observers to recognize and categorize
TABLE 37.1. International Clinical Diabetic Retinopathy Disease

Severity Scale
Proposed Disease Severity Level
Findings Observable Upon

Dilated Ophthalmoscopy
No apparent retinopathy
No abnormalities
Mild nonproliferative diabetic

retinopathy
Microaneurysms only
Moderate nonproliferative
diabetic retinopathy
More than just

microaneurysms but less
than severe NPDR
Severe nonproliferative diabetic

retinopathy
Any of the following:

More than 20 intraretinal
hemorrhages in each of
4 quadrants
Denite venous beading in
2+ quadrants
Prominents IRMA in 1+
quadrant
And no signs of
proliferative retinopathy
Proliferative diabetic retinopathy
One or more of the following:

Neovascularization
Vitreous/preretinal
hemorrhage
403
TABLE 37.2. International Clinical Diabetic Macular Edema Disease Severity Scale
Findings Observable Upon Dilated

Ophthalmoscopy
Diabetic macular edema apparently

absent
No apparent retinal thickening or hard exudates

in posterior pole
Diabetic macular edema apparently

present
Some apparent retinal thickening or hard exudates

in posterior pole
If Diabetic Macular Edema is Present, it can be Categorized as Follows:

Findings Observable Upon Dilated

Ophthalmoscopy *
Diabetic macular edema present
Mild diabetic macular edema

Some retinal thickening or hard exudates in posterior
pole but distant from the center of the macula
Moderate diabetic macular edema
Retinal thickening or hard exudates approaching the
center of the macula but not involving the center
Severe diabetic macular edema
Retinal thickening or hard exudates involving the
center of the macula
SECTION 5
* Hard exudates are a sign of current or previous macular edema. Diabetic macular edema is dened as
retinal thickening and this requires a three-dimensional assessment that is best performed by a dilated
examination using slit-lamp biomicroscopy and/or stereo fundus photography.
404
levels of retinopathy and the presence of most DME resulting in

more appropriate and consistent referrals to treatment centers.
DR, and 5000 new cases of blindness are estimated to occur

each year.22
INCIDENCE AND PREVALENCE
RISK FACTORS
There are no national data on the prevalence or incidence of

diabetic retinopathy in the US. Population studies have been
performed on select populations such as inhabitants of
Rochester, Minnesota; Framingham, Massachusetts; Pittsburgh,
Pennsylvania; Mexican-Americans in San Antonio; blacks in
the Bahamas and Pima Indians.1218 The largest populationbased study using fundus photography to document diabetic
retinopathy was the Wisconsin Epidemiologic Study of
Diabetic Retinopathy (WESDR). It identied previously diagnosed diabetics in 11 counties in Wisconsin. Because the determination of insulin dependency can be difcult, the WESDR
investigators divided the diabetics by age of diagnosis into two
cohorts. The rst cohort was composed of patients whose
disease was diagnosed prior to the age of 30. The second cohort
was a stratied random sample of patients whose diagnosis was
made after the age of 30. The prevalence of diabetic retinopathy
determined between 1980 and 1982 was 50.1%. The prevalence
of PDR with high-risk characteristics was 2.2%.19,20 Youngeronset patients with diabetes had the highest prevalence of any
retinopathy, proliferative retinopathy, and macular edema.
The 4-year incidence of developing retinopathy was 40.3%,
whereas the incidence of developing PDR with high-risk
characteristics was 2.4%. The younger-onset group had the
highest incidence rate of progression to PDR, whereas the olderonset patients with diabetes had the highest incidence of
macular edema.
Although WESDR provides the only population-based data
for calculating rates, there are caveats common to this
epidemiologic study: The population in Wisconsin is relatively
homogeneous ethnically, consisting of few blacks, Hispanics,
and Asians. Since the study was based on previously diagnosed
individuals with diabetes, rates on type 1 may be skewed,
because up to one-half of all persons with type 2 are undiagnosed.21 Generalizing the WESDR data to the US population,
11 000 new cases of DME, 22 000 new cases of proliferative
A number of risk factors have been identied for the

development and progression of diabetic retinopathy.
DURATION
The strongest predictor is duration of diabetes. In younger
type-1 group in WESDR, the prevalence of any retinopathy
was 8% at 3 years, 25% at 5 years, 60% at 10 years, and 80% at
15 years. The prevalence of PDR was 0% at 3 years and
increases to 25% at 15 years.23 In the Pittsburgh Epidemiology
of Diabetes Complications Study (PEDCS) for patients with
type-1 diabetes, longer duration of diabetes was also observed
in those with PDR than in those with no retinopathy.24
Among groups comprising mostly type-2 diabetes, retinopathy was more frequent earlier after the diagnosis; 23% had
retinopathy at 3 years and 2% had proliferative retinopathy.25
However, after 20 years of duration of disease, smaller
proportion of older-onset individuals had any proliferative
retinopathy than in the younger-onset group.
The incidence of developing retinopathy also increases with
increasing duration. The 4-year incidence of developing
proliferative retinopathy in the WESDR younger-onset group
increased from 0% during the rst 5 years to 27.9% during
years 1314 of diabetes. After 15 years, the incidence of
developing PDR remained stable. In a cohort study of patients
with type-1 diabetes from the Joslin Clinic, the incidence rate
(cases of PDR/1000 person-years) for development of PDR
was 1.5 in patients with less than 10 years of diabetes, rises to
30 during the second decade of diabetes, and remains at this
level for the next 25 years.26
In the WESDR older-onset group, the 4-year incidence of
developing proliferative retinopathy in those with less than
5 years of follow-up was 2%. In the Rochester, Minnesota
study, the 20-year cumulative incidence of PDR was 4% and
2% in obese and nonobese patients with type-2 diabetes,27
respectively. Among Mexican-American patients with type-2

diabetes in the San Antonio study, duration was again
signicantly associated with the development of retinopathy.
HYPERGLYCEMIA
The relationship between hyperglycemia and retinopathy has
been strongly documented. In WESDR,28 the level of glycemic
control was measured by the level of hemoglobin Alc.
Among patients with IDDM less than 18 years of age, those
who developed retinopathy had higher levels of glycosylated
hemoglobin than those who did not develop retinopathy
(10.4% vs 12.1%, p < 0.001).29 No association was found in
the older age groups between Hg Alc and the development of
any retinopathy or proliferative diabetic retinopathy. In the
Pittsburgh Prospective Insulin-Dependent Diabetes Cohort
Study, seven of 62 newly diagnosed patients with IDDM
developed retinopathy. Those who developed retinopathy had
a higher level of Hg Alc than those who did not (13.0 vs 11.7%,
p < 0.05).30
Among older-onset patients with diabetes, subject in the
highest quartile of glycosylated hemoglobin are 2.5 times as
likely to have retinopathy as those in the lowest quartile.
This relationship existed even after controlling for duration
of diabetes. In the Rochester, Minnesota study of type-2
diabetes, after controlling for other factors, elevated fasting
blood sugar was associated with an increased risk of developing
diabetic retinopathy and PDR.31 Evidence for the role of
hyperglycemia in the development of retinopathy is present
also in experimental studies in animals; poorly controlled
diabetic animals are more likely to develop retinopathy.32
Several earlier clinical trial studies attempted to demonstrate
the effect of tight glycemic control but were difcult to interpret,
because of small sample sizes, short follow-ups, and failure to
adjust for patients with different levels of retinopathy.3335 To
address these issues, the Diabetes Control and Complications
Trial (DCCT) was designed and executed in 1441 patients with
type-1 diabetes.36 The DCCT asked whether (1) intensive treatment of glycemia would prevent or delay the progression of early
non-PDR (primary prevention cohort), and whether (2) intensive
glycemic control would prevent the progression of early
retinopathy to more advanced forms of retinopathy (secondary
intervention) cohort.
In the primary prevention cohort, the cumulative incidence
of a three-step increase in retinopathy level sustained over
6 months was quite similar between the two groups, during
the rst 36 months. From that point on, there was a persistent
decrease in the intensive group. From 5 years onward, the
cumulative incidence was ~50% less in the intensive group.
During a mean follow-up of 6 years, retinopathy developed
in 23 patients in the intensive group and 91 in the conventional
group. Intensive therapy reduced the mean risk of retinopathy
by 76% (95% condence interval: 6285).
In the secondary intervention cohort, the intensive group
had a higher cumulative incidence of sustained progression
during the rst year. However, by 36 months, the intensive
group had lower risks of progression. Intensive therapy reduced
the risk of progression by 54% (95% condence interval: 3966).
In addition, the risk for proliferative diabetic retinopathy, severe
NPDR, and laser photocoagulation was also reduced.
The protective effect of intensive therapy for retinopathy
was found to be consistent in all subgroups. It was also protective against neuropathy, microalbuminuria, and albuminuria and
reduced the development of hypercholesterolemia. Cardiovascular disease was reduced by 57% in the EDIC study.36a The
incidence of severe hypoglycemia was three times higher in the
intensive group which poses a signicant problem. Because of
the associated dangers of hypoglycemic episodes, the DCCT

recommends that therapy should be individualized for each
patient.
After 6.5 years of follow-up, the DCCT ended, and all
patients were encouraged to maintain strict control of blood
sugar. These patients are followed in the Epidemiology of
Diabetes Interventions and Complications trial (EDIC), which
includes 95% of DCCT subjects, half from each treatment
group. A total of 12941335 patients have been examined annually in the EDIC. Further progression of diabetic retinopathy
during the rst 4 years of the EDIC was 6677% less in the
former intensive treatment group than in the former conventional treatment group.37 The benet persists even at 7 years.38
This benet included an effect on severe diabetic retinopathy,
including severe nonproliferative diabetic retinopathy, proliferative diabetic retinopathy, clinically signicant macular edema,
and the need for focal or scatter laser therapy. The decrease in
the mean hemoglobin A1C (HbA1C) from 9% to ~8% did not
drastically reduce the progression of diabetic retinopathy in the
former conventional treatment group, nor did the increase in
HbA1c from ~7% to ~8% drastically accelerate diabetic retinopathy in the former intensive treatment group. Thus, it takes
time for improvements in control to negate the long lasting
effects of prior prolonged hyperglycemia, and once the biological
effects of prolonged improved control are manifest, the benets
are long lasting. Furthermore, the total glycemic exposure of the
patient (i.e., degree and duration) determines the degree of
retinopathy observed at any one time.
In 2005, the DCCT/EDIC study group reported that
intensive treatment reduced the risk of any cardiovascular
disease event by 42% (95% CI: 96%; p = 0.02) and the risk
of nonfatal myocardial infarction, stroke, or death from
cardiovascular disease by 57% (95% CI, 1279%; p = 0.02).39
For the rst time, intensive treatment of hyperglylcemia was
proven to be benecial in reducing the risk of cardiovascular
disease in persons with diabetes.
In 1998, the United Kingdom Prospective Diabetes Study
Group (UKPDS)40 compared the effects of intensive bloodglucose control with either sulfonylurea or insulin and
conventional treatment on the risk of microvascular and
macrovascular complications in patients with type-2 diabetes
in a randomized controlled trial. It similarly found that intensive control substantially decreases the risk of microvascular
complications for these patients. There was a 25% reduction
in the risk of the any diabetes-related microvascular endpoint,
including the need for retinal photocoagulation in the intensive
treatment group compared to the conventional treatment group.
After 6 years of follow-up, a smaller proportion of patients in
the intensive treatment group than in the conventional group
had a two-step progression (worsening) in diabetic retinopathy
(p < 0.01). Epidemiologic analysis of the UKPDS data showed
a continuous relationship between the risk of microvascular
complications and glycemia, such that for every percentage
point decrease in hemoglobin A1C (e.g., 98%), there was a 35%
reduction in the risk of microvascular complications.
In summary, intensive treatment of type-1 diabetes delays
the onset and slows the progression of diabetic retinopathy,
nephropathy, neuropathy, hypercholesterolemia, and cardiovascular disease. While the DCCT also conrmed the early
worsening of retinopathy with intensive glycemic control seen
in the early treatment it showed that tight control still leads
to subsequent protection. The benets of tight glucose
control were similarly found in patients with type-2 diabetes.
The results of both the DCCT and UKPDS show that while
intensive therapy of glucose reduces the risk of the development and progression of diabetic retinopathy, it does not
prevent retinopathy completely.
CHAPTER 37
405
BLOOD PRESSURE
HYPERLIPIDEMIA
Epidemiologic observations suggest that hypertension increases

the risk of diabetic retinopathy and macular edema. In WESDR,
progression of retinopathy was associated with higher diastolic
blood pressure (BP) at baseline and an increase in diastolic
BP over a 4-year follow-up period.41 Among older-onset
patients, increased diastolic BP was associated with a higher
incidence of macular edema.42
The UKPDS reported the effectiveness of tight BP
control.43,44 It randomized 1148 hypertensive patients with
type-2 diabetes to less tight (<180/105 mmHg) and tight BP
control (<150/85 mmHg) with the use of an angiotensin
converting enzyme (ACE) inhibitor or a b-blocker. Patients in
the tight control group had a 34% reduction in progression
of retinopathy and a 47% reduced risk of deterioration in
visual acuity of three lines compared with the less tight
control group. There were also reductions in deaths related to
diabetes and stroke and no clear adverse reactions to tight BP
control. There was no difference in the efcacy of ACE
inhibitors or b-blockers with regard to progression of DR in
type-2 diabetics, suggesting that BP control and not the type
of medication is most important in those with hypertension.45
The EUCLID study group investigated the effect of lisinopril
in patients with type-1 diabetes who were normotensive
and normo- or microalbuminuric.46 After 2 years, the clinical
trial showed a statistically signicant 50% (p = 0.02) reduction
in retinopathy progression by at least one level; 13.2% of
159 patients on lisinopril progressed versus 23.4% of 166
patients on placebo. After adjusting for center and glycemic
control, the protective effect (0.55, p = 0.06) was similar
but not statistically signicant. This borderline effect in this
study and the ndings from ABCD may be due to the small
benet of incremental lowering of BP in normotensive
patients.
Dyslipidemia is a known risk factor for diabetic renal disease,

but the effect of serum lipids on DR and macular edema is
still under investigation.47,5458 There is observational evidence
that elevated lipids may increase the morbidity of macular
edema, and affect the risk of diabetic retinopathy severity.
Among insulin-using patients in WESDR, the presence of retinal hard exudates was signicantly associated with increased
serum cholesterol levels.59 Likewise, patients in ETDRS who
had elevated serum cholesterol or low-density lipoprotein
levels at baseline were more likely to have retinal hard exudates
than those with normal levels.60 Development of retinal hard
exudates was also 50% more likely in those patients with
elevated serum total cholesterol or triglyceride levels. Because
the risk of loss in visual acuity was correlated with the degree
of retinal hard exudates, reducing serum lipid levels in patients
with diabetes and retinopathy may be particularly important.
In addition, severe hard exudates can lead to the development
of subretinal brosis, a complication that can lead to permanent loss of vision.61 Whether intensive lipid-lowering therapy
will reduce the severity of retinopathy or the resultant losses
in visual acuity remains to be tested in prospective trials.
SECTION 5
IMPAIRED RENAL FUNCTION
406
Retinopathy and nephropathy are two important microvascular

complications in diabetic patients with hyperglycemia and
hypertension.47 The Microalbuminuria Collaborative Study
Group found that retinopathy was not an independent predictor of albuminuria, but the ETDRS and the WESDR study
found that the presence and severity of DR are still indicators
of the risk of developing proteinuria.4850 Conversely,
proteinuria is a known predictor of the development of PDR
in type-1 diabetics51 and gross proteinuria is also associated
with a 95% increased risk of developing DME among persons
with type-1 diabetes.52 There is controversy as to whether this
association is due to hyperglycemia, or whether nephropathy
is an independent risk factor for diabetic retinopathy.
Use of angiotensin converting enzyme inhibitors (ACE-I)
slow the progression of nephropathy. Serum pro-renin concentrations have recently been directly correlated with the
severity of DR and components of the reninangiotensin
system (RAS) have been found in the eye. These observations
imply that the use of ACE-I may also protect against the
development and progression of DR.53 However, the association
between the RAS and the development and progression of DR
is not straightforward, and initial results looking at the
influence of ACE-I upon DR in normotensive diabetics have
had equivocal results. A large randomized, double-masked,
placebo-controlled trial examining the efcacy of ACI-I (and
ACE receptor blockers) in both type-1 and -2 diabetics is currently underway. Patients with refractory retinopathy and
macular edema should have an evaluation of their renal
status.
VESSEL CALIBER
WESDR showed that for patients with type-1 diabetes, larger
retinal arteriolar caliber is an independent predictor of
incident retinopathy in children and adolescents.62 For type-2
diabetics, variations in retinal vascular caliber are related to a
number of systemic and ocular factors but further evaluation
may provide insight into diabetic changes as well.63
DEMOGRAPHICS
The prevalence of PDR is higher in young males, but there is
no difference in the incidence of progression between the
sexes.18 Among older diabetics, there is no sex differential. In
the WESDR younger-onset group, children 1012 years old
compared to those less than 10 years old have a 4-year relative
risk of 3.6. This increased risk is believed to be due to the
10- to 12-year-old children passing through puberty during
the 4-year period of follow-up. Prior to puberty, children rarely
develop diabetic retinopathy regardless of the duration of
diabetes. Among type-2 diabetes, younger age at examination
was a strong risk factor for the 4-year progression of diabetic
retinopathy. PDR was not seen in those whose age at examination was older than 75 years of age.
Pima Indians have the highest prevalence and incidence of
type 2 diabetes. The 20-year cumulative incidence was 14%.26
Mexican-Americans also have a high prevalence of type-2
diabetes and diabetic retinopathy when compared with nonHispanic whites. They have three to ve times the prevalence
of type-2 diabetes and are more likely to develop any retinopathy and severe retinopathy (preproliferative and proliferative
retinopathy).18,48
The epidemiology of diabetic retinopathy in Asian-Americans
is limited. In one study of Japanese-Americans living in Seattle,
the prevalence of diabetic retinopathy was reported to be 11.5%.
Additional epidemiologic studies will be needed.64
There are not many studies of the prevalence of the disease
in African-Americans, but the 9-year diabetic retinopathy incidence in the Barbados Eye study, a population with similar
ancestry, was 39.6% (38.0% for minimum, 9.0% for moderate,
and 2.6% for severe/proliferative DR). Of persons with preexisting DR at baseline, 8.2% progressed to proliferative DR.
The CSME incidence was 8.7%. All incidences tended to
GENETICS
In studies of identical twins, the retinopathy was observed
to have similar onset and severity.67,68 HLA-DR antigens have
been examined in WESDR and at the Joslin Clinic. After controlling for duration of diabetes, diastolic BP, proteinuria, and
history of hypertension, the DR4 allele was associated with
an increased risk of proliferative retinopathy in WESDR. The
Joslin Clinic study showed an increased risk of PDR in DR3 and
DR4 homozygote that was neutralized in the presence of
myopia. DR3 and DR4 heterozygote showed decreased risk.69
Other studies have shown no association between HLA
antigens and diabetic retinopathy.70,71
PREGNANCY
There are few studies on the effect of pregnancy on diabetic
retinopathy.7274 One review reported that 8% of women with
minimal to no retinopathy had progression during their pregnancy; if proliferative retinopathy was present, 25% progressed.75 In one prospective study, the risk of progression of
retinopathy was 2.3 higher during pregnancy as compared to
controls.76 The role of tight control that is instigated at the
beginning of pregnancy may play an important role in this
accelerated progression of retinopathy during pregnancy.77,78
Progression of retinopathy during pregnancy was proven to be
secondary to both the tight control and to the pregnancy itself.79
ALCOHOL
In a case-control study examining the relationship between
alcohol and retinopathy of IDDM from the Joslin Clinic, the
percentage of subjects consuming alcohol was similar in those
subjects with and without proliferative diabetic retinopathy.33
In a prospective study of 296 diabetic men, the relative risk in
heavy drinkers of developing severe retinopathy was 3.5 (95%
condence interval, 1.28.4).80
SOCIOECONOMIC STATUS
One case-control study reported an association between proliferative retinopathy and working-class occupational status
and lower income in patients with IDDM.81 Among MexicanAmericans in San Antonio, SES determined by Duncans socioeconomic index, education, and income was not associated
with retinopathy status.82 Similarly, in Oklahoma Indians
socioeconomic factors also were not found to be a risk factor
for retinopathy.36
Key Features: Medical Risk Factors for Progression of
Diabetic Retinopathy
Hyperglycemia. Observational and clinical trials support for

benecial effects of achieving tight glucose control for
reducing the risk of diabetic retinopathy by 3570%
Hypertension. Modest reduction in both systolic and diastolic
BP result in reduction in diabetic retinopathy progression
Hyperlipidemia. Observational data to suggest that progression
of diabetic retinopathy and the development of macular edema
may result from dyslipidemia. Clinical trials are underway
Pregnancy may increase the risk of progression of diabetic
retinopathy
TREATMENT AND FUTURE DIRECTIONS

Because diabetic retinopathy is a signicant source of visual
loss among diabetics, the National Eye Institute (NEI) and
other research institutions sponsor several multicenter clinical
trials to determine the optimal management regimen for
patients with diabetic retinopathy.
DIABETIC RETINOPATHY STUDY

The Diabetic Retinopathy Study (DRS) was the rst multicenter randomized controlled clinical trail in ophthalmology.83
The study addressed the question of whether photocoagulation
therapy was benecial in patients with diabetic retinopathy in
over 1700 patients with severe nonproliferative or proliferative
diabetic retinopathy.
The study showed a reduction, after only 2 years, in the
cumulative event rate (visual acuity <5/200, at two consecutive
4-month follow-up visits) from 16.3% in untreated eyes to
6.4% in treated eyes (z = 5.5). This early benet persisted
at 5 years; the difference between the treated group compared
to the control group was even greater (z = 11.0). In addition,
argon laser compared to xenon arc photocoagulation was
found to cause fewer side effects. The study further identied
eyes that were at high risk for severe visual loss and for which
photocoagulation was of particular benet. The features of
these eyes can be summarized as follows: (1) neovascularization
of the disk, severity greater than standard photo 10A; (2) any
neovascularization of the disk if accompanied by vitreous
or preretinal hemorrhage; and (3) vitreous hemorrhage
accompanied by one-half disk area of neovascularization
elsewhere.84
EARLY TREATMENT DIABETIC RETINOPATHY

STUDY
The ETDRS was designed to determine when in the course of
diabetic retinopathy it is most effective to initiate scatter or
pan-retinal photocoagulation, whether photocoagulation is
effective in the treatment of DME, and whether aspirin treatment is effective in altering the course of diabetic retinopathy.
The results from the ETDRS showed that early treatment
compared with deferral of photocoagulation until high-risk
characteristics were observed is associated with a small
reduction in the incidence of severe visual loss. The 5-year
rates of severe visual loss were 2.6% in the early-treatment
group versus 3.7% in the deferral of treatment group. The
relative risk of severe visual loss in eyes randomized to early
photocoagulation compared to eyes assigned to deferral was
0.77 (99% condence interval, 0.561.06). Furthermore,
scatter laser photocoagulation is probably not benecial for
eyes with mild or moderate nonproliferative diabetic
retinopathy. However, in those with type-2 diabetes, additional
analyses of visual outcome in ETDRS patients with severe
NPDR to non-high-risk PDR suggest that the recommendation
to consider scatter photocoagulation prior to the development
of high-risk PDR is particularly appropriate for patients with
type-2 diabetes. The risk of severe vision loss or vitrectomy was
reduced by 50% in those who were treated early compared
with the deferral until high-risk PDR developed.
Regarding the management of DME, the ETDRS demonstrated that eyes with CSME should be considered for
treatment. Eyes assigned to immediate focal photocoagulation
were about half as likely to double their visual angle (12% in
those treated versus 24% in those assigned to deferral, z = 2.58)
at 3 years.85 It also showed that aspirin did not affect the
course of retinopathy.
CHAPTER 37
increase with diabetes duration.65 A more recent study of

whites, African-Americans, Hispanics, and the Chinese in the
multiethnic study of atherosclerosis (MESA) found race not
to be a signicant risk factor in diabetic retinopathy.66
407
DIABETIC RETINOPATHY VITRECTOMY

STUDY
not delay the progression to sight threatening macular

edema.92a
The Diabetic Retinopathy Vitrectomy Study (DRVS) was a

multicenter, randomized clinical trail that addressed the risks
and benets of performing pars plana vitrectomy in eyes with
severe proliferative diabetic retinopathy. The DRVS was divided
into three studies. The rst study, DRVS-Group N, was a
natural history study to examine the course of severe PDR
managed by conventional therapy. It showed that decreases in
visual acuity were more likely during the rst year than the
second year of follow-up. The second study, DRVS-Group H,
examined the timing of vitrectomy in eyes with severe vitreous
hemorrhage of less than 6 months duration. Four-year
follow-up showed that the proportion of eyes with visual acuity
of 10/20 or better was higher (p <0.05) in the early vitrectomy
group than in the deferral group.86 The benet of early
vitrectomy to patients with IDDM, but not to patients with
NIDDM, remained after 4 years of follow-up. The third study,
DRVS-Group NR, was a randomized clinical trail comparing
early vitrectomy with conventional management in eyes with
extensive, active neovascular, or brovascular proliferations
and useful vision. After 4 years of follow-up, the percentage
of eyes with visual acuity of 10/20 or better was 44% in the
early-vitrectomy group and 28% in the conventionalmanagement group (p < 0.05). The proportion with very poor
visual outcome was similar in the two groups. The advantage
of early vitrectomy tended to increase with increasing severity
of new vessels. In the group with the least severe new vessels,
no advantage of early vitrectomy was apparent. Thus, the
decision to perform early vitrectomy on eyes with severe PDR
and good vision remains complex.
CORTICOSTEROIDS
Corticosteroids, a class of substances with antiinflammatory
properties, have been demonstrated to inhibit the expression of
the VEGF gene and reduce the induction of VEGF by proinflammatory mediators in a time and dose-dependent manner.93,94
Multiple case series using 4 mg/0.1 mL of intravitreal triamcinolone suggest efcacy in the treatment of macular edema,
however, elevated intraocular pressure and cataract formation
are important side effects and their long-term effect on functional outcome is still unknown.95,96 To investigate the safety
and efcacy in DME, the NIH is sponsoring a clinical trial
investigating the efcacy of intravitreal triamcinolone for
DME through the Diabetic Retinopathy Clinical Research
Network.97
ANTI-VEGF TREATMENTS
A number of anti-VEGF antibodies are under investigation.
(Macugen) Pegaptanib is a 28-base oligonucleotide ligand
(aptamer) that binds VEGF and was approved for treatment
of neovascular AMD.98 It is currently being investigated in a
phase-3 clinical trial on DME. Lucentis (ranibizumab) is an
antibody fragment directed against VEGF. There are plans to
investigate this product in the treatment of DME.99,100 Avastin,
(bevacizumab) the larger compound from which Lucentis is
derived is also being evaluated for its effect on macular edema
as well as neovascular proliferation.101
VITRECTOMY
SECTION 5
PKC INHIBITORS
The relation between metabolic abnormalities and diabetic
microvascular complications has long been suspected but the
underlying pathologic mechanisms are not clearly understood.
Early data suggest that protein kinase C (PKC) is an important
factor. Hyperglycemia increases diacylglycerol (DAG), an activator of PKC. There are multiple isoforms of PKC but PKC-b2
isoform is preferentially activated in tissues that usually are
damaged in diabetes such as retina and kidney. Ruboxistaurin
(RBX LY333531) is a specic inhibitor of PKC-b87 and has been
found to block vascular complications of diabetes, including
abnormalities in retinal blood flow, neovascularization, and
VEGF-mediated effects on permeability in animal models.8890
Early phase-1 safety studies with orally administered RBX
showed that it was well tolerated with no signicant adverse
effects. It also showed that while it did not prevent progression
of retinopathy to proliferative disease it was associated with
less visual loss especially in patients with clinically signicant
macular edema.91,92 A phase-2 study showed that RBX did
Various literature reports suggest that vitrectomy surgery may

be helpful in eyes with refractory macular edema.102 In some
eyes, tangential tractional forces from the vitreous may be the
reason for visual loss. The role of the vitreous has been
described in cystoid macular edema secondary to uveitis, retinitis pigmentosa, and aphakia.103 Diabetic eyes with macular
edema may have a lower rate of posterior vitreous separation
(20%) than those without macular edema (55%).104 Tangential
vitreomacular traction may arise from contraction of the
premacular hyaloid membrane and cause increased permeability of the retinal vasculature or retinal detachment. In a
recent study using optical coherence tomography to evaluate
the macula in eyes with DME and thickened posterior hyaloid,
a shallow macular traction detachment was observed in eight
of nine eyes.
These observations have led some investigators to recommend vitrectomy for patients with refractory DME
(Table 37.3).105109 In every series, eyes that improved with
vitrectomy had an intact/attached macular posterior vitreous
TABLE 37.3. Published Reports of Vitrectomy for Diffuse Macular Edema

Study
Eyes
Vitreous Findings
Previous Focal
Photocoagulation (%)
Lewis et al
10
Thickened hyaloid
90%
80%
60%
Van Effenterre et al
22
Thickened hyaloid
64%
45%
86%
Harbour et al
408
Resolution of
Edema (%)
>2 Lines of Snellen

Acuity Increase (%)
Thickened hyaloid
57%
57%
57%
Tachi and Ogino
58
Attached hyaloid
19%
98%
53%
Pendergrast
59
Thickened hyaloid
86%
73%
47%

hyaloid attachment. However, one cannot conclude that eyes
with diffuse macular edema undergoing vitrectomy had a more
favorable clinical course than eyes that did not undergo
vitrectomy. The natural history of these eyes is unknown
because the study did not provide a comparison group. The
efcacy of vitrectomy surgery will likely require investigation
with a randomized clinical trial.
SUMMARY
Diabetic retinopathy is a major cause of blindness in the
United States. It occurs in both insulin-dependent diabetes
mellitus and noninsulin-dependent diabetes mellitus. Because
NIDDM accounts for 90% of diagnosed cases of diabetes, most
cases of PDR are due to NIDDM even though the retinopathy

is more severe in IDDM.
Many risk factors for diabetic retinopathy have been studied.
The most important risk factors are duration of diabetes and
hyperglycemia. Further studies are necessary to improve these
patients overall health and quality of life. With these needs in
mind, the Diabetic Retinopathy Clinical Research Network97
was formed in September 2002 and is funded by the NEI. It is
a collaborative network to facilitate the identication, design,
and implementation of multicenter clinical research initiatives
focused on diabetes-induced retinal disorders. It currently
includes over 150 participating sites (ofces) with over 500
physicians throughout the United States and will likely head
signicant advances in the near future.

Group: Causes and prevalence of visual
impairment among adults in the United
States. Arch Ophthalmol 2004; 122:477485.
2. Kahn HA, Hiller R: Blindness caused by
diabetic retinopathy. AM J Ophthalmol
1974; 78:5867.
3. Palmberg PF: Diabetic retinopathy.
Diabetes 1977; 26:703709.
Group: The prevalence of diabetic
retinopathy among adults in the United
States. Arch Ophthalmol 2004;
122:552563.
5. Klein R, Klein BEK, Moss SE: Visual
impairment in diabetes. Ophthalmology
1984; 91:19.
6. Sussman EJ, Tsiaras WG, Soper KA:
Diagnosis of diabetic eye disease. JAMA
1982; 247:32313234.
7. Moss SE, Klein R, Kessler SD, Richie KA:
Comparison between ophthalmoscopy and
fundus photography in determining severity
of diabetic retinopathy. Ophthalmology
1985; 92:6267.
8. Early Treatment Diabetic Retinopathy Study
Research Group: Fundus photographic risk
factors for progression of diabetic
retinopathy. ETDRS report number 12.
9. The Diabetic Retinopathy Study Research
Group: A modication of the Airlie House
Classication of diabetic retinopathy. DRS
report #7. Invest Ophthalmol Vis Sci 1981;
21:210226.
10. Fukuda M: Clinical arrangement of
classication of diabetic retinopathy.
Tohoku J Exp Med 1983;
141(Suppl):331335.
11. Wilkinson CP, Ferris FL 3rd, Klein RE, et al:
Proposed international clinical diabetic
retinopathy and diabetic macular edema
disease severity scales. Ophthalmology
2003; 110:16771682.
12. Dwyer MS, Melton LG, Ballard DJ, et al:
Incidence of diabetic retinopathy and
blindness: a population-based study in
Rochester, Minnesota. Diabetes Care 1985;
8:316.
13. Kahn HA, Leibowitz HM, Ganley JP, et al:
The Framingham Eye Study. I. Outline and
major prevalence ndings. Am J Epidemiol
1977; 106:1732.
14. Orchard TJ, Dorman JS, Maser RE, et al:
Prevalence of complications in IDDM by
sex and duration. Pittsburgh Epidemiology
of Diabetes Complications Study II.
Diabetes 1990; 39:11161124.
15. Haffner SM, Fong D, Stern MP, et al:

Diabetic retinopathy in Mexican Americans
and non-Hispanic whites. Diabetes 1988;
37:878884.
16. Bennett PH, Rushforth NB, Miller M, et al:
Epidemiologic studies of diabetes in the
Pima Indians. Recent Prog Horm Res 1976;
32:333376.
17. Wong TY, Klein R, Islam A, et al: Diabetic
retinopathy in a multi-ethnic cohort in the
United States. Am J Ophthalmol 2006;
141:539541.
18. Varma R, Torres M, Pena F, et al:
Prevalence of diabetic retinopathy in adult
Latinos: the Los Angeles Latino eye study.
Ophthalmology 2004; 111:12981306.
19. Diabetic Retinopathy Study Research
Group: Four risk factors for severe visual
loss in diabetic retinopathy: the third
report from the Diabetic Retinopathy Study.
20. Roy MS, OColmain BJ, Klein BE, et al:
The prevalence of diabetic retinopathy
among adult type 1 diabetic persons in the
United States. Arch Ophthalmol 2004;
122:546551.
21. Harris MI, Hadden WC, Knowler WC,
Bennett PH: Prevalence of diabetes and
impaired glucose tolerance and plasma
glucose levels in the U.S. population aged
2074 yr. Diabetes Care 1987; 36:523534.
22. Javitt JC, Canner JK, Sommer A: Cost
effectiveness of current approaches to the
control of retinopathy in Type 1 diabetics.
23. Klein R, Klein BE, Moss SE, et al: The
Wisconsin epidemiologic study of diabetic
retinopathy. II. Prevalence and risk of
diabetic retinopathy when age at diagnosis
is less than 30 years. Arch Ophthalmol
1984; 102:520525.
24. Kostraba JN, Klein R, Dorman JS, et al:
The epidemiology of diabetes
complications study. IV. Correlates of
diabetic background and proliferative
retinopathy. Am J Epidemiol 1991;
133:381391.
25. Klein R, Klein BEK, Moss SE, et al: The
retinopathy. III. Prevalence and risk of
diabetic retinopathy when age at diagnosis
is 30 or more years. Arch Ophthalmol 1984;
102:527532.
26. Krolewski AS, Warram JH, Rand LI, et al:
Risk of proliferative diabetic retinopathy in
juvenile-onset type I diabetes: a 40-year
follow-up study. Diabetes Care 1986;
9:443452.
27. Ballard DJ, Melton LJ, Dwyer MS, et al:

Risk factors for diabetic retinopathy: a
population-based study in Rochester,
Minnesota. Diabetes Care 1986;
9:334342.
28. Klein R, Klein BEK, Moss SE, et al:
Glycosylated hemoglobin predicts the
incidence and progression of diabetic
retinopathy. JAMA 1988; 250:28642871.
29. Kostraba JN, Klein R, Dorman JS, et al:
The epidemiology of diabetes
complications study. IV. Correlates of
diabetic background and proliferative
retinopathy. Am J Epidemiol 1991;
133:381391.
30. DAntonio JA, Ellis D, Doft BH, et al:
Diabetes complications and glycemic
control. The Pittsburgh prospective insulindependent diabetes cohort study status
report after 5 yr of IDDM. Diabetes Care
1989; 12:694697.
31. Nelson RG, Wolfe JA, Horton MB, et al:
Proliferative retinopathy in NIDDM.
Incidence and risk factors in Pima Indians.
Diabetes 1989; 38:435440.
32. Engerman RL, Bloodworth JMB, Nelson SL:
Relationship of microvascular disease in
diabetes to metabolic control. Diabetes
1977; 26:760769.
33. The Kroc Collaborative Study Group: Blood
glucose control and the evolution of
diabetic retinopathy and albuminuria: a
preliminary multi-center trail. N Eng J Med
1984; 311:365372.
34. Lauritzen T, Frost-Larsen K, Larsen HW,
et al: Effect on one year of near-normal
blood glucose levels on retinopathy in
insulin-dependent diabetics. Lancet 1983;
1:200204.
35. Dahl-Jorgensen K, Brinchmann-Hansen O,
Hansen KF, et al: Rapid tightening of blood
glucose control leads to transient
deterioration of retinopathy in insulindependent diabetes mellitus: the Oslo
Study. Br Med J 1985; 290:811815.
36. Diabetes Control and Complications Trial
Research Group: The effect of intensive
treatment of diabetes on the development
and progression of long-term complications
in insulin-dependent diabetes mellitus.
N Engl J Med 1993; 329:977-986.
36a. Nathan DM, Cleary PA, Backlund JY, et al:
Intensive diabetes treatment and
cardiovascular disease in patients with
type 1 diabetes. N Engl J Med 2005;
353:26432653.
37. The Diabetes Control and Complications
Trial/Epidemiology of Diabetes Intervention
CHAPTER 37
REFERENCES
409
38.
39.
40.
41.
42.
43.
SECTION 5
44.
45.
46.
47.
48.
49.
50.
410
and Complications Research Group:

Retinopathy and nephropathy in patients
with type 1 diabetes four years after a trial
of intensive therapy. N Engl J Med 2000;
342:381389.
Trial/Epidemiology of Diabetes Intervention
and Complications Research Group. Effect
of intensive therapy on the microvascular
complications of type 1 diabetes mellitus.
JAMA 2002; 287:25632569.
Nathan DM, Cleary PA, Backlund JY, et al:
Intensive diabetes treatment and
cardiovascular disease in patients with type
1 diabetes. N Engl J Med 2005;
353:26432653.
Intensive blood-glucose control with
sulphonylureas or insulin compared with
conventional treatment and risk of
complications in patients with type 2
diabetes (UKPDS 33): UK Prospective
Diabetes Study (UKPDS) Group. Lancet
1998; 352:837853.
Klein R, Klein BE, Moss SE,
epidemiologic study of diabetic
retinopathy: XVII. The 14-year incidence
and progression of diabetic retinopathy and
associated risk factors in type 1 diabetes.
Ophthalmology 1998; 105:18011815.
Klein R, Klein BE, Moss SE, Cruickshanks KJ:
The Wisconsin epidemiologic study of
diabetic retinopathy. XV. The long-term
incidence of macular edema.
Tight blood pressure control and risk of
macrovascular and microvascular
complications in type 2 diabetes: UKPDS 38.
UK Prospective Diabetes Study Group.
BMJ 1998; 317:703713.
The United Kingdom Prospective Diabetes
Study Group: Tight blood pressure control
and risk of macrovascular and microvascular
complications in type 2 diabetes. United
Kingdom prospective diabetes study report
number 38. BMJ 1998; 317:703713.
The United Kingdom Prospective Diabetes
Study Group: efcacy of atenolol and
captopril in reducing risk of macrovascular
and microvascular complications in type 2
diabetes. United Kingdom prospective
diabetes study report number 39. BMJ
1998; 317:713720.
Chaturvedi N, Sjolie AK, Stephenson JM,
et al: Effect of lisinopril on progression of
retinopathy in normotensive people with
type 1 diabetes. The EUCLID Study Group.
EURODIAB controlled trial of lisinopril in
insulin-dependent diabetes mellitus. Lancet
1998; 351:2831.
Sjolie AK, Chaturvedi N, Fuller J: Effect of
lisinopril on progression of retinopathy and
microalbuminuria in normotensive subjects
with insulin-dependent diabetes mellitus.
Ugeskr Laeger 1999; 161:949952.
Cusick M, Chew EY, Hoogwerf B, et al:
Risk factors for renal replacement therapy
in the Early Treatment Diabetic Retinopathy
Study (ETDRS), early treatment diabetic
study report no. 26. Kidney Int 2004;
66:11731179.
The Microalbuminuria Collaborative Study
Group: Predictors of the development of
microalbuminuria in patients with type I
diabetes mellitus. A seven-year prospective
study. Diabet Med 1999; 16:918925.
Klein R, Klein BEK, Moss SE, et al: The
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
retinopathy: V. Proteinuria and retinopathy

in a population of diabetic persons
diagnosed prior to 30 years of age.
In: Freidman EA, LEsperance FA, eds.
Diabetic renal-retinal syndrome 3.
Orlando, FL: Grune & Stratton; 1986.
Mathiesen ER, Ronn B, Storm B, et al:
The natural course of microalbuminuria in
insulin-dependent diabetes: a 10 year
prospective study. Diabet Med 1995;
12:482487.
Klein R, Klein BEK, Moss SE,
epidemiologic study of diabetic retinopathy,
XVII: the 14 year incidence and progression
of diabetic retinopathy and associated risk
factors in type I diabetes. Ophthalmology
1998; 105:18011815.
Pradhan R, Fong D, March C, et al:
Angiotensin-converting enzyme inhibition
for the treatment of moderate to severe
diabetic retinopathy in normotensive type 2
diabetic patients: a pilot study. J Diabetes
Complications 2002; 16:377381.
Ravid M, Brosh D, Ravid-Safran D, et al:
Main risk factors for nephropathy in type 2
diabetes mellitus are plasma cholesterol
levels, mean blood pressure, and
hyperglycemia. Arch Intern Med 1998;
158:9981004.
Chowdhury TA, Hopkins D, Dodson PM,
Vadis GC: The role of serum lipids in
exudative diabetic maculopathy: is there a
place for lipid lowering therapy? Eye 2002;
16:689693.
Harold M, Marmion VJ, Gough KR:
A double-blind controlled trial of clobrate
in the treatment of diabetic retinopathy.
Diabetes 1969; 84:285291.
Cullen JF, Ireland JT, Oliver MF:
A controlled trial of atromid therapy in
exudative diabetic retinopathy. Trans
Duncan LJ, Cullen JF, Ireland JT: A threeyear trial atromid therapy in exudative
diabetic retinopathy. Diabetes 1968;
17:458467.
Klein BE, Moss SE, Klein R, Surawicz TS:
The Wisconsin epidemiologic study of
diabetic retinopathy. XIII. Relationship of
serum cholesterol to retinopathy and hard
exudate. Ophthalmology 1991;
98:12611265.
Chew EY, Klein ML, Ferris FL 3rd, et al:
Association of elevated serum lipid levels
with retinal hard exudate in diabetic
retinopathy. Early treatment diabetic
retinopathy study (ETDRS) report 22.
Fong DS, Segal PP, Myers F, et al:
Subretinal brosis in diabetic macular
edema. ETDRS report 23. Early Treatment
Diabetic Retinopathy Study Research Group.
Alibrahim E, Donaghue KC, Rogers S, et al:
Retinal vascular caliber and risk of
retinopathy in young patients with type 1
diabetes. Ophthalmology 2006;
113:14991503.
Klein R, Klein BE, Moss SE, et al: Retinal
vascular caliber in persons with type 2
diabetes: the Wisconsin epidemiological
study of diabetic retinopathy: XX.
Ophthalmology 2006; 113:14881498.
Fujimoto WY, Leonetti DL, Kinyoun JL, et al:
Prevalence of complications among
second-generation Japanese-American
men with diabetes, impaired glucose
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
tolerance, or normal glucose tolerance.

Diabetes 1987; 36:730739.
Leske MC, Wu SY, Hennis A, et al: Nineyear incidence of diabetic retinopathy in
the Barbados eye studies. Arch Ophthalmol
2006; 124:250255.
Wong TY, Klein R, Islam FM, et al: Diabetic
retinopathy in a multi-ethnic-cohort in the
United States. Am J Ophthalmol 2006;
141:446455.
Leslie RD, Pyke DA: Diabetic retinopathy in
identical twins. Diabetes 1982; 31:1921.
Tattersall RB, Pyke DA: Diabetes in
identical twins. Lancet 1972; 2:11201125.
Rand LT, Krolewski AS, Aiello LM, et al:
Multiple factors in the prediction of risk of
proliferative diabetic retinopathy. N Engl J
Med 1985; 113:14331438.
Cove DH, Walker JM, Mackintosh P, et al:
Are HLA types or Bf alleles markers for
diabetic retinopathy? Diabetologica 1980;
19:402403.
Jervell J, Solheim B: HLA-antigens in long
standing insulin dependent diabetics with
terminal nephropathy and retinopathy with
and without loss of vision. Diabetologia
1979; 17:391.
Dibble CM, Kochenour NK, Worley RJ,
et al: Effect of pregnancy on diabetic
retinopathy. Obstet Gynecol 1982;
59:699704.
Soubrane G, Canivet J, Coscas G:
Influence of pregnancy on the evolution of
background retinopathy. Int Ophthalmol
1985; 8:249255.
Moloney JBM, Drury IM: The effect of
pregnancy on the natural course of diabetic
retinopathy. Am J Ophthlamol 1982;
93:745756.
Rodman HM, Singerman LJ, Aiello LM,
et al: Diabetic retinopathy and its
relationship to pregnancy. In: Merkatz ER,
Adams PAJ, eds. The diabetic pregnancy
a perinatal perspective. New York: Grune &
Stratton; 1979:73-91.
Klein BEK, Moss SE, Klein R: Effect of
pregnancy on progression of diabetic
retinopathy. Diabetes Care 1990; 12:3440.
Phelps RK, Sakol P, Metzger BE, et al:
Changes in diabetic retinopathy during
pregnancy correlates with regulation o
fhyperglycemia. Arch Ophthalmol 1986;
104:18061810.
Chew EY, Metzger BE, Mills JL, et al:
Metabolic control and progression of
retinopathy: the diabetes in early
preganancy study: national institute of child
health and human development. Diabetes
in Early Pregnancy Study. Diabetes Care
1993; 18:631637.
Trial Research Group: Effect of pregnancy
on microavscular complications of Diabetes
Control and Complications Trial. Diabetes
Care 2000; 23:10841091.
Young RJ, McCulloch DK, Prescott RJ,
Clarke BF: Alcohol: another risk factor for
diabetic retinopathy? Br Med J 1984;
288:10351037.
Hanna AK, Roy M, Zinman B, et al:
An evaluation of factors associated with
proliferative diabetic retinopathy. Clin Invest
Med 1985; 8:109116.
Haffner SM, Hazuda HP, Stern MP, et al:
Effect of socioeconomic status on
hyperglycemia and retinopathy levels in
Mexican Americans with NIDDM. Diabetes
Care 1989; 12:128134.
83. The Diabetic Retinopathy Study Research

Group: Preliminary report on effects of
photocoagulation therapy. Am J
84. DRS Research Group: Four risk factors for
severe visual loss in diabetic retinopathy.
85. Early Treatment Diabetic Retinopathy Study
Research Group: Photocoagulation for
diabetic macular edema. ETDRS report
number 1. Arch Opthalmol 1985;
103:17961806.
86. DRVS Research Group: Early vitrectomy for
severe vitreous hemorrhage in diabetic
retinopathy. Four year results of a
randomized trial. DRVS report 5. Arch
Ophthalmol 1990; 108:958964.
87. Jirousek MR, Gillig JR, Gonzalez CM, et al:
(S)-13-[(dimethylamino)methyl]10,11,14,15-tetrahydro-4,9:16, 21dimetheno-1H,
13H-dibenzo[e,k]pyrrolo[3,4-h][1,4,13]
oxadiazacyclohexadecene-1,3(2H)-d ione
(LY333531) and related analogues: isozyme
selective inhibitors of protein kinase C
beta. J Med Chem 1996; 39:26642671.
88. Ishii H, Jirousek MR, Koya D, et al:
Amelioration of vascular dysfunctions in
diabetic rats by an oral PKC beta inhibitor.
Science 1996; 272:728731.
89. Aiello LP, Bursell SE, Clermont A, et al:
Vascular endothelial growth factor-induced
retinal permeability is mediated by protein
kinase C in vivo and suppressed by an
orally effective beta-isoform-selective
inhibitor. Diabetes 1997; 46:14731480.
90. Danis RP, Bingaman DP, Jirousek M, Yang Y:
Inhibition of intraocular neovascularization
caused by retinal ischemia in pigs by
PKCbeta inhibition with LY333531. Invest
91. The PKC-DRS Study Group: The effect of
ruboxistaurin on visual loss in patients with
moderately severe to very severe
nonproliferative diabetic retinopathy initial
results of the Protein Kinase C beta
92.
92a.
93.
94.
95.
96.
97.
98.
99.
Inhibitor Diabetic Retinopathy Study (PKCDRS) multicenter randomized clincal trial.

Diabetes 2005; 54:21882197.
Aiello LP, Clermon A, Arora V, et al:
Inhibition of PKC b by oral administration of
Ruboxistaurin is well tolerated and
ameliorates diabetes-induced retinal
hemodynamic abnormalities in patients.
PKCDMES Study Group. Effect of
ruboxistaurin in patients with diabetic
macular edema: thirty-month results of the
randomized PKCDMES clinical trial. Arch
Ophthalmol 2007; 125:318324.
Nauck M, Roth M, Tamm M, et al: Induction
of vascular endothelial growth factor by
platelet-activating factor and plateletderived growth factor is downregulated by
corticosteroids. Am J Resp Cell Mol Biol
1997; 16:398406.
Nauck M, Karakiulakis G, Perruchoud A,
et al: Corticosteroids inhibit the expression
of the vascular endothelial growth factor
gene in human vascular smooth muscle
cells. Euro J Pharmacol 1998;
341:309315.
Martidis A, Duker J, Greenberg P, et al:
Intravitreal triamcinolone for refractory
2002; 109:920927.
Jonas J, Kreissig I, Sofker A, Degenring R:
Intravitreal injection of triamcinolone for
diffuse diabetic macular edema. Arch
Ophthalmol 2003; 121:5761.
Diabetic Retinopathy Clinical Research
Network. Available: http://www.drcr.net.
The Eyetech Study Group: Preclinical and
phase 1A clinical evaluation of an antiVEGF pegylated aptamer (EYE001) for the
treatment of exudative age-related macular
Nguyen QD, Tatlipinar S, Shah SM, et al:
critical stimulus for diabetic macular
edema. Am J Ophthalmol 2006;
142:961969. Epub 2006 Aug 2.
100. Chun DW, Heier JS, Topping TM, et al:

A pilot study of multiple intravitreal
injections of ranibizumab in patients with
center-involving clinically signicant
2006; 113:17061712.
101. Avery RL: Regression of retinal and iris
neovascularization after intravitreal
bevacizumab (Avastin) treatment. Retina
2006; 26:352354.
102. Hikichi T, Fujio N, Akiba J, et al:
Association between the short-term natural
history of diabetic macular edema and the
vitreomacular relationship in type II
diabetes mellitus. Ophthalmology 1997;
104:473478.
103. Schepens CL, Avila MP, Trempe CC: Role
of the vitreous in cystoid macular edema.
104. Nasarallah FP, Jalkh AE, Van Coppenole F,
et al: The role of vitreous in diabetic
macular edema. Ophthalmology 1988;
95:13351339.
105. Lewis H, Abrams GW, Blumenkranz MS,
Campo RV: Vitrectomy for diabetic macular
traction and edema associated with
posterior hyaloidal traction. Ophthalmology
1992; 99:753759.
106. Harbour JW, Smiddy WE, Flynn HW Jr,
Rubsamen PE: Vitrectomy for diabetic
macular edema associated with a
thickened and taut posterior hyaloid
membrane. Am J Ophthalmol 1996;
121:405443.
107. Ikeda T, Sato K, Katano T, Hayashi Y:
Vitrectomy for cystoid macular oedema
with attached posterior hyaloid membrane
in patients with diabetes. Br J Ophthalmol
1999; 83:1214.
108. Tachi N, Ogino N: Vitrectomy for diffuse
macular edema in cases of diabetic
retinopathy. Am J Ophthalmol 1996;
122:258260.
109. Tachi N: Surgical management of macular
edema. Semin Ophthalmol 1998; 13:2030.
CHAPTER 37
411
CHAPTER
38
Epidemiology of Age-related Macular Degeneration

Johanna M. Seddon and Lucia Sobrin
Age-related macular degeneration (AMD) is the leading cause of

irreversible blindness in older individuals in all developed
countries around the world.1,2 It can adversely affect activities of
daily living, rendering it more difcult or impossible to read,
write, and drive, and thus forcing many individuals in their retirement years to lose their independence. The prevalence of AMD
is increasing as the proportion of our elderly population rises.
The dry or nonexudative forms of this disease, which comprise about 85% of the cases, are generally not reversible,
although rates of progression to more advanced disease can be
altered. For the remaining 15% of cases with neovascular or wet
disease, new and more effective treatment strategies have been
developed, some of which can improve vision. The established
demographic risk factors including increasing age and a family
history of the disease are not modiable. Possible methods to
decrease the incidence of this disease to date are to refrain from
cigarette smoking, maintain a normal weight, and get adequate
exercise.36 Nutritional factors and eating a healthy diet are also
important.710 A multivitamin/mineral supplement reduces risk
of progression by 25% over ve years, for individuals with
intermediate disease or advanced disease in one eye. Foods rich
in lutein and zeaxanthin and omega-3 fatty acids are also potentially benecial. Genetic variants which increase susceptibility
to AMD have been reported.
This chapter reviews the classication and denition of
macular degeneration, its frequency, and the known and
potential environmental and genetic factors associated with the
occurrence of this prevalent condition.
CLASSIFICATION AND DEFINITION

Macular degenerative changes have typically been classied into
two clinical forms: dry or wet; and the latter form is also called
exudative. Both types can lead to visual loss. In the dry form,
visual loss is usually gradual. Ophthalmoscopy reveals yellow,
subretinal deposits called drusen or retinal pigment irregularities
including hyperpigmentation or hypopigmentation changes.
Each of these signs can be further subdivided into various
categories according to the number and size of the lesions.
Drusen, which become confluent, can evolve into drusenoid
retinal pigment epithelial detachments; many of these lesions
progress to geographic atrophy. Geographic atrophy can involve
both the central and non-central regions of the macula. In the
wet or exudative form, vision loss can appear to occur suddenly,
when a choroidal neovascular membrane leaks fluid or blood
into the subpigment epithelial or subretinal space. Serous
retinal pigment epithelial detachments often, but not always,
advance to the neovascular stage. Late AMD includes two
advanced forms of AMD usually associated with visual loss:
geographic atrophy and neovascular disease. This phenotypic
heterogeneity has led to the use of various denitions of AMD,

and, as a result, difculties with comparisons among studies.
It is important for investigators to standardize denitions of
a disease and its subtypes in order to enhance comparability and
to promote collaborative efforts.11 The Age-Related Eye Disease
Study (AREDS) enrolled participants into four groups ranging
from no disease (Category 1) to late stages (Category 4), which
included central atrophy based on macular appearance and
visual acuity, neovascular disease, as well as AMD due to visual
loss without these late stages.12 The Clinical Age-Related
Maculopathy Grading System (CARMS) classies individuals
into ve groups, with Grade 4 dened as central or non-central
geographic atrophy, and Grade 5 as neovascular disease, and
there are no visual acuity criteria.13 The classication of agerelated maculopathy will no doubt change in the future as
genetic and epidemiologic studies provide further insight into
the pathogenesis of this disease, and subcategories of AMD are
better dened.
PREVALENCE
Several population-based studies have provided information on
the prevalence of AMD: The National Health and Nutrition
Examination Survey (NHANES)14,15 the Framingham Eye Study
(FES),16 the Chesapeake Bay Watermen Study,17 the Beaver Dam
Eye Study (BDES),18 as well as studies outside the United States
including the Rotterdam Study in the Netherlands,19 and the
Blue Mountains Study in Australia.20 Prevalence rates are
quite variable for all types of AMD combined, because of differences in denitions of AMD, but are more consistent for
advanced AMD.
The BDES18 found that the early forms are much more
common than the late stages of ARM, and both types increase
in frequency with increasing age. The prevalence of late ARM
was 1.6% overall and the prevalence of late ARM rose to 7.1%
in persons who were 75 or older. Total prevalence of AMD in
the USA was also estimated in 2004 using pooled ndings from
seven large population-based studies both inside and outside
the USA, and applying those prevalence rates to the US
population.21 This meta-analysis by the Eye Diseases
Prevalence Group calculated the overall prevalence of
neovascular AMD and/or geographic atrophy to be 1.47% of the
US population aged 40 years or older. This is more than 1.75
million individuals affected with advanced AMD in the USA,
with an estimated increase of 50% to 2.95 million by 2020.
INCIDENCE
Incidence rates have been estimated in various populations.22,23
The FES used the age-specic prevalence data to estimate 5-year
413
PRINCIPLES OF EPIDERMIOLOGY
incidence rates of AMD, according to the denition of AMD in
that study. These estimates were 2.5%, 6.7%, and 10.8% for
individuals who were 65, 70, and 75 years of age, respectively.22
The BDES determined the 5-year cumulative incidence of
developing early and late AMD in a population of 3583 adults
(age range 43 to 86 years).23 Incidence of early AMD increased
from 3.9% in individuals aged 43 to 54 years to 22.8% in
persons 75 years of age and older. The overall 5-year incidence
of late AMD was 0.9%. Persons 75 years of age or older had a
5.4% incidence rate of late AMD. More recently, the
Copenhagen City Eye Study found the 14-year incidences of
early and late AMD to be 31.5% and 14.8%, respectively, in 946
patients who were 60 to 80 years old at baseline.24
PSYCHOSOCIAL IMPACT
Patients with visual loss due to AMD and other medical
problems often report AMD as their worst medical problem and
have a diminished quality of life.25 In a study of well-being,
patients with AMD had lower scores than did those with
chronic obstructive pulmonary disease and acquired immune
deciency syndrome.26 The AREDS study showed that
progression to advanced AMD had a signicant impact on
vision-related quality of life, as did loss in vision of at least three
lines.27 The largest impact on vision-targeted quality of life
occurred in persons who lost vision in both eyes. Such an
impact on the patients psychosocial well-being and activities of
daily living underscores the growing importance of this disease
on the expanding elderly population.
SOCIODEMOGRAPHIC RISK FACTORS
stages were too low to make statistical comparisons. In the

Baltimore Eye Survey, AMD accounted for 30% of bilateral
blindness among whites and for 0% among blacks.28 Data from
a population-based study in Barbados, West Indies,29 revealed
that age-related macular changes occurred commonly but at a
lower frequency than in predominantly white populations in
other studies. The prevalence of ARM was also compared by
geographic region and ethnicity in Southern Colorado and
Central Wisconsin.30 Late stage AMD was signicantly less
frequent among Hispanics in Colorado compared with nonHispanic whites in Beaver Dam (odds ratio [OR] of 0.07; 95%
condence interval [CI] of 0.01 to 0.49). Overall, the literature
to date suggests that early ARM in blacks and Hispanics is less
common than among non-Hispanic whites, and advanced stages
of AMD are much less frequent in these groups compared with
non-Hispanic whites.
EDUCATION AND SOCIOECONOMIC STATUS

Persons with higher levels of education had a slightly reduced
risk of neovascular AMD in the Eye Disease Case-Control
Study (EDCCS), but the association did not remain statistically
signicant after multivariate modeling.31 In the Beaver Dam
Eye Study, no association was found between education,
income, employment status, marital status, and the incidence
of maculopathy.32 Furthermore, no associations were noted in
another case-control study33 or in the FES,16 although different
denitions of macular degeneration were used in those reports.
Education was inversely related to AMD in a case-control study
within the AREDS population.34
OCULAR RISK FACTORS
SECTION 5
AGE
All studies demonstrate that the prevalence, incidence, and
progression of all forms of AMD rise steeply with increasing
age.1618,23 By 75 years of age and older, 7.1% of the population
have late age-related maculopathy (ARM) or AMD compared
with 0.1% in the age group of 43 to 54 years and 0.6% among
people aged 55 to 64.
SEX
In the Beaver Dam Eye Study, while controlling for age, there was
no overall difference in the frequency of AMD between men and
women.18 However, exudative macular degeneration was more
frequent in women 75 years or older compared with men of that
age (6.7% vs. 2.6%, P = .02).18 A similar nding was observed in
the FES.16 In NHANES III, men, regardless of race and age, had
a lower prevalence of AMD than did women.15 Incidence rates
within the Beaver Dam population also suggest a gender
difference. After adjusting for age, women 75 years of age or older
had approximately twice the incidence of early ARM compared
with men of that age.23 In the Blue Mountains Eye Study, there
were consistent, although not signicant, sex differences in
prevalence for most lesions of ARM, with women having higher
rates for AMD and soft indistinct drusen but not for retinal
pigmentary abnormalities.20 No gender differences were seen in
the Rotterdam Study.19 Residual confounding by age in the broad
age category 75 and older may partially explain these
differences. However, true gender differences may exist, and
further research is needed to conrm and expand these ndings.
RACE/ETHNICITY
414
In the NHANES III study,15 a higher frequency of early ARM

was reported in whites compared with blacks. Rates for late
IRIS COLOR
Investigators have postulated that higher levels of ocular
melanin may be protective against light-induced oxidative
damage to the retina. To date, the literature is inconclusive
about the relationship between iris color and AMD. Darker
irides have been found to be protective in some33,3536 but not
other studies.31,3739 Differences in studies may be related partly
to the use of different denitions of disease, different number
and types of other risk or protective factors evaluated, and
residual confounding by ethnicity in some studies.
REFRACTIVE ERROR
Several case-control studies have shown an association between
AMD and hyperopia.31,33 The potential problem with some of
these studies is the setting (ophthalmology practices) in which
they were conducted. Because ophthalmology practices tend to
contain higher percentages of myopic patients, controls selected
from such practices would tend to have a higher prevalence of
myopia than would the general population. However, the
population-based Rotterdam Study also showed an association
between hyperopia and both incident and prevalent ARM.40
This association therefore might implicate structural and
mechanical differences which render some eyes predisposed to
maculopathy.
CUP:DISC RATIO
The EDCCS demonstrated that eyes with larger cup:disc ratios
had a reduced risk of exudative AMD. This effect persisted even
after multivariate modeling,31 adjusting for known and
potential confounding factors. Whether this nding, which is
consistent with the association with hyperopic refractive error
Epidemiology of Age-Related Macular Degeneration
LENS OPACITIES
The literature has not shown a consistent relationship between
the presence of cataract and AMD. FES investigators found no
relationship,41 whereas data from the NHANES Study did
support a relationship between AMD and lens opacities.42 In
the Beaver Dam Eye Study, nuclear sclerosis was associated
with increased odds of early ARM (OR of 1.96; 95% CI of 1.3 to
3), but not of late ARM. 43
On the other hand, investigators have postulated that
cataract surgery may increase the risk for AMD, perhaps
because the cataractous lens can block damaging ultraviolet
light. Inflammatory changes after cataract surgery may also
increase risk of progression of early to late ARM. In the
NHANES, aphakia was associated with a twofold increased risk
of AMD (OR of 2; 95% CI of 1.44 to 2.78).14 Another study
evaluated 47 patients with bilateral, symmetric, early AMD,
who underwent extracapsular cataract extraction with
intraocular lens implantation in one eye. Progression of AMD
occurred more often in the surgical eyes compared with the
fellow eyes.44 In the Beaver Dam Eye Study, previous cataract
surgery at baseline was associated with a statistically signicant
increased risk of development of late ARM (OR of 2.8; 95% CI
of 1.03 to 7.6).37 The risk for late ARM with a history of
cataract surgery at baseline persisted at the ten year follow-up,
with a risk ratio of 3.81 (95% CI 1.897.69).45 The Blue
Mountains Eye Study found a three fold risk in the 10-year
incidence of late stage ARM in nonphakic eyes when compared
with phakic eyes.46 This effect also persisted in an analysis of
pooled ndings from the BDES and the Blue Mountains Eye
Study.47 A study of postmortem eyes was suggestive of an
increase in disciform scars in eyes with cataract extraction and
implantation of an intraocular lens.48
However, an analysis of AREDS data showed that there
was no correlation between cataract surgery and advanced
neovascular AMD, although there was a small risk of
advancement of geographic atrophy.49 For geographic atrophy,
cataract surgery was associated with a 50% increased risk, and
the effect was marginally statistically signicant (RR of 1.47, CI
of 0.99 to 2.17). Further matched-pair analyses were undertaken to validate these initial ndings comparing 342 patients
with intermediate or advanced AMD who had cataract surgery
after study entry and before developing AMD and those who did
not have surgery. There was no increased risk of progression to
neovascular AMD among those who had cataract surgery.
ENVIRONMENTAL, MEDICAL, AND

NUTRITIONAL FACTORS
SMOKING
The preponderance of epidemiologic evidence indicates a strong
positive association between both wet and dry AMD and
smoking. Two large prospective cohort studies have evaluated
the relationship between smoking and wet AMD and dry AMD
associated with visual loss.3,50 In the Nurses Health Study,
women who currently smoked 25 or more cigarettes per day had
a relative risk (RR) of 2.4 (95% CI of 1.4 to 4), and women who
were past smokers had an RR of 2.0 (95% CI of 1.2 to 3.4)
for AMD compared with women who never smoked.3 Risk
increased as pack-years of smoking increased indicating a dosedependent relationship. Risk for AMD remained elevated for
many years after smoking cessation. Results were consistent for
various denitions of AMD, including wet AMD, dry AMD
with different levels of visual loss, and for different denitions

of smoking. Among women, it was estimated that 29% of the
AMD cases in that study could be attributed to smoking.3 These
results were supported by a study among men participating in
the Physicians Health Study,50 suggesting that smoking is an
important, independent, avoidable cause of AMD.
Recently reported pooled data on 9523 adults from three
populations living in Australia, The Netherlands, and the
United States support the body of evidence indicating that
smoking is related to an increased risk of incident AMD.4 In the
US Twin Study of Age-Related Macular Degeneration current
smokers had a 1.9-fold increased risk of AMD while past
smokers had about a 1.7-fold increased risk.5
Mechanisms by which smoking may increase risk of developing macular degeneration include its adverse effect on blood
lipids by decreasing levels of HDL and increasing platelet
aggregation and brinogen, increasing oxidative stress and lipid
peroxidation, and reducing plasma levels of antioxidants.3
BODY MASS INDEX

Evidence for a role of increased body mass index (BMI) on the
development and progression of AMD is growing. In one
prospective cohort study of the rate of progression to advanced
AMD, higher BMI increased the risk for progression to the
advanced forms of AMD.6 Relative risk was 2.35 (95%
condence interval [CI], 1.27-4.34) for a body mass index of at
least 30, and 2.32 (95% CI, 1.32-4.07) for a body mass index of
25 to 29, relative to the lowest category (<25) after controlling
for other factors (P = .007 for trend). In that study there was
also about a two-fold increased risk for progression to advanced
AMD for abdominal obesity as measured by both waist
circumference and waist-hip ratio.6 An observational analysis of
a randomized clinical trial found that a signicant association
between late AMD and greater body mass index (1.05 per
1 kg/m, 1.001 to 1.10, P = .05)].51 In AREDS, greater BMI was
signicantly associated the incidence of central geographic
atrophy (OR, obese vs. nonobese, 1.93; 95% CI, 1.25-2.65).52 In
the Physicians Health Study, the relationship of BMI with dry
ARM was J-shaped, with both the leanest individuals and obese
individuals at increased risk.53 In a French population-based
study, obese subjects had a 2.29-fold (CI: 1.00-5.23) and 1.54fold (CI: 1.05-2.26) increased risk of late AMD and pigmentary
abnormalities in comparison with lean subjects.54
CHAPTER 38
mentioned earlier, is meaningful in terms of the mechanisms

associated with the development of AMD, awaits further study.
CARDIOVASCULAR DISEASES
AMD and cardiovascular disease may have common
antecedents.55 The presence of atherosclerotic lesions,
determined by ultrasound, was examined in relation to risk of
macular degeneration in a large population-based study
conducted in the Netherlands.56 Results obtained from this
cross-sectional study showed a 4.5-fold increased risk of late
macular degeneration (dened as geographic atrophy or
neovascular macular degeneration as determined by grading of
fundus photographs) associated with plaques in the carotid
bifurcation and a twofold increased risk associated with plaques
in the common carotid artery. Lower-extremity arterial disease
(as measured by the ratio of the systolic blood pressure level of
the ankle compared with the arm) was also associated with a
2.5 times increased risk of AMD. In addition, a case-control
study found a relationship between AMD and history of one or
more cardiovascular diseases.33 The NHANES-I study reported
a positive association between AMD and cerebrovascular
disease, but positive associations with other vascular diseases
did not reach statistical signicance.57 A Finnish study reported
a signicant correlation between the occurrence of AMD and
415
the severity of retinal arteriosclerosis.58 However, some studies
found that persons who reported a history of CVD did not have
a signicantly greater risk of AMD.31,58,59
BLOOD PRESSURE AND HYPERTENSION

The role of blood pressure in the etiology of AMD remains
unclear. There was a small, statistically signicant relationship
between AMD and systemic hypertension in two crosssectional population-based studies.57,60 One case-control study
found that persons with AMD were signicantly more likely to
be taking antihypertensive medication.61 Also, a signicant
relationship was found between AMD and diastolic blood
pressure measured several years before the eye examination
in the FES.62 The Beaver Dam Study reported that systolic
blood pressure was associated with incidence of RPE
depigmentation.59 In the Macular Photocoagulation Study,
there was an increased incidence of exudative AMD associated
with hypertension, in the second eye of individuals with
exudative AMD in one eye at baseline (relative risk of 1.7; 95%
CI of 1.2 to 2.4).63
Cross-sectional56,58 and case-control studies,31 as well as one
prospective study59 in which duration of hypertension was not
taken into account, did not show an increased risk of late AMD
associated with current hypertension or systolic or diastolic
blood pressure. However, in the EDCCS, a trend for an
increased risk associated with higher systolic blood pressure
was evident.31 Evidence suggests a possible mild to moderate
association between elevated blood pressure and AMD.
SECTION 5
CHOLESTEROL LEVELS
There is some evidence linking cholesterol level to AMD, but
not all results are consistent. The EDCCS reported a statistically
signicant four-fold increased risk of exudative AMD associated
with the highest serum cholesterol level (>4.88 mmol/L), and a
twofold increased risk in the middle cholesterol level group,
compared with the lowest cholesterol level group, controlling
for other factors.31 No signicant association was noted between
AMD and cholesterol level in the FES.62 A study of plasma
cholesterol and fatty acid levels found no difference between
65 cases of exudative AMD and control pairs.64 The Beaver
Dam Study found that early AMD was related to low total
serum cholesterol levels in women and men older than age 75.
Furthermore, men with early AMD had higher high-density
lipoprotein-cholesterol (HDL-C) and lower total cholesterol/
HDL-C ratios.59,65 Slightly, but not signicantly, increased risk
of wet AMD was seen with increasing triglyceride level in the
EDCCS,31 but this nding was not conrmed in the Rotterdam
Study56 or the Beaver Dam Study59 (both of which had small
numbers of exudative AMD cases and therefore limited power).
In a case-control study to assess the risk of AMD in patients
who were taking statins, short term and medium term statin
use was not associated with a decreased risk of AMD.66
DIETARY FAT INTAKE
416
Dietary fat intake was associated with a slightly elevated risk of

exudative AMD in the Dietary Ancillary Study of the EDCCS.
This association was primarily due to vegetable fat rather than
animal fat. For omega-3 fatty acid intake, an inverse association, or a protective effect with higher intake, was found in the
multivariate model controlling for other factors.67,68 A prospective study of dietary fat intake and AMD found that total
fat intake was positively associated with AMD.69 A high intake
of sh was associated with a 35% lower risk of AMD (risk ratio
of 0.65, 95% CI of 0.46 -0.91). In the Beaver Dam Study,
persons in the highest quintile of saturated fat and cholesterol

intake compared with the lowest quintile had 80% and 60%
increased risk, respectively, for early AMD.70
A prospective cohort study also supported the role of dietary
lipids. Higher total fat intake increased the risk of progression to
the advanced forms of AMD, with a risk ratio of 2.9.71 Saturated,
monounsaturated, polyunsaturated, and transunsaturated fats
increased the likelihood of progression for the highest fat-intake
quartile relative compared to the lowest fat-intake quartile, after
controlling for other factors Higher sh intake was associated
with a lower risk of AMD progression among subjects. Similar
results regarding sh intake were seen in a twin study.5 Increased
intake of sh reduced risk of AMD, particularly for two or more
servings per week. Dietary omega-3 fatty intake was inversely
associated with AMD comparing the highest vs lowest quartile.
Reduction in risk of AMD with higher intake of omega-3 fatty
acids was seen primarily among subjects with low levels (below
median) of linoleic acid intake, an omega-6 fatty acid.
There is consistent evidence that omega-3 fatty acids may
reduce risk of AMD from multiple studies with different designs
and different study populations. AREDS 2 will test this
hypothesis in a randomized trial.
DIABETES AND HYPERGLYCEMIA

Many studies have investigated the relationship between
diabetes and/or hyperglycemia and AMD, and most have found
no signicant relationships.31,33,58,62,63 The Beaver Dam Study
found no overall association between early or late AMD and
diabetes or glycosylated hemoglobin, a measure of glycemia,
although a positive association was found between glycosylated
hemoglobin and exudative AMD only in older men. However,
sample sizes in these subgroup analyses were very small.72
Based on the scant literature to date, the association between
hyperglycemia or diabetes and AMD is uncertain. Difculties
with these studies include the uncertainty of diagnosing AMD
in the presence of diabetic retinopathy and many studies of
AMD exclude persons with diabetic retinopathy.
REPRODUCTIVE AND RELATED FACTORS

The EDCCS showed a marked decrease in the risk of
neovascular AMD among postmenopausal women who used
estrogen therapy.31 The odds of neovascular AMD were 0.3
(95% CI of 0.1 to 0.8) in current users of estrogen therapy.
Former use of estrogen therapy was also associated with reduced
risk (OR of 0.6; 95% CI of 0.3 to 1). In an ancillary study to the
Womens Health Initiative, 4262 women sixty-ve years and
older were randomized to treatment with conjugated equine
estrogens (CEE), CEE with progestin, and placebo.73 Treatment
with CEE alone or CEE and progestin did not affect early or late
stage AMD. They did nd that the treatment with CEE and
progestin may reduce the risk of soft drusen or neovascular
AMD. Snow et. al. found that women with age-related
maculopathy (ARM) who had used postmenopausal estrogen
therapy in the past had signicantly lower odds of advanced
ARM than nonusers, after controlling for other risk factors (OR
of 0.5, 95% CI of 0.30 to 0.98).74 No relationship was found in
the Beaver Dam Study between years of estrogen therapy and
exudative AMD.75 However, there were few cases of late AMD
in that study (n = 49). The Blue Mountains Eye Study reported
no relationship between AMD and hormone replacement
therapy or early menopause, although there was a small
decrease in risk of early ARM with increasing number of years
between menarche and menopause.76
A nested case-control study within the Rotterdam Study 77
showed that risk of AMD was almost twice that among women
who had undergone menopause before 45 years of age compared

with those who had their menopause at 45 years of age or later
(OR of 1.9; 95% CI of 1 to 3.8). A protective effect of estrogen
on AMD cannot be ruled out, and further research is warranted.
SUNLIGHT
The literature to date regarding the association between
sunlight exposure and AMD is conflicting. Overall, the data do
not support a strong association between ultraviolet (UV)
radiation exposure and risk of AMD, although a small effect as
well as an adverse effect of blue light exposure is possible.
In a study of 838 Maryland Watermen,78 sunlight exposure
was assessed by detailed interview and eld measurements. A
modest, positive relationship between blue light or visible light
exposure over the preceding 20 years and risk of advanced AMD
was seen, with an odds ratio of 1.36 (95% CI of 1 to 1.85) for
each 0.1 increase in Maryland Sun-Years. No adverse effects
were observed for UV-A or UV-B exposure. However, only eight
men had advanced AMD (geographic atrophy or exudative
disease). In the Beaver Dam Eye Study,79 no relationship was
seen between advanced AMD or early ARM and UV-B exposure,
but the effects of UV-A or blue light were not assessed. A
twofold increased risk of advanced AMD was associated with
increased time spent outdoors in the summer. With ten years of
follow-up, participants who experienced more than ten severe
sunburns during their youth were more likely than those who
experienced one or no burns to develop drusen with a 250micron diameter or larger.80 The risk ratio was 2.52 (95% CI of
1.294.94).
The EDCCS also evaluated crude measures of sunlight
exposure.31 No association was seen between exudative AMD
and leisure time spent outdoors in summer. Advanced AMD
was not associated with leisure time spent outdoors in the
winter, occupational sunlight exposure, or the use of sunglasses
or hats with brims. In an Australian case-control study,81 a
greater proportion of people with advanced AMD reported
higher sensitivity to sunburn compared with the control group.
The controls actually had greater median hours of sun exposure
than did the cases. The authors suggested that sun-sensitive
individuals may be at increased risk of AMD, although they
tend to avoid sun exposure.
Conflicting results in these studies exemplify the difculties
encountered with studying this complex exposure. These include
challenges in measurement of acute and chronic lifetime
exposure and the effect of potential confounding variables, such
as sun sensitivity and sun-avoidance behaviors.
ANTIOXIDANTS
Antioxidants including vitamin C (ascorbic acid); vitamin E
(alpha-tocopherol); and the carotenoids, including alphacarotene, beta-carotene, cryptoxanthin, lutein, and zeaxanthin
may be relevant to AMD due to their physiologic functions and
the location of some of these nutrients in the retina. Trace
minerals like zinc, selenium, copper, and manganese may also
be involved in antioxidant functions of the retina.82
Theoretically, antioxidants could prevent oxidative damage to
the retina, which could, in turn, prevent development of
AMD.83 Damage to retinal photoreceptor cells could be caused
by photo-oxidation or by free radical-induced lipid
peroxidation.84 This could lead to impaired function of the
retinal pigment epithelium and, eventually, to degeneration
involving the macula. The deposit of oxidized compounds in
healthy tissue may result in cell death because they are
indigestible by cellular enzymes.83 Antioxidants may scavenge,
decompose, or reduce the formation of harmful compounds.
In the Dietary Ancillary Study of the EDCCS, an inverse

association between exudative AMD and dietary intake of
carotenoids from foods was observed.7 A high intake of green
leafy vegetables containing the carotenoids lutein and
zeaxanthin was associated with a reduction in the risk of
exudative AMD. Intake of vitamin C was associated with a
small but nonsignicant reduction in risk. Intake of vitamin A
or vitamin E was not associated with a reduction in risk.7 In
this cohort, persons with higher serum levels of carotenoids
(sum of serum lutein/zeaxanthin, beta-carotene, alphacarotene, cryptoxanthin, and lycopene levels) had a greatly
reduced risk of exudative AMD.8 Persons with higher individual
serum levels of lutein/zeaxanthin, beta-carotene, alphacarotene, and crypotxanthin had reduced risks of exudative
AMD. The study did not nd a statistically signicant
protective effect for serum levels of vitamin C, vitamin E, or
selenium individually, but when these were combined into an
antioxidant index with carotenoids, there was a signicant
reduction in risk of exudative AMD with increasing levels of the
index.
A cross-sectional study using NHANES-I data examined the
relationship between the prevalence of any AMD and vitamins
A and C intake. A weak protective effect was seen with
increased consumption of fruits and vegetables rich in vitamin
A.57 The Beaver Dam Study found no effect of supplemental
antioxidant vitamins alone or in combination on risk of early or
late ARM.85 However, in a case-control study nested within that
study, a low serum level of one carotenoid, lycopene, was
associated with presence of any AMD.9 Another study reported
a protective effect for any AMD among those who had higher
serum vitamin E and among those who had higher values for an
antioxidant index of vitamins C, E, and beta-carotene, but no
protective effect was seen for vitamin supplementation.86 A
study of plasma levels of vitamins A and E and ve carotenoids
found no relationship with exudative AMD in 65 case-control
pairs.64 Increased blood levels of carotenoids and antioxidant
vitamins were also related to decreased risk of exudative AMD
in other reports.8788
Overall, results from observational studies suggest that diets
rich in antioxidant-rich fruits and vegetables are related to
a lower risk of exudative AMD. A prospective follow-up of
women in the Nurses Health Study and men in the Health
Professionals Follow-up Study found suggests a protective role
for fruit intake on the risk of neovascular ARM.89 The
Carotenoids in Age-Related Eye Disease Study (CAREDS) found
that a diet rich in lutein plus zeaxanthin may protect against
intermediate AMD in healthy women younger than 75 years
old.90 In the Rotterdam Study, a high dietary intake of beta
carotene, vitamins C and E, and zinc was associated with a
substantially reduced risk of AMD in elderly persons.91
A small randomized trial demonstrated less visual loss due to
AMD and less accumulation of drusen in the group of 97
patients assigned to high-dose zinc supplementation, compared
with 84 patients in the placebo group.92 However, another small
randomized trial found that zinc supplementation had no shortterm effect on the course of disease in 112 patients with wet
AMD.93 The Beaver Dam Study found a weak protective effect
of zinc intake on early ARM.85 The EDCCS did not nd any
signicant relationships between serum zinc levels or zinc
supplementation and risk of exudative AMD.31 A prospective
study of zinc intake, moderate zinc intake, either in food or in
supplements, was not associated with a reduced risk of AMD.94
A supplement containing antioxidant vitamins and minerals
has been shown to reduce the risk of AMD in a large
randomized clinical trial, the Age-Related Eye Disease Study,
sponsored by the National Eye Institute.10 In this study,
supplementation with a high dose of zinc plus antioxidant
CHAPTER 38
417
vitamins C and E and beta-carotene reduced the incidence and
progression of AMD, but not lens opacities. Based on evidence
from observational studies regarding the potential protective
effects of lutein7,88 and omega-3 fatty acids,5,6771 a new clinical
trial called AREDS 2 is underway to test supplements
containing these nutrients.
SECTION 5
ALCOHOL INTAKE
Studies that have examined the relationship between AMD
and alcohol consumption have yielded mixed results. In the
EDCCS, no signicant relationship between alcohol intake
and exudative AMD was noted in univariate analyses.31 In a
separate multivariate analysis, alcohol intake appeared to be
associated with a decreased risk of disease in the highest
quartile of intake compared with nondrinkers.95 Another casecontrol study found a nonsignicant association between
current daily alcohol intake and AMD, with a suggestion of a
nonlinear trend of higher risk of AMD in persons who had ve
drinks or more per day and a slightly lower risk in persons who
had one or two drinks per day compared with nondrinkers.96 In
a case-control study using NHANES-I data, moderate wine
consumption was associated with decreased risk of developing
AMD, although the analysis did not control for the potential
confounding effects of smoking.97
In population-based cross-sectional studies, there is conflicting
evidence for an association between alcohol and AMD. The Los
Angeles Latino Eye Study found heavy alcohol consumption,
particularly beer, was associated with a greater risk of having
advanced AMD (OR 2.9).98 The Beaver Dam Study found a
slightly increased risk for retinal pigment degeneration in persons
who consumed beer in the past year.99 With the ten years of
follow-up, they found that heavy drinking appears to be related to
an increased risk of late ARM, although the exposure and outcome
were infrequent and the effect was based on few exposed cases (RR
of 6.94; 95% CI of 1.85 to 26.1).100 Neither the Beaver Dam Study
nor the Blue Mountains Eye Study found an increased risk for
AMD related to total alcohol intake. Two prospective studies of
alcohol consumption and the risk of age-related macular
degeneration have failed to nd any association between alcohol
intake and risk of AMD.101,102 In summary, evidence suggests that
alcohol intake has little effect on AMD, although the possible
influence of heavy intake requires further study.
GENETICS
418
It is now known that genetic factors play a role in the etiology

of AMD. The evidence leading up to recent discoveries includes
the demonstration of familial aggregation,103104 large twin
studies,105,106 a case-control study,33 and a segregation
analysis.107 In one study,103 rst-degree relatives of cases with
AMD were compared with rst-degree relatives of control
subjects without AMD. The prevalence of medical record
conrmed age-related maculopathy was signicantly higher
among rst-degree relatives of all case probands (23.7%)
compared with rst-degree relatives of control probands (11.6%)
with an age- and sex-adjusted OR of 2.4 (95% CI of 1.2 to 4.7).
When relatives of cases with exudative disease were evaluated,
the OR was 3.1 (95% CI of 1.5 to 6.7) for relatives of cases
compared with relatives of controls. These results suggested
that macular degeneration has a familial component and that
genetic or shared environmental factors, or both, contribute to
its development. In another study,104 20 of 81 siblings of
affected patients had AMD compared with only 1 of 78 siblings
of control subjects. These studies supported the familial
aggregation of this disease. Additional evidence for a genetic
component was suggested by a segregation analysis involving
the Beaver Dam population.107 In a case-control study, cases

were twice as likely to report a family history of this disease.33
Several genetic linkage studies have also been performed to
try to identify regions of the genome that would merit further
exploration. While almost every chromosome has been implicated in linkage studies for AMD, the most reproducible linkage
peaks have been on chromosomes 1q and 10q and this has been
recently conrmed in a meta-analysis.108113 There is also some,
albeit weaker, evidence for linkage on chromosomes 2p, 3p, 4q,
12q, and 16q.108 The most promising developments in AMD
genetics research have occurred recently in the context of
association analyses. A complete review of the genetics of AMD
up to July 2006 was recently reported.114 Following is a summary of the ndings.
Complement Factor H
In March 2005, three separate groups reported in Science on a
common coding variant, Y402H, in the complement factor H
(CFH) gene on chromosome 1 (1q31) that increases the risk of
developing AMD.115117 The studies estimated the odds ratio
associated with this variant for all categories of AMD to be
between 2.45 to 3.33. The odds ratios were higher, between 3.5
and 7.4, for advanced dry and wet forms of AMD. CFH inhibits
the formation and accelerates the decay of alternative pathway
C3 convertases and serves as a cofactor for the factor-1 mediated
cleavage and inactivation of C3b.118 The Y402H single
nucleotide polymorphism (SNP) is within the CFH binding site
for heparin and C-reactive protein. Binding to these sites
increases the afnity of CFH for complement protein C3b,
which in turn increases the ability of CFH to inhibit
complements effects. Previous to the discovery of the strong
association between AMD and this CFH variant, the inflammatory cascade had already been postulated as an important
component in the pathophysiology of AMD.55,119123 The
discovery of the risk allele lends further support to this theory.
Many studies have veried the importance of the Y402H
variant.124133 In the Chinese population, even though the
frequency of this polymorphism is low, it was still signicantly
associated with neovascular AMD.130 In Japanese patients,
while the single Y402H allele was not associated with AMD,
two haplotype blocks in CFH were associated with AMD.131
Other variants within CFH have been discovered, 124,132,133
including the noncoding variant rs1410996.132
One study has found that the G allele of ERCC6
(chromosome 10) is associated with a risk of AMD and possibly
interacts with a SNP in CFH to influence AMD susceptibility.134 Another group reported that CRP (C-reactive protein)
haplotypes conferring high levels of CRP signicantly increased
the effect of CFH Y402H.135
LOC387715/HTRAI
The LOC387717 locus on chromosome 10 has been the second
gene to be convincingly implicated in the risk of AMD development. The rst study reporting on this locus found the strongest
association for SNP rs10490924 within LOC387715.110 The
odds ratio for this allele was 5.03. More recently, two groups
have reported that the AMD LOC387715 signal is narrowed to
a single nucleotide polymorphism in the promoter region of
HTRA1, a serine protease gene on chromosome 10q26.136137
Preliminary analysis of lymphocytes and retinal pigment
epithelium from three AMD patients revealed that the risk
allele was associated with elevated expression levels of HTRA1
mRNA and protein.137
Factor B (BF) and complement component 2 (C2)

The BF and C2 genes are found on chromosome 6 within the
major histocompatibility complex (MHC) class III region. They
Complement Factor 3 (C3)

Complement factor 3 is the component of the complement
pathway which has been most recently associated with agerelated macular degeneration. The common functional
polymorphism rs2230199 (Arg80Gly) in the C3 gene was
independently reported in two association studies.139,140
Other Potentially Implicated Genes

Prior to the discovery of the CFH gene, there were several genes
that were the subject of repeated studies in AMD. ABCA4 is
mutated in Stargardt disease, a hereditary macular dystrophy, and
codes for the ATP-binding transporter protein involved in
photoreceptor vitamin A transport. Several studies showed an
association with ABCA4 variants, including variants G1961E
and D2177N, and risk of AMD.141142 while others have
not.143145 ABCA4 may be involved in a small number of AMD
cases. There have also been mixed results for the apolipoprotein
E (APOE) gene.146151 The most consistent effect in this gene has
been a protective effect of the epsilon 4 allele, although the effect
has not reached statistical signicance in some studies.147148
Other genes with mixed results include: ELOVL4 (another gene
implicated in Stargardts disease), ACE (angiotensin-converting
enzyme), and the bulin 6 gene (which plays a role in the stability
of extracellular matrix [ECM] complexes).152156
Many candidate genes which have been investigated because
of a biological hypothesis for their involvement in AMD have
been associated with AMD in one or two studies. VEGF,
vascular endothelial growth factor, is one such gene; it is located
on chromosome 6p and is denitively involved in the
pathogenesis of neovascular AMD.157 Two studies have found
associations with AMD.153,158 Given the critical role of the
human leukocyte antigen (HLA) genes in the immune response,
they have also been chosen for candidate gene association
studies.159 The TLR4 (toll-like receptor 4) gene on chromosome

9 is involved in inflammation pathways and phagocytosis of
photoreceptor outer segments by the RPE and one study found
a relationship with AMD.160 All of these associations remain to
be independently replicated.
Gene-Environment Interactions
Since the discovery of the association between the CFH gene
and AMD, there have been a few epidemiological studies
examining the relationships between environmental risk
factors, the Y402H polymorphism and AMD. One group from
the UK found that the association between Y402H and both
geographic atrophy and choroidal neovascularization was
similar in smokers and nonsmokers, although heavier smokers
with the CC genotype appeared to be at particular risk.161
Another study reported that that the susceptibility to advanced
AMD associated with CFHY402H is modied by BMI, and both
BMI and smoking increased risk of advanced AMD within the
same genotype.129
Similarly, another case-control analysis found that current
cigarette smoking and body mass index were independently
related to AMD, controlling for the LOC387715 genotype. 162
Statistical interactions between smoking and either the
CFHY402H or LOC387715 A69S genotypes were not
observed.129,162 A study of patients in the Netherlands found
that elevated erythrocyte sedimentation rate levels, elevated
CRP levels and smoking further increased the risk of AMD
among CFH Y402H homozygotes.134
SUMMARY
Age-related macular degeneration affects a large proportion of
the elderly population and is influenced by both environmental
and genetic factors. Modiable risk factors include smoking,
body mass index, antioxidants, and possibly omega-3 fatty
acids. The discovery of AMD genetic risk factors has begun
and our understanding of how genes and gene-environment
associations impact AMD onset and progression will expand
rapidly over the next several years. This will lead to new
insights into the mechanisms involved in development and
progression of AMD and better ways to prevent and treat this
common disease.
CHAPTER 38
act in the same pathway as CFH. An association between these

two genes and AMD has been reported.132,138 The initial nding
was that of one common risk haplotype with an odds ratio of
1.32 and two protective haplotypes with odds ratios of 0.36 and
0.45.138 The rst protective haplotype contained the L9H
variant of BF and the E318D variant of C2 while the second
protective haplotype contained the R32Q variant of BF and a
variant in intron 10 of C2.
REFERENCES
1. Friedman DS, OColmain BJ, Munoz B,
et al. Prevalence of age-related macular
degeneration in the United States. Arch
Ophthalmol 2004; 122: 564572.
2. National Advisory Eye Council (US): Vision
research: A national plan 19992003.
Bethesda, MD, US Department of Health
and Human Services, 1998.
3. Seddon JM, Willett WC, Speizer FE,
Hankinson SE: A prospective study of
cigarette smoking and age-related macular
degeneration in women. JAMA 1996;
276:11411146.
4. Tomany SC, Wang JJ, Van Leeuwen R,
et al. Risk factors for incident age-related
macular degeneration: pooled ndings from
3 continents. Ophthalmology 2004;
111:12801287.
5. Seddon JM, George S, Rosner B. Cigarette
smoking, sh consumption, omega-3 fatty
acid intake, and associations with agerelated macular degeneration: the US Twin
Age-Related Macular Degeneration. Arch
Ophthalmol 2006; 124:9951001.
6. Seddon JM, Cote J, Davis N, et al.
7.
8.
9.
10.
Progression of age-related macular

degeneration: association with body
mass index, waist circumference, and
waist-hip ratio. Arch Ophthalmol 2003;
121:785792.
Seddon JM, Ajani UA, Sperduto RD, et al:
Dietary carotenoids, vitamins A, C, and E,
and advanced age-related macular
degeneration. JAMA 1994; 272:14131420.
Eye Disorders Case-Control Study Group:
Antioxidant status and neovascular agerelated macular degeneration. Arch
Ophthalmol 1993; 111:104109.
Mares-Perlman JA, Brady WE, Klein R,
et al: Serum antioxidants and age-related
macular degeneration in a populationbased case-control study. Arch Ophthalmol
1995; 113:15181512.
Age-Related Eye Disease Study Research
Group. A randomized, placebo-controlled,
and vision loss: AREDS report no. 8. Arch
Ophthalmol 2001; 119:14171436.
11. Seddon J, Gragoudas E, Egan K:

Standardized data collection and coding in
eye disease epidemiology: The uveal
melanoma data system. Opthalmic Surg
1991; 22:127136.
Group. The Age-Related Eye Disease Study
(AREDS). AREDS report no. 1. Control Clin
Trials 1999; 20:573600.
13. Seddon JM, Sharma S, Adelman RA.
Evaluation of the clinical age-related
maculopathy staging system.
Ophthalmology 2006; 113; 260266.
14. Ganley J, Roberts J: Eye conditions and
related need for medical care among
persons 174 years of age, United States,
197172. Vital and Health Statistics, Series
11. No. 228. Washington, DC, DHHS
Publication No. (PHS) 831678, March
1983.
15. Klein R, Rowland ML, Harris MI:
Racial/ethnic differences in age-related
maculopathy. Third National Health and
Nutrition Examination Survey.
419
SECTION 5
420
16. Leibowitz HM, Krueger DE, Maunder LR,

et al: The Framingham Eye Study
Monograph. Surv Ophthalmol 1980;
24(S):335610.
17. Bressler NM, Bressler SB, West SK, et al:
The grading and prevalence of macular
degeneration in Chesapeake Bay Watermen.
18. Klein R, Klein BEK, Linton KLP: Prevalence
of age-related maculopathy.
19. Vingerling JR, Dielemans I, Hofman A, et al:
The prevalence of age-related maculopathy
in the Rotterdam Study. Ophthalmology
1995; 102:205210.
20. Mitchell P, Smith W, Attebo K, Wang JJ:
Prevalence of age-related maculopathy in
Australia: The Blue Mountains Eye Study.
Ophthalmology 1995; 102:14501060.
21. Eye Diseases Prevalence Research Group.
Prevalence of age-related macular
degeneration in the United States. Arch
Ophthalmol 2004; 122:564572.
22. Podgor MJ, Leske MC, Ederer F: Incidence
estimates for lens changes, macular
changes, open-angle glaucoma and
diabetic retinopathy. Am J Epidemiol 1983;
118:208212.
23. Klein R, Klein BEK, Jensen C, et al: The
ve-year incidence and progression of agerelated maculopathy. Ophthalmology 1997;
104:721.
24. Buch H, Nielsen NV, Vinding T, et al.
14-year incidence, progression, and visual
morbidity of age-related maculopathy: the
Copenhagen City Eye Study.
25. Davis C, Lovie-Kitchin J, Thompson B:
Psychosocial adjustment to age-related
macular degeneration. J Vis Impair Blind
1995; 1:1627.
26. Williams RA, Brody BL, Thomas RG, et al:
The psychosocial impact of macular
degeneration. Arch Ophthalmol 1998;
116:514520.
27. Lindblad AS, Clemons TE: Responsiveness
of the National Eye Institute Visual Function
Questionnaire to progression to advanced
age-related macular degeneration, vision
loss, and lens opacity: AREDS Report no.
14. Arch Ophthalmol 2005; 23:12071214.
28. Sommer A, Tielsch JM, Katz J, Quigley HA,
et al: Racial differences in the causespecic prevalence of blindness in East
Baltimore. N Engl J Med 1991;
325:14121417.
29. Schachat AP, Hyman L, Leske C, Connell
AMS, et al: Features of age-related macular
degeneration in a black population. Arch
Ophthalmol 1995; 113:728735.
30. Cruickshanks KJ, Hamman RF, Klein R,
Nondahl DM, et al: The prevalence of agerelated maculopathy by geographic region
and ethnicity. Arch Ophthalmol 1997;
115:242250.
31. The Eye Disease Case-Control Study
Group: Risk factors for neovascular agerelated macular degeneration. Arch
Ophthalmol 1992; 110:17011708.
32. Klein R, Klein BEK, Jensen SC, et al: The
relation of socioeconomic factors to agerelated cataract, maculopathy and impaired
vision. Ophthalmology 1994;
101:19691979.
33. Hyman LG, Lilienfeld AM, Ferris FL, et al:
Senile macular degeneration: A casecontrol study. Am J Epidemiol 1983;
118:213227.

Group. Risk factors associated with agerelated macular degeneration. A casecontrol study in the Age-Related Eye
Disease Study: AREDS report no. 3.
Ophthalmology 2000; 107:22242232.
35. Holz FG, Piguet B, Minassian DC, et al:
Decreasing stromal iris pigmentation as a
risk factor for age-related macular
degeneration. Am J Ophthalmol 1994;
117:1923.
36. Sandberg MA, Gaudio AR, Miller S, et al:
Iris pigmentation and extent of disease in
patients with neovascular age-related
macular degeneration. Invest Ophthalmol
1994; Vis Sci 35:27342740.
37. Klein R, Barbara BEK, Jensen SC,
Cruikshanks KJ: The relationship between
ocular factors to the incidence and
progression of age-related maculopathy.
38. Khan JC, Shahid H, Thurlby DA et al. Age
related macular degeneration and sun
exposure, iris coulour, and skin sensitivity
to sunlight. Br J Ophthalmol 2006; 90:2932.
39. Tomany SC, Klein R, Klein BE, et al. The
relationship between iris color, hair color,
and skin sun sensitivity and the 10-year
incidence of age-related maculopathy: the
Beaver Dam Eye Study. Ophthalmology
2003; 110:15261533.
40. Ikram MK, van Leeuwen R, Vingerling JR,
et al. Relationship between refraction and
prevalent as well as incident age-related
maculopathy: the Rotterdam Study. Invest
Ophthalmol 2003; Vis Sci 44:3778782.
41. Sperduto R, Hiller R, Seigel D: Lens
opacities and senile maculopathy. Arch
Ophthalmol 1981; 99:10041008.
42. Liu IY, White L, LaCroix AZ: The association
between age-related macular degeneration
and lens opacities in the aged. Am J Public
Health 1989; 79:765769.
43. Klein R, Klein BE, Wang Q, et al: Is agerelated maculopathy associated with
cataracts? Arch Ophthalmol 1994;
112:191196.
44. Pollack A, Marcovich A, Bukelman A, et al:
Age-related macular degeneration after
extracapsular cataract extraction with
intraocular lens implantation.
Ophthalmology 1996; 103:15461554.
45. Klein R, Klein BE, Wong TY, et al: The
association of cataract and cataract
surgery with the long-term incidence of
age-related maculopathy: the Beaver Dam
eye study. Arch Ophthalmol 2002;
120:15511558.
46. Cugati S, Mitchell P, Rochtchina E, et al:
Cataract surgery and the 10-year incidence
of age-related maculopathy: The Blue
Mountains Eye Study. Ophthalmology,
2006. In Press (E-Pub).
47. Wang JJ, Klein R, Smith W, et al: Cataract
surgery and the 5-year incidence of latestage age-related maculopathy: pooled
ndings from the Beaver Dam and Blue
Mountains eye studies. Ophthalmology
2003; 110:19601967.
48. van der Schaft TL, Mooy CM, de Brujin WC,
et al: Increased prevalence of disciform
macular degeneration after cataract
extraction with implantation of an
intraocular lens. Br J Ophthalmol 1994;
78:441445.
49. Guttman C, Ferris FL: No link between
cataract surgery and AMD, study shows.
Ophthalmology Times 28, 1 March 2003.
50. Christen WG, Glynn RJ, Manson JE, et al:

A prospective study of cigarette smoking
and risk of age-related macular
degeneration in men. JAMA 1996;
276:11471151.
51. Klein R, Deng Y, Klein BF, et al:
Cardiovascular disease, its risk factors and
treatment, and age-related macular
degeneration: Womens Health Initiative
Sight Exam ancillary study. Am J
Ophthalmol 2007; 143:473483.
52. Clemons TF, Milton RC, Klein R, et al: Risk
factors for the incidence of advanced agerelated macular degeneration in the AgeRelated Eye Disease Study (AREDS):
AREDS report no.19. Ophthalmology 2005;
112:553559.
53. Schaumberg DA, Christen WG, Hankinson
SE, Glynn RJ: Bosy mass index and the
incidence of visually signicant age-related
maculopathy in men. Arch Ophthalmol
2001; 119:12591265.
54. Delcourt C, Michel F, Colvez A, et al:
Associations of cardiovascular disease and
its risk factors with age-related macular
degeneration: the POLA study. Ophthalmic
Epidemiol 2001; 8:237249.
55. Snow KK, Seddon JM: Do age-related
macular degeneration and cardiovascular
disease share common antecedents?
Ophthalmic Epidemiol 1999; 6:125143.
56. Vingerling JR, Dielemans I, Bots ML, et al:
Age-related macular degeneration is
associated with atherosclerosis: The
Rotterdam Study. Am J Epidemiol 1995;
142:404409.
57. Goldberg J, Flowerdew G, Smith E, et al:
Factors associated with age-related
macular degeneration: An analysis of data
from the rst National Health and Nutrition
Examination Survey. Am J Epidemiol 1988;
128:700710.
58. Hirvela H, Luukinen H, Laara E, et al: Risk
factors of age-related maculopathy in a
population 70 years of age or older.
59. Klein R, Klein BEK, Franke T: The
relationship of cardiovascular disease and
its risk factors to age-related maculopathy.
60. Sperduto RD, Hiller R: Systemic
hypertension and age-related macular
degeneration in the Framingham Eye Study.
61. Delaney WV, Oates RP: Senile macular
degeneration: A preliminary study. Ann
62. Kahn HA, Leibowitz HM, Ganley JP, et al:
The Framingham Eye Study II: Association
of ophthalmic pathology with single
variables previously measured in the
Framingham Heart Study. Am J Epidemiol
1977; 106:3341.
63. Macular Photocoagulation Study Group:
Risk factors for choroidal
neovascularization in the second eye of
patients with juxtafoveal or subfoveal
choroidal neovascularization secondary to
age-related macular degeneration. Arch
Ophthalmol 1997; 151:741747.
64. Sanders TAB, Haines AP, Wormald R, et al:
Essential fatty acids, plasma cholesterol,
and fat-soluble vitamins in subjects with
age-related maculopathy and matched
control subjects. Am J Clin Nutr 1993;
57:428433.
65. Klein R, Klein BEK, Jensen SC: The relation
of cardiovascular disease and its risk
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
factors to the 5-year incidence of agerelated maculopathy: The Beaver Dam Eye
Smeeth L, Cook C, Chakravarthy U, et al:
A case control study of age related macular
degeneration and use of statins. Br J
Ophthalmol 2005; 89:11711175.
Seddon J, Ajani U, Sperduto R, et al:
Dietary fat intake and age-related macular
1994; 25:2003.
Seddon JM, Rosner B, Sperduto RD, et al:
Dietary fat and risk for advanced agerelated macular degeneration. Arch
Ophthalmol 2001; 119:11911199.
Cho E, Hung S, Willett WC, et al:
Prospective study of dietary fat and the risk
of age-related macular degeneration. Am J
Clin Nutr 2001; 73:209218.
Mares-Perlman JA, Brady WE, Klein R:
Dietary fat and age-related maculopathy.
Seddon JM, Cote J, Rosner B: Progression
of age-related macular degeneration:
association with dietary fat,
transunsaturated fat, nuts, and sh intake.
Klein R, Klein BEK, Moss SE: Diabetes,
hyperglycemia and age-related
maculopathy: The Beaver Dam Eye Study.
Haan MN, Klein R, Klein BE, et al: Hormone
therapy and age-related macular
degeneration: the Womens Health Initiative
Sight Exam Study. Arch Ophthalmol 2006;
124:988992.
Snow KK, Cote J, Tang W, et al:
Association between reproductive and
hormonal factors and age-related
maculopathy in postmenopausal women.
Am J Ophthalmol 2002; 134:842848.
Klein BEK, Klein R, Jensen SC, et al: Are
sex hormones associated with age-related
maculopathy in women? The Beaver Dam
Eye Study. Trans Am Ophthalm Soc 1994;
92:285289.
Smith W, Mitchell P, Wang JJ: Gender
estrogen, hormone replacement and agerelated macular degeneration: Results from
the Blue Mountains Eye Study. Aus N Z J
Ophthalmol 1997; 25 (Suppl 1):S13S15.
Vingerling JR, Dielemans I, Witteman JC,
et al: Macular degeneration and early
menopause: A case-control study. BMJ
1995; 310:15701571.
Taylor HR, West S, Munoz B, et al: The
long-term effects of visible light on the eye.
Cruickshanks KJ, Klein R, Klein BEK:
Sunlight and age-related macular
degeneration: The Beaver Dam Eye Study.
Tomany SC, Cruickshanks KJ, Klein R,
et al. Sunlight and the 10-year incidence of
age-related maculopathy: the Beaver Dam
Eye Study. Arch Ophthalmol 2004;
122:750757.
Darzins P, Mitchell P, Heller RF: Sun
exposure and age-related macular
degeneration: An Australian case-control
study. Ophthalmology 1997; 104:770776.
Hung S, Seddon JM: The relationship
between nutritional factors and age-related
macular degeneration. In Bendich A,
Deckelbaum R (eds): Preventive Medicine:
The Comprehensive Guide for Health
Professionals. Totowa, NJ, Humana Press,
1997; pp 245265.
83. Seddon JM, Hennekens CH: Vitamins,

minerals and macular degeneration. Arch
Ophthalmol 1994; 112:176179.
84. Anderson RE, Kretzer FL, Rapp LM: Free
radicals and ocular disease. Adv Exp Med
Biol 1994; 366:7386.
85. Mares Perlman JA, Brady WE, Klein R,
et al: Association of zinc and antioxidant
nutrients with age-related maculopathy.
86. West S, Vitale S, Hallfrisch J, et al: Are
antioxidants or supplements protective for
age-related macular degeneration? Arch
Ophthalmol 1994; 112:222227.
87. Mares-Perlman JA, Fisher AI, Klein R:
Lutein and zeaxanthin in the diet and
serum and their relation to age-related
maculopathuy in the third national health
and nutrition examination survey. Am J
Epidemiol 2001; 153:424.
88. Richer S, Stiles W, Statkute L, et al:
Double-masked, placebo-controlled,
randomized trial of lutein and antioxidant
supplementation in the intervention of
atrophic age-related macular degeneration:
the Veterans LAST study (Lutein
Antioxidant Supplementation Trial). 2004;
75:216230.
89. Cho E, Seddon JM, Rosner B, et al:
Prospective study of intake of fruits,
vegetables, vitamins, and carotenoids and
risk of age-related maculopathy. Arch
Ophthalmol 2004; 122:883892.
90. Moeller SM, Parekh N, Tinker L, et al:
Associations between intermediate agerelated macular degeneration and lutein
and zeaxanthin in the Carotenoids in Agerelated Eye Disease Study (CAREDS):
ancillary study of the Womens Health
Initiative. Arch Ophthalmol 2006;
124:11511162.
91. Van Leeuwen R, Boekhoorn S, Vingerling JR,
et al: Dietary intake of antioxidants and risk
of age-related macular degeneration. JAMA
2005; 294:31013107.
92. Newsome DA, Swartz M, Leone NC, et al:
Oral zinc in macular degeneration. Arch
Ophthalmol 1988; 106:192198.
93. Stur M, Titti M, Reitner A, et al: Oral zinc
and the second eye in age-related macular
1996; 37:12251235.
94. Cho E, Stampfer MJ, Seddon JM, et al:
Prospective study of zinc intake and the
risk of age-related macular degeneration.
Ann Epidemiol 2001; 11; 328336.
95. Ajani UA, Willett W, Miller D, et al: Alcohol
consumption and neovascular age-related
macular degeneration. [Abstract] Am J
Epidemiol 1993; 138:646.
96. Vinding T, Appleyard M, Nyboe J, et al:
Risk factor analysis for atrophic and
exudative age-related macular
degeneration: An epidemiologic study of
1000 aged individuals. Acta Ophthalmol
1992; 70:6672.
97. Obisesan TO, Hirsch R, Kosoko O, et al:
Moderate wine consumption is associated
with decreased odds of developing agerelated macular degeneration in NHANES-1.
J Am Geriatr Soc 1998; 46:17.
98. Fraser-Bell S, Wu J, Klein R, et al:
Smoking, alcohol intake, estrogen use, and
age-related macular degeneration in
Latinos: the Los Angeles Latino Eye Study.
99. Ritter LL, Klein R, Klein BE, et al: Alcohol
use and age-related maculopathy in the
100.
101.
102.
103.
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
115.
116.
Beaver Dam Eye Study. Am J Ophthalmol

1995; 120:190196.
Klein R, Klein BE, Tomany SC, et al: Ten-year
incidence of age-related maculopathy and
smoking and drinking: the Beaver Dam Eye
Study. Am J Epidemiol 2002; 156:589598.
Ajani UA, Christen WG, Manson JE, et al:
A prospective study of alcohol
consumption and th risk of age-related
macular degeneration. Ann Epidemiol 1999;
9:172177.
Cho E, Hankinson SE, Willett WC, et al:
Prospective study of alcohol consumption
and the risk of age-related macular
118:681688.
Seddon JM, Ajani UA, Mitchell BD: Familial
aggregation of age-related maculopathy.
Am J Ophthalmol 1997; 123:199206.
Silvestri G, Johnston PB, Hughes AE: Is
genetic predisposition an important risk
factor in age-related macular
degeneration? Eye 1994; 8:564568.
Seddon JM, Cote J, Page WF, et al: The
US twin study of age-related macular
degeneration: relative roles of genetic and
environmental influences. Arch Ophthalmol
2005; 123:321327.
Hammond CJ, Webster AR, Snieder H,
et al: Genetic influence on early age-related
maculopathy: a twin study. Ophthalmology
2002; 109:730736.
Heiba IM, Elston RC, Klein BEK, et al:
Sibling correlations and segregation
analysis of age-related maculopathy: The
Beaver Dam Eye Study. Genet Epidemiol
1994; 11:5167.
Fisher SA, Abecasis GR, Yashar BM, et al:
Meta-analysis of genome scans of agerelated macular degeneration. Hum Mol
Genet 2005; 14:22572264.
Iyengar SK, Song D, Klein BE, et al:
Dissection of genome-wide scan data in
extended families reveals a major locus
and oligogenic susceptibility for age-related
macular degeneration. Am J Hum Genet
2004; 74:2039.
Jakobsdottir J, Conley YP, Weeks DE, et al:
Susceptibility genes for age-related
maculopathy on chromosome 10q26. Am J
Hum Genet 2005; 77:389407.
Majewski J, Schultz DW, Weleber RG, et al:
Age-related macular degeneration a
genome scan in extended families. Am J
Hum Genet 2003; 73:540550.
Seddon JM, Santangelo SL, Book K, et al:
A genome-wide scan for age-related
macular degeneration provides evidence
for linkage to several chromosomal regions.
Am J Hum Genet 2003; 73:780790.
Weeks DE, Conley YP, Tsai HJ, et al: Agerelated maculopathy: an expanded
genome-wide scan with evidence of
susceptibility loci within 1q31, 10q26, and
17q25 regions. Am J Hum Genet 2004;
75:174189.
Haddad S, Chen CA, Santangelo SL, et al:
The genetics of age-related macular
degeneration: a review of progress to date.
Edwards AO, Ritter R, Abel KJ, et al:
Complement factor H polymorphism and
age-related macular degeneration. Science
2005; 308:421424.
Haines JL, Hauser MA, Schmidt S, et al:
Complement factor H variant increases the
Science 2005; 308:419421.
CHAPTER 38
421
SECTION 5
422
117. Klein RJ, Zeiss C, Chew EY, et al:

Complement factor H polymorphism in
age-related macular degeneration. Science
2005; 308:385389.
118. Soames CJ, Sim RB: Interactions between
human complement components factor H,
factor I and C3b. Biochem J 1997;
326:553561.
119. Donoso LA, Kim D, Frost A, et al: The role
of inflammation in the pathogenesis of agerelated macular degeneration. Surv
Ophthalmol 2006; 51:137152.
120. Gurne DH, Tso MO, Edward DP, et al:
Antiretinal antibodies in serum of patients
with age-related macular degeneration.
121. Killingsworth MC, Sarks JP, Sarks SH, et al:
Macrophages related to Bruchs membrane
in age-related macular degeneration. Eye
1990; 4:613621.
122. Mullins RF, Russell SR, Anderson DH, et al:
Drusen associated with age and agerelated macular degeneration contain
proteins common to extracellular deposits
associated with atherosclerosis, elastosis,
amyloidosis and dense deposit disease.
FASEB J 2000; 14:835846.
123. Seddon JM, Gensler G, Milton RC, et al:
Association between C-reactive protein and
age-related macular degeneration. JAMA
2004; 291:704710.
124. Hageman GS, Anderson DH, Johnson LV,
et al: A common haplotype in the
complement regulatory gene factor H
(HF1/CFH) predisposes individuals to agerelated macular degeneration. Proc Natl
Acad Sci USA; 2005; 102:72277232.
125. Magnusson KP, Duan S, Sigurdsson H,
et al: CFH Y204H confers similar risk of
soft drusen and both forms of advanced
AMD. PLOS Med 2006; 3:e5.
126. Souied EH, Leveziel N, Richard F, et al:
Y402H complement factor H polymorphism
associated with exudative macular
degeneration in the French population. Mol
Vis 2005; 11:11351140.
127. Zareparsi S, Branham KE, Li M, et al:
Strong association of the Y204H variant in
complement factor H at 1q32 with
susceptibility to age-related macular
degeneration. Am J Hum Genet; 2005;
77:149153.
128. Schaumberg DA, Christen WG, Kozlowski P,
et al: A prospective assessment of the
Y204H variant in complement factor H,
genetic variants in C-reactive protein and
47:23362340.
129. Seddon JM, George S, Rosner B, et al:
CFH gene variant, Y402H, and smoking,
body mass index, environmental
associations with advanced age-related
macular degeneration. Hum Hered 2006;
61:157165.
130. Lau L-I, Chen S-J, Cheng C-Y, et al:
Association of the Y204H polymorphism in
complement factor H gene and
degeneration in Chinese patients. Invest
131. Okamoto H, Umeda S, Obazawa M, et al:
Complement factor H polymorphisms in
Japanese population with age-related
macular degeneration. Mol Vis; 2006;
12:156158.
132. Maller J, George S, Purcell S, et al:
Common variation in three genes, including
133.
134.
135.
136.
137.
138.
139.
140.
141.
142.
143.
144.
145.
146.
147.
a noncoding variant in CFH, strongly

influences risk of age-related macular
degeneration. Nat Genet 2006;
38:10551059.
Li M, Atmaa-Sonmez P, Othman M, et al:
CFH haplotypes without the Y204H coding
variant show strong association with
susceptibility to age-related macular
38:10491054.
Tuo J, Ning B, Bojanowski CM, et al:
Synergic effect of polymorphisms in ERCC6
5 flanking region and complement factor H
on age-related macular degeneration
predisposition. PNAS 2006; 103:92569261.
Despriet DDG, Klaver CCW, Witteman JCM,
et al: Complement factor H polymorphism,
complement activators, and risk of agerelated macular degeneration. JAMA 2006;
296:301309.
DeWan A, Liu M, Harman S, et al: HTRA1
promoter polymorphism in wet age-related
macular degeneration. Science, 2006. In
Press (EPub).
Yang Z, Camp NJ, Sun H, et al: A variant of
the HTRA1 gene increases susceptibility to
age-related macular degeneration. Science,
2006. In Press (Epub).
Gold B, Merriam JE, Zernant J, et al:
Variation in factor B (BF) and complement
component 2 (C2) genes is associated with
age-related macular degeneration. Nat
Genet 2006; 38:458462.
Yates JR, Sepp T, Matharu, et al:
Complement C3 variant and the risk of
age-related macular degeneration. N Engl J
Med 2007; 357:553561.
Maller JB, Fagerness JA, Reynolds RC,
et al: Variation in complement factor 3 is
associated with risk of age-related macular
39:12001201.
Allikmets R: Further evidence for an
association of ABCR alleles with agerelated macular degeneration. The
International ABCR Screening Consortium.
Am J Hum Genet 2000; 67:487491.
Shroyer NF, Lewis RA, Yatsenko AN, et al:
Cosegregation and functional analysis of
mutant ABCR (ABCA4) alleles in families
that manifest both Stargardt disease and
age-related macular degeneration. Hum
Mol Gent 2001; 10:26712678.
Guymer RH, Heon E, Lotery AJ, et al:
Variation of codons 1961 and 2177 of the
Stargardt disease gene is not associated
Rivera A, White K, Stohr H, et al: A
comprehensive survey of sequence
variation in the ABCA4 (ABCR) gene is
Stargardt disease and age-related macular
degeneration. Am J Hum Genet 2000;
67:800813.
Schmidt S, Postel EA, Agarwal A, et al:
Detailed analysis of allelic variation in the
ABCA4 gene in age-related maculopathy.
44:28682875
Baird PN, Guida E, Chu DT, et al: The
epsilon2 and epsilon4 alleles of the
apolipoprotein gene are associated with
age-related macular degeneration. Invest
Souied EH, Benlian P, Amouyel P, et al: The
epsilon4 allele of the apolipoprotein E gene
as a potential protective factor for
exudative age-related macular
148.
149.
150.
151.
152.
153.
154.
155.
156.
157.
158.
159.
160.
161.
162.

2000; 41:244247.
Zareparsi S, Reddick AC, Branham KE,
et al: Association of apolipoprotein E alleles
with susceptibility to age-related macular
degeneration in a large cohort from a single
center. Invest Ophthalmol Vis Sci 2004;
45:13061310.
Schultz DW, Klein ML, Humpert A, et al:
Lack of an association of apolipoprotein E
gene polymorphisms with familial agerelated macular degeneration. Arch
Ophthalmol 2003; 121:679683.
Seddon JM, De La Paz M, Clements K,
et al: No association between
apolipoprotein E and advanced age-related
macular degeneration. Am J Hum Gen
1996; 59:388.
Wong TY, Shnakar A, Klein R, et al:
Apolipoprotein E gene and early agerelated maculopathy: the Atherosclerosis
Risk in Communities Study. Ophthalmology
2006; 133:255259.
Ayyagari R, Zhang K, Hutchinson A, et al:
Evaluation of the ELOVL4 gene in patients
Ophthalmic Genet 2001; 22:233239.
Haines JL, Schnetz-Boutaud N, Schmidt S,
et al: Functional candidate genes in agerelated macular degeneration: signicant
association with VEGF, VLDLR, and LRP6.
47:329335.
Hamdi HK, Reznik J, Castellon R, et al:
Alu DNA polymorphism in ACE gene is
protective for age-related macular
degeneration. Biochem Biophys Res
Commun 2002; 295:668672.
Abecasis GR, Yashar BM, Zhao Y, et al:
Age-related macular degeneration: a highresolution genome scan for susceptibility
loci in a population enriched for late-stage
disease. Am J Hum Genet 2004;
74:482494.
Stone EM, Braun TA, Russell SR, et al:
Missense variations in bulin 5 gene and
age-related macular degeneration. N Engl J
Med 2004; 351:346353.
Ferrara N: Vascular endothelial growth
factor: basic science and clinical progress.
Endocr Rev 2004; 25:581611.
Churchill AJ, Carter JG, Lovell HC, et al:
VEGF polymorphisms are associated with
degeneration. Hum Mol Genet 2006;
15:29552961.
Goverdhan SV, Howell MW, Mullins RF,
et al: Association of HLA class I and class
II polymorphisms with age-related macular
degeneration. Invest Ophthalmol Vis Sci.
2005; 46:17261734.
Zareparsi S, Reddick AC, Branham KE,
et al: Toll-like receptor 4 variant D299G is
associated with susceptibility to agerelated macular degeneration. Hum Mol
Genet 2005; 14:14491455.
Sepp T, Khan JC, Thurlby DA, et al:
Complement factor H variant Y402H is a
major risk determinant for geographic
atrophy and choroidal neovascularization.
47:536540.
Francis PJ, George S, Schulz DW, et al:
The LOC387715 gene, smoking, body
mass index, environmental associations
with advanced age-related macular
degeneration. Hum Hered 2007;
63:212218.
SECTION 6
CORNEA AND CONJUNCTIVA

Edited by William J. Power and Dimitri T. Azar
CHAPTER
39
Anatomy and Cell Biology of the Cornea,

Supercial Limbus, and Conjunctiva
Ilene K. Gipson and Nancy C. Joyce
The tissues at the ocular surface include the cornea, conjunctiva, and the intervening zone of the limbus; the regions are
shown diagrammatically and histologically in Figure 39.1. The
primary function of the entire region is to refract and transmit
light to the lens and retina. Although the cornea and its surface
tear lm constitute the tissue actually performing the tasks, the
limbus and conjunctiva support the cornea in these important
functions. Because the cornea is such a major functional tissue
of the eye and because damage to or disease of the cornea has
serious visual consequences, its structure, function, and
pathology have received much attention. Increased interest in
the limbus and conjunctiva has yielded new information
regarding the important supportive functions of the tissues
surrounding the cornea. This chapter reviews the anatomy and
cell biology of the three regions of the ocular surface, including
some of the recently observed structural and cell biologic
features. In the previous edition of this text, specic chapters
dealt with the cell biology of the corneal epithelium, the corneal
stroma and its connective tissue, and the corneal endothelium.
In this volume, these topics are combined and the publication
space is restricted. Thus, for more complete details regarding
the cell and molecular biology of these tissue regions, refer to a
previous edition of this text.1 More complete details regarding
the gross anatomy of the region also are available.2,3
CORNEA
The cornea is a highly specialized tissue that refracts and transmits light to the lens and retina. In humans, it is about twice as
thick at the periphery than at the center (1 mm compared with
0.5 mm).46 The tissue of the cornea appears simple in
composition because it is composed only of an outer stratied
squamous nonkeratinized epithelium, an inner dense connective tissue stroma with its resident broblast-like keratocytes,
and a monolayered cuboidal endothelium bordering the anterior
chamber (see Fig. 39.1). The cornea, however, actually is highly
ordered and complexly arranged in comparison with other
tissues of the body. Its transparency, avascularity, and highly
ordered structure make it unique among all tissues of the body.
Cells of all layers interact with and influence each others
functions. They do not act alone, but mediators (cytokines)
expressed by one cell type influence cells of adjacent layers.
EPITHELIUM
The surface of the cornea is covered by a stratied squamous
nonkeratinizing epithelium, which in humans, rodents, and
rabbits has ve to seven cell layers. The epithelium is 5052 mm
thick. The corneal epithelium has functions unique to it and
functions that are common to all other epithelia of the body.
Several of its unique functions include light refraction and

transmittance and survival over an avascular bed. The unique
function of light refraction is brought about by its absolutely
smooth, wet apical surface and its extraordinarily regular
thickness. Transparency of the epithelium to light appears to be
brought about by scarcity of cellular organelles and possibly by
high concentrations of enzyme crystallins.7 The epithelium has
specialized metabolic characteristics that allow it to exist over
an avascular connective tissue.8 Protection of these unique and
vital functions is provided by a high density of sensory nerves
that send unmyelinated endings to terminate within the
suprabasal and squamous cells of the epithelium. The density
of nerve endings per unit area appears to be 300400 times that
of the epidermis.9 The epithelium also has a rapid and highly
developed ability to respond to wounds, and it is maintained by
centripetal movement of cells derived from an adult stem-cell
population located in the basal layer of the limbal epithelium
(see further ahead).
In addition to its specialized functions, the corneal epithelium has the routine housekeeping functions of all epithelia
that border the outside world. The layers of cells provide a
barrier to fluid loss and pathogen entrance and resist abrasive
pressure by tightly adhering to one another and to the
underlying connective tissue stroma.
The stratied epithelium includes three or four layers of
outer flat squamous cells termed squames, one to three layers
of midepithelial cells termed wing cells because of their rounded
cell body and lateral winglike cellular processes, and a layer of
columnar basal cells (Fig. 39.2). The latter secrete and maintain
the epitheliums basement membrane, which, compared with
that of the other stratied epithelia (i.e., epidermis), is smooth
or planar and nonundulating. This smooth or planar characteristic may support the regular thickness of the epithelium
over the entire cornea.
Like all other stratied epithelia, the epithelium of the cornea is self-renewing, turning over in humans and rats in
~57 days.10 Basal cells are the mitotically active layer; as
they divide, daughter cells begin their movement off the basement membrane toward terminal differentiation and desquamation from the apical surface. It was thought that one
daughter cell resulting from a division moved off the basal
lamina, leaving one daughter cell in place to undergo mitosis
again.11 More recent data using bromodeoxyuridine (BrdU)
labeling indicate that the two progeny of a single division move
together toward the apical surface.12
All cell layers of epithelium have a sparse accumulation of
cytoplasmic organelles. Endoplasmic reticulum and mitochondria are sparsely distributed around the cytoplasm, with a
Golgi apparatus present in a supranuclear position, particularly
in the basal cell layer (Fig. 39.3). In the apical cell layers, Golgi
423

FIGURE 39.1. Diagram and light micrographs
of ocular surface tissues. Boxes (AC)
correspond to regions in the light micrographs
(ac) at the right; all are sections of human
tissue, as is (d), which shows a higher
magnication of conjunctival epithelium.
(a) Section through the central cornea. a,
epithelium; b, Bowmans layer; c, lamellar
stroma; d, Descemets membrane; e,
endothelium. 120. (b) Section through the
limbus. The large arrow designates the end of
Bowmans layer and the small arrow the
position of the rst blood vessel encountered
outside the corneal stroma. 48. (c) Section of
bulbar conjunctiva. Note the highly vascularized
connective tissue. 120. (d) Section of bulbar
conjunctiva demonstrating the presence of
numerous goblet cells (arrows) within the
stratied epithelium and the cellular nature of
the connective tissue of the substantia propria
(arrowheads). 300.
a
SECTION 6
424
cisternae and small membrane-bound vesicles consistent in

size and structure with Golgi-associated vesicles are especially
prominent (see Fig. 39.3).
Of the three cytoplasmic lament types within all cells, actin
laments, keratin laments and microtubules, keratin or
intermediate laments are the major type within the cytoplasm
of cells of the corneal epithelium. On electron micrographs,
the cell cytoplasm of all layers of the corneal epithelium appears
full of these laments, and keratin proteins, which polymerize
to form the laments, are among the most abundant proteins of
the tissue. The keratin family of proteins that form intermediate laments is a complex family of ~30 polypeptides,
which are of two classes: type I, or acidic; and type II, or neutral
and basic. The intermediate laments within ectodermally
derived epithelia are formed by the pairing of two specic
keratin proteins, one from each class. In the corneal epithelium,
as basal cells differentiate to apical cells, two keratin pairs are
expressed sequentially. First, K5 and K14 are expressed in basal
cells; subsequently, suprabasal cells express K3 and K12.1315

K12, a 64-kDa keratin, is believed to be cornea specic.16 The
cytokeratin laments not only increase the tensile strength of
the epithelial cells but also, by keeping the nucleus and other
organelles in their proper positions, affect the overall
organization of the cell. They also provide a scaffold upon which
associated proteins are organized and regulated to control cell
metabolic and homeostatic activities.17 Another major role of
the intermediate laments of the corneal epithelium is to
provide the cytoskeletal component of the system that anchors
cells tightly to one another and to their substrate through the
desmosome and hemidesmosome (Figs 39.4 and 39.5). Such
tight anchorage is critical to a stratied epithelium that borders
the outside world and is subject to the abrasive pressures from
lid movement and eye rubbing.
Actin laments, as with all cells, are present throughout the
cytoplasm of cells of the corneal epithelium. They are particularly prevalent as a network along the apical cell membranes
FIGURE 39.2. Sections of corneal epithelium as seen by light (inset)

and electron microscopy showing supercial, wing, and basal cell
layers and Bowmans layer (bl). In the electron micrograph, note the
surface microplicae and interdigitating cell membranes with electrondense desmosomes. Electron-lucent proles of endoplasmic reticulum
are widely scattered, primarily within basal and wing cells. Electrondense hemidesmosomes are prominent along the basal cell
membrane of the columnar basal cells adjacent to the basal lamina.
300; inset 2700.
of the epithelium, where they extend into microplicae

(Fig. 39.6), and at the junction of the lateral membranes, where
they are associated with adherens and tight junctions.18 The
actin lament system is particularly important in providing
the cytoskeletal connection of cell adhesion molecules, such as
the integrins and cadherins, and the cytoskeletal component of
adherens and tight junctions in epithelia.
Composed of both a- and b-subunits of the proteins known
as tubulins, microtubules are the third major cytoskeletal
element within all cells.19 Although they are not obvious on
electron micrographs of corneal epithelia, they are obvious
within the spindles of mitotic basal cells, where they provide
the cytoskeletal framework for chromosome segregation. They
do not appear to play a signicant role in corneal epithelial
wound healing, indicating that they are not required for epithelial migration and that mitosis is not required for epithelial
wound coverage.20
The corneal epithelium, like all other epithelia, has intercellular junctions that function not only in cell adhesion but
also in cell communication and barrier formation. Four junction types are present (see Fig. 39.4). Desmosomes, which are
present along the lateral membranes of all corneal epithelial
cells, function in cell-to-cell adherence; adherens junctions,
which are present along the lateral membrane of the apical cells
of the epithelium, function to maintain cell-to-cell adherence
in the region of the tight junctions; the tight junctions are
present along with adherens junctions in apical cell lateral
membranes, where they function to provide a paracellular
permeability barrier; and gap junctions, which function in cellto-cell communication, allow intercellular passage of small
molecules up to 2000 Da. The latter are present along lateral
membranes of all cells of the epithelium. Basal cells have gap
junctions with a different molecular composition (connexin 43)
than suprabasal cells (connexin 50). For a more complete
description of the molecular composition of the four junction
types, see Gipson and Sugrue1 and Alberts and colleagues.21
Molecules present along cell membranes also function in cellto-cell adhesion. Two types of cell adhesion molecules in membranes of corneal epithelial cells outside specialized junction
regions are cadherins (specically, E-cadherin)4 and several of
the integrin heterodimers.22
The two surfaces of the corneal epithelium, the apical and
basal surfaces, have specializations indicative of their roles in
the epithelium. The apical surface is specialized to maintain the
tear lm and mucous layer23 and, with that layer, provides the
extraordinarily smooth refractive surface of the cornea. To
facilitate this function, the apical cell membrane has short
ridgelike folds, termed microplicae, that form regular undulations of the membrane when viewed in cross-section (see
Fig. 39.6). In addition, microvilli (nger-like projections of the
membrane) up to ~1 mm in length are present. These two
membrane specializations presumably supply an increased
surface area for adherence of the mucous layer of the tear lm.
Scanning electron microscopic studies of the corneal surface
demonstrate that apical cells scatter electrons to varying degrees
(see Fig. 39.6). Cells that scatter electrons to a lesser degree are
termed dark cells. Light cells, which scatter electrons to a
greater degree, have a higher density of surface microplicae and
microvilli.24 It has been hypothesized that the dark cells with
fewer surface membrane specializations represent the oldest
cells of the ocular surface and therefore are about to
desquamate.25 The undulating, specialized apical membrane
bears a prominent glycocalyx that is intimately associated with
the tips of the microplicae and with the mucous layer of the tear
lm (see Fig. 39.6). The corneal cells express three membranespanning mucins, designated MUC1, MUC4, and MUC16,
which are present in the apical cell membrane. The latter is a
major component of the glycocalyx and is particularly prevalent
on the tips of microplicae.26
The basal surface of the epithelium is specialized to provide
tight anchorage of the epithelium to the stroma.27,28 A series of
linked structures, termed the anchoring complex, extends from
the cytoplasm of the basal cell, through the basal cell
membrane, then through the basal lamina and into the anterior
of Bowmans layer at the anterior region of the stroma. The
structures of the anchoring complex visible by electron
microscopy include keratin laments that insert into the
hemidesmosome plaque; the hemidesmosome, which is the
specialized anchoring junction on the basal membrane;
anchoring laments, which extend from the hemidesmosome
to the basement membrane; and anchoring brils, which extend
from the basement membrane into Bowmans layer. These
anchoring brils form an intertwining network and terminate
distal to the basement membrane in anchoring plaques. The
linked structures and their molecular components are shown
diagrammatically and by electron microscopy in Figure 39.5.
CHAPTER 39
Anatomy and Cell Biology of the Cornea, Supercial Limbus, and Conjunctiva
STROMA
The corneal stroma (see Fig. 39.1a) is the connective tissue
located between the epithelial basal lamina and Descemets
membrane, the thick extracellular matrix secreted by the endothelial monolayer. The stroma comprises ~90% of the corneal
thickness and includes both Bowmans membrane and the
lamellar stroma. The major functions of the stroma are to
425

FIGURE 39.3. Electron micrographs
demonstrating aspects of the ultrastructure of
the corneal epithelium of apical cells (a) and
wing cells (b and c). (a) Portion of an apical cell
and the cell immediately beneath it. Note
microplicae (mp) and Golgi vesicles (gv).
21 000. (b) Elaborate interdigitation of
membranes of adjacent cells, characteristic of
wing and squamous cells, shows
mitochondrion (m), Golgi apparatus (g), and
rough endoplasmic reticulum (rer). 21 000.
(c) Higher-magnication electron micrograph
demonstrating that the cytoplasm of epithelial
cells is rich in keratin laments (kf). 42 000. All
these micrographs show the presence of the
cell-to-cell adhesion junctions known as
desmosomes (d), which are present along
interdigitating cell membranes. Desmosomes of
corneal epithelia appear similar to those of all
other stratied squamous epithelium.
SECTION 6
426
maintain the proper curvature of the cornea as the primary lens

of the eye, to provide mechanical resistance to intraocular
pressure, and to transmit light into the eye without signicant
absorbance. Corneal transparency is dependent on the
maintenance of a low level of stromal hydration and on the
orderly arrangement of collagen bers within the stroma.
BOWMANS MEMBRANE
Bowmans membrane (Fig. 39.7) is an 8- to 10-mm acellular
zone of randomly arranged collagen brils that forms an
interface between the basal lamina of the epithelium and the
subjacent lamellar stroma. Constituents of this layer are
believed to be synthesized and secreted by both epithelial cells
and stromal keratocytes.29,30 Bowmans membrane contains
several collagen types, including types I, V, and VII,31,32 and
proteoglycans, such as chondroitin sulfate proteoglycan.33 Both
Bowmans membrane and the lamellar stroma contain brils
composed of collagen types I and V; however, the brils in
Bowmans membrane are smaller in diameter (~20 nm) than
those in the stroma (2530 nm).34 Fibril diameter appears to be
regulated by the relative ratio of type V to type I collagen,
the greater the amount of type V, the smaller the bril
diameter.35,36 Studies using competitive polymerase chain
reaction to quantify messenger RNA (mRNA) from avian
corneal epithelial cells and stromal broblasts indicate that the

amount of mRNA for type V collagen relative to that for type I
collagen is higher in epithelial cells than in stromal
broblasts.31 This nding suggests that epithelial cells
synthesize and secrete type I and V collagen bers in Bowmans
membrane and that the higher ratio of type V to type I collagen
produced by these cells accounts for the smaller bril diameter
in Bowmans layer. Type VII collagen-containing anchoring
brils connect epithelial hemidesmosomes to anchoring
plaques located 12 mm into the anterior portion of Bowmans
membrane. These anchoring brils intertwine with type I
brillar collagen, forming a network that stabilizes the
association between the surface epithelium and the underlying
lamellar stroma.18,37 Bowmans layer is prominent in primates,
including humans, but is thin or nonexistent in other
mammals. The specic function of this layer is not clearly
understood, but its feltwork of collagen brils may stabilize the
transition between the epithelial and stromal layers, ensure
adhesion of the overlying epithelial cells to the stromal matrix,
and contribute to the smooth curvature of the corneal surface.
LAMELLAR STROMA
The lamellar stroma is the thick collagenous layer posterior
to Bowmans membrane. Collagen types I and V are the
Intermediate filaments
Keratins
Hemidesmosome
Bullous Pemphigoid Antigen
a6b4 Integrin
Anchoring Filaments
Laminin V
Bsement Membrane
Laminin
H. Sulf. Proteoglycan
Collagen VII Globular Domain
Anchoring Fibril
Collagen VII HElical Domain
Anchoring Plaque
Laminin
Collagen VII Globular Domain
FIGURE 39.5. Electron micrograph demonstrating adhesion complex

of the corneal epithelium. The linked structures of the complex and
their known molecular components are identied. 165 000.
predominant brillar collagens in the lamellar stroma, although

small amounts of other brillar collagens, such as type III, also
may be present.38 The stroma contains collagen type XII, which
cross-links brillar collagens, and type VI, which forms
microbril networks.39 Keratan sulfate proteoglycans are the
Antibody against connexin 43 was provided by DA

Goodenough, PhD. (b) Reproduced from Zieske JD,
Bukusoglu G, Gipson IK: Enhancement of vinculin
synthesis by migrating stratied squamous epithelium.
J Cell Biol 1989; 109:571.
predominant proteoglycans within the corneal stroma.40

Lumican and keratocan are the core proteins of keratan sulfate
proteoglycans, with lumican regulating keratocan expression.40
As shown in Figure 39.8a, the constituents of the lamellar
stroma are organized precisely. The basic structural unit of the
brillar collagens is tropocollagen, an asymmetric molecule
~300 nm long and 1.5 nm in diameter. Fibrillar collagens are
composed of three polypeptide chains coiled in a triple helix.
These molecules polymerize to form elongated collagen brils
with diameters of 2530 nm. The uniformity of collagen bril
diameter appears to result from specic interactions between
type V collagen, located toward the center of the bril, and type
I collagen, on the bril exterior. As mentioned previously, the
relative ratio of type V to type I collagen appears to regulate
bril diameter. The interbrillar distance also is highly uniform
and may be maintained by apposing interactions at the bril
surface. In the chick cornea, type XII collagen binds to type I on
the bril exterior and may form lateral bridges between brils,
thus limiting interbrillar distance.41 Proteoglycans bind to the
exterior surfaces of collagen brils. The polyanionic nature of
the glycosaminoglycan side chains attracts cations and water
molecules and may exert a swelling pressure on the collagen
brils, which is balanced by the interactions between collagen
types I and XII. Microbrils composed of type VI collagen also
associate with type I collagen,42,43 but the specic function of
these brils is not known. Collagen brils are packed in parallel
bundles extending from limbus to limbus, and the bundles are
arranged in layers, or lamellae. The stroma of the human eye
contains 200250 lamellae. Lamellae in the middle and
posterior regions of the stroma are arranged at approximate
right angles, whereas those in the anterior stroma are arranged
at less than right angles. The small diameter of the collagen
brils and their close, regular packing creates a lattice or threedimensional diffraction grating.44 The lattice theory of
Maurice45 suggests that the ability of the cornea to scatter 98%
CHAPTER 39
FIGURE 39.4. Cell-to-cell junctions of the

corneal epithelium as demonstrated by electron
microscopy (a, c, e) and immunolocalization of
cell-to-cell junction components (b, d, f). (a)
Areas of apparent membrane fusion at the tight
junction are obvious. 66 000. (b) Vinculin, a
component of the adherens junction, can be
seen in the immunofluorescence micrograph on
the lateral membranes of apical cells (arrows).
600. (c) Desmosomes are prominent along cell
membranes. 66 000. (d) Localization of the
desmosome component desmoplakin is
demonstrated in the immunofluorescence
micrograph. 1000. (e) A gap junction (arrow)
between two basal cells is shown in the
electron micrograph. 105 000. (f) The
immunolocalization of the gap junction protein
connexin 43 is shown in a section of chick
corneal epithelium. Note the punctate binding,
particularly along the membranes of basal cells.
427
FIGURE 39.6. Micrographs showing

specialization of the apical membrane of apical
cells of the ocular surface. (a) Electron
micrograph of mucin layer preserved on apical
membranes of guinea pig conjunctiva. Note
microplicae (mp) in cross-section and electron
density of the glycocalyx (gc) region at the tips
of the microplicae. Note tight junction (tj)
between adjoining cells. 56 000. (b and c) Low
(b, 750) and high (c, 6200) magnication
scanning electron micrographs of apical cells of
a rabbit cornea. In (b), cells vary in the amount
to which they scatter electrons, leading to a
mosaic with cobblestone appearance. In
(c), this degree of scatter correlates to the
density of microplicae on the surfaces of the
cells. (d) Immunofluorescence micrograph
demonstrating specic molecules along the
apical membrane. Cells in this section of human
cornea have been labeled with antibody to the
membrane-spanning mucin termed MUC1.
A similar pattern of labeling is seen with
antibodies to MUC16. 300.
SECTION 6
FIGURE 39.7. (a) Bowmans membrane (BM)

forms an acellular interface between the basal
cells of the epithelium (E) with its basement
membrane (straight arrow) and the lamellar
stroma (curved arrow). Note the relative
thickness and feltwork-like appearance of
Bowmans membrane. 5800. (b) The random
arrangement of collagen brils (arrowheads) is
shown. Also note the close association of the
hemidesmosomal structures (large arrow) on
the basal aspect of the epithelial cells, the
highly organized extracellular matrix (small
arrow) of these cells, and Bowmans
membrane. 31 000.
428
FIGURE 39.9. Low-magnication electron micrograph illustrating the

posterior portion of the cornea. The stroma (S) is closely associated
on its posterior-most aspect with Descemets membrane (DM), the
thick extracellular matrix secreted by the endothelial cells (EN). The
endothelium is the monolayer of cells located at the posterior of the
cornea; it acts as a barrier between the aqueous humor and overlying
corneal tissues. 750.
ENDOTHELIUM
FIGURE 39.8. Sections of corneal stroma showing collagen bundles

arranged in lamellae (L), which are oriented at different angles.
(a) Micrograph illustrating the stacked lamellae and long, attenuated
processes (arrowheads) of the stromal broblasts (F) located between
the lamellae. 4800. (b) Collagen bundles in the upper lamella (L) are
sectioned crosswise, whereas those in the lower lamella are sectioned
at an angle. Junctions between the cytoplasmic processes of
neighboring broblasts form a network of communicating cells. 13 000.
of incoming light results from equal spacing of the collagen

bers. Scattered light waves interact in an ordered fashion,
eliminating destructive interference. The lamellar organization
of the stroma also produces a uniform tensile strength across
the cornea, withstanding intraocular pressure and maintaining
appropriate corneal curvature.
The matrix components of the lamellar stroma are secreted
and maintained by stromal broblasts, also known as
keratocytes. As shown in Figure 39.8b, these long, attenuated
cells are arranged parallel to the corneal surface and are located
between the collagen lamellae. The keratocyte cell body
contains an elaborate rough endoplasmic reticulum and Golgi
apparatus, reflecting its active synthetic function. Keratocytes
extend slender cytoplasmic processes and can form gap
junctions with neighboring cells, resulting in a network of communicating cells.46 An ultrastructural study47 of human cornea
demonstrated the presence in central stroma of unmyelinated
nerve bers that run parallel to the collagen bundles, pass
through Bowmans membrane and the basal lamina of the
epithelium, and associate with subepithelial cells. Nerve bers
were found to invaginate stromal keratocytes as well as corneal
epithelial cells. This nding suggests that nerves may mediate
information exchange between the epithelium and stroma
under certain conditions, such as corneal wounding. Recently,
bone marrow-derived cells were demonstrated in the corneal
stroma.48 These cells were of both monocytic and myeloid
lineages, demonstrating surface markers of the dendridic cell
(antigen-presenting cells) and monocytes/macrophage type. It is
not clear whether these cells function in immunologic defense
or play a role in the induction of tolerance and the immuneprivileged state of the cornea.48
MORPHOLOGIC AND ULTRASTRUCTURAL

CHARACTERISTICS
The average density of corneal endothelial cells at birth is
~4000 cells/mm2.51 Each cell is 46 mm thick, ~20 mm wide,
and has a surface area of ~250 mm2. Scanning electron microscopy of the monolayer surface (Fig. 39.10) reveals that cells
assume a hexagonal shape and possess numerous lateral,
interdigitating cellular processes.5153 These processes increase
the area of contact between neighboring cells and resemble
interlocking ngers. Numerous small microvilli are present on
the posterior (apical) cell surface, which faces the aqueous
humor. This surface appears to be covered by a mucinous layer
~0.5 mm thick.54 MUC1 is at least one component of this layer
and is believed to have a protective function.55 A single, centrally located cilium, ~27 mm long, has been observed on the
apical surface of peripheral cells. This cilium exhibits the
ultrastructural characteristics of other primary cilia,56 but its
function in corneal endothelium is unclear.
The ultrastructural features of the endothelium reflect its
functions.57 Numerous mitochondria within the cytoplasm
indicate that these cells are metabolically active (Fig. 39.11).
The cytoplasm also contains extensive rough and smooth
endoplasmic reticulum, numerous ribosomes, and a prominent
Golgi apparatus reflective of a high level of protein synthesis
(Fig. 39.12). A circumferential band of actin-containing microlaments is located beneath the apical plasma membrane at the
cell periphery. These microlaments help maintain cell shape
and mediate cell movement.5860 An intermediate lament
CHAPTER 39
The endothelium (Fig. 39.9) is the single layer of cells located at

the posterior of the cornea that forms a barrier between the
corneal stroma and anterior chamber. The endothelial cell
monolayer acts as a leaky barrier, permitting the passage of
nutrients from the aqueous humor into the avascular
cornea.49,50 The endothelium is responsible for maintaining the
relatively low level of stromal hydration required for corneal
transparency. The tendency of the corneal stroma to swell is
balanced by removal of excess stromal fluid via the activity of
ionic pumps located at the endothelial plasma membrane. The
relatively high extracellular ion concentration produced by
these pumps draws water from the stroma, thus maintaining
the highly organized collagen lamellar structure required for
corneal transparency. The endothelium also secretes components of the thick basal lamina, termed Descemets membrane, which lies between the endothelium and posterior
stroma.
429
FIGURE 39.10. Scanning electron micrograph of the surface of the

corneal endothelium illustrating the hexagonal shape of the cells as
well as the other surface features, including nuclei (N) that bulge from
the cell surface, a single cilium (C), and long lateral projections (BB)
that bridge from one cell onto the body of adjacent cells.
SECTION 6
From Svedbergh B, Bill A: Scanning electron microscopic studies of the corneal

endothelium in man and monkeys. Acta Ophthalmol Scand 1972; 50:321.
430
network comprised primarily of vimentin forms a basket-like

structure that surrounds the nucleus and anchors at the apical
junctions.5962 This network appears to be responsible for
nuclear centration and, in part, for maintenance of cellcell
junction stability. Relatively little is known about the molecular
basis of adhesion of the endothelium to Descemets membrane.
Focal areas of increased electron density are present on the
cytoplasmic aspect of the basal plasma membrane63 and may
represent a form of adhesion plaque anchoring the endothelium
to Descemets membrane. Cytoplasmic processes extend from
the basal aspect of the cells and penetrate Descemets
membrane, possibly contributing to increased adhesiveness of
the monolayer.63 Alpha-v beta-5 integrin has been identied in
the endothelium of human corneas by immunocytochemistry64
and may help mediate endothelial adhesion to Descemets
membrane.
Focal tight junctions (Fig. 39.12) on the apical aspect of the
lateral membranes are small areas in which the outer leaflets
of the plasma membranes of adjacent cells appear to fuse,
obliterating the extracellular space.63,6567 These junctional
complexes do not form belts or rings extending around the cell,
as are found in many epithelia.68,69 Rather, they occur as small
zones of membrane fusion around the cell circumference. There
have been few studies to specically identify the protein
constituents of tight junctions in corneal endothelium;
however, it is known that ZO-1 (zonula occludens-1)60,70 and
occludin60 are components of these structures. In fact, focal
tight junctions can be visualized by the discontinuous
immunostaining of ZO-1. Electrical resistance across the endothelial monolayer is low (73 6 W/cm2)71 compared to that
across the corneal epithelium (1.69.1 KW/cm2), reflecting the
different organization of tight junctions in these two tissues.71,72
Adherens junctions are also located at the lateral plasma
membrane. These appear to be composed of N-cadherin,73
alpha- and beta-catenin, and plakoglobin.60 Gap junctions
FIGURE 39.11. Low-magnication transmission electron micrograph

illustrating the general orientation and ultrastructural features of the
corneal endothelium and Descemets membrane. A band of actincontaining laments, termed the terminal web (tw), is present in the
anterior aspect of the cells and excludes other cell organelles.
Junctional complexes located on the apical aspect of the lateral
plasma membranes are visible at this low magnication as a terminal
bar (tb). The intercellular border (ic) formed between adjacent cells is
long and sinuous. A, anterior chamber; E, endothelium; n, nucleus;
D, Descemets membrane. Bar = 1 mm.
From Iwamoto T, Smelser GK: Electron microscopy of the human corneal
endothelium with reference to transport mechanisms. Invest Ophthalmol Vis Sci
1965; 4:270.
(Fig. 39.13) are located at all levels of the lateral plasma

membrane below the tight junctions.63,65 These junctions stain
positively for connexin-43,74,75 possess a characteristic pentalaminar structure, and are the site of electrical and metabolic
coupling, which facilitates cell-to-cell communication.75,76
BARRIER FUNCTION
As an avascular tissue, the cornea receives oxygen mainly from
the tear lm,77 but its nutritional requirements are met via the
aqueous humor. As such, the glucose, amino acids, vitamins,
etc., needed by the epithelial cells and stromal keratocytes
must traverse the corneal endothelial monolayer. This nutrient
transport occurs primarily via a paracellular route, i.e., solutes
move between the cells rather than by being actively
FIGURE 39.12. Anterior portion of an endothelial cell illustrating the

extensive endoplasmic reticulum (ER) with its associated ribosomes
as well as focal tight junctional complexes (arrows) and apical folds
(AF) where adjacent cells interdigitate. AC, anterior chamber; TW,
terminal web. 80 000.
From Hirsch M, Renard G, Faure JP, et al: Formation of intercellular spaces and
junctions in regenerating rabbit corneal endothelium. Exp Eye Res 1976; 23:385.
FIGURE 39.14. The pump-leak hypothesis. When the rate of fluid

leakage into the stroma is balanced by the rate of fluid pumped out of
the stroma, normal corneal architecture and thickness are maintained.
Adapted from Waring GO III, Bourne WM, Edelhauser HF, Kenyon KR: The
corneal endothelium: normal and pathologic structure and function.
the requirement that fluid must move between the interdigitating lateral membranes, helps prevent bulk fluid
movement across the endothelial monolayer.
FIGURE 39.13. Electron micrograph of gap junctional complexes

illustrating the characteristic regular spacing of the connexin crossbridges that draw adjacent plasma membranes into close apposition.
Inset, Arrowheads indicate areas in which the gap between cell
membranes is clearly visible. 180 000.
From Leuenberger PM: Lanthanum hydroxide tracer studies on rat corneal
endothelium. Exp Eye Res 1973; 15:85.
transported through them. This form of transport requires that

the endothelial monolayer be leaky to substances within the
aqueous humor, but not permit bulk fluid flow into the corneal
stroma. The barrier to bulk flow of fluid from the aqueous
humor to the stroma is formed primarily by the focal tight
junctions of the endothelium. Experiments with molecular
tracers indicate that small molecules do not penetrate the tight
junctions, but enter the intercellular spaces by leaking around
them.63,65,67,69 Gap junctions and the sinuous, elaborate interdigitation of the lateral plasma membranes together may form
a secondary barrier to fluid flow.78 Gap junctions narrow the
width between apposing cell membranes from the normal
intercellular gap of 2540 nm to ~3 nm.6567 Narrowed intercellular spaces produced by the formation of gap junctions, plus
Transparency is essential for the function of the cornea as the

primary lens of the eye. Transparency results from the
uniformity of the tissue elements comprising the cornea and
from the regularity of their spatial organization. Precise
arrangement of the collagen bundles within the corneal stroma
is especially important for corneal clarity.45 This precise
arrangement depends to a great extent on the maintenance of a
relatively low level of stromal hydration. Proteoglycans
associated with the collagen brils within the stroma bind
water, producing a natural pressure gradient across the
endothelial monolayer. In addition, loss of integrity of the endothelial cell layer can hydrate the stroma. The disorganization
of collagen brils, which results from stromal swelling, causes
light absorbance, corneal clouding, and reduced vision.
The requirement that the endothelium permits passage of
nutrients into the cornea and, at the same time, maintain a
barrier to the free flow of water into the stroma presents an
interesting cell biological paradox. The pump-leak hypothesis
has attempted to resolve this paradox. It states that the rate of
leakage of water and solutes into the corneal stroma is balanced
by the rate of pumping of excess water from the stroma back to
the aqueous.5,79 As long as the equilibrium suggested by this
hypothesis is maintained, the corneal stroma remains relatively
dehydrated and corneal clarity is maintained. Figure 39.14
illustrates this equilibrium. Any imbalance between the rate of
CHAPTER 39
PUMP FUNCTION
431

fluid leak into and the rate of ionic pumping of fluid out of the
cornea results in corneal swelling and loss of visual acuity.
The endothelium maintains a low level of stromal hydration
by the activity of ionic pumps, which mediate the transfer of
Na+, K+, Mg+, Cl, and HCO3. Fluid flow from the stroma to
aqueous humor appears to be secondary to electrolyte movement;80 however, the specic mechanism by which movement
of electrolytes is coupled to the movement of water is not
completely understood.8082 Metabolic energy is needed to
maintain normal corneal thickness, indicating that at least
part of the mechanism regulating stromal hydration involves
an active process.83,84 The fluid pump is dependent on the
presence of Cl and HCO3 and can be slowed by carbonic
anhydrase inhibitors.82,85 A number of anion transport
mechanisms have been identied in corneal endothelium,
including a basolateral Na+K+2Cl co-transporter86 and a
Na+HCO3 co-transporter.82,87 In addition, the water channel
protein, aquaporin-1 (AQP1), has been localized to the plasma
membrane of corneal endothelium;88 however, it is currently
unclear how it functions in fluid transport in this tissue.80,89,90
SECTION 6
MONOLAYER REPAIR
432
Corneal endothelial cells are capable of normal division during

fetal development; however, the total corneal endothelial cell
reserve is limited, because cell division in adult cells either does
not occur at all or occurs at a rate too slow to efciently replace
dead or injured cells.51,9193 At birth, endothelial cell density is
35004000 cells/mm2, whereas, in adults it is reduced to
14002500 cells/mm2.51 Cell density begins to decrease during
fetal development due to both a rapid growth in corneal size and
the limited mitotic activity that occurs after the second trimester. Once rapid corneal growth subsides, cell density continues to decrease, but at a slower rate. Beginning at about the
second year of life, decreased cell density is directly related to
endothelial cell loss and the inability of the endothelium to
reproduce in numbers sufcient to keep pace with this loss.91,92
The overall rate of cell loss accelerates if the endothelium is
injured as the result of trauma, disease, or dystrophy.9395
Polymegathism, i.e., heterogeneity in cell size, increases in
the endothelium with age and as the result of damage caused
by trauma, corneal infection, or disease.9699 Cell size can
become heterogeneous for several reasons. When the endothelium is injured or when cells are lost due to normal attrition,
repair of the defect in the monolayer occurs mainly through
enlargement and spreading of neighboring cells, causing cells to
be larger in these areas.94,96,100103 In addition, the number of
multinucleated cells93,104 and cells with more than 4N DNA
content105,106 increase with age, producing a population of very
large cells. Increased heterogeneity in cell shape, i.e., pleomorphism, also occurs with age or trauma.52,107109 As the number
of cells within the monolayer decreases and the size of cells
enlarges, there is a decrease in the percentage of hexagonal cells
within the monolayer. As polymegathism and pleomorphism
increase, the endothelial monolayer can become destabilized. It
is well known that a regular hexagonal pattern provides the
greatest cellular packing with an optimal cell-to-membrane
ratio. Irregular cell sizes and shapes can increase surface tension
within the monolayer, producing decreased geometric and
architectural stability.
When cell numbers are reduced as the result of aging or
trauma and the remaining cells become larger and more pleomorphic, the ability to maintain or restore normal barrier and
pump function can be compromised.110112 With decreased
monolayer stability, permeability increases and the cornea
can swell. Decompensation, i.e., loss of monolayer integrity
and function, can occur when cell density falls below
300400 cells/mm2 or when the mean cell size reaches

~30003500 mm2.49,93 Because of the stressed state of the
endothelial monolayer under these conditions, the leak rate of
fluid into the stroma becomes greater than the pump rate of
fluid flow out of the stroma, producing stromal edema and
corneal clouding. At present, full-thickness transplantation is
the normal recourse for reestablishing corneal clarity and visual
acuity following decompensation of the corneal endothelium.
PROLIFERATIVE CAPACITY
Investigators are currently re-examining the relative proliferative capacity of corneal endothelial cells.113115 There is some
evidence to suggest that these cells can divide in vivo, but
at a very slow rate;104,116,117 however, the well-established
observation that endothelial cell density decreases with age
strongly suggests that, if endothelial cells do divide in vivo,
the rate of cell division does not keep pace with the rate of cell
loss. The ability to grow human corneal endothelial cells in
tissue culture without requiring viral oncogene protein
expression70,118,119 clearly indicates that these cells retain
proliferative capacity that can be harnessed under appropriate
conditions. Comparative studies of cell-cycle protein expression
in corneal cells suggest that endothelial cells in vivo are arrested
in G1-phase of the cell cycle.120,121 Thus, the limited proliferation observed in this tissue in vivo appears to be due, at
least in part, to microenvironmental conditions that actively
maintain the endothelium in a nonproliferative state.
A number of mechanisms appear to contribute to inhibition
of corneal endothelial cell proliferation in vivo. One is the
apparent lack of positive growth factor simulation. A number of
growth factors have been detected in aqueous humor,122125 and
corneal endothelial cells themselves both synthesize a number
of growth factors and express growth factor receptors.123,126,127
However, they do not appear to readily divide despite the
potential for autocrine or paracrine stimulation. Another
inhibitory mechanism appears to be suppression of S-phase
entry by transforming growth factor-b2 (TGF-b2). A role for
this cytokine in negatively regulating proliferation of corneal
endothelium is supported by the fact that TGF-b2 is present in
relatively high concentration in aqueous humor128,129 and that
corneal endothelial cells express the receptor types required to
transmit a TGF-b2-induced signal.130 Studies in cultured
corneal endothelial cells have demonstrated that both
exogenous TGF-b2 and TGF-b2 in aqueous humor suppress
S-phase entry.131,132 Contact inhibition is another mechanism
that suppresses proliferation in corneal endothelial cells. This
has been shown using an ex vivo wound model in which
treatment of the endothelium with the calcium/magnesium
chelator ethylenediaminetetraacetic acid (EDTA) releases
cellcell junctions and promotes cell division.133 The existence
of a connection between cellcell contacts and growth
inhibition in corneal endothelium is demonstrated by the fact
that expression of p27kip1, a protein that inhibits movement
from G1- to S-phase of the cell cycle, is upregulated when
cultured endothelial cells reach confluence.134,135
There are also intrinsic, age-related factors that affect the
ability of human corneal endothelial cells to proliferate. A
common nding has been that cells cultured from young donors
grow more robustly and can be passaged more times than cells
from older donors.136138 Results from both ex vivo wound
models139 and cell culture studies114,140 provide evidence that:
(1) Fewer cells from older donors are responsive to mitogenic
stimulation; (2) Those cells that retain the ability to respond to
mitogens generally require stronger stimulation than cells from
younger donors; and (3) Cells from older donors respond more
slowly to mitogens than cells from younger donors. Recent
relative percent of cells exhibiting replication competence was
signicantly higher in peripheral cornea compared with central
cornea, regardless of donor age. In corneas from older donors,
central endothelial cells exhibited the lowest percent of
replication competent cells. Interestingly, staining for
senescence-associated beta-galactosidase (SA-b-Gal)146 demonstrated an age-related increase in the relative percent of
endothelial cells exhibiting senescence-like characteristics, with
the percent of cells staining positively for SA-b-Gal highest in
the central region of corneas from older donors. Together, these
studies suggest that, with donor age, central endothelial cells
become senescent and die, and that there may be a slow
centripetal movement of cells from the periphery to replace
them. It is not yet clear whether peripheral cells are recruited
from the periphery as the result of slow cell movement and
rearrangement or whether peripheral cells may divide to help
replace cells lost from central endothelium. Recent studies147
have demonstrated telomerase activity in the far peripheral
region of human corneal endothelium, suggesting that this
region contains progenitor (stem-like) cells. It remains to be
clearly demonstrated that stem-like cells could act as a source
of cell renewal for corneal endothelium.
FIGURE 39.15. Micrograph illustrating Descemets membrane (DM)

located between the posterior aspect of the corneal stroma (S) and
the underlying endothelium (EN). Two regions of Descemets
membrane are apparent in adult corneas. The anterior banded region
(A) is secreted by the endothelial cells during fetal development and is
more highly organized than the posterior amorphous region (P),
which is secreted after birth. The posterior region increases in
thickness with age as a result of continued synthesis of its
constituents by the endothelium throughout life. 9600.
studies have shown that gene transfer to ex vivo human corneal

endothelium of E2F2, a transcription factor whose activity is
required for entry into S-phase, is able to induce proliferation
and increase endothelial cell density.141 This gene therapeutic
method was able to induce proliferation in cells from both
young and older donors; however, the kinetics of the induction
appeared to be age-dependent.
Morphometric studies142,143 have documented that there are
differences in endothelial cell density in peripheral versus central cornea, indicating a nonuniform distribution of endothelial
cells across the cornea. Cells in the far peripheral region of the
endothelium close to Schwalbes line exhibited a higher cell
density than cells in the paracentral or central regions.143
Although cell densities from all regions decreased with age, the
rate of decrease in density was slowest in the peripheral region.
Recently, the intriguing question has been raised whether there
may be a difference in relative proliferative capacity between
endothelial cells located in central cornea and those located in
the periphery. Tissue culture studies by Konomi et al144 and
ex vivo cornea studies by Mimura and Joyce145 have demonstrated that human corneal endothelial cells cultured from both
the central and peripheral regions exhibit proliferative capacity,
regardless of donor age. In the ex vivo cornea studies, the
Descemets membrane is the thick extracellular matrix

synthesized and secreted by the corneal endothelium
(Fig. 39.15). In adults this matrix consists of two layers. An
anterior, banded layer is formed during fetal development and
consists of highly organized collagen lamellae and proteoglycans. A posterior amorphous layer is synthesized after birth
and is less organized than the fetal layer. Adult Descemets
membrane contains bronectin, laminin, type IV and type VIII
collagen, as well as heparan sulfate and dermatan sulfate
proteoglycan. How these constituents are assembled to form
the highly ordered lattice of the fetal membrane and the more
randomly organized adult membrane remains unresolved.148
Corneal endothelial cells slowly synthesize and secrete basement membrane material throughout life. In young adults the
posterior layer measures ~2 mm, but increases to ~10 mm in
older individuals. The positive correlation between age and
Descemets membrane thickness149,150 suggests that there is
little, if any, destruction of previously formed basement membrane material. This provides a type of historic record of corneal
endothelial function and has been used to study the development of endothelial diseases or dystrophies. By comparing the
morphology and thickness of Descemets membrane in normal
and diseased corneas, it is possible to determine the relative
point in time in which the ability of corneal endothelial cells to
synthesize and secrete normal Descemets membrane is
compromised. Individual endothelial cells can produce excess
extracellular matrix material, resulting in the formation of
focal or nodular thickenings in Descemets membrane. These
thickenings, called HassallHenle bodies or warts, are frequently found in cells at the corneal periphery.51 Similar
structures are termed guttatae when they are located centrally
within the cornea.151 The number of these focal thickenings
increases with age, in certain endothelial dystrophies, such
as Fuchs dystrophy,152154 and as the result of inflammation.155
CHAPTER 39
DESCEMETS MEMBRANE
LIMBUS
The various anatomic denitions of the limbus include the
anatomists limbus, the pathologists limbus, the histologists
limbus, and the surgeons limbus.2,156 The various denitions
and the various angles of lines drawn on sections or diagrams of
cross-sections of the region indicate that there are no denite
433

FIGURE 39.16. Micrographs of the limbal
epithelium and subjacent connective tissue.
The basal cells of the region are smaller and
less columnar than those of the cornea. (a)
Light micrograph showing the absence of
Bowmans layer and the cellular stroma and
blood vessels (bv). 300. (b) Scanning electron
micrograph of palisades of Vogt (pv); limbal
epithelium was removed with
ethylenediaminetetraacetic acid to demonstrate
the folds or ridges in the connective tissue.
Remnant epithelium (e) and denuded corneal
basement membrane (cbm) are labeled. 2000.
(c) Electron micrograph of basal cells of the
limbal region. Note their small size and the
hemidesmosomes (hd) present along the
undulating basement membrane. Small, retelike
pegs (pg) of stromal matrix extend into the
epithelium. 6000.
SECTION 6
reliable boundaries to the zone. The broadest denition of the

limbus is the zone between a line drawn between the termini
of Bowmans layer and Descemets membrane, which forms
the anterior border, and a line that passes parallel but 1 mm
posterior to the anterior line, passing through the posterior end
of Schlemms canal. In this denition, both Schlemms canal
and the trabecular meshwork are within the limbus. This
section reviews aspects of the limbus relevant to ocular surface
function, specically the supercial region, including the
epithelium and loose connective tissue overlying the interface
of the connective tissue at the corneoscleral junction (see
Fig. 39.1b). The limbal region has been termed the transition
zone between cornea, conjunctiva, and sclera, and although that
may be an apt description, the specialized characteristics of the
limbal epithelium and its immediate subjacent connective
tissue indicate that the region has specialized functions supporting the cornea and that it may be a barrier to conjunctival
overgrowth of the cornea.
EPITHELIUM
434
The epithelium of the limbus has many features common to

corneal epithelium. It is a stratied squamous nonkeratinizing
epithelium but has several more cell layers than corneal
epithelium.2,157 Cell junctions in the limbus are similar to
those in the cornea, and the apical and basal specializations
present in the limbus are the same as those in the cornea. The
basal cells of the limbal epithelium appear unique and are
believed to be stem cells for maintenance of the corneal epithelium. (For review, see Gipson and Sugrue.1) The cells appear
smaller and less columnar and have more cytoplasmic
organelles (Fig. 39.16). The basal cells sit on a basement

membrane that is not flat and planar like that of the cornea;
peglike interdigitations of the epithelium and the stroma are
present (see Fig. 39.16).2 Probably the best indication to date
that the adult stem cells are a subpopulation of limbal basal
cells is their long 3H-thymidine label retention time, indicating
slower passage through the cell cycle than basal cells of the
cornea and conjunctiva.11 Other recent data using GFP mice
show a slow centripetal movement of successive progeny of
the limbal basal cells toward the center of the cornea.158 There
are differences in keratin expression in basal cells compared
with suprabasal cells of the limbus and cells of the corneal
epithelium;16 and they also show enhanced presence of certain
metabolic enzymes and proteins, including a-enolase, cytochrome oxidase, Na+,K+-ATPase, carbonic anhydrase, metallothionein, and glucose transporter.159162 Recent data suggest
that the stem cell population binds antibodies to the membrane-transporter protein designated ABCG2.163 Another
characteristic of the region is that ocular surface tumors occur
primarily in the limbal area and rarely are found on the cornea.
Taken together, these data and those from experiments
demonstrating centripetal migration of cells from the limbal
region into the cornea over time indicate that the limbal basal
cells are the stem cells of the corneal epithelium. Further
evidence that these basal cells are important to maintenance of
the corneal epithelium comes from clinical data that
demonstrate the effectiveness of limbal transplantation in the
treatment of persistent, nonhealing corneal problems.164,165 In
addition, these basal cells are protected by pigmentation and are
present within deep crypts in the limbal connective tissue,
termed the palisades of Vogt (see Fig. 39.16b).
Anatomy and Cell Biology of the Cornea, Superficial Limbus, and Conjunctiva
FIGURE 39.17. Micrographs demonstrating

regions of bulbar conjunctiva. In the bulbar
region, particularly in the nasal zone, goblet
cells are dense. They can occur in crypts or
groups, which have the appearance of acini, as
demonstrated in the light micrograph (a). 750.
(b) Electron micrograph of the apical region of
two adjacent goblet cells. Note the microvilli
(mv) on the surfaces of cells and the brillar
pattern in the mucin packets (mp). 21 000.
(c) Low-magnication electron micrograph of
nongoblet cells in the conjunctiva. Note the
vesicles (v) and granules (g) in the apical region
of cells and clumping of keratin laments into
bundles (k). 6000.
The connective tissue underlying the limbal epithelium is loose

and less organized than the corneal stroma, and Bowmans layer
is not present. Although the molecular composition of the
two matrices appears to correspond, an exception is absence of
the keratan sulfate proteoglycan (lumican).166 Cellular elements
within the limbal stroma are more diverse than in the corneal
stroma. In addition to broblasts, melanocytes, mast cells,
lymphocytes, and plasma cells occur routinely. A major
difference between the limbal stroma and that of the cornea is
the presence of blood and lymphatic vessels that loop into the
area of the limbal stroma. These vessels include capillaries,
small arterioles and venules, and large lymphatics. Bundles of
unmyelinated nerves also are present. The palisades of Vogt,
large folds of matrix, are a unique characteristic of this area (see
Fig. 39.16b). The outward folds of connective tissue are large
enough to accommodate small blood vessels, lymphatics, and
nerves, and crypts of limbal epithelium reach down into the
palisades of Vogt. The deep housing of the limbal epithelium
in the folds not only may protect the stem cell population but
also may increase the surface area for accommodating a large
cell population and increase exposure to vascularly derived
nutrients and effector molecules. In addition to the large
macroscopic folds of the palisades of Vogt, tiny rete (peglike
folds or outpockets of stroma) begin in the peripheral stroma
and extend through the limbal region into the conjunctiva (see
Fig. 39.16c). These rete may increase the surface area of the
basal cell membrane of basal cells and may provide for
additional anchoring strength in a region where hemidesmosomes are not as extensive.167
CONJUNCTIVA
GENERAL CHARACTERISTICS AND
DESCRIPTION OF REGIONS
The conjunctiva is the mucous membrane that covers the inner
surfaces of the upper and lower lids and extends to the limbus
on the surface of the globe. The two major functions of this
tissue, besides connecting the lids to the globe, are provision of
mucus for the tear lm and protection of the ocular surface
from pathogens through immune tissue. The ducts of the
lacrimal, accessory lacrimal, and meibomian glands enter the
conjunctival epithelium and deliver their respective products to
the tear lm. Three regions within the conjunctiva are
recognized: the palpebral or tarsal region, which lines the inner
surface of the lids; the fornical region, which lines the upper and
lower surfaces of the recess or cul-de-sac known as the fornix;
and the bulbar region, which lines the surface of the sclera
between the fornix and the limbus. The conjunctiva has two
structural components throughout all regions: the surface
epithelium and the substantia propria (Fig. 39.17; see
Fig. 39.1c).
CHAPTER 39
CONNECTIVE TISSUE
435

FIGURE 39.18. Distribution of messenger RNA
for the mucin MUC5AC using a 35S-labeled
probe. (a) Low-magnication micrograph
demonstrating dense signal distributed in
patches (arrows) within the conjunctival
epithelium. (b) Higher magnication of the
epithelium demonstrating label specically over
goblet cells. (c) The control sense probe
showed no binding. Bars = 100 mm.
SECTION 6
EPITHELIUM
436
Conjunctival epithelium is unique among stratied nonkeratinizing epithelia in that it has goblet cells intercalated
within it (see Figs 39.1 and 39.17). The goblet cells are the
major producers of mucins for the tear lm. Reports of the
number of cell layers in the stratied epithelium vary, especially
regarding fornical and bulbar areas. These variations may result
from different degrees of stretch on the tissue at time of xation
for histologic study. Cell layers of the palpebral conjunctival
epithelium do not vary as much, perhaps because the substantia
propria is not as loose and contractile at xation. Reports have
varied from 2 or 3 cell layers to 1012, the latter number of
layers being present at the lid margin near the junction with the
epidermis covering the external lid. Langerhans cells are
present within the conjunctival epithelium.48
Compared with cells of the corneal epithelium, the stratied
cells of the conjunctiva have more cytoplasmic organelles.
Keratin laments in these cells are not as dispersed as those in
corneal cell cytoplasm and often appear in bundles (see
Fig. 39.17). Keratin proteins expressed by stratied conjunctival
epithelial cells also are different, with the keratin pairs K4 and
K13, and K3 and K19. K7 is expressed by goblet cells.168 Cell-tocell junctions and cell-to-substrate junctions appear similar in
corneal, limbal, and conjunctival epithelia, except that gap
junction proteins in the conjunctiva have not been
characterized.169171
The apical cells of the stratied epithelium have numerous
small vesicles within their cytoplasm (see Fig. 39.17). It has
been proposed that these vesicles (which bind Alcian blue
and periodic acid-Schiff stains, indicating a highly glycosylated
content) deliver mucins onto the ocular surface and thus
represent a second source of mucus for the tear lm.
Reports indicate that the stratied epithelium is expressing
membrane-spanning mucins MUC1,26 MUC4,172,173 and
MUC16.26
The goblet cells that are intercalated within the stratied
epithelium of the human conjunctiva occur as individual cells;
in rodents, they occur as clusters.174 In humans, there is a regional
variation in goblet cell distribution pattern and density,174 the
highest density being in the palpebral region near the tear
drainage punctum and in the midfornix. In some regions,
especially the temporal bulbar conjunctiva, goblet cell density is

so great that the cells appear to be clustered and arranged in
acini.175 Goblet cells of the conjunctiva are plump and lack the
goblet stem a thin cytoplasmic extension to the basement
membrane, that is obvious in intestinal goblet cells. Mucin
packets that ll the cytoplasm of goblet cells appear electron
lucent; however, a ne lamentous network can be discerned
within the packets (see Fig. 39.17b). Studies have demonstrated
that a major mucin gene expressed by the conjunctival goblet
cell is the large gel-forming mucin MUC5AC173 (Fig. 39.18).
Tight junctions appear to be present between goblet cells and
adjacent stratied cells (see Fig. 39.17).
With the accumulation of data indicating that the basal cells
of the limbal epithelium are the stem cells for the corneal
epithelium, interest has been generated in the location of the
stem-cell population in the conjunctiva. If stem cells are present
within the conjunctiva, do the stratied epithelial cells and the
goblet cells derive from the same stem cell population? Data
suggest that slow-cycling stem-like cells are present in the
fornical region of the rabbit conjunctiva.176 More recently, data
from observation of conjunctival epithelial cells of GFP mice,
and of BRDU-labeled cells suggest that epithelial stem cells
in bulbar conjunctiva are evenly distributed.177 In addition,
clonal cultures of conjunctival epithelium injected subdermally
into nude mice produce cysts that contain both goblet cells and
stratied cells, indicating that stem cells of the conjunctival
epithelium are pluripotent and can give rise to both cell
types.178 It is not known what causes divergence of the
differentiation pathway to give rise to the two cell types.
The connective tissue of the substantia propria of the
conjunctiva is similar to that of the supercial limbus; immune
cells are especially abundant in its loose and highly vascularized
connective tissue. Lymphocytes, mast cells, plasma cells, and
neutrophils are common cell types in its matrix.157 In fact, the
substantia propria has been described as having two layers: an
inner brous layer and an outer lymphoid layer. Although the
lymphoid layer has dense accumulations of lymphocytes, these
do not form lymph nodules. The accumulation of the lymphoid
tissue, in addition to the phagocytic abilities of the conjunctival
epithelium, demonstrates the function of the tissue in dealing
with infectious agents.3
1. Gipson IK, Sugrue SP: Cell biology of the

corneal epithelium. In: Albert DM, Jakobiec
FA, eds. Principles and practice of
ophthalmology, basic sciences.
Philadelphia, PA: Lippincott; 1994.
2. Hogan MJ, Alvarado JA, Wedell JE:
Histology of the human eye. Philadelphia,
PA: W B Saunders; 1971.
3. Jakobiec FA, Iwamoto T: Ocular adnexa:
introduction to lids, conjunctiva and orbit.
In: Jakobiec FA, ed. Ocular anatomy,
embryology, and teratology. Philadelphia,
PA: Harper & Row; 1982.
4. Gipson IK: Anatomy of the conjunctiva,
cornea, and limbus. In: Smolin G, Thoft RA,
eds. The cornea. Boston, MA: Little, Brown
and Company; 1994.
5. Maurice DM: The cornea and the sclera.
3rd edn. London: Academic Press; 1985.
6. Warwick R: Eugene Wolffs anatomy of the
eye and orbit. 7th edn. London: Chapman
& Hall Medical; 1997.
7. Sax CM, Salamon C, Kays WT, et al:
Transketolase is a major protein in the
mouse cornea. J Biol Chem 1996;
271:3356833574.
8. Friend J, Hassell J: Biochemistry of the
cornea. In: Smolin G, Thoft R, eds. The
cornea. Boston, MA: Little, Brown; 1994.
9. Muller LJ, Marfurt CF, Kruse F, Tervo TM:
Corneal nerves: structure, contents and
function. Exp Eye Res 2003; 76:521542.
10. Hanna C, OBrien JE: Cell production and
migration in the epithelial layer of the
cornea. Arch Ophthalmol 1960; 64:536.
11. Lavker RM, Dong G, Cheng SZ, et al:
Relative proliferative rates of limbal and
corneal epithelia. Implications of corneal
epithelial migration, circadian rhythm, and
suprabasally located DNA-synthesizing
keratinocytes. Invest Ophthalmol Vis Sci
1991; 32:18641875.
12. Beebe DC, Masters BR: Cell lineage and
the differentiation of corneal epithelial cells.
37:18151825.
13. Sun T-T, Green H: Cultured epithelial cells
of cornea, conjunctiva and skin: absence of
marked intrinsic divergence of their
differentiated states. Nature 1976; 269:489.
14. Doran TI, Vidrich A, Sun T-T: Intrinsic and
extrinsic regulation of the differentiation of
skin, corneal and esophageal epithelial
cells. Cell 1980; 22:1725.
15. Kurpakus MA, Stock EL, Jones JCR:
Expression of the 55-kD/64-kD corneal
keratins in ocular surface epithelium. Invest
16. Liu CY, Zhu G, Converse R, et al:
Characterization and chromosomal
localization of the cornea-specic murine
keratin gene Krt1.12. J Biol Chem 1994;
269:2462724636.
17. Green KJ, Bohringer M, Gocken T, Jones
JC: Intermediate lament associated
proteins. Adv Protein Chem 2005;
70:143202.
18. Gipson IK, Anderson RA: Actin laments in
normal and migrating corneal epithelial
16:161166.
19. Amos LA, Schlieper D: Microtubules and
maps. Adv Protein Chem 2005; 71:257298.
20. Gipson IK, Westcott MJ, Brooksby NG:
Effects of cytochalasins B and D and
colchicine on migration of the corneal
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
epithelium. Invest Ophthalmol Vis Sci 1982;

22:633642.
Alberts B, Bray D, Lewis J, et al: Molecular
biology of the cell. 3rd edn. New York:
Garland; 1994.
Stepp MA, Rajendran V, Spurr-Michaud S,
Gipson IK: Integrins in the wounded and
unwounded stratied squamous epithelium
of the cornea. Invest Ophthalmol Vis Sci
1993; 34:18291844.
Nichols BA, Chiappino ML, Dawson CR:
Demonstration of the mucous layer of the
tear lm by electron microscopy. Invest
Pster RR: The normal surface of corneal
epithelium: a scanning electron
microscopic study. Invest Ophthalmol Vis
Sci 1973; 12:654668.
Pster RR: The normal surface of
conjunctiva epithelium. A scanning electron
microscopic study. Invest Ophthalmol
1975; 14:267279.
Gipson IK, Hori Y, Argueso P: Character of
ocular surface mucins and their alteration in
dry eye disease. Ocul Surf 2004; 2:131148.
Gipson IK, Spurr-Michaud SJ, Tisdale AS:
Anchoring brils form a complex network in
human and rabbit cornea. Invest
Gipson IK, Grill SM, Spurr SJ, Brennan SJ:
Hemidesmosome formation in vitro. J Cell
Biol 1983; 97:849857.
Rodrigues M, Waring G III, Hackett J, et al:
Cornea. In: Jakobiec F, ed. Ocular anatomy,
embryology, and teratology. Philadelphia,
Tisdale AS, Spurr-Michaud SJ, Rodrigues
M, et al: Development of the anchoring
structures of the epithelium in rabbit and
human fetal corneas. Invest Ophthalmol Vis
Sci 1988; 29:727736.
Gordon MK, Foley JW, Birk DE, et al: Type
V collagen and Bowmans membrane.
Quantitation of mRNA in corneal epithelium
and stroma. J Biol Chem 1994;
269:2495924966.
Nakayasu K, Tanaka M, Konomi H, Hayashi
T: Distribution of types I, II, III, IV and V
collagen in normal and keratoconus
corneas. Ophthalmic Res 1986; 18:110.
Li Y, Yi Y, Feng G, et al:
Glycosaminoglycans associated with
corneal transparency. Eye Sci 1991;
7:5862.
Birk DE, Fitch JM, Linsenmayer TF:
Organization of collagen types I and V in
the embryonic chicken cornea. Invest
Birk DE, Fitch JM, Babiarz JP, et al:
Collagen brillogenesis in vitro: interaction
of types I and V collagen regulates bril
diameter. J Cell Sci 1990; 95(Pt 4):649659.
Linsenmayer TF, Fitch JM, Birk DE:
Heterotypic collagen brils and stabilizing
collagens. Controlling elements in corneal
morphogenesis? Ann N Y Acad Sci 1990;
580:143160.
Keene DR, Sakai LY, Lunstrum GP, et al:
Type VII collagen forms an extended
network of anchoring brils. J Cell Biol
1987; 104:611621.
Trelstad RL, Hayashi K, Toole BP: Epithelial
collagens and glycosaminoglycans in the
embryonic cornea. Macromolecular order
and morphogenesis in the basement
membrane. J Cell Biol 1974; 62:815830.
39. Zimmermann DR, Trueb B, Winterhalter KH,

et al: Type VI collagen is a major
component of the human cornea. FEBS
Lett 1986; 197:5558.
40. Carlson E, Liu C-Y, Chikama T-I, et al:
Keratocan, a cornea-specic keratan
sulfate proteoglycan, is regulated by
lumican. J Biol Chem 2005;
280:2554125547.
41. Olsen B, McCarthy M: Extracellular matrix:
molecular structure of the sclera, cornea,
and vitreous body. In: Jakobiec FA, ed.
Principles and practice of ophthalmology.
Philadelphia, PA: WB Saunders; 1994.
42. Kobayashi M: Tissue localization and
possible function of type VI collagen.
J Anat 1995; 70:298306.
43. Nakamura M, Kobayashi M, Hirano K, et al:
Assembly of 100 nm periodic brils (type VI
collagen) in human infant corneal stroma.
44. Berman E: Cornea. In: Berman E, ed.
Biochemistry of the eye. New York:
Plenum; 1991.
45. Maurice DM: The structure and
transparency of the cornea. J Physiol 1957;
136:263.
46. Watsky MA: Keratocyte gap junctional
communication in normal and wounded
rabbit corneas and human corneas. Invest
47. Muller LJ, Pels L, Vrensen GF:
Ultrastructural organization of human
corneal nerves. Invest Ophthalmol Vis Sci
1996; 37:476488.
48. Hamrah P, Liu Y, Zhang Q, Dana MR: The
corneal stroma is endowed with a
signicant number of resident dendritic
44:581589.
49. Waring GO 3rd, Bourne WM, Edelhauser
HF, Kenyon KR: The corneal endothelium.
Normal and pathologic structure and
function. Ophthalmology 1982; 89:531.
50. Tuft S, Coster D: The corneal endothelium.
Eye 1990; 4:389424.
51. Svedbergh B, Bill A: Scanning electron
microscopic studies of the corneal
endothelium in man and monkeys. Acta
52. Rao GN, Lohman LE, Aquavella JV: Cell
size-shape relationships in corneal
endothelium. Invest Ophthalmol Vis Sci
1982; 22:271274.
53. Davanger M, Olsen EG: The corneal
endothelial cell interface. Acta Ophthalmol
(Copenh) 1985; 63:109115.
54. Kim EK, Cristol SM, Kim HL, et al: The
mucinous layer of corneal endothelial cells.
Yonsei Med J 2000; 41:651656.
55. Jung SE, Seo KY, Kim H, et al: Expression
of MUC1 on corneal endothelium of
human. Cornea 2002; 21:691695.
56. Gallagher BC: Primary cilia of the corneal
endothelium. Am J Anat 1980;
159:475484.
57. Marshall J, Grindle CF: Fine structure of the
cornea and its development. Trans
58. Olsen EG, Davanger M, Moen T: The role of
microlaments in the healing of the corneal
endothelium. Acta Ophthalmol (Copenh)
1985; 63:104108.
59. Barry PA, Petroll WM, Andrews PM, et al:
The spatial organization of corneal
endothelial cytoskeletal proteins and their
CHAPTER 39
REFERENCES
437
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
SECTION 6
71.
72.
73.
74.
75.
76.
438
relationship to the apical junctional

complex. Invest Ophthalmol Vis Sci 1995;
36:11151124.
Petroll WM, Hsu JK, Bean J, et al: The
spatial organization of apical junctional
complex-associated proteins in feline and
human corneal endothelium. Curr Eye Res
1999; 18:1019.
Hayashi K, Sueishi K, Tanaka K, Inomata H:
Immunohistochemical evidence of the
origin of human corneal endothelial cells
and keratocytes. Graefes Arch Clin Exp
Ophthalmol 1986; 224:452456.
Risen LA, Binder PS, Nayak SK:
Intermediate laments and their
organization in human corneal endothelium.
28:19331938.
Iwamoto T, Smelser GK: Electron
microscopy of the human corneal
endothelium with reference to transport
mechanisms. Invest Ophthalmol Vis Sci
1965; 4:270.
Rayner SA, Gallop JL, George AJ, Larkin
DF: Distribution of integrins alpha v beta 5,
alpha v beta 3 and alpha v in normal
human cornea: possible implications in
clinical and therapeutic adenoviral
infection. Eye 1998; 12(Pt 2):273277.
Leuenberger PM: Lanthanium hydroxide
tracer studies on rat corneal endothelium.
Exp Eye Res 1973; 15:8591.
Hirsch M, Renard G, Faure JP, Pouliquen Y:
Formation of intercellular spaces and
junctions in regenerating rabbit corneal
endothelium. Exp Eye Res 1976;
23:385397.
Kreutziger GO: Lateral membrane
morphology and gap junction structure in
rabbit corneal endothelium. Exp Eye Res
1976; 23:285293.
Hirsch M, Renard G, Faure JP, Pouliquen Y:
Study of the ultrastructure of the rabbit
corneal endothelium by the freeze-fracture
technique: apical and lateral junctions. Exp
Eye Res 1977; 25:277288.
Montcourrier P, Hirsch M: Intercellular
junctions in the developing rat corneal
endothelium. Ophthalmic Res 1985;
17:207215.
Chen KH, Azar D, Joyce NC:
Transplantation of adult human corneal
endothelium ex vivo: a morphologic study.
Cornea 2001; 20:731737.
Lim JJ, Fischbarg J: Electrical properties of
rabbit corneal endothelium as determined
from impedance measurements. Biophys J
1981; 36:677695.
Klyce SD, Wong RK: Site and mode of
adrenaline action on chloride transport
across the rabbit corneal epithelium. J
Physiol 1977; 266:777799.
Beebe DC, Coats JM: The lens organizes
the anterior segment: specication of
neural crest cell differentiation in the avian
eye. Dev Biol 2000; 220:424431.
Joyce NC, Harris DL, Zieske JD: Mitotic
inhibition of corneal endothelium in
neonatal rats. Invest Ophthalmol Vis Sci
1998; 39:25722583.
Williams K, Watsky M: Gap junctional
communication in the human corneal
endothelium and epithelium. Curr Eye Res
2002; 25:2936.
Rae JL, Lewno AW, Cooper K, Gates P:
Dye and electrical coupling between cells
of the rabbit corneal endothelium. Curr Eye
Res 1989; 8:859869.
77. Weissman BA, Fatt I, Rasson J: Diffusion of

oxygen in human corneas in vivo. Invest
78. Watsky MA, McCartney MD, McLaughlin
BJ, Edelhauser HF: Corneal endothelial
junctions and the effect of ouabain. Invest
79. Maurice DM: The location of the fluid pump
in the cornea. J Physiol 1972; 221:4354.
80. Fischbarg J: On the mechanism of fluid
transport across corneal endothelium and
epithelia in general. J Exp Zoolog A Comp
Exp Biol 2003; 300:3040.
81. Fischbarg J, Diecke FP: A mathematical
model of electrolyte and fluid transport
across corneal endothelium. J Membr Biol
2005; 203:4156.
82. Bonanno JA: Identity and regulation of ion
transport mechanisms in the corneal
endothelium. Prog Retin Eye Res 2003;
22:6994.
83. Harris J, Nordquist L: The hydration of the
cornea. I. The transport of water from the
84. Harris J: Symposium on the cornea.
Introduction: factors influencing corneal
hydration. Invest Ophthalmol Vis Sci 1962;
1:151.
85. Diecke FP, Wen Q, Sanchez JM, et al:
Immunocytochemical localization of Na+HCO3- cotransporters and carbonic
anhydrase dependence of fluid transport in
corneal endothelial cells. Am J Physiol Cell
Physiol 2004; 286:C1434C1442.
86. Kuang K, Li Y, Wen Q, et al: Corneal
endothelial NKCC: molecular identication,
location, and contribution to fluid transport.
Am J Physiol Cell Physiol 2001;
280:C491C499.
87. Sun XC, Bonanno JA, Jelamskii S, Xie Q:
Expression and localization of Na(+)HCO(3)() cotransporter in bovine corneal
endothelium. Am J Physiol Cell Physiol
2000; 279:C1648C1655.
88. Li J, Kuang K, Nielsen S, Fischbarg J:
Molecular identication and
immunolocalization of the water channel
protein aquaporin 1 in CBCECs. Invest
89. Sun XC, Allen KT, Xie Q, et al: Effect of
AQP1 expression level on Co(2)
permeability in bovine corneal endothelium.
42:417423.
90. Verkman AS: Aquaporin water channels
and endothelial cell function. J Anat 2002;
200:617627.
91. Murphy C, Alvarado J, Juster R, Maglio M:
Prenatal and postnatal cellularity of the
human corneal endothelium. A quantitative
histologic study. Invest Ophthalmol Vis Sci
1984; 25:312322.
92. Bahn CF, Glassman RM, MacCallum DK, et
al: Postnatal development of corneal
1986; 27:4451.
93. Landshman N, Ben-Hanan I, Assia E, et al:
Relationship between morphology and
functional ability of regenerated corneal
1988; 29:11001109.
94. Capella JA: Regeneration of endothelium in
diseased and injured corneas. Am J
Ophthalmol 1972; 74:810817.
95. Ikebe H, Takamatsu T, Itoi M, Fujita S:
Changes in nuclear DNA content and cell
size of injured human corneal endothelium.
Exp Eye Res 1988; 47:205215.
96. Honda H, Ogita Y, Higuchi S, Kani K: Cell

movements in a living mammalian tissue:
long-term observation of individual cells in
wounded corneal endothelia of cats. J
Morphol 1982; 174:2539.
97. Laing RA, Sandstrom MM, Berrospi AR,
Leibowitz HM: Changes in the corneal
endothelium as a function of age. Exp Eye
Res 1976; 22:587594.
98. Glasser DB, Matsuda M, Gager WE,
Edelhauser HF: Corneal endothelial
morphology after anterior chamber lens
103:13471349.
99. Hirose N, Shimomura Y, Matsuda M, et al:
Corneal endothelial changes associated
with herpetic stromal keratitis. Jpn J
100. Sherrard ES: The corneal endothelium in
vivo: its response to mild trauma. Exp Eye
Res 1976; 22:347357.
101. Doughman D, VanHorn D, Rodman W, et
al: Human corneal endothelial layer repair
during organ culture. Arch Ophthalmol
1976; 94:1791.
102. VanHorn D, Hyndiuk R: Endothelial wound
repair in primate cornea. Exp Eye Res
1975; 21:113.
103. Matsubara M, Tanishima T: Wound-healing
of corneal endothelium in monkey: an
autoradiographic study. Jpn J Ophthalmol
1983; 27:444450.
104. Laing RA, Neubauer L, Oak SS, et al:
Evidence for mitosis in the adult corneal
endothelium. Ophthalmology 1984;
91:11291134.
Cytofluorometric nuclear DNA determinations
on human corneal endothelial cells. Exp Eye
Res 1984; 39:497504.
Age-dependent changes in nuclear DNA
content and cell size of presumably normal
human corneal endothelium. Exp Eye Res
1986; 43:251258.
107. Rao GN, Shaw EL, Arthur E, Aquavella JV:
Morphological appearance of the healing
corneal endothelium. Arch Ophthalmol
1978; 96:20272030.
108. Ling TL, Vannas A, Holden BA: Long-term
changes in corneal endothelial morphology
following wounding in the cat. Invest
109. Schultz RO, Matsuda M, Yee RW, et al:
Corneal endothelial changes in type I and
type II diabetes mellitus. Am J Ophthalmol
1984; 98:401410.
110. ONeal MR, Polse KA: Decreased
endothelial pump function with aging.
27:457463.
111. Macdonald JM, Geroski DH, Edelhauser
HF: Effect of inflammation on the corneal
endothelial pump and barrier. Curr Eye Res
1987; 6:11251132.
112. Carlson KH, Bourne WM, McLaren JW,
Brubaker RF: Variations in human corneal
endothelial cell morphology and
permeability to fluorescein with age. Exp
Eye Res 1988; 47:2741.
113. Joyce NC: Proliferative capacity of the
corneal endothelium. Prog Retin Eye Res
2003; 22:359389.
114. Joyce N, Zhu C: Human corneal endothelial
cell proliferation: potential for use in
regenerative medicine. Cornea 2004;
23(Suppl 1):S8.
115. Joyce NC: Cell cycle status in human
116.
117.
118.
119.
120.
121.
122.
123.
124.
125.
126.
127.
128.
129.
130.
131.
132.
corneal endothelium. Exp Eye Res 2005;

81:629638.
Singh G: Mitosis and cell division in human
corneal endothelium. Ann Ophthalmol
1986; 18:8890.
Treffers WF: Human corneal endothelial
wound repair. In vitro and in vivo.
Engelmann K, Bohnke M, Friedl P: Isolation
and long-term cultivation of human corneal
endothelial cells. Invest Ophthalmol Vis Sci
1988; 29:16561662.
Yue BY, Sugar J, Gilboy JE, Elvart JL:
Growth of human corneal endothelial cells
in culture. Invest Ophthalmol Vis Sci 1989;
30:248253.
Joyce NC, Meklir B, Joyce SJ, Zieske JD:
Cell cycle protein expression and
proliferative status in human corneal cells.
37:645655.
Joyce NC, Navon SE, Roy S, Zieske JD:
Expression of cell cycle-associated
proteins in human and rabbit corneal
endothelium in situ. Invest Ophthalmol Vis
Sci 1996; 37:15661575.
Tripathi RC, Borisuth NS, Tripathi BJ:
Detection, quantication, and signicance
of basic broblast growth factor in the
aqueous humor of man, cat, dog and pig.
Exp Eye Res 1992; 54:447454.
Schulz MW, Chamberlain CG, de Iongh RU,
McAvoy JW: Acidic and basic FGF in
ocular media and lens: implications for lens
polarity and growth patterns. Development
1993; 118:117126.
Arnold DR, Moshayedi P, Schoen TJ, et al:
Distribution of IGF-I and -II, IGF binding
proteins (IGFBPs) and IGFBP mRNA in
ocular fluids and tissues: potential sites of
synthesis of IGFBPs in aqueous and
vitreous. Exp Eye Res 1993; 56:555565.
Araki-Sasaki K, Danjo S, Kawaguchi S, et
al: Human hepatocyte growth factor (HGF)
in the aqueous humor. Jpn J Ophthalmol
1997; 41:409413.
Wilson SE, Schultz GS, Chegini N, et al:
Epidermal growth factor, transforming
growth factor alpha, transforming growth
factor beta, acidic broblast growth factor,
basic broblast growth factor, and
interleukin-1 proteins in the cornea. Exp
Eye Res 1994; 59:6371.
Wilson SE, Walker JW, Chwang EL, He YG:
Hepatocyte growth factor, keratinocyte
growth factor, their receptors, broblast
growth factor receptor-2, and the cells of
the cornea. Invest Ophthalmol Vis Sci
1993; 34:25442561.
Jampel HD, Roche N, Stark WJ, Roberts
AB: Transforming growth factor-beta in
human aqueous humor. Curr Eye Res 1990;
9:963969.
Cousins SW, McCabe MM, Danielpour D,
Streilein JW: Identication of transforming
growth factor-beta as an immunosuppressive
factor in aqueous humor. Invest Ophthalmol
Vis Sci 1991; 32:22012211.
Joyce NC, Zieske JD: Transforming growth
factor-beta receptor expression in human
38:19221928.
Harris DL, Joyce NC: Transforming growth
factor-beta suppresses proliferation of
rabbit corneal endothelial cells in vitro. J
Interferon Cytokine Res 1999; 19:327329.
Chen KH, Harris DL, Joyce NC: TGF-beta2
in aqueous humor suppresses S-phase
133.
134.
135.
136.
137.
138.
139.
140.
141.
142.
143.
144.
145.
146.
147.
148.
149.
entry in cultured corneal endothelial cells.

40:25132519.
Senoo T, Obara Y, Joyce NC: EDTA: a
promoter of proliferation in human corneal
2000; 41:29302935.
Kim TY, Kim WI, Smith RE, Kay EP:
Differential activity of TGF-beta2 on the
expression of p27Kip1 and Cdk4 in actively
cycling and contact inhibited rabbit corneal
endothelial cells. Mol Vis 2001; 7:261270.
Joyce NC, Harris DL, Mello DM:
Mechanisms of mitotic inhibition in corneal
endothelium: contact inhibition and
TGF-beta2. Invest Ophthalmol Vis Sci
2002; 43:21522159.
Baum JL, Niedra R, Davis C, Yue BY: Mass
culture of human corneal endothelial cells.
Hyldahl L: Primary cell cultures from human
embryonic corneas. J Cell Sci 1984;
66:343351.
Nayak SK, Binder PS: The growth of
endothelium from human corneal rims in
tissue culture. Invest Ophthalmol Vis Sci
1984; 25:12131216.
Senoo T, Joyce NC: Cell cycle kinetics in
corneal endothelium from old and young
donors. Invest Ophthalmol Vis Sci 2000;
41:660667.
Zhu C, Joyce NC: Proliferative response of
corneal endothelial cells from young and
older donors. Invest Ophthalmol Vis Sci
2004; 45:17431751.
McAlister JC, Joyce NC, Harris DL, et al:
Induction of replication in human corneal
endothelial cells by E2F2 transcription
factor cDNA transfer. Invest Ophthalmol Vis
Sci 2005; 46:35973603.
Schimmelpfennig BH: Direct and indirect
determination of nonuniform cell density
distribution in human corneal endothelium.
25:223229.
Amann J, Holley GP, Lee SB, Edelhauser
HF: Increased endothelial cell density in the
paracentral and peripheral regions of the
135:584590.
Konomi K, Zhu C, Harris D, Joyce NC:
Comparison of the proliferative capacity of
human corneal endothelial cells from the
central and peripheral areas. Invest
Mimura T, Joyce N: Replication
competence and senescence in human
corneal endothelium from central and
peripheral areas. Invest Ophthalmol Vis Sci
2006; 47:1387.
Dimri GP, Lee X, Basile G, et al: A
biomarker that identies senescent human
cells in culture and in aging skin in vivo.
92:93639367.
Whikehart DR, Parikh CH, Vaughn AV, et al:
Evidence suggesting the existence of stem
cells for the human corneal endothelium.
Mol Vis 2005; 11:816824.
Sawada H, Konomi H, Hirosawa K:
Characterization of the collagen in the
hexagonal lattice of Descemets
membrane: its relation to type VIII collagen.
J Cell Biol 1990; 110:219227.
Murphy C, Alvarado J, Juster R: Prenatal
and postnatal growth of the human
Descemets membrane. Invest Ophthalmol
Vis Sci 1984; 25:14021415.
150. Johnson DH, Bourne WM, Campbell RJ:

The ultrastructure of Descemets
membrane. I. Changes with age in normal
corneas. Arch Ophthalmol 1982;
100:19421947.
151. Lorenzetti DW, Uotila MH, Parikh N,
Kaufman HE: Central cornea guttata.
Incidence in the general population. Am J
Ophthalmol 1967; 64:11551158.
152. Goar E: Dystrophy of the corneal
epithelium (corneal guttata), with a report of
a histological examination. Am J
153. Bourne WM, Johnson DH, Campbell RJ:
membrane. III. Fuchs dystrophy. Arch
Ophthalmol 1982; 100:19521955.
154. Brooks AM, Grant G, Gillies WE: A
comparison of corneal endothelial
morphology in cornea guttata, Fuchs
dystrophy and bullous keratopathy. Aust N
Z J Ophthalmol 1988; 16:93100.
155. Waring GO, Font RL, Rodrigues MM,
Mulberger RD: Alterations of Descemets
membrane in interstitial keratitis. Am J
Ophthalmol 1976; 81:773785.
156. Jakobiec FA, Ozanics V: General
topographic anatomy of the eye. In:
Jakobiec FA, ed. Ocular anatomy,
embryology and teratology. Philadelphia,
157. Srinivasan BD, Jakobiec FA, Iwamoto T:
Conjunctiva. In: Jakobiec FA, ed. Ocular
anatomy, embryology, and teratology.
Philadelphia, PA: Harper & Row; 1982.
158. Nagasaki T, Zhao J: Centripetal movement
of corneal epithelial cells in the normal
adult mouse. Invest Ophthalmol Vis Sci
2003; 44:558566.
159. Zieske JD, Bukusoglu G, Yankauckas MA:
Characterization of a potential marker of
corneal epithelial stem cells. Invest
160. Hayashi K, Kenyon KR: Increased
cytochrome oxidase activity in alkaliburned corneas. Curr Eye Res 1988; 7:131.
161. Lutjen-Drecoll E, Steuhl P, Arnold WH:
Morphologische besonderheiten der
conjunctiva bulbi. In: Marquardt R, ed.
Chronische Conjunctivitis-trockenes Auge.
New York: Springer Wien; 1982.
162. Steuhl K-P, Thiel H-J: Histochemical and
morphological study of the regenerating
corneal epithelium after limbus-to-limbus
denudation. Graefes Arch Clin Exp
Ophthalmol 1987; 225:5358.
163. Budak MT, Alpdogan OS, Zhou M, et al:
Ocular surface epithelia contain ABCG2dependent side population cells exhibiting
features associated with stem cells. J Cell
Sci 2005; 118:17151724.
164. Thoft RA, Friend J: The X,Y,Z hypothesis of
corneal epithelial maintenance. Invest
165. Kenyon KR, Tseng SC: Limbal autograft
transplantation for ocular surface disorders.
166. SundarRaj N, Chao J, Gregory JD, Damle
SP: Ocular distribution of keratan sulfates
during pre- and post-natal development in
rabbits. J Histochem Cytochem 1986;
34:971976.
167. Gipson IK: The epithelial basement
membrane zone of the limbus. Eye 1989;
3:132140.
168. Krenzer KL, Freddo TF: Cytokeratin
expression in normal human bulbar
conjunctiva obtained by impression
CHAPTER 39
439
SECTION 6
cytology. Invest Ophthalmol Vis Sci 1997;

38:142152.
169. Srinivasan BD, Worgul BV, Iwamoto T,
Merriam GR: The conjunctival epithelium. II.
Histochemical and ultrastructural studies
on human and rat conjunctiva. Ophthalmic
Res 1977; 9:6579.
170. Dilly PN: On the nature and the role of the
subsurface vesicles in the outer epithelial
cells of the conjunctiva. Br J Ophthalmol
1985; 69:477481.
171. Greiner JV, Weidman TA, Korb DR,
Allansmith MR: Histochemical analysis of
secretory vesicles in nongoblet conjunctival
epithelial cells. Acta Ophthalmol (Copenh)
1985; 63:8992.
440
172. Gipson IK, Inatomi T: Mucin genes

expressed by the ocular surface epithelium.
173. Inatomi T, Spurr-Michaud S, Tisdale AS,
et al: Expression of secretory mucin
genes by human conjunctival epithelia.
37:16841692.
174. Gipson IK, Tisdale AS: Visualization of
conjunctival goblet cell actin cytoskeleton
and mucin content in tissue whole mounts.
Exp Eye Res 1997; 65:407415.
175. Kessing S: Mucous gland system of the
conjunctiva: a quantitative normal
anatomical study. Acta Ophthalmol Suppl
1968; 95:1.
176. Wei Z-B, Cotsarelis G, Sun T-T, Lavker RM:

Label-retaining cells are preferentially
located in fornical epithelium: implications
for conjunctival epithelial homeostasis.
36:236246.
177. Nagasaki T, Zhao J: Uniform distribution of
epithelial stem cells in the bulbar
conjunctiva. Invest Ophthalmol Vis Sci
2005; 46:126132.
178. Wei Z-G, Lin T, Sun T-T, Lavker RM: Clonal
analysis of the in vivo differentiation
potential of keratinocytes. Invest
CHAPTER
40
Corneal Form and Function: Clinical Perspective

Stephen D. Klyce, Claes H. Dohlman, and Carlos E. Martinez
INTRODUCTION
The cornea forms the anterior meniscus-shaped transparent
portion of the ocular globe; it serves as the principal refractive
element in the eye, while maintaining a highly impermeable
barrier between the eye and the environment. The cornea is
avascular, meeting its oxygen requirements largely from the
atmosphere by diffusion across the tear film and epithelium;
conversely, it derives most of its additional nutritional
requirements from the aqueous humor arising from across the
corneal endothelium. The epithelium of the cornea provides
the major barrier to tear-borne pathogens, while the corneal
endothelium is principally responsible for maintaining the
hydration and clarity of the corneal stroma.
This chapter reviews aspects of corneal form and function
most relevant to clinical practice to understand the impact of
corneal diseases and surgical interventions on this unique
tissue.
STROMAL ARCHITECTURE AN
INVITATION TO EDEMA
Most of the connective tissue in the body, including the
sclera, is composed of a dense mat of interweaving collagen
bers, which limits swelling to some extent. By contrast, the
corneal stroma can swell to several times its normal thickness;
stromal edema can cloud the cornea. This ability to swell is
due in part to the anatomic framework of the corneal stroma
which consists of very long, thin, and striated type I collagen
brils. These are laid down in a remarkably parallel and
equidistant fashion.1 The brils appear to span, without
branching, from limbus to limbus and are organized into
bundles, called lamellae, arrayed so that adjacent layers lie at
acute angles with one another. Between the brils, however, and
in close association with them, proteoglycans and other
proteins constitute the ground substance. The carbohydrate
components of the proteoglycans consist of glycosaminoglycans (keratan sulfate, chondroitin sulfate, and dermatan
sulfate), which are mainly responsible for the unique waterholding capacity of the stroma (Fig. 40.1). Glycosaminoglycans
are polyelectrolytes that, when placed in an aqueous environment, tend to occupy a large molecular volume, resulting in
stromal swelling if they are not restricted. The swelling tendency is caused by the xed anionic groups of polyelectrolytes,
which promote swelling by long-range electrostatic repulsion,
the Donnan effect, and, to a lesser degree, colloid osmotic
pressure.2,3
The tendency of the stroma to swell has been called the
swelling pressure (SP), so named because when the stroma is
denuded of its limiting cell layers and placed in normal saline,

a mechanical force can be used to prevent the tissue from
swelling. Swelling pressure measured in vitro and in vivo is
~55 mmHg at normal thickness.4,5 If the stroma is allowed to
swell abnormally (edema), the swelling pressure drops rapidly;
conversely, if the stroma is allowed to dry (because of tear
lm breakup), the swelling pressure rises exponentially. The
swelling force can also be measured as the force required
preventing saline from being sucked out of a cannula placed in
the stroma. In this case, the expansive tendency develops a
negative pressure (imbibition pressure, IP), which has the
same numeric value as the swelling pressure in vitro. In vivo,
IP is modied by the intraocular pressure (IOP) according to
the following equation6:
IP = IOP SP
[1]
As discussed later, this relation has important consequences:

as long as IP is negative, fluid is not apt to collect within the
highly resistive epithelium to produce edema. When the IOP
rises (as with glaucoma) so that it is equal to or exceeds the
stromal swelling pressure, the IP rises above zero, and epithelial
edema can ensue.7
The swelling pressure is the force that moves fluid from one
place in the stroma to another whenever there is a gradient in
hydration. In fact, in the cornea in vivo, the anterior stroma
normally is at a lower hydration than the posterior stroma as
a result of differences in the permeability characteristics of
the epithelium (low permeability) and endothelium (high
permeability).8 This gradient in stromal hydration is sustained
because of the high viscosity of the ground substance, which
has been measured as the hydraulic conductivity by Fatt and
Goldstick.9 This high viscosity, which retards bulk fluid
FIGURE 40.1. Keratan sulfate is the main glycosaminoglycan

(mucopolysaccharide) of the stroma and is part of proteoglycan
molecules. Because of their polyanionic properties, the
glycosaminoglycans occupy large molecular volumes, which explains
the high water content of the tissue.
441

movement in the stroma (as well as in the epithelial cells), is
the underlying factor predisposing the cornea to dellen
formation in response to tear lm defects.
CONTROL OF CORNEAL HYDRATION

Because the stroma normally is kept in a relatively deturgesced
and, therefore, optically transparent state, a dehydrating mechanism must exist to counterbalance the swelling pressure and
to maintain normal thickness. It was originally proposed that
stromal thickness was maintained passively by the impermeability of the corneal membranes to salt and an excess of
solute concentration in the aqueous humor and tears over
that in the stroma.10 It was soon shown, however, that no
biologic membrane is truly impermeable to tissue electrolytes.
Both of the corneal cell layers are now known to be permeable
to solutes and, although they act as passive (non-energyconsuming) imperfect semipermeable barriers, there must be
a mechanism to maintain the ion content of the stroma.
This mechanism is assisted by the barrier properties of the
corneal cell layers their resistance to electrolytes, which is
highest in the epithelium and ~100 times lower in the endothelium.11 In the epithelium, the supercial squamous cells
provide the major barrier to the ion flow because this is the
locus of the tight junctional complex that completely surrounds the margins of every supercial cell.12 The restriction
of ions means that rapid water movement is similarly restrained
for osmotic reasons. Subsequently, Davson13 and Harris and
Nordquist14 demonstrated that corneas swollen overnight in
the cold at temperatures that slowed their metabolic processes
were able to deturgesce the following day when rewarmed
to body temperature. From this observation, it became clear
that either the corneal epithelium or the corneal endothelium,
or both must expend metabolic energy in the process of
maintaining normal corneal hydration and transparency.
Trenberth and Mishima15 developed rabbit corneal isolation
chamber techniques to show that such a mechanism resided in
the endothelium and was shut down by the transport inhibitor
ouabain.
SECTION 6
ENDOTHELIAL FLUID PUMP A

MISNOMER
442
The early experiments showed clearly that the endothelium

uses metabolic energy to transfer fluid from the stroma to the
aqueous humor, and Maurice16 coined the term endothelial
fluid pump. Maurice, however, envisioned this pump, not
as a literal entity that packaged fluid and removed it from the
stroma, but as a consequence of an active ion transport process
associated with the endothelium. This process would obey the
laws of thermodynamics to move an as yet to be identied ion
and co-ion (to maintain electroneutrality) out of the stroma
to lower the osmotic pressure of the connective tissue. The
consequence of this process in the normal tissue would lead
to a steady-state situation in which a constant solute gradient
would be maintained across the corneal cell layers that would
balance the swelling pressure and prevent the imbibition of
water. The endothelial fluid pump terminology was adopted,
however, and was soon accompanied by the term fluid leak.
This provides an easily understood concept fluid leaks in and
is pumped out but confuses the underpinnings of corneal
hydration control because water actually always is close to
equilibrium in living systems, obeying the osmotic force
gradients across membranes developed and maintained by
active ion transport processes.
It was clear that active ion transport is involved in the control
of corneal hydration, particularly because the active transport of
water had never been demonstrated. The model for the control
of corneal hydration that held the most promise postulated
that one or more ion pumps located in the endothelium
would transport solute out of the stroma to balance the solute
that leaked in across the imperfect semipermeable corneal
endothelium.
EPITHELIAL TRANSPORT PROPERTIES

Donn and associates17,18 were the rst to search for ion transport processes across the cornea. These investigators demonstrated the active transport of Na+ by the rabbit cornea in vitro.
Curiously, the transport not only was associated with the
epithelium but also was oriented in the wrong direction: from
tears to stroma. This anomaly did not help with the understanding of corneal hydration control because, as noted
previously, one could eliminate the epithelium entirely and
demonstrate the control of stromal thickness by the
endothelium.
Later studies with in vitro rabbit corneas showed that the
epithelium also actively transports Cl in the secretory direction
(from stroma to tears).19 This transport is regulated by the badrenergic receptoradenylate cyclase complex, with possible
control by sympathetic nerve bers.20 Under normal resting
conditions simulating the in vivo corneal situation, the outward active Cl transport process competes with the inward
active Na+ transport process, leading to a net outward transport
of solute and fluid that could account for a corneal thinning
rate of 1.3 mm/h.21
It was concluded that, although the epithelium is capable of
thinning the cornea at a rate perhaps 30 times lower than that
demonstrated for the endothelium, at least the epithelium is
operating in synergy and not compounding the solute removal
task of the endothelium.
ENDOTHELIAL TRANSPORT PROPERTIES

The search for a candidate for the endothelial transport function was arduous. Because of the leaky nature of the cell layer,
passive unidirectional fluxes of small ions are large and apt to
mask a net ion flux, which is the hallmark of active transport
when all forces across a cell layer are eliminated. Furthermore,
the corneal endothelium generates an electrical potential of
only 500 mV across a membrane resistance of ~50 Wcm2. This
electromotive force was difcult to measure and neutralize
accurately, further obscuring the nature of the active ion
transport process. With renements in technology, however,
it was possible to show that the corneal endothelium appears
to accomplish hydration control with transport of the bicarbonate ion from stroma to aqueous.22,23 This is accompanied
by the transport of Na+ in the same direction.24
For an active ion transport mechanism to drive water
movement osmotically, the activity must lead to a reduction
or increase in osmotic pressure in a sequestered space. With an
intact epithelium, the corneal stroma could provide such a
space. A reduction of solute concentration in the stroma by as
little as 1% (3 mOsm) below that of the aqueous humor is sufcient to counterbalance stromal swelling pressure. The fluidpumping ability of the corneal endothelium, however, appears
to remain intact despite removal of most of the stroma.16,25,26
Therefore, it would appear that the sequestered space needed for
the endothelial transport function must be the cell interior or
perhaps the intracellular space. Nevertheless, the corneal endothelium secretes bicarbonate and Na+ into the aqueous humor,
and the direction and magnitude of this energy-consuming
process are adequate for the regulation of corneal stromal
hydration.

that might occur with severe corneal pathology leading to
scarring and opacication are beyond this framework.
Because the normal corneal stroma swells only in thickness,
its diameter and anterior curvature remaining relatively
constant with edema, there is a linear relation between
thickness, q, and water content or hydration, H:
H = 8q 0.7
[2]
As noted previously, stromal swelling pressure, SP, is the

driving force for water movements in the stroma and has been
shown to be related to hydration, H, as follows:
SP = g eH
where g is a constant. This exponential relation is the reason

that swelling pressure drops rapidly with hydration. The next
equation relates the flow conductivity for water within the
stroma to local stromal hydration. This is known as the
hydraulic conductivity, Lp, and changes even more rapidly than
swelling pressure with changes in hydration:
Lp = r H4
EVAPORATION AND INTRAOCULAR

PRESSURE
When the eye is open, evaporation occurs at a rate of
~2.5 mL/cm2 per hour,27 and this can account for the cornea
thinning some 5% during awake periods compared with periods
of sleep. In a normal eye with ample tear secretion, this
evaporation is of little importance for the overall corneal fluid
balance. In a dry eye or an eye with insufcient blink function,
exposure and evaporation can have devastating consequences.
Also, in the normal eye, the IOP does not have a signicant
effect on corneal hydration so long as it is in the normal range.
If the IOP is more than 5060 mmHg, however, or if the stroma
is swollen from endothelial dysfunction or decompensation, the
IOP becomes a major factor in the development of epithelial
edema and reduction of vision (discussed later).
MODEL FOR THE CONTROL OF CORNEAL

HYDRATION
With the endothelial ion transport process identied, the
opportunity is presented to develop a holistic model of corneal
hydration dynamics that considers all the major flows and
forces underlying homeostasis. A full characterization of corneal hydration dynamics must consider at least cell layer barrier
(passive permeability) properties as well as any active ion
transport processes, the swelling pressure and flow characteristics of the stroma, and external forces, such as evaporation
and IOP (Fig. 40.2). Klyce and Russell8 considered these and
developed and tested such a model applying the formalism of
Kedem and Katchalsky,28 which is based on the thermodynamics of irreversible processes. In the absence of more
detailed information regarding specic cell pathways for ion and
water flow, the corneal epithelium and endothelium were
regarded as thin, semipermeable membranes that embrace the
active ion transport processes identied for these layers. The
corneal stroma, which often is considered a large, well-stirred
compartment with respect to corneal permeability studies, was
modeled with the KedemKatchalsky equations as well, with
the addition of the well-characterized relations that associate
stromal thickness to its hydrodynamic properties that change
reversibly with edema. These concepts show that water
movements in the stroma are greatly retarded compared to free
solution, and the viscous nature of the stromal ground
substance permits large gradients in hydration to develop under
abnormal circumstances. However, irreversible stromal changes
[4]
where r is a constant. Fluid flow from point to point, Jv, in the

stroma is the product of the driving force, SP, and the flow
conductivity, Lp. Hence, one can write the following:
Jv = Lp SP
[5]
The consequence of this relation favors corneal stromal

homeostasis with regard to hydration. When the corneal stroma
swells above normal hydration, the swelling pressure (its
gradient drives fluid flow) falls rapidly (see Eqn [3]). When the
corneal stroma thins below normal hydration, its hydraulic
conductivity (permissivity to water flow) falls rapidly (see Eqn
[4]). We can combine the overall effect in Eqn [5] to indicate
that corneal thickness changes are most dynamic when the
stroma is near normal hydration and less dynamic as it either
swells or thins. Clinically, this is the reason that steep
hydration gradients can exist from place to place in the stroma,
as with dellen or focal edema.
The model for corneal hydration control proposed by Klyce
and Russell8 was able to achieve a constant corneal thickness
and mimic the corneal response to several well-documented
observations. For example, the model matched the rate at which
the cornea thinned after cold swelling and the rate of stromal
swelling that occurred when the endothelial transport system
was inhibited by ouabain. The model was also used to show
that the cornea swells in a predictable fashion during hypoxia,
such as occurs under a tight-tting contact lens due to the
accumulation of lactate within the stroma.29
CORNEAL EDEMA
While the above formalism is developed to understand the
normal regulation of corneal hydration, this knowledge can be
extended to learn the corneal hydration response to pathology.
Chronic corneal edema develops as a consequence of endothelial dysfunction, regardless of whether the original clinical
condition was dystrophy, inflammation, or trauma. The
increased permeability of this cellular layer, its decreased ion
transport function, or both can lead to the subsequent corneal
changes. In mild cases, increased stromal thickness occurs,
with initially little consequence to vision. In advanced cases,
epithelial edema ensues, which rapidly decreases visual acuity.
Late in the course of the disease, painful bullous changes can
develop (bullous keratopathy). If the natural clinical course is
not interrupted by keratoplasty, a thick subepithelial pannus
eventually develops, leading to the disappearance of the bullae
and of the discomfort. Vision at this stage usually is reduced to
CHAPTER 40
FIGURE 40.2. Hydration of the cornea is kept in balance by opposing

forces: the stromal swelling pressure and the endothelial pump. The
epithelium and the endothelium restrict rapid fluid movements (see
text). IOP, intraocular pressure; SP, swelling pressure.
[3]
443
the hand-movement level because of epithelial and stromal

scar formation (Figs 40.3 and 40.4).
Acute corneal edema, which can result from contact lens
wear or angle-closure glaucoma, follows a different path of
physiologic development and is usually reversible.
DEVELOPMENT OF EDEMA
Endothelial Changes
SECTION 6
FIGURE 40.3. Natural history of progressive

corneal edema. (a) Cornea guttata of Fuchs
dystrophy often begins in young adulthood and
progresses slowly over decades. Some edema
(thickening) gradually ensues for a long time
without affecting vision. (b) Epithelial edema
begins in midlate life, first as fine microcysts
that distort the surface and cause reduction of
vision. (c) Frank epithelial edema with visible
blebs, opacity, and gross surface irregularities.
(d) End-stage chronic edema after many years
of often painful bullous epithelium. A
connective tissue pannus has formed between
the epithelium and Bowmans layer. At this
stage, the cornea is opaque, but the epithelium
has scarred down, and the pain is gone.
The endothelium under stress changes in a few nonspecic but

characteristic ways. Thus, in acute inflammation or in trauma,
rapid cell degeneration and cell death can occur in a focal
manner that is then repaired by sliding and rearrangement of
neighboring cells. The resulting endothelium is characterized by
decreased cell number and enlarged and irregularly shaped cells
(polymegathism and pleomorphism).30
In chronic inflammation, endothelial cells can undergo
brous metaplasia31; this can result in a brous membrane between Descemets membrane and the endothelium. In Fuchs
dystrophy, the cells exhibit a change in form and show vacuoles,
phagocytized pigment, and periodic acid-Schiff stain-positive
material that is eventually deposited on Descemets membrane.
These irregular depositions become visible as the characteristic
warts (guttae) over which the endothelial cells eventually
become attenuated. Even in advanced cases of Fuchs dystrophy,
however, the endothelial surface appears intact.32 For a review
of this subject, see Waring and colleagues.33
in the posterior direction (corneal anterior curvature and

diameter remain normal), its thickness increases, especially
centrally, because the peripheral corneal swelling appears to be
limited somewhat by structural restriction imposed by the
limbus. This flattening of the posterior surface can throw
Descemets membrane into multiple folds that become visible
as striae on slit-lamp microscopy. Usually, there is little tissue
reaction to the swelling; it is only in massive, chronic edema
that scarring of the tissue eventually develops, more markedly
in the posterior layers and especially in the folds created by
Descemets membrane.
Epithelial Edema
Epithelial edema resulting from endothelial dysfunction,
elevated IOP, or both is predominantly extracellular.34 Thus,
Stromal Edema
444
When the endothelial cell density falls below a critical level

(200400 cells/mm2), the ability of the endothelium to maintain stromal hydration begins to falter, and stromal edema
develops gradually. The two opposing forces the osmotic pressure developed by the endothelial ion transport and the stromal
swelling pressure remain in balance, but the osmotic gradient
established by the endothelial pump must diminish with
reduced transport function and possibly greater ion leakage
across the endothelium. Because the stroma can swell only
FIGURE 40.4. Massive bullous keratopathy after cataract extraction.

fluid begins to accumulate in the space between the basal
epithelial cells, stretching the bridging desmosomes. Later in
the process, these fluid-lled spaces enlarge to form ne
blisters, visible as microcystic edema in the slit lamp. Finally,
larger bullae develop, characteristic of bullous keratopathy.
Epithelial edema rarely involves the anterior-most squamous
cells of the epithelium (Fig. 40.5).
The underlying pathophysiologic mechanism appears to
involve a forward movement of stromal fluid and aqueous,
generated by the IOP. Thus, if the endothelial functional
reserve falls below a certain level, leading to edema and a
reduction in stromal swelling pressure to below the value of
the IOP, fluid from the aqueous can collect.8 Because the otherwise healthy epithelium has such a high resistance to
electrolytes and to the flow of water, the fluid can be trapped
within the epithelium, resulting in the formation of cysts and
bullae. The anterior-most wing cells generally are unaffected,
suggesting that the resistance to this anterior fluid movement
is situated primarily in this layer.
The IOP can be higher than the stromal swelling pressure,
resulting in epithelial edema, in several different clinical settings. On one end of the spectrum, poor endothelial function,
even with normal IOP, is sufcient to cause epithelial edema
(e.g., Fuchs dystrophy, aphakic or pseudophakic edema). On
the other end of the spectrum, even with normal endothelium,
very high IOP (such as occurs in angle-closure glaucoma) also
can cause epithelial edema (Fig. 40.6). Between these two
extremes, edema can result from various combinations of
endothelial dysfunction and elevated pressure8 (Fig. 40.7).
The concept that the IOP is the driving force behind the fluid
movement in epithelial edema is supported particularly by the
fact that in phthisis with marked hypotony, epithelial edema
does not occur, no matter how damaged the endothelium or
how thick the stroma.
Evaporation can be a balancing factor in borderline epithelial
edema. Commonly in early edema, vision is blurred in the
morning but clears as the day progresses. The lack of evaporation when the lids are closed during the night allows fluid
to accumulate in the epithelium. After opening the eyes, evaporation results in slight hypertonicity of the tear lm, which
in turn extracts water from the epithelium, clearing the vision.
This state can last for months, rarely years, but eventually the
edema worsens.
FIGURE 40.5. Development of epithelial edema. When the intraocular

pressure (IOP) overpowers the endothelial pump, fluid is pushed into
the epithelium, resulting in edema. This can occur with a normal
endothelium and very high pressure (acute glaucoma). In chronic
edema, the IOP is normal, but the endothelium is severely
dysfunctional (see text).
The edema that occasionally is seen as a result of contact lens

wear differs in many respects from the forms of edema
described previously. The symptoms and signs usually are
attributable to hypoxia, hypercapnia (elevated CO2 tension), or
both under the contact lens. Therefore, gas permeability of
the lens and tear fluid exchange are the most important
parameters in maintaining normal fluid balance in the cornea.
FIGURE 40.6. Corneal thickness and IOP as they relate to epithelial

edema. Such edema is expected to the right of the solid line. The
circle indicates normal values.
Data from Ytteborg J, Dohlman CH: Corneal edema and intraocular pressure. II.
Clinical results. Arch Ophthalmol 1965; 74:477; and Klyce SD, Beuerman RW:
Structure and function of the cornea. In: Kaufman HE, Barron BA, McDonald
MB, eds. The cornea. 2nd edn. Boston: Butterworth-Heinemann; 1998.
CHAPTER 40
Edema with Contact Lens Wear
FIGURE 40.7. (a) and (b) Epithelial edema. The

fluid is pushed into the epithelium, resulting in
distention of the intercellular spaces. (a) 500;
(b) Electron micrograph, 2000.
Courtesy of Toichi Kuwabara, MD.
445

well preserved in mild or moderate stromal edema,41 and
backscattering toward the source is minimal.42 In more
advanced and long-standing edema, however, irregular fluid
accumulations occur in the stroma that can reduce
transparency.43 Later, stromal scarring and posterior irregular
astigmatism from folds in Descemets membrane reduce
vision further.
In Sattlers veil, reduction in vision usually is minor and is
caused by a diffraction phenomenon that occurs within the
slightly swollen epithelial cells.44 True epithelial edema, in
contrast to Sattlers veil and stromal edema, can reduce vision
early and profoundly. The fluid accumulations between or
within the epithelial cells markedly increase light scattering.
Even more important are the minute surface irregularities
from the edema that break up the smoothness normally
provided by the corneal tears and consequently lead to blurring
of the retinal image (Fig. 40.8). In general, in a patient with
vision reduced by corneal pathology, there often is a tendency
for the clinician to overestimate the contribution of opacities
within the tissue and to underestimate the role of surface
irregularities. A hard contact lens refraction or corneal
topography analysis should settle the question of the influence
of surface irregularities on visual acuity.
CLINICAL EVALUATION OF EDEMA
SECTION 6
FIGURE 40.8. Irregular Placidos mires in chronic edema, indicating

the substantial role of the surface abnormalities in reduction of vision.
446
Mild stromal edema is common in soft contact lens wear.

Epithelial hypoxia causes lactic acid buildup in the stroma35
and reduced pH,36 which in turn may affect endothelial
performance. The lactic acid accumulation, however, raises
stromal osmotic pressure, drawing in water for osmotic reasons
noted previously, and pH effects on the corneal endothelium
may not occur in acute situations.30 This thickening of the
stroma usually is clinically acceptable because it has little
influence on visual acuity or contrast sensitivity.37 Epithelial
edema, on the other hand, resulting from hard contact lens
overwear (Sattlers veil, epithelial bedewing) can be debilitating
visually, although it is readily reversible on removal of the lens.
Histologically, the location of the fluid collection is primarily
intracellular, in contrast to the primarily extracellular edema in
Fuchs dystrophy or similar conditions, and may be related to
induced abnormalities caused by hypoxia and lactate
accumulation.29 For a review of corneal pathophysiology and
contact lens wear, see Bruce and Brennan.38
Visual Acuity in Edema

Because of its surface smoothness and its transparency, the
cornea normally allows a remarkably sharp image to be
focused on the retina. In general, these optical qualities can
be reduced by opacities within the tissue (stroma or epithelium)
or by surface irregularities in the form of gross astigmatism
(e.g., keratoconus) or minute central irregularities (e.g., bullous
keratopathy).
Normal stromal transparency has been difcult to explain
in view of the internal fluctuations in refractive index between
the stromal components. It has been proposed, however, that
as long as the collagen brils are parallel and equidistant and
the distance between them is less than half the wavelength of
light (~2000 ), light scattering should be at a minimum and
transparency preserved.39,40 In the normal corneal stroma, the
collagen brils are spaced some 600 from center to center and
are closer together than half the wavelength of light, which
explains the optical qualities of the tissue. Transparency is still
To evaluate a case of corneal edema fully its cause, extent,

and prognosis it is advisable to conduct the investigation in
a sequential and systematic manner.
History
As with any other ophthalmic condition, a detailed patient
history is mandatory. Because most cases of chronic edema
result from endothelial malfunction, it is important to nd
out whether there has been a family history of dystrophy or
trauma or whether inflammation with a red eye has been
experienced by the patient in the past. Other issues that should
be covered include the degree and duration of visual loss,
whether vision is worse in the morning but clears during the
day (possibly indicating Fuchs dystrophy), whether episodes of
blurred vision have been separated by long periods of normal
vision (possible herpes virus infection), whether the symptoms
are unilateral or bilateral, and the results of previous
examinations by other physicians. A detailed history is usually
strongly suggestive of the cause of the condition.
Slit-Lamp Microscopy
The introduction of the slit lamp in 1911 contributed enormously to our ability to diagnose corneal disease, especially
edema. The epithelium should be intact and free of fluorescein
or rose bengal stain. In early epithelial edema, this layer appears
more gray or full than the normal, optically empty cell layer.
Later, a microcystic appearance may ensue patchwise, central,
or uniform that initially may be visible only in retroillumination. In full-blown bullous keratopathy, the variably
sized cysts are obvious to inspection, as is the degree of opacity
within the thickened epithelium. In end-stage severe edema, a
pannus layer of connective tissue may be visible between the
epithelium and Bowmans membrane.
The stroma usually is thickened somewhat, as indicated by
posterior striae from folds in Descemets membrane. As mentioned previously, the edematous stroma can be optically clear
for a long time, but in severe chronic edema, scar formation
gradually develops, creating haze. Guttate appearance of the
posterior surface, often with pigment (the hallmark of Fuchs
dystrophy), is apparent in specular reflection using a wide angle
between the light source and the observer. Stromal vas-
FIGURE 40.10. Specular microscopy of the endothelium in a cornea

with guttae but no overt edema. The black dots indicate areas where
the endothelium has been lifted posteriorly and out of focus by the
excrescences on Descemets membrane.
TABLE 40.1. Endothelial Cell Count (per mm2)

b
40001500
15001000
FIGURE 40.9. Specular microscopy of the corneal endothelium.

(a) The endothelium in an adult, 3200 cell/mm2. (b) The endothelium
in a graft that has suffered a rejection episode but is still clear,
500 cell/mm2, large, and irregular.
1000500
<500
Normal
Low
Borderline
Specular Microscopy
The technique of observing and recording the morphology of
the corneal endothelium with high magnication in vivo was
introduced by Maurice in 1968.45 The subsequent developments of contact and noncontact specular microscopes,
applicable in the clinic, as well as the various types of analyses
of cell density and cell shape, are covered in Chapter 60. This
technique allows the clinician to follow the status of the
endothelium in dystrophy, in degeneration, before and after
surgery, after trauma, and in donor corneas, among other
situations.
It generally is agreed that human endothelium has little or no
ability to divide after birth. Therefore, in aging or in an accelerated fashion after injury or disease, the endothelium loses
cells without replacement. The normal endothelial cell count
is 30003500 cells/mm2 in young adults, decreasing to about
two-thirds that value in old age.46 After injury, the endothelium heals by sliding, rearrangement, and irregular enlargement of adjacent cells, usually from a large surrounding
area.47 The result is decreased cell count, often only regionally
rather than uniformly across the whole back surface of the
cornea, rendering specular microscopy variable. The degree of
pleomorphism and polymegathism of the endothelial cells
usually is as indicative of the endothelial health and reserve as
is the cell count (Fig. 40.9).
In clinical practice, it has proved difcult to predict physiologic function on the basis of cell density or morphology.48
Cell counts down to only a few hundred per square millimeter
have been observed in edema-free corneas and grafts, whereas
many cases of frank edema have had much higher counts. This
discrepancy at times can be explained by the variability of
morphology across the cornea; at other times, there may be
factors not detectable with a specular microscope.
Usually edema
Fuchs dystrophy can be diagnosed early with specular

microscopy, even before guttae become visible in the slit
lamp.49,50 The excrescences from Descemets membrane cause
the overlying endothelium to bulge posteriorly out of focus,
resulting in multiple round black areas of different diameters
(Fig. 40.10). It can be argued, however, that with our present
state of knowledge, very early diagnosis of Fuchs dystrophy is
not essential because the condition takes decades to develop
and because no preventive treatment is available. The visible
endothelial cells in Fuchs dystrophy are usually enlarged
and have irregular cell borders.
Specular microscopy has become a useful tool in the evaluation of surgical procedures with respect to their trauma
to the corneal endothelium. Thus, by following the cell counts,
the value of various techniques of cataract surgery and different
models of intraocular lenses has been determined with much
greater precision than was possible when relying only on statistics on the incidence of edema after years of postoperative
follow-up. Thus, switching from an intracapsular to an extracapsular cataract extraction technique has not been
accompanied by increased cell loss. In one study done in 1981,
99 consecutive cases of intracapsular cataract extraction
resulted in a 17% cell loss, whereas the same number of extracapsular cases had a 17% loss, both series with lens implantation.51 Phacoemulsication in the anterior chamber proved
more traumatic, with a cell loss of up to 30%.52 The protective
effect of sodium hyaluronate during anterior segment surgery
a milestone in ophthalmic surgery development has also been
demonstrated with exactness by specular microscopy.53
Although the correlation between endothelial cell count and
frank corneal edema is poor, there remains a linear relation
between the two54; therefore, specular microscopy can have considerable predictive value before and after surgery (Table 40.1).
The technique is gaining increasing popularity in clinical
CHAPTER 40
cularization and keratic precipitates indicate previous or present

inflammation and, if unilateral, often are suggestive of herpes
virus infection. The corneas should be photographed, with
and without slit beam, for documentation.
447
TABLE 40.2. Indications for Simultaneous Keratoplasty in

Cataract Surgery
instrument with the patients eye. As with most technology,

costs to purchase have been reduced and major improvements
have occurred in the quality of both live and stored images.
Frank epithelial edema, or

Blurry vision in the morning, or
Corneal thickness above 0.70 mm, or
Endothelial cell density less than 500 cells/mm2
practice, especially when the endothelium can be expected preoperatively to have borderline function. This is particularly
pertinent when cataract extraction is contemplated in Fuchs
dystrophy or after trauma. With a very low cell count and
abnormal corneal thickness, a simultaneous keratoplasty might
be indicated (Table 40.2).
Specular microscopy also is a valuable tool in keratoplasty.
Postoperatively, several investigators noted 1520% cell loss
within the rst 3 months, and attrition continued for several
years.33 Mean cell counts in clear grafts were 10002000
cells/mm2.55,56 From a practical point of view, specular
microscopy probably has become most useful in the evaluation
of donor corneas, especially for screening of donor material for
Fuchs dystrophy. Mild forms of this dystrophy are common,
and such donor grafts would be expected to have shorter
survival times, especially if grafted into patients with edema.
Slit-lamp examination alone of donor corneas with Fuchs
dystrophy is of limited use because the tissue is usually swollen
when harvested, which makes guttae hard to detect. Most eye
banks in the United States now routinely perform specular
microscopy on donor corneas.
SECTION 6
Corneal Confocal Microscopy
448
The corneal confocal microscope has become a useful diagnostic tool for evaluating the cornea with capabilities that
surpass those of the specular microscope. While the magnication of the confocal microscope is similar to that of the
specular microscope, the resolution of the confocal is higher.
In the confocal mode, the narrow illumination and viewing
paths are at an angle to one another. This reduces the impact
of contrast loss from tissue light scatter along the illumination
pathway. The early development work was done largely through
the pioneering efforts of Lemp, Masters, Cavanagh, and
Jester,57,58 who demonstrated the potential for the use of confocal microscopy in corneal research. Today the confocal microscope is available for clinical application. Designed specically
for imaging the cornea, scanning spot and scanning slit models
have been developed. The scanning spot models applanate the
cornea and provide imaging throughout the cornea. The
scanning slit model has a noncontacting objective with a 2 mm
working distance that is optically coupled to the cornea by
means of a globule of gel. This reduces the intense specular
reflection from the epithelial surface. The confocal microscopes
provide high-resolution images at various corneal depths with
optical sections of less than 10 um.
Corneal confocal microscopes can provide other useful
clinical information.59 Applications have included evaluation of
corneal wound healing responses after refractive surgery
(e.g., amount and localization of haze after PRK, thickness of
stromal bed and interface artifacts after LASIK), endothelial cell
analysis (e.g., cell density, polymegathism, and pleomorphism),
and identication of corneal pathology (e.g., differentiation
between herpetic keratitis and Acanthamoeba infections. In the
past this technology has had limited success in the clinical
setting, because of the high purchase cost, lack of high-contrast
images, and difculty in manually aligning the axis of the
Measurement of Corneal Thickness

As mentioned previously, the thickness of the stroma (or the
whole cornea) is directly related to the dehydrating function
of the endothelium.60 Therefore, the functional status of the
endothelium and its reserve capacity can be measured by
pachometry, the technique of recording corneal thickness
in vivo. The normal thickness of the central cornea is
0.510.52 mm according to most investigators.61 As mentioned
previously, the exact relation between stromal thickness and
stromal swelling pressure (assumed to be equal to opposing
endothelial pump pressure) has been determined.4 Pachometry
used to be primarily a laboratory tool in studies of corneal
physiology. Because of the need to determine the status of the
endothelium in a number of clinical situations, however, several
different types of pachometers have been developed for clinical
use. The HaagStreit optical pachometer is attachable to its
slit lamp and employs an image-splitting principle. This
technique can be exact (2% error), but reading the endpoint is
difcult, and substantial practice is required to achieve reliable
data.62
Both the specular microscope and the corneal confocal microscope also can be used to measure corneal thickness. When
a contact microscope is focused on the endothelium, the distance between the applanating lens and the endothelium is
displayed automatically in some models. This technique is
exact and is easily learned by a photographer. Nonapplanating
lens confocal microscopy measurement of corneal thickness is
also possible with ancillary apparatus that measures the
distance between the lens and the globe.
Ultrasound pachometers are most commonly used not only
for measuring edema, but also for measuring corneal thickness
prior to performing refractive surgery.63 They are easy to handle,
but the ultrasound beam must be directed perpendicularly to
the cornea. The machines are designed to record only those
reflections received from within a few degrees of the
perpendicular.
Automatic slit scanning corneal topographers are also in use
for measuring corneal thickness. These not only provide central
corneal thickness, but can present a thickness prole that
extends out to the limbus. These data are useful for detecting
the distribution of thinning in keratoconus.
Clinically, pachometry has been used to follow endothelial
function after cataract surgery59 and after penetrating
keratoplasty.64,65 More important is the use of pachometry in
evaluating the functional reserve of the endothelium in a clear
cornea in which some stromal edema is suspected but no
epithelial edema has yet appeared. This has particular
importance for the prognosis of Fuchs dystrophy when cataract
surgeries is contemplated but when there is uncertainty about
whether the endothelial layer could withstand further
manipulation (see Table 40.2). A central reading close to the
normal 0.50.6 mm measurement is reassuring, but a value of
~0.7 mm or above measures borderline decompensation with
risk of frank epithelial edema (discussed later).
Endothelial Permeability
Because the leaky corneal endothelium maintains corneal
deturgescence by virtue of its transport and permeability
properties, study of its permeability may yield clinically useful
information. Fluorescein has been chosen for the test
substance, and the passage of dye across the endothelium
can be measured with a sensitive fluorometer.66,67 Fluorescein
can be driven into the cornea by iontophoresis and its
CLINICAL CONDITIONS LEADING TO EDEMA

The clinical disease entities that can cause corneal edema are
covered fully elsewhere in this text and are not reviewed here.
Therefore, the diagnoses are merely listed, and the reader is
advised to turn to the appropriate section for further
information.
The most common conditions are the following:
1. Inflammation (particularly after infection)
2. Graft rejection
3. Endothelial dystrophy
a. Fuchs dystrophy
b. Congenital hereditary endothelial dystrophy
c. Chandlers syndrome and similar conditions
4. Dysgenesis
5. Keratoconus (hydrops)
6. Trauma
a. Intraocular surgery
b. Other trauma (mechanical, chemical)
7. Acute glaucoma
8. Contact lens overwear
MEDICAL TREATMENT OF EDEMA

Some forms of corneal edema respond to nonsurgical measures,
but most cases do not. The cases in which medical treatment is
effective are usually caused by inflammation particularly
postinfection and steroids can be useful in this setting. In
chronic noninflammatory edema, some amelioration of
symptoms can be achieved by a soft contact lens or hypertonic
agents or, rarely, by reducing the IOP. In most patients with
advanced edema, however, only keratoplasty is curative.
Suppression of Inflammation
If given early enough in the disease process, corticosteroids
can be highly effective in reversing corneal edema resulting
from inflammation. Infections, particularly herpetic but also
bacterial or fungal, are often severe enough to affect the
function of the endothelium immediately. The microbes may
be eliminated promptly with appropriate antibiotics, but the
subsequent cascade of postinfectious inflammatory events can
result in edema of such severity that it becomes irreversible
unless corticosteroids are instituted. Energetic efforts to reverse
inflammatory edema are important because severe, irreversible
edema is not easily treated surgically. The long-term prognosis
after keratoplasty is poor for an eye that has been severely
inflamed.
Ophthalmologists are often reluctant to give steroids to
patients with acute infections as long as microorganisms may
still be alive, considering the steroids lowering of the hosts
resistance. These fears are often exaggerated, and it can be
much worse to allow inflammatory destruction to advance
FIGURE 40.11. Common gestalt of topical corticosteroid treatment in

corneal inflammation, especially in stromal herpes. Frequent
instillation of a steroid preparation rarely is necessary for many days,
particularly in cases of lymphocytic response. The drug then can be
tapered quickly, but a subsequent long trial of very low dose or
infrequent treatments may be necessary.
unchecked for a long time than to give steroids early. In patients

with bacterial infections, it is usually safe to start topical
steroids simultaneously with appropriate antibiotics or to
wait 23 days at most. In patients with stromal herpes and
developing edema, steroids should be given promptly unless an
active dendritic epithelial process is also present. Fungal
infections such as fusarium keratitis should not be treated with
steroids.
In patients with corneal edema caused by inflammation,
steroids should be administered topically and only rarely
systemically. Steroids given by mouth, even in high doses, result
in only a low concentration in the endothelium much lower
than what can be achieved by topical treatment. Only in severe
keratouveitis may steroids be added systemically and with
caution, owing to the side effects. The most commonly used
steroids in drop form are 0.1% dexamethasone (as alcohol or
phosphate) and 1% prednisolone (as acetate). One-eighth
percent prednisolone and 0.1% fluorometholone have a weaker
effect. The acetate form penetrates the epithelium more readily
than do the other derivatives.72 The general treatment principle
should be to hit hard (administer a strong concentration of
steroids four to eight times per day) for a few days and then to
taper rapidly. Small doses of a weak steroid solution may then
be necessary for months (Fig. 40.11). It is important not to give
more than is clinically necessary. The latter goal certainly is not
always obvious in the clinical setting, and in general, steroid
treatment of corneal diseases requires considerable experience.
In corneal edema caused by dystrophy, corticosteroids are
without value. Thus, in Fuchs dystrophy, steroid treatment has
no effect on corneal thickness.73 In obscure cases, especially in
unilateral edema with no signs of guttate lesions, an
inflammatory component (sometimes herpes) is possible, and a
shorter trial of topical steroids is reasonable. Steroid drops given
three to four times per day for 2 or 3 weeks should settle
whether the edema is reversible. Small doses of antiviral or
antibiotic agents may also be indicated prophylactically.
Corneal edema of a graft caused by immunologic rejection
constitutes a special case of inflammation. This topic is covered
in more detail in Chapters 65 and 77, but a few points deserve
mention here. When sensitized lymphocytes attack the graft
endothelium, destruction often is rapid, and recognition and
treatment often late. Therefore, the patient should be taught to
respond to symptoms of decreased vision, redness, or
discomfort by contacting the surgeon within 24 h. Topical
steroids should be instituted as rapidly as possible (e.g., a
strong-concentration solution every hour while awake for a few
days). Systemic steroids rarely are indicated. After a few days,
the medication can be tapered fairly quickly, even if frank
CHAPTER 40
diffusion into the aqueous followed. It is safer and simpler, however, to ingest the dye systemically and measure the forward
diffusion across the endothelium.68 Results of this test have
reconrmed the key role of the endothelial barrier function in
corneal deturgescence. Also, in this way, fluorophotometry
promises to be a useful diagnostic technique in the clinic.69
For instance, after keratoplasty, the endothelial transfer
coefcient has been found to be increased initially but to return
gradually to near normal, parallel to the reduction of thickness.70 In early Fuchs dystrophy, fluorophotometry has
indicated that the swelling is due primarily to a decrease in
the endothelial pump function rather than to a breakdown of
the barrier function.71 Mishima has provided a review of the
theoretical background and clinical applications.69
449

edema remains. The lymphocytes are quickly eradicated, but it
takes some time for surviving endothelial cells to recover and
ll in defects. If the patient still has a clear lens, long-term use
of steroids will eventually result in a cataract. It was shown that
after keratoplasty to treat keratoconus, it takes only ~800 drops
of 0.1% dexamethasone to initiate posterior subcapsular
cataract in half of cases.74 Obviously, the IOP must also be
followed during such steroid therapy. If discovered early, most
graft rejections can be reversed.
Lowering of Intraocular Pressure

Because epithelial edema and the concomitant reduction of
vision are the result of a situation in which the IOP overpowers
the stromal swelling pressure, it seems logical to try to reduce
the IOP. In most cases of chronic edema (e.g., Fuchs, aphakic,
and pseudophakic edema), however, the tension is normal,
and it is not possible to lower the pressure safely to a stable
single-digit level that would be necessary to reduce the edema.
In acute glaucoma with high pressure and epithelial edema,
prompt lowering is mandatory. In most cases, the endothelium
is normal, and the epithelial edema disappears rapidly once
the pressure has been reduced to the level of 50 mmHg or less.
Only in situations of moderate tension elevation combined
with a marginally decompensated cornea can long-term pressure lowering be of clinical value. This scenario rarely occurs,
but it can be seen in grafts bordering on failure and also in some
dystrophies, such as Chandlers syndrome.75 The usual antiglaucomatous medication can be employed. In the long term,
however, edema tends to worsen gradually, and manipulation
of the pressure becomes increasingly futile.
SECTION 6
Hypertonic Agents
When hypertonic drops or ointments are instilled in eyes with
early epithelial edema, some clearing of vision often can be
achieved.10 Enough water is extracted temporarily out of the
epithelium to smooth the surface and reduce the diffraction of
light by the microcysts. Five percent sodium chloride drops for
daytime use and ointment of the same strength at night are
commonly used. Sucrose at 40% concentration (sticky),
anhydrous glycerol (painful), and a hair dryer (dusty) have
also been recommended but have no advantage over sodium
chloride. Similarly, a dry climate causes more rapid evaporation of the tear lm than does humid air. These measures
are often effective in morning edema in Fuchs dystrophy,
when they help to speed up the clearing of vision. Hypertonic
agents thus can be effective for months, rarely years, until the
edema has become constant and irreversible. Even when in
doubt about the efcacy of hypertonic agents, these agents are
almost always harmless to try.
Stromal edema, in contrast to epithelial edema, is not
affected by the topical use of hypertonic agents. The volume of
stromal water is simply too massive and too rapidly replenished
across the leaky endothelium to be influenced.
Soft Contact Lenses
450
Chronic corneal edema, once it has reached the bullous stage,

is not only blinding but also painful. The tugging of the blebs
and their corneal nerve endings by the lids during blinking
made life miserable for many patients before keratoplasty
became successful. In cases in which there is likely to be little
or no recovery of vision with transplantation but in which
comfort is important, a soft hydrophilic contact lens often is
an excellent tool. Vision is likely to decrease somewhat but
marked comfort is restored in about three-quarters of patients.76
I prefer the thick, high-water-content lenses that are left in
place around the clock for months at a time. Antibiotics are
probably not necessary but can be given in low doses for the rst
few weeks. The incidence of infection is unknown but probably

low. Soft contact lenses have become a valuable therapeutic
modality in cases of painful bullous keratopathy in which
keratoplasty is not indicated.
SURGICAL TREATMENT
Before the 1950s there was simply no cure for chronic edema,
and patients often had to live out their lives in blindness and
frequently in pain. The rst successful transplantations for
edema were reported in 1952,77 and others followed rapidly.78
Since that time, progress has been rapid because of greater care
in handling the endothelium, larger grafts, the availability of
steroids, and the development of ner suture material.79 Today,
most patients with corneal edema can be helped by keratoplasty
a remarkable surgical success story (see Fig. 40.10).
CORNEAL TOPOGRAPHY
As an optical component of the visual system, the swelling
properties of the cornea discussed above relate to light scatter.
In summary, mild epithelial edema can produce the symptoms
of halos around bright lights or Sattlers Veil, while moderate
stromal edema can also decrease visual acuity primarily
through light scatter, although this does not become signicant
until swelling of 70% is achieved. The most critical element
in preserving corneal optics is the status of the corneal surface
and tear lm. Disruption of the tear lm or irregularities in
the corneal epithelial surface such as caused by basement
membrane dystrophy, bullous keratopathy, infectious keratitus,
trauma, ectatic degenerative disease (keratoconus, pellucid
marginal degeneration, terriens marginal degeneration, and
keratoglobus), and keratectasia and other complications subsequent to refractive surgery can all cause signicant visual
loss. Corneal topographers have emerged as a powerful tool
with which to assess the etiology of factors that degrade vision
by producing irregularities on the corneal surface that lead to
optical aberrations.
The corneal tear lm/air interface provides about two-thirds
of the vergence of the eye. Thus, it plays a critical role in the
quality of the optics of the eye. Furthermore, because of this
property, even small amounts of surface distortion can greatly
reduce the quality of the retinal image. Direct examination of
the corneal surface with the biomicroscope does not provide
enough resolution to detect vision-reducing irregular astigmatism. Although retinoscopy provides a greater sensitivity to
irregular astigmatism, the distortion seen in the retinal reflex
(e.g., scissoring and distorted shadows) does not always indicate
the nature or the location of the irregular astigmatism. The
interpretation of retinoscopy is, therefore, subjective and details
of the origin of image blur are absent.
Nevertheless, it has been known for more than 300 years
that one can study the corneal curvature through observation of
reflected geometric patterns from the corneal surface.80
Reflection techniques, such as the Placido disk, keratometry,
photokeratoscopy, and corneal topography all arise from this
principle. However, it was not until the development of corneal
topography that clinicians were provided with easily understood color-coded maps of corneal curvature as well as quantitative indices of irregular astigmatism that correlate with
potential visual acuity. This allows the clinician to evaluate the
entire cornea both qualitatively and quantitatively.
A thorough understanding of the fundamentals and applications, as well as limitations of corneal topography, is of great
importance to the anterior segment surgeon. The remainder of
this chapter explores the background and fundamentals of corneal topography analysis appropriate to the clinical audience.

This is followed by a concise pictorial essay of corneal shapes
that are commonly seen clinically. Finally, a brief discussion of
future developments is presented.
BACKGROUND
FIGURE 40.12. The NIDEK PKS-1000 photokeratoscope provided a

Placido disk photograph of the corneal surface.
that such patients could lose more lines of best spectacle

corrected vision than one could account for by increased
cylinder alone. A contact lens over refraction would often
provide the clue that irregular astigmatism in the graft might
be more debilitating to visual acuity than induced cylinder.
To examine irregular astigmatism, a larger portion of the
corneal surface had to be analyzed than could be measured
with the keratometer. For this, a Placido disk/camera system
was developed the photokeratoscope (Fig. 40.12). Two of the
most widely distributed photokeratoscopes were the Nidek
PKS 1000 (Nidek Co.) and the Corneascope (Kera Corp.).
These devices produced a rapid print of a Placido disk image
from the patients cornea. Interpretation was accomplished by
visual inspection of the mires. Mires became closely spaced in
the areas of the cornea that had a high curvature such as
the region of the cone in keratoconus, more broadly spaced in
areas of lower power, and irregular near areas of tight sutures
(Fig. 40.13). The reading of photokeratoscopy was indeed
subjective, nevertheless the information was clinically useful.
Photokeratoscopy did, however, have another limitation; that
is, the devices did not cover the central area of the cornea well,
leaving much of the corneal area important to visual acuity
unanalyzed. Additionally, the devices were not able to cover
the corneal periphery, which limited usefulness in the contact
lens tting area.
With the increased practice of refractive surgery in the early
1980s, the entire corneal surface topography had to be
accurately and objectively evaluated. Doss and associates83,84
were among the rst to publish a method for the automatic
scanning and calculation of corneal power from a
photokeratoscope. Klyce extended this approach to the Nidek
photokeratoscope and explored methods that might be used for
the quantitative presentation of corneal topography to the
clinical audience (Fig. 40.14).85 This work culminated in the
CHAPTER 40
The study of corneal topography dates back to 1619 when

Father Christopher Scheiner realized that one could estimate
corneal curvature by comparing the reflection of a window on
the corneal surface to that on a series of different sized
marbles.80 The Placido disk, introduced in 1880 by Antonio
Placido, consists of a circular target of alternating white and
black rings or mires with a central aperture through which
one can view its virtual image. This image is formed by
reflection of the target from the surface of the tear lm. The size
and the shape of the image features depend on the fact that
convex mirrors will produce a magnication that varies directly
with their radius of curvature. A highly curved surface with
a short radius of curvature will have a low magnication,
making the virtual image of a Placido target appear small. In
contrast, a surface with less curvature will have a greater magnication, making a virtual image appear large. Corneas are
more complex than convex mirrors in their shape; therefore,
the curvature and thus magnication can vary considerably
from one segment to another. However, it is important to note
that all corneas are convex in shape; this includes both normal
prolate corneas as well as oblate corneas that may have
undergone refractive surgery.
In the late nineteenth century, Helmholtz developed the
ophthalmometer.81 This instrument was extremely difcult to
use; however, out of this invention grew the rst clinical
keratometer, which was introduced by Javal and Schioetz for
the measurement of anterior corneal curvature. The modern
keratometer measures corneal curvature along the orthogonal
steep and flat principal meridians by manual rotation of a
dial; the autokeratometer generally improves the repeatability of
the measurements between observers. The sites of measurement are four positions on the corneal surface ~34 mm apart,
depending on the underlying curvature of the cornea. Using
the standard keratometric index, the radius of curvature for
two orthogonal meridians is then converted into dioptric
powers (see later, Eqn [6]). Because the keratometer is designed
to make its measurements from only four positions on the
surface, this device can be used to accurately reproduce the
curvatures of only spheres and ellipsoids, and it does this with
an accuracy better than 0.25 D. However, because it can
measure only spherocylindrical surfaces, the keratometer
cannot be used to detect the myriad of shapes that corneas
can exhibit. The assumption that the cornea can be modeled as
an ellipsoid is at best an approximation even for normal
corneas. For corneas with irregular astigmatism, only gross
amounts can be appreciated with the keratometer and then
described with the general sign irregular mires. Nevertheless,
keratometry remains very useful for anterior segment applications such as intraocular lens (IOL) calculations and contact
lens tting for corneas that are not diseased or affected by
surgery or trauma.
Despite the surgical success enjoyed with corneal transplantation, the development of ever-improving methodologies for
the preservation of donor eyes, and the ever-expanding network
of eye banks worldwide, the transplant surgeon is still faced
with one major hurdle: eliminating or reducing the postoperative astigmatism of the graft. To meet this challenge,
Troutman82 and other pioneers had to rely on the keratometer
to measure induced regular astigmatism, and a good result
was a graft with less than 4 D of regular astigmatism and corrected vision better than 20/40. However, it was often observed
451

FIGURE 40.13. Photokeratographs with the
NIDEK PKS-1000. (a) Mild keratoconus. Note
the characteristic pear-shaped inner mires.
(b) Penetrating keratoplasty. Although the
central mires look fairly circular, there are large
amounts of cylinder and irregular astigmatism
in this early postoperative examination. Note
also that the early Placido disk
photokeratoscopes do not provide good central
corneal coverage.
FIGURE 40.14. Wire mesh stereo pairs to

represent corneal topography an initial
approach to displaying shape.6 The heights are
amplified; a normal cornea would have a flat
appearance with this method. (a) Cornea after
radial keratotomy. Note the bowl shape created
by amplifying the corneal height. (b) Stereo
pairs of mild keratoconus.
SECTION 6
FIGURE 40.15. The color-coded contour map

of surface powers for a normal cornea
introduced by Maguire and associates (lefthand panel).7 This early version used manually
traced enlargements from the NIDEK
photokeratoscope and predated widespread
use of color monitors and printers. A numeric
map is shown in the right-hand panel.
452
introduction of the color-coded contour map for the

presentation of corneal surface powers by Maguire and
associates86 (Fig. 40.15), which has now become the standard
display for corneal topographers.
The initial work with manually analyzed photokeratoscope
photographs demonstrated the clinical and research value of
corneal topography and led to several observations on normal
and abnormal corneas (see Fig. 40.15).8790 However, it was
not until the advent of the affordable personal computer that
this technology was commercialized. The rst commercial
device available was the Corneal Modeling System (CMS)

(Computed Anatomy, Inc, New York) in 1988, which at a cost
of $80 000 was accessible only to major clinical research
centers. This device used image analysis techniques to capture
and process both Placido disk images from the corneal surface
as well as cross sectional slit images that could be used to
model both surfaces of the cornea and to provide pachometry.
The scanned slit images constituted a feature that was discontinued in order to reduce the costs and make the product
clinically accessible.
FIGURE 40.16. Influence of mire spacing on spatial resolution in

videokeratoscopy. Dotted line, model surface of a central island of
elevated power, simulating an unwanted feature that is sometimes
seen in refractive surgery. In this test, there is a sharp transition
between curvatures (40.08/41.91 D). Solid line, response of a
videokeratoscope that samples every 180 m on a 40-D surface.
Dashed line: response of a videokeratoscope that samples every
270 m on a 40-D surface. Note that sparse sampling can smooth out
irregular astigmatism in the corneal surface.
After Belin MW, Ratliff CD: Evaluating data acquisition and smoothing functions
of currently available videokeratoscopes. J Cataract Refract Surg 1996; 22:421.
BASIC PRINCIPLES
Instrumentation
The CMS was the rst of a growing number of devices for
measuring corneal topography, and this class of machine
employing the videocapture of Placido disk images has become
known as a videokeratoscope. The more generic term for
devices that measure corneal shape is corneal topographer.
This device displays results in the form of a color-coded contour
map. There can be differences in the results obtained with the
various videokeratoscopes. The area of the analyzed cornea
varies for the different machines depending on the type of
Placido target used. Those devices with a long working distance
have a large diameter and part of the target is always masked by
eclipse of the brow and nose, whereas those with a compact
cone and short working distance do not suffer from peripheral
data loss to shadows. Automatic alignment and focus or compensation for misalignment is critical for corneal topographers
with short working distances. The ability for a corneal
topographer to show ne detail is somewhat variable owing to
differing resolutions (spacing between mires or mire edges). An
example of this is shown in Figure 40.16, where wider than
optimal spacing of mires results in smoothing of the devices
response to curvature change.91
In addition, reconstruction algorithms vary among the
devices, and this can degrade the nominal accuracy of 0.25 D,
particularly in the corneal periphery.86,92 With Placido disk
reflection corneal topography, there is no exact equation or
set of equations that can be used. Each device using this principle must make a series of approximating calculations, which
can be quite accurate93; in early implementations, this has on
occasion produced misleading results such as the presentation
of a keratoconus pattern where none exists.94 Abnormalities
of the tear lm can result in poor quality mires and misleading
results as well, calling for scrutiny of the mire processing for
accuracy assurance. Despite its limitations, Placido disk-based
systems remain the most successful methodology for corneal
topography analysis because of their sensitivity and
reproducibility.
As mentioned earlier, the use of slit beam technology can
provide the opportunity to analyze both the outer and inner
surfaces of the cornea. Since both of these refracting surfaces

as well as corneal thickness come into play when calculating
total corneal power, measurement of the position of the surfaces
directly would appear to provide an advantage. Additionally,
because each surface can be measured directly with slit beam
technology, no approximation errors should arise as with the
Placido disk-based devices. There are, however, disadvantages to
this approach as well. Because the cornea is in motion from
muscle tremor, pulse, and xation nystagmus, the entire image
must be captured in a minimum time of 30 ms. Capturing
successive slit images over a period of time longer than this
requires either an eye tracking system or a post capture image
registration technique to avoid movement artifact. The former
is too expensive for routine clinical use and the latter has
proven to be impractical because of the absence of registration
landmarks on the clear cornea. The second limitation of the slit
image technique is that measurement of the position of the
surface directly cannot lead to the same measurement sensitivity as measurement of the position of a reflected image.
Notwithstanding these comments, slit scanning has proved to
be a valuable adjunct to obtain corneal thickness proles, while
corneal topography can be provided with the traditional
Placido disk approach.
Another approach that avoids the uncertainties in the calculations of corneal topography is rasterstereography.9597 With
this approach, fluorescein is rst instilled in the tear lm, and
a grid or raster pattern is projected with cobalt blue light onto
the anterior surface of the eye. Images are then captured simultaneously from two directions and processed using triangulation methodology to reconstruct the shape of the cornea. This
method is also less sensitive than corneal topography for the
reason given earlier, and this factor along with the inconvenience of having to instill fluorescein to make the measurement reduces its usefulness. Nonetheless, rasterstereography
can supply corneal shape data of very irregular corneas, such as
corneal transplants, over a broad area extending well out past
the limbus.
Perhaps the most accurate methodology that can be used to
measure shape is interferometry.98100 Interference techniques
are used in the optical industry to detect aberrations of lenses
and mirrors to subwavelength accuracies. In essence, a reference surface (or its hologram) is compared with the measured
surface, and interference fringes are produced as a result of
differences in the two shapes. With respect to the measurement
of corneal shape, there is such a wide variation in the shapes of
corneas, even among those that are normal, that it is difcult
for a single interference device to represent all the variations.
Examples of interference devices include a phase modulated
laser holography-based device101,102 and an acoustic holographic
technique.103 This approach has not yet been found practical
for the measurement of corneal topography.
Methods of Power Calculation

Corneal topography devices measure the shape or curvature of
the corneal surface. A corneal topographer does not measure
beyond the surface, and therefore the corneal power reported by
the device is based on a series of geometric calculations and
assumptions. The convention that has been adopted is the
same as that used for decades, when only the keratometer (which
like the corneal topographer does not measure beyond the
corneal surface) was available for the measurement of corneal
curvature. This convention leads to the following expression:
P = 0.3375 / Rc
CHAPTER 40
[6]
where P is the corneal power in diopters; Rc is the local radius

of curvature in meters; and the keratometric constant, 0.3375,
is the difference between the refractive index of air and the
453
SECTION 6

refractive index of an equivalent cornea with the thickness
and back surface effects for the average cornea considered.
Although this relationship has been adopted because of its
widespread clinical use, it should be remembered that it will not
accurately reflect changes in corneal power after refractive
surgery since this may alter corneal thickness and may preserve the curvature of the endothelial surface. The effect
usually leads to an overestimation of the refractive change by
11.4%. If this method of power calculation is not appropriate in
a given situation, it is possible to change the settings in the
corneal topographer to report radius of curvature in millimeters,
rather than power in diopters. The surface radius of curvature
measurement can then be used with the measured corneal
thickness and endothelial curvature to obtain a more correct
approximation of corneal power after refractive surgery.
However, note that keratometry will not provide accurate
readings from the central area of the surgical cornea; alternative
methods must be used (see further ahead).
Corneal power, when calculated from front surface curvature, might best be called keratometric power, because it derives
from the keratometric index. However, there are other considerations in the estimation of corneal power (and curvature)
that will affect the result. Because the keratometer measures the
curvature at only two points on each of two meridians, the
meridians can only be interpreted as circular arcs. For this
reason, keratometers were calibrated with reference spheres and
the results reported were very accurate for these spheres.
Calibrating corneal topographers with the same approach leads
to representation of corneal power as axial or spherically based.
Although this representation is preferred for routine clinical
diagnosis, details of corneal topography that are important to
understand certain aberrations that occur after corneal surgery
can be made more apparent with other methods.
Corneal power can also be calculated from local curvature
data.85 However, a more elegant and precise method for calculating local corneal curvature has been shown to be through
the use of the instantaneous radius of curvature.104 This
method may be preferable for evaluating shape changes after
refractive surgery, for example, but it suffers from system noise
due to its extreme sensitivity to small changes in radius of
curvature over short distances.
A nal method that may be used to calculate corneal power
is that of refractive power, which is calculated from ray tracing
and Snells law. This has the effect of showing the residual
spherical aberration of the corneal surface.93,105 This
information is useful to the optical scientist, but is of little
value to the clinician, because it displays an optical aberration
that is believed to be further compensated for if not eliminated
by the native lens of the eye and by neural processing.
A nal word about power calculation methods in corneal
topography is in order. The reconstruction algorithms among
corneal topographers may vary, and measurement accuracy
generally is lower in the corneal periphery.106 However, in the
central 3 mm diameter of the cornea, all of the methods provide
nearly the same result for the same cornea; it is this portion of
the cornea that is most important to the formation of the image
on the retina. Clinically, the axial method for power calculation
is preferred. It is a direct representation of corneal shape
without the confusion of spherical aberration or measurement
noise.
Presentation Methods
454
Devices that map the cornea collect information from

thousands of data points. In order for these devices to be useful,
the data must be distilled into a form that is unambiguous and
rapidly interpreted. Initially, a common presentation technique
was numeric, with powers displayed on a geographic rep-
resentation of the collection site (see Fig. 40.15). Another

approach,85 is a three-dimensional wire mesh model of surface
powers presented as stereo pairs (see Fig. 40.14). Although this
technique conveyed some topographic information, the
information content was disappointingly low and difcult to
appreciate by the clinical audience.
A major breakthrough in the clinical use of corneal topography analysis came with the idea of the color-coded contour
map of corneal powers (see Fig. 40.15).86 A color spectrum was
chosen so that cool colors were associated with low corneal
powers and warm colors were associated with high corneal
powers. Only a few distinct colors were chosen over the
central range of corneal powers so that a specic power range
could be easily identied. Normally occurring corneal powers
were assigned color values in the green part of the spectrum
to further assist in the identication of normal versus abnormal with this emerging clinical test. It was rapidly apparent
that along with color association, the contours of the color
maps provided diagnostic capability through pattern recognition
and this combination of factors combined with the appropriate
scale (see later) would achieve wide acceptance and use.
Additional options have enhanced the utility of corneal topography. Plotting the contour map directly on the video image
of the patients eye was helpful to convey scale and position of
topographic features and can be particularly useful when
evaluating post-penetrating-keratoplasty or post-cataractsurgery corneas. Plotting metric scales (rectangular or polar) on
the display is likewise helpful to locate the position and meridian of salient features. The ability to view multiple patient
examinations simultaneously can permit a comparison of several eyes with a similar disease or can present the time course
of a diseased (Fig. 40.17) or postsurgical cornea (Fig. 40.18).
Difference maps are quite useful to demonstrate the early
postoperative effect of a surgical procedure, to examine the
effects of wound healing over time, or to watch for progression
(keratoconus; see Fig. 40.17) or regression (central island after
excimer laser; Fig. 40.19) of specic topographic features.
Power displays of the corneal surface are useful to understand corneal optics, but there are a number of situations in
which height or actual corneal shape would provide useful
information. For example, in the sculpting of corneal tissue
with the excimer laser, a true height map would be essential
to detail the effect of the removal of tissue by photoablation.
Heights can be presented directly by the corneal topography
units that do not use reflection keratoscopy; corneal
topographer data can also be used, but the algorithms used to
calculate height must be carefully validated.
Finally, ray tracing diagrams can be used to subjectively
visualize the optical quality of the cornea that is being
examined. Maguire and associates107 used such techniques to
calculate the modulation transfer function of surgical and
nonsurgical corneas and showed the effects on the blur of eye
chart symbols. Such routines are currently available on corneal
topographers to help assess the optical quality of the corneal
surface (Fig. 40.20).
Standardized Scales
The clinical utility of corneal topographer technology depends
to a large extent on how well the color-coded maps can be
interpreted. Two aspects are involved: (1) color association, with
the cool, blue end of the spectrum representing low powers
and the warm end of the spectrum representing high powers;
and (2) pattern recognition, with the contours representing
specic topographic entities. Despite initial efforts at
standardization in corneal topography by national and international organizations, none have been nalized or adopted.
Since standards are essential for the comparison and sharing

FIGURE 40.17. Progression of keratoconus in
a young male patient. Upper left, 8/88; lower
left, 7/89; upper right, 1/90; lower right, 7/92.
of information, those that have been proved and published in

the peer reviewed literature are set forth in the following.
A number of different color scales have been used since the
original introduction of the International Standard or Absolute
Scale (Fig. 40.21).7 This scale spanned a range of corneal
powers from 9 to 101.5 D, with 1.5 D intervals in the middle of
the range and 5 D intervals at each end. Wilson and co-workers
introduced a more practical scale (the Klyce/Wilson scale),
which ranges from 28 to 65.5 D in equal 1.5 D intervals.108
With the advent of refractive surgical corrective procedures
for high myopia, it was important to make the lower range of
the scale have uniform intervals to prevent masking of irregular
astigmatism in the central, surgically flattened cornea. This
scale has been revised slightly to form a universal standard
seeking adoption by the American National Standards
Institute.109
Even with this alteration, it was often argued that the 1.5 D
interval was so wide that important features in corneal topography may be hidden between contours. The diagnostic
adequacy of the Klyce/Wilson scale was evaluated in a clinical
series that included normal corneas, contact lens-wearing
corneas, early to moderate and advanced keratoconus, penetrating keratoplasties, extracapsular cataract surgery, excimer
laser photorefractive keratectomy, radial keratotomy, aphakic
epikeratoplasty, and myopic epikeratoplasty. It was found that
the correct interpretation for all cases could be made with the
1.5 D scale without resorting to a 1 D or lower interval scale.108
Additionally, the 1.5 D scale proved broad enough to cover the
full range of powers encountered in the study. The routine use
of a xed standard scale showing only adequate detail and not
redundant information or extraneous noise is essential for
efcient and accurate clinical interpretation.
CHAPTER 40
FIGURE 40.18. Videokeratograph of the left

eye of a myopic patient after laser in situ
keratomileusis (LASIK). Upper left, preoperative
eye. Follow-up: lower left, 2 weeks; upper right,
2 months; lower right, 6 months after surgery.
This format is useful to study topographic
stability.
455

FIGURE 40.19. Difference map of the right eye
of a patient after LASIK. (a) Note the
astigmatism in the preoperative cornea. (b) One
month after surgery, the cornea shows a central
steep area central island within the ablated
zone. (c) Difference map. Difference maps
should always be used when diagnosing a
central island because the preoperative
astigmatism may appear as a central island in
the postoperative cornea. In that case, a
difference map would show no central island.
c
b
SECTION 6
FIGURE 40.20. Ray tracing is being used to

evaluate corneal optics from topography. The
figure represents the intensity of light focused
on the retina as two point sources move from
convergence to separation. The distance of
separation at which the two peaks in the
distribution can be resolved by the eye is
related to visual acuity.
456
It has been tradition that corneal topographers provide

adaptable scales that are self-adjusting to the range of powers
found for a given cornea. The use of such scales runs counter to
standardization in corneal topography and is misleading in
interpretation. Such scales make grossly irregular corneas look
uncomplicated and quite normal corneas look complex with
extensive amounts of irregular astigmatism. Such adaptive
scales should be avoided, except as an adjunct to examine
details of corneal topography.
User adjustable scales are often also available on corneal topographers. These may be necessary to suit a specic application
or device, but it is recommended that in the learning phase a
user become accustomed to a single xed scale.
easily yield numeric measures such as keratometry. Therefore,

in order to complement the information provided by the color
maps, a number of quantitative indices have been developed
and are derived from the corneal topography data les. In
addition to corneal power(s), each topographer will have
calculated the three dimensional shape of the cornea, generally
as a set of heights from a plane normal to the corneal vertex
corresponding with measurement sites on the mires. Most
often, 256360 sites are measured for each mire along
semimeridians, like spokes of a wheel, giving denser coverage
in the central cornea than in the periphery. These data can thus
be accessed, and indices can be derived to supplement the
corneal maps.
Quantitative Indices
Keratometry
The color-coded contour map of corneal power distribution

provides a powerful method for diagnostics through the
association of particular colors with specic corneal powers and
the recognition of patterns from the contours. Although these
maps are based on measured data, they do not by themselves
Simulated keratometry (SimK) was one of the rst indices to be

broadly available.87 This index aims to simulate the readings
that a keratometer would yield; that is, the maximum power of
the surface along any axis and the power orthogonal to that
power. These were designated SimK1 and SimK2 and were

FIGURE 40.21. An eye with 2.5 D of regular
oblique astigmatism shown on different scales.
Upper left, absolute scale. Note that the two
ends of the range are in 5-D intervals. Lower
left, Klyce/Wilson scale. All the contour intervals
are 1.5 D; because of the narrow range of
powers in this cornea, this map and the
absolute scale map are identical. Upper right,
normalized scale. This is a self-adapting scale
that makes clinical interpretation difficult
because it can overemphasize small changes in
a nearly normal cornea, as it does here. This
scale can also deemphasize topographic
details in a cornea with large amounts of
irregular astigmatism. Lower right, adjustable
scale. This allows the user to develop a special
purpose scale.
parameter, called the average corneal power (ACP), was

compared with simulated keratometry values for normal
corneas, astigmatic corneas (cylinder 1.5 D), and corneas that
had undergone radial keratotomy (RK) or photorefractive
keratectomy (PRK). No disparity was found between the SimK
readings and ACP values for normal or astigmatic corneas, but
a disparity of 0.5 D or more was found for signicant numbers
of RK (7%) and PRK (25%) eyes.117
Therefore, where keratometry readings are used for refractive
power calculations, better measurements can be obtained by
appropriate calculations using corneal topographer data to
estimate the average curvature of the central cornea. These
values can be particularly important to improve the accuracy of
IOL calculations for certain cases after keratorefractive surgery.
Measures of irregular astigmatism

Irregular astigmatism is the remainder after subtracting sphere
and cylinder from a corneal power map. With reference to
optical analytic tools, irregular astigmatism is equivalent to the
higher-order (HO) terms in the surface tting Zernike
polynomial series; hence, irregular astigmatism is also referred
to as the HO aberrations, which include the familiar
components: spherical aberration and coma. In this discussion,
irregular astigmatism will be the preferred terminology, since
the HO aberrations obtained with the Zernike method do not
capture all of the corneal aberrations associated with visual
function, particularly in aberrated eyes.118 On the other hand,
with the very sensitive topographers available, some of the
irregular astigmatism may be clinically insignicant; indeed, a
certain amount is present in even normal corneas (see later).
Clinically the locus of irregular astigmatism and its impact on
vision are assessed with a contact lens over refraction. Reducing
the power of the irregular corneal surface to a thin tear
meniscus beneath a contact lens proves the etiology of reduced
acuity but does little to display the nature of the aberrations.
Initially, the clinician could instill fluorescein under a rigid
contact lens, which is helpful to depict gross shape anomalies
such as keratoconus, but the specic character of the irregular
astigmatism in an individual cornea needs to be known in order
to manage cases of reduced acuity. Corneal transplants, ocular
trauma, cataract surgery, and even scleral buckles can produce
vision impairing irregular astigmatism. Keratorefractive surgery
CHAPTER 40
written in standard notation as: 42.5 85/43.25 175, for

example, with the power units given either in diopters or
optionally in millimeters. These measurements are collected
from the data present on corneal topographer mires that
represent positions on a cornea that would be similar to those
positions of the keratometer mires on the same cornea, a
separation distance of 34 mm. Cylinder and spherical
equivalent are easily calculated from the SimK values and are
often provided with the printout of the color map.
Several investigators have pointed out that keratometry may
only be valid on normal corneas that do not have irregular
astigmatism, since taking measurements along only two axes
makes the assumption that the measured surface is either
spherical or ellipsoidal. Keratometry has long been used in
refractive power calculations for IOLs and has been shown to be
adequate for normal corneas.110,111 However, studies have
shown that the predictability of standard IOL power calculations can be reduced in patients who have irregular astigmatism or who have undergone radial keratotomy.112,113 By its
design, simulated keratometry based on corneal topography
data will suffer the same consequence as its progenitor, and
better estimates of central corneal curvature seemed necessary
to improve the predictability of IOL calculations.
Maloney114 introduced methodology to nd the best t
spherocylinder to the corneal topographer central mires. A more
direct approach was taken by Celikkol,115 who found a good
correlation between the average power of the third corneal
topographer mire and refractive accuracy after IOL implantation
in eyes that had undergone refractive surgery. Fourier analysis
has also been used to separate corneal power into spherical,
cylindrical, and irregular astigmatic components,116 and this
sophisticated approach may nd practical application with
clinical testing. As mentioned earlier, the density of sampled
data points is greater in the center of the cornea than in the
periphery owing to the usual procedure of radial sampling from
the center to the limbus. In fact, over sampling occurs in the
innermost mire region, because the same pixel may be sampled
multiple times owing to overlapping scans. In order to
compensate for this fact and to provide a good estimate of
refractive power, an algorithm was developed117 that produced
an area compensated average of corneal power from the central
cornea demarcated by the apparent entrance pupil. This
457
SECTION 6
458
can produce aberrations unique to that procedure, such as

central islands with excimer laser area ablations, decentered
treatment areas, and undesired spherical aberration.
Viewing the color-coded contour maps has permitted
classication of corneal shapes in both normal and abnormal
eyes. However, a most useful complement to corneal topography analysis has been the development of indices that
are predictive of the potential visual acuity (PVA) of an eye
based on the topographic character of the analyzed corneal
surface. The rst such index that was developed along these
lines was the surface regularity index (SRI).86,119 SRI is
calculated from a summation of local power fluctuations along
256 equally spaced semimeridians on the central 4 mm of the
cornea. SRI increases with increasing irregular astigmatism and
approaches zero for a smooth corneal surface. There was a
statistically signicant correlation found between the SRI and
best spectacle corrected visual acuity in a prospective clinical
study that included eyes of normals, keratoconics, and
transplants (r = 0.80, P <.001). The SRI and the PVA that
is derived from SRI have been useful clinically as a guide to
the optical performance that might be associated with a
particular irregular astigmatism and as a quantitative index
for monitoring the effect of refractive surgical procedures in
clinical studies.
Another useful quantitative descriptor is the surface asymmetry index (SAI). The SAI is a centrally weighted summation
of differences in corneal power between corresponding points
180 apart on 128 equally spaced meridians crossing the
corneal topographer mires.86,119 SAI approaches zero for a
perfectly radially symmetric surface and increases as the corneal
shape becomes more asymmetric within specic meridians.
Since the normal cornea usually has a high degree of central
radial symmetry, the SAI is a useful quantitative parameter for
monitoring changes that occur in patients following refractive
surgery. For example, decentration of a refractive procedure will
cause an increase in the SAI value. In addition, since the
steepening that occurs with keratoconus is generally located off
center, SAI increases greatly in these cases. As SAI increases,
there is an associated decrease in vision, although the
correlation is not as strong as with SRI.
Since the early work on topographic indices, a number of
additional variables have been used for the clinical assessment
of irregular astigmatism. Seiler has calculated the spherical
aberration from corneal topography examinations of post-PRK
corneas. He showed that corneal topography can be used to
calculate spherical aberration and that after PRK, the amount of
spherical aberration correlates well not only with measured
glare visual acuity but also with best spectacle corrected visual
acuity.120 Similar results have been found using the Zernike
polynomial method to examine spherical aberration after
LASIK.
Because of the lack of standardization in the eld, an index
developed with one topography unit will not be directly comparable to an index developed on another unit, even if the same
equations are used for its calculation. However, this limitation
can be overcome using a Fourier ltering technique, which has
been demonstrated using a wide mire corneal topographer and
a ne mire corneal topographer.121 Hence, it is instructive to
review the indices developed for the TMS 1, and some of these
are discussed here. The coefcient of variation of corneal power
(CVP) is a measure of the distribution of corneal powers in a
topography examination over the entrance pupil. This measure
and the standard deviation of corneal power (SDP) are related
in that the CVP is equal to the SDP divided by the average
corneal power. They both increase as the range of powers
increases in the measured topography. Examples of high CVP
and SDP are keratoconus corneas, transplants, and trauma
cases. Studies show that CVP, which is calculated from the

powers limited to the parts of the cornea ahead of the entrance
pupil, is often the strongest correlation to best spectacle
corrected visual acuity in refractive surgical eyes.
Some have pointed out the potential benet of a fortuitous
bifocal cornea following refractive surgery122 in patients who are
provided with near and distance vision. In reality, more often
surgical corneas are multifocal, or more accurately, varifocal and
this may lead to problems with decreased contrast sensitivity
and visual acuity. Furthermore, with a decentered procedure
untreated peripheral cornea can intrude upon the entrance
pupil producing annoying visual symptoms. Varifocality has
been found to correlate with best spectacle corrected visual
acuity; it may be expected to be a sensitive correlate to contrast
sensitivity as well due to increased image blur. With better
understanding of physiological optics, some procedures such as
conductive keratoplasty are offering simultaneous functional
near and far vision.123
Another measure of irregular astigmatism is the elevation
depression magnitude (EDM). Whereas the SRI is a measure of
high-frequency distortion, EDM is a measure of low frequency
distortion. In essence, it is a measure of the size and power of
the bumps and pits in the topography. It has been used as a
measure of central islands after refractive surgery.
The irregular astigmatism index (IAI)124 is an area compensated average summation of inter ring power variations
along every meridian for the entire corneal surface analyzed. It
is analogous to the SRI, but, whereas the SRI is calculated for
the central cornea to be more representative of Snellen acuity,
the IAI is calculated from the whole analyzed surface to be
more representative of overall corneal irregular astigmatism.
IAI is particularly high in corneal transplants shortly after
surgery; persistence often heralds suboptimal best spectacle
corrected vision.
The analyzed area (AA) gives the fraction of the corneal area
covered by the mires that could be processed. AA is lower than
normal for corneas with gross, irregular astigmatism, which
causes the mires to break up and not be resolved. A lower than
normal AA is found with early postoperative corneal transplants, advanced keratoconus, and trauma. AA can also be
artifactually low when the eyes are not opened wide.
The corneal asphericity index (CAI) is a quantitative descriptor that indicates the eccentricity of the central cornea. CAI
is calculated by tting an ellipse to the average curve obtained
from the 256 semimeridians out to the twenty-fth mire. The
CAI for 22 control corneas was reported to be 0.33 0.26 (SD),
which corresponds with the prolate shape of the normal central
cornea.125 This value is useful in contact lens tting and for
differentiating between normal corneas and corneas flattened
by myopic refractive surgery.
CLINICAL APPLICATIONS OF CORNEAL

TOPOGRAPHY
Topography of the Normal Cornea
The normal cornea tends to exhibit a great deal of variation
from one individual to another as well as asymmetry from one
area of the cornea to another. A thorough understanding of
the topography of the normal cornea is of fundamental importance to distinguish them from those corneas affected by
trauma, surgery, or disease. In addition, the topography of the
normal cornea must be understood in relation to vision in
the design and planning of corneal surgery.
For almost 100 years, it has been known that the normal
cornea is aspheric with the central cornea being steeper than
the periphery (Fig. 40.22). This change in curvature compensates somewhat for spherical aberration in the eye. In 1989,

FIGURE 40.22. All the characteristics of
normal corneas are illustrated in this pair of
maps from the same person. The corneas are
steeper centrally than peripherally; the contours
are relatively regular; the left and right corneas
have a similar mirror image symmetry, and the
specific color and contour pattern of this
individual are quite unique, like a fingerprint.
Dingeldein and Klyce126 studied eyes with uncorrected vision

of 20/20, no history of contact lens wear, and no evidence of
other corneal abnormalities. They found that the average
central corneal power to be 42.84 D and that the corneas did
flatten progressively toward the limbus. However, the degree
and the rate of flattening as well as the location of the area of
shortest radius of curvature varied widely from one subject to
another. Curiously, in none of these normal eyes did the cornea
flatten more rapidly temporally than nasally. Additionally, when
viewed with the color-coded contour map, each normal cornea
shows a unique pattern, like a ngerprint. Moreover, the
topography of fellow eyes tends to be mirror images of each
other (enantiomorphs). A nal characteristic of normal corneas
is that even with these variations between individuals, normal
corneas are relatively smooth in keeping with their optical
performance requirements.
Corneal Topography and Astigmatism

Regular astigmatism
Naturally occurring regular astigmatism reveals itself in corneal topography as a bow-tie pattern (Fig. 40.23). When Bogan
and associates examined the ne detail of normal corneal
topography with an expanded (0.4 D contour interval) scale,
they found that 22% of the corneas had round patterns, 21%
had oval patterns, 7% had irregular patterns, and 50% had
bow-tie patterns.127 Corneas that had measurable amounts of
keratometric cylinder also exhibited the bow-tie pattern,
which conrms the use of corneal topography to detect
cylinder. If one were to make a contour map of a sphere with a
small amount of cylinder, a fan shaped gure would result;
however, as noted earlier, the cornea is naturally a prolate
ellipsoid with cylinder added to that geometry. The 3 to 4 D
of flattening occurring between the corneal center and the
CHAPTER 40
FIGURE 40.23. With-the-rule cylinder has a

characteristic bow-tie pattern. Again, note the
mirror image symmetry.
459

periphery turns the fan shape into the bow-tie conguration
seen in the contour maps.
Corneal topography can provide quantitative measures of
astigmatism. Dingeldein and co-workers showed that there is a
high degree of correlation between the weighted average
powers from photokeratoscopes and the average keratometric
powers (r = 0.96, P <.001).87 Later Wilson and Klyce showed a
good correlation between a similar measure, the simulated
keratometry from corneal topography, to the keratometric
values.119 Sophisticated Fourier decomposition techniques116 as
well as polynomial tting methods can also be used to measure
corneal astigmatism. To correlate astigmatism to refractive
cylinder, such calculations are usually performed only for the
portion of the cornea over the entrance pupil. However, it is
important to note that the magnitude of refractive astigmatism
does not always agree with corneal astigmatism; the lens and
the macula can occasionally be responsible for all or a part of
refractive astigmatism.
SECTION 6
Irregular astigmatism
460
Irregular astigmatism takes many forms and has many causes,

and it can be dened as any aberration that diminishes vision.
It was noted earlier that Bogan and associates found some
topographic patterns in normal 20/20 eyes that exhibited
irregular contours.127 Irregular contours seen with nonstandard,
high-sensitivity topography scales is not pathognomonic for
irregular astigmatism, although the structural detail seen with
this magnication is unique from one person to another.
However, using a standard scale, such the 1.5 D contour
interval suggested for use with corneal topographers, irregular
contours become appreciable only when there is a concomitant
visual decit and, as mentioned earlier, the extent of irregular
astigmatism can be measured with a topographic quantity
such as the SRI.
Irregular astigmatism rarely occurs naturally, although it is
often a component of the corneal ectasias. Irregular astigmatism is associated most often with trauma and ocular surgery. Irregular astigmatism that is radially symmetric, such as
spherical aberration, has much less impact on vision than
corneal asymmetries such as coma. The patient with central
keratoconus will achieve functional spectacle vision much
longer than a patient with a similarly advanced cone in the
more typical inferior position. Likewise, a patient with contact
lens warpage consisting of central flattening may tolerate
this without complaint, whereas a patient with asymmetric
contact lens warpage will often complain of spectacle blur.
Signicant irregular astigmatism can often be corrected with
rigid contact lens wear as long as the lenses are tolerated.
Beyond this, corneal transplantation may be required, except
where topographic analysis suggests that asymmetric astigmatic
keratotomy (AK) may improve corneal shape. Phototherapeutic
keratectomy (PTK) which uses the excimer laser and a
smoothing agent has had some success smoothing highly
irregular corneas. Customized ablation with an excimer laser
has made strides toward improving vision in eyes with smaller
amounts of irregular astigmatism.
and clinical keratoconus (an area of corneal steepening with

one or more of the classical clinical signs: corneal thinning,
scissoring of the light reflex on retinoscopy, Vogts striae, or
Fleischers ring). Somewhat arbitrarily, one can further divide
clinical keratoconus into mild, moderate, and advanced (see
Fig. 40.17) based on variables such as the change in corneal
power from the base of the cone to its apex. However,
keratoconus is often a continuously progressive disease and,
therefore, discrete classication may only be appropriate until
quantitative measures of corneal involvement are agreed upon.
Keratoconus suspects are detected most easily with corneal
topography: it is the most sensitive means for screening.130
Previously, clinicians used distortion of keratometer or
keratoscope mires and scissoring of the light reflex during
retinoscopy as signs of the irregular astigmatism often present
with preclinical (suspect) keratoconus. Relatively rapid changes
in refraction were often noted to accompany keratoconus in
the progressive phase. However, with corneal topography, the
keratoconus suspect cornea is easily identied using the
standard 1.5 D interval scale (see Fig. 40.17), even though by
denition there are no other clinical signs and the only visual
symptom may be spectacle blur from associated irregular
astigmatism. The earliest topographic signs are recognized by
a localized area of corneal steepening some two or more
contour intervals above the surrounding topography. Although
atypical inferior steepening131 is a common nding with
keratoconus suspect cases, the steepest part of the cone in
clinical keratoconus may be found in any quadrant; indeed, it
may be located centrally or superiorly.132 Keratoconus is
almost always bilateral, although one cornea is almost always
more involved than the other (Fig. 40.24). In a few cases
(2.44%),133,134 where one eye has moderate to advanced
involvement, keratoconus is unilateral in its topographic
appearance. It is likely that a fraction of these cases will develop
keratoconus in the eye that appears normal at a later time. It is
important to note that the metabolic underpinnings leading to
ectasia in one eye are certain to be present in the contralateral
eye even when clinical signs are not currently evident. When
keratoconus is bilateral, the cone apex seems to be located in
the two eyes at corresponding positions. If the cone is
inferotemporal in one eye, it is inferotemporal in the other eye;
if the cone is central in one eye, it is central in the other eye.
Pseudokeratoconus
The presence of evidence from corneal topography alone is
not sufcient to make the diagnosis of clinical keratoconus
because of the possible confounding influences of contact lensinduced warpage (Fig. 40.25), misalignment artifact (not a
consideration with modern corneal topographers),94,135 tear
meniscus artifact from excessive tearing or the addition of
a viscous articial tear solution, or inadvertent external
pressure on the globe. All of these situations can lead to a
pseudokeratoconus; that is, a topographic pattern and sometimes retinoscopic ndings similar to those seen in clinical
keratoconus. The latter three artifacts can be eliminated by
repeated corneal topography, but contact lens warpage can
persist for weeks or months (see later).
Corneal Topography in Ectasias

Keratoconus
Pellucid marginal corneal degeneration
Keratoconus is the most prevalent of the ectatic corneal

dystrophies128 and has a reported incidence in the general population as high as 0.6%. It is characterized as a noninflammatory localized thinning disorder that can lead to anterior
protrusion of the cornea and the development of visual impairment through irregular astigmatism and stromal scarring over
the visual axis. This condition can be classied into two groups:
keratoconus suspect129 (a local area of mild corneal steepening)
Whereas keratoconus produces an ectasia from a focal thinning

of the corneal stroma, pellucid marginal corneal degeneration,136 which is much less common, is associated with a 1 to
2 mm wide band of inferior corneal thinning usually near the
limbus from the 4 to the 8 oclock meridian. Because of the
difference in thinning pattern from keratoconus, pellucid marginal corneal degeneration produces a characteristic arcuate
band like ectasia of the inferior cornea with marked flattening

FIGURE 40.24. Keratoconus is almost always
bilateral, and one cornea is more involved than
the other (in this case OD). Note that, while OS
at first impression looks fairly normal, the lazy
eight astigmatic pattern is a common
characteristic of keratoconus.
of the central cornea along the vertical meridian (Fig. 40.26),

which is quite different from the conical protrusion seen with
keratoconus. Usually, against-the-rule cylinder is present. This
topographic pattern can be present before the classic inferior
thinning pattern of pellucid marginal corneal degeneration is
detectible.137 While the perilimbal band-like thinning usually
occurs inferiorly, the degenerative thinning with pellucid
marginal corneal degeneration can appear in other quadrants as
well, with a concomitant change in cylinder axis.
However, in some patients, central corneal topography may

appear relatively normal when the peripheral area of thinning
is small or when the thinning extends around the entire
circumference of the cornea.138
Keratoglobus and Posterior Keratoconus. Both keratoglobus
and posterior keratoconus can produce corneal astigmatism,
but the rarity of these conditions has limited topographic
characterization.
CHAPTER 40
FIGURE 40.25. This patient has corneal

topography that might be misinterpreted as
keratoconus. However, this person wears
contact lens and has contact lens warpage.
Automated Screening for Keratoconus

Terriens marginal corneal degeneration
Terriens marginal corneal degeneration is also uncommon
and involves perilimbal corneal thinning that can lead to
topographic changes depending on the extent of thinning
involved. The most frequent topographic pattern is the presence
of a high amount of against the rule astigmatism, which is
present when there is superior or inferior thinning (Fig. 40.27).
Because of the widespread application of refractive surgery for

the correction of ametropia, corneal topography has been able
to provide an important role in preoperative evaluation. As
noted below, a high incidence of keratoconus has been found
in the population of patients who elect refractive surgery. This
situation provided the rst opportunity to characterize and
recognize a corneal topographic abnormality with articial
461
SECTION 6
FIGURE 40.26. Pellucid marginal corneal degeneration is

characterized by the central flattening in the vertical meridian and the
inferior band of steepening.
462
intelligence techniques. With the variety of topography instruments now available and in the absence of universally adopted
standards, it is not possible to use the color-coded map by itself
to differentiate between the normal aspherical and sometimes
asymmetric topography and the abnormal cornea. Keratoconus
often rst appears on corneal topography as a localized area of
steepening, inferiorly displaced. It would seem useful to provide
a quantitative method to discriminate corneas with such
steepening from those with clinical keratoconus.
Rabinowitz and McDonnell139 were the rst to use a numeric
method to detect keratoconus systematically with data from
corneal topography. They examined dioptric power differences
between the superior and inferior paracentral corneal regions,
which were designated I-S values, the central corneal power or
MaxK, and the differences in power between the two eyes. They
considered that if the central corneal power is greater than
47.2 D or if the I-S value is greater than 1.4 D, then the cornea
could be considered keratoconus suspect. Further, if the central
corneal power is greater than 48.7 or the I-S value is greater
than 1.9, then the cornea could be classied as keratoconus.
Although these criteria are able to distinguish the topography
of keratoconus corneas from normal corneas, its specicity
was not optimized. The numeric approach has been extended
with the use of topographic indices calculated from corneal
powers and areas, which are used as the input to an expert
system classier.124 The eight topographic indices are: SimK1
and SimK2 (simulated keratometric steep and flat axis powers),
the surface asymmetry index (SAI), the differential sector index
(DSI), the opposite sector index (OSI), the center/surround
index (CSI), the irregular astigmatism index (IAI), and the
analyzed area (AA). The system was trained with examinations
of 22 clinical keratoconus (a mixture of mild, moderate, and
advanced stages) and 78 nonkeratoconus corneas (normals,
regular astigmatism, keratoplasty, epikeratoplasty (EPIK),
photorefractive keratectomy, radial keratotomy, contact lensinduced warpage, astigmatic keratotomy, scarred corneas,
postretinal detachment surgery, postcataract surgery, and
keratomileusis). The purpose of including the abnormal corneas
in the nonkeratoconus group was to permit the detection
system to discriminate these from clinical keratoconus.
Analysis yielded the keratoconus prediction index (KPI), which,
in turn, was introduced to a binary decision tree to differentiate
between central and peripheral keratoconus. Validation of this
FIGURE 40.27. Terriens marginal corneal degeneration. The

prominent feature here is the marked against-the-rule astigmatism.
approach was done with a second set of topographic examinations consisting of 28 keratoconus corneas and 72 normal
and abnormal nonkeratoconus corneas.
The results of Maeda and associates124 showed a sensitivity
of 100% in the training set and 89% in the validation set. Three
keratoconus corneas that were diagnosed with clinical
keratoconus (based on the medical records) were not conrmed
by the classication scheme; however, because two of these
maps resembled contact lens-induced corneal warpage and one
resembled pellucid marginal degeneration, this discrepancy
seemed acceptable. Specicity was 96% in the training set and
99% in the validation set. All of the false positive results that
were classied as keratoconus involved eyes that had undergone keratoplasty and that had the corneal steepening characteristic of keratoconus. The occurrence of these false negative
and false positive results in the validation set seemed reasonable given the fact that the classication of the validation set
was done with no more information than was available from
the data supplied by a single corneal topographer examination.
This method has been compared with the Rabinowitz/
McDonnell method as well as with a simple method based on
keratometry readings alone using a sample of examinations
independent from the groups upon which any of the methods
were based.140 The sensitivity for keratometry was 84%, for
the Rabinowitz McDonnell method was 96%, and for the expert
classier system was 98%, whereas for specicity the three
tests had values of 86%, 85%, and 99%, respectively. The performance of the expert classier system in terms of specicity
was signicantly better than either of the two other methods
(P = 0.001).
The aforementioned methods have used indices derived from
the surface power values available from corneal topographers.
Schwiegerling and Greivenkamp141 proposed the use of Zernike
polynomials to t the actual three dimensional shape data to
detect keratoconus. Although the method showed some merit,
it did not appear to be as sensitive or specic to detect
keratoconus as the discriminant analysis approach mentioned
earlier.
The automatic detection of keratoconus is a good rst step
in the development of a topographic classication system. The
approach has been be extended with the use of a neural network that is able to classify a number of categories of corneal
topography in addition to keratoconus, including normals,
transplants, and astigmats.142,121
Corneal Topography in Cataract Surgery
Postoperative uses
Corneal topography has a wide range of applications in the

preoperative as well as the postoperative management of the
cataract surgery patient.
Preoperative Uses. Studies using corneal topography have
shown that for phacoemulsication, smaller, temporal, and
scleral incisions cause less induced astigmatism.143148 Clear
cornea incisions tend to produce astigmatism 90 from the
incision and peripheral corneal flattening. Thus, superior clear
corneal incision is only recommended for patients with signicant preexisting with-the-rule astigmatism. Planned extracapsular cataract extraction (ECCE) may yield different results
in that the superior approach may result in less astigmatism,149
and surgically induced corneal cylinder occurs most frequently
along the axis where the incision is placed.150 Small incision
techniques have greatly reduced the impact of the surgery on
corneal topography.
Corneal topography data have also been used in the calculation of IOL power. In the normal spherical or spherocylindrical
corneas, the simulated keratometry values obtained from
corneal topography are in agreement with standard keratometry
values.110,151 Nevertheless, the usefulness of corneal topography
in the calculation of IOL power is controversial. Cuaycong110
and Antcliff152 found measures of corneal power from corneal
topography to yield smaller errors in predicted postoperative
refraction than when using keratometry. However, Husain
found that standard keratometry was more accurate than the
corneal topography or keratometric equivalent.111 Nevertheless,
corneal topography is most useful in the calculation of IOL
power for eyes with irregular surfaces, such as diseased or postsurgical corneas. After refractive surgery, standard keratometry
may overestimate the power of the central cornea and thus
result in signicant amounts of postoperative hyperopia. This is
primarily because the central cornea flattens with myopic
refractive surgery and becomes progressively steeper peripherally.
Keratometry readings tend to measure points peripheral to this
central area of flattening and produce articially greater
keratometric readings (Fig. 40.28).113,153 This problem is not
limited to refractive surgical corneas154,155 and probably affects
all patients after refractive surgery. In the case of RK, Celikkol
has found that use of the mean power from ring 3 of the TMS
1 for keratometric power yields postsurgical refractive results
closer to the ideal than standard keratometry.115
Corneal topography is the most sensitive way to examine

the entire cornea for changes induced by cataract surgery. Vass
and co-workers have shown that corneal topography and
keratometry have comparable sensitivities and specicities in
detecting corneal changes in the paracentral cornea induced by
cataract surgery.156 Nevertheless, they showed that cataract
surgery frequently results in peripheral corneal changes, irregular astigmatism, and asymmetric regular astigmatism that
cannot be detected or measured using keratometry. Thus,
corneal topography should be performed in every patient with
decreased visual acuity after cataract surgery with an otherwise
normal examination in order to rule out irregular astigmatism
as the cause.
Postoperatively, corneal topography can be used to detect
irregular astigmatism as well as direct suture removal. The
aforementioned quantitative indices can be used to measure as
well as follow the amount of astigmatism present after
surgery.150 In the future, it may be possible to perform corneal
topography intraoperatively during combined procedures as
well as planned cataract surgery in order to maximize the
refractive benets of cataract extraction and clear lens
extractions.
Corneal Topography in Penetrating Keratoplasty
CHAPTER 40
Greater tissue availability, improvements in grafting techniques, and better tissue preservation have greatly increased
the success rate of penetrating keratoplasty (PKP). However, the
post-PKP cornea frequently shows excessive amounts of
regular as well as irregular astigmatism. Irregular astigmatism
can result from wound conguration abnormalities such
as ovality/overcut or dehiscence, a thin recipient cornea, graft
elevation, or uneven tension on interrupted sutures
(Fig. 40.29).157 Efforts to improve the refractive results after
PKP have included using contact lenses to mold the corneal
surface after PK in order to reduce astigmatism and increase
the regularity of the cornea,158 adjustable running sutures,159
and selective suture removal (Fig. 40.30).160,161 Nevertheless,
sometimes each of these measures can lead to a further increase
in astigmatism. Thus, corneal topography should be used to
direct these interventions in an effort to maximize the
reduction in astigmatism. Furthermore, in those cases where
the previous measures failed, corneal topography has been used
FIGURE 40.28. In this decentered excimer laser PRK procedure,

keratometry may give an erroneous result. The crosses are placed at
the approximate position where keratometer mires would read the
steep axis.
FIGURE 40.29. Irregular astigmatism after penetrating keratoplasty

due to uneven suture tension.
463

night vision in keratorefractive surgery patients.120,173178 These
corneal topography studies have contributed to further the
development of keratorefractive surgery and have helped in the
understanding of the optical performance of the postoperative
cornea.179
The topography of the different refractive procedures and
their complications and also the role that corneal topography
has played in the development of each procedure are now
described.
Radial keratotomy
FIGURE 40.30. Selective suture removal can improve topography.

