Journal of Infection (2008) 56, 1e12

www.elsevierhealth.com/journals/jinf

REVIEW

Legionella spp. and Legionnaires disease

B.M.W. Diederen*

Regional Laboratory of Public Health Haarlem, Boerhaavelaan 26, 2035 RC Haarlem, The Netherlands

Accepted 19 September 2007

Available online 5 November 2007

KEYWORDS Summary Infection with Legionella spp. is an important cause of community- and hospital-
Legionella; acquired pneumonia, occurring both sporadically and in outbreaks. Infection with Legionella
Legionella spp. ranks among the three most common causes of severe pneumonia in the community set-
pneumophila; ting, and is isolated in 1e40% of cases of hospital-acquired pneumonia. There are no clinical
Diagnosis; features unique to Legionnaires disease. Macrolides and fluoroquinolones are the most widely
Review; used drugs in treatment. The availability of a good diagnostic repertoire, suitable for accu-
Treatment; rately diagnosing LD, constitutes the basis for the early recognition and treatment of the indi-
Legionnaires Disease vidual patient as well as for effective measures for prevention and control. This review
summarizes the available information regarding the microbiology, clinical presentation, diag-
nosis and treatment of LD, with an emphasis on the laboratory diagnosis of infection with
Legionella spp.
2007 The British Infection Society. Published by Elsevier Ltd. All rights reserved.

Introduction Prevention, Joseph McDade and Charles Shepard, an-

nounced that they had discovered the etiologic agent, a fas-
tidious Gram-negative bacillus.2 Because of the historical
An explosive, common-source outbreak of pneumonia
association with the American Legion convention, this dis-
caused by a previously unrecognized bacterium affected
ease was called Legionnaires disease (LD) and the etiologic
primarily persons attending an American Legion convention
agent Legionella pneumophila. L. pneumophila belongs to
in Philadelphia in July, 1976. Thirty-four of 221 cases were
the family Legionellaceae. Use of an antibody test for the
fatal. Despite intensive laboratory investigation, the cause
disease soon learned that several prior unsolved outbreaks
of the outbreak was undetected for months. An epidemi-
of pneumonia had been LD, including outbreaks in the 1950s
ologic investigation determined that the disease most likely
and 1960s. In addition, members of the genus had been iso-
was airborne and focused primarily at one convention
lated some 25 years earlier from sporadic cases of pneumo-
hotel, which subsequently had to be closed because of
nia.3 An unsolved outbreak of a non-pneumonic febrile
adverse publicity.1 About 6 months later, two investigators
illness was also found to have resulted from exposure to
at the United States Centers for Disease Control and
Legionella bacteria; this illness was called Pontiac fever,
after Pontiac, Michigan, where this had occurred.4 As
* Tel.: 31 23 530 7800; fax: 31 23 530 7805. with LD, prior epidemics of Pontiac fever had occurred as
E-mail address: bramdiederen@gmail.com early as 1949 without solved etiology.

0163-4453/$30 2007 The British Infection Society. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.jinf.2007.09.010
2 B.M.W. Diederen

There are three lessons from the Philadelphia outbreak.

