(Advances in Cell Aging and Gerontology 17) Mark P. Mattson (Eds.) - Sleep and Aging-Academic Press, Elsevier (2005)

TABLE OF CONTENTS
Preface vii
Chapter 1
Evolutionary Aspects of Sleep and Its REM and NREM States . . . . . . . . . . . . 1
J. Lee Kavanau
Chapter 2
Sleep Disturbances in Aging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Marci M. Loiselle, Melanie K. Means, and Jack D. Edinger
Chapter 3
Sleep and Learning in Animal Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Barry W. Row and David Gozal
Chapter 4
Sleep Disordered Breathing in the Geriatric Patient Population . . . . . . . . . . . . 79
Alon Y. Avidan
Chapter 5
Sleep Associated Endocrine and Immune Changes in the Elderly . . . . . . . . . . . . 113
Boris Perras and Jan Born
Chapter 6
Neurotrophic Factors and Sleep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Mark P. Mattson
Chapter 7
Sleep and Aging: Molecular Approaches within a Systems
Neurobiology Context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
Akira Terao and Thomas S. Kilduff
List of Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
v
PREFACE
Abnormalities in sleep can be symptoms or causes of a variety of disorders, many

of which are associated with aging and age-related disease. For example, insomnia is
a major component of clinical depression and anxiety disorders, and neurodegen-
erative disorders such as Alzheimers and Parkinsons diseases. Another sleep
disorder of considerable concern is apnea which may lead to brain damage and even
sudden death. Sleep disorders in the elderly may involve changes in neural circuits
that regulate dierent states of sleep including rapid eye movement (REM) and slow
wave sleep (SWS), and circadian rhythms. The neurochemical basis of sleep
disorders is beginning to be understood and this information is being used to
develop new ways of preventing and treating sleep disorders. In this volume of
Advances in Cell Aging and Gerontology entitled Sleep and Aging, experts in brain
aging and sleep, review key aspects of this complex and very important topic.
The rst chapter, written by J. Lee Kavanau, provides a fascinating account of
evolutionary aspects of sleep and its dierent states. Questions such as why is sleep
necessary for the survival of mammals and what is the role of sleep in memory, are
considered. Marci Loiselle and her colleagues then provide a review of sleep
disturbances in aging including how sleep patterns are altered in aging, the dierent
types of sleep disorders, and the evaluation and treatment of these disorders.
Abnormalities in breathing patterns, such as sleep apnea, are common in the elderly.
Alon Avidan reviews the characteristics and causes of sleep disordered breathing and
current approaches for the medical management of these disorders. Considerable
evidence is consistent with important roles for sleep in learning and memory. Barry
Row and David Gozal provide a timely review of the results of studies in animals
that reveal the eects of sleep on neural systems involved in synaptic plasticity and
memory, and the cellular and molecular mechanisms by which sleep aects these
processes. The endocrine and immune systems are altered during aging in ways that
reduce ones resistance to stress, infections, and other environmental factors. Boris
Perras and Jan Born describe how endocrine and immune functions change during
sleep and wakefulness, and how the normal sleep-related uctuations are aected
by aging. Finally, I review emerging evidence that suggests important roles for
neurotrophic factors in sleep and disorders of aging that involve abnormalities in
sleep. By virtue of its roles in learning and memory, neurogenesis, depression and
neurodegenerative disorders, brain-derived neurotrophic factor appears to be a
particularly important neurotrophic factor in sleep regulation.
Advances are being made in the understanding of the molecular and
cellular biology of sleep, the roles of aging and disease in sleep disturbances, and
how diseases of aging cause or exacerbate disorders such as insomnia and sleep
apnea. This volume of ACAG will provide graduate students, postdoctoral fellows,
and senior investigators in the eld of sleep research with a valuable source of
vii
viii Preface
information that will benet their research. In addition, neuroscientists, neurologists,

and psychiatrists will learn that many of the neuronal circuits and neurochemicals
that are altered in age-related neurodegenerative and psychiatric disorders are the
same as those involved in sleep disorders. Indeed, emerging ndings from preclinical
and clinical studies suggest that treatments benecial for depression (serotonin
reuptake inhibitors, for example) may also be eective for treating insomnia and
neurodegenerative disorders. Sleep well and you will age gracefully.
MARK P. MATTSON, PhD

Advances in
Cell Aging and
Gerontology
Evolutionary aspects of sleep and

its REM and NREM states
J. Lee Kavanau
Department of Organismic Biology, Ecology and Evolution University of California, Los Angeles,
CA 90095-1606, USA. Tel: 1-310-825-3474; fax: 1-310-206-3987.
E-mail address: lkavanau@biology.ucla.edu
Contents
1. Introduction
2. Evolutionary origin of sleep
2.1. Schwartzs admonition and Rauscheckers Fundamental Dogma
2.2. Restful waking, detailed focal vision, and conicting brain activities
2.3. Primitive sleep obviated potentially conicting brain activities
2.4. Vertebrates that never sleep
2.5. Need for sleep in congenitally and adventitiously blind mammals
2.6. Earliest sleep
2.7. Complex retinal processing
3. Warm-bloodedness and selective pressures for REM and NREM sleep
3.1. Historical
3.2. Sleep was non-uniform before the origin of REM and NREM states
3.3. Presence and absence of thermoregulation during sleep states
3.4. Implications for the evolution of sleep states
3.5. Inuences of ambient temperature
3.6. Actions of brain waves during waking
3.7. Actions of brain waves during sleep
3.8. Long-term memory maintenance
3.9. Basic links between muscle-controlling and visual circuits
3.10. Absence or lesser amounts of REM sleep in marine mammals and birds
3.11. Origin of fast waves of REM sleep
4. Evolutionary origin of sleep within a brain-wave paradigm
5. Sequential cycling of NREM and REM sleep
6. Overview
Advances in Cell Aging and Gerontology, vol. 17, 132

2005 Elsevier B.V. All Rights Reserved.
DOI: 10.1016/S1566-3124(05)17001-1
2 J. L. Kavanau
1. Introduction
Darwins discoveries in the eld of evolution did much more than open our eyes
to long-term inuences of the environment on higher plants and animals. They led
to enormously fruitful reorientations of studies in all elds of biology. Viewing
biological phenomena from an evolutionary perspective frequently yields insights
beyond those that can otherwise be discerned. However, paraphrasing Damasios
(1999) remark: one could almost say that, until the last decade, neuroscience and
cognitive science have proceeded as if Darwin never existed. In this chapter, I employ
a Darwinian approach to the origin and functions of sleep and sleep states.
Studies with the goal of identifying selective pressures that might have given
rise to sleep and its rapid-eye-movement (REM) and non-rapid-eye-movement
(NREM) states, could cast light on their functions. Being mostly medically oriented,
current studies of sleep have somewhat dierent objectives. Attention is directed
primarily to mechanisms, for example, inuences of various brain secretions on
vigilance states. While signicant medical progress, with great practical benets,
has been made along these lines, implications of the underlying mechanisms
for sleeps functions have been tenuous or non-existent. Further, one must
distinguish between the functions of sleep and the activities that occur during
sleep, and their benets. Sometimes, the functions have been equated with the
activities and benets.
Some treatments included evolutionary considerations, but they infrequently
probed underlying selective pressures. Almost all attention has been directed toward
comparative aspects, such as the degrees to which sleep and its REM and NREM
states occur in various species, and how they correlate with brain structures,
anatomy, behavior, and ecology. These analyses inevitably lean heavily on
information gleaned from evolutionary endpoints, perhaps millions of years
removed from the selective pressures of origin, and often arrived at along dierent
routes. Taken at face value, these endpoints can be misleading.
Even before probing for the basic or primitive function(s) of sleep, the
evolutionist anticipates that it will be expressible in terms of maintenance of an
overall high eciency of brain operation. No one would doubt that normal
waking brains operate at high levels of eciency, with responses to threatening
circumstances having the highest priority. Accordingly, the provision of a suitably
modied alternative vigilance state sleep in some animals, very likely functions
to maintain brain operations at overall high levels of eciency, by subsuming
non-urgent activities that cannot be accomplished eciently during a continuous
waking state.
2. Evolutionary origin of sleep
2.1. Schwartzs admonition and Rauscheckers Fundamental Dogma
In addressing the basic function of sleep referred to by some as the supreme

mystery namely, the function that provided the evolutionary selective pressure for
Evolutionary Aspects of Sleep and Its REM and NREM States 3
sleeps origin, it is essential to heed Schwartzs (1997) admonition to sleep

researchers. Certain phenomena that characterize a given state, . . . may mirror
aspects of the mechanism for generating the state and its attendant phenomenology
rather than the function of the state (original italics). This caution is highly relevant
for sleep studies, because some researchers assume that a sucient knowledge
of the neurological and physiological mechanisms that initiate, maintain, and
terminate sleep, will reveal sleeps basic function sometimes enunciated as
function and mechanism questions ultimately merge. However, the proposed basic
function discussed below is unrelated to sleeps neurological (brainstem activating
systems) and physiological control mechanisms (see Kavanau, 1997). The latter are
manifested chiey by cyclic alterations in brain neuromodulatory secretions (such as
acetylcholine and serotonin) (see Stickgold, 1998) during the four principal vigilance
states, that is, REM and NREM sleep, and active and restful wakefulness.
It is a property of brain evolution, driven by the adaptive advantages of eciency,
that any given region of the brain typically carries out more than one function. This
adaptation largely underlies the fundamental dogma of neuroscience, as
characterized by Rauschecker (1995), namely, that long-term memories are stored
by means of synaptic modications in the same distributed assembly of brain
structures that process and analyze the events and relations to be remembered
(Squire, 1986; Ungerleider, 1995). In a striking example of such circuit
multifunctionality, one not only becomes blind to colors (achromatopsia), after
severe damage to brain regions that process colors, some of those so aicted do not
even remember that colors exist (Zeki, 1999).
2.2. Restful waking, detailed focal vision, and conflicting brain activities
Restful waking, including readiness, undoubtedly was the phylogenetically

antecedent vigilance state to primitive sleep (Kavanau, 1997). During restful waking,
the brain continues to receive and process sensory inputs, and full vigilance is
maintained, but there is no voluntary activity. Absence of voluntary activity
amounts to reduced interference with other activities of the waking brain, of which
sensory processing is a major component. Even when an activating stimulus is
received during restful waking, it merely initiates a state of readiness, with lowered
sensory thresholds. Restful waking evolved in many animals that previously were
active continuously. For greatest eectiveness, periods of restful waking likely were
relatively lengthy and spent in safe retreats. Inasmuch as daily lightdark cycles
provide natural constraining guides for alternate periods of activity and inactivity, at
its inception, restful waking, in any given species, probably was channeled largely
into lengthy nocturnal or diurnal periods.
Vertebrates would already have been engaging in daily cycles of activity and
restful waking when selective pressures for primitive sleep arose. Evolutionary
progression toward primitive sleep would have begun when animals with
comparatively simple lifestyles evolved increasing complexity and detailed focal
vision (vision that recreates a complex scene). Such vision requires enormous
amounts of neural processing (Llinas and Pare, 1991) the combining of a very great
4 J. L. Kavanau
number of incredibly specic bits and pieces of visual features, formed by

decomposition processes that begin in the photoreceptor cells (rods and cones). This
combining involves vastly more, and more complex, operations than are required by
other sensory modalities. Despite its complexity and enormous requirements, visual
processing is carried out largely at a low level, without visual attention (but not
without potential interference with other waking brain activities).
In the macaque monkey (Macaca fascicularis), for example, a complex mosaic
covering roughly 60% of the brains surface regions is involved in visual processing
(Van Essen et al., 1992). The mammalian visual system is said to have a nearly
innite capacity to recognize patterns and objects. Without focusing attention on
any specic region of a scene, one becomes aware of the space-lling presence of
almost limitless numbers of objects, of all sizes, shapes, and colors, in all imaginable
relationships with everything in view simultaneously, and despite, or perhaps
facilitated (Ross and Ma-Wyatt, 2004) by, saccades at 3/s. In the ancient times
when such detailed focal vision was being acquired, animals would have been
engaging increasingly in multifarious activities and wide-ranging movements,
frequently exposed to new visual experiences. In such a lifestyle, lifelong retention of
greatly increased stores of memories would have been crucial.
It has been postulated that, in animals achieving detailed focal vision, with the
accompanying needs for large stores of memories, the parallel-processing capacity of
the brain was becoming excessively taxed, because of developing conicts between
enormous and urgent demands of complex visual analysis and split-second control of
movements, on the one hand, and non-urgent learning and memory processing, on
the other (Kavanau, 1997). The brains of these animals could no longer meet crucial,
largely unpredictable, hazards and routine needs, while at the same time, meeting
needs to acquire, establish (consolidate), and reinforce large stores of long-term
memories, with all neural activities in given categories sharing circuitry in
corresponding dedicated brain regions. In other words, an adaptation (circuit
multifunctionality) that had conferred great eciency before the evolution of
detailed focal vision, would have become increasingly maladaptive as a more
complex visual lifestyle evolved, had not compensating adaptations evolved in
parallel, namely, restful waking at rst, followed by primitive sleep.
2.3. Primitive sleep obviated potentially conflicting brain activities
The selective pressure for the origin of primitive sleep may have been the need to
ameliorate the developing conicts discussed earlier, by achieving a more profound
state of brain unresponsiveness to external occurrences during memory processing
than exists during restful waking. By relieving the brain of extensive needs to
process and respond to environmental, chiey detailed visual, inputs during a
portion of the 24-h cycle, memory processing could have proceeded without
impediment during that portion. As a result, those mnemonic activities that could
be delayed with minimal survival risk, such as establishing new memories
and reinforcing existing long-term memories, came to be carried out during the
new portion, namely, primitive sleep (Kavanau, 1997). It is a reasonable assumption
that sleep in present-day reptiles has changed very little from the primitive sleep of
their ancestors.
In line with the above proposals, the reason why invertebrates lacking
image-forming eyes, such as many mollusks, echinoderms, worms, and the like,
need no sleep, is because their sensory input processing during activity and
restful waking interferes minimally, or not at all, with other vital brain activities.
As indicated above, these would be largely memory consolidation and processing,
and other maintenance activities. In other words, because processing of non-visual
sensory information engages much less brain tissue, and in a very much simpler
manner than detailed visual input, it does not come into conict with other vital
brain activities. In this regard, with eyelids closed, or under anesthesia, region 18
of the feline visual cortex shows intrinsic patterns of EEG activity resembling
those produced by certain visual stimuli (see Ringach, 2003). These patterns
presumably reect ongoing unimpeded reinforcement of component circuits of visual
memories.
The key, overt, adaptive changes that accompanied selection for primitive sleep
probably were: (a) to close eyelids that previously were transparent and purely
protective, but were in the process of becoming increasingly opaque; and/or (b) to
retire to secure quarters, often in dim light or darkness. With the exclusion of light,
and without a need to process detailed visual inputs, and with correspondingly
decreased attentiveness to other sensory inputs, the sleeping brain would have been
almost totally occupied with some of those previous neural activities of waking that
could be delayed with minimal risk.
That the primitive sleep state subjected animals to greater risks than spending the
same periods awake, would not have been determining. The critical factor vis-a`-vis
natural selection would have been, whether primitive sleeps adaptive advantages
outweighed the greater risks entailed. The proposed adaptive advantages were the
maintenance of great eciency of brain function, both awake and asleep highly
ecient sensory processing and responding when awake, and highly eective
memory processing when asleep. Primitive sleep would have compensated for
a gradually developing potential conict, namely, the need to accomplish all
these interrelated neural activities in partially shared circuitry during a continuous
waking state.
In this connection, Moorcroft (1989) made a general suggestion similar to that
proposed here and earlier (Kavanau, 1996, 1997), namely, that sleep provides a
period when certain activities can occur most easily and most eciently. In a
similar vein, Maquet (2001) suggested that [s]leep could be a privileged period for
memory consolidation. . . . Some mental dysfunctions may exert their inuences
through reducing the eciency of brain operations. Thus, the cerebellum may
expedite the automation of motor and cognitive skills. Rutherford (2003) proposes
that cerebellar dysfunction slows or prevents this automation. Either circumstance
could take a toll on cerebral performance, aecting connections between the senses
and physical functions as well as the ability to organize, create, and complete
thoughts and tasks. This certainly seems to be the case for the cognitive and motor
functioning of patients who have cerebellar dysfunctions.
6 J. L. Kavanau
In incipient stages of sleeps evolution, the brain functions that now occur
independently during waking and sleep probably occurred competitively. However,
compelling evidence indicates that the conicting needs now have become
incompatible during wakefulness. Some memory consolidation involving detailed
visual discriminations cannot occur during waking. It absolutely requires sleep
(Gais et al., 2000; Stickgold et al., 2000). This unequivocal establishment of a
specic vision-related activity that requires sleep lends strong support to the
foregoing paradigm for the role of detailed focal vision in sleeps origin. This
paradigm already is persuasively supported by ndings that sleep occurs only
in animals with complex image-forming eyes, as opposed to eye spots and
light-sensitive pigment cups and tubes, that to sleep, many animals must block
visual input, by closing their eyelids or other means, and that, in some
animals, unihemispheric sleep automatically ensues when one eyelid is closed (see
Kavanau, 1997).
One should not lose sight of the circumstance that, despite the potential neural
processing conicts referred to earlier, continuous dynamic interactions of sensory
inputs with ongoing neural activities are intrinsic to waking brain function. Even
at its most basic levels the central nervous system (CNS) is not organized to yield
particular responses to particular stimuli, but instead, to accomplish particular
objectives. Rather than mirroring the external world, the CNS embodies a dialog
between internal states, generated by intrinsic electrical activity of nerve cells and
their connectivity, and sensory information. With respect to the visual input, which
is of greatest interest here, activity in the feline cortex depends not only on the
nature of a visual stimulus, but also on the cortical state at the time of stimulation
(see Ringach, 2003). Sensory stimuli gain their signicance by virtue of triggering a
preexisting disposition of the brain to be active in a particular way. If a stimulus is
not put in the context of thalamocortical reality by becoming correlated temporally
with ongoing neural activity, it does not exist as a functionally meaningful event
(Llinas and Pare, 1991; additional Refs. in Kavanau, 1997).
This proposal for the origin and function of primitive sleep does not
preclude subsequent or concomitant evolution of secondary functions that may
have become essential. Indeed, for birds and most mammals, secondary functions
of sleep come into play, some of which are discussed below, as well as deep-
seated rhythmical changes that engage many physiological systems (Vertes, 1990;
Everson, 1995).
2.4. Vertebrates that never sleep
The lifestyles of vertebrates that never sleep are fully consonant with the above
proposals. As would be expected, since genetically blind, cold-blooded vertebrates
that live in caves have no visual input, there can be no visual processing conict,
so no sleep is needed. However, many continuously swimming shes that possess
detailed focal vision, such as tunas and many sharks, do not sleep, either. Their
lack of a need for sleep can be attributed to their lifestyle, in which needs
for processing both sensory information, predominantly visual, and long-term
experiential memories, are greatly reduced and, therefore, do not conict with each
other. These reductions owe to the following aspects of shark and tuna behavior
and ecology: (1) their visual input is greatly reduced or absent during lengthy periods
of both diurnal and nocturnal activity; (2) schooling greatly reduces needs for
environmental sensory information, particularly visual; (3) their circuitry for most
inherited memories needs no supplemental reinforcement, because it is maintained
through almost continuous use; and (4) because they lead a comparatively routine,
monotonous existence in essentially featureless, open waters, they acquire, and have
need to reinforce, relatively few experiential memories. Analogous circumstances
could account for the ability of migratory birds to y for days without rest or sleep
(Kavanau, 1998a, 2001a).
2.5. Need for sleep in congenitally and adventitiously blind mammals
In view of the proposed role of detailed focal vision in the evolution of sleep,
some comments are in order concerning the need for sleep in congenitally and
adventitiously blind mammals. The resolution of the seeming disparity hinges on
three principal factors: (a) the same amount of visual cortex, in need of extensive
reinforcement, exists in both sighted and blind animals; (b) the high metabolic rates
of endothermy (loosely speaking, warm-bloodedness), with an accompanying
relatively high rate of degradation and functional depletion of molecules essential
for synaptic function, and a correspondingly more frequent need for replenishment
of these molecules by reinforcement during sleep; and (c) the great plasticity and
high adaptability of visual cortices which, when unused for vision, take over the
implementation of other sensory modalities (Rauschecker, 1995). Secondary roles of
sleep rest, rejuvenation, etc. far removed from the functions of origin, also come
into play.
Concerning item (a), many neurological and neurophysiological studies in
non-genetically blind humans and/or cats have given uniform results. Although
the optic nerve, optic chiasm, and lateral geniculate nucleus may degenerate,
neocortical visual regions undergo no size reduction and show no evidence of
organic change. They remain highly active, metabolically and electrically, with
highest activity in the striate and prestriate regions (Wanet-Defalque et al., 1988;
Yaca et al., 1999; additional Refs. in Kavanau, 2001b). Moreover, in monkeys
and cats visually deprived since birth, the electrical activity that develops spon-
taneously in neurons of visual regions resembles that in non-deprived animals
(Roder et al., 1997).
The observations of item (a) probably are accounted for partly by the
development of cross-modal compensatory plasticity encompassed under item (c).
Evidence has been mounting of a great potential for such plasticity in vertebrate
visual neocortical regions, many of which take on supplemental auditory and tactile
functions. Many visual regions also normally process auditory and somatosensory
inputs (Rolls, 1991; Van Essen et al., 1992). These circumstances are well exemplied
by human cortical visual regions. Thus, the level of activity in the primary and
secondary visual cortices of congenitally and adventitiously blind subjects during
8 J. L. Kavanau
various auditory and tactile tasks is higher than in sighted, blindfolded controls
(Sadato et al., 1993), including the activation of cortical visual associative areas of
congenitally blind subjects during auditory localization tasks (Aziz-Sultan
et al., 1997).
Additionally, cross-modal compensatory plasticity is known in ferrets, Syrian
hamsters, rats, and blind mole rats (Rauschecker, 1991; Roe et al., 1992; Doron
and Wollberg, 1994). To these ndings, one can add the responsiveness of all
neurons in some regions of the visual cortices of cats to auditory stimuli, and
improved auditory localization by visually deprived animals (Diamond, 1982;
Rauschecker, 1991; Van Essen et al., 1992; Yaca et al., 1999). In essence, loss of
visual input to mammalian visual cortical regions does not lead to their inactivity
or size reduction; rather, the regions take on highly active roles in the service of
other sensory modalities (see also, Kavanau, 1999, 2001b).
2.6. Earliest sleep
With the knowledge that the occurrence of sleep, or lack thereof, in shes is
dependent upon their behavior and ecology, and that marine vertebrates preceded
land vertebrates, it is likely that the earliest vertebrate sleep occurred in shes.
Prominent among modern-day shes with detailed focal vision that engage in sleep
are many teleosts (bony shes) that occupy complex coral reef habitats. Teleosts,
however, date back only about 235 My (million years) to the early Triassic, so
the earliest primitive sleep in shes might have originated much earlier, say in
small, shallow-water or reef sharks the earliest jawed shes living in complex
habitats, during or more recently than the Ordovician period, about 450 My ago
(Kavanau, 1998a).
Sleep in invertebrates may have an even more ancient origin. Many trilobites,
which existed from the early Cambrian (543 My ago) to the late Permian (248
My ago), had advanced visual systems, mostly paired compound eyes. In many
cases, a huge variation in size and form of the eyes with eyes secondarily lost
in bottom dwellers at great depths seemingly was related to their behavior and
ecology. In the free-swimming trilobite, Opipeuter, the large eyes dominated the
head (cephalon), providing a 360 eld of vision (Levi-Setti, 1993). The lifestyles
of some of these trilobites may have been suciently complex to require sleep.
Among mollusks, sleep is known only in cephalopods, probably the most
intelligent invertebrates, many of which have large camera-type eyes. Ancestors of
the chambered, primitive, external-shelled Nautilus, date back to the Silurian
period (435 My ago), when they were dominant and a top predator. The nautilus is
still a major predator of crustaceans in the deep sea. Its brain is very complex,
though less specialized than in other cephalopods. The modern nautilus very likely
sleeps. It is inert and withdrawn in its shell during inactivity, with respiratory
movements scarcely perceptible (see Kavanau, 1998b). As with the trilobites,
lifestyles of some ancestral nautiluses also may have been suciently complex to
require sleep.
2.7. Complex retinal processing
As with most biological phenomena, there are complicating circumstances.