Top, Presuture removal. Bottom, Reduced astigmatism after removal.
to guide the placement of arcuate incisions and compression

sutures for the reduction of post-PKP astigmatism.162
SECTION 6
Corneal Topography in Refractive Surgery

General considerations
464
Corneal topography should always be performed before

refractive surgery in order to detect preexisting corneal
abnormalities such as irregular astigmatism, ectasias such as
keratoconus, and contact lens-induced corneal warpage. These
abnormalities are frequently undetected without corneal
topography. If undiagnosed, surgery on corneas with these conditions can lead to devastating consequences, such as exacerbation of astigmatism and loss of best corrected vision with
the development of kerectasia. In addition, if a keratoconus
pattern is seen preoperatively in a contact lens wearer, corneal
topography should be evaluated over time to differentiate true
keratoconus from contact lens-induced changes.163 This
becomes more important when one considers the fact that these
abnormalities may be overrepresented in the refractive surgery
population owing to self-selection.164 Postoperatively, corneal
topography has been used to evaluate decentration of the
refractive surgery,165167 fluctuating vision,168 multifocality,169
regression,164,170 induced astigmatism, and central islands
that can result after refractive surgery; such undesirable features can go undetected with other clinical diagnostic
modalities.86,88,159,171,172 This information has been used to
explain complaints such as glare, halos, and difculty with
While RK was rarely performed after the successful introduction of excimer laser techniques, it is instructive to recant
the history of its development for what was learned about the
optics of the cornea. In 1981, 3 years after Bores performed the
rst radial keratotomy in the United States, a multicentered
clinical trial of a single standardized technique of radial
keratotomy was initiated. This became known as the
Prospective Evaluation of Radial Keratotomy (PERK) study.180
After 4 years, using data from manifest refractions, the PERK
study group found that the older the patient and the smaller
the clear zone, the greater were the refractive effects of the
surgery.181 Concurrently, they reported a 2.5% incidence of
loss of two lines or more in best spectacle corrected visual
acuity as well as patient complaints of glare and problems
with night driving; they speculated that these complaints
were linked to irregular astigmatism. Atkin and associates182
showed decreased glare contrast sensitivity after RK. Further
studies evaluating the topography of the post-RK cornea were
clearly needed.
Computer assisted corneal topographic analysis was not
commercially implemented until 1988.183 Thus, Rowsey and
co-workers,184 using a nine-ring photokeratoscope, keratometry,
and refraction, showed that a greater amount of flattening
was achieved in eyes with smaller clear zones and in older
patients. Later, they examined the influence of the preoperative
topography on the refractive change after RK.185 They found
that all preoperative corneas had a prolate (steeper centrally
than peripherally) shape and became oblate (steeper peripherally than centrally) postoperatively. They also demonstrated
that other factors were also involved in the prediction of refractive change after RK, namely, the preoperative corneal curvature (less effect with steeper preoperative corneas) and
horizontal corneal diameter (less effect for smaller diameter
corneas).
Although Rowsey did not comment on the etiology of the
vision loss or on the shape of the mires for those patients, it is
well known that RK can introduce irregular astigmatism.186188
Corneal topography of patients from the PERK study has
shown various degrees of irregular astigmatism with incision
sites being evident in some cases even a decade after surgery.
Other refractive problems after radial keratotomy include
glare, halos, diurnal changes of refraction and vision, regular
and irregular astigmatism, and early as well as late progressive
hyperopia.186,187,189191
The problem of diurnal fluctuations in vision and refraction
has been reported by many authors.100,168,192,193 It occurs in
1.960% of RK patients and may involve a myopic or a
hyperopic shift.168 This change in refraction is thought to be
related more to a change in corneal hydration and not to a
change in intraocular pressure.100,194198 However, the mechanism is not entirely understood. Most studies based on standard keratometry have not shown any correlation between
variations in refractive error and changes in keratometry.
However, this is most likely because keratometry does not accurately reflect the average corneal curvature over the entrance
pupil. McDonnell and associates168 used corneal topography to
Astigmatic keratotomy
Symmetric astigmatism of a magnitude and axis that matches
the magnitude and axis of refractive cylinder and which is too
great in magnitude to be corrected with excimer laser techniques can be treated with the methodology developed by
Thornton and others.204 Semiradial, transverse, and trapezoidal
incisions have been used to neutralize different amounts of
astigmatism.205 However, in cases where the astigmatism is
asymmetric, corneal topography should be used to direct the
relaxing incisions (Fig. 40.31). Software based on corneal
FIGURE 40.31. Asymmetric oblique regular astigmatism induced by

cataract surgery. Keratometry would not be sufficient to detect this
asymmetry; corneal topography was used to plan the procedure that
involved asymmetric relaxing incisions.
topography can be used to determine the best position and

conguration of the relaxing incisions in order to achieve the
desired result as long as refractive and topographic cylinder are
equivalent.206 Furthermore, astigmatic keratotomy can be
combined with radial keratotomy in the treatment of myopic
astigmatism, but higher amounts of irregular astigmatism may
result from this combined procedure than if each procedure is
done alone.207 Other methods to evaluate the topographic
results of AK include nite element modeling of the eye.208
Using this technique, Hanna and co-workers evaluated the
incision variables and their effect on the curvature of the incised
and unincised meridians: length (longer incisions cause more
steepening of unincised meridian), distance from the center of
the cornea (incisions further from the center cause less
flattening of the incised meridian), and depth (deeper incisions
cause more effect).208
In the surgical management of symmetric astigmatism, accurate determination of the axis of the steep axis is important,
because a small angular error will result in a relatively large
reduction of the anticipated induced cylinder. There are at
least three sources of error in measuring the steep axis with
keratometry, the rst of which is eliminated with corneal topography. Having been taught that eyes are most likely to exhibit
with the rule astigmatism, there is a natural tendency when
using the manual keratometer to report axis 90 rather than, for
example, axis 8595. Small errors in axis alignment produce
signicant undercorrection in cylinder. Use of the objective
measure of the SimK provided with corneal topography removes
this error of bias. However, the corneal topographer does not
eliminate the error in steep axis angular measurement caused
by head rotation in the head rest or incyclotorsion that can
occur from the stress of the clinical environment. Therefore,
surgeons must adopt strategies to carefully mark the cylinder
axis directly on the bulbar conjunctiva prior to astigmatic
keratotomy. Again, it is important to ensure that the corneal
cylinder matches the refractive cylinder (as it usually does)
before the surgical plan is implemented.
Epikeratoplasty
Epikeratoplasty was rst intended for the correction of aphakia
but was later applied to the correction of myopia and utilized
as an onlay lamellar patch to flatten the keratoconus cornea.
Computerized corneal topography was used to study the effect
of this procedure on the surface of the cornea when applied
to each of the three types of patients mentioned earlier. When
used for keratoconus, it causes compression of the cone and
flattening of both the anterior and posterior aspects of the
cornea, whereas when used in myopia, it causes flattening of
the anterior refracting surface of the cornea only.172 For aphakia,
the lenticle results in a steeper anterior surface.172 In addition,
it was through the use of corneal topography of eyes that had
undergone epikeratoplasty for myopia that decentration was
recognized as a complication of keratorefractive surgery.154 In
addition, this type of analysis revealed that it was important to
increase the size of the optical zone in order to maximize
refractive results.154
Corneal topography was helpful in identifying both regular
and irregular astigmatism after this procedure. Corneal topography was also used to show that moderate amounts of
astigmatism were compatible with good Snellen visual acuity209
and to study the effects of this type of astigmatism on optical
performance after epikeratophakia.210
CHAPTER 40
show that most people with diurnal fluctuations in visual acuity

had postoperative corneal topographies that had dumbbell
shaped or split optical zones. Patients with large, round central
optical zones proved to be largely immune to the problem of
diurnal fluctuations. With the use of corneal topography, Lucci
and associates199 examined the diurnal fluctuations in refraction and average corneal power of a subset of patients in the
PERK Study at their 10-year follow-up and conrmed that the
continued morning to evening increase in myopia was due to
central corneal steepening.
On a more long term basis, progressive hyperopia has been
reported and conrmed by the 10-year follow-up on the patients
in the PERK study.200 Corneal topography has been used to
analyze the specic regional changes that occur in the post-RK
cornea.191 It seems that early on after surgery, the central cornea
appears steeper than the midperipheral cornea whereas the
peripheral cornea is steeper than both. In time, the central
cornea flattens faster than the midperiphery so that it is no
longer relatively steeper.
In order to look at the possible etiology of glare and halos
after RK, Applegate and co-workers used corneal topography to
show an increase in the amount of spherical like as well as
coma like aberrations produced by RK.175 Several reports have
described the multifocality of the cornea after RK.122,169,201 This
multifocality can result in increased depth of focus and amelioration of presbyopia for some people and in decreased contrast
sensitivity in others.202,203 This multifocality may be the result
of the regional changes in curvature with time after surgery as
explained earlier.112 In addition, the increased spherical
aberration that can result from the gradient in curvature from
the treated area to the untreated area has been implicated in the
loss of contrast sensitivity after radial keratotomy.202
Hexagonal keratotomy
Hexagonal keratotomy was aimed at achieving a steeper cornea
for the treatment of hyperopia, presbyopia, and the overcorrection of RK. However, the procedure was plagued with
465
FIGURE 40.32. Myopic excimer laser PRK with uniform central

corneal flattening.
FIGURE 40.33. Hyperopic excimer laser PRK showing steepening of
the central cornea.
complications, including corneal ectasia, glare, photophobia,

polyopia, fluctuation in vision, overcorrection, irregular
astigmatism, corneal edema, corneal perforation, bacterial
keratitis, cataract, and endophthalmitis.211 Because of this,
hexagonal keratotomy has been largely abandoned.
SECTION 6
Photorefractive keratectomy
466
In 1983, Trokel ablated the cornea of freshly enucleated, bovine

eyes using an argon fluoride 193 nm excimer laser and showed
that this laser might be used for radial keratotomy because it
resulted in sharp grooves without thermal damage to adjacent
structures.212 At that time, Trokel also suggested that the laser
could be applied using a circular mask with a graded intensity
from center to edge in order to steepen or flatten the cornea.
Subsequent studies demonstrated its superiority over longer
wavelength excimer lasers in the production of corneal incisions
and its ability to flatten the cornea in order to correct myopia
(Fig. 40.32).213222 Today, photorefractive keratectomy has also
been used in the treatment of hyperopia (Fig. 40.33),223,224 as
well as combined myopic225 and hyperopic astigmatism.226
Initial ablations performed on rabbits were difcult to evaluate topographically227; however, improvements in laser technology and data collection allowed the evaluation of important
topographic data in primates.228 This led to further renements in both the surgical approach and in laser technology.
Corneal topography has subsequently been used to evaluate
regression, decentration, multifocality, and the optical performance of the post-PRK cornea.
Regression is a drift over time of the postoperative refraction
toward the preoperative refraction. In the early studies, it was
found that the post-PRK cornea was more likely to regress for
higher attempted corrections than for lower ones.229 This
regression is believed to occur secondary either to thickening
of the epithelial layer or to deposition of stromal collagen or
to both. For PRK corneas, stabilization tends to occur at
~ 6 months166 but can take up to 1 year.165,229 In some corneas,
regression is such that it leads to complete loss of the ablation,
but this is rare. It appears that communication between the
epithelial cells and the stroma is important in regression,230 and
there is growing evidence that the release of cytokines for the
epithelium as it is being scraped from the stromal surface may
be the cell signal that leads to anterior stroma keratocyte
apoptosis and subsequent stromal remodeling.231 In procedures
where Bowmans layer is not ablated (e.g., laser in situ
keratomileusis (LASIK)), regression is not as great a problem
and stabilization may occur within weeks after the procedure is
FIGURE 40.34. Myopic excimer laser PRK showing moderate

decentration (~0.7 mm from the pupil center). The pupil is indicated by
the black dotted outline.
performed.165 Surface ablation (PRK) became more widely

practiced with the development of devices with which to
remove the epithelium as a clean sheet. With the introduction
of Mitomycin C to block keratocyte activation, the incidence of
postsurgical stromal haze formation has been greatly reduced.
Mitomycin C was rst proposed to block scar formation after
PRK in the rabbit model.232
Decentration of the ablation is another problem encountered
after PRK. It was rst recognized as a complication with the
topographic analysis of epikeratophakia233 but can affect almost
all keratorefractive procedures. Uozato and Guyton suggested
that the center entrance pupil should be used as the center of
the ablation.234 Since then, decentration has been dened as the
distance of the apparent center of the ablation to the apparent
center of the pupil as viewed with corneal topography. When
severe, decentration can result in monocular diplopia, glare,
ghost images, astigmatism, poor visual acuity, and poor contrast sensitivity, particularly for small diameter ablations.235
Decentration is difcult or impossible to detect with traditional
tools,90,171 but it can be seen easily and can be measured with
topographic analysis (Fig. 40.34).170,236238 Using corneal
topography, Wilson and associates170 found that the average
decentration after PRK for the LSU phase IIA study was

FIGURE 40.35. Irregularities in the treated area
after excimer laser PRK for myopia. A
difference map must be used to document
induced irregular astigmatism. Here the
preoperative cornea (upper left) is subtracted
from a postoperative examination (lower left).
The difference map (right panel) shows no
significant induced irregular astigmatism; the
treatment was uniform despite the topographic
appearance of the ablation.
Phototherapeutic keratectomy
Partial thickness corneal scars treated previously with lamellar
keratoplasty can now be treated with phototherapeutic
keratectomy. Lamellar keratoplasty may induce large amounts
of irregular astigmatism. PTK can remove the scar noninvasively without as great a risk of inducing irregular astigmatism.
In addition, PTK can be used to treat recurrent erosions241 or
treat corneal irregularities after EKC or as a result of a corneal
dystrophy. In general, PTK leads to an improvement in corneal
topography but can result in a decrease in best spectacle corrected Snellen visual acuity.242 As noted above, customized corneal ablations may improve corneal optics when the aberrations
are not too severe.
Automated lamellar keratotomy

Automated lamellar keratotomy (ALK) originated from myopic
keratomileusis (MKM) developed by Barraquer.243 The
technique for MKM was very difcult, and the predictability of
the nal refraction was a problem. Furthermore, it was clear
that even patients with excellent postoperative visual acuity
could show signicant amounts of irregular astigmatism244 and
complain occasionally of glare and image distortion.90 Maguire
and associates90 reported a patient after MKM who experienced
marked visual distortion despite a normal slit lamp examination and a smooth corneal surface. Color-coded maps generated by computer analysis were used to show the degree of
irregular astigmatism, which was not evident from simple
inspection of the keratoscope photographs.90
The modern automated microkeratome for ALK has led to a
more regular treated zone. However, large amounts of irregular
astigmatism can occur (Fig. 40.37). Although strides in the
development of microkeratomes continue to be made, the
complication rate and the introduction of signicant amounts
of irregular astigmatism can limit its use.
CHAPTER 40
0.79 0.11 mm (range of 0.032.1 mm, 79% < 1 mm). With

improvements in technique, decentration was reduced to
0.47 0.06 mm for LSU phase IIB, but since that early report,
considerable improvements have been made in technique as
accurate alignment is particularly important for hyperopic
refractive corrections as well as for correction of higher-order
aberrations.
Multifocality of the postoperative cornea can reduce the
optical quality of the cornea. Seiler and co-workers have shown
that spherical aberration resulting from the gradient in refraction at the edge of the treated zone correlates highly with best
spherical corrected visual acuity in normal eyes and with
measured glare visual acuity in patients with PRK.120 Martinez
and co-workers have shown that both coma and spherical
aberration are increased by PRK, and this is dependent on pupil
size and attempted correction.173 These changes in higher-order
aberrations may account for problems with night driving, halos,
and loss of contrast sensitivity experienced by some patients
after refractive surgery.
Multifocality can also result from unequal ablation within
the treated area. In some cases, areas of local contiguous elevated power within the ablation zone and 2 mm or more in
diameter are seen in the postoperative topography and have
been called central islands.237,239 Clinically, central islands can
cause decreased vision, monocular diplopia, or decreased contrast sensitivity or create apparent over and undercorrections.
They occur rarely with the latest generations of excimer laser
but have been observed infrequently with LASIK procedures in
addition to the surface ablation techniques. Their etiology
remains controversial but may be due to degraded laser optics,
beam blockage by the plume of photodisrupted tissue, and
external hydration that results in unequal laser delivery to the
cornea.240 Corneal topography can be used to diagnose as well
as follow central islands after PRK. When diagnosing central
islands or other irregular astigmatism, one should use
difference maps, because preexisting irregular astigmatism may
appear accentuated after PRK (Fig. 40.35). True central islands
tend to resolve by 18 months after PRK (Fig. 40.36), although it
is tempting to remove these earlier with an estimated
enhancement procedure to improve visual performance.
Laser in situ keratomileusis

When PRK was rst introduced, patients could undergo unpredictable amounts of regression accompanied by prolonged
periods of rehabilitation and corneal haze. Myopic corrections
above 6 D were problematic, although some degree of success
467

FIGURE 40.36. Central island formation and
resolution after excimer laser PRK. Upper left,
Preoperative examination. Lower left, 1 month.
Upper right, 7 months. Lower right, 10 months.
FIGURE 40.37. Irregular astigmatism induced by ALK. Dislocated flap

or jamming of the microkeratome during the procedure can reduce
best-corrected spectacle visual acuity.
tends to stabilize within weeks. Solomon has shown that wound

healing in LASIK is very rapid, whereas PRK has a prolonged
response that lasts 46 months.232 This might explain the
difference in the amount of regression.
Irregular astigmatism can be a signicant problem after
LASIK, but its incidence varies widely.245,246 LASIK increases all
quantitative measures (CVP, EDM, and SRI) of irregular astigmatism.247 The increase in these indices can be comparable to
PRK. The amount of irregular astigmatism has been determined by the quality of the microkeratome cut. Moreover,
wrinkles on the flap and misalignment in repositioning the
flap may introduce some additional irregular astigmatism.
However, with the improvements in flap creation noted above,
signicant irregular astigmatism after LASIK or surface ablation
is rare. It is noted that spherical aberration and coma are
specic representations of irregular astigmatism, and it is known
that decentration leads to the induction of coma, and small
diameter treatment zones lead to the induction of unwanted
amounts of spherical aberration.
SECTION 6
Others
468
was achieved by successive myopic treatments with several

optical zone diameters (multizone, multipass). Furthermore,
reepithelialization requires several days during which vision
was reduced and many patients experienced discomfort. This
helped to promote LASIK, which is a combination of PRK and
ALK.230 During this procedure, a flap of the anterior cornea is
created with the microkeratome, and the underlying stroma is
ablated to produce the myopic correction; following this, the
flap is repositioned. Recovery of good to excellent visual acuity
is often fairly immediate and less discomfort is generally
experienced. While LASIK has had its share of complications
related to the mechanical microkeratome (buttonholes, free
caps, and wrinkles) and to interface opacities due to epithelial
ingrowth, debris, and haze formation, the introduction of the
femptosecond laser for the creation of the LASIK flap has
reduced many of these problems.
Using topography to measure the average central corneal
power, we have found that LASIK may result in greater stability
of refraction than PRK (see Fig. 40.18). In particular, refraction
The intracorneal ring, introduced by Kera Vision Corporation,

was originally applied to the correction of refractive error in
ametropes.248,249 Induced astigmatism and long term compatibility with corneal functional anatomy are issues of concern, but the potential for reversibility of the procedure is a
denite advantage. As well, intracorneal rings are being used to
improve vision in keratoconus patients. Intracorneal implant
technology has included devices such as deep stromal polyacrylate lenses, midstromal hydrogel lenses, and epistromal
collagen gels. Concerns with these devices include anterior
corneal nutrition, long term stability, and long term biocompatibility. Tissue necrosis can produce scarring, opacity, and
severe irregular astigmatism. Nevertheless, opaque annular
corneal inlays that have fenestrations for nutritional concerns
and a small 1.6 mm central clear zone are being explored to
improve near vision in presbyopes.
Corneal Topography and Contact Lenses

Contact lenses continue to be a good cosmetic alternative to
spectacles. Despite the incidence of complications that can have

devastating consequences to vision (e.g., amoebic, fungal, and
microbial keratitis), they can be a safer and more efcacious
modality than refractive surgery. As materials used in their
manufacture have improved in oxygen permeability (Dk),
biocompatibility, and wear comfort, adverse events associated
with contact lens wear are on the decline.
Effect of contact lens wear on corneal topography
FIGURE 40.38. Large amounts of irregular corneal astigmatism can

be induced with contact lens wear.
lenses can be tted to flatten the cornea in the case of myopic

correction, this change of shape is generally not permanent in
the absence of wearing a maintenance lens.257 In the presence
of irregular astigmatism, contact lenses are particularly useful
not only to reduce the visual distortion from the corneal
surface but can in some cases reduce the extent of irregular
astigmatism. Examples of this include the effect of contact
lens wear with clinical keratoconus, where the force of the
lens can reduce the ectasia and the use of contact lenses
after penetrating keratoplasty. It has been suggested that
irregular astigmatism can be reduced by tting corneal
transplant eyes with rigid lenses.158 Subsequent studies have
concluded that it is safe to t such eyes with contact lenses
after penetrating keratoplasty and that the topography remains
stable with time.258,259
Using corneal topography to t contact lenses

The use of keratometry readings to t contact lenses is traditional and is usually sufcient for successful wear. However,
as shown earlier, the normal cornea is neither spherical nor is it
completely symmetric. The amount that the corneal shape
deviates from an ideal geometric shape is related to the
adequacy of the t with conventional lenses. It has become
clear with the availability of computerized corneal topography
analysis that there is a new capability that can be used to
improve contact lens tting. Corneal topographers mathematically reconstruct the shape of the cornea; these data can
be used to evaluate the relationship between the contact lens
and the corneal surface by simulating the familiar fluorescein
examination (Fig. 40.39).260 With this facility it is possible to
observe the t of a number of trial lenses without testing all of
these directly in the patients eye.
Contact lens tting has been constrained to an art form in
the past, because with only the keratometer to measure corneal
shape, there was insufcient information available. To utilize
corneal topography more fully in this area, a number of contact
lens tting programs have been developed. Whereas early
software programs have not been very successful,261,262 subsequent versions include expert tting systems with a wide
variety of commercially available lenses that have met with a
measure of clinical success in both normal and pathologic
corneas.263,264
CHAPTER 40
Contact lenses embed themselves in the tear lm and are

held in place by the force of capillary attraction. The primary
resting position is a function of the relationship between the
shape of the contact lens and the shape of the cornea. Gravity
and the action of the lids act to displace the contact lens from
the primary resting position. Owing to thermodynamics, in the
absence of friction, the lens would always center itself in the
same position on the corneathe position of minimum entropy
where the overall space between the lens and the cornea is at a
minimum. Capillary attraction is a surface tension effect and
provides a force of negative pressure between the contact lens
and the cornea. This force plus the additional force of the lids
can cause the contact lens to alter the shape of the cornea.
When this contact lens-induced shape change is unintended, it
is generally referred to as corneal warpage (Fig. 40.38). There
are, however, situations in which this shape change is
intentional and these will be discussed later.
As mentioned earlier, contact lens-induced corneal warpage
can produce corneal topography that is indistinguishable from
preclinical and mild clinical keratoconus, and this has been
called pseudokeratoconus. Patients who have with the rule
corneal astigmatism and who wear decentered contact lenses
appear to be at risk. This combination of factors can flatten the
area of the cornea in the semimeridian under which the contact
lens is displaced and steepen the opposite semimeridian; the
result can resemble keratoconus.94 Such a case was presented in
Figure 40.25. These patients can complain of spectacle blur,
because asymmetric cylinder in the corneal surface cannot be
corrected with eye glasses. Whereas one management approach
is to discontinue contact lens wear altogether, a more satisfactory approach for patients with rigid gas permeable (RGP)
contact lenses has been to ret with a daily wear high water
content lens that can allow the corneal topography to return to
normal.250 This strategy did not work well for patients who
were wearing polymethylmethacrylate contact lenses and were
switched to RGP lenses.251
Contact lens-induced warpage is a concern in the preoperative screening of refractive surgical patients, not only
because warpage can resemble keratoconus but also because
contact lens-induced warpage can destabilize the refraction of
the eye. Contact lenses, both rigid and soft, can alter corneal
curvature. After discontinuation of contact lens wear, the
average length of time for corneal curvature to stabilize in
patients that were symptomatic for rigid contact lens-induced
warpage can be almost 15 weeks and for soft contact lensinduced warpage, the average time for stabilization was
5 weeks.163 The change in corneal power was often more than
1 D, and there were examples of both flattening and steepening.
Hence, to maximize predictability in a refractive surgical
procedure it is clear that, for patients with a history of contact
lens wear-even those who are asymptomatic252 -one should
obtain repeated refractions or, better still, repeated corneal
topography examinations until normal topography and
stabilization of refraction have been achieved.
Contact lenses have also been used to intentionally mold
the cornea, both to correct refractive error (orthokeratology
253255
) as well as to improve the optics of the cornea in the
presence of irregular astigmatism. Although there is renewed
interest in orthokeratology,256 it appears that, while contact
469
CONCLUSIONS
FIGURE 40.39. Simulation of the fluorescein pattern of a trial contact

lens for a cornea with mild keratoconus.
It has been over a century since Javal and Schioetz introduced

their keratometer to the clinical practice of ophthalmology.
Now, owing to the computer revolution, the extensive analysis
of corneal curvature with corneal topography has become a
common diagnostic test and the clinical applications of this
technology are numerous. It is not often that a medical advance
answers more questions than it poses, but when used
appropriately, corneal topography can provide a clarity of
perception that makes classication of corneal topography
nearly intuitive. Modern corneal topography had its beginnings
in the refractive surgery clinical research laboratories; it has
succeeded beyond expectations in its initial goal to provide
topographic analysis for keratorefractive surgery. As the eld
has matured, more accurate, smaller, and less expensive corneal
topographers have appeared. Finally, by combining slit beam
and Placido technology, we are getting close to being able to
measure separately the shape and refractive properties of both
surfaces of cornea and lens in order to compare these data to the
aberrometry measurements made from the entire eye!
ACKNOWLEDGMENT
Supported in part by the National Eye Institute, Bethesda, Maryland
(R01EY003311 and P30EY002377).
SECTION 6
REFERENCES
470
1. Kuwabara T: Current concepts in anatomy

and histology of the cornea. Contact
Intraocular Lens Med J 1978; 4:101104.
2. Hodson S: Why the cornea swells. J Theor
Biol 1971; 33:419427.
3. Maurice DM: The cornea and sclera. In:
Davson H, ed. The eye. 3rd edn. New York:
Academic Press; 1984:121.
4. Hedbys BO, Dohlman CH: A new method
for the determination of the swelling
pressure of the corneal stroma in vitro. Exp
Eye Res 1963; 2:122129.
5. Klyce SD, Dohlman CH, Tolpin DW: In vivo
determination of corneal swelling pressure.
Exp Eye Res 1971; 11:220229.
6. Hedbys BO, Mishima S, Maurice DM: The
imbibition pressure of the corneal stroma.
Exp Eye Res 1963; 2:99111.
7. Ytteborg J, Dohlman CH: Corneal edema
and intraocular pressure. II. Clinical results.
8. Klyce SD, Russell SR: Numerical solution
of coupled transport equations applied to
corneal hydration dynamics. J Physiol
1979; 292:107134.
9. Fatt I, Goldstick TK: Dynamics of water
transport in swelling membranes. J Colloid
Sci 1965; 20:434400.
10. Cogan DG, Kinsey VE: The cornea. V.
Physiologic aspects. Arch Ophthalmol
1942; 28:661667.
11. Maurice DM: The permeability to sodium
ions of the living rabbits cornea. J Physiol
(Lond) 1951; 112:367369.
12. Klyce SD: Electrical proles in the corneal
epithelium. J Physiol 1972; 226:407429.
13. Davson H: The hydration of the cornea.
Biochem J 1955; 59:2428.
14. Harris JE, Nordquist LT: The hydration of
the cornea. Am J Ophthalmol 1955;
40:100110.
15. Trenberth SM, Mishima S: The effect of
ouabain on the rabbit corneal endothelium.
16. Maurice DM: The location of the fluid pump

in the cornea. J Physiol (Lond) 1972;
221:4354.
17. Donn A, Maurice DM, Mills NL: Studies on
the living cornea in vitro. I. Method and
physiologic measurements. Arch
Ophthalmol 1959; 62:741747.
18. Donn A, Maurice DM, Mills NL: Studies on
the living cornea in vitro. II. The active
transport of sodium across the epithelium.
19. Klyce SD, Neufeld AH, Zadunaisky JA: The
activation of chloride transport by epinephrine
and Db cyclic-AMP in the cornea of the
rabbit. Invest Ophthalmol 1973; 12:127139.
20. Klyce SD, Beuerman RW, Crosson CE:
Alteration of corneal epithelial transport by
sympathectomy. Invest Ophthalmol Vis Sci
1985; 26:434442.
21. Klyce SD: Transport of Na, Cl, and water by
the rabbit corneal epithelium at resting
potential. Am J Physiol 1975;
228:14461452.
22. Hodson S, Miller F: The bicarbonate ion
pump in the endothelium which regulates
the hydration of the rabbit cornea. J
Physiol 1976; 263:563577.
23. Hull DS, Green K, Boyd M, Wynn HR:
Corneal endothelium bicarbonate transport
and the effect of carbonic anhydrase
inhibitors on endothelial permeability and
fluxes and corneal thickness. Invest
24. Lim JJ, Ussing HH: Analysis of presteadystate Na+ fluxes across the rabbit corneal
endothelium. J Membr Biol 1982;
65:197204.
25. Fischbarg J, Lim JJ, Bourget J: Adenosine
stimulation of fluid transport across rabbit
corneal endothelium. J Membr Biol 1977;
35:95112.
26. Hodson S, Miller F, Riley M: The electrogenic
pump of rabbit corneal endothelium. Exp
Eye Res 1977; 24:249253.
27. Iwata S, Lemp MA, Holly FJ, Dohlman CH:

Evaporation rate of water from the
precorneal tear lm and cornea in the
28. Kedem O, Katchalsky A: Thermodynamic
analysis of the permeability of biological
membranes to non-electrolytes. Biochim
Biophys Acta 1958; 27:229246.
29. Klyce SD: Stromal lactate accumulation
can account for corneal edema osmotically
following epithelial hypoxia in the rabbit.
J Physiol 1981; 321:4964.
30. Spencer WH: Ophthalmic pathology: an
atlas and textbook. Philadelphia: WB
Saunders; 1985:259 pp.
31. Michels RG, Kenyon KR, Maumenee AE:
Retrocorneal brous membrane. Invest
Ophthalmol 1972; 11:822831.
32. Iwamoto T, DeVoe AG: Electron
microscopic studies on Fuchs combined
dystrophy. I. Posterior portion of the
cornea. Invest Ophthalmol 1971; 10:928.
33. Waring GO, Bourne WM, Edelhauser HF,
Kenyon KR: The endothelium: normal and
pathologic structure and function.
34. Iwamoto T, DeVoe AG: Electron
microscopic studies on Fuchs combined
dystrophy. II. Anterior portion of the cornea.
35. Smelser G, Chen D: Physiological changes
in cornea induced by contact lenses. Arch
Ophthalmol 1955; 53:676679.
36. Bonanno J, Polse K: Corneal acidosis
during contact lens wear: effects of
hypoxia and carbon dioxide. Invest
37. Lancon M, Miller D: Corneal hydration,
visual acuity and glare sensitivity. AMA
38. Bruce AS, Brennan NA: Corneal
pathophysiology with contact lens wear.
39. Maurice DM: The structure and
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
transparency of the cornea. J Physiol 1957;

136:263268.
Miller D, Benedek GB: Intraocular light
scattering. Springeld, IL: Charles C
Thomas; 1973.
Zucker B: Hydration and transparency of
corneal stroma. Arch Ophthalmol 1966;
75:228231.
Grifths S, Drasdo N, Barnes D, Sabell A:
Effect of epithelial and stromal edema on the
light scattering properties of the cornea. Am J
Optom Physiol Opt 1986; 63:888894.
Goldman JN, Benedek GB, Dohlman CH,
Kravitt B: Structural alterations affecting
transparency in swollen human corneas.
Lambert SR, Klyce SD: The origins of Sattlers
veil. Am J Ophthalmol 1981; 91:5156.
Maurice DM: Cellular membrane activity in
the corneal endothelium of the intact eye.
Experientia 1968; 24:10941095.
Yee RW, Matsuda M, Schultz RO,
Edelhauser HF: Changes in the normal
corneal endothelial cellular pattern as a
function of age. Curr Eye Res 1985;
4:671678.
Bourne WM, Kaufman HE: Specular
microscopy of human corneal endothelium
in vivo. Am J Ophthalmol 1976;
81:319323.
Sato T: Studies on the endothelium of the
corneal graft. Jpn J Ophthalmol 1978;
22:114.
Bigar F, Schimmelpfennig B, Hurzeler R:
Cornea guttata in donor material. Arch
Ophthalmol 1978; 96:653655.
Burns RR, Bourne WM, Brubaker RF:
Endothelial function in patients with cornea
guttata. Invest Ophthalmol Vis Sci 1981;
20:7785.
Bourne WM, Walker RR, Liesegang TJ, et
al: Corneal trauma in intracapsular and
extracapsular cataract extraction with lens
99:13751376.
Kraff MC, Sanders DR, Lieberman HL:
Specular microscopy in cataract and
intraocular lens patients. Arch Ophthalmol
1980; 98:17821784.
Miller D, Stegman R: Use of sodium
hyaluronate in human intraocular lens
implantation. Ann Ophthalmol 1981;
13:811815.
Olsen T: Corneal thickness and endothelial
damage after intracapsular cataract
extraction. Acta Ophthalmol 1980;
58:424433.
Bourne WM: Morphologic and functional
evaluation of the endothelium of
transplanted human corneas. Trans Am
Matsuda M, Sada T, Manabe R: Long-term
observations of the graft endothelium with
different postoperative courses. Jpn J
Ophthalmol 1983; 27:556566.
Cavanagh HD, El-Agha MS, Petroll WM,
Jester JV: Specular microscopy, confocal
microscopy, and ultrasound biomicroscopy:
diagnostic tools of the past quarter century.
Cornea 2000; 19:712722.
Li J, Jester JV, Cavanagh HD, et al: On-line
3-dimensional confocal imaging in vivo.
41:29452953.
Kaufman SC, Musch DC, Belin MW, et al:
Confocal microscopy: a report by the
American Academy of Ophthalmology.
60. Hedbys BO, Mishima S: The thicknesshydration relationship of the cornea. Exp
Eye Res 1966; 5:221228.
61. Kremer FB, Walton P, Gensheimer G:
Determination of corneal thickness using
ultrasonic pachometry. Ann Ophthalmol
1985; 17:506507.
62. Mishima S: Corneal thickness. Surv
63. Thornton SP: A guide to pachometers.
64. Ginsberg SP, Capella JA, Kaufman HE:
Normal patterns of corneal thinning after
penetrating keratoplasty using refrigerated
cryopreserved donor material. Am J
Ophthalmol 1972; 73:555559.
65. Ehlers N: Graft thickness after penetrating
keratoplasty. Acta Ophthalmol 1974;
52:893903.
66. Maurice DM: A new objective
fluorophometer. Exp Eye Res 1963; 2:33.
67. Mishima S: In vivo determination of
fluorescein permeability of the corneal
endothelium. Arch Ophthalmol Rev Gen
Ophthalmol 1975; 35:191199.
68. Sawa M, Araie M, Nagataki S: Permeability
of the human corneal endothelium to
fluorescein. Jpn J Ophthalmol 1981;
25:6063.
69. Mishima S: Clinical investigations on the
corneal endothelium. Am J Ophthalmol
1982; 93:129.
70. Sawa M, Araie M, Tanishima T: Permeability
of the corneal endothelium to fluorescein: a
follow-up of keratoplasty cases. Jpn J
Ophthalmol 1982; 26:326337.
71. Wilson SE, Bourne WM, OBrien PC,
Brubaker RF: Endothelial function and
aqueous humor flow rate in patients with
Fuchs dystrophy. Am J Ophthalmol 1988;
106:270278.
79:7881.
73. Wilson SE, Bourne WM, Brubaker RF: Effect
of dexamethasone on corneal endothelial
function in Fuchs dystrophy. Invest
74. Donshik P, Cavanaugh D, Boruchoff SA,
Dohlman CH: Posterior subcapsular
cataracts induced by topical
corticosteroids following keratoplasty for
keratoconus. Ann Ophthalmol 1981;
13:2932.
75. Chandler PA: Essential atrophy of the iris
with glaucoma. In: Epstein DL, ed.
Chandler and Grants glaucoma.
Philadelphia: Lea & Febiger; 1986:396417.
76. Dohlman CH, Boruchoff SA, Mobilia EF:
Complications in the use of soft contact
lenses in corneal disease. Arch Ophthalmol
1973; 90:367371.
77. Stocker FW: Successful corneal grafts in a
case of endothelial and epithelial
dystrophy. Am J Ophthalmol 1952;
35:349352.
78. Fine M, West CE: Late results of
keratoplasty for Fuchs dystrophy. Am J
Ophthalmol 1971; 72:109114.
79. Abbot RL, Forster RF: Determinants of
graft clarity in penetrating keratoplasty.
80. Rabinowitz YS, Wilson SE, Klyce SD: Color
atlas of corneal topography. New York:
Igaku Shoin Medical Publishers; 1993;
135 pp.
81. Helmholtz HV: Hanndbuch der

physiologieschen Optik. Hamburg,
Germany: Leopold Voss; 1909.
82. Troutman RC: Surgical correction of high
astigmatic corneal errors after successful
keratoplasty. Adv Ophthalmol 1970;
27:170179.
83. Rowsey JJ, Reynolds AE, Brown R:
Corneal topography: corneascope. Arch
Ophthalmol 1981; 99:10931180.
84. Doss JD, Hutson RL, Rowsey JJ, Brown R:
Method for calculation of corneal prole
and power distribution. Arch Ophthalmol
1981; 99:12611265.
85. Klyce SD: Computer assisted corneal
topography: high resolution graphical
presentation and analysis of keratoscopy.
25:14261435.
86. Maguire LJ, Singer DE, Klyce SD: Graphic
presentation of computer-analyzed
keratoscope photographs. Arch
Ophthalmol 1987; 105:223230.
87. Dingeldein SA, Klyce SD, Wilson SE:
Quantitative descriptors of corneal shape
derived from computer assisted analysis of
photokeratographs. Refract Corneal Surg
1989; 5:372378.
88. Maguire LJ, Klyce SD, Singer DE, et al:
Corneal topography in myopic patients
undergoing epikeratophakia. Am J
Ophthalmol 1987; 103:404416.
89. McDonald MB, Sanders DR, Keates RH,
et al: Epikeratophakia for keratoconus: the
nationwide study. Arch Ophthalmol 1986;
104:12941300.
90. Maguire LJ, Klyce SD, Sawelson H, et al:
Visual distortion after myopic
keratomileusis: computer analysis of
keratoscope photographs. Ophthalmic
Surg 1987; 18:352356.
91. Belin MW, Ratliff CD: Evaluating data
acquisition and smoothing functions of
currently available videokeratoscopes. J
Cataract Refract Surg 1996; 22:421426.
92. Wilson SE, Klyce SD: Advances in the
analysis of corneal topography. Surv
Ophthalmol 1991; 35:269277.
93. Wang J, Rice DA, Klyce SD: A new
reconstruction algorithm for improvement
of corneal topographical analysis. Refract
Corneal Surg 1989; 5:379387.
94. Mandell RB, Chiang CS, Yee L: Asymmetric
corneal toricity and pseudokeratoconus in
videokeratography. J Am Optom Assoc
1996; 67:540547.
95. Warnicki JW, Rehkopf PG, Curtin DY, et al:
Corneal topography using computer
analyzed rasterstereographic images. Appl
Opt 1988; 27:11351142.
96. Arffa RC, Warnicki JW, Rehkopf PG:
Corneal topography using
rasterstereography. Refract Corneal Surg
1989; 5:414417.
97. Naufal SC, Hess JS, Friedlander MH,
Granet NS: Rasterstereography-based
classication of normal corneas. J Cataract
Refract Surg 1997; 23:222230.
98. Rottenkolber M, Podbielska H: High
precision TwymanGreen interferometer for
the measurement of ophthalmic surfaces.
99. Smith TW: Corneal topography. Doc
Ophthalmol 1997; 43:249276.
100. MacRae S, Rich L, Phillips D, Bedrossian
R: Diurnal variation in vision after radial
keratotomy. Am J Ophthalmol 1989;
107:262267.
CHAPTER 40
471
SECTION 6
472
101. Burris TE, Baker PC, Ayer CT, et al:

Flattening of central corneal curvature with
intrastromal corneal rings of increasing
thickness: an eye bank eye study. J
Cataract Refract Surg 1993;
19(Suppl):182187.
102. Shack R, Baker R, Buchroeder R, et al:
Ultrafast laser scanner microscope.
J Histochem Cytochem 1979;
27:153159.
103. Smolek MK: Holographic interferometry of
intact and radially incised human eye bank
corneas. J Cataract Refract Surg 1994;
20:277286.
104. Roberts C: Analysis of the inherent error of
the TMS 1 topographic modeling system in
mapping a radially aspheric surface.
Cornea 1995; 14:258265.
105. Wang JW, Silva DE: Wave front
interpretation with Zernike polynomials.
Appl Opt 1980; 19:15101515.
106. Roberts C: The accuracy of power maps to
display curvature data in corneal
topography systems. Invest Ophthalmol Vis
Sci 1994; 35:35253532.
107. Maguire LJ, Zabel RW, Parker P, et al:
Topography and raytracing analysis of
patients with excellent visual acuity 3
months after excimer laser photorefractive
keratectomy for myopia. Refract Corneal
Surg 1991; 7:122128.
108. Wilson SE, Klyce SD, Husseini ZM:
Standardized color coded maps for corneal
topography. Ophthalmology 1993;
100:17231727.
109. Smolek MK, Klyce SD, Hovis JK: The
Universal Standard Scale: proposed
improvements to the ANSI standard
corneal topography map. Ophthalmology
2002; 109:361369.
110. Cuaycong M, Gay C, Emery J, et al:
Comparison of the accuracy of
computerized videokeratography and
keratometry for use in intraocular lens
calculations. J Cataract Refract Surg 1993;
19(Suppl):178181.
111. Husain SE, Kohnen T, Maturi R, et al:
Computerized videokeratography and
keratometry in determining intraocular lens
calculations. J Cataract Refract Surg 1996;
22:362366.
112. Modorati G, Pierro L, Brancato R:
Preoperative astigmatic influence on the
predictability of intraocular lens power
calculation. J Cataract Refract Surg 1990;
16:591593.
113. Koch DD, Liu JF, Hyde LL, et al: Refractive
complications of cataract surgery after
radial keratotomy. Am J Ophthalmol 1989;
108:676682.
114. Maloney RK, Bogan SJ, Waring GO III:
Determination of corneal image properties
from corneal topography. Am J Ophthalmol
1993; 115:3141.
115. Celikkol L, Pavlopoulos G, Weinstein B,
et al: Calculation of intraocular lens power
after radial keratotomy with computerized
videokeratography. Am J Ophthalmol 1995;
120:739750.
116. Hjortdal JO, Erdmann L, Bek T: Fourier
analysis of video keratographic data: a tool
for separation of spherical, regular
astigmatic and irregular astigmatic corneal
power components. Ophthalmic Physiol
Opt 1995; 15:171185.
117. Maeda N, Klyce SD, Smolek MK,
McDonald MB: Disparity between
keratometry readings and corneal power
118.
119.
120.
121.
122.
123.
124.
125.
126.
127.
128.
129.
130.
131.
132.
133.
134.
135.
136.
137.
within the pupil after refractive surgery for

myopia. Cornea 1997; 16:517524.
Smolek MK, Klyce SD: Zernike polynomial
fitting fails to represent higher order
aberrations in the eye. Invest Ophthalmol
Vis Sci 2003; 44:46764681.
Wilson SE, Klyce SD: Quantitative
descriptors of corneal topography: a clinical
study. Arch Ophthalmol 1991; 109:349353.
Seiler T, Reckmann W, Maloney RK:
Effective spherical aberration of the cornea
as a quantitative descriptor in corneal
topography. J Cataract Refract Surg 1993;
19:155165.
Klyce SD, Karon MD, Smolek MK: Screening
patients with the corneal navigator.
J Refractive Surg 2005; 21:S617S622.
McDonnell PJ, Garbus J, Lopez PF:
Topographic analysis and visual acuity after
radial keratotomy. Am J Ophthalmol 1988;
106:692695.
McDonald MB, Durrie D, Asbell P, et al:
Treatment of presbyopia with conductive
keratoplasty: six-month results of the
1-year United States FDA clinical trial.
Cornea 2004; 23:661668.
Maeda N, Klyce SD, Smolek MK,
Thompson HW: Automated keratoconus
screening with corneal topography
analysis. Invest Ophthalmol Vis Sci 1994;
35:27492757.
Maeda N, Klyce SD, Hamano H: Alteration
of corneal asphericity in rigid gas
permeable contact lens induced warpage.
CLAO J 1994; 20:2731.
Dingeldein SA, Klyce SD: The topography
of normal corneas. Arch Ophthalmol 1989;
107:512518.
Bogan SJ, Waring GO, Ibrahim O, et al:
Classication of normal corneal topography
based on computer assisted
videokeratography. Arch Ophthalmol 1990;
108:945949.
Kennedy RH, Bourne WM, Dyer JA: A 48
year clinical and epidemiologic study of
keratoconus. Am J Ophthalmol 1986;
101:267273.
Anonymous: Nomenclature for keratoconus
suspects. Refract Corneal Surg 1993;
9:220223.
Anonymous: Opinions. Keratoconus,
videokeratography, and refractive surgery.
Grandon SC, Weber RA: Radial keratotomy
in patients with atypical inferior steepening.
Wilson SE, In DT, Klyce SD: The corneal
topography of keratoconus. Cornea 1991;
10:28.
Lee LR, Hirst LW, Readshaw G: Clinical
detection of unilateral keratoconus. Aust N
Z J Ophthalmol 1995; 23:129133.
Holland D, Maeda N, Hannush SB, et al:
Unilateral keratoconus: incidence and
comparison of quantitative topographic
indices with bilateral keratoconus. Invest
Ophthalmol Vis Sci 1995; 36:S308.
Hubbe RE, Foulks GN: The effect of poor
xation on computer-assisted topographic
corneal analysis: pseudokeratoconus.
Ophthalmology 1994; 101:17451748.
Krachmer JH: Pellucid marginal corneal
96:12171221.
Maguire LJ, Klyce SD, McDonald MB,
Kaufman HE: Corneal topography of
pellucid marginal degeneration.
138. Wilson SE, Lin DTC, Klyce SD, Insler MS:

Terriens marginal degeneration: corneal
topography. Refract Corneal Surg 1990;
6:1520.
139. Rabinowitz YS, McDonnell PJ: Computer
assisted corneal topography in
keratoconus. Refract Corneal Surg 1989;
5:400408.
140. Maeda N, Klyce SD, Smolek MK:
Comparison of methods for detecting
keratoconus using videokeratography. Arch
Ophthalmol 1995; 113:870874.
141. Schwiegerling J, Greivenkamp JE:
Keratoconus detection based on
videokeratoscopic height data. Optom Vis
Sci 1996; 73:721728.
142. Maeda N, Klyce SD, Smolek MK: Neural
network classication of corneal
topography: preliminary demonstration.
36:13271335.
143. Martin RG, Sanders DR, Miller JD, et al:
Effect of cataract wound incision size on
acute changes in corneal topography. J
Cataract Refract Surg 1993;
19(Suppl):170177.
144. Hayashi K, Hayashi H, Nakao F, Hayashi F:
The correlation between incision size and
corneal shape changes in sutureless
cataract surgery. Ophthalmology 1995;
102:550556.
145. Nizzola F, Rossetti L, Torrazza C, et al:
Topographic analysis of induced changes
in corneal astigmatism after
phacoemulsication. Invest Ophthalmol Vis
Sci 1995; 36:S42.
146. Anders N, Pham DT, Hafezi N, Wollensak J:
Influence of the site and the depth of the
incision in the no stitch technique on
wound strength and postoperative
astigmatism in cataract surgery. Invest
147. Kohnen T, Dick B, Jacobi K: Comparison of
the induced astigmatism after temporal
clear corneal tunnel incisions of different
sizes. J Cataract Refract Surg 1995;
21:417424.
148. Kohnen T, Mann PM, Husain SE, et al:
Corneal topographic changes and induced
astigmatism resulting from superior and
temporal scleral pocket incisions.
149. Wong HC, Davis G, Della N: Corneal
astigmatism induced by superior versus
temporal corneal incisions for extracapsular
cataract extraction. Aust N Z J Ophthalmol
1994; 22:237241.
150. Mafra CH, Dave AS, Pilai CT, et al:
Prospective study of corneal topographic
changes produced by extracapsular
cataract surgery. Cornea 1996;
15:196203.
151. Zhang Z, Liu Y, Lin Z, et al: Comparison of
corneal topography and keratometer in
patients with cataract preoperative and
postoperatively. Yan Ke Xue Bzo 1994;
10:8589.
152. Antcliff RJ, Bell J, Flanagan DW:
Comparison of the accuracy of
computerized videokeratography for use in
the SRK II formula for lens calculations. Eur
J Impl Refr Surg 1995; 7:288293.
153. Markovits AS: Extracapsular cataract
extraction with posterior chamber
intraocular lens implantation in a postradial
keratotomy patient. Arch Ophthalmol 1986;
104:329331.
154. Odenthal MTP, van Marle GW, Pameyer JH,
155.
156.
157.
158.
159.
160.
161.
162.
163.
164.
165.
166.
167.
168.
169.
170.
171.
et al: Refractive errors requiring IOL

exchange in cataract surgery after excimer
laser PRK for myopia. Invest Ophthalmol
Vis Sci 1996; 37:S573.
Lesher MP, Schumer DJ, Hunkeler JD, et
al: Phacoemulsication with intraocular lens
implantation after excimer photorefractive
keratectomy: a case report. J Cataract
Refract Surg 1994; 20(Suppl):265267.
Vass C, Menapace R: Computerized
statistical analysis of corneal topography
for the evaluation of changes in corneal
shape after surgery. Am J Ophthalmol
1994; 118:177184.
Hoppenreijs VP, Van Rij G, Beekhuis WH, et
al: Causes of high astigmatism after
penetrating keratoplasty. Doc Ophthalmol
1993; 85:2134.
Wilson SE, Friedman RS, Klyce SD:
Contact lens manipulation of corneal
topography after penetrating keratoplasty:
a preliminary study. CLAO J 1992;
18:177182.
Lin DTC, Wilson SE, Reidy JJ, et al: An
adjustable single running suture technique
to reduce postkeratoplasty astigmatism: a
preliminary report. Ophthalmology 1990;
97:934938.
Lin DTC, Wilson SE, Klyce SD, et al:
Topographic changes that occur with 100
suture removal following penetrating
keratoplasty. Refract Corneal Surg 1990;
6:2125.
Burk LL, Waring GO III, Radjee B, Stulting
RD: The effect of selective suture removal
on astigmatism following penetrating
keratoplasty. Ophthalmic Surg 1988;
19:849854.
Koffler BH, Smith VM: Corneal topography,
arcuate keratotomy, and compression
sutures for astigmatism after penetrating
keratoplasty. J Refract Surg 1996;
12:S306S309.
Wilson SE, Lin DTC, Klyce SD, et al: Rigid
contact lens decentrationa risk factor for
corneal warpage. CLAO J 1990;
16:177182.
Wilson SE, Klyce SD: Screening for corneal
topographic abnormalities before refractive
surgery. Ophthalmology 1994;
101:147152.
Pallikaris IG, Siganos DS: Excimer laser in
situ keratomileusis and photorefractive
keratectomy for correction of high myopia.
J Refract Corneal Surg 1994; 10:498510.
Klyce SD, Smolek MK: Corneal topography
of excimer laser photorefractive
19(Suppl):122130.
Maloney RK: Corneal topography and
optical zone location in PRK. Refract
Corneal Surg 1990; 6:363371.
McDonnell PJ, McClusky DJ, Garbus JJ:
Corneal topography and fluctuating visual
acuity after radial keratotomy.
Maguire LJ, Bourne WM: A multifocal lens
effect as a complication of radial
keratotomy. Refract Corneal Surg 1989;
5:394399.
Wilson SE, Klyce SD, McDonald MB, et al:
Changes in corneal topography after
excimer laser photorefractive keratectomy
for myopia. Ophthalmology 1991;
98:13381347.
Binder PS: Optical problems following
refractive surgery. Ophthalmology 1986;
93:739745.
172. Reidy JJ, McDonald MB, Klyce SD:

Epikeratophakia: corneal topography.
173. Martinez CE, Applegate RA, Howland HC,
et al: Changes in corneal aberration
structure after photorefractive keratectomy.
Invest Ophthalmol Vis Sci 1996; 37:S933.
174. Applegate RA, Howland HC, Buettner J, et
al: Corneal aberrations before and after
radial keratotomy (RK) calculated from
videokeratometric measurements. In vision
science and its applications. Topical
meeting, Santa Fe, The Optical Society of
America; 1994.
175. Applegate RA, Howland HC, Buettner J, et
al: Changes in the aberration structure of
the RK cornea from videokeratographic
measurements. Invest Ophthalmol Vis Sci
1994; 35(Suppl):1740.
176. Applegate RA, Howland HC: Refractive
surgery, optical aberrations, and visual
performance. J Refract Surg 1997;
13:295299.
177. Greivenkamp JE, Schwiegerling J, Miller
JM, Mellinger MD: Visual acuity modeling
using optical raytracing of schematic eyes.
Am J Ophthalmol 1995; 120:227240.
178. Maguire LJ, Zabel RW, Parker P, et al:
Topography and raytracing analysis of
patients with excellent visual acuity 3
Surg 1991; 7:122128.
179. Maguire LJ: Topographical principles in
keratorefractive surgery. Int Ophthalmol
Clin 1991; 31:16.
180. Waring GO III: Refractive keratotomy for
myopia and astigmatism. St Louis: Mosby
Year Book; 1992:887897.
181. Waring GO III, Lynn MJ, Fielding B, et al:
Results of the prospective evaluation of
radial keratotomy (PERK) study four years
after surgery for myopia. JAMA 1990;
263:10831091.
182. Atkin A, Asbell P, Justin N: Radial
keratotomy and glare effects on contrast
sensitivity. Doc Ophthalmol 1986;
62:129148.
183. Gormley DJ, Gersten M, Koplin RS, et al:
Corneal modeling. Cornea 1988; 7:3035.
184. Rowsey JJ, Balyeat HD, Monlux RD, the
PERK Study Group: Prospective evaluation
of radial keratotomy: photokeratoscope
corneal topography. Ophthalmology 1988;
95:322334.
185. Rowsey JJ, Waring GO III, Monlux RD, et
al: Corneal topography as a predictor of
refractive change in the prospective
evaluation of radial keratotomy (PERK)
study. Ophthalmic Surg 1991; 22:370380.
186. Olson RJ, Biddulph MC: Hyperopia,
anisometropia, and irregular astigmatism in
a patient following revisional radial
keratotomy. Ophthalmic Surg 1992;
23:782783.
187. Diamond S: Present status of radial
keratotomy myopia surgery: aerospace
considerations. Aviat Space Environ Med
1990; 61:732734.
188. Steinberg EB, Wilson LA, Waring GO III, et
al: Stellate iron lines in the corneal
epithelium after radial keratotomy. Am J
Ophthalmol 1984; 98:416421.
189. Alio J, Ismail M: Management of radial
keratotomy overcorrections by corneal
sutures. J Cataract Refract Surg 1993;
19:595599.
190. Deitz MR, Sanders DR: Progressive
191.
192.
193.
194.
195.
196.
197.
198.
199.
200.
201.
202.
203.
204.
205.
206.
207.
hyperopia with long term follow-up of radial

keratotomy. Arch Ophthalmol 1985;
103:782784.
Moreira H, Fasano AP, Garbus JJ, et al:
Corneal topographic changes over time
after radial keratotomy. Cornea 1992;
11:465470.
Santos V, Waring GO III, Lynn M, et al:
Morning to evening change in refraction,
corneal curvature, and visual acuity 2 to 4
years after radial keratotomy in the PERK
Wyzinski P, ODell L: Diurnal cycle of
refraction after radial keratotomy.
Maloney RK: Effect of corneal hydration
and intraocular pressure on keratometric
power after experimental radial keratotomy.
Busin M, Arffa RC, McDonald MB,
Kaufman HE: Change in corneal curvature
with elevation of intraocular pressure after
radial keratotomy in the primate eye. CLAO
J 1988; 14:110112.
Simon G, Legeais JM, Parel JM: Radial
keratotomy, corneal hydration and
intraocular pressure: experimental study.
J Fr Ophthalmol 1993; 16:657662.
Ehud A, Renan N, Mordechai S, Isaac A:
The relation between diurnal changes in
intraocular pressure and refractive power
after radial keratotomy. Metab Pediatr Syst
Schanzlin DJ, Santos VR, Waring GO III, et
al: Diurnal change in refraction, corneal
curvature, visual acuity, and intraocular
pressure after radial keratotomy in the
PERK Study. Ophthalmology 1986;
93:167175.
Lucci L, Martinez C, Klyce SD, et al:
Topographic correlates to diurnal variations
in vision in a 10 year PERK cohort. In 1995
pre academy meeting of the ISRS. Atlanta,
GA, October 2628, 1995; 116117.
Waring GO III, Lynn MJ, McDonnell PJ:
radial keratotomy (PERK) study 10 years
after surgery. Arch Ophthalmol 1994;
112:12981308.
Moreira H, Garbus JJ, Lee M, et al:
Multifocal corneal topographic changes
after radial keratotomy. Ophthalmic Surg
1992; 23:8589.
Hemenger RP, Tomlinson A, Caroline PJ:
Role of spherical aberration in contrast
sensitivity loss with radial keratotomy.
30:19972001.
Hemenger RP, Tomlinson A, McDonnell PJ:
Explanation for good visual acuity in
uncorrected residual hyperopia and
presbyopia after radial keratotomy. Invest
Thornton SP: Astigmatic keratotomy: a
review of basic concepts with case reports.
Lindstrom RL, Lindquist TD: Surgical
correction of postoperative astigmatism.
Cornea 1988; 7:138148.
Pallikaris I, Xiras M, Naoumidis L, Siganos
D: Arcuate transverse keratotomy with a
mechanical arcutome based on
videokeratography. J Refract Surg 1996;
12:S296S299.
McDonnell PJ, Caroline PJ, Salz J: Irregular
astigmatism after radial and astigmatic
keratotomy. Am J Ophthalmol 1989;
107:4246.
CHAPTER 40
473
SECTION 6
474
208. Hanna KD, Jouve FE, Waring GO III, Ciarlet

PG: Computer simulation of arcuate
keratotomy for astigmatism. Refract
Corneal Surg 1992; 8:152163.
209. Maguire LJ: Corneal topography of patients
with excellent Snellen visual acuity after
epikeratophakia for aphakia. Am J
Ophthalmol 1990; 109:162167.
210. Camp JJ, Maguire LJ, Cameron BM, Robb
RA: A computer model for the evaluation of
the effect of corneal topography on optical
performance. Am J Ophthalmol 1990;
109:379386.
211. Basuk WL, Zisman M, Waring GO III, et al:
Complications of hexagonal keratotomy.
212. Trokel SL, Srinivasan R, Braren BA:
Excimer laser surgery of the cornea. Am J
Ophthalmol 1983; 96:710715.
213. Aron Rosa DS, Boerner CF, Bath P, et al:
Corneal wound healing after excimer laser
keratotomy in a human eye. Am J
Ophthalmol 1987; 103:454464.
214. Krueger R, Trokel S, Schubert H:
Interaction of ultraviolet laser light with the
26:14551464.
215. Marshall JST, Rothery S, Schubert H: An
ultrastructural study of corneal incisions
induced by an excimer laser at 193 nm.
216. Puliato CA, Steinert RF, Deutsch TF, et al:
Excimer laser ablation of the cornea and
lens: experimental studies. Ophthalmology
1985; 92:741748.
217. LEsperance FA Jr, Taylor DM, Del Pero RA,
et al: Human excimer laser corneal surgery:
preliminary report. Trans Am Ophthalmol
Soc 1988; 86:208275.
218. McDonald MB, Kaufman HE, Frantz JM,
et al: Excimer laser ablation in a human
eye: case report. Arch Ophthalmol 1989;
107:641642.
219. McDonald MB, Frantz JM, Klyce SD, et al:
Central photorefractive keratectomy for
myopia: the blind eye study. Arch
Ophthalmol 1990; 108:799808.
220. McDonald MB, Liu JC, Byrd TJ, et al:
Central photorefractive keratectomy for
myopia. Ophthalmology 1991;
98:13271337.
221. Seiler T, Kahle G, Kriegerowski M: Excimer
laser (193 nm) myopic keratomileusis in
sighted and blind human eyes. Refract
Corneal Surg 1990; 6:165173.
222. Seiler T, Wollensak J: Myopic
photorefractive keratectomy with the
excimer laser: one year follow-up.
223. Dausch D, Smecka Z, Klein R, et al:
Excimer laser photorefractive keratectomy
for hyperopia. J Cataract Refract Surg
1997; 23:169176.
224. Dausch D, Klein R, Schroder E: Excimer
laser photorefractive keratectomy for
hyperopia. J Cataract Refract Surg 1993;
9:2028.
225. Gallinaro C, Toulemont P, Cochener B,
Colin J: Excimer laser photorefractive
keratectomy to correct astigmatism.
226. Dausch D, Klein R, Schroder E, Niemczyk
S: Photorefractive keratectomy of
hyperopic and mixed astigmatism.
227. McDonald MB, Beuerman R, Falzoni W,
et al: Refractive surgery with the excimer
laser. Am J Ophthalmol 1987; 103:469.

One year refractive results of central
photorefractive keratectomy for myopia in
the nonhuman primate cornea. Arch
Ophthalmol 1990; 108:4047.
229. Kremer FB, Dufek M: Excimer laser in situ
11:S244S247.
230. Pallikaris IG, Papatzanaki ME, Stathi EZ, et
al: Laser in situ keratomileusis. Laser Surg
Med 1990; 10:463468.
231. Wilson SE, Chwang EL, Vital M, et al:
Epithelial injury induces keratocyte
apoptosis: hypothesized role for the
interleukin 1 system in the modulation of
corneal tissue organization and wound
healing. Exp Eye Res 1996; 62:325327.
232. Schipper I, Suppelt C, Gebbers JO:
Mitomycin C reduces scar formation after
excimer laser (193 nm) photorefractive
keratectomy in rabbits. Eye 1997;
11:649665.
233. Maguire LJ, Klyce SD, Singer DE, et al:
Corneal topography in myopic patients
undergoing epikeratophakia. Am J
Ophthalmol 1987; 103:404416.
234. Uozato H, Guyton DL: Centering corneal
procedures. Am J Ophthalmol 1987;
103:264275.
235. Maloney RK: Corneal topography and optical
zone location in photorefractive keratectomy.
236. Cavanaugh TB, Durrie DS, Reidel SM, et al:
Topographical analysis of the centration of
excimer laser photorefractive keratectomy.
J Cataract Refract Surg 1993; 19:S136S143.
237. Lin DTC, Sutton HF, Berman M: Corneal
topography following excimer photorefractive
keratectomy for myopia. J Cataract Refract
Surg 1993; 19:S149S154.
238. Amano S, Tanaka S, Shimizu K:
Topographical evaluation of centration of
excimer laser myopic photorefractive
20:616619.
239. Parker PJ, Klyce SD, Ryan BL, et al:
Central topographic islands following
photorefractive keratectomy. Invest
240. Oshika T, Klyce SD, Smolek MK, McDonald
MB: Corneal hydration and central islands
after excimer laser photorefractive
24:15751580.
241. Lohman CP, Sachs H, Marshall J, Gabel
VP: Excimer laser phototherapeutic
keratectomy for recurrent erosions: a
clinical study. Ophthalmic Surg Lasers
1996; 27:768772.
242. Hersh P, Burnsten Y, Carr J, et al: Excimer
laser phototherapeutic keratectomy:
surgical strategies and clinical outcomes.
Ophthalmology 1996; 103:12101222.
243. Barraquer JI: Keratomileusis for myopia
and aphakia. Ophthalmology 1981;
88:709715.
244. Swinger CA, Barker BA: Prospective
evaluation of myopic keratomileusis.
245. Brint SF, Ostrick DM, Fisher C, et al: Six
month results of the multicenter phase I
study of excimer laser myopic
keratomileusis. J Cataract Refract Surg
1994; 20:610615.
246. Salah T, Waring GO III, Maghraby AE, et al:
Excimer laser in situ keratomileusis under a
corneal flap for myopia of 2 to 20 diopters.
Am J Ophthalmol 1996; 121:143156.
247. Martinez CE, Klyce SD, Waring GO III,

Maghraby AE: The topography of LASIK.
In: Pallikaris I, Siganos DS, eds. LASIK.
Thorofare, NJ: Slack 1997; 339357.
248. Fleming JF, Reynolds AE, Kilmer L, et al:
The intrastromal corneal ring: two cases in
rabbits. J Refract Surg 1987; 3:227240.
249. Assil KK, Barrett AM, Fouraker BD,
Schanzlin DJ: One year results of the
intrastromal corneal ring in nonfunctional
human eyes. Intrastromal Corneal Ring
Study Group. Arch Ophthalmol 1995;
113:159167.
250. Calossi A, Zanella SG, Verzella F: Corneal
warpage resolution after retting an RGP
contact lens wearer into hydrophilic high
water content material. CLAO J 1996;
22:242244.
251. Novo AG, Feldman ST, Pavlopoulos G:
Corneal topographic changes after retting
polymethylmethacrylate contact lens
wearers into rigid gas permeable materials.
CLAO J 1995; 21:4751.
252. Ruiz Montenegro J, Mafra CH, Wilson SE,
et al: Corneal topographic alterations in
normal contact lens wearers.
253. Polse KA, Brand RJ, Schwalbe JS, et al:
The Berkeley Orthokeratology Study. II.
Efcacy and duration. Am J Optom Physiol
Optics 1983; 60:187198.
254. Polse KA, Brand RJ, Keener RJ, et al: The
Berkeley Orthokeratology Study. III. Safety.
Am J Optom Physiol Optics 1983;
60:321328.
255. Swarbrick HA: Orthokeratology review and
update. Clin Exp Optom 2006; 89:124143.
256. Heng LS, Khoo CY: Can contact lenses
control the progression of myopia?
Singapore Med J 1994; 35:367370.
257. Polse KA, Brand RJ, Vastine DW, Schwalbe
JS: Corneal change accompanying
orthokeratology: plastic or elastic? Results
of a randomized controlled clinical trial.
258. Gomes JA, Cohen EJ, Rapuano CJ:
Topographic stability and safety of contact
lens use after penetrating keratoplasty.
CLAO J 1996; 22:6469.
259. Sperber LT, Cohen EJ, Lopatynsky MO:
Corneal topography in contact lens wearers
following penetrating keratoplasty. CLAO J
1995; 21:183190.
260. Klyce SD, Estopinal HA, Gersten M, et al:
Fluorescein exam simulation for contact
lens tting. Invest Ophthalmol Vis Sci 1992;
33:S697.
261. Szczotka LB, Lass JH, Capretta DM:
Clinical evaluation of a computerized
topography software method for tting rigid
gas permeable contact lenses. CLAO J
1994; 20:231236.
262. Donshik PC, Luistro AE, Reisner DS: The
use of computerized videokeratography as
an aid in tting rigid gas permeable contact
lenses. Trans Am Ophthalmol Soc 1996;
94:135143.
263. Szczotka LB: Clinical evaluation of a
topographically based contact lens tting
software. Optom Vis Sci 1997; 74:1419.
264. Srivannaboon S, McDonald MB, Doubrava
M, Klyce SD: A prospective clinical trial
comparing a topographically guided
articial intelligence software system
versus clinical expertise for tting normal
and pathological corneas with contact
lenses. Invest Ophthalmol Vis Sci 1997;
38:S1089.
CHAPTER
41
Ocular Surface Epithelial Stem Cells and Corneal

Wound Healing Response to Injury and Infection
Leonard P. K. Ang and Dimitri T. Azar
INTRODUCTION
The ocular surface is a complex biological continuum responsible for the protection of the cornea and maintenance of
corneal clarity. The precorneal tear lm, neural innervation and
the protective blink reflex help sustain an environment
favorable for the epithelial cell layers. The ocular surface,
comprising corneal, limbal and conjunctival epithelia, is selfrenewing. Ocular surface stem cells are responsible for the
maintenance and regeneration of the ocular surface epithelium.
These play an important role in the wound healing process, and
in the regeneration of the epithelium following injury from
accidental trauma, surgery or infection.
STEM CELLS
Stem cells are present in all self-renewing tissues of the body.
Stem cells are a small, quiescent subpopulation of cells within
a given tissue.1,2 Stem cells are highly proliferative and selfrenewing, and are responsible for the continued replacement
and regeneration of tissues, thereby maintaining a steady-state
population of healthy cells within tissues during the lifespan
of the organism. At steady state, stem cells remain fairly
dormant and replicate infrequently, but when the need for
tissue regeneration arises, proliferation may be rapidly induced.
Relative dormancy minimizes the possibility of replication
errors during cell division, which can result in mutations.
Stem cells give rise to transient amplifying cells that proliferate
rapidly, ensuring prompt regeneration of the tissue
(Fig. 41.1).35 Transient amplifying cells in turn give rise to
postmitotic cells, and nally to terminally differentiated cells.
These progenitor cells have a long life span, potentially exceeding that of the organism, and show little evidence of aging.
These properties make stem cells also more prone to developing
neoplastic lesions.
Adult corneal and conjunctival stem cells represent the
earliest progenitor cells responsible for the homeostasis and
regeneration of the ocular surface. An intricate balance of
intrinsic and extrinsic factors modulates stem-cell proliferation
and differentiation, eventually resulting to terminally
differentiated cells that bear the phenotypic characteristics of
the tissue.
corneal epithelium could be replenished from the adjacent

conjunctival epithelium.68 Conjunctival transdifferentiation,
in which conjunctival epithelial cells differentiate into a corneal
epithelial cell phenotype, was proposed as a mechanism to
explain replenishment of the corneal epithelium.68 Subsequent
studies have argued against the concept of conjunctival
transdifferentiation, as conjunctival tissue rarely resulted in
complete corneal epithelial replacement.920
The idea that limbal epithelial cells are involved in
regeneration of epithelial cells of the cornea was proposed by
Davanger and Evensen in 1971.21 In heavily pigmented
eyes, they observed pigmented epithelial lines migrating from
the limbal region to the central cornea during healing of corneal epithelial defects. Limbal basal epithelial cells appeared to
be the least differentiated cells of the corneal epithelium.
Schermer et al found a 64 kDa keratin, called K3, among
SC
Self renewal
TAC3
TAC2
TAC3
PMC
LIMBAL STEM CELLS

TDC
Differentiated cells located supercially in the corneal

epithelium are constantly lost, and are replaced by basal cells
entering the differentiation pathway.35 Previous reports
suggested that conjunctival and corneal epithelial cells arose
from a common progenitor cell type, and that depletion of the
FIGURE 41.1. Schematic diagram showing hierarchy of stem cell

(SC), transient amplifying cell (TAC1, TAC2, and TAC3 ), postmitotic cell
(PMC), and terminally differentiated cell (TDC). A self-renewal process,
possibly by asymmetric division, maintains the stem cell population.
475

compounds compared to cells of the peripheral or central cornea
(Fig. 41.3).13,16,28 The limbal basal cells proliferative response
was maintained over a prolonged period, demonstrating a
signicantly greater proliferative reserve than cells in the central
corneal epithelium. The label-retaining cells present in the
limbus exhibited properties expected of stem cells.
Exactly how a population of stem cells is maintained is
unclear. A stem cell may divide symmetrically, giving rise to a
transient amplifying cell and producing a daughter stem cell
which replenishes the stem cell pool. Alternatively, regeneration
of stem cells could occur by de-differentiation of early transient
amplifying cells back to stem cells.
Stem cells have the highest growth potential under in vitro
cell culture conditions, and regions enriched in stem cells
display greater colony-forming ability. They can continue to
divide in vitro for at least 120160 generations.29,30 Limbal
epithelial cells display greater in vitro proliferative capacity than
central and peripheral corneal cells, consistent with the
presence of stem cells in the limbus.3141 Culture conditions
in vitro do not entirely mimic the original microenvironment of
these cells, as indicated by their eventual senescence. Therefore,
the true proliferative reserve of stem cells relative to the lifespan
of the organism is impossible to determine at present.
Clinical evidence also supports the limbal region to be the
site of corneal stem cells.36,4245 Destruction of the limbal
epithelium by physical or chemical insult induces a stem celldecient state. Clinical features of limbal stem cell deciency
include abnormal wound healing with persistent or recurrent
epithelial defects, conjunctivalization (conjunctival epithelial
ingrowth), vascularization, loss of corneal clarity and chronic
inflammation. Additionally, the limbus is the most common
site of ocular surface neoplasias. They likely arise from altered
growth behavior of undifferentiated progenitor cells, suggesting
that a corneal intra-epithelial neoplasm is essentially a stemcell tumor.
Transient amplifying cells play an important role in wound
healing. When slow-cycling limbal stem cells are activated by a
demand for tissue regeneration, such as wounding, they give
rise to daughter transient amplifying cells that migrate centrally
or supercially to replenish the population of corneal epithelial
cells.5 Transient amplifying cells have shorter cell cycle times,
resulting in rapid cell division, and have a limited proliferative
capacity. They probably undergo a predetermined number of
cell divisions before differentiating into postmitotic cells which
in turn terminally differentiate and replenish the diminished
epithelial cell population.
A hierarchy of cells extends from the limbus to the central
cornea. Early transient amplifying cells, located adjacent to
limbal stem cells, have a greater proliferative capacity than later
transiently amplifying cells which are migrating from the
periphery toward the center of the cornea. Cells in the central
cornea are mainly postmitotic cells with no capacity for cell
division. These ndings are consistent with growth responses
in vitro, where limbal and peripheral corneal cells generate large
colonies and are easily serially cultivated, whereas central corneal
cells are less clonogenic, and cannot be subcultured more than
once.31,32,34,36
FIGURE 41.2. Schematic diagram showing the location of corneal

epithelial stem cells (SC) in the basal layer of the limbus. Solid arrows
denote the centripetal (horizontal) migration of limbal-derived TA cells
which progresseively lose their proliferative potential; dashed arrows
denote the (vertical) migration of cells into the suprabasal
compartment to become terminally differentiated (TD). C: cornea;
CC: central cornea; Cj: conjunctiva; DM: Descemet membrane;
L: limbus; PC: peripheral cornea.
UNIQUE PROPERTIES OF THE LIMBUS
SECTION 6
differentiated corneal epithelial cells.22 This cornea-specic

keratin was expressed in differentiated cells in the suprabasal
limbal layer, and throughout the corneal epithelium, but was
essentially absent in limbal basal cells, suggesting that they
represented a more primitive, nondifferentiated subpopulation
that did not express this cytokeratin. Kurpakus et al demonstrated that the cornea-specic keratin K12, expressed in the
suprabasal cells of the limbus and throughout the entire corneal
epithelium, was also absent from the limbal basal cells.23,24
They also demonstrated that stem or stem-like cells found
throughout the basal layer or the limbal and corneal epithelium
during embryonic development were later sequestered in the
limbus.2325
Current evidence indicates that corneal epithelial cells arise
from specic progenitor cells located in the basal cell layer
of the limbus (Fig. 41.2).3,5,1320 Upon a demand for tissue
regeneration, for example, following injury, limbal stem cells
are stimulated to divide and differentiate into transient
amplifying cells.3,5 These transient amplifying cells migrate
supercially to the suprabasal limbus, as well as centrally to
form the basal layer of the corneal epithelium. Transient
amplifying cells increase rapidly in number to replace injured or
dead cells within the tissue. These cells differentiate into
postmitotic cells, which in turn differentiate further into
terminally differentiated cells. These eventually migrate
supercially and take on the nal phenotypic characteristics of
the tissue. As their names imply, postmitotic and terminally
differentiated cells are incapable of cell division.
The observation that slow-cycling cells were restricted to a
subset of limbal epithelial basal cells provided strong support for
the limbal stem cell hypothesis.2628 One of the most reliable
ways to identify epithelial stem cells takes advantage of their
slow turnover or slow cycling nature, which can be identied
experimentally as label retaining cells.26,27 Continuous
administration of tritiated thymidine for a prolonged period
labels replicating DNA in all cells that undergo cell division,
including slow-cycling cells. During a prolonged chase period in
the absence of tritiated thymidine, radioactive label in the
DNA of rapidly dividing cells is diluted by incorporation of
nonradioactive thymidine. Slow-cycling cells, presumably stem
cells, retain most of the previously incorporated isotope after
a 48 week chase period.26,27 Using this technique, Cotsarelis
et al observed retention of tritiated thymidine in limbal basal
cells, suggesting that these may represent corneal stem cells.28
This small subpopulation of normally slow-cycling limbal
basal epithelial cells demonstrated a greater proliferative
response to wounding and to stimulation by tumor promoting
476
Adapted from: Lavker RM, Sun TT: Epithelial stem cells: the eye provides a
vision. Eye 2003; 17:937-942. Figure 1.
Since the corneal epithelium must provide a transparent

medium for vision, it is devoid of pigmentation, and has a
smooth stromalepithelial junctional structure. As such,
corneal epithelial cells are vulnerable to shearing injury because
of their poor adhesion to the underlying stroma, as evident in
patients with recurrent corneal erosions following relatively
minor corneal injuries.
Ocular Surface Epithelial Stem Cells and Corneal Wound Healing Response to Injury and Infection
FIGURE 41.3. Autoradiograms of corneal (a, c,

and e) and limbal (b, d, and f) epithelia that
have been exposed to a single (c and d) or a
2-day treatment (e and f) of phorbol ester. The
response of corneal and limbal epithelia to
petrolatum treatment is shown in a and b. Note
the low level of [3H]thymidine ([3H]TdR)
incorporation (arrow) in unperturbed corneal
epithelium (a) and limbal epithelium (b). A single
exposure of phorbol myristate (TPA) results in
marked increases in [3H]TdR incorporation in
corneal (c) and limbal epithelia (d). After 2 days
of TPA treatment, both regions show a decrease
in [3H]TdR incorporation.
Stem cells in the body are usually located in deeper tissue

layers, presumably for protection. The anatomical structure of
the limbus is signicantly different from the adjacent cornea
because it need not be transparent. It is well suited to harbor
and protect the corneal stem-cell population. The limbal
epithelium is 810 cell layers thick, compared with ve layers
in the corneal epithelium. The limbus tends to be heavily
pigmented, especially in pigmented races; this may protect
basal cells from the carcinogenic effects of ultraviolet
radiation.28,46 In addition, the palisades of Vogt have an
undulating epithelialstromal junction, which provides greater
adhesion properties, thereby rendering the limbal epithelium
resistant to shearing forces. These folds also greatly increase the
surface area of the basal cells. The stromal component of the
limbus is well innervated, and is supplied by a rich vascular
network, allowing regulation of limbal stem-cell growth and
proliferation through various cytokine- and neural-mediated
pathways. An appropriate stromal micro-environment (stem
cell niche) is important for correctly regulating stem cell
activity.
stromal environment is rarely complete.10,48 Conjunctival

epithelium transplanted onto the cornea of limbal stem celldecient patients retained many characteristics of conjunctival
tissue, such as its glycogen content and goblet cells. 49,50
The evaluation of cytokeratin expression under identical cell
culture conditions provided direct evidence for separate lineages
of conjunctival and corneal cells.47 Conjunctival epithelial cells
expressed K4 and K13 cytokeratins, whereas corneal epithelial
cells expressed K3 and K12.36,47 Conjunctival and corneal epithelial cell suspensions were injected subcutaneously into the
flanks of athymic mice.37,47 Cysts resulting from injection of
limbal and corneal epithelial cells retained features of normal
corneal epithelium, a stratied squamous epithelium without
goblet cells, whereas cysts derived from conjunctival epithelial
cells displayed normal conjunctival morphology, a stratied
epithelium interspersed with numerous goblet cells. Current
evidence suggests that conjunctival epithelial stem cells are
bipotent, and can give rise to both nongoblet epithelial cells and
goblet cells.35,37,47
CONJUNCTIVAL STEM CELLS
LOCATION OF CONJUNCTIVAL STEM

CELLS
The conjunctival epithelium extends from the corneal limbus

to the lid margin, where it gradually merges with the
keratinized, stratied squamous epithelium of the eyelid. The
conjunctival epithelium provides a mechanical and immunological barrier to injury and infection, and its numerous mucinsecreting goblet cells contribute to the production and stability
of the tear lm.
Corneal and conjunctival epithelia are now believed to arise
from different stem-cell populations.47 Evidence shows that
transdifferentiation of conjunctival epithelial cells in a corneal
Conjunctival stem cells are likely to be scattered throughout the

various regions of the conjunctiva (i.e., bulbar, forniceal and
palpebral), although the forniceal conjunctiva appears to be a
site that is enriched in conjunctival stem cells.35,36,49 A greater
number of slow-cycling cells (a property of stem cells) were
found in the forniceal epithelium compared with the bulbar and
palpebral conjunctiva (Fig. 41.4).49 Forniceal basal cells also
displayed a greater and more sustained proliferative response
than cells from other regions following injury or stimulation
with a tumor promoting compounds (Fig. 41.5). Further
CHAPTER 41
Adapted from: Lavker RM, Wei ZG, Sun TT: Phorbol

ester preferentially stimulates mouse fornical
conjunctival and limbal epithelial cells to proliferate in
vivo. Invest Ophthalmol Vis Sci 1998; 39:301-307.
Figure 2.
477
SECTION 6
FIGURE 41.4. Schematic diagram showing the relative densities of

label-retaining cells in the palpebral, forniceal and bulbar conjunctiva
in the mice model. The highest concentration is noted in the forniceal
conjunctiva, which is believed to be the site enriched in conjunctival
stem cells. E: epidermis; T: transitional zone between palpebral
conjunctiva and epidermis (muco-cutaneous junction); P: palpebral
conjunctiva; F: fornix conjunctiva; B: bulbar conjunctiva; L: limbus;
C: cornea.
evidence was provided by in vitro studies which showed that

forniceal conjunctival cells had greater proliferative capacities
compared to the other regions.35,36 The mucocutaneous
junction at the lid margin might also be a site enriched in
conjunctival stem cells, which may be important for the
replacement of the palpebral and forniceal conjunctival
epithelium.51
The fornix is well suited to house and protect conjunctival
stem-cell populations from extrinsic injury, as it is located well
within the upper and lower recesses created by the closely
apposed eyelid and globe, and further from the external
environment than the other conjunctival regions. The network
of collagen and elastic bers in the stroma protects the
epithelial cells from shearing and mechanical forces. The fornix
is also the most richly vascularized and innervated region of the
conjunctiva, allowing prompt response to cytokine or neural
stimuli.
EPITHELIALSTROMAL INTERACTIONS
AND THE STEM-CELL
MICROENVIRONMENT
Both intrinsic factors (inherent to the cell), and extrinsic factors
(environmental factors surrounding the cell) are thought to be
involved in the regulation of stem cells.52,53 Schoeld proposed
FIGURE 41.5. Autoradiograms of the response of the bulbar (a, d, and g), fornical (b, e, and h), and palpebral (c, f, and i) epithelia to a single
exposure (d, e, and f) and a 2-day exposure (g, h, and i) of phorbol ester. Response of bulbar, fornical, and palpebral epithelia to petrolatum
treatment is shown in a,b and c. Note the low level of [3H]thymidine ([3H]TdR) incorporation (arrows) in unperturbed fornical epithelium compared
with bulbar and palpebral epithelia (a, b, and c). A single exposure of phorbol myristate (TPA) (d, e, and f) results in marked increases in [3H]TdR
incorporation in all three conjunctival epithelia, most notably in the fornical epithelium (e). Note the marked decrease in [3H]TdR incorporation in
bulbar and palpebral epithelia after 2 days of TPA treatment (g and i), whereas the fornical epithelia (h) has a higher proliferative prole.
478
Adapted from: Lavker RM, Wei ZG, Sun TT: Phorbol ester preferentially stimulates mouse fornical conjunc-tival and limbal epithelial cells to proliferate in vivo. Invest
Ophthalmol Vis Sci 1998; 39:301-307. Figure 1.
IDENTIFICATION OF EPITHELIAL STEM

CELLS
One of the most reliable ways to identify epithelial stem cells
takes advantage of their slow turnover or slow cycling nature,
which can be identied experimentally as label retaining
cells.26,27 This may be determined by a continuous administration of tritiated thymidine, followed by a prolonged chase
period, and identifying the slow-cycling label-retaining cells that
retained the previously incorporated isotope. The in vitro
proliferative capacity of cells has also been used to distinguish
stem cells from other cells.29,30 Three types of keratinocytes
with different capacities for proliferation have been identied
from the human epidermis: holoclones, meroclones, and
paraclones. The holoclone, which has the highest proliferative
capacity and is able to undergo 120160 divisions with less
than 5% terminally differentiated colonies, is considered a stem
cell. Although no denite stem-cell marker currently exists,
various putative markers for limbal stem cells have been
proposed. These include the nuclear protein p63,64 alphaenolase,14,65,66 high levels of a6-integrin in combination with
low to undetectable expression of transferrin receptor (CD71),67
the absence of connexins 43 and 50,68 and more recently, the
ABCG2 transporter.69 Slow-cycling label-retaining cells in the
mouse cornea limbus were also found to be enriched in cells
that expressed high levels of b1 and b4 integrins and little a9
integrin.70,71
FIGURE 41.6. X, Y, and Z hypothesis of corneal epithelial

maintenance
X= proliferation of basal cells
Y= centripetal movement of cells
Z=cell loss from the surface
From: Thoft RA, Friend J: The X, Y, Z hypothesis of corneal epithelial
maintenance. Invest Ophthalmol Vis Sci 1983; 24:1442-1443.
India ink particles phagocytosed by basal cells of normal

corneas migrated centripetally from the limbal region to the
central area, at ~123 mm/week.73 The limbal basal cells, the site
of corneal stem cells, give rise to basal cells that migrate onto
the cornea, constantly renewing the supply of basal cells. These
cells, which do not originally express the 64 KD keratin, slowly
migrate across the corneal basement membranes and upward,
and begin to express the 64 KD keratin.
Thoft rst proposed the X, Y, Z hypothesis of corneal
epithelial maintenance (Fig. 41.6).72 He suggested that the
maintenance of the corneal epithelium could be viewed as a
result of three separate, independent mechanisms. X represented the proliferation of basal epithelial cells, Y represented
the proliferation and centripetal migration of peripheral cells,
and Z referred to the epithelial cell loss from the surface.
Corneal epithelial maintenance, which involved a balance of
these processes, was dened by the equation: X+Y=Z. It is
estimated that the corneal epithelium is constantly renewed
every 710 days. Following corneal injury with resultant epithelial cell loss, the regenerative mechanisms designed to
replace the corneal epithelium are set into motion, with
resultant centripetal movement of the cells from the periphery
to the central area. Other investigators have also demonstrated
this migration of epithelial cells from the peripheral cornea and
limbus.33,7375 The corneal epithelium is therefore maintained
by a balance of cell shedding, basal cell division and renewal of
basal cells by centripetal migration of new basal cells from the
limbal stem cells.
CORNEAL WOUND HEALING RESPONSE

The primary function of the corneal epithelium is to form a
barrier to invasion of the eye by pathogens and for uptake of
excess fluid from the stroma. Injury to the cornea may be
accidental or iatrogenic in origin. Various surgical procedures
may result in corneal wounds or abrasions. Excimer laser
refractive surgery is another important cause of iatrogenically
induced corneal wounds. The process of wound healing
involves a complex cascade of events that eventually results in
wound repair and reestablishment of the normal structure and
function of the cornea.
MAINTENANCE OF THE CORNEAL

EPITHELIUM
EPITHELIAL WOUND HEALING
The corneal epithelium is maintained by a constant cycle of

shedding of supercial cells, proliferation of cells in the basal
layer, as well as the slow migration of basal cells toward the
centre of the cornea.72 It has previously been demonstrated that
With the advent of refractive surgical procedures, such as

photorefractive keratectomy (PRK) and laser in situ
keratomileusis (LASIK), there has been strong interest in the
study of healing of corneal epithelial wounds. Accidental injury
CHAPTER 41
that stem cells existed in a microenvironment that helped

maintain their undifferentiated state.53 Limbal basal cells
express higher levels of epidermal growth factor receptor (EGFR)
levels compared to the more differentiated cells of the central
cornea, which may serve to allow these cells to respond more
rapidly to various growth factors during development and
following perturbations, such as wounding.54
Limbal basal cells have been found to express intermediate
laments, cytokeratin 19, vimentin, a6b4-integrin, metallothionein, transferrin receptor, and a protein bound by monoclonal antibody AE1.5557 Intermediate laments are involved in
maintenance of cell cytoarchitecture, and may play a role in
anchorage of these cells to the underlying tissues. This expression prole is unique to limbal basal cells, and differs from that
of the surrounding basal cells. In addition, limbal basal cells
express higher concentrations of metabolic enzymes, such as
NaKATPase, cytochrome oxidase, and carbonic anhydrase,
reflecting the different physiologic properties of these cells.3,58,59
Stromalepithelial interactions are believed to be extremely
important in supporting normal corneal function, and
regulating the limbal stem-cell population. Intercellular
communications between the corneal stromal and epithelial
cells that are critical during early development, homeostasis,
and wound healing, are mediated by a variety of cytokines and
growth factors, such as transforming growth factor-b (TGF-b),
platelet-derived growth factor B (PDGF-B) and interleukin-1
(IL-1).6062 Hepatocyte growth factor (HGF), expressed by
corneal broblasts, and keratinocyte growth factor (KGF),
expressed mostly by limbal broblasts, play important roles in
the regulation of proliferation, motility and differentiation
during epithelial stem-cell division in wound healing. 6163
479
FIGURE 41.7. Scanning electron micrograph of corneal epithelial cells

migrating to cover an epithelial abrasion.
SECTION 6
From Pster RR: The healing of corneal epithelial abrasions in the rabbit: a
scanning electron microscope study. Invest Ophthalmol Vis Sci 1975;
14:648.
480
or abrasion of the corneal epithelium results in a prompt

healing response to cover the exposed basement membrane
with cells.
After abrasion, mitosis ceases and the cells at the wound edge
retract, and lose their hemidesmosomal attachments to the
basement membrane. During the rst 46 h after an epithelial
injury, there is an initial latent phase where no appreciable
decrease in size of the wound occurs. The basal and squamous
cells in the vicinity of the wound show thickening and
separation. Neutrophils accumulate along the wound edge ~3 h
after injury, as does thinning of the epithelium to a single layer
of flattened cells.76 The leading edge of the migrating cells is
only one cell thick. The cells enlarge, and the epithelial sheet
begins to migrate by ameboid movement across the defect until
it is completely covered. The edges of the cell membranes ruffle
and send out lopodia and lamellipodia toward the center of
the wound (Fig. 41.7).77
Corneal epithelial defects, irrespective of the nature of injury,
result in a fairly consistent pattern of re-epithelization.36,76
A circumferential migration of three to six convex leading fronts
of migrating epithelial sheets from the limbus continue to
advance and progress towards the center.76 The advancing
fronts of epithelium eventually meet and merge imperceptibly
to repopulate the entire surface.74 The wound is covered by a
multilayered sheet made up of both basal and squamous
cells.78,79 After wound closure, mitosis restores the epithelium
to its normal conguration (Fig. 41.8). The healing process
occurs rapidly. An experimental epithelial wound 6 mm in
diameter is closed within 48 h, and the rate of epithelial cell
migration is 6080 mm/h.80,81
The basal epithelial cells play a key role in proliferating and
covering the epithelial defect. The ultimate source of these cells
arise from the limbal basal stem cells that are activated to help
regenerate and repopulate the surface.6 Lavker et al suggested
the mechanism of centripetal migration was the inward draw-
FIGURE 41.8. Corneal wound healing following a 4 mm diameter

abrasion. Cross-sectional view of the wound margin, and the
microphotographic appearance of the defect at the indicated times
after wounding.
Adapted from: Beuerman RW, Thompson HW: Molecular and cellular responses
of the corneal epithelium to wound healing. Acta Ophthalmol Suppl 1992; 712.
Figure 1.
ing of cells by preferential desquamation of central corneal

epithelial cells, rather than the cells forcing their way toward
the center.46 It is interesting to note that the healing rates for
larger (8 mm diameter) corneal epithelial defects were more
rapid (mean rate 0.91 mm2/h) than for smaller (4 mm diameter)
defects (mean rate 0.37 mm2/h). This is attributed to a greater
proliferative response of cells in the peripheral cornea and
limbus than in the central corneal.81 The histological appearance of regenerated limbal epithelium resembles corneal and
not conjunctival epithelium.82
An important aspect of wound healing is the reformation of
adhesion complexes to the underlying connective tissue.
Wounding of the epithelium results in disassembly of the
hemidesmosomes of the remaining epithelial cells, which
allows these cells to migrate over the wounded surface. The
leading edge of the migratory cells form focal linkages from
cytoplasmic actin laments to extracellular matrix proteins like
bronectin, brinogenbrin, laminin, tenascin and integrins.83
Reformation of the adhesion complexes gradually occurs from
STROMAL WOUND HEALING

Keratocytes are responsible for the maintenance and regeneration of the corneal stroma. After injury, keratocytes are capable
of phagocytosis of collagen brils and synthesis and secretion
of collagen, glycosaminoglycan ground substance, collagenase,
and collagenase inhibitors.8789 Stromal wound healing involves
resynthesis and cross-linking of collagen, alterations in
proteoglycan synthesis, and gradual wound remodeling, leading
to restoration of tensile strength.
A cascade of responses of cytokines leads to important
changes in the stroma that contribute to wound healing. Within
hours, polymorphonuclear cells appear around areas of cellular
necrosis in a corneal wound, followed thereafter by monocytes.
Immediately following injury, initial keratocyte apoptosis and
necrosis occurs. Within 12 h, proliferation and migration of
residual activated keratocytes occurs. The proliferating keratocytes are believed to give rise to activated keratocytes,
broblasts, and myobroblasts that repopulated the depleted
stroma.90 Fibroblasts and myobroblasts have the ability to
establish and maintain intercellular communication with
themselves and nonactivated keratocytes, which may be critical
in the wound healing process.91 Stromal keratocytes lose their
interconnections and undergo morphologic changes, including
hypertrophy, proliferation, and nally reformation of cellular
processes and connecting gap junctions.92 TGF-b has been
found to signicantly reduce corneal brosis. These early
changes contribute to other responses associated with stromal
remodeling, epithelial healing, production of altered extracellular matrix and wound contraction.93
Corneal wound healing is a complex cascade mediated by
cytokines, growth factors, and chemokines. These complex
functions may be modulated by cytokines from the epithelium,
inflammatory cells and other keratocytes.94,95 The healing
process is initiated immediately after injury through the
release of multiple cytokines and growth factors, such as IL-1,
tumor necrosis factor-a (TNF-a), bone morphogenic proteins
2 and 4 (BMB), epidermal growth factor (EGF), and PDGF
from the corneal epithelium and epithelial basement
membrane.9698
The early phases of wound healing also involve degradation
and removal of damaged tissue orchestrated by the plasminogen-activator/plasmin system, collagenolytic metalloproteinases and other enzymes.99 Activation of the plasmin
plasminogen system is needed for the progression of normal
healing.100 An increase in polymorphonuclear leukocytes in the
cornea often coincides with enhanced production of matrix
metalloproteinases (MMPs). MMPs may be involved in normal
epithelial migration, the initial stromal degradation during the
inflammatory response, and the ultimate remodeling of the
extracellular matrix.
MMPs are responsible for the initial rate-limiting cleavage of

collagen molecules, and changes in expression of these
collagenolytic/gelatinolytic enzymes occur in healing or
ulcerating corneal wounds.101,102 Following corneal wounding,
MMP-2 expression is increased and much of it appears in the
active form. These changes persist for at least 7 months,
suggesting that MMP-2 is involved in the prolonged process of
collagen remodeling in the stromal repair tissue. MMP-9 is
expressed in the epithelial layer of the repair tissue and is
believed to be involved in the degradation of the epithelial
basement membranes that precedes corneal ulceration, as well
as in controlling resynthesis of the basement membrane.102 It is
possible that these proteolytic enzymes may play a role in the
short-term and long-term stromal remodeling in the normal
cornea. The MMP/tissue inhibitor of metalloproteinase (TIMP)
systems may play an important role in the early stages of
corneal wound healing as well as in scar formation and clearing
after excimer laser keratectomy.103,104
Among the many mediators involved in regulating wound
healing, IL-1a appears to play a special role in orchestrating
wound healing by modulating many key processes involved
in stromal healing after its release triggered by epithelial cell
injury or death.105
Return of normal structure and function may take months,
or even years, in some eyes, depending on the nature of injury
or surgery.
WOUND HEALING RESPONSE TO

CORNEAL INFECTION
RISK FACTORS FOR CORNEAL INFECTION
Because an intact corneal epithelial surface, with its tight
junctions formed by desmosomes and hemidesmosomes, is the
main line of defense against microbial infection, an important
complication arising from a breech in the integrity of the
corneal epithelium is infectious keratitis. There are several
other mechanisms that protect the surface of the eye from
infectious agents. The eyelid provides a physical barrier to
protect against organisms gaining direct access to the eye. The
tear lm contains antimicrobial enzymes, secretory immunoglobulins and complement components, such as lysozyme,
lactoferrin, and betalysins. The normal ocular flora provides a
balance to prevent overgrowth of exogenous organisms. The
conjunctiva contains subepithelial mucosal-associated lymphoid tissue with a collection of lymphoid cells. These factors
serve to protect the ocular surface against infection.
Any alteration of the local or systemic defense mechanism
may predispose the eye to infection. Disruption of the corneal
epithelium may be caused by trauma, contact lens wear or
from chronic bullous keratopathy, which creates a portal of
entry for microbial organisms. Other predisposing factors
include eyelid abnormalities (e.g., trichiasis, entropion, ectropian or lagophthalmos), tear-lm abnormalities (e.g., Sjgrens
syndrome), exposure keratopathy, neuropathic keratopathy,
ocular surface disorders (e.g., StevensJohnson syndrome,
chemical injury), and chronic steroid use. Systemic conditions
that may predispose to corneal infection include diabetes
mellitus, and systemic immunodeciency.
CHAPTER 41
the periphery toward the center.84 After wound healing, the

adhesion of the epithelium is re-established by formation of
new hemidesmosomes in the basal cell layer. The location of
these hemidesmosomes corresponds precisely to the locations
of anchoring brils in the basement membrane.
In corneal wounds where the basement membrane is not
damaged, a normal epithelium with adhesion complexes is
formed soon after. In the situation when the basement
membrane is removed, the epithelium must lay down new
basement membrane after healing and development of normal
adhesion complexes may be delayed for more than 12
months.85,86 This is particularly relevant in excimer laser
procedures such as PRK, where destruction of the basement
membrane and supercial stroma occurs, which results in
delayed corneal healing.
CORNEAL WOUND HEALING FOLLOWING

INFECTION
In bacterial keratitis, entry of organisms results in diffusion of
toxins and enzymes. Polymorphonuclear leukocytes arrive at
the corneal wound site. Stromal damage from bacterial and
neutrophil enzymes facilitates progressive bacterial invasion of
481

the cornea. There may be progressive tissue necrosis resulting
in sloughing of the epithelium and stroma, which varies with
the virulence of the organism and toxin production. The necrotic
base of the ulcer is surrounded by heaped-up tissue. The host
cellular and humoral immune defense mechanisms retard
bacterial replication, promote phagocytosis of the organism and
cellular debris, and halt destruction of stromal collagen.
In the healing phase, the epithelium resurfaces the central
area of ulceration and the necrotic stroma is replaced by scar
tissue produced by broblasts. The reparative broblasts are
derived from histiocytes and keratocytes that have undergone

transformation. New epithelium slowly resurfaces the irregular
base. The physiologic processes involved in corneal wound
healing are similar to what are described above. Bowmans layer
does not regenerate but is replaced with brous tissue. New
blood vessels are directed toward the area of ulceration to
deliver humoral and cellular components to promote healing.
These gradually disappear and may leave ghost vessels. The
brous scar tissue results in corneal opacity, which may
gradually fade over time.
SECTION 6
REFERENCES
482
1. Potten CS, Morris RJ: Epithelial stem cells

in vivo. J Cell Sci Suppl 1988; 10:4562.
2. Spangrude GJ, Heimfeld S, Weissman IL:
Purication and characterization of mouse
hematopoietic stem cells. Science 1988;
241:5862.
3. Dua HS, Azuara-Blanco A: Limbal stem
cells of the corneal epithelium. Surv
Ophthalmol 2000; 44:415425.
4. Thoft RA, Wiley LA, Sundarraj N: The
multipotential cells of the limbus. Eye 1989;
3(Pt 2):109113.
5. Tseng SC: Concept and application of
limbal stem cells. Eye 1989; 3(Pt
2):141157.
6. Huang AJ, Watson BD, Hernandez E,
Tseng SC: Induction of conjunctival
transdifferentiation on vascularized corneas
by photothrombotic occlusion of corneal
neovascularization. Ophthalmology 1988;
95:228235.
7. Tseng SC, Farazdaghi M: Reversal of
conjunctival transdifferentiation by topical
retinoic acid. Cornea 1988; 7:273279.
8. Kruse FE, Chen JJ, Tsai RJ, Tseng SC:
Conjunctival transdifferentiation is due to
the incomplete removal of limbal basal
31:19031913.
9. Kinoshita S, Friend J, Thoft RA: Biphasic
cell proliferation in transdifferentiation of
conjunctival to corneal epithelium in
24:10081014.
10. Tseng SC, Hirst LW, Farazdaghi M, Green
WR: Goblet cell density and vascularization
during conjunctival transdifferentiation. Invest
11. Chen WY, Mui MM, Kao WW, et al:
Conjunctival epithelial cells do not
transdifferentiate in organotypic cultures:
expression of K12 keratin is restricted to
corneal epithelium. Curr Eye Res 1994;
13:765778.
12. Moyer PD, Kaufman AH, Zhang Z, et al:
Conjunctival epithelial cells can resurface
denuded cornea, but do not
transdifferentiate to express corneaspecic keratin 12 following removal of
limbal epithelium in mouse. Differentiation
1996; 60:3138.
13. Lehrer MS, Sun TT, Lavker RM: Strategies
of epithelial repair: modulation of stem cell
and transit amplifying cell proliferation.
J Cell Sci 1998; 111(Pt 19):28672875.
14. Chung EH, Bukusoglu G, Zieske JD:
Localization of corneal epithelial stem cells
in the developing rat. Invest Ophthalmol Vis
Sci 1992; 33:21992206.
15. Dua HS: Stem cells of the ocular surface:
scientic principles and clinical
applications. Br J Ophthalmol 1995;
79:968969.
16. Lavker RM, Wei ZG, Sun TT: Phorbol ester

preferentially stimulates mouse fornical
conjunctival and limbal epithelial cells to
proliferate in vivo. Invest Ophthalmol Vis
Sci 1998; 39:301307.
17. Tseng SC: Regulation and clinical
implications of corneal epithelial stem cells.
Mol Biol Rep 1996; 23:4758.
18. Thoft RA: The role of the limbus in ocular
surface maintenance and repair. Acta
Ophthalmol Suppl 1989; 192:9194.
19. Daniels JT, Dart JK, Tuft SJ, Khaw PT:
Corneal stem cells in review. Wound Repair
Regen 2001; 9:483494.
20. Moore JE, McMullen CB, Mahon G,
Adamis AP: The corneal epithelial stem
cell. DNA Cell Biol 2002; 21:443451.
21. Davanger M, Evensen A: Role of the
pericorneal papillary structure in renewal of
corneal epithelium. Nature 1971;
229:560561.
22. Schermer A, Galvin S, Sun TT:
Differentiation-related expression of a major
64K corneal keratin in vivo and in culture
suggests limbal location of corneal epithelial
stem cells. J Cell Biol 1986; 103:4962.
23. Kurpakus MA, Stock EL, Jones JC:
Expression of the 55-kD/64-kD corneal
keratins in ocular surface epithelium. Invest
24. Kurpakus MA, Maniaci MT, Esco M:
Expression of keratins K12, K4 and K14
during development of ocular surface
epithelium. Curr Eye Res 1994; 13:805814.
25. Rodrigues M, Ben Zvi A, Krachmer J, et al:
Suprabasal expression of a 64-kilodalton
keratin (no. 3) in developing human corneal
epithelium. Differentiation 1987; 34:6067.
26. Bickenbach JR, Mackenzie IC: Identication
and localization of label-retaining cells in
hamster epithelia. J Invest Dermatol 1984;
82:618622.
27. Bickenbach JR, McCutecheon J,
Mackenzie IC: Rate of loss of tritiated
thymidine label in basal cells in mouse
epithelial tissues. Cell Tissue Kinet 1986;
19:325333.
28. Cotsarelis G, Cheng SZ, Dong G, et al:
Existence of slow-cycling limbal epithelial
basal cells that can be preferentially
stimulated to proliferate: implications on
epithelial stem cells. Cell 1989;
57:201209.
29. Barrandon Y, Green H: Cell size as a
determinant of the clone-forming ability of
human keratinocytes. Proc Natl Acad Sci
USA 1985; 82:53905394.
30. Barrandon Y, Green H: Three clonal types
of keratinocyte with different capacities for
multiplication. Proc Natl Acad Sci USA
1987; 84:23022306.
31. Ebato B, Friend J, Thoft RA: Comparison of
central and peripheral human corneal
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
epithelium in tissue culture. Invest

Ebato B, Friend J, Thoft RA: Comparison of
limbal and peripheral human corneal
Buck RC: Cell migration in repair of mouse
corneal epithelium. Invest Ophthalmol Vis
Sci 1979; 18:767784.
Lindberg K, Brown ME, Chaves HV, et al: In
vitro propagation of human ocular surface
epithelial cells for transplantation. Invest
Pellegrini G, Golisano O, Paterna P, et al:
Location and clonal analysis of stem cells
and their differentiated progeny in the
human ocular surface. J Cell Biol 1999;
145:769782.
Wei ZG, Wu RL, Lavker RM, Sun TT: In vitro
growth and differentiation of rabbit bulbar,
fornix, and palpebral conjunctival epithelia.
Implications on conjunctival epithelial
transdifferentiation and stem cells. Invest
Wei ZG, Lin T, Sun TT, Lavker RM: Clonal
analysis of the in vivo differentiation
potential of keratinocytes. Invest
Kruse FE, Tseng SC: A serum-free clonal
growth assay for limbal, peripheral, and
central corneal epithelium. Invest
Kruse FE, Tseng SC: Proliferative and
differentiative response of corneal and
limbal epithelium to extracellular calcium in
serum-free clonal cultures. J Cell Physiol
1992; 151:347360.
Kruse FE, Tseng SC: Differing regulation of
proliferation of limbus and corneal
epithelium caused by serum factors.
Ophthalmologe 1993; 90:669678.
Kruse FE, Tseng SC: Growth factors
modulate clonal growth and differentiation
of cultured rabbit limbal and corneal
34:19631976.
Chen JJ, Tseng SC: Corneal epithelial
wound healing in partial limbal deciency.
31:13011314.
Chen JJ, Tseng SC: Abnormal corneal
epithelial wound healing in partial-thickness
removal of limbal epithelium. Invest
Dua HS, Saini JS, Azuara-Blanco A, Gupta
P: Limbal stem cell deciency: concept,
aetiology, clinical presentation, diagnosis
and management. Indian J Ophthalmol
2000; 48:8392.
Huang AJ, Tseng SC: Corneal epithelial
wound healing in the absence of limbal
32:96105.
46. Lavker RM, Dong G, Cheng SZ, et al:

Relative proliferative rates of limbal and
corneal epithelia. Implications of corneal
epithelial migration, circadian rhythm, and
suprabasally located DNA-synthesizing
keratinocytes. Invest Ophthalmol Vis Sci
1991; 32:18641875.
47. Wei ZG, Sun TT, Lavker RM: Rabbit
conjunctival and corneal epithelial cells
belong to two separate lineages. Invest
48. Thoft RA: Conjunctival transplantation as
an alternative to keratoplasty.
49. Wei ZG, Cotsarelis G, Sun TT, Lavker RM:
Label-retaining cells are preferentially
located in fornical epithelium: implications
on conjunctival epithelial homeostasis.
36:236246.
50. Huang AJ, Tseng SC, Kenyon KR:
Morphogenesis of rat conjunctival goblet
29:969975.
51. Wirtschafter JD, Ketcham JM, Weinstock RJ,
et al: Mucocutaneous junction as the major
source of replacement palpebral
conjunctival epithelial cells. Invest
52. Zieske JD: Perpetuation of stem cells in the
eye. Eye 1994; 8(Pt 2):163169.
53. Schoeld R: The stem cell system. Biomed
Pharmacother 1983; 37:375380.
54. Zieske JD, Wasson M: Regional variation in
distribution of EGF receptor in developing
and adult corneal epithelium. J Cell Sci
1993; 106(Pt 1):145152.
55. Kasper M, Moll R, Stosiek P, Karsten U:
Patterns of cytokeratin and vimentin
expression in the human eye.
Histochemistry 1988; 89:369377.
56. Kasper M: Patterns of cytokeratins and
vimentin in guinea pig and mouse eye
tissue: evidence for regional variations in
intermediate lament expression in limbal
epithelium. Acta Histochem 1992;
93:319332.
57. Lauweryns B, van den Oord JJ, Missotten L:
The transitional zone between limbus and
peripheral cornea. An immunohistochemical
study. Invest Ophthalmol Vis Sci 1993;
34:19911999.
58. Steuhl KP, Thiel HJ: Histochemical and
morphological study of the regenerating
corneal epithelium after limbus-to-limbus
denudation. Graefes Arch Clin Exp
Ophthalmol 1987; 225:5358.
59. Hayashi K, Kenyon KR: Increased
cytochrome oxidase activity in alkaliburned corneas. Curr Eye Res 1988;
7:131138.
60. Li DQ, Tseng SC: Differential regulation of
cytokine and receptor transcript expression
in human corneal and limbal broblasts by
epidermal growth factor, transforming
growth factor-alpha, platelet-derived
growth factor B, and interleukin-1 beta.
37:20682080.
61. Wilson SE, He YG, Weng J, et al: Effect of
epidermal growth factor, hepatocyte growth
factor, and keratinocyte growth factor, on
proliferation, motility and differentiation of
human corneal epithelial cells. Exp Eye Res
1994; 59:665678.
62. Wilson SE, Liu JJ, Mohan RR: Stromalepithelial interactions in the cornea. Prog
Retin Eye Res 1999; 18:293309.
63. Li DQ, Tseng SC: Differential regulation of

keratinocyte growth factor and hepatocyte
growth factor/scatter factor by different
cytokines in human corneal and limbal
broblasts. J Cell Physiol 1997;
172:361372.
64. Pellegrini G, Dellambra E, Golisano O, et al:
p63 identies keratinocyte stem cells. Proc
Natl Acad Sci USA 2001; 98:31563161.
65. Chung EH, DeGregorio PG, Wasson M,
Zieske JD: Epithelial regeneration after
limbus-to-limbus debridement. Expression
of alpha-enolase in stem and transient
amplifying cells. Invest Ophthalmol Vis Sci
1995; 36:13361343.
66. Zieske JD, Bukusoglu G, Yankauckas MA:
67. Tani H, Morris RJ, Kaur P: Enrichment for
murine keratinocyte stem cells based on
cell surface phenotype. Proc Natl Acad Sci
USA 2000; 97:1096010965.
68. Matic M, Petrov IN, Chen S, et al: Stem
cells of the corneal epithelium lack
connexins and metabolite transfer capacity.
Differentiation 1997; 61:251260.
69. Budak MT, Alpdogan OS, Zhou M, et al:
Ocular surface epithelia contain
ABCG2-dependent side population cells
exhibiting features associated with stem
cells. J Cell Sci 2005; 118(Pt 8):17151724.
70. Pajoohesh-Ganji A, Pal-Ghosh S,
Simmens SJ, Stepp MA: Integrins in slow
cycling corneal epithelial cells at the limbus
in the mouse. Stem Cells 2005; Nov 10.
Epub ahead of print.
71. Pajoohesh-Ganji A, Ghosh SP, Stepp MA:
Regional distribution of alpha9beta1
integrin within the limbus of the mouse
ocular surface. Dev Dyn 2004;
230:518528.
72. Thoft RA, Friend J: The X, Y, Z hypothesis
of corneal epithelial maintenance. Invest
73. Buck RC: Measurement of centripetal
migration of normal corneal epithelial cells
in the mouse. Invest Ophthalmol Vis Sci
1985; 26:12961299.
74. Haddad A: Renewal of the rabbit corneal
epithelium as investigated by
autoradiography after intravitreal injection
of 3H-thymidine. Cornea 2000;
19:378383.
75. Sharma A, Coles WH: Kinetics of corneal
epithelial maintenance and graft loss. A
population balance model. Invest
76. Dua HS, Forrester JV: Clinical patterns of
corneal epithelial wound healing. Am J
Ophthalmol 1987; 104:481489.
77. Pster RR: The healing of corneal epithelial
abrasions in the rabbit: a scanning electron
microscope study. Invest Ophthalmol 1975;
14:648661.
78. Cintron C, Hassinger LC, Kublin CL,
Cannon DJ: Biochemical and ultrastructural
changes in collagen during corneal
wound healing. J Ultrastruct Res 1978;
65:1322.
79. Beuerman RW, Thompson HW: Molecular
and cellular responses of the corneal
epithelium to wound healing. Acta
80. Crosson CE, Klyce SD, Beuerman RW:
Epithelial wound closure in the rabbit
cornea. A biphasic process. Invest
81. Matsuda M, Ubels JL, Edelhauser HF.

A larger corneal epithelial wound closes at
a faster rate. Invest Ophthalmol Vis Sci
1985; 26:897900.
82. Kinoshita S, Kiorpes TC, Friend J, Thoft RA:
Limbal epithelium in ocular surface wound
healing. Invest Ophthalmol Vis Sci 1982;
23:7380.
83. Berman M, Leary R, Gage J: Evidence for a
role of the plasminogen activator plasmin
system in corneal ulceration. Invest
84. Khodadoust AA, Silverstein AM, Kenyon DR,
Dowling JE: Adhesion of regenerating
corneal epithelium. The role of basement
65:339348.
85. Gipson IK, Spurr-Michaud S, Tisdale A,
Keough M: Reassembly of the anchoring
structures of the corneal epithelium during
wound repair in the rabbit. Invest
86. Azar DT, Hahn TW, Jain S, et al: Matrix
metalloproteinases are expressed during
wound healing after excimer laser
keratectomy. Cornea 1996; 15:1824.
87. Cintron C, Hassinger LC, Kublin CL,
Cannon DJ: Biochemical and ultrastructural
changes in collagen during corneal wound
healing. J Ultrastruct Res 1978; 65:1322.
88. Girard MT, Matsubara M, Kublin C, et al:
Stromal broblasts synthesize collagenase
and stromelysin during long-term tissue
remodeling. J Cell Sci 1993;
104(Pt 4):10011011.
89. Johnson-Wint B: Regulation of stromal cell
collagenase production in adult rabbit
cornea: in vitro stimulation and inhibition by
epithelial cell products. Proc Natl Acad Sci
USA 1980; 77:53315335.
90. Andresen JL, Ehlers N: Chemotaxis of
human keratocytes is increased by plateletderived growth factor-BB, epidermal growth
factor, transforming growth factor-alpha,
acidic broblast growth factor, insulin-like
growth factor-I, and transforming growth
factor-beta. Curr Eye Res 1998; 17:7987.
91. Watsky MA: Keratocyte gap junctional
communication in normal and wounded
rabbit corneas and human corneas. Invest
92. Lemp MA: Cornea and sclera. Arch
Ophthalmol 1976; 94:473490.
93. Wilson SE, Mohan RR, Ambrosio R:
Corneal injury. A relatively pure model of
stromal-epithelial interactions in wound
healing. Methods Mol Med 2003; 78:6781.
94. Matsubara M, Girard MT, Kublin CL, et al:
Differential roles for two gelatinolytic
enzymes of the matrix metalloproteinase
family in the remodelling cornea. Dev Biol
1991; 147:425439.
95. Strissel KJ, Rinehart WB, Fini ME: A
corneal epithelial inhibitor of stromal cell
collagenase synthesis identied as
TGF-beta 2. Invest Ophthalmol Vis Sci
1995; 36:151162.
96. Wilson SE, Chen L, Mohan RR, et al:
Expression of HGF, KGF, EGF and receptor
messenger RNAs following corneal
epithelial wounding. Exp Eye Res 1999;
68:377397.
97. Wilson SE, He YG, Weng J, et al: Epithelial
injury induces keratocyte apoptosis:
hypothesized role for the interleukin-1
system in the modulation of corneal tissue
organization and wound healing. Exp Eye
Res 1996; 62:325327.
CHAPTER 41
483
SECTION 6
98. Jester JV, Huang J, Petroll WM,

Cavanagh HD: TGFbeta induced
myobroblast differentiation of rabbit
keratocytes requires synergistic TGFbeta,
PDGF and integrin signaling. Exp Eye Res
2002; 75:645657.
99. Berman M, Leary R, Gage J: Evidence for a
role of the plasminogen activator plasmin
system in corneal ulceration. Invest
100. Kao WW, Kao CW, Kaufman AH, et al:
Healing of corneal epithelial defects in
plasminogen- and brinogen-decient
484
mice. Invest Ophthalmol Vis Sci 1998;

39:502508.
101. Fini ME, Girard MT: Expression of
collagenolytic/gelatinolytic
metalloproteinases by normal cornea.
31:17791788.
102. Fini ME, Girard MT, Matsubara M:
Collagenolytic/gelatinolytic enzymes in
corneal wound healing. Acta Ophthalmol
Suppl 1992; 70:2633.
103. Azar DT, Hahn TW, Jain S, et al: Matrix
metalloproteinases are expressed during
wound healing after excimer laser

104. Ye HQ, Azar DT: Expression of gelatinases
A and B, and TIMPs 1 and 2 during corneal
wound healing. Invest Ophthalmol Vis Sci
1998; 39:913921.
105. West-Mays JA, Sadow PM, Tobin TW, et al:
Repair phenotype in corneal broblasts is
controlled by an interleukin-1 alpha
autocrine feedback loop. Invest Ophthalmol
Vis Sci 1997; 38:13671379.
CHAPTER
42
Corneal Examination, Specular and Confocal

Microscopy, UBM, OCT
Ula V. Jurkunas and Kathryn Colby
The cornea is a transparent, supercial tissue, and various

examination techniques readily identify corneal pathology. The
use of different light wavelengths, illumination and magnication modes in the instruments described below allows the
examiner to delineate both morphologic and functional changes
in corneal tissue.
PENLIGHT EXAMINATION
Useful information about the cornea can often be gathered
using a simple penlight in an illuminated room before the slitlamp (SL) examination. For example, inflammation of the
eyelids, conjunctiva, sclera, episclera, or anterior uvea are often
better noted with this exam. Penlight examination may also
reveal a localized or general opacication or vascularization of
the cornea. Corneal surface irregularities, such as abrasions,
can be diagnosed by scanning the cornea with a penlight,
looking for an irregular light reflection (Fig. 42.1).
SL MICROSCOPY
The techniques of anterior segment examination were crude at
best until the development of the rst SL by Gullstrand in
1911.1 The combination of Gullstrands focal illumination
source and the Czapski microscope was the rst successful

system that allowed the light beam to be projected at various
angles and focused at different depths of the eye.1 Since then,
renement of SL biomicroscopy has led to universal acceptance
of this modality. Current widely used models of SL include
those made by HaagStreit, Zeiss, Bausch & Lomb Thorpe, and
Nikon.1 The SL is a compound binocular microscope which
provides variable magnication for delivery of the brightest
possible image. It allows maximum magnication of 40 and
resolution of 20 mm (Table 42.1).2
The SL uses corneal properties to transmit and reflect light.
Since the cornea is translucent, most of the light encountered is
transmitted, but some is scattered and reflected back, enabling
the visualization of the tissue which is not crystal clear. Due
to the difference in refractive indices (RI) between two
interfaces, such as airtear or stromaaqueous, the light that is
reflected back and/or scattered forms an image, such as seen
during SL examination. When angle of incidence of light is
equal to the angle of reflection, the incident light forms a bright
reflex called specular reflection.2 In clinical practice the two
most commonly used images are the surface corneal light reflex
from the epithelial surface and the specular reflex from the
corneal endothelium. Light can also be reflected in a diffuse
manner when the angles of incidence and reflection are not
equal, as seen in light scattering. Surface irregularities such as
scarring or epithelial edema reflect light in a variety of directions giving the tissue its opaque appearance. The more dense
the opacity the greater the scattering of the reflected light. The
terms nebula, macula and leukoma represent the continuum of
opacity density with the latter denoting a most opaque white
scar.2 The internal corneal reflection of light from the epithelial
basement membrane or Descemets membrane is called
sclerotic scatter. The light can also be transmitted back through
the cornea after it has been reflected from surrounding ocular
structures such as the iris and lens, highlighting corneal guttae
or epithelial edema by retroillumination.3
TABLE 42.1. Comparison of Different Instrument Optical

Properties2,7,9,24
FIGURE 42.1. Irregular corneal reflex indicating a corneal abrasion.
Instrument
Maximum
Magnification
Resolution
View
Slit Lamp
40
20 mm
Transverse
Specular
microscope
500
25 mm
Lateral
Axial
Confocal
microscope
600
12 mm lateral
510 mm axial
Lateral
Axial
485
CLINICAL TECHNIQUES
The SL examination is a dynamic process that allows an in vivo
three-dimensional view of the structures of the cornea. By
varying the illumination technique and the aperture of the slit
beam, one can achieve different methods of examining the
cornea. Two particularly useful techniques include specular
reflection and retroillumination.
otherwise. For example, a small foreign body can be identied

within a dense area of corneal inflammation when viewed by
this method.
Sclerotic Scatter
This technique is used as the means of initial overview of the

anterior ocular structures and provides a general survey and
localization of abnormalities. The slit beam should be opened
to the maximal width. The intensity of the light beam can be
adjusted to minimize patient discomfort. The initial inspection
should be started with low magnication. The light beam can
be rotated at various angles in order to delineate the shape and
the extent of any abnormalities.
When the slit beam is directed at the limbus, the opaque sclera
scatters the light, directing part of it inside the corneal stroma.
Based on the principle of total internal reflection of light, the
light travels through the entire stroma and is reflected back and
forth from the anterior and posterior corneal surfaces. When no
opacity is present, the halo of light emerges back 360 around
limbus. Any opacity impedes the spread of light inside the
cornea and reflects the scattered light back to the examiner. By
this method faint opacities such as subepithelial inltrates or
mild epithelial edema due to Fuchs endothelial dystrophy can
be readily observed. This technique may be used early in the
examination, to acclimatize the patient to the light before it is
directed into the pupil.
Direct Focal Illumination
Specular Reflection
In order to zoom in on the area of interest, the slit beam is

narrowed and placed at an angle to produce a well illuminated
and magnied optical section. By rotating the illumination
arm, the examiner can obtain vertical, horizontal and oblique
sections. When the illumination arm is placed at an angle of 45
from the observation axis, the information from the particular
optical section is maximized (Fig. 42.2). The depth of the
abnormality can be determined and, therefore, localized to a
particular corneal layer. Abnormal depressions and elevations,
corneal thinning or changes in corneal shape or contour can be
elucidated by observing differential bending of the thin beam of
light. By moving the slit systematically across the cornea, serial
optical sections are viewed, and a mental construct of corneal
pathology is created.
The microscope is placed directly in line with the angle of

reflection, so angle of incidence of the slit beam is equal to the
angle of microscope observation.3 The posterior specular
reflection is at the interface between the endothelium and the
aqueous and is formed due to difference in RI between two
structures. Because the difference in RI between the aqueous
and endothelium is small, only 0.022% of total incident light is
reflected, thus forming only a faint image.2 Since the cornea is
curved, narrowing the beam to ~0.5 mm eliminates the surrounding glare from other tissues. If the beam is too narrow, not
enough corneal surface is illuminated to examine the mosaic
pattern. To start, the beam of light is projected to the central
cornea from the temporal side, then moved towards the
periphery. The slit-beam width should be constantly adjusted
to avoid bright reflections from the tear lm. The magnication can be adjusted for maximum identication of cellular
structures.13
Diffuse Illumination
Indirect Proximal Illumination

This technique aids in identifying the details of the pathology
within the opacity. The beam is shortened to 23 mm,
broadened and focused to the area adjacent to the opaque area.
The light beam will undergo internal reflection within the
cornea and will be scattered as it hits the opacity. The reflected
light will readily highlight the internal structure of the opacity,
and will aid in identifying the details that could be missed
Retroillumination from the Iris

By placing the slit beam at an angle of 60 from the microscope,
transmitted light through the cornea is reflected back from the
iris or surrounding structures. It allows objects whose RI differ
from the surrounding medium to be observed. Subtle anterior
basement membrane changes or endothelial guttae are often
best visualized using this technique (Fig. 42.3).1,2
SECTION 6
ADVANTAGES AND DISADVANTAGES
486
FIGURE 42.2. The slit-lamp beam strikes the cornea in an oblique

fashion. The scattering properties of the cornea give the viewer an
optical cross-section. In this case, folds in Descemets membrane can
be seen.
SL biomicroscopy is the most commonly used and readily

available method of examining the cornea. It is a fast screening
technique that does not require additional technical training.
The utilization of dyes, most commonly fluorescein, readily
provides the examiner with additional information about the
different patterns of surface disruption differentiating, for
example, exposure to keratopathy from superior limbic keratoconjunctivitis. Rose Bengal dye, which stains devitalized
epithelial cells or cells lacking mucin, will stain the areas
affected from tear deciency states when the epithelium is still
intact and fluorescein staining is negative.
SL biomicroscopy is limited in its resolution of structures.
The light reflected from other corneal structures anterior to the
area of interest can obscure ne detail. Also, other corneal pathology, such as edema or scarring can altogether block the image.
The SL has a number of disadvantages for examining the endothelium, including low intensity of reflected light, low magnication of the cells and illumination of only a small area at any
one time.2 In general these disadvantages are far outweighed
Corneal Examination, Specular and Confocal Microscopy, UBM, OCT

of the endothelium. Wider slit illumination also causes morphological distortion of endothelial cells.7 Technological improvement in the design of the objective lens has led to the use of
wide-eld SMs without compromising the quality of the image.
In contrast to the SL, the modern SM is designed with separate
illumination and light-viewing pathways, so that reflection from
anterior corneal structures is reduced.7,12,14 Also, a scanning
system developed by Koester captures the entire endothelial
mosaic without overtly compromising resolution.8,10,14
CLINICAL ANALYSIS
Qualitative
The parameters evaluated are cell morphology, cell boundaries
and intersections, posterior corneal surface, and additional
structures.
Cell morphology
FIGURE 42.3. Illumination from the fundus provides an orange

background for the early signs of lattice dystrophy of the cornea.
Normal corneal endothelial cells are hexagonal in shape and

form a regular pattern of contiguous cells. Normally, the cells
are of same size: cell-side lengths are equal and the intersection
between all cell sides is ~120 (Fig 42.4).7 With age cells
become larger, there is more variation in cell wall intersection
angles, with overall tendency towards pleomorphism, or variation
in cell shape (Table 42.2).15 Cell shape can also change from
by the easy use and ready availability, making SL biomicroscopy

the most common corneal examination technique.
SPECULAR MICROSCOPY
FIGURE 42.4. (a) SM image of corneal endothelium with

morphometric analysis. This is a normal measurement of endothelial
cell density (CD) and coefcient of variation (CV). SD, standard
variation. (b) Specular microscopy image of corneal guttae
represented as hyporeflective areas (arrow).
TABLE 42.2 Specular Microscopy Parameters7,15

Qualitative
Parameter
Quantitative
Parameter
Normal Values
Quantitative
Equations
Cell size
Cell density
(cell
count)
Cell area
Children
3500 cells/mm
Adults
2400 cells/mm
106/mean
cell area
106/mean
cell
density
Variation in
cell size
polymegathism
Coefficient
of variation
Less 0.3
Mean cell
area/SD
of mean
cell area
Variation in cell
shape
pleomorphism
Percentage
in hexagonal cells
100%
N/A
CHAPTER 42
It was David Maurice in 1968 who rst photographed the

posterior corneal surface of an enucleated rabbit eye, and
published a specular image of the endothelium, captured by the
instrument he called a specular microscope.4,5 Liang and
colleagues subsequently published photomicrographs with
improved resolution that showed individual endothelial cell
boundaries and intracellular structures.6,7 In the 1980s Koester
developed a wide-eld specular microscope (SM) that captured
the entire endothelial mosaic with good resolution by scanning
narrow endothelial layer zones and projecting them on the
same lm.8 Since then specular microscopy has gained widespread clinical acceptance as a method for evaluating endothelial cell density and morphology.5,7,9
The advent of SM has greatly improved the study of human
corneal endothelial morphology and allowed quantication of
endothelial changes. Both types of SM, contact and noncontact,
produce an image superior to the one that is obtained by the SL.
Specular microscopy provides a larger overlapping image of
endothelial cell layer, higher magnication and less interference
from patients eye movement (Table 42.1). The latter advantage
is mostly seen in contact SM, although it can also be seen with
noncontact SM, if the nal image alignment is automated.7,10
Computer-assisted morphometric analysis is a powerful tool
present in most SMs that standardizes cell counting and
analysis, image and data management and provides data
storage.7,1012 Some machines have a pachymeter attached for
measurement of corneal thickness.13
SM captures the specular reflection of light formed at the
optical interface between the endothelium and the aqueous
humor. By increasing the angle of incidence of the illuminating
source, the width of the slit beam can also be increased to image
a wider area of endothelium. By increasing the width of the slit
beam, scattering of light from structures anterior to endothelium
produces more diffuse illumination of the surrounding ocular
tissues and a consequent decrease in the contrast and denition
487

hexagonal to elongated, as seen in the apex of keratoconus.
Also, rounded, square and triangular cells have been noted,
without clear clinical signicance of such changes.7
Cell boundaries and intersections

The variability in cell boundary intersections (such that angles
between the walls deviate from 120), signies thermodynamic
instability of the endothelium.15 Such boundary formation is
usually formed by cells in transition brought by loss of nearby
cells.15
Posterior corneal surface

The examination of boundary between endothelium and
aqueous humor depicts a silhouette of the posterior corneal
surface which can be smooth or irregular. Hyperreflective
excrescences on Descemets membrane due to guttae can be
depicted against the dark background.7,16
Additional structures
Corneal guttae appear as hyporeflective oval-to-round areas
with central highlight at the apex (Fig. 42.4). The surrounding
endothelial cells are hyperreflective in relation to guttae since
the endothelium overlying the guttae is out of plane of focus
and appears as dark or absent.16 Evaluation of endothelial
mosaic and structures around it can distinguish most common
corneal endothelial disorders (Table 42.3).15,17
Quantitative
Wide-eld SBM permits evaluation of both central and
peripheral cornea and studies of regional variability.18 The morphologic parameters can be quantied and are summarized in
Table 42.2.
SECTION 6
The SM provides a detailed picture of both the cell density and

morphology of the corneal endothelium. The wide-eld view
provides an image of the entire mosaic permitting a study of
regional variability. In clinical practice the SM is useful for
diagnosis of patients with unilateral corneal edema and no
corneal gutta visible on SL exam of the contralateral eye.19 If
reduced cell counts and abnormal morphology are evident on
specular microscopy, the diagnosis of endothelial cell dysfunction can be made (Table 42.3).19
The main disadvantage of the SM is still image resolution,
which is limited by light scattering of the surrounding corneal
488
tissue. In advanced corneal edema or scarring, the endothelial

mosaic can not be visualized with the SM. In order to discern
the cellular detail, the endothelial layer has to be smooth and in
the same plane of focus.16,20 Otherwise the cells are seen as
hyporeflective images with no additional information on their
morphology. As described above, corneal guttae manifest as
confluent reflex-free areas (Fig. 42.4). By changing a plane of
focus it is possible to discern endothelial cells on top of these
excrescences, but these cells are not accounted for in the
measurement of the cell count.16,21 Also, SM has disadvantages
of cost, availability and a need of image interpretation by an
expert or a corneal specialist.
CONFOCAL MICROSCOPY
The confocal microscope (CM), invented in 1957 by Marvin
Minsky, began to be used for imaging the cornea in vivo in
the mid 1990s.9,22,23 The improved optics of CM allow the
magnied imaging of all corneal layers in x,y,z axes over time,
rendering a truly novel four-dimensional in vivo microscopy.9,23
The improvement in lateral (x, y) and axial (z) resolution with
the CM was achieved by eliminating scattered light that is not
in the focal plane of the imaged object (Table 42.1). Diffractionlimiting pinhole or slit apertures focus the light source and the
image on the same focal plane, thus creating a confocal
image.24 As a result, out-of-focus reflected signals are excluded.
Because such apertures create a very small eld of view of only
a single spot on the cornea, the instrument has to scan the
whole sample by moving both the illuminator and the detector
in a synchronous fashion. By simply varying the plane of focus,
the source and detector scan the tissue along z axis and provide
magnied coronal sections at a variable depth.5,14,24 The speed
at which a single eld is scanned at a constant depth provides
the time resolution feature of the microscope. In order to reconstruct a real-time view on the screen5,8,24 rapid image capture
systems are necessary during scanning to avoid interference
from a patients pulse, respiration and ocular microsaccades.24
The newest models contain optical pachymetry, video recording
and automated cell analysis features (Fig. 42.5).22
2. Nipkow disk
1. Light
source
5. Cornea
4. Objective
lens
3. Beam
splitter
TABLE 42.3 Specular Microscopy Finding in Posterior Corneal

Disorders1517,20
Specular
Findings
Fuchs
PPMD
ICE
Cells
Mosaic
Decreased
endothelial
cell density
Polymegathism
Pleomorphism
Disrupted by
guttae
Pleomorphism
Irregular large
cells with
scalloped
edges as
dark rings
around
lighter
center
Vesicles/pits
in DM
Disrupted by
crater-like
focal lesions
Enlarged ICE cells

White reflections
in center of
dark cells
dark-light
reversal
No disruption
from craters or
guttae
Direct
Video
viewing
camera
6. Front surface
Mirror
FIGURE 42.5. A disk with pinholes arranged with conjugate

symmetry is used to provide confocal point source and point
detectors. A full eld of view is obtained in real time (i.e., >30
frames/sec) by rotating the disk at high speed to scan the specimen.
The image can be viewed directly or imaged using a video camera
and recorded on videotape or continuously displayed on a monitor, or
both.
CLINICAL APPLICATION
Two types of CM are available, which differ in the type of
scanning system employed. Tandem-scanning CMs (TSCMs)
and slit-scanning CMs (SSCMs) scan the sample either by a
plate of pinholes in the former or by thin optical slits in the
latter.24
SSCM is particularly well suited for imaging transparent
tissues such as cornea because the slit height can be varied to
achieve optimal image clarity.9 The adjustment of the slit adjusts
the depth of focus at z axis and minimizes the surrounding
noise. Also, the slit provides greater signal and higher image
clarity for a particular video frame. For example, using this
device imaging of the poorly reflecting epithelial wing cell layer
is possible.25 In a commercially available SSCM (the Confoscan
3 (NIDEK Technologies), Fig. 42.6), the distance immersion
principle is applied and a noncontact floating probe is utilized.
No alteration in tissues and improved patient comfort render
this technique safer and potentially more popular.9 The major
disadvantage of this SSCM is the inability to quantify the z axis
when a noncontact probe is utilized. Since a xed zero point is
not available, the assessment of depth of the image becomes
arbitrary. In clinical situations, the assessment of keratocyte
count post-refractive surgery or detection of post-LASIK flap
depth might be better performed by the contact method of

a TSCM.2628
TSCM scans the cornea using Napkow disk containing
~64 000 diffraction-limited pinholes of 20 mm arranged in
spirals. This arrangement allows a z axis resolution of 9 mm and
x, y axes resolution of less than 1 mm. Since the light transmission though the pinholes is low (less than 1%), the loss of
luminance limits the resolution of low-contrast structures. The
TSCM that is commercially available (Advanced Scanning, Ltd.)
utilizes dipping cone objectives. Since the position of the focal
plane relative to the objective lens can be varied, recorded and
converted to z axis position, the depth of the tissue focal plane
can be accurately calibrated.25 In order to reduce surrounding
noise of micromovements, a real-time digitizing system has
been developed that averages sequential frames and saves the
best-quality images. Confocal microscopy through focusing
(CMTF) is a powerful tool which enables imaging of entire
cornea in a highly rapid manner (in ~8 s). By moving the focal
plane of the objective lens through the cornea and capturing
the image focal plane in digitized manner, hundreds of images
are created with available on-screen three-dimensional
reconstruction.25
By measuring the distance between the intensity peaks on
the CMFT intensity curve, tissue and sublayer thicknesses can
FIGURE 42.6. ConfoScan 3 (Nidek

Technologies) image of corneal endothelium.
Automated cell count is performed after the
region of interest (ROI) position and dimension
is selected by the user. By default, the cells are
lled with different colors, depending on their
dimensions and cell side numbers. Percentages
of polymegathism and pleomorphism are
calculated.
Courtesy of Jose de la Cruz, MD. New York Eye and
Ear Inrmary, New York, New York.
Cell Count: 1960 [cell/mm2]

Normal: 2015-3552 [cell/mm2]
Polymegathism: 34.4%
Normal: <30%
Cell sides statistics (N =124)
CHAPTER 42
Cell density and area statistics (N =124)
Pleomorphism: 60.5%
Normal: >59.6%
ConfoScan 3 Imaging system rev. 3.1 Nidek Technologies s.r.i.
489

be assessed.25 In pre- and post-LASIK patients the effects of
laser ablation on wound healing can be quantied by observing
flap thickness, tissue loss and regrowth, and keratocyte morphology and density.25,27,28
The role of CM in diagnosis and management of atypical
corneal ulcers such as Acanthamoeba, Microsporidium and
fungal keratitis has received widespread attention, but has not
gained widespread utilization.29 TSMF has been showed to be
particularly promising in indentifying the location and depth of
Acanthamoeba trophozoites and cysts when employed by
experienced observers.25,2932 Due to complexity of result
interpretation, CM use remains in the hands of a few academic
centers.
One of the main advantages of the CM over the SM is its
ability to image the endothelium through a hazy cornea, as is
seen in corneal edema.33,34 Also, earlier detection of Descemets
membrane alterations have been reported with CM as opposed
to SM and SL.35,36 On the hand, most of the studies comparing
confocal and specular microscopy nd both instruments to be
equal in their clinical application.24,12,14,35,37 The comparison of
SM and CM on evaluation of corneal endothelium is given in
Table 42.4.
Newer advances in in vivo microscopy have combined
Heidelberg retina tomography (HRT II) and the Rostock cornea
module (Heidelberg Engineering GmbH) (HRT II/RCM) into a
digital confocal laser scanning microscope.38 After a polymethylmethacrylate (PMMA) plate contacts the patients ocular
surface, layer-by-layer three-dimensional images are created and
displayed in a computer monitor. Preliminary studies of HRT
II/RCM have conrmed superior image contrast in evaluation
of corneal and conjunctival layers on a cellular level, including
visualization of corneal keratocytes, Langerhans cells, meibomian
glands and goblet cells.38,39

In summary, the optics of the CM allows improved resolution
and magnication of the human corneal structures. Both cost
and result interpretation in diagnostic use of CM preclude its
widespread utilization in clinical practice, rendering CM
primarily as a research tool at present.24 On the other hand,
ever-evolving technological aspects in confocal imaging will
most likely turn this technique into a routine practice in
anterior segment evaluation in the future.
SECTION 6
OPTICAL COHERENCE TOMOGRAPHY
490
The cross sectional imaging by optical coherence tomography

(OCT) has been available for in vivo examination of posterior
segment since 1990s. First described by Huang, and later

developed into SL-adapted OCT system by Izatt and colleagues,
this new technique led to the novel and noninvasive diagnostic
evaluation of the anterior segment in 1994.4042
The OCT utilizes an infrared diode light source (wavelength
of 830 or 1310 nm) and a Michelson-type interferometer that
detects differential light backscattering from the tissue microstructures. The amplitudes and delays in tissue reflections are
scanned by the reference mirror and the interferometric signal
is simultaneously recorded.40 The OCTs false-color images
denote the regions with strongest reflection with red and white
colors and the regions with weakest reflection with blue and
black.43 The OCT imaging of transparent corneal tissues in
cross section is comparable to histopathologic sectioning, and it
provides resolution of 1020 mm (Fig. 42.7).44 Additional quantitative information, such as structural dimensions and backscattering amplitudes, are readily available by this noninvasive,
noncontact technique.41 The clinically relevant measurements
possible by OCT include corneal thickness, iris thickness, lens
thickness, anterior chamber depth, anterior angle-chamber
angle dimensions and anterior and posterior radii of corneal
curvature with subsequent calculation of corneal refractive
power. In order to image the cornea at a close-up view, higher
spatial frequencies are used to sample the eye. The reflectivity
prole in the longitudinal scan direction can be numerically
tted and extrapolated to the cornea and its substructure (i.e.,
epithelial layer) thickness measures by measuring the distances
between the optical signals. Because the image contrast is high
between the cornea and the surrounding media, the strong
reflections between anterior and posterior corneal surfaces are
recorded and the corneal thickness measurements are estimated in submicrometer scale (Fig. 42.7).41,45 A relatively low
contrast between the corneal epithelium and Bowmans layer
reduces the accuracy of thickness estimates and the epithelial
layer measures might vary by 10 mm between the images.41 In
order to determine true corneal thickness, the OCTs pachymetric measurements have to be corrected by factoring in the
refractive index of the cornea, assumed to be 1.3853.42
OCT has been shown to be a useful tool when examining a
normal cornea. Similarly to ultrasound biomicroscopy (UBM),
but in a less invasive manner, OCT allows identication and
monitoring of intraocular masses and tumors.41 The
relationships between the cornea, anterior chamber angle and
lens can be assessed. At this point, intraocular lenses can not be
visualized in vivo with OCT because they consist of homogenous material with smooth surfaces. On the other hand, the
TABLE 42.4 Comparison of Specular Microscopy and Confocal Microscopy12,14,24,35,37

Advantages
Disadvantages
Clinical use
Specular
Wider field of view

More accurate endothelial cell counts
Easier to use
Quality of images does not depend on
patients movement
Cornea has to be transparent
Most cases of endothelial cell

assessment
Confocal
Higher image resolution (axial and

lateral) and contrast
Higher magnification
Can visualize endothelium in
edematous cornea
Real-time endothelial cell assessment
Eye movement interferes with exam

Narrower field of view
Less accurate endothelial cell count in
presence of guttae
Cellular organelles not routinely seen
Expensive not routinely used for
clinical practice yet
Earlier diagnosis of dystrophies

(can visualize DM thickening)
Use in severe cases of corneal edema
Other corneal structures assessed at
the same time
b
FIGURE 42.7. OCT image of cornea with scan width of 12 mm and
scan depth 4 mm. (a) Contour of normal cornea is clearly identiable
against the dark background. The amount of backscatter from within
the nominally transparent cornea decreases from central to peripheral
cornea. The intensity of the signal increases at the corneoscleral
limbus as it approaches highly scattering opaque sclera. Central
corneal thickness is measured to be 561 mm. (b) Image of the corneal
edema. The backscatter intensity is increased in the central region.
Central corneal thickness is measured to be 811 mm.
Courtesy of David Huang, MD, PhD. Doheny Eye Institute, Los Angeles, CA.
capsular bag can be identied, especially in the setting of

posterior capsular opacication. Although lens densitometric
analysis for the objective grading of cataract formation has been
explored, that model still remains experimental and has not
been widely utilized.41,42
The availability of highly accurate biometric measures makes
OCT an invaluable tool in intraocular implant power calculations and for tting of contact lenses.
In refractive surgery realm the OCTs pachymetric analysis
has been utilized for intraoperative and perioperative corneal
thickness measurement.45 The flap-interface reflectivity enables
the measurements of flap thickness up to 15 months postoperatively.46 The intraoperative measurements of residual
stromal thickness by online OCT coupled with excimer laser
(Online OCP, 4Optics AG) are advantageous due to the employment of the noncontact method and the ability to obtain continuous measurements of central corneal thickness throughout
the procedure, thus improving the intraoperative safety of
keratorefractive surgery.47,48
Some authors claim that changes in corneal shape and contour following refractive surgery, as well as wound healing effects
that alter light scattering characteristics of collagen brils and
keratocytes, can be reliably imaged and quantied by OCT.42,46
In pathologic conditions, OCT can identify and delineate the
extent and depth of calcied lesions, dystrophic opacities, and
lesions whose accurate assessment is precluded by corneal
edema or haze (Fig. 42.7).43,49 In the diffusely hazy cornea, a
descemetocoele or corneal perforation can be detected. Crosssectional images in postoperative patients after deep lamellar
keratoplasty may be used to monitor grafthost junction in
initially edematous grafts.43

OCT provides direct quantitative measurements of ocular
tissues in cross section. It does not require immersion or direct
contact with the ocular surface and does not disrupt the tissues
under investigation. Patient discomfort is also minimal. Optical
UBM
The rst use of ultrasound in ophthalmology was in 1956 by
Mundt and Hughes.50 Since then, standardization of both A and
B-scan instruments has led to the widespread utilization of
ultrasound for intraocular and orbital disorders.51 In 1990s
Pavlin and associates popularized the use of UBM, which
greatly enhanced the resolution of anterior segment structures
and lesions.52 In UBM observation of living tissues at a microscopic level is similar to optical biomicroscopy, hence the term.
The essential components of UBM are identical to a conventional B-mode imaging system except for the signicantly
higher operating frequencies and subsequently short wavelengths (less than 0.2 mm). The resulting longitudinal ultrasound waves carry properties similar to light rays since they can
be refracted and reflected. The reflected waves are referred to as
an echo when they hit back the source of the emitted energy
(i.e., the transducer or probe).5153
UBM utilizes high-frequency transducers in 40100 MHz range
(frequencies greater than 20 KHz are inaudible to humans) to
provide resolution of 2060 mm and depth of tissue penetration
of ~4 mm. The lateral resolution of the ultrasound system can
be related to the full width of the ultrasound beam at halfmaximum amplitude (FWHM) and expressed by the equation:
FWHM = cf/(vd)=lambda (f-number)
where c is the speed of sound (a speed of 1 640 m/s is generally
used for cornea and sclera), f is the focal length of the transducer, v is the frequency of ultrasound, d is the diameter of the
focused transducer, lambda is the wave length, and f-number is
the ratio of the focal length to the diameter of the transducer. By
selecting appropriate frequency and an f-number, the variably
high resolution can be achieved. For example, 60 mm resolution
can be achieved by operating frequency of 60 MHz and an
f-number of 2.0. By increasing the resolution, tissue penetration
is compromised, due to tissue ultrasound attenuation
coefcients that increase with frequency. Therefore, for the
60 MHz frequency, penetration is ~5 mm. The optimization of
transducer parameters is essential in creating the best image
quality, and it is achieved by a compromise between resolution,
contrast and depth of eld (range of depth over which the beam
remains well focused).51
CHAPTER 42
sectioning of the ocular surface encompasses most of the

anterior segment and provides a general view of the structural
relationships between anterior segment components. The
resolution is comparable to the CM. Still most images are not
able to distinguish Bowmans and Descemets membranes
when no pathological thickening is present.42 Similarly to the
previously discussed instrument limitations, patient micromovements during data acquisition might cause image resolution degradation requiring averaging of serial measurements.41
When performing biometric analysis, strictly axial measurements should be obtained, as off-axis images contain optical
distortions arising from surfaces crossed by a nonperpendicular
beam of the OCT.42 In summary, OCT provides an optical
biopsy of the anterior segment structures and supplements
corneal examination with a unique and noninvasive method.
The utilization of UBM in anterior segment examination is
most applicable when corneal opacication or total internal
reflectivity precludes the visualization of the ocular structures.
In the normal cornea, three highly reflective surface echoes are
produced by the epithelium, Bowmans membrane and the
Descemets membrane/endothelial complex.53 Corneal stroma
491

has low regular reflectivity and it is lower than that found in the
more irregular collagen distribution of the sclera. The difference
of reflectivity between the corneal stroma and the sclera allows
for denition of the corneoscleral junction.
In corneal edema, the separation of the corneal lamellae by
fluid enhances the stromal reflectivity. In bullous keratopathy
the epithelial echo becomes more irregular, and the separation
of epithelium from the stroma becomes readily visible. Other
causes of increased stromal reflectivity are due to deposition of
higher reflectivity material in-between the corneal lamellae and
disruption of their regular (usually weakly reflective) structure,
as seen in scarring, inflammation and dystrophic material
accumulation. Areas of calcication are highly reflective, and
produce complete shadowing of structures behind it.53
When corneal opacication is present it is difcult to assess
the underlying cornea and the anterior chamber. UBM allows
the assessment of corneal anatomy despite corneal edema as
seen in Descemets detachment (Fig. 42.8).53,54 In corneal transplantation, UBM aids in examination of grafthost junction,
evaluation of wound gaping, apposition of Descemets membrane between grafthost junction and presence of iris adhesions
or angle closure.55 The intraocular lens haptic position can be
identied by UBM and can facilitate the preoperative planning
for lens exchange.56
UBM creates a qualitative representation of iris, ciliary body,
lens and anterior chamber structures, as well as accurately measures these structures (Fig. 42.9).51,52 The utilization of short
wavelengths and improved resolution of UBM over the conventional ultrasound led to a more accurate measurement of small
diameter structures, such cornea, iris, ciliary body, sclera. The
differential reflectivity of corneal layers enables the measurement of stromal thickness, epithelial thickness, and depth of
intracorneal incisions.57 After LASIK and the photorefractive
keratectomy, the alteration in epithelial layer, Bowmans layer
and stromal layer reflectivity can be picked up by UBM and
can aid in assessing results and complications of refractive
surgery.57,58
FIGURE 42.8. UBM image demonstrating Descemets detachment.

Epithelial layer and Bowmans membrane create two highly reflective
lines. Highly reflective image of Descemets membrane and
endothelium is separated from corneal stroma. Edematous cornea has
thickened stroma with higher than usual stromal reflectivity.
Courtesy of Lois Hart. Massachusetts Eye and Ear Inrmary, Boston, MA.
SECTION 6
492
UBM produces cross-sectional images of the anterior globe and

allows observation of living tissues at a magnied level. In UBM
the integrity of the imaged structures and their relationship to
one another is preserved. The vast majority of clinical applications that UBM can potentially be employed were beyond the
scope of this text. Still, the new advances in transducer sensitivity are in evolution to improve the tissue penetration and
image resolution of UBM technology for corneal and anterior
segment examination.
KERATOMETRY (OPHTHALMOMETRY)
In 1916, Scheiner59 noticed that shiny glass balls of different
radii produce reflected images of different sizes. This prompted
him to make a series of balls of progressively larger curvatures.
To perform keratometry, Scheiner would match the size of the
image of the window frame reflected form a subjects cornea
with that produced by one of the calibrated balls.
The next major advance in keratometer was a magnication
system introduced by Ramsden.60 Ramsden also introduced the
doubling device, in which the examiner matches the corneal
reflection to itself, thus eliminating annoying eye movement.
The cornea acts as a convex mirror and produces an erect and
virtual image of the illuminated target placed near the patients
cornea. Keratometry allows the operator to measure the size of
the reflected image precisely. The device then converts image
size to corneal radius using the following relations:
b
FIGURE 42.9. UBM image showing iris and ciliary body cyst (c). Note
cyst walls and lack of internal echoes indicating fluid inside the cysts.
(a) Axial view. (b) Transverse view.
Courtesy of Lois Hart. Massachusetts Eye and Ear Inrmary, Boston, MA.

Corneal radius = (2[cornea-to-mire distance] [corneal image
size])/mire size
Corneal refractive power = 03375/corneal radius in meters
The range of most keratometers covers all patients, except
those with extreme keratoconus and cornea plana. The technique used to extend the range of the keratometer to include
these special patient groups uses a spherical lens mounted over
the central aperture of the keratometer mire. The cornea is then
measured in the usual way, and the value is multiplied by a
constant unique to the auxiliary lens.
The examiner can utilize the keratometer to evaluate the
quality of the corneal surface as well as the dioptric curvature of
the anterior cornea. The keratometric mires that are reflected
from the cornea fall on an area of 3.03.5 mm and the resulting
keratometric measurement does not represent the curvature of
the entire cornea.61 The quality of mire overlap can distinguish
between regular and irregular astigmatism. When the mires do
not overlap perfectly, and/or have irregular shape, one should
suspect an ocular surface irregularity or keratoectasia. In
keratoconus, there is steepening and thinning of the paracentral
cornea, that manifest with irregular astigmatism and high
keratometric values.61
FIGURE 42.10. Computerized keratometry device, which uses 18

concentric Placido rings that produce a reflected image that covers
almost the entire cornea. The corneal curvature in different parts of
the cornea is color coded.
Courtesy of Tomey Technology, INC., Cambridge, MA.
Modern video techniques can freeze a reflected corneal image

and use the information in that image to approximate the
corneal shape. Once the image is captured on a video screen, a
computer can measure the image and calculate the radius of
curvature. Corneal topography creates a map representation of
the corneal surface. Most commonly used topographers are
based on mire arrangement similar to a Placido disc. A series of
illuminated rings is projected onto a cornea and the reflected
images are captured on the video screen. A computer analysis
reports the radius of curvature in any portion of the cornea and
produces color-coded dioptric maps of the corneal surface.62 A
standard topogram gives a clear cylinder axis and amount of
corneal astigmatism in diopters (e.g., simulated keratometry
values (SIM K) (Fig. 42.10). A reasonably accurate assessment
of irregular astigmatism can be achieved by observation of color
map or by using numerical indices (e.g., surface regularity index
(SRI) or surface asymmetry index (SAI)) provided by some
analyzers.62
In slit-scanning corneal topography (SSCT), the machine
projects a series of slit beams across the cornea. Each of the slit
images is captured and analyzed by the computer software. The
information is used to calculate the shape and corneal thickness
between the captured slit sections. The commercially available
SSCT, the Orbscan, creates a graphic data output of anterior
corneal curvature, posterior corneal curvature and regional map
of corneal thickness in addition to mean axial corneal power.63
There is a multitude of clinical applications of corneal topography that aid in diagnosis and management of corneal
abnormalities. Irregular astigmatism from corneal scarring,
keratoectasia, trauma, surgery or postinflammatory conditions
can be readily depicted by keratography (Fig. 42.11).62 Topography may explain why best-corrected acuity does not improve
with spectacle refraction in patients with irregular corneas.
The detection of irregular astigmatism may herald an early
ectasia that is a contraindication to the refractive surgery.64 In
Orbscan systems the anterior and posterior differences in the
best-t spheres, mean axial dioptric maps and pachymetry maps
can aid in detection and monitoring of patients with keratoconus
(Fig. 42.12). The representative maps after the LASIK surgery
for myopia show characteristic flattening in the mean axial
power map, as opposed to steepening in keratoconus maps
(Figs 42.12 and 42.13). Topography aids in the determination of
selective suture removal in corneal transplant patients and
surgical planning of astigmatism by both incisional and laserassisted surgery (Fig. 42.14).62
In summary, the advent of refractive surgery created a niche
for the advances in evaluation and measurement of corneal
shape and power. Such methods have evolved from keratometry
to keratoscopy to videokeratoscopy and to SSCT. Despite the
advances in this area, the ever-evolving instrumentation is still
needed to combat the inaccuracies and inefciencies of the
existing technology.
CHAPTER 42
TOPOGRAPHY (VIDEOKERATOSCOPY)
FIGURE 42.11. Orbscan mean axial

keratometric map (left) of oblique against the
rule astigmatism of 13.2 D in a patient with
Terriens marginal degeneration located
superiorly. Corneal thickness map (right) shows
normal central thickness of 559 mm (green
color) with marked thinning superiorly (red
color) due to ectasia.
493

FIGURE 42.12. Orbscan topography of
keratoconus. (a) Early keratoconus of the right
eye with inferior steepening in the mean axial
keratometric map (bottom left). The area of
steepening coincides with area of thinning
(bottom right). (b) Advanced keratoconus of the
left eye of the same patient. Anterior best-t
sphere float (top left) and posterior best-t
sphere float (top right) show markedly higher
elevation compared to the right eye (a) with
less advanced keratoconus. Advanced
steepening denoted with red colors in the mean
axial keratometric map (bottom left) coincides
with thinning in the thickness map (bottom
right).
SECTION 6
494
FIGURE 42.13. Orbscan topography of postmyopic LASIK treatment.

Anterior best-t sphere float (top left) shows concentric elevation in the
paracentral area denoted with yellow color, and central elevation
denoted with red color in the posterior best-t sphere map (top right).
The pachymetry map shows central thinning (bottom right), while mean
axial keratometric map shows characteristic central corneal flattening
(bottom left). The constellation of the ndings above distinguishes this
keratometric map from the one seen in keratoconus.

FIGURE 42.14. Mean axial keratometric maps of Orbscan topographer.
(a) Against-the-rule astigmatism in corneal transplant. (b) After suture
removal at 3 and 9 oclock, the astigmatism becomes with-the-rule in
the same patient due to remaining tight sutures at 6 and 12 oclock.
1. Tate GW, Sar A: The slit lamp: history,

principles, and practice. In: Tasman W,
Jaeger EA, eds. Duanes clinical
ophthalmology. Philadelphia: Williams &
Wilkins; 2004:144.
2. Waring GO: Slit-lamp microscopy of the
cornea. In: Leibowitz HM, Waring GO, eds.
Corneal disorders. Philadelphia: W.B.
Saunders; 1998:3481.
3. Jurkunas U, Colby K: Evaluation of the
corneal endothelium. Tech Ophthalmol
2005; 3:1926.
4. Maurice DM: Cellular membrane activity in
the corneal endothelium of the intact eye.
Experientia 1968; 24:10941095.
5. Masters BR: David Maurices contributions
to ophthalmic instrumentation: roots of the
scanning slit confocal microscope. Exp Eye
Res 2004; 78:315326.
6. Liang RA, Sandstrom MM, Leibowitz HM:
In vivo photomicrography of the corneal
endothelium. Arch Ophthalmol 1975;
93:143145.
7. Laing RA, Oak SS, Leibowitz HA:
Specialized microscopy of the cornea. In:
Leibowitz HM, Waring GO, eds. Corneal
disorders. Philadelphia: W.B. Saunders;
1998:83122.
8. Koester CJ: Scanning mirror microscope
with optical sectioning characteristics:
applications in ophthalmology. Appl Optics
1980; 19:17491757.
9. Masters BR, Bohnke M: Three-dimentional
confocal microscopy of the living human
eye. Annu Rev Biomed Eng 2002;
4:6991.
10. Ohno K, Nelson LR, McLaren JW, et al:
Comparison of recording systems and
analysis methods in specular microscopy.
Cornea 1999; 18:416423.
11. Geroski DH, Edelhauser HF: Morphometric
analysis of the corneal endothelium.
Specular microscopy vs. alizarin red
staining. Invest Ophthalmol Vis Sci 1989;
30:254259.
12. Hara M, Morishige N, Chikama T, Nishida T:
Comparison of confocal biomicroscopy
and noncontact specular microscopy for
evaluation of the corneal endothelium.
Cornea 2003; 22:512515.
13. Modis L, Langenbucher A, Seitz B: Corneal
endothelial cell density and pachymetry
measured by contact and noncontact
specular microscopy. J Cataract Refract
Surg 2002; 28:17631769.
14. Cavanagh HD, El-Agha MS, Petroll WM,
Jester JV: Specular microscopy, confocal
microscopy, and ultrasound biomicroscopy.
Cornea 2000; 19:712722.
15. Yeh PC, Colby K: Corneal endothelial
dystrophies. In: Foster CS, Azar DT,
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
Dohlman CH, eds. Smolin and Thofts the

cornea. Philadelphia: Williams & Wilkins;
2005:849874.
Wilson SE, Bourne WM: Fuchs dystrophy.
Cornea 1988; 7:218.
Colby K: Corneal endothelial dystrophies.
Contemp Ophthalmol 2004; 3:18.
Amann J, Holley GP, Lee SB, Edelhauser
HF: Increased endothelial cell density in the
paracentral and peripheral regions of the
135:584590.
Abbott RL, Fine BS, Webster RG, et al:
Specular microscopic and histologic
observations in nonguttate corneal
endothelial degeneration. Ophthalmology
1981; 88:788800.
Borboli S, Colby K: Mechanisms of
disease: Fuchs endothelial dystrophy.
Ophthalmol Clin North Am 2002;
15:1725.
Olsen T: On the signicance of a low
endothelial cell density in Fuchs
endothelial dystrophy. A specular
microscopic study. Acta Ophthalmol 1980;
58:111116.
Minsky M: Memoir on inventing the
confocal scanning microscope. Scanning
1988; 10:128138.
Petran M, Hadravsky M, Egger MD,
Galambos R: Tandem-scanning
reflected-light microscope. J Opt Soc Am
1968; 58:661664.
Jalbert I, Stapleton F, Papas E, et al: In vivo
confocal microscopy of human cornea.
Cavanagh HD, Ladage PM, Petroll WM,
Jester JV: Clinical confocal microscopy. In:
Foster CS, Azar DT, Dohlman CH, eds.
Smolin and Thofts the cornea.
Philadelphia: Williams & Wilkins;
2005:187198.
McLaren JW, Nau CB, Kitzmann AS,
Bourne WM: Keratocyte density:
comparison of two confocal microscopes.
Eye Contact Lens 2005; 31:2833.
Erie JC, Nau CB, McLaren JW, et al:
Long-term keratocyte decits in the corneal
stroma after LASIK. Ophthalmology 2004;
111:13561361.
Avanduk AM, Senft CJ, Emerab S, et al:
Corneal healing after uncomplicated LASIK
and its relationship to refractive changes: a
six-month prospective confocal study.
45:13341339.
Kaufman SC, Musch DC, Belin MW,
et al: Confocal microscopy. A report
by the American Academy of
Ophthalmology. Ophthalmology 2004;
111:396406.
30. Winchester K, Mathers WD, Sutphin JE,

Daley TE: Diagnosis of Acanthamoeba
keratitis in vivo with confocal microscopy.
Cornea 1995; 14:1017.
31. Mathers WD, Sutphin JE, Folberg R, et al:
Ophthalmol 1996; 121:129142.
32. Mathers WD, Nelson SE, Lane JL, et al:
Conrmation of confocal microscopy
diagnosis of acanthamoeba keratitis using
polymerase chain reaction analysis. Arch
Ophthalmol 2000; 118:178183.
33. Kaufman SC, Beuerman RW, Kaufman HE:
Diagnosis of advanced Fuchs endothelial
dystrophy with the confocal microscope.
Am J Ophthalmol 1993; 116:652653.
34. Chiou AGY, Kaufman SC, Beuerman RW,
et al: Confocal microscopy in posterior
polymorphous corneal dystrophy.
35. Mustonen RK, McDonald MB,
Srivannaboon S, et al: In vivo confocal
microscopy of Fuchs endothelial
dystrophy. Cornea 1998;
17:493503,416423.
36. Chiou AGY, Kaufman SC, Beuerman RW,
et al: Confocal microscopy in cornea
guttata and Fuchs endothelial dystrophy.
37. Klais CMC, Buhren J, Kohnen T:
Comparison of endothelial cell count
using confocal and contact specular
microscopy. Ophthalmologica 2003;
217:99103.
38. Bochert R, Zhivov A, Kraak R, et al:
Contribution to comprehension of image
formation on confocal microscopy of
cornea with Rostock cornea module.
Br J Ophthalmol 2005; 89:13511355.
39. Zhivov A, Stave J, Vollmar B, Guthoff R: In
vivo confocal microscopic evaluation of
Langerhans cell density and distribution in
the normal human corneal epithelium.
243:10561061.
40. Huang D, Swanson EA, Lin CP, et al:
Optical coherence tomography. Science
1991; 254:11781181.
41. Izatt JA, Hee MR, Swanson EA, et al:
Micrometer-scale resolution imaging of the
anterior eye in vivo with optical coherence
tomography. Arch Ophthalmol 1994;
112:15841589.
42. Hoerauf H, Wirbelauer C, Scholz C, et al:
Slit-lamp-adapted optical coherence
tomography of the anterior segment.
238:818.
43. Hirano K, Ito Yasuki, Suzuki T, et al: Optical
coherence tomography for the noninvasive
CHAPTER 42
REFERENCES
495
44.
45.
46.
47.
48.
SECTION 6
49.
496
evaluation of the cornea. Cornea 2001;

20:281289.
Wirbelauer C, Winkler J, Bastian GO, et al:
Histopathological correlation of corneal
diseases with optical coherence
tomography. Graefes Arch Clin Exp
Ophthalmol 2002; 240:727734.
Muscat S, McKay N, Parks S, et al:
Repeatability and reproducibility of corneal
thickness measurements by optical
coherence tomography. Invest Ophthalmol
Vis Sci 2002; 43:17911795.
Wirbelauer C, Pham DT: Monitoring corneal
structures with slitlamp-adapted optical
coherence tomography in laser in situ
2004; 30:18511860.
Wirbekauer C, Pham DT: Continuous
monitoring of corneal thickness changes
during LASIK with online optical coherence
pachymetry. J Cataract Refract Surg 2004;
30:25592568.
Thompson RW, Choi DM, Price MO, et al:
Noncontact optical coherence tomography
for measurement of corneal flap and
residual stromal bed thickness after laser in
situ keratomileusis. J Refract Surg 2003;
19:507515.
Wirbelauer C, Pham DT: Imaging and
quantication of calcied corneal lesions
with optical coherence tomography. Cornea
2004; 23:439442.
50. Mundt GH Jr, Hughes WF Jr: Ultrasonics in

ocular diagnosis. Am J Ophthalmol 1956;
41: 488.
51. Pavlin CJ, Foster FS: Ultrasound
biomicroscopy of the eye. In: Byrne SF,
Green RL, eds. Ultrasound of the eye and
orbit. 2nd edn. St Louis: Mosby Inc;
2002:111, 223235.
52. Pavlin CJ, Harasiewicz K, Sherar MD,
Foster FS: Clinical use of ultrasound
biomicroscopy. Ophthalmology 1991;
98:287295.
53. Pavlin CJ, Foster FS: Corneal disease. In:
Ultrasound biomicroscopy of the eye. New
York: Springer; 1995:155169.
54. Avitabile T, Russo V, Ghirlanda R, et al:
Corneal oedemas: diagnosis and surgical
planning with ultrasound biomicroscopy.
55. Madhavan C, Basti S, Naduvilath TJ,
Sangwan VS: Use of ultrasound
biomicroscopic evaluation in preoperative
planning of penetrating keratoplasty.
Cornea 2000; 19:1721.
56. Rutnin SS, Pavlin CJ, Slomovic AR, et al:
Preoperative ultrasound biomicroscopy to
assess ease of haptic removal before
penetrating keratoplasty combined with
lens exchange. J Cataract Refract Surg
1997; 12:239243.
57. Reinstein DZ, Silverman RH, Raevsky T, et
al: Arc-scanning very high-frequency digital
58.
59.
60.
61.
62.
63.
64.
ultrasound for 3D pachymetric mapping of

the corneal epithelium and stroma in laser
in situ keratomileusis. J Refract Surg 2000;
16:414430.
McWhae J, Willerscheidt A, Gimbel H,
Freese M: Ultrasound biomicroscopy in
refractive surgery. J Cataract Refract Surg
1994; 20:493497.
Scheiner C: Occlusion fundamentum
opticum. Innsbruck, Austria; 1619.
Levene JR: Clinical refraction and visual
science. Woburn, MA: Butterworths;
1977:128.
Krachmer JH, Mannis MJ: Refraction of the
abnormal cornea. In: Krachmer JH, Mannis
MJ, Holland EJ, eds. Cornea. 2nd edn.
Philadelphia: Elsevier Mosby;
2005:167170.
Maguire LJ: Keratometry,
photokeratoscopy, and computer-assisted
topographic analysis. In: Krachmer JH,
Mannis MJ, Holland EJ, eds. Cornea. 2nd
edn. Philadelphia: Elsevier Mosby;
2005:171184.
Cairns G, McGhee NJ: Orbscan
computerized topography: attributes,
applications, and limitations. J Cataract
Refract Surg 2005; 31:205220.
Donnenfeld E, Wu H, McDonnell P,
Rabinowitz Y: Keratoconus and corneal
ectasia after LASIK. Letter. J Cataract
Refract Surg 2005; 31:20352037.
CHAPTER
43
Corneal Dysgeneses, Dystrophies, and

Degenerations
Kenneth R. Kenyon, Tomy Starck, Glen Cockerham, and Peter S. Hersh
Corneal Dysgeneses
Abnormalities of size and curvature
Absence of cornea
Microcornea
Simple megalocornea
Anterior megalophthalmos
Cornea plana
Mesenchymal dysgeneses
Posterior embryotoxon
Axenfelds anomaly and syndrome
Reigers anomaly and syndrome
Posterior keratoconus
Congenital central corneal opacity (Peters anomaly)
Sclerocornea
Congenital anterior staphyloma
Congenital hereditary endothelial dystrophy
Corneal Dystrophies
Anterior dystrophies
Epithelial basement membrane dystrophies (map-dotngerprint)
Hereditary epithelial dystrophy (Meesmann, Stocker-Holt)
Lisch corneal dystrophy
Corneal dystrophies of Bowmans layer
Vortex dystrophy (Fleischers)
Anterior mosaic crocodile shagreen (Vogts)
Idiopathic band keratopathy
Stromal dystrophies
Granular dystrophy (Groenouws Type I)
Lattice dystrophy
Macular dystrophy (Groenouws Type II)
Polymorphic stromal dystrophy
Gelatinous Drop-like dystrophy
Central crystalline dystrophy (Schnyder)
Marginal crystalline dystrophy (Bietti)
Central cloudy dystrophy (Francois)
Posterior amorphous stromal dystrophy
Congenital hereditary stromal dystrophy
Posterior mosaic crocodile shagreen
Fleck dystrophy (Francois-Neetens)
Pre-Descemets dystrophies
Cornea farinata
Grayson-Wilbrandt dystrophy
Deep filiform dystrophy
Endothelial dystrophies
Congenital hereditary endothelial dystrophy
Cornea guttata
Late hereditary endothelial dystrophy (Fuchs)
Posterior polymorphous dystrophy
Iridocorneal endothelial syndrome
Noninflammatory corneal ectasias

Keratoconus
Pellucid marginal degeneration
Keratoglobus
Corneal Degenerations
Peripheral degenerations
Corneal arcus (juvenilis and senilis)
White limbal girdle of Vogt
Idiopathic furrow degeneration
Furrow degeneration associated with systemic disease
Postirradiation thinning
Terriens marginal degeneration
Moorens ulcer
Central or diffuse degenerations
Iron lines
Coats white ring
Lipid degeneration
Amyloid degeneration
Spheroid degeneration (climatic droplet keratopathy, keratinoid
degeneration)
Band keratopathy
Salzmanns nodular degeneration
Corneal keloid
Conjunctival Degenerations
Pterygium
Pinguecula
The dysgeneses, dystrophies, and degenerations of the cornea

account for a broad spectrum of ocular abnormalities, ranging
from clinical curiosities to sight-threatening anomalies. Knowledge of these entities has traditionally accrued through clinical
study and examination of histopathologic specimens. Within
the past decade, discovery of the specic gene mutations for
corneal stromal dystrophies on human chromosome 5 has
advanced the study of corneal disorders into the exciting age of
molecular genetics. (The recent monograph edited by M. Wang1
is especially current and comprehensive in this regard.)
Dysgeneses of the cornea are developmental disorders,
sometimes inherited, resulting in congenital malformations.
Corneal dysgeneses may be unilateral or bilateral and are
nonprogressive. The central, peripheral, or entire cornea, as
well as other ocular structures, may be affected. Occasionally,
associated systemic abnormalities are present.
A corneal dystrophy generally exhibits a familial pattern, is
bilateral if not symmetric, and does not appear to be secondary
to any environmental or systemic factor. Dystrophies tend to
497

manifest relatively early in life and are variably progressive.
Abnormalities generally affect the central cornea and are noninflammatory in origin. Senescence may encourage deterioration of the dystrophic cornea but is not a primary cause of
the disorder. Each unique dystrophy exhibits characteristic
histopathologic features.
Corneal degenerations, in contrast to dysgeneses and dystrophies, appear to have no developmental or hereditary pattern
and may be unilateral or bilateral. A degeneration is often a
manifestation of aging, inflammation, or environmental insult
and, therefore, usually occurs later in life than a dystrophy.
Degenerations most often begin in the peripheral cornea,
although central vision eventually may be affected. Inflammation sometimes is involved early in the degenerative process
and may be accompanied by corneal vascularization. In some
cases, these inflammatory processes are associated with
systemic disease (e.g., collagen vascular disorders).
CORNEAL DYSGENESES
ABNORMALITIES OF SIZE AND CURVATURE
Absence of Cornea
Complete absence of the cornea is rare. In such cases, there is
variable absence of other anterior ocular structures derived from
surface ectoderm, and the eye consists of a sclera-like enclosure
lined with neural ectoderm. Ultrasonography should aid in
differentiating this entity from cryptophthalmos. True
cryptophthalmos, also known as ablepharon, occurs when the
lids fail to form. The cornea and conjunctiva are exposed and
undergo metaplastic changes to form skin.2 This condition is
rare and is usually transmitted as an autosomal recessive trait.3
The term cryptophthalmos syndrome has been used to describe
the association of the ocular ndings with extraocular
abnormalities, such as craniofacial anomalies, syndactyly, spina
bida, cleft lip and palate, genitourinary and cardiac anomalies,
and mental retardation.4
SECTION 6
Microcornea
The term microcornea (Fig. 43.1, left) implies a corneal diameter of less than or equal to 10 mm. The size of a normal
newborn cornea measures ~10 mm in horizontal diameter,
whereas the size of a normal adult cornea measures ~12 mm in
diameter. The vertical diameter almost always is exceeded by
~1 mm by the horizontal diameter. The cornea usually reaches
adult size by 2 years of age.5 Microcornea can occur either
unilaterally or bilaterally and is thought to occur secondary to
an arrest in corneal growth after the fth month of fetal
development. The eye may be otherwise normal, but often
other ocular abnormalities, such as colobomas, are present. Just
as megalocornea is associated occasionally with anterior
megalophthalmos, microcornea often accompanies anterior microphthalmos, with crowding of the anterior segment structures
commonly resulting in angle-closure glaucoma.6 Microcornea
can also be seen in nanophthalmos and as part of many other

anterior segment dysgeneses.
The microcornea is generally clear, with normal histologic
architecture, and in the absence of other ocular abnormalities,
vision may be good. Numerous somatic abnormalities have
been described in conjunction with microcornea and anterior
microphthalmos, including dwarsm and EhlersDanlos
syndrome.7
Simple Megalocornea
Simple megalocornea (see Fig. 43.1, right) is a nonprogressive,
usually symmetric, inherited condition in which the cornea and
limbus are enlarged without evidence of previous or concurrent
ocular hypertension. The diameter of the cornea is 13 mm or
greater, but the corneal thickness and histologic anatomy are
normal. Although X-linked recessive inheritance is most common with 90% of all cases found among males, all modes of
inheritance have been reported.810 Female carriers may have
slightly enlarged corneas.
The condition has been mapped to the long arm of the Xchromosome (Xq21.3-q22 and Xq12-q26).11
Simple megalocornea can be differentiated from congenital
glaucoma by the clarity of the cornea and by the normal intraocular pressure and normal optic nerve in simple megalocornea.
Moreover, the megalocornea demonstrates normal endothelial
cell population densities on specular microscopy, whereas in
congenital glaucoma, these are diminished, ostensibly because
of corneal distention.12 Studies have also suggested using
A-scan ultrasonography to highlight the pathognomonic
biometric ndings of megalocornea not present in glaucoma:
markedly increased anterior chamber depth, posterior lens and
iris positioning, and short vitreous length.13 Although some
authorities suspect that megalocornea may represent arrested
congenital glaucoma, a single case reporting the histopathology
of megalocornea did not disclose any of the characteristic angle
abnormalities of congenital glaucoma. Both conditions, however,
have been reported in the same family and in the same person.14,15
Simple megalocornea also must be differentiated from
keratoglobus (see section on Noninflammatory Corneal Ectasias).
Anterior Megalophthalmos
In comparison with simple megalocornea, eyes with anterior
megalophthalmos have enlargement of the lens-iris diaphragm
and ciliary ring in addition to the cornea.16 A large myopic
astigmatic refractive error often results from the abnormal
optical architecture. The iris may exhibit transillumination
defects as a result of attenuation of the dilator muscle. Because
of the abnormal spatial relations of structures in the anterior
segment and stretching of the zonules, iridodonesis, phakodonesis, and lens subluxation or dislocation may occur; the
latter may result in secondary lens-induced glaucoma. The lens,
furthermore, may become prematurely cataractous.
Marfans syndrome,17 Aperts syndrome,18 and mucolipidosis
type II19 have been found in association with this disorder.
FIGURE 43.1. Left, Microcornea. A young child
had a cornea 9.5 mm in diameter and subtle
peripheral sclerocornea. Right, Megalocornea.
Light microscopy of a 62-year-old man with
corneal diameters of 13 mm. Note the anterior
segment with no abnormalities (except beveled
scar of cataract incision and surgical aphakia).
H & E 3.
Right, From Wood WJ, Green WR, Marr WG:
Megalocornea: a clinico-pathologic clinical case report.
Md State Med J 1974; 23:5760.
498
Corneal Dysgeneses, Dystrophies, and Degenerations
Disorder
Gene Location
Cornea plana
Chromosome 12
Corneal dystrophy of Bowmans layer

type I
Unknown
Corneal dystrophy of Bowmans layer

type II
Chromosome 5
Granular dystrophy
Chromosome 5q22-32
Avellino dystrophy
Chromosome 5
Lattice dystrophy I
Chromosome 5
Lattice dystrophy II
Chromosome 9q34
Macular dystrophy
Chromosome 16q22
Congenital hereditary endothelial

dystrophy
Chromosome 20
Posterior polymorphous dystrophy
Chromosome 20q11
Cornea Plana
In cornea plana, the cornea is flat with a corneal curvature
of less than 43 D. The radius of curvature may reach levels as
low as 2030 D, similar to that of the sclera.2022 Peripheral
scleralization of the cornea is almost always present, and the
condition is indistinguishable clinically from peripheral sclerocornea. The limbal landmarks are also obscured, simulating
microcornea.
In cornea plana, the anterior chamber is shallow by virtue
of the low corneal dome. Refractive abnormalities vary from
hyperopia of 7 D to myopia of 9 D, depending on the globe
dimensions and corneal curvature.23 This condition also
features concurrent anterior segment abnormalities,24 including
iris colobomas, congenital cataract, and occasional posterior
segment colobomas. The distortion of the cornea, along with
concomitant sclerocornea, leads to a decrease in corneal transparency. This nonprogressive condition is more commonly
bilateral and asymmetric. Most cases are sporadic, with both
dominant and recessive inheritance pedigrees reported.25,26
Genetic linkage analysis has mapped the gene to the long arm
of chromosome 12 (Table 43.1).27 The embryologic explanation
for sclerocornea lies in the absence of the limbal anlage. The
formation of the limbal anlage occurs between the seventh and
tenth gestational weeks, allowing neural crest mesenchymal
cells to differentiate into either sclera or cornea and to induce a
corneal curvature that exceeds the scleral.28 With its absence,
the normal interface between sclera and cornea is disrupted,
and the normal surface curvature is flattened.
Histopathologic studies of sclerocornea have revealed morphologic features resembling scleral tissue. The stroma consists
of irregularly arranged collagen brils with an increased diameter anteriorly, in contrast to the normal cornea.29
Treatment is limited to correction of any refractive error30; for
cases with signicant central corneal opacication, penetrating
keratoplasty is indicated.31 The prognosis is guarded, however,
because of a high incidence of glaucoma, a common association
with other ocular anomalies, and an increased risk of graft
allograft rejection.32,33
MESENCHYMAL DYSGENESES
The spectrum of congenital eye ndings subsumed by the term
mesenchymal dysgenesis historically has been known by a
variety of names, including mesodermal dysgenesis and anterior

segment cleavage syndrome. A number of pathogenetic theories
have been advanced, all based on concepts of anterior segment
embryogenesis. The somewhat archaic term anterior segment
cleavage syndrome, for instance, implies abnormal separation
of developing tissues (e.g., the lens vesicle).34 With increased
knowledge of ocular embryology, however, the more current
term mesenchymal dysgenesis has been devised to reflect a
developmental arrest and incomplete central migration of
neural crest cells and corneogenic mesoderm.35
Neural crest cells migrate into the developing anterior segment in three waves, contributing to the corneal endothelium36
and trabecular meshwork, stromal keratocytes, and iris, respectively. Arrest at any of these stages may bring about the
recognized clinical dysgenesis syndromes. In addition to this
developmental arrest, secondary anterior displacement of the
lens-iris diaphragm may account for other abnormalities.37,38
Whatever the exact pathogenesis, because corneal and iris
tissues are likely derived at least in part from the neural crest39
rather than from mesoderm, and because tissues of other origin
(e.g., the ectoderm-derived lens) may also be involved, this
heterogeneous group of congenital anomalies may be described
best by the broader term mesenchymal dysgeneses.40 The
mesenchymal dysgeneses may affect the periphery of the anterior segment, manifest only central pathologic changes, or affect
the entire anterior segment. For simplicity, this spectrum of
disorders can be categorized in a stepladder classication scheme
as suggested by Waring and associates (Fig. 43.2).41 Rarely,
however, a case specically conforms to only one of these entities.
Posterior Embryotoxon
The simplest dysgenesis of the anterior segment periphery
is posterior embryotoxon, the anterior displacement and
enlargement of Schwalbes line, appearing as an irregular,
circumferential ridge on the posterior surface of the cornea just
inside the limbus (Fig. 43.3). Gonioscopy shows that it juts into
the anterior chamber, and the adjacent uveal trabecular
meshwork may appear dense.41 Posterior embryotoxon occurs
in 1015% of normal eyes.42 A prominent Schwalbes line may
be associated with other disorders, including primary congenital
glaucoma,43 Alagilles syndrome (arteriohepatic dysplasia),44
megalocornea, aniridia, corectopia, and Noonans syndrome.45
Axenfelds Anomaly and Syndrome

Axenfelds anomaly results when posterior embryotoxon is
accompanied by abnormal iris strands crossing the anterior
chamber angle to attach to a prominent Schwalbes line46 (see
Fig. 43.3). If glaucoma also is present (secondary to angle
abnormality), the condition is called Axenfelds syndrome.47
Reigers Anomaly and Syndrome

Reigers anomaly is present if hypoplasia of the anterior iris
stroma is found with the changes typical of Axenfelds
anomaly.48,49 This anomaly is associated with glaucoma in
~60% of patients, which may result from incomplete development of the aqueous outflow system.50 Various systemic
associations have been described, such as Downs syndrome,
EhlersDanlos syndrome, Franceschettis syndrome, Noonans
syndrome, Marfans syndrome, oculodentodigital dysplasia, and
osteogenesis imperfecta. Reigers syndrome (see Fig. 43.3)51 is
present when the eye anomaly is accompanied by skeletal
abnormalities, such as maxillary hypoplasia, microdontia, and
other limb and spine malformations. Mutations in the PITX2
and FOXC1 genes have been identied both in Axenfeld-Rieger
syndrome as well as in Peters anomaly (Table 43.1).52
An examination that includes gonioscopy and tonometry is
essential to making the differential diagnosis and to determining
CHAPTER 43
TABLE 43.1. Genetic Linkage Analysis of Corneal Dysgeneses

and Dystrophies
499

FIGURE 43.2. Composite illustration of the
anatomic ndings in mesenchymal dysgenesis
of the ocular segment. The stepladder table
demonstrates the spectrum of anatomic
combinations of terms by which they are
commonly known. The markers in the table
indicate the corresponding anatomic
component in the illustration. The central
abnormalities occur because of focal absence
or attenuation of the endothelium.
SECTION 6
From Waring GO III, Rodrigues MM, Laibson PR:

Anterior chamber cleavage syndrome: a stepladder
classication. Surv Ophthalmol 1975; 20:3.
whether the intraocular pressure is elevated. The pneumotonometer or Tonopen is preferable to other applanation
instruments because the presence of associated corneal
abnormalities or small radius of corneal curvature may give
false intraocular pressure readings. Assessment of the optic
nerve is critical to determining the overall visual prognosis and
deciding on the course of future treatment.
Medical therapy can be useful when intraocular pressure is
particularly high and temporizing measures are needed. This
disorder has a generally poor surgical prognosis, both for
glaucoma control and for corneal opacities, if present. Achieving
a balance between chronic medications and performing surgery
is uniquely difcult. The advent of effective use of antimetabolites for ltration in children may favor of surgery when the
optic nerve is threatened signicantly. Nevertheless, this type of
treatment in children remains a substantial concern as the eyes
mature.
Posterior Keratoconus
500
Posterior keratoconus5356 has no relation to anterior keratoconus. It consists of a discrete indentation of the posterior
cornea with a variable degree of overlying stromal haze and may
represent the mildest variant of Peters anomaly. Some attribute

the cause to an abnormal migration or differentiation of the
secondary mesenchyme that normally forms the corneal
stroma.57 Posterior keratoconus tends to be sporadic, unilateral,
and relatively central. In some cases, pigment surrounds the
edges of the posterior depression, suggesting previous contact to
the iris. On histologic examination, Descemets membrane may
be thinned, with concomitant endothelial abnormalities in the
focally abnormal area55 (Fig. 43.4).
Although the irregularity of the posterior cornea may affect
vision to some extent, the anterior surface is normal unless
there is sufcient posterior thinning to cause ectasia. Rarely, the
entire posterior cornea has increased curvature.56 Because
vision usually is acceptable, keratoplasty rarely is indicated.
Congenital Central Corneal Opacity (Peters

Anomaly)
Peters anomaly is a congenital central corneal opacity with
corresponding defects in the posterior stroma, Descemets
membrane, and endothelium.34,41,58 Most cases of Peters
anomaly are sporadic, although both recessive and irregular
dominant inheritances have been described. Eighty percent of
reported cases are bilateral. Mutations have thus far been

described in four genes including PAX6 for aniridia, PITX2 and
FOXC1 for Axenfeld-Rieger syndrome, and CYP1B1 for primary
congential glaucoma.52,59
Although Peters anomaly generally is characterized by a
central corneal leukoma, two clinical variants have been
recognized.60 Peters anomaly type I (see Fig. 43.4) is almost
an extension of posterior keratoconus, showing the typical
posterior nebular opacity in the pupillary axis, with the
additional feature of iris strands that cross the anterior chamber
from the iris collarette to the margin of the posterior defect. The
lens usually remains clear and is positioned normally.
Associated anomalies, such as microcornea, sclerocornea, and
infantile glaucoma, may be present, but for the most part, no
other ocular or systemic abnormalities are present.
In Peters anomaly type II (Fig. 43.5), the lens is abnormal
either in position or in transparency, in addition to the central
corneal opacity and iridocorneal synechiae. Centrally, the posterior cornea and lens may be adherent, and there may be an
anterior polar cataract. This type more commonly is bilateral,
and almost every involved case shows severe ocular and
systemic malformations.61 Nearly 5070% of patients with

Peters anomaly have concomitant glaucoma. Other associated
ocular abnormalities include microcornea, microphthalmos,
cornea plana, sclerocornea, colobomas, aniridia, dysgenesis of
the angle and iris and persistent hyperplastic primary vitreous.
Systemic associations include developmental delay, congenital
heart disease, external ear abnormalities, central nervous
system structural abnormalities, genitourinary abnormalities,
hearing loss, cleft lip and palate and spinal defects.61
Histopathologic changes are present in all layers of the
cornea.37,38,6265 Often, the anterior changes, which include
disorganization of the epithelium, brovascular pannus, and
loss of Bowmans layer as a result of long-standing edema, are
secondary to the posterior abnormalities. Fluid lakes are also
present in the affected edematous stroma.
In the peripheral and unaffected areas, the corneal endothelium forms a continuous monolayer, and Descemets membrane is of normal, uniform thickness (~5 mm). In the area of
defect, however, endothelium and Descemets membrane can
terminate abruptly or be severely attenuated. The affected
Descemets membrane consists of multiple laminations of base-
CHAPTER 43
FIGURE 43.3. Top left and right, Posterior

embryotoxon demonstrating an anteriorly and
centrally displaced Schwalbes line. Middle,
Axenfelds anomaly. Markedly dense and
advanced Schwalbes line (left) accompanied
by adherent abnormal iris processes bridging
the anterior chamber (right). Bottom, Riegers
syndrome. Left, Multiple facial anomalies such
as telecanthus, low nasal bridge, and maxillary
hypoplasia. Right, This same patient exhibits
posterior embryotoxon, hypoplasia of the
anterior iris stroma, corectopia, and peripheral
anterior synechiae.
501
FIGURE 43.4. Posterior keratoconus. (a) Light

micrograph of a keratoplasty specimen shows
the posterior central stromal defect (between
arrowheads) devoid of Descemets membrane
and endothelium. H & E 15. (b) By scanning
electron microscopy, the posterior central
defect (asterisk) is prominently displayed and
appears lined by brous tissue (31).
(c) Higher-power scanning electron microscopy
of the posterior brous tissue shows the loose
collagenous network of this layer (310).
(d) Phase-contrast microscopy of the posterior
cornea discloses only attenuated broblastic
cells (asterisk) covering the posterior stromal
surface. PPDA250. (e) Transmission electron
microscopy of the area in (d) shows loosely
aggregated collagen brils of normal
dimensions and thin broblasts (F) (12 600).
SECTION 6
502
ment membrane-like material, with interspersed collagen brils

and ne laments. Because such abnormal material is elaborated by the corneal endothelium, a broblastic metaplasia of
the endotheliogenic mesenchyme is likely, as is thought to
occur in a number of corneal conditions in which the endothelium is similarly disturbed to secrete a posterior collagen
layer.66
The lens abnormalities in Peters anomaly are characterized
histologically by a stalk-like connection between the lens and
the posterior corneal defect, suggesting primary incomplete separation of the lens vesicle. Alternatively, there may be contact of
a morphologically intact lens to the posterior cornea, suggesting
subsequent anterior displacement of a normally developed lens.
There are several reasonable explanations for a central
corneal leukoma of the Peters anomaly variety. One is incomplete central migration of corneogenic mesenchyme (i.e.,
neural crest cells), accounting for posterior endothelial and
stromal defects.41 This is corroborated by the nding of abnormally large stromal collagen brils of 360600 in some
patients. A similar abnormality of mesenchymal development
is found in sclerocornea and congenital hereditary endothelial
dystrophy.40 Another explanation of posterior corneal leukoma
of a Peters type anomaly is an in utero subluxation of the lens,
either before or after its full development, in either case
interrupting the normal migration or function of the developing
endothelium.
Historically, the internal ulcer of von Hippel has been
grouped with Peters anomaly, but the former is probably
an intrauterine inflammatory condition rather than a true

developmental defect.40
The clinical management of these patients is complex and
difcult, and the ~35% success of keratoplasty is usually related
to the control of concomitant glaucoma.67
Sclerocornea
In sclerocornea (Fig. 43.6), the limbus is not well dened because opaque scleral tissue with ne vascular conjunctival
arcades extends into the peripheral cornea. A broad range of
corneal involvement is possible, the most extreme of which is
complete scleralization of the cornea. Ninety percent of cases
are bilateral, although the disorder generally is asymmetric.
Most cases are sporadic; there is no known heredity. Sclerocornea is nonprogressive and must be differentiated from
interstitial inflammatory conditions and arcus juvenilis
(congenital peripheral lipid deposition, also known as anterior
embryotoxon). Sclerocornea is associated with cornea plana in
~80% of patients.68 Other related ocular abnormalities include
microphthalmos, iridocorneal synechiae, persistent pupillary
membrane, dysgenesis of angle and iris, congenital glaucoma,
coloboma, and posterior embryotoxon of the fellow eye.69
Numerous sSomatic abnormalities are also associated, including mental retardation, deafness, and craniofacial, digital,
and skin abnormalities.68
Ultrastructural studies40,70,71 have shown the involved
stroma to assume the morphologic features of scleral tissue,
with irregularly arranged collagen brils of variable and im-
mensely enlarged diameter for corneal tissue (up to 1500 ,

comparable to normal scleral collagen). The precise lamellar
organization of normal corneal stroma is not present; thus,
optical clarity is not achieved. Various abnormalities of the
endothelium and Descemets membrane exist, from attenuation to focal absence. Descemets membrane generally is
thin, with multilaminar deposition of basement membrane-like
collagen.
Pathophysiologically, sclerocornea may result from developmental arrest of limbal anlage responsible for both limbal
differentiation and corneal curvature during neural crest
migration, as is seen in the other mesenchymal dysgeneses.40
be adherent to the posterior cornea. Anterior staphyloma may

result from intrauterine inflammation or maldevelopment.72 In
the latter situation, there is no histologic evidence of
inflammation, and there is failure of migration of mesenchymal
tissues that would usually form the posterior corneal structures,
iris, and angle. This maldevelopment, probably coupled with
increased intraocular pressure caused by the angle abnormality,
leads to corneal opacity and thinning and to prominent
buphthalmic enlargement of the entire anterior segment.
Hereditary cases have been reported. Prognosis for keratoplasty
as well as preservation of any functional vision is dismal.
CHAPTER 43
FIGURE 43.5. Peters anomaly. Top left,

Clinical photograph of a typical bilateral case
with large dense central leukomas that was
successfully treated by penetrating keratoplasty
with optical iridectomy of the fellow eye.
Top center, A more diffuse corneal opacity in a
7-month-old infant. Top right, Intraoperative
photograph demonstrates adhesion of the lens
to the posterior cornea as a corneal button
(grasped with forceps) is trephined. No iris
could be identied. Middle left, Light
micrograph of a corneal button showing a
posterior central depression in which lodged
the cataractous lens (L). H & E 10. Middle
center, Higher magnication of light microscopy
of the posterior cornea adjacent to the central
stromal defect demonstrates fragments of
presumed lens capsule and lens epithelium
(between arrowheads) immersed in the stromal
collagen. PAS 200. Middle right, Phasecontrast microscopy of the same cornea
resolves the thin and undulating Descemets
membrane, which terminates (arrow) at the site
of keratolenticular apposition. PPDA 250.
Bottom inset, Phase-contrast micrograph of
central area of the cornea devoid of
Descemets membrane and lined by lens
capsule (arrowhead), lens epithelium (LE), and
cataractous lens cortex (asterisk). PPDA 250.
Bottom, Transmission electron micrograph of
the same area discloses numerous broblastic
cells (F) in the posterior stroma, lined by a
uniform, 8-mm-thick lens capsule (LC) and lens
epithelium (LE) (4000).
Congenital Hereditary Endothelial Dystrophy

Congenital Anterior Staphyloma
Anterior staphyloma (Fig. 43.7) is a congenital opacity of one
or both corneas, which become protuberant, are often lined
with iris tissue, and are associated with an extremely disorganized anterior segment. As in Peters anomaly, the lens may
Given the neural crest mesenchymal origin of the corneal

endothelium, we consider congenital hereditary endothelial
dystrophy (CHED) to also represent a variant of mesenchymal
dysgenesis. Detailed discussion of the condition is presented
here in the section devoted to Corneal Endothelial Dystrophies.
503
SECTION 6
FIGURE 43.6. Sclerocornea. Top left, Moderate

corneal haze in a partially affected patient. Top
right, In this advanced bilateral case with
multiple congenital abnormalities, the entire
cornea is scleried, and the ne vascular
arcades extend centrally from the conjunctiva
and sclera. Middle left, Light micrograph of
anterior cornea shows disorganization of the
epithelium, fragmentation of Bowmans layer
(b), and interstitial vascularization (v). PPDA
350. Middle right, Transmission electron
microscopy discloses a disorganized array of
collagen brils that measure as much as three
times normal diameter (52 500). Bottom inset,
Light microscopy of the posterior cornea shows
irregularly thick and wavy stromal lamellae (S).
Descemets membrane could not be clearly
identied. PPDA 350. Bottom left,
Transmission electron micrograph of the same
area discloses rudimentary Descemets
membrane (DM) with notable absence of
endothelial cells (4000). Bottom right, Highermagnication electron micrograph of the area
circled in bottom left gure reveals multilaminar
basement membrane material interspersed with
ne laments (75 000).
CORNEAL DYSTROPHIES
ANTERIOR DYSTROPHIES
The anterior corneal dystrophies (Fig. 43.8) are conned to the
epithelium, basement membrane, and Bowmans layer.
Epithelial Basement Membrane Dystrophy

(MapDotFingerprint)
504
Disorders involving the epithelium and its basement membrane

may have a variable clinical appearance but probably involve a
common pathophysiology and clinical course. Because the
predominant abnormality involves the basement membrane
complexes that mediate the tight attachment between the
epithelium and Bowmans layer, the clinical manifestations of
these conditions predictably involve frequent recurrent erosions
and occasional persistent defects of the corneal epithelium.
The appellation of mapdotngerprint dystrophy is appropriately descriptive of the biomicroscopically visible features
of intraepithelial microcysts (dots), subepithelial ridges

(ngerprints), and geographic opacities (maps)7390 (Figs 43.9
and 43.10).7390 Family studies have revealed a probable dominant inheritance, with variable penetrance.91 Other clinical
studies are more consistent with a degeneration that is rather
highly prevalent in the general population.80
The symptoms of recurrent erosion can become prominent
in early adulthood through middle age and range from mild
early morning irritation to painful, erosive episodes. Irregular
corneal astigmatism with complaints of distortion or ghost
images also occasionally develop secondary to plaque-like
accumulations of subepithelial cellular debris, basement
membrane, and collagen.
The degree of clinical symptoms, however, often does not
parallel the extent of abnormal slit-lamp ndings. Because of
the presumed primary abnormality in the epithelial basement
membrane, even minor trauma can cause a major epithelial
breakdown, with impaired subsequent healing. In a patient who
has had a trivially traumatic or seemingly spontaneous erosive
episode, meticulous examination of the symptomatic eye, as

well as the fellow eye, should be performed in an attempt to
disclose an underlying dystrophy. Careful inspection of the
fluorescein-stained tear lm for localized irregularity, instability,
or negative staining (focal dark areas where epithelial elevation
thins the overlying tear lm thereby reducing fluorescence) as
well as retroillumination at high magnication through a
dilated pupil are helpful in uncovering these often subtle abnormalities in a patient who complains of spontaneous irritation.
Hykin and colleagues92 prospectively examined 117 patients
with histories of recurrent corneal erosions. They found that 23
had only epithelial basement membrane dystrophy with no
history of trauma. Seventy-ve patients had histories of trauma
but no slit-lamp evidence of dystrophy. Williams and Buckley93
stated that mapdotngerprint dystrophy is the most common

cause of recurrent erosion in general practice.
Many ultrastructural studies of mapdotngerprint
dystrophy have disclosed a discontinuous multilaminar,
thickened basement membrane under the abnormal
epithelium.73,88,89 This abnormal basement membrane
sometimes contains an admixture of collagenous and cellular
debris suggestive of prior breakdown episodes. More widespread
coalescence of this subepithelial material gives the clinical maplike picture. Other congurations of aberrant basement
membrane and brillar collagen can be found extending in
ridges into the epithelial layers, thus explaining the ngerprint
pattern. Epithelial microcysts actually are pseudocystic
collections of cellular and amorphous debris within the
CHAPTER 43
FIGURE 43.7. Congenital anterior staphyloma.

Top left, A 1-year-old girl was born with anterior
staphyloma of the right eye and anterior
segment mesodermal dysgenesis of the left.
The right eye shows enormous proptosis of the
enlarged and scleralized cornea. The axial
length is elongated to 24 mm as a result of
disproportionate enlargement of the anterior
segment. Top right, The left eye immediately
after penetrating keratoplasty and anterior
segment reconstruction. Middle left, Light
microscopy of keratoplasty specimen shows
secondary epithelial metaplasia into keratinized
stratied squamous epithelium. Middle center,
Involved stroma of the same specimen
assumes the morphologic features of scleral
tissue with the presence of abundant blood
vessels. H & E 75. Middle right, Transmission
electron microscopy of the corneal stroma
discloses abnormally thick (440 ) collagen
brils (43 400). Bottom inset, Light microscopy
of posterior cornea demonstrates pigmented
epithelium of the iris apposed to Descemets
membrane (asterisks). PAS 75. Bottom,
Transmission electron microscopy of this same
area discloses iris pigment epithelial cells and
stromal tissue lining the posterior corneal
surface (6400).
505

FIGURE 43.8. Characteristic corneal changes
in various types of corneal stromal dystrophy.
SECTION 6
Adapted from and courtesy of Dr A Bron. From Coney

A, Miller J, Krachmer JH: Corneal diseases. In:
Goldberg M, ed. Genetic and metabolic eye disease.
Boston: Little, Brown; 1974:283-285.
506
FIGURE 43.9. Mapdotngerprint dystrophy. Top left and center, Clinical photograph of a 42-year-old woman with nontraumatic erosions
shows characteristics of map dystrophy with supercial geographic haze interrupted by clear areas and few dots. Top right, Light microscopy of
the clinical dot pattern reveals a large debris-containing intraepithelial cyst. PPDA 400. Bottom left, Enhanced transillumination view of the dot
pattern. Bottom right, Transmission electron microscopy of the evolving cyst that results from cellular dissolution leaving residual nonspecic
cytoplasmic granular debris (asterisks) (8000).
epithelial layer. Their shape changes with time, because they

are formed from entrapped cellular material deeper within the
epithelium. As they travel to the surface, they may coalesce
with other cysts and, nally, break through the surface, giving
rise to an erosive episode.
The primary defect in mapdotngerprint dystrophy is
presumably the synthesis of abnormal basement membrane
and adhesion complexes by the dystrophic epithelium
(Fig. 43.11). Unable to form proper hemidesmosomes or
anchoring brils, the epithelium undergoes recurrent
subclinical or overt episodes of dysadhesion. This periodic liftoff allows debris to accumulate subepithelially, providing an
even less adequate substrate on which the already abnormal
basement membrane must form. Moreover, intraepithelial
extensions of abnormal basement membrane and collagenous
material may block the normal surface migration of maturing
epithelial cells, allowing the formation of encysted collections
of debris. Thus, the cycle is to a degree self-perpetuating, with

primary faulty epithelial adhesion secondarily causing abnormal epithelial maturation that, in turn, exacerbates the accumulation of abnormal basement membrane and collagenous
debris and leads to further worsening of epithelial adhesion.
Careful dbridement of severely aberrant epithelium and, in
some cases, supercial keratectomy to remove subepithelial
debris are aids to conservative therapy with lubricants,
hypertonic saline ointment, patching, or bandage soft contact
lenses. When used as prophylaxis for recurrent erosion, one
study showed no difference between bland ointment and
hypertonic saline ointment.92 McLean and associates94
recommended the use of a needle to perform anterior stromal
reinforcement or puncture. Pathologic studies of anterior
stromal puncture demonstrate activated keratocytes, new
basement membrane with type IV collagen, and production of
new bronectin and laminin.95 Excimer laser phototherapeutic
CHAPTER 43
507

FIGURE 43.10. Mapdotngerprint dystrophy.
Top left and center, Two variants of ngerprint
dystrophy show subepithelial ridges and appear
refractile against the red fundus reflection. Top
right, Under direct illumination, otherwise faintly
visible ngerprint lines are enhanced with
fluorescein staining and cobalt light. Irregular
corneal tear lm and abnormal tear breakup are
evident. Bottom, upper inset, Phase-contrast
photomicrograph illustrates a prominent
intraepithelial ngerprint extension (arrow) from
the subepithelial zone with marked
rearrangement of basal epithelium. PPDA
1200. Bottom, Transmission electron
microscopy of this same area discloses
collagenous and granular composition of the
subepithelial material, as well as cellular
elements (asterisk), and an elaborate
multilaminar basement membrane (bracketed
area) loosely apposed to an undulating basal
cell membrane (7000). Bottom, lower inset,
Higher magnication of the bracketed area in
bottom gure resolves typical redundant
laminations of the basement membrane
(asterisks), underdeveloped hemidesmosomes
(encircled areas), and absence of anchoring
brils. Ep, basal epithelium (40 000).
SECTION 6
Top left and center, Courtesy of Dr L Hirst.
FIGURE 43.11. Theorized pathogenesis of epithelial basement membrane dystrophy. Epithelial cells produce an abnormal multilaminar
basement membrane, both in the normal location and intraepithelially. As the intraepithelial basement membrane thickens, it blocks the normal
migration of epithelial cells toward the surface. Trapped epithelial cells degenerate to form intraepithelial microcysts that slowly migrate to the
surface. The abnormal basement membrane produces map and ngerprint changes, and microcysts produce the dot pattern seen clinically.
From Waring GO III, Rodrigues MM, Laibson PR, et al: Corneal dystrophies. I. Dystrophies of the epithelium, Bowmans layer and stroma. Surv Ophthalmol 1978;
23:71.
508
keratectomy (PTK) to stimulate diffuse microcicatrization at

the surface of Bowmans layer is probably the treatment of
choice when the erosion involves the visual axis.96,97
Similar ngerprint, map, and intraepithelial microcyst
changes may develop after traumatic, infectious, or ulcerative
conditions, and particularly in patients with chronic epithelial
edema where repeated liftoff of the epithelial sheet allows the
interposition of material that can again thwart the development

of proper basement membrane adhesion complexes.
Hereditary Epithelial Dystrophy (Meesmann,

StockerHolt)
The corneal dystrophy of Meesmann98104 and of StockerHolt105
is a dominantly inherited abnormality of the corneal

contain a dense intracellular substance of unknown composition.102 Older individuals may complain of mild erosive
symptoms and minimally decreased acuity.
In 1955, Stocker and Holt105 similarly described a dominantly inherited condition in patients 7 months to 70 years of
age, characterized by gray, punctate, scattered corneal opacities
that, with focal illumination, appeared as minute droplets.
Histopathologically, PAS-positive thickening of basement
membrane was present overlying a normal-appearing Bowmans
layer. In some patients, this nodular thickening of the basement
membrane produced an irregular epithelial surface.
Lisch Corneal Dystrophy

Lisch corneal dystrophy (LCD) is characterized by band-shaped
and whorled microcysts within the corneal epithelium.107 The
inheritance pattern is compatible with either X-chromosomal
or pseudoautosomal dominant transmission, and linkage with
chromosome Xp22.3 has been indentied.108 On clinical
examination, opaque grey lesions resembling whorls, bands or
strands separated by clear areas are present within the corneal
epithelium. Densely packed clumps of microcysts within the
opacities are evident on retroillumination. Vision is affected
if the lesions involve the visual axis, but corneal erosions are
infrequent. Light and electron microscopy demonstrate cycoplasmic vacuolization of all epithelial cells within the affected
areas. Removal of affected epithelium by supercial keratectomy may improve vision.
FIGURE 43.12. Hereditary epithelial dystrophy (Meesmanns; StockerHolt). Top left, Broad slit-lamp photograph discloses myriad small,
clear to gray-white punctate opacities in the interpalpebral zone. Top
right, The intraepithelial vesicles stand out with retroillumination.
Bottom, Transmission electron micrograph of the corneal epithelium
shows an intraepithelial pseudocyst containing desquamated cellular
debris (18 000).
epithelium, rst described clinically by Pameijer in 1935.106 A

possibly recessive form also has been reported. Mutations
within the genes coding for epithelial keratins, specically K3
and K12 located within chromosome 17q12, have been
described.103,104
Clinically, asymptomatic intraepithelial cysts are biomicroscopically evident within the rst months of life as myriad
small clear to gray-white punctate opacities in the interpalpebral zone of the cornea (Fig. 43.12). The cysts are uniform
in size and shape, and few may stain with fluorescein.74 Occasionally, the opacities also are noted at the level of Bowmans
layer, although histopathologically, Bowmans layer is not
abnormal. It has been demonstrated that the cysts actually are
accumulations of degenerated cellular material and basement
membrane-like debris surrounded by adjacent cells. Although
cells in Meesmanns dystrophy contain material that stains
with the periodic acid-Schiff (PAS) stain, they do not contain
excessive glycogen as was believed previously; rather, they
Corneal dystrophies primarily affecting Bowmans layer have

recently undergone reclassication. Previous clinical and
histopathologic reports were reviewed by Kuechle and colleagues,109 who proposed that the dystrophy reported by Reis110
in 1917 and subsequently described by Buecklers111 in 1949 be
renamed cornea dystrophy of Bowmans layer type 1 (CDB1)
The anterior honeycomb-shaped corneal dystrophy described by
Thiel and Behnke112 in 1967 has been classied as CDB 2. Both
CDB types are inherited as autosomal dominant trait. For
CDB 1 a mutation at R124L within the TGF-beta induced
gene on chromosome 5q31 has been reported,113 and for CDB 2
mutation at R555Q locus within the BIG-H3 gene on chromosome 5q114,115 is known (Table 43.1). Both dystrophies are
usually bilateral and symmetric and become evident in the rst
or second decade of life as painful recurrent erosive episodes.
Patients develop decreased visual acuity because of anterior
scarring and surface irregularity. It appears that the recurrent
erosions occur earlier and with fairly more marked visual loss in
patients with CDB 1117118 (Fig. 43.13).
Slit-lamp examination of the cornea shows an irregular
epithelium with diffuse, irregular, patchy geographic opacities at
the level of Bowmans layer. As time passes, central opacities
develop as a reticulated pattern spreading into the midperiphery
with a diffuse supercial stromal haze. The clinical appearance
of these dystrophies is similar, and differentiation can be made
only with light microscopy or electron microscopy.
On light microscopy, the region of Bowmans layer is replaced
with a brocellular scar tissue that has an undulating sawtoothlike conguration. This conguration is not specic to CDB
type I or II and may be seen in both variants. CDB type I stains
positively with Massons stain, whereas CDB type II is only
equivocally positive.
Transmission electron microscopy of CDB type I reveals rodlike, electron-dense paracrystalline structures at Bowmans
layer, similar to those observed in granular dystrophy.118,119
Instead, the 915 nm diameter curly bers,120 once thought
to be characteristic of ReisBcklers dystrophy, appear only in
the region of Bowmans layer in CDB type II.
CHAPTER 43
Corneal Dystrophies of Bowmans Layer
509

FIGURE 43.13. (a) Corneal dystrophy of
Bowmans layer type I. Top left, Clinical
photograph of the eye of a 20-year-old woman
with recurrent epithelial erosions who had been
followed since age 7 after being diagnosed with
ReisBcklers dystrophy. Top right, Slit-lamp
photograph shows a reticular pattern of gray
ring-like supercial opacities. Middle left inset,
Phase-contrast microscopy of supercial
keratectomy specimen reveals prominent
subepithelial deposits replacing Bowmans layer
(arrowhead). PPDA 375. Middle, Transmission
electron microscopy conrms deposits as rodlike paracrystalline structures (7500). Bottom,
High-magnication electron micrograph of
random electron-dense deposits at Bowmans
layer and supercial stroma (17 750).
Continued
SECTION 6
The pathogenesis of CDB is unknown. The primary lesion of

CDB type II may be due to fragmentation of the collagen brils
of Bowmans layer, and the epithelial lesion may occur
secondarily. Alternatively, immunofluorescent localization of
laminin and bullous pemphigoid antigen suggests a primarily
epithelial disease.121 Concomitant abnormalities in the
epithelial basement membrane account for recurrent erosive
episodes.73,122125
Treatment of these dystrophies varies from early conservative
therapy for recurrent erosions to supercial keratectomy, either
mechanical126 or by excimer laser PTK, for corneal scarring and
opacication.127,128 These methods are helpful in managing the
visual aspects of this disorder and always should be attempted
before lamellar or penetrating keratoplasty.126 Recurrences after
keratoplasty and after supercial keratectomy have been
described.129,130
Vortex Dystrophy (Fleischers)
510
The terms vortex corneal dystrophy and cornea verticillata of

Fleischer have been applied to the nding of pigmented, whorlshaped lines in the corneal epithelium.131,132 Because this same
corneal abnormality is evident in Fabrys disease, it is now

thought that these patients may have been asymptomatic
female carriers of X-linked Fabrys disease. In general, similar
whorl-like corneal lesions are evident in patients taking
chloroquine, amiodarone,133 phenothiazines, or indomethacin.
Striate melanokeratosis and ngerprint dystrophic changes can
also mimic the vortex pattern. In the absence of these factors,
however, a thorough survey of family members should be made
to exclude Fabrys disease.
Anterior Mosaic Crocodile Shagreen (Vogts)

Anterior mosaic crocodile shagreen appears as bilateral,
polygonal, grayish-white opacities in the deep layers of the
epithelium and in Bowmans layer.134,135 These opacities are
usually axial and separated by clear cornea. Because visual
acuity usually is not affected, treatment is not indicated.
Limited histologic study has revealed interruptions of Bowmans
layer and interposition of connective tissue between it and
the epithelium. It is unclear whether mosaic crocodile shagreen
is an actual corneal dystrophy or an age-related process.
A juvenile form of anterior mosaic crocodile shagreen may
occur in association with megalocornea, peripheral band keratopathy, and iris malformation. Similar changes may also arise
in posttraumatic conditions.
The so-called anterior mosaic pattern is a different entity, in
which a delicate polygonal pattern is seen after topical instillation of fluorescein. The anatomic explanation for this pattern
is not clear.
keratopathy with both childhood and senile forms has been

described without obvious associated cause. In clinical appearance, the inherited form is identical to that which occurs
secondarily (see section on Corneal Degenerations).
CHAPTER 43
FIGURE 43.13 (contd). (b) Corneal dystrophy

of Bowmans layer type II. Top left and right,
Clinical photographs of the eyes of two patients
with recurrent erosions exhibiting a diffuse
supercial haze. Middle left, Light microscopy
demonstrates the sawtooth conguration of
accumulated subepithelial material with an
irregular basal epithelial layer. H & E 220.
Middle right, Phase-contrast microscopy
reveals a prominent deposit of subepithelial
brocellular tissue (asterisk) with a distorted
Bowmans layer. PPDA 300. Bottom inset,
Phase-contrast microscopy demonstrates
degeneration of dark-staining basal cells and
fragmentation of Bowmans layer (asterisk) by
nodular brous pannus. PPDA 800. Bottom
left, Transmission electron microscopy conrms
thin remnants of a disarrayed Bowmans layer
(B) and apparent continuity (arrowheads)
between basal cell cytoplasm (Ep) and
degenerate cellular debris (D) within Bowmans
layer. Basement membrane complexes
(encircled area) are discontinuous and lack
anchoring brils (30 000). Bottom right, Highmagnication electron micrograph of brillar
deposits resolves as masses of irregular curled
laments, 6-8 nm in diameter (63 000).
STROMAL DYSTROPHIES
Granular Dystrophy (Groenouws Type I)
Idiopathic Band Keratopathy

Band-shaped keratopathy is a deposition of calcium in the
interpalpebral basal epithelium and Bowmans layer.136 Most
often, calcium deposition is secondary to a chronic ocular disease, such as uveitis, or to a systemic disease, such as hypercalcemia or chronic renal disease. An inherited type of band
Granular corneal dystrophy manifests in the rst decade of life

and is transmitted as an autosomal dominant trait (Table 43.2).
The lesions are sharply demarcated, milky, opaque gures
resembling snowflakes or bread crumbs and are conned to the
axial portion of the cornea, usually beginning in the most
supercial portion of the stroma (Fig. 43.14). During their
511
TABLE 43.2. Corneal Stromal Dystrophies

Characteristics
Granular
Lattice
Macular
Genetics
Autosomal dominant
Autosomal dominant
Autosomal recessive
Onset
Early adolescence
First decade of life
First decade
Vision
Good until middle age
Early reduction with obvious clouding Significantly reduced by 3040 years;

by 20 years; 20/200 by 50 years
finger counting by 50 years
Symptoms
Minimal inflammation and

irritation
Severe recurrent erosions
Mild recurrent erosive symptoms
Opacities
Grayish opaque granules;

bread crumbs; sharp
borders
Grayish pipe cleaner linear,

branching, threads; dots and flakes;
distinct borders
Grayish opaque spots; indistinct borders
Intervening stroma
Clear
Relatively clear
Diffusely cloudy
Distribution of opacities
Axial only; periphery clear
Entire cornea with dots; linear

opacities central; periphery usually
clear; progress to central disciform
by middle age
Entire cornea to limbus, but

most dense centrally
Histopathology
Discrete, hyaline, granulated
Large hyaline lesions with scattered

fibrillar material; also subepithelial
Diffuse, granular, nonhyaline material,

especially associated with keratocytes
Histochemistry
Masson: brilliant red

PAS: negative
Masson: redpurple
PAS: positive
Congo red: positive
Birefringent
Two-color dichroism
Thioflavin-T fluoro
Colloidal iron: positive

Alcian blue: positive
Electron microscopy
Rod-shaped, electron-dense,
crystal structure (100500 mm);
keratocytes normal; endothelium
normal
Random fibrils (80 diameter);

electron-dense; keratocytes normal;
endothelium normal
Diffuse vesicles, fibrillar material in

stroma and Descemets; keratocytes
and endothelium distended by
membrane-bound vacuoles with
fibrillogranular material
Defect
Structural protein: hyaline

degeneration of collagen?
Structural protein: primary

amyloidosis of cornea
Metabolic: defective acid

mucopolysaccharide metabolism;
localized mucopolysaccharidosis
SECTION 6
From Kenyon KR, Hersh PS, Starck T, Fogle JA: Corneal dysgeneses, dystrophies, and degenerations. In: Duane T, (ed.): Clinical ophthalmology, vol 4. Philadelphia:
JB Lippincott; 1981, p 23.
512
evolution, they may extend more posteriorly. Between the dense

opacities, the intervening cornea is characteristically clear.
Jones and Zimmerman137 noted that the opacities consist of
areas of hyaline degeneration in which stromal bers appear
granular. Histologically, the deposits stain red with Massons
trichrome stain and are less PAS-positive and less birefringent
than the normal stroma. Reticulin stains demonstrate
numerous argyrophylic bers. Using histochemical techniques,
Garner138 concluded that the deposits consist mainly of noncollagenous protein-containing tryptophan, arginine, tyramine,
and sulfur-containing amino acids and postulated that the
abnormal proteins originated from the epithelium, keratocytes,
and extracorneal sources. Rodrigues and co-workers139 found
immunofluorescent evidence of a microbrillar protein, a poorly
characterized glycoprotein, and a Luxol fast blue-staining
phospholipid. More recently, keratoepithelin deposits have been
immunohistochemically identied.140 An epithelial origin of
the deposits based on light and electron microscopic studies of
corneas with recurrent granular dystrophy has been suggested.
Transmission electron microscopy demonstrates rod-shaped or
trapezoidal extracellular structures 100500 m wide with
faintly visible periodicity. Keratocytes, endothelium, and
Descemets membrane appear unaffected.141
Several atypical variants of granular dystrophy have been
distinguished from the classic form. The rst group has been
described as the supercial variants.142146 Careful review of
the descriptions, clinical photographs, light microscopic, and
transmission electron microscopic ndings reveals that they

are likely to represent corneal dystrophies of CDB type I
(see section on Corneal Dystrophies of Bowmans Layer). The
so-called corneal dystrophy of WaardenburgJonkers143 was
later proved by WittebolPost and colleagues145 to be identical to
CDB type II. These variants have an earlier onset and more
severely decreased visual acuity than those of typical granular
dystrophy. Most have increased recurrent erosive episodes. On
clinical examination, large rings and disks within the supercial
stroma with stellate gures extending to deeper stroma characterize granular dystrophy, whereas snowflake-like opacities or
confluent rings conned to Bowmans layer forming a diffuse
supercial stromal haze characterize CDB type I.
A second variant of granular dystrophy has been described in
a group of patients tracing their ancestry to Avellino, Italy.146148
These patients exhibit an appearance similar to typical granular
dystrophy along with axial anterior stromal haze and the
presence at midstroma of discrete linear opacities. On histologic
and ultrastructural analysis, two groups of deposits are found.
The rst are in Bowmans layer and supercial stroma and
display a classic granular dystrophy staining reaction for keratoepithelin. The second exhibit lattice-like amyloid deposits.
Similar to granular and lattice corneal dystrophies, transmission is autosomal dominant and localizes to the transforming growth factor beta-induced (Big-H3) gene on
chromosome 5q31.
Linkage of granular corneal dystrophy to a locus in the region

FIGURE 43.14. Granular corneal dystrophy.
Top, Three different clinical congurations of
granular dystrophy. Top left, Densely axial
nontranslucent gray-white deposits simulating
bread crumbs. Top center, More discrete and
well-dened round and oval shapes with clear
stroma between lesions. Top right, Christmas
tree-like opacities with moderate anterior
stromal scarring. Middle left, Retroillumination
emphasizes the optical clarity of intervening
stroma between granular opacities. Middle
center, Light microscopy of irregularly shaped
hyaline deposits is accentuated with Massons
trichrome stain. (220). Middle right, Light
microscopy of a patient with severe recurrent
erosion reveals a supercial deposit evolving to
break the epithelial surface. PAS 220. Bottom
left, Transmission electron microscopy shows
relatively normal epithelium (E) and basement
membrane (arrowhead) anterior to large
electron-dense deposits (asterisk) within
Bowmans layer and anterior stroma. 15 000.
Bottom right, Higher-magnication transmission
electron microscopy of granular deposits shows
the characteristic homogeneous rod-shaped
paracrystalline structure (50 000).
5q2232 on chromosome 5 was rst established in an analysis

of 124 blood samples from a single Danish pedigree of seven
generations149 (see Table 43.1). The markers IL9 and D5S436
flanked the disease locus most closely.149 Subsequently, chromosome linkage analysis of families with lattice corneal dystrophy type I, CDB type II, and Avellino dystrophy also mapped
these disease-causing genes to the transforming growth factor
beta-induced (Big-H3) gene on chromosome 5q31119,150157 (see
Table 43.1). This suggests that either a corneal gene family
exists in this region, or that these corneal dystrophies represent
allelic heterogeneity (different mutations within the same gene
manifest as different phenotypes) of the fundamentally same
disease.
Granular dystrophy does not require keratoplasty as often as
the other familial dystrophies because visual acuity may remain
adequate if clear spaces in the stroma coincide with the visual
axis. Recurrent erosions may occur when deposits involve the
basement membrane zone, but this happens less commonly
than in lattice dystrophy. If vision is reduced markedly, the
surgical management varies based on the depth and extent of
the stromal lesions. If the opacities are extremely supercial,
then supercial keratectomy or lamellar keratectomy can be
performed.158 Excimer laser PTK has been successfully used

to treat supercial granular dystrophy.159162 A preoperative
myopic refractive error is desirable because a shift toward
hyperopia has been reported after such treatment.162165 When
deep stromal lesions occur, the treatment of choice is either
deep lamellar or penetrating keratoplasty.
As in the other familial dystrophies, recurrence in the graft
(usually anterior and peripheral) or after supercial keratectomy150 may take place several years later, suggesting that
the granular deposits are either the result of some acquired
metabolic disturbance in the transplanted corneal tissue or the
product of abnormal epithelium.166169 Interestingly, LASIK is
apparently specically contraindicated in Avellino dystrophy
due to the postoperative increase of deposits within the
interface and posterior stroma.170
CHAPTER 43
Middle left, Courtesy of Dr Lawrence Hirst.
Lattice Dystrophy
Lattice dystrophy (see Table 43.2) is an autosomal dominant
condition characterized by pathognomonic branching
pipestem lattice gures within the stroma (Fig. 43.15).
Symptoms usually begin in the rst decade of life and include
decreased vision as well as recurrent erosions because of
513

FIGURE 43.15. Lattice corneal dystrophy. Top
left and center, Slit-lamp photography
demonstrates pathognomonic branching lattice
gures throughout the stroma. Top right, Light
microscopy of a cornea in a patient with
multiple episodes of recurrent erosions
discloses an irregular epithelial layer, partial
absence of Bowmans layer (arrowhead), and
predominantly subepithelial amyloid deposits
(asterisk). PAS 220. Middle left, Congo red
stain of a fusiform lesion that distorts the
normal stromal lamellar architecture (55).
Middle center, Corneal amyloid shows
birefringence and dichroism under the
polarizing microscope (20). Middle right,
Transmission electron microscopy of basement
membrane complexes reveals basement
membrane irregularity and discontinuity
resulting from underlying amyloid brils
(21 300). Bottom left, Transmission electron
micrograph of stroma shows normal collagen
brils and keratocytes with electron-dense
material abnormally dispersed extracellularly.
16 000. Bottom right, High-magnication
transmission electron micrograph resolves
lattice material as masses of ne amyloid brils,
80-100 in diameter (circled area) (43 400).
SECTION 6
Top left, Courtesy of Dr W. J. Stark.
514
subepithelial and stromal accumulations of amyloid material.

In time, the condition progresses to involve marked opacication of the axial stroma, as well as of the supercial layers,
leaving the limbus relatively free. At this stage, because the
cornea also shows a supercial haze, it becomes difcult
to visualize typical lattice lesions, and hence examination of
younger affected family members is useful. Amyloid accumulation under the epithelium gives rise to poor epithelialstromal adhesion with consequent recurrent erosion
syndrome.73 The dystrophy advances inexorably, and by 40
years of age or earlier, these problems become markedly aggravated, causing considerable discomfort and visual incapacity.
Many published reports have documented the nature of the
corneal deposits in lattice dystrophy. In 1961, Jones and Zimmerman137 and others suggested that the disorder was due to
amyloid degeneration of the stromal collagen bers. In 1967,
Klintworth171 conrmed that the disorder was a familial form of
amyloidosis limited to the cornea and showed that the brillar
material stained with Congo red and exhibited the birefringence
and dichroism typical of amyloid. On transmission electron
microscopy, the ne, electron-dense brils, 80100 in

diameter, are similar to those of known amyloid brils. Using
fluorescence microscopy, staining with thioflavin-T is helpful in
further characterizing the amyloid material, as are immunofluorescent studies using antihuman amyloid antisera.172
Evaluation of corneas with typical lattice dystrophy has demonstrated the presence of the amyloid P (AP) component, but
staining for amyloid A (AA) protein has remained controversial.173175 The corneal endothelium and Descemets
membrane are not involved. Moreover, amyloid deposits have
not been found in other excised tissues from patients with
typical lattice dystrophy.171
The specic cause of the amyloid deposits is unclear. They
may be secondary to collagen degeneration, perhaps from
lysosomal enzymes elaborated by abnormal keratocytes. An
alternative theory holds that abnormal keratocytes actually
produce the abnormal amyloid substance, although this process
is not ultrastructurally evident.
Classic lattice corneal dystrophy (LCD type I), granular
dystrophy, and Avellino dystrophy have been independently
CHAPTER 43
FIGURE 43.16. Lattice corneal dystrophy and

systemic amyloidosis (Meretojas syndrome).
Top left, A 73-year-old woman with typical
mask-like facies, including skin thickening,
prominent blepharochalasis, depressed
eyebrows, and bilateral facial nerve palsies. Top
right, Slit-lamp view of lattice lines beginning at
the periphery and sparing the visual axis.
Middle left, Light microscopy of conjunctival
biopsy shows continuous subepithelial layer
(asterisks) of extracellular material. PPDA 300.
Middle right, Transmission electron microscopy
of this biopsy demonstrates masses of ne
amyloid brils (asterisk) beneath the epithelial
basement membrane (arrows). E, epithelium
(8700). Bottom left, Transmission electron
microscopy of skin biopsy reveals deposition of
extracellular material (asterisk) immediately
beneath the normal epithelial basement
membrane (8100). Bottom right, Similar
deposits (asterisks) are found associated with
the perineurium and endoneurium of peripheral
nerves (12 060).
linked to chromosome 5q150151 (see Table 43.1). Folberg and

associates176 suggested that the morphologic distinction between LCD type I and granular corneal dystrophy is not as clear
as previously believed. They found evidence of granular deposits
in LCD type I families, and vice versa. This evidence of
histologic overlap strongly suggests that these dystrophies are
caused by mutations within the same gene.
Treatment is symptomatic, depending on visual acuity and
patient discomfort. Recurrent erosions are treated either conventionally or with supercial keratectomy to remove sub-
epithelial amyloid accumulations. Excimer laser PTK has been

described as an optional treatment for recurrent erosions and
supercial opacities.159,160,162 Penetrating keratoplasty in this
condition carries an excellent prognosis, although recurrence of
the dystrophy in the graft may take place.177178
In LCD type II (also termed Meretojas syndrome or amyloidotic polyneuropathy type IV, Fig 43.16) rst described by
Meretoja in 1969 in a large series of Finnish patients, is
systemic amyloidosis associated with lattice dystrophy.179183
The onset of clinical corneal changes usually occurs later, and
515
SECTION 6
516
FIGURE 43.17. Lattice corneal dystrophy variants. Top left and right, Clinical appearance of two patients with axial thick ropy lattice lines and a
dense nodular opacication. Bottom left, Numerous thinner lines are more easily observed by retroillumination. Bottom right, Light microscopy of
a unique large amyloid deposit at the posterior half of the cornea. Bowmans layer and epithelium are intact. H & E 170.
Bottom right, From Hida T, Tsubota K, Kigasawa T, et al: Clinical features of a newly recognized type of lattice corneal dystrophy. Am J Ophthalmol 1987; 104:241248.
Copyright by The Ophthalmic Publishing Company.
erosive episodes are less common. Systemic manifestations

include progressive cranial and peripheral neuropathy and skin
changes, such as lichen amyloidosis and cutis laxa, resulting
in mask-like facies. Other variable features include polycythemia vera and ventricular hypertrophy. Biomicroscopically,
the lattice lines are fewer, thicker, more radially oriented and
involve mainly the periphery of the cornea, with relative central
sparing. Amorphous deposits are fewer and more conned in
distribution than they are in LCD type I. Open-angle glaucoma
and pseudoexfoliation with or without glaucoma are found
commonly.183
Histologic examination of the LCD type II cornea reveals
characteristic amyloid deposits forming a layer beneath a
normal-appearing Bowmans layer and focally within the
stroma. Deposits also may be found in arteries, basement mem-
branes, skin, peripheral nerves, and sclera. The amyloid in this

systemic disorder may differ from classic lattice dystrophy,
showing loss of Congo red staining after treatment with permanganate.174
With the aid of immunohistochemistry, LCD type II can be
diagnosed and differentiated from type I in tissue sections, even
retrospectively.184186 Using antibodies raised to a chymotryptic
fragment inclusive of the carboxy terminal half of gelsolin as
well as adjacent to and inclusive of the codon 187 mutant 7
11 kDa fragment, immunoreactivity was detected in the skin
and conjunctival amyloid in LCD type II.184 The amyloid within
the cornea in type II reacted nonhomogeneously with the
antigelsolin antibody but not with the antibodies produced to
the amino and carboxy terminals of gelsolin.184 The mutation
involves a guanine-to-adenine substitution at nucleotide 654,
TABLE 43.3. Comparison of Inherited Varieties of Corneal Amyloidosis

Characteristics
Lattice Corneal Dystrophy
Familial Subepithelial
Amyloidosis
Type I
Type II
Type III
Usual age at onset
<10 years
>20 years
>40 years
<20 years
Visual acuity
Markedly impaired by age

4060 years
Usually good until after age

65 years
Impaired after 60 years
Markedly impaired by
age 1030 years
Systemic amyloidosis
No
Yes
No
No
Mode of inheritance
Autosomal dominant
Autosomal dominant
Autosomal recessive?
Autosomal recessive
Facies
Normal
Masklike facial expression,

blepharochalasis, floppy
ears, protruding lips
Normal
Normal
Nervous system
Normal
Cranial and peripheral nerve

palsies
Normal
Normal
Skin
Normal
Dry, itchy, and lax with

amyloid deposits
Normal
Normal
Cornea
Delicate interdigitating
network of filaments;
no lines present at early
stage; lines difficult to see
at late stage
Thick and radially oriented

lines
Thick lines
Multiple prominent
subepithelial nodules
Episodic corneal erosion
Yes
Yes
No
No
resulting in an asparagine-187 variant of gelsolin.182,186190 The

gelsolin gene (type II) has been localized to the long arm of
chromosome 9 (9q34)189 (see Table 43.1).
Other atypical variants of lattice dystrophy, as well as rare
cases of unilateral lattice dystrophy, have also been
reported191192 (Fig. 43.17). The former, termed LCD type III,
characterized by a probable autosomal recessive inheritance
pattern, has thicker lattice deposits within the corneal
epithelium, onset later in life without systemic involvement or
episodic recurrent corneal erosions. Histologically, there is
absence of subepithelial deposits with a normal epithelium and
Bowmans layer. The stromal deposits are larger than in LCD
types I and II, which correlates with the thicker lattice lines
clinically evident in this variant. Immunohistochemical
analysis has revealed positive staining for AP protein but only
weak staining for AA protein.191 Families with LCD types IIIA,
IV, VI, and VII have also been described.193 Confocal microscopy
may be helpful in distinguishing amyloid deposits and nerve
devenerataion in LCD from other entities, such as infectious
crystalline dystrophy, ananthamoeba and fungal hyphae.194,195
The cornea may also develop secondary amyloid deposits
after various chronic ocular diseases, but such deposits
generally are insignicant clinically (see section on Corneal
Degenerations). The differences between the inherited varieties
of corneal amyloidosis are summarized in Table 43.3.
Macular Dystrophy (Groenouws Type II)

Among the classic corneal dystrophies, macular corneal
dystrophy (MCD), unlike granular and lattice dystrophies, is an
autosomal recessive disorder and is much less common (see
Table 43.2). It usually begins in the rst decade of life and leads
to progressive visual deterioration as the stroma becomes
generally cloudy, with superimposed dense, gray-white spots
(Fig. 43.18). Unlike granular dystrophy, these macular spots
have indenite edges, and the intervening stroma is not clear.
Young patients exhibit axial lesions in the supercial layers of
the cornea, but with time, lesions approach the periphery and
extend throughout the entire stromal thickness. Corneal thinning conrmed by central pachometry has been documented.196
Also, unique to macular corneal dystrophy is primary involvement of the endothelium as evidenced clinically by the presence
of guttate changes of Descemets membrane (Fig. 43.19).
The lesions in macular corneal dystrophy stain intensely
with alcian blue and colloidal iron, minimally with PAS, and
not at all with Massons trichrome. Birefringence is decreased.
The lesions have been histochemically identied as an abnormal
keratan sulfate-like glycosaminoglycan that accumulates
extracellularly within the stroma and Descemets membrane
and intracellularly within keratocytes and endothelium.197
As consistent with an autosomal recessively inherited
condition, macular dystrophy presumably results from
deciency of a hydrolytic enzyme (sulfotransferase) and may
thus be considered a localized mucopolysaccharidosis.198,199
The effect of altered glycosaminoglycan metabolism is evident
at the cellular level; on transmission electron microscopy,
keratocytes and endothelial cells exhibit distention of roughsurfaced endoplasmic reticulum cisternae. With the acridine
orange technique, compensatory generalized hyperactivity of
the lysosomal enzyme system has been demonstrated. Eventually, the accumulated undigested storage products engorge
the cells, and the cells ultimately degenerate or rupture. The
derivation of these intracytoplasmic storage vacuoles from
endoplasmic reticulum suggests that the biochemical lesion in
macular dystrophy occurs at a different metabolic location than
in the systemic mucopolysaccharidoses because in the latter,
storage products accumulate within lysosome-like intracytoplasmic vacuoles associated with the Golgi complex.199
Snip and associates200 were able to determine that the storage
phenomenon affecting endothelium and Descemets membrane
is likely also primary because the intracellular and extracellular
lesions appear ultrastructurally comparable with those evident
in the keratocytes and stroma.
Two subtypes of MCD have been immunohistochemically
identied.201203 MCD type I is the most prevalent and is
CHAPTER 43
From Hida K, Tsubota Kigasawa K, et al: Clinical features of a newly recognized type of lattice dystrophy. Am J Ophthalmol 1987; 104:241248, 1987. Published with
permission from The American Journal of Ophthalmology. Copyright by The Ophthalmologic Publishing Company.
517
SECTION 6
FIGURE 43.18. Macular corneal dystrophy. Top

left, Clinical appearance of cornea features
diffuse stromal haze with a ground-glass
appearance extending to the limbus. Centrally,
superimposed dense gray-white spots with
indistinct edges are observed. Top right, In a
clinical variant, larger, amorphous lesions
involve only the central cornea, sparing the
limbus. Bottom, right inset, By phase-contrast
microscopy, the epithelium is seen to be
irregular. Fibrocellular pannus intervenes
between the epithelium and the unaffected
Bowmans layer (asterisks). Several extensively
vacuolated keratocytes are evident in the
anterior stroma. PPDA 250. Bottom, main
gure, Transmission electron microscopy
demonstrates irregular thinning and breaks
(arrowheads) of basement membrane. Within
the fragmented Bowmans layer, subepithelial
cells are distended by membrane-limited
intracytoplasmic inclusions containing ne
granular and reticular material (12 000).
Bottom, left inset, Higher-magnication
transmission electron microscopy resolves the
reticular pattern of accumulated intracellular
material (43 500).
518
characterized by the absence of antigenic keratan sulfate (aKS)

from both cornea and serum. In fact, it may represent a more
widespread systemic disorder of keratan sulfate metabolism.201,202 In MCD type II, the serum aKS level is normal,
and the corneal accumulations react with the antikeratan sulfate antibody.204,205 MCD types I and II cannot be distinguished
based on clinical characteristics. Moreover, some genealogic
studies have shown patients of both types sharing common
ancestors.202 Linkage analysis has mapped MCD type I to the
16q22 locus of chromosome 16206 (see Table 43.1). In addition,
a peak LOD score of 2.50 at a recombination fraction of 0.00
was obtained for the type II families by use of the identical
marker. These ndings raise the possibility that MCD types I
and II may be due to the same genetic locus.206 Recent studies
have identied a variety of mutations in a carbohydrate sulfotransferase gene (CHST6) encoding corneal glucosamine
N-acetyl-6-transferase.207210
Treatment of macular dystrophy is either deep lamellar
or penetrating corneal transplantation. In Saudi Arabia, MCD
accounts for 87% of penetrating keratoplasties performed for
classic corneal dystrophies,211 indicative of a remarkably high
prevalence of the affected gene within a dened population. As
with most other stromal dystrophies, recurrence in the graft has

been reported.212,213
Polymorphic Stromal Dystrophy

Polymorphic stromal dystrophy is a manifestation of corneal
amyloid clinically distinct from the lattice dystrophies and
gelatinous drop-like dystrophy. Thomsitt and Bron214 described
patients with a variety of posterior stromal opacities consistent
with the dystrophic changes reported in 1939 by Pillat.215 Axial
polymorphic opacities resembling stars and snowflakes were
noted, as well as branching lamentous opacities in the posterior cornea. The lesions were gray-white and mildly refractile
when viewed with direct light but appeared transparent in
retroillumination. As the intervening stroma is clear, vision is
minimally affected. Histochemistry and electron microscopy
demonstrate typical amyloid deposits within the opacities.216
The late onset, lack of progression, and absence of familial
association distinguish this condition from LCD.
Gelatinous Drop-Like Dystrophy

Gelatinous drop-like dystrophy (Fig. 43.20) is yet another
clinical manifestation of primary, localized corneal amyloidosis
CHAPTER 43
FIGURE 43.19. Macular corneal dystrophy. Top

right inset, By light microscopy, a large
extracellular accumulation of abnormal material
(large asterisk) is shown. Descemets
membrane (bracketed area) is enormously
thickened to ~40 m with additional guttate
excrescences (small asterisk). Endothelial cells
(arrowheads) are greatly reduced in number and
remarkably attenuated. Alcian blue 500. Main
gure, Transmission electron microscopy
demonstrates extensive Descemets membrane
abnormalities. The anterior banded portion
(bracketed area) is of normal thickness and
conguration. Posteriorly, Descemets
membrane (DM) is thickened, here measuring
20 m, and is honeycombed by ne granular
material anteriorly with banded collagen gures
(circled area) and multiple basement membrane
laminations predominantly posteriorly. A
degenerating cellular process is evident
centrally (asterisk) and an attenuated
endothelial cell (E) is shown posteriorly. S,
stroma (8900). Bottom left insets, Highmagnication transmission electron microscopy
(left) and scanning micrograph (right) of area
circled in main gure reveal fusiform longspacing collagen congurations with ~1100-
macroperiodicity (27 000; 10 000).
FIGURE 43.20. Gelatinous drop-like dystrophy. Central mulberry-like opacity has protuberant subepithelial mounds that appear white on focal
illumination (left) and semitransparent on retroillumination (right). Minor stromal neovascularization is present superonasally.
519
SECTION 6
520
FIGURE 43.21. Central crystalline dystrophy (Schnyders). Top left, Clinical appearance of the eye of an 8-year-old boy includes an axial ringshaped opacity formed by densely packed, ne, needle-shaped polychromatic crystals. Associated genu valgum was present in this pedigree.
Top right, In a different variant, more extensive involvement of the central cornea and associated arcus senilis was present. Bottom inset, Frozen
section of keratoplasty specimen of supercial cornea reveals the epithelium (a) unstained, Bowmans layer (b) with intense lipid deposits, and
stroma (s) with scattered lipid staining. Oil red O stain 350. Bottom left, Electron micrograph of basal epithelium and Bowmans layer reveals
vacuolated corneal epithelium (E), thickened basement membrane (arrows), and distorted, vacuolated Bowmans zone (B) with polygonal proles
(10 000). Bottom right, High-magnication transmission electron microscopy of the same area discloses multiple polygonal spaces (asterisks),
typical of cholesterol crystal ghosts (25 000).
Bottom, From Burns RP, Connor W, Gipson I: Cholesterol turnover in hereditary crystalline corneal dystrophy of Schnyder. Trans Am Ophthalmol Soc 1978; 76:184.
Central Crystalline Dystrophy (Schnyder)

Central crystalline dystrophy, an autosomal dominantly
inherited disorder that occurs in early life and is occasionally
congenital, was initially described by Schnyder in 1929229
(Fig. 43.21). Previously considered to be nonprogressive after
childhood, subsequent reports have documented signicant
progression.230 The responsible gene is located in chromosome
1 (1p34.136)231 and as the B120 gene located within this area
is associated with lipid production within broblasts, it is
speculated that lipid production by keratocytes may be
causative.232
The main feature of the disease is a bilateral, axial, ringshaped corneal opacity consisting of polychromatic crystals.
The yellow-white opacity is noted in Bowmans layer and the

anterior stroma. The epithelium is normal, and the uninvolved
stroma also appears normal, although in time a diffuse stromal
haze can develop.233 In some patients, small white opacities
are scattered throughout the stroma. Schnyders dystrophy
without crystals is not uncommon and is usually unrecognized.234 Histologic examination using lipid stains on
frozen sections reveals neutral fats and cholesterol.235 The
clinically apparent crystals correspond to cholesterol accumulations, both within keratocytes and extracellularly. Neutral
fat is distributed within the stroma among the collagen brils.
Both the limbal girdle of Vogt and corneal arcus may also be
associated. The disease is presumably a localized defect of lipid
metabolism; systemic hypercholesterolemia, xanthelasma, and
genu valgum can occur. Hence it is important to perform cholesterol and lipid studies to detect elevated serum lipid levels
and concomitant cardiovascular disease.236
Excimer laser PTK may have a potential role in patients with
signicant decreased visual acuity consequent to anterior
stromal opacities.237 Some patients with more severe opacities
may require corneal grafting. Recurrence of cholesterol crystals
may occur in lamellar or penetrating grafts.235
Marginal Crystalline Dystrophy (Bietti)

Bietti238 described crystalline deposits in the paralimbal anterior
corneal stroma in conjunction with intraretinal crystals and
retinitis pigmentosa (RP). Inheritance is autosomal recessive.
Mutations within the CYP4V2 gene of chromosome 4q354qtel
may adversely affect fatty acid metabolism within ocular structures.239 In a series of 200 patients, the prevalence of the
dystrophy was 3% in nonsyndromic RP and 10% in autosomal
RP.240
Yellowish-white subepithelial crystals are typically found in
the corneal periphery, especially at the superior and inferior
limbus.241,242 They may be overlooked due to their small size
(less than 15 mm), are asymptomatic and do not affect vision.
Crystals are also located within all layers of the retina with
atrophy of the retinal pigment epithelium and choroidal
sclerosis. EOG and ERG deterioration precede vision loss and
nyctalopia by years. Lipid inclusions and crystals are found
within keratocytes and conjunctival broblasts, as well as
within choroidal broblasts and circulating lymphocytes.
Currently there is no treatment.
Central Cloudy Dystrophy (Franois)

Franois243 described eight patients having polygonal opacities
with intervening clear zones resembling cracks of the axial
posterior stroma superimposed on nebular stromal haze
(Fig. 43.22). Strachan reported involvement in three consecutive
FIGURE 43.22. Central cloudy dystrophy
(Franois). This 62-year-old man with visual
acuity of 20/30 in both eyes demonstrates
clouding of the central cornea into segmental
areas of opacication with intervening clear
tissue (left). Slit-lamp view (right) features
mainly posterior opacities that extend forward
but become much less dense.
CHAPTER 43
characterized by deposition of large amounts of amyloid

beneath the corneal epithelium. First reported by Nakaizumi217
in Japanese patients, this dystrophy remains common in Japan
(estimated prevalence 1 in 33 000) but is rare in Western
countries. The disorder is bilateral and noninflammatory and
may exhibit autosomal recessive inheritance.218220 The responsible gene has been identied as membrane component,
chromosome 1, surface marker 1 (M1S1) on the short arm of
chromosome 1.221223
Photophobia and epiphora commence in the rst decade of
life with progressive visual deterioration during the teens. The
clinical presentation is bilateral but often asymmetrical as
multiple supercial yellow to milky white mulberry-like focal
nodular depositions elevating the epithelium and supercial
stroma. Other variations include band keratopathy and either
localized or more diffuse opacity. In longstanding cases, corneal
sensation may be decreased, as stromal vascularization and
scarring lead to severe vision loss.
Histopathologic specimens have demonstrated mounds of
amyloid interposed between the irregular epithelium and degenerated Bowmans layer as well as fusiform LCD-like deposits
in the deeper stroma.224225 The specic type of corneal amyloid
remains to be determined, but immunostaining of the deposits
is mildly positive for amyloid AL and AP, as well as lactoferrin,
apolipoproteins J and E and gelsolin but negative for amyloid
AA, AF, AB, and keratin.226
Corneal opacities and scars usually necessitate supercial
keratectomy in mild cases and either lamellar or penetrating
keratoplasty in more advanced conditions, typically by age 30.
However, recurrence is common as subepithelial haze evolves
into nodular deposits.227 Hypothesizing that a dystrophic corneal epithelium is responsible for secretion of the amyloid
accumulations, Shimazaki combined limbal stem cell transplantation with keratoplasty with seemingly improved prolongation of corneal clarity.228
521

FIGURE 43.23. Posterior mosaic crocodile
shagreen (Vogt). Clinical photography (left) of a
55-year-old asymptomatic woman
demonstrates bilateral central opacication
compromising the entire corneal thickness.
Broad slit-lamp photography (right) discloses
multiple small, fluffy, and indistinct grayish
areas in a polygonal pattern separated by clear,
crack-like zones.
generations and suggested autosomal dominant inheritance.244

Vision is affected minimally, and as patients remain asyptomatic, no therapy is required. Confocal microscopy reveals
refractile keratocytes and extracellular matrices separated by
dark striae.245 Histopathology is limited, but alcian bluepositive mucopolysaccharides and lipid-like inclusions within
the stroma by light and electron microscopy have been
described.246 The differential diagnosis of such polymorphous
stromal opacities includes posterior crocodile shagreen, macular
corneal dystrophy, icthyosis, fleck corneal dystrophy, lecithin
cholesterol acetyltransferase deciency, GraysonWilbrandt
pre-Descemet dystrophy and systemic mucopolysaccharidosis.
SECTION 6
Posterior Amorphous Stromal Dystrophy

This autosomal dominant disorder was rst described in 1977
in a family spanning three generations as symmetric gray-white,
sheet-like posterior stromal opacities centrally and extending
peripherally to the limbus.247 Corneal thinning was also present
in the more advanced cases. Findings in other reported pedigrees included: (1) both centroperipheral and peripheral forms;
(2) hyperopia with corneal flattening; (3) iris abnormalities,
including glassy sheets on the iris surface, corectopia, and
pseudopolycoria; and (4) iris processes extending to Schwalbes
line.248
The sheet-like opacities may be irregular and broken with
clear intervening stroma. Descemets membrane and endothelium may be indented by the opacities, and focal endothelial
abnormalities have been observed. Despite high astigmatism
and amblyopia, vision is usually only mildly affected.
In the keratoplasty specimen from a 5-year-old child,
fracturing of the posterior stromal collagen lamellae, a thin
Descemets membrane, and focal attenuation of endothelial
cells were evident.249 Ultrastructural studies showed disorganization of the posterior stromal collagen. Because the cornea is
structurally abnormal and is thin and flat, the opacities appear
stable throughout life, the iris is affected, and the changes have
been found in a child as young as 6 months of age, this disorder
may be more appropriately classied as a mesenchymal dysgenesis rather than a dystrophy250 (see section on Mesenchymal
Dysgeneses).
Congenital Hereditary Stromal Dystrophy
522
Congenital hereditary stromal dystrophy is characterized by

bilateral flaky or feathery clouding of the stroma, present either
at birth or within the rst years of life.251 Both the peripheral
and the central cornea are affected, the latter more severely. It
is autosomal dominantly inherited. Genome-wide screening of
three generations of a family showed linkage to chromosome
12q22. Mutations have been described in the DCN gene which

encodes for decorin, a dermatan sulfate proteoglycan which is
important in collagen brillogenesis.252
Electron microscopy has revealed abnormally small stromal
collagen brils with disordered lamellae, suggesting a disorder
in collagen brogenesis. Penetrating keratoplasty is usually
successful but recurrence is possible.
Posterior Mosaic Crocodile Shagreen

Posterior crocodile shagreen (Fig. 43.23) is a bilateral condition
marked by a series of small gray polygonal patches of various
sizes, separated by dark regions, at the level of Descemets
membrane.253 The condition may be a variant of central cloudy
dystrophy of Francois. Transmission electron microscopic
studies have demonstrated the grayish opacities of posterior
mosaic crocodile shagreen to correspond with sawtooth-like
congurations of the corneal collagen lamellae (see section on
Anterior Mosaic Crocodile Shagreen). Because vision is not
compromised, no treatment is required.
Fleck Dystrophy (FranoisNeetens)

First described by Francois and Neetens in 1957,254 fleck
dystrophy is a rare, autosomal dominant disorder is detectable
early in life and congenital in some patients255,256 (Fig. 43.24).
The genetic defect occurs on chromosome 2q35 where the
PIP5K3 gene, a member of the phosphoinositide 3-kinase
family, regulates the function of multivesicular bodies within
endosomes.257,258 Subtle grayish specks are present in all layers
of both corneas, and some appear as rings with relatively less
opacied centers. As the stroma is otherwise clear, patients
have no visual disability apart from mild photophobia. Confocal
microscopy demonstates highly reflective material within heratocytes. Histopathologic examination has revealed abnormal
keratocytes that on transmission electron microscopy contain a
brillogranular substance within intracytoplasmic vacuoles.259
Histochemical staining shows glycosaminoglycans and lipids
within these vacuoles.
PRE-DESCEMETS DYSTROPHIES
The pre-Descemets category of dystrophy has several rare
entities, all generally compatible with good vision and comfort.
A clear pattern of heredity is not always obvious.
Cornea Farinata
Cornea farinata260 is often a routine nding in older people and
therefore may represent a degenerative process rather than a
dystrophic one. Visual acuity is not usually decreased. Small

FIGURE 43.24. Fleck dystrophy (FranoisNeetens). Top, Retroillumination (left) and slitlamp view (right) demonstrate discrete flattened
white flecks with comma, wreath, or dot
conguration present throughout the entire
stroma. Bottom inset, Light microscopy of the
posterior cornea illustrates positive staining for
acid mucopolysaccharide limited to a swollen
keratocyte (circled area). Colloidal iron 500.
Bottom, Transmission electron microscopy of a
markedly vacuolated keratocyte lled with
brillogranular (F) or lipid (L) substances. There
are no extracellular abnormalities except for an
accumulation of the ne granular material
(asterisk) and occasional foci of long-spacing
collagen (square) (14 400).
gray punctate opacities can be seen in the pre-Descemets

membrane area of the stroma on retroillumination. Sometimes,
larger and more polymorphous types of comma, circular, linear,
liform, and dot-like opacities are observed as well. The
opacities may be distributed axially or annularly. Similar preDescemets opacities may be found in association with
ichthyosis.261
Deep Filiform Dystrophy

The deep liform dystrophy of Maeder and Danis264 consists of
multiple liform gray opacities in the pre-Descemets area that
affect the entire width of the cornea except for the perilimbal
region. The original case occurred in a middle-aged woman with
keratoconus. The histopathology has not been documented.
This disorder may represent a degeneration rather than a
dystrophy.
CHAPTER 43
All, From Nicholson DH, Green WR, Cross HE, et al:

A clinical and histopathological study of FrancoisNeetens speckled corneal dystrophy. Am J Ophthalmol
1977; 83:554560. Copyright by The Ophthalmic
Publishing Company.
GraysonWilbrandt Dystrophy
Grayson and Wilbrandt262 described asymptomatic opacities
that were slightly larger and more diffusely scattered than those
in cornea farinata and that were distributed axially and
paraaxially (Fig. 43.25). Familial associations were documented.
Curran and associates263 described the ultrastructure of
abnormal keratocytes anterior to Descemets membrane as
containing membrane-bound intracytoplasmic vacuoles that
included brillogranular material and electron-dense lamellar
lipid bodies.
ENDOTHELIAL DYSTROPHIES
Congenital Hereditary Endothelial Dystrophy
Initially described by Maumenee265 in 1960, this congenital
disorder of the endothelium has been mapped in its autosomal
dominant form to the pericentromeric area of chromosome 20,
within the same region assigned to posterior polymorphous
dystrophy whereas its autosomal recessive form has been
localized to a distinct area of chromosome 20p266267 (see Table
523
SECTION 6
FIGURE 43.25. GraysonWilbrandt dystrophy.

Top left, Slit-lamp photograph shows discrete
pleomorphic opacities in the pre-Descemet
area that have comma-shaped, circular, linear,
liform, and dot-like congurations. The
intervening stroma is clear. Top right, Phasecontrast microscopy demonstrates the refractile
vacuolar inclusions (arrows) within a deep
keratocyte. Descemets membrane (bracketed
area) is uniformly normal, and endothelial cells
(E) are artifactitiously vacuolated. Toluidine blue
micrograph of a keratocyte lled with vacuoles
that have clear to brillogranular material (F),
pleomorphic substances (arrowhead), and dark
electron-dense bodies (asterisk). The
surrounding stroma (S) is normal (12 000).
Inset, High-magnication transmission electron
micrograph resolves pleomorphism of
accumulated material and the presence of
membranous lamellas (arrow) (40 000).
524
43.1). Congenital hereditary endothelial dystrophy (CHED) is

usually characterized clinically by diffuse, bilaterally symmetric
corneal edema (Fig. 43.26).268 The autosomal recessive variety
(CHED 2) is present at birth and is relatively nonprogressive.
Symptoms of discomfort are not prominent despite profound
epithelial and stromal edema. Nystagmus is common. A
dominantly inherited form (CHED 1) is less severe, developing
in the rst or second year of life, and in contrast to the recessive
variety, progressive photophobia and tearing are the initial
symptoms. Nystagmus generally is absent. Some CHED
patients have no affected family members and may represent
autosomal recessive cases or a new mutation. CHED has also
been sporadically associated with sensorineural hearing loss, nail
anomalies, corneal amyloidosis and corpus callosum agenesis.
As in all instances of congenital corneal clouding, it is important to rule out congenital glaucoma. Corneal diameters and
globe axial length remain normal in CHED, whereas the
corneal thickness is greatly increased, thereby artifactiously
elevating applanation measurements of intraocular pressure.
The combination of congenital glaucoma and congenital hereditary endothelial dystrophy may occur and should be suspected
when persistent and total corneal opacication fails to resolve
after normalization of intraocular pressure.
The degree of edematous corneal clouding varies from a mild
haze to a milky, ground-glass opacication. Epithelial microbullae may be obvious, and stromal thickness may be increased
threefold or more. Uniform thickening of Descemets membrane sometimes is evident on clinical examination, but no
guttata are apparent. Interstitial inflammation and secondary
vascularization are absent. Apart from rare congenital
glaucoma, there are no other associated ocular anterior or

posterior segment abnormalities.
Histologic study269273 reveals nonspecic anterior and
stromal changes consistent with long-standing secondary
edema: basal epithelial cell swelling, basement membrane
thickening and disruptions, and irregularities of Bowmans layer
with pannus formation. It may be signicant, however, that in
some patients, ultrastructural examination discloses greatly
enlarged stromal collagen brils sometimes measuring as much
as 600 in diameter. Descemets membrane is uniform in a
given specimen; it may display diffuse thinning of 3 m to
massive thickening of 40 m (normal thickness is 35 m in
neonates and 810 m in adults). The anterior banded layer of
Descemets membrane always is present and of relatively usual
thickness; however, the posterior layer consists of multilaminar
basement membrane-like material with ne laments and of
collagen brils with a 550 and 1100 banded conguration.
With the exception of the lack of guttata, these ndings are
similar to those in Fuchs dystrophy and thus represent another
example of posterior collagen layer formation by either
primarily or secondarily abnormal endothelium.36,65,273 We
postulate that in patients with thin Descemets membranes,
complete endothelial loss occurred in utero such that only the
fetal anterior portion of Descemets membrane was secreted.272
In contrast, thickened Descemets membranes may be the
product of dystrophic but persistent endothelium having
secreted a hypertrophic posterior collagen layer. The posterior
collagenous layer of Descemets membrane in congenital
hereditary endothelial dystrophy contains collagen types IV
and laminin.274 This distribution of collagen within the pos-
terior collagenous layer supports previous morphologic observations of broblast-like change of the endothelium. CHED
endothelial cells also stain positively for cytokeratin intermediate laments, typically found in cells of epithelial origin. A
similar pattern is seen in the dytrophic endothelium of
posterior polymorphous corneal dystrophy.275
Evaluation and management of young children with CHED
is challenging and best performed in collaboration with a
pediatric ophthalmologist. Visions are difcult to measure in
the rst years of life. Despite the special and multiple
challenges of pediatric keratoplasty, clear corneal transplants
and improved vision are possible, but amblyopia is too often a
limiting factor.276,277 Better surgical and visual outcomes may be
attained if corneal transplantation is delayed until fusion is lost,
as determined by nystagmus or exophoria changing to exotropia.
The frequent nding of enlarged stromal collagen brils
suggests some primary developmental abnormality of both
keratocytes and endothelium, perhaps qualifying this disorder

as another example of mesenchymal dysgenesis.40
Cornea Guttata
Cornea guttata is usually initially evident as a primary condition in middle to older age groups. Slit-lamp examination
reveals a typical beaten-metal appearance of Descemets membrane (Fig. 43.27). These wart-, anvil-, or mushroom-shaped
excrescences are abnormal elaborations of basement membrane
and brillar collagens by distressed or dystrophic endothelial
cells. The endothelial cells over these excrescences become
attenuated and eventually die prematurely. The lesions often
are located in the axial areas of the cornea and may be
distributed sparsely. Brownish pigmentation often is seen at the
level of the guttata (clinically misnamed pigmented guttata),
as this in fact represents pigment phagocytosis by the endothelium.
CHAPTER 43
FIGURE 43.26. Congenital hereditary

endothelial dystrophy. Top left, Clinical
photograph of a mildly affected 20-year-old
woman shows diffuse corneal haze and visual
acuity of 20/200. Top center, On slit-lamp
biomicroscopy, diffuse edematous thickening of
the corneal stroma is evident in the same
patient. Top right, Comparison of similarly
prepared survey light micrographs of congenital
hereditary endothelial dystrophy (a) and normal
human cornea (b). Note the extraordinary
increase in the thickness of the stroma in the
former. H & E 60. Bottom, upper right inset,
Light micrograph of the edematous stroma
demonstrates vesicular water clefts (asterisks).
PAS 200. Bottom, upper middle inset, Electron
micrograph of the central stroma shows the
cross-sectioned collagen brils to have
enlarged diameters (~500 ) with some at
700 (arrowheads) (45 000). Bottom, main
gure, Transmission electron micrograph of the
posterior cornea. The anterior portion of
Descemets membrane (DM) appears to have
banding of normal thickness, but the posterior
collagenous layer is markedly thickened
(815 mm). In addition, an abnormal posterior
collagenous layer is present (asterisk). No
endothelial cells are present. S, stroma; AC,
anterior chamber (10 240). Bottom, lower
inset, At higher magnication, the components
of the posterior collagenous layer are visible as
ne laments (~12 nm diameter) interspersed
with basement membrane-like material
(asterisk) (50 000).
525
SECTION 6
FIGURE 43.27. Cornea guttata. Top left, Slitlamp photography shows stromal edema and
folds in Descemets membrane with metalbeaten appearance. Top right, Extensive
endothelial guttae are demonstrated by
retroillumination. Left, upper inset, By light
microscopy, excrescences (arrows) of
Descemets membrane are evident with loss of
endothelial cells. PAS 100. Left, middle inset,
Specular photomicrograph of the endothelial
mosaic represents such guttae as dark holes.
Bottom right, Transmission electron micrograph
features a thickened Descemets membrane
with individual guttae (asterisk) (3000). Left,
bottom inset, At higher magnication, the
guttata are resolved as ne laments, multiple
segments of basement membrane material, and
collagen in long-spacing conguration
(arrowheads) (40 500).
526
Guttata located in the periphery of the cornea may be seen

even in young patients; these are called HassallHenle bodies
and are of no clinical concern. If, however, the guttata become
more numerous and central, this may portend functional compromise of the endothelial cells to the extent that their barrier
and pump functions become insufcient. In this event, stromal
edema occurs, followed by epithelial edema and bullous keratopathy, and the condition may then be appropriately termed
Fuchs dystrophy. The presence of central guttata without
edema does vary with age: 3% of patients between ages 20 and
40 years, 10% of patients over 40, and 18% of patients over 50.
However, in individual cases mild to moderate guttata can
remain stationary for years without obligate dystrophic
progression.
Secondary cornea guttata is usually associated with degenerative corneal disease, trauma, or inflammation. The corneal
endothelial cells may be affected adversely by iritis, deep

stromal inflammation or infection, and anterior segment
surgery. In severe inflammation, the endothelial mosaic may be
affected by edema of the endothelial cells,278 a condition
resembling cornea guttata. On removal of the causative agent,
the pseudoguttata subside, whereas the guttae of true cornea
guttata are permanent.
The normal endothelial pattern can be well demonstrated
with nitroblue tetrazolium stain. The cells form a uniform
mosaic. If trypan blue stain is used, the decreasing endothelial
viability is noted by staining of the nuclei. An abnormal endothelial cell population is suggested by abnormally sized and
shaped cells (polymegathism and pleomorphism), numerous
guttata, and areas of Descemets membrane that are not covered
by cells. Specular microscopy also can be used to study in vivo
the size, shape, and number of endothelial cells.279
CHAPTER 43
FIGURE 43.28. Late hereditary endothelial

dystrophy (Fuchs). Top left, upper and lower,
Clinical photographs of moderately advanced
cases illustrate severe stromal edema and
surface irregularities secondary to epithelial
microcysts and coalescent bullae. Top right
inset, Light microscopy demonstrates
intraepithelial edema, thickening of the
basement membrane, subepithelial bullae
(asterisk), and brocellular pannus with an
adjacent break in Bowmans layer. H & E 350.
Top right, Transmission electron micrograph of
basal epithelial cells and Bowmans layer shows
multilaminar basement membrane complexes
(BM), the sequela of chronic epithelial edema
(25 000). Bottom, main gure, Transmission
electron micrograph of posterior cornea shows
unremarkable stroma (S) and anterior
Descemets membrane (D), but remarkable
thickening of the posterior Descemets
membrane to 12 m with additional
superimposition of large guttae (G). The
remaining endothelial cells (En) are severely
degenerated and attenuated (5000). Bottom
inset, By scanning electron microscopy, the
comparable picture shows disconnected and
enormously attenuated endothelial cells (En)
and numerous exposed mushroom-shaped
excrescences (asterisk) projecting from the
posterior collagenous layer (1600).
Pachymetry of the corneal stroma is extremely helpful in

monitoring the functional status of the endothelium. Cornea
guttata per se does not require treatment. If serial pachymetry
measurements over time remain stable, then the patient can
be reassured of retaining vision and deferring corneal transplantation. In contrast, if endothelial decompensation and stromal
edema progress to visually incapacitating and/or painful epi-
thelial edema, then medical measures or corneal transplantation may be indicated.
Late Hereditary Endothelial Dystrophy (Fuchs)

First described in 1910, Fuchs dystrophy (Fig. 43.28)280 usually
is seen in the fth or sixth decade of life and is more common
in women. It is bilateral and commonly of dominant
527
SECTION 6
528
inheritance.281283 The dystrophy has been linked to

chromosome 1p34.3p32 in the region of the COL8A2 gene,
which encodes the alpha-2 chain type VIII collagen, a
component of endothelial basement membrane.284,285 Serial
analysis of gene expression demonstrates diminished expression of mitochondrial pump function and antiapoptotic defense
genes.286 The fundamental functional defect is progressive
deterioration of the endothelium. The endothelial cells in
adults lack signicant mitotic capability, and as they undergo
attrition, the surviving cell population must enlarge and spread
to maintain an intact monolayer and to remain functionally
competent as a barrier and pump in maintaining corneal
deturgescence. Thus, as in patients with cornea guttata, serial
pachymetry, specular microscopy279 and confocal microscopy
are helpful in following the disease process. Specular
microscopy demonstrates dark areas, corresponding to the
guttata, within the normally uniform endothelial cell mosaic.
Both discrete (guttate) and diffuse thickening of Descemets
membrane usually develop, with progressive endothelial
degeneration and dysfunction leading to advancing stromal
edema and decreased corneal transparency.
Clinically signicant edema starts axially and spreads peripherally. As stromal edema progresses to involve the
epithelium, microbullous elevations of the epithelium produce
irregularities of the tear lm with conseqent decreased visual
acuity, and in time, macrobullous keratopathy erupts with
profound visual compromise. When these epithelial blisters
rupture, patients experience foreign body sensation or pain,
which may be symptomatically relieved by lubricants,
occlusion, or a bandage soft contact lens. On histologic
examination, the sequelae of chronic epithelial and stromal
edema are prominent. Anteriorly, abnormalities of the
basement membrane adhesion complexes develop because of
repeated liftoff of the edematous epithelium.287 There are
occasional breaks in Bowmans layer, and subepithelial debris
and brovascular pannus collect in the zone of bullous edema.
The most striking abnormality (especially vivid with PAS
staining) is diffuse thickening of Descemets membrane (often
to 20 mm or more) and posteriorly projecting excrescences,
corresponding to clinically apparent guttata. The remaining
endothelial cells are flattened and attenuated, corresponding to
the clinically observed decrease in cell population density with
compensatory increased cell size and polymegathism.
Histologic evidence of abnormal endothelial cell function is
apparent many years before the clinical signs of cornea guttata
and thickened Descemets membrane appear.288
Ultrastructural examination shows the newly deposited
abnormal portion of Descemets membrane to consist of
bundles and sheets of widely spaced banded collagen and
multiple laminations of basement membrane material. This
abnormal posterior collagenous layer of Fuchs dystrophy can be
considered analogous to the deposition of excess collagen and
basement membrane material found in other circumstances of
the endothelial distress syndrome.36,66
A number of surgical considerations arise in patients with
endothelial dystrophy. Excimer laser vision correction should be
undertaken cautiously and only in patients with mild stages of
the dystrophy, as nonresolving corneal edema after routine
LASIK surgery has occurred.289 As both cataract and Fuchs
dystrophy tend to be progressive disorders of older age, the
approach to such patients must be individualized, favoring
cautious cataract surgery alone (i.e., copious use of viscoelastics,
minimal phakoemulsication energy and duration) in younger
patients with milder stages of dystrophy (i.e., no epithelial
edema and pachymetry < 640 mm).290 Patients with more
advanced corneal changes and denser cataract benet from the
combined or triple procedure of penetrating keratoplasty with
extracapsular cataract extraction and posterior chamber IOL

implantation. Fuchs dystrophy is a common indication for
corneal transplantation (~1520% of cases) in predominantly
caucasian patients but is far rarer in the oriental population.291
Keratoplasty is generally very successful as 5- and 10-year graft
survival rates are 97% and 90%, respectively.292 The recent and
ongoing development of deep lamellar endothelial keratoplasty
is especially exciting as this approach allows tissue specic
replacement of diseased endothelium and Descemets membrane without surface or stromal incision or sutures, thereby
producing more rapid visual rehabilitation and less postoperative astigmatism.293 In rare cases for which keratoplasty is
not indicated due to other vision-limiting factors (e.g., advanced
glaucoma or optic atrophy), then anterior stromal puncture,
surface cauterization or amnion membrane transplantation
may afford symptomatic relief.
Posterior Polymorphous Dystrophy

Posterior polymorphous corneal dystrophy (PPCD), originally
described by Koeppe,294 is a bilateral, usually dominantly
transmitted corneal dystrophy that may be stationary or only
slowly progressive, such that affected patients generally retain
normal visual acuity and demonstrate no stromal edema or
vascularization.295297 Based on recent genetic studies, the
Human Genome Nomenclature Committee has identied
several loci for PPCD284,297,298: PPCD 1 on chromosome 20q11;
PPCD 2 in the COL8A2 gene on chromosome 1p, which is also
associated with Fuchs endothelial dystrophy; and PPCD 3 on
chromosome 10. Transcripts of all three identied genes are
present in corneas. However, other unidentied genes appear to
be involved as some cases of clinically and pathologically typical
PPCD do not map to these loci. The co-occurrence of both
PPCD and keratoconus has been reported. Mutations in the
VSX1 homeobox gene may be responsible for both phenotypes.
The condition is characterized by polymorphous opacities,
some of which are vesicular or annular with surrounding halos,
at the level of Descemets membrane (Figs 43.29 and 43.30).
Broad peripheral anterior synechiae also is a characteristic
feature, present in up to 27% of patients.296,299 Although careful
biomicroscopy usually is adequate to establish the diagnosis,300
specular microscopy301 and confocal microscopy may be helpful
in differentiating posterior polymorphous dystrophy from other
corneal endothelial disorders (Fig. 43.31). Iris abnormalities
include corectopia, papillary ectropion and rare Descemetization of the iris surface by endothelial outgrowth. Elevated
intraocular pressure has been described in 14% of 59 PPCD
patients.295 Elevated intraocular pressures were present in 62%
of patients with PPCD undergoing keratoplasty, all of whom
displayed iridocorneal adhesions. The differential diagnosis of
PPCD includes Fuchs endothelial dystrophy, iridocorneal
endothelial (ICE) syndrome, tears in Descemets membrane and
interstitial keraitis.
Numerous histologic studies have demonstrated endothelial
cells that morphologically and immunopathologically resemble
epithelium302304 (see Figs 43.29 and 43.30). These cells contain
epithelial keratin and are connected by well-developed
desmosomes. Scanning electron microscopy of the posterior cell
membrane reveals myriad microvilli, again suggestive of an
epithelium-type cell. Ultrastructural studies have also revealed
some endothelial cells that resemble broblasts.295 Descemets
membrane is also pathologic as the normal anterior banded
zone is accompanied by a thickened posterior collagenous layer
comprised of disorganized collagen and basement membrane
material. An aberrant developmental differentiation of the
endotheliogenic mesenchyme (neural crest) has been
suggested,39 possibly similar to the pathogenesis of the
iridocorneal endothelial syndromes.
CHAPTER 43
FIGURE 43.29. Posterior polymorphous

dystrophy. Top left, Broad slit-lamp illumination
reveals multiple coalescent posterior corneal
vesicles with surrounding halos. Top right, In a
similar case, retroillumination highlights bandlike and polymorphous congurations of the
posterior cornea. Middle left, Scanning electron
microscopy discloses an epithelial-like cell with
characteristic myriad microvilli lining the
posterior corneal surface (1000). Bottom,
Transmission electron micrograph illustrates
other features of these multilayered cells, such
as desmosomal attachments (circles), and
bundles of cytoplasmic laments (arrows). Dm,
Descemets membrane; Ac, anterior chamber
(19 000). Middle right, Higher-magnication
transmission electron microscopy shows details
of the microvilli as seen in transverse and
longitudinal sections. Note resolution of the
central lamentous core typical of cilia
(87 500).
In some cases, endothelial decompensation occurs and

stromal edema develops with decreased vision necessitating
keratoplasty. Preexisting glaucoma and iridocorneal adhesions
are implicated in graft failiure. The dystrophy can recur in
grafts. Short-term followup suggests LASIK is safe and effective
in mildly affected PPCD patients.305
Iridocorneal Endothelial Syndrome

Chandlers syndrome, essential iris atrophy, and iris nevus
or CoganReese syndrome have been regarded as variations
of a single disease process and pathogenetic mechanism,
the so-called iridocorneal endothelial (ICE) syndrome

(Fig. 43.32).306310 These conditions are predominantly
sporadic, almost always unilateral and generally arise in early
adulthood, usually in women. Evidence of herpes simplex viral
DNA or antigens has been detected in ocular tissues of ICE
syndrome patients, although the signicance is unclear.
Typical of Chandlers syndrome is corneal edema secondary
to endothelial abnormality, usually accompanied by ipsilateral,
unilateral glaucoma. The degree of corneal edema is severe
relative to the level of intraocular pressure. The various iris
changes (stromal thinning, full-thickness holes, nevi, and broad,
529

FIGURE 43.30. Posterior polymorphous
dystrophy. Top right, Slit-lamp photograph
demonstrates focal thickenings (arrows) of
varying size at the level of Descemets
membrane. Upper center inset, Phase-contrast
photomicrograph of guttae changes in
Descemets membrane with many irregularly
shaped excrescences and deteriorating
endothelial cells. PPDA 400. Lower center
inset, Phase-contrast photomicrograph shows
the posterior stromal pit as an infolding of
Descemets membrane. Note continuity of the
brocellular tissue (lling central cavity) with
posterior collagen layer of Descemets
membrane (DM). PPDA 400. Bottom, main
gure, Transmission electron micrograph shows
Descemets membrane to have a normalappearing anterior-banded zone (diamond) and
an extremely thickened posterior portion
(asterisk). Normal endothelium is absent;
instead, epithelium-type cells with numerous
microvillous projections, desmosomal
attachments (arrows), and aggregates of
keratobrils are seen. 12 800. Middle right,
Scanning electron microscopy discloses a
geographic area of endothelial cell
degeneration exposing a brillar posterior
collagen layer. The remaining cells are
congured bizarrely, with extended cytoplasmic
processes (540).
SECTION 6
Lower center inset, From Henriquez AS, Kenyon KR,

Dohlman CH, et al: Morphologic characteristics of
posterior polymorphous dystrophy. A study of nine
corneas and review of the literature. Surv Ophthalmol
1984; 29:139.
530
tenting peripheral anterior synechiae) vary, depending on the

subcategory of ICE syndrome. Campbell and colleagues307 have
proposed that the primary abnormality resides in the corneal
endothelium, which, besides malfunctioning and allowing
corneal edema in Chandlers syndrome, tends to grow across
angle structures and iris surface, elaborating a Descemets
membrane-like tissue. Contraction of the membrane then leads
not only to anterior synechiae but also to pupillary distortion
and the iris abnormalities seen to a greater or lesser extent in all
of the ICE syndromes. Glaucoma is common, as high as 77% in
one large series, all accompanied by iridocorneal adhesions.308
Mechanisms for glaucoma include iridocorneal adhesions and
overgrowth of trabecular meshwork with ectopic endothelium.
Chandlers syndrome must be differentiated from Fuchs
dystrophy and posterior polymorphous dystrophy. The latter
may also be considered within the spectrum of ICE syndromes
because a similar pathogenic defect in the corneal endothelium

may be implicated, possibly reflecting abnormal proliferation or
induction of embryonic neural crest cells.39 In contrast to the
ICE syndromes, however, posterior polymorphous dystrophy is
familial and bilateral and without similar iris ndings, except
peripheral anterior synechiae in some patients. In addition to
posterior polymorphous dystrophy, the abnormal beaten-metal
appearance of Descemets membrane in Chandlers syndrome
resembles that of Fuchs dystrophy. Specular microscopy may
prove useful in better describing the in vivo characteristics of
these conditions.311
Histopathologic study of Chandlers syndrome has revealed
abnormalities in the mesenchymally derived cells lining the
cornea, trabecular meshwork, and anterior iris surface.312314
The corneas typically exhibit a posterior collagenous layer
containing collagen types III, IV, V, and VIII. The abnormal
single- or multilayered endothelium extends from the cornea

over the trabecular meshwork and, in some specimens, onto the
iris. Elaboration of excessive multilaminar basement membrane
by flat and discontinuous corneal endothelial cells is further
evidence of an endothelial distress syndrome.315
Medical and surgical control of glaucoma in ICE syndrome is
difcult with poorer success rates compared to other types of
glaucoma. Trabeculectomy with 5-fluoruracil or mitomycin-C
and aqueous shunt surgery have been described, with multiple
surgical procedures and revisions frequently required. Penetrating keratoplasty is indicated for visually signicant corneal
edema. Clear corneal transplants have moderately high prognosis, although chronic intraocular inflammation, glaucoma and
rejection compromise the long term success.316,317 Recurrence
of abnormal endothelium may occur in the grafted cornea.
NONINFLAMMATORY CORNEAL ECTASIAS

Keratoconus
Keratoconus is a bilateral, noninflammatory condition characterized by axial ectasia of the cornea (Fig. 43.33). Reported
estimates of its frequency vary widely, but most range between
50 and 230 per 100 000 population.318 Keratoconus occurs in

all races and has a female preponderance. Despite being common, its causes and pathogenesis remain poorly understood.
Although familial in nature, no exclusive pattern of inheritance
exists. Large studies have suggested that the frequency of
inheritance is 68%.319 Feder believes that it is reasonable to tell
patients that the chances of a blood relative developing
symptoms of the disease are less than 1 in 10.320
Subtle irregular astigmatism often is the rst clinical nding
in keratoconus, and this is evidenced by a distortion of the
corneal image as noted with the Placido disk, retinoscope,
keratometer, keratoscope, and computerized keratographs.
Among computer assisted corneal topography devices, the
development of the scanning slit combined with Placido disk
instrumentation (Orbscan II, Bausch & Lomb) is especially
useful, particularly in the early diagnosis of subclinical cases
which might by all other criteria appear appropriate for excimer
laser vision correction. Keratoconus manifests at puberty and
can progress either slowly, stabilizing over the course of ~10
years, or relatively rapidly, requiring keratoplasty. Thinning of
the cornea with conical protrusion of the apex occurs, such that
in downgaze, the lower lid is distorted by the cone (Munsons
CHAPTER 43
FIGURE 43.31. Posterior polymorphous dystrophy. Pathognomonic specular microscopy patterns occurring in the same patient range. From
mild polymegathism and pleomorphism (top left) to discrete geographic areas (arrows) with minimal residual normal-appearing endothelial cells
(asterisks) (top and bottom right). Bottom left, Grouped vesicles make up dark rounded central areas (asterisk) surrounded by halos of abnormal
endothelial cells.
531
SECTION 6
FIGURE 43.32. Iridocorneal endothelial

syndrome. Top right, Clinical appearance of a
cornea in a 48-year-old woman with unilateral
stromal edema, peripheral anterior synechia, iris
atrophy, and glaucoma. Bottom left, Phasecontrast photomicrograph illustrates posterior
stroma (S), relatively normal anterior
Descemets membrane (D), and the ~10-mmthick posterior collagen layer (bracketed area).
The endothelial layer is irregular and
discontinuous. PPDA 400. Main gure,
Transmission electron micrograph shows the
corresponding area with posterior stroma (S),
ultrastructurally normal Descemets membrane
(D), thick posterior collagen layer (between
arrows), and an attenuated endothelial cell
(9000). Middle right inset, Higher magnication
of area indicated by asterisk in main gure to
resolve basement membrane-like material, ne
laments, and long-spaced banding patterns of
the posterior collagen layer (50 000). Lower
right inset, Scanning microscopy of
keratoplasty specimen shows an attenuated
endothelial cell (En) extending numerous
cytoplasmic processes (2000).
532
sign; Table 43.4). Ultrasonic pachymetry, especially if performed

serially over time is, in conjunction with topography, an important
objective means of monitoring the progression of the disease.
Two types of cones have been described: a well-demarcated
nipple-shaped cone and a larger, oval or sagging cone.321 The
apex of the nipple cone usually is slightly inferonasal, whereas
the oval-shaped cone often is slightly displaced to the inferotemporal quadrant and extends closer to the periphery. The
apex of the cone often exhibits subepithelial scarring. Vertical
stress lines (Vogts striae) are seen deep in the affected stroma.
Increased visibility of the corneal nerves and Fleischers iron
ring are additional diagnostic signs. The latter is caused by a
deposition of hemosiderin pigment deep in the epithelium and
Bowmans layer at the base of the cone.
An early histopathologic change is focal disruption of
Bowmans layer,322 which is replaced in affected areas with
keratocytes and collagenous material. The epithelium is

irregular in thickness and has an abnormal basement
membrane in areas where Bowmans layer is destroyed.323,324
Stromal changes, even in areas of extreme thinning, are
nonspecic.
Acute hydrops may occur when Descemets membrane is
stretched beyond its elastic breaking point. Such a rupture leads
to sudden, profound corneal edema. Endothelium bridges the
gap in 68 weeks, with resultant stromal deturgescence and
residual stromal scarring of varying severity. Although the
abrupt onset of often massive corneal edema and profound
visual loss is startling, management remains conservative medical therapy (patching, bandage soft contact lens, lubricants,
hypertonic agents and occasionally topical steroids) plus
reassurance that the situation should resolve spontaneously
within 3 months. Thus there is never urgent indication for
CHAPTER 43
FIGURE 43.33. Keratoconus. Top left, Clinical

photograph in lateral projection demonstrates
extreme anterior protrusion of the markedly
ectatic cornea. Top right, Munsons sign. The
V-shaped conformation of the lower lid is
produced by the ectatic cornea in downgaze.
Upper middle left, Acute hydrops due to a
break in Descemets membrane is
accompanied by extreme stromal and epithelial
edema. Upper middle right, Keratoscopic view
of typical egg-shaped appearance of the
central corneal mires caused by inferotemporal
steepening. Lower middle left, Corneal
retroillumination is a useful technique to identify
the position and extent of the cone. Bottom
left, Fish-mouth break in Descemets
membrane remains after resolution of corneal
hydrops. Bottom right inset, Light micrograph
of a cornea with healed hydrops shows a ledge
formed by detached Descemets membrane (D)
and endothelium (e). New regenerated
endothelium (ne) lines the anterior surface of
the ledge and posterior stroma (s). AC, anterior
chamber. Phase-contrast, PPDA 400. Bottom
right, Electron micrograph of area indicated by
square in inset demonstrates normal
ultrastructure of endothelium (e) and
Descemets membrane (D). A thin basement
membrane (BM) is subjacent to the new
endothelium (NE). K, keratocyte (8100).
533
TABLE 43.4. Differential Diagnosis of Noninflammatory Corneal Ectasias

Keratoconus
Pellucid Marginal
Degeneration
Keratoglobus
Posterior
Keratoconus
Frequency
Most common
Less common
Rare
Least common
Laterality
Usually bilateral
Bilateral
Bilateral
Usually unilateral
Age at onset
Puberty
Age 20 to 40 years
Usually at birth
Birth
Thinning
Inferior paracentral
Inferior band 1 to 2 mm wide
Greatest in periphery
Paracentral posterior
excavation
Protrusion
Thinnest at apex
Superior to band of thinning
Generalized
Usually none
Iron line
Fleischers ring
Sometimes
None
Sometimes
Scarring
Common
Only after hydrops
Mild
Common
Striae
Common
Sometimes
Sometimes
None
SECTION 6
From Feder RS: Noninflammatory ectatic disorders. In: Krachmer JH, Mannis MJ, Holland EJ, (eds.): Cornea. 2nd edn. St Louis: CV Mosby; 2005
keratoplasty, rather elective keratoplasty should be reserved for

cases that persist longer than 34 months. Ultrastructural
examination in areas of healed hydrops has shown the torn
edges of Descemets membrane to have retracted as scrolls and
the disrupted endothelium to have migrated across the exposed
surface of posterior stroma, depositing new Descemets
membrane material and renewing continuity of the endothelial
monolayer.325
Keratoconus can occur in association with a variety of ocular
and systemic diseases, including atopic dermatitis,326 vernal
catarrh,327 Downs syndrome,328,329 retinitis pigmentosa,330
infantile tapetoretinal degeneration,331 Marfans syndrome,332
and aniridia,333 and in patients with blue sclera, EhlersDanlos
syndrome, and osteogenesis imperfecta type I. The association
with atopy and vernal keratoconjunctivitis has led to speculation
that frequent, vigorous eye rubbing may aggravate, accelerate, or
even cause keratoconus.334,335 Some investigators, moreover,
have inferred contact lens wear as causative.336
Initial treatment requires astigmatic spectacle correction or
various combinations of rigid gas permeable and/or toric soft
contact lens that compensates for the irregular corneal
astigmatism. When lens t or comfort becomes a problem due
to a focal elevated pannus over the apex of the cone, supercial
keratectomy may be performed to smooth the corneal surface.
Excimer laser phototherapeutic keratectomy may also be
cautiously applicable in similar circumstances.337 Thermokeratoplasty generally is only a temporary measure because
resteepening, scarring, or persistent epithelial defects usually
ensue.338 and, like epikeratoplasty, is currently seldom performed (except occasionally for persistent hydrops). Contact
lens-intolerant patients with clear central corneas may benet
from reduction of astigmatism by surgical insertion of
intracorneal ring segments (INTACS).339 Deep lamellar
keratoplasty is highly useful in special situations of clinical
circumstance (e.g., Downs syndrome) or surgical difculty (e.g.,
thin corneal periphery requiring large diameter keratoplasty
with consequent high rejection risk). In most cases, however,
penetrating keratoplasty remains a highly successful procedure
of choice for long-term visual rehabilitation of advanced
cases.340 Postkeratoplasty myopia can be reduced by using the
same-sized donor and host, if the anterior lens-to-retina length
is not less than 20.19 mm.341,342
Pellucid Marginal Degeneration
534
Pellucid marginal degeneration is characterized as a bilateral,

peripheral corneal ectasia with an inferior band of corneal
thinning 12 mm in width. The area of thinning usually

extends from the 4 oclock to the 8 oclock position and is
located 12 mm central to the inferior limbus. The protruding
cornea usually is central to the area of thinning and remains of
normal thickness (see Table 43.4). The abnormal contour
usually induces a shift in the axis of astigmatism from againstthe-rule superiorly, to with-the-rule, at the top of the protrusion
(Fig. 43.34). There is no sex or racial predilection, nor does it
appear to be inherited. These corneas are clear, avascular, and
without apical scarring, lipid deposition, or iron ring. Corneal
topography is an especially useful diagnostic aid. There is some
consensus that keratoconus, keratotorus, keratoglobus, and
pellucid degeneration are related because these different
conditions have been found to coexist in families. Histopathologic reports demonstrate similar abnormalities in these
various disorders.
Because of extremely abnormal corneal topography, the
treatment of pellucid zone degeneration is difcult. Contact
lens wear should be attempted initially. If the patient is contact
lens intolerant, a large-diameter, penetrating keratoplasty may
be performed.343 Alternatively, tectonic lamellar grafting of
the thinned periphery may be followed by a central penetrating
keratoplasty.344 Krachmer345 has suggested that thermokeratoplasty may be a reasonable alternative.
Keratoglobus
Keratoglobus (Fig. 43.35) is a rare bilateral condition resembling
megalocornea, with the exception that the cornea in
keratoglobus is uniformly thinned, particularly peripherally (see
Table 43.4). Corneal scarring may be seen, but an iron ring is
not observed. No denitive inheritance pattern has been
demonstrated. A familial association between keratoconus and
keratoglobus has been made.346 Acquired keratoglobus has been
described in association with hyperthyroidism and after
unilateral preexisting keratoconus.347,348 Rupture of Descemets
membrane may occur, as in keratoconus, but this is not usually
the case.346,349 Especially in cases associated with Ehlers
Danlos syndrome type VI, patients must be cautious to avoid
even minor ocular trauma because rupture of the globe can
occur easily, and repair is difcult. Unlike keratoconus,
keratoglobus is not associated with atopy.
Spectacle correction may help to achieve functional vision in
addition to providing protection from corneal rupture. Contact
lens t is difcult, and surgical intervention should be considered when functional vision cannot be obtained. In general,
surgery should be delayed, when possible. Large-diameter,

FIGURE 43.34. Pellucid marginal degeneration.
Left top and bottom, Clinical photographs
feature corneal ectasia occurring above the
narrow band of clear, thin, nonvascularized
cornea that parallels the inferior limbus
(arrows). Right top and bottom, Slit-lamp view
and corresponding illustration show normal
corneal thickness central and peripheral to the
band of thinning (arrow).
CHAPTER 43
Bottom right, From Krachmer JH, Feder RS, Belin MW:

Keratoconus and related noninflammatory corneal
thinning disorders. Surv Ophthalmol 1984; 28:293.
FIGURE 43.35. Keratoglobus. Left, Clinical photograph of acquired keratoglobus shows globoid protrusion of clear, diffusely thin cornea.
Corneal thickness is one-third normal. Right, Horizontal pupiloptic nerve section of this eye reveals bulging cornea and deep anterior chamber.
The entire cornea is about one-third normal thickness, except in extreme periphery nasally and temporally. H & E 4.
Right, From Jacobs DS, Green WR, Maumenee AE: Acquired keratoglobus. Am J Ophthalmol 1974; 77:393399. Copyright 1974, Elsevier Science.
535

tectonic penetrating or preferably lamellar keratoplasty followed
by smaller-diameter, central penetrating keratoplasty may be
appropriate. Alternative surgical approaches include crescentic
lamellar keratoplasty, thermokeratoplasty, epikeratoplasty, and
wedge or crescentic resection.
CORNEAL DEGENERATIONS
PERIPHERAL DEGENERATIONS
Corneal Arcus (Juvenilis and Senilis)
Corneal arcus appears as a whitish ring of the peripheral cornea
separated from the limbus by a clear zone (Fig. 43.36). Arcus
juvenilis is sometimes called anterior embryotoxon. Both the
juvenile and the adult forms represent paralimbal stromal
accumulations of cholesterol esters, triglycerides, and
phospholipids.350352 Patients younger than 40 years of age with
corneal arcus are at increased risk of coronary artery disease and
should be evaluated for hyperlipoproteinemia. Hyperlipoproteinemia types 2 and 3 are associated with premature arcus
formation. These diseases may be primary or secondary.
Diseases causing a rise in b-lipoproteins include nephrotic
syndrome, hypothyroidism, increased cholesterol intake,
obstructive jaundice, and diabetic ketoacidosis. Rare lipoprotein
disorders causing arcus or generalized corneal clouding include
lecithin cholesterol acyltransferase deciency and Tangier
disease.
In histologic sections, the deposits appear wedge-shaped and
are most prominent near Bowmans layer and Descemets
membrane. Abnormalities in blood lipids may be concomitant
in younger patients displaying corneal arcus or Schnyders
crystalline corneal dystrophy.
White Limbal Girdle of Vogt
Idiopathic Furrow Degeneration

Thinning of the cornea in older people in the area of an arcus
senilis sometimes occurs. There is no tendency for this thinned
area to perforate, and no vascularization develops. Visual acuity
is generally not affected.
Furrow Degeneration Associated with Systemic

Disease
Focal or extensive ring-type epithelial defects and sterile
ulceration near the limbus can accompany certain systemic
diseases, such as rheumatoid arthritis, Wegeners granulomatosis, polyarteritis nodosa, relapsing polychondritis,
systemic lupus erythematosus, and other collagen vascular
diseases (Fig. 43.37). The treatment of such immunogenic
diseases is discussed elsewhere.
Postirradiation Thinning
Noninflammatory corneal excavation at the limbus may occur
after high local doses of b-radiation.354
Terriens Marginal Degeneration

Terriens marginal degeneration is an uncommon but distinct
variety of marginal thinning of the cornea355 (Fig. 43.38). It is
usually bilateral, although often asymmetric, and is seen
mainly in younger men. The condition progresses slowly over
the course of years and generally starts superiorly as a marginal
opacication. Stromal thinning and ectasia develop with an
intact epithelium, and there is a lucid zone between the
advancing edge and the limbus. A yellow border of lipid is
present characteristically at the advancing edge. Vessels traverse
the furrow and pass beyond it. Difculties arise from the
induced corneal astigmatism, and minor trauma may result in
rupture if thinning is sufcient. Most cases are noninflammatory, although patients with recurrent inflammation
SECTION 6
The white limbal girdle of Vogt type II is a common nding in

patients older than 45 years. It is a white opacity in the medial
and temporal limbal regions and may be mistaken for corneal
arcus. Fine white lines extend irregularly from the limbus. A
clear interval may be present between the girdle and the limbus.
The limbal girdle is not associated with inflammation, is not
vascularized, and does not progress. The incidence of Vogts
limbal girdle increases with age, reaching ~55% at 4060 years
of age.353 Sugar and Kobernick353 described the pathologic change

in Vogts limbal girdle type II as a subepithelial hyperelastosis
with degeneration similar to that in pingueculae and pterygia.
The type I limbal girdle is likely to be more closely related to
early calcic band keratopathy because it appears, as Vogt
described it, as a white band with clear holes at several points
and separated from the sclera by a clear interval.
536
FIGURE 43.36. Corneal arcus. Left, The arcus lipoides shows a dense white annular opacity of the peripheral stroma. Right, At higher
magnication, an intervening zone of clear stroma separates lipid deposition from the limbus.
have been described.356 Electron microscopy demonstrates

collagen precursors, stromal ground substance, and possibly
lipid phagocytized by histiocytic cells with high lysosomal
activity.355 Therapy is limited to tectonic grafting to prevent or
to repair perforation of thinned areas.
Moorens Ulcer
Moorens ulcer (Fig. 43.39) is probably best considered a
localized inflammatory ulceration rather than a degeneration of
the corneal periphery. It must be differentiated, however, from
entities such as Terriens marginal degeneration. It is a diagnosis of exclusion; other serious systemic connective tissue
diseases with generalized vasculitis and collagen destruction
must be excluded before the diagnosis of Moorens ulcer is
made.
The disease can be divided into two groups: primary and
secondary. Primary Moorens ulcer is the classic idiopathic
variety, whereas the secondary Mooren ulcer may be associated
with different insults to the cornea. A Mooren-like ulcer has
been reported after cataract surgery,357 penetrating keratoplasty,358 corneal trauma and chemical burns,359 herpes zoster
virus infection,360 syphilis, and tuberculosis. An association has
been reported between Moorens ulcer and hepatitis C
infection.361
Moorens ulcer, in contrast to typical degenerations, is

characterized by a fulminating, centrally progressive, and
painful inflammation occurring more often in males.362366 The
leading edge of the ulcerative process often undermines the
more central corneal stroma. Two types of Moorens ulcer have
been described. A benign type is seen in older patients. This
type is usually unilateral and responds to treatment more often
than the more severe type that occurs in younger patients.365
The latter variety is relentlessly progressive and often bilateral.
Young Nigerians have exhibited a severe form of Moorens ulcer,
with a rapid progression to perforation and marked involvement
of the limbal sclera and episclera in a necrotizing process.363
Histologic studies reveal necrosis of collagen tissue, with
vessels and chronic inflammatory cells in the adjacent limbal
margin. Autoantibodies to human epithelium have been
demonstrated.364 Polymorphonuclear leukocytes intensely
inltrate the zone of active ulceration, suggesting that acute
inflammatory cells play a role in the collagenolytic process.367
The results of treatment have not been encouraging. A
stepladder approach to therapy is recommended.366 Initially,
patients are aggressively treated with topical steroids every hour
in addition to topical antibiotics for prophylaxis. If there is no
evidence of clinical improvement, conjunctival excision is
performed.368 Generally, most patients with unilateral disease
CHAPTER 43
FIGURE 43.37. Furrow degeneration associated with systemic disease. Top left, Clinical photograph of a patient with polyarteritis nodosa shows
a full-ring ulcer and associated lipid deposition near the limbus. Top right, Light microscopy of the same eye illustrates peripheral corneal
thinning (arrows) corresponding to the area of clinical ulceration. H & E 3. Bottom left, Severe necrotizing vasculitis of a medium-caliber artery
conrms the diagnosis. H & E 100. Bottom right, Light microscopy of a 47-year-old woman with rheumatoid arthritis who developed a
perforated marginal corneal ulcer shows adherent iris incorporated into a brous scar. PAS 32.
537

FIGURE 43.38. Terriens marginal
degeneration. Top left, Clinical photograph of a
patient with extensive thinning of the peripheral
stroma extending circumferentially from the
9-oclock to the 2-oclock position. Top right,
Higher magnication discloses vascularization
of the involved stroma with lipid deposition at
the advancing edge. Bottom inset, Light
microscopy reveals numerous foamy histiocytic
cells and blood vessels (asterisk) within the
anterior stroma. H & E 300. Bottom,
Transmission electron micrograph shows
histiocytic cells laden with neutral lipid
inclusions (circled area). Several reactive
broblasts and chronic inflammatory cells also
are seen (5000).
SECTION 6
or bilateral nonsimultaneous disease respond to this approach.

In more aggressive cases, perforation may occur from collagenolytic processes or secondary infection, especially in the
potentiating presence of topical corticosteroids. Tissue adhesive
and lamellar grafting may be necessary in the event of perforation.369 Systemic immunosuppression may be of value in
patients with progressive disease.370374
CENTRAL OR DIFFUSE DEGENERATIONS

Iron Lines
Iron deposition in the cornea occurs secondarily in a number of
clinical settings375376:
HudsonSthli line
Normal aging cornea
Adjacent to ltering bleb
Ferrys line377
Stockers line
Adjacent to head of pterygium
Fleischers ring
Base of keratoconus cone
Histologic examination reveals hemosiderin deposition in
the basal corneal epithelial cells.377 The pathogenesis of corneal
iron lines is unclear, although they may be related to chronic
abnormalities of tear flow. Iron lines do not affect vision and are
asymptomatic.
Coats White Ring
538
This small corneal opacity usually is located in an area that

previously harbored a foreign body.378,379 The opacity, which
contains iron, appears as a small granular oval ring when
viewed with the slit lamp. It was originally thought to be lipid

in nature but probably contains iron from the foreign body.379
The condition causes no symptoms and requires no therapy.
Lipid Degeneration
Lipid degeneration (Fig. 43.40) is characterized clinically by the
accumulation of a yellow or cream-colored diffuse or crystalline
material in the corneal stroma, which may be abnormally thick
or thin. There is typically a history of prior corneal inflammatory episodes with resultant stromal vascularization. The
lipid accumulations are, therefore, of a secondary nature, with
extravasation of cholesterol and fatty acids from the vessels.
Lipid keratopathy has been reported after hydrops380 and as a
nding with no clear antecedent corneal damage or
vascularization.381,382
Amyloid Degeneration
Acquired corneal amyloidosis can be associated with intraocular
disease or may be secondary to corneal trauma.383386 Such
amyloid deposition may also occur as a result of long-standing
diseases, such as retrolental broplasia, trachoma, glaucoma,
uveitis, bullous keratopathy, keratoconus, and leprosy.387 These
corneal amyloid lesions consist of salmon-pink to yellow-white,
raised, fleshy masses that create a nodular surface (Fig. 43.41)
and that may be amenable to treatment by supercial keratectomy. The cornea may be vascularized, depending on other
factors. The deposits seen in lattice and gelatinous dystrophies
CHAPTER 43
FIGURE 43.39. Moorens ulcer. Top left,

Clinical photograph of a 55-year-old man with
painful and rapidly progressive ulcerative
keratitis. Top right, Same patient 15 days after
conjunctival resection reveals marked
improvement with decreased inflammatory
response and arrest of ulceration. Bottom inset,
Phase-contrast micrograph of the stroma at the
margin of the ulcerating area includes abrupt
termination of Bowmans layer (arrow) with
numerous acute inflammatory cells. PPDA
micrograph of the area in bottom inset resolves
multiple intrastromal inflammatory cells actively
engaged in degranulation and phagocytosis.
Note the remnants of the epithelial basement
membrane (arrowheads). E, epithelium; B,
Bowmans layer (7500).
also are amyloid in nature, but those conditions are primary

disorders.
Polymorphic stromal degeneration is another manifestation
of primary localized amyloid deposition in the cornea. Thomsitt
and Bron388 described patients with a variety of posterior
stromal opacities consistent with the type of dystrophic change
reported in 1940 by Pillat.389 They described axial polymorphic
star- and snowflake-shaped and branching lamentous stromal
opacities, some of which indented the anterior surface of
Descemets membrane, causing an apparent irregularity of the
posterior corneal surface. Punctate opacities were polymorphic,
gray-white, and somewhat refractile when examined directly

but were transparent in retroillumination. Because intervening
stroma appeared clear, visual acuity was not markedly affected.
Histochemical staining and electron microscopy have shown
the deposits to be composed of amyloid.390,391 The late
appearance of the linear opacities, the lack of progression, and
the apparent nonfamilial pattern help to distinguish this
condition from lattice dystrophy. Amyloid degeneration also
occurs in association with spheroid degeneration.392,393
The amyloid material in the cornea is identical to that in
other organs. It stains with Congo red, displays birefringence
539
SECTION 6
FIGURE 43.40. Lipid degeneration. Top left,

Clinical photograph of a dense white deposition
of lipid with feathery edges occurring in
association with superior limbic pannus. Note
that this eye has previously undergone an
intracapsular cataract extraction and secondary
implantation of an anterior chamber intraocular
lens. Top right, Same patient after keratoplasty
shows a clear graft with residual opaque lipid
deposition at the periphery. Upper middle left,
Light microscopy shows an intact Bowmans
layer with multiple clear vacuoles within the
stroma (asterisks). 400. Upper middle right,
Phase-contrast microscopy includes numerous
ne osmiophilic deposits (circled area) within
Bowmans layer. PPDA 800. Lower middle,
Transmission electron micrograph of the same
area discloses confluent globular empty spaces
below Bowmans layer (B) as well as some
electron-dense complex lipid deposits (arrows).
E, epithelium (40 000). Bottom right,
Transmission electron micrograph of the
anterior stroma illustrates the same type of
deposits without disruption or other
abnormality of keratocytes (K). 12 000. Bottom
left, At higher magnication, lipid deposits of
~1-m diameter have the characteristics of
saturated neutral fats (asterisk (40 000).
and two-color dichroism with the polarizing microscope, and is

fluorescent with thioflavin-T stain and ultraviolet light. Amyloid contains protein, carbohydrate, and polysaccharide components as well as a-chain immunoglobulins. Ultrastructural
study reveals short brils, 90100- in diameter, in a random
pattern of aggregation within a granular background.
Spheroid Degeneration (Climatic Droplet

Keratopathy, Keratinoid Degeneration)
540
Keratinoid degeneration,393 climatic droplet keratopathy,394407

proteinaceous degeneration,398 Labrador keratopathy,400402 and
chronic actinic keratopathy402 are likely all similar nonhereditary
degenerations related to geographic or climatic conditions.405407

Spheroid degeneration has been classied into three basic
types. Type 1 occurs bilaterally in the cornea without evidence
of other ocular pathology. Type 2, or secondary, spheroid
degeneration occurs in the cornea in association with other
ocular pathology. Type 3 is the conjunctival form of the degeneration and may occur concurrently with types 1 and 2.
This degeneration is characterized by yellow, oily-appearing
subepithelial droplets within the interpalpebral ssure,
generally beginning at the periphery (Fig. 43.42). These droplets
may replace Bowmans layer or may lie deeper. Types of spheroid
degeneration resulting from a local disease or chronic irritant
FIGURE 43.41. Amyloid degeneration. Left, In a patient with long-standing herpes keratitis and subsequent corneal scarring and vascularization,
supercial irregular amyloid deposits developed. Right, Light microscopy of corneal specimen discloses characteristic birefringent, Congo redpositive amyloid deposits (100).
CHAPTER 43
FIGURE 43.42. Spheroid degeneration. Top

left, Clinically, numerous spheroidal deposits
appear over the anterior stroma (arrows). Top
right, Histologic section reveals multiple
densely staining spherules beneath the
distorted epithelium and within the anterior
stroma. H & E 20. Bottom, Survey
transmission electron micrograph shows
spheroidal deposits as extracellular
accumulations of electron-dense material with
variably crystalline structure. Lipid substances
and blood vessels are also evident (5000).
Bottom inset, High-magnication transmission
electron micrograph of a spheroidal deposit
shows variable electron density with a
crystalline fragment similar to calcium
(40 000).
may be unilateral and involve the central cornea as well as the

periphery. Spheroid degeneration has been described in association with lattice dystrophy.
Electron microscopy reveals that the lesions appear to
develop from extracellular material deposited on collagen brils.
Some suggest that this material is secreted by abnormal
brocytes, forming collagenous spheroids.403 An interaction
between ultraviolet light and plasma proteins within the stroma
has also been proposed to result in the abnormal deposits.408

The deposits are PAS-negative but stain positively with rhodamine B hence the designation keratinoid, even though keratin
is not present. The condition probably is related to elastotic
degeneration of collagen, as in conjunctival pingueculae.407
The conjunctiva may become involved with spheroid degeneration, often in association with pingueculae. Patients with
spheroid degeneration do not usually have symptoms, but if
541
SECTION 6
FIGURE 43.43. Band keratopathy. Left, In a 42-year-old woman with chronic uveitis, band keratopathy has resulted in epithelial erosion with a
persistent central defect. Center, Light microscopy discloses dense staining within Bowmans layer. Alizarin red, 40. Right, Transmission
electron micrograph resolves the ne crystalline characteristic and extreme electron density of calcium or hydroxyapatite particles (70 000).
aggravating local factors exist, the disorder may rapidly progress

and predispose to spontaneous sterile ulceration and secondary
microbial keratitis.409
General guidelines for approaching the rehabilitation of
patients with climatic droplet keratopathy with or without
associated cataract have been proposed.411 If the patient is
aphakic or pseudophakic or has a clear lens, excimer laser PTK
may provide primary visual rehabilitation. Because most
patients usually are older and have signicant cataract, the
therapeutic options include cataract extraction without addressing the corneal opacity; combined penetrating keratoplasty
and cataract extraction; combined lamellar keratoplasty and
cataract extraction; or PTK followed by cataract extraction at a
later date. If postoperative PTK is anticipated at the time the
surgical decision is made, it is preferable to begin with PTK and
then to perform cataract extraction ~3 months later. The
combination of lamellar keratoplasty with cataract extraction
avoids the risk of immunologic rejection but is technically more
difcult. This procedure should be limited to cases in which
the anterior pathology is too deep for effective removal with
PTK and there are major contraindications to penetrating
keratoplasty. Penetrating keratoplasty with cataract extraction
has a high risk of failure because these patients usually have
poor goblet cell function, ocular surface wetting, and abnormal
lidglobe relations. This combination of procedures should be
reserved for cases with both supercial and deep scarring and
for those in which there is a reasonable tear lm function and
in which lidglobe anomalies either are absent or have been
addressed with oculoplastic repair.
Band Keratopathy
542
Band keratopathy (Fig. 43.43) can arise from localized ocular

inflammation or systemic disease. Hydroxyapatite deposits
of calcium carbonate accumulate in the epithelial basement
membrane, Bowmans layer, and supercial stroma.411,412
Calcic degenerations, phthisis bulbi, necrotic intraocular
neoplasms, and conditions in which bone is formed in other
parts of the eye are common associations.413
Band keratopathy is conned to the interpalpebral ssure
area. A lucid interval separates the calcic band from the
limbus. Small defects in the band are scattered throughout and
probably represent areas where nerves penetrate Bowmans
layer.
Histopathologically, the earliest changes consist of basophilic

staining of the basement membrane of the epithelium; this
is followed by involvement of Bowmans layer with calcium
deposition and eventual fragmentation.
The factors that stimulate precipitation of calcium salts in the
interpalpebral region of the anterior corneal layers are thought
to involve gaseous exchanges at the corneal surface, leading to
decreased carbon dioxide levels and elevated pH.412 Anatomic
peculiarities in the basement membrane and Bowmans layer
invite calcium deposition, as does the decreased content of acid
mucopolysaccharides in an edematous cornea.414 The calcic
deposits caused by local disease usually are extracellular. In
systemic hypercalcemia, the deposits are intracellular.
Band keratopathy can also result from deposition of urates in
the cornea415; these customarily are brown, instead of the graywhite usually seen in calcic band keratopathy.
The instillation of mercury-containing eye drops, as in
glaucoma and dry-eye states, has a circumstantial relation to the
development of band keratopathy in some patients.416418 The
dry-eye condition itself, through concentration of tear calcium,
also may encourage its deposition near the corneal surface.
Conditions that can result in band keratopathy include the
following:
Hypercalcemia419424
Sarcoidosis (rare)
Fanconis disease
Stills disease (nongranulomatous uveitis)
Hypercalcemia (uremia, parathyroid adenoma)
Hypophosphatasia
Multiple myeloma
Discoid lupus erythematosus
Hyperphosphatemia
Vitamin D toxicity
Metastatic disease (lung and bone disease with increased
calcium)
Ichthyosis
Ocular disease
Chronic nongranulomatous uveitis (juvenile rheumatoid
arthritis)
Prolonged glaucoma
Long-standing corneal edema
Degenerated globe (phthisis bulbi)
Spheroid degeneration
FIGURE 43.44. Salzmanns nodular degeneration. Top left and center, Clinical photographs of two different patients with the classic bluish-gray
elevated paraaxial nodules with sparing of the remainder of the cornea. Right, Higher-magnication slit-lamp photograph emphasizes the
minimal vascularization of the underlying stroma.
Norries disease
Toxic and mercury vapors
Irritants and exposure
Spheroid degeneration
Noncalcic band keratopathy (urate deposits)415
Idiopathic causes
Band keratopathy can be treated by supercial keratectomy

either with or without application of the calcium-binding agent,
ethylenediaminetetra-acetic acid (EDTA). After instillation of
topical anesthesia, EDTA 0.4% may be applied to the deepithelialized cornea. Supercial keratectomy is then performed
by carefully stripping the calcic scale with forceps and by performing blunt dissection with dry cellulose sponges.425 Band
keratopathy has also been treated successfully using excimer
laser PTK.426
Salzmanns Nodular Degeneration

Salzmanns nodular degeneration is noninflammatory and
creates multiple, blue-white, supercial corneal nodules,
usually in the midperiphery (Fig. 43.44). The nodules may
be related to previous inflammation, especially phlyctenular
disease, vernal keratoconjunctivitis, trachoma, or lues and
interstitial keratitis. It has also been reported in patients with
epithelial basement membrane dystrophy, contact lens wear,
and keratoconus and after corneal surgery.427 Although patients
do not often have symptoms, they may develop recurrent
epithelial erosion or decreased vision.
The nodules represent focal areas of subepithelial brocellular avascular pannus, replacing Bowmans layer and
superimposed on a normal stroma. Transmission electron
microscopy has shown reduplication of the epithelial basement
membrane in some patients.427
Treatment may include simple stripping of the focal nodules
by supercial keratectomy. Lamellar or penetrating keratoplasty
rarely is required for visual rehabilitation.
Corneal Keloid
Corneal keloids may be found in either the central or the
peripheral cornea and resemble the nodules in Salzmanns
degeneration. They occur as hypertrophic scars after corneal
injury, inflammation, or surgical trauma. Keloid-like lesions

have also been reported in early life without antecedent trauma.
Immunohistochemical and electron microscopic studies have
demonstrated the presence of myobroblasts in these lesions,
differentiating them from Salzmanns nodules.428
CONJUNCTIVAL DEGENERATIONS
Pterygium
Pterygia are triangular, brovascular connective tissue overgrowths of bulbar conjunctiva onto the cornea (Fig. 43.45).
They are located horizontally in the interpalpebral ssure on
either the nasal or the temporal side of the cornea. A pigmented
iron line (Stockers line) may be seen in advance of a pterygium
on the cornea. The location of the pterygium is determined by
exposure to ultraviolet energy, the amount of which varies
with the geographic latitude.429 True pterygia are found only in
the interpalpebral ssure. Wearing glasses can decrease their
incidence because the ultraviolet transmission is decreased.
A pterygium may progress slowly toward axial cornea or may
become quiescent. Indications of activity are corneal epithelial
irregularity, opacication of Bowmans layer, and prominence of
active blood vessels and inflammation.
Histopathologic examination reveals the subepithelial tissue
to exhibit elastotic degeneration of collagen, resulting from
breakdown of the collagen and destruction of Bowmans membrane.430 The subepithelial material stains for elastin but is not
sensitive to elastase.
Generally, pterygium excision is indicated if the visual axis
is threatened or if the pterygium causes extreme irritation.
A pterygium that recurs after excision does so within several
weeks, starting from the excised conjunctival border. The rate of
recurrence is signicant as high as 40% when a bare scleral
excision is performed. This rate usually is reduced when surgery
is followed by b-radiation treatment with strontium-90.
Treatment with autologous conjunctival transplantation431433
has been shown to decrease the incidence of recurrence to ~5%,
as has adjunctive treatment with mitomycin drops.434,435
Pseudopterygia occur after chemical injury, corneal
ulceration, or other inflammatory problems in which the conjunctiva becomes scarred and drawn on the cornea. A probe can
CHAPTER 43
543
FIGURE 43.45. Pterygium. Top left, Clinical appearance of a typical interpalpebral pterygium shows extension of the brovascular conjunctival
tissue on to clear cornea. Top right, Light microscopy of the limbus features a subepithelial mound of inflammatory tissue invading the cornea
(20). Bottom, Histologic sections show elastotic degeneration of collagen bers (circled area, left gure) and positive stain for elastin (asterisk,
right gure). Phosphotungstic acid-hematoxylin 375; elastin stain 40.
be passed between this conjunctival bridge and the sclera, a

feature that distinguishes pseudopterygia from true pterygia.
Pinguecula
Like pterygia, pingueculae likely represent an age-related
degeneration associated with ultraviolet and general environ-
mental exposure. Pingueculae appear as raised, cream-colored,

white, or chalky perturbations of the conjunctiva adjacent to the
limbus and within the palpebral ssure. Occasionally, they
become inflamed but generally do not require treatment. As in
the case of pterygia, pingueculae may represent elastotic
degeneration of the substantia propria of the conjunctiva.430
SECTION 6
REFERENCES
544
1. Wang MX, ed: Corneal dystrophies and

degenerations: a molecular genetics
approach. Am Acad Ophthalmol 2003.
2. Waring GO, Shields JA: Partial unilateral
cryptophthalmos with syndactyly,
brachycephaly, and renal anomalies. Am J
3. Sugar HS: The cryptophthalmos-syndactyly
syndrome. Am J Ophthalmol 1968; 66:897.
4. Ide CH, Wollschlaeger PB: Multiple
congenital abnormalities associated with
cryptophthalmos. Arch Ophthalmol 1969;
81:638.
5. Friede R: Surface area of cornea and sclera
in embryos and in newborn infants and its
relation to megalocornea in adults.
Z Augenheilkd 1933; 81: 213.
6. Lowe RF: Aetiology of the anatomical basis
of primary angle-closure glaucoma:
biomedical comparisons between normal
eyes and eyes with primary angle-closure
glaucoma. Br J Ophthalmol 1970; 54:161.
7. Durham DG: Cutis hyperelastica

(EhlersDanlos syndrome) with blue
scleras, microcornea, and glaucoma. Arch
8. Meire FM, Bleeker-Wagemaker EM, Oehler
M, et al: X-linked megalocornea: ocular
ndings and linkage analysis. Ophthalmic
Pediatr Genet 1991; 12:153.
9. Kraft SP, Judisch GF, Grayson DM:
Megalocornea: a clinical and echographic
study of an autosomal dominant pedigree.
J Pediatr Ophthalmol Strabismus 1984;
21:190.
10. Meire FM: Megalocornea: clinical and
genetic aspects. Doc Ophthalmol 1994;
87:1.
11. Mackey CA, Buttery RG, Wise GM, Denton
MJ: Description of X-linked megalocornea
wih identication of the gene locus. Arch
12. Skuta GL, Sugar J, Ericson ES: Corneal
endothelial cell measurements in
13.
14.
15.
16.
17.
18.
megalocornea. Arch Ophthalmol 1983;

101:51.
Meire FM, Delleman JW: Biometry in Xlinked megalocornea: pathognomonic
ndings. Br J Ophthalmol 1994; 78:781.
Franois J: La gonioscopie. III. Colobome
de liris. Adv Ophthalmol 1955; 4:44.
Malbran E, Dodds R: Megalocornea and its
relation to congenital glaucoma. Am J
Vail DT Jr: Adult hereditary anterior
megalophthalmos sine glaucoma: a denite
disease entity, with special reference to
extraction of the cataract. Arch Ophthalmol
1931; 6:39.
Duke-Elder S: System of ophthalmology.
Normal and abnormal development. Part 2.
Congenital deformities. London: Kimpton;
1963:501.
Calamandrei D: Megalocornea in due
pazienti con syndrome craniosinostocia.
Q Ital Ophthalmol 1950; 3:278.
19. Libert J, Vanhoof F, Farreaux JP,

Tousseant D: Ocular ndings in I cell
disease (mucolipidosis type II). Am J
20. Erikkson AW, Lehmann W, Forsius H:
Congenital cornea plana in Finland.
Clin Genet 1973; 4:301.
21. Desvignes P, Pouliquen Y, Legras M,
Guyot JD: Aspect iconographique dune
cornea plana dans une maladie de Lobstein.
Arch Ophtalmol (Paris) 1967; 72:585.
22. Rubel E: Kongenitale familiare flachheit der
kornea (cornea plana). Klin Monatsbl
23. Duke-Elder S: System of ophthalmology.
Normal and abnormal development. Part 2.
Abnormalities in the curvature of the
cornea: cornea plana. London: Kimpton;
1963:505.
24. Kolkott W: Cornea plana und Mikrocornea
periplana. Klin Monatsbl Augenheilkd 1937;
98:372.
25. Howard RO, Abrahams IW: Sclerocornea.
26. Elliot JH, Feman SS, ODay DM, Garber M:
Hereditary sclerocornea. Arch Ophthalmol
1985; 103:676.
27. Tahvanainen E, Forsius H, Karila E, et al:
Cornea plana congenita gene assigned to
the long arm of chromosome 12 by linkage
analysis. Genomics 1995; 26:290.
28. Friedman AH, Weingeist S, Brackup A,
Marinoff G: Sclerocornea and defective
mesodermal migration. Br J Ophthalmol
1975; 59:683.
29. Kanai A, Wood TC, Polack FM, Kaufman HE:
The ne structure of sclerocornea. Invest
30. Dada VK, Verma LK, Sachdev MS:
Comprehensive visual and cosmetic
rehabilitation of cornea plana. Cornea
1988; 7:102.
31. Wood T, Kaufman HE: Penetrating
keratoplasty in an infant with sclerocornea.
32. Gregersen E: Keratoplasty in babies with
sclerocornea: to do or not to do? Report of
an unsuccessful case. Dev Ophthalmol
1981; 5:52.
33. Kampik A, Lund OE, Halbig W: Penetrating
keratoplasty in congenital corneal
opacities. Klin Monatsbl Augenheilkd 1986;
188:188.
34. Reese AB, Ellsworth RM: The anterior
chamber cleavage syndrome. Arch
35. Kenyon KR: Mesenchymal dysgeneses of
the cornea. Metab Ophthalmol 1978;
2:173.
36. Waring GO, Bourne WM, Edelhauser HF,
Kenyon KR: The corneal endothelium:
normal and pathologic structure and
function. Ophthalmology 1982; 89:531.
37. Townsend WM, Font RL, Zimmerman LE:
Congenital corneal leukomas. II.
Histopathologic ndings in 19 eyes with
central defect in Descemets membrane.
38. Townsend WM, Font RL, Zimmerman LE:
Congenital corneal leukomas. III.
Histopathologic ndings in 13 eyes with
noncentral defect in Descemets
77:400.
39. Bahn CF, Falls HF, Varley GA: Classication
of corneal endothelial disorders based on
neural crest origin. Ophthalmology 1984;
91:558.
40. Kenyon KR: Mesenchymal dysgenesis in

Peters anomaly, sclerocornea and
congenital endothelial dystrophy. Exp Eye
Res 1975; 21:125.
41. Waring GO III, Rodrigues MM, Laibson PR:
Anterior chamber cleavage syndrome: a
stepladder classication. Surv Ophthalmol
1975; 20:3.
42. Forsius H, Eriksson A, Fellman J:
Embryotoxon cornea posterius in an
isolated population. Acta Ophthalmol
(Copenh) 1964; 42:42.
43. Schwartz DE: Noonans syndrome
associated with ocular abnormalities. Am J
44. Brodsky MC, Cunniff C: Ocular anomalies
in the Alagille syndrome (arteriohepatic
dysplasia). Ophthalmology 1993;
100:1767.
45. Maumenee AE: Further observations on the
pathogenesis of congenital glaucoma. Am
46. Axenfeld T: Embryotoxon corneae
posterius. Dtsch Ophthalmol Gesamte
1920; 42:301.
47. Sugar HS: Juvenile glaucoma with
Axenfelds syndrome: a histologic report.
48. Dark AJ, Kirkham TH: Congenital corneal
opacities in a patient with Riegers anomaly
and Downs syndrome. Br J Ophthalmol
1968; 52:631.
49. Henkind P, Siegel IM, Carr RE: Mesodermal
dysgenesis of the anterior segment:
Riegers anomaly. Arch Ophthalmol 1965;
73:810.
50. Shields MB, Buckley E, Klintworth GK,
Thresher B: Axenfeld-Rieger syndrome:
a spectrum of developmental disorders.
51. Tabbara KF, Khouri FP, Kaloustian VM der:
Riegers syndrome with chromosomal
anomaly (report of a case). Can J
52. Honkanen RA, Nishimura D, Swiderski R,
et al: A family with Axenfeld-Rieger
syndrome and Peters anomaly caused by a
point mutation (Phe112Ser) in the FOXC1
gene. Am J Ophthalmol 2003; 135:368.
53. Jacobs HB: Posterior conical cornea. Br J
54. Williams R: Acquired posterior
keratoconus. Br J Ophthalmol 1987; 71:16.
55. Wolter JR, Haney WP: Histopathology of
keratoconus posticus circumscriptus. Arch
56. Ingram HU: Keratoconus posticus. Trans
57. Krachmer JH, Rodrigues MM: Posterior
keratoconus. Arch Ophthalmol 1978;
96:1867.
58. Peters A: Uber angeborene Defektbildung
der Descemetschen Membran. Klin
Monatsbl Augenheilkd 1906; 44:27, 105.
59. Doward W, Perveen R, Lloyd IC, et al:
A mutation in the REIG1 gene associated
with Peters anomaly. J Med Genet 1999;
36:152.
60. Townsend WM: Congenital anomalies of
the cornea. In: Kaufman HE, Barron BA,
McDonald MB, Waltman SR, eds. Cornea.
New York: Churchill Livingstone; 1989:333.
61. Traboulsi EI, Maumenee IH: Peters
anomaly and associated congenital
malformations. Arch Ophthalmol 1992;
110:1739.
62. Nakanishi I, Brown SI: The histopathology
and ultrastructure of congenital central
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
corneal opacity (Peters anomaly). Am J

Stone DL, Kenyon KR, Green WR, Ryan SJ:
Congenital central corneal leukoma (Peters
anomaly). Am J Ophthalmol 1976; 81:173.
Fogle JA, Green WR, Kenyon KR, et al:
Peripheral Peters anomaly: a
histopathologic case report. J Pediatr
Townsend WM: Congenital corneal
leukomas. I. Central defect in Descemets
77:80.
Waring GO, Laibson PR, Rodrigues M:
Clinical and pathological alterations of
Descemets membrane: with emphasis on
endothelial metaplasia. Surv Ophthalmol
1974; 18:325.
Yang LL, Lamber SR, Lynn MJ, Stulting RD:
Long-term results of corneal graf survival in
infants and children with Peters anomaly.
Ophthalmology 106:833, 1999.
Howard RO, Abrahams IW: Sclerocornea.
Goldstein JE, Cogan DG, Kaufman HE:
Sclerocornea and associated congenital
anomalies. Arch Ophthalmol 1962;
67:761.
Kanai A, Wood TC, Polack FM, et al: The
ne structure of sclerocornea. Invest
Rodrigues MM, Calhoun J, Weinreb S:
Sclerocornea with unbalanced trans
location (17p, 10q). Am J Ophthalmol
1974; 78:49.
Weinzenblatt S: Congenital malformations
of cornea associated with embryonic arrest
of ectodermal and mesodermal structures.
Fogle JA, Kenyon KR, Stark WJ, Green WR:
Defective epithelial adhesion in anterior
corneal dystrophies. Am J Ophthalmol
1975; 79:925.
Bron AJ, Tripathi RC: Cystic disorders of
the corneal epithelium. I. Clinical aspects.
Guerry D III: Fingerprint-like lines in the
Guerry DUP III: Observations on Cogans
microcystic dystrophy of the corneal
epithelium. Trans Am Ophthalmol Soc
1965; 63:320.
DeVoe AG: Certain abnormalities of
Bowmans membrane with particular
reference to ngerprint lines in the cornea.
Cogan DG, Donaldson DD, Kuwabara T,
Marshall D: Microcystic dystrophy of the
corneal epithelium. Trans Am Ophthalmol
Soc 1964; 62:213.
Cogan DG, Kuwabara T, Donaldson DD,
Collins E: Microcystic dystrophy of the
cornea: a partial explanation for its
pathogenesis. Arch Ophthalmol 1974;
92:470.
Laibson PR: Microcystic corneal dystrophy.
Franceschetti A: Hereditare rezidivierende
Erosion der Hornhaut. Z Augenheilkd 1928;
66:309.
Chandler PA: Recurrent erosion of the
Bron AJ, Brown NA: Some supercial
corneal disorders. Trans Ophthalmol Soc
UK 1971; 91:13.
Tripathi RC, Bron AJ: Ultrastructural study
of nontraumatic recurrent corneal erosion.
CHAPTER 43
545
SECTION 6
546
85. Trobe JD, Laibson PR: Dystrophic changes

in the anterior cornea. Arch Ophthalmol
1972; 87:378.
86. Tripathi RC, Bron AJ: Cystic disorders of
the corneal epithelium. II. Pathogenesis.
87. Brown N, Bron A: Recurrent erosion of the
88. Brodrick JD, Dark AJ, Peace GW: Fingerprint
dystrophy of the cornea: a histologic study.
89. Rodrigues MM, Fine BS, Laibson PR,
Zimmerman LE: Disorders of the corneal
epithelium: a clinicopathologic study of dot,
geographic, and ngerprint patterns. Arch
90. Brown NA, Bron AJ: Supercial lines and
associated disorders of the cornea. Am J
91. Laibson PR, Krachmer JH: Familial
occurrence of dot (microcystic), map,
ngerprint dystrophy of the cornea. Invest
92. Hykin PG, Foss AE, Pavesio C, et al:
The natural history and management of
recurrent corneal erosion: a prospective
randomized trial. Eye 1994; 8:35.
93. Williams R, Buckley RJ: Pathogenesis and
treatment of recurrent erosion. Br J
94. McLean EN, MacRae SM, Rich LF:
Recurrent erosion: treatment by anterior
stromal puncture. Ophthalmology 1986;
93:784.
95. Katsev DA, Kincaid MC, Fouraker BD, et al:
Recurrent corneal erosion: pathology of
corneal puncture. Cornea 1991; 10:418.
96. Orndahl MJ, Fagerholm PP: Treatment of
corneal dystrophies with phototherapeutic
keratectomy. J Refract Surg 1998; 14:129.
97. Dinh R, Rapuano CJ, Cohen EJ, Laibson PR:
Recurrence of corneal dystrophy after
excimer laser phototherapeutic
keratectomy. Ophthalmol 1999; 106:1490.
98. Meesmann A, Wilke F: Klinische und
anatomische Untersuchungen uber eine
bisher unbekannte, dominant vererbte
Epitheldystrophie der Hornhaut. Klin
Monatsbl Augenheilkd 1939; 103:361.
99. Kuwabara T, Ciccarelli EC: Meesmanns
corneal dystrophy: a pathological study.
100. Thiel HJ, Behnke H: Uber die
Variationsbreite der hereditaren
Hornhautepitheldystrophie (Typ MeesmannWilke). Ophthalmologica 1968; 155:81.
101. Burns RP: Meesmann corneal dystrophy.
102. Nakanishi I, Brown SI: Ultrastructure of the
epithelial dystrophy of Meesmann. Arch
103. Irvine AD, Corden LD, Swensson O, et al:
Mutations in cornea-specic K3 or K12
cause Meesmann corneal dystrophy. Nat
Genet 1997; 16:184.
104. Coleman CM, Hannush S, Covello SP, et al:
A novel mutation in the helix termination
motif of keratin K12 in a US family with
Meesmann corneal dystrophy. Am J
105. Stocker FW, Holt LB: Rare form of
hereditary epithelial dystrophy: genetic,
clinical and pathologic study. Arch
106. Pameijer JK: Ueber eine fremdartige
familiare oberflachliche
Hornhautveranderung. Klin Monatsbl
107. Lisch W, Steuhl KP, Lisch C, et al: A new

band-shaped and whorled microcystic
dystrophy of the corneal epithelium. Am J
108. Lisch W, Buettner A, Oeffner F, et al: Lisch
corneal dystrophy is genetically distinct
from Meesman corneal dystrophy and
maps to Xp22.3. Am J Ophthalmol 2000;
130:461.
109. Kuchle M, Green WR, Volker HE, et al:
Reevaluation of corneal dystrophies of
Bowmans layer and the anterior stroma
(Reis-Bcklers and Thiel-Behnke): a light
and electron microscopic study of eight
corneas and a review of the literature.
Cornea 1995; 14:333.
110. Reis W: Familiare, fleckige
Hornhautentartung. Dtsch Med
111. Buecklers M: Ueber eine weitere familiare
Hornhautdystrophie (Reis). Klin Monatsbl
112. Thiel HJ, Behnke H: Eine bisher
unbekannte, subepitheliale hereditare
Hornhaut dystrophie. Klin Monatsbl
113. Ridgeway AE, Akhtar S, Munier FL, et al:
Ultrastructural and molecular analysis of
Bowmans layer corneal dystrophies: an
epithelial origin? Invest Ophthalmol Vis Sci
2000; 41:3286.
114. Yee RW, Sullivan LS, Lai HT, et al: Linkage
mapping of Thiel-Behnke corneal dostrphy
(CDB 2) to chromosome 10q23-q24.
Genomics 1997; 46:152.
115. Small KW, Mullen L, Barletta J, et al:
Mapping of Reis-Bcklers corneal
dystrophy to chromosome 5q. Am J
116. Winkelman JE, Wittebol-Post D,
Delleman JW: Ein Beitrag zur
Hornhautdystrophie Reis-Bcklers. Klin
117. Weidle EG: Klinische und feingewebliche
Abgrenzung der Reis-Bcklerschen
Hornhautdystrophie. Klin Monatsbl
118. Wittebol-Post, Pels E: The dystrophy
described by Reis and Bcklers: separate
entity or variant of the granular dystrophy?
119. Yamaguchi T, Polack FM, Rowsey JJ:
Honeycomb-shaped corneal dystrophy:
a variation of Reis-Bcklers dystrophy.
Cornea 1982; 1:71.
120. Perry HD, Fine BS, Caldwell DR:
Reis-Bcklers dystrophy: a study of eight
cases. Arch Ophthalmol 1979; 97:664.
121. Lohse E, Stock EL, Jones JC, et al:
Reis-Bcklers corneal dystrophy:
immunofluorescent and electron
microscopic studies. Cornea 1989; 8:200.
122. Rice NSC, Ashton N, Jay B, Blach RK:
Reis-Bcklers dystrophy: a
clinicopathologic study. Br J Ophthalmol
1968; 52:577.
123. Jones ST, Stauffer LK: Reis-Bcklers
corneal dystrophy: a clinicopathological
study. Trans Am Acad Ophthalmol
124. Akiya S, Brown SI: The ultrastructure of
Reis-Bcklers dystrophy. Am J Ophthalmol
1971; 72:549.
125. Hogan MJ, Wood I: Reis-Bcklers corneal
dystrophy. Trans Ophthalmol Soc UK
1971; 91:41.
126. Wood TO, Fleming JC, Dotson RS, Cotten
MS: Treatment of Reis-Bcklers corneal
127.
128.
129.
130.
131.
132.
133.
134.
135.
136.
137.
138.
139.
140.
141.
142.
143.
144.
145.
146.
dystrophy by removal of subepithelial brous

tissue. Am J Ophthalmol 1978; 85:360.
Orndahl M, Fagerholm P, Fitzsimmons T,
et al: Treatment of corneal dystrophies with
excimer laser. Acta Ophthalmol (Copenh)
1994; 72:235.
Rapuano CJ, Laibson PR: Excimer laser
phototherapeutic keratectomy. CLAO J
1993; 19:235.
Caldwell DR: Postoperative recurrence of
Reis-Bcklers dystrophy. Am J Ophthalmol
1978; 35:567.
Olson RJ, Kaufman H: Recurrence of ReisBcklers corneal dystrophy in a graft. Am J
Franceschetti AT: La cornea verticillata
(Gruber) et ses relations avec la maladie de
Fabry. Ophthalmologica 1968; 156:232.
Franois J: Glycolipid lipoidosis. In:
Symposium on surgical and medical
management of congenital anomalies of
the eye. Transactions of the New Orleans
Academy of Ophthalmology. St Louis: CV
Mosby; 1968.
DAmico DJ, Kenyon KR, Ruskin JN:
Amiodarone keratopathy: a drug-induced
lipid storage disorder. Arch Ophthalmol
1981; 99:257.
Tripathi RC, Bron AJ: Secondary anterior
crocodile shagreen of Vogt. Br J
Pouliquen Y, Dhermy P, Presles D, Tollard MF:
De generescence en Chagrin de crocodile
de Vogt ou denenerescence en mosaique
de Valerio. Arch Ophthalmol (Paris) 1976;
36:395.
Charney SM: Idiopathic band keratopathy.
Jones ST, Zimmerman LE: Histopathologic
differentiation of granular, macular and
lattice dystrophies of the cornea. Am J
Garner A: Histochemistry of corneal
granular dystrophy. Br J Ophthalmol 1969;
53:799.
Rodrigues MM, Streeten BW, Krachmer JH,
et al: Microbrillar protein and phospholipid
in granular corneal dystrophy. Arch
Konishi M, Yamada M, Nakamura Y,
Mashima Y: Immunohistology of
keratoepithelin in corneal stromal
dystrophies associated with R124
mutations of the BIGH3 gene. Curr Eye Res
2000; 21:891.
Akiya S, Brown SI: Granular dystrophy of
the cornea: characteristic electron
microscopic lesion. Arch Ophthalmol 1970;
84:179.
Rodrigues MM, Gaster RN, Pratt MV:
Unusual supercial confluent form of
granular corneal dystrophy. Ophthalmology
1983; 90:1507.
Waardenburg PJ, Jonkers GH: A specic
type of dominant progressive dystrophy of
the cornea developing after birth. Acta
Pouliquen Y, Dhermy P, Monteil C, et al:
La dystrophie de Groenouw I dans sa
forme superelle. Bull Soc Mem Fr
Ophtalmol 1977; 89:327.
Wittebol-Post D, Van Schooneveld MJ,
Pels E: The corneal dystrophy of
Waardenburg-Jonkers. Ophthalmic Pediatr
Genet 1989; 10:249.
Folberg R, Alfonso E, Croxatto O, et al:
Clinically atypical granular corneal
dystrophy with pathologic features of
147.
148.
149.
150.
151.
152.
153.
154.
155.
156.
157.
158.
159.
160.
161.
162.
163.
164.
lattice-like amyloid deposits.