First and foremost was the dismissal of the idea that Table 1 Currently known Legionella species
infectious diseases and their causes are a thing of the L. adelaidensis
past or even that all infectious diseases are known. Second, L. anisa
LD occurs as a consequence of altering the environment for L. beliardensis
human benefit. Cases of LD have been traced to a wide L. birminghamensis
variety of man-made water sources, including cooling L. bozemanae
towers, whirlpools and spas, fountains, ice machines and L. bozemanii
vegetable misters. A third lesson is how difficult it can be to L. brunensis
break out of traditional patterns of thought when con- L. busanensis
fronted with the unknown. In Philadelphia, initially, viral L. cherrii
and toxic etiologies were sought, because of the clinical L. cincinnatiensis
resemblance of the pneumonia to severe influenza. When
L. drancourtii
bacteria that had the morphology of Gram-negative bacilli
L. drozanskii
were subsequently demonstrated with electron microscopy
L. dumoffii
in lung tissues of fatal LD cases, the significance of the
observation was obscured by the consideration that they L. erythra
were merely secondary invaders.3 L. fairfieldensis
This review summarizes the available information L. fallonii
regarding the microbiology, clinical presentation, diagnosis L. feeleii
and treatment of LD, with an emphasis on the laboratory L. geestiana
diagnosis of infection with Legionella spp. L. gormanii
L. gratiana
L. gresilensis
Microbiology
L. hackeliae
L. israelensis
The family Legionellaceae consists of the single genus
L. impletisoli
Legionella. Legionella are Gram-negative coccobacilli
that measure 0.3e0.9 mm in width and 2e20 mm in length. L. jamestowniensis
L. jordanis
In tissue and clinical specimens, the organisms are cocco-
bacillary, measuring 1e2 mm. Elongated filamentous forms L. lansingensis
may be seen after growth on some culture media. Soluble L. londiniensis
iron and l-cysteine are required for optimal growth and L. longbeachae
for the initial isolation of the bacterium from both clinical L. lytica
and environmental sources. Iron, l-cysteine, a-ketogluta- L. maceachernii
rate, and charcoal-containing yeast extract agar buffered L. micdadei
with an organic buffer (BCYEa agar) is the preferred growth L. moravica
medium for clinical isolation. To support bacterial growth, L. nautarum
the pH of the agar is critical and should be adjusted to pH
L. oakridgensis
6.9 by adding N-2-acetamino-2-aminoethansulfonic acid
(ACES).5 Proteins rather than carbohydrates are used as L. parisiensis
L. pittsburghensis
an energy source. Legionnellae are obligate aerobes, and
L. pneumophila
grow at temperatures ranging from 20  to 42  C. Clinically
important Legionella species grow best at 35  C in humidi- L. quateirensis
fied air on BCYEa medium, usually in 2e5 days after inocu- L. quinlivanii
lation of plates. An incubation of up to 10 days may very L. rowbothamii
rarely be required for the isolation of unusual Legionella L. rubrilucens
species. L. sainthelensi
The number of recognized species and serogroups of the L. santicrucis
genus Legionella continues to increase (Table 1). There are L. shakespearei
currently more than 50 known species. L. pneumophila L. spiritensis
comprises at least 16 different serogroups.6 Some legionel- L. steigerwaltii
lae cannot grow on routine Legionella-specific media and L. taurinensis
have been called Legionella-like amoebal pathogens L. tucsonensis
(LLAPs). These organisms have been isolated and main- L. wadsworthii
tained by cocultivating the bacteria with their protozoan L. waltersii
hosts. One LLAP strain was isolated from the sputum L. worsleiensis
of a pneumonia patient by enrichment in amoebae and is L. yabuuchiae
considered an extremely rare human pathogen.7
Source: German Collection of Microorganisms and Cell Cultures.
Legionella spp. are ubiquitous. They are found in natural http://www.dsmz.de/microorganisms/bacterial_nomenclature_
aquatic environments (streams, rivers, ponds, lakes and info.php?genusZLEGIONELLA.
thermal pools) in moist soil and in mud. They have even
Legionella ssp. and LD
been found in the canopy of the rain forest. The organisms
are able to survive in moist environments for long periods of
time and can withstand temperatures of 0e68  C and a pH
range of 5.0e8.5 They can survive chlorination and thus
enter water supply systems and proliferate in thermal
habitats, including air-conditioning cooling towers, hot wa-
ter systems, shower heads, taps, whirlpool spas and respi-
ratory ventilators. The organisms are found in biofilms on
the surfaces of these systems, where they are far less sus-
ceptible to the effects of biocides and chlorine.
Most cases of legionellosis can be traced to man-made
aquatic environments where the water temperature is
higher than ambient temperature. The growth of Legionella
spp. is aided by co-existing micro-organisms, which provide
nutrients, and free-living amebae, in which the Legionella
spp. can reside and multiply. The presence of the bacteria
in an aquatic environment and warm water temperature
are two factors that can increase the risk of LD. The third
component is the presence of nutritional factors that allow
the bacteria to amplify. Legionellae survive in aquatic and
moist soil environments as intracellular parasites of free-
living protozoa.8 Thermally altered aquatic environments
can shift the balance between protozoa and bacteria, re-
sulting in rapid multiplication of legionellae. We should
be aware that Legionella is a common colonizer of water
distribution systems, similar to other potentially patho-
genic bacteria and fungi. Although multiple strains may col-
onize water-distribution systems, only a few specific
species and types will cause disease in patients exposed
to the water.
LD is a major concern of public health professionals and
individuals involved with the construction or maintenance
of water systems, such as air-conditioning systems, circu-
lating water systems and cooling towers. Legionellosis is
generally considered a preventable illness because control-
ling or eliminating the bacterium in certain reservoirs will
(in theory) prevent disease. The linkage between specific
levels of colonization and risk of LD remains controversial.9
The concept of a preventable illness has resulted in a num-
ber of guidelines and control strategies aimed at reducing
the risk of legionellosis in building water systems. The fac-
tors that lead to outbreaks or cases of LD are not com-
pletely understood, but certain events are considered
prerequisites for infection. These include the presence of
virulent bacteria in an aquatic environment, amplification
of the bacterium to an unknown infectious dose, and trans-
mission of the bacteria via aerosol to a human host that is
susceptible to infection.
L. pneumophila serogroup 1 caused the 1976 Philadel-
phia outbreak and is the cause of approximately 90% of
all cases of LD from which a bacterial strain was isola-
ted.10e12 However, although L. pneumophila serogroup 1 ac-
counts for the majority of American and European
Legionella isolates, in Australia and New Zealand, L. pneu-
mophila serogroup 1 accounts for only approximately 50%
of cases of community-acquired legionellosis, while L. long-
beachae accounts for approximately 30% of cases.11
L. pneumophila serogroup 1 can be further divided into
multiple subtypes using a variety of serologic, other pheno-
typic, and genetic methods. One particular subtype of
L. pneumophila serogroup 1 causes 67%e90% of cases of
LD, and 85% of cases due to L. pneumophila serogroup 1;
3
this subtype is distinguished by its reactivity with a particu-
lar monoclonal antibody, and it is variously termed the Pon-
tiac, the Joly monoclonal type 2 (MAb2), or the Dresden
monoclonal type 3/1 (MAb 3/1) subtype.13
The main reason for genotyping Legionella pneumophila
is to help identify environmental sources giving rise to cases
of LD, thus allowing control measures to be implemented and
further cases prevented. As L. pneumophila is quite common
in the environment, a wide range of methods have been de-
veloped in an attempt to differentiate between strains. The
suitability of the various methods available depends, in large
part, on the context within which they are to be applied.14
These include ribotyping, amplified fragment length poly-
morphism (AFLP) analysis, pulsed-field gel electrophoresis
(PFGE), restriction fragment length polymorphism (RFLP)
analysis, restriction endonuclease analysis (REA), multi locus
sequence typing (MLST) and arbitrarily primed PCR.14,15 One
of these methods, a single-endonuclease, amplified frag-
ment length polymorphism analysis method by which the
patterns are resolved by standard agarose electrophoresis,
was adopted as an international standard and is now widely
used by members of the European Working Group for Legion-
ella Infections (EWGLI).14,15 Recently, a scheme for the
sequence-based typing (SBT) of L. pneumophila that uses
the sequences of six genes was described.16 The use of
neuA as a seventh allele for typing significantly increased
the index of discrimination.17 This modification to the stan-
dard method is proposed as the method of choice in the ep-
idemiological investigation of L. pneumophila infections.17 It
is now available through the website of the European Work-
ing Group on Legionella Infections (EWGLI) (http://www.
ewgli.org). Compared to DNA fragment-based techniques
DNA sequencing offers much greater reproducibility. It
should be noted, however, that none of these typing methods
are entirely perfect: they sometimes fail in correctly identi-
fying the environmental source of disease.18 SBT does not yet
offer a flawless solution to problems of reproducibility and
standardisation in Legionella pneumophila genotyping. In
an international external quality assessment (EQA) program
to assess the performance of laboratories in genotyping Le-
gionella pneumophila isolates using the standard EWGLI
SBT protocol, the number of centres achieving 100% score,
for all loci tested, rose successively from 50% for the first
EQA distribution, to 56% for the second EQA distribution, to
76% for the third EQA distribution.19
Clinical spectrum
Legionellosis classically presents as two distinct clinical
entities, LD, a pneumonia with severe multisystem dis-
ease,2 and Pontiac fever, a self-limited flu-like illness.4 In
addition, many persons who are infected with legionellae,
as proven by seroconversion, will remain asymptomatic.20
LD is transmitted from the environment to man by inhala-
tion of an infectious aerosol.21 In an unknown fraction of
cases, microaspiration of contaminated water into the
lungs could be the mode of nosocomial transmission.22 Mul-
tiple examples of exclusive aerosol transmission of LD ex-
ist, especially in epidemics where a cooling tower, water
spa, water fountain, or water mister are the source of
disease.23
4 B.M.W. Diederen
There is no agreed-upon definition of Pontiac fever.24e26
The diagnosis is usually made on the basis of epidemiologic,
clinical, clinical laboratory, and environmental microbiol-
ogy findings. Epidemiologic and clinical findings usually in-
clude a common-source outbreak of a short incubation
period and a short-duration, non-fatal, non-pneumonic ill-
ness characterized by malaise, myalgia, and fever. Age,
gender and smoking do not seem to be risk factors. To
date, there is no consensus on the duration of the incuba-
tion period, on its clinical symptoms, nor on the causal spe-
cies of Legionella.24,26 Also usually required are proven
exposure to an aerosolized environmental source contain-
ing Legionella bacteria and development of antibodies to
the isolated bacterium in a significant fraction of affected
people. Because of its benignity and lack of specific clinical
features, the occurrence of Pontiac fever is probably often
undiagnosed and is therefore less reported than LD.
Whether Pontiac fever is actually due to Legionella infec-
tion is even still a matter of debate.26 The very short incu-
bation period is too brief to allow high-grade bacterial
multiplication in the lung or elsewhere in the body. In addi-