Retinal processing of visual information, at some level, supplements central
processing in virtually all sighted vertebrates. The more we learn, the greater the
active role the retina appears to assume in perception. However, varying degrees
of specicity, integrative capacity, and structural complexity exist in dierent
retinas, together with varying degrees of lesser capacity for visual processing in the
rest of the brain. Inasmuch as some aspects of visual information are processed in
the retina (which is a part of the brain), the amount needing processing in central
brain regions is reduced. Since retinas are far from central regions, where most
memory reinforcements take place, central reinforcements can occur with little or no
conict. Indeed, in some vertebrates employing this strategy, retinas have been
described as true peripheral brains (Jerison, 1973).
In its original function or adaptive value, complex retinal processing probably
reduced response delays in avoiding predators and capturing prey. Such reductions
would have been facilitated to a considerable degree by analysis of critical
components of visual information immediately upon receipt in the retinas namely,
those components that allow the detection and evaluation of sources of potentially
threatening movements. Thus, honing of the image dening precise edges, and
focusing on ne details starts at the rst retinal synaptic level. This is not to imply
that central brain regions do not also play vital roles in generating responses to
visual input. At the very least, though, there is less redundancy in the information
that is conveyed by the optic nerves (of which there are more than a million in
primates) from the retinas to central brain regions. Response eciency is
correspondingly increased.
By reducing the amount of central processing of visual input, complex retinal
processing might well have delayed a need for sleep in ancestral cold-blooded
vertebrates possibly in those small Ordovician sharks mentioned before. In turn,
this would have delayed the build-up of conicts with memory circuit reinforcement
during restful waking and activity. However, even if such a sleep-delaying
inuence occurred, it usually would have lost its eectiveness as evolving brains
became increasingly complex. It now may eliminate a need for sleep, but not
restful waking, in only a few non-marine vertebrates and, presumably, in their
close relatives. These are the bullfrog (Rana catesbiana), the sea turtle (Caretta
caretta), the tortoise (Testudo denticulata), and the American alligator (Alligator
mississipiensis) (Refs. in Campbell and Tobler, 1984).
Whereas most mammals have a greatly reduced dependence on complex retinal
processing, it plays a major role in the visually highly procient, fast-moving birds
whose brains and faculties are specialized for high-speed decision making.
The tremendous information-processing capacities of mammals, that make possible
intelligent decisions, confer no advantage to birds, in critical, life-threatening
situations that require decision-making at speeds too great for the use of such
capacities (Kavanau, 1987). The strong selections for both reduced response delays
and increased response speeds in birds, not only led to retention of a high degree of
10 J. L. Kavanau
complex retinal processing, it also led to the most rapid neural transmission
(achieved in large, myelin-coated nerves), and the highest tolerable core temperatures
(nerve conduction speed increases with temperature, which is up to 4 C greater
during ight than rest). Modern birds possess a highly procient, very fast-acting,
essentially superreptilian visual system (Refs. in Kavanau, 1987).
A relatively high degree of complex retinal processing has persisted or been
reestablished in only a few species of mammals: it is known in the rabbit (Lepus
cuniculus); ground squirrel (Citellus mexicanus); rat (Rattus norvegicus); and in
the domestic cat (Felis catus), and surely exists in some related forms (Refs. in
Kavanau, 1997). While it has become a rule that characters of living forms, once lost,
are not regained, properties of eyes are among the notable exceptions.
The adaptations of vertebrate eyes for diurnality and nocturnality are thought to
. . . come and go in evolution as mutatory capacity and ecological expedient direct
(Walls, 1967). Even avian ight has been lost and regained (Paul, 2002; Kavanau,
in preparation).
3. Warm-bloodedness and selective pressures for REM and NREM sleep
3.1. Historical
An early attempt to correlate the evolution of sleep states with warm-bloodedness

was made by Allison and Van Twyver (1970). They proposed that NREM sleep
arose roughly 180 My ago in monotremes (egg-laying mammals), presently
represented by the echidna and platypus. REM sleep sleep researchers proverbial
riddle wrapped in a mystery inside an enigma was thought to have evolved some
50 My later, before the ancestral branching from pantotheres into marsupials and
placentals in the early Cretaceous. On the contrary, Walker and Berger (1980)
assumed that REM sleep was primitive in cold-blooded vertebrates, based entirely
upon the supercial resemblance of the absence of control of body temperature.
They proposed that NREM sleep and warm-bloodedness evolved together in
mammals millions of years later.
Siegel et al. (1996) proposed that . . . since the divergence of monotreme. . . and
nonmonotreme mammals 130 My ago there has been a dierentiation of this
primordial sleep state into two states [REM and NREM]. . . . Subsequently, they
(Siegel et al., 1998) echoed the WalkerBerger paradigm, suggesting that REM
sleep. . .may have had its origin in reptilian species. The above proposals were not
based on functional analyses or evolutionary tenets. Rather, they were arrived at by
according great weight to comparative aspects of todays evolutionary endpoints,
the anatomical, behavioral, physiological, and ecological characteristics, as they
occur across the spectrum of mammalian species.
3.2. Sleep was non-uniform before the origin of REM and NREM states
According to the above proposals, an ancient sleep state in one scenario REM
sleep, in another NREM sleep, in a third, primordial sleep evolved to include
both REM and NREM states. In considering the feasibility of these scenarios, an
answer was sought to the following general question. How could selective pressures
acting on animals engaged in a uniform condition of sleep have led to its evolution
into two dierent states? Such a result, it was concluded, could not have been
produced by selective pressures, alone. The very beginning and end of a uniform
sleep period might have undergone dierential selection because, as times of
opposite transitions, they dier initially from each other. But the period of sleep
itself, could not have been selected for dierentially unless it were adaptively non-
uniform. In other words, if no part of the period of primitive sleep were more
adaptive than any other part, there could have been no selection.
Since any initial departure from an adaptively uniform period of primitive sleep
could not have arisen solely through the actions of selective pressures, dierences
in sleeps adaptedness must have been caused initially by some other means.
Environmental inuences are the only feasible candidates. Once such dierences
arose, selection could have acted. Accordingly, it was concluded that primitive sleep
in evolving mammalian ancestors must have been adaptively non-uniform before
the origin of the REM and NREM states.
One environmental variable stood out as most likely to have been responsible for
a non-uniformity of primitive sleep. Knowledge of it was a guide to how its inuence
might have aected the physiology of sleeping mammalian ancestors evolving warm-
bloodedness. In turn, this led to proposals for a chain of other evolutionary
adaptations. The further topics addressed include the origin of the REM and NREM
states from sleep that was already non-uniform, how these sleep states appear to
relate to warm-bloodedness, major actions that are believed to take place during
sleep states, how these actions may contribute to the maintenance of long-term
memories, roles of REM sleep in reinforcing memory circuits for muscle
contractions and vision, and a basis for several cycles of the NREM and REM
states occurring during a nights sleep. For expository convenience, the modes of
origin of the three salient properties of REM sleep are dealt with in sequence.
However, these properties are revealed to be causally related, and are expected to
have evolved in parallel.
3.3. Presence and absence of thermoregulation during sleep states
Ambient temperature probably was responsible for an adaptively non-uniform

condition of sleep in animals evolving warm-bloodedness. And it surely is no
coincidence that a basic dierence between REM and NREM sleep, as well as
between cold-bloodedness (ectothermy) and warm-bloodedness, is whether or not
internal control of body temperature occurs, that is, whether it is actively maintained
at a constant, elevated level by physiological and behavioral means. With this
further realization, a substantial foundation was laid for elucidating the path of
evolution of the REM and NREM sleep states.
Further progress followed from considering the status of internal body
temperature control during sleep states of adult mammals, and how this control
and the sleep states, themselves, are inuenced by ambient temperature. During
12 J. L. Kavanau
NREM sleep and waking, adult mammals maintain their deep-body temperature
close ( 2 C) to a species-specic level, known as the core value. The regulatory
system behind this thermostatic (set-point) control must have deep roots, because
local heat or cold application at virtually any site in the CNS, from the spinal cord to
the cortex, initiates widespread, coordinated responses to restore the core value
(Gordon and Heath, 1986).
On the contrary, during REM sleep, when much of the electrophysiology of
NREM sleep has become altered, set-point control of body temperature, in response
to ambient temperatures outside the core-value range, does not occur (Parmeggiani,
1980). Although some actions to initiate temperature regulation might take place,
just as during NREM sleep and waking, the skeletal muscles employed in such
regulation are atonic (lacking tone) and unable to respond. Moreover, the
responsiveness of preoptic hypothalamic temperature-sensitive neurons reaches its
lowest level then (Parmeggiani, 1980). But autonomic processes involving smooth
muscles and other cellular basal metabolic processes continue, as during NREM
sleep and waking. Some control of body temperature by sweat-gland activity can
occur during REM sleep, in response to ambient temperature changes, but the
control is less sensitive than during NREM sleep (Bach et al., 2002).
Adult mammals also usually pass from one sleep state to the other, following
certain changes in ambient temperature. When the ambient value is at and near the
core value, REM sleep prevails, without set-point temperature control. But at
ambient values above or below the core value, NREM sleep ensues, with body
temperature controlled (Parmeggiani, 1980).
3.4. Implications for the evolution of sleep states
These thermoregulatory relationships have signicant implications for the

evolution of sleep states in mammalian ancestors evolving warm-bloodedness.
Since environmental temperatures over the course of the daily 24-h cycle are not
uniform, neither could the sleep of our ancient ancestors exposed to such changing
temperatures been uniform, a specic accounting for which is elucidated in the
following. Only for animals sleeping in caves, deep in burrows, or at aquatic depths,
with ambient temperature unvarying, could a lengthy period of early sleep have
been uniform. Accordingly, if the primitive ancestors of mammals evolving warm-
bloodedness were exposed to dierent ranges of ambient temperatures while they
slept, these dierent exposures could have provided the basis for the evolution of
a non-uniform condition of sleep, laying the foundation for the origin of the REM
and NREM states.
Before venturing to decide whether these ancestors, dating back roughly 220 My,
were exposed to ambient temperatures during sleep, we turn to the knowledge of
their behavior and ecology. Although for many My they had been diurnal, predation
pressure from the keen-sighted, bipedal, more agile and mobile dinosaurs newly
arrived on the scene but achieving dominance over all terrestrial vertebrates within a
span of 12 My had forced them to conne their activities to the night, but not
precluding dim portions of dawn and dusk. Though imposed at rst, this lifestyle,
which may have lasted 140 My an incomprehensible time span for us proved to
be a boon. It was favored by the uniquely warmer, on average, moist and moderate,
greenhouse, mid-to-late Triassic climates (240200 My ago), and the absence of
reptilian competitors in insect-rich, nocturnal niches (Crompton et al., 1978). This
very eective ecological partitioning allowed our ancestors to ourish as
contemporaries of the dinosaurs, rather than their competitors. Elsewhere,
prevailing views of this diurnal-to-nocturnal evolutionary transition, sometimes
referred to colorfully as the bottleneck theory, have been reviewed (Kavanau,
1997, 2002a).
During this lengthy nocturnal sojourn, mammalian ancestors became dramati-
cally reduced in size. They evolved from large (up to 250 kg), day-active,
carnivorous, advanced, mammal-like reptiles (therapsids), that for several My
were dominant, and the largest terrestrial carnivores. Most of these mammal-like
reptiles became extinct at the end of the Permian period. Survivors evolved
into relatively inconspicuous, nocturnal, terrestrial-arboreal, shrew-to-rat sized
(2040 g), mammalian insectivores, a representative of which is depicted in Fig. 1.
Fig. 1. Artists rendition of a representative nocturnal, insectivorous, primitive mammal from the late
Triassic, about 200 My ago. By Hermine Kavanau, after a sketch of the animals form by Crompton et al.
(1978).
14 J. L. Kavanau
Warm-bloodedness, coupled with great agility in terrestrial-arboreal habitats,

permitted these small, nocturnal ancestors to exploit prey beyond the capacity of
their amphibian competitors. At rst, these comprised only frogs and toads, later
(mid Jurassic) they included salamanders and newts.
These late Triassic, insectivorous mammals rank among the smallest in
mammalian history, belonging to such unfamiliar groups as triconodonts,
symmetrodonts, conodunts, and haramyids. Paralleling, and intimately related to
the decrease in size, was the evolution, or merely further progression, of warm-
bloodedness (McNab, 1978). Further progression cannot be ruled out entirely,
because diurnal, advanced, cynodont-therapsids might possibly have achieved a
primitive grade of warm-bloodedness (Bennett and Ruben, 1986) before they
retreated into nocturnal niches.
A 3- to 4-fold relative increase in brain size occurred while in nocturnal niches,
tracing to an expansion and extensive recomposition of the reptilian forebrain,
culminating in the emergence of a neocortex. These changes accompanied, and
probably owed, not to greater intelligence, but to rened skeletomuscular
coordination in a terrestrial-arboreal habitat, and improvements in the olfactory,
auditory, and tactile senses, that are highly adaptive for a nocturnal existence
(Crompton and Jenkins, 1978). As early as the late Triassic period, these adap-
tations are believed to have compensated for the extensive restriction of visual
information at night (Jerison, 1973).
In subsequent visual system evolution, some of the unique specializations of
the mammalian retina arose as further adaptations to a continuing nocturnal
existence. The fact that mammals are the only major vertebrate group to generally
lack color vision, also harks back to that multi-My period of colorless nocturnal
activity (Walls, 1967). Aside from such adaptations, the Mesozoic era seems to
have been, comparatively speaking, a time of quiet evolutionary conservatism
for mammals in nocturnal habitats. Thus, they were much less diverse than their
post-Mesozoic descendants in less restrictive diurnal habitats, despite having existed
for more than twice as long.
Most pertinent to the present considerations, it is very likely that these relatively
tiny ancestors were exposed to changing ambient temperatures while they slept.
They did not live in caves or at aquatic depths, and there is no evidence for sleeping
in burrows, though use of shade and dense, secluded cover would have been
obligatory. Also, their teeth, legs, and prehensile feet were not adapted for digging.
The latter were specialized for rapid locomotion over rough, uneven ground,
for maintaining stability on steep inclines, and for climbing in shrubs and
trees (Crompton et al., 1978), adaptations that also would have facilitated the
manipulation of plant nesting materials.
The prevailing moderate, appreciably warmer climates would not have conferred
advantages on, nor selected for, animals that slept in burrows. On the contrary, in
the pronounced wet seasons, with monsoonal precipitation, and a very lengthy
mid and late Carnian (228220 My ago) pluvial period, burrows and their
inhabitants would have been highly vulnerable to ooding and drowning. Indeed,
such mishaps befell small, early-Triassic, burrowing therapsids, Trirachodon, that
had lived millions of years earlier (Groenewald et al., 2001). However, being
herbivores rather than insectivores, Trirachodon were not in the mammalian lineage.
Just as today, ancient ancestors of mammals probably were adept at nding or
constructing, and inhabiting, nests and other secluded cover that were inconspicuous
and/or inaccessible to potential predators, and provided a favorable milieu during
inhospitable external conditions (Lillegraven et al., 1979).
Before treating the circumstances postulated to have led to the origin of REM
and NREM sleep, some additional background information is desirable. Reptiles
get most of their bodily heat directly from the sun, and the temperature of all
except the largest of them tends to track the ambient value. While largest reptiles
achieve some degree of independence from ambient temperatures, most small
representatives are lethargic, or even helpless, in the cold of the night, and must
await being warmed by the sun to become active. Even during activity, however,
reptiles rely on a non-sustainable, anaerobic metabolism for almost all activities,
and they generally fatigue rapidly as a result of cellular lactic acid accumulation.
Mammals and birds, on the other hand, raise their body temperature mostly by
producing heat internally through prodigious rates of cellular oxygen consumption
and, as noted earlier, usually maintain a constant, elevated, species-specic core
value. This often requires bodily heating at some times, and cooling at others, and is
very costly energetically. The greater the skin, deep-body, or brain temperatures
deviate from their set-point values, the greater the magnitude of the heating and
cooling recovery eorts referred to as proportional control (Gordon and Heath,
1986). This ability to control body temperature, and an accompanying increased
aerobic scope for metabolism, confer great advantages. Animals so endowed usually
can sustain relatively high levels of activity full muscular power, with fully
functional sensory faculties in hunting or escaping for lengthy periods at any time
of the day or night, and in all seasons. They also can forage widely and undertake
long migrations.
The core temperatures of living primitive mammals, such as monotremes, are
in the range 3032 C, those of marsupials (giving pouch birth) in the range
3436.5 C, and those of most other mammals in the range 3638 C (Dawson, 1973).
All groups, however, are considered to be well adapted, because a suciently
high core value, and an accompanying nely tuned chemostatic system, have
great adaptive value. The specic values of the appropriate core temperatures
in mammals are greatly dependent on the niche occupied. The core value of the
shrew-to-rat sized ancestors of mammals, conned to nocturnal activity for tens of
My, is believed to have attained 2831 C, but not necessarily entailing an increase in
resting metabolic rate. Only after dinosaurs became extinct during the late
Cretaceous (9565 My ago), were many mammals able to resume daytime activity,
probably several times independently, and achieve todays higher core temperatures
and much greater sizes (Heath, 1968; Crompton et al., 1978).
By then, there had been considerable evolutionary radiation among marsupials
and placentals. Some descendants, including most families of insectivores, remained
in their more conservative nocturnal niches, and retained ancestral, reptilian, low
metabolic rates. Indeed, most shrews continue to rely more on their senses of
16 J. L. Kavanau
hearing, smell, and touch than on vision (Lillegraven et al., 1979). Additionally, eyes
of monotremes have retained their reptilian organization (Walls, 1967). Among all
living mammals, a marsupial, the common opossum, Didelphis virginiana,
most merits the title living fossil, with a skull reminiscent of a Triassic
mammal-like reptile, and a typically reptilian threat posture and manner of attack.
(MacLean, 1986).
3.5. Influences of ambient temperature
This brings us to a signicant juncture for the primitive ancestors of mammals