Rosenwasser GO, Sucheski BM, Rosa N,
et al: Phenotypic variation in combined
granular-lattice (Avellino) corneal dystrophy.
Holland EJ, Daya SM, Stone EM, et al:
Avellino corneal dystrophy: clinical
manifestations and natural history.
Eiberg H, Moller HU, Berendt I, et al:
Assignment of granular corneal dystrophy
Groenouw type I to chromosome 5q. Eur J
Hum Genet 1994; 2:132.
Stone EM, Mathers WD, Rosenwasser GO,
et al: Three autosomal dominant corneal
dystrophies map to chromosome 5q.
Nature Genet 1994; 6:47.
Gregory CY, Evans K, Bhattacharya SS:
Genetic renement of the chromosome 5q
lattice corneal dystrophy type I locus to
within a 2 cm interval. J Med Genet 1995;
32:224.
Munier FL, Korvatska E, Djemi A, et al:
Keratoepithelin mutations in four 5q31linked corneal dystrophies. Nat Genet
1997; 15:247.
Afshari NA, Mullally JE, AFshari MD, et al:
Survey of patients with granular, lattice,
Avellino and Reis-Bucklers corneal
dystrophies for mutations in the BIGH3
and gelsolin genes. Arch Ophthalmol 2001;
119:16.
Ellies P, Renard G, Valleix S, et al: Clinical
outcomes of eight BIGH3-linked corneal
dystrophies. Ophthalmology 2002; 109:793.
Klintworth GK, Bao W, Afshari NA: Two
mutations in the TGFBI (BIGH3) gene
associated with lattice corneal dystrophy in
an extensively studied family. Invest
Dighiero P, Neio F, Ellies, et al: Histologic
phenotype-genotype correlation of corneal
dystrophies associated with eight distinct
mutations in the TGFBI gene.
Aldave AJ, Gutmark JG, Yellore VS, et al:
Lattice corneal dystrophy associated with
the Ala546Asp and Pro551Gln missense
changes in the TGFBI gene. Am J
Lyons CJ, McCartney AC, Kirkness CM,
et al: Granular corneal dystrophy: visual
results and pattern recurrence after lamellar
or penetrating keratoplasty. Ophthalmology
1994; 101:1812.
Stark WJ, Chamon W, Kamp MT, et al:
Clinical follow-up of 193-nm ArF excimer
laser photokeratectomy. Ophthalmology
1992; 99:805.
Stein HA, Cheskes A, Stein RM:
Phototherapeutic keratectomy. In: Excimer:
fundamentals and clinical use. Thorofare,
NJ: Slack; 1994.
Hahn TW, Sah WJ, Kim JH:
Phototherapeutic keratectomy in 9 eyes
with supercial corneal diseases. Refract
Corneal Surg 1993; 9:115.
Nassaralla BA, Garbus J, McDonell PJ:
Phototherapeutic keratectomy for granular
and lattice corneal dystrophies at 1.5 to 4
years. J Refract Surg 1996; 12:795.
Campos M, Nielsen S, Szerenyi K, et al:
Clinical follow-up of phototherapeutic
keratectomy for treatment of corneal
opacities. Am J Ophthalmol 1993; 115:433.
Hersh PS, Spinak A, Garrana R, et al:
Phototherapeutic keratectomy: strategies
165.
166.
167.
168.
169.
170.
171.
172.
173.
174.
175.
176.
177.
178.
179.
180.
181.
182.
and results in 16 eyes. Refract Corneal

Surg 1993; 9(Suppl):S90.
Fagerholm P, Fitzsimmons TD, Orndahl M,
et al: Phototherapeutic keratectomy: long
term results in 166 eyes. Refract Corneal
Surg 1993; 9(Suppl):S76.
Tripathi RC, Garner A: Corneal granular
dystrophy: a light and electron
microscopical study of its recurrence in a
graft. Br J Ophthalmol 1970; 54:361.
Brownstein S, Fine BS, Sherman ME,
Zimmerman LE: Granular dystrophy of the
cornea: light and electron microscopic
conrmation of recurrence in a graft. Am J
Rodrigues MM, McGavic JS: Recurrent
corneal granular dystrophy: a
clinicopathologic study. Trans Am
Stuart JC, Mund ML, Iwamoto T, et al:
Recurrent granular corneal dystrophy. Am J
Wan XH, Lee HC, Stulting RD, et al:
Exacerbation of Avellino corneal dystrophy
after laser in situ keratomileusis. Cornea
2002; 21:223.
Klintworth GK: Lattice corneal dystrophy:
an inherited variety of amyloidosis
restricted to the cornea. Am J Pathol 1967;
50:371.
Bowen RA, Hassard DTR, Wong VG, et al:
Lattice dystrophy of the cornea as a variety
of amyloidosis. Am J Ophthalmol 1970;
70:822.
Mondino BJ, Raj CVS, Skinner M, et al:
Protein AA and lattice corneal dystrophy.
Wheeler GE, Eiferman RA:
Immunohistochemical identications of the
AA protein in lattice dystrophy. Exp Eye
Res 1983; 36:181.
Gorevic PD, Rodrigues MM, Krachmer JH,
et al: Lack of evidence for AA reactivity in
amyloid deposits of lattice corneal
dystrophy and corneal amyloid
98:216.
Folberg R, Stone EM, Shefeld VC, et al:
The relationship between granular, lattice
type I, and Avellino corneal dystrophies: a
histopathologic study. Arch Ophthalmol
1994; 112:1080.
Lorenzetti DWC, Kaufman HE: Macular and
lattice dystrophies and their recurrences
after keratoplasty. Trans Am Acad
Meisler DM, Fine M: Recurrence of the
clinical signs of lattice corneal dystrophy
(type I) in corneal transplants. Am J
Meretoja J: Genetic aspects of familial
amyloidosis with corneal lattice dystrophy
and cranial neuropathy. Clin Genet 1973;
4:173.
Purcell JJ, Rodrigues M, Chisthi MI, et al:
Lattice corneal dystrophy associated with
familial systemic amyloidosis (Meretojas
syndrome). Ophthalmology 1983; 90:1512.
Starck T, Kenyon KR, Hanninen LA, et al:
Clinical and histopathological studies of
two families with lattice corneal dystrophy
and familial systemic amyloidosis
(Meretoja syndrome). Ophthalmology 1991;
98:1197.
Akiya S, Nishio Y, Ibi K, et al: Lattice
corneal dystrophy type II associated with
familial amyloid polyneuropathy type IV.
183. Meretoja J: Comparative histopathological

and clinical ndings in eyes with lattice
corneal dystrophy of the two different
types. Ophthalmologica 1972; 165:15.
184. Rodrigues MM, Rajagopalan S, Jones K,
et al: Gelsolin immunoreactivity to corneal
amyloid, wound healing, and macular and
granular dystrophies. Am J Ophthalmol
1993; 115:644.
185. Kivela T, Tarkkanen A, McLean I, et al:
Immunohistochemical analysis of lattice
corneal dystrophy type I and II. Br J
186. Sunada Y, Shimizu T, Nakase H, et al:
Inherited amyloid polyneuropathy type IV
(gelsolin variant) in a Japanase family. Ann
Neurol 1993; 33:57.
187. Levy E, Haltia M, Fernandez-Madrid I, et al:
Mutation in gelsolin gene in Finnish
hereditary amyloidosis. J Exp Med 1990;
172:1865.
188. Hiltunen T, Kiuru S, Hogell V, et al: Finnish
type of familial amyloidosis: cosegregation
of Asp 187 to Asn mutation of gelsolin with
the disease in three large families. Am J
Hum Genet 1991; 49:522.
189. Wiens A, Marles S, Safneck J, et al:
Exclusion of the gelsolin gene on 9q32-34
as the cause of familial lattice corneal
dystrophy type I. Am J Hum Genet 1992;
51:156.
190. Kangas H, PaunioT, Kalkkinen N, et al: In
vitro expression analysis shows that the
secretory form of gelsolin is the sole source
of amyloid in gelsolin-related amyloidosis.
Hum Mol Gent 1996; 9:1237.
191. Hida T, Proia AD, Kigasawa K, et al:
Histopathologic and immunochemical
features of lattice corneal dystrophy type
III. Am J Ophthalmol 1987; 104:249.
192. Hida T, Tsubota K, Kigasawa K, et al:
Clinical features of a newly recognized type
of lattice corneal dystrophy. Am J
193. Klintworth G: The molecular genetics of the
corneal dystrophies. Front Biosci 2003;
1:687.
194. Chiou AG, Beuerman RW, Kaufman SC,
Kaufman HE: Confocal microscopy in
lattice cornea dystrophy. Graefes Arch Clin
195. Rothstein A, Auran JD, Wittpenn JR, et al:
Confocal microscopy in Meretoja
Syndrome. Cornea 2002; 21:364.
196. Donnenfeld ED, Cohen EJ, Ingraham HJ,
et al: Corneal thinning in macular corneal
dystrophy. Am J Ophthalmol 1986; 101:112.
197. Garner A: Histochemistry of macular
dystrophy. Invest Ophthalmol 1969; 8:475.
198. Klintworth GK, Smith CF: Macular corneal
dystrophy: studies of sulfated
glycosaminoglycans in corneal explant and
confluent stromal cell cultures. Am J Pathol
1977; 89:167.
199. Klintworth GK, Vogel FS: Macular corneal
dystrophy: an inherited acid
mucopolysaccharide storage disease of the
corneal broblast. Am J Pathol 1964;
45:565.
200. Snip RC, Kenyon KR, Green WR: Macular
corneal dystrophy: ultrastructural pathology
of corneal endothelium and Descemets
membrane. Invest Ophthalmol 1973; 12:88.
201. Deepak E, Thonar EJ-MA, Srinivasan M,
et al: Macular corneal dystrophy of the
cornea: a systemic disorder of keratan
sulfate metabolism. Ophthalmology 1990;
97:1194.
CHAPTER 43
547
SECTION 6
548
202. Jonasson F, Oshima E, Thonar E, et al:

Macular corneal dystrophy in Iceland: a
clinical, genealogic and
immunohistochemical study of 28 patients.
203. Yang CJ, SundarRaj N, Thonar EJ, et al:
Immunohistochemical evidence of
heterogeneity in macular corneal dystrophy.
204. Klintworth GK, Meyer R, Dennis RF, et al:
Macular corneal dystrophy: lack of keratan
sulfate in serum and cornea. Ophthalmic
205. Thonar EJ, Meyer R, Dennis RF, et al:
Absence of normal keratan sulfate in the
blood of patients with macular corneal
dystrophy. Am J Ophthalmol 1986;
102:561.
206. Vance JM, Jonasson F, Lennon F, et al:
Linkage of a gene for macular corneal
dystrophy to chromosome 16. Am J Hum
Genet 1996; 58:757.
207. Akama TO, Nishida K, Nakayama J, et al:
Macular corneal dystrophy type I and type
II are caused by distinct mutations in a new
sulfotransferase gene. Nature Genet 2000;
26:237.
208. El-Ashry MF, El-Aziz MM, Wilkins S, et al:
Identication of novel mutations in the
carbohydrate sulfotransferase gene
(CHST6) causing macular corneal
dystrophy. Invest Ophthalmol Vis Sci 2002;
43:377.
209. Abbruzzese C, Kuhn U, Molina F, et al:
Novel mutations in the CHST6 gene
causing macular corneal dystrophy. Clin
Genet 2004; 65:120.
210. Aldave AJ, Yellore VS, Thonar EJ, et al:
Novel mutations in the carbohydrate
sulfotransferase gene (SHST6) in American
patients with macular corneal dystrophy.
211. Al-Swailen SA, Al-Rajhi AA, Wagoner MD:
Penetrating keratoplasty for macular
corneal dystrophy. Ophthalmology 2005;
112:220.
212. Robin AL, Green WR, Lapsa TP, et al:
Recurrence of macular corneal dystrophy
after lamellar keratoplasty. Am J
213. Marcon AS, Cohen EJ, Rapuano CJ,
Laibson PR: Recurrence of corneal stromal
dystrophies after penetrating keratoplasty.
Cornea 2003; 22:19.
214. Thomsitt J, Bron AJ: Polymorphic stromal
dystrophy. Br J Ophthalmol 1975; 59:125.
215. Pillat A: Zur rage der familiaeren
Hornhautentartung. Ueber eine eigenartige
tiefe schollige und periphere gitterfoermige
familiaere Hornhautdystrophie. Klin
216. Krachmer JH, Dubord PJ, Rodrigues MM,
Mannis MJ: Corneal posterior crocodile
shagreen and polymorphic amyloid
101:54.
217. Nakaizumi K: A rare case of corneal
dystrophy. Nippon Ganka Gakkai Zasshi
1914; 18:949.
218. Santo RM, Yamaguchi T, Kanai A, et al:
Clinical and histopathologic features of
corneal dystrophies in Japan.
219. Gartry DS, Falcon MG, Cox RW: Primary
gelatinious drop-like keratopathy. Br J
220. Ide T, Nishida K, Maeda N, et al:
A spectrum of clinical manifestations of
221.
222.
223.
224.
225.
226.
227.
228.
229.
230.
231.
232.
233.
234.
235.
236.
237.
238.
gelatinous drop-like dystrophy in Japan.

Ren Z, Lin P, Klintworth GK, et al: Allelic
and locus heterogeneity in autosomal
recessive gelatinous drop-like dystrophy.
Hum Gen 2002; 110:568.
Akhtar S, Bron AJ, Qin X, et al: Gelatinous
drop-like corneal dystrophy in a child with
developmental delay: clinicopathologic
features and exclusion of M1S1 Gene. Eye
2005; 19:198.
Tian X, Fujiki K, Li Q, et al: Compound
heterozygous mutations of M1S1 gene in
gelatinous drop-like dystrophy. Am J
Kirk JG, Rabb M, Hattenhauer J, et al:
Primary familial amyloidosis of the cornea.
Trans Am Acad Ophthalmol Otolaryngol
1973; 77:411.
Stock EL, Kielar RA: Primary familial
amyloidosis of the cornea. Am J
Mondino BJ, Rabb MF, Sugar J, et al:
Primary familial amyloidosis of the cornea.
Shimazaki J, Hida T, Inoue M, et al: Longterm follow-up of patients with familial
subepithelial amyloidosis of the cornea.
Shimazaki J, Shimmura S, Tsubota K:
Limbal stem cell transplantation for the
treatment of subepithelial amyloidosis of
the cornea (gelatinous drop-like dystrophy).
Cornea 2002; 21:177.
Schnyder WF: Mitteilung ueber einen
neuen Typus von familiaerer
Hornhauterkrankung. Schweiz Med
Ingraham HJ, Perry HD, Donnenfeld ED:
Progressive Schnyders corneal dystrophy.
Shearman AM, Hudson TJ, Andersen JM,
et al: The gene for Schnyders crystalline
corneal dystrophy maps to human
chromosome 1p34.1-p36. Hum Mol Genet
1996; 5:1667.
Takeuchi TT, Furihata M, Heng HHQ, et al:
Chromosomal mapping and expression of
the human B120 gene. Gene 1998;
213:189.
Burns RP, Connor W, Gipson I: Cholesterol
turnover in hereditary crystalline corneal
dystrophy of Schnyder. Trans Am
Weiss JS: Schnyder crystalline dystrophy
sine crystals: recommendation for a
revision of nomenclature. Ophthalmology
1996; 103:465.
Garner A, Tripathi RC: Hereditary crystalline
stromal dystrophy of Schnyder. II.
Histopathology and ultrastructure. Br J
Bron A, Williams HP, Carruthers ME:
Hereditary crystalline stromal dystrophy of
Schnyder. I. Clinical features of a family
with hyperlipoproteinaemia. Br J
Paparo LG, Rapuano CJ, Raber IM, et al:
Phototherapeutic keratectomy for Schnyder
crystalline corneal dystrophy. Cornea 2000;
19:343.
Bietti G: Ueber familares Vorkommen von
Retinitis punctata albescens (verbunden
mit Dystrophia marginalis cristallinea
cornea), Glitzern des Glaskoerpers und
anderen degenerativen
Augenveraenderungen. Klin Monatsbl
239. Li A, Jiao X, Munier FL, et al: Bietti

crystalline corneoretinal dystrophy is
caused by mutations in the novel gene
CYP4V2. Am J Hum Genet 2004; 74:817.
240. Mataftsi A, Zografos L, Milla E, et al: Biettis
crystalline corneoretinal dystrophy: a crosssectional study. Retina 2004; 24:416.
241. Welch RB: Biettis tapetoretinal
degeneration with marginal corneal
dystrophy: crystalline retinopathy. Trans Am
242. Kaiser-Kupfer MI, Chan CC, Markello TC,
et al: Clinical, biochemical and pathologic
correlations in Biettis corneal dystrophy.
243. Franois J: Une novelle dystrophie
heredofamiliale de la cornee. J Genet Hum
1956; 5:189.
244. Strachan IM: Cloudy central corneal
dystrophy of Franois: ve cases in the
same family. Br J Ophthalmol 1969; 53:192.
245. Kobayashi A, Sugiyama K, Huang AJW:
In vivo confocal microscopy in patients
with central cloudy dystrophy of Francois.
246. Karp CL, Scott IU, Green WR, et al: Central
cloudy corneal dystrophy of Francois: a
cliinicopathologic study. Arch Ophthalmol
1997; 115:1058.
247. Carpel EF, Sigelman RJ, Doughman DJ:
Posterior amorphous corneal dystrophy.
248. Dunn SP, Krachmer JH, Ching SST: New
ndings in posterior amorphous corneal
dystrophy. Arch Ophthalmol 1984; 102:236.
249. Johnson AT, Folberg R, Vrabec MP, et al:
The pathology of posterior amorphous
corneal dystrophy. Ophthalmology 1990;
97:104.
250. Grimm BB, Waring GO, Grimm SB:
Posterior amorphous corneal dysgenesis.
251. Witschel H, Fine BS, Grutzner P, et al:
Congenital hereditary stromal dystrophy of
the cornea. Arch Ophthalmol 1978;
96:1043.
252. Bredrup C, Knappskog PM, Majewski J,
et al: Congenital stromal dystrophy of the
cornea caused by a mutation in the decorin
gene. Invest Ophthalmol Vvis Sci 2005;
46:420.
253. Goodside V: Posterior crocodile shagreen:
a corneal dystrophy. Am J Ophthalmol
1958; 46:748.
254. Franois J, Neetens A: Lheredo-dystrophie
mouchetee du parenchyme corneen. Acta
Genet Med Gemellol (Roma) 1957; 6:387.
255. Streeten BW, Falls HF: Hereditary fleck
dystrophy of the cornea. Am J Ophthalmol
1961; 51:275.
256. Purcell JJ Jr, Krachmer JH, Weingeist TA:
Fleck corneal dystrophy. Arch Ophthalmol
1977; 95:440.
257. Jiao X, Munier FL, Schorderet DF, et al:
Genetic linkage of Francois-Neetens fleck
(mouchetee) corneal dystrophy to
chromosome 2q35. Hum Genet 2003;
112:593.
258. Li S, Tiab L, Jiao X, et al: Mutations in
PIP3K3 are associated with FrancoisNeetens mouchetee fleck corneal
dystrophy. Am J Hum Genet 2005; 77:54.
259. Nicholson DH, Green WR, Cross HE, et al:
A clinical and histopathological study of
Franois-Neetens speckled corneal
260. Pippow G: Zur Erbbedingheit der Cornea
farinata. (Mehlstaubartige
261.
262.
263.
264.
265.
266.
267.
268.
269.
270.
271.
272.
273.
274.
275.
276.
277.
278.
Hornhautdegeneration). Graefes Arch

Franceschetti A, Maeder G: Dystrophie
profonde de la cornee dans un cas
ditchtyose congenitale. Bull Soc
Ophthalmol Fr 1954; 67:146.
Grayson M, Wilbrandt H: Pre-Descemet
Curran RE, Kenyon KR, Green WR: PreDescemets membrane corneal dystrophy.
Maeder G, Danis P: Sur une nouvelle forme
de dystrophic cornenne (dystrophia
liformis profunda corneae) associe un
keratocone. Ophthalmologica 1947;
114:246.
Maumenee AE: Congenital hereditary
corneal dystrophy. Am J Ophthalmol 1960;
50:1114.
Toma NM, Ebenezer ND, Inglehearn CF,
et al: Linkage of congenital hereditary
endothelial dystrophy to chromosome 20.
Hum Mol Genet 1995; 4:2395.
Callaghan M, Hand CK, Kennedy SM, et al:
Homozygosity mapping and linkage
analysis demonstrate that autosomal
recessive congenital hereditary endothelial
dystrophy (CHED) and autosomal dominant
CHED are genetically distinct. Br J
Judisch GF, Maumenee IH: Clinical
differentiation of recessive congenital
hereditary endothelial dystrophy and
dominant hereditary endothelial dystrophy.
Kenyon KR, Maumenee AE: The
histological and ultrastructural pathology of
congenital hereditary corneal dystrophy: a
case report. Invest Ophthalmol 1968;
7:475.
Pearce WG, Tripathi RC, Morgan G:
Congenital endothelial corneal dystrophy:
clinical, pathological and genetic study.
Antine B: Histology of congenital hereditary
corneal dystrophy. Am J Ophthalmol 1970;
69:964.
Kenyon KR, Maumenee AE: Further studies
of congenital hereditary endothelial
dystrophy of the cornea. Am J Ophthalmol
1973; 76:419.
Rodrigues MM, Waring GO, Laibson PR,
Weinreb S: Endothelial alterations in
congenital corneal dystrophies. Am J
Sekundo W, Marshall GE, Lee WR, et al:
Immunoelectron labelling of matrix
components in congenital hereditary
endothelial dystrophy. Graefes Arch Clin
Cockerham GC, Laver NV, Hidayat AA,
McCoy DL: An immunohistochemical
analysis and comparison of posteriour
polmorphous dystrophy with congential
hereditary endothelial dystrophy. Cornea
2002; 21:787.
Adamis AP, Filatov V, Tripathi BJ, Tripathi RC:
Fuchs endotheliual dystrophy of the
cornea. Surv Ophthalmol 1993; 38:149.
Biswas S, Munier FL, Yardley J, et al:
Missense mutations in COL8A2, the gene
encoding the alpha-2 chain of type VIII
collagen, cause two forms of corneal
endothelial dystrophy. Hum Mol Genet
2001; 10:2415.
Wolter JR: Secondary cornea guttata in
interstitial keratopathy. Ophthalmologica
1964; 148:289.
279. Bourne WM, Kaufman HE: Specular

microscopy of human corneal endothelium
in vivo. Am J Ophthalmol 1976; 81:319.
280. Fuchs E: Dystrophia epithelialis corneae.
Graefe Arch Klin Exp Ophthalmol 1910;
76:478.
281. Cross HE, Maumenee AE, Cantolino SJ:
Inheritance of Fuchs endothelial dystrophy.
282. Krachmer JH, Purcell JJ Jr, Young CW,
Bucher KD: Corneal endothelial dystrophy:
a study of 64 families. Arch Ophthalmol
1978; 96:2036.
283. Al-Rajhi AA, Wagoner MD: Penetrating
keratoplasty in congenital hereditary
endothelial dystrophy. Ophthalmology
1997; 104:956.
284. Schaumberg DA, Moynes AL, Gomes JAP,
Dana MR: Corneal transplantation in young
children with congenital hereditary
endothelial dystrophy. Am J Ophthalmol
1999; 127:373.
285. Gottsch JD, Sundlin OH, Liu SH, et al:
Inheritance of a novel COL8A2 mutation
denes a distinct early onset subtype of
Fuchs corneal dystrophy. Invest
286. Gottsch JD, Bowers AL, Margulies EH, et al:
Serial analysis of gene expression in the
corneal endothelium of Fuchs dystrophy.
287. Kenyon KR: The synthesis of basement
membrane by the corneal epithelium in
bullous keratopathy. Invest Ophthalmol
1969; 8:156.
288. Bourne WM, Johnson DH, Campbell RJ:
membrane. III. Fuchs dystrophy. Arch
Ophthalmol 1982; 100:1952.
289. Dastjerdi MH, Sugar A: Corneal
decompensation after laser in situ
keratomileusis in Fuchs endothelial
dystrophy. Cornea 2003; 22:379.
290. Seitzman GD, Gotsch JD, Stark WJ:
Cataract surgery in patients with Fuchs
corneal dystrophy. Expanding
recommendations for cataract surgery
without simultaneous keratoplasty.
291. Brady SE, Rapuano CJ, Arentsen JJ, et al:
Clinical indications for and procedures
associated with penetrating keratoplasty.
292. Thompson RW, Price MO, Bowers PJ,
Price FW: Long-term graft survival after
penetrating keratoplasty. Ophthalmology
2003; 110:1396.
293. Terry MA, Ousley PJ: Rapid visual
rehabilitation after endothelial transplants
with deep lamellar endothelial keratoplasty
(DLEK). Cornea 2004; 23:143.
294. Koeppe L: Klinische Beobachtungen mit
der Nertspaltlampe und der
Hornhautmikroskop. Graefes Arch Klin Exp
295. Cibis GW, Krachmer JA, Phelps CD,
Weingeist TA: The clinical spectrum of
posterior polymorphous dystrophy. Arch
296. Krachmer JH: Posterior polymorphous
corneal dystrophy: a disease characterized
by epithelial-like endothleial cells which
influence management and prognosis.
297. Shimizu S, Krafchak C, Fuse N et al: A
locus for posterior polymorphous corneal
dystrophy (PPCD3) maps to chromosome
10. Am J Med Genet 2004; 130A:372.
298. Heon E, Mathers WD, Alward WL, et al:

Linkage of posterior polymorphous corneal
dystrophy to 20q11. Hum Mol Genet 1995;
4:485.
299. Rodrigues MM, Phelps CD, Krachmer JH,
et al: Glaucoma due to endothelialization of
the anterior chamber angle: a comparison
of posterior polymorphous dystrophy of the
cornea and Chandlers syndrome. Arch
300. Cibis GW, Tripathi RC: The differential
diagnosis of Descemets tears (Haabs
striae) and posterior polymorphous
dystrophy bands: a clinicopathologic study.
301. Hirst L, Waring GO: Clinical specular
microscopy of posterior polymorphous
endothelial dystrophy. Am J Ophthalmol
1983; 95:143.
302. Henriquez AS, Kenyon KR, Dohlman CH,
et al: Morphologic characteristics of
posterior polymorphous dystrophy. A study
of nine corneas and review of the literature.
303. Threlkeld AB, Green WR, Quigley HA, et al:
A clinicopathologic study of posterior
polymorphous dystrophy: implications for
pathogenetic mechanism of the associated
glaucoma. Trans Am Ophthalm Soc 1994;
92:133.
304. Rodrigues MM, Sun TT, Krachmer J,
Newsome D: Epithelialization of the corneal
endothelium in posterior polymorphous
dystrophy. Invest Ophthalmol Vis Sci 1980;
19:832.
305. Moshirfar M, Barsam CA, Tanner MC: Laser
in situ keratomileusis in patients with
posterior polymorphous dystrophy. Cornea
2005; 24:230.
306. Chandler PA: Atrophy of the stroma of the
iris. Endothelial dystrophy, corneal edema
and glaucoma. Am J Ophthalmol 1956;
41:607.
307. Campbell DG, Shields MB, Smith TR: The
corneal endothelium and the spectrum of
essential iris atrophy. Am J Ophthalmol
1978; 86:317.
308. Shields MB, Campbell DG, Simmons RJ:
The essential iris atrophies. Am J
309. Shields MB: Progressive essential iris
atrophy, Chandlers syndrome, and the iris
nevus (CoganReese) syndrome: a
spectrum of disease. Surv Ophthalmol
1979; 24:3.
310. Yanoff M: Iridocorneal endothelial
syndrome: unication of a disease
spectrum. Surv Ophthalmol 1979; 24:1.
311. Brooks AMV, Grant G, Gillies WE:
Differentiation of posterior polymorphous
dystrophy from other posterior corneal
opacities by specular microscopy.
312. Patel A, Kenyon KR, Hirst LW, et al:
Clinicopathologic features of Chandlers
syndrome. Surv Ophthalmol 1983;
27:327.
313. Eagle RC, Font RL, Yanoff M, Fine BS: The
iris naevus (CoganReese) syndrome: light
and electron microscopic observations.
314. Rodrigues MM, Streeten BW, Spaeth GL:
Chandlers syndrome as a variant of
essential iris atrophy. A clinicopathologic
study. Arch Ophthalmol 1978; 96:643.
315. Kenyon KR, Van Horn DL, Edelhauser HF:
Endothelial degeneration and posterior
collagenous proliferation in aphakic bullous
CHAPTER 43
549
316.
317.
318.
319.
320.
321.
322.
323.
324.
325.
326.
327.
328.
329.
330.
331.
SECTION 6
332.
333.
334.
335.
336.
337.
338.
550
keratopathy. Am J Ophthalmol 1978;

85:329.
Crawford GJ, Stulting RD, Cavanagh HD,
Waring GO: Penetrating keratoplasty in the
management of iridocorneal endothelial
syndrome. Cornea 1989; 8:34.
Chang PC, Soong HK, Couto MF, et al:
Prognosis for penetrating keratoplasty in
iridocorneal endothelial syndrome. Refract
Corneal Surg 1993; 9:129.
Harstein J: Keratoconus and contact
lenses. JAMA 1969; 208:539.
Kennedy R, Bourne WM, Dyer J: A 48-year
clinical and epidemiologic study of
101:267.
Feder R: Noninflammatory ectatic
disorders. In: Krachmer JH, Mannis MJ,
Holland EJ, eds. Cornea. 2nd edn.
Perry HD, Buxton JN, Fine BS: Round and
oval cones in keratoconus. Ophthalmology
1980; 87:905.
Chi HH, Katzin HM, Teng CC:
Histopathology of keratoconus. Am J
Pataa C, Joyon L, Roucher F:
Ultrastructure du keratocone. Arch
Ophtalmol (Paris) 1970; 30:403.
Pouliquen Y, Graf B, De Kozak Y, et al:
Etude morphologique du keratocone. Arch
Ophtalmol (Paris) 1970; 30:497.
Stone DL, Kenyon KR, Stark WJ:
Ultrastructure of keratoconus with healed
hydrops. Am J Ophthalmol 1976; 82:450.
Spencer WH, Fisher JJ: The association of
keratoconus with atopic dermatitis. Am J
Rahi A, Davies P, Ruben M, et al:
Keratoconus and coexisting atopic disease.
Cullen JF, Butler HG: Mongolism and
keratoconus. Br J Ophthalmol 1963;
47:321.
Pierse K, Eustace P: Acute keratoconus in
mongols. Br J Ophthalmol 1971; 55:50.
Streiff EB: Keratocone et retinite
pigmentaire. Bull Mem Soc Fr Ophtalmol
1952; 65:323.
Alstrom CH, Olson O: Heredo-retinopathia
congenitalis: monohybride recessiva
autosomalis. Hereditas Genetiskt Arkiv
1957; 43:1.
Austin MG, Schaefer RF: Marfans
syndrome with unusual blood vessel
manifestations. Arch Pathol 1957; 64:205.
Franceschetti A: Keratoconus. In: King JH,
McTigue JW, eds. The cornea world
congress. Washington DC: Butterworths;
1965.
Koenig SB, Smith RW: Keratoconus and
corneal hydrops associated with
compulsive eye rubbing. Refract Corneal
Surg 1993; 9:383.
Cameron JA, Al-Rajhi AA, Badr IA: Corneal
ectasias in vernal keratoconjunctivitis.
Macsai MS, Varley GA, Krachmer JH:
Development of keratoconus after contact
lens wear. Arch Ophthalmol 1990; 108:534.
Rapuano CJ: Excimer laser
phototherapeutic keratectomy: long-term
results and practical considerations.
Cornea 1997; 16:151.
Colin J, Cochener B, Savary G, et al:
INTACS inserts for treating keratoconus:
one year results. Ophthalmology 2001;
108:1409.
339. Fogle JA, Kenyon KR, Stark WJ: Damage

to epithelial basement membrane by
thermokeratoplasty. Am J Ophthalmol
1977; 83:392.
340. Krachmer JH, Feder RS, Belin MW:
Keratoconus and related noninflammatory
corneal thinning disorders. Surv
341. Wilson SE, Bourne WM: Effect of recipientdonor trephine size disparity on refractive
error in keratoconus. Ophthalmology 1989;
96:299.
342. Perry HD, Foulks GN: Oversize donor
button in corneal transplantation surgery
for keratoconus. Ophthalmic Surg 1987;
18:751.
343. Varley GA, Macsai MS, Krachmer JH: The
results of penetrating keratoplasty for
pellucid marginal degeneration. Am J
344. Kremer I, Sperber LTD, Laibson PR:
Pellucid marginal degeneration treated by
lamellar and penetrating keratoplasty. Arch
345. Krachmer JH: Pellucid marginal corneal
96:1217.
346. Cavara V: Keratoglobus and keratoconus.
347. Jacobs DS, Green WR, Maumenee AE:
Acquired keratoglobus. Am J Ophthalmol
1974; 77:393.
348. Pouliquen Y, Dhermy P, Espinasse MA,
Savoldelli M: Keratoglobe. J Fr Ophtalmol
1985; 8:43.
349. Grayson M: Acute keratoglobus. Am J
350. Cogan DG, Kuwabara T: Arcus senilis: its
pathology and histochemistry. Arch
351. Andrews JS: The lipids of arcus senilis.
352. Walton KW: Studies on the pathogenesis of
corneal arcus formation. I. The human
corneal arcus and its relation to
atherosclerosis as studied by
immunofluorescence. J Pathol 1973; 111:263.
353. Sugar HS, Kobernick S: The white limbus
girdle of Vogt. Am J Ophthalmol 1960;
50:101.
354. Macfaul PA, Bedford MA: Ocular
complications after therapeutic irradiation.
355. Suveges I, Levai G, Alberth B: Pathology of
Terriens disease: histochemical and
electron microscopic study. Am J
356. Austin P, Brown SI: Inflammatory Terriens
marginal corneal disease. Am J Ophthalmol
1981; 92:189.
357. Salamon SM, Mondino BJ, Zaidman GW:
Peripheral corneal ulcers, conjunctival
ulcers and scleritis after cataract surgery.
358. Mondino BJ, Hofbauer JD, Foos RY:
Moorens ulcer after penetrating keratoplasty.
359. Mondino BM: Inflammatory diseases of the
peripheral cornea. Ophthalmology 1988;
95:463.
360. Mondino BJ, Brown SI, Mondzekewski JP:
Peripheral corneal ulcers with herpes
zoster ophthalmicus. Am J Ophthalmol
1978; 86:611.
361. Wilson SE, Lee WM, Murakami C, et al:
Mooren-type hepatitis C virus-associated
corneal ulceration. Ophthalmology 1994;
101:737.
362. Edwards WC, Reed RE: Moorens ulcer:

a pathologic case report. Arch Ophthalmol
1968; 80:361.
363. Kietzman B: Moorens ulcer in Nigeria. Am
364. Schaap OL, Feltkamp TEW, Breebaart AC:
Circulating antibodies to corneal tissue in a
patient suffering from Moorens ulcer (ulcus
rodens corneae). Clin Exp Immunol 1969;
5:365.
365. Wood TO, Kaufman HE: Moorens ulcer.
366. Brown SI, Mondino BJ: Therapy of
Moorens ulcer. Am J Ophthalmol 1984;
98:1.
367. Foster CS, Kenyon KR, Greiner J, et al:
The immunopathology of Moorens ulcer.
368. Feder RS, Krachmer JH: Conjunctival
resection for the treatment of the
rheumatoid corneal ulceration.
369. Kenyon KR: Decision-making in the therapy
of external eye disease. Ophthalmology
1982; 89:44.
370. Foster CS: Systemic immunosuppressive
therapy for progressive bilateral Moorens
ulcer. Ophthalmology 1985; 92:1436.
371. Foster CS: Immunosuppressive therapy for
external ocular inflammatory disease.
372. Tauber J, Sainz de la Maza M,
Hoang-Xuan T, Foster CS: An analysis of
therapeutic decision making regarding
immunosuppressive chemotherapy for
peripheral ulcerative keratitis. Cornea 1990;
9:66.
373. Zhao JC, Jin XY: Immunological analysis
and treatment of Moorens ulcer with
cyclosporin A applied topically. Cornea
1993; 12:481.
374. Hollan EJ, Olsen TW, Ketcham JM, et al:
Topical cyclosporin A in the treatment of
anterior segment inflammatory disease.
Cornea 1993; 12:413.
375. Gass JD: The iron lines of the supercial
cornea. Arch Ophthalmol 1964; 71:348.
376. Barraquer-Somers E, Chan CC, Green WR:
Corneal epithelial iron deposition.
377. Ferry AP: A new line of the supercial
cornea: occurrence in patients with ltering
blebs. Arch Ophthalmol 1968; 79:142.
378. Coats G: Small supercial opaque white
rings in the cornea. Trans Ophthalmol Soc
UK 1912; 32:53.
379. Nevins RC Jr, Davis WH Jr, Elliott JH:
Coats white ring of the cornea: unsettled
metal fettle [correspondence]. Arch
380. Shapiro LA, Frakas TG: Lipid keratopathy
following corneal hydrops. Arch
381. Fine BS, Townsend WM, Zimmerman LE,
Lashkari MH: Primary lipoidal degeneration
of the cornea. Am J Ophthalmol 1974; 78:12.
382. Friedlaender MH, Cavanagh HD, Sullivan WR,
et al: Bilateral central lipid inltrates of the
383. Stafford WR, Fine BS: Amyloidosis of the
cornea: report of a case without
conjunctival involvement. Arch Ophthalmol
1966; 75:53.
384. McPherson SD Jr, Kiffney GT Jr, Freed CC:
Corneal amyloidosis. Am J Ophthalmol
1966; 62:1025.
385. Garner A: Amyloidosis of the cornea. Br J
386. Ramsey MS, Fine BS, Cohen SW:

Localized corneal amyloidosis: case report
with electron microscopic observations.
387. Rodrigues MM, Zimmerman LE: Secondary
amyloidosis in ocular leprosy. Arch
388. Thomsitt J, Bron AJ: Polymorphic stromal
dystrophy. Br J Ophthalmol 1975; 59:125.
389. Pillat A: Zur Frage der familren
Hornhautentartung: Ueber eine eigenartige
tiefe schollige und periphere gitterrmige
famire Hornhautdystrophie. Klin Monatsbl
390. Mannis MJ, Krachmer JH, Rodrigues MM,
Pardos GJ: Polymorphic amyloid
degeneration of the cornea: a clinical and
histopathologic study. Arch Ophthalmol
1981; 99:1217.
391. Matta CS, Tabbara KF, Cameron JA, et al:
Climatic droplet keratopathy with corneal
amyloidosis. Ophthalmology 1991; 98:192.
392. Santo RM, Yamaguchi T, Kanai A:
Spheroidal keratopathy associated with
subepithelial corneal amyloidosis: a
clinicopathologic case report and a
proposed new classication for spheroidal
keratopathy. Ophthalmology 1993;
100:1455.
393. Garner A: Keratinoid corneal degeneration.
394. Freedman A: Climatic droplet keratopathy.
I. Clinical aspects. Arch Ophthalmol 1973;
89:193.
395. Garner A, Morgan G, Tripathi RC: Climatic
droplet keratopathy. II. Pathologic ndings.
396. Anderson J, Fuglsang H: Droplet
degeneration of the cornea in North
Cameroon: prevalence and clinical
appearances. Br J Ophthalmol 1976;
60:256.
397. Ahmad A, Hogan M, Wood I, Ostler HB:
Climatic droplet keratopathy in a 16-yearold boy. Arch Ophthalmol 1977; 95:149.
398. Christensen GR: Proteinaceous corneal
degeneration: a histochemical study. Arch
399. Freedman A: Labrador keratopathy. Arch
400. DAlena P, Wood IS: Labrador keratopathy:
a microscopic study. Am J Ophthalmol
1972; 74:430.
401. Johnson CJ, Ghosh M: Labrador
keratopathy: clinical and pathological
ndings. Can J Ophthalmol 1975; 10:119.
402. Klintworth GK: Chronic actinic keratopathy:
a condition associated with conjunctival
elastosis (pingueculae) and typied by
characteristic extracellular concretions.
Am J Pathol 1972; 67:327.
403. Hanna C, Fraunfelder FT: Spheroidal

degeneration of the cornea and
conjunctiva. II. Pathology. Am J Ophthalmol
1972; 74:829.
404. Fraunfelder FT, Hanna C: Spheroidal
degeneration of cornea and conjunctiva.
III. Incidences, classication and etiology.
405. Young YDH, Finlay RD: Primary spheroidal
degeneration of the cornea in Labrador and
Northern Newfoundland. Am J Ophthalmol
1975; 79:129.
406. Fraunfelder FT, Hanna C, Parker JM:
Spheroid degeneration of the cornea and
conjunctiva. I. Clinical course and
characteristics. Am J Ophthalmol 1972;
74:821.
407. Rodrigues MM, Laibson PR, Weinreb S:
Corneal elastosis: appearance of band-like
keratopathy and spheroidal degeneration.
408. Johnson GJ, Overall M: Histology of
spheroidal degeneration of the cornea in
Labrador. Br J Ophthalmol 1978; 62:53.
409. Ormerod DL, Dahan E, Hagele JE, et al:
Serious occurrences in the natural history
of advanced climatic keratopathy.
410. Badr IA, Al-Rajhi A, Wagoner MD, et al:
Phototherapeutic keratectomy for climatic
droplet keratopathy. J Refr Surg 1996;
12:114.
411. Pouliquen Y, Haye C, Bisson J, Offret G:
Ultrastructure de la keratopathie en
bandelette. Arch Ophtalmol (Paris) 1967;
27:149.
412. OConnor GR: Calcic band keratopathy.
413. Zeiter HJ: Calcication and ossication in
ocular tissue. Am J Ophthalmol 1962;
53:265.
414. Barber CW: Physiological chemistry of the
415. Fishman RS, Sunderman FW: Band
keratopathy in gout. Arch Ophthalmol
1966; 75:367.
416. Kennedy RE, Roca PD, Landers PH:
Atypical band keratopathy in glaucomatous
patients. Am J Ophthalmol 1971; 72:917.
417. Kennedy RE, Roca PD, Platt DS: Further
observations on atypical band keratopathy
in glaucoma patients. Trans Am Ophthalmol
Soc 1975; 72:107.
418. Lemp MA, Ralph RA: Rapid development
of band keratopathy in dry eyes. Am J
419. Walsh FB, Howard JE: Conjunctival and
corneal lesions in hypercalcemia. J Clin
Endocrinol 1947; 7:644.
420. Walsh FB, Murray RG: Ocular
manifestations of disturbances in calcium
421.
422.
423.
424.
425.
426.
427.
428.
429.
430.
431.
432.
433.
434.
435.
metabolism. Am J Ophthalmol 1953;

36:1657.
Lessel S, Norton EWD: Band keratopathy
and conjunctival calcication in
hypophosphatasia. Arch Ophthalmol
1964; 71:497.
Berkow JW, Fine BS, Zimmerman LE:
Unusual ocular calcication in
hyperparathyroidism. Am J Ophthalmol
1968; 66:812.
Schumacher H, Scheler F: Metastatische
Kalzizierungen an Kornea und Konjunktiva
bei chronischer Niereninsufzienz. Klin
Porter R, Crombie AL: Corneal calcication
as a presenting and diagnostic sign in
hyperparathyroidism. Br J Ophthalmol
1973; 57:665.
Wood TO, Walker GG: Treatment of band
keratopathy. Am J Ophthalmol 1975;
80:553.
OBrart DP, Gartry DS, Lohmann CP, et al:
Treatment of band keratopathy by excimer
laser phototherapeutic keratectomy:
surgical techniques and long term
follow-up. Br J Ophthalmol 1993; 77:702.
Wood TO: Salzmanns nodular
degeneration. Cornea 1990; 9:17.
Holbach LM, Font RL, Shivitz IA, Jones DB:
Bilateral keloid-like myobroblastic
proliferations of the cornea in children.
Adamis AP, Starck T, Kenyon KR: The
management of pterygium. Ophthalmol Clin
North Am 1990; 3:611.
Austin P, Jakobiec FA, Iwamoto T:
Elastodysplasia and elastodystrophy as
pathologic bases of ocular pterygium and
pinguecula. Ophthalmology 1983; 90:96.
Serrano F: Plastia conjunctival libre en la
cirugia del pterigion. Arch Soc Am Oftal
Optom 1977; 12:97.
Kenyon KR, Wagoner MD, Hettinger ME:
Conjunctival autograft transplantation for
advanced and recurrent pterygium.
Starck T, Kenyon KR, Serrano F:
Conjunctival autograft for primary and
recurrent pterygia: surgical technique and
problem management. Cornea 1991;
10:196.
Singh G, Wilson MR, Foster CS: Mitomycin
eye drops as treatment for pterygium.
Singh G, Wilson MR, Foster CS: Long-term
follow-up study of mitomycin eye drops as
adjunctive treatment for pterygia and its
comparison with conjunctival autograft
transplantation. Cornea 1990; 9:331.
CHAPTER 43
551
CHAPTER
44
Keratoconus and Corneal Noninflammatory

Ectasias
Elisabeth J. Cohen, MD
INTRODUCTION
Keratoconus is a progressive, noninflammatory, thinning
disorder causing irregular corneal astigmatism of unknown
cause. Pellucid marginal degeneration (PMD) is a less common,
but related condition with peripheral corneal thinning.
Keratoconus, typically, has a teenage onset and is usually bilateral, but often asymmetric. It occurs in ~1 in 2000, affecting
on the order of 150 000 people in the USA.1 As it is not rare, it
is important to consider the diagnosis in young adults with
increasing myopic astigmatism, especially if it is difcult to get
20/20 visual acuity with a manifest refraction or the patient
complains of decreased quality of vision due to monocular
diplopia, haloes, or ghost images. It is critical to make the
diagnosis of keratoconus, as it is a contraindication for most
refractive surgery. In addition, in recent years it has been
recognized that patients who are suspects for keratoconus,
i.e., have forme fruste keratoconus, on the basis of abnormal
corneal topography alone with normal slit-lamp biomicroscopy
exams, are not good candidates for refractive surgery. These
patients are at increased risk of postoperative complications,
particularly corneal ectasia.
After a teenage onset, keratoconus typically progresses and
then stabilizes in the fourth decade. However, relatively little
has been published about keratoconus in older patients. In a
study of the demographics of keratoconus, less than 10% of
patients were over age 50.2 It was concluded that either keratoconus is associated with premature death or that older patients
are followed locally and not at referral centers, which is probably
more likely. Recent publications regarding patients followed for
more that ten years after penetrating keratoplasty (PK) for
keratoconus have shown evidence of late increases in astigmatism and inferior peripheral thinning and steepening consistent
with progressive keratoconus in the host.3 Therefore, it is
possible that keratoconus progresses for a longer time than is
often thought, at least in some patients. Generally, patients with
onset before 18 years of age have worse disease with shorter
time to PK than patients who are older at the time of diagnosis.4
Keratoconus is usually an isolated condition, but it may be
associated with other conditions, including Down syndrome,
atopy, and floppy eyelids. Eye rubbing is a common denominator in all these conditions, which may be why they are
associated with keratoconus.5 Keratoconus can be associated
with Leber s congenital amaurosis and connective tissue
diseases including mitral valve prolapse. It may also be
accompanied by rare genetic diseases.68 There is a positive
family history of keratoconus in ~10% of patients.7
Keratoconus is routinely referred to as a form of corneal
ectasia,8 but this may not be an accurate term to use. Ectasia is
associated with stretching and should result in increased

corneal surface area, which does not occur in keratoconus.9
There is steepening in the area of the cone and compensatory
flattening elsewhere, usually superiorly, in keratoconus.
CLINICAL SIGNS
Key Features
Diagnosis of keratoconus is made by the presence of thinning

and protrusion on slit-lamp examination.
Forme fruste keratoconus, or keratoconus suspect, is
diagnosed by abnormal topography only and a normal slitlamp examination.
Vogts striae and Fleischer ring are other frequent ndings.
Hydrops usually resolves over several months and is not an
indication for emergent penetrating keratoplasty.
The diagnosis of keratoconus is based on a careful slit-lamp

biomicroscopy examination. The clinical diagnosis requires the
presence of localized corneal thinning and protrusion, typically
inferiorly or centrally. These early slit-lamp signs can be subtle
and easily missed. It is helpful to observe the shape of the
anterior cornea with a narrow slit beam to detect mild ectasia
where the cornea bulges forward (Fig. 44.1a). In keratoconus the
area of maximal thinning corresponds to the area of maximal
ectasia. After identifying the area of ectasia, one can look
carefully for thinning in that location by comparing the
thickness of the cornea above and below there using a narrow
slit beam. The area of corneal steepening and ectasia is more
obvious on corneal topography than by slit-lamp examination
(Fig. 44.1b). If the topography is suspicious for keratoconus, one
should reexamine the patient by slit lamp to look for minimal
ectasia and thinning that may have been overlooked on initial
examination. The slit-lamp ndings are critical: if abnormal,
the patient has keratoconus, and, if normal in the presence of
abnormal topography, the patient is a keratoconus suspect, also
known as forme fruste keratoconus.
Vogts striae and Fleischer ring are present on slit-lamp
examination with mild to moderate disease. Vogts striae are
relatively frequent. They are vertical folds at the level of the
posterior stroma and Descemets membrane in the area of
maximal thinning that are best seen with a wide slit-lamp beam
(Fig. 44.2). These are stress lines that can be made to disappear
with gentle pressure at the limbus. The Fleischer ring of iron
deposition in the corneal basal epithelium surrounds part, or
all, of the cone. It is best seen with a broad slit beam and the
cobalt-blue light, but is not always present (Fig. 44.3).
553

FIGURE 44.1. (a) Slit-lamp examination of
keratoconus patient shows maximal thinning
which coincides with maximal ectasia. Note
abnormal curvature of cornea is evident by
shape of anterior image of slit beam.
(b) Topography of same patient shows inferior
steepening that is more readily appreciated
than by slit-lamp examination. Superior
changes are artifact.
SECTION 6
FIGURE 44.2. Vogts striae are vertical folds in the deep cornea that
are best seen with a wide slit beam.
554
With more advanced disease anterior corneal scarring

develops, progresses and reduces vision.10 Scarring may have a
reticular pattern and corresponds on histopathology with
brosis in areas where there are breaks in Bowmans layer
(Fig. 44.4). Elevated scars can develop in the area of maximal
ectasia and limit contact lens tolerance. Signicant scarring
can develop in advanced disease, which limits best-corrected
visual acuity with rigid gas-permeable (RGP) lenses, and is an
indication for PK. Scarring usually coincides with the area of
maximal ectasia so will be more visually signicant in central
than inferior cones.
Some clinical signs of advanced disease associated with
keratoconus are less helpful in diagnosing keratoconus than in
the past due to the widespread availability of slit-lamp biomicroscopes, autorefractors and corneal topography which
make early diagnosis much easier. Munsons sign refers to the
V-shaped protrusion of the lower lid observed in downgaze
caused by the ectatic cornea. Rizzutis sign, describing light
focused on the nasal limbus with lateral illumination, and the
Charleux oil droplet sign by retroillumination are primarily of
historical interest. Scissoring of the light reflex on retinoscopy
is also rarely observed in the era of autorefractors.
Hydrops is a manifestation of advanced keratoconus in which
there is a sudden loss of vision usually associated with pain
caused by breaks in Descemets and acute, marked corneal
edema, often with fluid clefts in the stroma, involving a variable
area of the cornea (Figs 44.5a and b). It is not an indication for
PK. It usually resolves over a period of weeks to months and
results in corneal scarring and flattening, with or without corneal neovascularization. Although hydrops is frequently treated
FIGURE 44.3. The Fleischer ring of iron in the epithelium is seen with
Cobalt-blue illumination.
FIGURE 44.4. In advanced keratoconus supercial scarring develops

which can have a reticular pattern and is associated with brosis
within areas of breaks in Bowmans layer.
with topical hyperosmotic agents, aqueous suppressants,

cycloplegia, antibiotics and steroids, it is not clear that any of
these shorten the time until resolution.11 Rarely, hydrops is
complicated by corneal perforation which may be treated with
tissue adhesive or require PK, and even less often by corneal
infection. Sometimes after hydrops resolves there is additional
inferior scarring, the cornea flattens, and the visual acuity
improves.
Keratoconus and Corneal Noninflammatory Ectasias
CORNEAL TOPOGRAPHY
Key Features
Corneal topography is very helpful in the diagnosis, evaluation

and management of keratoconus.
It also is important in research to advance understanding of
the disease.
The diagnosis of forme fruste keratoconus by topography has
evolved over time.
Computer-assisted corneal topography has revolutionized the

diagnosis, evaluation, and management of keratoconus. Prior to
corneal topography, corneal curvature measured by keratometry
and keratoscopes was used to diagnose and follow keratoconus.
Keratometry measures usually four points on the central cornea
3 mm apart and assumes a regular spherocylindrical shape of
the cornea. Irregular mires and skewed axes of astigmatism
(i.e., the steep axes are not along one meridian and are at an
angle to each other) are signs of irregular astigmatism
suggestive of keratoconus. Inferior corneal steepening measured
by keratometry in upgaze compared to primary position is also
associated with keratoconus. Placido-based keratoscopy provides qualitative evidence of irregular astigmatism and localized
corneal steepening. Amsler in 1938 used a photographic placido
disk to evaluate very early keratoconus determined by small
amounts of skewed astigmatism where the horizontal axis
deviated by only 185! The photokeratoscope in the 1970s
provided qualitative information about corneal curvature only
beyond the central 3 mm.
Since 1990 computer-assisted corneal topography or videokeratoscopy has become the standard way to evaluate corneal
curvature and has vastly improved our ability to diagnose and
treat keratoconus as well as many other corneal conditions
affecting the shape of the ocular surface.5,12 Scanning slittopography devices (Orbscan, Bausch and Lomb, Claremont,
CA) provide information on the anterior elevation, posterior
elevation, and pachymetry, but there is some concern about the
accuracy of the posterior elevation or curvature.7 Placido-based
systems are used more commonly. Color coded maps makes
them relatively easy to interpret. There are different scales that
can be used. One is the absolute scale with 1.5 D steps where
each color is always associated with the same corneal curvature
in diopters. Another is a normalized scale where the range of
colors is used to cover the range of powers in the specic cornea
being imaged. The size of the steps varies depending on the

shape of the cornea and can be very small, 0.5 D or less. The
color of a given power varies from one eye and map to another.
There is some controversy as to what scale is most appropriate
to use. In the late 1990s it was thought that the normalized
scale would detect inferior steepening and suspect keratoconus
in too many patients,5 but with the increasing problem of
ectasia after LASIK, more recently 0.5 D intervals have been
recommended specically to detect forme fruste keratoconus.13
In order to judge the quality of corneal topographic color-coded
maps it is important to review the reflection of the rings on the
cornea imaged at the same time. By looking at the rings one can
determine whether the image was well centered on the cornea
and the number and quality of complete rings (Fig. 44.6). If the
eye is misaligned, for example by the patient looking up, regular
astigmatism will look irregular and can be mistaken for
keratoconus.
There is considerable variation in slit lamp and topographic
ndings in keratoconus. Typically, the apex of the cone where
there is maximal thinning, ectasia, and steepening is located
inferocentrally, but it often can be inferotemporal or inferonasal. By slit-lamp examination advanced cones have been
analyzed as round, nipple, central and oval, sagging, inferior
cones.14 The location of corneal steepening is more obvious by
topography. Central nipple cones by slit lamp are associated
with central steepening and peripheral flattening by topography
(Fig. 44.7). Low, sagging, oval cones are associated with inferior
steepening and superior flattening (Fig. 44.8). The approach to
contact lens tting and surgical planning varies depending on
CHAPTER 44
FIGURE 44.5. (a) In hydrops there is sudden corneal edema due to breaks in Descemets. (b) In the same patient, by narrow slit beam, fluid
clefts within the stroma are present.
FIGURE 44.6. Example of inferocentral cone. It is important to

examine the image of the reflected rings from the cornea in order to
judge the quality of the topography with regard to the number of
complete rings and the centration of the image on the cornea.
555
FIGURE 44.7. Paracentral steepening is present in a patient with a

paracentral nipple cone. The flat K is 52 D corresponding to a radius
of curvature of 6.50 mm. The steep K is 65 D. Trial lens tting with a
Rose K lens was started using a 6.50 mm base curve lens. The lens
prescribed had a 6.70 mm. Base curve, flatter than the flat K.
FIGURE 44.9. This is an example of the topography in a case with an

asymmetric bowtie pattern with skewed radial axes of astigmatism
(SRAX).
SECTION 6
FIGURE 44.8. Inferior steepening and superior flattening are present

in this patient with an inferior cone.
556
the location of the cone. In addition many keratoconus patients

and suspects have steepening in an asymmetric bow tie pattern
with skewed radial axes (SRAX) of astigmatism5,13 (Fig. 44.9).
Software has been developed to generate quantitative indices
from topographic images to distinguish normal from
keratoconus suspect corneas. These indices have evolved over
time and continue to do so. In 1995 Rabinowitz indices
included central corneal power more than 47.2 D, inferior
minus superior (I-S value) asymmetry above 1.2 D, Sim K
astigmatism greater than 1.5 D, and SRAX more than 21.15 By
2004 Levy concluded that from studies of familial keratoconus
that IS greater than 0.8 D should be considered as abnormal
and suggestive of forme fruste keratoconus.13 Maeda and Klyce
indices distinguish keratoconus from a variety of conditions.16
The KISA% index (K (IS) Astigmatism (Sim K1Sim K2)
SRAX 100) has been developed by Rabinowitz and is helpful
for analyzing the evolution of suspect patterns over time.7,17 A
novel approach recently reported uses maps of mean curvature,
determined at each point by averaging the curvature along two
principal perpendicular directions, to topographically characterize, diagnose and follow keratoconus and pellucid without
indices.18 Corneal topography is indispensable in the diagnosis
of forme fruste keratoconus. Various indices may be helpful, but
they must be used in conjunction with a thorough clinical
examination and good clinical judgment in the evaluation of
patients for refractive surgery. Over time, as the problem of post
LASIK ectasia has increased, our understanding has grown and
the threshold for diagnosing forme fruste keratoconus has

decreased and may continue to do so.
Corneal topography is not only helpful in the diagnosis of
keratoconus and contact lens tting but in many other ways. It
has improved our understanding of the disease by showing it is
more often bilateral than previously thought. Patients with socalled unilateral disease often have topographic evidence of
forme fruste keratoconus in their normal eye. In one large
series, one third of unilateral cases developed bilateral disease,
and abnormal topographic ndings of higher IS values and
asymmetric bow tie with SRAX were predictive of developing
keratoconus in the second eye.19 Topography is also helpful in
evaluating disease progression in affected eyes over years. The
disease has a familial predisposition more often than previously
recognized if one includes family members with forme fruste
disease. Topography is also very helpful in patient education.
One can compare the image to a geographic map where the
lines are closer together on the side of hills. The obvious
irregularity makes it easier to explain why vision can not be
corrected with glasses or soft contact lenses and why gaspermeable lenses are necessary to create a normal-shaped
anterior surface of the eye with tears lling in the space between
the back of the contact lens and the irregular cornea.
ETIOLOGY AND PATHOGENESIS

Key Features
The cause of keratoconus is unknown.

Eye rubbing is associated with keratoconus.
Long-time use of PMMA lenses is at most a very uncommon
cause.
Biochemical abnormalities and genetics of keratoconus appear
to be complex and are under investigation.
The cause of keratoconus is unknown and probably multifactorial. This is subject of ongoing research on a number of
fronts. It is thought that mechanical trauma and chronic
epithelial injury are involved in the pathogenesis of keratoconus. Eye rubbing is frequent among keratoconus patients.20
The association of atopy with keratoconus is via eye rubbing.8

pathogenesis of keratoconus it would be helpful to routinely
obtain baseline corneal topography prior to initial contact lens
tting in all patients.
Biochemical and molecular abnormalities in keratoconus
are the subject of much ongoing research. Thinning is thought
to be due to an increase in degradative enzyme activity and a
decrease in a number of enzyme inhibitors. Kenney proposed
a unifying working hypothesis for the pathogenesis of
keratoconus.28 First, there is abnormal processing of free
radicals and superoxides generated by UV light exposure in
keratoconus corneas. Second, there is a build-up of destructive
aldehydes in the cornea due to reduced aldehyde dehydrogenase
activity resulting in oxidative stress and damage. Third, cells
that are damaged irreversibly undergo apoptosis resulting in
thinning. Fourth, cells that are damaged reversibly undergo
wound healing which involves upregulation of degradative
enzymes and leads to focal areas of thinning and scarring.
Recent work by this group has shown that keratoconus corneal
buttons obtained at the time of PK (with advanced disease) have
more mitochondrial DNA damage than do normal corneas and
hypothesize that this is both due to oxidative stress and may
add to it.29
The genetic basis of keratoconus is another area of active
investigation and great interest given the potential for gene
therapy of corneal diseases.6 In contrast to granular and lattice
stromal corneal dystrophies, abnormalities in the transforming
growth factor beta-induced gene (BIGH3) are not the cause of
keratoconus.30 One rare form of keratoconus associated with
abnormal retinal function and posterior polymorphous
dystrophy has been determined to be associated with a mutation in the retinal transcription factor VSX1 gene.31 The genetic
basis for most keratoconus remains unknown and appears to be
heterogeneous and complex.32 Recent work by Rabinowitz to
create a data base of genes expressed in human keratoconus
corneas obtained at the time of PK has found abundant expression of a novel gene (designated KC6) of unknown function and
absent expression of another gene, Aquaporin 5, which involves
water channels. The signicance of these ndings is yet to be
determined. There were many genes expressed involved in
apoptosis. Comparison to genes expressed in normal corneas
has not yet been done.
DIFFERENTIAL DIAGNOSIS
Key Features
Pellucid marginal degeneration is in the differential diagnosis of

keratoconus and is distinguished from it by the location of
thinning inferior to the area of maximal ectasia.
Keratoglobus and posterior keratoconus are very different, less
common, nonprogressive congenital disorders.
The major differential diagnosis of keratoconus is PMD, a

similar, but less common noninflammatory ectatic disorder.
Keratoglobus and posterior keratoconus are often included in
the differential diagnosis, but they are very rare, nonprogressive,
congenital conditions that are readily distinguished from
keratoconus. In keratoglobus there is limbus-to-limbus thinning and ectasia with the maximal thinning in the midperiphery. Keratoglobus is associated with connective tissue
disorders such as EhlersDanlos syndrome type VI much more
frequently than keratoconus. Keratoglobus is a true ectasia with
corneal stretching resulting in increased surface area.9 The
cornea may also be enlarged. Keratoglobus patients, especially
those with blue sclera and systemic connective tissue disease,
are at risk for spontaneous corneal rupture, probably because of
CHAPTER 44
Asymmetric keratoconus has been attributed to asymmetric eye

rubbing.21 Chronic eye rubbing is frequent in keratoconus with
not only atopic disease but also floppy eyelids, contact lens
wear, Down syndrome and Lebers congenital amaurosis.22 It is
important to ask keratoconus patients if they rub their eyes. If
patients rub their eyes due to ocular itching, topical mast-cell
stabilizers/antihistamines should be prescribed for treatment.
They should be advised to stop rubbing their eyes. Articial
tears and cool compresses can also be helpful in reducing
eye rubbing.
The floppy-eyelid syndrome is relatively common in
keratoconus, including patients who are not obese. In one series
10% of keratoconus patients had floppy eyelids.23 Keratoconus
tends to be worse on the side the patient with floppy eyelids
sleeps due to the increased mechanical trauma to that side.24
Its easy to diagnose a floppy eyelid: when one everts the upper
lid in the course of a routine eye examination to check for
papillary changes on the superior tarsus, the lid everts
spontaneously. It is important to recognize this condition so it
can be treated with eye shields or taping the lid shut at bedtime,
or by lid-shortening procedures. Floppy-eyelid syndrome
typically occurs in obese patients who need to be evaluated for
sleep apnea. It is unclear whether or not floppy-eyelid syndrome
in nonobese keratoconus patients is associated with sleep
apnea, but medical referral for possible work-up for this
potentially serious and treatable condition is appropriate.
The role of contact lenses, specically polymethylmethacrylate (PMMA) hard contacts, in the cause of keratoconus has
long been debated. Reviewing the evidence suggests that hard
contact lens use is probably a very uncommon cause of
keratoconus. In 1968 Hartstein reported four patients who
developed keratoconus after wearing hard contact lenses, but
two patients had steep keratometry when they were t and two
patients were teenagers so the evidence for the hard contacts
causing keratoconus was relatively weak.25 In 1978 Gasset
reported a large series of patients in which 26.5% (43/162) of
keratoconus patients had a history of wearing PMMA contacts
prior to the diagnosis of keratoconus compared to only 1 patient
(of 1248 controls) who wore soft contact lenses.26 Possible
explanations given for the frequent use of PMMA contact lenses
prior to the diagnosis of keratoconus were that many patients
were at the age when keratoconus typically begins and that
often patients become more myopic prior to the diagnosis. This
topic resurfaced when Macsai reported a retrospective study of
keratoconus patients diagnosed in the 1980s and observed that
patients who wore contacts (89% PMMA for an average of
twelve years) prior to the diagnosis of keratoconus were
signicantly older than those who had not (age 32 years vs 19
years, p<.0001).27 In addition 75% of patients with a history of
contact wear had central cones, compared to 80% without a
contact lens history who had inferior cones (p<.0001). One
limitation of this study is that most of the patients were not
examined by one of the authors, who were cornea specialists,
prior to contact lens tting so that subtle slit-lamp signs of
keratoconus could have been missed. In addition, corneal
topography was not available so it was not possible to diagnose
forme fruste keratoconus in the patients prior to contact lens
tting. However, it is possible that PMMA lenses cause
mechanical trauma contributing to the development of
keratoconus. Many patients who wear contacts also rub their
eyes after removing them. PMMA lenses, however, are a very
uncommon cause of keratoconus, since only 10% of contact
lens wearers in the USA use RGP lenses, and very few wear
PMMA lenses. Concern about this possibility is one reason to
ret people who wear PMMA contacts with gas-permeable
lenses, although a more important reason is to avoid hypoxia.
To further elucidate the possible role of contact lens use in the
557

FIGURE 44.10. (a) In pellucid marginal
degeneration the maximal thinning is inferior to
the area of maximal ectasia. This is best seen
with a narrow slit beam. (b) Corneal topography
in pellucid has a butterfly pattern with
steepening at 4 and 8 oclock in addition to far
inferiorly. Superiorly and inferocentrally there is
flattening. A low sagging cone can have a
similar appearance.
a
SECTION 6
corneal stretching.33 In this series hydrops was very common,

occurring in almost all eyes. The mainstay of treatment for
keratoglobus is spectacles due to the difculty and risks
associated with contact lenses or surgery. If surgery is necessary
for repair of a perforated cornea, a lamellar tectonic limbus-tolimbus procedure is usually done rst.34 Alternative approaches
to support the thin peripheral cornea in keratoglobus in the
absence of perforation have been tried using an onlay of corneal
tissue.35
Posterior keratoconus is an entirely unrelated condition. It is
a mild form of anterior segment dysgenesis in which there is
localized area of posterior increased corneal curvature resulting
in mild corneal thinning with or without scarring.36 The
anterior surface of the cornea is minimally involved, so vision
is usually not greatly affected.
PMD differs from keratoconus in that the area of maximal
thinning is typically below the area of maximal ectasia whereas
in keratoconus the two coincide (Fig. 44.10a). In PMD there is
a band of thinning usually inferiorly within 12 mm of the
limbus, but the thinning can be located elsewhere, even
superiorly. There is a characteristic butterfly or crab-claw
pattern on corneal topography with radial steepening at 4 and
8 oclock and far inferiorly as well as flattening along the 90
axis centrally and superiorly (Fig. 44.10b). Low-sagging cones
can have a similar topographic appearance. The distinction
between PMD and a very low-sagging cone is made by careful
slit-lamp examination, although sometimes it can be difcult to
be certain whether the maximal ectasia is above or coincident
with the area of maximal thinning. The term pellucid means
clear, and it is generally true that in PMD there are no Vogt
striae or Fleisher ring and deep stromal scarring is mild unless
there is a history of hydrops.37,38
MANAGEMENT OF KERATOCONUS AND

PELLUCID
Key Features
558
Rigid gas-permeable lenses do not slow progression of the

disease.
Rigid gas-permeable lenses are indicated to improve vision
when it is inadequate with glasses or soft contact lenses.
Contact lens tting in keratoconus is an art and a science.
Topography and trial lens tting are very helpful. Expertise in
lens tting enhances the success rate.
When patients are intolerant of contact lenses or have reduced
vision due to scarring they are candidates for surgery.
Penetrating keratoplasty is highly successful in keratoconus.
New surgical approaches include intracorneal ring segments
(INTACS) for mild to moderate disease and deep anterior
lamellar keratoplasty.
Gas permeable lenses are indicated and the mainstay of treatment to correct irregular astigmatism caused by keratoconus
and pellucid when vision is inadequate with spectacles or soft
contact lenses. When gas-permeable lenses fail due to decreased
vision or contact lens intolerance, then surgery is indicated. PK
has the highest success rate in keratoconus. It is much more
problematic in pellucid due to difculty getting beyond peripheral thinning. New, promising modalities in the surgical
management of keratoconus and pellucid include intracorneal
ring segments (INTACS) and deep anterior lamellar
keratoplasty (DALK). Despite the achievement of usually
excellent visual acuity in keratoconus, there is good data to
suggest that this condition adversely impacts the quality of life
of patients similar to those patients with grade 34 age-related
macular degeneration.39 Further improvements in management
are necessary.
CONTACT LENSES
It is a common misconception that gas-permeable lenses are
indicated to prevent progression of keratoconus. There is no
evidence to support this. They are indicated only when patients
are dissatised with their vision with glasses or soft contacts. If
a patient has good vision in one eye due to asymmetric disease,
as is often the case, it is often easier for the patient to adjust to
one gas-permeable lens in the eye that needs it rather than to
t both eyes at once. Due to the superior comfort of soft
contacts, many patients prefer to sacrice some vision and not
wear gas-permeable lenses; they are often reassured to learn
that they do not need to wear gas-permeable contacts for
treatment. Advances in frequent replacement toric soft contact
lenses have facilitated their use in patients with mild
keratoconus. Trial lenses can be ordered on the basis of the
patients manifest refraction using minus cylinder. Patients who
wear gas-permeable lenses in only one eye may develop
unilateral, reversible ptosis in that eye. There is some evidence
to suggest that gas-permeable lenses may cause disease
progression. If they are t relatively steep, they can cause
ectasia, and if they are t relatively flat, they can cause apical
scarring, so they should not be used to prevent disease
progression.4042 In addition, patients tend to rub their eyes
after removing gas-permeable contact lenses, and this eye
rubbing may aggravate the condition.
Gas-permeable lenses are the mainstay of treatment for
keratoconus. The expertise of the contact lens tter is an
important factor in the success of contact lenses. Many patients
referred as contact lens failures for PK can often be ret with
contacts successfully.43,44 An attempt at contact lens tting is
almost always indicated prior to PK, especially if the problem is
contact lens intolerance and not decreased visual acuity with
contacts. Successful lens tting is time-consuming for both the
patient and the tter. PK does not always eliminate the need for

t with them due to extreme steepening, but have good visual
acuity.51 They have one diameter (14.3 mm) and a limited range
of base curves (6.508.10 mm). They have relatively low gas
permeability and tend to be tight tting with limited
movement. Use of Softperm lenses has been associated with
corneal swelling and endothelial cell loss due to hypoxia.52,53
Due to concern regarding a tight t and hypoxia, a relatively flat
lens with some movement and even a small amount of edge
pucker represents a desirable t, as long as the patient is
comfortable.46 These lenses tighten up quickly, even in the
course of trial lens tting in the ofce. Patients must be
followed regularly for chronic hypoxic complications such as
neovascularization. In addition giant papillary conjunctivitis is
relatively frequent.50 Regular enzymatic cleaning can be helpful
to prevent and treat this problem. These lenses are cleaned
using soft contact lens solutions. The lenses are fragile and tend
to break along the junction between the gas-permeable center
and soft hydrogel skirt. They are more expensive than gaspermeable lenses and need to be replaced more frequently.
Despite these limitations, patients often strongly prefer these
contacts to gas-permeable lenses due to their increased comfort.
SynergEyes (SynergEyes Inc., Carlsbad, CA) hybrid lenses are
new, more gas permeable, and more durable lenses that are
increasing the success of hybrid lenses.
Other approaches to contact lens tting in keratoconus
include piggyback lenses and scleral lenses. Piggyback lenses
involve the use of a soft lens as a carrier for an RGP lens.
The availability of highly gas-permeable lenses for both
components has decreased the hypoxia associated with this
approach in the past. In addition, a daily-disposable soft lens
decreases the care necessary and probably improves safety.
Nonetheless, people who require this approach are often candidates for surgery. Scleral contacts were used before the
development of corneal lenses, and now with the availability of
RGP scleral lenses there is renewed interest in them. They can
be successful in keratoconus patients who are intolerant of gaspermeable corneal contact lenses as well as other patients
with severe ocular surface disease. To date their use in the
USA has required lens tting at the Boston Foundation for
Sight where Perry Rosenthal, MD developed these lenses
(www.bostonsight.org).
PK
Corneal transplantation is the standard treatment for
keratoconus patients who have decreased visual acuity with
RGP lenses due to corneal scarring or who are contact lens
intolerant. Corneal transplantation has the highest success rate
in keratoconus with clear grafts obtained in over 95% of cases.
In the USA it is the fourth most common indication for PK
after pseudophakic bullous keratopathy, regraft and Fuchs
corneal dystrophy.54 Despite the very high rate of graft clarity
and excellent visual outcome, the recovery of vision is
prolonged, astigmatism is common, graft rejection episodes are
relatively frequent, contact lenses may be necessary, and
patients are at life-long risk for graft rejection and traumatic
wound dehiscence. Other surgical options currently being
evaluated as an alternative to PK for keratoconus include
INTACS, for early and moderate disease, and DALK. In PMD
corneal transplantation is more problematic than in
keratoconus due to peripheral corneal thinning, and a variety of
other techniques have been used.
The frequency of and risk factors for having a PK in
keratoconus following referral to tertiary referral centers have
been studied.4,5557 The likelihood of undergoing PK was ~20%
in studies with four years of follow-up, but was almost 65% in
CHAPTER 44
contact lenses in keratoconus patients, since over a third of

patients wear them postoperatively either due to astigmatism,
anisometropia or for convenience because they wear them in
the other eye.43,45
There are many approaches to tting gas-permeable contact
lenses in keratoconus. There is controversy between the concepts of apical clearance, vaulting and relatively steep t versus
apical touch, bearing or support and relatively flat t. Some
degree of apical touch is tolerated and often desirable in
keratoconus.46 Relatively flat-tting lenses are used most
commonly.47 Corneal topography and in the recent past
keratometry are helpful in lens selection for trial lens tting.
Corneal topography is less accurate in highly irregular corneas
such as keratoconus than in normal eyes, but it still is very
useful in initial trial lens selection and retting. The approach
to lens tting varies depending on if there is a central, round,
nipple cone or a low, sagging, oval cone.14 Multicurve lenses
designed for keratoconus with a steep central curve and wide
flat peripheral curves such as the Soper cone and McGuire
lenses work best in central cones. In recent years, the Rose K
lens is one of these lenses that has been used with success and
improved comfort in patients with advanced central cones.48
Adequate peripheral clearance determined by the pattern of
fluorescein staining has been associated with good contact lens
comfort.49 Even when the base curve of the initial trial lens is
chosen on the basis of the flat peripheral curvature, flatter
lenses are often necessary to obtain a comfortable t without
central microbubbles or discomfort superiorly suggestive of a
tight t (Fig. 44.7).
In general, standard spherical lenses are used successfully to
t many keratoconus patients. They are indicated when there is
corneal steepening inferiorly, as is most commonly the case.
There is corneal flattening superiorly in these patients. The
initial trial lens can be selected on the basis of the flat curvature
above. Larger diameter (usually 9.09.6 mm) lenses are used
more often in these patients than in patients with central cones
in whom smaller lenses (usually 8.58.8 mm) t better. In
patients with PMD or low sagging cones with a similar butterfly
pattern on topography large, relatively flat lenses are used. The
special design Dyna intra-limbal lenses (Lens Dynamics Inc.,
Golden, CO), a large 11.2 mm diameter lens designed for
irregular corneas has been helpful in patients with PMD and
similar topography.50
Fitting gas-permeable lenses is both an art and a science.46
Tight lenses are a frequent cause of lens intolerance. They get
predictably more uncomfortable as the day goes on. Patients
often feel the lens more above near the upper lid than below
because the cornea is flatter superiorly and the lens is tighter
there. It is helpful to t lenses without topical anesthesia,
although one has to wait for reflex tearing to subside in order to
judge the t. It is important to listen to the patient as to where
he or she feels the lens most and then compare the curvature of
the cornea by topography to the base curve of the lens and make
adjustments accordingly to improve the lens corneal alignment.
Loose lenses cause variable discomfort. Both tight and loose
lenses can cause central supercial punctate keratitis. In
general, moderate- or high-Dk lens materials should be used.
Sometimes the material makes a difference in contact lens
tolerance, and one should consider changing the material if the
t seems optimal and yet the patient is uncomfortable.
Moderately high-Dk lenses tend to wet better and develop
coating less than very high-Dk lenses.
Softperm lenses (Wesley Jessen Corp, Des Plaines, Il) are a
hybrid lens with a gas-permeable lens center and a hydrogel
skirt. They can be used with good success in keratoconus when
patients are intolerant of gas-permeable lenses or can not be
559
SECTION 6

a series with a longer average follow-up of almost 11 years,
despite attempted contact retting prior to keratoplasty.4,55,56
Steep baseline keratometry was a risk factor for PK in all series.
Black race was identied as a signicant risk factor for PK in
one series.4 In PMD the majority of patients are managed
medically with spectacles or contact lenses, in part because of
increased difculty and worse prognosis for surgery, as well as
the absence of central cornea scarring.58,59
Special considerations in performing a phakic PK in
keratoconus include preoperative planning of graft size and
efforts to prevent positive pressure during surgery. The size of
the host trephination is determined by the extent of thinning
and ectasia observed at the slit lamp. One-quarter millimeter
larger or sometimes same-size donor buttons are used to reduce
postoperative myopia. Despite preoperative digital massage and
intravenous mannitol there is often positive pressure during
surgery due to low scleral rigidity and scleral collapse. Reverse
Trendelenburg position is helpful to reduce positive pressure,
especially in obese patients.
A xed dilated pupil of unknown etiology (UrretsZavalia
syndrome) has been reported after PK for keratoconus, but is
rare.60 It is associated with iris ischemia and is more likely
related to intraoperative positive pressure than postoperative
pressure spikes which are uncommon in keratoconus.61,62
In pellucid corneal degeneration, larger grafts, often
decentered inferiorly are necessary to get beyond the area of
inferior thinning.63 When this is not possible peripheral lamellar procedures are used, recently in combination with PK.64,65
Despite the high success rate and excellent visual acuity after
PK for keratoconus, meticulous, indenite postoperative care
is necessary. Rejection episodes occur in ~30% of patients,
but rarely result in graft failure.45,6668 Larger graft size (host
greater than or equal to 8.25 mm) is a risk factor for rejection.
There has been concern that bilateral transplantation increases
the risk for rejection, but this has been shown not to be the case
in more recent studies.68,69
Astigmatism after PK for keratoconus is relatively common.
It can usually be managed with RGP lenses, but relaxing
incisions, compression sutures, laser refractive surgery, and/or
wedge resections are sometimes done. Progressive late astigmatism is also relatively common more than 10 years after PK.
It is associated most frequently with thinning of the inferior
grafthost junction and host periphery due to progressive
disease rather than recurrent keratoconus in the graft. In a
series of keratoconus patients followed for 20 years after PK,
keratometric astigmatism was stable for the rst seven years
after suture removal and then increased progressively in
association with thinning of the grafthost junction consistent
with disease progression in the host.3 This has been observed by
others and should be looked for by careful slit-lamp examination.70 Late recurrences of keratoconus within the donor
cornea conrmed by histopathology after repeat corneal
transplantation have been reported, but are infrequent.71
OTHER SURGERY
Currently, INTACS and deep anterior lamellar keratoplasty are
alternative surgical approaches for keratoconus, but other
procedures have also been tried. Epikeratoplasty was performed
in the past, but lost favor due to poor visual outcome. Excimer
laser procedures are generally contraindicated in keratoconus,
except occasionally excimer phototherapeutic keratoplasty
(PTK) can be performed to remove elevated nodular scars, when
they can not be readily shaved off with a blade, in order to
improve contact lens tolerance. Riboflavin/ultraviolet-A-induced
collagen crosslinking has been reported to stop progression of
keratoconus and induce regression in some patients.72
Some success with INTACS to reduce myopia and astigmatism and improve uncorrected and spectacle corrected vision in
keratoconus without central scarring has been reported.73 The
role of INTACS in keratoconus remains unknown, but patients
with milder keratoconus appear to have a better outcome.74
Complications are relatively frequent in patients with moderate
to advanced disease, especially thinning over the implants
resulting in their exposure.75 Optimal indications and technique for INTACS for keratoconus remain to be determined.
Deep anterior lamellar keratoplasty, an alternative to fullthickness PK, is gaining popularity. It has the advantage of
avoiding the risk of endothelial rejection, since the healthy
endothelium of the patient is left in place. Although the
procedure is technically challenging, early reports suggest that
the visual results may be as good as after PK.76
CONCLUSIONS
Keratoconus is a condition affecting ~150 000 people in the
USA. Much is known about the diagnosis and management,
and yet there are many unresolved issues regarding the
pathogenesis and treatment which are areas of current research.
The focus in caring for keratoconus patients is visual rehabilitation. In most cases this is achieved by correcting irregular
astigmatism using contact lenses and by surgery when contacts
fail. Patients are often myopic and require comprehensive eye
care. Intraocular pressure (IOP) measurement may be falsely
low due to corneal thinning in keratoconus. However, optic
nerve changes suspicious for glaucoma should be further
evaluated as these patients may be susceptible to progressive
glaucomatous optic neuropathy at low pressures.
Although the prognosis is good for patients with keratoconus,
it is of great concern that the disease appears to have a decidedly
negative impact on the quality of life.39 Studies have tried to
address so-called difcult personality traits associated with
keratoconus with variable results.7780 It is very possible that
personality issues, if they exist, are secondary to the condition.
Improved management of this disease may lessen the burden
for patients and improve not only their vision but also their
quality of life.
REFERENCES
560
1. Kennedy RH, Bourne WM, Dyer JA:

A 48-year clinical and epidemiologic study
of keratoconus. Am J Ophthalmol 1986;
101:267273.
2. Pobelle-Frasson C, Velou S, Huslin V, et al:
Keratoconus: what happens with older
patients? J Fr Ophthalmol 2004;
27:779782.
3. Toledo JA, Paz MF, Barraquer RI, Barraquer
J: Long-term progression of astigmatism
after penetrating keratoplasty for
keratoconus. Cornea 2003; 22:317323.
4. Tuft SJ, Moodaley LC, Gregory WM et al:

Prognostic factors for the progression of
keratoconus. Ophthalmology 1994;
101:439447.
5. Bawazeer AM, Hodge WG, Lorimer B: Atopy
and keratoconus: a multivariate analysis.
6. Rabinowitz YS: Keratoconus. Surv
Ophthalmol 1998; 42:297319.
7. Rabinowitz YS: The genetics of
keratoconus. Ophthalmol Clin N Am 2003;
16:607620.
8. Feder RS, Kshettry P: Noninflammatory

ectatic disorders. In: Krachmer JH,
Mannis MJ, Holland EJ, eds. Cornea.
Philadelphia: Elsevier Mosby; 2005:955974.
9. Smolek MK, Klyce SD: Is keratoconus a
true ectasia? Arch Ophthalmol 2000;
118:11791186.
10. Zadnik K, Barr JT, Edrington TB, et al:
Corneal scarring and vision in keratoconus:
a baseline report from the Collaborative
Longitudinal Evaluation of Keratoconus
(CLEK) Study, Cornea 2000; 19:804812.
11. Grewal S, Laibson PR, Cohen EJ,

Rapuano CJ: Acute hydrops in the cornea
ectasias: associated factors and outcomes.
Tr Am Ophthalmol Soc 1999; 97:187198.
12. Corneal topography. American Academy of
Ophthalmology. Committee on ophthalmic
procedures assessment cornea panel.
Ophthalmology 1999; 106:16281638.
13. Levy D, Hutchings H, Rouland JF, et al:
Videokeratographic anomalies in familial
keratoconus. Ophthalmology 2004;
111:867874.
14. Perry HD, Buxton JN, Fine BS: Round and
oval cones in keratoconus. Ophthalmology
1980; 87:905-909.
15. Rabinowitz YS: Videokeratographic
indices to aid in screening for keratoconus.
J Refract Surg 1995; 11:371379.
16. Maeda N, Klyce SD, Smolek MK:
Comparison of methods for detecting
keratoconus using videokeratography.
Arch Ophthalmol 1995;113:870874.
17. Rabinowitz YS, Rasheed K: KISA% index:
a quantitative videokeratography algorithm
embodying minimal topographic criteria for
diagnosing keratoconus. J Cataract Refract
Surg 1999; 25:13271335.
18. Tang M, Shekhar R, Miranda D, Huang D:
Characteristics of keratoconus and pellucid
marginal degeneration in mean curvature
maps. Am J Ophthalmol 2005;
140:9931001.
19. Li X, Rabinowitz YS, Rasheed K, Yang H:
Longitudinal study of the normal eyes in
unilateral keratoconus patients.
20. Tretter T, Rabinowitz YS, Yang H, et al:
Aetiological factors in keratoconus.
21. Jafri B, Lichter H, Stulting RD: Asymmetric
keratoconus attributed to eye rubbing.
Cornea 2004; 23:560564.
22. Krachmer JH: Eye rubbing can cause
23. Culbertson WW, Tseng SCG: Corneal
disorders in floppy eyelid syndrome.
Cornea 1994; 13:3342.
24. Donnenfeld ED, Perry HD, Gibralter RP,
et al: Keratoconus associated with floppy
eyelid syndrome. Ophthalmology 1991;
98:16741678.
25. Harstein J: Keratoconus developed in 4
patients wearing conreal contacts. Arch
Ophthalmol 1968; 80:345246.
26. Gasset AR, Houde WL, GarciaBengochea M: Hard contact lens wear as
an environmental risk in keratoconus.
27. Macsai MS, Varley GA, Krachmer JH:
Development of keratoconus after contact
lens wear. Arch Ophthalmol 1990;
108:534538.
28. Kenney MC, Brown DJ, Rajeev B:
The elusive causes of keratoconus: a
working hypothesis CLAO 2000; 26:1013.
29. Atiliano SR, Coskun P, Chwa M, et al:
Accumulation of mitochondrial DNA
damage in keratoconus corneas. Invest
30. Udar N, Kenney MC, Chalukya M, et al:
Keratoconus No association with the
transforming growth factor beta-induced
gene in a cohort of American patients.
Cornea 2004; 23:1317.
31. Heon E, Greenberg A, Kopp KK, et al:
VSX1: a gene for posterior polymorphous
dystrophy and keratoconus. Hum Mol
Genet 2002; 11:10291036.
32. Rabinowitz YS, Dong L, Wistow G: Gene

expression prole studies of human
kertaoconus cornea for the NEIBank:
a novel cornea-expressed gene and the
absence of transcripts for aquaporin 5.
46:12391246.
33. Cameron JA: Keratoglobus. Cornea 1993;
12:124130.
34. Macsai MS, Lemley HL, Schwartz T:
Management of oculus fragilis in
Ehlers-Danlos Type VI. Cornea 2000;
19:104107.
35. Kanellopoulos AJ, Pe LH: An alternative
surgical procedure for the management of
keratoglobus. Cornea 2005; 24:10241026.
36. Waring GO, Rodrigues MM, Laibson PR:
Anterior chamber cleavage syndrome.
A stepladder classication. Surv
37. Schlaeppi V: La dystrophie marginale
inferieure pellucide de la cornee. Probl
Actuels Ophthalmol 1957; 1:672677.
38. Cameron JA: Deep corneal scarring in
pellucid marginal corneal degeneration.
Cornea 1992; 11:309310.
39. Kymes SM, Walline JJ, Zadnik K, et al:
Quality of life in keratoconus. Am J
Ophthalmol 2004; 138:527535.
40. McMonnies CW: Keratoconus fittings.
Eye Contact Lens 2004; 30: 147155.
41. McMonnies CW: The biomechanics of
keratoconus and rigid contact lenses.
Eye Contact Lens 2005; 31:8092.
42. Barr JT, Wilson BS, Gordon MO, et al:
Estimation of the incidence and factors
predictive of corneal scarring in the
Collaborative Longitudinal Evaluation of
Keratoconus (CLEK) Study. Cornea 2006;
25:1625.
43. Smiddy WE, Hamburg TR, Kracher GP,
Stark WJ: Keratoconus Contact lens or
keratoplasty? Ophthalmology 1988;
95:487492.
44. Belin MW, Fowler WC, Chambers WA:
Keratoconus Evaluation of recent trends
in the surgical and nonsurgical correction
of keratoconus. Ophthalmology 1988;
95:335339.
45. Silbiger JS, Cohen EJ, Laibson PR: The
rate of visual recovery after penetrating
keratoplasty for keratoconus. CLAO 1996;
22:266269.
46. Lembach RG: Use of contact lenses for the
management of keratoconus. Ophthalmol
Clin N Am 2003; 16:383394.
47. Edrington TB, Szczotka LB, Barr JT, et al:
Rigid contact lens tting relationships in
keratoconus. Optom Vis Sci 1999;
76:692699.
48. Betts AM, Mitchell GL, Zadnik K: Visual
performance and comfort with the Rose K
lens for keratoconus. Optom Vis Sci 2002;
79:493501.
49. Edrington TB, Gundel RE, Libassi DP, et al:
Variables affecting rigid contact lens
comfort in the Collaborative Longitudinal
Evaluation of Keratoconus (CLEK) Study.
Optom Vis Sci 2004; 81:182188.
50. Ozbek Z, Cohen EJ: Use of the intralimbal
rigid gas permeable lenses for pellucid
marginal degeneration, keratoconus, and
after penetrating keratoplasty. Eye Contact
Lens 2006; 32:3336.
51. Chung CW, Santim R, Weng WJ, Cohen EJ:
Use of Softperm contact lenses when rigid
gas permeable lenses fail. CLAO 2001;
27:202208.
52. Owens H, Watters G, Gamble G: Effect of

Soptperm lens wear on corneal thickness
and topography: a comparison between
and keratoconic and normal corneae.
CLAO 2002; 28:8387.
53. Edmonds CR, Wung SF, Husz MJ,
Pemberton B: Corneal endothelial cell
count in keratoconus patients after contact
lens wear. Eye Contact Lens 2004;
30:5458.
54. Cosar CB, Sridhar MS, Cohen EJ, et al:
Indications for penetrating keratoplasty and
associated procedures, 1996-2000. Cornea
2002; 21:148151.
55. Lass JH, Lembach RG, Park SB, et al:
Clinical management of keratoconus.
56. Sray WA, Cohen EJ, Rapuano CJ,
Laibson PR: Factors associated with the
need for penetrating keratoplasty in
57. Reeves SWW, Stinnett S, Adelman RA,
Afshari NA: Risk factors for the progression
to penetrating keratoplasty in patients with
140:607611.
58. Sridhar MS, Mahesh S, Bansal AD, et al:
Pellucid marginal corneal degeneration.
Ophthalmology 2004; 111:11021107.
59. Tzelikis PF, Cohen EJ, Rapuano CJ, et al:
Management of pellucid marginal corneal
degeneration. Cornea 2005; 24:555560.
60. Jastaneiah S, Al-Towerki A, Al-Assiri A:
A xed dilated pupil after penetrating
keratoplasty for macular corneal dystrophy
and keratoconus. Am J Ophthalmol 2005;
140:484489.
61. Tuft SJ, Buckley RJ: Iris ischaemia
following penetrating keratoplasty for
keratoconus (Urrets-Zavalia syndrome).
Cornea 1995; 14:618622.
62. Chien AM, Schmidt CM, Cohen EJ, et al:
Acute glaucoma after penetrating
keratoplasty. Am J Ophthalmol 1993;
115:711714.
63. Varley GA, Macsai MS, Krachmer JH:
The results of penetrating keratoplasty for
pellucid marginal corneal degeneration.
Am J Ophthalmol 1990; 110:149152.
64. Cameron JA: Results of lamellar crescentic
resection for pellucid marginal corneal
113:296302.
65. Rasheed K, Rabinowitz YS: Surgical
treatment of advanced pellucid marginal
degeneration. Ophthalmology 2000;
107:18361840.
66. Olson RJ, Pingree M, Ridges R:
Penetrating keratoplasty for keratoconus:
a long-term review or results and
complications. J Cataract Refract Surg
2000; 26:987991.
67. Javadi MA, Motlagh BF, Jafarinasab MR,
et al: Outcomes of penetrating keratoplasty
in keratoconus. Cornea 2005; 24:941946.
68. Donshik PC, Cavanagh HD, Goruchoff AS,
Dohlman CH: Effect of bilateral and
unilateral grafts on the incidence of
rejections in keratoconus. Am J Ophthalmol
1979; 87:823826.
69. Musch DC, Meyer RF: Risk of endothelial
refection after bilateral penetrating
96:11391143.
70. Lim L, Pesudovs K, Goggin M, Coster DJ:
Late onset post-keratoplasty astigmatism
in patients with keratoconus. Br J
Ophthalmol 2004; 88:371376.
CHAPTER 44
561
SECTION 6
71. Bourges JL, Savoldelli M, Dighiero P, et al:

Recurrence of keratoconus characteristics.
Ophthalmology 2003; 110:19201925.
72. Wollensak G, Spoerl E, Seiler T:
Riboflavin/ultraviolet-A-induced collagen
crosslinking for the treatment of keratoconus.
Am J Ophthalmol 2003; 135:620627.
73. Colin J, Cochener B, Savary G, et al:
INTACS inserts for treating keratoconus.
Ophthalmology 2001; 108:14091414.
74. Levinger S, Pokroy R: Keratoconus
managed with intacs. Arch Ophthalmol
2005; 123:13081314.
562
75. Kanellopoulos AJ, Pe LH, Perry HD,

Donnenfeld ED: Modied intracorneal ring
segment implantations (INTACS) for the
management of moderate to advanced
76. Watson SL, Ramsay A, Dart JK, et al:
Comparison of deep lamellar keratoplasty
and penetrating keratoplasty in patients
with keratoconus. Ophthalmology 2004;
111:16761682.
77. Farge EJ, Baer PE, Adams GL, Paton D:
Personality correlates of keratoconus.
In: Fann WE, ed. Phenomenology and
treatment of psychophysiological disorders.

Bridgeport, Ct: Spectrum Publications;
1982:243250.
78. Mannis MJ, Morrison TL, Zadnik K, et al:
Personality trends in keratoconus. Arch
Ophthalmol 1987; 105:798800.
79. Swartz NG, Cohen EJ, Scott DG, et al:
Personality and keratoconus. CLAO 1990;
16:6264.
80. Giedd KK, Mannis MJ, Mitchell GL,
Zadnik K: Personality in keratoconus in a
sample of patents derived form the
internet. Cornea 2005; 24:301307.
CHAPTER
45
Corneal Manifestations of Metabolic Disease

Kristin M. Hammersmith and Christopher J. Rapuano
The optical clarity of the cornea permits visualization of subtle

deposition of metabolites not possible in other tissues of the
body. This chapter concentrates on disorders of metabolism
that have clinically observable changes in the cornea that may
help to establish or conrm a systemic diagnosis. Such disorders may indicate disturbance in aspects of metabolism
involving amino acids, lipids, or complex carbohydrates. Most
metabolic diseases are inherited on an autosomal recessive
basis. Hunters syndrome and Fabrys disease are two notable
exceptions, both of which are X-linked. Generally, the metabolic
disorder is a result of an enzymatic deciency causing accumulation of substrate either locally or after transport in the
blood stream.
As an organizational device we divide our descriptions into
disorders of metabolism involving amino acids, lipids, complex
carbohydrates, purines, and metals.
features can include variable degrees of mental retardation,

seizures, and multiple congenital anomalies.5
Ocular symptoms and ndings occur early in tyrosinemia
type 2 and may even be the presenting manifestation of the
disease. Keratoconjunctivitis with photophobia may appear
before the patient is 2 weeks of age.6,7 Corneal opacities are
bilateral supercial punctate crystalline deposits that may
assume a dendritiform pattern leading to an ulcerative keratitis
and a mistaken diagnosis of herpes simplex virus (HSV) keratitis. Peripheral corneal vascularization may develop (Fig. 45.2).
In contrast to HSV, the dendritiform lesions are often bilateral
and do not have terminal bulbs and corneal sensation is
normal.2 The ocular ndings may appear months before the
hyperkeratotic skin lesions on the hands and feet, which can be
DISORDERS OF AMINO ACID METABOLISM

TYROSINEMIA (TYROSINOSIS)
Key Features
Autosomal recessive
Two forms: type 1, most common, no corneal involvement
type 2, can develop superficial punctate crystalline lesions,
often in a dendritiform pattern, that may mimic HSV keratitis
Tyrosine is an organic amino acid precursor in the metabolic

pathways of amines, which include thyroid hormones and the
neurotransmitters epinephrine, norepinephrine, dopamine, and
tyramine. Tyrosine can be derived from phenylalanine by
hydroxylation or from proteins (Fig. 45.1). Elevated levels of
serum tyrosine can occur in transient neonatal tyrosinemia
as well as in two autosomal recessive conditions: tyrosinemia
type 1 (hepatorenal tyrosinemia) and tyrosinemia type 2
(oculocutaneous tyrosinemia, RichnerHanhart syndrome).
Tyrosinemia type 1, the more common variant, is caused by
a deciency of fumarylacetoacetate hydrolase. Corneal
changes have not been reported with this disorder.1 Tyrosinemia
type 2 (RichnerHanhart syndrome) is due to a deciency
of hepatic tyrosine aminotransferase, the gene for which is
located on chromosome 16q22.2 In this disorder, serum tyrosine levels range from 2.5 to 25 times normal. Urinary tyrosine
and several of its metabolites are found in high concentrations.3
Differentiation from tyrosinemia type 1 can be made by the lack
of hepatorenal disease and the lack of the inhibitor effect of the
patients serum on D-aminolevulinic acid dehydrogenase
activity, which is specic to hepatorenal tyrosinemia.4 Systemic
FIGURE 45.1. Points of enzymatic deficiencies in organic amino acid

metabolic pathways in tyrosinemia type 2 and alkaptonuria.
563
FIGURE 45.2. Corneal changes in tyrosinemia type 2. (a) The left eye shows
peripheral neovascularization, marked irregularity of the epithelium, patchy
opacities, and loss of corneal transparency. (b) The right eye shows even
more extensive involvement. (c) After 6 weeks of therapy, there is marked
clearing of the lesions.
(ac) From Goldsmith LA: Cutaneous changes in errors of amino acid metabolism:
tyrosinemia, phenylketonuria, and argininosuccinic aciduria. In: Fitzpatrick TB, Eisen AZ,
Wolff K, et al, eds. Dermatology in general medicine. 3rd edn. New York: McGraw-Hill;
1987:1636. Copyright 1987 by McGraw-Hill, Inc. Used by permission of McGraw-Hill
Book Company.
SECTION 6
painful enough to prevent walking. The skin lesions begin as

bullae and erosions that progress to white-yellow hyperkeratotic
plaques and papules.8 Treatment with dietary restriction of
phenylalanine and tyrosine results in complete reversal of
both ocular and dermatologic abnormalities (Fig. 45.3).2,7,9
Nystagmus, strabismus, conjunctival thickening, and cataract
have also been reported in association with RichnerHanhart
syndrome, as has variable mental retardation.
Transient neonatal tyrosinemia is a temporary biochemical
abnormality affecting premature infants or infants who ingest a
high-protein diet such as evaporated milk formula. Crystalline
corneal opacities characteristic of tyrosinemia type 2 have been
reported in transient neonatal tyrosinemia. The subepithelial
crystals can completely reabsorb within 5 days of normalization
of plasma tyrosine concentrations.10
ALKAPTONURIA (OCHRONOSIS)
a
Key Features
564
Autosomal recessive
The yellowbrown pigmentation that develops in sclera,
cartilage and tendons is termed ochronosis
Alkaptonuria is a rare, recessively inherited disorder of amino

acid metabolism caused by the lack of the hepatic and renal
enzyme homogentisate 1,2-dioxygenase (HGO). The locus for
this enzyme is on chromosome 3q21-23.11 Large amounts of
FIGURE 45.3. (a) Diffuse plantar hyperkeratosis in an adult with

tyrosinemia. (b) The hyperkeratosis cleared on a low-tyrosine, lowphenylalanine diet without topical treatment.
(a and b) From Goldsmith LA: Cutaneous changes in errors of amino acid
metabolism: tyrosinemia, phenylketonuria, and argininosuccinic aciduria.
In: Fitzpatrick TB, Eisen AZ, Wolff K, et al, eds. Dermatology in general
medicine. 3rd edn. New York: McGraw-Hill; 1987:1639. Copyright 1987
by McGraw-Hill, Inc. Used by permission of McGraw-Hill Book
Company.)

homogentisic acid are excreted unchanged into the urine, which
turns dark black if the urine becomes alkaline from standing
exposed to air. There is a low prevalence of alkaptonuria
(1:100 000250 000) in most ethnic groups, with the exception
of Slovaks, in whom the incidence rises to 1:19 000.12
The normal metabolism of phenylalanine and tyrosine
produces homogentisic acid, which is cleaved by HGO (see
Fig. 45.1). In the absence of HGO, homogentisic acid is excreted
into the urine, where it is oxidized into a melanin-like product.
Ochronosis is the name given to the ochre, yellow-brown
pigmentation that begins to appear in the collagen of the sclera,
cartilage, and tendons toward the end of the second decade
of life (Fig. 45.4). Tracheal, bronchial, laryngeal, costal, and
auricular cartilages are involved as is the dura mater.13,13a Over
a period of time, the pigment deposition increases and takes on
a darker bluishblack appearance that may be mistaken for
melanoma.14 The arthritis, which is the major clinical morbidity, has an insidious, progressive character that may begin
in the fourth decade of life.15,16 It may lead to incapacitating
kyphosis and joint immobility. Cardiovascular disease, atherosclerosis, prostatic, and renal lithiasis may also occur.
The cornea and more characteristically the sclera become
pigmented in ochronosis without causing a decrease in visual
function.17 Corneal involvement, when it occurs, is usually
limited to the deposition of ne, brown, oil-like droplets at the
level of Bowmans layer and the anterior stroma near the limbus
in the horizontal meridians.14,16 By slit-lamp examination,
scleral as well as conjunctival and subconjunctival pigment is
seen in the nasal and temporal interpalpebral area anterior to
the insertions of the horizontal muscles in patches, ecks, or
spots.14 Over a period of time, the patches coalesce to form
triangular, deeply pigmented areas in the same location as
pingueculae.
FIGURE 45.4. Alkaptonuria showing the yellow-brown pigmentation

in the paralimbal cornea with a predilection of deposition anterior to
the insertion of the horizontal muscle tendons.
From Donaldson DD: Atlas of external diseases of the eye. 2nd edn. Cornea and
sclera. St Louis: CV Mosby; 1980:65.
CYSTINOSIS
Key Features
Autosomal recessive
There are three forms of cystinosis: infantile (most severe),
intermediate, and adult (least severe)
Corneal deposition of multiple fine, needle-shaped refractile
crystals are noted in all three forms
When advanced, filamentary keratopathy, recurrent erosions
and decreased vision from dense crystals can develop
Renal failure develops within the first decade in the infantile
form, typically requiring kidney transplantation
FIGURE 45.5. Cystinosis. Fine, needle-shaped refractile crystals can

be seen within the corneal stroma. All forms of the disease show
corneal changes.
From Mandel ER, Wagoner MD: Atlas of corneal disease. Philadelphia: WB
Saunders;1989:47.
Cystinosis is an autosomal recessive hereditary disorder in

which free cystine accumulates intracellularly within
lysosomes.18 The exact metabolic defect is not yet dened but is
known to involve defective lysosomal cystine transport.19 The
cystinosis gene has been linked to markers on the short arm of
chromosome 17.20 There are three forms of cystinosis based on
age of onset: infantile, intermediate, and adult. All forms of
cystinosis show characteristic corneal deposition of ne, needleshaped refractile crystals, which have been described as tinsellike opacities pathognomonic of the condition.21 While corneal
ndings are the same in the various forms, the systemic clinical
ndings vary widely. A routine ophthalmic examination may
lead to the diagnosis (Fig. 45.5).22
Infantile nephropathic cystinosis is the most common and
severe form of cystinosis. Children present toward the end of
the rst year of life with recurrent fever and dehydration. By the
rst year, Fanconis syndrome is established, involving a complex dysfunction of the proximal renal tubules and metabolic
bone disease. There is also associated renal rickets, which may

be related to impaired renal conversion of 25-hydroxyvitamin
D3 to 1,25-dihydroxyvitamin D3.23 Failure to thrive is a prominent feature of nephropathic cystinosis. Affected children may
have pale irides24 and hypopigmentation of skin and hair for
their race.25 Over time as crystalline accumulation increases,
glomerular damage occurs, resulting in severe renal failure by
the age of 10 years. Renal transplantation is life-saving but does
not prevent progressive damage to other organs, including
the eyes.26
The characteristic needle-like refractile corneal crystals present in the rst year of life, usually preceding the full-blown
renal disease. Crystals are initially deposited in the peripheral
and anterior corneal stroma. With age, deposition proceeds posteriorly and centripetally, so that by the age of 7 years, crystals
CHAPTER 45
565
SECTION 6
TABLE 45.1. Clinical Manifestations of the Three Major Forms of Cystinosis

Manifestation
Infantile Nephropathic
Intermediate Juvenile
Benign Adult
Presenting signs
Failure to thrive; fever; dehydration;

Fanconis syndrome; rickets; photophobia
Renal dysfunction
Incidental finding on eye examination
Age at presentation
Late infancy
18 mo17 yr
Incidental finding at any age
Ocular findings
Corneal crystals by age 1 yr; increase with age

Patchy peripheral retinal pigment epithelial atrophy
Corneal crystals
Variable retinal findings
Corneal crystals
None
Growth
Third percentile
Nearly normal
Normal
Free cystine content

of leukocytes
80 times normal
30 times normal
can be found within or on the endothelial surface. The depth

of stromal deposition is greater in the periphery and symmetric
between the two eyes. The crystals appear more dense and
larger in the anterior stroma.27 In a study of children, who had
undergone renal transplantation, all showed full-thickness
corneal involvement that was so dense in one 18-year-old
patient that it resembled mutton fat keratitic precipitates. Of
the same group, two patients had band keratopathy.24 Crystals
can also be seen in the iris and lens.
Corneal thickness is increased in patients with nephropathic
cystinosis, even at a young age.28 This may be an indication of
subclinical dysfunction of the endothelial or epithelial cells.
Corneal sensation is also signicantly reduced.29 Photophobia
and sensitivity to glare are common and related to light
scattering by the crystals rather than to retinal problems.30
Glare disability is correlated with age and the density of
clinically observable corneal crystals.31 With time, however,
recurrent erosions may become a major clinical problem
requiring bandage contact lenses or corneal transplant for the
relief of symptoms.30,32,33 Supercial punctate keratopathy and
lamentary keratopathy are more common in older patients.
Band keratopathy, corneal neovascularization, and posterior
synechiae are also seen in older patients.34 Symmetric patchy
depigmentation of the peripheral retina is a constant nding in
nephropathic cystinosis and may be noted even before the
corneal changes are visible.35
Visual acuity is generally normal in the early stages of
cystinosis. As patients are living longer after renal transplantation, impaired visual function may result from abnormal
retinal function, posterior synechiae, glaucoma,36 and
hemorrhagic retinopathy.26,30 Progressive neurologic dysfunction, primarily motor incoordination and hypotonia, may
occur in young adults.39
An intermediate juvenile form of cystinosis exists with the

onset of renal dysfunction between 18 months and 17 years of
age with corneal crystals and variable retinopathy. Growth is
nearly normal.
The adult form is characterized by the asymptomatic presence of corneal crystals without retinopathy or renal dysfunction.37 Diagnosis can be conrmed by conjunctival biopsy.
The diagnosis of cystinosis can be made by measuring the
free cystine content in leukocytes or cultured amniotic cells.
Those with infantile nephropathic cystinosis have values 80
times normal (Table 45.1).38
Treatment is symptomatic, addressing electrolyte and uid
imbalances caused by the renal disease, vitamin D therapy for
rickets, and thyroid hormone for hypothyroidism. When renal
function fails, dialysis or renal transplantation is frequently
necessary in patients between 6 and 12 years of age. After
corneal transplantation the grafted cornea generally remains
free of crystals.30,33,39 Oral cysteamine is a more specic therapy
to reduce intracellular cystine levels. This therapy has met with
success in maintaining renal glomerular function and improving growth but does not relieve the symptoms of the Fanconi
syndrome.40,41 Similarly, it has not been shown to retard the
rate of corneal crystal accumulation even after 8 years of
therapy.42 In contrast, topical cysteamine given hourly as eye
drops in doses ranging from 0.1 to 0.5% has been successful in
reducing the number of corneal crystals in treated eyes of young
children.42,43 Reducing the frequency of instillations to four
times a day over a 7-month period was not as successful in
reducing crystal formation, although it did reduce photophobia
in a 21-year-old.44 The higher concentration, 0.5%, was more
effective than the 0.1% concentration in reducing crystals in
older patients. The best therapy may be prophylaxis with the
early institution of cysteamine eye drops (Fig. 45.6).45
FIGURE 45.6. Slit-lamp photograph of a

placebo-treated right eye (a) showing crystals
in the central cornea compared with the
cysteamine-treated left eye of a 26-month-old
child showing no crystals (b).
(a and b) From Kaiser-Kupfer MI, Gazzo MA,
Datiles MB, et al: A randomized placebo-controlled
trial of cysteamine eye drops in nephropathic
cystinosis. Arch Ophthalmol 1990; 108:689.
Copyright 1990, American Medical Association.
566
DISORDERS OF LIPOPROTEIN AND LIPID

METABOLISM
HYPERLIPOPROTEINEMIAS
Lipoproteins in plasma allow the transport of cholesterol,
triglycerides, phospholipids, and proteins throughout the
systemic circulation. Lipoproteins consist of chylomicrons, very
low-density lipoproteins (VLDLs), LDLs, and HDLs. Dietary
lipids are packaged into chylomicrons during absorption by
intestinal mucosa. They are modied and selectively transported to specic tissues under the control of lipid-cleaving
enzymes (lipases), plasma lipoproteins, and the corresponding
lipoprotein receptors on tissues. Apolipoproteins on the surface
of lipoprotein particles assist in the directed transport and
uptake of nutritive lipid and protein at specic tissue sites
throughout the body. At least ve different electrophoretic
patterns of abnormal elevations of lipoprotein levels have been
described that may be associated with secondary systemic and
ocular defects (Table 45.2).
Type 1 hyperlipoproteinemia (hyperchylomicronemia), a rare
autosomal recessive disorder, is characterized by a massive
elevation of plasma chylomicron levels and a corresponding
increase in triglyceride levels.48 The genetic locus is 8p22. This
disorder may be associated with hepatosplenomegaly, repeated
episodes of abdominal pain, central nervous system dysfunction, recurrent pancreatitis, lipemia retinalis, and palpebral
or diffuse eruptive xanthomas. Early atherosclerosis and corneal
arcus are usually not features.49 Type 2 hyperlipoproteinemia
This diverse group of disorders affecting lipoprotein and lipid

metabolism frequently has associated ocular manifestations.
DYSLIPOPROTEINEMIAS
Key Features
Disorders of lipid metabolism such as lecithin-cholesterol
acyltransferase (LCAT) deficiency, Tangier disease, and fish eye
disease are important to recognize as they may be associated with
coronary artery and peripheral vascular disease.
The dyslipoproteinemias are a group of lipid metabolism disorders that include the hyperlipoproteinemias, lecithincholesterol acyltransferase (LCAT) deciency, Tangier disease
(familial high-density lipoprotein (HDL) deciency), and sh
eye disease. The recognition of these ocular ndings, especially
corneal arcus and xanthelasma, is important, as they may be
associated with coronary artery and peripheral vascular
disease.46 The presence of corneal arcus in men under 50 years
of age may be a harbinger of developing coronary artery disease
(Fig. 45.7).47
FIGURE 45.7. Corneal arcus. (a) Note the clear

zone between the corneal limbus and the
peripheral stromal deposition of phospholipid,
cholesterol esters, and triglycerides. (b) The
lipid deposition is limited to the paralimbal
cornea. The central cornea is clear.
Characteristic
Type I:
Hyperchylomicronemia
Type II: Hyperbetaand Prebetalipoproteinemia
Type III:
Broad-Beta
Disease
Type IV:
Hyperprebetalipoproteinemia
Type V: Hyperprebetalipoproteinemia and

Hyperchylomicronemia
Elevated
lipoprotein
Chylomicrons;
triglycerides
LDL; VLDL
Abnormal
chylomicron
remnant removal;
VLDL: triglyceride
>30%
VLDL
Triglycerides; apoprotein
abnormalities
Skin
Xanthomas
Xanthomas
Xanthomas
Xanthomas
Eyes
Lipemia retinalis
Lipemia retinalis
Lipemia retinalis
Corneal arcus
Early
Early
Atherosclerosis
Other findings
Hepatosplenomegaly;
pancreatitis
Inheritance
Rare, autosomal
recessive;
secondary
CHAPTER 45
TABLE 45.2. Classification of the Major Types of Hyperlipoproteinemia
Hepatosplenomegaly
Autosomal
dominant or
secondary
Abbreviations: LDL, low-density lipoprotein; VLDL, very low density lipoprotein.
Autosomal
recessive;
secondary
Autosomal
dominant;
secondary
567

(hyperbetalipoproteinemia and prebetalipoproteinemia) can
occur as an autosomal dominant disorder or it may be
secondary to hypothyroidism, dysgammaglobulinemia, and
hepatic and renal disease.50 This disorder results in elevation of
LDL levels alone or in combination with elevated VLDL levels.
Corneal arcus, xanthelasma, conjunctival xanthomas, and
coronary artery disease occur.
Type 3 hyperlipoproteinemia (familial dysbetalipoproteinemia; broad-beta disease) is inherited autosomal recessively,
but secondary cases have also been described. This disorder
results from a mutation in the apolipoprotein E gene, linked
to 19q13.2.51 Early atherosclerosis and xanthomas are the
major clinical features of this disorder. Characteristic palmar
(xanthochromia striatum palmaris) or tuboeruptive xanthomas,
typically on the elbows, may develop.52 Ocular ndings can
include early corneal arcus and lipemia retinalis.
Type 4 hyperlipoproteinemia (hyperprebetalipoproteinemia)
is characterized by an elevation of VLDL levels and may be
transmitted by autosomal dominant inheritance, although this
disorder can also be related to obesity and diabetes mellitus.53
Corneal arcus and xanthelasma are usually not prominent
clinical features of this disorder.
Type 5 hyperlipoproteinemia (hyperprebetalipoproteinemia
and hyperchylomicronemia), like type 1, results in marked elevation in triglyceride levels, but also often has other associated
apoprotein abnormalities.52 Eruptive xanthomas, lipemia
retinalis, and hepatosplenomegaly may occur. As with type 1,
corneal arcus and vascular disease are not prominent features.
The corneal involvement by types 2 and 3 hyperlipoproteinemias is usually limited to premature development of corneal arcus. Histopathologic evaluations demonstrate peripheral
lipid deposition in the corneal stroma, Bowmans layer, and
Descemets membrane. An intervening clear space between the
limbus and the arcus opacity as well as central corneal sparing
is characteristic. The arcus may rst appear in the superior
cornea, then inferiorly, and progress to become conuent.
Successful treatment and control of elevated lipoprotein levels
in patients with hyperlipoproteinemias does not reverse corneal
arcus once it has developed (see Fig. 45.7).53
HYPOLIPOPROTEINEMIAS
The hypolipoproteinemias are also disorders of lipid catabolism
but result in abnormal reductions of circulating lipoprotein
levels. This group of disorders includes BassenKornzweig disease (abetalipoproteinemia), familial hypobetalipoproteinemia,
LCAT deciency, Tangier disease, and sh eye disease.54,55
The ocular manifestations of BassenKornzweig disease and
familial hypobetalipoproteinemia are primarily retinal abnormalities and are not discussed further in this chapter, except to
mention a report of moderate diffuse opacication of the cornea
in a case of a possible familial variant of apolipoprotein A.56
Lecithin-Cholesterol Acyltransferase Deciency

The enzyme LCAT, encoded on chromosome 16q22, is a plasma
enzyme which catalyzes cholesterol ester formation from
cholesterol.57 LCAT deciency is an autosomal recessive metabolic disorder that results in the accumulation of unesteried
cholesterol within tissues, particularly in blood vessels and the
bone marrow (Table 45.3).58 Levels of HDLs as well as apolipoproteins A-I and A-II are reduced in this disorder, and levels of
VLDLs and LDLs are typically elevated. Levels of serum cholesterol and triglycerides are often above normal. Other common
systemic manifestations include anemia, renal insufciency,
and accelerated atherosclerosis.59
The cornea may develop a dense peripheral arcus and diffuse
stromal haze due to the deposition of multiple, ne, grayish
opacities (Fig. 45.8). The visual acuity is relatively unaffected.60
Anterior and posterior crocodile shagreen is present in the midcorneal periphery.61 The corneal opacities are usually noted by
the early teenage years. Heterozygous carriers may have an
increased incidence of arcus-like changes, but this association
lacks sufcient sensitivity or specicity to be of value
diagnostically. Although the composition of the opacities is not
known, pathologic evaluation has demonstrated collections of
tiny vacuoles in Bowmans layer containing many electrondense particles.62 A recent case report presented light and
transmission electron microscopic ndings of a patient with
LCAT deciency.63 All stromal layers had extracellular vacuoles
SECTION 6
TABLE 45.3. Hypoproteinemias With Significant Corneal Involvement

Characteristic
Lecithin Cholesterol
Acyltransferase Deficiency
Tangier Disease
Accumulated
metabolite
Unesterified cholesterol
Cholesterol ester
Serum cholesterol
Free cholesterol:cholesteryl ester

ratio increased
Low
High
Triglycerides
Above normal
VLDL
High
Normal
High
LDL
Normal or high
Low
Normal
Very high
HDL
Low
Very low
Very low
Apolipoproteins
A-I and A-II
Low
Low
Low
Atherosclerosis
Accelerated
Other
Cornea
568
Fish Eye Disease
Proteinuria
Peripheral neuropathy
Anemia
Orange tonsils
Arcus and diffuse stromal haze;

crocodile shagreen appears in
second decade of life
Fine diffuse clouding may be

detectable in first decade of life
but generally in fifth decade
Abbreviations: VLDL, very low density lipoprotein; LDL, low-density lipoprotein; HDL, high-density lipoprotein.
Diffuse stromal clouding; denser yellowgray peripheral opacification appears

in second decade of life; impairs vision

acuity. Confocal microscopy, reported from one affected patient,
demonstrated lipid deposits as small granular bodies, which
were fairly uniformly distributed in the stroma. This also
showed an unaltered sub-basal nerve plexus, which was
contrary to expectation.84
Neuropathy is a common feature of the disease, affecting
peripheral nerves. Corneal exposure due to seventh-nerve palsy
and lagophthalmos may be the most clinically signicant ocular
nding with secondary exposure keratopathy (Fig. 45.9). Punctal
occlusion by cautery or plugs ameliorates corneal drying.
There are no reports of coronary artery disease occurring in
homozygous patients younger than 40 years.78 It has been
suggested that other coexisting alterations such as low serum
cholesterol levels, thrombocytopenia, and decreased platelet
adhesiveness may have counterbalancing antiatherosclerotic
effects in Tangier homozygotes.79
Courtesy of Ernst J Schaefer, MD.
with acid mucopolysaccharide contents measuring up to

2.5 m. Amyloid deposits were also noted predescemetally.
Other ocular ndings can include venous dilatations, angioid
streaks, and peripapillary mottling and hemorrhage.64
Tangier Disease (Familial High-Density Lipoprotein

Deciency)
Tangier disease was initially described in 1961 as a familial HDL
deciency occurring in two residents of Tangier Island, Virginia.65
The rare disorder is transmitted by autosomal recessive
inheritance and is characterized by a deciency or complete lack
of plasma HDLs (see Table 45.3). The genetic locus is 9q31.66
Heterozygous carriers of the disease typically have reduced but
measurable levels of HDLs on electrophoretic study.67,68
HDLs in these patients, when detectable, have altered compositions such as markedly reduced levels of apolipoproteins
A-I and A-II.6971 In addition, other lipid abnormalities may
also be present, including reduced plasma cholesterol levels,
mild elevations of serum triglyceride levels, and abnormal
plasma levels of chylomicron metabolites.72
Clinical features include neuropathy, hepatosplenomegaly,
lymphadenopathy, hyperplastic yellow-orange tonsils, and the
late presence of corneal clouding. Lipid deposition, presumably
cholesterol esters, has been identied within histiocytes,
Schwanns cells, or broblasts in various tissues including
skin, nerves, cardiac valves, tonsils, spleen, liver, gastrointestinal tract mucosa, bone marrow, cornea, and conjunctiva.69,7379 Clinical manifestations of Tangier disease have
been reported in patients as young as 5 years.65
Ocular manifestations described include corneal clouding
and mottling of the retinal pigment epithelium. Orbicularis
oculi weakness, lagophthalmos, and exposure keratopathy as
well as ocular motility disturbances are secondary to the neuropathy.73,74,77,80,81 The corneal opacication appears to be caused
by lipid accumulation, particularly of esteried cholesterol and
phospholipids.82 Corneal clouding most frequently can be
detected by slit-lamp examination in affected individuals older
than 40 years, but it has also been found in childhood.69,73 The
clouding may be diffuse or localized in the stroma, typically
more central than peripheral and more posterior than
anterior.73,79,83 Small posterior stromal dot-like opacities distributed randomly or in a whorl pattern and peripheral corneal
haze along the horizontal meridian may also be seen.69,83
Neither type of corneal opacication signicantly reduces visual
Fish Eye Disease

Fish eye disease is an autosomal recessive disorder that was rst
described in 1979.85 Unlike Tangier disease and LCAT
deciency, patients with this disorder are unable to esterify
cholesterol within HDLs in spite of near-normal LCAT activity
as measured by the endogenous plasma cholesterol esterication rate.86,87 Molecular defects in the LCAT gene have been
associated with sh eye disease, explaining the almost absent
LCAT activity, when measured with exogenous HDL analogs as
substrate.88 Levels of HDL cholesterol and lipoproteins apo A-I,
A-II, and D are dramatically reduced, and LDL triglyceride
levels are strikingly high (see Table 45.3). The effect of these
changes on the incidence of atherosclerosis is not clear.
Characteristically, the cornea is diffusely cloudy with small,
dot-like gray-white-yellow opacities deposited in all layers of the
cornea except the epithelium.85 The peripheral cornea appears
more opaque than the central zone and may contain a thin,
supercial, yellow, ring-shaped opacity ~1 mm from the limbus
(Fig. 45.10). Gradual progression of the corneal changes can
cause marked visual impairment during the second decade
of life.89
LYSOSOMAL STORAGE DISEASES

Disorders of catabolic lysosomal enzymes make up a diverse
group of diseases including the mucolipidoses, mucopolysaccharidoses, galactosialidoses, sphingolipidoses, gangliosidoses,
mannosidosis, and fucosidosis. These disorders are characterized by their major storage product, although abnormal
accumulation of other metabolic compounds may occur,
e.g., complex lipids may be detected ultrastructurally in mucopolysaccharidoses.90 All are inherited as autosomal recessive
conditions with the exception of Fabrys disease and Hunters
syndrome, which are X-linked.
Although therapeutic options have previously been limited,
bone marrow transplantation has been successful in modifying
the natural history of lysosomal and peroxisomal storage
diseases.91 Cord-blood transplants of stem cells from unrelated
donors has also proved to be an effective treatment for patients
with Hurler s syndrome.92 Targeted treatments for the
lysosomal storage disorders, via enzyme replacement and/or
substrate depletion have also been successfully used for some
clinical serotypes.93
CHAPTER 45
FIGURE 45.8. Lecithin-cholesterol acyltransferase deficiency with a

dense peripheral arcus and diffuse stromal haze. Vision is relatively
unaffected.
MUCOLIPIDOSES (OLIGOSACCHARIDOSES)
The mucolipidoses are a group of lysosomal storage diseases
characterized by the accumulation of oligosaccharides. The
mucolipidoses may occur sporadically or may be transmitted by
autosomal recessive inheritance.94
569
FIGURE 45.9. Tangier disease. (a) Right eye of a 52-year old woman
shows very subtle powdery corneal stromal clouding. Vision is 20/15.
The corneal clouding is subtle in this patient and must be astutely
looked for despite the fact that she has had neuropathy for more than
20 years. (b) Slit-beam photograph shows mild fluorescein staining of
the epithelium at the junction of the mid and lower thirds of the
cornea. (c) Her major ocular problems are due to lagophthalmos from
facial nerve palsy. The irregular corneal light reflex indicates a rough
epithelial surface in the lower third of the cornea. (d) Corneal stromal
opacification is mild, generally only observed on slit-lamp
examination.
(d) Courtesy of Ernst J Schaefer, MD.
SECTION 6
570
FIGURE 45.10. Fish eye disease. The cornea shows diffuse clouding
with moderate reduction in vision due to stromal opacification. There
is denser yellow-gray peripheral opacification.
Courtesy of Harry Koster, MD, Yves Pouliquen, MD.
Mucolipidosis type 1 (dysmorphic sialidosis, Spranger s

syndrome) is caused by a mutation in the gene encoding
neuraminidase, located on 6p21.3.95 There are two subtypes.
Patients with sialidosis type 1 have decreased visual acuity with
an associated macular cherry-red spot, myoclonus, and gait
abnormalities.9698 Sialidosis type 2, with infantile onset, is
more severe and results in dysmorphic Hurler-like facies
(prominent brow, hypertrichosis, frontal bossing, and saddle
nose), organomegaly, mental retardation, dysostosis multiplex,
sensorineural hearing loss, and progressive neurologic
decline.99101 Most affected patients with either subtype do
not survive past adolescence or early adulthood.
Ocular manifestations of sialidosis type 2 include spoke-like
lenticular opacities, a macular cherry-red spot, and tortuous
retinal and conjunctival vessels.57 Fine corneal epithelial

and stromal opacities can occur but do not typically produce
signicant corneal clouding.102 Histopathologic study has
revealed single membrane-bound inclusions similar to those
of mucopolysaccharidoses in corneal epithelium and
keratocytes.57 As in sphingolipidoses, rare intracellular membranous lamellar inclusions may also be present.
Mucolipidosis type 2 (I-cell or inclusion cell disease) is caused
by an abnormality of N-acetylglucosamine phosphotransferase,
the gene for which is located on chromosome 12.103 Clinical
features include dysmorphic facies with gingival hyperplasia,
skeletal deformities, organomegaly, short stature, and mental
retardation. Orbital changes include hypoplastic orbits with
hypoplasia of the supraorbital ridges and prominence of the
globes. Other ocular changes can include glaucoma,
megalocornea, or corneal clouding. The cornea usually remains
clear in early life, but ~40% of patients later develop abnormal
stromal granularity and mild opacity.104 Keratocytes and
broblasts of the corneal stroma and conjunctiva have
membrane-bound vacuoles with ne brillogranular and
irregular membranous lamellar inclusions. These cells are thus
referred to as inclusion cells or I cells. Such I cells have been
found in mucolipidoses, types 2 and 3.
Mucolipidosis type 3 (pseudo-Hurlers polydystrophy) is
also caused by a deciency of N-acetylglucosamine phosphotransferase, but its features, although similar to those of
type 2, are less severe.102 Fine opacities of the corneal stroma
may also be present, but these do not signicantly affect visual
acuity (Fig. 45.11).102 Disk edema, surface wrinkling maculopathy, and hyperopic astigmatism are sometimes seen.105
Mucolipidosis type 4 (Bermans syndrome) is an autosomal
recessive neurodegenerative disorder, characterized by psychomotor retardation and ophthalmic abnormalities due to a
deciency in the gene encoding mucolipin-1, which is linked to
chromosome 19p13.3p13.2.106 Patients have prominent
diffuse corneal clouding that is present at birth or appears in
early infancy.107 A mild variant can also present later in child-

various tissues throughout the body. All the sphingolipidoses,
with the exception of Fabrys disease, are transmitted by
autosomal recessive inheritance. Those disorders with corneal
involvement are included in this chapter.
Fabrys Disease
Key Features
FIGURE 45.11. Mucolipidoses type 3, pseudo-Hurlers polydystrophy.

There is diffuse corneal stromal clouding. The epithelium is regular.
Courtesy of Trexler M Topping, MD.
hood with corneal clouding having the appearance of cornea

verticillata and retinal dystrophy.108 The vacuoles within the
corneal epithelium are either single membrane-bound vesicles
containing brillogranular material suggestive of mucopolysaccharides or membranous lamellar bodies consistent with
phospholipids.107,109 Conjunctival cytologic studies, which
reveal characteristic lysosomal inclusions on light and electron
microscopy, may help conrm the disorder.110
Mechanical epithelial dbridement or penetrating keratoplasty does not prevent recurrent corneal opacication after
reepithelialization or replacement of the graft epithelium by
cells of host origin.111 Transplantation of epithelium through
conjunctival allografts, particularly limbal grafts, which
theoretically may allow normal stem cell repopulation, may
offer more prolonged maintenance of corneal clarity.112 Other
features include cataract, optic nerve atrophy, attenuated retinal
vasculature, and electroretinographic abnormalities suggestive
of a retinal degeneration.110,112,113
GALACTOSIALIDOSIS
Galactosialidosis (neuraminidase deciency with B-galactosidase
deciency, Goldbergs syndrome, GoldbergCotlier syndrome,
sialidosis type 2 juvenile onset) is an autosomal recessive
disorder that is caused by combined deciency of the lysosomal
enzymes beta-galactosidase and alpha-neuraminidase and
linked to chromosome 20.114,115 Systemic features include facial
dysmorphism, mental retardation, seizures, skeletal deformities, ataxia, hearing loss, and myoclonus. The characteristic
ocular nding is reduced visual acuity associated with a macular
cherry-red spot, but mild degrees of corneal clouding may
occur.116 An adult form of galactosialidosis, characterized by
ne corneal opacities, a cherry red spot and optic atrophy, has
also been described and may be diagnosed by a conjunctival
biopsy.117
SPHINGOLIPIDOSES
The sphingolipidoses are a group of lipid storage diseases
presumably caused by deciencies of specic hydrolytic
enzymes that result in the accumulation of lipids within
X-linked condition
Cornea verticillata, fine powdery epithelial deposits in a spokelike pattern in the inferior cornea, develop in both the affected
males and carrier females
Cornea verticillata rarely affects vision
Cornea verticillata can also develop secondary to medication
use, such as amiodarone, indomethacin, chloroquine and
phenothiazines
Fabrys disease, an X-linked disorder, is caused by a deciency

of a-galactosidase A. This results in the accumulation of
ceramide trihexoside in all areas of the body, but predominantly
within lysosomes of vascular endothelial and smooth muscle
cells. Renal failure and cardiovascular complications are
common in adult life.
Recurrent episodes of pain in the peripheral extremities,
associated with fever and sedimentation rate elevation, are
typical in affected males beginning in childhood.118,119
Sphingolipid deposition within vascular endothelium results in
the characteristic small dot-like skin lesions, referred to as
angiokeratoma corporis diffusum, that become manifest in the
bathing trunk area around the time of puberty and become
more elevated and hyperkeratotic with time.
Corneal involvement is typical in Fabrys disease in both the
affected hemizygous males and the female carriers. The ne
powdery opacities of the corneal epithelium or subepithelium
usually develop in early infancy and are best seen by retroillumination during slit-lamp examination.120 They can be
observed by the age of 4 years in hemizygotes and by the age of
10 years in heterozygotes, in whom the corneal involvement is
often more pronounced. They occur in a whorl or vortex
distribution (cornea verticillata) like force lines in a magnetic
eld (Fig. 45.12). They are usually inferior and typically cream
colored but range from white to golden-brown or appear very
faint.121 They do not affect vision, although vision can become
seriously affected by vascular accidents in both the retina and
the central nervous system. Confocal microscopy demonstrates
the accumulated glycosphingolipids as intracellular inclusions
in the corneal and conjunctival epithelial cells and epithelial
basement membrane.122 Striate melanokeratosis and various
medications, including indomethacin, chloroquine, amiodarone
(Fig. 45.13), and phenothiazines can cause corneal epithelial
changes that may mimic the cornea verticillata of Fabrys
disease.
Other ocular manifestations include telangiectasia and tortuosity of conjunctival and retinal vasculature, seen in 70%
of affected males as opposed to 25% of carrier females
(Fig. 45.14).123 These changes may precede the corneal manifestations. Lenticular changes can include a characteristic
granular anterior subcapsular wedge-shaped or propeller-shaped
lens opacity as well as a posterior linear whitish deposit of
granular material at or near the posterior capsule, which
sometimes resembles a herpetic dendrite.
Characteristic Maltese cross-pattern birefringent intracellular
inclusions can be histopathologically identied after biopsy of
the conjunctiva, skin, kidney, or other blood vessel-containing
tissue.124 Electron microscopy can conrm the typical lamellar
inclusion bodies within these cells.
CHAPTER 45
571
FIGURE 45.12. (a) Subepithelial whorl pattern of corneal opacities in a man with Fabrys disease. (b) Corneal involvement in his sister. (c) Fine
whorls may give the appearance of force lines in a magnetic field.
SECTION 6
(a and b) Reprinted from Miller CA, Krachmer JH: Corneal diseases. In: Renie WA, ed. Goldbergs genetic and metabolic eye disease. 2nd edn. Boston: Little, Brown;
1986:350.
FIGURE 45.13. Amiodarone, a cardiac antiarrhythmic, causes a

vortexshaped keratopathy that is reversible with cessation of the drug
therapy.
572
FIGURE 45.14. Conjunctival telangiectasia with microaneurysmal

dilatation of the vessels. Similar changes can be seen in the mouths
of patients with Fabrys disease. Similar conjunctival vascular changes
have been reported in patients with fucosidosis.184,185
Reprinted with permission from Miller CA, Krachmer JH: Corneal diseases. In:
Renie WA, ed. Goldbergs genetic and metabolic eye disease. 2nd edn. Boston:
Little, Brown; 1986.
Multiple Sulfatase Deciency

Multiple sulfatase deciency (MSD) (metachromatic leukodystrophyAustins juvenile form) is a disorder that combines
features of metachromatic leukodystrophy and mucopolysaccharidoses. This autosomal recessive disorder results from
a deciency of arylsulfatases A, B, and C.126 Consequently,
excessive amounts of sulfatide accumulate within tissues.
The genetic locus is 3p26.127 In the classic MSD presentation,
this disorder is characterized by the development of facial
dysmorphism, skeletal abnormalities, ichthyosis, and early
psychomotor retardation. Ocular ndings are similar to those
of mucopolysaccharidoses, including rare corneal clouding,
grayish cherry-red spot, optic atrophy, and pigmentary retinal
degeneration.128,129
GANGLIOSIDOSES
Gangliosides are glycosphingolipids that contain sialic acid in
their oligosaccharide chain. They occur in high concentration
in the brain in nerve ending membranes and synaptic
membranes.
GM1 Gangliosidosis Type 1 (Generalized

Gangliosidosis)
GM1 gangliosidosis type 1 is an autosomal recessive disorder
caused by deciency of the enzyme b-galactosidase, which is
encoded on chromosome 3.130 In GM1 gangliosidosis type 1,
there is an accumulation of ganglioside in the central nervous
system as well as accumulation of the glycosaminoglycan
keratan sulfate in the liver and spleen. The clinical picture is
one of initially normal development, followed by severe
neurologic decline. There are at least ve subtypes of GM1
gangliosidosis caused by variable residual activity of the mutant
enzyme against different substrates. Thus some patients have
severe neurologic decit and early death, whereas others have
severe bony abnormalities, facial dysmorphism, and normal
intelligence.
Ocular ndings include macular cherry-red spots, nystagmus, strabismus, retinal hemorrhages, and optic atrophy. Some
patients also have mild, diffuse corneal clouding and numerous
histopathologically detectable intracytoplasmic vacuoles within
all layers of the cornea.131
GM2 Gangliosidosis Type 2 (Sandhoffs

Disease)
The GM2 gangliosidoses comprise three distinct genetic
disorders, TaySachs disease, Sandhoff s disease, and the GM2activated protein deciency. In contrast to TaySachs disease,
which is the most common ganglioside storage disease and
occurs from a deciency of hexosaminidase A, Sandhoff s
disease results from a deciency of both hexosaminidase A and
B. Severe central nervous system dysfunction occurs owing to
the accumulation of GM2 gangliosides in neurons. Membranebound vesicles have been detected within keratocytes by
histopathologic and ultrastructural evaluation, although the
cornea may appear clear clinically.132 Other ocular features may
also include tapetoretinal degeneration, optic atrophy, and the
presence of a macular cherry-red spot. Involvement of white
matter of the optic radiations has been shown by magnetic
resonance imaging.133
MUCOPOLYSACCHARIDOSES
Key Features
All mucosaccharidoses are autosomal recessive except

Hunters syndrome, which is X-linked
Corneal clouding is a significant feature of all
mucosaccharidoses except Hunters and San Filippos
syndromes
The mucopolysaccharidoses (MPSs) result from lysosomal

enzyme deciencies affecting the degradation of glycosaminoglycans (mucopolysaccharides). All these disorders,
except for X-linked Hunters syndrome, are transmitted by
autosomal recessive inheritance. The clinical manifestations
vary for each MPS type because of differences in the specic
enzyme defect or in the tissue localization of the involved
enzyme. Ten enzyme deciencies have been identied that give
rise to different syndromes.134 There are geographical variations
in incidence of the different MPS types.135 The cornea is
frequently affected by abnormal glycosaminoglycan metabolism
because glycosaminoglycans make up the corneal stromal
ground substance and constitute 44.5% of the dry weight of
the cornea. Corneal keratan sulfate makes up 50% of the
ground substance; it differs from keratan sulfate in cartilage.
Chondroitin and chondroitin sulfate A each constitute ~25% of
the ground substance. Chondroitin is found only in the
cornea.136 The diagnosis of a disorder of mucopolysaccharide
catabolism is made on the basis of characteristic clinical
ndings in association with demonstration of the enzymatic
deciency or detection of elevated levels of urinary
glycosaminoglycans (Table 45.4).
HURLERS SYNDROME (MPS I-H)

Hurlers syndrome, results from deciency of the enzyme
a-L-iduronidase, which is required for the breakdown of heparan
sulfate and dermatan sulfate. In its severe form as Hurlers
syndrome, clinical manifestations are usually evident by the
rst year of life. Characteristic features include short stature,
dysostosis multiplex, facial dysmorphism (coarse facies, prominent forehead, hypertelorism, anteverted nostrils, hypertrichosis, synophrys, and depressed nasal bridge), and progressive
psychomotor retardation (Fig. 45.15). Other ndings can
include hepatosplenomegaly, neurosensory hearing loss, joint
stiffness, umbilical hernia, and cardiac defects. Death often
occurs in the second decade of life due to recurrent pneumonia
or heart failure. The diagnosis can be conrmed by measurement of a-L-iduronidase activity in isolated peripheral leukocytes or cultured dermal broblasts or amniotic cells.
Diffuse corneal clouding usually becomes apparent by the age
of 3 years and may present with photophobia. The ne, punctate corneal opacities are usually distributed throughout the
stroma (Fig. 45.16a), although they are most pronounced
centrally and are best visualized by slit-lamp examination
(Fig. 45.16b).137 Although penetrating keratoplasty can restore
corneal clarity in severe cases of corneal clouding, visual acuity
is often limited in these patients because of associated optic
nerve or retinal disease. Glaucoma may also be present and may
be difcult to diagnose and monitor because of corneal opacication and thickening.135 Progressive retinopathy with vascular
narrowing, hyperpigmentation of the fundus, and later bone
spicule formation occur. Papilledema and optic atrophy are
common.138 Bone marrow transplantation can improve some of
the ocular manifestations such as corneal clouding, optic nerve
edema, and retinopathy.139 Cord-blood transplants from
unrelated donors have been demonstrated to be effective in the
CHAPTER 45
The DNA sequence for human a-galactosidase has been

isolated and enzyme replacement therapy, utilizing recombinant technology, has been demonstrated to be a safe and
effective treatment.125
573
TABLE 45.4. Major Clinical Findings in the Mucopolysaccharidoses

Mucopolysaccharidosis
Clinical
Corneal
Clouding
Glaucoma
Retinal
Bone
Spicules*
Optic
Atrophy
Skeletal
Involvement
Retardation
Major
Compound
Stored*
Age at
Onset
Death
MPS I-H
Hurlers
Diffuse,
progressive;
onset age
1 yr
Late
++ Gargoyle
Severe
HS, DS
612 mo
By teens
MPS I-S
Scheies
Diffuse; onset
after 4 yr
++
Late
HS, DS
57 yr
Normal
MPS I-HS
Hurler-Scheie
Diffuse; onset
2 yr
Late
HS, DS
24 yr
MPS II-A
Hunters A
Often
HS, DS
Under
1 yr
Before
age
15 yr
MPS II-B
Hunters B
Late
Often
HS, DS
4 yr
3060 yr
MPS III-A,
-B, -C, -D
Sanfilippos
Rare
HS
26 yr
2030 yr
MPS IV Morquio
Diffuse; after
age 10 yr
++
KS, CS
12 yr
Varies
MPS VI
MaroteauxLamy
Diffuse;
punctate
Rare
Dwarf +
DS
23 yr
MPS VII Slys
HS, DS
SECTION 6
Modified from Lang GE, Maumenee IH: Retinal dystrophies associated with storage diseases. In Newsome DA (ed): Retinal Dystrophies and Degenerations. Philadelphia,
Lippincott-Raven, 1988, pp 320321.
Abbreviations: HS, heparan sulfate; DS, dermatan sulfate; KS, keratan sulfate; CS, chondroitin sulfate; MPS, mucopolysaccharidosis.
*Night blindness is associated with retinal bone spicules.
Retinal involvement occurs when HS is stored.
574
FIGURE 45.15.
Hurlers syndrome,
mucopolysaccharidosis (MPS) I-H.
Short stature and
facial dysmorphism
are characteristic of
patients with MPS I-H.
Facial features are
coarse, the nares are
anteverted, and the
brows are heavy and
close, with wideset
eyes. The abdomen is
protuberant with an
umbilical hernia.
Courtesy of Trexler M
Topping, MD.
treatment of these patients, as it favorably alters the natural

history of Hurlers syndrome.92
Patients with Hurlers syndrome are at increased risk for
cardiovascular collapse or laryngospasm during general
anesthesia. Difcult or failed intubations are common in
children with mucopolysaccharidoses.140 Pharyngeal secretions
may be excessive and may be managed by large doses of

atropine in the preinduction period and avoidance of narcotics
in the postoperative period.141
SCHEIES SYNDROME (MPS I-S)

Scheies syndrome, previously referred to as MPS V, also results
from a deciency of the enzyme a-L-iduronidase, but it is the
least severe form of MPS. Clinical manifestations include
coarse facies, clawlike hand deformities, carpal tunnel
syndrome, hernias, neurosensory hearing loss, joint stiffness,
and cardiac abnormalities. In contrast to Hurlers syndrome,
mental retardation, dwarsm, and early death are usually not
features.
Ocular abnormalities include corneal clouding (Fig. 45.17),
optic nerve head swelling or late optic atrophy, and pigmentary
retinal degeneration.135 The corneal clouding is usually progressive and diffusely involves the stroma, particularly in the
corneal periphery and posterior stromal regions. Ultrastructural
analysis of corneas from patients with MPS I-H and MPS I-S
has found a greater range and size of collagen bril diameter, the
presence of brous long-spacing collagen, vacuolated stromal
cells, and disrupting sulfated glycosaminoglycan deposits compared with normal corneal stroma.142,143 Glaucoma also occurs
more frequently than in Hurlers syndrome. Success of penetrating keratoplasty for corneal opacication may be limited by
coexisting optic nerve and retinal disease.
HURLERSCHEIE SYNDROME (MPS I-HS)

The activity of the enzyme a-L-iduronidase and systemic
manifestations in patients with HurlerScheie syndrome are
FIGURE 45.16. Hurlers syndrome, Corneal clouding increases over time with punctate opacification of the stroma (a), which can be best seen
by retroillumination at the slit lamp (b).
FIGURE 45.17. Scheies syndrome, MPS I-S. Corneal clouding is fine,

diffuse, and slightly more prominent in the peripheral stroma.
polysaccharides have been shown histologically in corneas

that appeared clear clinically.136
HUNTERS SYNDROME
Hunters syndrome is the only MPS that is transmitted by
X-linked recessive inheritance. It is caused by a deciency of the
enzyme iduronate-2-sulfatase and results in the accumulation
of heparan sulfate and dermatan sulfate within tissues.
Several allelic variants have been identied, with varying
levels of severity. In the more severe form, Hurlers-like features
may occur, including deafness, coarse facies, short stature,
mental retardation, hepatosplenomegaly, cardiac disease, and
death within the second decade of life. Hirsutism and smooth,
pinpoint dermal elevations are also frequent.
Ocular ndings in either type often include papilledema
(Fig. 45.18), optic atrophy, and pigmentary retinal degeneration. Corneal clouding may be detectable by slit-lamp
examination, but it is not clinically signicant. However, muco-
SANFILIPPOS SYNDROME (MPS III TYPES A

THROUGH D)
Sanlippos syndrome is caused by one of four different
enzymatic defects of heparan sulfate catabolism, all of which
have different loci. Type A is caused by a deciency of the
enzyme heparan sulfate N-sulfatase (heparan-S-sulfaminidase)
and is the most severe form. Type B is due to deciency of
N-acetylglucosaminidase. Type C results from deciency of
N-acetyltransferase (acetyl-CoA-a-glucosamide-N-N-acetyltransferase). Type D occurs due to a deciency of the enzyme Nacetylglucosamine-6-sulfate sulfatase.
Clinical manifestations of Sanlippos syndrome include
deafness, coarse facies, short stature, joint stiffness, mild
hepatosplenomegaly, and severe mental retardation.
CHAPTER 45
intermediate between those occurring in Hurlers syndrome

and those in Scheies syndrome.135 Corneal clouding is usually
progressive and typically requires penetrating keratoplasty for
visual rehabilitation within the rst decade of life. Other ocular
abnormalities include glaucoma, optic atrophy, and retinal
pigmentary degeneration.
FIGURE 45.18. Hunters syndrome, MPS II. A clinically clear cornea

allows excellent visualization of papilledema in a 13-year old boy with
MPS II.
575

Ocular ndings are usually limited to pigmentary retinal
degeneration and corresponding electroretinogram changes.144
Optic atrophy rarely occurs. Corneal opacication has been
reported, but is not a prominent feature.
MORQUIOS SYNDROME (MPS IV)

Morquios syndrome results from the accumulation of keratan
sulfate secondary to a deciency of the enzyme
N-acetylgalactosamine-6-sulfate sulfatase. The gene for Morquio
MPS IV-A maps to 16q24.3.145 Historically, MPS IV was divided
into IVA and IVB; however, MPS IVB is now considered a
variant of GM1-gangliosidosis.135
Skeletal deformities are the most prominent clinical features;
these include dysostosis multiplex, dwarsm, pectus
carinatum, kyphoscoliosis, short rst metacarpal bones, genu
valgum, and other deformities (Fig. 45.19). Spinal cord compression may develop owing to vertebral abnormalities. Aortic
valvular disease and recurrent pneumonia are common.
Intelligence is normal.
Ocular ndings are limited to papilledema and diffuse,
stromal corneal clouding (Fig. 45.20). It appears as a myriad
of white minute dots in the stroma.146 The corneal opacity,
however, usually does not necessitate penetrating keratoplasty.
Associated lenticular opacities have been described.147
Ocular manifestations include progressive corneal clouding

with increased corneal thickness (Fig. 45.22), papilledema, and
optic atrophy. Retinal involvement has not been reported.
Diffuse corneal opacities can necessitate penetrating keratoplasty. A case of repeat opacication of the corneal graft has
been reported.148 However, clear grafts have been reported in a
patient after bone marrow transplantation 13 years
postkeratoplasty.149
SLYS SYNDROME (MPS VII)

Slys syndrome results from deciency of the enzyme
b-glucuronidase, which is normally encoded for by a gene on
chromosome 7.137 The lack of b-glucuronidase can be documented in cultured broblasts. Metachromatic Alder granules
have been identied in leukocytes of affected patients.
Clinical manifestations include mental retardation,
dysostosis multiplex, hepatosplenomegaly, frequent respiratory
infections, and umbilical hernias. Ocular ndings can include
mild corneal opacities, papilledema, and retinal pigmentary
MAROTEAUXLAMY SYNDROME (MPS VI)
SECTION 6
MaroteauxLamy syndrome is caused by deciency of the

enzyme arylsulfatase B (N-acetylgalactosamine-4-sulfate
sulfatase). Elevated levels of dermatan sulfate and heparan sulfate
are excreted in the urine. Prominent intracellular inclusions are
often present in circulating leukocytes (Fig. 45.21).
Systemic changes can resemble those of Hurlers syndrome,
but intellectual function is usually preserved. Clinical features
include cardiac anomalies, dwarsm, and other skeletal
deformities. Spinal cord compression may occur owing to
vertebral anomalies. Meningeal involvement can cause
hydrocephalus.
576
FIGURE 45.19.
Morquios syndrome,
MPS IV. Dwarfism,
short neck, pectus
carinatum,
kyphoscoliosis, and
other skeletal
abnormalities are
characteristic of
MPS IV.
FIGURE 45.20. Morquios syndrome, MPS IV. Patchy dotlike and

diffuse corneal clouding gives the cornea a ground-glass appearance.
Courtesy of Trexler M
Topping, MD.
FIGURE 45.21. MaroteauxLamy syndrome, MPS VI. Circulating

leukocytes often have prominent inclusions.

FIGURE 45.22. MaroteauxLamy syndrome,
MPS VI. Corneal clouding varies from mild (a)
to moderate (b) in type A disease. (c and d) The
peripheral corneal clouding is much more dense
and visible to the unaided eye in type B
disease.
degeneration, as well as late optic atrophy.100 Corneal clouding

may be severe enough to require transplantation.150
of the corneal epithelium removes the corneal haze. The

stroma and endothelial layers of the cornea do not appear to
be involved.
MISCELLANEOUS STORAGE DISEASES

Mannosidosis and fucosidosis are caused by deciencies of
a-mannosidase and a-fucosidase, respectively. In patients with
fucosidosis, corneal opacities in a verticillatapattern and conjunctival and retinal vessel tortuosity (Figs 45.15 and 45.18),
have been noted.151 Cataracts, specically spoke-like posterior
cortical opacication, have been observed in mannosidosis.152
XERODERMA PIGMENTOSUM
Xeroderma pigmentosum is an autosomal recessive disorder
caused by a deciency of DNA repair mechanisms that predisposes affected individuals to radiation-induced damage.
DISORDERS OF NUCLEIC ACID

METABOLISM
Gout is a group of diseases of humans that result in a variable
combination of clinical ndings including increased concentrations of serum urate, deposits of monosodium urate
monohydrate in and around the joints of the extremities
(tophi), renal disease, and uric acid urolithiasis. Hyperuricemia
may result from an increased rate of uric acid production and by
diminished clearance by the kidneys.
Ocular ndings in gout are rare, but histological or ultrastructural features of tophi have been demonstrated in the
conjunctiva, cornea, lateral canthus, brow and orbit.153,154
Recently, urate crystals were reported in the iris and anterior
chamber.154 Ferry et al reported on the ocular abnormalities
in 69 patients with severe gout.155 The most common nding
was red eyes (62%), pingueculae (25%), elevated IOP (14%),
asteroid hyalosis (4%), and corneal crystals (one patient). Band
keratopathy (Fig. 45.23) consisting of urate crystals has been
reported.153 By slit-lamp examination, ne golden-yellow
scintillating crystals were found to be present in the epithelial
and subepithelial regions extending to the limbus; these were
more numerous in the interpalpebral space. On morphologic
examination, hexagonal, octagonal, or cylindrical crystals were
found within the nuclei of epithelial cells. In cross-section, the
crystals demonstrated a regular lattice structure. The cytoplasmic structure of the cells was normal.156 Simple scraping
CHAPTER 45
GOUT
FIGURE 45.23. Urate band keratopathy in a 68-year-old man with

gout. The patient had dry eyes with a Schirmer test with anesthetic of
2-mm wetting. Vision was 20/50 and returned to 20/20 after the
epithelium was scraped.
577

Various dermatologic manifestations of sun-exposed areas
include hypopigmentation, hyperpigmentation, hyperkeratosis,
and neoplasia (Fig. 45.24).157 Excision of multiple basal or
squamous cell carcinomas of the eyelids often produces
secondary eyelid deformities. Recurrent corneal ulcerations may
develop.158 Fibrovascular pannus of the cornea may develop as
well as squamous cell carcinoma (Fig. 45.25). Avoidance of sun
exposure and the generous use of sunblock are essential to
reduce cumulative actinic damage and subsequent neoplasia.
DISORDERS OF MINERAL METABOLISM

WILSONS DISEASE
Key Features
SECTION 6
Autosomal recessive.
The KayserFleischer ring, copper deposition in the deep,
peripheral cornea, can be seen early in the disease process.
As the disease progresses, it becomes more prominent. It can
change in color from a yellow-green and gold hue to deep
brown. It may initially only be seen using gonioscopy, while
later it is obvious with the naked eye.
Patients typically present with neurological symptoms. The
ophthalmologist may be consulted to evaluate the patient for a
KayserFleischer ring.
If treatment is started prior to significant liver and neurological
damage, patients often do very well. The KayserFleischer ring
may regress or disappear with treatment.
FIGURE 45.24. Xeroderma pigmentosum on sun-exposed areas of

skin showed hypopigmentation and hyperpigmentation with varied
intensities of pigmentation. The deformity of the left part of the
patients nose is due to surgical excision of a tumor.
578
From Calonge M, Foster CS, Rice BA, et al: Management of corneal

complications in xeroderma pigmentosum. Cornea 11:175, 1992.
FIGURE 45.25. Right eye of the patient shown in Figure 45.29 shows
corneal stromal scarring with peripheral pannus and lipid deposition.
The bulbar and tarsal conjunctiva are injected with notable
telangiectasia of vessels.
From Calonge M, Foster CS, Rice BA, et al: Management of corneal
complications in xeroderma pigmentosum. Cornea 1992; 11:175.
Wilsons disease, an autosomal recessive disorder, is the most

common genetic disorder of copper metabolism. The gene for
Wilsons is closely linked to the esterase D locus near 13q14.159
Both the biliary excretion of copper and its incorporation into
ceruloplasmin, the copper transport enzyme, are severely
impaired in Wilsons disease, leading to progressive accumulation of copper in the liver. Accumulation of copper also
occurs in the brain, especially in the basal ganglia. Neurologic
symptoms of tremor, dysarthria, or choreoathetosis may be
presenting signs, usually after puberty, but these can develop as
late as 60 years of age.160
KayserFleischer rings are copper deposits in Descemets
membrane of the cornea (Fig. 45.26). They are rst seen by
gonioscopy at the upper and lower limbal edges of Descemets
membrane. While they do not affect vision, they are an important diagnostic sign and management indicator. With time they
extend to the full corneal circumference and change from a
lighter yellow-green-gold color to deep brown, visible to the
unaided eye. They are found in most patients with neurologic
manifestations of Wilsons disease and in ~95% of all Wilsons
patients. The absence of a KayserFleischer ring does not
exclude the diagnosis.161 It may be absent in up to 30% of young
patients presenting with acute liver disease and up to 60% of
patients in the presymptomatic stage.162 The rings may fade or
disappear after treatment.162163 A KayserFleischer ring may
also be seen in primary biliary cirrhosis, familial cholestatic
cirrhosis, neonatal liver disease, and multiple myeloma.164167
Sunower cataracts are disk-shaped axial opacities with
spoke-like deposits radiating peripherally. The deposits are
brilliantly colored immediately below the anterior and posterior
lens capsules. They occur in a minority of patients and do not
impair vision. They may disappear within a few years of starting
D-penicillamine therapy.168
The diagnosis of Wilsons disease is most reliably made by
liver biopsy demonstrating greatly increased copper levels.
Administration of D-penicillamine is a standard treatment for
Wilsons disease. Other therapies include trientine, zinc, and
tetrathiomolybdate. If therapy is instituted before severe hepatic
and neurologic damage have occurred, patients can enjoy a
normal life span and good health.

FIGURE 45.26. KayserFleischer ring in
Wilsons disease. (a and b) These patients have
full 360-degree corneal involvement as is
classically described. (c) This patient has
involvement of only the upper cornea. (d) The
slit-lamp view of the patient in (c). The arrow
points to the KayserFleischer ring in
Descemets membrane.
(ad) From Wiebers DO, Hollenhorst RW, Goldstein

NP: The ophthalmologic manifestations of Wilsons
disease. Mayo Clin Proc 1977; 52:414.
HEMOCHROMATOSIS
Hemochromatosis, a condition resulting from the excessive
accumulation of iron in various organs, manifests as cirrhosis,
diabetes mellitus, cardiomyopathy, hyperpigmentation,
arthritis, and hypogonadism. Iron overload can occur as a
consequence of excessive absorption of iron or after repeated
transfusions. Idiopathic hemochromatosis is transmitted by
autosomal recessive inheritance.
With excessive iron deposition, brown pigmentation may
appear at the eyelid margin and in the perilimbal conjunctiva
encroaching on the peripheral corneal limbus.169 The inferior
cornea is usually affected more than the upper. Histopathologic
examination has conrmed the presence of iron in the corneal
epithelium in affected patients.
MISCELLANEOUS DISORDERS
SCHNYDERS CRYSTALLINE DYSTROPHY
Autosomal dominant
Deposition of cholesteral and lipid in the cornea
During young adulthood, the main feature is a prominent
corneal arcus
During mid adulthood, the entire corneal stroma becomes
diffusely cloudy, greater centrally than in the mid-periphery
The classic central superficial corneal crystals may only be
noted in half of patients with Schnyders. They can be seen as
early as the first decade
Schnyders crystalline dystrophy is transmitted by autosomal

dominant inheritance and characterized by bilateral deposition
of cholesterol and lipid in the cornea. The gene has been
mapped to chromosome 1p36.2-36.3. The words largest
pedigree of patients has a Swede-Finn heritage.
The dystrophy progresses with age. In the third decade of life,
a peripheral corneal arcus may be prominent. By the fth decade
the stroma becomes diffusely cloudy (Fig. 45.27). As the patients
age, corneal sensation decreases and visual acuity worsens as
the central cornea becomes more hazy from cholesterol depo-
FIGURE 45.27. Cornea of a 78-year old woman with dense arcus and
diffuse stromal lipid deposition.
From Weiss JS: Schnyders dystrophy of the cornea. A SwedeFinn connection.
Cornea 1992; 11:98.
sition. The opacication can involve the central cornea in a

disc-shaped pattern or the paracentral area in a ring distribution
(Fig. 45.28). The corneal opacities, even if prominent on
slit-lamp examination, do not greatly reduce visual acuity in
most patients. In a study of 33 patients with Schnyders
crystalline dystrophy, only 51% actually had clinical evidence of
corneal crystalline deposits.170 If cholesterol crystals are absent,
the disease may be very difcult to diagnose. A subtle corneal
haze best seen by retro-illumination may be the only sign
(Fig. 45.29).171 Crystals in the subepithelial or Bowmans layer
may lead to epithelial destabilization and corneal epithelial
erosion. The crystalline opacities may disappear but recur over
a period of years.172
Various systemic lipid abnormalities occur in association
with Schnyders crystalline dystrophy, the most frequent of
which is familial hypercholesterolemia.173 The differential
CHAPTER 45
Key Features
579
FAMILIAL DYSAUTONOMIA (RILEYDAY

SYNDROME)
Key Features
FIGURE 45.28. Schnyders crystalline dystrophy. Patient with ring of

fine crystalline deposits.
Autosomal recessive
Decreased corneal sensation and severe dry eyes can lead to
severe ocular surface disease and even corneal erosion,
ulceration and perforation
Aggressive treatment with ocular lubricants, punctal occlusion
and possibly autologous serum eye drops may be necessary
RileyDay syndrome is a rare autosomal recessive disorder

that is characterized by sensory and autonomic nervous system
dysfunction.177,178 Approximately 1 in 3700 patients of
Ashkenazic Jewish or Eastern European ancestry is affected.179
The familial autonomia gene has been mapped to polymorphic
markers in the q31 to q33 region of chromosome 9. A deciency
of the enzyme dopamine b-hydroxylase results in elevation of
levels of homovanillic acid.180
Clinical manifestations include paroxysmal hypertension,
emotional lability, increased sweating, an absence of fungiform
papillae on the tongue, and coldness of the distal
extremities.181182 Excessive drooling is common due to salivary
gland hypersecretion. This may be attributable to salivary gland
denervation supersensitivity, a mechanism present in the
cardiovascular system and pupil in familial dysautonomia.183
The characteristic ophthalmic signs are reduced corneal
sensation and a lack of tearing, which may lead to exposure
keratopathy, corneal erosion, and eventual ulceration
(Fig. 45.30) and perforation.184 Early intervention with punctal
occlusion using silicone plugs or cautery can prevent signicant corneal morbidity and preserve vision. Autologous serum
eye drops may also be helpful. Other features including
blepharoptosis, anisocoria, tortuosity of retinal vasculature,
myopia, and anisometropia may also be present.184,185 Prenatal
diagnosis of familial dysautonomia in families with a previously
affected child can be performed using linkage analysis.
SECTION 6
FIGURE 45.29. Schnyders crystalline dystrophy with a circular

pericentral non crystalline opacity
diagnosis includes systemic disorders affecting lipid metabolism,

such as LCAT deciency, sh eye disease, Tangier disease, as
well as those disorders with corneal crystals, such as cystinosis,
multiple myeloma and gout.
There are no local or systemic treatments that halt the
progression. Phototherapeutic keratectomy can be used to treat
subepithelial crystals if they are affecting vision.174 Penetrating
keratoplasty can be performed successfully in advanced cases.
Histochemical evaluations of excised penetrating keratoplasty
buttons have demonstrated crystalline deposits of cholesterol
or cholesterol esters within Bowmans layer, supercial corneal
stroma, and anterior sclera.175 Recurrent deposition of
crystalline lipids may eventually occur in corneal allografts after
penetrating keratoplasty.176
AMYLOIDOSIS
Familial systemic amyloidosis may occur in association with
lattice corneal dystrophy, cranial nerve palsies, peripheral neuropathy, and skin changes. Meretojas syndrome, an autosomal
dominant disorder, is further described in Chapter 241
(Vascular Lesions of the Orbit).
580
FIGURE 45.30. Familial dysautonomia, RileyDay syndrome. The

patient is a young girl with a neurotrophic corneal epithelial defect that
resulted in sterile ulceration and stromal scarring.
From Mandel ER, Wagoner MD: Atlas of corneal disease. Philadelphia: WB
Saunders; 1989:48.
1. Burns RP: Tyrosinemia. In: Gold DH,

Weingeist TA, eds. The eye in systemic
disease. Philadelphia: JB Lippincott;
1990:327329.
2. Macsai MS, Schwartz TL, Hinkle D, et al:
Tyrosinemia Type II: nine cases of ocular
signs and symptoms. Am J Ophthalmol
2001; 132:522527.
3. Buist NRM, Kennaway N, Fellman JG:
Tyrosinemia Type II: hepatic cytosol
tyrosine amino-transferase deciency. In:
Bickel H, Wachtel U, eds. Inherited
diseases of amino acid metabolism.
Stuttgart: George Thieme; 1985:203235.
4. Grenier A, Lescault A, Laberge C, et al:
Detection of succinylacetone and the use
of its measurement in mass screening for
hereditary tyrosinemia. Clin Chim Acta
1982; 123:93.
5. Fuerst D: Tyrosinemia in mink. In: Tabbara
KF, Cello RM, eds. Animal models of ocular
disease. Springeld, IL: Charles C Thomas;
1984:207213.
6. Burns RP, Gipson IK, Murray MJ:
Keratopathy in tyrosinemia. Birth Defects
1976; 12:169.
7. Roussat B, Fournier F, Besson D, GoddeJoly D: A propos de deux cas de
tyrosinose de type II (syndrome de RichnerHanhart). Bull Soc Ophtalmol Fr 1988;
88:751.
8. Tallab TM: Richner-Hanhart syndrome:
importance of early diagnosis and early
intervention. J Am Acad Dermatol 1996;
35:857.
9. Farage IT: Dietetic therapy of RichnerHanhart syndrome. J R Soc Med 1993;
86:495.
10. Driscoll DJ, Jabs EW, Alcorn D, et al:
Corneal tyrosine crystals in transient
neonatal tyrosinemia. J Pediatr 1988;
113:91.
11. Jonocha S, Wolz W, Srsen, et al: The
human gene for alkaptonuris (AKU) maps
to chromosome 3q. Genomics 1994;
19:34.
12. Zatkova A, deBernabe DB, Polakova H,
et al: High frequency of alkaptonuria in
Slovakia: evidence for the appearance of
multiple mutations in HGO involving
different mutational hotspots. Am J Hum
Genet 2000; 67:13331339.
13. Goldsmith LA: Cutaneous changes in errors
of amino acid metabolism: Alkaptonuria.
In: Fitzpatrick TB, Eisen AZ, Wolff K, et al,
eds. Dermatology in general medicine.
3rd ed. New York: McGraw-Hill;
1987:16421646.
13a. LaDu BN: Alkaptonuria. In: Scriver CR,
Beaudet AL, Sly WS, Valle D, eds. The
metabolic basis of inherited disease. New
York: McGraw-Hill; 1989:775.
14. Garrett EE: Ocular ochronosis with
alkaptonuria. Am J Ophthalmol 1963;
55:617.
15. Royer et Rollin MJ: Les manifestations
oculaires de lochronose. Bull Soc
Ophtalmol Fr 1965; 65:500.
16. Daicker B, Riede UN: Histologische und
ultrastrukturelle Befunde bei
alkaptonurischer Ochronosis oculi.
17. Allen RA, OMalley C, Straatsma BR:
Ocular ndings in hereditary ochronosis.
18. Wong VG, Kuwabara T, Brubaker R:

Intralysosomal cystine crystals in
cystinosis. Invest Ophthalmol Vis Sci 1970;
9:83.
19. Gahl WA, Tietze F, Bashan N, et al: Cystine
transport is defective in isolated leukocyte
lysosomes from patients with cystinosis.
Science 1982; 217:1263.
20. Cystinosis Collaborative Research Group:
Linkage of the gene for cystinosis to
markers on the short arm of chromosome
17. Nat Genet 1995; 10:246.
21. Cogan DG, Kuwabara T: Ocular pathology
of cystinosis. Arch Ophthalmol 1960; 63:75.
22. Dodd MJ, Pusin SM, Green WR: Adult
cystinosis: a case report. Arch Ophthalmol
1978; 96:1054.
23. Morris RC, Sebestian AA: Renal tubular
acidosis and Fanconi syndrome. In:
Stanbury JB, Wyngaarden JB, Fredrickson
DS, et al, eds. The metabolic basis of
inherited disease. New York: McGraw-Hill;
1983:1808.
24. Yamamoto GK, Schulman JD, Schneider
JA, et al: Long-term ocular changes in
cystinosis: Observations in renal transplant
recipients. J Pediatr Ophthalmol
25. Kenyon KR, Sensenbrenner JA: Electron
microscopy of cornea and conjunctiva in
childhood cystinosis. Am J Ophthalmol
1974; 78:68.
26. Kaiser-Kupfer MI, Caruso RC, Minkler DS,
et al: Long-term ocular manifestations of
nephropathic cystinosis. Arch Ophthalmol
1986; 104:706.
27. Melles RB, Schneider JA, Rao NA, Katz B:
Spatial and temporal sequence of corneal
crystal deposition in nephropathic
cystinosis. Am J Ophthalmol 1987;
104:598.
28. Katz B, Melles RB, Schneider JA, Rao N:
Corneal thickness in nephropathic
cystinosis. Br J Ophthalmol 1989; 73:665.
29. Katz B, Melles RB, Schneider JA: Corneal
sensitivity in nephropathic cystinosis. Am J
30. Katz B, Melles RB, Schneider JA:
Recurrent crystal deposition after
keratoplasty in nephropathic cystinosis. Am
31. Katz B, Melles RB, Schneider JA: Glare
disability in nephropathic cystinosis. Arch
Ophthalmol 1987; 105:1670.
32. Kaiser-Kupfer MI, Datiles MB, Gahl WA:
Corneal transplant in a twelve-year-old boy
with nephropathic cystinosis. Lancet 1987;
1:331.
33. Elder MJ, Aslin CL: Recurrent corneal
erosion in cystinosis. J Pediatr Ophthalmol
34. Tsilou ET, Rubin BI, Reed G: Age-Related
prevalence of anterior segment
complications in patients with infantile
nephropathic cystinosis. Cornea 2002;
21:173.
35. Schneider JA, Wong V, Seegmiller JE: The
early diagnosis of cystinosis. J Pediatr
1969; 74:114.
36. Trauner DA, Chase C, Scheler J, et al:
Neurologic and cognitive decits in children
with cystinosis. J Pediatr 1988; 112:912.
37. Schneider JA, Schulman JD: Cystinosis. In:
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.

1983:18441866.
Jonas AJ, Conley SB, Marshall R, et al:
Nephropathic cystinosis with central
nervous system involvement. Am J Med
1987; 83:966.
Kaiser-Kupfer MI, Datiles MB, Gahl WA:
Clear graft two years after keratoplasty in
nephropathic cystinosis. Am J Ophthalmol
1988; 105:318.
Gahl WA, Reed GF, Thoene JG, et al:
Cysteamine therapy for children with
nephropathic cystinosis. N Engl J Med
1987; 316:971.
Kimonis VE, Troendle J, Rose SR, et al:
Effects of early cysteamine therapy on
thyroid function and growth in
nephropathic cystinosis. J Clin Endocrinol
Metab 1995; 80:3257.
Kaiser-Kupfer MI, Fujikawa L, Kuwabara T:
Removal of corneal crystals by topical
cysteamine in nephropathic cystinosis.
N Engl J Med 1987; 316:775.
Jones NP, Postlethwaite RJ, Noble JL:
Clearance of corneal crystals in
nephropathic cystinosis by topical
cysteamine 0.5%. Br J Ophthalmol 1991;
75:311.
MacDonald IM, Noel LP, Mintsioulis G, et
al: The effect of topical cysteamine drops
on reducing crystal formation within the
cornea of patients affected by
nephropathic cystinosis. J Pediatr
Ophthalmol Strabismus 1990; 27:272.
Kaiser-Kupfer MI, Gazzo MA, Datiles MB,
et al: A randomized placebo-controlled trial
of cysteamine eye drops in nephropathic
cystinosis. Arch Ophthalmol 1990; 108:689.
Bron AJ: Corneal changes in the
dislipoproteinaemias. Cornea 1989; 8:135.
Rosenman RH, Brand RJ, Sholtz RI, et al:
Relation of corneal arcus to cardiovascular
risk factors and the incidence of coronary
disease. N Engl J Med 1974; 291:1322.
Brunzell JD: Familial lipoprotein lipase
deciency and other causes of the
chylomicronemia syndrome. In: Scriver CR,
Schaefer EJ, Levy RI: Pathogenesis and
management of lipoprotein disorders.
N Engl J Med 1985; 312:1300.
Schechter RJ: Hyperlipoproteinemia. In:
Gold DH, Weingeist TA, eds. The eye in
systemic disease. Philadelphia: JB
Lippincott; 1990:342343.
Breckenridge WC, Little JA, Alaupovic P,
et al: Lipoprotein abnormalities associated
with a familial deciency of hepatic lipase.
Atherosclerosis 1982; 45:161.
Brewer HB, Zech LA, Gregg RE, et al: Type
III hyperlipoproteinemia: diagnosis,
molecular defects, pathology and
treatment. Ann Intern Med 1983; 98:623.
Sugar J: Metabolic diseases of the cornea.
Schaefer EJ, Ordovas JM: Diagnosis and
management of high density lipoprotein
deciency states. In: Miller NE, Tall AR,
eds. High density lipoproteins and
atherosclerosis III. New York: Elsevier
Science; 1992:235.
CHAPTER 45
REFERENCES
581
SECTION 6
582
55. Schaefer EJ, Genest JJ, Ordovas JM, et al:

Familial lipoprotein disorders and
premature coronary artery disease.
Atherosclerosis 1994; 108(Suppl):S41.
56. Gustafson A, McConathy W, Alaupovic P,
et al: Identication of lipoprotein families in
a variant of human plasma apolipoprotein A
deciency. Skand J Clin Lab Invest 1979;
39:377.
57. Azoulay M, Henry I, Tata F, et al: The
structural gene for lecithin:cholesterol acyl
transferase (LCAT) maps to 16q22.
Ann Hum Genet 1987; 51:129136.
58. Schechter RJ: Hyperlipoproteinemia. In:
59. Neumann MJ, Whisiker-Lewis S: Lecithincholesterol acyltransferase deciency: rst
report of case in United States citizen.
J Am Osteopath Assoc 1994; 94:333.
60. Gjone E: Familial lecithin cholesterol
acyltransferase (LCAT) deciency. Birth
Defects 1982; 18:423.
61. Vrabec MP, Shapiro MB, Koller E, et al:
Ophthalmic observations in lecithin
cholesterol acyltransferase deciency. Arch
62. Bethell W, McCulloch C, Ghosh M: Lecithin
cholesterol acyltransferase deciency: light
and electron microscopic ndings from two
corneas. Can J Ophthalmol 1975; 10:494.
63. Viestenz A, Schlotzer-Schrehardt U,
Hofmann-Rummelt C, et al: Histopathology
of corneal changes in Lecithin-Cholesterol
Acyltransferase Deciency. Cornea 2002;
21:834-837.
64. Horven I, Egge K, Gjone E: Corneal and
fundus changes in familial LCAT deciency.
65. Fredrickson DS, Altrocchi PH, Avioli LV,
et al: Tangier disease. Ann Intern Med
1961; 55:1016.
66. Rust S, Walter M, Funke H, et al:
Assignment of Tangier disease to
chromosome 9q31 by a graphical linkage
exclusion strategy. Nature Genet 1998;
20:9698.
67. Fredrickson DS, Young O, Shiratori T, et al:
The inheritance of high-density lipoprotein
deciency (Tangier disease). J Clin Invest
1964; 43:228.
68. Hoffman HN, Fredrickson DS: Tangier
disease (familial high-density lipoprotein
deciency). Clinical and genetic features in
two adults. Am J Med 1965; 39:582.
69. Assmann G, Schmitz G, Brewer HB:
Familial high-density lipoprotein deciency:
Tangier disease. In: Scriver CR, Beaudet
AL, Sly WS, Valle D, eds. The metabolic
basis of inherited disease. New York:
McGraw-Hill; 1989:12671282.
70. Assmann G, Simantke O, Schaefer HE,
Smootz E: Characterization of high-density
lipoproteins in patients heterozygous for
Tangier disease. J Clin Invest 1977;
60:1025.
71. Schaefer EJ, Kay LL, Zech LA, Brewer HB
Jr: Tangier disease. High-density
lipoprotein deciency due to defective
metabolism of an abnormal apolipoprotein
A-I (ApoA-I Tangier). J Clin Invest 1982;
70:934.
72. Herbert PN, Forte T, Heinen RJ, et al:
Tangier disease. One explanation of lipid
storage. N Engl J Med 1978; 299:519.
73. Chu FC, Kuwabara T, Cogan DG, et al:
Ocular manifestations of familial high-
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
density lipoprotein deciency (Tangier

disease). Arch Ophthalmol 1979; 97:1926.
Ferrans VJ, Fredrickson DS: The pathology
of Tangier disease. A light and electron
microscopic study. Am J Pathol 1975;
78:101.
Marbini A, Gemignani F, Ferrarini G, et al:
Tangier disease. A case with sensorimotor
distal polyneuropathy and lipid
accumulation in striated muscles and vasa
nervorum. Acta Neuropathol (Berl) 1985;
67:121.
Pressly TA, Franklin J, Alpert M, et al:
Cardiac valvular involvement in Tangier
disease. Am Heart J 1987; 113:200.
Pressly TA, Scott WJ, Ide CH, et al: Ocular
complications of Tangier disease. Am J
Med 1987; 83:991.
Schaefer EJ, Zech LA, Schwartz DE, et al:
Coronary heart disease: prevalence and
other clinical features in familial highdensity lipoprotein deciency (Tangier
disease). Ann Intern Med 1980; 93:261.
Waldorf DS, Levy RI, Fredrickson DS:
Cutaneous cholesterol ester deposition in
Tangier disease. Arch Dermatol 1967;
95:161.
Haas LF, Austad WI, Bergin JD: Tangier
disease. Brain 1974; 97:351.
Herbert PN, Assmann G, Gotto A, et al:
Familial lipoprotein deciency:
Abetalipoproteinemia,
hypobetalipoproteinemia, and Tangier
disease. In: Stanbury JB, Wyngaarden JB,
Fredrickson DS, et al, eds. The metabolic
basis of inherited disease. New York:
McGraw-Hill; 1983:589621.
Winder AF, Alexander R, Garner A, et al:
The pathology of cornea in Tangier disease
(familial high density lipoprotein deciency).
J Clin Pathol 1996; 49:407.
Pratt-Johnson JA: Studies on the anatomy
and pathology of the peripheral cornea. Am
Herrmann WA, Winker von Mohrenfels C,
Lohmann CP: Confocal microscopy and
corneal manifestations of Tangier diseases.
Cornea 2004; 23:825827.
Carlson LA, Phillipson B: Fish eye disease:
a new familial condition with massive
corneal opacities and dyslipoproteinemia.
Lancet 1979; 2:921.
Cheung MC, Wolf AC, Lum KD, et al:
Distribution and localization of
lecithin:cholesterol acyltransferase and
cholesteryl ester transfer activity in A-Icontaining lipoproteins. J Lipid Res 1986;
27:1135.
Fielding CJ, Fielding PE: Cholesterol
transport between cells and body uids.
Role of plasma lipoproteins and the plasma
cholesterol esterication system. Med Clin
North Am 1982; 66:363.
Kuivenhoven JA, Stalenhoef AF, Hill JS,
et al: Two novel molecular defects in the
LCAT gene are associated with sh eye
disease. Arterioscler Thromb Vasc Biol
1996; 16:294.
Carlson LA: Fish eye disease: a new
familial condition with massive corneal
opacities and dyslipoproteinemia. Clinical
and laboratory studies in two aficted
families. Eur J Clin Invest 1982; 12:41.
Van Hoof F: Mucopolysaccharidoses and
mucolipidoses. J Clin Pathol 1971;
27(Suppl):64.
Krivit W, Lockman LA, Watkins PA, et al:
The future for treatment by bone marrow
92.
93.
94.
95.
96.
97.
98.
99.
100.
101.
102.
103.
104.
105.
106.
107.
108.
transplantation for adrenoleukodystrophy,

metachromatic leukodystrophy, globoid cell
leukodystrophy and Hurler syndrome.
J Inher Metab Dis 1995; 18:398.
Staba SL, Escolar ML, Poe M, et al: Cordblood transplants from unrelated donors in
patients with Hurlers syndrome. N Engl J
Med 2004; 350:19601969.
Pastores GM, Barnett NL: Current and
emerging therapies for the lysosomal
storage disorders. Exp Opin Emer Drugs.
2005; 10:891902.
Dangel ME, Mauger T: Mucolipidoses. In:
Bonten E, van der Spoel A, Fornerod M,
et al: Characterization of human
lysosomal neuraminidase denes the
molecular basis of the metabolic storage
disorder sialidosis. Genes Dev 1996;
10:31563169.
Rapin I, Goldscher S, Katzman R, et al:
The cherry red spot-myoclonus syndrome.
Am J Neurol 1978; 3:234.
OBrien JS: Neuraminidase deciency in
the cherry red spot-myoclonus syndrome.
Biochem Biophys Res Commun 1977;
79:1136.
Thomas GH, Tipton RE, Chien IE, et al:
Sialidase (alpha-N-acetyl neuraminidase)
deciency: the enzyme defect in an adult
with macular cherry-red spots and
myoclonus without dementia. Clin Genet
1978; 13:369.
McKusick VA, Neufeld EF: The
mucopolysaccharide storage diseases. In:
1983:751777.
Lang GE, Maumenee IH: Retinal
dystrophies associated with storage
disease. In: Newsome DA, ed. Retinal
dystrophies and degenerations.
Philadelphia: Lippincott-Raven;
1988:319340.
Spencer WH: Ophthalmic pathology. An
Atlas and Textbook. Philadelphia: WB
Saunders; 1985:346352.
Dangel ME, Mauger T: Mucolipidoses. In:
Caneld WM, Bao M, Pan J, et al:
Mucolipidosis II and mucolipidosis IIIA are
caused by mutations in the GlcNAcphosphotransferase alpha/beta gene on
chromosome 12p. (Abstract) Am J Hum
Genet 1998; 63:A15.
Libert J, Van Hoof F, Farriaux J, Toussaint
D: Ocular ndings in I-cell disease
(mucolipidosis type II). Am J Ophthalmol
1977; 83:617.
Traboulsi EI, Maumenee IH:
Ophthalmologic ndings in mucolipidosis III
(pseudo-Hurler polydystrophy). Am J
Bargal R, Avidan N, Ben-Asher E, et al:
Identication of the gene causing
mucolipidosis type IV. Nature Genet 2000;
26:118121.
Merin S, Livni N, Berman ER, Yatsiv S:
Mucolipidosis IV: Ocular, systemic, and
ultrastructural ndings. Invest Ophthalmol
Vis Sci 1975; 14:437.
Casteels I, Taylor DSI, Lake SD, et al:
Mucolipidosis, Type IV: Presentation of a
109.
110.
111.
112.
113.
114.
115.
116.
117.
118.
119.
120.
121.
122.
123.
124.
125.
126.
127.
mild variant. Ophthal Pediatr Genet 1992;

13:205.
Kenyon KR, Maumenee IH, Green WR,
et al: Mucolipidosis IV: Histopathology of
conjunctiva, cornea, and skin. Arch
Smith JA, Chan CC, Goldin E, et al:
Noninvasive diagnosis and ophthalmic
features of mucolipidosis type IV.
Lake BD, Milla PJ, Taylor DSI, Young EP: A
mild variant of mucolipidosis type 4 (ML4).
Birth Defects 1982; 18:391.
Dangel ME, Bremer DL, Rogers GL:
Treatment of corneal opacication in
mucolipidosis IV with conjunctival
99:137.
Riedel KG, Zwaan J, Kenyon KR: Ocular
abnormalities in mucolipidosis IV. Am J
Patel MS, Callahan JW, Zhang S, et al:
Early-infantile galactosialidosis: prenatal
presentation and postnatal follow-up. Am J
Med Genet 1999; 85:38.
Mueller OT, Henry WM, Haley LL, et al:
Sialidosis and galactosialidosis:
chromosomal assignment of two genes
associated with neuraminidase deciency
disorders. Proc Natl Acad Sci USA 1986;
83:1817.
Goldberg MF, Cotlier E, Fichenscher LG,
et al: Macular cherry-red spot, corneal
clouding, and beta-galactosidase
deciency: clinical, biochemical and
electron microscope study of a new
autosomal recessive storage disease. Arch
Intern Med 1971; 128:387.
Usui T, Sawaguchi S, Abe H, et al:
Conjunctival biopsy in adult form
galactosialidosis. Br J Ophthalmol
77:165167.
Spence MW, Clarke JTR, DEntremont DM,
et al: Angiokeratoma corporis diffusum
(Anderson-Fabry disease) in a single large
family in Nova Scotia. J Med Genet 1978;
15:428.
Gadoth N, Sandbank U: Involvement of
dorsal root ganglia in Fabrys disease.
J Med Genet 1983; 20:309.
Spaeth GL, Frost P: Fabrys disease: Its
ocular manifestations. Arch Ophthalmol
1965; 74:760.
Sher NA, Letson RD, Desnick RJ: The
ocular manifestations in Fabrys disease.
Mastropasqua L, Nubile M, Lanzini M, et al:
Corneal and conjunctival manifestations in
Fabry disease: in vivo confocal microscopy
study. Am J Ophthalmol 2006;
141:709718.
Sher NA, Letson RD, Desnick RJ: The
ocular manifestations in Fabrys disease.
Libert J, Tondeur M, Van Hoof F: The use
of conjunctival biopsy and enzyme analysis
of tears for diagnosis of homozygotes and
heterozygotes with Fabrys disease. Birth
Defects 1976; 12:221.
Wilcox WR, Banikazemi M, Guffon N, et al.
Long-term safety and efcacy of enzyme
replacement therapy for Fabry disease. Am
J Hum Genet 2004; 75:6574.
Austin JH: Studies in metachromatic
leukodystrophy. Arch Neurol 1973;
28:258.
Dierks T, Schmidt B, Borissenko LV, et al:
Multiple sulfatase deciency is caused by
128.
129.
130.
131.
132.
133.
134.
135.
136.
137.
138.
139.
140.
141.
142.
143.
144.
mutations in the gene encoding the human

C-alpha-formylglycine generating enzyme.
Cell 2003; 113:435444.
Cogan DG, Kuwabara T, Moser H:
Metachromatic leukodystrophy.
Bateman JB, Philippart M, Isenberg SJ:
Ocular features of multiple sulfatase
deciency and a new variant of
metachromatic leukodystrophy. J Pediatr
OBrien JS: The gangliosidoses. In:
1983:945.
Emery JM, Green WR, Wyllie RG, Howell
RR: GM1-gangliosidosis: ocular and
pathological manifestations. Arch
Brownstein S, Carpenter S, Polomeno RD,
Little JM: Sandhoffs disease (GM2
gangliosidosis type 2): histopathology and
ultrastructure of the eye. Arch Ophthalmol
1980; 98:1089.
Koelfen W, Freund M, Jaschke W, et al:
GM-2 gangliosidosis (Sandhoffs disease):
two year follow-up by MRI. Neuroradiology
1994; 36:152.
Neufeld EF, Muenzer J: The
mucopolysaccharidoses. In: Scriver CR,
Ashworth JL, Biswas S, Wraith E, et al:
Mucopolysaccharidoses and the eye.
Survey of Ophthalmology 2006; 51:117.
Waltman SR, Hart WM Jr: The cornea. In:
Moses RA, Hart WM Jr, eds. Adlers
physiology of the eye. St Louis: CV Mosby;
1987:36.
Frangieh GT, Traboulsi EI, Kenyon KR:
Mucopolysaccharidoses. In: Gold DH,
Weingeist TA, eds. The eye in systemic
disease. Philadelphia: JB Lippincott;
1990:372376.
Collins MLZ, Traboulsi EI, Maumenee IH:
Optic nerve head swelling and optic
atrophy in the systemic
mucopolysaccharidoses. Ophthalmology
1990; 97:1445.
Summers CG, Purple RL, Krivit W, et al:
Ocular changes in the
mucopolysaccharidoses after bone marrow
transplantation. A preliminary report.
Walker RWM, Darowski M, Morris P, et al:
Anaesthesia and mucopolysaccharidoses.
Anaesthesia 1994; 49:1078.
Kelly TE, McKusick VA:
Mucopolysaccharidoses I-H. In:
Fraunfelder FT, Roy HT, eds. Current ocular
therapy 3. Philadelphia: WB Saunders;
1990:147.
Quantock AJ, Meek KM, Fullwood NJ, et
al: Scheies syndrome: the architecture of
corneal collagen and distribution of corneal
proteoglycans. Can J Ophthalmol 1993;
28:266.
Rummelt V, Meyer HJ, Naumann GOH:
Light and electron microscopy of the
cornea in systemic mucopolysaccharidosis
Type I-S (Scheies syndrome). Cornea
1992; 11:86.
Caruso RC, Kaiser-Kupfer MI, Muenzer J,
et al: Electroretinographic ndings in the
mucopolysaccharidoses. Ophthalmology
1986; 93:16121616.
145. Baker E, Guo XH, Orsborn AM, et al: The

Morquio A syndrome
(mucopolysaccharidosis IV-A) gene maps
to 16q 24.3. Am J Hum Genet 1993; 52:96.
146. Ghosh M, McCulloch C: The Morquio
syndrome: light and electron microscopic
ndings from two corneas. Can J
147. Olsen H, Baggesen K, Sjolie AK: Cataracts
in Morquio syndrome
(mucopolysaccharidosis IV A). Ophthalmic
148. Schwartz MF, Werblin TP, Green WR:
Occurrence of mucopolysaccharide in
corneal grafts in the Maroteaux-Lamy
syndrome. Cornea 1985; 4:58.
149. Ucakhkhan OO, Brodie SE, Desnick R,
et al: Long-term follow-up of corneal graft
survival following bone marrow
transplantation in the Maroteaux-Lamy
syndrome. CLAO J; 234 2001.
150. Bergwerk KE, Falk RE, Glasgow BJ, et al:
Corneal transplanatation in a patient with
mucopolysaccharidosis type VII (Sly
disease). Ophthalmic Genet 2000;
21:1720.
151. Snyder RD, Carlow TJ, Ledman J, Wenger
DA: Ocular ndings in fucosidosis. Birth
Defects 1976; 12:241.
152. Arbisser AL, Murphree AL, Garcia CA, et al:
Ocular ndings in mannosidosis. Am J
Ophthalmol; 1976; 82:46571.
153. Fishman RS, Sunderman FW: Band
keratopathy in gout. Arch Ophthalmol
1966; 75:367.
154. Coassin M, Piovanetti O, Stark WJ, et al:
Urate deposition in the iris and anterior
chamber. Am Acad Ophthalmol 2006;
113:462465.
155. Ferry AP, Sar A, Melikian HE: Ocular
abnormalities in patients with gout. Ann
Ophthalmol 1985; 17:632635.
156. Slansky HH, Kuwabara T: Intranuclear urate
crystals in corneal epithelium. Arch
157. Dollus H, Porto F, Caussade P,
Speeg-Schatz C, et al: Ocular
manifestations in the inherited DNA repair
disorders. Surv Ophthalomol 2003;
48:107122.
158. Calonge M, Foster CS, Rice BA, et al:
Management of corneal complications in
xeroderma pigmentosum. Cornea 1992;
11:173.
159. Bonne-Tamir B, Farrer LA, Frydman M,
Kanaaneh H: Evidence for linkage between
Wilson disease and esterase D in 3
kindreds: detection of linkage for an
autosomal recessive disorder by the family
study method. Genet Epidemiol 1986; 3:201.
160. Danks DM: Disorders of copper transport.
In: Scriver CR, Beaudet AL, Sly WS, Valle
D, eds. The metabolic basis of inherited
disease. New York: McGraw-Hill;
1989:14111431.
161. Demirkiran M, Jankovic J, Lewis RA, et al:
Neurologic presentation of Wilson disease
without Kayser-Fleischer rings. Neurology
1996; 46:1040.
162. Lossner A, Lossner J, Bachmann H, Zotter
J: The Kayser-Fleischer ring during longterm treatment in Wilsons disease
(hepatolenticular degeneration). A follow-up
study. Graefes Arch Clin Exp Ophthalmol
1986; 224:152.
163. Esmaeli B, Burnstine MA, Martonyi CL,
et al: Regression of Kayser-Fleischer rings
during oral zinc therapy: correlation with
CHAPTER 45
583
164.
165.
166.
167.
168.
169.
170.
SECTION 6
171.
584
systemic manifestations of Wilsons

disease. Cornea 1996; 15:582.
Kaplinsky C, Sternlieb I, Javitt N, Rotem Y:
Familial cholestatic cirrhosis associated
with Kayser-Fleischer rings. Pediatrics
1980; 65:782.
Lipman RM, Deutsch TA: A yellow-green
posterior limbal ring in a patient who does
not have Wilsons disease. Arch
Ophthalmol 1990; 108:1385.
Tauber J, Steinert RF: Pseudo-KayserFleischer ring of the cornea associated with
non-Wilsonian liver disease. Cornea 1993;
12:74.
Dunn LL, Annabele WL, Kliegman RM:
Pigmented corneal rings in neonates with
liver disease. J Pediatr 1987; 110:771.
Wiebers DO, Hollenhorst RW, Goldstein
NP: The ophthalmologic manifestations of
Wilsons disease. Mayo Clin Proc 52:409.
Davies G, Dymock J, Harry J, Williams R:
Deposition of melanin and iron in ocular
structures in haemochromatosis. Br J
Weiss JS: Schnyder crystalline dystrophy
sine crystals. Recommendation for a
revision of nomenclature. Ophthalmology
1996; 103:465.
Weiss JS: Schnyders dystrophy of the
cornea: A Swede-Finn connection. Cornea
1992; 11:93.
172. Chern KC, Meisler DM: Disappearance of

crystals in Schnyders crystalline corneal
dystrophy after epithelial erosion. Am J
173. Bron AJ, Williams HP, Carruthers ME:
Hereditary crystalline stromal dystrophy of
Schnyder. I. Clinical features of a family
with hyperlipoproteinemia. Br J Ophthalmol
1972; 56:383.
174. Dinh R, Rapuano CJ, Cohen EJ, et al:
Recurrence of corneal dystrophy after
keratectomy. Ophthalmology 1999;
106:14901497.
175. Weller RO, Rodger FC: Crystalline stromal
dystrophy: histochemistry and
ultrastructure of the cornea. Br J
176. Ehlers N, Matthiesson ME: Hereditary
crystalline corneal dystrophy of Schnyder.
177. Brunt PW, McKusick VA: Familial
dysautonomia: a report of genetic and
clinical studies, with a review of the
literature. Medicine 1970; 49:343.
178. Gitlow SE, Bertani LM, Wilk E, et al:
Excretion of catecholamine metabolites by
children with familial dysautonomia.
Pediatrics 1970; 46:513.
179. Eng CM, Slaugenhaupt SA, Axelrod FB,
et al: Prenatal diagnosis of familial
180.
181.
182.
183.
184.
185.
dysautonomia by analysis of linked

CA-repeat polymorphisms on
chromosome 9q 31-q 33. Am J Med
Genet 1995; 59:349.
Axelrod FB, Iyer K, Fish I, et al: Progressive
sensory loss in familial dysautonomia.
McKusick VA: Mendelian Inheritance in
Man. Baltimore: Johns Hopkins University
Press; 1990:1382.
Riley CM: Familial dysautonomia: clinical
and pathophysiological aspects. Ann NY
Acad Sci 1974; 228:283.
Mass E, Wolff A, Gadoth N: Increased
major salivary gland secretion in familial
dysautonomia. Dev Med Child Neurol 1996;
38:133.
Goldberg MF, Payne JW, Brunt PW:
Ophthalmologic studies of familial
dysautonomia: the Riley-Day
syndrome. Arch Ophthalmol 1968;
80:732.
Worobec-Victor SM, Bain MAB:
Oculocutaneous genetic diseases. In: Renie
WA, ed. Goldbergs genetic and metabolic
eye disease. Boston: Little, Brown and
Company, 1988:515.
CHAPTER
46
Immunologic Disorders of the Conjunctiva,

Cornea, and Sclera
C. Stephen Foster
Key Features
The eye may be affected by immunologically driven

inflammation
Understanding the mechanism of the immune reaction helps to
guide therapy
The immune dysregulation may be from a systemic, potentially
lethal disorder
The eye can be affected by any of the immunologic hypersensitivity reactions, and understanding the mechanism of a
particular patients inflammatory problem lays the groundwork
for correct treatment. The diagnostic pursuit of a mechanistic
understanding of a patients inflammatory problem is, at the
very least, sight saving and even may be life saving.
In this chapter, ocular diseases are grouped by the primary
Gell, Coombs, and Lackmann hypersensitivity reactions that
exist at the heart of the inflammatory mechanism. The four
types of hypersensitivity reactions rarely exist in pure form (i.e.,
in isolation from each other) in human pathologic states; it is
typical for hypersensitivity reactions to have more than one of
the classic Gell and Coombs responses to the inflammatory
problem. In cases in which it is known to occur, these combinations of types of mechanisms are pointed out in the various
ocular diseases presented and discussed in this chapter.
OCULAR ALLERGIC DISORDERS

SEASONAL ALLERGIC CONJUNCTIVITIS
Seasonal allergic rhinitis and seasonal allergic conjunctivitis
(SAC) are caused by a pure type 1 hypersensitivity mechanism.
Indeed, these are perhaps the only ocular inflammatory diseases
to satisfy all of Kochs postulates, as re-phrased by Witebsky, for
proving that an inflammatory problem is immunologic: (1) one
or more causative antigens has been identified (e.g., ragweed
pollen), (2) the details of the immunologic response to the
antigen have been elucidated (e.g., immunoglobulin E
(IgE)ragweed antibody production), (3) an animal model of the
disorder has been produced, and (4) adoptive transfer between
syngeneic animals has been accomplished.
People who develop SAC are atopic. Hay fever (either seasonal or perennial), asthma, and eczema or atopic dermatitis
are considered the major atopies, and idiopathic urticaria,
nonhereditary angioedema, and food allergies are classified as
minor atopies. Together, the atopies affect 1020% of the
general population, and males are over-represented in that
number. SAC occurs in a large proportion of people afflicted by
seasonal rhinitis. Diagnosing the disorder usually is not
difficult. The patient complains of the typical symptoms of
ocular itch and watering, often in association with the sneezing

and nasal congestion symptoms of seasonal allergic rhinitis.
The seasonal influence on the appearance and disappearance of
the symptoms is obvious from the history, and a positive family
history of atopy is obtained in ~70% of patients with SAC.
Signs of ocular inflammation, even during the time of maximal
symptoms, usually are unimpressive. The globes usually are not
obviously inflamed. Indeed, the conjunctiva may appear totally
white and quiet. Further inspection by biomicroscopy, however,
often reveals mild edema of the bulbar conjunctiva and signs of
inflammation in the tarsal conjunctiva, both upper and lower.
Increased mucus is found in the preocular tear film and in the
inferior fornix.
Appropriate laboratory testing to establish the diagnosis
includes a quantitative serum IgE level, analysis of conjunctiva
(tarsal conjunctival scraping or conjunctival biopsy) with
specific attention to the presence of mast cells and eosinophils,
particularly in the epithelium, and skin testing for hypersensitivity to ubiquitous environmental allergens. The latter is
most appropriately performed by skilled allergists.
Treatment of SAC should include the following steps: environmental controls, mast cell-stabilizing agents throughout the
patients known allergy seasons, therapy for the nose with mast
cell stabilizers (e.g., azelastine) and aerosolized corticosteroid
(e.g., fluticasone propionate), systemic antihistamines when
environmental allergen exposure is unavoidable, topical combination antihistamine and mast cell stabilizers (e.g., olopatidine), and desensitization immunotherapy.
The value of the involvement of an expert allergist in the
care of a patient with SAC and other ocular allergic disorders
cannot be overemphasized. The allergist is a better environmental detective than most ophthalmologists and can provide
the patient with specific instructions for environmental control
procedures, ranging from specific air-conditioning units,
furnace air-filtering devices, air purifiers, mattress and bedding
material alterations, and specific housecleaning techniques as
well as issues relating to existing carpeting and pets, the elimination of which is sometimes essential, particularly in the more
severe forms of allergy. The allergist also is the best resource for
expert skin testing and identification of the allergens responsible
for provoking episodes of SAC and can determine whether the
use of systemic antihistamines and the embarkation on the
lengthy road of desensitization immunotherapy are appropriate.
Topical antihistamines may be helpful, temporarily, in
patients with mild SAC, but because these agents competitively
inhibit only one mediator liberated by the mast cells, they are
not as effective a therapeutic strategy as are mast cell-stabilizing
agents during long-term therapy. The latter, used correctly,
stabilize mast cell membranes, and inhibit degranulation of all
the mast cell mediators, thereby preventing major SAC attacks,
585
FIGURE 46.1. Giant papillary conjunctivitis, stage 2. By slit-lamp

biomicroscopy with white light in the absence of fluorescein dye, the
large (greater than 1 mm in diameter) papillae in the upper (lower
portion of the lid) tarsal conjunctiva are evident, but just barely.
Compare this with the ease of viewing of these papillae after the
instillation of 2% fluorescein dye (see Fig. 46.2).
The diagnosis of GPC usually is not difficult to make.

Symptoms of decreasing contact lens tolerance and increasing
mucus production in a previously successful contact lens wearer
are the primary features that should stimulate suspicion of
GPC. Some degree of ocular itch may be present, and examination of the upper tarsal conjunctiva discloses conjunctival
hyperemia and tarsal papillae greater than 1 mm in diameter
(Fig. 46.1). The geographic extent of the papillary response and
the size of the papillae, as well as the presence or absence of
epithelial erosions on the apices of the papillae, are important
features guiding therapy. Two percent fluorescein dye instilled
into the preocular tear film, with subsequent eversion of the
upper eyelid and examination of the tarsal conjunctiva with
cobalt blue-filtered light, facilitates the recognition of lowprofile papillae because the fluorescein dye outlines the
macropapillae as it lies in the valleys at their bases (Fig. 46.2).
The dye also shows stained epithelial defects at the apices of
macropapillae (Fig. 46.3).
Treatment of GPC may be difficult in the patient who is
determined to continue with contact lens wear. In most cases,
and limiting the amount of each of the hundreds of mediators

typically released by mast cells when they degranulate. Thus,
4% disodium cromoglycate, 2% nedocromil sodium, and
lodoxamide have been shown to be safe and effective in the
treatment of patients with SAC.1,2 Additionally, the dual acting
(H1 inhibition plus mast cell stabilization) agents, olopatidine,
azelastine, and epinastine 0.05% are even more user friendly in
the case of patients with ocular allergy because of twice daily
dosing effectiveness.
SECTION 6
GIANT PAPILLARY CONJUNCTIVITIS
586
Spring3 was the first to describe the condition now known as

giant papillary conjunctivitis (GPC). In 1974, he reported that
78 of 170 soft contact lens wearers developed an allergic reaction
on the upper tarsal conjunctiva, presenting with complaints
of contact lens intolerance and excessive mucus production.
Allensmith and co-workers4 more definitively described this
disorder and called it GPC because of the appearance of papillae
in the upper tarsal conjunctiva; these papillae grew larger when
the condition was left untreated. Biopsy of the conjunctival
papillae discloses mast cells in the conjunctival epithelium and
substantia propria, eosinophils in the same sites, and occasionally
basophils in the conjunctival epithelium or substantia propria.
Mast cell participation in GPC is substantially greater than
would first appear to be the case on the basis of light microscopic observations; ultrastructural studies show many more mast
cells than can be observed by light microscopy, with ultrastructural evidence of mast cell degranulation, leaving mast cell
remnants that cannot be seen at the light microscopy level.5
GPC develops as a result of tarsal conjunctival sensitization
to allergenic material present on the surface of the contact lens,
coupled with the trauma to the upper tarsal conjunctiva associated with the excursion of the eyelid over the soft lens at each
blink, an event that occurs 10 000 to 16 000 times each day.
Scanning electron microscopy studies show that within 8 h of
wear, the contact lens becomes coated with a material
composed of mucus, protein, bacteria, cells, cell debris, and airborne pollutants.6 Vigorous contact lens hygiene, with lens
cleaning and enzyme treatment, fails to remove this lens
coating completely, and successive days of wear results in a
steady buildup of the lens coat.
FIGURE 46.2. Same patient, same date, same flipped upper lid as
shown in Figure 46.1, after instillation of 2% fluorescein dye,
photographed using light filtered with a cobalt blue filter. Note the
collection of the fluorescein dye in the valleys between the papillae,
which outlines the bases of the periphery, making the detection of the
geographic extent of the bumps much easier.
FIGURE 46.3. Same patient, same eye, same flipped upper lid as
shown in Figures 46.1 and 46.2. Note the subtle staining of the apices
of some of the papillae.
Immunologic Disorders of the Conjunctiva, Cornea, and Sclera

Modifying the contact lens edge-design or using a different
polymer material often is helpful in the care of patients with
GPC of moderate severity, in whom the aforementioned steps
have not been successful in achieving lens comfort. Prescribing
a new lens of different polymer and edge-design modifies the
conjunctival trauma and allergen adherence profile in comparison with the original lens that resulted in the GPC. This,
combined with a vigorous lens hygiene program, may be all that
is required for a contact lens-intolerant GPC patient.
If all these steps result in improved patient comfort and lens
tolerance, but the patient still has distracting symptoms, then a
mast cell stabilizing agent may be added.
FIGURE 46.4. Vernal keratoconjunctivitis with the obvious giant

papillae, or cobblestones, in the upper tarsal conjunctiva.
FIGURE 46.5. Vernal keratoconjunctivitis. Note the bulbar

conjunctival injection, the milky edema of the conjunctiva, and the
characteristic ropy mucous thread on the cornea.
however, patient education about proper lens hygiene, environmental controls, modification of contact lens material and
design, and limitation of contact lens use can keep patients in
their contact lenses. Proper contact lens hygiene involves
vigorous daily cleaning with a soft contact lens cleaning agent,
hydrogen peroxide sterilization, lens storage in preserved saline
solution, and protein enzyme treatment at least twice a week;
some patients require enzymatic treatment of the lens every
other day. Conjunctival irrigation with the contact lens in place
four to eight times a day with unpreserved saline solution may
be of some benefit, and contact lens wearing time should be
kept to 50 h/week or less if the patient expects to remain a
successful contact lens wearer for many years. Finally, the
patient must be educated that the contact lens should be
considered a disposable device and encouraged to replace the
contact lens frequently. Low-cost disposable lenses may be ideal
for this reason, but even for those patients who cannot achieve
satisfactory wear with one of the available disposable lenses,
contact lens replacement every 48 weeks is recommended in
patients with GPC.
Vernal keratoconjunctivitis (VKC) is an allergic conjunctival

inflammatory disorder with (in most cases) an associated
secondary keratopathy. VKC is characterized by the classic
hallmark of giant papillae, usually in the upper tarsal conjunctiva but in some cases in the conjunctiva at the corneoscleral limbus. It is a disease predominantly of young men,
with pronounced seasonal (spring) influence, probably secondary to vernal allergens, but perennial forms exist as well. It
also can affect women, and some patients do not outgrow their
VKC. A personal or family history of atopy usually is uncovered,
and in many cases, specific allergens to which the patient is
sensitive can be determined by history and by scratch and prick
allergen skin testing. One particularly notorious provocative
allergen in patients with VKC is the house dust mite and its
feces.
The predominant symptom of VKC is itching. As a rule, the
patient spontaneously complains of profound itching. Excessive
tearing, mucus production, photophobia, and burning or foreign
body sensation are common symptoms.
The classic sign of palpebral VKC is the giant papillae or
cobblestone in the upper tarsal conjunctiva (Fig. 46.4). These
papillae markedly increase the mass of the upper lid, and hence
ptosis is an additional typical sign. Inflammation of the bulbar
conjunctiva is variable, but a ropy, lardaceous thread almost
invariably can be found in the inferior fornix (Fig. 46.5).
Patients with VKC also may develop the mucus fishing
syndrome because of this elastoid, irritating mucous thread,
with the result that an especially ocularly pernicious conspiracy
between these two problems is established.
The keratopathy of VKC typically begins as a diffuse superficial punctate keratitis. If the inflammation continues with an
outpouring of inflammatory mediators into the tear film and
with associated epithelial toxicity and possibly conspiracy from
the mechanical effects of the large papillae, a frank epithelial
defect appears next. These defects have been termed shield
ulcers because of their position and morphology (Fig. 46.6).
Epithelial defects are trophic, defying the therapeutic strategies
that usually are successful in healing corneal abrasions or
epithelial defects (i.e., lubrication, patching, wearing a soft
contact lens). The longer such trophic defects persist, the higher
is the likelihood of eventual stromal ulceration and permanent
corneal scarring. Secondary microbial infestation also may
complicate this condition (Fig. 46.7). Successful treatment of
such defects invariably requires control of the ocular
inflammatory problem (discussed later).
The limbal form of VKC was first described by Arlt in 1846,7
pre-dating the description by von Graefe of the palpebral
form by 25 years.8 This form, common in highly pigmented
people, is characterized by the presence of large papillae in
the conjunctiva at the corneoscleral limbus, with associated
collections of inflammatory cells rich in eosinophils at the
apices of the limbal papillae, the so-called HornerTrantas
CHAPTER 46
VERNAL KERATOCONJUNCTIVITIS
587
FIGURE 46.6. Shield ulcer in a patient with vernal

keratoconjunctivitis. This ulcer had persisted for 4 months before
referral.
SECTION 6
FIGURE 46.7. Vernal keratoconjunctivitis with a persistent shield

corneal ulcer treated with topical corticosteroids. Note the suppurative
keratitis in the central cornea caused by Candida albicans infection.
588
FIGURE 46.8. Limbal vernal keratoconjunctivitis. Note the white

Horner-Trantas dots on the apices of the limbal papillae.
FIGURE 46.9. Atopic keratoconjunctivitis with massive inflammatory

mound formation at the limbus.
dots (Fig. 46.8). In especially severe forms of limbal VKC,

the steady accumulation of inflammatory cells may result
in formation of a frank mound on the peripheral cornea
(Fig. 46.9).
The histopathology of the conjunctival papillae discloses
not only the cells typically associated with allergic reactions
(mast cells and eosinophils) but also large collections of
mononuclear cells, fibroblasts, and newly secreted collagen.
This tremendous influx of cells and collagen formation
increases the mass of the upper lid in palpebral VKC. The
histopathologic and immunopathologic characteristics of the
tissues has led some authorities9,10 to conclude that VKC is
not a pure type 1 Gell and Coombs hypersensitivity reaction,
but rather a combination of both type 1 and type 4 reactions.
Immunohistochemical studies show that the mononuclear
cells are rich in helper (CD4) T cells and that the cytokines
produced by the inflammatory cells are, among other things,
inducing abnormal expression of class 2 HLA glycoproteins
on conjunctival epithelium and stromal cells.9
Epidemiology
VKC has a worldwide distribution, with pronounced regional
variations in prevalence. It is most common in the
Mediterranean region and Central and South America. It is
relatively rare in North America and Northern Europe. It may
represent as much as 3% of serious ophthalmic disease in some
regions,11 whereas in Northern Europe and North America, the
prevalence is ~1 in 5000 cases of eye disease.12 VKC has been
reported to affect patients from 1 month to more than 70 years
of age, but at least 50% of the patients in most reported series
are between 5 and 25 years of age. In most patients, the disease
resolves spontaneously within 10 years of onset. It has been
associated with keratoconus, atopy, and atopic cataract. In a
study by Dart,13 78% of 120 patients with VKC developed the
disorder before the age of 16 years. Dart found that the corneal
complications of VKC in this population occurred almost
exclusively in patients with palpebral or mixed palpebral and
limbal VKC. He found no differences in serum or tear IgE levels
among VKC patients with the various forms of the disease; the
VKC patients did have higher than normal levels of IgE, and
specific IgE to cat dander and to house dust mites was detected.
Twenty-seven percent of the study population lost vision as
a result of VKC, and Dart commented that therapeutic
complications are also common and may lead to blindness.
Although it usually is an unpleasant, expensive, timeconsuming exercise, policing the patients environment, and
scrupulously cleaning it of all potential allergen provocateurs
is critical to the long-term stability of patients with VKC.
Involvement by an expert allergist is essential; this allergist
should perform not only the appropriate patch, scratch, and
prick tests as well as the serum radioallergosorbent test (RAST)
but also the environmental detective work and motivational
and educational work necessary for a successful environmental
control program. Obviously, the family must be convinced of
the long-term benefits, not only to the patient but also to the
family as a whole, before they will seriously embark on a
complex program that sometimes involves removal of expensive
carpeting, installation of air conditioning, installation of airfiltering systems in the home heating system, removal of
beloved pets, and other measures. The wisdom, importance,
and usefulness of this component of the patients care cannot
be overemphasized.
Systemic medication
Systemic antihistamine therapy is superior to topical ocular
antihistamine therapy in patients with the more complicated
allergic eye diseases, primarily because these diseases last so
long but also because these allergic patients sometimes become
sensitized to the preservatives present in the commercially
available ocular antihistamines. The use of cetirizine, fexofenadine, or loratidine usually is sufficient. In the patient with
a significant neuroconjunctivitis component of itch-scratchitch, the use of slowly escalating doses of hydroxyzine
(beginning with 50 mg at bedtime and slowly increasing as
needed) may interrupt this pernicious cycle.
Systemic desensitization immunotherapy may be indicated
in the patient who has striking sensitivity to a limited number
of allergens. Performing desensitization immunotherapy on a
patient with ocular allergy is not easy, however, and some
features of this practice are different from the typical practice of
desensitization immunotherapy in the patient with allergies not
affecting the eyes (see Atopic keratoconjunctivitis).
Ocular therapy
Mast cell-stabilizing agents are the mainstay of successful
treatment of patients with VKC. They have been shown,
unequivocally, in randomized, placebo-controlled clinical
trials,14,15 to be both safe and effective. Mast cell-stabilizing
agents available for ocular use in the United States include 4%
disodium cromoglycate, 0.1% lodoxamide, olopatidine, epinastine,
and azelastine have H-1 receptor blocking (antihistamine)
activity plus mast cell stabilization properties and are effective
with twice daily application. Nedocromil sodium, 2% eye drops,
is available in some of the Western European countries.
Topical corticosteroid therapy is required for breakthrough
attacks of highly active VKC inflammation after the patient has
encountered a stimulating allergen. The so-called pulse therapy
strategy involves administration of 1% prednisolone sodium
phosphate, 1% prednisolone acetate, or 1% rimexolone four
times daily for 2 days, with subsequent tapering to three times
daily for the succeeding 2 days, twice daily for 2 days after that,
once daily for an additional 3 days, and subsequent discontinuation thereafter. Long-term low-dose maintenance topical
steroid therapy is inappropriate.
Adjunctive ocular therapy may be required for secondary
infection, for extreme mucus production and mucous plaque
formation on shield ulcers (e.g., with 1020% N-acetylcysteine
drops, four times daily), and for persistent epithelial defect (e.g.,
prolonged bandage soft contact lens, fibronectin, epidermal
growth factor). The key to healing a persistent epithelial defect

is control of the associated inflammation.
ATOPIC KERATOCONJUNCTIVITIS
Atopic keratoconjunctivitis was defined by Hogan16 in 1952 as
allergic keratoconjunctivitis occurring in association with
atopic dermatitis (eczema). This definition, although imprecise
(patients with SAC, after all, are atopic), is in common usage
and is used here to connote the patients with the most severe
form of atopic ocular disease seen in association with eczema.
The argument by some physicians that, other types of atopic
conjunctivitis, such as chronic allergic conjunctivitis or
perennial atopic conjunctivitis, also are atopic ocular diseases
and therefore can be confused with atopic keratoconjunctivitis
is not a constructive one, particularly in view of the fact that in
those latter disorders, keratitis or significant keratopathy is not
part of the clinical picture. Corneal disease is, however, typical
of patients with atopic keratoconjunctivitis.
The term atopy originally was coined by Coca and Cooke in
1923.17 It is derived from the Greek atopos, meaning out of
place, and it is defined by a group of findings occurring in
patients with a family history of allergic disease. These findings
include hay fever, bronchial asthma, and atopic dermatitis as
the major atopies; and food allergies, urticaria, and nonhereditary angioedema as minor atopies. About 1020% of the
general population is affected by one or another of the atopic
disorders.18 The reported incidence of ocular involvement in
atopic dermatitis is between 25% and 42%.19,20 This represents
a substantial number of people who are at risk of bilateral
blinding corneal complications from this complex inflammatory disorder. Atopic keratoconjunctivitis is always bilateral.
The symptoms include ocular itch, burning and foreign body
sensation, excessive tearing, and an abundant mucoid
discharge. Cicatrizing conjunctivitis may develop with chronic
conjunctival inflammation, and lid dermatitis and chronic
blepharitis with lid thickening and meibomian gland dysfunction are typical. Loss of vision occurs as a result of corneal
scarring and neovascularization.21 The disorder has been
neglected somewhat in the ophthalmic literature, and this is
especially regrettable because successful care of the atopic
patient is complex and commonly fragmented, with a failure to
provide long-term care by the physicians involved.
Atopic patients may have multiple immunologic derangements, but one of the most notable is a defect in the T cells
responsible for regulating IgE production to one or more
allergens (ubiquitous environmental materials usually not
associated with allergy).
Therapy for AKC must occur in concert with an allergist, and
allergen avoidance is critical. Systemic signal transduction
inhibitor therapy (cyclosporine) can be life transforming, and
similar treatment of the surface (tacrolimus for the skin and
cyclosporin for the eye), combined with long-term mast cellstabilizing therapy (e.g., olopatidine), can result in a very
limited need for the use of topical corticosteroid eye drop
therapy for breakthrough episodes of AKC.
CHAPTER 46
Therapy
Allergen avoidance
TYPE 2 HYPERSENSITIVITY DISEASES OF

THE EYE
Although no disease affecting the eye has been definitively
proved to occur on the basis of a Gell and Coombs type 2
hypersensitivity reaction, the ocular consequences of cicatricial
pemphigoid, dermatitis herpetiformis, and pemphigus vulgaris
are believed to occur on this basis. This belief is based on the
nonocular findings of autoantibody deposition at the site of
disease activity22 and on the in vitro findings that such
589

antibodies are pathogenic.23 Furthermore, the immunopathologic characteristics of biopsy sampled ocular tissue
affected by these disorders is essentially the same as those
characteristics seen in skin affected by these blistering and
scarring autoimmune diseases. Hence, although circumstantial,
the evidence is strong that these diseases represent type 2
inflammatory disorders affecting conjunctiva.
OCULAR CICATRICIAL PEMPHIGOID

Ocular cicatricial pemphigoid (OCP) is a systemic autoimmune
disease with both ocular and non-ocular manifestations. Unlike
its blistering cousin, bullous pemphigoid, cicatricial pemphigoid
produces scarring of the affected skin. The so-called Brusting
Perry dermatitis of cicatricial pemphigoid typically is confined
to the scalp and thorax. Scarring also is the typical consequence
of the inflammation that affects conjunctiva and other mucous
membranes. Indeed, the disease can be fatal when it produces
strictures from scarring in the esophagus or in the trachea.
Epidemiology
Although the estimated prevalence of this disease is only 1 in
20 000 ophthalmic cases,24 in fact, the disease is probably more
common than is recognized, because, the earliest stages of the
disease are underrecognized, and patients sometimes are
treated for chronic conjunctivitis for many years before the
latter stages of the disease become grossly apparent. OCP has a
worldwide distribution, affects all races, and affects females to
a slightly greater extent than males. Although, it is said to be a
disease of old age (60s and 70s at disease onset), again, because
of the subtle nature of the subepithelial fibrosis in the earliest
stages of the disease, many cases probably begin when patients
are in their 40s and 50s.
SECTION 6
Pathogenesis
590
At least two forms of OCP exist: idiopathic and drug induced.

Studies strongly suggest that, just as in the case of so many
other autoimmune diseases, a two-hit hypothesis best fits the
available evidence regarding the pathogenesis. The first hit is
a genetic predisposition; in the case of OCP, the susceptibility
gene is at or closely linked to the HLA-DQw7 gene, and
people carrying this gene are at about a 9.6% relative risk of
developing OCP.25
The second hit in the development of pemphigoid most
probably is contact with a triggering or exciting agent in the
genetically susceptible person. Such an agent might be a virus,
as yet undefined; contact with a specific drug also may provoke
the onset of the autoimmunity in the genetically susceptible
person. One notable example of this occurred in Great Britain
in the 1970s when the new b-blocking agent for treatment of
systemic hypertension, practolol, was introduced. Shortly after
its introduction, an epidemic of cicatrizing conjunctivitis with
features indistinguishable from idiopathic cicatricial pemphigoid appeared in British ophthalmic clinics.26 Similarly, socalled pseudopemphigoid or drug-induced pemphigoid is seen
most typically in patients who have been receiving topical
medication for the treatment of glaucoma. Such cicatrizing
conjunctivitis cases have been seen after the topical use of
pilocarpine, epinephrine, timolol, and echothiophate iodide.27
Autoantibodies to a component in the conjunctiva, in the
vicinity of the lamina lucida of the basement membrane zone
are produced by people with OCP, and although the traditional
techniques for detecting autoantibodies fail to detect circulating
antibasement membrane zone antibodies in many OCP
patients, more sensitive techniques have disclosed the presence
of such antibodies in 100% of the patients tested thus far.28 Our
research aimed at defining the OCP antigen has disclosed that
the OCP antigen is different from the relevant antigen for

bullous pemphigoid. The antigen is a 205-kDa protein, the b4
peptide of a6 b4 integrin.29 The precise location of the antibody
binding site is in the intracytoplasmic portion of the b4 peptide
in the basal epithelial cell. 30
Just as in bullous pemphigoid, the autoantibody in OCP is
pathogenic in cicatricial pemphigoid. Antibody deposition, with
subsequent complement activation, probably results in a
cascade of events, including signal transduction disturbance
across the BMZ and basal epithelial cell, mast cell degranulation from anaphylatoxin, with subsequent effects on
conjunctival vasculature from the vasoactive amines liberated
from the mast cells, recruitment of macrophages and
lymphocytes, liberation of cytokines from these inflammatory
cells, vascular damage and conjunctival epithelial damage from
these cytokines, upregulation of class 2 glycoprotein expression
on conjunctival epithelium and fibroblasts, with possible subsequent contribution by these class 2 glycoprotein-bearing cells
to the inflammatory process, and fibroblast activation with
abnormal type 3 collagen secretion and with subsequent
cicatrization.31 Enormous numbers of immunologically active
cells are present in the substantia propria of patients with OCP,
and the predominant cells present are helper T lymphocytes
and macrophages.32 Systemic immunologic derangements are
present as well, including slightly abnormal proportions of
circulating helper T cells and evidence of systemic immunoreactivity with elevated levels of soluble interleukin-2 receptors,
elevated levels of soluble CD8 glycoprotein, and elevated levels
of tumor necrosis factor-a in the serum (unpublished
observations). These findings emphasize the unequivocal fact
that OCP is a systemic autoimmune disease. Even today, some
ophthalmologists still harbor the mistaken belief that OCP is a
local ocular problem that can be treated with local ocular
measures. This is not correct, and attempts to treat this disease
through local measures invariably result eventually in loss of
vision for the patient. The disease is systemic and must be
treated systemically.
Clinical Features
We initially described four clinical stages of OCP,33 and
subsequently described a more refined, precise staging system
that can enhance detection of even subtle disease progression.34
Stage 1 of the disease is characterized by chronic conjunctivitis
with mild conjunctival or corneal epitheliopathy, or both, and
subtle subepithelial fibrosis of the conjunctiva (Fig. 46.10).
FIGURE 46.10. Ocular cicatricial pemphigoid. Flipped upper lid

demonstrating the subepithelial fibrosis of the superior tarsal
conjunctiva.
FIGURE 46.12. Ocular cicatricial pemphigoid, stage 2. Note the

subepithelial fibrosis under the tarsal conjunctival epithelium, with
formation of an extensive feltwork of the new collagen, and the loss
of the normal depth of the inferior fornix (i.e., fornix foreshortening).
The latter is easily overlooked. It is best seen under the

epithelium of the upper or lower tarsal conjunctiva and presents
as fine white striae (Fig. 46.11) that, as they accumulate, may
coalesce to form a fine feltwork. Stage 2 is characterized by
progression of the cicatrizing process, with contraction of
the newly formed collagen, which foreshortens the inferior
fornix. The subepithelial fibrotic striae still may be seen
under the epithelium of the tarsal conjunctiva (Fig. 46.12).
Stage 3 is characterized by formation of the first obvious
symblepharon (Fig. 46.13). Stage 4, or end-stage disease, is
virtually untreatable and consists of a totally dry eye with
ankyloblepharon and keratinization of the cornea (Fig. 46.14).
Associated keratopathy, beginning with epitheliopathy and
progressing to corneal scarring and neovascularization, may
begin in stage 2.
A common myth about OCP is that it is a dry-eye syndrome
and, more specifically, that it is a mucin-deficient dry-eye
syndrome. This is not true until the later stages of the disease.
Indeed, in the earlier stages of the disease, patients with OCP
actually have an overproduction of mucus. This mucin, mixed
with proteins and nucleic acids from damaged cells, adheres to
the ocular surface epithelium and produces areas over which
FIGURE 46.13. Ocular cicatricial pemphigoid, stage 3. The cicatrizing

process has progressed to the point of formation of symblephara.

leatherization of the ocular surface and the adhesion of the lids to the
globe.
the tear film breaks up more rapidly than it would normally

(Fig. 46.15). The findings of a relative decrease in the numbers
of goblet cells in conjunctival biopsy specimens from patients
with OCP and of a more rapid tear film breakup time were
features that led Lemp35 to conclude, erroneously, that this
disease is a mucin-deficient dry-eye syndrome. It is true that as
the cicatrizing process progresses, with deformation of lash
follicles and compromise of lacrimal ductules and meibomian
gland ductules, aqueous and oil (and eventually mucin)
deficiencies begin to appear. With the appearance of the secondary phenomena of trichiasis, distichiasis, and tear film
abnormalities, the blinding consequences are corneal epitheliopathy, epithelial defect formation, secondary infection with
corneal ulceration, and corneal neovascularization (Fig. 46.16).
Non-ocular manifestations of cicatricial pemphigoid include
mucosal lesions in the nose, mouth, pharynx, trachea, esophagus, anus, urethra, and vagina. Any patient in whom OCP
is suspected must be questioned carefully about the presence of
dysphasia or difficulty breathing. If the physician believes that
even the remote possibility exists of esophageal or tracheal
involvement, careful endoscopic evaluation for this possibility
is essential.
CHAPTER 46

fibronic striae under the inferior tarsal conjunctiva.
591
FIGURE 46.15. Conjunctival biopsy of a patient with ocular cicatricial

pemphigoid. Note the normal surface specializations of the epithelium
(inferior half of the photograph) with microplicae and microvillae. The
superior half of the photograph, however, shows a thick mucous mat
adherent to the ocular surface epithelium.
FIGURE 46.16. Ocular cicatricial pemphigoid with extensive

keratopathy, corneal scarring, and neovascularization.
TABLE 46.1. Differential Diagnosis of Cicatrizing Conjunctivitis

Cicatrizing pemphigoid
Atopic keratoconjunctivitis
Ocular rosacea
Scleroderma
Corynebacterium diphtheriae conjunctivitis
Chemical burn
Squamous cell carcinoma
Intraepithelial epithelioma
StevensJohnson syndrome
Lyells syndrome
Sarcoidosis
Trachoma
Adenovirus conjunctivitis
Trauma
FIGURE 46.17. Conjunctival biopsy of a patient with active ocular

cicatricial pemphigoid, shown with immunofluorescence microscopy.
The primary antibody used on this specimen was antiimmunoglobulin
A. The bright apple-green, linear, continuous line of fluorescence of
the epithelial basement membrane zone shows that this patient has
large amounts of immunoglobulin-A deposited at the epithelial
basement membrane zone, an abnormal finding.
SECTION 6
Sebaceous carcinoma
592
immunoelectron microscopic techniques in cases in which

the traditional immunofluorescence analysis is negative.
Diagnosis
A variety of disorders can cause subepithelial fibrosis of the
conjunctiva with or without inflammation (Table 46.1), many
of which can be excluded on the basis of history. The definitive
establishment of a diagnosis of OCP requires the demonstration, by immunopathologic technique, of immunoreactant
(immunoglobulin or complement component) deposition at
the epithelial basement membrane zone of biopsy sampled
affected inflamed conjunctiva (Fig. 46.17). Performing
immunopathologic studies on sampled conjunctiva is difficult
because of differing requirements from immunopathology of
skin, kidney, liver, and other organs and because few centers
with skilled ocular immunopathologists exist. In addition to
the usual requirements of extremely careful, expert handling
and processing, the availability of a full panel of antibody
reagents, including antibodies to serve as positive and negative
controls, is essential. A further advantage is the ability to
analyze the tissue by enhanced immunoperoxidase and even
Treatment
Systemic therapy
Caring for patients with cicatricial pemphigoid is difficult and
multidisciplinary, requiring the close involvement of both the
ophthalmologist and a chemotherapist. The ophthalmologists
role is to apprise the chemotherapist of the state of inflammatory activity in the conjunctiva. The chemotherapists role is
to modify treatment based on therapeutic response and
systemic drug tolerance. A hand-in-glove relationship between
the ophthalmologist and an oncologist or hematologist can be
an effective collaboration. Randomized, controlled clinical trials
have shown that the systemic chemotherapeutic agents are both
safe and effective, when used properly, in ~90% of patients with
OCP.33 The current therapeutic recommendations remain the
same as previously described36: if the definitively diagnosed
pemphigoid clearly is active (inflamed) and progressive, the use
of dapsone is recommended, provided the patient is not allergic
to sulfa-containing drugs and is not deficient in glucose-6-
TABLE 46.2. Therapy for Ocular Cicatricial Pemphigoid

Agent
Initial Dose
Maximal Dose
Dapsone
25 mg b.i.d.
150 mg/day
Methotrexate
7.5 mg once weekly
15 mg once weekly
Azathioprine
2 mg/kg/day
3 mg/kg/day
Cyclophosphamide
2 mg/kg/day
3 mg/kg/day
Prednisone (adjunctive)
1 mg/kg/day
1 mg/kg/day
Cytosine arabinoside
0.2 mg/kg; 5 days/month
0.3 mg/kg; 5 days/month
Adjunctive ocular therapy

Removal of lashes that are abrading the cornea is an essential
part of the care of patients with OCP. Epilation is temporarily
effective, but the lash that has re-grown after epilation in fact
may produce more damage than the original lash that was
epilated; the re-grown lash produces a stabbing-type injury to
the cornea, and the re-growth is short and stiff, whereas the
original lash may have been a long, supple lash that was
brushing over the surface of the cornea. Permanent destruction
of the lens follicles is the ideal treatment, although accomplishing this can be difficult. Electrolysis, cryoablation, and
marginal lid rotation and follicle extirpation surgeries all have
been employed with varying degrees of success. Keratinized
posterior lid margin conjunctiva may respond to topical retinoid
therapy provided the immunologically driven conjunctival
inflammation has been brought completely under control with
immunosuppressive therapy. The effect of topical retinoids is
variable and often unimpressive, however, and lid margin
mucous membrane grafting procedures may be required to treat
extensive keratinization of tarsal conjunctiva effectively.37
Aqueous deficiency should be treated with punctal occlusion
and attention to the lids and meibomian glands (warm
compresses and meibomian gland massage with lid hygiene,
with or without systemic tetracycline therapy), and ocular
lubricants, preferably without preservative, may be indicated.
Corneal hypoesthesia and lagophthalmos should be treated
with tarsorrhaphy.
Corneal surgery is highly ill-advised in patients with
advanced OCP. The outcome almost invariably is the formation
of an epithelial defect, with subsequent stromal ulceration and
perforation. Penetrating keratoplasty may be used to treat a select

few patients with corneal scarring but with good lid function
and reasonably good tear production. In most patients with
pemphigoid and significant corneal pathology, however, corneal
sensibility is profoundly impaired, the eye is very dry, and normal
lid function is impaired. For these patients, keratoprosthesis is
the only realistic surgical alternative that holds any hope for
visual rehabilitation.
PEMPHIGUS VULGARIS
Pemphigus vulgaris is a blistering autoimmune skin disease
that in the past was universally fatal. It affects people of all ages
and has widespread geographic distribution, but Ashkenazi Jews
are markedly overrepresented in groups of patients with
pemphigus vulgaris. This link led to the discovery of the HLA
association with this disease; susceptibility to the disease is
carried by the HLA-DQ3 gene DQw8. The relevant antigen of
pemphigus vulgaris is the intercellular cement of epithelial cells
(an adhesion molecule or cadherin previously designated PVA
for pemphigus vulgaris antigen) and now known to be desmoglein 3, a member of the cadherin transmembrane adhesion
molecule family which interacts with plakoglobin in the
desmosome. Additional targets of the autoimmune autoantibody response are probably operative as well, most
particularly anticholinergic receptor antibodies.38
The circulating autoantibody (IgG, IgA, or IgM) deposits on
this 130-kDa glycoprotein, activating complement, releasing
plasminogen activator, forming plasmin, and dissolving the
bonds between adjacent epithelial cells. The epithelial cells then
separate from one another (acantholysis) forming blisters under
the surface epithelium. These blisters rupture, leaving an intact
layer of basal epithelium attached to a normal basement
membrane at the base of the blister.
Ocular Manifestations
Lid skin manifestations of pemphigus vulgaris are not rare, but
frank conjunctival manifestations are. Conjunctival bullae with
subsequent rupture and erosions can occur, however; but these
heal with no evidence of the subepithelial fibrosis, fornix foreshortening, and symblepharon formation typical of OCP.39 Rare
cases of the latter have been reported, perhaps because of
secondary infection or because the patient had concomitant
pemphigus vulgaris and OCP.40
CHAPTER 46
phosphate dehydrogenase. If the response to dapsone is

incomplete, once-weekly methotrexate or daily azathioprine or
mycophenolate mofetil is added to the therapeutic program. If
there is no response to the dapsone, it is discontinued, and
either azathioprine or mycophenolate mofetil or once-weekly
methotrexate is used. Patients who fail these regimens and
those who have extremely active, rapidly progressive OCP are
treated with high-dose systemic prednisone and either daily oral
cyclophosphamide or intravenous-pulse cyclophosphamide.
After the inflammation is controlled, the daily prednisone is
tapered, changed to alternate-day therapy, tapered further, and
eventually discontinued within 3 months. Selected patients
who also failed to respond to these treatment regimens adequately
have responded to subcutaneous cytosine arabinoside. This
approach to OCP therapy has been reserved for otherwise resistant cases because of its logistically cumbersome administration
requirements. Cyclosporine is remarkably ineffective in treating
OCP. The chemotherapeutic drugs used to treat this disease
and their starting dosages and dose ranges are listed in
Table 46.2. Finally, intravenous biologic therapy with intravenous
immunoglobulin or with intravenous daclizumab.
Treatment
In the past, patients with pemphigus vulgaris died of sepsis and
electrolyte imbalance from the extensive fluid loss from open
skin lesions. High-dose systemic corticosteroid therapy changed
that, and the introduction of combined cytotoxic immunosuppressive therapy with corticosteroid therapy appears to have
induced cases of long-term remission. The cytotoxic agents
593

employed in the care of patients with pemphigus vulgaris have
included azathioprine, cyclophosphamide, methotrexate and
intravenous immunoglobulin.
DERMATITIS HERPETIFORMIS
Ocular Manifestations
Dermatitis herpetiformis is an autoimmune blistering dermatosis characterized by a pruritic eruption, usually of the scalp,
buttocks, lower back, and extensor surfaces of the arms.
Mucous membrane involvement, including ocular involvement,
is rare but can occur. The disease is chronic, with remissions
and exacerbations. The autoantibody generally is of the IgA
class, and immunochemical studies of biopsy sampled affected
skin show deposition of IgA and complement in a granular (or
in rare cases, linear) pattern at the dermalepidermal junction.
The target autoantigen is tissue transglutaminase 2. Glutensensitive enteropathy is associated strikingly with dermatitis
herpetiformis, and there is an HLA-DR3 and HLA-B8 genetic
association. The ocular manifestations reported are those of a
chronic or recurrent cicatrizing conjunctivitis producing
subepithelial fibrosis and symblepharon.41
Treatment
Dapsone is an extraordinarily effective therapy for dermatitis
herpetiformis. The usual caveats apply regarding sulfa allergy and
glucose-6-phosphate dehydrogenase deficiency. Starting with doses
of 50 mg/day and escalating to, in most cases, 150200 mg/day
is typical. Concomitant systemic corticosteroid therapy and
antihistamine therapy also may be used, and dietary analysis
with gluten restriction is employed by many dermatologists in
the care of patients with dermatitis herpetiformis.
TYPE 3 HYPERSENSITIVITY DISEASES OF

THE EYE
SECTION 6
MOORENS ULCER
594
The evidence that Moorens ulcer has a type 3 hypersensitivity

mechanism as at least part of its etiopathogenesis is not vast.
This disease is included here, however, because some
circumstantial evidence for this type of mechanism exists and
because the disease has some striking similarities to the
peripheral ulcerative keratitis (PUK) associated with some of the
classic type 3 circulating immune complex systemic disorders,
such as rheumatoid arthritis and polyarteritis nodosa (PAN).
Moorens ulcer, in fact, was first described by Bowman in
1849,42 and McKenzie43 later remarked on chronic serpiginous
ulcer of the cornea in 1854. Moorens name, however, became
attached to this curious disorder because of his publication of
cases in 1867.44 The disease is rare, with fewer than 200 cases
described in the worlds ophthalmic literature; some of the
cases included in reports of Moorens ulcer are in fact not true
Moorens ulcer but rather cases in which PUK was the
presenting manifestation of an occult systemic disease.
Moorens ulcer is, by definition, idiopathic, occurring in the
absence of any diagnosable systemic disorder that could be
responsible for the progressive destruction of the cornea. It also
is strictly a PUK, with no associated scleritis. This latter point
is of substantial importance because so many of the misdiagnosed cases were in patients who had PUK in association
with adjacent scleritis, necrotizing or otherwise. Wood and
Kaufman45 emphasized the distinction between limited
Moorens ulcer, unilateral, usually occurring in older people,
and bilateral malignant Moorens ulcer, which typically was
relentlessly progressive despite all previously described
treatments.
Interestingly, my experience has been just the opposite; most

of my patients with progressive, bilateral Moorens ulcer have
been older than 40 years,46 and a careful review of all reported
cases suggests that the bilateral form of the disease is
overrepresented in the older patients in the worlds reported
cases of Moorens ulcer.47
This latter form of the disease was observed to occur most
often in younger males. An especially interesting group of young
African males has been described with bilateral Moorens ulcer,
and some evidence was collected suggesting that helminthiasis
was associated with the disease and that progression of PUK
was halted by local ocular therapeutic measures combined with
systemic therapy for the helminthiasis.48 Although Schazlin49
and others have conjectured that the Ascaris and Ancylostoma
species parasites caused the Moorens ulcer, possibly through
antigenantibody reactions to helminth toxins deposited in the
peripheral cornea, helminth infestation is epidemic in the
countries where these cases of Moorens ulcer in young African
males have occurred, and yet the disease, even in these endemic
areas of ascariasis, is rare. Other putative causes of some cases
of Moorens ulcer have included trauma, herpes, hepatitis C
virus, and ocular surgery. Again, with the exception of the cases
that follow trauma, it is difficult to indict the other agents as
true causes of PUK associated with Moorens ulcer simply
because herpes simplex virus, varicella-zoster virus, and
cataract surgery themselves can be associated with subsequent
PUK, often with associated necrotizing scleritis. This is not the
entity that Mooren described.
Clinical Features
Moorens ulcer is a PUK that begins in clear cornea at the
corneoscleral limbus and progresses centrally, circumferentially,
and posteriorly through the cornea, leaving a thinned, vascularized corneal residua in its wake (Fig. 46.18a). The edge of the
progressive ulcer is undermined, and the extent of dissolution
of the cornea may be surprising as one gently probes the
overhanging edge of the ulcer with a thin probe (Fig. 46.18b).
White blood cell infiltrates in the corneal stroma in advance of
the edge of the ulcer are characteristic. The condition is painful,
and the pain is usually well out of proportion to obvious signs
of ocular (conjunctival) inflammation. A low-grade iritis may be
present, and spontaneous or traumatic perforation may occur.
The epithelium of the central edge of the ulcer remains intact;
conjunctival epithelium covers the thinned, vascularized cornea
left in the wake of the advancing ulcer. This may give the
impression that there is a progressive corneal thinning with
associated keratitis but without an epithelial defect. This is not
true, and the presence of the narrow, crescent-shaped epithelial
defect can be seen after instillation of 2% fluorescein eye drops,
with subsequent forced lid closure for 30 s and with evaluation
of the eye using cobalt blue light. The destructive process
progresses slowly, with eventual destruction of the entire extent
of the cornea, leaving thinned, vascularized residua. The
sometimes unbearable pain the patient has experienced
throughout this saga, suddenly vanishes when the process has
finally swept over the entire geographic extent of the cornea, but
this process may take 418 months.
Pathogenesis
Immunoglobulins and complement are found in the peripheral
cornea of patients with PUK typical of Moorens ulcer, and this
has suggested the possibility that Moorens ulcer represents a
type 3 immune complex deposition hypersensitivity reaction.
As with all type 3 diseases, the predominant cell attracted to the
site of immune complex deposition is the neutrophil, which is
found in great abundance in the area of corneal destruction.
The corneal destruction results from liberation of proteases and
FIGURE 46.18. (a) Moorens ulcer. Note the peripheral ulcerative keratitis that has begun at the 3 oclock position and progressed clockwise to
the 6:30 position, counterclockwise to the 1 oclock position, and centrally to involve (apparently) ~2.5 mm of the cornea. Note also the
inflammatory infiltrates in advance of the edge of the ulcer. (b) Probing of the edge of this ulcer demonstrates that the extent of the undermining
and hence of destruction of the corneal stroma is astonishingly much greater than was apparent on inspection by slit-lamp biomicroscopy.
Diagnosis
The differential diagnosis of PUK is shown in Table 46.3.
Scrapings for culture generally establish an infectious cause in
infectious PUK, although I have experience with three patients
whose PUK was caused by a limbal vasculitis secondary to
herpes simplex virus; the diagnosis was established only after
conjunctival resection and analysis, by immunohistologic techniques, for the presence of herpes antigens. The vasculitic
collagen diseases typically cause an associated scleritis with the
PUK that may accompany them. But because each of these may
produce isolated PUK, and because PUK may be the presenting
manifestation of one of these potentially lethal systemic
vasculitic disorders and may precede obvious nonocular clinical
manifestations by many months, any patient with presumed
Moorens ulcer should be evaluated for the possibility of an
occult systemic vasculitic disease. Moorens ulcer can then be
appropriately diagnosed after these other disorders are excluded
(see Table 46.3).
TABLE 46.3. Differential Diagnosis of Peripheral Ulcerative

Keratitis
Ocular, Infectious
Any microbe
Ocular, Noninfectious
Moorens ulcer
Systemic, Infectious
Gonococcus
Bacillary infection
Acquired immunodeficiency syndrome
Tuberculosis
Syphilis
Systemic, Noninfectious
Rheumatoid arthritis
Systemic lupus erythematosus
Polyarteritis nodosa
Wegeners granulomatosis
Relapsing polychondritis
Behets disease
Sarcoidosis
Inflammatory bowel disease
Rosacea
CHAPTER 46
collagenase from the neutrophil granules. The substantia

propria of the conjunctiva adjacent to the ulcerating cornea is
filled with plasma cells. Immunoglobulins found in the peripheral cornea may be manufactured locally by these plasma
cells in the conjunctiva. Schaap and associates50 found
circulating autoantibodies to cornea in patients with Moorens
ulcer, and Berkowitz and colleagues51 found circulating immune
complexes in patients with PUK associated with Moorens ulcer.
Interestingly, removal of conjunctiva adjacent to the ulcerating
cornea, combined with resection of the necrotic, ulcerating
cornea and application of tissue adhesive to exclude neutrophils
from access to the region, results in an instantaneous cessation
of corneal destruction. This surgical procedure is curative in
most patients with the limited, unilateral form of Moorens
ulcer.52 The destructive process resumes, however, in patients
with the bilateral form once the conjunctiva has re-grown to the
limbus and has been re-populated with the immunocompetent
cells responsible for antibody and cytokine production. These
patients usually require systemic cytotoxic chemotherapy to
stop the progressive destruction.
Evidence of autoimmunity directed against an antigen or
antigens from corneal stroma in a patient with bilateral
Moorens ulcer in 197953 have been found, and Gottsch and
associates have identified a specific corneal protein as the
probable target autoantigen in this autoimmune disease.54
Treatment
The recommended initial treatment of progressive Moorens
ulcer (Fig. 46.19) is wide conjunctival resection to bare sclera,
extending at least to the 2 oclock position on either side of the
peripheral ulcer and posteriorly for 4 mm. This is followed by
resection of the overhanging lip of ulcerating cornea and
application of tissue adhesive, with soft contact lens application
and with subsequent topical 1% prednisolone sodium
phosphate application four times daily. Others have advocated
aggressive (every 1530 min) topical steroid therapy, describing
595

has become so extensive that surgical efforts at rehabilitation
are required.
Rehabilitative surgical therapy for this disease generally
requires two procedures: (1) lamellar tectonic grafting, followed
by (2) definitive central penetrating grafting. The lamellar graft
is required because of lack of sufficient peripheral cornea into
which a penetrating graft might be secured.
COLLAGEN VASCULAR DISEASES

Rheumatoid Arthritis
SECTION 6
FIGURE 46.19. Moorens corneal ulcer with perforation, after

conjunctival resection, ulcer dbridement, and application of
cyanoacrylate tissue adhesive at the perforation site at the 2 oclock
position on the limbus.
596
good results with this approach in patients with the limited

form of Moorens ulcer.52 Systemic tetracycline may be used
for its anticollagenolytic properties, as may topical 1%
medroxyprogesterone, eight times daily.
Bilateral progressive Moorens ulcer almost always requires
systemic cytotoxic chemotherapy to stop the progressive
corneal destruction. I typically treat bilateral cases that have not
progressed too extensively with the aforementioned approach
(i.e., conjunctival resection with tissue adhesive, soft contact
lens, and topical steroids) first. With the first appearance of
recurrence of keratitis, however, I will institute therapy with
systemic methotrexate (7.515 mg once weekly), azathioprine
(2 mg1 kg1 day1), or cyclophosphamide (2 mg1 kg1 day1).
This approach has been highly successful.53 A small number of
reports have been made of successful therapy of unilateral
Moorens ulcer treated with topical cyclosporine.55 Systemic
therapy for the bilateral cases typically is continued for 6
months, with subsequent attempts at tapering and discontinuation of the medication.
The anamnesis and the ocularly pernicious nature of this
disease are documented amply in the ophthalmic literature.
Attempts at reparative corneal surgery almost always are
unsuccessful unless the underlying Moorens disease activity
has been controlled. Even when the disease has burned itself
out, attempts at corneal grafting usually are associated with
recurrence of Moorens ulcer and destruction of the graft.56 An
example of this latter circumstance was seen in a patient57 who
underwent grafting 15 years after Moorens ulcer had destroyed
his cornea. This patient was not immunosuppressed before
grafting, and Moorens ulcer recurred in the edge of the
transplant. The graft perforated, the patient developed
endophthalmitis, and the eye was enucleated. The immune
system has a remarkable memory, and this is a vivid illustration
of that. Patients with burned-out Moorens ulcer should be
immunosuppressed before cataract surgery or corneal grafting
procedures. Furthermore, I disagree vigorously with the
statement by Schazlin47 that since the value of immunosuppressive therapy is less clear than other treatments, they are
recommended only in the severest and most resistant cases.
On the contrary, it is believed that the evidence for the efficacy
of systemic immunosuppressive chemotherapy for progressive
bilateral Moorens ulcer is strong and that the evidence suggests
that such treatment should be employed sooner rather than
later in the care of such patients, before the corneal destruction
Rheumatoid arthritis is a crippling, potentially lethal collagen

vascular disease that can affect the eye in many ways, the most
common of which is through autoimmune damage to the
lacrimal gland with resultant dry-eye syndrome. It is possible
(even likely) that some of the ocular devastation and morbidity
associated with rheumatoid arthritis could be prevented if
patients with the earliest ocular manifestations of this disease
were treated more aggressively. For example, since rheumatologists have turned to high-dose intravenous-pulse steroid
therapy and to once-weekly methotrexate therapy earlier in the
care of patients with the nonocular manifestations of
rheumatoid arthritis, fewer of the serious, late ocular
manifestations of this disease have been seen. I have cared
for one patient with rheumatoid arthritis who developed
pronounced, acute dacryoadenitis with associated profound
keratoconjunctivitis sicca (KCS); this patients lacrimal gland
inflammation resolved with systemic corticosteroid therapy,
with resultant resolution of the KCS. Two additional
recurrences of this phenomenon were similarly treated with like
outcomes.
Keratoconjunctivitis Sicca
The KCS associated with rheumatoid arthritis can be highly
destructive. It usually manifests with the typical dry-eye
symptoms of foreign body sensation, blepharitis, excessive
mucus production, and a sensation of dryness or paradoxical
tearing, all of which progress the longer the patient is awake.
The typical signs observable on evaluation include a diminished
marginal tear strip, superficial punctate keratitis, and ocular
surface epitheliopathy with rose bengal staining in the typical
horizontally oriented diamond pattern in the intrapalpebral
fissure (Fig. 46.20). In especially severe or neglected cases, the
superficial punctate keratitis may progress to filamentary
FIGURE 46.20. Keratoconjunctivitis sicca after instillation of 1% rose

bengal solution. The photograph has been taken with a red-free light
source. Note the punctate staining of the corneal epithelium in the
interpalpebral fissure.
keratopathy, epithelial defect formation, stromal ulceration, and

corneal perforation (Fig. 46.21).
Treatment for KCS always should include vigorous attention
to the meibomian glands and lid margins. In KCS, meibomian
gland dysfunction nearly always is present, and a deficient oil
layer component to the preocular tear film is at least as
important (if not more so) than a deficient aqueous layer in the
abnormal function of the tear film. Lid hygiene with vigorous
warm compresses and lid massage, with or without the use of a
systemic tetracycline derivative, therefore is indicated in the
care of patients with KCS. Additionally, I am relatively
aggressive about puncta occlusion in these patients. If Schirmer
values (after the use of topical anesthetic and subsequent
inferior fornix blotting with tissue) are consistently 3 mm or
less, he occludes the puncta permanently with a hyphrecator.
For Schirmer values of 310 mm, I insert punctum plugs into
the inferior puncta. Artificial tears are still used, but better
insights gained through recent studies may indicate that use of
commercially available artificial tears may be more harmful
than helpful.58 In any case, my current practice is to be cautious
with the use of artificial tears and to have the patient
experiment with one commercial product in one eye and
another in the other eye, trying various preparations until the
best preparation is found. The deleterious effects of preservatives in some of the commercial preparations, as compared
with the lack of convenience and risk associated with the
preparations not containing preservatives, should be weighed
by both the physician and the patient. Topical cyclosporine
(Restasis ) twice daily can result in decreased lacrimal gland
lymphocytic infiltration and increased tear production in
patients with active dacryoadenitis but residual glandular acini
capable of function after resolution of imflammation.
Additional tear conservation methods, including specially
prepared glasses with side shields, the use of room humidifiers
in dry climates and during the winter, and even marginal
tarsorrhaphy, are other appropriate therapeutic measures that
may be employed, depending on the severity of the case.
Episcleritis and Scleritis

Between 1981 and 1996, 172 patients with scleritis and 94
patients with episcleritis were cared for by me, and these
patients formed the basis of our text on the sclera.59 Three of
the 94 patients with episcleritis had rheumatoid arthritis, and
32 of the 172 patients with scleritis had rheumatoid arthritis.
The episcleritis in the patients with rheumatoid arthritis was
FIGURE 46.22. Diffuse scleritis. Note the slightly purple contribution

to the red appearance of the eye.
relatively trivial in all cases, although it generally required the

use of a systemic nonsteroidal antiinflammatory drug (NSAID)
for resolution, unlike typical idiopathic episcleritis. Topical
corticosteroids are never indicated in the care of a patient with
episcleritis because the evidence is abundant that, although
they may be effective temporarily, their use prolongs the natural
history of resolution of episcleritis and makes each recurrence
more difficult to treat through a rebound phenomenon.59
Topical NSAIDs are also ineffective in treating episcleritis.60
Scleritis, on the other hand, is an extremely serious problem
that threatens the patients vision. It always requires systemic
treatment, and topical steroids may be indicated as adjunctive
therapy. Additionally, it is critical that the ophthalmologist
distinguish between episcleritis and true scleritis because the
latter carries serious nonocular implications. Patients with
rheumatoid arthritis who develop true scleritis must be watched
carefully because they are at higher risk of developing necrotizing scleritis and subsequent potentially lethal systemic
vasculitis.6163 Distinguishing between episcleritis and scleritis
can be difficult. I agree with Watson and Hazelman64 that the
presence of scleral edema is the sine qua non for establishing
that a patient has scleritis. All of the other discriminatory signs
and symptoms are important, but the critical issue is whether
one can determine that scleral edema is present. Patients with
scleritis commonly complain of severe pain; those with
episcleritis rarely do. Patients with scleritis have an intensely
red eye with a violaceous or purple hue (Fig. 46.22), whereas
patients with episcleritis have pink or bright-red conjunctival
and episcleral inflammation from dilation of the vessels in the
superficial episcleral vascular plexus (Fig. 46.23). These dilated
vessels typically blanche with the use of topical phenylephrine
(5% drops), whereas the dilated vessels in the deep episcleral
vascular plexus associated with scleritis often remain dilated
after the use of such drops. Palpating the globe through the
closed lids or through the lids while the patient is looking in one
gaze or another generally elicits tenderness to the touch in the
patient with true scleritis and little discomfort in the patient
with episcleritis. Slit-lamp biomicroscopy with a thin slit beam
is the only way to determine whether underlying scleral edema
exists. If it does, the beam is bowed forward as it makes its
excursion across the scleral surface (focusing through the conjunctiva and the blanched conjunctival and episcleral vessels
after the use of topical phenylephrine).
In 32 patients with true scleritis associated with rheumatoid
arthritis, 11 had diffuse scleritis, five had nodular scleritis, 11
had necrotizing scleritis, four had scleromalacia, and one had
CHAPTER 46
FIGURE 46.21. Keratoconjunctivitis sicca with filament formation in

the supernasal sector of the peripheral and midperipheral corneal
epithelium.
597
FIGURE 46.23. Diffuse episcleritis. Note the brilliant red characteristic

of the conjunctival inflammation.
FIGURE 46.24. Necrotizing scleritis. Note the loss of sclera

superiorly, down to choroid, and the associated extensive avascular
area temporal to this area of near perforation in the right eye.
posterior scleritis. The patients with necrotizing scleritis

presented the most difficulty in management and experienced
the most extensive nonocular systemic vasculitic problems
(Fig. 46.24). Associated uveitis carried an especially poor visual
prognosis.65
Systemic therapy always was necessary for the successful
treatment of patients with scleritis. Systemic NSAIDs with or
without short-term high-dose systemic corticosteroid therapy is
the appropriate initial treatment in most patients with scleritis,
with the notable exception of those with necrotizing scleritis.
Patients with rheumatoid arthritis who develop necrotizing
scleritis are at high risk of dying from a vasculitic event within
25 years of onset of the necrotizing scleritis61,62 unless they are
treated with systemic immunosuppressive chemotherapy.63
Cyclophosphamide (2 mg1 kg1 day1) is probably most effective
for the ocular and nonocular consequences of rheumatoid
vasculitis, although methotrexate (7.515 mg once weekly),
azathioprine (2 mg1 kg1 day1), and cyclosporine (5 mg1 kg1
day1) may also work. The choice of drug and management of
the patient should involve close collaboration between
chemotherapist and ophthalmologist. High-dose short-term
systemic prednisone may be indicated for temporary control
while chemotherapy induction is in progress, with subsequent
(within 4 weeks) prednisone taper, a switch to alternate-day

dosing, and eventual (within 3 months) discontinuation of the
prednisone.
Oral NSAIDs are usually sufficient for treatment of diffuse or
nodular scleritis. Just as in the case of NSAID response to
arthritis, however, the response of the patient with scleritis to
the NSAID initially chosen is unpredictable. I typically try at
least three NSAIDs in succession before concluding that a
patients scleritis is NSAID resistant. A list of the NSAIDs I
have used, with dosages, is given in Table 46.4. Eight of 11
rheumatoid arthritis patients with diffuse scleritis responded to
NSAID therapy alone.
Peripheral Ulcerative Keratitis

Peripheral corneal ulceration may be seen in patients with
rheumatoid arthritis in association with adjacent severe scleritis, usually necrotizing, but PUK may develop in the absence
of clinically obvious scleritis (Fig. 46.25). The pathogenesis of
these lesions has a vasculitic basis, with immune complex
localization in peripheral cornea and limbal vessels, chemotaxis
of inflammatory cells (particularly neutrophils and histiocytes),
and inflammatory cell enzyme liberation with resultant
collagen and proteoglycan destruction. The PUK typically
SECTION 6
TABLE 46.4. Nonsteroidal Antiinflammatory Drugs
598
Trade Name
Generic Name
Dose
Dolobid
Diflunisal
500 mg b.i.d.
Naprosyn
Naproxen
250500 mg b.i.d.
Indocin
Indomethacin
75 mg SR b.i.d.
Motrin
Ibuprofen
800 mg t.i.d.
Feldene
Piroxicam
20 mg q.d.
Butazolidin
Phenylbutazone
100 mg t.i.d.
Nalfon
Fenoprofen
600 mg t.i.d.
Voltaren
Diclofenac
75 mg b.i.d.
Tolectin
Tolmetin
400 mg t.i.d.
Meclomen
Meclofenamate
100 mg q.i.d.
Ansaid
Flurbiprofen
100 mg t.i.d.
Orudis
Ketoprofen
100 mg t.i.d.
FIGURE 46.25. (a) Peripheral ulcerative keratitis in a patient with rheumatoid arthritis. Note the 360-degree ring infiltrate of inflammatory cells
with associated destruction of the peripheral cornea. (b) Same eye as shown in (2), after conjunctival resection, ulcer dbridement, application of
cyanoacrylate adhesive, and application of a soft contact lens.
Marginal Furrow
Marginal corneal thinning without obvious inflammatory cell
infiltration into the cornea and without an overlying epithelial
defect may occur in patients with rheumatoid arthritis.66 The
cause of these marginal furrows, which often are in the inferior
aspect of the cornea, is unknown. The corneal epithelium is
intact over the progressively thinning cornea, and changing
degrees of corneal astigmatism may result. The lesions rarely
progress to the point of threatened perforation, do not
vascularize, and have no known effective treatment. Systemic
collagenase inhibitors, such as tetracycline and doxycycline,
may slow progression of these lesions.
Corneoscleral Ulceration after Cataract Surgery

Peripheral and, more rarely, central corneal ulceration and
necrotizing scleritis have been reported in patients with
rheumatoid arthritis (usually associated with KCS).6771
Although Smith and Schanzlin67 impugn unrecognized dry eye
and epithelial damage at the time of cataract surgery, with
delayed treatment of postoperative epithelial defects as the
primary culprit in these cases, in many cases (particularly those
in which the pathology is strictly peripheral), the pathogenesis
appears to come from a slow accumulation of immune com-
plexes and immune complex-mediated vasculitic damage after

surgical trauma. The evidence for this mechanism of the
destructive lesions is particularly compelling in cases in which
the onset of PUK and necrotizing scleritis was delayed by weeks
or months after the cataract operation. Conjunctival resection
and removal of ulcerating scleral tissue at the time of scleral
transplantation in some of these patients has disclosed classic
vasculitis identical to that seen in scleral biopsy specimens from
patients with rheumatoid arthritis who developed idiopathic
necrotizing scleritis.71 Corneal ulcers in patients with
rheumatoid arthritis after cataract surgery may progress to
descemetocele formation, perforation, and loss of the eye.6770
In cases of central corneal ulceration, aggressive treatment of
dry eye and exposure, through tarsorrhaphy and the use of
tissue adhesive if the progression of the central ulcer is rapid, is
essential if the globe is to be preserved. In patients with peripheral ulceration with or without necrotizing scleritis, systemic
immunosuppression after resection of conjunctiva and
demonstration of unequivocal vasculitis is indicated if the globe
is to be salvaged.
Systemic Lupus Erythematosus

Episcleritis and Scleritis
Episcleritis or scleritis may occur in patients with systemic
lupus erythematosus (SLE) and may be the initial manifestation
of the disease. True scleritis is a reasonably accurate guide to the
presence of significant systemic activity in patients with SLE
and resolves only with adequate control of systemic disease
activity; it does not respond to topical therapy. In my practice,
one of 94 patients with episcleritis had SLE, and seven of 172
patients with scleritis had SLE.59 The episcleritis was more
persistent than idiopathic episcleritis, and the patient insisted,
eventually, on systemic therapy. Systemic NSAIDs usually
eliminate the episcleritis completely, with the side benefit of
eliminating or decreasing the arthralgias experienced in a
significant proportion of patients with SLE. Systemic
hydroxychloroquine (200 mg twice daily by mouth) also was
highly effective in eliminating the episcleritis from one SLE
patient who was placed on this drug primarily for the skin and
constitutional symptoms associated with the SLE. Five of seven
SLE patients with scleritis responded completely to oral
NSAIDs, whereas two patients required the addition of
prednisone to the treatment regimen. No patient with SLEassociated scleritis required the institution of immunosuppressive therapy for control of the scleritis.
CHAPTER 46
progresses both centrally and circumferentially, relentlessly,

unresponsive to commonly used topical forms of therapy for
corneal ulceration. The extraocular significance of PUK is
underrecognized; its significance is the same as that of necrotizing scleritis (see earlier) even if evidence of vasculitis at the
time of PUK development is lacking. Treatment for these
lesions is the same as that for necrotizing scleritis, that is, with
systemic immunosuppressive therapy after wide conjunctival
resection, ulcer dbridement, application of tissue adhesive, and
the application of a soft contact lens. Topical (medroxyprogesterone, 1%, every 2 h while awake) and systemic (tetracycline, 250 mg by mouth four times daily) collagenase inhibitors
or collagenase synthetase inhibitors may be used to some effect,
although it is negligent not to immunosuppress patients with
rheumatoid arthritis who develop PUK or necrotizing scleritis.
The most effective drug, as stated previously in the discussion
of necrotizing scleritis, is cyclophosphamide, but depending on
whether one or both eyes is involved and depending on the
speed with which the process is progressing, once-weekly
methotrexate or daily azathioprine or cyclosporine may be used
instead.
599
Keratitis
Corneal manifestations of SLE are confined primarily to ocular
surface epitheliopathy, although PUK has been described, and
keratitis with neovascularization has been described in patients
with discoid SLE. KCS, however, is extremely common in SLE
patients with inadequately controlled systemic disease.
Pillat72 reported superficial keratitis in one of 16 SLE
patients, and Gold and co-workers73 found a 6.5% incidence of
keratitis in an SLE outpatient population. Spaeth74 found that
88% of SLE patients hospitalized at the National Institutes of
Health had superficial punctate keratitis with fluorescein corneal staining, even though their Schirmer values were normal.
Reeves75 reported the occurrence of deep bilateral segmental
interstitial keratitis and subsequent recurrent iritis in a patient
who, 1 year later, developed cutaneous, articular, and hematologic manifestations of SLE. Halmay and Ludwig76 had described similar findings in SLE patients.
Analysis of the entire population of SLE patients presenting
to me between 1981 and 1996 found that 16 of 47 patients had
corneal complications from the disease and that 62.5% of these
patients had associated KCS.
Therapy for the corneal disease associated with SLE includes
control of the underlying SLE disease activity as well as
adjunctive therapy for the eye. The well-known approach to
KCS (see the previous discussion of KCS associated with
rheumatoid arthritis), as well as the brief, judicious use of
topical corticosteroids for associated true inflammation of the
cornea, is the appropriate approach.
SECTION 6
Polyarteritis Nodosa
600
Cornea External ocular manifestations of PAN include ptosis,

exophthalmos, extraocular muscular paresis, episcleritis,
conjunctival chemosis, scleritis, and PUK. Corneal and scleral
lesions in this disease are highly destructive and progressive
unless the correct diagnosis is made and control of the underlying systemic disease is achieved.
PUK may occur in PAN and may be its presenting
manifestation.77 A clinical characteristic of PUK in these cases
is corneal ulceration at the corneoscleral limbus that is
progressive centrally and circumferentially. This ulceration is
associated with ocular pain and inflammation and with
undermining of the central edge of the ulcer, resulting in an
overhanging lip of cornea producing a peripheral ulcer that
morphologically resembles Moorens ulcer. Adjacent sclera has
been involved in all of the reported cases, and this involvement
is a distinguishing characteristic from idiopathic localized PUK
of Moorens corneal ulcer.
Local therapy for these ulcers has failed routinely. Diagnosis
of the underlying systemic condition and institution of
adequate systemic therapy to control the disease and the
destructive ocular lesions are essential. It is possible, however,
to retard progressive corneal destruction in cases of PUK
associated with PAN using local therapy while the underlying
systemic disease is being controlled. Conjunctival resection,
ulcer dbridement, application of cyanoacrylate tissue adhesive
to the ulcer bed and to a small rim of surrounding normal
cornea and sclera, and application of a continuous-wear
bandage soft contact lens can be used to delay the degradation
process while the patient is being immunosuppressed. The use
of topical corticosteroids to inhibit inflammatory cell activity
and migration does not appear particularly effective and may be
harmful because it inhibits new collagen formation. Inhibitors
of collagenase synthesis, such as 1% medroxyprogesterone
drops, and competitive inhibitors of collagenase, such as
N-acetylcysteine (20% drops) and systemic tetracycline derivatives, are adjunctive forms of therapy that may retard ulcer
progression while the systemic disease is being controlled.
The histopathology of the corneal lesions of PAN shows

almost exclusively a neutrophil infiltrate of the cornea, with
large numbers of plasma cells, histiocytes, and lymphocytes
and with various numbers of eosinophils in the adjacent
conjunctiva. Microvasculopathy typically is demonstrable in the
conjunctiva and in the sclera that is resected from areas of
associated necrotizing scleritis. The mechanisms involved in
the production of corneal and scleral lesions in PAN appear the
same as those associated with the vasculitis lesions extraocularly: immune complex deposition and inflammatory cell
chemotaxis into the area of deposition of immune complexes,
with vessel and adjacent tissue damage from the digestive
enzymes liberated from the phagocytes attracted to the area of
immune complex deposition.
Sclera
The scleritis of PAN may be diffuse, nodular, or necrotizing. It
is always painful and never responds to local therapy or to oral
NSAIDs. Successful resolution of PAN always requires the
use of systemic cytotoxic immunosuppressive agents, and
cyclophosphamide is the most effective agent. PAN is a lethal
disease that, if left untreated, has a 5 year mortality rate of
87%.78 Treatment with systemic corticosteroids improves the
5 year survival rate to only 50%,79 whereas the use of oral
cyclophosphamide results in induction of permanent remission
in most cases and in a 5 year survival rate of 80%.79 The
ophthalmologist who diagnoses the patient with scleritis with
occult PAN not only has provided the opportunity to save
the patients eye but also has set the stage for saving the
patients life.
In my practice, PAN was diagnosed in two of 172 patients
with scleritis.59 One patient had diffuse scleritis, and the other
had necrotizing scleritis; both cases were unilateral. PUK
accompanied the necrotizing scleritis.
Wegeners Granulomatosis
Ocular involvement commonly occurs in Wegeners granulomatosis (5060% of cases), and the ocular lesion may be the
presenting symptom. Focal ocular manifestations of Wegeners
granulomatosis include conjunctivitis, episcleritis, scleritis, and
PUK. Conjunctivitis and episcleritis are the most common, and
if the underlying Wegeners granulomatosis is not diagnosed,
these relatively benign ocular manifestations become chronic
and eventually may be accompanied by the more ominous PUK
and necrotizing scleritis.
Cornea
PUK is reported with increasing frequency as the initial
significant manifestation of Wegeners granulomatosis.80 It
commonly is preceded by localized conjunctivitis or episcleritis,
followed by the onset of true scleritis and the development of
intrastromal peripheral corneal inflammatory infiltrates. Pain
may be mild or severe. The corneal ulceration develops with
breakdown of the peripheral corneal epithelium, and the
crescentic peripheral corneal ulcer progresses centrally and
circumferentially, just as do that of Moorens ulcer and that of
PUK associated with PAN (Fig. 46.26). The ocular lesion is
relentlessly progressive despite local medical and surgical
therapy; control of the ocular lesion depends entirely on control
of the underlying systemic disease. Strategies to delay the
disease process, as mentioned previously in the discussion of
PUK associated with PAN, are appropriate while systemic
immunosuppressive chemotherapy is being instituted. In my
practice, three of seven patients with Wegeners granulomatosis
presented with PUK as the initial manifestation of their occult
Wegeners granulomatosis. Careful review of systems and
laboratory pursuit of the diagnosis (before the advent of the

otherwise, who fail to exhibit ANCA normalization while on
systemic chemotherapy.
Histopathologic analysis of resected conjunctiva from the
area adjacent to PUK rarely is diagnostic, but biopsy of sclera in
patients with necrotizing scleritis commonly is diagnostic. In
the presence of a typical granulomatous inflammatory reaction
with epithelioid cells and multinucleated giant cells,85
eosinophil infiltration may have special diagnostic significance,
especially if these cells are activated.
Sclera
antineutrophil cytoplasmic antibody (ANCA) testing) resulted

in a definitive establishment of the diagnosis in all cases,
through established histopathologic criteria. With the advent of
ANCA testing, ophthalmologists should be able to establish the
diagnosis of Wegeners granulomatosis more quickly, and less
ocular damage and less permanent renal or pulmonary damage
should result.81
Multiple studies have confirmed that the presence of ANCA
is highly sensitive and specific for Wegener s granulomatosis.8284 The specificity is ~99%, and the sensitivity
depends on the extent of disease and is 96% for active
generalized disease, 67% for local regional disease, and 32% for
patients in remission after initial loco-regional symptoms.
Experience with ANCA testing in a large population of patients
with scleritis demonstrates the extraordinary utility of this test
in diagnosing occult manifestations.81 Indeed, every patient
with scleritis should undergo ANCA testing as part of the
diagnostic laboratory evaluation. Microscopic polyangiitis and
crescentic glomerulonephritis also can produce a positive
ANCA test, and patients with the so-called perinuclear ANCA
(P-ANCA) staining pattern usually have these other two
diseases. Patients with Wegeners granulomatosis can also show
this pattern, although they usually show the diffuse granular
cytoplasmic pattern of staining (C-ANCA). Specific autoantibody analysis by ELISA discloses that autoantibodies
directed against myeloperoxidase (MPO) accounts for the
perinuclear pattern of stain on immunofluorescence (IF)
analysis, while antibodies directed against proteinase 3 (PR3)
account for the cytoplasmic pattern of IF staining.
ANCA testing is important not only diagnostically but also
therapeutically. Patients with Wegeners granulomatosis are
treated to bring about a total resolution of ocular and nonocular
manifestations of the disease and, if at all possible (short of
producing systemic toxicity), to normalize the serum ANCA.
Failure of the ANCA to disappear from the patients serum is
associated with a significant risk of reactivation of Wegeners
disease after discontinuation of cyclophosphamide or other
immunosuppressant agents. In several patients with the limited
form of Wegeners granulomatosis and with ocular manifestations, all ocular and sinus inflammation was abolished, and
remission of disease was maintained for more than 1 year, with
subsequent discontinuation of systemic cyclophosphamide; but
in the patients in whom the ANCA did not normalize,
Wegeners disease activity recurred, not only in the eye but also
in the lungs or kidneys. Thus, extreme vigilance is essential in
treating patients with Wegeners granulomatosis, limited or
CHAPTER 46
FIGURE 46.26. Necrotizing scleritis and peripheral ulcerative keratitis

in a patient with Wegeners granulomatosis.
The scleritis associated with Wegeners granulomatosis may be

diffuse, nodular, or necrotizing. In my practice, three of seven
patients with Wegeners granulomatosis and scleritis exhibited
diffuse scleritis, one had nodular scleritis, and three developed
necrotizing scleritis.81 As mentioned previously, specimens of
ulcerating sclera commonly show neutrophils in the area of
active scleral degradation, fibrinoid necrosis, and surrounding
granulomatous inflammation, with epithelioid cells and
multinucleated giant cells (Fig. 46.27). If intrascleral vessels are
obtained in the biopsy specimen, true necrotizing vasculitis
with inflammatory cell infiltration into the vascular wall and
fibrinoid necrosis of the vessel may be seen. Even if true
vasculitis cannot be seen, a constellation of histopathologic
features should make the ophthalmologist and pathologist
particularly suspicious of Wegeners granulomatosis.85
I have never seen a patient with Wegeners granulomatosis
with associated PUK who did not have involvement of the
adjacent sclera. As in patients with Wegeners disease who have
PUK, systemic therapy is the key to salvage of the globe in those
with scleritis. Systemic cyclophosphamide has been shown,
unequivocally, to be the treatment of choice for patients with
Wegeners granulomatosis; the 5 year mortality rate for patients
with this disease who are not treated or who are treated with
only systemic prednisone is 95%,86 whereas the 5 year mortality
rate for patients treated with systemic cyclophosphamide is
10%.86 Cyclophosphamide is started at a dose of 2 mg1 kg1
day1, with dosing restricted to breakfast and with encouragement of high levels of fluid intake in the afternoon and evening.
Monitoring of the ANCA levels, the peripheral hemogram, liver
enzymes, blood urea nitrogen, chest radiograph, sinus films,
and ocular lesions longitudinally is recommended. The drug
dosage is adjusted according to clinical disease activity, ANCA
levels, and systemic tolerance of the medication. A particular
cause of concern is when the patient appears clinically to be in
FIGURE 46.27. Hematoxylin and eosin stain showing the

histopathology of a scleral biopsy in a patient with Wegeners
granulomatosis. Note the extensive array of multinucleated giant cells.
601

remission but the ANCA levels remain high, breakthrough
relapse attacks or relapse of potentially lethal Wegener s
granulomatosis activity after drug cessation is common in these
patients. Ordinarily, however, cyclophosphamide should be
tapered and withdrawn after a 6 month1 year period of total
clinical quiescence and replaced by longterm once a week
methotrexate maintenance therapy. Azathioprine maintenance
is also acceptable. Patients who relapse are retreated with
cyclophosphamide. This disease does not respond well to
trimethoprim and sulfa combinations. Adjunctive prednisone is
used as needed for disease control when bone marrow tolerance
of cyclophosphamide is at its limit and clinical remission has
not been achieved. Azathioprine, methotrexate, chlorambucil,
and cyclosporine are alternative therapies; scattered anecdotal
reports have been made of successful induction of remission
using these drugs in patients intolerant to cyclophosphamide.
Pneumocystis carinii prophylaxis is generally accomplished with
twice weekly trimethoprimsulfa preparations.
three patients with necrotizing scleritis, in whom high-dose

systemic steroids, penicillamine, methotrexate, and azathioprine
had failed to control the progressively destructive inflammation.
The need for immunosuppressive therapy to control some
cases of relapsing polychondritis is not surprising; it is an
autoimmune disease in which autoantibodies to type 2 collagen
(which exists in both sclera and cartilage) and cell-mediated
immunity to cartilage components have been found. It is
interesting in this regard that sclera and cartilage share a
common phylogenetic origin.
Scleritis is a marker of severity of the underlying disease in
patients with connective tissue and vasculitic syndromes, and
as in other vasculitides, such as rheumatoid arthritis, Wegeners
granulomatosis, PAN, and Behets disease, the onset of necrotizing scleritis is a reliable sign of potentially lethal systemic
vasculitis and a clear indication for immunosuppressive
chemotherapy.
Progressive Systemic Sclerosis
SECTION 6
Relapsing Polychondritis
602
Relapsing polychondritis can be fatal when it affects the trachea

or the kidney. The most obvious manifestations of this disease
usually are in the external ear and in the cartilage of the nose
(Fig. 46.28), but the eye can be involved, and when it is,
necrotizing scleritis with or without PUK is the most ominous
manifestation. Because the disease is rare and the multisystem
involvement may not be simultaneous, the definitive diagnosis
of relapsing polychondritis often is delayed; in this case, the
prognosis is poor.
In 11 of my cases of relapsing polychondritis with scleritis,87
achieving complete control of necrotizing scleritis associated
with relapsing polychondritis was extremely difficult. Three of
these patients had necrotizing scleritis, and two had PUK.
Scleritis was bilateral in four patients, and the ocular inflammation was the presenting manifestation of the relapsing
polychondritis in three patients. Seven developed auricular
chondritis, six developed nasal chondritis, six developed arthritis,
two developed tracheal chondritis, four developed damage to the
cochleovestibular system, one developed renal involvement,
and one developed central nervous system vasculitic
manifestations of the disease. Cytotoxic immunosuppressive
chemotherapy was required to bring about total resolution of
the destructive inflammation in seven of the 11 patients.
Although dapsone is often effective in the care of patients with
relapsing polychondritis in whom auricular or nasal chondritis
is the primary manifestation, it was ineffective in all of the
patients studied who had necrotizing or nodular scleritis and
was effective in only two of the four patients with diffuse
scleritis. Cyclophosphamide was the only effective drug in the
FIGURE 46.28. Facial profile of a patient with relapsing

polychondritis. Note the destruction of the nasal cartilage from
episodes of nasal chondritis.
Progressive systemic sclerosis, or scleroderma, is associated

with subepithelial fibrosis of the conjunctiva, KCS, and
blepharophimosis. The tear insufficiency has been well documented, and it is treated using the usual techniques, mentioned
previously in the discussion of KCS in rheumatoid arthritis.
The fornix foreshortening requires no treatment, but simply the
recognition that progressive systemic sclerosis is associated
with subepithelial fibrosis of the conjunctiva, which should not
make the ophthalmologist consider the associated diagnosis of
cicatricial pemphigoid.8890 Chapter 330 illustrates both the
ocular and the non-ocular salient clinical findings in patients
with progressive systemic sclerosis.
StevensJohnson Syndrome
StevensJohnson syndrome (SJS), or erythema multiforme
major, is a life-threatening systemic illness most commonly
precipitated by a type 3 hypersensitivity reaction to a microbe.91
Mycoplasma pneumoniae is a common cause of this disease in
children, and children affected by SJS caused by M. pneumoniae
may die of the consequences of the SJS or may die of the
undiagnosed pneumonia. Herpes simplex virus is a common
and underrecognized cause of precipitation of SJS, and other
viruses, including polio, vaccinia, variola, and mumps, have
been associated with fulminant SJS. Mycobacterium
tuberculosis and various other microbiologic agents also have
been implicated. In truth, many cases of SJS that are blamed
on medication probably have occurred as a result of a
hypersensitivity reaction to the microbe for which the patient
was taking the medication and not from the medication itself,
which typically is blamed. Some drugs, however, are clearly
implicated in cases of SJS. Notable examples include
sulfonamides, tetracycline, penicillin, NSAIDs, allopurinol,
barbiturates, and various immunizing vaccines.
The disease is characterized by the systemic manifestations
of fever, malaise, headache, loss of appetite, nausea, and
vomiting. The dermatologic manifestation is a generalized
erythematous papular eruption, including involvement of the
soles of the feet and the palms of the hands, with eventual
emergence of the so-called target or iris or bulls-eye lesion with
an erythematous center surrounded by a zone of relatively
normal-appearing skin and then by an erythematous ring
outside of that (Fig. 46.29). Mucous membranes typically are
involved in SJS, with nose and mouth the most common sites
affected, but vagina, anus, and conjunctiva are affected in a high
percentage of cases. The mucosal involvement is one of bullae
formation and rapid rupture of these bullae, with subsequent
scarring in the area of the epithelial erosions (Fig. 46.30). The
nails are also affected by the disease (Fig. 46.31).
FIGURE 46.29. Erythema multiforme major. Note the classic target

lesions affecting the palms of the hands.
FIGURE 46.31. Fingernail involvement in erythema multiforme major.

Note also the typical skin lesions.
FIGURE 46.30. Erythema multiforme major with oral mucosal

involvement and mild conjunctival involvement.
The best evidence suggests that the immunopathology of SJS

is a combined mechanism involving circulating IgA-containing
immune complexes and a lymphocytic vasculitis in areas where
these IgA complexes lodge in vessel walls.92 A lymphocytic
microvasculitis is an immunopathologic characteristic of the
lesions of SJS.
mian gland duct compromise, misdirection of lash follicles with

resultant trichiasis and distichiasis, and chronic keratopathy
secondary to KCS, meibomian gland dysfunction, the abrading
action of the misdirected lashes, and the abrading action of
the keratinization of tarsal conjunctiva posterior to the grayline
of both the upper and the lower lids (Fig. 46.33). Corneal
epithelial defect formation, neovascularization, and stromal
ulceration with stromal scarring or perforation are the blinding
consequences of this cruel disease.
CHAPTER 46
FIGURE 46.32. Ocular involvement in StevensJohnson syndrome,

with subepithelial fibrosis, fornix foreshortening, and dense
symblepharon formation.
Treatment
Conjunctivitis, conjunctival bullae formation, keratopathy, and
secondary infection are the typical acute ocular manifestations
of SJS. These manifestations may clear without chronic
sequelae, but in severe cases of SJS, the chronic consequences of
the acute exanthem are the features with which the
ophthalmologist and the patient must struggle for the rest of
the patients life. The subepithelial fibrosis of the conjunctiva
produces these chronic consequences, with many features
similar to those of stage 3 or 4 cicatricial pemphigoid. Specifically, the subepithelial fibrosis produces conjunctival fornix
foreshortening, symblepharon formation (Fig. 46.32), meibo-
General supportive topical antibiotic and corticosteroid therapy

during the acute phases of SJS is the therapy most typically
employed for the eyes. However, increasing interest in acute
intervention with brief high dose corticosteroid therapy93
(provided the patient is not in sepsis) and/or with IV-Ig therapy
(provided the patient is not dehydrated) is gaining in
popularity.94 Acute care of the ocular manifestations of SJS may
appropriately include careful ocular hygiene with regard to
crusts and mucus, judicious use of topical corticosteroids,
vigilance for the formation of adhesions between raw surfaces,
gentle separation of such adhesions when observed, and use of
topical antibiotics for prophylaxis of infection.
603
SECTION 6
FIGURE 46.33. Typical keratinization of the tarsal conjunctiva for

~2 mm posterior to the mucocutaneous junction, lower lid, in a
patient with Stevens-Johnson syndrome.
604
Treatment of the chronic ocular consequences appropriately

includes control or correction of the trichiasis and distichiasis,
preferably through permanent destruction of the follicles of the
aberrant lashes, treatment of KCS if it exists (SJS patients are
often not tear deficient), treatment to the extent possible of
meibomian gland dysfunction and meibomian duct obstruction
through the use of warm compresses and lid massage, and
treatment of keratopathy secondary to the keratinized posterior
lid margins. Topical retinoids may or may not be helpful in the
latter regard; lubrication with ointments may help in some
cases. When the problem is severe, however, removal of the
keratinized tissue and replacement by mucosal membrane
grafting is the most definitive and effective therapy for this
problem (Fig. 46.34).
Scleral lens therapy, for protection from abrading lashes and
keratinized lid margins and for retention of fluid in the space
formed between the cornea and the posterior surface of the
scleral lens, can be extremely helpful, not only for chronic
management but also for postoperative protection of limbal
stem cell and corneal grafts. After all, placement of delicate
limbal stem cell allografts or the epithelium of a corneal
transplant into the same hostile environment that has resulted
in recurrent or persistent corneal epithelial defects is unlikely to
succeed in repopulating the patients ocular surface with
healthy cells that can resist the assault of offending lashes, KCS,
and keratinized posterior lid margins unless some protection is
provided against these assailants.
Immunosuppressive therapy has no role in the care of
patients with the chronic consequences of SJS. Whether such
treatment might be of benefit during the acute phases of the
disease is unknown and probably cannot be answered unless or
until a proper prospective study of the immunopathogenesis of
the disease is performed. In many cases (e.g., in patients in
whom a microbe-like herpesvirus or M. pneumoniae has
stimulated the SJS), acute intervention with immunosuppressive agents probably would be contraindicated. There is
one extremely rare case, however, in which immunosuppressive
chemotherapy is effective in the care of patients with SJS:
immunologically driven, truly recurrent SJS. This rare
phenomenon, well described in skin and in oral mucosa,90 also
has been described in nine of my patients with recurrent
conjunctival inflammation for many years after the acute
exanthem of SJS.95 Immunohistochemical studies of the
affected conjunctiva that continued to exhibit pronounced
inflammation after appropriate control of all the potentially
FIGURE 46.34. A patient with StevensJohnson syndrome who had

extensive keratinization of the tarsal conjunctiva is shown after
resection of that keratinization and performance of buccal mucosal
membrane grafting to both upper and lower lids.
confounding variables, such as lid margin keratinization, sicca

syndrome, meibomian gland dysfunction, trichiasis, and
distichiasis, disclosed IgA deposition in vessel walls and
lymphocytic microvasculitis. Immunosuppressive chemotherapy in these nine patients resulted in abolition of the
recurrent attacks of chronic inflammation that was producing
progressive scarring of the conjunctiva.
Lyells Syndrome
Lyells syndrome, or toxic epidermal necrolysis (TEN), can
occur as a result of Staphylococcus aureus infection in infants
and young children. The staphylococcal toxin production of
TEN produces the so-called scalded skin syndrome, with
generalized peeling of the epidermis in large geographic areas
of the skin and of the mucous membranes.96 Ocular
manifestations in this group are not extremely common but
can occur and include mucopurulent conjunctivitis with
conjunctival scarring and keratopathy. The more common form
of TEN, however, seen in older people, usually is not
staphylococcal but may be microbe related nonetheless. The
medication used often is blamed for the exanthem, however,
and such medications have included penicillin, allopurinol,
NSAIDs, sulfonamides, and phenytoin. As is the case in SJS,
TEN is systemic and the patient may die. Indeed, the mortality
rate for both SJS and TEN approaches 30%,97 and this is
probably underappreciated by most ophthalmologists.
The immunopathogenesis of TEN probably begins with drugskin binding with an aberrant immune response to this bound
form of the drug and with resultant attack on skin and mucous
membrane. Complement and immunoglobulin deposition
within the epidermis and mucosa occurs, with resultant
inflammatory cell infiltration and damage secondary to the
inflammatory cells.
Headache and malaise precede the appearance of bullae on
the skin. Unlike the tense subepidermal bullae of SJS, the bullae
of TEN are intraepidermal and hence are flaccid. The bullae
rupture, large expanses of epidermis are lost, and mucous
membrane lesions often develop in the nose, mouth, trachea,
esophagus, and conjunctivae.
The ocular manifestations of TEN essentially are identical to
those of SJS. Conjunctivitis with epithelial erosions and
subepithelial fibrosis and fornix foreshortening, symblepharon
formation, trichiasis, distichiasis, and meibomian duct and, in
FIGURE 46.35. Blinding keratopathy consequences include scarring

and neovascularization in a patient with toxic epidermal necrolysis.
some cases, lacrimal duct obstruction occur. Keratopathy with

corneal ulceration, neovascularization, perforation is typical, as
in the case of SJS (Fig. 46.35).
Treatment
The treatment of TEN is the same as that of SJS, with acute
supportive care, antibiotic prophylaxis, ocular hygiene,
attention to adhesions, and judicious use of topical corticosteroids. Care of the chronic consequences are as described
previously for SJS.
TYPE IV HYPERSENSITIVITY REACTIONS

PHLYCTENULOSIS
bacterial proteins. Although the disease is still seen today in

American Eskimos and in disadvantaged cultures in which
tuberculosis is still prevalent, the occasional case seen in more
developed societies is found generally in association with
staphylococcal proteins. Candida and Coccidioides species and
lymphogranuloma venereum have also been associated with
phlyctenulosis.98
The epithelium overlying this inflammatory lesion generally
develops a defect at the apex during the course of ~1 week, with
subsequent ulceration of the lesion and the residua of a limbal
scar (Fig. 46.37). When the cornea has been involved to a
significant degree, the patients clinical picture may evolve to
that of Saltzmanns nodular corneal degeneration (Fig. 46.38).
Topical corticosteroids are the mainstay of treatment for
phlyctenulosis. The lesions are exquisitely sensitive to topical
steroids, and the equivalent of 1 drop of 1% prednisolone
sodium phosphate twice daily is generally effective within 48
to 96 h. Concomitant topical antibiotic therapy and lid hygiene
to control the almost invariably associated staphylococcal
blepharitis and meibomianitis is mandatory. The application of
CHAPTER 46
Phlyctenulosis, which was relatively common when tuberculosis was prevalent, is relatively rare in developed countries
today. The lesions, which are granulomatous, typically appear
at the corneoscleral limbus (Fig. 46.36), but in severe disease,
they may occur in conjunctiva posterior to the limbus, in the
cornea, or both. Organisms cannot be demonstrated in these
lesions, which are believed to be secondary to a classic Gell
and Coombs type IV delayed-type hypersensitivity reaction to
FIGURE 46.37. Limbal scarring at the 3 oclock and 5 oclock

positions in a patient who previously had phlyctenulosis. Note also
that this patient now has Saltzmanns nodular degeneration with
corneal lesions in the periphery and midperiphery of the cornea, in the
7 oclock sector.
FIGURE 46.36. Phlyctenulosis. Note the bowing forward of the thin

slit beam as it sweeps across the phlyctenular conjunctival lesion near
the corneoscleral limbus.
FIGURE 46.38. Saltzmanns nodular degeneration with a solitary

lesion at the 6 oclock periphery.
605
CORNEAL TRANSPLANT REJECTION
FIGURE 46.39. Drug allergy contact dermatitis affecting the lids of

the right eye.
warm compresses with lid massage and lid scrubs twice daily,
along with an antibiotic such as bacitracin ointment instilled
into the cul-de-sac twice daily, typically is sufficient.
DRUG ALLERGY (CONTACT

HYPERSENSITIVITY)
SECTION 6
Type 4 contact hypersensitivity reactions to ocular medications

are not rare, and the most culpable medications in provoking
such reactions in the susceptible patients conjunctiva and lid
skin are neomycin, atropine, penicillin, and antazoline. The
diagnosis must first be suspected, based on a history of ocular
medication use and the clinical appearance (Fig. 46.39), which
typically includes erythematous, scaly dermatitis, affecting
upper and lower eyelids, with the lower lid skin (and even down
onto the cheek) being more affected than the upper. Resolution
of this problem, including the erythematous conjunctiva,
within 4 days of cessation of medication use offers strong
circumstantial evidence of the accuracy of the diagnosis.
Definitive diagnosis requires patch testing, with the allergen
applied to the skin under an occlusive patch for 48 h. The site
of application is unremarkable 24 h later but shows a classic
delayed-type hypersensitivity response with erythema and
induration 4872 h after application.
Cool compresses usually are indicated, along with withdrawal of the medication, in the treatment of patients with
contact dermatitis. Occasionally, 0.5% hydrocortisone skin
cream applied to the affected skin area is indicated, depending
on the severity of the allergic reaction and on the social
circumstances of the patient.
Corneal transplantation, like any other solid organ transplantation, provokes a systemic immune response in the
recipient, contrary to the mistaken beliefs of Sir Peter Medawar
dating from his important observations in the early 1950s.99
The cornea and the anterior chamber are immunologically
privileged by virtue of the avascularity of the cornea and by
virtue of the anterior chamber route of class 2 HLA glycoprotein
presentation of alloantigens from the transplanted cornea and
not by virtue of antigen invisibility from the systemic immune
system. Rapid, potent systemic immune responses occur after
foreign antigen presentation into the cornea or anterior
chamber, but curiously, the systemic immune response results
predominantly in a tolerating sequence of events that actively
produces the immunologic privilege of these sites.100104 Such
immunologic privilege includes the tolerance usually enjoyed by
the antigens on corneal allografts.
As with transplant rejection of other solid organs, however,
transplant rejection of transplanted corneas represents primarily a type 4 hypersensitivity reaction. Such a reaction is
mediated primarily by T cells, although lymphokines liberated
by these T cells may recruit other cell types that can participate
in the damage to the graft. Such surrogate effector cells include
natural killer cells and macrophages. In extremely rare cases,
hyperacute rejection of corneal grafts may occur in the
previously sensitized recipient by virtue of the presence of
preexisting antibodies in the blood of the recipient that react
with donor tissue. In this so-called antibody-dependent cellular
cytotoxicity reaction, neutrophils, macrophages, and killer K
cells participate with the preformed antibodies in destroying the
graft cells.
Treatment of corneal graft rejections appropriately includes
the administration of corticosteroids through all routes and
with sufficient dosage to reverse the rejection reaction.
Aggressive topical corticosteroid administration (e.g., 1%
prednisolone sodium phosphate hourly while awake and
dexamethasone phosphate ointment at bedtime) generally is
sufficient to reverse corneal transplant rejection reactions in
patients with reactions of mild to moderate severity (Figs 46.40
and 46.41), whereas this treatment combined with subconjunctival corticosteroid (dexamethasone sodium phosphate,
4 mg) and systemic corticosteroid (prednisone, 60 mg by
mouth each morning for 5 days with subsequent tapering
and discontinuation) often is required to treat a severe rejection
reaction effectively. In a small number of desperate cases
606
FIGURE 46.40. (a) Corneal transplant endothelial rejection with dramatic stromal swelling in the inferior half of the graft. (b) Same eye as shown
in (a), higher magnification, illustrating the classic endothelial rejection line.

involving one-eyed patients who had undergone multiple
corneal transplantations, I have employed intraocular
dexamethasone sodium phosphate (400 g) and systemic
therapy with immunosuppressive regimens (azathioprine,
2 mg1 kg1 day1, combined with cyclosporine, 5 mg1 kg1
day1) and in one case have salvaged a patients corneal graft
with the use of intravenous anti T-cell antibodies.
In the high-risk corneal graft recipient, HLA typing with
subsequent screening of the recipients serum for preformed
antibodies against candidate corneal grafts is appropriate, as is
obtaining a cornea from a donor who is as closely HLA matched
as possible. Preemptive treatment with solid organ immunosuppressive regimens, including combination prednisone,
cyclosporine, and azathioprine therapy, coupled with topical
corticosteroid and cyclosporine therapy, is employed in such highrisk recipients. The value of HLA matching for such patients
has been shown in a well-designed Dutch study.105 A multicentric trial to answer this question in the American population
concluded that HLA matching provided no significant
advantage for the high-risk corneal graft recipient but that
ABO matching did.106 A randomized study comparing topical
cyclosporine with placebo for such patients was discontinued
when interim data analysis disclosed that there was no
statistically significant benefit derived from topical cyclosporine
and therapy. The sponsoring pharmaceutical company,
regrettably, has declined to publish these negative results.
GRAFT-VERSUS-HOST DISEASE
CHAPTER 46
Graft-versus-host disease (GVHD), after bone marrow

transplantation, occurs because the donor T lymphocytes recognize the multiple differences (even in closely matched HLA
donors and recipients) among the various polymorphic minor
histocompatibility antigens on recipient tissues, and these
donor lymphocytes then attack the recipients cells. It is not
clear why, but the primary targets of this immunologic attack
are skin (Fig. 46.42), liver, intestine, oral mucosa, conjunctiva,
lacrimal gland, vaginal mucosa, and esophageal mucosa. Acute
GVHD develops in 3545% of bone marrow recipients; the
occurrence of GVHD has a profound influence on patients
survival rates; ~90% of patients who have little to no acute
GVHD survive, and only 45% of patients with moderate to
severe acute GVHD survive.107 The primary effector cells in the
affected tissue of patients with GVHD are T-lymphocytes,
specifically CD8 lymphocytes.
Multiple immunosuppressive chemotherapeutic regimens

have been tested for their efficacy in prevention of acute GVHD.
Combination preemptive therapy with methotrexate, cyclophosphamide, and cyclosporine has been the most thoroughly
studied regimen. Methotrexate typically is given immediately
after the bone marrow transplantation, and cyclosporine is
sometimes administered subsequently. At least 3 months of
therapy is required for a therapeutic effect.
Chronic GVHD develops 315 months after bone marrow
transplantation in ~45% of bone marrow recipients.108 In
addition to T-cell infiltration into the target tissues, complement and autoantibody deposition is found at the
dermalepidermal junction and in conjunctiva. Chronic GVHD
patients experience recurrent and sometimes fatal bacterial
infections. The disease may be preventable through the
aggressive use of prednisone and other immunosuppressants
after bone marrow transplantation. The most effective prevention of chronic GVHD is prevention of acute GVHD
because patients who do not experience acute GVHD have only
a 25% risk of developing chronic GVHD, whereas those who
experience acute GVHD have a 6080% probability of
developing chronic GVHD.104 Older bone marrow recipients
are at higher risk of developing late-onset, chronic GVHD, but
the form of the disease seen in these recipients generally is
less severe and more amenable to immunosuppressive
chemotherapy.
Treatment of chronic GVHD is with aggressive immunosuppressive chemotherapy, typically employing a polypharmacologic approach with systemic corticosteroids and one
or more immunosuppressants. The most commonly used
immunosuppressants with good effect have been azathioprine
and cyclosporine.108
The external immunologic ocular manifestations of GVHD
are extraordinary and ocularly devastating, producing profound
morbidity for patients suffering from this problem. The most
impressive ocular manifestation is KCS. The severity of the
KCS usually is extreme, and treating it can present a major
challenge (Fig. 46.43). The usual treatment modalities are
described in the section on KCS (see earlier), but increase in
aggressiveness of therapy generally occurs much more rapidly in
patients with GVHD. It is not uncommon for these patients to
have profound corneal epitheliopathy and to develop epithelial
erosions, persistence of these erosions, and subsequent stromal
ulceration. Permanent puncta occlusion and tarsorrhaphy
FIGURE 46.41. Same eye as shown in Figure 46.40, after 1 week of

intensive (hourly) topical 1% prednisolone sodium phosphate therapy
with partial resolution of the transplant rejection.
FIGURE 46.42. Typical skin manifestations of graft-versus-host

disease.
607

particularly the retina, nearly impossible at times. As with the
immunologic attack on the conjunctiva, the only effective
treatment of these manifestations is systemic treatment that
brings the other systemic manifestations of GVHD under
control.
SUMMARY
FIGURE 46.43. Profound corneal scarring and neovascularization as

a consequence of the severe keratoconjunctivitis sicca associated
with graft-versus-host disease.
should be performed very early in the course of the physicians

care of the GVHD patient who develops KCS.
An underrecognized ocular consequence of GVHD is the
immunologically mediated T-lymphocyte attack on the
conjunctiva, with resultant chronic conjunctivitis.9 This
immunologically driven inflammation responds only to
systemic therapeutic techniques that bring the systemic
manifestations of the GVHD under control (see previous
discussion). Adjunctive therapy that has had some ameliorating
effect has been the use of topical 2% cyclosporine drops four
times daily.
Other underrecognized consequences of GVHD are cataract,
uveitis, and retinitis. The basis of these ocular manifestations
is not clear, and they are typically unrecognized because of the
profound KCS that is blamed for the patients symptoms and
that makes adequate examination of the intraocular structures,
The eye can be affected by any of the immunologic

hypersensitivity reactions, and this chapter attempts to describe
the current state of knowledge regarding the immunologic
inflammatory diseases of the conjunctiva, cornea, and sclera.
Some of the diseases, such as Moorens ulcer and
phlyctenulosis, are strictly ocular. Others, such as scleritis
associated with long-standing rheumatoid arthritis, are ocular
manifestations of preexisting, previously diagnosed and treated
systemic disorders. A major point of emphasis in this chapter,
however, is the idea that some potentially lethal systemic
diseases may be silent or occult systemically but may produce
an ocular inflammatory lesion that is the first obvious clinical
manifestation of the disease. A second point of emphasis in this
chapter has been the idea that the onset of necrotizing scleritis
or of PUK in a patient with previously diagnosed systemic
disease indicates a distinct change in the character of the
underlying systemic disease: the vasculitic component to the
disease now should be foremost in the minds of the physicians
caring for such patients because failure to recognize this
important harbinger of necrotizing scleritis and PUK for
subsequent potentially fatal systemic lesions may place the
patient in jeopardy. Finally, a third major point of emphasis
is the notion that the physician who investigates a patient
thoroughly in an effort to understand the immunologic
mechanisms central to the ocular inflammatory lesion is the
physician who is best prepared to care for the patient. Similar
remarks can be made for most of the uveitic syndromes and the
diseases associated with vasculitis, but the material
encompassed in this chapter is restricted to the anterior
segment.
SECTION 6
REFERENCES
608
1. Greenbaum J, Cockcroft D, Hargreave FE,

Clin Immunol 1977; 59:437439.
2. Blumenthal M, Casale T, Dockhorn R, et al:
Efficacy and safety of nedocromil sodium
ophthalmic solution in the treatment of
seasonal allergic conjunctivitis. Am J
Ophthalmol 1992; 113:5663.
3. Spring TF: Reaction to hydrophilic lenses.
Med J Aust 1974; 1:499.
4. Allansmith MR, Korb DR, Greiner JV, et al:
Giant papillary conjunctivitis in contact lens
wearers. Am J Ophthalmol 1977;
83:697708.
5. Henriquez AS, Baird RS, Korb DR,
Allansmith MR: Histology of hard and soft
contact lens-associated giant papillary
conjunctivitis. Ann Ophthalmol 1980;
12:929933.
6. Fowler SA, Allansmith MR: Evolution of soft
contact lens coatings. Arch Ophthalmol
1980; 98:9599.
7. Arlt F: Physiologisch und pathologisch
anatomische Bemerkungen ber die
Bindehaut des Auges. Prager
Vierteljahrschrift 1846; 4:73.
8. von Graefe A: Klinische Vortrge ber
Augenheilkunde. Germany: Hirschberg;
1871:21.
9. Bahn AK, Fujikawa LS, Foster CS: T-cell

subsets and Langerhans cells in normal
and diseased conjunctiva. Am J
Ophthalmol 1982; 94:205212.
10. Allansmith MR: Vernal conjunctivitis. In:
Duane T, ed. Clinical ophthalmology. New
York: Harper & Row; 1978.
11. Beigleman MN: Vernal conjunctivitis. New
York: University of Southern California
Press; 1950.
12. Jones BR, Andrews BE, Henderson WG,
Schofield PB: The pattern of conjunctivitis
at Morefields during 1956. Trans
13. Dart JKG: The epidemiology of vernal
keratoconjunctivitis. Proceedings of The
Second Fisons International
Ophthalmology Workshop. Bollington,
Cheshire, UK: Pennine; 1989:2637.
14. Foster CS, Duncan J: A controlled clinical
trial of disodium cromoglycate therapy in
Ophthalmol 1980; 90:175181.
15. Foster CS: The sodium cromolyn
collaborative study group: evaluation of
topical cromolyn sodium in the treatment of
vernal keratoconjunctivitis. Ophthalmology
1988; 95:194201.
16. Hogan MJ: Atopic keratoconjunctivitis.
17. Coca AF, Cooke RA: On the classification

of the phenomenon of hypersensitiveness.
J Immunol 1923; 8:163.
18. Broude LS, Chandler JW: Atopic corneal
disease. Int Ophthalmol Clin 1984;
24:145.
19. Jay JL: Clinical features and diagnosis of
adult atopic keratoconjunctivitis and the
effect of treatment with sodium
cromoglycate. Br J Ophthalmol 1981;
65:335.
20. Garrity JA, Liesegang TJ: Ocular
complications of atopic dermatitis. Can J
21. Foster CS, Calonge M: Atopic
97:9921000.
22. Beutner EH, Chorzelski TP, Jordon RE:
Autosensitization in Pemphigus and
Bullous Pemphigoid. Springfield, IL:
Charles C Thomas; 1970:185.
23. Gammon WR, Merritt CC, Lewis DM,
et al: An in vitro model of immune
complex-mediated basement membrane
zone separation caused by pemphigoid
antibodies, leukocytes and complement.
J Invest Dermatol 1982; 78:285290.
24. Lever WF, Talbott JH: Pemphigus: a
historical study. Arch Dermatol Syph 1942;
46:800823.
25. Ahmed AR, Foster CS, Zaltas M, et al:

Association of DQw7 (DQb10301) with
ocular cicatricial pemphigoid. Proc Natl
Acad Sci USA 1991; 88:1157911582.
26. Wright P: Cicatrizing conjunctivitis. Trans
27. Hirst LW, Werblin T, Novak M, et al: Druginduced cicatrizing conjunctivitis simulating
ocular pemphigoid. Cornea 1982; 1:121.
28. Ahmed AR, Kahn KNA, Wells P, et al:
Preliminary serological studies comparing
immunofluorescence assay with
radioimmunoassay. Curr Eye Res 1989;
8:10111019.
29. Tyagi S, Bhol K, Natarajan K, et al: Ocular
cicatricial pemphigoid antigen: partial
sequence and characterization. Proc Soc
Natl Sci 1996; 93:1471414719.
30. Kumari S, Bhol KC, Simmons RK, et al:
Identification of ocular cicatricial
pemphigoid antibody binding site(s) in
human b4 integrin. Invest Ophthalmol Vis
Sci 2001; 42:379385.
31. Dutt JE, Ledoux D, Baer H, Foster CS:
Collagen abnormalities in conjunctiva of
patients with cicatricial pemphigoid. 1996;
Cornea 15:606611.
32. Rice BA, Foster CS: Immunopathology of
cicatricial pemphigoid affecting the
conjunctiva. Ophthalmology 1990; 97:
14761483.
33. Foster CS: Cicatricial pemphigoid [Thesis
for the American Ophthalmological Society].
Trans Am Ophthalmol Soc 1986; 84:
527663.
34. Tauber J, Jabbur N, Foster CS: Improved
detection of the disease progression in
ocular cicatricial pemphigoid. Cornea 1992;
11:446451.
35. Lemp MA: The mucin-deficient dry eye. Int
37. Shore JW, Foster CS, Westfall CT, Rubin
PAD: Results of buccal mucosal grafting for
patients with medically controlled cicatricial
pemphigoid. Ophthalmology 1992; 71:417.
38. Nguyen VT, Ndoye A, Shultz LD, et al:
Antibodies against keratinocyte antigens
other than desmogleins 1 and 3 can induce
pemphigus vulgaris-like lesions. J Clin
Invest 2000; 106:14671479.
39. Bean SF, Holubar K, Gillet RB: Pemphigus
involving the eyes. Arch Dermatol 1975;
111:1484.
40. Buhac J, Bhol K, Padilla T, et al:
Coexistence of pemphigus vulgaris and
ocular cicatricial pemphigoid. J Am Acad
Dermatol 1996; 34:884886.
41. Foster CS: The eye in skin and mucous
membrane disorders. In: Tasman W, Jaeger
EA, eds. Duanes clinical ophthalmology.
Philadelphia: JB Lippincott; 1995:141.
42. Bowman W: The parts concerned in the
operations of the eye (1849), cited by
Nettleship E: chronic serpiginous ulcer of
the cornea (Moorens ulcer). Trans
43. McKenzie H: Diseases of the eye. London,
Longman, Rees, Orme, Brown & Green,
1854:631.
44. Mooren A: Ulcus Rodens.
Ophthalmiatrische Beobachtungen. Berlin:
A Hirschwald; 1867:107110.
45. Wood TO, Kaufman HE: Moorens ulcer. Am
46. Soukiasian SH, Foster CS: Moorens ulcer:

diagnostic problems in clinical
ophthalmology. Philadelphia: WB Saunders;
1994:220227.
47. Chow CYC, Foster CS: Moorens ulcer. Int
48. Keitzman B: Moorens ulcer in Nigeria. Am J
49. Schazlin D: Moorens ulceration. In: Smolin
G, Thoft R, eds. The cornea. Boston: Little,
Brown; 1987:321327.
50. Schaap OL, Feltkamp TEW, Breebaart AC:
Circulating antibodies to corneal tissue in a
patient suffering from Moorens ulcer. Clin
Exp Immunol 1969; 5:365370.
51. Berkowitz PJ, Arentsen JJ, Felberg NT,
Laibson PR: Presence of circulating
immune complexes in patients with
peripheral corneal disease. Arch
52. Brown SI, Mondino BJ: Therapy of
Moorens ulcer. Am J Ophthalmol 1984;
98:16.
53. Foster CS: Systemic immunosuppressive
ulcer. Ophthalmology 1985; 92:1436.
54. Gottsch J, Liu S, Minkovitz J, Goodman D,
et al: Autoimmunity to a cornea-associated
stromal antigen in patients with Moorens
ulcer. Int Ophthalmol Vis Sci 36:15411547,
1995.
55. Wakefield D, Robinson LP: Cyclosporine
therapy in Moorens ulcer. Br J Ophthalmol
1987; 71:415417.
56. King JH: Destructive marginal ulceration: a
saga of surgical therapy. Trans Am
57. Foster CS, Kenyon KR, Griner J, et al: The
immunopathology of Moorens ulcer. Am J
Ophthalmol 1979; 88:149159.
58. Gilbard J, Rossi SR, Heyda KG: Ophthalmic
solutions, the ocular surface, and a unique
therapeutic artificial tear formulation. Am J
Ophthalmol 1989; 107:348355.
59. Foster CS, Sainz de la Maza M: Sclera.
Berlin: Springer; 1994.
60. Lyons CJ, Hakin KN, Watson PG: Topical
flurbiprophen: an effective treatment for
episcleritis? Eye 1990; 4:521525.
61. Watson PG, Hayreh SS: Scleritis and
episcleritis. Br J Ophthalmol 1976;
60:163191.
62. McGavin DDM, Williamson J, Forrester JV,
et al: Episcleritis and scleritis: a study of
their clinical manifestations and association
with rheumatoid arthritis. Br J Ophthalmol
1976; 60:192226.
developing necrotizing scleritis or peripheral
ulcerative keratitis: effects of systemic
immunosuppression. Ophthalmology 1984;
91:12531263.
64. Watson PG, Hazelman BL: The sclera and
systemic disease. Philadelphia: WB
Saunders; 1976.
65. Sainz de la Maza M, Foster CS, Jabbur NS:
Scleritis-associated uveitis. Ophthalmology
104:5863, 1996.
66. Grayson M: Marginal furrows: a
characteristic corneal lesion of rheumatoid
arthritis. Arch Ophthalmol 1968; 79:563.
67. Smith RE, Schanzlin DJ: Rheumatoid
diseases. In: Smolin G, Thoft R, eds. The
cornea. Boston: Little, Brown; 1987:327344.
68. Cohen KL: Sterile corneal perforation after
cataract surgery in Sjgrens syndrome.
69. Gelender H: Descemetocele after

intraocular lens implantation. Arch
70. Insler MS, Boutros G, Boulware DW:
Corneal ulceration following cataract
surgery in patients with rheumatoid arthritis.
J Am Intraocul Implant Soc 1985; 11:594.
71. Sainz de la Maza M, Foster CS:
Necrotizing scleritis after ocular surgery:
a clinicopathologic study. Ophthalmology
1991; 98:17201726.
72. Pillat A: ber das Vorkommen von Choroiditis
bei Lupus Erythematodes. Graefes Arch Clin
73. Gold DH, Morris DA, Henkind P: Ocular
findings in systemic lupus erythematosus.
74. Spaeth GL: Corneal staining in systemic
lupus erythematosus. N Engl J Med 1967;
276:1168.
75. Reeves JA: Keratopathy associated with
systemic lupus erythematosus. Arch
76. Halmay O, Ludwig K: Bilateral band-shaped
deep keratitis and iridocyclitis in systemic
lupus erythematosus. Br J Ophthalmol
1964; 48:558.
77. Wise GN: Ocular periarteritis nodosa. Arch
78. Fronert PP, Scheps FG: Long-term followup studies of periarteritis nodosa. Am J
Med 1967; 43:8.
79. Leib ES, Restivo C, Paulus HE:
Immunosuppressive and corticosteroid
therapy of polyarteritis nodosa. Am J Med
1979; 67:941.
80. Fauci AS, Wolff SM: Wegeners
granulomatosis: studies in 18 patients and
review of the literature. Medicine 1973;
52:535.
81. Soukasian SH, Foster CS, Niles JL,
Raizman MB: Diagnostic value of
antineutrophil cytoplasmic antibodies in
scleritis associated with Wegeners
granulomatosis. Ophthalmology 1992;
99:125132.
82. Ludemann G, Gross WL: Autoantibodies
against cytoplasmic structures of neutrophil
granulocytes in Wegeners granulomatosis.
Clin Exp Immunol 1987; 6:350.
83. Savage CS, Winearls CG, Jones S, et al:
Prospective study of radioimmunoassay for
antibodies against neutrophil cytoplasm in
diagnosis of systemic vasculitis. Lancet
1987; 1:1389.
84. Niles JL, McCloskay RT, Ahmed MF, et al:
Wegeners granulomatosis autoantibody is
a novel antineutrophil serine protease.
Blood 1989; 74:1888.
85. Ahmed M. Niffenegger JH, Jakobiec FA,
et al: Diagnosis of limited ophtalmic
Wegeners granulomatosis: distinctive
pathologic features with ANCA test
confirmation. Int Ophthalmol 2007; Jun 23;
[epub ahead of print].
86. Fauci AS, Haynes BF, Katz P, Wolff SM:
Wegeners granulomatosis: prospective
clinical and therapeutic experience with
85 patients over 21 years. Ann Intern Med
1983; 98:76.
87. Hoang Xuan T, Foster CS, Rice BA: Scleritis
in relapsing polychondritis. Ophthalmology
1990; 97:892898.
88. Kirkham TH: Scleroderma in Sjgrens
syndrome. Br J Ophthalmol 1969;
53:131.
89. Stucci CA, Geiser JD: Manifestations
oculares de la sclerodermie generalise.
CHAPTER 46
609
90.
91.
92.
93.
94.
95.
SECTION 6
96.
610
(Points communs avec le syndrome de

Sjgren). Doc Ophthalmol 1967; 22:71.
Horan EC: Ophthalmic manifestations of
progressive systemic sclerosis. Br J
Stevens AM, Johnson FC: A new eruptive
fever associated with stomatitis and
ophthalmia: Report of two cases in
children. Am J Dis Child 1922; 24:526533.
Bean SF, Quezada RK: Recurrent oral
erythema multiforme: clinical experience
with 11 patients. JAMA 1983;
249:28102812.
Tripathi A, Ditto AM, Grammer LC, et al:
Corticosteroid therapy in an additional 13
cases of Stevens-Johnson syndrome: a
total series of 67 cases. Allergy Asthma
Proc 2000; 21:101105.
Hynes AY, Kafkala C, Daoud YJ, Foster CS:
Controversy in the use of high-dose
systemic steroids in the acute care of
patients with Stevens-Johnson syndrome.
Int Ophthalmol Clin 2005; 45:2548.
Foster CS, Fong LP, Azar D, Kenyon KR:
Episodic conjunctival inflammation after
Stevens-Johnson syndrome.
Lyell A: Toxic epidermal necrolysis: a
reappraisal. Br J Ophthalmol 1979; 100:69.
97. Westly ED, Wechsler HL: Toxic epidermal

necrolysis. Arch Dermatol 1984; 120:721.
98. Thygeson P: Observations on nontuberculous phlyctenular
keratoconjunctivitis. Trans Am Acad
99. Medawar PB: Immunity to homologous
transplanted to brain, to subcutaneous
tissue and to the anterior chamber of the
100. Wetzig RP, Foster CS, Greene MI: Ocular
immune responses. I. Priming of A/J mice
in the anterior chamber with
azobenzenearsonate-derivatized cells
induces second-order-like suppressor
T cells. J Immunol 1982; 128:17531757.
101. Foster CS, Wetzig RP: Immune reactions
in the eye. Surv Immunol Res 1982;
1:93108.
102. Foster CS, Monroe JG, Campbell R, et al:
Ocular immune responses. II. Priming of A/J
mice in the vitreous induces either
enhancement of or suppression of
subsequent hapten-specific DTH responses.
J Immunol 1985; 136:27872791.
103. Streilein JW, Niederkorn JY, Shadduck JA:
Systemic immune unresponsiveness
induced in adult mice by anterior chamber
104.
105.
106.
107.
108.
presentation of minor histocompatibility

antigens. J Exp Med 1980; 152:1121.
Kaplan HJ, Streilein JW, Stevens TR:
Transplantation immunology of the anterior
chamber of the eye. II. Immune response to
allogeneic cells. J Immunol 1973; 115:805.
Koch-van-Alphen CC, Volker-Dieben HJ,
DAmaro J: Results of HLA typing in
corneal transplantation. Fortschr
Storb R, Prentice RL, Buckner CD, et al:
Graft-versus-host disease and survival in
patients with aplastic anemia treated by
marrow grafts from HLA-identical
siblings: Beneficial effect of a protective
environment. N Engl J Med 1983;
308:302.
Storb R, Prentice RL, Sullivan KM, et al:
Predictive factors in chronic graft-versushost disease in patients with aplastic
anemia treated by marrow transplantation
from HLA-identical siblings. Ann Intern Med
1983; 98:461.
Sullivan KM, Shulman HM, Storb R, et al:
Chronic graft-versus-host disease in 52
patients: adverse natural course and
successful treatment with combination
immunosuppression. Blood 1981;
57:267276.
CHAPTER
47
Allergic and Toxic Reactions: The Immune

Response
Mark B. Abelson, Gail L. Torkildsen, and Ira J. Udell
Key Features
Ocular allergies affect an estimated 20% of the population in

industrialized countries
Types of ocular allergies include:
Allergic conjunctivitis (seasonal and perennial) most
common, mild, caused by environmental allergens such as
ragweed, tree pollen, animal dander, and dust mites.
Characterized by ocular itching, hyperemia, chemosis,
eyelid edema, and tearing. Treatments include allergen
avoidance, cold compresses, antihistamines, mast cell
stabilizers, combination antihistamine/mast cell stabilizers,
NSAIDs, and corticosteroids.
Atopic keratoconjunctivitis rare, serious, chronic
condition seen in 2440% of patients with atopic
dermatitis. Characterized by itching, redness, photophobia,
keratopathy, corneal ulcers, keratoconus, anterior polar
cataracts, mucous discharge, atopic blepharitis.
Treatments include topical corticosteroids and mast cell
stabilizers.
Vernal keratoconjunctivitis rare, serious, usually seen in
warm climates in males ages 320 years with family history
of atopy. Characterized by ptosis, ropy mucous discharge,
photophobia, large, nonuniform cobblestone papillae,
HornerTrantas dots, limbal nodules, neovascularization,
corneal shield ulcers, and itching. Treatments include
allergen avoidance, cold compresses, antihistamines,
corticosteroids, and mast cell stabilizers.
Giant Papillary Conjunctivitis is not a true allergic reaction but
an inflammatory reaction of the upper tarsal conjunctiva
associated with the presence of contact lenses, surgical suture
barbs, and ocular prostheses.
Toxic keratoconjunctivitis is an ocular toxic reaction due to use
of certain drugs, vehicles, and preservatives.
THE ALLERGIC RESPONSE

Ocular allergies can range from mild (as in seasonal and
perennial allergic conjunctivitis (SAC and PAC)) to severe and
vision threatening, as in atopic and vernal keratoconjunctivitis
(AKC and VKC). Allergic diseases affect an estimated 20% of the
population in developed countries worldwide, including an
estimated 22 million people in the United States.1 Data from
epidemiological studies in a number of countries suggest that
the prevalence of allergic diseases has increased substantially
since the 1940s.2 This rising prevalence of asthma and allergic
diseases has not been denitively explained, but researchers
have proposed causes including increasing air pollution and
diesel exhaust, genetics, and the so-called hygiene hypothesis,
which proposes that atopic diseases are prevented by infections
in early childhood.
T-cells play an important role in the allergic response. Upon

activation, naive T-helper (Th0) cells differentiate into Th1 or
Th2 cells. Th1 cells secrete interferon (IFN)-g, tumor necrosis
factor (TNF), and lymphotoxin, and are associated with cellmediated immunity. Th2 cells secrete interleukin (IL)-4, IL-5,
and IL-13. Th2 cells are important for immunity and resistance
to parasitic infection and are associated with allergy and asthma.
The Th1/Th2 balance, which determines the type of immune
response the body will mount in response to a given allergen,
forms the basis of the hygiene hypothesis. As allergen
sensitivity develops, the balance in this specialization shifts
toward higher Th2 and lower Th1 levels. According to the
hygiene hypothesis, a lack of triggers for Th1-type immune
response, such as exposure to infections, endotoxins, and dirt
in childhood, would result in a preponderance of Th2-type
immune responses responsible for allergic disease.
The primary factor that influences this differentiation is the
presence of certain cytokines at the time of T cell activation.
Cells exposed to IL-12 produce IFN-g and become Th1 cells,
whereas T cells exposed to IL-4, a product of other CD4+ T
cells and mast cells, tend to become Th2 cells.3 In vitro studies
show that Th1- and Th2-type immune responses downregulate
each other when activated. For example, typical Th2-type
cytokines like IL-4 and IL-10 inhibit the production of Th1derived cytokines such as IFN-g, and vice versa.4
Genetics also play a role in the predisposition to allergic
diseases. The risk for atopic disease is doubled in children who
have one parent with a history of atopy, and it is more than 50%
if both parents have such history. Although the specic type of
allergy expressed by individuals may differ within a family, the
incidence of allergic disorders is approximately three times
higher in atopic families than in nonatopic families.5 Furthermore, children who have both a maternal and a paternal family
history of atopy generally manifest allergy before puberty.6
Several genes are suspected to be associated with atopy, such as
5q3133. It appears that children do not inherit the allergic
disorder per se, but instead an allergic predisposition that can
be contributed equally by both parents, suggesting that this is
an autosomally carried trait.6 Nongenetic risk factors include
small family size, higher socioeconomic status, use of
antibiotics, and residing in an urban environment.
THE ROLE OF MAST CELLS IN THE ALLERGIC

RESPONSE
The mast cell mediates type I (immediate) hypersensitivity
reactions. Mast cells are characterized as containing either
tryptase (T) or tryptase/chymase (TC) based on immunohistochemical staining of these endoproteases. Both types
of mast cells develop from the same CD34+ mononuclear
611

precursor. Normally, the conjunctival mast cells are mostly the
TC subtype.7 Mast cells of the T subtype are increased in
epithelial and subepithelial layers in SAC and PAC and much
more in VKC. In AKC, the TC subtype predominates which
may be responsible for brosis.8 Mast cell heterogeneity can be
seen across species, which means that drawing conclusions
based on animal data must be done cautiously.9 In normal
human patients, their distribution is limited to the substantia
propria of the conjunctiva, whereas in patients with VKC,
mast cells are also found in the conjunctival epithelium.10
Approximately 50 million mast cells can be found in the ocular
and adnexal tissues of the human eye.
The allergic cascade begins when antigen binds and crosslinks with two immunoglobulin (Ig)-E receptors located on the
surface of conjunctival mast cells. The cell membrane surface of
a mast cell has as many as 500 000 IgE receptors, 10% of which
are occupied in vivo.11 In GPC, 30% of mast cells are degranulated, and in VKC up to 80% appear to be degranulated.12 IgE
molecules, the major homocytotropic antibodies, may adhere to
these surface receptors. The allergenIgE antibodymast cell
union results in the activation of a serine esterase, initiating a
change in the Fc portion of the IgE molecule, which is attached
to the mast cell membrane.13 This event leads to an intracellular biochemical cascade resulting in mast cell degranulation and the release of histamine, eosinophil cationic protein
(ECP), high molecular weight neutrophil chemotactic factor, and
platelet activating factor (PAF). These agents attract eosinophils
and neutrophils (cells that contain secondary mediators), that
then restore homeostasis or produce tissue alterations in
chronic allergic disease. The signs and symptoms of an acute
allergic reaction are the result of this intricate network of
mediator interaction (Fig. 47.1).
MAST CELL AND EOSINOPHIL MEDIATORS

Mediators have been categorized into three different groups
based on their mode of action. The rst group includes substances such as histamine and the prostaglandins that mediate
their actions by binding to a specic cell membrane receptor.
The second group comprises substances of cell or plasma origin
that directly damage tissues, including eosinophil major basic

protein (EMBP) and complement. The third group consists of
chemotactic factors that attract cells such as eosinophils and
macrophages to the inflammatory site and includes the
arachidonic acid metabolites.14
Histamine
Histamine, stored in granules of mast cells and basophils, is the
central mediator of ocular allergy and inflammation. The
conjunctiva has at least two histamine receptors, H1 and H2.
Selective H1-receptor activation results mainly in itching,15
while selective H2-receptor activation primarily elicits redness.16
Thus, it appears that H1 and H2 receptors may be associated
with neuronal tissue and vascular tissue, respectively. Instillation
of histamine into the eye reproduces the exact clinical picture
of acute allergic conjunctivitis in a dose-dependent fashion:
itching, redness, chemosis, tearing, and eyelid swelling.17
Histamine levels were not found to be consistently elevated
in patients with allergic conjunctivitis; however, they were
found to be elevated in tear samples from patients with VKC
(16 ng/mL) (normal, 5ng/mL).18 One study examined the
presence of histaminase activity in human tears after in vivo
conjunctival allergen challenge. Histaminase inactivation
resulted in a 15-fold elevation of histamine recovery.19 These
results demonstrate the presence of histaminase activity in
human tears and suggest that histaminase activity may have
confounded the role of histamine in ocular allergic disorders
other than VKC. The nding of insufcient histaminase
activity in tears of patients with VKC may play some as yet
unknown role in the cause of the disease.20
Eosinophils
Eosinophils have a dual role in allergic disease: either spurring
inflammation or acting to quiet it. EMBP causes mast cell
degranulation and corneal damage. ECP and eosinophil derived
neurotoxin (EDN) can also cause corneal damage. Other
eosinophil mediators modulate the mast cell response through
negative feedback: histaminase inactivates histamine, phospholipase inactivates PAF, and aryl sulfatase inactivates certain
leukotrienes.13 PAF is a potent eosinophil and neutrophil
FIGURE 47.1. Illustration of the allergic

response.
Histamine
Endothelial cells
Fibroblasts
SECTION 6
IL-3
IL-4
IL-4
IL-5
B cell
E-selectin ICAM
VCAM
Blood vessel
612
Basophil
APC
Eosinophil
IL-3, IL-4
IL-10
Th2 cell
IgE
Mast cell
IL-3, IL-4,
IL-6, IL-8,
IL-13, TNFa,
Histamine
Adapted from McGill, J I et al. Br J Ophthalmol 1998;82:1203-1214
Histamine
Nerve
Allergic and Toxic Reactions: The Immune Response
ALLERGIC CONJUNCTIVITIS
It is estimated that 90% of cases of ocular allergies are in the
form of seasonal and perennial allergic conjunctivitis (SAC and
PAC).24 While allergic conjunctivitis is a fairly mild disease, it
can result in considerable costs in terms of lost productivity and
decreases in quality of life. Many allergic patients report sleep
disruption, daytime fatigue, learning impairment, decreased
cognitive function, and decreased productivity. The economic
burden is substantial as well, with an estimated 3.5 million lost
workdays, 2 million missed school days, and 28 million days of
restricted activity or productivity loss per year in the US due to
allergies.25
SAC is a recurrent condition caused by airborne allergens, such
as pollens from ragweed, grasses, or trees. There are regional
variations throughout the United States and most fluctuate
seasonally. PAC results from allergens that exist year round,
including dust mites, animal dander, or mold. Whether seasonal
or perennial, allergens come into contact with the tear lm and
then traverse the conjunctiva to join IgE antibodies attached to
conjunctival mast cells. This allergenIgE antibodymast cell
union results in mast cell degranulation, release of chemical
mediators, and the manifestation of ocular allergic signs and symptoms: itching, redness, chemosis, tearing, and eyelid swelling.
In the mild-to-moderate forms, allergic conjunctivitis can be
described as a series of acute type I hypersensitivity reactions
including mast cell degranulation and the release of preformed
mediators, which are responsible for the signs and symptoms
presented by the patient. The mast cells also releases enzymes,
such as histaminase, that act as control mechanisms to shut
off the release of mediators so that the overall effect is discrete
and self-limiting. If the allergic stimulation continues, additional
mast cell degranulation, cellular inltrates, and inflammation
results. These are the late-phase allergic reactions, as opposed to
the early-phase response observed with the milder forms.
CLINICAL FEATURES AND DIAGNOSIS

Itching is the hallmark symptom of ocular allergy. Its absence
makes the diagnosis suspect. Symptoms of redness, tearing,
burning, and sinus pressure behind the eyes and ears can also
be present. Allergic conjunctivitis is an intermittent condition
that may not be manifested at the time of the ophthalmic
examination. It is important to inquire specically about environmental triggers such as cats, trees, grasses, ragweed, dust
and molds, as well as provoking factors like lawn mowing, pet
exposure, or camping. Patients who take anti-allergic medications systemically or nasally need to be asked specically about
eye symptoms, since it is estimated that 90% of patients with
allergic rhinitis have ocular symptoms as well.26
Sometimes, a chart review will reveal that ofce visits occur
during the same month every year. The month or season that
the patient experiences symptoms is important to diagnosing
their allergic trigger. The patient is usually an excellent source
for identifying the allergen to which she or he is sensitive.
For example, in the north-eastern United States, March is
commonly tree pollen season, May is grass season, and midAugust through mid-September is ragweed season. Additionally,
a personal or family history, either current or in childhood, of
eczema, asthma, rhinitis, or other atopic history may be
indicative of ocular allergy. When questioned, patients may
deny asthma but respond positively to questions related to
wheezing in cold air or upon exertion.
Perennial ocular allergens such as dust, pet hair and dander,
mites or mold can present a constant aggravation. If a patient
experiences intense itching, redness, and discomfort while
indoors, and during times of the year inconsistent with seasonal
allergens, then perennial ocular allergies may be involved. In
one investigation, the dust collected from nearly all of the
homes sampled contained pet allergens, even though less than
half of the homes had pets.27
Since clinical signs may be absent at the time of an ofce
visit, gathering a medication history of allergy medications,
inhalers, nasal sprays, and over-the-counter eye drops is
important. These medications may only be used seasonally or
as needed. A patient may be symptomatic, but the adnexal
tissue may appear normal. Most patients, however, have some
microchemosis and dilatation of conjunctival vessels, or
swelling of the lids (Fig. 47.2). Microchemosis is evident only
with careful slit-lamp examination. With the slit beam
narrowed, a slight elevation of the conjunctiva can be seen. An
additional clue is to notice the conjunctival redness prior to
dilation and compare it to the conjunctiva after dilation. Venous
CHAPTER 47
chemotactic factor21 and an inflammatory mediator that modulates vascular permeability. PAF has been found in basophils,
mast cells, eosinophils, monocytes, polymorphonuclear leukocytes, and macrophages. Arachidonic acid is broken down by
cyclooxygenase into prostaglandins and thromboxanes, which
produce itching and conjunctival redness. Leukotrienes are produced from the breakdown of arachidonic acid by lipoxygenase,
which act by recruiting macrophages.
Eosinophil inltration is observed in both VKC and contact
lens-associated giant papillary conjunctivitis (GPC), suggesting
that the tissue damage seen in these diseases may be, in part,
the result of eosinophilic mediators.10 Patients with VKC have
increased tear levels of both major basic protein and Charcot
Leyden crystal protein, with the magnitude of the increase
correlating with the severity of the disease.22 EMBP, the highly
toxic substance released by eosinophils, is thought to be linked
to the corneal damage in VKC, that is, keratitis and shield ulcer
formation.22
Signicant major basic protein deposition in conjunctival
specimens from patients with VKC, corneal vernal plaque
specimens, and GPC has been seen with no correlation between
the intensity of major basic protein deposition and the severity
of disease.23 Thus, in both VKC and GPC, the release of major
basic protein from eosinophils can contribute to the sustained
mast cell degranulation that occurs, leading to a more severe and
long-lasting process than that found in acute allergic conjunctivitis.
FIGURE 47.2. Classic acute allergic conjunctivitis. Conjunctival

hyperemia is evident as is chemosis. Hyperemia and swelling of the
lids may or may not be present on examination, but patients will often
associate their exacerbation with knuckle rubbing. The characteristic
symptom reported by patients is itching.
613
SECTION 6
614
congestion can lead to the allergic shiners often seen in

children. Nasal symptoms are so common that sometimes just
listening to the sniffles points a clinician in the right direction.
Watching for the allergic salute (nasal rubbing) while taking a
history is also important.
An acute reaction may yield a clear or white exudate, whereas
a chronic reaction is characterized by a mucopurulent, thicker,
stringier exudate. Pallor of the palpebral conjunctiva may occur
as a result of edema. A papillary vasodilatory reaction may
occur with the absence of giant papillae. Excessive conjunctival
chemosis can lead to corneal dellen; however, the limbus
and cornea are usually normal. The presence of purulent
discharge, follicles, cobbles, or keratitis is inconsistent with a
sole diagnosis of allergic conjunctivitis. A complaint of ocular
itching that, with further questioning, specically involves the
lid and lid margin may suggest a meibomian gland dysfunction
or allergic blepharodermatitis.
The ocular allergic reactions elicited by animal dander, dust
mites, and molds are similar to those seen in seasonal allergic
conjunctivitis. The conjunctival and periorbital lid swelling can
be impressive and the conjunctiva may actually balloon beyond
the lids.
The differentiation of allergic conjunctivitis from dry eye
syndrome hinges on the presence or absence of the distinguishing characteristic of allergy, which is ocular itching. Both
conditions may produce mild conjunctival vasodilatation and
a burning sensation with intermittent exacerbations. Not
infrequently, these diseases occur concurrently. Patients with
dry eye may be more susceptible to allergic conjunctivitis owing
to decreased tear lm production and a decreased ability to
wash away and dilute airborne allergens, thus acting as a barrier
to the adherence of allergens to the conjunctival surface.
Corneal and conjunctival staining, tear meniscus level assessment, Schirmers testing, and tear break-up time evaluations
may aid in properly categorizing dry eye. Ocular allergic
symptoms rarely include foreign body sensation, although this
is common in dry eye. Another distinguishing factor may be the
type of mucus present. In allergy, it tends to be thin and clear,
rather than the stringy white mucus associated with dry eye.
If the diagnosis is in doubt, a conjunctival scraping positive
for eosinophils is certainly indicative of allergy. However,
because eosinophils may exist more deeply in the conjunctiva,
a negative scraping is inconclusive.28 High levels of total and
specic serum and tear levels of specic IgE have a strong correlation with ocular allergy. Tear cytology positive for eosinophils
would suggest that an IgE-mediated response is present.
Neutrophils may also be seen. A positive skin test for specic
allergens has been shown to be between 71 and 87% predictive for
positive ocular reactivity.29,30 Finally, a positive ocular challenge
to the allergen is most decisive for a positive diagnosis,30,31 but
may be impractical unless all other tests are inconclusive.
HISTOPATHOLOGY AND PATHOGENESIS

Most animal work using models of ocular anaphylaxis
corroborate the clinical picture of mild to moderate ocular
allergic reactions, with 70100% of mast cells degranulated,32
no effect on goblet cells,33 and signicant early-phase increase
in neutrophils.34 Mast cells have been shown to regranulate
within 24 h.35 Macroscopically, eyes can also appear normal,
despite degranulated mast cells and an increase in microscopic
interstitial edema.36
The conjunctiva is capable of mounting immediate immune
responses to external insult. There is strong evidence of a latephase reaction as well.37 In an allergen challenge study, the
numbers of neutrophils, eosinophils, lymphocytes, and monocytes found in conjunctival scrapings were quantied and
correlated with the clinical prole, total serum IgE, and serum
IgE to rye I antigen. Signicant increases in neutrophils of
patients occurred after 20 min (P <.001), and in eosinophils at
6 h (P <.005), compared with values of control subjects. Thus,
signicant inflammatory changes in conjunctival scrapings are
present long after allergen exposure has ended.
The late phase of allergic conjunctivitis is manifested by
either a sustained or discrete second peak of allergic response
424 h after allergen exposure.38 Tear cytology has also shown
increases in eosinophils or neutrophils in allergen-challenged
human eyes long after the immediate reaction had disappeared.39
A recent study40 using a modied Conjunctival Allergen
Challenge (CAC) design30 showed that some individuals who
were sensitive to low doses of allergen had a late-phase reaction.
When subjects were given an allergen challenge and then
rechallenged 24 h later, these individuals showed heightened
ocular itching and redness. An anti-inflammatory agent
(corticosteroid) was shown to inhibit signs and symptoms of the
late phase component. This modied CAC model will further
elucidate the late phase phenomenon and may develop into a
tool for assessing other antiinflammatory agents.
SACs and PACs are type 1 hypersensitivity reactions. Allergens
penetrate the conjunctival epithelium and bind to IgE receptors
on mast cells. This leads to mast cell degranulation and release
of chemical mediators including histamine. Histamine causes
itching, increases in vascular permeability and recruitment of
immune cells. Preformed mediators include histamine, tryptase,
and heparin. Mediators formed in response to allergen binding
include prostaglandins (from membrane-bound arachidonic acid),
thromboxanes, leukotrienes, PAF, cytokines, chemokines, and
growth factors. Histamine, bradykinin, and prostaglandins
stimulate pain and itching.
TREATMENT
No patient should suffer unnecessarily, and treating ocular
allergies represents a tremendous opportunity to improve the
quality of patients lives. The rst and foremost step in the treatment of allergic conjunctivitis is the removal of the offending
allergen, if possible. The severity of the allergic condition is
directly proportional to the level and duration of exposure to the
allergen. Depending on the allergen to which the patient is
sensitized, limiting time spent outdoors, using air conditioners,
or avoiding animal dander will all dramatically improve the
condition. Tear substitutes can be used to dilute the allergen
and wash some allergen away.
In the short term, using over-the-counter H1 antihistamine
vasoconstrictor combinations (as needed) can successfully
alleviate the primary symptom, itching, and the primary signs,
redness and swelling. A study of the effects of antazoline
phosphate 0.5% in combination with naphazoline hydrochloride
0.05% in the allergen challenge model of allergic conjunctivitis
showed signicant relief of itching, redness, chemosis, and lid
swelling immediately and 2 h after administration of the drug.31
However, these agents are short-acting and may lead to rebound
vasodilation.
A more effective therapeutic option for allergic conjunctivitis
is a combination antihistamine plus mast cell stabilizer, such as
olopatadine, epinastine, ketotifen, or azelastine. Olopatadine
0.1% is the most frequently prescribed combination antihistaminic agent with mast cell stabilizing properties. It attaches to
the H1 receptor site to prevent histamine from binding and
therefore provides initial relief of itching. Owing to its mast cell
stabilization, IgE cross-linking is prevented, histamine release is
halted, and further mast cell degranulation is inhibited. In this
way, olopatadine provides both quick and long-lasting relief
from the signs and symptoms of allergic conjunctivitis. The
ATOPIC KERATOCONJUNCTIVITIS
Atopic keratoconjunctivitis (AKC) represents the ocular
manifestation of atopy, a hereditary condition characterized by
FIGURE 47.3. Atopic keratoconjunctivitis. Note the slight corneal

haze, the conjunctival changes, and the accompanying blepharitis
marked by structural changes at the lid margin and loss of lashes.
eczema, rhinitis, asthma, and atopic dermatitis. The term

atopy was originally paraphrased by Coca and Cooke52 in 1923
from its Greek equivalent meaning out of the ordinary.52 In
1953, Hogan was credited for rst describing AKC as a distinct
entity, characterized by severe, chronic external ocular inflammation associated with atopic dermatitis.53 Although atopy is
common, AKC is rare. Atopic dermatitis is present in ~3%
of the general population.54 Ocular involvement in atopic
dermatitis is estimated to occur in 2540% of patients.55,56 AKC
is characterized by atopic dermatitis of the eyelids as well as by
papillary conjunctivitis, disruption of the corneal epithelium,
and in severe cases, conjunctival and corneal scarring (Fig. 47.3).
Treatment may need to be administered for years, and patients
commonly develop complications from both the disease and the
therapeutic interventions.