tion, the absence of pneumonia, short duration of illness,
recurrent milder illness with rechallenge, and complete re-
covery without antibiotic treatment also make Legionella
infection very unlikely. Experts believe that pontiac fever
is probably due to exposure to a toxic mix of live and
dead microorganisms and their products, including endo-
toxin made by non-Legionella bacteria, as well as low-
dose live or dead Legionella bacteria incapable of causing
pneumonia in the affected host.26 However, the high fre-
quency of urine antigen positivity (36%) in an Oklahoma
City Pontiac fever outbreak provided reasonable evidence
that the Pontiac fever outbreak was due, in part, to inhala-
tion of live or dead L. pneumophila serogroup 1 and not just
from inhalation of non-Legionella bacterial endotoxin.25
Cases of LD may be sporadic or occur as part of an
outbreak. Sporadic cases are reported throughout the year,
but most cases of epidemic infection occur in the summer
and autumn, presumably because warmer weather encour-
ages proliferation of the bacteria in water. It is possible that
a recent rise in the number of sporadic cases in England and
the Netherlands may be associated with certain weather
conditions.27,28 In moderate climates many cases are travel-
associated. The disease tends to occur in the middle-aged
and elderly, in people who have impaired respiratory and
cardiac function, heavy smokers or immunocompromised
patients.29,30 In a recent prospective case-control study a
history of diabetes mellitus, current tobacco smoking, trav-
elling abroad, spending one or more nights away from
home not leaving the country and being a driver by profession
were independent risk factors for acquiring LD.31 Interest-
ingly, LD patients who had travelled abroad during their incu-
bation period differed from LD patients who had not. They
appeared healthier than non- or domestic travellers with re-
spect to a history of coronary disease, pulmonary disease,
current use of corticosteroids or immunosuppressives and
any medication.
The incubation time of LD is between 2 and 10 days.
Among patients of the Bovenkarspel outbreak the reported
incubation period was 2e19 days (median 7 days). In 22
cases (16%) the time before onset of illness exceeded
10 days.20 A prodromal illness may occur, lasting for hours
to several days, with symptoms of headache, myalgia, as-
thenia, and anorexia. Cavitary lung disease is relatively
common in immunocompromised patients, especially in
some groups of solid organ transplant patients, and requires
prolonged therapy.32 Extrapulmonary infection caused by
Legionella spp., a very rare occurrence, is now recognised
to occur in the presence and absence of LD, especially in
severely immunocompromised patients.32
It is not possible to clinically distinguish patients with LD
from patients with pneumococcal pneumonia. Several pro-
spective studies have shown that the two diseases have
nearly identical clinical and radiological findings, and that
non-specific laboratory test results cannot differentiate
between the two diseases.23,33e35 Non-specific features of
LD include fever, non-productive cough, myalgias, rigors,
dyspnea and diarrhea.36 Neurological symptoms range
from headache and lethargy to encephalopathy. Change
in mental status is the most common neurologic abnormal-
ity.37 Suspicion should be raised in cases of pneumonia and
the presence of headache, confusion, hyponatremia, ele-
vated creatine kinase.38 Also, the diagnosis becomes more
likely if an acute consolidating pneumonia fails to respond
to several days of b-lactam antimicrobial therapy, or if
the pneumonia is severe enough to require intensive care
unit hospitalization.23 Epidemiologic clues might include
use of a hot tub or recreational spa; recent pneumonia of
a co-worker, relative, or fellow traveler; and recent plumb-
ing work done at home or work. The non-specific presenta-
tion of LD makes clinical diagnosis very difficult and
mandates empiric therapy for this disease in most patients
with community-acquired pneumonia (CAP) of uncertain
etiology. The key to diagnosis is performing appropriate
microbiologic testing.
In addition to L. pneumophila, 20 Legionella species
have been documented as human pathogens on the basis
of their isolation from clinical material. Pneumonia due to
non-pneumophila Legionella species resembles, both clini-
cally and radiographically, that due to L. pneumophila.39
Like L. pneumophila, other Legionella species are inhabi-
tants of natural and man-made aqueous environments.
The majority of confirmed infections involving non-pneu-
mophila Legionella species has occurred in immunosup-
pressed patients.
LD is considered a very rare cause of pneumonia in
children; most children with LD are immunosuppressed.40
All cases of LD in neonates were hospital-acquired, and
most patients had potential risk-factors including pre-
maturity, bronchopulmonar dysplasia, and corticosteroid
use.
Mortality rates are highly variable, ranging from less
than 1% to as high as 80%, depending on the underlying
health of the patient, the promptness of specific therapy,
and whether the disease is sporadic, nosocomial, or part of
a large outbreak.10 Fatality rates of nosocomial disease
have declined by more than 50% in the United States over
the past 20 years; a similar but less dramatic decrease in
death rates of community-acquired cases has also been ob-
served.10 The declines in mortality rates appear to result
from better and faster disease recognition, especially
through use of the urinary antigen test, and more wide-
spread use of empiric therapy for pneumonia that includes
drugs active against L. pneumophila.
Legionella ssp. and LD
Diagnosis
Although diagnostic methods have improved since L. pneu-
mophila was first described in 1976, no currently available
test is able to diagnose all Legionella spp. in a timely fash-
ion with a high degree of sensitivity and specificity. Most of
the data are applicable to L. pneumophila, since sensitivity
and specificity estimates for non-pneumophila species are
not known.
Culture
Isolation of Legionella spp., which has a specificity of 100%,
is considered the gold standard for diagnosis of LD. Culture
diagnosis requires special media, adequate processing of
specimens, and technical expertise. Several days are re-
quired to obtain a positive result, with most Legionella
spp. colonies being detected within 7 days. Species other
than L. pneumophila may grow at a slower rate and may
therefore be detectable only after 10 days of incuba-
tion.23,41 In addition, some Legionella spp. have unusual
colony morphology and may therefore be more easily over-
looked. The standard medium used to culture Legionella is
BCYE agar supplemented with f-ketoglutarate, with or
without antimicrobial agents. The antibiotics most com-
monly added are polymyxin to control Gram-negative
growth, anisomycin against yeasts, and cefamandole or
vancomycin against Gram-positive bacteria. Vancomycin
should be chosen if culture is aimed at species other than
L. pneumophila, because cefamandole inhibits some
Legionella spp. that do not produce beta-lactamases.42
The identification and speciation of cultured strains of
Legionella species is often difficult, and even the more suc-
cessful chromatographic classification techniques have
struggled to discriminate newly described species.43 A se-
quence-based genotypic classification scheme based on the
mip gene is able to accurately discriminate among most Le-
gionella species.43 A sequence based classification scheme
based on the smaller 5S rRNA gene (104 bp) and partial 16S
rRNA gene sequencing, while less discriminatory than the
mip gene, enables species-specific identification of most
Legionella species implicated in human disease.44,45
Legionellae can be isolated from a variety of sample
types, although lower respiratory tract secretions (e.g.,
sputum and bronchoscopy samples) are the samples of
choice. Culture yield depends on the severity of illness,
with the lowest yield (15%e25%) in mild pneumonia and the
highest yield (>90%) for severe pneumonia causing re-
spiratory failure.46 A major limitation of sputum culture is
that fewer than one-half of patients with LD produce spu-
tum.46e49 Some patients with LD produce sputum that has
relatively little purulence; these samples may be rejected
by laboratories that discard sputum samples containing
few polymorphonuclear leukocytes. However, Ingram and
Plouffe demonstrated that up to 84% of L. pneumophila-
positive samples would have been discarded by using estab-
lished sputum purulence screens and they recommend
acceptance of all specimens submitted for Legionella cul-
ture.50 Estimated sensitivities of sputum culture range
from <10% to 80% and vary according to different compar-
ison standards and by individual laboratories.23,46 In
5
practice, the better results are likely to be achieved only
by laboratories with a special interest in Legionella
infection.
Serology
The indirect immunofluorescence assay (IFA) was used to
detect antibodies in patients from the Philadelphia outbreak
and was instrumental in determining the cause of the
illnesses. Since then, a number of serologic test methodol-
ogies has been developed to detect antibodies to Legionella
spp.51 Of the various antibody detection methods that are
available, IFA and enzyme-linked immunosorbent assays
(ELISA) are the most commonly used. Nowadays, ELISA assays
are preferred by many laboratories over IFA testing because
they are less subjective, thought to be more accurate than
IFA testing and have the potential for automated performan-
ce.41,52e54 The reported sensitivities of serological assays
vary substantially, from 41% to 94%.41 A recent report of
a study in which serum samples from outbreak-related LD
patients from The Netherlands were used110 showed sensitiv-
ities of 64%, 61%, and 44%, respectively, for an ELISA, IFA, and
a rapid microagglutination test. In ELISA, half of patients
showed a seroconversion in IgM and the other half a serocon-
version in IgG. Other studies have shown that, in many
patients with legionellosis, the immune response is primarily
IgM and that IgM tests must thus be included for optimal sen-
sitivity. Seroconversion may take several weeks, which is
a major limitation of serological testing. Approximately
25%e40% of patients with LD seroconvert within the first
week after the onset of symptoms.41 In most cases, a 4-fold
increase in antibody titer is detected within 3e4 weeks,
but in some cases, this may take more than 10 weeks.55
Acute-phase reciprocal IFA antibody titers of 256 in the
presence of pneumonia were once considered sufficient for
a presumptive diagnosis, but this has been shown to be unre-
liable, given the high prevalence of Legionella antibody pos-
itivity in persons without clinical evidence of legionellosis.56
The specificity of seroconversion using L. pneumophila se-
rogroup 1 antigen in IFA has been reported to be approxi-
mately 99%.51,57 A disadvantage of serological testing is the
inability to accurately detect all Legionella species and se-
rogroups. Although seroconversion to L. pneumophila se-
rogroup 1 is generally regarded as being highly diagnostic,
the sensitivity and specificity of seroconversion to other spe-
cies and serogroups has not been rigorously confirmed.46 In
summary, a diagnosis by a fourfold immunoglobulin G (IgG)
or IgM titer increase can only be made retrospectively, and
rarely influences the initial treatment of the patient. There-
fore there is a need for additional tests to diagnose LD in the
early stage of disease.
Detection of Legionella antigen in urine
The detection of Legionella antigenuria has been used al-
ready, shortly after the first outbreak in Philadelphia.58
Legionella antigenuria can be detected as early as 1 day af-
ter onset of symptoms and persists for days to weeks. In one
instance, excretion of antigen was documented to occur for
more than 300 days.59 The antigen detected is a component
of the lipopolysaccharide portion of the Legionella cell wall
6 B.M.W. Diederen
and is heat stable.60,61 The urinary antigen tests combine