during evolution of warm-bloodedness in mid-to-late Triassic climates. These
ancestors probably were exposed to average ambient temperatures above their
core values of only 2831 C during most or all of daytime sleep requiring bodily
cooling; to average temperatures below core values during most or all of nocturnal
activity requiring bodily heating; and to temperatures traversing core values
while asleep during, or before or after, changes between day and night requiring
neither heating nor cooling. With tens of My in uniquely favorable climates,
traversals of core values by ambient temperatures during some times of sleep
would have been almost a certainty during one or another evolutionarily lengthy
time spans, in the mild winter and/or summer seasons.
Guided by temperature-regulating responses of living mammals which, as noted
earlier, probably are deep-rooted, and need to conserve energy, which always
is critical, the following occurrences were likely. Mechanisms of set-point
temperature-control in these daytime-sleeping ancestors, even in the early stages of
evolution of warm-bloodedness, would have become relaxed or disabled when
ambient temperatures traversed the core-value range. In this range, there would
have been no need to control body temperature. If this relaxed or disabled condition
were adaptive, reasons for which will be adduced in the following, sleep would
have come to exist in two states, just as today, one with, the other without, set-
point control of body temperature. This, then, is the specic basis for the earlier
expectation that sleep would have been non-uniform in mammalian ancestors
evolving warm-bloodedness, exposed to daily ambient temperature cycles.
The above scenario assumes that a certain physiological requirement was satised.
As noted, during most or all of the day, when the tiny ancestors of mammals are
thought to have been asleep in nests or shade of other secluded cover, average
microhabitat temperatures probably were above their core values. Accordingly,
these exposed ancestors would have had to maintain their body temperatures in
a core range below ambient values for relatively long periods. The principal way
they could have accomplished this would have been through cooling by evaporative
water loss.
To achieve this would have required such mechanisms as increased respiratory
rate, panting, and unregulated diusion of water through skin and respiratory
passages. Sweating was unlikely because it is weakly developed or absent in
carnivores (Cossins and Bowler, 1987). In terms of water consumption, these
methods can be very costly. The unique mid-to-late Triassic climates (and continuing
favorable climates through the Jurassic and on into the Cretaceous), with monsoonal
precipitation, lengthy Carnian pluvial period, and moderate temperature regimes,
also assume great signicance in this regard. Their eect would have been to relieve
constraints on evaporative cooling and reduce needs for it during extended periods
of daytime sleep, both by providing moist conditions and comparatively moderate
maximum ambient temperatures.
For illustration, some inuences of moderate nest shading on present-day, 24-h
variations in ambient temperature and relative humidity are indicated in Fig. 2,
with sensors for temperature and relative humidity in the nest, that is, in a
partially foliage-shaded, wire-mesh enclosure (Kavanau and Rischer, 1973). Sunrise
began at about 5:40 a.m. with the enclosure temperature at about 15 C. Temperature
increased very gradually for 2 h to about 17.5 C. Thereafter, it increased rapidly
to about 25 C at 9:10 a.m., peaking at about 27 C at 3:10 p.m. After that, it dropped
rapidly to about 20 C at sunset, at about 6:10 p.m. On typical autumnal nights the
drop in enclosure temperature between sunset and sunrise was about 5 C. The
signicance of these present-day examples of weather variables and their changes
during the course of a day are limited in their implications for the warmer, on
average, more equable, Triassic climates. Only the lags in changes between light
level, enclosure temperature, and relative humidity may be roughly indicative for a
moderately sheltered nest.
It can be suggested that: (a) nest temperature after dawn, whatever its absolute
value, would have remained fairly constant, rising at most 2 C in the rst 2 h;
(b) although the greatest heat stress probably would have been in mid afternoon,
nest temperatures would have dropped relatively rapidly in the next 3 h; and (c) the
temperature in the nest probably was higher during late dusk than during early
dawn. Changes in weather variables induced by the eclipse suggest that nest
temperatures might have lagged light-level changes by about 1/2 h, and similarly for
the lag of nest relative humidity values, after changes in nest temperature.
The predominantly opposite changes in relative humidity with temperature largely
reect the relationship that, when the amount of water vapor remains unchanged,
the relative humidity decreases as the temperature rises, and vice versa. Also to be
taken into consideration are the circumstances that multivariable mechanisms
of acclimation would have played a role, possibly including toleration of limited
periods of late-afternoon mild dehydration during sleep.
Many living mammals maintain core temperatures that are well below daytime
ambient values. But this is achieved through various highly specialized anatomical,
physiological, and behavioral adaptations, even in most inhospitable climates. In
Mesozoic times, it was probably mostly the uniquely favorable climates that made it
possible. Further indicative of the key role played by water relations in the evolution
of warm-bloodedness, is a remarkable structural adaptation with beginnings also
thought to trace to Triassic ancestors. This is the development of a series of thin,
often highly convoluted sheets of bone in enlarged nasal passageways the turbinals
or respiratory conchae. These lie directly in the path of respired air, and are essential
for the high lung ventilation rates of warm-bloodedness. They are almost ubiquitous
among mammals and birds, but absent in ectotherms, which possess only the
18 J. L. Kavanau
Fig. 2. Daytime course of meteorological variables in a rooftop enclosure at UCLA, on the day of 60%
solar eclipse (09/11/1969). The recording occurred while studying the activity of an antelope ground
squirrel, Ammospermophilus leucurus, in a running-wheel enclosure. Each dot in the running record gives
instantaneous speed and direction at 2.4-s. intervals. Inuence of the partial eclipse on weather variables
reveals inherent interrelationships. Curve for light intensity is at left, with illuminance levels marked at
40,000 and 100,000 lux (meter-candles). Corresponding twilight curves are at the bottom and top, at
greater magnication, marked at 10-lux values. The curve for temperature within the partially shaded
enclosure is also at left, with corresponding increasing values indicated from left to right along abscissa.
Note the gradual, essentially monotonic, increase in temperature in the morning, the lagging arc of
temperature decrease following the partial eclipse, and the continued, essentially monotonic, increase until
mid afternoon, when it decreases again, as the light level declines rapidly (h indicated at right ordinate).
Percent relative humidity is recorded to left, with values declining rapidly from about 8:00 to 10:00 a.m.,
with the lowest values at about 2:30 p.m., scaled on abscissa from right to left at bottom left. The arc
of increasing humidity lags the arc of causative temperature decrease, following the reduced light level
(from Kavanau and Rischer, 1973).
olfactory counterparts (Hillenius, 1994). They recover much water from expired air,
and would have contributed greatly to both water and heat conservation by
Mesozoic mammalian ancestors during their nocturnal activity periods.
How did natural selection acting on non-uniform sleep, including periods with
and without control of body temperature, lead to the other salient properties of
REM and NREM sleep? In exploring this matter, it is essentially the dierences
between REM and NREM sleep for which an accounting is sought. Inasmuch as
almost all set-point mechanisms of body temperature control directly or indirectly
involve a variety of multifaceted behavioral and autonomic muscle contractile
systems including panting, respiratory rate, piloerection, shivering, licking, and
spreading of saliva the absence of a need for set-point temperature control during
certain periods of sleep in ancestors of mammals evolving warm-bloodedness, would
have translated directly into the absence of a need for almost all skeletal muscle
contractions.
But the very circumstance of skeletal muscles having been unused during certain
daily periods of sleep, if adaptive, would be expected to lead, evolutionarily
speaking, to relaxation and eventual absence of skeletal muscle tone (atonia) during
these periods the putative forerunners of REM sleep. Accordingly, assuming their
adaptedness, two salient properties of REM sleep in adult mammals have been
accounted for, namely, absences of both set-point temperature regulation and
skeletal muscle tone. Accounting for the third salient property, predominance of
fast brain waves ( 14 and up to 800 cycles/s Hz) of relatively low voltage in
electroencephalograms (EEGs) of REM sleep, as opposed to predominance of
slow waves ( 14 Hz) of greater voltage in EEGs of NREM sleep (Fig. 3), requires
an introductory discussion. This concerns some major actions eected by brain
waves during waking and sleep. As we shall see, it is to the great adaptedness of the
presence and actions of fast waves in REM sleep, that we can trace the adaptedness
of the other two salient properties aforementioned.
3.6. Actions of brain waves during waking
Both slow and fast brain waves are present during waking. The fast waves
greatly predominate, and all waves are of relatively low voltage. In fact, the EEGs
of most waking mammals are very similar to those of their REM sleep (Fig. 3).
Unsurprisingly then, the brain is believed to be engaged in many of the same
activities during these two vigilance states. Wakefulness, of course, is highly adaptive
in obvious ways. Accordingly, one can suspect, with considerable justication, in
view of the brain-wave similarities, that REM sleep probably is adaptive for at
least one of the same reasons as is wakefulness. As proposed above, the key to REM
sleeps principal adaptive feature, like one of the features of wakefulness, lies in
the presence and actions of fast waves, specically their roles in assembling and
reinforcing mnemonic neural circuitry.
Of pertinence to these actions, von der Malsburg (1981) proposed his cell
assembly hypothesis, according to which the thoughts, perceptions, and actions
(hereafter referred to as events) of conscious activity are brought about through
20 J. L. Kavanau
Fig. 3. Representative electroencephalographic records (auricularoccipital electrodes) during 30-s

periods of wakefulness and three sleep states. The wakefulness curve is most complex, and consists
mostly of asynchronous fast waves (1430 Hz beta waves, 3080 Hz gamma waves, and higher frequencies
up to 800 Hz). The slow waves during waking are obscured by the more abundant fast waves. Stage 2
NREM sleep (about 4550% of sleep time) consists largely of spindles, i.e., waxing and waning,
synchronized, relatively high-voltage, waves at 714 Hz, in sequences recurring every 310 s. Stage 4
NREM sleep (about 1323% of sleep time) consists largely of synchronized, relatively high-voltage, delta
waves (0.54.5 Hz). The REM state consists mainly of asynchronous fast waves, with minor slow-wave
components. Scale bars are for 1 s and 50 mv (by permission of Prof. M. Mahowald).
the temporally-correlated ring (mobilization and synchronous activation) of

participating neural circuits. Many following experiments, most notably by
G. Buzsaki, R. Eckman, C. M. Gray, R. Llinas, W. Singer, and M. Steriade,
implicated fast brain waves in this assembly and activation process. It is thought
that, on a time scale of 1030 ms, these waves coordinate and synchronously bind
the outputs of distributed component circuits of the events mostly circuits for
cognition, the senses, and detecting and eecting motion. The expression of the
assembled, bound contents of these circuits in higher vertebrates can be regarded
as the neuronal correlate of consciousness. The binding action applies both to
events that are perceived, and those that are in long-term memory, accounting
both for their real-time occurrence and recall, together with reinforcement of the
participating neural circuitry. These actions of fast waves are referred to here as
coordinated reinforcement.
Based on the analysis that follows, it is proposed that the less abundant slow
waves of waking primarily activate and reinforce only individual, distributed
component circuits of memories, without binding them into meaningful events.
It is largely the strengths of the individual component circuits internal synapses
that are reinforced, maintaining the circuits in a competent state. These slow-wave
actions are referred to here as uncoordinated reinforcement. Consciousness is not
penetrated by slow-wave activations of individual component circuits (essentially
meaningless fragments of memories) because the components do not become

bound into meaningful events. It appears that the slow and fast waves of waking act
cooperatively, the former maintaining the competence of component circuits
(uncoordinated reinforcement), and the latter assembling and reinforcing the
specic combinations of these circuits that participate in real-time events and
fully-formed memories (coordinated reinforcement).
3.7. Actions of brain waves during sleep
A major basis for dierences between NREM and REM sleep, as regards roles
in the long-term maintenance of memories, hinges on the well-established
circumstance mentioned earlier: individual component circuits of long-term
memories in brains of advanced organisms came to exist at widely separated
locations as the brains organization and specializations evolved. As a result of
having become distributed in this way during their establishment in memory
(the initial phase of memory formation, known as consolidation), contents of
component neural circuits have to be brought together and recombined to express
fully-formed memories in which they participate. Following the present analysis,
ability to accomplish these vital long-term memory maintenance processes
most eectively during sleep, rather than only during waking, awaited the evolu-
tion of warm-bloodedness, with the presence and actions of fast waves during
REM sleep.
Concerning the occurrence and actions of brain waves in reptiles, slow waves
generally predominate during rest and sleep. Fast waves, thought to accomplish
coordinated reinforcement, make their appearance during arousal and wakefulness
(reviewed in Kavanau, 1997, 2002a). Accordingly, it would follow from this
treatment that the basic function of the slow waves that characterize reptilian sleep
and NREM sleep in mammals and birds (aside from roles in memory consolidation)
is uncoordinated reinforcement, dened and discussed before as it is thought to
occur during waking. In this reinforcement process, weakened synaptic strengths
within individual component circuits become strengthened. Except that slow waves
are thought to act preferentially on most recently used or recalled circuits, they
probably act without selectivity on individual component circuits of all memories
(Kavanau, 2002a). Following this analysis, uncoordinated reinforcement would
have been the original sleep-associated mechanism of long-term memory mainte-
nance of cold-blooded vertebrates, inherited by mammals and birds as a chief
action of slow brain waves during NREM sleep.
It is well established that stimulation of circuits that control skeletal muscle
contractions during both the sleep of cold-blooded vertebrates and NREM sleep of
warm-blooded vertebrates, fails to produce contractions, because the innervated
muscles are hypotonic (partial loss of tone) at these times. Only uncoordinated
twitches often occur. Although the hypotonic muscles do not contract, the activated
component circuits that control their contractions become reinforced. Similar
stimulation and reinforcement of the individual component circuits of memories of
thoughts and perceptions also lead to reinforcement. However, they fail to lead to
22 J. L. Kavanau
awareness. But the latter failure is for a dierent reason than the failure of muscle
contractions. As suggested above, it is because mere fragments of memories (analogs
of muscle twitches) cannot penetrate consciousness, just as uncoordinated muscle
twitches cannot lead to contractions.
Fast waves of REM sleep also lead to reinforcement of circuits controlling skeletal
muscle contractions, but the muscles do not contract then, either because they are
completely without tone. But, contrary to circumstances during NREM sleep, fully-
formed, long-term memories of thoughts and perceptions are assembled, bound, and
reinforced by the fast waves (coordinated reinforcement) of REM sleep, just as
during waking. The crucial departure from the aftermath during waking, is that
reinforcement of events during REM sleep becomes manifested only as dreams.
Several known dierences from waking conditions, not mutually exclusive, that
result in these events being expressed only as dreams, are: atonia in large muscle
masses; a high (but not complete) degree of perceptual isolation from external
stimuli; all the dream events originating within the brain; and reduced behavioral
responsiveness, as characterized by sleep researchers (e.g., Rechtschaen, 1998).
The latter, perhaps, was achieved during evolution of REM sleep primarily by
selection for lower voltage fast waves, of somewhat dierent composition from those
of waking (Kavanau, 2002a) (see Fig. 3, noting voltage-scale dierences).
3.8. Long-term memory maintenance
Although reptilian sleep closely resembles NREM sleep of warm-blooded

vertebrates, albeit without either the ability to keep the locomotor muscle-contractile
system on standby or set-point control of body temperature, reptiles also maintain
memories over the long term. Unless this is achieved solely by fast waves during
waking, which is unlikely (recall that sleep is thought to relieve an otherwise
overloaded brain of some of its non-urgent waking activities), it would follow
that the reptiles primitive mechanism of uncoordinated reinforcement during
sleep also maintains long-term memories, though probably less eectively than by
coordinated reinforcement when awake. Patients with brain pathologies, who have
lacked REM sleep and dreaming for years, without overt evidence of impairment,
also have this ability.
A minority of sleep researchers regard this phenomenon retention of fully-
formed long-term memories in the absence of REM sleep and fast waves as a
persuasive counterexample to the widely-held concept that REM sleep is involved in
consolidation and long-term maintenance of memories. Adopting this extreme
position overlooks both the fundamentally conservative nature of evolutionary
processes and the brains high degree of plasticity. Retention of long-term memories
in these conditions can, instead, be accounted for by persistence of the less eective,
primitive reptilian mechanism of slow-wave (uncoordinated) memory reinforcement,
as discussed earlier. In other words, when the most eective sleep-associated memory
maintenance mechanism, that of NREM followed by REM sleep, fails, the brain
falls back on its primitive, less eective safeguard, the memory maintenance
mechanism of NREM sleep a phenomenon that tends to conrm the thesis that, of
the two sleep states of mammals and birds, the NREM state most approximates
ancestral conditions. One should mention, in this connection, that minor
components of fast waves (e.g., ripples) persist during NREM sleep (see
Kavanau, 2002a,b). These could be responsible for small amounts of coordinated
reinforcement, probably contributing to NREM dreams, which are more thoughtlike
than those of REM sleep (Solms, 2000).
The primitive, retained reptilian mechanism of long-term memory maintenance
by slow waves during sleep hinges on an indirect strengthening of residual links
between individual, distributed, component circuits which, taken together, comprise
fully-formed memories. This mechanism of strengthening residual links (synaptic
connections) appears to depend on one of the most widely studied phenomena
in neuroscience, treated recently by Zhou et al. (2003). When two synapsed neurons
are activated in sequence within a few milliseconds, long-term changes (potentia-
tion or depression) are induced in their linking synapses. The type of change
is determined by the order of activation of the pre- and postsynaptic neurons.
With individual links reinforced indirectly in this way during NREM sleep, even
though with less overall eectiveness than by coordinated reinforcement of fully-
formed memories during REM sleep, long-term memories are stored and available
for recall. It is typically the case in such circumstances, that the more recent the
last recall or expression of a memory to which the links contribute, the stronger
the links are.
3.9. Basic links between muscle-controlling and visual circuits
A fundamental basis for the very close interlinking of circuits that control skeletal
muscles and achieve vision, and for vivid visual accompaniments of movements in
dreams, probably owes to the following suggested correlation (Kavanau, 2002a).
Vision in vertebrates did not, as one might assume, evolve primarily for the sake of
the multiple advantages of excellent sight. Rather, it evolved primarily in response
to crucial, primordial, perceptual needs for precisely timed control of bodily
movements in hunting and escape (Milner and Goodale, 1995). This very close
association has recently received convincing experimental conrmation. Many
neurons in parietal, prefrontal, and motor cortical areas are activated both by
the particular movements that they control and by the visual stimuli that trigger
these movements (Toth and Assad, 2002). Such relationships are strikingly
illustrated in humans: the so-called mirror neurons in brain regions active during
a subjects motor activity, also are active when the subject views someone else
engaged in the same activity (Buccino et al., 2001). As a result of this intrinsic
linkage, when either type of circuitry (motor or visual) is activated during REM
sleep in the process of coordinated reinforcement and dreaming, the other type
often is activated as well.
Other signs of this primordial association also are evident. In many situations,
neuronal activity in primary visual cortex (region V1) is better correlated with
perception than with activity in higher visual regions. And region V1 has unique
properties, well-suited for participating in the control of behavioral movements: it is
24 J. L. Kavanau
the rst cortical region to receive aerents from the lateral geniculate nucleus; it is
the largest visual cortical region; it has the highest resolution retinotopic map
and smallest receptive elds; and it has independent input from both eyes
(Paradiso, 2002).
In view of the above, the very close ties between visual and movement perceptions
in dreams, and also between synaptic reinforcements that have the highest priority
during REM sleep, should be no cause for surprise (see Kavanau, 2002b). And the
critical need for skeletal muscle-controlling circuitry to be maintained in a high
degree of competence should apply with equal force to many visual (visuomotor)
circuits. In this connection, Solms (2000) has observed, [f ]ew people would disagree
that the average NREM dream is more thoughtlike than the average REM
dream. The above considerations help to account for this phenomenon. Since
memories for movements and vision have highest priority for reinforcement
during REM sleep, it falls to many thoughtlike memories to be reinforced during
NREM sleep, presumably mediated by the ripples and other minor fast-wave
components.
3.10. Absence or lesser amounts of REM sleep in marine mammals and birds
The pertinence of REM sleep for reinforcement of circuitry for vision and
gross bodily movements can be gauged by reference to the practices of mammals
that lack REM sleep. Several marine mammals, including dolphins, sea porpoises,
and white whales, are continually in motion, with their skeletal muscles and visual
systems in virtually continuous use. In this mode of life there is no need for
ancillary reinforcement of visual and muscle-controlling circuits. Accordingly,
there is no need for a sleep state in which gross bodily movements are inhibited
and, therefore, no need for, nor engagement in, REM sleep. These animals
engage only in NREM sleep, which usually occurs in only half of their brains at a
time (unihemispheric sleep), and with only one eye closed (Mukhametov, 1984;
Lyamin et al., 2002).
Moreover, these marine mammals have excellent memories. Their otherwise
normal existence with no evident memory handicap not only establishes that REM
sleep is not essential for their memory acquisition and maintenance, but also that
NREM sleep alone, achieves all mnemonic-supporting functions to the required
degree. These circumstances support the proposal that the original mnemonic
functions of REM sleep primarily supplemented those of NREM sleep.
The REM sleeps origin did not entail loss of the capability of NREM sleep to
reinforce circuitry, it merely accomplished a more eective specialization of these
capabilities, namely, coordinated reinforcement of fully-formed memories and links
between them.
The same selective forces, with very similar results, appear to have been in eect
for birds. Avian species that put their antigravity muscles to the greatest and most
lengthy continuous use need the least REM sleep, together with much lesser needs
for hypotonia in only 335% of REM episodes and with atonia rare (Amlaner
and Ball, 1994). This probably is because synaptic strengths in their heavily used
skeletal muscle controlling circuits automatically become reinforced during

frequent vigorous use in ight. As a result, they have lesser or no need for further
reinforcement during REM sleep (Kavanau, 1996), which averages only about 25%
of that in most mammals (Amlaner and Ball, 1994).
Support also comes from the other extreme. Birds that put their ight muscles to
the least use need the most REM sleep. Flightless birds, leading an existence not
unlike that of mammals, fall into this category. So, REM sleep amounts to 16.2%
of total sleep in the little penguin, Eudyptula minor (Refs. in Amlaner and Ball, 1994;
Kavanau, 1996). Moreover, many birds also engage in unihemispheric NREM sleep.
Besides occurring at rest, this has been observed in some gull species on ights to the
roost, and it may also occur on migratory ights, the analog of unihemispheric sleep
in continuously active marine mammals (Refs. in Amlaner et al., 1994; Kavanau,
2001a).
Similar changes in brain waves present during waking, as compared to sleep (or
closed eyelids), occur in many other vertebrates. But they are exhibited most
strikingly in birds, in which the optic nerves decussate (cross over in the optic
chiasma) virtually completely or incompletely in most adults. In many birds, brain
unihemispheric sleep can be turned on and o merely by blocking vision in one
eye, oering additional support for the proposed role of detailed focal vision in
sleeps origin. Thus, a NREM-like sleep state, during which memory processing
presumably is taking place at a high level, is initiated or terminated in either brain
hemisphere, when the contralateral eyelid is closed or opened. Also REM sleep
occurs only during the relatively brief periods (measured in s) when eyelids of both
eyes are closed. Avian sleep is so closely associated with eyelid closure that it is
asserted that eyelids close only in sleep, and that blinking in some species is a
good behavioral index of sleep (Refs. in Kavanau, 1997).
3.11. Origin of fast waves of REM sleep
With this preamble, I consider the origin of the third salient, and proposed highly
adaptive, property of REM sleep, the presence and action of fast waves. As noted
earlier, with warm-bloodedness evolving, and the daytime-sleeping ancestors of
mammals exposed to ambient temperatures in their core ranges, set-point control of
body temperature would have ceased and skeletal muscles would have become
atonic, just as during REM sleep today. Since fast waves bring about the thoughts,
perceptions, and actions of the consciously active animal, they could not have been
present during primitive sleep. Had they been present, consciousness would not have
been lost (see, also, further discussion). This is the basis for the conclusion drawn
here, that slow waves activate only component circuits (fragments) of memories that
cannot penetrate consciousness. But upon attainment of a condition with set-point
control of body temperature suspended, and with skeletal muscles atonic, fast waves
no longer would have been maladaptive and selected against, since they could no
longer have disrupted sleep by stimulating gross movements.
Inasmuch as fast waves are thought to reinforce fully-formed, long-term
memories that control muscle contractions, rather than only individual fragments,
26 J. L. Kavanau
direct fast-wave reinforcement during atonic primitive sleep would have been
much more eective than indirect reinforcement by slow waves during hypotonic
periods. Accordingly, selective pressures, which favor most eective or ecient
mechanisms, would have favored persistence of the fast waves of waking during
those periods of primitive sleep in which muscles were atonic the putative
forerunners of REM sleep.
But anything adaptive for reinforcing circuits of long-term memories that
control muscle contractions, also would be expected to be adaptive for reinforcing
circuits for other fully-formed, long-term memories those of thoughts and per-
ceptions. In that event, just as seen above, selection would have even more heavily
favored sleep with fast waves present, thereby making possible reinforcement of
circuits for all events, but without conscious awareness. Awareness at these
times would have been only non-conscious, namely, dreaming, as accounted for
earlier.
Absence of set-point temperature control and skeletal muscle tone during REM
sleep may not, in themselves, be adaptive. In fact, their apparent maladaptedness, in
that they appear to leave the sleeper more vulnerable during REM sleep, has puzzled
sleep researchers ever since these properties were discovered. The properties appear
to gain adaptedness the greater vulnerability aspect being obviated largely by the
use of safe sleeping quarters (as discussed earlier) mainly by virtue of enabling the
persistence of fast waves during REM sleep. In this way, appreciable additional
periods beyond wakefulness are created for the actions of the putatively most highly
eective mechanism for reinforcing long-term memories, namely, for coordinated
reinforcement by fast waves.
4. Evolutionary origin of sleep within a brain-wave paradigm
It was concluded earlier, that the selective pressure for the origin of sleep was the
need to ameliorate conicts between the brains need to meet crucial, largely
unpredictable, moment-to-moment needs and, at the same time, meet needs to
acquire, consolidate, and reinforce memories, all being accomplished largely in
shared neural regions. The acquisition of sleep by ancient vertebrates is thought to
have resolved these conicts. In the light of the foregoing treatments of brain waves,
the resolution can now be dealt with more explicitly.
If the aforementioned conicts that led to selection of sleeps origin arose
from the increasing waking need to acquire and maintain large stores of memories
of new and old waking experiences, the new vigilance state of primitive sleep
should have provided the unrestricted opportunity to fulll this need. Since
the fast waves that activate and temporally bind distributed component circuits
of memories produce awareness during waking, it follows, as noted before, that
fast waves could not have been present during the evolving vigilance state of
primitive sleep.
Accordingly, at the neurophysiological level, the selective pressure for primitive
sleep would have favored a progressively reduced presence of fast waves during
restful waking, culminating in sleep. Indeed, this is what happens when we fall
asleep. Our vigilance declines monotonically in progressing from NREM stages 1 to 4