AKC can appear in childhood and continue for 45 decades, at
which time the disease can spontaneously resolve.54 The peak
incidence is reported to occur between the ages of 30 and
50 years57 with an age range between 9 and 76 years.58 The
symptoms of AKC last throughout the year and are almost
always bilateral.54 Patients regularly complain of moderate to
severe itching, and may also describe a burning sensation,
photophobia, and blurring of vision. Tearing is frequently noted
and can be accompanied by a distinct mucous discharge that is
mucopurulent, thick, ropy, and white.54
Gross examination of the ocular adnexae may reveal
indurated lid margins that are often thickened and scaly, and
there may be secondary blepharitis. Exudative, vesicular, or
crusted lesions may be observed elsewhere from atopic
dermatitis. Maceration of the inner or outer canthi may be
observed,59 and punctal stenosis can occur.57 An additional lid
fold (Dennies line) may be present.
The conjunctiva may be pale in comparison with that seen in
the other allergic disorders;54,58 however, limbal hyperemia and
chemosis can be seen with exacerbations of the disease.
Papillary hypertrophy is prominent in the inferior palpebral
conjunctiva, but not exclusively so. In contrast, VKC and GPC
tend to have papillae in the superior tarsus.54,58 Conjunctival
scarring is frequently a serious consequence of AKC, and
shrinkage of the inferior fornix may result. The characteristic
increase in tearing associated with AKC may be a result of the
loss of the inferior cul-de-sac. Subepithelial brosis, forniceal
CHAPTER 47
olopatadine molecule is the most researched of all topical ocular

anti-allergic agents, and has been shown to be the most effective
treatment for ocular allergy.41 A new formulation, olopatadine
0.2%, has been introduced as the rst and only available
antiallergic agent indicated for once-daily dosing. This product
has demonstrated enhanced efcacy and extended duration of
action (up to 24 h) (Vogelson, Abelson), while still remaining
safe and well-tolerated in both adults and children (Lichtenstein).
The once-a-day formulation will provide increased convenience
and compliance for patients.
The classic H1 antihistamines, levocabastine and emedastine,
have been shown to signicantly inhibit ocular itching, but these
are combined with a-adrenergic agents in order to maximize the
alleviation of conjunctival redness. The topical antihistamine
levocabastine has been shown to be effective in the relief of
itching.42,43
Topical nonsteroidal antiinflammatory drugs such as ketorolac
and flurbiprofen inhibit the activity of the cyclooxygenase
pathway and the production of prostaglandins. They inhibit
vasodilation and edema and are useful in reducing itching and
conjunctival injection, but not very helpful in ridding the eye of
excess immune cells.
Single-action mast cell stabilizers include cromolyn sodium
4%, pemirolast potassium 0.1%, nedocromil, and lodoxamide
0.1%. Cromolyn and lodoxamide are indicated for the treatment
of VKC and AKC, but not allergic conjunctivitis. Cromolyn in
vitro inhibited activation of neutrophils, eosinophils, and monocytes, thus blocking tissue destruction.44 This may explain why
this drug is more effective in chronic cell-mediated disorders
such as VKC and atopic conjunctivitis and not as effective in
the type 1 hypersensitivity allergic conjunctivitis. Cromolyn has
also been shown to inhibit the release of substance P and other
neuropeptides from nerve endings.45
Pemirolast has been demonstrated to be upto 100 times more
potent than cromolyn in vitro and in animal studies, and in
clinical trials, has been shown to be effective at completely
resolving itching in some patients.46
Nedocromil has been shown to be more effective than
cromolyn in stabilizing mucosal mast cells, and this difference
may account for its effectiveness in a disease in which cromolyn
has not shown efcacy.47 Nedocromil has also been shown to
block eosinophils, neutrophils, macrophages, monocytes, and
platelets.48 In clinical trials, nedocromil was more effective than
cromolyn in resolution of itching, hyperemia, epithelial keratitis,
corneal pannus, and other symptoms in patients with ocular
allergy.49
The clinical efcacy of lodoxamide may be superior to that
seen with cromolyn, as has been reported in the literature for
the treatment of AKC, VKC, and GPC.50,51 It is unclear how
effective lodoxamide is for seasonal allergic conjunctivitis. One
clinical study of lodoxamide showed it to be more effective than
the placebo in treating seasonal allergic conjunctivitis during
peak pollen exposure.51
Steroids, such as loteprednol etabonate, are an option for
allergic conjunctivitis but are usually reserved for AKC and
VKC. Corticosteroids inhibit the production of arachidonic
acid itself, reducing the production of all three eicosanoids
(prostaglandins, thromboxanes and leukotrienes) reducing
redness, edema, and inflammation. However, their long term
use can be associated with side effects including delayed wound
healing, increased intraocular pressure, local immunosuppression,
and resultant superinfection and induction of cataractogenesis.
615
SECTION 6

shortening, and symblepharon are seen in patients with AKC.58
In this stage of the disease, both the bulbar and the tarsal
conjunctiva appear hyperemic.
Corneal scarring, suppurative keratitis, and keratoconus are
the major reasons for loss of vision in patients with AKC and
may require immediate attention. Corneal involvement can begin
as supercial peripheral keratitis, with or without inltrates.
Gelatinous inltration, opacication, HornerTrantas dots, and
true cysts may be seen.53 Generally, patients with AKC display
punctate epithelial keratopathy and intraepithelial microcysts.59
Ulcers resulting from AKC are typically ovoid, horizontally
oriented, and have irregular borders. Peripheral corneal neovascularization is a prominent sign.59 Severe cases are marked by a
hazy vascularized cornea, interfering with normal visual
function.60 Keratoconus is estimated to be found in 25% of patients
with atopic dermatitis61 and in 16% of patients with AKC.59
Cataractogenesis is estimated to occur in ~810% of patients
with atopic dermatitis54,61 and is unique to AKC. Patients
treated with or without steroids may develop cataracts. Almost
90% of the AKC-associated cataracts are bilateral and are either
anterior subcapsular (shield-like) or posterior polar. They have
been observed as early as the teenage years, and the rate of
progression can vary from several months to many years. The
fundus may show degenerative changes and retinal detachment
may occur.
Contact lenses are often not tolerated by patients with atopy,
but if contact lenses are worn, signs of GPC may be superimposed on those of AKC. A study of the clinical features
of AKC revealed that 95% of the patients surveyed had
concomitant eczema and 87% had concomitant asthma.59 Hay
fever, migraine headaches, and rhinitis have also been reported
as part of this symptom complex.58
The ocular inflammation is perennial, although exacerbations
related to airborne allergens are common. Animal dander was
suspected as the precipitative factor in 51%, dust-type allergens
were thought to be causative in 43%, and food allergies were
suspected of being responsible in 35% of patients questioned
in one study.59 Patients diagnosed with the condition are
encouraged to control their environment as much as is practical
and to identify and avoid offending allergens if possible.
Patients should be questioned extensively about other atopic
conditions and about a family history of atopy. Patches of dry,
erythematous, pruritic skin may be ignored by patients who fail
to recognize it as eczema. Secondary staphylococcal infection of
the lids may complicate the AKC-associated blepharitis.
Laboratory tests measuring serum and tear IgE levels may
conrm the diagnosis of AKC if the levels are elevated. In the
serum, IgE is normally less than 100 IU/mL.58 Conjunctival
scrapings that contain eosinophils and mononuclear cells are
also indicative of AKC, although this procedure is not encouraged because it may promote additional scarring.
AKC may not always be the sole allergic disorder presenting;
some investigators feel that it can occur in conjunction with
acute allergic conjunctivitis and even VKC.58 The presence of
severe itching and induration of the lids is indicative of AKC
and can be used to differentiate it from chronic blepharitis.57
Moreover, herpes simplex and staphylococcal infections are not
uncommon in patients with AKC and must also be considered.57,62
616
The underlying mechanism of AKC is thought to be both a type

1 hypersensitivity and a type 4 delayed hypersensitivity
response.57 Histological evaluation of conjunctival biopsies from
patients with AKC reveals an elevated number of mast cells and
inltration by basophils, eosinophils, and lymphocytes. Mast
cells are found in much higher densities in the conjunctival
epithelium of patients with AKC and tryptase and heparin may

contribute to papillae formation and conjunctival scarring.63
Pronounced degranulation of eosinophils and neutrophils has
been reported, contributing to the clinical picture of the disease.62
The epithelium of patients with AKC appears convoluted, and
small brovascular stalks have been described.63 Goblet cell
proliferation and epithelial pseudotubular formation have also
been noted in conjunctival biopsies of patients with AKC.64
Differences have been shown among the conjunctival T-cell
subpopulations characteristic of patients with AKC, normal
individuals, and patients with ocular cicatricial pemphigoid
with respect to the antigen-recognizing receptor TCR. TCR
a- or b-containing T cells are found in the normal conjunctiva
and in the conjunctiva of patients with ocular cicatricial
pemphigoid; TCR g or d predominated, suggesting that TCR
g or d may play a role in the autoimmune diseases but not the
allergic disorders. TCR g or d cells were not found in patients
with AKC, but an increase in TCR a- or b-containing cells was
noted in the substantia propria when compared with that of
normal individuals. This could prove to be a method of
differentiating between severe AKC and the other cicatrizing
disorders. In addition, a statistically signicant increase in
Langerhans cells (CD1+) was found in the epithelium and in
the substantia propria of patients with AKC.65
IgE levels are elevated in both the serum and the tears of
patients with AKC. Although these levels generally do not
correlate with the severity of the disease,59serum IgE has been
observed to decrease when the patient goes into remission.
Although exposure to allergens is associated with exacerbation
of the disease, allergen-specic (pollen, mite, and cat) IgE levels
were not elevated in the tears of patients with AKC compared
with those of nonatopic controls. Allergen-specic IgE levels
were elevated in the serum, however. Despite the increase in
circulating IgE antibodies, the number of IgE-bearing lymphocytes in circulation appears to remain within normal limits. By
contrast, the number of complement-bearing lymphocytes has
been reported to increase.
The cellular immune response appears to be dysfunctional in
patients with AKC, as evidenced by the absence of type 4 delayed
hypersensitivity responses to Candida and streptokinasestreptodornase antigens and the inability of some patients to
become sensitized to dinitrochlorobenzene,66 depressed mitogenic T-cell responses to phytohemagglutinin,67 and an increased
susceptibility to fungal and viral diseases.68 The circulating
immune cell prole includes fewer peripheral T cells but increased
B cells and eosinophil counts. The classic theories attempting
to explain the pathogenesis of AKC have been based on the
dysfunction of the cellular aspect of the immune system,
postulating that regulation of IgE synthesis is not properly
maintained by the T-cell population. Excessive binding by the
overabundant IgE molecules to resident mast cells may induce
a somewhat continual release of histamine and other mediators,
producing the clinical picture observed with this disease. Indeed,
the dense numbers of mast cells in AKC biopsies have been
observed to be in various states of degranulation.64
Early experiments in animals identied immune response
genes, called Ir genes, which were linked to the major histocompatibility complex of these animals.6971 The Ir gene was
found subsequently to control the function of T-cells and related
intercellular interactions.71 Later research on IgE has revealed
similar relationships in humans and the role of the T cell on IgE
expression.72 Combined with the strong family history associated with the disease and the observation that, under histological evaluation, the human conjunctival epithelium stains
intensely for HLA-DR glycoproteins,64 these ndings suggest
that the actual site of the aberration in AKC lies at the
chromosomal level.
The mast cell stabilizer cromolyn sodium has been reported to

be effective in the treatment of AKC.60 One study reported a
reduction in itching, watering, and photophobia in 66% of
patients taking one to two drops qid.72 Another trial revealed a
statistically signicant improvement in photophobia, discharge,
papillary hypertrophy, limbal changes, and corneal changes
compared with placebo treatment.74 In contrast, a clinical trial
concurrently treating one eye with 2% cromolyn sodium and
the fellow eye with 1% medrysone every 6 h showed cromolyn
to be ineffective, whereas medrysone was highly effective in
improving both objective and subjective parameters.75 Package
labeling advises that the efcacy of cromolyn sodium is
dependent on its instillation at regular intervals. Furthermore,
a loading period of up to 2 weeks may be necessary before the
complete drug effect is observed. Other mast cell stabilizer
treatment possibilities include lodoxamide, pemirolast, or
nedocromil. For some patients, mast cell stabilizer/
antihistamines are often effective treatments for signs and
symptoms of AKC.
Typically, topical vasoconstrictorantihistamine combinations
are benecial for the relief of symptoms in the less severe forms
of the disease. Symptomatic relief has also been achieved to
some extent by lowering tear pH with saline irrigation and the
use of mucolytic agents54 and cold compresses.60 Systemically
administered antihistamines may alleviate the symptoms, but
may also induce dry eye symptoms, which could complicate the
ocular condition.76,77 The use of systemic antihistamines is
usually reserved for those with atopic conditions affecting
other organ systems. Terfenadine (60120 mg bid), astemizole
(10 mg/day), and hydroxyzine hydrochloride (50 mg at bedtime
with gradual introduction into daytime with dosage escalation)
were found to be effective.58
Corticosteroids bring about the most dramatic improvement
in symptoms, but these agents must be used with caution
because of the increased risks. Pulse-type therapy is an alternative available to the practitioner. Topical steroids in strong
concentrations (1%) are used up to eight times per day for
periods not exceeding 7 days, followed by a rapid tapering.
Continuous wear soft contact bandage lenses used in combination with 0.125% prednisolone acetate instilled qid have
been successful in the treatment of AKC-related epithelial
defect.58 In severe recalcitrant cases of AKC, systemic steroids
(such as a medrol pack) may be used.
The treatment of corneal ulceration with antibiotics and the
treatment of blepharitis, resulting from opportunistic infection
of the compromised lid epithelium, should not be ignored. The
prevalence of such conditions was demonstrated by the isolation
of Staphylococcus aureus from 67.6% of the lids of patients with
AKC in a study by Tuft and colleagues.59
Application of topical tacrolimus on eyelid skin may be
effective for treatment of severe atopic dermatitis of the eyelids,
and may have secondary benets for AKC.78 Because a deciency in the suppressor T-cell population has been implicated
in the failure to arrest IgE responses, drugs geared toward
modulating the numbers, maturity, and function of this immune
cell subpopulation were evaluated with some success in the
1970s and 1980s. Indeed, the numbers of CD8 or cytotoxic
suppressor T-cells do not differ between AKC and normal
patients in conjunctival biopsies, whereas CD3, CD4, and CD5
T-cell subpopulations are greatly increased in AKC.64 Further
understanding of the roles of the various control mechanisms of
the immune system with respect to atopy may warrant
additional investigation into therapeutic agents of this type.
Topical cyclosporine A 0.05% seems to be safe and have some
effect in alleviating signs and symptoms of severe AKC
refractory to topical steroid treatment.79 Oral cyclosporine at

a dosage of 400 mg/day has been used in combination with
topical treatments in patients whose serum IgE levels were
exceedingly high.58 The use of lymphokines is also being explored
in the treatment of immune disorders. The identication of a
subset of the T-cell population that produces IgE-binding factors
when stimulated by glycosylation-enhancing factor but
produces IgE-suppressive factors when exposed to glycosylationinhibiting factor suggests that recombinant human IgEsuppressive factors can be developed for therapeutic purposes.72
Regardless of the therapeutic choice made to treat the ocular
symptoms, the systemic nature of the disease warrants a
multidisciplinary approach to ensure complete and efcacious
control of AKC. Other targets for therapeutic intervention
include inhibition of tryptase, cyclooxygenase, leukotrienes,
bradykinins, platelet activating factor and IgE. Cytokine
antagonism and agonism, T-cell or eosinophil inhibition and
adhesion molecule antagonism might provide potent new
modes of treatment.
VERNAL KERATOCONJUNCTIVITIS
Vernal, derived from the Greek meaning occurring in the
spring, is a rare, serious form of ocular allergy. The disease
characteristics include a predilection for warm rather than cold
climates, a frequent family and personal history of atopic
disease, a higher than 2:1 frequency in males over females, and
an early onset, with remission by the late teens, very frequently
at the onset of puberty.80,81
Vernal keratoconjunctivitis (VKC) has a hereditary predisposition with exogenous factors, such as climate, season, and
allergen exposure, determining the likelihood and severity of
this disease. Arid areas with the potential for wind and desert
storms, such as the Middle East and North Africa, have the
highest incidence of VKC. Dahan82 noted a high frequency of
VKC in black patients in South Africa, almost all of whom had
the limbal form. Patient histories in Israel and in Egypt had
shown VKC to be year-round and rarely associated with atopy
in these areas.83,84

The clinical manifestations of VKC include raised conjunctival
cobblestones over the upper tarsal plate, and almost never over
the lower plate, with no signicant conjunctival hyperemia
(Fig. 47.4). Bulbar conjunctival cobblestones, papules, or
follicles are almost never observed. The cobbles are large and
pleomorphic, and rarely are they evenly distributed. Bulbar
conjunctival vasodilatation is diffuse and presents as pink,
rather than red. HornerTrantas dots, rst described in the
1880s,85 are chalk-white, raised supercial inltrates straddling
the limbus with no specic meridional predilection (Fig. 47.5).
Gelatinous, translucent, globular deposits at the limbus vary
greatly in size and shape, from a 2 mm circle to an arc to a 360o
ring. Diffuse keratitis is present in more severe cases. The
shield ulcer, a central ovoid epithelial defect with a white brin
coating, is well delineated with no surrounding haze (Fig. 47.6).
These ulcers are almost never associated with iritis. Copious,
tenacious cordlike mucus with highly elastic properties is
always present in VKC. The common and often debilitating
symptoms of VKC are itching, photophobia, and pain.
The full manifestation of VKC provides few diagnostic
difculties. The large cobblestones of the upper tarsal plate are
pathognomonic for this disease, but they do require lid eversion
to be identied. Thus, this is an indispensable component of
the external ophthalmic examination. These cobbles differ from
those in GPC by being dramatically larger in height and breadth
CHAPTER 47
TREATMENT
617
FIGURE 47.4. Vernal conjunctivitis. The palpebral form is

characterized by enlarged papillae, referred to as cobblestone papillae
because of their shape and size, that are almost always conned to
the upper tarsus. The conjunctiva is often pink.
SECTION 6
FIGURE 47.5. Vernal conjunctivitis. The limbal form exhibits

HornerTrantas dots, which are white inltrates found at the limbus
that can vary greatly in size and specic location or pattern. In more
severe cases, diffuse keratitis may be observed.
618
and varying in shape in a different pattern than the homogeneous cobbles seen in GPC or in the follicles in viral
conjunctivitis. Scarring is not present, regardless of the number
of years that VKC has been present. If scarring exists, it is more
suggestive of the Arlt lines found in trachoma. The development of ptosis is related to the presence of keratitis and
photophobia, producing a protective response. Additionally, it
can be caused by the increased bulk of upper tarsal conjunctiva
or myositis of the levator muscle. Ptosis and conjunctivitis in
combination can also be found in trachoma, chlamydia, GPC,
herpes zoster, and follicular conjunctivitis.
The pattern of vasodilatation seen in VKC is nondescript but
gives the conjunctiva a pink color rather than the red observed
in severe corneal ulcers and infectious conjunctivitis. Mild to
moderate chemosis, sometimes visible only with a slit-lamp as
pinkish fluid slightly separating the conjunctiva from the
underlying episclera, is commonly seen in VKC, rather than the
ballooning chemosis of acute conjunctivitis. However, other
forms of conjunctivitis also present with this mild chemosis.
The gelatinous nodules of limbal VKC are vascular and rapid
in onset and respond promptly to topical steroids, factors that
FIGURE 47.6. Vernal conjunctivitis. The shield ulcer is the most

serious consequence of vernal conjunctivitis. It is a centrally located,
white, brinous defect in the corneal epithelium. It lacks the
surrounding haze often seen with other ulcer types and is rarely
accompanied by iritis.
differentiate them from other limbal tumors. The chalk-white

HornerTrantas dots are elevated and straddle the limbus,
unlike immune marginal inltrates, which have a surrounding
lucid area, involve the corneal stroma, and are separated from
the limbus by ~1 mm. The shield shape of the central corneal
ulcer with its overlying white plaque differs from that of other
corneal ulcers in the absence of a surrounding haze, iritis, and
purulent discharge. However, the keratitis of VKC is diffuse and
provides no help with differential diagnosis.
Itching is pathognomonic for all allergic disease. Yet, the
intensity of the itching seen in patients with VKC requires
vigorous knuckle rubbing, and this observation is very typical of
this disease. The lids, other than manifesting mild edema, are
not involved. There is usually no excoriation of the lateral
canthal area or associated allergic blepharitis.
A careful and complete ocular examination, including upper
and lower lid eversion, careful limbal evaluation, attention to
the nature of the mucus, observation by slit-lamp of chemosis,
absence of conjunctival scarring, lid involvement, follicles, and
pretragal adenopathy, will inevitably lead to a correct diagnosis
of VKC.

The cobblestones in VKC represent dramatic collagen
proliferation and ground substance and cellular accumulation.
Mast cells are found in increased numbers, 80% degranulated,
located more supercially in the conjunctiva, and more likely
to be found in 1-m light microscopic or electron microscopic
sections.10,86 Eosinophils are also found in increased numbers
and located more supercially; thus, they are frequently available for recovery in scrapings. VKC is the only ocular surface
disorder in which greater than two eosinophils can be found
per 25-power objective eld.28 Mast cells, lymphocytes,
macrophages, basophils, and rarely, polymorphonuclear cells
are also present.
The rst stage in the development of VKC is heralded by a
prehypertrophic phase of hyperemia and a thin, milky-white
pseudomembrane. Subsequently, hypertrophic changes occur
that are related to a stromal inltration with large papillae
covered by an epithelial monolayer with mucoid degeneration
in the crypts between papillae. The early cellular and vascular
phase is replaced by collagen deposition, hyaluronization,
TREATMENT
Therapy for both AKC and VKC, as for all allergic disease,
should be aimed primarily at the identication of the allergen
and, when possible, its elimination or avoidance. Although
these patients generally have multiple sensitivities to allergens
such as grasses, dust, and mites, avoidance can be extremely
TABLE 47.1. Mediators Identied in Vernal Conjunctivitis and

Their Effects
Mediator
Levels
Origin
Effect
Histamine
Mast cell
Itch/vasodilatation/chemosis
EMBP
Eosinophil
Keratitis/ulcer
Tryptase
Mast cell
Effect on ground substance?
PGF
Mast cell
Vasoactivity; chemotaxis
ECF
Eosinophils
Mast cell Eosinophil

chemotaxis
ECF, eosinophil chemotactic factor; EMBP, eosinophil granule major basic

protein; PGF, prostaglandin F.
helpful in acute exacerbations. The use of air conditioning with

the appropriate lters can also be helpful. Avoidance of wind,
which is usually pollen bearing, and use of glasses or goggles to
decrease airborne allergen contact should also be suggested. As
mentioned previously, in North Africa, the wind coming off the
desert has long been associated with increased incidence and
exacerbations of VKC. A nal suggestion to the patient is to
limit digital manipulation and knuckle rubbing of the eye as
much as possible, as this has been shown to cause mechanical
release of mast cell mediators.
Cold compresses are helpful in treating AKC and VKC,
perhaps related to a vasoconstricting effect or to some minor
role in mast cell stabilization. Elimination of the allergen will
yield dramatic benets, particularly in the presence of shield
ulcers and keratitis. The use of bandage lenses should be
avoided since they can trap allergen in the precorneal tear lms
and worsen keratitis.
Tear substitutes are helpful because of their barrier function,
their allergen dilutional function, and their irrigating ability.
The use of vasoconstrictors can inhibit vascular transudation,
thus decreasing chemosis.100 Although rebound vasodilatation
does not occur with the ocular use of vasoconstrictors, overuse
must be avoided to prevent tachyphylaxis and medicamentosa.101
Topical antihistamines can provide short term itching relief
for VKC patients by virtue of their H1 activity.15 In combination
with vasoconstrictors,31 these may provide adequate control in
mild conditions or allow for a decrease in the dosage of, or a
delay in the use of, topical steroids. A topical antihistamine
mast cell stabilizer, such as olopatadine, can also be applied
owing to its H1 activity and inhibition of mast cell
degranulation. Olopatadine 0.1% has also been shown to reduce
the number of goblet cells in brush cytologic specimens of VKC
patients after 2 months of treatment, which, in turn, decreased
the amount of mucus discharge.102
The double-edged sword of steroids is acutely evident in VKC
therapy. Therapeutic response to topical steroids can be
dramatic. However, the potential for superinfection and delayed
wound healing as well as cataract and glaucoma development
must be taken into account. For these reasons, pulse therapy of
a topical steroid such as prednisolone phosphate 1%, six to eight
times per day for up to 1 week, followed by rapid tapering to the
lowest levels needed for patient functioning, should be
prescribed. Steroids should not be used to eliminate the last
vestige of vasodilatation or itching, nor should the clinician
expect immediate resolution of the cobbles. Cobbles can remain
for many months without creating clinical problems. Surgical
removal of these cobbles with cryotherapy should be avoided,
because the resultant scarring of the conjunctiva can lead to lid
and tear lm abnormalities that will persist as a life-long
problem after the VKC has spontaneously resolved. Patching
with antibioticsteroid combinations is highly effective in
CHAPTER 47
decreased vascularity, and an overall decrease in inflammatory

cells. The limbal papillae, although differing in clinical appearance, undergo the same pathologic transformation.87 Horner
Trantas dots consist mainly of eosinophils and degenerated
cellular debris, but they may also contain polymorphonuclear
cells and lymphocytes.85
VKC represents a cell mediated hypersensitivity reaction by
T-cells. The increased number of degranulated mast cells and
their more supercial location was an important nding by
Allansmith and colleagues in implicating the central role of
these cells in VKC. The epithelium of normal subjects
contained neutrophils and lymphocytes but not plasma cells,
eosinophils, mast cells, or basophils. All these cells are found
in the conjunctival epithelium of the vernal patient. The
substantia propria of patients with VKC has substantially fewer
lymphocytes and more eosinophils, basophils, and mast cells.10
Leonardi and colleagues observed the role of mast cells in VKC
by observing the increase of mast cells in the stroma and
epithelium and the increase of thick, abnormal collagen bers
in seven eyes of patients with VKC.88
A comparison of the percentage of degranulated mast cells
in VKC and GPC showed little difference.10 The fact that
histamine levels are dramatically elevated in VKC and not in
GPC suggests that there is a higher level of histamine release
from the mast cells of patients with VKC.18,89
There is increased mucus production in allergic conjunctivitis, but more so in VKC. The mucus in VKC differs in
consistency from the thin strands seen in allergic conjunctivitis.
The thick, tenacious, ropy strands have an elastic quality
described as the Maxwell Lyon sign. This chewing gum-like
mucus contains inflammatory cells, specically, dramatic
numbers of eosinophils and their CharcotLeyden granules.90
Mucopolysaccharides, possibly hyaluronic acid, have been
noted by Neumann and Blumenkrantz.91 In certain patients,
the pH of the secretions has been noted to be highly alkaline.92
Such tear pH elevations have been found otherwise only in
severe rosacea keratitis.93
The central role of the mast cell and its mediators in VKC is
well established. The increased number and supercial location
of mast cells, dramatically degranulated, with increased tear
histamine and IgE levels94 leave little doubt that these factors
contribute to the pathogenesis of this disease. It remains
unknown why certain individuals respond to allergens with
such severity, and undoubtedly many other factors are involved.
Tear assays have shown 10-fold increases in tear histamine
levels from patients with VKC. Histaminase inactivation in
acute allergic conjunctivitis led to 10-fold increase in recovery
of histamine from tears (levels similar to the highest levels
found in patients with VKC).19 This suggests the possibility of
a histaminase dysfunction contributing to the high histamine
levels seen in VKC. In support of this theory, blood histaminase
levels were found to be reduced in patients with VKC.95,96
In addition to histamine, allergen-specic IgE antibodies
in tears and serum, tear IgG,97 and tear tryptase levels98 are
elevated in patients with VKC. Tear EMBP has been eluted from
shield ulcers.22 The mucoid plaque overlying the shield ulcer
has been shown to contain eosinophils and their granules.99
Table 47.1 summarizes the mediator effects presently elucidated in VKC.
619

treating shield ulcers, and in recalcitrant cases of shield ulcer
plaque, debridement is highly effective with or without
amniotic membrane.
Studies have shown the ability of cromolyn to often decrease,
and occasionally eliminate, the amount of or need for steroid
use in certain patients with VKC.84,103,104 Other agents used
in VKC treatment include topical lodoxamide,50 nedocromil,47
levocabastine,42 ketotifen,105 and cyclosporine.106,107 Topical
cyclosporine 2% provides a marked reduction in the symptoms
and signs of VKC, and is helpful as a steroid-sparing agent. It is
an immunomodulator inhibiting the clonal expansion of the
helper T subset of lymphocytes and the release of interleukins.
SECTION 6
GIANT PAPILLARY CONJUNCTIVITIS
620
Giant Papillary Conjunctivitis (GPC) is not a true allergic

reaction, but is instead an inflammatory reaction of the upper
tarsal conjunctiva associated with the presence of contact
lenses, surgical suture barbs, and ocular prostheses.108 Initially,
GPC was mistakenly classied as an allergic reaction because of
its vague resemblance to VKC. Both diseases are characterized
by a papillary reaction on the upper tarsus, but the similarity
ends there. The papillae of GPC are small and even, between
0.3 and 1 mm in diameter. In VKC they are large (greater than
1mm in diameter), irregular, and often have mucus between
them. VKC patients report severe itching, whereas there is
limited itching seen with GPC. The epidemiology of the two
diseases is also quite different. While VKC is commonly seen in
young males and generally resolves after puberty, GPC, though
it is often seen in both female and male children and teenagers,
may affect patients at any age. In addition, histopathological
examinations show no increase in histamine or eosinophil
levels in GPC, which are two hallmark signs of chronic ocular
allergy.109
GPC is characterized by similar copious tearing, foreign body
sensation, copious production of mucus, and the proliferation of
subepithelial collagen, leading to the eruption of giant papillae.110
Patients may have blurred vision and excessive contact lens
movement serious enough to cause intolerance to lens wear.
Both VKC and GPC arise from similar underlying pathophysiologic mechanisms. VKC and GPC are best explained
by a hyperactivity of resident mast cells, lymphocytes, and
broblasts, and the proliferation of collagen and formation of
conjunctival papillae. Advances in theory have been translated
to rational treatment.
GPC and similar conjunctival papillary disorders are the
result of: (1) genetics, (2) the appropriate triggering agent, (3)
sufcient duration of exposure, (4) sufcient area of exposure
on the conjunctival surface, and (5) the particular geometry of
the exposure. Thus, the onset of disease is the result of two
broadly dened factors-genetics and physical trauma.
Retrospective epidemiologic studies of GPC have revealed a
strong association of GPC and wearing contact lenses
, especially hydrogel lenses.111 Younger patients were shown to
have a higher risk of developing GPC. Gender and tear lm
breakup time were not found to be associated with the condition. GPC is almost exclusively bilateral with a mean onset
time of 31.4 months after commencing lens wear. A retrospective study of the personal histories of patients with GPC
disclosed a higher incidence of atopy. Like PAC, GPC showed a
bimodal distribution, with peaks in the spring and late summer
and autumn. Patients with GPC reported a higher incidence of
allergy to pollen as well as to drugs and medications, but the
only statistically signicant discriminator between patients
with GPC and comparison patients was sensitivity to
thimerosal. The seasonal onset of GPC diagnoses in 1987 and
1988, and the increase in reported allergies within the GPC
group, suggests a strong association between atopy and the

development of GPC.112
GPC can result from an exposed suture end that abrades the
upper palpebral conjunctiva.113 This syndrome consists of a
mucoid ocular discharge with blurred vision, a foreign body
sensation, upper lid edema, and blepharoptosis concomitant
with giant papillae of the upper palpebral conjunctiva. Removal
of the offending suture(s) results in resolution of the papillae
and symptoms.

Contact lens-associated GPC is observed in wearers of both
rigid gas-permeable and soft lenses, although the incidence is
greater in wearers of soft contact lenses. Supercial neovascularization, contact lens-associated superior limbic keratoconjunctivitis, and GPC are all associated with wearing soft contact
lenses.114 Furthermore, mechanical irritation from the largediameter soft contact lenses and a tendency for these lenses to
be coated with mucoprotein deposits increase the potential
problems.115
Signs and symptoms of GPC range from minimal discomfort
upon inserting or removing contact lenses to complete
intolerance of the contact lenses.108 The earliest symptoms of
GPC (which precede signs) are irritated eyes when the lenses are
removed, accumulations of mucus in the nasal corner of the
eye, and slight blurring of vision due to coatings on the surface
of the contact lens. Symptoms of more advanced GPC may
include foreign body sensation and reports of mucus gluing the
eyes shut during sleep.
Signs of GPC may range from mild hyperemia of the upper
tarsal conjunctiva, with strands of mucus streaking the otherwise smooth conjunctival surface, to the presence of milky-white
discharge covering broad areas of giant papillae. Early in the
progress of GPC, the conjunctiva remains translucent, but careful
observation reveals that the conjunctiva is somewhat thickened.
As GPC progresses further and inltration by inflammatory cells
continues, the conjunctiva acquires a more opaque appearance.
Conjunctival papillae larger than 0.3 mm in diameter are
abnormal.110 In GPC, as these abnormally large papillae
emerge, they push aside the normal smaller papillae, their
apexes flatten, and there may also be conjunctival ulceration.
The precise location of papillae varies with the type of contact
lens worn by the patient.116 Wearing soft contact lenses tends to
induce papillae that appear rst in the upper zone of the tarsal
area and then progress toward the lid margin. Enlarged papillae
may be seen throughout the upper tarsal conjunctiva in
advanced cases of GPC. Wearing hard contact lenses tends to
induce fewer and smaller papillae, with crater-like flattened
(rather than rounded) tops. Those papillae are likely to be found
rst along the lid margins.

Disruptions of normal host defenses are present in both VKC
and GPC. Although patients with VKC and GPC are found to
have normal tear concentrations of tear lysozyme, for example,
tear concentrations of lactoferrin are reduced in VKC and
GPC.117 Lactoferrin, an essential component of the nonspecic
immune protection of the external eye, is reduced in patients
with active VKC and GPC, although patients with inactive GPC
had normal tear levels of lactoferrin.118 The contribution of
reduced lactoferrin to the onset and course of VKC or GPC
remains unclear. It is possible, however, that decreased
lactoferrin in the tears of patients with VKC and GPC somehow
contributes to the increased ocular inflammation and to the
troublesome bacterial contamination of worn contact lenses.
TREATMENT
Successful treatment depends on early recognition of the
condition, although signs and symptoms will resolve if the
patient refrains from wearing the contact lens. The rst concern
is the prevention of GPC. Prevention depends on: (1)
encouraging strict lens hygiene, and (2) prescribing the
appropriate lens type and edge design. Treatment of GPC
likewise depends on nding the appropriate lens material and
design and encouraging proper lens hygiene. Treatment also
requires proper therapy to control the conjunctival
inflammatory response.
Regarding hygiene, lens cleaning agents and saline solution
for rinsing and storing lenses should be thimerosal free.
Enzymatic cleaning of the lens with papain preparations is
essential to minimize the accumulation of lens coatings and to
remove build-up of environmental antigens that may adhere to
the lens coating. Finally, lenses should be replaced frequently.
Fluorescein staining of the apices of enlarged papillae, heavy
mucus, signicant conjunctival hyperemia, and movement of
the lens on blinking (decentering) are all indications that the
patient should discontinue wearing contact lenses until the
signs and symptoms of GPC resolve. A marked increase in
immunoglobulin deposition and enhanced IgG:IgA (P <.001) is
common to high-water content lenses (especially those of
nonionic composition) used on an extended wear basis, when
compared with low-water content lenses used on a daily wear
basis. It is thus hypothesized that use of high-water content
lenses on an extended wear basis leads to a greater degree of
inflammatory or immune stress.121 However, frequent changing
of the lens material or design allows patients to continue
wearing contact lenses. Because GPC seems to occur less
frequently with hard than with soft contact lens wear, if GPC
develops with soft contact lens wear, changing to the rigid gaspermeable lenses may resolve the problem.
Hydrogel contact lenses appear to result in an overall
prevalence of GPC of ~20%. The stiffness of the material,
rather than thicker edges, is presumed to be the principal factor
behind the higher incidence of GPC in silicone hydrogel lens
wearers. Changes to lens design and the introduction of a
steeper base curve may reduce the incidence levels previously
reported.
There is no need to wait until the enlarged papillae regress
before reintroducing contact lenses. In fact, it may take several
months or even years for the enlarged papillae to disappear. In
the absence of signs indicating the need for withdrawing lenses,

contact lenses may be reintroduced 35 days after symptoms
such as hyperemia and excessive mucus production have been
resolved.
Unlike in the treatment of VKC, topical corticosteroids have
not proved particularly effective in the treatment of GPC. A
short course of corticosteroids may quiet the inflammation
before long-term management of the disease is begun.
TOXIC KERATOCONJUNCTIVITIS
Toxic keratoconjunctivitis (keratoconjunctivitis medicamentosa) is one of the most frequently encountered problems in
the subspecialty of cornea and external disease. Taking a careful
history is critical to establishing correct diagnoses in
ophthalmology. Potent medications used inappropriately can
result in toxic or hypersensitivity reactions. For example, a
patient with an underlying dry eye problem may be
misdiagnosed as having viral conjunctivitis. Various antibiotics
or antiviral agents may be prescribed without subsequent
improvement or with worsening of the initial problem. Owing
to a deteriorating clinical condition, the practitioner may alter
therapy using other agents possibly aggravating the clinical
picture. As a general rule, if a patient has been treated with
multiple medications over the course of weeks to months, with
no apparent improvement or worsening of the original
complaint, all topical medications should be discontinued for at
least a few weeks. This allows the practitioner to re-establish a
baseline clinical status and may result in clinical resolution.
Toxic symptoms may range from mild irritations to
ulcerative keratitis with potentially sight-threatening visual
consequences. By denition, a toxic substance is poisonous and
may cause a disturbance of structure or function.122 Although
irritation implies inflammation, generally speaking, toxicity
and irritation are terms that may be used interchangeably.
Differentiating between an allergic and a toxic reaction may be
difcult because some drugs may elicit both reactions with
similar biochemical mechanisms. In general, allergic reactions
require repeated exposure to the sensitizing agent and a
sufcient amount of time to elapse for sensitization of the
immune system. This time period may range from 510 days to
years, depending on the potency of the sensitizing agent and the
susceptibility of the exposed individual. Patch testing may help
differentiate allergy from toxicity; however, false-positive and
false-negative test results may frustrate the clinician in a
situation in which a careful history and examination are more
likely to establish the etiology.
Factitious (self-induced) disease results from mechanical
trauma or toxicity from eye drop abuse. The incidence of
iatrogenic keratoconjunctivitis was found to be 13% at one
tertiary center.123 Healing was prolonged, taking 793 (median
28.5) days. From a research standpoint, toxicities from drop
abuse are often not documented in the medical chart and not
reported. Factitious disease should only be considered after
iatrogenic causes have been investigated.
In addition to redness, the conjunctival tissue response to
toxic agents may result in follicular or papillary excrescences.
Typically, the reaction is more prominent in the inferior bulb,
fornix, and tarsal conjunctiva. Conjunctival scrapings may reveal
mononuclear cells, a few neutrophils, and mucus. Eosinophils,
the hallmark of an allergic reaction, are generally absent unless
a combined allergic and toxic mechanism is present. Epithelial
cells, mononuclear cells, and polymorphonuclear cells may show
toxic large basophilic cytoplasmic granules.122
The most common manifestation of corneal toxicity is a coarse
punctate epithelial keratopathy. Heaped-up opaque epithelium,
swirl patterns, and pseudodendrites may occasionally develop.122
CHAPTER 47
Eosinophil degranulation commonly occurs in VKC and GPC.

EMBP is elevated in conjunctival tissues of patients with VKC
and GPC.119 The cytotoxic effects of these cationic proteins are
almost certain to play a major role in the conjunctival inflammatory reaction and the subsequent deposition of collagen in
the pathogenesis of VKC and GPC.
The signicance of atopy in GPC is emphasized by the
nding of elevated tear concentrations of IgG and IgE in GPC,
perhaps owing to the presence of antigenic coatings on the
surface of the worn contact lens. Tear IgE levels in patients with
GPC were signicantly increased, especially in the more
symptomatic eye (geometric mean of 6.9 IU/mL, P <.01), compared with those in a control group who also wore contact
lenses (2.1 IU/mL). Increased tear IgG levels (50.7 g/mL, P <.01)
were found in the more symptomatic eyes of patients with
GPC.120 In eight of the 18 patients, tear IgM was measurable
(> 4.7 g/mL), whereas none of the patients in the control groups
had detectable amounts of IgM in their tears. Studies with
transferrin as a marker for the vascular leakage of serum proteins
into the tears showed that local production was responsible for
the increased tear immunoglobulin levels.
621

These morphologic patterns have been reported most commonly in cases of idoxuridine toxicity, an antiviral now used
infrequently. Possibly the leading cause of pseudodendritic keratitis is timolol, a b-blocker used in the treatment of glaucoma.
Toxic ulcerative keratitis is the most severe form of corneal
toxicity. Schwab and Abbott have reported on 19 such cases, of
which ve were factitious and 14 iatrogenic.124 In this series,
the corneal epithelial defects were typically oval, occurred
inferonasally, and had gray rolled edges with surrounding
intense supercial keratitis. The lusterless conjunctiva and
cornea stained well with both rose bengal and fluorescein dyes.
An additional corneal abnormality, punctate marginal keratitis,
may be a result of allergic hypersensitivity or, on occasion, a
hypersensitivity reaction to topical drugs such as gentamicin
(the most common offending agent), atropine, anesthetics, and
epinephrine.125 Multiple small perilimbal inltrates are evident
circumferentially with a characteristically clear zone noted
between the inltrates and the limbus.
Cicatrizing conjunctival and keratinizing changes may
develop, particularly in patients using glaucoma medications
and antiviral agents. These changes, which may completely
mimic ocular cicatricial pemphigoid, include punctal occlusion,
canalicular obstruction, fornix and tarsal conjunctiva scarring,
corneal vascularization, and keratinization.122,126129
Conjunctival and Tenons capsule specimens from glaucoma

patients have shown a signicant increase in the number of
macrophages, lymphocytes, mast cells, and broblasts, whereas
goblet cells were decreased in patients who took at least two
glaucoma medications for 1 year.130 This data suggest that
glaucoma therapy may induce inflammation and may worsen
the prognosis for future glaucoma surgery.
Drug-induced ocular cicatricial pemphigoid or drug-induced
pseudopemphigoid has been reported by a number of
investigators.126129 Topical drugs implicated in this reaction
include miotics (echothiophate iodide, pilocarpine), sympathetic agents (dipivefrin hydrochloride), b-blockers (timolol
maleate), and antivirals (idoxuridine, trifluorothymidine).
Fortunately, this side effect occurs rarely.
Numerous studies have to date been unable to differentiate
between cicatricial pemphigoid and drug-induced cicatricial
pemphigoid on the basis of histopathologic, ultrastructural, or
immunofluorescent criteria.126,127,131133 Tauber reported a 26%
incidence of glaucoma in their patients with cicatricial
pemphigoid (29 of 111 patients).134 In this study, 27 of 29
patients had a history of glaucoma medication use, suggesting
that long-standing glaucoma therapy may induce or increase
the susceptibility of certain individuals to the development of
ocular cicatricial pemphigoid.
SECTION 6
REFERENCES
622
1. Davies RJ, Rusnak C, Devalia JL: Why is

allergy increasing? Environmental factors.
Clin Exp Allergy 1998; 28:814.
2. Howarth PH: Is allergy increasing? early
life influences. Clin Exp Allergy
1998;28(Suppl 6):27.
3. King MR, Ismail AS, Davis LS, Karp DR:
Oxidative stress promotes polarization of
human T cell differentiation toward a T
helper 2 phenotype. J Immunol 2006;
176:27652772.
4. Lucey DR, Clerici M, Shearer GM: Type 1
and type 2 cytokine dysregulation in human
infectious, neoplastic, and inflammatory
diseases. Clin Microbiol Rev 1996;
9:532562.
5. Wyman M: Autumnal Catarrh (Hay Fever).
Cambridge, England: Hurd & Houghton;
1872.
6. Cooke RA, VanderVeer A: Human
sensitization. J Immunol 1916; 1:201.
7. Irani AA, Butrus SI, Tabbara KF, Schwartz
LB: Human conjunctival mast cells:
distribution of MCT and MCTC in vernal
keratoconjunctivitis and giant papillary
86:3440.
8. Baddeley SM, Bacon AS, McGill JI, et al:
Mast cell distribution and neutral protease
expression in acute and chronic allergic
conjunctivitis. Clin Exp Allergy 1995;
25:4145.
9. Katz HR, Stevens RL, Austen KF:
Heterogeneity of mammalian mast cells
differentiated in vivo and in vitro. J Allergy
Clin Immumol 1985; 19:143155.
10. Allansmith MR, Baird RS, Greiner JV: Vernal
conjunctivitis and contact lens-associated
giant papillary conjunctivitis compared and
contrasted. Am J Ophthalmol 1979; 87:544.
11. Metzger H, Bach MK: The receptors for IgE
in mast cells and basophils: Studies on IgE
binding and on the structure of the
receptor. In: Bach MK, ed. Immediate
hypersensitivity: modern concepts
developments. New York: Dekker; 1978.
12. Henriquez AS, Kenyon KR, Allansmith MR:

Mast cell ultrastructure: comparison in
contact lens associated giant papillary
conjunctivitis and giant papillary
conjunctivitis. Arch Ophthalmol 1981;
99:12661272.
13. Allansmith MR: Immunology of the eye. In:
The Eye and Immunology. St Louis: CV
Mosby; 1982.
14. Udell IJ, Abelson MB: Chemical mediators
of inflammation. Int Ophthalmol Clin 1983;
23:15.
15. Weston JH, Udell IJ, Abelson MB: H1
receptors in the human ocular surface.
20(Suppl):32.
16. Abelson MB, Udell IJ: H2 receptors in the
human ocular surface. Arch Ophthalmol
1981; 99:302.
17. Abelson MB, Allansmith MR: Histamine and
the eye. In: Silverstein AM, OConnor GR,
eds. Immunology and immunopathology of
the eye. New York: Masson; 1979.
18. Abelson MB, Baird RS, Allansmith MR: Tear
histamine levels in vernal conjunctivitis and
ocular inflammations. Ophthalmology 1980;
87:812.
19. Berdy GJ, Levene RB, Bateman ST:
Identication of histaminase activity in
human tears after conjunctival antigen
challenge. Invest Ophthalmol Vis Sci 1990;
31(Suppl):65.
20. Abelson MB, Leonardi AA, Smith LM, et al:
Histaminase activity in patients with vernal
keratoconjunctivitis. Ophthalmology.
1995;102:19581963.
21. Tamura N, Agrawal D, Suliman FA, et al:
Effects of platelet-activating factor on the
chemotaxis of normodense eosinophils
from normal subjects. Biochem Biophys
Res Commun 1986; 142:638.
22. Udell IJ, Gleich GJ, Allansmith MR, et al:
Eosinophil granule major basic protein
and Charcot-Leyden crystal protein in
human tears. Am J Ophthalmol 1981;
92:824.
23. Trocme SD, Kephart GM, Allansmith MR,

et al: Conjunctival deposition of eosinophil
granule major basic protein in vernal
keratoconjunctivitis and contact lensassociated giant papillary conjunctivitis. Am
24. Abelson MB, Chapin MJ: Current and
future topical treatments for ocular allergy.
Comp Ophthalmol Update 2000;
1:303317.
25. Schoenwetter WF, Dupclay L Jr,
Appajosyula S, Botteman MF, Pashos CL:
Economic impact and quality-of-life burden
of allergic rhinitis. Curr Med Res Opin
2004; 20:305317.
26. Berger W, Abelson MB, Gomes PJ, et al:
Effects of adjuvant therapy with 0.1%
olopatadine hydrochloride ophthalmic
solution on quality of life in patients with
allergic rhinitis using systemic or nasal
therapy. Ann Allergy Asthma Immunol 2005;
95:361371.
27. Arbes SJ, Cohn RD, Yin M, Muilenberg ML,
et al: Dog allergen (Can f 1) and cat
allergen (Fel d 1) in US homes: results from
the National Survey of Lead and Allergens
in Housing. J Allergy Clin Immunol; 2004;
114:111117.
28. Abelson MB, Udell IJ, Weston JH:
Conjunctival eosinophils in allergic ocular
disease. Arch Ophthalmol 1983; 101:631.
29. Leonardi A, Fregona IA, Gismondi M, et al:
Correlation between conjunctival
provocation test (CPT) and systemic
allergometric tests in allergic disease. Eye
1990; 4:760.
30. Abelson MB, Smith LM, Chambers W:
Conjunctival allergen challenge: a clinical
approach to studying allergic conjunctivitis.
31. Abelson MB, Paradis A, George MA, et al:
The effects of Vasocon-A in the allergen
challenge model of acute conjunctivitis.
32. Allansmith MR, Baird RS, Henriquez AS:
Regeneration of mast cells after ocular
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
anaphylaxis. In: Symposium on

Immunology and Immunopathology of the
Eye. New York: Masson; 1985.
Allansmith MR, Baird RS, Bloch KJ: Effect
of anaphylaxis on conjunctival goblet cells.
Allansmith MR, Baird RS, Greiner JV, et al:
Late-phase reactions in ocular anaphylaxis
in the rat. J Allergy Clin Immunol 1984;
73:49.
Allansmith MR, Baird RS, Henriquez AS,
et al: Sequence of mast cell changes in
ocular anaphylaxis. Immunology 1983;
49:281.
Allansmith MR, Bloch KJ, Baird RS, et al:
Ocular anaphylaxis: induction by local
injection of antigen. Immunology 1981;
44:623.
Bonini S, Bonini S, Vecchione A, et al:
Inflammatory changes in conjunctival
scrapings after allergen provocation in
humans. J Allergy Clin Immunol 1988;
82:462.
Leonardi A, Briggs RM, Bloch KJ, et al:
Clinical patterns of ocular anaphylaxis in
guinea pigs passively sensitized with IgG1
antibody. Ophthalmic Res 1990; 22:95.
Bonini ST, Bonini SE, Todini V, et al:
Persistent inflammatory changes in the
conjunctival late phase reaction of humans.
29(Suppl):230.
Gomes PJ, Yunes S, Welch D, Abelson MB:
Evaluation of the effect of multiple doses of
PredForte (prednisolone acetate 1%) in the
prevention of conjunctival allergen
challenge-induced ocular allergic signs and
symptoms. Invest Ophthalmol Vis Sci 2006;
47: 4978.
Lichtenstein SJ, Abelson MB:
Pharmacology, clinical efcacy and safety
of olopatadine hydrochloride. Expert Rev
Clin Immunol 2006; 2:341351.
Smith LM, Southwick PC, DeRosia DR,
et al: The effects of levocabastine, a new
highly potent and specic histamine H1
receptor antagonist, in the ocular allergen
challenge model of allergic conjunctivitis.
Abelson MB, George MA, Schaefer K,
Smith LM: Evaluation of new ophthalmic
antihistamine 0.05 percent levocabastine in
the clinical allergen challenge model of
allergic conjunctivitis. J Allergy Clin
Immunol 1994; 94(3 Pt 1):458464.
Kay AB, Walsh GM, Moqbel R: Disodium
cromoglycate inhibits activation of human
inflammatory cells in vitro. J Allergy Clin
Immunol 1987; 80:1.
Foreman JC: Substance P and calcitonin
gene-regulated peptide: Effects on mast
cells and in human skin. Int Arch Allergy
Appl Immunol 1987; 82:366.
Abelson MB, Berdy GJ, Mundorf T, et al:
Pemirolast study group. Pemirolast
potassium 0.1% ophthalmic solution is an
effective treatment for allergic
conjunctivitis: a pooled analysis of two
prospective, randomized, double-masked,
placebo-controlled, phase III studies.
J Ocul Pharmacol Ther 2002;
18:475488.
Stockwell A, Easty D: Double-blind group
comparative trial of 2 percent nedocromil
sodium eye drops with placebo eye drops
in the treatment of seasonal allergic
conjunctivitis. Invest Ophthalmol Vis Sci
1988; 29:229.
48. Gonzalez JP, Brogden RN: Nedocromil

sodium: a preliminary review of its
pharmacodynamic and pharmacokinetic
properties, and therapeutic efcacy in the
treatment of reversible obstructive airway
disease. Drugs 1987; 34:560.
49. Verin PH, Dicker ID, Mortemousque B:
Nedocromil sodium eye drops are more
effective than sodium cromoglycate eye
drops for the long-term management of
vernal keratoconjunctivitis. Clin Exp Allergy
1999, 29:529536.
50. Collum LMT: A three-month clinical
evaluation of topical lodoxamide
tromethamine for the control of allergic
conjunctivitis. Twenty-sixth International
Congress of Ophthalmology Abstracts
SP-177, 1990.
51. Fahy G, Easty DL, Collum L, et al: Doublemasked efcacy and safety evaluation of
lodoxamide 0.1 percent ophthalmic
solution versus opticrom 2 percent-A
multi-centre study. Ophthalmol Today 341,
1988.
52. Coca AF, Cooke RA: On the classication
of the phenomena of hypersensitiveness.
J Immunol 1923; 8:163.
53. Hogan MJ: Atopic keratoconjunctivitis. Am
54. Allansmith MR, Abelson MB: Immunologic
diseases. In: Smolin G, Thoft RA, eds. The
cornea. Boston: Little, Brown; 1983.
55. Garrity JA, Liesegang TJ: Ocular
complications of atopic dermatitis. Can J
56. Rich LF, Hanin JM: Ocular complications
of atopic dermatitis and other eczemas. Int
57. Ehlers AH, Donshik PC: Allergic ocular
disorders: a spectrum of diseases. CLAO J
1992; 18:117.
58. Foster CS, Cologne M: Atopic
97:992.
59. Tuft SJ, Kemeny DM, Dart JK, et al: Clinical
features of atopic keratoconjunctivitis.
60. Terry JE: Diseases of the cornea. In:
Bartlett JD, Jaanus SD, eds. Clinical ocular
pharmacology. 2nd edn. Boston:
Butterworths; 1989.
61. Friedlaender MH: Allergy and immunology
of the eye. San Francisco: Harper & Row;
1979.
62. Trocme AD, Raizman AM, Bartley GB:
Medical therapy for ocular allergy. Mayo
Clin Proc 1992; 67:557.
63. Morgen SJ, Williams JH, Walls AF, et al:
Mast cell hyperplasia in atopic
keratoconjunctivitis. Eye 1991; 5:729.
64. Foster CS, Rice BA, Dutt JE:
Immunopathology of atopic
98:1190.
65. Soukasian SH, Rice B, Foster CS, et al:
The T cell receptor in normal and inflamed
conjunctiva. Invest Ophthalmol Vis Sci
1992; 33:453.
66. McGeady SJ, Buckley RH: Depression of
cell-mediated immunity in atopic eczema.
67. Rachelefsky GS: Defective T cell function in
atopic dermatitis. J Allergy Clin Immunol
1976; 57:569.
68. Lobitz WC, Honeyman JF, Winkler NW:
Suppressed cell-mediated immunity in two
adults with atopic dermatitis. Br J Dermatol
1972; 86:317.
69. McDevitt HO, Chinitz A: Genetic control of

the antibody response: relationship
between immune response and
histocompatibility (H2) type. Science 1969;
163:1207.
70. McDevitt HO, Landy M: Genetic control of
immune responsiveness. New York:
Academic Press; 1972.
71. Schreffler DC, David CS: The H2 major
histocompatibility complex and the I
immune response region: genetic variation,
function and organization. Adv Immunol
1975; 20:125.
72. Ishizaka K: Regulation of IgE synthesis. In:
Debelic M, ed. 20 Years with IgE-New
Prospects. Kingston Upon Thames,
England, Medicom, 1987.
73. Goen TM, Sieboldt K, Terry JE: Cromolyn
sodium in ocular diseases. J Am Optom
Assoc 1986; 57:526.
74. Allansmith MR, Ross RN: Ocular allergy
and mast cell stabilizers. Surv Ophthalmol
1986; 30:229.
75. Vajpayee RB, Gupta SK, Uppal RK:
Evaluation of sodium cromoglycate and
medrysone in the management of atopic
keratoconjunctivitis: a double-masked
clinical study. ANZ J Ophthalmol 1986;
14:251.
76. Welch D, Ousler G, Abelson M, Wilcox K:
Ocular drying associated with oral
antihistamines (loratadine) in the normal
population an evaluation of exaggerated
dose effects. Lacrimal Gland, Tear Film,
and Dry Eye Syndromes 2002;
3:10511055.
77. Ousler GW, Wilcox KA, Gupta G, Abelson
MB: An evaluation of the ocular drying
effects of 2 systemic antihistamines:
loratadine and cetirizine hydrochloride.
Ann Allergy Asthma Immunol 2004;
93:460464.
78. Rikkers SM, Holland GN, Drayton GE, et al:
Topical tacrolimus treatment of atopic
eyelid disease. Am J Ophthalmol 2003;
135:297302.
79. Akpek EK, Dart JK, Watson S, et al:
A randomized trial of topical cyclosporin
0.05% in topical steroid-resistant atopic
111:476482.
80. Beigelman MN: Vernal conjunctivitis.
New York: University of Southern California
Press; 1950.
81. Allansmith MR, Frick OL: Antibodies to
grass in vernal conjunctivitis. J Allergy
1963; 34:535.
82. Dahan E, Appel R: Vernal conjunctivitis in
the black child and its response to therapy.
83. Neumann E, Gutmann MJ, Blumenkrantz N,
Michaelson IC: A review of four hundred
cases of vernal conjunctivitis. Am J
84. Hennawi ML: Clinical trial with 2 percent
sodium cromoglycate (Opticrom) in vernal
64:483.
85. Trantas: Sur le catarrhe printanier. Arch
Ophthalmol (Paris) 1910; 30:593.
86. Allansmith MR, Baird RS: Percentage of
degranulated mast cells in vernal
conjunctivitis and giant papillary
conjunctivitis associated with contact lens
wear. Am J Ophthalmol 1981; 91:71.
87. Hogan HJ, Zimmerman LE: Ophthalmic
pathology. 2nd edn. Philadelphia: WB
Saunders; 1962.
CHAPTER 47
623
SECTION 6
624
88. Leonardi A, Abatangelo G, Cortivo R,

Secchi AG: Collagen types I and III in giant
papillae of vernal keratoconjunctivitis. Br J
Ophthalmol 1995; 79:482485.
89. Abelson MB, Soter NA, Simon MA, et al:
Histamine in human tears. Am J
90. Duke-Elder S: Allergic keratoconjunctivitis.
In: Duke-Elder S, ed. System of
ophthalmology, diseases of the outer
eye. Part 1. London: Henry Kimpton;
1965:475.
91. Neumann E, Blumenkrantz N:
Mucopolysaccharide in the secretion of
vernal conjunctivitis: Its use as a diagnostic
test. Br J Ophthalmol 1959; 43:46.
92. Hosford GN, Hicks AM: Hydrogen ion
concentration of tears: its relation to certain
ocular symptoms and to conjunctival and
corneal lesions. Arch Ophthalmol 1935;
13:14.
93. Abelson MB, Sadun AA, Udell IJ, et al:
Alkaline tear pH in ocular rosacea. Am J
94. Ballow M, Mendelson L: Specic
immunoglobulin E antibodies in tear
secretions of patients with vernal
66:112.
95. Mukhopadhyay K, Pradhan SC, Mathur JS,
et al: Studies on histamine and histaminase
in spring catarrh (vernal conjunctivitis). Int
Arch Allergy Appl Immunol 1981; 64:464.
96. Abelson MB, Leonardi AA, Smith LM, et al:
Histamine activity in patients with vernal
102:19581963.
97. Sompolinsky D, Samra Z, Zavaro A, et al:
Allergen-specic immunoglobulin E
antibodies in tears and serum of vernal
conjunctivitis patients. Int Arch Allergy Appl
Immunol 1984; 75:317.
98. Butrus SI, Ochsner KI, Abelson MB, et al:
The level of tryptase in human tears.
99. Golubovic S, Parunovic A: Vernal
conjunctivitis A cause of corneal mucoid
plaques. Fortschr Ophthalmol 1986;
83:272.
100. Abelson MB, Allansmith MR, Friedlaender
MH: Effects of topically applied ocular
decongestant and antihistamine. Am J
101. Abelson MB, Butrus SI, Weston JH, et al:
Tolerance and absence of rebound
vasodilation following topical ocular
decongestant usage. Ophthalmology 1984;
91:1364.
102. Corum I, Yeniad B, Bilgin LK, Ilhan R:
Efciency of olopatadine hydrochloride
0.1% in the treatment of vernal
keratoconjunctivitis and goblet cell density.
J Ocul Pharmacol Ther 2005; 21:400405.
103. Foster SC, Duncan BA: Randomized

clinical trial of topically administered
cromolyn sodium for vernal
1980; 90:175.
104. Tabbara KF, Arafat NT: Cromolyn effects of
vernal keratoconjunctivitis in children. Arch
105. Mikuni I, Fujiwara T, Togawa K, et al:
Therapeutic effects of a new, anti-allergic
ophthalmic preparation. Tokai J Exp Clin
Med 1982; 7:279.
106. BenEzra D, Peer J, Brodsky M, et al:
Cyclosporine eyedrops for the treatment of
severe vernal keratoconjunctivitis. Am J
107. Secchi AG, Tognon MS, Leonardi A: Topical
use of cyclosporine in the treatment of
108. Allansmith MR, Korb DR, Griener JV, et al:
Giant papillary conjunctivitis in contact lens
wearers. Am J Ophthalmol 1977; 83:697.
109. Greiner JV: Giant papillary conjunctivitis. In:
Abelson MB, ed. Allergic diseases of the
eye. Philadelphia: WB Saunders;
2001:144164.
110. Korb KJ, Greiner JV, Finnemore VM,
Allansmith MR: Biomicroscopy of papillae
associated with soft contact lens wearing.
111. Hart DE, Schkolnick JA, Bernstein S, et al:
Contact lens-induced giant papillary
conjunctivitis: a retrospective study. J Am
Optom Assoc 1989; 60:195.
112. Begley CG, Riggle A, Tuel JA: Association
of giant papillary conjunctivitis with
seasonal allergies. Optom Vis Sci 1990;
67:192.
113. Jolson AS, Jolson SC: Suture barb giant
papillary conjunctivitis. Ophthalmic Surg
1984; 15:139.
114. Key JE II: Are hard lenses superior to soft?
Arguments in favor of hard lenses. Cornea
1990; 9(Suppl 1):S9; discussion S15.
115. Abelson MB, Richard K: What we know
and dont know about GPC. Rev
Ophthalmol 1994; 1:664665.
116. Greiner AV, Korb DR, Allansmith MR:
Pathogenesis of contact lens papillary
conjunctivitis: A hypothesis. In: OConnor
GR, Chandler AW, eds. Advances in
immunology and immunopathology.
Seattle: Masson; 1985:302304.
117. Rapacz P, Tedesco J, Donshik PC, Ballow
M: Tear lysozyme and lactoferrin levels in
giant papillary conjunctivitis and vernal
conjunctivitis. CLAO J 1988; 14:207.
118. Ballow M, Donshik PC, Rapacz R,
Samartino L: Tear lactoferrin levels in
patients with external inflammatory ocular
disease. Invest Ophthalmol Vis Sci 1987;
28:543.
119. Trocme SD, Kephart GM, Allansmith MR,

et al: Conjunctival deposition of eosinophil
granule major basic protein in vernal
keratoconjunctivitis and contact lensassociated giant papillary conjunctivitis. Am
120. Donshik PC, Ballow M: Tear
immunoglobulins in giant papillary
conjunctivitis induced by contact lenses.
121. Jones B, Sack R: Immunoglobulin
deposition on soft contact lenses:
relationship to hydrogel structure and mode
of use and giant papillary conjunctivitis.
CLAO J 1990; 16:43.
122. Wilson FM II: Adverse external ocular
effects of topical ophthalmic medications.
123. Wilson FM: Adverse external ocular effects
of topical ophthalmic therapy: an
epidemiological, laboratory and clinical
study. Trans Am Ophthalmol Soc 1983;
81:854965.
124. Schwab IR, Abbott RL: Toxic ulcerative
keratopathy. An unrecognized problem.
125. Stern GA, Killingsworth DW: Complications
of topical antimicrobial agents. Int
126. Patten JT, Cavanagh HD, Allansmith MR:
Induced ocular pseudopemphigoid. Am J
127. Lass JH, Thoft RA, Dohlman CH:
Idoxuridine-induced conjunctival cicatrization.
128. Udell IJ: Trifluridine-associated conjunctival
cicatrization. Am J Ophthalmol 1985;
99:363.
129. Kristensen EB, Norn MS: Benign mucous
membrane pemphigoid. 1: Secretion of
mucus and tears. Acta Ophthalmol
(Copenh) 1974; 52:266.
130. Sherwood MB, Grierson I, Millar L: Longterm morphologic effects of antiglaucoma
drugs on the conjunctiva and Tenons
capsule in glaucomatous patients.
131. Pouliguen Y, Patey A, Foster CS: Druginduced cicatricial pemphigoid affecting the
conjunctiva. Ophthalmology 1986; 93:775.
132. Leonard JN, Hobday CM, Haffenden GP:
Immunofluorescent studies in ocular
cicatricial pemphigoid. Br J Dermatol 1988;
118:209.
133. Foster CS: Cicatricial pemphigoid. Trans
134. Tauber J, Melamed S, Foster CS:
Glaucoma in patients with ocular cicatricial
pemphigoid. Ophthalmology 1989; 96:33.
CHAPTER
48
Lid Inflammations
Audrey S. Chan and Kathryn A. Colby
Lid inflammation or blepharitis, a common problem in

ophthalmic practice,1 is also a frequent cause for visits to
physicians. In a general practice, 2.3% of visits were for ocular
problems, 70% of which were diagnosed as bacterial conjunctivitis, allergic conjunctivitis, meibomian cyst, or
blepharitis.2 Blepharitis is a condition that can often be
controlled but never permanently cured, because the underlying
factors of sebaceous gland dysfunction and skin flora cannot be
permanently and irrevocably altered. Chronic blepharitis can
lead to severe dry eye, trichiasis, lid notching, reduced corneal
sensation, corneal scarring with neovascularization and
marginal keratitis. There is considerable amount of overlap of
symptoms among blepharitis and other inflammatory disorders
affecting the lids.
ANATOMY OF SEBACEOUS GLANDS IN

THE EYELID
Sebaceous glands are present in the eyelids as both meibomian
glands and the glands of Zeis. The sebaceous glands in the lid
are embryologically derived from a common pilosebaceous unit
that differentiates during the second month of gestation. Unlike
the glands of Zeis, which are associated with cilia, meibomian
glands are modied sebaceous units that lack hair follicles.
They are vertically oriented in parallel rows through the tarsus.
There are ~2022 meibomian glands on the lower lids and
2224 on the upper lids. The glands of Zeis are located on the
lid margin anterior to the opening of the meibomian glands.
Absence of meibomian glands may be a marker for ectodermal
dysplasia syndrome.3
A disruption in normal meibomian gland function leads to a
variety of disorders that affect the ocular surface, the most
common of which is blepharitis, which causes secondary
evaporative dry eye. The meibomian glands play a critical role
in maintaining tear lm homeostasis and stability. The
meibomian secretions produce the oily outer layer of the tear
lm, which prevents evaporation of the aqueous tear layer and
creates an optically smooth surface. The structural and
refractive integrity of the ocular surface depends on the quality
of meibomian secretions. The symptoms of ocular burning,
stinging, and irritation are the result of tear lm instability
causing evaporative dry eye. In this chapter the treatment and
diagnosis of the many disorders stemming from lid
inflammation will be discussed.
COMPOSITION OF MEIBOMIAN GLAND

SECRETIONS
Sebaceous glands are holocrine glands. Each meibomian gland
consists of several acini connected by a long central duct that
opens at the lid margin. Each acinus is lined by cuboidal

epithelium that houses storage granules containing lipid
material. When the cell becomes engorged with lipid, the
nucleus of the cell becomes pyknotic and the apex of the cell
membrane ruptures into the lumen of the gland opening; the
cell spills forth its lipid and cellular contents into the duct.
The lipid material flows to the orice of the gland and onto the
skin or into the tear lm. The formation of sebum is dependent
on cellular proliferation. There is also increasing evidence
that meibomian gland secretion is modulated via neuronal,
hormonal and vascular influences.4,5 Vasoactive intestinal
peptide (VIP) innervation has been shown to be present in acinar
cells and also provides innervation to the lacrimal glands.6
Studies of meibomian gland growth and differentiation in cell
culture may provide better understanding of meibomian gland
function.7,8
Stagnation of meibomian gland secretion results in several
histopathologic features. Features of meibomian gland dysfunction include signs of obstruction and dilatation of the ducts,
enlargement of acini with cystic degeneration and squamous
metaplasia, foreign-body reaction and granuloma formation,
a mild increase in inflammatory cells, and abnormal
keratinization.9 Obstruction of the meibomian gland ducts with
stagnation of secretions may cause increased pressure within
the ducts, thus inhibiting cellular differentiation and causing
squamous metaplasia. Keratinization of the meibomian gland
ductal epithelium may be the initiating event.912
The chemical composition of meibum has been extensively
studied. Meibomian gland secretion is distinct from other forms
of sebum in that it has a relatively low melting point of
1932C, which allows the secretion to remain fluid at lid
temperature.13 Approximately 84% of meibum is composed of
nonpolar lipid wax esters and sterol esters. Cholesterol is the
main esteried ester with relatively longer carbon chains of 20
carbons or more in comparison to wax esters.14 Triglycerides
and free fatty acids are found in smaller amounts in meibum.
Unsaturated fatty acids are particularly important to the
maintenance of normal meibum properties. Solidied paste-like
meibum such as the type often found in blepharitis patients
contains relatively low concentrations of unsaturated fatty
acids.15 These lipid abnormalities may account for many of the
signs and symptoms of chronic blepharitis, such as tear-lm
instability and inspissation of secretions.
One study conducted by Shine and McCulley investigated the
composition of polar lipids from patients with meibomianitis.
They found that patients with meibomianitis had higher
concentrations of an unknown type of polar lipid and polar
lipids with unsaturated fatty acids and amide acyl chains when
compared with normal patients or patients with other forms of
chronic blepharitis. From these results they hypothesized that
625
CORNEA & CONJUNCTIVA
FIGURE 48.1. Hordeolum. There is focal inflammation and abscess formation around the mouth of a meibomian gland (a) and around a lash
follicle/gland of Zeis (b). The abscesses may be single or multiple (c), involving more than one lid.
changes in the polar lipid phase of the tear lm could lead to

tear-lm instability and thus dry eye symptoms.16
The lipid abnormalities may be physiologic or induced by
bacterial lipases.17 Work has also been directed toward analysis
of meibomian secretions and interactions with bacterial
lipases.1821 S. epidermidis and S. aureus produce triglyceride
lipase and cholesterol and wax esterase. These exoenzymes
hydrolyze wax and sterol into free fatty acids thereby altering
the composition of meibum. Increased levels of free cholesterol
from esterase activity have been found to stimulate proliferation
of S. aureus in vitro.22 Low levels of tetracyclines used for the
treatment of blepharitis have been found to inhibit bacterial
lipases reducing free fatty acid production thereby changing the
composition of meibum and inhibiting S. aureus proliferation.23
INFLAMMATIONS OF THE SEBACEOUS

GLANDS
SECTION 6
HORDEOLUM
626
An acute focal inflammation of the eyelid may occur when a

meibomian or a Zeis gland becomes infected (Fig. 48.1). This is
called a hordeolum, or in the vernacular, a stye. It represents
an acute pyoderma. The most common bacterial cause is
Staphylococcus aureus. The process is usually self-limited,
with spontaneous drainage of the abscess and resolution within
57 days. Warm compresses are helpful in localizing the
inflammation. More rapid drainage of a hordeolum can be
promoted by nicking the pustule at the mouth of the occluded
orice using the sharp tip of a needle or blade and then applying
focal pressure to express the pus (Fig. 48.2). No anesthesia is
required. Systemic antibiotics are unnecessary unless there is
signicant cellulitis, in which case a semisynthetic penicillin,
erythromycin, or clindamycin should be administered.
Preseptal and orbital cellulitis are discussed in Chapter 239.
CHALAZION
A chalazion is a granuloma that develops around a sebaceous
gland in the lids as a foreign-body reaction to sebum released
into the surrounding tissue. It may evolve from a hordeolum
or may occur secondary to inspissation of sebum at the opening of a gland with engorgement and rupture of the gland
contents into the surrounding tissue (Fig. 48.3). Histopathologic evaluation of chalazion contents reveals histiocytes,
multinucleated giant cells, lymphocytes, plasma cells,
polymorphonuclear leukocytes, and eosinophils.24 The acute
inflammatory process may be intense, creating enormous lid
edema that may spread to the opposite lid and sometimes
across the bridge of the nose to the lids of the other eye. The
local lymphatic congestion can be differentiated from cellulitis
FIGURE 48.2. A hordeolum can be drained by nicking the pustule at

the gland opening and expressing the purulent contents.
by the absence of tense erythema, pain, or leukocytosis. The

focal inflammation around the involved gland may cause
pointing of the lesion through the skin anteriorly or into the
subconjunctivaltarsal space, where it may eventually drain
spontaneously or persist as a chronically inflamed granuloma.
In the chronic phase, a chalazion may appear as a quiet but
large swelling of the lid. Large lesions of the upper lid may cause
distortion of vision from induced astigmatism. As with
treatment of a hordeolum, warm compresses are useful in
trying to localize the inflammation and cause spontaneous
drainage. Topical antibiotics cannot directly affect the inflammation inside the gland but are an adjunctive therapy in trying
to decrease the local bacterial flora. Many chalazions, especially
if they are small and of short duration, will be cured or
improved within a month of medical treatment (Table 48.1).25
Rosacea is common in patients with chalazions.26
Treatment of the chalazion in the chronic phase consists of
surgical drainage or intralesional steroids. When the transconjunctival surgical approach is used, it is important to make the
incision(s) perpendicular to the lid margin, parallel to the orientation of the meibomian glands. Horizontal conjunctivaltarsal
Lid Inflammations
incisions will create scar tissue across the ducts of the

meibomian glands, resulting in blockage of sebum in the
proximal ducts. The sebum then backs up, causing rupture of
the ducts and more granuloma formation, which perpetuates
rather than cures the problem. Incisions can be made in the
skin, especially if the chalazion has already ruptured through
the skin. The incision in the skin should be parallel to the lid
margin to minimize visible scarring. Excision of a chalazion
using a carbon dioxide laser and curettage has been reported.27
Triamcinolone acetonide, a soluble aqueous steroid suspension, in a concentration of 5 mg/mL, can be injected directly
into the center of a chalazion. The total volume varies from
0.05 to 0.2 mL, which is injected transdermally or directly
perpendicularly through the conjunctiva and tarsus.24 The
advantage to the latter approach is that it decreases the risk
of dermal depigmentation and atrophy, which sometimes
accompany intradermal use of fluorinated corticosteroids.28,29
The injections can be repeated anywhere from 2 days to
1 month apart if the initial injection does not result in complete resolution of the lesion. Success rates vary from 76% to
93%.24,3032 A recent retrospective case series from Jules Stein
Eye Institute reported an 83% success rate after triamcinolone
acetonide injection of primary or recurrent chalazia.33 Hard
lesions present for more than 6 months are less likely to
respond.31 Accidental intravitreal injection of a steroid,
resulting in macular pucker and optic atrophy, has been
reported.34
Care must be taken not to mistake sebaceous cell carcinoma
of the eyelid for recurrent chalazions. If any doubt exists,
especially in an older patient, the excised material should be
sent for histopathologic examination.3537
BLEPHARITIS
Key Features
Blepharitis is a common chronic eye disease.

The goal of treatment is management of symptoms; the
condition cannot be cured.
New treatment for dry eye address the inflammation of the
ocular surface now known to play a role in this condition.
Classication of Blepharitis
Blepharitis can be broadly classied anatomically as either
anterior or posterior, anterior blepharitis comprising staphylococcal and seborrheic forms and posterior blepharitis
primarily involving the meibomian glands. Many other
classication schemes have been proposed based on clinical
symptoms and ndings.38
Mathers and Choi evaluated 513 patients to create a classication tree to separate blepharitis and dry eye conditions into
clinically relevant groups based on objective physiologic
measurements.39 According to their classication scheme, nine
categories were established based on meibomian gland drop
out, lipid viscosity, lipid volume, Schirmer testing, and tear
evaporation. Interestingly, the presence or absence of bacteria
was not included as one of their objective measurements,
arguing that the mere presence of bacteria does not necessarily
indicate infection. Although Staphylococcus aureus can be
found more frequently in blepharitis,40 they contend that
bacterial infection alone may not be causative, but instead may
secondarily exert some effect on meibomian glands resulting in
blepharitis.
CHAPTER 48
FIGURE 48.3. A chalazion is a granulomatous

reaction to the inspissated secretions of
meibomian gland. It appears as a swollen
tumor mass involving the eyelid. It may be
associated with local inflammation around the
mouth of a single gland (a). It may show no
external signs of inflammation (b), with only a
slight engorgement of vessels on the tarsal
conjunctiva (c). Chalazions may be very small,
presenting as only small granulomas at the
posterior lid margin (d). They may involve
multiple glands and different lids (b). They may
cause blurring of vision owing to induced
astigmatism from the pressure of the mass on
the cornea. The patient in b and c complained
of visual acuity that was reduced to 20/25; this
returned to 20/15 when the chalazion was
excised.
TABLE 48.1. Treatment Options for Chalazions

Treatment
Duration
Infection
Size
Hardness
Compresses
Less than 6 wk
Yes
Smallmedium
Soft
Intralesional
steroids/incision
2 wk6 mo
No/yes
Smalllarge
Rubbery
Incision/drainage
Longer than 6 mo
Yes
Smalllarge
Hard
627
SECTION 6
TABLE 48.2. Clinical Spectrum of Blepharitis
628
Type
Major Feature
Common Associations
Dry Eye
Microbiology
Staphylococcal
Acute inflammation
80% female
50% keratitis sicca
Staphylococcus aureus
common
Seborrheic blepharitis
Oily, greasy scales

around lashes
Spotty glandular
involvement
Meibomian seborrhea
Excess secretions
from glands
Bacterial flora within

normal limits
Seborrhea
Solidied with plugged

secretions
Bacterial flora within

normal limits
Rosacea
Facial dermal involvement
Seborrhea/rosacea
Chalazions
Keratitis
Using cluster analysis Mathers and Choi were able to devise

a decision tree to place patients into distinct categories based on
the results of the objective tests. Meibomian gland drop out was
found in only two groups, which they identied as (1) rosacea
and (2) obstructive meibomian gland dysfunction (MGD)
dry-eye patients. Both of these groups had obstructive MGD but
differed by lipid volume and viscosity. According to the cluster
analysis, seborrheic MGD was divided into three groups, with
only one group having a low tear-evaporation rate. It was
previously thought that seborrheic patients have low tearevaporation rates owing to the presence of excess lipid
secretion. However, in this study the authors found two
subgroups of seborrheic patients with high lipid volumes with
high tear evaporation and dry eye. Evaporation was an
important variable in the study that helped to classify patients
into relevant groups. Through the use of physiologic
parameters, the authors hope to shed insight into the
underlying pathologic mechanisms of blepharitis and dry eye.
McCulley and colleagues suggest categorizing blepharitis
into six groups to facilitate a rational approach to investigation
and therapy17 (Table 48.2). The group that they called
staphylococcal tended to have more acute lid inflammation of
shorter duration. Eighty percent of the group consisted of
women. Keratitis sicca affected 50% of the group. The results
of lid cultures were positive for S. aureus in 46% of patients
versus 15% in the control group. Interestingly, fully 90% of
both control and staphylococcal patients had positive
S. epidermidis cultures.
McCulley and colleagues seborrheic blepharitis group had
oily, greasy scales and crusting around the cilia with spotty
involvement of clusters of glands. A third group had combined
seborrheic and staphylococcal involvement. A fourth group
had seborrheic blepharitis with excess secretions from the
meibomian glands (meibomian seborrhea), and a fth group
had seborrheic blepharitis with secondary inflammation of the
meibomian glands with solidied secretions within the ductules
that were difcult or impossible to express. There was no
increase in recovery of bacteria in these patients compared with
normal controls. McCulley and colleagues nal group was
described as having primary meibomitis with associated
generalized dermal involvement in the form of acne rosacea or
seborrheic dermatitis. These patients, with what McCulley and
colleagues called meibomian keratoconjunctivitis, had a
marked instability of the tear lm. They tended to have the
most severe dry eye signs and symptoms.41
In clinical practice, patients do not tend to fall neatly into a
pigeonhole category (Fig. 48.4) but present on a continuum
between categories. As their blepharitis waxes and wanes over
Yes
years, they may sometimes t into each and every category.

Nevertheless, the concept of multiple types of blepharitis is
useful in sorting out the various components and approaches to
therapy (Table 48.3).
Role of Staphylococcus Aureus

Traditionally, the cause of blepharitis has been attributed to
S. aureus,4244 despite the fact that S. aureus can not always be
recovered on culture.45 However, despite decades of study, its
contribution to blepharitis remains poorly dened. Cultures of
normal lids and those with blepharitis reveal similar, very
frequent colonization with coagulase-negative staphylococcal
species, Propionibacterium acnes, and Corynebacterium
species.4649 S. aureus is not disproportionately represented in
patients with blepharitis,49 unless the group with S. aureus has
been deliberately subdivided from the larger group.17 In one
large study, however, quantitative growth of S. aureus was
signicantly heavier in patients with blepharitis.49
The role of staphylococcal toxins in blepharitis is still not
well established despite extensive research in toxin production.
a-Lysin from bacterial strains isolated from patients with
blepharitis produced dermal necrosis when tested in rabbits,50
but a-lysin was also found to be produced by all isolates of
S. aureus-colonizing lids of normal controls as well as by
patients with blepharitis.51 Enhanced cell-mediated immunity
to S. aureus was demonstrated in 40% of patients with chronic
blepharitis but not among normal controls.52 The same group
of patients was tested with intradermal thiomersal and only an
expected 6% showed cell-mediated immunity to thiomersal,
suggesting that they were not hyperimmune.53
Mondino and coworkers developed a rabbit model of
staphylococcal blepharitis and postulated that hypersensitivity
to the S. aureus cell wall, particularly to ribitol teichoic acid,
plays a role in the pathogenesis of staphylococcal blepharitis.54
Rabbits immunized with cell wall or ribitol teichoic acid
also developed 45 mm of peripheral corneal vascularization,
and several developed corneal phlyctenules or catarrhal
inltrates.54 Subcutaneous vaccination with S. aureus phage
lysate did not prevent development of phlyctenules or
blepharitis in rabbits given topical applications of viable
S. aureus in both eyes. In fact, the control group was less
affected than the vaccinated group.55
DEMODEX
Demodectic mites can be found inhabiting hair follicles
(Demodex folliculorum) and sebaceous glands (D. brevis). Their
role in causing blepharitis has not been well established,12,5659
Lid Inflammations
FIGURE 48.4. (a) This patient with chronic blepharitis shows the typical heavy crusting and scales along the bases of the eyelashes. There is a
fairly uniform swelling to the lids with a chronic spotty redness to the lid margins. (b) Patients with long-standing blepharitis will demonstrate
patchy loss of lashes (madarosis) and whitening or loss of pigmentation in lashes (poliosis), as well as chronic crusting and scale formation along
the bases of the lashes. There may be patchy focal involvement, with some portions of the lids affected more than others. a and b show the
right and left eyes, respectively, of the same patient. (c) Focal pouting of the individual meibomian glands may be seen, accompanied by
telangiectasia of the lid margin. (d) The upper and lower lids may be asymmetrically involved. (e) Patients with chronic staphylococcal
blepharitis frequently have a dry eye, as demonstrated here by rose Bengal staining. They are also susceptible to peripheral marginal ulcers of
the cornea (b and c). It is postulated that hypersensitivity to components of the Staphylococcus aureus cell wall plays a role in the pathogenesis
of staphylococcal blepharitis and peripheral corneal inltrates.
Major Findings
Treatment
Scales and crusts around lashes
Lid hygiene: warm compresses, lid scrubs, dilute shampoo; topical antibiotics with efcacy against
Staphylococcus species
Ocular irritation
Test for dry eye (Schirmer with anesthesia); articial tears; punctal occlusion; Restasis
Focal swelling
Hot compresses; manual expression of glands
Recurrent hordeola and rosacea
Oral tetracycline and derivatives; topical metronidazole
Meibomian gland dysfunction
Warm compresses, oral omega-3 fatty acids, Restasis
but it is tempting to postulate that in heavy infestations they

could cause mechanical plugging of the gland orices and
secondary blepharitis.
D. folliculorum and D. brevis, the hair follicle mites, are the
most common ectoparasites of humans. They are colorless,
spindle-shaped, and only 0.30.4 mm long. The anterior third
of their bodies has four pairs of very short legs, which limits
their mobility. They are almost always found with their
posterior down in the hair follicles, especially in the nasolabial
folds, the nose, and the eyelids. They feed on the cells of the
follicular or sebaceous glandular epithelium by piercing the cell
wall with their convex U-shaped chelicerae. Their complete life
cycle is ~15 days. The female mite lays her eggs deep in the
gland. The larvae are conveyed passively with secreted sebum
into the pilosebaceous canal. Nymphs move in the dark into
another follicle.57,60
Demodex can be identied in normal-appearing eyelids by
epilating lashes and observing the mites clinging to the lashes
under the microscope (Fig. 48.5). They cannot be seen by slit
lamp. They have been noted by electron microscopy beside an
eyelash at the lid margin (Fig. 48.6).61 The presence of
translucent cylinders resembling clear plastic insulation or

cuffs enclosing the base of a lash for a distance of ~1 mm is
suggestive of the presence of Demodex in the follicle. Such
cylinders were demonstrated in 26% of patients with blepharitis
without Demodex and in 44% of patients with Demodex. They
were found in almost 66% of patients with a heavy infestation
of Demodex. There was no correlation with the nding of
scales, hyperemia of the lid margin, clubbed hairs, or itching.57
The incidence of Demodex infestation increases with age. It is
rarely seen in children62 but involves virtually everyone older
than 70 years of age.57,58,62,63 Patients with rosacea and perioral
dermatitis frequently have signicant mite infestation of the
face.64 Some authors attribute the lesions of rosacea to a cellmediated immune response to D. folliculorum because
inflammatory inltrates, including helperinducer T cells, can
be found around the mites.65
Treatment for Demodex of the eyelids is problematic and its
necessity is questionable in the rst place.57,66 Substances
effective in killing the organisms are simply unusable in ophthalmic practice because they are highly toxic, irritating, and
malodorous. Absolute and ethyl alcohol, ether, xylol, acetone,
CHAPTER 48
TABLE 48.3. Treatment of Blepharitis
629
FIGURE 48.5. A Demodex mite is colorless and spindle-shaped. The

anterior section of the body has four pairs of very short legs, which
limits the mites mobility.
From Smolin GR, Tabbara K, Whitcher J: Lids. In: Infectious diseases of the eye.
Baltimore: The Williams & Wilkins Company; 1984.
FIGURE 48.7. Early rosacea may be easily overlooked as being a

ruddy complexion. The ne, blotchy inflammation of the skin over the
malar areas and the nose is typical of early rosacea. There may be
little lid inflammation with slightly increased prominence of
conjunctival veins.
and benzene kill Demodex within minutes.57 Danish ointment,

which contains 14% sulfur as potassium polysuldes, also kills
the organisms within minutes.57 In a recent study,67 treatment
with 2% mercury oxide ointment was reported to be successful
in reducing Demodex concentrations. The mites can survive in
concentrations of metronidazole that are unachievable in
serum.68
SECTION 6
ROSACEA
Rosacea is a very common chronic inflammatory disorder of
the midline facial skin and blush area of the chest, with an
onset mainly between the ages of 30 and 50 years, although it
can also occur during childhood. In ocular rosacea, women are
affected slightly more often than men, but the disease is often
more severe in men. The early stages of rosacea consist of facial
erythema. This may be overlooked as high coloring or a ruddy
complexion (Fig. 48.7). The next stage includes the development of ne telangiectasias, especially around the nose, and
recurrent episodes of inflammatory papules and pustules
(Fig. 48.8). Severe involvement results in facial disgurement
from rhinophyma and markedly dilated supercial telangiectatic blood vessels on the nose, cheeks, and chin (Fig. 48.9).69
Ocular involvement is common, affecting up to 58%.66 Ocular
signs and symptoms may precede signicant skin changes in up
to 20% of cases.70 Keratoconjunctivitis sicca is much more
common in patients with rosacea26 (36.6%) compared to agematched and sex-matched controls (4.1%).71 Patients with
FIGURE 48.8. More pronounced rosacea gives rise to inflammatory

papules and pustules.
Courtesy of Curatek Pharmaceuticals and Arthur Sober, MD.
rosacea have a high incidence of chalazion formation. In a series

of patients older than 19 years of age who were scheduled for
chalazion excision, 57% had rosacea.26
The ocular signs of rosacea are similar to those of chronic
blepharitis with chronic low-grade conjunctivitis and tear-lm
instability giving rise to ocular surface irritation and
FIGURE 48.6. (a) Scanning electron

micrograph of the lower eyelid of a 35-year-old
woman who underwent a full-thickness lidshortening procedure reveals the dome-shaped
tail of a Demodex folliculorum mite contiguous
with an eyelash. (b) Higher magnication
reveals the characteristic annular bands of the
abdomen.
630
(a and b) From English FP, Zhang GW, McManus DP,

Campbell P: Electron microscopic evidence of acarine
infestation of the eyelid margin. Am J Ophthalmol
1990; 109:239240.
Lid Inflammations
From Browning DJ, Proia

AD: Ocular rosacea. Surv
Ophthalmol 1986;
31:145158.
irregularity. Patients with rosacea have a tendency toward

disproportionate conjunctival hyperemia. They frequently
complain of foreign body sensation, burning, tearing or redness
that may be worse toward the end of the day, and contact lens
intolerance. The most common ocular manifestations include
meibomian gland dysfunction, lid telangiectasis, conjunctival
hyperemia, and blepharoconjunctivitis. Corneal changes result
from involvement of the lid and conjunctiva. Initially there
is marginal vascular inltration,72 followed by the formation of
supercial peripheral corneal neovascularization. As the disease
progresses, patients will develop subepithelial inltrates that
appear near the limbus as round, oval or linear in shape. In
severe cases of rosacea, there may be peripheral corneal
vascularization (Figs 48.10a and b), thinning, ulceration, and
even perforation (see Fig. 48.10c and d) with serious visual and
ocular morbidity. Vitritis, which was not explained by other
causes, has been reported in two cases.66
The pathogenesis of rosacea remains unclear although it is
commonly thought of as inflammatory in nature. There is a
genetic predilection, and the disorder is common in people of
Celtic and Northern European ancestry.73 It has been reported
in blacks74,75 and Japanese.76 One hypothesis is that rosacea
represents a dermal dystrophy in which there are degenerative

changes in perivascular collagen that lead to small vessel dilatation and eventually to incompetence of the vessels. Subsequent leakage of potentially inflammatory substances into the
perivascular space leads to lymphedema and to the formation of
papules, pustules, and lupoid nodules.77,78 Recent studies have
documented increased levels of proinflammatory mediators
such as interleukin 1-alpha and matrix metalloproteinase-9
in the tears of rosacea patients compared with age-matched
controls, which seem to support the theory that rosacea is
inflammatory in nature.7981
Patients with rosacea are also thought by some to have
lability of vascular regulatory mechanisms, accounting for
their tendency to flush. Patients with rosacea are twice as likely
to have a migraine compared with a control group.82,83
Histopathologic study of conjunctivae in subjects with rosacea
shows attenuation and inltration by inflammatory cells,
mainly helperinducer T (CD4) cells, phagocytic cells, and
antigen-presenting (CD14, Mac-1) cells. The substantia propria
corneae contains large subepithelial inltrates of inflammatory
cells and sometimes granulomas. The mechanism involved
resembles a type IV hypersensitivity reaction.84 Other studies
linking rosacea to Helicobacter pylori are inconclusive.
TREATMENT OF ROSACEA AND CHRONIC

BLEPHARITIS
LID HYGIENE
Treatment of rosacea and chronic blepharitis is multifaceted.
Since there is no cure for rosacea, patient education is the key
to controlling symptoms. Lid hygiene is important in reducing
the oil and blepharitic scales around the cilia. This can be
accomplished in various ways. Simple bathing with warm to
hot water held against the lids with a washcloth will hydrate
and loosen brinous scales and mucus and heat the meibomian
gland contents to a more liquid state. The patient should be
instructed to brush the bases of the lashes with the cloth to
mechanically dbride each lash. The patient may also be
instructed to gently press against the meibomian glands, rolling
a nger toward the lid margin trying to express the glandular
secretions. Using one or two cotton-tipped applicators against
the tarsal plates, the physician can gently but forcibly massage
FIGURE 48.10. Ocular rosacea may have

ocular involvement more severe than facial and
lid involvement. Therapy tends to be chronic.
(a) A patient with rosacea after diagnosis and
treatment with oral tetracycline and low-dose
topical steroids shows quiet peripheral
neovascularization. The patient had systemic
hyperlipidemia with secondary deposition of
lipid in his peripheral cornea. (b) After cessation
of therapy, the patient had a flare-up of his
condition that responded to resumption of
therapy with retention of excellent vision.
(c) Peripheral corneal ulceration may occur,
leading to perforation (d). Corneal
transplantation may be necessary and has a
good prognosis if the underlying disease can
be controlled with topical and oral medication.
CHAPTER 48
FIGURE 48.9. Severe

rosacea results in
facial disgurement
from rhinophyma and
markedly dilated
supercial
telangiectatic blood
vessels. Ocular
involvement is
common.
631

FIGURE 48.11. (a and b) Expression of
inspissated meibomian glandular secretions is
helpful in controlling chronic blepharitis. A
bimanual technique, placing cotton-tipped
applicators on both sides of the lids, may also
be used.
SECTION 6
TABLE 48.4. Comparison of Pharmacologic Aspects of Tetracyclines Used in the Treatment of Rosacea and Blepharitis
632
Tetracycline
Minocycline
Doxycycline
Daily dosage
250 mg2 g
50200 mg
50200 mg
Serum half-life
8h
16 h
18 h
Absorption from small bowel
Fair; better on an empty stomach
Excellent
Excellent
Excretion
Urine
Urine; metabolized
Feces
Use in renal failure
Avoid
Avoid with caution
None
Unusual side effects
Pancreatitis; colitis
Vertigo, tinnitus, skin, nail, and scleral

pigmentation
sebum and debris from the glands (Fig. 48.11). The patient may
use baby shampoo diluted with water to scrub the lid margins
and lashes with a washcloth, cotton-tipped applicators, or even
their ngertips. The patient should understand that the goal is
to clean the bases of the lashes and the lid margins, not just the
skin of the eyelids. In a study of eyelid-cleaning regimens in
contact lens wearers with chronic blepharitis, hypoallergenic
bar soap, baby shampoo, and commercial lid scrubs were all
shown to be effective in improving the slit-lamp ndings.
Patients preferred the commercial lid scrub because of
convenience and ease of use.85 Antibiotic ointments with
efcacy against staphylococci, such as bacitracin and erythromycin,
are useful in controlling the more acute bacterial overgrowth
component of the disease but are not effective in eradicating
the severe forms of the disease.46,48 If one subscribes to the
Demodex gland-blocking theory of blepharitis, then one can
postulate that nightly application of ointment to the lid margins
mechanically impedes migration of nymphs from one follicle
to another, thus reducing the infestation. Patients who wear
mascara and eye make-up should replace their products,
especially mascara, on a regular basis to reduce the likelihood of
reinoculating their lids with contaminated cosmetics.86
TREATMENT OF THE DRY EYE

Aggressive treatment of evaporative dry eye associated with
blepharitis is important to relieve the major symptoms of
stinging and burning. Articial tears and ointments should be
used as needed. If they fail to control symptoms, a Schirmer
test with anesthetic should be performed and temporary
punctal occlusion tried if basal tear secretion is low. If there is
success with temporary punctal occlusion and no overflow
tearing, then punctal cautery or silicone plugs should be
seriously considered for the highly symptomatic patient.
Topical cyclosporine A 0.05%, Restasis, an immunomodulator that inhibits activation of T-lymphocytes, has been
studied for use in dry-eye patients and patients with meibomian
gland dysfunction. The suppression of T-cell activation reduces
cytokine production and release of inflammatory mediators.
Although its exact mechanism of action in promoting tear-lm

stability is not fully understood, it has been hypothesized that
Restasis decreases meibomian gland inflammation, thus
reducing the duct obstruction that predisposes to bacterial
colonization. In clinical studies, 15% of Restasis-treated
patients had an increase in Schirmer scores of 10 mm or more
compared with controls.87 Topical cyclosporine for the
treatment of meibomian gland dysfunction was evaluated in a
randomized prospective study.88 The investigators found a 50%
reduction in the number of meibomian gland inclusions at
3 months compared with placebo, suggesting that topical
cyclosporine A may be useful in the treatment of posterior
blepharitis.
Oral omega-3 fatty acid supplementation has been
investigated as an adjunctive treatment for dry eye based on
large epidemiologic studies showing a decrease in dry eye
symptoms among women with a higher intake of foods rich in
omega-3 fatty acids.89 Meibomian glands require essential fatty
acids to produce meibum. Increased dietary intake of omega-3s
as found in certain sh and flax seed oil, has recently been
shown to affect the polar lipid proles of meibum as observed
by high performance liquid chromatography (HPLC).90
Eicosapentaenoic acid (EPA), a long chain omega-3 fatty acid,
blocks the gene expression of proinflammatory cytokines such
as tumor necrosis factor-alpha (TNF-a) and interleukin 1-alpha
that may play a role in meibomitis. Further investigation is
needed to clarify the relevance of dietary omega-3s on the
treatment of blepharitis and dry eye.
TETRACYCLINE
Tetracycline and its derivatives are very useful in treating
rosacea (Table 48.4). Tetracycline is an antibiotic that is
bacteriostatic in usual doses. It inhibits protein synthesis by
binding on the 30S ribosomes. This is similar to the action of
aminoglycosides. It has a broad spectrum of activity against
gram-positive, gram-negative, aerobic, and anaerobic bacteria;
spirochetes; mycoplasmas; rickettsiae; chlamydiae; and some
protozoa.91 Tetracycline, when administered orally, is absorbed
Lid Inflammations
FIGURE 48.12. A patient with rosacea before (a) and after 6 months
of metronidazole topical therapy (b).
(a and b) Courtesy of Curatek Pharmaceuticals and Arthur Sober, MD.
of matrix metalloproteinase-9, which has been implicated in

delayed corneal wound healing. Doxycycline is the most potent
inhibitor, followed by minocycline and tetracycline, which
corresponds to their ability to bind Zn2+. It is postulated that
the inhibitory mechanism is through tetracycline binding of
essential Zn2+ in corneal collagenase.105 This is compatible
with the ndings that low oral dosage levels could cause
complete cessation of lipase production in sensitive strains of
S. epidermidis.103
Tetracycline may be administered in various dosage schedules. Because rosacea is a chronic condition and because gastrointestinal side effects are dose-related, patients are typically
started at low doses such as doxycycline 100 mg twice a day or
tetracycline hydrochloride 250500 mg once or twice a day for
several months. It takes ~6 weeks for symptomatic improvement. In a study comparing doxycycline and tetracycline
therapy, greater symptomatic relief was seen in the tetracyclinetreated group after 6 weeks, but there was no difference between
the groups after 3 months.106 The drug should not be taken
at bedtime, because it may cause reflux esophagitis or stomach
irritation. The drugs are slowly tapered as clinical ndings
warrant.
OTHER THERAPIES
Strongly fluorinated topical steroids should not be used on the
face, especially in rosacea, because the steroids themselves may
cause a confusing picture of steroid-induced rosacea-like
dermatitis.107,108
Topical metronidazole is highly effective in treating rosacea
dermatitis, showing signicant improvements in over 70% of
patients (Fig. 48.12).109111 Metronidazole is a broad-spectrum
antibiotic and antiparasitic agent that has antiinflammatory
and perhaps immunosuppressive effects.112 It probably has little
effect against Demodex.68 It is not currently available in an
ophthalmic preparation, but careful application of the gel to the
lid margins has been shown to signicantly improve the
adnexal changes in ocular rosacea. It did not signicantly alter
the surface disease.113 Metronidazole reduces potent inflammatory mediators in skin in which palmitoleic acid is
present.114 Studies need to be done to show whether controlling
the facial dermal aspects of the condition has any effect on the
ocular manifestations.
CHAPTER 48
in the proximal small bowel and reaches peak levels in 13 h

after administration. There are three groups of tetracyclines,
differentiated by their pharmacology and duration of action.
Tetracycline hydrochloride is typical of the group of short-acting
compounds. It is inexpensive, most commonly used, and most
poorly absorbed. Milk and milk products as well as polyvalent
cations such as calcium, iron, aluminum, and magnesium
inhibit its absorption. The drug is concentrated in the liver
and excreted in the bile. Minocycline and doxycycline are longacting analogues with half-lives of 16 and 18 h, respectively,
versus 8 h for tetracycline. They are absorbed almost completely. Except for doxycycline, they are excreted mainly in the
urine.92
The side effects of tetracycline are chiefly gastrointestinal
and are dose-related. Diarrhea is related to changes in bowel
flora, which are least pronounced with doxycycline because it
is well absorbed. The diarrhea usually subsides when the
antibiotic is stopped; however, pseudomembranous colitis has
been reported.93 Tetracycline-induced pancreatitis has also been
reported.94 Tetracyclines should not be given to patients with
renal failure. These agents should be used cautiously in patients
with hepatic disease because they have been associated with
hepatic toxicity.92 Tetracycline crosses the placenta and
accumulates in fetal bones and teeth.93 It causes a permanent
gray-brown or yellowish discoloration of growing teeth, which
appears to be dose-related.94 Doxycycline does not bind with
calcium to the same degree as other tetracyclines and may cause
less dental discoloration.95 Tetracyclines should be avoided by
pregnant or lactating women and by children younger than 8
years of age.
Allergic reactions that occur with tetracyclines include
urticaria, xed-drug eruptions, periorbital edema, and morbilliform rashes. An allergy to one analogue implies an allergy
to all. Photosensitivity is not an allergic reaction, rather a toxic
one. Superinfection occurs; the most common is oral or vaginal
moniliasis, which can be treated with specic topical
medication. Minocycline can cause vertigo and tinnitus.92
It may cause fatty inltration of the liver, intrahepatic
cholestasis, and acute hepatitis.96 Prolonged administration of
minocycline can cause nail, skin, and scleral pigmentation
that is usually reversible.97 Minocycline has been associated
with a reaction of fever, arthritis/arthralgia, and livedo
reticularis.98 The syndrome was associated with a high titer of
serum perinuclear antineutrophil cytoplasmic antibodies
(p-ANCA) and antimyeloperoxidase (anti-MPO) antibody.
Symptoms resolved after stopping the drug but returned
when minocycline was restarted.92 Benign intracranial hypertension has been reported.99 The use of outdated tetracycline
can cause damage to the renal tubules, resulting in Fanconis
syndrome.100 The formulations producing this syndrome have
been modied, so that it is unlikely to occur in the future.101
It has been reported that women on oral contraceptives have
become pregnant while taking tetracycline.102
Oral tetracycline has become the treatment of choice for
rosacea blepharitis or meibomian keratoconjunctivitis.76,92,103
Its mechanism of action cannot be fully explained by its
antibacterial effects because 75% of S. epidermidis strains are
resistant to tetracycline.19 Tetracycline was found to cause
signicant decreases in the production of bacterial lipase in
sensitive and resistant strains of S. epidermidis without
decreases in bacterial growth. S. aureus showed parallel
decreases in lipase production and growth. McCulley showed
that low doses of tetracycline inhibited bacterial lipase
production by ~30%.23 Lipases act on wax and sterol esters to
release free fatty acids that can affect the solubility of other
lipids in the tear lm or contribute to ocular inflammation.104
Tetracyclines can inhibit collagenase, decreasing the activity
CONCLUSION
Lid inflammation is a problem that is commonly encountered
in a general ophthalmic practice. The severity of symptoms falls
into a range of mild ocular irritation to severe discomfort and
reduced vision. It is therefore important to understand the
633

alterations in the normal physiology of the meibomian glands
that result in lid inflammation in order to target therapies
effectively. Along with patient education, encouragement and
proper treatment, blepharitis can be successfully controlled in
most patients to improve their overall quality of life. As our

understanding of the etiology of blepharitis improves, so will
our future therapies for the treatment of this chronic ocular
condition.
SECTION 6
REFERENCES
634
1. Zug KA, Palay DA, Rock B: Dermatologic

diagnosis and treatment of itchy red
eyelids. Surv Ophthalmol 1996; 40:293.
2. McDonnell PJ: How do general
practitioners manage eye disease in the
community? Br J Ophthalmol 1988; 72:733.
3. Bonnar E, Logan P, Eustace P: Absent
meibomian glands: a marker for EEC
syndrome. Eye 1996; 10:355.
4. Chung CW, Tiggs M, Stone RA: Peptidergic
innervation of the primate meibomian
gland. Invest Ophthalmol Vis Sci 1996;
37:238245.
5. Kirch W, Horneber M, Tamm ER:
Characterization of meibomian gland
innervation in the cynomolgus monkey
(Macaca fasiccularis). Anat Embryol 1996;
193:365375.
6. Seifert P, Spitznas M: Immunocytochemical
and ultrastructural evaluation of the
distribution of nervous tissue and
neuropeptides in the meibomian gland.
234:648656.
7. Maskin SL, Tseng SC: Culture of rabbit
meibomian gland using collagen gel. Invest
8. Maskin SL, Tseng SCG: Clonal growth and
differentiation of rabbit meibomian gland
epithelium in serum-free culture: differential
modulation by EGF and FGF. Invest
9. Gutgesell VJ, Stern GA, Hood CI:
Histopathology of meibomian gland
dysfunction. Am J Ophthalmol 1982; 94:383.
10. Jester JV, Nicolaides N, Smith RE:
Meibomian gland dysfunction. I: keratin
protein expression in normal human and
rabbit meibomian glands. Invest
11. Jester JV, Nicolaides N, Smith RE:
Meibomian gland dysfunction. II: the role of
keratinization in a rabbit model of MGD.
12. Nicolaides N, Santos EC, Smith RE, et al:
Meibomian gland dysfunction. III:
meibomian gland lipids. Invest Ophthalmol
Vis Sci 1989; 30:946.
13. McCulley JO, Shine EW: The lipid layer of
tears: dependent on meibomian gland
14. Tiffany JM: The lipid secretion of the
meibomian glands. Adv Lipid Research
1987;22:162.
15. Shine WE, McCully JP: Association of
meibum oleic acid with mebomian
seborrhea. Cornea 2000; 19:7274.
16. Shine WE, McCulley JP: Meibomianitis,
polar lipid abnormalities. Cornea 2004;
23:781783.
Classication of blepharitis. Ophthalmology
1982; 89:1173.
18. Shine WE, McCulley JP: Meibomian gland
triglyceride fatty acid differences in chronic
blepharitis patients. Cornea 1996; 15:340.
19. Dougherty JM, McCulley JP: Bacterial
lipases and chronic blepharitis. Invest
20. Osgood JK, Dougherty JM, McCulley JP: The

role of wax and sterol esters of meibomian
secretions in chronic blepharitis. Invest
21. Shine WE, McCulley JP: The role of
cholesterol in chronic blepharitis. Invest
22. Shine WE, Silvaney R, McCulley JP:
Relation of cholesterol-stimulated
Staphylococcus aureus growth to chronic
blepharitis. Invest Ophthalmol Vis Sci 1993;
34:22912296.
Meyer DR: The role of tetracyclines in
chronic blepharitis: inhibition of lipase
production in Staphylococci. Invest
24. McLean IW, Burnier MN, Zimmerman LE, et
al: Tumors of the eye and ocular adnexa.
Washington: Armed Forces Institute of
Pathology;1994.
25. Bohigian GM: Chalazion: a clinical
evaluation. Ann Ophthalmol 1979; 11:1397.
26. Lempert SL, Jenkins MS, Brown SI:
Chalazia and rosacea. Arch Ophthalmol
1979; 97:1652.
27. Korn EL: Laser chalazion removal.
28. Cohen BZ, Tripathi RC: Eyelid
depigmentation after intralesional injection
of a fluorinated corticosteroid for chalazion.
29. Jakobiec FA, Silvers D: Reply. Am J
Ophthalmol 88:270, 1979
30. Dua HS, Nilawar DV: Nonsurgical therapy
of chalazion. Am J Ophthalmol 1982;
94:424.
31. Castren J, Stenborg T: Corticosteroid
injection of chalazia. Acta Ophthalmol
1983; 61:938.
32. Mohan K, Dhir SP, Munjal VP, et al: The use
of intralesional steroids in the treatment of
chalazion. Ann Ophthalmol 1986; 18:158.
33. Simon GJ, Huang L, Nakra T, et al:
Intralesional triamcinolone acetonide
injection for primary and recurrent chalazia:
is it really effective? Ophthalmology 2005;
112:913917.
34. Nawrocki J, Swietliczko I: Pars planaVitrektomie nach akzidenteller Injektion von
Depot-Steroiden in den Glaskorperraum.
Klin Monatsbl Augenheilkd 1990; 97:519.
35. Khan JA, Doane JF, Grove AS Jr:
Sebaceous and meibomian carcinomas of
the eyelid: recognition, diagnosis, and
management. Ophthalmic Plast Reconstr
Surg 1991; 7:61.
36. Yeatts RP, Waller RR: Sebaceous
carcinoma of the eyelid: pitfalls in
diagnosis. Ophthalmic Plast Reconstr Surg
1985; 1:35.
37. Doxanas MT, Green WR: Sebaceous gland
carcinoma. Review of 40 cases. Arch
Classication of chronic blepharitis.
39. Mathers WD, Choi D: Cluster analysis of
patients with ocular surface disease,
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
blepharitis, and dry eye. Archives of

Ophthalmology 2004; 122:17001704.
Groden LR, Murphy B, Rodnite J, Genvert
GI: Lid flora in blepharitis. Cornea 1991;
10:5053.
McCulley JP, Sciallis GF: Meibomian
1977; 84:788.
Thygeson P: The etiology and treatment of
blepharitis: a study in military personnel.
Milit Surg 1946; 98:191.
Thygeson P: Etiology and treatment of
blepharitis: a study in military personnel.
Thygeson P: Complications of
staphylococcal blepharitis. Am J
Scobee RG: The role of the meibomian
glands in recurrent conjunctivitis. Am J
Seal DV, Barrett SP, McGill JI: Etiology and
treatment of acute bacterial infection of the
external eye. Br J Ophthalmol 1982;
66:357.
Dougherty JM, McCulley JP: Comparative
bacteriology of chronic blepharitis. Br J
Bowman RW, Dougherty JM, McCulley JP:
Chronic blepharitis and dry eyes. Int
Groden LR, Murphy B, Rodnite J, et al: Lid
flora in blepharitis. Cornea 1991; 10:50.
Valenton MJ, Okumoto M: Toxin-producing
strains of Staphylococcus epidermidis
(albus): isolates from patients with
staphylococcic blepharoconjunctivitis. Arch
Seal D, Ficker L, Ramakrishnan M, et al:
Role of staphylococcal toxin production in
blepharitis. Ophthalmology 1990; 97:1684.
Ficker L, Ramakrishnan M, Seal D, et al:
Role of cell-mediated immunity to
staphylococci in blepharitis. Am J
Seal D, Ficker L, Wright R, et al: The case
against thiomersal. Lancet 1991; 338:315.
Mondino BJ, Caster AI, Dethlefs B: A rabbit
model of staphylococcal blepharitis. Arch
Giese MJ, Adamu SA, Pitchekian-Halabr H,
et al: The effect of Staphylococcus aureus
phace lysate vaccine on a rabbit model of
staphylococcal blepharitis, phlyctenulosis,
and catarrhal inltrates. Am J Ophthalmol
1996; 122:245.
English FP, Nutting WB: Demodicosis of
ophthalmic concern. Am J Ophthalmol
1981; 91:362.
Norn MS: Demodex folliculorum: incidence
and possible pathogenic role in the human
eyelid. Acta Ophthalmol 1970; 108:85.
Roth AM: Demodex folliculorum in hair
follicles of eyelid skin. Ann Ophthalmol
1979; 11:37.
Kemal M, Sumer Z, Toker MI, et al: The
prevalence of Demodex folliculorum in
blepharitis patients and the normal
population. Ophthalmic Epidemiol 2005;
12:287290.
60. Rufli T, Mumcuoglu Y: The hair follicle mites

Demodex folliculorum and Demodex brevis:
biology and medical importance: a review.
Dermatologica 1981; 162:1.
61. English FP, Zhang GW, McManus DP,
Campbell P: Electron microscopic evidence
of acarine infestation of the eyelid margin.
Am J Ophthalmol 1990; 109:239;1981;
91:362.
62. Coston TO: Demodex folliculorum
blepharitis. Trans Am Ophthalmol Soc
1967; 65:361.
63. Czepita D, Kuzna-Grygiel W, KosikBogacka D: Investigations on the
occurrence as well as the role of Demodex
follicuforum and Demodex brevis in the
pathogenesis of blepharitis. Klin Oczna.
2005; 107:8082; (Article in Polish).
64. Rufli T, Mumcuoglu Y, Cajacob A, et al:
Demodex folliculorum: Zur
Atiopathogenese und Therapie der
Rosazea und der perioralen Dermatitis.
Dermatologica 1981; 162:12.
65. Rufli T, Buchner SA: T-cell subsets in acne
rosacea lesions and possible role of
Demodex folliculorum. Dermatologica
1984; 160:1.
66. Browning DJ, Proia AD: Ocular rosacea.
67. Rodriguez AE, Ferrer C, Alio JL: Chronic
blepharitis and Demodex. Arch Soc Exp
Oftalmol 2005;80:635642.
68. Persi A, Rebora A: Metronidazole and
Demodex folliculorum. Acta Derm Venereol
(Stockh) 1981; 61:182.
69. Ellis CN, Stawiski MA: The treatment of
perioral dermatitis, acne rosacea, and
seborrheic dermatitis. Med Clin North Am
1982; 66:819.
70. Borrie P: Rosacea with special reference to
its ocular manifestations. Br J Dermatol
1954; 65:458.
71. Lemp MA, Mahmood MA, Weiler HH:
Association of rosacea and
1984; 102:556.
72. Duke-Elder: System of ophthalmology.
Disease of the outer eye. St Louis: Mosby;
1965:498527.
73. Gupta AK, Chaudhry MM: Rosacea and its
management: an overview. J Eur Acad
74. Browning DJ, Rosenwasser G, Lugo M:
Ocular rosacea in blacks. Am J Ophthalmol
1986; 101:441.
75. Rosen T, Stone MS: Acne rosacea in
blacks. J Am Acad Dermatol 1987; 17:70.
76. Urabe H, Koda H: Perioral dermatitis and
rosacea-like dermatitis: clinical features and
treatment. Dermatologica 1976; 152:155.
77. Marks R: The problem of rosacea BMJ
1976; 1:94.
78. Marks R, Ellis J: Comparative effectiveness
of tetracycline and ampicillin in rosacea.
Lancet 1971; 2:1049.
79. Barton K, Dagoberto CM, Nava A,
Pflugfelder SC: Inflammatory cytokines in
the tears of patients with ocular rosacea.
Ophthalmology1997; 104:18681874.
80. Sobrin L, Liu Z, Monroy DC, et al:

Regulation of MMP-9 activity in human tear
fluid and corneal epithelial culture
supernatant. Invest Ophthal Vis Sci 2000;
41:17031709.
81. Alfonso AA, Sobrin L, Monroy DC, et al:
Tear fluid gelatinase B activity correlates
with IL-1alpha concentraion and fluorescein
clearance in ocular rosacea. Invest
82. Tan SG, Cunliffe WJ: Rosacea and
migraine. BMJ 1976; 1:21.
83. Berg M, Liden S: An epidemiological study
of rosacea. Acta Derm Venereol (Stockh)
1989; 69:419.
84. Hoang-Xuan T, Rodriguez A, Zaltas MM,
et al: Ocular rosacea. A histologic and
immunopathologic study. Ophthalmology
1990; 97:1468.
85. Key JE: A comparative study of eyelid
cleaning regimens in chronic blepharitis.
CLAO J 1996; 22:209.
86. Wilson LA, Julian AJ, Ahearn DG: The
survival and growth of microorganisms in
mascara during use. Am J Ophthalmol
1975; 79:596.
87. Sall K, Stevenson OD, Mundorf TK, Reis
BL: Two multicenter, randomized studies of
the efcacy and safety of cyclosporine
ophthalmic emulsion in moderate to
severe dry eye disease. CsA Phase 3
Study Group. Ophthalmology 2000;
107:631639.
88. Perry HD, Doshi-Carnevale S, Donnenfeld
ED, Solomon R et al: Efcacy of
commercially available topical cyclosporine
a 0.05% in the treatment of meibomian
gland dysfunction. Cornea 2006;
25:171175.
89. Miljanovic B, Trivedi KA, Dana MR, et al:
82:887893.
90. Sullivan BD, Cermak JM, Sullivan RM, et al:
Correlations between nutrient intake and
the polar lipid proles of meibomian gland
secretions in women with Sjogrens
syndrome. Adv Exp Med Biol 2002;
506:441447.
91. Standiford HC: Tetracyclines and
chloramphenicol. In: Mandell GL, Douglas
RG Jr, Bennett JE, eds. Principles and
practice of infectious disease. 3rd edn.
New York: Churchill Livingstone;
1990:284295.
92. Salamon SM: Tetracyclines in
29:265.
93. Gorbach SL, Bartlett JG: Anaerobic
infections. N Engl J Med 1974; 290: 1289.
94. Elmore MF, Rogge JD: Tetracycline-induced
pancreatitis. Gastroenterology 1981;
81:1134.
95. Forti G, Benincori C: Doxycycline and the
teeth. Lancet 1969; 1:782.
96. Malcolm A, Heap TR, Eckstein RP, et al:
Minocycline-induced liver injury. Am J
Gastroenterol 1996; 91:1641.
97. Angeloni VL, Salasche SJ, Ortiz R: Nail,

skin, and scleral pigmentation induced by
minocycline. Cutis 1987; 40:229.
98. Elkayam O, Yaron M, Caspi D: Minocycline
induced arthritis associated with fever,
livedoreticularis and p-ANCA. Ann Rheum
Dis 1996; 55:769.
99. Goulden V, Glass D, Cunliffe WJ: Safety of
long term high dose minocycline in the
treatment of acne. Br J Dermatol 1996;
134:693.
100. Fulop M, Drapkin A: Potassium-depletion
syndrome secondary to nephropathy
apparently caused by out-dated
tetracycline. N Engl J Med 1965; 272:986.
101. Whelton A: Tetracyclines in renal
insufciency: resolution of a therapeutic
dilemma. Bull N Y Acad Med 1978; 54:223.
102. Bacon JF, Cheneld GM: Pregnancy
attributable to interaction between
tetracycline and oral contraceptives. BMJ
1980; 280:293.
103. Dougherty JM, McCulley JP, Silvany RE, et
al: The role of tetracycline in chronic
32:2970.
104. McCulley JP, Sciallis GF: Meibomian
1977; 84:788.
105. Burns FR, Stack MS, Gray RD, et al:
Inhibition of puried collagenase from
alkali-burned rabbit cornea. Invest
106. Frucht-Pery J, Sagi E, Hemo I, et al:
Efcacy of doxycycline and tetracycline in
ocular rosacea. Am J Ophthalmol 1993;
116:88.
107. Leyden JJ, Thew M, Klingman AM: Steroid
rosacea. Arch Dermatol 1974; 110:619.
108. Fisher AA: Steroid rosacea: a friendly
pharmacist syndrome. Cutis 1987; 40:209.
109. Bleicher PA, Charles JH, Sober AJ: Topical
metronidazole therapy for rosacea. Arch
Dermatol 1987; 123:609.
110. Aronson IK, Rumseld JA, West DP, et al:
Evaluation of topical metronidazole gel in
acne rosacea. Drug Intell Clin Pharm 1987;
21:346.
111. Lowe NJ, Henderson MD, Millikan LE, et al:
Topical metronidazole for severe and
recalcitrant rosacea: a prospective open
trial. Cutis 1989; 43:283.
112. Grove D, Mahmoud AAF, Warren KS:
Suppression of cell-mediated immunity by
metronidazole. Int Arch Allergy Appl
Immunol 1977; 54:422.
113. Barnhorst DA, Foster JA, Chern KC, et al:
The efcacy of topical metronidazole in the
treatment of ocular rosacea.
114. Akamatus H, Oguchi M, Nishijimas S, et al:
The inhibition of free radical generation by
human neutrophils through the synergistic
effects of metronidazole with palmitoleic
acid: a possible mechanism of action of
metronidazole rosacea and acne. Arch
Dermatol Res 1990; 282:449.
CHAPTER 48
Lid Inflammations
635
CHAPTER
49
Viral Disease of the Cornea and External Eye

Deborah Pavan-Langston
Viral infections of the external eye range from the benign to

malignant, from a transient keratitis of mononucleosis to the
progressive sarcoma associated with AIDS. Transient acute
follicular conjunctivitis, with or without keratitis, may be seen
with any of the DNA or RNA agents. The former include the
herpes viruses, adenoviruses, poxviruses, and papillomaviruses.
The latter include the paramyxoviruses (measles, mumps, and
Newcastle disease), the retrovirus (HIV), the picornavirus of
acute hemorrhagic conjunctivitis, the togaviruses (rubella and
arbovirus), and the orthomyxovirus (influenza). With the exception of congenital rubella, the RNA viruses tend to be the more
benign, and the DNA viruses more associated with a notable
ocular morbidity and loss of vision. Many, but not all, of the
organisms that most seriously affect the anterior segment are
amenable to therapy; however, some progress despite our best
efforts, and others need only palliative treatment. Diagnostic
tests and currently available antiviral and antiinflammatory
agents are briefly reviewed, followed by a discussion of the
clinical disease and management of the major anterior ocular
viral infections. Antiviral drugs are discussed in detail in
Chapter 20.
DIAGNOSTIC TESTS
Diagnosis of ocular viral disease is usually made based on
clinical impression only. When objective data are needed,
however, the four most commonly used approaches are: (1)
examination of skin, conjunctival, or corneal scrapings
(herpetic, adenoviral, and pox infections); (2) molecular and
immunologic assays; (3) viral culture; and (4) measurement of
circulating antibodies. A fth approach is histopathologic study
of tissue obtained at keratoplasty. This is done almost
exclusively in herpetic disease as other forms of viral disease are
not sufciently severe to warrant biopsy or are obtained only as
postmortem specimens.15
External ocular scrapings may be taken with a sterile platinum
spatula or the edge of a curved surgical blade, then these are
smeared on a slide and stained for light microscopic examination. The cellular inflammatory reaction indicative of viral
infection is predominantly a monocytic white cell inltrate.
The simplest cytopathologic examination uses heat xation of
the slide followed by Giemsas staining. Herpetic infections
caused by herpes simplex virus (HSV) and herpes varicellazoster virus (VZV) are characterized by some multinucleated
epithelial cells with ballooning degeneration and a mixed mononuclear and polymorphonuclear leukocyte (PMN) reaction. As
Giemsas stain obscures nuclear detail, the eosinophilic viral
inclusion bodies of Lipschutz, also called Cowdry-A inclusions,
are best seen in the epithelial cells after the slide is xed in
Bouins solution or 95% ethanol and stained with the
Papanicolaou method. The inclusion bodies appear as an

eosinophilic intranuclear mass within a clear halo and are
associated with clumping or margination of the basophilic
chromatin on the nuclear membrane. Fluorescent antibody
staining may also reveal herpetic antigen in the nucleus or
cytoplasm. This technique is rapid and as reliable as tissue
culture recovery but is not generally available outside of major
medical centers because an ultraviolet microscope is required.
The stained preparations are also unstable, which prevents
storage for later evaluation.14,6
Polymerase chain reaction (PCR) has also become a useful
and sophisticated tool for diagnosis of numerous herpetic and
other diseases. PCR has been used to detect both HSV and VZV
in the tear lm and corneas of patients.2,79 PCR has further
been used to identify HSV DNA in irido-corneal-endothelial
and in PosnerSchlossman syndromes.10,11 While Kaye has
reported nding HSV DNA by PCR in corneal lesions unrelated
to HSV, the rst two reports raise the specter of HSV as an
etiologic agent in syndromes not previously associated with this
infectious agent.12,13
Multiplex PCRs (mPCRs) have been successfully developed
for detection of DNA and RNA agents in the investigation of
congenital infection and an mPCR for the viruses most
commonly requested in a diagnostic virology laboratory (CMV,
EpsteinBarr virus (EBV), enterovirus, HSV-1, HSV-2, and
varicella-zoster virus).14 Nested PCR was performed as the most
sensitive assay currently available, and detection of the
amplicons using hybridization to labeled probes and enzymelinked immunosorbent assay detection was incorporated into
three of the four assays. In a number of cases this technique
reveals an agent not diagnosed clinically which affected subsequent treatment and course.
A new, easy to use, multipotential derivative of PCR is the
Smartcycler II (Cepheid, Sunnyvale, CA) real-time PCR system
for detecting HSV-1, VZV, adenovirus, and Chlamydia
trachomatis in ocular infections.15 This test may be performed
quickly in a small conventional laboratory or ofce with results
comparable to those of a central molecular laboratory.
Sensitivity for adenovirus, HSV-1, VZV, and Chlamydia
trachomatis were 85%, 91%, 100%, and 95% respectively while
specicity was 98%, 100%, 100%, and 100%.15
Enzyme-linked immunosorbent assay (ELISA) tests are used
not only for viral antigen detection but are particularly useful
for detection of IgM in the presence of IgG. Only IgM antibodies, if present in the serum, are bound to the solid phase and
are therefore easily detectable. This is important diagnostically
as IgM antibodies appear early during infection and last only a
few weeks whereas IgG comes after 1 or 2 weeks but lasts for
years. Quantitative documentation of a fourfold rise in either
IgM or IgG strongly supports a diagnosis. Serum should be
637
SECTION 6
638
drawn as soon as possible in the acute illness and again 23

weeks later for comparative titers. Finding a positive IgM in a
single specimen may also be diagnostic in a very ill patient, e.g.,
disseminated varicella, or a patient with ongoing infection, e.g.,
HIV. Elevation of IgM may also indicate re-activation of a latent
infection. IgM detection is most useful in diagnosis of VZV,
EBV, CMV, measles, rubella, coxsackie viruses, and hepatitis. As
IgM does not cross the placental barrier, nding IgM in a
newborn is diagnostic of intrauterine infection.1,1619
Immunologic ELISA diagnostic kits are available for specic
diagnosis of HSV infections in ocular scrapings: the Herpchek
kit (Dupont) and the Virogen kit may be purchased
commercially.20 The Virogen test relies on the cross-linking
of latex particles to produce an agglutination reaction visible
to the human eye without magnication; Herpchek uses an
immunostaining system that relies on a color change for
antigen detection. The Herpchek is 100% specic and has a
sensitivity of 99% compared with the shell vial tissue culture
method but because of the equipment needed require the
facilities of medical centers. The Virogen is easy to set up in an
ofce but has about 26% sensitivity.20,21
Serologic testing is also the principal method of diagnosis in
EBV infection. Rapid diagnosis of acute EBV infectious mononucleosis can usually be made on clinical grounds, atypical
lymphocytosis, and a positive rapid heterophile.2224 The
Monospot test has numerous pitfalls but tests for viral capsid
antigen (VCA), EB nuclear antigen (EBNA) (see EBV ahead)
also of use particularly in late diagnostic testing where only IgG
VCA and EBNA are high for several years but early antigen (EA)
and IgM VCA are negative. In HIV testing, the conventional
ELISA is used to screen for anti-HIV antibodies and the Western
blot test, a more sophisticated and complex antibody assay,
used for conrmation of diagnosis in those screening positive by
ELISA.25 The HIV Oraquick rapid HIV-1 testing (OraSure
Technologies, Inc., Bethlehem, PA) (blood) has been found to be
a highly reliable rapid test preferred by patients in screening
programs and enhancing the effectiveness of screening
programs.26
In adenoviral infections, Giemsa-stained smears of conjunctival exudates reveal lymphocytes and degenerated epithelial
cells with a few polymorphonuclear leukocytes (PMNs). If the
reaction is so acute as to induce pseudomembrane formation, a
PMN response will predominate over the mononuclear. No
light-microscope-visible inclusion bodies are formed. As with
herpetic infections, fluorescent antibody testing or ELISA may
be used for rapid denitive diagnostic testing on scrapings taken
during the rst week of infection, but the drawbacks noted with
the use of these techniques in herpetic disease also apply here
thus favoring the Smartcycler II.15,1,27
Smallpox and vaccinia have only recently become infectious
agents of concern again, the former having been declared extinct
by 1980. Now, however, it is a potential bioterrorism agent and
its preventative, vaccinia vaccination, make both these agents
of importance in ocular disease such as cellulitis, conjuctivitis,
and acute or chronic keratitis or iritis. In pox infections, acute
disease is characterized by an outpouring of PMNs followed
by a mononuclear reaction days later. Giemsa-stained smears
may reveal diagnostic eosinophilic bodies of Guarnieri in the
cytoplasm of epithelial cells. Diagnostic tests of use are not
routine serologic testing but ELISA, radioimmunoassay, or
monoclonal antibody assays.1,2831
Viral recovery on tissue culture is the most denitive method
of diagnostic testing but, unfortunately, may take several days
to become positive and is not widely available outside of major
medical centers.1 Ocular or periocular lesions are swabbed with
calcium-alginate-tipped applicators and eluted either into viral
carrier medium or into viral monolayer tubes and sent directly
to the laboratory, where the carrier medium is inoculated into

cultures and the inoculated cell monolayers incubated at 37C.
Once a virus is recovered in culture, its precise identity is
conrmed by serologic testing using antibody specically
directed against the suspected agent.
HSV and the poxviruses grow on almost any cell monolayer
such as human embryonic or rabbit kidney or chick embryo.
The recovery rate from acutely infected ulcers is about 70% if
the specimen is taken within 23 days of the appearance of the
lesion, as the viral titer is highest just before, and as the corneal
or skin lesions appear and then decreases as the clinical ndings
become more prominent.30 The use of antiviral agents prior to
culture will drop recovery rate to 4% even in early disease.32
VZV and adenovirus are more fastidious; therefore, cultures
must be done early in disease, and these require cells of human
origin. They may be isolated directly from ocular cultures, but
because of their fastidious nature diagnosis is usually by
immunofluorescence, viral neutralization or PCR.13,15,37
CMV has been isolated from virtually all forms of body fluid
from tears to blood, or breast milk. Transmission is via the
congenital, oral, and sexual routes, blood transfusion and tissue
transplantation. Diagnosis, however, is usually made by tests
such as immunoassays, and PCR testing as discussed above.
EBV and human immunodeciency virus (HIV) are generally
not grown in culture for diagnostic purposes.1
ANTIVIRAL DRUGS
Twenty antiviral drugs are currently FDA-approved for clinical
use. Half of those are for the treatment of HIV infections
(acquired immune deciency syndrome). The others are used
for herpes virus (e.g., herpes simplex virus, varicella zoster
virus, and cytomegalovirus), hepatitis B virus, hepatitis C
virus, or influenza virus infections. Recent studies have focused
on antiviral therapies for virus infections that appear amenable
to antiviral drug treatment, as well as for virus infections
for which, to date, no antiviral drugs have been approved, e.g.,
adenoviruses, human herpes virus type 6, poxviruses, corona
virus, severe acute respiratory syndrome, and hemorrhagic fever
viruses.38 A vaccine has been approved for human papilloma
viruses related to cervical carcinoma.
There are nine antiviral drugs with proven efcacy in ocular
viral disease: idoxuridine (IDU, Herplex), vidarabine (ara-A, Vira
A), trifluridine (TFT, F3T, Viroptic), acyclovir (ACV, Zovirax),
famciclovir (FCV, Famvir), and valacyclovir (VCV, Valtrex) and
bromovinyldeoxyuridine (BVDU, Brivudine). Ganciclovir (DHPG,
Cytovene), foscarnet (PFA, Foscavir), and HPMPC (Cidovir)
have specialized roles. All but BVDU are approved by United
States Food and Drug Administration (FDA) in one or more
forms: drops, ointments, pills, or for injection.3942 Because
of greater convenience and overlapping efcacies two antiherpes
drugs are no longer commercially available: IDU and
vidarabine. BVDU is licensed throughout Europe. This chapter
discusses these various antiviral agents where pertinent in
various clinical therapy sections5,43 (see Chapter 20).
CORTICOSTEROIDS: PROS, CONS,

INITIATION, AND WITHDRAWAL
TECHNIQUE
Corticosteroids (steroids), specically the glucocorticoids, are of
use in those viral diseases characterized by vision-threatening
immunologic keratitis or keratouveitis seen in certain cases of
stromal HSV, VZV, or adenoviral disease. These drugs interfere
with the distribution and function of immunologically
competent lymphocytes, amoeboid white cell migration, and
release of white cell digestive enzymes. Topical steroids inhibit

as to the etiology: chronic HSV or the steroid. Abel et al have
reported that the normal tear concentration of calcium and
phosphate are near spontaneous precipitation levels.49,50 While
the added burden of phosphate when delivered as a steroid salt
has never been proved to cause band keratopathy, the index of
suspicion is raised such that acetate forms of topical steroids
might be advisable in patients showing tendencies toward
calcium precipitation.
Oral steroids given short-term may be necessary in situations
such as iritis in the presence of a melting ulcer with topical
steroids not being initiated until some control over the ulcer
has been achieved. Common dosing schedule is 20 mg of
prednisone po with meals tid for 5 days, then bid for 5 days, and
nally q AM (to minimize adrenal suppression) for 4 days.
Medical history to ascertain any contraindications to oral
steroid should be taken before starting any such regimen.
Cyclosporin A (Cy A), FDA-approved for dry inflammatory
disease, may be useful in controlling herpetic inflammation
without the risk of elevating intraocular pressure.51 It has been
reported, however, that herpetic epithelial keratitis persisted in
a corneal graft until Cy A was discontinued suggesting that the
drug has the same potential to enhance the infectious
component of herpetic disease as steroids.52 Cy A should,
therefore, be covered with antiviral agent prophylactically. In
another study, 10 patients with HSK using 2% Cy A drops qid
and ACV 3% ointment ve times per day for 2 months had
complete resolution of the stromal disease, vision increased by
at least two Snellen lines in eight of the 10 patients and there
were no episodes of epithelial infection.51 Heilingenhaus study
of 18 patients with stromal HSK treated with Cy A drops and
acyclovir ointment reported that the condition can be treated
successfully with CsA eyedrops, especially in nonnecrotizing
disease. Some cases of necrotizing keratitis required a
combination of CyA and steroids, but the latter at lower dose
than without Cy A. Cy A, then, may be particularly helpful in
the presence of steroid glaucoma, herpetic corneal ulcers, and to
taper off topical corticosteroids. Additional use of oral acyclovir
may be judicious to prevent recurrence of epithelial HSV
keratitis.53
HERPETIC DISEASE OF THE ANTERIOR

SEGMENT
The herpes viruses of interest in the context of anterior segment
disease are HSV, VZV, EBV, and CMV.
HERPES SIMPLEX VIRUS (HSV)

Ocular HSV is a multifaceted disease capable of inducing the
most difcult complications through both infectious and
immune pathogenetic mechanisms.
Epidemiology
HSV is the most common infectious cause of corneal blindness
in the developed world, with up to 500 000 cases diagnosed
annually in the United States alone.5 There are a number of
excellent ocular epidemiologic studies which emphasize the
importance and morbidity of this disease.30,54,55 A study out
of Moorelds Eye Hospital in London sampled their clinic
population to ascertain the incidence with herpetic eye disease
of the anterior segment. Age, gender predominance, and incidence of bilateral disease among these patients had not changed
in 20 years. The most common disease form was stromal
keratitis, with signicant morbidity and visual loss. They also
noted that HSV disease of the anterior segment utilizes only 1%
of out-patient clinic resources but 17% of external disease
specialists clinic time. There was a disturbing and statistically
CHAPTER 49
antibody-forming cells and cell-mediated immunity (CMI) in

the cornea and iris. As there is little effect on the number of
immunocompetent cells in the draining lymph nodes, the host
is still capable of immune reaction upon reduction or cessation
of steroid therapy.4446
Steroids interfere not only with corneal neovascularization but
also with mucopolysaccharide and collagen formation, substances
critical to the integrity of the corneal structure. Medroxyprogesterone acetate (Provera) is a mild steroid that suppresses both
latent and active collagenases, thus differing from other steroids
by not interfering with collagen synthesis yet still suppressing
CMI and neovascularization.47 At 1.0% concentration, medroxyprogesterone has an antiinflammatory efcacy roughly
equivalent to 0.12% prednisolone. It appears to be a safer topical
steroid for use if these drugs must be used to suppress immune
reaction in the presence of corneal thinning or melting.
The factors in support of steroid use in the eye are inhibition
of (1) white cell inltration, (2) release of toxic hydrolytic
enzymes, (3) scar-tissue formation, and (4) neovascularization.
Because these drugs induce vascular constriction, they keep the
eye white and quiet, thus enhancing patient comfort. On the
negative side of steroid administration are (1) suppression of the
normal immune-inflammatory response, allowing spread of
potentially supercial viral infection, inhibition of collagen
synthesis in corneal ulceration, (2) opening the eye to opportunistic bacterial or fungal infection through suppression of the
immune defense system, and (3) steroid-induced glaucoma and
cataract.
It may be difcult to withdraw treatment once a patient is
committed to topical steroid therapy. Too abrupt cessation often
results in rebound inflammation. A useful rule of thumb in
tapering a patient from topical steroid therapy is never reduce
dosage more than 50 percent of the current level of therapy.
The higher the dosage the more rapid the taper and, conversely,
the lower the dosage the more prolonged the withdrawal period.
If a patient is at risk of scarring involving the visual axis, it is
preferable to initiate steroid therapy to inhibit structural
damage and then slowing wean the dose down to whatever level
is necessary to keep the disease quiescent. Periodic attempts at
further tapering should be made over time. As an example, a
patient may require 1% prednisolone qid to bring an immune
keratitis under control over a 12-week period. Dosage may
then drop to tid or bid (but no lower) over the next 34 weeks,
then down to everyday over several weeks before switching to
0.12% prednisolone qid so that the decrease is not more than
50% of the total dose. The physician is then positioned to take
the patient down through a more prolonged withdrawal using a
much weaker drug concentration. Excellent alternative medium
to weak strength steroids with less tendency to cause glaucoma
and cataract are rimexolone 1% (Vexol), lotoprednol 0.5%
(Lotemax), and the weaker lotoprednol 0.2% (Alrex) and
fluoromethalone 0.1% or 0.25% or 0.1% ointment. Not
infrequently, a rebound may begin during withdrawal, signaling
a need to go back up to the next higher dosage for a longer
period of time. It is not uncommon, nor necessarily worrisome,
if a patient is unable to discontinue topical steroids altogether.
Many patients do very well on one drop a day, every other day,
or even once-weekly using 0.12% prednisolone to hold an
otherwise scarring immune reaction in check.
It should be noted that patients with a history of previous
HSV epithelial keratitis and nonwhite patients are more likely
to develop HSV epithelial keratitis during treatment of stromal
keratouveitis with steroids.48 Such patients should be put on
prophylactic oral antiviral therapy during the period of steroid
use or at least until the dose is down to once daily or less.
The issue of band keratopathy being so frequent in HSV
patients chronically on topical steroids has raised the question
639
SECTION 6
640
signicant correlation between total length of follow-up and

reduced visual acuity. While the prevalence of anterior segment
herpetic eye disease appears not to have increased in incidence
the visual prognosis had worsened.29
In a more recent study in France, the overall incidence of
HSV keratitis was 31.5/100 000 person-years (p = 0.05). The
incidence was 13.2/100 000 person-years for new cases (p = 0.05)
and 18.3/100 000 person-years for recurrences (p = 0.05).56 The
most frequent types were dendritic keratitis (56.3%, n = 153),
stromal keratitis (29.5%, n = 81), and geographic keratitis
(9.8%, n = 27). Associated with 35% of the keratitis cases were
conjunctivitis (18.8%, n = 67), uveitis (11.8%, n = 42), and/or
lid involvement (8.6%, n = 31).56
Humans are the only natural reservoir of herpes. Sources of
infection are by direct contact with infected lesions, by salivary
droplets from children and adults with active disease (cold
sores), and via the saliva or fomites of asymptomatic, virusshedding carriers.57,58 The marked frequency of trigeminal HSV
and VZV was demonstrated in ganglia removed at autopsy from
immunocompetent individuals with no history of HSV
infections who were screened by PCR for latent HSV-1 and
(VZV) DNA. HSV-1 DNA was found in the vast majority of
samples (> 90%) and VZV DNA in ~50% of samples. Both
DNA types were distributed throughout each latently infected
ganglion.55 Peposes recent report on the changing epidemiology
and new emerging disease patterns reveals an epidemic increase
in genital HSV-2 (30% increase in type 2 antibodies in the
United States since 1976).59 Approximately, one in four people
in the United States over age 30 is infected with HSV-2. In
contrast to developing nations where HSV is acquired early in
life and is ubiquitous, primary acquisition of herpes simplex
type 1 is becoming progressively delayed in many industrialized
countries. Changes in sexual behavior among young adults have
been associated with a recent increase in genital HSV-1
infection, resulting from oralgenital rather than genitalgenital
contact which strongly suggests that we will or may already be
seeing an increase in the incidence of type 2 HSV keratitis.
Multiple recurrences are far more common with genital or
oral HSV in comparison with the recurrence of ocular disease.
The reported rates of HSV-2 genital recurrence are 0.33/month
(89% in 24 of 27 patients), for orolabial HSV 0.12/month (42%
in ve of 12 patients) in one study, and only 40% over a 5 year
period for another study on ocular HSV epithelial recurrence
rates.5961 The elderly (> 60 years) appear particularly
susceptible to microbial keratitis with HSV being the most
common cause (8% out of 62% positive cultures or PCR in 190
patients) and causing greater damage than that seen in younger
populations.62 Further, it has been found that recurrent
epithelial herpes is frequently associated with corneal reinfection with a different HSV-1 strain with PCR typing of
strains from successive recurrences revealing 37% were of a
strain different from that causing the previous recurrence.63
Iatrogenic sources of patient infection are the physicians
unwashed hands and the contaminated Schiotz or applanation
tonometer head. HSV is viable for up to 2 h on a dry tonometer
head and up to 8 h on the one kept moist. Dry wiping and a
variety of ophthalmic solutions such as anesthetics and dilating
agents have minimal antiviral effect. Swabbing the tonometer
with 70% isopropyl alcohol is 100% effective in killing virus,
and this should be done between patients along with hand
washing with a soapy solution. Dipping the tip into Dakins
solution followed by careful rinsing and wiping dry with a sterile
pad is also 100% effective. By the age of 5 years, at least 60% of
all children have been infected with type 1 (oral) HSV, usually
through the mouth or nares, with only about 6% developing
clinically apparent primary disease. Less than 4% of primary
HSV presents as overt ocular disease.58 The oral and nasal portals
of entry allow the virus access to the trigeminal ganglion, which

also innervates the eye. The vast majority of rst ocular (not
primary) or recurrent ocular herpes infections are due to reactivation of latent trigeminal ganglion virus with subsequent
appearance of the virus in the eye alone, or associated with an
eruption of cold sores around the mouth or nose.
The spectrum and recurrence of (HSV) keratitis in children
and adolescents has been reported in a retrospective cohort
study of 23 patients under age 16 years and diagnosed with
HSV keratitis. Dendritic or punctate epithelial keratitis and
stromal keratitis occurring concurrently with epithelial keratitis
was seen in 14 patients (61%). Six patients (26%) had HSV
keratitis OU. Eleven of these 23 children (48%) developed
recurrent HSV keratitis at a median of 15 months after the rst
documented episode and in three patients, amblyopia occurred.
The study concluded that children with herpetic keratitis may
frequently have bilateral ocular involvement, like adults are at
risk for recurrent keratitis, and in addition have the added risk
of amblyopia.64
HSV tear shedding studies have revealed that people with no
history of ocular herpes may have HSV in the tear lm and
patients with known history of HSV keratitis show no greater
risk than the rest of the population of shedding asymptomatically.8,65 Tear antibodies to HSV-1 may be detected in
the absence of detectable serum or parotid saliva antibodies.
The former nding suggests that the ocular surface may be the
initial infection site for this virus in some healthy, clinically
normal subjects, and the latter suggests that there is a
preferential homing of committed B lymphocytes to different
mucosal surfaces.66 The signicance of these ndings in terms
of latency and recurrent disease is yet to be determined.
Other recent studies on the tears and saliva in 50
asymptomatic patients determined DNA copy number and
frequency of shedding by real-time PCR sampling bid over 30
days.67 While only 74% of the patients were sero-positive for
HSV-1 (by IgG ELISA and neutralizing antibodies), 98% shed
virus at least once from either or both saliva and tears during
the test period. This indicated that the percentage of
asymptomatic subjects who intermittently shed HSV-1 DNA in
tears or saliva was higher than the percentage of subjects with
positive ELISA or neutralization antibodies to HSV. As most
HSV transmission occurs during asymptomatic shedding, it is
easy to see why the frequency of this infection is so widespread.
Other epidemiologic characteristics of ocular herpes have
been noted. Liesegang and co-workers reported an incidence of
8.4 new cases of rst ocular herpes per 100 000 person-years
during the period from 1950 to 1982.68 In 1980, the overall
prevalence of a history of ocular herpes was 149 cases per
100 000 population. Initial ocular HSV episodes included
incidences of 54% blepharitis or conjunctivitis, 63% epithelial
keratitis, 6% stromal keratitis, and 4% uveitis. Age-adjusted
rates by sex were comparable and there were no seasonal trends
in incidence, although rates increased with time. In some
contrast, a study by Bell et al of 141 patients with documented
infectious dendritic/geographic keratitis revealed that there was
a signicant predominance of men in the population of patients
over 40 years of age.69 Of the 65 patients who suffered more
than one episode, 34% had a mean recurrence rate of one or
more episodes per year and 68% had more than one episode
within 2 years of each other. The cold weather months of fall
and winter correlated with increased herpetic recurrences and
flu-like viral respiratory infections. The epidemiologic report by
Shuster et al on ocular HSV in 119 patients revealed that 24%
had a recurrence within 1 year of the rst ocular occurrence
and 33% had them within 2 years.70 There was a positive
correlation between short intervals between past attacks and
short intervals between future recurrences.
HSV Vaccine
Although much interest and work is now being conducted with
an aim to developing either a vaccine to prevent a primary
infection or recurrent disease, at present, neither passive immunization nor existing circulating antibodies have signicant
influence on the development of disease.7375 Neutralizing antibody titers in the serum remain constant during and between
recurrences or may fluctuate to high levels in the absence of an
episode of infectious disease.7679 Approaches to the induction
of protective responses by altering innate and adaptive immunity using novel vaccines such as recombinant viral vaccine
vectors and DNA vaccines specically tested in models of HSV
infections of the eye may offer some true efcacy.73 As Nesburn
et al have noted, however, most vaccines fail to induce local
ocular immune responses and, without adjuvant, may induce a
state of immunological tolerance. Using epitope-based vaccines
delivered via the ocular mucosal (OM) route.74 The generation
of local and systemic peptide- and virus-specic T cells
conrmed the potent immunogenicity of peptides-CpG2007
formulation when applied via this route and suggest the clinical
feasibility of developing an OM delivery system using epitopebased vaccine.
Clinical Disease
Ocular herpetic disease may be primary neonatal, primary, or
recurrent. Primary disease is infectious disease of the nonimmune host, and recurrent disease occurs in the immune (or
previously immune) host and may be either infectious or
immune, or both (Table 49.1).
TABLE 49.1. Classication of Anterior Herpes Simplex Disease

I. Primary Infection
A. Neonatal
B. Primary (children, adults)
II. Recurrent Infection
A. Blepharitis
B. Conjunctivitis
C. Infectious dendritic or geographic epithelial keratitis
D. Sterile corneal neurotrophic ulceration
E. Stromal keratitis primarily immune
1. Interstitial keratitis
2. Immune rings
3. Limbal vasculitis
4. Disciform keratitis
F. Endotheliitis/trabeculitis primarily immune
G. Iridocyclitis primarily immune
Neonatal HSV infection

Approximately 1 in 10 000 infants is born with neonatal HSV,
20% type 1 and 80% type 2. The clinical manifestations of this
rare but usually devastating disease include local skin, eye, or
oral infection, central nervous system (CNS) disease, and disseminated HSV in which the visceral organs are also affected.
The local forms may accompany either of the other two forms.
Despite antiviral therapy earlier studies have reported a mortality rate of ~57% in disseminated disease and 10% in CNS
disease, with a very high rate of CNS damage in survivors.80,81
More recent studies by Freij et al and by Kimberlin et al
reported that while most exposure to the virus occurs in an
infected birth canal, 5% of infants acquire the infection in
utero.82,83 After an incubation period which can last as long as
24 weeks, neonatal HSV disease then manifests in one of three
ways: (1) disseminated disease, with visceral organ involvement
(including infection of the brain in two-thirds to three-quarters
of patients), (2) central nervous system disease (with no other
visceral organ involvement, but with skin lesions in two-thirds
of patients), or (3) disease limited to the skin, eyes, and/or mouth
(i.e., SEM disease).The mortality rate is 31% for disseminated
infection and 6% for localized central nervous system disease
with long-term neurologic sequelae seen in 17% and 70% of
survivors, respectively. Diagnosis is made by isolating the virus
from skin lesions or other involved sites. PCR detection
of viral DNA in cerebrospinal fluid (CSF) or serum is now the
diagnostic test of choice for central nervous system or
disseminated neonatal herpes. Supportive measures and
neuroimaging studies are often required.
Treatment is with high-dose intravenous acyclovir (60 mg1
1
kg day1 in three divided doses) for 3 weeks, with adjustments
made for infants with renal or hepatic insufciency. Infants
with disease localized to the skin, eyes, and mucous membranes can be treated for 2 weeks if the CSF PCR reaction assay
is negative for HSV DNA. Cesarean delivery will prevent
infection in infants when women have active lesions at the
onset of labor. Suppressive acyclovir therapy beginning at
36 weeks gestation is often prescribed for women with frequent
recurrences of genital herpes. Neonates delivered through an
infected birth canal should be screened between 24 and 48 h
of age with viral cultures of eyes, nasopharynx, mouth, and
rectum. If positive, they should be treated with acyclovir even
if asymptomatic.
Kimberlin et als study on the natural history of neonatal
HSV in the era of acyclovir was somewhat discouraging. The
investigators found that comparisons between patients treated
between 19811988 and 19891997 revealed that the mean
time between the onset of disease symptoms and initiation of
therapy has not changed signicantly from the early 1980s to
the late 1990s. It is this delay in thinking of HSV in any acutely
ill infant, not so much the lack of excellent treatment, that
causes the unchanged incidence of morbidity and mortality in
neonatal HSV.
Acute neonatal ocular HSV is most frequently a conjunctivitis often associated with ulcerative keratitis, which may
be diffuse microdendrites or serpiginous epithelial defects or
simply a punctate keratitis. Stromal involvement at or within
days of birth may occur but is extremely rare and usually
indicates intrauterine infection rather than infection during
passage through an infected birth canal.84,85 Other ocular complications that may occur acutely but more commonly after the
acute phase has passed include necrotizing chorioretinitis,
cataracts, optic neuritis, variable forms of comitant and noncomitant strabismus due to CNS damage and phthisis. The
appearance of opsoclonus may also be an early sign of HSV CNS
infection and warrants full radiologic and infectious disease
workup.86
CHAPTER 49
In contrast to the foregoing studies, the HEDS report on

recurrences as a risk factor noted that a history of epithelial
keratitis within the past year was not a risk factor for recurrent
epithelial keratitis, but that previous, especially multiple,
episodes of stromal keratitis markedly increased the probability
of subsequent stromal keratitis.71 Similarly, another HEDS study
differed in its interpretation of the role of factors stimulating
re-activation.72 Psychological stress, systemic infection, sunlight exposure, menstrual period, contact lens wear, and eye
injury were recorded on a weekly log. The exposure period was
considered to be the week before symptomatic onset of a
recurrence. No association was found between any of the other
exposure variables and recurrence. Psychological stress did not
appear to be a trigger of recurrences of ocular HSV disease. It
was concluded that recall bias substantially overestimated the
importance of factors such as systemic illness or stress that did
not have a causal association with ocular HSV.
641
SECTION 6
The diagnosis of ocular HSV must be considered in any

infant with nonpurulent conjunctivitis or keratitis, particularly
those in whom there has been fetal monitoring with a scalp
electrode, as this group appears to be at particular risk for all
forms of neonatal HSV infections.81,87 Evidence of focal ulcerative dermatitis should be sought by thorough examination of
the entire infant, as this may assist in the ocular diagnosis and
save the childs life through institution of earlier therapy.
Diagnostic scrapings and cultures and radiologic imaging as
described earlier are useful; serial serologic tests will conrm
diagnosis but only belatedly.
642
Therapy of neonatal ocular herpes

An emergency pediatric or infectious disease consultation
should be obtained. Although there is only a 6% incidence
of corneal scarring, about 37% of these children will have
visual acuity of less than 20/200 due to the long-term sequelae
of other ocular forms of HSV, which are best treated
systemically.80,88 This is quite independent of the life-saving
aspects of such therapy.
Therapy of focal ocular disease is with oral antivirals such as
acyclovir (ACV) and topical antivirals regardless of whether the
etiologic agent is type 1 or type 2 HSV. Both types are equally
sensitive to the commercially available agents. In a study to
evaluate the use of oral acyclovir in pediatric patients with HSV
keratitis, Schwartz and Holland reported seven pediatric
patients with ages ranging from 6 weeks to 5 years at the time
of presentation (mean 1.9 years).89 All patients received oral
ACV; six of seven patients also received topical antiviral medications. Three of seven patients had topical antiviral therapy
fail before being placed on oral acyclovir and the remaining four
patients were placed on oral acyclovir primarily. All patients
resolved their keratitis. Three patients were maintained on
prophylactic dosage of oral ACV because of recurrent disease or
because they had been chronically treated with topical corticosteroids for immune stromal keratitis. There were no adverse
drug reactions. The authors concluded that oral acyclovir is
effective and safe in treating HSV infectious epithelial keratitis
in pediatric patients. It is benecial in treating infectious epithelial keratitis and as prophylaxis while treating with topical
corticosteroids for immune stromal keratitis or for preventing
recurrent infectious epithelial keratitis.
Kimberlin et al reported use of intravenous ACV, 60 mg/kg
for 21 days in 72 infants under the age of 28 days.90 Six patients
dropped their absolute neutrophil count (ANC) to 5001000/mm3.
In all cases, the ANC recovered during continuation of acyclovir
at the same dosage or after completion of acyclovir therapy,
and there were no apparent adverse sequelae of the transient
neutropenia. However, decreasing the acyclovir dosage or
administering granulocyte colony-stimulating factor should be
considered if the ANC remains below 500/mm3 for a prolonged
period.
1. Intravenous acyclovir (60 mg1 kg1 day1 in three divided
doses) for 2 weeks if disease localized to eyes, skin, and
mucous membranes and CSF PCR reaction assay is
negative for HSV DNA.
2. Decreasing the acyclovir dosage or administering granulocyte
colony-stimulating factor should be considered if the ANC
remains below 500/mm3 for a prolonged period.90
3. Topical trifluridine 9x/day 14 days.
4. Topical antibiotic drops such as Polytrim qid.
5. Periocular skin lesions should be kept clean with warm
sterile compresses two to three times daily and the lesions
should be kept dry between compresses. Some physicians
advocate topical antibiotic ointments such as ophthalmic
bacitracin. This is effective in minimizing secondary
infection.85
Primary ocular HSV (POHSV)

Primary ocular HSV (POHSV) is an acute rst HSV infection of
the nonimmune host. It is to be differentiated from a rst
ocular occurrence in a patient previously infected with HSV
elsewhere, e.g., orally, as the latter patients exhibit disease
similar to recurrent forms. POHSV may present as a blepharitis, conjunctivitis, keratoconjunctivitis without, but more
commonly with, signicant periorbital skin involvement, and,
rarely, iridocyclitis.5,9193 As noted, < 4% of POHSV presents
as overt disease. But whether overt or not, it would appear that
all patients, once infected with HSV at any site, become viral
carriers with the agent residing in a latent state in the trigeminal ganglia and, if there has been an infectious keratitis, the
cornea.5,9499
Clinically, overt disease begins 39 days after exposure to an
infected carrier and typically manifests itself as an intense,
occasionally hemorrhagic vesicular (blistering) periocular
dermatitis or blepharitis, follicular conjunctivitis that may be
pseudomembranous and/or have geographic ulceration, corneal
ulceration, iritis, and a nonsuppurative preauricular adenopathy. The skin eruption remains fairly localized to the periocular
area in the immunocompetent host and is a self-limited disease
that resolves entirely without scarring and often without
specic therapy (Fig. 49.1).5,92,100102
A keratitis will develop in more than 60% of patients and,
due to the lack of immune inhibition, is usually atypical. There
may initially be diffuse punctate staining that converts within
24 h to multiple diffuse microdendritic epithelial defects, or
there may be serpiginous ulcers without clear-cut branching
effect covering much of the corneal surface (Fig. 49.2).
Iritis is uncommon in POHSV but, if it occurs, may produce
extensive atrophic damage.91
Therapy of POHSV Specic therapy is highly successful and
often results in healing with little to no scarring owing to
absence of a preprogrammed immune response. The most
commonly used drug is acyclovir because of ease of use and
compliance as well as efcacy. For ambulatory patients, therapy
is tailored according to age. For children less than 8 years of age,
oral acyclovir is administered at a dosage of 20 mg/kg every 8 h.
A pediatric suspension of 200 mg/5 mL is available. For children
FIGURE 49.1. Acute primary herpes simplex virus (HSV) blepharitis

with extensive vesicular eruption of the lids and the periorbital area
but no ocular involvement. Skin lesions healed within 3 weeks without
scarring.
times daily until resolved. Recurrences are managed in a similar

manner. Some physicians administer oral acyclovir at the doses
noted above in order to prevent frequent recurrences. Ocular
and mucocutaneous HSV infections in the immunocompromised host can be treated with either intravenous acyclovir
or one of the orally bioavailable antiviral therapies. For
hospitalized patients, therapy consists of IV acyclovir at 5 mg/kg
every 8 h for 74 days.104
1. Periocular dermatitis or blepharitis only:
A. Prophylactic acyclovir 400 mg po two to three times
daily (dependent on patient weight) for 10 days or until
lesions scabbed over. Use pediatric suspension
(200 mg/5 mL) at dose of 20 mg/kg in children less
than 60 lb (28 kg.) (1 lb = 2.2 kg).
B. Warm wet soaks 5 min bid with drying of lesions
allowed between soaks. General good hygiene. Topical
ophthalmic bacitracin may be used on badly ulcerated
skin.
2. Corneal ulceration (dendrites, geographic ulcers):
A. Oral antivirals for 10 days to 3 weeks as in 1.A above.
B. Add topical trifluridine ve to nine times daily if
patient immunocompromised and therapeutic response
to oral antivirals not satisfactory by 5 days of oral
treatment.
C. Topical antibiotics bid for corneal ulceration.
D. Cycloplegics, cyclopentolate, or scopolamine bid if iritis
is present.
E. Fox shield or hand restraints in young children as
needed.
b
FIGURE 49.2. (a) Acute primary HSV epithelial dendritic keratitis with
diffuse serpiginous ulcers. (b) Same eye 3 weeks later with complete
epithelial healing and a clear stroma.
60 lb or over dosage can be the standard 400 mg po tid to 5id

either by pills or pediatric suspension, with the higher doses
being used in patients with altered immune systems, e.g., excema,
asthma, or immunosuppression. Topical trifluridine (Viroptic)
ve to nine times daily for 23 weeks may be added after about
5 days if response is slow or steroids are in use.43,103,104
Although no large controlled studies have been performed
with valaciclovir or famciclovir, these drugs have been shown to
be clinically equal to acyclovir systemically and, valciclovir,
superior topically. However, they are notably more expensive.
For difcult cases, however, it is noteworthy that famciclovir
has a longer intracellular half-life than acyclovir, and valciclovir,
which is hydrolyzed back to ACV results in ve times the
bioavailability of the latter drug.105107 Loutsch et als rabbit
study on oral FCV treatment showed that the drug in doses
comparable to 120 mg bid in humans signicantly reduced
the severity of corneal lesions, reduced the number of HSV-1
genomes in the TG, improved survival, and therefore may
be benecial in reducing the morbidity of HSV keratitis in
the clinic.
Dosage recommendations have not been established for
young children. For postpubertal children, dosage should mirror
that of adults. Valaciclovir is administered at 500 mg twice
daily. Famciclovir is administered at 125 mg three times daily.
Herpes simplex keratoconjunctivitis is treated with topical
triflurothymidine. Two drops are applied to the infected eye ve
Latency Within the rst 48 h of primary infection the virus

travels as an unenveloped particle by retrograde axoplasmic flow
to the sensory (trigeminal) ganglia, ciliary ganglia, mesencephalic nucleus of the brain stem, and in some cases, to the
sympathetic ganglia where it enters a latent or dormant
state.48,94,95,101 The trigeminal ganglion (TG) is one of the main
sources of virus for recurrent disease particularly of the ocular
adnexa and cornea. Re-activation of HSV latent in the ciliary or
superior cervical ganlion of the EdingerWestfall nucleus is
more likely to result in iritis. Virus in the suprachiasmatic or
paraventricular nuclei may cause retinitis or encephalitis.75,108
As noted, HSV may enter the TG via any of the three major
divisions of the trigeminal nerve, mandibular, maxillary, or
ophthalmic. Thus, an initial orofacial infection with HSV may
establish latent virus in the ophthalmic division of the
trigeminal ganglion and subsequently lead to ocular infection
after re-activation of virus with spread down the nerves to the
eye rather than those to the nose or mouth, or down all three
(Fig. 49.3).29,109,110 Bilateral disease is rare, occurring in about
2% of patients.101
Further, several studies indicate that the cornea may also
serve as a site of viral latency capable of re-activation and that
serial recurrences are due to re-activation of the same latent
virus, not new exogenous infection.98,111114 In a study by Rong
et al, corneal specimens from 18 patients with quiescent herpes
simplex keratitis (HSK) were obtained at keratoplasty. PCR
amplication followed by Southern blot hybridization detected
HSV-1 genome in these human corneal samples. The DNA
sequences from either the TK or the LAT gene were identied
in 15 of 18 HSK corneas (83%). These data indicate that the
HSV genome was retained, at least in part, in human corneas
during quiescent HSV infection, giving further support to the
concept of corneal extraneuronal latency which may, in turn, be
one source of herpetic disease in previously nonherpetic grafted
eyes. This has further been conrmed by Remeijer et al and
others.12,96,99,112114 It has also been shown that viral infection
CHAPTER 49
Recurrent ocular herpes
643
SECTION 6
FIGURE 49.3. Distribution of the rst, second, and third divisions of

the trigeminal nerve to the eye, forehead, periorbital area (V-I), the
maxillary area (V-II), and the mandibular area (V-III). HSV or VZV may
travel anterograde down any division. A large posterior nerve trunk
leads from the TG to the central nervous system: pons, midbrain, and
spino-mesancephalic pain pathway critical in zoster ophthalmicus.
644
of periocular tissue and subsequent disease development occurs

by zosteriform spread from the cornea to the periocular tissue
via the trigeminal ganglion rather than by direct spread from
cornea to the periocular skin.115 In some contrast, Polcicova et
al constructed an HSV mutant with very limited capacity to
move in neuronal axons.116 The virus replicated and spread
normally in the mouse corneal epithelium and to the trigeminal
ganglia. However, it was unable to return from ganglia to
the cornea and failed to cause periocular skin disease, which
requires zosteriform spread from neurons. It did, however, cause
infectious keratitis demonstrating that herpes keratitis can
occur without anterograde transport from ganglia to the cornea,
and that at least some local recurrences are mediated by virus
persistent in the cornea.
Whether latency is a static or smoldering but dynamic state
has not yet been determined. Data indicate that at least
the ICPO or junctional region of the viral genome is retained in
the latent state and that viral RNA transcribed from the
opposite region on the DNA strand may be involved in but
not required for maintenance of latency or prevention of reactivation.5,99,111,117,118 Both animal and human studies indicate
that re-activation from latency to infectious virus production
results in the creation of large amounts of this RNA (antisense
RNA). This leads to a cascade of events resulting in the
development of viral polypeptides and ultimately to infectious
progeny virions.119,120
HSV re-activating factors The mechanisms of viral reactivation are not well understood. A common theory is that
the ganglion (and possibly corneal) cells harboring the HSV
genome are in a nonpermissive state for replication during
latency. A yet-to-be-established stimulus then allows the
ganglion cell to become permissive for active HSV replication.
On a cellular level, this may translate into a deciency of
immune competence or intracellular messenger systems.
A wide variety of trigger factors have been reported as stimulating re-activation; all appear to be various forms of physical
or emotional stress or immunoincompetence. Reported trigger
factors include fever, ultraviolet light (sunlight, tanning
machines), cold wind, systemic illness, surgery, menstruation,
emotional stress, minor local trauma, trigeminal root ganglion
section for trigeminal neuralgia, and immunosuppression, from
either endogenous disease or iatrogenic drug management of
disease requiring immunosuppression for control.79,121128 As
noted in the section on Epidemiology a 2000 HEDS study
discounted most of these factors as meaningful and interpreted
the data as overreporting on the part of the study cohort.72 The
issue of trigger factors is unresolved and many patients can
clearly predict recurrent disease following a given event.
A more recently proposed HSV trigger factor is topical antiglaucoma drugs, particularly the prostaglandin analogs. Morales
et al reported two cases of HSV blepharitis with latanaprost
use, and Kaufman et al found that experimental HSV keratitis
was worsened and recurred more often with latanaprost
therapy.129,130 Schumer, however, has noted that the causal
relationship has not been scientically established between this
drug therapy and rare side effects; repeated re-challenging with
masked controls in a clinical study is required.131 Further, two
more recent studies negate these reports.132,133 Using a medical
claims database, Bean et al screened 93 869 eligible glaucoma
patients using 21 different ocular hypotensive agents. In all,
411 patients had an ocular HSV episode; 272 of 411 patients
had at least one ocular hypotensive agent dispensed prior to the
OHSV event but not preceding the event by less than 7 days.
The overall HSV event rate was just 0.11%, and there was no
signicant association between HSV event rates and agent use.
Prevalence rates were slightly more than 160/100 000 population from 1999 to 2001 which is no higher than that found in
the general population.
Laser photokeratectomy has also been strongly implicated as
a risk factor for HSV and thus far there have been no reports
refuting this.134139 Studies show that the type of laser is
irrelevant and that patients with a history of ocular HSV who
are undergoing laser procedures should be protected with
prophylactic oral antiviral therapy starting at least 24 h pre-op
and continuing for abut 2 weeks. The author has had one
patient with no history of ocular HSV develop severe dendritic
and ultimately stromal HSV keratitis following argon laser
iridectomy in one eye and YAG iridectomy in the other (DPL
unpublished 1990). This implies not only that any type of laser
treatment of the eye may reactivate latent HSV but that all laser
patients should be closely monitored for re-activation of latent
ocular HSV even in the absence of a history of the disease.
HSV blepharitis and dacryoadenitis

Recurrent HSV blepharitis in the immunocompetent patient
differs from primary disease in being focal cluster(s) of vesicles
or weeping ulcers along the lid margin or skin of the lids, which
shed virus for only 23 days and last for only 57 days
(Fig. 49.4) in comparison to primary disease which is diffuse,
sheds virus for 10 days and resolves over 23 weeks. There may
or may not be involvement of other ocular structures.140,141 The
nasolacrimal system may be partially or completely scarred
over, leaving the patient with epiphora which is usually
FIGURE 49.4. Acute recurrent HSV blepharitis with multiple early

clustered vesicles. Recurrent disease remains focal in
immunocompetent patients.
Therapy of herpetic blepharitis and postherpetic lacrimal

obstruction: topical and systemic antivirals:
1. In the absence of ocular involvement, therapy is more to
protect the globe as acyclovir 400 mg po tid5id or
prophylactic trifluridine ve to six times a day or until the
lesions have scabbed. HSV blepharitis is a supercial
infection, unlike herpes zoster, and it heals without
scarring (Fig. 49.5).
There have now been clinical reports with demonstrated
efcacy in preventing herpetic blepharitis, Tsao et al
reported an 11-year-old boy with monthly bilateral
recurrent HSV type 1 blepharitis for more than 10 years.
He had a normal immunological examination. Only
topical acyclovir ointment treatment proved adequate for
controlling the monthly recurrent disease without corneal
involvement or other sequelae.143 In lieu of topical ACV
which is not available in the USA, oral ACV 400 mg three
to ve times daily or trifluridine 1% 59 times daily for
1014 days should resolve the acute infection Studies in
therapy of recurrent herpes simplex labialis with oral ACV
may also be reasonably applied to therapy of patients with
herpetic lid involvement. Spruance et al studied 174
nonimmunocompromised patients and demonstrated that
initiating ACV 400 mg po ve times daily for 5 days
within 1 h of the rst sign or symptoms of labial
recurrence signicantly hastened resolution of lesions and
duration of pain.144 It did not, however, actually block the
development of lesions or affect maximal lesion size.
Nonetheless, it was felt that oral ACV signicantly
alleviated some of the annoying clinical manifestations of
HSV labialis. In the HEDS study on oral acyclovir (400 mg
po bid) for the prevention of recurrent herpes simplex the
cumulative probability of a recurrence of primarily
orofacial HSV disease was lower in the acyclovir group
than in the placebo group (19% versus 36%, P <0.001).145
2. Postherpetic lacrimal obstruction must be corrected
surgically if irrigation and probing does not sufce. The
FIGURE 49.5. Fourth recurrence of acute HSV vesicular lesion just

below the right lower lid in a 1-year-old child. Lesions resolved within
2 weeks and recurred frequently during the rst 2 years of life but
without direct ocular involvement.
study by McLean et al on 160 such patients with a mean

age of 31 years and all with a history typical of primary
herpes simplex blepharoconjunctivitis showed that
procedures that use any remaining unaffected portion of
canaliculi, such as DCR, with anterograde or retrograde
intubation, as well as the more commonly used Lester
Jones tube are justied in curing this problem.146
HSV conjunctivitis
An acute primary or recurrent follicular conjunctivitis due to
HSV may be seen with or without preauricular lymph node
involvement and without involvement of other ocular
structures.147,148 Signs and symptoms include a watery discharge and a mononuclear cell inltrate in scrapings; in some
patients, dendritic or dendrogeographic ulcers on the conjunctiva are revealed by rose bengal or fluorescein stain (Fig.
49.6). Rarely a pseudomembrane will form, and there may be
some atypical supercial punctate keratitis or more severe
corneal disease. It has also been reported that HSV may occur
in conjunctival flaps appearing as round ulcers and culture
positive for HSV.149 Diagnosis in the absence of other ocular
signs of herpes may be made based on index of suspicion from
a previous history of herpes, the appearance of herpetic lesions
elsewhere on the face, or resistance to routine antibiotic therapy.
CHAPTER 49
amenable to treatment with irrigation and probing but may

require surgical repair. There has been one case reported of
herpes simplex virus dacryoadenitis in an immunocompromised patient. Higher-dose po acyclovir was required for
therapy.142
Therapy of herpetic conjunctivitis Treatment is full antiviral

therapy with either oral medications such as acyclovir 400 mg
po tid 1014 d or trifluridine drops (see Table 49.1). The
resolution of disease usually occurs within a week to 10 days.
HSV viral keratitis

Classication of HSV keratitis The varying forms of HSV
keratitis have been debated for decades with not much difference made among the proposed systems during those years. In
1967, Patterson et al differentiated epithelial from stromal
disease and the need for antivirals in the former and steroids
in the latter.150 PavanLangston then described a system of
645
FIGURE 49.6. Acute HSV recurrent corneal dendrogeographic ulcer

with extension onto the conjunctiva.
primary and recurrent disease and disease categories of the lids,

conjunctiva and, in the cornea, going from most supercial to
deep ocular structures: epithelial infection, trophic postinfectious ulcers, stromal immune disease (interstitial and disciform), and uveitis.151 Liesegang subsequently further rened
these categories with a number of subcategories including
disciform as central endotheliitis.152 Most recently, Holland and
Schwartz proposed a classication system based on the
anatomy and pathophysiology of the specic forms of HSV
keratitis.153 Anatomically, the primary level of corneal
involvement whether epithelium, stroma, or endothelium was
described. Pathophysiologically, the cause of the inflammation
whether immunologic, infectious, or neurotrophic, was determined. They concluded that there are four major categories of
HSV keratitis: (1) infectious epithelial keratitis, which is made
up of cornea vesicles, dendritic ulcer, geographic ulcer, and
marginal ulcers, (2) neurotrophic keratopathy, which includes
punctate epithelial erosions and neurotrophic ulcer, (3) stromal
keratitis which is subdivided into necrotizing stromal keratitis
and immune stromal keratitis, and (4) endotheliitis, which has
three clinical presentations: disciform, diffuse, and linear.
SECTION 6
Corneal epithelial infectious ulcers
646
Symptomatically, patients often present complaining of tearing,

photophobia, irritation, and occasionally blurred vision. As the
ocular infection may present as conjunctivitis or blepharitis
alone prior to onset of keratitis, the history should include
previous corneal ulcers, iritis, nasal or oral cold sores, genital
ulceration, recent use of topical or systemic steroids or
immunosuppressive agents, and potential for immunologic
deciency states (AIDS, organ transplantation, chronic eczema,
asthma, other atopy, malignancy, blood dyscrasia).
Bilateral ocular HSV fortunately is not common and is not on
the increase.29,154 In a study on 30 patients with keratitis, OU
40% were atopic and stromal keratitis occurred in 40% of the
eyes and recurrent ulceration in 68%. Seventeen percent of
patients lost vision to 6/60 or less due to corneal scarring. In a
larger study on the clinical outcomes of HSV in 544 patients
only 1.3% had bilateral disease. In these seven patients, the age
at onset of keratitis ranged from 7 weeks to 46 years, with a
mean of 19.3 years. However, ve patients had systemic atopy,
two patients had severe ocular rosacea, and two had systemic
immune disorders.155 It was also noted that patients with
bilateral herpetic corneal infections had more protracted clinical
FIGURE 49.7. Recurrent HSV dendritic limbal ulceration. The ulcer

shows the classic terminal bulbs not seen in herpes zoster dendrites.
courses than patients with unilateral disease, and a higher

incidence of complications such as opacication, neovascularization, and corneal thinning or perforation. Further, long-term
prophylactic antiviral (ACV 400 mg po bid) treatment reduced
the incidence of recurrence in four of the bilateral group of
patients to an average of 1/1.7 years.
Dendritic (branching), dendrogeographic, or geographic
ulcerations of the corneal epithelium are caused by live HSV
replication in these cells. The disease may initially appear as a
punctate keratitis evolving into a dendritic keratitis, which may
progress to the dendrogeographic or geographic state. Steroid
enhancement of a dendritic process often results in a geographic
ulcer, which is commonly referred to as an ameboid ulcer,
indicating enhancement by steroid therapy. Dendritic ulcers
may be single or multiple and have linear branches that
characteristically end in bead-like extensions called terminal
bulbs, unlike the narrow, trailing ends of herpes zoster dendrites
(Fig. 49.7).50,156 Rose bengal stain will highlight not only the
areas of epithelial absence but also the swollen infected
epithelial cells surrounding the ulcerated area. Conversely,
fluorescein dye stains primarily the central ulcerated area in the
absence of cells but may reveal punctate staining over the
surrounding damaged epithelium.157 Under the dendrite, it is
not uncommon to note a faint stromal inltrate in the shape of
the epithelial lesion. This may represent diffusion of soluble
antigen in the stroma, which ultimately leaves a dendritiform
ghost scar in the form of the previous dendrite.
Corneal anesthesia may give some clue to diagnosis in
equivocal cases. In a study Kodama et al evaluated 25 control
eyes with normal corneas, eight eyes with possible herpes
simplex keratitis, and 48 eyes with corneal lesions: recurrent
erosion, supercial punctate keratitis, marginal ulcer, and
follicular keratoconjunctivitis in whom herpes simplex keratitis
was not suspected. On immunofluorescent staining, all 25
control subjects had negative reactions, all eight eyes suspected
of having herpes simplex keratitis had positive reactions, and
23% of the 48 eyes not suspected of having herpes simplex
keratitis had positive reactions. Of the 11 eyes not suspected of
having herpes simplex keratitis but that had positive reactions
on immunofluorescent staining, nine had recurrent erosions
and the remaining two eyes had supercial punctate keratitis.
Of the eight eyes with possible herpes simplex keratitis, seven
(88%) had decreased corneal sensitivity. Of the 11 eyes not

suspected of having herpes simplex keratitis but that had
positive reactions on immunofluorescent staining, eight (73%)
had decreased corneal sensitivity. Of the 37 eyes not suspected
of having herpes simplex keratitis that had negative reactions
on immunofluorescent staining, 11 (30%) had decreased
corneal sensitivity.158
Van Horn et al have reported little inflammatory cell reaction
in the form of PMNs, few to no lymphocytes, but many free
viruses lying in intracellular and extracellular locations,
particularly in the basal epithelium.159 The mechanism for the
branching or dendritic formation has never been adequately
explained. There appears to be no relationship to the neuronal
distribution, and current evidence suggests that it may simply
be related to the viral pattern of linear spread by contiguous cell
to cell movement.160
In milder epithelial keratitis, stromal involvement is usually
absent or conned to the previously mentioned dendritiform
stromal haze localized to the subepithelial area. However, as
discussed later under factors affecting the severity of HSV
keratitis, depending on HSV strain virulence and host immune
response, there may be considerable stromal edema and iritis
associated with epithelial disease. In these eyes, stromal immune
reaction with consequent scarring is more likely. If the
ulcerative lesion lies within 2 mm of the limbus, it is usually
slower to heal than is a central herpetic ulcer. Similarly, these
peripheral lesions are predisposed to chronic trophic ulceration
for reasons that are unknown.
Contrary to common opinion, recurrent epithelial infections
are not always due to the patients original HSV strain.
Remeijer et al have reported a study on 30 patients in whom
sequential corneal HSV-1 isolates revealed that 63% were
genotypically the same from recurrence to recurrence while 37%
were actually genetically different.63 Four of 11 patients in this
latter group had undergone keratoplasty between recurrences
but otherwise no other risk factor could be identied for
infection with exogenous strains.
TABLE 49.2. Herpes Simplex Keratouveitis Treatment Summary

1. Infectious periocular dermatitis or corneal ulcers
A. Oral antivirals: acyclovir (Zovirax) 400 mg, tid5id, or
famciclovir (Famvir) 125250 mg po bid or valaciclovir
(Valtrex) 500 mg po bid 714 d (add Topical antiviral prn
immunosuppression/eczema)
B. Topical antiviral: 1421 d
1. Trifluridine (Viroptic) 1 GTT 9 /d 5 d 5 /d 11 d if
improving
C. Antibiotic, e.g., Polysporin ung, HS, or quinolone gtt BID)
D. Treat nonscarring recurrent dendrites each occurrence
(14 d)
2. Neurotrophic ulcer (sterile epithelial defects, hypesthesia)
A. Nonpreserved lubricants (articial tears, tear gel/ung
46 /d): Abiotic, e.g., PS ung, or quinolone
B. T-SCL (Permalens, Kontur, Acuvue, Focus Night & Day):
Polytrim gtt bid, quinolone
C. Tissue glue (Dermabond, Epidermglu): PRN thinning (Plano
T SCL)
D. Mild steroid PRN stromal immune edema: Antibiotic gtt
3. Immune limbitis, stromal IK, Wessley rings, disciform edema,
endotheliitis
A. Steroid gtts* qid: qod in slow taper for inflammation
B. Acyclovir 400 mg po bid or famciclovir 125 mg bid 1 yr +
Prophylaxis: stromal recurrence or scarring dendrite
4. Iridocyclitis
A. Steroid* gtts with slow taper as in 3A above
B. PO prednisone 60 mg 20 mg/d over 14 d PRN ulcer
(Rx Ulcer as above)
C. Acyclovir 400 mg po bid prophylaxis
5. Graft survival
A. Interrupted sutures
B. Intense post-op steroid
C. Post-op acyclovir 1218 M
D. Antivirals c rejection rx
*Strong steroid = prednisolone, 0.1%, dexamethasone.
Moderate steroid = rimexolone (Vexol), lotoprednol 0.5% (Lotemax).
Weak steroid = lotoprednol 0.2% (Alrex), FML 0.1% gtt, or 1% ung.
The vast majority of patients will respond well to topical or oral

antivirals alone as outlined in Table 49.2. Occasional patients
are prone to frequent recurrences of infectious keratitis, i.e.,
more than two to three times per year.
Earlier masked controlled trials by Collum et al and Hung
et al indicated that oral ACV doses ranging from 400 mg 5id to
200 mg po ve times daily for two to three weeks were effective
in resolving infectious herpetic epithelial keratitis.161,162 The
reported improvement in stromal keratitis and uveitis in
Schwabs report may well be due to secondary improvement of
deeper inflammatory reaction after healing of epithelial
disease.163 Patients were receiving concomitant topical steroids.
This study did also indicate that patients with multiple
previous recurrences became free of disease so long as oral ACV
was given. Patients had been carried for up to 17 months on
dosages of 200400 mg daily with only one breakthrough when
a patient was reduced to a 200 mg daily dosage. Three patients
had prompt recurrence of herpetic epithelial and stromal
disease on discontinuation of ACV but were subsequently
controlled with reinstitution of the oral medication.
In more recent studies, Colin et al found similar results in
open study of acyclovir treatment (200 mg po qid) ranging
from 1 to 24 months. Forty patients were treated because of
frequently recurring herpetic keratitis, 36 were treated because
of known risk factors, 23 were treated after ocular surgery
needing topical corticosteroid therapy in the postoperative
period, and 91 patients were treated after penetrating keratoplasty for herpetic keratitis. Oral acyclovir signicantly lowered
the incidence of recurrences of herpetic keratitis in selected

patients.164
In a self-controlled, prospective and retrospective study, Simon
and Pavan-Langston have reported that oral ACV 400 mg bid for
up to 4 years signicantly reduces the incidence of recurrent
ocular infections in grafted and nongrafted corneas without
adverse side effects.167 These doses correspond with those found
in similar long-term studies on genital (upto 9 years) and labial
HSV.144,165167
Similar dosing of acyclovir, 400 mg po tid, was the one used
in the HEDS 2000 study on oral acyclovir for herpes simplex
virus eye epithelial keratitis and stromal keratitis.168 This randomized, double-masked clinical trial enrolled 703 immunocompetent patients with prior HSV eye disease within the
preceding year; assigned 357 patients to receive oral acyclovir,
800 mg/day, and 346 to receive placebo. Follow-up was a 12month treatment period for the development of HSV eye
disease. The data revealed that the cumulative probability of a
recurrence of any type of ocular HSV disease during the 1 year
treatment period was 19% in the acyclovir group compared with
32% in the placebo group. The greatest benet was among
patients with the highest number of prior episodes of ocular
HSV disease. The benet in preventing stromal keratitis was
seen solely among patients with a history of stromal keratitis.
There was no therapeutic effect of the antiviral on active
stromal keratitis.
A summary of the general data-based consensus is that the
dose for acyclovir in acute infectious HSV keratitis is 400 mg po
CHAPTER 49
Therapy of infectious epithelial herpes in

immunocompetent patients
647

tid5id, or for immunosuppressed or atopic patients 5id for 10
days to 3 weeks depending on response.5,169,170 (see section on
Therapy of HSV in Immunocompromised, ACV-Resistant, and
Atopic Patients below). Physicochemical methods of removing
infected corneal epithelium (debridement) are effective, but
adjunctive virucidal agents are needed to avert recrudescent
epithelial keratitis. Although there is some positive evidence,
whether debridement in combination with antiviral therapy is
more benecial than antiviral chemotherapy alone but this
remains inconclusive.170 What has been shown, however, is
that denuded epithelial basement membrane is partially
disassembled and may extend healing time.171
1. Gentle debridement (moist Weck cell sponge) especially in
atopic patients.
2. Oral antiviral: acyclovir 400 mg po tid to 5id for 1014 days.
Alternative oral antivirals: famciclovir 250 mg po bid or
valaciclovir 500 mg bid. Alternatively, trifluridine drops
(TFT) 9id5id over 14 days are very effective but less
convenient. Start sid if healing.
3. Topical antibiotic bid while ulcers present.
4. Mydriatic-cycloplegic (homatropine, cyclopentolate) bid prn
iritis.
5. Treat each infectious recurrence as it occurs unless scarring
occurs. If scarring treat acute infection and then put
patient on 1 year (or longer if needed) of prophylactic oral
antiviral.
6. In atopic or immune-compromised patients with frequent
recurrences, patients with bilateral disease, stromal
disease, or during high-risk periods, especially fever,
chemotherapy, ocular or other surgery in patients with a
history of ocular herpes, initiate antiviral prophylaxis
(400 mg po bid or famciclovir 125250 mg po bid) for
1 year or longer, or for high-risk periods, treat for 23
weeks beyond that period and at tid5id doses.169,155,172
Valaciclovir is best avoided in immune-compromised
patients because of risk of hemolytic uremia.
7. In the absence of stromal immune disease or iritis, steroids
are contraindicated in the presence of infectious epithelial
keratitis.
SECTION 6
Neurotrophic keratopathy and trophic ulceration
648
As a consequence of protracted or repeated episodes of herpetic

keratitis, either simplex or zoster (among many other causative
illnesses), a condition known as neurotrophic keratopathy often
develops. It is most commonly dened as an abnormality in
the peripheral or central nervous system which results in a
dysfunctional ocular surface.173 It has been shown repeatedly
that dysfunctional corneal innervation results in tissue degeneration, i.e., neurotrophic keratopathy. A sensory nerve supply
is crucial for optimal tissue function. There are many other
proposed and very probable factors involved in this condition
including dry eye due to reduced lacrimal secretions, inadequate
and infrequent blink reflex, defective epithelial cell metabolism
with consequent inability to withstand even mild to moderate
adverse events, and loss of the trophic interactions and
neurochemicals due to damage or loss of the corneal nerves.
Corneal neuronal anatomy, function, and mechanism of
neurotrophic disease To understand and treat this condition
some knowledge of the corneal neuronal anatomy and function
is needed. Postmortem studies reveal that nerve ber bundles in
the subbasal plexus of the human cornea form a regular dense
meshwork with equal density over a large central and
centralperipheral area. Because of their size, the majority of
the bers can be classied as C-bers which are nonmyelinated
bers in the cornea but fast conductors of sensory neuronal
signals.174 In studies of explanted trephinated human corneas
with Fuchs dystrophy taken at keratoplasty staining by the

Life/Dead-Kit gave an accurate picture of the nerves in the
central human cornea (Fig. 49.8). The thin sensory nerves,
derived from the rst and to some extent second division of
the trigeminal nerve, run parallel to Bowmans layer in the
subepithelial plexus.175 Passing vertically through Bowmans
layer, A-delta- and C-bers can be distinguished by ber
diameter, and, while running in the basal epithelial plexus, by
their spatial arrangement. A-delta-bers run straight and
parallel to the Bowman layer underneath the basal cell layer.
C-bers, after a short run parallel to the Bowman layer, send off
multiple branches penetrating epithelial cell layers, ending
blindly in the supercial cells.
While few stromal nerves can be visualized even by tandem
confocal microscopy, the critical subbasal nerve ber bundles
made of axons coated by a Schwann cell sheath can be detected
fairly easily, serve as landmarks for evaluating corneal nerve
density, and demarcate the epithelium from the stroma. Studies
by Muller et al reveal that there are about 6000 nerve bundles
in the human subbasal plexus each of which gives off up to
seven axons resulting in between 19 00044 000 axons in the
sub-basal plexus. These in turn give off 1020 nerve terminals
which may be extrapolated to result in 315 000630 000 or
roughly 7000 nociceptors (pain receptors and transmitters)
per mm2 making the cornea the most highly innervated
structure in the body (including tooth dentine).176
Interactions among corneal cell types make critical contributions to development, homeostasis, and wound healing in
the cornea.177 Many interactions are mediated by cytokines,
growth factors, and chemokines. The best characterized are
stromalepithelial interactions between epithelial cells and
stromal cells such as keratocytes, keratoblasts, and myobroblasts. Interactions also occur between corneal nerves and
epithelial cells and between corneal cells (epithelial cells and
stromal cells) and corneal immune cells. Epithelial to stromal
interactions are mediated by cytokines, such as interleukin-1
(IL-1) and soluble Fas ligand that are178 released by corneal
epithelial cells in response to injury. Other, yet to be identied,
cytokine systems may be released from the unwounded corneal
epithelium to regulate keratocyte viability and function. IL-1
appears to be a master regulator of corneal wound healing
that modulates functions such as matrix metalloproteinase
FIGURE 49.8. Corneal neuronal anatomy showing stromal nerves

penetrating Bowmans layer to branch and form the critical subbasal
plexus which, in turn, sends branches vertically upward between the
basal cells to the wing cell layer.
Adapted from Muller L et al, Architecture of Human Corneal Nerves IOVS 38 (5):
985994, 1997.

Goins has recently reported a uselful algorithm for the
Diagnosis and Treatment of Neurotrophic Keratopathy much
of which the author (DP-L) has adapted below.173
A. Slit-lamp exam
Stage 1: conjunctival hyperemia, diffuse interpalpebral
punctate stain, epithelial roughening, possible stromal
vascularization.
Stage 2: persistant oval interpalpebral epithelial defect (PED)
with thickened, gray, rolled edges, possible anterior
chamber cell and flare, possible stromal edema.
Stage 3: enzyme release from corneal and inflammatory
cells resulting in stromal lysis, melting, and possible
progression to perforation.
B. Schirmer test
1. Often reduced especially with topical anesthesia.
2. Reduced naso-lacrimal reflex tearing (especially with
zoster).
C. Aesthesiometry
1. Quantication of ve different sectors of the cornea
may be done most easily with Cochet-Bonnet
aesthesiometer, or, if not available a standardized
3 cm 8 pound nylon line afxed to a wooden or plastic
handle (DP-L).
D. Tandem scanning confocal microscopy
1. Shortened nerve stubs in subbasal plexus.
2. Absent or severely reduced subbasal nerve plexus bers.
Loss of corneal sensation and structural damage result in
recurrent epithelial erosions, persistent sterile ulceration, and
in some patients, melting of the corneal stroma, which may
ultimately lead to perforation. Stromal scarring and neurotrophic changes inhibit epithelialization even without stromal
inflammation or viral replication although presence of either of
the latter will obviously contribute to worsening the trophic
disease.188,189
Corneal epithelial cells do not interdigitate with their
basement membrane but simply lie on top, attached by only
electron-dense hemidesmosomes.190,171 Such basement membrane damage, under the best of circumstances, takes at least
1215 weeks to heal, during which time closure of the overlying
epithelial defect is retarded.191,192 During this period of
ulceration, diseased epithelium and leukocytes attracted by
chemotaxis may release collagenases, proteases, and various
other cytokines and chemokines with consequent collagen
melting and thinning of the cornea (Fig. 49.9). Persistent
epithelial ulceration is further aggravated by poor wetability of
CHAPTER 49
production, HGF and KGF production, and apoptosis of

keratocyte cells following injury. Although investigations are
limited, it is likely that there are interactions between corneal
endothelial cells and keratocytes in the posterior stroma.
The existence and function of these nerves, along with a few
sympthetic nerves which also feed the cornea, is critical to
the health of the various corneal cells and, therefore, healing
abilities of the corneal tissues. The corneal neurons, epithelium, and stromal keratocytes are all mutually supportive as
prime sources of at least 17 different neuropeptides and neurotransmitters. These include, among others, substance P (SP),
calcitonin gene-related peptice (CGRP), epithelial growth factor
(EGF), pituitary adenylate cyclase-activating peptide (PAPCAP),
vasoactive intestinal polypeptide (VIP), opioid growth factor
(OGF), neurotensin (NT), galanin, the neurotrophins: a. nerve
growth factor (NGF), b. NT-3, c. NT-4, and d. brain derived
neurotrophic factor (BDNF) from the sensory nerves.176,179181
The neurotrophins are exclusively located in the basal limbal
epithelium making it likely that they specically regulate
corneal epithelial stem cells. The sympathetic nerves supply
acetylcholine (ACH), norepinephrine, and neuropeptide Y
(NPY). Additional neurochemicals are supplied from the conjunctival, lacrimal, accessory and orbital glands, and the corneal
cells themselves.
The function sources of these neurochemicals are shown in.
They are the movers and shakers in increasing cell mitogenesis
and migration, DNA synthesis, neurite extension and survival,
keratocyte proliferation, regulation of epithelial stem cells,
keratocyte modulation of epithelial cell function, endothelial
protection from oxidative stress, and they decrease apoptosis.
Within 1224 h of corneal nerve impairment or loss, the
epithelial cells swell and lose their microvilli, and begin to
slough at an accelerated rate. Histology shows wing and basal
cell loss of tonolaments and decreased surface membrane
interdigitations, abnormally rounded cells, and dilated intercellular spacing.176,182,183 Clinically, a supercial punctate
keratitis develops. Denervation of the cornea clearly impairs the
ability of the epithelium to heal after injury and newly healed
tissue is at high risk of spontaneous breakdown. Tandem
scanning confocal microscopy of human corneas reveals the
denervation seen post-HSV keratitis with reduced ber count
and shortening of the bers themselves.176184
A further complication of corneal and/or anterior segment
denervation is dry eye.185 In studies on unilateral dysfunction
of the rst (ophthalmic) division of the trigeminal nerve
(neurotrophic keratitis) reduced aqueous tear production was
noted. Heigle et al reported sensation in the brow and upper lid
skin, nasal mucosa and cornea was absent on the affected
side of patients with neurotrophic keratitis, but was intact in
groups without keratopathy.186 Schirmer test values were also
signicantly reduced (p < or = 0.05) in eyes with neurotrophic
keratitis while rose-bengal staining scores were signicantly
increased compared with the control groups. The authors
concluded that neurotrophic keratitis is associated with reduced
cutaneous, nasal mucosal and corneal sensation on the affected
side, resulting in marked reduction in aqueous tear production
due to loss of the nasal-lacrimal reflex, all of which contributes
to corneal epithelial pathology in neurotrophic keratitis.
Confocal microscopy of dry eyes in Sjgrens syndrome, another
syndrome with decreased corneal sensation, shows distinct
changes in corneal nerves including increased nerve number,
tortuosity, and branching, suggesting an attempted nerve
regeneration. There was a strong correlation between the
abnormal nerve morphology and the degree of dry eye providing
evidence for the abnormal corneal sensation in dry eye which
could also be interpreted as neuronal changes preceding and
causing dry eye.187
FIGURE 49.9. HSV postherpetic trophic ulceration with 50% depth

stromal melt.
649
SECTION 6
650
such a damaged corneal surface and abnormal cellular metabolism due to disruption of the neurosensory pathways.177,193
The greatest danger, then, posed by trophic ulceration comes
with persistence of the epithelial defect over several weeks or
months. Collagenolytic activity with melting and potential
perforation become more likely the longer the ulcer is present.
This is particularly true in men and in postmenopausal women
and if the ulcer is located centrally away from peripheral
neovascularization that scars but also heals.194,195 Trophic
ulcers may usually but not always be distinguished clinically
from actively infected geographic ulcers. The former have gray
thickened borders owing to the piling up of epithelial cells that
are unable to move across the damaged ulcer base. Actively
infected geographic ulcers have discrete, flat edges that may
change in conguration as the infectious process spreads. It
should be noted, however, that a study by Fukada et al using
real-time PCR HSV genome was found in the tears of six of
six patients diagnosed with persistant epithelial defects, i.e.,
trophic ulceration.196 This is a good rationale to keep all of these
patients on long-term prophylactic oral antivirals such as ACV
400 mg po bid.
Therapy of neurotrophic keratitis and ulceration. Treatment is aimed at protecting the corneal surface and damaged
basement membrane because of the neurochemical and
mechanical nature of the problem. Multiplying virus does not
cause trophic disease so there is no need for antiviral therapy
except, as noted later, prophylactically if steroids are in use.
Similarly, cauterization will worsen the condition by further
damaging the basement membrane. Scraping the epithelium
will usually result in the regrowth of these cells up to the border
of the ulcer, but they are then still unable to adhere to the
damaged basement membrane and will not move across to close
the defect which may be made worse by the debridement.
Therapeutic approaches used include treatment of any meibomian gland dysfunction, copious lubrication with unpreserved
articial tears, gels and/or ointments, lateral tarsorrhaphy,
therapeutic soft contact lenses, tetracyclines, suppression of
inflammation, prophylactic oral antivirals, autologous serum
drops, tissue adhesive with soft contact lens, amniotic membrane transplant (AMT) or AM contact lens (ProKera) or
occasionally, conjunctival transplant, conjunctival flap, penetrating keratoplasty, or keratoprosthesis.
A. Treat meibomian gland dysfunction (MGD) to stabilize
MG secretions and tear lm:
1. Lid hygiene: warm soaks, baby shampoo lid margins qd.
2. Doxycycline 100 mg po qd not within 2 h of food or
pills containing calcium 6 months or longer. Avoid in
patients on coumadin.
3. Metrogel to lids and face qd 6+ months if rosacea
present.
4. Instruct patients with incomplete blinking to blink
correctly and completely.
B. Lubrication and antibiotics:
1. Copious lubrication with nonpreserved or oxidative
preserved articial tears or gel four to six times daily,
antibiotic ointment qhs, e.g., bacitracin or erythromycin.
Do not use more toxic antibiotic agents such as
aminoglycocides or fluoroquinelones unless otherwise
indicated.
2. Check for adequacy of Bells reflex. If eyes do not roll
up on lid closure, consider tear gel followed by taping
lids at night.
3. Punctal plugs.
4. Cholinergic enhancers to increase exocrine function
(tears + saliva) with cevimeline (Evoxac) 30 mg po tid
or pilocarpine (Salagen) 510 mg po tid.176,184
C. Therapeutic soft contact lenses: High-water-content soft
contact lenses are commonly used to promote healing of

the corneal surface. These bandage or therapeutic contact
lenses (B-SCL or T-SCL) include the Permalens
(Coopervision), Kontur (Kontur Kontacts), Acuvue (J&J),
Focus Night & Day (CIBA).5,197,173,198 They are usually
worn for several months as prophylaxis against or actual
treatment of neurotrophic ulceration. They should not be
removed at the time of each exam as this disrupts the
healing process. It is more judicious to put in an anesthetic
drop (to float the lens and reduce any remaining
sensation), and slide the lens temporally using a sterile
cotton tip applicator while the patient looks nasally. It is
all right to put in fluorescein stain as this will wash out
within a day. They should also not be removed for cleaning
and sterilizing unless deposits build up or the lens tightens
grip and irritates the eye. In the latter case, a new,
sometimes flatter lens may be applied. The lenses are
thought to work by splinting and protecting the cornea
from abrasive lid action and by keeping the ulcer well
lubricated by inhibiting tear lm evaporation. Patient
tolerance of these lenses may be increased by use of
cycloplegia with homatropine or other cycloplegics.
Cyclopentolate may not be advisable, as this drug tends to
attract PMNs.198,99 They should also be lubricated with
unpreserved articial tears several times daily and
prophylactic antibiotic drops such as sulfacetamide or
polymyxin-trimethoprim (Polytrim) bid.
D. Steroids: In the presence of stromal inflammatory reaction,
without ulceration and corneal thinning, mild steroid such
as 0.125% prednisolone, rimexolone (Vexol), or lotoprednol
(Lotemax 0.5% or Alrex 0.2%). If melting occurs, 1%
medroxyprogesterone (Provera), a steroid that will not
inhibit collagen synthesis, may be used two to ve times
daily to quiet the inflammatory response but be prepared
to take further steps as outlined below.47
Medroxyprogesterone is a fairly weak steroid, about the
strength of 0.2% lotoprednol. This may be made up as
sterile 1% eye drops by a compounding pharmacy. An
inflamed eye will heal more slowly than one that has been
quieted with steroid therapy.189,194,195 Inflammation causes
increased prostaglandin and cyclic AMP which in turn
decreases cell growth and mitosis. If healing is successful
the lens may be removed and topical lubrications with
bland ophthalmic ointment or articial tears continued for
a minimum of 6 months or forever to prevent lid action
from rubbing the newly healed epithelium from a possibly
still-fragile ulcer base.
E. Antivirals: If HSV stromal disease is present oral antiviral
such as acyclovir 400 mg po bid, famciclovir 125250 po
bid, or valaciclovir 500 mg po bid for 1 year or longer
should be initiated not because they will affect a current
recurrence but will serve as prophylaxis against future
stromal disease. This is not indicated in zoster, however,
as the dose required to prevent any recurrence is very high
and would likely not be safe over a long period of time.
Further, there is no evidence that zoster prophylaxis would
work.5,40,134,168,200203
F. Tetracyclines: Oral and/or topical tetracyclines will not
only stabilize the meibomian gland secretions and tear lm
breakup time, they inhibit matrix metalloproteinases via
restricted gene expression of neutrophil collagenase and
epithelial gelatinase. Further, they suppress alpha-1 trypsin
degradation and scavanging of reactive oxygen species. All
of this is by mechanisms independent of their
antimicrobial properties.204 The oral tetracyclines most
commonly used are doxycycline or minocycline 100 mg po
qd, not within 2 h of ingestion of something containing
calcium which inactivates the drugs. Through a

compounding pharmacist topical oxytetracycline may be
made up and used bid and qhs.
G. If the epithelium becomes progressively unhealthy looking,
a lateral tarsorrhaphy or botox-induced ptosis is advisable.
Animal study shows that corneal denervation results in
dry eyes and neurotrophic breadown which is responsive to
lateral tarsorrhaphy.183 In one study on 77 patients,
indications for a tarsorrhaphy widely varied and included
such entities as persistent epithelial defects associated with
neurotrophic ulcers, penetrating keratoplasty (PK),
postinfection, exposure keratopathy, surgery other than PK,
dry eye syndrome, and radiation keratopathy.205 The
epithelial defects in 70 of the 77 eyes completely resolved,
and the difference between the duration of the signs and
symptoms before tarsorrhaphy and time-to-healing after
tarsorrhaphy was statistically signicant (p = 0.01); 31%
of the tarsorrhaphies were temporary and 69% were
permanent.
H. Tissue adhesives (glue): If the patient is not fortunate
enough to heal the ulcer and stromal lysis with progressive
thinning becomes apparent (Fig. 49.10), consider
cyanoacrylate tissue adhesive (glue) such as Indermil
(Sherwood, Davis & Geck, St Louis, MO), Histoacryl
(B Braun, Melsungen, Germany available in Canada),
Nexacryl (Closure Medical, Raleigh, NC), or 206208
The technique of applying tissue adhesive is generally
uncomplicated. With the patient under topical anesthesia
and with a lid speculum in place, the epithelium is gently
debrided from the ulcer edge and base, the area to be glued
carefully dried with Weck-cel sponges, and the polyethylene
applicator tube or a 25-gauge needle containing liquid
tissue adhesive gently touched in concentric circles to the
edge of the ulcer. The physician works from the center or
until the area is entirely covered with adhesive. Sterile
saline solution may then be dripped on to the eye at the
end of the procedure to hasten polymerization. A soft
therapeutic contact lens is then applied for continuous
wear to protect the lids from irritation by the irregular
anterior surface of the glue. The Plano T lens works well
in this situation as it is slightly thicker than the Permalens
and stands up better against the rough glue. Antibiotic
FIGURE 49.10. HSV trophic ulcerative keratitis with a deep melt lled
with sterile tissue adhesive. A contact lens is to be placed over
adhesive to prevent irritation of lids.
drops should be instilled twice daily, and any steroid drops

warranted for inflammation may be used with safety
greater than in the presence of an open ulcer. With time
the ulcer often heals, and new epithelium will slide in
underneath the adhesive and ultimately dislodge it, leaving
behind a scarred but intact eye amenable to surgery for
visual restoration.
It should be remembered, however, that it does not
always work and the ulcer can enlarge and dislodge the
glue. In these cases, corneal lens with cyanoacrylate glue is
a temporizing procedure only, buying time to allow healing
secondary to medical treatment of the underlying
condition, or allowing surgery to be elective and under
more optimal conditions once inflammation has been
reduced and the integrity of the globe restored.206
I. Autologous serum drops: These are made from the
patients own serum and contain collagenase inhibitors
such as alpha-1, alpha-2 macroglobulins and, in their
undiluted form, were showned to heal trophic herpetic
ulcers in 1973.209 Since then, a number of studies have
reported the usefulness of this approach. Two studies from
Boninis group using murine nerve growth factor (NGF)
(110 micrograms of highly puried murine NGF in 50 mL
of physiological saline) to treat human anesthetic
neurotrophic ulcers found that corneal healing began
214 days after starting treatment of 10 times daily for
2 days and then six times daily until the ulcers healed. All
patients had complete healing of their corneal ulcers after
10 days to 6 weeks of treatment. Corneal sensitivity
improved in 13/14 eyes, and returned to normal in two
of the 13 eyes. Corneal integrity and sensitivity were
maintained during the follow-up period (range,
312 months).210,211
Tsubota et al used 20% solution of autologous serum in
saline 610 times daily in 16 neurotrophic eyes with
persistent epithelial defects (PED).212,213 The concentration
of the natural serum healing factors, vitamin A, epidermal
growth factor (EGF), and transforming growth factor-beta
(TGF-beta) was measured at 1 week and 1 month, stored
in the refrigerator and 1 and 3 months in the freezer and
all were found to be stable at all time periods. Autologous
serum healed 44% of persistent defects within 2 weeks and
63% within 1 month. The remainder did not resolve in
that period.
J. Surgical techniques to enhance healing of the recalcitrant
neurotrophic ulcer:
1. New ocular surface re-construction including
techniques such as amniotic membrane
transplantation, limbal stem cell transplant procedures,
transplantation of cultivated oral mucosal or limbal
stem cell sheets.214217 Amniotic membrane
transplantation has become increasingly popular as
several studies of its success have been
reported.214,217219 Both multilayer and single layer
transplants have been successful in healing herpetic
and other causes of trophic ulceration. In a study by
Solomon et al 34 eyes with descemetocele or
perforation received three to four layers in the ulcer bed
and then a total corneal covering with a large piece of
AMT anchored with 10-0 nylon. Successful healing
occurred in 28 of 34 eyes (82.3%). Of these successful
cases, 23 eyes needed only one AMT procedure, ve
eyes needed two procedures for success. In ve eyes, a
subsequent surgical procedure such as penetrating
keratoplasty or lid surgery was needed. Failure occurred
in six eyes with rheumatoid arthritis, neurotrophic
keratopathy, or graft melting.219 Over several weeks to
CHAPTER 49
651

months an AMT will gradually dissolve leaving behind,
in the vast majority of cases, a smooth, healed surface
with some restoration of stromal thickness.
Perforated corneas may be treated with greater
benet using a combination of tissue glue and AMT.
Hicks et al reported 14 such eyes. Grafts with brin
sealant showed a success rate of 92.9% (13/14 eyes)
compared to 73.7% (14/19 eyes) for amniotic grafts
alone.220 If there was severe limbal damage, a success
rate of only 20% (1/5) was observed. Perforations up to
3 mm were safely managed by brin glue and AMT
leading to rapid re-construction of the corneal surface
(34 weeks healing time), gave good nal functional
result and allowed keratoplasty to be done in more
favorable conditions.
Amniotic membrane is commercially available as is a
new amniotic membrane contact lens (Amniogaft,
Prokera, Biotissue Inc.). The latter may be inserted in a
minor surgical room with topical anesthesia and its
conformer ring removed a few weeks later leaving the
AMT in place.
2. Conjunctival flaps (CF) are still preferred by some
surgeons but are becoming less frequent as new
technology evolves that does not cause the notable loss
of vision found with CFs. In one study, 48 patients had
total CF and 13 had partial CF for severe bullous
keratopathy, chronic graft failure, (not candidates for
keratoplasty) (19), herpes zoster ophthalmicus (7),
chronic ulcerative keratitis (14), neurotrophic keratitis
(2), and herpes simplex keratitis (9). Successful
outcomes were achieved in 54 out of 61 eyes. The
primary complication was flap retraction in the
remaining TCF and PCF eyes. The authors concluded
that conjunctival flaps are underused and should be
considered for such conditons as bullous keratopathy,
neurotrophic keratitis, recalcitrant keratitis, and
persistent nonhealing epithelial defects.221
3. Patch grafts are good alternatives to penetrating
keratoplasty in corneas that have perforations too large
to be glued. The eye then has time to be treated
medically until a more auspicious time for penetrating
keratoplasty may be done.
4. Penetrating keratoplasty and other surgical procedures
are discussed below under Surgical Intervention in
Ocular Herpes.
SECTION 6
HSV immune keratitis
652
Herpetic immune keratitis may present as (1) limbal vasculitis,

(2) Wessley immune rings, (3) necrotic interstitial keratitis (IK),
(4) disciform edema, and (5) endotheliitis with or without
trabeculitis; all are primarily immune in nature.5,118,152,207,222226
Endotheliitis may be further subdivided into three clinical
forms: disciform, diffuse, and linear.153
HSV keratitis is thought to be a hypersensitivity reaction to
the xed herpes antigen within the stromal keratocytes or broblasts or in the endothelium. Other evidence, however, suggests
that it may also result from reaction to active viral infection in
the stroma or endothelium secondary to re-activation of
trigeminal ganglion latent HSV or corneal latent HSV.12,13,227231
Intact virus particles have been demonstrated in the stroma of
patients by electron microscopy and by tissue cocultivation.
Additionally, Pavan-Langston et al have demonstrated retention
of herpetic DNA in all layers of the cornea for at least 90 days
postinfection, well into the latency period.98,228230,232
Factors which influence the severity of herpes simplex
keratitis are in all likelihood also involved in herpes zoster
keratitis, but, for lack of a good animal model and difculties
encountered in working with human disease, have not been

so well studied.
In his review of multiple studies on the role of viral and host
genes in herpes simplex 1 keratitis Brandt reported that viral
gene products interact with each other, and with host proteins
and these interactions are critical in determining the nature
of the infection.233 The grouping of genes encoded by each
particular strain is critical, and how these genes work together
and with host proteins determines the acute and long-term
course of the disease. Recent data show that different viral
genes cooperate to influence disease severity and conrm that
the constellation of genes within a particular strain determines
the disease phenotype. New methods are now used to test the
role of viral genes in virulence.
Different viral strains may also induce ocular disease of
differing severity and frequency. The more destructive forms
of corneal stromal disease have been associated with HSV
strains that produce larger amounts of glycoprotein in the virus
envelope. These viroglycoproteins are particularly effective
inducers and targets of humoral and cell-mediated immune
responses.234,235 Additionally, data indicate that the clinical
response to topical steroids may also be determined by a viral
genome.236 Therefore, both viral and host immune factors
appear to play key roles in determining the clinical expression
of HSV ocular disease.
Viral replication in the corneal epithelium triggers proinflammatory cytokine release both by infected cells and their
uninfected neighbors. IL-1alpha and TNF-alpha are upregulated
in corneas in mice experiencing recurrent HSK. Anti-IL-1 and
anti-TNF-alpha antibody administration both resulted in
signicantly decreased virus-induced corneal opacity between
7 and 21 days after UV-B exposure indicating that these
cytokines, among others, play important roles in the pathogenesis of recurrent disease and that neutralization of specic
proinflammatory cytokines may have potential therapeutic
value.237,238 It has also been noted that absence of MIP-2 and
MIP1 alpha and TNF prevents the development of stromal
keratitis.239,240
Both corneal cells and neutrophils contribute to the cytokines
causing inflammatory disease. Transient corneal haze may
appear during the acute infection. This can be secondary not
to immune reaction but to the effect of toxic cytokines on
endothelial function.242 Defending the cornea is the early
production of interferons alpha, beta, and gamma which inhibit
viral replication.243
In immune diseases Langerhans cells move into the area of
infection, acquire viral antigen which they transport to the local
lymph nodes and present them to T-cells thus attracting them
to the site of infection in what appears to be a type 4 hypersensitivity response mediated largely by CD4+ cells although
other mechanisms have been proposed.244248 With ongoing
immune reaction or intermittent immune keratitis, the corneal
scarring worsens, diminishing vision, and in some patients,
causing blindness. Other study results indicate that viralinduced tissue damage can be caused by bystander cells, but
these fail to control infection. HSV-reactive CD8+ T cells are
capable of ocular virus clearance, possibly through a combination of corneal and peripheral nervous system antiviral effects.
Immune CD8+ T cells initiate viral clearance from the eye,
but this appears to result by the T cells acting at sites both local
and distal to the cornea.249,250 It is proposed that CD8+ T cell
control is expressed in the trigeminal ganglion, serving to
inhibit a source of virus to the cornea. CD8+ cells, however, are
apparently not involved in lesion development. CD4+ T cells
both virus-specic and nonspecic are the pathogenic T cells
capable of causing scarring disease. Targeting specic cell types
in HSK may be a useful future therapeutic approach.249,251254
FIGURE 49.11. Acute HSV limbal vasculitis from 2:30 to 4:30 and
from 8:30 to 11 oclock with central edematous stromal disciform
edema in eye with combined-mechanism immune disease
(lymphocyte-mediated disciform keratitis and antigen-antibodycomplement (AAC)-mediated vasculitis).
FIGURE 49.12. HSV anterior stromal immune ring of Wessley under

healthy intact epithelium.
Limbal vasculitis (local Arthus reaction) tends to be sectoral

and located at the limbus with minor invasion of the corneal
stroma. These areas of focal hyperemia and edema do not
invade the cornea and with or without treatment, will resolve
over time without scarring. The immune basis for the limbitis,
however, makes it amenable to more rapid resolution if topical
steroids are used (Fig. 49.11).5,226
Wessley immune rings precipitate (antigen-antibody) in the
anterior to mid-stroma and may have a hazy edema within the
ring and/or attract neovascularization. These translucent central
inltrates with a circumferential opaque ring histologically form
along a line of altered keratocytes and ground substance with
inltration of inflammatory cells. Herpes-virus particles in the
corneal stroma were seen by electron microscopy, but these
virus particles were abnormal, noninfective forms such as empty
capsids and incomplete virions.252254 Herpes-virus antigens
were in corneal keratocytes and stroma, primarily in association
with the herpes virions and surrounding vacuoles in the keratocyte nuclei and in the stroma in the area of degenerating
keratocytes (Fig. 49.12).
IK is characterized by necrotic, blotchy, cheesy-white stromal
inltrates. Holbach et al have reported detecting HSV antigen
signicantly more often in human corneas with ulcerative
necrotizing stromal keratitis (IK) than nonulcerative, nonnecrotizing keratitis (disciform) which may account for the greater
scarring tendencies of the former.255 He also reported the
histopathology on a perforated, necrotizing HSV IK in a corneal
graft showed viral inclusion bodies and particles in the endothelium and in the stromal keratocytes (Figs 49.13 and
49.14).222
In a chart review study to determine the etiology of IK of
97 patients 55 had active ISK, dened by stromal inflammation
without ulceration within 1 year of presentation. Forty-two
patients had inactive ISK, dened by evidence of past stromal
inflammation, including stromal scarring, stromal thinning,
ghost vessels, and re-duplication of Descemets membrane
without active inflammation for 1 year before presentation.254
HSV accounted for 71.4% of unilateral active ISK. Idiopathic
accounted for 14%, and varicella-zoster virus accounted for
8.6% in this group. HSV was the etiologic factor of 50.0%
of inactive unilateral cases, whereas 33% were idiopathic. In
FIGURE 49.13. Severe HSV necrotizing interstitial keratitis involving

entire cornea with deep neovascularization moving in 360o.
A conjunctival flap was placed to quiet the process.
bilateral inactive disease syphilis accounted for 48% and the rest
were idiopathic. We know, however, from Souza et als study of
bilateral HSV keratitis that there was stromal keratitis in nine
eyes (64.3%), necrotizing stromal keratitis in ve eyes (35.7%),
and progressive endotheliitis in two eyes (14.2%).155 This
should be factored in when considering the incidence of
bilateral HSV IK.
It is occasionally difcult to distinguish IK from a secondary
bacterial or fungal infection, although the former is far more
indolent than the latter two.229,255 In HSV-IK after several weeks
of smoldering, dense leashes of deep neovascularization move
in as if in pursuit of the antigenic inltrates. Both immune
rings and IK tend to scar signicantly despite steroid therapy.
Limbal vasculitis may induce a dellen effect because of the
local edema but otherwise resolves spontaneously or with the
assistance of steroid therapy and with little to no scarring. As
CHAPTER 49
653
FIGURE 49.15. Focal edematous acute HSV central disciform

keratitis without necrosis or neovascularization.
SECTION 6
654
c
FIGURE 49.14. (a) Acute HSV focal interstitial keratitis resembling
bacterial inltrate (arrow). (b) Same eye 5 years later with acute
recurrent limbal dendritic ulcer with typical terminal bulbs (arrow).
(c) Same eye 6 weeks later showing well-healed epithelium in the area
of the previous dendritic ulcer and faceting with minimal scar in the
area of the old interstitial keratitis.
with all forms of stromal keratitis, disciform disease may occur

without obvious epithelial infection and may be due to immune
reaction to subclinical ganglionic re-activation with migration
of virus to the cornea, or it may be due to smoldering activity of
genome retained in the cornea with antigenic alteration of the
surface membrane of corneal cellular elements. Viral antigen
and cellular antigen elicit a host inflammatory response characterized by granulomatous foci and/or lymphocytes and plasma
cells, and ultimately macrophages and PMNs.157,229,231,255,256
Clinically milder disciform keratitis is a diffuse or focal diskshaped area of stromal edema without necrosis or neovascularization (Fig. 49.15). There may be focal keratitic precipitates
(KPs) made up of plasma cells and lymphocytes clinging to the
endothelium of the edematous area. The KPs may be difcult to
see except at the edge of the lesion. There may be no anterior
chamber reaction. In more moderately active disease, edema,
and folds in Descemets membrane are seen indicating
endothelial decompensation due to toxic inflammatory reaction
with fluid entering the cornea in abnormally great amounts.
Neovascularization and iritis may be present. In the most
severe forms of combined disciform reaction, there is an associated necrotizing IK with diffuse edema, ulcerating bullous
keratopathy, mutton fat KPs, necrotic stromal thinning,
frequently melting, severe iritis, and ultimately severe scarring
(Figs 49.16 and 49.17).
An unusual form of HSV disciform disease, sectorial
keratitis, has been reported in six patients (seven eyes) between
the ages of 21 and 50 years.257 The corneal haze and edema was
most common in the superior corneal quadrants, located in the
deep corneal layers. An anterior nongranulomatous uveitis was
present in all cases and one case was bilateral. Glaucoma
developed in six of the seven eyes and ve patients developed
corneal scarring, localized in the anterior and mid-stroma. Five
patients responded well to topical steroid therapy and antiviral
prophylaxis but the sixth required immunomodulating treatment with systemic methotrexate, cyclosporine, and prednisone.
Wilhelmus et al have studied the changes in corneal
thickness by ultrasonic pachymetry during the course of herpes
simplex virus disciform corneal edema in 35 patients.
Measurements during follow-up and at clinical resolution of
inflammation were compared with the initial corneal thickness
and revealed that corneal thickness of disciform stromal
keratitis decreased a signicant 15% during follow-up, and that
a signicant 65% of this decrease occurred during the rst
2 weeks of treatment.258
HSV endotheliitis, trabeculitis and secondary glaucoma
Herpetic endotheliitis is considered by some clinical investigators to be a form of disciform disease, a clinical entity all of
its own, or, conversely, that all disciform edema is secondary to
endotheliitis.5,48,151,153,259263 These cases may present as acute
corneal edema with keratic precipitates and may have minimal
FIGURE 49.17. Large HSV infectious geographic ulcer overlying

necrotic stromal interstitial keratitis formed as a partial immune ring in
an eye with combined-mechanism herpetic disease (infectious ulceration
and AAC-mediated interstitial keratitis and Wessley immune ring).
or no flare and cells in the anterior chamber or a severe

keratouveitis which may be largely obscured by the corneal
edema.262 All of these cases responded to topical steroid therapy.
Sundmacher et al cultured the aqueous fluid from 33 HSV and
four zoster patients with focal endotheliitis, prolonged
disciform keratitis, iritis, and glaucoma.261 Nine taps from eight
HSV patients yielded herpes simplex virus. In the case of one
patient, two taps were positive at 14 days interval. External
ocular cultures were all negative. The culture-positive cases
had three things in common: (1) Secondary glaucoma was
uniformly present. This in itself was felt to be an indication for
culturable herpes simplex virus in the aqueous. (2) Three
clinical pictures could be differentiated biomicroscopically: focal
iritis, peripheral endotheliitis, and prolonged disciform keratitis.
(3) In some cases, tissue damage from immune reactions seems
to be more important for the functional outcome than tissue
damage by viral cytolysis itself. All eyes responded to treatment
with topical steroids and prophylactic trifluridine.
Other cases may be preceded by dendritic ulceration or
marked elevation in intraocular pressure (IOP) prior to onset of
FIGURE 49.18. Acute HSV progressive endotheliitis similar to graft

rejection with a lymphocyte line on the endothelium and focal fullthickness stromal edema medial to the lymphocyte line.
the corneal edema. One case of a 62-year-old man presenting

with acute glaucoma, endotheliitis, KPs, and corneal stromal
edema had tissue excised at trabeculectomy. The histopathology
proved highly immunoreactive for HSV, thus revealing the virus
as directly involved in the pressure-elevating trabeculitis.264
A second form of endotheliitis is linear endotheliitis. This
entity appears clinically as a line of KPs on the peripheral
corneal endothelium that progresses centrally and is followed
by peripheral stromal and epithelial edema.265 There may be
associated ocular pain, redness, and photophobia. This form too
usually responds well to topical steroids with prophylactic oral
antivirals. Anterior chamber reaction may be minimal to nonexistent, although the author has had a patient who, 2 weeks
after a small dendrite resolved, presented with pressure of
50 mmHg (trabeuclitis) and early linear endotheliitis. Despite
aggressive topical and sub-conjunctival steroids progression of
this rejection line across the cornea with endothelial damage
and overlying stromal edema could not be stopped although the
pressure did come under control with that treatment and antiglaucoma agents. Ultimately, the entire cornea was permanently
edematous and he underwent keratoplasty (Pavan-Langston, D
unpublished) (Fig. 49.18).
HSV endotheliitis may involve the peripheral cornea and
trabecular meshwork (TM) together or inflammation may be
conned just to the TM (trabeculitis) within the angle of the
eye. This focal inflammation may be present in a totally quietappearing eye but with acute secondary glaucoma due to the
swelling and obstruction of the trabecular meshwork by
inflammatory debris. Several investigators noted that elevated
IOP was strongly predictive of the presence of infectious herpes
virus in the aqueous humor of these patients.259,261263,266 Two
separate reports of HSV endotheliitis, one PCR-proven, failed to
respond to steroids but did respond to combined topical and
systemic antivirals (oral ACV).263,266 Concomitant antiglaucoma medications should also be used until topical steroids
have been able to suppress the inflammatory reaction clogging
the trabecular meshwork. Laser trabeculoplasty is not indicated
in this condition and may cause increased scarring. The
glaucomatous episodes may be transient or may ultimately lead
to permanent damage with severe secondary glaucoma. Despite
the association with viral particles in the aqueous of at least
some patients, topical antiviral therapy has little to no effect but
oral antivirals such as ACV 400 mg qid are advisable because of
the high chance of live virus in the eye.
CHAPTER 49
FIGURE 49.16. HSV dense disciform scar with supercial and deep
neovascularization.
655
SECTION 6
Therapy with topical steroids such as 1% prednisolone four

to six times/day will usually bring the IOP down within a few
days. This helps diagnostically by indicating that one is not
dealing with a steroid glaucoma in a patient already on steroids.
If the pressure goes up in the face of increased steroid therapy,
the physician should consider that this is a steroid glaucoma
and reverse the course of treatment. Along with steroids and oral
antivirals, it is advisable to give antiglaucoma drops if the pressure is too high to be let go for 24 days. This may temporarily
obscure the diagnosis by bringing the pressure down, however.
656
Histopathology of chronic HSV keratitis Characteristic corneal

changes in corneal buttons taken at keratoplasty include
changes at all levels. Most common is destruction of Bowmans
membrane in about 90% of patients. In studies by Dawson and
colleagues, Hogan, and Easty and co-workers from 3040%
of patients demonstrated epithelial loss, thickened basement
membrane, marked inltration of inflammatory cells in the
anterior stroma, subepithelial brovascular pannus, stromal
neovascularization, and breaks in or re-duplication of
Descemets membrane.12,228,229,255,267,268 About 20% of eyes had
retro-corneal membranes. In some patients, virus particles were
noted only in the anterior stroma where Bowmans membrane
had been destroyed; in others HSV was found in midstroma or
just anterior to Descemets membrane. In the study by Dawson
and colleagues, steroids had been used as part of the long-term
therapy preoperatively and may have played a role in the deeper
penetration of virus in some cases.
Holbach et al have reported that granulomatous reaction
from Bowmans through to Descemets is the characteristic
tissue response in HSV stromal keratitis in both necrotizing
and nonnecrotizing disease.255 T-cell lymphocyte-mediated
immunity was felt to play a major role in the pathogenesis of
this immune reaction. In addition, these immune foci were
associated with HSV antigens located in all corneal layers but
primarily the deep stroma and adjacent endothelial cells. This
rst report of endothelial antigen in humans conrms the
earlier report of endothelial HSV in experimental keratitis.269
Virus particles are seen in less than 20% of herpetic buttons
studied by electron microscopy but under proper culture
conditions may be isolated in up to 66% of corneal buttons
placed in organ culture for up to 11 days. Brik et al reported
cases of calcic ring deposits in the stroma while patients were
on topical and/or systemic antiviral therapy.228 At keratoplasty,
all buttons, including a fourth not operated on at the time
of the report (DP-L, unpublished), had numerous mature and
immature virus particles despite the antiviral therapy thus
indicating that current therapeutic antiviral agents have little
effect even on viral particle-related stromal keratitis.
Therapy for herpetic interstitial keratitis, immune rings,
limbal vasculitis, and disciform keratitis/endotheliitis and
traveculitis is outlined below .
Therapeutic guidelines for management of all forms of HSV
stromal keratitis are similar. If a patient has never been on
steroids and the disease is mild, every effort should be made to
avoid introduction of these agents into the therapeutic regimen.
Once a patient has been treated with steroids, it may be difcult
to withdraw these drugs and any subsequent immune reaction
will, in all likelihood, require the use of steroids to be brought
under control. As the stromal and endothelial diseases are
primarily immune in etiology, there is no established role for
antivirals except for prophylaxis against potential spontaneous
reaction of scarring epithelial infection and recurrence of
stromal keratitis.
1. In mild, nonprogressive, off-visual-axis cases, no treatment
but articial tears for lubrication of potentially unhealthy
epithelium.
2. Steroids:
a. In moderate to severe cases, especially with
neovascularization, start with the lowest dose of
steroids needed to bring process under control, e.g., 1%
prednisolone or one of the newer steroids (lotoprednol,
rimexolone) qid6id, begin to taper slowly as disease
comes under control. Progress downward with stronger
steroid to weaker steroid. The lower the dose of steroid,
the longer it is used. Below qd go to qod, tiw etc., or to
a weaker steroid over several months as needed. If the
inflammatory process begins to re-activate with steroid
dose reduction these drops should be increased to the
previous level or higher for a longer period of time
before further attempts at tapering are initiated. Some
patients will never be able to go below qd dosing to
keep their disease under control and some may need to
be at qid levels for many weeks before tapering may be
done without rebound inflammation. Monitor for side
effects or treat steroid glaucoma if it occurs.
b. 0.5 percent cyclosporin A drops qid are an effective
alternative to the usual steroids particularly in patients
with difcult steroid glaucoma. Oral antivirals and
topical antibiotic should be used prophylactically.
3. Oral antiviral prophylaxis is an integral part of treatment
of stromal disease. Dosing: ACV 400 mg po bid, or FCV
125250 mg bid, or VCV (immunocompetent patients
only) 500 mg po bid for 1 year or more.
4. a. Secondary glaucoma may be due to trabeculitis in
which case therapy is topical steroid such as 1%
prednisolone qid and, if warranted, a nonprostamide
analog* antiglaucoma drop such as alphagan or a
b-blocker. *Not conclusively proven that these agents
precipitate herpetic disease.
b. If the pressure does not respond to steroid therapy
within a week, or if the patient was already on a steroid
and the pressure goes up because the rise was
(unknowingly) due to steroid glaucoma in the rst
place, reduce or stop the steroids and continue
glaucoma treatment. It is often difcult to know
which way to go with the steroid if the patient is
already on it even in low dose. It is advisable to tell the
patient at the beginning that treatment may have to be
reversed if the response is not satisfactory.
c. Oral antivirals such as acyclovir 400 mg po tid or qid
are advisable because of the high association of this
type of glaucoma with live virus in the eye.
5. Cycloplegics as needed for iridocyclitis.
6. If an associated iridocyclitis is moderate to severe, use
steroids as described in 2 above, for weeks or months, with
oral antivirals for 1 year, or TFT drops 5id (not more than
3 weeks) as prophylaxis against recurrent stromal disease
or if steroids are used more than twice daily. Topical
antibiotic ointment as lubricant and as antibacterial
prophylaxis for topical steroids.
7. Articial tears or gel (no preservative or oxidative
preservative) four to six times/day to lubricate roughened
or hypesthetic cornea.
8. If the epithelium is ulcerated and melting, reduce or stop
topical steroids. If iritis or trabeculitis must be treated
urgently, use systemic prednisone (2030 mg orally twice
daily for 710 days), then taper over 7 days to control
intraocular inflammation until the epithelium is healed,
glued, or otherwise under control and topical steroids may
be started.
9. If, with treatment, the eye remains uninflamed with little
or no steroid for several months but vision is poor due to
HSK scarring, keratoplasty may reasonably be performed. If

the cornea is anesthetic or hypesthetic, and the eye is
chronically inflammed consider the Boston
keratoprosthesis in preference over keratoplasty.
Management of combined epithelial and stromal

disease
Not infrequently, patients may present with combined immune
keratitis, and a viral-infected epithelial ulcer or a sterile trophic
healing defect. The physician may be placed in a difcult
therapeutic decision. If the surface disease is thought to be
infectious, full antiviral therapy should precede or at least start
simultaneously with any use of topical steroids. This contains
the infectious process before any aggravating effects of steroids
may occur. Should the ulcer worsen because of the steroids,
their frequency or concentration should be reduced until the
ulcers are under control and healing. If steroids are chosen to
control the stromal inflammation, the surface process may
worsen, if it is infectious.
If the surface process is an indolent trophic ulcer, the usual
ophthalmic steroid drops may enhance the chances of stromal
melting. As discussed previously, milder steroids such as
loteprednol or rimexilone or use of 1% medroxyprogesterone
acetate drops may be advisable in a situation in which the
stromal reaction must be controlled or is interfering with
healing of the trophic ulcer. In addition, the usual soft contact
lens, prophylactic antibiotics and prophylactic oral antivirals,
and lubricants should be used to treat the trophic ulcer. If
melting does progress, cyanoacrylate tissue adhesive and other
steps, as discussed in the section on Trophic Postinfectious
Ulcers, should be taken.
HSV iridocyclitis
Recurrent nongranulomatous anterior uveitis may be caused by
HSV. This uveitis may occur prior to any known herpetic ocular
disease or may be associated with an active keratitis. The
etiology of this entity is not well established. Intact virus
particles have been isolated from the aqueous humor and from
a retro-corneal membrane, but there is also clearly an immune
inflammatory component.261,266,270273 Despite recent advances
in measuring anti-HSV antibodies and viral DNA in ocular
fluids, diagnosis remains largely clinical.274 Iritis in an eye with
a previously known herpetic keratitis should be considered
herpetic until proved otherwise by examination or laboratory
testing.
Clinically, the iritis may be focal or diffuse. In focal
involvement there may be scattered areas of swollen hyperemic
iris, an irregular pupil and localized posterior synechiae. Cells,
flare, and ne or heavy keratic precipitates may be present and
FIGURE 49.19. Acute HSV keratouveitis with mutton-fat keratic

precipitates on the corneal endothelium and extensive anterior
chamber cell and flare reaction.
associated with a concomitant endotheliitis with focal corneal

edema (Fig. 49.19). On retro-illumination pigment epithelial
defects may be detected. More commonly herpetic iritis is
diffuse with histopathology indicating widespread inltration
of iris stroma with lymphocytic cells. In these cases, there
may be severe cell and flare reaction in the anterior chamber
with brin, hypopyon, iris edema, synechia formation, and
secondary glaucoma due to associated trabeculitis. There has
also been a case report of a hypersensitivity iris granuloma
mimicking malignant melanoma of the iris in a patient with
recurrent HSV keratitis. Benign granuloma should be considered in the differential diagnosis of iris melanoma in patients
with ocular inflammatory disease.271
Therapy of herpetic iridocyclitis Treatment at present is
nonspecic. Suppression of the inflammatory reaction with
topical corticosteroids is still the currently advised therapy.
Again, if the patient has never been on steroids every effort
should be made to avoid their use by managing patients having
milder disease with cycloplegics alone. Cyclopentolate is
probably not advised in these situations because of its tendency
to attract PMNs. Homatropine, scopolamine, or atropine does
not attract white blood cells, is effective in blocking ciliary
spasm, and is sufciently strong to prevent most synechia
formation. If, however, the inflammatory reaction is more
severe and there is progressive aqueous cells and flare or
synechia formation, or the patient has been controlled with
only steroids previously, topical steroids should be reinstituted
using a regimen similar to that described in treatment of
immune keratitis. Starting dosages and frequency of steroids are
compatible with the relative severity of disease and prophylactic
antivirals and antibiotics should be used accordingly with
continued cycloplegic mydriatic agents.
Experience in treating HSV iridocylitis with oral ACV is
limited and the subject of the Herpetic Eye Disease Study
(HEDS) in which a masked, placebo-controlled clinical trial of
oral ACV 400 mg 5id for 10 weeks was used in the treatment
of this condition.272 Unfortunately, the number of patients
recruited was too small to reach statistical signicance but there
was a trend toward some efcacy when ACV was used in
patients also on topical steroids and prophylactic TFT.
A retrospective uncontrolled trial by Schwab indicated that
oral ACV 200 mg po ve times daily for 23 weeks is effective
CHAPTER 49
Although all forms of immune ocular HSV disease are thought

to be generally responsive to steroids, there is often residual
lipid-like deposition in interstitial keratitis and brotic scarring
in immune rings. Vision may, therefore, be compromised
despite treatment if the central visual axis is involved.
Interstitial keratitis, immune rings, and limbal vasculitis all
tend to resolve spontaneously in several weeks to months but
with scarring especially with IK. Neovascularization should
regress to ne or ghost vessels. An overlying irregular
astigmatism will often smooth out with time and if there is no
recurrence of disease. Stromal disciform disease, being largely a
lymphocyte-mediated inflammatory reaction, is highly steroidsensitive and may clear with little scarring or leave a notable
gray haze primarily in the anterior stroma. If a major attack on
the endothelium is allowed to continue unmitigated by therapy,
a potentially reversible bullous keratopathy may be converted
to a permanent state owing to irreversible damage to the
endothelium.
657

in resolving infectious herpetic epithelial keratitis.163 The
reported improvement in stromal keratitis and iridocyclitis in
Schwabs report may well be due to secondary improvement of
deeper inflammatory reaction after healing of epithelial disease.
Patients were receiving concomitant topical steroids.
1. Cycloplegics only for mild iritis and coupled with below
treatment for more severe disease.
2. Topical steroid therapy may range from agents such as
prednisolone 1% or dexamethasone 0.1% every 3 h for
severe iritis to once per day for mild disease. Once the
inflammation has come under control, switch to a
medium strength steroid such as rimexolone or
lotoprednol bidtid and begin gradual, slow tapering off.
Total cessation should be attempted when thrice a day
treatment is reached.
3. If the cornea has ulcerated or is melting, topical steroids
should be reduced or stopped, and prednisolone (2030 mg
orally) may be given twice daily for 714 days, then tapered
off over 10 days. Appropriate treatment for the corneal
condition is given during this period.
4. Oral antiviral agent prophylaxis bid (ACV, FCV, VCV),
daily plus antibiotic qd are advisable if topical steroids are
used more often than bidtid.
5. Nonprostaglandin agonists (not conclusively proved that
these agents precipitate herpetic disease) such as alphagan
or b-blockers should be given, if secondary glaucoma is
present.
SECTION 6
Ocular HSV in immunocompromised patients

Atopy and allergy affect more than 15% of the world population
with some studies having shown that upto 30% of the US
population has some form of allergy.275 In studies of atopic
disease around the Mediterranean and in Europe, the incidence
is increasing signicantly. In the Aegean area the prevalence
of allergic rhinitis, allergic conjunctivitis, and allergic skin
disorders (urticaria and atopic dermatitis) were found to be 13,
13, and 24%, respectively. The increase of prevalence for
asthma was 2.6%, for allergic rhinitis was 9%, and for allergic
conjunctivitis was 0.6%.276 In France, the prevalence of atopic
dermatitis is 25/100 at any age (69/100 before the age of
15).277 With a rst-degree parent exhibiting AD, asthma or
rhinitis, the risk of developing asthma is 40 p 100. In Taiwan,
the prevalence of symptoms of asthma, allergic rhinitis, and
atopic eczema in young teenagers increased by 37%, 51%, and
193% in 2004.278
To this number of immunocompromised patients we must
also add the growing numbers of patients with HIV infection or
AIDS. Global HIV infections have doubled since 1995.279281
While the actual prevalence in the United States is difcult to
determine in 2002, the CDC estimated that ~900 000 people
were infected with HIV and 25% of these did not know they
were infected.282 And to that, we may add the many thousands
of organ transplant patients and blood dyscrasia patients who
are iatrogenically immunosuppressed.
All of these patients are particularly susceptible to HSV

infections, because of a steady decline in the absolute number
of CD4+ T-lymphocytes and other immune malfunction.
These individuals may develop severe local herpetic vesicular
eruptions accompanied by fever and secondary bacterial skin
infections or widespread herpetic infection known as Kaposis
varicellaform eruption, or in the case of atopic dermatitis,
eczema herpeticum (Fig. 49.20).169,283286 Bilateral herpetic
keratitis is particularly common in patients with atopic or other
immunosuppressive disease.60,155,169,287 There is frequently
stromal scarring and delayed epithelial healing in the atopic
patients despite adequate antiviral therapy and an unusually
severe keratitis and poor therapeutic response to topical
antivirals. As a result, systemic ACV is recommended as part
of the therapeutic regimen in these patients. Some studies
reported a predilection for marginal as opposed to central
epithelial keratitis, epithelial ulcers more resistant to therapy
with median healing time after initiation of topical antivirals
alone being 3 weeks compared with less than 2 weeks in
immunocompetent patients, and little stromal scarring.288292
HSV types 1 and 2 may cause dendritic keratitis simultaneously.293 While the incidence of HSV was no greater than in
the general population, recurrences of herpetic disease were
more frequent in the HIV population with some patients having
two to three recurrences over an average period of 17
months.225,288,292 These recurrences also tended to be lengthier
than the initial episodes. Immunocompetent patients generally
had a recurrence-free interval of at least 18 months.69,70,154 In
the report of six patients by Young and co-workers, the former
beneted from the use of oral ACV in more rapid resolution of
their recurrent episodes of infectious disease.292
More recently, Hodge and Margolis have reported a large,
controlled study on ocular HSV in HIV patients.294,295 There
were 1800 HIV+ patient visits and 48 200 HIV-negative
control visits. While they also found no increased incidence of
HSV keratitis, unlike the earlier reports there was no signicant
difference between HIV and control patients in lesion type
(epithelial and stromal), lesion location (peripheral versus
central), response time to topical TFT or oral ACV therapy
(HIV+ = 17 d, HIV = 18 d), ultimate visual outcome, and
time to rst recurrence rate (347 d HIV+ versus 321 d HIV).
Only the recurrence rate was signicantly higher in HIV+
patients, being on average 1/587 d for HIV+ and 1/1455 d for
HIV patients.
One optimistic note is that since the late 1990s, widespread
use of highly active antiretroviral therapy (HAART) has altered
the spectrum, and reduced the incidence of ocular involvement
in developed countries.296 We may, therefore, look forward to
fewer cases of AIDS-related ocular HSV.
Therapy of HSV in immunocompromised, ACV-resistant, and
atopic patients As Margolis et al have reported, individuals
with atopic dermatitis are unusually susceptible to HSV
infection and may develop dissemination (eczema herpeticum).
FIGURE 49.20. (a) Acute recurrent HSV

blepharitis in an eczema patient resembles
primary herpetic disease owing to immune
system dysfunction. (b) Same patient 5 weeks
later after topical antiviral therapy was
administered to protect the eye.
658
It is important for clinicians to be cognizant of the unique

features of herpetic disease in immunocompromised patients,
as their clinical courses and management appears to differ from
that of immunocompetent patients.301303 The role of corticosteroids has not been determined but at the present time would
appear to be inadvisable except in more severe visionthreatening corneal immune reactions in patients responding
well to antiviral therapy.
Surgical Intervention in Ocular HSV

During the 1980s and 1990s, changes in both medical and
surgical management of ocular herpetic disease greatly reduced
the need for penetrating keratoplasty (PK) and improved the
prognosis for surgical intervention when it occurs. This
includes the advent of more judicious use of corticosteroids,
antiviral drugs, therapeutic soft contact lenses, tissue adhesives,
and amniotic membrane transplants. Lateral tarsorrhaphy may
be used adjunctively with a therapeutic lens to protect and
heal re-calcitrant ulceration or prevent breakdown of unhealthy
surface epithelium. Lamellar keratoplasty has had an unfavorable past record because of poor visual results but is still
used with success in patients with quiet scarring conned to
the anterior half of the stroma.304306 The conjunctival flap or
conjunctival transplant is now largely reserved to resolve acute
disease in inflamed, ulcerated, thinning corneas that cannot be
controlled with medical or other therapeutic measures such as
amniotic membrane transplant mentioned previously.221,307,308
See Therapy of HSV Neurotrophic Ulcers above. Even a flap is
not totally protected from recurrence of infectious HSV ulcers,
however. It is advisable to use oral antivirals for several months
in patients with re-calcitrant disease or prone to infectious
recurrence.309 PK is the procedure of choice in the visual
restoration of patients with signicantly scarred or chronically
inflamed herpetic eyes. Two epidemiologic studies, however,
have shown a marked decrease in the need for keratoplasty over
the past decades. From 1972 to 2001, HSK was declined as an
indication for PKP at UCSF. The authors felt that it was
unlikely that this decline was the result of improved diagnostic
accuracy as detection of HSV DNA in corneal buttons with a
clinical diagnosis of HSK was similar at the beginning and end
of the study period.310 In a second study on indications for graft
and regraft, viral keratitis (5.9% of 784 patients) included both
herpes simplex and herpes zoster and showed a statistically
signicant decreasing trend using regression analysis
(p <0.005). In the regraft subgroup, viral keratitis accounted for
21.2% as the underlying primary diagnosis and was the leading
indication for regrafting although declining, was the leading
primary diagnosis.311
While the incidence of HSV keratoplasty is decreasing, the
improved medical management and surgical techniques now
justify this surgery, not only for visual purposes but also to
remove inflammation-inciting, viral antigenic material lodged
in the cornea. Nonetheless, the 5 year success rate of HSV
grafts is well below that of such conditions as keratoconus and
corneal dystrophies.311313 It is of interest that the return of
corneal sensitivity after PK for HSV is negligible compared
with PKs done for dystrophies or bullous keratopathy. This
anesthesia should be borne in mind in terms of protecting the
HSV-graft with lubricants.314
Langston et al reported a clear graft success rate for at least
210 years postoperatively in herpetic eyes as being related to
(1) reduced or absent inflammation, (2) minimal deep vascularization, (3) use of 100 or ner nylon sutures, and (4) use of
very high doses of postoperative topical steroids.315 Recurrence
of dendritic disease in the graft averaged 15% within 2 years,
and postoperative complications such as rejection, synechia
formation, and glaucoma were signicantly lower in the highsteroid dosage groups. Boisjoly and colleagues study revealed
that factors in graft rejection were recurrence of herpetic
keratitis, transplant sizes greater than 8 mm, recipients younger
than 60 years of age and vascularization of the corneal bed
(Fig. 49.21).316 Tuppin et al in a survey that included 8904 eyes
that underwent keratoplasty between 2000 and 2002 in France,
reported the main risk factors for all groups were vascularization in more than two quadrants, planned recipient diameter
over 8.5 mm, immunologic disorders, and previous ocular
CHAPTER 49
Further, they may develop severe and bilateral herpetic ocular

disease. The keratitis is commonly complicated by stromal
scarring and slow epithelial healing despite topical antiviral
therapy.169
In their report on management of acyclovir-resistant HSV,
Chihikuri and Rosen noted that in immunocompetent patients,
nonocular HSV is controlled fairly rapidly by the human hosts
immune system, and recurrent lesions are small and shortlived often. When treated with antiviral agents, these patients
rarely develop resistance to these drugs. However, immunocompromised patients are often unable to control HSV infection
resulting in frequent and serious re-activations. Treatment in
these patients may be limited because immunocompromised
hosts may develop disease refractory to antiviral drug therapy.
Physicians must be prepared to deal with both receptive and
refractory HSV disease. Current recommendations for atopic
and more severely immunosuppressed patients:297,298,169
1. Gentle debridement with moist Weck cell sponge. Culture
if possible or use Smartcycler PCR test to conrm
diagnosis.
2. In all patients weighing > ~25 kg, oral ACV therapy
should be initiated at a dose of 400 mg po 5id. If the
response is poor, the dose of oral ACV should be increased
to 800 mg ve times a day. If no response is seen after
57 days, ACV dosage should be raised to 800 mg po 5id
or, alternatively, more potent antiviral FCV 500 mg po tid
or VCV 1 g po tid initiated along with topical trifluridine
six to nine times daily. Watch for corneal toxicity after
12 weeks.
In children, if IV therapy is needed, the dose is ACV
20 mg/kg q 8 h for 710 days. This is followed by oral ACV
400600 mg/day in 24 doses.299
3. If there is still no response in adult or child, it is not likely
that the lesions will respond to chemically and structurally
related drugs such as VCV or FCV. Therapy with IV
foscarnet 40 mg/kg three times per day or 60 mg/kg twice
daily, should be given for 10 days or until complete
resolution of the lesions.
4. If foscarnet fails to achieve clinical clearing, consideration
should be given to use of IV cidofovir or application of
compounded 13% topical cidofovir ointment. Vidarabine
is reserved for situations in which all of these therapies
fail.
5. If lesions re-occur in the same location following clearing,
the patient should be started on high-dose oral ACV
800 mg ve times daily, or IV foscarnet (40 mg/kg tid or
60 mg/kg bid) as soon as possible. If lesions occur in a
different location, the patient should be treated initially
with standard doses of oral ACV (200 mg ve times daily)
and the above protocol followed should there be clinical
failure.298
The above therapeutic regimens are not yet
FDA-approved for specic ocular use. Topical antivirals
may be discontinued after 34 weeks of therapy upon
healing of the corneal ulcers.
6. Topical alpha-2A interferon treatment of HSV resistant to
multiple antiiviral drugs has also been reported as
therapeutically effective.300
659
SECTION 6
FIGURE 49.21. Acute rejection of penetrating keratoplasty in a

herpetic eye showing multiple endothelial keratic precipitates.
Rejection responded to hourly dexamethasone over a several-day
period with gradual tapering over several weeks.
660
surgery. Independently associated with a signicantly increased

risk of corneal regraft (P < 0.05) was, among others, herpes
simplex keratitis.317 In contrast, Cohen et al found no
correlation between vascularization and rejection and reported
a success rate as 85% in herpetic eyes, the vast majority of
which were uninflamed at the time of surgery.318 The incidence
of recurrent dendritic keratitis was similar (19%) when steroids
were used with or without prophylactic antivirals thus agreeing
with the study of Fine and Cignetti.319 Cohen et al did note,
however, that 32% of herpetic eyes undergoing therapy for
rejection developed infectious epithelial herpetic ulcerations,
strongly indicating that antiviral prophylaxis should be used in
those patients in whom allograft rejection is occurring and is
under treatment with corticosteroids. This conrmed previous
observations and recommendations by Cobo and associates that
withholding antivirals during the postoperative period had no
adverse effect on the rate of herpetic recurrence but that
antiviral prophylaxis was necessary to prevent herpetic
recurrence in the face of graft rejection under steroid therapy.320
Other studies by Ficker et al on the effects of changing
management in improving prognosis for keratoplasty in
herpetic eyes noted that our improved use of corticosteroids and
prophylactic antivirals increased the success rate in grafting
inflamed eyes to one comparable to that for uninflamed eyes.321
This is contrary to earlier ndings reported by Polack and
Kaufman and Foster and Dunkin and in agreement with those
of Langston et al, and Rice and Jones who also recommended
intensive postoperative antiinflammatory therapy.315,322324
Ficker et al expected long-term survival for rst grafts both in
quiet and inflammed herpetic eyes was 70% if done after
1979.321 They used survival curves not utilized by previous
investigators. They also reported increased success rates with
the use of interrupted sutures, extracapsular cataract techniques where extraction was indicated, prompt removal of
loose sutures (a trigger factor for both rejection and herpetic
recurrence), adequate topical steroid therapy to assure a
quiescent eye, and antiviral prophylaxis during intensive steroid
treatment for rejection (but none was deemed necessary as
routine postoperative management). Survival of secondary grafts
was signicantly worse than for primary grafts, but preoperative
vascularization was not found to be a risk factor. The incidence
of HSV recurrence was 15% over a mean 4 year period, with

89% of these being epithelial and 11% stromal.
This supports the recommendation by Barney and Foster
and Simon and Pavan-Langston that oral ACV 400 mg bid
be used for 1218 months postoperatively to prevent recurrence
of infectious keratitis in the graft, thus decreasing the failure
rate.267,325
In a survey taken from 36 active cornea surgeons in Great
Britain and the United States, risk factors and management
showed a wide variation in preferences and certainly the need
for greater use of postoperative antiviral prophylaxis in patients
with a history of ocular herpes. Other results from the study
were factors considered to be high risk for corneal graft
rejection: previous corneal graft rejection in the operated eye
(97%), signicant corneal vessels (97%), and previous herpetic
eye disease (94%). The preferred preoperative treatment in
high-risk patients included no treatment (47%), topical
corticosteroids (33%), and oral prednisolone (22%). In postoperative high-risk patients, 100% of surgeons used topical
and 42% used oral corticosteroids. Immunosuppression was used
by 44% of respondents, the majority (92%) using cyclosporine
A. In previous herpes simplex patients, 47% of surgeons used
oral and 14% used topical antivirals preoperatively, while 75%
used oral and 47% used topical postoperatively.326
HSV trigeminal ganglion latency as a source of recurrent
infection was already well documented when reports of HSV
corneal latency and persistance of HSV in ocular tissues began
to appear in the late 1980s.79,98,119,120,152,230,232,327 Further in
1991, Mannis et al reported the occurrence of HSV between
3 and 11 months postoperatively in grafts in eyes with no
history of HSV thus raising the index of suspicion for this
infection for all cases of late-onset epithelial defects in any
graft.328 The detection of HSV DNA in two corneal donor
buttons that had degenerated during storage preoperatively and
in the failed graft of one of the donor pairs used also raised
suspicions of the origin of recurrent HSV in grafts not always
coming from the host.114
More recent studies on the relationship between HSV and
graft success or failure have shed considerable light on the cause
of some failures and the need for certain forms of postoperative
management. Liekfeld et al performed micro-ELISA assay for
HSV, VZV, and CMV on aqueous taken from 24 herpetic eyes
and found antibodies against HSV in 50%, HSV and VZV in
25%, VZV alone in 3.6%, and no antibodies against either in
22%. They suggested the need for peri- and postoperative use of
antivirals as advisable but also that about one-fourth of patients
may not need this prophylaxis.329 PCR studies for viral DNA
on 31 herpetic corneal buttons taken at keratoplasty and 78
nonherpetic eyes revealed that in the HSV specimens one-third
were positive for HSV-1, 3% for HSV-2, and 19% for VZV.330 In
nonherpetic specimens HSV-1 DNA was detected in 17%
including eight grafts that failed without clinically obvious
HSV or a history thereof. Similarly a report of four nonherpetic
patients with unexplained primary graft failure were found to be
culture positive for external HSV, PCR-positive in aqueous,
corneal graft, and iris tissue testing.331
The origin of HSV in recurrent disease then may or may not
always be re-activation of the patients own re-activated latent
HSV and may occur in patients with no history of ocular or
other herpes. The study on causes of primary graft failure (PGF)
by Cockerham et al using clinical, histologic, immunohistochemical, PCR and, occasionally, transmission electron
microscopic studies revealed no evidence of HSV type 2 or VZV
in any cornea. All control corneas were negative for viral DNA.
Herpes simplex virus type 1 DNA was present in 33% of
patients with PGF. Herpetic stromal keratitis was found in
some failed corneas. The authors felt that the lack of HSV in
the paired recipient suggested viral importation into the donor

cornea.112 Rezende et al reported 14 patients with new-onset
HSV in recent grafts in patients with no history of herpetic
disease.332 Remeijer et als study on 30 nongrafted patients
revealed that 63% were genotypically the same from recurrence
to recurrence. However, 37% were actually genetically different
suggesting a signicant number of recurrent HSK cases may be
exogenously acquired.63 Several other studies on grafted patients
have conrmed these ndings of both host and recipient may
serve as the source of the virus.54,96,333,334 Animal studies by
Zheng et al on the phenomenon of transmission of HSV
via keratoplasty revealed that corneas from latently infected
rabbits contain HSV-1 DNA that can replicate after induced
reactivation and that viral migration may be both anterograde
or retrograde between donor cornea and recipient rim and
trigeminal ganglion.335,336 Further, lamellar keratoplasty induces
HSV-1 shedding and recurrent epithelial lesions in rabbits
latently infected with HSV-1 preoperatively but not in
uninfected control animals.337
In outcome studies on the AlphaCor articial cornea
implantation in patients with and without a history of ocular
herpes simplex virus, the extensive lamellar corneal surgery
involved may precipitate re-activation of latent HSV with the
resulting inflammation reducing device biointegration and
facilitating melting of corneal stromal tissue anterior to the
device. It was concluded that prior HSV was a contraindication
to use of this form of articial cornea.338
In contrast, if the patient has had HSV or HZO (zoster), one
or more grafts that have failed, or an anterior segment with
a nonhealing, even inflammed trophic ulcer, the Boston
keratoprosthesis is now becoming the standard of care for
these otherwise inoperable, high-risk, or highly inflamed
herpetic eyes.339341 This prosthesis has been used with a high
rate of success in very high-risk patients, both with HSV and
zoster (VZV).
In summary, it would appear that four major factors emerge
as key to long-term survival of keratoplasty in herpetic eyes:
(1) use of ne, interrupted sutures (less trauma on removal),
(2) intensive postoperative topical steroids to suppress inflammation; taper over 1 year, (3) ACV 400 mg po bid for 1218
months (or longer as needed) prophylaxis against recurrence of
infection in the graft, (4) full antiviral prophylaxis (po or topical)
during intensive topical steroid therapy for rejection. Because of
the signicantly worse survival rate in regrafts, it is emphasized
that the importance of immediate and intensive treatment
of complications of rst grafts be addressed by all treating
physicians. Despite the above the graft fails, the Boston
keratoprosthesis has proved a highly successful alternative in
complicated re-grafts.34
VARICELLA-ZOSTER VIRUS (VZV)

OPHTHALMICUS
Varicella (chickenpox) and herpes zoster (shingles) are two distinct
clinical diseases caused by the same organism, the varicellazoster virus (VZV). The viruses are identical antigenically and
on a molecular biologic level. Transmission of VZV from a
herpes zoster patient often results in classical varicella.5
Ocular Varicella
Positive VZV seroconversions approach 100% by age 60 years in
the United States, with an estimated 2 800 000 cases of varicella
occurring annually prior to release of the varicella vaccine for
children or adults with no history of chickenpox. Although
varicella is usually a mild illness, complications leading to
morbidity and mortality are signicant and the disease is worth
preventing. The vaccine offers close to 100% protection from
severe chickenpox and 90% protection from illness. Waning of

immunity after vaccination, particularly in children, has not
been a signicant problem although there are cases of
chickenpox in those previously vaccinated (see section on
Vaccine below).342 The major prevaccine complications,
pneumonia and encephalitis, accounted for 100200 deaths per
year.343349 Varicella represents a patients rst encounter with
the organism. Local infection of the nasopharynx, or rarely via
the skin is followed by waves of viremia and seeding of the
reticuloendothelial cells, skin, viscera, and ganglia.345,350,351
Ocular manifestations of varicella may be either those of
congenital varicella syndrome or those of the more common
generalized varicella most commonly seen in young
children.347,348,352 The congenital varicella syndrome is the
result of maternal varicella infection during pregnancy, most
frequently during the rst or second trimester.353 Systemic
ndings may include hemiparesis, bulbar palsies, cicatricial
skin lesions in a dermatomal distribution, developmental delay,
and learning difculties. Ocular ndings may include chorioretinitis, optic nerve atrophy or hypoplasia, congenital cataract,
and Horner s syndrome. These congenital malformations
appear largely the result of the high afnity of VZV for the
nervous system.353,354 There is no denitive therapy for
congenital varicella syndrome. As 516% of women of childbearing age are susceptible to varicella, the adult vaccine now
approved by the FDA is the best hope to minimalize or possibly
even eradicate this devastating illness.355 An alternative is, of
course, early and intensive treatment of maternal varicella with
oral antivirals such as acyclovir 800 mg po 5id, famciclovir
500 mg po tid, or valaciclovir 1 g po tid for 710 days and hope
that this will prevent malformation.
Classical varicella or chickenpox has an incubation period of
about 2 weeks after exposure before the onset of the viremia
that produces fever, malaise, and an infectious mucocutaneous
exanthem. This maculopapulovesicular rash appears in successive crops, so lesions in various stages are present simultaneously. The infectious period is ~1 week after the appearance
of each crop of lesions or until the cutaneous sores crust over.356
Occasionally, the vesicular rash may involve the lid margins,
and more rarely, vesicular lesions may appear on the conjunctiva. These are usually unilateral, small phlyctenule-like
lesions that may erupt most commonly at the corneal
limbus.347,351,352,357 It is unclear whether these are due to live
virus or an immune phlyctenule-like reaction, or both
(Fig. 49.22). They may resolve without problem or may become
pustular punched-out, dark red painful ulcers with swollen
margins and with secondary inflammatory reaction in the eye.
Varicella keratitis may develop either as an infectious
supercial punctate keratitis or with branching dendritic
lesions.358360 Varicella dendrites may be distinguished from
those of herpes simplex in that they are ne, nonulcerated,
linear lesions that lack the classic terminal bulbs of HSV
dendrites and appear to be heaped up on intact underlying
epithelium (Fig. 49.23). Gentle blotting of these varicella
lesions will commonly leave no underlying ulcer, whereas
removal of HSV dendrites will leave a positive-staining, fullthickness epithelial defect. There is local anesthesia in the
area of the varicella dendrite, and these lesions may very rarely
expand to form geographic epithelial defects.
Weeks to months after the initial episode of infectious
varicella, a patient may develop infectious varicella dendritic
keratitis that may run a course of successive crops of dendrites
similar to the successive crops of lesions seen during the acute
dermatitis. A healthy 10-year-old child developed chronic
recurrent varicella virus keratitis with pseudodendrites after
recovery from systemic varicella. The debrided pseudodendrites
were repeatedly positive for VZV DNA and negative for HSV
CHAPTER 49
661

extraocular muscle palsies, internal ophthalmoplegia, cataract,
chorioretinitis, and optic neuritis.351,352,364367
Therapy of ocular varicella
FIGURE 49.22. Acute varicella limbal phlyctenule (arrow) appearing

during the course of disseminated chickenpox.
SECTION 6
FIGURE 49.23. Acute varicella dendritiform ulcers developing 6 weeks

after resolution of chickenpox. Three similar episodes occurred over a
several-month period, two in association with a mild stromal disciform
edema. The entire process resolved without scarring.
662
From deFreitas D, Kelly L, Pavan-Langston D, et al: Late-onset varicella-zoster

dendritic keratitis. Cornea 11:471, 1992.
DNA. The lesions responded to oral acyclovir and topical corticosteroid drops but recurrences occurred once the medications
were discontinued. Varicella virus epithelial keratitis can be a
recurrent condition requiring prolonged therapy.361 Similarly,
several weeks to months after the acute epithelial disease an
immune disciform keratitis similar to that seen in HSV disease
may develop.362 This disciform reaction is usually mild and
steroid-responsive but may be recurrent and cause some
scarring before resolving over a 2- to 5-month period. One case
of perforation of an ulcer of unknown etiology (but 7 months
after uncomplicated varicella) proved, on PCR and E/M study, to
be due to delayed primary varicella keratitis.363
Other less frequently seen manifestations of ocular varicella
are iritis (occasionally brinous), lid necrosis, IK with neovascularization, neurotrophic ulceration with corneal melting,
Therapy of varicella lid lesions, conjunctival phlyctenules, or

epithelial keratitis is not conclusively established, but current
recommendations are acyclovir in doses appropriate for the
patients age and weight. Adults (> 40 kg) are 800 mg po 5id for
710 days and children 20 mg/kg po qid 5 days.42 The American
Association of Pediatrics does not approve of routine treatment
of all cases of chickenpox but just in chronic dermatitis,
pulmonary disease, and (not FDA approved) ocular lesions.367
TFT has also been reported as useful in resolving the lesions,
but it is also likely that the lesions resolve without any treatment.362 The disciform keratitis or iritis is commonly managed
with mild topical steroid, 0.125% prednisolone, one to four times
daily with tapering over a several week to 2- to 5-month period as
described earlier in this chapter. The disciform disease of
varicella may recur a number of times before becoming quiescent.178 There is no consensus on the treatment of this optic
neuritis and the current attitude, other than oral acyclovir, is therapeutic abstention because of a rapid spontaneous improvement.
Varicella vaccine In 1995, the FDA approved a live, attenuated
VZV vaccine, the Oka strain, to immunize healthy people of
all ages (infancy to adulthood) who have not had previous
varicella and thereby reduce the incidence of varicella and its
complications.342,368370 How long the vaccine is effective is as
yet not well established. It is known, however, that while the
postvaccination antibodies decline over the years in adults and
leukemic children, this does not happen in healthy children
studied for 1020 years.370372 Cell-mediated immunity may
continue to protect long after antibody levels are low but
periodic booster shots of the vaccine may also be necessary to
maintain protection much as we repeat booster tetanus and
other vaccines. The recent observation that vaccinated children
may, in fact, develop acute varicella showed that the children
immunized 3 years prior to exposure were at greatest risk of
developing disease, but also that vaccinated children where
much less likely to develop moderate or severe disease.373
Perhaps an even more important question, is does the
vaccine reduce the incidence of zoster? Given the increasing
average age of our population the effect of vaccination on
preventing zoster is now well elucidated. Studies on the effect
of giving live, attenuated VZV vaccine (Oka strain) showed a
signicant increase in VZV cell-mediated immunity (CMI) in a
healthy, elderly (5575 years old) population.374,375 No relationship between vaccine dose and the intensity of the specic
response was noted. In leukemic children, the incidence of
zoster was 15% in unvaccinated controls compared to just 3%
in vaccinated patients possibly reflecting a lower rate of latency
after vaccination as there is no skin infection compared to that
seen in natural disease.376,377 Data in the normal population
indicate that the incidence of zoster is much lower in vaccinated
healthy children and adults compared to those who suffered
a natural infection.378 Further, there has however been one
case of zoster ophthalmicus reported in a 3-year-old child
5 years after vaccination but it could not be proven whether this
was due to the vaccine strain or a natural strain of VZV.379 In
another report, however, a child developing zoster sclerokeratitis
and uveitis 3 years after vaccination proved to have wild-type
virus, not re-activation of the vaccines Oka strain.380
But the most important study to date has just recently been
reported. Oxman et al tested the hypothesis that vaccination
against VZV would decrease the incidence, severity, or both of
zoster and postherpetic neuralgia among older adults. Enrolled
were 38 546 adults 60 years of age or older in a randomized,
HERPES ZOSTER (HZ) AND HERPES ZOSTER

OPHTHALMICUS (HZO)
Epidemiology and Incidence
Herpes zoster is a recurrent infection with varicella virus,
endogenous (reactivated latent dorsal ganglionic virus). Up to
20% of the worlds population will suffer from zoster at some
time in life. Approximately 50% of individuals reaching
90 years of age will have had HZ. In about 6%, a second attack
may occur (usually decades after the rst). Patients with HZ can
transmit the virus to a nonimmune individual causing varicella
but HZ is not contracted from individuals with varicella. An
estimated 1 million new cases of infectious zoster occur
annually in the United States of which ~10% involve the eye or
adnexae.355,381,382,36,383,384 The reported annual frequency of
herpes zoster in the American and British populations in the
1970s and 1980s fell within a range of 24%.385390 HopeSimpsons British study indicated that the annual attack rate
rises steadily between 0 and 19 years of age, plateauing at three
cases/1000 between ages 20 and 49, and then rising sharply to
10/1000 by the eighth decade of life.388 More worrisome are
gures reported in 1995 indicating that in the United States the
number of people over the age of 65 will double and the number
of octogenarians will go from 3.5 million to 8.8 million thus
increasing the potential number of new cases of zoster.390,391
In more recent studies, data reveal either different or
conrmatory results. In the United States, 9152 incident cases
of HZ (3.2/1000 person-years, P<0.05) were reported.392 The
annual HZ rates/1000 person-years were higher among females
(3.8) than males (2.6) (P<0.0001). Disease rates rose sharply
with age, and were highest among individuals over age 80
(10.9/1000 person-years, P<0.01). Among patients with
evidence of recent care for transplantation, HIV infection, or
cancer the incidence of HZ per 1000 person-years (10.3) was
greater than for patients without recent care for these
conditions (3.0) (P<.0001). The authors noted that the overall
incidence of HZ reported in the present study was found to be
similar to rates observed in US analyses conducted 1020 years
earlier, after age- and sex-standardizing estimates from all
studies to the 2000 US population. The higher rate of HZ
in females compared with males was conrmed in genderspecic age-standardized incidence rates of shingles in 14 000+
patients calculated from 19942001 and showed a consistent
female excess in each year (average annual excess 28%).393
Studies on racial aspects of zoster reveal that Black and Asian
racial groups from tropical regions as opposed to temperate
zones were each signicantly associated with younger age at
zoster onset. Black and Asian patients did not signicantly
differ in age or sex, however.394 Schmader et als studies on
racial incidence clearly show that Caucasians are four times
more likely than Blacks to develop zoster.395
While the adult vaccine may represent a major breakthrough

in eliminating or reducing the severity of illness from this
potentially devastating virus there is still one major concern. It
is hypothesized that exposure to varicella may boost immunity
to latent VZV and that the vaccination-associated decrease in
varicella disease will cause the incidence of HZ to increase.
Jumaan et al looked at the incidence of herpes zoster, before
and after varicella-vaccination-associated decreases in the incidence of varicella from 19922002.396 They found that the
vaccination-associated decrease in varicella disease did not
result in an increase in the incidence of HZ. The age-adjusted
rates of chickenpox decreased from 2.63 cases/1000 personyears during 1995 to 0.92 cases/1000 person-years during 2002
while the incidence rates of HZ fluctuated only slightly over
that time (4.05 cases/1000 person-years in 1992, and 3.71
cases/1000 person-years in 2002).
Conversely, Yih et al found that between 1998 and 2003,
varicella incidence declined from 16.5/1000 to 3.5/1000 (79%)
overall with 66% decreases for all age groups except adults
(27% decrease).397 However, overall herpes zoster occurrence
increased from 2.77/1000 to 5.25/1000 (90%) (P < 0.001).
Annual age-specic rates were somewhat unstable, but all
increased, and the trend was signicant for the 2544 year and
65+ year age groups. The authors concluded that as varicella
vaccine coverage in children increased, the incidence of varicella
decreased and the occurrence of herpes zoster increased. We
await the effect of adult vaccination on these contrasting
studies. These ndings agree with those of the Varicella Active
Surveillance Project indicating that HZ may be increasing
among adults. The authors were concerned that booster costeffective interventions that meet or exceed the level of
protection provided by that immunologic boosting that existed
naturally in the community in the prelicensure era could not
be met.398
Herpes Zoster Ophthalmicus (HZO)

The Mayo Clinic studies on prognostic factors in zoster severity
noted an incidence of trigeminal nerve zoster (herpes zoster
ophthalmicus, HZO; Table 49.3) of 6.1%, 9.3%, and 16%,
which are at the low end of the 856% range in other
TABLE 49.3. Complications of Herpes Zoster Ophthalmicus

(Incidence Values in %)*
Lids
Entropion (6)
Scarring of both upper and
lower lids (3)
Cicatricial ectropion (2)
Ptosis (marked) (1)
Total (12)
Corneal
Acute epithelial keratitis
Pseudodendritic keratitis (8)
Punctate epithelial keratitis (14)
Mucous plaques (2)
Total (22)
Disciform keratitis (20)
Neurotrophic keratitis (12)
Acute anterior stromal
inltrates (8)
Sclerokeratitis (2)
Late dendritic keratitis (1)
Perforation (1)
Total (44)
Sclera
Scleritis (3)
Episcleritis (1)
Total (4)
Canalicular Scarring (2)
Iridocyclitis
Diffuse (38)
Sectoral iris atrophy (17)
Localized (2)
Total (57)
Glaucoma (Secondary) (12)
Persistent (2)
Cataract (8)
Neuroophthalmic Involvement
Cranial nerve palsy (3)
Contralateral hemiplegia (2)
Segmental cerebral arteritis (2)
Total (7)
Postherpetic Neuralgia (17)
*86 patients total. Some had more than one complication.

Adapted from Womack L, Liesegang T: Complications of herpes zoster
ophthalmicus. Arch Ophthalmol 101:4245, 1983.
CHAPTER 49
double-blind, placebo-controlled trial of a live attenuated

Oka/Merck VZV vaccine of greater strength than that used in
children.355 The vaccine signicantly reduced morbidity from
herpes zoster and postherpetic neuralgia. There were 957 conrmed cases of herpes zoster (315 among vaccine recipients and
642 among placebo recipients) and 107 cases of postherpetic
neuralgia (27 among vaccine recipients and 80 among placebo
recipients). The zoster vaccine reduced the burden of illness due
to herpes zoster by 61.1% (P < 0.001), reduced the incidence of
postherpetic neuralgia by 66.5% (P < 0.001), and reduced the
incidence of herpes zoster by 51.3% (P < 0.001). Reactions at
the injection site were more frequent among vaccine recipients
but were generally mild. This vaccine has received FDA
approval. The potential impact on the epidemiology of the
childhood and adult vaccines is discussed ahead.
663

reports.388,391,399401 Womack and Liesegangs 19751980 study
of 86 cases of HZO revealed a predominance in female patients
and in the left eye and a peak incidence in the seventh and
eighth decades of life.402 This differed from other series showing
a male predominance or equal distribution between the sexes
and a peak incidence in the fourth or fth through the seventh
decade of life.386,403405,401 Of the Mayo Clinic series of 64
patients with acute HZO, 72% had ocular or adnexal disease, a
gure signicantly higher than the 50% usually noted in the
literature.402 As Caucasians are four times more likely to suffer
zoster than African-Americans, they are signicantly more
likely to develop HZO.395
Occult malignancy has long been a concern of physicians
with newly diagnosed zoster. Scheie noted that among
hospitalized patients zoster occurred in 0.2% of those without
malignant disease and in 0.85% of those with malignancy.406
This does not take into account, however, the many thousands
of patients with zoster who are not hospitalized and who do not
have malignant disease. The relationship with occult
malignancy has never been adequately demonstrated, although
there is an increased incidence in those patients with overt
malignant disease such as leukemia or lymphoma. Buntinx et
al did a retrospective cohort study on 311 000 Belgians and
found that in patients with and without herpes zoster only
above the age of 65 years was there a signicant increase of
cancer emergence in the whole group and in females but not in
males. No difference could be identied in the rst year after
the herpes zoster infection. The authors concluded that their
results did not justify extensive testing for cancer in herpes
zoster patients. In contrast, a 5 year retrospective study in
India on a 399 zoster patient cohort reported that within 3 years
of zoster, three patients developed acute leukemia, two
developed mycosis fungoides, and 17 developed HIV, suggesting
that physicians must not forget that malignant disease or
immunosuppressive disease may develop in patients not long
after zoster infection.407 Risk factors, then, underlying development of zoster, ophthalmic or elsewhere include advancing
age, female gender, ethnicity, genetic susceptibility, lack of
exogenous boosting of immunity from varicella contacts, underlying cell-mediated immune disorders, mechanical trauma,
psychological stress, immunotoxin exposure, clinical depression
(living alone) but probably not occult malignancy.390,400,408411
SECTION 6
Pathogenesis and histopathology of ocular and

central nervous system (CNS) HZ
664
During the primary disease, varicella, the virus gains access to

the trigeminal sensory ganglia by viremia and retrograde
migration from the skin where, like HSV, it enters a latent state
(Fig. 49.3). The complete viral genome becomes latent in most
dorsal root and cranial ganglia: 6590% of trigeminal, 5080%
of thoracic, and 70% of geniculate, thus making the trigeminal
form the most common in the body.36,383,400,412 Attempts to
recover VZV from ganglia at autopsy have been unsuccessful
although the virus has been recovered if active VZV existed in
the corresponding dermatome at the time of death.413,414
During active infection, it is known that virus replicates both in
neural and nonneuronal satellite cells. By using a combination
of in situ PCR and in situ hybridization, VZV DNA has been found
to exist only in the neuronal nuclei during latent infection.415
Although it is clear that competent CMI is essential to
prevent clinical VZV disease, the immune response does not
completely prevent re-activation. It appears that subclinical
infections occur in both immunocompetent and immunocompromised patients several times during life.416,417 This appears
to reboot the VZV CMI system such that the CMI acts to
inhibit spread of VZV within the ganglion and subsequent
spread to the skin or eye. During these subclinical re-activations
there is also a rapid rise in glycoprotein VZV antibody level with

the greatest increase in gp98 and gp62 which persist for at least
2 years.418 Occasionally, there may be partial breakthrough in
the form of recurrent dermatomal pain, corneal pseudodendrites, or even a marked uveitis. With or without a previous
history of overt herpes zoster, these VZV re-activations without
dermal eruption are classied as zoster sine herpete (discussed
below). During these contained recurrences, there is a marked
boost in T-cell response and incomplete VZV DNA is present
in peripheral mononuclear cells.419,420 The ability of CMI to
continue these recurrences determines whether a patient will
sometime later have a full-blown attack of herpes zoster.
This latent DNA is capable of re-activation to the infectious
state after disturbance of the hostparasite relationship years
to decades after the original infection. Virions have been
detected by electron microscopy in neurones following VZV
infection, and biopsy of an involved dermatome on the rst
day of eruption showed loss of nerve ber staining, both of
which suggest that nerve involvement precedes that of the
skin.95,413,421,422
In anterograde spread, the virus travels down the fth nerve
axons to the skin resulting in a demyelination, granulomatous,
mononuclear cellular inltration, and consequent brotic
scarring of peripheral nerves and end organs (skin and eye)
affected. Acute skin and conjunctival vesicles and dendritic
ulcers are infectious in etiology. The skin eruption is simultaneous with a strong VZV-specic T-cell proliferation. Current
evidence as to the pathogenesis of stromal disease indicates
that, as in herpes simplex, AAC-mediated reaction is responsible for the Wessely immune rings, necrotizing interstitial
keratitis, and limbal vasculitis (local Arthus reaction).400,423426
The diffuse or local gray stromal edema of disciform disease
and iritis are thought to be primarily lymphocyte-mediated
delayed hypersensitivity reactions to virus or viral antigen. The
chronic HZO keratitis may be noted to have a giant-cell
reaction at the level of Descemets membrane and VZV DNA
may be detected in human corneas at least 8 years after the
acute event.427,428 In addition, mechanical healing problems
(trophic ulcers) may result from the abnormal precorneal tear
lm encountered with neurotrophic changes, exposure, and
scarred meibomian gland orices. The neurotrophic changes
and density of corneal anesthesia are in all likelihood a
function of the ganglionic and brainstem damage with subsequent corneal denervation (see HSV Neurotrophic Ulcers
above).429431
Re-activating VZV also causes inflammation and hemorrhagic necrosis often associated with neuritis, leptomeningitis,
segmental myelitis, and related motor and sensory root
degeneration.430 This retrograde viral spread to the brainstem
and spinal cord via a large branch from the TG damages the
corresponding mesancephalic sensory (pain) nucleus as well as
to regional arteries and the CNS (Fig. 49.3).432,433 Postmortem
exam of those who die of VZV shows satellitosis, lymphocytic
inltration and necrosis of the ganglia, and viral particles in
granulomatous arteritis.434 Cerebrovascular accident is not
infrequent.435 Virus particles have also been found in acute VZV
in the trigeminal ganglion and its axons, in CNS tissues, and in
the arterial walls of ocular and CNS tissues by electron
microscopy and immunofluorescence.429 There is VZV DNA
in peripheral mononuclear cells.436 Postmortem exam of the
Gasserian ganglion (trigeminal) and the brainstem mesencephalic
(trigeminal) nucleus in three old HZO patients who died years
after the acute infection revealed the primary lesions in the
semilunar ganglion varied with the length of the clinical course.
Secondary changes in the brainstem from pons to the second
cervical segment of the cord showed degeneration, inflammation, and glial nodules in the mesencephalic trigeminal
Neuronal relationships of trigeminal zoster (HZO)

The thoracic dermatomes are the most commonly affected,
followed closely in frequency by the trigeminal cranial nerve
(Fig. 49.3). The ophthalmic division of this nerve is affected
about 20 times more often than are the second or third
divisions.441 In the ophthalmic division, the frontal nerve is the
most frequently affected, via its supraorbital and supratrochlear
branches the frontal nerve innervates the upper eyelid, the
forehead, and some superior conjunctiva. The primary sensory
nerve to the eyeball is the nasociliary branch, which supplies
the lacrimal sac, the conjunctiva, the skin of both lids, and the
root of the nose via the infratrochlear nerve. The nasal branches
of this nerve, however, along with the sympathetic branches
from the ciliary ganglion, innervate the sclera, cornea, iris,
ciliary body, and choroid via the long and short ciliary nerves (as
well as the less critical but diagnostically helpful side of the tip
of the nose). Involvement of the tip of the nose is called
Hutchinsons sign, a sign taken to indicate that the eye may
be seriously involved by VZV because of the involvement of
the nasal branch of the nasociliary nerve.442 By direct neural
connection to the many external and internal ocular structures,
and by direct spread through the orbital tissues to other cranial
and autonomic nerves and to the central nervous system nuclei
of ocular and orbital nerves, the zoster virus is able to cause a
wide variety and severity of disease (Table 49.4) gives a more
complete listing of the complications of HZO, but in brief
the most common are cicatricial lid retraction or loss,
paralytic ptosis, conjunctivitis, episcleritis, scleritis, keratitis
(infectious or immune), iridocyclitis, retinitis, choroiditis, optic
neuritis, optic atrophy, retrobulbar neuritis, Argyll Robertson
pupil, exophthalmos, extraocular muscle palsies, and
glaucoma.5,351,387,404,406,400,423,445447,448,443
Equally important, as noted above, is the large branch sent
from the TG to the brainstem (mesencephalic nucleus) from
the pons to the level of C-2.443,431 Inflammatory destruction in
this area appears to be directly related to neurotrophic keratopathy (NTK) and postherpetic neuralgia (PHN) (Fig. 49.3).
HZO clinical disease

Clinical disease may not only accompany the acute disease but
recur periodically or smoulder chronically for months to years
in any of its many forms. Zoster infections occur by one of
two mechanisms: (1) re-activation in the trigeminal sensory
ganglion of latent virus or (2) re-introduction of exogenous virus
through direct or indirect contact with either a chickenpox or
zoster patient. The incubation for endogenous zoster is not
known, but in those patients exposed to chickenpox, incubation
varied between a few days and 2 weeks. The illness may begin
with headache, malaise, dysesthesia but rarely fever, followed
2448 h later by neuralgia and dysesthesia, and 23 days after
that by hyperemic, hyperesthetic edema of the involved
dermatome, which erupts with multiple crops of watery blisters
that continue to form over 35 days.
Dermatitis Occasionally, the dermatitis may never develop (see
Zoster sine herpete below); VZV DNA has been isolated from
the aqueous in an idiopathic keratouveitis without skin
TABLE 49.4. Incidence of Herpes Zoster Ophthalmicus Corneal

Lesions
Finding
Incidence (%)
Punctate epithelial keratitis
51
Pseudodendrites
50
Anterior stromal inltrates
41
Keratouveitis endotheliitis
34
Neurotrophic keratitis
25
Delayed mucous plaques (pseudodendrites)
13
Exposure keratitis
11
Disciform keratitis
10
Serpiginous ulceration
Scleral keratitis
Delayed limbal vasculitis
CHAPTER 49
nucleus.431 A granulomatous intracranial arteritis was also

noted and has been reported by others.432,433
Hedges and Alberts study on the histopathology of acute
and chronic HZO revealed, in the acute cases, normal corneal
stroma, nongranulomatous inflammation of the iris, ciliary
body with extension into the choroid, macrophage and other
inflammatory cell inltration of the trabecular meshwork; all
of these abnormalities appeared to be reversible.428 In other
patients, a severe retinitis was noted over areas of intense
granulomatous choroiditis containing epithelioid and giant cells
with areas of hemorrhage. Similarly the optic nerve, meninges,
and central retinal vessels could be involved in a granulomatous
inflammation with the primary site of involvement being the
optic nerve itself and secondarily the posterior ciliary nerves.
Several studies on chronic HZO described a keratitis
characterized by various combinations of epidermalization of
epithelium, lipid keratopathy, intense stromal vascular scarring,
and giant cell granulomatous reaction to Descemets
membrane.427,428,437,438 Naumann et al also reported a lymphocytic inltration of the posterior ciliary nerves and vessels,
chronic inflammation and vasculitis of the iris and ciliary body
with patchy necrosis of the iris and pars plicata, perivascular
cufng by chronic inflammatory cells in the retina, and granulomatous choroiditis with one case involving a giant cell
granulomatous arteritis.439 It was suggested that the extraocular
muscle palsies and orbital edema seen in HZO may be the
result of perineuritis and perivasculitis associated with the
generalized orbit inflammation.
Wenkel et al used immunohistochemistry and in situ
hybridization to detect VZV DNA in ve of 14 corneal buttons
taken from HZO patients at keratoplasty or enucleation. In a
larger study on nine eyes and four corneal buttons from 13
patients with HZO, they found vascularization of the corneal
stroma (11 of 13), granulomatous reaction to Descemets
membrane (eight of 13), fusiform-shaped ciliary scarring (ve
of nine), optic neuritis (four of nine), and perineuritis (eight of
nine) and perivasculitis (eight of nine) of the long posterior
ciliary nerves and arteries. Zoster antigen was detected in two
patients with acute infection one and seven days after acute
onset. VZV-DNA was identied in seven patients up to 10 years
after acute HZO. It was found in corneal epithelial cells (two of
13), corneal stroma (ve of 13), but none in the endothelium. It
was also in inflammatory inltrate of the anterior chamber (one
of nine), episclera (two of nine), posterior ciliary nerves (one of
nine) and arteries (ve of nine), optic nerve (ve of nine), and
adjacent leptomeninges (two of nine). The authors concluded
that persistent viral genomes accompanied by gene expression
or slow viral replication, appeared to be responsible for the often
smoldering panophthalmitis and the chronic recurrent
keratouveitis in patients with HZO. Localization of viral DNA
in vascular structures indicated vasculitis in the pathogenesis of
some ocular ndings of HZO.427,438,440
<1
Adapted from Leisegang T: Corneal complications from herpes zoster

ophthalmicus. Ophthalmology 92:316, 1985.
665

lesions.420,449 Steroid treatment failed but oral ACV resolved the
disease in 3 weeks. More commonly typical ocular disease may
occur. Within 23 days of neuralgia around the eye and forehead,
hot, flushed hyperesthesia and edema of the dermatome(s)
develop, and the patient erupts with multiple crops of clear
vesicles from which virus may be cultured for ~35 days. The
vesicles then become turbid and yellow and then crust over as
scabs. Two to 20% of patients will have several vesicles scattered
elsewhere on the body indicating viremia.450,451 Unlike, herpes
simplex skin infections, herpes zoster involves epidermis down
to corium and forms deep eschars that may leave permanent
pitted scars which may perfectly map out the dermatome. The
severity of the skin and periocular involvement may be so
severe as to resemble a bacterial orbital cellulitis with a notable
contralateral sterile cellulitis in the adnexae of the contralateral
eye as well. The acute inflammatory period lasts 814 days,
with the lesions considered infectious until they have scabbed
over. The deep skin ulceration may take many weeks to heal
and result in either little scarring or the equivalent of thirddegree burns with signicant loss and scarring of tissues
(Fig. 49.24).5,351,400,404,406
Dermal scarring is frequently pigmented in the early phase,
after which time it becomes very pale and white or silvery in
advanced cases. These scarred areas are anesthetic to pinprick,
although paradoxically the area is often hyperesthetic. This
increased sensitivity to tactile stimulus often causes the patient
difculty in contact with clothing or care of the skin or hair.
It is also of note that HSV may closely mimic VZV dermatitis.
In one case report, PCR was used to prove that the zoster
dermatitis was, in fact due to HSV.452
or papillary hypertrophy with or without regional adenopathy,

and rarely severe necrotizing membranous inflammation. In
one case of a 2-year-old child who had been vaccinated against
VZ, the presenting nding was a painful and diffuse subconjunctival hemorrhage that appeared before any of the classic
signs of the illness occurred.453 The episcleritis is often sectoral
and may be flat or nodular. Similarly, the scleritis tends to be
focal and may involve several areas, particularly in the
perilimbal region, either as a flat or a nodular process
(Fig. 49.25). The episcleritis or scleritis may occur during the
acute disease or several months after the cutaneous eruption
has cleared. Episcleritis, in particular, tends to recur. As the
scleritis resolves scleral thinning is frequently noted.
Occasionally, patients may develop a striking complication of
HZO characterized by 360o of perilimbal vasculitis, which
results in anterior segment ischemic necrosis. This may occur
months after the initial acute disease (Fig. 49.26). This virus
should be considered in any patient presenting with episcleritis,
Conjunctivitis, episcleritis, and scleritis Conjunctival inflammation is extremely common and characterized by watery
hyperemia occasionally with petechial hemorrhages, follicular
SECTION 6
FIGURE 49.25. Acute herpes zoster nodular scleritis with onset

several months after an acute attack of HZO. Scleritis responded to
mild topical steroid and oral ibuprofen and resolved to leave moderate
focal scleral thinning.
a
666
FIGURE 49.24. (a) Acute herpes zoster ophthalmicus (HZO) with

crops of fresh and crusted vesicles at different stages of evolution.
(b) Same patient 5 weeks later showing residual ptosis and partial
third nerve palsy apparent on attempted upgaze. Palsy resolved
completely over a 1-month period.
FIGURE 49.26. Acute severe HZO 360 occlusive limbal vasculitis.

Anterior ischemic necrosis resulted but responded to intensive topical
and systemic steroids.
Dendritic keratitis Nearly two-thirds of HZO patients develop

a keratitis often associated with marked decrease in corneal
sensation due either to signicant corneal damage or to the
necrotic ganglionitis, damage to the mesancephalic nucleus in
the brainstem, or local corneal deneravation.431,173,176,422 The
keratitis may precede or follow the neuralgia or skin lesions by
several days and may assume a variety of forms (Table 49.5).
Most common are the infectious forms of epithelial keratitis.
This may present as a ne or coarse punctate epithelial keratitis
with or without stromal edema and may give the cornea a
ground-glass appearance. There may be group vesicle formation, dendritic in pattern, which appears to be layered on the
corneal surface and may be easily mistaken for HSV
keratitis.452,455 These vesicles may be differentiated from HSV
dendrites in that VZV dendrites lack the rounded terminal
bulbs at the end of the branches, and when they are wiped from
the corneal epithelium they tend to leave behind a layer of
intact epithelium rather than the full-thickness ulcers noted
with HSV.359,456,457
Zaal et al reported a longitudinal study on VZV DNA on the
conjunctival and corneal surfaces associated with acute HZO
of less than 7 days duration.458 At entrance into the study, 19 of
the 21 patients were PCR positive for VZV DNA; six had no
ocular inflammation. All were treated with 1 g VCV tid for 10
days. Continued weekly testing revealed that DNA could be
detected from 234 days after rash onset in different study
patients with duration of DNA presence being longer with age
over 66 years. The authors concluded that VZV DNA shedding
is highly variable, age-dependent, and probably related to host
immune status.
Zoster has been isolated from these dendritic lesions by
Pavan-Langston and McCulley and demonstrated by immunofluorescence by Uchida and associates (Fig. 49.27).359,360 The
former investigators described lesions as medusa-like in
pattern, gray, and linear with tapering ends, and appearing to be
painted on the surface of the cornea. These lesions cleared
rapidly on either idoxuridine (IDU) or steroid therapy alone,
leaving no or mild anterior stromal nebulae. Piebenga and
Laibson describe similar lesions in HZO patients, noting the
supercial plaque-like formations without terminal bulbs,
which stained poorly with fluorescein.457
Liesegang has reported pseudodendrites in HZO patients,
noting lesions similar to those described earlier, which appeared
at 215 days after the onset of acute illness.424 The lesions were
usually peripheral, stained only moderately with rose bengal
or fluorescein, were broader and more plaque-like than HSV
dendrites without central ulceration, and were frequently
stellate. He too was able to culture VZV from four of nine
patients. Cytologic study of corneal scrapings from these lesions
revealed multinucleated giant cells and intranuclear inclusions.
Topical antiviral agents were not used, and steroids were
ineffective. The lesions, as in all other reports, were self-limited
and cleared within a few days, leaving behind mild anterior
stromal inltrates in 52% of patients (Fig. 49.28).
Delayed pseudodendrites (Liesegangs delayed corneal
mucous plaques appear in about 13% of patients, most often
TABLE 49.5. HZO Treatment Summary

1. Acute HZO: If possible start antiviral < 72 h p onset of rash
onset or continued rash activity. All three drugs = for acute
HZO. Famvir + Valtrex best re PHN
A. Antivirals of choice
1. Famvir 500 mg po or Valtrex 1 g po tid 7 d
2. Second line = acyclovir (Zovirax) 800 mg po 5 id 7 d
3. Antivirals are not indicated after acute HZO has passed
even with steroid use except possibly for recurrent HZO
dendrites
4. Antibiotic (e.g., Polysporin ung, HS, or quinolone bid)
PRN keratitis
B. Pain control and PHN inhibition: Tricyclic antidepressant
(TCA). Start with antivirals or asap.
1. Nortriptylene or desipramine 2575 mg qHS acute
disease. Taper up over days to best tolerated dose.
2. Non-narcotic and narcotic (oxycodone-CR) analgesics
PRN. See below
C. Immune keratitis and/or iritis: Topical steroid, antibiotic.
Dilation PRN. Dendritic keratitis: no rx or topical or po
antivirals. +/ effect
D. Neurotrophic ulcer: Lubricate. lateral or Botox tarsorrhaphy.
T-SCL (Kontur, Permalens, Acuvue, Focus Night & Day) with
antibiotic gtts.** Mild steroid PRN. Glue (Dermabond
Epidermglu) with Plano T SCL PRN melt. Amniotic memb.
Tx, patch graft, keratoplasty (failure risk), Boston
keratoprosthesis
2. Acute or chronic pain (PHN): Combined therapy additive or
synergistic.
A. TCA as above
B. Neurontin 12003600 mg/d in divided doses (600 mg pills)
months. Taper up over 23 w until relief or intolerant, or
Lyrica 300600 mg/d
C. Lidocaine 5% oint. or emla cream (lidocaine-prilocaine)
q 612 h. Or lidocaine skin patches 12 h on, 12 h off
D. Oxycodone-CR 1040 mg po bid PRN (slow release opioid)
E. Benadryl 2550 mg po HS (itch)
F. Marcaine supra + infraorbital + trochlear nerve block. Botox
A map injections
3. Late keratouveitis: As under acute but antivirals not indicated
CHAPTER 49
scleritis, or iritis. A 9-year-old child presented with scleritis,

marginal keratitis, mild glaucoma, marked iritis, and rare skin
vesicles 3 years after varicella vaccination. PCR study of DNA
from vesicle fluid taken revealed wild-type varicella zoster virus
(VZV) DNA, not the Oka strain of the vaccine.454 The
episcleritis, scleritis, and vasculitis all appear to be variably
responsive to topical or, if necessary, systemic steroids. The
nodular scleritis may also respond dramatically to long-term
therapy with nonsteroidal antiinflammatory agents such as
ibuprofen 300 mg po tid.103,406,446
FIGURE 49.27. Acute HZO with zoster infectious viral dendritiform

lesions appearing as transient, positive-staining branching gures
without terminal bulbs. Lesions resolved without therapy.
667
SECTION 6
FIGURE 49.28. Delayed zoster pseudodendrites with onset 11 months

after acure herpes zoster ophthalmicus. Polymerase chain reaction
assay revealed varicella-zoster virus DMA in the lesions.
812 weeks after the acute event but years later in some
cases.424,456 They cause a foreign body sensation and are
elevated, coarse, gray-white, swollen epithelial cells piled in
plaques or a dendritiform shape on the surface of the cornea.
They are both migratory and transitory and are usually
associated with a neurotrophic keratitis (75%) or previous
corneal inflammation (100%). The tear lm is unstable. Unlike
studies by Pavan-Langston and McCulley these plaques were
culture-negative, and cytologic study showed ballooning
degeneration of the epithelial cells with occasional giant cells.
The plaques reportedly did not respond to antiviral, steroid, soft
contact lens, acetylcysteine, or lubrication therapy but were felt
to be self-limited as in the case of the early pseudodendrite. In
subsequent work Pavan-Langston et al have reported PCR
detection of VZV-DNA in several of these late-onset pseudodendrites in immunocompetent and immunocompromised
patients.459 Electron microscopy of a corneal button taken from
a lymphoma patient with pseudodendrites revealed numerous
mature and immature viral particles in the epithelial basal cells
and myeloid bodies in the keratocytes. Systemic ACV and
topical TFT had little to no effect but there was a resolution
on topical 3% vidarabine ointment 5id.362,459 Engstrom and
Holland have reported a case of chronic zoster dendriform
lesions in an HIV patient. Immunofluorescence revealed VZV
and therapy with TFT and oral ACV failed.460 There was a
therapeutic response to topical ara-A ointment, however.
It is the authors experience that response to therapy is
variable and unpredictable. Full-dose oral antivirals or topical
vidarabine or TFT may or may not succeed. If one fails, the
physician may succeed with another or the dendrites often
simply self-resolve.461
668
Stromal keratitis: disciform, interstitial keratitis, immune rings,

limbal vasculitis Stromal forms of VZV disease are clinically
difcult if not impossible to differentiate from stromal HSV.
While the data available on zoster immune keratitis is nowhere
near that of HSV, because of the similarity of the infectious
agents, it seems reasonable to assume that the mechanisms of
disease (with the exception of acute and postherpetic neuralgia)
are the same or closely related.462 Anterior stromal inltrates,
single or multiple in the anterior stroma, usually appear
between the second and the third weeks after onset of acute
disease if there has been epithelial infectious keratitis. These
probably represent an immune response to soluble viral antigen
diffusing into the anterior stroma or possibly viral cytotoxicity.
Stromal keratitis with or without endotheliitis is seen in

about one-third of patients and may have its onset immediately
with acute disease or several weeks to months later. Frequently,
focal or diffuse keratic precipitates, folds in Descemets
membrane, and diffuse or focal stromal edema appear. IOP is
elevated in 33% of these patients, presumably due to an
associated trabeculitis with inflammatory cells blocking the
trabecular meshwork in a mechanism similar to that discussed
earlier in this chapter. A severe vasculitis with hypopyon or
hyphema and pars plicata ischemia may develop, leading to
anterior segment ischemia.424,463 The endotheliitis may result
in signicant loss of endothelial cells with permanent corneal
decompensation.464,465
Disciform immune keratitis is similar in appearance to HSV
focal or diffuse edema with or without endothelitis. It appears
most commonly at 34 months after the acute event but may
occur with the rash and a keratouveitis. There may be fullthickness stromal edema and associated interstitial keratitis,
Wessley rings, or limbal vasculitis although any of these keratitis
forms may occur independently. If there is an associated interstitial keratitis, there is increased chance of deep neovascularization with lipid deposition and brovascular scarring. The
response of this lymphocytic/plasma cell/macrophage immune
reaction to steroids is rapid to moderate initially, but tapering
of steroids is frequently slow with many patients requiring
minimal daily doses to prevent rebound immune disease
(Fig. 49.29).5,439,466
Maudgal has reported varicella-zoster virus in the human
corneal endothelium and studies by Wenkel et al have revealed
both chronic viral and gene persistance in zoster corneas for as
long as 10 years after acute disease. Nine eyes and four corneal
buttons surgically obtained from 13 patients with HZO were
examined at different timepoints after clinical onset of HZO
(range, 1 day19 years; median, 36 months). Histopathologic
changes associated with HZO included corneal stromal vascularization (11 of 13), granulomatous reaction to Descemets
membrane (eight of 13), fusiform-shaped ciliary scarring (ve of
nine), optic neuritis (four of nine), perineuritis (eight of nine)
and perivasculitis (eight of nine) of the long posterior ciliary
nerves and arteries. VZV antigen was detected in two acute
HZO patients 1 and 7 days after onset of disease, and VZVDNA was identied in seven patients up to 10 years after onset
of HZO in corneal epithelial cells (two of 13), corneal stroma
(ve of 13), inflammatory inltrate of the anterior chamber (one
of nine), episclera (two of nine), posterior ciliary nerves (one of
nine) and arteries (ve of nine), optic nerve (ve of nine), and
adjacent leptomeninges (two of nine).427,438,467
In a study of 14 keratectomy specimens from patients with
a history of HZO,7 (50%) were positive for VZV DNA. The
positive ndings in the specimens correlated with the clinical
ndings of uveitis (3/3) and chronic stromal keratitis (4/4).
Patients with stromal scarring, granulomatous keratitis, and
neurotrophic ulcers had negative ndings. The largest interval
between the initial appearance and detection of viral DNA was
51 years. The authors concluded that VZV DNA is not
detectable in the cornea in every patient and at every stage of
zoster keratitis possibly due to the low number of VZV particles
present in the cornea or the lack of viral DNA in the
keratocytes. They could not conclude whether VZV-related
keratopathy is caused by an immunologic response to a viral
antigen, the viable virus itself, or both.468
As with iritis (see below), disciform keratitis may occur as a
manifestation of zoster sine herpete (no rash). VZV DNA has
been detected in the aqueous of such a patient while HSV and
CMV assays were negative.469 The corneal edema slowly
resolved while the patient was treated with famciclovir 500 mg
po tid for 2 months.
b
FIGURE 49.29. (a) Acute HZO focal interstitial keratitis with
neovascularization. (b) Same eye 10 years later with resolution of
interstitial keratitis leaving residual crystalline (?cholesterol) deposits
in the anterior stroma.
Peripheral ulcerative keratitis is commonly associated with

collagen vascular disease. It has been reported, however, by
Liesegang in 7% of his HZO patients.423 Within the rst
5 months of the acute disease the peripheral cornea may
develop an acute stromal edema with an overlying crescentshaped ulcer having smooth edges and a gray-white base. There
may be a mild uveitis, and the corneal sensation is almost
invariably markedly decreased. Seventy-ve percent of these
ulcers are thin and vascularized and 25% perforate if untreated.
It is felt that these changes are due to a delayed Arthus reaction
similar to that of limbal vasculitis or to a cell-mediated immune
disease of the limbal vessels.470,471 Fortunately, these patients
are responsive to topical steroid therapy but some may require
tissue adhesive and a therapeutic soft lens to enhance healing.
Neurotrophic keratopathy Neurotrophic keratitis is seen in
nearly 50% of HZO patients, the result of corneal anesthesia
secondary to the VZV trigeminal ganglionitis and
mesancephalic/pons nuclear damage, aqueous tear deciency

due to loss of the nasolacrimal reflex, and decreased blink
reflex.35,176,186,402,424,431,443,472 The mechanisms involved are
reviewed in detail earlier in the section on Neurotrophic
Keratopathy and Trophic Ulceration (Fig. 49.8).
Corneal anesthesia may be evident at the time of onset of
HZO or may develop over weeks to years. As shown in and
discussed earlier in the section on HSV Neurotrophic
Keratopathy, the neurochemicals and consequent devastation to
neuronal cellular function and existence present a challenge to
the treating physician. Confocal microscopy studies showing
damage to or complete loss of the corneal subbasal nerve plexus
in both HSV and HZO demonstrate physically in the living
patient why the corneas become anesthetic and unable to
heal.176,184 About 25% of all HZO patients will develop clinical
signs of neurotrophic keratitis due to permanent corneal anesthesia. Early clinical ndings include a dull or irregular corneal
surface with mild coarse punctate epithelial keratitis.399,400,423
This is followed by a gray diffuse epithelial haze or edema with
ne intraepithelial vesicles. As noted, the tear lm is highly
unstable and blink frequency is reduced in these anesthetic
eyes, thus further aggravating the condition. As the corneal
epithelium becomes progressively unhealthy, oval epithelial
neurotrophic defects frequently develop in the palpebral ssure
or lower corneal area with subsequent melting and corneal
thinning.173 This complication of neurotrophic keratitis is
frequently seen in patients who had previous keratouveitis
(80%) and in Liesegangs study there was an accompanying
exposure pattern in 40% of patients so affected (Figs 49.30
and 49.31).423
Treatment of the anesthetic eye with unhealthy epithelium is
reviewed and summarized in detail under HSV Neurotrophic
Keratopathy. Exposure keratitis is frequently seen in patients
with neurotrophic keratitis and may develop anywhere from the
period of the acute illness to several years later. It is most
commonly seen, however, within a few months after the onset
of disease. There may be cicatricial retraction of the upper lid,
thus making blink ineffective and complete lid closure
impossible. The lower lid may be similarly involved with
additional ndings in either or both lids of thickening, irregular
margins, meibomian gland dysfunction, trichiasis, punctal
occlusion, ectropion, or entropion.
Warm wet compresses applied to the ulcerated skin may help
reduce local cellulitis and secondary infection with consequent
decreased scarring of the periocular skin and lids. In severe
cases, cicatricial retraction of the lids may require full-thickness
skin grafts or tarsorrhaphy for the protection of the ocular
structures.381,400,473475
Iridocyclitis The anterior uveal tract is second only to the
cornea in frequency of involvement in HZO. In Womack and
Liesegangs studies uveitis was seen in 43% of their patients, a
gure in general agreement with other studies, as were the
characteristic ndings of vascular dilatation, posterior synechiae,
striate keratopathy, pigment iris atrophy, sectoral pigment iris
atrophy, and sphincter damage.402 Fluorescein angiography
revealed occlusion of iris vessels at the sites of atrophy. This
differs from the iris atrophy seen with HSV, which causes
sharply dened borders and scalloped margins with the iris
arterioles patent in the involved areas. HZO may cause sectoral
iris atrophy in 1725% of patients.402,445477 Involvement may
occur early or many months or years after the onset of acute
disease and independent of corneal activity. It is felt to be an
immune reaction probably due to direct invasion of the uveal
structure by infectious VZV. Late-onset uveitis may be due to
immune reaction to the antigenic residua of this virus. VZV
DNA has been detected in the aqueous.449
CHAPTER 49
669
FIGURE 49.30. HZO indolent trophic ulcer with thickened edges

surrounding a damaged basement membrane in an anesthetic cornea.
SECTION 6
FIGURE 49.31. Chronic herpes zoster keratitis with active necrotic

interstitial keratitis, soft neurotrophic ulceration, and supercial
vascular pannus.
670
Clinically, the iridocyclitis may be characterized simply by

photophobia in a white and quiet eye with just a few cells and
flare in the anterior chamber. More severe disease will involve
ciliary flush, miosis, deep ocular pain, visual decrease, keratic
precipitates, iris edema and hyperemia, and anterior and
posterior synechiae secondary to a brinous exudate. In
comparing the clinical courses and outcome of 40 HSV and 24
HZO uveitis patients Miserocchi et al noted that the HSV
tended to be remitting and recurrent and chronic in VZV
patients (P = 0.046). Secondary glaucoma was the most
frequent ocular complication in both groups (54% HSV, 38%
VZV) and 25% of VZV patients developed posterior pole
complications (cystoid macular edema, epiretinal membrane,
papillitis, retinal brosis, and detachment) compared with 8%
of HSV patients (P = 0.069). Treatment modalities were
generally similar in the two groups, although HSV patients
required periocular and systemic steroids more frequently (60%
versus 25%; P = 0.01) but the percentage of legally blind eyes
FIGURE 49.32. Histopathology of herpes zoster trabeculitis showing

extensive leukocytic clogging of the trabecular meshwork resulting in
secondary glaucoma which is responsive to topical steroid therapy.
at the end of follow-up was essentially the same (HSV, 20%;

VZV, 21%).478
The characteristic nding in HZO uveitis histopathologically
is the plasma cell/lymphocyte inltration of the posterior ciliary
nerves and vessels particularly in uveal structures. Zoster iritis
differs from HSV iritis in that the former is chiefly a vasculitis,
whereas the latter is primarily a diffuse lymphocytic inltrate of
iris stroma. The iris focal or sector atrophy seen in 1725% of
HZO is the result of the localized ischemic necrosis similar to
that seen after excessive diathermy and muscle detachment in
retinal surgery.402,445,477 Acute or late-onset uveitis may be due
to immune reaction to the antigenic residua of this virus or to
new virus production.440 Additional studies indicated that absence
or poor VZV delayed hypersensitivity response correlates with
increased severity of VZV uveitis and may also prove to be a
useful diagnostic test in zoster sine herpete.479 Hypopyon,
hyphema, glaucoma, and ultimate phthisis bulbi may all result
from this severe zoster iridovasculitis.351,445
As with corneal disease this immune inflammatory reaction
is sensitive to therapy with topical and, if needed, short-term
systemic steroids. The usual gradual tapering regimen must be
used once the disease is under control to avoid rebound intraocular reaction. Some patients may need to be indenitely
on the equivalent of 0.125% prednisolone (or rimexolone, or
lotoprednol) daily or even every 23 days to prevent recurrence.
Glaucoma and hypotony Marked elevation of IOP may be seen
in ~10% of patients, especially those with peripheral endotheliitis with an associated trabeculitis, with isolated trabeculitis and no corneal involvement, or secondary to cicatricial
closure of the angle. Endotheliitis may involve the peripheral
cornea and trabecular meshwork (TM) together or inflammation may be conned just to the TM (trabeculitis) within the
angle of the eye. An accompanying secondary glaucoma is likely
due to the swelling and obstruction of the trabecular meshwork
by inflammatory debris similar to that seen with HSV
(Fig. 49.32).264 Occasionally, a marked decrease in IOP due to
necrosis of the pars plicata of the ciliary body, with subsequent
decreased aqueous production, is more than counterbalanced by
impairment of outflow facility by clogging of the trabecular

meshwork with inflammatory cellular debris. Depending on the
balance between decreased aqueous production and decreased
outflow, IOP may be low, normal, or elevated. Because this is an
inflammatory glaucoma, miotics should not be used. Topical
beta-blockers, prostaglandin analogs, systemic or topical
carbonic anhydrase inhibitors may all be used effectively to treat
both the acute and the chronic aspects of this secondary
glaucoma.
Zoster sine herpete (ZSH)
Muscle palsies and myositis

Palsies of nerves III, IV, and VI are not frequent and, with the
exception of an almost invariable residual partial ptosis, these
almost invariably resolve independent of initial severity
(Fig. 49.33).406,485487 Per Archambault et al in HZO muscle
paresis the third nerve appears to be the most commonly
affected and the fourth nerve the least.488 They reported six
patients with HZO in whom spontaneous recovery occurred in
four patients. Shin et al have reported external ophthalmoplegia
and exophthalmos in a case of HZO. Intramuscular injections
of dexamethasone were given for 10 days, followed by prednisone, 15 mg for 2 weeks, and 10 mg for 2 weeks. The patient
was fully recovered by 6 months after the onset of the cutaneous
lesion.489 It is the authors experience, however, that such
aggressive therapy is unnecessary. In her experience either no
treatment (other than the antivirals and management of other
anterior segment disease) or a 2 week course of prednisone,
20 mg po tid tapered to 20 mg q AM, is sufcient even with
total ophthalmoplegia, as this is essentially a self-resolving
paralysis.
Other palsies include internuclear ophthalmoplegia, isolated
iris sphincter paralysis, and Horners syndrome.487,490494 It is
felt that the extraocular muscle palsies and orbital edema seen
FIGURE 49.33. Recent herpes zoster ophthalmicus with residual total

third, fourth, and sixth nerve paralysis frsulting in a frozen globe. The
paralysis resolved spontaneously over an 8 month period.
in HZO may be the result of perineuritis and perivasculitis

associated with the generalized orbit inflammation.439,488,492
Fortunately, the vast majority of muscle palsies resolve
spontaneously.
Myositis is another cause of extraocular muscle dysfunction
(EOM) in HZO. EOM palsies in these patients are traditionally
interpreted as diseases of III, IV, or VI cranial nerves. Orbital
myositis is diagnosed only rarely. In separate case reports, a
patient with HZO and external ophthalmoplegia was found to
have ocular myositis demonstrated by MR imaging.495,496
Treatment with acyclovir and cortisone resulted in a rapid
improvement of the ophthalmoplegia. Ocular myositis, then,
is an important differential diagnosis to inflammatory involvement of the cranial nerves III, IV, and VI.
Posterior pole involvement

While discussion of HZ posterior pole involvement is beyond
the scope of this chapter, the list includes optic neuritis, central
retinal vein occlusion, central retinal artery occlusion,
necrotizing retinitis (acute retinal necrosis or progressive outer
retinal necrosis), delayed thrombophlebitis, optic neuropathy,
and localized arteritis with or without exudates.497,498
Acute and postherpetic neuralgia (PHN) and itch (PHI)

In a 1957 study of 916 zoster patients seen between the years
1935 and 1949, deMoragas and Kierland reported an incidence
of 12.5% acute neuralgia in those patients less than 20 years of
age, ~40% in those in the third and fourth decades of life, and
only 20% in those in the sixth and seventh decades.399 In
contrast, within this same population the incidence of chronic
postherpetic neuralgic (PHN) pain lasting more than 1 year fell
to less than 4% in the under 20 years age group, and to about
10% in those patients in the third and fourth decades of life, but
rose to nearly 50% in those patients in the sixth and seventh
decades of life. This concurs with Womack and Liesegangs
nding of postherpetic neuralgia persisting in 17% of their
patients, with the most severe cases being in those patients in
the sixth through eighth decades of life. Four of their patients
had been treated with prednisone in the course of disease.402
Ragozzino and co-workers found a 9.3% incidence also in the
older patient population.451
Scott et als study on shingles and PHN in 165 East
Londoners revealed that the prevalence of PHN was 30% at
CHAPTER 49
Iritis, disciform keratitis, and facial palsy (Bells palsy) may all
be due to zoster sine herpete (ZSH). This entity is dened as
reactivated VZV which causes only neurologic symptoms such
as dermatomal neuralgia or neuropathy, and, on occasion, ocular
inflammation. There is, by denition, no rash.419,420,449,469
Any unexplained acute granulomatous iritis, with or without
elevated IOP, should be suspect for HZO. In studies on nine
suspected ZSH patients with iritis and secondary glaucoma, the
aqueous humor was positive for VZV-DNA in the early stages
of disease and pigmented KPs and typical sectoral iris atrophy
were residua.480 The aqueous was also positive for VZV-DNA in
a 65-year-old man with bilateral granulomatous iritis,
secondary glaucoma unresponsive to steroid therapy, and no
rash.481 The inflammatory disease responded promptly to
acyclovir po but there was residual iris sector atrophy. Facial
palsy (Bells palsy) may also be due to ZSH.482 Thirteen acute
facial palsy patients with PCR positive for VZV-DNA were
treated within 7 days of onset using acyclovirprednisone
therapy and all recovered completely.
Noncontact in vivo photomicrography on seven patients with
ZSH revealed that all patients had corneal epithelial changes at
presentation and went on to develop new ones over 2 weeks.
The smallest lesions noted were 1025 mm in diameter and
large foci were 100200 mm. Three of three corneas tested were
PCR positive for VZV-DNA.483 This was also noted in a similar
study on patients with overt HZO.484 In the latter study, two
patients developed pseudodendrites and some developed white
plaques but there were no ulcers. Disciform keratitis may also
occur as a manifestation of zoster sine herpete. VZV-DNA had
been detected in the aqueous of such a patient while HSV and
CMV assays were native.469
Treatment for ZSH is similar to that of the complications
seen with HZO including oral antivirals and PHN inhibitors.
671
SECTION 6
672
6 weeks, 27% at 12 weeks, 15.9% at 6 months, and 9% at 1 year.