reasonable sensitivity and high specificity with rapid re-
sults. It has revolutionized the laboratory diagnosis of LD,
making it the most common laboratory test for diagno-
sis.12,62 Application of the test to specimens other than
urine might prove useful for rapid diagnosis of Legionnaires
disease63 but need further validation. In Europe, the pro-
portion of cases diagnosed by the urinary antigen detection
has seen a dramatic increase since 1998,64 (Table 2).
In 1995, they represented 15% of diagnosed cases; in 1998
this had risen to 33%, in 2004 to 74%, and in 2006 more
than 90% of all cases.
Commercial kits that use both radioimmunoassay (RIA)
and enzyme immunoassay (EIA) methodologies have been
available for several years and have similar performance
characteristics.23 Agglutination assays have also been intro-
duced, but they have not demonstrated acceptable sensitiv-
ity and specificity.65 In addition, immunochromatographic
assays have been developed that have similar sensitivity
and specificity to EIA assays.66,67 The majority are most sen-
sitive for the detection of the Pontiac (MAb 2) monoclonal
antibody type of L. pneumophila serogroup 1 (up to 90%), less
sensitive for other monoclonal antibody types of L. pneumo-
phila serogroup 1 (w60%), and poorly sensitive (w5%) for
other L. pneumophila serogroups and other Legionella spe-
cies.68,69 Because the majority (about 90%) of cases of com-
munity-acquired LD are caused by the Pontiac subtype of
L. pneumophila serogroup 1, the average sensitivity of this
test is in the range of 70%e80%. An important feature of
these assays is their high specificity (>99%), which is a
requirement when testing a relatively rare disease.
The sensitivity of urinary antigen detection appears to
be associated with the clinical severity of disease.70 Yzer-
man et al. tested two enzyme immunoassays (Binax and Bi-
otest) and one immunochromatographic assay (Binax NOW),
using urine samples from outbreak-related LD patients. The
Binax EIA, Biotest EIA, and Binax NOW assay showed overall
sensitivities of 69, 71, and 72%, respectively. When the
tests were performed with concentrated urine samples,
the overall sensitivities increased to 79, 74, and 81%, re-
spectively. A statistically significant association was found
between clinical severity and test sensitivity for all tests.
For patients with mild LD, the test sensitivities ranged
from 40 to 53%, whereas for patients with severe LD who
Table 2 Cases and proportion by main method of diagnosis 1995e2004
Main method L. pneumophila sg1 L. pneumophila other
of diagnosis serogroup or serogroup
not determined
Isolation 2564 796
Antigen detection: urinary 14,000 1474
Serology: sero-conversion 1523 1289
Serology: single high titre 1480 1279
Antigen detection: respiratory 128 104
PCR 24 169
Other 59 90
Not known 294 147
Adapted from reference,64 with permission.
needed immediate special medical care, the sensitivities
reached 88e100%. These findings have implications for
the diagnostic process in patients with mild pneumonia
and suggest that patients with mild pneumonia may go un-
der diagnosed if urine antigen tests alone are used. The use
of concentrated urine samples increased sensitivity without
a significant decrease in specificity. Since this concentra-
tion step is time- and labor-intensive, some laboratories
only use this approach in case of equivocal results.
Another association between test sensitivity and certain
defined subpopulations was described by Helbig et al.66 The
clinical utility of Legionella urinary antigen assays for the
diagnosis of Legionnaires disease was assessed by using
samples from 317 culture-proven cases. The sensitivities
of the Binax EIA and Biotest EIA were found to be 94% for
travel-associated infection and 87 and 76% for commu-
nity-acquired infection but only 44 and 46% for nosoco-
mially acquired infection, respectively. This appears to be
caused primarily by the different serotype distribution in
these different categories: travel-associated and commu-
nity-acquired cases are caused predominantly by MAb
2-positive strains.
Several newly immunochromatographic urinary antigen
tests for the detection of L. pneumophila serogroup 1 in
urine have been developed recently.71,72 The Binax NOW
urinary antigen test, in concordance with the findings of
previous studies, has excellent sensitivity and specificity.
The performance of some new tests are below the accept-
able level for diagnostic assays.71
Detection of Legionella nucleic acid
The first assay designed to detect the DNA of L. pneumo-
phila was a radiolabeled ribosomal probe specific for all
strains of Legionella spp. (Gen-Probe, San Diego, Calif.).
Researchers reported varying sensitivity and specificity for
this assay.73e75 The use of the probe at one hospital
resulted in 13 false-positive cases76 and the assay was
removed from the market soon after this pseudo-outbreak.
PCR enables specific amplification of minute amounts
of Legionella DNA and can provide results within a short time
frame. It also has the potential to detect infections caused by
any Legionella species and serogroup. Legionella PCR is only
available in a limited number of laboratories that mostly use
Other Legionella All Legionella
species or species cases
not known
168 3528
393 15,867
455 3267
425 3184
21 253
54 247
27 176
281 722
Legionella ssp. and LD
a variety of in-house or commercial assays.23,41,77 A new com-
mercial test (BD ProbeTec ET L. pneumophila; Becton Dickin-
son) that detects serotypes 1e14 of L. pneumophila in
sputum is now cleared by the FDA, but published data on per-
formance characteristics are lacking.
Real-time PCR has added benefits to routine diagnosis,
as it minimizes manual time for the PCR and makes the use
of post-PCR analysis superflues. Diagnostic PCR assays have
principally targeted specific regions within 16S rRNA
genes,78e85 the 23S-5S spacer region,86 5S rDNA,87,88 or
the macrophage inhibitor potentiator (mip) gene.43,89e92
Thus far, encouraging results obtained mostly from in vitro
evaluations and small patient series have been reported.
Legionella DNA can be detected in urine, serum, and leuko-
cyte samples obtained from patients with LD with sensitiv-
ities of 30%e86%.87,93e95
The application of PCR to non-respiratory samples seems
particularly attractive, because this will circumvent the
problem of patients who do not produce sputum. The
sensitivity of the detection of Legionella DNA in serum is
relatively low (w50e60%) in LD patients, but might be
higher (w70e90%) in those patients with more severe dis-
ease.93,94 The proof for its presumed utility would lie in
a prospective study to evaluate the value of Legionella-
specific PCR on serum samples in patients with pneumonia.
When testing samples from the lower respiratory tract,
PCR has repeatedly been shown to have a sensitivity equal
to or greater than culture.46,84,85 Indeed, PCR is considered
the test of choice for patients who produce sputum by some
authors.46 However, a number of false-positive results have
been reported, both with commercially available tests and
with in house tests.23,84 A problem with the interpretation
of these false positive results is the question whether these
are truly false-positive or whether the reference method
failed, e.g. because less common legionellae are not as eas-
ily detected by conventional methods. It is difficult to solve
this issue and at present, there are no well designed studies
available that have determined the exact sensitivity and
specificity of Legionella PCR in patients with pneumonia
of unknown etiology. The quality performance of 46 partici-
pating laboratories for the detection of Legionella spp. by
two quality control (QC) exercises was investigated in
2004 and 2005.96 In-house methods were used by 93% of
participants. The rate of false positivity (panel members
negative for Legionella spp.) ranged from 4.0% in 2004 to
8.2% in 2005. The occurrence of false-positive Legionella
testing demonstrates the value of routine confirmatory test-
ing procedures, because such protocols can be beneficial in
rapidly detecting problems with diagnostic assays. Laborato-
ries should comply with stringent quality control require-
ments, and this QC study underlines that NAAT have not
yet been properly standardized in all laboratories. Labora-
tory workers and clinicians must be cautious when interpret-
ing results obtained from these types of assays and should not
hesitate to question results which are unexpected based on
clinical presentation and local epidemiology.
Treatment
LD is regarded by many as a plague with a high mortality. It
must be remembered that the original cases were patients
7
with severe disease who did not receive appropriate
antibiotics. In this millennium, mortality has decreased
with the increased index of suspicion by physicians, early
empirical treatment with antibiotics that cover Legionella
spp. and the advent of rapid laboratory tests. Delay in
starting appropriate therapy has been associated with in-
creased mortality.97 Benin et al. reported data from
a large-scale study by the Centers for Disease Control and
Prevention showing a decrease in the case-fatality rate
for community-acquired Legionella pneumonia from 26%
to 10% for the period 1980e1998.10
This finding is in accordance with recent studies of
patients with outbreak-related LD who received rapid
diagnoses by means of urine antigen testing; these studies
have reported case-fatality rates of 0%e5.5%.98e100
The choice of empiric antibiotic therapy for CAP is based
on the intention of providing optimal therapy, the epide-
miological features of various microorganisms, and an
inference of the most likely pathogen.101e103 Empirical an-
tibiotic therapy should target primarily S. pneumoniae be-
cause of its high incidence. In both seriously ill patients
and those suspected of having LD, antibiotic therapy should
also target L. pneumophila. Empirical therapy should be re-
placed with pathogen-directed therapy when a causative
agent is identified.
The intracellular location of the pathogen is relevant
for the efficacy of the antimicrobial agent. Antimicro-
bial agents that achieve intracellular concentrations higher
than the minimal inhibitory concentration (MIC) are
more effective than antibiotics with poor intracellular
penetration.104 Thus, macrolides, quinolones, tetracy-
clines, and rifampicin are most likely to be efficacious
(Table 3). There is debate as to whether rifampin pro-
vides additional benefit to patients with LD105,106; co-
administration of rifampicin is of questionable benefit and
is not recommended.107
A number of small uncontrolled, or underpowered pro-
spective controlled studies of the treatment of LD exist.
Prospective, adequate-size clinical trials of antimicrobial
therapy for LD have not yet been performed. Three
observational studies have evaluated the efficacy of macro-
lides (erythromycin, clarithromycin and azithromycin) ver-
sus levofloxacin.98,108,109 Time to defervescence was
shorter in patients on levofloxacin therapy in two studies.
Length of hospital stay was significantly shorter for patients
treated with levofloxacin in three studies. The overall mor-
tality was 4.5% for the macrolide group and 1.1% for the lev-
ofloxacin group but this difference was not statistically
significant. Because these studies are not randomized, the
possible superiority of levofloxacin therapy over therapy
with the older macrolides should be interpreted with cau-
tion. It is possible that concomitant early recognition of Le-
gionella spp. and a trend toward levofloxacin treatment in
the past few years coincided to achieve less morbidity and
fewer fatalities and that levofloxacin is not a better treat-
ment than macrolides.
In the absence of adequate-size human studies, decisions
about potential antimicrobial efficacy for LD are mostly
made on the basis of experimental animal and cell culture
studies. The ability of a drug to inhibit or kill intracellular
L. pneumophila usually correlates well with its clinical effec-
tiveness for LD.104e106 Similarly, therapy studies using
8 B.M.W. Diederen