(see Kavanau, 1997), often described merely as an overt increase in the threshold for
responses or arousal. During the new state of primitive sleep, only slow waves would
have been present, and distributed component circuits of memories would have
received only uncoordinated reinforcement. The matter of memory consolidation
during sleep, suggested to include restructuring new memory representations by
Wagner et al. (2004), is considered elsewhere (see Kavanau, 2002b).
5. Sequential cycling of NREM and REM sleep
In the earlier discussion, it was proposed, that individual component circuits of

long-term memories of mammals are kept competent by the actions of slow waves
during NREM sleep, while fully-formed memories are assembled, bound, and
reinforced by the actions of fast waves during REM sleep, acting on these competent
circuits. It follows that the most eective reinforcement of fully-formed, long-term
memories of mammals during sleep probably is achieved by fast waves acting on
circuits kept competent by slow waves REM sleep following NREM sleep
(coordinated following uncoordinated reinforcement). It is reasonable to anticipate
that some minimal time would be required to achieve an eective measure of
reinforcement during each sleep state.
These relationships have implications for the phenomenon of sequential cycling
of NREM and REM states during a nights sleep, regarded by Rechtschaen (1998)
as one of the major mysteries of sleep. One would expect four or ve 90-min
cycles occurring in sequence in humans, with the length of each sleep state
fullling the minimum, to be much more eective than one long biphasic period
lasting 68 h, with the lengths of each state greatly exceeding the minimum. Since
multiplication and reduction of physiological processes and structural elements
are among the most common outcomes of evolutionary processes (Kavanau, 1990),
these circumstances could account for the partitioning of sleep into multiple
NREMREM cycles.
Other workers have also proposed that REM and NREM sleep have
complementary eects, but only in the context of memory consolidation, which
has received the lions share of attention. They also proposed that the sleep states
must act serially in order to accomplish consolidation (e.g., Gais et al., 2000). This
requirement also would favor the evolution of multiple NREMREM cycles during
lengthy periods of sleep. One recent study (Mednick et al., 2003) illustrates the
need for both NREM and REM sleep to learn and improve performance on a visual
texture discrimination task. On the day after learning the task, after a nights sleep,
repeated testing revealed deterioration of performance. A 60-min midday nap, rich
in NREM but poor in REM sleep, reversed the deterioration (restored performance)
but failed to produce signicant improvement over baseline. But a comparable
nap, rich in both NREM and REM sleep, facilitated learning in a manner closely
resembling that seen after a full 8-hour night of sleep.
28 J. L. Kavanau
6. Overview
In brief recapitulation, the neural adaptation of circuit multifunctionality

conferred great eciency of brain function at early stages of evolution, in animals
that led relatively simple lives with few needs for long-term memories, such as
many lower invertebrates. As more complex lifestyles and detailed focal vision
evolved, needs for self-initiated and reexive activities increased, and recognition
of many locales, conspecics, and other forms of life became essential. These
developments were accompanied by greatly expanded needs for neural processing
supporting sensory and motor activities, and establishing and storing long-
term memories. Since these categories of neural processing occur largely in shared
regions, brain functioning would have become increasingly maladaptive had
the evolution of these more complex lifestyles not been accompanied by the
evolution of compensating adaptations. These coevolving adaptations consisted
of: rst, restful waking; second, primitive sleep; and lastly, fully developed sleep, with
specialized REM and NREM states, that contribute to the maintenance of great
eciency of brain function. The only animals with detailed focal vision that can
achieve highly ecient brain function without sleep, are those in which demands on
brain function are greatly reduced in consequence of routine, monotonous, almost
purely reexive lifestyles, with few needs for acquiring experiential long-term
memories. The best known animals in this continuously active category are tunas
and many sharks.
In this treatment, the utility of an evolutionary approach has been illustrated
always seeking and mindful of dierential adaptedness and underlying selective
pressures to guide research into the origin and functions of sleep and its REM and
NREM states. It was of paramount importance in this pursuit to recognize that
physiological and behavioral adaptive responses to many types of selective pressures
are basically highly conservative, usually occur gradually and continuously, and
typically are restricted to modifying or co-opting existing substrates, rarely eecting
new mechanisms. It was equally important to be mindful of the often overlooked
circumstance that, though counterexamples unequivocally invalidate mathematical
and logical theorems, they do not dictate failure in the much more complex
biological realm. Otherwise, since counterexamples can be deceptive, and are by
no means rare (such as blind mammals that need sleep, sighted shes that need no
sleep, marine mammals that need no REM sleep, and patients who retain fully-
formed long-term memories in the absence of REM sleep and fast waves), one
may be too easily diverted from pursuing valid lines of analysis. Evolutionary
processes sometimes achieve adaptations that, on the face of it, appear to be
unattainable.
References
Allison, T., Van Twyver, H., 1970. The evolution of sleep. Nat. Hist. 79, 5665.
Amlaner, C.J., Jr., Ball, N.J., 1994. Avian sleep. In: M.H. Kryger, T. Roth, W.C. Dement (Eds.),
Principles and Practice of Sleep Medicine. W.B. Saunders, Philadelphia, pp. 8194.
Aziz-Sultan, A., Weeks, R.A., Tian, B., Cohen, I.G., Rauschecker, J.P., Hallett, M., 1997. Auditory
localization demonstrates cross modal plasticity in congenitally blind subjects. Neurol. 84(Suppl. 2),
A219.
Bach, V., Telliez, F., Libert, J.-P., 2002. The interaction between sleep and thermoregulation in adults and
neonates. Sleep Med. Rev. 6, 481492.
Bennett, A.F., Ruben, J.R., 1986. The metabolic and thermoregulatory status of therapsids. In: N. Hotton,
III, P.D. MacLean, J.J. Roth, E.C. Roth (Eds.), The Evolution and Biology of Mammal-like Reptiles.
Smithsonian Institute Press, Washington, D.C.
Buccino, G., Binkofski, F., Fink, G.R., Fadiga, L., Fogassi, L., Gallese, V., Seitz, R.J., Zilles, K.,
Rizzolatti, G., Freund, H.-J., 2001. Action observation activates premotor and parietal areas in a
somatotropic manner: An fMRI study. Eur. J. Neurosci. 13, 400404.
Campbell, S.S., Tobler, I., 1984. Animal sleep: a review of sleep duration across phylogeny. Neurosci.
Biobehav. Rev. 8, 269300.
Cossins, A.R., Bowler, K., 1987. Temperature Biology in Animals. Chapman and Hall, New York.
Crompton, A.W., Jenkins, F.A., Jr., 1978. Mesozoic mammals. In: V. Maglio, H.B.S. Cooke (Eds.),
Evolution of African Mammals. Harvard University Press, Cambridge, pp. 4655.
Crompton, A.W., Taylor, C.R., Jagger, J.A., 1978. Evolution of homeothermy in mammals. Nature 272,
333336.
Damasio, A., 1999. The Feeling of What Happens. Harcourt Brace, New York.
Dawson, T.J., 1973. Primitive mammals. In: G.C. Whittow (Ed.), Comparative Physiology of
Thermoregulation, Vol III. Academic Press, New York, pp. 146.
Diamond, I.T., 1982. Changing views of the organization and evolution of the visual pathways.
In: A.R. Morrisson, P.L. Strick (Eds.), Changing Concepts of the Nervous System. Academic Press,
New York, pp. 201233.
Doron, N., Wollberg, Z., 1994. Cross-modal neuroplasticity in the blind mole rat Spalax ehrenbergi:
A WGA-HRP tracing study. NeuroReport 5, 26972701.
Everson, C.A., 1995. Functional consequences of sustained sleep deprivation in the rat. Behav. Brain Res.
69, 4354.
Gais, S., Plihal, W., Wagner, U., Born, J., 2000. Early sleep triggers memory for early visual
discrimination skills. Nature Neurosci. 3, 13351339.
Gordon, C.J., Heath, J.E., 1986. Integration and central processing in temperature regulation. Ann. Rev.
Physiol. 48, 595612.
Groenewald, G.H., Welman, J., MacEachern, J.A., 2001. Vertebrate burrow complexes from the early
triassic cynognathus zone (Driekoppen Formation, Beaufort Formation) of the Karoo Basin, South
Africa. Palaios 16, 148160.
Heath, J.E., 1968. The origins of thermoregulation. In: E.T. Drake (Ed.), Evolution and Environment.
Yale University Press, New Haven, pp. 336355.
Hillenius, W.J., 1994. Turbinates in therapsids: evidence for late Permian origins of mammalian
endothermy. Evol. 48, 207229.
Jerison, H.J., 1973. Evolution of the Brain and Intelligence. Academic Press, New York.
Kavanau, J.L., 1987. Lovebirds, Cockatiels, Budgerigars, Behavior and Evolution. Science Software
Systems, Inc., Los Angeles.
Kavanau, J.L., 1990. Conservative behavioural evolution, the neural substrate. Anim. Behav. 39, 758767.
Kavanau, J.L., 1996. Memory, sleep, and the dynamic stabilization of neiral circuitry: evolutionary
perspectives. Neurosci. Biobehav. Rev. 20, 289311.
Kavanau, J.L., 1997. Memory, sleep, and the evolution of mechanisms of synaptic ecacy maintenance.
Neurosci. 79, 744.
Kavanau, J.L., 1998a. Vertebrates that never sleep: implications for sleeps basic function. Brain Res. Bull.
40, 269279.
Kavanau, J.L., 1998b. Memory, sleep, and the evolution of mechanisms of synaptic ecacy maintenance.
UCLA Document Services, Los Angeles, pp. 1146.
Kavanau, J.L., 1999. Adaptations and pathologies linked to dynamic stabilization of neural circuitry.
Neurosci. Biobehav. Rev. 23, 635648.
30 J. L. Kavanau
Kavanau, J.L., 2001a. Brain-processing limitations and selective pressures for sleep, sh schooling, and
avian ocking. Anim. Behav. 62, 12191224.
Kavanau, J.L., 2001b. Reinforcement of memory during sleep. In: C.A. Shaw, J.C. McEachern (Eds.),
Toward a Theory of Neuroplasticity. Psychology Press, Philadelphia, pp. 5163.
Kavanau, J.L., 2002a. REM and NREM sleep as natural accompaniments of the evolution of warm
bloodedness. Neurosci. Biobehav. Rev. 26, 889906.
Kavanau, J.L., 2002b. Dream contents and failing memories. Archives Italiennes de Biologie 140,
109127.
Kavanau, J.L., Rischer, C.E., 1973. Ground squirrel behaviour during a partial solar eclipse. Bollettino di
Zoologia 40, 217221.
Levi-Setti, R., 1993. Trilobites, 2nd ed. University of Chicago Press, Chicago.
Lillegraven, J.A., Kielan Jaworowska, Z., Clemens, W.A., 1979. Mesozoic Mammals. University of
California Press, Berkeley.
Llinas, R.R., Pare, D., 1991. Of dreaming and wakefulness. Neuroscience 44, 521535.
Lyamin, O.I., Mukhametov, L.M., Siegel, J.M., Nazarenko, E.A., Polyakova, I.G., Shpak, O., 2002.
Unihemispheric slow wave sleep and the state of the eyes in a white whale. Behav. Brain Res. 129,
125129.
Maquet, P., 2001. The role of sleep in learning and memory. Science 294, 10481051.
MacLean, P.D., 1986. Neurobehavioral signicance of the mammal-like reptiles (therapsids). In:
N. Hotton, III, P.D. MacLean, J.J. Roth, E.C. Roth (Eds.), The Evolution and Biology of
Mammal-Like Reptiles. Smithsonian Institute Press, Washington, pp. 122.
McNab, B.K., 1978. The evolution of endothermy in the phylogeny of mammals. Amer. Nat. 112, 121.
Mednick, S., Nakayama, K., Stickgold, R., 2003. Sleep-dependent learning: a nap is as good as a night.
Nature Neurosci. 6, 697698.
Milner, A.D., Goodale, M.A., 1995. The Visual Brain in Action. Oxford University Press, New York.
Moorcroft, W.H., 1989. Sleep, Dreaming, and Sleep Disorders. University Press of America, New York.
Mukhametov, L.M., 1984. Sleep in marine mammals. Exp. Brain Res. Supp. 8, 227236.
Paradiso, M.A., 2002. Perceptual and neuronal correspondence in primary visual cortex. Curr. Opinion
Neurobiol. 12, 155161.
Parmeggiani, P.L., 1980. Temperature regulation during sleep: a study in homeostasis. In: J. Orem, C.D.
Barnes (Eds.), Physiology in Sleep. Academic Press, New York, pp. 98143.
Paul, G.S. 2002. Dinosaurs of the Air, the Evolution and Loss of Flight in Dinosaurs and Birds. Johns
Hopkins University Press, Baltimore.
Rauschecker, J.P., 1991. Mechanisms of visual plasticity: Hebb synapses, NMDA receptors and beyond.
Physiol. Rev. 71, 587615.
Rauschecker, J.P., 1995. Developmental plasticity and memory. Behav. Brain Res. 66, 712.
Rechtschaen, A., 1998. Current perspectives on the function of sleep. Perspect. Biol. Med. 41, 359387.
Ringach, D.L., 2003. Neuroscience: states of mind. Nature 425, 912913.
Roder, B., Rosler, F., Hennighausen, E., 1997. Dierent cortical activation patterns in blind and sighted
humans during encoding and transformation of haptic images. Psychophysiol. 34, 292307.
Roe, A.W., Pallas, S.L., Kwon, Y.H., Sur, M., 1992. Visual projections routed to the auditory pathway in
ferrets: Receptive elds of visual neurons in primary auditory cortex. J. Neurosci. 12, 36513664.
Rolls, E.T., 1991. Information processing in the temporal lobe visual cortical areas of macaques. In:
M.A. Arbib, J.-P. Ewert (Eds.), Visual Structures and Integrated Functions. Springer-Verlag, Berlin,
pp. 339352.
Ross, J., Ma-Wyatt, A., 2004. Saccades actively maintain perceptual continuity. Nature Neurosci. 7,
6569.
Rutherford, D.R., 2003. Brains on the mind. Sci. Amer. Dec., 21.
Sadato, N., Pascual-Leone, A., Grafman, J., Ibanez, V., Deiber, M.-P., Dold, G., Hallett, M., 1993.
Activation of the primary visual cortex by Braille reading in blind subjects. Nature 380, 526528.
Schwartz, W.J., 1997. On the neurobiology of sleep and sleep disorders not yet known. In: W.J. Schwartz
(Ed.), Sleep Science: Integrating Basic Research and Clinical Practice. Karger, New York, pp. 18.
Siegel, J.M., Manger, P.R., Nienhuis, R., Fahringer, H.M., Pettigrew, J.D., 1996. The echidna
Tachyglossus aculeatus combines REM and non-REM aspects in a single sleep state. J. Neurosci.
16, 35003506.
Siegel, J.M., Manger, P.R., Nienhuis, R., Fahringer, H.M., Pettigrew, J.D., 1998. Monotremes and the
evolution of rapid eye movement sleep. Phil. Trans. Royal Soc. (London) B353, 11471157.
Solms, M., 2000. Forebrain mechanisms of dreaming are activated from a variety of sources. Behav. Brain
Sci. 23, 10351040.
Squire, L.R., 1986. Mechanisms of memory. Science 232, 16121619.
Stickgold, R., 1998. Sleep: o-line memory processing. Trends Cognitive Sci. 2, 484492.
Stickgold, P., James, L., Hobson, J.A., 2000. Visual discrimination requires sleep after learning. Nature
Neurosci. 3, 12371238.
Toth, L.J., Assad, J.A., 2002. Dynamic coding of behaviourally relevant stimuli in parietal cortex. Nature
415, 165168.
Ungerleider, L.G., 1995. Functional brain imaging studies of cortical mechanisms for memory. Science
270, 769775.
Van Essen, D.C., Anderson, C.H., Felleman, D.J., 1992. Information processing in the primate visual
system: an integrated systems perspective. Science 255, 419423.
Vertes, R.P., 1990. Brain mechanisms of slow-wave sleep and REM sleep. In: W. Klemm, R.P. Vertes
(Eds.), Brainstem Mechanisms of Behavior. Wiley, New York, pp. 535583.
Von der Malsburg, C., 1981. The correlation theory of brain function. MPI Biophysical Chemistry,
Internal Report 812. Reprinted in: E. Domay, J.L. van Hemmen, K. Schulten (Eds.), Models of
Neural Networks II (1994), Springer-Verlag, Berlin, pp. 95119.
Wagner, U., Gais, S., Haider, H., Verleger, R., Bord, J., 2004. Sleep inspires insight. Nature 427, 352355.
Walker, J.M., Berger, R.J., 1980. Sleep as an adaptation for energy conservation functionally and
physiologically related to hibernation and shallow torpor. Prog. Brain Res. 53, 255278.
Walls, G.L., 1967. The Vertebrate Eye and Its Adaptive Radiation. Hafner, New York.
Wanet-Defalque, M.-C., Veraart, C., De Volder, A., Metz, R., Michel, C., Dooms, S., Gonet, A., 1988.
High metabolic activity in the visual cortex of early blind human subjects. Brain Res. 446, 369373.
Yaca, R., Yinon, U., Wollberg, Z., 1999. Auditory activation of cortical visual areas in cats after early
visual deprivation. Eur. J. Neurosci. 11, 13011312.
Zeki, S., 1999. Inner Vision: An Exploration of Art and the Brain. Oxford University Press, New York.
Zhou, Q., Huizhong, W., Poo, M.-m., 2003. Reversal and stabilization of synaptic modications in a
developing visual system. Science 300, 19531957.
Advances in
Cell Aging and
Gerontology
Sleep disturbances in aging

Marci M. Loisellea, Melanie K. Meansa,b
and Jack D. Edingera,b,*
a
Duke University Medical Center, Durham, NC, USA
b
Department of Veterans Affairs Medical Center, Durham, NC, USA
*Correspondence address: J. D. Edinger, VA Medical Center, 508 Fulton Street, Durham,
NC 27705, USA. Tel: 1-919-286-0411; fax: 1-919-416-5832.
E-mail address: jack.edinger@duke.edu
Contents
1. Introduction
2. Normal sleep and the effects of the aging process
2.1. Age-related sleep and circadian changes
2.2. Gender differences in sleep and aging
2.3. Nocturia
3. Epidemiology
3.1. Prevalence
3.2. Risk factors
3.3. Morbidity
3.4. Mortality
4. Common sleep disorders in older age groups: nature and etiology
4.1. Sleep-disordered breathing (SDB)
4.2. Primary insomnia
4.3. Circadian rhythm disorders
4.4. Sleep-related movement disorders
4.5. Secondary sleep disturbances
5. Assessment and diagnosis of sleep problems in older adults
5.1. Clinical interview
5.2. Psychological assessment
5.3. Sleep logs
5.4. Actigraphy
5.5. Polysomnography (PSG)
5.6. Multiple sleep latency test (MSLT)
6. Treatment of sleep disturbances in older adults
6.1. Sleep-disordered breathing
6.2. Positive pressure devices
6.4. Secondary insomnia

Published by Elsevier B.V.
DOI: 10.1016/S1566-3124(05)17002-3
34 M. M. Loiselle et al.