Age and severity of acute pain were signicant risk factors for
persistence of pain beyond 3 months. Viremia was detected at
presentation in 66% of patients and was also signicantly
associated with the persistence of pain at 6 months or beyond.
Antiviral agents were administered to only 50% of those at
highest risk of PHN because of presentation longer than 72 h
after the onset of rash. Few patients were given the more potent
prodrugs, valaciclovir and famciclovir. The authors concluded
that treatment of acute shingles in this community-based study
was suboptimal in 50% of cases and that more accurate
prediction of which subset of elderly patients are most at risk of
PHN may enable targeted prescribing of the more potent drugs,
valaciclovir and famciclovir, to those patients most likely to
benet.499
The risk of PHN in those over 60 years is emphasized in a
study by Bowsher et al on 1071 men (n = 534) and women
(n = 537) with a median age of 80 years.410 Twenty-four percent
had had shingles at a median age of 60 for both sexes and 15%
of those who had had shingles, two-thirds of whom were
female, developed PHN, dened as pain persisting for more
than 3 months. In 17 of 62 PHN patients, it was ongoing for
years. New independent risk factors for PHN were (1) female
gender, and (2) living alone at the time of HZ acquisition (P =
0.009). Extrapolating the prevalence gures to the whole UK
population, of whom 9.28 million were over 64 in 1992, it was
conservatively estimated that at any one time, some 200 000
people in the UK had PHN. If we extrapolate these gures to the
United States, using 100 000 000 people over the age of
60 years, at any given time 2 155 170 Americans have PHN.
PHN symptomatology includes constant or intermittent
aching or burning, sudden lancinating pain, allodynia (nonpainful stimuli perceived as painful), and a constant or
intermittent itching in the involved area (PHI). Risk factors for
PHN and local sensory denervation include older age (5080
years), greater prodromal and acute disease pain, marked rash
severity, rash on head, face, or neck (particularly ophthalmic),
clinical depression, adverse psychosocial factors, failure to
treat with valciclovir or famciclovir, and viremia.499504 In their
studies on hallmarks of PHN in 113 acute HZ patients Haanpa
et al noted that when rst seen, 87 (77%) patients reported
ongoing pain and 48/107 (45%) had allodynia (pain from
nonpainful stimuli). Twenty-eight (25%) patients had pain at 3
months (and were considered to have developed PHN), while 14
(12%) patients had pain at 6 months. Mechanical allodynia and
pinprick hypesthesia were strongly associated with the
development of PHN and were considered risk factors by these
authors for predicting PHN. By contrast, lack of allodynia
during the acute stages of HZ predicted uneventful recovery by
3 months.500
The incidence of postherpetic neuralgia is higher in the HZO
population than in those persons with other forms of zoster
especially those over the age of 50 years.385,386,451,505 Because of
its different clinical manifestations long-term neuralgia has
been dened as that which lasts longer than six months or that
which persists for more than a month beyond the usual healing
of an inciting lesion.87,505507 Chronic pain is associated with
quiescent behavior, which is often overlooked by the treating
physician. Such behavior includes sleep disturbance, lassitude,
anorexia, weight loss, constipation, and in place of anxiety,
depression. By Watson et als observations of 100 patients with
HZO, ~10 will have severe pain for 1 month, 5 will have severe
pain for 3 months, and two to three will be in severe pain for
more than 1 year after the acute episode (see Table 49.5).
There are many proposed mechanisms of zoster-induced
pain, the majority of which appear valid and contribute to
the many variations in the forms of pain. The acute pain
experienced by patients during the early phases of HZO is

attributed, in part, to swelling of the trigeminal nerve (fth
nerve) in association with a lymphocytic inltrate as well as the
pain of inflammatory reaction in and around the eye itself.508
This pain is often accompanied by sympathetic hyperactivity
such as tachypnea, tachycardia, diaphoresis, mydriasis, and an
effect characterized by anxiety.
Bowsher et al have reported that there is increasing evidence
that the majority of acute trigemial neuralgia (TGN) cases
result from vascular compression of the fth nerve at its point
of entry into the pons as surgical decompression will relieve the
pain and restore the sensory decit.509 They correctly felt that
TGN is (at least in part) a disorder of central processing, the
term being taken to include the oligodendroglial-sheathed
proximal segment of the nerve but erroneously stated that no
anatomical abnormalities of the (peripheral) trigeminal nerve
have demonstrated implying the pain is entirely central in
origin.
As noted by Feller et al the peripheral nerve injury that occurs
during the acute phase of herpes zoster (HZ) leads to an
abnormal tonic impulse discharge from primary nociceptive
afferent neurons (peripheral pain-transmitting neurons) which
induce slow temporal summation.510 This wind-up phenomenon is responsible for continuous partial depolarisation of
second-order neurons with increased spontaneous impulse
discharge and expanded receptive elds within the dorsal horn
nociceptive neurons.The abnormal central processing results in
neuropathic pain, characterized by spontaneous pain, hyperalgesia and allodynia which is typical of PHN. In addition, tonic
input from nonnociceptive AB afferent neurons, supported by
sympathetic efferent activity, contribute to the development and
maintenance of neuropathic pain in general, and a burning
sensation in particular.
Both central (see the section on Pathogenesis and
Histopathology of Ocular and Central Nervous System (CNS)
HZ) and peripheral nerve abnormalities in HZ and HZO have
been well demonstrated in PHN and PHI. One proposed
mechanism for the nociceptor hyperactivity (peripheral nerve
pain transmitter neurons) is that the period of acute disease
initiates peripheral nociceptor-evoked CNS hyperexcitability
and axonal lesions that may induce the growth of ectopically
discharging nociceptor nerve endings.476,511At autopsy, patients
with PHN have histopathologic lesions both in the peripheral
nerves and in the dorsal horn and spinal cord, changes not seen
in VZV patients who recover without PHN (Fig. 49.3).443
Oaklander et al have reported a 39-year-old woman with
severe postherpetic itch (PHI), but no postherpetic neuralgia,
after ophthalmic zoster. Over the course of a year, she painlessly
scratched through her frontal skull into her brain. Quantitation
of epidermal neurites in immunolabeled skin biopsies demonstrated loss of 96% of epidermal innervation in the area of PHI.
Quantitative sensory testing revealed loss of most sensory
modalities except for itch associated with this loss of peripheral
sensory neurons. Proposed mechanisms included spontaneous
hyperactivity of hypo-afferented central itch-specic neurons,
selective preservation of peripheral itch-bers from neighboring
unaffected dermatomes, and/or an imbalance between excitation and inhibition of second-order sensory neurons.512 In a
separate study on the post-zoster density of remaining nerve
endings in human skin with and without PHN she found that
17 of 19 subjects without PHN had more than 670 neurites/
mm2 skin surface area and 18 of 19 subjects with PHN had 640
or fewer neurites/mm2. She concluded that PHN may be a
phantom-skin pain associated with loss of nociceptors (pain
transmission neurons), and that absence of pain after shingles
may require the preservation of a minimum density of primary
nociceptive neurons.513
An autopsy done on a patient with severe ophthalmic PHN

for more than 10 years revealed, unlike Reske-Nielson (as discussed before), no CNS morphologic abnormalities.431,443,514 There
was, however, severe pathologic damage to the ophthalmic and
supraorbital nerves including brosis, demyelination, loss of myelinated bers, and shift in ber diameter to pain-transmitting
small-diameter neurons. This suggested another mechanism
for PHN and that was a dying back form of pathology expressed
primarily in the periphery. As in Oaklanders study, there was
also pathologic damage to the contralateral ophthalmic and
supraorbital nerves, although not as severe. This side had no
PHN.514,515 In their review of the heterogeneous pattern of
sensory dyfunction with PHN Pappagallo et al concluded that
despite a common cause, the patterns of sensory abnormalities
differ between groups with facial or truncal PHN and between
groups with recent or more chronic PHN. This indicated that
the relative contributions of peripheral and central mechanisms
to the pathophysiology of pain differ among subjects and may
vary over the course of PHN.516
In an unusual case study done over 11 years, Gilden et al
reported an immunocompetent elderly woman with PHN.
Blood mononuclear cells (MNCs) contained varicella-zoster
virus (VZV) DNA on two consecutive occasions but random
testing after treatment with famciclovir to relieve pain did not
detect VZV-DNA. The patient started and stopped famciclovir
ve times. Pain always recurred within 1 week, blood MNCs
contained VZV genome, and there was increased cell-mediated
immunity to VZV on all ve occasions. It was felt that chronic
VZV ganglionitis-induced PHN best explained all of these
phenomena but that recurrent ganglionitis was not the only
mechanism of PHN. It has also been the authors experience
in ve cases that full-dose treatment for 12 weeks using
famciclovir or valaciclovir resolved recurrent PHN.517,518 Vafai
et al had also noted that VZV-specic proteins were found in
the monocytes of PHN patients months or years after the acute
disease indicating persistence, re-activation, and expression
of VZV in the chronic pain patients.519
The pathophysiology of postherpetic neuralgia in all its many
forms, then, is not fully understood. The brosis occurs in the
peripheral nerve and there is markedly increased sensitivity of
the skin nociceptors resulting in the spontaneous ring of pain
bers or the induction of pain by a mild stimulus which would
not cause pain in a normal situation (allodynia). There is also
brosis in the dorsal root ganglion, and the structure of the
dorsal horn of the spinal cord is disordered without atrophy of
the dorsal horn as well as damage to decrease or increase in
peripheral small bers and reduction in large bers. The small
bers are excitory, transmit pain, and contain substance P, a
tachykinin. Recurrent ganglionitis with reappearance of VZVDNA is another reported mechanism. In brief, postherpetic
neuralgia appears to be the result of disordered ber input into
a diseased dorsal root ganglion and dorsal horn (for HZO the
TG) and on to the cerebral cortex as well as abnormally
heightened skin nociceptor sensitivity and, in some cases,
reactivated ganglionitis.431,472,500,502,514,517,518,520522
The importance of understanding and addressing early on the
prevention and management of PHN cannot be emphasized
enough. As Oster et al have noted, older persons (age >65 years)
with PHN experience long-standing, severe, and debilitating
pain and poor health-related quality of life.523 Dissatisfaction
with treatment is high because many physicians do not make
the effort to work out the optimal medication regimen to relieve
this very debilitating disease.
HZO in HIV/AIDS patients

HZO tends to be severe no matter what the cause of marked
immunosuppression. With the advent of the AIDS epidemic,
physicians were seeing increasing numbers of HZO patients in

the younger age groups than previously reported. In 1984, Cole
and co-workers reported four cases of HZO in four males with
AIDS ranging in age from 2641 years.524 Interestingly, only one
of these patients was treated with systemic antiviral therapy
(ara-A IV). All received topical steroids for ocular immune
reactions, and one received systemic prednisone therapy, yet all
recovered without immediate adverse consequences of the
HZO. Similarly, Seiff et al in 1988 reported two more cases of
HZO in AIDS patients 24 and 49 years of age, both of whom
were successfully treated with ACV IV.525 As the reactivation of
VZV in immunocompetent patients under the age of 50 years
is uncommon, it has become a guideline that zoster in patients
under age 40 who have risk factors by history should be
evaluated for the possible co-existence of HIV infection. In a
study on 399 HZ patients in India, 56 had associated HIV and
by 3 years follow-up, 17 more had developed HIV infection.408
Friedmann-Kein and colleagues noted that 8% of AIDS
patients with Kaposis sarcoma had a past history of herpes
zoster, an incidence seven times that of age-matched controls
(ages 2452).526 In a prospective study by Sandor and coworkers on herpes zoster patients under the age of 45 year, 75%
had AIDS risk factors and all tested positive for HIV
antibody.527,528 This indicates that any patient in a high-risk
group, a homosexual, an IV drug abuser, or persons with
hemophilia A, that oral or IV acyclovir or oral famciclovir
therapy should be initiated immediately along with highly
active anti-retroviral therapy (HAART). Valaciclovir has been
implicated in cases of thrombocytopenic purpura/hemolytic
uremic syndrome in severely ill HIV patients.391,529 It is not
approved for use in immunocompromised patients.
Studies on the influence of the introduction of HAART on
the incidence of HZ revealed that HZ was highest in the rst
6 months of treatment in patients at late stage of HIV infection
than at 12 months (17/100 person years (PY) versus 3/100
PY, P < 0.0001). Over a 9 year period 103 patients out of 716
developed HZ (5.67/100 person years). Baseline CD4+
lymphocyte count was the most signicant risk factor
associated with development of HZ and HZ was associated with
an increased risk for HIV progression, but not mortality.530
These ndings were conrmed and extended by Gebo et al.
They found that zoster infection rates have not changed during
the HAART era and that a signicant number of patients
develop complications, particularly PHN, which is unusual
given the young age of the patient population.531 Vanhems et al
have also reported that the incidence of herpes zoster is less
likely than other opportunistic infections to be reduced by
highly active antiretroviral therapy.532 Evans has reported that
since the advent of HAART the effects of HIV on the eye have
been less in countries where such treatment is available but
even in such situations ophthalmic zoster can occur at higher
CD4 cell counts and can still cause problems.533
HZO in the uncontrolled (untreated) HIV/AIDS patient is
characterized by marked and prolonged dermatitis, keratitis,
iritis, and PHN.400,460,524,531 HZO, as most forms of zoster
disease, is normally considered an infrequently recurring
illness. In the face of AIDS, however, recurrence may be more
frequent than previously seen. Litoff and Catalano reported
treating a 40-year-old HIV-positive woman with HZO with a
10 day course of IV ACV (4 g/day).534 There was complete
resolution of disease, but 3 weeks later the patient represented
with recurrence of her dendritic lesions in both eyes and
another vesicular eruption in the trigeminal dermatome.
Additionally, she had a marked zoster vasculitis and retinitis.
She underwent a repeat 10 day course of IV ACV (4 g/day) and
a 5 day course of IV methylprednisolone sodium succinate
(250 mg q6h). The ocular lesions gradually resolved, although
CHAPTER 49
673

visual acuity remained poor in the eye with severe retinitis. She
was discharged on maintenance therapy with oral ACV 800 mg
ve times daily and topical cycloplegia and fluorometholone
(FML). She remained asymptomatic in the 6 weeks of follow-up.
As in HSV, then, HZO in the face of AIDS-induced immunosuppression manifests a more difcult clinical entity both in
terms of management and in the risk of recurrent disease.
Long-term, low-dose acyclovir has been reported as successful
in preventing reactivation of VZV after hemopoietic stem cell
transplantation in 247 allogeneic recipients.535 These patients
are at extremely high risk of post-transplant antiviral prophylaxis with low-dose oral aciclovir 400 mg/day, was administered
until immunosuppression was discontinued and the CD4+ cell
count exceeded 200/mm3. Viral reactivation was successfully
suppressed by acyclovir prophylaxis, with only one case of
breakthrough infection. The data support previous ndings
that acyclovir prophylaxis prevents VZV reactivation, although
the long-term incidence is not affected as infection occurs once
prophylaxis is discontinued. Such infections, however, are mild
and localized.
Medical Treatment of Acute HZO and PHN

Antiviral drugs
SECTION 6
HZO is clinically unresponsive to the commercially available

topical antiviral medications despite susceptibility in vitro and
cannot compare to systemic antivirals such as acyclovir.440,536
Zaal et al have reported that at 3 months post onset of HZO
patients who received 3% topical ACV had longer durations of
periocular lesions and signicantly more visual loss compared
to the group receiving oral ACV, and that all patients put on
combined topical ACV and dexamethasone drops developed
chronic disease.440
674
Acyclovir Because of the high complication rate in HZO,

several studies have been conducted comparing ACV to placebo
or other antiviral therapy. ACV 800 mg po 5id for 710 d,
induces prompt resolution of rash and pain, induces more rapid
healing, reduces the duration of viral shedding, and the
duration of new vesicle formation. As important, there is also a
signicant reduction in the incidence and severity of acute
dendritiform keratopathy, scleritis, episcleritis, iritis, the
incidence (but not the severity if it occurred) of corneal stromal
immune keratitis, and the incidence of late-onset ocular inflammatory disease, e.g., episcleritis, scleritis, iritis.537541 Effect on
PHN was variable with some reports showing no efcacy, and
others notable decrease in severity and incidence although more
recent studies indicate that aciclovir is inferior to other oral
antivirals in prevention of PHN.391,537544 As noted below,
famciclovir, valaciclovir, and brivudine are superior in their
inhibiting effect on PHN.
It should be noted that, with ACV, gastrointestinal upset,
particularly diarrhea, has been a distressing side effect in certain
individuals. This appears to be due to lactose intolerance, which
is found in North American adults in ~75% of Native
Americans and blacks, 51% of hispanics, and 21% of whites.
ACV tablets contain lactose; intolerance to lactose is a common
cause of intolerance to milk and milk products owing to lack of
the intestinal enzyme lactase. Manka has reported reversal of
this oral ACV side-effect by administration of oral lactase in the
form of one Lactaid caplet ve times daily po.545
Valaciclovir (VCV) Valaciclovir is a prodrug of acyclovir
synthesized to enhance GI uptake of ACV. It is hydrolyzed back
to ACV resulting in ve times the bioavailability of the latter
drug.200,546 The valine and valinevaline ester prodrugs of ACV
penetrated the anterior segment of the eye signicantly better
than acyclovir alone, probably via a carriermediated transport

mechanism. Vitreous levels of the prodrugs were not
measurable.547
Clinical studies comparing valciclovir 1.0 g po tid with acyclovir
800 mg po 5 day for 7 or 14 days in 1141 immunocompetent
zoster patients (35 with HZO) revealed the drug-equivalency
in acceleration of dermal healing and reduction of duration of
viral shedding, but also that VCV was signicantly better in acute
pain resolution and reduced duration of PHN up to one year of
follow-up.201,548 Data from 14 days of treatment did not differ
from that of 7 days. Further, studies on postherpetic neuralgia
(PHN) revealed that the median time to pain resolution was 38
days with VCV and 51 days with acyclovir (P< 0.03). Other
studies support the high efcacy of valaciclovir in herpes zoster
particularly if started within 72 h of rash onset.381,400,549551
The only ocular study compared ACV with VCV in 121
immunocompetent patients with acute HZO, and reported an
incidence of keratitis, uveitis, and episcleritis and long-term
complications that was similar in both groups.552 These included
acute pain in about two-thirds of each group, conjunctivitis
(54% and 52%, respectively), supercial keratitis (39% and 48%,
respectively for punctate keratitis; 11% in each group for
dendritic keratitis), stromal keratitis (13% in each group), and
uveitis (13% and 17%, respectively). Most patients reported
prodromal pain and pain duration and severity. After
1 month, 25% of patients in the valaciclovir group and 31%
in the acyclovir group still had pain. Neither group had any
incidence of neurotrophic keratitis or scleritis. The authors
concluded that VCV was a valid alternative to ACV in treatment of HZO but like famciclovir (below) was superior in acute
and long-term pain inhibition and in patient compliance with
only tid dosing. The absence of neurotrophic keratopathy is in
marked contrast to this and other authors experience, however.
Valaciclovir prophylaxis against recurrent HSV genitalis with
doses of 500 mg or 1 g/day is as effective as 400 mg twice daily
of acyclovir which suggests that prophylaxis against ocular HSV
would follow a comparable dose and be used for 1 year or
possibly more. Tolerance to valaciclovir, like its active metabolite aciclovir, is generally good. Central neurological toxicity may
be observed with high doses, but regresses on withdrawal.553 It
is of note, however, that some severely immunocompromised
HIV patients have developed thrombocytopenic purpura/
hemolytic-uremic syndrome with a few deaths.391 As a result
this drug is not FDA approved for use in immunocompromised
patients but is approved for therapy of varicella zoster and
genital HSV.
Famciclovir (FCV) Famciclovir, a third-generation nucleoside,
is the orally bioavailable diacetyl ester of the active antiviral,
penciclovir. It is similar to ACV in mechanism of action and
antiviral activity against HSV 1 + 2, VZV, and EBV but superior
in GI absorption: 77% versus only 30% for ACV.40,107,554556 The
drug is FDA approved for treatment of herpes zoster infection
at doses of 500 mg tid for 7 days preferably starting within 72 h
of onset of rash. Clinical studies indicate that famciclovir
accelerates healing time as well as ACV on less frequent dosing.
Further, Dworkin et al have reported that treatment of acute
herpes zoster patients with famciclovir signicantly reduces
both the duration and prevalence of PHN.504 The overall
efcacy has been conrmed by others.550,557564
The masked, controlled HZO study by Tyring et al revealed
that famciclovir 500 mg tid was well tolerated and demonstrated efcacy similar to acyclovir 800 mg ve times daily.565
Of the 454 patients enrolled, those experiencing severe or one
or more ocular manifestations were similar for famciclovir
(142/245, 58.0%) and acyclovir (114/196, 58.2%). There was
Brivudin While oral brivudin is not available in this country, it

is licensed throughout Europe. Studies have shown that this
drug is as effective and safe as famciclovir and valaciclovir and
superior to acyclovir in therapy of acute HZ and HSV. Because
brivudin has a markedly higher anti-VZV potency than oral
acyclovir, valacyclovir, or famciclovir it is given in a single dose
regimen of 125 mg po qd for 7 days.543,546,566
Wood et als study in 1076 acute HZ patients on the effect
of the time of initiation of treatment, early (<48 h) versus late
(4872 h), on prolonged pain revealed that acyclovir signicantly shortened the time to complete resolution of zosterassociated pain compared with placebo and that valaciclovir was
superior to acyclovir in this regard even when therapy was
delayed up to 72 h after rash onset. Median times to complete
resolution of pain were 28 and 62 days, respectively, for patients
treated with acyclovir and placebo within 48 h and 28 and 58
days, respectively, for those treated later. In the valaciclovir
versus acyclovir study (in patients 50 years of age), the
corresponding gures were 44 and 51 days for patients treated
early and 36 and 48 days for those treated later (P<0.05 for
all).567 This agrees with the report by Lilie et al that valaciclovir,
famciclovir, and brivudin are comparably effective in the
reduction of the incidence and/or prevention of zosterassociated pain and PHN and that acyclovir is not.543
Corticosteroids In the era of oral antivirals the most denitive
studies on the role of steroids in this illness are two large
controlled clinical trials combining acyclovir with corticosteroids. Two hundred and eight immunocompetent patients
over 50 years of age with localized herpes zoster of less than
72 h were enrolled in one study by Whitley et al.568 Acyclovir or
a matched placebo was given po, 800 mg ve times daily, for
21 days. Prednisone or a matched placebo was given po at
60 mg/day for the rst 7 days, 30 mg/day for days 814 and
15 mg/day for days 1521. Patients receiving both had a
moderate but statistically signicant acceleration in the rate of
cutaneous healing and relief of acute pain. There was also an
improved quality of life: less need for analgesics, more uninterrupted sleep, shorter time to resumption of usual activities.
The study by Wood et al produced similar results. Neither study
demonstrated any effect on PHN.568,569 This was in agreement
with the report by Gross et al.546
Corticosteroids can be considered as soon after diagnosis as
possible for patients with at least moderately severe pain or rash
as well as for patients with VZV-induced facial paralysis and
cranial polyneuritis to improve motor outcomes, although the
benet of this treatment has not been fully studied.570 Because
of the potential risks of systemic steroids in acute zoster, the
well-known adverse side effects of these drugs, and the risk of
disseminated infectious disease, the use of systemic steroids
must be carefully assessed. It would appear advisable to limit
the use of systemic corticosteroids to those nonimmunocompromised, nondiabetic patients suffering acute pain uncontrolled by day three of opioid the vasculitic complications of
herpes zoster ophthalmicus such as severe scleritis, uveitis,
and orbital apex syndrome and those at minimal risk for
adverse steroid reaction to achieve faster improved quality
of life.466,511,546,568,569,571
Pharmacologic agents for PHN or PHI

A recent meta-analysis on analgesic therapy in PHN by
Hempenstall et al revealed that there is evidence to support
the use of the following orally administered therapies: tricyclic
antidepressants, strong opioids, gabapentin, tramadol, and

pregabalin. Topical therapies associated with efcacy were
lidocaine 5% ointment, EMLA (lidocaine-prilocaine cream), or
lidocaine patch. To this list the author adds diphenhydramine
(Benadryl) cream or pills (2550 mg qdbid) for PHI and
marcaine injection of the supra- and infraorbital and supranasal
nerve or, lastly, Botox A map injections for PHN unresponsive
to other therapies.521,572578,570
Tricyclic antidepressants (TCAs) in postherpetic neuralgia
The use of psychotropic medication for treatment of chronic
pain has become an integral part of the multimodal treatment
of chronic neuralgia of numerous etiologies.381,579,580 The
mechanism of action is through blocking the reuptake of norepinephrine and serotonin thereby relieving pain by increasing
the inhibition of spinal neurons involved in conscious pain
perception.516
Watson et al in a placebo-controlled study reported the
signicant clinical value of the tricyclic antidepressant (TCA)
amitriptyline (AT) 25150 mg po daily, in elderly patients
suffering from permanent postherpetic neuralgia.581,582 There
was a drug response rate of 60%, with very severe pain
becoming mild but not entirely relieved. In 66% of the patients,
pain was reduced from severe to mild within 3 weeks, yet serum
amitriptyline levels were below those associated with antidepressant activity. Increased dosage produced increased pain
in some patients, indicating a therapeutic window for the
analgesic dosage of the TCAs. In a subsequent double-blind,
crossover trial of AT versus nortriptylene (NT) in 33 patients,
67% had at least a good response to AT or NT, or both. The
authors concluded that, while there were fewer side effects with
NT, AT and NT had a similar analgesic action for most
individuals without any antidepressant effect.583 Analgesia is
achieved at levels half of that required for antidepressant effect
and occurred within 2 weeks of initiating treatment.584
Similarly, there are reports that the combination of either
doxepin or amitriptyline with a narcotic analgesic reduced pain
intensity more than either an antidepressant or a narcotic drug
alone in patients suffering chronic neuralgic pain.521,574576
TCA treatment outcome studies by Bowsher et al on 279
patients revealed a number of important ndings.579,585 The
time at which TCA treatment is begun is by far the most critical
factor. If started between 3 and 12 months after acute HZ onset,
more than two-thirds obtain pain relief; between 13 and 24
months, two-fths (41%); and more than 2 years, one-third.
Background and paroxysmal pain disappear earlier and are more
susceptible of relief than allodynia. Twice as many (86%) of
PHN patients without allodynia obtain pain relief with TCA
treatment than those with (42%). The use of ACV for acute HZ
cuts by more than half the time-to-relief of PHN by TCAs, and
PHN patients with burning pain are signicantly less likely to
obtain pain relief with TCAs than those without (P<0.0001).
This indicates that while TCAs are extremely important in
treating various forms of PHN, other agents may need to be
added to cover unresponsive symptoms.
Because there are little data suggesting notable efcacy
differences among amitriptyline, doxepin, imipramine,
nortriptyline, or desipramine for treatment of pain, selection
of one or more of these drugs may depend on the side effects.
The tertiary amines, amitriptyline, imipramine, and doxepin,
have more anticholinergic, cardiac, and central nervous system
effects than the demethylated secondary amines, nortriptyline
and desipramine. Therefore, in the more vegetative patient,
desipramine may be the least sedating, whereas an agitated
patient may benet more from amitriptyline or doxepin.
Nortriptyline is the drug of choice in patients with bradycardia
CHAPTER 49
also no signicant difference between groups for visual acuity

loss. Effect on PHN was not reported.
675
SECTION 6
or heart block and the best tolerated by patients.580 Common

dosage administration is nortriptylene, desipramine, amitriptyline, or doxepin, 25 mg po at bedtime, increasing increments
every 57 days to a stable dose of 25100 mg daily as tolerated.
The drugs are usually given at bedtime to take advantage of the
sedating effect.516
Alternative therapy includes the combination of a neuroleptic
agent such as perphenazine, a tranquilizer, and amitriptyline
(Triavil 210 or Estrafon 210 tablets) three to four times daily
for those chronic pain patients with a mixture of anxiety and
agitation with symptoms of depression. This combination
therapy is particularly effective in patients under the age of
60 years. In the older age group, however, there is an increased
risk of occasionally irreversible tardive dyskinesia (involuntary
movements of the face or extremities) and other extrapyramidal
reactions such as motor restlessness, oculogyric crisis, or
opisthotonos.580,586,587 Upon the appearance of any suggestion
of these symptoms, the medication should be stopped
immediately.
Despite the side effects discussed previously, the use of
antidepressants or antidepressantantianxiety combination
medications has proved to be highly effective in situations in
which other medical management has failed.
676
Anticonvulsants The anticonvulsant gabapentin (Neurontin)

600 mg po bid-6id is frequently very effective at controlling
PHN as a single agent and may be given intermittently or
continuously for months to years as tolerated or needed.588591
The mechanism of action appears to be multifactorial. A review
by Taylor et al includes the following: gabapentin (1) increases
the concentration and probably the rate of synthesis of GABA
in brain, which may enhance nonvesicular GABA release during
seizures; (2) binds with high afnity to a novel binding site in
brain tissues that is associated with an auxiliary subunit of
voltage-sensitive Ca2+ channels; (3) may modulate certain types
of Ca2+ current; (4) reduces the release of several monoamine
neurotransmitters; (5) inhibits voltage-activated Na+ channels;
and (6) increases serotonin concentrations in human whole
blood, which may be relevant to neurobehavioral actions.592
Studies in rats indicate that gabapentin interacts with a
novel-binding site on cortical neurons that may be associated
with the L-system amino acid transporter of brain cell
membranes.593
A literature search found two published studies of the efcacy
of gabapentin in a total of 563 patients with PHN that had
persisted for at least 3 months after the healing of herpes zoster
rash.594,595 Using maximum target dosages of 18003600 mg/
day, gabapentin produced signicant reductions in mean daily
pain scores compared with placebo and enhanced overall quality
of life in patients with chronic PHN. It is especially effective
against allodynia, neuralgia not particularly responsive to TCAs.
Further, its use has been shown to decrease opioid dosing.
Tapering may be attempted periodically to see if a lower dose
controls symptomatology with time.
Berry et al have reported a double-blind, placebo-controlled
crossover study measuring the effect of a single dose of oral
gabapentin (900 mg) on HZ pain and allodynia. Pain severity
decreased by 66% with gabapentin compared to 33% with
placebo. Reductions in allodynia area and severity, and overall
pain relief, were also greater with gabapentin.596 In working out
the best starting dose, Jean et al enrolled 61 subjects (32 male/
29 female) in the gabapentin study. The intensity of pain was
greatly improved in all three groups after three days of
treatment and there was no statistically signicant difference
among subjects taking 200 mg, 400 mg, or 600 mg with respect
to dizziness, drowsiness, or fatigue. The authors felt that as
starting with a dose of 200 mg/day did not offer a better
reduction of side effects, that 600 mg/day gabapentin would be

a safe and effective starting dose for patients with postherpetic
neuralgia.597 This could then be worked up to as high as
3600 mg/day as needed.
Pregabalin (Lyrica) has a similar pharmacological prole to
that of its predecessor gabapentin, but showed greater analgesic
activity in rodent models of neuropathic pain. The mechanism
of action is unclear, although it may reduce excitatory
neurotransmitter release by blocking voltage-gated calcium
channels. Oral pregabalin 150600 mg/day, bid to tid (600 mg
maximum/d), was superior to placebo in relieving pain and
improving pain-related sleep interference in double-blind,
placebo-controlled, studies in a total of 776 patients with PHN.
In two studies, signicant improvements in daily mean pain
scores were apparent on the rst or second day of treatment with
pregabalin tid. The drug was generally well tolerated (maximum
600 mg/day) in clinical trials that enrolled most elderly PHN
patients. Dizziness, somnolence, and peripheral edema were
the most common adverse events.598600
Both gabapentin and pregabalin are reported as safe, tolerated
well, and efcacious. They are both FDA approved for treatment of PHN.
Combination therapy and opioids For patients not responding
satisfactorily to single-agent therapy, gabapentin and a TCA
such as nortriptylene or desipramine may be combined for
additive effect. If the combination is still not totally effective or
if one or both drugs are not tolerated in treatment of PHN,
slow-release opioids such as oxycodone-CR (Oxycontin-CR)
1020 mg po q 12 h may be added or given as a single agent to
give added relief. Controlled-release oxycodone is an effective
analgesic for control of steady or paroxysmal pain, and
allodynia.583,588,591 Because of controlled-release (CR), there is
no high and, therefore, little chance of addiction when taken
by the appropriate oral route. The combination of either
doxepin or amitriptyline (AT) with a narcotic analgesic reduced
pain intensity more than either an antidepressant or a narcotic
drug alone in patients with chronic neuralgia.575 Because of
ATs less desirable side effect prole, combination of nortriptylene or desipramine with an opioid analgesic for recalcitrant
cases seems more desirable, e.g., nortriptylene 50 mg q hs and
oxycodone slow release (Oxycontin-SR 1040 mg po q 12 h)
plus gabapentin or pregabalin. Again, periodic tapering
should be attempted as PHN may decrease spontaneously
over time.
Topical analgesics Topical lidocaine is often very effective for
allodynia and itch as well as ache and lancinating pain and may
be added to any or all of the above pain agents.573,600602 The
mechanism of action is by blocking the voltage-gated Na+
channels on excitable membranes thereby preventing the
generation and conduction of nerve impulses.573 In Wasner et
als study of topical lidocaine in PHN, they found that patients
with impairment of nociceptor function (pain transmitter
neurons) had signicantly greater pain reduction under
lidocaine versus placebo. Patients with preserved and sensitized
nociceptors demonstrated no signicant pain relief. They concluded that at least some PHN pain is caused by pathologically
sensitized nociceptors due to a loss of function of cutaneous Cnociceptors within allodynic skin. Patients responded well to
topical lidocaine even if the skin was completely deprived of
nociceptors.578
Forms of lidocaine which are most useful for HZO are
lidocaine 5% ointment or EMLA (lidocaine-prilocaine) cream
which may be applied q 46 h as needed (DPL unpublished).
The 5% lidocaine patch is also effective but sometimes difcult
to apply on the face or in the hair.602

Capsaicin cream is prescribed by some physicians but is
generally poorly tolerated and has fallen out of use with the
availability of the lidocaine agents.
Other forms of acute pain and PHN control

Nerve blocks of the supra- and infraorbital and supranasal
nerves with bupivacaine, epinephrine, and methylprednisolone
may be given for severe acute HZO pain with notable
efcacy.577,603,604 Botulinum map injections are effective in 50%
of intractable HZO PHN patients.572,605
Surgical procedures in HZO
FIGURE 49.34. Acute HZO in a patient with chronic lymphatic

leukemia showing partial lid loss with exposure and neurotrophic
keratitis.
CHAPTER 49
The most common indications for a surgical procedure in

herpes zoster ophthalmicus are exposure keratopathy and
anesthetic cornea. If lid closure is good but the tear meniscus
low or unstable, punctal plugs may sufce and the other steps
for dry eye listed under HSV Neurotrophic Ulcers above should
be taken. In the more severe forms of the illness, signicant
scarring or partial destruction of lid tissue will interfere with
blinking and normal lid closure during sleep. If lid structures
are basically intact but closure is incomplete due to scarring, a
lateral or lateral and medial tarsorrhaphy with ointment q hs
should sufce to protect the globe. If lid tissue has actually been
lost, plastic reconstruction involving the swinging of flaps may
be necessary. This should be done by a surgeon experienced
with such procedures as the remaining lid tissues are often
friable and hold sutures poorly (Figs 49.34 and 49.35).
In the partially or totally anesthetic cornea where the
epithelium is gray, unhealthy, or prone to recurrent breakdown,
partial tarsorrhaphy is also indicated. Wherever medical and
lateral tarsorrhaphy is used, an open area should be left between
the lids to allow the physician an adequate view of the underlying globe. All of the therapeutic steps outlined in the section
on HSV Neurotrophic Keratopathy should be followed (Fig. 49.36).
If corneal melting occurs, sealing the area with the sterile
cyanoacrylate glue (Dermabond, Epidermglu, see section on
HSV therapy of trophic ulcers) and inserting a Plano T
therapeutic lens to cover the rough surface of the glue are
indicated. The patient should be informed that this gluing
procedure is not FDA-approved for the eye, however. With time
the cornea will almost invariably heal under the glue and the
glue will dislodge spontaneously or a corneal pannus will grow
in and heal the surface under the glue. Good alternative but
more extensive surgical procedures include the pulling down of
a conjunctival flap if the tissue has not been too severely scarred
by the disease or the placing of conjunctival transplants from
the contralateral eye. Stem cell transplantation from the
contralateral eye is another viable option.606
Keratoplasty is potentially complicated in herpes zoster ophthalmicus. Anesthetic corneas heal poorly, and the transplanted
eye is prone to melting and superinfection. A cornea that has
scarred but retained a reasonable amount of sensation is perhaps
the best situation in which keratoplasty might succeed. Any
major surgical procedure, transplantation, or cataract extraction, should be deferred, if possible, until the eye is uninflamed.
The longer it is deferred beyond this, the better. Cataract
extractions which must be performed for visually signicant
cataracts are best done with a posterior limbal incision taking
care to avoid incisions in the anesthetic cornea. The visual
acuity can be diminished by the lipid keratopathy that persists
after the herpes zoster ophthalmicus has subsided. Argon laser
has been employed to diminish the deposition of lipid.607
If keratoplasty is performed, however, a lateral tarsorrhaphy
should be done at the same time to protect the graft. Reed et al
performed penetrating keratoplasty on 12 patients with HZO,
ve of whom had neurotrophic ulceration.608 Lateral tarsorrhaphy was performed in 10 patients to prevent postoperative
FIGURE 49.35. Close-up of the patient in Figure 49.34 showing

extensive neurotrophic ulceration and exposure keratitis with vascular
pannus moving in 360.
breakdown of the corneal epithelium. At average follow-up of

3 years, 83% of the grafts were clear and 75% of the eyes had
vision at 20/80 or greater. Similarly, Marsh and Cooper reported
that tarsorrhaphy led to rapid epithelial healing in grafted zoster
eyes, vision of 6/12 or better in six of seven eyes undergoing
keratoplasty, and clear grafts for 29 years of follow-up.475
677
2.
3.
4.
5.
FIGURE 49.36. Lateral tarsorrhaphy was performed to protect a
cornea left anesthetic from HZO.
6.
7.
Neovascularization was closed with the argon laser preoperatively.

Eighteen additional patients with herpes zoster ophthalmicus
underwent successful cataract extraction, 12 receiving posterior
chamber implants. Trabeculectomies were successful in seven
zoster glaucoma patients, although ve subsequently developed
cataracts. In a review of 15 patients, (12 VZV, three varicella
keratopathy) Tanure et al reported placing lateral tarshorrhaphies
at the time-off of our grafts.609 Three patients had steroidresponsive rejections, two failures (one primary and one uneurotrophic). At an average follow-up of 50 months, 87% of grafts
remained clear and best corrected visual acuity was 20/100
or better in 53% of eyes. The authors felt that useful visual
rehabilitation could be achieved in VZV eyes but that careful
postoperative follow-up, frequent lubrication, and lateral tarsorrhaphies to protect the surface were major factors in enhancing
chances of successful outcome.
A newer procedure, the Boston keratoprosthesis, holds great
promise for success in many of the most severe cases,
however.341,610
8.
9.
days. Less protection against PHN compared to FCV or

VCV. For immunocompromised patients, acyclovir can be
given at dosages of 10 mg/kg intravenously every 8 h for
7 days in adults and 500 mg/m2 every 8 h for children
under 12 years old (FDA-approved dosages). FCV but not
VCV may be used in immunocompromised patients.
Simultaneously with antivirals start nortriptylene or
desipramine 2575 mg po q hs for 3 months or longer to
inhibit acute pain and minimize or prevent PHN
particularly in patients > 50 years old. Add gabapentin
(Neurontin) 600 mg bid to 600 mg bid as needed.
For dendritic keratitis, topical antibiotic. Antivirals are
variably effective and dendrites often self-resolve.
For moderate to severe corneal or scleral inflammatory
disease, topical steroids ranging from 0.125% prednisolone
two to four times daily, up to 0.1% dexamethasone in a
frequency as disease warrants. Taper over a several week
period.
If disease is mild or there is no pain or ocular involvement,
warm compresses or Burows solution compresses 15 min,
four times daily until scabbing has cleared.
For iritis, cycloplegia (homatropine, atropine) and topical
steroids as needed.
For acute neuralgia nonnarcotic or narcotic analgesics on
days one through three. If there is no resolution of pain or
increase in neuralgia, add a systemic steroid, prednisone,
in the immunocompetent patient at a dosage of 20 mg
orally three times a day for 4 days, 20 mg orally twice a
day for 4 days, and 20 mg orally q AM for 4 days.
Continue po antiviral. Systemic steroids may also be useful
for severe orbital edema with superior orbital ssure
syndrome, facial nerve paralysis, and to hasten the
patients return to a more comfortable, functional life.
Contraindications are hypertension, diabetes, gastritis,
osteoporosis. Risks associated with the use of
corticosteroids must be carefully evaluated, and they
should not be used in the absence of concomitant antiviral
therapy, there is no effect on PHN or PHI.
Articial tears (nonpreserved) and ointments for exposure
or hypesthetic keratitis, dry eye, or unstable tear lm.
Supra- and infraorbital, sna superior trochlear bupivicaine
block for severe pain uncontrolled by medical means. It is
best if it is administered by a physician experienced in
this area.
SECTION 6
Summary of therapeutic approaches to HZO
678
Prevention Recent studies have shown that the VZV vaccine in

children begins to lose efcacy after a few years and should
probably be repeated.611 The study on VZV vaccination to
prevent zoster has now been completed. Oxman et al enrolled
38 546 adults 60+ years of age in a randomized, masked,
placebo-controlled trial of an investigational live attenuated
Oka/Merck VZV vaccine (zoster vaccine).355 The use of the
zoster vaccine reduced the burden of illness due to herpes zoster
by 61% (P<0.001), reduced the incidence of postherpetic
neuralgia by 66% (P<0.001), and reduced the incidence of
herpes zoster by 51% (P<0.001). This vaccine has been
approved by the FDA for adults over 60 years. HZO occurred in
10% of the ~8000 HZO patients in the study, the duration of
efcacy of the vaccine yet known. It is clear, however, that
Zostavax is a very important advance in the therapy
(prevention) of zoster.
Acute disease
1. Antiviral pills: Famciclovir FCV) 500 mg po tid or
valaciclovir (VCV) 1 g po tid for 7 days, preferably starting
within 72 h of rash onset. Second-line antiviral pill if rst
two not available, acyclovir 800 mg po 5x/day for 710
Long-term or chronic problems

1. For exposure or corneal ulceration use a high-water-content
therapeutic soft contact lens (Permalens, Kontur or other
listed under HSV Neurotrophic Ulcer), with or without
tissue adhesive (Dermabond, Epidermglu), punctal plugs,
lateral tarsorrhaphy, conjunctival flap, or amniotic
membrane, corneal or Boston keratoprosthesis transplant
as described in text (under HSV and HZO) (Fig. 49.37).
2. Late pseudodendritic keratitis: antibiotic lubrication only
or, for persistant lesions, there is variable response to
topical or systemic antivirals. Trial and error must be used.
3. Immune stromal disciform, IK, limbal vasculitis, episcleritis,
scleritis: Depending on the severity of the inflammation,
starting therapy may range from 1% prednisolone or 0.1
dexamethasone every two h while awake to just bidtid.
Tapering the dose in 50% reduction steps begins as the
immune disease lessens and is continued over several
weeks to months. by switching to weaker dilutions of
prednisolone, e.g., 1/8%, or to intermediate-strength
rimexolone or lotoprednol or qd to qod. The latter three
steroids have less propensity to elevating the intraocular
pressure.
4. Iritis therapy is similar to that of stromal immune

keratitis. Treat any secondary glaucoma. Secondary
glaucoma: (a) trabeculitis therapy is strong topical steroids
with glaucoma drops. Pressure should drop quickly (days) if
inflammatory trabeculitis is the cause and is responding to
steroids. (b) Glaucoma due to debris or partial synechial
angle closure is treated with drops such as beta blockers,
e.g., timoptic, betoxalol, levobunalol, or alpha adrenergics,
e.g., brimonidine, or carbonic anhydrase inhibitors, e.g.,
brinzolamide, dorzoalamide, or prostaglandin inhibitors,
e.g., latanaprost, or bimatoprost may be used once or twice
daily alone or in combination with other drug groups just
named. Prostaglandin inhibitors may, however, increase
inflammation. Antivirals are not indicated.
5. For acute pain and chronic postherpetic neuralgia (PHN),
use tricyclic antidepressants (nortriptylene or desipramine)
alone or in combination with gabapentin (Neurontin) and/
or narcotic analgesics, per text and topical analgesics (#6).
6. For PHN and/or PHI (itch) lidocaine 5% ointment EMLA
cream q 412 h lidocaine. Benadryl cream to involved skin.
Oral benadryl 2550 mg may also help PHI. The duration
of treatment is several months to several years.
7. Bupivicaine or botulinum injections per text for intractable
pain.
EPSTEINBARR VIRUS OCULAR INFECTIONS

The EpsteinBarr Virus (EBV), a ubiquitous member of the
herpesvirus family, is the etiologic agent of or is causally
associated with infectious mononucleosis, nasopharyngeal
carcinoma, Hodgkins disease, and African Burkitts lymphoma.
This DNA virus has an afnity for some epithelial cells of the
naso-oropharynx, and B-lymphocytes, transforming them in
vitro into lymphoblasts capable of continuous cultivation.
This limited tissue tropism is a result of the limited cellular
expression of its receptor. By the third decade of life, 90%
of adults are positive for EBV antibody.612615 EBV infection in
childhood usually produces little overt clinical disease. In
adolescence or adulthood, however, the characteristic picture of
infectious mononucleosis (IM) is fever, extensive lymphadenopathy, sore throat, lymphocytosis, hepatitis, pericarditis, polyarthritis, myositis, and follicular conjunctivitis.615617
Transmission is by upper respiratory droplet contamination.
EBV-specic antibodies are present in 5085% of children under
the age of 4 years who live under low socioeconomic conditions
and in 2682% of college students.24,613,618621
The traditional serologic diagnosis of IM has depended on
detection of the heterophile antibody. The Monospot test is also
highly reliable, but the very useful and readily available tests
are those for antibody to EBV capsid antigen (VCA) and EBV
nuclear antigen (EBNA).24,615,618 Patients with acute IM will have

elevated VCA antibodies, but EBNA antibodies will be absent
until several weeks or months after the acute infection. Both
antibodies are detectable throughout life. A recent EBV infection
may, therefore, be diagnosed by the presence of elevated VCA
antibodies in association with absent or rising EBNA antibodies.
Conversely, the presence of anti-VCA and anti-EBNA
antibodies at comparable levels is indicative only of a previous
infection with EBV at an undetermined time in the past.
The ocular manifestations of IM encompass a wide range of
anterior segment and neuroophthalmic ndings.24,60,615,618620,622,623
Neuroophthalmic manifestations include papilledema, optic
neuritis, and cranial nerve palsies. Anterior segment ndings
include follicular conjunctivitis, flat or nodular scleritis, episcleritis, subconjunctival hemorrhage, iridocyclitis, oculoglandular syndrome, infectious dendritic epithelial keratitis and
stromal keratitis, and hemorrhagic or follicular conjunctivitis
that may become membranous.624 In one 14-year-old girl EBV
presented as exophthalmos and ocular muscle swelling.625 EBV
genome was detected in activated T lymphocytes and local
muscle biopsies. The patient responded to immunosuppressive
treatment with steroid and cyclophosphamide.
Other patients are not so fortunate. A 41-year-old male
presented with conjunctival swelling of his left upper eyelid and
increased over 2 months despite topical corticosteroid treatment.
Conjunctival biopsy revealed a highly malignant, CD3+ and
BCL2+ extranodal T-cell lymphoma. All cells were positive for
EBV RNA. Despite initial response to chemotherapy the lesion
recurred and the patient died from malignant marrow
inltration.626 Woog et al have reported 8 patients (5 male,
3 female) with natural killer/T-cell lymphoma (NKTL) involving
the orbit and/or ocular adnexa.627 The mean age at presentation
was 45 years (range, 2665). CD3, CD56, and EBV-encoded
mRNA were found in every patient. Four of the 8 patients (50%)
with NKTL involving the orbit or ocular adnexa had systemic
involvement at presentation. Five of the 8 patients (62.5%) had
concurrent sinonasal involvement, whereas 3 (37.5%) had
orbital involvement alone. Various chemotherapeutic regimens
typically employed in the treatment of non-Hodgkins lymphoma, steroids, surgical intervention, and radiation were used
but seven (87.5%) of the patients died 5 weeks to 13 months
after presentation, and only 1 (12.5%) is alive without disease at
5-year follow-up.
Meisler et al have reported EBV Parinauds oculoglandular
syndrome.622 In another orbital case EBV dacryoadenitis
resulted in severe keratoconjunctivitis sicca KCS in a 10-yearold male.628 Serologic and immunohistologic data conrmed
the diagnosis and the histopathology of the lacimal gland was
similar to that of primary Sjgrens syndrome. Treatment with
acyclovir and cyclosporin A was highly successful.
CHAPTER 49
FIGURE 49.37. (a) Chronic HZO with a melting neurotrophic ulcer over focal interstitial keratitis. (b) Application of tissue adhesive and a
therapeutic soft lens to seal the thinning trophic ulcer. (c) At 9 weeks after placement of the tissue adhesive, the epithelium has healed
completely, the adhesive has dislodged, the soft contact lens has been removed, and the inflammation has resolved. Pannus completely covers
the cornea.
679
FIGURE 49.38. Chronic EpsteinBarr virus keratitis showing anterior

and midstromal discrete slightly irregular whitish focal inltrates similar
to those seen in adenoviral keratitis.
SECTION 6
Courtesy of Kirk Wilhelmus, MD, Cullen Eye Institute, Dallas.
680
The epithelial keratitis may be punctate or mimic HSV

with multiple microdendritic ulcers. EBV has been cultured
from or detected by ELISA and PCR testing in corneal,
conjunctival, and tear lm samples from such patients with
microdendrites.24,60,615,618620,622,623,629632 While patients have
resolved disease on topical ACV or TFT it is unclear whether
these agents actually influenced the course of infection.
Stromal keratitis is not common but has been reported by a
number of investigators.60,615,618,620,623,629,631,632 By the classication of Matoba and McCulley there are two forms of EBV
stromal keratitis.619 The anterior stromal form has highly
demarcated areas of discrete, granular, circular, or ring-shaped
opacities ranging in size from 0.1 to 2.0 mm. These may be
distributed diffusely across the cornea and associated with
variable degrees of supercial and deep neovascularization. The
intervening stroma is spared, with consequent preservation of
relatively good to normal vision. The overlying epithelium is
usually intact or has mild punctate granularity. These anterior
stromal lesions may be differentiated from adenovirus-induced
opacities because the latter are soft inltrates that usually
develop sequentially to an overlying punctate epithelial keratitis
and tend to be conned to the subepithelial and anterior
stromal areas. EBV inltrates were discrete-edged, multifocal,
and pleomorphic, involving the anterior and mid-stroma and
could develop in the absence of any epithelial keratitis
(Fig. 49.38). The second form of EBV keratitis is a blotchy,
peripheral full-thickness or deep stromal inltration that is
clinically similar to syphilitic interstitial keratitis or HSV stromal
keratitis.
The corneal disease may be either unilateral or bilateral with
the symptoms of irritation, photophobia, watering, and the
signs of conjunctival hyperemia and mild blurring of vision.
Onset is ~14 weeks after onset of the acute IM or flu-like
illness compatible with IM. There are rare reports of patients
who experience recurrent bouts of EBV keratitis in both eyes.
Four years after the onset of initial corneal EBV keratitis one
patient also developed a nodular scleritis in one eye.620,629631

The keratitis is felt to be based on an immune reaction to EBV
or EBV-antigen-bearing cells located in the cornea. EBV is
known to persist in the host in the form of a chronic low-grade
infection as opposed to true latency. There is, therefore, chronic
virus shedding and therefore persistent stimulation of the
immune system. Because of the probable immune etiology of
the keratitis, topical steroids are highly effective in patients with
vision-disturbing disease. Other patients will, however, recover
spontaneously over a several-month period without therapy.
The uveal tract may also be involved, as manifested by an
acute iritis responsive to topical steroids or a recalcitrant chronic
smoldering focal or panchorioretinitis with development of
secondary cataract and macular edema. This EBV intraocular
inflammatory disease may develop several months after the
onset of the acute IM. The uveitis may or may not respond well
to topical or systemic steroids. Morishima et al have reported a
case of uvieitis associated with chronic active EBV infection in
a 7-year-old girl.633 The patient had fever and hepatosplenomegaly
followed by laft facial nerve palsy. Eye exam showed right
iridocyclitis and bilateral optic disc edema. EBV antibody titers
were highly elevated. The patient responded dramatically well
in all parameters to treatment with topical and systemic
steroids, interleukin-2, and splenectomy.
Wong et al reported three cases of chronic EBV systemic
infection with ocular manifestations ranging from keratitis and
iritis responsive to 1% prednisolone drops qid over a 3-week
period to severe panuveitis recalcitrant to topical or systemic
steroids.623 This patient also failed to respond to 10 months of
oral ACV 600 mg po ve times daily, but upon the addition of
topical ACV to this regimen, gradually cleared without recurrence
over the ensuing 5 months. The relationship of the antiviral
therapy to recovery from this disease is not established.
Treatment
Treatment for nonmalignant ocular EBV disease is a twopronged approach. Topical steroids for keratitis, iritis, scleritis,
and other anterior segment disease coupled often with systemic
treatment of the systemic disease with immunosuppressants
and, in some cases, splenectomy. Malignant disease is surgical
and chemotherapy with a poor prognosis for survival.
CYTOMEGALOVIRUS INFECTIONS
CMV is the most common virus known to be transmitted
in utero, with an incidence of infection of 2.2% of all live
births. By far the most common ocular manifestation is
chorioretinitis.617 With the advent of the AIDS epidemic and
pre-HAART (highly active antiretroviral therapy) this became a
particularly threatening illness, with up to 45% of all AIDS
patients developing this blinding CMV opportunistic infection
of the retina. The time period from 1993 to 1996 is classied as
pre-HAART and the period from 1997 to 1999 as the HAART
era, the era which has made a great difference in the incidence
and severity of this disease. Salzberger et al, for the purpose of
their study, have classied 19931996 as pre-HAART and
19971999 as the HAART era.634 They have reported that the
incidence of CMV disease declined rapidly and signicantly
from 7.34 cases per 100 patient years (py) in the pre-HAART
era to 0.75 cases per 100 py in the HAART era. The median
survival time in the pre-HAART era was 9.5 months; the
median survival was not yet reached 4 years of follow-up in
the HAART era. The only risk factors influencing survival were
CD4-cell count and antiretroviral therapy before and after
diagnosis of CMV disease.
Asymptomatic CMV infection is common in the immunocompetent segment of our population; 50100% of adults have
antibody indicative of previous infection.635637
Anterior segment disease in CMV infection is more common
than previously thought.638 A case of follicular conjunctivitis in
an otherwise healthy patient suffering from CMV mononucleosis syndrome has been reported, and transmission of
CMV through contact with another infected person with CMV
conjunctivitis has been suspected.285,639 In AIDS patients, there
have been reports of CMV infection of the conjunctiva.640
Corneal manifestations include asymptomatic linear and
stellate endothelial deposits in a reticular pattern and most
often in the inferior cornea. Brody et al have reported corneal
changes in 81% of 32 eyes with CMV retinitis; some had
mild iritis.638 Histopathologically, these are macrophages and
brin adherent to the endothelium with no evidence of CMV
infection indicating the corneal ndings may be useful in
warning of the presence of CMV retinitis but are secondary only
to the inflammation.641 As the corneal ndings do not resolve
after effective treatment of the retinitis, they cannot be used as
guidelines to the activity status of recurrent retinitis.642
CMV is shed in the tears and may be the means of transmission of disease in some patients.639,643 The transmission of
CMV by corneal transplantation has been reported by Holland
et al.644 Of 25 seronegative patients receiving corneal tissue
from CMV-positive donors, two (8%) seroconverted. Additionally,
of 11 patients seronegative preoperatively and receiving tissue
from seronegative donors, one seroconverted to be positive for
CMV antibody. No patient who was seroconverted, however,
had febrile illness or other signs of clinically overt disease and
all three grafts remained clear.
There is one case report of PCR+ CMV anterior uveitis and
glaucoma secondary to trabeculitis successfully treated with IV
and then oral ganciclovir. IOP decreased to normal and the
intraocular inflammation declined. After cessation of ganciclovir
administration, however, both recurred. Continuous administration may be required to control CMV infection in an
immunocompetent patient.645
Treatment
It is beyond the scope of this chapter to discuss therapy of CMV
ocular disease, as it is essentially treatment of HIV and/or CMV
retinitis which are discussed in other chapters in this series.
ADENOVIRAL OCULAR DISEASE

Epidemiology
The adenoviruses (Ad) make up a group of some 51 morphologically similar but antigenically distinct DNA viruses that
share a common group complement-xing antigen.646 They are
extremely stable, ether-resistant organisms, ubiquitous throughout the world, and causing infections of the upper respiratory
tract and the eye.27,614,647,648 These agents cause a number of
highly infectious, sometimes fatal diseases that affect multiple
organs, most commonly those associated with respiratory, genitourinary, and gastrointestinal tracts and the ocular surface.646,649
Many factors have driven a search for effective topical and systemic antivirals to adenoviruses. These include patient morbidity, economic losses and chronic visual disturbances associated
with epidemic keratoconjunctivitis (EKC) and pharyngoconjunctival-fever (PCF). There has also been a worrisome
recent trend of high morbidity and rising mortality associated
with systemic adenoviral infections in the immunosuppressed.649
As there is no known animal carrier of this virus, humans are
the only reservoir. Serologic studies have shown that there is a
low level of natural immunity in the general population of the
United States and European countries to offer protection

against EKC. This lack of immune protection would appear
to account for ocular adenovirus infections occurring primarily
in epidemic form, e.g., there is < 10% immunity for Ad8. The
pattern of adenoviral disease in Asia and Africa is that of an
endemic disease with 2585% of the general population having
positive serology to Ad8. Clinical cases of ocular adeno infection
tend to be sporadic.650,651 The incidence of acute disease in the
general population is generally low (0.031.10%). In situations
where there is close contact, however, the attack rate is high
(1032%), e.g., camps, home, prisons. The primary mode of
spread appears to be direct contact with contaminated
secretions on such surfaces as towels, bed linens, clothing, soap,
swimming pools, physical intimacy, and probably salivary and
nasal air droplet.27,100,648 It would also appear that there are
continual changes within the genome of adenoviruses but that
these are usually contained within the population at a
subclinical level. On occasion, however, a genotypic change may
enhance pathogenicity such that clinical disease in sporadic or
epidemic form ensues.27
Other settings which provide a unique opportunity for starting
an epidemic are the ophthalmologists ofce or hospital setting
through use of a contaminated tonometer on sequential
patients without adequate sterilization, after use on one patient
infected with Ad this hardy virus is passed from eye to eye.
Another excellent mode of spread is the contaminated hands of
the doctor or staff; a single handshake with an infected patient
will spread the organism. Further risk is in the waiting room.
Ad may survive for hours in a desiccated but viable form on the
furniture and magazines patients share.27,100,652654 Disease
transmission may be prevented by physicians and paramedical
personnel by frequent and adequate handwashing and by the
cleansing of ophthalmic instruments, especially tonometers, between patient examinations (alcohol swabs or Dakins solution
followed by careful rinse). This is particularly important when
examining any patient with a red eye or during times of Ad
epidemic in the community. Patients should be advised,
whether in the home or institutional setting, to avoid close
personal contact for at least 2 weeks and to use their own towels
and facecloths, avoiding sharing with those who are not
infected.655,656 Hard and soft contact lens sterilization studies
using Ad types 8 and 19 have shown that both viruses survive
hydrogen peroxide and heat sterilization systems. This strongly
suggests that contact lens wearers who contract acute
adenoviral ocular infection should simply dispose of their
contacts and buy new ones after the illness has resolved.657
Clinically, adenoviral ocular infections have four basic modes
of presentation: pharyngoconjunctival fever (PCF), epidemickerato-conjunctivitis (EKC), acute nonspecic follicular conjunctivitis (NCF), and chronic keratoconjunctivitis.
Of the four, EKC tends to be the most debilitating and often
localized just to the eyes. The Ad ocular infection is controlled
initially by the innate immune response on the ocular surface.
In studies on response to low-level Ad5 infection, conjunctival
epithelial cells showed upregulation of IFN-associated genes.
The peptide products of two of these, IP-10 and I-TAC, are
directly active against Ad3, and IP-10 is active against Ad5, two
of the milder forms of keratoconjunctivitis. However, the ocular
tropism and severity of disease caused by Ad8 and Ad19 may
be due in part to their resistance to these defensin-like
chemokines.658
It is also important to note that Ad infections may mimick
herpes keratitis. Chodosh et al reported 6 cases of adenoviruspositive, HSV-negative dendritic, geographic, or dendrogeographic
ulcerative keratoconjunctivitis.659 Serotypes isolated were 3, 8,
and 19. It is also reported by the national surveillance of ocular
CHAPTER 49
681

infectious diseases in Japan that 4.3% of cases of epidemic
keratoconjunctivitis (EKC) diagnosed clinically were caused by
herpes simplex virus (HSV).660 Chlamydia is another key part of
the differential diagnosis of acute adenoviral conjunctivitis.
SECTION 6
Pharyngoconjunctival Fever (PCF)
682
PCF is commonly caused by adenoviruses types 3, 4, and 7,

but it has been associated with types 1, 5, 6, and 14 as well,
having been isolated from conjunctiva, nasopharynx, and
feces.27,30,614,646648,661 PCF is an acute and highly infectious
illness characterized by fever, pharyngitis, acute follicular conjunctivitis, which may be hemorrhagic, and regional lymphoid
hyperplasia with tender enlarged preauricular adenopathy. It is
seen predominantly in the young and institutionalized people,
with epidemics occurring within families, schools, and military
organizations.
The incubation period after exposure is 512 days (most
commonly 8 days), at which time the patient experiences a
sudden or gradual onset of fever that may range from
100104F, lasting up to 10 days.27,646 Associated with the fever
are myalgia, malaise, and often, gastrointestinal disturbances.
The pharyngitis may be mild or quite painful and is typically a
reddened posterior oropharynx covered with glassy follicules
with nontender cervical lymphadenopathy. Transmission is by
contact with infected upper respiratory droplets or fomites or
through swimming pools. Communicability is 100% during the
rst few days to 0% by 1015 days after the onset of symptoms.
Initial ocular symptoms range from slight itching and
burning to marked irritation and tearing but little photophobia.
The lids are swollen within 48 h (Fig. 49.39). The conjunctivitis
presents initially as a diffuse hyperemia, being maximal in the
lower fornix but extending throughout the palpebral mucosa
and onto the bulbar conjunctiva. It may be sufciently boggy to
give a slightly gelatinous appearance to this tissue, and follicle
formation, although not invariable, is always more severe in the
lower lid than in the upper (Fig. 49.40). The discharge is serous,
and there may be slight crusting on the lids. In the absence of
a pseudomembrane mucopurulence is absent. Scrapings will
reveal a predominantly mononuclear exudate without characteristic features. If a membrane is present the predominant
inflammatory cell type is the PMN. The lower lid may be
somewhat tender to palpation and occasionally ecchymotic,
giving the patient the appearance of recent orbital trauma. PCF
is most frequently bilateral, with one eye having onset 13 days
prior to the second. In this event the second eye involved has
less severe disease than the rst.
A few days to a week after the onset of symptoms a punctate
keratitis may appear. This begins as small epithelial dots that
stain with fluorescein and progresses to combined epithelial and
subepithelial focal whitish lesions that may or may not stain
and nally to nonstaining subepithelial inltrates. Virus may be
cultured during the acute epithelial stage, but the stromal
inltrates are thought to be immune complexes against residual
viral antigen. These inltrates are usually scattered primarily in
the central corneal area. The entire illness is generally acute and
transient, resolving over a few days to 3 weeks, although the
subepithelial inltrates may last for several months and, if in
the visual axis, may cause glare or diminished vision. In the
absence of positive virus cultures taken during the rst 810
days, proof of diagnosis beyond clinical impression must rely on
paired blood specimens, the rst drawn about 1 week after the
onset of symptoms and the second to third week later. A fourfold or greater increase in humoral antibody to Ad, usually
measured by complement xation, indicates recent infection
with the organism. Other diagnostic tests for adenovirus have
been discussed earlier and include the ofce or small lab-oriented
Smartcycler II (Cepheid, Sunnyvale, CA) real-time PCR system,
FIGURE 49.39. Acute adenoviral conjunctivitis with classic blearyeyed appearance due to lid edema, conjunctival chemosis, and
hyperemia.
FIGURE 49.40. Acute adenoviral conjunctivitis with extensive

chemosis and hyperemia. The cornea is not yet involved.
and the more laboratory-oriented ELISA, electron microscopy,

and immunofluorescence testing.1,15,646
Epidemic Keratoconjunctivitis (EKC)

In relation to the eye the more serious of the adenoviral
illnesses is EKC.27,614,646,647,662664 This entity is generally
associated with Ad, types 8 and 19, but has also been reported
with multiple serotypes including types 24, 711, 14, 16, and
29.656,665,666 Serotypes other than 8 and 19 may produce a
similar clinical picture to the latter but do not have the
tendency to widespread epidemic.
Epidemic transmission within the community, as discussed
above, follows a pattern in families and other close groups
associated by personal contact and, particularly with type 19,
sexual activity. Harnet et al have reported that the peak
incidences of genital infection with Ad type 19 in both men and
women corresponded with those of eye infection with the same
virus in the general community.667 Clinically, the disease
usually attacks young adults during the fall and winter months
and is unilateral in two-thirds of patients. It produces few to no
systemic symptoms.27,614,648,655,663 This differentiates it from
PCF, which is usually bilateral and associated with fever and
FIGURE 49.41. Acute severe adenoviral conjunctivitis with true

inflammatory membranes and symblepharon formation.
FIGURE 49.43. Resolving adenoviral keratitis with anterior stromal

round whitish inltrates. These may occasionally merge to form gures
simulating focal HSV anterior stromal scars.
and may cause a two to three lines decrease in vision and

photophobia for months or even years (Fig. 49.43). Rarely, the
keratitis may have full-thickness inltrates with Descemet
folds, epithelial erosions, and iritis. These cases usually resolve
completely without consequence. In a few patients who do not
have terribly severe disease, however, the inltrates leave
anterior stromal scars with sharp borders and some slight
depression causing irregular astigmatism. These lesions may
last for many years and notably disturb vision.30,668
Diagnostic techniques for Ad have been discussed earlier
in this chapter but include the Smartcycler real-time PCR,
cytologic scrapings that reveal a mixed lymphocytic and PMN
inltrate and degenerated epithelial cells without inclusion
bodies. Viral cultures are positive 82% of the time if taken
during the rst week of the disease but by the end of the third
week less than 25% of the patients will continue to spill
virus.1,15,27,646 Alternative diagnostic techniques include paired
blood specimens with the rst drawn within 7 days of the onset
of symptoms and the second to third week later. A four-fold
CHAPTER 49
sore throat. The incubation period after exposure is about

8 days, at which time there may be the sudden onset of acute
tearing; foreign body sensation; marked conjunctival diffuse
hyperemia, occasionally with glassy chemosis; follicular and
papillary conjunctival response, with or without hemorrhage;
mild photophobia; and tender preauricular nodes. In those
patients in whom the disease goes on to bilaterality the second
eye becomes involved within 45 days but generally much less
severely than the rst eye, probably due to partial immune
protection of the host.
The conjunctival involvement may become so severe as to
develop marks of subconjunctival hemorrhaging, pseudomembranes or true membranes that are friable and may bleed,
or symblepharon formation that may scar the lid to the globe
(Fig. 49.41). The patients are moderately uncomfortable at this
time, but more severe pain comes with the development of
keratitis that occurs in ~80% of patients and begins around the
eighth day. This adenoviral keratitis is heralded by marked
discomfort, photophobia, lacrimation, and blepharospasm. These
symptoms persist until the acute epithelial phase subsides,
usually within a week or two, by which time the conjunctivitis
has also begun to resolve.
The keratitic disease is commonly divided into four
stages.27,30,614,648,664666 Stage 1 is a diffuse, ne, supercial
epithelial punctate keratitis caused by live virus. This moves
quickly to stage 2, which is a coalescence of these lesions to
staining focal punctate white epithelial lesions that stain with
fluorescein. Within 24 to 48 h these areas become combined
epithelial and subepithelial areas as stage 3, and over the next
few days the disease enters stage 4, which is characterized by
subepithelial white macular lesions that no longer stain with
fluorescein. The keratitis typically involves the central cornea
in clumps or rows of macular opacities but may reach the
periphery (Fig. 49.42). Occasionally, lesions coalesce to form
scallop-edged nummular opacities 12 mm in diameter that in
severe cases may mimic HSV keratitis. In the presence of a lid
membrane, a mechanical geographic ulcer also mimicking HSV
keratitis may develop.664,665 Electron microscopic study of
corneal specimens from patients in the acute epithelial stage
has demonstrated intact virus in the epithelial cells.663 This
area of virus replication is believed to establish the antigen
for the delayed hypersensitivity reaction that produces the
subepithelial inltrates of the later and more chronic stages.
In the vast majority of patients, the conjunctivitis has resolved
within 23 weeks but the subepithelial corneal inltrates reach
a maximal density in the third and fourth weeks of the infection
FIGURE 49.42. Marked adenoviral anterior stromal inltrates

developed 23 weeks after acute onset of infection and gradually
faded over 2 years.
683

or greater increase in humoral antibody to Ad is indicative of
recent infection. Other rapid laboratory diagnostic tests similar
to those discussed under PCF and earlier in this chapter under
Diagnostic Tests include immunofluorescence, ELISA, and
electron microscopy.
Nonspecic Follicular Conjunctivitis

NFC due to ocular adenovirus infection may occur in children
or adults and may be caused by many of the serotypes that also
cause EKC or PCF. As keratitis does not develop and the
conjunctivitis remains mild, ophthalmologists frequently do
not see these patients because they are handled by pediatricians
or family physicians. The clinical disease resolves without
residua over a 7- to 10-day period but also serves as the reservoir
of the adenovirus serotypes which may ultimately spark a more
severe widespread epidemic in the community.27,648 Both HSV
and Chlamydia should be considered in the differential diagnosis.
Chronic Adenoviral Keratoconjunctivitis

Although rare, chronic keratoconjunctivitis caused by a variety
of adenovirus serotypes has been reported. Because of its
atypical nature it is not frequently recognized. The syndrome is
characterized by a prolonged course of intermittent exacerbation of tearing, redness, and photophobia. The clinical history
will almost invariably reveal an episode of acute conjunctivitis
several months in the past. In the chronic state the cornea may
or may not have subepithelial opacities or active focal
supercial keratitis. Adenovirus types 2, 3, 4, and 19 have been
isolated as late as 12 months after the onset of chronic
keratoconjunctivitis, cases with either active epithelial keratitis,
recurrent conjunctivitis with subepithelial opacities, or chronic
recurrent papillary conjunctivitis. The total duration of disease
may well exceed 1 1/2 years.669671 As the conjunctival reaction
is primarily papillary, the clinician may be confused by the
absence of follicles in what is, in fact, a viral disease. The
diagnosis may be made by virus isolation from cornea or
conjunctiva or by testing for serotype-specic neutralizing and
hemagglutination-inhibition antibody in the absence of other
bacterial, viral, or toxic systemic illness that might mimic
chronic adenoviral keratoconjunctivitis. There appears to be no
relation to the early use of steroids and the establishment of
chronic adenoviral ocular disease.
SECTION 6
Treatment of Ocular Adenoviral Disease
684
The treatment of acute adenoviral ocular disease is still

controversial. The development of effective antivirals has
proven to be a complex task owing to the fact that multiple and
often genetically divergent Ad serotypes can cause similar
diseases. There is no licensed systemic or topical treatment in
the USA or Europe. However, many compounds have been
explored for activity against Ad, and some have been evaluated
clinically in either a topical setting for ocular disease or in the
setting of systemic treatment in the face of life-threatening
adenovirus infections.649
Most currently available antivirals are ineffective, although
the virus is sensitive in vitro to trifluridine.661,672,673 Cidofovir,
an antimetabolite, which is FDA approved for therapy of CMV
infection, also holds considerable hope as the rst effective
antiviral agent in ocular adenoviral disease.674677 In the
adenoviral rabbit model topical 0.5% cidofovir bid for 7 days
showed signicant antiviral activity against Ad serotypes 1, 5,
and 6.677
Antiviral prophylaxis to reduce community epidemics is a
key public health goal. Studies against Ad5 infection using 0.5
and 1% bid dosing conferred effective prophylaxis against viral
infection with 1% cidofovir drops eliminating all replication
after day zero.676 Additional reports reveal that topical 0.5%
cidofovir bid for 7 days had signicant antiviral activity against

Ad-1, Ad5, and Ad6 further supporting the broad-spectrum
activity of this drug.677 While the results of clinical studies have
yet to be reported, because of its FDA approval for CMV
retinitis, cidofovir has the potential for off label use or FDA
approval as a topical agent for adeno infection if future clinical
data are positive. As the subepithelial corneal opacities seen
in postacute disease are immunologic in origin and the
conjunctival disease is self-limited, cidofovir therapy does not
appear to have a role in adenoviral ocular sequelae. A possible
exception to this is proven chronic adenoviral conjunctivitis
(see above).
Topical steroid therapy may have a role in patients with
severe conjunctival reactions such as marked inflammation,
edema, pseudomembrane, or early symblepharon formation.
These drugs do provide dramatic symptomatic relief with
decreased inflammation and will also result in resolution of
corneal inltrates during the convalescent period. Steroids have
no benecial therapeutic effect, however, on the ultimate
clinical outcome. Laibson et al have shown that the subepithelial inltrates recur when steroids are discontinued and
that only time will ultimately resolve their presence.665 More
recent studies show that topical steroids reduce conjunctivitis
and subepithelial inltrates, regardless of which strength
streroid is used, 0.12% prednisolone, 0.1% fluoromethalone, or
1% rimexolone, they all increase viral replication and duration
of viral shedding in the Ad type 5 rabbit ocular model.678,679
This suggests that in a human population, topcial steroids may
increase the risk of viral transmission in community epidemics.
Antiviral prophylaxis with an agent such as cidofovir drops may,
in the future, allow safer use of steroids in those patients who
would most benet.
Without any treatment the corneal inltrates will almost
invariably recede spontaneously over a period of weeks, months,
or rarely years and vision will improve. Inltrates appear to be
the results of T-lymphocytes attracted to viral antigen in the
cornea. They may usually be suppressed by topical steroids, but
until the inciting antigen washes out over a several-month
period, the inltrates will simply reappear when lymphocyte
suppression is released on discontinuation of steroids. As noted
under EKC keratitis above in a few patients the inltrates leave
anterior stromal scars with sharp borders that may last for years
and notably disturb vision.30,668 The author has, however,
seen one patient who had severe, vision-debilitating (20/80 OU)
subepithelial inltrates 3 years after the acute disease; the
inltrates were unresponsive to steroids and non-steroidal
agents (DPL unpublished).
1. Antivirals are ineffective, with the possible exception of
cidofovir (Cidovir) drops but clinical trials have not been
completed.
2. Mild topical steroids (0.125% prednisolone, rimexolone 1%
or lotoprednol 0.2% or 0.5% q.d. tid for 14 week with
taper) relieve symptoms and inltrates temporarily; reserve
for severe cases only, i.e., photophobia, membrane or
pseudomembrane formation or visual loss from inltrates.
Contagion period will last longer than the usual 12 days
with the use of steroids acutely.
3. Cycloplegia prn iritis (rare).
4. Topical antibiotic ointment to lubricate and protect cornea
in presence of membranes.
5. Ice packs, antipyretics, and dark glasses as needed.
6. Prophylaxis against disease spread, by careful washing of
hands and instruments by medical personnel working on
the eye.
7. Infected medical and other personnel should terminate
their duties immediately for 12 days after onset of disease
when virus transmission becomes unlikely.
8. Infected patients should avoid oral or close contact with

family members or associates, not attend work or school
for the rst 12 days of illness, and use separate linens.
Wash hands frequently with antiseptic soap.
ACQUIRED IMMUNODEFICIENCY SYNDROME

(AIDS)
HIV infection affects 850 000 to 950 000 persons in the United
States alone and tens of millions world-wide.680 As the virus
replicates in CD4 T lymphocytes, the agent is transmitted by
blood, blood products, and other body fluids such as semen,
breast milk, saliva and tears, and urine. Recent estimates from
the World Health Organization (WHO) and Joint United
Nations Program on HIV/AIDS indicate that in the past 20
years more than 50 million people have been infected worldwide and 22 million have died of this disease.681 It is estimated
that 15 00020 000 new infections occur daily. The Centers for
Disease Control and Prevention estimates that 300 000
individuals in the United States are unaware that they are
currently infected with HIV and that over 23 000 health care
workers have AIDS.682,683 The epidemic increasingly affects
women, minorities, persons infected through heterosexual
contact, and the poor.684
The effect of HIV infection on the cellular immune system is
characterized by striking immunologic abnormalities in the
infected T-helper lymphocyte population. With progression
of disease there is reversal of the normal T-lymphocyte helper/
suppressor (T4/T8) ratio from a normal of 1.13.5 to levels far
below 1.0. There is also a resulting reduced lymphokine
production, inhibition of mitigen and antigen response,
depressed clonal expansion, and decreased ability to assist
B-lymphocytes in immunoglobulin production. The ocular
disease seen in AIDS is in part related to the nding that
B-lymphocytes in AIDS patients are polyclonally activated and
spontaneously secrete antibody. This results in elevated total
serum immunoglobulin levels, primarily IgG and IgA, resulting
in circulating immune complexes that ultimately infarct small
blood vessels. These same B-lymphocytes do not, however,
respond to the normal signals for proliferation and differentiation and do not usually respond to common immunizations or new antigens. Monocytes lose their chemotactic
migratory abilities and their ability to kill certain target cells
and to secrete interleukin-1. Natural killer cell immune
surveillance and virus-specic T-cytotoxic lymphocyte function
are also impaired. The progressive decline of the immune
system results in the eye, among other organs of the body, being
subjected to multiple opportunistic infections and malignant
diseases not seen prior to the AIDS epidemic, afflictions
normally held in check by an intact immune surveillance
system.285,685687
Ocular complications were seen in about 75% of AIDS patients
and have both diagnostic and prognostic signicance.688 They
are generally divided into four categories: retinal microangiopathy, opportunistic infection, tumors, and neuroophthalmological disease.689 Since the late 1990s widespread
use of HAART has markedly reduced the incidence of ocular
involvement in developed countries, especially CMV retinitis.
Unfortunately, immune recovery uveitis secondary to HAART
has emerged as a common vision-threatening condition. Visual
loss is minimized by early diagnosis and periocular steroids.296
Other common ndings are in the posterior segment and
include cotton-wool spots (immune complex infarctions),
retinal hemorrhages, Roths spots, microaneurysms, ischemic
maculopathy, retinal periphlebitis, and papilledema. Etiologic
agents of opportunistic infectious retinitis include HSV, VZV,
CMV, Cryptococcus, Toxoplasma, Candida, Mycobacterium
avium-intracellulare, atypical mycobacterial and extrapulmonary tuberculosis, Microsporidia, and potentially any other
infectious agent known. AIDS patients with ocular manifestations are often signicantly more immunosuppressed than
those without eye ndings.290,296,529,685,690692
Anterior segment ndings are less common and remain little
changed in the pre-HAART and HAART eras at 5.3% and
neuro-ophthalmic disease at 19%.693 Other ndings include
conjunctivitis, herpes-like ulcerations (HSV culture negative),
diffuse interstitial keratitis, Kaposis sarcoma, and opportunistic infection with HSV, VZV, and a variety of bacterial and
fungal organisms. The conjunctivitis seen in AIDS patients is
nonspecic with diffuse hyperemia, irritation, and tearing. This
is transient in nature and requires no specic therapy other
than ocular decongestants. Cases of severe bilateral ulcerative
keratitis similar to herpetic geographic ulceration have been
reported in AIDS patients. Immunofluorescence studies on
these corneas taken at autopsy, however, failed to reveal any
HSV antigen, thus raising the question whether this was truly
HSV or secondary to invasion by HIV. Brody et al have reported
21 AIDS patients with CMV retinitis, 90% of whom had
corneal endothelial deposits which were opaque, linear flecks
arranged in a reticular fashion. HIV+ patients with these
ndings have an 81% chance of active CMV retinitis and should
be evaluated for such.638 HZO, discussed above, is a well-known
presenting symptom in previously undiagnosed AIDS patients
as well as those with active disease.296,689691,693,694 The course
and management of HSV and HZO in AIDS patients is discussed in their respective sections above.
Other than the keratitis described previously, AIDS patients
may also develop diffuse punctate keratitis that is transient and
associated with marked anterior iridocyclitis that requires
intensive topical steroid therapy. Peripheral corneal ulceration
(PUK) similar to that seen in other immunologic disorders in
which circulating immune complexes are found has now been
reported in a patient with AIDS-related complex. This PUK was
felt to be due to high levels of circulating immune complexes
that created anterior segment microinfarctions affecting the
integrity of the peripheral cornea, a mechanism felt to be
operative in the AIDS/CMV retinitis.687,695,696 AIDS PUK
appears to be responsive to sealing with tissue adhesive and
placement of therapeutic soft contact lenses with prophylactic
antibiotic drops and cycloplegia. With neovascularization of the
stroma over several weeks the ulcer healed, dislodging the
glue and leaving behind an area with intact epithelium. Any
of the above conditions are worsened by the 18% incidence of
clinically signicant dry eye seen in these HIV patients.691
In the early 1980s pre-HAART era Kaposis sarcoma occurred
in ~9% of AIDS patients and could involve the eyelid or
conjunctiva.697700 By the late 1980s Kaposis sarcoma occurred
in ~20% of AIDS patients.701 However, by the 1990s HAART
era, the incidence dropped dramatically in the industrialized
countries.702 The malignancy is believed to be caused by the
human herpesvirus type 8 (HHV8) as this virus is found in 90%
of tissue samples taken from AIDS-associated Kaposis
sarcoma.703 Conjunctival involvement is more frequently found
in the inferior cul-de-sac. It may be missed without retraction
of the lower lid on examination. Sarcoma of the lid presents as
a bright red subconjunctival mass that may appear to be a
subconjunctival hemorrhage but is really a lymphomatous mass
(Fig. 49.44). The masses may be focal nodules or diffuse
inltrative lesions. Dugel and colleagues reported nding
multiple immature retrovirus particles in conjunctival nodules
in eyes with conjunctival Kaposis sarcoma but no particles
within the sarcoma tissue itself.698 Nakamura and co-workers
showed that HIV, type 1 or 2, releases a growth factor that
greatly enhances the growth of Kaposis sarcoma cells in
CHAPTER 49
685
Molluscum Contagiosum (MC)
SECTION 6
FIGURE 49.44. Kaposis sarcoma in an AIDS patient. Such sarcomas

typically appear in the lower or medial fornix and are soft and deep
purple-red.
culture.704 Elucidating the possible role of retroviruses in growth

of Kaposis sarcoma may give key information to the pathogenesis of this progressively more common ocular malignancy.
The most effective therapy for the sarcoma is treatment of the
undelying HIV disease itself. Local cryotherapy, radiotherapy,
and local excision have only transient efcacy.
HIV-1 has been isolated from multiple ocular tissues
including the tears, conjunctiva, cornea, iris, vitreous, and
retina.700,704710 This poses an epidemiologic concern not only
in that the eye may be an as-yet-unproven source of spread of
disease but also in the implications for corneal transplantation.
Eye banks now screen all potential donors for HIV-1, as well as
hepatitis virus, CreutzfeldJakob, and other quietly communicable diseases. With 25 000 corneal transplants being performed
annually in the United States, the risk of transplanting HIV via
donor corneal tissue despite negative serologic testing is a valid
concern to physicians and patients alike. In a pre-HAART era
mathematical model, Goode et al calculated that the risk of a
patient undergoing corneal transplantation receiving a donor
from an HIV-infected patient with negative serology was only
0.03%, a number which should be even lower in the HAART era
but not yet reported.711
The therapy of AIDS is discussed more extensively under
retinal disease elsewhere in this series. At present, there is still
no means of truly curing patients infected with HIV, but with
the advent of highly active antiretroviral therapy (HAART), a
combination of three or more anti-HIV agents, the natural
history of the disease has changed and the long-term survival
rate greatly increased.686,712 Patients live longer with higher
CD4 cell counts and little to no detectable virus load, thus
vastly improving the quality of life for these patients. Some have
been able to partially reconstitute their CD4 T-lymphocyte
immune system to help further in prevention of opportunistic
infection or malignancy.687,691
POX VIRUSES: MOLLUCUM CONTAGIOSUM,

VARIOLA, AND VACCINIA
686
The poxviruses, which include the now-extinct variola

(smallpox), its derivative vaccinia, and molluscum contagiosum,
are a group of large DNA viruses that share a common group
antigen. Their primary afnity is for the skin.30,648,713715
This cutaneous disease is generally limited exclusively to humans

although there are a few isolated reports of MC occurring in
birds, chimpanzees, dogs, and horses. Transmission requires
direct contact with infected hosts or contaminated fomites. The
virus is found worldwide but has a higher incidence in children,
sexually active adults, and those who are immmunodecient
whether by AIDS or atopy.614,716,717 The virus causes growth of
benign, self-limited papular eruptions of multiple, small, pink
umbilicated tumors on the skin and mucous membranes.
Ocular ndings are single or multiple umbilicated, wart-like
growths along the lid margins and a serous follicular conjunctivitis, punctate keratitis, superior corneal vascular pannus,
and cicatricial punctal occlusion, all the result of virus shed into
the tear lm.30,285,648,715,718 Lesions may also occur several
millimeters away from the lid margins yet still cause a follicular
conjunctivitis which is culture positive for MC.100,614,714,719 MC
lesions conned to cornea or conjunctiva alone are rare but not
unheard of. They are generally seen in patients with immune
dysfunction.717
The lesions themselves consist of acanthotic epidermis with
central craters lled with epithelial cells containing intracytoplasmic inclusion bodies. Immunohistochemical study of
biopsy specimens shows T-lymphocytes and a few macrophages
consistently present in the adjacent dermis and epidermis but
not inltrating the MC lesions themselves.720 There is cross
reactivity by T cell antibody to the MC bodies. Giemsa stains
reveal the viral particle inclusions 1224 h after infection.
Unlike many other poxviruses molluscum contagiosum virus
cannot be grown productively in tissue culture and does not
produce long-term cytopathic effects that can be passaged to
fresh tissue cultures.1,714
Treatment of MC
In the immunocompetent patient treatment may range from
nothing as the lesions often resolve spontaneously to simple
excision, or laser therapy. The keratoconjunctivitis resolves
with removal of the MC lesions. Successful treatment of MC
with pulsed dye laser over a 28 month period in 43 patients has
been reported.721 There were no complications noted, all 1250
lesions resolved, and 35% of patients had no new lesions after
two treatments.
Chemotherapy is another potential therapy. Cidofovir, the
broad-spectrum antiviral agent effective against herpes, and Ad
is also effective against pox viruses such as vaccinia, cowpox,
and monkey pox in animal models, against variola in vitro, and,
in human studies against molluscum contagiosum. It is
currently proposed as formulated in gel or cream form or as
intranasal aerosol or peroral as a lipid prodrug against these
latter infections as might be indicated.722,723
HIV patients are particularly severely affected by this virus
and tend to develop clusters of growth around the eyes which
recur despite repeated surgical treatment. Again, as with all
HIV-related eye disease the new combination of reverse transcriptase inhibitors and protease inhibitors (HAART) along with
topical antiviral such as cidofovir is probably the most effective
means of reversing all manifestations of this disease.296,686,712
Vaccinia
Vaccinia, an organism occasionally used in some laboratory
studies and, until recently, of little clinical concern, is a close
relative of variola virus (smallpox) and replaced cowpox in the
1800s as the live virus vaccine against variola. Of little clinical
concern until 2001 bioterrorism has now put vaccinia and its
potential ocular and other complications back on the list of
infections of concern.713,724726 Because the disease against
which it was used, variola, was considered extinct and the risk
of vaccination outweighed the benets, especially in children

and the immunocompromised, compulsory childhood smallpox
vaccination in the United States was stopped in 1972, vaccination of health care workers in the 1976, and of the military
in the early 1990s.343 However, due to growing concerns about
the potential use of smallpox as an agent of bioterrorism, in
2002 the US Department of Health and Human Services
reinstated smallpox vaccination (Dryvax, Wyeth) for US military
personnel and bioterrorism rst-responder units, followed by
primary and ancillary healthcare personnel.343,725,727,728
Adverse reactions to smallpox vaccination include vaccinia
necrosum, eczema vaccinatum, postvaccinial encephalitis,
generalized vaccinia, and accidental inoculation of other areas
of the body, including the eye.5,729732 Smallpox vaccination in
the preoutbreak setting is contraindicated in persons who have:
1) history of atopic dermatitis (eczema), 2) active acute, chronic,
or exfoliative skin conditions, 3) pregnant or lactating women
4) immunocompromised as a result of HIV, autoimmune
conditions, cancer, radiation treatment, immunosuppressive
medications, or other immunodeciencies, 5) allergy to
smallpox vaccine-component, 6) taking topical ocular steroid
medications, 7) aged < 18 years.724,733
With an overall complication rate of 0.004% and one death
per million vaccinations, vaccination with vaccinia is
considered a relatively safe and effective preventative against
smallpox.31,730 The complication rate is higher, however, than
with other vaccines and can be quite severe. Most complications occur in the person who has been vaccinated. Because it
is a live virus vaccine, however, the virus can be inadvertently
transmitted from the vaccinee to other sites in his/her own body
or to others in contact. The inoculation site can shed infectious
virus up to 21 days, until the dried scab detaches.5,729,731,732,734
Autoinoculation or cross-inoculation is the most common
route of spread to the eye from the patients vaccination site via
contaminated ngers to to his/her own face or to that of another
person in close contact.31,730,735 The incidence is low at one
case of ocular vaccinia per 40 000 vaccinations. Other routes of
transmission include health care workers carrying virus on their
clothes, or fomites from the nasopharynx of vaccinees. Secondary
cases manifest between eight and 18 days after exposure. A
primary accidental self-inoculation may appear between ve
and 11 days. Dissemination of vaccinial disease is expected to
be minimized by techniques not used in the previous
vaccination era: use of an occlusive dressing at the vaccination
site, and infection-control procedures including hand and
equipment hygiene and sterilization procedures.5,736
Filmore et al have reported the ocular complications in the
Department of Defense Smallpox Vaccination Program:
20022003. Of 450 293 vaccinations given, there were 16 conrmed or probable cases of ocular vaccinia, with an incidence of
3.6/100 000 inoculations. Of these cases, 12 (75%) were seen in
the vaccinees, and 4 (25%) in close contacts. Of the 12 selfinoculation cases, 7 (58.3%) were seen in individuals receiving
the vaccine for the rst time (primary vaccination) and 3
(25.0%) were seen in individuals previously vaccinated
(re-vaccination).737
bution. Vaccinia lesions can produce severe lid swelling and

periorbital erythema in a true orbital cellulitis. There is often
preauricular and/or submandibular lymphadenopathy. Eyelid
lesions can progress to scarring, madarosis, and be accompanied
by symblepharon formation.727,735,738740
The differential diagnosis of vaccinia lesions of the eyelid
or ocular adnexae includes molluscum contagiosum, keratoacanthoma, bacterial blepharitis, and herpes simplex or
varicella zoster virus infection.740 Diagnosis can be made by
obtaining scrapings and swabs of lesions and ocular discharge.
Smears of mucopurulent discharge from infected individuals
show numerous polymorphonuclear cells. Scrapings of vaccinial
lesions show epithelial cells containing Guarnieri bodies,
eosinophilic cytoplasmic inclusion bodies that are characteristic
of vaccinia. Viral culture is also used for diagnosis of
vaccinia.1,727 Rapid laboratory diagnosis of vaccinia infection
using real-time PCR may allow rapid analysis of autoclaved
suspensions, thereby limiting contact with infectious samples.741
Vaccinial keratitis results from live viral invasion of the
cornea, causing a supercial punctate keratitis that stains with
rose bengal early and fluorescein later in the clinical course.
Stromal involvement may consist of either subepithelial opacities
or deeper abscesses within the corneal stroma.5,727,735,739,742
Keratic precipitates may be present and there may be an
associated iritis.726 Disciform and necrotizing stromal keratitis
may occur up to 23 months after primary infection, and are
thought to be immune-mediated responses to virus and viral
antigens.743,744 Corneal perforation is possible if stromal
keratitis is left untreated.
The acute inflammatory disease lasts for 1014 days and
then heals with varying amounts of scarring. Corneal complications may be infectious epithelial keratitis or a late-onset
immune stromal keratitis. The latter may develop 23 months
after the original infection. Fortunately, vaccinial keratitis is
uncommon. In 1970, Ruben and Lane indicated that, not only
were the ocular complications of vaccination infrequent, they
were not notably vision threatening.735 The incidence of
keratitis was 1.2 cases per million primary vaccinations. Of 328
cases of ocular vaccinia, 70% were primary vaccinees, 58% of
whom were under 4 years of age. The time of onset ranged from
1 to 15 days postexposure with the majority being between
3 and 11 days. Only 22 cases involved the cornea and only 2%
of noncorneal cases had residual ocular damage, none severe.
Eighteen per cent (4/22) of the keratitis patients had residual
minor scarring. Treatment involved 336 of 348 patients
receiving vaccinia immune globulin (VIG) and 28 patients
received idoxuridine. It was not stated whether VIG was also
given to these patients or whether they had corneal involvement.
The authors concluded that ocular involvement was more
severe in primary vaccinees that in revaccinated patients, that
the residua in non-corneal cases was strikingly low, and that reexam of the corneal cases 5 years later revealed either no
residua, minor corneal scarring, and one case with a few ghost
vessels at the limbus and one with minor subepithelial opacity
which responded to steroid drops three times weekly.5
CHAPTER 49
Therapy of ocular vaccinia

Clinical disease
The most common form of ocular vaccinia is lid and
conjunctival involvement and is similar to that seen on the arm
at the site of the intentional vaccination: formation of vesicles
that progress to indurated pustules that umbilicate, scab and
scar leaving deep, depigmented pock marks in the skin (Fig.
49.44). Vaccinia lesions may be differentiated from those of
herpes simplex or zoster in that the latter two have a clear
vesicle stage which then scabs without going through a pustular
stage and, in the case of zoster, respect a dermatomal distri-
Although there are no reported masked, controlled human

studies on the efcacy of antivirals or vaccinia immune globulin
(VIG) on ocular vaccinia VIG is licensed for the treatment of
complications of vaccinia vaccination. This immunoglobulin
fraction of plasma from persons recently vaccinated with the
smallpox vaccine is currently administered IM, but a new formulation will allow intravenous administration. It is available
only through an Investigational New Drug (IND) protocol from
the Centers for Disease Control and Prevention in Atlanta
(Clinician Information Line at 877-554-4625).31,682,724,745 It
687
SECTION 6
688
has been effective for the treatment of eczema vaccinatum,

and some cases of progressive vaccinia. As noted by Ruben and
Lane, use of VIG appears to have no adverse effect even if used
in patients with keratitis. Further supportive anecdotal evidence
for IM VIG was reported by Kempe who treated two brothers
with identical cases of vaccinia keratitis.746 The brother treated
with IM VIG improved in 24 h and healed with no scar. The
untreated brother took 4 weeks to heal and was left with corneal
scarring.
Although no topical antiviral is FDA-approved for the treatment of ocular vaccinia, IDU, trifluridine, or cidofovir drops or
vidarabine ointment have been shown to be effective in animal
and uncontrolled human reports. Acyclovir is not effective
against vaccinia by in vitro antiviral screening.38,673,723,747752
Because of the limited clinical information available in
treating ocular vaccinia, the following recommendations have
been made by the Centers for Disease Control and Prevention
after consultation with an outside panel of corneal and external
disease and infectious disease specialists, based upon principles
routinely employed in the treatment of other viral diseases
of the ocular surface.31,343,724,745,737 Many of these would apply
to ocular variola as well should such a need arise. Clinical
experience after these guidelines were issued indicated that
trifluridine 9x/day was a more effective dose (DPL unpublished).
There has been no recent experience with vidarabine.
I. Blepharitis
1. Mild (few pustules, mild edema, no fever)
Consider prophylaxis of the conjunctiva and cornea:
Adults: Trifluridine (Viroptic) drops 9x/day for 2 weeks;
Children: Vidarabine 3% ointment tid for 2 weeks;
topical antibiotic to the conjunctiva.
2. Severe (pustules, edema, hyperemia, lymphadenopathy,
cellulitis, fever)
VIG 100 mg/kg IM; repeat in 48 h if not improved
Children: Vidarabine 3% ointment tid for 2 weeks
II. Conjunctivitis with or without blepharitis but without
keratitis
1. Mild (mild hyperemia and edema, no membranes or
focal lesions)
2. Severe: (marked hyperemia, edema, membranes, focal
lesions, lymphadenopathy, fever)
VIG 100 mg/kg; repeat in 48 h if not improved. Adults:
Trifluridine (Viroptic) drops 9x/day for 2 weeks;
Topical antibiotic to the conjunctiva
III. Keratitis with mild or no blepharitis or conjunctivitis
1. Mild (gray epitheliitis, no ulcer, no stromal haze or
inltrate)
Topical antibiotic gtts qid or ointment bid for 10 days+
or as needed.
2. Moderate (ulcer, but no stromal haze or inltrate)
Same treatment as mild, but use topical antibiotic qid
for 10 days or until ulcer healed
3. Severe (ulcer, stromal haze or inltrate)
Same treatment as for mild, but after epithelium is
healed (at about 7 days) add moderate-low dose steroid
to decrease immune reaction (e.g., prednisone 1/8% or
Vexol and taper slowly)
Mydriatic
IV. Keratitis with severe blepharitis and/or conjunctivitis
1. Trifluridine (Viroptic) drops 9x/day for 2 weeks;
Children: Vidarabine 3% ointment bid for 2 weeks
2. Topical antibiotic gtts qid or as indicated for 10+ days

3. Consider VIG IM 100 mg/kg one dose, consider repeat
in 48 h if no improvement
4. After epithelium is healed (at about 7 days) add
moderate-low dose steroid to decrease immune
reaction, if present (e.g., prednisone 1/8% or Vexol and
taper slowly)
Mydriatic
V. Iritis
1. Treat as for other eye conditions above
2. After corneal epithelium is healed add moderate-low
dose steroid to decrease immune reaction (e.g.,
prednisone 1/8% or Vexol and taper slowly)
Mydriatic.
Variola (Smallpox)
Variola, once considered an extinct threat to the world population has again come into prominance as a threat in the form
of a bioterrorist weapon.31,713,729 After an 812 day incubation
period systemic smallpox has a two to three day prodrome of
flu-like illness followed by abrupt onset of viral shedding and
severe illness with high fever, myalgia, headache, prostration,
and often severe abdominal pain. A maculopapular rash appears
in one to two days in the oropharynx, face, and arms and
spreads centrally, rapidly becoming vesicular and then pustular
sometimes associated with hemorrhage. The lesions are
numerous, rm, elevated and involve the palms and soles. Most
deaths occur during the second week of illness.31,713,726,729 The
lesions scab and fall off over 3 weeks leaving deep, depigmented,
pitted scars. Case-fatality rates reach 2035% among
unvaccinated individuals.
Serious ocular complications, including eyelid and conjunctival infection, corneal ulceration, disciform keratitis, iritis,
optic neuritis, and blindness may occur. About 59% of patients
with smallpox develop ocular complications.753,754 About 5 days
after the onset of clinical disease an exanthematous watery
conjunctivitis may develop and frequently clear without
complication. In a few patients, however, pustules appeared on
the bulbar conjunctiva (Fig. 49.45). These are painful with great
FIGURE 49.45. Umbilicated pustules of ocular vaccinia. Kissing

lesion of upper and lower lid of right eye.
From Pavan-Langston, D Ocular Viral Diseases. In Antiviral Agents and Viral
Diseases of Man, ed. Galasso, G, Merigan, T, Buchanan, R. Raven Press,
New York, 253304, 1979.
Treatment
Specic treatment is currently not established. Suspect cases
should be placed in a negative-pressure room, if available and
vaccinated as soon as possible, especially if the illness is in early
stage.713,753,754 Adequate hydration and nutrition are important
as much fluid is lost through fever and weeping lesions. There
are no data showing that prophylaxis or treatment of active
disease with VIG has any effect.5,713 Guidelines given in the
section on Therapy of Ocular Vaccinia above should be followed.
One promising but unproven treatment, however, is systemic
and topical cidofovir. As noted above under Molluscum Contagiosum this broad-spectrum antiherpes agent is also effective
against poxviruses such as vaccinia, cowpox, and monkey
pox in animal models, and against variola in vitro.722,755,756
Isothiazole thiosemicarbazone given soon after documented
exposure may prevent death but not disgurement.757
Penicillinase-resistant antimicrobial agents should be used if
the skin lesions are secondarily infected or if infection is near
or involves the eyes. Daily cleansing of the eyes and lids is
important ito minimize scarring.713
PAPILLOMA VIRUSES (HUMAN PAPILLOMA

VIRUS, HPV)
Neoplastic and Nonneoplastic Relationships5
Of the 70 different strains of HPV, the ones most associated
with ocular disease are HPVs 6, 11, 16, and 18 which infect
mucosal areas and HPVs 14 which infect skin sites.758 This
DNA virus is ubiquitous and spread by contact. After initial
inoculation the virus grows to create a lesion that may lead to
further autoinoculation causing a multicentric infection.
Conjunctival intraepithelial neoplasia (CIN) is of particular
interest. In studies on 10 consecutive patients who underwent
excision of CIN and ve non-CIN control patients, reverse
transcriptase in situ polymerase chain reaction (PCR) technique
was used to search for the presence of HPV mRNA.759 HPV 16
DNA and mRNA were found in ve CIN specimens, and HPV
18 DNA and mRNA were present in the remaining ve CIN
specimens. Further, in each of the CIN specimens, 2040% of
the dysplastic cells expressed the HPV E6 region. Niether HPV
DNA nor mRNA were detected in any of the control specimens
or in any of the clinically uninvolved conjunctival specimens
(P < 0.001).
HPV may also be associated with a variety of other conjunctival conditions. In a PCR study of 96 neoplastic and
nonneoplastic lesions and 19 conjunctival samples free from
overt disease HPV types 16 and 18 DNA were identied in 57%
of in situ squamous cell carcinoma, in 55% of invasive squamous
cell carcinoma, in 20% of climatic droplet keratopathy, in 35%
of scarred corneas, and in 32% of normal conjunctival tissue
obtained during routine cataract extractions.760 It is this type of
data that makes the exact relationship between viral papillomas
and neoplastic transformation unclear.
Clinical Disease
The papillomas are fleshy, pinkish-red, shiny, and elevated with
prominent internal blood vessels present as multiple vascular
loops within a brovascular core.761 The lesions may be broad
and low (sessile) or pedunculated on a stalk and located on the
palpebral, forniceal, or bulbar conjunctiva, on the lacrimal

puncta or caruncle, or within the canaliculus. Limbal or coreal
involvement is unusual and may be associated with brovascular
pannus or punctate keratitis. In HIV+ or other immunocompromised patients lesions may be bilateral, multiple, and
large. Symptoms may vary from none to irritation, foreign body
sensation, tearing, itching, mucoid discharge, photophobia, and
blurred vision.
Diagnosis is made by clinical observation and histopathology
of excised specimens.
Treatment
Many viral papillomas regress spontaneously over 12 years
making observation of asymptomatic or mildly symptomatic
patients a good option. Common therapy, however, for those
who need treatment of conjunctival lesions is surgical excision,
cryotherapy, or both. Unfortunately, with either of these
techniques alone, seeding may take place during the procedure(s)
resulting in recurrence of infection. The most effective method
to prevent this is a combination of technique by freezing the
entire lesion and gently lifting it slightly from the surface to
permit excision of the papilloma, stalk and base plus some
surrounding normal tissue. Then apply double freeze-thaw to
the base of the area excised. Electrodesiccation or heat cautery
is useful for lid papillomas. In the former, an electric needle is
inserted in the lesion and heat applied until the tissue begins to
bubble. The lesion is then curretted. With heat cautery, the
lesion is excised and cautery applied to the base.
Because of equivocal success in treating ocular papillomatosis
chemotherapy is now of interest. Initial clinical trials point to
the efcacy of topical cidofovir (HPMPC 1% ointment) in the
treatment of pharyngeal, laryngeal, and anogenital HPV
infections.38,762,763 HPMPC (cidofovir), among others, is now
being pursued in the topical treatment of the papilloma viruses.
The recent FDA approval of a cervical cancer vaccine against
papilloma virus 16 may offer some future aid in preventing
ocular neoplastic disease, especially by the two highest risk
cervical, laryngeal, and ocular strains, HPV 16 (66.7%) and
HPV 18 (19.4%).764,765
PARAMYXOVIRUS OCULAR DISEASE

The paramyxoviruses include the RNA viruses of measles
(rubeola), mumps, and Newcastle disease.674,766770 Measles
infection of the eye results in an acute catarrhal conjunctivitis,
SPK, and occasionally Kopliks spots on the conjunctiva or
semilunar fold. There is often severe photophobia, which is
self-limited and leaves no visual decit. Rarely an immune
interstitial keratitis may occur. In their report of a measles
epidemic in 61 army recruits Kayikcioglu et al found 65% had
measles conjunctivitis with bulbar and tarsal conjunctival
hyperemia, 8% had increased mucous secretion, 57% had
supercial punctate corneal epithelial and subepithelial lesions
which stained with fluorescein, and 22% had subconjunctival
hemorrhages, half of whom OU. Diclofenac treatment yielded
no diffference in the healing time between the two eyes, about
4 days (P = 0.75).771 Recently, it has been reported that during
the rst 2 weeks of infection topical ketorolac 0.5 and 0.1%
indomethacin are signicantly more effective than articial
tears in decreasing conjunctival hyperemia but had no effect on
symptoms of burning, foreign body, or photophobia.772
In immunocompromised patients or in patients living in
developing countries where there are nutritional deciencies
measles keratitis may be a blinding disease. A generalized
measles infection may result in severe keratitis, keratomalasia,
pneumonia, myocarditis, encephalitis, and death. Tuberculosis
has been known to reactivate following measles infection in
CHAPTER 49
inflammatory reaction and purulent discharge, often extending

to the cornea, causing inflammation, scarring, and even perforation with loss of the eye. Bacterial infection is not infrequent
in these corneal ulcers and contributes to the ocular damage if
untreated.
Diagnosis is usually clinical in time of epidemic but objective
tests are similar to those of vaccinia above.
689

developing countries.767 Diagnosis is made by clinical impression and conrmed by isolation of virus from the throat, blood,
or mucous membranes or by determination of humoral antibody response. There is no specic treatment and no cure.
Attenuated measles vaccine is highly effective, however, in
producing adequate prophylaxis against this potentially lethal
illness.
Mumps virus may involve the ocular adnexae, causing a
severe dacryoadenitis, sudden orbital pain, and swelling with a
lacrimal fossa mass. A catarrhal conjunctivitis is frequent, and
a punctate epithelial keratitis or severe stromal keratitis with
decrease in vision may develop along with severe photophobia
and lacrimation but amazingly little pain. The stromal disciform keratitis is often unilateral and may begin within a week
of onset of the epithelial disease. Despite marked stromal
edema the disease ultimately resolves spontaneously.769,770
Diagnosis is made based on clinical impression, by isolation of
the virus from saliva and tears, and by humoral antibody
detection.
Specic therapy is not available, although cycloplegic agents
may relieve the discomfort of ciliary spasm. Mumps may also
induce episcleritis, scleritis, uveitis, and a variety of posterior
segment inflammatory lesions and extraocular muscle palsies.
Severe intraocular inflammatory disease is usually seen only in
immunosuppressed patients. Topical steroids and cycloplegia
may be useful, but systemic steroids may only serve to disseminate disease.
Newcastle disease virus causes a limited infection seen primarily in poultry workers and laboratory technicians. Clinical
ndings are a unilateral follicular conjunctivitis with mild tearing
and preauricular adenopathy. There may be a ne punctate
epithelial keratitis with occasional subepithelial inltrates. The
illness is self-limiting with no sequelae and does not require
therapy.766
FIGURE 49.46. Acute hemorrhagic conjunctivitis showing a solid

sheet of subconjunctival blood under the superior conjunctiva.
Punctate keratitis developed 4 days later.
The disease resolves spontaneously within 24 days and is

completely gone within 10 days of onset. Therapy is purely
supportive with bedrest, analgesics, and cool compresses.
Antibiotics and steroids have no established role. Because of the
epidemic nature of this illness work continues toward its prevention by vaccine. Langford et al have reported that adjuvant
MDP-induced conjunctivitis increased bloodconjunctival
barrier (BCB) permeability and anti-EV70-neutralizing activity
in tears of seropositive rabbits. This suggests that immunization
with inactivated EV70 could provide systemic and ocular
protection during natural EV70 infection.778
SECTION 6
PICORNAVIRUSES
690
Acute hemorrhagic conjunctivitis (AHC) is a highly contagious

ocular infection caused by the enteroviruses (EVs), members of
the picornavirus family. The EVs include several well-known
RNA organisms: poliovirus, coxsackievirus A and B, and the
echoviruses. The specic EV most commonly associated with
AHC is EV70, but reports from the Far East indicate that other
picornaviruses not cross-reacting with known EVs may also
induce the disease.614,648,773775 Because the virus is difcult to
isolate, reverse transcription-PCR has been used successfully to
detect EV70 specically in patients with AHC who were
culture-negative.776
AHC may be distinguished from other external ocular
infections by its proclivity for widespread epidemic proportions
and its clinical presentation. It may, during times of epidemic,
afflict from tens of millions of people in densely populated
humid areas of the Far East to several hundred people in
Western countries.777
The incubation period following exposure is 12 days, followed
by the sudden onset of ocular foreign body sensation, itching,
photophobia, profuse tearing, and lid edema. Progression of
disease is rapid over the ensuing 24 h with development
of hyperemic conjunctival chemosis and characteristic subconjunctival petecchial or sheet-like hemorrhages that appear
as concentric ridges encircling the corneal limbus (Fig. 49.46).
There is frequently an associated supercial punctate keratitis
and preauricular adenopathy, and the entire clinical picture may
initially resemble acute adenoviral keratoconjunctivitis.
Systemic symptoms may or may not be present and include
malaise, myalgia, and upper respiratory tract symptoms similar
to influenza.614,648,775 Rarely, there may be a radiculomyelitis.
TOGAVIRUSES
The togaviruses include the agents of rubella (German measles)
and the arbovirus group B infections of yellow fever, dengue,
and sandfly fever. All viruses of this group may cause
conjunctival hyperemia, lid edema, photophobia, and lacrimation. Clinical disease resolves spontaneously and does not
require therapy.
Congenital rubella syndrome is the result of maternal
infection with this virus during the rst or second trimester
of pregnancy. Ocular ndings include corneal scarring, keratoconus, cataracts, glaucoma, retinopathy, microcornea, microphthalmia, iris hypoplasia, and subretinal neovascularization.
The incidence of keratoconus in these patients is much higher
than in the general population, and the patients may develop
full-blown hydrops.648,779781
Acquired German measles may also produce ocular disease in
70% of children and adults with this viral exanthem. The
incubation period after exposure is 57 days, at which time an
onset of a mild catarrhal or follicular conjunctivitis frequently
occurs. In 2% of patients a ne punctate epithelial keratitis
associated with photophobia and tearing will develop. The
corneal lesions are central and have their onset ~1 week after
appearance of the rash.782,783 Late ocular disease has not been
reported, and the acute disease is self-limited, requiring no
therapy.
SUMMARY
Of all body organs, the eye is perhaps the most frequently
affected by the ravages of local or systemic infectious disease. It

is also an organ amenable to the development of new diagnostic
and therapeutic technologies. The eye remains the premier
testing ground for many drugs under evaluation today in both
experimental and human clinical studies. The development of
new animal models of human ocular disease such as HSV, VZV,
and adenovirus has provided invaluable information as to
disease mechanism, evolution, and management. With the

development of more specic topical and systemic antiviral
agents and the multiple viral infections manifested by ocular
disease, the prospects for new effective therapies pertinent both
to this and to other organ systems remains highly promising in
the near and distant future.
1. White DFF (ed): Laboratory diagnosis of

viral disease. In: San Diego, CA: Academic
Press; 1994:191218
2. Baron E, Peterson L, Finegold S (eds):
Laboratory methods in basic virology. In:
St Louis, MO: Mosby-Year Book; 1994;
634688.
3. Hooks J: Ocular virology. In: Tabbara K,
Hyndiuk R, eds. Boston, MA: Little Brown
and Co; 1996:99114.
4. Kinchington P: Virology. In: Smolin G, Thoft
R, eds. Boston, MA: Little Brown and Co;
1994:169182.
ocular anterior segment: basic science and
clinical disease. In: Foster S, Azar D,
Dohlman C, eds. Philadelphia, PA:
Lippincott, Williams & Wilkins;
2005:297397.
6. Fields B, Knipe D, Howley P: Virology. 3rd
edn. Philadelphia, PA: Lippincott, Williams
& Wilkins; 1996.
7. Tei M, Nishida K, Kinoshita S: Polymerase
chain reaction detection of herpes simplex
virus in tear fluid from atypical herpetic
epithelial keratitis after penetrating
122:732735.
8. Yamamoto S, Shimomura Y, Kinoshita S,
et al: Detection of herpes simplex virus
DNA in human tear lm by the polymerase
chain reaction. Am J Ophthalmol 1994;
117:160163.
9. Yamashita T: [Diagnostic tests: Varicellazoster virus], Nippon Rinsho 2005;
63(Suppl 7):294296.
10. Alvarado J, Underwood J, Green R, et al:
Detection of herpes simplex viral DNA in
the iridocorneal endothelial syndrome. Arch
Ophthalmol 1994; 112:16101609.
11. Yamamoto S, Pavan-Langston D, Tada R, et
al: Possible role of herpes simplex virus in
the origin of Posner-Schlossman syndrome.
Am J Ophthalmol 1995; 119:796798.
12. Kaye SB, Baker K, Bonshek R, et al:
Human herpesviruses in the cornea.
13. Kaye SB, Lynas C, Patterson A, et al:
Evidence for herpes simplex viral latency in
the human cornea. Br J Ophthalmol 1991;
75:195200.
14. McIver CJ, Jacques CF, Chow SS, et al:
Development of multiplex PCRs for
detection of common viral pathogens and
agents of congenital infections. J Clin
Microbiol 2005; 43:51025110.
15. Kowalski R, Thompson P, Kinchington P,
et al: Evaluation of the Smartcycler II
system for the real-time detection of
viruses and chlamydia from ocular
specimens. Arch Ophthalmol 2006;
in press.
16. Gillo V, Fedorko D Witebsky F: The clinician
and the microbiology laboratory. In:
Mandell G, Bennett J, Dolin R, eds.
Principles and practices of infectious
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
diseases. Philadelphia, PA: Elsevier;

2005:203242.
Gleaves C, Hondinka R, Johnston S, et al:
Cumtech 15A. In: Baron E, ed. Washington,
DC: American Public Health Associan;
1994:3856.
Liesegang T: Ocular Virology. In: Albert D,
Jakobiec F, eds. Philadelphia, PA: WB
Saunders; 2000:171198.
Britt W: Human cytomegalovirus. In:
Goldman L, Ausiello D, eds. Philadelphia,
PA: Saunders; 2004:19931995.
Kowalski R, Gordon Y: Evaluation of
immunologic tests for the detection of
ocular herpes simplex virus.
Lee S, Gregory A, Storch M, et al:
Comparative laboratory diagnosis of
experimental herpes simplex keratitis. Am J
Papesch M, Watkins R: Epstein-Barr virus
infectious mononucleosis. Clin Otolaryngol
Allied Sci 2001; 26:38.
Kieff E: Infectious mononucleosis: EpsteinBarr infection. In: Goldman L, Ausiello D,
eds. Philadelphia, PA: Saunders;
19951999, 2004.
Matoba A: Epstein-Barr virus diseases. In:
Pepose J, Holland G, Wilhelmus K, eds.
St Louis, MO: Mosby; 1996:958969.
Hazeltine W: Silent HIV infections. N Engl J
Med 1989; 320:1487.
Liang TS, Erbelding E, Jacob CA, et al:
Rapid HIV testing of clients of a mobile
STD/HIV clinic. AIDS Patient Care STDS
2005; 19:253257.
Gordon Y, Aoki K, Kinchington P:
Adenovirus keratoconjunctivitis. In: Pepose
J, Holland G, Wilhelmus K, eds. St Louis,
MO: Mosby; 1996:877894.
Behbehani A: Poxviruses. In: Lennette E,
Lennette D, Lennette E, eds. Washington,
1995:511520.
Claoue C, De Cock R: The spectrum of
herpes simplex virus disease of the anterior
segment in the 1990s. Acta Ophthalmol
Scand 1996; 74:407410.
Pavan-Langston D: Herpes simplex,
adenovirus, Epstein Barr virus, pox. in
Major Ocular Viral Infections: In: Galasso G,
Whitley R, Merigan R, eds. New York:
Raven Press; 1997:187229.
Semba R: The ocular complications of
smallpox and smallpox immunization. Arch
Ophthalmol 2003; 121:715719.
Coleman V, Thygeson P, Dawson C, et al:
Isolation of virus from herpetic keratitis:
influence of idoxuridine on isolation rate.
Cohen P: Tests for detcting herpes simplex
virus and varicella zoster infections.
Dermatol Clin 1994; 12:5168.
Pavan-Langston D, Yamamoto S, Dunkel E:
Delayed herpes zoster pseudodendrites.
35. Mahalingam R, Wellish M, Wolfe W, et al:

Latent varicella-zoster viral DNA in human
trigeminal and thoracic ganglia. N Engl J
Med 1990; 323:627631.
36. Mahalingam R, Wellish M, Lederer D, et al:
Quantitation of latent varicella-zoster virus
DNA in human trigeminal ganglia by
polymerase chain reaction. J Virol 1993;
67:23812384.
37. Mann L, Woods G: Rapid diagnosis of viral
pathogens. Clin Lab Med 1995;
15:389405.
38. DeClercq E: Recent highlights in the
development of new antiviral drugs. Curr
Opin Microbiol 2005; 8:552560.
39. US Pharmcopeia Valtrex. In: Greenwood
Village. Colorado: Thomson Micromedex
Inc; 2005:15011508.
40. US Pharmcopeia Famvir. In: Greenwood
Village. Colorado: Thomson Micromedex
Inc; 2005:13901392.
41. US Pharmcopeia Acyclovir. In:
Greenwood City. Colorado: Thomson
Micromedex Inc; 2005:2432.
42. Handbook of Antimicrobial Therapy. 17th
edn. In: The Medical Letter on Drugs
and Therapeutics. New Rochelle: Pages;
2005.
43. Pavan-Langston D: Ocular pharmacology
of antiviral drugs. In: Jaeger E, Tasman W,
Wilhelmus K, eds. Phildelphia, PA:
Lippincott, Williams & Wilkins; 2004:124.
44. Foster CS: Nonsteroidal anti-inflammatory
and immunosuppressive agents. In:
Lamberts D, Potter D, eds. Boston, MA:
Little, Brown; 1987:173.
45. Kleinert R, Palmer M: Antiinflammatory
agents. In: Tabbara K, Hyndiuk R, eds.
46. Pavan-Langston D, Dunkel E:
Corticosteroids, immunosuppressive
agents, and non-steroidal anti-inflammatory
drugs. In: Pavan-Langston D, Dunkel E,
eds. Handbook of Ocular Drug Therapy.
47. Lass J, Berman M, Campbell R, et al:
Treatment of experimental herpetic
interstitial keratitis with
mexroxyprogesterone. Arch Ophthalmol
1980; 98:520525.
48. Wilhelmus KR, Dawson CR, Barron BA,
et al: Risk factors for herpes simplex virus
epithelial keratitis recurring during
treatment of stromal keratitis or
iridocyclitis. Herpetic Eye Disease Study
Group. Br J Ophthalmol 1996; 80:696672.
49. Abel R, M T, Forstot S: Calcic band
keratopathy. Arch Ophthalmol 1991;
109:1072.
50. Pavan-Langston D: Herpetic infections.
In: Smolin G, Thoft R, eds. Boston, MA:
51. Gunduz K, Ozdemir O: Topical cyclosporin
as an adjunct to topical acyclovir treatment
in herpetic stromal keratitis. Ophthalmic
Res 1997; 29:405408.
CHAPTER 49
REFERENCES
691
SECTION 6
692
52. Field A, Gottsch J: Persisting epithelial

herpes simplex keratitis while on
cyclosporin-A ointment. ANZJ
Ophthalmol1995; 23:333334.
53. Heiligenhaus A, Steuhl KP: Treatment of
HSV-1 stromal keratitis with topical
cyclosporin A: a pilot study. Graefes Arch
54. Remeijer L, Osterhaus A, Verjans G: Human
herpes simplex virus keratitis: the
pathogenesis revisited. Ocul Immunol
Inflamm 2004; 12:255285.
55. Cohrs RJ, Laguardia JJ, Gilden D:
Distribution of latent herpes simplex virus
type-1 and varicella zoster virus DNA in
human trigeminal ganglia. Virus Genes
2005; 31:223227.
56. Labetoulle M, Auquier P, Conrad H, et al:
Incidence of herpes simplex virus keratitis in
France. Ophthalmology 2005; 112:888895.
57. Nahmias AJ, Alford CA, Korones SB:
Infection of the newborn with herpesvirus
hominis. Adv Pediatr 1970; 17:185226.
58. Nahmias AJ, Roizman B: Infection with
herpes-simplex viruses 1 and 2.3. N Engl J
Med 1973; 289:781789.
59. Pepose JS, Keadle TL, Morrison LA: Ocular
herpes simplex: changing epidemiology,
emerging disease patterns, and the
potential of vaccine prevention and therapy.
Am J Ophthalmol 2006; 141:547557.
60. Lafferty W, Coombs R, Benedetti J, et al:
Recurrences after oral and genital herpes
simplex viral infection. N Engl J Med 1987;
316:1444.
61. Wilhelmus K: The epidemiology of ocular
infection. In: Jaeger E, Tasman W, eds.
Philadelphia, PA: Lippencott-Raven; 1998.
62. Butler TK, Spencer NA, Chan CC, et al:
Infective keratitis in older patients: a 4 year
review, 19982002. Br J Ophthalmol 2005;
89:591596.
63. Remeijer L, Maertzdorf J, Buitenwerf J,
et al: Corneal herpes simplex virus type 1
superinfection in patients with recrudescent
herpetic keratitis. Invest Ophthalmol Vis Sci
2002; 43:358363.
64. Chong EM, Wilhelmus KR, Matoba AY, et al:
Herpes simplex virus keratitis in children.
Am J Ophthalmol 2004; 138:474475.
65. Kaye S, Madan N, Dowd T, et al: Ocular
herpes virus shedding. Br J Ophthalmol
1990; 74:114.
66. Coyle P, Sibony P: Viral antibodies in
normal tears. Invest Ophthalmol Vis Sci
1988; 29:1552.
67. Kaufman HE, Azcuy AM, Varnell ED, et al:
HSV-1 DNA in tears and saliva of normal
adults. Invest Ophthalmol Vis Sci 2005;
46:241247.
68. Liesegang TJ, Melton LJ 3rd, Daly PJ,
et al: Epidemiology of ocular herpes
simplex. Incidence in Rochester, Minn,
19501982. Arch Ophthalmol 1989;
107:11551159.
69. Bell D, Holman R, Pavan-Langston D:
Epidemiologic aspect of herpes simplex
keratitis. Ann Ophthalmol 1982; 14:421.
70. Shuster J, Kaufman H, Nesburn A:
Statistical analysis of the rate of recurrence
of herpes virus ocular epithelial disease.
71. HEDS: Predictors of recurrent herpes
simplex virus keratitis. Herpetic Eye
Disease Study Group. Cornea 2001;
20:123128.
72. HEDS: Psychological stress and other
potential triggers for recurrences of herpes
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
simplex virus eye infections. Herpetic Eye

Disease Study Group. Arch Ophthalmol
2000; 118:16171625.
Koelle DM, Ghiasi H: Prospects for
developing an effective vaccine against
ocular herpes simplex virus infection. Curr
Eye Res 2005; 30:929942.
Nesburn AB, Ramos TV, Zhu X, et al: Local
and systemic B cell and Th1 responses
induced following ocular mucosal delivery
of multiple epitopes of herpes simplex virus
type 1 glycoprotein D together with
cytosine-phosphate-guanine adjuvant.
Vaccine 2005; 23:873883.
Cunningham AL, Taylor R, Taylor J, et al:
Prevalence of infection with herpes simplex
virus types 1 and 2 in Australia: a
nationwide population based survey, Sex
Transm Infect 2006; 82:164168.
Buddingh GJ, Schrum DI, Lanier JC, et al:
Studies of the natural history of herpes
simplex infections. Pediatrics 1953;
11:595610.
Centifanto-Fitzgerald Y, et al: Ocular
disease pattern induced by HSV is
genetically determined by specic region of
viral DNA. J Exp Med 1982; 155:475.
Hammer H, Dobozy A: Cell mediated
immunity to herpes virus type I in patients
with recurrent corneal herpes simplex. Acta
Openshaw H, Asher L, Wohlenberg C, et al:
Acute and latent infection of sensory
ganglia with herpes simplex virus: immune
control and virus reactivation. J Gen Virol
1979; 44:205.
Azazi M, Gunilla M, Forsgren M: Late
ophthalmologic manifestations of neonatal
herpes simplex virus infection. Am J
Kimberlin DW, Lin CY, Jacobs RF, et al:
Natural history of neonatal herpes simplex
virus infections in the acyclovir era.
Pediatrics 2001; 108:223229.
Freij BJ: Management of neonatal herpes
simplex virus infections. Indian J Pediatr
2004; 71:921926.
Kimberlin DW, Whitley RJ: Neonatal
herpes: what have we learned. Semin
Pediatr Infect Dis 2005; 16:716.
Hutchison D, Smith R, Haughton D:
Congenital herpetic keratitis. Arch
Nahmias AJ, Visintine AM, Caldwell DR,
et al: Eye infections with herpes simplex
viruses in neonates. Surv Ophthalmol 1976;
21:100105.
Krolczyk S, Pacheco E, Valencia P, et al:
Opsoclonus: an early sign of neonatal
herpes encephalitis. J Child Neurol 2003;
18:356358.
Pavan-Langston D: Living with Herpes.
In: Doubleday, ed. New York;1983.
Donoval BA, Passaro DJ, Klausner JD:
The public health imperative for a neonatal
herpes simplex virus infection surveillance
system. Sex Transm Dis 2006; 33:170174.
Schwartz G, Holland G: Oral acyclovir for
the management of herpes simplex
keratitis in children. Ophthalmology 2000;
107:278282.
Kimberlin DW, Lin CY, Jacobs RF, et al:
Safety and efcacy of high-dose
intravenous acyclovir in the management of
neonatal herpes simplex virus infections.
Pediatrics 2001; 108:230238.
Chang T, Brewer L, Hooper P: Primary
herpes simplex iridocyclitis with iris
atrophy. Arch Ophthalmol 1993; 111:2526.

92. D P-L: Major ocular viral infections: Herpes
simplex, adenovirus, Epstein Barr virus,
pox. In: Galasso G Mt, Whitley R, eds.
New York: Raven Press; 1997:187228.
93. Yao Y, Inoue Y, Shimomura Y, et al:
Primary herpes simplex virus infection with
geographic conjunctival ulceration. Am J
Ophthalmol 1994; 118:670671.
94. Baringer J, Swoveland P: Recovery of
Herpes simplex virus from human
trigeminal ganglions. N Engl J Med 1973;
288:648.
95. Bastian F, Rabson A, Yee C, et al:
Herpesvirus hominus: isolation from human
trigeminal ganglion. Science 1972; 178:306.
96. Remeijer L, Maertzdorf J, Doornenbal P,
et al: Herpes simplex virus 1 transmission
through corneal transplantation. Lancet
2001; 357:442.
97. Sabbaga E, Pavan-Langston D, Bean K, et al:
Detection of HSV nucleic acid sequences
in the cornea during acute and latent
ocular disease. Exp Eye Res 1988; 47:545.
98. Pavan-Langston D: Herpes simplex virus
ocular infections: current concepts of
acute, latent and reactivated disease. Trans
99. Ohashi Y, Yamamoto S, Nishida K, et al:
Demonstration of herpes simplex virus
DNA in idiopathic corneal endotheliopathy.
Am J Ophthalmol 1991; 112:419423.
100. Asbell P: Viral conjunctivitis. In: Hyndiuk R,
Tabbara K, eds. Boston, MA: Little, Brown;
1996:453470.
101. Hyndiuk R, Glasser D: Herpes simplex
keratitis. In: Tabbara K, Hyndiuk R, eds.
102. Barnes S, Pavan-Langston D, Azar D:
Microbial keratitis, in Principles and
Practice of Infectious Diseases. In: Mandell
G, Bennett J, Dolin R, eds. Philadephia, PA:
Elsevier. 2005:13951406.
cornea and external eye. In: Albert D,
Jakobiec F, eds. Philadelphia, PA: WB
Saunders; 2000:846893.
104. Whitley RJ: Herpes simplex virus in
children. Curr Treat Options Neurol 2002;
4:231237.
105. Burnette T, deMiranda P: Purication and
characterization of an enzyme from rat live
that hydrolyzes 256487, the L-valyl ester
prodrug of acyclovir (Zovirax). Antiviral Res
1993; 20 (Suppl 1):115.
106. Weller S, Blum M, Smiley M: Phase I
pharmacokinetics of the acyclovir prodrug,
valacyclovir. Antivir Res 1993; 20(Suppl 1):
144.
107. Perry CM, Wagstaff AJ: Famciclovir. A
and therapeutic efcacy in herpesvirus
infections. Drugs 1995; 50:396415.
108. Labetoulle M, Kucera P, Ugolini G, et al:
Neuronal propagation of HSV1 from the
oral mucosa to the eye. Invest Ophthalmol
Vis Sci 2000; 41:26002606.
109. Claoue C, Hill T, Blyth W, et al: Clinical
ndings after zosteriform spread of herpes
simplex virus to the eye of the mouse.
Curr Eye Res 1987; 6:281.
110. Whitley RJ, Roizman B: Herpes simplex
virus infections. Lancet 2001;
357:15131518.
111. Pavan-Langston D: Major ocular viral
infections. In: Galasso G, Whitley R,
Merrigan T, eds. New York: Raven Press;
183, 1990.
112. Cockerham GC, Bijwaard K, Sheng ZM,

et al: Primary graft failure: a
clinicopathologic and molecular analysis.
Ophthalmology 2000; 107: 20832090;
113. Borderie VM, Meritet JF, Chaumeil C, et al:
Culture-proven herpetic keratitis after
penetrating keratoplasty in patients with no
previous history of herpes disease. Cornea
2004; 23:118124.
114. Cleator G, Klapper P, Dennett C, et al:
Corneal donor infection by herpes simplex
virus: herpes simplex virus DNA in donor
corneas. Cornea 1994; 13:294304.
115. Summers BC, Margolis TP, Leib DA:
Herpes simplex virus type 1 corneal
infection results in periocular disease by
zosteriform spread. J Virol 2001;
75:50695075.
116. Polcicova K, Biswas PS, Banerjee K, et al:
Herpes keratitis in the absence of
anterograde transport of virus from sensory
ganglia to the cornea. Proc Natl Acad Sci
USA 2005; 102:1146211467.
117. Davido DJ, Leib DA: Analysis of the basal
and inducible activities of the ICPO
promoter of herpes simplex virus type 1.
J Gen Virol 1998; 79(pt 9):20932098.
118. Naito J, Mott KR, Osorio N, et al: Herpes
simplex virus type 1 immediate-early
protein ICP0 diffuses out of infected rabbit
corneas. J Gen Virol 2005; 86:29792988.
119. Croen K, Ostrove J, Dragovic L, et al:
Latent Herpes simplex virus in human
trigeminal ganglia: detection of an
immediate early gene anti-sense
transcript by in situ hybridization. N Engl J
Med 1987; 317:1427.
120. Stevens J, Wagner E, Devi-Rao G, et al:
RNA complimentary to a herpes virus alpha
gene mRNA is prominent in latently infected
neurones. Science 1987; 235:1056.
121. Baringer J: The virology of Herpes simplex
virus infection in humans. Surv Ophthalmol
1976; 21:171.
122. Blyth W, Hill T, Field H, et al: Reactivation
of Herpes simplex virus infection by
ultraviolet light and possible involvement of
prostaglandins. J Gen Virol 1976; 33:547.
123. Klein R: Pathogenic mechanisms of
recurrent herpes simplex virus infections.
Arch Virol 1976; 51:1.
124. Klein R: The pathogenesis of acute, latent,
and recurrent herpes simplex infections.
Arch Virol 1982; 72:143168.
125. Kwon B, Gangarosa L, Burch K, et al:
Induction of ocular herpes simplex virus
shedding by iontophoresis of epinephrine
into rabbit cornea. Invest Ophthalmol 1981;
21:442.
126. Laibson P, Kilbrick S: Reactivation of
herpetic keratitis by epinephrine in rabbit.
127. Nesburn A, Dickenson M, Radnoti M, et al:
Experimental reactivation of ocular herpes
simplex in rabbits. Surv Ophthalmol 1976;
21:185.
128. Openshaw H, Puga A, Notkins A: Herpes
simplex virus infection in sensory ganglia:
immune control, latency, and reactivation.
Fed Proc 1979; 38:26602664.
129. Kaufman HE, Varnell ED, Toshida H, et al:
Effects of topical unoprostone and
latanoprost on acute and recurrent herpetic
keratitis in the rabbit. Am J Ophthalmol
2001; 131:643646.
130. Morales J, Shihab ZM, Brown SM, et al:
Herpes simplex virus dermatitis in patients
131.
132.
133.
134.
135.
136.
137.
138.
139.
140.
141.
142.
143.
144.
145.
146.
147.
148.
149.
using latanoprost. Am J Ophthalmol 2001;

132:114116.
Schumer RA, Camras CB, Mandahl AK:
Putative side effects of prostaglandin
analogs. Surv Ophthalmol 2002; 47(Suppl 1):
S219.
Bean G, Reardon G, Zimmerman TJ:
Association between ocular herpes simplex
virus and topical ocular hypotensive
therapy. J Glaucoma 2004; 13:361364.
Gordon YJ, Yates KA, Mah FS, et al:
The effects of Xalatan on the recovery of
ocular herpes simplex virus type 1 (HSV-1)
in the induced reactivation and spontaneous
shedding rabbit models. J Ocul Pharmacol
Ther 2003; 19:233245.
Asbell P: Valaciclovir for prevention of
recurrent herpes simplex virus eye disease
after Excimer laser photokeratectomy.
Trans Am Ophthalmologic Soc 2000;
98:285303.
Dhaliwal DK, Romanowski EG, Yates KA,
et al: Valacyclovir inhibition of recovery of
ocular herpes simplex virus type 1 after
experimental reactivation by laser in situ
2001; 27:12881293.
de Rojas Silva MV, Diez-Feijoo E, Javaloy
J, et al: Prophylactic perioperative antiviral
therapy for LASIK in patients with inactive
herpetic keratitis. J Refract Surg 2006;
22:404406.
Reed SY, Shin DH, Birt CM, et al: Herpes
simplex keratitis following argon laser
trabeculoplasty. Ophthalmic Surg 1994;
25:640.
Vrabec M, Durrie D, Chase D: Recurrence
of herpes simplex after excimer laser
keratectomy. Am J Ophthalmol 1992;
114:9697.
Deai T, Fukuda M, Tomoda Y, et al:
Excimer laser photokeratectomy
reactivates latent herpes simplex virus.
Egerer U-D, Stary A: Erosive-ulcrative
herpes simplex blepharitis. Arch
Jakobiec F, Srinivasan D, Gamboa E:
Recurrent herpetic angular blepharitis in an
adult. Am J Ophthalmol 1979; 88:744.
Foster WJ Jr, Kraus MD, Custer PL: Herpes
simplex virus dacryoadenitis in an
immunocompromised patient. Arch
Ophthalmol 2003; 121:911913.
Tsao CH, Chen CY, Yeh KW, et al: Monthly
recurrent herpes simplex virus blepharitis in
a boy for more than 10 years. Infection
2003; 31:257259.
Spruance S, Stewart J, Rowe N, et al:
Treatment of recurrent herpes simplex
labialis with oral acyclovir. J Infect Dis
1990; 161:185.
HEDS: Acyclovir for the prevention of
recurrent herpes simplex virus eye disease.
Herpetic Eye Disease Study Group. N Engl
J Med 1998; 339:300306.
McLean CJ, Rose GE: Postherpetic
lacrimal obstruction. Ophthalmology 2000;
107:496499.
Ameye C, Maudgal P, Missotten L: Primary
ocular HSV infections in adults. In: Maudgal
P, Missotten L, eds. The Netherlands,
Dordrecht: W Junk; 1985:133.
Brown DC, Nesburn AB, Nauheim JS, et al:
Recurrent herpes simplex conjunctivitis.
Rosenfeld SI, Alfonso EC, Gollamudi S:
Recurrent herpes simplex infection in a
150.
151.
152.
153.
154.
155.
156.
157.
158.
159.
160.
161.
162.
163.
164.
165.
166.
167.
168.
conjunctival flap. Am J Ophthalmol 1993;

116:242244.
Patterson A: Management of ocular herpes
simplex. Br J Ophthalmol 1967; 51:494495.
Pavan-Langston D (ed): Diagnosis and
infection. Boston, MA: Little, Brown;
1975:1935.
Liesegang T: Ocular herpes simplex
infection: pathogenesis and current
therapy. Mayo Clin Proc 1988; 63:1092.
Holland EJ, Schwartz GS: Classication of
herpes simplex virus keratitis. Cornea
1999; 18:144154.
Wilhelmus KR, Falcon MG, Jones BR:
Bilateral herpetic keratitis. Br J Ophthalmol
1981; 65:385387.
Souza PM, Holland EJ, Huang AJ: Bilateral
herpetic keratoconjunctivitis.
Ahonen R, Vannas A: Clinical comparison
between herpes simplex and herpes zoster
ocular infections. In: Maudgal P, Missotten
L, eds. The Netherlands, Dordrecht: Fr. W.
Junk; 1985:389.
Jones B, Falcon M, Williams H, et al:
Symposium on herpes simplex eye
disease: objectives in therapy. Trans
Kodama T, Hayasaka S, Setogawa T:
Immunofluorescent staining and corneal
sensitivity in patients suspected of having
herpes simplex keratitis. Am J Ophthalmol
1992; 113:187189.
Van Horn DL, Edelhauser HE, Schultz RO:
Experimental herpes simplex keratitis.
Early alterations of corneal epithelium
and stroma. Arch Ophthalmol 1970;
84:6775.
Baum J: Morphogenesis of the dendritic
gure in herpes simplex keratitis: a
negative study. Am J Ophthalmol 1970;
70:722.
Collum LM, McGettrick P, Akhtar J, et al:
Oral acyclovir (Zovirax) in herpes simplex
dendritic corneal ulceration. Br J
Ophthalmol 1986; 70:435438.
Hung S, Patterson A, Clark D, et al: Oral
acyclovir in management of dendritic
herpetic corneal ulceration. Br J
Schwab IR: Oral acyclovir in the
infections. Ophthalmology 1988;
95:423430.
Colin J, Robinet A, Malet F: Preventive
treatment of herpetic keratitis with
acyclovir tablets. J Fr Ophthalmol 1993;
16:69.
Goldberg L, Kaufman R, Kurtz T, et al:
Continuous ve-year treatment of patients
with frequently recurring genital herpes
simplex virus infection with acyclovir.
J Med Virol 1993; 1:4550.
Honda M, Okuda T, Hasegawa T, et al:
Effect of long-term, low-dose acyclovir
suppressive therapy on susceptibility to
acyclovir and frequency of acyclovir
resistance of herpes simplex virus type 2.
Antivir Chem Chemother 2001;
12:233239.
Simon AL, Pavan-Langston D: Long-term
oral acyclovir therapy. Effect on recurrent
infectious herpes simplex keratitis in
patients with and without grafts.
Ophthalmology 1996; 103:13991405.
HEDS: Oral acyclovir for herpes simplex
virus eye disease: effect on prevention of
CHAPTER 49
693
169.
170.
171.
172.
173.
174.
175.
176.
177.
178.
179.
180.
181.
182.
SECTION 6
183.
184.
185.
186.
187.
188.
189.
694
190.
epithelial keratitis and stromal keratitis.

Herpetic Eye Disease Study Group. Arch
Ophthalmol 2000; 118:10301036.
Margolis TP, Ostler HB: Treatment of ocular
disease in eczema herpeticum. Am J
Ophthalmol 1990; 110:274279.
Wilhelmus KR: The treatment of herpes
simplex virus epithelial keratitis. Trans Am
Iglesia DD, Stepp MA: Disruption of the
basement membrrane after corneal
debridement. Invest Ophthalmol Vis Sci.
2000; 41:10451053.
Labetoulle M: The latest in herpes simplex
keratitis therapy. J Fr Ophtalmol 2004;
27:547557.
Goins K: New insights into the diagnosis
and treatment of neurotrophic keratopathy.
Ocular Surface 2005; 3:96110.
Muller LJ, Vrensen GF, Pels L, et al:
Architecture of human corneal nerves.
Guthoff RF, Wienss H, Hahnel C, et al:
Epithelial innervation of human cornea: a
three-dimensional study using confocal
laser scanning fluorescence microscopy.
Cornea 2005; 24:608613.
Muller LJ, Marfurt CF, Kruse F, et al:
Corneal nerves: structure, contents and
Wilson S, Netto M, Ambrosio R Jr.
Cornealk cells: chatty in development,
homeostatis, wound healing, and disease.
Am J Ophthalmol 2003; 136:530536.
Netto MV, Mohan RR, Ambrosio R Jr, et al:
Wound healing in the cornea: a review of
refractive surgery complications and new
prospects for therapy. Cornea 2005;
24:509522.
Lindsay RM, Harmar AJ: Nerve growth
factor regulates expression of neuropeptide
genes in adult sensory neurons. Nature
1989; 337:362364.
Suuronen EJ, Nakamura M, Watsky MA,
et al: Innervated human corneal equivalents
as in vitro models for nerve-target cell
interactions. FASEB J 2004; 18:170172.
Wilson SE, Liu JJ, Mohan RR: Stromalepithelial interactions in the cornea. Prog
Retin Eye Res 1999; 18:293309.
Beuerman RW, Schimmelpfennig B:
Sensory denervation of the rabbit cornea
affects epithelial properties. Exp Neurol
1980; 69:196201.
Lim CH: Innervation of the cornea of
monkeys and the effects of denervation.
Br J Physiol Opt 1976; 31:3842.
Rosenberg ME, Tervo TM, Muller LJ, et al:
In vivo confocal microscopy after herpes
Gilbard JP, Rossi SR: Tear lm and ocular
surface changes in a rabbit model of
neurotrophic keratitis. Ophthalmology
1990; 97:308312.
Heigle TJ, Pflugfelder SC: Aqueous tear
production in patients with neurotrophic
Zhang M, Chen J, Luo L, et al: Altered
corneal nerves in aqueous tear deciency
viewed by in vivo confocal microscopy.
Cornea 2005; 24:818824.
Cavanagh C, Colley A, Pihlaja D: Persistent
corneal epithelial defects. Int Ophthalmol
Clin 1979; 19:197.
Eakins KE, Whitelocke RA, Bennett A, et al:
Prostaglandin-like activity in ocular
inflammation. Br Med J 1972; 3:452453.
Khodadoust AA, Silverstein AM, Kenyon
191.
192.
193.
194.
195.
196.
197.
198.
199.
200.
201.
202.
203.
204.
205.
206.
207.
208.
209.
210.
DR, et al: Adhesion of regenerating corneal

epithelium. The role of basement membrane.
Cavanagh H: Herpetic ocular disease:
therapy of persistent epithelial defects. Int
Kaufman H: Epithelial erosion syndrome:
metaherpetic keratitis. Am J Ophthalmol
1964; 57:983.
VanBijsterveld O, deKoning E: Trophic
herpetic corneal ulcers. In: Blodi F, ed.
New York: Churchill Livingstone; 1984:29.
Cavanagh HD, Colley A, Pihlaja DJ:
Persistent corneal epithelial defects. Int
Cavanagh HD, Colley AM: The molecular
basis of neurotrophic keratitis. Acta
Fukuda M, Deai T, Hibino T, et al:
Quantitative analysis of herpes simplex
virus genome in tears from patients with
herpetic keratitis. Cornea 2003; 22:S55S60.
Kaufman H, Uotila M, Gasset A, et al: The
medical uses of soft contact lenses. Trans
Am Acad Ophthalmol Otolaryngol 1971;
75:361.
Karlgard CC, Jones LW, Moresoli C: Survey
of bandage lens use in North America,
OctoberDecember 2002. Eye Contact
Lens 2004; 30:2530.
Tsia I, Till G, Marak G Jr. Effects of
mydriatic agents on neutrophil migration.
Ophthalmol Res 1988.
Beutner KR: Valacyclovir: a review of its
antiviral activity, pharmacokinetic
properties, and clinical efcacy. Antiviral
Res 1995; 28:281290.
Beutner KR, Friedman DJ, Forszpaniak C,
et al: Valaciclovir compared with acyclovir
for improved therapy for herpes zoster in
immunocompetent adults. Antimicrob
Mindel A: Long-term clinical and
psychological managaement of genital
herpes. J Med Virol 1993; 1:3944.
Strauss S, Croen K, Sawyer M, et al:
Acyclovir suppression of frequently
recurring genital herpes: efcacy and
diminishing need during successive years
of treatment. JAMA 1988; 260:22272230.
Ralph RA: Tetracyclines and the treatment
of corneal stromal ulceration: a review.
Cornea 2000; 19:274277.
Cosar CB, Cohen EJ, Rapuano CJ, et al:
Tarsorrhaphy: clinical experience from
a cornea practice. Cornea 2001;
20:787791.
Vote BJ, Elder MJ: Cyanoacrylate glue for
corneal perforations: a description of a
surgical technique and a review of the
literature. Clin Experiment Ophthalmol
2000; 28:437442.
Pavan-Langston D: Herpes keratitis
Clinical features, diagnosis, treament and
prophylaxis. In: Swartz M, Remington J,
eds. Boston, MA: Blackwell Science
Publishers; 2000:298324.
Refojo M, Dohlman C, Koliopoulos J:
Adhesives in ophthalmology: a review.
Berman MB, Barber JC, Talamo RC, et al:
Corneal ulceration and the serum
antiproteases. I. Alpha 1-antitrypsin. Invest
Ophthalmol 1973; 12:759770.
Lambiase A, Manni L, Rama P, et al:
Clinical application of nerve growth factor
on human corneal ulcer. Arch Ital Biol 2003;
141:141148.
211. Lambiase A, Rama P, Bonini S, et al:

Topical treatment with nerve growth factor
for corneal neurotrophic ulcers. N Engl J
Med 1998; 338:11741180.
212. Tsubota K: Ocular surface management in
corneal transplantation, a review. Jpn J
Ophthalmol 1999; 43:502508.
213. Tsubota K, Goto E, Shimmura S, et al:
Treatment of persistent corneal epithelial
defect by autologous serum application.
Ophthalmology 1999; 106:19841989.
214. Dekaris I, Gabric N, Mravicic I, et al:
Multilayer vs. monolayer amniotic
membrane transplantation for deep corneal
ulcer treatment. Coll Antropol 2001;
25(Suppl):2328.
215. Dogru M, Tsubota K: Current concepts in
ocular surface reconstruction. Semin
216. Kruse FE, Rohrschneider K, Volcker HE:
Techniques for reconstruction of the
corneal surface by transplantation of
preserved human amniotic membrane.
Multilayer amniotic membrane
transplantation for reconstruction of deep
corneal ulcers. Ophthalmology 1999;
106:15041510; discussion 1511.
218. Kenyon KR: Amniotic membrane: mothers
own remedy for ocular surface disease.
Cornea 2005; 24:639642.
219. Solomon A, Meller D, Prabhasawat P, et al:
Amniotic membrane grafts for nontraumatic
corneal perforations, descemetoceles, and
deep ulcers. Ophthalmology 2002;
109:694703.
220. Hick S, Demers PE, Brunette I, et al:
Amniotic membrane transplantation and
brin glue in the management of corneal
ulcers and perforations: a review of 33
cases. Cornea 2005; 24:369377.
221. Alino AM, Perry HD, Kanellopoulos AJ,
et al: Conjunctival flaps. Ophthalmology
1998; 105:11201123.
222. Holbach LM, Asano N, Naumann GO:
Infection of the corneal endothelium in
herpes simplex keratitis. Am J Ophthalmol
1998; 126:592594.
223. Kaufman H, Rayeld M: Viral conjunctivitis
and keratitis. In: Kaufman H, Mcdonald M,
Barron B, et al, eds. New York: Churchill
224. Pavan-Langston D: Herpes simplex of the
ocular anterior segment. Curr Clin Top
Infect Dis 2000; 20:298324.
225. Pepose J: Herpes simplex keratitis: role of
viral infection versus immune response.
Survey Ophthalmol 1991; 35:345352.
226. Tabbara KF: Treatment of herpetic keratitis.
227. Abghari S, Stulting R: Recovery of Herpes
simplex virus from ocular tissues of latently
infected mice. Invest Ophthalmol Vis Sci
1988; 29:239.
228. Brik D, Dunkel E, Pavan-Langston D:
Herpetic keratitis: persistence of viral
particles despite topical and systemic
antiviral therapy. Report of two cases and
review of the literature. Arch Ophthalmol
1993; 111:522527.
229. Dawson C, Togni B, Moore T Jr: Structural
changes in chronic herpetic keratitis:
studied by light and electron microscopy.
230. Rong BL, Dunkel EC, Pavan-Langston D:
Molecular analysis of acyclovir-induced
suppression of HSV-1 replication in rabbit
231.
232.
233.
234.
235.
236.
237.
238.
239.
240.
241.
242.
243.
244.
245.
246.
247.
corneal cells. Invest Ophthalmol Vis Sci

1988; 29:928932.
Tullo A, Easty D, Shimeld C, et al: Isolation
of Herpes simplex virus from corneal discs
of patients with chronic stromal keratitis.
Trans Ophthalmol Soc UK 1985; 104:159.
Rong BL, Pavan-Langston D, Weng QP,
thymidine kinase and latency-associated
transcript gene sequences in human
herpetic corneas by polymerase chain
reaction amplication. Invest Ophthalmol
Vis Sci 1991; 32:18081815.
Brandt CR: The role of viral and host genes
in corneal infection with herpes simplex
virus type 1. Exp Eye Res 2005;
80:607621.
Kaufman H, Centifanto-Fitzgerald Y, Varnell
E: Herpes simplex keratitis. Ophthalmology
1983; 90:700.
Spear P: Glycoproteins specied by
Herpes simplex viruses. In: Roizman B, ed.
New York: Plenum Press; 1985:315.
Kaufman H, Varnell E, Centifanto Y, et al:
Effect of the herpes simplex virus genome
on the response of infection to
corticosteroids. Am J Ophthalmol 1985;
100:114.
Keadle T, Laycock K, Morris J, et al:
Therapeutic vaccination with vhs(-) herpes
simplex virus reduces the severity of
recurrent herpetic stromal keratitis in mice.
J Gen Virol 2002; 83(Pt 10):23612365.
Keadle TL, Usui N, Laycock KA, et al: IL-1
and TNF-alpha are important factors in the
pathogenesis of murine recurrent herpetic
stromal keratitis. Invest Ophthalmol Vis Sci
2000; 41:96102.
Tumpey TM, Cheng H, Yan XT, et al:
Chemokine synthesis in the HSV-1-infected
cornea and its suppression by interleukin10. J Leukoc Biol 1998; 63:486492.
Yan XT, Tumpey TM, Kunkel SL, et al: Role
of MIP-2 in neutrophil migration and tissue
injury in the herpes simplex virus-1-infected
39:18541862.
Sagoo P, Chan G, Larkin DF: Inflammatory
cytokines induce apoptosis of corneal
endothelium through nitric oxide. Invest
OBrien WJ, Guy J, Taylor JL: Pathogenesis
of corneal oedema associated with herpetic
eye disease. Br J Ophthalmol 1990;
74:723730.
Vollstedt S, Arnold S, Schwerdel C, et al:
Interplay between alpha/beta and gamma
interferons with B, T, and natural killer cells
in the defense against herpes simplex virus
type 1. J Virol 2004; 78:38463850.
Streilein JW, Dana MR, Ksander BR:
Immunity causing blindness: ve different
paths to herpes stromal keratitis. Immunol
Today 1997; 18:443449.
Tumpey TM, Fenton R, Molesworth-Kenyon
S, et al: Role for macrophage inflammatory
protein 2 (MIP-2), MIP-1alpha, and
interleukin-1alpha in the delayed-type
hypersensitivity response to viral antigen.
J Virol 2002; 76:80508057.
Ellison AR, Yang L, Cevallos AV, et al:
Analysis of the herpes simplex virus type 1
UL6 gene in patients with stromal keratitis.
Virology 2003; 310:2428.
Verjans GM, Remeijer L, Mooy CM, et al:
Herpes simplex virus-specic T cells
inltrate the cornea of patients with
herpetic stromal keratitis: no evidence for
248.
249.
250.
251.
252.
253.
254.
255.
256.
257.
258.
259.
260.
261.
262.
263.
264.
265.
autoreactive T cells. Invest Ophthalmol Vis

Sci 2000; 41:26072612.
Biswas PS, Rouse BT: Early events in HSV
keratitissetting the stage for a blinding
disease. Microbes Infect 2005; 7:799810.
Banerjee K, Biswas PS, Kumaraguru U,
et al: Protective and pathological roles of
virus-specic and bystander CD8+ T cells
in herpetic stromal keratitis. J Immunol
2004; 173:75757583.
Osorio Y, Cai S, Hofman FM, et al:
Involvement of CD8+ T-cells in
exacerbation of corneal scarring in mice.
Curr Eye Res 2004; 29:145151.
Banerjee K, Biswas PS, Rouse BT:
Elucidating the protective and pathologic
T cell species in the virus-induced corneal
immunoinflammatory condition herpetic
stromal keratitis. J Leukoc Biol 2005;
77:2432.
Meyers-Elliott RH, Chitjian PA: Induction of
cell-mediated immunity in herpes simplex
virus keratitis. Kinetics of lymphocyte
transformation and the effect of antiviral
antibody. Invest Ophthalmol Vis Sci 1980;
19:920929.
Meyers-Elliott RH, Pettit TH, Maxwell WA:
Viral antigens in the immune ring of herpes
simplex stromal keratitis. Arch Ophthalmol
1980; 98:897904.
Schwartz GS, Harrison AR, Holland EJ:
Etiology of immune stromal (interstitial)
Holbach LM, Font RL, Naumann GO:
Herpes simplex stromal and endothelial
keratitis. Granulomatous cell reactions at
the level of Descemets membrane, the
stroma, and Bowmans layer.
Meyers RL, Pettit TH: The pathogenesis of
corneal inflammation due to herpes
simplex virus. I. Corneal hypersensitivity in
the rabbit. J Immunol 1973; 111:10311042.
Baltatzis S, Romero-Rangel T, Foster CS:
Sectorial keratitis and uveitis: differential
diagnosis. Graefes Arch Clin Exp
Wilhelmus KR, Sugar J, Hyndiuk RA, et al:
Corneal thickness changes during herpes
simplex virus disciform keratitis. Cornea
2004; 23:154157.
Robin JB, Steigner JB, Kaufman HE:
Progressive herpetic corneal endotheliitis.
Am J Ophthalmol 1985; 100:336337.
Sundmacher R: A clinico-virologic
classication of herpetic anterior segment
disease with special reference to intraocular
herpes. In: Sundmacher R, Munich JF, eds.
Bergmann Verlag; 1981:203.
Sundmacher R, Neumann-Haefelin D: Herpes
simplex virus isolations from the aqueous
humor of patients suffering from focal iritis,
endotheliitis, and prolonged disciform
keratitis with glaucoma (authors transl). Klin
Sutcliffe E, Baum J: Acute idiopathic corneal
endotheliitis. Ophthalmology 1984; 91:1161.
Ohashi Y, Yamamoto S, Nishida K, et al:
Demonstration of herpes simplex virus
DNA in idiopathic corneal endotheliopathy.
Am J Ophthalmol 1991; 112: 419423.
Amano S, Oshika T, Kaji Y, et al: Herpes
simplex virus in the trabeculum of an eye
with corneal endotheliitis. Am J Ophthalmol
1999; 127:721722.
Olsen TW, Hardten DR, Meiusi RS, et al:
Linear endotheliitis. Am J Ophthalmol 1994;
117:468474.
266. Cheng CK, Chang SW, Hu FR: Acyclovir

treatment for linear endotheliitis on grafted
corneas. Cornea 1995; 14:311315.
267. Easty D, Shimeld C, Claoue C, et al:
Herpes simplex virus isolation in chronic
stromal keratitis: human and laboratory
studies. Curr Eye Res 1987; 6:69.
268. Hogan M: Corneal transplantation in the
treatment of herpetic disease of the
cornea. Am J Ophthalmol 1957;
43:147.
269. Pavan-Langston D, Nesburn AB: The
chronology of primary herpes simplex
infection of the eye and adnexal glands.
270. Collin H, Abelson M: Herpes simplex virus
in human cornea, retrocorneal brous
membrane and vitreous. Arch Ophthalmol
1976; 94:1726.
271. Gupta K, Hoepner J, Streeten B:
Pseudomelanoma of the iris in herpes
simplex keratoiritis. Ophthalmol 1986;
93:1524.
272. HEDS: A controlled trial of oral acyclovir
for iridocyclitis caused by herpes simplex
virus. The Herpetic Eye Disease Study
114:10651072.
273. Pavan-Langston D, Brockhurst RJ: Herpes
simplex panuveitis. A clinical report. Arch
Ophthalmol 1969; 81:783787.
274. Gaynor BD, Margolis TP, Cunningham ET
Jr. Advances in diagnosis and management
of herpetic uveitis. Int Ophthalmol Clin
2000; 40:85109.
275. Bielory L: Allergic and immunologic
disorders of the eye. Part II: ocular allergy.
276. Demir E, Tanac R, Can D, et al: Is there an
increase in the prevalence of allergic
diseases among schoolchildren from the
Aegean region of Turkey? Allergy Asthma
Proc 2005; 26:410414.
277. Taieb A: Atopic dermatitis: denition,
epidemiology, natural history, severity and
scores. Ann Dermatol Venereol 2005;
132(Spec No 1):1S351S43.
278. Yan DC, Ou LS, Tsai TL, et al: Prevalence
and severity of symptoms of asthma,
rhinitis, and eczema in 13- to 14-year-old
children in Taipei, Taiwan. Ann Allergy
Asthma Immunol 2005; 95:579585.
279. Global. HIV infections doubled since 1995.
AIDS Policy Law 2005; 21:1,4.
280. Del Rio C, Curran J: Epidemiology and
prevention of acquired immunodeciency
syndrome and human immunodeciency
virus infection. In: Mandell G, Bennett J,
Dolin R, eds. Philadelphia, PA: Elsevier;
2005:14771506.
281. Merson M, Piot P: Global perspectives on
human immunodeciency virus infection
and acquired immunodeciency syndrome.
In: Mandell G, Bennett J, Dolin R, eds.
Philadelphia, PA: Elsevier; 2005:14651477.
282. Fleming PL, Byers R, Sweeney P, et al:
HIV prevalence in the United States. In:
Fleming PL, Byers R, Sweeney P, et al, eds.
9th Conference on Retroviruses Seattle,
WA, February 2428, 2002.
283. Kaplan MH, Sadick N, McNutt NS, et al:
Dermatologic ndings and manifestations
of acquired immunodeciency syndrome
(AIDS). J Am Acad Dermatol 1987;
16:485506.
284. Leyden J, Baker B: Localized herpes
simplex infections in atopic dermatitis. Arch
Dermatol 1979; 115:311314.
CHAPTER 49
695
SECTION 6
696
285. Ryan-Graham MA, Durand M, PavanLangston D: AIDS and the anterior segment.
286. Vestey J, Howie S, Norval M, et al: Immune
responses to herpes simplex virus in
patients with facial herpes simplex and
those with excema herpeticum. Br J
Dermatol 1988; 118:775779.
287. Garrity J, Liesegang T: Ocular complications
of atopic dermatitis. Can J Ophthalmol
1984; 19:21.
288. Binder P: Review of treatment of ocular
Herpes simplex infections in the neonate
and immunocompromised host. Cornea
1985; 3:178.
289. Mines J, Kaplan H: Acquired immune
deciency syndrome (AIDS): the disease
and its ocular manifestations. Int Ophthal
Clin 1986; 26:7385.
290. Ryan M, Pavan-Langston D, Durand M:
AIDS and the anterior segment. In: Cl Io,
ed. Philadelphia, PA: Lippencott-Raven;
1997:241264.
291. Cunningham ET Jr, Margolis TP: Ocular
manifestations of HIV infection. N Engl J
Med 339:236244.
292. Young T, Robin J, Holland G, et al: Herpes
simplex keratitis in patients with acquired
immune deciency syndrome. Ophthalmol
1989; 96:14761481.
293. Rosenwasser G, Greene W: Simultaneous
herpes simplex types 1 and 2 keratitis in
acquired immune deciency syndrome. Am
J Ophthalmol 1992; 113:102103.
294. Hodge W, Margolis T: Herpes simplex virus
keratitis among patients who are positive or
negative for human immunodeciency
virus. Ophthalmology 1997; 104:120124.
295. Hodge W, Seiff S, Margolis T: Ocular
opportunistic infection incidences among
patients who are HIV positive compared to
patients who are HIV negative.
296. Ng WT, Versace P: Ocular association of
HIV infection in the era of highly active
antiretroviral therapy and the global
perspective. Clin Experiment Ophthalmol
2005; 33:317329.
297. Balfour H: Resistance of herpes simplex to
acyclovir. Ann Int Med 1983; 98:404.
298. Chilukuri S, Rosen T: Management of
acyclovir-resistant herpes simplex virus.
Dermatol Clin 2003; 21:311320.
299. Ramos-Gomez F, Greenspan D, Oleske J:
Pediatric HIV: orofacial manifestations.
In: Oleske J, Denny T, eds. New York:
New Jersey Medical School; 1994:15.
300. Minkovitz J, Pepose J: Topical interferon
alpha-2A treatment of herpes simplex
keratitis resistant to multiple antiviral
medications in an immunosuppressed
patient. Cornea 1995; 14:326330.
301. Corey L, Holmes K: Therapy for human
immunodeciency virus infection what
have we learned? New Engl J Med 1996;
335:11421144.
302. Hammer S, Katzenstein D, Hughes M,
et al: A trial comparing nucleoside
monotherapy with combination therapy
in HIV-infected adults with CD4 cell
counts from 200 to 500 per cubic
millimeter. New Engl J Med 1996;
335:10811090.
303. Nimura M, Nishikawa T: Treatment of
excema herpeticum with oral acyclovir. Am
J Med 1988; 85:4954.
304. Colin J, Baikoff G, Sourdille P: Lamellar
graft in the treatment of herpes simplex
305.
306.
307.
308.
309.
310.
311.
312.
313.
314.
315.
316.
317.
318.
319.
320.
321.
322.
323.
324.
325.
ocular (authors transl). J Fr Ophtalmol

1978; 1:501505.
Ludwig K, Bechmann M, Welge-Lussen U,
et al: New automated microkeratome for
trepanation of lamellar keratoplasty.
Weiss N, Jones B: Problems of corneal
grafting in herpetic keratitis. Ciba Found
Symp 1973; 15:220.
Gundersen T: Conjunctival flaps in
treatment of corneal disease. Arch
Thoft RA: Conjunctival transplantation.
Rosenfeld S, Alfonso E, Gollamudi S:
116:242243.
Branco BC, Gaudio PA, Margolis TP:
Epidemiology and molecular analysis of
herpes simplex keratitis requiring primary
penetrating keratoplasty. Br J Ophthalmol
2004; 88:12851288.
Al-Yousuf N, Mavrikakis I, Mavrikakis E,
et al: Penetrating keratoplasty: indications
over a 10 year period. Br J Ophthalmol
2004; 88:9981001.
Epstein R, Seedor J, Dreizen N, et al:
Penetrating keratoplasty for herpes simplex
keratitis and keratoconus: allograft rejection
and survival. Ophthalmology 1987;
94:935944.
Rapuano C, Cohen E, Brady S, et al:
Indications for and outcomes of repeat
penetrating keratoplasty. Am J Ophthalmol
1990; 109:689695.
Mathers WD, Jester JV, Lemp MA: Return
of human corneal sensitivity after
penetrating keratoplasty. Arch Ophthalmol
1988; 106:210211.
Langston RH, Pavan-Langston D:
Penetrating keratoplasty for herpetic
keratitis: decision-making and management.
Boisjoly H, Roy R, Dube I, et al: HLA-A, B
and DR matching in corneal transplantation.
Tuppin P, Poinard C, Loty B, et al: Risk
factors for corneal regraft in patients on the
French waiting list. Cornea 2004;
23:704711.
Cohen EJ, Laibson PR, Arentsen JJ:
Corneal transplantation for herpes simplex
95:645650.
Fine M, Cignetti F: Penetrating keratoplasty
in herpes simplex keratitis. Arch
Ophthalmol 1977; 95: 1755.
Cobo LM, Coster DJ, Rice NS, et al:
Prognosis and management of corneal
transplantation for herpetic keratitis. Arch
Ophthalmol 1980; 98:17551759.
Ficker L, Kirkness C, Rice N, et al:
Changing management and improved
prognosis for corneal grafting and herpes
simplex keratitis. Ophthalmology 1989;
96:1587.
Foster C, Dunkin J: Penetrating
keratoplasty for herpes simplex keratitis.
Polack F, Kaufman H: Penetrating
keratoplasty in herpetic keratitis. Am J
Ophthalmol 1972; 73:908913.
Rice N, Jones B: Problems of corneal
grafting in herpetic keratitis. In: Jones B,
ed. Amsterdam: Elsevier; 1973:221.
Barney N, Foster C: A prospective
randomized trial of oral acyclovir following
326.
327.
328.
329.
330.
331.
332.
333.
334.
335.
336.
337.
338.
339.
340.
341.
342.
343.
penetrating keratoplasty for herpes. Cornea

1994; 13:232236.
Koay PY, Lee WH, Figueiredo FC: Opinions
on risk factors and management of corneal
graft rejection in the United Kingdom.
Cornea 2005; 24:292296.
Croen K, Ostrove J, Dragovic L, et al:
Patterns of gene expression and sites of
latency in human ganglia are different for
varicella-zoster and herpes simplex viruses.
85:97739777.
Mannis M, Plotnik R, Schwab I, et al:
Herpes simplex dendritic keratitis after
111:480484.
Liekfeld A, Jaeckel C, Pleyer U, et al:
Analysis of the aqueous humor in
keratoplasty patients with keratitis. Initial
results. Ophthalmologe (German) 2001;
98:456459.
van Gelderen B, Van der Lelij A, Treffers W,
type 1, 2 and varicella zoster virus DNA in
recipient corneal buttons. Br J Ophthalmol
2000; 84:12381243.
De Kesel R, Koppen C, Leven M, et al:
Primary graft failure caused by herpes
simplex virus type 1. Cornea 2001;
20:187190.
Rezende RA, Uchoa UB, Raber IM, et al:
New onset of herpes simplex virus
epithelial keratitis after penetrating
137:415419.
Robert PY, Adenis JP, Denis F, et al:
Transmission of viruses through corneal
transplantation. Clin Lab 2005; 51:419423.
Robert PY, Adenis JP, Pleyer U: [How safe
is corneal transplantation?]. Klin Monatsbl
Augenheilkd 2005; 222:870873.
Zheng X: Reactivation and donor-host
transmission of herpes simplex virus after
corneal transplantation. Cornea 2002;
21(7 Suppl):S90S93.
Zheng X, Loutsch J, Shimomura Y, et al:
Reactivation of herpes virus after lamellar
keratoplasty. Jpn J Ophthalmol 1999;
43:257261.
Zheng X, Marquart M, Loustch J, et al: HSV1 migration in latently infected and naive
rabbits after penetrating keratoplasty. Invest
Hicks CR, Crawford GJ, Tan DT, et al:
Outcomes of implantation of an articial
cornea, AlphaCor: effects of prior ocular
herpes simplex infection. Cornea 2002;
21:685690.
Kahn B, Pavan-Langston D: Clinical
manifestations and treatment modalities in
Herpes simplex virus of the ocular anterior
segment. In: Gragoudas E, Vattas D, Hatten
M, eds. Philadelphia, PA: Lippencott,
Williams & Wilkins; 2004; 44:103134.
Ma JJ, Graney JM, Dohlman CH: Repeat
penetrating keratoplasty versus the Boston
keratoprosthesis in graft failure. Int
Kahn B, Aqualvella J, Harissi-Dagher M, et
al: The Boston keratoprosthesis in herpetic
keratitis. Arch Ophth 2007; 125:745749.
Gershon AA: The current status of live
attenuated varicella vaccine. Arch Virol
Suppl. 2001; 16.
Vaccinia (smallpox) vaccine:
recommendations of the Advisory
Committee on Immunization Practices.
CDC MMWR, 50: No. RR-2001; 10:122.
344. Fleisher G, Henry W, McSorley H, et al: Life

threatening complications of varicella. Am J
Dis Child 1981; 135:896.
345. Gershon A, Forghani B: Varicella-zoster
virus. In: Lennette E, Lennette D, Lennette
E, eds. Washington, DC: Am Public Health
Association; 1995:601613.
346. Gershon AA: Varicella vaccine: rare serious
problemsbut the benets still outweigh
the risks. J Infect Dis 2003; 188:945947.
347. Preblud S, DAngela I: Age specic risks of
varicella complication. Pediatrics 1981;
68:14.
348. Preblud S, DAngelo I: Chicken pox in the
United States, 19721977. J Infect Dis
1979; 140:257263.
349. Vazquez M, LaRussa PS, Gershon AA,
et al: Effectiveness over time of varicella
vaccine. JAMA 2004; 291:851855.
350. Cherry J: Chicken pox transmission. JAMA
1983; 250:2060.
351. Wilson FI: Varicella and Herpes zoster
ophthalmicus. In: Tabbara K, Hyndiuk R, eds.
352. Edwards T: Ophthalmic complications of
varicella. J Pediatr Ophthalmol 1965; 2:37.
353. Lambert SR, Taylor D, Kriss A, et al: Ocular
manifestations of the congenital varicella
syndrome. Arch Ophthalmol 1989; 107:52.
354. Gershon A: Varicella in mother and infant.
In: Krugman S, Gershon A, eds. New York:
Alan R. Liss Inc; 1975.
355. Oxman MN, Levin MJ, Johnson GR, et al:
A vaccine to prevent herpes zoster and
postherpetic neuralgia in older adults.
N Engl J Med 2005; 352:22712284.
356. Gershon A, Steinberg S: Antibody
responses to varicella-zoster virus and the
role of antibody in host defense. Am J Med
Sci 1981; 282:1217.
357. Pavan-Langston D, Dunkel E: Varicellazoster of the anterior segment. In: Pepose J,
Holland G, Wilhelmus K, eds. Philadelphia,
PA: Mosby Inc; 1996:946985.
358. Nesburn A, Borit A, Pentelei-Molnar J,
et al: Varicella dendritic keratitis. Invest
359. Pavan-Langston D, McCulley JP: Herpes
zoster dendritic keratitis. Arch Ophthalmol
1973; 89:2529.
360. Uchida Y, Kaneko M, Hayashi K: Varicella
dendritic keratitis. Am J Ophthalmol 1980;
89:259.
361. Magone MT, Nasser RE, Cevallos AV, et al:
Chronic recurrent varicella-zoster virus
keratitis conrmed by polymerase chain
reaction testing. Am J Ophthalmol 2005;
139:11351136.
362. deFreitas D, Sato E, Kelly L, et al: Delayed
onset varicella keratitis. Cornea 1992;
11:471474.
363. Power WJ, Hogan RN, Hu S, et al: Primary
varicella-zoster keratitis: diagnosis by
polymerase chain reaction. Am J
Ophthalmol 1997; 123:252254.
364. Robb R: Cataracts acquired following
varicella infection. Arch Ophthalmol 1972;
87:352.
365. Strachman J: Uveitis associated with
chicken pox. J Pediatr 1955; 46:327.
366. Taylor D, Ffytche T: Optic disc pigmentation
associated with a eld defect following
chicken pox. J Pediatr Ophthalmol 1976;
13:80.
367. Roelandt V, Fayol L, Hugonenq C, et al:
Retrobulbar optic nevritis and chicken pox:
a case report in a child. Arch Pediatr 2005;
12:278280.
368. CDC, C, Prevention. fDCa: Licensure of

varicella virus vaccine, live. MMWR 1995;
44:264.
369. Gershon A: Varicella-zoster virus: prospects
for control. Adv Pediatr Infect Dis 1995;
10:93124.
370. Gershon AA: Live-attenuated varicella
vaccine. Infect Dis Clin North Am 2001;
15:6581, viii.
371. Asano Y, Suga S, Yoshikawa T, et al:
Experience and reason: twenty-year
follow-up of protective immunity of the Oka
strain live varicella vaccine. Pediatrics
1994; 94:524526.
372. Kuter B, Weibel R, Guess H, et al:
Oka/Merck varicella vaccine in healthy
children: nal report of a 2-year efcacy
study and 7-year follow-up studies.
Vaccine 1991; 9:643647.
373. Galil K, Lee B, T, S, et al: Outbreak of varicella
at a day-care center despite vaccination.
N Engl J Med 2002; 347:19091915.
374. Arvin AM, Pollard RB, Rasmussen LE, et al:
Cellular and humoral immunity in the
pathogenesis of recurrent herpes viral
infections in patients with lymphoma. J Clin
Invest 1980; 65:869878.
375. Trannoy E, Berger R, Hollander G, et al:
Vaccination of immunocompetent elderly
subjects with a live attenuated Oka strain
of varicellazoster virus: a randomized,
controlled, dose-response trial. Vaccine
2000;; 18:17001706.
376. Brunell P, Taylor-Wiedeman J, Geiser C,
et al: Risk of herpes zoster in children with
leukemia: varicella vaccine compared with
history of chickenpox. Pediatrics 1986;
77:5356.
377. Hardy I, Gershon A, Sterinberg S, et al: The
incidence of zoster after immunization with
live attenuated varicella vaccine: a study in
children with leukemia. N Engl J Med 1991;
325:15451550.
378. White C: Letter to the Editor. Pediatrics
1992; 89:354.
379. Matsubara K, Nigami H, Harigaya H, et al:
Herpes zoster in a normal child after
varicella vaccination. Acta Paediatrica
Japonica 1995; 37:648650.
380. Naseri A, Margolis TP: Varicella zoster virus
immune recovery stromal keratitis in a
patient with AIDS. Br J Ophthalmol 2001;
85:13901391.
381. Gnann JW Jr, Whitley RJ: Clinical practice.
Herpes zoster. N Engl J Med 2002;
347:340346.
382. Schmader KE: Epidemiology and impact on
quality of life of postherpetic neuralgia and
painful diabetic neuropathy. Clin J Pain
2002; 18:350354.
383. Mahalingam R, Cabirac G, Wellish M, et al:
In-vitro synthesis of functional varicella
zoster and herpes simplex viral thymidine
kinase. Virus Genes 1990; 4:105120.
384. Mahalingam R, Wellish M, Wolf W, et al:
Latent varicella-zoster viral DNA in human
trigeminal and thoracic ganglia. N Engl J
Med 1990; 323:627631.
385. Cobo M, Foulks G, Liesegang T, et al:
Observations on the natural history of
1987; 6:195199.
386. Donahue J, Choo P, Manson J, et al: The
incidence of herpes zoster. Arch Int Med
1995; 155:16051609.
387. Harding S, Lipton J, Wells J: Natural history
of herpes zoster ophthalmicus: predictors
of postherpetic neuralgia and ocular
388.
389.
390.
391.
392.
393.
394.
395.
396.
397.
398.
399.
400.
401.
402.
403.
404.
405.
406.
407.
408.
involvement. Br J Ophthalmol 1987;

71:353358.
Hope-Simpson R: The nature of herpes
zoster: a long term study and a new
hypothesis. Proc Soc Med 1965; 58:9.
US Department of Health and Human
Services, N: External ocular infections and
inflammatory disease. Washington, DC:
National Eye Institute; 1983:2472.
Weller T: Varicella-zoster virus. History,
perspective, and evolving concerns.
Neurology 1995; 45(Suppl 8):S9S10.
Pepose J: The potential impact of varicella
vaccine and new antivirals on ocular
disease related to varicella-zoster virus. Am
J Ophthalmol 1997; 123:243249.
Insinga RP, Itzler RF, Pellissier JM, et al:
The incidence of herpes zoster in a United
States administrative database. J Gen
Intern Med 2005; 20:748753.
Fleming DM, Cross KW, Cobb WA, et al:
Gender difference in the incidence of
shingles. Epidemiol Infect 2004; 132:15.
Nagasako EM, Johnson RW, Grifn DR,
et al: Geographic and racial aspects of
herpes zoster. J Med Virol 2003; 70(Suppl 1):
S20S23.
Schmader K, George LK, Burchett BM,
et al: Racial differences in the occurrence
of herpes zoster. J Infect Dis 1995;
171:701704.
Jumaan AO, Yu O, Jackson LA, et al:
Incidence of herpes zoster, before and after
varicella-vaccination-associated decreases
in the incidence of varicella, 19922002.
J Infect Dis 2005; 191:20022007.
Yih WK, Brooks DR, Lett SM, et al: The
incidence of varicella and herpes zoster in
Massachusetts as measured by the
Behavioral Risk Factor Surveillance System
(BRFSS) during a period of increasing
varicella vaccine coverage, 19982003.
BMC Public Health 2005; 5:68.
Goldman GS: Universal varicella
vaccination: efcacy trends and effect on
herpes zoster. Int J Toxicol 2005;
24:205213.
deMoragas J, Kierland R: The outcome of
patients with herpes zoster. Arch Dermatol
1957; 75:193.
Liesegang T: Varicella-zoster virus eye
disease [Review]. Cornea 1999; 18:511531.
Severson EA, Baratz KH, Hodge DO, et al:
Herpes zoster ophthalmicus in Olmsted
county, Minnesota: have systemic antivirals
made a difference? Arch Ophthalmol 2003;
121:386390.
Womack L, Liesegang T: Complications of
herpes zoster ophthalmicus. Arch
Burgoon C Jr, Burgoon J, Baldridge G: The
natural history of herpes zoster. JAMA
1957; 164:265.
Edgerton A: Herpes zoster ophthalmicus:
report of cases and review of literature.
Scott T: Epidemiology of herpetic
infections. Am J Ophthalmol 1957;
43:134.
Scheie H: Herpes zoster ophthalmicus.
Tr Ophth Soc UK 1970; 90:899.
Tunsuriyawong S, Puavilai S: Herpes
zoster, clinical course and associated
diseases: A 5-year retrospective study at
Tamathibodi Hospital. J Med Assoc Thai
2005; 88:678681.
Buntinx F, Wachana R, Bartholomeeusen S,
et al: Is herpes zoster a marker for occult
CHAPTER 49
697
409.
410.
411.
412.
413.
414.
415.
416.
417.
418.
419.
420.
421.
SECTION 6
422.
423.
424.
425.
426.
427.
428.
698
or subsequent malignancy? Br J Gen Pract

2005; 55:102107.
Thomas SL, Hall AJ: What does
epidemiology tell us about risk factors for
herpes zoster? Lancet Infect Dis 2004;
4:2633.
Bowsher D: The lifetime occurrence of
Herpes zoster and prevalence of postherpetic neuralgia: A retrospective survey
in an elderly population. Eur J Pain 1999;
3:335342.
Miller A: Selective decline in cellular
immune response to varicella-zoster in the
elderly. Neurology 1980; 30:582.
Gilden D, Varai A, Shtram Y, et al: Varicellazoster virus DNA in human sensory ganglia.
Nature 1983; 306:478480.
Esiri M, Tomlinson A: Herpes zoster:
Demonstration of virus in trigeminal nerve
and ganglion by immunofluorescence and
electron microscopy. J Neurol Sci 1972;
15:3538.
Bastian F, Rabson A, Yee C, et al: Herpes
virus varicella isolated from human dorsal
root ganglia. Arch Pathol 1973; 97:331336.
Dueland A: Latency and reactivation of
varicella zoster virus infections. Scand J
Infect Dis Suppl 1996; 100:4650.
Wilson A, Sharp M, Koropchak C, et al:
Subclinical varicella-zoster virus viremia,
herpes zoster, and T-lymphocyte immunity
to varicella-zoster viral antigens after bone
marrow transplantation. J Infect Dis 1992;
165:119126.
Gilden D, Dueland A, Devlin M, et al:
Varicella-zoster virus reactivation without
rash. J Infect Dis 1992; 166(Suppl 1):
S30S34.
Weigle KA, Grose C: Molecular dissection
of the humoral immune response to
individual varicella-zoster viral proteins
during chickenpox, quiescence, reinfection,
and reactivation. J Infect Dis 1984;
149:741749.
Gilden D, Wright R, Schneck S, et al:
Zoster sine herpete: a clinical variant. Ann
Neurol 1994; 35:530533.
Lewis G: Zoster sine herpete. Br Med J
1958; 2:418.
Hassegawa T: Further electron microscopic
observations on herpes zoster virus. Arch
Dermatol 1970; 103:45.
Muller SA, Winkelmann RK: Cutaneous
nerve changes in zoster. J Invest Dermatol
1969; 52:7177.
Liesegang T: Corneal complications from
herpes zoster ophthalmicus. Ophthalmol
1985; 92:316.
Liesegang T: Herpes zoster ophthalmicus.
Pavan-Langston D: Cornea and external
disease, in manual of ocular diagnosis and
therapy. 5th edn. In: Pavan-Langston D, ed.
Philadelphia, PA: Lippincott, Williams &
Wilkins; 2002:67129.
Starr C, Pavan-Langston D: Varicella-zoster
virus: mechanisms of pathogenicity and
corneal disease. Ophthalmol Clin North Am
2002; 15:715.
Wenkel H, Rummelt C, Rummelt V, et al:
Detection of varicella zoster virus DNA and
viral antigen in human cornea after herpes
zoster ophthalmicus. Cornea 1993;
12:131137.
Hedges T III, Albert D: The progression of
the ocular abnormalities of herpes zoster:
histopathologic observations of nine cases.
429. Nagashima K, Nakazawa M, Endo H:

Pathology of the human spinal ganglia in
varicella-zoster virus infection. Acta
Neuropath (Berl) 1975; 33:105117.
430. Ghatak N, Zimmerman H: Spinal ganglion
in herpes zoster. Arch Pathol 1973;
95:411415.
431. Reske-Nielsen E, Oster S, Pedersen B:
Herpes zoster ophthalmicus and the
mesencephalic nucleus. A
neuropathological study. Acta Pathol
Microbiol Immunol Scand [A] 1986;
94:263269.
432. Linnemann CC Jr, Alvira MM: Pathogenesis
of varicella-zoster angiitis in the CNS. Arch
Neurol 1980; 37:239240.
433. Rosenblum WI, Hadeld MG:
Granulomatous angiitis of the nervous
system in cases of herpes zoster and
lymphosarcoma. Neurology 1972;
22:348354.
434. Doyle P, Gibson G, Dohlman C: Herpes
zoster ophthalmicus with contralateral
hemiplegia: Identication of cause. Ann
Neurol 1983; 14:8485.
435. Bhat G, Mathur D, Saxena G, et al:
Granulomatous angiitis of the central
nervous system associated with herpes
zoster. J Assoc Physicians India 2002;
50:977978.
436. Gilden D, Devlin M, Wellish M, et al:
Persistance of varicella-zoster virus DNA in
blood mononuclear cells of patients with
varicella or zoster. Virus Genes 1989;
2:299305.
437. Green W, Zimmerman L: Granulomatous
reaction of Descemets membrane. Am J
438. Wenkel H, Rummelt V, Fleckenstein B, et al:
Detection of varicella zoster virus DNA and
viral antigen in human eyes after herpes
zoster ophthalmicus. Ophthalmology 1998;
105:13231330.
439. Naumann G, Gass J, Font R:
Histopathology of herpes zoster
ophthalmicus. Am J Ophthalmol 1968;
65:533.
440. Zaal MJ, Maudgal PC, Rietveld E, et al:
Chronic ocular zoster. Curr Eye Res 1991;
10(Suppl):125130.
441. Juel-Jensen B, MacCallum F: Herpes simplex,
varicella and zoster: clinical manifestations
and treatment. In: Juel-Jensen B, ed.
Philadelphia, PA: Lippincott; 1972.
442. Zaal MJ, Volker-Dieben HJ, DAmaro J:
Prognostic value of Hutchinsons sign in
acute herpes zoster ophthalmicus. Graefes
Arch Clin Exp Ophthalmol 2003;
241:187191.
443. Watson C, Deck J, Morshead C, et al:
Post-herpetic neuralgia: further postmortem studies of cases with and without
pain. Pain 1991; 44:105117.
444. Thomas M, Robertson W: Dermal
transmission of a virus as a cause of
shingles. Lancet 1968; 2:1349.
445. Liesegang T: Varicella-zoster virus:
systemic and ocular features. J Am Acad
Dermatol 1984; 11:165.
446. Pavan-Langston D: Ophthalmic zoster. In:
Cp W, Gershon A, eds. Amsterdam:
Elsevier Science; 2001:119129.
447. Pavan-Langston D: Clinical manifestations
and therapy of herpes zoster ophthalmicus.
Compr Ophthalmol Update 2002;
3:217225.
448. Obata H, Yamagami S, Saito S, et al: A
case of acute dacryoadenitis associated
449.
450.
451.
452.
453.
454.
455.
456.
457.
458.
459.
460.
461.
462.
463.
464.
465.
466.
467.
468.
with herpes zoster ophthalmicus. Jpn J

Ophthalmol 2003; 47:107109.
Yamamoto S, Tada R, Shimomura Y, et al:
Detecting varicella-zoster virus DNA in
iridocyclitis using polymerase chain
reaction: a case of zoster sine herpete.
Dolin R, Reichman R, Mazur M, et al:
Herpes zoster varicella infections in
immunosuppressed patients. Ann Int Med
1978; 89:375.
Ragozzino M, Melton M, Kurland L, et al:
Population based study of herpes zoster
and its sequellae. Medicine 1982; 61:310.
Yamamoto S, Shimomura Y, Kinoshita S,
et al: Differentiating zosteriform herpes
simplex from ophthalmic zoster. Arch
Ophthalmol 1994; 112:15151516.
Binder NR, Holland GN, Hosea S, et al:
Herpes zoster ophthalmicus in an
otherwise-healthy child. J Aapos 2005;
9:597598.
Naseri A, Good WV, Cunningham ET Jr:
Herpes zoster virus sclerokeratitis and
anterior uveitis in a child following varicella
vaccination. Am J Ophthalmol 2003;
135:415417.
Forrest W, Kaufman H: Zosteriform herpes
simplex. Am J Ophthalmol 1976; 81:86.
Marsh R, Cooper M: Ophthalmic zoster:
mucous plaque keratitis. Br J Ophthalmol
1987; 71:725728.
Piebenga L, Laibson P: Dendritic lesions in
herpes zoster ophthalmicus. Arch
Zaal M, Volker-Dieben H, Wienesen M,
et al: Longitudinal analysis of varicellazoster virus DNA on the ocular surface
associated with herpes zoster
ophthalmicus. Am J Ophthalmol 2001;
131:2529.
Pavan-Langston D, Yamamoto S, Dunkel
EC: Delayed herpes zoster
pseudodendrites. Polymerase chain
reaction detection of viral DNA and a role
for antiviral therapy. Arch Ophthalmol 1995;
113:13811385.
Engstrom RE, Holland GN: Chronic herpes
zoster virus keratitis associated with the
acquired immunodeciency syndrome. Am
J Ophthalmol 1988; 105:556559.
Chern KC, Conrad D, Holland GN, et al:
Chronic varicella-zoster virus epithelial
keratitis in patients with acquired
immunodeciency syndrome. Arch
Ophthalmol 1998; 116:10111017.
Straus S: Clinical and biological differences
between recurrent herpes simplex virus
and varicella-zoster virus infections. JAMA
1989; 262:34553458.
Crock G: Clinical syndromes of anterior
segment ischaemia. Tr Ophthalmol Soc UK
1967; 87:513.
Marsh R: Ophthalmic herpes zoster. Br J
Hosp Med 1976; 15:609618.
Reijo A, Antti V, Jukka M: Endothelial cell
loss in Herpes zoster keratouveitis. Br J
Liesegang T: Varicella zoster viral disease.
In: Tasman W, Haeger E, eds. Philadelphia,
PA: Lippincott-Raven; 1991:110.
Maudgal PC, Missotten L, De Clercq E,
et al: Varicella-zoster virus in the human
corneal endothelium: a case report. Bull
Soc Belge Ophtalmol 1980; 190:7186.
Mietz H, Eis-Hubinger AM, Sundmacher R,
et al: Detection of varicella-zoster virus
DNA in keratectomy specimens by use of
469.
470.
471.
472.
473.
474.
475.
476.
477.
478.
479.
480.
481.
482.
483.
484.
485.
486.
487.
the polymerase chain reaction. Arch

Ophthalmol 1997; 115:590594.
Silverstein BE, Chandler D, Neger R, et al:
Disciform keratitis: a case of herpes zoster
sine herpete. Am J Ophthalmol 1997;
123:254255.
Jones D: Herpes zoster ophthalmicus. In:
Golden B, ed. Springeld, IL: Charles C.
Thomas; 1974:198.
Mondino B, Brown S, Mondzelewski J:
Peripheral corneal ulcers with herpes
zoster ophthalmicus. Am J Ophthalmol
1978; 86:611.
Head H, Campbell A: The pathology of
Herpes zoster and its bearing on sensory
localization. Brain 1900; 23:379.
Mackie I: Role of the corneal nerves in
destructive disease of the cornea. Trans
Waring G, Ekins M: Corneal perforation in
herpes zoster ophthalmicus caused by
eyelid scarring with exposure keratitis. In:
Sundmacher R, ed. Munich: J F Bergman;
1981:469.
Marsh RJ, Cooper M: Ocular surgery in
ophthalmic zoster. Eye 1989; 3(Pt 3):
313317.
Liesegang TJ: Diagnosis and therapy of
herpes zoster ophthalmicus.
Marsh R, Easty D, Jones B: Iritis and iris
atrophy in herpes zoster ophthalmicus. Am
Miserocchi E, Waheed NK, Dios E, et al:
Visual outcome in herpes simplex virus and
varicella zoster virus uveitis: a clinical
evaluation and comparison. Ophthalmology
2002; 109:15321537.
Kezuka T, Sakai J, Minoda H, et al: A
relationship between varicella-zoster
virus-specic delayed hypersensitivity and
varicella-zoster virus-induced anterior uveitis.
Kashiwase M, Sakai J, Usui M: Uveitis
associated with zoster sine herpete.
Diagnosis and clinical ndings. Nippon
Ganka Gakkai Zasshim (Japanese) 2000;
104:97102.
Nakamura M, Tanabe M, Yamada Y, et al:
Zoster sine herpete with bilateral ocular
involvement. Am J Ophthalmol 2000;
129:809810.
Furuta Y, Ohtani F, Mesuda Y, et al:
Early diagnosis of zoster sine herpete and
antiviral therapy for the treatment of facial
palsy. Neurology 2000; 55:708710.
Tabery H: Corneal epithelial keratitis in
herpes zoster ophthalmicus: delayed and
sine herpete. A non-contact
photomicrographic in vivo studyin the
human cornea. Eur J Ophthalmol 2002;
12:267275.
Tabery H: Morphology of epithelial keratitis
in herpes zoster ophthalmicus. A noncontact photomicrographic in vivo study in
the human cornea. Acta Ophthalmol Scand
2000; 78:651655.
Eaglstein WH, Katz R, Brown JA: The effect
of early corticosteroid therapy on the skin
eruptions and pain of herpes zoster. JAMA
1970; 211:16811683.
Pavan-Langston D: Varicella-zoster
ophthalmicus. in Ocular Viral Disease,
Internat. Ophthalmol. Clin, Ed. PavanLangston, D, Boston, Little Brown and Co;
1975:171185.
Marsh R, Dudley B, Kelly V: External
ocular motor palsies in ophthalmic zoster:
488.
489.
490.
491.
492.
493.
494.
495.
496.
497.
498.
499.
500.
501.
502.
503.
504.
a review. Br J Ophthalmol 1977;

61:677682.
Archambault P, Wise JS, Rosen J, et al:
Herpes zoster ophthalmoplegia. Report of
six cases. J Clin Neuroophthalmol 1988;
8:185193.
Shin HM, Lew H, Yun YS: A case of
complete ophthalmoplegia in herpes zoster
ophthalmicus. Korean J Ophthalmol 2005;
19:302304.
Al-Abdulla N, Rismondo V, Minkowski J,
et al: Herpes zoster vasculitis presenting as
giant cell arteritis with bilateral internuclear
ophthalmoplegia. Am J Ophthalmol 2002;
134:912914.
Hallas P: Pupillary paresis. A rare
complication of varicella zoster eye
infection. Ugeskr Laeger (Danish) 2001;
163:58355836.
Pandey PK, Chaudhuri Z, Sharma P:
Extraocular muscle and facial paresis in
herpes zoster ophthalmicus. J Pediatr
Pandey PK, Garg D, Bhatia A, et al:
Horners syndrome and sixth nerve palsy
due to herpes zoster ophthalmicus arteritis.
Eye 2005; 19:224226.
Sodhi P, Goel J: Presentations of cranial
nerve involvement in two patients with
Herpes zoster ophthalmicus. J Commun
Dis 2001; 33:130135.
Krasnianski M, Sievert M, Bau V, et al:
External ophthalmoplegia due to ocular
myositis in a patient with ophthalmic
herpes zoster. Neuromuscul Disord 2004;
14:438441.
Kawasaki A, Borruat FX: An unusual
presentation of herpes zoster
ophthalmicus: orbital myositis preceding
vesicular eruption. Am J Ophthalmol 2003;
136:574575.
Engstrom R, Holland G, Margolis T, et al:
The progressive outer retinal necrosis
syndrome. A variant of necrotizing herpetic
retinopathy in paitents with AIDS.
Ophthalmology 1994; 101:14881502.
Oueghlani E, Baglivo E, Durakovic O, et al:
Varicella-zoster virus retinitis: successful
evolution with a combination of antiviral
therapies. Klin Monatsbl Augenheilkd 2005;
222:264266.
Scott FT, Leedham-Green ME, BarrettMuir WY, et al: A study of shingles and the
development of postherpetic neuralgia in
East London. J Med Virol 2003; 70(Suppl 1):
S24S30.
Haanpaa M, Laippala P, Nurmikko T:
Allodynia and pinprick hypesthesia in acute
herpes zoster, and the development of
postherpetic neuralgia. J Pain Symptom
Manage 2000; 20:5058.
Zaal M, Volker-Dieben H, DAmaro J: Risk
and prognostic factors of postherpetic
neuralgia and focal sensory denervation:
a prospective evaluation in acute herpes
zoster ophthalmicus. Clin J Pain 2000;
16:345351.
Oaklander AL, Bowsher D, Galer B, et al:
Herpes zoster itch: preliminary
epidemiologic data. J Pain 2003; 4:338343.
Opstelten W, Mauritz JW, de Wit NJ, et al:
Herpes zoster and postherpetic neuralgia:
incidence and risk indicators using a
general practice research database. Fam
Pract 2002; 19:471475.
Dworkin RH, Boon RJ, Grifn DR, et al:
Postherpetic neuralgia: impact of
famciclovir, age, rash severity, and acute
505.
506.
507.
508.
509.
510.
511.
512.
513.
514.
515.
516.
517.
518.
519.
520.
521.
522.
523.
524.
pain in herpes zoster patients. J Infect Dis

1998; 178(Suppl 1):S76S80.
Watson C: Postherpetic neuralgia: the
importance of preventing this intractable
end-stage disorder Review. J Infect Dis
1998; 178(Suppl 1): S91S94.
Pearce JM: Post herpetic neuralgia.
J Neurol Neurosurg Psychiatry 2005;
76:572.
Watson P, Evans R: Postherpetic neuralgia.
A review. Arch Neurol 1986; 43:836840.
Harding S, Lipton J, Wells J, et al: Relief of
acute pain in herpes zoster ophthalmicus
by stellate ganglion block. Br Med J 1986;
292:1428.
Bowsher D: Trigeminal neuralgia: an
anatomically oriented review. Clin Anat
1997; 10:409415.
Feller L, Jadwat Y, Bouckaert M: Herpes
zoster post-herpetic neuralgia. Sadj 2005;
60:432,436437.
Liesegang TJ: Ophthalmic herpes zoster:
diagnosis and antiviral therapy. Geriatrics
1991; 46:6466, 6971.
Oaklander AL, Cohen SP, Raju SV:
Intractable postherpetic itch and cutaneous
deafferentation after facial shingles. Pain
2002; 96:912.
Oaklander AL: The density of remaining
nerve endings in human skin with and
without postherpetic neuralgia after
shingles. Pain 2001; 92:139145.
Watson CP, Oaklander AL, Deck J: The
neuropathologyof herpes zoster with
particular reference to postherpetic
neuralgia and its pathogenesis. In: Watson
CP, Gershon AA, eds. Amsterdam:
Elsevier-Science; 2001:168182.
Oaklander AL, Romans K, Horasek S, et al:
Unilateral postherpetic neuralgia is
associated with bilateral sensory neuron
damage. Ann Neurol 1998; 44:789795.
Pappagallo M, Oaklander AL, QuatranoPiacentini AL, et al: Heterogenous patterns
of sensory dysfunction in postherpetic
neuralgia suggest multiple
pathophysiologic mechanisms.
Gilden DH, Cohrs RJ, Hayward AR, et al:
Chronic varicella-zoster virus ganglionitis
a possible cause of postherpetic neuralgia.
J Neurovirol 2003; 9:404407.
Gilden DH, Cohrs RJ, Mahalingam R: VZV
vasculopathy and postherpetic neuralgia:
progress and perspective on antiviral
therapy. Neurology 2005; 64:2125.
Vafai A, Murray R, Wellish M, et al:
Expression of varicella-zoster virus and
herpes simplex virus in normal human
trigeminal ganglia. Proc Natl Acad Sci USA
1988; 85:23622366.
Bennett G: Hypotheses on the pathogenesis
of herpes zoster-associated pain. Annals of
Neurol 1994; 35(Suppl):S38S41.
Watson CP: Management issues of
neuropathic trigeminal pain from a medical
perspective. J Orofac Pain 2004;
18:366373.
Watson CP, Morshead C, Van der Kooy D,
et al: Post-herpetic neuralgia: post-mortem
analysis of a case. Pain 1988; 34:129138.
Oster G, Harding G, Dukes E, et al: Pain,
medication use, and health-related quality
of life in older persons with postherpetic
neuralgia: results from a population-based
survey. J Pain 2005; 6:356363.
Cole E, Meisler D, Calabrese D, et al:
Herpes zoster ophthalmicus and acquired
CHAPTER 49
699
525.
526.
527.
528.
529.
530.
531.
532.
533.
534.
535.
SECTION 6
536.
537.
538.
539.
540.
541.
700
immune deciency syndrome. Arch

Ophthalmol 1984; 102:1027.
Seiff S, Margolis T, Graham S, et al: Use of
intravenous acyclovir for treatment of
herpes zoster ophthalmicus in patients at
risk for AIDS. Ann Ophthalmol 1988;
20:480.
Freidmann-Kein A, LaFleur F, Gendler E,
et al: Herpes zoster: a possible clinical sign
for development of acquired immune
deciency syndrome in high risk individuals.
J Am Acad Dermatol 1986; 14:10231028.
Sandor E, Millman A, Croxson S, et al:
Herpes zoster ophthalmicus in patients at
risk for the acquired immune deciency
syndrome (AIDS). Am J Ophthalmol 1986;
101:153.
Sandor E, Schnall C, Croxson T, et al:
Herpes zoster ophthalmicus in young
patients and its correlation with HTLV-III
sensitivity. ARVO Abstracts 1987; 28:46.
Pepose JS: External ocular herpesvirus
infections in immunodeciency. Current Eye
Research 1991; 10(Suppl):8795.
Hung CC, Hsiao CF, Wang JL, et al: Herpes
zoster in HIV-1-infected patients in the era
of highly active antiretroviral therapy: a
prospective observational study. Int J STD
AIDS 2005; 16:673676.
Gebo KA, Kalyani R, Moore RD, et al: The
incidence of, risk factors for, and sequelae
of herpes zoster among HIV patients in the
highly active antiretroviral therapy era.
J Acquir Immune Dec Syndr 2005;
40:169174.
Vanhems P, Voisin L, Gayet-Ageron A, et al:
The incidence of herpes zoster is less likely
than other opportunistic infections to be
reduced by highly active antiretroviral
therapy. J Acquir Immune Dec Syndr
2005; 38:111113.
Evans BG: Management of blinding
disease: loss of immunity and
superinfection. Eye 2005; 19:10351036.
Litoff D, Catalano R: Herpes zoster optic
neuritis in human immunodeciency
virus infection. Arch Ophthalmol 1990;
108:782.
Thomson KJ, Hart DP, Banerjee L, et al:
The effect of low-dose aciclovir on
reactivation of varicella zoster virus after
allogeneic haemopoietic stem cell
transplantation. Bone Marrow Transplant
2005; 35:10651069.
McGill J, Chapman C: Comparison of
topical acyclovir with steroids in the
treatment of herpes zoster keratouveitis.
Borruat F, Buechi E, Piguet F, et al:
Prevention of ocular complications of
herpes zoster ophthalmicus by adequate
treatment with acyclovir. Klin Monatsbl
Augenheilkd 1991; 198:358360.
acyclovir in the therapy of acute herpes
zoster ophthalmicus: an interim report.
acyclovir in the treatment of acute herpes
zoster ophthalmicus. Ophthalmology 1986;
93:763.
Harding S, Porter S: Oral acyclovir in
1991; 10:177182.
Hoang-Xuan T, Buchi ER, Herbort CP, et al:
Oral acyclovir for herpes zoster
99:10621071.
542. Aylward G, Claoue C, Marsh R, et al:

Influence of oral acyclovir on ocular
complications of herpes zoster
ophthalmicus. Eye 1994; 8:7074.
543. Lilie HM, Wassilew S: The role of antivirals
in the management of neuropathic pain in
the older patient with herpes zoster. Drugs
Aging 2003; 20:561570.
544. McGill J, White J: Acyclovir and postherpetic neuralgia and ocular involvement.
Br Med J 1994; 309:11241128.
545. Manka R: Exogenous lactase in the
treatment of oral acyclovir intolerance. Am
546. Gross G, Schofer H, Wassilew S, et al:
Herpes zoster guideline of the German
Dermatology Society (DDG). J Clin Virol
2003; 26:277289; discussion 291273.
547. Dias C, Nashed Y, Atluri H, et al: Ocular
penetration of acyclovir and its peptide
prodrugs valacyclovir and val-valacyclovir
following systemic administration in
rabbits: An evaluation using ocular
microdialysis and LC-MS. Curr Eye Res
2002; 25:243252.
548. Colin J, Prisant O, Cochener B, et al: A
double blind randomized trial to compare
the efcacy and safety of valaciclovir and
acyclovir for treatment of herpes zoster
107:15071511.
549. Fiddian A: Antiviral drugs in development
for herpes zoster. Scand J Infect Dis Suppl
1996; 100:5154.
550. Tyring S, Beutner K, Tucker B, et al:
Antiviral therapy for herpes zoster:
randomized controlled clinical trial of
valacyclovir and famciclovir therapy in
immunocompetent patients 50 years and
older. Arch Fam Med 2000; 9:863869.
551. Tyring S, Douglas JJ, Corey L, et al:
A randomized, placebo-controlled
comparison of oral valaciclovir and
acyclovir in immunocompetent patients
with recurrent genital herpes infections.
Arch Dermatol 1998; 134:185191.
Comparison of the efcacy and safety of
valaciclovir and acyclovir for the treatment
of herpes zoster ophthalmicus.
Ophthalmology 2000; 107:15071511.
553. Lebrun-Vignes B: [Valaciclovir]. Ann
554. Boyd M, Safrin S, Kern E: Penciclovir: a
review of spectrum of activity, selectivity,
and cross-resistance pattern. Antiviral
Chem. Chemother 1993; 4(Suppl 1):2536.
555. Goldani LZ: Treatment of severe infectious
mononucleosis with famciclovir. J Infect
2002; 44:9293.
556. Vere Hodge R, Cheng Y-C: The mode of
action of penciclovir. Antiviral Agents
Chemother 1993; 4(Suppl 1):1324.
557. deGreef H: Famciclovir, a new oral
antiherpes drug: results of the rst
controlled clincial study demontrating its
efcacy and safety in the treatment of
uncomplicated herpes zoster in
immunocompetent patients. Int J
558. deGreef H: Famciclovir, a new oral antiviral
drug: its efcacy and safety in the
treatment of uncomplicated herpes zoster
in immunocompetent patients. Int J
559. Gheeraert P, Group: Efcacy and safety of
famciclovir in the treatment of
uncomplicated herpes zoster. Program and
560.
561.
562.
563.
564.
565.
566.
567.
568.
569.
570.
571.
572.
573.
574.
575.
576.
577.
Abstracts, 32nd ICAAC. Washington, DC:

1992:1108.
Saltzman R, Jurewicz R, Boon R: Safety of
famciclovir in patients with herpes zoster
and genital herpes. Antimicrob Agents
Chemother 1994; 38:24542457.
Tyring S: Efcacy of famciclovir in the
treatment of herpes zoster. Semin Dermatol
1996; 15(2 Suppl 1):2731.
Tyring S: Famiciclovir therapy (Famvir)
for herpes simplex and herpes zoster
infections. Skin Therapy Lett 2001;
6:12,5.
Tyring S, Belanger R, Bezwoda W, et al: A
randomized, double-blind trial of
famciclovir versus acyclovir for the
treatment of localized dermatomal herpes
zoster in immunocompromised patients.
Cancer Invest 2001; 19:1322.
Tyring SK, Diaz-Mitoma F, Shafran SD,
et al: Oral famciclovir for the suppression
of recurrent genital herpes: the combined
data from two randomized controlled trials.
J Cutan Med Surg 2003; 7:449454.
Tyring S, Engst R, Corriveau C, et al:
Famciclovir for ophthalmic zoster: a
randomised aciclovir controlled study.
Wassilew SW, Wutzler P: Oral brivudin in
comparison with acyclovir for herpes
zoster: a survey study on postherpetic
neuralgia. Antiviral Res 2003; 59:5760.
Wood MJ, Shukla S, Fiddian AP, et al:
Treatment of acute herpes zoster: effect of
early (<48 h) versus late (4872 h) therapy
with acyclovir and valaciclovir on prolonged
pain. J Infect Dis 1998; 178(Suppl 1):
S81S84.
Whitley R, Weiss H, Gnann JJ, et al:
Acyclovir with and without prednisone for
the treatment of herpes zoster. A
randomized, placebo-controlled trial. The
National Institute of Allergy and Infectious
Diseases Collaborative Antiviral Study
Group. Ann Intern Med 1996; 125:376383.
Wood M, Johnson R, McKendrick M, et al:
A randomized trial of acyclovir for 7 days
and 21 days with and without prednisolone
for treatment of acute herpes zoster. N Engl
J Med 1994; 330:896900.
Dworkin RH, Johnson RW, Breuer J, et al:
Guidelines for the Management of Herpes
Zoster. Clin Infect Dis 2007; 44:S1S26.
Merselis J, Kaye D, Hook E: Disseminated
herpes zoster. Arch Intern Med 1961;
113:679.
Argoff CE: A focused review on the use of
botulinum toxins for neuropathic pain. Clin
J Pain 2002; 18:S177S181.
Argoff CE, Katz N, Backonja M: Treatment
of postherpetic neuralgia: a review of
therapeutic options. J Pain Symptom
Manage 2004; 28:396411.
Hempenstall K, Nurmikko TJ, Johnson RW,
et al: Analgesic therapy in postherpetic
neuralgia: a quantitative systematic review.
PLoS Med 2005; 2:164.
Levin J, Gordon N, Smith R, et al:
Desipramine enhances opiate
postoperative analgesia. Pain 1986;
27:45.
Urban B, France R, Steinberger E, et al:
Longer term use of narcotic/antidepressant
medication in the management of phantom
limb pain. Pain 1986; 24:191.
Hardy D: Relief of pain in acute herpes
zoster by nerve blocks and possible
578.
579.
580.
581.
582.
583.
584.
585.
586.
587.
588.
589.
590.
591.
592.
593.
594.
595.
596.
597.
598.
prevention of post-herpetic neuralgia. Can

J Anaesth 2005; 52:186190.
Wasner G, Kleinert A, Binder A, et al:
Postherpetic neuralgia: topical lidocaine is
effective in nociceptor-deprived skin.
J Neurol 2005; 252:677686.
Bowsher D: The effects of preemptive
treatment of postherpetic neuralgia with
amitriptyline: a randomized, double-blind,
placebo-controlled trial. J Pain Sympt
Manage 1997; 13:327331.
Satterthwaite J, Tollison C, Kriegel M: Use
of tricyclic antidepressants for the
treatment of intractable pain. Compr Ther
1990; 16:10.
Watson C, Evans R, Reed K: Amitriptyline
vs placebo in post-herpetic neuralgia.
Neurology 1982; 32:671.
Watson C, Vernich L, Chipman M, et al:
Nortriptyline versus amitriptyline in
postherpetic neuralgia: a randomized trial.
Neurology 1998; 51:11661171.
Watson CP, Vernich L, Chipman M, et al:
Nortriptyline versus amitriptyline in
postherpetic neuralgia: a randomized trial.
Neurology 1998; 51:11661171.
Kvinesdal B, Molin J, Froland A, et al:
Imipramine treatment of painful diabetic
neuropathy. JAMA 1984; 251:1727.
Bowsher D: Factors influencing the
features of postherpetic neuralgia and
outcome when treated with tricyclics. Eur J
Pain 2003; 7:17.
Hurtig H: Fluphenazine and post-herpetic
neuralgia. JAMA 1990; 263:2750.
Murphy T: Post-herpetic neuralgia. JAMA
1989; 262:34783481.
Dworkin R, Schmader K: Treatment and
prevention of postherpetic neuralgia.
Review. Clin Infect Dis 2003; 36:877882.
Galen B: Neuropathic pain of peripheral
origin: advances in pharmacologic
treatment. Neurology 1995; 45(Suppl 9):
S17S25.
Panlilio LM, Christo PJ, Raja SN: Current
management of postherpetic neuralgia.
Neurologist 2002; 8:339350.
Berger A, Dukes E, McCarberg B, et al:
Change in opioid use after the initiation of
gabapentin therapy in patients with
postherpetic neuralgia. Clin Ther 2003;
25:28092821.
Taylor CP, Gee NS, Su TZ, et al: A
summary of mechanistic hypotheses of
gabapentin pharmacology. Epilepsy Res
1998; 29:233249.
US Pharmcopeia Gabapentin. In:
Greenwood Village. Colorado: Thomson
Micromedex Inc; 2005.
Stacey BR, Glanzman RL: Use of gabapentin
for postherpetic neuralgia: results of two
randomized, placebo-controlled studies. Clin
Ther 2003; 25:25972608.
Rowbotham M, Harden N, Stacey B, et al:
Gabapentin for the treatment of
postherpetic neuralgia: a randomized
controlled trial. JAMA 1998; 280:18371842.
Berry JD, Petersen KL: A single dose of
gabapentin reduces acute pain and
allodynia in patients with herpes zoster.
Neurology 2005; 65:349350.
Jean WH, Wu CC, Mok MS, et al: Starting
dose of gabapentin for patients with postherpetic neuralgiaa dose-response study.
Acta Anaesthesiol Taiwan 2005; 43:7377.
Frampton JE, Foster RH: Pregabalin: in the
treatment of postherpetic neuralgia. Drugs
2005; 65:111118; discussion 119120.
599. Hadj Tahar A: Pregabalin for peripheral

neuropathic pain. Issues Emerg Health
Technol 2005; 14.
600. Dworkin R, Corbin A, Young JJ, et al:
Pregabalin for the treatment of postherpetic
neuralgia: A randomized, placebo-controlled
trial. Neurology 2003; 60:12741283.
601. Argoff CE: Targeted topical peripheral
analgesics in the management of pain. Curr
Pain Headache Rep 2003; 7:3438.
602. Argoff CE, Galer BS, Jensen MP, et al:
Effectiveness of the lidocaine patch 5% on
pain qualities in three chronic pain states:
assessment with the Neuropathic Pain
Scale. Curr Med Res Opin 2004; 20(Suppl
2):S21S28.
603. Gain P, Thuret G, Chiquet C, et al: Frontal
and nasal nerve blocks in the treatment of
severe pain in acute ophthalmic zoster.
Anesth Analg 2002; 95:503.
604. Gain P, Thuret G, Chiquet C, et al: Facial
anesthetic blocks in the treatment of acute
pain during ophthalmic zoster. J Fr
Ophtalmol (French) 2003; 26:714.
605. Borodic GE, Acquadro MA: The use of
botulinum toxin for the treatment of chronic
facial pain. J Pain 2002; 3:2127.
606. Sangwan VS, Matalia HP, Vemuganti GK,
et al: Early results of penetrating
keratoplasty after cultivated limbal
epithelium transplantation. Arch
Ophthalmol 2005; 123:334340.
607. Marsh RJ, Marshall J: Treatment of lipid
keratopathy with the argon laser. Br J
Ophthalmol 1982.
608. Reed J, Joyner S, Knauer W III: Penetrating
keratoplasty for Herpes zoster keratopathy.
609. Tanure M, Cohen E, Grewal S, et al:
Penetrating keratoplasty for varicella-zoster
virus keratopathy. Cornea 2000; 19:135139.
610. Pavan-Langston D, Dohlman CH: Boston
keratoprosthesis treatment of herpes zoster
neurotrophic keratopathy. Am J Ophthal
2007; in press.
611. Wack R: An outbreak of varicella despite
vaccination. N Engl J Med 2003;
348:14051407.
612. Aaberg T, OBrien W: Expanding
ophthalmologic recognition of Epstein-Barr
virus infections. Am J Ophthalmol 1987;
104:420.
613. Henle G, Henle W: Observations on
childhood infections with the Epstein-Barr
virus. J Infect Dis 1970; 121:303310.
614. Ogawa G, Vastine D: Nonherpetic viral
keratitis. In: Tabbara K, Hyndiuk R, eds.
615. Johanssen E, Schooley R, Kaye K: EpsteinBarr Virus. In: Mandell G, Bennett J, Dolin
R, eds. Philadelphia, PA: Elsevier, Churchill
616. McCallum R: Infectious mononucleosis and
the Epstein-Barr virus. J Infect Dis 1970;
121:347.
617. Stagno S, Whitley R: Herpes virus infections
of pregnancy. Parts I and II: Cytomegalovirus
and Epstein Barr virus infections. N Engl J
Med 1985; 313:12701275,13271232.
618. Henle W, Henle G: Serodiagnosis of
infectious mononucleosis. Resident Staff
Phys 1981; 27:37.
619. Matoba A, McCulley J: Epstein-Barr virus
and its ocular manifestations. In: Darrell R,
ed. Philadelphia, PA: Lea & Febiger; 1985.
620. Matoba A, Wilhelmus K, Jones D: EpsteinBarr viral stromal keratitis. Ophthalmology
1986; 93:746-750.
621. Niederman J, Evans A, Subrahmanyan L,

et al: Prevelance, incidence and
persistence of EB virus antibody in young
adults. N Engl J Med 1970; 272:361.
622. Meisler D, Bosworth D, Krachmer J: Ocular
infectious mononucleosis manifested as
Parinauds oculoglandular syndrome. Am J
623. Wong K, DAmico D, Hedges TI, et al:
Ocular involvement associated with chronic
Epstein-Barr virus disease. Arch
624. Kanafani ZA, Bashur Z, Kanj SS: Acute
Epstein-Barr virus infection causing
bilateral conjunctival hemorrhages. South
Med J 2005; 98:390391.
625. Sugiyama K, Ito M, Ichimi R, et al: A case
of Epstein-Barr virus infection with
exophthalmos and ocular muscle swelling.
Acta Paediatr Jpn 1997; 39:694697.
626. Widmer S, Tinguely M, Egli F, et al: Lethal
Epstein-Barr virus associated NK/T-cell
lymphoma with primary manifestation in the
conjunctiva. Klin Monatsbl Augenheilkd
2005; 222:255257.
627. Woog JJ, Kim YD, Yeatts RP, et al: Natural
killer/T-cell lymphoma with ocular and
adnexal involvement. Ophthalmology 2006;
113:140147.
628. Merayo-Lloves J, Baltalzis S, Foster CS:
Epstein-Barr virus dacryoadenitis resulting
in keratoconjunctivitis sicca in a child. Am J
Ophthalmol 2001; 132:922923.
629. Pflugfelder S, Crouse C, Atherton S:
Ophthalmic manifestations of Epstein-Barr
virus infection. Int Ophthal Clin 1993;
33:95101.
630. Pflugfelder S, Huang C, Crouse C: EpsteinBarr virus keratitits after a facial chemical
peel. Am J Ophthalmol 1990; 110:571.
631. Pflugfelder S, Tseng S, Pepose J, et al:
Epstein-Barr virus infection and
immunologic dysfunction in patients with
aqueous tear deciency. Ophthalmology
1990; 97:313.
632. Pinnolis M, McCulley J: Nummular keratitis
associated with infectious mononucleosis.
633. Morishima N, Miyakawa S, Akazawa Y,
et al: A case of uveitis associated with
chronic active Epstein-Barr virus infection.
634. Salzberger B, Hartmann P, Hanses F, et al:
Incidence and prognosis of CMV disease in
HIV-infected patients before and after
introduction of combination antiretroviral
therapy. Infection 2005; 33:345349.
635. Nankervis G, Kumar M: Diseases produced
by cytomegaloviruses. Med Clin North Am
1978; 62:1021.
636. Onorato I, Morens D, Martone W, et al:
Epidemiology of cytomegalovirus
infections. Recommendation for prevention
and control. Rev Infect Dis 1985; 7:479.
637. Crumpacker C, Wadwha S:
Cytomegalovirus. In: Mandell G, Bennett J,
Dolin R, eds. Philadelphia, PA: Elsevier,
638. Brody JM, Butrus SI, Laby DM, et al:
Anterior segment ndings in AIDS patients
with cytomegalovirus retinitis. Graefes Arch
639. Garau J, Kabins S, DeNassquo S, et al:
Spontaneous cytomegalovirus
mononucleosis with conjunctivitis. Arch Int
Med 1977; 137:1631.
640. Brown HH, Glasgow BJ, Holland GN, et al:
Cytomegalovirus infection of the
CHAPTER 49
701
641.
642.
643.
644.
645.
646.
647.
648.
649.
650.
651.
652.
653.
654.
SECTION 6
655.
656.
657.
658.
659.
660.
702
conjunctiva in AIDS. Am J Ophthalmol

1988; 106:102104.
Walter K, Coulter V, Palay D, et al: Corneal
endothelial deposits in patients with CMV
retinitis. Am J Ophthalmol 1996;
121:391396.
Mitchell S, Barton K, Lightman F: Corneal
enothelial changes in CMV retinitis. Eye
1994; 8:4143.
Weller T: The cytomegaloviruses: ubiquitous
agents with protean clinical manifestations.
N Engl J Med 1971; 285:203.
Holland E, Bennett S, Brannian R, et al:
The risk of cytomegalovirus transmission
by penetrating keratoplasty. Am J
Mietz H, Aisenbrey S, Ulrich Bartz-Schmidt
K, et al: Ganciclovir for the treatment of
anterior uveitis, Graefes Arch Clin Exp
Ophthalmol 2000; 238:905909.
Baum S: Adenovirus. In: Mandell G,
Bennett J, Dolin R, eds. Philadelphia, PA:
Elsevier, Churchill Livingstone;
2005:18351840.
Duke-Elder S: The adenoviruses. In:
Duke-Elder S, ed. St Louis, MO: CV
Mosby; 1965:348.
Gordon Y: Adenovirus and other
nonherpetic viral diseases. In: Smolin G,
Thoft R, eds. Boston, MA: Little, Brown;
1994:215228.
Kinchington PR, Romanowski EG, Jerold
Gordon Y: Prospects for adenovirus
antivirals. J Antimicrob Chemother 2005;
55:424429.
Guyer B, ODay D, Hierholzer J, et al:
Epidemic keratoconjuncitvitis: types 8 and
19. J Infect Dis 1975; 132:142.
Tullo A: The Adenoviruses. In: Easty Dl, ed.
Chicago, IL: Year Book Medical;
1985:257270.
Azar M, Dhaliwal K, Bower K, et al: Possible
consequences of shaking hands with your
patients with epidemic keratoconjunctivitis.
Am J Ophthalmol 1996; 121:711712.
Gordon Y, RY, G, E, R, et al: Prolonged
recovery of desiccated adenoviral
serotypes 5, 8, and 19 from plastic and
metal surfaces in vitro. Ophthalmology
1993; 100:18351839.
Sprague JB, et al: Epidemic
keratoconjunctivitis: A severe industrial
outbreak due to adenovirus type 8. N Engl
J Med 1973; 289:13411346.
Laibson P, Ortolan G, Dupre-Strachan S:
Community and hospital outbreak of
epidemic keratoconjunctivitis. Arch
ODay D, Guyer B, Hierholzer J, et al:
Clinical and laboratory evaluation of
epidemic keratoconjunctivitis due to
adenoviruses type 8 and 19. Am J
Kowalski RP, Sundar-Raj CV, Romanowski
EG, et al: The disinfection of contact lenses
contaminated with adenovirus. Am J
Ophthalmol 2001; 132:777779.
Harvey SA, Romanowski EG, Yates KA,
et al: Adenovirus-directed ocular innate
immunity: the role of conjunctival defensinlike chemokines (IP-10, I-TAC) and
phagocytic human defensin-alpha. Invest
Chodosh J, Miller D, Stroop W, et al:
Adenoviral epithelial keratitis. Cornea 1995;
14(2):167174.
Uchio E, Takeuchi S, Itoh N, et al: Clinical
and epidemiological features of acute
661.
662.
663.
664.
665.
666.
667.
668.
669.
670.
671.
672.
673.
674.
675.
676.
677.
follicular conjunctivitis with special

reference to that caused by herpes simplex
virus type 1. Br J Ophthalmol 2000;
84:968972.
Waring GI, Laibson P, Satz J, et al: Use of
vidarabine in epidemic keratoconjunctivitis
due to adenovirus types 3, 7, 8 and 19. Am
J Ophthalmol 1976; 82.
Dawson C, Darrell D: Infections due to
adenovirus type 8 in the United States. I.
An outbreak of epidemic keratoconjunctivitis originating in physicians ofce.
N Engl J Med 1963; 268:1031.
Dawson C, Hanna L, Togni B: Adenovirus
type 8 infections in the United States. IV.
Observations on the pathogenesis of
lesions in severe eye disease. Arch
Dawson C, Hanna M, Wood T, et al:
Adenovirus type 8 keratoconjunctivitis in
the United States. III. Epidemiologic,
Clinical and Microbiologic features. Am J
Laibson P, Dhiri S, Oconer J, et al: Corneal
inltrates in epidemic keratoconjunctivitis:
response to double blind corticosteroid
therapy. Arch Ophthalmol 1970; 84:36.
Murrah W: Epidemic keratoconjunctivitis.
Harnett G, Newnham W: Isolation of
adenovirus type 19 from the male and
female genital tracts. Br J Vener Dis 1981;
57:55.
Laibson P: Adenoviral keratoconjunctivitis.
in Ocular Viral Disease, Int Ophthalmol Clin,
Ed. Pavan-Langston, D, Boston, Little
Brown and Co. 1975; 15:187201.
Boniuk M, Philips C, Friedman J, et al:
Chronic adenovirus type 2 keratitis in man.
N Engl J Med 1965; 273:924.
Darougar S, Quinlan J, Gibson J, et al:
Epidemic keratoconjunctivitis and chronic
papillary conjunctivitis in London due to
adenovirus type 19. Br J Ophthalmol 1977;
61:76.
Pettit T, Holland G: Chronic keratoconjunctivitis associated with ocular adenovirus
infection. Am J Ophthalmol 1979; 88:748.
Hecht S, Hanna L, Sery T, et al: Treatment
of epidemic keratoconjunctivitis with
idoxuridine (IUDR). Arch Ophthalmol 1965;
73:49.
Pavan-Langston D, Dohlman C: A double
blind clinical study of adenine arabinoside
therapy of viral keratoconjunctivitis. Am J
Gordon Y, Romanowski E, Araullo-Cruz T:
HPMPC, a broad-spectrum topical antiviral
agent, inhibits herpes simplex virus type 1
replication and promotes healing of
dendritic keratitis in the rabbit ocular
model. Cornea 1994; 13(6):516520.
Romanowski E, Gordon Y, Araullo-Cruz T,
et al: The antiviral resistance and
replication of cidofovir-resistant adenovirus
variants in the New Zealand White rabbit
ocular model. Invest Ophthalmol Vis Sci
2001; 42:18121815.
Romanowski EG, Yates KA, Gordon YJ:
Antiviral prophylaxis with twice daily topical
cidofovir protects against challenge in the
adenovirus type 5/New Zealand rabbit ocular
model. Antiviral Res 2001; 52:275280.
Romanowski EG, Gordon YJ: Efcacy of
topical cidofovir on multiple adenoviral
serotypes in the New Zealand rabbit ocular
model. Invest Ophthalmol Vis Sci 2000;
41:460463.
678. Romanowski EG, Roba LA, Wiley L, et al:

The effects of corticosteroids of adenoviral
replication. Arch Ophthalmol 1996;
114:581585.
679. Romanowski EG, Yates KA, Gordon YJ:
Topical corticosteroids of limited potency
promote adenovirus replication in the
Ad5/NZW rabbit ocular model. Cornea
2002; 21:289291.
680. Chou R, Huffman LH, Fu R, et al: Screening
for HIV: a review of the evidence for the
US. Preventive Services Task Force. Ann
Intern Med 2005; 143:5573.
681. CDC, for Disease Control and Prevention:
The global HIV and AIDS epidemic. MMWR
2001; 50:434439.
682. CDC, ACoIP: Surveillance of health care
workers with HIV/AIDS. Available:
http://www.cdc.gov/hiv/pubs/facts/hchsurv.
htm Accessed 13 Sept 2002.
683. Walensky R, Losina E, Steger-Craven K,
et al: Identifying undiagnosed human
immunodeciency virus. Arch Int Med
2002; 162:887892.
684. Karon JM, Fleming PL, Steketee RW, et al:
HIV in the United States at the turn of the
century: an epidemic in transition. Am J
Public Health 2001; 91:10601068.
685. Dunn J, Holland G: Human
immunodeciency virus infection and AIDS.
In: Tabbara K, Hyndiuk R, eds. Boston, MA:
686. Maldarelli F: Diagnosis of human
immunodeciency virus infection. In:
Mandell G, Bennett J, Dolin R, eds.
Philadelphia, PA: Elsevier, Churchill
687. Sterling T, Chaisson R: General Clinical
Manifestations of Human
Immunodeciency Virus Infection (Including
Acute Retroviral Syndrome, and Oral,
Cutaneous, Renal, Ocular, and Cardiac
Diseases). In: Mandell G, Bennett J, Dolin
R, eds. Philadelphia, PA: Elsevier, Churchill
688. Guex-Crosier Y: [Diagnosis and treatment of
ocular viral infections in AIDS patients]. Rev
Med Suisse Romande 1998; 118:941947.
689. Meyer D: Eye signs that alert the clinician
to a diagnosis of AIDS. Sadj 2005;
60:386387.
690. Chan CC, Zhang M: [Ocular manifestations
in HIV/AIDS]. Zhonghua Yan Ke Za Zhi
2005; 41:563571.
691. Kahraman G, Krepler K, Franz C, et al:
Seven years of HAART impact on
ophthalmic management of HIV-infected
patients. Ocul Immunol Inflamm 2005;
13:213218.
692. Grossniklaus H, Frank K, Tomsak R:
Cytomegalovirus retinitis and optic neuritis
in acquired immune deciency syndrome.
693. Mesaric B, Lisic M, Kniewald T, et al:
[Ocular manifestations in patients with
human immunodeciency virus infection
before and after the introduction of highly
active antiretroviral therapy]. Lijec Vjesn
2005; 127:123128.
694. Melbye M, Grossman RJ, Goedert JJ, et al:
Risk of AIDS after herpes zoster. Lancet
1987; i:728730.
695. Pepose J, Holland G, Nestor M, et al:
Acquired immune deciency syndrome:
pathogenic mechanisms of ocular disease.
696. Pflugfelder S, Saulson R, Ullman S:
Peripheral corneal ulceration in a patient
697.
698.
699.
700.
701.
702.
703.
704.
705.
706.
707.
708.
709.
710.
711.
712.
713.
714.
with AIDS-related complex. Am J

Dugel P, Gill P, Frangieh G, et al: Ocular
adnexal Kaposis sarcoma in acquired
immunodeciency syndrome. Am J
Dugel P, Gill P, Frangieh G, et al: Particles
resembling retrovirus in conjunctival Kaposis
sarcoma. Am J Ophthalmol 1990; 110:86.
Holland GN, Gottlieb MS, Yee RD, et al:
Ocular disorders associated with a new
severe acquired cellular immunodeciency
syndrome. Am J Ophthalmol 1982;
93:393402.
Pepose JS, Holland GN, Nestor MS, et al:
Acquired immune deciency syndrome.
Pathogenic mechanisms of ocular disease.
Zuccati G, Tiradritti L, Mastrolorenzo A,
et al: AIDS-related Kaposis sarcoma of the
eye. Int J STD AIDS 1991; 2:136137.
Sissolak G, Mayaud P: AIDS-related
Kaposis sarcoma: epidemiological,
diagnostic, treatment and control aspects
in sub-Saharan Africa. Trop Med Int Health
2005; 10:981992.
Pak F, Pyakural P, Kokhaei P, et al: HHV-8/
KSHV during the development of Kaposis
sarcoma: evaluation by polymerase chain
reaction and immunohistochemistry. J
Cutan Pathol 2005; 32:2127.
Nakamura S, Salahuddin S, Biberfeld P,
et al: Kaposis sarcoma cells. Long term
culture with growth factor from retrovirusinfected C.D. #4 +T. Science 1989;
242:426.
Cantrill H, Henry K, Jackson B, et al:
Recovery of human immunodeciency virus
from ocular tissues in patients with
acquired immune deciency syndrome.
Conway M, Insler M: Identication and
incidence of human immunodeciency
virus antibodies and hepatitis B virus
antigens in corneal donors. Ophthalmology
1988; 95:1463.
Doro S, Navia B, Kahn A, et al:
Conrmation of HTLV-III virus in cornea.
Fujikawa L, Salahuddin S, Dharam A, et al:
HTLV-III in the tears of AIDS patients.
ODay D: The risk posed by HTLV-IIIinfected corneal donor tissue. Am J
Salahuddin S, Palestine A, Hecke R, et al:
Isolation of the human T-cell leukemia/
lymphotrophic virus type III from the
Goode S, Hertzmark E, Steinert R, et al:
Adequacy of the ELISA test for screening
corneal transplant donors. Am J
Kilby J, Eron J: Novel therapies based on
mechanisms of HIV-1 cell entry. N Engl J
Med 2003; 348:22282238.
Damon I: Orthopoxviruses Vaccinia
(Smallpox Vaccine), Variola (Smallpox),
Monkeypoxs, Cowpox. In: Mandell G,
2005:17421752.
Damon I: Other Poxviruses that infect
humans: parapoxvirus, molluscum
contagiosum, and tanapox. In: Mandell G,
2005:17531756.
715. Pepose J, Esposito J: Molluscum

contagiosum, orf, and vaccinia. In:
Pepose J, Holland G, Wilhelmus K, eds.
St Louis, MO: Mosby; 1996:846856.
716. Hanson D, Diven D: Molluscum
contagiosum. Review. Dermatol Online J
2003; 9: 2.
717. Ingraham H, Schoenleber D: Epibulbar
molluscum contagiosum. Am J Ophthalmol
1998; 125:394396.
718. North R: Presumptive viral
keratoconjunctivitis, mononucleosis and
the oncogenic viruses. In: Pavan-Langston
D, ed. Boston, MA: Little, Brown; 1975:211.
719. Gonnering R, Kronish J: Treatment of
periorbital molluscum contagiousum by
incion and curettage. Ophthalmologic Surg
1988; 19:325329.
720. Charteris D, Bonshek R, Tullo A:
Ophthalmic molluscum contagiosum:
clinical and immunopathological features.
721. Hancox J, J J, S M: Treatment of
molluscum contagiosum with the pulsed
dye laser over a 28-month period. Cutis
2003; 71:414416.
722. De Clercq E: Cidofovir in the therapy and
short-term prophylaxis of poxvirus
infections. Trends Pharmacol Sci 2002;
23:456458.
723. Neyts J, ED, C: Therapy and short-term
prophylaxis of poxvirus infections: historical
background and perspectives. Antiviral Res
2003; 57:2533.
724. Cono J, Casey CG, Bell DM: Smallpox
vaccination and adverse reactions.
Guidance for clinicians. MMWR Recomm
Rep 2003; 52:128.
725. Henderson D, Inglesley T, Bartlett J,
et al: Smallpox as a biological weapon:
medical and public health management
(consensus statement). JAMA 1999;
281:21272137.
726. Maki DG: National preparedness for
biological warfare and bioterrorism:
smallpox and the ophthalmologist. Arch
Ophthalmol 2003; 121:710711.
727. Brauner S, Pavan-Langston D: Small Pox,
Vaccinia, and the Eye. In: Miller J, ed.
Philadelphia, PA: Lippincott, Williams &
Wilkins; 2006.
728. CDC, CfDCaP: Smallpox fact sheet,
smallpox overview, 2002. Available:
http://www.cdc.gov/smallpox 2002.
729. Breman J, Henderson D: Diagnosis and
management of smallpox. N Engl J Med
2002; 346:13001308.
730. Lane JM, Ruben FL, Neff JM, et al:
Complications of smallpox vaccination,
1968. N Engl J Med 1969; 281:12011208.
731. Mack T: A different view of smallpox and
vaccination. N Engl J Med 2003;
332(18),348(5):460463.
732. Neff J, Lane J, Fulginiti V, et al: Contact
vacciniaTransmission of vaccinia from
smallpox vaccination. JAMA 2002;
288:19011905.
733. Bozzette S, Boer R, Bhatnaga V, et al:
A model for a smallpox-vaccination policy.
N Engl J Med 2003; 348:416425.
734. Lane JM, Millar JD: Routine childhood
vaccination against smallpox reconsidered.
N Engl J Med 1969; 281:12201224.
735. Ruben F, Lane J: Ocular vaccinia:
epidemiologic analysis of 348 cases. Arch
736. Sepkowitz K: How contagious is vaccinia?
N Engl J Med 2003; 348:439446.
737. Fillmore GL, Ward TP, Bower KS, et al:

Ocular complications in the Department of
Defense Smallpox Vaccination Program.
Ophthalmology 2004; 111:20862093.
738. Ellis P, Winograd L: Current concepts of
ocular vaccinia. Trans Proc Coast
Otoophthalmol Soc 1963; 44:141148.
739. Jones B, Al-Hussaini M: Therapeutic
considerations in ocular vaccinia. Trans
Ohthalmol Soc UK 1964; 83:613631.
740. Pepose JS, Margolis TP, Pavan-Langston
D: Ocular complications of smallpox
vaccination. Am J Ophthalmol 2003;
136:343352.
741. Espy M, Uhl J, Sloan L, et al: Detection of
vaccinia virus, herpes simplex virus,
varicella-zoster virus, and Bacillus anthracis
DNA by LightCycler polymerase chain
reaction after autoclaving: implications for
biosafety of bioterrorism agents. Mayo Clin
Proc 2002; 77:624628.
742. Jones B, Galbraith J, Al-Hussaini M:
Vaccinial keratitis treated with interferon.
Lancet 1962; 1:875.
743. Perera C: Vaccinial disciform keratitis. Arch
Ophthalmol 1940; 24:352356.
744. Rennie A, Cant J, Foulds M, et al: Ocular
vaccinia. Lancet 1974; 273275.
745. CDC, ACoIP: Summary of October 2002
ACIP Smallpox Vaccination
Recommendations. Atlanta: Centers for
Disease Control and Prevention. Available:
http://www.bt.cdc.gov/agent/smallpox/vacc
ination/acip-recs-oct2002.asp 2002.
746. Kempe C: Studies on smallpox and
complications of smallpox vaccination.
Pediatrics 1960; 7689.
747. Elion G, Rideout J, de Miranda P, et al:
Biological activities of some purine
arabinosides. Ann NY Acad Sci 1975;
255:468480.
748. Elion GB, Furman PA, Fyfe JA, et al: The
selectivity of action of an antiherpetic
agent, 9-(2-hydroxyethoxymethyl) guanine.
Reproduced from Proc Natl Acad Sci USA
1977; 74:57165720; Rev Med Virol 1999;
9:147152;discussion 152143.
vidarabine. Arch Ophthalmol 1976;
94:1363.
trifluorthymidine. Arch Ophthalmol 1976;
94:1976.
751. Jack M, Sorenson R: Vaccinial keratitis
treated with IDU. Arch Ophthalmol 1963;
69:730732.
752. Snoeck R, Holy A, Deworlf-Peeters C, et al:
Antivaccinia activities of acyclic nucleoside
phosphonate derivatives in epithelial cells
and organotypic cultures. Antimicrob
Agents Chemo 2002; 46:33563361.
753. Chirambo MC, Benezra D: Causes of
blindness among students in blind school
institutions in a developing country. Br J
Ophthalmol 1976; 60:665668.
754. Koplan J, Hicks J: Smallpox and vaccinia
in the United States 1972. J Infect Dis
1974; 129:224.
755. De Clercq E: Cidofovir in the treatment of
poxvirus infections. Antiviral Res 2002;
55:113.
756. Snoeck R, Holy A, Dewolf-Peeters C, et al:
Antivaccinia activities of acyclic nucleoside
phosphonate derivatives in epithelial cells
and organotypic cultures. Antimicrob
CHAPTER 49
703
SECTION 6
757. Sharma R: Clinical assessment of an

isothiazole thiosemicarbazone against
smallpox. J Indian Med Assoc 1968;
51:610615.
758. Bonnez W, Reichman R: Papillomaviruses.
In: Mandell G, Bennett J, Dolin R, eds.
Philadelphia, PA: Elsevier, Churchill
759. Scott I, Karp C, Nuovo G: Human
papillomavirus 16 and 18 expression in
conjunctival intraepithelial neoplasia.
760. Karcioglu Z, Issa T: Human papilloma virus
in neoplastic and non-neoplastic conditions
of the external eye. Br J Ophthalmol 1997;
81:595598.
761. Prasad C: Pathobiology of human papilloma
viruses. Clin Lab Med 1995; 15(3):685704.
762. Andrei G, Snoeck R, Schols D, et al:
Induction of apoptosis by cidofovir in
human papillomavirus (HPV)-positive cells.
Oncol Res 2000; 12:397408.
763. Simonart T, Boelaert JR, Mosselmans R,
et al: Antiproliferative and apoptotic effects
of iron chelators on human cervical
carcinoma cells. Gynecol Oncol 2002;
85:95102.
764. Raley JC, Followwill KA, Zimet GD, et al:
Gynecologists attitudes regarding human
papilloma virus vaccination: a survey of
Fellows of the American College of
Obstetricians and Gynecologists. Infect Dis
Obstet Gynecol 2004; 12:127133.
765. Sowjanya AP, Jain M, Poli UR, et al:
Prevalence and distribution of high-risk
human papilloma virus (HPV) types in
704
766.
767.
768.
769.
770.
771.
772.
773.
774.
invasive squamous cell carcinoma of the

cervix and in normal women in Andhra
Pradesh, India. BMC Infect Dis 2005;
5:116.
Greyson M: Acute and chronic follicular
conjunctivitis. In: Greyson M, ed. St Louis,
MO: Mosby; 1983:114.
Haltier M, Tarkkanen A, Vaheri A, et al:
Measles retinopathy: measles retinopathy
during immunosuppression. Br J
Meyer R, Sullivan J, Oh J: Mumps
78:1022.
Mickatavatge R, Amadur J: A case report
of mumps keratitis. Arch Ophthalmol 1963;
69:758.
Riffenburgh R: Ocular manifestations
of mumps. Arch Ophthalmol 1961;
66:739.
Kayikcioglu O, Kir E, Soyler M, et al: Ocular
ndings in a measles epidemic among
young adults. Ocul Immunol Inflamm 2000;
8:5962.
Toker MI, Erdem H, Erdogan H, et al: The
effects of topical ketorolac and
indomethacin on measles conjunctivitis:
randomized controlled trial. Am J
Ophthalmol 2006; 141:902905.
Baum J: Hemorrhagic conjunctivitis: a new
type of epidemic viral keratoconjunctivitis.
Mitsui Y, Kajima M, Matsumura K, et al:
Hemorrhagic conjunctivitis: a new type of
epidemic viral keratoconjunctivitis. Jpn J
775. Rosa R, Alfonso E: Enterovirus

keratoconjunctivitis. In: Pepose J, Holland
G, Wilhelmus K, eds. St Louis, MO: Mosby;
1996:895904.
776. Uchio E, Yamazaki K, Aoki K, et al:
Detection of enterovirus 70 by polymerase
chain reaction in acute hemorrhagic
122:273275.
777. Whitcher J, Schmidt N, Malbrouk R, et al:
Acute hemorrhagic conjunctivitis in Tunisia.
778. Langford MP, Orillac R, Chen D, et al:
Systemic and ocular antibody responses to
inactivated acute hemorrhagic
conjunctivitis (AHC) virus; enterovirus 70
(EV70). Ocul Immunol Inflamm 2003;
11:197209.
779. Boger W III, Peterson R, Robb R:
Keratoconus and acute hydrops with
congenital rubella syndrome. Am J
780. Boniuk V: Rubella and congenital rubella
syndrome. in Ocular Viral Disease, Internat.
Ophthalmol. Clin.,Ed. Pavan-Langston, D,
Boston, Little, Brown, 1975:229242.
781. Boniuk V: Rubeola (measle, morbilli). in
Ocular Viral Disease, Internat. Ophthalmol.
Clin. Ed. Pavan-Langston D, ed. Boston,
MA: Little, Brown; 1975:243256.
782. Hara J, Fujimoto F, Ishibashi T, et al:
Ocular manifestations of the 1976 rubella
epidemic in Japan. Am J Ophthalmol 1979;
87:642.
783. Smolin G: Report of a case of rubella
keratitis. Am J Ophthalmol 1972; 74:36.
CHAPTER
50
Bacterial, Chlamydial, and Mycobacterial

Infections
Francis Mah
Bacterial colonization of the eyelid and conjunctiva is normal

and helps reduce opportunities for pathogenic strains from
gaining a foothold. Host defense mechanisms can be overcome,
however, and lead to serious ocular morbidity if not treated
properly. Although the clinical manifestations of conjunctival
and corneal infections may be characteristic of certain pathogens, further laboratory evaluation with cultures and antibiotic
susceptibility testing provide a denitive diagnosis and more
focused treatment after empirical therapy has been initiated.
PATHOGENESIS
OCULAR DEFENSE MECHANISMS
Several mechanisms work in concert to protect the eye from
infection.1 The bony orbital rim and eyelids protect the eye
from injury and foreign materials. Normal eyelid function and
the flushing action of the tears remove debris and microbes
from the ocular surface and prevent desiccation of the corneal
epithelium. The tear lm contains substances that have antimicrobial properties. Lactoferrin binds iron used in microbial
metabolism, augments antibody function, and modulates complement activity. Beta-lysin induces rupture of bacterial cell
membranes, while lysozyme degrades cell walls of Gram-positive
bacteria. Found in higher concentrations in the tears than in
serum, secretory IgA prevents bacterial adherence to epithelium.
The squamous epithelium of the conjunctiva and cornea
provides a mechanical barrier against microbial invasion.
Corneal epithelial cells as well as keratocytes secrete cytokines
that activate immune defenses.2 Langerhans cells are antigenpresenting cells found in the peripheral corneal epithelium
which activate T-cells when foreign antigens are processed.
Conjunctiva-associated lymphoid tissue are sites of antigen
processing in the conjunctiva resulting in activation of T-cells
as well as production of IgA by plasma cells. The bulbar and
palpebral conjunctivae contain lymphatics which drain into the
preauricular, parotid, and submandibular lymph nodes.
AGENTS
The most common bacterial organisms found in the normal
flora of the eyelids and conjunctiva are Staphylococcus epidermidis, Staphylococcus aureus, and Corynebacterium species.3
Less commonly identied are Propionibacterium acnes, Moraxella
catarrhalis, Streptococcus species, Haemophilus influenzae,
Proteus species, and Micrococcus species.3,4 The spectrum of
organisms in the normal flora varies with age and geography. At
birth through the vaginal canal, Staphylococcus epidermidis,
Staphylococcus aureus, streptococci, and Escherichia coli are
transferred from mother to child. With increasing age, more
Gram-negative organisms are isolated but S. epidermidis,

S. aureus, and Corynebacterium species remain predominant.
Once host defense mechanisms are overwhelmed, these
commensal organisms could produce disease but pathogenic
organisms are typically involved in conjunctivitis and keratitis.
MECHANISM OF DISEASE
An intact surface epithelium provides an effective barrier to
microbial penetration, and a defect in the epithelial layer is
usually required. Some organisms such as Neisserria gonorrhea,
Neisseria meningitides, Corynebacterium diptheriae, Listeria,
and Shigella can directly penetrate an intact epithelium.
Infection of the ocular surface begins with adherence to surface
epithelium. Bacteria express adhesins which are proteins that
bind to receptors on the host cell surface.5 Pseudomonas
aeruginosa utilize virulence factors such as pili and glycocalyx
for adhesion and flagella for motility.6 Biolm production by
bacteria may prevent phagocytosis and enhance adhesion to
contact lens material.7 Microbial invasion into corneal stroma
is aided by a variety of bacterial enzymes and toxins. Pseudomonas aeruginosa produces elastase, alkaline protease, LasA
protease, and protease IV which degrade not only collagen, but
also immunoglobulins, complement, and cytokines.8,9 Staphylococcus aureus can produce a variety of toxins including alphatoxin which can induce severe corneal inflammation and
epithelial erosions.10 Once microbial invasion occurs, the hostderived inflammatory response contributes to further corneal
destruction.11 Release of lysozomal enzymes and oxygen free
radicals from PMN can damage stromal collagen, glycosaminoglycans and disrupt stromal keratocytes contributing to corneal
necrosis and edema.12
DIAGNOSTIC TECHNIQUES
The diagnosis of bacterial infection is conrmed by culture
identication of the pathogenic organism. In cases of routine
suspected bacterial conjunctivitis, microbiologic studies may
have a limited role as these cases are self-limited and respond
well to empiric broad-spectrum therapy. Bacteriologic studies
should be mandatory in certain clinical settings such as
neonatal conjunctivitis, hyperacute conjunctivitis, chronic
conjunctivitis, and persistent refractory acute conjunctivitis.
For the best bacterial yield, cultures should be obtained prior
to initiating antibiotic therapy. Conjunctival cultures should be
performed without topical anesthetic by swabbing the
conjunctival surface with a sterile calcium alginate or Dacron
swab that has been moistened with trypticase soy broth. The
specimen should then be inoculated on to blood agar, chocolate
agar, and mannitol salt agar. ThayerMartin medium should
705

also be inoculated if gonococcal infection is suspected. If
chlamydial infection is suspected, a swab should be placed in
Chlamydia transport media for PCR processing. Conjunctival
smears are obtained after cultures have been taken. A platinum
spatula is used to rmly scrape the anesthetized palpebral
conjunctiva. The scrapings can then be evaluated by Grams,
Giemsas, or other stains as clinically indicated.
Routine culture of corneal infections is not the usual practice
in the community.13,14 A small peripheral ulcer may be treated
empirically, but a large, purulent, central ulcer that extends to
the middle to deep stroma should be cultured. In addition,
ulcers that are clinically suspicious for fungal, mycobacterial, or
amoebic infections or are unresponsive to initial broad spectrum antibiotics warrant cultures. Topical anesthesia with
proparacaine hydrochloride is preferred since it has less antibacterial properties compared with other topical anesthetics.15
A sterile platinum spatula is used to scrape the leading edge as
well as the base of the ulcer while carefully avoiding contamination from the lids and lashes. Organisms such as Streptococcus pneumoniae are more readily recovered from the ulcer
edge while other organisms such as Moraxella are recovered
from the base.16 The scrapings are inoculated into solid media
(blood, chocolate, mannitol, Sabourauds agar) by streaking a
row of Cs onto its surface. New material is recovered for each
row. Scrapings are also placed on microscope slides and stained
as above. Special stains include ZiehlNeelsen acid-fast stain
for Mycobacterium, Actinomyces, and Nocardia. Acridine
orange is a fluorescent dye that may be helpful in identifying
bacteria when yields are low, but this stain does not yield
classication information that Grams stain provides.
In cases of deep stromal suppuration that is not readily
accessible or a progressive microbial keratitis unresponsive to
therapy, a corneal biopsy may be warranted.17 A round 23 mm
sterile disposable skin punch is used to incise the anterior
corneal stroma and lamellar dissection is performed with a
surgical blade. The specimen is then ground in a mortar with
trypticase soy broth and plated on media.
CLINICAL PRESENTATION
CONJUNCTIVAL INFECTIONS
SECTION 6
Mucopurulent (Acute) Bacterial Conjunctivitis
706
Bacterial conjunctivitis can be divided into subtypes based on

the degree of purulence (mucopurulent, purulent) as well as the
onset and duration of symptoms (hyperacute, acute, chronic).
The most common form of bacterial conjunctivitis is the acute
mucopurulent form. Onset is rapid with conjunctival
hyperemia and mucopurulent discharge. The fellow eye may
become affected in 23 days from cross-inoculation. Other signs
include subconjunctival hemorrhage, velvety palpebral papillary
reaction, chemosis, and punctate epithelial keratopathy. Symptoms consist of foreign body sensation, tearing, and matting of
eyelids. Vision is usually normal or minimally affected. The
most common causative pathogens are Staphylococcus aureus,
Streptococcus pneumoniae, and Haemophilus influenzae.18
Streptococcus viridans and Streptococcus pyogenes can also
produce an acute conjunctivitis, and the latter can induce a
membranous reaction usually on the bulbar conjunctiva.
S. aureus conjunctivitis can become chronic due to its afnity
for the eyelid margin and resultant blepharitis. Gram-negative
rods other than Haemophilus species rarely cause acute
conjunctivitis in an immunocompetent patient.
In children, Streptococcus pneumoniae and Haemophilus
influenzae are most commonly isolated.19 Nonencapsulated
H. influenzae conjunctivitis can be associated with otitis media
(conjunctivitis-otitis syndrome).20 H. influenzae conjunctivitis
tends to last longer and can be characterized by petechial conjunctival hemorrhages and perilimbal corneal inltrates.
Encapsulated H. influenzae type b rarely causes conjunctivitis;
however, infection with this species has been associated with
disseminated disease including periorbital cellulitis, sepsis, and
meningitis requiring systemic antibiotic therapy. In Brazil,
H. influenzae biotype III (previously called H. aegyptius) has
been identied as the cause of a fulminant purpuric sepsis
following a purulent conjunctivitis in children.21
Although mild cases of acute conjunctivitis resolve on their
own, topical antibiotic therapy speeds resolution and reduces
morbidity.22,23 Several factors should determine antibiotic
selection including severity of the infection, spectrum of antibiotic susceptibility, patient compliance, and cost. In routine
cases of acute bacterial conjunctivitis, empirical therapy using
broad-spectrum antibiotics for 710 days results in rapid resolution. Sulfacetamide 10% or 15% is inexpensive and effective
against the common causative organisms. Those with sulfa
allergy respond well with topical trimethoprimpolymyxin B or
neomycinpolymyxin Bgramicidin. Chloramphenicol is an
effective broad-spectrum agent but is not widely used due to
concerns of aplastic anemia following topical use.24 Empirical
use of fluoroquinolones should be reserved for more serious
ocular infections such as bacterial keratitis so that selection for
drug-resistant strains is avoided. Lack of improvement from
initial therapy warrants further laboratory studies and tailoring
treatment according to the culture diagnosis.
Purulent (Hyperacute) Bacterial Conjunctivitis

Purulent conjunctivitis has a more rapid and severe onset than
acute conjunctivitis and is characterized by preauricular adenopathy, lid edema, conjunctival hyperemia, chemosis, and
copious amounts of purulent discharge. A membrane or
pseudomembrane may be present contributing to ocular discomfort. This condition is rapidly progressive leading to corneal
inltration, abscess formation, and perforation if left untreated.
Neisseria gonorrhoeae, which can penetrate an intact corneal
epithelium, is the most common causative pathogen. Gonococcal conjunctivitis is a sexually transmitted disease and
symptoms of cervicitis or urethritis may be elicited from the
history. Neisseria meningitides is a less frequent cause of purulent conjunctivitis and is classied into primary (exogenous)
and secondary (endogenous) disease. Timely diagnosis and
treatment of meningococcal conjunctivitis is indicated since
meningococcemia and meningitis may result.25
Smears and cultures should be obtained in all cases of purulent conjunctivitis. Grams stain of conjunctival scrapings reveals
Gram-negative diplococci within polymorphonuclear leukocytes
as well as adherent to epithelial cells. Material for culture
should be directly inoculated from the eye onto chocolate agar
and ThayerMartin media and incubated in 5% CO2 at 37C.
Neisseria conjunctivitis requires systemic therapy. In adults,
gonococcal conjunctivitis is treated with a single dose of
ceftriaxone 1 g IM.26 Patients with penicillin allergy can be
given ciprofloxacin 500 mg orally, ofloxacin 400 mg orally, or
spectinomycin 2 g IM.27 Gonococcal conjunctivitis with corneal
ulceration warrants hospital admission for treatment with
intravenous ceftriaxone 1 g every 12 h for 3 days. Adjunctive
therapy includes frequent irrigation of purulent material from
the conjunctival surface and fornices to remove inflammatory
cells, proteolytic enzymes and debris that may contribute to
corneal thinning. Treatment for Chlamydia trachomatis should
also be instituted since co-infection is not uncommon.28 A
single dose of azithromycin 1 g orally or doxycycline 100 mg
orally twice a day for 7 days is recommended. Treatment
of Neisseria meningitides conjunctivitis with penicillin
G 300 000 IU kg1 day1 or ceftriaxone 100 mg kg1 day1 has
Bacterial, Chlamydial, and Mycobacterial Infections
Neonatal Conjunctivitis (Ophthalmia Neonatorum)

Neonatal conjunctivitis is any conjunctivitis occurring within
the rst month of life. This condition is an ocular emergency
requiring prompt diagnosis and treatment. Neonatal conjunctivitis can be caused by bacterial, viral and chemical agents. The
most common infectious cause is Chlamydia trachomatis
which has been isolated from 10% to 55% of all cases of
neonatal conjunctivitis.32 Clinical signs can appear unilaterally
or bilaterally 514 days after delivery. The presentation varies
from mild hyperemia with scant mucoid discharge to severe
hyperemia, purulent discharge, chemosis, and pseudomembranes. Unlike adult chlamydial conjunctivitis, a follicular
response is absent due to the immature lymphoid system in the
neonate. Systemic infections such as pneumonitis and otitis
media may occur with chlamydial conjunctivitis, and corneal
scarring and neovascularization may develop in severe cases.
Neisseria gonorrhoeae causes 1% of cases of neonatal conjunctivitis in developed countries.32 The typical presentation of
gonococcal conjunctivitis is a sudden onset of severe bilateral
eyelid edema, hyperemia, chemosis, and profuse purulent discharge that may appear 17 days after delivery. Since Neisseria
gonorrhoeae can penetrate an intact epithelium, corneal
ulceration, perforation, and endophthalmitis can be seen in
untreated cases. Disseminated infection includes arthritis,
meningitis, and sepsis. Other bacterial organisms causing neonatal conjunctivitis include Gram-positive organisms such as
Staphylococcus aureus, Streptococcus pneumoniae, and
Streptococcus viridans. Gram-negative organisms include
Haemophilus influenzae, Escherichia coli, Enterobacter species,
Klebsiella species, and Proteus species. Presentation of nongonococcal conjunctivitis is typically later than that of chlamydial or gonococcal disease. Pseudomonas aeruginosa is a rare
cause of nongonococcal conjunctivitis and may result in corneal
ulceration and perforation. Chemical conjunctivitis often
results after application of topical drops. Silver nitrate 1%
introduced by Cred can cause conjunctivitis in 90% of
neonates.33 This conjunctivitis begins a few hours after delivery
with mild conjunctival injection and watery discharge and
typically resolves in 12 days.
Laboratory evaluation with smears and cultures is mandatory
in neonatal conjunctivitis. A preliminary diagnosis can be
achieved using Grams and Giemsa staining. Intracellular
Gram-negative diplococci within polymorphonuclear leukocytes
are revealed in gonococcal conjunctivitis. In addition to neutrophils, lymphocytes and plasma cells, neonatal chlamydial
conjunctivitis smears, unlike the adult form, show basophilic
intracytoplasmic inclusions. A chemical conjunctivitis shows
neutrophils with occasional lymphocytes. Cell culture identication is the gold standard; however, rapid detection tests for
Chlamydia trachomatis such as immunofluorescent antibody
staining, enzyme immunoassay and polymerase chain reaction
assay are also available.3436
Although topical therapy for chlamydial neonatal conjunctivitis may be adequate, the potential for infection at other sites
warrants systemic therapy. Oral erythromycin base or ethylsuccinate 50 mg kg1 day1 divided into four equal doses for
1014 days is recommended by the CDC.27 A repeat course at
the same dosage can be administered if needed. Due to
increasing resistance to penicillin, tetracycline, and fluoroquinolones, gonococcal conjunctivitis is treated with ceftriaxone
2550 mg/kg up to 125 mg as a single dose. Hospitalization and
hourly irrigation of the conjunctival fornices until the purulence
subsides are recommended. Ocular prophylaxis immediately
after delivery is directed against gonococcal conjunctivitis. The

recommended agents include silver nitrate 1% aqueous solution, erythromycin 0.5% ointment, or tetracycline 1% ointment
administered from single-use ampules or tubes. This intervention, however, is not effective in preventing chlamydial conjunctivitis.37 Povidone-iodine has been shown to be effective against
both Neisseria gonorrhoeae and Chlamydia trachomatis.38 Its
decreased expense and increased clinical spectrum compared to
the other agents make povidone-iodine a viable alternative,
especially in developing countries.39
Chronic Bacterial Conjunctivitis

Conjunctivitis that persists for 4 or more weeks is considered
chronic. This condition is typically bilateral with symptoms
that include foreign body sensation, matting of eyelashes, and
minimal discharge. Findings include an associated blepharitis
with thickened lid margins, telangiectasia, and pouting of
meibomian gland orices. Mild conjunctival injection, papillary
or follicular reaction, and scant mucopurulent discharge can
also be noted. The most common cause of chronic conjunctivitis
is Staphylococcus aureus. This organism releases exotoxins
that produce ulcerations of the lid margin as well as a punctate
epithelial keratopathy. Hypersensitivity reaction to S. aureus
results in marginal corneal inltrates as well as conjunctival
and corneal phlyctenules. Moraxella lacunata causes a chronic
angular blepharoconjunctivitis that features crusting and
ulceration of the medial and outer canthi. Staphylococcus aureus
blepharoconjunctivitis often accompanies this condition.
Enteric Gram-negative bacteria are more frequently associated
with chronic than acute conjunctivitis. These organisms include
Proteus species, Klebsiella pneumoniae, Serratia marcescens,
and Escherichia coli.
A persistent conjunctivitis unresponsive to empiric treatment requires culture of the eyelid and conjunctiva after a brief
washout period. Antibiotic susceptibility testing should then
provide a guide to appropriate antimicrobial therapy. Management also includes lid hygiene with lid scrubs, lid massage, and
warm compresses. Evaluation of the lacrimal system with
irrigation and culture is also warranted since the canaliculi and
lacrimal sac can be a chronic bacterial reservoir providing a
source of re-infection. Irrigation with the appropriate antibiotic
and oral therapy is then begun if an infectious agent is found.
Staphylococcal hypersensitivity reactions of the cornea may
respond to a mild topical corticosteroid if lid hygiene and antibiotics have no effect.
Chlamydial Conjunctivitis
Chlamydiae are obligate intracellular bacteria that have a
unique growth cycle. These organisms alternate between two
morphologic forms. The elementary body, which is metabolically inactive, is the extracellular and infectious form. The EB
binds to a mucosal epithelial cell and is taken up into intracellular inclusions. The EB then differentiates to the reticulate
body, which is the metabolically active form, and divides by
binary fusion. The RB then differentiates back to the infectious
EB which is released upon lysis of the host cell.
Chlamydia trachomatis causes ocular infection in two distinct
epidemiologic patterns. Trachoma is a chronic keratoconjunctivitis that is spread from eye to eye and is caused by repeated
infection with serotypes A, B, Ba, and C. The second pattern is
caused by sexual transmission of serotypes D through K
resulting in adult inclusion conjunctivitis. Neonatal chlamydial
conjunctivitis described previously is also caused by these
sexually transmitted strains.
Other chlamydial organisms cause ocular disease less
frequently. Parinaud oculoglandular syndrome has been associated with Chlamydia trachomatis serotypes L1, L2, and L3,
CHAPTER 50
been described.29,30 Prophylaxis for close contacts of patients

with rifampin 600 mg twice a day for 2 days or a single dose of
ciprofloxacin 500 mg is also recommended.31
707

which cause lymphogranuloma venereum, a sexually transmitted disease marked by painful inguinal lymphadenopathy.
Chlamydia pneumoniae and Chlamydia psittaci have been
identied as causes of follicular conjunctivitis.40
SECTION 6
Trachoma
708
Trachoma was once common in North America and Europe;

however, as living standards improved, trachoma disappeared in
these regions.41,42 In developing areas such as parts of Africa,
the Middle East, Southeast Asia, and South America, trachoma
remains a major public health problem. In these endemic areas
where overcrowding and decreased sanitation are prevalent, the
disease is transmitted directly or indirectly from person to
person from ocular secretions via infected materials such as
hands, towels, or clothing. Flies have also been identied as a
vector for disease transmission.43 Most children in these areas
are infected by the age of 2 and provide a reservoir for further
disease propagation.
Acute infection by C. trachomatis serotypes A, B, Ba, or C
causes a mucopurulent conjunctivitis characterized by lymphoid follicles, particularly of the superior tarsal conjunctiva,
and by diffuse papillary hypertrophy. Tender preauricular
adenopathy may also be present. Follicles may not be seen in
children younger than 2 years of age and a papillary reaction
predominates. Repeated infections produce a chronic inflammatory state resulting in conjunctival brosis and scarring.
Inflammation also affects the cornea with epithelial keratitis
and supercial vascular pannus. Tarsal follicles become necrotic
and heal, forming linear or stellate scars (Arlts line). Involution
of follicles at the limbal conjunctiva results in shallow
depressions known as Herberts pits. The progressive scarring
distorts the lids causing entropion and trichiasis. Aqueous tear
deciency can also result from damage to goblet cells and
lacrimal glands. These changes induce constant trauma to the
cornea leading to ulceration, opacication, and blindness.
Several grading systems have been devised in an effort to
standardize diagnosis for public health control programs and
research. The MacCallan classication staged the disease by
conjunctival changes and has been widely used in the past.
This classication system, however, lacks prognostic value as it
does not address inflammation severity, corneal opacity, nor
visual impairment.44 The modied WHO system was developed
to address these issues and emphasizes the intensity of
inflammatory disease based on the presence of tarsal follicles (F)
and papillary hypertrophy (P). Conjunctival scarring (C),
trichiasis/entropion (T/E), and corneal scarring (CC) were also
graded.45 A simplied WHO system followed which was aimed
for use by nonspecialists and is widely used in endemic areas.46
This system assesses the presence or absence of ve signs.
TF: trachomatous inflammation follicular. Five or more
follicles at least 0.5 mm in diameter on the central
area of the upper tarsal conjunctiva.
TI: trachomatous inflammation intense. Pronounced
inflammatory thickening of the upper tarsal
conjunctiva obscuring 50% or more of the normal
deep tarsal vessels.
TS: trachomatous scarring. Presence of visible scars on
the tarsal conjunctiva.
TT: trachomatous trichiasis. At least one lash touching
the eyeball.
CO: corneal opacity. Any corneal opacity blurring the
pupil margin.
Clinical diagnosis requires at least two of the following clinical
signs: follicles on the upper tarsal conjunctiva, limbal follicles
or Herberts pits, typical conjunctival scarring, or vascular
pannus marked on the superior limbus.45 Cytologic studies with
Giemsa staining can reveal intracytoplasmic inclusion bodies

(HalberstaedterProwazek bodies), but these are usually not
seen in chronic trachoma. Other diagnostic tests include
immunofluorescence, enzyme immunoassays, and nucleic acid
amplication by polymerase chain reaction.36,47,48
Management of trachoma in endemic areas consists of a multifaceted effort to prevent blindness. The WHO recommends
the SAFE strategy to treat different stages of trachoma and to
address risk factors that may increase disease transmission.49
This acronym stands for: surgery for trichiasis/entropion,
antibiotics for active disease, facial cleanliness, and environmental improvements. Surgical correction of lid deformities
can limit or prevent corneal damage and blindness. Though
several methods are utilized, tarsal rotation procedures provide
the best result.50 Active trachoma is treated with topical tetracycline twice daily for 6 weeks. To decrease infection in endemic
areas, the WHO recommends mass antibiotic distributions in
any community with a prevalence of follicular trachoma in
children of 10% or greater.51 Treatment is repeated annually for
3 years until prevalence falls below 5%. Mass treatment with a
single dose of azithromycin has been shown to decrease prevalence in the short term.52 Infection in children has been shown
to return 24 months after single-dose azithromycin treatment,
suggesting that repeat treatment is necessary to fully eliminate
disease.53
Adult Inclusion Conjunctivitis

Adult inclusion conjunctivitis is a sexually transmitted disease
caused by C. trachomatis serotypes D through K and presents as
a mucopurulent conjunctivitis affecting one or both eyes
beginning 12 weeks after inoculation. This condition is most
prevalent in sexually active adolescents and young adults who
can present with a concurrent urethritis or cervicitis. Signs and
symptoms can be mild and persist for months if left untreated.
Findings include a palpable preauricular adenopathy, scanty
mucopurulent discharge, and a follicular response that is most
prominent in the inferior conjunctival fornices. Corneal
changes include supercial punctate keratitis, supercial
micropannus, marginal inltrates, and subepithelial inltrates
resembling adenoviral keratitis.
Systemic antibiotic therapy is indicated to cover nonocular
chlamydial disease such as urethritis, epididimytis, cervicitis,
endometritis, and salpingitis. Oral treatment with a 3-week
course of tetracycline 250 mg four times a day, doxycycline
100 mg twice a day, or erythromycin 500 mg four times a day is
recommended. A single oral dose of azithromycin 1 g has also
been found to be effective.54 Sexual partners of patients with
AIC should also be treated.
CORNEAL INFECTIONS
Bacterial keratitis can lead to severe visual disability and requires
prompt diagnosis and treatment. Sequelae can vary in severity
from corneal scarring to perforation, endophthalmitis, and loss
of the eye. Although the corneal surface is awash with microorganisms of the normal flora, an intact corneal epithelium and
ocular defense mechanism prevent infection from setting in.
While some organisms such as Neisseria gonorrhoeae, Neisseria
meningitides, Corynebacterium diptheriae, Listeria, and Shigella
can penetrate an intact epithelium, others require damage to
the epithelial layer to invade the cornea. Several risk factors
predispose the cornea to bacterial infection. Dry eyes from
Sjgren syndrome, StevensJohnson syndrome, or vitamin A
deciency can result in bacterial keratitis. Prolonged corneal
exposure from ectropion, lagophthalmos, or proptosis can lead
to secondary infection. Entropion and trichiaisis resulting in
epithelial defects put the cornea at risk. Neurotrophic
keratopathy from cranial neuropathy, or prior herpes simplex

or zoster infections predispose to secondary infections. Some
systemic conditions such as chronic alcoholism, severe
malnutrition, immunosuppressive drug use, immunodeciency
syndromes, and malignancy can impair immune defenses and
allow infection by unusual organisms. Prior ocular surgery such
as penetrating keratoplasty or refractive procedures are also risk
factors.55,56 Trauma is a common predisposing factor of bacterial
keratitis, especially for patients at the extremes of age and in
developing countries.57,58 Injury to the corneal surface and
stroma allows invasion of normal flora as well as organisms
harbored by foreign bodies.
Contact lens wear is an established risk factor for bacterial
keratitis.5963 All types of contact lenses can cause infection,
with extended-wear soft lenses conferring greater risk than daily
wear hard or soft lenses.64 Corneal changes from contact lens
use include an induced hypoxic and hypercapnic state
promoting epithelial cell desquamation and allowing bacterial
invasion.65 Contact lenses also induce dry eye and corneal
anesthesia. Overnight rigid gas-permeable lens use for orthokeratology has also been associated with bacterial keratitis.66
Although there are geographic variations in the order of
incidence, the most common pathogenic organisms associated
with bacterial keratitis include Staphylococcus species,
Streptococcus species, Pseudomonas aeruginosa, and enteric
Gram-negative rods.58,67,68 A 5-year review of bacterial keratitis
isolates from Pittsburgh showed a change in distribution with a
decrease in Gram-positive organisms while Gram-negative
isolates remained stable.69 In South Florida, an increase in
gram-positive isolates with a decrease in Gram-negative isolates
over a 30-year period has been reported.70 A similar trend has
also been reported in north China.71 Pseudomonas aeruginosa
is commonly associated with contact-lens-related bacterial
keratitis, causing up to two-thirds of cases, although a decline
in the frequency of P. aeruginosa isolates in these patients has
been noted.60,7274 Nontuberculous mycobacteria is being
reported with increasing frequency as a cause of infectious
keratitis after laser in situ keratomileusis.75,76 Although the
reported incidence of infection after LASIK is low, this condition
is a management challenge requiring proper diagnosis and
treatment.77
CLINICAL FEATURES
The presenting symptoms, clinical history, and exam ndings
may suggest an infectious keratitis but are not diagnostic for a
particular organism. The presenting signs of bacterial keratitis
vary depending on the virulence of the organism, duration of
infection, structural status of the cornea, and host inflammatory response.
Common presenting symptoms include pain, decreased
vision, tearing, and photophobia. Eyelid edema, conjunctival
hyperemia with a papillary reaction, and chemosis are typical
ndings. A corneal epithelial defect with adherent mucopurulent exudate and underlying stromal inltrate is a hallmark
sign for infectious keratitis. Multiple focal inltrates can be
seen with contact lens use or with polymicrobial infections.
Migration of inflammatory cells causes a diffuse cellular inltration adjacent to the ulcerated stroma. An anterior chamber
reaction can range from mild aqueous cells and flare to a
marked hypopyon. A cornea damaged from prior disease can
present with less distinct signs and symptoms. Preexisting
corneal scars, epitheliopathy, or inflammation confuse the
picture as do prior use of antibiotics and corticosteroids. On
examination, all ocular abnormalities should be documented in
detail to help track the clinical course on subsequent visits.
Repeat measurements of the size of the epithelial defect, the
depth of the stromal inltrate, and the severity of inflammation, can be used to assess the effectiveness of treatment.
Some clinical features may be characteristic of certain
bacteria. These features, however, are not pathognomonic and
should not supplant laboratory studies with smear and culture
in providing the best therapy. Staphylococcus species reside in
the normal ocular flora and often cause keratitis in compromised corneas. Patients present with well-dened gray-white
stromal inltrates with mild surrounding stromal edema.
Infection with S. aureus can lead to rapid suppuration with deep
stromal abscess while S. epidermidis may have a more indolent
course. Streptococcus pneumoniae keratitis is associated with a
central stromal inltrate with a leading edge forming a serpiginous contour. Deep stromal involvement results in stromal
edema and radiating folds in Descemets membrane. Hypopyon
is usually present as is a retrocorneal brin plaque. Streptococcus viridans is a common cause of infectious crystalline
keratopathy in which bacterial growth occurs between stromal
lamellae in the setting of a suppressed or impaired immune
response.78,79 Gram-negative bacteria such as Pseudomonas
aeruginosa are less frequent causes of ICK.80 Most cases occur
after penetrating keratoplasty with topical steroid use, but cases
after LASIK have also been reported.81
Pseudomonas aeruginosa keratitis is characterized by a
rapidly progressive stromal inltrate with an adherent mucopurulent exudate. A ground glass stromal edema as well as a
ring inltrate surrounds the ulcer and a hypopyon is usually
present. The yellow necrotic ulcer becomes increasingly suppurative with marked stromal thinning. Pseudomonas
aeruginosa elaborates proteases which directly degrade corneal
stroma leading to descemetocele formation or perforation. The
infection can also extend into adjacent sclera.82 Other Gramnegative organisms such as Proteus, Serratia, Enteroabacter,
and Klebsiella may present in a similar manner. Risk factors
for Serratia keratitis include an abnormal corneal surface,
contact lens wear, and use of topical medications.83 Klebsiella
keratitis is usually seen in debilitated or immunocompromised
patients.84 Moraxella is a Gram-negative diplobacillus that
colonizes the nasopharynx and causes keratitis after trauma in
patients with chronic alcoholism and malnutrition.85 Clinical
presentation varies from an indolent shallow peripheral
ulceration with mild anterior chamber reaction to a deep central
ulcer with severe stromal and anterior chamber reaction that
may perforate.86
Gonococcal conjunctivitis marked by a hyperacute onset of
severe hyperemia and copious purulent discharge can infect the
cornea leading to rapid stromal necrosis, perforation, and endophthalmitis. Bacillus species are identied in soil and have been
linked to ocular infection after trauma due to contaminated
metallic foreign bodies.87 In addition, Bacillus keratitis has been
associated with failure of contact lens disinfectant to kill
spores.88 Bacillus cereus keratitis is marked by a corneal ring
inltrate with rapid progression to stromal abscess, perforation
and intraocular extension. Its virulence may be associated with
the production of toxins including phospholipases, proteases,
hemolysins, enterotoxin, and emetic toxin.89
Corynebacterium diptheriae is a Gram-positive rod that can
penetrate an intact corneal epithelium. C. diptheriae keratitis
presents as a gray epithelial haze that can quickly progress to
stromal dissolution.90 Clostridium species, also Gram-positive
rods, are anaerobic bacteria that produce gas and as a result
have a characteristic clinical appearance of subepithelial or
anterior chamber air.91 Other anaerobic organisms associated
with keratitis include Propionibacterium, Peptostreptococcus,
and Prevotella species.92 Prior trauma, corneal surgery, contact
lens wear, and chronic topical steroid use predispose patients to
P. acnes keratitis which has an indolent clinical course.93
CHAPTER 50
709
SECTION 6

Nocardia are aerobic Gram-positive lamentous bacteria that
are partially acid-fast and a rare cause of infectious keratitis.
Nocardia asteroides is the most commonly encountered species.
Trauma with vegetative matter, dirt, or gravel as well as contact
lens use are common predisposing factors.94 The keratitis
presents as anterior stromal patchy inltrates arranged in a
wreath pattern. Satellite lesions may be seen along with a
hypopyon. The clinical picture resembles mycotic keratitis
which can delay proper diagnosis and treatment.
Bacterial keratitis from Mycobacterium is caused by
nontuberculous species, while primary corneal infection from
Mycobacterium tuberculosis and Mycobacterium leprae is rare.
Nontuberculous mycobacteria are aerobic acid-fast rods that are
ubiquitous and can be found in water, dust, soil, animals, milk,
and other foodstuffs. These organisms have lipid-rich cell walls
which contribute to their acid-fast staining characteristics. Nontuberculous mycobacteria cause an indolent keratitis usually
occurring after trauma or surgery, including cataract extraction
and penetrating keratoplasty, and rarely with contact lens
use.9598 The organisms implicated in keratitis include Mycobacterium chelonae, Mycobacterium fortuitum, Mycobacterium
gordonae, and Mycobacterium avium-intracellulare. Clinical
signs develop 28 weeks after the inciting event. The ulcer typically lacks suppuration and can be multifocal on presentation.
Unusual clinical presentations include a linear pseudodendritiform appearance, ring inltrate, and crystalline keratopathy.99
Nontuberculous mycobacterial keratitis has been reported with
increasing frequency after laser in situ keratomileusis including
several clusters of cases.76,100,101 In two recent reviews of postLASIK corneal infections, Mycobacterium represented the most
common etiologic organism.77,102 The isolated subtypes include the
fast-growing Mycobacterium chelonae, Mycobacterium abscessus,
Mycobacterium fortuitum, and Mycobacterium mucogenicum,
as well as the slow-growing Mycobacterium szulgai. Nontuberculous keratitis after LASIK is characterized by a delayed
onset with an indolent course. Time of onset from fast-growing
organisms averaged 3.4 weeks after the procedure while the
slow-growing M. szulgai can present 624 weeks after
surgery.102,103 Symptoms can range from a mild foreign body
sensation to pain, redness, photophobia, and decreased vision.
The inltrate, which can be multiple, begins in the interface
and spreads to adjacent stroma of the flap and stromal bed.
Anterior perforation through the flap can occur with progression of infection. The location can be central, paracentral,
or peripheral. In addition to a focal inltrate, a cracked windshield appearance of infectious crystalline keratopathy has been
reported.81,104
Appropriate laboratory studies should be performed if
nontuberculous keratitis is suspected. The flap should be lifted
to obtain sample unless the inltrate is in the periphery or has
perforated through the flap. In addition to Gram and Giemsa
stains, ZiehlNeelsen stain to check for acid fastness and
fluorochrome stain which reveals yellow-orange fluorescence
should be obtained to identify Mycobacterium on smears.
Careful microscope diagnosis is warranted since misdiagnoses
with Nocardia and Corynebacterium have occurred.105 Direct
inoculation of inspissated egg solid LowensteinJensen media
and Middlebrook 7H9 or 7H12 broth media isolates
Mycobacterium. Fast-growing Mycobacteria are culture positive
within 7 days while slow growers require a few weeks.
TREATMENT
ROUTES OF ADMINISTRATION
710
The topical application of drugs with eyedrops is the preferred

method of treatment of bacterial keratitis. Increased drug
penetration can be achieved by higher concentrations, more

frequent applications, and by the typical presence of an
epithelial defect. Fortied antibiotics are made by mixing the
powdered drug or diluting the parenteral form with articial
tears or balanced salt solution. These freshly prepared solutions
remain stable for up to a week without signicant loss of
activity. Although ointments prolong corneal contact time and
lubricate the ocular surface, peak corneal concentrations may be
limited when compared with solutions. Ointments can be used
as adjunctive therapy at bedtime in less severe cases.
Subconjunctival injections may not have a therapeutic
advantage over topical solutions.106 However, they may be
indicated in certain clinical situations such as imminent
perforation or spread of infection to adjacent sclera, especially
when patient compliance is an issue. Soft contact lenses and
collagen shields can act as drug delivery devices and aid in
sustaining high corneal drug levels.107,108 Bandage contact
lenses may also provide structural support to promote
reepithelialization. Systemic therapy is indicated for gonococcal
infections as well as for young children with severe
H. influenzae or P. aeruginosa keratitis. Systemic antibiotics are
also indicated for perforations and scleral involvement.
Empiric Therapy
Since bacterial keratitis can rapidly progress and threaten
vision, treatment should be begun when an infectious process
is suspected. Topical broad spectrum antibiotics are initially
used and later modied according to culture results, antibiotic
susceptibilities, and clinical response. For severe cases,
combination therapy with fortied beta-lactam (cefazolin
50 mg/mL) and aminoglycoside (tobramycin or gentamicin
14 mg/mL) provides adequate coverage of both Gram positive
and negative organisms that cause bacterial keratitis. Vancomycin (50 mg/mL) can be substituted for cefazolin in cases of
penicillin allergy or resistance to Enterococcus and Staphylococcus species. A loading dose is achieved with a drop every
5 min for ve applications. Antibiotic is then continued every
30 min to 1 h around the clock.
Single-agent therapy with fluoroquinolones has been shown
to be as effective as combination therapy in treating bacterial
keratitis.109111 The widespread use of the second- (ciprofloxacin
and ofloxacin) and third- (levofloxacin) generation fluoroquinolones has, however, led to the emergence of resistance in
several bacterial species including Staphylococcus aureus and
Pseudomonas aeruginosa.69,74,112114 The fourth-generation fluoroquinolones gatifloxacin and moxifloxacin have been developed
as a response to this rising resistance. They require two mutations to establish resistance and, therefore, are more effective
against Gram-positive organisms that already have a single mutation and are resistant to older-generation fluoroquinolones.115118
A favorable response to empiric therapy merits continuing
the treatment plan. Positive signs of clinical improvement
include decreased pain, decreased discharge, consolidation of
the stromal inltrate, decreased anterior chamber reaction, and
corneal reepithelialization. Culture and antibiotic susceptibility
results can be used to focus therapy against the offending organism or to discontinue unnecessary drugs. Clinical improvement
may not be seen during the rst 2 days due to increased inflammation and suppuration from bacterial exotoxins. Toxicity from
topical medications can also mask any changes. A lack of
improvement or clinical worsening after 48 h may warrant
repeat cultures, although concomitant antibiotic therapy will
decrease yields. Topical therapy can be tapered as the clinical
picture improves.
Management of nontuberculous mycobacterial keratitis after
LASIK can be challenging and requires aggressive treatment.
The flap should be lifted for smears and culture as well as for

soaking of the stromal bed and flap with antibiotics. Fortied
amikacin, clarithromycin, or azithromycin are the drugs of
choice.102,119 Fourth generation fluoroquinolones have also been
shown to be effective against mycobacterial keratitis.120,121
Combination therapy is recommended due to emergence of
resistance on monotherapy.103 Lack of clinical improvement
warrants repeat culture and tailoring of antibiotics accordingly.
Flap amputation may also be necessary to allow increased
antibiotic penetration.
Adjunctive Therapy
Bacterial keratitis is often associated with severe pain. Pain
control with analgesics may provide not only comfort, but also
increased compliance with the difcult regimen of around the
clock topical drops. Cycloplegic agents can also be used to
decrease discomfort from ciliary spasm and to prevent
synechiae formation. Cyanoacrylate glue can be used to reinforce an area of corneal thinning, a descemetocele, or a small
perforation. A bandage contact lens is placed after the glue
hardens. This procedure allows for further treatment of the
infection and inflammation while postponing surgery. A corneal
patch graft is an alternative for small perforations while larger
necrotic perforations require a therapeutic penetrating keratoplasty. Maximal topical antibiotic therapy as well as systemic
antibiotics is given preoperatively.
Corticosteroids may play a limited role in treating bacterial
keratitis with its potential for reducing the host inflammatory
response and resultant corneal scarring. Adverse effects of
corticosteroids include inhibition of corneal wound healing,
promotion of stromal thinning and perforation, potentiation of
microbial replication and recrudescence of infection, secondary
glaucoma, and cataract formation. Despite its theoretical
advantages, studies have not shown a consistent or signicant
benecial effect of corticosteroids on clinical outcome.122 Prior
use of corticosteroids in eyes with preexisting corneal disease
increased the risk of ulcerative keratitis. Worsening or recrudescence of Pseudomonas keratitis has been reported after the
addition of topical steroids.123,124 Guidelines regarding the
optimal use of corticosteroids are lacking; however, certain
recommendations have been proposed: (1) steroids should not
be used initially or if the eye is improving, (2) steroids should be
used after several days of antibiotics if there is persistent
inflammation, (3) continue use of concomitant antibiotics, and
(4) steroids should not be used if there is corneal thinning.125
1. Mannis MJ, Smolin G: Natural defense

mechanisms of the ocular surface. In:
Pepose JS, Holland GN, Wilhemus KR, eds.
Ocular infection and immunity. St Louis:
Mosby; 1996:185190.
2. Akpek EK, Gottsch JD: Immune defense
at the ocular surface. Eye 2003; 17:
949956.
3. Osato MS: Normal ccular flora. In:
Pepose JS, Holland GN, Wilhemus KR, eds.
Ocular infection and immunity. St Louis:
Mosby; 1996:191199.
4. Singer TR, Isenberg SJ, Apt L: Conjunctival
anaerobic and aerobic bacterial flora in
paediatric versus adult subjects. Br J
Ophthalmol 1988; 72: 448451.
5. OBrien TP, Hazlett LD: Pathogenesis of
ocular infection. In: Pepose JS, Holland GN,
Wilhemus KR, eds. Ocular infection and
immunity. St Louis: Mosby; 1996:200214.
6. OBrien TP: Management of bacterial
keratitis: beyond exorcism towards
consideration of organism and host factors.
Eye 2003; 17: 957974.
7. Zegans ME, et al: The role of bacterial
biolms in ocular infections. DNA Cell Biol
2002; 21: 415420.
8. Engel LS, et al: Pseudomonas aeruginosa
protease IV produces corneal damage
and contributes to bacterial virulence.
Invest Ophthalmol Vis Sci 1998; 39:
662665.
9. Matsumoto K: Role of bacterial proteases
in pseudomonal and serratial keratitis. Biol
Chem, 2004; 385: 10071016.
10. Dajcs JJ, et al: Corneal virulence of
Staphylococcus aureus in an experimental
model of keratitis. DNA Cell Biol 2002; 21:
375382.
11. Dana MR, Qian Y, Hamrah P: Twenty-veyear panorama of corneal immunology:
transplant rejection. Cornea 2000; 19:
625643.
12. Hazlett LD: Corneal response to
Pseudomonas aeruginosa infection. Prog
Retin Eye Res 2004; 23:130.
13. McDonnell PJ, et al: Community care of

corneal ulcers. Am J Ophthalmol 1992;
114: 531538.
14. Charukamnoetkanok P, Pineda R 2nd.
Controversies in management of bacterial
keratitis. Int Ophthalmol Clin 2005;
45:199210.
15. Badenoch PR, Coster DJ: Antimicrobial
activity of topical anaesthetic
preparations. Br J Ophthalmol 1982;
66:364367.
16. OBrien TP: Bacterial keratitis. In: Foster CS,
Azar DT, Dohlman CH, eds. Smolin and
Thofts the cornea. Philadelphia: William &
Wilkins;. 2005:235288.
17. Alexandrakis G, et al: Corneal biopsy in
the management of progressive microbial
129:571576.
18. Seal DV, Barrett SP, McGill JI: Aetiology
and treatment of acute bacterial infection
of the external eye. Br J Ophthalmol 1982;
66:357360.
19. Gigliotti F, et al: Etiology of acute
conjunctivitis in children. J Pediatr 1981;
98:531536.
20. Bodor FF, et al: Bacterial etiology of
conjunctivitis-otitis media syndrome.
21. Brazilian purpuric fever: Haemophilus
aegyptius bacteremia complicating purulent
conjunctivitis. MMWR Morb Mortal Wkly
Rep 1986; 35:553554.
22. Gigliotti F, et al: Efcacy of topical antibiotic
therapy in acute conjunctivitis in children.
J Pediatr 1984; 104:623626.
23. Leibowitz HM: Antibacterial effectiveness of
ciprofloxacin 0.3% ophthalmic solution in
the treatment of bacterial conjunctivitis.
Am J Ophthalmol 1991; 112(4 Suppl):
29S33S.
24. Fraunfelder FT, Bagby GC Jr, Kelly DJ:
Fatal aplastic anemia following topical
administration of ophthalmic
chloramphenicol. Am J Ophthalmol 1982;
93:356360.
25. Barquet N, et al: Primary meningococcal
conjunctivitis: report of 21 patients and
review. Rev Infect Dis 1990; 12:838847.
26. Haimovici R, Roussel TJ: Treatment of

gonococcal conjunctivitis with single-dose
intramuscular ceftriaxone. Am J
Ophthalmol 1989; 107:511514.
27. Centers for Disease Control and Prevention.
Sexually transmitted diseases treatment
guidelines 2002. Morb Mortal Wkly Rep
2002; 51(No RR-6): 3041.
28. Judson FN: The importance of coexisting
syphilitic, chlamydial, mycoplasmal, and
trichomonal infections in the treatment of
gonorrhea. Sex Trans Dis 1979; 6(2 Suppl):
112119.
29. Orden B, et al: Primary meningococcal
conjunctivitis. Clin Microbiol Infect 2003;
9:12451247.
30. Andreoli CM, et al: Primary meningococcal
conjunctivitis in an adult. Cornea 2004;
23:738739.
31. Rosenstein NE, et al: Meningococcal
disease. N Engl J Med 2001;
344:13781388.
32. Hammerschlag MR, Rapoza P: Neonatal
conjunctivitis. In: Pepose JS, Holland GN,
Wilhemus KR, eds. Ocular infection and
immunity. St Louis: Mosby; 1996:831842.
33. Nishida H, Risemberg HM: Silver nitrate
ophthalmic solution and chemical
conjunctivities. Pediatrics 1975;
56:368373.
34. Roblin PM, et al: Comparison of two rapid
microscopic methods and culture for
detection of Chlamydia trachomatis in
ocular and nasopharyngeal specimens from
infants. J Clin Microbiol 1989; 27:968970.
35. Stenberg K, et al: Culture, ELISA and
immunofluorescence tests for the diagnosis
of conjunctivitis caused by Chlamydia
trachomatis in neonates and adults. Apmis
1990: 98:514520.
36. Talley AR, et al: The use of polymerase
chain reaction for the detection of
chlamydial keratoconjunctivitis. Am J
Ophthalmol 1992; 114:685692.
37. Hammerschlag MR, et al: Efcacy of
neonatal ocular prophylaxis for the
prevention of chlamydial and gonococcal
conjunctivitis. N Engl J Med 1989;
320:769772.
CHAPTER 50
REFERENCES
711
SECTION 6
712
38. Benevento WJ, et al: The sensitivity of

Neisseria gonorrhoeae, Chlamydia
trachomatis, and herpes simplex type II to
disinfection with povidone-iodine. Am J
Ophthalmol 1990; 109:329333.
39. Isenberg SJ, Apt L, Wood M: A controlled
trial of povidone-iodine as prophylaxis
against ophthalmia neonatorum. N Engl
J Med 1995; 332:562566.
40. Lietman T, et al: Chronic follicular
conjunctivitis associated with Chlamydia
psittaci or Chlamydia pneumoniae. Clin
Infect Dis 1998; 26:13351340.
41. Mabey DC, Solomon AW, Foster A:
Trachoma. Lancet 2003; 362:223229.
42. Allen SK, Semba RD: The trachoma
menace in the United States, 18971960.
43. Miller K, et al: Pesky trachoma suspect
nally caught. Br J Ophthalmol 2004;
88:750751.
44. Solomon AW, et al: Diagnosis and
assessment of trachoma. Clin Microbiol
Rev 2004; 17:9821011.
45. Dawson CR, Jones BR, Tarizzo ML: Guide
to trachoma control in programs for the
prevention of blindness. Geneva: World
Health Organization; 1981.
46. Thylefors B, et al: A simple system for the
assessment of trachoma and its
complications. Bull World Health Organ
1987; 65:477483.
47. Uyeda CT, et al: Rapid diagnosis of
chlamydial infections with the MicroTrak
direct test. J Clin Microbiol 1984;
20:948950.
48. Mabey DC, Robertson JN, Ward ME:
Detection of Chlamydia trachomatis by
enzyme immunoassay in patients with
49. Report of the Second Meeting of the WHO
Alliance for the Global Elimination of
Trachoma. Geneva: World Health
Organization; 1998.
50. Reacher MH, et al: A controlled trial of
surgery for trachomatous trichiasis of the
upper lid. Arch Ophthalmol 1992;
110:667674.
51. Report of the Eighth Meeting of the WHO
Alliance for the Global Elimination of
Blinding Trachoma. Geneva: World Health
Organization; 2004.
52. Solomon AW, et al: Mass treatment with
single-dose azithromycin for trachoma.
N Engl J Med 2004; 351:19621971.
53. Chidambaram JD, et al: Effect of a single
mass antibiotic distribution on the
prevalence of infectious trachoma. JAMA
2006; 295:11421146.
54. Katusic D, et al: Azithromycin vs doxycycline
in the treatment of inclusion conjunctivitis.
Am J Ophthalmol 2003; 135:447451.
55. Siganos CS, Solomon A, Frucht-Pery J:
Microbial ndings in suture erosion after
1997; 104:513516.
56. Sampath R, Ridgway AE, Leatherbarrow B:
Bacterial keratitis following excimer laser
photorefractive keratectomy: a case report.
Eye 1994; 8( Pt 4): 481482.
57. Parmar P, et al: Microbial keratitis at
extremes of age. Cornea 2006;
25:153158.
58. Laspina F, et al: Epidemiological
characteristics of microbiological results on
patients with infectious corneal ulcers: a
13-year survey in Paraguay. Graefes Arch
59. Dixon JM, et al: Complications associated

with the wearing of contact lenses. JAMA
1966; 195:901903.
60. Alfonso E, et al: Ulcerative keratitis
associated with contact lens wear. Am J
Ophthalmol 1986; 101:429433.
61. Schein OD, et al: The incidence of microbial
keratitis among wearers of a 30-day
silicone hydrogel extended-wear contact
lens. Ophthalmology 2005; 112:21722179.
62. Driebe WT Jr: Present status of contact
lens-induced corneal infections. Ophthalmol
Clin North Am 2003; 16:485494, viii.
63. Mah-Sadorra JH, et al: Trends in contact
lens-related corneal ulcers. Cornea 2005;
24:5158.
64. Liesegang TJ: Contact lens-related
microbial keratitis. Part I. Epidemiology.
Cornea 1997; 16:125131.
65. Liesegang TJ: Contact lens-related
microbial keratitis. Part II. Pathophysiology.
Cornea 1997; 16:265273.
66. Hsiao CH, et al: Infectious keratitis related
to overnight orthokeratology. Cornea 2005;
24:783788.
67. Keay L, et al: Microbial keratitis predisposing
factors and morbidity. Ophthalmology
2006; 113:109116.
68. Bourcier T, et al: Bacterial keratitis:
predisposing factors, clinical and
microbiological review of 300 cases. Br J
Ophthalmol 2003; 87:834838.
69. Goldstein MH, Kowalski RP, Gordon YJ:
Emerging fluoroquinolone resistance in
bacterial keratitis: a 5-year review.
Ophthalmology 1999; 106:13131318.
70. Forster RK: Conrad Berens Lecture. The
management of infectious keratitis as we
approach the 21st century. CLAO J 1998;
24:175180.
71. Sun X, et al: Distribution and shifting trends
of bacterial keratitis in north China (198998).
72. Cohen EJ, et al: Corneal ulcers associated
with cosmetic extended wear soft contact
lenses. Ophthalmology 1987; 94:109114.
73. Cohen EJ, et al: Trends in contact lensassociated corneal ulcers. Cornea 1996;
15:566570.
74. Alexandrakis G, Alfonso EC, Miller D:
Shifting trends in bacterial keratitis in south
Florida and emerging resistance to
fluoroquinolones. Ophthalmology 2000;
107:14971502.
75. Solomon R, et al: Infectious keratitis after
laser in situ keratomileusis: results of an
ASCRS survey. J Cataract Refract Surg
2003; 29:20012006.
76. Freitas D, et al: An outbreak of
Mycobacterium chelonae infection after
LASIK. Ophthalmology 2003; 110:276285.
77. Chang MA, Jain S, Azar DT: Infections
following laser in situ keratomileusis:
an integration of the published literature.
78. Sridhar MS, et al: Infectious crystalline
keratopathy in an immunosuppressed
patient. CLAO J 2001; 27:108110.
79. Sharma N, et al: Infectious crystalline
keratopathy. CLAO J 2000; 26:4043.
80. Khater TT, Jones DB, Wilhelmus KR:
Infectious crystalline keratopathy caused
by gram-negative bacteria. Am J
Ophthalmol 1997; 124:1923.
81. Alvarenga L, et al: Infectious post-LASIK
crystalline keratopathy caused by
nontuberculous mycobacteria. Cornea
2002; 21:426429.
82. Raber IM, et al: Pseudomonas corneoscleral

ulcers. Am J Ophthalmol 1981; 92:353362.
83. Mah-Sadorra JH, et al: Serratia corneal
ulcers: a retrospective clinical study.
Cornea 2005; 24:793800.
84. Pinna A, et al: Detection of virulence
factors in a corneal isolate of Klebsiella
pneumoniae. Ophthalmology 2005;
112:883887.
85. Marioneaux SJ, et al: Moraxella keratitis.
Cornea 1991; 10:2124.
86. Garg P, et al: Treatment outcome of
Moraxella keratitis: our experience with
18 cases a retrospective review. Cornea
1999; 18:176181.
87. Choudhuri KK, et al: Clinical and
microbiological prole of Bacillus keratitis.
Cornea 2000; 19:301306.
88. Donzis PB, Mondino BJ, Weissman BA:
Bacillus keratitis associated with
contaminated contact lens care systems.
Am J Ophthalmol 1988; 105:195197.
89. Pinna A, et al: Bacillus cereus keratitis
associated with contact lens wear.
Ophthalmology 2001; 108:18301834.
90. Chandler JW, Milam DF: Diphtheria corneal
ulcers. Arch Ophthalmol 1978; 96:5356.
91. Stern GA, Hodes BL, Stock EL: Clostridium
perfringens corneal ulcer. Arch Ophthalmol
1979; 97:661663.
92. Brook I: Ocular infections due to anaerobic
bacteria. Int Ophthalmol 2001; 24:269277.
93. Underdahl JP, et al: Propionibacterium
acnes as a cause of visually signicant
corneal ulcers. Cornea 2000; 19:451454.
94. Sridhar MS, et al: Ocular nocardia
infections with special emphasis on the
cornea. Surv Ophthalmol 2001;
45:361378.
95. Sudesh S, et al: Mycobacterium chelonae
infection in a corneal graft. Arch
Ophthalmol 2000; 118:294295.
96. Servat JJ, et al: Mycobacterium
chelonaeMycobacterium abscessus
complex clear corneal wound infection with
recurrent hypopyon and perforation after
phacoemulsication and intraocular lens
implantation. J Cataract Refract Surg 2005;
31:14481451.
97. Mah-Sadorra JH, Cohen EJ, Rapuano CJ:
Mycobacterium chelonae wound ulcer after
clear-cornea cataract surgery. Arch
Ophthalmol 2004; 122:18881889.
98. Malecha MA, Doughman DJ:
Mycobacterium chelonae keratitis
associated with soft contact lens wear.
CLAO J 2002; 28:228230.
99. OBrien TP, Matoba AY: Nontuberculous
mycobacterial diseases. In: Pepose JS,
Holland GN, Wilhemus KR, eds. Ocular
infection & immunity, St Louis: Mosby;
1996:10331041.
100. Chandra NS, et al: Cluster of
Mycobacterium chelonae keratitis cases
following laser in-situ keratomileusis. Am J
Ophthalmol 2001; 132:819830.
101. Holmes GP, et al: A Cluster of cases of
Mycobacterium szulgai keratitis that
occurred after laser-assisted in situ
keratomileusis. Clin Infect Dis 2002;
34:10391046.
102. John T, Velotta E: Nontuberculous (atypical)
mycobacterial keratitis after LASIK: current
status and clinical implications. Cornea
2005; 24:245255.
103. Fulcher SF, et al: Delayed-onset
mycobacterial keratitis after LASIK. Cornea
2002; 21:546554.
112.
113.
114.
115.
116.
117.
118.
bacterial corneal ulcers. Ciprofloxacin

Bacterial Keratitis Study Group.
Ophthalmology 1996; 103:18541862;
Marangon FB, et al: Ciprofloxacin and
levofloxacin resistance among methicillinsensitive Staphylococcus aureus isolates
from keratitis and conjunctivitis. Am J
Ophthalmol 2004; 137:453458.
Garg P, Sharma S, Rao GN: Ciprofloxacinresistant Pseudomonas keratitis.
Ophthalmology 1999; 106:13191323.
Chaudhry NA, et al: Emerging ciprofloxacinresistant Pseudomonas aeruginosa. Am J
Ophthalmol 1999; 128:509510.
Kowalski RP, et al: Gatifloxacin and
moxifloxacin: an in vitro susceptibility
comparison to levofloxacin, ciprofloxacin,
and ofloxacin using bacterial keratitis
isolates. Am J Ophthalmol 2003;
136:500505.
Parmar P, et al: Comparison of topical
gatifloxacin 0.3% and ciprofloxacin 0.3%
for the treatment of bacterial keratitis. Am J
Ophthalmol 2006; 141:282286.
Dajcs JJ, et al: Effectiveness of
ciprofloxacin, levofloxacin, or moxifloxacin
for treatment of experimental
Staphylococcus aureus keratitis. Antimicrob
Aliprandis E, et al: Comparative efcacy of
topical moxifloxacin versus ciprofloxacin
and vancomycin in the treatment of
119.
120.
121.
122.
123.
124.
125.
P. aeruginosa and ciprofloxacin-resistant

MRSA keratitis in rabbits. Cornea 2005;
24:201205.
Karp CL, et al: Infectious keratitis after
LASIK. Ophthalmology 2003; 110:503510.
Abshire R, et al: Topical antibacterial
therapy for mycobacterial keratitis:
potential for surgical prophylaxis and
treatment. Clin Ther 2004; 26:191196.
Hyon JY, et al: Comparative efcacy of
topical gatifloxacin with ciprofloxacin,
amikacin, and clarithromycin in the
treatment of experimental Mycobacterium
chelonae keratitis. Arch Ophthalmol 2004;
122:11661169.
Wilhelmus KR: Indecision about
corticosteroids for bacterial keratitis: an
evidence-based update. Ophthalmology
2002; 109:835842;quiz 843.
Burns RP: Pseudomonas aeruginosa
keratitis: mixed infections of the eye. Am J
Ophthalmol 1969; 67:257262.
Harbin T: Recurrence of a corneal
pseudomonas infection after topical steroid
therapy: report of a case. Am J Ophthalmol
1964; 58: 670674.
Stern GA, Buttross M: Use of
corticosteroids in combination with
antimicrobial drugs in the treatment of
infectious corneal disease. Ophthalmology
1991; 98:847853.
CHAPTER 50
104. Umapathy T, et al: Non-tuberculous

mycobacteria related infectious crystalline
keratopathy. Br J Ophthalmol 2005;
89:13741375.
105. Garg P, Athmanathan S, Rao GN:
Mycobacterium chelonei masquerading as
Corynebacterium in a case of infectious
keratitis: a diagnostic dilemma. Cornea
1998; 17:230232.
106. Baum J, Barza M: Topical vs
subconjunctival treatment of bacterial
90:162168.
107. Willoughby CE, Batterbury M, Kaye SB:
Collagen corneal shields. Surv Ophthalmol
2002; 47:174182.
108. Kalayci D, et al: Penetration of topical
ciprofloxacin by presoaked medicated soft
contact lenses. CLAO J 1999; 25:182184.
109. OBrien TP, et al: Efcacy of ofloxacin vs
cefazolin and tobramycin in the therapy for
bacterial keratitis. Report from the Bacterial
Keratitis Study Research Group. Arch
Ophthalmol 1995; 113:12571265.
110. Ofloxacin monotherapy for the primary
treatment of microbial keratitis: a doublemasked, randomized, controlled trial with
conventional dual therapy. The Ofloxacin
Study Group. Ophthalmology 1997;
104:19021909.
111. Hyndiuk RA, et al: Comparison of
ciprofloxacin ophthalmic solution 0.3% to
fortied tobramycin-cefazolin in treating
713
CHAPTER
51
Fungal Keratitis
Denis ODay
The prognosis for fungal infections in the cornea has markedly

improved in recent years. A better understanding of the clinical
setting of fungal keratitis, the principles of treatment, the availability of more effective antifungal agents, and the role of surgery
have undoubtedly contributed to this more optimistic view.
TABLE 51.1. Risk Factors for Fungal Keratitis

Hot, humid climate
Agricultural workers
Injury with a vegetable foreign body
EPIDEMIOLOGY
Nonpenetrating trauma
The normal cornea is remarkably resistant to fungal infection.