Table 3 Preferred therapy for Legionnaires Disease

Conclusions
a b c
Antimicrobial agent Dosage Route
When systematically sought, Legionella species are consis-
d
Macro-azalides tently recognized as one of the more common causes of
Azithromycin 500 mg every 24 h IV, p.o. pneumonia. The failure to diagnose LD in routine clinical
Clarithromycin 500 mg every 12 h IV, p.o. practice largely depends on 3 factors: the inability to clin-
Spiramycin 1.5 M IU every 8 h IV ically and radiographically distinguish LD from other causes
6e9 M IU p.o. of pneumonia, the omission to order specific diagnostic
(total daily dose) tests for Legionella infection, and the shortcomings of
Erythromycin 1000 mg every 6e8 h IV, p.o. available diagnostic tests. During an epidemic or in a setting
Tetracyclines with an unusual high prevalence, a specificity of 100% is not
Doxycyclinee 200 mg every 24 h IV, p.o. an essential prerequisite for a diagnostic test. However,
when the prevalence of infection is low, even a modest
Fluorochinolones loss of specificity will result in many false-positive findings.
Ciprofloxacin 400 mg every 8e12 h IV The sensitivity of diagnostic tests for LD is usually in the
500e750 mg every 12 h p.o. 60e70% range, and does not exceed 90% for any test
Gatifloxacinf 200e400 mg every 24 h IV, p.o. used. Therefore it appears that none of the individual diag-
Gemifloxacinf 320 mg every 24 h p.o. nostic tests fulfils the needs of both clinicians and microbi-
Levofloxacin 500e750 mg every 24 h IV, p.o. ologists, and the examination of different specimen types
Moxifloxacin 400 mg every 24 h IV, p.o. with several tests in parallel is recommended.
Ofloxacin 400 mg every 12 h IV, p.o. The amount of microbiological work-up should be de-
Ketolides termined by the severity of the pneumonia. The incidence
Telithromycinf 800 mg every 24 h p.o. of LD is higher among patients with severe pneumonia, and
all patients with pneumonia who are admitted to an
Adapted from reference.106 intensive care unit should therefore be tested for this
a
Some antibiotics are not commercially available in selected
infection. Patients with pneumonia that does not respond
countries.
b to therapy with beta-lactam antibiotics or the combination
With some of these drugs, dosage adjustments have to be
made for renal insufficiency. Therapy duration may need to with aminoglycosides, or patients with severe underlying
be considerably longer for patients with lung abscesses, empy- disease, should also be tested for Legionella spp. Culture
ema, endocarditis, or extrathoracic infection. diagnosis remains the gold standard for diagnosis of LD
c
Oral therapy can be used in patients after response to IV an- and is the most specific diagnostic procedure, but its rela-
tibiotic therapy (switch therapy), or in mild cases that do not tively low sensitivity and the reliance on the availability
require hospitalisation. of a lower respiratory tract sample make it inadequate as
d
In mild cases other oral macrolides are also effective: josa- a sole diagnostic test.
mycin (1000 mg every 12 h), roxithromycin (150 mg every 12 h), Between 1995 and 2005 a total of 4056 culture confirmed
dirithromycin (500 mg every 24 h).
e cases of LD from European countries were notified to the
Given in one or two divided doses.
f European Working Group for Legionella Infections (EWGLI).
Because of a lack of clinical experience, their use is only
recommended in mild to moderate cases.