6.7. Sleep in institutionalized older adults
7. Summary
1. Introduction
For a variety of reasons, aging is accompanied by an increasing susceptibility to

sleep disturbances. Normal age-related changes in the human sleep system increase
the propensity for developing sleep problems in late life. With advancing age
comes increased risks of medical illnesses, psychiatric problems, and polypharmacy,
all of which may adversely impact sleep. Multiple life changes including retirement
from long-term employment, chronic medical illness, loss of a loved one, and
reduced social contacts may dramatically alter the rest-activity cycle of the older
individual. This chapter begins with a discussion of age-related sleep changes.
Subsequently, epidemiology of sleep disturbances in older adults is presented,
followed by a review of sleep disorders commonly found in this age group. Finally,
assessment, diagnosis, and treatment of sleep disorders are reviewed, with a focus on
issues relevant to older adults.
2. Normal sleep and the effects of the aging process
Human sleep patterns reect characteristic changes in brain wave activity that
can be quantied via electroencephalography (EEG) and categorized into two
distinct brain states: non-rapid eye movement (NREM) sleep and rapid eye
movement (REM) sleep. The NREM sleep consists of stages 1, 2, 3, and 4, reecting
a continuum of lighter to deeper sleep. Stage 1 is a transitional period of light sleep
that occurs when falling asleep. Stage 2 comprises the largest portion of sleep
throughout the night. Together, stages 3 and 4 represent deep sleep, during which
slow wave EEG activity dominates. The REM sleep is distinguished by vivid dream
activity, rapid eye movements, muscle atonia, and an EEG pattern that resembles
wakefulness. Adults typically cycle through deepening stages of NREM sleep
followed by an episode of REM sleep every 90120 min throughout the night.
Greater amounts of deep sleep (stages 3 and 4) occur in the rst half of the night,
with increased REM sleep in the latter half. Although the exact functions of sleep
are unknown, sucient amounts of both NREM and REM sleep are necessary for
full restorative benet (Morin, 1993).
2.1. Age-related sleep and circadian changes
Aging is associated with reliable changes in sleep/wake architecture that are

believed to reect neuronal degeneration in brain mechanisms responsible for sleep
Sleep Disturbances in Aging 35
100%
% of sleep period
80%
Wake
60%
Stage 1
40%
REM
20% Stage 2
Stages 3 & 4
0%
16-19 20-29 30-39 40-49 50-59 60-69 70-79
Age
Fig. 1. Objective sleep changes with aging. Data reported in Williams et al. (1974).
physiology (Prinz et al., 1990). As depicted in Fig. 1, older adults have increased
light sleep (stages 1 and 2) and corresponding reductions in deep sleep (Williams
et al., 1974). Compared to younger adults, older adults take longer to fall asleep
and spend more time awake during the night; their sleep is characterized by
frequent arousals and stage shifts reecting sleep fragmentation (Morgan, 2000;
Floyd, 2002). Although they spend longer periods of time in bed, they experience less
overall sleep time. In addition, age-related changes in breathing and ventilatory
responses may predispose older adults to sleep-disordered breathing (SDB) (Janssens
et al., 2000).
Older adults also experience changes in their endogenous circadian rhythms
thought to be the result of deterioration of the suprachiasmatic nucleus in the
hypothalamus (Bliwise, 2000b; Van Someren, 2000). These changes are characterized
by a attening of the circadian amplitude and a weakening of circadian entrain-
ment (Monk, 1989). Consequently, older adults are more sensitive to disturbances
in circadian timing and thus more susceptible to jet lag and shift work disorders
(Monk, 1989; Morgan, 2000). Age-related circadian changes cause a forward shift
in the sleep phase, associated with earlier bed and rising times. This advanced
sleep phase in part may explain the complaint of early morning awakening
common in this population.
From the above discussion, it is clear that age-related sleep changes result in a
deterioration of the nighttime sleep pattern. It is important to note, however, that the
process of aging does not guarantee the development of a sleep problem. Most
older adults experience the aforementioned age-related sleep and circadian changes,
whereas not all older adults complain of sleep disturbance. Thus, it appears that
sleep changes due to normal aging are predisposing but not sucient for the
development of sleep problems.
2.2. Gender differences in sleep and aging
Men have greater diculty maintaining sleep with more wake time during the
night, more light sleep, and a higher incidence of primary sleep pathologies than do
women (Gottlieb, 1990; Floyd, 2002). Despite these dierences, women are more
likely to report sleep problems and to use sleep medications (Bliwise, 2000b). Studies
also suggest that circadian phase advances are greater in women than in men
(Bliwise, 2000b). Furthermore, menopause aects sleep patterns via hormonal
mechanisms and is commonly associated with complaints of insomnia, sleep
disruption, and fatigue (Moe, 1999).
2.3. Nocturia
Nocturia, the need to urinate multiple times during the night with a consequent
disruption of sleep, increases signicantly with age and is very common in older
adults. By age 60, over half of all individuals experience nocturia, and this problem
increases to at least 80% of 80-year-olds (Donahue and Lowenthal, 1997; Jennum,
2002). Causes of nocturia include excessive nighttime urine production, reduced
bladder capacity, prostate problems, or other underlying medical conditions
(Jennum, 2002). In some cases, nocturia may be a result of untreated SDB (Bliwise,
2000b; Russo-Magno et al., 2001). Whatever its etiology, nocturia is a common
cause of sleep maintenance complaints in older adults (Bliwise, 2000b).
3. Epidemiology
3.1. Prevalence
Complaints about sleep disturbances and associated waking symptoms are

relatively prevalent in society at large and are particularly common among older
adults. More than half of community-dwelling adults over the age of 65 report
chronic sleep diculties (Prinz et al., 1990; Foley et al., 1995). Close to 50% of this
age group report insomnia symptoms such as diculty initiating sleep, diculty
sustaining sleep, waking too early in the morning, or feeling unrefreshed upon
awakening (Ganguli et al., 1996; Babar et al., 2000; Ohayon et al., 2001; Ohayon and
Vecchierini, 2002). Moreover, roughly 5% of the older adult population experience
the onset of such sleep symptoms each year, and once present, such symptoms tend
to persist over time (Ganguli et al., 1996; Foley et al., 1999). In addition, complaints
of excessive daytime fatigue and sleepiness resulting from nocturnal sleep
disturbances are also relatively common in aged populations. In fact, between 8
and 25% of all adults 65 and older complain of daytime sleepiness that disrupts
normal diurnal functioning (Ganguli et al., 1996; Hays et al., 1996; Foley et al.,
2001).
Given these nocturnal and diurnal complaints, it is not surprising that various
common sleep disorders are especially prevalent in older age groups. For example,
a greater proportion of middle-aged and elderly adults meet formal criteria for
an insomnia diagnosis than do younger age groups (Ohayon and Vecchierini, 2002).
Likewise, sleep apnea, characterized by SDB and associated daytime sleepiness,
is present in as many as 24% of community-dwelling, 33% of acutely hospitalized,
and 42% institutionalized individuals over 65 years of age (Ancoli-Israel, 1989;
Ancoli-Israel et al., 1991b). Restless legs syndrome and periodic limb movements are
common sleep-related movement disorders that respectively occur in 28% (Clark,
2001) and 45% (Ancoli-Israel et al., 1991a) of the aged population. Finally, disorders
involving unusual sleep-related events such as dangerous dream enactments or
motor paralysis upon awakening also show relative propensities to occur in older
adults (Wing et al., 1999; Mahowald and Schenck, 2000).
3.2. Risk factors
Since subjective sleep/wake complaints and various sleep disorders are relatively
prevalent among older adults, it is tempting to conclude that age operates as an
independent risk factor for sleep/wake pathology. However, aging itself is often
accompanied by marked changes in vocational status, social activity, health,
and lifestyle in general (Bliwise, 2000b; Krystal et al., in press). Each of these age-
related factors may have a powerful inuence on sleep/wake functioning
independent of the aging process alone. As such, age, per se, may not represent a
specic risk factor for either sleep complaints or specic sleep disorders. Rather, age
may only serve as a proxy for other factors that more directly enhance risk for such
conditions.
Clearly, complaints of insomnia and daytime somnolence are strongly inuenced
by various age-related risk factors. Declining health status often accompanies
aging and appears to operate as an independent risk factor for sleep/wake
dysfunction. Indeed, a variety of studies have shown that perceptions of poor health
as well as medical conditions such as angina, respiratory disorders, arthritis, chronic
pain, visual impairment, and nocturia all enhance risk for both nighttime sleep
disturbance and daytime sleepiness in older age groups (Blazer et al., 1995; Foley
et al., 1995; Newman et al., 1997; Maggi et al., 1998; Whitney et al., 1998; Roberts
et al., 1999; Bliwise, 2000b; Zizi et al., 2002; Byles et al., 2003). Additionally, various
studies have shown that the presence of psychiatric disorders and depression in
particular enhance risks for sleep/wake complaints in aged samples (Blazer et al.,
1995; Morgan and Clarke, 1997; Newman et al., 1997; Schechtman et al., 1997;
Maggi et al., 1998; Whitney et al., 1998; Roberts et al., 1999; Byles et al., 2003).
Furthermore, somatic and mental disorders seemingly account for a substantial
proportion of the increased sleep/wake dysfunction among older adults as
compared to younger age groups. In fact, when community-dwelling older adults
are thoroughly screened to exclude those with sleep-disruptive medical and
psychiatric conditions, fewer than 4% report signicant sleep-related diculties
(Vitiello et al., 2002).
However, even healthy older adults may be subject to certain age-related risk
factors that enhance their chances of suering sleep/wake diculties. In particular,
psychosocial and behavioral changes that accompany aging may have signicant
eects on sleep/wake functioning. Death of a spouse leading to extended
bereavement has been noted as a common cause of sleep diculty in older
populations (Hall et al., 1997). Many older adults face a reduction in their social and
physical activities after retiring from long-term employment. Such individuals
often incur disturbance as a result of their limited physical and social activity
(Ohayon et al., 2001). Furthermore, the loss of a daily schedule accompanying
retirement often results in excessive free time for older adults. Those who attempt
to ll some of this time by spending excessive time in bed or taking daytime naps
often develop sleep diculties as a result of these poor sleep hygiene practices
(Ohayon et al., 1996).
3.3. Morbidity
Considerable morbidity accompanies the sleep/wake disturbances in older

adults. Within this population, those who suer from persistent insomnia typically
report reduced physical functioning, reduced energy for activities, and a global
reduction in their quality of life in general (Schubert et al., 2002). Perhaps as a
result of such changes, those older adults with insomnia have a greater risk for
depression than do their counterparts without sleep/wake complaints (Roberts et al.,
1999). Nocturnal sleep diculties also represent a safety concern for older adults
in that those with insomnia are at increased risk for falls and hip fracture both as
a function of the sleep disturbance alone as well as due to sedating medications
they take to remedy these diculties (Brassington et al., 2000; Byles et al., 2003;
Cumming and Le Couteur, 2003). In addition, nocturnal sleep diculties have been
linked to the subsequent cognitive decline of older adults (Cricco et al., 2001; Jelicic
et al., 2002). In turn, sleep diculties among demented community-dwelling older
adults rank high among the leading reasons for their nursing home placement
(Pollak et al., 1990).
Like insomnia, daytime sleepiness may contribute to considerable morbidity in
older age groups. Complaints of daytime sleepiness have been linked to a reduction
in productivity, vigilance, activity level, and general functional status in those aged
65 and older (Gooneratne et al., 2003). As is the case with insomnia, daytime
sleepiness seemingly operates as an independent risk factor for cognitive decline in
older adults (Foley et al., 2001; Ohayon and Vecchierini, 2002). Excessive daytime
sleepiness has also been linked to increased risk for cardiovascular events such as
myocardial infarction and congestive heart failure (Newman et al., 2000). However,
it should be noted that excessive daytime sleepiness is a common symptom of
SDB (Foley et al., 2003), a primary sleep disorder recognized as contributing to
hypertension and cardiovascular disease in older adults (Nieto et al., 2000; Wang
et al., 2002). Hence, the link between daytime sleepiness and cardiovascular disease
seen in older age groups is likely mediated by occult SDB.
3.4. Mortality
A very limited number of studies have examined the eects of insomnia and its
treatment on risk for mortality among older adults. One study (Pollak et al., 1990)
that statistically controlled for other risk factors showed severe insomnia among
men was related to increased risk for mortality during a 3-year follow-up period.
In a more recent study (Manabe et al., 2000), insomnia was found to be a signicant
risk factor for mortality among older, chronically institutionalized patients.

By contrast, three similarly well-controlled studies (Morgan and Clarke, 1997;
Althuis et al., 1998; Kripke et al., 2002) failed to demonstrate a signicant
relationship between the presence/absence or frequency of insomnia and subsequent
mortality over 4 to 6-year follow-up periods. However, one (Morgan and Clarke,
1997) of these studies did nd that mortality risk was signicantly related to
incident insomnia during the follow-up period. Another longitudinal study (Rumble
and Morgan, 1992) showed that mortality risk over a 6-year follow-up period
was signicantly elevated in older adults who used medications other than
traditional hypnotics to aid their sleep. Yet, given the inconsistent ndings of
these various studies, it remains questionable whether insomnia per se enhances
mortality risk.
A similar statement can be made about mortality risk associated with daytime
sleepiness. Whereas several survey studies (Ganguli et al., 1996; Hays et al., 1996;
Newman et al., 2000) of community-dwelling older adults in the U.S. showed a link
between daytime sleepiness and mortality, a recent survey study (Rockwood et al.,
2001) conducted in Canada failed to show that daytime sleepiness enhances
mortality risk. However, when daytime sleepiness reects underlying SDB, this
symptom should be considered seriously, since sleep apnea syndromes do appear
to enhance mortality risk among older age groups (Lavie et al., 1995; Ancoli-Israel
et al., 2003b).
4. Common sleep disorders in older age groups: nature and etiology
Sleep disturbances in older adults are often multifactorial in nature. Contributing

factors may include age-related changes in sleep architecture, medical and
psychiatric conditions, medication and substance use, primary sleep disorders,
psychosocial factors, and poor sleep hygiene (Vitiello, 1997). Common primary sleep
disturbances in this age group include SDB, insomnia, circadian rhythm disorders,
and sleep-related movement disorders. Secondary sleep disturbances, which are
the result of or aggravated by a medical disorder, psychiatric disorder, medication
use, and/or substance use, are also commonly seen in older age groups.
4.1. Sleep-disordered breathing (SDB)
Studies show that SDB (see Chapter 6) encompasses a variety of nocturnal

breathing disorders that disrupt the sleep pattern and typically cause excessive
daytime sleepiness. Sleep apnea is the most common form of SDB. It is characterized
by episodes of nocturnal breathing cessation (apnea) occurring repeatedly and is
associated with a reduction in oxygen level and brief arousals from sleep. Apnea
can be the result of upper airway collapse (obstructive sleep apnea), complete
cessation of respiratory eort due to a depressed central nervous system drive
(central sleep apnea), or a combination of the two (mixed sleep apnea).
This SDBs increased prevalence in older adults is due to multiple factors. The
incidence of SDB is higher in patients with neurological disorders, cardiac
conditions, cerebrovascular disease, and severe metabolic disorders, all of which
are more prevalent in older adults (Janssens et al., 2000). Snoring and obesity,
both risk factors for SDB, increase in prevalence through age 70 (Bliwise, 2000b;
Janssens et al., 2000). Although SDB is more common in men, postmenopausal
women are at an increased risk compared to their premenopausal counterparts
(Moe, 1999).
Primary insomnia is a multifaceted disorder characterized by diculty initiating

or maintaining sleep, unwanted early morning awakenings, and/or unrefreshing
sleep with associated daytime impairment. It may be precipitated by an identiable
event but persists when maladaptive habits and conditioned environmental cues
perpetuate sleep disruption after the precipitating event is no longer present (Morin
and Edinger, 2003). A comprehensive model of the development and maintenance of
insomnia is helpful in understanding the etiology of this problem (Spielman and
Glovinsky, 1997). According to this model, predisposing factors increase the
likelihood of insomnia developing, but are not sucient in and of themselves to
trigger the insomnia. For example, individuals who have an innate propensity
toward physiological or cognitive arousal or feelings of anxiety or depression may
have a lower threshold for the development of insomnia. However, insomnia
typically develops after precipitating factor(s) temporally and proximally cause an
acute sleep disturbance. Insomnia may be precipitated by a variety of events,
conditions, or circumstances, such as stress, environmental factors (e.g., nocturnal
light or noise, sleeping away from home), changes to the sleep/wake cycle (e.g.,
shiftwork, jet lag), medical or psychiatric illnesses, or ingestion of sleep-disrupting
substances (e.g., caeine, stimulants, alcohol). As precipitating factors resolve (either
fully or partially), insomnia assumes a chronic status with perpetuating factors
maintaining sleep disruption. These factors include behavioral and conditioning
factors such as poor sleep habits and are found to be as relevant in older adults as
they are for their younger counterparts (Bliwise, 1993).
Individuals with circadian rhythm disturbances experience a disparity between

their endogenous sleep/wake systems and the sleep/wake schedules they desire or
are required to follow (Morin and Edinger, 2003). Examples of circadian rhythm
disturbances include those resulting from frequent changes in schedules and
alterations of daily routines, such as from jet lag or shift work. Circadian rhythm
disturbances also aect individuals who demonstrate a persistent advance or delay
in the endogenous rhythm relative to the desired sleep schedule as well as
individuals who obtain their daily sleep complement by means of multiple nap
periods throughout the 24-h day. Individuals presenting with circadian rhythm
disturbances may complain that their sleep does not occur at a desirable time for
them or that their sleep is disrupted. They may also complain of insomnia, excessive
sleepiness, and general daytime dysfunction (Myers and Badia, 1995).
Given age-associated biologic, environmental, and psychosocial changes, older
adults are particularly vulnerable to advanced sleep phase syndrome, jet leg, and
irregular sleep/wake schedule disorders. Available data suggest that older adults
sleep/wake schedules are phase advanced approximately 90 mins (Neylan et al.,
1996). This phase advance causes older adults to become sleepy much earlier
(e.g., between 6:00 and 9:00 p.m.) and to waken much earlier (e.g., between 2:00 and
5:00 a.m.) than desired or required. In addition, older adults are relatively intolerant
of shifts in their sleep/wake schedules as might occur during a trip across several
time zones (Bliwise, 1993; Neylan et al., 1996). Finally, an irregular sleep/wake
schedule, characterized by random sleep throughout the 24-h day in place of a single
consolidated overnight sleep period, can emerge due to an absence of daytime
structure.
4.4.1. Restless legs syndrome (RLS)

The RLS is characterized by unpleasant sensations, typically in the lower limbs,
leading to an almost irresistible urge to move (Feinsilver, 2003). Movement tends to
momentarily relieve the sensations, but diculty with sleep initiation is often
present. There is an increased incidence of RLS in older adults due to the higher
prevalence of medical disorders and medication use that can aggravate or cause
the condition. Medical conditions that are associated with RLS include anemia,
renal failure, and diabetic peripheral neuropathy (Milligan and Chesson, 2002).
Medications and substances that exacerbate RLS include antidepressants, caeine,
diuretics, antihistamines, and bronchodilators (Bliwise, 1993; Chokroverty, 1994a;
Neubauer, 1999).
4.4.2. Periodic limb movement disorder (PLMD)

Most individuals with RLS (85%) also suer from periodic limb movement
disorder (Milligan and Chesson, 2002). The PLMD is characterized by periodic and
stereotypic leg movements during sleep approximately every 2040 s (Ancoli-Israel,
1997; White and Mitler, 1997). These repetitive leg movements are often associated
with arousals from sleep. Individuals often are unaware of these repetitive twitches
during sleep, but they may complain of excessive daytime sleepiness or unrefreshing
sleep. The higher incidence of PLMDs in older adults appears to be related to
metabolic, vascular, and neurological factors (Prinz et al., 1990). Proposed causal
factors for PLMD include venous insuciency in the lower limbs, cycles in arterial
blood pressure, decits in central dopaminergic transmission, and osteoarthritic
changes (Bliwise, 1993, 2000b; Chokroverty, 1994b).
4.4.3. REM behavior disorder (RBD)

The RBD is characterized by restoration of muscle tone and elaborate activity
during REM sleep (e.g., kicking, punching, jumping out of bed, talking) representing
dream enactment (Espie, 2000). Sleep disruption is often less of a concern than injury
to self or bed partners during these nocturnal events (Mahowald and Schenck, 2000).
Arousals from these episodes are usually rapid with full dream recall and no
resulting confusion (Mahowald and Schenck, 2000). Acute forms of RBD are
associated with alcohol intoxication/withdrawal, excessive caeine intake, or use of
tricyclic antidepressants and monoamine oxidase inhibitors. Chronic and late onset
RBD are associated with organic brain factors and may herald the onset of
Parkinsons disease (Mahowald and Schenck, 2000). Chronic RBD is much more
prevalent in males, suggesting that hormonal inuences may be a contributing factor
to this disorder (Mahowald and Schenck, 2000).
4.5. Secondary sleep disturbances
4.5.1. Sleep disturbance secondary to medical disorders

Many age-dependent medical conditions have the capacity to impair sleep, such as
nocturia, gastrointestinal disturbance, chronic pain, menopause, cerebrovascular
disease, diabetes, pulmonary disease, rheumatoid arthritis, renal disorders, and
osterarthritis (Bliwise, 1993). Sleep disturbances can arise from the physiological
mechanisms of the disease itself, from symptoms associated with the disease,
from treatment of the medical disorder, and/or the resulting psychological stress
produced (Lichstein, 2000). In addition, the sleep disturbance itself can have a
negative impact on the course of the medical illness (Chokroverty, 1994a). Most
patients with sleep disturbances secondary to a medical condition present with
insomnia but may have a mixture of insomnia and hypersomnia (Chokroverty,
1994a).
4.5.2. Sleep disturbance secondary to medications and/or substance use