However, when normal host defenses are breached either
through underlying disease or trauma, fungi can readily invade
the cornea. Fungal infections occur all over the world but the
incidence is highest in warm and humid climates and particularly in a rural environment. Unfortunately, it is in these climatic
conditions that population densities tend to be greatest while at
the same time resources for effective treatment are scarce.
Over the last several decades, concise epidemiological studies
have emerged from many of these locations and these paint a
rather frightening picture of the devastating consequences of
microbial infections in the cornea and of fungal infections in
particular.16 In India, for example, a recent study suggests that
~1% of the population or ~9 million people will acquire
microbial keratitis in 10 years and 50% of these cases will be
fungal.7 Extrapolating these figures to countries in the
equatorial belt provides a useful estimate of the magnitude of
the problem. When coupled with what is known about the
dearth of effective treatment in these countries, they offer a
glimpse of the potentially devastating effects of corneal
blindness from fungal infections in these populations.
Although there are a myriad of potential fungal pathogens,
the overwhelming majority of cases are caused by a small
number of filamentous fungi and yeasts. Evidence developed
from a review of published cases and case series some years ago
demonstrated the importance of the molds, Fusarium and
Aspergillus and the Candida species of yeasts as principal
pathogens.8 This analysis has now been confirmed by
numerous case series from around the world.16 Far behind
these in fourth place, according to one study, sits infection with
Curvularia sp.9 It is now clear that while many other fungi can
invade the cornea, the incidence of such cases is very much
lower and although there are exceptions, for the most part these
organisms tend to be less virulent. The large series of cases now
being reported from countries in the equatorial belt stand in
contrast to the paucity of cases in the temperate zones. The
spectrum of fungi is also different. Candida species become the
most frequent isolates; whereas Fusarium is rarely reported.
When considering corneal infections of any kind, an
important issue to consider is the question of risk factors
(Table 51.1). Agricultural workers seem to be at greatest risk
Immunocompromised patient
Topical steroid administration
Dry eye
Neurotrophic cornea
HIV/AIDS
and the inciting agent is usually minor corneal trauma with

a vegetable foreign body.4 Studies performed in tropical and
temperate climates show the rate of microbial corneal
ulceration in a tropical climate to be 30 times that in the
more temperate zone.6 Since about half of these ulcers are
fungal in tropical zones where the population is predominantly
rural, living in a tropical climate and working in agriculture
are clearly significant risks, possibly amplified by the lack of
safety precautions in the less advanced agricultural societies.4,10
In a study reported several years ago, Gopithanan reported that
amongst 1353 cases, males were much more likely to be
involved than females by a factor of 2.5 to 1. Almost 30% were
agricultural workers, the remainder working in menial labor
or were unemployed.4
The injury that facilitates fungal infection is usually
nonpenetrating and may actually be quite trivial even though
it is caused by vegetable material. Other risk factors for fungal
infection include an immunocompromised host, topical or
systemic corticosteroid administration, dry eye and neurotrophic cornea. Fungal keratitis is relatively rare amongst
individuals who are human immunodeficiency virus (HIV)
positive.11
CLINICAL FEATURES
This brief epidemiological background provides a practical basis
for reviewing the clinical features of fungal keratitis and its
management. Keratitis caused by filamentous fungi and
yeasts present differently and are managed in different ways.
The approach in this section is to consider each separately.
715
FIGURE 51.1. Early stage of keratomycosis due to Acremonium

species following a corneal abrasion. Note shallow central stromal
ulceration with underlying grey infiltrate.
SECTION 6
FIGURE 51.3. Severe Fusarium solani keratitis with well-marked

immune ring.
716
FIGURE 51.2. Aspergillus flavus keratitis. Note slightly elevated fungal

plaque on the corneal surface.
FILAMENTOUS FUNGAL INFECTION

The onset is usually insidious. There is usually a history of
minor trauma; however, not uncommonly, the patient delays
seeking attention as the eye remains asymptomatic or only
mildly so. Even if seen immediately after the injury, evidence of
infection is unlikely to be visible for several days to a week and
the epithelium may even heal over the area of the precipitating
trauma. It is only after several days or in some cases a week or
longer that the patient becomes more aware of pain and
discomfort or notices a decrease in vision or a red eye and seeks
medical attention.
In this early period, there may be minimal anterior stromal
infiltrate and possibly a mild anterior chamber reaction
(Fig. 51.1). During this stage, the appearance may be easily
misconstrued as a bacterial infection. Gradually the infiltrate
increases in density and depth (Fig. 51.2), and so-called
FIGURE 51.4. Fungal keratitis with large hypopyon and endothelial

plaque.
satellite lesions may be observed. These are small focal areas

of infiltrate separated from the main lesion by clear cornea.
Occasionally during this early stage, with high magnification,
fungal filaments can be seen coursing through clear cornea on
the periphery of the lesion.
A partial or complete immune ring may develop as the
infection takes hold (Fig. 51.3). With intensification of the
inflammation, hypopyon can appear and may enlarge rapidly
(Fig. 51.4). Untreated, frank ulceration develops as the fungus
invades the deeper layers of the cornea. Although most
filamentous fungi appear to have the ability to pass through
Descemets membrane, the presence of an enlarging hypopyon
does not necessarily imply anterior chamber invasion. This
serious complication however, becomes more likely with the
development of a white plaque on the endothelial surface
underlying the lesion (Fig. 51.4).
Fungal Keratitis
TABLE 51.2. Principles of Management

Avoid specific treatment prior to definitive diagnosis
Topical therapy is the most effective
Systemic (oral) supplementation can be useful
Avoid steroids (topical and systemic)
Move promptly to therapeutic keratoplasty if progressing despite
treatment
FIGURE 51.5. Candida albicans keratitis in a patient with severe

keratoconjunctivitis sicca.
YEAST INFECTIONS
PRINCIPLES OF MANAGEMENT
The principles of management are listed in Table 51.2. A
prolonged period of treatment is likely to be required since the
action of most agents is fungistatic. Therefore, before
committing a patient to treatment that may also have toxic side
effects, it is important first to make a definitive diagnosis if at
all possible. The topical route has proven to be the most
effective route of administration. Subconjunctival injection of
antifungal agents is generally ineffective as well as carrying the
risk of significant toxicity. In some instances, particularly with
deep corneal infection and when anterior chamber invasion is
FIGURE 51.6. Combined infection with Streptococcus species and

Fusarium solani in a patient with chronic herpes simplex keratitis.
suspected, systemic (usually oral) therapy can be useful.1520

Corticosteroids should be avoided if at all possible, and in those
cases where antifungal treatment is appearing to be ineffective,
it is important to move to therapeutic keratoplasty promptly
rather than to wait until the disease is far advanced.
DIAGNOSIS
Stained smears from the lesion are helpful, but the diagnosis is
on firmest ground when an isolate is recovered. Fungi can be
visualized with most stains including Gram, Geimsa, Gomori
methenamine silver, potassium hydroxide (KOH) preparation,
periodic acid-Schiff (PAS), acridine orange, and calcofluor. A
study in 2002 showed that of these, the Gram, Geimsa, KOH,
and the calcofluor all had a high percentage of positivity
indicating that any of them can be used in an attempt to
identify the inciting organism.4 Recently polymerase chain
reaction (PCR) has been advocated as a means to increase the
speed of diagnosis.21
For isolate recovery, it is important always to consider the
possibility of multiple isolates (Fig. 51.6). Particularly with
trauma, bacterial and fungal infection may coexist and
occasionally two fungi may be isolated. Also, while the presence
of many of the features described above is said to be
characteristic of a fungal infection and should alert the viewer
to that possibility, they are by no means pathognomonic; the
clinical signs of fungal keratitis infection can be mimicked by
infection with other organisms. For these reasons it is preferable
to attempt recovery on both bacterial and fungal media and
CHAPTER 51
The clinical features of yeast infections are usually quite

dissimilar from those caused by filamentous fungi. The most
common yeast to infect the cornea is Candida albicans but
other members, particularly C. parapsilosis may at times be
isolated.12
Corneas infected with yeast are more likely to be abnormal
immunologically or to have defective defense mechanisms.
Thus, they usually occur in neurotrophic corneas, in patients
using high-dose corticosteroids (local or systemic), in the
presence of severe dry eye from various causes, or in topical
anesthetic abuse. Yeasts, as a rule, are not deeply invasive.
Yeasts tend to colonize on the corneal surface where they can
exist as white plaques or to penetrate only into the superficial
layers of the stroma (Fig. 51.5). In most instances the infection
proceeds very slowly. However, occasionally, a Candida albicans
strain may have the potential to invade deeply into the corneal
stroma, producing frank corneal ulceration and even invading
the anterior chamber. Evidence from animal studies suggest
that this is more likely to occur in the presence of steroid
administration.13 Yeast infections, usually with Candida
species, can also occur following corneal transplant surgery
when they appear around a suture placed in the cornea.
Candida albicans is also implicated as one of the inciting
organisms for crystalline keratopathy. Candida parapsilosis is
occasionally responsible for fungal infections particularly in
patients with HIV infection and following corneal
transplantation.14
717

even on occasions to consider the possibility of Acanthamoeba
infection.
A standard approach attempts isolate recovery on sheep
blood agar, Sabourauds agar and if possible a liquid fungal
medium such as brainheart infusion broth. With suspected
mycotic infections, the C streak technique in which a series of
Cs are marked on the plate is particularly useful in identifying
the site of inoculation. While making a culture, airborne fungal
spores may alight on the plate, and since the types of fungal
organisms that are likely to be isolated are also known
environmental inhabitants, there is always a risk of
contamination in the course of attempting isolate recovery. By
making a C streak one can be reasonably certain that the
organism has been deposited in the C streak by the instrument
scraping the cornea. In the humid tropical localities where
fungal keratitis is most prevalent, this risk is probably
accentuated.
Since some fungi are slow to grow, an effort should be made
to keep the plates for longer than the usual period of
observation. Placing a wet 4 4 sponge with the plate in a
sealed plastic bag can help prolong its life.
TABLE 51.3. Antifungal Agents of Doubtful Value

Nystatin
Thiabendazole
Thiomersal
Naftate
Flucytosine
Econazole
Many others
TABLE 51.4. Antifungal Agents of Proven Value

POLYENES AZOLES
Natamycin Fluconazole
Amphotericin B Itraconazole
Ketoconazole
Miconazole
TREATMENT
SECTION 6
Available Antifungal Agents
718
Although a multitude of antifungal agents are available for

treating fungal infections in the cornea, many are of doubtful
value (Table 51.3). These agents lack published evidence of
efficacy. They will not be considered further. There are,
however, a number of agents of proven value including the
polyenes, natamycin and amphotericin B, the imidazoles;
miconazole, clotrimazole and ketoconazole and the azoles;
fluconazole and itraconazole (Table 51.4).
Polyenes act by binding to the sterol moiety (ergosterol) in the
cell wall of susceptible fungi.22 This interaction forms pores,
increasing the permeability of the cell wall and leading to
leakage of small molecules. Amphotericin B, the oldest effective
antifungal agent, is particularly useful in the treatment of
Candida sp. infections, against which it exhibits a fungicidal
action. Amphotericin B has considerable toxicity when
administered systemically. The systemic preparation is used to
formulate the medication for topical use. When used in a
concentration of 0.15% it is both efficacious and well
tolerated.22,23 The principal application of Amphotericin B in
the cornea is for yeast infections. Natamycin, also a polyene
and the only ophthalmic preparation specifically developed for
treating oculomycoses, is most effective against filamentous
fungi. It has been particularly valuable in treating infections
caused by Fusarium and Aspergillus spp. Natamycin (Natacin)
is only available as a 5% suspension.23
The remainder of the agents with proven efficacy belong to
the azole group of compounds. Fifty years ago the antifungal
activity of these azoles was first discovered. Since then,
successive generations have shown progressive improvement in
both spectrum of activity and pharmacologic characteristics.
They act by inhibiting ergosterol synthesis in the cell wall.24
The earlier azoles were difficult to solubilize for ophthalmic use
and penetrated the cornea poorly. Miconazole, an imidazole,
was the first azole compound to demonstrate clinical efficacy.
Even though its phamacokinetic profile was quite poor, it was
found to be useful in treating yeast infections. Clotrimazole,
another imidazole with a pharmacokinetic profile similar to
miconazole, has been used topically with success in the
treatment of filamentous fungal infections as well as those
caused by yeasts.22 Ketoconazole, yet another imidazole,
showed a slightly broader spectrum of activity with clinical case
reports of efficacy against Aspergillus, Fusarium, and Curvularia
Chlotrimazole
ocular infections.25 Its principal ophthalmic use was

systemically to augment topical therapy.
Based on the experience with imidazoles a new group of
compounds, the triazoles, has been developed. The triazoles
have three nitrogens in the azole ring. Triazoles, like the
imidazoles, inhibit the sterol 14a-demethylase (cytochrome
P450-dependent enzyme system). This impairs ergosterol
synthesis, leading to accumulation of methyl sterols that impair
the function of membrane-bound enzyme systems, thus
inhibiting fungal growth. As a class, the triazoles exhibit
considerably improved antifungal activity and have virtually
replaced the imidazoles. Fluconazole (Diflucan) is the first
antifungal agent with a good phamacokinetic profile and a low
incidence of systemic side effects. It is water-soluble and
exhibits low protein binding.24 Uptake and persistence in all
ocular tissues is excellent when administered systemically.24
The systemic preparation can be administered topically as
a 2 mg/mL drop. Fluconazole is most active against yeasts.
Another triazole, itraconazole (Sporanox), by contrast has very
different pharmacological characteristics, being poorly soluble
in water and exhibiting high protein binding.23 Nevertheless, it
has proven efficacy not only against yeasts but also against
certain filamentous fungi and in particular Aspergillus sp.24 For
corneal infections, its principal use has been systemically to
augment topical therapy. One study in animals suggests a
possible role via the subconjunctival route.26
A third group of agents are those with potential value (Table
51.5). They include chlorhexidine gluconate, a steadily growing
number of triazole derivatives, among them voriconazole,
posaconazole, and ravuconazole and the echinocandin,
caspofungin. Chlorhexidine gluconate, as a 0.2% topical
preparation, has been shown to have efficacy against Aspergillus
and Fusarium species in a randomized controlled trial reported
in 1998.27 The triazole derivatives are just now being
introduced. Voriconazole, with which there is greatest
experience to date, is a structural derivative of fluconazole but
appears to be more effective in inhibiting ergosterol synthesis.28
The oral bioavailability of voriconazole is ~96% while protein
binding is 58%.29 It can be administered topically, systemically,
by intraocular injection and by anterior chamber lavage. Eye
Fungal Keratitis
Chlorhexidine gluconate 0.2%

Triazole derivatives
Voriconazole (FDA approved 2002)
Posaconazole
Ravuconazole
Pneumocandins
Caspofungin
drops (1%) can be prepared by diluting the IV preparation in

distilled water. They deliver high concentrations of the drug
to the cornea.30,31 Based on recent case reports, voriconazole
appears to have a remarkably broad spectrum of activity, which
includes against Candida, Fusarium, Aspergillus, Curvularia,
Paecelomyces, and Scedosporium apiospermum.2933
Posaconazole, ravuconazole, and caspofungin are promising
antifungal agents under study. However, experience is too
limited at this stage to determine their ultimate role in the
treatment of corneal infection.34 The promising experience to
date with voriconazole and the continuing development of new
members of the triazoles offer hope of considerably more
effective antifungal agents in the future. Although not
specifically developed for ophthalmic use, they are likely to be
modified successfully as others have been in the past.
Routes of Administration
Three potential routes of administration of antifungal agents
have been advocated for the treatment of corneal infections;
topical, subconjunctival, and systemic. For all agents so far
developed, the topical route is preferred. Studies have shown
that most current topical antifungal agents share similar
characteristics.24 Since, with the sole exception of natamycin,
they have been derived from preparations intended for systemic
use, the pharmacokinetic profile has been a secondary
consideration. Achieving a topical preparation for these agents
is often a challenge. The corneal epithelium tends to be a
barrier to penetration and bioavailability tends to be medium to
low at best.35 Despite these obstacles therapeutic levels can be
achieved with intensive topical administration.36 Natacin is a
special case. Although formulated specifically for the eye, its
profile is poor. Also, the maximum achievable concentration of
a 5% suspension is necessary to provide therapeutic corneal
levels.36
From time to time, the subconjunctival route for
administration has been advocated. However, animal studies
have shown that the pharmacokinetics by this route is almost
universally quite poor with either no penetration into the
cornea or rapid corneal clearance with only transient levels of
drug. Only itraconazole showed evidence of significant
penetration into the cornea and even then a minimum of twice
a day injections would seem to be required in order to maintain
adequate levels.26 However, a recent report suggests clinical
efficacy with subconjunctival fluconazole.37 Several of these
agents (particularly amphotericin B and natamycin) exhibit
considerable toxicity when injected subconjunctivally.
Systemic therapy generally has little to add to the treatment
of fungal infection in the cornea except when the infection is
deep seated and intraocular penetration of the fungus is
suspected. In such cases, the newer azole compounds seem to
offer the best hope of significant efficacy. Fluconazole is
relatively nontoxic when given systemically and has been
shown to provide excellent corneal penetration.38 Itraconazole,

likewise has low toxicity. Although specific data are lacking,
it is believed to have good corneal penetration. It also has a
broader spectrum of activity than fluconazole. Voriconazole, has
also been shown to have good levels in ocular tissues following
systemic administration. Based on experience with fluconazole,
it is reasonable to expect therapeutic levels to be achievable
also, although concentrations in the cornea have yet to be
reported.29 The older imidazole compounds seem to have only
a modest therapeutic benefit at best, when given systemically in
corneal disease. Amphotericin B administered systemically does
not provide therapeutic levels in the cornea.22
As a principle, the topical route is preferred for all agents.
There is some clinical and experimental evidence that the
subconjunctival route may have some value with itraconazole
and fluconazole. For systemic therapy, fluconazole,
itraconazole, and voriconazole appear to be the only agents
likely to be of value.
Management of Filamentous Fungi Infections

As noted earlier, the most common isolates in filamentous
fungal infections belong to Fusarium and Aspergillus species.
Among the remainder, Curvularia species are the next most
definable group. All three usually respond well to natamycin.
Initially, the identity of the fungus though recognizable as
filamentous on smear or culture may not be known. There may
be a considerable delay until spores, necessary for full
identification, appear. Fortunately there is strong evidence to
support the initial use of natamycin regardless of the type of
filamentous fungus isolated.22 Experience and some
experimental data support the administration of natamycin
5% suspension every hour for the first 48 h to establish a
loading dose in the cornea.36 The frequency can then be reduced
to every hour during waking hours while the cornea is observed
closely for evidence of resolution. Even when the infecting
fungus is susceptible the response is likely to be slow. Careful
documentation of the clinical findings is essential, aided by slitlamp photography if available.
Subconjunctival itraconazole (100 mg) given at least once
per day may be considered especially for Aspergillus species
infections with deep corneal involvement, but its efficacy has
not been established in clinical cases. Of greater value and
especially when anterior chamber invasion is suspected, is the
administration of a systemic triazole. Until recently,
itraconazole and fluconazole have been the agents most likely to
be recommended. However, recent experience with clinical
cases suggests that voriconazole may be more efficacious
against a widening spectrum of filamentous organisms in such
cases than either itraconazole or fluconazole.
Identification of the fungus places treatment on firmer
grounds. Prior experience can now serve as a guide to therapy.
Susceptibility testing is of uncertain value and is not routinely
recommended. If the infection does not appear to be responding
to natamycin or is worsening, there are now several options.
Topical 1% itraconazole has been reported to be effective against
a broad range of filamentous fungi.39 There is also evidence for
the value of topical chlorhexidine gluconate.27 Most promising,
however are the reports of success with topical voriconazole.
While it is still too early to define with certainty the appropriate
indications, topical voriconazole should be strongly considered
in the face of apparent resistance to natamycin.
CHAPTER 51
TABLE 51.5. Antifungal Agents of Potential Value
Management of Yeast Infections

Candida species, almost always C. albicans, are the
predominant cause of yeast infections in the cornea.
Amphotericin B 0.15% is fungicidal against Candida albicans
at concentrations achievable in the cornea by topical
719

administration. It is unusual to find strains resistant to
amphotericin B. While Candida species are susceptible to
natamycin and the azoles, these agents should be reserved
as second-line treatment. The evidence suggests that
amphotericin B is therapeutically superior, probably on account
of its fungicidal activity.2224
Initially amphotericin B 0.15% should be administered
topically every hour for at least 24 h after which the rate can be
reduced to the waking hours while the cornea is observed
closely. In those rare instances where deep invasion in the
cornea is occurring despite treatment, or anterior chamber
involvement is suspected, systemic therapy should be strongly
considered. Fluconazole is efficacious against Candida species
and does provide high levels in the cornea and anterior chamber
when given systemically or by mouth. Voriconazole also has
similar characteristics and should be considered. However, it is
important to emphasize that the use of systemic agents is rarely
needed in the treatment of yeast infections.
Duration of Treatment
Fungal infections in the cornea tend to develop slowly and
likewise to regress slowly. A prolonged period of treatment, in
the region of 46 weeks, is usually needed in the best
circumstances and in some instances an even longer period is
necessary. Unfortunately, susceptibility testing is not a strong
guide to therapeutic efficacy with antifungal agents, so clinical
acumen combined with careful observation is needed to
determine when the infection is coming under control. After the
initial period of intensive topical therapy the rate can be reduced
progressively to four or five times per day. This rate should be
continued until clear evidence of resolution is seen. Once the
infection has resolved, the agent can be discontinued without
further tapering of the dose.
SECTION 6
Surgical Management
720
Until very recently, therapeutic keratoplasty during the acute

phase of the infection was considered a practical alternative to
medical therapy. Several series have reported remarkable cure
rates.4048 However, with the introduction of new more effective
and broad-spectrum antifungal agents, surgical intervention in
the form of a therapeutic keratoplasty during the acute stage of
infection is much less likely to be necessary. Keratoplasty is now
most frequently performed for visual restoration after the
infection has resolved. There are, however, a number of
circumstances where therapeutic keratoplasty is indicated
during the acute phase of the infection. These include
unavailability of effective antifungal therapy, progression of the
infection despite pharmacologic treatment, impending or actual
perforation, and progression of the infection to involve the
limbus and adjacent sclera (Fig. 51.7). Fortunately, corneal
perforation is an unusual event during the initial presentation
of a fungal infection in the cornea unless the injury has
penetrated into the anterior chamber or a significant delay has
occurred between initiation of the infection and presentation for
treatment. Even in the face of a corneal perforation, it may be
possible to delay penetrating keratoplasty by placing a
therapeutic contact lens and applying glue in order to continue
medical therapy. An important indication for therapeutic
keratoplasty is progression of the corneal lesion toward the
limbus or evidence of scleral invasion. The latent period
between invasion of the corneal stroma with fungi and the
appearance of inflammation can be quite long so that fungal
hyphae may already have invaded apparently uninvolved
cornea. For this reason when determining the size of the graft
for a therapeutic keratoplasty in an active fungal infection, a
generous free margin of apparently normal cornea is desirable.
Performing a keratoplasty before the limbus is involved can
FIGURE 51.7. Fungal keratitis secondary to infection with

Cladosporium species adjacent to the limbus.
avoid the need for a corneoscleral graft, which carries a much

less optimistic prognosis.
Fungi may gain access to the anterior chamber and anterior
ocular structures by direct spread from the corneal lesion or by
implantation following a penetrating injury. Fortunately,
anterior chamber invasion is not common but when it does
occur the prognosis worsens considerably. This is particularly
true of infection with Fusarium solani and Aspergillus species.
Recognition of intraocular spread can be quite difficult.
Indications that intraocular invasion may be occurring include
worsening pain, an increase in the size of hypopyon, the
development of endothelial plaques, an increase in the intensity
of ciliary flush, the presence of iris nodules, and the
development of fungal glaucoma.
Fungal invasion of the anterior chamber is a very serious
development. Until fairly recently the only realistic surgical
approach was a large penetrating keratoplasty to excise visible
fungal material, combined with irrigation of the anterior
chamber with amphotericin B, and, if the lens is involved, an
extracapsular procedure to preserve the posterior capsule. Now,
with better medical therapeutic options available as a result of
the development of the newer triazoles, a nonsurgical approach
is feasible at least in the initial stages.
The management of therapeutic penetrating keratoplasty for
fungal keratitis is complicated by the desire to avoid the use of
corticosteroids if at all possible because of their deleterious
effect on fungal growth and in promoting fungal tissue
invasion.49 Fortunately, topical cyclosporine can be used in
place of corticosteroids and has been effective in managing
grafts postoperatively. However, it is prudent to remember that
graft failure may be a trade off for eradicating the fungal
infection so that a repeat penetrating keratoplasty may be
required.
SUMMARY
The outlook for fungal infections in the cornea has improved
immeasurably in the past few years with the introduction
and more widespread use of effective antifungal therapy.
Nevertheless, in developing countries where the disease is
Fungal Keratitis
most prevalent, the incidence of corneal blindness as a result
of fungal keratitis remains high. For the foreseeable future
treatment of fungal keratitis will have to take advantage of
developments with systemic agents since there is no evidence of

any interest on the part of drug companies in developing agents
specifically for ophthalmic use.
1. Bharathi MJ, Ramakrishnan R, Vasu S, et

al: Epidemiological characteristics and
laboratory diagnosis of fungal keratitis. A
three-year study. Indian J Ophthalmol 2003;
51:315321.
2. Leck AK, Thomas PA, Hagan M, et al:
Aetiology of suppurative corneal ulcers in
Ghana and south India, and epidemiology
of fungal keratitis. Br J Ophthalmol 2002;
86:12111215.
3. Thomas PA, Leck AK, Myatt M:
Characteristic clinical features as an aid to
the diagnosis of suppurative keratitis
caused by filamentous fungi. Br J
Ophthalmol 2005; 89:15541558.
4. Gopinathan U, Garg P, Fernandes M, et al:
The epidemiological features and laboratory
results of fungal keratitis: a 10-year review
at a referral eye care center in South India.
Cornea 2002; 21:555559.
5. Srinivasan M, Gonzales CA, George C, et
al: Epidemiology and aetiological diagnosis
of corneal ulceration in Madurai, south
India. Br J Ophthalmol 1997; 81:965971.
6. Gonzales CA, Srinivasan M, Whitcher JP,
Smolin G: Incidence of corneal ulceration in
Madurai district, South India. Ophthalmic
Epidemiol 1996; 3:159166.
7. Whitcher JP, Srinivasan M: Corneal
ulceration in the developing
world a silent epidemic. Br J Ophthalmol
1997; 81:622623.
8. ODay DM: Selection of appropriate
antifungal therapy. Cornea 1987;
6:238245.
9. Wilhelmus KR, Jones DB: Curvularia
keratitis. Trans Am Ophthalmol Soc 2001;
99:111130.
10. Wilhelmus KR: Climatology of
dematiaceous fungal keratitis. Am J
Ophthalmol 2005; 140:11561157.
11. Mselle J: Fungal keratitis as an indicator of
HIV infection in Africa. Trop Doct 1999;
29:133135.
12. Bourcier T, Touzeau O, Thomas F, et al:
Candida parapsilosis keratitis. Cornea
2003; 22:5155.
13. ODay DM, Head WS, Robinson RD, et al:
Contact lens-induced infection a new
model of Candida albicans keratitis. Invest
14. Parrish CM, ODay DM, Hoyle TC:
Spontaneous fungal corneal ulcer as an
ocular manifestation of AIDS. Am J
Ophthalmol 1987; 104:302303.
15. Thomas PA, Kalavathy CM, Abraham DJ:
Oral ketoconazole in keratomycosis. Indian
J Ophthalmol 1987; 35:197203.
16. Rajasekaran J, Thomas PA, Kalavathy CM,
et al: Itraconazole therapy for fungal
keratitis. Indian J Ophthalmol 1987;
35:157160.
17. Thomas PA, Abraham DJ, Kalavathy CM:
Oral itraconazole therapy for mycotic
keratitis. Mycoses 1988; 31:271279.
18. Tanure MA, Cohen EJ, Sudesh S, et al:
Spectrum of fungal keratitis at Wills Eye
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
Hospital, Philadelphia, Pennsylvania.

Cornea 2000; 19:307312.
Maichuk IuF, Diadina UV: Itraconazole in the
treatment of ophthalmomycoses. Antibiot
Khimioter 1994; 39:5456; [Russian].
ThakarM: Oral fluconazole therapy for
keratomycosis. Acta Ophthalmol (Copenh)
1994; 72:765767.
Kumar M, Mishra NK, Shukla PK: Sensitive
and rapid polymerase chain reaction based
diagnosis of mycotic keratitis through single
stranded conformation polymorphism. Am J
Ophthalmol 2005; 140:851857.
ODay DM: Fungal Keratitis. In: Wilhelmus
K, ed. Ocular infection and immunology.
Philadelphia, PA: Mosby;1993.
ODay DM, Head WS: Advances in the
management of keratomycosis and
acanthamoeba keratitis. Cornea 2000;
19:681687.
ODay DM, Head WS: Ocular pharmacology
of antifungal drugs. In: Tasman W, Jaeger
EA, eds. Duanes foundation of clinical
ophthalmology. Philadelphia, PA:
Lippincott-Raven; 1997.
Johns KJ, ODay DM: Pharmacologic
management of keratomycoses. Surv
Ophthalmol 1988; 33:178188.
Klippenstein K, Robinson RD, Williams TE,
et al: The qualitative evaluation of the
pharmacokinetics of subconjunctivally
injected antifungal agents in rabbits.
Cornea 1993; 12:512516.
Rahman MR, Gordon JJ, Rabiul H, et al:
Randomised trial of 0.2% chlorhexidine
gluconate and 2.5% natamycin for fungal
keratitis in Bangladesh. Br J Ophthalmol
1998; 82:919925.
Sanati H, Belanger P, Fratti R, Ghannoum
M: The effect of the new triazole,
voriconazole (UK-109,496), on the
interactions of Candida albicans and
Candida krusei with endothelial cells. J
Chemother 1998; 10:716.
Hariprasad SM, Mieler WF, Holt ER, et al:
Determination of vitreous, aqueous, and
plasma concentration of orally administered
voriconazole in humans. Arch Ophthalmol
2004; 122:4247.
Sponsel W, Chen N, Dang D, et al: Topical
voriconazole as a novel treatment for fungal
keratitis. Antimicrob Agents Chemother
2006; 50:262268.
Prats CH, Tello PL, San Jose AB, et al:
Voriconazole in fungal keratitis caused by
scedosporium apiospermum. Ann
Pharmacother 2004; 38:414417; Epub
2004 Jan 30.
Durand ML, Kim IK, DAmico DJ, et al:
Successful treatment of Fusarium
aspergillus endophthalmitis with
voriconazole plus caspofungin. Am J
Ophthalmol 2005; 140:552554.
Anderson KL, Mitra S, Salouti R, et al:
Fungal keratitis caused by Paecilomyces
lilacinus associated with a retained
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
intracorneal hair. Cornea 2004;

23:516521.
rabbit model. Antimicrob Agents Chemother
2005; 49:13591363.
ODay DM, Ray WA, Head WS, Robinson
RD: Influence of the corneal epithelium on
the efficacy of topical antifungal agents.
25:855859.
ODay DM, Head WS, Robinson RD, et al:
Corneal penetration of topical amphotericin
B and natamycin. Curr Eye Res 1986;
5:877882.
Yilmaz S, Maden A: Severe fungal keratitis
treated with subconjunctival fluconazole.
Am J Ophthalmol 2005; 140:454458.
ODay DM, Foulds G, Williams TE, et al:
Ocular uptake of fluconazole following oral
administration. Arch Ophthalmol 1990;
108:10061008.
Kalavathy CM, Parmar P, Kaliamurthy J, et
al: Comparison of topical itraconazole 1%
with topical natamycin 5% for the treatment
of filamentous fungal keratitis. Cornea
2005; 24:449452.
Jones BR, Jones DB, Richards AB: Surgery
in the management of keratomycosis. Trans
Ophthalmol Soc U K 1970; 89:887897.
Polack FM, Kaufman HE, Newmark E:
Keratomycosis. Medical and surgical
treatment. Arch Ophthalmol 1971;
85:410416.
Singh G, Malik SR: Therapeutic
keratoplasty in fungal corneal ulcers. Br J
Polack FM: Diagnosis and treatment of
keratomycosis. Int Ophthalmol Clin 1973;
13:7591.
Singh G, Malik SR, Bhatnagar PK:
Therapeutic value of keratoplasty in
keratomycosis. An experimental study. Arch
Forster RK, Rebell G: Therapeutic surgery in
failures of medical treatment for fungal
keratitis. Br J Ophthalmol 1975;
59:366371.
Sanitato JJ, Kelley CG, Kaufman HE:
Surgical management of peripheral fungal
keratitis (keratomycosis). Arch Ophthalmol
1984; 102:15061509.
Wong TY, Fong KS, Tan DT: Clinical and
microbial spectrum of fungal keratitis in
Singapore: a 5-year retrospective study. Int
Ophthalmol 1997; 21:127130.
Rao GN, Garg P, Sridhar MS: Penetrating
keratoplasty in infectious keratitis. In:
Brightbill FS, ed. Corneal surgery: theory,
technique & tissue. St Louis, MO: Mosby;
1999:518525.
ODay DM, Ray WA, Robinson RD, et al:
Efficacy of antifungal agents in the cornea.
II. Influence of corticosteroids. Invest
CHAPTER 51
REFERENCES
721
CHAPTER
52
Acanthamoeba Keratitis
Hassan Alizadeh, Jerry Y. Niederkorn, and James P. McCulley
Overview
The most important aspect in the management of Acanthamoeba
keratitis is prevention. The frequent association between the
disease and homemade saline prepared from distilled water and
salt tablets have led several companies to withdraw salt tablets
from the market voluntarily. Therefore, it is recommended that
contact lenses should not be stored or cleaned in sterilized saline
solution. The care of contact lenses is also important for
preventing the disease. Commercially available heat sterilization
units are effective methods for sterilizing contact lenses and
result in killing both Acanthamoeba trophozoites and cysts.
In this case, the patient should be advised not to contaminate
the lenses after heat sterilization and before reinsertion onto the
eye. Moreover, it has been shown that exposure to hydrogen
peroxide in its active state for 2 h is an effective method for
killing both trophozoites and cysts. Deactivation of hydrogen
peroxide with deactivating agents, however, should be used after
the lens has been in hydrogen peroxide for more than 2 h.
Preservatives that are found in cold sterilization solutions are also
effective in killing the organisms. However, 4-h exposure to these
solutions is necessary to kill both trophozoites and cyst.
Early diagnosis is invaluable and is associated with a relatively
satisfactory outcome, while the consequences of late diagnosis
can be devastating.
Acanthamoeba keratitis is a vision-threatening infection caused

by a free-living, pathogenic amoeba. Corneal infection was
first recognized in 1973, and the main risk factor was found
to be contact lens wear. Characteristic symptoms include
disproportionately severe ocular pain, a paracentral ring-like
stromal infiltrate, epithelial ulcers, and resistance to many
antimicrobial agents. Typical treatment consists of around-theclock, hourly, topical applications of chlorhexidine, Brolene, or
polyhexamethylene biguanide (PHMB), alone or in
combination. This regimented therapy may continue for weeks
and slowly taper to maintenance therapy for a year or more.
Early diagnosis is invaluable and is associated with a relatively
satisfactory outcome, while the consequences of late diagnosis
can be devastating.
isoenzyme patterns also change when strains are cultured under

different conditions.2,3 During the late 1990s, attempts to
rectify the problems in classification focused on nuclear rRNA.
Under this system, Acanthamoeba species were divided into
12 sequence types, and in combination with morphological
features, helped to create the most current phylogenetic tree.4,5
BIOLOGY
Acanthamoeba spp. are ubiquitous in nature and have been
isolated from a wide variety of environments.613 Acanthamoebae exist in two stages; the motile free-living trophozoite
and the dormant cyst (Fig. 52.1). The trophozoites of at least
seven species of Acanthamoeba are the causative agents of
disease and range from 1025 mm in size. They are easily
identified at the genus level due to the presence of spiny surface
pseudopodia, called acanthapodia, from which their name is
derived.
In unfavorable conditions, such as starvation, hyperosmolarity, desiccation, freezing, and extreme temperatures
or extreme pH, the trophozoite will encyst.1417 The cyst is
812 mm in diameter and is encapsulated within a double
wall that is primarily comprised of cellulose and proteins.
The cyst form is extraordinarily resistant to extreme conditions, such as alterations in osmolarity, pH, desiccation,
freezing, or chemical antimicrobial agents.15,17 More recently
it has been reported that Acanthamoeba castellanii cysts
withstood heating up to 65C for 5 min and produced viable,
proliferating trophozoites. Moreover, A. castellanii cysts
survived five cycles of freezethawing and were resistant
to both 250 krads of g irradiation and 800 mJ/cm2 of UVB
irradiation and very high doses of UV and gamma radiation.17,18
Such resistance is beneficial for the organism as it enhances
CLASSIFICATION
Acanthamoeba was first discovered in 1930 by Castellani while
examining cultures of the yeast Cryptococcus pararoseus. In
1991, Visvesvara created a new species classification scheme
dividing Acanthamoeba spp. into three different subgroups
based on morphology, isoenzyme analysis, and serology.1 This
scheme is widely used today and places Acanthamoeba into the
suborder Acanthopodina and the genus Acanthamoeba.
However, many species share antigenic determinants, and
Trophozoites
Cysts
FIGURE 52.1. Photomicrographs of Acanthamoeba castellanii

trophozoites and cysts. Photos were taken at 100. Bar = 12 mm.
723

FIGURE 52.2. Clinical photographs showing
early stages of Acanthamoeba keratitis: (ac)
showing diffuse epithelial lesions,
(d) showing diffuse subepithelial lesions.
Parmar ND, Awwad ST, Petroll WM, et al: Tandem
scanning confocal corneal microscopy in the diagnosis
of suspected Acanthamoeba keratitis. Ophthalmology
(in press).
its survival during periods of very harsh environmental

conditions, which might include immunological attack.
SECTION 6
EPIDEMIOLOGY
At least eight species of Acanthamoeba have been reported to
cause keratitis: A. castellanii, A. polyphaga, A. hatchetti, A.
culbertsoni, A. rhysodes, A. lugdunesis, A. quina, and A.
griffini.18,19 The first case of Acanthamoeba keratitis was
described by Jones in 1975.19,20 However, only 10 cases of
Acanthamoeba keratitis were reported between 1975 and
1981.19,21 Since 1981, the number of Acanthamoeba keratitis
cases has increased gradually, and more than 100 cases were
reported during the late 1980s in the United States. In the
United Kingdom, ~400 cases of Acanthamoeba keratitis have
been diagnosed since 1977.19,22
The true incidence of Acanthamoeba keratitis is not known.
However, more than 750 cases of Acanthamoeba keratitis
have been diagnosed worldwide.19,21 Since the disease is closely
associated with contact lens wear, it is estimated that 1.652.01
cases per million contact lens wearers occurred between 1985
and 1987.19 However, Mathers et al19,23 suggested that the
incidence of Acanthamoeba keratitis might be as high as 1 per
10 000 contact lens wearers per year.
Key Features: Potential Risk Factors in Acanthamoeba
Keratitis
Damage to the corneal epithelium

Contact lenses
Lack or low level of anti-Acanthamoeba secretory IgA in the
tear
Contaminated water or solutions
Host susceptibility?
Defects in host defense mechanisms?
CLINICAL CHARACTERISTICS
724
More recently a total of 83 patients (85 eyes) with

Acanthamoeba keratitis have been reviewed by Hargrave
et al.24 The average duration of disease at entry into the trial
was 68 days. Forty-one patients had a history of right eye
involvement, 40 had their left eye affected, and two had

bilateral disease. The major clinical symptoms were severe
or moderate pain, decreased vision, redness, irritation, foreign
body sensation, photophobia, mucous discharge, and tearing.
The overwhelming majority of patients (95%) had ocular pain.
The most common predisposing condition was the use of
contact lenses (75%). One of the specific and important
symptoms of Acanthamoeba keratitis is the severity of the
pain in the early stage of infection. This symptom nonetheless
is atypical as compared with that of other corneal infections
such as herpes simplex keratitis.
Acanthamoeba keratitis can have many appearances, but the
signs of disease at an early stage (12 weeks) include diffuse
epithelial lesions (Fig. 52.2ac). Moreover, the occurrences of
diffuse subepithelial lesions indicate early stage of
Acanthamoeba keratitis (Fig. 52.2d).
Patients presenting with corneal epithelial abnormalities
exhibited a broad spectrum of pathologic findings. These
include epithelial haze with elevated lines, epithelial defects,
microcysts, pseudodendrites, and punctuate epithelial erosions
(Fig. 52.3a). Subsequently, the patchy stromal infiltrates extend
and form crescents or ring infiltrates (Fig. 52.3b). The ring
initially located in the central cornea, becomes more
circumscribed and dense at later stages of infection. Stromal
infiltrates other than ring infiltrates included large patchy
anterior stromal infiltrates, satellite lesions, and rarely, stromal
ulceration with melting. These clinical signs appeared at
early or moderately advanced stages of infections. As the
disease progresses, however, there may be a progressive loss
of the corneal stroma with the formation of descemetocele
and possible perforation.
Stromal involvement before ring abscess development varies
tremendously. The most characteristic form of stromal disease
occurs late and is recognizable by a ring infiltrate. Recurrent
epithelial breakdown, overlying ring infiltrates, and abscesses
lead to frequent diagnoses of herpes simplex keratitis. These
manifestations are indistinguishable from the ring abscesses
caused by herpes simplex virus. Indeed, many previously
misdiagnoses of herpes ring abscesses probably were caused
by Acanthamoeba. The cause of dense ring infiltrates in the
corneas of patients with Acanthamoeba keratitis is not
known. It has been suggested that ring infiltrates result from
FIGURE 52.3. Clinical photographs showing diverse spectrum of Acanthamoeba keratitis: (a) segmental subepithelial involvement, (b) classical
ring infiltrate, and (c) radial keratoneuritis.
the release of various proteolytic enzymes by infiltrating

neutrophils that cause collagenolysis of the stroma. However,
it has been reported that elaboration of collagenolytic enzymes
by Acanthamoeba trophozoites not only degrades collagen
in vitro but also produces ring infiltrates in the corneas of rats,
which mimic Acanthamoeba keratitis in humans.25 A not
uncommonly encountered sign is that of infiltration along the
nerves by the organisms, creating what appears to be radial
keratoneuritis (Fig. 52.3c). This finding probably accounts
for the severe pain that is characteristic of Acanthamoeba
keratitis. In vitro studies indicated that Acanthamoeba
trophozoites are chemotaxically more attracted to the
endothelium that arises from the neural crest than the
epithelium.
Visual acuity varies among individuals, ranging from a visual
acuity of 20/20 to counting fingers and no light perception.
These variations depend on stages of infection, duration of
corneal involvement, timing of medical treatment, and
susceptibility of Acanthamoeba to drug treatments. In many
cases of successful treatment, however, a corrected visual
acuity (20/20) was achieved after either antiamoeba treatment
or keratoplasty.26,27 More recently Parmar et al28 have studied
63 cases of Acanthamoeba keratitis and concluded that the
earlier diagnosis of Acanthamoeba keratitis was a key factor
and contributed to better visual outcome.
Key Features: Clinical Characteristics of Acanthamoeba
Keratitis
Severe ocular pain

Epithelial irregularity and dendriform pattern
Severe anterior and posterior scleritis
Diffuse or nodular scleritis
Stromal infiltrates (single, multiple, and ring shape)
Anterior uveitis (transient hypopyon)
Variable, persistent, or recurrent epithelial erosion
Satellite lesion
Radial keratoneuritis
PATHOGENESIS
The pathogenic cascade of Acanthamoeba keratitis involves
a series of processes that include: (1) binding of the trophozoites
to the corneal epithelial cells via lectinglycoprotein
interactions,2933 (2) generation of cytopathic factors that
destroy the corneal epithelium and stromal cells,34,35 (3) production of proteolytic enzymes that facilitate the invasion and
penetration of trophozoites through the basement membrane and
stroma,34,3639 (4) elaboration of collagenolytic enzymes that
degrade types I and IV collagens, which constitute the
corneal matrix,25,38 (5) activation of corneal membrane
metalloproteinases, and (6) induction of radial neuritis.
There are only a few isolated reports of Acanthamoeba

trophozoites penetrating the endothelial cell layer and entering
the interior of the eye.20,40,41 More recently, it has been
shown that A. castellanii is capable of penetrating Descemets
membrane and entering the anterior chamber of Chinese
hamster eye. However, the aqueous humor contains factors
that either induce encystment or kill Acanthamoeba
trophozoites.42
Once inside the stroma, trophozoites incite an inflammatory response that is comprised of macrophages and
neutrophils. In vitro studies have shown that macrophages are
capable of killing Acanthamoeba trophozoites in an antibodydependent mechanism.43 Depletion of conjunctival macrophages with clodronate-encapsulated liposomes prior to
infection with A. castellanii produces a severe infection that
does not resolve.44 This finding suggests that macrophages are
important for limiting the severity and duration of disease.
However, they do not prevent the initial corneal infection.29
This is consistent with in vitro studies, which have
demonstrated that macrophage activation is key for successful
elimination of the trophozoites.45 As a result, resting
macrophages are unable to effectively kill trophozoites, and
in fact, succumb to trophozoite-mediated killing.
Neutrophils have also been shown to be effective at killing
Acanthamoeba trophozoites and are very abundant in corneas
afflicted with Acanthamoeba keratitis.43,46,47
Presently, there are no known antimicrobials for eliminating
cysts from the eye. Moreover, the use of dexamethasone, a
topical steroid typically used to control pain and inflammation
after eye surgery, has been shown to induce conversion of
the dormant cysts to infectious trophozoites.48 Further
compounding the problem is the fact that cysts appear to
evade elimination by the immune system. It is unclear what
role such antibodies might have in human patients with
Acanthamoeba keratitis. Once in the stroma, the cysts must
then evade the macrophages and neutrophils that respond
to the infection.49 It has been shown that anti-Acanthamoeba
IgA antibodies specific for molecules on the surface of
trophozoites and present in the tears of orally immunized
animals inhibit adhesion of the parasites to the corneal
epithelium and prevent the crucial first step in the development
of corneal infection.34,5052 Acanthamoeba trophozoites express
the mannose-binding receptor, which facilitates adhesion of
the parasite to mannosylated proteins on corneal epithelial
cells.53 It has been shown that free mannose strongly inhibits
the binding of Acanthamoeba trophozoites to the corneal
epithelial and stromal cells.33 Moreover, the presence of
mannose in the medium inhibits parasite-mediated cytolysis
of corneal cells in short-term in vitro assays.32,33 However,
it has been shown that trophozoites exposed to free mannose
for 48 h or longer are induced to release a soluble 133-kDa
cytolytic factor that mediates contact-independent cytolysis
CHAPTER 52
Parmar ND, Awwad ST, Petroll WM, et al: Tandem scanning confocal corneal microscopy in the diagnosis of suspected Acanthamoeba keratitis. Ophthalmology (in press).
725

of corneal epithelial cells in vitro.33,35 Therefore, the parasites
binding to the mannose receptors induces the generation of
cytopathic factors that destroy the corneal epithelial and
stromal cells and is an important step in the pathogenicity of
Acanthamoeba keratitis.
Our findings with Acanthamoeba keratitis suggest that if
the mucosal immune system is preferentially activated and
is directed against pathogen-derived proteases that contribute
to the pathogenesis of corneal disease, it may be possible to
mitigate corneal disease even if the microorganism is resistant
to immune elimination or is capable of evading the immune
response. This concept is not limited to Acanthamoeba
keratitis and may be applicable to other pathogens that
elaborate bacterial toxins or pathogenic proteases.
LABORATORY DIAGNOSIS
SECTION 6
It is important to obtain a laboratory diagnosis for

Acanthamoeba as early as possible, because the therapy is most
effective when initiated early and requires prolonged treatment.
Scraping and biopsy can be performed if epithelial or
subepithelial abnormality is observed. Scraping of the cornea
with a sterile scalpel should be performed under local
anesthesia, and the scraping materials can be used either for
slide preparation or culture in a nonnutrient agar plate as
described previously.54
Culturing the specimen on a confluent lawn of Escherichia
coli plated on nonnutrient agar (1.5%) is another diagnostic
technique. Enteric gram-negative bacteria, such as E. coli or
Enterobacter species, are a food source for Acanthamoeba.54
Trophozoites can be identified under the microscope for the
presence of contractile vacuoles, which disappear and reappear
quickly.
If materials for culturing the parasite are not available, tissue
specimens, corneal smears, contact lenses, and swabs may be
kept in Pages saline and sent to the laboratory. Corneal
scrapings, contact lenses, and filtered contact lens solution
can be cultured on nonnutrient agar plates at 35C for
710 days. A longer period is necessary to culture cysts to
transform to the trophozoite stage. For wet mount preparation,
contact lens solution is best centrifuged at 250 g, and the
sediment transferred to a slide and covered with a coverslip.
The slide should be kept in a covered Petri dish and examined
by phase microscopy.
Acanthamoeba trophozoites and cysts can be identified
in corneal scrapings or smears by staining with Gram,
GiemsaWright, or Wheatly trichrome stains. Trophozoites and

cysts stain purple with GiemsaWright staining. The
trophozoite is characterized by a large single nucleus and
spindle-like pseudopodia. It is much easier to recognize the
cysts, which are double-walled, with the inner wall having
a variety of polygonal shapes. With Wheatly trichrome staining,
however, the trophozoite cytoplasm stains green, the karyosome stains red, and the cyst walls stain green. Moreover,
cysts and trophozoites can be differentiated with calcofluor
white staining. Under fluorescent microscopy, the cyst walls
stain green and the trophozoites stain red.55
A corneal biopsy should be considered if the epithelium
is intact but the stromal lesion is active. Again, nonnutrient
agar plates containing a lawn of E. coli can be used to culture
the organisms in biopsy specimens. Other staining procedures such as hematoxylin and eosin, periodic acid-Schiff,
calcofluor white, or fluorescein-labeled antibodies against
Acanthamoeba can be used to stain organisms in biopsy
specimens. Moreover, electron microscopy techniques can be
applied to identify the parasite in the corneal tissues. More
recently, tandem confocal microscopy has been used in
diagnosis and managing Acanthamoeba keratitis because of
the ability to detect the organism in the cornea in vivo. This
technique can also be used to monitor patients who have been
treated for Acanthamoeba keratitis. In these patients,
organisms are absent, but typically create lacunae in the
corneal stroma.23,56,57 Although confocal microscopy is a very
sensitive and powerful technique for identifying acanthamoebae in the cornea, the current level of resolution
available with in vivo confocal microscopy limits its ability to
make an accurate diagnosis. Mathers et al23 reported that
confocal microscopy can be used to distinguish between
inflammatory cells, fungi, and Acanthamoeba. However, they
found that it was not possible to discriminate between
macrophages and Acanthamoeba trophozoites. More recently,
Parmar et al28 reported that confocal microscopy can
demonstrate both the cyst and often also the trophozoite
forms of Acanthamoeba in suspected keratitis. It also shows
the enlarged corneal nerves accompanying radial neurokeratitis
and the characteristic honeycomb-patterned intrastromal
microcavities seen during the late stages of the disease
(Fig. 52.4).
More recently, restriction enzyme analysis of either
mitochondrial DNA or cellular DNA was applied to differentiate species of parasites. However, these characteristics
were not correlated with morphologic identification of different
726
FIGURE 52.4. Tandem-scanning confocal micrographs of Acanthamoeba keratitis (width of each micrograph represents 400 mm). (a) Typical cyst
(15-30 mm) is seen as a bright oval particle, the double-walled nature partially apparent (to the left of the main cyst) through the oblique optical
section taken in this image. (b) Typical rounded cyst (arrowhead) and linearly elongated trophozoite adjacent to and presumably ingesting a
subepithelial corneal nerve (arrow). (c) Two cysts (arrows) surrounded by a nonreflecting circumferential cavity. The stroma has several areas in
the background that are optically lucent distributed in a honeycomb pattern, reflecting microcavities of focal stromal tissue destruction.
Parmar ND, Awwad ST, Petroll WM, et al: Tandem scanning confocal corneal microscopy in the diagnosis of suspected Acanthamoeba keratitis. Ophthalmology (in press).
Acanthamoeba infection can mimic several other kinds of
keratitis, including viral, bacterial, or fungal keratitis. This
condition is frequently misdiagnosed as herpes simplex virus
keratitis. Minor trauma can be associated with both the onset
of the dendritic form of herpes simplex virus keratitis as well
as Acanthamoeba keratitis. However, contact lens wear is not
a frequent predisposing factor in herpes simplex virus keratitis.
Clinically, during the early stages of infection with herpes
simplex virus keratitis, the dendritic lesion is ulcerated,
whereas in Acanthamoeba keratitis, the pseudodendrite is
elevated at the center of the cornea, and the epithelium has
a gelatinous appearance. In contrast to the stromal keratitis in
herpes simplex virus infection, small stromal infiltrates are
present in Acanthamoeba infection. However, at more advanced
stages of the disease, which are accompanied by ulceration and
loss of stromal tissue, the keratitis may resemble herpes
simplex virus infection.
The history of trauma and chronic infection associated with
fungal or bacterial keratitis may mimic Acanthamoeba
infections. However, the severe pain, radial keratoneuritis, and
the annular infiltrate associated with Acanthamoeba keratitis
may aid in the clinical differentiation. Auran et al12 reviewed
several cases and reported that in some cases of Acanthamoeba
keratitis, the appearance of satellite lesions increased the
suspicion of fungal infection. They concluded that in many
cases, Acanthamoeba infection was considered only when
treatment with antibacterial, antifungal, and antiviral drugs
were ineffective to control the infection. Moore54 reported that
fungal and mycobacterial keratitis is slowly progressive and
nonsuppurative. In contrast to Acanthamoeba keratitis,
however, severe pain and ring infiltrate are normally absent in
patients with fungal and mycobacterial keratitis. Moreover,
keratitis resulting from bacterial infections are suppurative,
have stromal infiltrate(s), may be accompanied with anterior
chamber reaction, and develop more rapidly than
Acanthamoeba infection.
TREATMENT
has been reported by several investigators using a combination

of antibiotic, antiviral, antiparasitic, antifungal, and
antiprotozoal drugs.12 Ocular application of several drugs
demonstrated the efficacy of diamidine derivatives such as
0.15% dibrom-propamidine (Brolene ointment), 0.1%
propamidine isethionate (Brolene solution pentamidine
isethionate 0.050.1% (Pentam 300). The sterile pentamidine
isethionate powder can be mixed with artificial tears and
applied topically as recommended for Brolene solution.
Moreover, imidazole derivatives, such as miconazole (10
mg/mL) are used for treatment of Acanthamoeba keratitis. The
intravenous preparation can be applied topically on the cornea
at a dose of 1 drop every 2 h. Clotrimazole (1.0%) also has been
recommended for treatment of Acanthamoeba keratitis. The
sterile powder can be mixed with artificial tears to obtain a 1.0%
solution. One drop of clotrimazole solution can be applied to
the cornea every 2 h. Successful treatment with aminoglycoside derivatives such as Neosporin (neomycinpolymyxin
Bgramicidin), with or without Brolene, has been reported.54,59
Oral administration of ketoconazole (Nizoral) at a dose of
200600 mg/day is also recommended.
More recently, PHMB has been reported to kill
Acanthamoeba cysts and trophozoites effectively. Patients have
been successfully treated with a topical application of 0.02%
PHMB, 624 times/day without any toxic effect to the corneal
epithelium.57,60 Initially, the drug was given 13 times every
hour, ranging from 6 to 24 times/day. This regimented therapy
may continue for weeks and slowly tapered to maintenance
therapy for a year or more.
Treatment Options
Drug Database
The limited number of cases of Acanthamoeba keratitis makes

it difficult to organize a trial to determine an effective
treatment. In vitro sensitivity testing has shown that a few
agents are effective in killing the trophozoites, but the cysts are
resistant and a higher concentration of these drugs is necessary
to kill them. Successful treatment of Acanthamoeba keratitis
Epithelial debridement
Cryotherapy
Keratoplasty
0.15% dibrom-propamidine (Brolene ointment)

0.1% propamidine isethionate (Brolene solution)
Miconazole (10 mg/mL)
Clotrimazole (1.0%)
Neosporin (neomycinpolymyxin Bgramicidin), with or without
Brolene,
Oral administration of ketoconazole (Nizoral) at a dose of
200600 mg/day
Polyhexamethylene biguanide (PHMB) 0.02%
ACKNOWLEDGMENT
This work was supported by NIH grant EY09756, NIH grant R24 EY016664,
and an unrestricted grant from Research to Prevent Blindness, Inc, New
York, NY.
CHAPTER 52
species.58 Mathers et al23 demonstrated that polymerase

chain reaction (PCR) is more sensitive than histological
techniques to identify Acanthamoeba in the corneal tissues
of Acanthamoeba patients. Currently, the identification of
Acanthamoeba species is based on cyst morphology.
REFERENCES
1. Visvesvara GS: Classification of
Acanthamoeba. Rev Infect Dis 1991;
13(Suppl 5):S369S372.
2. Page FC: Taxonomic criteria for limax
amoebae, with descriptions of 3 new
species of Hartmannella and 3 of
Vahlkampfia. J Protozool 1967; 14:499521.
3. Page FC: Re-definition of the genus
Acanthamoeba with descriptions of three
species. J Protozool 1967; 14:709724.
4. Gast RJ: Development of an
Acanthamoeba-specific reverse
dot-blot and the discovery of a new

ribotype. J Eukaryot Microbiol 2001;
48:609615.
5. Stothard DR, Schroeder-Diedrich JM,
Awwad MH, et al: The evolutionary history
of the genus Acanthamoeba and the
identification of eight new 18S rRNA gene
sequence types. J Eukaryot Microbiol 1998;
45:4554.
6. Wang SS, Feldman HA: Isolation of
hartmannella species from human throats.
N Engl J Med 1967; 277:11741179.
7. Kingston D, Warhurst DC: Isolation of

amoebae from the air. J Med Microbiol
1969; 2:2736.
8. Cerva L, Serbus C, Skocil V: Isolation of limax
amoebae from the nasal mucosa of man.
Folia Parasitol (Praha) 1973; 20:97103.
9. Brown TJ, Cursons RT, Keys EA: Amoebae
from antarctic soil and water. Appl Environ
Microbiol 1982; 44:491493.
10. Visvesvara GS, Mirra SS, Brandt FH, et al:
Isolation of two strains of Acanthamoeba
castellanii from human tissue and their
727
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
SECTION 6
24.
25.
26.
27.
728
pathogenicity and isoenzyme profiles.

Rivera F, Medina F, Ramirez P, et al:
Pathogenic and free-living protozoa
cultured from the nasopharyngeal and oral
regions of dental patients. Environ Res
1984; 33:428440.
Auran JD, Starr MB, Jakobiec FA:
Acanthamoeba keratitis. A review of the
literature. Cornea 1987; 6:226.
Paszko-Kolva C, Yamamoto H, Shahamat M,
et al: Isolation of amoebae and
Pseudomonas and Legionella spp. from
eyewash stations. Appl Environ Microbiol
1991; 57:163167.
Griffiths AJ, Hughes DE: Starvation and
encystment of a soil amoeba Hartmannella
castellanii. J Protozool 1968; 15:673677.
Chambers JA, Thompson JE: A scanning
electron microscopic study of the
excystment process of Acanthamoeba
castellanii. Exp Cell Res 1972; 73:415421.
Weisman RA: Differentiation in
Acanthamoeba castellanii. Annu Rev
Microbiol 1976; 30:189219.
Cordingley JS, Wills RA, Villemez CL:
Osmolarity is an independent trigger of
Acanthamoeba castellanii differentiation.
J Cell Biochem 1996; 61:167171.
Aksozek A, McClellan K, Howard K, et al:
Resistance of Acanthamoeba castellanii
cysts to physical, chemical, and
radiological conditions. J Parasitol 2002;
88:621623.
Schaumberg DA, Snow KK, Dana MR: The
epidemic of Acanthamoeba keratitis: where
do we stand? [see comments]. Cornea
1998; 17:310.
Jones DB, Visvesvara GS, Robinson NM:
Acanthamoeba polyphaga keratitis and
Acenthamoeba uveitis associated with fatal
meningoencephalitis. Trans Ophthalmol
Soc U K 1975; 95:221232.
Martinez AJ, Janitschke K: Acanthamoeba,
an opportunistic microorganism: a review.
Infection 1985; 13:251256.
Kumar R, Lloyd D: Recent advances in the
treatment of Acanthamoeba keratitis. Clin
Infect Dis 2002; 35:434441.
Mathers WD, Sutphin JE, Folberg R, et al:
Ophthalmol 1996; 121:129142.
Hargrave SL, McCulley JP, Husseini Z:
Results of a trial of combined propamidine
isethionate and neomycin therapy for
Acanthamoeba keratitis. Brolene Study
Group. Ophthalmology 1999; 106:952957.
He YG, Niederkorn JY, McCulley JP, et al:
In vivo and in vitro collagenolytic activity of
Acanthamoeba castellanii. Invest
Moore MB, McCulley JP: Acanthamoeba
keratitis associated with contact lenses:
six consecutive cases of successful
management. Br J Ophthalmol 1989;
73:271275.
Mannis MJ, Tamaru R, Roth AM, et al:
Acanthamoeba sclerokeratitis. Determining
diagnostic criteria. Arch Ophthalmol 1986;
104:13131317.
28. Parmar ND, Awwad ST, PetrollWM, et al:

Tandem scanning confocal corneal
microscopy in the diagnosis of suspected
Acanthamoeba keratitis. Ophthalmology
2007; 113:538547.
29. Niederkorn JY, Alizadeh H, Leher H,
McCulley JP: The pathogenesis of
Acanthamoeba keratitis. Microbes Infect
1999; 1:437443.
30. Panjwani N, Zhao Z, Baum J, et al:
Acanthamoebae bind to glycolipids of
rabbit corneal epithelium. Infect Immun
1992; 60:34603463.
31. Panjwani N, Zaidi TS, Gigstad JE, et al:
Binding of Pseudomonas aeruginosa to
neutral glycosphingolipids of rabbit corneal
epithelium. Infect Immun 1990; 58:114118.
32. Morton LD, McLaughlin GL, Whiteley HE:
Adherence characteristics of three strains
of Acanthamoeba. Rev Infect Dis 1991;
13(Suppl 5):S424.
33. Panjwani N, Zhao Z, Baum J, et al:
Acanthamoebae bind to rabbit corneal
epithelium in vitro. Invest Ophthalmol Vis
Sci 1997; 38:18581864.
34. Cao Z, Jefferson DM, Panjwani N: Role of
carbohydrate-mediated adherence in
cytopathogenic mechanisms of
Acanthamoeba. J Biol Chem 1998;
273:1583815845.
35. Leher H, Silvany R, Alizadeh H, et al:
Mannose induces the release of cytopathic
factors from Acanthamoeba castellanii.
36. Mitra MM, Alizadeh H, Gerard RD,
Niederkorn JY: Characterization of a
plasminogen activator produced by
Acanthamoeba castellanii. Mol Biochem
Parasitol 1995; 73:157164.
37. Hadas E, Mazur T: Biochemical markers of
pathogenicity and virulence of
Acanthamoeba sp. strains. Parasitol Res
1993; 79:696698.
38. Mitro K, Bhagavathiammai A, Zhou OM,
et al: Partial characterization of the
proteolytic secretions of Acanthamoeba
polyphaga. Exp Parasitol 1994;
78:377385.
39. Hadas E, Mazur T: Proteolytic enzymes of
pathogenic and non-pathogenic strains of
Acanthamoeba spp. Trop Med Parasitol
1993; 44:197200.
40. Heffler KF, Eckhardt TJ, Reboli AC,
Stieritz D: Acanthamoeba endophthalmitis
in acquired immunodeficiency syndrome.
Am J Ophthalmol 1996; 122:584586.
41. Moshari A, McLean IW, Dodds MT, et al:
Chorioretinitis after keratitis caused by
Acanthamoeba: case report and review of
the literature. Ophthalmology 2001;
108:22322236.
42. Clarke DW, Alizadeh H, Niederkorn JY:
Failure of Acanthamoeba castellanii to
produce intraocular infections. Invest
43. Stewart GL, Shupe K, Kim I, et al:
Antibody-dependent neutrophil-mediated
killing of Acanthamoeba castellanii. Int J
Parasitol 1994; 24:739742.
44. van Klink F, Taylor WM, Alizadeh H, et al:
The role of macrophages in Acanthamoeba
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
keratitis. Invest Ophthalmol Vis Sci 1996;

37:12711281.
Marciano-Cabral F, Toney DM: The
interaction of Acanthamoeba spp. with
activated macrophages and with
macrophage cell lines. J Eukaryot Microbiol
1998; 45:452458.
McCulley JP, Alizadeh H, Niederkorn JY:
Acanthamoeba keratitis. Clao J 1995;
21:7376.
Ferrante A, Abell TJ: Conditioned medium
from stimulated mononuclear leukocytes
augments human neutrophil-mediated
killing of a virulent Acanthamoeba sp.
Infect Immun 1986; 51:607617.
McClellan K, Howard K, Niederkorn JY,
Alizadeh H: Effect of steroids on
Acanthamoeba cysts and trophozoites.
42:28852893.
Hurt M, Proy V, Niederkorn JY, Alizadeh H:
The interaction of Acanthamoeba castellanii
cysts with macrophages and neutrophils.
J Parasitol 2003; 89:565572.
Leher HF, Alizadeh H, Taylor WM, et al:
Role of mucosal IgA in the resistance to
Acanthamoeba keratitis. Invest Ophthalmol
Vis Sci 1998; 39:26662673.
Leher H, Kinoshita K, Alizadeh H, et al:
Impact of oral immunization with
Acanthamoeba antigens on parasite
adhesion and corneal infection. Invest
Leher H, Zaragoza F, Taherzadeh S, et al:
Monoclonal IgA antibodies protect against
Acanthamoeba keratitis. Exp Eye Res
1999; 69:7584.
Yang Z, Cao Z, Panjwani N: Pathogenesis
of Acanthamoeba keratitis: carbohydratemediated hostparasite interactions. Infect
Immun 1997; 65:439445.
Moore MB: Acanthamoeba keratitis. Arch
Ophthalmol 1988; 106:11811183.
Silvany RE, Luckenbach MW, Moore MB:
The rapid detection of Acanthamoeba in
paraffin-embedded sections of corneal
tissue with calcofluor white. Arch
Ophthalmol 1987; 105:13661367.
Cavanagh HD, Petroll WM, Alizadeh H,
et al: Clinical and diagnostic use of in vivo
confocal microscopy in patients with
corneal disease. Ophthalmology 1993;
100:14441454.
Winchester K, Mathers WD, Sutphin JE,
Daley TE: Diagnosis of Acanthamoeba
keratitis in vivo with confocal microscopy.
Cornea 1995; 14:1017.
Visvesvara GS, Jones DB, Robinson NM:
Isolation, identification, and biological
characterization of Acanthamoeba
polyphaga from a human eye. Am J Trop
Med Hyg 1975; 24:784790.
Moore MB, McCulley JP, Luckenbach M,
et al: Acanthamoeba keratitis associated
with soft contact lenses. Am J Ophthalmol
1985; 100:396403.
Larkin DF, Kilvington S, Dart JK: Treatment
of Acanthamoeba keratitis with
CHAPTER
53
Interstitial Keratitis
Elmer Y. Tu
INTRODUCTION
The term interstitial keratitis (IK) was rst introduced by John
Hutchinson in the 1850s in his description of eye ndings in
syphilis. As used today, it refers to any nonulcerative inflammatory process of the corneal stroma, a common feature to a
varied list of disorders. In each specic disease, IK represents
a different balance of appropriate and inappropriate (hypersensitivity) inflammatory response, often resulting in unwanted
morbidity.
DESCRIPTION
The hallmark of IK is stromal inflammation without epithelial
or endothelial involvement. Epithelial edema and other
secondary effects of stromal inflammation as well as thinning
after its resolution are consistent with a diagnosis of IK. Stromal
vascularization is another prominent, but not universal, feature
of IK usually following the onset of corneal inltration. These
abnormally permeable vessels may allow stromal extravasation
of lipid that remains for years. With resolution, these vessels
become inactive leaving another hallmark of IK, ghost vessels
(Fig. 53.1).
IK can be unilateral or bilateral and involve the entire cornea
or be restricted to a sectoral, peripheral, or central location. The
inflammation may be confluent or occur in patches. These
areas of inflammation may affect a single depth, multiple levels
FIGURE 53.1. Inactive syphilitic keratitis. Not lipid deposition

superiorly, faint ghost vessels centrally (small arrows) and ghost
vessels in relief in an area of pre-Descemets scarring. (Inset) Confocal
microscopy showing deep vessels and stagnant erythrocyte flow.
or the entire thickness of the cornea. The pattern of involvement may be helpful in establishing an early differential diagnosis.
INCIDENCE
IK is now a relatively uncommon disease in the US, accounting
for only 3% of all reported diagnoses for keratoplasty.1 This also
reflects the low incidence of visual loss seen with most treated
IK. IK has been synonymous with syphilitic keratitis,
accounting for almost 90% of reported cases. The introduction
of penicillin in 1943 led to a dramatic drop from ~600 to
2.6 cases per 100 000 of acquired syphilis in 1998.2 By 1998,
Schwartz et al found that of 97 cases of IK, only two cases
represented active syphilis and 16 represented inactive syphilis
with an overall incidence of 18%. Herpes simplex was the most
common cause of IK (35%) while an almost equal number was
categorized as idiopathic (32%).3
BACTERIA-RELATED IK
SYPHILIS
Syphilis is a systemic infection caused by Treponema pallidum.
Acquired syphilis is contracted through intact mucous
membranes or small breaks in the skin. Congenital infection
occurs transplacentally producing the classic triad of deafness,
malformed (Hutchinsons) teeth, and saddle-nose deformity.
Ophthalmic involvement includes scleritis, uveitis, chorioretinitis, bone spiculization and a pigmentary salt and pepper
retinopathy.4
90% of Syphilitic IK occurs in congenital syphilis.5 It presents
with a peak incidence between 9 and 11 years of age, but may
occur between 5 and 20 years of age.6 Bilateral involvement is
seen in 7580% of cases and may occur sequentially or months
to years apart.6,7 Patients present with pain, photophobia and
blurred vision. Patchy inltrates may be single or multifocal and
involve the deep and/or anterior stroma. The inflammation may
begin centrally or as a marginal keratitis. Untreated, inflammation progresses followed by stromal neovascularization
(Fig. 53.1). Intracorneal hemorrhage produces a characteristic
salmon-patch lesion. Resolution leaves nonspecic scarring of
the cornea while ghost vessels are a hallmark of previous IK.
Deeper inflammation may create branching rolled scrolls of
Descemets membrane (Fig. 53.2). IK associated with acquired
syphilis is similar to congenital syphilis, but is, in general,
milder. Only about 40% of acquired syphilitic IK is bilateral.
Diagnosis of ocular disease is based on serologic testing.
Nontreponemal tests such as the VDRL (venereal disease
research laboratory) and RPR (rapid plasma reagin) are rapid,
but remain positive for a lifetime despite adequate treatment.
729
FIGURE 53.2. Inactive syphilitic keratitis. Characteristic branching

scrolls of Descemets membrane with deep scarring. Inset:
Photomicrograph of another patient showing pre-Descemets
vascularization (arrow) and thickened Descemets membrane
(asterisks) with guttae (double arrows). (Hematoxylin and Eosin;
original magnication: 40).
Courtesy of Deepak Edward, MD.
Treponemal tests such as the FTA-ABS (fluorescent treponemal

antibody absorbed) or MHA-TP (microhemagglutination assayT pallidum) tests are required to conrm the results because of
the high-positive rate of nontreponemal tests.5
Antibiotic treatment has little effect on either form of
syphilitic IK implying that the etiology is immunologic rather
than infectious. Topical corticosteroids have shortened the
course of inflammation, reduced or eliminated corneal neovascularization and improved visual outcomes in syphilitic IK.
Treatment of systemic disease consists of 2.4 million units
of intramuscular benzathine penicillin for primary, secondary,
and early latent syphilis (Table 53.1). Two additional doses at
one week intervals are recommended for late or tertiary
syphilis. Neurosyphilis requires benzyl penicillin intravenous
34 million units every 4 h or 2.4 million units of intramuscular procaine penicillin with probenicid for 1014 days.5
Doxycycline or tetracycline may serve as an alternative for
neurosyphilis, but is not recommended.
LYME DISEASE
Lyme disease is also caused by a spirochete, Borrelia burgdorferi,
carried by the tick Ixodes scapularis. The pathognomonic skin
lesion, erythema migrans, occurs in the area of the tick bite and
TABLE 53.1. Treatment of Interstitial Keratitis

Causative Agent
Local Treatment
Primary Systemic
Therapy
Secondary Systemic
Therapy
Syphilis
Corticosteroids alone
Benzathine penicillin IM
Latent/ Tertiary requires 3 weekly doses
Neurosyphilis
Benzyl PCN IV or procaine PCN IM
Doxycycline
Lyme disease
Corticosteroids +
Systemic therapy
doxycycline, amoxicillin, cephalosporins
IV penicillin
cephalosporins
Tuberculosis
Corticosteroids +
Systemic therapy
Combination of isoniazid, ethambutol,

rifampin, pyrazinamide
Cycloserine, ethionamide,
fluoroquinolones, aminoglycosides
Leprosy
Corticosteroids +
Systemic therapy
Combination therapy
Rifampicin
Ofloxacin
Minocycline
Daspone
Clofamizine
Levofloxacin
Sparfloxacin
Clarithromycin
Brucellosis
Systemic therapy
? Corticosteroids
Tetracycline
Cephalosporins
Rifampin
TMP-SMX
Aminoglycosides
Leptospirosis
Corticosteroids +
Systemic Therapy
Amoxicillin/clavulanate
Doxycycline
Cogans Syndrome
Corticosteroids
High-dose corticosteroids for

vestibuloauditory symptoms
Methotrexate
Azathioprine
Cyclophosphamide
Cyclosporine
Contact lens keratitis
Corticosteroids or
Supportive
None
None
Sarcoidosis
Corticosteroids
Corticosteroids
Mycosis fungoides
Systemic therapy
Radiation
Psoralen
PUVA
a-interferon
SECTION 6
Bacterial
Immune
730
TABLE 53.1. Treatment of Interstitial Keratitiscontd

Causative Agent
Local Treatment
Primary Systemic
Therapy
Secondary Systemic
Therapy
Onchocerciasis
Systemic therapy
Ivermectin
?Doxycycline
Suramin
Diethylcarbamazine
Acanthamoeba
Biguanide- Chlorhexidene
or PHMB
DiamidinePropamidine
Itraconazole (if needed for eye disease)
Neomycin (topical)
Triazoles (topical)
Leishmaniasis
Systemic therapy
Cutaneous
Pentavalent antimony compounds
Visceral
Miltefosine
Amphotericin B
Pentamidine
Paromycin
T. cruzi
Systemic therapy
Benznidazole
+Nifurtimox
T. brucei
Systemic therapy
Pentamidine or suramin
CNS
Malersoprol
CNS
Eflornithine + Nifurtimox
Microsporidiosis
Fumagillin +/Systemic therapy
Fumagillin
Albendazole
Itraconazole
Sulsoxazole
Propamidine
Herpes simplex
Corticosteroids +Trifluridine
or Systemic therapy
Acyclovir
Valacyclovir
Famvir
Herpes zoster
Corticosteroids +/
Topical Acyclovir or Vidarbine
Acyclovir
Valacyclovir
Famvir
EpsteinBarr
Lubrication
Corticosteroids (if needed)
Supportive
HLTV-1
Corticosteroids
Supportive
Antiretroviral trials underway
Mumps
Corticosteroids
Supportive
Measles
Corticosteroids
Treat malnutrition/hypovitaminosis A
Vaccinia/ variola
Corticosteroids + Trifluridine
or Vidarbine
VIG-Vaccinia Immune Globulin

(may worsen keratitis)
Parasitic
Trypanosomiasis
resolves spontaneously. Hematogenous spread then results in

systemic neurologic and rheumatologic symptoms. Ocular
ndings include episcleritis, conjunctivitis, uveitis, retinal
edema and papilledema associated with pseudotumor cerebri.
IK is uncommon, but manifests as single or multiple nummular
stromal lesions, normally bilateral.8 Treatment of the
underlying infection will lead to resolution of the IK while
corticosteroids help control the acute inflammation.9
MYCOBACTERIA
TUBERCULOSIS
Ocular disease caused by Mycobacterium tuberculosis is usually
a result of direct hematogenous spread to the eye. The most
common manifestations are iritis, chorioretinitis or vasculitis.
IK is found in only 0.2% of patients, but can more commonly
produce an anterior segment granuloma, scleritis, or
phlyctenulosis.10 Tuberculous IK starts in the peripheral
stroma, remaining marginal or progressing centrally to involve
a sector of the cornea (Fig. 53.3). The disease is normally
unilateral,11 and may be an extension of tuberculous scleritis.12
Left untreated, stromal neovascularization and corneal scarring

may occur. Diagnosis relies on demonstrating systemic
tuberculosis. Topical corticosteroids limit local inflammation,
but denitive treatment requires a multi-drug regimen for
active tuberculosis.
LEPROSY
Leprosy is caused by the bacteria Mycobacterium leprae and
takes two forms, lepromatous and tuberculoid. Nerve
involvement is common to both, but lepromatous leprosy
includes the widespread skin disease potentiated by an
underlying defect in cell-mediated immunity. Lepromatous IK is
uncommon in the US, found in only 23% of patients. More
common manifestations include iritis, keratopathy and corneal
hypesthesia secondary to the nerve involvement.13 The risk of
keratopathy persists after treatment because of lagophthalmos.
IK is usually bilateral and involves the superior cornea with
supercial neovascularization. Readily available treatment
makes signicant visual disability in the US is rare. Worldwide,
however, visual disability related to leprosy has a total incidence
of ~250 000.14,15
CHAPTER 53
Viral
731
FIGURE 53.3. Tuberculous interstitial keratitis. Slit beam shows

patchy inltration at several different levels of the cornea. Inset:
Sectoral involvement in this hematogenously disseminated disease.
OTHER BACTERIA
Passed through unpasteurized milk, Brucella is a gram-negative
intracellular coccobacilli which causes vague constitutional
symptoms of malaise and fever. Ocular involvement results in
uveitis, optic neuritis and choroiditis. A nummular IK has been
described consisting of coin-shaped subepithelial inltrates.16
Tetracycline is the primary therapy. Cephalosporins, rifampin,
trimethoprimsulphamethoxazole (TMPSMX) or aminoglycosides are added for chronic recurrent disease.17
Systemic leptospirosis, caused by another spirochete, usually
produces a flu-like syndrome, but may also result in classic
Weil disease (acute hepatorenal failure with fever). Ocular ndings
are predominantly anterior and posterior uveitis, but also include
optic neuritis, and IK. The IK is described as large central corneal
edema with Descemets folds and keratic precipitates.18 Corneal
scarring and adjacent limbal vascularization is also seen.
Diagnosis is largely by blood culture and treatment consists of
amoxicillin/clavulanate and corticosteroids.
IMMUNE-RELATED IK
SECTION 6
COGANS SYNDROME
732
David Cogan was the rst to characterize a form of nonsyphilitic IK in association with vestibuloauditory symptoms in
1945.19 Cogans Syndrome is rare and is classically described
in young, Caucasian adults. In typical Cogans Syndrome,
patients exhibit progressive, bilateral, sensorineural hearing
loss often with Meinere-like vestibular symptoms in association with IK. Corneal inflammation may begin months to years
after auditory symptoms and appear initially as a peripheral,
subepithelial keratitis (Fig. 53.4). With repeated recurrences,
stromal inflammation and mild vascularization characteristic
of IK emerges. Atypical Cogans syndrome is the association of
vestibuloauditory symptoms with other forms of ocular
inflammation including scleritis (Fig. 53.4), episcleritis, recurrent
conjunctivitis, uveitis, retinal vasculitis or optic disk edema and
carries a higher likelihood of systemic manifestations.20
Aortitis, similar to that seen with Takayasus arteritis, has
been reported in 10% of patients and can result in aortic
aneurysms or valvular disease.21,22 Neurologic symptoms, both
central and peripheral, as well as musculoskeletal symptoms
including polyarthralgias and myalgias have been reported.23
Cogans has also been reported in association with systemic
FIGURE 53.4. Cogans Syndrome. (Top) Subtle peripheral inltrates

(large arrow) seen in typical Cogans syndrome. (Bottom) Perilimbal
inltrates (arrows) seen adjacent to scleritis associated with the
atypical form.
Courtesy Howard H. Tessler, MD.
autoimmune disease including Wegener s granulomatosis,

sarcoidosis, various arthritides, and inflammatory bowel disease.
The rate of irreversible deafness is greater than 50% making
early treatment critical. In a patient with IK, any clinical suspicion of vestibuloauditory dysfunction should lead to audiometric testing. Studies have detected autoantibodies against
inner ear and cornea,24,25 but the correlation with active IK is
inconsistent.24 High-resolution magnetic resonance imaging
(MRI) demonstrates vestibular and cochlear inflammation in
active disease and shows obstruction of semicircular canals in
postinflammatory disease.24,26
Corneal symptoms are normally responsive to topical corticosteroids, but recur once withdrawn. Initial treatment for
vestibuloauditory involvement is high-dose (1 mg kg1 day1)
systemic corticosteroids and is most effective early in its course.
Treatment is continued for up to 6 months if effective. Alternative immunosuppressives such as methotrexate, azathioprine,
cyclophosphamide and cyclosporine (systemically and locally)
have been utilized with varying success.27,28 Cochlear implants
are helpful in restoring some function in cases proceeding to
complete deafness.29
CONTACT LENS-RELATED IK
Contact lens-induced IK (CL-related IK) is a corneal hypersensitivity response to contact lens-adherent proteins, solutions
MISCELLANEOUS INFILTRATIVE IK
Sarcoidosis is a granulomatous inflammatory disorder which
may cause a variety of ocular disease, primarily uveitis. IK is
rare but has been reported in children and as part of a Coganslike syndrome involving vestibuloauditory symptoms. Topical
corticosteroids are effective in treatment of the corneal lesions
while systemic steroids are effective for systemic disease.31,32
Mycosis fungoides is a malignant dermatologic T-cell
lymphoma which can involve the face and eyelids. Necrotizing
corneal disease has been described as well as a focal unilateral
IK. Rheumatoid arthritis is associated with sterile keratolysis,
peripheral ulcerative keratitis as well as an immune IK. One of
the treatments for rheumatoid arthritis, gold injections, can
create an idiosyncratic IK responsive to topical steroids.33
PARASITE-RELATED IK
ONCHOCERCA
Onchocerciasis (river blindness) affects 20 million people and is
a major cause of worldwide blindness.34 Its causative organism,
a larial parasite, Onchocerca volvulus is carried by Simulium
blackflies which infect humans during a blood meal. The infection is rare in North America and should only be considered
with a history of travel to endemic areas of Africa, Central and
South America. The adult worm releases microlariae which
migrate into multiple organs including eyelids and conjunctiva.
There, the microlariae migrate centripetally into the corneal
stroma where they are, initially, well tolerated. Death of the
microlariae initiates a sclerosing keratitis (Fig. 53.5) beginning
peripherally and progressing centrally until vision is obscured.
Panstromal neovascularization follows the inflammation.
FIGURE 53.5. Onchocerciasis. Peripheral confluent sclerosing

keratitis with central sparing is seen early and progresses centrally to
obscure vision.
Courtesy Hugh R. Taylor, MD
Diagnosis is through direct observation of coiled microlariae

in the cornea or anterior chamber. Traditional treatment is with
ivermectin (single dose of 150 mg/kg) against the microlariae.
Recent landmark work has established that antigens from the
endosymbiotic Wolbachia bacteria activate a Toll-like receptor 4
which may, in turn, initiate the well-established inflammatory
cascade seen in Onchocerca IK.35 Since the bacteria are integral
to the Onchocerca life cycle, treatment directed toward
Wolbachia with doxycycline has been shown to decrease larial
load and reduce corneal inflammation.36,37
ACANTHAMOEBA
90% of Acanthamoeba keratitis is associated with contact lens
wear and 90% with soft contact lens wearers. Because the
amoeba resides in sources of fresh water, a history of exposure
while wearing contacts should raise clinical suspicion.38,39 Early
symptoms are nonspecic irritation, foreign body sensation
with minimal visual disturbance. Paradoxical corneal hypesthesia may occur in association with severe pain, photophobia
and visual loss. Corneal ndings usually begin with a groundglass epitheliitis followed by stromal invasion signied by
pathognomonic radial neuritis. Further progression results in a
patchy central, anterior stromal nummular-type IK sometimes
more evident after initiation of treatment and nally a characteristic immune ring inltrate (Fig. 53.6). The infection is
most often confused with herpes simplex keratitis with which
it shares many clinical features.
Diagnosis utilizes a number of modalities including confocal
microscopy, culture on charcoal or non-nutrient agar with an
E.coli overlay, calcafluor white stain, and occasionally corneal
biopsy. The keratitis is steroid responsive, but, when used without
antiacanthamoebal agents, likely encourages proliferation of the
organism. Treatment includes epithelial dbridement and a
topical biguanide such as PHMB (polyhexamethylene biguanide;
Bacquacil) or chlorhexidene. These are effective against both the
trophozoite form (and perhaps the cyst form).40 The addition of
propamidine, a diamidine, effective against the trophozoite can
be benecial early.41 The use of neomycin, clotrimazole topically
or itraconazole systemically have also been reported. Keratoplasty
is required in medical treatment failures.
LEISHMANIASIS
Leishmaniasis is a protozoan infection caused by the species in
the genus Leishmania and transmitted by a sandfly vector. The
infection follows two different patterns, cutaneous and visceral42,43
with the cutaneous form further subdivided into American
(New World) and Old World. IK is seen most commonly with
American (mucocutaneous) leishmaniasis characterized by a
parrots beak nose. This nding is secondary to infection and
ulceration of the nasopharyngeal mucosa and underlying
cartilage.43 The IK clinically resembles syphilitic IK with sectoral
or diffuse involvement of the cornea followed by dense neovascularization. Direct contact with eyelid lesions can lead to a
suppurative ulcer and corneal abcess formation. Unlike
syphilitic IK, treatment of the underlying infection is curative.43
First-line therapy are the pentavalent antimony compounds,
sodium stibo-gluconate and meglumine antimoniate, followed
by amphotericin B, pentamidine, and paromycin.44 Miltefosine
has some efcacy in the visceral form.
CHAPTER 53
or the lens material itself. The IK will normally present as a

focal or multifocal dense bright white opacity in the peripheral
anterior cornea without a signicant epithelial defect.
Photophobia is severe with pain. The IK may also present as a
more central nummular or circinate pattern of IK resulting in
scarring and stromal neovascularization.30 CL-related IK is
exquisitely sensitive to topical corticosteroids, but must be
distinguished from infectious keratitis. Signs which support the
diagnosis of aseptic keratitis include an intact epithelium, multiple
foci, peripheral or perilimbal location and a quiet anterior
chamber. Recurrences mandate a change in wear regimen.
TRYPANOSOMIASIS
The protozoa Trypanosoma causes two major forms of human
disease, American Chagas disease (T. cruzi) and African sleeping sickness (T. brucei). They are transmitted by the reduviid
733

FIGURE 53.6. Different clinical presentations of
Acanthamoeba. (Top Left) Classic late ring
inltrate with central stromal edema. (Top right)
Confocal microscopy of showing numerous
double-walled cysts. Inset: Numerous doublewalled cysts in a corneal scraping (DiffQuick
stain). (Bottom Left) Patchy central interstitial
keratitis seen in a patient during treatment.
(Bottom Right) Classic radial neuritis is a
hallmark of the disease.
Courtesy of Joel Sugar, MD
bug and the tsetse fly, respectively.42 Ocular ndings in Chagas

disease are related to direct inoculation of the conjunctiva
(Romanas sign) or the periocular region, but keratitis is
uncommon. T. brucei causes either an acute, fulminant East
African form or a more chronic West African form. Initial
infection results in involvement of lymph nodes, liver and
spleen, eventually progressing on to the central nervous system
producing hallmark neurologic decits and somnolence. Unlike
Chagas disease, T. brucei is thought to invade ocular structures. Acute corneal ulceration has been reported as well as IK
with corneal inltration and neovascularization. Treatment of
ocular manifestations is directed toward the underlying infection.
Chagas disease is treated with benznidazole and nifurtimox
while African sleeping sickness is treated in the hematologic
stage with pentamidine or suramin. The CNS stage requires
malersoprol or eflornithine sometimes in combination with
nifurtimox.44
SECTION 6
MICROSPORIDIA
734
The causal agents of microsporidiosis are obligate intracellular

protozoa increasing in incidence secondary to systemic
immunosuppression and AIDS45 (Fig. 53.7). Transmission is
either by direct contact or through inhalation. Systemic
infection is spread by infection of macrophages and subsequent
hematogenous dissemination. Corneal involvement may take
two forms, an epitheliopathy or stromal keratitis. The stromal
form occurs in immunocompetent hosts while the more common epitheliopathy is classically described, but not restricted
to, immunocompromised individuals.46 Stromal keratitis
resembles herpes simplex virus (HSV) stromal keratitis with a
nonsuppurative, inltrative central corneal inflammation. This
can lead to neovascularization, ulceration and/or perforation.
Vittaforme (formerly Nosema) corneae is the most commonly
associated species. Treatment is with topical fumagillin 70 mg/mL
and systemic albendazole, although corneal transplantation is
often required for cure. In settings with altered cellular immunity
such as AIDS, corneal infection causes minimal inflammation
with a central, raised, punctate epitheliopathy (Fig. 53.7). The
most reported species are from the genus Encephalitozoon.
Treatment consists of dbridement, topical fumagillin and oral
albendazole.47,48 Itraconazole, sulsoxazole, and propamidine
have also been used with some success.
FIGURE 53.7. Epithelial Microsporidia. The cornea exhibits a

punctate, raised, central epitheliopathy. Inset: Characteristic
intracellular ovoid protozoa seen on a corneal scraping.
VIRAL IK
HERPES SIMPLEX STROMAL KERATITIS
Herpes simplex Stromal Keratitis (HSK) is the most common
form of IK encountered in the US3 and is also a leading cause of
corneal blindness in the third world. The majority of corneal
infections are related to HSV-1 or the labial form of herpes, but
HSV-2 or genital herpes is also reported.
Corneal disease may affect any level of the cornea and is
usually unilateral. Stromal keratitis normally occurs after one
or more repeated episodes of surface disease. Disciform keratitis
is characterized by central stromal edema with mild cellular
inltration, a mild iritis, keratic precipitates and Descemets
folds (Fig. 53.8). A focal or multifocal pattern of stromal inflammation with either peripheral or central involvement may also
occur, attracting stromal neovascularization. Other forms
include a marginal keratitis, an uncommon necrotizing stromal
keratitis, characterized by ulceration with epithelial defect and
an immune Wesley-type ring comparable to late-stage
acanthamoebal keratitis.
The pathophysiology of HSK-related damage is, in large part,
immunologic. Studies have demonstrated resident HSV-specic
CD4+ T cells in the corneal stroma in HSK and a number
of pro-inflammatory cytokines.49 The Herpetic Eye Disease
Study (HEDS) established that corticosteroids, when used in
conjunction with trifluridine, signicantly shortened the course
of stromal keratitis over trifluridine alone.50 The role of active
viral replication is less clear, since attempts to isolate active
virus from HSK has been inconclusive. Supporting a role for
active viral replication, the HEDS group found that long-term
suppression with oral acyclovir was benecial in decreasing the
number of recurrences of stromal and other forms of keratitis.51
It appears that a limited viral reactivation may be required to
trigger the self-amplifying corneal hypersensitivity response.
The current treatment regimen for stromal keratitis is the use
of topical corticosteroids with trifluridine and/or oral antivirals
for prophylaxis. Long-term oral therapy is indicated in patients
with frequent recurrences to lessen visual complications.52
years.54 The disciform keratitis in HZV is similar to that seen

in HSV and follows several weeks to months after the acute
HZO. Corneal disease can precede or occur without skin
disease.55 As in HSV keratitis, HZV patients develop a
sometimes dense neurotrophic keratopathy producing severe
complications. A late IK may occur with stromal scarring and
neovascularization which likely shares the immunologic nature
of HSK. The mainstay of HZV epitheliitis and stromal keratitis
is corticosteroids.56 The role of topical antivirals is unclear but
they are effective in chronic epithelial disease.
EPSTEINBARR VIRUS
Epstein-Barr virus (EBV) is another member of the
Herpesviridae family and is the agent responsible infectious
mononucleosis in the US. EBV infection causes conjunctivitis,
dacryoadenitis and iritis. Corneal disease may include dendritic
keratitis, stromal and an epidemic keratoconjunctivitis-like
pattern of subepithelial inltrates. Peripheral geographic
changes with stromal edema have also been described. Stromal
keratitis can present as multiple coin-like lesions with a clear
intervening stroma multiple granular circular opacities of the
anterior and mid-stroma, or a multifocal full-thickness IK with
mild stromal vascularization (Fig. 53.9).57,58 Evidence for EBVassociated ocular disease is largely circumstantial because
diagnosis relies mainly on systemic clues and serologic testing.
Treatment is with lubrication and topical corticosteroids.
HERPES ZOSTER
CHAPTER 53
Like herpes simplex, the varicellazoster (HZV) virus can cause

a range of ocular diseases. As chickenpox, HZV produces lid
and conjunctival disease. In adults, reactivation manifests as
herpes zoster ophthalmicus (HZO) when involving the
ophthalmic division of the trigeminal nerve with greater than
50% of patients developing corneal involvement.53 Both forms
may produce a self-limited punctate keratopathy or a
dendritiform (pseudodendrite) keratitis which normally resolves
without sequelae. Immunocompromised patients can, however,
harbor active virus in the epithelial layer over months or
FIGURE 53.8. Photomicrograph showing HSV interstitial keratitis.

Note the corneal vascularization (arrows) and giant cell (arrowhead).
(Hematoxylin and Eosin; original magnication: 40).
FIGURE 53.9. EBV keratitis. (Top) Nummular inltrates sometimes

seen after cessation of systemic infection are usually not vision
threatening. (Bottom) Coin-shaped lesions (small arrows) with clear
intervening stroma with a perilimbal interstitial keratitis (large arrow).
Courtesy of Deepak Edward, MD.
Courtesy of Joel Sugar, MD.
735
HUMAN T-LYMPHOCYTIC VIRUS TYPE I

Human T-lymphocytic Virus Type I (HTLV-1) is the causative
agent in adult T-cell leukemia and neuromyelopathy
(HAM/TSP) and is endemic in several regions including the
Caribbean, Central and South America, Japan, and Africa.
Uveitis and keratoconjunctivitis sicca are common. IK is
usually bilateral and largely asymptomatic appearing as whitish
or gray anterior stromal opacities and are located peripherally.59
The incidence is signicantly higher in patients with a neuromyelopathy (HAM/TSP).59 The lesions respond well to topical
corticosteroids.
MISCELLANEOUS VIRUSES
Mumps may cause a number of corneal complications
including stromal keratitis, endotheliitis, corneal edema and
uveitis.60 Ocular involvement is normally unilateral with onset
of symptoms 1 to 11/2 weeks after the onset of the characteristic
parotitis followed by resolution 12 weeks later. Neovascu-
larization and corneal scarring is uncommon. Topical corticosteroids may hasten the resolution.61 Measles (rubeola) may
cause a conjunctivitis with characteristic bluewhite centered
red spots (similar to true Kopliks spots) found in the conjunctiva.
A punctate epitheliopathy is usually mild and nonvisionthreatening.62 In developing countries, stromal keratitis,
ulceration, perforation and secondary infection may be seen in
the setting of malnutrition and vitamin A deciency.63
Prior to its eradication, variola (smallpox) was a major cause
of blindness. The virus causes a pustular, scarring rash.
Subepithelial inltrates and keratitis originate as a direct
extension from limbal pustules. Disciform keratitis may follow
resolution of the rash and lead to visually signicant corneal
scarring.64 Smallpox vaccinations utilized another orthopoxvirus, vaccinia, which can cause a keratitis similar to variola
by direct contact innoculation. Disciform keratitis has also been
reported 23 months after immunization as well as a Coganslike syndrome.64 Topical antivirals and immune globulin
may be used to treat or prevent involvement of the cornea.65
Corticosteroids can reduce subsequent scarring.
SECTION 6
REFERENCES
736
1. Maeno A, Naor J, Lee HM, et al: Three

decades of corneal transplantation:
indications and patient characteristics.
Cornea 2000; 19:711.
2. Primary and secondary syphilis United
States, 1998. MMWR Morb Mortal Wkly
Rep 1999; 48:873878.
3. Schwartz GS, Harrison AR, Holland EJ:
Etiology of immune stromal (interstitial)
4. Margo CE, Hamed LM: Ocular syphilis.
5. Kiss S, Damico FM, Young LH: Ocular
manifestations and treatment of syphilis.
Semin Ophthalmol 2005; 20:161167.
6. Knox CM, Holsclaw DS: Interstitial keratitis.
7. Hariprasad SM, Moon SJ, Allen RC,
Wilhelmus KR: Keratopathy from congenital
syphilis. Cornea 2002; 21:608609.
8. Miyashiro MJ, Yee RW, Patel G, Ruiz RS:
Lyme disease associated with unilateral
interstitial keratitis. Cornea 1999; 18:115116.
9. Zaidman GW: The ocular manifestations of
Lyme disease. Int Ophthalmol Clin 1997;
37:1328.
10. Donahue HC: Ophthalmologic experience
in a tuberculosis sanatorium. Am J
Ophthalmol 1967; 64:742748.
11. Thompson MJ, Albert DM: Ocular
tuberculosis. Arch Ophthalmol 2005;
123:844849.
12. Tabbara KF: Ocular tuberculosis: anterior
segment. Int Ophthalmol Clin 2005; 45:5769.
13. Lewallen S, Tungpakorn NC, Kim SH,
Courtright P: Progression of eye disease in
cured leprosy patients: implications for
understanding the pathophysiology of
ocular disease and for addressing eyecare
needs. Br J Ophthalmol 2000; 84:817821.
14. Lewallen S, Courtright P: Blindness in
Africa: present situation and future needs.
15. Whitcher JP, Srinivasan M, Upadhyay MP:
Corneal blindness: a global perspective.
Bull World Health Organ 2001; 79:214221.
16. al-Kaff AS: Ocular Brucellosis. Int
17. Tabbara KF, al-Kassimi H: Ocular
brucellosis. Br J Ophthalmol 1990;
74:249250.
18. Mancel E, Merien F, Pesenti L, et al: Clinical

aspects of ocular leptospirosis in New
Caledonia (South Pacic). Aust N Z
J Ophthalmol 1999; 27:380386.
19. Cogan DG: Syndrome of nonsyphlitic
interstitial keratitis and vestibuloauditory
symptoms. Arch Ophthalmol 1945;
33:144149.
20. Haynes BF, Kaiser-Kupfer MI, Mason P,
Fauci AS: Cogan syndrome: studies in
thirteen patients, long-term follow-up, and
a review of the literature. Medicine
(Baltimore) 1980; 59:426441.
21. Grasland A, Pouchot J, Hachulla E, et al:
Typical and atypical Cogans syndrome: 32
cases and review of the literature.
Rheumatology (Oxford) 2004; 43:10071015.
22. Smith JC, Peck JE, Ray LI, Smith EC:
Aortoarteritis and sensorineural hearing
loss in an adolescent black male. Am J
Otolaryngol 2004; 25:370376.
23. Orsoni JG, Zavota L, Vincenti V, et al: Cogan
syndrome in children: early diagnosis and
treatment is critical to prognosis. Am J
Ophthalmol 2004; 137:757758.
24. Helmchen C, Arbusow V, Jager L, et al:
Cogans syndrome: clinical signicance of
antibodies against the inner ear and cornea.
Acta Otolaryngol 1999; 119:528536.
25. Lunardi C, Bason C, Leandri M, et al:
Autoantibodies to inner ear and endothelial
antigens in Cogans syndrome. Lancet
2002; 360:915921.
26. Casselman JW, Majoor MH, Albers FW: MR
of the inner ear in patients with Cogan
syndrome. AJNR Am J Neuroradiol 1994;
15:131138.
27. Shimura M, Yasuda K, Fuse N, et al:
Effective treatment with topical cyclosporin
A of a patient with Cogan syndrome.
28. Riente L, Taglione E, Berrettini S: Efcacy
of methotrexate in Cogans syndrome.
J Rheumatol 1996; 23:18301831.
29. Pasanisi E, Vincenti V, Bacciu A, et al:
Cochlear implantation and Cogan
syndrome. Otol Neurotol 2003; 24:601604.
30. Braude LS, Sugar J: Circinate-pattern
interstitial keratopathy in daily wear soft
contact lens wearers. Arch Ophthalmol
1985; 103:16621665.
31. De Smedt S, Ayliffe W: Interstitial keratitis

and deafness in a patient with cutaneous
sarcoidosis. Bull Soc Belge Ophtalmol
2001:1518.
32. Lennarson P, Barney NP: Interstitial keratitis
as presenting ophthalmic sign of
sarcoidosis in a child. J Pediatr Ophthalmol
Strabismus 1995; 32:194196.
33. Zamir E, Read RW, Affeldt JC, et al: Gold
induced interstitial keratitis. Br J
Ophthalmol 2001; 85:13861387.
34. Pearlman EaL, JH: Keratitis due to
Onchocerciasis. Ophthalmol Clin 1994;
7:641648.
35. Saint Andre A, Blackwell NM, Hall LR, et al:
The role of endosymbiotic Wolbachia
bacteria in the pathogenesis of river
blindness. Science 2002; 295:18921895.
36. Hoerauf A, Mand S, Volkmann L, et al:
Doxycycline in the treatment of human
onchocerciasis: Kinetics of Wolbachia
endobacteria reduction and of inhibition of
embryogenesis in female Onchocerca
worms. Microbes Infect 2003; 5:261273.
37. Hoerauf A, Mand S, Fischer K, et al:
Doxycycline as a novel strategy against
bancroftian lariasis-depletion of Wolbachia
endosymbionts from Wuchereria bancrofti
and stop of microlaria production. Med
Microbiol Immunol (Berl) 2003; 192:211216.
38. Kilvington S, Gray T, Dart J, et al:
Acanthamoeba keratitis: the role of
domestic tap water contamination in the
United Kingdom. Invest Ophthalmol Vis Sci
2004; 45:165169.
39. Radford CF, Minassian DC, Dart JK:
Acanthamoeba keratitis in England and
Wales: incidence, outcome, and risk
factors. Br J Ophthalmol 2002; 86:536542.
40. Kosrirukvongs P, Wanachiwanawin D,
Visvesvara GS: Treatment of acanthamoeba
keratitis with chlorhexidine. Ophthalmology
1999; 106:798802.
41. Duguid IG, Dart JK, Morlet N, et al:
Outcome of acanthamoeba keratitis treated
with polyhexamethyl biguanide and
propamidine. Ophthalmology 1997;
104:15871592.
42. Lam S: Keratitis caused by leishmaniasis or
trypanosomiasis. Ophthalmol Clin North
Am 1994; 7:635639.
51. Barron BA, Gee L, Hauck WW, et al:
Herpetic eye disease study. A controlled
trial of oral acyclovir for herpes simplex
stromal keratitis. Ophthalmology 1994;
101:18711882.
52. Oral acyclovir for herpes simplex virus eye
disease: effect on prevention of epithelial
keratitis and stromal keratitis. Herpetic Eye
Disease Study Group. Arch Ophthalmol
2000; 118:10301036.
53. Liesegang TJ: Corneal complications from
herpes zoster ophthalmicus.
54. Chern KC, Conrad D, Holland GN, et al:
Chronic varicella-zoster virus epithelial
keratitis in patients with acquired
immunodeciency syndrome. Arch
Ophthalmol 1998; 116:10111017.
55. Silverstein BE, Chandler D, Neger R,
Margolis TP: Disciform keratitis: a case of
herpes zoster sine herpete. Am J
Ophthalmol 1997; 123:254255.
56. Karbassi M, Raizman MB, Schuman JS:
Herpes zoster ophthalmicus. Surv
Ophthalmol 1992; 36:395410.
57. Matoba AY, Jones DB: Corneal
subepithelial inltrates associated with
systemic Epstein-Barr viral infection.
58. Matoba AY, Wilhelmus KR, Jones DB:
Epstein-Barr viral stromal keratitis.
59. Merle H, Cabre P, Olindo S, et al: Ocular

lesions in 200 patients infected by the
human T-cell lymphotropic virus type 1 in
martinique (French West Indies). Am J
Ophthalmol 2002; 134:190195.
60. Riffenburgh RS: Ocular manifestations of
mumps. Arch Ophthalmol 1961;
66:739743.
61. Singh K, Sodhi PK: Mumps-induced
corneal endotheliitis. Cornea 2004;
23:400402.
62. Kayikcioglu O, Kir E, Soyler M, et al: Ocular
ndings in a measles epidemic among
young adults. Ocul Immunol Inflamm 2000;
8:5962.
63. Foster A, Yorston D: Corneal ulceration in
Tanzanian children: relationship between
measles and vitamin A deciency. Trans R
Soc Trop Med Hyg 1992; 86:454455.
64. Semba RD: The ocular complications of
smallpox and smallpox immunization. Arch
Ophthalmol 2003; 121:715719.
65. Jack MK, Sorenson RW: Vaccinial keratitis
treated with IDU. Arch Ophthalmol 1963;
69:730732.
CHAPTER 53
43. Roizenblatt J: Interstitial keratitis caused by

American (mucocutaneous) leishmaniasis.
44. Croft SL, Barrett MP, Urbina JA:
Chemotherapy of trypanosomiases and
leishmaniasis. Trends Parasitol 2005;
21:508512.
45. Rastrelli PD, Didier E, and Yee RW:
Microsporidial keratitis. Ophthalmol Clin
North Am 1994; 7:617633.
46. Chan CM, Theng JT, Li L, Tan DT:
Microsporidial keratoconjunctivitis in
healthy individuals: a case series.
Ophthalmology 2003; 110:14201425.
47. Gritz DC, Holsclaw DS, Neger RE, et al:
Ocular and sinus microsporidial infection
cured with systemic albendazole. Am J
Ophthalmol 1997; 124:241243.
48. Yee RW, Tio FO, Martinez JA, et al:
Resolution of microsporidial epithelial
keratopathy in a patient with AIDS.
49. Koelle DM, Reymond SN, Chen H, et al:
Tegument-specic, virus-reactive CD4
T cells localize to the cornea in herpes
simplex virus interstitial keratitis in humans.
J Virol 2000; 74:1093010938.
50. Wilhelmus KR, Gee L, Hauck WW, et al:
Herpetic eye disease study. A controlled
trial of topical corticosteroids for herpes
1994; 101:18831895; discussion 9596.
737
CHAPTER
54
Recurrent Corneal Epithelial Erosion

Natalie A. Afshari, Sherman W. Reeves, Kenneth R. Kenyon, and Roberto Pineda
Recurrent erosion of the corneal epithelium is a clinical

syndrome of multiple etiologies, characterized by inadequate
epithelialstromal attachments, resulting in episodic dysadhesion and defects of the corneal epithelium. These episodes
are typically acute, with symptoms ranging from mild ocular
irritation to extreme pain. Most frequently associated with prior
trauma or underlying corneal disease, the condition may be a
prolonged and debilitating condition, frustrating for both the
patient and the treating physician.
EPIDEMIOLOGY
To date, only case reports and small case series have provided
epidemiologic data on recurrent corneal erosions as no population-based studies currently exist. The condition appears to
occur across all ages, with reports from childhood to the late
eighth decade; however, the average age is the mid-fth
decade,1,2 with a slight female predominance13 but no racial
predisposition.
Trauma and epithelial basement membrane dystrophy (mapdot ngerprint) disease are the most common underlying
etiologies of recurrent erosion, with trauma accounting for
4569% of cases and epithelial basement membrane dystrophy
associated with 2030% of cases.14 Incidence estimates of
recurrent erosion following traumatic corneal abrasion have
ranged from 5% to 25%.46 Other dystrophic and degenerative
diseases account for a minority of cases.
CORNEAL EPITHELIAL WOUND HEALING

The process of normal corneal epithelial healing provides
insight into the aberrant wound healing response of recurrent
epithelial erosion syndrome.
The reestablishment of the corneal surface after epithelial
disturbance normally occurs in three stages.7 First, the formation of a single-cell leading wound edge signals the migratory
stage of wound healing, which reestablishes the cellular barrier.8
In stage two, the proliferative stage, mitosis occurs in the cells
surrounding the epithelial defect, resulting in the restoration of
the normal epithelial layer architecture.9 In the nal stage, the
epithelial basement membrane is deposited and adhesive complexes are regenerated.10,11
EPITHELIAL MIGRATION
During the rst 46 h after an epithelial injury, no appreciable
decrease in wound size occurs. This is referred to as the latent
phase. It is characterized by increased intracellular protein synthesis, actin lament polymerization, and reorganization from
the apical to the basal region of cells. The basal and squamous
cells in the vicinity of the wound thicken and separate.

Neutrophil accumulation, mainly from the tear lm layer,
occurs along the wound edge at ~3 h,12 as does thinning of the
epithelium to a single layer of flattened cells. Noncellular
constituents of the epithelium change as well, including
increased concentrations of bronectin, brinogen, and brin
on the corneal surface in 18 h.
A linear phase follows, as flattened epithelial cells move
across the epithelial defect until it is completely covered, a
process usually complete by the fourth or fth day after injury,
depending on initial wound size.5 The formation of lamellipodia and lopodia marks the beginning of cell migration.
Gipson and colleagues have demonstrated actin laments at
the leading edges of migrating cells and within the podial
extensions.13 It is thought that epithelial cells migrate en
masse as a continuous sheet, with most cells retaining their
positions relative to each other.6 Dura and Forrester studied
human corneal epithelial defects with limbal involvement
and demonstrated that a preferential circumferential migration
of a population of cells occurs along the limbus, from both
ends of the limbal defect.14 This circumferential migration
continues until the advancing ends meet to reestablish a
confluent epithelial cover for the limbus.
EPITHELIAL PROLIFERATION
Basal epithelial cells are the key participants in the proliferative
process that restores the epithelial defect. Stem cells at the
limbus are the immortal progenitor cells responsible for corneal
epithelial cell replacement, as their mitosis renews the stem cell
population itself and amplies epithelial cell production.15
Evidence for the existence and location of limbal corneal stem
cells includes 3H-thymidine labeling showing high rates of cell
mitosis at the limbus;16 a large epithelial wound, in which the
edge of the defect is closer to the limbus, heals at a faster rate
than a small central wound;17 and the histopathologic appearance of regenerated limbal epithelium resembles corneal and
not conjunctival epithelium.18
The exact role of growth factors in epithelial proliferation and
healing has yet to be elucidated. The interactions between
epithelial cells and extracellular matrix proteins and proteases
are mediated in part through endogenously produced peptide
growth factors. Epidermal growth factor, transforming growth
factor-a and -b, and nerve growth factor are among several
proteins being investigated. Additionally, the upregulation of
matrix metalloproteinases has been demonstrated in response
to cytokine stimulation in corneal wounds.19 These enzymes,
which degrade extracellular matrix and basement membrane
constituents, may play a role in basement membrane and
stromal remodeling.20,21
739
SECTION 6
740

FIGURE 54.1. Morphology and clinical appearance of a normal cornea, corneal epithelial erosion, persistent defect, and ulceration. (a, Left) A
Normal corneal displays regular epithelial layers, a uniform continuous basement membrane, and Bowmans layer plus stroma devoid of
inflammatory cells. (Right) Phase-contrast microscopy discloses these same features. Paraphenylenediamine 125. (b, Left) Recurrent epithelial
erosion involves epithelial defects associated with aberrant and discontinuous basement membrane plus subepithelial fibrocellular pannus, but
without a Bowmans layer defect or inflammation. (Right) Clinical features of posttraumatic recurrent epithelial erosion include an extensive area
of loose epithelial adhesion with devitalized shreds of the epithelial sheet floating on the tear film. (c, Left) A persistent epithelial defect is notable
for a thickened and nonmotile epithelial cell layer at the edge of the defect, defects of basement membrane and Bowmans layer, and some
inflammatory cells. PMN, polymophonuclear neutrophil. (Right) Neurotrophic keratitis in an anesthetic cornea displays a typical persistent
epithelial defect with gray, thickened margins of immobile epithelium overlying a somewhat edematous but structurally intact stroma. (d, Left)
Sterile ulceration involves enzymatically mediated degradation n of Bowmans layer and stroma, usually in the presence of acute and chronic
inflammatory cells. (Right) Biomicroscopy of a sterile ulcer reveals features of a persistent epithelial defect, plus stromal ulceration as evident in
the thinning of the slit-lamp light beam.
(ad, Left) From Kenyon KR, Wagoner MD: Conjunctival and corneal injuries. In: Shingleton BJ, Hersh PS, Kenyon KR, eds. Eye trauma. St Louis, Mosby-Year Book;
1991.
Newly regenerated epithelium must anchor itself to the underlying connective tissue before the healing process is complete.
Transient attachments occur during cell migration. Damaging
of the epithelium causes disassembly of the remaining adjacent
epithelial cell hemidesmisomes. This disassembly allows adjacent cells to migrate over the wounded surface. Re-formation of
the adhesion complexes gradually occurs; starting at the periphery and progressing centrally.5 The migrating cells lack hemidesmisomes and must use an alternative mode for developing
adhesions.22 Focal contacts at the leading edge of epithelium
form by linkages from the cytoplasmic actin lament to
extracellular matrix proteins like bronectin, brinogen-brin,
laminin, and tenascin. These attachments are mediated by cell
surface adhesion molecules. Fibronectin receptor (integrin) is
associated with the formation of these contacts.23 Degradation
of the contacts by proteases allows restructuring of these contacts. Urokinase-type plasminogen activator activates plasminogen to plasmin, which in turn cleaves brinogen and brin,
allowing release of the leading edge of epithelium.14,23
The presence of an intact epithelial basement membrane
plays a crucial role in adhesion of the newly regenerated epithelium to the underlying Bowmans membrane. In cases where
epithelial cells are scraped but the basement membrane complex remains intact, new anchoring cellular adhesions are
formed by the seventh day postinjury. However, with deeper
corneal injuries where the corneal stroma itself is exposed, the
epithelium may not be fully adherent for up to 7 weeks after the
injury.5,11
PATHOGENESIS
Under normal conditions, the corneal epithelium is rmly
anchored to the underlying Bowman layer and stroma by
specialized attachment complexes. This epithelial basement
membrane complex, comprised of the hemidesmisomes of the
basal epithelial cell plasma membrane (BM), the directly
apposed extracellular collagenous basement membrane, and
its attendant anchoring brils, is responsible for tight adhesion
of the corneal epithelium to the stroma (Fig. 54.1a). Consequently, any traumatic, dystrophic, or degenerative process of
the BM can predispose to defective epithelial adhesion and
repetitive breakdown of the epithelial cell layer; and hence, the
so-called recurrent erosion syndrome (see Fig. 54.1b). In
situations where the ocular surface is further compromised by

extensive epithelial damage, limbal stem cell deciency,
protracted inflammation, sensory denervation, tear deciency,
corneal exposure and stromal scarring or ulceration, the
problems of adhesion are compounded by a failure of epithelial
mitosis and migration, resulting in a persistent epithelial defect
(see Fig. 54.1c). Such epithelial defects frequently occur in the
setting of chronic inflammation and, in fact, may be selfperpetuating since inflammatory cells inhibit epithelial wound
healing. This vicious circle of a persistent epithelial defect
and chronic inflammation is of further concern as it predisposes
to enzymatically mediated sterile ulceration and potential
thinning of the corneal stroma (see Fig. 54.1d).
Although the exact mechanism by which the epithelial
anchoring process fails is not known, increased activity of
matrix metalloproteinases-2 and -9 has been demonstrated in
patients with recurrent epithelial erosions. With the ability to
dissolve anchoring brils and basement membrane, aberrant
activity of these enzymes may play a role in the pathophysiology
of this process.2426
ETIOLOGY
Although recurrent epithelial erosions are associated with many
different diseases, the common denominator is the ultrastructural abnormality in one or more of the components of the
epithelial cell attachment complex. A classication of epithelium and basement membrane complex disorders that result
in these alterations is summarized in Table 54.1. As a specic
example, primary epithelial disorders, such as epithelial
basement membrane dystrophy (mapdotngerprint dystrophy), predispose to the recurrent erosion syndrome, and
acquired epithelial disorders, as occur with fth cranial nerve
palsy, lead to a neurotrophic keratitis with persistent epithelial
defect. Similarly, a primary disorder of the basement membrane
complex, as in diabetes mellitus, may result in a persistent
epithelial defect, whereas an acquired basement membrane
complex disorder, most commonly after a mechanical abrasion,
produces only recurrent corneal erosion. Combined epithelial
and basement membrane complex disorders can also develop in
more widespread ocular surface damage, such as chemical burn.
Depending on the severity and the circumstance, both recurrent
erosion and persistent defects of the epithelium may be
encountered.
CHAPTER 54
EPITHELIAL ADHESION
741
TABLE 54.1. Classification of Corneal Epithelial and Basement

Membrane Disorders
Primary Epithelial Disorders
Epithelial basement membrane dystrophy (mapdotfingerprint)
Meesman's dystrophy
Acquired Epithelial Disorders
Neurotrophic keratitis (fifth-nerve palsy)
Neuroparalytic keratitis (seventh-nerve palsy)
Infectious and postinfectious keratitis
Herpes simplex and zoster
Lid anomalies
Primary keratoconjunctivitis sicca
Sjgren's syndrome
Graft-versus-host disease
Contact lens keratopathy and overwear
The clinical signs evident with slit-lamp examination vary

from focal supercial punctate keratitis to an area of devitalized, disadhesive epithelium to a full-thickness epithelial
defect (Fig. 54.2). Patients lacking an obvious cause for
erosion should be carefully examined with flouroscein staining
and retroillumination for signs of epithelial basement membrane dystrophy in the asymptomatic eye. In corneas with
suspected erosion but lacking in obvious biomicroscopic
evidence of an epithelial defect, the presence of occult epithelial
adhesion is detected at the slit lamp by use of a dry cellulose
surgical sponge (Weck-cel) rubbed gently and tangentially over
the area of suspect epithelium. If the intact epithelia sheet is
moveable (positive adhesion test), then the lack of adequate
epithelialstromal adhesion is certain. This simple diagnostic
test may also be of use in the preoperative evaluation of laser
vision correction patients.27 In addition to epithelial basement
membrane dystrophy, other dystrophic processes and acquired
corneal disease, as detailed in Table 54.1, provide settings in
which corneal erosions are encountered.2730
MEDICAL THERAPY
Rosacea blepharokeratitis
Primary Basement Membrane Complex Disorders
Diabetes mellitus
ReisBcklers dystrophy
Aquired Basement Membrane Complex Disorders
Mechanical injury (especially shear or tangential)
Combined Epithelial and Basement Membrane Complex
Disorders
Chemical and thermal injury
Other limbal stem cell deficiency diseases (primary or acquired)
Band keratopathy
Posttraumatic or postsurgical scar
Salzmann's nodular degeneration
Radiation
Collagen vascular diseases
Cicatricial pemphigoid
Erythema multiforme
SECTION 6
Keratomalacia
742
DIAGNOSIS
Clinically, the recurrent erosion syndrome is characterized by
abrupt transition from an otherwise asymptomatic eye to one
with a varied degree of distress. Symptoms may range from
occasional mild irritation to recurrent attacks of acute ocular
pain, tearing, and redness. Onset typically occurs during sleep
or at awakening.
The initial evaluation must be directed toward establishing
the underlying mechanism in order to select appropriate therapy and afford accurate prognosis. Though several categories
of disease processes may cause recurrent erosions, by far the
most prevalent are those related to trauma and epithelial basement membrane dystrophy. A history of prior corneal abrasion,
especially a shearing injury, as from a tree branch or ngernail
scratch, can often be elicited. Prior symptomatic episodes
may have resolved spontaneously or have received only brief
patching or therapeutic soft contact lens (SCL) therapy.
In most cases, conservative management results in resolution

of the epithelial defect. Thus in the acute setting, a relatively
small (less than one corneal quadrant) and clean (minimal
epithelial debris and no stromal inflammatory inltrate or
anterior chamber reaction) is simply and appropriately treated
by frequent application of articial tears by day and mild antibiotic ointment (erythromycin or bacitracin) at bedtime to provide appropriate lubrication and antibiotic prophylaxis.
Nonsteroidal antiinflammatory agents (NSAIDs) (diclofenac
(Voltaren), ketorolac (Acular), bromfenac (Xibrom)) used as
frequently as every 2 h afford pain relief such that cycloplegia
is seldom warranted. Cautious and judicious use of topical
steroids (prednosolone acetate 1% or fluorometholone) are
applicable with close monitoring in the setting of marked
inflammation causing increased patient discomfort and
retarding epithelial wound healing, but only in the absence of
risk factors (e.g., contact lens-related abrasion) and/or clinical
suspicion of infection. If the epithelial defect is larger and the
patient is extremely uncomfortable, then continuous pressure
patching during the 24- to 72-h interval of epithelial wound
healing may also be employed. Finally, the presence of corneal
stromal inltrate and/or anterior chamber reaction
disproportionate to the extent of surface injury should signal
the possibility of concomitant infection, thereby mandating
performance of microbiological cultures, plus avoidance of
corticosteroids, patching and/or bandage contact lens
application, and mandatory follow-up within 24 h.
The use of a therapeutic SCL may be applicable even in
the acute management setting as the simultaneous use of
topical NSAIDs facilitates avoidance of the so-called tight lens
syndrome (acute steepening of SCL t resulting in decreased
lens movement, inflammatory debris accumulation beneath
lens, increased inflammation, and major pain). Thus a relatively flat (base curve >8.6 mm), plano or low minus power,
high water content SCL can be inserted to improve patient
comfort, cosmesis, and visual function, while simultaneously
facilitating wound healing by protecting the vulnerable
epithelium from the windshield wiper debridement action of
the blinking eyelids. As continuously worn SCL may increase
the risk of microbial keratitis,31 a topical antibiotic
(fluoroquinolone) should be prescribed one or twice daily.
Frequent instillation of unpreserved articial tears or saline
solution will eliminate inflammatory debris and improve
patient comfort. The SCL should then be continuously retained
(with appropriate replacement) for at least 1 to 2 weeks or, if
necessary, for as long as 2 months, since such extended time

may be required for restoration of tight epithelialstromal
adhesion.
After reepithelization either with or without the aid of SCL
therapy, lubricating ointments should be continued at bedtime
for a period of ~8 weeks until the newly restored epithelium
has been able to fully complete reconstruction of basement
membrane complexes. For this purpose, the use of nonpreserved petrolatum ointments is preferable, as we nd them
as effective as hypertonic preparations, which can be irritating.
Close follow-up is mandatory especially for those patients
with concomitant ocular surface diseases who use multiple
topical medications, especially steroids, because the risk of
microbial superinfection is increased. While generally safe and
well tolerated with short-term use, chronic use of topical
nonsteroidal agents, especially in the setting of ocular surface
disease, may also inhibit epithelial healing and thereby increase
the risk for corneal ulceration.32,33
SURGICAL THERAPY
For more recalcitrant erosive cases, surgical therapy, ranging
from simple debridement, anterior stromal puncture, or supercial epithelial keratectomy to phototherapeutic keratectomy
or even surface cautery may be indicated.
CHAPTER 54
FIGURE 54.2. Clinicopathologic correlations of recurrent erosion and persistent defects of corneal epithelium. (a) Epithelial basement membrane
dystrophy. (Left) Biomicroscopy of microcystic (Cogans) variant of dystrophy exhibit debris-filled opaque intraepithelial microcysts. (Middle)
Large cysts have erupted to erode the epithelial surface, staining with fluorescein in a pseudodendrite pattern. (Right) Phase-contrast
microscopy of a superficial keratectomy specimen discloses a disrupted epithelial layer organization, as intraepithelial pseudocysts contain
devitalized epithelial debris. Paraphenylenediamine 125. (b) ReisBcklers dystrophy. (Left) Clinical features include diffuse, reticulated
subepithelial pannus with overlying epithelial irregularity predisposing to erosion. (Middle) After removal of corneal epithelium in preparation for
superficial keratectomy, the irregularly reticulated pattern of the pannus is evident. (Right) Light microscopy of a keratoplasty specimen
demonstrates epithelial irregularity, focal fibrocellular pannus and fractures of Bowmans layer. Paraphenylenediamine 125. (c) Diabetes mellitus.
(Left) An adhesion defect of epithelial basement membrane to Bowmans layer results in the ability of the entire epithelial sheet to be easily
dislodged from the stromal surface, as in the course of vitreoretinal surgery. (Middle) In this patient with diabetes-related corneal anesthesia, the
postvitrectomy recovery was prolonged by a persistent epithelial defect. (Right) Phase-contrast microscopy of the epithelial sheet (as clinically
depicted in (c, left)) resolves the intact epithelial sheet with an abnormally thickened and duplicated basement membrane layer remaining
adherent to epithelium rather than attached to Bowmans layer. Paraphenylenediamine 125.
DEBRIDEMENT
If lubrication, pressure patching and/or bandage soft contact
lens therapy are insufcient for cases involving extensive epithelial deterioration with residual associated cellular debris,
then simple localized debridement readily provides a smooth
basement membrane or Bowmans layer substrate for healthy
epithelium to resurface and readhere. Debridement may also
be appropriately performed at the time of initial evaluation if
extensive sheets of devitalized and disadhesive epithelium are
evident. This minimalist procedure, performed at the slit
743
SECTION 6
744

FIGURE 54.3. Technique of epithelial debridement after recurrent erosion. (a) Devitalized epithelium and debris adherent to the damaged
basement membrane surface inhibit restoration of intact basement membrane and recovery of tight epithelialstromal adhesion. (b) After the
application of topical anesthetic, a dry cellulose sponge is used to sweep aside nonadherent epithelium and debris. (c) Jewelers forceps are
employed to remove loose shards of marginal epithelium. (d) The surface of Bowmans layer is polished with a dry celluose sponge. Topical
antibiotic, steroid, and cycloplegic agents are applied followed by a pressure patch. If the epithelial defect persists beyond 72 h, then the patch
is replaced by a soft contact lens with continuation of the same medical therapy in decreasing doses for 68 weeks.
(ad, Left) From Kenyon KR, Wagoner MD: Conjunctival and corneal injuries. In: Shingleton BJ, Hersh PS, Kenyon KR, eds. Eye trauma. St Louis, Mosby-Year Book,
1991.
ANTERIOR STROMAL PUNCTURE

Rarely, when conventional medical therapy and epithelial
debridement fail, the anterior stromal puncture technique
described by McLean and co-workers35 may be used (Fig. 54.4).
This, technique employs numerous (20100 depending on area
to be treated) lightly applied micropunctures into the anterior
stroma using a no. 18 or 20 hypodermic needle to encourage
epithelial adhesion by inciting focal microcicatrization to spot
CHAPTER 54
lamp under topical anesthesia, is best accomplished with

gentle scrub using a cellulose sponge (Weck-cel) and removal
of the epithelial fragments with jewelers forceps (Fig 54.3).
Sharp instruments such as scalpel blades or chemical
cauterization such as iodine are never appropriate. The
adjunctive medical and/or SCL therapy is then instituted as
previously described. Although debridement assists in healing
of the acute erosive episode, but it may have limited efcacy
in preventing recurrence.34
FIGURE 54.4. Technique of anterior stromal puncture. (Top left) Multiple superficial punctures with a disposable (no. 18 or 20) needle are used to
stimulate microcicatrization between the epithelium, Bowmans layer, and the anterior stroma. (Top right) A patient with recurrent erosion suitable
for anterior stromal puncture displays a focal area of nonadherent epithelium. (Bottom left and right) At the completion of the procedure, multiple
superficial punctures are evident within and around the area of defective epithelium.
(Top left) From Kenyon KR and Wagoner MD: Conjunctival and corneal injuries. In: Shingleton BJ, Hersh PS, Kenyon KR, eds. Eye trauma. St Louis, Mosby-Year
Book, 1991. (Top right, bottom left and right) Courtesy of Dr S M MacRae.
745

weld the epithelium to stroma. While the large-gauge needle
nearly eliminates the risk of perforation, the alternative use of
an inexpensive, commercially available instrument36 or bending
the tip of the needle, congured like a cystatome, also helps
produce small punctures of consistent depth. The treatment is
performed directly over areas of defective epithelium or over the
dysadhesive areas of the cell sheet (using Weck-cel test, 27), and
should extend 12 mm beyond the erosive focus into the normal surrounding tissue. This technique, which is also quickly,
easily and safely performed at the slit lamp under topical
anesthesia, is best suited for patients with single erosive areas
(typically following mechanical trauma) occurring outside the
SECTION 6
Limbus
746
FIGURE 54.5. Technique of superficial keratectomy. (Top left) Schematic summary of the technique presents epithelial debridement with a dry
celluose sponge (a) thus exposing subepithelial pannus (b), which can then be stripped using jewelers forceps (c). (Top right) A patient with
ReisBcklers dystrophy (as depicted in Fig. 54.2b) undergoes removal of the epithelium overlying the pannus, using a dry celluose sponge.
(Middle left) Cellophane-like fibrocellular membranes are stripped from the stromal surface with jewelers forceps. (Middle right) At the conclusion
of the procedure, smooth stromal surface is restored (compare with the preoperative appearance of the same patient in Fig. 54.2b). (Bottom left)
Fluorescein staining of the tear film in this same patient before the superficial keratectomy discloses marked surface irregularity, thereby
decreasing vision and predisposing to epithelial erosion. (Bottom right) After superficial keratectomy, the surface quality of the regenerated
epithelium is markedly improved.
SUPERFICIAL EPITHELIAL KERATECTOMY

Particularly with multifocal disease or when there is excessive
aberrant epithelial basement membrane and subepithelial
collagenous pannus (most commonly in mapdotngerprint
dystrophy but also in numerous other supercial dystrophic
or degenerative disorders), we favor the extremely straightforward procedure of supercial epithelial keratectomy
(Fig. 54.5).39,40 Especially when the visual axis is involved,
patients may experience not only erosive symptoms but also
visual distortions, typically monocular diplopia or ghost
imaging, due to irregular surface topography.
The procedure is best performed with the use of the operating
microscope and topical anesthesia. The area appropriate for
debridement is identied with fluorescein staining and/or the
Weck-cel adhesion test. The epithelium and any loosely
adherent subepithelial debris are removed by gently scraping
with a dry cellulose surgical sponge (Weck-cel) and/or disposable
scarier blade (Grieshaber 681.01 or Beaver 59). The corneal
surface is then kept dry in order to visualize any topographic
irregularities, which may represent focal areas of more substantial basement membrane or pannus accumulations. With
the leading edge of a dry cellulose sponge, cleavage planes can
be identied between the anterior abnormal accumulations and
the underlying Bowman layer or stroma. Jewelers forceps may
then be used to strip these cellophane-like membranes from
the surface. Further polishing of the peeled surface is
accomplished with additional cellulose sponges or careful
scraping. Frequently, and especially in cases of epithelial basement membrane dystrophy, the smooth and glistening surface
of the intact Bowmans layer can be visualized and used as
an appropriate therapeutic end point. Sharp dissection (as in
lamellar keratectomy) or diamond burr polishing is never
appropriate. Furthermore, although some authors recommend
debridement broadly over the entire corneal surface, we perform
only focal debridement and supercial keratectomy in areas
of denite biomicroscopic abnormality and thereby specically
seek to spare any damage to the limbal epithelium. Postoperatively, routine medical management with either pressure
patching or therapeutic SCL typically produces excellent visual
and symptomatic improvement.
PHOTOTHERAPEUTIC KERATECTOMY
In recent years, the excimer laser phototherapeutic keratectomy
has become an established treatment alternative for recurrent
epithelial erosions. After the epithelium is debrided (either
manually or with laser), Bowmans layer is partially removed
(to a depth of 1015 mm) with excimer pulses utilizing a broad
(>6 mm diameter) treatment zone. Such minimalist treatment
produces negligible refractive change,37,38 and the treatment
can also be combined with photorefractive keratectomy in
patients with concomitant myopia. As discussed elsewhere,
deeper ablation depths may be required in patients whose
erosions are caused by deeper anterior stromal abnormality
such as ReisBuckler, granular and lattice dystrophies.3941 In
such cases, hyperopic refractive shifts must be anticipated
postoperatively.
Experimental studies of epithelial wound healing have

suggested that corneal epithelium may migrate more slowly
over irregular corneal surface, such as that produced by manual
keratectomy, compared with the rate of migration over a more
regular surface.42 Phototherapeutic keratectomy performed by
the excimer laser yields a smoother surface contour than
manual keratectomy.43 Animal models show increased rates of
epithelial wound healing after excimer laser versus manual
keratectomy.44 Another presumed advantage of this technique
is increased safety for treatment within the papillary margin,
possibly affording a lower incidence of postprocedural stromal
haze.45
Little evidence exists to suggest phototherapeutic keratectomy
markedly lessens erosion recurrences over other surgical techniques.46 Especially in the highly prevalent situations of epithelial basement membrane dystrophies, the ability of manual
supercial keratectomy to retain an intact Bowmans layer
for subsequent healthy reepithelialization and without unanticipated refractive consequences remains highly advantageous.
Thus, as straightforward techniques such as supercial
keratectomy remain both simple and economical for such
purposes, we tend to reserve phototherapeutic keratectomy for
patients who have failed more conventional and less costly
therapies, or those with subsurface abnormalities.
SURFACE CAUTERY
Finally, as described by Salleras and later modied by Wood,47
surface cautery or diathermy, while having no application in
the ocular surface management of any eye with visual potential,
may be useful in providing relief of painful erosive or bullous
keratopathy in poor visual prognosis eyes that are neither tolerant of bandage SCL therapy nor candidates for penetrating
keratoplasty. With the operating microscope and topical
cocaine anesthesia, ~100 focal cautery burns are applied over
the entire corneal surface, followed by the cellulose sponge
debridement of the devitalized epithelium. Pressure patching
or therapeutic soft contact lens with topical antibiotic, or both,
is continued for 12 weeks until the epithelium has recovered.
TABLE 54.2. Therapy of Recurrent Corneal Erosion

Diagnosis and Primary Therapy
Determine cause
Debride devitalized tissue
Culture (as indicated)
Pressure patch or therapeutic SCL
Medical Therapy
CHAPTER 54
visual axis. In erosions directly involving the visual axis, where

even nebular stromal haze might compromise vision, micropuncture should be employed with extreme caution or avoided.
Following stromal puncture, a therapeutic SCL is usually
applied with appropriate adjunctive medical therapy.
Anterior stromal micropuncture has also been described
using Nd:YAG laser,37,38 but efcacy and follow-up data are
limited.
Lubricants, hypertonics, viscoelastic substances

Topical antibiotic, steroid, NSAID, cycloplegic (as indicated)
Pressure patch or therapeutic SCL
Surgical Therapy
Debridement
Anterior stromal puncture
Superficial epithelial keratectomy
Excimer phototherapeutic keratectomy
Surface cautery
747
SUMMARY
Recurrent epithelial erosion syndrome is characterized by repeated
sloughing of the corneal epithelium. Symptoms range from
mild irritation to severe ocular distress, and classically occur on
waking from sleep. The syndrome is thought to result from
inadequate epithelialstromal attachments, and most commonly
occurs after epithelial trauma or in association with anterior

basement membrane dystrophy. While most acute erosive episodes will resolve with lubrication and/or bandage contact lens
therapy, adjunctive surgical debridement, stromal micropuncture, and manual or excimer laser keratectomy may indicated
and effective. A summary of diagnosis and primary therapy,
medical therapy, and surgical therapy is outlined in Table 54.2.
SECTION 6
REFERENCES
748
1. Reidy JJ, Paulus MP, Gona S: Recurrent

erosions of the cornea: epidemiology and
treatment. Cornea 2000; 19(6):761771.
2. Heyworth P, Morlet S, Rayner S, et al:
Natural history of recurrent erosion
syndrome a 4 year review of 117 patients.
BJOG 1998; 82:2628.
3. Kang PC, Zlogar DF, Adelman RA,
Afshari NA: Recurrent corneal erosion: a
study of 259 cases. Invest Ophthalmol Vis
Sci 2003; 44:EAbstract 809.
4. Thygeson PT: Observations on recurrent
erosion of the cornea. Am J Opthalmol
1959; 47:48.
5. Eke T, Morrison DA, Austin DJ, Recurrent
symptoms following traumatic corneal
abrasion: prevalence, severity, and the
effect of a simple regimen of prophylaxis.
Eye 1999; 13:345347.
6. Wenne LE: Recurrent corneal erosion after
trauma: a statistical study. Ann Ophthalmol
1985; 17:521524.
7. Dua HS, Gomes JAP, Singh A: Corneal
epithelial wound healing. Br J Ophthalmol
1994; 78:401.
8. Crosson CE, Klyce SD, Beuerman RW:
Epithelial wound closure in the rabbit
cornea: a biphasic process. Invest
9. Thompson HW, Malter JM, Steinemann TL,
Beurerman RW: Flow cytometry
measurements of the DNA content of
corneal epithelial cells during wound
32:433.
10. Kenyon KR: The synthesis of basement
membrane by the corneal epithelium in
bullous keratopathy. Invest Ophthalmol
1969; 8:156.
11. Khodadoust AA, Silverstein AM, Kenyon KR,
corneal epithelium. Am J Ophthalmol 1968;
65:339.
12. Dua HS, Forrester JV: Clinical patterns of
corneal epithelial wound healing. Am J
13. Gipson IK, Spurr-Michaud S, Tisdale A,
Kkeough M: Reassembly of the anchoring
structures of the corneal epithelium during
wound repair in the rabbit. Invest
14. Dua HS, Forrester JV: The corneoscleral
limbus in human corneal epithelial
wound healing. Am J Ophthalmol 1990;
10:646.
15. Tseng SC: Concept and application of
limbal stem cells. Eye 1989; 3:141.
16. Cotsarelis G, Cheng S, Dong G, et al:
stimulated to proliferate: Implications on
epithelial stem cells. Cell 1989; 57:201.
17. Matsuda M, Ubels JL, Edelhauser HF: A
larger corneal epithelial wound closes at a
faster rate. Invest Ophthalmol Vis Sci 1985;
26:897.
18. Kinoshita S, Kiorpes TC, Friend J, Thoft RA:

Limbal epithelium in ocular surface wound
23:73.
19. Ottino P, Taheri F, Bazen HE: Plateletactivating factor induces the gene
expression of TIMP-1, -2, and PAI-1:
imbalance between the gene expression of
MMP-9 and TIMP-1 and -2. Exp Eye Res
2002; 74:393402.
20. Mullholand B, Tuft SJ, Khaw PT: Matrix
metalloproteinase distribution during early
corneal wound healing. Eye 2005; 19:
584588.
21. Matsubara M, Girard MT, Kublin CL,
Cintron C, Fini ME: Differential roles for two
gelatinolytic enzymes of the matrix
metalloproteinase family in the remodeling
cornea. Dev Biol 1991; 147:425439.
22. Gipson IK: Adhesive mechanisms of the
corneal epithelium. Acta Ophthalmol 1992;
70:13.
23. Berman M: The pathogenesis of corneal
epithelial defects. Acta Ophthalmol 1989;
67:55.
24. Garrana RMR, Zieske JD, Assouline M,
Gipson IK: Matrix metalloproteinases in
epithelia from human recurrent corneal
erosion. IVOS 1999; 40(6):12661270.
25. Alfonso AA, Sobrin L, Monroy DC, et al:
Tear fluid gelatinase B activity correlates
with IL-1a concentration and fluorescein
clearance in ocular rosacea. Invest
26. Aitken DA, Beirouty ZA, Lee WR:
Ultrastructural study of the corneal
epithelium in the recurrent erosion syndrome.
27. Kenyon KR, Paz H, Greiner JV, Gipson IK:
Corneal epithelial adhesion abnormalities
associated with laser in situ keratomileusis.
28. Fogle JA, Kenyon KR, Stark WJ, et al:
corneal epithelium. Am J Ophthalmol
1975; 79:925.
29. Cogan DG, Donaldson DD, Kuwabara T:
Microcystic dystrophy of the corneal
epithelium. Trans Am Ophthalmol Soc
1964; 62:213.
30. Kenyon KR, Wafai Z, Michels RC, et al:
Corneal basement membrane abnormality
in diabetes mellitus. Invest Ophthalmol Vis
Sci 1978; 17:245.
31. Thoft RA, Mobilia EF: Complications with
therapeutic extended wear soft contact
lenses. Int Ophthalmol Clin 1981;
21:197208.
32. Lu KLa, Wee WRa b, Sakamoto Ta,
McDonnell PJ: Comparison of in vitro
antiproliferative effects of steroids and
nonsteroidal antiinflammatory drugs on
human keratocytes. Cornea 1996;
15:185190.
33. Congdon NG, Schein OD, von Kulajta P,
et al: Corneal complications associated
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
with topical ophthalmic use of nonsteroidal

antiinflammatory drugs. J Cataract Refract
Surg 2001; 27:622631.
Brown N, Bron A: Recurrent erosion of the
McLean EN, McRae SM, Rich LF:
Recurrent erosion: treatment by stromal
puncture. Ophthalmology 1986; 93:784.
Rubinfeld RS, Laibson PR, Cohen EJ,
Arentsen JJ, Eagle RC Jr. Anterior stromal
puncture for recurrent erosions: further
experience and new instrumentation.
Cavanaugh TB, Lind DM, Cutarelli PE, et al:
Photothereapautic keratectomy for
recurrent erosion syndrome in anterior
basement membrane dystrophy.
Seitz B, Langenbucher A, Hafner A,
Kus MM, Naumann GO: Phototherapeutic
keratectomy for recurrent corneal erosion
syndrome (e-PTK). Report on 116
consecutive interventions. Ophthalmologe
2002; 99:703708.
Zuckerman SJ, Aquavella JV, Park SB:
Analysis of the efcacy and safety of
excimer laser PTK in the treatment of
corneal disease. Cornea 1997; 15:9.
Rapuano CJ: Excimer laser phototherapeutic
keratectomy: long-term results and practical
considerations. Cornea 1997; 16:151.
Starr M, Donnenfeld E, Newton M, et al:
Excimer laser phototherapeutic
Lance SE, Capone AJ, Raj NS, et al:
Diamond burring and surgical keratectomy:
Morphologic comparison in the rabbit. Arch
Gartry D, Muir KM, Marshall J: Excimer
laser treatment of corneal surface
pathology: a laboratory and clinical study.
Reidy JJ, Jacobson MS, Thompson HW,
et al: Comparison of corneal epithelial
wound healing after photorefractive and
lamellar keratectomy. J Refractive Surg
1996; 12:352.
Sridhar MS, Rapuano CJ, Cosar CB,
Cohen EJ, Laibson PR: Phototherapeutic
keratectomy vsersus diamond burr
polishing of Bowmans membrane in the
treatment of recurrent corneal erosions
associated with anterior basement
membrane dystrophy. Ophthalmology
2002; 109:674.
Ohman L, Fagerholm P: The influence of
excimer laser ablation on recurrent corneal
erosions: a prospective randomized study.
Cornea 2002; 17:349.
Salleras A: Bullous keratopathy. In: King JH,
McTigue JW, eds. The Cornea World
Congress. Washington DE: Butterworths;
1965:292299.
CHAPTER
55
Persistent Epithelial Defects

Matthew A. Dahlgren, Avninder Dhaliwal, and Andrew J. W. Huang
INTRODUCTION
When the cornea is wounded, the healing process is usually
initiated to reestablish epithelial continuity. Persistent (or
nonhealing) epithelial defects (PEDs) occur when the corneal epithelium fails to regenerate steadily over a corneal wound within
due course (usually less than 2 weeks in normal corneas). Only
after the epithelium has completely covered a tissue defect can
appropriate adhesions to the underlying stroma begin to
develop. Recurrent epithelial breakdowns may occur as a result
of poor adhesion. The potential causes of PEDs are myriad.
Regardless of the cause, PEDs may progress through the subjacent corneal stroma and eventually lead to stromal ulceration.
As our understanding of the mechanisms of corneal epithelial
homeostasis and related tissue reaction to injury continues to
evolve, so will our therapeutic regimens become more targeted
towards the underlying mechanisms.
EPITHELIAL HOMEOSTASIS
The human corneal epithelium consists of ve to six layers of
stratied squamous cells, with three subdivisions: supercial
cells (outer 23 cell layers), wing cells (middle 23 cell layers)
and basal cells (innermost 1 cell layer). In an X, Y, Z hypothesis,
Thoft postulated that the forces X (which represents the proliferation of basal cells) and Y (which represents centripetal
migration of cells) must be balanced by Z (the loss of surface
epithelium) during homeostasis.1 The basal cells comprise
mitotically active cells, with higher proliferative capacities at
the corneal periphery, known as the limbus. The limbus, a
transitional zone between avascular corneal and vascular
conjunctival tissues, contains corneal epithelial stem cells (SC)
which undergo mitosis, giving rise to daughter cells (known as
transiently amplifying cells, TAC) in the basal epithelium. As
basal cells migrate supercially, they lose their mitotic ability
(postmitotic cells, PMC) and differentiate into wing and then
supercial cells (terminally differentiated cells, TDC). The
supercial cells desquamate into the tear lm and are steadily
replenished by more migrating cells. The entire process takes
~714 days.2 The process of desquamation is generally
regulated by apoptosis (programmed cell death).3
As mentioned above, as epithelial cells migrate centripetally
and supercially they also become further differentiated. Various
types of adhesion complexes are formed: desmosomes and tight
junctions in supercial cells; desmosomes and gap junctions in
wing cells; desmosomes, hemidesmosomes, and gap junctions
in basal cells. It is these hemidesmosomes, through anchoring
brils and plaques, which mediate attachment of the basal
epithelium to the basement membrane.4 Tight adhesion of the
corneal epithelium to the underlying stroma is crucial for
maintaining the corneal surface integrity. When only corneal

epithelium is removed, the regenerating epithelial cells can
migrate along an intact underlying basement membrane. The
new epithelial sheet does not develop tight adhesions via basement membrane complexes until 7 days. However, when the
corneal epithelium along with its underlying stroma is removed,
regenerating epithelial cells cover the defect in ~4 days but do
not adhere to the underlying stroma until 8 weeks or later. This
observation indicates that delayed development of epithelial
adhesion to the underlying stroma may play a signicant role in
recurrent epithelial breakdowns and PEDs.5
NORMAL CORNEAL RESPONSE TO INJURY

After corneal injuries, both disrupted epithelial cells and exposed
underlying basement membrane or stroma are responsible for
generating various signals to surrounding cells and into the
overlying tear lm.6 Although the precise mechanisms by which
these signals lead to epithelial migration and healing are poorly
understood, it has been shown that various factors responsible
for epithelial healing are elevated in the wounded cornea and tear
lm. Amongst these are bronectin (FN),7 vinculin,8 plasmin,
tissue plasminogen activator (tPA),9 urokinase plasminogen activator (uPA),9 cell surface receptors such as integrins,10 epidermal
growth factor (EGF),11 EGF receptors,11 nerve growth factor
(NGF),12 and alpha-enolase.13 Since these factors are active in
normal epithelial healing in response to injury, many of them
have been targeted as potential therapeutic options for PEDs.
The rst step in healing an epithelial defect involves epithelial migration to populate the defect. EGF, which is present
in epithelial cells and the tear lm, promotes epithelial
proliferation and migration. After epithelial injury, increased
levels of the glycolytic enzyme, alpha-enolase, are found in the
mitotically and metabolically active limbal basal cells,
suggesting their role as the epithelial TACs.13 High levels of alphaenolase are found up to 4 weeks after wounding, further
suggesting its role in epithelial restratication.13 Time-lapse
phase-contrast cinematography studies have demonstrated that
the migration of epithelial cells is not simply an X and Y phenomenon.14 That is, epithelial cells migrate centripetally, but in
a vortex type pattern while at the same time migrating supercially. EGF, which is present in epithelial cells and the tear
lm, promotes epithelial proliferation and migration.
EXTRACELLULAR MATRIX IN CORNEAL

WOUND HEALING
FN is a multifunctional extracellular matrix protein. FN binds
to specic molecules such as cell surface receptors, collagen, and
brin to facilitate cell adhesion.7 Two forms of FN are present
749

on the wound surface. The soluble plasma bronectin (pFN) in
the tear lm originating from the conjunctival vessels covers
exposed basement membrane of the denuded corneal surface
creating a temporary scaffold for migrating epithelium.15 The
insoluble cellular bronectin (cFN) derived from stromal
keratocytes is localized under the migrating epithelium and disappears after wound closure.7 The resorption of the FN is
mediated by proteolytic enzymes such as plasmin. Other basement membrane matrix components such as collagens, laminins,
keratan sulfate proteoglycan, and heparan sulfate proteoglycans
are also important to corneal epithelial cell proliferation, differentiation, and migration.
Shortly after injury, epithelial cells at the edge of the injury
change their cytoskeleton to form pseudopodia16 and cell surface integrins are upregulated (a process mediated by EGF).10
These actin-rich pseudopodia adhere via integrins to the FN
matrix, a process mediated by vinculin.8 Actin-mediated contraction moves the epithelium in to the denuded area, directed
by chemotactic factors.17 This migration requires the constant
destruction and reformation of attachments between the epithelial cells and their scaffold, a process mediated by several
serine proteases including plasmin, tPA, and uPA.9
Once the basal cells have spread over the defect they lay down
a basement membrane to which they form permanent attachments. Basement membrane constituents such as laminin 1 and
laminin 518 are laid down by migrating epithelium. At the same
time, and possibly influenced by the aforementioned cellmatrix
interactions, cellcell adhesion complexes begin to be formed.
This suggests that even after the epithelium has begun to ll in
the defect, a steady interaction between epithelium and basement membrane or stroma is required for epithelial maturation.
The natural process of supercial migration and differentiation
ensues to form a complete epithelial layer.
PROTEOLYTIC ENZYMES AND

METALLOPROTEINASES
SECTION 6
THE PLASMINOGEN ACTIVATOR/PLASMIN

SYSTEM
750
Plasmin is generated by plasminogen activators (PAs) via proteolysis. After an epithelial injury, FN at the leading edge of the
defect is degraded by the increased production of plasmin,
thereby compromising FNs adhesive functions. FN fragments
can compete with intact FN and other ligands for the epithelial
integrin receptors or other adhesion receptors, leading to an
epithelial defect. Plasmin generated in the stroma may also
contribute to collagenase activation, brinolysis, and eventual
stromal ulceration.19,20
Plasmin and PA have been found in tears. There are two wellknown types of plasminogen activator: the tPA and the uPA.
Tissue-type PA is derived from the conjunctiva or lacrimal gland
and is the predominant type of PA in normal tears.21 In contrast,
uPA is derived from the corneal or conjunctival epithelial cells
and has been found to increase in rabbit tears and human
corneas with injury and inflammation.21,22 Both PAs are
synthesized as monomer enzymes. While tPA has some biologic
activity as a monomer, a dimer uPA is needed to achieve full
enzymatic activity. The primary role of tPA is in brinolysis23
and it is activated upon binding to brin, whereas uPA is involved
in proteolytic events such as epithelial migration, matrix degradation, and tumor invasion. Plasmin is one of a number of
enzymes that can convert latent uPA to its active form.24 The
release of cellular uPA causes degradation of FN, which is needed
by the migrating corneal epithelial cells for cell attachment.25
Corneal stromal ulceration has been correlated with the conversion of uPA from the latent to the active form.26 When uPA is
located at the leading edge of epithelium after corneal wounding,

a secondary epithelial defect can occur as a result of further
degradation of the subepithelial brin or FN by plasmin activation.19,27 Furthermore, uPA can induce corneal neovascularization
in rabbits.28 In turn, the vascular response may mitigate the
severity of subsequent stromal ulceration.27
Plasmin and PA can be inhibited by endogenous inhibitors such
as a-2 antiplasmin, a-2 macroglobulin, and a-1 antitrypsin.22,29,30
These inhibitors are also present in tears and can be increased
by ocular inflammation associated with increased conjunctival
vessel permeability. The balance between plasmin/plasminogen
activator and their inhibitors may thus modulate the process of
epithelial wound healing.
MATRIX METALLOPROTEINASES
Matrix metalloproteinases (MMPs) are enzymes capable of breaking down extracellular matrices. MMPs, such as collagenases,
gelatinases, and stromelysins are involved in the stromal
remodeling during corneal wound healing31 and may contribute
to the development of PED and stromal ulceration. Many cell
types are involved in the complex processes of MMP production
and related matrix degradation.
Collagenases that degrade type I collagen are present in
broblasts, capillary endothelial cells, macrophages/monocytes,
and PMNs of corneal ulcers.32 Cytokines from corneal epithelial
cells and macrophages/monocytes can modulate collagenase
production by stromal keratocytes.32 Latent collagenases may be
activated by plasmin20 and result in the progression from a PED
to stromal ulceration. While collagenases cleave collagen bril,
gelatinases denature triple helix chains of collagen. Following
collagenase activation, both MMP-2 (78 kDa gelatinase) and
MMP-9 (92 kDa gelatinase) degrade collagen types I, II, and III.33
In addition, these two enzymes have specicity for native collagen types VI, V, and VII.34 MMP-2 is found to be upregulated in
human corneal epithelium with recurrent erosions.35 Loss of
the epithelial basement membrane after epithelial wounding
has been noted to precede stromal ulceration.3639 Dissolution
of the basement membrane may be crucial in initiating stromal
ulceration and the process can be activated by the remaining
corneal cells without the involvement of inflammatory cells.
There is evidence suggesting that MMP-9 may be responsible
for degrading the basement membrane, while activated MMP-2
may be present in the basal epithelial cells to degrade the
epithelial anchoring system and contribute to the recurrent
epithelial breakdown.3335 Stromelysins degrade proteoglycans,
FN, and laminin.40
Naturally occurring inhibitors can modulate the production
and activation of the MMPs.20 Tissue inhibitors of matrix metalloproteinases (TIMPs) and a-2 macroglobulin are two principal
inhibitors of MMPs. At least two types of TIMPs have been
known, TIMP-1 and TIMP-2. Both are synthesized and secreted
by many cell types. TIMP-1 inhibits collagenases, gelatinases, and
stromelysins. In addition to having potent inhibitory activity
against MMPs, TIMP-2 has also been effective in ameliorating
corneal ulceration and perforation in rabbits after an alkali
injury.41
GROWTH FACTORS
Endogenous peptide growth factors such as EGF, transforming
growth factor alpha (TGF-a), transforming growth factor beta
(TGF-b), insulin-like growth factor (IGF-1), and NGF are
involved in the complex interactions among epithelial cells,
extracellular matrix proteins, and proteolytic enzymes and they
may be responsible for the development of PED or subsequent
stromal ulceration.12,42,43
PATHOLOGIC RESPONSE TO CORNEAL

WOUNDING
As discussed, the normal healing process of the corneal epithelium is quite complex, involving matrix proteins, their corresponding integrin receptors, growth factors, and numerous
proteolytic enzymes. Any deviation of the normal response to
epithelial injury as noted above can result in a PED. Unregulated
proteolytic degradation of stromal collagens and/or extracellular
matrix components may be associated with a PED and lead to
subsequent corneal ulceration.2126,5254 For example, dysregulation of the plasminogen activator/plasmin system at the leading
wound edge leads to the prolonged presence of plasmin at the
ocular surface5557 and results in compromised epithelial attachments to the subjacent FN, thereby producing a PED.19 In
addition, plasmin can cleave complement C3 to generate C3a,
a chemotactic factor for polymorphonuclear leukocytes (PMNs).42
These inflammatory cells elaborate various lysosomal hydrolytic
enzymes and further contribute to stromal melting.54 Plasmin
may also activate latent collagenases to dissolve the epithelial
basement membrane complex and lead to PED or stromal
ulceration.
Other than the biochemical basis of epithelial/stromal wound
healing, a biomechanical homeostasis between ocular surface
and the surrounding lids/adnexae is of utmost importance to
ensure proper corneal wound healing. Consequently, a wide
variety of conditions that can lead to PEDs can be divided into
the following categories: (1) defective epithelial adhesion or
abnormal proliferation and migration (limbal stem cell deciency),
(2) ocular surface inflammation, (3) neurotrophic cornea,
(4) mechanical irritations, and (5) idiopathic/hereditary ocular
surface disorders (Table 55.1).
1A. DEFECTIVE EPITHELIAL ADHESION

After successful closure of an epithelial defect over a temporary
matrix, a permanent basement membrane is produced by the
basal epithelial cells. Formation of hemidesmosomes is necessary for permanent adhesion. Therefore, any condition that causes
a prolonged disruption of basement membrane or inability to
form attachments to it will result in defective epithelial adhesion
and consequently PEDs. In cases of recurrent corneal erosions,
increased levels of MMPs have been described,35 which cause
degradation of epithelial basement membrane as well as the
anchoring brils through which the epithelium attaches to the
basement membrane. In epithelial basement membrane dystrophy, there is a reduplication of epithelial basement membrane
and defective adhesion of the epithelium to its basement membrane.58 Toxic keratopathy, from topical anesthetics or preserv-
atives, can be induced by poor adhesion caused by disruption of

hemidesmosomes and/or impairment of epithelial migration
due to various causes such as disruption of vinculin/actinmediated migration and poor cellcell adhesion from topical
anesthetics or preservatives.59,60 Several corneal degenerations
including Salzmanns nodular degeneration, corneal scarring,
and band keratopathy also can cause PEDs due to defective epithelial adhesion from absent or abnormal basement membrane.61
1B. LIMBAL STEM CELL DEFICIENCY

As discussed, the rst step in lling in an epithelial defect
involves migration of epithelial cells. In his hypothesis, Thoft
described the necessary balance between production and loss of
epithelium, however, it was not until relatively recently that the
corneoscleral limbus was identied as the site of corneal epithelial SC.6264 After loss of limbal SC, corneal epithelium may
continue to regenerate for up to 6 months, corresponding to the
time that the TACs can continue to replenish the pool of
corneal epithelium.65 However, epithelial progenitor cells are
eventually not available to supply new TACs. As a result, defective epithelial regeneration or PED may arise. More devastatingly,
without a steady supply of corneal epithelial cells, conjunctival
epithelial cells can invade the cornea and lead to an untoward
conjunctivalization of the corneal surface, a hallmark of limbal
stem cell deciency.65
2. OCULAR SURFACE INFLAMMATION

Injury to the corneal epithelium sets off a cascade of events
which is only partially understood. Among this cascade, there is
the release of several inflammatory cytokines, notably
interleukin-1 (IL-1) and tumor necrosis factor alpha (TNF-a).66
If there is a breach in the basement membrane, these cytokines
exert their effects on subjacent keratocytes leading to several
processes. The keratocytes, in response to IL-1, produce mediators
of epithelialstromal interaction such as hepatocyte growth factor
(HGF) and keratocyte growth factor (KGF).67 These growth factors
act to control the proliferation, migration, and differentiation
of the overlying epithelial cells.68 IL-1 released by damaged
epithelial cells also promotes production of various proteases69
and inflammatory cell chemotactic factors70 to help with
remodeling of the corneal stroma. As mentioned above, these
proteases (including MMPs and plasminogen activators) break
down the epithelialmatrix interactions or the substrates
required (FN), leading to PED. Under normal circumstances,
once the epithelium has healed and IL-1 and TNF-a have
abated, the inflammatory processes will cease. However, this is
not always the case under pathological conditions. In inflammatory conditions such as rosacea, chemical burns, infectious
keratitis, and others, there is an increase in levels of various
proteolytic enzymes as discussed.21,55 Although multifactorial,
there is mounting evidence that keratitis from dry eye syndromes
is largely due to an induced decrease in antiinflammatory cytokines and an increase in proinflammatory cytokines (such as
IL-1 and TNF-a) as well as proteolytic enzymes.66 As a result,
various inflammatory conditions may lead to PED as a result
of interruption of the normal corneal healing processes by
inflammatory cytokines.
CHAPTER 55
EGF is produced by the lacrimal glands and secreted into tear

fluids.4345 In vitro, EGF can stimulate the DNA synthesis of
epithelial cells and stromal broblasts, as well as the synthesis
of FN by epithelial cells. It can enhance activities of the FN
receptors,46 and act as a chemotactic factor for human corneal
epithelial and stromal cells while not directly promoting the
epithelial migration.46,47 TGF-a is crucial for inducing the
synthesis of extracellular matrix components after corneal
wounding and can modulate the effects of EGF.47,48 In the presence
of substance P, a neuropeptide, EGF synergistically enhances
epithelial migration, possibly via the tyrosine kinase pathway.49,50
Substance P and IGF-1 together can enhance corneal epithelial
attachment to FN and corneal epithelial migration.51 In vitro,
NGF has been found to be produced and released by human and
rat epithelial cells. Corneal epithelial cells also express NGF
receptors. In vivo, antibody neutralization of NGF results in
delayed corneal epithelial healing.12
3. NEUROTROPHIC CORNEAS
It is well established that corneal innervation is necessary for
normal epithelial homeostasis and wound healing, in addition
to detection of sensory input. A bidirectional control of epithelial
proliferation has been proposed with sensory neuromediators
promoting epithelial mitosis and sympathetic neuromediators
751
TABLE 55.1. Principal Causes of PED

Etiology
Epithelial/limbal
Common Disease Entities
Management
EBMD*
Preservative free tears or medications
Recurrent erosions
Topical lubricants (gels/ointments)
Posttraumatic scar
Punctal occlusion
Salzmanns nodular degeneration
PTK
Band keratopathy
Bandage soft contact lens
Bullous keratopathy
Amniotic membrane graft
Toxic medicamentosa
Limbal stem cell transplant
Malnutrition (vitamin A deciency)

Limbal stem cell deciency
Inflammatory
Keratoconjunctivitis sicca
Oral tetracyclines
Ocular rosacea
Punctal occlusion
Chemical/thermal injury
Topical corticosteroid
Postinfectious keratitis
Topical cyclosporine A
Autoimmune disorders
Topical medroxyprogesterone
Sjgren syndrome
Amniotic membrane graft
Mucous membrane pemphigoid
Systemic immunosuppression
Conjunctival resection
Graft vs host disease

Peripheral ulcerative keratitis
Moorens ulcer
Neurotrophic
Diabetes mellitus
Punctal occlusion
Herpes simplex
Tarsorrhaphy
Herpes zoster
Conjunctival flap
RileyDay syndrome
Autologous serum drops
Anesthetic abuse
NGF
Postradiation
Postkeratoplasty
SECTION 6
Mechanical
Entropion/ectropion
Bandage soft contact lens
Lagophthalmos
Oculoplastic surgeries
Trichiasis
Botulinum toxin
Blepharospasm
Pseudomembranes/tarsal scar
Trachoma
Factitious
Idiopathic
Aniridia
Treat as limbal stem cell deciency
Corneal stromal dystrophies
Treat as recurrent erosions, PTK, or corneal transplant
*EBMD, epithelial basement membrane dystrophy.
Corticosteroids should be used judiciously under close observation with PED.
752
reducing mitosis.71 As a result, conditions such as diabetes

mellitus and herpes zoster ophthalmicus which lead to sensory
denervation would be expected to lead to an imbalance and
potentially PEDs. After denervation, there is a decrease in production of several constitutively expressed neuropeptides that
are usually increased in wounded corneas such as substance P,
calcitonin gene-related peptide (CGRP), and several neurotrophins (including NGF).72 Substance P has been shown to increase
epithelial adhesion to FN through upregulation of integrin.51

Substance P also seems to have effects on epithelial proliferation and migration, and the effects are synergistic with EGF and
IGF.51 CGRP may also play a role in epithelial proliferation and
migration73 but its role is less clear. NGF is a neurotrophin that
has been shown to promote recovery of nerves after injury,74
allowing reestablishment of substance P production. In addition, NGF appears to have direct effects on corneal epithelium

by enhancing epithelial proliferation and migration.75 Loss of
the expression of these factors can impair the ability to heal a
wounded cornea, leading to PEDs.
4. MECHANICAL IRRITATIONS
When there is persistent irritation of the ocular surface, as can
be seen with abnormal lid pathologies such as trichiasis, lid
mal-positioning, blepharospasm, foreign body, or other causes
there may be an increase in epithelial turnover. In many cases,
the underlying problem (such as after zoster infection or
neurosurgical intervention) and a dry or inflammatory ocular
surface (erythema multiforme, mucous membrane pemphigoid,
trachoma) may cause not only eyelid abnormality but cause a
neurotrophic cornea. The mechanical factors causing epithelial
trauma may lead to focal or diffuse limbal stem cell attrition. In
any case, if the increase in loss of epithelium exceeds the ability
to heal, a PED will ensue.
5. IDIOPATHIC/HEREDITARY OCULAR
SURFACE DISORDERS
A vast array of hereditary or idiopathic disorders can cause
delayed healing of epithelial defects by an equally vast array of
pathogeneses. For instance, PEDs in RileyDay syndrome are a
predominantly neurotrophic etiology, while corneal stromal and
epithelial basement membrane dystrophies belong to the category of aberrant epithelial adhesion due to abnormal basement
membrane and adhesion complexes. PEDs in aniridia are largely
due to deciency of limbal SC. In general, most conditions in
this category also fall into one of the above categories and can
be addressed as such.
FIGURE 55.1. A PED with irregular epithelial borders and without

stromal inltrate was noted in a patient with systemic vitamin A
deciency due to chronic alcoholism. There were deep stromal folds
and corneal edema. Peripheral corneal vascularization was also
noted. The PED was refractory to antibiotics and corticosteroids. The
PED and night blindness resolved after using systemic vitamin A and
nutrition supplements.
As listed in Table 55.1, the causes of PEDs and related stromal

ulceration are diverse. Ideally each case of PED could be attributed to a defect in normal epithelial response to injury, or the
perpetuation of a pathologic response to injury as outlined above.
However, our current understanding of these conditions does
not always allow us to do so. PEDs after herpetic keratitis can
de due to at least two of the listed categories, namely neurotrophic and inflammatory. Similarly, PEDs after alkali burns can
be caused by limbal deciency or intense inflammation. In addition, many of the conditions likely cause PEDs by disturbing
multiple steps of normal epithelial reaction to injury. Therefore,
effective treatment algorithms rely on a stepwise approach and
specic targeting of underlying etiologies (Table 55.1).
Determining the etiology of a PED should begin by obtaining
a thorough patient history and performing careful examination.
Previous ocular surgery, infection, and trauma are important in
establishing a diagnosis. Topical ophthalmic medications and
preservatives must be considered as potential toxins to the
ocular surface, and should be discontinued if possible. Diabetes
mellitus, malnutrition (Fig. 55.1), autoimmune disorders, and
other systemic ailments need to be managed appropriately as
they are often associated with delayed healing of the corneal
epithelium. Many oral medications have anticholinergic and
antihistamine properties which may disrupt the lacrimal
functional unit.
Prudent external observation of the patient may reveal a
seventh nerve palsy, subtle lagophthalmos, incomplete blinking,
or blepharospasm. The eyelids should be examined for structural
abnormalities, blepharitis, rosacea, and meibomian gland dysfunction. The tarsal and bulbar conjunctiva should be inspected
for mechanical factors such as pseudomembranes or scarring
causing chronic irritation to the ocular surface. Corneal sensa-
FIGURE 55.2 A central corneal epithelial defect with underlying

stromal opacity was noted in a patient after herpes zoster
ophthalmicus with an anesthetic cornea. The PED was noted to have
slightly raised epithelial edges, characteristic of neurotrophic epithelial
defects. The lesion failed to respond to medical treatment and
eventually required tarsorrhaphy to facilitate the epithelial wound
healing.
tion should be checked before anesthetic instillation, especially

if herpes simplex or herpes zoster (Fig. 55.2) is suspected. Finally,
slit lamp biomicroscopy of the ocular surface may provide clues
to the etiology and management of the disease. Once diagnosis
is obtained, appropriate targeted treatment can be administered.
In addition to correcting the underlying disorder, PED is principally treated with ocular surface lubrication. We recommend
frequent use of preservative-free articial lubricants to avoid
potential toxicity from preservatives. Patients and their caregivers should be instructed on how and when to instill eye
drops, and the physician must realize that compliance is often
a problem. Management of PED is often frustrating for the
patient and challenging for the physician due to frequent
CHAPTER 55
TREATMENT FOR PED
753

treatment failures and recurrences. Various medical and
surgical options are discussed below.
NONSURGICAL MANAGEMENT
doses of doxycycline (20 mg twice a day) may be just as effective

as standard dosing85 if side effects are problematic. Topical tetracycline can be formulated in a 1% suspension or 3% ointment
for local treatment.
Contact Lens
Antiinflammatory Agents
Extended-wear therapeutic contact lenses can serve as a bandage

to protect the healing epithelium from mechanical trauma from
the eyelids. This treatment is very useful for PED when palpebral
conjunctival scarring or lid marginal abnormalities are present.
Bandage lenses may also protect a fragile epithelium from
sloughing off, as seen in recurrent erosion disorders. Finally
bandage lenses may provide signicant ocular surface comfort
for the patient with PED.
Therapeutic soft contact lenses usually have to be worn for at
least 2 weeks and sometimes up to 3 months to ensure healing.
A lens with high oxygen permeability (Dk/L) should be chosen
to minimize potential corneal complications. An extended-wear
soft contact lens should have a Dk/L of at least 87 109 cm2 mL
O2/s mL mmHg to avoid corneal edema and neovascularization.76
Silicone matrix hydrogel thin lenses with high water content
generally provide the highest Dk/L. They can be changed on a
weekly basis, however if the PED is not closed they are probably
best left untouched as long as they are free of mucus and debris.
The risk for corneal infection in extended versus daily wear is
ve times as high,77 therefore prophylactic topical antibiotic or
frequent follow up is advisable.
Collagen shields made of porcine scleral tissue or bovine
dermis were initially developed in 1984 by Fyodorov.78 These
substrates gained popularity because the collagen was able to
support corneal epithelial cell growth in culture. They are
currently labeled for ocular surface protection following surgery,
but are also used for traumatic epithelial defect. Stored dry, they
are usually soaked in an antibiotic solution before being placed
on the cornea much like a soft contact lens. The advantage is
that the collagen increases the bioavailability of the antibiotic
(or other drug) to the cornea. However, the collagen shield
biodegrades in 1272 h, which limits its usefulness in chronic
epithelial defects. One study showed they were not helpful in
treating PED following penetrating keratoplasty.79
Gas-permeable extended-wear scleral lenses have been
reported to heal PED in some refractory cases, however there is
a high risk of microbial keratitis with this treatment.80 These
lenses also require custom tting and are difcult to insert,
thereby limiting their utility in PED.
Caution should be exercised when using antiinflammatory

therapy to inhibit migration of inflammatory cells and limit
tissue inflammation. Topical corticosteroids may have a role in
PED therapy only when there is concomitant ocular inflammation. Externally, this includes active mucous membrane pemphigoid, peripheral ulcerative keratitis, chemical burns, and
StevensJohnson syndrome (erythema multiforme). The use of
topical corticosteroids to limit inflammation for alkali burns
remains controversial. Corticosteroids work by inhibiting phospholipase A2, an enzyme at the beginning of the inflammatory
cascade. Patients should be warned about possible cataract and
glaucoma complications. Corticosteroids should be used judiciously with frequent follow-up, as they may increase stromal
ulceration by inhibiting collagen synthesis and cause increased
risk of microbial keratitis.
Specially compounded preservative-free topical steroids,
medroxyprogesterone 1% and methylprednisolone 1% may
prevent stromal lysis, but must be kept refrigerated to avoid
contamination. Medroxyprogesterone prevents stromal melting
by inhibiting local collagenases that degrade the corneal stroma,
while at the same time exhibiting a mild antiinflammatory
property.8689 Both doxycycline and corticosteroids can inhibit
MMP-9. In patients with recurrent corneal erosions unresponsive
to the conventional therapy, administration of oral doxycycline
and topical corticosteroids reduces pain and heals epithelial
defects within 210 days.83 Methylprednisolone also provides
symptomatic relief and resolution of laments in severe keratoconjunctivitis associated with Sjgren syndrome.90 These two
steroid preparations should only be considered when a PED is
associated with intense surface or intraocular inflammation.
Topical cyclosporine, mycophenolate mofetil, and tacrolimus,
are available as alternative or simultaneous antiinflammatory
treatment. Cyclosporine, most commonly used, can be found
commercially (Restasis 0.05%), but higher doses can be specially
compounded at a 0.52% concentration used 24 times per day.
Oral immunosuppressive therapy is required when there is
active systemic inflammatory disease such as rheumatoid arthritis, Wegeners granulomatosis, and recalcitrant scleritis.
Serum and Fibronectin

SECTION 6
Tetracyclines
754
Topical and systemic tetracyclines can effectively inhibit MMPs

in animal and human subjects in a mechanism independent of
their antimicrobial activity.81 High levels of MMPs cause
corneal stromal lysis via collagen degradation and injury to the
epithelial basement membrane adhesion complexes resulting in
poor epithelial adherence. MMP-9, a gelatinase in corneal
epithelial cells, has been detected at the edges of nonhealing
corneal ulcers.82 Oral doxycycline at 50 mg twice a day has been
demonstrated to inhibit MMP-9, resulting in rapid healing and
preventing recurrences of recurrent corneal erosions.83
Adjunctive treatment with systemic tetracycline 250 mg four
times a day was shown to be benecial in a series of patients
with PED.84 Concomitant rosacea or meibomian gland dysfunction should also improve with tetracycline therapy. These
antibiotics should not be used in pregnant women or children
due to the risk of permanent discoloration of teeth. In adults,
common side effects are gastrointestinal irritations, photosensitivity, and mucosal yeast infections. The typical initial dose of
doxycycline for ocular rosacea is 100 mg twice a day for 46 weeks,
then tapering to a maintenance dose of 50100 mg a day. Lower
Autologous serum drops contain neurotrophic factors which have

been shown to promote healing in PED unresponsive to conventional treatment with a success rate of 5681%.9193 These
factors which promote epithelial healing are thought to include
EGF, broblastic growth factor (FGF), NGF, FN, vitamin A,
substance P, and corneal collagenase inhibitors.91,93 The drops
are easily prepared by drawing 3050 mL of blood from the
patient and centrifuging the tube to separate the blood components. The serum is removed in a sterile environment and
mixed with balanced salt solution to a concentration of 20%
with or without antibiotics. The drops must be kept refrigerated
to avoid contamination. Umbilical cord serum, also thought to
contain several of these important epithelial promoters, may also
be used. One prospective randomized controlled clinical study
demonstrated that human umbilical cord 20% serum had a faster
healing rate than autologous serum in patients with PED.94
FN is a glycoprotein found in serum that helps cellular adhesion and binds collagen during wound healing. Small case series
have reported success with use of topical FN for resolution of
PED.95,96 However, a moderate-size randomized double-blinded
control trial in 1995 showed that topical FN was no better than

vehicle or placebo in decreasing epithelial defect size after
21 days of treatment.97 Due to conflicting reports regarding its
efcacy, exogenous pFN is no longer used widely for PEDs.
Lecithin-bound superoxide dismutase reduces potentially toxic
free oxygen radicals produced by neutrophils by converting superoxide radicals to hydrogen peroxide. This topical medication at
a concentration of 0.1% was shown to promote epithelial
healing in chronic noninfectious inflammatory corneal ulcers.98
tive approach in some patients with PED from exposure keratopathy.111 The neurotoxin works by inhibiting the release of
acetylcholine at the neuromuscular junction to induce clinical
paralysis for ~3 months. This is not the procedure of choice for
a monocular patient as the induced ptosis will obstruct the
visual axis or for younger children with amblyogenic potential.
Persistent blepharospasm, which can also cause PED, can be
relieved by botulinum toxin injection into the obicularis oculi
muscle.
Growth Factors
SURGICAL MANAGEMENT
Punctal Occlusion
Punctal occlusion via plugs is a quick ofce procedure that will
rapidly facilitate increased surface hydration and decrease tear
osmolarity.107,108 Absorbable (i.e., collagen) and nonabsorbable
(i.e., silicone) plugs exist which can be placed in the lower
and/or upper puncta. Silcone plugs are most often used for dry
eye syndrome, and have an 8186% success rate of improved
subjective symptoms and decreased staining in patients.109,110
Success was only 40% for PED; however, there were only ve
patients with that diagnosis in the study.110 The most common
complication from silicone plugs is extrusion seen ~50% of the
time, with even higher rates of extrusion on second or third placement.109,110 Other potential complications include epiphora,
subconjunctival hemorrhage, conjunctival erosion, pygogenic
granuloma, and migration into the canaliculus. If punctal plugs
are found to be benecial, the puncta may also be permanently
closed by electrocautery, thermocautery, or argon laser to avoid
many of the potential complications.
Botulinum Toxin
Injection of a small amount of botulinum toxin (510 units)
into the upper eyelid to paralyze the levator muscle is an effec-
Tarsorrhaphy
Partial or complete tarsorrhapy is a low-risk surgical option for
PED. It has been shown to be very benecial in postkeratoplasy
patients112 and is useful in cases of chronic neurotrophic and
neuroparalytic keratitis. Noncompliant and debilitated patients
who cannot reliably put in therapeutic topical medications
perhaps prot from this intervention the most. Techniques vary,
but care must be taken to avoid suture chang the cornea epithelium and iatrogenic trichiasis when a temporary or permanent
tarsorrhaphy is constructed. One small randomized prospective
study showed there was no statistical difference in epithelialization between lateral tarsorrhaphy and amniotic membrane
transplantation for PED.113
Amniotic Membrane
The preserved amniotic membrane is nonantigenic, and is composed of a single epithelial cell layer, a basement membrane, and
an avascular stroma. The thick basement membrane is very
similar to the epithelial basement membrane and acts as a matrix
substrate for epithelial adhesion and migration. Several series of
reports have shown that amniotic membrane transplantation
(AMT) can be a successful surgical treatment for PED refractory
to medical treatment.114117 Typically, the fresh or frozen amniotic
tissue is grafted with stromal side down over the PED with 10-0
nylon suture or brin adhesive glue (Tisseal glue). For PED with
stromal thinning or minute perforation, multilayer amniotic
membrane grafting can be performed.118 In this procedure, the
membrane is cut into several pieces and layered on top of each
other or folded in multiple layers, to match the ulcer bed. Only
the top layer is sutured to the host. Often a lateral tarsorrhaphy
is performed simultaneously or a bandage soft contact lens is
used until complete reepithelialization.
The amniotic membrane can also be used as a patch, in
which the basement membrane side is oriented down.119 This
technique may be useful for PED without ulceration, as the
membrane consistently dissolves or is easily removed after epithelialization. Amniotic membrane patching may permit better
visual clarity compared to grafting.
Successful epithelialization with AMT has ranged from 73%
to 82% in these obstinate cases of PED.113,115,119 The tissue
works because the basement membrane acts as a scaffold to
support epithelial adhesion, growth, and differentiation.120,121
The amniotic membrane suppresses TGF-b signaling and may
release important epithelial growth factors as well.122 In addition, the avascular stroma contains various protease inhibitors
which reduce ocular surface inflammation and vascularization,
and may restore corneal stromal thickness when ulceration is
concomitantly present.
CHAPTER 55
Application of growth factors to promote corneal wound healing

and to treat PED or stromal ulcers has been of major interest in
ophthalmology. Although numerous growth factors have been
characterized and puried, only EGF, NGF, and IGF-1 have been
investigated clinically for the corneal epithelial wound healing.
EGF stimulates mitosis of corneal epithelial cells in vitro and
in vivo. It can also enhance epithelial migration, stimulate wound
healing, and improve stromal wound strength.47,55,99 Both mousederived EGF (mEGF) and recombinant human EGF (hEGF)
have been studied experimentally. In rabbit corneas with alkali
burn, mEGF enhances epithelial wound healing; however, it does
not prevent recurrent erosions and subsequent epithelial breakdown.100 In humans, mEGF signicantly accelerates the epithelial healing of various nondystrophic corneal diseases.101,102 The
therapeutic effect of mEGF seems to be inversely related to the
magnitude of stromal damage. There is a signicant homology
between mouse and human EGF with the same receptors and
identical biologic activity for both forms.101 In rabbit corneas after
anterior keratectomy and alkali burns, hEGF accelerates corneal
reepithelialization. A double-masked multicenter clinical trial
showed that recombinant hEGF accelerated the healing of traumatic epithelial defects when compared with control treatment.103
NGF is a polypeptide that is important for neuronal health
and stability. A study showed that NGF 200 mg/ml for 2 weeks
completely resolved a PED in 45 consecutive eyes with neurotrophic keratitis unresponsive to other medical therapies.104,105
Some patients in the study had improved corneal sensitivity,
suggesting that NGF restored function of injured neurons. The
exact mechanism of NGF is not well understood.
Topical administration of substance P and IGF-1 to a patient
with PED results in complete epithelial healing within 1 week
with nonrecurrence in the following 8 months.106
Conjunctival Flap
While cosmesis may be superior with AMT due to increased
clarity and reduction of vascularization, partial or full
(Gundersen flap) conjunctival autografting is another surgical
option for PED.123126 A thin free or rotational conjunctival flap
without Tenons capsule is dissected. Meticulous care must be
taken not to buttonhole the harvested graft. The remaining corneal
epithelium must be removed entirely to prevent postoperative
755

cyst formation. The flap is sutured over the defect. The
most common complication (11%) is conjunctival retraction
requiring surgical revision.126 This procedure should not be
considered in patients with active conjunctival disease, with
functioning ltering bleb, or if future glaucoma ltering surgery
is planned.
Cyanoacrylate Glue
As discussed, stromal ulceration, descemetocele, and perforation
may accompany a chronic PED. Tissue adhesive (cyanoacrylate
glue) application to the thinned or perforated cornea frequently
negates the need for emergency tectonic corneal transplantation. The glue will polymerize immediately upon contact with
water, so care must be taken to distribute the glue evenly in the
desired area while at the slit lamp or under operating microscope. A bandage soft contact lens is placed over the glue patch
for patient comfort (Fig. 55.3). The glue may induce stromal
neovascularization, which will help prevent further ulceration.127
The glue will extrude or can be removed once epithelialization
occurs, and optical corneal transplantation can be considered.
Phototherapeutic Keratectomy
As in treatment for recurrent erosions, the excimer laser may be
effective in surgically managing PEDs unresponsive to conventional treatment by laser ablation of the basement membrane
and supercial Bowmans layer.128,129 The exact mechanism by
which phototherapeutic keratectomy (PTK) exerts its effect
remains unclear; however, it probably modies the basement
membrane and Bowmans layer to facilitate stronger cell anchorage.128 In a randomized prospective trial comparing epithelial
removal only with epithelial removal followed by excimer laser
ablation, those patients undergoing PTK had fewer recurrences
and better symptomatic relief.130 Instead of LASIK, PTK is the
preferred procedure for patients with recurrent corneal erosions
or corneal stromal dystrophies, since the treatment may correct
the refractive error and epithelial pathologies simultaneously.
In summary, treatment failure or recurrence of PED may occur
after enduring effort with these medical and surgical therapies.
FIGURE 55.3 A patient with severe Sjgren syndrome had developed

large corneal perforation and necessitated patch therapeutic
keratoplasty. A PED recurred on the patch graft and led to a crescentshaped corneal perforation. Tissue adhesive was applied to seal the
perforation and a bandage contact lens was placed over the patch
graft and adhesive to facilitate healing and reduce discomfort.
An intact healthy epithelial layer is crucial in preventing ocular

infection and progressive stromal ulceration. It is also vital for
the survival of optical corneal transplantation. Limbal stem cell
deciency may be the underlying pathology present in these
refractory cases; if so, stem cell transplantation should be considered. The above treatment strategies can be employed while
nurturing the grafted tissue. Keratoprosthetic surgery may also
be considered in chronic ocular surface disease that is nonrespondent to transplantation.
SECTION 6
REFERENCES
756

of corneal epithelial maintenance. Invest
2. Hanna C, Bicknell DS, OBrien JE: Cell
turnover in the adult human eye. Arch
Ophthalmol 1961; 65:695698.
3. Estil S, Primo EJ, Wilson G: Apoptosis in
shed human corneal cells. Invest
4. Gipson I, Spurr-Michaud S, Tisdale A:
Hemidesmosomes and anchoring bril
collagen appear synchronously during
development and wound healing. Dev Biol
1988; 126:253262.
5. Khodadoust AA, Silverstein AM, Kenyon DR,
corneal epithelium. The role of basement
65:339348.
6. Tuft SJ, Gartry DS, Rawe IM, Meek KM:
Photorefractive keratotomy: implications of
corneal wound healing. Br J Ophthalmol
1993; 77:243247.
7. Gipson IK, Watanabe HI, Zieske JD:
Corneal wound healing and bronectin. Int
8. Zieske JD, Bukusoglu G, Gipson IK:
Enhancement of vinculin synthesis by
9.
10.
11.
12.
13.
14.
migrating stratied squamous epithelium.

J Cell Biol 1989; 109:571576.
Zieske JD, Bukusoglu G: Effect of
protease inhibitors on corneal epithelial
migration. Invest Ophthalmol Vis Sci 1991;
32:20732078.
Murakami J, Nishida T, Otori T: Coordinated
appearance of beta 1 integrins and
bronectin during corneal wound healing.
J Lab Clin Med 1992; 120:8693.
Hongo M, Itoi M, Yamaguchi N, Imanishi J:
Distribution of epidermal growth factor
(EGF) receptors in rabbit corneal epithelial
cells, keratocytes and endothelial cells,
and the changes induced by transforming
growth factor-b1. Exp Eye Res 1992;
54:916.
Lambiase A, Manni L, Bonini S, et al:
Nerve growth factor promotes corneal
healing: structural, biochemical, and
molecular analyses of rat and human
41:10631069.
Chung E, DeGregorio PG, Wasson M,
Zieske J: Epithelial regeneration after
limbus-to-limbus debridement. Invest
Murakami J, Morimoto K, Nishida T, Otori T:
15.
16.
17.
18.
19.
20.
Movement of corneal epithelium of rats in

situ observed by time-lapse
cinematography. Curr Aspects Ophthalmol
1991; 1:315319.
Nishida T, Nakagawa S, Nishibayashi C,
et al: Fibronectin enhancement of corneal
epithelial wound healing in rabbits in vivo.
Ponti A, Machacek M, Gupton SL, et al:
Two distinct actin networks drive the
protrusion of migrating cells. Science 2004;
305:17821786.
Cameron JD, Hagen ST, Watereld RR,
Furcht LT: Effects of matrix proteins on
corneal epithelial cell adhesion and
migration. Curr Eye Res 1998; 7:293301.
Lampe PD, Nguyen BP, Gil S, et al: Cellular
interaction of integrin a3B1 with laminin 5
promotes gap junctional communication.
J Cell Biol 1998; 143:17351747.
Berman M: The pathogenesis of corneal
epithelial defects. Acta Ophthalmol Suppl
1989; 192:5564.
Berman M: Collagenase and corneal
ulceration. In: Woolley DE, Evanson JM,
eds. Collagenase in normal and
pathological connective tissues.
Chichester, UK: Wiley; 1980.
21. Hayashi K, Berman M, Smith D, et al:

Pathogenesis of corneal epithelial defects:
role of plasminogen activator. Curr Eye Res
1991; 10:381398.
22. Berta A, Tozser T, Holly FJ: Determination
of plasminogen activator activities in
normal and pathological human tears. The
signicance of tear plasminogen activators
in the inflammatory and traumatic lesions
of the cornea and the conjunctiva. Acta
Ophthalmol 1990; 68:508514.
23. Tripathi RC, Tripathi BJ, Park JK:
Localization of urokinase-type
plasminogen activation in human eyes an
immunocytochemical study. Exp Eye Res
1990; 51:545552.
24. Vaheri A, Tapiovaara H, Myohanen H, Bitik J:
Regulation of the pericellular activation of
plasminogen and its role in tissuedestructive processes. Acta Ophthalmol
Suppl 1992; 202:3941.
25. Morimoto K, Mishima H, Nishida T, Otori T:
Role of urokinase type plasminogen
activator (u-PA) in corneal epithelial
migration. Thromb Haemost 1993;
69:387391.
26. Wang HM, Berman M, Law M: Latent and
active plasminogen activator in corneal
ulceration. Invest Ophthalmol Vis Sci 7985;
26:511524.
27. Conn H, Berman M, Kenyon K, et al:
Stromal vascularization prevents corneal
ulceration. Invest Ophthalmol Vis Sci 1980;
19:362370.
28. Berman M, Winthrop S, Ausprunk D, et al:
Plasminogen activator (urokinase) cause
vascularization of the cornea. Invest
29. Kruithot EK: Plasminogen activator
inhibitors a review. Enzyme 1988;
40:113121.
30. Tervo T, van Setton GB: Aprotinin for
inhibition of plasmin on the ocular surface:
principles and clinical observations. In:
Beuerman W, Kaufman HE, eds. Healing
processes of the cornea. Woodlands, TX:
Portfolio Publishing; 1989.
31. Gross J: An essay on biological
degradation of collagen. In: Hay ED (ed)
Cell biology of the extracellular matrix.
New York: Plenum Press; 1981.
32. Wagoner MD, Kenyon KR: Distribution of
collagenase and cell types in sterile
ulceration of human corneal grafts. Acta
33. Fini ME, Girard MT, Matsubara M:
Collagenolytic/gelatinolytic enzymes in
corneal wound treating. Acta Ophthamol
Suppl 1992; 202:2633.
34. Matsubara M, Zieske JD, Fini ME:
Mechanism of basement membrane
dissolution preceding corneal ulceration.
32:32213237.
35. Garrana RM, Zieske JD, Assouline M,
Gipson IK: Matrix metalloproteinases in
epithelia from human recurrent corneal
erosion. Invest Ophthalmol Vis Sci 1999;
40:12661270.
36. Pster RR, Burstein N: The alkali burned
cornea 1. Epithelial and stromal repair. Exp
Eye Res 1976; 23:519535.
37. Matsuda H, Smelser GK: Epithelium and
storms in alkali-burned corneas. Arch
Ophthalmol 1973; 89:396401.
38. Berrman M, Leary R, Gage J: Evidence for
a role of the plasminogen activator-plasmin
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
system in cortical ulceration. Invest

Berman M: The pathogenesis of epithelial
defects and stromal ulceration. In:
Cavanaugh D, ed. The cornea:
transactions of the Wall Congress on the
Cornea III. New York: Raven; 1988.
Chin JR, Murphy G, Werb Z: Stromelysin,
a connective-tissue degrading
metalloproteinase secreted by stimulated
rabbit synovial broblasts in parallel with
collagenase. Biosynthesis, isolation,
characterization, and substrates. J Biol
Chem 260:1236712376, 1985.
Murphy G, Allan JA, Willenbrock F, et al:
The N-terminal domain of tissue inhibitor of
metalloproteinases retains metalloproteinase
inhibitory activity. Biochemistry 1991;
30:80978102. (Erratum appears in
Biochemistry 1991; 30:10362.)
Murphy CJ, Marfurt CF, McDennott A, et al:
Spontaneous chronic corneal epithelial
detects (SCCED) in dogs: clinical features,
innervation, and effect of topical SP, with or
without IGF-1. Invest Ophthalmol Vis Sci
2001; 42:22522261.
van Setten GB, Tervo K, Virtarer I, et al:
Immunohistochemical demonstration of
epidermal growth factor in the lacrimal and
submandibular glands of rats. Acta
Ophthalmol 1990; 68:477480.
van Setten GB, Viinika L, Tervo T, et al:
Epidermal growth factor is a constant
component of normal human tear fluid.
227:184187.
Ohashi Y, Motokura M, Kinoshita Y, et al:
Presence of epidermal growth factor in
human tear. Invest Ophthalmol Vis Sci
1989; 30:18791882.
Nishida T, Nakamura M, Murakami J, et al:
Epidermal growth factor stimulates corneal
epithelial cell attachment to bronectin
through a bronectin receptor system.
33:24642469.
Schultz G, Chegini N, Grant M, et al:
Effects of growth factors on corneal wound
healing. Acta Ophthalmol Suppl 1992;
202:6066.
Mishima H, Nakamura M, Murakami J,
et at: Transforming growth factor-beta
modulates effects of epidermal growth
factor on corneal epithelial cells. Curr Eye
Res 1992; 11:691696.
Ofuji K, Nakamura M, Nishida T: Signaling
regulation for synergistic effects of
substance P and insulin-like growth
factor-1 or epidermal growth factor on
corneal epithelial migration. Jpn J
Nakamura M, Nishida T, Ofujik K, et al:
Synergistic effect of substance P with
epidermal growth factor on epithelial
migration in rabbit cornea. Exp Eye Res
1997; 65:321329.
Nakamura M, Chikama T, Nishida T:
Up-regulation of integrin alpha 5
expression by combination of substance P
and insulin-like growth factor-1 in rabbit
corneal epithelial cells. Biochem Biophys
Res Commun 1998; 246:777782.
Kessler E, Mondino BJ, Brown SI: The
corneal response to Pseudomonas
aeruginosa: histopathological and
enzymatic characterization. Invest
53. Berman M: Regulation of collagenase.

Therapeutic considerations. Trans
54. Kenyon KR: Inflammatory mechanism in
corneal ulceration. Trans Am Ophthalmol
Soc 1985; 83:610663.
55. Tervo T, et al: Wound healing of the ocular
surface. Ann Med 1992; 24:1927.
56. Tervo T, Tervo K, van Setten GB, et al:
Plasminogen activator and its inhibitor in
the experimental corneal wound. Exp Eye
Res 1989; 48:445449.
57. Dano K, et al: Plasminogen activators,
tissue degradation, and cancer. Adv
Cancer Res 1985; 44:139266.
58. Fogle JA, Kenyon KR, Stark WR, Green WR:
corneal dystrophies. Am J Ophthalmol
1975; 79:925940.
59. Dass BA, Soong HK, Lee B: Effects of
proparacaine on actin cytoskeleton of
corneal epithelium. J Ocular Pharmacol
1988; 4:194197.
60. Burstein NL: Corneal cytotoxicity of
topically applied drugs, vehicles and
preservatives. Surv Ophthalmol 1980;
25:1530.
61. Kenyon KR: Recurrent corneal erosion:
pathogenesis and therapy. Int Ophthalmol
Clin 1979; 19:169195.
Differentiation-related expression of a
major 64D corneal keratin in vivo and in
culture suggests limbal location of corneal
epithelial stem cells. J Cell Biol 1986;
103:4962.
63. Ebato B, Friend J, Thoft RA: Comparison
of central and peripheral human corneal
57:201209.
65. Huang AJ, Tseng SC: Corneal epithelial
32:96105.
66. Pflugfelder S, Jones C, Ji Z, et al: Altered
cytokine balance in the tear fluid and
conjunctiva of patients with Sjogrens
syndrome keratoconjuncitivitis sicca.
Curr Eye Res 1999; 19:201211.
67. Weng J, Mohan RR, Li Q, Wilson SE: IL-1
upregulates keratinocyte growth factor
and hepatocyte growth factor mRNA and
protein production by cultured stromal
broblast cells: interleukin-1 beta
expression in the cornea. Cornea 1996;
15:465471.
68. Wilson SE, He YG, Weng J, et al: Effect of
epidermal growth factor, hepatocyte growth
factor and keratinocyte growth factor, on
proliferation, motility, and differentiation of
human corneal epithelial cells. Exp Eye Res
1994; 59:665678.
69. West-Mays JA, Strissel KJ, Sadow PM,
Fini ME: Competence for collagenase gene
expression by tissue broblasts requires
activation of an interleukin 1 alpha
autocrine loop. Proc Natl Acad Sci 1995;
92:67686772.
70. Tran MT, Tellaetxe-Isusi M, Elner V, et al:
Proinflammatory cytokines induce RANTES
and MCP-1 synthesis in human corneal
CHAPTER 55
757
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
SECTION 6
82.
83.
84.
85.
86.
87.
758
keratocytes but not in corneal epithelial

cells. Beta-chemokine synthesis in corneal
37:987996.
Cavanagh HD, Colley AM: The molecular
Ophthalmol 1989; 192:115134.
Mishima S: The effects of the denervation
and the stimulation of the sympathetic and
trigeminal nerve on the mitotic rate of the
corneal epithelium in the rabbit. Jpn J
Mikulec AA, Tanelian DL: CGRP increases
rate of corneal re-epithelialization in an in
vitro whole mount preparation. J Ocul
Donnerer J, Amann R, Schuligoi R,
Skotsch G: Complete recovery by nerve
growth factor of neuropeptide content and
function in capsaicin-impaired sensory
neurons. Brain Res 1996; 741:103108.
Kruse FE, Tseng SC: Growth factors
modulate clonal growth and differentiation
of cultured rabbit limbal and corneal
34:19631976.
Holden BA, Mertz GW: Critical oxygen
levels to avoid corneal edema for daily and
extended wear contact lenses. Invest
Poggio EC, Glynn RJ, Schein OD, et al: The
incidence of ulcerative keratitis among
users of daily-wear and extended-wear soft
contact lenses. N Engl J Med 1989;
321:779783.
Fyodorov SN, Ivashina AI, Bagrov SN:
Efciency of collagen covers: application in
cases in keratotomy. In: Fydorov, SN, ed.
Eye microsurgery. Moscow: Research
Institute of Eye Microsurgery; 1984.
Groden LR, White W: Porcine collagen
corneal shield treatment of persistent
epithelial defects following penetrating
keratoplasty. CLAO J 1990; 16:9597.
Rosenthal P, Cotter JM, Baum J: Treatment
of persistent corneal epithelial defect with
extended wear of a fluid-ventilated gaspermeable scleral contact lens. Am J
Ophthalmol 2000; 130:3341.
Ralph RA: Tetracyclines and the treatment
of corneal stromal ulceration: a review.
Cornea 2000; 19:274277.
Fini ME, Parks WC, Rinehart WB: Role of
matrix metalloproteinases in failure to
re-epithelialize after corneal injury. Am J
Pathol 1996; 149:12871302.
Dursun D, Kim MC, Solomon A, Pflugfelder
SC: Treatment of recalcitrant recurrent
corneal erosions with inhibitors of matrix
metalloproteinase-9, doxycycline, and
corticosteroids. Am J Ophthalmol 2001;
132:813.
Perry HD, Kenyon KR, Lamberts DW, et al:
Systemic tetracycline hydrochloride as
adjunctive therapy in the treatment of
persistent epithelial defects.
Yoo SE, Lee DC, Chang MH: The effect of
low-dose doxycycline therapy in chronic
meibomian gland dysfunction. Korean J
Ophthalmol 2005; 19:258263.
Dohlman CH, Doane MG: Some factors
influencing outcome after keratoprosthesis
surgery. Cornea 1994; 13:214218.
Crabb C: Endocrine influences on
ulceration and regeneration in the alkaliburned cornea. Arch Ophthalmol 1977;
95:18661870.
88. Newsome D, Gross J: Prevention by

medroxyprogesterone of perforation in the
alkali-burned rabbit cornea: inhibition of
collagenolytic activity. Invest Ophthalmol
Vis Sci 1977; 16:2131.
89. Hicks CR, Crawford GJ: Melting after
keratoprosthesis implantation: the effects
of medroxyprogesterone. Cornea 2003;
22:497500.
90. Marsh P, Pflugfelder SC: Topical
nonpreserved methylprednisolone therapy
for keratoconjunctivitis sicca in Sjogren
syndrome. Ophthalmology 1999;
106:811816.
91. Tsubota K, Goto E, Shimmura S,
Shimazaki J: Treatment of persistent
corneal epithelial defect by autologous
serum application. Ophthalmology 1999;
106:19841989.
92. Ferreira de Souza R, Kruse FE, Seitz B:
Autologous serum for otherwise therapy
resistant corneal epithelial defects
prospective report on the rst 70 eyes. Klin
93. Poon AC, Geerling G, Dart JK, et al:
Autologous serum eyedrops for dry eyes
and epithelial defects: clinical and in vitro
toxicity studies. Br J Ophthalmol 2001;
85:11881197.
94. Vajpayee RB, Mukerji N, Tandon R, et al:
Evaluation of umbilical cord serum therapy
for persistent corneal epithelial defects.
Br J Ophthalmol 2003; 87:13121316.
95. Nishida T, Ohashi Y, Awata T, Manabe R:
Fibronectin a new therapy for corneal
trophic ulcer. Arch Ophthalmol 1983;
101:10461048.
96. Phan TM, Foster CS, Boruchoff SA, et al:
Topical bronectin in the treatment of
persistent corneal epithelial defects and
trophic ulcers. Am J Ophthalmol 1987;
104:494501.
97. Gordon JF, Johnson P, Musch DC: Topical
bronectin ophthalmic solution in the
treatment of persistent defects of the
corneal epithelium. Chiron Vision
Fibronectin Study Group. Am J Ophthalmol
1995; 119:281287.
98. Shimmura S, Igarashi R, Yaguchi H, et al:
Lecithin-bound superoxide dismutase in
the treatment of noninfectious corneal
ulcers. Am J Ophthalmol 2003;
135:613619.
99. Feldman ST: The effect of epidermal
growth factor on corneal wound healing:
practical considerations for therapeutic
use. Refract Corneal Surg 1991;
7:232239.
100. Phan TM, Foster CS, Shaw CD, et al:
Topical bronectin in an alkali burn model
of corneal ulceration in rabbits. Arch
Ophthalmol 1991; 109:414419.
101. Brazzell RK, Stern ME, Aquavella JV, et al:
Human recombinant epidermal growth
factor in experimental corneal wound
32:336340.
102. Daniele S, Frati L, Fiore C, Santoni G: The
effect of the epidermal growth factor (EGF)
on the corneal epithelium in humans.
Graetes Arch Clin Exp Ophthalmol 1979;
210:159165.
103. Labella F; Alcuaz A: The effects of epithelial
growth factor (EGF) on the healing time of
traumatic epithelial corneal ulcers. A
multicenter double-blind clinical study.
30(Suppl):151.
104. Bonini S, Lambiase A, Rama P, et al:

for neurotrophic keratitis. Ophthalmology
2000; 107:13471351.
105. Lambiase A, Rama P, Bonini S, et al:
for corneal neurotrophic ulcers. N Engl J
Med 1998; 338:12221223.
106. Lee CH, Whiteman AL, Murphy CJ et al:
Substance P, insulinlike growth factor I, and
surface healing. Arch Ophthalmolol 2002;
120:215217.
107. Yen MT, Pflugfelder SC, Feuer WJ: The
effect of punctal occlusion on tear
production, tear clearance, and ocular
surface sensation in normal subjects. Am J
Ophthalmol 2001; 131:314332.
108. Gilbard JP, Rossi SR, Azar DT, Heyda KG:
Effect of punctal occlusion by Freeman
silicone plug insertion on tear osmolarity in
dry eye disorders. CLAO J 1989; 15:216218.
109. Balaram M, Schaumberg DA, Dana MR:
Efcacy and tolerability outcomes after
punctal occlusion with silicone plugs in dry
eye syndrome. Am J Ophthalmol 2001;
131:3036.
110. Tai MC, Cosar CB, Cohen EJ, et al: The
clinical efcacy of silicone punctal plug
therapy. Cornea 2002; 21:135139.
111. Ellis MF, Daniel M: An evaluation of the
safety and efcacy of botulinum toxin
type A (BOTOX) when used to produce a
protective ptosis. Clin Exp Ophthalmol
2001; 29:394399.
112. Panda A, Pushker N, Bageshwar LM:
Lateral tarsorrhaphy: is it preferable to
patching? Cornea 1999; 18:299301.
113. Khokhar S, Natung T, Sony P, et al:
Amniotic membrane transplantation in
refractory neurotrophic corneal ulcers: a
randomized, controlled clinical trial. Cornea
2005; 24:654660.
114. Heinz C, Eckstein A, Steuhl KP, Meller D:
Amniotic membrane transplantation for
reconstruction of corneal ulcer in graves
ophthalmopathy. Cornea 2004;
23:524526.
115. Prabhasawat P, Kosrirukvongs P,
Booranapong W, Vajaradul Y: Single and
multilayer amniotic membrane
transplantation for persistent corneal
epithelial defect with and without stromal
thinning and perforation. Br J Ophthalmol
2001; 85:14551463.
116. Ivekovic B, Tedeschi-Reiner E, Petric I,
et al: Amniotic membrane transplantation
for ocular surface reconstruction in
neurotrophic corneal ulcera. Coll Antropol
2002; 26:4754.
117. Lee SH, Tseng SC: Amniotic membrane
transplantation for persistent epithelial
defects with ulceration. Am J Ophthalmol
1997; 123:303312.
106:15041510.
119. Chen HJ, Pires RT, Tseng SC: Amniotic
membrane transplantation for severe
neurotrophic corneal ulcers. Br J
Ophthalmol 2000; 84:826833.
120. Boudreau N, Werb Z, Bissell MJ:
Suppression of apoptosis by basement
membrane requires three-dimensional
tissue organization and withdrawal from
the cell cycle. Proc Natl Acad Sci 1996;
93:35093513.
125.
126.
127.
128.
conjunctival flaps on the cornea: a review

of 122 Consecutive Cases. Int Ophthalmol
Clin 1970; 10:329325.
Gundersen T, Pearlson HR: Conjunctival
flaps for corneal disease: their usefulness
and complications. Trans Am Ophthalmol
Soc 1969; 67:7895.
Alino AM, Perry HD, Kanellopoulos AJ, et al:
Conjunctival flaps. Ophthalmology 1998;
105:11201123.
Fogle JA, Kenyon KR, Foster CS: Tissue
adhesive arrests stromal melting in the
89:795802.
Dausch D, Landesz M, Klein R, Schroder E:
Phototherapeutic keratectomy in recurrent
corneal epithelial erosion. Refract Corneal

Surg 1993; 9:419424.
129. John ME, van der Karr MA, Noblitt RL,
Boleyn KL: Excimer laser phototherapeutic
keratectomy for treatment of recurrent
corneal erosion. J Cataract Refract Surg
1994; 20:179181.
130. Ohman L, Fagerholm P: The influence of
excimer laser ablation on recurrent corneal
erosions: a prospective randomized study.
Cornea 1998; 77:349352.
CHAPTER 55
121. Meller D, Tseng SCG: Conjunctival

epithelial cell differentiation on amniotic
membrane. Invest Ophthalmol Vis Sci
1999; 40:878886.
122. Sato H, Shimazaki J, Shinozaki K: Role of
growth factors for ocular surface
reconstruction after amniotic membrane
transplantation. Invest Ophthalmol Vis Sci
1998; 39:S248.
123. Gundersen T: Conjunctival flaps in the
treatment of corneal disease with
reference to a new technique of
application. Arch Ophthalmol 1958;
60:880888.
124. Paton D, Milauskas AT: Indications,
surgical technique, and results of thin
759
CHAPTER
56
Chemical Injuries of the Eye

Michael D. Wagoner, Samar Al-Swailem, Sabah Al-Jastaneiah, and Kenneth R. Kenyon
Chemical injuries are potentially devastating ocular surface

injuries that can result in permanent visual impairment.113
Most victims of chemical injury are young with exposure
occurring in industrial accidents, at home, and in criminal
assaults.1416
The most important agents causing chemical injuries of the
eye are summarized in Table 56.1.1 Alkali injuries tend to occur
more commonly than acid injuries.1416 Commonly encountered causes of alkali injury include ammonia, lye,
potassium hydroxide, magnesium hydroxide, and lime.1 Of
these, ammonia and lye tend to produce the most serious
injuries.810 Magnesium hydroxide found in reworks may
combine with thermal injury to produce a particularly
devastating injury.17 Lime, particularly in the form of plaster, is
the most commonly encountered alkali injury; fortunately, it
tends to cause less severe injury.1
The most commonly encountered acid injuries include those
from sulfuric, sulfurous, hydrofluoric, nitrous, acetic, chromic,
and hydrochloric acids.1 Sulfuric acid injury is the most
commonly seen, usually after battery explosions.1 Sulfuric acid
reacts with water in the precorneal tear lm to produce heat and
charring of the corneal and conjunctival epithelium.1 It seldom
produces serious injury unless there is additional damage from
thermal injury or high-velocity penetration of a foreign body
into the eye.1 The most severe acid injuries are associated with
hydrofluoric acid.18 Because of its low molecular weight and small
molecular size, hydrofluoric acid penetrates readily into the
corneal stroma and anterior segment, producing severe injury.18
PATHOPHYSIOLOGY
Key Features: Pathophysiology
The severity of a chemical injury is related to the surface area

of contact and the degree of penetration.
Limbal stem cells are the cells most qualied to restore
functional competence of the corneal epithelial surface after
injury.
Corneal repair by keratocytes consists of both synthesis and
degradation of stromal collagen.
Persistent inflammation may delay reepithelialization and shift
the net balance of corneal repair toward progressive stromal
ulceration.
The severity of ocular injury after chemical exposure is related

to the surface area of contact and the degree of penetration.113
In general, alkalis tend to penetrate more readily than acids.712
Depending on the degree of penetration, there may be damage
to the corneal and conjunctival epithelium, the limbal stem

cells, stromal keratocytes, stromal nerve endings, endothelium,
lens, ciliary body, and vascular endothelium of the conjunctiva,
episclera, iris, and ciliary body.1013 Acid injuries tend to remain
conned to the ocular surface and produce more supercial
damage.7
The three major principles guiding evaluation and management of chemical injury are based on addressing each of the
three main pathophysiologic mechanisms contributing to the
nal outcome: (1) regeneration of ocular surface epithelium
and its state of differentiation, (2) stromal matrix remodeling,
including repair and degradation, and (3) inflammation. These
pathophysiologic processes have been discussed extensively
elsewhere but are summarized briefly in the remainder of this
section.
EPITHELIAL INJURY, REPAIR, AND

DIFFERENTIATION
In 1977, Thoft and Friend coined the term ocular surface to
emphasize the interdependence of the stratied, nonkeratinizing epithelium of the cornea and conjunctiva.19 Numerous
studies have supported the hypothesis that centripetal
movement of cells from the peripheral cornea, limbus, or
conjunctiva is responsible for normal and posttraumatic
replacement of corneal epithelium.2024 Consistent with
observations in other self-renewing tissues in which cell
populations are maintained by stem cells,25 the theory of the
existence and localization of a limbal stem-cell population that
maintains normal and replaces injured corneal epithelium was
advanced26 and supported by subsequent scientic studies.2736
It is now well recognized that both conjunctival epithelium3235
and limbal stem-cell populations29 may resurface the
chemically injured corneal epithelium. Although partial
transdifferentiation of conjunctival epithelium to corneal
epithelium may be possible,3335 compelling evidence suggests
that conjunctiva-derived epithelium never fully expresses
corneal epithelial phenotypic features.19,33,36,37 Reliance on
conjunctival epithelium to resurface the cornea after chemical
injury is associated with delayed reepithelialization, supercial
and deep stromal vascularization, persistence of goblet cells in
the corneal epithelium, and poor epithelium-basement
membrane adhesion.3739 Clinical and experimental studies
have provided strong evidence that limbal stem cells are the
cells most qualied to restore the functional competence of the
corneal epithelial surface after injury.29,40,41 Reestablishment of
a phenotypically normal corneal epithelial surface with limbal
stem cell-derived cell populations is the rst major principle
in the therapeutic management of severe chemical injuries.
761
TABLE 56.1 Common Causes of Chemical Injury

Class
Compound
Common Sources and Uses
Comments
Alkali
fumes
Ammonia (NH3)
1. Fertilizers
2. Refrigerants
3. Cleaning agents (7% solution)
1. Combines with water to form NH4OH

2. Very rapid penetration
Lye (NaOH)
1. Drain cleaners
Potassium hydroxide
(KOH)
1. Caustic potash
1. Penetrates almost as rapidly

as ammonia
1. Severity similar to that of lye
Magnesium hydroxide
(Mg(OH)2)
1. Sparklers
1. Produces combined thermal and

alkali injury
Lime (Ca(OH)2)
1. Plaster
1. Most common cause of chemical injury

in workplace
2. Poor penetration
3. Toxicity increased by retained
particulate matter
2. Mortar
3. Cement
4. Whitewash
Acid
Sulfuric acid (H2SO4)
1. Industrial cleaner
2. Battery acid
1. Combines with water to produce

corneal thermal injury
2. May have associated foreign body
battery acid
Sulfurous acid
(H2SO3)
1. Formed from sulfur dioxide

(SO2) by combination with
corneal water
2. Fruit and vegetable preservative
3. Bleach
4. Refrigerant
1. Penetrates more easily than other acids
Hydrofluoric acid
(HF)
1.
2.
3.
4.
5.
1. Penetrates easily
2. Produces severe injury
Acetic acid
(CH3COOH)
1. Vinegar 410%
Glass polishing
Glass frosting
Mineral refining
Gasoline alkylation
Silicone production
2. Essence of vinegar 80%

3. Glacial acetic acid 90%
1. Mild injury with less than 10%

contamination
2. Severe injury with higher concentration
Chromic acid
(Cr2O3)
1. Used in the chrome- plating

industry
1. Chronic exposure produces chronic

conjunctivitis with brown discoloration
Hydrochloric acid (HCl)
1. Used as a 3238% solution
1. Severe injury only with high

concentration and prolonged exposure
Adapted from McCulley JP: Chemical injuries. In: Smolin G, Thoft RA, (eds.): The cornea: scientific foundations and clinical practice, 2nd edn. Boston, Little, Brown; 1987.
SECTION 6
CORNEAL STROMAL MATRIX INJURY,

REPAIR, AND ULCERATION
762
The maintenance and regeneration of the corneal stroma is

the primary responsibility of the keratocyte.42 After chemical
injury, these pluripotential cells are capable of phagocytosis
of collagen brils and synthesis and secretion of collagen
glycosaminoglycan ground substance, collagenase, and
collagenase inhibitors.4345 These complex functions may be
modulated by cytokines from the epithelium, inflammatory
cells, and other keratocytes.4648 A closely related family of
enzymes, called matrix metalloproteinases (MMP), are
responsible for the initial rate-limiting cleavage of collagen
molecules.49 After chemical injury, degradation of the basement
membrane collagen may be initiated by MMP9 (formerly,
collagenase types IV and V).45,49,50 This step always precedes
subsequent degradation of the corneal stromal matrix by
MMP1 (formerly, keratocyte-derived collagenase type I)49,51,52
and MMP8 (formerly, polymorphonuclear leukocyte-derived
collagenase type I).53,54 MMP1 and MMP8 may be detectable
as soon as 9 h after injury,55 but signicant amounts are not
detected until 1421 days later,56 which corresponds to the
same period when keratocyte synthesis of type I collagen
peaks.44 During this active period of remodeling, excessive

degradation of the matrix by MMP1 and MMP8, relative to
type I collagen synthesis, may result in enzymatic degradation
of the corneal stroma, a process referred to as sterile corneal
ulceration. Exploitation of known pharmacologic intervention,
which helps shift the balance toward repair, rather than
ulceration, is the second major principle in the management
of severe chemical injuries.
INFLAMMATION
The third pathophysiologic component influencing the course
of the severe chemical injury is inflammation.57 The
association of inflammatory cell inltration (especially with
polymorphonuclear leukocytes) into the corneal stroma with
sterile corneal ulceration is well documented.57 The cessation of
sterile corneal ulceration after exclusion of inflammatory cells
from the stroma also is established.5860 Persistent inflammation may delay reepithelialization61 and perpetuate
continued recruitment of additional inflammatory cells.62
Rigorous control of inflammation is the third major principle
in the therapeutic management of severe chemical injuries.

FIGURE 56.1. (a) Thoft - grade I. (b) Thoft grade II. (c) Thoft - grade III. (d) Thoft - grade
IV.
From Wagoner MD, Kenyon KR: Chemical injuries. In:
Shingleton BJ, Hersh PS, Kenyon KR, eds. Eye
trauma. St Louis: CV Mosby; 1990.
CLINCAL COURSE AND EVALUATION

Key Features: Clinical Course and Evaluation
Chemical injuries can be tentatively graded on the basis of

presumptive limbal stem cell damage, as indirectly determined
by assessment of limbal ischemia.
The healing pattern of the ocular surface epithelium can
subsequently be used to make a more definitive determination
of the severity of limbal stem cell injury.
Extensive limbal stem cell injury will ultimately result in either
development of fibrovascular pannus or sterile corneal
ulceration.
McCulley has divided the clinical course of chemical injuries

into four distinct pathophysiologic and clinical phases: the
immediate, acute (days 07), early repair (days 721), and late
repair (day 21 to several months later) phases.1
IMMEDIATE PHASE
The clinical ndings present immediately after chemical
injury are related to the extent of ocular surface involvement,
the depth of penetration, and the relative toxicity and concentration of the injurious compound. The extent of surface
involvement can be determined by the size of the corneal and
conjunctival epithelial defects. The depth of corneal and
intraocular penetration can be estimated by evaluating
corneal clarity, intraocular inflammation, intraocular pressure,
and lens clarity. The depth of ocular surface penetration, and
possible limbal stem-cell damage, can be evaluated indirectly
by assessment of vascular ischemia and necrosis of limbal and
bulbar conjunctiva.
The modied Hughes classication8,9,63 correlates the degree
of limbal ischemia with prognosis. In recognition that the
relation between the clinical appearance of limbal ischemia
and prognosis is attributable to the actual damage of limbal
stem cells, a newer classication scheme based on the degree
of limbal stem-cell loss has been proposed).4 A grade I injury
(see Fig. 56.1a) involves little or no loss of limbal stem cells
and presents with little or no evidence of ischemia. A grade II
injury (see Fig. 56.1b) involves subtotal loss of limbal stem
cells and presents with ischemia of less than one-half of the

limbus. A grade III injury (see Fig. 56.1c) involves total loss of
limbal stem cells with preservation of the proximal conjunctival epithelium and presents with ischemia of one-half
to the entire limbus. A grade IV injury (see Fig. 56.1d) involves
total limbal stem-cell loss as well as loss of the proximal
conjunctival epithelium and presents with extensive damage to
the entire anterior segment. In the absence of specic stains
for limbal stem cells, a denitive diagnosis of the actual extent
of limbal stem-cell loss can be made only after several weeks
of observation of the epithelial healing patterns.
ACUTE PHASE
During the rst week, important parameters that should be
monitored include evidence of reepithelialization (or lack
thereof), intraocular pressure, and progressive ocular inflammation grade I injuries tend to heal without incident
(Fig. 56.2a) during this period. There is slow but progressive
reepithelialization in grade II injuries (see Fig. 56.2b), although
the sector of limbal stem-cell loss may show little or no
reepithelialization. Grade III and IV injuries (see Fig. 56.2c,d)
show no reepithelialization. Even in the most severe chemical
injuries, there is little or no collagenolytic activity during this
phase.55,56 The intraocular pressure may rise during the rst
1224 h as a result of distortion of the trabecular meshwork
and release of prostaglandins, and may remain elevated because
of persistent inflammation.64,65 Within 1224 h after chemical
injury, inltration of the peripheral cornea with polymorphonuclear leukocytes and other inflammatory cells begins.66
CHAPTER 56
EARLY REPAIR PHASE

During the early repair phase, epithelial migration continues
in less severe injury (grade II) but remains delayed in more
severe injuries (grades III and IV). In grade III injuries, the
limbal appearance may be relatively normal and the cornea
may be reasonably clear, but the presence of complete limbal
stem-cell loss is suggested by the absence of any corneal
reepithelialization. In grade IV injuries, there often is persistence of both the conjunctival and corneal epithelial defects
and no change in the ischemic appearance of the bulbar conjunctiva. In severe chemical injuries, a second wave of inflam-
763
SECTION 6
764
FIGURE 56.2. (a) Type I healing pattern:

normal epithelial recovery. After a grade I injury
with no limbal stem-cell loss, complete
reepithelialization with a normal corneal
phenotype is complete within 1 week. (b) Type
II healing pattern: delayed differentiation. After
a grade II injury with limbal stem-cell loss from
the 7 to 9oclock position, there is delayed
reepithelialization through the late repair phase,
with development of superficial vascular
pannus and impaired corneal epithelial
differentiation in the affected quadrant. (c) Type
III healing pattern: fibrovascular pannus. After a
grade III injury with complete limbal stem-cell
loss, progressive corneal epithelialization with
conjunctivally derived epithelium over a
4month period results in fibrovascular pannus
covering the entire cornea. (d) Type IV healing
pattern: sterile corneal ulceration. After a grade
IV injury with complete limbal stem-cell loss, as
well as loss of conjunctival epithelium and
vascularity in the entire nasal quadrant, there is
no corneal reepithelialization. Sterile corneal
ulceration of the nasal and inferior corneal
stroma begins ~3 months after injury.
matory cell inltration begins after 7 days and continues to

progress over the next several weeks.66 Progressive inflammation
persists as long as the epithelial defect persists.67
stroma has not been initiated by the beginning of the early

repair phase, it will occur inevitably within the next several
weeks or months (see s. 56.2d).
LATE REPAIR PHASE
MEDICAL THERAPY
By the beginning of the late repair phase, corneal inflammation,

collagen synthesis, and collagenase activity are peaking,44,56 and
the balance of repair and collagenolysis is influenced greatly by
the status of the corneal epithelium.68 Based on the clinical
behavior of the ocular surface epithelium by this stage, it is
possible to conrm the healing pattern that is being established,
predict the prognosis with reasonable certainty, and make a
denitive decision regarding the need for surgical intervention.1
A type I healing pattern (normal epithelial recovery)
corresponds to a grade I limbal stem-cell injury in that
restoration of an intact and phenotypically normal corneal
epithelial surface (see Fig. 56.2a) has occurred by this stage.
A type II healing pattern (delayed differentiation) corresponds
to a grade II limbal stem-cell injury. Because there has been
partial loss of limbal stem cells, there still may be a sectorial
corneal epithelial defect in the quadrant corresponding to
limbal stem-cell loss, with completed reepithelialization in
quadrants with adequate limbal stem-cell reserves. With
subsequent evolution of the clinical course, reepithelialization
with conjunctival epithelium may occur in this quadrant,
along with supercial vascular pannus (see Fig. 56.2b). In the
absence of any reepithelialization by this stage, it is virtually
certain that complete limbal stem cell loss has occurred.
Both healing patterns that are determined by this rate-limiting
deciency of stem cells are unsatisfactory. A type III healing
pattern (brovascular pannus) corresponds to a grade III
injury in which there has been complete loss of limbal stem
cells but preservation of source of proximal conjunctival epithelium. Although delayed reepithelialization with conjunctival
epithelium may occur in the next few weeks or months, it will
be with brovascular pannus (conjunctivalization) of the ocular
surface, and the ultimate outcome is a tectonically stable but
scarred and vascularized cornea (see Fig. 56.2c). A type IV
healing pattern (sterile corneal ulceration) corresponds to a
grade IV injury in which there has been complete loss of limbal
stem cells and proximal conjunctival epithelium with ischemic
necrosis. If progressive enzymatic destruction of the corneal
Key Features: Medical Therapy
Topical corticosteroids indirectly facilitate reepithelialization by

ameliorating inflammation-induced delays in epithelial
migration.
Doxycycline is the most potent clinically available collagenase
inhibitor.
Systemic and topical ascorbate may facilitate stromal collagen
repair.
Management of the severely chemically injured eye must be

directed toward: (1) promoting ocular surface epithelial recovery
with proper phenotypic transdifferentiation, (2) augmenting
corneal repair by supporting keratocyte collagen production
and minimizing ulceration related to collagenase activity, and
(3) controlling inflammation.
The rapidity with which pH abnormalities of the ocular
surface are neutralized has a signicant impact on the
subsequent clinical course. Because of the deep location and
relatively protected position of the limbal stem cells,69 it is
possible that prompt removal of slowly penetrating chemical
agents may avert permanent damage and ensure a type I or type
II healing pattern, rather than a type III or IV healing pattern
that inevitably will occur if all limbal stem cells are lost. Early
attempts at irrigation by the patient and coworkers usually are
inadequate, permitting signicant penetration. Copious
irrigation with any nontoxic irrigating solution must be
immediately initiated on presentation, irrespective of the prior
history of irrigation.70 Irrigation for a minimum of 30 min and
checking the pH of tears for evidence of neutrality is
recommended. If this is not achieved, irrigation must be
continued. Failure to achieve neutrality often is evidence of a
retained reservoir of chemical in the eye. This is particularly
true in plaster injuries, in which particles embedded in the
upper tarsal conjunctiva can provide continued slow release of
PROMOTE EPITHELIAL WOUND HEALING

AND DIFFERENTIATION
The recovery of an intact and phenotypically normal corneal
epithelium is the rate-limiting determinant of prognosis of a
chemical injury. Initially, aggressive medical therapy is
indicated to facilitate reepithelialization. In grade I or II
injuries, there may be reduction in morbidity associated with
slow reepithelialization related to poor medical management of
the epithelial defect and its associated inflammation. In
presumed grade III injuries, aggressive medical therapy is
warranted in the fortunate event that the severity of the injury
has been overestimated and some residual limbal stem cells are
available to contribute to corneal epithelial recovery. In grade IV
injuries, it can be assumed from the beginning that medical
therapy alone is doomed to failure, and early planning of
surgical intervention is warranted.
Tear Substitutes
Severe chemical injuries most commonly occur in young
patients in whom tear production usually is adequate. In the
event that the patient has a preexisting deciency of aqueous
tear production, unpreserved tear substitutes may be used
facilitate reepithelialization. The use of topical viscoelastics
may be useful in facilitating corneal epithelial migration in
grade I and II injuries and in minimizing conjunctival scarring
and symblepharon formation after grade III and IV injuries.75
After reepithelialization, frequent administration of unpreserved
tear substitutes and administration of ointments at bedtime
may be necessary to benet persistent keratopathy and
recurrent epithelial erosions.
Occlusive Therapy
Although there is a theoretical advantage to protecting the
migrating epithelium from the windshield-wiper effect of the
eyelids, occlusive therapy (patching, taping) is of little use in the
acute care of the chemically injured eye. If epithelial defects
persist into the early and late repair phases, the cause usually is
persistent inflammation or limbal stem-cell deciency, both of
which are unresponsive to occlusive therapy.
Bandage Soft Contact Lens

Therapeutic soft contact lenses and collagen shields tend to be
poorly tolerated in the acutely inflamed, chemically injured
eye.76 As with occlusive therapy, there is no theoretical
advantage to using therapeutic soft contact lens therapy for the
management of a persistent epithelial defect in this setting.
Autologous Serum
The benecial effect of topical bronectin77 and epidermal
growth factor78 in promoting reepithelialization in experimental
clinical trials suggested a potential role for autologous serum in
the treatment of persistent epithelial defects.79 Although autologous serum is not commercially available, it can be prepared
in standard clinical laboratories by previously described
methods.79 To date, a prospective clinical trial evaluating the
efcacy of autologous serum in promoting reepithelialization
has not been evaluated in prospective clinical trials.
Retinoic Acid
Retinoic acid may play a role in the late management of
persistent ocular surface abnormalities.80 In grade I and II
injuries, there may be persistent goblet cell dysfunction, tear
lm instability, and poor ocular surface wetting, even after
reepithelialization is complete. In these cases, retinoic acid may
promote goblet cell recovery and improve ocular surface
function.81 In grade III injuries, in which reepithelialization
must come from conjunctival epithelium, retinoic acid may
promote partial, but not complete, transdifferentiation of the
conjunctival epithelium to a corneal epithelial phenotype.81
SUPPORT REPAIR AND MINIMIZE

ULCERATION
Ascorbate
Ascorbate is an essential water-soluble vitamin that is a cofactor
in the rate-limiting step of collagen formation.82 Damage to the
ciliary body epithelium by intraocular chemical injury results in
decreased secretion of ascorbate and a reduction in its
concentration in the anterior chamber.83 This may lead to
impaired collagen synthesis by keratocytes.84 Both topical and
systemic ascorbate have been shown to decrease the incidence
of sterile corneal ulceration after chemical injury.8487 Topical
application is superior to systemic supplementation.87 Ascorbate
has no effect on the progression of established stromal ulceration,
emphasizing the importance of early supplementation.86
Collagenase Inhibitors
Early investigators reported clinical efcacy of several
collagenase inhibitors, including cysteine, acetylcysteine,
sodium ethylenediaminetetraacetic acid (EDTA), calcium
EDTA, and penicillamine, in the experimental and clinical
management of severe chemical injuries.88 The efcacy of these
topical collagenase inhibitors is limited by its instability, poor
corneal penetration, toxicity, and relative weak potency.88
Tetracycline derivatives are efcacious in reducing
collagenase activity.89 This effect is independent of its
antimicrobial properties and probably is due to chelation of zinc
at the active site of the collagenase enyzme.90 Doxycycline is the
most potent tetracycline collagenase inhibitor.91 In vitro,
doxycycline is 33 to 180fold more potent collagenase
inhibitor than acetylcysteine,91 and its systemic use has largely
replaced acetylcysteine as the collagenase inhibitor of choice in
the treatment of chemical injuries. Tetracycline derivates have
been demonstrated to be efcacious in preventing sterile corneal
ulceration in experimental alkali injuries,92,93 although
demonstration of clinical efcacy in prospective clinical trials
has not been done.
Several promising synthetic collagenase inhibitors may
provide an even more effective pharmacologic tool.9497
Synthetic thiol and carboxylpeptide collagenase inhibitors are
10 000-fold more potent collagenase inhibitors in vitro than
acetylcysteine.98100 Recombinant tissue inhibitors of
metalloproteinases (TIMP) have anticollagenolytic properties
similar to those of synthetic thiol inhibitors.97 These synthetic
inhibitors are effective in reducing the incidence of corneal
thinning, ulceration, and perforation in experimental chemical
injuries.9496 To date, the results of clinical trials with synthetic
thiols or TIMP are not available.
CHAPTER 56
alkali into the tear lm. Using topical anesthesia, all particles
should be removed with ne forceps or by scraping with a
disposable scalpel (e.g., BardParker No. 15 blade). The benets
of paracentesis and irrigation of the anterior chamber remain
uncertain.7173 Limited data suggest that such maneuvers must
be performed within 1731572 min to have any benecial effect,
thus limiting their practicality.
Dbridement of necrotic corneal epithelium is necessary
to allow proper reepithelialization, irrespective of the severity
of the injury.74 In addition, it is important to dbride necrotic
conjunctival tissue because this tissue has been shown to be a
nidus of continued inflammation from retained caustic materials,
a site of accumulation and sustained release of inflammatory
cells, and a source of detrimental proteolytic enzymes.74
765
CONTROL INFLAMMATION
Corticosteroids
Corticosteroids traditionally have been the mainstay of therapy
for the reduction of tissue injury related to acute or chronic
inflammatory conditions. Although it is well recognized
clinically that there is little risk of sterile ulceration in the rst
week after chemical injury, whether or not corticosteroids are
used, unfounded fears that corticosteroids may delay
reepithelialization and potentiate sterile corneal ulceration has
resulted in an unfortunate reluctance to use these potent
inflammation inhibitors in the acute chemically injured eye.
Corticosteroids have no adverse effect on the rate of epithelial
wound healing.101 By decreasing inflammatory cell inltration,
they may facilitate migration indirectly by partially
ameliorating inflammation-induced delays in corneal epithelial
migration,61 and suppress sterile ulceration indirectly by
reducing one source of proteolytic enzymes.53,54 Corticosteroids
do interfere with stromal repair by impairing both keratocyte
migration and collagen synthesis.102,103 Fortunately, the
deleterious effects of corticosteroids do not become apparent
until the early repair phase, when their favorable contributions
to reduction of inflammation and collagenase inhibition104,105
are offset by their interference with collagen synthesis, thereby
shifting the balance toward ulceration rather than repair.
The key to successful corticosteroid use is to maximize the
antiinflammatory effect during the window of opportunity in
the rst 710 days, when there is little risk associated with
corticosteroid use. Later, when corticosteroid-related complications are more likely to occur, therapy can be modied by
tapering corticosteroids while monitoring for evidence of corneal
thinning or by substituting progestational steroids (discussed
later), nonsteroidal antiinflammatory drugs (NSAIDs; discussed
later), or both.
Progestational Steroids
Progestational steroids have less antiinflammatory potency
than do corticosteroids but have only a minimal effect on stromal repair and collagen synthesis.105,106 Medroxyprogesterone
1% has been shown experimentally to inhibit collagenase and
reduce ulceration after chemical injury, suppress corneal neovascularization, and minimally suppress stromal wound repair.105108
Progestational steroids may be substituted for corticosteroids
after 1014 days, when suppression of inflammation still is
required but interference with stromal repair is undesirable.
SECTION 6
NSAIDs
Early experimental trials on the efcacy of NSAIDs suggest a
possible role for these compounds in the management of
chemical injuries.109,110 NSAIDs may prove to be an effective
additive for corticosteroids in the rst week and a substitute or
additive for progestational steroids after the rst week if similar
antiinflammatory properties, such as those demonstrated after
cataract surgery,111 or experimental lens-induced uveitis,112 are
applicable in the acutely chemically injured eye. The effect of
NSAIDs on stromal wound repair, collagen synthesis,
collagenolytic activity, and inhibition of neovascularization has
not been addressed adequately.
Citrate
766
Citrate is a calcium chelator that decreases the membrane and

intracellular levels of calcium, resulting in impaired
chemotaxis, phagocytosis, and release of lysosomal enzymes of
polymorphonuclear leukocytes.113 In experimental chemical
injuries, early administration of citrate reduces early-phase and
late-phase inflammatory cell inltration by 63% and 92%,
respectively,67,113 and signicantly reduces the incidence of
corneal ulceration.114116 As with ascorbate, topical

administration is superior to systemic administration.116
SURGICAL THERAPY
Key Features: Surgical Therapy
Early surgical intervention with conjunctival/Tenons

advancement, amniotic membrane transplantation, and/or
limbal stem cell transplantation may be necessary to achieve
successful reepithelialization.
Late surgical intervention with penetrating or lamellar
keratoplasty may be necessary for visual rehabilitation, but this
should only be performed after appropriate rehabilitation of the
ocular surface has been achieved.
Two of the most important advances in the past two decades in

the management of chemical injury have been the identication
of the seminal role stem cells in the maintenance and
regeneration of the ocular surface and the application of ocular
surface transplantation techniques that are able address specic
stem cell and structural abnormalities at each stage of the
evolving clinical course after severe chemical injuries.117
Ocular surface transplantation techniques that are useful in
achieving the objective of promoting epithelial wound healing
and transdifferentiation include conjunctival and Tenons
advancement (tenoplasty) for immediate reestablishment of limbal
vascularity and proximate source of epithelium for denuded
corneal surface,118 amniotic membrane transplantation to
facilitate migration of surviving limbal stem cells,119124 limbal
stem-cell transplantation for early or late reestablishment of
a phenotypically normal corneal epithelial surface,40 and
conjunctival125 or mucosal membrane transplantation98,126 for
reestablishment of the conjunctival fornices and normal
lidglobe apposition. Reestablishment of a phenotypically
normal corneal epithelial surface and an anatomically normal
conjunctival fornix and lidglobe relationships is mandatory for
success of subsequent attempts at visual rehabilitation with
corneal transplantation.
CONJUNCTIVAL AND TENONS

ADVANCEMENT (TENOPLASTY)
The use of conjunctival and Tenons advancement, or
tenoplasty, is based on the principle of using vital connective
tissue within the orbit to reestablish limbal vascularity and
to facilitate corneal reepithelialization with conjunctival
epithelium.118 Although this technique is almost uniformly
successful in reducing the likelihood of anterior segment
necrosis and sterile corneal ulceration, it is less successful in
ensuring adequate or appropriate differentiated corneal
epithelial recovery. As such, this technique is recommended to
facilitate initial stabilization of a grade IV injury.
TISSUE ADHESIVE
Tissue adhesives are effective in arresting further sterile corneal
ulceration and in maintaining the integrity of the globe,60 when
all other measures have failed to prevent this untoward event.
It is best reserved for use for impending or actual perforations
that are 1 mm or smaller, and remains preferable to emergency
tectonic procedures. A tectonic penetrating keratoplasty may be
used to preserve the integrity of the globe for larger perforations
(>1 mm) that cannot be adequately addressed with tissue
adhesives.
AMNIOTIC MEMBRANE TRANSPLANTATION

Human amniotic membrane is the thin, semitransparent
innermost layer of the fetal membrane.120 It consists of an
avascular stromal matrix, a thick basement membrane, and an
epithelial monolayer.120 The tissue may be transplanted to the
corneal surface with the basement membrane oriented upwards
or downwards. When used with the basement membrane
oriented downward, the amniotic membrane acts like a biologic
bandage contact lens or an onlay (patch) graft, promoting
epithelialization beneath the membrane.120,121 The amniotic
membrane invariably dissolves in a few weeks, which may be
earlier than desired if reepithelialization is not complete. When
used with basement membrane oriented upward it acts like an
inlay graft, with the amniotic membrane functioning as a new
basement membrane which promotes epithelialization over its
surface.120,121 The amniotic membrane becomes incorporated
into the substratum of the newly formed epithelium and
persists for months. Once it is reabsorbed, it is replaced by new
brotic stromal tissue that partially conserves the corneal
thickness but does not always possess the same transparency
of healthy stroma.122 When used with a combination of an
inner amniotic membrane with the basement membrane
oriented upward and an outer amniotic membrane with the
basement membrane oriented downward, reepithelialization is
sandwiched between two basement membranes.
Irrespective of the transplantation technique, amniotic tissue
may facilitate reepithelialization if complete or partial limbal
stem-cell function is present.123 Reepithelialization is facilitated
by providing a basement membrane substrate for induction of
epithelial migration and adhesion and through the release of
growth factors that facilitate proliferation of limbal stem cells
and transient amplifying cells.119124 Amniotic tissue may
reduce inflammation mechanically by excluding inflammatory
cells from the corneal stroma120,124 and biochemically through
release of factors that suppress proinflammatory cytokines,99,100
facilitate apoptosis of inflammatory cells,120,124 and reduce
epithelial and keratocyte apoptosis.120,124 Amniotic tissue may
also reduce corneal angiogenesis120,124 and conjunctival scar
formation after chemical injury.124
Although the use of amniotic membrane for the treatment
of ocular surface disease is not new, its modern use was
reintroduced by Kim and Tseng in 1995.119 Since then, multiple
investigators have reported favorable results with the amniotic
membrane transplantation in the treatment of chemical
injuries.124,127130 Successful reepithelialization has been
reported in acute phase treatment of grade II and III injuries
with both onlay and inlay graft techniques.124,127130
Disappointing results have been reported in the treatment of
grade IV injuries,131,132 presumably due to the lack of efcacy of
this procedure in situations where total limbal stem-cell loss
has occurred.
Amniotic membrane transplantation may use in the late
rehabilitation phase, either alone or in conjunction with limbal
stem-cell transplantation.123,124,133 In cases of incomplete limbal
stem-cell loss, it may be effective in the treatment of persistent
epithelial defects, recurrent epithelial erosions, and persistent
epitheliopathy, and in the reduction of chronic inflammation.
It may also be combined with other reconstructive procedures
of the ocular surface, such as the lysis of symblepharon. In
cases of complete limbal stem-cell function, it may be used

in conjunction with limbal stem-cell transplantation.
LIMBAL STEM-CELL TRANSPLANTATION

Limbal stem-cell transplantation was developed by Kenyon and
Tseng40 as a modication of the original conjunctival transplantation technique of Thoft.125 This technique is the best
method of reestablishing a phenotypically correct corneal
epithelial surface early in the clinical course of a grade III or IV
injury, thereby preventing the problems associated with a type
III (brovascular pannus) or type IV (sterile corneal ulceration)
healing pattern. In addition, this procedure may be performed
later in the clinical course either alone, or as part of a staged
procedure with penetrating keratoplasty, in order to improve
ocular surface function and provide visual rehabilitation,
respectively.
In unilateral cases of chemical injury or asymmetric chemical
injuries, conjunctival limbal autograft transplantation (CLAU)
is usually performed by harvesting contralateral limbal stem
cells from the uninjured or less injured fellow eye and
transferring them to the injured or more injured eye.40
Ipsilateral CLAU is an option for treatment of limited, sectorial
limbal stem-cell deciency, with transfer of tissue from an
uninjured area in the opposite meridian of the damaged limbus
to the decient area.134
In severe bilateral injuries, limbal allograft transplantation
from a living relative135138 or a cadaver donor are the only viable
options.139141 Living-related conjunctival limbal allograft
transplantation (lr-CLAG) is technically the same procedure as
CLAU with the exception that the limbal stem cells are
harvested from a close relative and transferred to the injured
eye(s).135138 The risk of allograft rejection is higher after this
procedure than with penetrating keratoplasty, possibly due to a
higher concentration of transplanted antigens in the peripheral
cornea from which stem cells are harvested, and the
transplantation of vascular tissue into a vascular bed.142
Prevention of rejection and maintenance of a stable stem cell
population has been reported with the use prolonged topical and
systemic immunosuppression.143 One specic recommended
protocol is indenite use of topical corticosteroids and
cyclosporine, and triple immunosuppressive therapy with
tapering doses of prednisone, cyclosporine A, and azathioprine
for at least 1224 months.143 Despite systemic immunosuppression, rejection rates as high as 2513533%136 have been
reported. Increasing the dosage of the immunosuppressive
regimen may successfully reverse immunological rejection in
some cases.135,136 Keratolimbal allograft transplantation (KLAT)
is a technique for transferring limbal stem cells from a donor
cadaver to treat severe bilateral injuries.139141 As with lr-CLAG,
there is a signicant risk of graft rejection,139,143 but successful
preservation of limbal stem-cell function and corneal clarity has
been reported with prolonged, aggressive topical and systemic
immunosuppression.139,143
Ex vivo expansion of limbal stem cells is currently being
investigated as an improvement of existing limbal stem-cell
transplantation techniques.144146 This procedure involves the
dissection of a small piece of donor limbal tissue, growth and
expansion of viable limbal stem cells in culture, and
transplantation of the epithelial sheet to the recipient eye. To
date, technical difculties in cell culture and transfer of
the epithelial sheet to the recipient eye have limited the
applicability of this technique. The successful application of ex
vivo expansion after successful solutions have been found for
the technical roadblocks will allow reduction in the number of
limbal stem cells that must be harvested to restore normal
function. It will therefore be possible to restore function in all
CHAPTER 56
The tissue adhesive may be removed or allowed to extrude

spontaneously after 68 weeks, when a secure brovascular scar
has formed and eliminated the risk of subsequent stromal
ulceration. Unfortunately, this brovascular scar may impair
vision and worsen the prognosis for subsequent penetrating
keratoplasty.
767
FIGURE 56.3. (a) This 5-year-old boy

sustained a severe alkali injury to the left eye
with total ocular surface epithelial loss and
extensive limbal and scleral ischemia. (b)
Despite maximal medical therapy, including
corticosteroids and ascorbate, an extensive
epithelial defect persisted. Limbal autograft
transplantation was performed 4 weeks after
injury with slow resolution of the epithelial
defect. (c) The ocular surface epithelium is
intact and stable 6 months after injury, and
stromal neovascularization and edema have
subsided, albeit with dense scarring. (d) A deep
lamellar keratoplasty was performed 1 year
after injury. Visual acuity is 20/60 2 years later,
and the cornea remains stable, uninflamed, and
avascular without topical medications or
lubricants.
From Wagoner MD, Kenyon KR: Chemical injuries. In:
Shingleton BJ, Hersh PS, Kenyon KR, eds. Eye
trauma. St Louis: CV Mosby; 1990.
but the most severe bilateral injuries by autograft

transplantation, thereby eliminating the risk of immunological
rejection. In cases where harvesting of limbal stem cells is
required from a living relative or a cadaver, it is possible that the
risk of rejection will be reduced if antigen-presenting
Langerhans cells are eliminated during cell culture and
expansion and only epithelial cells are transplanted.147
MUCOSAL MEMBRANE TRANSPLANTATION
SECTION 6
Mechanical abnormalities of the bulbar and palpebral

conjunctiva related to progressive scarring include restriction of
extraocular movement, fornix foreshortening and obliteration,
symblepharon formation, incomplete lid closure, cicatricial
entropion, trichiasis, and lid margin keratinization. In some
unilateral cases, bulbar conjunctival transplantation may correct
many of these abnormalities.125 A more effective approach,
especially in bilateral cases, is to use mucosal membrane grafts
to reconstruct the fornix and restore normal lidglobe
relations.98,126 Although such grafts do not restore the corneal
epithelial functions provided by limbal stem-cell transplantation, the harvesting of mucosal grafts from nasal mucosa
may improve impaired goblet cell function of the conjunctiva.
768
PENETRATING KERATOPLASTY
An optical penetrating keratoplasty may be attempted after
appropriate rehabilitation of the ocular surface has been
achieved (Figure 56.3).148,149 There are advocates of performing
limbal stem-cell transplantation prior to penetrating
keratoplasty150,151 or doing the procedures simultaneously in
order to facilitate more rapid visual rehabilitation.152 The
prognosis for successful penetrating keratoplasty is related to
the original severity and sequelae of the chemical injury, as well
as the adequacy of restoration of normal phenotypic functions
of the ocular surface.148152
Kuckelkorn153 and Redbrake154 have reported success in both
the acute and chronic chemically injured eye with largediameter (11 to 12mm) penetrating keratoplasty. In the acute
setting, this approach is highly successful in the management
of impending or actual corneal perforation, although the longterm prognosis for graft clarity is poor.153 As a technique for
combined late rehabilitation of persistent limbal stem-cell
dysfunction and corneal opacication, the prognosis for graft
clarity also is guarded, mainly because of the complication of

allograft rejection of the donor limbal stem cells, rather than
endothelial rejection.154
KERATOPROSTHESIS
Keratoprosthesis may be useful in bilateral, severe chemical
injury in which the prognosis is hopeless for penetrating
keratoplasty because of irreparable damage to the ocular
surface, or in unilateral cases that have experienced repeated
immunologic endothelial rejection. Although the success rate
has been poor in the past, improved keratoprosthesis design and
better postoperative management now offer an improved
prognosis.155157
SPECIFIC THERAPY
The most common therapeutic mistakes in the management of
severe chemical injuries of the eye are the failure to diagnose
and treat persistent limbal stem-cell dysfunction properly at an
early stage and to try aggressively to control ocular
inflammation. After immediate irrigation and dbridement of
necrotic corneal epithelium and bulbar conjunctival tissue, the
recommended therapy for severe chemical injuries (grades II
through IV) is summarized as follows.
Acute Phase
1. Topical corticosteroids every 12 h
2. Topical sodium ascorbate 10% every 2 h
3. Topical sodium citrate 10% every 2 h
4. Topical tetracycline 1% ointment four times a day
5. Topical cycloplegics as needed
6. Topical antiglaucoma medications as needed
7. Systemic sodium ascorbate 2 g orally four times a day
8. Systemic doxycycline 100 mg orally twice a day
9. Consider amniotic membrane transplantation (grade II
and III)
10. Consider conjunctival and Tenons advancement
(grade IV)
Early Repair Phase
1. Discontinue or taper (with close observation) topical
corticosteroids
2. Begin progestational steroids (Provera 1%), NSAIDs, or
both, topically every 12 h

3. Continue topical and systemic sodium ascorbate
4. Continue topical sodium citrate
5. Continue topical tetracycline and systemic doxycycline
Late Repair Phase
1. Taper medical therapy after reepithelialization is complete
(grade I or II)
2. Limbal stem-cell transplantation +/ amniotic membrane
transplantation (for grade III or IV injuries)
3. Tectonic procedures (tissue adhesive, small- or largediameter keratoplasty), if necessary
Late Rehabilitation
1. Ocular surface reconstruction (amniotic membrane
transplantation, conjunctival transplantation, mucous
membrane transplantation)
2. Limbal stem-cell transplantation (conjunctival limbal
autograft, living-relative conjunctival limbal allograft,
keratolimbal allograft)
3. Penetrating keratoplasty
4. Keratoprosthesis
CONCLUSION
For many years, the main focus of investigation and therapy for
severe chemical injury was on control of inflammation and
regulation of the delicate balance between corneal collagen synthesis and collagenolytic activity by carefully selected medical
intervention.1 In the past two decades, improved understanding
of the importance of the ocular surface has led to the
development of ocular surface transplantation techniques that
may help restore depleted limbal stem-cell populations (limbal
autograft and allograft transplantation), augment the function
of limbal stem cells (amniotic membrane transplantation),
restore limbal vascularity (conjunctival and Tenons advancement), and correct mechanical and anatomic disturbances of
the bulbar and palpebral conjunctiva (conjunctival, mucous
membrane, and amniotic membrane transplantation).
1. McCulley JP: Chemical injuries. In: Smolin

G, Thoft RA, eds. The cornea: scientic
foundation and clinical practice. 2nd edn.
2. Wagoner MD, Kenyon KR: Chemical
injuries of the eye. I. Etiology,
pathophysiology, and primary intervention.
In: Kuhn N, Pieramici D, eds. Ocular
trauma: the essentials. New York/Stuttgart:
Thieme Medical Publishers; 2002:7778.
injuries of the eye. I. Etiology,
pathophysiology, and primary intervention.
In: Kuhn N, Pieramici D, eds. Ocular
trauma: the essentials. New York/Stuttgart:
Thieme Medical Publishers; 2002:335349.
4. Wagoner MD: Chemical injuries of the eye.
injuries. In: Shingleton BJ, Hersh PS,
Kenyon KR, eds. Ocular trauma. St Louis:
Mosby; 1991:6368.
6. Pster RR: Chemical corneal burns. Int
7. Friedenwald JS, Hughes WF, Hermann H:
Acid burns of the eye. Arch Ophthalmol
1946; 35:98108.
8. Hughes WF: Alkali burns of the cornea.
II. Clinical and pathologic course. Arch
Ophthalmol 1946; 36:189214.
9. Hughes WF: Alkali burns of the cornea.
I. Review of the literature and summary of
present knowledge. Arch Ophthalmol 1946;
35:423426.
10. Grant WM, Kern HL: Action of alkalis on
the corneal stroma. Arch Ophthalmol 1955;
54:931934.
11. Matsuda H, Smelser GK: Endothelial cells
in alkali-burned corneas: ultrastructural
alterations. Arch Ophthalmol 1973;
89:402409.
12. Matsuda H, Smelser GK: Epithelium and
stroma in alkali-burned corneas. Arch
Ophthalmol 1973; 89:396401.
13. Kuckelkorn R, Kottek A, Reim M:
Intraocular complications after severe
chemical burns-incidence and surgical
treatment. Klin Monatsbl Augenheilkd 1994;
205:8692.
14. Morgan SJ: Chemical burns of the eye:
causes and management. Br J Ophthalmol
1987; 71:854857.
15. Kuckelkorn R, Luft I, Kottek AA, et al:

Chemical and thermal eye burns in the
residential area of RWTH Aachen: analysis
of accidents in l year using a new
automated documentation of ndings. Klin
16. Shingleton BJ: Eye injuries. N Engl J Med
1991; 325:408413.
17. Harris LH, Cohn K, Galin MA: Alkali injury
from reworks. Ann Ophthalmol 1971;
3:849851.
18. McCulley JP: Ocular hydrofluoric acid
burns: animal model, mechanism of injury
and therapy. Trans Am Ophthalmol Soc
1990; 88:649684.
19. Friend J, Thoft RA: Functional competence
of regenerating ocular surface epithelium.
17:134139.
20. Friedenwald JS: Growth pressure and
metaplasia of conjunctival and corneal
epithelium. Doc Ophthalmol 1951;
56:184192.
21. Friedenwald JS, Buschke W: Some factors
concerned in the mitotic and woundhealing activities of the corneal epithelium.
42:371383.
22. Hanna C, Bicknell DS, OBrien JE: Cell
turnover in the adult human eye. Arch
Ophthalmol 1961; 65:695698.
23. Kinoshita S, Friend J, Thoft RA: Sex
chromatin of donor corneal epithelium in
rabbits. Inv Ophthalmol Vis Sci 1981;
21:434441.
pericorneal papillary structure in renewal of
corneal epithelium. Nature 1971;
229:560561.
25. Cairnie AB, Laila PK, Osmond DG, eds:
Stem cells of renewing cell populations.
New York: Academic Press; 1976.
26. Thoft RA, Friend J: The X,Y,Z hypothesis of
corneal epithelial maintenance. Invest
27. Lavker RM, Doug G, Cotsarellis G, Sun TT:
Limbal basal epithelial cells display
characteristics consistent with stem cells
from various stratifying epithelia. Invest
Ophthalmol Vis Sci 1988; 29(suppl):191.
28. Zeiske JD, Bukusglu G, Yankauekas MA,
et al: Alpha-enolase is restricted to basal
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
cell of stratied squamous epithelium. Dev

Biol 1992; 151:1826.
Thoft RA, Wiley LA, Sundarraj N: The
multipotential cells of the limbus. Eye 1989;
3:109113.
Cotsarellis G, Cheng SZ, Dong G, et al:
103:4962.
Schermer A, Galvin S, Sun TT:
Differentiation-related expression of a major
suggests limbal location of corneal epithelial
stem cells. J Cell Biol 1986; 103:4962.
Zieske JD, Bukusoglu G, Yankauckas MA:
Shapiro MS, Friend J, Thoft RA: Corneal
re-epithelialization from the conjunctiva.
21:135142.
Harris TM, Berry ER, Pakurar AS, Sheppard
LB: Biochemical transformation of bulbar
conjunctiva into corneal epithelium: an
electrophoretic analysis. Exp Eye Res
1985; 41:597606.
Kinoshita S, Friend J, Thoft RA: Biphasic
cell proliferation in transdifferentiation of
conjunctival to corneal epithelium in
24:10081014.
Thoft RA, Friend J: Biochemical
transformation of regenerating ocular
surface epithelium. Invest Ophthalmol Vis
Sci 1977; 16:1420.
Tseng SCG, Hirst LW, Farazdaghi M, Green
WR: Goblet cell density and vascularization
during conjunctival transdifferentiation.
25:11681176.
Gartaganis SP, Margaritis LH, Koliopoulous
JX: The corneal epithelium basement
membrane complexes after alkali burns: an
ultrastructural study. Ann Ophthalmol 1987;
19:263268.
Huang AJW, Tseng SCG: Corneal epithelial
epithelium. Invest Ophthalmol Vis Sci 199l;
32:96105.
CHAPTER 56
REFERENCES
769
SECTION 6
770
40. Kenyon KR, Tseng SCG: Limbal autograft

41. Tseng SCG, Tsai RJ: Limbal transplantation
for ocular surface transplantation: a review.
Fortschr Ophthalmol 1991; 88:236242.
42. Cintron C, Hassinger LC, Kublin CL, et al:
Biochemical and ultrastructural changes in
collagen during corneal wound healing.
J Ultrastruct Res 1978; 65:1322.
43. Ishizaki M, Zhu G, Haseba T, et al:
Expression of collagen I, smooth muscle
alpha-actin, and vimentin during the
healing of alkali-burned and lacerated
34:33203328.
44. Katakami C, Fujisawa K, Sahori A, et al:
Localization of collagen (I) and collagenase
mRNA by in situ hybridization during
corneal wound healing after
epikeratophakia or alkali-burn. Jpn J
45. Matsubara M, Girard MT, Kublin CL, et al:
Differential roles for two gelatinolytic
enzymes of the matrix metalloproteinase
family in the remodeling cornea. Dev Biol
1991; 147:425439.
46. Johnson-Wint B: Autocrine regulation of
collagenase production by passaged
corneal stromal cells in vitro. Invest
47. Johnson-Wint B: Regulation of stromal cell
collagenase production in adult rabbit
cornea: in vitro stimulation and inhibition by
epithelial cell products. Proc Natl Acad Sci
USA 1980; 77:55315535.
48. Johnson-Wint B, Bauer EA: Stimulation of
collagenase synthesis by a 20,000-dalton
epithelial cytokine. J Biol Chem 1985;
260:20802085.
49. Fini ME, Yue BY, Sugar J: Collagenolytic/
gelatinolytic metalloproteinases in normal
and keratoconus corneas. Curr Eye Res
1992; 11:849862.
50. Garbisa S, Ballin M, Daga- Giordini D,
et al: Transient expression of type IV
collagenolytic metalloproteinase by human
mononuclear phagocytes. J Biol Chem
1986; 261:23692375.
51. Gross J: An essay on biological
degradation of collagen. In: Hay ED, ed.
Cell biology of the extracellular matrix. New
York: Plenum; 1982:217258.
52. Johnson-Muller B, Gross J: Regulation of
corneal collagenase production: epithelialstromal cell interactions. Proc Natl Acad
Sci USA 1978; 75:44174421.
53. Lazarus GS, Brown RS, Daniels JR, et al:
Human granulocyte collagenase. Science
1968; 159:14831485.
54. Kao WWY, Ebert J, Kao CWC, et al:
Development of monoclonal antibodies
recognizing collagenase from rabbit PMN:
the presence of this enzyme in ulcerating
corneas. Curr Eye Res 1986; 5:801805.
55. Gordon JM, Bauer EA, Eisen AZ:
Collagenase in the human cornea:
immunologic localization. Arch Ophthalmol
1980; 98:341345.
56. Berman MB, Dohlman CH, Davison PF,
et al: Characterization of collagenolytic
activity in the ulcerating cornea. Exp Eye
Res 1971; 11:225227.
57. Kenyon KR: Inflammatory mechanisms in
corneal ulceration. Trans Am Ophthalmol
Soc 1985; 83:610663.
58. Kenyon KR, Berman M, Rose J, Gage J:
Prevention of stromal ulceration in the
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
alkali-burned rabbit cornea by glued-on

contact lens: evidence for the role of
polymorphonuclear leukocytes in collagen
degradation. Invest Ophthalmol Vis Sci
1979; 18:570587.
Foster CS, Zelt RP, Mai-Phan T,
Kenyon KR: Immunosuppression and
selective inflammatory cell depletion:
studies on a guinea pig model of corneal
ulceration after ocular alkali burning.
Fogle JA, Kenyon KR, Foster CS: Tissue
adhesive arrests stromal melting in the
89:795802.
Wagoner MD, Kenyon KR, Gipson IK, et al:
Polymorphonuclear neutrophils delay
corneal epithelial wound healing in vitro.
25:12171220.
Cavanagh HD, Pihlaja D, Thoft RA,
Dohlman CH: The pathogenesis and
treatment of persistent epithelial defects.
Trans Am Acad Ophthalmol Otolargyngol
1976; 81:754769.
Thoft RA: Chemical and thermal injury. Int
Paterson CA, Eakins KE, Paterson RM,
et al: The ocular hypertensive response
following experimental acid burns in the
rabbit eye. Invest Ophthalmol Vis Sci 1979;
6774.
Paterson CA, Pster RR: Intraocular
pressure changes after alkali burns. Arch
Ophthalmol 1974; 91:211218.
Pster RR, Haddox JL, Sommers CI: Alkalidegraded cornea generates a low
molecular weight chemoattractant for
polymorphonuclear leukocytes. Invest
Paterson CA, Williams RN, Parker AW:
Characteristics of polymorphonuclear
leukocyte inltration into the alkali-burned
eye and influence of sodium citrate. Exp
Eye Res 1984; 39:701708.
Kenyon KR: Decision making in the therapy
of external eye disease: noninfected corneal
ulcers. Ophthalmology 1982; 89:4451.
Goldberg MF, Bron AJ: Limbal palisades of
Vogt. Trans Am Ophthalmol Soc 1982;
80:155171.
Herr RD, White GL Jr, Bernhisel K, et al:
Clinical comparison of ocular irrigation
fluids following chemical injury. Am J
Emerg Med 1991; 9:228231.
Paterson CA, Pster RR, Levinson RA:
Aqueous humor pH changes after
experimental alkali burns. Am J Ophthalmol
1975; 79:414419.
Bennett TO, Peyman GA, Rutgard J:
Intracameral phosphate buffer in alkali
burns. Can J Ophthalmol 1978; 13:9395.
Grant WM: Experimental investigation of
paracentesis in the treatment of ocular
ammonia burns. Arch Ophthalmol 1950;
44:399404.
Reim M, Teping C: Surgical procedures in
the treatment of most severe eye burns:
revival of the articial epithelium. Acta
Reim M: A new treatment concept in
severe caustic injuries and thermal burns of
the eye. Fortschr Ophthalmol 1989;
86:722726.
Wentworth JS, Paterson CA, Wells JT,
et al: Collagen shields exacerbate
ulceration of alkali-burned rabbit corneas.
77. Gordon JF, Johnson P, Musch DC: Topical

bronectin ophthalmic solution in the
treatment of persistent defects of the
cornea epithelium. Am J Ophthalmol 1995;
119:281287.
78. Dellaert MM, Casey TA, Wiffen S, et al:
Influence of topical human epidermal
growth factor on post-keratoplasty reepithelialization. Br J Ophthalmol 1997;
81:391395.
79. Tsubota K, Goto E, Shimmura S, Shimazaki
J: Treatment of persistent corneal epithelial
defects by autologous serum application.
Ophthalmology 1999; 106:19841989.
80. Soong K, Martin NF, Wagoner MD, et al:
Topical retinoid therapy for squamous
metaplasia of various ocular surface
disorders. Ophthalmology 1988;
95:14421446.
81. Tseng SCG, Farazdaghi M: Reversal of
conjunctival transdifferentiation by topical
tretinoin. Cornea 1988; 7:273279.
82. Barnes MJ: Function of ascorbic acid in
collagen metabolism. Ann N Y Acad Sci
1975; 258:264277.
83. Pster RR, Friend J, Dohlman CH: The
anterior segments of rabbits after alkali
burns. Arch Ophthalmol 1971; 86:189193.
84. Pster RR, Paterson CA: Additional clinical
and morphological observations on the
favorable effect of ascorbate in
experimental ocular alkali burns. Invest
85. Pster RR, Paterson CA: Ascorbic acid in
the treatment of alkali burns of the eye.
86. Pster RR, Paterson CA, Hayes SA: Effect
of topical l0% ascorbate solution on
established corneal ulcers after severe
alkali burns. Invest Ophthalmol Vis Sci
1982; 22:382385.
87. Pster RR, Paterson CA, Spiers JW, et al:
The efcacy of ascorbate treatment after
severe experimental alkali-burns depends
upon the route of administration. Invest
88. Brown SI, Hook CW: Treatment of corneal
destruction with collagenase inhibitors.
1971; 75:11991207.
89. Golub LM, McNamara TF, DAngelo G,
Greenwald RA: A non antibacterial
chemically-modied tetracycline inhibits
mammalian collagenase activity. J Dent
Res 1987; 66:13101314.
90. Brion M, Lambs L, Berthon G: Metal iontetracycline interactions in biological fluids.
Part 5. Formation of zinc complexes with
tetracycline and some of its derivatives and
assessment of their biological signicance.
Agents Actions 1985; 17:229242.
91. Lindy O, Sorsa T, Saari H, et al:
Doxycycline inhibits the serpinase activity
of human neutrophil collagenase. [Abstract]
Rome, The International Congress on
Inflammation. Biennial forum for basic and
clinical research. October 611, 1991.
92. Perry HD, Hodes LW, Seedor JA, et al:
Effect of doxycycline hyclate on corneal
epithelial wound healing in the rabbit alkaliburn model: preliminary observations.
Cornea 1993; 12:379382.
93. Seedor JA, Perry HD, McNamara TE,
et al: Systemic tetracycline treatment of
alkali-induced corneal ulceration in rabbits.
94. Paterson CA, Wells JG, Koklitis PA, et al:
Recombinant tissue inhibitor of
95.
96.
97.
98.
99.
100.
101.
102.
103.
104.
105.
106.
107.
108.
109.
110.
metalloproteinases type-1 suppresses

alkali-burn-induced corneal ulceration in
35:677684.
Schultz GS, Strelow S, Stern GA, et al:
Treatment of alkali-injured rabbit corneas
with a synthetic inhibitor of matrix
metalloproteinases. Invest Ophthalmol Vis
Sci 1992; 33:33253331.
Wentworth JS, Paterson CA, Gray RD:
Effect of a metalloproteinase inhibitor on
established corneal ulcers after alkali burn.
33:21742179.
Gray RD, Miller RB, Spatola AF: Inhibition
of mammalian collagenase by thiolcontaining peptides. J Cell Biochem 1986;
32:7177.
Naumann GOH, Lang GK, Rummelt V,
Wigand ME: Autologous nasal mucosa
transplantation in severe bilateral
conjunctival mucus deciency syndrome.
Solomon A, Rosenblatt M, Monroy D,
et al: Suppression of interleukin 1(and
interleukin 1b in human limbal epithelial
cells cultured on the amniotic membrane
stromal matrix. Br J Ophthalmol 2001;
85:444449.
Kim JS, Kim JC, Na BK, et al: Amniotic
membrane patching promotes healing and
inhibits proteinase activity on wound
healing following acute alkali burn. Exp Eye
Res 2000; 70:329327.
Ho PC, Elliott JH: Kinetics of corneal
epithelial regeneration. II. Epithelial growth
factor and topical corticosteroids. Invest
Beams R, Linabery L, Grayson M: Effect of
topical corticosteroids on corneal wound
strength. Am J Ophthalmol 1968;
66:11311133.
Gasset AR, Lorenzetti DW, Ellison EM,
Kaufman HE: Quantitative corticosteroid
effect on corneal wound healing. Arch
Ophthalmol 1969; 81:589591.
Koob TF, Jeffrey JJ, Eisen ZZ: Regulation
of human skin collagenase activity by
hydrocortisone and dexamethasone in
organ culture. Biochem Biophys Res
Commun 1974; 61:10831088.
Phillips K, Arffa R, Cintron C, et al: Effects
of prednisolone and medroxyprogesterone
on corneal wound healing, ulceration, and
neovascularization. Arch Ophthalmol 1983;
101:640643.
Gross, J, Azizkhan RG, Biswas C, et al:
Inhibition of tumor growth, vascularization,
and collagenolysis in the rabbit cornea by
medroxyprogesterone. Proc Natl Acad Sci
USA 1981; 78:11761180.
Lass JH, Campbell RC, Rose J, et al:
Medroxyprogesterone on corneal
ulceration: its effects after alkali burns on
rabbits. Arch Ophthalmol 1981;
99:673676.
Newsome DA, Gross JA: Prevention by
medroxyprogesterone of perforation of the
alkali-burned rabbit cornea: inhibition of
collagenolytic activity. Invest Ophthalmol
Vis Sci 1977; 16:2131.
Struck HG, Geiser H, Block HU, et al:
Leukotriene antagonist S 872419 for earlyphase treatment of chemical burn in the
rabbit eye. Eur J Ophthalmol 1991;
l:137141.
Struck HG, Franke C, Tost M, Taube C:
Animal experiment and clinical studies of
111.
112.
113.
114.
115.
116.
117.
118.
119.
120.
121.
122.
123.
124.
125.
126.
the use of anti-inflammatory agents in

chemical burns of the eye. Klin Monatsbl
Augenheilkd 1988; 193:401406.
Flach AJ, Lavelle CJ, Olander KW, et al:
The effect of ketorolac tromethamine 0.5%
in reducing postoperative inflammation
after cataract extraction and intraocular
lens implantation. Ophthalmology 1988;
95:12791284.
Chang MS, Chiou GC: Prevention of lens
protein-induced ocular inflammation with
cyclooxygenase and lipoxygenase
inhibitors. J Ocul Pharmacol 1989;
5:353360.
Pster RR, Haddox JL, Dodson RW,
Deshazo WF: Polymorphonuclear leukocyte
inhibition by citrate, other heavy metal
chelators, and trifluoperazine: evidence to
support calcium binding protein
involvement. Invest Ophthalmol Vis Sci
1984; 25:955970.
Pster RR, Nicolaro ML, Paterson CA:
Sodium citrate reduces the incidence of
corneal ulcerations and perforations in
extreme alkali-burned eyes: acetylcysteine
and ascorbate have no favorable effect.
21:486490.
Pster RR, Haddox JL, Lank KM: Citrate or
ascorbate-citrate treatment of established
corneal ulcers in the alkali injured rabbit
eye. Invest Ophthalmol Vis Sci 1988;
29:11101115.
Pster RR, Haddox JL, Paterson CA: The
efcacy of sodium citrate in the treatment
of severe alkali burns of the eye is
influenced by the route of administration.
Cornea 1982; 1:205211.
Tseng SCG: Concept and application of
Reim M, Overkamping B, Kuckelkorn R:
Two years experience with Tenonplasty.
Kim JC, Tseng SCG: Transplantation of
preserved human amniotic membrane for
ocular surface reconstruction in severely
damaged rabbit corneas. Cornea 1995;
14:473484.
Dua HS, Gomes JAP, King AJ,
Maharajan VS: The amniotic membrane in
ophthalmology. Br J Ophthalmol 2004;
49:5177.
Letko E, Stechschulte SU, Kenyon KR,
et al: Amniotic membrane inlay and overlay
grafting for corneal epithelial defects and
stromal ulcers. Arch Ophthalmol 2001;
119:659653.
Gris O, Wolley-Dod C, Guell JL, et al:
Histopathologic ndings after amniotic
membrane graft in the human cornea.
Tseng SCG, Prabhasawat P, Barton K, et al:
Amniotic membrane transplantation with or
with out limbal autograft for corneal surface
reconstruction in patients with limbal stem
cell deciency. Arch Ophthalmol 1998;
116:431441.
Meller D, Renato TF, Pires RT, et al:
acute chemical or thermal burns.
Thoft RA: Conjunctival transplantation.
Kuckelkorn R, Wenzel M, Lamprecht J,
et al: Autologous transplantation of nasal
mucosa after severe chemical and thermal
eye burns. Klin Monatsbl Augenheilkd
1994; 204:155161.
127. Ucakhan OO, Koklu G, Firat E: Nonpreserved human amniotic membrane

transplantation in acute and chronic
chemical eye injuries. Cornea 2002;
21:169172.
128. Sridhar MS, Bansal AK, Sangwan VS, Rao
GN: Amniotic membrane transplantation in
acute chemical and thermal injury. Am J
Ophthalmol 2000; 130:134137.
129. Kobayashi A, Shirao Y, Toshita T, et al:
Temporary amniotic membrane patching for
acute chemical burns. Eye 2003;
17:149158.
130. Shimazaki J, Yang H-Y, Tsubota K:
ocular surface reconstruction in patients
with chemical and thermal burns.
Ophthalmology 1997; 104:20682076.
131. Joseph A, Dua HS, King AJ: Failure of
amniotic membrane transplantation in the
treatment of acute ocular burns. Br J
Ophthalmol 2001; 85:10651069.
132. Panda A, Nainiwal SK, Sudan R: Failure of
amniotic membrane transplantation in the
treatment of acute ocular burns. Br J
Ophthalmol 2002; 86: 831.
133. Meallet MA, Espana EM, Grueterich M,
with conjunctival limbal autograft for total
limbal stem cell deciency. Ophthalmology
2003; 110:15851592.
134. Nishiwaki-Dantas MC, Dantas PEC, Reggi
JRA: Ipsilateral limbal translocation for the
treatment of partial limbal deciency
secondary to ocular alkali burn. Br J
Ophthalmol 2001; 85:10311033.
135. Daya SM, Ilari L: Living related conjunctival
limbal allograft for the treatment of stem
cell deciency. Ophthalmology 2001;
108:126134.
136. Rao SK, Rajagopal R, Sitlakshmi G,
Padmanabhan P: Limbal allograft from
related live donors for corneal surface
reconstruction. Ophthalmology 1999;
106:822828.
137. Tsubota K, Shimmura S, Shinozaki N,
et al: Clinical application of living-related
conjunctival-limbal allograft. Am J
Ophthalmol 2002; 133:134135.
138. Tsai RJ, Tseng SCG: Human allograft limbal
transplantation for corneal surface
reconstruction. Cornea 1994;
13:389400.
139. Solomon A, Ellies P, Anderson DF, et al:
Long term outcome of keratolimbal
allograft with or without penetrating
keratoplasty for total limbal stem cell
deciency. Ophthalmology 2002;
109:11591166.
140. Espana EM, Grueterich M, Ti S, Tseng
SCG: Phenotypic study of a case receiving
a keratolimbal allograft and amniotic
membrane for total limbal stem cell
deciency. Ophthalmol 2003; 110:481486.
141. Meisler DM, Perez VL, Proudt J: A devise
to facilitate limbal stem cell procurement
from eye bank donor tissue for keratolimbal
allograft procedures. Am J Ophthalmol
2005; 139:212214.
142. Williams KA, Brereton HM, Aggarwal R,
et al: Use of DNA polymorphisms and
polymerase chain reaction to examine the
survival of human limbal stem cell allograft.
Am J Ophthalmol 1995; 120:342350.
143. Holland EJ, Schwartz GS: Changing
concepts in the management of severe
ocular surface disease over twenty-ve
years. Cornea 2000; 19:688698.
CHAPTER 56
771
SECTION 6
144. Pellegrini G, Traverso CE, Franzi AT, et al:

Long term restoration of damaged corneal
surfaces with autologous cultivated corneal
epithelium. Lancet 1997; 349:990993.
145. Lindberg K, Brown M E, Chaves HV, et al:
In vitro propagation of human ocular
surface epithelial cells for transplantation.
34:26722679.
146. Koizumi N, Cooper LJ, Fullwood NJ, et al:
An evaluation of cultivated corneal limbal
epithelial cells using cell-suspension
culture. Invest Ophthalmol Vis Sci 2002;
43:21142121.
147. Lavker RM, Tseng SCG, Sun TT: Corneal
epithelial stem cells at the limbus: looking
at some old problem from a new angle.
Exp Eye Res 2004; 78:433446.
148. Brown SI, Bloomeld SE, Pearce DB:
A follow-up report on transplantation of the
772
149.
150.
151.
152.
alkali-burned cornea. Am J Ophthalmol

1974; 77:538542.
Abel R Jr, Binder PS, Polack FM, Kaufman
HE: The results of penetrating keratoplasty
after chemical burns. Trans Am Acad
Ophthalmol Otolaryngol 1975;
79:OP584OP595.
Kenyon KR, Tseng SCG: Limbal autograft
Frucht-Perry J, Siganos CS, Solomon A,
et al: Limbal cell autograft transplantation
for severe ocular surface disorders. Graefes
236:582587.
Yao Y, Zhang B, Zhour P, Jiang J:
Autologous limbal grafting combined with
deep lamellar keratoplasty with severe
chemical or thermal burns at lat stage.
Ophthalmology 2002; 109:20112017.
153. Kuckelkorn R, Redbrake C, Schrage NF,

Reim M: Keratoplasty with ll12 mm
diameter for management of severely
chemically-burned eyes. Ophthalmologe
1993; 90:683687.
154. Redbrake C, Buchal V, Reim M:
Keratoplasty with a scleral rim after most
severe eye burns. Klin Monatsbl
Augenheilkd 1996; 208:145151.
155. Dohlman CH, Schneider HA, Doane MG:
Prosthokeratology. Am J Ophthalmol 1974;
77:694700.
156. Cardona H, DeVoe AG:
Prosthokeratoplasty. Trans Am Acad
Prosthokeratoplasty. Am J Ophthalmol
1974; 77:694700.
CHAPTER
57
Wetting of the Ocular Surface and Dry-Eye

Disorders
Eva-Marie Chong, Mona Harissi-Dagher, and Reza Dana
INTRODUCTION
A smooth, stable, and re-wettable ocular surface is essential for
good vision and comfort. By denition, dry eye refers to disorders
of the tear lm due to reduced tear production and/or excessive
tear evaporation associated with symptoms of ocular discomfort.1 Discomfort related to dry-eye disease may reduce quality
of life and may be more than a nuisance to many patients. As an
illustration of this point, a time-trade-off utility study showed
that patients with severe dry eyes were willing to trade 1.6 years
of their expected 10-year longevity to be free of the condition.
This is comparable to that reported by patients suffering from
moderate to severe (class III/IV) angina.2 The economic burden
of dry-eye disorders is reflected in loss of productivity, number
of physicians ofce visits, and the multibillion dry-eye therapeutic
industry.
Over 60 years ago, Henrik Sjgren described a disease characterized by autoimmune damage to lacrimal gland tissue, decreased
tear secretion, and ocular surface disease, that he termed keratoconjunctivitis sicca (KCS).3 KCS has since become a generic
term, and various classication systems for dry-eye syndrome
have emerged in recent years in an effort to facilitate the study
of this group of disorders (Table 57.1).
Recent progress in the understanding of the inflammatory and
immunological pathophysiology of dry-eye disease has led to a
shift in the paradigm for evaluating the ocular surface and in
the management of this common, chronic problem. This chapter
reviews the tear lm components, the pathophysiologic mechanism, diagnosis, and treatment of dry-eye disorders.
factor (EGF), brinogen, basic broblast growth factor (bFGF),

neuropeptides, and mucins (Table 57.4).
Classically, the tear lm has been described as a three-layered
structure with a lipid layer (0.1 mm thick), an aqueous layer
(710 mm thick), and a mucinous layer (0.21.0 mm thick). This
view of the three-layer architecture has been replaced by the concept of an integrated aqueous and mucin gel with a graded concentration of mucins under a dynamic lipid layer. Conceptually,
however, the triple-layer structure remains useful in dening
the pathophysiology of dry-eye disorders.
TABLE 57.1. Denitions of Dry-Eye Syndrome

KCS (Sjgren)3
Autoimmune disorder involving lacrimal gland
Tear volume deciency
Ocular surface disease
National Eye Institute/Industry Workshop (Lemp)4
Disorder of the tear lm due to tear deciency or excessive
evaporation
Damage to the interpalpebral ocular surface
Associated with symptoms of discomfort
Lacrimal Keratoconjunctivitis (LKC) (Pflugfelder et al)5
Dysfunction of the integrated lacrimal unit
Unstable tear lm
EPIDEMIOLOGY
Altered tear composition
The most comprehensive epidemiologic studies available show

a prevalence of at least 912 million people in the USA with
signicant dry-eye syndrome.9 Dry eye is more prevalent amongst
the elderly, particularly postmenopausal females (Table 57.2).
Other associated risk factors for dry eye are listed in Table 57.3.
Ocular surface and glandular inflammation

Discomfort
Dysfunctional Tear Syndrome (DTS) (McDonnell et al)6
DTS without associated lid margin disease
DTS with associated lid margin disease
TEAR FILM COMPOSITION

The tear lm is a complex fluid mixture secreted by the main
and accessory lacrimal glands, meibomian glands, goblet cells of
the conjunctiva, corneal and conjunctival surface epithelia, and
the vasculature of the conjunctiva. It is essential for the optical
properties of the cornea and in protecting, nourishing, and maintaining the health of the ocular surface. Tear lm composition
may vary, even among healthy individuals, by ethnicity, gender,
and age. The tear lm is a matrix-like structure composed of
water, electrolytes, antimicrobial molecules, immunoglobulins,
immunomodulators, growth factors such as epithelial growth
DTS with disturbance of distribution of tears

Dry Eye Workshop DEWS (Dogru et al)7
Triple Classication of Dry Eye for Practical Clinical Use
(Murube et al)8
Etiopathogenesis
Affected Glands ALMEN (Aquo-serous deciency, Lipodeciency,
Mucodeciency, Epitheliopathy, Non-lacrimal affected exocrine
glands)
Severity
773
TABLE 57.2. Major Epidemiologic Studies of Dry-Eye Disorders in the USA

Reference
Study Parameter
Study Population
Prevalence
Nurses Health Study

(Schaumberg et al)10
Severe symptoms,
clinical diagnosis
N = 39 876
US women
5.7% <50 years old

9.8% >75 years old
7.8% age adjusted
Salisbury Maryland Study

(Schein et al)11
Symptoms of DES
N = 2520
White Men and women
Age 65-84
14.6% >1 symptom DES
Beaver Dam Eye Study

(Moss et al)12
Self report of symptoms of DES
N = 3722
Men and women
Age 48-91
14.4% total population

16.7% among women
Womens Health Study

(Schaumberg et al)13
Use of HRT and DES symptoms

or clinical Dx
N = 25 556
Postmenopausal women
Increased DES with HRT

Odds ratio 1.69 estrogen and
1.29 estrogen/progesterone
vs no HRT
15% increase in DES each
3 year duration of HRT
TABLE 57.3. Common Risk Factors of Dry-Eye Disorders

Age
Gender
Menopause
Hormone replacement therapy
Smoking
Autoimmune disorders
Contact lens wear
Medications
Exposure and eye lid/blinking disorders
SECTION 6
Environmental
774
Mucus exists on the ocular surface in a structureless continuum, in granules arranged in clusters or sheets, and as ne
strands. It functions to lubricate the ocular surface and to trap
and eliminate foreign matter. Foreign particles are captured within
ne networks of conjunctival mucus which collapse and are
swept toward the medial canthus with each blink.
The measured thickness of the mucous layer is variable. It is
estimated that the adsorbed precorneal mucous layer is between
0.02 and 0.04 mm thick and that the supercial layer of dilute
mucin is ~4 mm thick.22 Using transmission electron microscopy, Nichols and co-workers showed that the mucous layer
overlying the cornea measures between 0.6 and 1.0 mm thick,
whereas the layer overlying the conjunctiva measures 2 mm to
as much as 7 mm thick in certain regions.23 It may be impossible to dene the precise thickness of the mucous layer because
the surface mucus is hydrated, and the transition between the
mucous layer and the aqueous layer is probably a gradual one,
forming a gel-like structure.24
MUCIN LAYER
AQUEOUS LAYER
The properties of the epithelial cell membrane enable wetting of

the ocular surface. In humans, the supercial cell membranes
of the cornea and conjunctiva are packed densely with microplicae and microvilli; these surface structures play a role in
creating an evenly distributed mucous layer. Located within the
conjunctival epithelium are membrane-bound subsurface vesicles that contain high-molecular-weight glycoproteins.14 These
vesicles rise to the tear-side surface of the cell and fuse with the
cell membrane, forming a glycocalyx. The mucus is in continuity
with the cell membrane glycocalyx. Mucins are critical in water
retention for wetting of the ocular surface. MUC5A is one of the
mucins secreted specically by the goblet cells. MUC 1 and
MUC 4 are membrane-spanning mucins that help in wetting of
the ocular surface and are expressed by nongoblet cell epithelia.
Conjunctival goblet cells are found throughout the bulbar and
palpebral conjunctiva, denser nasally than temporally.15 Some
cells secrete mucus directly onto the ocular surface and others
secrete into crypts that rise to the ocular surface.16 More is now
known about the regulation of mucus secretion by goblet cells.
It appears that there is a discharge of mucus from the goblet
cells when the electrolyte composition of the adjacent fluid
differs from that of normal tear fluid,17 or when the osmolarity
increases.18,19 In addition, sympathetic and parasympathetic nerves
are located adjacent to conjunctival goblet cells, and sensory stimulation of the cornea causes goblet cell mucus discharge.20,21
The aqueous layer is secreted by the main and accessory lacrimal

glands. Classically, the lacrimal secretory system has been divided
anatomically and functionally into two parts: (1) the basic
secretors (goblet cells, accessory lacrimal glands, and oil glands)
and (2) the reflex secretor (the main lacrimal gland).25
The main lacrimal gland is an exocrine gland located in the
superotemporal fossa of the orbit. It is divided into two lobes
the deeper orbital lobe and the more accessible palpebral lobe.
The acini secrete the basic tear aqueous components and the
secretory ducts modify the electrolyte and water concentration
in tears. Contractile myoepithelial cells surround the lacrimal
acini. Receptors for parasympathetic neurotransmitters such as
acetylcholine and vasointestinal peptide (VIP) have been found
in these cells, but their precise role in regulating secretion is still
being elucidated. The variability of output of the gland is
enormous from 2 mL/min to copious amounts associated with
reflex tearing in response to noxious or emotional stimuli.
The innervation of the lacrimal gland is by parasympathetic,
sympathetic, and sensory nerves. Parasympathetic neurotransmitters such as VIP and acetylcholine predominate and parasympathetic bers traveling with the lacrimal nerve stimulate
lacrimal gland secretion.26 Sympathetic nerves have norepinephrine and neuropeptide Y (NPY) neurotransmitters. Antibodies
to substance P and calcitonin gene-related peptide (CGRP) have
been found in the sensory nerves.
Wetting of the Ocular Surface and Dry Eye Disorders
TABLE 57.4. Proteins, Hormones, Neurotransmitters, and Other

Nonelectrolyte Components Identied in the Tear Film
TABLE 57.4. Proteins, Hormones, Neurotransmitters, and Other

Nonelectrolyte Components Identied in the Tear Filmcontd
a1-Antichymotrypsin
Interleukin-1b
a1-Antitrypsin
Interleukin-8
a2-Macroglobulin
Lactoferrin
Albumin
Leukotrienes
Antichlamydial factor
Lysosomal enzymes
Anticomplement factor
Lysozyme
Apolipoprotein D
Mac2-binding protein
b-1,4-Galactosyltransferase
Monoglycerides
b-Amyloid protein precursor
Monomeric immunoglobulin A
b-Lysin
Mucins Pseudomonas aeruginosa binding component(s)
b2-Macroglobulin
Peroxidase
bFGF
Plasmin
Ceruloplasmin
Plasminogen activator
Complement components C1q, C3, C3 activator, C3c, factor B, C4,

C5, and C9
Polar lipids
Convertase decay-accelerating factor
Polymeric immunoglobulin A
Prolactin
Cystatin
Properdin factor B
Cystatin D
Prostaglandin E2
Cystatin-related protein
Retinoic acid
Diesters
Secretory component
Diglycerides
Secretory immunoglobulin A
3,4-Dihydroxyphenylacetic acid
Specic leukocyte protease inhibitor
Dopamine
Sterol esters
Elastase
Superoxide
Endothelin-1
Tear lipocalins
Eosinophil cationic protein
Tenascin
Epidermal growth factor
Testosterone
Estrogens
Transferrin
Fibronectin
Transforming growth factor
Free alcohols
Transforming growth factor-1
Free fatty acids
Transforming growth factor-2
Free sterols
Tryptase
Granulocyte-monocyte colony-stimulating factor
Tumor necrosis factor
Group II phospholipase A2
Hepatocyte growth factor
Vitronectin
Histamine
CHAPTER 57
Triglycerides
Glucose
Wax esters
Homovanillic acid
Those components in bold are known to be, or may possibly be, secreted by
the lacrimal gland. The italicized lipids originate from the meibomian gland. The
information in this table was obtained from Tiffany JM: Physiological functions of
the meibomian glands. Prog Ret Eye Res 1995; 14:47 and Sullivan DA: Ocular
mucosal immunity. In: Ogra PL, Mestecky J, Lamm ME, et al, eds. Handbook of
mucosal immunology. 2nd edn. Orlando, FL: Academic Press; 1998.
Hydrocarbons
Immunoglobulin E
Immunoglobulin G
Immunoglobulin M
Inter-a-trypsin inhibitor
Interferon
Interleukin-1a
Continued
The named accessory glands include the glands of Krause,

Wolfring, and Manz. These are located in the conjunctival fornices and along the perilimbal arcade. They are thought to be
important in the basal secretion of tears. Androgen and estrogen
receptors have been found on these glands though androgen
775

function appears to be more critical. Recently, it has become clear
that the main and accessory lacrimal glands routinely work simultaneously with one another.27 Main and accessory lacrimal gland
secretion is regulated both centrally and peripherally. The average
tear flow in humans is ~1.2 mL/min and ranges between 0.5 and
2.2 mL/min.26,28 The volume of the tear fluid averages ~7 mL.26,28
About 1.1 mL of this total volume lies in the preocular lm
within the palpebral ssure, ~2.9 mL within the marginal strips,
and ~4.5 mL within the fornices. Ehlers calculated the thickness of the precorneal tear lm, based on quantitative data, to
average ~79 mm. He demonstrated that the precorneal tear lm
is thickest (~8.7 mm) immediately after blinking and gradually
thins over a 30-s period to a thickness of ~4.5 mm.29 The rate
of thinning is fastest in the 5 s immediately after the blink. This
thinning occurs as well due to evaporation,29 but is probably
related to the drainage of tears that commences on lid opening.30
Secretion of tears is also under hormonal influence, primarily
androgens. It is thought that this accounts, in part, for the
difference in tear composition between male and female gender.
Disruption of the hypothalamicpituitarygonadal axis results
in atrophy of the lacrimal gland, a decrease in fluid and protein
secretion and apoptotic cellular changes.31 Receptors for other
hormones have been identied in lacrimal tissue, including that
for alpha-MSH, prolactin, estrogens, progestin, glucocorticoids,
retinoic acid, and insulin.
Many signal transduction pathways regulating lacrimal gland
secretion have been demonstrated.32 Neurohormonal regulation
of lacrimal gland secretion interacts with the immune system.
Additional cells in the lacrimal gland include CD4 and CD8
lymphocytes, dendritic cells, macrophages, and mast cells. In
humans, plasma cells account for more than 50% of all the
mononuclear cells in lacrimal tissue.33 As an example of how
neurohormonal factors interact with the immune system, the
secretion of secretory IgA (sIgA), the predominant immunoglobulin in tears, is under the control of androgenic steroids. The
regulation of sIgA secretion is augmented or suppressed by neurotransmitters, cytokines, and secretagogs.34
The tear lm has a unique electrolyte composition that differs
from both the aqueous humor and serum. The potassium concentration is higher in the tear lm, averaging ~23 mmol/L,35
compared with ~5 mmol/L in the aqueous humor,36 and
4.5 mmol/L in serum.37 The osmolarity of the normal human
tear lm averages 302 6 (SD) mOsm/L.38 Tear osmolarity
seems to be lowest in the morning after prolonged lid closure,38
and increases somewhat as the day progresses.39
SECTION 6
LIPID LAYER
776
The most supercial layer of the tear lm is produced by the

meibomian glands in the tarsal plate, which secrete sebaceous
material at the mucocutaneous junction of the lid margin.
Blinking compresses and stretches this secretion over the tear
lm to create and maintain the supercial oily layer. The glands
of Zeis (sebaceous) and Moll (sweat) are located more anteriorly
in the lid margin and are associated with the cilia.40 The supercial oily layer measuring only 0.1 mm41 has a major role in
retarding evaporation from the tear lm.42 The lipid layer
prevents tear spillage from the ocular surface, prevents eyelid
skin damage by tears, and forms a protective seal over the ocular
surface during sleep.43 The functions of stabilizing the ocular
surface by preventing evaporation and enhancing the solubility
of tear components have also been proposed.
The meibomian glands are organized from acini of epithelial
cells which empty into ducts lined by ductal epithelium. The
ductal epithelium modulates tear lm osmolarity, pH, and composition. The germinal basal layers of ductal cells do not contain lipid. As they migrate toward the center of the duct, lipid is
synthesized and stored in secretory granules. Lipid secretory

granules stored in the acini are secreted in a holocrine fashion,
with the disintegration of mature cells lining the duct. The
orice of the meibomian glands lies just behind the mucocutaneous junction along the lengths of the tarsal plates. Blinking
controls the release of meibomian secretions into the tear lm.
The lipid is liquid at body temperature, and with each blink it
is spread across the ocular surface. The rate of synthesis of
meibomian gland lipid and the rate of secretion of the meibomian glands are influenced by hormones. Androgen receptors
mRNA and receptor proteins were discovered in the nuclei of
the meibomian acini. Enzymes involved in androgen metabolism
such as 5-alpha-reductase affect meibomian gland activity.
Rabbits, dogs, and humans treated with topical androgens have
increased production and secretion of meibomian glands, qualitative changes in meibomian gland fluid and prolonged tear
breakup time (TBUT). Conversely, androgen deciency and premature ovarian failure in humans have been associated with
altered lipid composition of tears, decreased TBUT, and functional
dry eye.32,4447
Very little is known about the neural regulation of meibomian glands. Nerve endings containing VIP and NPY have been
found around the acini and vasculature of the glands.46 The
composition of meibomian gland fluid is listed in Table 57.4.
Wax monoesters and sterol esters make up 6070% of the lipid.
Other than the preputial gland, the sebaceous glands in humans
are not neurally control.
PATHOLOGY OF THE OCULAR SURFACE IN

DRY-EYE DISORDERS
Sjgrens description of the ocular surface disease of KCS is yet
to be surpassed.3,4850 Morphologically, the conjunctiva is
affected before the cornea in KCS. Initially, there is loss of
conjunctival goblet cells50,51 and edema of the conjunctival
stroma.3,52 This is followed by intercellular edema in the deeper
layers of the conjunctival epithelium3,52,53 and then by intracellular edema as the disease progresses.47 The cytoplasmic density
of conjunctival epithelial cells53 is diminished with blunting and
loss of cell surface microplicae and eventual disruption of the
cell surface membrane.50,52,53 This is accompanied by increased
conjunctival epithelial cell desquamation.3,52 Squamous metaplasia of the conjunctiva occurs with further decrease in conjunctival goblet cell density50,5456 and it has been proposed that
this skin-like change in the ocular surface is a protective response
that inhibits further fluid loss from the ocular tissues.
Sjgren was rst to recognize that the severity of conjunctival
disease varied with the topographic location.48 He found that
the conjunctival epithelial disease and goblet cell loss were more
advanced within the interpalpebral ssure compared with bulbar
conjunctiva covered by the lids. Also, the epithelial disease in
the exposed nasal conjunctiva was typically more advanced than
disease in exposed temporal conjunctiva.49 Impression cytology
demonstrates that goblet cell loss in KCS is greater in the bulbar
conjunctiva than in the palpebral conjunctiva54 and greater in
the nasal than temporal bulbar conjunctiva.
Although the cornea is the tissue at risk for nearly all the
important and sight-threatening complications of dry eye and
pathology in the cornea correlates with high symptomatology,
most of the information we have on dry-eye pathogenesis is derived
from the tears and conjunctiva. The cornea is more resistant
than the conjunctiva to disease in KCS. Sjgren rst noted that
rose bengal (RB) would commonly stain the entire bulbar conjunctiva within the exposure zones, while staining only the
inferior cornea.3
Recently, it was shown that patients with moderate dry eye
had higher conjunctival HLA-DR-positive cells compared with

Autoimmune
Exocrinopathy
Aqueous
Insufficiency
Drying Environmental &

Physical Factors
Hyperosmolarity
Microtrauma
Hypaesthesia
Hormones & systemic

immunity
Altered
Growth
Factors
Meibomian Gland
Disease
Ocular Surface
Disease
Inflammation
Cytokines &
Chemokines
IL-1, IL-6, IL-8, TNF-a
Leukocyte
activation
Adhesion &
costimulatory
factors (ICAM-1)
Apoptosis
Proteases
Mucin
alterations
Neurogenic
Inflammation
Vascular
permeability
factors (VEGF)
FIGURE 57.1. Pathophysiology of dry-eye disorders.
controls, with HLA-DR expression pattern in mild and moderate

dry eyes reflecting disease progression.57
FIGURE 57.3. A patient with early KCS. The tear lm appears normal
before the instillation of dyes. The diagnosis of KCS was based on a
sandy-gritty irritation that was worse toward the end of the day, RB
staining of the nasal bulbar conjunctiva within the exposure zone, and
elevated tear lm osmolarity.
PATHOPHYSIOLOGY OF DRY-EYE
DISORDERS
FIGURE 57.2. Mechanisms for elevated tear lm osmolarity result in

the surface disease of KCS.
FIGURE 57.4. A patient with late KCS. A healed corneal perforation is

evident inferiorly within the palpebral ssure. Many patients with late
KCS develop a ptosis such as that seen in this patient.
AQUEOUS TEAR DEFICIENCY

Lacrimal Gland Disease
The most signicant anatomic cause of lacrimal gland dysfunction is damage from an autoimmune mechanism. Dry-eye
disease from this mechanism is called KCS (Figs 57.3 and 57.4).
Primary Sjgrens syndrome exhibits a classic triad consisting of
(1) KCS and xerostomia, (2) positive exocrine gland (salivary or
lacrimal) biopsy, and (3) autoantibodies. Secondary Sjgrens syndrome patients have an associated connective tissue disease such
as rheumatoid arthritis, systemic lupus erythematosus, scleroderma, polymyositis, or primary biliary cirrhosis among others
(Table 57.5).61
Histologically, lacrimal gland tissue from these patients shows
mononuclear cell inltration with lymphocytes, both CD4
T-helper cells and IgG-producing B cells, some follicle formation,
plasma cells, and atrophy of secretory epithelial tissue with deposition of collagen.47,62,63 This is analogous to changes occurring
in the salivary glands. Patients with primary Sjgrens syndrome
are more likely to have anti-La (SS-B) antibodies, lack antibodies
to salivary gland ducts, and have a high frequency of histocompatibility antigen HLA-DR3.6466 The actual mechanism by
CHAPTER 57
The etiology of dry-eye disease is multifactorial (Fig. 57.1). The

ocular surface, the tear-secreting glands, the neural innervations,
and the neuroendocrine factors function as an integrated unit.
When dysfunctional, this unit results in an unstable tear lm
causing ocular surface disease. Age, decrease in supportive factors
(androgen hormones), systemic inflammatory disease (rheumatoid arthritis), ocular surface diseases (HZV), trigeminal nerve
severing (laser in situ keratomileusis, LASIK), and efferent
cholinergic nerve disruption (anticholinergic drugs) are causes
of dysfunction of this integrated unit.
The current paradigm in the pathophysiology of dry eye integrates neurohormonal with immune factors that alter the lacrimal
gland tear production. Qualitative and quantitative alterations
in tears start an inflammatory cascade on the ocular surface,
which is exacerbated by long-term changes in the local epithelia,
meibomian glands, and goblet cells. The chronic inflammatory
microenvironment is amplied by costimulatory molecules, such
as ICAM-1, recruits more immune cells and induces apoptosis
of the conjunctival epithelium. The abnormal ocular surface fails
to wet properly, and a vicious cycle of inflammation is amplied
involving both soluble and cellular mediators.58,59 Clinical and
basic research suggests that this inflammation plays an important
role in the pathogenesis of dry eye (Fig. 57.2).60
777
TABLE 57.5. Summary of Revised International Classication

Criteria for Sjgrens Syndrome72
I. Ocular symptoms: a positive response to at least one of the
following questions:
1. Have you had daily, persistent, troublesome dry eyes for
more than 3 months?
2. Do you have a recurrent sensation of sand or gravel in the
eyes?
3. Do you use tear substitutes more than 3 times a day?
II. Oral symptoms: a positive response to at least one of the
following questions:
1. Have you had a daily feeling of dry mouth for more than
3 months?
2. Have you had recurrent or persistently swollen salivary
glands as an adult?
3. Do you frequently drink liquids to aid in swallowing dry
food?
III. Ocular signs: that is, objective evidence of ocular involvement
1. Schirmers I test, performed without anesthesia (5 mm in
5 min)
2. Rose bengal score or other ocular dye score (4 according to
van Bijstervelds scoring system)
IV. Histopathology: in minor salivary glands focal lymphocytic
sialoadenitis with a focus score 1, dened as a number of
lymphocytic foci (which are adjacent to normal-appearing
mucous acini and contain more than 50 lymphocytes) per
4 mm2 of glandular tissue
V. Salivary gland involvement: objective evidence of salivary gland
involvement dened by a positive result for at least one of the
following diagnostic tests:
1. Unstimulated whole salivary flow (1.5 mL in 15 min)
2. Parotid sialography showing the presence of diffuse
sialectasias
3. Salivary scintigraphy showing delayed uptake, reduced
concentration and/or delayed excretion of tracer
shown to activate inflammatory pathways in epithelial and

inflammatory cells. The production of stress-activated protein
kinases such as p38, involved in the mitogen-activated protein
kinase (MAPK) signaling pathway, c-jun NH(2) terminal kinase
(JNK), matrix metalloproteinases e.g., MMP-1,3, 9 and MMP13,
inflammatory cytokines such as IL-1, tumor necrosis factor
(TNF) alpha has been implicated in the pathogenesis of dry eye
via ocular surface inflammation.5,70,71
NEUROTROPHIC KERATITIS
Collins and associates showed that bilateral corneal anesthesia
induced by topical proparacaine decreases the blink rate by
~30%.73 Fresh tears are spread from the marginal tear strips to
the interpalpebral ocular surface only by blinking, and since the
stability of the tear lm is nite, a decreased blink rate after
bilateral corneal anesthesia can attain clinical signicance in
selected patients. Contact lens wear and neurotrophic keratitis
are two conditions among many others exhibiting decreased
corneal sensation along with decreased reflex tearing.74,75 The
effect of contact-lens-related diminished corneal sensitivity is
cumulative. It is common to see patients who have worn hard
lenses for longer than 15 years develop lens intolerance and
complain of dryness, requiring cessation of lens wear.
Although bilateral corneal anesthesia is necessary to affect
the blink rate, neurotrophic keratitis often develops with unilateral loss of corneal sensation (Table 57.6). Intact corneal
sensation partially drives tear secretion, and a decreased corneal
sensation leads to decreased tear secretion.76 Neurotrophic keratitis is, in part, a dry-eye disorder; studies have noted that, like
eyes in KCS, these eyes show abnormal RB staining, decreased
conjunctival goblet cell density, decreased corneal epithelial
glycogen, and conjunctival epithelial cell abnormalities.77
The trigeminal nerve exerts an independent trophic influence
on the cornea that may occur via axonally transported neurotransmitters and neuropeptides.78 There is evidence for such an
effect in multiple systems,79 and it appears that an analogous
mechanism plays a role in the corneal changes of neurotrophic
keratitis. Specically, there is a decrease in corneal acetylcholine, substance P, and CGRP. These substances stimulate
VI. Autoantibodies: presence in the serum of the following

autoantibodies:
1. Antibodies to Ro(SSA) or La(SSB) antigens, or both
TABLE 57.6. Major Causes of Decreased Corneal Sensation
SECTION 6
Neurotrophic keratitis (damage to the fth nerve)
778
which these inflammatory cells destroy lacrimal gland tissue,

thereby decreasing aqueous tear secretion, is unknown. The
cytokine balance in the tears and conjunctiva of Sjgrens syndrome patients are altered with enhanced expression of proinflammatory cytokines (including interleukin-1 (IL-1)).67
Decreased Lacrimal Gland Secretion and

Hyperosmolarity
There is now considerable evidence to support the theory that
elevated tear lm osmolarity is the link between decreased tear
secretion and ocular surface disease.18,19,51,52 The most powerful
evidence was obtained by developing and studying rabbit models
for dry-eye disease. In these rabbit models, the surface disease
of KCS is dependent on and proportional to increases in tear
lm osmolarity and the duration of disease.51,52,68
Besides the effects of evaporation on tear lm osmolarity, the
osmolarity of the lacrimal gland fluid itself actually increases
with decreased rates of lacrimal gland secretion.69 The hyperosmolarity in KCS is thought to be from increased electrolytes,
particularly sodium.5 Hyperosmolar saline on the mouse ocular
surface and on human cultured corneal epithelial cells has been
Corneal surgery
Limbal incisions
Penetrating keratoplasty
Lamellar keratoplasty
Radial keratotomy
Excimer laser surgery
Herpes zoster virus
Topical medications
b-Blockers
Atropine
Diabetes
Contact lens wear
Aging
Chronic ocular surface inflammation

cell mitosis, and their loss likely plays a signicant role in the
decreased corneal mitosis and healing problems seen in this
condition.80
POSTSURGICAL
Sensory denervation of the ocular surface after corneal incisional
surgery such as LASIK disrupts ocular surface tear dynamics
and causes irritation symptoms.81 LASIK signicantly alters the
TBUT, Schirmer test values, and basal tear secretion. Dry-eye
symptoms are common after LASIK surgery. Patients with preexisting aqueous deciency as demonstrated by Schirmer test
values less than 10 mm are at particular risk of experiencing dry
eye.82 Unrecognized lipid tear deciency can also be a factor
contributing to dry eye after LASIK.83 However, Toda et al demonstrated that preexisting dry eye does not affect the safety and
efcacy of LASIK.84
LIPID LAYER ABNORMALITIES
(Fig. 57.6). The 2540% of patients with meibomian gland dysfunction have an evaporative form of dry eye.88
Blepharitis associated with meibomitis was noted in 3.6% of
2520 subjects aged 65 and older in the population-based study.11
Those patients were twice as likely to have dry-eye symptoms
as those without signs of meibomitis. Patients with blepharitis
have increased bacterial loads on their eyelid margins, most
commonly Staphyloccocus epidermis, Proprionibacterium acnes,
and Corynebacterium sp.88a Similarly, a large proportion of patients
with aqueous decient dry-eye syndrome (as seen in Sjgrens
syndrome) have meibomian gland dysfunction, likely because of
the proinflammatory environment of the preocular tear lm in
these patients.
Regardless of cause, the stasis of oil within the inspissated
gland as seen in meibomian gland dysfunction results in an
inflammatory response in and around the gland, and inflammation spills over to involve the ocular surface, which may exacerbate
dry eye. Qualitative and quantitative abnormality in meibum
affects tear lm stability and consequently, ocular surface wetting.
Evaporative Dry Eye and Meibomian Gland

Dysfunction
MUCIN INSUFFICIENCY
Meibomian gland dysfunction results as a sequel or in association with meibomitis. In patients with meibomian gland
dysfunction (e.g., in association with facial rosacea) there is
progressive stenosis or closure of the meibomian gland orices
(Fig. 57.5).85 Stenosis of the meibomian gland openings
increases tear lm evaporation86 and results in an increase in
tear lm osmolarity68,87 contributing to ocular surface disease
The ocular surface epithelium expresses at least three major

mucin genes. MUC5AC is a gel-forming mucin secreted by the
goblet cells. Loss of goblet cells and derangement in conjunctival epithelium such as in burns, ocular cicatricial pemphigoid,
StevensJohnson syndrome, and vernal catarrhal create dry eye
by altering the quality and quantity of mucin of the ocular
surface.89 Altered levels of cytokine and growth factors such as
CHAPTER 57
FIGURE 57.5. In the normal lid margin (a),

meibomian gland orices are visible and the
margin is free of blood vessels. The normal
meibomian gland piano-key pattern is visible
beneath the tarsal conjunctiva when the lid is
pulled down (b). With meibomian gland orice
stenosis, the orice is no longer visible, but oil
can be expressed when pressure is applied to
the lid (c). With meibomian gland orice
closure, the orice is no longer visible and oil
cannot be expressed (d). There is distortion
and obliteration of the normal meibomian gland
piano-key pattern beneath the tarsal
conjunctiva.
FIGURE 57.6. Meibomian gland dysfunction

can be graded by examining the meibomian
gland orice. Patent orices rst become
stenosed and then close.
779

EGF in the tear lm impair ocular surface epithelium differentiation, causing hyperplastic epithelium and loss of protective
surface molecules such as MUC 1 and MUC 4.
TABLE 57.7. Conditions Associated with Dry-Eye Syndrome

Environmental factors
Reduced humidity, wind, heating, air conditioning
MECHANICAL
Exogenous irritants and allergens
Large Palpebral Widths and Abnormal Blinking

The normal blink rate averages 1015 times per minute90,91 and
blinking is critically important in spreading freshly secreted
lacrimal gland fluid.92 Abnormal eyelid apposition such as
ectropion, entropion, lagophthalmos, eyelid coloboma, or
irregular lid margin from ocular rosacea and blepharitis increase
tear evaporation and create areas of dryness.
PARADIGM AND CLINICAL EVALUATION

It is helpful clinically to recognize that any condition that
decreases tear secretion or increases tear lm evaporation has
the potential to increase tear lm osmolarity and create the
surface disease of KCS.93 Dry eye syndrome (DES) can be
pathogenically divided into three (aqueous, lipid, mucin), and
functionally into two, distinct groups (evaporative and tear
decient). However, there is considerable overlap between those
with meibomian gland dysfunction and aqueous tear deciency,
and those with allergic eye disease (Fig. 57.7).
Smoking
Systemic medications
Antiarrythmics
Anticholinergics
Antidepressants
Antihistamines
b-Blockers
Chemotherapy
Diuretics
Hormones
Isotretinoin
Systemic retinoids
Systemic viral infections
HIV
Hepatitis C
CLINICAL EVALUATION
EBV
History
Most common symptoms of dry-eye syndrome include foreignbody sensation, burning, stinging, itching, dryness, soreness,
heavy lids, photophobia, ocular fatigue, and reflex tearing. A
history of exacerbation by activities such as reading, computer
use, airplane flights, and arid or windy environments is often
elicited. These activities are associated either with decreased
blinking and/or increased evaporative loss. Other pertinent historical data such as alleviation by articial tears, the frequency
of instillation and use of overi the-counter or prescribed medications is essential. Table 57.7 lists the common medications that
can exacerbate dry eye. Enquire about contact lens use, ocular
allergy, infections (HZV, HSV, HIV, and EBV) and eyelid, corneal
and prior refractive surgery. Relevant medical history includes the
presence of connective tissue diseases and other autoimmune
diseases, diabetes mellitus, thyroid eye disease, sarcoidosis,
Parkinsons syndrome, Bells palsy, bone marrow transplant and
associated graft versus host disease, periorbital radiation or
cranial tumors.
Systemic inflammatory disease

Sjgren syndrome
Systemic lupus erythematosus
Scleroderma
Contact lens wear
Blepharitis and meibomitis
Rosacea
Lacrimal gland inltration
Lymphoma
Sarcoidosis
Hemochromatosis
SECTION 6
Amyloidosis
Cicatrizing conjunctivitis
Ocular cicatricial pemphigoid
DRY EYE
Atopic keratoconjunctivitis
Deficient Aqueous
Tear Production
Increased Evaporative Loss

Corneal surgery
Sjgren
Syndrome
Non-Sjgren
Syndrome
Blephantis/
Meibomian Gland
Dysfunction
Exposure
Other factors
1. Contact lenses
2. Blink abnormality
3. Environmental
LASIK
PRK
Penetrating keratoplasty
primary secondary
780
Lacrimal gland
Neurotrophic
absence
Reflex
infiltration
disease
Lacrimal obstruction
Cicatricial changes
FIGURE 57.7. Modied Lemp classication of dry-eye syndrome.94
Trauma
Chemical
Thermal
Continued
TABLE 57.7. Conditions Associated with Dry-Eye

Syndromecontd
laments, mucous plaques, xerosis, pannus formation, thinning,

dellen, inltrates, ulceration, scarring, and neovascularization.
The anterior chamber is usually quiet in dry-eye disorders.
Radiation
Neurological conditions
Parkinsons disease
Bells palsy
RileyDay syndrome
Examination
External observation of the face for rosacea and other skin,
eyelid, and adnexal disorders is an essential component of the
ocular examination (Fig. 57.8). The lids and lashes are examined
for the presence of lagophthalmos, infrequent blinking, floppy
eyelids, lid retraction, entropion, ectropion, notching, or blepharospasm, trichiasis, and distichiasis. Anterior and posterior
lid margins examination include observation of the meibomian
gland architecture and openings, presence of blepharitis, telangiectasia, scurf, and position and size of lacrimal puncta. The
adnexa are checked for enlargement of the lacrimal glands and
proptosis. Cranial nerves V and VII function are tested as well.
Tear lm is evaluated by noting the height of the tear meniscus
and presence of foam or debris suggestive of meibomian gland
dysfunction. The conjunctiva is examined for injection,
pinguecula/pterygium, papillae, or follicles. The inferior fornix
and tarsal conjunctiva are observed for erythema, chalasis, xerosis,
mucous threads, scarring, keratinization, foreshortening,
symblepharon, and presence of masses which interfere with tear
distribution. The bulbar conjunctiva is examined for punctate
staining with RB, fluorescein, or lissamine green dyes, hyperemia,
and xerosis. Attention is then directed to the cornea to note the
presence of interpalpebral punctuate epithelial erosions or defects,
staining with RB or fluorescein dye, as well as the presence of
Ancillary Tests
Although at this time the history is the most useful diagnostic
tool, certain tests can help stage and characterize the condition.
TBUT test, ocular surface dye staining pattern by RB
(Fig. 57.9), fluorescein (Fig. 57.10), or lissamine green and the
Schirmer tests performed in this sequence are useful. When
trigeminal nerve dysfunction is suspected, corneal sensation can
be assessed prior to instillation of any eyedrops. A laboratory
evaluation of autoimmune disorders is considered in patients
with a suggestive history such xerostomia, signicant constitutional symptoms, arthralgia, myalgia, Raynauds phenomenon,
vasculitic type rashes, history of multiple miscarriage, in young
patients with signicant dry eyes or with family history of an
autoimmune disorder.
Tear Breakup Time

Noninvasive fluorescein TBUT is one of the measures of tear
lm stability.9597 It is performed by moistening a fluorescein
strip with sterile nonpreserved saline and applying it to the
inferior tarsal conjunctiva. Fluoresceinanesthetic combination
drops are not suitable for this purpose.94 The precorneal tear
lm is examined using a broad beam of the slit lamp with a blue
FIGURE 57.9. RB staining typical for moderate KCS. The conjunctiva

stains more than the cornea, and the nasal conjunctiva stains more
than the temporal conjunctiva.
FIGURE 57.8. Rosacea, as evidenced here by rhinophema,

telangeictasia, and papules, is an important contributing factor for
dry-eye disease.
FIGURE 57.10. Fluorescein staining typical for KCS.
CHAPTER 57
Eyelid malposition
781

lter. The time lapse between the last blink and the appearance
of the rst randomly distributed dark discontinuity in the
fluorescein stained tear lm is the TBUT. Breakup times less
than 10 s are considered abnormal.94 A rapid TBUT is observed
in both aqueous tear deciency and meibomian gland disease in
the presence of lipid instability.94
Ocular Surface Staining

1. Fluorescein dye stains permeate intercellular junctions of
corneal and conjunctival epithelia which have been
disrupted.98 One to two percent sodium fluorescein
solution is used to stain the tear lm. One to 2 min after
instilling the dye, the ocular surface is examined through a
biomicroscope using a cobalt blue lter. Staining is more
intense when it is observed with a yellow lter.
2. Rose Bengal (RB) stains the conjunctiva more intensely
than the cornea (Figs 57.11 and 57.12). The dye stains
ocular surface cells that lack a mucous coating, as well as
debris in the tear lm.98 A red-free lter makes
examination easier.93 van Bijsterveld developed a scoring
system for RB dye that divides the ocular surface into three
zones: nasal bulbar conjunctiva, cornea, and temporal
bulbar conjunctiva.99 Each zone is given a score ranging
from zero to 3, with zero indicating no staining and 3
indicating essentially confluent staining. Scores for each
eye are totaled according to this system, and scores of 3.5
or greater indicate dry eye. The van Bijsterveld scoring
system has been used in clinical studies and as a clinical
tool in following dry-eye patients although other scoring
systems have also been adopted, that are more precise,
such as the National Eye Institute system.
3. Lissamine green stains dead and degenerated cells and
mucus. The staining quality of lissamine green in similar
to that of RB but is less irritating on the ocular surface. It
is not currently commercially and is not useful in
evaluating corneal epithelial disease.94
Schirmer Test
Although not diagnostic in itself, Schirmer testing is commonly
used to evaluate and conrm aqueous tear production. It is
performed by placing a narrow lter-paper strip in the inferior
cul-de-sac. Aqueous tear production is measured by the length
in millimeters that the strip wets during the test period, generally 3 or 5 min.100 Schirmer testing may be performed with or
without the use of topical anesthesia. The Schirmer test with
anesthesia, also referred to as a basic secretion test, was found
to yield more variable results than the Schirmer test done
without anesthesia.100 Excess fluid from the topical anesthetic
should be gently removed from the cul-de-sac prior to insertion
of the lter paper. Normal results are often greater than 10 mm;
results of 5 mm or less for the Schirmer test without anesthesia
are generally considered abnormal and serially consistent results
are highly suggestive. Lamberts and co-workers found that 15%
of normal subjects had basic Schirmer test results of 3 mm or
less.101 A comparable diagnostic cutoff value has not been
agreed on for the Schirmer test with anesthesia. There is still no
consensus on age-adjusted cutoff values for Schirmer testing.4
Sensitivity of the Schirmer has been noted to be poor, ranging
from 10% to 25%. Many patients with dry-eye disease yield
false-negative results.102 The physician must not rely solely on
Schirmer test results to diagnose dry-eye disease.
Fluorescein Clearance Test

Fluorescein clearance test (FCT) has been found to correlate
with dry eye by assessing the clearance or turnover of tears on
the ocular surface.77 A measured amount of fluorescein dye is
instilled onto the ocular surface and clearance of the dye is
assessed visually by comparing the residual dye of the Schirmer
strip placed in the inferior tear meniscus to a standard color
scale.77 This test provides an assessment of aqueous tear production, tear volume, and tear drainage. It correlates better with
the severity of ocular irritation symptoms and corneal fluorescein
staining than the Schirmer test.94
SECTION 6
FIGURE 57.11. Rose Bengal staining in early,

moderate, and late KCS.
FIGURE 57.12. Rose Bengal staining in early,

moderate, and late meibomitis and meibomian
gland dysfunction. Early in the disease,
inflammation predominates, whereas late in the
disease, dry-eye disease predominates. As
inflammation resolves (with treatment or
disease progression), staining of the inferior
and superior bulbar conjunctiva clears. Dry-eye
disease in these patients is based on increased
tear lm evaporation from a decient tear lm
lipid layer (right).
782
Other Tests
Conjunctival impression cytology used to analyze goblet cell
density, conjunctival epithelial cell morphology and differentiation is useful in research settings to follow therapeutic response.
Recently, a number of investigators have used impression cytology techniques to identify novel markers of disease such as
chemokine recptors and adhesion factors. Meibography can be
performed by transilluminating the lower eyelid with a cold light
source to evaluate the extent of glandular acinar dropout, which
has been reported to correlate with elevated tear osmolarity.86
TABLE 57.8 Differential Diagnosis for Possible Dry-Eye

Syndrome
Lacrimal drainage obstruction
Allergic conjunctivitis
Nocturnal lagophthalmos
Superior limbal keratitis
Dry eyelid skin
Tarsal foreign body
Other research-oriented tests include measuring tear osmolality

with an osmometer, tear proteomics including lactoferrin levels,
tear levels of cytokines, for example, TNF, EGF, and ICAM-1.
Lactoferrin is the most abundant tear protein that is secreted by
the lacrimal gland.57,94,103
Tear lactoferrin concentrations are reduced in Sjgren syndrome104 and correlate with the severity of ocular surface RB
staining.105 The Lactocard solid-phase enzyme-linked immunosorbent assay (Touch Scientic, West Chester, PA) takes
1015 min. It replaced the Lactoplate measurement which took
3 days. In a multicenter study, sensitivity, or positivity in the
presence of KCS, was 0% for mild KCS, 26.3% for moderate
KCS, and 83.3% for severe KCS. For patients in whom a diagnostic test is most needed, the sensitivity is less than 50%.
Specicity, or negativity in the absence of disease, was 98.9%.106
The Touch Tear analyzer (Raleigh, NC) is the only commercially
available tear immunoassay for lactoferrin that has been
approved by the Food and Drug Administration.94,107
Recently, a Dry Eye Workshop (DEWS) comprised of international leaders in dry-eye pathogenesis and therapy was formed
and charged with identifying the most sensitive and specic
markers of different forms of dry eye.
Natural History of Dry-Eye Syndrome

Dry eye is considered a chronic pain disorder by some authorities. Although there is often some associated neuropathy, pain
can be signicant to patients. Mild dry eye is often benign and
characterized by ocular irritation but no sight-threatening complications. However, those with moderate to severe dry eye can
develop reversible conjunctival squamous metaplasia and punctate epithelial erosions of the conjunctiva and cornea. Rarely,
patients with severe dry eye will develop complications such as
ocular surface keratinization; corneal ulceration, scarring, thinning, or neovascularization; microbial keratitis; and sterile corneal
keratolysis with possible perforation and severe visual loss.
Patients at maximal risk of perforation are the elderly patients
with rheumatoid disease and secondary Sjgrens syndrome and
those with severe exposure keratopathy.
The patient complaining of chronic eye irritation can have any
one of many causes for the ocular discomfort (Table 57.8). Often
a dry-eye disorder is the culprit, but frequently there are other
causes instead, or compounding dry eye. The physician must
rst determine a differential diagnosis for the patients discomfort and then narrow this list to a specic diagnosis or diagnoses.
Identifying causative factors such as environmental exposure,
prolonged visual efforts, as well as circumstances that improve
the symptoms help in making the diagnosis. Clinical observations and tests also aid in narrowing the diagnosis. For example,
prolonged driving at night, or long hours in front of a computer
terminal are associated with diminished blinking and enhanced
tear evaporation, characteristic of dry eye. Conversely, intense
Mucus shing syndrome

Blepharospasm
itching is rarely due to dry eye alone and suggests an atopic

etiology. It is important to determine the basis for the symptoms because, as discussed later, therapy varies.
TREATMENT
Patient education about the natural history and chronicity of
the dry-eye disorder is crucial to successful management of this
condition. Emphasis should be on setting realistic expectations
of managing a chronic condition rather than providing a cure.
Patients may rarely require professional counseling to help deal
with chronic pain. Alleviation of modiable factors such as air
drafts and humidity of surroundings is essential. Elimination of
responsible medications may be considered, if safe, though this
is often not practical. The widespread use of computers has led
to increasing awareness of computer vision syndrome and ways
to address it are promoting the ergonomics of computer workstations, special computer reading aids, and altering work habits.108
For example, lowering the terminal to maintain a lower position
of the upper eyelid diminishes globe exposure and often helps
alleviate symptoms. Medical therapies are used in most cases to
treat dry eyes. Patients with systemic autoimmune disease such
as rheumatoid arthritis should be managed with their internist
as antiinflammatory and immunosuppressive therapy may be
appropriate. Surgery is reserved only for patients with moderate
to severe disease in whom medical treatment was insufcient to
alleviate the symptoms (Tables 57.9 and 57.10).
WARM COMPRESSES AND LID HYGIENE

Warm compresses are indicated in patients with meibomitis or
meibomian gland dysfunction. We instruct patients to place a
clean washcloth under hot water and then apply it to closed lids
while massaging both upper and lower lids with their ngertips.
The massaging action combined with heat helps express lipid into
the tear lm, preventing retention of lipid within the meibomian gland which may be a stimulus for inflammation. We
recommend that patients perform this procedure once or twice
a day, for at least 10 min each time. Lid hygiene consists of
cleansing the crust and scurf from the eye lashes. Traditionally,
diluted baby shampoo has been advocated as it removes excess
oil and debris along the eyelid margins without irritating the
ocular surface if inadvertently splashed into the eye. In general,
while lid hygiene measures can help the management of these
patients, long-term compliance can be problematic. Patients
reluctant to comply with this are reminded that in the long
term, these time-consuming procedures could assist them in
better management of their condition and symptoms.
CHAPTER 57
Novel Laboratory Tests
783
TABLE 57.9. Treatment of DES
ANTIINFLAMMATORY THERAPY
Environmental modication: environmental humidity, goggles
Cyclosporin A
Lid hygiene and warm compresses for blepharitis

Medical
Articial tears, gels, emulsions, ointment
Topical steroids
Cyclosporine
Doxycycline
Punctal occlusion
Others
Autologous serum
Secretagogs
Essential fatty acids
Contact lens for lamentary keratitis
Surgery
Lid malposition repair
Tarsorrhaphy
Punctoplasty
Upper eyelid Botox to induce protective ptosis
Limbal stem cell transplant
Inflammation is seen consistently in different forms of dry eye

and dry-eye-associated complications, and many patients respond
therapeutically to antiinflammatory treatments. Topical cyclosporin A, a fungal-derived molecule, used extensively in organ
transplantation, was rst used in a canine model of KCS and
found to prevent T cell activation and inflammatory cytokine
production.94 It received FDA approval in early 2003 for patients
with moderate to severe aqueous insufciency as a result of
inflammation. Topical cyclosporin decreases ocular surface
inflammation and results in an improvement in Schirmer test
results and punctuate staining.109,110 An increase in goblet cell
numbers in both non-Sjgrens and Sjgrens syndrome dry eye,
as well as a decrease in epithelial cell turnover has also being
reported110a Cyclosporin also has been documented to be helpful
in healing paracentral rheumatoid corneal ulceration.111 Topical
application of the 0.1% emulsion for up to 3 years was found to
be safe in phase III studies, although currently only the 0.05%
emulsion is commercially available. (The most common side
effects are burning (11%), stinging upon instillation (4%), and
conjunctival hyperemia (3.4%).112) No serious systemic side effects
were seen. Onset of therapeutic benets of topical cyclosporin is
not immediate and is typically achieved in about a month.
Many clinicians use a concurrent mild steroid to quieten the
inflammatory component while awaiting the onset of topical
cyclosporins antiinflammatory effect.
Topical Corticosteroid
TABLE 57.10. Emerging Treatments
Novel immunosuppressives and immunomodulators
Topical hormonal therapy
Antiinflammatory agents
Tear component replacement
Lactoferrin
Endothelin-1
HGF
Lysozyme
TGFa
NGF
Lipocalin
TGFa
Retinoids
EGF
TGFb
Meibomian lipids
Secretogogs
Lacrimal
SECTION 6
Pilocarpine and Cevimeline

Mucin
P2Y2 agonists (INS365)
TEAR SUPPLEMENTATION
784
The use of articial tears, emulsions, gels, and ointments can

be very useful. Hypotonic and electrolyte-balanced tear substitutes are preferable, and nonpreserved forms are recommended
if tears are to be used more than four to six times a day, to minimize the chances of preservative-induced toxicity to surface
epithelial cells. If symptoms or signicant surface drying persists despite the above measures, or if the patient is unable or
unwilling to instill tears at the required frequency, punctal occlusion can be considered. Spectacle side shields, moisture inserts,
and moisture chambers are noninvasive therapies that can be
used to decrease evaporation, but are often poorly accepted due to
poor cosmesis.
Corticosteroids have been reported to decrease ocular irritation

symptoms, decrease corneal fluorescein staining, and improve
lamentary keratitis. Low-dose corticosteroid therapy can be
used at infrequent intervals for short-term (2 weeks) suppression
of discomfort and epithelial disease secondary to inflammation.
Topical corticosteroids are often used before or in conjunction
with starting topical cyclosporin therapy, with a brief overlap
period of a few weeks. The long-term side effects of corticosteroids,
including cataract and steroid response glaucoma preclude their
long-term use for management of dry eyes and patients should
always be monitored for complications.
TETRACYCLINES
Systemic tetracyclines are useful in treating posterior blepharitis
or meibomitis with or without ocular rosacea.113118 In addition
to its antibacterial properties, tetracycline inhibits collagenase
activity119121 and decreases leukocyte chemotaxis and phagocytosis.122125 These antiinflammatory properties are attributed
to the effect of tetracycline on meibomitis. Tetracycline is contraindicated in pregnant or nursing women. Patients with gastric
ailments can be given low-dose doxycycline (2550 mg/day) instead
of the more common dose of 50100 mg/day. Minocycline
(100 mg/day) is also effective and only uncommonly causes the
dermal photosensitivity that is seen with tetracycline. A trial of
24 months of treatment is often helpful, although some patients
are on doxycycline indenitely. Long-term use of antibiotics
should be used with caution in view of potential alteration of
intestinal microflora, changes in the immune system with
serious associated risks.125a
AUTOLOGOUS SERUM DROPS

Autologous serum tears have been reported to improve ocular
irritation symptoms and conjunctival and corneal dye staining
in patients with Sjgren syndrome.47 Since serum also contains
a variety of growth factors, it can also be useful in neurotrophic
dry eye.
Filamentary keratopathy can be treated with removal of the

laments with a jewelers forceps or cotton tip applicator and
instillation of topical mucolytic agents such as 1020% acetylcysteine. Soft contact lenses are effective in preventing recurrence
of lamentary keratopathy although they are not well tolerated
in patients with severe dry eyes as they can exacerbate dryness.
Contact lenses should generally be avoided in neurotrophic
keratopathy.94
SECRETAGOGS
Oral medications such as cholinergic agonists, pilocarpine
(Salagen) and cevimeline (Evoxac), have been approved by the
Food and Drug Administration to treat the symptoms of dry
mouth in patients with Sjgren syndrome. These medications
bind to muscarinic receptors and are thought to improve secretion of salivary, sweat, and lacrimal glands. Most clinical studies
demonstrate greater improvement in dry mouth than dry
eye.126128 Pilocarpine, at a dose of 5 mg orally four times a day,
improved patients ability to focus their eyes during reading and
reduced symptoms of blurred vision compared to placebo-treated
patients.128 Excessive sweating occurred in over 40% of patients,
limiting the tolerance to this medication. Cevimeline, another
oral cholinergic agonist, may have fewer adverse systemic side
effects and be better tolerated than oral pilocarpine due to more
selective receptor binding. It also has been found to improve
ocular irritation symptoms and aqueous tear production at the
30-mg dose, but has not been approved by the DFA for dry-eye
treatment.128
OMEGA-3 FATTY ACIDS

Tears contain essential fatty acids, both omega-3 and omega-6,
which are not manufactured by the body and only obtained
through diet. Essential fatty acids are found in various foods,
such as flaxseed, blackcurrant seed, canola oil, walnuts, soy,
and mainly cold-water sh including mackerel, tuna, salmon,
sardines, and herring. A recent study evaluated whether high
intake of omega-3-containing foods has a potentially protective
role in dry eye. As part of the Womens Health Study (WHS),
32 470 women aged 4584 years that provided information on
diet and DES, were studied cross-sectionally. After adjustment
for demographic factors, hormone therapy, and total fat intake,
this study showed that a higher ratio of omega-6 to omega-3 fatty
acid consumption was associated with a signicantly increased
risk of DES and suggested that a higher dietary intake of omega3 fatty acid may decrease incidence of DES in women.129
SURGICAL TREATMENT
PUNCTAL OCCLUSION
William Beetham introduced the use of punctal occlusion as a
treatment for dry eye in 1935.130 Beetham described the use of
electrocautery and presented data showing that this procedure
reduced ocular surface disease as indicated by staining. Subsequent studies performed in the 1980s conrmed his
ndings.131,132 Dohlman hypothesized that punctal occlusion is
helpful for dry eye by decreasing elevated lm osmolarity;133
this was later demonstrated by Gilbard and associates.134,135
Punctal occlusion probably decreases tear lm osmolarity by

increasing the tear volume. It is useful in patients with aqueous
tear deciency, neurotrophic keratopathy and those with incomplete eyelid closure such as after blepharoplasty. Temporary collagen plugs may be used as therapeutic trials. Plugs are usually
made of silicone or thermal labile polymer. Retention of plugs
can be an issue for certain patients and permanent punctual
closure using thermal cautery or laser may be indicated in these
cases. A study by Balaram et al showed that nearly one-third of
punctual plugs are spontaneously extruded 6 months after
insertion; subsequent re-plugging is associated with an even
higher rate of plug loss.136 Cauterization may be done at the slit
lamp or with the aid of magnifying loupes. Lidocaine (24%) is
injected or applied on a pledget to the area around the punctum.
After conrmation of anesthesia, a low-temperature cautery is
inserted into the distal canaliculus and turned on for 1.52 s.
The instrument is then withdrawn from the punctum. It is
helpful to compress the wire loop on the cautery tip before use.
This narrows the loop and permits the distal canaliculus and
punctum to close more adequately in response to the heat.
Usually the inferior puncta are closed rst, and if necessary
based on the persistence of symptoms, the superior puncta can
be closed later.
SURGERY
Many patients with severe dry-eye disease develop a protective
ptosis. Perhaps this develops because these patients have a tendency to rub their eyes. Lid inflammation and associated edema
can also contribute to this. A smaller palpebral ssure width
decreases the evaporative stress on the tear lm and ocular
surface. Likewise, tarsorrhaphy surgically decreases the interpalpebral surface area. It is used as a last resort in severe dry-eye
disease, usually in the context of a persistent epithelial defect or
corneal ulceration.
Amniotic membrane, conjunctival, limbal stem cell, parotid
duct, and salivary gland transplantation, as well as keratoprosthesis are surgical alternatives used in the most severe circumstances and will not be discussed further in this chapter.
CORRECTION OF LID ABNORMALITIES