Table 4 Legionella isolates by species between 1995 and

2005
a guinea pig model of LD correlate quite well with drug effec- L. pneumophila 3867
tiveness for the treatment of the disease in humans. In vitro L. anisa 3
data suggest that newer macrolides (azithromycin and clari- L. bozemanii 19
thromycin) and many fluoroquinolone agents show the best L. brunensis 1
activity against Legionella species. Additionally, these L. cincinnatiensis 2
agents have fewer side effects than erythromycin. Newer L. dumoffii 5
macrolides and levofloxacin are licensed by the Food and L. feeleii 2
Drug Administration for the treatment of LD and are con- L. gormanii 3
sidered preferable to erythromycin. Newer macrolides L. jordanis 1
(especially azithromycin) have been shown to have some ad- L. longbeachae 18
ditional beneficial effect. However, the lack of an intrave- L. micdadei 24
nous formulation in some European countries has limited L. parisiensis 1
the use of newer macrolides in severely ill patients. The du- L. sainthelensi 1
ration of therapy has to be decided on an individualized ba- Legionella spp 109
sis.106 Clinical judgement must be used to establish the Total 4056
optimal length of treatment, but usually a 7e14 day course
of therapy is sufficient to cure most patients. However, ther- The Legionella species in the table constitute reports where
the country has not specified the species of the isolate. Data
apy duration may need to be considerably longer for patients
provided by Dr. C.A. Joseph, European Working Group for
with lung abscesses, empyema, endocarditis, or extrathora-
Legionella Infections (EWGLI).
cic infection.
Legionella ssp. and LD
L. pneumophila constituted 98.0% (3867/3947) of the iso-
lates for whom the species was known (Table 4). In The Neth-
erlands, LD is a notifiable disease and a supplementary
nationwide source identification program is operative (Be-
monsterings Eenheid Legionella-pneumonie; BEL project).
In this survey, 172 culture-confirmed LD cases were identi-
fied between August 2002 and October 2006. L. pneumophila
constituted 98.8% (170) of the isolates. Serogroup 1 was the
predominant serogroup (86.6%), and serogroups 2 (2.3%), 3
(1.7%), 6 (2.3%), and 14 (5.2%) accounted for the remaining
serogroups. (E. Yzerman, J. Bruin; Regional Laboratory of
Public Health Haarlem, Haarlem, The Netherlands, personal
communication). In regions where L. pneumophila serogroup
1 are numerically the most important Legionella species
causing disease, urinary antigen detection is recommended
for patients with severe CAP and in patients where this infec-
tion is clinically or epidemiologically suspected. In case of
a negative antigen test, Legionella-specific PCR should be
considered in severe pneumonia of unknown etiology. In re-
gions where legionellae other than L. pneumophila se-
rogroup 1 are important pathogens, current urinary antigen
tests are still useful but should never be used as the sole di-
agnostic tool. The availability of a good diagnostic reper-
toire, suitable for accurately diagnosing LD, constitutes the
basis for the early recognition and treatment of the individ-
ual patient as well as for effective measures for prevention
and control.
Prospective studies are needed that directly compare
multiple PCR assays available against reference tests in
larger samples. The results of such comparisons would be
more helpful to clinicians and microbiologists, who are
faced with having to choose among many new tests, and
might help to establish standard PCR methods that are
robust enough to be used outside the setting of certain
specialised reference laboratories.
Acknowledgements
I thank Prof. Dr. Jan Kluytmans (Laboratory for Microbiology
and Infection Control, Amphia Hospital, Breda, The Nether-
lands and Department of Medical Microbiology and Infection
Control, VU University Medical Center, Amsterdam, The
Netherlands), Dr. Marcel Peeters (Laboratory for Medical
Microbiology and Immunology, St. Elisabeth Hospital,
Tilburg, The Netherlands) and Prof. Dr. Christina Vanden-
broucke-Grauls (Department of Medical Microbiology and
Infection Control, VU University Medical Center, Amster-
dam, The Netherlands) for critically reviewing the
manuscript.
References
1. Fraser DW, Tsai TR, Orenstein W, Parkin WE, Beecham HJ,
Sharrar RG, et al. Legionnaires disease: description of an ep-
idemic of pneumonia. N Engl J Med 1977;297:1189e97.
2. McDade JE, Shepard CC, Fraser DW, Tsai TR, Redus MA,
Dowdle WR. Legionnaires disease: isolation of a bacterium
and demonstration of its role in other respiratory disease.
N Engl J Med 1977;297:1197e203.
3. Winn WC. Legionnaires disease: historical perspective. Clin
Microbiol Rev 1988;1:60e81.
9
4. Glick TH, Gregg MB, Berman B, Mallisson G, Rhodes WW,
Kassanoff I. Pontiac fever: an epidemic of unknown etiology
in a health department: I. Clinical and epidemiologic aspects.
Am J Epidemiol 1978;107:149e60.
5. Maiwald M, Helbig JH, Luck PC. Laboratory methods for the
diagnosis of Legionella infections. J Microbiol Methods
1998;33:59e79.
6. Benson RF, Fields BS. Classification of the genus Legionella.
Semin Respir Infect 1998;13:90e9.
7. Rowbotham TJ. Legionella-like amoebal pathogens. In:
Barbaree JM, Breiman RF, Dufour AM, editors. Legionella: cur-
rent status and emerging perspectives. Washington, D.C.:
American Society for Microbiology; 1993. p. 137e40.
8. Rowbotham TJ. Preliminary report on the pathogenicity of Le-
gionella pneumophila for freshwater and soil amoebae. J Clin
Pathol 1980;33:1179e83.
9. ONeill E, Humphreys H. Surveillance of hospital water and
primary prevention of nosocomial legionellosis: what is the
evidence? J Hosp Infect 2005;59:273e9.
10. Benin AL, Benson RF, Besser RE. Trends in Legionnaires dis-
ease, 1980e1998: declining mortality and new patterns of
diagnosis. Clin Infect Dis 2002;35:1039e46.
11. Yu VL, Plouffe JF, Pastoris MC, Stout JE, Schousboe M,
Widmer A, et al. Distribution of Legionella species and se-
rogroups isolated by culture in patients with sporadic commu-
nity-acquired legionellosis: an international collaborative
survey. J Infect Dis 2002;186:127e8.
12. Joseph JA. European Working Group for Legionella Infections.
Legionnaires disease in Europe 2000e2002. Epidemiol Infect
2004;132:417e24.
13. Helbig JH, Bernander S, Castellani PM, Etienne J, Gaia V,
Lauwers S, et al. Pan-European study on culture-proven le-
gionnaires disease: distribution of Legionella pneumophila
serogroups and monoclonal subgroups. Eur J Clin Microbiol In-
fect Dis 2002;21:710e6.
14. Harrison TG, Fry NK, Afshar B, Bellamy W, Doshi N,
Underwood AP. Typing of Legionella pneumophila and its
role in elucidating the epidemiology of Legionnaires Disease.
In: Cianciotto NP, Kwaik Y, Edelstein PS, Fields BS, Geary DF,
Harrison TG, et al., editors. Legionella: state of the art
30 years after its recognition. Washington, D.C.: American
Society for Microbiology; 2006. p. 94e9.
15. Fry NK, Bangsborg JM, Bergmans A, Bernander S, Etienne J,
Franzin L, et al. Designation of the European Working
Group on Legionella Infections (EWGLI) amplified fragment
length polymorphism types of Legionella pneumophila
serogroup 1 and results of intercentre proficiency testing
using a standard protocol. Eur J Clin Microbiol Infect Dis
2002;21:722e8.
16. Gaia V, Fry NK, Afshar B, Luck PC, Meugnier H, Etienne J,
et al. Consensus sequence-based scheme for epidemiological
typing of clinical and environmental isolates of Legionella
pneumophila. J Clin Microbiol 2005;43:2047e52.
17. Ratzow S, Gaia V, Helbig JH, Fry NK, Luck PC. Addition of
neuA, the gene encoding N-acylneuraminate cytidylyl trans-
ferase, increases the discriminatory ability of the consensus
sequence-based scheme for typing Legionella pneumophila
serogroup 1 strains. J Clin Microbiol 2007;45:1965e8.
18. Fendukly F, Bernander S, Hanson HS. Nosocomial Legion-
naires disease caused by Legionella pneumophila serogroup
6: implication of the sequence-based typing method (SBT).
Scand J Infect Dis 2007;39:213e6.
19. Afshar B, Fry NK, Bellamy W, Underwood AP, Harrison TG.
Members of the European Working Group for Legionella Infec-
tions. External quality assessment of a DNA sequence-based
scheme for epidemiological typing of Legionella pneumophila
by an international network of laboratories. J Clin Microbiol
2007;45:3251e6.
10
20. Boshuizen HC, Neppelenbroek SE, van Vliet H, Schellekens JF,
den Boer JW, Peeters MF, et al. Subclinical Legionella infec-
tion in workers near the source of a large outbreak of legion-
naires disease. J Infect Dis 2001;184:515e8.
21. Fraser DW. Legionellosis: evidence of airborne transmission.
Ann N Y Acad Sci 1980;353:61e6.
22. Marrie TJ, Haldane D, MacDonald S, Clarke K, Fanning C, Le