Individuals with substance-induced or medication-induced sleep disorders present
with complaints of insomnia, hypersomnia, parasomnias, or a mixture of these
symptoms (Morin and Edinger, 2003). Common substances that impair sleep
include alcohol, caeine, and nicotine. Although the use and abuse of alcohol
declines with age, alcohol is frequently used by older adults to address their sleep
problems (Prinz et al., 1990; Espie, 2000). This self-medication is problematic since
alcohol can increase the severity of SDB, cause fragmented sleep, and lead to sleep
disruption when discontinued (Chokroverty, 1994b). In addition to alcohol, older
adults may use stimulants such as caeine or nicotine that increase sleep onset
latency, interfere with sleep maintenance, and increase sleep fragmentation
(Ancoli-Israel, 1997).
Polypharmacy is especially prevalent in older adults. The number of medications
ingested is positively related to the severity of reported sleep disturbances in
older adults (Bliwise, 1997; Bliwise and Breus, 2000). Sleep disruption may be due
to direct eects on sleep architecture or to side eects of the medications
(Bliwise and Breus, 2000). A higher risk of sleep disturbance is not only due to the
increased utilization of prescription medication in this population, but also to
reduced drug metabolism, absorption, distribution, and elimination rates in
older adults (Lichstein, 2000). The main classes of prescription drugs that can
cause sleep disruption are antidepressants (particularly SSRIs), decongestants,
antihypertensives, bronchodilators, diuretics, beta-blockers, and corticosteroids
(Neubauer, 1999; Lichstein, 2000). Additionally, sedative-hypnotic use is especially
prevalent in older adults and, as discussed later, is associated with a number of
adverse side eects.
4.5.3. Sleep disturbance secondary to psychiatric disorders

Common psychiatric disorders in older adults that aect sleep include adjustment
disorders, depression, bereavement, anxiety, dementia, and delirium (Bliwise,
2000b). Adjustment disorders can be caused by retirement, institutionalization,
relocation, nancial strain, health concerns, and holocaust trauma (Bliwise,
2000b). Bereavement and depression usually involve some form of sleep disturbance,
and sleep disturbance is a major risk factor for the subsequent development
of depression (Reynolds, 1996). Changes in sleep architecture caused by depression
include reduced total sleep time, sleep eciency, and slow-wave sleep and
increased awakenings and REM density (Wooten, 1994). Additionally, anxiety
disorders can increase sleep latency and lead to insomnia complaints. Psychosocial
factors, such as social isolation, boredom, and inactivity, can negatively impact
sleep as well.
Dementia and delirium increase in prevalence in older age groups. Although
Parkinsons disease, Huntingtons disease, and vascular dementia are all common
causes of dementia in old age, 70% of dementias result from Alzheimers disease
(Bliwise, 1993). Sleep in Alzheimers dementia shows a progressive deterioration
in sleep architecture and circadian rhythm as the disease progresses (Reynolds,
1996). Sleep is marked by increased frequency and duration of awakenings,
decreased slow-wave sleep, randomly-distributed sleep over a 24-h period, and
increased daytime napping (Prinz et al., 1990; Wauquier, 1993). Sundowning,
frequently seen in the middle to late stages of dementing disorders, reects the
onset or exacerbation of delirium during the night. It involves behaviors such as
reduced responsiveness to external stimuli, agitation and restlessness, disruptiveness,
emotional and perceptual disturbances, and disorganized thinking and speech
(Prinz et al., 1990; Neylan et al., 1996).
5. Assessment and diagnosis of sleep problems in older adults
Because many factors can contribute to sleep disturbances in older adults, a

thorough assessment of a sleep complaint addresses environmental, behavioral,
medical, psychological, and pharmacological factors (Morin and Edinger, 2003). In
addition to a comprehensive clinical interview, assessment procedures may involve a
physical examination, a psychological evaluation, patient self-monitoring, and/or
sleep laboratory testing to assist in diagnostic clarication.
5.1. Clinical interview
The initial assessment interview should elicit information about the duration,
course, severity, and frequency of the sleep problem. It is also important to ascertain
an understanding of contributing and alleviating factors, daytime consequences/
impairment, and responses to previous treatment. Assessment of poor sleep habits,
disruptive environmental inuences on sleep, sleep schedules, beliefs about sleep,
and family history of sleep disturbance should routinely be assessed (Reynolds, 1996;
White and Mitler, 1997). Inquiry about life events, medical illness, psychiatric
disturbance, and medication or substance use that coincided with the onset of the
sleep diculty can assist in determining etiological factors. Because individuals are
often unaware of sleep behaviors such as snoring, breathing cessation, or nocturnal
movements, interviewing the bed partner can contribute greatly to diagnostic
decision-making.
Medical history should include a review of systems most commonly associated
with sleep diculties, such as rheumatologic (e.g., arthritis, bromyalgia) pulmonary
(e.g., asthma, COPD), cardiac (e.g., heart disease), gastrointestinal (e.g., reux,
peptic ulcer disease), neurological (e.g., seizure disorder), endocrine (e.g.,
hyperthyroidism), and chronic pain disorders. Results of a medical history and
exam may uncover the need for laboratory testing (e.g., thyroid function, ferritin
levels in suspected sleep movement disorders). As mentioned previously, numerous
medications and substances cause or exacerbate sleep complaints. Current medi-
cation usage and use of illicit substances also should be evaluated. Additionally,
some individuals may resort to self-help remedies (e.g., melatonin, alcohol,
antihistamines) to promote sleep. Finally, inquiry regarding psychiatric symptoms
and history can help determine the inuence of psychological factors on the sleep
complaint and the need for further psychological assessment (discussed next).
5.2. Psychological assessment
Because of the elevated prevalence of anxiety and depression among older

adults and the association between psychopathological symptoms and sleep
disturbance, assessment of psychological functioning is important in assessing
sleep complaints. Brief psychological screening measures such as the Beck
Depression Inventory (Beck et al., 1979) and the State-Trait Anxiety Inventories
(Spielberger, 1983) are easily administered and may provide important supple-
mental information not apparent from the clinical interview. A mental status
exam also can be easily administered and may be useful to screen for decits in
cognitive functioning. In some cases, a more thorough psychological assessment with
instruments such as the Minnesota Multiphasic Personality Inventory-2 (MMPI-2)
may be warranted. The MMPI-2 is an extensive psychological questionnaire that
produces personality proles for a wide range of psychopathology in addition to
information on response biases such as attempts to deny or exaggerate symptoms
(Butcher et al., 1989). However, it is a lengthy questionnaire that requires
interpretation by a trained psychologist.
Other psychometric assessment instruments developed specically for use

with sleep-disordered patients can provide useful collateral information. A variety
of self-report sleep questionnaires have been developed to measure hyperarousal,
daytime sleepiness, sleep hygiene behaviors, sleep quality, satisfaction with
sleep, functional impairment, sleep-related beliefs and attitudes, and insomnia
severity (Sateia et al., 2000; Savard and Morin, 2002). These questionnaires are not
immune to subjective biases inherent in self-report methodology, but their utility
may be maximized when combined with other methods such as sleep logs or
actigraphy.
5.3. Sleep logs
Sleep logs provide a practical, simple, and non-intrusive measure of an

individuals perceived sleep pattern over an extended period of time. A typical sleep
log contains questions about bed and wake times, number and duration of nighttime
awakenings and daytime naps, sleep latency, use of sedatives or alcohol, and ratings
of sleep quality and restorativeness (see Fig. 2). This information easily can be
gathered over several weeks to provide information about the variability in sleep/
wake patterns, severity of sleep disturbance, and subjective measures of sleep
quantity and quality. Its main applications are in individuals with insomnia and
circadian rhythm disorders, as it may reveal important diagnostic issues such as
circadian changes, insucient sleep, or perpetuating factors (Savard and Morin,
2002). Sleep logs have been used successfully in older adult populations and also
provide a reliable measurement of treatment outcome (Espie, 2000).
5.4. Actigraphy
The actigraph is a portable behavioral assessment device that derives estimates of

sleep and wakefulness from limb movement activity. It consists of a recording device
that is usually worn on the non-dominant wrist in a manner similar to a wristwatch.
Microprocessors contained in the device enable the storing and downloading of data
into a computer software program that then estimates common sleep/wake
parameters. Actigraphy can be used to inexpensively and unobtrusively collect sleep/
wake data in the home environment over extended periods of time. Current
standards of practice recommend a minimum recording period of three days with
concurrent use of a sleep log to assist in determining sleep periods (Standards of
Practice Committee of the American Academy of Sleep Medicine, 2003). When
combined with information from a clinical interview and sleep log, actigraphy
contributes additional objective data regarding sleep phase changes and variability
of sleep patterns over time. This information can assist in diagnosis, document
severity of the sleep problem, measure treatment outcome, and monitor patient
compliance to treatment recommendations (Ancoli-Israel et al., 2003a; Standards
of Practice Committee of the American Academy of Sleep Medicine, 2003).
Actigraphy has been used successfully in older adults and nursing home patients; in
these populations actigraphy may be particularly useful in documenting the nature
46
DAY OF THE WEEK MONDAY TUESDAY WEDNESDAY THURSDAY FRIDAY SATURDAY SUNDAY
Calendar Date
Yesterday I napped from ___ to ___
(note time of all naps).
Last night I took __ mg of ____ or

____ of alcohol.
Last night I turned off the lights and

attempted to fall asleep at
_____a.m./p.m.
After turning off the lights, it took me
about ___ minutes to fall asleep.
I woke from sleep ___ times. (Do not
M. M. Loiselle et al.
count your final awakening here)
My awakenings lasted ____ minutes.
(List each awakening separately)
Today I woke up at ____a.m./p.m.
NOTE: This is your final awakening.
Today I got out of bed at

____a.m./p.m.
I would rate the quality of last nights
sleep as:
1 = very poor 4 = good
2 = poor 5 = excellent
3 = fair
When I awoke today I felt:
1 = not at all rested 4 = rested
2 = slightly rested 5 = well rested
3 = somewhat rested
Fig. 2. Sample Sleep Log.

of the nocturnal sleep disturbance, assessing the 24 h sleep/wake activity pattern, and
identifying unreported daytime napping (Ancoli-Israel et al., 2003a; Standards of
Practice Committee of the American Academy of Sleep Medicine, 2003).
5.5. Polysomnography (PSG)
The PSG is a gold standard objective test of diagnostic sleep assessment.

This typically includes physiological measurements of EEG, electro-oculography,
chin electromyography, electrocardiography, respiration, oxygen saturation, and
leg muscle activity obtained during an individuals normal sleep period. Recordings
are usually conducted in a sleep laboratory environment, although portable
recorders now allow for physiological data collection to occur in other settings.
This PSG is essential in appropriately diagnosing SDB and is usually indicated
to diagnose PLMD and RBD (American Sleep Disorders Association Standards
of Practice Committee, 1997). However, PSG is not recommended for routine
diagnosis of RLS, insomnia, and circadian rhythm disturbances (American
Sleep Disorders Association Standards of Practice Committee, 1997; Chesson
et al., 2000). These disorders usually can be diagnosed appropriately on the basis of
clinical information; however, PSG may be considered in these cases if there is
evidence of underlying sleep pathologies (e.g., SDB, periodic limb movements)
or a failure to respond to treatment. Due to the extensive nature of this
assessment method, PSG may be dicult to conduct in demented patients (Janssens
et al., 2000).
5.6. Multiple sleep latency test (MSLT)
The MSLT is a laboratory-based objective assessment test to evaluate daytime

sleepiness (Carskadon and Dement, 1977). During this procedure, the patient
is oered approximately ve separate 20-min nap periods every two hours
throughout the day. Sleep latency for each nap opportunity is recorded including
the presence or absence of any REM periods. A mean sleep latency 5 min is
indicative of excessive daytime sleepiness, whereas latencies of 10 min are
considered normal. The MSLT testing may have clinical utility in documenting
sleepiness associated with SDB, PLMD, insomnia, and circadian rhythm disorders
(Thorpy, 1992) but is not considered a necessary part of the diagnostic evaluation
in these disorders.
6. Treatment of sleep disturbances in older adults
6.1. Sleep-disordered breathing
Treatments for SDB include symptom-focused interventions designed to maintain

airway patency during sleep (positive pressure devices and oral appliances) and
treatments to reduce the severity of apneic events (surgical and behavioral
interventions). These treatments are reviewed briey here and discussed in more
detail in Chapter 6.
6.2. Positive pressure devices
Positive pressure devices such as positive airway pressure (PAP) are the treatment
of choice for SDB. By delivering air pressure via a nasal mask, PAP prevents airway
obstruction during sleep and restores normal sleep and breathing patterns. Because
many patients use PAP only partially or are unable to use it long-term, compliance
with this treatment is a critical issue (Grunstein and Sullivan, 2000). Preliminary
studies in older adults with SDB suggest that a brief educational intervention
can enhance PAP use and that higher compliance is associated with greater
improvements in cognitive functioning (Aloia et al., 2001, 2003).
6.2.1. Oral appliances

Oral appliances include a variety of devices designed to enlarge and maintain
airway space by repositioning the mandible and/or advancing the tongue (Lowe,
2000). They are generally indicated for patients with mild SDB or those with more
severe SDB who have failed PAP (Thorpy et al., 1995). Because sucient dentition is
needed to t most of these devices, use in older adults with dentures or otherwise
poor dentition is limited (Lowe, 2000).
6.2.2. Surgical interventions

A number of surgical techniques that attempt to treat SDB by correcting
anatomical abnormalities or removing upper airway tissue have been developed
(Riley et al., 2000). However, because older adults in general have a greater risk of
surgical complications and because data regarding the eectiveness of surgical
treatments for SDB in this age group are lacking, surgery is usually not considered in
this age group (Janssens et al., 2000).
6.2.3. Behavioral interventions

Behavioral interventions such as weight loss and sleep position training are often
used as adjunctive therapy to PAP or in cases where SDB is of mild severity
(Sanders, 2000; Cartwright, 2003). Because SDB is often more severe when the
individual is sleeping in a supine position, sleep position training encompasses a
variety of techniques designed to encourage non-supine sleep (Sanders, 2000;
Cartwright, 2003). Sleeping on a wedge or elevating the head of the bed can help
maintain an open airway during sleep. Positional retraining can be achieved by use
of a position alarm that signals and awakens the patient after movement to the
supine position, or by tennis ball therapy, sewing a vertical pocket on the back of a
T-shirt and lling it with tennis balls to keep the patient from rolling over on his/her
back at night.
The main approaches to treating insomnia are pharmacological and behavioral

interventions. Pharmacological treatments include medications that provide
symptomatic relief to the sleep problem. Behavioral treatments, however, are
designed to reestablish normal sleep/wake functioning, with the goal of eradicating
the behavioral and conditioning factors that perpetuate insomnia. This section
reviews the most common behavioral and pharmacological interventions used to
treat insomnia in older adults.
6.3.1. Behavioral treatments

6.3.1.1. Stimulus control therapy (SCT)
The SCT is a structured behavioral regimen based on the assumption that both
the timing (bedtime) and setting (bed/bedroom) are associated with repeated
unsuccessful sleep attempts and over time become conditioned cues that perpetuate
insomnia (Bootzin, 1972). The goal of SCT is to re-associate the bed/bedroom with
successful sleep attempts by curtailing sleep-incompatible activities that occur in
the bedroom and by enforcing a consistent sleep/wake schedule. Patients are
instructed to go to bed only when sleepy, establish a standard wake-up time, get
out of bed whenever awake for extended periods, refrain from napping, and avoid
sleep-incompatible behaviors (e.g., reading, watching TV, eating, worrying) in the
bed and bedroom.
6.3.1.2. Sleep restriction therapy (SRT)

The SRT reduces sleep disturbance by restricting time in bed each night to
match the patients actual sleep requirement (Spielman et al., 1987). Spending too
much time in bed fragments the sleep pattern and creates excessive time awake
each night. By restricting time in bed, mild sleep deprivation may be induced;
subsequently, sleep drive is increased, wakefulness is reduced, and the sleep pattern is
consolidated.
6.3.1.3. Sleep hygiene

This approach involves education about healthy sleep behaviors and promotion
of sleep-conducive environmental conditions and habits (Hauri, 1991). Patients are
encouraged to exercise daily, eliminate the use of caeine, alcohol, and nicotine, eat
a light snack at bedtime, and ensure that the sleeping environment is quiet, dark, and
comfortable. Sleep hygiene is seldom used as a primary intervention, but is often
included with other interventions such as SC or SRT.
6.3.1.4. Relaxation therapies

A variety of relaxation strategies (e.g., progressive muscle relaxation, passive
relaxation, autogenic training, biofeedback, imagery training, meditation) have been
used to treat insomnia (Stepanski, 2000; McCrae and Lichstein, 2001). The goal of
these therapies is to reduce or eliminate the physiological arousal (e.g., muscle
tension) and/or cognitive arousal (e.g., racing thoughts) associated with insomnia.
Treatment typically entails specic therapeutic exercises and relaxation skills
training. Once the patient gains mastery of relaxation skills, these techniques can
be used to facilitate sleep initiation by reducing sleep-related performance anxiety
and bedtime arousal.
6.3.1.5. Cognitive-behavioral therapy (CBT)

The CBT is a multi-component treatment approach combining cognitive
therapy strategies with behavioral therapies such as SC, SRT, and sleep hygiene
(Morin, 1993). The cognitive component addresses sleep-related misconceptions
and dysfunctional beliefs about sleep, whereas the behavioral component targets
the behavioral and conditioning mechanisms that perpetuate insomnia.
6.3.1.6. Sleep education

Although not a formal intervention strategy, sleep education may be benecial to
older adults who develop unrealistic expectations about sleep. Some retired older
adults believe that since they are retired, they should be able to sleep as long as they
wish. Older adults also may mistakenly believe that they should be obtaining a
certain number of hours of sleep per night, with minimal or no awakenings. In these
cases, it is usually helpful to provide education about the range of human sleep
requirements and the negative eects of too much time in bed. These patients should
be advised that several brief nighttime awakenings are normal and are not likely
to aect their daytime functioning.
6.3.1.7. Effectiveness of behavioral therapies

Behavioral treatments for older adults with insomnia produce signicant
improvements in sleep patterns that are durable over time (Pallesen et al., 1998;
Martin et al., 2000; Stepanski et al., 2003). Many older adults prefer behavioral
treatment to pharmacotherapy because it is perceived to have greater benets and
less side eects (Morin et al., 1992).
6.3.2. Pharmacotherapy
6.3.2.1. Sedative-hypnotic medications
Pharmacological approaches for insomnia typically include benzodiazepine
receptor agonists (e.g., temazepam, zolpidem, zaleplon), sedating antidepressants
(e.g., trazodone, amitriptyline, mirtazapine), or antihistamines (e.g., diphenhydra-
mine). Although these medications eectively treat sleep disturbance, older
adults are at increased risk for drug interaction, accumulation, and toxicity due
to polypharmacy and age-related changes in pharmacokinetics and pharmacody-
namics (Gottlieb, 1990; Monane, 1992; Buysse and Reynolds, 2000). Thus, older
adults are particularly susceptible to the adverse eects of these medications, which
may include daytime sedation, respiratory depression, increased risk of falls, and
impairments in memory, cognition, and psychomotor functioning (Buysse and
Reynolds, 2000; Roehrs and Roth, 2000; Petrovic et al., 2003). Long-term use of
benzodiazepines may result in such iatrogenic eects as drug tolerance, dependence,

hangover, amnesia, and rebound insomnia upon discontinuation (Roehrs and Roth,
2000; Petrovic et al., 2003). Furthermore, there are limited data demonstrating the
ecacy and safety of long-term benzodiazepine use. For these reasons, treatment is
rarely recommended for extended periods (Buysse and Reynolds, 2000; Petrovic
et al., 2003). Although sleep medications may be helpful in treating short-term or
intermittent insomnia, they often provide little lasting benet to patients with
chronic sleep disturbances. Unlike behavioral treatments that address underlying
mechanisms that ostensively cause and sustain insomnia, sedative-hypnotic
medications provide only symptomatic relief.
6.3.2.2. Melatonin
Melatonin is an endogenous hormone produced by the pineal gland. Released in a
circadian pattern during dark cycles, it functions to synchronize circadian rhythms.
Because older adults show age-related reductions in melatonin levels, melatonin
replacement has been tested as a treatment for insomnia. Although melatonin
supplementation in older adults with sleep complaints may improve sleep measures
(Buysse and Reynolds, 2000; Olde Rikkert and Rigaud, 2001) it is seldom
recommended as a long-term treatment for sleep diculties. The FDA does not
regulate the purity, quality, and exact dosage of over-the-counter melatonin
preparations, and little information exists concerning the safety, side eects, and
long-term use of this substance (Buysse and Reynolds, 2000; Martin et al., 2000).
6.4. Secondary insomnia
The treatment of older adults with sleep disturbance secondary to medical or

psychiatric disorders or medication/substance use is of particular importance, given
the prevalence and severity of secondary insomnia in this population (Lichstein,
2000). The primary treatment approach should address the causative disorder.
However, once the primary disorder is successfully treated, insomnia may fail to
remit as it becomes perpetuated by behavioral and conditioning factors. Secondary
insomnia has been successfully treated in older adults using behavioral treatments
described above (Lichstein et al., 1998, 2000).
6.5.1. Bright light therapy

Bright light therapy delivered via a specially designed light box shifts
desynchronized circadian rhythms by exposing the individual to light of a specied
intensity and duration. Bright light ( 2500 lux) for a few hours in the evening over a
number of days advances the circadian phase. This therapy is benecial in treating
advanced sleep phase syndrome and sleep maintenance insomnia in older adults
(Campbell et al., 1993; Murphy and Campbell, 1996; Chesson et al., 1999a;
Van Someren, 2000). Bright therapy also has been associated with cognitive
performance improvements in older adults (Murphy and Campbell, 1996).
6.5.2. Other circadian modulators

Manipulation of other circadian modulators such as body temperature and
physical activity may be benecial in improving sleep patterns in older adults. Warm
baths prior to bedtime have been shown to improve sleep continuity and quality in
older adults (Dorsey et al., 1996, 1999; Liao, 2002). Similarly, physical activity has
been associated with a reduction in sleep fragmentation (Van Someren et al., 1997),
improvements in subjective sleep quality (King et al., 1997), and increased slow wave
sleep (Naylor et al., 2000). The precise mechanisms of how these treatments
inuence circadian systems are unknown. In general, these interventions have not
been rigorously tested, and more information concerning their eectiveness and
application are needed.
6.6.1. Periodic limb movement disorder (PLMD) and restless legs syndrome (RLS)
Studies show PLMD and RLS are typically treated pharmacologically with
medications designed to reduce movement activity and/or improve sleep quality.
Dopaminergic agents (e.g., levodopa, pergolide, pramipexole) are the most
successful treatments (Chesson, et al., 1999b; Edinger, 2003). Other agents that
have shown treatment ecacy include opioids (e.g., oxycodone, propoxyphene),
benzodiazepines (e.g., clonazepam), and anticonvulsants (e.g., carbamazepine,
gabapentin). Iron supplementation can eectively treat RLS secondary to iron
deciency (Chesson et al., 1999b). Eective therapeutic management may also
include treatment of contributing medical disorders or discontinuation of
medications such as antidepressants that can exacerbate movement disorders.
6.6.2. REM behavior disorder (RBD)