Oculoplastic surgery may correct eyelid misalignment such
as entropion, ectropion, and lagophthalmos to decrease tear
evaporation and improve symptomatology. Patients with
chronic meibomian gland dysfunction may have trichiatic
lashes, and their correction is helpful in decreasing surface
epitheliopathy.
CONCLUSION
Dry eye is a chronic condition with signicant personal, medical, and economic burden. Great strides have been made in our
understanding of its pathogenesis and more exciting therapies
are in the horizon.
CHAPTER 57
BANDAGE SOFT CONTACT LENS
ACKNOWLEDGMENTS
Portions of this chapter were taken from Gilbard JP. Dry eye disorders. In:
Albert DM, Jakobiec FA, eds. Principles and practice of ophthalmology.
Phildelphia: WB Saunders; 1994:257276, and the authors would like to
thank Jeffrey P Gilbard, MD for his valuable comments on this chapter.
785
SECTION 6
REFERENCES
786
1. Haas EB: The pathogenesis of

keratoconjunctivitis sicca. Ophthalmologica
1964; 147:118.
2. Schiffman RM, Walt JG, Jacobsen G, et al:
Utility assessment among patients with dry
eye disease. Ophthalmology 2003;
110:14121419.
3. Sjgren H: Keratoconjunctivitis sicca. In:
Ridley F, Sorsby A, eds. Modern trends in
ophthalmology. London: Butterworth;
1940:403413.
4. Lemp MA: Report of the National Eye
Institute/Industry Workshop on Clinical trials
in dry eyes. CLAO 1995; 2122121232.
5. Pflugfelder SC, Stern ME, Beuerman RW,
eds: Dysfunction of the lacrimal functional
unit and its impact on tear lm stability and
composition. In: Dry eye and ocular surface
disorders. Dekker; 2004:7576.
6. McDonnell PJ, Doyle JJ, Stern L, et al:
A modied Delphi technique to obtain
consensus on the treatment of
dysfunctional tear syndrome. Invest
Ophthalmol Vis Sci. 2004;45: E-Abstract
3909. Poster presented at 2004 ARVO
Annual Meeting. Ft Lauderdale, FL,
2004.
7. Dogru M, Stern ME, Smith JA, et al:
Changing trends in the denition and
diagnosis of dry eyes. Am J Ophthalmol
2005; 140:507508.
8. Murube J, Nemeth J, Hoh H, et al: The
triple classication of dry eye for practical
clinical use. Eur J Ophthalmol 2005;
15:660667.
9. Schaumberg DA, Sullivan DA, Dana MR:
Epidemiology of dry eye syndrome. Adv
Exp Med Biol 2002; 506(Pt B):989998.
10. Schaumberg DA, Sullivan DA, Buring JE,
Dana MR: Prevalence of dry eye syndrome
among US women. Am J Ophthalmol 2003;
136:318326.
11. Schein OD, Munoz B, Tielsch JM, et al:
Prevalence of dry eye among the elderly.
Am J Ophthalmol 1997; 124:723728.
12. Moss SE, Klein R, Klein BE: Prevalence of
and risk factors for dry eye syndrome. Arch
Ophthalmol 2000; 118:12641268.
13. Schaumberg DA, Buring JE, Sullivan DA,
Dana MR: Hormone replacement therapy
and dry eye syndrome. JAMA 2001;
286:21142119.
14. Dilly PN, Mackie IA: Surface changes in the
anesthetic conjunctiva in man, with special
reference to the production of mucus from
a non-goblet cell source. Br J Ophthalmol
1981; 65:833842.
15. Kessing SV: Topographical quantitative
studies of the conjunctival goblet cells.
Arch Ophthalmol 1968; 95(Suppl):36.
16. Greiner JV, Allansmith MR: Effect of contact
lens wear on the conjunctival mucous
system. Ophthalmology 1981; 88:821832.
Ophthalmic solutions, the ocular surface,
and a unique therapeutic articial tear
formulation. Am J Ophthalmol 1989;
107:348355.
18. Gilbard JP: Effect of hyperosmolarity on
ocular surface epithelium in vivo. In:
Proceedings of the VIIth Congress of the
European Society of Ophthalmology,
Helsinki, May 2125, 1984. Helsinki:
European Society of Ophthalmology;
1985:354358.
19. Huang AJW, Belldegrun R, Hanninen L,

et al: Effects of hypertonic solutions on
conjunctival epithelium and mucinlike
glycoprotein discharge. Cornea 1989;
8:1520.
20. Dartt DA, McCarthy DM, Mercer HJ, et al:
Localization of nerves adjacent to goblet
cells in rat conjunctiva. Curr Eye Res 1995;
14:9931000.
21. Kessler TL, Mercer HJ, Zieske JD, et al:
Stimulation of goblet cell mucous secretion
by activation of nerves in rat conjunctiva.
Curr Eye Res 1995; 14:985992.
22. Dohlman CH, Lemp MA, English FP: Dry
eye syndromes. Int Ophthalmol Clin 1970;
10:215251.
23. Nichols BA, Chiappino ML, Dawson CR:
Demonstration of the mucous layer of the
tear lm by electron microscopy. Invest
24. Prydal JI, Artal P, Woon H, Campbell FW:
Study of the human precorneal tear lm
thickness and structure using laser
interferometry. Invest Ophthalmol Vis Sci
1992; 33:20062011.
25. Jones L: The lacrimal secretory system and
its treatment. Am J Ophthalmol 1966;
62:4760.
26. Botelho SY, Hisada M, Fuenmayor N:
Functional innervation of the lacrimal gland
in the cat: origin of secretomotor bers in
the lacrimal nerve. Arch Ophthalmol 1966;
76:581588.
27. Scherz W, Dohlman CH: Is the lacrimal
gland dispensable? Keratoconjunctivitis
sicca after lacrimal gland removal. Arch
Ophthalmol 1975; 93:281283.
28. Mishima S, Gasset A, Klyce SD, Baum JL:
Determination of tear volume and tear flow.
29. Ehlers N: The thickness of the precorneal
tear lm. Acta Ophthalmol 1965;
81(Suppl):1134.
30. Doane MG: Blinking and the mechanics of
the lacrimal drainage system.
31. Sullivan DA, Wicham LA, Krenzer KL, et al:
Aqueous tear deciency in Sjogrens
syndrome: possible causes and potential
treatment. In: Pleyer U, Hartmann C, Sterry
W, eds. Oculodermal diseases
immunology of bulous oculo-mucocutaneous disorders. Buren, The
Netherlands, Acolus Press, 199:95.
32. Dartt DA, Sullivan DA: Wetting of the ocular
surface. Albert, Jakobiec, eds. 960981.
33. Wieczorek R, Jakobiec FA, Sacks EH,
Knowles DM: The immunoarchictecture of
the normal human lacrimal gland.
Relevancy for understanding pathologic
conditions. Ophthalmology 1988;
95:100109.
34. Sullivan DA: Ocular mucosal immunity.
In: Ogra PL, Mesteck J, Lamm ME, et al,
eds. Handbook of mucosal immunology.
2nd edn. Orlando, FL: Academic Press;
1998.
35. Gilbard JP: Human tear lm electrolyte
concentrations in health and dry-eye
disease. Int Ophthalmol Clin 1994;
34:2736.
36. Lam K-W, Lee P-F, Fox R: Aqueous
ascorbate concentration in hereditary
buphthalmic rabbits. Arch Ophthalmol
1976; 94:15651567.
37. Van Haeringen NJ: Clinical biochemistry of

tears. Surv Ophthalmol 1981; 26:8496.
38. Terry JE, Hill RM: Human osmotic tear
pressure: diurnal variations and the closed
39. Farris RL, Stuchell RN, Mandel ID: Tear
osmolarity variation in the dry eye. Trans
40. Bron AJ, Tripathi RC, Tripathi BJ, eds: The
ocular appendages: eyelids, conjunctiva
and lacrimal apparatus. In: Wolffs anatomy
of the eye and orbit. London: Chapman &
Hall Medical; 1997:3084.
41. Holly FJ, Lemp MA: Tear physiology and
dry eyes. Surv Ophthalmol 1977; 22:6987.
42. Iwata S, Lemp MA, Holly FJ, Dohlman CH:
Evaporation rate of water from the
precorneal tear lm and cornea in the
43. Tiffany JM: Physiological functions of the
meibomian glands. Prog Ret Eye Res 1995;
14:37.
44. Wickham LA, Gao J, Toda I, et al:
Identication of androgen, estrogen and
progesterone receptor mRNAs in the eye.
45. Rocha EM, Wickham LA, da Silveira LA,
et al: Identication of androgen receptor
protein and 5 alpha-reductase mRNA in
human ocular tissues. Br J Ophthalmol
2000; 84:7684.
46. Sullivan DA, Rocha EM, Ullman MD, et al:
Androgen regulation of the meibomian
gland Adv Exp Med Biol 1998;
438:327331.
47. Chung CW, Tigges M, Stone RA:
Peptidergic innervation of the primate
meibomian gland. Invest Ophthalmol Vis
Sci 1996; 37:238245.
48. Sjgren H: A new conception of
keratoconjunctivitis sicca. (Hamilton JB,
translator). Sydney: Australasian Medical
Publishing; 1943.
49. Sjgren H: Some problems concerning
keratoconjunctivitis sicca and the siccasyndrome. Acta Ophthalmol 1951;
29:3347.
50. Sjgren H, Bloch KJ: Keratoconjunctivitis
sicca and the Sjgren syndrome. Surv
51. Abdel-Khalek LMR, Williamson J, Lee WR:
Morphological changes in the human
conjunctival epithelium. II. In
keratoconjunctivitis sicca. Br J Ophthalmol
1978; 62:800806.
52. Gilbard JP, Rossi S, Gray K: A new rabbit
model for keratoconjunctivitis sicca. Invest
53. Gilbard JP, Rossi SR, Gray KL, et al: Tear
lm osmolarity and ocular surface disease
in two rabbit models for
keratoconjunctivitis sicca. Invest
54. Meyer E, Scharf Y, Schechner R, et al:
Light and electron microscopic study of the
conjunctiva in sicca syndrome.
55. Ralph RA: Conjunctival goblet cell density
in normal subjects and in dry eye
syndromes. Invest Ophthalmol Vis Sci
1975; 14:299302.
56. Nelson JD, Havener VR, Cameron JD:
Cellulose acetate impressions of the ocular
surface. Dry eye states. Arch Ophthalmol
1983; 101:18691872.
57. Rolando M, Barabino S, Mingari C, et al:

Distribution of conjunctival HLA-DR
expression and the pathogenesis of
damage in early dry eyes. Cornea 2005;
24:951954.
58. Nelson JD, Wright JC: Conjunctival goblet
cell densities in ocular surface disease.
59. Solomon A, Dursun D, Liu Z, et al: Pro- and
anti-inflammatory forms of interleukin-1 in
the tear fluid and conjunctiva of patients
with dry-eye disease. Invest Ophthalmol Vis
Sci 2001; 42:22832292.
60. Kunert KS, Tisdale AS, Stern ME, et al:
Analysis of topical cyclosporine treatment
of patients with dry eye syndrome: effect
on conjunctival lymphocytes. Arch
Ophthalmol 2000; 118:14891496.
61. Pflugfelder SC, Solomon A, Stern ME: The
diagnosis and management of dry eye: a
twenty-ve-year review. Cornea 2000;
19:644649.
62. Bloch K, Buchanan W, Wohl M, et al:
Sjgrens syndrome. A clinical,
pathological, and serological study of 62
cases. Medicine 1965; 44:187231.
63. Font R, Yanoff M, Zimmerman LE: Benign
lymphoepithelial lesion of the lacrimal gland
and its relationship to Sjgrens syndrome.
Am J Clin Pathol 1967; 48:365376.
64. Williamson J, Gibson AAM, Wilson T, et al:
Histology of the lacrimal gland in
keratoconjunctivitis sicca. Br J Ophthalmol
1973; 57:852858.
65. Kassan SS, Gardy M: Sjgrens syndrome:
an update and overview. Am J Med 1978;
64:10371046.
66. Fox RI, Howell FV, Bone RC, et al: Primary
Sjgren syndrome: clinical and
immunopathologic features. Semin Arthritis
Rheum 1984; 14:77105.
67. Akata F, Pflugfelder SC, Lee SF, et al:
Immunocytologic features of lacrimal gland
biopsies in Sjgrens syndrome. Invest
68. Gilbard JP, Dartt DA: Changes in rabbit
lacrimal gland fluid osmolarity with flow
rate. Invest Ophthalmol Vis Sci 1982;
23:804806.
69. Pflugfelder SC, Jones D, Ji Z, et al: Altered
cytokine balance in the tear fluid and
conjunctiva of patients with Sjogrens
syndrome keratoconjunctivitis sicca. Curr
Eye Res 1999; 19:201211.
70. Luo L, Li DQ, Corrales RM,
Pflugfelder SC: Hyperosmolar saline is a
proinflammatory stress on the mouse
ocular surface. Eye Contact Lens 2005;
31:186193.
71. Li DQ, Chen Z, Song XJ, et al: Stimulation
of matrix metalloproteinases by
hyperosmolarity via a JNK pathway in
human corneal epithelial cells. Invest
72. Vitali C, Bombardieri S, Jonsson R, et al:
Classication criteria for Sjogrens
syndrome: a revised version of the
European criteria proposed by the
American-European Consensus Group. Ann
Rheum Dis 2002; 61:554558.
73. Collins M, Seeto R, Campbell L, et al:
Blinking and corneal sensitivity. Acta
Ophthalmol 1989; 67:525531.
74. Millodot M: Corneal sensitivity. Int
75. Millodot M: Clinical evaluation of an
extended wear lens. Int Contact Lens Clin
1984; 11:16.
76. Jordan A, Baum J: Basic tear flow. Does it

exist? Ophthalmology 1980; 87:920930.
77. Gilbard JP, Rossi SR: Tear lm and ocular
surface changes in a rabbit model of
neurotrophic keratitis. Ophthalmology
1990; 97:308312.
78. Beuerman RW, Schimmelpfennig B:
Sensory denervation of the rabbit cornea
affects epithelial properties. Exp Neurol
1980; 69:196201.
79. Markelonis GJ, Oh TH: A protein fraction
from peripheral nerve having neurotrophic
effects on skeletal muscle cells in culture.
Exp Neurol 1978; 58:285295.
80. Cavanagh HD, Colley AM: The molecular
Ophthalmol 1989; 192(Suppl):115134.
81. Battat L, Macri A, Dursun D, Pflugfelder SC:
Effects of laser in situ keratomileusis on
tear production, clearance, and the ocular
surface. Ophthalmology 2001;
108:12301235.
82. Yu EY, Leung A, Rao S, Lam DS: Effect of
laser in situ keratomileusis on tear stability.
Ophthalmology 2000; 107:21312135.
83. Di Pascuale MA, Liu TS, Trattler W,
Tseng SC: Lipid tear deciency in
persistent dry eye after laser in situ
keratomileusis and treatment results of new
eye-warming device. J Cataract Refract
Surg 2005; 31:17411749.
84. Toda I, Asano-Kato N, Hori-Komai Y,
Tsubota K: Laser-assisted in situ
keratomileusis for patients with dry eye.
Tear lm and ocular surface changes
after closure of the meibomian gland
orices in the rabbit. Ophthalmology 1989;
96:11801186.
86. Robin JB, Jester JV, Nobe J, et al: In vivo
transillumination biomicroscopy and
photography of meibomian gland
dysfunction. Ophthalmology 1985;
92:14231426.
87. Mishima S, Maurice DM: The oily layer of
the tear lm and evaporation from the
corneal surface. Exp Eye Res 1961;
1:3945.
88. Shields WJ, Mathers WD, Roberts J, et al:
Criteria for the evaluation of lid margin in
31(Suppl):483.
88a. Groden LR, Murphy B, Rodnite J,
Genvert GI: Cornea. Lid flora in blepharitis.
Cornea 1991; 10:5053.
89. Gulati A, Dana MR: Keratoconjunctivitis
sicca: clinical aspects. In: Foster CS,
DT Azar, Dohlman CH, eds. Smolin and
Thoftss the cornea: scientic foundations
and clinical practice. 4th edn. Philadelphia,
PA: Lippincott Williams & Wilkins; 2005.
90. McEwen WK: Secretion of tears and
blinking. In: Davson H, ed. The eye. New
York: Academic Press; 1962:271305.
91. Carney LG, Hill RM: The nature of normal
blinking pattern. Acta Ophthalmol 1982;
60:427.
92. Mishima S: Some physiological aspects of
the precorneal tear lm. Arch Ophthalmol
1965; 73:233241.
93. Gilbard JP: Tear lm osmolarity and
11:243250.
94. Matoba AY, Harris DJ, Meisler DM, et al:
Preferred practice pattern, dry eye
syndrome. American Academy of
Ophthalmology; 2003.
95. Gilbard JP, Rossi SR, Gray Heyda K: Tear

lm and ocular surface changes after
closure of the meibomian gland orices in
the rabbit. Ophthalmology 1989;
96:11801186.
96. Mengher LS, Bron AJ, Tonge SR,
Gilbert DJ: A non-invasive instrument for
clinical assessment of the pre-corneal tear
lm stability. Curr Eye Res 1985; 4:1.
97. Mengher LS, Bron AJ, Tonge SR,
Gilbert DJ: Effect of fluorescein instillation
on the pre-corneal tear lm stability. Curr
Eye Res 1985; 4:912.
98. American Academy of Ophthalmology:
Comprehensive adult medical eye
evaluation, preferred practice pattern. San
Francisco: American Academy of
Ophthalmology; 2000.
99. Van Bijsterveld OP: Diagnostic tests in the
sicca syndrome. Arch Ophthalmol 1985;
82:1014.
100. Balik J: The lacrimal fluid in
keratoconjunctivitis sicca: a quantitative
and qualitative investigation. Am J
Ophthalmol 1952; 35:17731782.
101. Lamberts DW, Foster CS, Perry HD:
Schirmer test after topical anesthesia and
the tear meniscus in normal eyes. Arch
Ophthalmol 1979; 97:10821085.
102. Farris RL, Gilbard JP, Stuchell RN, et al:
Diagnostic tests in keratoconjunctivitis
sicca. CLAO J 1983; 9:2328.
103. Holm S: Keratoconjunctivitis sicca and the
sicca syndrome. Acta Ophthalmol 1949;
33(Suppl):1.
104. Scherz W, Doane MG, Dohlman CH: Tear
volume in normal eyes and
keratoconjunctivitis sicca. Graefes Arch
105. Gilbard JP, Farris RL, Santamaria J II:
Osmolarity of tear microvolumes in
1978; 96:677681.
106. Shearn MA: Ocular aspects. In: Smith LH,
ed. Sjgrens syndrome. Philadelphia, PA:
107. McCollum CJ, Foulks GN, Bodner B,
et al: Rapid assay of lactoferrin in
keratoconjunctivitis sicca. Cornea 1994;
13:505508.
108. Blehm C, Vishnu S, Khattak A, et al:
Computer vision syndrome: a review. Surv
Ophthalmol 2005; 50:253262.
109. Gunduz K, Ozdemir O: Topical cyclosporin
treatment of keratoconjunctivitis sicca in
secondary Sjogrens syndrome. Acta
110. Sall K, Stevenson OD, Mundorf TK,
Reis BL: Two multicenter, randomized
studies of the efcacy and safety of
cyclosporine ophthalmic emulsion in
moderate to severe dry eye disease.
Cyclosporin A Phase 3 Study Group.
110a.Kunert KS, Tisdale AS, Gipson IK:
Goblet cell numbers and epithelial
proliferation in the conjunctiva of patients
with dry eye syndrome treated with
cyclosporine. Arch Ophthalmol 2002;
120:330337. Erratum: Arch Ophthalmol
2002.
111. Kervick GN, Pflugfelder SC, Haimovici R,
et al: Paracentral rheumatoid corneal
ulceration. Clinical features and
cyclosporine therapy. Ophthalmology 1992;
99:8088.
112. Barber LD, Pflugfelder SC, Tauber J,
Foulks GN: Phase III safety evaluation of
CHAPTER 57
787
113.
114.
115.
116.
117.
118.
119.
120.
SECTION 6
121.
788
cyclosporine 0.1% ophthalmic emulsion

administered. Ophthalmology 2005;
112:17901794.
Jenkins MS, Brown SI, Lempert SL, et al:
Ocular rosacea. Am J Ophthalmol 1979;
88:618622.
McCulley JP: Blepharoconjunctivitis. Int
Salamon SM: Tetracyclines in
29:265275.
McCulley JP, Dougherty JM: Blepharitis
associated with acne rosacea and
seborrheic dermatitis. Int Ophthalmol Clin
1985; 25:159172.
Browning DJ, Proia AD: Ocular rosacea.
Bowman RW, Miller KN, McCulley JP:
Diagnosis and treatment of chronic
blepharitis. In: Wagner MD, ed. Clinical
modules for ophthalmologists. Module 10:
focal points 1989. San Francisco: American
Academy of Ophthalmology; 1989.
Golub LM, Lee HM, Lehrer G, et al:
Minocycline reduces gingival collagenolytic
activity during diabetes. Preliminary
observations and a proposed new
mechanism of action. J Periodont Res
1983; 18:516526.
Golub LM, Ramamurthy N, McNamara TF,
et al: Tetracyclines inhibit tissue
collagenase activity: a new mechanism
in the treatment of periodontal disease.
J Periodont Res 1984; 19:651655.
Golub LM, Wolff M, Lee HM, et al: Further
evidence that tetracyclines inhibit
collagenase activity in human crevicular

fluid and from other mammalian sources.
J Periodontal Res 1985; 20:1223.
122. Martin RR, Warr GA, Couch RB, et al:
Effects of tetracycline on leukotaxis.
J Infect Dis 1974; 129:110116.
123. Belsheim J, Gnarpe H, Persson S:
Tetracyclines and host defense
mechanisms: interference with leukocyte
chemotaxis. Scand J Infect Dis 1979;
11:141145.
124. Esterly NB, Koransky JS, Furey NL, et al:
Neutrophil chemotaxis in patients with
acne receiving oral tetracycline therapy.
Arch Dermatol 1984; 120:13081313.
125. Elewski BE, Lamb BAJ, Sams WM, et al: In
vivo suppression of neutrophil chemotaxis
by systemically and topically administered
tetracycline. J Am Acad Dermatol 1983;
8:807812.
125a.Velicer CM, Heckbert SR, Lampe JW,
et al: Antibiotic use in relation to the risk of
breast cancer. JAMA. 2004; 291:827835.
126. Vivino FB, Al-Hashimi I, Khan Z, et al.
Pilocarpine tablets for the treatment of dry
mouth and dry eye symptoms in patients with
Sjogren syndrome: a randomized, placebocontrolled, xed-dose, multicenter trial. P9201 Study Group. Arch Intern Med 1999;
159:174181.
127. Fox RI, Konttinen Y, Fisher A: Use of
muscarinic agonists in the treatment of
Sjogrens syndrome. Clin Immunol 2001;
101:249263.
128. Petrone D, Condemi JJ, Fife R, et al:
A double-blind, randomized, placebo-
129.
130.
131.
132.
133.
134.
135.
136.
controlled study of cevimeline in Sjogrens

syndrome patients with xerostomia and
keratoconjunctivitis sicca. Arthritis Rheum
2002; 46:748754.
Miljanovic B, Trivedi KA, Dana MR, et al:
82:887893.
Beetham WP: Filamentary keratitis. Trans
Tuberville AW, Frederick WR, Wood TO:
Punctal occlusion in tear deciency
syndromes. Ophthalmology 1982;
89:11701172.
Willis RM, Folberg R, Krachmer JH, et al:
The treatment of aqueous-decient dry eye
with removable punctal plugs: a clinical
and impression-cytologic study.
Dohlman CH: Punctal occlusion in
keratoconjunctivitis sicca. Ophthalmology
1978; 85:12771281.
Gilbard JP, Rossi SR, Azar DT, et al: Effect
of punctal occlusion by Freeman silicone
plug insertion on tear osmolarity in dry eye
disorders. CLAO J 1989; 15:216218.
Gilbard JP: Tear lm osmolarity and
11:243250.
Balaram M, Schaumberg DA, Dana MR:
Efcacy and tolerability outcomes after
punctal occlusion with silicone plugs in dry
eye syndrome. Am J Ophthalmol 2001;
131:3036.
CHAPTER
58
Tumors of the Cornea and Conjunctiva

Mona Harissi-Dagher and Kathryn Colby
INTRODUCTION
Tumors of the cornea and conjunctiva often present the ophthalmologist with a difficult diagnostic and therapeutic
challenge. If one considers the close topographic relationship
between the cornea and the conjunctiva and their shared
ectodermal origin, it becomes apparent that it is difficult to
discuss tumors in one area in isolation from the other. Tumors
of the ocular surface are classified according to the type of cell
from which they originate, including epithelium, melanocytes,
lymphocytes, vascular endothelium, and mesenchymal cells.
Despite their proximity, the conjunctiva and cornea have
important differences in anatomy and, therefore, in the cellular
components that cause tumors. The cornea is composed of
nonkeratinizing squamous epithelium that covers a nonvascularized collagenous layer, whereas the conjunctiva has a
squamous cuboidal cell layer that covers a rich connective
tissue composed of delicate substantia propria with abundant
blood vessels, lymphatic channels, and terminal nerve endings.1
The corneal limbus is a transition zone from which the corneal
epithelial stem cells originate. It is not surprising given the
amount of normal physiologic cellular activity that occurs in
this region that this is the preferred site for tumors, especially
those of epithelial origin. The lamellar arrangement of the
corneal stroma and the condensation of the outer layer into
Bowmans layer protect the cornea from deep invasion by many
tumors that arise in the conjunctiva.
When evaluating a patient with a suspicious lesion of the
ocular surface, the clinician should obtain a careful history;
specifically the duration of the lesion, its growth characteristics,
and the presence of systemic diseases such as HIV and xeroderma
pigmentosum should be determined. History of sun exposure
and climate of origin can be relevant as can a prior history of
skin cancers. The physical examination should include evaluation of the preauricular, submandibular, and cervical nodes and
a complete eye examination with meticulous evaluation of the
entire conjunctival (bulbar and palpebral) surface. Serial photographs are helpful in documenting stability or progression of a
lesion and specialized studies such as ultrasound biomicroscopy
may be useful in certain cases.
In this chapter, special emphasis is placed on neoplasms of
the cornea and conjunctiva, but for completeness, common
congenital, inflammatory, and degenerative lesions that simulate
neoplasms are discussed. Attention is drawn to the ability
of tumors in this area to mimic more benign conditions, thereby leading the clinician into a false sense of security. Tumors
originating in adjacent tissues such as the lacrimal gland and
orbit and from distant metastases may also present initially in
the cornea and conjunctiva requiring proper diagnosis and
management.
CONGENITAL LESIONS
EPIBULBAR CHORISTOMA
A choristoma is a congenital tumor-like growth that contains
displaced epithelial cells and other dermis-like elements not
normally indigenous to the site in which they are found. Although
present at birth, a choristoma in the eyelid or orbit may not be
apparent until later in life. Four types are generally recognized
as the prototypic choristomas2: simple choristomas (Fig. 58.1),
dermoids, dermolipomas, and complex choristomas. The most
common episcleral choristoma is a dermoid. Dermoids and
dermolipomas may coexist with other systemic malformations,
which include Goldenhars syndrome (oculoauriculovertebral
dysplasia), mandibulofacial dysostosis (Treacher Collins syndrome,
Franceschettis syndrome), and band-like cutaneous nevus and
central nervous system dysfunction (Solomons syndrome, linear
nevus sebaceous of Jadassohn).
DERMOID
Dermoids of the conjunctiva are usually well-circumscribed,
solid benign tumors. They are smooth, elevated, porcelain white,
round to oval lesions. They arise most commonly inferotemporally at the limbus3; however, isolated involvement of the
cornea is recognized.1 Smaller dermoids may be tan-colored and
somewhat inconspicuous, whereas larger lesions (greater than
10 mm) tend to be whiter and can protrude from the palpebral
aperture (Fig. 58.1).1 Most lesions are asymptomatic, yet secondary
complications include irritation, disturbance of the precorneal
tear film, and inadequate lid closure. Corneal astigmatism
caused by dermoids can lead to amblyopia. Histopathologically,
they occupy the superficial outer third of the sclera and are composed of ectodermal (keratinized epithelium; hairs; sebaceous
and sudoriferous glands; nerves; smooth muscle; and less frequently teeth) and mesodermal (fibrous tissue, fat, blood vessels,
and cartilage) elements combined in different proportions. The
presence of these dermal elements and the association with lid
colobomas in some cases have led to the theory that dermoids
or dermolipomas may result from faulty development of the lid
folds with entrapment of dermal elements within the sclera.1
Indications for surgical intervention include amblyopia,
secondary astigmatism, increased growth, and encroachment
onto the visual axis, dellen formation, inadequate lid closure,
and cosmesis.2 The removal of dermoids can be associated with
significant complications such as reduced ocular motility, astigmatism, and globe penetration.1 Ultrasound biomicroscopy may
help in the preoperative evaluation of limbal dermoids. If the
decision is made to excise the lesion, careful dissection in a
lamellar fashion or a corneoscleral lamellar keratoplasty is
789

FIGURE 58.1. Dermoid of the conjunctiva
arising from the inferior fornix and protruding
through the palpebral fissure. These lesions
typically occur on the temporal bulbar
conjunctiva.
performed depending on tumors size and depth.3,4 For isolated

corneal dermoids, the physician should be prepared to do a
penetrating keratoplasty at the time of surgery.5 Early removal
is advocated, to avoid rapid enlargement of the tumor and to
improve the postoperative visual outcome.3 Preoperative astigmatism persists postoperatively after limbal dermoid surgery
with little change in its orientation or amount regardless of
patients age. A relaxing incision or other refractive surgical techniques may be considered.
DERMOLIPOMA
SECTION 6
Dermolipomas are lesions similar to dermoids that tend to have

more yellowish color and are generally located in the superotemporal fornix, near the lacrimal gland and lateral rectus
(Fig. 58.2). They do not usually extend posterior to the orbital
rim but may be contiguous with orbital fat. Occasionally,
dermolipomas may distort the lateral canthal region. They may
be bilateral or multiple, and they are firmly adherent to the
underlying conjunctiva.6 The differential diagnosis includes
prolapsed orbital fat, the palpebral lobe of the lacrimal gland,
lymphangioma, and lymphoma. Histopathologically, the lesions
are similar to dermoids, except for the predominance of fat.
Ultrasonography should be performed preoperatively to investigate the degree of posterior scleral involvement. Generally, no
treatment is required, but if excision is necessary, caution
should be exerted not to damage the lacrimal gland or rectus

muscles. Subtotal excision of the mass may produce acceptable
results,6 although remnants of the excised dermoid may cause
considerable postoperative inflammation.
COMPLEX CHORISTOMA
Variable combinations of ectopic tissues in various configurations
characterize complex choristomas. Usually on the superotemporal globe, they may contain cartilage, bone, lacrimal gland
lobules, hair follicles, hair, sebaceous glands, and adipose tissue.
Slow growth may be exhibited, particularly during puberty, but
the potential for malignant degeneration is minimal. Excision
should be used with caution especially because the glandular
elements may extend deep into the cornea and sclera. Clinically,
these lesions may appear similar to dermoids and dermolipomas.
However, when acinar elements predominate, they can assume
a fleshier, vascularized appearance with raised translucent
nodules; this is referred to as ectopic lacrimal gland.7 An epibulbar
lacrimal choristoma is richly vascularized and may become
inflamed.
EPIBULBAR OSSEOUS CHORISTOMA

Osseous choristomas are solitary nodules of bone surrounded
by fibrous tissue that resemble conjunctival dermoids. They
tend to be more discrete and to have sharper edges than the
lesions previously described (Fig. 58.3). They are also located
superotemporally. The cornea is typically spared, and the lesions
are usually located 510 mm behind the limbus, which is a useful feature differentiating these from other conjunctival lesions.
The size and shape may vary, and adherence to the bulbar
conjunctival may occur. Most are composed of mature, compact
bone surrounded by additional choristomatous elements. These
are stationary lesions, and surgery may be indicated to improve
cosmesis or for histologic diagnosis.
DEGENERATIVE LESIONS
PINGUECULAE AND PTERYGIA
790
FIGURE 58.2. Dermolipoma of the conjunctiva. The lesion tends to

arise from the supertemporal bulbar conjunctiva. The yellowish
appearance is secondary to increased sebaceous material within the
lesion, which distinguishes them histopathologically from dermoids.
Often, fine hairs protrude from their surface; however, they are usually
asymptomatic.
A pinguecula (Latin pingueculus, meaning fatty) is a yellowish

raised growth located nasally or temporally on the bulbar conjunctiva in the palpebral fissure. It is commonly thought to
be the precursor of a pterygium (Fig. 58.4). A pterygium is
indistinguishable from a pinguecula in its earliest stages. It is a
wedge-shaped fibrovascular growth of conjunctiva that extends
onto either side of the cornea. As the mound of tissue grows in
size and moves toward the cornea, it acquires its characteristic
appearance, and the involved bulbar conjunctiva becomes
FIGURE 58.3. Epibulbar osseous choristoma in a 1-year-old girl. This

lesion, owing to its extraordinary size, required surgical excision;
however, smaller lesions may be asymptomatic and managed
conservatively.
muscle movement. In one study correlating pterygium size and

induced corneal astigmatism,12 it was found that once pterygia
reach a critical size (extension to >45% of the corneal radius),
they induce visually significant asymmetric with-the-rule astigmatic changes. These changes may be detected only by corneal
topography and not by subjective refraction.
A small pterygium with mild symptoms of photophobia and
redness can often be managed with the use of topical preservative-free lubricants, vasoconstrictors, and an intermittent
mild steroid. Ultraviolet blocking spectacles may help prevent
progression.9,13 The surgical management of pterygia has been
complicated in the past by high incidence of recurrence, especially
using a bare scleral approach.1416 In cases of recurrence, 97%
are within 12 months of pterygium removal.1719 Addition of a
conjunctival autograft has reduced the recurrence rate to ~5%.18
Amniotic membrane transplantation may be used as an alternative to conjunctival autografts and medical tissue adhesives
have recently been used in the management of pterygia.20,21
Adjunctive therapies including radiotherapy22,23 and chemotherapy agents such as mitomycin C have been used but carry
long-term risks of scleral melting and possible limbal stem cell
damage.2427
A pterygium must be distinguished from other lesions that
can mimic its appearance, most notably pseudopterygium, a fibrovascular scar arising in the bulbar conjunctiva and extending
into the cornea. Unlike a true pterygium, a pseudopterygium is
a result of previous external ocular inflammation, such as chemical
burns, trauma, or infection. It can also occur following extensive
cryotherapy. The absence of organization into recognizable parts,
the tendency of occurring outside the interpalpebral space, and
the lack of adhesion to the limbus clearly differentiates these
lesions from true pterygia.
EPITHELIAL CYST
Conjunctival inclusion cyst is a relatively common lesion that
may be congenital or acquired, the latter being more common.
The most frequent cause of a large acquired congenital cyst is
the implantation of conjunctival epithelium after surgery,
trauma, or conjunctival inflammation (Fig. 58.5).28 Many small
cysts are likely formed by apposition of conjunctival folds.
Usually conjunctival inclusion cysts fail to progress in size but
on occasion they may develop into enormous translucent cysts
or may cause intracorneal pseudohypopyon.29,30 Typically, these
increasingly taut. A pigmented iron line, known as Stockers

line, may be seen in advance of a pterygium on the cornea. Both
lesions are similar in pathogenesis. Their location in the interpalpebral zone and their occurrence in older individuals raised
in warmer climates suggest that pinguecula and pterygia are
degenerative lesions causally related to prolonged actinic exposure.8,9 However, the finding of p53 overexpression (the most
common marker of human neoplastic growth) in the epithelium
of both primary and recurrent pterygium specimens suggests
that pterygium is a growth disorder rather than a degeneration.10,11
The decision to remove a pterygium depends on several
factors, including the symptoms the patient is experiencing and
the desire for cosmetic improvement. These lesions may become
episodically inflamed, may grow to occlude the visual axis, may
induce astigmatism, or may cause restriction of extraocular
CHAPTER 58
FIGURE 58.4. Pterygium extending onto the cornea. These lesions

typically arise in the nasal bulbar conjunctiva and assume a triangular
configuration with the base toward the nose. Growth occurs from the
apex of the lesion toward the visual axis.
FIGURE 58.5. Typical conjunctival inclusion cyst. The lesion is usually

well circumscribed and contains translucent fluid.
791

cavities contain desquamated cellular debris and chronic inflammatory cells. The presence of a double layer of nonkeratinizing
cuboidal epithelium suggests a cyst of ductal origin. Most conjunctival cysts can be treated adequately by simple excision.
Drainage alone is insufficient because fluid may accumulate
again unless the cyst is excised. This should be considered if the
diagnosis is doubtful, or if the cyst interferes with lid closure,
threatens the integrity of the ocular surface, interferes with
vision, causes discomfort, or is a cosmetic concern.28
TUMORS OF EPITHELIAL ORGIN

BENIGN EPITHELIAL TUMORS
Conjunctival Papilloma
Squamous papilloma
Papilloma is the term applied to a finger-like exophytic projection
of a cutaneous or mucosal surface. In the conjunctiva, papillomas
are benign proliferations of conjunctival surface epithelium
composed of a fibrovascular core that is covered by an acanthotic
squamous epithelium. The base of attachment to the underlying substantia propria and tenons can be narrow or broad,
giving rise clinically to either a pedunculated or a sessile appearance, respectively (Fig. 58.6a). Pedunculated papillomas are
usually viral in origin, whereas sessile papillomatous lesions in
elderly patients may represent intraepithelial neoplasm. A
combination of the translucency of the conjunctiva, the presence
of goblet cells within the covering epithelium, and the central
vascular core accounts for its characteristic glistening surface.
Squamous papillomas may arise anywhere on the conjunctiva
and may be unilateral or bilateral in presentation. The presentation of multiple bilateral squamous papillomas in a young
individual has been associated with infection by the human
papillomavirus (HPV) (Fig. 58.6bd). HPV types 6, 11, 16, 18,
and 33 have been associated with benign conjunctival
papilloma.3134 A majority of these lesions behave in a benign
fashion and have little tendency to undergo malignant proliferation; however, dysplastic changes can occur. Signs of dysplastic
change include increased keratinization, inflammation of the
adjacent conjunctiva with symblepharon formation, and spread
to the tarsal conjunctiva. Pain, bleeding due to telangiectatic
vessels, and secondary infection are common. Although they
SECTION 6
792
FIGURE 58.6. (a) Sessile papilloma of the conjunctiva. These lesions are typically located 12 mm from the limbus, as opposed to the
juxtalimbal position of CIN. The geometrically arranged red dots on the surface of the lesion are pathognomonic for a papilloma and are
secondary to central vasculature surrounded by thickened epithelium. Keratinization of the surface may be seen in both nondysplastic and
dysplastic forms of this lesion. (b) An exuberant example of recurrence of conjunctival papillomas in a 4-year-old boy 1 month after excision of
the original lesion at its base. (c) A closer view of the lesion from (b) showing the geometrically arranged red dots present throughout the lesion,
corresponding to the central vascular core seen histopathologically. (d) The appearance of the child after repeat excision and carbon dioxide
laser treatment of the lesions.
(a) Courtesy of Dr A Kaufman and Dr Richard Darrell, New York. (cd) Courtesy of James R Patrinely, Houston, TX.

are initially asymptomatic, as they grow, papillomas are often
associated with low-grade chronic papillary conjunctivitis or
punctate epithelial keratitis.35
Given the premalignant potential of these lesions and the
possible infectious nature of the process, treatment is indicated
for all symptomatic and asymptomatic conjunctival lesions that
are growing rapidly or not remitting spontaneously on careful
follow-up.35 The management of these squamous papillomas is
difficult.31 Although topical chemotherapy, including interferon
alpha and mitomycin C, has been proposed,36 simple excision
with cryotherapy to the base and surrounding epithelium remains
the most effective treatment. When these lesions are treated, it
is important to excise the lesion with an adequate surrounding
area of normal conjunctiva at the base, because recurrences are
frequent, particularly in children.37 Recently, CO2 and dye lasers
have been used to debulk these lesions. Recurrence is not
uncommon.38 Submitting tissue for detection of HPV is still
being evaluated in the management of these lesions.39 Oral
cimetidine (Tagamet) may act as an immunomodulator and
thereby provide a systemic method of treatment.40
Inverted conjunctival papilloma

A variant of the conjunctival squamous cell papilloma is the
inverted conjunctival papilloma, also known as benign mucoepidermoid carcinoma of the conjunctiva. It typically demonstrates
an endophytic growth pattern of the conjunctival epithelium,
instead of the usual exophytic pattern. In contrast to inverted
papilloma of the nasal cavity, which tend to be locally destructive
and can undergo malignant transformation, inverted papillomas of
the conjunctiva have been more indolent in their clinical behavior.
Owing to the presence of numerous mucus-secreting cells in
specimens, Jakobiec and associates41 have recommended the term
benign mucoepidermoid of the conjunctiva to describe this entity.
HPV genomic material was confirmed in a reported case using
in situ hybridization.38 Treatment is complete excision;
recurrence after excision has not been reported.
Benign Hereditary Intraepithelial Dyskeratosis
Pseudoepitheliomatous Hyperplasia
Pseudoepitheliomatous hyperplasia is a benign reactive proliferation of the conjunctival or corneal epithelium.43 The lesion
develops rapidly over weeks or months, usually because of irritation
in a setting of preexisting stromal inflammation by lesions such
as pinguecula or pterygium. The typical clinical appearance is a
white elevated mass with a hyperkeratotic surface and sometimes
an area of central umbilication similar to that seen in keratoacanthomas (Fig. 58.8). Jakobiec43 has pointed out that the
clinical distinction between pseudoepitheliomatous hyperplasia
and squamous cell carcinoma is difficult when the former occurs
at the limbus; however, a history of rapid proliferation and a lack
of regularly arranged vascular fronds are useful in distinguishing
this lesion from squamous papilloma and squamous carcinoma.
This lesion does not degenerate to a carcinomatous condition,
and simple excision is usually curative.
Keratoacanthoma
Keratoacanthoma most frequently presents as a rapidly growing
benign lesion of the eyelid; however, it can arise in the bulbar
conjunctiva, the temporal limbus being affected most often.44
Its importance lies in the need to distinguish it, along with
pseudoepitheliomatous hyperplasia, from squamous cell
carcinoma.45,46 The rapidity with which this lesion grows in the
conjunctiva is a clue to the diagnosis; squamous cell carcinoma
usually develops in a more indolent fashion and lacks the rapid
onset of keratoacanthoma. Although keratoacanthomas have
the potential for spontaneous regression, excision is justified by
the risk of malignant transformation, and for cosmesis.46
CHAPTER 58
Benign hereditary intraepithelial dyskeratosis is a bilateral disorder inherited as an autosomal dominant trait with a high
degree of penetrance. It is a very rare condition that usually
appears during the first decade of life and affects consanguineous
kindreds of individuals known as Haliwa Indians, an interracial
mixture of black, white, and Native American people residing in

the counties of Halifax and Washington, North Carolina.42 Benign
hereditary intraepithelial dyskeratosis has been described in other
areas of the United States and in patients who are not of Haliwa
ancestry.
The affected individuals develop bilateral elevated hyperplastic lesions typically beginning at the limbus in a V-shaped
pattern.42 Foci of whiteness and dilated vessels are other
accompanying features (Fig. 58.7). The lesions usually do not
extend centrally toward the visual axis or threaten vision, but
corneal opacification and marked loss of vision can occur.
Associated leukoplakic lesions of the oropharynx and buccal
mucosa may be observed.
FIGURE 58.7. (a) Bilateral dilatation of the conjunctival blood vessels associated with benign hereditary intraepithelial dyskeratosis. (b) Closer
view of the right eye showing white perilimbal plaque with adjacent hyperemia of the conjunctiva.
From Shields CL, Shields JA, Eagle RA: Hereditary benign intraepithelial dyskeratosis. [Photo essay] Arch Ophthalmol 1987; 105:422-423. Copyright 1987, American
Medical Association.
793
a
FIGURE 58.8. Pseudoepitheliomatous hyperplasia (PEH) of the
conjunctiva. Characteristically, this lesion can be difficult to distinguish
clinically from CIN. The history of a rapid onset points toward the
diagnosis of PEH.
PREINVASIVE EPTHELIAL TUMORS
SECTION 6
Conjunctival and Corneal Intraepithelial Neoplasia
794
Conjunctival and corneal intraepithelial neoplasia are uncommon

tumors that have stimulated considerable debate about their
classification, natural history, and treatment.4751 Although
multiple terms, such as Bowens disease, conjunctival dysplasia,
intraepithelial epithelioma, and dyskeratosis, have been used in
the past to name these lesions, most clinicians now use the
term conjunctival intraepithelial neoplasia (CIN).48,51,52 CIN
refers to a neolapstic proliferation of a noninvasive dysplastic
squamous epithelium. In the older population, it is the third most
common ocular tumor after choroidal melanoma and intraocular lymphoma.52
The cause of CIN is uncertain and possibly multifactorial.
This usually unilateral tumor frequently occurs in fair-skinned
men in their mid-60s who have a history of extensive solar
exposure.43,51 Ultraviolet light-induced mutations to the p53
tumor suppressor gene may play a role. In addition to the
connection between UV-B and CIN, other identified risk factors
include heavy smoking, previous exposure to petroleum derivatives, xeroderma pigmentosum, and HPV.53,54 HPV has been
identified in both benign (types 6, 8, and 11) and malignant
(types 16 and 18) conjunctival epithelial growths.3134,5560 Because
of a possible association, human immunodeficiency virus (HIV)
testing is advised in patients younger than 50 years in whom
CIN is diagnosed.61,62
Clinically, the most common appearance is a nodular translucent or gelatinous thickening of the conjunctiva with a
variable degree of keratinization. Diffuse forms of CIN are rare
but have been described.63 They can masquerade as a unilateral
chronic conjunctivitis64 thereby delaying the diagnosis.52 Less
frequently, a papillomatous configuration may be seen. More
than 95% of these lesions arise at the limbus in the interpalpebral zone (Fig. 58.9). The adjacent cornea may be affected
as a result of spreading of frosted epithelium anterior to Bowmans
layer beyond the main area of limbal thickening.48 Often this
sheet of tissue has a characteristic fimbriated margin with isolated
clusters of gray spots. When the involvement is limited only to
the corneal epithelium and the extent of the involved area is
disproportionate to the size of the limbal mass, the lesion is
usually called primary corneal dysplasia.43,48
Histologically, CIN usually connotes a partial-thickness to
full-thickness intraepithelial neoplasia, whereas carcinoma in
c
FIGURE 58.9. (a) CIN. A typical gelatinous mass arises at the limbus.
(b) Lesion typifying the papillomatous variety of CIN. The juxtalimbal
location is more characteristic of CIN than of a benign sessile
papilloma, which tends to occur a few millimeters from the limbus (see
Fig. 58.7a). (c) Leukoplakic variety of CIN.
situ has full-thickness involvement. Both lesions, however, are

characterized by an intact basement membrane without invasion of the underlying substantia propria.
Clinical and laboratory evidence has convincingly showed
that these lesions, dysplasia and CIN, are part of a single disease.
Moreover, the neoplastic potential for the different degrees of
dysplasia and CIN has been shown to be similar.51 Mauriello
and co-workers65 stressed that all intraepithelial neoplastic
lesions be segregated as either actinic keratosis or dysplasia. In
doubtful cases, a focal lesion with parakeratosis should be
MALIGNANT EPITHELIAL LESIONS

Squamous Cell Carcinoma
Squamous cell carcinoma occurs when neoplastic cells
penetrate the basement membrane and invade the conjunctival
stroma. Invasive squamous cell carcinoma is less frequent than
FIGURE 58.10. Invasive conjunctival squamous cell carcinoma

demonstrating a papillomatous configuration.
CIN. The incidence varies from 0.02 to 3.5 per 100 000.7778 CIN
is often the precursor of invasive squamous cell carcinoma, which
is the most common malignant tumor of the conjunctiva.
Clinically, its presentation is similar to that of CIN, with lesions
occurring most frequently at the limbus and appearing either
gelatinous, leukoplakic, or papilliform (Fig. 58.10). In contrast to
CIN, invasive squamous cell carcinoma can involve a greater
proportion of the limbal area and be slightly larger.51 Often an
engorged conjunctival blood vessel can be seen feeding the tumor
(Fig. 58.10). Sometimes features of microinvasion such as
adherence to underlying tissues may be present; however, this is
not always the case.51 The clinical diagnosis can be enhanced by
exfoliative or impression cytologic examination. The latter has a
sensitivity of 77% for diagnosing squamous cell carcinoma.77
Histopathologically, dysplastic epithelial cells penetrate the
underlying basement membrane, allowing free access for extension throughout the subconjunctival space. Primary or secondary
squamous cell carcinoma of the cornea usually proliferates in
the epithelium and, despite invasion through the epithelial
basement membrane, is usually inhibited by the compact architecture of Bowmans layer.79
The majority of these lesions are not locally aggressive, and
they can be successfully treated by local surgical excision combined with cryotherapy. Surgical excision of the suspicious area
using a no-touch technique is the standard approach to therapy,
especially if the entire lesion can be removed in its entirety. The
treatment of invasive squamous cell carcinoma should be more
aggressive than the treatment of CIN, using wider local excision
and deeper sclerectomy or keratectomy with triple freeze-thaw
cryotherapy.71 After removal of the lesion with 23 mm margins,
freezing of the remaining conjunctival margin and sublesional
base is accomplished with a nitrous oxide probe. Cryotherapy
destroys tumor cells by thermal disruption as well as resultant
local ischemia. This regimen has resulted in a reduction of the
recurrence rate from 40% to 10%.71
Therapy with antimetabolite agents has proven beneficial in
the adjunctive treatment of partially excised corneal epithelial
neoplasia, as well as initial therapy in recurrent disease. Other
possible indications are extensive disease with ill-defined borders,
or situations in which excessive conjunctiva would be removed
causing limbal stem cell deficiency. Both mitomycin C and
5-fluororacil have a selective effect on rapidly dividing tumor
cells. Adjunctive topical chemotherapy with mitomycin C
0.020.04% is tid or qid for 7- to 14-day courses with rest
periods in between treatment cycles because of dose-related
CHAPTER 58
classified as actinic keratosis, whereas a diffuse lesion without

parakeratosis should be classified as dysplasia. The importance
of distinguishing between these two clinicopathologic entities is
reflected by the fact that in one study, dysplasia lesions (62%)
recurred more often than actinic keratosis lesions (8.5%).65 The
degree of atypia did not appear to be associated with recurrence.
Because the squamous dysplasia arises from a single cell that
undergoes neoplastic transformation, these lesions are slowly
progressive.51,66 Most CIN lesions do not progress to invasive
squamous carcinoma.51 Although the histologic features of both
these lesions may be malignant, the potential for metastasis is
nil for the former and relatively rare for the latter.50 Intraocular
extension by a squamous cell carcinoma is very unusual and
difficult to diagnose but has recently been treated successfully
with proton beam therapy.66,67
Despite its low virulence, CIN has been difficult to cure. The
management of these lesions consists of a wide local excision
plus cryotherapy.43,66 During the excision, it is important to
incorporate a surrounding area of 2 mm of apparently uninvolved
conjunctival epithelium to secure adequate surgical margins
using a no touch technique during removal.43,68 This is the
most important factor in predicting recurrence.69 Rose bengal
may be used topically to delineate the extent of the abnormal
epithelium. The sclera may be left bare, although primary closure
with absorbable sutures may produce a better cosmetic result.
Larger defects may be filled with amniotic membrane. Some
authors recommend frozen section control to ensure this goal.66
Usually, the thickened portion of the lesion at the limbus is
excised first, followed by a corneal epithelial debridement or
superficial keratectomy in order to remove the frosted epithelium.
Eighteen percent alcohol can be used to facilitate removal of the
corneal epithelium. In general, it is best to avoid disturbing
Bowmans layer, which acts as a natural barrier to intracorneal
tumor extension. If the entire lesion is resected and both the cut
edges and the base of the resection are treated with adjunctive
cryotherapy, greater than 90% long-term tumor control is
achieved.7072 A double or triple freeze-thaw technique, typically
with a nitrous oxide probe, is used.66 Topical 5-fluorouracil and
mitomycin C drops have been used with success for primary
therapy or as adjuncts to surgery in CIN treatment.7375 CIN
may involve more than 50% of the cornea, and its excision may
deplete the reservoir of the limbal stem cell population. A
limbal autograft from the uninvolved eye provides an excellent
anatomic result, and a healthier ocular surface.76
The recurrence rates of both CIN and invasive squamous cell
carcinoma are mainly a function of the completeness of the
initial tumor resection. Although excised lesions with free surgical
margins have shown an ~5% recurrence rate, the recurrence
may be as high as 50% if the lesion is incompletely excised.51
Topical mitomycin C is useful for the treatment of incompletely
excised or recurrent lesions. In addition, topical chemotherapy
with 5-fluorouracil or interferon alpha 2b has also been reported
with fewer and less severe side effects.36 For the topical treatment of CIN, mitomycin C is most effective followed closely by
5-fluorouracil, and then interferon alpha 2b.36 The clinical
appearance, presence of invasion, degree of dysplasia, and cell
type do not affect the prognosis.48 The slow growth of the recurrent
lesions and the ever-present malignant potential warrant lifelong annual follow-up.69
795

local toxicity.36 5-Fluororacil 1% in an artificial tear base is used
tid to qid for 1-2 week cycles until sloughing of the epithelium
occurs. A rest period then follows to allow regeneration of the
epithelium, followed by additional courses of therapy if
neoplastic disease remains. Recombinant interferon-alpha 2b
has been used successfully in the treatment of corneal and CIN
with an initial injection of 3 million international units (IU)
followed by topical interferon-alpha 2b drops (1 million IU per
mL) qid. If clinical response is noted by 1 week, topical therapy
was continued until resolution of the CIN. If minimal response
is seen at 1 week, subconjuntival and perilesional injections
are performed three times weekly until clinical resolution.
Regression of biopsy-proven conjunctival CIN has been reported
after 6 weeks of topical therapy with Cidofovir 2.5 mg/mL, one
drop every 2 h initially with a weekly taper in frequency over the
next 6 weeks. It is sometimes possible to treat cases of focal
intraocular invasion with local resection combined with an
iridocyclectomy80; however, more often enucleation is the only
alternative. The presence of orbital invasion is an indication for
exenteration.
CIN and squamous cell carcinoma are considered low-grade
malignancies. Recurrence is influenced by the integrity of surgical
margins reinforcing the need for wide margins and histopathologic examination of all edges of the excised specimen.
The degree of histopathologic atypia and the presence of subepithelial neoplastic cells (squamous cell carcinoma) also influence
the recurrence rate. Intraocular invasion and metastasis are
uncommon.77 Conjunctival squamous cell carcinoma is a relatively low-grade malignancy but appears to be particularly
aggressive in HIV and xeroderma pigmentosum patients. Highfrequency ultrasound may help delineate the extent of scleral
and intraocular spread in suspected cases. Invasive conjunctival
squamous cell carcinoma is not usually associated with regional
or distant metastases.
SECTION 6
Mucoepidermoid Carcinoma
Mucoepidermoid and spindle cell carcinoma are rare variants of
squamous cell carcinoma that can arise in the conjunctiva. In
contrast to the relatively benign course of squamous cell carcinoma, these entities tend to be more locally aggressive and cause
greater complications. Clinically, it is not possible to distinguish
these two epithelial neoplasms from their more indolent counterpart, and the diagnosis is most often made retrospectively after
the recurrence of what was previously thought to be squamous
cell carcinoma. The histologic appearance of the early tumor
may be identical to that of squamous cell carcinoma because
the mucin-producing elements may not be expressed until there
is intraocular invasion.50 Histologic stains directed to mucin
should always be used on squamous lesions to rule out mucoepidermoid carcinoma.81
Mucoepidermoid carcinoma is the most common malignant
tumor of the major salivary glands. The conjunctival occurrence
is extremely unusual.81 In contrast to squamous cell conjunctiva,
which usually arises at the limbus, mucoepidermoid carcinoma,
can arise anywhere on the conjunctiva, and may invade the
underlying tissues early in the course. This disease is difficult to
control with almost universal recurrence.
The treatment of both of these rarer epithelial neoplasms is

a combination of wide local excision and cryotherapy.85 The
propensity for these conjunctival tumors to recur early necessitates
close follow-up. The presence of intraocular or orbital spread is
an indication for enucleation or exenteration, respectively. Distant
metastasis and death have been reported.86
GLANDULAR TUMORS OF THE CONJUNCTIVA

Sebaceous Cell Carcinoma
The ability of sebaceous cell carcinoma to show intraepithelial,
pagetoid spread to the conjunctiva and masquerade as less
ominous conditions such as chalazion, blepharoconjunctivitis,
conjunctivitis, and superior limbic keratitis underscores the
importance of including this entity in a discussion of conjunctival and corneal tumors (Fig. 58.11).87,88
Since this tumor usually begins in the meibomian glands it
will be covered in detail in the eyelid malignancy chapter.
Sebaceous cell carcinoma may also originate from the other
structures in the lids associated with sebaceous glands, such as
the glands of Zeis at the lid margin and pilosebaceous units in
the brow and caruncle.89 Some evidence suggests that sebaceous
cell carcinoma may arise de novo in the conjunctiva as well.90
The difficulty in making the diagnosis clinically is reflected
in a delay between presentation and diagnosis of 13 years.91-93
Clinically, the presence of an atypical unilateral conjunctivitis
that fails to resolve with conventional therapy behooves the
physician to consider the diagnosis of sebaceous cell carcinoma
and perform a biopsy. Map biopsies of both involved and
clinically normal conjunctiva are important in determining the
extent of intraepithelial spread of this lesion.94
The histopathologic diagnosis of sebaceous cell carcinoma
can be difficult, and the tumor is easily confused with squamous
cell and basal cell carcinomas.95,96 The diagnosis is based on the
presence of sebaceous elements within the tumor. The demonstration of these elements by lipid stains (oil red-O) on frozen
sections or ultrastructural studies can be helpful in differentiating
this tumor from basal and squamous cell carcinomas. The
treatment is primarily surgical, with wide local excision of the
Spindle Cell Carcinoma
796
Spindle cell carcinoma of the conjunctiva is a rare aggressive

epithelial neoplasm that may arise in the epibulbar conjunctiva,
limbus, or cornea. It tends to be very locally invasive and can
cause significant ocular morbidity.8284 Histopathologically, this
neoplasm can be confused with fibrosarcoma, spindle cell melanoma, leiomyoma, or rhabdomyosarcoma. The use of electron
microscopy and immunohistochemical markers can be helpful
in differentiating this tumor from other simulating lesions.43
FIGURE 58.11. Sebaceous cell carcinoma of the conjunctiva in a

65-year-old woman with a 10-year history of conjunctivitis. Note the
absence of lashes on the lower eyelid secondary to invasion by the
tumor.
From Margo CE, Lessner A, Stern GA: Intraepithelial sebaceous cell carcinoma
of the conjunctiva and the skin of the eyelid. Published courtesy of
FIGURE 58.13. Racial melanosis. Circumlimbal distribution of flat

golden-brown pigmentation typically fades toward the fornices.
FIGURE 58.12. Oncocytoma of the caruncle. This lesion typically
arises in the caruncle; however, it may occur in the lacrimal gland, the
conjunctiva, or the eyelid.
BENIGN PIGMENTED LESIONS

BENIGN ACQUIRED MELANOSIS (RACIAL
MELANOSIS)
Oncocytoma (Oxyphilic Adenoma)

Oncocytomas are rare tumors that frequently originate in the
caruncle, and are derived from degenerated ductal epithelial
cells. Other sites of ocular involvement are the lacrimal glands,
lacrimal ducts, accessory lacrimal glands of the conjunctiva,
and holocrine glands of the upper lid.97
They account for 38% of caruncular lesions.98 These tumors
arise typically in older patients, with a slight female predominance.
Clinically, they appear as small cystic masses and are characteristically yellow to red to tan (Fig. 58.12).97 These tumors are
generally benign, but isolated reports of malignant oncocytomas
exist in extracaruncular locations. There tends to be a greater
propensity toward malignancy and orbital invasion if these
tumors arise in the lacrimal gland. These lesions are usually
asymptomatic and are removed for cosmetic reasons. Wide local
excision is adequate, and recurrence is rare.
TUMORS OF NEUROECTODERMAL ORIGIN

Pigmented lesions of the conjunctiva may be divided into melanocytic or nonmelanocytic lesions. Pigmentation of melanocytic
tumors of the conjunctiva is due either local melanocytes
proliferation (melanocytosis) or to an increased production of
pigment granules by the resident melanocytes. The pathogenic
mechanism that initiates either of these processes is unclear;
however, exposure to sunlight has been implicated as a triggering factor.
Melanocytes are cells of neural crest origin that migrate to
mucus membranes, including the conjunctiva, and to the skin
during embryogenesis. They reside in the basal layers of the
conjunctival epithelium near the basement membrane.
Melanosomes within melanocytes synthesize and store melanin
and are therefore responsible for mucus membrane and skin
pigmentation. Melanin may be released from melanocytes and
taken up by epithelial cells. Skin and mucous pigmentation in
individuals with darker complexions is usually due to increased
synthesis and release of melanin, rather than to an increase in
melanocyte proliferation.99
Benign acquired melanosis is a common condition seen primarily in middle-aged darkly pigmented individuals (Fig. 58.13).
It is characterized by flat, dusty, light brown perilimbal and
interpalpebral conjunctival pigmentation that generally fades
toward the fornices, where there is less exposure to ultraviolet
radiation. It is a bilateral condition; however, involvement may
be asymmetric. Because of their intraepithelial location, these
pigmented lesions are freely mobile over the globe. The pigmentation is due to increased metabolic activity of the resident
melanocytes with an increase deposition of melanin granules in
the basal layer of the conjunctival epithelium.99 There is virtually
no potential for malignant degeneration. In the skin the equivalent
condition is called an ephelis (freckle). This condition requires
no treatment other than periodic observation.1
OCULAR MELANOCYTOSIS
Congenital melanosis of the deep conjunctiva, episclera, or
superficial sclera is more common in the black, Hispanic, and
Asian populations. These pigmented lesions appear bluish or
slate-gray and are usually unilateral. They are deep and immobile.
Ocular melanosis consists of focal proliferation of subepithelial melanocytes. The melanocytosis may also affect the
uvea, meninges, and orbital soft tissue. Fifty percent of patients
with ocular melanocytosis have ipsilateral dermal melanocytosis
(nevus of Ota), a proliferation of dermal melanocytes in the
periocular skin of the first and second dermatomes of the trigeminal nerve. The combination of ocular and cutaneous
pigmentation is referred to as oculodermal melanocytosis.
Glaucoma secondary to hyperpigmention of the trabecular
meshwork occurs in the affected eye in 10% of patients. Afflicted
white patients have an increased risk of developing malignant
melanoma in the skin, conjunctiva, uvea, or orbit. The lifetime
risk of uveal melanoma in a patient with ocular melanocytosis
is ~1 in 400, much greater than the risk of 1 in 13 000 of the
general population.
CHAPTER 58
lesion until the surgical margins are histopathologically clear

of tumor.
CONJUNCTIVAL NEVI
Benign melanocytic nevi are the most common pigmented
lesions of the conjunctiva. Thirty percent of these nevi may be
797
PREINVASIVE EPTHELIAL TUMORS

PRIMARY ACQUIRED MELANOSIS
SECTION 6
FIGURE 58.14. Benign nonpigmented conjunctival nevus. The

presence of an epithelial cyst within the lesion is suggestive of a
benign conjunctival nevus.
798
lightly pigmented, or even nonpigmented. Pigmentation may

increase during puberty, and previously nonpigmented lesions
can become pigmented. Conjunctival nevi are usually solitary,
well circumscribed, and freely mobile. A nevus near the limbus
is usually almost flat. Those appearing elsewhere tend to be
elevated. The most frequent locations include the bulbar conjunctiva, plica, caruncle, and the lid margin (Fig. 58.14). The
presence of cystic spaces within these lesions is diagnostic and
suggests a benign growth pattern; it is particularly important
to recognize this feature in the nonpigmented variety .100,101
The normal basal layer of the conjunctival epithelium is
made up of 5% melanocytes.102 Nevi are formed by an abnormal
benign proliferation of melanocytes with retraction of their
dendritic processes, giving them a more rounded appearance.100,102
This lesion usually makes its appearance in the first two
decades of life, which is an important historical distinguishing
feature since primary acquired melanosis (PAM) and conjunctival melanoma are rarely reported in the young. The natural
history of this lesion is for the abnormal melanocytes to
proliferate in the basal layer of the epithelium and form clusters
known as junctional nests. Junctional nevi may occasionally be
difficult to differentiate from PAM due to histopatholgic similarities but the age of presentation helps in making the diagnosis. In the second to the third decade of life, the cells sprinkle
down into the underlying substantia propria to form a compound nevus. Eventually the epithelial and junctional component
regress completely, and by the third to fourth decade, nests of
nevi cells are left in the substantia propria, resulting in a subepithelial nevus.100,103
Nevi are benign and require only periodic follow-up. Biopsy
should be performed and cryotherapy applied to the cut edges if
inclusion cysts are not found and the lesion is suspicious. The
presence of nevi on the palpebral conjunctiva or the forniceal
conjunctiva is rare and should alert the clinician to the possibility of conjunctival melanoma. It is estimated that ~20% of
conjunctival melanomas arise from preexisting nevi.104 Gerner
and colleagues105 suggested that nevi should be treated according to the following rules: (1) nevi of the fornix and tarsal conjunctiva should be excised; (2) bulbar conjunctival nevi (including
limbal and caruncular nevi) should be excised if showing
significant growth, neovascularization or nutrient vessels, inflammation, or increased or changed pigmentation; (3) excision should
be performed in all cases of recurrences; and (4) all excised tissue
should be histopathologically examined.
This acquired pigmentation of the conjunctival epithelium is

analogous to lentigo maligna of the skin (Hutchinsons freckle),
a preinvasive intraepidermal lesion of sun-exposed skin. PAM
refers to a unilateral, flat, pigmented lesion of the conjunctival
epithelium with irregular margins. These lesions are freely
mobile and may involve any part of the conjunctiva; therefore
inspection of the entire conjunctival surface, including the
palpebral and forniceal areas, is essential.99
Intraepithelial conjunctival melanocytes proliferate in middleaged white individuals for unknown reason. Most idiopathic
types of acquired melanosis remain benign, but cases that show
cellular atypia can be a precursor of conjunctival melanoma.
Malignant transformation should be suspected when a lesion
shows nodularity, enlargement, or increased vascularity. Clinical
features cannot distinguish precancerous lesions; this can only
be done by biopsy to assess histological atypia. PAM without
atypia has an extremely low risk of progression to melanoma.
PAM with atypia will progress to invasive melanoma in ~46%
of patients. Pagetoid spread by epithelioid melanocytes and fullthickness replacement of the epithelium are the most important
predictors of subsequent invasive melanoma (7590% cases).
Complete excision with tumor-free margins is essential for
PAM with atypia. In the setting of diffuse PAM, excision of any
nodularity is crucial. Multiple map biopsies of the remaining
conjunctiva help in assessing the extent of the disease. Cryotherapy, radiotherapy, and mitomycin C are useful adjuncts.
Mitomycin C is particularly appealing in patients with diffuse
disease.106 Mitomycin C 0.020.04% is given qid for a total
of 14 weeks, in cycles.36 Because of the potential for malignant
transformation and the possibility of recurrence, patients with
PAM with atypia should undergo careful follow-up, ocular and
adnexal examination, and photographic documentation several
times per year.
MALIGNANT PIGMENTED LESIONS

MALIGNANT MELANOMA
Malignant melanoma of the conjunctiva is an uncommon
malignancy accounting for only 1% of all ocular malignancies
and only 25% of all ocular melanomas. It is much less frequent
than uveal and skin melanomas. The incidence of PAM with
atypia or with malignant melanoma of the conjunctiva was
estimated to be 0.05 per year per 100 000 inhabitants of the
Danish population.107 It occurs predominantly in whites and is
very rare in non-Caucasians.106 It usually develops in the early
50s with no clear sex predilection. The incidence of conjunctival melanoma is increasing for white men above the age
of 60 years. A similar increase in the incidence of skin
melanoma has been noted. This has been related to an increase
in sun exposure.99
This tumor may originate de novo, from preexisting nevi,
and from PAM with atypia. Approximately 75% of conjunctival
melanomas arise in a setting of PAM; 20% of the remaining
patients have a history or microscopic evidence of a benign
conjunctival nevus.108,109 Primary malignant melanoma of the
cornea is extremely rare; most are secondary to extension from
neighboring conjunctiva and usually involve the superficial
layers of the cornea anterior to the basement membrane.
The clinical presentation can be variable and is dependent on
the antecedent status of the conjunctiva. Melanomas that arise
without a preexisting conjunctival nevus tend to occur at the
limbus and are thought to have initially a short horizontal
FIGURE 58.15. Nodule of malignant melanoma (arrow) of the

conjunctiva arising in a patient with preexisting PAM.
Courtesy of Frederick A. Jakobiec, MD.
growth phase followed by a rapid vertical growth phase (Fig.

58.15). Melanoma arising in a preexisting nevus is often heralded
by growth of the lesion or by increased vascularity. These
lesions can be managed with wide local excision using a no
touch technique with adjuvant cryotherapy. In the case of PAM,
the onset of malignant degeneration is often indicated by the
development of nodular thickening in a previously flat area of
pigmentation. Other features of malignant degeneration include
increased vascularity, fixation of the conjunctiva to the underlying sclera, and hemorrhage.
In two studies, the 5-year survival rate after surgery and/or
radiotherapy was 84%, and the 10-year survival rate was 71%.107,110
The 5-year recurrence rate in both studies was 39%. The overall
tumor-related mortality rate for conjunctival melanoma ranges
between 8% and 25%. This increases to 40% if the tumor arose
from PAM with an intraepithelial pagetoid growth pattern.103,104
Tumor thickness, mixed cell type, and lymphatic invasion on
histologic examination were found to increase the death rate.110
Other poor prognostic features include tumor in unfavorable
locations (caruncle, palpebral conjunctiva, or forniceal
conjunctiva); moderate to severe atypia; a paucity of small
polyhedral cells in the tumor; invasion of deeper ocular tissues;
greater than five mitotic figures per 10 high-power fields; and
lack of an inflammatory response induced by the tumor.100
Recurrence has been estimated to occur in 35% of patients, at
an average of 3.54.5 years after primary treatment.111
Metastatic spread occurs in 16 % of patients at 5 years, 26% at
10 years and 32% at 15 years, more commonly following
recurrent melanoma.111 The most frequent site of metastasis
was the lung, followed by the liver, brain, and bone. It is
important to palpate the regional lymph nodes, because spread
to the ipsilateral preauricular, intraparotid, submandibular, and
cervical nodes is well recognized.100 Once metastasis has
occurred, the survival rate declines markedly. Sentinel lymph
node mapping and selective lymphadenectomy may help detect
early metastasis.99 PET scan has an unclear role in screening for
metastasis.
In the past, exenteration was commonly performed for conjunctival melanoma but current therapy is less drastic. However, the surgical technique used to manage conjunctival
melanoma is critical, as incomplete tumor removal may lead to
spread through local lymphatic channels and increases the risk
of recurrence.68,111 The importance of appropriately aggressive
surgical management cannot be overstated. Incisional biopsies
of areas suspicious for malignancy (i.e., areas of pigmentation
that are elevated or those with increased vascularity) should be
avoided as they can lead to seeding of tumor cells throughout

the rest of the ocular surface. Instead, complete excision of any
nodular or vascularized conjunctiva, with a superficial lamellar
dissection of the underlying sclera as needed for adherent
tumors, using the no-touch technique is essential.68 Extension
of the tumor onto the corneal epithelium can be managed using
alcohol-assisted epithelial removal, so that nodular areas of the
tumor are removed en bloc. One should avoid breaching Bowmans
membrane, as this structure represents a natural barrier against
tumor extension into the corneal stroma.
Ideally, one should aim for complete excision of all conjunctival pigment. However, there are cases in which the ocular
surface pigmentation is too extensive to allow complete
removal.106 In the setting of diffuse disease, complete excision
of the most suspicious regions (i.e., nodular or heavily vascularized
areas) is complemented by map biopsies of remaining areas of
flat pigment to determine the extent of underlying PAM with
atypia.
Following tumor excision and map biopsies, double freezethaw
cryotherapy is applied to the cut edges of conjunctiva and to the
scleral base in areas suspicious for malignancy. In the setting of
diffuse disease, it may not be advisable to apply cryotherapy to
all pigmented areas of the bulbar conjunctiva. Instead,
cryotherapy should be applied to any nodular or vascularized
areas (i.e., those that are most suspicious for malignancy) during
the initial surgery. One can then await the histopathology determination of cellular atypia before proceeding with adjuvant
treatment of the areas of flat pigmentation, either additional
cryotherapy if the atypia is localized to a few spots only or
topical chemotherapy in the setting of diffuse atypia. We do apply
cryotherapy at the time of initial surgery to flat areas of pigment
in the upper palpebral conjunctiva or in the superior fornix, as
these areas will be less accessible to topical chemotherapy than
flat pigment on the bulbar conjunctiva or in the lower palpebral
conjunctiva.
Adjunct topical chemotherapy is useful in the management
of ocular surface neoplasia.36 Mitomycin C is the agent most
commonly used. The advantage of topical chemotherapy is that
it treats the entire ocular surface. Cryotherapy, on the other
hand, only treats the areas to which it is applied. Mitomycin C
is generally well tolerated when used on an intact ocular surface.
We wait at least 46 weeks following the initial surgery before
prescribing topical mitomycin C to allow adequate time for
complete healing of the ocular surface. Punctal plugs are inserted
into the upper and lower punctum to reduce systemic absorption of the topical chemotherapy. Female patients of childbearing age should be cautioned to use appropriate measures to
prevent pregnancy while using mitomycin C.
We typically give 0.04% mitomycin C four times daily for 1
week followed by a 3-week holiday. We attempt to repeat this
cycle a total of three times. Side effects include toxic
conjunctivitis in virtually all patients.99 This can be managed
with lubrication and topical steroid-antibiotic ointment. Fewer
patients will experience a toxic keratitis. Long-term limbal stem
cell deficiency remains a concern. Less common side effect
include an increase in intraocular pressure or a mild anterior
chamber reaction. Patients should be cautioned to avoid contact
with their skin, as a contact dermatitis can occur if the medication is allowed to run on to the periocular skin. Scleral
melting, which has been reported following application of
mitomycin C to the bare sclera in other settings, is rare when
this medication is used as discussed above on an intact ocular
surface.
Map biopsies of the entire conjunctival surface are repeated
several months after the last course of mitomycin C to assess
for residual cellular atypia. Further cryotherapy can be applied if
atypia persists.
CHAPTER 58
799
TABLE 58.1 Massachusetts Eye and Ear Infirmary Approach to Ocular Surface Tumors
1. Complete ophthalmic exam including palpation of regional lymph nodes and inspection of entire conjunctival surface, including lid eversion
to inspect entire upper fornix.
2. Photos to document disease state.
3. Complete excision of any areas suspicious for malignancy (i.e., nodules or areas of increased vascularity) including lamellar scleral
dissection, if needed, using a no-touch technique with 2 mm margins; Curettage of involved corneal epithelium, assisted by 18% alcohol to
allow en bloc removal of lesions located at the limbus.
4. Map biopsies of remaining flat areas of pigment if pigment too extensive to allow complete removal.
5. Double freeze-thaw cryotherapy to areas suspicious for malignancy (elevated lesions or lesions with increased vascularity).
6. Pass off surgical instruments that have handled areas of potential malignancy.
7. Primary closure of conjunctiva using absorbable sutures. May consider amniotic membrane graft if defect too large to close primarily.
8. If focal PAM with atypia found within diffuse pigmentation: double freeze-thaw cryotherapy to areas of atypia. If diffuse atypia: topical
Mmitomycin C 0.04% qid for 1 week, followed by 3 week holiday. Repeat for a total of three cycles if possible.
9. Map biopsies 23 months after last course of Mmitomycin C; additional cryotherapy if atypia persists.
Patients with melanoma are followed closely (every 4 months)

indefinitely. They are counseled to return immediately should
anything unusual occur. Our approach to pigmented conjunctival tumors is summarized in Table 58.1.
At present, orbital exenteration is rarely performed, as this
procedure is not associated with increased patient survival.112
Exenteration is reserved for aggressive ocular surface or orbital
disease that cannot be controlled locally. It should not be performed in cases where metastases have already occurred unless
other indications are present (pain, cosmetic concerns).
differentiating features. In cases in which there is confusion about

the diagnosis, a biopsy should be performed. The name pyogenic
granuloma is a misnomer, as neither granulomatous inflammation nor suppuration is a feature of this condition. These
lesions can usually be treated adequately by simple excision
combined with cautery to the base of the lesion.43 Topical
steroid administration before excision may cause a marked
reduction in the size of a pyogenic granuloma and even a complete
regression of smaller lesions.114
KAPOSIS SARCOMA
VASCULAR AND MESENCHYMAL TUMORS
SECTION 6
MESENCHYMAL TUMORS
800
The loose connective tissue of the substantia propria contains a

wide variety of tissue elements including blood vessels, nerves,
and lymphatics that infrequently cause a number of subepithelial lesions. The most common subepithelial tumor in
one large series was lymphoma113; rarer lesions reported in this
location include neurofibroma, schwannoma, rhabdomyosarcoma,
myxoma, xanthomatous lesions, hemangiopericytoma, and
metastatic lesions.
PYOGENIC GRANULOMA
Pyogenic granuloma is an inflammatory vascular response of
tissue that has usually sustained a previous insult, typically
either inflammatory or traumatic. In the conjunctiva, it occurs
most frequently after strabismus surgery, but other events such
as inflammation (secondary to chalazia, microbial infection, or
pterygia), limbal surgery for pterygium or squamous cell
carcinoma, phthisis, and chemical burns can predispose to the
development of a pyogenic granuloma. However, there are
reports of its presenting in the conjunctiva without any preceding
incident.43 Reported cases of pyogenic granulomas involving
primarily the cornea are few.
Pyogenic granulomas are typically raised, red, smoothsurfaced lesions with a narrow base and usually develop over a
relatively short period. Their appearance can sometimes be
confused with Kaposis sarcoma of the conjunctiva; however,
the rate of onset and the clinical circumstances can be useful
Initially described as idiopathic multiple pigmented sarcoma of

the skin by Kaposi in 1872, this once-rare neoplasm became a
major cause of morbidity and mortality in immunosuppressed
populations, especially patients with acquired immunodeficiency
syndrome (AIDS).115 The current pathogenetic concept is that
an initial infection by herpes virus 8 transforms normal mesenchymal cells, sensitizing them to high levels of cytokines
present during HIV infection.116 Ocular adnexal Kaposis sarcoma
had become an increasingly common tumor occurring in ~20%
of AIDS patients and systemic sarcoma.117 Of those, 20% have
conjunctival involvement. Kaposis sarcoma presenting as an
initial manifestation of AIDS has been reported.118 The clinical
presentation is of a reddish or bluish, painless, vascular conjunctival lesion that may be diffuse or nodular (Fig. 58.16).119,120
With the emergence of the highly active antiretroviral therapy era,
it is apparent that the incidence of Kaposi sarcoma, in patients
with AIDS is declining, especially in regions of the world where
these regimens are routinely available.121
Treatment options include surgical excision, cryotherapy,
irradiation, chemotherapy, and immunotherapy.116 Dugel and
associates122 found a better rate of success for surgical excision
or cryotherapy, or both, for stage I and II tumors. Kaposis sarcoma
lesions usually respond rapidly to irradiation, but occasionally
excision is necessary to rule out other diagnoses or for patients
with functional or cosmetic difficulties.64,122 The possibility of
occult HIV disease should be entertained in a young person
with an atypical hordeolum or subconjunctival hemorrhage, as
Kaposis sarcoma sometimes mimics these common lesions
and may represent the initial presenting sign of AIDS. The
patient with ocular lesions must also be evaluated for lifethreatening visceral disease.

lymphomas. Biopsy is necessary to establish the diagnosis, and
a systemic evaluation is required in all affected patients to
exclude systemic involvement. Biopsy specimens require special
handling to complete histochemical and immunologic studies.
Reactive lymphoid hyperplasia is a benign lesion occurring in
patients over 40 years of age. Clinically, it presents as a minimally
elevated salmon colored tumor with a pebbly appearance corresponding to follicle formation. Lymphoid hyperplasia may resolve
spontaneously, but it can be treated with local excision, topical
corticosteroids, or radiation. Although lymphoid hyperplasia
presents with as a benign polyclonal lymphoid lesion, these
lesions have the potential to develop into systemic lymphomas;
therefore a long-term medical follow-up is required.123,124
Lymphoma
FIGURE 58.16. Kaposis sarcoma of the conjunctiva with invasion of
the adjacent cornea.
Courtesy of Frederick A. Jakobiec, MD, and Stephen Foster, MD, Massachusetts
Eye and Ear Infirmary, Boston.
LYMPHOID TUMORS
Lymphangiectasia
Lymphangiectasia appears on the eye as a group of irregularly
dilated lymphatic channels of the bulbar conjunctiva. It may be
a developmental anomaly or may follow trauma or inflammation. Anomalous communication with a venule can lead to
spontaneous blood-filled lymphatic vessels.
Lymphangioma
Lymphangiomas are proliferations of lymphatic channel
elements. This hamartoma is present at birth and may slowly
enlarge. Intralesional hemorrhage, producing a chocolate cyst,
may occur.
LYMPHOID HYPERPLASIA
CHAPTER 58
A number of benign and malignant lymphoid lesions can

involve the conjunctiva, many of which have overlapping
clinical and pathologic features. Lymphoid lesions of the conjunctiva usually present as a painless salmon-pink infiltration
of the inferior fornix in patients in their fifth or sixth decade of
life.123 Most conjunctival lesions are localized and not
associated with systemic disease, in contrast with preseptal skin
Conjunctival lymphomas are rare comprising ~1.5% of all conjunctival tumors.113 Half of all conjunctival lymphocytic
proliferations in adults are lymphomas. Most patients are over
50 years of age or are immunosuppressed. Patients often present
with a salmon pink mass of the conjunctiva that is freely
movable on the globe. Aside from lack of lymphoid follicles and
absence of vascularity, lymphoma has the same clinical appearance as lymphoid hyperplasia. Conjunctival lymphoid infiltrate
are usually unilateral at presentation. In a small percentage of
patients initially unilateral conjunctival disease may evolve into
bilateral involvement over time. The presentation is bilateral in
2031% of patients, and the majority of patients do not have
disseminated disease when they are initially diagnosed
(Fig. 58.17).123,124
Monoclonal proliferations are more worrisome for malignancy while polyclonal proliferations are more benign. The
majority of conjunctival lymphoid proliferations behave in a
benign fashion; however, lymphoid tumors of the conjunctiva
are associated with systemic lymphoma in 31% of patients.
Systemic lymphoma is found more often in those patients with
forniceal or midbulbar conjunctival involvement and in those
with multiple conjunctival tumors. Until more accurate
methods are developed to determine the future clinical behavior
of benign lymphoid lesions, periodic follow-up is advised in
these patients.124
The most common subtype of conjunctival lymphoma is the
extranodal marginal-zone B-cell lymphoma (EMZL), which was
termed mucosa-associated lymphoid tissue (MALT) in older
publications. Some lymphomas are limited to the conjunctiva
while others occur in conjunction with systemic malignant
lymphoma. A neoplastic lymphoid lesion of the conjunctiva is
FIGURE 58.17. Bilateral conjunctival lymphoma arising superonasally in the right eye (a) and on the inferior tarsal conjunctiva of the left eye (b).
Imaging failed to demonstrate orbital involvement, and systemic work-up was negative.
801

generally a monoclonal proliferation of B-cells. A diffuse sheet
of monotonous small round or cleaved lymphocytes is more
characteristic of a low-grade malignant lymphoma. High-grade
lymphomas are readily recognized as malignant by virtue of
their nuclear features and high mitotic rate.
Incisional biopsy is required in all patients to help in determining the malignant potential. The most accurate means of
making the diagnosis and predicting the eventual clinical outcome is through a study of the cytomorphologic features of the
lesion.124,125 Immunophenotypic analysis, either by flow cytometry
of fresh unfixed tissue or by immunoperoxidase staining, may
demonstrate B-cell monoclonality by revealing the light chain
predominance. More sophisticated molecular techniques may
show monoclonality by revealing immunoglobulin gene
rearrangements within tumor cells. Polymerase chain reaction
heteroduplex analysis established a diagnosis of conjunctival
B-cell lymphoma in the absence of supporting histology and
immunohistochemistry studies.126 However, although these
techniques are helpful they are not definitive.
The clinical, radiological, and histopathologic differentiation
of benign from malignant lymphoid proliferations is difficult,
and herein lays the dilemma in managing these patients.
Patients should be referred for a systemic evaluation including
physical examination with palpation of regional lymph nodes, a
chest roentgenogram, a complete blood count, serum protein
electrophoresis, computed tomography of the abdomen, bone
marrow biopsy, bone scan, liverspleen scan and PET scan to
exclude systemic involvement. An imaging study of the orbits
with thin cuts and with attention to the lacrimal gland is

helpful in confirming the localization of the tumor to the conjunctiva. Because development of systemic disease may occur
years after the diagnosis of primary conjunctival lymphomas,
repeat systemic evaluation is recommended at 6 months interval
for at least 5 years.123,124
Therapeutic options of conjunctival lymphoma include external
beam radiotherapy, brachytherapy, cryotherapy, intralesional
interferon injections, systemic rituximab, and observation.127 The
predominant treatment choice for all types of conjunctival
lymphoma is radiotherapy. The long-term complications are the
main concern of this type of treatment. These side effects include
keratoconjunctivitis sicca, cataract, and retinopathy. However, for
intermediate-grade and high-grade tumors, radiotherapy remains
the treatment of choice for local control. Brachytherapy has also
been shown to provide good control but it is also associated with
high risk of complications. Cryotherapy is another option for local
treatment of conjunctival lymphoma but its efficacy is not well
established. The use of interferon injections and systemic
rituximab are both being used as first-line treatments of
conjunctival lymphomas as well as for recurring tumors in patients
who cannot tolerate the side effects of radiotherapy. Patients with
systemic disease are candidates for systemic chemotherapy in
addition to any local treatment. Spontaneous regression of these
lesions has been described. Therefore if the lymphoma is localized
to the conjunctiva and is low-grade, a reasonable option may be
to defer treatment and observe for signs of recurrence or progression of the tumor.127
SECTION 6
REFERENCES
802
1. Spencer WH, Zimmerman LE: Conjunctiva.

In: Spencer WH, ed. Ophthalmic pathology:
an atlas and textbook. 3rd edn.
Philadelphia: WB Saunders; 1985:109228.
2. Mansour AM, Barber JC, Reinecke RD,
Wang FM: Ocular choristomas. Surv
Ophthalmol 1989; 33:339358.
3. Burillon C, Durand L: Solid dermoids of the
limbus and cornea. Ophthalmologica 1997;
211:367372.
4. Mader TH, Stulting D: Technique for the
removal of limbal dermoids. Cornea 1998;
17:6667.
5 Golubovic S, Latkovic Z, HorvaticObradovic M: Surgical treatment of large
corneal dermoid. Doc Ophthalmol 1995;
91:2532.
6. Fry CL, Leone CR Jr: Safe management of
dermolipomas. Arch Ophthalmol 1994;
112:11141116.
7. Roth DB, Shields JA, Shields CL, Eagle RC
Jr: Lacrimal gland choristoma of the
conjunctiva simulating a squamous cell
carcinoma. J Pediatr Ophthalmol
8 Karai I, Horiguchi S: Pterygium in welders.
9. Taylor HR, West SK, Rosenthal FS, et al:
Corneal changes associated with chronic
UV irradiation. Arch Ophthalmol 1989;
107:1481.
10. Tan DTH, Lim ASM, Goh H, Smith DR:
Abnormal expression of the p53 tumor
suppressor gene in the conjunctiva of
patients with pterygium. Am J Ophthalmol
1997; 123:404-405.
11. Austin P, Jakobiec FA, Iwamoto T:
Elastodysplasia and elastodystrophy as
the pathologic bases of ocular pterygia
and pinguecula. Ophthalmology 1983;
90:96.
12. Lin A, Stern G: Correlation between

pterygium size and induced corneal
astigmatism. Cornea 1998; 17:2830.
13. Char DH: Corneal tumors. In: Smolin G,
Thoft RA, eds. The cornea. 2nd edn.
14. Kamel S: Pterygium. Its nature and a new
line of treatment. Br J Ophthalmol 1946;
30:549.
15. Jaros PA, DeLuise VP: Pingueculae and
pterygia. Surv Ophthalmol 1988; 33:41.
16. Tan DTH, Chee SP, Dear KBG, Lim ASM:
Effect of pterygium morphology on
pterygium recurrence in a controlled trial
comparing conjunctival autografting with
bare sclera excision. Arch Ophthalmol
1997; 115:12351240.
17. Hirst LW, Sebban A, Chant D: Pterygium
recurrence time. Ophthalmology 1994;
101:755758.
18. Kenyon KR, Wagoner MD, Hettinger ME:
19. Vastine DW, Stewart WB, Schwab IR:
Reconstruction of the periocular mucous
membrane by autologous conjunctival
transplantation. Ophthalmology 1982;
89:1072.
20. Sippel KC, Ma JJ, Foster CS: Amniotic
membrane surgery. Curr Opin Ophthalmol
2001; 12:269281.
21. Chan SM, Boisjoly H: Advances in the use
of adhesives in ophthalmology. Curr Opin
Ophthalmol 2004; 15:305310.
22. Alaniz-Camino F: The use of postoperative
beta radiation in the treatment of pterygia.
23. Aswad MI, Baum J: Optimal time for
postoperative irradiation of pterygia.
24. Adams A, Starck T, Kenyon KR: The

management of pterygium. Ophthalmol Clin
North Am 1990; 3:611.
25. Chen PP, Ariyasu RG, Kaza V, et al: A
randomized trial comparing mitomycin C
and conjunctival autograft after excision of
primary pterygium. Am J Ophthalmol 1995;
120:151.
26. Manning CA, Kloess PM, Diaz MD, Yee
RW: Intraoperative mitomycin in primary
pterygium excision. Ophthalmology 1997;
104:844848.
27. Panda A, Das GK, Tuli SW, Kumar A:
Randomized trial of intraoperative
mitomycin C in surgery for pterygium. Am J
Ophthalmol 1998; 125:5963.
28. Gloor P, Horio B, Klassen M, Eagle RC Jr:
Conjunctival cyst. Arch Ophthalmol 1996;
114:1020.
29. Bloomfield SE, Jakobiec FA, Iwamoto T:
Traumatic intrastromal corneal cyst.
30. Williams BJ, Durcan FJ, Mamalis N,
Veiga J: Conjunctival epithelial
inclusion cyst. Arch Ophthalmol 1997;
115:816.
31. Lass JH, Grove AS, Papale JJ, et al:
Detection of human papillomavirus DNA
sequences in conjunctival papilloma. Am J
32. Pfister H, Fuchs PG, Vlcker HE: Human
papillomavirus DNA in conjunctival
papilloma. Graefes Arch Clin Exp
33. Naghashfar Z, McDonnell PJ, McDonnell
JM, et al: Genital tract papillomavirus type
6 in recurrent conjunctival papilloma. Arch
Ophthalmol 1986; 104:1814.
34. McDonnell PJ, McDonnell JM, Kessis T,
et al: Detection of human papillomavirus
type 6/11 DNA in conjunctival papillomas
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
by in situ hybridization with radioactive

probes. Hum Pathol 1987; 18:1115.
Miller DM, Brodell RT, Levine MR: The
conjunctival wart: report of a case and
review of treatment options. Ophthalmic
Surg 1994; 25:545548.
Poothulil AM, Colby KA: Topical medical
therapies for ocular surface tumors. Semin
Ophthalmol 2006; 21:161169.
Wilson FM II: Papilloma. In: Fraunfelder FT,
Roy FH, eds. Current ocular therapy 3.
Philadelphia: WB Saunders; 1990:286288.
Chang T, Chapman B, Heathcote JG:
Inverted mucoepidermoid papilloma of the
conjunctiva. Can J Ophthalmol 1993;
28:184186.
Nakamura Y, Mashima Y, Kameyama K,
et al: Detection of human papillomavirus
infection in squamous tumours of the
conjunctiva and lacrimal sac by
immunohistochemistry, in-situ hybridization
and polymerase chain reaction. Br J
Ophthalmol 1997; 81:308313.
Shields CL, Lally MR, Singh AD, et al: Oral
cimetidine (Tagamet) for recalcitrant, diffuse
conjunctival papillomatosis. Am J
Ophthalmol 1999; 128:362364.
Jakobiec FA, Harrison W, Aronian D:
Inverted mucoepidermoid papillomas of the
epibulbar conjunctiva. Ophthalmology
1987; 94:283.
Reed JW, Cashwell LF, Klintworth GK:
Corneal manifestations of hereditary benign
intraepithelial dyskeratosis. Arch
Jakobiec FA: Corneal tumors. In: Kaufman
HE, Barron BA, McDonald MB, et al, eds.
The cornea. New York: Churchill
Roth AM: Solitary keratoacanthoma of the
85:647.
Walter WL: Pseudo-epitheliomatous
hyperplasia of the conjunctiva; a lesion
simulating squamous-cell carcinoma. Am J
Schellini SA, Marques ME, Milanezi MF,
Bacchi CE: Conjunctival keratoacanthoma.
Cockerham GC, Cockerham KP: Epithelial
tumors involving the oular surface. In:
Foster CS, Azar DT, Dohlman CH, eds.
Smolin and Thofts the cornea. 4th edn.
Philadelphia: Williams & Wilkins;
2004:769780.
Waring GO III, Roth AM, Ekins MB: Clinical
and pathological descriptions of 17 cases
of corneal intraepithelial neoplasia. Am J
Dark AJ, Streeten BW: Preinvasive
carcinoma of the cornea and conjunctiva.
Ni C, Searl SS, Kriegstein HJ, Wu BF:
Epibulbar carcinoma. Int Ophthalmol Clin
1982; 22:1.
Erie JC, Campbell RJ, Liesegang TJ:
Conjunctival and corneal intraepithelial and
invasive neoplasia. Ophthalmology 1986;
93:176.
Lee GA, Hirst LW: Ocular surface
squamous neoplasia. Surv Ophthalmol
1995; 39:429450.
Napora C, Cohen EJ, Genvert GI, et al:
Factors associated with conjunctival
intraepithelial neoplasia: a case control
study. Ophthalmic Surg 1990; 21:27.
McDonnell JM, Mayr AJ, Martin WJ: DNA
of human papillomavirus type 16 in
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
dysplastic and malignant lesions of the

conjunctiva and cornea. N Engl J Med
1989; 320:1442. Erratum: N Engl J Med
1989; 321:63.
Lauer SA, Malter JS, Meier JR: Human
papillomavirus type 18 in conjunctival
intraepithelial neoplasia. Am J Ophthalmol
1990; 110:23.
McDonnell JM, Carpenter JD, Jacobs P,
et al: Conjunctival melanocytic lesions in
children. Ophthalmology 1989; 96:986.
Odrich MG, Jakobiec FA, Lancaster WD,
et al: A spectrum of bilateral squamous
conjunctival tumors associated with human
papillomavirus type 16. Ophthalmology
1991; 98:628.
McDonnell JM, Wagner D, Ng ST, et al:
Human papillomavirus type 16 DNA in
ocular and cervical swabs of women with
genital tract condylomata. Am J
Shibata DK, Arnheim N, Martin WJ: Detection
of human papilloma virus in parraffinembedded tissue using the polymerase
chain reaction. J Exp Med 1988; 167:225.
McDonnell JM, McDonnell PJ, Sun YY:
Human papillomavirus DNA in tissue and
ocular surface swabs of patients with
conjunctival epithelial neoplasia. Invest
Karp CL, Scott IU, Chang TS, Pflugfelder
SC: Conjunctival intraepithelial neoplasia: a
possible marker for human
immunodeficiency virus infection? Arch
Ophthalmol 1996; 114:257261.
Waddell KM, Lewallen S, Lucas SB, et al:
Carcinoma of the conjunctiva and HIV
infection in Uganda and Malawi. Br J
Ophthalmol 1996; 80:503508.
Khalil MK, Pierson RB, Mihalovits H, et al:
Intraepithelial neoplasia of the bulbar
conjunctiva clinically presenting as diffuse
papillomatosis. Can J Ophthalmol 1993;
28:287290.
Goldberg SH, Riedy DK, Lanzillo JH:
Invasive squamous cell carcinoma of the
palpebral conjunctiva. Graefes Arch Clin
Mauriello JA Jr, Napolitano J, McLean I:
Actinic keratosis and dysplasia of the
conjunctiva: a clinicopathological study of
45 cases. Can J Ophthalmol 1995;
30:312316.
Char DH: Clinical ocular oncology. New
York: Churchill Livingstone; 1989:6387.
Ramonas KM, Conway RM, Daftari IK,
et al: Successful treatment of intraocularly
invasive conjunctival squamous cell
carcinoma with proton beam therapy. Arch
Ophthalmol 2006; 124:126128.
Shields CL, Shields JA: Overview of tumors
of the conjunctiva and cornea. In: Foster
Thofts the cornea. 4th edn. Philadelphia:
Williams & Wilkins; 2004:737756.
Tabin G, Levin S, Snibson G, et al: Late
recurrences and the necessity for long-term
follow-up in corneal and conjunctival
intraepithelial neoplasia. Ophthalmology
1997; 104:485492.
Divine RD, Anderson RL: Nitrous oxide
cryotherapy for intraepithelioma of the
conjunctiva. Arch Ophthalmol 1983;
101:782.
Fraunfelder FT, Wingfield DL: Management
of intraepithelial conjunctival tumors and
squamous cell carcinomas. Am J
72. Peksayar G, Soyturk MK, Demiryont M:

Long-term results of cryotherapy on
malignant epithelial tumors of the
107:337.
73. Frucht-Pery J, Sugar J, Baum J, et al:
Mitomycin C treatment for conjunctivalcorneal intraepithelial neoplasia: a
multicenter experience. Ophthalmology
1997; 104:20852093.
74. Grossniklaus HE, Aaberg TM Sr: Mitomycin
C treatment of conjunctival intraepithelial
neoplasia. Am J Ophthalmol 1997;
124:381383.
75. Yeatts RP, Ford JG, Stanton CA, Reed JW:
Topical 5-fluorouracil in treating epithelial
neoplasia of the conjunctiva and cornea.
Ophthalmology 1995; 102:13381344.
76. Meallet MA, Espana EM, Grueterich M,
with conjunctival limbal autograft for total
limbal stem cell deficiency. Ophthalmology
2003; 110:15851592.
77. Yang J, Foster CS: Squamous cell
carcinoma of the conjunctiva. Int
Ophthalmol Clin 1997; 7384.
78. Sun EC, Fears TR, Goedert JJ:
Epidemiology of squamous cell
conjunctival cancer. Cancer Epidemiol
Biomarkers Prev 1997; 6:7377.
79. Cha SB, Shields CL, Shields JA, et al:
Massive precorneal extension of squamous
cell carcinoma of the conjunctiva. Cornea
1993; 12:537540.
80. Glasson WJ, Hirst LW, Axelsen RA, Moon
M: Invasive squamous cell carcinoma of
the conjunctiva. Arch Ophthalmol 1994;
112:13421345.
81. Carrau RL, Stillman E, Canaan RE:
Mucoepidermoid carcinoma of the
conjunctiva. Ophthal Plast Reconstr Surg
1994; 10:163168.
82. Cohen BH, Green WR, Iliff NT, et al:
Spindle cell carcinoma of the conjunctiva.
83. Ni C, Guo BK: Histological types of spindle
cell carcinoma of the cornea and the
conjunctiva. Chin Med J 1990; 103:915.
84. Slusker-Shternfeld I, Syed NA, Sires BA:
Invasive spindle cell carcinoma of the
conjunctiva. Arch Ophthalmol 1997;
115:288289.
85. Schubert HD, Farris RL, Green WR: Spindle
cell carcinoma of the conjunctiva. Graefes
233:5253.
86. Seregard S, Kock E: Squamous spindle cell
carcinoma of the conjunctiva; fatal
outcome of a pterygium-like lesion. Acta
87. Wagoner MD, Beyer CK, Gonder JR, Albert
DM: Common presentations of sebaceous
gland carcinoma of the eyelid. Ann
88. Jakobiec FA, Brownstein S, Albert W, et al:
The role of cryotherapy in the management
of conjunctival melanoma. Ophthalmology
1982; 89:502.
89. Ni C, Searl SS, Kuo PK, et al: Sebaceous
cell carcinomas of the ocular adnexa. Int
90. Margo CE, Lessner A, Stern GA:
Intraepithelial sebaceous cell carcinoma of
the conjunctiva and the skin of the eyelid.
91. Ginsberg J: Present status of meibomian
gland carcinoma. Arch Ophthalmol 1965;
73:271.
CHAPTER 58
803
SECTION 6
92. Straatsma BR: Meibomian gland tumors.

93. Sweebe EC, Cogan DG: Adenocarcinoma
of the meibomian gland; a pseudochalazion entity. Arch Ophthalmol 1959;
61:282.
94. Margo CE, Grossniklaus HE: Intraepithelial
sebaceous neoplasia without underlying
invasive carcinoma. Surv Ophthalmol 1995;
39:293301.
95. Doxanas MT, Green WR: Sebaceous gland
carcinoma; review of 40 cases. Arch
96. Yeatts R, Waller R: Sebaceous carcinoma
of the eyelid: Pitfalls in diagnosis.
Ophthalmic Plast Reconstr Surg 1985;
1:35.
97. Rodgers IR, Jakobiec FA, Krebs W, et al:
Papillary oncocytoma of the eyelid. A
previously undescribed tumor of apocrine
gland origin. Ophthalmology 1988;
95:1071.
98. Shields CL, Shields JA, White D, et al:
Types and frequency of lesions of the
caruncle. Am J Ophthalmol 1986; 102:771.
99. Nguyen ATQ, Colby KA: Pigmented tumors
of the cornea and conjunctiva. In: Foster
Thofts the cornea. 4th edn. Philadelphia:
Williams & Wilkins; 2004:769780.
100. Jakobiec FA, Folberg R, Iwamoto T:
Clinicopathologic characteristics of
premalignant and malignant melanocytic
lesions of the conjunctiva. Ophthalmology
1989; 96:147.
101. Tanaka M, Okisaka S, Nakayasu K, Kanai
A: Clinical importance of inclusion cysts of
nevi in the bulbar conjunctiva. Nippon
Ganka Gakkai Zasshi 1995; 99:190194.
102. Margo CE, Grossniklaus HE: Ocular
histopathology: a guide to differential
diagnosis. Philadelphia: WB Saunders;
1991:133139.
103. Folberg R, McLean IW: Primary acquired
melanosis and melanoma of the conjunctiva:
terminology, classification, and biologic
behavior. Hum Pathol 1986; 17:652.
104. Folberg R, McLean IW, Zimmerman LE:
Conjunctival melanosis and melanoma.
105. Gerner N, Norregaard JC, Jensen OA,
804
106.
107.
108.
109.
110.
111.
112.
113.
114.
115.
116.
Prause JU: Conjunctival naevi in Denmark

19601980; a 21-year follow-up study. Acta
Colby KA, Nagel DS: Conjunctival
melanoma arising from diffuse primary
acquired melanosis in a young black
woman. Cornea 2005; 24:352355.
Norregaard JC, Gerner N, Jensen OA,
Prause JU: Malignant melanoma of the
conjunctiva: occurrence and survival
following surgery and radiotherapy in a
Danish population. Graefes Arch Clin Exp
Ophthalmol 1996; 234:569572.
Folberg R, McLean IW, Zimmerman LE:
Malignant melanoma of the conjunctiva.
Hum Pathol 1985; 16:136.
Jakobiec FA, Rini FJ, Fraunfelder FT, et al:
Cryotherapy for conjunctival primary
acquired melanosis and malignant
melanoma; experience with 62 cases.
Paridaens ADA, Minassian DC, McCartney
ACE, Hungerford JL: Prognostic factors in
primary malignant melanoma of the
conjunctiva: a clinicopathological study of
256 cases. Br J Ophthalmol 1994;
78:252259.
Shields CL, Shields JA, Gunduz K, et al:
Conjunctival melanoma. Risk factors for
recurrence, exenteration, metastasis, and
death in 150 consecutive patients. Arch
Ophthalmol 2000; 118:14971507.
Paridaens ADA, McCartney ACE, Minassian
DC, Hungerford JL: Orbital exenteration in
95 cases of primary conjunctival malignant
melanoma. Br J Ophthalmol 1994;
78:520528.
Grossniklaus HE, Green WR, Luckenbach
M, et al: Conjunctival lesions in adults: a
clinical and histopathological review.
Cornea 1987; 6:78.
Cameron JA, Mahmood MA: Pyogenic
granulomas of the cornea. Ophthalmology
1995; 102:16811687.
Bedrick JJ, Savino PJ, Schatz NJ:
Conjunctival Kaposis sarcoma in a patient
with myasthenia gravis. Arch Ophthalmol
1981; 99:1607.
Brun SC, Jacobiec FA: Kaposis sarcoma of
the ocular adnexa. Int Ophthalmol Clin
1997; 37:2538.
117. Shuler JD, Holland GN, Miles SA, et al:

Kaposi sarcoma of the conjunctiva
and eyelids associated with the
acquired immunodeficiency
syndrome. Arch Ophthalmol 1989;
107:858.
118. Kurumety UR, Lustbader JM: Kaposis
sarcoma of the bulbar conjunctiva as an
initial clinical manifestation of acquired
immunodeficiency syndrome. Arch
119. Macher AM, Palestine A, Masur H, et al:
Multicentric Kaposis sarcoma of the
conjunctiva in a male homosexual with the
acquired immunodeficiency syndrome.
120. Weiter JJ, Jakobiec FA, Iwamoto T: The
clinical and morphologic characteristics of
Kaposis sarcomas of the conjunctiva. Am
121. Orem J, Otieno MW, Remick SC: AIDSassociated cancer in developing nations.
Curr Opin Oncol 2004; 16:468476.
122. Dugel PU, Gill PS, Frangieh GT, Rao NA:
Treatment of ocular adnexal Kaposis
sarcoma in acquired immune deficiency
syndrome. Ophthalmology 1992;
99:11271132.
123. Shields CL, Shields JA, Carvalho C, et al:
Conjunctival lymphoid tumors: clinical
analysis of 17 cases and relationship to
systemic lymphoma. Ophthalmology 2001;
108:979984.
124. McNally L, Jakobiec FA, Knowles DM II:
Clinical, morphological, immunophenotypic,
and molecular genetic analysis of bilateral
ocular adnexal lymphoid neoplasms in 17
patients. Am J Ophthalmol 1987; 103:555.
125. Ellis JH, Banks PM, Campbell RJ,
Liesegang TJ: Clinical correlation with the
working formulation classification and
immunoperoxidase staining of paraffin
sections. Ophthalmology 1985; 92:1311.
126. Strauss EC, Warren JF, Margolis TP,
Holsclaw DS: Diagnosis of conjunctival
B-cell lymphoma by polymerase chain
reaction heteroduplex analysis. Am J
Ophthalmol 2003; 136:207209.
127. Tsai PS, Colby KA: Treatment of
conjunctival lymphomas. Semin
Ophthalmol 2005; 20:239246.
CHAPTER
59
Lamellar Keratoplasty
John Goosey
HISTORICAL PERSPECTIVE
Key Features: Historical Perspective
The rst successful human corneal transplant was a lamellar

keratoplasty performed by von Hippel in 1888 using rabbit
donor tissue.
Lamellar keratoplasty remained the most popular form of
corneal transplant through the rst decade of the twentieth
century and was endorsed by notable luminaries including
Fuchs and Elshnig.
As penetrating keratoplasty results improved lamellar
keratoplasty became less popular and by the 1930s most
surgeons preferred penetrating keratoplasty over lamellar
keratoplasty. Today lamellar keratoplasty represents only 35%
of all corneal transplants.
In 1824, Reisinger reported a technique for replacing opacied

human corneal tissue with clear tissue from an animal. Using
an experimental rabbit model, he described a healing process in
which donor tissue appeared to mold together with recipient
tissue. This union of tissues suggested a plastic process and led
him to coin the term keratoplasty.1 Reisinger never perfected his
keratoplasty technique but his report inspired a wave of further
experiments. In 1840, Mhlbauer acting on ideas espoused by
von Walther developed an animal model for lamellar keratoplasty
using triangular-shaped grafts.2 After several decades of experimental failure, von Hippel reported the rst successful human
lamellar keratoplasty in 1888 using rabbit donor tissue.3 von
Hippel was one of the luminaries of nineteenth century keratoplasty. He designed a mechanical circular trephine that rotated
automatically by means of a spring in its head and produced a
vertical partial depth incision for lamellar keratoplasty. When
used during penetrating keratoplasty, von Hippels trephine
incised the cornea with minimal pressure on the eye, thereby
avoiding extrusion and damage to underlying ocular tissue. von
Hippels trephine would later be used by Zirm to perform the
rst successful penetrating keratoplasty on 7 December 1905.4
However, von Hippels early failure with penetrating keratoplasty led him to emphasize the superiority of lamellar grafting.
He also advocated the use of heteroplastic donor tissue after
experiencing an endophthalmitis using human donor tissue and
because of his influence xenotransplantation remained popular
through the rst decade of the twentieth century. Successful
lamellar grafting using von Hippels technique was reported by
others, including Fuchs, who reported 50 viable lamellar grafts.5
In 1908, Anton Elschnig stated that lamellar grafting was the
only technique that provided reliable visual outcomes.6 During
this time an important modication of von Hippels lamellar technique included the use of homoplastic grafts. Magitot reported
a lamellar homograft that remained clear for 1 year after surgery.7

He also developed a preservation technique for cadaveric cornea
using human serum that allowed tissue to be preserved several
days before use.8
In 1914, just 6 years after asserting that lamellar keratoplasty
represented the only reliable technique for corneal grafting, Anton
Elschnig reported his rst successful penetrating keratoplasty in
a patient with interstitial keratitis.9 When Elschnig reported his
series of 174 penetrating transplants in 1930, he was convinced
that penetrating keratoplasty was superior to lamellar keratoplasty even though only 22% of his penetrating grafts were
successful.10 Elschnigs clinic in Prague represented an epicenter
for experimentation in keratoplasty and his endorsement of
penetrating keratoplasty sparked renewed interest in the technique. Lamellar keratoplasty became less popular and by the late
1930s was rarely performed by corneal surgeons with the notable
exception of Louis Pauque. Although he was impressed when
he witnessed his rst penetrating keratoplasty during a visit in
1937 to Elschnigs clinic, Pauque maintained that lamellar
keratoplasty avoided the complications of secondary glaucoma
and cataract that occurred with penetrating keratoplasty. As a
result, Pauque developed and rened his technique for lamellar
keratoplasty between 1943 and 1947, culminating in the
landmark work Les Greffes de la Corne.11 In spite of Pauques
enthusiasm for lamellar keratoplasty the procedures popularity
continued to decline and by the late 1970s lamellar keratoplasty
represented only 38% of all corneal grafts.12
Today, penetrating keratoplasty remains the most popular
method for the visual rehabilitation of patients requiring transplantation. Most corneal surgeons avoid lamellar keratoplasty
because of its perceived surgical difculty and the widely held
belief that penetrating keratoplasty yields superior outcomes.
However, recent improved microsurgical techniques and instrumentation have led to increased interest in lamellar keratoplasty.
The growing popularity of refractive lamellar procedures such as
LASIK has allowed corneal surgeons to gain familiarity with
microkeratomes and experience the benets of lamellar surgery.
Lamellar keratoplasty involves the removal and replacement
of diseased or deformed anterior corneal tissue (epithelium,
Bowmans layer, and stroma) while maintaining the hosts
Descemets layer and endothelium. The advantages of lamellar
surgery have long been recognized. Since lamellar keratoplasty is
an extraocular procedure, complications such as endophthalmitis,
expulsive hemorrhage, glaucoma, and cataract formation are mitigated. However, the most important advantage is the host endothelium is spared, eliminating the possibility of endothelial
rejection, the leading cause of graft failure following penetrating
keratoplasty. Penetrating grafts demonstrate a pronounced
endothelial cell loss in the rst year and progressive endothelial
cell loss each following year.13,14 Sadly, progressive endothelial
805
SECTION 6
FIGURE 59.1. Cornea of a patient who underwent RK surgery followed by LASIK and then PRK surgery. (a) A sclerotic scatter slit-lamp
photograph depicting diffuse corneal scarring. Note the multiple RK incisions in the periphery and the diffuse central stromal haze that
developed after PRK surgery was performed over a LASIK flap. (b) The slit-lamp photograph of (a). The depth of the RK incisions and the
stromal haze are evident. (c) A sclerotic scatter slit-lamp photograph showing the surgical treatment of the corneal scar of (a) and (b) 2 months
after lamellar keratoplasty. (d) A slit-lamp photograph depicting the barely discernable interface between the lamellar graft and the recipient bed.
Arrows point to small particles within the interface.
cell loss will result in eventual failure of all penetrating grafts if

the patient lives long enough. Alternatively, lamellar keratoplasty
reduces endothelial cell density by ~10% after 1 year followed
by normal physiological cell loss.1517 Therefore, it is reasonable
to consider lamellar keratoplasty when eyes with healthy endothelium and anterior corneal pathology require transplantation.
The indications, technique, outcomes, and complications of
lamellar keratoplasty are covered in this chapter.
INDICATIONS
Key Features: Indications
806
Lamellar keratoplasty is indicated when corneas with healthy

endothelium and pathology existing anterior to Descemets
membrance require transplantation.
Lamellar keratoplasty avoids the complications of endothelial
graft rejection and delayed endothelial cell loss seen with
penetrating keratoplasty.
Lamellar keratoplaty is an alternative to penetrating
keratoplasty for the surgical treatment of keratoconus, corneal
complications following refrective surgery and corneal scars
after infectious keratitis.
Lamellar keratoplasty is indicated when a cornea with healthy

endothelium and pathology existing anterior to Descemets membrane requires transplantation. Lamellar keratoplasty can be
performed for optical or tectonic reasons. One of the most
common indications for optical lamellar keratoplasty is contactlens-intolerant keratoconus. Iatrogenic keratoectasia following
LASIK surgery is another indication for lamellar keratoplasty. After
lamellar keratoplasty, such patients are visually rehabilitated
without worry of future endothelial graft reactions. Visually signicant scars following PRK, LASIK, or RK complications can
be successfully managed with optical lamellar keratoplasty
(Fig. 59.1). The grossly fluctuating vision commonly seen after
RK is also ameliorated following lamellar keratoplasty. Optical
lamellar keratoplasty is also useful for the treatment of anterior
stromal scars following trauma or infectious keratitis. Corneal
dystrophies such as ReisBckler, Salzmanns nodular dystrophy
and lattice, granular, or macular dystrophy can be treated with
lamellar keratoplasty. ReisBckler dystrophy commonly recurs in
lamellar grafts but this problem is safely managed by replacing
the graft.18 The treatment of macular dystrophy with lamellar
keratoplasty should be limited to those cases without advanced
endothelial involvement.
Tectonic lamellar keratoplasty reestablishes the corneas structural integrity. Tectonic grafts may be large in diameter covering
FIGURE 59.2. Depicted are electron
photomicrographs of anterior and posterior
stroma. (a) The interlacing lamellae found in the
anterior one-third of the corneal stroma. The
interweaving conguration makes manual
lamellar dissection difcult at this level. (b) The
parallel arrangement of the posterior lamellae.
Photomicrographs are courtesy of Dr Jan Bergmanson.
SURGICAL PROCEDURE
Key Features: Surgical Procedure
The cohesive force necessary to separate corneal lamellar

varies in different areas of the corneal stroma. The cohesive
strength of the peripheral and anterior stroma is greater than
the central and posterior stroma. Therefore manual lamellar
dissection is optimal when performed in the posterior stroma
and becomes increasingly easy as one dissects from the
periphery towards the center.
Use of the microkeratome and articial anterior chamber
facilitates the preparation of donor lamellar tissue that has an
optically smooth surface.
A variety of surgical approaches have been described for lamellar

keratoplasty. The advantages and shortcomings of these various
techniques are best appreciated with a basic understanding of the
biomechanical properties and morphology of the corneal stroma.
The stroma comprises 90% of the total corneal thickness. A
recent study of normal corneal stroma found that it consists of
242 + 6 lamellae.22 Other reports state that stroma contains
300500 lamellae.23 For the purpose of developing an optimal
lamellar procedure, the orientation of corneal lamellae in dif-
ferent areas of the corneal stroma is more important than their

exact number (Fig. 59.2). Anteriorly, the lamellae are interwoven and often oriented obliquely to the corneal surface. In
the posterior two-thirds of the stroma, the lamellae are oriented
parallel to the corneal surface.24 These ndings explain why
mechanical shearing of corneal lamellae is much more difcult
in the anterior stroma than in the posterior stroma. Collagen
interweaving is also more extensive in the corneal periphery than
in its center.25 This explains the observation that the cohesive
strength of the corneal stroma is greater in the corneal periphery
than in the central cornea.26 Awareness of the relative cohesive
strength of different areas of corneal stroma is useful for optimizing lamellar surgery. Different lamellar keratoplasty techniques that apply this knowledge are described below.
MELBRANS PEELING OFF

TECHNIQUE27,28
The procedure begins with a partial depth trephination centered
over the area of host corneal tissue to be removed. A minimal
trephination depth of 300 mm is recommended to insure that the
lamellar dissection begins within posterior stromal lamellae.
Lamellar dissection of the anterior stroma is very difcult due
to the interlacing of those lamellae and even if the pathology is
only present within these lamellae the trephine depth should be
set to assure reaching the level of posterior stroma. This will
facilitate the lamellar dissection.
Following partial depth trephination, a 0.12 forceps is used to
pull the inner edge of the trephine incision centrally so that the
depth or base of the incision is visualized. A microsurgical blade
(Grieshaber 681.21) is used to carefully cut the deep lamellae
immediately anterior to the base of the trephination. This initial
lamellar dissection only extends 3 mm axially from the inner edge
of the trephine incision. No further central extension of the
lamellar dissection should be performed until this 3 mm dissection is completed along the periphery of the inner aspect of the
trephine cut for 270. Dissection of the peripheral interlacing
corneal lamellae reduces the cohesive force required to extend
the lamellar dissection axially and allows for a much easier and
safer dissection of the central posterior lamellae.
After completion of the 270 peripheral lamellar dissection, the
central corneal lamellae can be peeled away by using a Pollack
forceps or another similar two-point xation forceps. Using the
Pollack forceps sustained upward, central traction is applied to
the inner edge of the keratectomy tissue. The posterior lamellar
CHAPTER 59
the entire surface of the recipient cornea or they may be specically shaped to cover only the structurally weakened area of the
recipient cornea. Some of the more creative shapes for tectonic
lamellar keratoplasty include horseshoe- and crescent-shaped
grafts.19 Large limbal to limbal onlay tectonic grafts are sometimes useful in the treatment of advanced cases of keratoglobus.
Tectonic grafts are indicated in the treatment of peripheral, noninflammatory thinning disorders such as Terriens marginal
degeneration or pellucid marginal degeneration. Tectonic grafts
are also used to treat peripheral ulcerative keratitis in autoimmune disorders such as Moorens ulcer or rheumatoid arthritis.
The autoimmune condition must be resolved in order to ensure
graft survival.20,21 Neurotrophic corneal melt from diabetes or
herpes zoster can be managed with a tectonic graft. But the
underlying reason for the corneal melt must be treated or the
tectonic graft will fail as a result of corneal melting. In the case
of herpes zoster, inferior corneal melting is commonly associated
with exposure keratitis secondary to lagophthalmus, so the
tectonic graft should be combined with a lateral tarsarrophy.
807
SECTION 6
FIGURE 59.3. (a) An intraoperative photograph demonstrating the snow-white bers that appear as traction is applied during the peeling off
process of a lamellar keratectomy in a moderately advanced keratoconic eye. Note that the lamellar microsurgical blade is angulated so that it
teases away the white bers and remains anterior to the recipient bed making a microperforation very unlikely. (b) An intraoperative photograph
of lamellar keratectomy in an advanced keratocone. Note the lamellar keratectomy is performed with traction applied by a Pollack forceps and a
microsurgical blade is not necessary.
808
bers at the base of the lamellar keratectomy will turn snowwhite as they begin to shear (Fig. 59.3). A ne sweeping motion
with a dull microsurgical blade placed above the base of the white
bers is used to tease the bers away from the stromal bed and
extend the keratectomy centrally. As the central part of the
dissection is reached, the cohesive forces of the posterior lamellae
are easily overcome by mechanical traction alone and no cutting
with a blade is required. The safety advantage of omitting the
use of a knife over the central cornea is obvious but equally
important is that the peeling method of dissection has a natural
tendency to remain in the same lamellar plane resulting in a
very smooth dissection. In advanced cases of keratoconus, it is
common to bare Descemets membrane as the central stroma is
peeled off the apex of the cone. At this point of the dissection,
the use of a blade is both unnecessary and unsafe. As the
dissection passes over the central cornea and moves toward the
inferior edge of the trephination, additional lamellar dissection
with the microsurgical blade may be necessary. Vannas scissors
are used to complete the removal of the lamellar keratectomy
tissue once the inferior edge of the trephination is reached.
After the lamellar dissection is completed, the recipient bed is
inspected. If additional abnormal stromal tissue is present, it
can be removed by making a second deeper lamellar dissection.
Descemets membrane can be completely bared if necessary but
this is rarely worth the risk of penetrating into the anterior
chamber.
AIR DISSECTION TECHNIQUE

Deep lamellar keratoplasty refers to anterior lamellar keratoplasty
in which the lamellar dissection extends posteriorly to the level
of Descemets membrane and all posterior stroma is removed
from the recipient bed. The technique involving intrastromal air
injection to facilitate the deep lamellar dissection was originally
described by Archila29 and later modied by Anwar.30 Anwars
big bubble procedure requires accurate trephination of the host
cornea to access deep lamellar tissue at the pre-Descemets
level. Intraoperative ultrasonic pachymetry and a microcalibrated
trephine, such as the Hanna or Krumeich trephine, are useful
to obtain a pre-Descemets depth for the base of the keratotomy.
A 30-gauge needle is bent ~5 mm from the tip in such a fashion
that the terminal segment angles upward 60 and the bevel faces
down. The needle is placed on a 3 mL air syringe and then introduced bevel down into the base of the keratotomy incision. The
needle is advanced obliquely 34 mm from the keratotomy edge
in a plane just anterior to Descemets membrane. The needle is
advanced bevel down to avoid penetrating Descemets membrane
and obliquely to avoid the thinner central cornea. About 1 mL
of air is injected into the posterior stroma. If the needle has been
accurately positioned at the pre-Descemets level, a large bubble
will ll the space between the stroma and Descemets membrane.
A partial-thickness anterior keratectomy is manually performed
leaving a layer of posterior stroma intact anterior of the bubble.
Then aqueous is drained through a paracentesis. A sharp-tipped
blade is used to carefully penetrate the remaining stromal layers
at the center of the cornea. A wire spatula is inserted into this
central opening and advanced along the cleavage plane created
by the air until its tip reaches the peripheral edge of the trephination groove. The spatula is lifted anteriorly to slightly tent the
residual stromal bers and safely separate them from Descemets
membrane so they can be excised. This maneuver is repeated for
360 until all the deepest stromal layers are circularly excised
to leave a bare Descemets membrane recipient bed.
Deep lamellar keratoplasty has also been described using
balanced salt solution and viscoelastic materials instead of air
to help separate the posterior stroma from Descemets
membrane.3133
LAMELLAR POCKET TECHNIQUE

Following partial-depth trephination, the depth of the lamellar
keratoplasty must be established. The base of the annular
keratectomy incised by the trephine is inspected by opening the
keratectomy with a 0.12 forceps. The back (dull) part of a super
blade knife can be used to bluntly dissect to the desired depth.
Another approach is to use intraoperative ultrasonic pachymetry
and a micrometer-adjusted diamond knife to precisely establish
the initial depth of the lamellar keratoplasty. A third approach
for dening the depth of the keratectomy has been described by
Melles.34 This technique requires lling the anterior chamber
with air. A lamellar dissector or dull knife is placed at the base
of the annular keratectomy. At the airendothelium interface, a
specular light reflex is created as the dissector indents and enters
the stroma. The portion of the cornea posterior to the dissector
is seen as a dark band between the tip of the dissector and the light
reflex. The width of the dark band is the thickness of posterior
corneal tissue which will remain if a lamellar keratectomy is
initiated at that particular level. Once the depth of the keratectomy has been established, a Thornton ring is used to xate
and pressurize the eye and a lamellar pocket is dissected with a
Pauque knife or similar instrument. This pocket runs parallel
to the posterior lamellae ~23 mm central to the edge of the base
of the keratectomy incision and has a width of ~4 mm. Finally,
Troutman corneal splitters or similar lamellar dissectors are
inserted into the rst lamellar pocket and the posterior lamellae
are separated using a rocking fan-like motion taking care to
maintain the dissection in a uniform horizontal plane until the
entire area within the trephination has been dissected. Additional
undermining peripheral to the trephination edge for 12 mm in
the same lamellar plane provides a pocket to aid in dovetailing
the donor graft to the recipient bed. The lamellar keratectomy
is excised with Vannas scissors to expose the recipient bed.
MICROKERATOME TECHNIQUE
The popularity of refractive lamellar keratoplasty has led to the
development of microkeratomes that are reliable and easy to use.
The speed and quality of a properly performed microkeratome
lamellar keratectomy cannot be duplicated with manual techniques. Microkeratomes are designed to cut a corneal disk with
parallel faces and if a corneal surface that is not uniform is cut
with a microkeratome the surface irregularities are transferred to
the host bed.35 The use of a microkeratome to prepare the lamellar
bed in a moderate to advanced keratoconus is contraindicated
since apical thinning in such corneas could result in a corneal
perforation with potentially disastrous consequences.
Anterior stromal opacities which have a uniform thickness can
be resected very efciently with a microkeratome. One problem
when using the microkeratome for lamellar resection is the
inconsistency in the diameter of the resected tissue. Barraquer
described a combination technique to overcome this problem.
The central microkeratome resection was circumscribed by a
larger trephination. Manual lamellar dissection was performed
from the edge of the microkeratome resection to the edge of the
trephination incision. This technique offers the advantage of a
smooth central host bed created by the microkeratome and a
consistent and controlled bed diameter can be matched with
donor tissue obtained by a similar technique.
LAMELLAR DISSECTION FOR TECTONIC

GRAFTS
Lamellar dissection of abnormal tissue for tectonic purposes
often requires an innovative approach which is dictated by the
degree and location of the corneal pathology. Typically, the size
and the shape of the recipient bed are determined only after the
abnormal tissue has been completely removed. A reliable technique
for lamellar tissue resection is to outline the area to be removed
with a diamond knife set to an appropriate depth. The lamellar
resection of abnormal tissue should extend ~1 mm into the
healthy recipient tissue to ensure a recipient bed with safe margins.
Supplementary undermining of the peripheral margins of the
recipient bed will assist in securing the donor graft.
DONOR PREPARATION
The donor tissue should have a clear, healthy, Bowmans layer and
stroma. Since the quality of endothelium is not an issue, the
selection of donor tissue is expanded and tissue can be utilized
that would not be used for penetrating keratoplasty. However,
lamellar donor tissue is screened for potentially transmissible
diseases using the same criteria required for penetrating keratoplasty donor tissue.
Donor tissue preparation can be achieved with a number of
techniques including free-hand dissection using a whole globe,
manual removal of endothelium and Descemets membrane from
corneal scleral rim, microkeratome preparation using a whole globe
or articial anterior chamber or tissue lathing on a Barraquer
cryolathe. When free-hand dissection using a whole globe is used
to prepare donor tissue it is helpful to increase the intraocular
pressure of the donor globe prior to the lamellar dissection. This is
easily done by injecting air through the optic nerve with a 30-gauge
needle. A diamond knife is then set to the proper depth to obtain
the desired thickness of the lamellar graft. The diamond knife
is then used to cut a 120 arc along the limbus of the donor globe.
An anterior lamellar dissector is inserted into the base of the
limbal incision and the posterior lamellae are separated using a
side-to-side sweeping motion. Care is taken to keep the dissector
in the same horizontal plane until the dissection is complete from
limbus to limbus for the whole diameter of the donor cornea.
Trephination is then performed to obtain a donor graft with the
desired diameter. If a special shape of donor graft is desired
(crescent or horseshoe), the entire diameter of the donor cornea
is removed following the donor lamellar dissection and then
this oversized graft is placed over the host recipient bed and a
graft is fashioned to t the shape of the recipient bed.
The technique of donor material preparation by manual removal
of Descemets membrane and endothelium from a corneal scleral
rim is facilitated by using the vital dye trypan blue. The trypan
blue aids in the visualization of endothelium and Descemets
membrane, which can then be removed with a nontoothed forceps
and brisk rubbing with a Weck cell sponge. A full-thickness
stromal lamellar graft is obtained and can be trephined to the
desired diameter.33
Donor tissue preparation with a microkeratome is gaining
popularity as an increasing number of corneal surgeons become
familiar with the Moria microkeratome and articial chamber
to perform the DSAEK (Descemets stripping automated endothelial keratoplasty) procedure. The Moria system is also an
excellent method for preparing anterior lamellar keratoplasty
tissue (Fig. 59.4). The microkeratome is available with a range
of heads for cutting tissue from 100 to 475 mm in thickness.
The Moria microkeratome can be used with a whole globe, but
corneal scleral rims can also be used with the Moria articial
chamber. The system facilitates donor tissue preparation with a
smooth surface and requires a minimal learning curve. Once the
tissue is cut, it can easily be trephined to the desired diameter.
The Barraquer cryolathe can also be used to prepare smooth
lamellar donor tissue with precise thickness and diameter. Use
of corneal press to obtain uniform hydration and thickness prior to
cryolathing enhances the quality and precision of the lathed tissue.
The donor tissue can be frozen for long-term storage or placed in
tissue culture medium if surgery is planned within 1 week.
OUTCOMES AND COMPLICATIONS
CHAPTER 59
Key Features: Outcomes and Complications
Malbrans peeling off technique is the safest lamellar

keratoplasty procedure and yields postoperative visual results
equal to the more risky deep lamellar techniques.
Suboptimal visual acuity following lamellar keratoplasty is often
blamed on interface haze but inferior visual results also occur if
the posterior curvature of the recipient bed is abnormal. This
abnormal posterior curvature may be the best explanation for
poor visual outcomes following lamellar keratoplasty for
keratoconus even when the interface is perfectly clear.
809
FIGURE 59.4. The Moria ALTK system is depicted. This system

allows for the preparation of the lamellar keratoplasty tissue using a
donor corneal scleral rim which is secured onto the articial chamber
platform. This particular Moria system shows a CB microkeratome
attached to the articial chamber. Different heads can be attached to
the CB microkeratome that allow for preparation of different
thicknesses of lamellar donor tissue. The diameter of the donor tissue
is determined by adjusting the height of the donor tissue protruding
above the articial chamber opening.
SECTION 6
A variety of techniques for lamellar keratoplasty have been

reported. An analysis of the results and complication rates of these
techniques provides an evidence-based approach for choosing
the optimal procedure. In 1972, Malbran reported his peeling
off technique in 115 keratoconus eyes with no perforations of
Descemets membrane.28 Using Malbrans technique and
cryolathed donor tissue, 91% of 23 eyes with contact lens
intolerant keratoconus obtained 20/30 or better corrected vision.
No intraoperative complications were encountered in this study,
which was consistent with Malbrans experience.36 Mixed
results have been reported with air dissection techniques, which

can be complicated by incomplete dissection in the area of most
stromal scarring.37 When Price evaluated this technique, three
of 10 patients experienced a perforation of Descemets membrane
requiring conversion to penetrating keratoplasty.38 Lamellar
surgical techniques using viscoelastic materials are limited.
Manche reported excellent vision in two patients using sodium
hyaluronate-assisted lamellar keratoplasty.39 Melles used hydroxypropylmethyl cellulose for viscodissection and reported perforation of Descemets membrane in ve of 25 eye bank eyes and
one of three patients undergoing his technique required conversion
to penetrating keratoplasty secondary to rupture of Descemets
membrane.33 Hydrodissection or fluid-assisted lamellar
keratoplasty resulted in 20/30 or better corrected vision in 96%
of 26 eyes with moderate to advanced keratoconus but two eyes
in this study required conversion to penetrating keratoplasty due
to perforation of Descemets membrane.31 A hydrodelamination
technique was used to treat 120 eyes with anterior corneal
pathologies and the average postoperative acuity was 20/30.32
Unfortunately, 39.2% of the eyes in this study experienced perforation of Descemets membrane. However, all eyes were
successfully treated with the lamellar keratoplasty technique and
no difference in visual acuity or endothelial cell count was reported
in the punctured versus nonpunctured group.
These outcomes indicate that Malbrans technique is the
safest procedure and yields postoperative visual results equal to
the more risky deep lamellar techniques. The nal visual results
with deep lamellar keratoplasty do not justify the risk for
Descemets membrane perforation. The main objective of lamellar keratoplasty should be to obtain a smooth host bed and not
complete removal of posterior lamellae to the level of
Descemets membrane. Deep lamellar dissection to Descemets
membrane need only be done when the level of pathology
requires it and such deep dissection should not be the goal of every
lamellar keratoplasty. Proponents of deep lamellar keratoplasty
argue that they see less interface haze with lamellar keratectomy
to Descemets level. However, more recent studies indicate
excellent results with lamellar keratoplasty, in which a deep
dissection is performed leaving 5060 mm of posterior stroma
and not bearing Descemets membrane.40 The decreased visual
results following lamellar keratoplasty are not always secondary
to interface haze. Inferior visual acuity after lamellar keratoplasty
is just as likely to result from an abnormal posterior curvature.
This is certainly the best explanation for inferior visual results
obtained when lamellar keratoplasty is used to treat keratoconus.
The posterior elevation changes in keratoconus, which can be
810
FIGURE 59.5. (a) A sclerotic scatter slit-lamp photograph showing a clear lamellar graft in a 16-year-old keratoconic patient. (b) The slit-lamp
photograph of the same patient demonstrating the crystal interface between the graft and the patients recipient bed.
topographically measured with Orbscan, are especially difcult
to eliminate with lamellar keratoplasty in older patients. These
same changes are much easier to reverse in younger patients,
which explains why younger patients achieve better and faster
visual results following lamellar keratoplasty than older patients.
These ndings suggest that lamellar keratoplasty should be considered as an earlier treatment option in keratoconic patients in
order to optimize results with lamellar keratoplasty and avoid
penetrating keratoplasty (Fig. 59.5).36
The most serious intraoperative complication with lamellar
keratoplasty is perforation of Descemets membrane. This usually
occurs with deep lamellar dissection but rarely may occur during
trephination. If the perforation site is small enough, the procedure can be continued by reforming the anterior chamber with
air. Rarely, suturing the perforation site is needed. If air is left in
the anterior chamber at the completion of the procedure, the
pupil should be pharmacologically dilated to avoid pupillary block
glaucoma. Perforation of Descemets membrane during the
procedure can also lead to the accumulation of aqueous fluid in
the interface between the host bed and the lamellar graft creating
what is called a pseudo anterior chamber (Fig. 59.6). This complication is managed by lling the anterior chamber completely
with air and then draining the aqueous fluid from the space
between the graft and host bed by opening the anterior operative
wound with a 30-gauge cannula.
Another early postoperative complication is failure of the graft
to epithelialize. This may be due to preexisting ocular surface
disease. Treatment of the lid margins with lid scrubs and oral
tetracycline may be indicated. If lagophthalmus is present, taping
of the lids or a lateral tarsarrophy may be required. The use of
preservative-free articial tears may be helpful. Topical steroids
should be avoided until the graft is completely re-epithelialized.
FIGURE 59.6. Slit-lamp photograph shows aqueous humor within the

grafthost interface as a result of a microperforation of Descemets
membrane which occurred during the lamellar dissection of a patient
undergoing lamellar keratoplasty for the treatment of corneal scarring
following RK. The aqueous-lled space called a pseudo-anterior
chamber resolved spontaneously after 2 weeks.
Epithelial ingrowth under the graft can lead to a graft melt and
may require elevating the graft and debriding the ingrowth.41
Stromal graft reactions occur rarely. They present as subepithelial
opacities similar in appearance to the subepithelial inltrates
that occur with adenovirus infection but without a history of
conjunctivitis. Unlike endothelial graft reactions, stromal graft
reactions are not sight threatening and are managed with
topical steroids.
1. Reisinger F: Die Keratoplastik: ein Versuch

zur Erweiterung der Augenheil kunst.
Bayerische Annalen 1824; 1:207215.
2. Mhlbauer FX: Ueber Trasnplasntation der
Cornea (Gekronte Preisschrift, Jos.
Lindaauer, Mucich 1840). In: Schmidt CC,
ed. Jahrbcher der in- und auslndischen
gesammten Medicin. Leipzig: Otto
Weigand; 1842: 267288.
3. von Hippel A: Eine neue methode der
Hornhauttransplantation. Arch Ophthalmol
1888; 34:105130.
4. Zirm E: Eine rfolgreiche totale
Keratoplastik. Archiv Ophthalmol 1906;
64:580593.
5. Fuchs E: Zur Keratoplastik. Ztschr fr
Augenheilk 1901; 5:15.
6. Elschnig A: In: Czermak W, ed. Die
Augenrtzlivhe Operationen. 2nd edn.
Berlin: Urban & Schwarzenberg;
1908:84109.
7. Magitot A: Transplantation of the human
cornea previously preserved in an
antiseptic fluid. JAMA 1912; 59:1821.
8. Magitot A: Recherches exprimentales sur
la survie possible de la corne conserve en
dehor de lorganisme et sur la keratoplastie
diffre. Ann docul. 1911; 146:134.
9. Elschnig A: ber Keratoplastik. Prag Med
Wschr. 1914; 36:405.
10. Elschnig A: Keratoplasty. Arch Ophthalmol
1930; 4:165173.
11. Pauque L, Sourdille GP, Offret G: Les
Greffes de la Core. Paris: Masson; 1948.
12. Terry MA: The evolution of lamellar grafting
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
techniques over twenty-ve years. Cornea

2000; 19:611616.
Bourne WM: Cellular changes in transplanted
human corneas. Cornea 2001; 20:560569.
Olson T: Postop changes in endothelial cell
density of corneal grafts. Acta Ophthal
1985; 59:863870.
Watson SL, Ramsay A, Dart JK, et al:
111:16761682.
Van Dooren BT, Mulder PG, Nieuvendaal
CP, et al: Endothelial cell density after deep
lamellar keratoplasty (Melles technique).
Am J Ophthalmol 2004; 137:397400.
Morris E, Kirwan JF, Sujatha S, Rostron CK:
Corneal endothelial specular microscopy
following deep lamellar keratoplasty with
lyophilized tissue. Eye 1998; 12:619622.
Olsen RJ, Kaufman HE: Recurrence of
Reis-Bcklers corneal dystrophy in a graft.
Schanzlin DJ, Sarno EM, Robin JB:
Crescentic lamellar keratoplasty for pellucid
marginal degeneration. Am J Ophthalmol
1983; 96:253254.
McDonnell P: Recurrence of Moorens ulcer
after lamellar keratoplasty. Cornea 1989;
8:191194.
Foster CS: Systemic immunosuppressive
ulcer. Ophthalmol 1985; 92:14361439.
Bergmanson JPG, Horne J, Doughty M,
et al: Assessment of the number of
23.
24.
25.
26.
27.
28.
29.
30.
31.
lamellae in the central region of the normal

human corneal stroma by transmission
electron microscopy. Eye Contact Lens
2005; 31:281287.
Maurice DM: The cornea and sclera. In:
Davson H, ed. The eye. Vegetative
physiology and biochemistry. Orlando, FL:
Academic Press; 1984:1158.
Komai Y, Ushiki T: The three-dimensional
organization of collagen bers in the human
cornea and sclera. Invest Ophthalmol Vis
Sci 1991; 32:22442258.
Polack FM: Morphology of the cornea. I.
Study with silver stains. Am J Ophthalmol
1961; 51:10511056.
Smolek MK, McCarey BE: Interlamellar
adhesive strength in human eye bank
31:10871095.
Polack FM: Lamellar keratoplasty:
Malbrans peeling off technique. Arch
Ophthalmol. 1971; 86:293295.
Malbran E, Stephani C: Lamellar
keratoplasty in corneal ectasias.
Archila E: Deep lamellar keratoplasty
dissection of host tissue with intrastromal
air injection. Cornea 1985; 3:217218.
Anwar M, Teichmann MD: Big-bubble
technique to bare Descemets membrane in
anterior lamellar keratoplasty. J Cataract
Refract Surg 2002; 28:398403.
Amayen AF, Anwar M: Fluid lamellar
keratoplasty in keratoconus. Opthalmology
2000; 107:7679.
CHAPTER 59
REFERENCES
811
SECTION 6
32. Suguita J, Kondo M, Monden Y, et al: Deep

lamellar keratoplasty with complete
removal of pathological stroma for vision
improvement. Br J Ophthalmol 1997;
81:184188.
33. Melles G, Remeijer L, Geerards A: A quick
surgical technique for deep, anterior
lamellar keratoplasty using viscodissection. Cornea 2000; 19:427.
34. Melles G, Rietveld F, Remeijer L, et al: A
technique to visualize corneal incision and
lamellar dissection depth during surgery.
Cornea 1999; 18:8086.
35. Barraquer J: Lamellar keratoplasty
(special techniques). Ann Ophthalmol 1972;
4:437469.
812
36. Benson W, Goosey C, Prager T, et al:

Visual improvement as a function of
time after lamellar keratoplasty for
116:207211.
37. Chau G, Dilly S, Sheard C, et al: Deep
lamellar keratoplasty with complete
removal of pathological stroma for vision
improvement. Br J Ophthalmol 1997;
81:184188.
38. Price FW Jr: Air lamellar keratoplasty.
39. Manche EE, Holland GN, Maloney RK:
Deep lamellar keratoplasty using
viscoelastic dissection. Arch Ophthalmol
1999; 117:15611565.
40. Marchini G, Mastropasqua L, Pedrotti E,

et al: Deep lamellar keratoplasty by
intracorneal dissection. Ophthalmol 2006;
113:12891300.
41. Busin M, Arffa RC, Sebastian A:
Endokeratoplasty as an alternative to
penetrating keratoplasty for the surgical
treatment of diseased endothelium: initial
results. Ophthalmology 2000;
107:20772082.
CHAPTER
60
Penetrating Keratoplasty
Leonard P. K. Ang, S. Arthur Boruchoff, and Dimitri T. Azar
INTRODUCTION
Penetrating keratoplasty involves surgical removal of diseased
or damaged cornea from the host and replacement with a fullthickness donor cornea. Depending on the location of the corneal abnormality, it may be sufcient to replace just the anterior
layers with anterior lamellar keratoplasty. When the endothelium is involved, replacement of the corneal tissue must include
the endothelium, and a penetrating keratoplasty is then usually
performed. The major goals of penetrating keratoplasty are to
improve visual acuity, to maintain the integrity of the eye and
to treat various infections or neoplastic conditions. Corneal
grafting techniques date back to the nineteenth century and the
earlier part of the twentieth century.1 von Hippel, in 1886, was
the rst person to successfully perform a lamellar corneal graft
in a human.2 Later, penetrating keratoplasty became more
common as techniques were improved and as realization of the
importance of the endothelium became more widespread.
Penetrating keratoplasty is now one of the most frequently
performed transplant procedures, exceeding those for other solid
organs such as the kidney, liver, and heart.36 In the United
Sates alone, a total of 46 625 corneal transplants were done in
the year 2002.4 Part of the reason for the increased frequency
may be attributed to improvements in keratoplasty techniques,
surgical instrumentation, and training of surgeons. In addition,
advances in graft immunology have signicantly improved graft
survival and expanded the indications for corneal transplantation. At the beginning of the century, corneal scars, and later,
keratoconus, were the major indications for penetrating keratoplasty. With a higher incidence of degenerative endothelial
disease, such as Fuchs dystrophy, as well as the rising incidence
of cataract-surgery-related endothelial dysfunction and corneal
edema, corneal edema-associated conditions have become the
major indications for penetrating keratoplasty.7,8
Penetrating keratoplasty is now one of the most successful
transplantations, with uncomplicated low-risk grafts having a
2-year survival rate of above 90%,910 and a 10-year success rate
of 7080%.11,12 Over the past few decades, penetrating keratoplasty has been much more commonly performed compared to
lamellar keratoplasty, owing to its remarkable surgical success.
However, in recent years, lamellar transplantation procedures
have started to achieve more prominence, owing to advances in
surgical techniques and instrumentation. Depending on the
location of the corneal abnormality, it may be sufcient to just
replace the anterior corneal layers (anterior lamellar keratoplasty)
or the posterior corneal stroma and endothelium (posterior
lamellar keratoplasty, deep lamellar endothelial keratoplasty, or
Descemets stripping endothelial keratoplasty).8,1317 In addition,
the use of semiautomated microkeratomes (automated lamellar
therapeutic keratoplasty) has helped to overcome some of the
technical difculties associated with lamellar surgery and to

enhance interface smoothness.18
Despite improvements in surgical technique and antirejection therapy, a large group of eyes with severe ocular surface
disease arising from the StevensJohnson syndrome, chemical
or thermal burns, or cicatricial pemphigoid, are poor candidates
for conventional keratoplasty alone. Corneal grafting in these
patients is associated with dismal results because of the severe
anterior segment damage and the fact that central keratoplasty
alone does not address the associated complications, such as
limbal stem cell deciency, poor tear function and lid abnormalities. In these patients, replacing the cornea with an articial
cornea or implant (keratoprosthesis) may be the only option for
visual rehabilitation.
Lamellar keratoplasty and keratoprosthesis surgery are
beyond the scope of this chapter, and will be discussed in other
chapters. This chapter reviews the indications, donor selection
and eye banking, preoperative considerations and preparation,
as well as the surgical techniques and complications related to
penetrating keratoplasty.
INDICATIONS FOR SURGERY

The main aims and surgical indications for penetrating keratoplasty include the following:
1. Optical. Improving visual acuity is the most common
indication for penetrating keratoplasty. The aim is to
remove corneal opacication and disease that is obscuring
the visual axis. Endothelial decompensation and corneal
edema arising from pseudophakic and aphakic bullous
keratopathy or Fuchs endothelial dystrophy may severely
degrade the vision, and cause glare, haloes, and ocular
discomfort. Penetrating keratoplasty may be performed for
these diseases, as well as diseases where irregularity of the
corneal surface or shape precludes good vision, such as in
keratoconus and marginal degenerations. The more the
cornea deviates from sphericity, the poorer is its ability to
act as a good refracting surface.
2. Tectonic. Penetrating keratoplasty may be used for the
restoration of normal corneal thickness or integrity of the
globe, such as in diseases causing corneal thinning and
perforation.
3. Therapeutic. Active corneal infections of bacterial, viral,
fungal, or parasitic origin, which are unresponsive to
medical treatment may also require surgical intervention.
The potential risk of perforation or intraocular spread
of infection may warrant performing a full-thickness
corneal transplantation. Corneal grafting may also be
used to eradicate benign or malignant tumors of the
cornea.
813

In terms of disease entities, developed countries with an established corneal program list pseudophakic or aphakic bullous
keratopathy, re-grafts, keratoconus, corneal scars, and corneal
dystrophies as the major indications for corneal grafting.1927 In
America, between 1990 and 2002, member eye banks of the Eye
Bank Association of America (EBAA) data showed that the most
common recipient diagnoses were pseudophakic bullous
keratopathy, followed by Fuchs endothelial dystrophy, repeat
corneal transplantation, and keratoconus.4 The order of frequency
for the various conditions varies from study to study.1927 Other
indications for penetrating keratoplasty include corneal ulcers
(bacterial, fungal, parasitic, or viral), corneal dystrophies and
degenerations, trauma, and any other causes of corneal
decompensation.
In less developed countries, corneal infections and scarring
were the main indications for penetrating keratoplasty.22 Socioeconomic progress and affluence, improvements in antimicrobial
therapy for corneal infections, renement of surgical techniques,
improved instrumentation, and the widespread acceptance of
cataract surgery have contributed to the changing trends of indications for penetrating keratoplasty. In the 1980s, the marked
increase in the number of cataract operations being performed
resulted in aphakic and pseudophakic bullous keratopathy
becoming the main indications for penetrating keratoplasty in
these countries.
The use of more efcacious antirejection therapies in the
form of topical and systemic immunosuppressive agents, such
as cyclosporine A, has signicantly improved graft survival among
re-grafted patients. As such, surgeons are now more inclined to
offer repeat corneal transplantation to these patients. Improved
surgical techniques have also led to the broadening of indications for penetrating keratoplasty to include conditions that
were previously deemed inoperable, such as iridocorneal endothelial syndrome,28 herpes zoster ophthalmicus,29 and marginal
degenerations.30
TABLE 60.1. What the Eye Bank Tells the Surgeon*

1. Name of (source) eye bank
2. Location of eye bank
3. Telephone number of eye bank
4. Eye bank identication number unique to each tissue graft
5. Type of preservation medium
6. Age of donor
7. Cause of death
8. Death date and time
9. Preservation date and time
10. Name of technician who enucleated, excised, and
evaluated the tissue
11. Slitlamp report/date
12. Specular microscopy report/date
13. Results of serologic testing
14. Results of corneal cultures (if performed)
*EBAA standards, 2005.
TABLE 60.2. Tissues Requiring Special Handling* (Potential

Danger to Eye Bank Personnel)
1. Active viral hepatitis
2. AIDS or HIV seropositivity
3. Viral encephalitis, active or of unknown origin
4. CreutzfeldtJakob disease
5. Rabies
SECTION 6
DONOR SELECTION AND EYE BANKING
814
Donor corneal tissue used for lamellar keratoplasty need not

contain viable cells. As such, lamellar grade tissues may be used
fresh, preserved by short-term methods, or frozen for long-term
storage. In contrast, the viability of the endothelium is a major
determining factor in the success of penetrating keratoplasty.
Donor tissue needs to be carefully chosen and meticulously stored.
The viability of the epithelium is less important, and some
surgeons actually prefer to remove these cells, so as to reduce
the antigenic stimulus for graft rejection. However, in conditions
where corneal stem cells are decient or absent, such as in severe
ocular surface disease arising from StevensJohnson syndrome
or chemical injury, retention of a healthy epithelium becomes
an important factor in the survival of the corneal graft.
The decision to use a specic tissue is made by the surgeon
after weighing multiple factors. To assist the surgeon, the eye
bank provides specic information about the donor and relevant
circumstances (Table 60.1).
The EBAA sets criteria for donor tissue selection and eye
banking in the US.31 An EBAA committee meets periodically to
review the most recent scientic data with regard to transmissibility of disease or techniques for evaluation of corneal integrity
and revises its criteria accordingly. All eye banks afliated with
the EBAA use the same criteria, although certain of the criteria
are not specied but are left to the discretion of the medical
director of the individual eye bank. Ultimately, the individual
surgeon makes the nal decision with regard to accepting
specic tissue (Table 60.2).
The donor tissue is screened by reviewing the donors medical
history and performing serological testing, so as to ensure that
severe or life-threatening infectious diseases, such as human

immunodeciency virus (HIV),32 viral disease (e.g., hepatitis B
and C) and prions, such as CreutzfeldJakob disease, are excluded
(Table 60.3).33,34 One of the main parameters in determining
suitability of the donor tissue is the viability of the endothelial
cell layer, which is determined by specular microscopy. The age
of the donor is not a critical factor in long-term graft survival,
although extremes of age pose unique problems that may compromise the success of the surgery.20,3439 Corneas obtained
from infants within the rst year of age are more flaccid, and
predispose the recipient to high myopia, while those from
donors older than 65 years of age are more likely to have endothelial abnormalities and dysfunction.
In order to preserve the quality of the endothelial cell layer,
the time interval between donor death and cornea harvesting or
enucleation, and the time to preservation in articial media
should be kept as short as possible.
Several methods of corneal preservation have been employed
to ensure endothelial cell viability. A freshly enucleated whole
globe may be stored in a moist chamber at 4oC.40,41 The globe is
kept in a sterile container with a gauze pad cushion, and excellent endothelial viability may be maintained for at least 2 days.
Corneoscleral buttons are now commonly stored in Optisol,42
which comprises a buffered solution containing various tissue
culture factors, antibiotics, and dextran. Endothelial viability can
be maintained for up to a week with this preservation method.
TABLE 60.3. Tissue Unsuitable as Donors*

Tissue from donors with the following are potentially health threatening for the recipient(s) or pose a risk to the success of the surgery and
shall not be offered for surgical purposes.
1. Death of unknown cause
2. CreutzfeldtJakob disease (CJD), variant CreutzfeldtJakob disease (vCJD), or family member with CJD
3. Death with neurologic disease of unestablished diagnosis
4. Dementia, unless due to cerebrovascular disease, brain tumor, or head trauma. Donors with toxic or metabolic-induced-dementia may be
acceptable pending documentation of consultation with the Medical Director. The approval of the Medical Director is required
5. Subacute sclerosing panencephalitis
6. Progressive multifocal leukoencephalopathy
7. Congenital rubella
8. Reyes syndrome
9. Active viral encephalitis or encephalitis of unknown origin or progressive encephalopathy
10. Active septicemia (bacteremia, fungemia, viremia)
11. Active bacterial or fungal endocarditis
12. Active viral hepatitis
13. Rabies
14. Intrinsic eye disease
a. Retinoblastoma
b. Malignant tumors of the anterior ocular segment or known adenocarcinoma in the eye of primary or metastatic origin
c. Active ocular or intraocular inflammation: conjunctivitis, scleritis, iritis, uveitis, vitreitis, choroiditis, retinitis
d. Congenital or acquired disorders of the eye that would preclude a successful outcome for the intended use (e.g., a central donor
corneal scar for an intended penetrating keratoplasty, keratoconus, and keratoglobus)
e. Pterygia or other supercial disorders of the conjunctiva or corneal surface involving the central optical area of the corneal button
15. Prior intraocular or anterior segment surgery
a. Refractive corneal procedures, e.g., radial keratotomy, lamellar inserts, etc.
b. Laser photoablation surgery is allowed to be used in cases of tectonic grafting and posterior lamellar procedures
c. Corneas from patients with anterior segment (e.g., cataract, intraocular lens, glaucoma ltration) surgery may be used if screened
by specular microscopy and meet the Eye Banks endothelial standards
d. Laser surgical procedures such as argon laser trabeculoplasty and retinal and panretinal photocoagulation do not necessarily
preclude use for penetrating keratoplasty but should be cleared by the Medical Director
16. Leukemias
17. Active disseminated lymphomas
18. Hepatitis B surface antigen-positive donors
19. Recipients of human pituitary-derived growth hormone during the years from 1963 to 1985
CHAPTER 60
20. HTLV-I or HTLV-II infection

21. Recipient of non-synthetic dura mater graft
22. Hepatitis C-seropositive donors
23. HIV-seropositive donors
22. HIV or high risk for HIV: Persons meeting any of the following criteria should be excluded from donation:
Behavioral/History Exclusionary Criteria
(May, 1994 CDC Guidelines)
a. Men who have had sex with another man in the preceding 5 years
b. Persons who have injected drugs for a nonmedical reason in the preceding 5 years including intravenous, intramuscular, or
subcutaneous injection of drugs
c. Persons with hemophilia or related clotting disorders who have received human-derived clotting factor concentrates
d. Men and women who have engaged in sex for money or drugs in the preceding 5 years
e. Persons who have had sex in the preceding 12 months with any person described in items ad above or with a person known or
suspected to have HIV, hepatitis B, or hepatitis C virus infection
Continued
815
TABLE 60.3. Tissue Unsuitable as Donors*Contd

f. Persons who have been exposed in the preceding 12 months to known or suspected HIV, HBV, and/or HCV-infected blood through
percutaneous inoculation or through contact with an open wound, nonintact skin, or mucous membrane
g. Inmates of correctional systems (including jail and prisons) and individuals who have been incarcerated for more than 72 consecutive
hours during the previous 12 months
h. Persons who have had close contact with another person having viral hepatitis within the 12 months preceding donation
i. Persons who have had or have been treated for syphilis or gonorrhea during the preceding 12 months
j. Persons within the last 12 months of donation who have undergone tattooing, acupuncture, ear or body piercing in which shared
instruments are known to have been used
Specic Exclusionary Criteria for Pediatric Donors
k. Children meeting any of the exclusionary criteria listed above for adults should not be accepted as donors
l. Children born to mothers with HIV infection or mothers who meet the behavioral or laboratory exclusionary criteria for adult donors
regardless of their HIV status should not be accepted as donors unless HIV infection can be denitely excluded in the child as follows:
Children >18 months of age who are born to mothers with or at risk for HIV infection, who have not been breast fed within the last
12 months, and whose HIV antibody tests, physical examination, and review of medical records do not indicate evidence of HIV
infection can be accepted as donors
m. Children <18 months of age who are born to mothers with or at risk for HIV infection or children of mothers with or at risk of HIV
infection who have been breast fed within the past 12 months should not be accepted as donors regardless of their HIV tests results
Laboratory and Other Medical Exclusionary Criteria
n. Persons who cannot be tested for HIV infection because of refusal, inadequate blood samples (e.g., hemodilution that could result in
false-negative tests), or any other reasons
o. Persons with a repeatedly reactive screening assay for HIV1 or HIV antibody regardless of the results of supplemental assays
p. Persons whose history, physical examination, medical records, or autopsy reports reveal other evidence of:
HIV infection or high-risk behavior, such as a diagnosis of AIDS, unexplained weight loss, night sweats, blue or purple spots on the
skin or mucous membranes typical of Kaposis sarcoma, unexplained lymphadenopathy lasting > 1 month, unexplained
temperature > 100.5F (38.6C) for > 10 days, unexplained persistent diarrhea, unexplained persistent cough or shortness of breath,
or opportunistic infections
Hepatitis B or C infection, which could include clinical signs and symptoms of hepatitis such as unexplained yellow jaundice or
hepatomegaly (record of laboratory data such as alanine aminotransferase (ALT), aspartate aminotransferase (AST), bilirubin, or
prothrombin time may assist in making a donor suitability determination)
25. Smallpox vaccine exclusionary criteria
a. Smallpox vaccination without complications
Potential donors who received the smallpox vaccine without complications shall be deferred until after the vaccination scab has
separated and the vaccination site appears to be healed and not inflamed, or for 21 days post-vaccination, whichever is the later date
b. Smallpox vaccination with complications that have resolved
Potential donors who received the smallpox vaccine and developed complications that have resolved shall be deferred for 14 days
after all vaccine complications have completely resolved, or for 21 days post-vaccination, whichever is the later date
c. Smallpox vaccination with complications that have not resolved
SECTION 6
Potential donors who received the smallpox vaccine and developed complications that have not resolved shall be deferred
d. Symptomatic contacts of recipients of smallpox vaccine
Potential donors who have had contact with someone who has received the smallpox vaccine shall be deferred in cases where the
donors have had recognizable signs or symptoms attributable to the virus within 14 days prior to donation
In Europe, longer preservation of corneal tissue for up to several

weeks has been employed.4345 Long-term storage of corneal
tissue may be performed by cryopreservation at 80oC. Use of
cryopreserved tissue is now mainly limited to lamellar transplantation or in emergency situations where fresh corneal tissue
is not available.46
RECIPIENT CONSIDERATIONS AND

PREOPERATIVE EVALUATION
816
Obtaining a thorough history and performing a complete preoperative ophthalmic examination is required prior to corneal
transplantation. The surgeon should ensure that he has exhausted

all medical measures as well as all optical means of visual correction (e.g., contact lenses) before recommending a penetrating
keratoplasty for the treatment of the corneal disease.
The assessment of the visual potential of the affected eye
includes obtaining a history of the patients vision prior to the
development of the corneal disease, as well as eliciting a history
of amblyopia, strabismus, glaucoma, or retinal problems. In cases
where visualization of the posterior segment is difcult because
of corneal edema or scarring, patients considered for transplantation should generally be tested for four-quadrant light projection, pupillary reflexes to light, and the presence of color vision,
SURGICAL TECHNIQUES
PREOPERATIVE PREPARATION
Penetrating keratoplasty is now often performed as an ambulatory surgical procedure. General anesthesia is the preferred
method of anesthesia for many surgeons. This is particularly
important in patients who are young, anxious, and when the duration of surgery is likely to be prolonged due to the complexity of
the case or when multiple procedures are planned. Alternatively,
regional anesthesia in the form of a peribulbar or retrobulbar
injection of a mixture of lidocaine and a longer-acting anesthetic (e.g., bupivacaine) may be used.
PREOPERATIVE MEDICATION
Depending on the phakic status of the eye or whether combined
procedures are planned, such as cataract removal, posterior
chamber, or sulcus xated intraocular lens (IOL) implantation,
preoperative mydriatics may be used prior to penetrating keratoplasty. When no additional procedures are contemplated, preoperative miotics may be used to provide added security to help
reduce the risk of lens injury, inadvertent IOL damage, and
vitreous prolapse. Some surgeons also administer an osmotic

agent, such as mannitol, prior to surgery, to help reduce the
vitreous volume and IOP, thereby reducing the risk of vitreous
prolapse and expulsive hemorrhage.
PATIENT PREPARATION
The patient is positioned on the operating table, with the globe
position directed vertically upwards. A Honan balloon or ocular
massage may be applied for several minutes to reduce the IOP
and reduce the risk of vitreous prolapse. The lid speculum chosen
to achieve adequate exposure should not cause signicant pressure on the globe. Superior rectus and inferior rectus traction
sutures may be used to aid in the posturing and stabilization of
the eye. Scleral support with a Flieringa ring may be used for
selected patients that have a greater tendency for scleral collapse, such as previously vitrectomized eyes, aphakic eyes, and
pediatric patients. The appropriately sized ring is sutured onto
the episclera ~34 mm from the limbus. The most commonly
used sizes are the 17- and 18-mm rings. The Flieringa ring should
be removed once adequate sutures (either four or eight sutures)
have been placed to maintain the circular shape of the graft.
GRAFT CENTRATION AND MEASUREMENT

The geometric and optical axes should be considered before
marking the center of the cornea. The geometric axis is determined by the intersection of the vertical and horizontal corneal
meridians, while the optical axis represents the center of the
pupil, which is often slightly nasally displaced. Most surgeons
prefer to use the geometric center of the cornea for centration of
the graft, although some surgeons prefer to position the graft
center slightly nasally to take into account the nasally displaced
pupil. Surgical callipers may be used to help in determining the
geometric center of the cornea. This point is marked with gentian violet ink. Radial markings may also be applied onto the
cornea using a 16 radial keratectomy (RK) marker, which serves
as a guide for the radial and symmetric placement of sutures. In
situations where the nature of the corneal pathology is marked
eccentric, it may be necessary to decenter the graft, such as in
advanced keratoconus with a hanging cone, peripheral disorders
such as pellucid marginal degeneration, and peripheral ulcerative
conditions.
The appropriate graft size is determined, based on the size of
the corneal pathology and the size of the cornea. In general,
small grafts less than 6.5 mm in diameter are optically less
satisfactory as they are associated with higher degrees of astigmatism. In addition, less endothelium is transplanted, which
would be less desirable for patients with endothelial disease. On
the other hand, large grafts greater than 8.5 mm in diameter are
associated with a higher risk peripheral anterior synechiae,
glaucoma, and allograft rejection due to their close proximity to
the limbal vasculature. The most common diameters of trephines
used to cut the recipient cornea range from 7.0 to 8.0 mm.
CHAPTER 60
thus denoting some degree of physiological retinal and optic

nerve function. Ancillary investigations such as B-scan ultrasonography, optical coherence tomography, and electrophysiological studies may be required to evaluate the visual potential
of the eye.
Glaucoma, if present, should be adequately treated prior to
corneal transplantation. Every attempt should be made to optimize the intraocular pressure (IOP) control prior to penetrating
keratoplasty, because keratoplasty may worsen glaucoma and
postoperative glaucoma jeopardizes graft clarity and is an
important risk factor for graft failure. Preexisting inflammation
should be controlled as well as possible before surgery. Preliminary treatment with topical steroids prior to surgery improves
the outcome in eyes that are chronically inflamed.
Careful attention should be paid to the regularity of the lid
margins, the adequacy of lid closure and blinking, as well as the
tear function. Lid abnormalities such as trichiasis, lagophthalmos,
entropion, and ectropion may interfere with graft epithelialization and should be corrected prior to penetrating keratoplasty.
Diminished tear production and dry eye may predispose to
corneal epithelial problems, and intact fth and seventh cranial
nerves are necessary for ensuring adequate corneal protection
and lid closure.
Ocular surface diseases arising from conditions such as
StevensJohnson syndrome, chemical injury and ocular cicatricial pemphigoid are often associated with limbal stem cell
deciency, conjunctivalization and corneal vascularization.
These eyes are also chronically inflamed. Epithelial healing is
very poor in these eyes, and limbal stem cell transplantation
may need to be performed prior to or at the time of penetrating
keratoplasty.
A general medical evaluation is required to assess the patients
anesthetic risks, and stabilization of these medical conditions is
particularly important if general anesthesia is planned. For repeat
grafts requiring systemic immunosuppression, one should exclude
diseases that contraindicate its use, such as active infection,
previous tuberculosis, and a history of viral hepatitis.
The needs of the patient (as determined by the patients age,
occupation, and daily activities), the patients motivation, and
the likelihood of compliance with postoperative care and followup should also be taken into consideration, as these play an
important role in the long-term outcome and success of the
surgery.
DONOR TREPHINATION
Having decided the size of the graft, the surgeon should remove
the donor button before removing the host button. The donor
button may be removed from a whole globe or more commonly,
from a corneoscleral button. The corneoscleral button is placed
endothelial side up on a Teflon block and trephination is performed with disposable trephines. Donor buttons trephined from
the posterior corneal surface are slightly smaller in diameter
than buttons trephined from the anterior surface. Thus, if
punching from the endothelial side, larger donor trephines are
needed compared to trephination from the anterior side of the
817

recipient cornea. Most surgeons routinely oversize the donor
button by 0.250.5 mm. The use of a 0.25 mm oversize in the
donor counteracts the smaller graft size produced by endothelial
trephination, and produces fairly equivalent matching of the
donor and recipient trephination. Oversizing by 0.5 mm decreases
the risk of postoperative glaucoma but increases the steepness
of the graft and myopia. In certain conditions, such as keratoconus, some surgeons prefer to use grafts of the same size as the
recipient, to help to reduce the degree of postoperative myopia.
The donor and recipient corneas may be removed with a variety
of trephines. The ideal trephination is one that produces a central, uniform cut, with vertically aligned edges, while avoiding
damage to the intraocular structures. The various trephination
systems may be broadly divided into suction-assisted trephines
or free-standing, handle-mounted trephines. Examples of suctionassisted trephines include the Baron Hessburg, Hanna, and
Krumeich trephination systems. The donor cornea is often
trephined on a Teflon block using a hand-held Troutman Punch
and Solan trephine, or suction-assisted trephines such as the
Hanna system.
Following donor cornea trephination, a generous amount of
viscoelastic and fluid is then applied onto the donor to protect
the endothelium, and the button is then passed to the scrub
nurse to be carefully laid aside until needed.
SECTION 6
RECIPIENT TREPHINATION
818
Because irregular trephination of the host bed and donor are

important factors in determining the degree of postoperative
astigmatism, the aim of recipient trephination is to ensure a
round, regular, and well-centered recipient bed. The surgeon
ensures that the trephine is held perpendicular to the cornea and
centered over it using the previously marked center as a guide.
The peripheral corneal ring between the blade and the limbus
should be even to ensure optimal centration. In order to stabilize the globe, the assistant exerts mild traction on the recti
sutures and slightly uplifts the globe, while the surgeon may
further grasp the limbus with a toothed forceps for additional
stability. The handle with attached trephine is then gently rotated,
with a mild downward compressive force. The cornea is cut by
making smooth back-and-forth rotations of the trephine around
its axis while applying rm even pressure. Excessive downward
pressure may result in corneal distortion, leading to an undercut, sloping trephination edge. The anterior chamber is entered
in a controlled manner, using a sharp blade held at an angle.
Viscoelastic is then introduced to reform the anterior chamber
and to discourage prolapse of the iris, lens, and vitreous, as the
remaining cornea is excised. The corneal excision is then completed with corneal scissors. Remnant tags of corneal tissue or
Descemets membrane are carefully trimmed flush with the
scissors.
The use of suction-assisted trephines, such as Baron-Hessburg
and Hanna trephination systems have the advantage of being
able to retain the position of the trephine during trephination
without undue downwards compression. Photogrammetric
analysis of host corneal trephination reveals that the Hanna
trephination system offers the greatest precision and produces
the most vertical cut.47 Use of these suction-assisted devices
also prevents an excessive outward shelving cut as the trephine
cuts deeper layers of the cornea with downward compression. In
addition, these trephine systems also achieve better centration
with the central gunsight of the Hanna trephine, and the crossed
centration site of the Barron-Hessburg trephine. However, adequate suction may not always be easily achievable, such as in
small eyes or deformed globes, or when the ocular surface is
very irregular.
After the host button has been excised, relevant concomitant

procedures, such as cataract extraction, IOL implantation, anterior vitrectomy, or pupil reconstruction are done at this point.
Modern anterior segment surgery aims to reconstruct and correct as much of the anterior segment abnormalities as is safely
possible. Lysis of peripheral anterior synechiae, reconstruction
of the iris and pupil margin, and IOL lens exchange may be
performed.
SUTURING
The donor button is then transferred onto the recipient bed. A
viscoelastic agent is applied into the anterior chamber as well as
the edges of the recipient corneal bed just prior to placement of
the donor button, so as to protect the endothelium of the donor
cornea against contact with other intraocular structures during
suturing. The use of viscoelastic agents helps to reduce donor
endothelial cell loss after keratoplasty.48,49 Likewise, its introduction into the anterior chamber angle to open the collapsed
angle during open-sky surgery may reduce the formation of
postoperative peripheral anterior synechiae.6
Using a corneal spatula, the donor cornea is scooped out
from the Teflon block and placed over the host bed. With a
double-toothed Polack forceps, the superior edge of the donor
cornea is grasped and the needle of a 10-nylon suture is passed
radially through the two tips of this forceps. The primary xation
of the graft is usually by four interrupted 100 nylon sutures,
placed in the four quadrants 90 apart. The second suture at
6 oclock is particularly important as it determines the nal
position and symmetric geometry of the donor cornea on the
recipient, and helps in minimizing postoperative astigmatism.
The remaining 3 and 9 oclock cardinal sutures are then placed.
The anterior chamber is reformed with viscoelastic and suturing
is continued, taking care to ensure even and radial distribution
of the sutures.
The graft may be secured in place with either continuous
sutures or interrupted sutures. The various methods of suturing
include placing either 16 symmetrically placed interrupted
sutures, single or double running continuous sutures, or a combination of four or eight interrupted sutures and a single continuous suture. Interrupted sutures are used if uneven wound
healing is anticipated, such as in scarred or necrotic areas of the
recipient bed, or if there are sections of vascularization. An
irregularly vascularized bed usually heals more rapidly in those
areas where vessels are present and more slowly in the avascular
areas. This allows for selective suture removal when suture
loosening occurs or to correct postoperative astigmatism. Similarly, more sutures may be required if the host bed is irregular in
thickness, such as when dealing with corneal melting or marginal
degeneration. Double running continuous sutures are used
when the host bed is uniformly thick and avascular, and the
wound is expected to heal evenly, such as in keratoconus or
bullous keratopathy. The major argument against the use of a
continuous suture is that if a single loop of the suture should
break, tear through, or loosen, it is not unusual to nd that several adjacent loops also loosen, leading to signicant wound
irregularity and astigmatism, or even wound dehiscence that
requires surgical repair.
Whatever the choice of suture technique opted for, sutures
should be passed deeply into the stroma at an equivalent depth
for both donor and recipient to allow accurate anteriorposterior
donorrecipient apposition. Deep stromal bites are taken at
~8090% depth, with care taken not to penetrate Descemets
layer, which would contribute to wound leak at the end of the
operation.50 Extra care should be taken to prevent under- or
overriding of the graft.
CONCOMITANT SURGICAL PROCEDURES

Modern anterior segment surgery attempts to repair as many
concomitant problems as are safely possible at the time of penetrating keratoplasty, such as problems related to the iris, lens,
vitreous, and retina.
LENS
If it is obvious that there is sufcient lens opacity to interfere
with vision, the lens should be extracted at the time of keratoplasty. The lens may also be removed if it does not yet interfere
with vision but progressive lens change has already been
documented, or if there is some degree of cataract and the angle
is very narrow.
In the presence of early corneal decompensation and cataract,
the decision has to be made to proceed with cataract extraction
alone, keratoplasty alone, or a combined procedure. If there is evidence of persistent corneal edema or edema that is worse in the
morning, the corneal pachymetry is greater than 0.600.65 mm,
and specular microscopy reveals low endothelial cell counts, then
keratoplasty is indicated. In the presence of a visually signicant
cataract, a combined procedure would be able to address both
issues. The prognosis of a properly performed triple procedure
(penetrating keratoplasty, extracapsular cataract extraction, and
implantation of an IOL) is as good as that for straightforward
penetrating keratoplasty.5254
Open-sky extracapsular cataract extraction is the procedure
of choice in these combined operations. Preoperative pupil
dilation and lowering of the IOP with intravenous mannitol and
Honan balloon application over the globe would help in minimizing the risk of inadvertent prolapse of the intraocular contents
and suprachoroidal hemorrhage.
Following excision of the diseased cornea, a capsulotomy is
performed and the nucleus is expressed. Removal of the remnant
cortical material should be done carefully, as there is a tendency
for the vitreous and capsule to bulge forward through the dilated
pupil. A posterior chamber IOL is then implanted into the bag
and the pupil miosed with acetylcholine (Miochol). The anterior
chamber is deepened with viscoelastic substance, and the rest of
the keratoplasty procedure is performed in the usual manner.
Surgeons should be aware of the increased risks associated with
the extra surgical manipulation in the open globe setting. Alternatively, phacoemulsication of the cataract may be performed
prior to recipient corneal trephination, to provide a more controlled removal of the cataract in a closed system.
IOL EXCHANGE
If an IOL is already present, the surgeon has to determine if this
lens is to be removed or retained. It is generally agreed that closedloop anterior chamber lenses should be removed and replaced.
The optic is rst cut from the haptics and removed, and the
remaining haptics are slowly rotated out of the cocoon that is
formed around it. At times, the residual haptics are tethered so
rmly that it may be safer to leave it in place, rather than risk
injury to the iris and angle. These remnant haptics do not pose
any problems to the eye or impair subsequent implantation of a
secondary IOL. The Kelman type three-footed anterior chamber
IOL has proven to be highly successful as a secondary IOL in
these situations.
In the situation where there is an intact posterior capsule,
many surgeons will prefer to use a posterior chamber IOL.55,56
In the absence of adequate capsular support, posterior chamber
IOLs may still be implanted by transscleral sulcus xation, or
by suturing the IOL to the posterior surface of the iris.5759
Posterior chamber IOLs are particularly useful in eyes with
compromised ltration angles as well as eyes that have insufcient iris tissue to permit safe implantation of an anterior
chamber IOL. The theoretical advantages of posterior chamber
IOLs compared with anterior chamber IOLs include providing
greater distance between the implant and the endothelium and
avoiding direct contact with the ltration angle. However, the
visual acuity, central corneal thickness, and the incidence of
glaucoma appear comparable in anterior chamber and posterior
chamber IOL lens placement.59 To date, there is no solid evidence
to suggest that posterior chamber IOLs give a better result.60
For transscleral sulcus xation of a posterior chamber IOL,
an IOL with a large optic (at least 6.5 mm) is preferred. Haptics
with eyelets or grooves would help in securing the haptics with
sutures. Prior to implantation, a core vitrectomy is performed to
prevent vitreous adherence to the haptics or xation sutures. A
nonabsorbable double-armed suture (100 Prolene) is tied to
each haptic, and each arm is then passed behind the iris through
the ciliary sulcus and sclera and tied externally, thus securing
each haptic in the ciliary sulcus. Iris-xated posterior chamber
IOLs come with four positioning holes on the optic, to allow a
double-armed 100 Prolene suture to be passed through and tied
onto the iris. Posterior chamber sutured IOLs require more
surgical manipulation and suturing, vitreous manipulation, and
prolonged operating time. As such the incidence of cystoid
macular edema and retinal detachment is signicantly higher.61,62
For iris-xated IOLs, the increased proximity of the lens to the
iris also leads to potential complications of this procedure, such
as peripheral anterior synechiae, iris chang and pigment
dispersion.63,64
In view of the potential complications that may arise from
exchanging a preexisting IOL, there is an increasing tendency to
leave well-tolerated lenses alone unless the indication for lens
removal and exchange is clear.65 The surgeon should rst determine if the corneal decompensation was a result of the IOL or
was contributed by its presence, and must always consider the
possibility that the manipulation required to remove certain
lenses may cause signicant trauma to the eye.
CHAPTER 60
All knots should be buried at the end of the procedure. It is

particularly important that the donor and recipient edges are
aligned properly and that neither an override nor an underride
exists. Poor apposition can interfere with reepithelialization of
the graft and may lead to postoperative astigmatism. The use of
an intraoperative keratoscope would aid in distributing the tension of the graft by adjusting the tightness of sutures, thereby
reducing postoperative astigmatism.
At the end of the procedure, the anterior chamber should be
reformed with viscoelastic substance or saline solution. Some
viscoelastic substance left in the anterior chamber at the end of
surgery does not pose signicant problems. Although the risk of
elevated IOP in the immediate postoperative period may be
increased, but such elevations are temporary and easily controlled
with medications, such as acetazolamide or timolol.51 Saline is
the safest of the solutions used to reform the anterior chamber, and
the ocular pressure may be easily adjusted. The wound is then
checked for water-tightness with a dry sponge. The surgery is completed with subconjunctival injections of antibiotics and steroids.
ANTERIOR SEGMENT AND IRIS

RECONSTRUCTION
Preexisting peripheral anterior synechiae may be lyzed and the
anterior chamber deepened by sweeping of the iris synechiae
with blunt instruments. Viscoelastic agents may also be used
to aid in the separation. Opening the angles allows better access
of aqueous to the trabecular meshwork and may reduce
819

FIGURE 60.1. (a) Aphakia with a large
coloboma. (b) Repair of the coloboma permits
insertion of a stable anterior chamber IOL.
postoperative glaucoma. Although these tissues may remain

anatomically separated at the end of surgery, how much of the
occluded angle remains functional in the long run remains
uncertain.
Iris defects arising from previous surgery, trauma or hereditary disease may be surgically repaired using nonabsorbable
sutures. Closing large iris defects and tightening the iris help to
reduce glare, and prevent progressive synechiae formation and
angle closure.66 In addition, one of the primary indications for
repairing an iris coloboma is to provide a stable platform on
which to situate an anterior chamber IOL (Fig. 60.1). Prolene
sutures are better suited for this purpose compared to nylon
sutures, because unlike nylon sutures, they do not degrade over
time. Iris and pupil reconstruction may be performed with a
100 or 110 Prolene suture with a round-bodied needle.
SECTION 6
POSTOPERATIVE MANAGEMENT
820
The postoperative care of corneal transplant patients is as

important as the surgery itself. During the postoperative period,
attention should be focused on the clarity of the graft, integrity
of the wound and sutures, status of the ocular surface, amount
of anterior chamber activity, IOP, and the presence of signs of
infection or rejection.
Most surgeons prescribe both steroids and antibiotics in the
immediate postoperative period. Topical steroids help to reduce
postoperative inflammation. Common steroids that are used
include prednisolone acetate 1% or dexamethasone 0.1% eyedrops. The postoperative regimen for steroids varies among surgeons. Immediately following surgery, they may be used every
23 h, and this is gradually tapered as the inflammation subsides.
Eyes that have undergone more complicated surgery would
require more intensive steroids initially to hasten resolution of
the inflammation. In general topical steroids should be used for
at least the rst 12 months after surgery, to reduce any risk of
rejection during this initial period. Some surgeons prefer to
maintain patients on weak steroids once or twice a day indenitely to reduce the risk of graft rejection. Because the longterm prognosis of grafts performed for corneal edema is worse
compared to other conditions like keratoconus, many surgeons
keep these patients on long-term steroids. Patients on steroids
should be monitored for any rise in IOP and the development of
cataract.
A broad-spectrum antibiotic is usually prescribed during the
initial period following keratoplasty. Some surgeons discontinue
antibiotics after several days or weeks. Additional medications
that may be required include the use of lubricating agents to
help protect the graft and enhance epithelial healing.
Special attention should be paid to the state of the intraocular
tension postoperatively, especially in all patients receiving
steroids. Elevated IOP frequently occurs in the immediate postoperative period following keratoplasty and may be attributed to
FIGURE 60.2. Iridocorneal synechiae (arrow). Signicant angle

closure may progress from this site.
ocular inflammation, retained viscoelastic substance, or presence of preexisting glaucoma. Any preexisting glaucoma may be
aggravated by keratoplasty, and mechanical problems such as
pupillary block or angle closure from synechiae or inflammation
must be specically searched for. Progressive zippering of the
angle may occur in aphakic patients (Fig. 60.2). Elevated IOP
should be treated with topical or systemic antiglaucoma
medications.
IMMUNOSUPPRESSION FOR HIGH-RISK

GRAFTS
Systemic immunosuppression should be considered for high-risk
grafts, particularly for patients that have had multiple repeat
grafts. Prior to starting the patient on systemic immunosuppression, it is important to exclude diseases where systemic
immunosuppression is contraindicated, such as hepatitis B, previous tuberculosis, signicant renal impairment, and others.
In the absence of any contraindication, oral corticosteroids
(e.g., prednisolone) may be started at an initial dose of 1 mg kg1
day1 and tapered over 3 weeks. Corticosteroid is used
concurrently with agents to protect the gastric mucosa. The role
of systemic cyclosporin (CSA) for high-risk grafts has been the
subject of controversy. Nonetheless, CSA is generally accepted
as the commonest form of systemic immunosuppression for
corneal grafting because it is more effective than oral corticosteroids. Hill, one of the proponents of the use of systemic CSA,
noted a signicantly improved rate of graft survival when it was
added to a regimen of topical or systemic steroids.67,68 However,
a more recent report suggested that the benet of CSA over
conventional therapy was moderate and was not statistically
signicant.69
COMPLICATIONS
INTRAOPERATIVE COMPLICATIONS
The complications that may occur during the procedure itself
are primarily related to instrumentation, to technical factors
such as irregular trephining of the donor or the host, and to
inadvertent surgical trauma to ocular tissues. Most of these
complications may be avoided if meticulous attention is paid to
the technical details and handling of the ocular tissues.
Injury to the iris or lens may occur during corneal trephination
or excision of the corneal button. Anterior chamber hemorrhage
occurs if the iris is inadvertently traumatized. Such bleeding often
stops spontaneously or when the IOP is returned following
closure of the globe.
The most dreaded and devastating of all complications is
expulsive suprachoroidal hemorrhage. The risk of this complication is higher in elderly patients, generalized atherosclerosis,
high myopia, glaucoma, and any sudden rise in IOP in the context of an open globe, such as sudden coughing at the time of
surgery. Measures to minimize the occurrence of this complication include proper positioning of the patient, adequate anesthesia and reducing the IOP prior to surgery with medication
(e.g., mannitol) and ocular massage.
POSTOPERATIVE COMPLICATIONS
Postoperative complications can be divided into those that occur
in the immediate postoperative period, and those that occur
weeks or months after surgery (see Table 60.4). Complications
occurring in the immediate postoperative period include wound
leakage, delayed epithelial healing, infection, suture-related
problems, increased IOP and primary graft failure. Complications that occur several weeks or longer after surgery include
wound dehiscence, glaucoma, cataract, suture-related problems,
high astigmatism, graft rejection, graft failure, and recurrence of

the original host pathology.
EARLY POSTOPERATIVE COMPLICATIONS

WOUND LEAKAGE
At the conclusion of surgery it is important to ensure proper
wound apposition at the grafthost junction and exclude any
possibility of leak. Wound leakage leading to shallowing of the
anterior chamber may become apparent in the immediate postoperative period and lead to anterior synechiae at the grafthost
junction with a subsequent risk of graft rejection. The most
common cause of wound leak is a broken, loose, or misplaced
suture.51,72 Full thickness sutures are more likely to leak and,
thus, such suturing is best avoided.51 If the anterior chamber is
formed and the leak is minimal, the use of a bandage lens may
be attempted. Resuturing may be necessary if excessive leakage
leads to the anterior chamber becoming flat.
PERSISTENT EPITHELIAL DEFECT

In a corneal graft, epithelialization is expected to be complete at
46 days postoperatively.73 Persistent epithelial defects may lead
to anterior stromal opacication, stromal melting, secondary
infection, suture loosening, vascularization, and reduced graft
survival. Careful attention to preservation of the donor epithelium at the time of graft surgery helps to maintain the donor
epithelium in the early postoperative period.
If the epithelium is slow to reestablish itself, every effort
should be made to diminish adnexal inflammation, to provide
adequate moisture, to control infection, and particularly to prevent the rubbing of vulnerable epithelium by irregular lid surfaces.
A soft contact lens or even tarsorrhaphy may be used. Finally,
one must also keep in mind the possibility that various topical
medications and preservatives (especially glaucoma medications)
may also contribute to a delayed epithelial healing and toxic or
whorl epitheliopathy (Fig. 60.3).
INFECTION
Endophthalmitis is a potentially devastating complication. The
incidence of endophthalmitis after penetrating keratoplasty
ranges from 0.2% to 2.0%.7376 Risk factors include concomitant
CHAPTER 60
Adverse effects of CSA include hypertension and nephrotoxicity, among others. CSA is initiated at a dose of 45 mg kg1 day1
and is subsequently adjusted to maintain a trough level of
200250 mg/L. Several authors have also reported the use of
topical CSA for high-risk grafts with variable success rates.70,71
Although its use is still not widely accepted, it may be considered
in selected high-risk cases where systemic immunosuppression
is contraindicated. Aside from mild epitheliopathy, no other
signicant ocular side effects have been reported.70
TABLE 60.4. Postoperative Complications of Keratoplasty

Early
Late
Wound leaks/suture problems
Wound dehiscence
Delayed epithelial healing
Cataract
Infection
Glaucoma
Elevated IOP
Suture abscess and infection
Hypotony
Astigmatism
Primary graft failure
Graft vascularization
Cystoid macular edema
Graft rejection
Late endothelial failure
Recurrence of disease
FIGURE 60.3. Supercial punctate keratitis. During the weeks after

keratoplasty, the epithelium is fragile and especially subject to
keratopathy from topical medications.
821

vitrectomy, aphakia, previous surgery or inflammation, and
steroid use. Possible sources of infection include contaminated
donor tissue, or contamination from the adjacent eyelids and
conjunctiva. Early infectious keratitis is rare, and may be related
to loose sutures that attract mucus debris and microbes. Early
recognition, removal of any predisposing factor (e.g., removal of
a loose suture) and treatment with intensive antibiotics is
critical for a good outcome.
on morphometric endothelial cell analysis cell counts. Grafts

can remain clear with cell counts of as low as 300500 cells/mm.51
There is an estimated mean cell loss of 612% following cataract extraction with IOL implantation, and the graft failure rate
ranges from 0% to 20% following surgery.51 The risk of a rejection episode is also minimized with the stepping-up of topical
steroid application in the perioperative period. These factors
should be taken into account when contemplating cataract
surgery after keratoplasty.
ELEVATED IOP
Elevated IOP in the early postoperative period frequently occurs
following keratoplasty. The absence of corneal edema does not
exclude the presence of high IOP. In the early or immediate
postoperative period, raised IOP may be related to residual viscoelastic in the anterior chamber (thus, the use of a dispersive
rather than a cohesive viscoelastic is preferred), intraocular inflammation and crowding of the anterior chamber angle. It should
be treated with topical or systemic IOP-lowering medications.
An elevated IOP that occurs later may be the result of prolonged
use of topical steroids, chronic synechial angle closure, and
exacerbation of preexisting glaucoma.
PRIMARY GRAFT FAILURE

Primary donor failure is dened as irreversible edema of the
graft occurring in the immediate postoperative period, which
presents as a cloudy graft that fails to clear. This condition is
relatively uncommon, and has a reported incidence of <5%.51
Other more common causes of graft edema include hypotony,
severe inflammation, and large epithelial defects. The principal
causes of primary donor failure are substandard donor tissue,
poor preservation, and surgical trauma. Once diagnosed, the
failed graft should be immediately exchanged for a new one to
facilitate rapid visual recovery.
LATE POSTOPERATIVE COMPLICATIONS
SECTION 6
GLAUCOMA
822
Glaucoma is frequently associated with graft failure after

keratoplasty.51,72,77 The incidence varies and factors such as
preexisting glaucoma, peripheral anterior synechiae, aphakia,
pseudophakia, and long-term use of topical steroids have been
linked to its occurrence. The use of a 0.5 mm oversized donor
button reduces the incidence of glaucoma postoperatively.51,72
Eyes with preexisting glaucoma should therefore be adequately
treated prior to surgery.
CATARACT
A preexisting cataract may progress rapidly after penetrating
keratoplasty, and cataract removal at the time of grafting (a
triple procedure) is usually warranted. Calculation of IOL power
in combined surgery poses a problem with keratometry, which
is often not possible in diseased corneas, and the refractive
result would be related to donor tissue curvature and other
surgical variables. Retrospective studies on average corneal curvatures after keratoplasty suggest an average value of 44 D, and
this average value may be used to calculate IOL power in triple
procedures.
If a cataract develops after keratoplasty, care should be taken
to evaluate the riskbenet ratio of cataract removal, with the
main risks being graft endothelial decompensation and
triggering a rejection episode. With regards to donor endothelial
status, the decision to perform cataract surgery should be based
SUTURE-RELATED COMPLICATIONS AND

INFECTION
Premature loosening of a suture, whether interrupted or continuous, may result in slight wound slippage and displacement of
the graft. Exposed sutures cause irritation and may incite
inflammation and vascularization around the suture. Loose or
exposed suture ends may attract mucus debris and microbes,
and predispose to suture-related infections.78,79
Other predisposing factors for graft-related infection include
persistent epithelial defects, keratoconjunctivitis sicca, graft
failure, previous herpetic disease, soft contact lens wear, and
chronic steroid use. Bacterial, viral, and fungal pathogens have
all been implicated in these infections.76,80 Infectious crystalline
keratopathy has been reported to occur almost exclusively in
corneal graft patients.81 This condition manifests as an indolent,
slowly progressive deep inltrate with crystalline deposits
emanating from a dense central cores. The organisms that have
been most commonly implicated in this infection are the Grampositive bacteria, most notably nutrient-variable Streptococcus
and Streptococcus pneumoniae.82,83
POSTOPERATIVE ASTIGMATISM
Graft astigmatism is an important cause of poor visual acuity
despite the presence of a clear graft. Considerable degrees of
astigmatism may develop after keratoplasty. An overall mean of
4.55.0 D has been reported after keratoplasty.72 Astigmatism
may be minimized with proper donor and recipient trephination,
careful avoidance of signicant grafthost misalignment, and a
consistent and even suturing technique. The use of intraoperative keratoscopy aims to reduce postoperative astigmatism, but
the results are sometimes unpredictable.84 Selective removal of
individual interrupted sutures85,86 and postoperative suture
adjustment of continuous sutures87,88 are the main methods of
correcting postoperative astigmatism.
GRAFT REJECTION
It is recognized that the relative immune privileged status of
the cornea accounts for the extraordinarily high success rates of
transplant procedures, which reach over 90% when performed
on an avascular noninflamed recipient bed.10 This characteristic
may be attributed to several factors including lack of corneal
vascularity, absence of corneal lymphatics, few antigen presenting cells, ocular expression of immunomodulatory factors and
neuropeptides that inhibit T cell and complement activation,
low expression of major histocompatibility complex (MHC)
antigen and a donor-specic anterior chamber-associated immune
deviation that develops after transplantation.89,90 Despite all
these, graft rejections do occur, especially in high-risk situations, and allograft rejection remains the most common cause
of late graft failure. Loss of immune privilege occurs in many
corneal disorders. Based on a prospective large-scale study, the
reported cumulative probability of graft rejection is ~21% in
10 years.11 In high-risk grafts, rejection rates can increase
ously failed allografts due to graft rejection.92,93 Corneal vascularization is the most common factor associated with graft
rejection.92,93 Immune-related graft failure rates of 2550% in
severely vascularized corneas were signicantly higher when
compared with rates of 010% in avascular corneas at 1 year
after keratoplasty.92 Patients with two repeat grafts had a
rejection rate of 40% compared to only 8% for patients without
previous grafts.92 Other predisposing factors include the presence of peripheral anterior synechiae at the grafthost junction,
young or pediatric recipients, glaucoma, previous anterior
segment surgery, large grafts, and concurrent inflammation.92,93
beyond 50%. Furthermore, ~25% of corneal transplant recipients

will experience at least one rejection episode.72
Allograft rejection is genetically determined activation and
modulation of the host immune system in response to an antigenically disparate donor. Immunologic disparity in humans
appears to be dened primarily by a group of genes located on
the sixth chromosome, the MHC. There are three classes of
histocompatibility antigens that are glycopeptides incorporated
into the cell membrane. Class I antigens are capable of inciting
strong immunologic reactions in the immunocompetent host
and are distributed on the surface of all nucleated cells. The
class I antigens are HLA-A, HLA-B, and HLA-C. HLA-D and
HLA-DO are known as class II antigens and are also capable of
eliciting a strong immunologic reaction in the host.90,91
The primary cellular constituents of the immune system are
B lymphocytes, T lymphocytes, and macrophages. Macrophages
are monocytes that occasionally play an effector role in cellmediated immunity but typically serve an intermediary role in
processing antigens. T lymphocytes are primarily responsible
for modulating and effecting cell-mediated immunity, whereas
B lymphocytes act as the primary cell population supporting
humoral immunity through the production of antibodies.
Although humoral immunity may play some role in transplant
reactions, it appears that cell-mediated immunity is more
prominent in allograft rejection.
The initial phase of corneal allograft rejection involves the
detection of foreign antigen by the host immune system. Antigenpresenting macrophages activate T lymphocytes to secrete
mediators that recruit and modulate the activities of other
macrophages and trigger lymphocytic proliferation in the
draining lymph nodes. When activated, cytotoxic T lymphocytes
return to the eye to destroy donor corneal cells. Although matching of MHC antigens between host and donor tissues has
reduced the rate of allograft rejection for perfused organ transplantation, the results have been less denitive with regards to
corneal transplantation. The results of the Collaborative Corneal
Transplant Study (CCTS) failed to demonstrate any statistically
signicant effect in altering the rate of corneal allograft rejection
through HLA cross-matching.9
Epithelial rejection is characterized by the appearance of a linear

opacity that migrates across the surface of the graft from one
edge to the other over the course of several days (Fig. 60.4). This
elevated line stains with fluorescein or rose bengal, and
represents a zone of donor epithelial replacement by recipient
epithelium.51,90,94 Once the recipient epithelial cells have been
completely replaced by donor epithelium, epithelial rejection
does not occur, and hence epithelial rejection is rare after the
rst year following graft surgery. Epithelial rejection accounts
for ~10% of all graft rejections.51 Although epithelial rejection
is usually asymptomatic and self-limited, treatment in the form
of topical steroids is required to prevent progression to other
forms of rejection.
RISK FACTORS
STROMAL REJECTION
Preoperatively, it is important to recognize risk factors that may

lead to graft rejection in the postoperative management of these
patients, and to educate transplant patients on early symptoms
of rejection. Denitions of high-risk grafts vary, but most
denitions include the presence of two or more quadrants of
corneal vascularization, or repeated corneal grafting for previ-
Isolated stromal rejection is uncommon51,90,94 and is often associated with either adjacent epithelial or adjacent endothelial
rejection. Stromal edema and inltration is present and often
adjacent to a vascularized site or an area of peripheral synechiae;
it may progress to a full-blown combined stromal and endothelial rejection if not aggressively treated.
CLINICAL FEATURES
The diagnosis of allograft rejection can only be made if the
donor corneal has remained clear for the rst 1014 days after
surgery. Patients usually complain of eye redness, photophobia,
blurring of vision and pain. Clinical signs include conjunctival
injection, graft edema, anterior chamber cells and flares and,
occasionally, elevated IOP, among others.90,93,94 Three types of
immunologic rejection have been identied: epithelial, stromal,
and endothelial. Most rejection episodes do not fall into discrete
categories but represent various combinations of all three.
FIGURE 60.4. (a) Epithelial rejection linear

distribution. (b) Epithelial rejection epidemic
keratoconjunctivitis-like deposits.
CHAPTER 60
EPITHELIAL REJECTION
823
FIGURE 60.5. (a) Graft rejection vascularization around the graft. (b) Endothelial rejection: (1) edema at the grafthost junction (curved white
arrow); (2) keratic precipitates (white arrow); and (3) Khodadousts line (this is a late manifestation) (black arrowheads). (c) Diffuse graft edema.
This rejection manifestation has the poorest prognosis for improvement.
ENDOTHELIAL REJECTION
Endothelial rejection is the most common and signicant type
of immunologic rejection, often leading to graft failure and
edema from endothelial damage.51,93,94 Patients with this form
of rejection may be more symptomatic because pain, redness,
and blurring of vision may be more marked. There are two
major manifestations of endothelial graft rejection (Fig. 60.5).
The rst manifestation is that of inflammation at the junction
of the graft and the host, with injection at the limbus and
affected area, pigmented keratic precipitates, endothelial haze,
localized edema, and mild to moderate anterior chamber flare
and cells. A later manifestation is a linear arrangement of the
keratic precipitates, commonly referred to as a Khodadoust line,
which may progress from the affected area of the grafthost
junction across the endothelium toward the other side of the
graft.72,95 Endothelial rejection can progress within days if left
untreated, and the localized corneal edema may eventually
become more diffuse and involve the entire graft. The presence
of generalized corneal edema with keratic precipitates, flare, and
cells represents the most serious of all the manifestations of
graft rejection and carries the poorest prognosis.
SECTION 6
SUBEPITHELIAL INFILTRATES
824
Some authors consider anterior nummular stromal inltrates as

a fourth type of rejection. These inltrates that closely resemble
those seen in nummular epidemic keratoconjunctivitis (but are
conned to the donor cornea) were originally described by
Krachmer and are referred to as Krachmers spots.51,72,90 They
are small, round, patchy inltrates measuring ~0.20.5 mm in
diameter that are diffusely located beneath Bowmans layer. The
frequency of subepithelial inltrates was noted to be 15%, which
occurred an average of 10 months after surgery. Subepithelial
inltrates usually clear with topical corticosteroid therapy but
may leave faint scars.
Allograft rejection can be reversed if diagnosed and treated
early. Graft failure is generally considered irreversible if the
signs of endothelial rejection do not improve within 34 weeks
of maximum therapy.72 The probability of reversing endothelial
rejection varies from 50% to 91%, depending on the severity of
the rejection, risk factors, and onset of treatment. The use of systemic immunosuppression helps to reduce the risk of rejection
in repeat grafts.72
GRAFT FAILURE
Graft failure refers to any graft that does not retain transparency
for adequate vision. Graft rejection is a major cause of graft
failure that is immunologically mediated, and is covered in the
previous section. Late endothelial failure occurs in a graft that
has been clear for many years and eventually fails without any
identiable cause. A decrease in endothelial cell counts beyond
what is necessary to maintain deturgescence is the cause of this
gradual corneal graft decompensation. The common denominator of graft failure is endothelial decompensation, which
occurs as a natural attrition of transplanted endothelial cells
over a period of time.
Recurrence of the underlying corneal disease may also contribute to corneal opacication. The recurrence of dystrophies in
the transplanted graft has been well documented in conditions
such as macular, lattice, granular, Reis Buckler, and posterior
polymorphous dystrophies.96
GRAFT SURVIVAL AND PROGNOSIS

The underlying corneal disease is a major factor in determining
penetrating keratoplasty success. Corneas that are avascular,
free of active inflammation, and have intact innervation are
associated with the best prognosis. These include eyes with
keratoconus, scars, and granular dystrophy. Conversely, corneal
vascularization, ocular surface inflammation, and impaired corneal innervation are poor prognostic factors.11,20,97,98 Examples
of such conditions include eyes with ocular surface disease
arising from StevensJohnson syndrome and chemical injury,
active corneal infection, and neuropathic keratopathy. The
presence of deep corneal vascularization has also been shown to
increase the risk of graft rejection. Patients with severe dry eye
are also more prone to problems post surgery.
Glaucoma has been shown to be one of the most important
limiting factors for graft success.11,20,97,98 As such, optimization
of the IOP control is of paramount importance before and after
penetrating keratoplasty. Uveitis is also associated with an
increased risk of graft failure. Where possible, surgery should be
deferred till the intraocular inflammation has completely resolved.
Previous anterior segment surgery confers a twofold increased
risk of graft failure.92
Previous keratoplasty is another important factor that may
affect graft success. The rate of graft failure secondary to allograft
rejection can be as high as 40% in patients with two or more
previous grafts.93 Larger grafts and more eccentrically placed
grafts are more likely to be associated with an increased risk of
graft rejection, because of the greater antigenic load, as well as
the closer proximity to the limbal vasculature. In addition,
larger grafts are associated with a greater risk of peripheral
anterior synechiae and glaucoma.
Risk factors for graft failure and graft survival rates are important considerations when evaluating patients for corneal grafting. Graft survival rates vary widely between reports due to
variations between the inherent characteristics of the study populations, study design, length of follow-up, and statistical methods
used.11,20,97101 The overall graft survival rates range from 76%
to 91% at 1 year, 45.5% to 72% at 5 years, and 69.8 to 79% at
10 years. Using multivariate analysis, Sit et al showed that the
risk factors that had signicant effects on graft survival were

preoperative diagnosis (RR = 5.79), postoperative graft neovascularization (RR = 2.24), preoperative peripheral anterior
synechia detected at the time of surgery (RR = 1.98), male gender
(RR = 1.75), occurrence of at least one postoperative rejection
episode (RR = 1.75), and age of the recipient (RR = 1.014).99
The widespread acceptance of cataract surgery and IOL
implantation, as well as advances in antirejection immunosuppressive therapy, have contributed to a steady rise in the number
of patients undergoing corneal grafting for bullous keratopathy
and previously failed grafts. Improvements in surgical instrumentation and graft immunology have signicantly improved
the survival of corneal grafts. These successes have led to a
greater number of diseases that can be effectively treated with
corneal transplantation.
1. Boruchoff SA, Thoft RA: Keratoplasty:

lamellar and penetrating. In: Smolin G,
Thoft RA, eds. The cornea. Boston, MA:
2. von Hippel A: On transplantation of the
cornea. Berichte Ophthalmol Gesellschaft
Herdelberg 1886; 18:54.
3. United Network for Organ Sharing: 2001
annual report. Richmond, VA: United
Network for Organ Sharing; 2001.
4. Eye Bank Association of America: 2002 eye
banking statistical report. Washington, DC:
Eye Bank Association of America; 2002.
5. Aiken-ONeill P, Mannis MJ: Summary of
comeal transplant activity Eye Bank
Association of America. Cornea 2002;
21:13.
6. Barron BA: Penetrating keratoplasty. In:
Kaufman HE, Barron BA, McDonald MB,
eds. The cornea. 2nd edn. St Louis, MO:
Butterworth-Heinemann; 1988: 805846.
7. Brady SE, Rapuano CJ, Arentsen JJ, et al:
Clinical indications for and procedures
associated with penetrating keratoplasty,
19831988. Am J Ophthalmol 1989;
108:118122.
8. Morris RJ, Bates AK: Changing indications
for keratoplasty. Eye 1989; 3(Pt 4):455459.
9. The Collaborative Corneal Transplantation
Studies Research Group: The collaborative
corneal transplantation studies (CCTS):
effectiveness of histocompatibility
matching in high-risk corneal
transplantation. Arch Ophthalmol 1992;
110:13921403.
10. Dana MR, Qian Y, Hamrah P: Twenty-veyear panorama of corneal immunology:
transplant rejection. Cornea 2000;
19:625643.
11. Ing JJ, Ing HH, Nelson LR, et al: Ten-year
postoperative results of penetrating
105:18551865.
12. Inoue K, Amano S, Oshika T, et al: A 10-year
review of penetrating keratoplasty. Jpn J
Ophthalmol 2000; 44:139145.
13. Ehrlich MI, Phinney RB, Mondino BJ,
Pettit TH: Techniques of lamellar
keratoplasty. Int Ophthalmol Clin 1988;
28:2429.
14. Melles GR, Eggink FA, Lander F, et al:
A surgical technique for posterior lamellar
15. Melles GR: Posterior lamellar keratoplasty.

Arch Soc Esp Oftalmol 2002; 77:175176.
16. Melles GR, Lander F, Nieuwendaal C:
Sutureless, posterior lamellar keratoplasty:
a case report of a modied technique.
Cornea 2002; 21:325327.
17. Watson SL, Ramsay A, Dart JK, et al:
111:16761682.
18. Alio JL, Shah S, Barraquer C, et al: New
techniques in lamellar keratoplasty. Curr
Opin Ophthalmol 2002; 13:224229.
19. Frucht-Pery J, Shtibel H, Solomon A, et al:
Thirty years of penetrating keratoplasty in
Israel. Cornea 1997; 16:1620.
20. Yamagami S, Suzuki Y, Tsuru T: Risk
factors for graft failure in penetrating
keratoplasty. Acta Ophthalmol Scan 1996;
74:584588.
21. Chen WL, Hu FR, Wang IJ: Changing
indications for penetrating keratoplasty in
Taiwan from 1987 to 1999. Cornea 2001;
20:141144.
22. Dandona L, Ragu K, Janarthanan M, et al:
Indications for penetrating keratoplasty in
India. Indian J Ophthalmol 1997; 45:163168.
23. Flowers CW, CK, McLeod SD, et al.
Changing indications for penetrating
keratoplasty, 19891993. Cornea 1995;
14:583588.
24. Dobbins KR, Price FW Jr, Whitson WE:
Trends in the indications for penetrating
keratoplasty in the midwestern United
States. Cornea 2000; 19:813816.
25. Ramsay AS, LW, Mohammed A: Changing
indications for penetrating keratoplasty in
the west of Scotland from 1970 to 1995.
Eye 1997; 11:357360.
26. Legeais JM, Parc C, dHermies F, et al:
Nineteen years of penetrating keratoplasty
in the Hotel-Dieu Hospital in Paris. Cornea
2001; 20:603606.
27. Cosar CB, Sridhar MS, Cohen EJ, et al:
Indications for penetrating keratoplasty and
associated procedures, 19962000. Cornea
2002; 21:148151.
28. Buxton JN, Lash RS: Results of penetrating
keratoplasty in the iridocorneal endothelial
98:297301.
29. Soong HK, Schwartz AE, Meyer RF, Sugar A:
Penetrating keratoplasty for corneal
scarring due to herpes zoster ophthalmicus.
30. Varley GA, Macsai MS, Krachmer JH: The

results of penetrating keratoplasty for
pellucid marginal corneal degeneration. Am
J Ophthalmol 1990; 110:149152.
31. ODay DM: Donor selection. In: Brightbill FS,
ed. Corneal surgery, theory, technique and
tissue. St Louis, MO: Mosby-Year Book;
1993:549562.
32. Caron MJ, Wilson R: Review of the risk of
HIV infection through corneal
transplantation in the United States. J Am
Optom Assoc 1994; 65:173178.
33. Hogan RN, Brown P, Heck E, Cavanagh HD:
Risk of prion disease transmission from
ocular donor tissue transplantation. Cornea
1999; 18:211.
34. Lang CJ, Heckmann JG, Neundorfer B:
Creutzfeldt-Jakob disease via dural and
corneal transplants. J Neurol Sci 1998;
160:128139.
35. Beck RW, Gal RL, Mannis MJ, et al: Is
donor age an important determinant of
graft survival? Cornea 1999; 18:503510.
36. Chipman ML, Willett P, Basu PK, Wolf A:
Donor eyes. A comparison of
characteristics and outcomes for Eye Bank
and local tissue. Cornea 1989; 8:6266.
37. Sharif KW, Casey TA: Penetrating
keratoplasty for keratoconus: complications
and long-term success. Br J Ophthalmol
1991; 75:142146.
38. Palay DA, Kangas TA, Stulting RD, et al:
The effects of donor age on the outcome of
penetrating keratoplasty in adults.
Ophthalmology 1997; 104:15761579.
39. Inoue K, Amano S, Oshika T, Tsuru T: Risk
factors for corneal graft failure and
rejection in penetrating keratoplasty. Acta
40. Kaufman HE, Varnell ED, Kaufman S, et al:
K-Sol corneal preservation. Am J
Ophthalmol 1985; 100:299304.
41. McCarey BE, Kaufman HE: Improved
corneal storage. Invest Ophthalmol 1974;
13:165173.
42. Kaufman HE, Beuerman RW, Steinemann TL,
et al: Optisol corneal storage medium. Arch
Ophthalmol 1991; 109:864868.
43. Frueh BE, Bohnke M: Prospective,
randomized clinical evaluation of Optisol vs
organ culture corneal storage media. Arch
Ophthalmol 2000; 118:757760.
44. Ehlers H, Ehlers N, Hjortdal JO: Corneal
transplantation with donor tissue kept in
organ culture for 7 weeks. Acta Ophthalmol
Scand 1999; 77:277278.
CHAPTER 60
REFERENCES
825
SECTION 6
826
45. Andersen J, Ehlers N: Corneal

transplantation using 4-week banked donor
material. Long-term results. Acta
46. Capella JA, Kaufman HE, Robbins JE:
Preservation of viable corneal tissue. Arch
Ophthalmol 1965; 74:669673.
47. Pflugfelder SC, Roussel TJ, Denham D,
et al: Photogrammetric analysis of corneal
trephination. Arch Ophthalmol 1992;
110:11601166.
48. Alpar JJ: The use of Healon in corneal
transplant surgery with and without
intraocular lenses. Ophthalmic Surg 1984;
15:757760.
49. Musch DC, Meyer RF, Sugar A, Soong HK:
Endothelial rejection following penetrating
keratoplasty using Healon or BSS. Cornea
1990; 9:305311.
50. Eve FR, Troutman RC: Placement of
sutures used in corneal incisions. Am J
Ophthalmol 1976; 82:786789.
51. Wilson SE, Kaufman HE: Graft failure after
penetrating keratoplasty. Surv Ophthalmol
1990; 34:325356.
52. Buxton JN: The triple procedure: corneal
graft, intracapsular cataract extraction and
intraocular lens. Contact Intraocular Lens
Med J 1980; 6:409.
53. Lindstrom RL, Harris WS, Doughman DJ:
Combined penetrating keratoplasty,
extracapsular cataract extraction, and
posterior chamber lens implantation. J Am
Intraocul Implant Soc 1981; 7:130132.
54. Kramer SG: Penetrating keratoplasty
combined with extracapsular cataract
extraction. Am J Ophthalmol 1985;
100:129133.
55. Zaidman GW, Goldman S: A prospective
study on the implantation of anterior
chamber intraocular lenses during
keratoplasty for pseudophakic and aphakic
bullous keratopathy. Ophthalmology 1990;
97:757762.
56. Kornmehl EW, Steinert RF, Odrich MG,
Stevens JB: Penetrating keratoplasty for
pseudophakic bullous keratopathy
associated with closed-loop anterior
chamber intraocular lenses. Ophthalmology
1990; 97:407412; discussion 413414.
57. Lass JH, DeSantis DM, Reinhart WJ, et al:
Clinical and morphometric results of
penetrating keratoplasty with one-piece
anterior-chamber or suture-xated
posterior-chamber lenses in the absence of
lens capsule. Arch Ophthalmol 1990;
108:14271431.
58. Price FW Jr, Whitson WE: Visual results of
suture-xated posterior chamber lenses
during penetrating keratoplasty.
Ophthalmology 1989; 96:12341239;
59. Soong HK, Musch DC, Kowal V, et al:
Implantation of posterior chamber
intraocular lenses in the absence of lens
capsule during penetrating keratoplasty.
60. Davis RM, Best D, Gilbert GE: Comparison
of intraocular lens xation techniques
performed during penetrating keratoplasty.
Am J Ophthalmol 1991; 111:743749.
61. Soong HK, Meyer RF, Sugar A: Posterior
chamber IOL implantation during
keratoplasty for aphakic or pseudophakic
corneal edema. Cornea 1987; 6:306312.
62. Heidemann DG, Dunn SP: Transsclerally
sutured intraocular lenses in penetrating
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.

113:619625.
Chu MW, Font RL, Koch DD: Visual results
and complications following posterior irisxated posterior chamber lenses at
penetrating keratoplasty. Ophthalmic Surg
1992; 23:608613.
Schein OD, Kenyon KR, Steinert RF, et al: A
randomized trial of intraocular lens xation
techniques with penetrating keratoplasty.
Ophthalmology 1993; 100:14371443.
Kozarsky AM, Stopak S, Waring GO 3rd,
et al: Results of penetrating keratoplasty
for pseudophakic corneal edema with
retention of intraocular lens.
Cohen EJ, Kenyon KR, Dohlman CH:
Iridoplasty for prevention of postkeratoplasty angle closure and glaucoma.
Hill JC: The use of cyclosporine in high-risk
107:506510.
Hill JC: Systemic cyclosporine in high-risk
keratoplasty. Short- versus long-term
therapy. Ophthalmology 1994; 101:128133.
Poon AC, Forbes JE, Dart JK, et al:
Systemic cyclosporin A in high risk
penetrating keratoplasties: a case-control
study. Br J Ophthalmol 2001;
85:14641469.
Belin MW, Bouchard CS, Frantz S,
Chmielinska J: Topical cyclosporine in
high-risk corneal transplants.
Zhao JC, Jin XY: Local therapy of corneal
allograft rejection with cyclosporine. Am J
Ophthalmol 1995; 119:189194.
Hodkin MJ: Complications of penetrating
keratoplasty. In: Kaufman HE, Barron BA,
McDonald MB, eds. The cornea. 2nd edn.
St Louis, MO: Butterworth-Heinemann;
1998.
Aiello LP, Javitt JC, Canner JK: National
outcomes of penetrating keratoplasty.
Risks of endophthalmitis and retinal
detachment. Arch Ophthalmol 1993;
111:509513.
Leveille AS, McMullan FD, Cavanagh HD:
Endophthalmitis following penetrating
90:3839.
Guss RB, Koenig S, De La Pena W, et al:
Endophthalmitis after penetrating
95:651658.
Cameron JA, Antonios SR, Cotter JB,
Habash NR: Endophthalmitis from
contaminated donor corneas following
1991; 109:5459.
Irvine AR, Kaufman HE: Intraolar pressure
following penetrating keratoplasty. Am J
Ophthalmol 1969; 68:835844.
Sonavane A, Sharma S, Gangopadhyay N,
Bansal AK: Clinico-microbiological
correlation of suture-related graft infection
Ophthalmol 2003; 135:8991.
Leahey AB, Avery RL, Gottsch JD, et al:
Suture abscesses after penetrating
Varley GA, Meisler DM: Complications of
penetrating keratoplasty: graft infections.
Gorovoy MS, Stern GA, Hood CI, Allen C:
Intrastromal noninflammatory bacterial
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
colonization of a corneal graft. Arch

Ophthalmol 1983; 101:17491752.
Matoba AY, OBrien TP, Wilhelmus KR,
Jones DB: Infectious crystalline
keratopathy due to Streptococcus
pneumoniae. Possible association with
serotype. Ophthalmology 1994;
101:10001004.
Meisler DM, Langston RH, Naab TJ, et al:
Infectious crystalline keratopathy. Am J
Ophthalmol 1984; 97:337343.
Troutman RC, Swinger CA, Belmont S:
Selective positioning of the donor cornea in
penetrating keratoplasty for keratoconus:
postoperative astigmatism. Cornea 1984;
3:135139.
Burk LL, Waring GO 3rd, Harris DJ Jr:
Simultaneous and sequential selective
suture removal to reduce astigmatism after
penetrating keratoplasty. Refract Corneal
Surg 1990; 6:179187.
Burk LL, Waring GO 3rd, Radjee B,
Stulting RD: The effect of selective suture
removal on astigmatism following
1988; 19:849854.
Van Meter WS, Gussler JR, Soloman KD,
Wood TO: Postkeratoplasty astigmatism
control. Single continuous suture
adjustment versus selective interrupted
suture removal. Ophthalmology 1991;
98:177183.
Lin DT, Wilson SE, Reidy JJ, et al: An
adjustable single running suture technique
to reduce postkeratoplasty astigmatism.
A preliminary report. Ophthalmology 1990;
97:934938.
Streilein JW: Immunobiology and
immunopathology of corneal transplantation.
Chem Immunol 1999; 73:186206.
Streilein JW: New thoughts on the
immunology of corneal transplantation. Eye
2003; 17:943948.
Braude LS, Chandler JW: Corneal allograft
rejection. The role of the major
histocompatibility complex. Surv
Ophthalmol 1983; 27:290305.
Maguire MG, Stark WJ, Gottsch JD, et al:
Risk factors for corneal graft failure and
rejection in the collaborative corneal
transplantation studies. Collaborative
Corneal Transplantation Studies Research
Group. Ophthalmology 1994;
101:15361547.
Tham VM, Abbott RL: Corneal graft
rejection: recent updates. Int Ophthalmol
Clin 2002; 42:105113.
Khodadoust AA, Silverstein AM:
Transplantation and rejection of individual
cell layers of the cornea. Invest Ophthalmol
1969; 8:180195.
Khodadoust AA, Silverstein AM: Local graft
versus host reactions within the anterior
chamber of the eye: the formation of
corneal endothelial pocks. Invest
Ophthalmol 1975; 14:640647.
Boruchoff SA, Weiner MJ, Albert DM:
Recurrence of posterior polymorphous
corneal dystrophy after penetrating
109:323328.
Kuchle M, Cursiefen C, Nguyen NX, et al:
Risk factors for corneal allograft rejection:
intermediate results of a prospective
normal-risk keratoplasty study. Graefes
240:580584.
100. Williams KA, Muehlberg SM, Coster DJ:
Visual outcome after corneal
transplantation. Australian Corneal Graft
Registry. Transplant Proc 1992; 24:178.
101. Dandona L, Naduvilath TJ, Janarthanan M,

et al: Survival analysis and visual outcome
in a large series of corneal transplants in
India. Br J Ophthalmol 1997; 81:726731.
CHAPTER 60
98. Boisjoly HM, Tourigny R, Bazin R, et al:

Risk factors of corneal graft failure.
Ophthalmology 1993; 100:17281735.
99. Sit M, Weisbrod DJ, Naor J, Slomovic AR:
Corneal graft outcome study. Cornea 2001;
20:129133.
827
CHAPTER
61
Endothelial Keratoplasty
Mark A. Terry
Overview
Endothelial keratoplasty (EK) transplant surgery represents the
selective replacement of recipient diseased endothelium without
touching the recipient corneal surface. The surgery has evolved
from a large 9 mm scleral access incision with removal of
posterior stromal tissue (PLK), to a short 5 mm scleral incision
with removal of only recipient Descemets membrane (Descemets
stripping endothelial keratoplasty). EK has become technically
easier with the use of viscoelastic material during preparation of
the recipient and use of a microkeratome for preparation of the
donor. The current visual results with EK rival those of standard
full-thickness PK, but with a much faster visual rehabilitation. The
postoperative corneal topography after EK is far superior to that
of standard PK, with a return to the premorbid normal
topography. EK results in a refractive neutral corneal transplant,
avoiding the problems associated with high astigmatism or
abnormal refractive spherical equivalent that is often found after
PK. Donor endothelial survival after EK is comparable to PK and
the graft rejection rate after EK may be lower. Prevention of the
major complications of donor dislocation and increased rate of
primary graft failure in EK is the goal of current research. EK is a
viable, and perhaps preferable, alternative to PK in the treatment
of eyes with endothelial dysfunction.
THE HISTORY OF ENDOTHELIAL

KERATOPLASTY
Endothelial keratoplasty (EK) is a relatively new eld of corneal
transplant surgery which involves the selective replacement of
the recipient diseased endothelium, leaving the normal anterior
surface of the cornea in place.1 Initially, there were two separate
approaches to EK. Several surgeons revived a technique described
long ago by Barraquer,2 Polack,3 and others4 whereby an anterior
flap was created either manually or with a microkeratome, the
flap was retracted, and the posterior recipient stroma was
trephined out. The tissue was replaced with a donor posterior
lamellar button and the overlying flap was then sutured into
place.58 This flap approach to posterior lamellar keratoplasty
(PLK) was attractive to corneal surgeons for several reasons. First,
the surgery involved familiar surgical techniques that had been
nely honed through the surgeons experience with full-thickness penetrating keratoplasty (PK). Second, the surgery utilized
a microkeratome dissection of both the donor and recipient
stromal tissue, and this held the promise of a smoother interface than manual lamellar surgery. And third, it was hoped that
the microkeratome-assisted PLK surgery would be faster and
more efcient than standard PK. Despite initial reports of
successful outcomes, flap PLK ultimately fell into disfavor due to
the problems associated with surface sutures, irregular astigmatism, flap problems, and unpredictable corneal topography.610
In 1993, Ko et al rst described a technique of EK utilizing a

scleral-limbal pocket approach for access to the diseased endothelium, rather than an anterior corneal flap.11 They were the
rst to show proof of concept with their successful animal studies.
In the pocket approach to EK, a limbal scleral incision is made,
a deep corneal lamellar pocket is formed, extending limbus to
limbus, and a central posterior disk of diseased endothelium
and posterior stromal tissue is excised. A similar diameter disk
of healthy donor posterior corneal tissue is then placed through
the pocket, into the recipient anterior chamber and up into the
recipient bed. The donor and recipient tissue are then coapted
using an air bubble in the anterior chamber to hold the tissue
temporarily in place. The air bubble is removed at the end of
surgery, whereby the endothelial pump mechanism and other
factors allow the donor tissue to self adhere. The appeal of this
technique was that it allowed endothelial replacement without
the need for surface corneal incisions or sutures. Without corneal
surface incisions or sutures, most of the postoperative complications of PK are avoided (e.g., suture-induced ulcers, vascularization, and rejection). In addition, the normal premorbid corneal
topography is restored, essentially eliminating the problems of
irregular astigmatism so frequently found with standard PK
surgery. The pocket approach to EK was further developed by
Gerrit Melles12 of the Netherlands who brought the technique
to fruition with the rst human limbal approach EK in 1998.13
Similar to the flap-technique terminology, he named this procedure posterior lamellar keratoplasty, or PLK. Melles initial
technique with humans involved a 9 mm superior incision and
use of an air bubble for recipient tissue dissections and resections,
as well as donor placement and attachment, using only air to
maintain surgical spaces throughout the entire operation.14 In
the absence of viscoelastic use, the PLK procedure performed by
Melles was technically difcult to the extreme. However, he
published the rst successful case and initial small series of cases
of EK in humans, clinically demonstrating the topographic
advantages of this approach with reasonable donor endothelial
cell survival.14 Mark Terry began laboratory work on this procedure in the United States in 1999 with the goal of making the
surgery easier. His laboratory work with the cohesive viscoelastic
Healon (Pzer, New York, NY) established the critical information that this cohesive viscoelastic could be utilized in EK to
stabilize the anterior segment and make the surgery easier, yet
be fully and completely removed from the eye without coating
the interface or jeopardizing donor tissue adhesion.15 Terry and
Ousley established a US prospective study of EK under institutional review board (IRB) approval, and after redesign of instrumentation, they performed the rst US EK in March of
2000.16 Terry renamed this procedure deep lamellar endothelial
keratoplasty (DLEK) in order to differentiate it from the PLK
flap technique, and also to facilitate more accurate Medicare
829
SECTION 6
830
coding.1518 As his DLEK series progressed, Terry simultaneously

established in 2000 the Endothelial Keratoplasty Group (EKG),
and trained 70 corneal surgeons free of charge in order to promulgate the development of EK with prospective IRB-approved
protocols throughout the United States and abroad.19
Gerrit Melles went on to modify his PLK procedure in 2002,
reducing the incision size to 5 mm and advocating that the tissue be folded in half for insertion. To protect the endothelium,
a thin layer of viscoelastic was placed on the endothelial surface,
with the resultant taco appearance containing the endothelium
on the inside. After insertion, the tissue was unfolded using
irrigation with balanced salt solution (BSS) and an air bubble.
His single case report demonstrated proof of concept for a folded
graft to clear the cornea, but the resultant endothelial cell count
was considerably less than his prior large incision series.14,20
Terry adopted Melles small incision idea, moved the incision to
the temporal side, and resubmitted his protocol to his IRB for
prospective analysis of the effect of folding of the graft on endothelial cell survival.21 As more surgeons adopted the small incision DLEK technique, one major problem encountered was the
donor tissue unfolding up-side-down, usually resulting in graft
failure. Terry was the rst to address this problem by advocating
that the donor tissue be over-folded into a 60%/40% taco
conguration prior to insertion. This technique change resulted
in nearly automatic unfolding of the donor tissue in the proper
orientation, nearly eliminating this complication.22,23 At this
same time, other surgeons in the EKG were also advocating
valuable technique changes to the DLEK procedure. Rob Shultze
from New York, Ken Goins from Iowa, and Francisco Sanchez
Leon from Mexico were the rst to begin using a Moria microkeratome to prepare the donor tissue for EK surgery.19 Thomas John
from Chicago described a method of staining the donor tissue
to aid in visualization for positioning24 and he also described a
novel method of phacoemulsicaton prior to DLEK when visualization was poor.25 Ashraf Amayem in Saudi Arabia (and
subsequently in Egypt) demonstrated that DLEK surgery was
particularly useful in the setting of a developing country, and
Rajesh Fogla began the rst prospective study in India. Dr
Amayem also demonstrated that DLEK surgery could be routinely
performed in severely damaged corneas (with preoperative vision
that was worse than count gures), combined with other vitreoretinal surgeries, and could result in good postoperative vision.26
Like all lamellar procedures, the DLEK procedure resulted in
a stromal interface between the donor and recipient tissues
which was not optically pure.27 It was apparent early on in the
development of DLEK that this interface was likely responsible
for at least some degradation of postoperative visual acuity, and
indeed, the average visual acuity reported in most DLEK series
was ~20/40 to 20/50, with a dearth of 20/20 results.14,17,21,23 In
an attempt to improve the smoothness of the recipient interface, Melles described a laboratory study whereby he stripped
Descemets membrane from the recipient and placed donor tissue
directly onto the posterior surface, eliminating the recipient
stromal dissection from the EK procedure.28 This accomplished
two key objectives: one, it produced an easier procedure and two,
it provided a possibly better optical interface. Laboratory work
by Terry et al has since conrmed by scanning electron microscopy
that the recipient bed after Descemets stripping is signicantly
smoother than after DLEK stromal dissections.29 The rst clinical results of the Descemetorhexis technique were reported by
Frank Price in 2005, and he renamed the surgery Descemets
stripping endothelial keratoplasty (DSEK).30 But the ease of the
DSEK technique came with the price of a dramatic increase in
dislocation rate compared with DLEK surgery, with Price and
other surgeons reporting a 50% or higher dislocation rate in
their initial cases. (Price F: Anterior and Posterior Lamellar
Keratoplasty Techniques, Skills Transfer Course for the AAO,

Chicago, 16 Oct 2005.) Much of the recent work in this eld
has been directed at reducing the rate of occurrence of this
signicant complication.
CLINICAL RESULTS WITH EK

In 2003, Terry published27 and presented for consideration ve
goals for the ideal EK procedure and described them as the
following: (1) a smooth surface topography without signicant
change in astigmatism from preop to postop, (2) a highly predictable and stable corneal power, (3) a healthy donor endothelium that resolves all edema, (4) a tectonically stable globe, safe
from injury and infection, and (5) an optically pure cornea. In
addition, a sixth goal was recently added, and that is: (6) A surgical technique that is quickly and easily acquired.31 Over the
past 6 years, the data that have been published independently
by Melles,14,20 Terry,1519,21,23 and Price30 in the evolving techniques of EK have strongly supported the achievement of the
rst four goals in this list. The fth goal of an optically pure
cornea to reach full visual potential has been more elusive, and
the new sixth goal is gradually being achieved through the
evolution of the procedure.
CORNEAL TOPOGRAPHY AND REFRACTIVE

POWER AFTER ENDOTHELIAL
KERATOPLASTY
PK has long been known to suffer from unpredictable postoperative corneal topography. The procedure is plagued by endemic
irregular astigmatism and a predicted corneal power which is
often no better than an educated guess.32,33 Nearly all of the
unpredictability of corneal topography and refractive power after
PK is attributable to the presence of surface corneal sutures and
vertical clear corneal incisions, and many of the publications
about PK over the past 25 years have been regarding suture and
incision manipulations to address these inherent problems.34
The data from EK surgery are best reported by the controlled,
prospective study of DLEK reported by Terry in 2005.3537 In his
prospective, IRB-approved study of 100 consecutive eyes, the
results at 6 months postoperatively of visual acuity, refraction,
and endothelial cell density were reported with 100% follow-up
of every eye.35 The results from this study demonstrate the early
topographic and refractive superiority of endothelial replacement by EK rather than PK. Unlike the average 46 D of
astigmatism that is found after standard PK surgery,3234 DLEK
surgery has an average of only 1.34 0.86 D of astigmatism, and
this represented only a +0.28 1.08 D of change from preoperative refractive astigmatism.35 In addition, an earlier paper
demonstrated that the average spherical equivalent after DLEK
is close to emmetropia (0.18 D), with only an insignicant
0.20 D change from preop measurements.38 Finally, the quality
of corneal topography after DLEK has been documented with the
absence of irregular astigmatism as measured by well-accepted
computerized corneal mapping indices.39,40 The average surface
regularity index (SAI) after DLEK surgery is normal at 0.94 0.18
and the average surface asymmetry index (SRI) is also normal at
0.77 0.51.19 The only data currently available for DSEK
surgery are from a retrospective study by Price which, as
expected, showed similar restored topography as DLEK with an
average postoperative refractive astigmatism of 1.5 0.94 D.30
It is evident, therefore, that EK surgery (whether DLEK or DSEK)
restores the normal corneal topography of the swollen cornea,
providing the patient with a superior and more predictive
refractive and topographic outcome than PK. An example of the
remarkable accuracy of refractive result after EK surgery
FIGURE 61.1. Eye with uncorrected visual acuity of 20/20 at 12 months after DLEK surgery combined with intraocular lens (IOL) exchange. Note
the excellent corneal topography and good donor endothelial cell density of 2282 cells/mm2 representing only a 17% cell loss from preop
measurements.
DONOR ENDOTHELIAL SURVIVAL AND

FUNCTION AFTER ENDOTHELIAL
KERATOPLASTY
The best reports on long-term donor endothelial survival after
PK are those of Bourne and his associates from Mayo Clinic.41,42
The overall donor cell loss after PK in these early studies has
been reported as 34% at 1 year, but this percentage includes cases
done prior to the use of Optisol preservation media, viscoelastic
use and other surgical innovations. More recent studies of PK
using Optisol preservation media have a reported donor
endothelial cell loss of ~1519% 1 year after surgery, and 34%
2 years after surgery.4345
In the prospective study of DLEK, Terry found that the average
endothelial cell density at 6 months was 2140 + 427 cells/mm2,
and that this represented a 25% cell loss from preop eye bank
measurements. Surprisingly, the cell loss from large incision
DLEK was not signicantly different from small-incision DLEK
where the tissue is folded and unfolded.35 Also, the stability of
the donor endothelial layer after DLEK is quite good with 2 year
data revealing an average cell count of 2151 + 457 cells/mm2
and was not signicantly different from what was found at
1 year postoperatively.37 The cell loss from DSEK has not been
reported in a large series, however the most recent data from the
initial 16 cases of DSEK by Gorovoy, report an average cell loss
of 40%.46 Whether this increased cell loss in DSEK is a function
of the procedure or simply intersurgeon variation remains to
be seen.
TECTONICALLY STABLE GLOBE

Two of the short-term liabilities of PK are a leaking wound postoperatively and the usual presence of a surface epithelial defect.
After EK, the scleral wound is secured with 23 interrupted
sutures and wound leaks have never been reported from this
highly beveled wound. In addition, in most cases of EK, the
recipient epithelial surface is intact the day after surgery and so,
unlike most cases of PK, the immediate protection of the epithelial surface is present.
Some of the long-term concerns with standard PK are the
liabilities which surface corneal sutures and incisions can present for the health and stability of the globe. Surface sutures can
induce vascularization and subsequent rejection of the graft.47
Corneal sutures can break and cause a bacterial keratitis which
can destroy not only the graft, but potentially the eye through
exogenous endophthalmitis.48,49 Finally, the vertical clear corneal wounds of the PK eye never heal with the same structural
strength as the virgin tissue, and the risk of a ruptured globe
from minor blunt trauma, even years after PK, is always present.5056 In DLEK and DSEK surgery, corneal surface sutures
are completely eliminated and the only wound is a short 5 mm
length incision placed in the well-vascularized region of the
scleral limbus. Minimal recipient tissue is excised, and the donor
tissue is securely scarred into place along the edges postoperatively by 3 months or earlier.26,36,57,58 At this point in time,
there have been no reports of trauma to a patient after endothelial keratoplasty, but given the highly beveled and short
length of the incision in this surgery, it seems unlikely that we
will encounter the devastating loss of vision that can occur with
globe rupture such as we have seen with PK.
INTERFACE CLARITY AND VISUAL FUNCTION

AFTER ENDOTHELIAL KERATOPLASTY
EK resolves corneal edema rather quickly and restores the premorbid corneal topography. Therefore, by avoiding the early
irregular astigmatism from sutures that we often see from PK,
EK allows faster visual rehabilitation than standard
PK.14,17,19,21,23,27,30,31,35,38,46 However, while the average visual
acuity of DLEK and DSEK surgery appears to be as good as (or
superior to) PK at 6 months postop, the evidence to date shows
that there are far fewer patients that achieve a level of 20/20
vision after EK surgery than what we have come to expect from
our PK surgery.14,19,21,26,3032,35,46,59,60 In the absence of retinal
macular disease, the explanation for this paucity of 20/20
results is most likely due to optical interference from the
horizontal stromal interface. Current work in the eld of EK is
directed at improving the optical qualities of the interface to
achieve the frequency of 20/20 vision that we get with PK
surgery. However, research with the femtosecond laser and the
microkeratome to improve the interface has not yet brought
this goal to fruition.6164 Most patients with EK surgery see
20/40 or better within 36 months after surgery, but in the
Terry series of his rst 100 DLEK cases, only one eye saw
20/20.35 Similarly, although the vision of 20/40 was common at
6 months after DSEK surgery, not a single eye of 50 DSEK cases
in the Price report saw 20/20.30 Potential vision may be mitigated by not only the interface, but also by the residual anterior
stromal tissue. Chronic edema from Fuchs dystrophy and other
endothelial disorders can create structural changes in the anterior stromal tissue, which may induce visual dysfunction such
as reduced contrast sensitivity and reduced Snellen visual acuity.
These structural changes have recently been measured as
increased light scatter compared to PK tissues, but further work
CHAPTER 61
combined with IOL exchange is shown in Figure 61.1 with a

DLEK eye at 1 year that sees 20/20 without glasses, with an
excellent topography and endothelial cell density.
831
FIGURE 61.2. Eye only 3 days after DSEK surgery combined with phacoemulsication cataract extraction with an uncorrected visual acuity of
20/50 and a corrected vision of 20/30 with a manifest refraction of 0.75 + 0.50 180. The air bubble left in the anterior chamber to stabilize the
graft is nearly absorbed.
in this area is required. (Patel SV, et al: Comparison of corneal

haze after DLEK and PK, Federated Scientic Session, Chicago,
15 Oct 2005). In our experience, the patients that have had one
eye with a PK and one eye with an EK overwhelmingly prefer
the EK eye, even in instances where the Snellen visual acuity
measurement is better in the PK eye (unpublished data). In
addition, after EK surgery, the visual recovery time can be
extremely fast, with many patients having useful vision within
just a few weeks after surgery (Fig. 61.2). This brings up the
practical point of EK surgery: a cornea that provides a normal
topography but a suboptimal stromal interface may be of greater
benet to the patient than a PK cornea which has irregular or
high astigmatism, but can still be corrected to 20/20 Snellen
visual acuity. Discussion of the benets and disadvantages of
both procedures (EK and PK) is warranted in the preoperative
counseling of endothelial dysfunction patients.
SECTION 6
EASE OF SURGERY
832
PK surgery is a skill which is introduced in residency training

and perfected in the corneal fellowship training. The techniques
utilized in PK, however, are not necessarily transferable to EK
surgery. The initial PLK procedure described by Melles was done
entirely under air, and the donor tissue was processed by the
surgeon from a whole eye.1214 The DLEK technique described
by Terry1519,2123,3538,65 was easier with the introduction of
viscoelastic and an articial anterior chamber for donor
cornealscleral tissue processing, but still required the separate
skill sets of deep lamellar stromal plane dissections. In DSEK
surgery, the recipient dissection is replaced with a simple stripping of Descemets membrane, making the surgery even easier.
Finally, the use of the microkeratome for donor preparation
further simplies the surgery by eliminating the need for any
manual stromal dissection. It is further anticipated that donor
tissue will soon be available precut by the distributing eye bank,
eliminating the cost and time involved for the donor preparation phase of EK surgery. All of these advances have made EK
surgery easier, but should not be misconstrued to imply that the
surgery is easy. Proper handling of the donor endothelium during
folding, insertion, and intraocular unfolding must be learned as
a new skill set, and air bubble manipulation during and after
surgery are equally important for donor tissue adherence and
tissue vitality. The transplant surgeon contemplating adding EK
to his/her surgical repertoire is well advised to read the literature, review the available teaching videos, and take an extensive
hands-on course in EK before attempting ones rst case.
COMPLICATIONS OF EK
The complications of PK surgery are well documented and
include: visually debilitating irregular astigmatism, sutureinduced infectious keratitis, ulceration, endophthalmitis, and
ruptured globes from minor blunt trauma.4756 In addition, the
incidence of graft rejection in grafts for Fuchs dystrophy is
reported at ~10%.60 The avoidance of these complications is
what makes EK surgery most appealing.
In his prospective study of DLEK surgery, Terry has reported
the small list of complications in the rst 6 months after
surgery. The most signicant risk is of donor dislocation from
the recipient bed, usually found the rst day after surgery. In the
Terry series of 100 eyes, the incidence of dislocation was 4%,
and all of the eyes regained corneal clarity with repositioning of
the tissue with an air bubble in the anterior chamber.35 In the
initial DSEK series, the incidence of dislocation was reported as
high as 50%, but the recent incidence is now reported at between
3% and 25%.29,46 While dislocated endothelial grafts can be
reattached, there is evidence that the further manipulation
involved in reattachment results in a further signicant reduction of donor endothelial cells compared to grafts which did not
require repositioning.36 Dislocation of the graft is now most
likely correlated with the degree of processing and surgical
trauma to the donor endothelial cells, but documentation of
this supposition is still required.
Similar to the medical management of PK eyes, the EK eye is
placed on topical steroids and these steroids can exacerbate or
induce elevated intraocular pressure with the potential of visual
eld loss. This occurred in four eyes in the Terry series and has not
been investigated in any published DSEK series at this time.35,36
Cataract formation after EK surgery has also been reported, and
steroids, rather than EK surgical trauma, have been implicated
as the etiology.35,36,57 Careful monitoring for drug-induced
complications of any EK eye on steroids is obviously indicated.
There is the possibility that the incidence of graft rejection
and failure may be less after EK surgery than after PK surgery.
Although, the primary antigenic stimulus for rejection is the
donor endothelium (present in both PK and EK eyes), the EK eye
Complications
Complications of EK
1. Dislocation of donor tissue
2. Primary graft failure rate higher than PK
3. Steroid-exacerbated glaucoma
Vs
Complications of PK
1. High astigmatism
2. Irregular astigmatism
3. Unpredictable topography and refractive error
4. Suture-related problems: vascularization, infection, ulceration,
rejection
5. Wound-related problems: poor healing, wound leak, infection,
late rupture
6. Endophthalmitis from retained suture fragments
7. Globe rupture from minor trauma
8. Steroid-exacerbated glaucoma
9. Graft rejection rate higher than EK
FUTURE ADVANCES IN EK
Cognizant of the limitations which the stromal interface of EK
surgery can place on nal visual acuity, it is appealing to explore
the possibility of a pure Descemets membrane transplantation,
providing the most optically smooth interface possible. While
Melles has shown that it is possible to strip donor Descemets
membrane and transplant it to a smooth stripped recipient bed
in the laboratory,67 this work has not been successfully repeated
in the clinical realm in any published reports. Pure Descemets
membrane is quite fragile and manipulations of donor tissue
which are well tolerated in the current EK techniques result in
wrinkles, folds, tears, and unacceptable endothelial cell death
when applied to pure Descemets transplantation. Hopefully,
newer techniques and instrumentation will overcome these
challenges.68,69
Recent work with in situ human corneal endothelial cell
regeneration is exciting and complementary to the evolution of
the surgical techniques of EK. In Boston, Dimitri Azars group
has shown success in amplication of endothelium on a stromal
carrier.70 This has the potential of taking the recipients peripheral endothelial cells, increasing the cell density in the laboratory,
and then retransplanting them back to the recipient central cornea
with EK; thus circumventing any issues of immune-mediated
graft rejection. In Japan, Tatsuya Mimuras ground-breaking laboratory work with human endothelial cell precursors, has successfully treated bullous keratopathy of the rabbit cornea with
the injection of precursor cells into the anterior chamber and
subsequent eye-down positioning.71 If we can extrapolate this
animal model to the clinical realm, then EK as we know it, may
be completely eliminated in the future by a simple injection of
precursor endothelial cells. Finally, it may be possible 1 day to
prevent the need for EK entirely by directly stimulating the
patients remaining endothelial cells to regenerate, utilizing viral
vectors to transfer genetic material which induces and controls
endothelial mitosis.72
These early years of the twenty-rst century herald an era
of custom keratoplasty, selectively treating only the diseased
portion of the cornea and leaving the normal portions intact.1
The technical and laboratory innovations which have taken place
this past decade and which are on the horizon, promise a leap
forward in our ability to provide quick and excellent visual rehabilitation for our patients suffering from endothelial dysfunction. It is the responsibility of each corneal surgeon to embrace the
new techniques and technologies that work, apply them to ones
individual clinical practice, and enjoy the benets endowed to
surgeon and patient alike.
ACKNOWLEDGMENT
Dr Terry has a small nancial interest in the specialized instruments used in
endothelial keratoplasty surgery. Bausch and Lomb Surgical (St Louis, MO)
manufactured and supplied the specially designed instruments free of charge.
CHAPTER 61
lacks many of the other inciting factors of rejection that are

present in PK surgery. There are no sutures in EK, no epithelial
rejection reactions, no induced vascularizations or ulcerations,
and no induced Purkinje cell processing in EK. In the absence of
these liabilities, the incidence of graft rejection loss over 2 years
in 56 eyes with DLEK surgery was less than 2%. (Terry MA, et al:
Deep Lamellar Endothelial Keratoplasty (DLEK): graft rejection
and failure is less likely than after PK surgery. Poster presentation,
annual meeting of the American Academy of Ophthalmology,
Chicago, 14 Oct 2005.) This perceived immunologic advantage
of EK surgery will require longer-term study of greater numbers
before general acceptance.
The nal complication of EK surgery is that of primary graft
failure. In the Terry series, there was one case (1%) out of
100 DLEK cases which required tissue replacement immediately
after surgery due to surgeon error.35,36 In the Price series of
50 DSEK cases, there were three cases (6%) which required a
second graft within 1 week of the rst graft.30 In the Gorovoy
series of just 16 DSEK eyes, there was one graft (6%) that was
replaced, due again to surgeon error.46 This incidence of tissue
failure requiring reoperation and second tissue is much higher
than in PK studies, and as such, has economic and patient safety
ramications. Early graft failure in EK is likely due to the surgeons
learning curve for endothelial manipulations, and as more novice
EK surgeons become active, the incidence of primary graft failure
is likely to rise dramatically. Indeed, in their initial experience
with DSEK at Moorelds in London, experienced anterior
lamellar surgeons most recently reported failure of ve out of
their initial 11 DSEK cases for a 45% rate of graft loss.66
Hopefully, as EK surgery becomes mainstream, and as we learn
more on how to protect endothelial vitality, the incidence of this
complication will return to PK levels.
REFERENCES
1. Terry MA: The evolution of lamellar grafting
techniques over 25 years. Cornea 2000;
19:611616.
2. Barraquer JI: The technique for penetrating
keratoplasty. In: Barraquer J, Rutlan J, eds.
Microsurgery of the cornea: an atlas and
textbook. Barcelona: Ediciones Scriba
1984; 289294.
3. Polack FM: Queratoplastia lamelar posterior.
Revista Peruana de Oftalmologia 1965;
2:6264.
4. Tillet CW: Posterior lamellar surgery. AJO

1956; 41:530533.
5. Jones DT, Culbertson WW: Endothelial
lamellar keratoplasty (ELK) [ARVO Abstract].
Invest Ophthalmol Vis Sci 1998; 39:S76.
Abstract nr 342.
6. Busin M, Arffa RC, Sebastiani A:
Endokeratoplasty as an alternative to
penetrating keratoplasty for the surgical
treatment of diseased endothelium: initial
results. Ophthalmology 2000; 107:20772082.
7. Azar DT, Jain S, Sambursky R, Strauss L:

Microkeratome-assisted posterior
keratoplasty. J Cataract Refract Surg 2001;
27:353356.
8. Ehlers N, Ehlers H, Hjortdal J, MollerPedersen T: Grafting of the posterior
cornea: description of a new technique
with 12-month clinical results. Acta
9. Join S, Azar DT: New lamellar keratoplasty
techniques: posterior keratoplasty and
833
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
SECTION 6
22.
23.
24.
25.
26.
27.
834
deep lamellar keratoplasty. Curr Opin

Ophthalmol 2001; 12:262268.
Silk W, Gupta P, Stevens G: Endothelial
lamellar keratoplasty: ten patient series
report. Ft Lauderdale, FL: ARVO abstract;
2002.
Ko W, Freuh B, Shield C, et al:
Experimental posterior lamellar
transplantation of the rabbit cornea (ARVO
abstract). Invest Ophthalmol Vis Sci 1993;
34:1102.
Melles GR, Eggink FA, Lander F, et al: A
surgical technique for posterior lamellar
Melles GR, Lander F, Beekhuis WH, et al:
Posterior lamellar keratoplasty for a case of
pseudophakic bullous keratopathy. Am J
Ophthalmol 1999; 127:340341.
Melles GR, Lander F, van Dooren BR, et al:
Preliminary clinical results of posterior
lamellar keratoplasty through a
sclerocorneal pocket incision.
Ophthalmology 2000; 107:18501856.
Terry MA, Ousley PJ: Endothelial
replacement without surface corneal
incisions or sutures: topography of the
deep lamellar endothelial keratoplasty
procedure. Cornea 2001; 20:1418.
Terry, MA, Ousley, PJ: Deep lamellar
endothelial keratoplasty in the rst United
States patients: early clinical results.
Cornea 2001; 20:239243.
Terry MA, Ousley PJ: Replacing the
endothelium without surface corneal
incisions or sutures: rst U.S. clinical series
with the deep lamellar endothelial
keratoplasty procedure. Ophthalmology
2003; 110:755764.
Terry MA: Endothelial replacement: the
limbal pocket approach. Ophthalmol Clin
North Am 2003; 16:103112.
Terry MA: Endothelial replacement surgery
In: Krachmer J, Mannis M, Holland E, eds.
Cornea: surgery of the cornea and
conjunctiva. 2nd edn. St Louis, MO:
Elsevier Mosby; 2005:17071718.
Melles GR, Lander F, Nieuwendaal C:
Sutureless, posterior lamellar keratoplasty:
a case report of a modied technique.
Cornea 2002; 21:325327.
Terry MA, Ousley PJ: Small incision deep
lamellar endothelial keratoplasty (DLEK):
6 months results in the rst prospective
clinical study. Cornea 2005; 24:5965.
Terry MA: Endothelial replacement: the
limbal pocket approach. Ophthalmol Clin
North Am 2003; 16:103112.
Terry MA, Ousley PJ: Rapid visual
rehabilitation with deep lamellar endothelial
keratoplasty Cornea 2004; 23:143153.
John T: Use of iodocyanine green in deep
lamellar endothelial keratoplasty. J Cataract
Refract Surg 2003; 29:437443.
John T: Upside-down phacoemulsication
in deep lamellar endothelial keratoplasty In:
John T, ed. Surgical techniques in anterior
and posterior lamellar corneal surgery. New
Delhi Jaypee Brothers Medical Publishers
(P) Ltd; 2005:372379.
Amayem AF, Terry MA, Helal MH, et al:
Deep lamellar endothelial keratoplasty
(DLEK): surgery in complex cases with
severe preoperative visual loss. Cornea
2005; 24:587592.
Terry MA: Deep lamellar endothelial
keratoplasty (DLEK): pursuing the ideal
goals of endothelial replacement. Eye 2003;
17:982988.
28. Melles GR, Wijdh RH, Nieuwendaal CP:

A technique to excise the descemets
membrane from a recipient cornea
(descemetorhexis). Cornea 2004;
23:286288.
29. Terry MA, Hoar K, Wall J, Ousley PJ: The
histological basis of donor dislocation in
endothelial keratoplasty surgery: DLEK and
DSEK. Cornea 2006; 25:926932.
30. Price FW, Price MO: Descemets stripping
with endothelial keratoplasty in 50 eyes:
a refractive neutral corneal transplant.
J Refract Surg 2005; 21:339345.
31. Terry MA: Endothelial keratoplasty: history,
current state and future directions. Cornea
(editorial) 2006; 25:873878.
32. Davis EA, Azar DT, Jacobs FM, Stark WJ:
Refractive and keratometric results after
the triple procedure: experience with early
and late suture removal. Ophthalmology
1998; 105:624630.
33. Pineros OE, Cohen EJ, Rapuano CJ,
Laibson PR: Triple vs nonsimultaneous
procedures in Fuchs dystrophy and
cataract. Arch Ophthalmol 1996;
114:525528.
34. Sugar A, Sugar J: Techniques in
penetrating keratoplasty: a quarter century
of development. Cornea 2000; 19:603608.
35. Terry MA, Ousley PJ: Deep lamellar
endothelial keratoplasty (DLEK): visual
acuity, astigmatism, and endothelial
survival in a large prospective series.
Ophthalmology 2005; 112:15411549.
endothelial keratoplasty (DLEK): early
complications and their management.
Cornea 2006; 25:3743.
37. Ousley PJ, Terry MA: Stability of vision,
topography, and endothelial cell density
from one year to two years after deep
lamellar endothelial keratoplasty
(DLEK) surgery. Ophthalmology 2005,
112:5057.
38. Terry MA, Ousley PJ: In pursuit of
emmetropia: spherical equivalent results
with deep lamellar endothelial keratoplasty.
Cornea 2003; 22:619626.
39. Dingeldein SA, Klyce SD, Wilson SE:
Quantitative descriptors of corneal shape
derived from computer-assisted analysis of
photokeratographs. Refract Corneal Surg
1989; 5:372378.
40. Wilson SE, Klyce SD: Quantitative
descriptors of corneal topography: a
clinical study. Arch Ophthalmol 1991;
109:349353.
105:18551865.
42. Bourne WM: Cellular changes in
transplanted human corneas. Cornea 2001;
20:560569.
43. Bourne WM, Nelson LR, Maguire LJ, et al:
Comparison of Chen medium and
Optisol-GS for human corneal preservation
at 4 degrees C. Results of transplantation.
Cornea 2001; 20:683686.
44. Lass JH, Bourne WM, Musch DC, et al:
A randomized, prospective, double-masked
clinical trial of Optisol vs DexSol corneal
storage media. Arch Ophthalmol 1992;
110:14041408.
45. Frueh BE, Bohnke M: Prospective,
randomized clinical evaluation of Optisol
vs organ culture corneal storage media.
46. Gorovoy MS: Descemet stripping

automated endothelial keratoplasty. Cornea
2006; 25:886889.
47. Stechschulte SU, Azar DT: Complications
after penetrating keratoplasty. Int
48. Akova YA, Onat M, Koc F, et al: Microbial
keratitis following penetrating keratoplasty.
49. Conno J, Brown SI: Bacterial
endophthalmitis associated with exposed
monolament sutures following corneal
99:111113.
50. Abou-Jaoude ES, Brooks M, Katz DG, et al:
Spontaneous wound dehiscence after
removal of single continuous penetrating
keratoplasty suture. Ophthalmology 2002;
109:12911296.
51. Tseng SH, Lin SC, Chen FK: Traumatic
wound dehiscence after penetrating
keratoplasty: clinical features and outcome
in 21 cases. Cornea 1999; 18:553558.
52. Perry HD, Donnenfeld ED: Expulsive
choroidal hemorrhage following suture
removal after penetrating keratoplasty.
53. Binder PS, Abel R Jr, Polack FM, et al:
Keratoplasty wound separations. Am J
Ophthalmol 1975; 80:109115.
54. Agrawal V, Wagh M, Krishnamachary M,
et al: Traumatic wound dehiscence after
penetrating keratoplasty. Cornea 1995;
14:601603.
55. Rehany U, Rumelt S: Ocular trauma
following penetrating keratoplasty:
incidence, outcome, and postoperative
recommendations. Arch Ophthalmol 1998;
116:12821286.
56. Elder MJ, Stack RR: Globe rupture
following penetrating keratoplasty: how
often, why, and what can we do to prevent
it? Cornea 2004; 23:776780.
57. Price MO, Price FW: Cataract progression
and treatment following posterior lamellar
30:13101315.
58. Kapur R, Sugar J, Edward DP: Deep
lamellar endothelial keratoplasty: histology
of complications in initial cases. Cornea
2006; 25:851854.
59. Pineros OE, Cohen EJ, Rapuano CJ,
Laibson PR: Triple vs nonsimultaneous
procedures in Fuchs dystrophy and
cataract. Arch Ophthalmol 1996;
114:525528.
60. Claesson M, Armitage WJ, Fagerholm P,
et al: Visual outcome in corneal grafts:
a preliminary analysis of the Swedish
Corneal Transplant Register. Br J
Ophthalmol 2002; 86:174180.
61. Terry MA, Ousley PJ, Wills B: A practical
femtosecond laser procedure for DLEK
endothelial transplantation: cadaver eye
histology and topography. Cornea 2005;
24:453459.
62. Soong HK, Mian S, Abbasi O, et al:
Femtosecond laser-assisted posterior
lamellar keratoplasty. Ophthalmology 2005;
112:4449.
63. Sarayba MA, Juhasz T, Chuck RS, et al:
Femtosecond laser posterior lamellar
keratoplasty: a laboratory model. Cornea
2005; 24:328333.
64. Kang PC, McEntire MW, Thompson CJ,
Moshirfar M: Preparation of donor tissue
for deep lamellar endothelial keratoplasty
(DLEK) using a microkeratome and articial
using a small-incision technique. Cornea
2006; 26:279283.
67. Melles GR, Lander F, Rietveld FJR:
Transplantation of descemets membrane
carrying viable endothelium through a
small scleral incision. Cornea 2002;
21:415418.
68. Shimmura S, Miyashita H, Konomi K, et al:
Transplantation of corneal endothelium with
Descemets membrane using a
hydroxyethyl methacrylate polymer as a
carrier. Br J Ophthalmology 2005;
89:134137.
69. Mimura K, Yamagami S, Yokoo S, et al:
Cultured human corneal endothelial cell
transplantation with a collagen sheet in a
rabbit model. Invest Ophthalmol Vis Sci

2004; 45:29922997.
70. Chen K-H, Azar D, Joyce NC:
Transplantation of adult human corneal
endothelium ex vivo: a morphologic study.
Cornea 2001; 20:731737.
71. Mimura T, Yokoo S, Araie M, et al:
Treatment of rabbit bullous keratopathy
with precursors derived from cultured
human corneal endothelium. Invest
72. McAlister JC, Joyce NC, Harris DL, et al:
Induction of replication in human corneal
endothelial cells by E2F2 transcription
factor cDNA transfer. Invest Ophthalmol Vis
Sci 2005; 46:35973603.
CHAPTER 61
anterior chamber system: endothelial cell

loss and predictability of lamellar thickness.
Ophthalmic Surg Lasers Imaging 2005;
36:381385.
endothelial keratoplasty: small incision
technique combined with
phacoemulsication and posterior chamber
intraocular lens implantation. In: John T, ed.
Surgical techniques in anterior and
posterior lamellar corneal surgery. New
Delhi Jaypee Brothers Medical Publishers
(P) Ltd; 2005:345364.
66. Mearza AA, Qureshi MA, Rostron CK:
Experience and early results of Descemets
stripping endothelial keratoplasty (DSEK)
835
CHAPTER
62
Complications of Corneal Transplantation and

Their Management
Paul D. OBrien and William J. Power
Key Features
The most common clinical indications for penetrating

keratoplasty in the United States and Canada are
pseudophakic bullous keratopathy, Fuchs dystrophy,
keratoconus, and aphakic bullous keratopathy
The factor most commonly associated with an increased risk
of allograft rejection is corneal vascularization
Any suture that I broken, loose, or associated with stromal
vascularization across the wound should be removed
immediately to lower the risk of rejection
Intensive topical steroid treatment should immediately be
commented for newly documented corneal graft failures
The success rate of rejection treatment is usually ~ 50%
Increasing corneal thickness measured by pachymetry is a
very useful way to document progressing graft failure
Refractive unpredictability after penetrating keratoplasty is
extremely common with most series documenting mean
cylinder of 4-5 Diopters and significant anisometropia
INTRODUCTION
Penetrating keratoplasty (PK) has become the most frequently
performed method of tissue transplantation in the United States.
The Eye Bank Association of America reports that 32 106 corneal donor grafts were used for transplantation in 2004 in the
United States, and since the establishment of this association
in 1961 more than 60 000 corneas have been provided for PK.1
The relative success of PK is attributed to continued advances
in surgical techniques and equipment, ocular pharmacology and
immunology, corneal storage, and eye banking procedures.2 The
most common clinical indications for PK in the United States
and Canada are pseudophakic bullous keratopathy (PBK), Fuchs
dystrophy, keratoconus, and aphakic bullous keratopathy (ABK).35
PK performed for PBK accounted for 20% of cases in the year
2000 according to the Eye Bank Association of America Survey
statistics, although there is a decreasing trend for this indication due to advances in phacoemulsication surgery and intraocular lens (IOL) designs. Corneal opacity as the end result from
trauma or inflammation secondary to infection in particular
herpes simplex virus (HSV), and, hereditary stromal dystrophies are other conditions that may benet from PK. In low-risk
clinical situations, such as keratoconus and Fuchs dystrophy,
PK is one of the most successful forms of transplantation with
a 10-year success rate of ~90% (with success dened as corneal
clarity).6 When all indications for PK are included, 10-year success rates ranging from 59% to 80% have been documented.68
This very high success rate in uncomplicated cases is despite
the fact that tissue matching is not routinely performed and
that systemic immunosuppression is rarely used unless

required in high-risk clinical situations.
PREOPERATIVE CONSIDERATIONS
Most preoperative considerations can be subclassied into donor
or recipient considerations. The identication of risk factors for
graft failure can help surgeons determine which eyes are at
increased risk, so that those eyes can be treated more aggressively. Graft failure can be broadly dened as any irreversible
change in the graft preventing recovery of useful vision.
DONOR CONSIDERATIONS
At present, one of the most important determinants of a successful corneal transplant program is high-quality eye bank
screening and processing of human donor corneal tissue. Donor
screening for infectious diseases is essential to protect recipients
from severe or life-threatening illnesses by donor tissue transmission of human immunodeciency virus (HIV) and prions.
Donor serology is used to screen for syphilis, hepatitis, and HIV.
Eye bank staff aim to detect donor risk behaviors for HIV by
taking a detailed sociobehavioral history.
Donor corneas from infants younger than 18 months of age
should be avoided for transplantation because corneal ectasia
and high myopia tend to develop in recipients after PK. There is
no upper age limit for donor tissue as long as the endothelial
cell count is adequate. Specular microscopy with corneal endothelial cell count on donor tissue of all ages is an important
preventive measure against primary donor failure. Adequate
donor cornea preservation before PK is essential because PK
requires a viable donor endothelial layer that can sustain the
surgical trauma and future cellular changes and cell loss. The
normal cornea enjoys a relative immune privilege,9 with a low
antigen load, few antigen-presenting cells, and a donor-specic
anterior chamber-associated immune deviation that develops
after transplantation. The benets of donorrecipient HLA matching to prevent allograft immune rejection and ultimate graft
failure are still controversial. The current state of knowledge
about tissue matching for corneal transplantation programs is
that ABO blood group matching, which can be achieved with
relatively little effort and expense, may be effective in reducing
the risk of graft failure at a currently acceptable cost in high-risk
patients.10
RECIPIENT CONSIDERATIONS
Repeat corneal transplantation6,11 and other high-risk factors,
such as recipients with signicant corneal vascularization,7,12,13
glaucoma,4,12 or peripheral anterior synechiae12,13 carry a poor
837

patients with HSV scarring.19 It seems prudent therefore to
prescribe this medication for every patient with HSV corneal
scarring for the rst postoperative year.
Ocular surface conditions such as cicatricial conjunctival diseases (e.g., ocular cicatricial pemphigoid and the StevensJohnsons
disease), severe dry eye (e.g., Sjgrens syndrome), or limbal stem
cell deciency conditions (e.g., congenital aniridia, chemical
burn) all carry with them a very bad prognosis for PK because
epithelial healing is very poor in such conditions (Figs 62.1a,b).
Limbal stem cell transplantation alone for congenital aniridia
keratopathy should be considered. For cicatricial conjunctival
diseases, either a two-step procedure with limbal stem cell transplantation followed by PK or a keratoprosthesis is preferable to
PK. Persistent epithelial defects, stromal melting, and corneal
perforation may well develop after PK in these patients.
Major mechanical or inflammatory lid abnormalities must be
noted and corrected before surgery. Patients with severe chronic
blepharitis and atopic blepharitis are at high risk for graft
infection and melt. These conditions must be under control at
the time of and after PK.
Graft Size Determination

This is based on the three main factors: the size of the recipient
cornea, the targeted disease, and the known risk of rejection
with increasing graft size. A normal-sized cornea with endothelial disease such as Fuchs dystrophy or PBK would typically be
trephinated with a 7.5 mm trephine. Although a larger recipient
size trephination would supply a larger quantity of healthy
endothelial cells, larger donor corneal size (> 8 to 8.5 mm) is
associated with a higher risk of rejection. Ectatic corneal pathologic processes, including keratoconus and pellucid marginal
degeneration, require prior recipient donor size measurements
at the slit-lamp because the area of corneal thinning is often
larger than that seen under the operating microscope. Most surgeons use a 0.25 mm oversize in the donor button to counteract the 0.2 mm difference in size produced by the endothelial
trephination of the donor cornea.
COMPLICATIONS
There is an array of complications associated with corneal transplant surgery. It is most convenient to classify them according
to when they occur, namely, intraoperatively, early in the postoperative period, or late in the postoperative period (Table 62.1).
SECTION 6
success rate of 4050%. One study14 that analyzed the risk factors
for graft failure found that any of the following factors: preoperative glaucoma, the presence of peripheral anterior synechiae,
deep stromal vascularization, or African-American race, each
independently doubled the relative risk of overall graft failure.
Regraft success rates decrease markedly after the third and
fourth regraftings compared with the rst and the second.15 The
mean corneal graft survival time also decreases with subsequent
grafting. The recipient with a previous history of graft failure
has often been exposed to poor prognostic factors such as the
presence of corneal vessels, peripheral anterior synechiae, raised
intraocular pressue, and prior immune allograft reaction mediators.11 Corneal vascularization causes the loss of the immune
privilege of the avascular cornea and increases exposure for nonself-antigens which in turn increases the likelihood for corneal
graft rejection. The presence of deep corneal vessels is associated with a twofold increased risk of allograft immune reaction
and failure.14,16 Rejection is the major cause for graft failure in
repeated transplantation. Close follow-up during the susceptible
period after repeated PK, extended use of adjuvant therapy, such
as topical corticosteroids, systemic or topical cyclosporin, and
specic treatment for a specic preoperative diagnosis may
improve the survival of repeated corneal grafts. The Collaborative
Corneal Transplantation Studies Research Group identied
additional risk factors for rejection such as a recipient age less
than 40 years, a combined surgery, and a graft diameter more
than 8 mm.17
Intraocular pressure (IOP) control prior to surgery is essential
because a preoperative diagnosis of glaucoma co-morbidity is a
major risk factor for corneal graft failure.4,12 Preoperative glaucoma is often made worse after PK by partial loss of the ltration
angle. The chronic use of corticosteroids after PK may also induce
glaucoma in patients who are steroid responders. Abnormally
low preoperative IOP, such as in patients with chronic uveitis,
may be further aggravated by the surgery, causing macular
edema and poor visual outcome.
Any intraocular inflammation must be controlled before PK
unless surgery is required in an emergency situation (e.g., corneal
perforation). The outcome of a PK is much poorer for patients
with active corneal herpetic inflammation than for patients with
a quiet herpetic corneal scar.18 A signicant reduction in the
number of HSV-related events (HSV recurrences and rejection
episodes) was detected when oral acyclovir 400 mg was administered twice daily during the rst year after keratoplasty in
838
FIGURE 62.1. These photographs show a patients eye (a) before and (b) after fluorescein staining of a persistent epithelial defect 6 weeks after
corneal grafting. The indication for performing a PK in this patient was corneal opacity and keratinization due to a chemical injury.
Complications of Corneal Transplantation and Their Management
TABLE 62.1. Complication Table

Intraoperative Complications
Damage to donor cornea

Inadvertent trephine reversal
Retained Descemets membrane
Irislens damage
Anterior chamber hemorrhage
Suprachoroidal expulsive hemorrhage
Postoperative Complications Early
Wound leak
Persistent epithelial defect
Suture-related problems
Loose sutures/inltrates/abscesses/wound
dehiscence
Infectious crystalline keratopathy

Endophthalmitis
Elevated IOP
Anterior synechiae
Pupillary block
UrretsZavalia syndrome
Late Complications
Immunologic rejection
Epithelial/stromal/endothelial
Endothelial failure
Postoperative astigmatism
Traumatic globe rupture
INTRAOPERATIVE COMPLICATIONS
Damage to the Donor Cornea
CHAPTER 62
It is necessary to trephinate the donor cornea before recipient

trephination. Any damage to the donor cornea, whether contamination or improper or incomplete trephination, warrants
cancellation of the surgery. Should the corneal button remain
within the trephine after punching, care should be taken not to
use instrumentation to get it to fall back into the well as this
can lead to inadvertent inversion of the corneal button. A
balanced salt solution (BSS) or optisol should be applied into the
trephine to coax the button to fall back on its own. Any
confusion as to the endothelial and epithelial side of the button
is impossible to conrm visually, and therefore a no-touch
approach to the corneal button is the best approach. Inadvertent
inversion of the button leads to primary graft failure. This
complication has been previously reported.20,21
Recipient Trephination
All trephine sizes should be double checked by a second observer,
thereby avoiding the disastrous complication of inadvertent
trephine reversal. Trephine sizes should also be double checked
under the microscope to ensure that the package labeling is correct. If the donor button is smaller than the recipient site, it
may be difcult to suture the button in place and secure a watertight wound. The tightened sutures may also result in shallowing of the anterior chamber (AC) and collapse of the trabecular
meshwork, resulting in raised IOP postoperatively.
FIGURE 62.2. This photograph shows the presence of a double AC

on the rst postoperative day. There is a second slit beam to the right
of the main slit beam, which represents the retained Descemets
membrane.
Retained Descemets Membrane

Inadvertent retention of Descemets membrane is surprisingly
easy in edematous corneas and cases of interstitial keratitis in
which Descemets membrane is thickened. This occurs when
incomplete penetration into the AC has occurred and the corneal
scissors remains above Descemets membrane during cutting.

The iris should be inspected and gently touched with a weck
sponge or lifted with forceps to ensure the absence of Descemets
membrane. This complication can be easily missed and will
839

result in a double AC on the rst postoperative day (Fig. 62.2).
If it occurs it results in graft failure due to endothelial damage
and it is therefore an indication for regrafting.
IrisLens Damage
The advent of viscoelastics has greatly decreased the prevalence
of intraocular damage during trephination. When it does occur
it is more likely to occur in thinned or perforated corneas. The
AC should be lled with viscoelastic through the paracentesis
prior to trephination. This can also be achieved by partial depth
trephination, a small scratch down incision, and deepening of
the AC with viscoelastic. Signicant iris damage during trephination can be repaired using 100 polypropylene suture.
Iridodialysis may occur during removal of an embedded IOL.
Damage to the anterior lens capsule is a serious intraoperative
complication that must be identied and treated immediately.
If the capsule has been opened then a combined cataract extraction with IOL placement must be performed in addition to the
PK (triple procedure).
AC Hemorrhage
This most commonly occurs in surgery on inflamed or perforated
eyes from iris vessel leakage. Slight bleeding usually stops spontaneously with closure of the eye and return of adequate IOP.
Closed-loop IOLs must be removed with care to avoid bleeding,
and when complete removal is impossible, it is safer to either
leave small pieces of the haptics in place or to cut the optic from
the haptic and slide the haptic through the synechial sheaths to
lower the risk of bleeding or iridodialysis. If the hemorrhage
persists in the presence of an adequate IOP, then it may need to
be controlled using cautery, compression with viscoelastic, or tamponade with Weck sponges soaked with epinephrine 1:1000.
Suprachoroidal Expulsive Hemorrhage
SECTION 6
The one nontechnical and most visually devastating complication during PK is suprachoroidal expulsive hemorrhage, which
may complicate 0.451.1% of operations.22,23 Eye-related risk
factors include increased axial length, the presence of an AC
IOL, and glaucoma. Patient-related risk factors include generalized
atherosclerosis, advanced age, hypertension, and intraoperative
tachycardia. The use of retrobulbar anesthesia (as opposed to
general anesthesia) is also associated with an increased risk of
suprachoroidal expulsive hemorrhage.23 Preoperative reduction
of the IOP and reduction of aqueous and vitreous volume with
decreased choroidal blood flow may help decrease the chance of
an expulsive hemorrhage.
POSTOPERATIVE COMPLICATIONS
Early Complications
The proper diagnosis and management of complications in the
critical early postoperative period is important as this may help to
prevent early graft failure and improve long-term graft survival.
Wound leaks
840
The prevention of wound leaks begins with good wound construction. Good donorrecipient apposition, symmetric tissue
distribution, and proper suture placement are all essential for
the construction of a watertight wound. When suturing begins,
the proper placement of the four cardinal sutures is essential. If
distribution of the donor tissue is not symmetric, the cardinal
sutures should be replaced until symmetric tissue distribution
is achieved.
During the early postoperative period, low IOP and/or the
presence of a shallow or flat AC suggests the possibility of a
wound leak. The best way to prevent wound leaks is to ensure
meticulous wound apposition at the end of the procedure. It is
best to test for small leaks using Seidels test, in which concentrated 2% fluorescein eyedrops are placed on the surface of
the cornea and the eye is examined under cobalt blue light. A
shallow AC may suggest the presence of a wound leak, however,
if the chamber remains formed and the leak controlled with
either a pressure patch or bandage contact lens, then resuturing
is not necessary. If nonsurgical attempts to seal the leak fail
after 2448 h, surgical repair is recommended. If the AC is flat
in the presence of a wound leak then resuturing of the wound is
recommended. Prolonged contact between the donor cornea and
the iris, lens, or IOL may result in irreversible endothelial cell
loss and consequently early graft failure.
A loose suture serves no useful purpose and should be
removed to reduce the risk of infection. If broken sutures occur
during the rst few weeks after grafting they should be replaced
using topical anesthesia at an operating microscope.
Persistent epithelial defects

The prevention of epithelial defects should be addressed preoperatively. Conditions such as dry eye, blepharitis, trichiasis,
ectropion, entropion, or lid mal-apposition should be corrected
prior to performing PK surgery. In patients in whom problems
with reepithelialization are anticipated, the use of a donor cornea
with good epithelium is essential. Additionally, careful handling
of the donor cornea intraoperatively is imperative to avoid damaging the epithelium. Good wound apposition and prevention of
an overriding edge leads to better tear-lm distribution and a
reduced incidence of epithelial defects.
The donor epithelium sloughs off the donor button with
time, and a complete epithelial defect is often seen on the rst
postoperative day. Preoperative donor factors increasing the risk
for the presence of an epithelial defect on the rst postopeartive
day include a longer time interval from preservation to surgery,24
longer death-to-enucleation time,25 and diabetes in the donor.25
Reepithelialization and the maintenance of an intact epithelium is critical for postoperative wound healing, graft survival,
and protection against infection and melting. The treatment of
a persistent epithelial defect in the grafted patient should therefore be performed with a degree of urgency. A permanent or
temporary tarsorrhaphy early in the postoperative period is the
most powerful prophylactic and treatment modality available
for epithelial defects. Alternatively, botulinum A toxin injected
into the levator muscle to induce a complete ptosis, may help
reduce the severity and persistence of an epithelial defect.
Botulinum toxin often takes 24 days to produce a ptosis but
has advantages over tarsorrhaphy in that it facilitates easier
instillation of eyedrops and allows the surgeon to assess the
graft by simply lifting the upper eyelid in the postoperative setting. For small epithelial defects pressure patching may improve
epithelialization by decreasing eyelid movement over the
healing surface. However, topical medication must be withheld
during this time.
The use of preservative-free medication is recommended to
reduce the risk of epithelial toxicity if a persistent epithelial defect
is present. Topical corticosteroids may also be associated with
delayed epithelial healing and in the presence of a persistent
epithelial defect the frequency of their instillation may need to
be decreased.
Many causes of lamentary keratitis may be present in the
grafted patient. These include keratoconjunctivitis sicca, neurotrophic keratopathy, recurrent erosions, blepharospasm, and the
presence of sutures. The laments, which consist of abnormal
collections of dead epithelial cells and mucus, are often seen at
a suture site. Patients often experience foreign body sensation and
grittiness. Intensive lubrication or removal of the laments with
forceps is often warranted if the symptoms are severe. Topical

acetylcysteine may also be useful because of its mucolytic effect.
Suture-related complications
The postoperative course of a PK procedure is often complicated
by suture-related problems. These can involve events such as
suture loosening (8%), epithelial erosions over sutures (11%),
FIGURE 62.4. Loose interrupted sutures (a) before and (b) after fluorescein staining.
CHAPTER 62
FIGURE 62.3. This photograph shows marked vascularization across

the hostdonor interface. Two sutures, which may have been
responsible for the vascularization, were removed from this site prior
to taking the photograph.
sterile inltrates (9%), secondary infections, corneal ulcerations

(3%), wound dehiscence with spontaneous wound leakage (2%),
and allograft rejection.26 Every patient is at risk for encountering suture-related complications immediately after PK. This risk
only comes to an end when all the sutures are removed. As a
general approach, suture removal at 1 year in vascularized recipients, and at 18 months in all other cases constitutes a good
policy. Wound dehiscence and/or large changes in keratometry
may still occur after this period. Any suture that is broken, loose,
or associated with stromal vascularization across the wound
(Fig. 62.3) should be removed immediately to lower the risk of
rejection. If interrupted sutures are used, the complication of a
loose suture poses little difculty as that suture can be easily
removed (Figs 62.4a,b). However, if a continuous suture is used
it may not be possible to remove the suture early in the postoperative period, and suture rotation may be required to help
retighten a loose quadrant of the graft (Figs 62.5a,b). The
alternative option is to bring the patient back to the operating
room for resuturing of the graft.
In children, earlier suture removal is advocated.27 In vascularized recipients, topical corticosteroids should be tapered to a
maximum of once daily for 1 year, and stopped after 15 months,
with a plan to remove all remaining sutures at 18 months. Early
removal of the sutures decreases the incidence of suture-related
problems. Topical corticosteroid therapy delays wound healing
and should therefore be discontinued a few months prior to suture
removal to allow maximum wound strength to be achieved.
Unfortunately, early cessation of corticosteroid therapy may give
the patient an increased susceptibility to immune rejection.
FIGURE 62.5. Loose continuous sutures (a) before and (b) after fluorescein staining.
841
Suture-related infection
Gram-positive organisms are the most common organisms
implicated in suture-related infections in the grafted patient, but
Gram-negative organisms or fungi may also be the causative
organisms.28 Suture-related infections, which may occur in 3%
of patients after PK,26 must be treated as an emergency as they
can lead to wound dehiscence, endophthalmitis, and corneal
scarring and thinning, all of which can contribute to eventual
graft failure. Treatment of a suture abscess begins with removal
and culture of the affected suture. Corneal scrapings for Gram
staining and culture should also be performed at this time to
help identify the causative organism. The patient should be
started on broad-spectrum fortied antibiotics such as ciprofloxacin, cefazolin, or gentamicin until the result of the culture
and antibiotic sensitivities are known. The use of topical
corticosteroids should be temporarily suspended in the early
stages of treatment. Once the infection is controlled, topical
corticosteroids may be recommenced cautiously.
Immune inltrates
Suture-related immune inltrates may mimic infectious inltrates. Some patients may mount an immunological reaction to
the suture material which results in inltrates along multiple
suture tracts. They are more commonly located on the host side
(recipient) of the grafthost interface, are usually not associated
with overlying epithelial defects and are often seen in patients

who have other risk factors for an immunologic reaction. They
can be treated with an intensied course of topical corticosteroids, beginning with drops every 2 h and ointment at night.
This complication occasionally necessitates eventual removal of
the sutures earlier than originally planned. The treatment may
be tapered over time depending on the patients response.
The development of discrete white dots in the donor epithelium in a 12 mm region central to the graft sutures was described
in 1980 by Kaye.29 The dots are not associated with staining
and the patient remains asymptomatic (Fig. 62.6). Their presence is not associated with rejection or infection. After suture
removal, the dots move centrally from their peripheral location
and gradually disappear over a period of 30 days.
Infectious crystalline keratopathy

This condition, which was rst reported in 1983 by Gorovoy
et al,30 is a distinctive clinical entity often seen in the grafted
patient (Fig. 62.7). It represents an inflammatory inltrate in
the corneal stroma associated with Gram-positive bacteria,
most notably Streptococcus pneumoniae, Gram-negative rods and
yeasts, as well as nontuberculous mycobacteria. This crystal-like
keratitis is indolent in nature, progressive and occurs beneath
an intact epithelium in the absence of clinically evident stromal
inflammation. It is associated with long-term corticosteroid use
and epithelial defects. Bacteria are thought to gain access to the
corneal stroma via epithelial ingrowth into a suture track or by
direct access through an epithelial defect. Administration of topical corticosteroids helps to protect the organisms from a marked
inflammatory response. Treatment involves administration of
an intensive topical fortied antibiotic regimen. Despite this,
repeat PK is often required in these patients.
Endophthalmitis
SECTION 6
FIGURE 62.6. There are Kaye dots present peripherally in the donor
cornea.
This is a potentially devastating complication after PK. Endophthalmitis rates as high as 0.77% were reported until the late
1980s by a retrospective study based on nationwide patient
registers,31 whereas a more recent study32 noted that the overall
incidence of post-PK endophthalmitis from 1972 to 2003 was
0.38%, but the trend in the last 3 years was for a reduction in
this rate to 0.2%. A high organism-proven endophthalmitis rate
of 93% has been reported,33 which may reflect the higher
bacterial load infections with post-PK endophthalmitis. Most
etiologic pathogens in that study were Gram-positive cocci. Of
these, Streptococcus species were the most frequent pathogens.
In the same study, Gram-negative bacteria were 100% sensitive
to both ceftazidime and gentamicin. Hence, a well-developed
antibiotic regimen should include vancomycin, which has excellent coverage for Gram-positive, and specically Streptococcus,
species and either ceftazidime or gentamicin for Gram-negative
bacteria. There has been a gradual increase over the last few years
in the percentage of cases of endophthalmitis post-PK caused by
fungal infection.34 However this apparent emergence of fungal
post-PK infection as a cause of 20% of cases of endophthalmitis
may be the result of the reduction of bacterial post-PK endophthalmitis that followed the widespread use of donor corneal
preservation medium supplemented with gentamicin and
streptomycin.
It must be reemphasized that the rate of endophthalmitis after
PK is higher than the rate quoted for cataract extraction alone.
Aseptic technique is therefore of utmost importance when
performing PK.
Elevated IOP
842
FIGURE 62.7. This photograph demonstrates the crystal-like or

snowflake-like opacities in the grafted cornea, which are a hallmark of
infectious crystalline keratopathy.
Early postoperative glaucoma, rst described in 1969 by Irvine

and Kaufmann,35 is important because elevated IOP may result
in endothelial cell damage. It is important to avoid retained vis-

coelastic in the AC at the end of the procedure and to reinflate
the eye with BSS. During the procedure, as soon as the AC is
able to maintain its depth, BSS should be used rather than
viscoelastic to reinflate the eye prior to suture tying. Other risk
factors for postoperative glaucoma include tight sutures, larger
trephine sizes, same-size donorhost trephination, intraocular
inflammation and anterior synechiae causing angle closure. If
peripheral anterior synechiae are present at the time of surgery,
it is prudent to try to break these in order to improve aqueous
outflow postoperatively. A steroid response can also be the cause
of early glaucoma, and topical antiglaucomatous agents such as
a beta-blocker, should be used judiciously to avoid long-standing
high pressure.
Anterior synechiae
Patients with anterior synechiae show a higher prevalence of
graft edema as well as secondary angle-closure glaucoma which
is often difcult to manage. Anterior synechiae may also expose
the endothelial cells to blood vessels, which may increase the
risk of graft rejection. The risk of this complication increases
with larger graft diameters and therefore peripheral iridectomies
may be required in such cases. Postoperative dilatation does increase AC depth, but it should be used with caution in patients
with keratoconus, because permanent mydriasis can result. The
aggressive control of postoperative inflammation is also
essential for the prevention of synechiae formation.
Pupillary block
UrretsZavalia pupil
UrretsZavalia syndrome,36 rst described in 1963, is essentially the presence of mydriasis occurring a few days after PK for
keratoconus. The syndrome includes iris stromal atrophy,
scattered pigment granules over the lens capsule and corneal
endothelium, ectropion uvea, and secondary glaucoma with
multiple posterior synechiae (Fig. 62.8). The incidence of this
complication in patients with keratoconus undergoing PK may
be ~6% of cases.37 The exact cause for this syndrome remains
unknown.
LATE POSTOPERATIVE COMPLICATIONS

Immunologic Rejection
Despite the corneas relative immunologic privilege, allograft
rejection remains the most signicant cause of graft failure
worldwide.38 The factor most commonly associated with an
increased risk of allograft rejection is corneal vascularization.
These vessels probably provide a route for allogeneic antigens to
reach the lymphatic tissue. Previous graft failure also increases
the probability of an immunologic rejection episode in repeat
PK.3,6 Large graft size and eccentric grafts (Fig. 62.9) may also
CHAPTER 62
The presence of a shallow AC and an intact securely closed

wound, conrmed by Seidels test, suggests the presence of pupillary block or a choroidal detachment. Pupillary block is usually
associated with an elevated IOP, whereas choroidal detachment
is usually associated with a low IOP. Vitreous protruding through
the pupil or the peripheral iridectomy, or the presence of posterior synechiae helps to conrm a diagnosis of pupillary block.
Fundal examination or B-scan ultrasonography help to diagnose
choroidal detachment. The appropriate medical treatment of
pupillary block is vigorous dilation of the pupil with mydriatic
and cycloplegic drops. Topical antiglaucomatous and corticosteroid medications are also frequently required. If this proves
unsuccessful, a peripheral iridectomy should be performed either
with a laser or surgically to reduce the risk of anterior synechiae
formation.
FIGURE 62.9. This gure shows an edematous failed graft, which is

located eccentrically.
FIGURE 62.8. This photograph shows evidence of UrretsZavalia

syndrome. Ectropion uveae, pigment deposition on the lens surface,
and an irregularly shaped dilated pupil are demonstrated. The graft
was intentionally decentered inferonasally because this patient had
keratoconus which was most marked in this region.
FIGURE 62.10. Faint subepithelial inltrates in a case of epithelial

rejection, which is similar to that seen in adenoviral keratitis.
843
SECTION 6
FIGURE 62.11. This photograph shows a Khoudadoust line in the

superior half of the grafted cornea. There is also marked corneal
edema present in the graft due to the endothelial rejection.
844
increase the risk of immunologic rejection as the grafthost

interface is nearer to the corneal limbus.12
Several types of corneal graft rejection have been described:
epithelial rejection lines, subepithelial inltrates, stromal rejection haze, endothelial rejection lines, and diffuse endothelial
rejection. Epithelial rejection is a benign situation characterized
by an elevated epithelial rejection line that stains with fluorescein or rose bengal. It is often asymptomatic and responds well
to mild steroid therapy, as do the subepithelial white inltrates
that are thought to be an immune reaction because they are
seen only in the donor tissue. These subepithelial inltrates
resemble those seen in adenovirus keratitis (Fig. 62.10). Stromal
rejection occurs very infrequently. On the contrary, an endothelial rejection line, referred to as the Khodadoust line,39 is present
and usually symptomatic at one time or another in up to 20%
of patients.16 During endothelial rejection, it is possible to directly
visualize linear or multifocal depositis of leucocytes adhering to
the endothelium and the loss of clarity resulting from edema
(Fig. 62.11). If left untreated, the endothelial rejection line
usually proceeds across the donor endothelium from a point of
origin at the graft wound, leaving damaged endothelium behind
it. It is often associated with an AC reaction.
Not all patients who experience an allograft rejection show a
greater than expected endothelial cell loss. The two main risk
factors for a high endothelial cell decline are a delay in diagnosis
and a recipient age of more than 60 years.40 Even if the delay
only exceeds 1 day, the outcome is signicantly worse. This
nding highlights the importance of patient education regarding
the symptoms of rejection and requirement for prompt attendance to the eye department if anything seems wrong.
Increasing corneal thickness measured by pachymetry is a
very useful way to document progressing graft failure. Ultrasound pachymetry permits reproducible measurement of
corneal thickness with a 1% precision in patients with PK.41 At
each postoperative time point up to 5 years, subsequent graft
survival is signicantly lower in patients with increased graft
thickness as compared with patients with normal or decreased
graft thickness.42 An increase in graft thickness above the upper
normal limit for the considered postoperative time point may
indicate a graft complication, even when slit-lamp examination
reveals no complications. These patients are at a greater risk of
failure and should be closely observed during follow-up. Another
study of high-risk patients showed that if the graft central
thickness exceeds 590 mm after 6 months, the overall failure
risk is increased.43 Naacke et al44 showed there was a signicant
difference in corneal thickness between patients with reversible
and irreversible rejection episodes. Therefore, when a rejection

episode occurs, the treatment should be more aggressive if the
graft central thickness measures more than 700 mm.
Intensive topical steroids should immediately be commenced
for newly documented corneal graft failures. If the corneal thickness is unchanged or becomes worse after a few weeks with
such treatment, it is unlikely that the graft will clear although
occasional exceptions do occur.
Treatment of corneal allograft immunologic rejection is more
likely to be effective if treatment is commenced promptly. The
success rate of rejection treatment is usually ~50%.44,45 Regarding
rejection reversibility, patients with keratoconus or Fuchs dystrophy have been shown to have the best prognosis and patients
with PBK have the worst.44 Donor variables appear to have no
influence on rejection reversibility. Patient education about
potential rejection symptoms again remains vitally important.
A decrease in visual acuity, irritation, redness, photophobia, and
tearing are the most commonly reported symptoms. Suspected
endothelial rejection episodes must be treated aggressively.
Immunosuppression, particularly with corticosteroids, is widely
accepted to be effective in the prevention and treatment of
rejection episodes. There is considerable debate on the optimal
administration route, dosage, and formulation. Systemic immunosuppression may be prescribed in high-risk cases as a means
of prophylaxis. Additionally, the indenite use of topical corticosteroid treatment is more commonly used in high-risk cases
compared to low-risk grafts.46 The treatment of corneal graft
rejection varies depending on the clinical presentation. Corneal
surgeons who were surveyed in the United Kingdom46 tended to
treat epithelial and stromal rejection on an outpatient basis
with frequent topical steroids. However, where there is evidence
of endothelial rejection, management strategy is more aggressive.
Up to 42% of surgeons prefer to admit these patients into the
hospital to treat them with topical prednisolone acetate 1% every
hour around the clock. In addition, systemic steroids are preferred over subconjunctival steroids. Immunosuppressants other
than steroids (cyclosporine A and azathioprine) are used infrequently. This regimen is then tapered slowly over a period of
weeks to months depending on the patients response. Several
clinicians give corticosteroids by a route other than topical at
the time of diagnosis of an immunologic endothelial rejection,
either an intravenous dose of methylprednisolone, oral prednisolone for a few days, or subconjunctival dexamethasone. A
randomized control trial by Hudde et al47 involving 36 patients
concluded that treatment of rejection with systemic steroids did
not offer signicant benet over local (topical and subconjunctival) steroids alone. This contradicts the ndings of a prospec-
FIGURE 62.12. This shows a markedly thickened graft with severe

epithelial and stromal edema.
Nonimmunologic Endothelial Failure

Graft endothelial failure can be the result of causes other than
allograft rejection. Early primary donor failure can result from
poor donor quality. Intraoperative trauma to the donor endothelium and a prolonged flat chamber in the immediate postoperative period can also result in irreversible corneal graft edema.
In the late postoperative period, some grafts fail due to the
natural attrition of endothelial cells that occur with age. A
decrease in endothelial cell density (ECD) affects the ability of
the endothelium to maintain its primary function, as is evidenced by an increase in corneal thickness as the ECD
decreases. This eventually leads to endothelial decompensation
and a hazy graft (Fig. 62.12), typically when the cell density
reaches 333500 cells/mm2.50,51 Under normal conditions, the
adult human cornea loses endothelial cells at a rate of 0.6% per
year.52 After intraocular surgery the rate of endothelial cell loss
is accelerated.53 Uncomplicated PK results in an endothelial cell
loss rate of 4.2% per year 510 years postkeratoplasty.4 This is
seven times the annual cell loss rate in normal adult human
eyes and may explain why late endothelial failure is the predominant cause of graft failure after the rst ve postoperative
years.54 One model of cell loss, described by Armitage et al,55
explained that if the critical density is taken to be
500 cells/mm2, it can be predicted that corneas with initial cell
densities lower than 2000 cells/mm2 could reach the critical
density, and therefore fail, in less than 20 years. With initial
densities above 2500 cells/mm2 the grafts should remain viable
for at least 30 years.
Postoperative Astigmatism
High degrees of regular and irregular astigmatism, frequently
associated with signicant myopia, hyperopia, and anisometropia
can lead to poor functional vision despite a clear corneal graft.
Although spectacle correction is the simplest method of addressing postoperative refractive error, contact lenses often provide
superior visual acuity and are frequently required in eyes with
moderate and severe anisometropia (> 3 D). Unfortunately,
contact lenses are often difcult to t, and they may induce
peripheral corneal neovascularization, leading to graft rejection
and failure. Factors contributing to signicant postoperative astigmatism in PK include trephination diameter disparity between
donor and recipient, central versus eccentric trephination, disparity between the donor and recipient tissue thickness, suturing
technique, recipient wound healing, timing of suture removal,
and wound disparity or override. Eccentric trephination is associated with severe astigmatism,56 and therefore a well-centered
graft is often the aim in PK. Many suturing techniques have
been described to help reduce the degree of postoperative astigmatism. Despite this, refractive unpredictability after PK is
extremely common with most series documenting mean cylinder
of 45 D and signicant anisometropia.57 Surgery may be required
to correct postoperative astigmatism. This should generally be
deferred until at least 12 months after PK and 3 months after

suture removal. Prior rejection episodes should be noted, and
the patient should be stable on minimal or no immunosuppressive agents. Relaxing incisions and compression sutures can
correct an average of 45 D of astigmatism.58 Wilkins et al59
evaluated the refractive effect of a standardized incision (paired
600 mm depth, 60 arc, 6.0 mm apart keratotomies), and found
the astigmatic effect to be proportional to the magnitude of the
preoperative cylinder. Therefore, arcuate nomograms for congenital astigmatism may be less useful in the management of
astigmatism in postkeratoplasty eyes. Photorefractive keratectomy has been used after corneal transplantation since the early
1990s.60 Unfortunately the use of PRK in PK eyes has been
associated with signicant regression, haze, and scarring. The
adjuvant use of mitomycin C 0.02% (0.2 mg/mL) is a promising
new method of scar prevention in high-risk eyes undergoing
PRK. Early reports of PRK with mitomycin C after PK have been
very positive.61,62 Because of the early difculties after PRK in
PK eyes, LASIK has become a popular modality for correcting
refractive error after corneal transplantation. The use of LASIK
after PK was rst reported by Arenas and Maglione in 1997.63
LASIK offers several advantages over PRK in the treatment of
myopia and astigmatism. These advantages include rapid visual
rehabilitation, decreased stromal scarring, minimal regression,
and the ability to treat a greater range of refractive disorders.6466
Most surgeons wait at least 1 year after PK and 36 months after
last suture removal or other refractive procedure prior to performing LASIK, to allow maximum wound stability and refractive
and topographical stability.6770 The corneal flap diameter should
be slightly larger than the diameter of the graft to allow the flap
to drape the wound which results in better wound apposition.71
The contraindications to performing LASIK after PK for residual
refractive errors/anisometropia include marked peripheral corneal
vascularization, thin host tissue, wound ectasia, a signicant graft
override or malapposition, and minimum central corneal thickness of less than 500 mm.72 Typical LASIK and postkeratoplasty
complications can occur after the laser procedure. Dehiscence of
the grafthost junction is rare but has been reported in an eye
that had LASIK 3 years after PK.73 If cataract is present, phacoemulsication with an appropriate lens implant may be a better
method of addressing anisometropia or ametropia.
Traumatic Globe Rupture

This complication of PK surgery is a life-long postoperative
concern. Pettinelli et al74 reported on a series of six patients who
sustained traumatic PK wound dehiscence 1019 years after
initial surgery. A study from New Zealand75 reported a traumatic rupture rate of 5.8% of which one-third occurred within
the rst postoperative month. Others have found lower rates of
rupture of 2.5%.76,77 The highest-risk period is the rst postoperative month because almost all of the wound strength is
derived from the sutures. The second most important high-risk
period is after removal of the graft sutures because the eye is
entirely dependent on wound healing. The use of protective eyewear during daytime and eye shields during sleep may therefore
be advised during these high-risk periods. The cornea never
regains its preoperative tensile strength, and lifelong caution
must be exercised. Some activities such as contact sports may
be unsafe to pursue following PK.
CHAPTER 62
tive trial by Hill et al48 that showed the superiority of using

systemic treatment over local steroid.
To prevent rejection in high-risk patients or to treat patients
with recurrent immunologic rejection episodes, topical 1%
cyclosporine may be an excellent alternative.49 It must be given
for several months up to 1 or 2 years.
REFERENCES
1. Frequently asked questions. Eye Bank
Association of America web site. Available:
http://www.restoresight.org/general/faqs.htm.
2. McNeill JI: Penetrating keratoplasty:
preoperative considerations; indications
and outcomes. In: Krachmer JH, Mannis MJ,

Holland EJ, eds. Cornea. St Louis, MO:
Mosby; 1997:3.
3. Price FW Jr, Whitson WE, Collins KS,
Marks RG: Five-year corneal graft
survival. A large, single-center patient

cohort. Arch Ophthalmol 1993;
111:799805.
845
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
SECTION 6
17.
18.
19.
20.
21.
22.
846

105:18551865.
Patel NP, Kim T, Rapuano CJ, et al:
Indications for and outcomes of repeat
penetrating keratoplasty, 19891995.
Thompson RW Jr, Price MO, Bowers PJ,
Price FW Jr: Long-term graft survival after
2003; 110:13961402.
Inoue K, Amano S, Oshika T, et al: A
10-year review of penetrating keratoplasty.
Williams KA, Muelberg SM, Bartlett CM,
et al, eds: The Australian Corneal Graft
Registry: 1999 report. Adelaide: Snap
Printing; 2000.
Streilein JW: Immunobiology and
immunopathology of corneal transplantation.
Chem Immunol 1999; 73:186206.
The collaborative corneal transplantation
studies (CCTS): Effectiveness of
histocompatibility matching in high-risk
corneal transplantation. The Collaborative
110:13921403.
Weisbrod DJ, Sit M, Naor J, Slomovic AR:
Outcomes of repeat penetrating
keratoplasty and risk factors for graft
failure. Cornea 2003; 22:429434.
Price FW Jr, Whitson WE, Johns S,
Gonzales JS: Risk factors for corneal graft
failure. J Refract Surg 1996; 12:134143;
discussion 143147.
Sit M, Weisbrod DJ, Naor J, Slomovic AR:
Corneal graft outcome study. Cornea 2001;
20:129133.
Price MO, Thompson RW Jr, Price FW Jr:
Risk factors for various causes of failure in
initial corneal grafts. Arch Ophthalmol
2003; 121:10871092.
Bersudsky V, Blum-Hareuveni T, Rehany U,
Rumelt S: The prole of repeated corneal
108:461469.
Boisjoly HM, Bernard PM, Dube I, et al:
Effect of factors unrelated to tissue
matching on corneal transplant endothelial
rejection. Am J Ophthalmol 1989;
107:647654.
Maguire MG, Stark WJ, Gottsch JD, et al:
Risk factors for corneal graft failure and
rejection in the collaborative corneal
transplantation studies. Collaborative
Group. Ophthalmology 1994;
101:15361547.
Cobo LM, Coster DJ, Rice NS, Jones BR:
Prognosis and management of corneal
transplantation for herpetic keratitis. Arch
Ophthalmol 1980; 98:17551759.
van Rooij J, Rijneveld WJ, Remeijer L, et al:
Effect of oral acyclovir after penetrating
keratoplasty for herpetic keratitis: a
placebo-controlled multicenter trial.
Ophthalmology 2003; 110:19161919;
discussion 1919.
Leger F, Mortemousque B, Morel D, et al:
Penetrating corneal transplant with
inadvertent corneal button inversion. Am J
Ophthalmol 2003; 135:9193.
Ohlrich S, Hirst LW, Harrison M, et al:
Inadvertent corneal button inversion during
11:586588.
Price FW Jr, Whitson WE, Ahad KA,
Tavakkoli H: Suprachoroidal hemorrhage in
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.

1994; 25:521525.
Ingraham HJ, Donnenfeld ED, Perry HD:
Massive suprachoroidal hemorrhage in
penetrating keratoplasty. Am J Ophthalmol
1989; 108:670675.
Kim T, Palay DA, Lynn M: Donor factors
associated with epithelial defects after
15:451456.
Chou L, Cohen EJ, Laibson PR,
Rapuano CJ: Factors associated with
epithelial defects after penetrating
keratoplasty. Ophthalmic Surg 1994;
25:700703.
Christo CG, van Rooij J, Geerards AJ, et al:
Suture-related complications following
keratoplasty: a 5-year retrospective study.
Cornea 2001; 20:816819.
Dana MR, Schaumberg DA, Moyes AL, et al:
Outcome of penetrating keratoplasty after
ocular trauma in children. Arch Ophthalmol
1995; 113:15031507.
Sonavane A, Sharma S, Gangopadhyay N,
Bansal AK: Clinico-microbiological
correlation of suture-related graft infection
Ophthalmol 2003; 135:8991.
Kaye DB: Epithelial response in penetrating
89:381387.
Gorovoy MS, Stern GA, Hood CI, Allen C:
Intrastromal noninflammatory bacterial
colonization of a corneal graft. Arch
Ophthalmol 1983; 101:17491752.
Aiello LP, Javitt JC, Canner JK: National
outcomes of penetrating keratoplasty.
Risks of endophthalmitis and retinal
detachment. Arch Ophthalmol 1993;
111:509513.
Taban M, Behrens A, Newcomb RL, et al:
Incidence of acute endophthalmitis
following penetrating keratoplasty: a
systematic review. Arch Ophthalmol 2005;
123:605609.
Kunimoto DY, Tasman W, Rapuano C, et al:
Endophthalmitis after penetrating
keratoplasty: microbiologic spectrum and
susceptibility of isolates. Am J Ophthalmol
2004; 137:343345.
Hassan SS, Wilhelmus KR: Eye-banking
risk factors for fungal endophthalmitis
compared with bacterial endophthalmitis
after corneal transplantation. Am J
Ophthalmol 2005; 139:685690.
Irvine AR, Kaufman HE: Intraolar pressure
Ophthalmol 1969; 68:835844.
Urrets-Zavalia A: Fixed dilated pupil, iris
atrophy and secondary glaucoma, a
distinct clinical entity following keratoplasty
in keratoconus. Am J Ophthalmology 1963;
56:257265.
Gasset AR: Fixed dilated pupil following
penetrating keratoplasty in keratoconus
(Castroviejo syndrome). Ann Ophthalmol
1977; 9:623626.
Williams KA, Roder D, Esterman A, et al:
Factors predictive of corneal graft survival.
Report from the Australian Corneal Graft
Registry. Ophthalmology 1992; 99:403414.
Khodadoust AA, Silverstein AM:
Transplantation and rejection of individual
cell layers of the cornea. Invest Ophthalmol
1969; 8:180195.
Claerhout I, Beele H, De Bacquer D,
Kestelyn P: Factors influencing the decline
in endothelial cell density after corneal
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
allograft rejection. Invest Ophthalmol Vis

Sci 2003; 44:47474752.
Touzeau O, Allouch C, Borderie V, et al:
Precision and reliability of Orbscan and
ultrasonic pachymetry. J Fr Ophtalmol
2001; 24:912921.
Borderie VM, Touzeau O, Bourcier T, et al:
Outcome of graft central thickness after
2005; 112:626633.
McDonnell PJ, Enger C, Stark WJ,
Stulting RD: Corneal thickness changes
after high-risk penetrating keratoplasty.
Collaborative Corneal Transplantation
Study Group. Arch Ophthalmol 1993;
111:13741381.
Naacke HG, Borderie VM, Bourcier T, et al:
Outcome of corneal transplantation
rejection. Cornea 2001; 20:350353.
Alldredge OC, Krachmer JH: Clinical types
of corneal transplant rejection. Their
manifestations, frequency, preoperative
correlates, and treatment. Arch Ophthalmol
1981; 99:599604.
Koay PY, Lee WH, Figueiredo FC: Opinions
on risk factors and management of corneal
graft rejection in the United kingdom.
Cornea 2005; 24:292296.
Hudde T, Minassian DC, Larkin DF:
Randomised controlled trial of
corticosteroid regimens in endothelial
corneal allograft rejection. Br J Ophthalmol
1999; 83:13481352.
Hill JC, Maske R, Watson P:
Corticosteroids in corneal graft rejection.
Oral versus single pulse therapy.
Inoue K, Amano S, Kimura C, et al:
Long-term effects of topical cyclosporine A
treatment after penetrating keratoplasty.
Bourne WM: Cellular changes in
transplanted human corneas. Cornea 2001;
20:560569.
Vail A, Gore SM, Bradley BA, et al:
Conclusions of the corneal transplant
follow up study. Collaborating Surgeons.
Bourne WM, Nelson LR, Hodge DO:
Central corneal endothelial cell changes
over a ten-year period. Invest Ophthalmol
Vis Sci 1997; 38:779782.
Ambrose VM, Walters RF, Batterbury M,
et al: Long-term endothelial cell loss and
breakdown of the bloodaqueous barrier in
cataract surgery. J Cataract Refract Surg
1991; 17:622627.
Patel SV, Hodge DO, Bourne WM: Corneal
endothelium and postoperative outcomes
15 years after penetrating keratoplasty. Am
J Ophthalmol: 2005; 139:311319.
Armitage WJ, Dick AD, Bourne WM:
Predicting endothelial cell loss and longterm corneal graft survival. Invest
van Rij G, Cornell FM, Waring GO 3rd,
et al: Postoperative astigmatism after
central vs eccentric penetrating
keratoplasties. Am J Ophthalmol 1985;
99:317320.
Clinch TE, Thompson HW, Gardner BP, et al:
An adjustable double running suture
technique for keratoplasty. Am J Ophthalmol
1993; 116:201206.
Hardten DR, Lindstrom RL: Surgical
correction of refractive errors after
penetrating keratoplasty. Int Ophthalmol
Clin 1997; 37:135.
66.
67.
68.
69.
70.
71.
lamellar keratoplasty and penetrating

23:11141118.
Zaldivar R, Davidorf J, Oscherow S: LASIK
for myopia and astigmatism after
penetrating keratoplasty. J Refract Surg
1997; 13:501502.
Barraquer CC, Rodriguez-Barraquer T:
Five-year results of laser in-situ
keratomileusis (LASIK) after penetrating
Buzard K, Febbraro JL, Fundingsland BR:
Laser in situ keratomileusis for the
correction of residual ametropia after
penetrating keratoplasty. J Cataract Refract
Surg 2004; 30:10061013.
Hardten DR, Chittcharus A, Lindstrom RL:
Long term analysis of LASIK for the
correction of refractive errors after
23:479489.
Nassaralla BR, Nassaralla JJ: Laser in situ
keratomileusis after penetrating keratoplasty.
J Refract Surg 2000; 16:431437.
Donnenfeld ED, Kornstein HS, Amin A, et al:
Laser in situ keratomileusis for correction of
myopia and astigmatism after penetrating
106:19661974; discussion 19741975.
72. Lam DS, Leung AT, Wu JT, et al: How long

should one wait to perform LASIK after
PKP? J Cataract Refract Surg 1998;
24:67.
73. Ranchod TM, McLeod SD: Wound
dehiscence in a patient with keratoconus
after penetrating keratoplasty and LASIK.
74. Pettinelli DJ, Starr CE, Stark WJ: Late
traumatic corneal wound dehiscence after
2005; 123:853856.
75. Elder MJ, Stack RR: Globe rupture
following penetrating keratoplasty: how
often, why, and what can we do to prevent
it? Cornea 2004; 23:776780.
76. Tseng SH, Lin SC, Chen FK: Traumatic
wound dehiscence after penetrating
keratoplasty: clinical features and outcome
in 21 cases. Cornea 1999; 18:553558.
77. Rehany U, Rumelt S: Ocular trauma
following penetrating keratoplasty:
incidence, outcome, and postoperative
recommendations. Arch Ophthalmol 1998;
116:12821286.
CHAPTER 62
59. Wilkins MR, Mehta JS, Larkin DF:

Standardized arcuate keratotomy for
postkeratoplasty astigmatism. J Cataract
Refract Surg 2005; 31:297301.
60. Campos M, Hertzog L, Garbus J, et al:
Photorefractive keratectomy for severe
postkeratoplasty astigmatism. Am J
Ophthalmol 1992; 114:429436.
61. Solomon R, Donnenfeld ED, Thimons J, et al:
Hyperopic photorefractive keratectomy
with adjunctive topical mitomycin C for
refractive error after penetrating
keratoplasty for keratoconus. Eye Contact
Lens 2004; 30:156158.
62. Solomon R, Donnenfeld ED, Perry HD:
Photorefractive keratectomy with mitomycin
C for the management of a LASIK flap
complication following a penetrating
63. Arenas E, Maglione A: Laser in situ
keratomileusis for astigmatism and myopia
after penetrating keratoplasty. J Refract
Surg 1997; 13:2732.
64. Lazzaro DR, Haight DH, Belmont SC, et al:
Excimer laser keratectomy for astigmatism
occurring after penetrating keratoplasty.
65. Parisi A, Salchow DJ, Zirm ME, Stieldorf C:
Laser in situ keratomileusis after automated
847
CHAPTER
63
Excimer Laser Phototherapeutic Keratectomy

Dimitri T. Azar, Jose de la Cruz, Robert P. Selkin, and Walter J. Stark
The argon fluorine (ArF) excimer laser emits high-energy ultraviolet (UV) radiation at a wavelength of 193 nm. It has demonstrated the ability to treat anterior opacities and irregularities of
the cornea with submicron precision. The use of the ArF excimer
laser to treat supercial corneal diseases and smooth corneal
surface irregularities is called phototherapeutic keratectomy
(PTK), and this laser was rst introduced in 1983 to perform
surface ablations of corneal tissue. In 1988, investigational protocols began to determine the efcacy of PTK in humans. Subsequently, in 1995, the Food and Drug Administration (FDA)
approved PTK for the VISX (Santa Clara, CA) and Summit
(Waltham, MA) excimer lasers. The unprecedented ability to
ablate corneal tissue without damage to adjacent, untreated
tissue provides the corneal surgeon with a powerful tool in the
treatment of anterior corneal diseases or irregularities. The
mainstay of treating supercial corneal disorders has been to
use surgical techniques such as supercial keratectomy, lamellar
keratoplasty, and penetrating keratoplasty. These procedures
may be avoided in many cases with the use of PTK.1 Precise
removal of epithelium, Bowmans membrane, and anterior
stroma provides a smooth template for reepithelialization.
Compared with the more irregular tissue edges produced by
diamond and steel blades, histologic specimens of PTK-treated
corneas show a smooth boundary between ablated and
nonablated tissue.13
CORNEAL TRANSPARENCY AND WOUND

HEALING
The interactions between the excimer laser and corneal tissue
are important determinants of the success of PTK. The cavity of
the excimer laser contains a mixture of a rare gas (argon, xenon,
or krypton) and a halogen (fluoride, chloride, or bromide).
Unstable rare gas-halide molecules, such as ArF, are created with
a high-voltage current.4 When these molecules dissociate, they
emit high-energy UV light with a wavelength depending on the
mixture (193 nm for ArF). The energy of the photons emitted is
greater than the energy required to break organic intermolecular
bonds. This process results in ablation of the corneal tissue, a
process called ablative photodecomposition.5 Secondary fluorescence is valuable in PTK to differentiate stromal from epithelial
tissue removal. The latter fluoresces in the visible spectrum
(blue), whereas the former fluoresces in the UV spectrum.
The tight packing and small diameter of stromal collagen
brils account for the transparency and minimal degree of light
scattering of the cornea. Stromal brils are arranged in ~250
stacked lamellae. Each lamella functions as a light scatterer, and
the scattering of different lamellae causes destructive interference
with one another.6 The thinness of the cornea further helps to
decrease light scattering. The diameter of scleral collagen brils
is much greater than that of corneal brils, and the arrangement

of scleral collagen bers is more disorganized, which makes the
sclera more opaque. An increase in the packing distance
between brils, which occurs in corneal edema, also decreases
corneal transparency.7 A functioning endothelium is thus also
necessary for corneal transparency.
After excimer laser photoablation, epithelialization typically
occurs within the rst postoperative week.13,8 A thin, electrondense layer called a pseudomembrane forms adjacent to the
ablated area. This pseudomembrane is believed to act as a template for epithelialization. Migrating, hyperplastic epithelial cells
cover the membrane within 13 days; mitosis occurs and leads
to multiple epithelial cell layers. The pseudomembrane is also
believed to act as a barrier to water and thus prevent corneal edema.
Reformation of epithelial adhesion to subepithelial tissue
occurs in 13 months. Hemidesmosomes, basal laminae, and
anchoring brils appear at approximately the same time.9
Hemidesmosomes are electron-dense adhesion complexes along
the cell membranes of basal epithelial cells. Basal laminae are
between the basal epithelial cells and anchoring brils.
Anchoring brils contain type VII collagen and form interlacing
networks with collagen in the anterior stroma. Epithelial breakdown can occur with poor anchoring bril complex formation.
Analyses of epithelial basement membrane zones in monkey
and human corneas have shown discontinuities 18 months
after excimer ablation.1013 It is possible that adhesion complex
abnormalities exist permanently.14
Stromal wound healing occurs with repopulation of activated
keratocytes under the wound area, which begins to occur at
about the third postoperative week. Collagen, extracellular matrix,
and glycoproteins are produced.15 Histologically, an increase in
rough endoplasmic reticulum and cytoplasmic vacuoles is seen.
Newly deposited collagen and proteoglycans lead to haze that
contributes to light scattering, also a result of activated keratocytes in the wound.14,16,17 Posterior and wide ablation areas have
been found to result in increased amounts of haze formation.
Dogru and colleagues18 treated 45 eyes of 33 patients with
stromal dystrophies, band keratopathy, and corneal scars. Three
months after PTK they found signicant improvements in
corneal sensitivity, tear lm break up time, lipid layer interference grades, and conjunctival squamous metaplasia. With these
results the authors concluded that by improving the smoothness and clarity of the supercial cornea by PTK, the stability of
the tear lm improves along with better mucin production from
a healthier epithelium.19
PREOPERATIVE EVALUATION
Visual acuity is evaluated without correction and with manifest
refraction. Pinhole or potential acuity meter may provide an
849
TABLE 63.1. Laser Parameters for Phototherapeutic

Keratectomy
Fluence
160 10 mJ/cm2
Repetition rate
5 Hz
Ablation rate
0.20-0.35 mm per pulse
Ablation diameter
5.56 mm including a 0.5 mm

transition zone
Ablation depths
Epithelium
40 mm (default value) or as
determined by pachymetry
Stroma
Depth of scar or opacity

(postoperative corneal thickness
should be at least 250 mm)
SECTION 6
From Ashraf F, Azar D, Odrich M: Clinical results of PTK using the VISX Excimer
laser. In: Azar DT, Steinert RF, Stark WJ, (eds.): Excimer laser phototherapeutic
keratectomy. Baltimore, MD: Williams & Wilkins; 1997: 169172.
estimate of visual potential, although hard contact lens refraction is often necessary in patients with irregular corneal surfaces.
Pupil size in room and near dark lighting is recorded. Slit-lamp
biomicroscopy, dilated fundus examination, intraocular pressure
measurement, keratometry, and computerized video topography
are performed. Ultrasonic pachymetry as well as optical pachymetry are important to measure the depth of intended
treatment. The type of pathology and its ablation characteristics
are also important preoperative considerations.
The patient receives a sedative such as diazepam preoperatively, unless contraindicated, in an effort to reduce the anxiety
and provide an optimal environment for treatment. The patient
is placed supine in the surgical chair, and the patients head is
positioned under the microscope. The patients head is rested
comfortably in a Vac-Pac headrest. A blinking red light from the
microscope will be visible to the patient, and the patient is told
to xate on this light, even though it may fade during treatment.
Topical proparacaine (0.5%) or tetracaine is usually sufcient for
PTK. The microscope is focused at a high magnication (18)
and then decreased to a lower magnication (12) before it is
nally centered on the entrance pupil.
Calibration of the laser is crucial before each treatment
session. The calibration procedure differs among various lasers.
For the VISX laser, the overall operation of the excimer laser is
conrmed by ablating a plano and a standard lens (4 D) into a
polymethylmethacrylate (PMMA) test block using PTK and
photorefractive keratectomy (PRK) software, respectively. A
lensometer is used to read the lens, and after achieving the
predetermined optical power, the value for the corneal ablation
rate is determined using standard nomograms. The appropriate
value is automatically incorporated into the computer program.
Table 63.1 summarizes PTK laser parameters with the VISX
laser. Nitrogen gas flow is avoided during the PTK procedure
because of potential corneal desiccation.20
INDICATIONS AND CONTRAINDICATIONS
850
PTK indications include anterior basement membrane dystrophy, Bowmans membrane dystrophies such as ReisBcklers,
and stromal dystrophies including lattice, Schnyders, and granular. Anterior corneal scars resulting from trauma, surgery, or
previous infections can be treated. Salzmanns nodules and broblastic nodules in patients with keratoconus are indications.
Band keratopathy can often be treated adequately with ethylenediaminetetraacetic acid (EDTA) chelation, although the
excimer laser PTK remains an option. However, PTK often
leaves an irregular base because of the nonuniformity of the

calcium band across the cornea.2123 Irregular corneal surfaces
in patients after removal of the pterygium and climatic droplet
keratopathy are further indications for treatment.
Absolute contraindications include immunocompromised
patients and patients with uncontrolled uveitis, severe blepharitis,
lagophthalmos, or severe dry eye. Microbial keratitis, including
infectious crystalline keratopathy, should be avoided because of
the potential risk of spreading microorganisms.24,25 Patients with
a history of herpes simplex keratitis are best avoided whenever
possible because of the possibility of recurrence, which has been
documented by several investigators.2630 Patients with collagen
vascular disease, diabetes, exposure keratopathy, and neurotrophic corneas should not be treated because of the possibility of
wound healing problems. The corneal stromal thickness after
surgery must be at least 250 mm, which makes signicant
corneal thinning a contraindication. Hyperopic patients are not
ideal candidates for PTK of the central cornea, because removal
of corneal tissue centrally results in central flattening and
increased hyperopia.
The nondominant eye is treated rst in patients with bilateral corneal disease. The fellow eye can be treated 36 months
after the rst eye. Furthermore, if re-treatment is required, a
6-month waiting period is advisable.
SURGICAL TECHNIQUES
ELEVATED OPACITIES AND CORNEAL
DYSTROPHIES
PTK starts with removal of the epithelium, which can be performed with the laser or manually with a BardParker blade. If
Bowmans layer is smooth relative to the epithelium, removal of
the epithelium is performed manually. In patients with an irregular Bowman membrane, the epithelial surface serves as a natural
masking agent and is removed with the laser; that is, epithelium
in relatively elevated areas is selectively ablated while epithelium
in depressed areas serves to cover these depressions. Masking
agents are important to use after removal of the epithelium.
These fluids serve to cover depressions and expose peaks. They
absorb laser energy, thus protecting depressions in the corneal
surface, while exposed areas are ablated. One percent hydroxymethylcellulose, 0.5% tetracaine, or Tears Naturale II are of
medium viscosity and are useful in covering valleys while
exposing peaks. Kornmehl and associates recommend the use of
Tears Naturale II in one study.31 Healon and 2% methylcellulose
have a high viscosity and may partially cover peaks, whereas
low-viscosity fluids allow valleys to be partly exposed. More
than a single agent may be used, depending on the irregularity
encountered.
Elevated central corneal nodules, such as Salzmanns nodules
or nodules associated with keratoconus, are especially challenging to the PTK surgeon (Fig. 63.1). Many surgeons recommend
manual keratectomy with a blade before PTK. This technique is
extremely valuable when a tissue plane can be found, thus
leaving a smooth corneal surface. The excimer laser can also be
used to debulk a lesion in cases in which a tissue plane cannot
be found. Removal of the epithelium over the nodule, while
leaving the surrounding epithelium alone, serves to protect the
surrounding cornea while the nodule is being treated. Masking
agents are applied after the elevated nodule has been removed
in order to attain a smooth surface. PTK surgeons have demonstrated a high rate of improvement in visual acuity after treatment of Salzmanns nodules and keratoconus-associated
nodules (Figs 63.2 and 63.3).
Corneal dystrophies have traditionally been treated by lamellar or penetrating keratoplasty. Epithelial, basement membrane,

Localized epithelial debridement
Surface modulators in annular furrow
Step 1
From Azar DT, Jain S, Stark W: Phototherapeutic

keratectomy. In: Azar DT, ed. Refractive surgery.
Stanford, CT: Appleton & Lange; 1996:504, 513.
Step 2
Laser Ablation 30-40 mm
FIGURE 63.1. Surgical technique of PTK for

elevated corneal nodules. Steps 16 are
schematically illustrated. The elevated opacity
noted preoperatively has disappeared.
Additional surface modulators to cover peak
Step 3
Step 4
PTK Laser stromal ablation
PRK or PAK for residual

myopia or astigmatism
Step 5
Step 6
FIGURE 63.2. (a and b) Slit-lamp photographs

of a 70-year-old man with multiple elevated
Salzmanns nodules in the left eye resulting in a
visual acuity of 20/200.
From Azar DT, Steinert RF, Stark WJ: Excimer laser
phototherapeutic keratectomy. Baltimore, MD: Williams
& Wilkins; 1997:129.
FIGURE 63.3. (a) Appearance 3 months after PTK. Note the smoothness of the central nodule compared to the neighboring superonasal
nodule. (b) Appearance 9 months after surgery, with 20/100 visual acuity. (c) Appearance 12 months postoperatively. Visual acuity of 20/70 was
achieved at that time.
CHAPTER 63
From Azar DT, Steinert RF, Stark WJ: Excimer laser phototherapeutic keratectomy. Baltimore, MD: Williams & Wilkins; 1997:130.
and anterior stromal dystrophies are amenable to PTK. The aim

of treatment is to ablate confluent opacities that obscure the
visual axis by removing the least amount of tissue possible to
achieve the desired visual result. Excessive treatment and removal
of all opacities are likely to result in a signicantly large hyperopic shift, with a clear central cornea and a poor uncorrected
visual outcome. These dystrophies often have the bulk of lesions
anteriorly. The middle and deep stroma have less opacities and
less confluence in most cases. Deeper lesions should not be
ablated. PTK is often a viable option in the treatment of anterior
corneal dystrophies and may avoid more invasive treatment in

patients who would otherwise need surgical keratoplasty. Dystrophies may recur after PTK in a manner similar to that after
penetrating keratoplasty. Recurrences are typically more supercial and are therefore amenable to re-treatment with PTK.
Furthermore, supercial recurrences in penetrating keratoplasty
can be treated with PTK.
Anterior basement membrane dystrophy, also referred to as
Cogans microcystic dystrophy or map-dot-ngerprint dystrophy,
is a bilateral condition with no denite inheritance pattern. It is
851

often encountered in a general ophthalmology practice and most
commonly affects patients older than 30 years of age, although
familial cases can affect patients younger than 10 years of
age.32,33 Maps are geographic gray lesions that are best seen with
broad illumination. Pathologically, these lesions are noted to be
multilaminar basement membrane with extensions of
abnormal basement membrane into the epithelium.34 Dots are
cysts that consist of cytoplasmic and nuclear debris and that vary
in size compared with the uniform cysts of Meesmanns dystrophy. Cysts form in areas where abnormal basement membrane
projects into the epithelium and epithelial cells become vacuolated and liqueed. Recurrent erosions are caused by ruptured
cysts. Fingerprint lines, composed of brillogranular material
projecting into the epithelium, are seen best with retroillumination. Standard treatment for recurrent erosions includes hypertonic solution or ointment, bandage contact lenses, epithelial
dbridement, supercial keratectomy, and anterior stromal micropuncture. PTK can be used for recalcitrant cases (Fig. 63.4).
Ohman and colleagues found a 74% success rate after initial
treatment in 76 eyes with recurrent erosions and a 92% success
rate after re-treatment.35 Fagerholm and associates reported a
success rate of 84% in 37 treated eyes, which increased to 100%
success after re-treatment.28
Meesmanns dystrophy, also known as juvenile hereditary
epithelial dystrophy, is an uncommon, autosomal dominant
dystrophy. Although this dystrophy can often be observed early
in life, visual acuity often remains good. Slit-lamp examination
reveals multiple, refractile intraepithelial cysts, which may be
seen best on retroillumination. The uniform, diffuse distribution
of the intraepithelial cysts differentiates this condition from mapdot-ngerprint dystrophy. Epitheliopathy related to contact
lenses, dry eye, and blepharitis is included in the differential
diagnosis, but these conditions are not familial.32 Photophobia
and pain may occur later in life as a result of recurrent erosions.
Histopathologically, the cysts are found to contain peculiar substance in the cytoplasm of the basal epithelium. Furthermore,
the basement membrane is found to be thickened. Treatment of
recurrent erosions is accomplished with hypertonic solution or
ointment, bandage contact lenses, or epithelial dbridement. As

described in the section on recurrent erosions, PTK has emerged
as an effective treatment option for recalcitrant recurrent erosions.
ReisBcklers dystrophy is an autosomal dominant dystrophy
involving Bowmans layer. Patients often present in the rst
decade with symptoms of ocular irritation, photophobia, and
lacrimation as a result of recurrent erosions. Recurrent erosions
decrease in frequency as Bowmans layer becomes replaced progressively with scar tissue, although visual acuity is compromised.3639 The central and mid-peripheral cornea are typically
involved with sparing of the peripheral cornea. A reticular opacication of the cornea, at the level of Bowmans membrane,
can be seen clinically. These gray-white opacities are best seen
with broad, oblique illumination. Corneal sensation may be
decreased. The etiology of this dystrophy is believed to be activation of anterior stromal keratocytes by abnormal basal epithelium, which leads to brous scar formation.32 Histologically,
Bowmans layer is found to be replaced by brocellular material.
The basement membrane is absent in certain areas.40,41 The
posterior epithelium has an irregular border. Loss of hemidesmosomes and disorganization of epithelialbasement membrane
adhesion complexes are noted. Recurrent erosions are treated as
already described. Before PTK, supercial keratectomy, lamellar
keratectomy, and penetrating keratoplasty were the surgical procedures performed for visually disabling ReisBcklers dystrophy.
PTK is useful in the recurrent erosions and visually disabling
opacities (Fig. 63.5). Stark and colleagues reported 100%
success in treating two eyes with this condition.1 Rapuano and
Laibsom, Hahn and associates, and Hersh and colleagues also
demonstrated 100% success in the treatment of ReisBcklers
dystrophy with PTK.4245
Granular dystrophy, Groenouws type I dystrophy, is an
autosomal dominant condition that is usually bilateral with
onset in the rst or second decade. Three variants are believed
to exist. Type I typically manifests in the rst decade of life and
is noted clinically to have progressive snowflake opacities, supercial stromal haze, ne punctate opacities, and recurrent erosions.
Patients often have reduced visual acuity by the fourth decade of
FIGURE 63.4. A 35-year-old man with

epithelial basement membrane dystrophy after
anterior stromal puncture (a) underwent PTK
with improvement of symptoms of recurrent
erosions (b). Visual acuity remained unchanged.
SECTION 6

& Wilkins; 1997:76.
852
FIGURE 63.5. A 26-year-old woman with ReisBcklers dystrophy (a). Following PTK, she developed minimal scarring at 1 month (b) and at
12 months (c), which are difcult to distinguish from disease recurrence.
From Azar DT, Steinert RF, Stark WJ: Excimer laser phototherapeutic keratectomy. Baltimore, MD: Williams & Wilkins; 1997:101.
loidosis. The cornea is affected in the third or fourth decade of

life. Fewer and thicker lattice lines occur than in type I dystrophy, and erosions are uncommon. Vision is not affected until
later in life.51 However, skin, peripheral nerves, and cranial nerves
may be affected, which typically occurs after corneal ndings
are detected. Lattice type III dystrophy has thick translucent
lattice lines and subepithelial opacities. Recurrent erosions do
not occur, and vision is often not affected until the eighth decade.52
Inheritance is autosomal recessive. Histopathologically, amyloid
deposits are found to occur in the cornea in the different varieties of lattice dystrophy. Immunofluorescence, Congo red, and
PAS staining conrm this nding. The amyloid exhibits dichroism and manifests green birefringence. Types I and III amyloid
deposits have both protein AA and AP, whereas type II has either
AA or AP. Treatment of lattice dystrophy can be accomplished
by means of PTK. Recurrent erosions can be treated with PTK
as well as surface irregularities and supercial deposits that
interfere with vision (Fig. 63.7). Stark and associates reported a
90% success rate in 11 eyes with PTK for lattice deposits.1
Orndahl and associates also reported a 90% success rate in 11
treated eyes.49
Recurrent granular and lattice dystrophy have been treated
with results comparable to the high success rate for primary
ReisBcklers dystrophy in which deposits are located at
Bowmans layer (Fig. 63.8).1 The more posteriorly located opacities of macular dystrophy should not be treated with PTK.
Penetrating keratoplasty still remains the treatment of choice
for visually disabling macular dystrophy.
FIGURE 63.6. A 42-year-old man with granular
dystrophy who underwent two PTK operations.
(a) Preoperative appearance with 20/200 visual
acuity. (b and c) Clinical appearance by direct
illumination and retroillumination, respectively,
3 months after the rst PTK. Visual acuity
became 20/1000. (d) Six months
postoperatively, with visual acuity of 20/600.
(e) Twelve months postoperatively, with visual
acuity of 20/200. (f and g) Direct illumination
and retroillumination of the eye, respectively,
12 months after the second PTK.