Fort-Jost S, et al. Control of endemic nosocomial Legion-
naires disease by using sterile potable water for high risk pa-
tients. Epidemiol Infect 1991;107:591e605.
23. Fields BS, Benson RF, Besser RE. Legionella and Legionnaires
disease: 25 years of investigation. Clin Microbiol Rev 2002;15:
506e26.
24. Tossa P, Deloge-Abarkan M, Zmirou-Navier D, Hartemann P,
Mathieu L. Pontiac fever: an operational definition for epide-
miological studies. BMC Public Health 2006;6:112.
25. Burnsed LJ, Hicks LA, Smithee LM, Fields BS, Bradley KK,
Pascoe N, et al. A large travel-associated outbreak of legion-
ellosis among hotel guests: utility of the urine antigen assay in
confirming Pontiac fever. Clin Infect Dis 2007;44:222e8.
26. Edelstein PH. Urine antigen tests positive for Pontiac fever:
implications for diagnosis and pathogenesis. Clin Infect Dis
2007;44:229e31.
27. Joseph CA, van der Sande M. Unexplained summer increase in
non-travel-related legionellosis in the UK and Netherlands.
Euro Surveill 2006;11:E061018.1.
28. Hicks LA, Rose Jr CE, Fields BS, Drees ML, Engel JP,
Jenkins PR, et al. Increased rainfall is associated with in-
creased risk for legionellosis. Epidemiol Infect 2007;135:
811e7.
29. Broome CV, Fraser DW. Epidemiologic aspects of legionellosis.
Epidemiol Rev 1979;1:1e16.
30. Marston BJ, Lipman HB, Breiman RF. Surveillance for Legion-
naires disease: risk factors for morbidity and mortality.
Arch Intern Med 1994;154:2417e22.
31. Den Boer JW, Nijhof J, Friesema I. Risk factors for sporadic
community-acquired Legionnaires disease. A 3-year national
case-control study. Public Health 2006;120:566e71.
32. Edelstein PH. Clinical features of Legionnaires Disease: a se-
lected review. In: Cianciotto NP, Kwaik Y, Edelstein PS,
Fields BS, Geary DF, Harrison TG, et al., editors. Legionella:
state of the art 30 years after its recognition. Washington,
D.C.: American Society for Microbiology; 2006. p. 3e7.
33. Edelstein PH. Legionnaires disease. Clin Infect Dis 1993;16:
741e9.
34. Roig J, Aguilar X, Ruiz Z, Domingo C, Mesalles E, Manterola J,
et al. Comparative study of Legionella pneumophila and
other nosocomial-acquired pneumonias. Chest 1991;99:344e
50.
35. Granados A, Podzamczer D, Gudiol F, Manresa F. Pneumonia
due to Legionella pneumophila and pneumococcal pneumo-
nia: similarities and differences on presentation. Eur Respir
J 1989;2:130e4.
36. Tsai TF, Finn DR, Plikaytis BD, McCauley W, Martin SM,
Fraser DW. Legionnaires disease: clinical features of the ep-
idemic in Philadelphia. Ann Intern Med 1979;90:509e17.
37. Morelli N, Battaglia E, Lattuada P. Brainstem involvement in
Legionnaires disease. Infection 2006;34:49e52.
38. Pedro-Botet ML, Sabria M. Legionellosis. Semin Respir Crit
Care Med 2005;26:625e34.
39. Muder RR, Yu VL. Infection due to Legionella species other
than L. pneumophila. Clin Infect Dis 2002;35:990e8.
40. Greenberg D, Chiou CC, Famigilleti R, Lee TC, Yu VL. Problem
pathogens: paediatric legionellosis-implications for improved
diagnosis. Lancet Infect Dis 2006;6:529e35.
41. den Boer JW, Yzerman EP. Diagnosis of Legionella infection in
Legionnaires disease. Eur J Clin Microbiol Infect Dis 2004;23:
871e8.
B.M.W. Diederen
42. Lee TC, Vickers RM, Yu VL, Wagener MM. Growth of 28 Legion-
ella species on selective culture media: a comparative study.
J Clin Microbiol 1993;31:2764e8.
43. Ratcliff RM, Lanser JA, Manning PA, Heuzenroeder MW.
Sequence-based classification scheme for the genus Legion-
ella targeting the mip gene. J Clin Microbiol 1998;36:
1560e7.
44. Murdoch DR, Light GJ, Jennings LC, Chambers ST. Sequence-
based classification scheme for the genus Legionella targeting
the 5S rRNA gene. J Clin Microbiol 1991;37:281.
45. Wilson DA, Reischl U, Hall GS, Procop GW. Use of partial 16S
rRNA gene sequencing for identification of Legionella pneu-
mophila and non-pneumophila Legionella spp. J Clin
Microbiol 2007;45:257e8.
46. Murdoch DR. Diagnosis of Legionella infection. Clin Infect Dis
2003;36:64e9.
47. Sopena N, Sabria-Leal M, Pedro-Botet ML, Padilla E,
Dominguez J, Morera J, et al. Comparative study of the clin-
ical presentation of Legionella pneumonia and other commu-
nity-acquired pneumonias. Chest 1998;113:1195e200.
48. Woodhead MA, Macfarlane JT. Legionnaires disease: a review
of 79 community acquired cases in Nottingham. Thorax 1986;
41:635e40.
49. Tsai TF, Finn DR, Plikaytis BD, McCaule W, Martin SM,
Fraser DW. Legionnaires disease: clinical features of the
epidemic in Philadelphia. Ann Intern Med 1979;90:509e17.
50. Ingram JG, Plouffe JF. Danger of sputum purulence screens in
culture of Legionella species. J Clin Microbiol 1994;32:
209e10.
51. Harrison TG, Taylor AG. The diagnosis of Legionnaires disease
by estimation of antibody levels. In: Harrison TG, Taylor AG,
editors. A laboratory manual for Legionella. Chichester, En-
gland: John Wiley and Sons, Ltd; 1988. p. 113e35.
52. Malan AK, Martins TB, Jaskowski TD, Hill HR, Litwin CM. Com-
parison of two commercial enzyme-linked immunosorbent
assays with an immunofluorescence assay for detection of
Legionella pneumophila types 1 to 6. J Clin Microbiol 2003;
41:3060e3.
53. Boshuizen HC, Den Boer JW, de Melker H, Schellekens JF,
Peeters MF, van Vliet JA, et al. Reference values for the
SERION classic ELISA for detecting Legionella pneumophila
antibodies. Eur J Clin Microbiol Infect Dis 2003;22:706e8.
54. Wreghitt TG, Nagington J, Gray J. An ELISA test for the detec-
tion of antibodies to Legionella pneumophila. J Clin Pathol
1982;35:657e60.
55. Monforte R, Estruch R, Vidal J, Cervera R, Urbano-Marquez A.
Delayed seroconversion in Legionnaires disease. Lancet 1988;
2:513.
56. Plouffe JF, File TM, Breiman RF, Hackman BA, Salstrom SJ,
Marston BJ, et al. Reevaluation of the definition of Legion-
naires disease: use of the urinary antigen assay. Clin Infect
Dis 1995;20:1286e91.
57. Wilkinson HW, Cruce DD, Broome CV. Validation of Legionella
pneumophila indirect immunofluorescence assay with epi-
demic sera. J Clin Microbiol 1981;13:139e46.
58. Berdal BP, Farshy CE, Feely JC. Detection of Legionella pneu-
mophila antigen in urine by enzyme-linked immunospecific
assay. J Clin Microbiol 1979;9:575e8.
59. Kohler RB, Winn WC, Wheat LJ. Onset and duration of urinary
antigen excretion in Legionnaires disease. J Clin Microbiol
1984;20:605e7.
60. Kohler RB, Zimmerman SE, Wilson E, Allen SD, Edelstein PH,
Wheat LJ, et al. Rapid radioimmunoassay diagnosis of Legion-
naires disease: detection and partial characterization of uri-
nary antigen. Ann Intern Med 1981;94:601e5.
61. Williams A, Lever MS. Characterisation of Legionella pneumo-
phila antigen in urine of guinea pigs and humans with Legion-
naires disease. J Infect 1995;30:13e6.
Legionella ssp. and LD
62. Formica N, Yates M, Beers M, Carnie J, Hogg G, Ryan N, et al.
The impact of diagnosis by legionella urinary antigen test on
the epidemiology and outcomes of legionnaires disease.
Epidemiol Infect 2001;127:275e80.
63. Wever PC, Notermans DW, Tulevski II, Schattenkerk JK, de
Jong MD. Detection of Legionella pneumophila serogroup 1
antigen in bronchoalveolar lavage fluid by an immunochroma-
tographic assay. J Clin Microbiol 2003;41:2265.
64. Joseph CA, Ricketts KD. Legionnaires Disease in Europe
1995e2004: a ten year review. In: Cianciotto NP, Kwaik Y,
Edelstein PS, Fields BS, Geary DF, Harrison TG, et al., editors.
Legionella: state of the art 30 years after its recognition.
Washington, D.C.: American Society for Microbiology; 2006.
p. 89e93.
65. Leland DS, Kohler RB. Evaluation of the L-CLONE Legionella
pneumophila Serogroup 1 Urine Antigen Latex Test. J Clin
Microbiol 1991;29:2220e3.
66. Helbig JH, Uldum SA, Bernander S, Luck PC, Wewalka G,
Abraham B, et al. Clinical utility of urinary antigen detection
for diagnosis of community-acquired, travel-associated, and
nosocomial Legionnaires disease. J Clin Microbiol 2003;41:
838e40.
67. Edelstein PH. Urinary antigen detection for Legionella spp. In:
Isenberg HD, editor. Clinical microbiology procedures man-
ual. 2nd ed. Washington: ASM Press; 2004. p. 11.4.1e6.
68. Harrison T, Uldum S, Alexiou-Daniel S, Bangsborg J,
Bernander S, Drasbrevear V, et al. A multicenter evaluation
of the Biotest Legionella urinary antigen EIA. Clin Microbiol
Infect 1998;4:359e65.
69. Domnguez J, Gali N, Blanco S, Pedroso P, Prat C, Matas L,
et al. Assessment of a new test to detect Legionella urinary
antigen for the diagnosis of Legionnaires disease. Diagn Mi-
crobiol Infect Dis 2001;41:199e203.
70. Yzerman EP, den Boer JW, Lettinga KD, Schellekens J,
Dankert J, Peeters M. Sensitivity of three urinary antigen tests
associated with clinical severity in a large outbreak of Legion-
naires disease in The Netherlands. J Clin Microbiol 2002;40:
3232e6.