Clonazepam, a long-acting benzodiazepine, is the primary treatment for
RBD. Other agents that aect the seroternergic or dopaminergic systems may be
benecial, although some antidepressants can aggravate REM motor activity
(Mahowald and Schenck, 2000). Recently, some data suggest that melatonin
may eectively treat RBD (Boeve et al., 2003). Due to the association between RBD
and neurodegenerative processes, brain imaging is indicated if there is evidence
of neurological dysfunction (Mahowald and Schenck, 2000). Adjunctive treatment
strategies should include avoidance of sleep deprivation and environmental
manipulations to increase safety in the bedroom (e.g., removing dangerous objects,
surrounding the bed with cushions, placing the mattress on the oor, protecting
windows) (Mahowald and Schenck, 2000; Stepanski et al., 2003).
6.7. Sleep in institutionalized older adults
Older adults living in supervised care facilities are particularly susceptible to

suering sleep disruption for a variety of reasons. First, sleep/wake dysregulation is
characteristic of dementing illnesses (Bliwise, 2000a). Second, institutionalized
individuals are likely to take multiple medications and suer from a number of
medical, neurological, and psychological disorders that disrupt sleep. Finally,
environmental factors such as low daytime light levels, long periods of time in bed,
inactivity, daytime napping, and noise often perpetuate sleep disturbances in
institutional settings (Bliwise and Breus, 2000).
Although much more research is needed to determine successful treatments
for sleep disturbance in institutionalized elderly, a number of experts have
made recommendations to maximize sleep potential in these patients (Bliwise,
2000a; Bliwise and Breus, 2000; Martin et al., 2000). One of the easiest and
most eective interventions is the provision of daytime light exposure to
strengthen the circadian rhythm, prevent napping, and improve sleep patterns
(Bliwise and Breus, 2000; Martin et al., 2000; Fetveit et al., 2003). Other
recommended interventions include increasing physical activity, adjusting
medications, providing stimulating daytime activities to prevent napping,
limiting time spent in bed during the day, encouraging routines for meals and
activities, matching roommates with similar schedules, providing an activity
time period during the night to reduce disruption to other residents, and minimizing
nighttime noise and light (Bliwise, 2000a; Bliwise and Breus, 2000; Martin et al.,
2000).
7. Summary
The prevalence of insomnia, SDB, circadian rhythm disorders, and sleep-related

movement disorders increases substantially in late-life. In addition, normal age-
related changes in sleep architecture create a fragmented and light sleep pattern
that sets the stage for sleep complaints in this population. Sleep disturbance in
older adults may be associated with poorer quality of life, dependence on hypnotic
medication, increased risk of falls, increased morbidity, impaired cognitive
performance, and daytime somnolence. A thorough evaluation of a sleep complaint
includes assessment of contributing medical, psychiatric, medication, behavioral,
and environmental factors. Once identied, sleep disorders can be eectively
treated. Treatments include medically-based interventions such as mechanical
devices for SDB and medications for movement disorders as well as behavioral
and environment interventions (e.g., behavioral treatments for insomnia and
bright light therapy for circadian disturbances). However, more information on
the application and eectiveness of these treatments specic to elderly populations
is needed and will only increase in importance as the population of older adults
grows.
References
Aloia, M., Di Dio, L., Ilniczky, N., Perlis, M., Greenblatt, D., Giles, D., 2001. Improving compliance with
nasal CPAP and vigilance in older adults with OAHS. Sleep & Breathing 5(1), 1321.
Aloia, M., Ilniczky, N., Di Dio, P., Perlis, M., Greenblatt, D., Giles, D., 2003. Neuropsychological
changes and treatment compliance in older adults with sleep apnea. Journal of Psychosomatic
Research 54(1), 7176.
Althuis, M., Fredman, L., Langenberg, P., Magaziner, J., 1998. The relationship between insomnia and
mortality among community-dwelling older women. Journal of the American Geriatrics Society
46(10), 12701273.
American Sleep Disorder Association Standards of Practice Committee 1997. Practice parameters for the
indications for polysomnography and related procedures. Sleep 20(6), 406422.
Ancoli-Israel, S., 1989. Epidemiology of sleep disorders. Clinics in Geriatric Medicine 5(2), 347362.
Ancoli-Israel, S., 1997. Sleep problems in older adults: Putting myths to bed. Geriatricse 52(1), 2030.
Ancoli-Israel, S., Cole, R., Chambers, M., Moorcroft, W., Pollack, C., 2003a. The role of actigraphy in the
study of sleep and circadian rhythms. Sleep 26(3), 342392.
Ancoli-Israel, S., DuHamel, E., Stepnowsky, C., Engler, R., Cohen-Zion, M., Marler, M., 2003b. The
relationship between congestive heart failure, sleep apnea, and mortality in older men. Chest 124(4),
14001405.
Ancoli-Israel, S., Kripke, D., Klauber, M., Mason, W., Fell, R., Kaplan, O., 1991a. Periodic limb
movements in sleep in community-dwelling elderly. Sleep 14(6), 496500.
Ancoli-Israel, S., Kripke, D., Klauber, M., Mason, W., Fell, R., Kaplan, O., 1991b. Sleep-disordered
breathing in community-dwelling elderly. Sleep 14(6), 486495.
Babar, S., Enright, P., Boyle, P., Foley, D., Sharp, D., Petrovitch, H., Quan, S., 2000. Sleep disturbances
and their correlates in elderly Japanese American men residing in Hawaii. Journals of Gerontology
Series A-Biological Sciences & Medical Sciences 55A(7), M406M411.
Beck, A., Rush, A., Shaw, B., Emery, G., 1979. Cognitive Therapy of Depression. Guilford Press, New
York.
Blazer, D., Hays, J., Foley, D., 1995. Sleep complaints in older adults: a racial comparison. Journals of
Gerontology Series A-Biological Sciences & Medical Sciences 50(5), M280284.
Bliwise, D., 1993. Sleep in normal aging and dementia. Sleep 16(1), 4081.
Bliwise, D., 1997. Sleep and aging. In: M. Pressman, W. Orr (Eds.), Understanding Sleep: The Evaluation
and Treatment of Sleep Disorders. American Psychological Association, Washington, D.C.
Bliwise, D., 2000a. Dementia. In: M. Kryger, T. Roth, W. Dement (Eds.), Principles and Practice of Sleep
Medicine, Third edition. W.B. Saunders Co., Philadelphia, pp. 10581071.
Bliwise, D., 2000b. Normal aging. In: M. Kryger, T. Roth, W. Dement (Eds.), Principles and Practices of
Sleep Medicine, Third edition. W.B. Saunders Co., Philadelphia, pp. 2642.
Bliwise, D., Breus, M., 2000. Insomnia in dementia and in residential care. In: K. Lichstein, C. Morin
(Eds.), Treatment of Late-Life Insomnia. Sage Publications, Thousand Oaks, pp. 321344.
Boeve, B., Silber, M., Ferman, T., 2003. Melatonin for treatment of REM sleep behavior disorder in
neurologic disorders: Results in 14 patients. Sleep Medicine 4(4), 281284.
Bootzin, R., 1972. Stimulus control treatment for insomnia. Proceedings of the 80th annual convention of
the American Psychological Association 7, 395396.
Brassington, G., King, A., Bliwise, D., 2000. Sleep problems as a risk factor for falls in a sample of
community-dwelling adults aged 6499 years. Journal of the American Geriatrics Society 48(10),
12341240.
Butcher, J., Dahlstrom, W., Graham, J., 1989. Minnesota Multiphasic Personality Inventory-2 (MMPI-2):
Manual for Administration and Scoring. University of Minnesota Press, Minneapolis.
Buysse, D., Reynolds, C., 2000. Pharmacologic treatment. In: K. Lichstein, C. Morin (Eds.), Treatment of
Late-Life Insomnia. Sage Publications, Thousand Oaks, pp. 231267.
Byles, J., Mishra, G., Harris, M., Nair, K., 2003. The problems of sleep for older women: Changes in
health outcomes. Age & Aging 32(2), 154163.
Campbell, S., Dawson, D., Anderson, M., 1993. Alleviation of sleep maintenance insomnia with timed
exposure to bright light. Journal of the American Geriatrics Society 41(8), 829836.
Carskadon, M., Dement, W., 1977. Sleep tendency: An objective measure of sleep loss. Sleep Research
6, 200.
Cartwright, R., 2003. Sleep apnea: A challenge for behavioral medicine. In: M. Perlis, K. Lichstein (Eds.),
Treating Sleep Disorders: Principles and Practice of Behavioral Sleep Medicine. John Wiley & Sons,
Hoboken, pp. 7799.
Chesson, A., Hartse, K., Anderson, W., Davila, D., Johnson, S., Littner, M., Wise, M., Rafecas, J., 2000.
Practice parameters for the evaluation of chronic insomnia. Sleep 23(12), 237241.
Chesson, A., Littner, M., Davila, D., Anderson, M., Grigg-Damberger, M., Hartse, K., Johnson, S.,
Wise, M., 1999a. Practice parameters for the use of light therapy in the treatment of sleep disorders.
Sleep 22(5), 641660.
Chesson, A., Wise, M., Davila, D., Johnson, S., Littner, M., Anderson, W., Hartse, K., Rafecas, J., 1999b.
Practice parameters for the treatment of restless legs syndrome and periodic limb movement disorder.
Sleep 22(7), 961968.
Chokroverty, S., 1994a. Sleep and other medical disorders. In: S. Chokroverty (Ed.), Sleep Disorders
Medicine: Basic Science, Technical Considerations, and Clinical Aspects. Butterworth-Heinemann,
Stoneham, MA, pp. 349368.
Chokroverty, S., 1994b. Sleep disorders in elderly persons. In: S. Chokroverty (Ed.), Sleep Disorders
Medicine: Basic Science, Technical Considerations, and Clinical Aspects. Butterworth-Heinemann,
Stoneham, MA, pp. 401415.
Clark, M., 2001. Restless legs syndrome. Journal of the American Board of Family Practice 14(5),
368374.
Cricco, M., Simonsick, E., Foley, D., 2001. The impact of insomnia on cognitive functioning in older
adults. Journal of the American Geriatrics Society 49(9), 11851189.
Cumming, R., Le Couteur, D., 2003. Benzodiazepines and risk of hip fractures in older people: a review of
the evidence. CNS Drugs 17(11), 825837.
Donahue, J., Lowenthal, D., 1997. Nocturnal polyuria in the elderly person. American Journal of the
Medical Sciences 314(4), 232238.
Dorsey, C., Lukas, S., Teicher, M., Harper, D., Winkelman, J., Cunningham, S., Satlin, A., 1996. Eects
of passive body heating on the sleep of older female insomniacs. Journal of Geriatric Psychiatry &
Neurology 9(2), 8390.
Dorsey, C., Teicher, M., Cohen-Zion, M., Stefanovic, L., Satlin, A., Tartarini, W., Harper, D., Lukas, S.,
1999. Core body temperature and sleep of older female insomniacs before and after passive body
heating. Sleep 22(7), 891898.
Edinger, J., 2003. Periodic limb movements: Assessment and management strategies. In: M. Perlis,
K. Lichstein (Eds.), Treating Sleep Disorders: Principles and Practice of Behavioral Sleep Medicine.
John Wiley & Sons, Hoboken, pp. 100117.
Espie, C., 2000. Assessment and dierential diagnosis. In: K. Lichstein, C. Morin (Eds.), Treatment of
Late-Life Insomnia. Sage Publications, Thousand Oaks, pp. 81108.
Feinsilver, S., 2003. Sleep in the elderly. What is normal? Clinics in Geriatric Medicine 19(1),
177188.
Fetveit, A., Skjerve, A., Bjorvatn, B., 2003. Bright light treatment improves sleep in institutionalised
elderly An open trial. International Journal of Geriatric Psychiatry 18(6), 520526.
Floyd, J., 2002. Sleep and aging. Nursing Clinics of North America 37(4), 719731.
Foley, D., Monjan, A., Masaki, K., Ross, W., Havlik, R., White, L., Launer, L., 2001. Daytime sleepiness
is associated with 3-year incident dementia and cognitive decline in older JapaneseAmerican men.
Journal of the American Geriatrics Society 49, 16281632.
Foley, D., Masaki, K., White, L., Larkin, E., Monjan, A., Redline, S., 2003. Sleep-disordered breathing
and cognitive impairment in elderly JapaneseAmerican men. Sleep 26(5), 596599.
Foley, D., Monjan, A., Simonsick, E., Wallace, R., Blazer, D., 1999. Incidence and remission of insomnia
among elderly adults: an epidemiologic study of 6,800 persons over three years. Sleep 22(Suppl. 2),
S366S372.
Foley, D., Monjan, A., Brown, S., Simonsick, E., Wallace, R., Blazer, D., 1995. Sleep complaints among
elderly persons: an epidemiologic study of three communities. Sleep 18(6), 425432.
Ganguli, M., Reynolds, C., Gilby, J., 1996. Prevalence and persistence of sleep complaints in a rural older
community sample: the MoVIES project. Journal of the American Geriatrics Society 44(7), 778784.
Gooneratne, N., Weaver, T., Cater, J., Pack, F., Arner, H., Greenberg, A., Pack, A., 2003. Functional
outcomes of excessive daytime sleepiness in older adults. Journal of the American Geriatrics Society
51(5), 642649.
Gottlieb, G., 1990. Sleep disorders and their management. Special considerations in the elderly. American
Journal of Medicine 88(3A), 29S33S.
Grunstein, R., Sullivan, C., 2000. Continuous positive airway pressure for sleep breathing disorders.
In: M. Kryger, T. Roth, W. Dement (Eds.), Principles and Practice of Sleep Medicine, Third Edition.
W.B. Saunders Co., Philadelphia, pp. 894912.
Hall, M., Buysse, D., Dew, M., Prigerson, H., Kupfer, D., Reynolds, C., 1997. Intrusive thoughts and
avoidance behaviors are associated with sleep disturbances in bereavement-related depression.
Depression & Anxiety 6(3), 106112.
Hauri, P., 1991. Sleep hygiene, relaxation therapy, and cognitive interventions. In: P. Hauri (Ed.), Case
Studies in Insomnia. Plenum Publishing, New York, pp. 6584.
Hays, J., Blazer, D., Foley, D., 1996. Risk of napping: Excessive daytime sleepiness and mortality in an
older community population. Journal of the American Geriatrics Society 44(6), 693698.
Janssens, J., Pautex, S., Hilleret, H., Michel, J., 2000. Sleep disordered breathing in the elderly. Aging-
Clinical & Experimental Research 12(6), 417429.
Jelicic, M., Bosma, H., Ponds, R., Van Boxtel, M., Houx, P., Jolles, J., 2002. Subjective sleep problems in
later life as predictors of cognitive decline. International Journal of Geriatric Society 17, 7377.
Jennum, P. 2002. Sleep and nocturia. BJU International 90(Suppl. 3), 2124.
King, A., Oman, R., Brassington, G., Bliwise, D., Haskell, W., 1997. Moderate-intensity exercise and self-
rated quality of sleep in older adults. A randomized controlled trial [comment]. JAMA 277(1), 3237.
Kripke, D., Garnkel, L., Wingard, D., Klauber, M., Marler, M. 2002. Mortality associated with sleep
duration and insomnia. Archives of General Psychiatry 59(2), 131136.
Krystal, A.D., Edinger, J.D., Wohlgemuth, W.K., Sharpe, J.M., 2004. Sleep and circadian rhythm
disorders. In: D.G. Blazer, D.C. Steens, E.W. Busse (Eds.), Textbook of geriatric psychiatry, third
edition. Washington, D.C.: American Psychiatric Press, pp. 339350.
Lavie, P., Herer, P., Peled, R., Berger, I., Yoe, N., Zomer, J., Rubin, A., 1995. Mortality in sleep apnea
patients: A multivariate analysis of risk factors. Sleep 18(3), 149157.
Liao, W., 2002. Eects of passive body heating on body temperature and sleep regulation in the elderly:
A systematic review. International Journal of Nursing Studies 39(8), 803810.
Lichstein, K., Riedel, B., Means, M., 1998. Psychological treatment of late-life insomnia. In: R. Schulz,
G. Maddox, M. Lawton (Eds.), Annual Review of Gerontology and Geriatrics. Springer, New York,
pp. 74110.
Lichstein, K., Wilson, N., Johnson, C., 2000. Psychological treatment of secondary insomnia. Psychology
& Aging 15(2), 232240.
Lowe, A., 2000. Oral appliances for sleep breathing disorders. In: T. Kryger, T. Roth, W. Dement (Eds.),
Principles and Practice of Sleep Medicine, Third Edition. W.B. Saunders Co., Philadelphia,
pp. 929939.
Maggi, S., Langlois, J., Minicuci, N., Grigoletto, F., Pavan, M., Foley, D., Enzi, G., 1998. Sleep
complaints in community-dwelling older persons: prevalence, associated factors, and reported causes.
Journal of the American Geriatrics Society 46(2), 161168.
Mahowald, M., Schenck, C. 2000. REM Sleep Parasomnias. In: M. Kryger, T. Roth, W. Dement (Eds.),
Principles and Practices of Sleep Medicine, Third Edition. W.B. Saunders Co, Philadelphia,
pp. 724741.
Manabe, K., Matsui, T., Yamaya, M., Sato-Nakagawa, T., Okamura, N., Arai, H., Sasaki, H., 2000. Sleep
patterns and mortality among elderly patients in a geriatric hospital. Gerontology 46(6), 318322.
Martin, J., Shochat, T., Ancoli-Israel, S., 2000. Assessment and treatment of sleep disturbances in older
adults. Clinical Psychology Review 20(6), 783805.
McCrae, C., Lichstein, K., 2001. Secondary insomnia: A heuristic model and behavioral approaches to
assessment, treatment, and prevention. Applied & Preventive Psychology 10(2), 107123.
Milligan, S., Chesson, A., 2002. Restless legs syndrome in the older adult: Diagnosis and management.
Drugs & Aging 19(10), 741751.
Moe, K., 1999. Reproductive hormones, aging, and sleep. Seminars in Reproductive Endocrinology 17(4),
339348.
Monane, M., 1992. Insomnia in the elderly. Journal of Clinical Psychiatry 53, 2328.
Monk, T., 1989. Circadian rhythm. Clinics in Geriatric Medicine 5(2), 331346.
Morgan, K. 2000. Sleep and aging. In: K. Lichstein, C. Morin (Eds.), Treatment of Late-Life Insomnia.
Sage Publications, Thousand Oaks, pp. 336.
Morgan, K., Clarke, D., 1997. Longitudinal trends in late-life insomnia: Implications for prescribing.
Age & Aging 26(3), 179184.
Morin, C. 1993. Insomnia: Psychological assessment and management. The Guilford Press, New York.
Morin, C., Edinger, J. 2003. Sleep disorders: Evaluation and diagnosis. In: M. Hersen,
S. Turner (Eds.), Adult Psychopathology and Diagnosis, Fourth Edition. John Wiley & Sons, Inc,
Morin, C., Gaulier, B., Barry, T., Kowatch, R., 1992. Patients acceptance of psychological and
pharmacological therapies for insomnia. Sleep 15(4), 302305.
Murphy, P., Campbell, S., 1996. Enhanced performance in elderly subjects following bright light treatment
of sleep maintenance insomnia. Journal of Sleep Research 5(3), 165172.
Myers, B., Badia, P., 1995. Changes in circadian rhythms and sleep quality with aging: Mechanisms and
interventions. Neuroscience & Biobehavioral Reviews 19(4), 553571.
Naylor, E., Penev, P., Orbeta, L., Janssen, I., Ortiz, R., Colecchia, E., Keng, M., Finkel, S., Zee, P., 2000.
Daily social and physical activity increases slow-wave sleep and daytime neuropsychological
performance in the elderly. Sleep 23(1), 8795.
Neubauer, D. 1999. Sleep problems in the elderly. American Family Physician, 59(9), 25512558,
25592560.
Newman, A., Enright, P., Manolio, T., Haponik, E., Wahl, P., 1997. Sleep disturbance, psychosocial
correlates, and cardiovascular disease in 5201 older adults: the Cardiovascular Health Study. Journal
of the American Geriatrics Society 45(1), 17.
Newman, A., Spiekerman, C., Enright, P., Lefkowitz, D., Manolio, T., Reynolds, C., Robbins, J., 2000.
Daytime sleepiness predicts mortality and cardiovascular disease in older adults. Journal of the
American Geriatrics Society 48(2), 115123.
Neylan, T., De May, M., Reynolds, C. 1996. Sleep and chronobiological disturbances. In: E. Busse, D.
Blazer (Eds.), The American Psychiatric Press Textbook of Geriatric Psychiatry, Second Edition.
American Psychiatric Association, Washington, DC.
Nieto, F., Young, T., Lind, B., Shahar, E., Samet, J., Redline, S., D Agostino, R., Newman, A.,
Lebowitz, M., Pickering, T., 2000. Association of sleep-disordered breathing, sleep apnea,
and hypertension in a large community-based study. Sleep Heart Health Study. JAMA 283(14),
18291836.
Ohayon, M., Caulet, M., Lemoine, P., 1996. The elderly, sleep habits and use of psychotropic drugs by the
French population. Encephale 22(5), 337350.
Ohayon, M., Vecchierini, M., 2002. Daytime sleepiness and cognitive impairment in the elderly
population. Archives of Internal Medicine 162, 201208.
Ohayon, M., Zulley, J., Guilleminault, C., Smirne, S., Priest, R., 2001. How age and daytime activities are
related to insomnia in the general population: Consequences for older people. Journal of American
Geriatrics Society 49, 360366.
Olde Rikkert, M., Rigaud, A., 2001. Melatonin in elderly patients with insomnia: A systematic review.
Zeitschrift fur Gerontologie und Geriatrie 34(6), 491497.
Pallesen, S., Nordhus, I., Kvale, G., 1998. Nonpharmacological interventions for insomnia in older adults:
A meta-analysis of treatment ecacy. Psychotherapy: Theory, Research, Practice, Training 35(4),
472482.
Petrovic, M., Mariman, A., Warie, H., Afschrift, M., Pevernagie, D., 2003. Is there a rationale for
prescription of benzodiazepines in the elderly? Review of the literature. Acta Clinica Belgica 58(1),
2736.
Pollak, C., Perlick, D., Linsner, J., Wenston, J., Hsieh, F., 1990. Sleep problems in the community
elderly as predictors of death and nursing home placement. Journal of Community Health 15(2),
123135.
Prinz, P., Vitiello, M., Raskind, M., Thorpy, M., 1990. Geriatrics: Sleep disorders and aging. New England
Journal of Medicine 323(8), 520526.
Reynolds, C. 1996. Sleep disorders. In: J. Sadavoy, L. Lazarus (Eds.), Comprehensive Review of Geriatric
PsychiatryII. American Psychiatric Association, Washington, DC.
Riley, R., Powell, N., Li, K., Guilleminault, C. 2000. Surgical therapy for obstructive sleep apnea-
hypopnea syndrome. In: M. Kryger, T. Roth, W. Dement (Eds.), Principles and Practice of Sleep
Medicine, Third Edition. W.B. Saunders Co, Philadelphia.
Roberts, R., Shema, S., Kaplan, G., 1999. Prospective data on sleep complaints and associated risk factors
in an older cohort. Psychosomatic Medicine 61(2), 188196.
Rockwood, K., Davis, H., Merry, H., Mac Knight, C., Mc Dowell, I., 2001. Sleep disturbances and
mortality: Results from the Canadian Study of Health and Aging. Journal of the American Geriatrics
Society 49(5), 639641.
Roehrs, T., Roth, T. 2000. Hypnotics: Ecacy and adverse eects. In: M. Kryger, T. Roth,
W. Dement (Eds.), Principles and Practice of Sleep Medicine, Third Edition. W.B. Saunders Co,
Philadelphia.
Rumble, R., Morgan, K., 1992. Hypnotics, sleep, and mortality in elderly people. Journal of the American
Geriatrics Society 40(8), 787791.
Russo-Magno, P., OBrien, A., Panciera, T., Rounds, S., 2001. Compliance with CPAP therapy in older
men with obstructive sleep apnea. Journal of the American Geriatrics Society 49(9), 12051211.
Sanders, M. 2000. Medical therapy for obstructive sleep apnea-hypopnea syndrome. In: M. Kryger,
T. Roth, W. Dement (Eds.), Principles and Practice of Sleep Medicine, Third Edition. W.B. Saunders
Co, Philadelphia.
Sateia, M., Doghramji, K., Hauri, P., Morin, C., 2000. Evaluation of chronic insomnia: An American
Academy of Sleep Medicine Review. Sleep 23(2), 243308.
Savard, J., Morin, C. 2002. Insomnia. In: M. Antony, D. Barlow (Eds.), Handbook of Assessment and
Treatment Planning for Psychological Disorders. Guilford Press, New York.
Schechtman, K., Kutner, N., Wallace, R., Buchner, D., Ory, M., 1997. Gender, self-reported depressive
symptoms, and sleep disturbance among older community-dwelling persons. FICSIT group. Frailty
and Injuries: Cooperative Studies of Intervention Techniques. Journal of Psychosomatic Research
43(5), 513527.
Schubert, C., Cruickshanks, K., Dalton, D., Klein, B., Klein, R., Nondahl, D., 2002. Prevalence of sleep
problems and quality of life in an older population. Sleep 25(8), 889893.
Spielberger, C. 1983. Manual for the State-Trait Anxiety Inventory. Consulting Psychologists Press,
Palo Alto.
Spielman, A., Glovinsky, P. 1997. The diagnostic interview and dierential diagnosis for complaints of
insomnia. In: M. Pressman, W. Orr (Eds.), Understanding Sleep: The Evaluation and Treatment of
Sleep Disorders. American Psychological Association, Washington DC.
Spielman, A., Saskin, P., Thorpy, M., 1987. Treatment of chronic insomnia by restriction of time in bed.
Sleep 10(1), 4556.
Standards of Practice Committee of the American Academy of Sleep Medicine, 2003. Practice parameters
for the role of actigraphy in the study of sleep and circadian rhythms: An update for 2002. Sleep 26(3),
337341.
Stepanski, E., Rybarczyk, B., Lopez, M., Stevens, S., 2003. Assessment and treatment of sleep disorders in
older adults: A review for rehabilitation psychologists. Rehabilitation Psychology 48(1), 2336.
Stepanski, E. 2000. Behavioral therapy for insomnia. In: M. Kryger, T. Roth, W. Dement (Eds.),
Principles and Practice of Sleep Medicine, Third Edition. W.B. Saunders Co, Philadelphia.
Thorpy, M., 1992. The clinical use of the Multiple Sleep Latency Test. Sleep 15(3), 268276.
Thorpy, M., Chesson, A., Derderian, S., Kader, G., Millman, R., Potolicchio, S., Rosen, G., Strollo, P.,
Wooten, V., 1995. Practice parameters for the treatment of snoring and obstructive sleep apnea with
oral applicances. Sleep 18(6), 511513.
Van Someren, E. 2000. Circadian and sleep disturbances in the elderly. Experimental Gerontology,
35(910), 12291237.
Van Someren, E., Lijzenga, C., Mirmiran, M., Swaab, D., 1997. Long-term tness training improves
the circadian rest-activity rhythm in healthy elderly males. Journal of Biological Rhythms 12(2),
146156.
Vitiello, M., 1997. Sleep disorders and aging: Understanding the causes. Journals of Gerontology Series
A-Biological Sciences & Medical Sciences 52(4), M189191.
Vitiello, M., Moe, K., Prinz, P., 2002. Sleep complaints cosegregate with illness in older adults: Clinical
research informed by and informing epidemiological studies of sleep. Journal of Psychosomatic
Research 53(1), 555559.
Wang, H., Zhang, X., Yang, Y., Yin, K., Huang, G., 2002. Relationship between sleep apnea hypopnea
syndrome and cardiovascular events in elderly Chinese snorers. Chinese Medical Journal 115(12),
18291832.
Wauquier, A., 1993. Aging and changes in phasic events during sleep. Physiology & Behavior 54(4),
803806.
White, J., Mitler, M. 1997. The diagnostic interview and dierential diagnosis for complaints of excessive
daytime sleepiness. In: M. Pressman, W. Orr (Eds.), Understanding Sleep: The Evaluation and
Treatment of Sleep Disorders. American Psychological Association, Washington, D.C.
Whitney, C., Enright, P., Newman, A., Bonekat, W., Foley, D., Quan, S., 1998. Correlates of daytime
sleepiness in 4578 elderly persons: the Cardiovascular Health Study. Sleep 21(1), 2736.
Williams, R., Karacan, I., Hursch, C. 1974. Electroencephalography (EEG) of human sleep: Clinical
applications. Wiley, New York.
Wing, Y., Chiu, H., Leung, T., Ng, J., 1999. Sleep paralysis in the elderly. Journal of Sleep Research 8(2),
151155.
Wooten, V. 1994. Sleep Disorders in Psychiatric Illness. In: S. Chokroverty (Ed.), Sleep Disorders
Medicine: Basic Science, Technical Considerations, and Clinical Aspects. Stoneham, MA:
Butterworth-Heinemann.
Zizi, F., Jean-Louis, G., Magai, C., Greenidge, K., Wolintz, A., Heath-Phillip, O., 2002. Sleep complaints
and visual impairment among older Americans: a community-based study. Journals of Gerontology
Series A-Biological Sciences & Medical Sciences 57(10), M691694.
Advances in
Cell Aging and
Gerontology
Sleep and learning in animal models