& Wilkins; 1997:108.
FIGURE 63.7. A 45-year-old man with a history

of progressive lattice dystrophy who underwent
PTK. (ac) Preoperative appearance of the
cornea showing progression of dystrophy over
12 years, resulting in 20/1000 visual acuity and
a 2+ mean haze score. (d) Retroillumination of
the eye 12 months after PTK showing residual
lattice dystrophy. Visual acuity improved to
20/200, with a mean haze score of 0.
a
CHAPTER 63
life.46 Type II dystrophy is less severe. Patients are affected usually

in the second decade of life; the opacities are less progressive;
and recurrent erosions are rare.47 Type III dystrophy is described
in infants who have recurrent erosions and a more supercial
variant.48 Pathologically, the deposits are noted to contain hyaline,
which stains with Masson trichrome and weakly with periodic
acid-Schiff (PAS). The source of the hyaline is unknown. Treatment is often not needed until the fourth decade of life, because
the intervening areas between deposits typically allow for good
vision. However, types I and III dystrophy may require earlier
intervention, especially for recurrent erosions. Supercial
deposits can be treated with PTK providing that at least 250 mm
of stroma remains postoperatively (Fig. 63.6). Rapuano and
associates treated six eyes with granular deposits with a success
rate of 83%.42,43 Stark and associates, as well as Orndahl and
colleagues (in a separate series), each treated four eyes with
granular dystrophy, and both groups attained success in 75% of
patients.1,49
Lattice dystrophy, also called BiberHaabDimmer dystrophy,
is believed to have three forms. Type I dystrophy, which is autosomal dominant, occurs in the rst decade of life with irregular
lines and dots in the anterior axial stroma.50 Central haze may
be noted, which increases with time and may affect vision.
Lattice lines are present at different layers of the stroma and may
reach the periphery of the cornea. Epithelial erosions may occur
in the second decade of life, and this may result in irregular
astigmatism and decreased vision. Type II lattice (Meretoja) is
also autosomal dominant and is associated with systemic amy-

& Wilkins; 1997:84.
853

FIGURE 63.8. A 48-year-old woman with a
history of recurrent granular dystrophy in a graft
presenting with 20/200 vision and a 1+ mean
haze score. (a) Clinical appearance of the
cornea prior to PTK by direct illumination.
(b) Clinical appearance of the cornea by
retroillumination. (c) Appearance 3 months after
PTK. Visual acuity improved to 20/80, with no
corneal haze. (d) One year after PTK; note the
increased scarring in the area of treatment.
a

& Wilkins; 1997:109.
FIGURE 63.9. Photograph of a patient with

corneal scar secondary to recurrent pterygium.
(a) Preoperative appearance. (b) Intraoperative
appearance. Note the difculties with bleeding.
(c) Two months after PTK. (d) Nine months
after PTK.
SECTION 6

& Wilkins; 1997:122.
854
Corneal scars may be classied as postinfectious, posttraumatic, trachomatous, climatic droplet keratopathy, and postsurgical. PTK has been used successfully for scars developing or
persisting after pterygium surgery (Fig. 63.9).53 Postinfectious
and posttraumatic scars are often deeper and ablate at different
rates than the surrounding, undamaged stroma, which leaves
an irregular surface. These scars have also been treated successfully with PTK, although not with the same success rate as
supercial pterygium-related opacication. Herpes-related scars
can be treated, although recurrences have been demonstrated in
a number of clinical studies.26,28,30 Topical and systemic antiviral coverage may be necessary to minimize these recurrences.
During stromal ablation, a 0.5 mm wide transition zone is
created between the ablated zone and normal corneal tissue to
provide a gradual transition. This is called standard taper
ablation. Stark and associates have described a modied taper
technique in which the patients eye is moved in a circular
fashion under the laser beam in order to polish the surface and
decrease central flattening. The perimeter of the ablation zone
is treated with a 20 mm deep, 2 mm diameter spot size to accomplish this purpose.1 In patients who are myopic with corneal
opacities or surface irregularities, PTK with PRK may be considered. However, the surgeon must take into account that each
12 mm of stroma removed in a 6 mm diameter ablation zone
may result in 1 D of hyperopia. Smaller diameter ablation zones
increase the hyperopic shifts for the same depth of ablation.
RECURRENCE OF DYSTROPHIES
AFTER PTK
Although initial PTK treatment is typically successful, the dystrophies can recur. Dinh et al54 reviewed the recurrence of
dystrophies after PTK in their institution. In there retrospective
study they evaluated the results of 50 PTK procedures in 43 eyes
with corneal dystrophies. ABM dystrophy recurred in 42% of
the eyes treated. For the group of eyes with ReisBcklers dystrophy 47% recurred after a mean of 22 months after PTK treatment.
Twenty-three percent with granular dystrophy recurred within a
TABLE 63.2. Clinical Results of Summit PTK (for FDA)

Best Corrected Visual Acuity
Preoperatively
(n = 387)
3 months
(n = 311)
6 months
(n = 262)
1 year
(n = 201)
20/40 or better
176 (54.5%)
187 (60.1%)
169 (65.5%)
135 (67.1%)
20/50 to 20/80
117 (30.2%)
69 (22.2%)
50 (19.1%)
38 (18.9%)
20/100 to 20/400
67 (17.3%)
33 (10.6%)
25 (9.5%)
14 (7%)
Less than 20/400
27 (7%)
22 (7.1%)
18 (6.9%)
14 (7%)
Overall Success Rates
Success
At 3 months
6 months
1 year
2 years
Eyes
(n = 333)
Eyes
(n = 276)
Eyes
(n = 206)
Eyes
(n = 94)
243 (73%)
74.6 (206%)
73.3 (151%)
72.3 (68%)
No change
35 (10.5%)
8.7 (24%)
11.2 (23%)
14.9 (14%)
Failure
55 (16.5%)
16.7 (46%)
15.5 (32%)
12.8 (12%)
mean of 40 months after PTK. The lattice dystrophy group

recurred at a rate of 14%, 6 months after initial PTK treatment.
No recurrences were noted in the eye of Schnyder crystalline
dystrophy.
RECURRENT EROSIONS
Conventional techniques to treat recurrent erosions include
hypertonic drops and ointment, bandage contact lenses, and
anterior stromal micropuncture, which can be repeated if necessary. PTK can be used for recalcitrant cases and can be performed
during a recurrent erosion episode or between episodes.55 It is
believed that treatment of Bowmans layer provides a new template for migrating epithelium and improved strength of the
hemidesmosomal adhesion complex. The success rate has been
extremely high in clinical studies. Ohman and colleagues described
three treatment techniques, including ablation of 3 or 5 mm
after removal of the epithelium or ablation of 20 mm directly on
the epithelium.35 Fagerholm and associates treated 37 patients
with 12 pulses after removal of the epithelium and only ablated
3 mm of Bowmans membrane.28 Six patients required retreatment, but no patients who were treated had recurrent episodes.
The maximum follow-up time was 28 months. Seiler noted
cases in which astigmatism resulted from treatment, although
Forster and associates reported no induced astigmatism when
only 15 pulses were used to treat only Bowmans membrane.55,56
It is clear from these and other studies that corneal ablations for
recurrent erosion syndrome should be limited to the anterior
3 mm of Bowmans layer.
CLINICAL OUTCOMES
The VISX FDA and Summit FDA results are presented in
Tables 63.2 and 63.3. The VISX FDA study included 269 primary eyes treated in 17 different facilities. Best spectacle-corrected
visual acuity (BSCVA) improved at least two lines in 53% and
dropped at least two lines in 8% at 1 year.57 Improvement of at
least three lines was noted in 41% at 1 year, while a decrease of
three lines or more occurred in 6.8%. Induced hyperopia of
+2.3 D was the average refractive change at 1 year. The Summit
Technology Excimer Laser study included 398 eyes.58 At 1 year
after PTK, a 22% increase of patients with at least 20/40 BSCVA
occurred. A decrease of 10% of patients with 20/100 vision was

also evident. Success of the procedure was dened as an increase
in two or more lines of BSCVA or signicantly improved subjective complaints in patients undergoing the procedure for
improved comfort. Success was achieved in 73.3%, and 85% of
treated patients reported that they would undergo PTK again.
Table 63.4 reveals the results of other major studies showing
the efcacy of PTK. Sanders gathered PTK cases performed with
the VISX laser from several centers and revealed an average best
corrected visual acuity (BCVA) increase of 1.8 lines
(P < 0.001).8 Forty-ve percent of patients gained two lines or
more of BCVA and 36% gained at least three lines of BCVA. Ten
percent lost two lines or more of BCVA, and 7% lost at least
three lines of BCVA. Three percent of the losses of BCVA were
believed to result from corneal surface irregularities. Campos
and associates treated 18 eyes with PTK, with follow-up of an
average of 8 months.22 Eleven of 18 patients demonstrated
improvement of uncorrected visual acuity, whereas two of 18
patients worsened. Five of 18 had no change. Irregular
astigmatism was believed to cause the loss of uncorrected visual
acuity in the two patients whose condition worsened. Induced
hyperopia was noted in 10 of 18 patients. Chamon and
associates treated 35 eyes with PTK and measured results in
terms of functional visual acuity, dened as the acuity with
either spectacles or contact lenses.23 Twenty-eight patients
gained one line of functional visual acuity, and one patient lost
one line. Four patients became intolerant of their contact lens.
CHAPTER 63
From Steinert RF: Clinical results with the Summit Technology Excimer laser. In: Azar DT, Steinert RF,
Stark WJ: Excimer laser phototherapeutic keratectomy. Baltimore, MD: Williams & Wilkins; 1997: 155166.
TABLE 63.3. Clinical Results of VISX PTK (for FDA) Visual

Efcacy
BSCVA gain of at least 2 lines = 53%
BSCVA gain of at least 3 lines = 41%
BSCVA loss of at least 2 lines = 8%
BSCVA loss of at least 3 lines = 6.8%
Abbreviations: PTK, phototherapeutic keratectomy; FDA, Food and Drug
Administration; BSCVA, best spectacle-corrected visual acuity.
From Ashraf F, Azar D, Odrich M: Clinical results of PTK using the VISX Excimer
laser. In: Azar DT, Steinert RF, Stark WJ: Excimer laser phototherapeutic
keratectomy. Baltimore, MD: Williams & Wilkins; 1997: 169172.
855
TABLE 63.4. Results of Major Clinical Studies

Hyperopic Shift
Study
ReisBcklers dystrophy
Eyes
1
Stark et al
35, 36
Rapuano
VISX
100
100
1.25
VISX
100
100
0.51.5
Summit
100
100
Up to 7
Summit
75
VISX
Stark et al
Rapuano
8.3
83
66
0.622
VISX
Hahn et al72
8.3
66
66
Up to 2
Summit
75
Up to 2
VISX and
Summit
Stark et al1
121
11
90
VISX
Campos et al20
10
100
100
38.2
VISX
Hersh38
100
11
12
90
Stark et al1
100
VISX
Rapuano35, 36
14
66
100
15.2
VISX
Zuckerman et al24
50
4.6
39
Up to 2
Summit
VISX and
Summit
100
Summit
Campos et al20
33
VISX
Hahn
12
100
Summit
Campos et al20
50
VISX
VISX
Fagerholm et al
15.2
66
Summit
Rapuano35, 36
100
VISX
Hahn40
100
Summit
38
Hersh
100
Summit
Fagerholm et al26
15.2
80
Summit
Campos et al
VISX
38
Hersh
100
Summit
Fagerholm et al26
18.3
87
Summit
76
16.3
74
VISX and
Summit
20
353
Ohman et al
92
26
Fagerholm et al
37
11.8
Forster30
9
35, 36
Rapuano
Zuckerman et al24
(after retreatment)
84
100
856
Hersh
26
SECTION 6
Rapuano35, 36
Recurrent erosions
10
40
Band keratopathy
38
Postpterygium scar
Orndahl
Posttrauma scar
Laser
Hersh38
Orndahl
Postinfectious scar
Range (D)
Hahn
39
Salzmanns
35, 36
Lattice
Success
(%)
40
Granular
Follow-Up
(months)
Summit
(after retreatment
89
100
11
82
(visual improvement)
79
(symptomatic recovery)
Three eyes did not show improvement and underwent subsequent keratoplasty. Stark and associates performed PTK on
31 patients and found that 21 of 27 patients with corneal scars
had a functional improvement in vision.1 Three patients did not
improve, and penetrating keratoplasty was performed. Three
Summit
VISX
VISX
other patients who did not improve delayed further treatment.

Hyperopic shifts in patients treated with a 0.5 mm tapered ablation were +5.7 1.14 D at 3 months (P < 0.001), +7.2 1.16 D
at 6 months (P < 0.001), and +5.9 2.36 D at 24 months
(P = 0.055). Six eyes treated with the modied taper technique
to attempt to reduce hyperopia had the following results: +7.1

2.32 D at 3 months (P < 0.05) and 2.7 2.09 D at 6 months
(P > 0.20). Rapuano treated 11 patients with PTK.42 Uncorrected
visual acuity improved two lines or more in 66% of patients,
whereas 33% had no change. BCVA improved two lines or more
in 66% and improved one line in the other 33%. An average
hyperopic shift of +1 D occurred, with a range of 5 to +4 D.
Another study by Rapuano and Laibson included 20 eyes treated
by PTK.43 Uncorrected visual acuity improved an average of two
lines with a range of 1 to +6 lines. BCVA improved an average
of 2.5 lines with a range of 3 to +7. One patients uncorrected
or BCVA decreased more than one line. Hyperopic shift averaged
1.06 D with a range of 4.75 to +9 D. Starr and associates treated
40 patients with PTK and monitored patients for a mean of
11.25 months.59 The average improvement of BSCVA was two
lines. Nineteen of 40 patients improved at least two lines of
visual acuity. A loss of BSCVA occurred in six patients. However, 60% of patients were at least 20/50 postoperatively, which
was an increase from 29% who were 20/50 or better preoperatively. Induced hyperopia was reported to average +2.81 D.
Several studies of PTK with the Summit laser have been published. Hersh and associates treated 12 eyes with PTK, 10 of
which experienced an improvement in BCVA.45 No patient
suffered a loss of BCVA postoperatively. Sixty-six percent of
patients were noted to have a hyperopic shift, with a mean change
of +5.4 D. Subjective improvement occurred in 10 patients (11
of 12 eyes). Orndahl and colleagues reported results of 33 eyes
treated with PTK, with a mean follow-up of 9 months.49
Twenty-three of 27 eyes, in which the goal was to improve
vision, achieved at least two lines of improvement. No loss of
BCVA occurred in any PTK-treated eye. After 1 year, the mean
induced hyperopic shift was +2 D with a range of 0 to +5.5 D.
Hahn and associates studied PTK in nine eyes of seven patients
and demonstrated successful results in treating anterior corneal
pathology.44 Uncorrected and corrected vision improved signicantly in eight of nine eyes. There was no loss of BSCVA in any
patient. Fagerholm and associates performed PTK on 166 eyes
with a mean follow-up of 15 months.28 Each patient and the
treating physician established an individual goal, including
improved vision, contact lens tting, wound healing, and
cosmesis. Improved vision was dened by at least three lines of
improvement on the Snellen chart, which was achieved by 84%
of patients.
PTK FOR THE TREATMENT OF PRK

COMPLICATIONS
Central islands are local steepenings of the central cornea compared with the surrounding ablated zone. Most commonly seen
in higher degrees of myopia, central islands can cause glare,
ghosting of images, myopia, and an overall qualitative decrease
in vision. Most central islands improve spontaneously during
the rst year after treatment.60 Patients with visually signicant
central islands that persist for more than 6 months may need
treatment.61 Using the PRK/PTK modality of the excimer laser,
a transepithelial approach (ablating the epithelium with the
laser) is used. The power difference of the central island must
be calculated from corneal topographic maps. A treatment zone
at least as large as the central island is used. Two hundred
pulses of PTK are programmed (anticipating a 50 mm epithelial
thickness) in a 66.5-mm diameter treatment zone. Microscope
illumination is reduced to the lowest level at which the treatment can be visualized. Epithelial fluorescence (blue range of
the spectrum) is noted when the laser beam ablates the epithelium. The procedure is stopped when a dark area of stromal
ablation is evident. After ablating the epithelium, PRK treatment is performed based on height and curvature data obtained
from corneal topography. Only 50% of associated myopia is

treated to avoid possible overtreatment. Alternatively, if stromal
haze is seen by slit-lamp examination, 515 mm of PTK is
performed.62
Corneal haze is part of the normal healing response after
PRK. Clinically signicant haze is more common in highly
myopic treatment.63 Similar to central islands, haze often clears
spontaneously during the rst year.6466 Steroids have been
reported to decrease haze, but haze is certainly not eliminated
by steroids. Patients with clinically signicant haze for longer
than 6 months may benet from PTK. The combined PRK/PTK
modality is used, with 200 pulses ablating 50 mm of epithelium.
Subepithelial haze is then ablated with the goal of ablating
5070% of haze, leaving the patient with trace, mottled haze
rather than no haze (which could result in a signicant overtreatment and thus, hyperopic shift). The diameter of the treatment is normally planned to be at least as large as the primary
procedure. The shoot and check technique (i.e., ablating and
then checking the patient at the slit-lamp and bringing the
patient back to the laser) is an essential component of this
procedure, which aims at minimizing the possibility of an overcorrection. A PRK approach, treating 50% of the associated
myopia, may also be used.
Another intraoperative flap complication during refractive
surgery is the development of doughnut-shaped or buttonholed
flap. A buttonhole flap is cut when the microkeratome blade
exits through the epithelium during mid-incision and then
reenters to complete the flap.67 Although most flap complications such as free caps and short flaps usually do not cause longterm loss of vision, buttonholed flaps are the flap complication
most likely to result in signicant visual loss. The conventional
management is to reapproximate the flap with the stroma and
defer laser ablation for several months. A new flap is created
and the refractive error treated. There are several disadvantages
of this strategy. This led Taneri and collegues to describe in a
case report an alternative technique of transepithelial PTK/PRK
assisted by the intraoperative use of mitomycin C (MMC) to
prevent haze and scar formation and recurrent epithelial
ingrowth. This procedure therefore is an alternative in managing
central epithelial ingrowth in a buttonholed LASIK flap. The
authors suggest that this method may offer faster visual
recovery and decreased risk for repeated buttonholes creation
compared with the widespread recutting of a new flap after a
several months.68
Irregular astigmatism is a possible complication of PRK.
Gibralter and Trokel reported a method of treating irregular astigmatism in a corneal graft by using the excimer laser.69 Smalldiameter ablations were used to neutralize the astigmatism by
treating steep areas of the cornea identied topographically. The
diameter and steepness of the irregular areas were determined
preoperatively. Residual myopia was corrected with PRK
parameters.
Decentered PRK treatments have been associated with halos,
glare, and refractive errors. The distinction between laser drift
and shift, as described by Azar and Yeh, may be useful in the
management of decentrations.70 Shift results from misalignment of the laser beam with the center of the entrance pupil.
Intraoperative drift occurs when the eye moves subtly during
treatment and results in a nonuniform ablation. Patients with
treatment decentration but no or low drift may benet from
transepithelial treatment centering the PTK and PRK on the
center of the entrance pupil (Fig. 63.10).71,72 The area of epithelial hypertrophy covering the edge of the ablation serves to
protect the area of previous ablation at the conclusion of the
PTK treatment while exposing the stroma of the neighboring
zone to be treated with the subsequent PRK. Patients with high
drift benet from centration over the entrance pupil and PTK
CHAPTER 63
857

FIGURE 63.10. (a) Preoperative topography
showing decentered treatment. The patient
underwent PTK treatment of the epithelium
centered around the pupillary area followed by
PRK treatment of the residual refractive error.
(b) Postoperative topography showing
recentration of the ablation zone.
& Wilkins; 1997:184.
with the use of masking agents to cover the valleys and expose
the peaks. Subsequent PRK is used to treat residual myopia.73
Masking agents are seldom needed to treat pure treatment
displacements not associated with intraoperative drift.
Successful treatment of a nasally decentered PRK with grade
2 haze at the temporal part of the ablation was described by
Talamo and Wagoner.74 A 6 mm diameter ablation was used for
PTK, and the epithelium was removed with the laser. All epithelium was removed temporally, but not nasally, to achieve more
temporal subepithelial ablation. Celluvisc was the masking
agent use, and additional pulses were used to decrease the
temporal haze.
SECTION 6
NEW MASKING AGENTS
858
The major inherent obstacle of PTK for the management of

corneal surface irregularities is that through photoablation these
irregularities are reproduced deeper within the stroma. Masking
techniques refer to photorefractive procedures utilizing various
masking means or so-called modulators to protect flatter corneal areas while steeper areas are excised with the excimer laser.
Kornmehl et al31 have shown that an ideal masking agent should
have moderate viscosity (between that of saline and 1% carboxymethylcellulose) and concluded that very viscous fluids would
not cover irregular surfaces uniformly whereas fluids of
inadequate viscosity would run off quickly exposing both peaks
and valleys thus resulting in irregular surfaces after ablation.
Numerous previous investigators have used different masking
agents as methylcellulose;45,7577 and sodium hyaluronate78,79 at
various concentrations or performed transepithelial (so that the
epithelium acts as a natural masking agent) treatments,5,6,9,10
and reported the benecial effect of masked PTK. Fasano et al75
reported that the ideal masking agent should have the same
ablation rate to that of the cornea, be biocompatible and adhere
well to the cornea.
PHOTOABLATABLE LENTICULAR
MODULATOR TECHNIQUE
This refers to the use of a modied collagen gel solution for the
photorefractive correction of corneal surface irregularities.80
The photoablatable lenticular modulator (PALM) gel similarly
to other collagen modulators8183 is thermo reversible. The gel
is in liquid state when heated to solidify to a rm gel as its
temperature lowers. Its use for masking purposes requires its
application onto the corneal stoma at a temperature of 49C
where it can be molded to form a stable lenticule that serves as
the nal masking agent. In studies by Pallikaris and colleagues,
they examined the possible implication of the PALM technique.84
They concluded that the use of the PALM gel did not seem to
seriously affect the healing process after PTK on rabbit corneas
as compared to controls.
Another potential molder, BioMask, has been evaluated.

BioMask (Mauerick Technologies Inc, Clearwater, FL, USA), a
porcine type I collagen has been tested to produce smooth corneas
during excimer laser PTK. Heated BioMask is a liquid that
solidies as it cools. In studies of live rabbit eyes with corneal
irregularities, the use of BioMask showed moderate success in
reproducing contact lens curvatures in the corneal stroma. The
authors concluded that BioMask is promising in the treatment
of supercial corneal irregularities.85
SELECTIVE ZONAL ABLATION

One of the most critical deciencies in PTK is the treatment of
irregular astigmatism. Irregular astigmatism is caused by, among
other causes, refractive surgery and corneal transplantation.
Alio and collegues86 performed a retrospective review of the use
of selective zonal ablation to treat varying degrees of irregular
astigmatism after refractive surgery. They divided the eyes into
two groups based on the type of irregularity on corneal topography. Based on the degree and location of the steep areas on
corneal topography, they used mathematical formulas to determine the exact size and depth of ablations for each of these
areas. In the group with the identiable pattern of irregularity,
there was a signicant improvement in the topographic corneal
uniformity index and in best corrected visual acuity. No signicant improvement was seen in the group of eyes with no identiable pattern of corneal irregularity. It is important to note that
this is a complex method described by the authors. Improvements in linking topographic analysis to excimer laser
treatments (such as wave-front analysis technology) will allow
for more predictable treatments for irregular astigmatism.87
POSTOPERATIVE CARE
Postoperatively, a prophylactic antibiotic ophthalmic ointment
(bacitracin or erythromycin) and a cycloplegic drop, such as homatropine, are placed in the eye, and a pressure patch is applied.
Alternatively, a bandage contact lens may be used, and the
patient is instructed to use a broad-spectrum topical antibiotic
such as Ocuflox or Ciloxan, and a topical steroid, including
prednisolone acetate 1% or fluorometholone 0.125% four times
a day. The steroid drop is tapered to once a day within 1 month,
whereas the antibiotic is stopped after epithelialization is complete. The possible complications of increased intraocular pressure, cataract, and potentiation of microbial infections, including
herpetic recurrence, often outweigh the benet of continued use
of topical steroids for a long period. Topical nonsteroidal antiinflammatory drugs (NSAIDs), such as diclofenac sodium
(Voltaren) and ketorolac (Acular), may help to control the pain
in the rst 2448 h.1,2,88 However, indiscriminate use may lead
to wound healing problems and sterile inltrates that may be
COMPLICATIONS OF PTK
HYPEROPIA
Flattening of the central cornea is the principal side effect of
PTK. The flattening may result in a large hyperopic shift, which
may require contact lens use postoperatively. Several explanations have been advanced for the induced hyperopia caused by
PTK. The release of ablation products (plume) toward the edge
of the ablation may protect the stroma at the periphery of the
ablation. More ablation may occur centrally with pathology that
thins toward the visual axis. Greater epithelial hyperplasia as
well as a thicker tear meniscus peripherally may play a role in
central flattening. Finally, since the laser beam is focused
centrally, the periphery receives an angled beam that is slightly
defocused and may cause decreased ablation peripherally.20
Use of masking agents during PTK and avoidance of treatment of deeper pathology, since there is a positive correlation
with the depth of ablation and induced hyperopia, can help to
reduce hyperopic shift, although it cannot by any means eliminate it.1 Shers buff and polish technique, moving the patients
head in a circular fashion under the laser beam of varying
aperture size, has had variable success in reducing hyperopic
shift.91 Sher also performed hyperopic ablations in certain cases
after PTK. Our modied taper technique, previously described,
is also a strategy that attempts to decrease hyperopic shift.1 The
importance of induced hyperopia should not be minimized.
Nevertheless, Zuckerman and associates have reported that
planned, desirable hyperopic shifts were possible in 44% of
patients undergoing PTK in order to minimize preoperative
myopia.26
MYOPIA
The treatment of paracentral or peripheral opacities may lead to
relative central steepening. Sher and colleagues reported a 3%
rate of induced myopia, whereas Campos and associates found
a rate of 16.6% myopic shift.22,91
DELAYED EPITHELIALIZATION
The importance of epithelialization cannot be overemphasized.
An intact epithelium is a barrier to the entry of microorganisms
into subepithelial corneal tissue. In addition, pain can be quite
severe after PTK because of the epithelial defect, and vision is
reduced. Recurrent erosions and persistent epithelial defects are

potential complications that are more often seen in patients
with preoperative epitheliopathy. PTK can possibly exacerbate
preoperative epitheliopathy by damaging Bowmans membrane
and causing inflammation that may hinder epithelialization.
Iatrogenic toxicity from postoperative drops may further
negatively affect wound healing. Collagen vascular diseases and
diabetes are possible systemic diseases that can affect wound
healing. A history of heavy alcohol intake in a patient who took
34 weeks to heal after PTK was reported by Chamon and
colleagues.23 Treatment of delayed epithelialization involves
decreasing active inflammation with topical steroids, bandage
contact lenses and lubrication, and punctal plugs that can also
be placed preoperatively in patients with signs and symptoms of
dry eye. If these techniques are not helpful, tarsorrhaphy can be
performed. However, preoperative treatment of epitheliopathy
can often be most helpful in preventing wound healing
difculties.
PAIN
Pain after PTK may be quite severe. Often, an already compromised cornea will demonstrate a greater amount of inflammation than a normal cornea undergoing excimer laser ablation.
Prostaglandins, thromboxanes, and leukotrienes generated from
arachidonic acid by way of the cyclooxygenase and lipoxygenase
pathways are believed to be signicant biologic mediators of
pain after excimer laser surgery. The use of topical NSAIDs has
been an important addition to pain management. Diclofenac
sodium (Voltaren), ketorolac tromethamine (Acular), and flurbiprofen sodium (Ocufen) have helped in pain control. Voltaren,
in high doses, inhibits the cyclooxygenase pathway and decreases
intracellular arachidonic acid, which in effect decreases substrates
available for the lipoxygenase pathway. Phillips and associates
found prostaglandin E2 levels to be signicantly lower in New
Zealand White rabbits treated with diclofenac sodium than controls after keratectomy.92 Szerenyi and colleagues reported
decreased prostaglandin E2 levels in rabbits after excimer laser
ablation.93 Stein and associates reported that patients receiving
ketorolac tromethamine after excimer ablation experienced signicantly less pain than controls and demonstrated an even greater
reduction in pain in patients receiving ketorolac tromethamine
preoperatively as well as postoperatively.94
Twenty-eight cases of subepithelial inltrates were reported
in a survey by Canadian PRK surgeons in patients using topical
NSAIDs pre- and post-PRK. None of these patients was using
topical steroids, which are believed to prevent inflammatory
cell margination and migration. Campos and associates found
fluorometholone to signicantly decrease leukocyte inltration
into rabbit corneas 1 day after excimer laser photoablation.
IRREGULAR ASTIGMATISM
Proper alignment of the microscope and laser beam is essential
in avoiding PTK decentrations, which can lead to irregular
astigmatism. The use of miotics preoperatively can potentially
cause superonasal displacement of the pupil, thus resulting in
decentration if the beam is centered at the entrance pupil.
Patient xation must be carefully monitored intraoperatively.
Treatment of diffuse pathology should be centered at the
entrance pupil with a large ablation zone. Care should be taken
in the treatment of focal nodules or pathology by centering on
the lesion itself. Masking agents help greatly in exposing elevations and covering depressions to achieve a smoother surface.
However, different ablation rates for long-standing scar tissue
and normal stroma can leave an irregular ablation area and
postoperative astigmatism.
CHAPTER 63
confused with infectious inltrates and development of

ulceration and perforation.89 NSAID-related inltrates typically
do not occur when topical steroids are also used. Systemic
sedativeanalgesics are also used as needed in the rst few days
postoperatively. In patients who are known or suspected to have
HSV keratitis, prophylactic treatment with topical and systemic
antivirals may reduce the risks of herpetic recurrences.
Epithelialization is often complete within 1 week after PTK.
Until this occurs, patients are examined every 2472 h. Patients
are also examined at 1 month, 3 months, 6 months, 12 months,
and annually after PTK. Each visit should address symptomatic
complaints and vision (with and without correction) as well as
detailed anterior segment evaluation. Several different studies
have reported delayed epithelialization in specic cases. Das et al,
reported delayed corneal epithelialization in patients with lattice
dystrophy but their study did not offer a consistent concept
explaining the reason for delayed epithelial wound healing.90
Another report by Zuckerman et al26 noticed a slight delay in
reepithelialization in cases of corneal leucoma patients.
Computer-assisted topography can also be performed at the
3-month visit and afterward, as needed.
859
RECURRENCE AND HAZE

Corneal dystrophies treated with PTK can recur. Recurrences,
such as in corneal transplants, may be anterior to the original
pathology. Re-treatment with PTK is a viable alternative, although
hyperopic shift and anisometropia may result from further PTK.
Manual keratectomy with a BardParker blade should be considered. Re-treatment with PTK may also lead to increased
haze, although haze often clears during the rst year.
BACTERIAL KERATITIS
The existence of an epithelial defect in what may be an already
compromised cornea is a risk factor for the development of
bacterial keratitis. In addition, the placement of a contact lens
over the epithelial defect postoperatively may enhance the risk
of infectious keratitis. These factors mandate the use of prophylactic topical, broad-spectrum antibiotics. The ability of bandage
contact lenses to help with wound healing and pain control must
be balanced against the small but potential risk of infectious
keratitis. Inltrates can also be caused by NSAIDs and contact
lenses. Suspicious lesions should be treated as if they are
infectious, similarly to patients with infectious keratitis who
have not had PTK. Al-Rajhi and associates reported three eyes
that developed bacterial keratitis of 258 total eyes undergoing
PTK.95 All three patients had climatic droplet keratopathy and
developed Gram-positive keratitis. The authors believe that the
keratitis is part of the natural history of the disease and
maintain that PTK did not enhance the risk.
VIRAL KERATITIS
Treatment of patients with a history of herpes simplex keratitis
should be avoided if possible. In cases that are treated, acyclovir
is a useful adjunct preoperatively and postoperatively. Campos
and associates described successful PTK in a patient with a
history of viral keratitis with oral acyclovir preoperatively and
10 days postoperatively.22 Topical antiviral agents may be considered, although toxicity to the corneal surface must be outweighed by the benets. Topical steroids should be used sparingly.
Zuckerman and associates treated four patients with a history
of herpetic keratitis.26 Two of the patients had recurrent keratitis,

although both were free of active disease for at least 1 year after
PTK; one at 2 weeks and the other at 6 weeks postoperatively.
Vrabec and colleagues reported on two patients with recurrent
dendritic ulcers in eyes treated with PTK for stromal scars secondary to herpes.27 McDonnell and associates performed excimer
laser astigmatic keratectomy, in a corneal graft, in a patient with
a history of herpetic keratitis and noted a recurrence of epithelial keratitis.30 Fagerholm and associates noted one patient who
had three recurrences of herpetic keratitis after PTK.28
GRAFT REJECTION
Hersh and associates have reported on a patient with recurrent
lattice dystrophy in a corneal graft, treated with PTK, who experienced graft rejection.96 Medical treatment resulted in successful suppression of the rejection. Epstein and associates treated
postoperative astigmatism in a corneal graft with the excimer
laser, and a rejection episode occurred that was also successfully
treated.97
CONCLUSION
In conclusion, a new era in the treatment of anterior corneal disease and surface irregularities is under way since the inception
of PTK for the treatment of corneal disorders in 1988. The risks
of intraocular surgery and systemic or peribulbar anesthesia
may be avoided as a result of the ability to avoid penetrating and
lamellar keratoplasty, which have been the surgical mainstay of
treating visually signicant corneal diseases. As the number of
PRK procedures continues to increase, PTK will be called upon
with increasing frequency to treat complications. The limits of
PTK must be clearly understood, and deep corneal pathology
should not be treated. Treatment must be individualized. Results
of clinical studies reveal a high success rate for various anterior
corneal diseases. The use of PTK should be viewed as an additional option to treat corneal disease, used in conjunction with
more traditional surgical techniques. Still, the use of PTK is an
exciting new alternative for corneal surgeons.
SECTION 6
REFERENCES
860
1. Stark WJ, Chamon W, Kamp MT, et al:

Clinical follow-up of 193-nm ArF excimer
laser photokeratectomy. Ophthalmology
1992; 99:805811.
2. Salz JJ, Maguen E, Macy JI, et al: Oneyear results of excimer laser photorefractive
Surg 1992; 8:270273.
3. Gaster RN, Binder PS, Coalwell K, et al:
Corneal surface ablation by 193 nm
excimer laser and wound healing in rabbits.
4. Abad JC, Krueger RR: Basic science and
principles. In: Talamo JH, Krueger RR, eds.
The excimer manual: a clinicians guide to
excimer laser surgery. Boston, MA: Little,
Brown; 1997:323.
5. Garrison BJ, Srinivasan R: Laser ablation of
organic polymers: microscopic models for
photochemical and thermal processes.
J Appl Phys 1985; 57:29092913.
6. Farrell RA: Corneal transparency. In:
Albert DM, Jakobiec FA, eds. Principles and
practice of ophthalmology: basic sciences.
Philadelphia, PA: WB Saunders; 1994:6481.
7. Olsen BR, McCarthy MT: Molecular
structure of the sclera, cornea, and vitreous
8.
9.
10.
11.
12.
13.
body. In: Albert DM, Jakobiec FA, eds.

Principles and practice of ophthalmology:
basic sciences. Philadelphia, PA: WB
Saunders; 1994:47, 48.
Sanders D: Clinical evaluation of
phototherapeutic keratectomy-VISX
twenty/twenty excimer laser. FDA, 1994.
Gipson IK, Spurr-Michaud SJ, Tisdale AS:
Hemidesmosomes and anchoring bril
collagen appear synchronously during
development and wound healing. Dev Biol
1988; 126:253262.
Marshall J, Trokel S, Rothery S, Schubert
H: An ultrastructural study of corneal
incisions induced by an excimer laser at
193 nm. Ophthalmology 1985; 92:749758.
Malley DS, Steinert RF, Puliato CA, et al:
Immunofluorescence study of corneal
wound healing after excimer laser anterior
keratectomy in the monkey eye. Arch
Ophthalmol 1990; 108:13161322.
Seiler T, Fantes T, Waring GO, Hanna KD:
Laser corneal surgery. In: Waring GO III, ed.
Refractive keratotomy. St Louis, MO:
Mosby; 1992:669745.
Hanna KD, Pouliquen YM, Savodelli M,
et al: Corneal wound healing in moneys 18
14.
15.
16.
17.
18.
19.

keratectomy. Refract Corneal Surg 1990;
6:340345.
Fountain TR, De la Cruz Z, Green WR, et al:
Reassembly of corneal epithelial adhesion
structures after excimer laser keratectomy
in humans. Arch Ophthalmol 1994;
112:967972.
Fantes FE, Hanna KD, Waring GO, et al:
Wound healing after excimer laser
keratomileusis (photorefractive
keratectomy) in monkeys. Arch Ophthalmol
1990; 108:665675.
Tuft SJ, Zabel RW, Marshall J: Corneal
repair following keratectomy. Invest
Courant D, Fritsch P, Azema A, et al:
Corneal wound healing after photo-keratomileusis treatment on the primate eye.
Lasers Light Ophthalmol 1990; 3:189195.
Dogru M, Katami, Miyashita M, et al:
Ocular surface changes after excimer laser
phototherapeutic keratectomy.
Ophthalmology 2000, 107:11441152.
Rapuano C: Excimer laser phototherapeutic
keratectomy. Curr Opin Ophthalmol 2001,
12:288293.
20. Azar DT, Jain S, Woods K, et al:

Phototherapeutic keratectomy: the VISX
experience. In: Salz JJ, McDonnell PJ,
McDonald MB, eds. Corneal laser
surgery. St Louis, MO: CV Mosby;
1995:213226.
21. Rapuano CJ: Excimer laser
phototherapeutic keratectomy. Int
22. Campos M, Nielson S, Szerenyi K, et al:
Clinical follow-up of phototherapeutic
keratectomy for treatment of corneal
opacities. Am J Ophthalmol 1993;
115:433440.
23. Chamon W, Azar DT, Stark WJ, et al:
Phototherapeutic keratectomy. Ophthalmol
Clin North Am 1993; 6:399413.
24. Gottsch JD, Gilbert ML, Goodman DF, et al:
Excimer laser ablative treatment of
microbial keratitis. Ophthalmology 1991;
98:146149.
25. Eiferman RA, Forgey DR, Cook YD:
Excimer laser ablation of infectious
crystalline keratopathy. Arch Ophthalmol
1992; 110:18.
26. Zuckerman SJ, Aquavella JV, Park SB:
Analysis of the efcacy and safety of
excimer laser PTK in the treatment of
corneal disease. Cornea 1996; 15:914.
27. Vrabec MP, Anderson JA, Rock ME, et al:
Electron microscopic ndings in a
cornea with recurrence of herpes simplex
keratitis after excimer laser
1994; 20:4144.
28. Fagerholm P, Fitzsimmons TD, Orndahl M,
et al: Phototherapeutic keratectomy: longterm results in 166 eyes. Refract Corneal
Surg 1993; 9(Suppl 2):7681.
29. Pepose JS, Laycock KA, Miller JK, et al:
Reactivation of latent herpes simplex virus
by excimer laser photokeratectomy. Am J
Ophthalmol 1992; 114:4550.
30. McDonnell PJ, Moreira H, Clapham TN,
et al: Photorefractive keratectomy for
astigmatism. Arch Ophthalmol 1991;
109:13701373.
31. Kornmehl EW, Steinert RF, Puliato CA:
A comparative study of masking fluids for
keratectomy. Arch Ophthalmol 1991;
109:860863.
32. Tuli S, Chang SW, Stark W, Azar DT:
Anterior corneal dystrophies: clinical
features. In: Azar DT, Steinert RF, Stark WJ,
eds. Excimer laser phototherapeutic
keratectomy. Baltimore, MD: Williams &
Wilkins; 1997:7397.
33. Laibson PR, Krachmer JH: Familial
occurrence of dot, map, and ngerprint
dystrophy of the cornea. Invest Ophthalmol
1975; 14:397399.
34. Broderick JD, Dark AJ, Peace GW:
Fingerprint dystrophy of the cornea. Arch
Ophthalmol 1974; 92:483489.
35. Ohman L, Fagerholm P, Tengroth B:
Treatment of recurrent erosions with the
excimer laser. Acta Ophthalmol 1994;
72:461463.
36. Reis W: Familidre, fleckige
homhautentartung. Dtsch Med Wochenschr
1917; 43:575.
37. Bucklers M: ber eine weitere familiare
hornhautdystrophie. Klin Monatsbl
38. Rice NS, Ashton N, Jay B, Black RK:
ReisBcklers dystrophy. Br J Ophthalmol
1968; 52:577603.
39. Jones ST, Stauffer LK: Reis-Bcklers

corneal dystrophy: a clinicopathologic
study. Trans Am Acad Ophthalmol
Otolaryngol 1970; 74:417426.
40. Kanai A, Kaufman HE, Polack FM: Electron
microscopic study of Reis-Bcklers
dystrophy. Ann Ophthalmol 1973;
5:953962.
41. Perry HD, Fine BS, Caldwell CR:
ReisBcklers dystrophy: a study of eight
cases. Arch Ophthalmol 1979; 97:664670.
42. Rapuano CJ, Laibson PR: Excimer laser
phototherapeutic keratectomy for anterior
corneal pathology. CLAO J 1994;
20:253257.
43. Rapuano CJ, Laibson PR: Excimer laser
1993; 19:235240.
44. Hahn TW, Sah WJ, Kim JH:
Phototherapeutic keratectomy in nine
eyes with supercial corneal diseases.
Refract Corneal Surg 1993; 9(Suppl
2):115118.
45. Hersh PR, Spinak A, Garrana R, Mayers M:
Phototherapeutic keratectomy: Strategies
and results in 12 eyes. Refract Corneal
Surg 1993; 9(Suppl):s90s94.
46. Waardenburg PJ, Jonkers GH: A specic
type of dominant progressive dystrophy of
the cornea developing after birth. Acta
47. Forsius H, Erickson AW, Karna J, et al:
Granular corneal dystrophy with late
manifestation. Ophthalmology 1983;
61:514528.
48. Rodrigues MM, Gaster RN, Pratt MV:
Unusual supercial confluent form of
granular corneal dystrophy. Ophthalmology
1983; 90:15071511.
49. Orndahl M, Fagerholm P, Fitzsimmons T,
Tengroth B: Treatment of corneal
dystrophies with excimer laser. Acta
Ophthalmol 1994; 72:235240.
50. Ramsey RM: Familial corneal dystrophy
lattice type. Trans Am Ophthalmol 1980;
64:53.
51. Asaoka T, Amano S, Sunada Y, Sawa M:
Lattice corneal dystrophy type II with
familial amyloid polyneuropathy type IV.
52. Hida T, Tsubota K, Kigasawa K, et al:
Clinical features of a newly recognized type
of lattice corneal dystrophy. Am J
Ophthalmol 1987; 104:241248.
53. Azar DT, Jain S, Stark W: Phototherapeutic
keratectomy. In: Azar DT, ed. Refractive
surgery. East Norwalk, CT: Appleton &
Lange; 1997:501517.
54. Dinh R, Rapuano CJ, Cohen EJ, et al:
Recurrence of conreal dystrophy excimer
laser phototherapeutic keratectomy.
Ophthalmology 1999, 106:14901497.
55. Forster W, Grewe S, Atzler U, et al:
Phototherapeutic keratectomy in corneal
diseases. Refract Corneal Surg 1993;
9(Suppl):s85s89.
56. Seiler T: Der excimer laser. Ein Instrument
fur die Hornhautchirurgie. Ophthalmologe
1992; 89:128133.
57. Ashraf F, Azar D, Odrich M: Clinical results
of PTK using the VISX excimer laser. In:
Azar DT, Steinert RF, Stark WJ, eds.
Wilkins; 1997:201205.
58. Steinert RF: Clinical results with the
Summit technology excimer laser. In:
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.

Wilkins; 1997:155156.
Starr M, Donnenfeld E, Newton M, et al:
Krueger RR, Saedy NF, McDonnell PJ:
Clinical analysis of topographic steep
central islands following excimer
photorefractive keratectomy (PRK). ARVO
Abstracts. Invest Ophthalmol Vis Sci 1994;
35(Suppl 4):1740.
Snibson GR, McCarty CA, Aldred GF, et al:
Retreatment after excimer laser
photorefractive keratectomy. Am J
Ophthalmol 1996; 121:250257.
Machat JJ: PRK retreatment techniques
and results. In: Machat JJ, ed. Excimer
laser refractive surgery: practice and
principles. Thorofare, NJ: Slack;
1996:215233.
Seiler T, Holschback A, Derse M, et al:
Complications of myopic photorefractive
keratectomy with the excimer laser.
Marshall J, Trokel SL, Rothery S: Long-term
healing of the central cornea after
photorefractive keratectomy using an
excimer laser. Ophthalmology 1988;
95:14111421.
Stein HA, Cheskes A, Stein RM:
Complications and their management.
In: Stein H, Cheskes A, Stein RM, eds.
The excimer: fundamentals and clinical
use. Thorofare, NJ: Slack; 1996.
Maguen E, Machat JJ: Complications of
photorefractive keratectomy, primarily with
the VISX excimer laser. In: Salz JJ, ed.
Corneal laser surgery. St Louis, MO: CV
Mosby; 1995:143158.
Wilson SE: LASIK: management of
common complications. Cornea 1998;
17:459467.
Taneri S, Koch JM, Melki S, Azar DT:
Mitomycin-C assisted photorefractive
keratectomy in the treatment of
buttonholed laser in situ keratomileusis
flaps associated with epithelial ingrowth.
31:20262030.
Gibralter R, Trokel SL: Correction of
irregular astigmatism with the excimer
laser. Ophthalmology 1994;
101:13101315.
Azar DT, Yeh PC: Corneal topographic
evaluation in decentration in
photorefractive keratectomy: treatment
displacement vs. intraoperative drift. Am J
Ophthalmol 1997; 124:312320.
Uozato H, Guyton DL: Centering corneal
surgical procedures. Am J Ophthalmol
1987; 103:264275.
Seiler T, Reckmann W, Maloney RK:
Effective spherical aberration of the cornea
as a quantitative descriptor in corneal
topography. J Cataract Refract Surg 1993;
19(Suppl):155165.
Azar DT, Stark WJ, Steinert RF: Surgical
management of PRK complications. In:
Wilkins; 1997:169172.
Talamo JH, Wagoner MD, Lee SL:
Management of ablation decentration
following excimer photorefractive
keratectomy. Arch Ophthalmol 1995;
113:706707.
CHAPTER 63
861
SECTION 6
75. Fasano PA, Moreira H, Mc Donnel PJ,

Sinbawy A: Excimer laser smoothing with a
reproducible model of anterior corneal
surface irregularity. Ophthalmology 1991;
98:17821785.
76. Gartry D, Kerr Muir M, Maarshall J: Excimer
laser treatment of corneal surface
pathology a laboratory and clinical study.
77. Fagerholm P, Fitzimmmons TD, Orndhal M,
et al: Phototherapeutic keratectomy: long
term results in 166 eyes. Refract Corneal
Surg 1993; 9:S76S81.
78. Alio JL, Belda JI, Shalaby MM: Correction
of irregular astigmatism with excimer laser
assisted by sodium hyaluronate.
Ophthalmology 2001; 108:12461260.
79. Dogru M, Katakami C, Yamanka A:
Refractive changes after excimer laser
phototherapeutic keratectomy. J Cataract
Refract Surg 2001; 27:686692.
80. Pallikaris IG, Katsanevaki VJ, Ginis HS: The
PALM technique as an alternative to
Customized Ablation. Semin Ophthalmol
2000; 15:160169.
81. De Vore DP, Scott JB, Nordquist RE, et al:
Rapidly polymerized collagen gel as a
smoothing agent in excimer laser
photoablation. J Refract Surg 1995;
11:5055.
82. Stevens SX, Bowyer BL, Sanchez-Thorin JC,
et al: The BioMask for treatment of corneal
surface irregularities with excimer Laser
862
83.
84.
85.
86.
87.
88.
89.
90.
phototherapeutic keratectomy. Cornea

1999; 18:155163.
Kremer F, Aronsky M, Bowyer BL,
Stevens SX: Treatment of corneal surface
irregularities using BioMask as an adjunct
to excimer laser phototherapeutic
Pallikaris IG, Ginis HS, Katsanevaki VJ:
Photoablated lenticular modulator PALM
technique: a review of 10 cases. Invest
Stevens SX, Bowyer BL, Sanchez-Thorin JC,
et al: The BioMask for treatment of corneal
surface irregularities with excimer laser
photokeratectomy. Cornea 1999, 18:155163.
Alio JL, Artola A, Rodriguez-Mier FA:
Selective zonal ablations with excimer laser
for correction of irregular astigmatism
induced by refractive surgery.
Ophthalmology 2000, 107:662673.
Rapuano et al 2000.
Arshinoff S, DAddario D, Sadler C, et al:
Use of topical nonsteroidal antiinflammatory drugs in excimer laser
photorefractive keratectomy. J Cat Refract
Surg 1994; 20(Suppl):216222.
Guidera AC, Luch JI, Udell IJ, et al:
Keratitis, ulceration and perforation
associated with topical nonsteroidal
antiinflammatory drugs. Ophthalmology
2001; 108:936944.
Das S, Langenbucher A, Seitz B: Delayed
healing of corneal epithelium after
91.
92.
93.
94.
95.
96.
97.
phototherapeutic keratectomy for lattice

dystrophy. Cornea 2005; 24:283287.
Sher NA, Bowers RA, Zabel RW, et al:
Clinical use of 193-nm excimer laser in the
treatment of corneal scars. Arch Opthalmol
1991; 109:491498.
Phillips AF, Szerenyi K, Campos M, et al:
Arachidonic acid metabolites after excimer
laser corneal surgery. Arch Ophthalmol
1993; 111:12731278.
Szerenyi K, Wang XW, Lee M, McDonnell PJ:
Topical diclofenac treatment prior to
excimer laser photorefractive keratectomy
in rabbits. Refract Corneal Surg 1993;
9:437442.
Stein R, Stein HA, Cheskes A, Symons S:
Photorefractive keratectomy and
postoperative pain. Am J Ophthalmol 1994;
117:403404.
Al-Rajhi AA, Wagoner MD, Badr IA, et al:
Bacterial keratitis following
phototherapeutic keratectomy. J Refract
Surg 1996; 12:123127.
Hersh PS, Jordan AJ, Mayers M: Corneal
graft rejection episode after excimer laser
phototherapeutic keratectomy. Arch
Ophthalmol 1993; 11:735736.
Epstein RJ, Robin JB: Corneal graft
rejection episode after excimer laser
phototherapeutic keratectomy. Arch
CHAPTER
64
Conjunctival Surgery
Christopher I. Zoumalan and Glenn C. Cockerham
INTRODUCTION
The conjunctiva, an epithelial-lined mucosal tissue covering the
ocular surface, serves several important functions. It acts as a
protective layer for underlying tissues, including the globe, and
by virtue of tight junctions between epithelial cells, retains
moisture by preventing evaporation of water vapor. Accessory
lacrimal glands continually lubricate the ocular surface. A rich
vascular network in conjunctiva and underlying episclera maintain homeostasis and enable rapid healing of injury or infection.
The benecial effects of an intact and functional conjunctiva
have long been recognized. The ability to easily mobilize and
translocate conjunctival tissue with an intact vascular pedicle
has led to several useful operations. More recently, conjunctival
replacement by heterologous tissues has been extensively
studied.
CONJUNCTIVAL FLAP
The use of a conjunctival graft to treat diseased or damaged
cornea was rst documented by Schoeler in Berlin in 1877. The
concept was further popularized several years later by Kuhnt in
1884.1 Later in 1912, Byers and Van Lint introduced its use in
cataract surgery, eviscerations, and corneoscleral lacerations.
Van Lint proposed placement of conjunctiva over cataract incisions to help retard infection and expedite healing.2 In 1927,
Green advocated surgical duplication of natures third process
of repair, or wound healing by conjunctival ingrowth, as benecial
for various corneal ulcerations, including herpetic and Moorens
ulcers, and for perforations.3 Haik, in 1954, noted during his
wartime experience, the benecial though temporary use of
fornix conjunctival flaps in the healing of severely diseased
eyes.4 Several years later, in 1958, Gunderson described a
technique for fashioning a thin flap by carefully dissecting
conjunctiva from the underlying Tenons fascia; this technique,
because of its success, remains in use today.5 Prior to the introduction of Gundersons flap, previous conjunctival flaps were often
temporizing measures, lasting a few days (such as purse-string
flaps pulled over the corneal center after a complete peritomy)
to a few months (advancement flaps of conjunctiva and Tenons
capsule drawn over a sector of cornea).5 Gundersons technique
of removing Tenons fascia was unique in that it reduced flap
contractility, allowing permanent coverage of diseased corneas.
There has been a considerable decrease in use of Gunderson
conjunctival flaps in the last several decades since the advent of
tissue adhesives, bandage contact lenses, and improved antibiotic,
antifungal, and antiviral therapy. These are alternatives that have
recently been made available for the management of corneal
injury. However, flaps are still an appropriate and reliable ther-
apy in selected cases.6 Gunderson flaps may be a temporizing

measure, followed by a penetrating keratoplasty at a later date,
or they may be denitive therapy for chronic conditions. In
addition to the total conjunctival flap described by Gunderson
to cover the entire cornea, there has been an introduction of
other types of flaps for specic uses. Partial advancement flaps,
created by undermining perilimbal conjunctiva to cover an
adjacent peripheral corneal abnormality, are designed to cover a
specic area of diseased cornea. The advantages, unlike a
Gunderson flap, are better visualization of the anterior chamber,
accurate IOP measurements, and suitability in patients with
short fornices.7 Racquet flaps are produced by rotating, or
swinging, a flap of limbal conjunctiva onto the cornea.8 Bucket
handle flaps can be used to provide a richly vascularized tissue
to cover the cornea without the extensive dissection necessary
for total flaps.9
Thick flaps, with the inclusion of Tenons capsule, have been
used in the treatment of corneal disorders leading to stromal
loss. Gunderson modied his own procedure by using thick
conjunctival flaps in the treatment of bullous keratopathy.
Khodadoust and Quinter found success in treating deeper
corneal ulcers and perforations with a thick partial conjunctival
flap which included Tenons capsule.7 Sanitato, Kelley, and
Kaufman have also advocated the use of thick conjunctival flaps
with a keratectomy in peripheral mycotic corneal abscesses,
where there is severe loss of stromal tissue.10 Most recently,
Geria et al reported success in using thick conjunctival flaps
with the use of deep, localized keratectomy in four patients with
infected corneal grafts.11
Conjunctival flaps restore ocular surface integrity, and by
virtue of a robust vascular supply help reduce inflammation and
provide metabolic support for healing. In essence, they function
as biological patches. Inflammation subsides with subsequent scar
formation over the area of refractory ulcerations and necrotic
areas. Prompt relief of pain is also achieved after flap placement.1214 A conjunctival flap provides a degree of tectonic support in thin corneas, although a flap as sole management is
contraindicated in actual perforation.6,15
Postoperative complications are relatively uncommon: these
include flap retraction, conjunctival buttonholes, persistent
epithelial defects (PEDs), epithelial inclusion cysts, and corneal
perforations. Buttonholes can be avoided at the time of the surgery
by avoiding excess tension on the conjunctival flap. The rate of
complications is largely affected by technical skill, patient population, severity of disease, and prior cataract or glaucoma surgery.
INDICATIONS
Indications for conjunctival flaps are listed in Table 64.1.
863
TABLE 64.1. Current Use of Conjunctival Flaps in Corneal

Diseases
Infectious
Herpetic
Fungal
Bacterial
a well vascularized cornea is vital in ridding the infection, since

fungi do not survive well in such oxygen-rich environments.
Thick visual-axis sparing, partial conjunctival flaps have also
shown success in medically resistant fungal keratitis (Candida
albicans) within a corneal graft.11 In essence, conjunctival flaps
permit access of cell-mediated immune modulators within the
cornea. In cases of central progressive fungal ulcers, penetrating
keratoplasty remains the treatment of choice.10
Parasitic
Bacterial Keratitis
Intensive topical and subconjunctival antibiotics are rst-line

treatment for bacterial keratitis. It is particularly important to
identify bacterial agents since virulent species, such as
Pseudomonas, can lead to rapid dissolution of corneal tissue with
descemetoceles and corneal perforations. Conjunctival flaps have
demonstrated utility in cases of progressive ulceration refractory
to medical management. Buxton and Fox reported resolution of
three of four cases of culture-proven Pseudomonas infections
managed by conjunctival flap; failure in the fourth case was largely
attributed to the scleral involvement at the time of surgery.20
Neurotrophic ulcer
Exposure keratopathy
Bullous keratopathy
Stromal Thinning Disorders
Peripheral ulcerative keratitis (PUK)
Moorens ulcer
Tectonic Support
Parasitic Keratitis
INFECTIOUS KERATITIS
SECTION 6
Herpetic Keratitis
Flaps remain useful in cases of herpetic ulcerative stromal
disease refractory to intensive medical management. Prior to
the advent of topical and systemic antiviral therapy, surgeons
used conjunctival flaps as a reliable way to control inflammation and provide pain relief. In a series of patients reported by
Wiedman and Gunderson in 1968, 60 of 177 (34%) conjunctival
flaps were performed for chronic herpetic keratitis.2 Despite
over a 30 year gap in data, flaps remain useful in cases of herpetic ulcerative stromal disease. Herpes simplex and herpes
zoster keratitis accounted for 11 of 33 cases requiring conjunctival flap in a published review by Insler and Pechous.16 Paton
and Milauskas reported 36 eyes with herpetic keratitis requiring
conjunctival flaps. Only four eyes had postoperative persistence
of keratitis, two of which had incomplete coverage caused by
buttonholes.17 In a series of 14 conjunctival flaps published by
Brown et al, nine flaps were performed for herpetic keratitis.18
However, all patients had PEDs with signicant ocular inflammation. After the use of various modalities, including multiple
medications, bandage contact lenses, and tissue adhesives, all
eyes were comfortable by 1 week after flap placement. No patient
had recurrence or deeper inflammation with a one- to 6-year
follow-up. Most recently, Alino et al in 1998 performed the
largest percentage of conjunctival flaps within their study of 61
eyes for patients with herpetic corneal disease (26%). None had
recurrence, and the only postoperative complication encountered
was flap retraction, 8.3% in full conjunctival flap versus 23.1%
in the partial conjunctival flap group. In theory, these percentages should be similar, but gravity and mechanical effect on
the horizontal partial flaps may have predisposed them to more
retraction than expected.19
Fungal Keratitis
864
Treating fungal keratitis is very difcult and challenging. Intensive medical management should be attempted for 1 week, and
if there is progression or lack or response, surgery is indicated.
Timing of the procedure is important to avoid a corneal or
anterior chamber abscess.10,20 Debridement of necrotic material
and lamellar keratoplasty in conjunction with a partial conjunctival flap has been suggested to prevent progression in cases of
peripheral fungal keratitis. Sanitato et al reported resolution of
three cases of fungal keratitis using this technique.10 Establishing
Acanthamoeba keratitis may be diagnosed late due to its similar

presentation as bacterial and herpetic keratitis. In a recent case
series, conjunctival flap with combined lamellar keratoplasty
has been shown to arrest inflammation in cases where medical
therapy provided little benet.21
PERSISTENT EPITHELIAL DEFECTS

Initial measures for treatment of PEDs or painful recurrent
erosions include a trial of lubrication, topical and systemic antiinflammatory medications, punctal plugs, bandage contact
lenses, and tarsorrhaphy. More recently, bronectin, growth
factors, and substance P have shown good results in the treatment of PEDs.2224 Conjunctival flaps can be of benet to those
unresponsive to the measures mentioned above. Replacement of
damaged corneal epithelium by vascularized conjunctiva stabilizes
the surface and prevents further erosions. Lugo and Arentsen
performed total conjunctival flaps in seven patients with neurotrophic ulcerations not amenable to lubricants, patching, and
contact lenses.25 The etiology of corneal anesthesia was secondary to herpes zoster ophthalmicus in six patients and trigeminal
nerve injury in one patient. The neurotrophic ulcer resolved in
all patients after flap placement, with thinning and clearing of
the flap within 3 months.
Epithelial breakdown in dry eye syndromes can occur from
a variety of causes. Management of persistent exposure keratopathy despite aggressive medical therapy can often be achieved
by a conjunctival flap. If there is no improvement with articial
tears, punctal occlusion, therapeutic lenses, and tarsorrhaphy,
conjunctival flaps are very effective in preventing stromal loss.
Exposure keratopathy was the indication in 16% of the patients
requiring conjunctival grafts from 1974 to 1980 in a study by
Hirst and colleagues.26
Bullous keratopathy has been successfully treated with conjunctival flaps, with relief of pain and cessation of recurrent erosions. As mentioned previously, Gunderson reported success
using thick conjunctival flaps (including Tenons capsule) with
a lamellar keratectomy to treat bullous keratopathy.27 On the
contrary, recent authors have found success in using a thin conjunctival flap alone.17,19
STROMAL THINNING DISORDERS

Conjunctival flaps have assumed a lesser role in the management of stromal thinning disorders since the advent of tissue
adhesives. From 1960 to 1974, 37% of the conjunctival flaps

were used for patients with corneal thinning or perforation.26
After the introduction of tissue adhesives in 1974, the reported
percentage of flaps used for these conditions has ranged from
8% to 10%.12,28 However, flaps are still used as adjunctive care
in cases of rheumatoid arthritis, systemic lupus erythematous,
and Moorens ulcer. They can be successfully used in melting
disorders with lamellar or full-thickness grafts to stabilize the
surface and prevent lytic destruction of grafts, but should not be
the sole treatment in cases of corneal perforations. Flaps can
help reform the anterior chamber, but aqueous leakage from the
wound is common. Additionally, the eye may develop endophthalmitis or secondary angle closure.
ALTERNATIVE TO ENUCLEATION
In the blind and painful eye, a conjunctival flap may be an alternative to enucleation, allowing retention of the globe with acceptable cosmesis. The flap can also provide a good surface for the
placement of a cosmetic scleral shell or painted contact lens.29
SURGICAL TECHNIQUE
TOTAL ADVANCEMENT FLAP (MODIFIED
GUNDERSON)
The technique of preparing a thin flap was rst described by
Gunderson in 1958 (Fig. 64.1). Numerous modications have
been proposed since then. The basic procedure is as follows:
FIGURE 64.1. Total (Gunderson) flap technique. Surgeons view.

(a) The eyelids are retracted with an eyelid speculum. With topical and
regional anesthesia, all corneal epithelium is removed with a surgical
blade. A 360 peritomy is performed. (b) The eye is maximally
infraducted by a 60 vicryl suture passed through superior limbus and
clamped inferiorly. A surgical marker delineates a 20 mm horizontal
line located ~1618 mm above the superior limbus. As a rule, allow
1.5 mm of superior conjunctiva for each 1 mm of the vertical diameter
of the cornea. (c) The conjunctival flap is carefully dissected free of
underlying Tenons fascia, taking care to avoid a buttonhole. The flap
is slowly retracted to the superior limbus with a moistened cotton
swab. (d) The dissected flap is placed over the cornea. 90 or 100
nylon sutures secure the inferior and superior edges to underlying
episclera without tension. The sutures are left in place to avoid flap
retraction.
1. Local anesthesia is achieved with lid block and retrobulbar

or peribulbar injection.
2. A lid speculum is placed.
3. All of the corneal epithelium is removed with a blade.
This technique allows flap adherence and prevents
epithelial cyst formation. Any necrotic tissue is also
removed. A supercial keratectomy is unnecessary.
4. A complete 360 peritomy is made. Cautery is used as
needed for hemostasis.
5. The eye is rotated downward with a traction suture
(60 silk) placed through the superior cornea at the
limbus. One can alternatively place a traction suture
through the superior rectus to infraduct the eye.
6. A surgical marking pen is used to mark the area to be
dissected. It is important to mobilize enough conjunctiva
to cover the cornea without tension or traction on the
flap, taking into consideration 1.5 mm of conjunctiva for
each millimeter of cornea to be covered. The vertical
height of the flap should be 1618 mm above the superior
limbus. Superior tarsal conjunctival may be used if
necessary, although it may increase the incidence of
postoperative ptosis. The horizontal extent of the flap
should be 1520 mm.
7. Conjunctiva is separated from Tenons capsule by a
subconjunctival injection of balanced salt solution, 2%
lidocaine with 1:100 000 epinephrine, or air. In order to
avoid a perforation in the flap, the needle is inserted
outside of the intended dissection.
8. The superior edge of the flap is incised horizontally. An
assistant provides exposure by elevating the flap with
smooth forceps. Blunt dissection, with the aid of a moist
cotton-tip applicator, or by spreading and cutting with
blunt scissors, is used to remove any fascial tags and
connections from the underside of the conjunctiva. The
flap should be kept moist with irrigation.
9. Enough conjunctiva should be undermined for a flap large
enough to cover the entire corneal surface. Sufcient
tissue must be available to cover the cornea without
stretch or tension.
10. 100 nylon sutures are used to secure the inferior
edge of the flap to the inferior limbus (mattress or
interrupted), incorporating episclera into the suture
bite. Episcleral anchoring is important to help prevent
any flap retraction. Care should be taken to avoid
inverting the conjunctival flap, which can implant
surface epithelium and lead to cyst formation. The
superior flap margin is secured with 100 nylon
sutures to episclera. The superior defect is left bare to
re-epithelialize.
11. It is preferable to have an intact corneal covering; however,
if the superior flap cannot be mobilized to completely
cover the corneal surface, the inferior edge of the flap may
be sutured onto the cornea to cover the remaining area.
Alternatively, inferior bulbar conjunctiva may be mobilized
and sutured to the superior flap along the cornea.
12. Antibiotic ointment and a pressure patch are applied
overnight. The sutures can be removed in ~1 month.
CHAPTER 64
PARTIAL ADVANCEMENT FLAP

Partial advancement flap technique is described as follows
(Fig. 64.2):
3. All of the corneal epithelium is removed with a blade. This
technique allows flap adherence and prevents epithelial
865
4.
5.
6.
7.
cyst formation. Any necrotic tissue is also removed using

lamellar keratectomy.
In cases of noninfectious peripheral thinning with a
threatened perforation, a partial flap can be used in
conjunction with a scleral or corneal patch graft.
A limbal peritomy is made, extending one clock hour to
either side of the corneal area to be covered. Blunt
dissection is used to undermine the conjunctiva from
Tenons capsule with the use of scissors. Sufcient tissue
must be available to cover the cornea without stretch or
tension.
The partial flap is secured to the corneal surface bed with
interrupted 100 nylon sutures. Care should be taken to
avoid placing any sutures in the visual axis.
Antibiotic ointment and a pressure patch are applied
overnight. The sutures can be removed in several weeks.
SECTION 6
COMPLICATIONS
Buttonholes are repaired using a 100 nylon suture on a tapered
vascular needle. If one occurs near the medial or lateral edge of
the flap, it may be possible to undermine enough conjunctiva to
place the defect away from the cornea surface medially or
laterally. Retraction usually occurs within 2 months of surgery,
resulting from the tension placed on the inferior anastomosis
and subsequent tearing away of the suture lines. Alino in 1998
found retraction in 11.4% of 61 eyes, all occurring within
1 month of surgery.19 Retraction can largely be avoided by secure
attachment of flaps to underlying episclera without tension. In
many instances, retraction can be managed by observation if
coverage is sufcient to suppress the inflammatory process.
Inadequate removal of corneal epithelium or inclusion of the
conjunctival epithelium can result in cyst formation. Cysts are
usually located at the limbus, and may be multiple and as large
as one third of the corneal circumference.17 They can be surgically excised if they are problematic. Needle decompression of
a cyst is a temporary solution.27
Postoperative ptosis of 13 mm may occur from downward
traction of the superior fornix when mobilizing large conjunctival flaps to cover a large portion of the cornea.17 Most patients
do tolerate the ptosis well, and it often resolves with time. However, it is important to avoid resecting Muellers muscle when
dissecting high in the superior fornix. Recurrences of infection,
though rare, can occur up to 2 years postoperatively.30 Cultureproven herpes simplex type I has been reported in two cases.
One case resolved with topical antivirals and corticosteroids
and the second required penetrating keratoplasty.31 Erosion of
conjunctival flaps has also been reported in cases of chronic
herpetic keratitis and Moorens ulcer.17

866
Amniotic membrane is a biological fetal tissue that has been

used for corneal and conjunctival reconstruction in a variety of
ocular surface disorders (Table 64.2). It has been used for almost
FIGURE 64.2. Partial conjunctival flap

technique. Surgeons view. (a) The eyelids are
retracted with an eyelid speculum. With topical
and subconjunctival or peribulbar anesthesia,
loose and necrotic epithelium is surgically
removed. (b) A partial flap is created by a
peritomy, followed by blunt dissection. (c) The
flap is advanced and secured over the affected
area without tension by 90 or 100 nylon
sutures.
TABLE 64.2. Current Uses of Amniotic Membrane

Transplantation in Ocular Surface Disorders
Conjunctival Lesions
Pterygium
Intraepithelial tumors/lesions
Cicatricial Ocular Disorders
Symblepharon
Scleromalacia
Corneal Reconstruction
Corneal ulcers
Nontraumatic perforations
Bullous keratopathy
100 years to aid in tissue regeneration of cutaneous and

mucosal lesions. Davis was rst to describe the use of amniotic
membrane for skin transplantation in 1910.32 De Rotth in 1940
and Sorsby in 1947 were the rst to use amniotic membranes
for ocular surface reconstruction.33,34 Its use was hampered by
problems with its processing and preservation. However, in
1995, Kim and Tseng helped reintroduce amniotic membrane
into ophthalmology, thus reviving interest for its use.35
Amniotic membrane is the innermost layer of the fetal membrane, consisting of a single layer of epithelial cells attached to
a thick basement membrane maintained structurally by an underlying avascular stromal matrix. Amniotic membrane is a useful
biological substrate for treating ocular surface disorders largely
because of its antiinflammatory, antiangiogenesis, and antibrotic effects. It promotes epithelial regeneration through
growth factors and prevention of apoptosis. The smooth
basement membrane serves as a compatible surface for epithelial migration. The tissue is slowly resorbed with time.3638
INDICATIONS
Amniotic membrane has two main indications: it can be used
as a graft (with basement membrane side facing up) or as a protective patch (with basement membrane down). Grafts are generally used when there is a tissue defect, such as neurotrophic
corneal ulcers or after resection of conjunctival or corneal tissue.
Several layers (up to three or four) can be used at times to cover
the bed of an ulcer. The objective is to achieve closure of the
wound as quickly as possible, while stimulating epithelium to
grow over the amniotic membrane. When used as a patch,
amniotic membrane effectively protects the underlying epithelium and aids in its regeneration. This is effectively used in persistent corneal epithelial defects, where the amniotic membrane
CONJUNCTIVAL GRAFTS
Amniotic membrane has been shown to be a suitable alternative
to conjunctival autografts in ocular surface disorders, especially
in those involving extensive resections of conjunctival lesions
such as pterygia, tumors, intraepithelial lesions, and symblepharon.36,3942 It may be employed when insufcient or inadequate conjunctiva exists adjacent to the scleral defect, as in
scarring from prior surgeries, injury, pemphigoid, or autoimmune disease. Amniotic membrane has recently been used for
treatment of scleromalacia, where the surrounding conjunctiva
was not suitable for grafting due to the size of the defect.43
Pterygium
Its use in primary pterygium excision has shown similar results
to conjunctival autografts, mainly ranging from 3.0% to 15%
recurrence rates and between 9.5% and 38% recurrence rates for
recurrent pterygia.39,40,4447 Amniotic membrane grafts can be
used as an alternative for pterygium excision, especially those
involving a large surface area. Adjunctive therapies such as beta
radiation and MMC can be avoided when using amniotic membrane alone, therefore, reducing the risk of serious complications
such as scleral necrosis.
Tumors and Intraepithelial Lesions

Amniotic membrane transplantation has been used successfully
in the treatment of ocular surface neoplasias, including conjunctival intraepithelial neoplasia, primary acquired melanosis,
and malignant melanoma. The advantages to amniotic membrane are that it can cover large areas of excision and heals well
despite adjunctive cryotherapy. Espana showed complete epithelial healing in all 16 patients who underwent single-layer amniotic
membrane grafts after removal of large neoplastic lesions up to
20 mm in square diameter.48
Cicatricial Ocular Surface Diseases

Amniotic membrane is also increasingly being used to rehabilitate
patients with conjunctival and corneal stem cell dysfunction
and deciency, in conjunction with limbal stem transplantation.
In patients with severe cicatricial ocular surface diseases, including chemical/thermal burns, StevensJohnsons syndrome, and
ocular cicatricial pemphigoid, amniotic membrane grafts have
been shown to be helpful, especially when early surgical intervention is taken to suppress ocular surface inflammation.49,50
Amniotic membrane has shown good results in restoring a deep
fornix after symblepharon lysis in 12 of 17 eyes.39 On the contrary, patients with severe dry eyes (Schirmers value less than
5 mm in 15 min) have been shown to respond poorly to amniotic
membrane grafts.51,52 Inadequate lubrication is a signicant risk
factor for surgical failure in such populations. But with sufcient tear lm production and with early surgical intervention,
amniotic membrane has properties to repopulate the diseased
external surface damaged from the chemical injury as seen in
burns, and allows signicant healing to occur.41,53
CORNEAL GRAFTS
1.5 mm in diameter, in 11 of 15 eyes (73%).55 For larger corneal

perforations, Hick proposed the use of brin glue with multilayer amniotic membrane in larger corneal perforations. In 14 eyes
with perforations up to 3 mm in diameter, 13 (92.9%) showed
overall success with the combined use of brin glue and
amniotic membrane.56
Bullous Keratopathy
Amniotic membrane has been successfully used to treat symptomatic bullous keratopathy, especially in cases of intractable
pain with poor visual potential or when penetrating keratoplasty
is not indicated. The treatment usually involves complete epithelial debridement of the affected area, with subsequent placement of an amniotic membrane graft. Gris showed complete
epithelialization in all ve bullous keratopathy cases within the
rst 16 days after the implants.57 Espana reported long-term
success in 17 of 18 eyes with 88% of the patient group achieving
immediate pain relief.58

Corneal epithelial defects usually heal well without any complications, but any insult that prevents proper epithelial wound
healing can lead to a condition called PED. Neurotrophic keratopathy, exposure keratopathy, and cicatricial ocular diseases,
such as chemical/thermal burns, StevensJohnsons syndrome,
and ocular cicatricial pemphigoid, can predispose patients to
PED. Current treatments include a trial of lubricants, topical and
systemic antiinflammatory medications, punctal plugs, bandage
contact lens, conjunctival flap, amniotic membrane transplantation, and tarsorrhaphy.45 More recently, epithelial healing has
been achieved with the use of bronectin, growth factors, and
substance P.2224 After failed medical therapy, amniotic membrane
transplantation has been shown to promote epithelial healing,
reduce vascularization, and yield good cosmesis. Prabhasawat
reported long-term healing in up to 82% of patients with PED,
with and without stromal thinning and perforation.44 The mean
healing time was observed at 2.1 weeks, with signicantly shorter
times in those that underwent multilayer amniotic membrane
versus single-layer transplantation.

SURGICAL PROCEDURE
Amniotic membrane transplantation surgical procedure is
described as follows (Fig. 64.3):
CHAPTER 64
protects the cornea from the mechanical forces of blinking and

exposure, allowing epithelial growth and adhesion.
Corneal Ulcers and Perforations

Amniotic membrane grafts have had success in treating nontraumatic corneal perforations and ulcers. One or more layers
are used for the treatment of corneal ulcers. Solomon et al had
a successful outcome in 28 of 34 eyes (82.3%) of patients with
deep ulcers and descemetoceles.54 RodriguezAres reported
successful long-term outcomes in corneal perforations, less than
FIGURE 64.3. Surgical technique of amniotic membrane graft

placement. (a) After removal of a pterygium, hemostasis of the bare
bed is achieved with light cautery. (b) Preserved or freeze-dried
amniotic membrane is placed over the bed, basement membrane
(epithelial) side up. The graft is oversized by 20%. Either tissue glue or
dissolvable sutures secure the graft to underlying episclera.
867

3. Conjunctiva is incised horizontally along the diseased
surface area. If Tenons fascia is disease-free, the
conjunctiva is undermined from Tenons fascia to allow the
tissue to retract to its normal anatomical position. The
adjacent extraocular muscles are identied and hooked.
If there is diseased conjunctiva near the muscle insertions,
careful blunt dissection of the abnormal tissue around the
muscle sheaths should be performed. Cautery is used for
hemostasis of the exposed sclera.
4. If cornea is involved, affected corneal epithelium is
removed with a blade. This technique allows better graft
adherence.
5. An amniotic graft is prepared from the preserved
membranes, by measuring it to be ~20% larger than the
corresponding area of conjunctival and/or corneal defect.
Sufcient tissue must be available to cover the area
without stretch or tension.
6. If the amniotic membranes function is to serve as a graft,
it should be placed basement membrane side up on the
scleral and/or corneal surface and secured to the
conjunctival edge with 80 vicryl sutures, incorporating
episclera into the suture bite. If the amniotic membranes
function is to serve as a patch, it should be placed
basement membrane side down (stromal matrix side up)
on the scleral and/or corneal surface and secured to the
conjunctival edge with 80 vicryl sutures, incorporating

episclera into the suture bite. Episcleral anchoring is
important to help prevent any graft or patch retraction.
7. Antibiotic ointment and a pressure patch are applied
overnight. The sutures can be removed in several weeks.
SUMMARY
Conjunctival flaps have been used in ophthalmic surgery for over
a century and remain a viable procedure in certain instances.
Although improvements in topical medications, bandage contact
lenses, and bioadhesives afford the surgeon additional treatment
options, the Gunderson flap and the partial conjunctival flap
are still reliable methods to arrest progression in inflammatory
and ulcerative conditions. In addition, flaps are generally helpful
in reducing chronic pain. Amniotic membrane is an alternative
approach for ocular surface reconstruction. It has demonstrated
utility in covering large conjunctival and corneal defects where
autografts or flaps are not possible. Amniotic membrane creates a
clinically useful, albeit temporary, antiangiogenic and antiinflammatory environment. It may also serve as a substrate for regeneration of normal phenotype epithelium. Advancements in
understanding the biology of corneal thinning and wound repair
will hopefully lead to additional renements in the surgical
management of these potentially blinding conditions.
SECTION 6
REFERENCES
868
1. Byers WGM: Conjunctival flaps in

ophthalmic surgery. Trans Am Ophthalmol
Soc 1909; 12:389.
2. Wiedman MS, Gundersen T: Conjunctival
flaps and cautery. Int Ophthalmol Clin
1968; 8:637653.
3. Green J: Conjunctivoplasty in certain
corneal affections. Trans Am Ophthalmol
Soc 1927; 25:39.
4. Haik GM: A fornix conjunctival flap as a
substitute for the dissected conjunctival
flap: a clinical and experimental study.
19541955; 52:497524.
5. Gunderson T: Conjunctival flaps in the
treatment of corneal disease with reference
to a new technique of application. Arch
6. Arentsen JJ, Morgan B, Green WR:
Changing indications for keratoplasty.
7. Khodadoust A, Quinter AP: Microsurgical
approach to the conjunctival flap. Arch
Ophthalmol 2003; 121:11891193.
8. Cies WA, Odeh-Nasrala N: The racquet
conjunctival flap. Ophthalmic Surg 1976;
7:3132.
9. Nichols BD: Conjunctival flaps. In:
Krachmer JH, Mannis MJ, Holland EJ, eds.
Cornea: surgery of the cornea and
conjunctiva. 3rd edn. St Louis, MO: Mosby;
1997:19031909.
10. Sanitato JJ, Kelley CG, Kaufman HE:
Surgical management of peripheral fungal
keratitis (keratomycosis). Arch Ophthalmol
1984; 102:15061509.
11. Geria RC, Wainsztein RD, Brunzini M, et al:
Infectious keratitis in the corneal graft:
treatment with partial conjunctival flaps.
Ophthalmic Surg Lasers Imaging 2005;
36:298302.
12. Kenyon KR, Wagoner MD, Hettinger ME:
13. Thoft RA: Conjunctival transplantation.

14. Coster DJ, Aggarwal RK, Williams KA:
Surgical management of ocular surface
disorders using conjunctival and stem cell
allografts. Br J Ophthalmol 1995;
79:977982.
15. Mannis MJ: Conjunctival flaps. Int
16. Insler MS, Pechous B: Conjunctival flaps
revisited. Ophthalmic Surg 1987; 18:455458.
17. Paton D, Milauskas AT: Indications, surgical
technique, and results of thin conjunctival
flaps on the cornea: a review of 122
consecutive cases. Int Ophthalmol Clin
1970; 10:329345.
18. Brown DD, McCulley JP, Bowman RW,
Halsted MA: The use of conjunctival flaps
in the treatment of herpes keratouveitis.
Cornea 1992; 11:4446.
19. Alino AM, Perry HD, Kanellopoulos AJ, et al:
Conjunctival flaps. Ophthalmology 1998;
105:11201123.
20. Buxton JN, Fox ML: Conjunctival flaps in
the treatment of refractory Pseudomonas
corneal abscess. Ann Ophthalmol 1986;
18:315318.
21. Cremona G, Carrasco MA, Tytiun A,
Cosentino MJ: Treatment of advanced
acanthamoeba keratitis with deep lamellar
keratectomy and conjunctival flap. Cornea
2002; 21:705708.
22. Spigelman AV, Deutsch TA, Sugar J:
Application of homologous bronectin to
persistent human corneal epithelial defects.
Cornea 1987; 6:128130.
23. Feldman ST: The effect of epidermal growth
factor on corneal wound healing: practical
considerations for therapeutic use. Refract
Corneal Surg 1991; 7:232239.
24. Brown SM, Lamberts DW, Reid TW, et al:
Neurotrophic and anhidrotic keratopathy
treated with substance P and insulinlike
growth factor 1. Arch Ophthalmol 1997;
115:926927.
25. Lugo M, Arentsen JJ: Treatment of

neurotrophic ulcers with conjunctival flaps.
Am J Ophthalmol 1987; 103:711712.
26. Hirst LW, Smiddy WE, Stark WJ: Corneal
perforations. Changing methods of
treatment, 19601980. Ophthalmology
1982; 89:630635.
27. Gunderson T, Pearlson HR: Conjunctival
flaps for corneal disease: their usefulness
and complications. Trans Am Ophthalmol
Soc 1969; 67:78.
28. Arentsen JJ, Laibson PR, Cohen EJ:
Management of corneal descemetoceles
and perforations. Ophthalmic Surg 1985;
16:2933.
29. Townsend WM: Conjunctival flaps. In:
Kaufman HE, ed. The cornea. New York,
NY: Churchill Livingstone; 1988:669681.
30. Rosenfeld SI, Alfonso EC, Gollamudi S:
116:242244.
31. Lesher MP, Lohman LE, Yeakley W, Lass J:
Recurrence of herpetic stromal keratitis
after a conjunctival flap surgical procedure.
Am J Ophthalmol 1992; 114:231233.
32. Davis JW: Skin transplantation. Johns
Hopkins Med J 1910; 15:307396.
33. De Rotth A: Plastic repair of conjunctival
defects with fetal membranes. Arch
Ophthalmol 1940; 23:522525.
34. Sorsby A, Symons HM: Amniotic
membrane grafts in caustic burns of the
eye burns of the second degree. Br J
35. Kim JC, Tseng SC: Transplantation of
preserved human amniotic membrane
for surface reconstruction in severely
damaged rabbit corneas. Cornea 1995;
14:473484.
36. Shimazaki J, Shinozaki N, Tsubota K:
Transplantation of amniotic membrane and
limbal autograft for patients with recurrent
pterygium associated with symblepharon.
45. Ma DH, See LC, Liau SB, Tsai RJ: Amniotic
membrane graft for primary pterygium:
comparison with conjunctival autograft and
topical mitomycin C treatment. Br J
Ophthalmol 2000; 84:973978.
46. Ivekovic R, Mandic Z, Saric D, Sonicki Z:
Comparative study of pterygium surgery.
47. Tekin NF, Kaynak S, Saatci AO, Cingil G:
Preserved human amniotic membrane
transplantation in the treatment of primary
pterygium. Ophthalmic Surg Lasers 2001;
32:464469.
48. Espana EM, Grueterich M, Sandoval H,
for bullous keratopathy in eyes with poor
visual potential. J Cataract Refract Surg
2003; 29:279284.
49. John T, Foulks GN, John ME, et al:
Amniotic membrane in the surgical
management of acute toxic epidermal
necrolysis. Ophthalmology 2002;
109:351360.
50. Di Pascuale MA, Espana EM, Liu DT, et al:
Correlation of corneal complications with
eyelid cicatricial pathologies in patients
with StevensJohnson syndrome and toxic
epidermal necrolysis syndrome.
51. Jain S, Rastogi A: Evaluation of the
outcome of amniotic membrane
transplantation for ocular surface
reconstruction in symblepharon. Eye 2004;
18:12511257.
52. Shimazaki J, Shimmura S, Fujishima H,
Tsubota K: Association of preoperative tear
53.
54.
55.
56.
57.
58.
function with surgical outcome in severe

StevensJohnson syndrome.
Ophthalmology 2000; 107:15181523.
Fournier JH, McLachlan DL: Ocular surface
reconstruction using amniotic membrane
allograft for severe surface disorders in
chemical burns: case report and review of
the literature. Int Surg 2005; 90:4547.
Solomon A, Meller D, Prabhasawat P, et al:
109:694703.
Rodriguez-Ares MT, Tourino R, LopezValladares MJ, Gude F: Multilayer amniotic
membrane transplantation in the treatment
of corneal perforations. Cornea 2004;
23:577583.
Hick S, Demers PE, Brunette I, et al:
ulcers and perforations: a review of 33
cases. Cornea 2005; 24:369377.
Gris O, Lopez-Navidad A, Caballero F, et al:
ocular surface pathology: long-term results.
Transplant Proc 2003; 35:20312035.
Espana EM, Prabhasawat P, Grueterich M,
for reconstruction after excision of large
ocular surface neoplasias. Br J Ophthalmol
2002; 86:640645.
CHAPTER 64
37. Lee HS, Kim JC: Effect of amniotic fluid in

corneal sensitivity and nerve regeneration
after excimer laser ablation. Cornea 1996;
15:517524.
38. Tsubota K, Satake Y, Ohyama M, et al:
Surgical reconstruction of the ocular surface
in advanced ocular cicatricial pemphigoid
and StevensJohnson syndrome. Am J
Ophthalmol 1996; 122:3852.
39. Solomon A, Pires RT, Tseng SC: Amniotic
membrane transplantation after extensive
removal of primary and recurrent pterygia.
40. Prabhasawat P, Barton K, Burkett G,
Tseng SC: Comparison of conjunctival
autografts, amniotic membrane grafts, and
primary closure for pterygium excision.
41. Tseng SC, Prabhasawat P, Lee SH:
conjunctival surface reconstruction. Am J
Ophthalmol 1997; 124:765774.
42. Azuara-Blanco A, Pillai CT, Dua HS:
ocular surface reconstruction. Br J
Ophthalmol 1999; 83:399402.
43. Oh JH, Kim JC: Repair of scleromalacia
using preserved scleral graft with amniotic
membrane transplantation. Cornea 2003;
22:288293.
44. Prabhasawat P, Tesavibul N,
Komolsuradej W: Single and multilayer
amniotic membrane transplantation for
persistent corneal epithelial defect with and
without stromal thinning and perforation.
Br J Ophthalmol 2001; 85:14551463.
869
CHAPTER
65
Ocular Surface Transplantation

William B. Lee and Ivan R. Schwab
The ocular surface, while initially considered as only an anatomical classication, is now known to serve as one functional
unit, which encompasses the entire external covering of the eye
including the epithelial layers of the cornea, limbus, palpebral,
and bulbar conjunctiva along with the overlying tear lm. Each
individual portion of the ocular surface carries its own important signicance for the overall health and maintenance of
the eye. Any process or disease that compromises the integrity
of any one of these structures can create an unstable ocular surface and promote a myriad of corneal and conjunctival ndings
ranging from a mild corneal abrasion to severe stem cell loss
that can ultimately result in decreased vision and potential blindness or loss of the eye. With our evolving knowledge of corneal
epithelial stem cells and advancing treatment options for ocular
surface disease, it has become more apparent that surgical
forms of treatment for severe ocular surface disease carry the
greatest promise in this challenging and often frustrating class of
disorders. This chapter focuses on stem cell theory and principle
as well as the surgical management of ocular surface diseases
using a variety of limbal stem cell transplantation techniques
for rehabilitation and restoration of the ocular surface. While
this chapter touches on conjunctival grafting and amniotic
membrane transplantation, other book chapters will discuss
these techniques in more detail.
OCULAR SURFACE PRINCIPLES

A healthy ocular surface requires a multitude of factors directly
and indirectly involving the eye and its surrounding structures.
Ocular surface maintenance is dependent on normal anatomy
of the external eye as well as intact neurophysiologic pathways
that maintain proper function and control over eyelid apposition and closure, lacrimal, accessory lacrimal, and meibomian
gland function, tear lm composition, and a healthy conjunctival, limbal, and corneal epithelium and stroma. While adequate health and function of all these structures is imperative
for a stable ocular surface, perhaps the most critical element of
these structures remains the corneal epithelial stem cells. Our
knowledge of ocular surface disorders has undergone tremendous evolution over the last half century as remarkable
advancements have been made with corneal epithelial stem cell
research and ocular surface surgical techniques.
Stem cells are dened as undifferentiated cells that are found
in all self-renewing tissues. They possess the ability to proliferate, produce differentiated daughter cells, self-maintain,
and regenerate after injury.18 Stem cell function and fate is
largely dependent on the stem cell niche, a surrounding microenvironment including interaction with extracellular matrix, local
growth factors, and surrounding cells.911 Our knowledge of
corneal epithelial stem cell origin, location, and function has

remained a relatively new discovery over the last few decades.
Perhaps one of the most important initial observations regarding
early theory of stem cell presence and function began when
Friedenwald observed that the corneal epithelium regenerated
fully after total de-epithelialization.12 In 1971, the stem cell
origin for corneal epithelium regeneration was proposed to
reside in the limbal papillary structures known as the palisades
of Vogt.13,14 While additional research suggested the same origin
of corneal epithelial stem cells, several studies have conrmed
the presence of these stem cells within the limbal basal epithelium by identication of cornea-specic keratins.4,1417 Corneaspecic keratins allow for localization to the limbal basal
epithelium; however, identication of a direct stem cell marker
remains in question.1820 While several reports now suggest that
p63, a transcription factor of p53, is a specic localizing marker
for corneal epithelial stem cells; not all studies conrm this
marker to be cell specic to the corneal stem cells.20
Corneal epithelial stem cells are the progenitor cells for the
entire corneal epithelial covering. It is believed that adult stem
cells are found in limited number with long life spans, slow cell
cycling capabilities including low mitotic activity, and are less
differentiated under normal homeostatic conditions.2126 However, they possess the ability to regenerate and repair tissue after
injury. When activated, they divide to produce daughter cells
known as transient amplifying cells that are responsible for
proliferation, differentiation, and migration in response to normal
physiologic renewal or repair in response to injury.19,18,27,28
Daughter cells in contrast have short life spans, rapid cell
cycling, and high mitotic activity (Fig. 65.1).15,11,2729 Additional
studies postulated the migratory pattern of epithelial cells following repair after injury.29,30 Thoft and Friend proposed the
XYZ hypothesis of corneal epithelial maintenance in which
basal epithelial cells (X) and cells from the periphery (Y) divide
and replace the desquamated surface cells (Z).30 Transient
amplifying cells migrate centripetally from the limbus and
vertically from the basal epithelial layers forward to promote
epithelial renewal.2931 This process of epithelial cell migration
is critical in maintenance of the corneal epithelial mass and its
ability to regenerate after injury.
When damage occurs either directly or indirectly to the
limbal stem cells or their stem cell niche, the ability for epithelial renewal and repair is lost and signicant complications
can occur resulting in ocular surface injury.
The diagnosis of ocular surface disorders including stem cell
deciency can present with a variety of symptoms and signs.
Symptoms at presentation may include foreign body sensation,
pain, dryness, increased tearing, redness, photophobia, decreased vision, and blepharospasm. A review of ocular history
871
CORNEAL AND CONJUNCTIVAL EPITHELIA
conjunctiva
limbus
peripheral cornea
central cornea
mature conjuntival cells
FIGURE 65.1. Schematic diagram depicting

the anatomical orientation of the conjunctiva,
limbus, and corneal epithelium including the
location of stem cells and transient amplifying
daughter cells within the basal corneal
epithelium.
transient amplifying cells

basal layer
stem cells
mature corneal cells
KEY
mature corneal cell AES positive
transient amplifying cells AEI/CK
19 positive
SECTION 6
stem cells
may elicit a history of painful corneal erosions resulting from

epithelial breakdown or chronic inflammatory episodes with
redness, dryness, and surface irritation. Biomicroscopy ndings
at the slit-lamp can reveal important ndings indicative of stem
cell disease within the eyelids, bulbar and palpebral conjunctiva,
and the corneal surface. Eyelid margins can develop trichiasis,
districhiasis, entropion, ectropion, chronic injection, keratinization, symblepharon, and ankyloblepharon. Conjunctival ndings
may include similar ndings of inflammation including subepithelial brosis and scarring, shortened fornices, hyperemia,
and symblepharon. Persistent conjunctival inflammation leads
to goblet cell and accessory gland damage resulting in mucin
tear deciency and aqueous tear deciency. Early slit lamp
ndings of the cornea include loss of palisades of Vogt, late
staining of the corneal epithelium with fluorescein, corneal
neovascularization, and localized or diffuse pannus formation
(Fig. 65.2a,b). As stem cell disease progresses the corneal epithelial mass cannot maintain its regenerative functions and the
epithelium becomes irregular and hazy. Punctate keratitis and
epithelial defects can develop with subsequent corneal scarring
and calcication. Presence of chronic epithelial defects can predispose to corneal ulceration, melting, perforation, and/or infection. As conjunctivalization of the cornea develops, histological
diagnosis of stem cell disease can be made by demonstrating
presence of conjunctival goblet cells in the corneal epithelium
with excisional corneal biopsy or impression cytology. Alcian
blue and periodic acid-Schiff stains identify glycosaminoglycans
within the goblet cells of the tissue specimens.
LIMBAL STEM CELL TRANSPLANTATION

Types of Limbal Stem Cell Deciency
Primary
Secondary
CAUSES
A multitude of conditions may lead to stem cell deciency and
adversely impact the ocular surface (Table 65.1). Stem cell
disorders can be grouped into primary and secondary causes.
Primary disorders represent congenital abnormalities, while
secondary disorders develop from external factors that promote
damage and progressive stem cell loss over time. Primary stem
cell diseases lead to improper development of the anterior segment and result in a dysfunctional or decient stem cell population. The stem cell deciency from these congenital conditions
may manifest at birth or show delayed progression later in life.
Examples of these disorders include aniridia, autosomal
dominant keratitis, and sclerocornea. Aniridia represents the
most common cause of congenital stem cell deciency. Other
congenital stem cell diseases include multiple endocrine
neoplasia, chronic mucocutaneous candidiasis, and ectodermal
dysplasia syndromes, of which over 150 separate forms exist.
Ichthyosis-related diseases such as keratitisichthyosisdeafness
(KID) syndrome lead to a severe stem cell deciency (Fig. 65.3).
FIGURE 65.2. (a) Late fluorescein staining of
the cornea from superior stem cell deciency
on slit lamp photography. (b) Severe corneal
neovascularization and pannus from stem cell
damage in TEN.
872
TABLE 65.1. Causes of Limbal Stem Cell Deciency

Primary Stem Cell Disease
Aniridia
Sclerocornea
Autosomal dominant keratitis
Multiple endocrine neoplasia
Ectodermal dysplasia syndromes
Secondary Stem Cell Disease
Direct stem cell loss
Alkali or acid injury
Thermal injury
Secondary, or acquired, stem cell disease most commonly

includes conditions that lead to direct stem cell loss such as
chemical, thermal, or radiation exposure. Autoimmune disorders
such as toxic epidermal necrolysis (TEN), mucous membrane
pemphigoid, StevensJohnson syndrome, and various collagen
vascular diseases represent potentially devastating secondary
causes of stem cell deciency. Additional secondary causes include
chronic nonautoimmune inflammatory disorders and iatrogenic
stem cell deciency. Chronic nonautoimmune inflammatory
conditions include damage from contact lens wear as well as
ocular surface diseases such as rosacea, Staphylococcus marginal disease, herpetic keratitis, pterygia, and limbal neoplasias.
Iatrogenic stem cell damage is typically seen following multiple
surgeries, cryotherapy to the limbus, widespread limbal conjunctival neoplasm excision (pterygia or conjunctival tumors),
and medication toxicity.
Radiation injury
Autoimmune disease
Toxic epidermal necrolysis
Mucous membrane pemphigoid
Collagen vascular diseases
Chronic non-autoimmune inflammatory disorders
Atopy
Ocular rosacea
Contact lens wear
Bacterial, viral, fungal keratitis-related stem cell loss
Staph. marginal disease
Pterygia/pseudopterygia
Xerophthalmia
Limbal neoplasm
Iatrogenic stem cell deciency
Multiple ocular surgeries
Excision of pterygia
Excision of limbal neoplasm
Cryotherapy/cyclophotocoagulation of limbus
SURGICAL ADVANCEMENTS
One of the rst reports of restored ocular surfaces occurred in
1951 when Hartman used a free conjunctival graft for correction of pterygia, pseudopterygia, and symblepharon.32 This
report suggested the benet of using conjunctiva for grafting
procedures and proposed the idea of using conjunctiva for
unilateral disease with harvesting from the contralateral eye in
certain cases.32 Conjunctiva is now commonly used to correct
conjunctival defects in a variety of conditions as well as an
adjunct for restoration of corneal epithelial stem cells in certain
ocular surface disease situations (see Table 65.2). In fact, autologous conjunctival grafting has now become the gold standard
for ocular surface diseases such as pterygia and pseudopterygia
surgical resection for many surgeons.3335 Despite the various
benets of conjunctiva for transplantation, a void remained for
surgical correction of stem cell diseases of the ocular surface.
Ocular surface rehabilitation has remained a difcult and
daunting task for treatment of the variety of stem cell disease
states. Advances in stem cell research have led to a better
understanding of surgical principles and techniques while
improving the surgical success of ocular surface diseases, and in
particular, stem cell deciency (see Table 65.2). While Barraquer
is credited as the rst surgeon to describe stem cell transplant
techniques in ocular surface chemical burns,36 Thofts work
remains the cornerstone for our understanding of ocular surface
Medication toxicity
Conjunctival Graft
Primary and recurrent pterygia
Cicatricial strabismus
Fornix reconstruction (unilateral disease)
Post-resection of benign/malignant conjunctival neoplasm
CHAPTER 65
TABLE 65.2. Indications for Ocular Surface Transplantation
Filtering bleb repair

Adjunct with limbal stem cell transplant surgery
Limbal Stem Cell Transplantation
Primary stem cell disease
Secondary stem cell disease
Direct stem cell loss
Autoimmune disease
FIGURE 65.3. Slit lamp photograph depicting corneal
neovascularization and stem cell deciency associated with KID
syndrome.
Chronic non-autoimmune inflammatory disorders

Iatrogenic stem cell deciency
873

disease and treatment.37,38 Thoft used autologous conjunctival
transplantation for the treatment of ve cases involving unilateral chemical corneal burns using four conjunctival grafts
from the uninvolved eye. Four 5 mm conjunctival grafts were
harvested from each of the four bulbar conjunctival quadrants
in the normal eye and secured to four quadrants of the diseased
limbus after a lamellar keratectomy and 360 limbal conjunctival resection. Three of the ve eyes showed signicant
improvement of the ocular surface and vision with one eye
reaching 20/30 at 8 months postoperatively.37 Thoft used a
keratoepithelioplasty in patients with bilateral stem cell disease,
a procedure which laid the ground-work for modern-day limbal
stem cell transplantation techniques (Fig. 65.4). The technique
SECTION 6
utilized four lenticules which included epithelium and a thin

layer of stroma harvested from the midperipheral limbus of a
donor globe. Each lenticule was secured around the corneoscleral limbus of the damaged ocular surface in four different
quadrants.38 While the keratoepithelioplasty procedure was the
rst attempt to transplant corneal epithelial stem cells in
patients with severe bilateral ocular surface disease, limbal stem
cell importance in ocular surface disease was not completely
understood at that time.
With anatomical conrmation of the precise location of the
corneal stem cells, additional techniques of limbal transplantation emerged as a treatment for ocular surface restoration.
Kenyon and Tseng built upon Thofts principles of conjunctival
transplantation to include limbal stem cells in their grafts for
the treatment of severe ocular surface diseases.39 This limbal
autograft technique called for transplantation of donor conjunctiva extending 0.5 mm onto the peripheral cornea from the
normal fellow to the diseased eye. Twenty of 21 eyes had
improved ocular surfaces and 17 eyes developed improved
visual acuity using this technique.39 Tsai and Tseng, followed by
Tsubota and colleagues, further modied the technique by using
a cadaveric keratolimbal graft harvested from a whole globe to
create a donor keratolimbal ring.40,41 The cadaveric tissue was
transplanted to the diseased corneoscleral limbus after appropriate preparation of the donor rim of tissue (Fig. 65.5). This
technique avoided the potential complications of autologous grafts
and prevented iatrogenic stem cell deciency in the normal
donor eye. Tsubota and colleagues described a novel surgical
technique known as limbal allograft transplantation. The
technique utilized a stored corneoscleral rim for transplantation
to the limbus of the diseased eye.41 Holland described a similar
technique in which he used two corneoscleral rims for transplantation with each rim divided into two, creating four cadaveric
tissue segments.42 The stem cells were harvested from all four
quadrants and three of the four healthiest segments were
chosen for transplantation to the diseased limbus. This method
completely surrounded the host limbus avoiding gaps of
exposed conjunctiva, and it provided 1.5 times more stem cells
to the host limbus than a single donor corneoscleral rim.42
b
FIGURE 65.4. Keratoepithelioplasty as described by Thoft. (a) Four
lenticules are harvested from a donor globe. (b) The lenticules are
secured to the diseased corneoscleral limbus in equidistant positions.
Reproduced from Am J Ophthalmol.
874
FIGURE 65.5. Schematic diagram depicting a KLAL using an entire

ring of donor keratolimbal tissue from the donor cadaveric globe. The
donor graft contains a portion of the peripheral donor cornea, the
donor limbus, and the anterior portion of donor conjunctiva and
episclera.
Limbal Stem Cell Transplant Techniques

1.
2.
3.
4.
5.
Conjunctival limbal autograft

Unilateral stem cell disease with normal fellow eye
Living related conjunctival limbal allograft
Unilateral stem cell disease with some degree of stem cell
deciency in fellow eye
Bilateral stem cell disease
Cadaveric keratolimbal allograft
Unilateral stem cell disease where use of living-related or
fellow eye raises concern for iatrogenic stem cell damage
Bilateral stem cell disease
Combined conjunctivalkeratolimbal allograft
Unilateral or bilateral stem cell disease with cicatricial
conjunctival disease including pronounced conjunctival
tissue loss
Ex vivo stem cell expansion
Unilateral or bilateral stem cell disease with avoidance of
cadaveric tissue use or iatrogenic stem cell damage
We currently utilize ve limbal stem cell transplantation procedures in the treatment of diseased ocular surfaces from stem
cell deciency. These include a conjunctival limbal autograft
(CLAU), living related conjunctival limbal allograft (LR-CLAL),
cadaveric keratolimbal allograft (KLAL), combined conjunctivalkeratolimbal allograft (C-KLAL), and ex vivo stem cell
expansion techniques.43
CLAU is the procedure of choice for unilateral limbal stem
cell disease where the fellow eye has completely normal stem
cells. The diseased eye is prepared for grafted tissue by creating

two limbal recipient beds. A superior and inferior conjunctival
peritomy are created, each extending four clock hours with
undermining of the conjunctiva to allow tissue recession. The
brovascular pannus and diseased epithelium can then be
removed in conjunction with diluted topical thrombin to assist
with hemostasis. Separate superior and inferior trapezoidalshaped grafts are harvested from the uninvolved eye including
conjunctival and limbal tissue. Each graft includes ~4 clock
hours of limbal tissue extending 0.5 mm into the peripheral
cornea in conjunction with ~6 mm of conjunctiva posteriorly.
The grafts are then transferred to the same anatomical location
of the stem cell-decient eye. The grafts can be secured to the
recipient bed with 100 nylon sutures (Fig. 65.6ac). Avoid
suture placement at the limbal margin in order to prevent any
additional stem cell damage. Advantages of this procedure include
lack of cadaveric corneal tissue and avoidance for the need of
systemic immunosuppression (Fig. 65.7ac). Disadvantages
include the risk of iatrogenic stem cell deciency created in the
previously normal donor eye.
LR-CLAL is indicated in situations involving unilateral stem
cell damage in which the fellow eye lacks completely normal
stem cell function or in cases of bilateral stem cell disease.
Severe stem cell deciency cases should have an alternative
procedure performed which can supply a greater population of
donor stem cells. This procedure requires two separate surgeries
involving two patients. Two trapezoidal grafts are harvested
from the living-related donor eye in a similar fashion to the
CLAU procedure. An irreversible letter is placed on the surface
of the grafts prior to complete removal and transfer to maintain
proper alignment. The tissue can be placed on glove paper and
FIGURE 65.6. Schematic diagram depicting a CLAU. (a) The recipient eye is prepared with a 360 conjunctival peritomy and conjunctival
resection is performed as shown. (b) One or two conjunctival grafts are harvested from the uninvolved eye (in CLAU), or a living-related donor
eye (in LR-CLAL), including conjunctiva and ~0.5 mm of peripheral cornea as shown. The grafts are marked for identication. (c) The donor
conjunctival graft is secured to the recipient bed with 100 nylon suture as shown.
CHAPTER 65
CURRENT SURGICAL TECHNIQUES
FIGURE 65.7. A patient with inferior sectoral stem cell deciency from an alkaline chemical burn is shown, (a) preoperatively, (b) 1 week
postoperatively following an inferior CLAU, and (c) 1 month postoperatively.
875
SECTION 6
immersed in colloidal storage solution during transfer. The

donor bed of the diseased eye is prepared in a similar fashion to
the diseased eye in CLAU. The grafts are secured to the host
tissue with 100 nylon in the same anatomical fashion. The
advantage of this procedure includes transplantation of conjunctiva with the stem cells, a useful concept in diseases with
conjunctival tissue loss compared to KLAL. Additional
advantages of this procedure include a potential for treatment in
bilateral disease and the lack of cadaveric tissue requirements.
Disadvantages include lack of efcacy in severe stem cell
deciency, required systemic and topical immunosuppression
unlike CLAU, and a risk of creating iatrogenic stem cell
deciency in the donor eye of a relative.
Cadaveric KLAL is indicated for bilateral stem cell deciency.
It may also be utilized in cases of unilateral stem cell deciency
in which a concern for inducing iatrogenic stem cell deciency is
present for the contralateral or living relatives eye. The
recipient bed is rst prepared by a 360 limbal conjunctival
peritomy with undermining of the conjunctiva. Minimal or no
conjunctival resection may be needed if signicant conjunctival
loss or symblepharon are present. A conjunctival resection of
4 mm or less is acceptable in cases with relatively healthy conjunctiva to allow for adequate recipient bed exposure.
Homeostasis can be achieved with topical thrombin, topical
sympathomimetics, and/or wet eld cautery. Next, a supercial
keratectomy is performed with removal of all brovascular
pannus and abnormal epithelium. A diamond-dusted corneal
burr can assist, with removal of diseased tissue with care to
avoid penetration into deep layers of the corneal stroma.
Stem cell harvesting in KLAL requires two corneoscleral
donor tissues preserved in eye bank solution with large scleral
rims measuring at least 14 mm in diameter. Routine keratoplasty
techniques using any trephination system can be employed to
remove a 7.58.0 mm central button within each donor for
adult cases and smaller sizes for pediatric cases. We harvest our
stem cells as described by Mannis and colleagues using a 22 mm
silicone orbital sizing sphere in adults and three 25-gauge
needles for xation of the rims.44 The posterior two-thirds of
each circular rim are dissected from the anterior one-third using
lamellar dissection with a rounded crescent blade (Fig. 65.8).
The posterior tissue is discarded and the residual anterior onethird of both circular rims is bisected with scissors making four
stem cell segments. The healthiest three pieces are then
876
FIGURE 65.8. A crescent blade is used to harvest limbal stem cells

from the anterior one-third of a cadaveric donor corneoscleral rim
using three 25-gauge needles to xate the donor rim on a 22 mm
silicone orbital sizing sphere.
FIGURE 65.9. A schematic diagram depicting a KLAL. After a 360

conjunctival peritomy and conjunctival resection, three of four
segments are positioned around the limbus and secured with 100
nylon sutures, one segment at a time with avoidance of gaps between
tissue segments.
fashioned in the recipient bed in the same anatomical alignment encircling the limbus. The stem cell segments are secured
one at a time at the four corners using 100 nylon to the host
corneal border, followed by securing the conjunctival borders.
Meticulous attention to avoid gaps in the three segments is
essential to avoid conjunctival extension toward the cornea
postoperatively (Fig. 65.9). A bandage contact lens and pressure
patch are applied after topical and subconjunctival medications
have been administered.
Advantages to KLAL include the delivery of a high number of
stem cells for transplantation to the diseased eye allowing for
increased efcacy in cases of bilateral stem cell disease. This
technique delivers 1.5 times more stem cells to the diseased eye
compared to previously reported techniques. It also avoids a risk
of iatrogenic stem cell deciency in donor eyes. KLAL can also
be effective for either complete stem cell deciency or localized
stem cell deciency. Disadvantages of KLAL include intense
systemic and topical immunosuppression requirements with a
potential for serious medication-induced side-effects. Adequate
systemic and topical immunosuppression is critical for survival
of a KLAL. Also KLAL success is decreased in cases with conjunctival inflammation, unstable tear lm, abnormal corneal
sensation, and presence of ocular surface keratinization. For eyes
with conjunctival inflammation, success of surgery is improved
if the inflammation is maximally controlled prior to surgery. Tear
lm and lid abnormalities must be corrected before attempting
KLAL as persistent surface defects following surgery hamper
surgical results. Neurotrophic corneas, ocular surface keratinization, and severe aqueous tear deciency are relative contraindications to KLAL as corneal sensation and a normal tear layer
are crucial for survival of the cadaveric donor stem cells.
Combined C-KLAL is indicated for unilateral or bilateral
stem cell disease in cases of cicatricial conjunctival damage
with pronounced conjunctival inflammation and tissue loss.
This technique is effective in stem cell deciency cases where
conjunctival loss is severe. The recipient bed is prepared in a
similar manner to KLAL; however, conjunctival tissue is
undermined rather than resected after the conjunctival peritomy to spare conjunctival tissue. Stem cells are harvested from
a living-related donor eye as with LR-CLAL and placed in
storage medium. Proper anatomical alignment during transfer

of the transplanted conjunctival tissue to the host is essential.
Stem cell harvesting from the cadaveric corneoscleral tissue is
performed as with KLAL surgery. Only one cadaveric circular
rim is needed with this technique because the conjunctival
segments will provide stem cells to the superior and inferior
four clock hours. After the cadaveric stem cells are obtained the
rim is bisected. The LR-CLAL segments are secured to the
superior and inferior limbus using 100 nylon suture as
described with LR-CLAL. The bisected cadaveric tissue is placed
along the bare nasal and temporal limbus adjacent to the conjunctival grafts. The segments can be trimmed for appropriate
sizing and secured to the recipient beds as with KLAL
(Fig. 65.10). It is essential to avoid gaps in the transplanted
tissue segments as conjunctiva will otherwise course through the
defects and create recurrent corneal vascularization and ocular
surface compromise. A bandage lens and pressure patch are placed
over the eye after topical and subconjunctival medications are
administered to allow for promotion of epithelialization.
Advantages of C-KLAL include the addition of conjunctival
tissue and stem cells to the diseased ocular surface rather than
conjunctiva alone or stem cells alone as in other stem cell
techniques. The procedure affords treatment for severe stem
cell disease including conditions where conjunctival tissue has
been destroyed by inflammation and brosis. Disadvantages of
this technique include the need for two surgeries. One surgery
requires harvesting of cadaveric stem cells and the other surgery
includes acquisition of autologous conjunctiva or conjunctiva
from a living relative without stem cell disease. Iatrogenic stem

cell damage is a risk to the uninvolved donor eye. Intense systemic and topical immunosuppression is also needed to assure
survival of transplanted tissue making the risk of medicationinduced side-effects a possibility. As with KLAL, tear lm and
lid abnormalities must be corrected before surgery to maximize
postoperative success. Neurotrophic corneas, ocular surface
keratinization, and severe aqueous tear deciency also represent
relative contraindications to surgery as with KLAL.
EX VIVO STEM CELL EXPANSION

The newest technique of stem cell transplantation surgery is ex
vivo stem cell expansion (Fig. 65.11). This technique utilizes a
small autologous limbal stem cell biopsy from a donor eye
which acts as the nidus for stem cell expansion. After harvesting stem cells from the biopsy, the cells are transported to the
laboratory and amplied as described by previous reports.4549
After amplication of the cultured stem cells, they can be placed
on a carrier-substrate where they are allowed to adhere for
several weeks in the laboratory, after which the entire complex
is transplanted onto the diseased ocular surface (Fig. 65.12a,b).
b
f
c
d
g
FIGURE 65.10. A schematic diagram depicting a C-KLAL procedure.

The conjunctival grafts are constructed rst as seen in Figures 65.6
ac. After dissection and subsequent bisection of a corneoscleral
donor rim, the anterior one-third of the two keratolimbal grafts are
secured along the nasal and temporal limbus with 100 nylon suture.
FIGURE 65.11. Schematic of ex vivo stem cell transplantation.

(a) Stem cells are harvested from a 2 mm2 limbal biopsy. (b) Harvested
stem cells are transported to the laboratory in storage medium.
(c) Stem cells and epithelial cells are grown in culture medium.
(d) Stem cells are selected from the formed colonies. (e) The pure
stem cell/epithelial cell culture is transported to amniotic membrane
and allowed to grow and attach. (f) The amniotic membrane and
attached stem cells are transplanted to a diseased recipient eye with
stem cell deciency with placement of the membrane across the
entire cornea or in a doughnut fashion at the encircling the limbus.
(g) The remaining cells can be frozen.
CHAPTER 65
FIGURE 65.12. (a) Slit lamp photography

depicting severe stem cell damage from an
alkaline chemical burn leading to severe
brovascular corneal pannus and dense
conjunctival brosis and symblephara.
(b) The same eye 1 week after ex vivo stem cell
expansion using a collagen carrier for
autologous stem cells from the fellow eye.
877
SECTION 6
878
The main advantage to this technique is the potential to

provide expanded limbal stem cells to a diseased ocular surface
with relative sparing of the donor eye stem cells. The use of a
small tissue biopsy of only 12 mm2 in this technique avoids
one of the main disadvantages to CLAU, KLAL, and LR-CLAL,
and C-KLAL, which is potential for iatrogenic stem deciency
in the donor eye. The disadvantages of this technique include
potential fragility of bioengineered tissue, expense, need for
laboratory space, and the need for research manpower for help
with stem cell amplication.
While the idea of cultured corneal epithelial stem cells was
considered in 1982,50 the rst clinical reports of cultured
autologous limbal stem cell transplantation did not occur until
1996 and 1997.45,51 Tor and colleagues rst reported success
with cultured autologous grafts delivered to the damaged eye
demonstrating improved ocular surfaces in three of four
patients with severe unilateral ocular surface disease.45
Pellegrini and colleagues found similar results with restored
ocular surfaces in two patients with severe unilateral stem cell
deciency using autologous cultured corneal epithelial stem
cells expanded in the laboratory and delivered to the diseased
eye on a cultivated corneal epithelial sheet attached to a
therapeutic bandage lens.51 Both groups conrmed that a small
12 mm2 limbal biopsy provides sufcient amounts of cultured
corneal epithelial cells to restore the entire corneallimbal
surface.45,51
A number of materials have been described to act as carriers
for expansion of limbal epithelial stem cells including corneal
stroma, collagen, de-epithelialized amniotic membrane,
therapeutic soft contact lenses, a carrier-free cultivated corneal
epithelial sheet or collagen gel, brin gel, and cross-linked gels
of bronectin and brin.11,45,48,49,5159 Several studies have
demonstrated success with improved ocular surfaces following
ex vivo expansion of limbal stem cells using these variety of
carriers. One study comparing different carrier substrates found
that de-epithelialized amniotic membrane was the best carrier
when compared to corneal stroma, collagen, collagen shields,
and soft contact lenses.49 Tsai was perhaps the rst to suggest
using amniotic membrane as a carrier for cultured limbal stem
cells.60 Once the stem cells are expanded in the laboratory
following a small limbal biopsy from a donor eye, the cells are
allowed to adhere to the amniotic membrane for several weeks,
after which the entire complex is transplanted to the diseased
ocular surface. Several groups have shown success with
amniotic membrane carriers indicating these carriers convey
durability and manipulability, while supplying a compatible
extracellular matrix for expanded stem cells with grafting
procedures (Fig. 65.13).11,45,48,49,5257 Schwab and colleagues
found success with this technique demonstrating ocular surface
improvement in 10 of 14 patients with mean follow-up of
13 months following a mixture of cultured autologous and allogeneic corneal epithelial stem cells on an amniotic membrane
carrier.48 Daya et al found improved ocular surfaces in seven of
10 eyes with a mean follow-up at 28 months.56 They cultivated
stem cells harvested from a corneoscleral rim and transplanted
the cultured cell sheet to the diseased recipient eye, followed by
coverage with amniotic membrane.56 Other groups have
reported additional benets of cultivating autologous oral
mucosal epithelial says on an amniotic membrane carrier.6164
The proposed advantage of cultivated oral mucosal epithelial cells
involves a more rapid epithelialization of the damaged ocular
surface just after transplantation of the expanded epithelial
cells. The process normally takes 714 days to occur following
transplantation, but oral mucosal epithelial cells promote rapid
re-epithelialization of the damaged ocular surface shortly after
transplantation.61 Rapid epithelialization reduces epitheliumrelated inflammation and epithelial defects thus lowering the
FIGURE 65.13. A photograph demonstrating an eye 1 week following

a simultaneous penetrating keratoplasty and ex vivo expanded stem
cells using an amniotic membrane carrier to correct severe stem cell
deciency and complete corneal opacication.
risk of infection and other ocular surface complications. Another

proposed advantage to cultivated oral mucosal epithelial cells
on amniotic membrane is the reduced risk of corneal scarring and
lipid deposition after transplantation.61 Nishida et al reported
on restored ocular surfaces in four patients with bilateral severe
stem cell deciency (three with mucous membrane pemphigoid
and one with StevensJohnson syndrome). The technique
required harvesting of autologous buccal mucosal epithelium
with expansion of the cells in the laboratory for 2 weeks as
suspended cells. An autologous oral epithelial-cell sheet was
produced and this matrix was transplanted over the diseased
ocular surface and allowed to heal without sutures.63
While a number of techniques and carriers have been
described for ex vivo stem cell expansion, we prefer the technique previously outlined by Schwab and colleagues.4749 A 2 mm2
conjunctival biopsy of the donor superotemporal conjunctiva is
obtained including limbal conjunctiva for harvesting of stem
cells.4749 The cells are transferred to a laboratory in cellular
transport medium and cultivated on modied amniotic membrane as described in previous reports.4749 This complex is
placed within enriched medium and the stem cells are allowed
to attach to the amniotic membrane over the next 1014 days.
The medium is changed every 2 days during the attachment
phase, and the graft is ready for transplantation after the allotted
time.4749 Surgical technique for transplanting the complex containing stem cells includes a 360 conjunctival peritomy and
a 24 mm conjunctival resection followed by removal of the
diseased corneal pannus and epithelium. The edges of the
amniotic membrane carrier are then sutured to the conjunctival
edges with 100 nylon suture and a bandage lens is placed to
allow for adhesion of the stem cells to the underlying tissue. A
bandaged soft contact lens or tissue adhesive can also be used to
promote adherence of the amniotic membrane complex to the
ocular surface. The lens is left in place for 23 months with
careful observation. All allogeneic ex vivo expansions require
systemic and topical immunosuppression as with other allogeneic limbal grafting procedures. Autologous ex vivo expanded
stem cell transplantation cases do not require oral immunosuppression, and are maintained on topical immunosuppression for an indenite amount of time.
SUMMARY
With the expanded knowledge of ocular surface disease and
stem cell anatomy and function, our ability to diagnose and

successfully treat severe ocular surface diseases has signicantly
improved in a relatively short time. Surgeries such as CLAU,
LR-CLAL, KLAL, and C-KLAL have provided new ways to treat
previously untreated ocular disorders. With increasing knowledge of stem cell principles and further evolution of limbal
stem cell transplant procedures, our capabilities to eradicate
damaged ocular surfaces and restore normal ocular architecture
will continue to improve. This expansive area of research has
now provided a basis for bioengineered stem cell tissue for
transplantation to damaged ocular surfaces. The reality of such

procedures is upon us with a number of facilities converting
previous theory and imaginative dreams into reality. Techniques
will continue to improve and evolve as continued research and
knowledge flourishes. Ex vivo stem cell expansion, gene
therapy, and a potential link to hematopoietic stem cells may
ultimately provide the cure for a class of diseases that many
eye care professionals previously found extremely frustrating
and challenging.
1. Lajtha LG: Stem cell concepts.

Differentiation 1979; 14:2334.
2. Leblond CP: The life history of cells in
renewing systems. Am J Anat 1981;
160:114158.
3. Potten CS: Epithelial proliferative
subpopulations. In: Lord BI, Potten CS,
Cole RJ, eds. Stem cells and tissue
homeostasis. Cambridge, UK: Cambridge
University Press; 1978.
4. Pster RR: Corneal stem cell disease:
concepts, categorization, and treatment by
auto- and homotransplantations of limbal
stem cells. CLAO 1994; 20:6472.
5. Tseng SCG: Concept and application of
6. Kinoshita S, Adachi W, Sotozono C, et al:
Characteristics of the human ocular surface
epithelium. Prog Ret Eye Res 2001;
20:639673.
7. Dua HS, Azuara-Blanco A: Limbal stem
cells of the corneal epithelium. Surv
Ophthalmol 2000; 44:415425.
8. Dua HS, Saini JS, Azuara-Blanco A,
Gupta P: Limbal stem cell deciency:
concept, aetiology, clinical presentation,
diagnosis, and management. Indian J
9. Potten CS, Loeffler M: Stem cells:
attributes, cycles, spirals, pitfalls and
uncertainties. Lessons for and from
the crypt. Development 1990;
110:10011020.
10. Watt FM, Hogan BL: Out of Eden: stem
cells and their niches. Science 2000;
287:14271430.
11. Grueterich M, Espana EM, Tseng SCG:
Ex vivo expansion of limbal epithelial stem
cells: amniotic membrane serving as a
stem cell niche. Surv Ophthalmol 2003;
48:631646.
12. Friedenwald JS: Growth pressure and
metaplasia of conjunctival and corneal
epithelium. Doc Ophthalmol 1951;
5:184192.
periocorneal papillary structure in renewal
of corneal epithelium. Nature 1971;
229:560561.
14. Goldberg MF, Bron AJ: Limbal palisades of
Vogt. Trans Am Ophthalmol Soc 1982;
80:155171.
Differentiation-related expression of major
suggests limbal location of corneal
103:4962.
57:201209.
17. Kasper M, Moll R, Stosiec P, Karsten U:

Patterns of cytokeratin and vimentin
expression in the human cycle.
Histochemistry 1988; 89:369373.
18. Pellegrini G, Golisano O, Paterna P, et al:
Location and clonal analysis of stem cells
and their differentiated progeny in the
human ocular surface. J Cell Biol 1999;
145:769782.
19. Pellegrini G, Dellambra E, Golisano O, et al:
p63 identies keratinocyte stem cells. Proc
Natl Acad Sci USA 2001; 98:31563161.
20. Salehi-Had H, Alvarenga LS, Isseroff R,
Schwab IR: Factors modulating p63
expression in cultured limbal epithelial
cells. Cornea 2005; 24:845852.
basal cells than can be preferentially
epithelial stem cells. Cells 1989;
57:201209.
22. Ebato B, Frind J, Thoft RA: Comparison of
central and peripheral human corneal
23. Kiritoshi A, SundarRaj N, Thoft RA:
Differentiation in cultured limbal epithelium
as dened by keratin expression. Invest
24. Lauweryns B, van den Oord JJ, Missotten L:
The transitional zone between limbus and
peripheral cornea. Invest Ophthalmol Vis
Sci 1993; 34:19911999.
25. Zieske JD, Bukusoglu G, Yankauchas MA:
26. Lindberg K, Brown ME, Chaves HV, et al:
In vitro preparation of human ocular
surface epithelial cells for transplantation.
34:26722679.
27. Daniels JT, Dart JK, Tuft SJ, et al: Corneal
stem cells in review. Wound Repair Regen
2001; 9:483494.
28. Kaur P, Li A, Redvers R, et al: Keratinocyte
stem cell assays: an evolving science.
J Investig Dermatol Symp Proc 2004;
9:238247.
29. Lehrer MS, Sun T-T, Lavker RM: Strategies
of epithelial repair: modulation of stem cell
and transient amplifying cell proliferation.
J Cell Sci 1998; 111:28672875.
of corneal epithelial maintenance (letter).
31. Dua HS, Gomes JA, Singh A: Corneal
epithelial wound healing. Br J Ophthalmol
1994; 78:401408.
32. Hartman DC: Use of free grafts in
correction of recurrent pterygia,
pseudopterygia and symblepharon.
California Med 1951; 75:279280.
33. Tananuvat N, Martin T: The results of

primary pterygium compared with
conjunctival autograft. Cornea 2004;
23:258263.
34. Hoffman RS, Power WJ: Current options in
pterygium management. Int Ophthalmol
Clin 1999; 39:1526.
35. Hirst LW: The treatment of pterygium. Surv
Ophthalmol 2003; 48:145180.
36. Holland EJ, Schwartz GS: The paton
lecture: ocular surface transplantation:
10 years experience. Cornea 2004;
23:425431.
37. Thoft RA: Conjunctival transplantation.
38. Thoft RA: Keratoepithelioplasty. Am J
39. Kenyon KR, Tseng SCG: Limbal autograft
40. Tsai RJF, Tseng SCG: Human allograft
limbal transplantation for corneal surface
reconstruction. Cornea 1994; 13:389400.
41. Tsubota K, Toda I, Saito H, et al:
Reconstruction of the corneal epithelium by
limbal allograft transplantation for severe
ocular surface disorders. Ophthalmology
1995; 102:14861495.
42. Holland EJ, Schwartz GS: The evolution
and classication of ocular surface
transplantation. In: Holland EJ, Mannis MJ,
eds. Ocular surface disease: medical
and surgical management. New York:
Springer; 2002.
43. Schwartz GS, Holland EJ: Classication
and staging of ocular surface disease. In:
Krachmer JH, Mannis MJ, Holland EJ, eds.
Cornea. 2nd edn. Philadelphia, PA: Elsevier
Mosby; 2005.
44. Mannis MJ, McCarthy M, Izquierdo L Jr:
Technique for harvesting keratolimbal
allografts from corneoscleral buttons. Am J
45. Tor H, Schwab IR, Isseroff R:
Transplantation of cultured autologous
limbal stem cells for ocular surface disease
(abstract). In Vitro 1996; 32:47A.
46. Rheinwald JG, Green H: Serial cultivation
of strains of human epidermal keratinocytes:
the formation of keratinizing colonies from
single cells. Cell 1975; 6:331343.
47. Schwab IR, Isseroff RR: Ex vivo stem cell
expansion. In: Holland EJ, Mannis MJ, eds.
Ocular surface disease: medical and surgical
management. New York: Springer; 2002.
48. Schwab IR, Reyes M, Isseroff RR:
Successful transplantation of
bioengineered tissue replacements in
patients with ocular surface disease.
Cornea 2000; 19:421426.
49. Schwab IR: Cultured corneal epithelia for
ocular surface disease. Trans Am
CHAPTER 65
REFERENCES
879
SECTION 6
50. Friend J, Kinoshita S, Thoft RA, et al:

Corneal epithelial cell cultures on stroma
carriers. Invest Ophthalmol Vis Sci 1982;
23:4149.
51. Pellegrini G, Traverso CE, Franzi AT, et al:
Long-term restoration of damaged corneal
surfaces with autologous cultivated
corneal epithelium. Lancet 1997;
349:990993.
52. Tsai RJ, Li LM, Chen JK: Reconstruction of
damaged corneas by transplantation of
autologous limbal epithelial cells. N Engl J
Med 2000; 342:8693.
53. Koizumi N, Inatomi T, Suzuki T, et al:
Cultivated corneal epithelial stem cell
transplantation in ocular surface disorders.
Ophthalmology 2001; 108:15691574.
54. Shimazaki J, Aiba M, Goto E, et al:
Transplantation of human limbal epithelium
cultivated on amniotic membrane for the
treatment of severe ocular surface
disorders. Ophthalmology 2002;
109:12851290.
55. Meller D, Pires RTF, Tseng SCG: Ex vivo
preservation and expansion of human
limbal epithelial stem cells on amniotic
880
56.
57.
58.
59.
60.
membrane cultures. Br J Ophthalmol 2002;

86:463471.
Daya SM, Watson A, Sharpe JR, et al:
Outcomes and DNA analysis of ex vivo
expanded stem cell allograft for ocular
surface reconstruction. Ophthalmology
2005; 112:470477.
Sangwan VS, Matalia HP, Vemuganti GK,
et al: Early results of penetrating
keratoplasty after cultivated limbal
epithelium transplantation. Arch
Ophthalmol 2005; 123:334340.
Rama P, Bonini S, Lambiase A, et al:
Autologous brin-cultured limbal stem cells
permanently restore the corneal surface of
patients with total limbal stem cell
deciency. Transplantation 2001;
72:14781485.
Han B, Schwab IR, Madsen TK, Isseroff RR:
A brin-based bioengineered ocular surface
with human corneal epithelial stem cells.
Cornea 2002; 21:505510.
Tsai RJF: Corneal surfaces reconstruction
by amniotic membrane with cultivated
autologous limbo-corneal epithelium. Invest
61. Hayashida Y, Nishida K, Yamato M, et al:

Ocular surface reconstruction using
autologous rabbit oral mucosal epithelial
sheets fabricated ex vivo on a temperatureresponsive culture surface. Invet
62. Nakamura T, Kinoshita S: Ocular surface
reconstruction using cultivated mucosal
epithelial stem cells. Cornea 2003;
22:S75S80.
63. Nishida K, Yamato M, Hayashida Y, et al:
Corneal reconstruction with tissueengineered cell sheets composed of
autologous oral mucosal epithelium. New
Eng J Med 2004; 351:11871196.
64. Inatomi T, Nakamura T, Koizumi N, et al:
Current concepts and challenges in ocular
surface reconstruction using cultivated
mucosal epithelial transplantation. Cornea
2005; 24:S32S38.
CHAPTER
66
Amniotic Membrane Surgery

Scheffer C. G. Tseng, Wei Li, Yukihiro Matsumoto, Yasutaka Hayashida,
and Victoria Casas
Overview
Transplantation of cryopreserved amniotic membrane (AM) is an
accepted standard surgical procedure to promote epithelial
healing and to reduce inflammation, scarring, and unwanted
blood vessels on the ocular surface. It can be used as a
temporary graft to suppress unwanted inflammation that may
threaten the wound healing and lead to chronic scarring in a
number of ocular surface diseases. It can also be used as a
permanent graft to replace the corneal and conjunctival epithelial
basement membrane and stroma that are damaged by diseases
or removed by surgeries. This chapter focuses on how AM
surgeries are presently performed with or without sutures and
summarize indications and contraindications of the procedure. It
is envisioned that other potential applications of AMT can be
expanded if action mechanisms can be fully explored in the
future.
PROPRIETARY DISCLOSURE SCGT and his family are more

than 5% shareholders of TissueTech, Inc, which owns US
Patents on the method of preparation and clinical uses of
cryopreserved human AM registered under the name of
AmnioGraft and ProKera, which are currently distributed by
Bio-Tissue, Inc, an afliated entity of TissueTech, Inc. Other
authors do not have any proprietary interest.
INTRODUCTION
Transplantation of cryopreserved amniotic membrane (AM) is
an accepted standard surgical procedure to promote epithelial
healing and to reduce inflammation, scarring, and unwanted
blood vessels on the ocular surface. It can be used as a temporary
graft to suppress unwanted inflammation that may threaten the
wound healing and lead to chronic scarring in a number of
ocular surface diseases. It can also be used as a permanent graft
to replace the corneal and conjunctival epithelial basement
membrane and stroma that are damaged by diseases or removed
by surgeries. In this chapter, we will focus on how AM surgeries
are presently being performed with or without the use of sutures
and then will summarize indications and complications of the
procedure. For more detailed information concerning clinical
studies and case illustration of each indication, the readers are
encouraged to consult with each individual publication cited
herein as well as in the following reviews.17
Histologically, the AM, or amnion, is the innermost layer of
the placental/fetal membrane, and consists of a simple epithelium, a prominent basement membrane and an avascular
stroma (c. average 75 mm). Historically, AM prepared by
different methods had been used mostly as a dressing in
several surgical subspecialties, including ophthalmology,
starting from the early twentieth century.5 The popularity of

transplanting AM as a graft for ocular surface reconstruction
has escalated since the report by Kim and Tseng in 1995.8
Nearly all (more than 150) clinical studies published thus far
are based on cryopreserved AM, and collectively show that
amniotic membrane transplantation (AMT) using cryopreserved AM is effective in facilitating epithelial wound healing
and in reducing stromal inflammation, scarring, and unwanted
new blood vessel formation.
ACTION MECHANISMS
Summary Box
AM Action Mechanisms
Provides a new basement membrane
Provides a new stroma that exerts
Antiinflammatory action
Antiscarring action
Antiangiogenic action
There are a number of action mechanisms that have been put

forth to explain how cryopreserved AM exerts these clinical
efcacies in ocular surface reconstruction. These actions have
recently been reviewed.9 Compositionally, AMs basement
membrane contains type IV collagen, laminin 1, laminin 5, and
collagen VII. The basement membrane side of amniotic tissue
is an ideal substrate for supporting adhesion, growth, and
differentiation of epithelial basal progenitor cells and may
facilitate migration of epithelial cells, and prevent epithelial
apoptosis. The AMs stroma contains unique extracellular matrix
components such as bronectin, type I, type III, and type V
collagens and such glycosaminoglycans/proteoglycans as hyaluronic acid, decorin, and biglycan, and nonglycosylated lumican,
of which the latter has been found to facilitate epithelial wound
healing,10 while exactly how other components may function
remains to be elucidated.
The AM contains several mitogenic growth factors, several
antiangiogenic and antiinflammatory proteins, and natural
inhibitors to various proteases. Although not innervated, AM
contains several neurotrophins11 and nerve growth factor.12 The
antiangiogenic action of AM may in part be explained by the
expression of pigment epithelium-derived growth factor13 found
in the amniotic basement membrane.
Several experimental studies have demonstrated that AM
indeed exerts potent antiinflammatory actions via suppression
of IL-1a and IL-1b expression,14 facilitation of rapid apoptosis
of polymorphonuclear neutrophils in eximer laser ablation,15,16
in alkali-injured rabbit corneas,17 CD20+ cells in human
881

chemical-burned corneas,18 and macrophages in rat corneas
with herpes simplex virus type 1-induced severe necrotizing
keratitis.19 In culture, human AM stromal matrix facilitates
apoptosis of murine macrophages even if they are activated by
IFN-g.20 Collectively, the above ndings provide evidence in
support of the effect of AM in suppressing both acute and
chronic inflammation caused by a number of insults.
Although the aforementioned antiinflammatory actions may
indirectly contribute to AMs antiscarring actions, several lines
of experimental evidence also support the notion that AM
exerts a direct antiscarring effect on broblasts. In culture, AM
stroma matrix suppresses TGF-b signaling, and, myobroblast
differentiation of human corneal broblasts, limbal broblasts,21
conjunctival broblasts, and pterygium body broblasts.22
Expression of TGF-b1 and deposition of collagen and bronectin are reduced by AM transplanted to the excimer laser
ablated corneal surface.23,24 Human AM transplanted into the
rabbit corneal stromal pocket reduces myobroblast differentiation elicited by invading epithelial cells and in a tissue
culture model of collagen gel contraction.25 The suppressive
effect in TGF-b signaling is not only pathologically important in
preventing scar formation, but also physiologically important in
maintaining the normal keratocyte phenotype. For example,
AM stromal matrix is capable of maintaining the characteristic
dendritic morphology and keratocan expression of human,26,27
murine,28 and monkey29 keratocytes in culture.
Combination of the aforementioned actions also explains
why AM is an ideal substrate to cultivate epithelial progenitor
cells of the conjunctiva, the cornea, the limbus, the oral
mucosa, corneal endothelial cells, and retinal pigment epithelial
cells in vitro. The resultant cultivated cells on AM have been
transplanted to treat human diseases.
CLINICAL USES
SECTION 6
TWO MODES OF AMNIOTIC MEMBRANE

TRANSPLANTATION
Clinical uses of AM for ocular surface reconstruction can be
categorized as a permanent graft (for host cells to grow over or
into the membrane) or as a temporary graft (i.e., dressing,
patch, or bandage for host cells to grow underneath the
membrane). In the former situation, the membrane is used to
ll in the tissue defect of the cornea or the conjunctiva so that
it will be integrated into the host tissue. In the latter situation,
the membrane is applied as if it were a bandage lens to cover
both the healthy host tissue and the site of interest so that
subsequently the epithelial healing is completed underneath
AM. Therefore, the membrane is invariably dissolved or
removed. Clinical uses of AM as either an ex vivo surrogate to
simulate the stem cell niche30 or as a carrier31,32 so as to expand
limbal epithelial stem cells are not covered in this chapter.
SURGICAL INDICATIONS
Tip Files
Two Major Modes of Amniotic Membrane Transplantation
As a temporary graft to suppress inflammation
As a permanent graft to encourage regeneration by restoring
basement membrane and stroma without inflammation and
scarring
As a Temporary Graft
882
Cryopreserved AM can be used as a temporary graft to suppress

inflammation on the ocular surface inflicted by various diseases
TABLE 66.1. Surgical Indications for Temporary AM Grafts

In Human Patients:
Acute chemical/thermal burns
Acute SJS with or without toxic epidermal necrolysis
Chronic recalcitrant keratitis caused by HZO, HSV, or vernal

keratitis
Persistent or recurrent epithelial defect (erosion)
High-risk corneal grafts (to reduce complication)
In conjunction with socket or fornix reconstruction (to prevent

lid/lash rubbing)
In Experimental Animals:
Excimer laser ablation (PRK/PTK) (to prevent haze)
Implantation of keratoprosthesis
and insults. As a result, the epithelial healing is facilitated with

minimal or no scarring. As shown in Table 66.1, clinical
indications include intense ocular surface inflammation and
epithelial erosion caused by acute chemical and thermal
burns,17,3336 and, acute inflammatory and ulcerative stage of
StevensJohnson syndrome (SJS) with or without toxic epidermal necrolysis (TEN).37,38 Chemical burns, especially alkali
burns, result in severe inflammation, which if left untreated,
invariably results in granulation and scarring. Scars on the
corneal surface threaten the vision, scars in the conjunctiva
cause motility restriction, and scars in the lids lead to exposure,
mechanical microtrauma (by misdirected lashes and keratinization), and dryness. Conventional therapies for chemical
burns have a limited success.39 For SJS with or without TEN,
the aforementioned cicatricial pathologies in the lid margin and
the tarsus correlate well with sight-threatening corneal complications.38 The conventional managements at Intensive Care
and Burn Units are directed to life-threatening problems, and
frequently are inadequate to address the ocular inflammation
and ulceration. Thus, patients are frequently left with a
blinding disease because of scarring-induced late complications.
For both of these devastating emergencies, cryopreserved AM
effectively reduces inflammation and facilitates epithelial
wound healing (Figs 66.1 and 66.2). Not only does it rapidly
restore comfort and vision, but also avoids late cicatricial
complications such as symblepharon, mechanical microtrauma
(by misdirected lashes and lid margin keratinization), motility
restriction, exposure, and dryness.
In addition, cryopreserved AM can also be used as a
temporary graft to treat chronic recalcitrant inflammation with
or without persistent epithelial defect caused by several ocular
surface diseases, including neurotrophic keratitis caused by
herpes zoster virus (HZO), herpes simplex virus (HSV),7,19 and
vernal keratoconjunctivitis.40 When epithelial defects are
accompanied by stromal ulceration, a temporary cryopreserved
AM graft is used together with a permanent graft to provide an
additional protection while ensuring epithelialization.41,42
Recently, cryopreserved AM has been advocated to promote
healing and reduce inflammation in high-risk eyes receiving
penetrating or lamellar keratoplasties.7
As stated above, in experimental animal models, cryopreserved AM as a temporary graft is effective in reducing
inflammation and corneal haze induced by excimer laser
ablation,15,16,4345 and to improve implantation of keratoprosthesis (Table 66.1).46 Except for a few patients with
10/0 Nylon sutures

placed on lid margin
Running suture, 8 to
10 episcleral bites,
10/0 Nylon
Amniotic Membrane
b
Doubled armed 4/0 black silk
FIGURE 66.2. Transplantation of cryopreserved

AM as a temporary graft to cover corneal and
conjunctival surfaces. Two large pieces of
cryopreserved AM are attached to the lid
margin by interrupted or running 100 nylon
sutures, reflected into the fornix by a muscle
hook, and secured to the fornix by two doublearmed 40 silk sutures to the skin with a bolster
(a). The rest of AM was overlapped on the
corneal surface and secured by a running 100
nylon in a purse-string fashion (see Fig.
66.1b,c). This technique is applied to treat
acute chemical burns or SJS (b), and usually
results in rapid healing, in this particular case
only in 1 week (c and d).
CHAPTER 66
FIGURE 66.1. Transplantation of cryopreserved AM as a temporary graft to cover the corneal surface. A semitransparent cryopreserved AM is
peeled off from the nitrocellulose paper (a). The membrane is laid down to cover the entire corneal surface with the stromal side down (b) and
can be sutured in a purse-string fashion using 100 nylon running with 68 episcleral bites (c). The suturing method is also schematically shown
in Figure 66.2. In this patient, two conjunctival limbal autografts (marked in blue) have already been transplanted from the fellow eye (b and c).
The membrane can also be introduced to the corneal surface via the insertion of ProKera (d). In this case, the corneal epithelial defect followed
by epithelial debridement heals as judged by fluorescein staining (e).
subepithelial opacities and myopic regression after photorefractive keratectomy (PRK),5,47 clinical efcacy has not been
thoroughly demonstrated in human patients for these indications.
As described in more detail below, cryopreserved AM as a
temporary graft can be delivered without sutures through a
symblepharon comformer termed ProKera. Besides the

aforementioned clinical effects, ProKeras PMMA conformer
ring can be used as a symblepharon ring and together with AM
to reduce conjunctival inflammation/swelling following
reconstruction in the orbit/socket, lids, or the fornix.
883
As a Permanent Graft
Cryopreserved AM can also be used as a permanent graft to
replace the decient or destroyed ocular surface tissue caused by
diseases or surgeries, and to promote regeneration rather than
repair of the ocular surface. As described in Action Mechanisms,
the basement membrane side of AM helps rapid epithelialization of the ocular surface, while the stromal side of AM exerts
antiinflammatory, antiscarring, and antiangiogenic effects to
help the newly reconstructed ocular surface heal with less
inflammation and scarring.
Contrary to conventional corneal transplantation, in which
allogeneic epithelial and mesenchymal cells are transplanted,
cryopreserved AM does not contain live cells, and hence the
healing depends on migrating host cells. Therefore, if the surrounding host tissue still retains healthy epithelial stem cells
and if the surrounding host stroma does not manifest persistent
inflammation, scarring, or ischemia, cryopreserved AM has
successfully treated a number of corneal and conjunctival
diseases listed in Table 66.2 regardless of the size of the tissue
defect to be covered.
For corneal surface reconstruction

Corneal ulcers of different depths, decemetocele and
perforation Corneal ulcers are serious and urgent clinical
problems that can be complicated by microbial infections and
threaten the patients vision. Corneal ulcers can be caused by
various insults, e.g., exogenously from chemical burns, infection, radiation, or surgeries, while endogenously from aging,
diabetes mellitus, viral (herpes) infection, and autoimmune
disorders, with the common denominator being neurotrophic
keratopathy.48 When all medical treatments fail and the ulceration persists (e.g., more than 3 weeks), conventional surgical
TABLE 66.2. Surgical Indications for Permanent AM Grafts
SECTION 6
I. Corneal Diseases
Persistent epithelial defects and ulcers
Descemetocele or perforation
Neurotrophic keratitis
Bullous keratopathy
Band keratopathy
Removal of scar
II. Conjunctival Diseases
Primary and recurrent pterygia
Pingueculae
Tumors
Conjunctivochalasis
Superior limbic keratoconjunctivis
Scars and symblepharon
Chemical burns, SJS, and pemphigoid
Leaking blebs
III. Other Diseases
884
Partial limbal stem cell deciency
Scleral melt/ischemia
Fornix reconstruction
Socket reconstruction
treatments include lamellar or full-thickness corneal transplantation (patch graft), tarsorrhaphy and conjunctival flap.
AMT offers the following advantages: avoidance of potential
allograft rejection and postoperative astigmatism of tectonic
corneal grafts, ease and convenience of use, feasibility in the
event of corneal tissue shortage, preservation of a better
aesthetic appearance.
A number of studies have shown that transplantation of
cryopreserved AM achieves an average success rate of 79%
(103/131 eyes ranging from 67% to 91%) and rapid healing,
usually completed in 14 weeks without recurrence, and
noninflamed ocular surfaces.41,4957 Variable success rates may
be attributed to different underlying etiologies and concomitant
or subsequent treatments in managing neurotrophic keratopathy. For example, AM rapidly dissolves (within 1 week) if
there is severe exposure and dryness caused by neurotrophic
keratopathy (Fig. 66.3). Although successful epithelial and
stromal healing can be achieved in a patient with a corneal ulcer
due to severe Graves ophthalmopathy despite topical medications and orbital decompression surgery,42 corneal surface
breakdown is likely to recur if severe neurotrophic keratopathy
is left untreated. Therefore, it is advised to perform punctal
occlusion before AMT, consider CL wear56 or temporary
tarsorrhaphy at the time of AMT, and add extended high DK
contact lens or permanent tarsorrhaphy after healing by AMT.
For deeper stromal ulcers down to descemetocele, multiple
layers of AM can be used to restore the normal corneal
thickness.50,54,5658 When there is frank perforation even up to
2 mm in diameter, AMT can be used to seal the ulcer56 with or
without additional tissue adhesive59 or brin glue.57,60 Corneal
ulcers caused by bacterial61,62 and fungal63 causes have been
successfully managed by AMT when the infection is controlled.
It should be noted that AMT alone is not sufcient to heal
corneal epithelial defects caused by total limbal stem cell
deciency or severe ischemia, e.g., chemical burns.
Symptomatic bullous keratopathy Bullous keratopathy, i.e.,
corneal edema, is a disorder caused by corneal endothelial
decompensation due to degeneration (Fuchs endothelial dystrophy), surgical trauma, intractable glaucoma, or previous corneal
graft failure. Patients with bullous keratopathy complain of
ocular pain and loss of vision. For those patients with potenital
vision, corneal transplantation is the treatment of choice.
However, for those who do not have a visual potential, relief of
pain and recurrent erosion will rely on several surgical
treatments including cauterization, anterior stromal puncture,
excimer laser photoablation, and conjunctival flap. AMT can
achieve a pain relief with an overall success rate of more than
85%.5,6467 The AM-covered corneal surface heals in 3 weeks
with reduced inflammation and only less than 10% of eyes
show recurrent surface breakdown.
Band keratopathy Band keratopathy occurs in a number of
corneal diseases characterized by chronic inflammation and
sometimes bullous keratopathy (Fig. 66.4). Patients with band
keratopathy complain of ocular irritation and experience
corneal surface erosion and microbial infection. Conventional
treatments include chelation by EDTA and supercial
keratectomy to remove supercial calcium deposit and corneal
stromal tissue. Without treatment, band keratopathy does not
show any remission and instead has a slowly progressive
clinical course. AMT has achieved a success rate of more than
90% in relieving patients pain, establishing a stable corneal
epithelium, and in some eyes improved vision.6870
Partial limbal stem cell deciency Corneal epithelial stem
cells are located exclusively at the limbus.71 Destructive loss of
FIGURE 66.3. Transplantation of cryopreserved AM as a permanent graft to cover the corneal surface. Depending on the depth of stromal loss,
one or multiple layers of cryopreserved AM can be used to restore the integrity. In this eye with HSV-induced necrotizing ulceration on the
grafthost junction with perforation (d and e, marked by an arrow). Multiple strips of AM are laid in the ulcer bed (a) and a larger piece of AM is
then laid down to cover both the ulcer and the healthy surface with the stromal side down (b), and sutured with a running 100 nylon suture to
secure it onto the surrounding healthy tissue (c). The excess membrane is trimmed off from the central cornea and the surrounding limbal region
(not shown). This technique has resulted in restoration of a normal corneal thickness and a noninflamed and healed surface (f).
CHAPTER 66
FIGURE 66.4. Transplantation of cryopreserved AM as a permanent graft to cover the corneal surface via a lamellar pocket. When the corneal
lesion such as band keratopathy (d and e) is removed by supercial keratectomy via a #64 blade, the denuded corneal surface of a bullous
corneal stroma can be restored by cryopreserved AM. Following a supercial trephination (a quarter turn), a lamellar pocket is created 360
towards the limbus by a crescent blade (a), AM is inserted into this pocket and secured by a running 100 nylon suture (b and c). This technique
results in a noninflamed, healed, and stable corneal surface (f).
885

the limbal epithelial stem cells and/or dysfunction of the limbal
stroma will lead to limbal stem cell deciency, characterized by
conjunctivalization of the cornea, i.e., the conjunctival epithelium
migrates to cover the corneal surface, which is accompanied by
vascularization, destruction of the basement membrane, chronic inflammation, and scarring of the cornea.72 Limbal stem cell
deciency can be caused by a number of corneal diseases such
as chemical and thermal burns, SJS, ocular pemphigoid, severe
microbial infections, radiation keratopathy, aniridia, etc.
Patients suffering from limbal stem cell deciency complain of
severe photophobia (light sensitivity) and severe loss of vision.
Without treatment, limbal stem cell deciency is progressively
worsened with time. Conventional corneal transplantation
invariably fails, as no stem cells are transplanted, and is frequently rejected due to corneal vascularization and inflammation. New surgical strategy resorts to transplantation of
autologous or allogeneic limbal epithelial stem cells.73,74 AMT
could restore 40% of rabbit corneas with limbal stem cell
deciency.8,75 This success may owe to the nding that AMT
alone can successfully reconstruct human corneas with partial
limbal stem cell deciency.7679 These results suggest that AM
helps expansion of residual limbal stem cells in these corneas so
as to avoid the need for limbal stem cell transplantation. That
also explains why AMT can be used to restore the donor eye
undergoing removal of the normal limbus and the recipient eyes
receiving conjunctival limbal autograft for unilateral total
limbal stem cell deciency.80
For conjunctival surface reconstruction

When a large conjunctival lesion is surgically removed, the
conjunctival defect is normally healed by the surrounding conjunctiva with granulation and scarring, frequently complicated
with cicatricial complications and aesthetic concerns. To avoid
such potential problems, conjunctival autograft from the same
eye or the fellow eye is used. However, some patients might not
have healthy conjunctival tissue to spare and further removal of
the uninvolved conjunctiva might put the patient at additional
risks. That is why AMT has been used as an alternative
substitute for conjunctival surface reconstruction.
Tip File
SECTION 6
If the surrounding host tissue is normal, i.e., with healthy stem cells
and the stroma is not inflamed or scarred, AMT alone is sufcient
to restore conjunctival surface without the concern of the size of
conjunctival defect created during the removal of the following
lesions.
886
Tumors AMT has been shown to restore the conjunctival

surface with minimal scar and inflammation when such tumors
as papilloma, nevus, melanosis, melanoma, intraepithelial
neoplasia, squamous cell carcinoma, and lymphoma are
removed.69,8185 During excision of conjunctival melanoma,
AMT can be accompanied by cryotherapy and topical chemotherapy.82 Importantly, impression cytology conrms that AMT
restores a normal conjunctival epithelial phenotype with
goblet cells.86
Conjunctivochalasis Conjunctivochalasis, dened as a conjunctival redundancy, is frequently seen in the older age group
as an elevation of the bulbar conjunctiva lying along the lower
lid margin.87 Conjunctivochalasis can cause dry eye at the mild
stage, epiphora at the moderate stage, and exposure problems at
the severe stage. Dry eye caused by conjunctivochalasis is
because aqueous tear spread and clearance are interfered, and
can clinically be differentiated from that caused by conventional
aqueous tear deciency.88 For patients with symptomatic conjunctivochalasis, AMT has been used to relieve these irritative
symptoms and restore a smooth and noninflamed conjunctival
surface when topical lubricants become unsuccessful and
punctal occlusion worsens dryness (Fig. 66.5).89,90
Scleral melt Successful reconstruction of conjunctival surface
and sclera has been achieved with AMT in a patient with
Marfans syndrome91 and in some postinfectious fungal scleritis
cases.63
In conjunction with other procedures to suppress

inflammation and scarring
Tip File
If the surrounding host conjunctival tissue is not normal, i.e., the
stroma is severely or diffusely inflamed or scarred, AMT can
restore the conjunctival surface with such adjunctive therapies as
subconjunctival injection of long-acting steroid or intraoperative
application of mitomycin C.
Pterygium Pterygium is characterized by progressive brovascular proliferation of the stroma and manifests dysfunction of
the adjacent limbal epithelial stem cells. The morphology of
subconjunctival brovascular tissue being translucent or not is
used to judge the aggressiveness of pterygium.92 Following the
removal of pterygium by a bare sclera technique, the denuded
conjunctival surface is treated with adjunctive therapies such as
topical application of mitomycin C or external beta irradiation
if left uncovered or is covered with a conjunctival autograft.93
Nevertheless, conjunctival autograft is not feasible in
patients with more than one pterygium in the eye, with recurrent pterygium after several excisions or following conjunctival
autograft, or with glaucoma where the donor site is reserved for
the prospective ltering procedure. For these clinical situations,
cryopreserved AM may be used as an alternative graft to
substitute conjunctival autograft (Fig. 66.6). For primary pterygium, AMT alone achieves an overall success rate either
better,94 compatible with,95 or worse96,97 than that of conjunctival autografts. However, when subconjunctival brovascular
tissue was more thoroughly removed and subconjunctival
injection of long-acting steroid is administered to the surrounding host conjunctiva, AMT can achieve the same low
recurrence as conjunctival autograft.98
For recurrent pterygium, AMT alone cannot attain the same
efcacy as conjunctival autografts, yielding as high as 1015%
recurrence rates.52,95,96,98 However, with intraoperative application of mitomycin C (0.04%) for 35 min, the recurrence rate
is reduced.99101 Intraoperative application of mitomycin does
not reduce the already low recurrence rates by AMT in primary
pterygium,102 but has been used in conjunction with AMT and
conjunctival autograft to restore ocular motility in multirecurrent pterygia.103,104
Symblepharon and fornix reconstruction in cicatricial
diseases Symblepharon and fornix obliteration are pathogenic
if they cause depletion of tear meniscus, exposure, aqueous tear
deciency, or restriction of ocular motility. AMT alone has been
used with an average success of ~70% in fornix reconstruction
and lysis of symblepharon caused by a number of cicatricial
keratoconjunctivitis.51,52,105,106111 However, eyes showing
partial success or failure are usually by multirecurrent pterygia,
SJS, or ocular cicatricial pemphigoid, suggesting that chronic
and persistent inflammation in the surrounding host conjunctiva continues to threaten transplanted AM. That is why
intraoperative application of mitomycin C (0.04% for 5 min) to
CHAPTER 66
FIGURE 66.5. Transplantation of cryopreserved AM as a permanent graft to cover the conjunctival surface with brin glue following
conjunctivochalasis removal. Following conjunctival peritomy, the redundant conjunctiva is readily recessed to expose the sclera due to the lack
of Tenon tissue. A piece of cryopreserved AM is laid down on the surface of the speculum (a and b). The thrombin solution is applied to the
denuded scleral surface (b) while the brinogen solution is applied to the stromal side of AM (c). A muscle hook is used to smoothen and spread
the brin gel underneath the membrane (d). This technique can help unwrinkle the conjunctiva resulting in a continuous and smooth tear
meniscus as shown by fluroescein staining before (e) and after surgery (f).
FIGURE 66.6. Transplantation of cryopreserved AM as a permanent graft to cover the conjunctival surface following pterygium removal. The
pterygium head is detached from the corneal surface (a), and the head and body are then removed (b). The denuded scleral defect is covered
by a piece of cryopreserved AM (c) and secured by interrupted 100 nylon sutures or 80 Vicryl sutures (d), or preferably by brin glue (also see
Fig. 66.5). This technique can restore a noninflamed conjunctival surface (e and f).
887

the fornix (not the bare sclera) is used as an adjunctive
treatment to reduce chronic conjunctival inflammation and
augment the efcacy of AMT in restoring a deep fornix after
symblepharon lysis.112,113
Uses in glaucoma Although AMT has been used to augment
the success of trabeculectomy in experimental rabbits,114 its
efcacy can also be enhanced by intraoperative application of
mitomycin C.115 That may explain why AMT might promote
the clinical success of trabeculectomy in conjunction with
intraoperative application of mitomycin C when trabeculectomy is performed for glaucoma.116,117 Although an early study
showed that AMT alone was not as efcacious as conjunctival
advancement in repairing glaucoma ltering bleb leaks,118 later
modication of the procedure has been made to achieve a high
success in eyes where conjunctival advancement is not possible
(Budenz, personal communication, February 2006).119
In conjunction with other surgical procedures to

correct ischemia
Conjunctival, scleral, or limbal stromal ischemia caused by
chemical/thermal burns, beta irradiation, or prolonged use of
antimetabolites such as mitomycin C is another major factor
threatening the benet of AMT. Ischemia can be corrected by
tenonplasty120 so that AMT can be used to manage ischemia
caused by chemical burns. AMT was found less efcacious than
conjunctival advancement118 in managing glaucoma leaking
blebs caused by mitomycin C, presumably causing ischemia in
the region. However, others have obtained an opposite
experience.5,119
In conjunction with other surgical procedures to

repopulate epithelial stem cells
SECTION 6
As mentioned before, corneas with total limbal stem cell

deciency cannot be restored by AMT alone and needs to be
resorted to, by transplantation of limbal epithelial stem cells, to
restore the normal corneal surface.
888
With conjunctival limbal autograft For unilateral total limbal

stem cell deciency, transplantation of autologous limbal stem
cells via the surgical procedure of conjunctival limbal autograft
is indicated.121 As stated above, because AMT helps expansion
of residual limbal epithelial stem cells, it can be used in
conjunction with conjunctival limbal autograft for the recipient
eye to augment its restorative efcacy (see example shown in
Fig. 66.1).78,100,122,123 For the same reason, AMT can also help
restore limbal integrity in the donor eye undergoing removal of
conjunctival limbal autograft.80 The overall success of the
procedure reaches more than 90%.
With keratolimbal allograft For bilateral total limbal stem cell
deciency caused by severe and advanced ocular pemphigoid
and SJS, AMT has been used in conjunction with keratolimbal
allograft.76,124 However, the overall success rate declines to 50%
presumably due to allograft rejection.125127 If these limiting
factors have been successfully managed, a normal corneal
epithelial phenotype can be achieved by keratolimbal allograft
and AMT for a total stem cell deciency.128,129
With conjunctival autograft To replace the missing conjunctival epithelial stem cells, conjunctival autograft with or without the limbal epithelium can also be used in conjunction with
AMT to restore conjunctival surfaces which have been severely
depleted by squamous metaplasia. As recurrent pterygium
frequently receives more than one surgery, and there is a great
deal of shortage of normal conjunctival adjacent to the diseased
area, it is theoretically advantageous to add a conjunctival autograft, which will bring in some healthy conjunctival epithelial
stem cells.
SURGICAL PROCEDURES
SOURCE OF CRYOPRESERVED AM
AMT for ocular surface reconstruction based on cryopreserved
AM was approved by Medicare as a standard surgical procedure
(CPT code 65780) for physicians and hospital facilities in
January 2004, and for ambulatory surgical centers in July 2005.
In addition, a supply code (HCPCS Level II V2790 Preserved
Human Amniotic Membrane) is available.
AMT is performed using standard surgical instruments and
microsurgical equipment. In the USA, AmnioGraft distributed
by Bio-Tissue, Inc. (Miami, FL), is the only cryopreserved AM
approved by the United States Food and Drug Administration
(FDA) as a graft for ocular surface reconstruction. Because this
cryopreservation method kills allogeneic amniotic cells in
AmnioGraft,130 it eliminates the need for immunosuppression
while maintaining the integrity of its cytokine-rich extracellular
matrix. The FDA classied cryopreserved AM as a Tissue
product when used as a surgical graft for ocular surface reconstruction.131 In contrast, the FDA ruled that the dry acellular
AM cannot be used as a graft without the premarket
approval.132
AmnioGraft is distributed in a foil package in a frozen state.
After thawing at the room temperature, it can be retrieved
aseptically from the inner clear plastic pouch and the
membrane is attached to one side of nitrocellulose paper. Once
transferred to the operating eld, the membrane can be easily
peeled off from the paper by two forceps grabbing the two
corners while the nurse peels the paper away (Fig. 66.1a). In
general, AmnioGraft is placed with the stromal (sticky) side on
the recipient bed; the side can be discerned by touching it with
the tip of a dry MicroSponge (Alcon Surgical, Fort Worth, TX).
ProKera, also manufactured by Bio-Tissue, Inc., is a class II
medical device and contains a piece of AmnioGraft clipped into
a concave dual ring system which conforms to the cornea like a
contact lens. ProKera was approved by the FDA in December
2003 to be used to deliver a temporary graft onto the corneal
surface without sutures (Fig. 66.1d). ProKera is also shipped in
the same manner and in the same medium in a similar foil
package, and can be retrieved from the package in the same
manner. ProKera can be easily inserted without sutures in the
ofce, and at the bedside of the emergency room, the intensive
care unit, or the burn unit where it may not be amenable to
bring the patient to the operating room due to medical reasons.
Therefore, ProKera facilitates the ease of patient care and
reduces the overall medical cost.
TECHNIQUES
As a Temporary Graft
With sutures
To secure AmnioGraft onto the ocular surface using sutures,
AMT is performed under local or general anesthesia depending
on the complexity of the disease.
To cover the corneal surface as a temporary graft for the
indications shown in Table 66.1, AmnioGraft (2.5 2.0 cm
size) is secured by a 100 nylon suture at 23 mm from the
limbus in a purse-string running fashion for a total of 810
episcleral bites (Fig. 66.1b,c, also schematically shown in Fig.
66.2a). To secure AmnioGraft as a temporary graft over both
corneal and conjunctival surfaces, especially for acute
Without sutures
ProKera is inserted with the aid of a lid speculum (Fig. 66.1d).
The size of ProKera, the inner diameter of either 15 or 16 mm,
is chosen to ensure that the PMMA ring is behind the tarsal
conjunctiva by judging the lid ssure when the eye is maximally
open so as to avoid dislodging or discomfort. While wearing
ProKera or AmnioGraft as a temporary graft, the corneal
epithelialization can be assessed by fluorescein staining133 (Fig.
66.1e) and the intraocular pressure can be monitored by
Tonopen134 without removing ProKera or AmnioGraft. Upon
complete healing, e.g., in 12 weeks, ProKera or AmnioGraft
can be easily removed from the ocular surface under a slit-lamp
microscope with forceps.
As a Permanent Graft
With sutures
For the corneal indications listed in Table 66.2, AmnioGraft
(2.5 2.0 cm or 2.0 1.5 cm size) can be used as a single layer
or multiple layers to ll in the stromal defect of an ulcer or
created following supercial keratectomy, depending on the
depth of the stromal loss. The orientation of the bottom layers
does not matter (Fig. 66.3a) while the top layer meant for
epithelialization is best to be placed with the stromal side down
and secured tightly to the corneal surface with 100 nylon
sutures, either interrupted or running (Fig. 66.3b). To ensure
that epithelialization will take place on the top, but not
underneath, the membrane, a lamellar pocket can be prepared
with crescent blade to allow insertion of the membrane (Fig.
66.4a). This technique can be used in several indications when
lesions are removed by supercial keratectomy such as bullous
keratopathy (Fig. 66.4).
For the conjunctival indications listed in Table 66.2,
AmnioGraft (size depending on the area of the defect to be
covered) can be used to substitute the conjunctival tissue.
Following the excision of a large conjunctival lesion, e.g.,
primary pterygium head and body (Fig. 66.6a,b, respectively),
the membrane is placed with the stromal side facing down (Fig.
66.6c). The membrane can be secured by 100 nylon sutures for
perilimbal bulbar regions and by interrupted 80 Vicryl for
forniceal regions. Multiple layers of AmnioGraft can also be
used to ll in the scleral defect (melt) in the same manner as
shown for the corneal stromal defect in Figure 66.3. When used
in conjunction with Tenonplasty, AmnioGraft is placed both
under and above the Tenon tissue. When used in conjunction
with conjunctival autograft, limbal conjunctival autograft or
allograft, or keratolimbal allograft, AmnioGraft is placed below
these grafts.
Without sutures
Topical anesthesia using 0.5% proparacaine hydrochloride, 0.5%
tetracaine hydrochloride, or 2% xylocaine jelly (AstraZeneca,
Wimington, DE) is needed if AMT is performed without
sutures.
Fibrin glue is used to secure AmnioGraft to the corneal or
conjunctival surface without sutures (Figs 66.5 and 66.6d).
Currently, brin glue uses in ophthalmology are considered
off-label, although they have been applied to conjunctival autograft135,136 and AM.57,60 There are two commercially available
brin glues, i.e., Tisseel and CoSeal (Baxter Biologics, Inc.). The
former requires prewarming in a thermal/stirrer provided
without costs by the manufacturer; the latter is ready for use
without warming. Both come with two components, thrombin
and brinogen. After the excision of a large conjunctival lesion,
e.g., conjunctivochalasis, AmnioGraft is laid with the stromal
surface up (Fig. 66.5a). Although the two components can be
delivered simultaneously via a provided dual injection syringe,
it appears to be more convenient and easier to control the time
of polymerization of brin by applying the thinner, clear thrombin
solution on the recipient bed, and the thicker (Fig. 66.5b) but
tawny brinogen solution on the stromal side of AmnioGraft
(Fig. 66.5c). A muscle hook is used to flatten and attach the
membrane onto the ocular surface with ease within 1520 s
(Fig. 66.5d). For fornix reconstruction, the membrane is attached
to the sclera rst before being afxed to the fornix and the
palpebral conjunctiva with brin glue.
LIMITATIONS AND CONTRAINDICATIONS

Tip File
Major Threats for AMT and Solutions
General Threats
(Common to All Surface
Surgeries)
Tear lm deciency and
exposure
Solution
Specic Threats
(Only to AMT)
Severe inflammation and
scarring
Ischemia
Squamous metaplasia
Exposed explant
Solution
Restore ocular surface

defense
Long-acting steroid,
intraoperative MMC
Tenonplasty
Conjunctival autograft
Limbal stem cell
transplantation
Lamellar corneal, scleral,
or pericardial graft
Like other tissue grafts on the ocular surface, the success of

AMT requires restoration of a noninflamed deep fornix and
effective ocular surface defense, e.g., normal lid blinking and
closure and being free of mechanical microtrauma.137 No
wonder, severe aqueous tear deciency, i.e., dry eye, has been
identied as one major limiting factor for the success of AMT
when performed in conjunction with transplantation of
autologous123 or allogeneic138 limbal epithelial stem cells. When
AmnioGraft or ProKera is used as a temporary graft, it may
require more than one application to suppress severe inflammation. Furthermore, if the membrane dissolves within 1 week,
it usually signies that the corneal surface is excessively
exposed, i.e., the tear lm cannot be effectively maintained.
This concern has been addressed by application of bandage
contact lens139 or simply by small temporary tarsorrhaphy.
CHAPTER 66
chemical/thermal burns or acute SJS with or without TEN, two

large pieces of AmnioGraft (3.5 3.5 cm size) are needed. One
piece is laid on the recipient bed with the stromal surface facing
down, and secured to the skin surface of the upper lid margin
by a 100 nylon suture placed in an interrupted or running
manner. AM is then tugged into the upper fornix with a muscle
hook and secured there by passing a double-armed 40 black
silk in a mattress fashion to the skin surface with a bolster
made of either cotton ball or 25 gauge IV tubing (Fig. 66.2). The
remaining AM is spread to cover the upper bulbar conjunctiva
and a part of the upper corneal surface. The other piece is
secured to the lower lid and the lower fornix in the similar
fashion, overlapped with the rst AmnioGraft on the corneal
surface, and secured by a 100 nylon suture placed in the same
manner as shown in Fig. 66.1b,c. A temporary tarsorrhaphy is
added to minimize the lid ssure if there is an exposure concern
due to large scleral show or infrequent blinking as a result of a
neurotrophic state.
889

Because AmnioGraft does not contain live cells, the ultimate
healing is dependent on the surrounding host tissue if used as
a permanent graft. Therefore, the clinical efcacy of AMT is
limited by several factors mentioned above. In such cicatricial
keratoconjunctivitis as current pterygium, ocular cicatricial
pemphigoid, SJS, chemical burns and trachoma, active
inflammation, scarring, or ischemia in the stroma present as
threats. As discussed above, AMT alone may not achieve a
satisfactory success without additional adjunctive measures
such as subconjunctival injection of long-acting steroid,
intraoperative application of mitomycin C, and tenonplasty.
If the surrounding conjunctival epithelial tissue is intrinsically abnormal, e.g., severe squamous metaplasia with frank
keratinization, the healing of AM-covered area may not be
normal unless AMT is performed in conjunction with
conjunctival autograft. If there is total limbal stem cell deciency, AMT alone is not sufcient to heal the persistent
corneal epithelial defect unless combined with transplantation
of limbal epithelial stem cells.
AmnioGraft is resilient to stretch but does not have a strong
rigidity; hence it is not amenable to use alone for covering
exposed synthetic glaucoma drainage implants or retinal explants.
However, AmnioGraft can assist the epithelial healing when a
much stronger tissue such as scleral graft, lamellar corneal

graft, or pericardium is used for these clinical problems. If there
is ischemia, it may be necessary to use a conjunctival autograft.
Although the method of manufacturing cryopreserved AM in
the form of AmnioGraft or ProKera follows Good Tissue
Practices, the tissue is not sterile (i.e., cannot be sterilized).
Careful donor selection and exclusive use of cesarean sectiondelivered placenta reduce the risk of infection.140 Although a
retrospective review of clinical and mircrobiological data veries
the sterility of the manufacturing method for AmnioGraft,141 it
is a good clinical practice to submit the culturing medium and
residual membrane to microbial cultures immediately after
surgery.
ACKNOWLEDGMENT
The development of sutureless ProKera is supported by SBIR Phase I
grant (R43 EY014768, R43EY015592 and R44 EY014768) from National
Institute of Health, National Eye Institute. Other works related to the study
of basic action mechanism of amniotic membrane (AM) were supported in
part by a research grant (RO1 EY06819 and RO1 EY015735) from National
Institute of Health, National Eye Institute and in part by research funding
from TissueTech, Inc., and by a fellowship grant from Ocular Surface
Research & Education Foundation, Miami, FL.
SECTION 6
REFERENCES
890
1. Kruse FE, Voelcker HE, Rohrschneider K:

Transplantation of amniotic membrane for
reconstruction of the eye surface.
2. Dua HS, Azuara-Blanco A: Amniotic
membrane transplantation. Br J Ophthalmol
1999; 83:748752.
3. Sippel KC, Ma JJK, Foster CS: Amniotic
membrane surgery. Curr Opin Ophthalmol
2001; 12:269281.
4. Tseng SCG: Amniotic membrane
reconstruction. Bioscience Rep 2002;
21:481489.
5. Dua HS, Gomes JA, King AJ, Maharajan VS:
The amniotic membrane in ophthalmology.
6. Bouchard CS, John T: Amniotic membrane
transplantation in the management of
severe ocular surface disease: indications
and outcomes. Ocul Surf 2004; 2:201211.
7. Kenyon KR: Amniotic membrane: mothers
own remedy for ocular surface disease.
Cornea 2005; 24:639642.
8. Kim JC, Tseng SCG: Transplantation of
preserved human amniotic membrane for
surface reconstruction in severely damaged
rabbit corneas. Cornea 1995; 14:473484.
9. Tseng SCG, Espana EM, Kawakita T, et al:
How does amniotic membrane work? Ocul
Surf 2004; 2:177187.
10. Yeh LK, Chen WL, Li W, et al: Soluble
lumican glycoprotein puried from human
amniotic membrane promotes corneal
epithelial wound healing. Invest Ophthalmol
Vis Sci 2005; 46:479486.
11. Uchida S, Inanaga Y, Kobayashi M, et al:
Neurotrophic function of conditioned
medium from human amniotic epithelial
cells. J Neurosci Res 2000; 62:585590.
12. Touhami A, Grueterich M, Tseng SC: The
role of NGF signaling in human limbal
epithelium expanded by amniotic
membrane culture. Invest Ophthalmol Vis
Sci 2002; 43:987994.
13. Shao C, Sima J, Zhang SX, et al:
Suppression of corneal neovascularization
by PEDF release from human amniotic
14.
15.
16.
17.
18.
19.
20.
21.
22.
membranes. Invest Ophthalmol Vis Sci

2004; 45:17581762.
Solomon A, Rosenblatt M, Monroy DC,
et al: Suppression of Interleukin-1a and
Interleukin-1b in the human corneal
epithelial cells cultured on the amniotic
membrane matrix. Br J Ophthalmol 2001;
85:444449.
Wang MX, Gray TB, Parks WC, et al:
Corneal haze and apoptosis is reduced by
amniotic membrane matrix in excimer laser
photoablation in rabbits. J Cat Refract Surg
2001; 27:310319.
Park WC, Tseng SCG: Modulation of acute
inflammation and keratocyte death by
suturing, blood and amniotic membrane in
PRK. Invest Ophthalmol Vis Sci 2000;
41:29062914.
Kim JS, Kim JC, Na BK, et al: Amniotic
membrane patching promotes healing and
inhibits protease activity on wound healing
following acute corneal alkali burns. Exp
Eye Res 1998; 70:329337.
Shimmura S, Shimazaki J, Ohashi Y,
Tsubota K: Antiinflammatory effects of
amniotic membrane transplantation in
ocular surface disorders. Cornea 2001;
20:408413.
Heiligenhaus A, Meller D, Meller D, et al:
Improvement of HSV-1 necrotizing keratitis
with amniotic membrane transplantation.
42:19691974.
Li W, He H, Kawakita T, et al: Amniotic
membrane induces apoptosis of interferongamma activited macrophages in vitro.
Exp Eye Res 2006; 82:28292.
Tseng SCG, Li D-Q, Ma X: Suppression of
transforming growth factor isoforms,
TGF-b receptor II, and myobroblast
differentiation in cultured human corneal
and limbal broblasts by amniotic
membrane matrix. J Cell Physiol 1999;
179:325335.
Lee S-B, Li D-Q, Tan DTH, et al:
Suppression of TGF-b signaling in both
normal conjunctival broblasts and
pterygial body broblasts by amniotic
23.
24.
25.
26.
27.
28.
29.
30.
31.
membrane. Curr Eye Res 2000;

20:325334.
Zhong Y, Zhou Y, Wang K: The preliminary
study of amniotic membrane on TGF-beta
1 expression after photorefractive
keratectomy. Yan Ke Xue Bao 1999;
15:215217, 235.
Zhong Y, Zhai Z, Zhou Y, et al: Effect of
amniotic membrane on expressions of
TGF-beta 1, collagens I, III and bronectin
in rabbit corneal healing after
photorefractive keratectomy. Yan Ke Xue
Bao 2000; 16:239242, 258.
Choi TH, Tseng SCG: In vivo and in vitro
demonstration of epithelial cell-induced
myobroblast differentiation of keratocytes
and an inhibitory effect by amniotic
membrane. Cornea 2001; 20:197204.
Espana EM, He H, Kawakita T, et al:
Human keratocytes cultured on amniotic
membrane stroma preserve morphology
and express keratocan. Invest Ophthalmol
Vis Sci 2003; 44:51365141.
Espana EM, Kawakita T, Liu CY, Tseng SCG:
CD-34 expression by cultured human
keratocytes is downregulated during
myobroblast differentiation induced by
TGF-beta1. Invest Ophthalmol Vis Sci
2004; 45:29852991.
Kawakita T, Espana EM, He H, et al:
Keratocan expression of murine
keratocytes is maintained on amniotic
membrane by downregulating TGF-beta
signaling. J Biol Chem 2005;
280:2708527092.
Kawakita T, Espana EM, He H, et al:
Preservation and expansion of primate
keratocyte phenotype by downregulating
TGF-B signaling in a low calcium serumfree medium. Invest Ophthalmol Vis Sci
2006; 47:19181927. Erratum in: Invest
ophthalmol Vis Sci 2006; 47:2279.
Tsai RJF, Li L-M, Chen J-K: Reconstruction
of damaged corneas by transplantation of
autologous limbal epithelial cells. N Eng J
Med 2000; 343:8693.
Schwab IR, Reyes M, Isseroff RR:
Successful transplantation of
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
bioengineered tissue replacements in

patients with ocular surface disease.
Cornea 2000; 19:421426.
Koizumi N, Inatomi T, Suzuki T, et al:
Cultivated corneal epithelial stem cell
transplantation in ocular surface disorders.
Ophthalmology 2001; 108:15691574.
Meller D, Pires RTF, Mack RJS, et al:
acute chemical or thermal burns.
Sridhar MS, Bansal AK, Sangwan VS,
Rao GN: Amniotic membrane transplantation
in acute chemical and thermal injury. Am J
Ophthalmol 2000; 130:134137.
Kobayashi A, Shirao Y, Yoshita T, et al:
Temporary amniotic membrane patching for
acute chemical burns. Eye 2003;
17:149158.
Pan DP, Li XX, Xu JF: Therapeutic effect of
ocular burn. Zhongguo Xiu Fu Chong Jian
Wai Ke Za Zhi 2003; 17:318320.
John T, Foulks GN, John ME, et al: Amniotic
membrane in the surgical management of
acute toxic epidermal necrolysis.
Di Pascuale MA, Espana EM, Liu DT, et al:
Correlation of corneal complications with
eyelid cicatricial pathologies in patients
with StevensJohnson syndrome and toxic
epidermal necrolysis syndrome.
Kim JS, Kim JC, Na BK, Jeong JM, Song
CY: Amniotic membrane patching promotes
healing and inhibits protease activity on
wound healing following acute corneal
alkali burns. Exp Eye Res 2000;
70:329333
Sridhar MS, Sangwan VS, Bansal AK,
Rao GN: Amniotic membrane
transplantation in the management of
shield ulcers of vernal keratoconjunctivitis.
Ophthalmology 2001; 108:12181222.
Solomon A, Meller D, Prabhasawat P, et al:
109:694703.
Heinz C, Eckstein A, Steuhl KP, Meller D:
reconstruction of corneal ulcer in graves
ophthalmopathy. Cornea 2004; 23:524526.
Woo HM, Kim MS, Kweon OK, et al: Effects
of amniotic membrane on epithelial wound
healing and stromal remodeling after
excimer laser keratectomy in rabbit cornea.
Lee JH, Lee HK, Kim JK, et al: Expression
of laminin-5 with amniotic membrane
transplantation in excimer laser ablated rat
corneas. Cabaract Refract Surg 2004;
30:21922199.
Choi YS, Kim JY, Wee WR, Lee JH: Effect
of the application of human amniotic
membrane on rabbit corneal wound healing
after excimer laser photorefractive
Stoiber J, Fernandez J, Kamiski S, et al:
Biological response to a supradescemetic
synthetic cornea in rabbits. Arch
Ophthalmol 2004; 122:18501855.
Lee HK, Kim JK, Kim EK, et al:
Phototherapeutic keratectomy with
amniotic membrane for severe subepithelial
brosis following excimer laser refractive
surgery. J Cataract Refract Surg 2003;
29:14301435.
48. Solomon A, Touhami A, Sandoval H, Tseng

SCG: Neurotrophic keratopathy: basic
concepts and therapeutic strategies. Comp
Ophthalmol Update 2000; 3:165174.
49. Lee S-H, Tseng SCG: Amniotic membrane
transplantation for persistent epithelial
defects with ulceration. Am J Ophthalmol
1997; 123:303312.
50. Kruse FE, Rohrschneider K, Vlcker HE:
106:15041511.
51. Azuara-Blanco A, Pillai CT, Dua HS:
ocular surface reconstruction. Br J
Ophthalmol 1999; 8339:399402.
52. Gabric N, Mravicic I, Dekaris I, et al:
Human amniotic membrane in the
reconstruction of the ocular surface.
Doc Ophthalmol 1999; 98:273283.
53. Chen H-J, Pires RTF, Tseng SCG: Amniotic
neurotrophic corneal ulcers. Br J
Ophthalmol 2000; 84:826833.
54. Hanada K, Shimazaki J, Shimmura S,
Tsubota K: Multilayered amniotic
ulceration of the cornea and sclera. Am J
Ophthalmol 2001; 131:324331.
55. Letko E, Stechschulte SU, Kenyon KR,
et al: Amniotic membrane inlay and overlay
grafting for corneal epithelial defects and
stromal ulcers. Arch Ophthalmol 2001;
119:659663.
56. Rodriguez-Ares MT, Tourino R, LopezValladares MJ, Gude F: Multilayer amniotic
membrane transplantation in the treatment
of corneal perforations. Cornea 2004;
23:577583.
57. Hick S, Demers PE, Brunette I, et al:
ulcers and perforations: a review of
33 cases. Cornea 2005; 24:369377.
58. Prabhasawat P, Tesavibul N,
Komolsuradej W: Single and multilayer
persistent corneal epithelial defect with and
without stromal thinning and perforation.
Br J Ophthalmol 2001; 85:14551463.
59. Su CY, Lin CP: Combined use of an
amniotic membrane and tissue adhesive in
treating corneal perforation: a case report.
60. Duchesne B, Tahi H, Galand A: Use of
human brin glue and amniotic membrane
transplant in corneal perforation. Cornea
2001; 20:230232.
61. Kim JS, Kim JC, Hahn TW, Park WC:
Amniotic membrane transplantation in
infectious corneal ulcer. Cornea 2001;
20:720726.
62. Chen JH, Ma DH, Tsai RJ: Amniotic
membrane transplantation for pseudomonal
keratitis with impending perforation. Chang
Gung Med J 2002; 25:144152.
63. Ma DH-K, Wang SF, Su WY, Tsai RJF:
Amniotic membrane graft for the
management of scleral melting and corneal
perforation in recalcitrant infectious scleral
and corneoscleral ulcers. Cornea 2002;
21:275283.
64. Pires RTF, Tseng SCG, Prabhasawat P, et al:
symptomatic bullous keratopathy. Arch
Ophthalmol 1999; 117:12911297.
65. Mrukwa-Kominek E, Gierek-Ciaciura S,
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
Rokita-Wala I, Szymkowiak M: Use of

treating bullous keratopathy. Klin Oczna
2002; 104:4146.
Mejia LF, Santamaria JP, Acosta C:
Symptomatic management of
postoperative bullous keratopathy with
nonpreserved human amniotic membrane.
Cornea 2002; 21:342345.
Espana EM, Grueterich M, Sandoval H, et al:
bullous keratopathy in eyes with poor
visual potential. J Cat Refract Surg 2003;
29:279284.
Anderson DF, Prabhasawat P, Alfonso E,
Tseng SCG: Amniotic membrane
transplantation after the primary surgical
management of band keratopathy. Cornea
2001; 20:354361.
Prabhasawat P, Kosrirukvongs P,
Booranapong W, Vajaradul Y: Amniotic
membrane transplantation for ocular
surface reconstruction. J Med Assoc Thai
2001; 84:705718.
Tosi GM, Traversi C, Schuerfeld K, et al:
Amniotic membrane graft: histopathological
ndings in ve cases. J Cell Physiol 2005;
202:852857.
Lavker RM, Sun T-T: Epidermal stem cells:
properties, markers and location. Proc Natl
Acad Sci USA 2000; 97:1347313475.
Puangsricharern V, Tseng SCG: Cytologic
evidence of corneal diseases with limbal
stem cell deciency. Ophthalmology 1995;
102:14761485.
Tseng SCG: Regulation and clinical
implications of corneal epithelial stem cells.
Mol Biol Rep 1996; 23:4758.
Holland EJ, Schwartz GS: Changing
concepts in the management of severe
ocular surface disease over twenty-ve
years. Cornea 2000; 19:688698.
Kim JC, Tseng SCG: The effects on
inhibition of corneal neovascularization
after human amniotic membrane
transplantation in severely damaged rabbit
corneas. Korean J Ophthalmol 1995;
9:3246.
Tseng SCG, Prabhasawat P, Barton K, et al:
Amniotic membrane transplantation with or
without limbal allografts for corneal surface
reconstruction in patients with limbal stem
cell deciency. Arch Ophthalmol 1998;
116:431441.
Anderson DF, Ellies P, Pires RT, Tseng SC:
partial limbal stem cell deciency. Br J
Ophthalmol 2001; 85:567575.
Gomes JA, dos Santos MS, Cunha MC,
for partial and total limbal stem cell
deciency secondary to chemical burn.
Sangwan VS, Matalia HP, Vemuganti GK,
Rao GN: Amniotic membrane
transplantation for reconstruction of
corneal epithelial surface in cases of partial
limbal stem cell deciency. Indian J
Ophthalmol 2004; 52:281285.
Meallet MA, Espana EM, Grueterich M, et al:
Amniotic membrane transplantation with
conjunctival limbal autograft for total limbal
stem cell deciency. Ophthalmology 2003;
110:15851592.
Tseng SCG, Prabhasawat P, Lee S-H:
conjunctival surface reconstruction. Am J
Ophthalmol 1997; 124:765774.
CHAPTER 66
891
SECTION 6
892
82. Shields CL, Shields JA, Armstrong T:

Management of conjunctival and corneal
melanoma with surgical excision, amniotic
membrane allograft, and topical
chemotherapy. Am J Ophthalmol 2001;
132:576578.
83. Paridaens D, Beekhuis H, van Den Bosch W,
et al: Amniotic membrane transplantation in
the management of conjunctival malignant
melanoma and primary acquired melanosis
with atypia. Br J Ophthalmol 2001;
85:658661.
84. Chen Z, Yan J, Yang H, et al: Amniotic
membrane transplantation for conjunctival
tumor. Yan Ke Xue Bao 2003; 19:145,
165167.
85. Dalla PG, Ghirlando A, Busato F, Midena E:
Reconstruction of conjunctiva with
amniotic membrane after excision of large
conjunctival melanoma: a long-term study.
Eur J Ophthalmol 2005; 15:446450.
86. Prabhasawat P, Tseng SCG: Impression
cytology study of epithelial phenotype of
ocular surface reconstructed by preserved
human amniotic membrane. Arch
Ophthalmol 1997; 115:13601367.
87. Meller D, Tseng SCG: Conjunctivochalasis:
literature review and possible
pathophysiology. Surv Ophthalmol 1998;
43:225232.
88. Di Pascuale MA, Espana EM, Kawakita T,
Tseng SC: Clinical characteristics of
conjunctivochalasis with or without
aqueous tear deciency. Br J Ophthalmol
2004; 88:388392.
89. Meller D, Maskin SL, Pires RTF, Tseng SCG:
symptomatic conjunctivochalasis refractory
to medical treatments. Cornea 2000;
19:796803.
90. Georgiadis NS, Terzidou CD: Epiphora
caused by conjunctivochalasis: treatment
with transplantation of preserved amniotic
membrane. Cornea 2001; 20:619621.
91. Rodriguez-Ares MT, Tourino R, Capeans C,
Sanchez-Salorio M: Repair of scleral
perforation with preserved scleral amniotic
membrane in Marfans syndrome.
92. Tan DTH, Chee S-P, Dear KBG, Lim ASM:
Effect of pterygium morphology on
pterygium recurrence in a controlled trial
comparing conjunctival autografting with
bare sclera excision. Arch Ophthalmol
1997; 115:12351240.
93. Di Pascuale M, Espana EM, Kawakita T,
Tseng SCG: Surgical management of
pterygium. Contemp Ophthalmol 2003;
4:213222.
94. Essex RW, Snibson GR, Daniell M, Tole DM:
Amniotic membrane grafting in the
surgical management of primary pterygium.
Clin Experiment Ophthalmol 2004;
32:501504.
95. Ma DH-K, See L-C, Liau S-B, Tsai RJF:
Amniotic membrane graft for primary
pterygium: comparison with conjunctival
autograft and topical mitomycin C
treatment. Br J Ophthalmol 2000;
84:973978.
96. Prabhasawat P, Barton K, Burkett G,
Tseng SCG: Comparison of conjunctival
autografts, amniotic membrane grafts and
primary closure for pterygium excision.
97. Tananuvat N, Martin T: The results of
primary pterygium compared with
98.
99.
100.
101.
102.
103.
104.
105.
106.
107.
108.
109.
110.
111.
112.
conjunctival autograft. Cornea 2004;

23:458463.
Solomon A, Pires RTF, Tseng SCG:
Amniotic membrane transplantation after
extensive removal of primary and recurrent
pterygia. Ophthalmology 2001;
108:449460.
Kawasaki S, Uno T, Shimamura I, Ohashi Y:
Outcome of surgery for recurrent pterygium
using intraoperative application of
mitomycin C and amniotic membrane
transplantation. Nippon Ganka Gakkai
Zasshi 2003; 107:316321.
Shimazaki J, Kosaka K, Shimmura S,
Tsubota K: Amniotic membrane
transplantation with conjunctival autograft
for recurrent pterygium 1. Ophthalmology
2003; 110:119124.
Miyai T, Hara R, Nejima R, et al: Limbal
allograft, amniotic membrane
transplantation, and intraoperative
mitomycin C for recurrent pterygium.
Ophthalmology 2005; 112:12631267.
Ma DH, See LC, Hwang YS, Wang SF:
Comparison of amniotic membrane graft
alone or combined with intraoperative
mitomycin C to prevent recurrence after
excision of recurrent pterygia. Cornea
2005; 24:141150.
Sangwan VS, Murthy SI, Bansal AK, Rao GN:
Surgical treatment of chronically recurring
pterygium. Cornea 2003; 22:6365.
Yao YF, Qiu WY, Zhang YM, Tseng SC:
Mitomycin C, amniotic membrane
transplantation and limbal conjunctival
autograft for treating multirecurrent pterygia
with symblepharon and motility restriction.
Graefes Arch Clin Exp Ophthalmol
2006; 244: 232236.
Honavar SG, Bansal AK, Sangwan VS, Rao
GN: Amniotic membrane transplantation for
ocular surface reconstruction in
Solomon A, Espana EM, Tseng SCG:
reconstruction of the conjunctival fornices.
Barabino S, Rolando M: Amniotic
membrane transplantation elicits goblet cell
repopulation after conjunctival
reconstruction in a case of severe ocular
cicatricial pemphigoid. Acta Ophthalmol
Scand 2003; 81:6871.
Barabino S, Rolando M, Bentivoglio G, et al:
Role of amniotic membrane transplantation
for conjunctival reconstruction in ocularcicatricial pemphigoid. Ophthalmology
2003; 110:474480.
Katircioglu YA, Budak K, Salvarli S,
Duman S: Amniotic membrane
transplantation to reconstruct the
conjunctival surface in cases of chemical
burn. Jpn J Ophthalmol 2003; 47:519522.
Zhou SY, Chen JQ, Chen LS, et al: Longterm results of amniotic membrane
transplantation for conjunctival surface
reconstruction. Zhonghua Yan Ke Za Zhi
2004; 40:745749.
Jain S, Rastogi A: Evaluation of the
outcome of amniotic membrane
reconstruction in symblepharon. Eye 2004;
18:12511257.
Tseng SCG, Di Pascuale MA, Liu D-Z, et al:
Intraoperative mitomycin C and amniotic
membrane transplantation for fornix
reconstruction in severe cicatricial ocural
113.
114.
115.
116.
117.
118.
119.
120.
121.
122.
123.
124.
125.
126.
127.
128.
surface diseases. Ophthalmology 2005;

112:896903.
Nava-Castaneda A, Tovila-Canales JL,
Monroy-Serrano MH, et al: Comparative
study of amniotic membrane
transplantation, with and without
simultaneous application of mitomycin C in
conjunctival fornix reconstruction. Arch Soc
Esp Oftalmol 2005; 80:345352.
Barton K, Budenz D, Khaw PT, Tseng SCG:
Glaucoma ltration surgery using amniotic
membrane transplantation. Invest
Demir T, Turgut B, Akyol N, et al: Effects of
amniotic membrane transplantation and
mitomycin C on wound healing in
experimental glaucoma surgery.
Fujishima H, Shimazaki J, Shinozaki N,
Tsubota K: Trabeculectomy with the use of
amniotic membrane for uncontrolled
glaucoma. Ophthalmic Surg Lasers 1998;
29:428431.
Inoue T, Hirata A, Kimura A, et al:
Mitomycin C trabeculectomy in an eye with
cicatricial conjunctiva following amniotic
membrane transplantation. Acta
Budenz DL, Barton K, Tseng SCG:
repair of leaking glaucoma ltering blebs.
Am J Ophthalmol 2000; 130:580588.
Kee C, Hwang JM: Amniotic membrane
graft for late-onset glaucoma ltering leaks.
Am J Ophthalmol 2002; 133:834835.
Reim M, Teping C: Surgical procedures in
the treatment of severe eye burns. Acta
Ophthalmol (Copenh) 1989; 67(Suppl):4754.
Kenyon KR, Tseng SC: Limbal autograft
Pires RTF, Chokshi A, Tseng SCG: Amniotic
membrane transplantation or limbal
conjunctival autograft for limbal stem cell
deciency induced by 5-fluorouracil in
glaucoma surgeries. Cornea 1999;
19:284287.
Santos MS, Gomes JA, Hofling-Lima AL,
et al: Survival analysis of conjunctival
limbal grafts and amniotic membrane
transplantation in eyes with total limbal
stem cell deciency. Am J Ophthalmol
2005; 140:223230.
Tsubota K, Satake Y, Ohyama M, et al:
Surgical reconstruction of the ocular
surface in advanced ocular cicatricial
pemphigoid and StevensJohnson
122:3852.
Tsubota K, Satake Y, Kaido M, et al:
Treatment of severe ocular surface
disorders with corneal epithelial stem-cell
transplantation. N Eng J Med 1999;
340:16971703.
Solomon A, Ellies P, Anderson DF, et al:
Long-term outcome of keratolimbal
allograft with or without penetrating
keratoplasty for total limbal stem cell
109:11591166.
Ikari L, Daya SM: Long-term outcomes of
keratolimbal allograft for the treatment of
severe ocular surface disorders.
Ophthalmology 2002; 109:12781284.
Espana EM, Grueterich M, Ti SE, Tseng
SC: Phenotypic study of a case receiving a
keratolimbal allograft and amniotic
membrane for total limbal stem cell
130.
131.
132.
133.
134.
135.
136.
137.
138.
through amniotic membrane patch by

fluorescein. Cornea 2005; 24:359360.
Yoshita T, Kobayashi A, Takahashi M,
Sugiyama K: Reliability of intraocular
pressure by Tono-Pen XL over amniotic
membrane patch in human. J Glaucoma
2004; 13:413416.
Koranyi G, Seregard S, Kopp ED: The cutand-paste method for primary pterygium
surgery: long-term follow-up. Acta
Uy HS, Reyes JM, Flores JD, Lim-BonSiong R: Comparison of brin glue and
sutures for attaching conjunctival
autografts after pterygium excision.
Espana EM, Di Pascuale M, Grueterich M,
et al: Keratolimbal allograft in corneal
reconstruction. Eye 2004; 18:406417.
Shimazaki J, Shimmura S, Fujishima H,
Tsubota K: Association of preoperative tear
function with surgical outcome in severe

Ophthalmology 2000; 107:15181523.
139. Gris O, Campo Z, Wolley-Dod C, et al:
Amniotic membrane implantation as a
therapeutic contact lens for the treatment
of epithelial disorders. Cornea 2002;
21:2227.
140. Khokhar S, Sharma N, Kumar H, Soni A:
Infection after use of nonpreserved human
amniotic membrane for the reconstruction
of the ocular surface. Cornea 2001;
20:773774.
141. Marangon FB, Alfonso EC, Miller D, et al:
Incidence of microbial infection after
amniotic membrane transplantation.
Cornea 2004; 23:264269.
CHAPTER 66
129.

110:481486.
Jurowski P, Gos I: Keratolimbal allografts
and multilayer amniotic membrane
transplantation in the treatment of ocular
surface disease due to chemical burns. Klin
Oczna 2004; 106:648652.
Kruse FE, Joussen AM, Rohrschneider K,
et al: Cryopreserved human amniotic
membrane for ocular surface
reconstruction. Graefes Arch Clin Exp
Ophthalmol 2000; 238:6875.
Tissue Action Plan, Tissue Reference
Group Annual Reports, and Federal
Register (69 Fed Reg. 68612 and 68643,
Nov. 24, 2004). Available at FDA website:
http://www.fda.gov/cber/tissue/trgfyrpts.htm.
FDA website: http://www.fda.gov/
cber/compl/ambio062305.htm.
Kobayashi A, Ijiri S, Sugiyama K, et al:
Detection of corneal epithelial defect
893
CHAPTER
67
Keratoprosthesis
Bilal F. Khan, Mona Harissi-Dagher, and Claes H. Dohlman
INTRODUCTION
The restoration of vision in patients with corneal blindness has
become increasingly successful with advances in standard penetrating keratoplasty progressing since the beginning of the twentieth
century. There remains a subset of patients with severe corneal
opacities in whom penetrating keratoplasty fails or carries a poor
prognosis.1,2 The outcome depends primarily on the severity of
the underlying disease, availability of donor tissue, tissue banking
facilities, funding and other factors. According to a large recent
study,3 grafts for all causes remained clear in 70% of the cases
after 5 years. In a separate study, only 20% of the rst regrafts
similarly survived, whereas all repeat regrafts failed in the same
5-year period.4 Access to corneal transplantation is even more
limited in the developing world. According to the World Health
Organization, more than 12 million people are blind in the world
due to corneal diseases, ve million of whom are bilaterally blind
from trachoma alone.5 Success of a standard corneal transplant
in these cases is very poor, and they are seldom performed.
Keratoprosthesis (KPro) can be considered as an alternative in
cases where standard corneal transplant has failed or has a very
low probability of success such as repeat graft failures and cicatrizing diseases. The very severe complications traditionally
associated with KPro surgery, namely corneal melt, extrusion
and endophthalmitis have recently been reduced to a signicant
extent.68 Over the last decade signicant progress has been
made in the KPro design, postoperative management, and identication of prognostic categories.2 This has led to change in the
mind-set of the corneal surgeon community and now KPros are
being viewed as an accepted treatment in certain categories like
repeated graft failure.
HISTORY
The concept of an articial cornea for the treatment of corneal
blindness was rst suggested in writing by the noted French
surgeon, Pellier de Quengsy, in 1789, at the time of the French
Revolution.9 He suggested the implantation of a glass lens held
by a silver ring in a completely opaque cornea to restore vision.
He proposed the surgical procedure and designed the necessary
instruments. The outcome was not reported in press.
Later in 1853, Nussbaum performed experimental work with
a glass stud in a rabbit eye and published human trials using a
quartz crystal implanted into the cornea.10 Over the next 50 years,
Heusser,11 Dimmer,12 Salzer,13 and von Hippel14 continued further efforts in design and insertion techniques. However, extremely
high incidences of early complications were associated with
these KPros, which typically failed owing to tissue necrosis,
with subsequent leak, infection, and extrusion of the device.
After 1906, when Zirm performed the rst successful

human-to-human corneal graft, attention became diverted
away from KPro development. Years later, however, it became
clear that penetrating keratoplasties would not be successful in
all cases in fact most failed during the rst 50 years of the
twentieth century.
In a renewed effort, Verhoeff,15 in 1920, reported on a single
case of insertion of a quartz button into a patients cornea.
However, it had to be removed shortly afterward. Similarly in
1935, Filatov implanted a full penetrating glass device into an
opacied cornea of a patient and covered it with a double
conjunctival flap postoperatively, achieving an ambulatory
vision of 1/200.16
After 1950, KPro research gained momentum. During World
War II, it was noted that polymethylmethacrylate (PMMA)
splinters imbedded in the corneas of pilots were well tolerated.
This led to experiments with implantation into animal corneas
by Wnsche,17 Stone and Herbert,18 and others. Soon, human
applications followed, and a few ophthalmologists, particularly
Cardona,19,20 attempted to use these new inert plastics for KPro
designs. Once again, however, many of these cases were associated with serious complications, and the procedure lost favor
with most surgeons. On the other hand, some persevered in
developing their techniques and rened them over the years,
achieving a measure of success.
The combined experience of surgeons preeminent in the
development of KPro surgery probably has not amounted to
more than ~40005000 cases during the past half-century a
small number when compared with the number of penetrating
keratoplasties carried out on a worldwide basis, presently over
100 000 cases/year. The published results of many of these KPro
series are difcult to interpret. This is mainly because visual
acuity was often recorded as a single outcome at one point in
time without an indication of length of follow-up or duration of
retention of the KPro. Also, much focus and attention has been
centered on the design and materials, whereas follow-up of
complications, such as glaucoma, retroprosthetic membranes,
melts, and vision-threatening retinal complications, have
received less attention. Likewise, the necessity for close followup and frequent revisions has not been sufciently emphasized,
and long-term outcomes have been underreported. The rate of
attrition and the incidence of eventual functional loss of vision
have not always been determined, nor were details of the
preoperative diagnosis of patients always included. Such details
are crucial because there are denitive subgroups of patients in
which KPro carries a much more favorable outlook than in
others. Despite these limitations, a considerable body of
knowledge in this complex eld has accumulated, leading to the
present advances.
895
KPro MATERIALS, DESIGNS, AND

IMPLANTATION TECHNIQUES CURRENTLY
IN USE
We are not able, within the space allocated, to do historical justice to the ingenuity and creativity of all the contributors over
time to the eld of KPro. The devices presented here are those
that have in the recent past been used in large numbers. This
effort is continued primarily in approximately a dozen centers
worldwide and encompasses diverse approaches. This should
not diminish the importance of smaller scale experimental
attempts that have often been quite ingenious.
CARDONA KPro
Cardona should be given the credit for being the rst leader in
the modern KPro era. He has persisted with his endeavors, contributing many models over several decades, and has pursued
perfection in the eld. The modied Cardona technique described
here is one of several designs developed by him.21 It is less
invasive than the nut and bolt KPro made of PMMA previously
described by him, in which the retention system is placed inside
the anterior chamber and behind the recipient cornea.19,20,22
removal, and anterior vitrectomy are then performed. The

PMMA cylinder is placed through the opening, and then the
layers of Teflon, Dacron, and tibial periosteum or fascia are
sequentially placed on top of the cornea securing each layer
separately to the underlying tissue. The bulbar conjunctiva is
reapproximated. In cicatricial diseases, the optical nub is
protruded through the lids.23
Osteoodonto KPro and Similar Devices

This original principle was introduced by Strampelli24 and has
been modied by a number of followers (Falcinelli et al,7,25,26
Marchi et al,27 Temprano,28,29 Grabber et al,30 Liu et al,31
Hille,32,33 and others). This is a two-step procedure using the
patients own tooth and alveolar bone as a support for a PMMA
stem (Fig. 67.2). Temprano uses a very similar technique28,29 in
which the patients tibial bone is utilized as the scaffolding.
Pintucci et al,34,26,35 Girard et al,36 Legeais et al,37 and others
have replaced the autologous tooth-derived skirt with biocolonizable porous plastic materials. Pintucci uses a PMMA core
with a Salvage lamentous Dacron fabric skirt for support
(Fig. 67.3).
Device
The KPro illustrated in Figure 67.1 has an optical cylinder made
of PMMA. It is 3.5 mm in diameter and is available in
7.59.5 mm length. The extra length is required for the throughthe-lid KPro approach. This technique is essentially a multilayering of different materials on top of the KPro. The Teflon
skirt is 8.5 mm wide and 0.3 mm thick with 1.8 and 0.5 mm
holes. There is a central 3.5 mm opening. The Dacron mesh is
22.0 mm wide with a 3.5 mm central opening.
Implantation Techniques
FIGURE 67.2. Schematic representation of a cross section of an
OOKP.
Reprinted from Hille K, Grabner G, Liu C, et al: Standards for modied
osteoodontokeratoprosthesis (OOKP) surgery according to Strampelli and
Falcinelli. The Rome-Vienna protocol. Cornea 2005; 24:895908.
SECTION 6
The patients own tibial periosteum or fascia is harvested. The

corneal epithelium is removed and the conjunctiva is also
removed out to the rectus muscle insertions. A central corneal
opening with radial incisions is made. Radial iridotomies, lens
896
FIGURE 67.1. Cardona KPro: Everything over the Cornea, used in

repeated graft failure, pseudophakic bullous keratopathy.
(1) Epithelium, (2) fascia lata, (3) Dacron mesh, (4) teflon skirt, and
(5) recipient cornea.
FIGURE 67.3. Assembled Pintuccis PMMA and Dacron KPro.
Reprinted from Cardona H: The cardona keratoprosthesis: 40 year experience.

Reprinted from: Pintucci S, Pintucci F, Caiazza: New Dacron tissue colonizable

keratoprosthesis: clinical experience. Br J Ophthalmol 1995; 79:825829.
Keratoprosthesis
Step 1A: ocular surface preparation for the OOKP

The ocular surface is prepared for receiving freshly harvested
autologous buccal mucosa by removing the corneal epithelium
and completing the resection of the conjunctiva and the tenons
layer up to the insertion of the recti muscles. The buccal mucosa is then secured to the episclera without stretching. If
healthy buccal mucosa is not available, mucosa from other
surfaces like the palate, lip, or vagina can be used.
Step 1B: device preparation

A tooth with the largest available root is removed with its
surrounding alveolar bone. Care is taken to preserve the
dentoalveolar ligament. In adentulous patients, tibial bone28,29
or tooth from a rst-degree relative can be used.7
This tooth is then ground to make a central opening for the
PMMA optical cylinder, which has a specic dioptric power.38 It
is 8.08.25 mm long, with a 6.0 mm anterior section and a
2.50 mm posterior section. The diameter of the posterior
section is 0.30.4 mm wider than that of the anterior section.
At least 1 mm of dentine is present surrounding the cylinder.
The dentine is then dried and the optical cylinder cemented to
it using acrylic cement (methyl methacrylate monomer).
The complex is then placed in a subcutaneous pouch for
~3 months. The pouch is made in the orbitozygomatic area
inferior to the lower lid of the opposite eye. This process enables
the OOKP to become vascularized and develop connective tissue
and periosteum growth. It also allows sufcient time to detect
any infections in the buccal mucosa.
Step 2: implantation techniques

After ~3 months in the pouch, the OOKP complex is removed.
At this time, the complex is expected to have complete connective tissue coverage. The buccal mucosa covering the ocular
surface is then opened and hinged downwards. A central corneal
opening, equal to the size of the posterior portion of the cylinder,
is made along with three radial corneal incisions starting from
the central opening. Similar to Cardonas technique, a complete
iridectomy, intracapsular lens removal, and anterior vitrectomy
are performed. This helps intraocular pressure control and
prevents retro KPro membrane formation. The radial incisions
are closed and the OOKP secured with sutures, with the dentine
surface facing the cornea. The buccal mucosa is then
reapproximated and a central opening is made (Fig. 67.4). The
globe is inflated with air and the patient is placed supine for
45 days. A scleral shield for cosmesis can be placed later.
This procedure is invasive but has a reputation for stability
and low rate of infection. The patients are followed regularly
and need one to two yearly spiral CT scans to assess dentine
resorption.
Pintucci KPro
Pintucci uses essentially the same technique as the OOKP
except that the supporting skirt is made of Salvage lamentous
Dacron fabric surrounding the PMMA cylinder, which
simplies the procedure (Fig. 67.3). A large number of Pintucci
devices have been implanted, particularly in India.
AlphaCor
Device
AlphaCor39,40 is made from poly(2-hydroxyethylmethacrylate),
previously called the Chirila KPro.39 Polymerization conditions
result in a central transparent optic surrounded by an annular
sponge skirt, by an interpenetration of the polymers at the
interface, allowing tissue ingrowth into the sponge pores. The
device is 7.0 mm in diameter.
Implantation techniques
An intrastromal pocket is dissected at ~50% depth within the
central cornea, and a central 3.5 mm diameter disk of stroma
posterior to the optic is removed at the time of implantation.
The IntraLase laser can also be used to create the pocket. The
pocket can also be made within an existing failed corneal
transplant. Adjunctive procedures such as lens removal or tube
implantation, and Gunderson flaps, are not routinely required.
The device is secured within its pocket and allowed to be tissue
colonized over 3 months (Fig. 67.5). After this, occasionally the
anterior stroma will have cleared sufciently to restore vision,
and refractive correction can be undertaken without requiring
further surgery. In most cases, however, residual scarring requires that stromal tissue anterior to the device optic be excised
(Fig. 67.6). Postoperative refractive correction with high oxygen
FIGURE 67.5.
Appearance at
completion of
AlphaCor implantation.
CHAPTER 67
Courtesy of Dr Celia Hicks

Lions Eye Institute, Perth,
Western Australia.
FIGURE 67.6.
Appearance 3 years
after AlphaCor
implantation, with the
device optic, a fullthickness corneal
replacement.
FIGURE 67.4. Eye with OOKP in place. The optic cylinder enters the
eye through the cornea and is apparent at the surface of the buccal
mucous membrane graft.
Reprinted from Hille K, Grabner G, Liu C, et al: Standards for modied
osteoodontokeratoprosthesis (OOKP) surgery according to Strampelli and
Falcinelli. The Rome-Vienna protocol. Cornea 2005; 24:895908.
Courtesy of Dr Celia Hicks

Lions Eye Institute, Perth,
Western Australia.
897

6 mm.
3 mm.
Initially the extrusion rates were as high as 13%, but subsequently, with more surgical experience and better implant
designs, the rate was reduced to ~3%.
Legeais and Lacombe

Over the years Legeais and Lacombe from France have contributed several innovative KPro materials and designs. They
have a rich experience of several hundred patients implanted
with these devices. Lacombe45 has used a PMMA KPro with
xation secured by a large posterior diameter support. Eventual
stability is provided by the brosis produced by the endothelialDescemet tissues. Legeais et al have reported two generations37,46 of KPros, which were implanted in the stroma. The
rst generation37 of KPros consisted of a PMMA optic attached
by clips to a colonizable expanded-polytetrafluoroethylene
(e-PTFE) skirt. Their second-generation KPro46 used polydimethylsiloxane (PDMS) coated with polyvinylpyrrolidone as the
optic and an 11 mm PTFE skirt molded to the optic with heat.
FIGURE 67.7. Champagne Cork KPro. (A) Hood, (B) stainless steel
loops inserted into the rim of the hood, and (C) anticonical shaft.
At the Massachusetts Eye and Ear Inrmary, we have for a

number of years used a PMMA KPro of double-plated collar
button design, which we have modied over the years
(Fig. 67.8).47,48
permeability contact lenses, that also provide surface protection, is commonly done.
Device
Worst KPro
The Worst KPro has been championed, particularly by Singh
in India, in a large number of cases. They have used the
Champagne Cork KPro41,42 which is secured with steel wires
(Fig. 67.7). The central PMMA cylinder is cone shaped with a
6.0 mm flange. The cone tapers from a central 3.0 mm to a
larger apex. The KPro is secured with eight 80 mm stainless steel
wires tied in pairs to the sclera.
SECTION 6
Former Soviet Union
898
Boston KPro
Reproduced from Worst JGF: Twenty three years of keratoprosthesis research:

present state of art. Refract Corneal Surg 1993; 9:188189.
In the former Soviet Union, a large number of patients, especially those with chemical burns, have been implanted with
devices of different designs by way of a PMMA optical core
developed by Moroz, Kalinnikov, Yakimenko, and others.37,43,44
With over 2000 cases, the Fyodorov Institute in Moscow, the
Filatov Institute in Odessa, and other former Soviet Union
centers, are believed to have the worlds most extensive
experience in KPro surgery.
Professor Moroz and Dr Kalinnikov have contributed several
innovative, alternative designs. One is based on the HEMA
monomer diffusing from a trephine opening in the center of a
graft (corneal or cartilage), thereby anchoring the poly-HEMA
central optics. Alternatively, PMMA can be incorporated centrally with a similar technique. The HEMA monomer, that has
inltrated the corneal tissue and polymerized there, anchors the
central optics.
Another approach has employed titanium haptics that are
inserted intrastromally, followed by exposure of the modied
graft to a mixture of formaldehyde, glutaraldehyde, and dextran
for a month, followed by 1 month of washout. This cross-linked
corneal stroma, resistant to swelling, with the titanium haptics
can then be stored for a long time. A central 2.2 mm trephine
opening is subsequently done, followed by insertion of a PMMA
stem. The whole complex is then inserted intralamellarly into
the patients corneal stroma.
Dr Yakimenko, at the Filatov Institute in Ukraine, has
reported 502 cases43 using their design of a central core PMMA
optical cylinder and a tantalumtitanium alloy haptic with wide
apertures for xation by tissue ingrowth, intrastromally.
The Boston KPro consists of two plates joined by a stem, which

constitutes the optical portion (Fig. 67.9). This is made to be
implanted into a corneal graft. The type I KPro, the most
frequently used type, has a front plate of 5.06.0 mm diameter
and has the appropriate dioptric powers polished into it,
depending on patients axial length and phakic status.49,50 The
back plate, to be secured onto the stem, is 7.08.5 mm in
diameter and has eight holes of 1.3 mm each. The holes have
proven to be of value because they facilitate nutrients from the
aqueous to reach the stroma and the keratocytes. In addition,
the holes allow aqueous to more rapidly replace fluid that has
evaporated from the corneal surface. A titanium-locking ring is
placed behind the back plate to prevent any intraocular
unscrewing of the device. The type II KPro is similar to type I,
except that it has a 2 mm long anterior nub designed to
protrude through the lids.
The type I Boston KPro is favored in eyes with reasonable
blink and tear secretion mechanisms (Fig. 67.9). The advantages of this design include a short optical stem, which provides
a good view with the slit lamp, a generous visual eld, and good
stability due to the wide plates that prevent tilting of the device
FIGURE 67.8. Designs of Boston KPro. Collar-button-shaped device.
Keratoprosthesis
then placed. This contact lens stays on for life with appropriate
replacements that may be necessitated by tears and losses.
Patients are also placed on lifelong light prophylactic topical
antibiotics, a fluoroquinolone, and vancomycin (14 mg/mL),
once daily.
In countries where the availability or cost of a well-tested
corneal graft is too great of a nancial imposition, the patients
own excised button can serve as carrier for the KPro.
Others
Many small-scale efforts, such as the ingenious Seoul-type

KPro, are also currently under way.51
FIGURE 67.9. Assembly of Boston KPro. (a) The stem of the

mushroom-shaped front plate passes through a trephinated 3 mm
central opening in a large corneal graft. (b) The posterior plate then
screws onto the exposed threads of the stem tightly to sandwich the
cornea between front and back plates. (c) A titanium-locking ring is
then placed on to the back surface of the back plate. The graft-KPro
combination is now ready to be sutured in place like a standard
transplant.
off the visual axis. The design also facilitates repair of necrosis
of tissue around the stem to occur. Patients with heavy exposure
to evaporation, such as end-stage ocular cicatricial pemphigoid
(OCP) and StevensJohnson syndrome (SJS), may still be
candidates for the procedure but would need the type II Boston
KPro. Its 2 mm nub extends through the lids along with
extensive tarsorraphy and a wedge resection of the upper lid
around the superior aspect of the nub. Other lid reconstruction
procedures are also required in order to avoid exposure of the
surrounding tissues.
Implantation technique
A fresh donor corneal graft with a 3.0 mm central opening is
slid on to the stem of the front plate (Fig. 67.9). The back plate
is then screwed on to the stem. This is followed by placement
of the titanium-locking ring behind the back plate. The
patients cornea is then trephined in the standard corneal
transplant manner. If necessary, cataract removal is performed.
The KPro cornea complex is sutured into the patient cornea as
in a standard corneal transplantation (Fig. 67.10). A soft contact
lens, usually a Kontur (Kontur Kontact Lens, Inc., Richmond,
CA) 16.0 mm diameter and 9.8 mm base curve, plano lens is
Indications and Prognostic Categories

KPro is still evolving as a procedure and it is difcult to give
clear and precise indications of its use. If a standard corneal
transplant has a good chance of giving longstanding vision, this
would be the preferred technique. However, if one or more graft
failures occur within months after surgery, reducing vision to
nger counting or less, a KPro may be considered.
Because the outcome of KPro surgery differs markedly among
various corneal diseases, the indication for such surgery should
be categorized accordingly (Table 67.1). In general, some criteria
must be fullled before qualifying for the procedure. First, endstage retinal disease, optic nerve disease or end-stage phthisis
constitute contraindication. Second, monocular status obviously poses more risk. In addition, young age or poor general
health should be taken into consideration. If the long-term
survival of the KPro is questionable, it follows that elderly
patients have a greater chance of trouble-free course than
younger patients do. Moreover, whether a procedure is advisable
is dependent not only on the patients condition but also on the
experience, interests and time commitment of the surgeon.
This said, guidelines can be suggested. The most risky category
is SJS and OCP, both of presumed autoimmune etiology. SJS
patients are often young, are usually binocularly involved and
are in desperate need of a KPro that should remain complication-free for many years. They should be approached with
caution, as this is a difcult goal to achieve. They often have
ongoing ocular inflammation, which increases the complication
rate postoperatively. Patients with OCP are usually older, which
works in their favor. All autoimmune etiology patients respond
to any surgical intervention with an intense and often
prolonged inflammation as well as with necrosis of the tissue
holding the KPro. In the through-the-lid technique, they are
also prone to postoperative skin retraction and glaucoma.52
Chemical burn patients can have good results after KPro
surgery, but these patients often have severe glaucoma and
abnormal sensitivity to pressure to begin with. This should not
necessarily be considered a contraindication to surgery,
however, since a glaucoma shunt procedure or cyclophotocoagulation can often save the situation.53
Contrary to the previous categories with history of inflammation, the use of KPro for graft failure with noninflammatory
CHAPTER 67
TABLE 67.1. Keratoprosthesis Prognostic Categories From

Best to Worst
1. Non-inflammatory conditions: graft failures in dystrophies,
degenerations, aniridia, trauma, etc.
2. Infectious: HSV, HZV, bacterial and fungal ulcers
3. Chemical burns
4. Autoimmune diseases: SJS, OCP, etc.
FIGURE 67.10. Successful implantation of Boston KPro after 7 years.
899

edema, dystrophies, degenerations, infection and trauma often
do very well.2 Blink mechanism and tear secretion are usually
normal. The frequency of severe postoperative uveitis or
medically uncontrollable glaucoma is low, and good vision is
usually restored even more rapidly than after a successful regraft.
The clear optics often allows excellent vision. The prognosis of
those with a history of herpetic keratitis may be less bright,54
however, our own experience with the Boston KPro in herpes
simplex and aniridia has been very encouraging.(unpublished).
Even in acutely inflamed herpetic keratitis with epithelial
defects, we have observed rapid cessation of the inflammation.
PATIENT EVALUATION
symblephara should be noted routinely. The corneal surface

should be judged for irregularity, keratinization, epithelial defects,
and sub-epithelial vascularization. Stromal opacity from scarring
or edema as well as any deep vascularization should be evaluated. Anterior chamber depth and the status of iris, pupil, and
lens (or intraocular lens) all merit detailed notes. The fundus is
often not observable in KPro candidates, but when possible, an
effort should be made to examine disk cupping and macular
changes. Disk cupping has, of course, high prognostic importance and may dictate aqueous shunt implantation. Gross changes
in the posterior pole, such as age-related macular degeneration,
are vital to observe. Special attention should be given to signs of
inflammation throughout the examination as its presence
severely influences the long-term prognosis of KPro surgery.
HISTORY
Taking a detailed history of the ocular condition, as well as any
important systemic disease is mandatory. This usually reveals
the underlying cause of the corneal condition whether dystrophy,
trauma, surgical, or infectious disease. Duration of symptoms,
laterality of the condition, and nature of condition, whether
episodic or progressive are all important to record. Details and
dates of previous surgery should be solicited. History of glaucoma is particularly important in predicting outcome, especially
following chemical burns (Table 67.1).
VISUAL ACUITY
Visual acuity should be recorded in the standard fashion using
a Snellen chart. Relative contributions of the cornea, cataract,
retina, or optic nerve are difcult to ascribe in severely damaged
eyes. If the corneal surface is highly irregular in the presence of
only moderate stromal opacities, a hard contact lens refraction
can be revealing. A standard visual eld test is rarely applicable
in these cases, but visual acuity with alternatively gross
projection of a strong light source is useful to assess. Testing
central xation, and particularly light projection nasally is often
helpful. If nasal projection is lost, end-stage glaucoma must be
suspected.
SECTION 6
INTRAOCULAR PRESSURE
Severe corneal damage often makes exact intraocular pressure
measurements impossible and precludes view of the optic
nerve. Recording intraocular pressure can be fraught with error.
Usually in severe corneal pathology, pneumotonometry is more
reliable than the applanation technique, but both can give
grossly erroneous readings. Simple digital palpation, even if
imprecise, is frequently the most dependable approach.
BLINK RATE AND TEAR SECRETION

On examination, blink mechanism and tear secretion are important factors in assessing KPro prognosis. Evaporative damage to
the corneal tissue around a KPro can be detrimental, especially
if a soft contact lens cannot be retained. Blink rate and completeness can be estimated when the patient does not feel
observed. Lagophthalmos and frank chronic exposure are
important to recognize. Tear secretion should be measured with
Schirmers test. Finally, tear breakup time may be valuable in
assessing the health of the ocular surface.
SLIT-LAMP EXAMINATION
900
This is the cornerstone of the patient evaluation. Eyelids should

be inspected for marginal incongruities. Conjunctival inflammation, surface keratinization, and fornix foreshortening or
SPECIAL EXAMINATIONS
Ultrasound examination is necessary in most cases. B-scan can
reveal a retinal detachment or massive debris behind an opaque
cornea or lens. While, a B-scan shows the presence or absence
of an intraocular lens, it cannot measure glaucomatous optic
nerve cupping with precision. If a glaucoma shunt has been
implanted previously, the B-scan can identify the fluid cleft over
the shunt plate. This indicates patency of the tube shunt, but it
does not rule out the presence of a dense capsule that may have
formed around the plate, obstructing flow and causing high
intraocular pressure. For implantation into an aphakic recipient
eye, an A-scan measurement of the axial length is also required
for proper selection of a KPro with the correct dioptric power.
DOCUMENTATION
Pre- and postoperative external photography and detailed
drawings of the eyes help document baseline and allow assessment of progress and outcome of the surgery.
POSTOPERATIVE CARE
Follow-up visits should be individualized, but typically, they
have to initially be frequent in order to evaluate for infection,
inflammation, or glaucoma. We routinely see our patients the
day after surgery, in a week, 3 weeks, and then monthly. After
6 months, the intervals can be 23 months. Intraocular pressure
elevation should be checked regularly and managed accordingly.
It is prudent to share the postoperative responsibility with a
glaucoma colleague if pressure is threatening.
Indenite prophylactic antibiotic use after surgery is advised.
Excluding patients with autoimmune diseases or severe
chemical burns, it seems adequate to treat patients with a
fourth-generation fluoroquinolone initially four times a day,
then rapidly decreasing to once daily for an indenite period. In
addition, one can use topical vancomycin 1.4%.7,55 once daily.
Since the inclusion of vancomycin in our treatment protocol,
we have not had any acute bacterial endophthalmitis among our
patients. Compliance is very important and nonobservance can
be problematic. Systemic antibiotics are recommended, such as
cephalexin 500 mg, 24 times a day for 1 week after the surgery
(penicillin allergy dictating a substitute).
Corticosteroid drops, usually as prednisolone acetate 1%, are
given as needed for postoperative inflammation. In cases with
severe inflammation, subtenon injection of 40 mg triamcinolone is recommended but not more frequently than every
23 weeks. Intraocular pressure elevation and fat necrosis can
be a complication of the subtenon injections.
In cases of herpes simplex, addition of systemic antivirals
(e.g., acyclovir 400 mg twice a day) is recommended on a
permanent basis.
Keratoprosthesis
COMPLICATIONS
In the past, it was primarily tissue necrosis around the device,
leak, extrusion, and/or endophthalmitis that brought a violent
end to the KPro effort, often with total loss of the eye. In most
cases, any severe complication (Table 67.2) is seen within the
rst year after surgery; however, the patient is never totally safe
and requires close monitoring.
During the last few decades, however, thanks to the work of
several groups of surgeons and investigators, the picture has
become much brighter. Several factors have improved the
outcome (Table 67.3).
TISSUE NECROSIS, MELT, AND EXTRUSION

Extrusions still occur in certain KPro varieties more than
others. Buccal mucosal resorption can occur in the OOKP
which, if the patient is followed regularly, can be identied early
and a new buccal graft can then be placed. Melt of the cornea
around the AlphaCor can occur and should be replaced with
fresh corneal tissue. Tissue necrosis and subsequent melt are
now rare with the Boston KPro. Should the melt occur, prompt
intervention is wise.56
epithelial defects, and stromal thinning can occur with longterm undesirable consequences. Therefore, the addition of a
soft contact lens after Boston and AlphaCor KPro surgery, and
its retention (with occasional replacement) for an indenite
time, has been of benet to the health of the tissue around the
device. The lens seems to diffuse the evaporative forces well and
allows better hydration of the exposed tissues.57
INFLAMMATION
In autoimmune diseases such as OCP, SJS, graft-versus-host
disease, severe uveitis, etc., a chronic low-grade intraocular
inflammation is a frequent complication. Consequently, a retroprosthetic membrane, vitreous opacities, epiretinal membranes,
retinal detachment, and angle closure glaucoma may supervene.
Corticosteroids are the standard treatment to suppress such
developments. Topical prednisolone drops are routine, sometimes augmented by peribulbar/subtenon injections of triamcinolone. Systemic steroids are less frequently used because of the
less favorable riskbenet ratio. Postoperative inflammation is
related to the degree of preoperative inflammation status and is
the single most important hindrance to postoperative good vision.
RETROPROSTHETIC MEMBRANE
Evaporation and irregular drying of corneal tissue around the
KPro can be a disturbing problem. Drying, dellen formation,
TABLE 67.2. Most Signicant Keratoprosthesis Complications

1. Tissue Necrosis
a. Melt of surrounding corneal tissue
b. Aqueous leak
c. Endophthalmitis
2. Postoperative Uveitis
a. Retroprosthetic membrane
b. Vitreous opacities
c. Macular edema
d. Retinal detachment
e. Phthisis
4. Glaucoma
5. Dentine Resorption in OOKP
TABLE 67.3. Factors Improving Keratoprosthesis Outcome

Design of device
Intraocular inflammation post-KPro surgery can lead to a retroprosthetic membrane which results in a decline in vision. Repeated
steroid injections (triamcinolone) are indicated at the rst sign of
such a membrane formation. Once formed, it is worthwhile to
open the membrane with Nd:YAG laser before it becomes too
thick or vascularized.58 Laser pulses with energy above 2.03.0 mJ
are inadvisable because the plastic can crack or become pockmarked. If the membrane becomes thick, leathery, and particularly if vascularized, a closed vitrectomy under high infusion
pressure and membranectomy are required to restore vision.59
INFECTIOUS ENDOPHTHALMITIS
This is the ultimate disaster after KPro surgery. Vision can be
lost permanently within 24 h. Even in recent times, endophthalmitis has occurred in OCP and SJS patients.6 The infectious
agents are almost always Gram-positive organisms. However,
bacterial endophthalmitis in the Boston KPro has now been
practically eliminated with the adherence to the present
regimen of a low level of antibiotic prophylaxis of vancomycin
and a fluoroquinolone.55 It is very important to impress upon
the patient that meticulous compliance for life is recommended.
Should an endophthalmitis still occur, tap and antibiotic injection are crucial immediately. An aqueous tap via the limbus for
smear and culture is followed by an injection of 1.0 mg vancomycin, 0.4 mg amikacin, and 0.4 mg dexamethasone. The
patient is hospitalized for topical and intravenous antibiotics. A
vitrectomy may be deemed necessary later. Useful vision is
rarely restored after an infectious endophthalmitis.
CHAPTER 67
SOFT CONTACT LENS LOSS
Identication of prognostic categories

Tissue coverage
STERILE UVEITISVITRITIS
Contact lens use
A sudden massive vitritis has been observed in some patients

with temporary reduction of vision to hand motions.60 This
vitritis masquerades as a bacterial endophthalmitis but with no
accompanying pain, tenderness, or redness. Pathogens are not
isolated in these cases. The patients are treated with topical and
peribulbar steroids. Within a few weeks, or months, the vitreous
clears and the vision returns back to the baseline level prior to
the event. We speculate that it is a sterile immune reaction
because, had the reaction been due to bacterial infection, most
of the vision would be wiped out.
Prophylactic antibiotics
Corticosteroids
Nd:YAG laser membranectomy
Glaucoma tube implant in severe cases
Repair techniques
Meticulous, regular follow-up
901
GLAUCOMA
CONCLUSION
With the drastic reduction in endophthalmitis, glaucoma is now

the most serious complication after KPro surgery. Its pathogenesis is probably multifactorial, but gradual closure of the
anterior chamber angle is the most likely cause. It is therefore
vital to monitor the intraocular pressure and nerve damage
postoperatively. Tonometers are virtually useless in this setting
and digital palpation of the globe is the main method available
to ascertain a rough estimate of intraocular pressure. Glaucoma
drops can penetrate into the eye and are effective. Oral carbonic
anhydrase inhibitors have the expected effect but should be
used with caution in patients with SJS, and completely avoided
in patients with sulfa allergy. When medical control of glaucoma
is insufcient, an Ahmed valve shunt or cyclophotocoagulation
may be indicated. Particularly vulnerable are the autoimmune,
chemical burns, or preexisting glaucoma patients.61
The ophthalmic surgeon facing the individual patient with

severe end-stage corneal opacity will have to choose between,
doing a standard corneal transplant, doing a KPro, or doing
nothing? The boundaries between these three alternatives are
always fluid depending on many factors, but it is clear that the
pendulum is swinging markedly in favor of the articial cornea.
In the US, in the early 1990s the skeptism about the safety of
KPros was so profound that very few devices were implanted. The
number in 1992 was less than 15, whereas in year 2005 the
number had reached to almost 300 a 20-fold increase.
This progress has been due to the collective effort of the few
groups worldwide that have been devoted to the long-term task
of making the KPro safer, cheaper, and simpler to implant and
manage. There is obviously much to be improved, however. The
earlier common problems of tissue necrosis, melt and leak
around the device are now largely eliminated. Likewise, the risk
of infection, which in the past was the most dreaded
complication, is now almost completely preventable as long as
a regimen of low-level prophylactic antibiotic drops can be kept
up. However, glaucoma is a long-term risk factor, which is still
not completely eradicated, despite progress with medication
and experimental valve shunts. Finally, long-term postoperative
intraocular inflammation in severe cases can be a stubborn
problem that cannot always be kept under control with steroids.
On the positive side, properly machined and polished PMMA
has such almost flawless optical quality and sturdiness that
postoperative vision can be excellent as long as the rest of the eye
allows it. The various KPros have advantages and disadvantages
but, on the whole, there is no question that this procedure has
already achieved a solid niche in corneal surgery and that its
future is promising.
RETINAL DETACHMENT
Retinal detachment is not a common complication. It can be
rhegmatogenous or tractional in nature. It is diagnosed by direct
visualization or by B-scan ultrasonography. Three-port vitrectomy is performed with or without silicone oil tamponade or
long-acting gases are used. The prognosis is ominous.59
DENTINE RESORPTION
Dentine resorption is a long-term problem in OOKP, resulting
in instability and extrusion of the device. However, this complication can be detected by spiral CT scans. If much dentine
resorption has taken place, the entire surgical procedure may
have to be repeated.
SECTION 6
REFERENCES
902
1. Ma JK, Graney J, Dohlman CH: Repeat

penetrating keratoplasty versus the boston
keratoprosthesis in graft failure. Inter
Ophthal Clin (Controversies in
Ophthalmology) 2005; 45:4959.
2. Yaghouti F, Nouri M, Abad JC, et al:
Keratoprosthesis: preoperative prognostic
categories. Cornea 2001; 20:1923.
3. Williams KA, Muehlberg SM, Lewis RF, et al:
How successful is corneal transplantation?
A report from the Australian Corneal Graft
Register. Eye 1995; 9:219227.
4. Besudsky V, Blum-Hareuveni T, Rehany U,
et al: The prole of repeat corneal
108:461469.
5. Thylefors B, Negrel AD, Pararajasegaram R,
et al: Global data on blindness:
reviews/analysis. Bull World Health Org
1995; 73:115121.
6. Nouri M, Terada H, Alfonso EC, et al:
Endophthalmitis after keratoprosthesis:
incidence, bacterial causes, and risk
factors. Arch Ophthalmol 2001;
119:484489.
7. Falcinelli G, Falsini B, Taloni M, et al:
Modied osteo-odontokeratoprosthesis for
treatment of corneal blindness. Arch
Ophthalmol 2005; 123:13191329.
8. Dohlman CH, Doane MG: Some factors
influencing outcome after keratoprosthesis
surgery. Cornea 1994; 13:214218.
9. Pellier de Quengsy G: Precis ou cours
doperation sur la chirurgie des yeux. Paris:
Didot; 1789.
10. Nussbaum N: Cornea Articialis, ein

Substitut fur die Transplantatio Cornea.
Deutsche Klinik 1853; 34:367.
11. Heusser J: Die Einheilung einer Cornea
articialis. Oesterr Ztschr Pract Med 1860;
26:424.
12. Dimmer F: Zwei Falle von Celluoidplattern
der Hornhaut. Klin Monatsbl Augenheilkd
1891; 29:104.
13. Salzer F: Uber den kunstlichen
Hornhautersatz. Wiesbaden; 1898.
14. von Hippel A: Uber die operative
Behandlung totaler stationarer HornhautTrubungen. Albrecht v. Graefes Arch Clin
Ophthal 1887; 23:79.
15. Verhoeff FH, cited in Cardona H:
Keratoprosthesis. Am J Ophthalmol 1962;
54:284.
16. Filatov VP: Alloplastik bei vollstandig
hoffmungslosem Leukomen. Sov Viest
Opht 1936; 9:400.
17. Wnsche G: Versuche zur totalen
Keratoplastie und zur Cornea Artecialis.
Arztliche Forschung 1947; 1:345.
18. Stone W Jr, Hebert E: Experimental study
of plastic material and replacement of the
cornea. Preliminary report. Am J
19. Cardona H: Keratoprosthesis: acrylic
optical cylinder with supporting
intralamellar plate. Am J Ophthalmol 1962;
54:284294.
20. Cardona H, De Voe AC: Postkeratoplasty.
Trans Am Acad Ophthalmol 1977;
83:271280.
21. Cardona H: The Cardona keratoprosthesis:

40 years experience. Refract Corneal Surg
1991; 7:468471.
22. Cardona H: Prosthokeratoplasty. Cornea
1983; 2:179183.
23. Cardona H, DeVoe AG: Prosthokeratoplasty.
1977; 83:271280.
24. Strampelli B: Osteo-odontokeratoprosthesis.
Ann Ottalmol Clin Ocul 1963; 89:10391044.
25. Falcinelli G, Missiroli A, Pettiti V, et al:
Osteo-odontokeratoprosthesis up to date.
In: Blodi F, Brancato R, (eds) Acta XXV
consilium ophthalmologicum. Milano:
Kugler & Ghedini; 1987.
26. Pintucci S, Pintucci F, Cecconi M, et al:
New Dacron tissue colonizable
keratoprosthesis: clinical experience. Br J
Ophthalmol 1995; 79:825829.
27. Marchi V, Ricci R, Pecorella I, et al:
Osteo-odonto-keratoprosthesis.
Description of surgical technique with
results in 85 patients. Cornea 1994;
13:125130.
28. Temprano J: Resultados a largo plazo de
Osteo-odonto-queratoprothesis y
queratoprotesis tibial. An Inst Barraquer
1998; 27(Suppl):5365.
29. Temprano J: Keratoprosthesis with tibial
auto graft. Refract Corneal Surg 1993;
9:192193.
30. Stoiber J, Csaky D, Schedle A, et al:
Histopathologic ndings in explanted
osteo-odontokeratoprosthesis. Cornea
2002; 21:400404.
Keratoprosthesis
42. Andel P, Worst J, Singh I: Results of
champagne cork keratoprosthesis in 127
corneal blind eyes. Refract Corneal Surg
1993; 9:189190.
43. Yakimenko S: Results of a PMMA/titanium
keratoprosthesis in 502 eyes. Refract
Corneal Surg 1993; 9:197198.
44. Moroz ZI: Articial Cornea. In: Fyodorov SN,
ed. Microsurgery of the eye: main aspects.
Moscow: Mir; 1987.
45. Lacombe E: Results of 30 keratoprosthesis
with retro corneal xation. J Fr Ophthalmol
1993; 16:426434.
46. Legeais JM, Renard G: A second generation
of articial cornea (Biokpro II ). Biomaterials
1998; 19:15171522.
Prosthokeratoplasty. Am J Ophthalmol
1974; 77:694670.
48. Dorzee MJ: Kratoprothse en acrylique.
Bull Soc Belge Ophtalmol 1955;
108:582593.
49. Barraquer J: Keratoplasty and
keratoprosthesis. Pocklington
Memorial Lecture delivered at the Royal
College of Surgeons of England on 5th
May, 1966. Ann R Coll Surg Engl 1967;
40:7181.
50. Doane MG, Dohlman CH, Bearse G:
Fabrication of a keratoprosthesis. Cornea
1996; 15:179184.
51. Kim MK, Lee JL, Wee WR, et al: Seoul-type
keratoprosthesis: preliminary results of the
rst 7 human cases. Arch Ophthalmol
2002; 120:761766.
52. Dohlman CH, Terada H: Keratoprosthesis in
syndrome. Adv Exp Med Biol 1998;
438:10211025.
53. Netland PA, Terada H, Dohlman CH:
Glaucoma associated with
keratoprosthesis. Ophthalmology 1998,
105:751757.
54. Dohlman CH, Abad JC, Dudenhoefer EJ,

Graney JM: Keratoprosthesis: beyond
corneal graft failure. In: Spaeth G, ed.
Ophthalmic surgery. Principles and
practice. Philadelphia, PA: WB Saunders;
2003:199207.
55. Dohlman CH, Nouri M, Barnes S, et al:
Prophylactic antibiotic regimens in
keratoprosthesis. Invest Ophthalmol Vis Sci
2003; 1455. ARVO abstract.
56. Dohlman Ch: Postoperative regimen and
repair of complications after
keratoprosthesis. Refract Corneal Surg
1993; 9:198.
57. Dohlman CH, Dudenhoefer EJ, Khan BF,
et al: Protection of the ocular surface after
keratoprosthesis surgery: the role of soft
contact lenses. CLAO J 2002; 28:7274.
58. Bath PE, McCord RC, Cox KC: Nd:YAG
laser discission of retroprosthetic
membrane: a preliminary report. Cornea
1983; 2:225228.
59. Ray S, Khan BF, Dohlman CH, et al:
Management of vitreoretinal complications
in eyes with permanent keratoprosthesis.
60. Nouri M, Durand ML, Dohlman CH: Sudden
reversible vitritis after keratoprosthesis. An
immune phenomenon? Cornea 2005;
24:915919.
61. Dohlman CH, Terada H: Keratoprosthesis in
syndrome. In: Sullivan D, ed. Lacrimal
gland, tear lm and dry eye syndromes II.
Basic science and clinical relevance. New
York: Plenum Publishing; Adv Exp Med Biol
1998; 438:10211025.
CHAPTER 67
31. Liu C, Herold J, Sciscio A, et al: Osteoodonto-keratoprosthesis surgery. Br J

32. Hille K: Keratoprothesen. Klin Aspekt.
33. Hille K, Grabner G, Liu C, et al: Standards
for modied osteo-odontokeratoprosthesis
(OOKP) surgery according to Strampelli
and Falcinelli. Cornea 2005; 24:895908.
34. Pintucci S, Pintucci F, Cecconi M, et al: The
dacron felt colonizable keratoprosthesis:
after 15 years. Eur J Ophthalmol 1996;
6:125130.
35. Pintucci S, Pintucci F, Caiazza S: The
dacron felt colonizable keratoprosthesis.
36. Girard LJ, Hawkins RS, Nieves R, et al:
Keratoprosthesis: a 12-year follow-up.
Trans Sect Ophthalmol Am Acad
37. Legeais JM, Renard G, Parel JM, et al:
Keratoprosthesis with biocolonizable
microporous fluorocarbon haptic.
Preliminary results in a 24-patient study.
38. Merlin U, Profazio V, Merlin F, et al: Optical
calculations in mesokeratoprosthesis. In:
Caramazza R, Versura P, eds. Biomaterials
in ophthalmology: an interdisciplinary
approach. Bologna: Studia E.R. Congressi;
1990:171176.
39. Crawford GJ, Hicks CR, Lou X, et al: The
Chirila keratoprosthesis: phase I human
clinical trial. Ophthalmology 2002;
109:883889.
40. Hicks CR, Crawford GJ, Lou X, et al:
Corneal replacement using a synthetic
hydrogel cornea, AlphaCor: device,
preliminary outcomes and complications.
Eye 2003; 17:385392.
41. Worst JG: Twenty-three years of
keratoprosthesis research: present state of
art. Refract Corneal Surg 1993; 9:188189.
903
SECTION 7
REFRACTIVE SURGERY
Edited by Dimitri T. Azar amd Jorge L. Ali
CHAPTER
68
History, Development, and Classication of

Refractive Surgical Procedures
Andre Cohen and Roger F. Steinert
The earliest report of surgery to change refractive error dates

back to 1885, when Schiotz1 used a limbal relaxing incision
(LRI). In 1894, Bates2 noticed flattening in the axis of traumatic
peripheral corneal scars. In 1898, Lans3 also described methods
to reduce astigmatism.
In 1939, Sato4 was the rst to describe posterior corneal incisions for the treatment of astigmatic error in keratoconus.
Satos technique led to corneal decompensation and was modied to anterior incisions by Yanaliev.5 He published his
results of 426 incisional refractive cases between 1969 and
1977.6 Fyodorov and Durnev79 continued to rene keratotomy
and popularized the technique. In 1978, Bores, Myers, and
Cowden were the rst in the United States to perform radial
keratotomy and report their results.10,11
Lamellar methods for refractive surgery were evolving
simultaneously. Jose Barraquer was the rst to correct refractive
error through lamellar surgery.1216
More recently, in addition to incisional and lamellar surgery,
the eld of refractive surgery has expanded to include intracorneal and intraocular implants.
Today, the eld of refractive surgery has evolved to include
many modalities for the treatment of different refractive errors.
In this chapter, we will review the history and development of
these techniques and at the same time classify them into
categories based on their mechanism of action.
CLASSIFICATION OF REFRACTIVE
SURGERY
There are several ways to classify types of refractive procedures.
Waring et al17 based his classication on the type of general
method of altering the refractive power of the cornea: lamellar,
keratotomy, keratectomy, collagen shrinkage, and penetrating
keratoplasty. Refractive procedures can also be grouped
according to the mechanism of action or on the anatomical
location of surgery.18 Table 68.1 illustrates our preferred method
of classication, a matrix of anatomical location and methodology. This method will be used throughout this chapter.
ADDITION
KERATOPHAKIA
The term keratomileusis comes from the Greek words kera
(cornea) and mileusis (carving). Barraquer developed this
technique in the late 1950s. The initial use of this procedure
was for correction of aphakia. Barraquer developed the rst
microkeratome in order to create a full corneal cap. Under the
cap, he placed a positive power shaped lenticule of donor
corneal stroma. The cap draped over the lenticule and was
sutured until healing occurred. The lenticule was shaped on a
TABLE 68.1. Proposed Classification of Keratorefractive Surgical Procedures

Optical Location
Supercial
Intrastromal
Addition
Subtraction
Epikeratophakia
PRK
Synthetic epikeratophakia
LASEK
Keratophakia
LASIK
Intracorneal lenses
Intrastromal laser
Pinhole aperture
Keratomileusis in situ
Relaxation
Coagulation-Compression
Corneal molding
Lamellar keratotomy
BKS keratomileusis
Classic myopic keratomileusis
Peripheral cornea
Intracorneal ring segments
Wedge resection
Radial keratotomy
Thermokeratoplasty
Hexagonal keratotomy
Compression sutures
Arcuate keratotomy
BKS, Barraquer-Krumeich-Swinger; LASEK, laser subepithelial keratomileusis; LASIK, laser stromal in situ keratomileusis; PRK, photorefractive keratectomy.
From Yanoff: Ophthalmology., 2nd edn., Copyright 2004 Mosby, Inc; and.
Azar DT:. Refractive Ssurgery. Ophthalmology., 2nd edition. St. Louis: Mosby;, 2004;128.
905
REFRACTIVE SURGERY
cryolathe, which stabilized the cornea by freezing it in order to
lathe cut it. The lenticule was then thawed and placed as a
sandwich inside the cornea. Because of high cost for the cryolathe, and the technical complexity of the surgery, keratomileusis was not widely adopted.
EPIKERATOPLASTY
This procedure, rst described by Kaufman in 1979,19 initially
corrected aphakia and later was applied to high myopia and
keratoconus. In this procedure, a lenticule is cut in the shape to
provide the required power and then used fresh or, more commonly, prepared and freeze dried for distribution at a central
commercial facility. At surgery, the patients epithelium is
removed and an area of stroma trephinated partial thickness
and dissected to create a peripheral ring-shaped pocket. The
edge of the lenticule is then sutured into the pocket and covered
by regrowth of the patients epithelium.
A 10-year retrospective study showed stable visual acuity with
good lenticule condition and minimal induced astigmatism.20
The advantages to this procedure were its potential
reversibility and the relative ease of use of tissue prepared and
shipped commercially. However, because of variable outcomes,
particularly in myopia, and loss of BCVA in many patients,
epikeratoplasty has largely been abandoned.
SECTION 7
INTRACORNEAL LENS
Several histological and clinical animal studies have assessed
the safety of intracorneal lenses.21 In 1966, Dohlman and
Brown published their work on buried corneal implants.22 In
1985, Choyce23 and Lindstrom and Lane24 implanted high
index of refraction polysulfone lenses. In both of these studies,
the barrier to diffusion of nutrients through the cornea caused
unacceptable toxicity. In 1990,25 Steinert et al showed that high
water content hydrogel lenses in aphakic correcting powers to
be safe and tolerated by the cornea, but the optical results were
disappointing in many cases. In 2002, Ismail26 tested Nutrapore, a microporous permeable hydrogel that allows flow of
corneal metabolites, and showed it to be safe and tolerated by
the stroma in rabbits.21
The intracorneal lens may or may not have the same
refractive index and water content of the cornea, depending on
whether its mechanism is to change the anterior curvature or
change the refractive power of the cornea. To alter the anterior
curvature, the implant must be placed under a lamellar flap that
drapes over the lens and follows its curvature.
In 2004, a retrospective study was published on outcomes of
Permavision lenses made of Nutrapore. The study was performed on 10 eyes with spherical hyperopia. The mean postoperative SE refraction was +0.03 0.36D, and the mean
uncorrected visual acuity was 0.85 0.13. The study concluded
these lenses to be safe and effective; however more long-term
follow-up is needed.21
A retrospective observational study on four hyperopic eyes
with implanted lenses found a signicant increase in higher
order aberrations. This nding emphasizes the importance of both
optical design and precise centration of an implanted lens.27
The ReVision Optics Corporation is developing a small
hydrogel intracorneal implant for the correction of presbyopia.
A slit lamp photo of human intracorneal implant is shown in
Figure 68.1.
INTRACORNEAL RINGS
906
In 1999 the US Food and Drug Administration approved

INTACs intracorneal ring segments made of polymethyl-
FIGURE 68.1. A 2 mm intracorneal hydrophilic intracorneal implant

designed to correct presbyopia.
Photo courtesy ReVision Optics.
methacralate for treatment of low myopia (<3 D of myopia

and < 1 D of astigmatism). In July 2004, the FDA approved
INTACs segments for keratoconus. The Ferrara Rings, which
have different dimensions than INTACs, are available
internationally but not in the US.
Forseto et al compared Intacs to LASIK and showed better
results with LASIK at 5 years. (See Adriana SF, Schor P, Belfort
R, Nose W: Segmentos de anilos corneales intrastromales para
la correccion de myopia baja: resultados a 5 anos. Presented at
ASCRS, San Diego, CA, 15 May 2004.) The principal current
use of INTACs is for post-LASIK ectasia and naturally occurring
keratoconus, but reports of outcomes are limited. Siganos et al28
published results of 33 keratoconus eyes. In this group, the
mean uncorrected visual acuity (UCVA) signicantly improved
in 33 eyes. Boxer Wachler29 reported results of 74 keratoconus
eyes, where mean best corrected logmar vision improved from
0.4120/501 0.48 to 0.2420/302 0.31. UCVA and astigmatism
also improved.
The concept of refractive change induced by intracorneal ring
segments is illustrated in Figure 68.2.
NONREFRACTIVE IMPLANTS
The AccuVision Corporation is sponsoring an investigation of
an intracorneal implant with a pinhole conguration. The
device is implanted in one eye to create increased depth of focus
and ameliorate presbyopia. This device is currently under
investigation with no published reports of outcomes at the
present time.
SUBTRACTION
MYOPIC KERATOMILEUSIS
This technique followed the principles of keratophakia, discussed above. Instead of adding an optically shaped donor cornea,
however, the cap obtained by the microkeratome was placed on
the cryolathe which froze the cornea to allow a lathe to thin the
disk centrally on its backside (Fig. 68.3). As with keratophakia,
myopic keratomileusis was a technically challenging procedure
and required great skill on the part of the surgeon.
Barraquer, Krumeich, and Swinger later developed a method
to shape the corneal cap without freezing (BKS system).30 After
History, Development, and Classication of Refractive Surgical Procedures
Intracorneal ring
flattening of
anterior lamellae
FIGURE 68.2. Intracorneal ring. After a peripheral circular lamellar

dissection, two polymethyl methacrylate ring segments of
predetermined diameter and thickness are inserted. The midperipheral
anterior lamellae are lifted focally by the ring segments, which results
in a compensatory flattening of the central anterior lamellae and hence
a decrease in the refractive power of the cornea.
FREEZE KERATOMILEUSIS
cutting a corneal cap with a microkeratome, the cap was

inverted, placed on a dye that was shaped according the desired
refractive change, stabilized with suction applied through
multiple ports in the dye, and then recut on the caps stromal
underside with a microkeratome. The reshaped cornea was
then sutured back onto the eye.31 Few reports were published for
manual keratomileusis in situ. In 1991, Bas and Nano
published a large series of patients and reported the procedure
to have unpredictable results.32 In addition, ArenasArchila
reported the results of their experiences with manual keratomileusis to be unsafe and unpredictable.33
In an effort to improve predictability and practicality, Ruiz
developed the technique of automated lamellar keratoplasty
(ALK) in the late 1980s. An automated microkeratome rst cut
a corneal cap. The suction ring was then adjusted and a second
pass of the microkeratome removed a lenticule of stroma from
the exposed corneal bed. The lenticule was discarded, and after
replacement of the cap, the resultant cornea was flatter, thereby
providing myopic correction.34 Predictability was better than
the earlier procedures but remained unsatisfactory, and the relatively small optical zone created by the second microkeratome
pass led to poor quality vision due to induced aberrations.
PHOTOREFRACTIVE KERATECTOMY
Trokel and Srinivisan were the rst to discover a new way of
corneal reshaping with use of the excimer laser. Trokel heard
that the ultrashort-wavelength emission of the excimer laser
was observed to remove corneal tissue in laser damage investigations. Curious about possible applications, Trokel enlisted
the assistance of Srinivasin, an IBM engineer who was studying
the excimer laser for engraving silicon computer chips. Their
pioneering work demonstrated that the argonfluoride (ArF)
excimer laser wavelength of 193 nm could cleanly remove
corneal stroma with minimal damage to the adjacent stroma.
Trokel realized the potential application for refractive corneal
surgery. Initially, the laser was to be used to create radial
keratotomy35,36 type incisions. This approach was rapidly
abandoned because of the tissue removal, which created a wide
kerf rather than an incision. Attention turned to removing,
through laser ablation, corneal tissue directly over the central
visual zone after rst removing the overlying epithelium, which
would regrow postoperatively. The ability to remove corneal
tissue on the outer surface while leaving a clear cornea was
previously impossible.37,38 This procedure was termed photorefractive keratectomy (PRK) by Trokel and Marshall, and after
extensive animal studies, formal human trials began in Europe
and the US.3943 The rst sighted human eye was treated by
McDonald and co-workers in 1988.44
The initial clinical excimer laser devices were made by
Summit, VISX, and Schwind. These all used powerful laser
cavities in order to create a broad beam that could be shaped by
apertures. Subsequent laser technology shifted to scanning slits
and spots. The Summit and VISX lasers were the initial lasers
to be approved for use by the FDA in the United States in 1995
and 1996, respectively.
Currently, in various lasers, the range of PRK approval by the
FDA is for 13 to +6 D and for 4 D of astigmatism.
LASER-ASSISTED IN SITU KERATOMILEUSIS
FIGURE 68.3. Freeze keratomileusis. A disc of parallel sides is

resected from the cornea with the microkeratome. After freezing the
disc, a lenticule of predetermined power is removed from the stromal
side with a lathe. The removed cornea is sutured back in place.
In 1990, Pallikaris rst used the excimer laser instead of the

second microkeratome pass for reshaping the stromal bed,45,46
introducing what was initially termed laser-assisted in situ
keratomileusis (LASIK). The term was later shortened to laser
in situ keratomileusis. Burrato and co-workers used the laser for
ablation to the posterior flap rather than the stromal bed, but
CHAPTER 68
INSERTION OF INTRACORNEAL RING
907
REFRACTIVE SURGERY
LASER-ASSISTED STROMAL IN SITU KERATOMILEUSIS

Flap lifted from eye
Lenticle removed using laser light

FIGURE 68.5. Docking of the IntraLase focusing cone into the
stabilizing ring in order to create a LASIK flap.
Photo courtesy of IntraLase Corporation.
Flap sewn back down
flap. A femtosecond laser lays down thousands of adjacent spots

in order to dene the intrastromal plane of the flap, and then
cut the side of the flap. The docking of the IntraLase laser is
shown in Figure 68.5.
As of 2006, LASIK is approved by the FDA in various devices
for a total range of 15 to +6 D of refractive error and for 6 D
of astigmatism. However, poorer optical performance and the
potential for inducing corneal ectasia with high myopic
corrections have led most surgeons to abandon higher level
corrections in most patients.
LASER SUBEPITHELIAL KERATOMILEUSIS
FIGURE 68.4. Laser-assisted stromal in situ keratomileusis. A flap

with parallel sides is lifted using the microkeratome. The excimer laser
is used to remove a lenticule of predetermined power from the
exposed corneal stroma. The flap, with its intact epithelium, is then
folded back, and as it drapes over the modied stromal surface, the
refractive power of the anterior corneal surface is modied. The
dotted area in the bottom panel corresponds to the lenticule that was
removed. Usually, no sutures are required.
A modication to PRK was introduced in 2003. In this technique, an epithelial flap is created and replaced after laser
ablation. In this technique, alcohol (typically 2030 s of a 20%
concentration of ethanol) is applied to the cornea to loosen the
epithelium. The rst laser subepithelial keratomileusis (LASEK)
was performed in 1996 at Mass Eye and Ear by Dmitri Azar.49
Camellin popularized the technique and named the procedure.
The theoretical advantages of this technique were avoidance of
flap related complications of LASIK, less haze as in PRK, and
quicker recovery.50 Other than avoiding the creation of a flap,
the latter two potential advantages have not been proven.
SECTION 7
EPI-LASEK
908
the results were not as promising as direct ablation of the

exposed corneal bed.47,48 As LASIK experience grew, surgeons
realized the advantage of using a hinged corneal flap rather than
a full free cap, improving safety as well as more accurate realignment of the flap. Microkeratomes were modied to create the
hinged flap automatically. The concept of myopic LASIK is
shown in Figure 68.4.
PRK decreased in popularity in the late 1990s as more
surgeons performed LASIK. The advantages were less postoperative pain, faster visual recovery, elimination of the
potential for haze and scarring as PRK heals, and a wider range
of treatment. Most studies showed better results in higher
myopia with LASIK than with PRK.
An alternative mechanism for flap creation utilizes laser
photodisruption instead of a cutting blade to create the LASIK
Similar to LASEK, an epithelial flap is mechanically created

with a microkeratome. As in LASEK, the epithelial flap is replaced
after stromal ablation. This mechanical technique avoids the
cellular toxicity effects of alcohol, with the hope that the flap of
epithelium will include basal lamina and be more likely to
survive than the alcohol loosened epithelial sheet of LASEK.
This has not been proven.
WAVE FRONT-GUIDED ABLATION

Until 2002, conventional treatment for lower order aberrations
was only available. With the introduction of wave front, higher
order aberrations could be treated. In 2002 and 2003, the FDA
approved waveguided treatment for the Alcon LADARWave
4000, Bausch & Lomb Technolas, and the VISX Star Wavescan.

In the Alcon LADARWave FDA trial, 79.9% of the wave front
treated eyes had uncorrected vision of 20/20 or better, 91.5%
with 20/25 or better, and 98.6% with 20/40 or better. Higher
order aberrations increased by 20% compared to 77% in the
conventional group. Furthermore, contrast sensitivity was
better in the wave front group.51
In the FDA trial for VISX, after excluding eyes needing
enhancement, 98% of eyes were 20/20 or better uncorrected at
12 months. In the Bausch & Lomb FDA studies, 91.5% of eyes
had UCVA of 20/20 at 6 months (BCS-refractive 135).
RADIAL KERATOTOMY
Partial thickness incisions
incisions
MULTIFOCAL ABLATIONS
The goal of a multifocal ablation is to compensate for presbyopia. Challenges in this approach are maintenance of optical
quality with precise centration, stability of the ablation contour
and the optical correction, and patient satisfaction with multifocal optics. In 2006, data from 100 eyes treated with multifocal
LASIK was presented at ASCRS. At 6-month follow-up there
were no retreatments. Of the 50 patients, 40 were 20/20 and J2
bilateral. Five patients were 20/25 and J2 bilateral, and ve were
20/25 and J3 bilateral. The presentation reported high patient
satisfaction. A prospective, nonrandomized trial of 300 eyes
was also introduced by the same group at ASCRS and the study
is ongoing.52
Compensatory flattening of the central cornea
INTRASTROMAL PHOTOABLATION
CRESCENTIC WEDGE RESECTION

Troutman described a technique of corneal wedge resection for
high astigmatism after penetrating keratoplasty.53 This technique is performed by removing a full-thickness crescent-like
portion of peripheral cornea in the flat meridian and then
resuturing the wound.
A retrospective study done in Rotterdam of 41 corneal wedge
resections showed a decrease from an average of 11.7 3.5 D, a
mean reduction of 8.2 D.54 Nevertheless, this technique is
unpredictable and not widely used. Its principle application is in
correction of high astigmatism after cornea transplantation.
RELAXATION
RADIAL KERATOTOMY
Sato4 in Japan was the rst to describe this technique. However,
his approach was to make radial incisions along the posterior
cornea. This led to an unacceptable rate of corneal edema due
to endothelial damage.
The concept of anterior radial keratotomy is illustrated in
Figure 68.6.
In 1981, the National Eye Institute funded the Prospective
Evaluation of Radial Keratotomy (PERK) study.55 It was formed
to determine outcomes of radial keratotomy. The study was a
nine center, self-controlled prospective clinical trial of radial
keratotomy performed in 1982 and 1983. It included 427
patients (793 eyes) with 2.00 to 8.75 D of myopia. The
technique was standardized to eight radial incisions using a
diamond knife. The incisions were spaced equidistant around a
central optical zone that extended to the limbus. Ultrasonic
FIGURE 68.6. Radial keratotomy. Partial-thickness incisions result in

ectasia of the paracentral cornea and compensatory flattening of the
central cornea.
pachymetry and preoperative cylcoplegic refraction helped

determine the blade length and the diameter of the clear zone,
respectively.56 The optical zones were 3.0, 3.5, or 4.0 mm; the
smallest corresponding to higher myopia.
Retreatments with eight more incisions were allowed in the
study, and second eyes could be done 1 year after the rst eye.
After initiation of the PERK study, two principal techniques
evolved. These were the American style (centrifugal) or
backcutting, which made incisions from the center to the
periphery and the Russian style (centripetal), or frontcutting
which made them from the limbus to the center.57 The Russian
style had a more reliable incision depth and better central
corneal flattening but was less safe than the American style.
Other techniques of radial keratotomy evolved including the
Genesis and Duotrack technique.58,59 This technique combined
the predictability of the Russian style and safety of the
American style.
Ten year results were published in 1994. Of the 793 eyes
receiving RK, 693 returned at 10 years for follow-up. Of the 675
eyes with refractive data, 38% were within 0.5D and 60%
within 1.00 D. Of the 310 rst operated eyes the mean refraction was +0.51D at 10 years, representing a signicant hyperopic shift from the mean refraction of 0.36 at 6-month
follow-up. Furthermore, between 6 months and 10 years 43% of
eyes had a hyperopic shift of 1.00 D or more. The 10-year data
also showed that among 681 eyes, uncorrected visual acuity was
20/20 or better in 53% and 20/40 or better in 85%. Loss of two
lines or more in Snellen acuity occurred in 3% of the total
793 eyes.60
CHAPTER 68
The theoretical potential of intrastromal ablation is the use precise photodisruption to remove tissue from inside the cornea,
causing the cornea to reshape itself and alter its optical properties.
Precise photodisruption requires an ultrafast (picosecond or
femtosecond) laser with precisely focused spot patterns. This
application has not reached clinical trials.
909
REFRACTIVE SURGERY
As evidenced by the PERK results, radial keratotomy was
unstable, and RK is less predictable than PRK and LASIK. RK
has been abandoned with the widespread availability of the
excimer laser.
ASTIGMATIC KERATOTOMY
Snellen, Schiotz, and Bates were the rst to correct corneal
astigmatism with transverse relaxing corneal incisions were. In
1898, Lans was the rst to perform a systematic study for the
correction of astigmatism.61 Others such as Lindstrom62 further
developed astigmatic keratotomy for treatment of corneal
astigmatism.
There have been many different AK techniques introduced
over the years. These include arcuate, transverse, trapezoidal,
and hexagonal keratotomy. An incision placed perpendicular to
the steep axis leads to flattening of the cornea by increasing the
radius of curvature in that meridian.6366 At the same time of
flattening, there is a coupling effect that steepens the opposite
meridian. The coupling ratio67 is the relationship between the
flattening of the incised axis to the opposite steepened axis. In
arcuate incisions parallel to the limbus, this ratio is ~1, meaning
that the overall spherical equivalent refractive error is unchanged.
Radial incisions alter the coupling ratio by as much as 5:1
when combined with straight (transverse) incisions. Therefore,
the coupling effect of astigmatic incisions can be reduced by
radial incisions.63 The steep meridian is flattened ve times as
much as the steeping of the flat meridian 90 away from the
incisions.
In general, the smaller the central optical zone, the more
powerful the effect but the higher the risk of inducing irregular
astigmatism. Furthermore, astigmatic keratotomy appears to
show better and more predictable results for idiopathic astigmatism compared to surgically induced astigmatism.68
ARCUATE AND TRANSVERSE KERATOTOMY

Arcuate incisions appeared to create more uniform relaxation of
the cornea69 and had a greater effect than straight incisions of
same chord length.68 The reason is that the ends of a straight
transverse incision moves away from the corneal center, with a
reduction in power. Merlin found that maximal flattening effect
occurred with two transverse incisions that totaled 110120.69
Adding multiple transverse or arcuate incisions is ineffective70
compared to two incisions. In determining the optimal clear
optical zone, several studies showed that the optimal effect
occurred when incisions were made at 5 mm.64,6971 Incisions
made with a smaller clear zone were associated with complications such as glare, ghost image aberrations, irregular
astigmatism, and overcorrection.64,65,7173
SECTION 7
TRAPEZOIDAL KERATOTOMY
This procedure, popularized by Ruiz, requires the surgeon to
make two pairs of semiradial incisions with two pairs of equally
spaced transverse incisions centered on the steep axis of the
cornea. These two transverse incisions are usually placed at 5
mm and 7 mm.74 The clear zone was determined by refraction,
keratometry, and sometimes age. Although some published
good results in reducing corneal astigmatism to as much as
56 D,7578 the surgery remained unpredictable and was not
better than astigmatic keratotomy.
HEXAGONAL KERATOTOMY
910
In 1952, Akiyama described using a hexagonal pattern of

intersecting incisions in the posterior cornea of rabbits.79 In
1986, Yamashita performed a similar procedure on the anterior

cornea in rabbits.80 In 1987, Mendez81 performed this procedure in humans for hyperopia. In his technique he intersected
the incisions which resulted in corneal instability. In 1989,
Jensen performed the technique without intersecting incisions.82
However the procedure was still complicated by instability of
corneal shape and irregular astigmatism.83 In the early 1990s
the procedure was further modied when tangential incisions
were placed peripheral to each hexagonal angle.84
This procedure was abandoned in the early 1990s because of
its continued unpredictability for the treatment of hyperopia.
LIMBAL RELAXING (PERIPHERAL CLEAR

CORNEA) INCISIONS
James Gills popularized this technique. The incisions are
performed at the 600 mm depth and placed just anterior to the
corneal limbus. The advantages are that it is more forgiving and
does not require complete centration at the axis. In contrast to
AK, LRI is more likely to preserve the optical quality of the
cornea. The increased distance from the center of the cornea
reduces irregular astigmatism and heals more reliably, particularly in the elderly cornea. It also means that the incisions
have less power and need to be longer than a more central
incision for the same effect. As a result, LRIs are mainly useful
for mild to moderate natural or cataract surgery-induced astigmatism. Corneal transplantation astigmatism is usually too high.
COLLAGEN SHRINKAGE
Gasset and Kaufman described a technique called thermokeratoplasty in 1975 for the treatment of keratoconus.85 In this technique, thermal energy was used to shrink the collagen within
the cornea. In 1980, Newmann et al coined the phrase radial
thermokeratoplasy, which was a technique used to treat hyperopia by heating the cornea to 600 celcius.86 Others tried to
perform thermal keratoplasty with carbon dioxide lasers, hot
copper wires, diode lasers, and cobalt:magnesium fluoride
lasers. These procedures were all proved unacceptable.87
HOLMIUM YAG LASER (NONCONTACT)

In 2000, the US Food and Drug Administration approved the
Sunrise Technologies Hyperion LTK (Laser Thermal
Keratoplasty) System. It was approved for temporary reduction
of hyperopia in patients with +0.75 to +2.50 D of MRSE with
less than or equal to 0.75 D of astigmatism, who were 40
years of age or older with documented stability of refraction for
the prior 6 months. The FDA approval document stated that
the magnitude of correction with this treatment diminished over
time, with some patients retaining some or all of their refractive
correction.88 Unpredictability and instability of results led to
discontinuation of the manufacturing of this system.
HO:YAG (CONTACT)
The contact handheld probe made by Summit emitted an
infrared pulse. Laser applications were placed onto the cornea
periphery. This device never received FDA approval.
CONDUCTIVE KERATOPLASTY
Conductive keratoplasty (CK) uses electrical conductive
properties to transfer energy through the stroma. This procedure uses radiofrequency energy. Eight to 32 spots in up to
three rings (6, 7, and 8 mm optical zones) are placed in the
peripheral cornea stroma using a ne tip, leading to steepening
Photo courtesy of Refractec.
of the central cornea and correction of hyperopia.89 The

procedure is shown in Figure 68.7.
The commercial device, known as the Viewpoint CK system,
was rst developed by Refractec (Irvine, CA) in 1993. In 2004,
it was approved by the FDA for the treatment of hyperopia in
patients over 40. In March 2004, it received approval for
presbyopia. Phase three clinical trials for CK post-LASIK are
ongoing.90
The 6-month data of the 1-year clinical trial showed 77% of
eyes were J3 or better uncorrected. 85% of all patients had
binocular distance of 20/25 or better and J3 or better
uncorrected. Furthermore, 66% of eyes treated were within
0.5 D of the intended refraction at 6 months.89 The hypothesis
is that the corneal contour after CK is partially multifocal,
leading to a better combined distance and reading vision
compared to monovision created by techniques such as PRK
or LASIK.
POSTERIOR CHAMBER INTRAOCULAR

LENS
MULTIFOCAL
Multifocal lenses are designed to spread the focal range from a
narrow zone to a broader one, in order to provide simultaneous
distance and near functional vision. The two most common
strategies are based on refractive and diffractive optics.
The Array lens (AMO) was the rst approved multifocal lens
in the United States. It is a refractive lens with progressive
zones that distribute ~50% of the light for distance, 35% for
near, and 15% for intermediate, with the exact distribution
varying depending on pupil size.91 Comparing bilateral implantation of the Array lens to a monofocal lens, the likelihood of
wearing glasses was 8% in the former and 32% in the latter.
However, as patients age and pupils become smaller the near
vision would decrease. In addition, glare (11% vs 1%) and haloes
(15% vs 6%) were increased in the Array versus the monofocal.92 A number of these lenses were explanted because of
nighttime haloes and other visual disturbances,93 and they did
not gain widespread acceptance.
Diffractive technology utilizes sharp concentric rings
imposed on a refractive base. The refractive base provides the
distance correction, while the diffraction divides the light
energy into several orders. The zero order (~41% of the light) is
not deviated and constitutes the distance focused image. The
rst order diffraction provides another 41% of light energy at a
ACCOMMODATIVE
The rst IOL attempting to achieve accommodation was a ringhaptic lens designed by Payer.96 Other lenses have been used
in Europe such as the Biocomfold lens, the Akkommodative
1CU,97,98 Synchrony (dual optic),99 and Morcher (Stuttgart,
Germany).100 The only FDA approved accommodating lens in
the United States is the EyeOnics Crystalens. It was approved
by the FDA in November 2003, with labeling that it provides
approximately D of accommodation. The Crystalens is a silicone
lens with an index of refraction of 1.43 and a 4.5-mm diameter
optic. The haptic ends are made of polyimide to encourage
strong xation in the equator of the capsular bag. The IOL
overall diameter is 11.5 mm and there is flexing of the lens at
the hinge. The flexed hinge allows for posterior vaulting of the
optic toward the posterior capsular bag.100 During accommodation, the optic is designed to move forward due to vitreous
pressure and tension changes on the capsule. The forward
motion of the optic increases the effective optical power.
PHAKIC IOLS
Phakic IOLs have numerous potential advantages in the
correction of refractive errors. In theory, any optical power can
be achieved, including correction of astigmatism if the lens is
rotationally stable. The optical correction remains stable, given
the absence of healing effects seen in cornea-based surgery. The
optical correction is immediate. The manufacturing of implants
is highly developed, beneting from the decades of experience
with IOLs and cataract surgery. However, many potential complications can occur with phakic IOLs, and no one style is immune
CHAPTER 68
FIGURE 68.7. Application of CK probe into corneal stroma.
closer focus, creating the near image. The remaining light

energy (18%) is in higher orders that remain unfocussed at all
times.91 Alcon has engineered a modication of the diffractive
rings known as apodization. In apodization, the height and
spacing of the diffractive rings is altered from the center to the
periphery, in an effort to reduce glare and halo at night when the
pupil enlarges.
The current generation of multifocal intraocular lenses
(IOLs) in the US are the refractive zonal-progressive ReZoom
(AMO) and the diffractive ReSTOR (Alcon) lenses.
Two types of ReSTOR lens were included in the FDA trials, a
single piece and three-piece IOL. Both types are made of hydrophobic acrylic. The FDA trial clearly showed less patient satisfaction with monocular versus binocular implantation. When
patients had binocular implantation, 88.1% of those with the
three piece achieved uncorrected distance vision of 20/25 or
better, and for those with the one piece lens, 88.4% were 20/25
or better. Near vision with best distance correction was J2 or better
in 90.5% of the three- piece lens group and 87% in the single
piece group. The best near acuity occurred at 31 cm (~12 in).94
The refractive ReZoom multifocal IOL was approved by the
FDA based on Array lens and Sensar lens data submitted by
AMO. No FDA information is available specic to the ReZoom
lens. However, European data of 200 patients showed that 93.3%
reported not needing spectacles for distance. 91.4% reported the
same for intermediate, and 66.7% for near. Furthermore, 93%
reported never or only occasionally wearing spectacles.95
Because the ReSTOR IOL has stronger near and weaker
intermediate uncorrected acuity, and ReZoom has stronger
intermediate and weaker near vision, some surgeons implant
one lens of each style in their patients. Other surgeons utilize a
strategy of bilateral implantation of the same lens design, but
offsetting the ReSTOR in a slightly hyperopic direction in one
eye to gain intermediate vision, or offsetting one ReZoom in a
slightly myopic direction to increase near vision.
911
REFRACTIVE SURGERY
to these problems which can become sight-threatening. These
include all potential complications of intraocular surgery,
including but not limited to endophthalmitis, glaucoma, chronic
inflammation, cataract, endothelial cell loss and corneal edema,
pupil deformity, and even retinal detachment. The crowding of
the anterior segment in hyperopic patients makes adds further
challenges compared to myopic eyes.
Phakic IOLs are xated in one of three locations: anterior
chamber angle-xated, anterior chamber iris-xated, or in the
posterior chamber between the iris and the crystalline lens. The
earliest iris-xated phakic IOL dates back to 1978. These lenses
were originally biconvex and led to endothelial failure and
corneal decompensation. A change to a concave/convex design
improved endothelial stability.101103 Another problem with
these lenses was ovalization of the pupil and glare.
The two phakic IOLs currently approved by FDA for use in
the United States are the Verisyse iris xated lens from AMO
and the Implantable Contact Lens (ICL) from Staar Surgical.19
Foldable and toric versions of the Verisyse are under development, as well as acrylic foldable angle-xated phakic IOLs.
SCLERAL EXPANSION
This procedure was rst attempted by Spencer Thornton, MD.
In this technique, the sclera was weakened by making eight or
more scleral incisions over the ciliary body (ACS: anterior

ciliary sclerotomy). Signicant adverse events such as anterior
segment ischemia occurred and ACS has been abandoned.
In 2001, Fukasaku and Marron published their work on
scleral silicone plugs into scleral incisions.104 Other attempts
include the PresVIEW scleral expansion bands and the use of
infrared laser energy to heat and change the shape of the sclera.
These techniques are based on the Schacar theory of accommodation, which has not gained wide acceptance. In brief, that
theory attributes the loss of accommodation to the slow
expansion of the crystalline lens with age, reducing tension on
the zonules. By expanding the sclera, zonular tension would be
increased and the ability of the ciliary muscle to cause shape
change in the lens would be restored.
CONCLUSION
In this chapter, we have provided a historical perspective into
the development of this eld. We have also classied the techniques available to the refractive surgeon of today. As we look at
all the different modalities available today and the history of
refractive surgery, it is apparent that tremendous strides have
been made upon the shoulders of Schiotz, Bates, Lans, Sato,
Fyodorov, and Barraquer and others who dedicated themselves
to advancement of refractive surgery.
SECTION 7
REFERENCES
912
1. Shiotz H: Ein Fali von hochgradigen

Hornhautastigmatus nach Stareextraction.
Besserung auf operativem Wege. Arch
Augenheilk 1885, 15:178181.
2. Bates WH: A suggestion of an operation to
correct astigmatism. Arch Ophthalmol
1894, 23:913.
3. Refractive surgery, basic and clinical
science course. San Francisco: American
Academy of Ophthalmology;
20052006:59.
4. Sato T: Treatment of conical cornea
(incision of Descemets membrane). Acta
Soc Ophthalmol Jpn 1939, 43:544555.
5. Yanaliev FS: Experience of surgical
treatment of myopia. Ann Ophthalmol
USSR 1979, 3:5255.
6. Choi DM, Thompson RW, Price FW:
Incisional refractive surgery. Curr Opin
Ophthalmol 2002, 13:237241.
7. Fyodorov SN, Agronovsky AA: Long term
results of radial keratotomy. L Ocul Ther
Surg 1982; 1:217223.
8. Fyodorov SN, Durnev VV: Operation of
dosaged dissection of corneal circular
ligament in cases of myopia of mild
degree. Ann Ophthalmol 1979;
11:18851889.
9. Durnev V: Decrease of corneal refraction by
anterior keratotomy method with the
purpose of surgical correction of myopia of
mild to moderate degree. In: Proceedings
of the rst congress of Ophthalmology of
Transcaucasia. Tbilisi, Georgia;
1976:129132.
10. Bores LD, Myers W, Cowden J: Radial
keratotomy: an analysis of the American
experience. Ann Ophthalmol 1981;
13:941948.
11. Bores LD: Historical review and clinical;
results of radial keratotomy. Int Ophthalmol
Clin 1983; 23:93118.
12. Barraquer JI: Oueratoplasia refractive.
Estudios Inform Oftal Inst Barraquer 1949;
10:221.
13. Bores L: Lamellar refractive surgery. In:

Bores L, ed. Refractive eye surgery.
Boston: Blackwell Scientic; 1993:324391.
14. Barraquer JI: Keratomileusis. Int Surg 1967;
48:103117.
15. Barraquer JI: Results of myopic
3:98101.
16. Barraquer JI: Method of cutting lamellar
grafts in frozen corneas: new orientations
for refractive surgery. Arch Soc Am
17. Waring GO III: Making sense of
keratospeak IV: classication of refractive
surgery, 1992. Arch Ophthalmol 1992;
110:13851391.
18. Barraquer JI: Cirugia Refractiva de la cornea.
Bogota, CO: Instituto Barraquer de
America; 1989:6785.
19. Barraquer JI: Keratomileusis and
keratophakia. In: Rycroft PV, ed. International
corneo-plastic conference (2nd: 1967:
Royal College of Surgeons of England).
Oxford: Pergamon; 1969:409443.
20. Cahill M, Condon P, Okeefe M: Long-term
outcome of epikeratophakia. J Cataract
Refract Surg 1999; 25:500507.
21. Michielaetto P, et al: Permavision
intracorneal lens for the correction of
hyperopia. J Cataract Refract Surg 2004;
30:21522157.
22. Dohlman CH, Brown SI: Treatment of
corneal edema with a buries implant. Trans
Am Acad Ophthalmol Otolarygol 1966;
70:267269.
23. Choyce DP: The correction of refractive
errors with polysulfone contact inlays.
A new frontier to be explored? Trans
24. Lane SS, Lindstrom RL: Polysulfone
intracorneal lenses. Int Ophthalmol Clin
1991; 31:3746.
25. Steinert RF, et al: Hydrogel intracorneal
lenses in aphakic eyes. Arch Ophthalmol
1996; 114:135141.
26. Ismail MM: Correction of hyperopia with

intracorneal implants. J Cataract Refract
Surg 2002; 28:527530.
27. Alio JL et al: Intracorneal hydrogel lenses
and corneal aberrations. J Refract Surg
2005; 21:247252.
28. Siganos SS, Kymionis GD, Kartakis N,
et al: Management of keratoconus with
Intacs. Am J Ophthalmol 2003;
1135:6470.
29. Boxer Wachler BS, Christie JP,
Chandra NS, et al: Intacs for keratoconus.
Ophthalmology 2003; 110:10311040.
30. Swinger CA, Krumeich JH, Cassidy D:
Planar Lemellar keratoplasty. 1986;
2:1724.
Academy of Ophthalmology,
20052006:106.
32. Bas AM, Nano HD: In situ myopic
keratomileusis: results in 30 eyes at
15 months. Refract Corneal Surg 1991;
7:223231.
33. Arena-Archilla E, Sanchez-Thorin JC,
Naranjo-Uibe JP, Jernandez-Lozano A:
Myopic keratomileusis in situ: A preliminary
report. J Cataract Refract Surg 1991;
17:424435.
34. Doane JF, Slade SG, Talamo JH, Azar DT:
Lamellar Refractive Surgery. Albert and
Jakobiec principles and practice of
ophthalmology. 2nd edn. Philadelphia:
WB Saunders; 2000.
35. Tenner A, Neuhann T, Schroeder E: Excimer
laser radial keratotomy in the living human
eye: a preliminary report. J Refract Surg
1988; 4:58.
36. Cotliar AM, Schubert HD, Manderl ER,
et al: Excimer laser radial keratotomy,
37. Marshall J, Trokel SL, Rothery S: Photo
ablative reproling of the cornea using an
excimer laser: photorefractive keratectomy.
Lasers Ophthalmol 1986; 1:2148.
38. Trokel SL: Evolution of excimer laser

corneal surgery. J Cataract Refract Surg
1989; 15:373383.
Central Photorefractive keratectomy for
myopia: The Blind Eye Study. Arch
Ophthalmol 1990; 108:799808.
40. Seiler T, Kahle G, Kriegerowski M: Excimer
laser (193 nm) myopic keratomileusis in
sighted and blind human eyes. Refract
Corneal Surg 1990; 6:165-173.
41. McDonald MB, Liu JC, Andreade H, et al:
Clinical results of 193 nm excimer laser
central photorefractive keratectomy for
myopia: the partially sighted and sighted
eye studies. Invest Ophthalmol Vis Sci
1990; 31(Suppl):245.
42. Taylor DM, LEsperance RA, Del Pero RA,
et al: Human excimer laser keratectomy:
a clinical study. Ophthalmology 1989;
96:654664.
43. Liu JC, McDonald MB, Vernell R, et al:
Myopic excimer laser photorefractive
keratectomy: an analysis of clinical
correlations. Refract Corneal Surg 1990;
6:3221328.
44. McDonald MB, Kaufman HE, Frantz JM,
et al: Excimer laser ablation in a human
45. Pallikaris IG, Papatzanaki ME, Stathi EZ,
et al: Laser in situ keratomileusis. Lasers
Surg Med 1990; 10:463468.
46. Pallikaris IG, Papatzanaki ME, Siganos DS,
et al: A corneal flap technique for laser in
situ keratomileusis. Human study. Arch
Ophthalmol 1991; 145:16991702.
47. Buratto L, Ferrari M: Excimer laser
intrastromal keratomileusis. Am J
Ophthalmol 1992; 113:291295.
48. Buratto L, Ferrari M: Excimer laser
intrastromal keratomileusis: Case reports.
49. Azar DT, Ang RT, Lee JB, et al: Laser
subepithelial keratomileusis: electron
microscopy and visual outcomes of flap
photorefractive keratectomy. Curr Opin
Ophthalmol 2001; 12:323328.
50. Cimberle M, Camellin M: LASEK technique
promising after 1 year of experience.
Ocular Surgery News 2000; 14:1417.
20052006:135.
52. Pinelli R, Jackson B, Agarwal A, Tamayo G:
Presby-LASIK: a corneal approach to
correct presbyopia. San Francisco:
ASCRS-ASOA; 2006.
53. Troutman RC: Microsurgery of the anterior
segment of the eye. St Louis: Mosby;
1977:263286.
54. Hoppenrijs VPT, Van Rij G, Beekhuis WH,
et al: Long-term results of corneal wedge
resections for the correction of high
astigmatism. Doc Ophthalmol 1990;
75:263273.
55. Waring GO III, Moftt SD, Gelender H, et al:
Rationale for and design of the National
Eye Institute prospective evaluation of
radial keratotomy (PERK) study.
Ophthalmology 1983, 90:4058.
56. Waring, et al: Results of the prospective
evaluation of radial keratotomy (PERK)
study one year after surgery.
Ophthalmology 1985; 92:177198, 307.
57. Francesconi CM, Azar DT, Talamo JH:
Incisional refractive surgery. Albert and
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68
69.
70.
71.
72.
73.
74.
75.
76.

Werblin TP, Stafforf GM: The Casebeer
system for predictable keratorefractive
surgery: one year evaluation of 205
consecutive eyes. Ophthalmology 1993;
10:10951102.
Berkeley RG, Sanders DR, Piccolo MG;
Effect of incision direction on radial
keratotomy outcome. J Cataract Refract
Surg 1991; 17:819823.
Waring GO III, Lynn MJ, McDonnell PJ:
radial keratotomy (PERK) study 10 years
after surgery. Arch Ophthalmol 1994;
112:12981308.
Azar T: Refractive surgery. Ophthalmology.
2nd edn. St Louis: Mosby;2004.
Lindstrom RL: The surgical correction of
astigmatism: a clinicians perspective.
Thornton SP: Astigmatic keratotomy: a
review of basic concepts with case reports.
Lindstrom RL, Lindquist TD: Surgical
correction of postoperative astigmatism.
Cornea 1988; 7:138148.
Park K, Lee JJ: Surgical correction of
astigmatism using paired T-incisions.
Korean J Ophthalmol 1989; 3:6164.
Thornton SP: Astigmatic keratotomy. In:
Thornton SP, ed. Radial and astigmatic
keratotomy. Thorofare, NJ:
Slack;1994:7388.
Binder PS, Waring GO III: Keratotomy for
astigmatism. In: Waring GO III, ed.
Refractive keratotomy for myopia and
astigmatism. St Louis: Mosby;
1992:11011119.
Doane JF, Slade SG, Talamo JH, Azar DT:
Lamellar refractive surgery. Albert and
Merlin U: Curve keratotomy procedure for
congenital astigmatism. J Refract Surg
1987; 3:9297.
Lindquist T, Rubenstein J, Rice S, et al:
Trapezoidal astigmatic keratotomy:
quantication in human cadaver eyes.
Lavery G, Lindstrom R: Clinical results
of trapezoidal astigmatic keratotomy.
J Refract Surg 1985; 1:7074.
Greene RB, Lindstrom RL: Astigmatic
keratotomy in the refractive patient: the
ARC-T study. In: Gills JP, ed. Surgical
treatment of astigmatism. Thorofare, NJ:
Slack; 1994:1126.
Price FW, Grene RB, Marks RG, et al:
Astigmatism reduction clinical trial: a
multicenter prospective evaluation of the
predictability of arcuate keratotomy. Arch
Ophthalmol 1995; 113:277282.
Ruiz LA: The astigmatic keratotomies
(Ruiz procedures). In: Boyd B, ed.
Refractive surgery with the masters. Vol 2.
Highlights of ophthalmology. FL: Coral
Gables; 1987:162193.
Greene RB, Lindstrom RL: Astigmatic
keratotomy in the refractive patient. The
ARC-T study. In: Gills JP, ed. Surgical
treatment of astigmatism. Thorofare, NJ:
Slack; 1994:1126.
Krachmer J, Ching S: Relaxing corneal
incisions for postkeratoplasty
astigmatism. Int Ophthalmol Clin 1983;
23:153159.
77. Lavery G, Lindstrom RL, Hofer L, et al:

Surgical management of corneal
astigmatism after penetrating keratoplasty.
78. Lavery GW, Lindstrom RL: Trapezoidal
astigmatic keratotomy in human cadaver
eyes. J Refract Surg 1985; 1:1824.
79. Akiyama K: Study of surgical treatment for
myopia. I. Posterior corneal incisions. Acta
Soc Ophthalmol Jpn 1952; 56:11421146.
80. Yamashita T, Schneider ME, Fuerst DJ,
et al: Hexagonal keratotomy reduces
hyperopia after radial keratotomy in rabbits.
J Refract Surg 1986; 2:261264.
81. Mendez A: Correcao da hipermetropia pela
ceratotomia hexagonal. In: Guimaraes R,
ed. Cirurgia refractiva. Rio de Janeiro,
Brazil: Piramide Livro Medico Editora;
1987:267269.
82. Jensen RP: Hexagonal keratotomy. Clinical
experience with 483 eyes. Int Ophthalmol
Clin 1991; 31:6973.
83. Basuk WL, Zisman M, Waring GO, et al:
Complications of hexagonal keratotomy.
84. Gilbert ML, Friedlander M, Granet N:
Corneal steepening in human eye bank
eyes by combined hexagonal and
transverse keratotomy. Refract Corneal
Surg 1990; 6:126130.
85. Gasset AR, Kaufman HE:
Thermokeratoplasty in the treatment of
79:226232.
86. Newmann AC, Fyodorov S, Sander DR:
Radial thermokeratoplasty for the
correction of hyperopia. Refract Corneal
Surg 1990; 6:404412.
87. Haw WW, Manche EE: Conductive
keratoplasty and laser thermal keratoplasty.
88. Department of Health and Human Services,
FDA Letter. Re: P990078 Sunrise
HYPERION LTK System. 3 Nov 1999.
89. McDonald MB, Durrie D, Asbell P, et al:
Treatment of Presbyopia with conductive
keratoplasty: six month results of the
1-year United States FDA clinical trial.
Cornea 2004; 23:661668.
90. Refractec website:
http://www.refractec.com/.
91. Olson RJ, Werner L, Mamalis N, Cionni R:
Perspectives-new intraocular lens
technology. Am J Ophthalmol 2005;
140:709716.
Academy of Ophthalmology;
20052006:164.
93. Mamalis N: Complication of foldable
intraocular lenses requiring explanation or
secondary intervention-2001 survey.
28:21932201.
94. Summary of safety and effectiveness data
for restor lens. Available: http://www.fda.gov/.
95. Young M: Cataract surgery innovations
head to head: the data on IOLs for
refractive lens exchange. Eye World 2006.
96. Payer H: Ringwulstlinse mit Zoomwirkung
zur Verstarkung einer Pseudoakkommodation
und deren Erklarung aus erweiterer
Akkomodationstheorie. Spektr Augenheilkd
1997; 11:8189.
97. Kuchle M, Langenbucher A, GusekSchneider GC, et al: First results of
implantation of a new potentially
accommodative posterior chamber
CHAPTER 68
History, Development, and Classification of Refractive Surgical Procedures
913
REFRACTIVE SURGERY
SECTION 7
intraocular lens. Kin Monatsbi Augenheilkd

2001; 218:603608.
98. Kuchle M, Langenbucher A, et al:
Comparison of 6-month results of
implantation of the 1CU accommodative
intraocular lens with conventional
intraocular lenses. Ophthalmology 2004;
111:318324.
99. Mcleod SD, Portney V, Ting A: A dual optic
accommodating foldable intraocular lens.
Br J Ophthalmol 2003; 87:10831085.
914
100. Dick HB: Accomodative intraocular lenses:

current status. Curr Opin Ophthalmol 2005;
16:826.
101. Maloney RK, Nguyen LH, John ME: Artisan
phakic intraocular lens for myopia: shortterm results of a prospective, multicenter
study. Ophthalmology 2002;
109:16311641.
102. Pop M, Payette Y: Initial results of
endothelial cell counts after Artisan lens for
phakic eyes: an evaluation of the United
States Food and Drug Administration

Ophtec Study. Ophthalmology 2004;
111:309317.
103. Guell JL, Vasquez M, Malacase F, et al:
Artisan Toric phakic intraocular lens for the
correction of high astigmatism. Am J
Ophthalmol 2003; 136:442447.
20052006:176.
CHAPTER
69
Optical Principles for Refractive Surgery

Pablo Artal
INTRODUCTION
Key Features: Optical Principles for Refractive Surgery
The main optical functions and parameters used to describe

the quality of the retinal image are introduced.
The optical aberrations in the normal eye depend on many
factors and conditions: individual to individual, pupil size, age,
accommodation, retinal eccentricity, etc.
In normal young subjects at the fovea, the average RMS of
higher order aberrations for a 5-mm pupil diameter is ~0.25 mm.
In most young eyes the aberrations are larger both in the
cornea and the lens than in the complete eye. This is due to an
active role of the lens to partially compensate for the
aberrations of the cornea.
Both corneal refractive surgery and the implantation of IOLs
may disrupt the naturally occurring balancing of aberrations
within the eye. Even in an ideal case where surgery would not
induce additional aberrations, the simple modication of one of
the components will cause in general a degradation of image
quality.
Refractive surgery involves a large range of procedures with the

general goal of improving the optical performance of a patients
eye. The success of each refractive procedure will depend on the
nal optical quality achieved after the surgery. Although refractive surgery is surely an art, the nal outcomes, and the continuous general improvements in the procedures, heavily rely on
a good understanding of the optical principles behind each treatment. Some important advances in this eld have been obtained
after a better understanding of the optical principles of the different techniques. In addition, the use of newly developed optical
technology is instrumental to the follow-up of different techniques
and the correct evaluation of nal outcomes.
During the last two decades, basic research on the optical
properties of the eye have provided us with a very clear and
complete picture of what the optical quality should be after
surgery. The best outcomes of any procedure will be those
mimicking the natural properties of a healthy young eye. In my
opinion, some current, and especially future surgical techniques,
will be based on attempts to reproduce the optics of normal eyes
to assure an excellent quality of vision.
Every modern refractive surgery technique considers more
than spherical refractive error or visual acuity. The best practitioners are now used to new concepts and ideas in ophthalmology, such as optical aberrations and quality of vision. These
are important new tools that will become common concepts in
the near future in this eld. In this context, I will cover in this
chapter, although briefly, some of the basic principles in
advanced optics that may be important for refractive surgery. In

addition, I will describe some of the properties of the optics in
the normal eye and several factors that affect it. Finally, some of
the optical changes that may appear after both corneal and
lenticular surgery will also be mentioned.
REVISION OF BASIC OPTICAL CONCEPTS

The treatment of the light as rays governed by the laws of the
geometrical optics was good enough for many situations in
ophthalmology for centuries. In fact, talking on distances of
objects, curvature, diopters, and the like has been nearly the
only optical knowledge for most clinicians in the past. However,
light is an electromagnetic wave with a wavelength that is not
negligible, so the wave character of the light and its consequences cannot be ignored in many conditions. In particular,
we need to consider wave optics (or physical optics) to correctly
describe and characterize the optical properties of the human
eye. Since a complete description of physical optics cannot be
included here, I refer the interested reader to some of the
numerous textbooks on basic optics (the book by Pedrotti and
Pedrotti1 is one of the possible choices for this level).
The image forming properties of any optical system and in
particular the eye can be described completely by the concept of
the wave-aberration. A previous denition must be introduced:
the wave front. It is dened as the surface with a constant optical path for a point object. This can be better understood with
some examples and noting that rays coming from a point object
are always normal (perpendicular) to the wave front. Figure 69.1a
shows a couple of simple examples: a plane wave front (in red)
perpendicular to the parallel rays (in blue) coming from a point
source at innite distance; and a spherical wave front produced
by a near object.
A perfect optical system, free of any aberration, has a
spherical wave front and forms perfect images. Every ray,
entering the pupil at any location, reaches the same point at the
focal plane. In this case, the image of a point will be only
affected by the effect of diffraction in the pupil (which cannot be
avoided since is due to the wave-like nature of the light). For a
circular aperture, this is known as the Airy-disk the site of
which depends on the pupil size, with large Airy disks occurring
with smaller pupils (Fig. 69.1b). However, in the particular case
of the human eye, imperfections in the refracting ocular surfaces
generate aberrations which produce a larger, and in general,
asymmetric retinal image. The images of a point source at the
retina are called point-spread function (PSF). Small, symmetric
and compact PSFs occur in good-quality eyes, while extended and
asymmetric PSFs are common in eyes with poor optical quality.
The wave-aberration is dened as the difference between the
perfect (spherical) and the real wave fronts for every point over
915
REFRACTIVE SURGERY
Perfect optical system

(without aberrations)
(point object at infinity)
Spherical wave-front
(near point object)
REALoptical system
(with aberrations)
retinal image
(PSF)
FIGURE 69.1. Schematic representation of some optical concepts: rays and wave fronts. (a) Parallel rays (in blue) from a point at innite
distance and associated wave front (red line), and rays originated in a near point source with spherical wave fronts. (b) Example of image
formation in a perfect (without aberrations) and in an aberrated optical system.
WAVE
ABERRATION
real
wave-front
retinal image
(PSF)
perfect wave-front
(sphere)
a
SECTION 7
FIGURE 69.2. (a) Schematic representation of the denition of wave-aberration (see text for details). (b) Example of the contribution of the
wave-aberration to the retinal image quality.
916
the eyes pupil (Fig. 69.2a). It is typically represented as a twodimensional map, where each gray, or color, level represents the
amount of wave-aberration, expressed either in microns or
number of wavelengths. An error smaller than l/4 (a quarter of
the wavelength) is considered to be insignicant. Although the
wave-aberration can be a quite complicated two-dimensional
function, it can be broken down through polynomials in a sum
of pure aberration terms (called modes). The lower degree terms
correspond to the well-known: defocus and astigmatism, while
the following terms are the higher order aberrations: coma,
spherical aberration, trefoil, among others. A convenient, and
now common, way to break down the wave-aberration is the
Zernike polynomial expansion.2 However, it is important to
consider that a small number of aberration modes do not
necessarily represent all the optical imperfections in the eye.
Figure 69.2b shows schematically the actual situation. A closer

look at the wave front aberration would shows higher spatial
frequency details, related with very high order aberrations and
scattered light. The effect of these small-scale details on the
retinal image is to produce structured details around the center
of the retinal image and/or a halo. The low to mid order
aberrations typically produce a more extended central part of
the retinal image. It is important to note, that what actually
affects the quality of vision is the complete retinal image.
If the wave-aberration is known for a particular eye, the PSF
can be computed for any desired condition of pupil size or
defocus. This is very convenient and useful to predict the optical performance under different conditions. From the PSF, a
useful single optical quality parameter, the Strehl ratio, is often
obtained as the quotient between the intensity peak in the eyes
PSF and the aberration-free (diffraction-limited) PSF. This

optical parameter is well correlated for some conditions with
visual performance.3 Another interesting feature is that by
performing a convolution operation it is possible to simulate
the retinal images of any test from knowledge of the waveaberration or the PSF. Figure 69.3 shows an example of the
retinal image of a letter chart. This type of representation can
be of use for the direct comparison of different clinical outputs,
for instance to predict visual acuity that can be compared with
the clinical data. Another related important function is the
modulation transfer function (MTF). It can be computed as the
modulus of the Fourier transformation of the PSF and carries
out information on how the optical system of the eye transmits
the contrast of gratings of different spatial frequencies. It is
conceptually the optical counterpart to the visual contrast
sensitivity function (CSF).
relative importance of higher order aberrations in normal eyes,

0.25 mm of defocus would be approximately equivalent to
0.25 D for that pupil size. Although this is very large in the
context of technical optics, it is modest in ophthalmic optics.
Beyond defocus and astigmatism, spherical aberration, coma
and trefoil are the most signicant aberrations in normal eyes.
It should be noted that in eyes with some pathologies high order
aberrations are much higher (for the same 5-mm pupil, values
of ~1 mm are not uncommon).
Despite the large individual variability it is interesting to
know the statistical properties of the eyes aberrations. Several
recent systematic measurements of ocular aberrations in
relatively large healthy populations addressed this issue.1315
These studies showed mirror symmetry between the
aberrations of left and right eyes of the same individual and the
average magnitude of the aberrations decreasing with increasing
Zernike order. When the Zernike coefcients were averaged
preserving their sign within the population, most of the mean
values were approximately zero, except for defocus, astigmatism,
and spherical aberration (with typically small positive values on
average).
Beyond the monochromatic aberrations, chromatic aberrations in optical systems arise from the dependence of refractive
index on wavelength. Chromatic aberrations are traditionally
divided into longitudinal chromatic aberration (LCA) and transverse chromatic aberration (TCA). The former is the variation
of axial power with wavelength while the latter is the shift of the
image across the image plane with wavelength. Both LCA and
TCA are widely studied issues,16,17 in the eye and may limit the
actual retinal image quality since the real world is usually
polychromatic and, therefore, its image becomes distorted in the
retina in a color-dependent fashion. The impact of different
refractive procedures on the eyes chromatic aberration, though
not yet extensively studied, may be of interest in the future.
MEASURING THE ABERRATIONS OF

THE EYE
SOURCES OF ABERRATIONS WITHIN

THE EYE
The wave-aberration of the eye can be measured using a variety

of different subjective and objective techniques. A noncomplete
list of different methods includes: the vernier alignment technique,4 the crossed cylinder aberroscope,5 estimates from
double-pass retinal images,6 the pyramid sensor,7 and the most
popular of the used methods today, the HartmannShack wave
front sensor.810 This consists of a microlenses array, conjugated
with the eyes pupil, and a camera placed at its focal plane. If a
plane wave front reaches the microlenses array, a perfectly
regular mosaic of spots is recorded. However, if an aberrated
wave front reaches the sensor, the pattern of spots is irregular.
The displacement of each spot is proportional to the derivative
of the wave front over each microlens area. These wave front
sensors provide useful information on the optical quality of the
eye. However, in eyes where very high order aberrations and/or
scattered light are prominent, they may overestimate retinal
image quality. In those cases, the direct recording of the retinal
image using the double-pass technique can provide complete
information on the eyes optical quality.11,12
A relevant question to refractive surgery is: where are the

sources of the aberrations present in the eye? This can be
answered by simultaneously measuring the aberrations induced
by the anterior surface of the cornea and the total ocular aberrations in the same eye. Then, the aberrations of the internal
ocular optics, i.e., those produced by the posterior corneal surface
and the lens, can be determined. This allows one to determine
the relative contributions of the different optical elements of the
eye to the total wave front. Refractive techniques will need and
will use such a detailed topographic structure of the eyes
aberrations in the future.
The aberrations associated with the anterior surface of the
cornea can be computed from its shape as measured with
corneal topographers. The simplest approach to calculate the
anterior corneal aberrations is to obtain a remainder lens by
subtracting the best conic surface t to the measured cornea,
and calculating the aberrations by multiplying the residual
surface prole by the refractive index difference between air and
the cornea. Another option providing better results is to trace
rays through the corneal surface to compute the associated
aberrations.18
From the wave aberrations of both the complete eye and the
cornea, the relative contributions of the different ocular surfaces
to retinal image quality can be evaluated. In particular, the
wave-aberration of the internal ocular optics, with the
crystalline lens as the main contributor, is estimated simply by
directly subtracting the corneal from the ocular aberrations. In
a simple model, the aberrations of the lens can be obtained by
direct subtraction of the aberrations of the cornea and the eye.
PSF
=
wave-aberration
FIGURE 69.3. From the wave-aberration, the PSF can be computed
to produce retinal image of any object (in this example a letter chart).
ABERRATIONS IN THE NORMAL EYE

The optical aberrations in the normal eye depend on many
factors and conditions. They vary from individual to individual,
with pupil size, the age of the subject, accommodation and retinal
eccentricity, among other factors. In normal young subjects at
the fovea, the average root mean square wave front error, or
RMS, of higher order aberrations for a 5-mm pupil diameter is
~0.25 mm. To provide the reader with an intuitive idea of the
CHAPTER 69
917
REFRACTIVE SURGERY
cornea
lens
eye
FIGURE 69.4. Example of wave-aberrations for the cornea, the lens

and the complete eye in one normal young subject. The associated
PSFs were calculated at the best image plane from the waveaberrations and subtend 20 min of arc of visual eld. The aberrations
of the internal optics compensate in part for the corneal aberrations.
We need to assume that the wave aberration remains approximately constant for different axial planes, i.e., from the corneal
vertex to the pupil plane. By applying this procedure, the
relative contribution of the aberrations of the cornea and the
lens in different eyes has been studied. Figure 69.4 shows, as an
example, the wave-aberrations and the associated PSFs for the
cornea, the lens, and the eye in a normal young eye. It can be
noticed that the magnitude of aberrations is larger both in the
cornea and the lens than in the eye. This is due to an active role
of the lens to partially compensate for the aberrations of the
cornea.19 It is remarkable that the magnitude of several aberration terms is similar for the two components, but they tend
to have opposite signs. This indicates that the internal optics
play a signicant role in compensating for the corneal aberrations in normal young eyes. It was recently shown that this
compensation is larger in the less optically centered eyes that
mostly correspond to hyperopic eyes.20 This suggested that the
distribution of aberrations between the cornea and lens allows
the optical properties of the eye to be relatively insensitive to
variations arising from eye growth or exact centration and alignment of the eyes optics relative to the fovea. This type of autocompensating mechanism renders the eyes optics robust
despite large variations in the ocular shape and geometry and
may have some potential implications in refractive surgery.
SECTION 7
TEMPORAL CHANGES OF THE EYES

ABERRATIONS
918
The aberrations of the eye change over time due to a variety of

factors. Accommodation,21 eye movements, changes in the tear
lm and humors produce rapid, although in general small,
changes of the aberrations.22 Due to these continuous changes
of the aberrations over time, an ideal, perfectly static correction
will not provide stable, aberration-free optics. For example,
when an eye that is perfectly corrected for distance vision
accommodates to near objects, the aberrations will change and
this eye will no longer be aberration-free. This indicates that,
due to the dynamic nature of the ocular optics, a static, perfect
correction (as attempted in customized refractive surgery)
would not remain perfect for every condition occurring during
normal accommodation.
Moreover on a much longer time scale, normal aging affects
different aspects of the ocular optics. Elderly eyes typically
experience increased light absorption by the ocular media,
smaller pupil diameters, increases of intraocular scatter and
nearly a complete reduction of their accommodative capabilities. In addition, we rst showed that the average MTF of
the eyes in a group of older subjects was lower than the average
MTF for a group of younger subjects.23 More recent measurements in a larger population show a nearly linear decline of
retinal image quality with age,24 suggesting a signicant increase in the optical aberrations of the eye with age, in agreement with other studies.25 Different factors contribute to the
increment of aberrations with age, such as changes in the
aberrations of the cornea26 and the lens or their relative contributions. During normal aging, the relatively small corneal
changes cannot account for the degradation in the retinal image
quality. However, the lens dramatically changes both its shape
and effective refractive index with age, leading to changes in its
aberrations.27 As the aberrations of the lens change with age,
the compensation of corneal and lenticular aberrations found in
young subjects is in part, or even completely, lost.28 This
explains the overall increase in aberration and the reduction of
retinal image quality throughout the life span.
REFRACTIVE SURGERY AND EYES

OPTICS
Refractive surgery procedures signicantly modify the optics of
the eye. The success of these procedures depends on the
resulting balance of the aberration of the normal eye. Both
standard LASIK and the implantation of intraocular lenses
(IOLs) tend to increase aberrations degrading the optical quality
of the eye.29,30 Moreover, the location of the aberrations within
the eye has important implications for current clinical procedures, such as wave front-guided refractive surgery and cataract
surgery. For example, customized ablation needs to be
performed based on the aberrations of the complete eye. If the
ablation is based on only the corneal aberrations, the nal
aberrations of the eye could be larger than before the ablation.
If an ideal ablation is performed, the eye becomes limited only
by diffraction. However, if the same ablation is performed,
correcting only the corneal aberrations, the remaining eye will
still have aberrations from internal structures that, in many
cases, can be more severe in the eye than before the treatment.
Another important example is cataract surgery with
implantation of IOLs. These IOLs usually have good image
quality when measured on an optical bench, but the nal
optical performance post-operatively was typically lower than
expected.30 The reason is that the ideal substitute for the
natural lens is not a lens with the best optical performance
when isolated, but one that is designed to compensate for the
aberrations of the cornea.31 IOLs should ideally be designed
with an aberration prole matching that of the cornea to maximize the quality of the retinal image. Current aspheric IOLs
address to some extent this problem, by partially correcting the
spherical aberration of the cornea. Future designs would
perhaps incorporate other aberrations, in particular coma.
Refractive surgery may disrupt the natural aberration compensation
Corneal refractive surgery
Cataract surgery
FIGURE 69.5. Optical changes induced in the cornea or the lens can
disrupt the natural compensation of aberration present in the young
eye. The goal of refractive surgery could be to preserve the natural
aberrations of normal eyes.

Figure 69.5 shows schematically how both corneal refractive
surgery and the implantation of IOLs may disrupt the naturally
occurring balance of aberrations within the eye. Even in an ideal
case where surgery would not induce additional aberrations, the
simple modication of one of the components will cause in
general a degradation of image quality. This will be important
for future advances in refractive surgery techniques. It has to be
well understood that the optics of the eye are nely tuned, and
changing only one part may severely affect the overall optical
performance. Surgeons should also keep in mind that poorly
balanced optics will result in low quality of vision. I personally

hope that advanced optical concepts will help refractive
surgeons better their outcomes, thus providing patients with
even improved vision.
ACKNOWLEDGMENTS
Part of the research described in this chapter has been supported by the
Ministerio de Educacin y Ciencia (MEC), Spain and by AMO Groningen
(The Netherlands).
REFERENCES
12. Daz-Doutn F, Benito A, Pujol J, et al:
Comparison of the retinal image quality
with a HartmannShack wavefront sensor
and a double-pass instrument. Invest
13. Porter J, Guirao A, Cox IG, Williams DR:
Monochromatic aberrations of the human
eye in a large population. J Opt Soc Am A
2001; 18:17931803.
14 Thibos LN, Hong X, Bradley A, Cheng X:
Statistical variation of aberration structure
and image quality in a normal population of
healthy eyes. J Opt Soc Am A 2002;
19:23292348.
15. Castejon-Mochon JF, Lopez N, Benito A,
Artal P: Ocular wave-front aberration
statistics in a normal young population.
Vision Res 2002; 42:16111617.
16. Wald G, Grifn DR: The change in
refractive power of the human eye in dim
and bright light. J Opt Soc Am 1947;
37:321366.
17. Marcos S, Burns SA, Moreno-Barriuso E,
Navarro R: A new approach to the study of
ocular chromatic aberrations. Vision Res
1999; 39:43094323.
18. Guirao A, Artal P: Corneal wave aberration
from videokeratography: accuracy and
limitations of the procedure. J Opt Soc Am
A 2000; 17:955965.
19. Artal P, Guirao A, Berrio E, Williams DR:
Compensation of corneal aberrations by
the internal optics in the human eye. J Vis
2001; 1:18.
20. Artal P, Benito A, Tabernero J: The human
eye is an example of robust optical design.
J Vis 2006; 6:17.
21. Artal P, Fernndez EJ, Manzanera S:
Are optical aberrations during
accommodation a signicant problem for
refractive surgery? J Refract Surg 2002;
18:S563S566.
22. Hofer HJ, Artal P, Singer B, et al: Dynamics
of the eyes wave aberration. J Opt Soc Am
A 2001; 18:497506.
23. Artal P, Ferro M, Miranda I, Navarro R:

Effects of aging in retinal image quality.
J Opt Soc Am A 1993; 10:16561662.
24. Guirao A, Gonzalez C, Redondo M, et al:
Average optical performance of the human
eye as a function of age in a normal
population. Invest Ophthalmol Vis Sci 1999;
40:203213.
25. Guirao A, Redondo M, Artal P: Optical
aberrations of the human cornea as a
function of age. J Opt Soc Am A 2000;
17:16971702.
26. McLellan JS, Marcos S, Burns SA: Agerelated changed in monochromatic wave
aberrations of the human eye. Invest
27. Glasser A, Campbell MCW: Presbyopia
and the optical changes in the human
crystalline lens with age. Vision Res 1998;
38:209229.
28. Artal P, Berrio E, Guirao A, Piers P:
Contribution of the cornea and internal
surfaces to the change of ocular
aberrations with age. J Opt Soc Am A
2002; 19:137143.
29. Moreno-Barriuso E, Lloves JM, Marcos S,
et al: Ocular aberrations before and after
myopic corneal refractive surgery: LASIKinduced changes measured with laser ray
tracing. Invest Ophthalmol Vis Sci 2001;
42:13961403.
30. Artal P, Marcos S, Navarro R, Miranda I,
Ferro M: Through focus image quality of
eyes implanted with monofocal and
multifocal intraocular lenses. Opt Eng 1995;
34:772779.
31. Guirao A, Redondo M, Geraghty E, et al:
Corneal optical aberrations and retinal
image quality in patients in whom
monofocal intraocular lenses were
implanted. Arch Ophthalmol 2002;
120:11431151.
CHAPTER 69
1. Pedrotti FL, Pedrotti LS: Introduction to

optics. 2nd edn. NJ: Prentice Hall; 1993.
2. Thibos LN, Applegate RA, Schwiegerling JT,
et al: Standards for reporting the optical
aberrations of eyes. In: Vasudevan L, ed.
OSA trends in optics and photonics, vision
science and its applications. Washington,
DC: Optical Society of America;
2000:232244.
3. Villegas EA, Gonzlez C, Bourdoncle B, et al:
Correlation between optical and
psychophysical parameters as a function of
defocus. Optom Vis Sci 2002; 79:6067.
4. Smirnov MS: Measurement of the wave
aberration of the human eye. Biozika
1961; 6:776795.
5. Howland HC, Howland B: A subjective
method for the measurement of
monochromatic aberrations of the eye.
J Opt Soc Am 1977; 67:15081518.
6. Iglesias I, Berrio E, Artal P: Estimates of the
ocular wave aberration from pairs of
double-pass retinal images. J Opt Soc Am
A 1998; 15:24662476.
7. Iglesias I, Ragazzoni R, Julien Y, Artal P:
Extended source pyramid wave-front
sensor for the human eye. Opt Express
2002; 10:419428.
8. Liang J, Grimm B, Goelz S, Bille JF:
Objective measurement of the WAs
aberration of the human eye with the use of
a HartmannShack sensor. J Opt Soc Am
A 1994; 11:19491957.
9. Liang J, Williams DR: Aberrations and
retinal image quality of the normal human
eye. J Opt Soc Am A 1997; 14:28732883.
10. Prieto PM, Vargas-Martn F, Goelz S, Artal P:
Analysis of the performance of the
HartmannShack sensor in the human eye.
J Opt Soc Am A 2000; 17:13881398.
11. Santamara J, Artal P, Bescs J:
Determination of the point-spread function
of the human eye using a hybrid opticaldigital method. J Opt Soc Am A 1987;
6:11091114.
919
CHAPTER
70
Corneal Topography and Wave Front Analysis

Damien Gatinel
CORNEAL TOPOGRAPHY
INTRODUCTION
Topography is a general term in geography, derived from the
Greek topos (place) and graphein (to draw). Corneal topography corresponds to the graphic representation of the geometrical properties of corneal surface. The cornea is a unique
organ whose function is tightly dependent on its shape; a
variation of the micron scale order can signicantly alter its
optical properties. Thus, measuring the shape, refractive power,
and the thickness of the cornea are crucial steps in diagnosing
corneal diseases and designing methods of vision correction.
Most routine ophthalmologic practices use a keratometer to
assess central corneal curvature for contact lens (CL) tting or
intraocular lens (IOL) power calculation prior to cataract
surgery. The evolution of refractive surgery procedures has even
accentuated the need for accurate analysis of the entire anterior
corneal surface. In the evaluation of patients for refractive
surgery, a keratometer is inadequate, since it provides local
measurements from the central 34 mm anterior corneal surface only. Over the past 20 years, the rapid development of
corneal topography was parallel with that of excimer laser
refractive surgery, and computerized videokeratography has
made topographic mapping of the power and shape of the
cornea a routine aspect of clinical practice.
Instruments that measure and describe the corneal surface
can be divided into two gen

Jakobiec 039 S Principles Amp Practice of Ophthalmology Volume One

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Jakobiec 039 S Principles Amp Practice of Ophthalmology Volume One

Uploaded by

Copyright:

Available Formats

Albert: Albert & Jakobiec's Principles &

Preface to the 3rd Edition

medicine, and genetics information has been incorporated into

Preface to the 1st Edition

with an anatomic woodcut (Fig. 2) and then lists in tabular

Preface to the 1st Edition

Handbuch. The nal volume of this great collective work, of

Preface to the 1st Edition

textbook or system previously published, in terms of quantity

Lloyd P Aiello MD PhD

Jorge L Ali MD PhD

Hassan Alizadeh PhD

Isabelle Audo MD PhD

Robert Bhisitkul MD PhD

Donald L Budenz MD MPH

Ravinder D Bhui BApSc in Elec Eng

Michael E Boulton PhD

Alan C Bird MD FRCS FRCOphth

Norman Paul Blair MD

Angela N Buffenn MD MPH

Reza Dana MD MSc MPH

M Ronan Conlon MD FRCSC

Joseph L Demer MD PhD

Associate Professor of Ophthalmology

Helen B Danesh-Meyer MBChB MD

Marshall G Doane PhD

Christopher Dodds MBBS MRCGP FRCA

Elizabeth A Davis MD FACS

Claes H Dohlman MD PhD

Frederick L Ferris III MD

Deborah I Friedman MD FAAN

Gary N Foulks MD FACS

Vamsi K Gullapalli MD PhD

Daniel G Green PhD

Franz Grehn Dr h.c.

Haiyan Gong MD MS PhD

Jack V Greiner DO PhD

Craig M Greven MD FACS

David B Granet MD FACS FAAP FAAO

Yasutaka Hayashida MD PhD

J Fielding Hejtmancik MD PhD

Bonnie A Henderson MD FACS

Peter G Hovland MD PhD

Peter S Hersh MD FACS

Harold G Jensen PhD

Douglas H Johnson MD (deceased)

Alon Kahana MD PhD

Tae-Im Kim MD PhD

Stephen R Klapper MD FACS

Barbara E K Klein MD MPH

Robert E Marc PhD

Jay Neitz PhD

Marja Mogk PhD

Maureen Neitz PhD

Peter A Netland MD PhD

Yen Hoong Ooi MD

Consultant Ophthalmic Surgeon

Michael A Sandberg PhD

Allan R Rutzen MD FACS

Alfredo A Sadun MD PhD

Malgorzata Rozanowska PhD

Peter A D Rubin MD FACS

Sarwat Salim MD FACS