71. Diederen BM, Peeters MF. Evaluation of two new immunochro-
matographic assays (Rapid U Legionella Antigen Test and SD
Bioline Legionella Antigen Test) for the detection of Legion-
ella pneumophila serogroup 1 antigen in urine. J Clin Micro-
biol 2006;44:2991e3.
72. Diederen BM, Peeters MF. Evaluation of the SAS Legionella
Test, a new immunochromatographic assay for the detection
of Legionella pneumophila serogroup 1 antigen in urine.
Clin Microbiol Inf 2007;13:86e8.
73. Doebbeling BN, Bale MJ, Koontz FP, Helms CM, Wenzel RP,
Pfaller MA. Prospective evaluation of the Gen-Probe assay
for detection of legionellae in respiratory specimens. Eur J
Clin Microbiol Infect Dis 1988;7:748e52.
74. Pasculle AW, Veto GE, Krystofiak S, McKelvey K, Vrsalovic K.
Laboratory and clinical evaluation of a commercial DNA probe
for the detection of Legionella spp. J Clin Microbiol 1988;27:
2350e8.
75. Wilkinson HW, Sampson JS, Plikaytis BB. Evaluation of a com-
mercial gene probe for identification of Legionella cultures.
J Clin Microbiol 1986;23:217e20.
76. Laussucq S, Schuster D, Alexander WJ, Thacker WL,
Wilkinson HW, Spika JS. False-positive DNA probe test for Le-
gionella species associated with a cluster of respiratory ill-
nesses. J Clin Microbiol 1988;26:1442e4.
77. Ginevra C, Barranger C, Ros A, Mory O, Stephan JL,
Freymuth F, et al. Development and evaluation of Chlamy-
lege, a new commercial test allowing simultaneous detection
and identification of Legionella, Chlamydophila pneumoniae,
and Mycoplasma pneumoniae in clinical respiratory specimens
by multiplex PCR. J Clin Microbiol 2005;43:3247e54.
11
78. Jonas D, Rosenbaum A, Weyrich S, Bhakdi S. Enzyme-linked
immunoassay for detection of PCR-amplified DNA of legionel-
lae in bronchoalveolar fluid. J Clin Microbiol 1995;33:
1247e52.
79. Reischl U, Linde HJ, Lehn N, Landt O, Barratt K,
Wellinghausen N. Direct detection and differentiation of Le-
gionella spp. and Legionella pneumophila in clinical speci-
mens by dual-color real-time PCR and melting curve
analysis. J Clin Microbiol 2002;40:3814e7.
80. Rantakokko-Jalava K, Jalava J. Development of conven-
tional and real-time PCR assays for detection of Legionella
DNA in respiratory specimens. J Clin Microbiol 2001;39:
2904e10.
81. Wellinghausen N, Frost C, Marre R. Detection of legionellae in
hospital water samples by quantitative real-time LightCycler
PCR. Appl Environ Microbiol 2001;67:3985e93.
82. Stolhaug A, Bergh K. Identification and differentiation of
Legionella pneumophila and Legionella spp. with real-
time PCR targeting the 16S rRNA gene and species identifi-
cation by mip sequencing. Appl Environ Microbiol 2006;72:
6394e8.
83. van der Zee A, Verbakel H, de Jong C, Pot R, Bergmans A,
Peeters M, et al. Novel PCR-probe assay for detection of
and discrimination between Legionella pneumophila and
other Legionella species in clinical samples. J Clin Microbiol
2002;40:1124e5.
84. Cloud JL, Carroll KC, Pixton P, Erali M, Hillyard DR. Detection
of Legionella species in respiratory specimens using PCR with
sequencing confirmation. J Clin Microbiol 2000;38:1709e12.
85. Templeton KE, Scheltinga SA, Sillekens P, Crielaard JW, van
Dam AP, Goossens H, et al. Development and clinical evalua-
tion of an internally controlled, single-tube multiplex real-
time PCR assay for detection of Legionella pneumophila and
other Legionella species. J Clin Microbiol 2003;41:4016e21.
86. Herpers B, de Jongh BM, van der Zwaluw K, van Hannen EJ.
Real-time PCR assay targets the 23S-5S spacer for direct de-
tection and differentiation of Legionella spp. and Legionella
pneumophila. J Clin Microbiol 2003;41:4815e6.
87. Murdoch DR, Walford EJ, Jennings LC, Light GJ, Schousboe MI,
Chereshsky AY, et al. Use of the polymerase chain reaction to
detect Legionella DNA in urine and serum samples from
patients with pneumonia. Clin Infect Dis 1996;23:475e80.
88. Kessler HH, Reinthaler FF, Pschaid A, Pierer K, Kleinhappl B,
Eber E, et al. Rapid detection of Legionella species in bron-
choalveolar lavage fluids with the EnviroAmp Legionella PCR
amplification and detection kit. J Clin Microbiol 1993;31:
3325e8.
89. Koide M, Saito A. Diagnosis of Legionella pneumophila infec-
tion by polymerase chain reaction. Clin Infect Dis 1995;21:
199e201.
90. Lindsay DS, Abraham WH, Fallon RJ. Detection of mip gene by
PCR for diagnosis of Legionnaires disease. J Clin Microbiol
1994;32:3068e9.
91. Wilson DA, Yen-Lieberman B, Reischl U, Gordon SM,
Procop GW. Detection of Legionella pneumophila by real-
time PCR for the mip gene. J Clin Microbiol 2003;41:3327e30.
92. Jaulhac B, Nowicki M, Bornstein N, Meunier O, Prevost G,
Piemont Y, et al. Detection of Legionella spp. in bronchoal-
veolar lavage fluids by DNA amplification. J Clin Microbiol
1992;30:920e4.
93. Diederen BM, de Jong CM, Marmouk F, Kluytmans JA,
Peeters MF, van der Zee A. Evaluation of real-time PCR for
the early detection of Legionella pneumophila DNA in serum
samples. J Med Microbiol 2007;56:94e101.
94. Diederen BM, Bruin JP, den Boer JW, Peeters MF, Yzerman EP.
Sensitivity of Legionella pneumophila DNA detection in serum
samples in relation to disease severity. J Med Microbiol 2007;
56:1255.
12
95. Helbig JH, Engelstadter T, Maiwald M, Uldum SA, Witzleb W,
Luck PC. Diagnostic relevance of the detection of Legionella
DNA in urine samples by the polymerase chain reaction. Eur
J Clin Microbiol Infect Dis 1999;18:716e22.
96. Mendonca R, Wallace P, NacKay WG, Rottiers S, Crevecoeur S,
Maes N, et al. External quality assessment for the detection of
Legionella pneumophila by nucleic acid amplification technol-
ogy (NAAT)- A European pilot program. In: Programme and ab-
stract book 21st meeting of the European Working Group for
Legionella infections, Lisbon, p. 54.
97. Heath CH, Grove DI, Looke DF. Delay in appropriate therapy of
Legionella-pneumonia associated with increased mortality.
Eur J Clin Microbiol Infect Dis 1996;15:286e90.
98. Mykietiuk A, Carratala J, Fernadez-Sabe N, Dorca J,
Verdaguer R, Manresa F, et al. Clinical outcomes for hospital-
ized patients with Legionella pneumonia in the antigenuria
era: the influence of levofloxacin therapy. Clin Infect Dis
2005;40:794e9.
99. Plouffe JF, Breiman RF, Fields BS, Herbert M, Inverso J,
Knirsch C, et al. Azithromycin in the treatment of Legionella
pneumonia requiring hospitalization. Clin Infect Dis 2003;37:
1475e80.
100. Yu VL, Greenberg RN, Zadeikis N, Stout JE, Khashab MM,
Olson WH, et al. Levofloxacin efficacy in the treatment of
community-acquired Legionellosis. Chest 2004;125:2135e9.
101. Niederman MS, Bartlett JG, Breiman RF, Mandell LA,
File Jr TM. Community-acquired pneumonia in adults: guide-
lines for management. The Infectious Diseases Society of
America. Clin Infect Dis 1998;26:811e38.
102. Mandell LA, Wunderink RG, Anzueto A, Bartlett JG,
Campbell GD, Dean NC, et al. Infectious Diseases Society of
America; American Thoracic Society. Infectious Diseases Soci-
ety of America/American Thoracic Society consensus
B.M.W. Diederen
guidelines on the management of community-acquired pneu-
monia in adults. Clin Infect Dis 2007;44(Suppl. 2):S27e72.
103. Woodhead M, Blasi F, Ewig S, Huchon G, Ieven M, Ortqvist A,
et al. European Respiratory Society; European Society of Clin-
ical Microbiology and Infectious Diseases. Guidelines for the
management of adult lower respiratory tract infections. Eur
Respir J 2005;26:1138e80.
104. Roig J, Rello J. Legionnaires disease: a rational approach to
therapy. J Antimicrob Chemother 2003;51:1119e29.
105. Edelstein PH. Antimicrobial chemotherapy for Legionnaires
disease: time for a change. Ann Intern Med 1998;129:
328e30.
106. Roig J, Rello J. Legionella: state of the art 30 years after its
recognition. In: Cianciotto NP, Kwaik Y, Edelstein PS,
Fields BS, Geary DF, Harrison TG, et al., editors. Treatment
of Legionnaires Disease. Washington, D.C.: American Society
for Microbiology; 2006. p. 8e14.
107. Grau S, Antonio JM, Ribes E, Salvado M, Garces JM, Garau J.
Impact of rifampicin addition to clarithromycin in Legionella
pneumophila pneumonia. Int J Antimicrob Agents 2006;28:
249e52.
108. Sabria M, Pedro-Botet ML, Gomez J, Roig J, Vilaseca B,
Sopena N, et al. LD Therapy Group. Fluoroquinolones vs mac-
rolides in the treatment of Legionnaires disease. Chest 2005;
128:1401e5.
109. Blazquez Garrido RM, Espinosa Parra FJ, Alemany Frances L, Ra-
mos Guevara RM, Sanchez-Nieto JM, Segovia Hernandez M,
et al. Antimicrobial chemotherapy for legionnaires disease:
levofloxacin versus macrolides. Clin Infect Dis 2005;40:800e6.
110. Yzerman EP, den Boer JW, Lettinga KD, Schel AJ,
Schellekens J, Peeters M, et al. Sensitivity of three serum an-
tibody tests in a large outbreak of Legionnaires disease in
The Netherlands. J Med Microbiol 2006;55:561e6.