Barry W. Rowa and David Gozala,b,*
a
Department of Pediatrics, Kosair Childrens Hospital Research Institute,
University of Louisville School of Medicine, Louisville, KY 40202, USA
b
Department of Pharmacology and Toxicology, Kosair Childrens Hospital Research Institute,
University of Louisville School of Medicine, Louisville, KY 40202, USA
*Correspondence address: D. Gozal, KCHRI, 570S. Preston Street, Suite 204, Louisville,
KY 40202, USA. Tel: 1-502-852-2323; fax: 1-502-852-2215.
E-mail address: david.gozal@lousiville.edu
Contents
1. Learning and memory
2. Learning, memory, and synaptic plasticity
3. Sleep and behavior
4. Learning-induced sleep alterations
5. Sleep deprivation and learning
6. Reactivation of neural activity patterns in sleep
7. Sleep and learning potential mechanisms
8. Summary
Over the last several decades, behavioral studies in both humans and animals
have implicated sleep in multiple forms of learning and memory, most notably
in those functions that involve the hippocampus. However, the proposed interaction
of sleep and its component stages with memory processes has been a source
of vigorous scientic controversy, primarily due to methodological limitations
in many of these studies (Vertes and Eastman, 2000; Siegel, 2001). Limitations
notwithstanding, animal experiments have been an essential component of
our eorts to dene the impact of sleep on the neural systems involved in learning,
memory, and synaptic plasticity. Animal learning models are especially important
in order to investigate the role of sleep at the cellular and molecular component
levels of synaptic plasticity, the fundamental biological substrates of learning
and memory.
Mammalian sleep can be broadly considered to comprise two distinct states,
namely, rapid-eye movement (REM) sleep (also referred to as paradoxical sleep)

DOI: 10.1016/S1566-3124(05)17003-5
62 B. W. Row and D. Gozal
and non-rapid-eye movement (NREM) sleep (also referred to as slow-wave

sleep). These two distinct states alternate at relatively regular intervals throughout
each sleep period, and are highly regulated by complex, yet mostly nonoverlapping
neuronal circuitry (Tobler, 1995). The REM sleep is characterized by bursts
of rapid eye movements, atonia of postural muscles, and a rapid desynchronous
cortical electroencephalogram (EEG) with the prevalence of a theta rhythm.
Neuronal activity in REM sleep is substantially similar to that seen during waking,
in that neocortical pyramidal cells are tonically depolarized, producing action
potentials at irregular intervals (Coenen, 1995). In the hippocampus, this tonic
depolarization results in synchronized rhythmic activity of 48 Hz, which has been
termed the theta frequency range (Coenen, 1995). In contrast, NREM sleep, which
consists of several distinct stages, is marked by substan-tially lower neuronal
activity than during waking or REM sleep. It is characterized by a slow cortical EEG
that progresses from intermittent spindling to a predominance of high-voltage
slow waves in the delta (0.54 Hz) and sigma frequency ranges (714 Hz), as well
as synchronous neuronal activity in the hippocampus, which are collectively indi-
cative of coordinated bursts of neuronal activity involving large population of
neurons in the cortex and other brain regions (Benington and Frank, 2003). This
type of synchronized rhythmic neuronal activity has been postulated to be
intrinsic to neurons of the cortex and thalamus, as well as other brain regions, and
is produced by the action of voltage-gated calcium and ion channels (Steriade et al.,
1993). The rhythmic neuronal activity of NREM sleep is suppressed by the release
of multiple neurotransmitters during waking, such as acetylcholine, norepinephrine,
serotonin, and histamine, that tonically depolarize neurons via activation of
IP3/DAG and cAMP second messenger systems (Steriade, 1994; McCormick and
Bal, 1997). In REM sleep, acetycholine alone activates IP3/DAG second messenger
systems and suppresses the intrinsic rhythmic neuronal activity, since release of other
neurotransmitters during this stage is minimal (Marrosu et al., 1995; Vazquez and
Baghdoyan, 2001). In contrast, these neurotransmitters are released at relatively
low levels during NREM sleep, resulting in hyperpolarization of neurons (Steriade,
2003).
In most species, NREM sleep makes up about 80% of total sleep time, and the
interval between REM-sleep episodes varies as a function of brain size, from less
than 10 min in mice to about 90 min in humans (Allison and Cicchetti, 1976; Tobler,
1995; Siegel, 2001). The functional signicance of the dierent stages is still unclear
and likely includes a number of dierent roles such as life sustaining, genetic
programming of innate behavior, brain maturation and developmental plasticity,
cortical homeostasis, oculomotor control, and brain repair (reviewed in Lee
Kavanau, 2002), in addition to its putative role as a modulator of learning, memory,
and synaptic plasticity.
1. Learning and memory
Although multiple taxonomies exist, learning and memory can be broadly

classied as comprising two distinct types, typically referred to as declarative and
Sleep and Learning in Animal Models 63
nondeclarative memory (Squire, 1992). Declarative memory (also referred to as

explicit or relational memory) stores information for facts, episodes, and
associations between arbitrarily dierent stimuli and is available for conscious
recollection. In contrast, nondeclarative (also referred to as implicit or procedural)
memory refers to simple forms of conditioning, priming, and the ability to
acquire skills and procedures, and typically involves unconscious processes
(Squire and Zola, 1996). Behavioral, electrophysiological, and anatomical studies
in animals, in conjunction with clinical studies in humans, have led to the proposal
that several neural systems in the brain exist and function simultaneously
and independently to process and store information in parallel (reviewed in
White and McDonald, 2002). The hippocampus and related medial temporal lobe
structures are essential for declarative memory, while the amygdala and matrix
compartment of the dorsal striatum (caudate-putamen) have primarily been
implicated in nondeclarative forms of memory (Squire, 1992; Squire and Zola, 1996).
This biological perspective has important implications for the study of sleep on
learning and memory, as evidence suggests that sleep and its component stages
do not aect all forms of learning in a heterogenous manner.
2. Learning, memory, and synaptic plasticity
Since the initial hypothesis by Ramon y Cajal that the strength of synaptic
connections is not xed, but it is plastic and malleable by neural activity, the
concept of synaptic plasticity has been central to the study of the neurobiology of
learning and memory (Llinas, 2003). Synaptic plasticity is generally thought
to underlie all forms of learning and memory, and consists of any activity-dependent
changes in the strength of connections between neurons, including alterations of
synaptic strength, the composition and number of receptor proteins, pre- and
postsynaptic signal transduction mechanisms, and the number and distribution of
synapses formed between neurons. Conceptual models of synaptic plasticity have
their origins in the theories of Donald Hebb, who postulated that neural
networks would have learning-related properties if only those synaptic inputs
which contribute more signicantly to activating the postsynaptic neuron, were
strengthened, while others were weakened (Hebb, 1949). Experimental ndings in
animals as well as humans suggest that several types of memory exist, and current
conceptual models of synaptic plasticity have attempted to account for the multiple
types of memory (for review see White and McDonald, 2002). The majority of
these models share a common framework, aptly stated as Activity-dependent
synaptic plasticity is induced at appropriate synapses during memory formation, and
is both necessary and sucient for the information storage underlying the type of
memory mediated by the brain area in which that plasticity is observed (Martin
et al., 2000). Although it is also sometimes suggested that plasticity is a general
property of the nervous system, the neuroplasticity that underlies information
storage in a particular system is thought to be localized exclusively within that
system.
3. Sleep and behavior
Sleep is vital for the optimal performance of learning tasks, although the degree to
which sleep is involved in actual memory consolidation, as well as the relative
contributions of REM and NREM sleep stages, is currently a matter of debate
(Vertes and Eastman, 2000; Siegel, 2001). Several lines of evidence have traditionally
been used to support the hypothesis that sleep modulates the reorganization or
consolidation of memories in animals. First, observations that alterations in sleep
architecture occur following learning and/or exposure to environments which
promote synaptic plasticity have led some investigators to hypothesize that sleep
participates in memory consolidation. Second, sleep deprivation following learning
tasks have been shown to interfere with the subsequent retention and performance of
learning tasks. More recently, electrophysiological studies of cellular activity during
sleep have demonstrated that patterns of waking neuronal activity are reactivated
following learning, prompting the suggestion that consolidation of information
occurs during subsequent sleep. Although these approaches are suggestive of sleep as
a modulator of learning and memory, the inherent limitations of these methods have
failed to denitively answer the role of sleep in learning and memory. However, more
recent studies using clever learning algorithms and sophisticated molecular
techniques are beginning to provide us with critical insights into the mechanisms
whereby sleep aects underlying static and dynamic synaptic plasticity (Tononi and
Cirelli, 2003; Huber et al., 2004).
4. Learning-induced sleep alterations
Examination of sleep states after training on learning tasks is based on the

hypothesis that learning requires increased memory processing, and therefore
should be augmented if sleep plays a role in memory consolidation. Enhancement
of REM sleep was rst reported following maze learning in the rat (Lucero, 1970).
Subsequent studies further showed that many dierent learning tasks and protocols,
as well as exposure to stimulating environments which induce synaptic modica-
tions (referred to as enriched environments), will alter the subsequent sleep
architecture in a number of mammalian species (Lucero, 1970; Gutwein and
Fishbein, 1980a; Hennevin et al., 1995). Early observations that training on both
appetitive and aversive tasks, such as maze, operant, avoidance, and classical
conditioning tasks are all associated with an increase in subsequent REM sleep in
animals, have led some investigators to conclude that learning exerts a general eect
on REM sleep (Fishbein et al., 1974; Pearlman and Becker, 1974; Smith and Lapp,
1986; Smith and Rose, 1997). Hennevin et al. (1995) reported that the changes
in post-learning REM sleep changes appeared to be related to both the acquisi-
tion and the degree of complexity of the task itself, since they were not typically
observed after a task had either been mastered or after the application of simple
learning paradigms (Hennevin et al., 1995; Benington and Frank, 2003). More
recently, studies in rodents illustrate that the eect of learning on subsequent sleep
may also be dependent on the neural systems involved. The Morris water maze
is one of the most widely used rodent learning tasks, and can be used to examine the
role of dierent neural systems via dierent congurations that share similar
sensory, motor, and motivational requirements (DHooge and De Deyn, 2001). In
the hippocampal-dependent spatial version of the Morris water maze, rodents learn
to nd a submerged, stable platform in a pool of opaque water that is located in a
constant position via distal, spatial cues. In the cued version of the Morris water
maze, which is hippocampus-independent, but sensitive to lesions of the striatum,
rodents learn to locate a visible platform, the position of which varies from trial to
trial. Smith and Rose (1997) found increases in REM sleep after training on the
hidden platform version of the Morris water maze. However, no changes were
observed after training on the visible (cued) version of the tasks (Smith and Rose,
1997). These ndings therefore suggested that the post-training enhancement of
REM sleep in rodents was selective for hippocampal-dependent tasks (Smith and
Rose, 1997).
Although ndings that REM sleep appears to be increased over baseline levels
during specic time periods following learning tasks in animals have been
consistently reported, there is considerable variability in the literature regarding the
timing of the phenomenon (Smith and Lapp, 1986; Smith, 1996b; Smith and Rose,
1997; Benington and Frank, 2003). This increased REM sleep that typically begins
after training and lasts for a limited period of time, has been termed the REM sleep
window and has been reported to occur as early as the rst 90 min of sleep
following a learning task, and as late as 6 days after completion of a learning task
(Smith, 1996b). In rodents, the REM sleep window varies substantially depending on
the species, strain, type of learning task, and whether a massed or distributed
learning protocol was used (Smith, 1985). This high degree of variability in the
timing of REM-learned task association has led some investigators to question the
validity of the phenomenon (Smith, 1996b; Siegel, 2001; Sanford et al., 2003a,b).
Changes in NREM sleep have been less widely reported following learning tasks
in rodents. However, some studies have shown that exposure to learning tasks, as
well as exposure to enriched environments, alter NREM sleep duration and intensity
(measured as the overall power of delta frequency), typically in conjunction with
increased REM sleep (Gutwein and Fishbein, 1980b; Mirmiran et al., 1982; van
Gool and Mirmiran, 1986; Giuditta et al., 1995; Piscopo et al., 2001). For example,
rats trained on a two-way active avoidance task displayed longer episodes of NREM
sleep, particularly in those sleep episodes that contained a subsequent transition to
REM sleep in the post-learning period (Ambrosini et al., 1995; Giuditta et al., 1995)
(sic). These observations have led some investigators to propose that memory
processing during sleep requires the initial participation of NREM sleep, in addition
to the subsequent involvement of REM sleep (Ambrosini and Giuditta, 2001).
It is important to note that many of the changes in sleep architecture that have
been reported to develop after exposure to learning tasks may not be directly related
to mnemonic processes themselves. Many of the studies have employed negatively
reinforced or aversive learning tasks, which substantially increase stress in
laboratory animals. Increased stress is capable of altering sleep, depending on the
intensity and duration of the stressor (Rampin et al., 1991; del et al., 1995; Siegel,
2001). Although studies showing that pseudo-conditioned animals (animals exposed

to similar environmental stimuli without a learning component) fail to show
similar alterations in sleep architecture, thereby suggesting that stress alone is
insucient to account for learning-induced alteration in sleep architecture, the
non-specic eect of stress induced by the learning tasks themselves cannot
denitively be excluded (Hennevin et al., 1995; Horne, 2000). Also, cholinergic
systems have been extensively implicated in learning and memory as well as sleep
regulation, and Bennington has suggested that strain dierences in cholinergic
activity might underlie increased learning ability and heightened REM sleep,
without the increases in post-learning REM sleep being necessarily involved in
memory processes (Shiromani, et al., 1991; Hasselmo, 1999; Oda, 1999; Benington
and Frank, 2003; Myhrer, 2003). Consistent with this conceptual framework,
variations in the septo-hippocampal cholinergic system have been found to have
functional consequences for dierent types of maze learning in mice, and has been
proposed as an alternative explanation for the strain-dependencies in REM-sleep
windows, as well as for the observation that baseline REM-sleep is positively
correlated with learning abilities in specic inbred lines of mice (Pagel et al.,
1973; Schwab et al., 1992; Schwegler et al., 1996; Benington and Frank, 2003).
Additionally, although increases in REM-sleep amounts have been reported in
several human studies (Peigneux et al., 2001), the existence of REM sleep-windows
in humans has not been well-documented and is still a matter of substantial
contention (Siegel, 2001).
5. Sleep deprivation and learning
Post-training sleep deprivation (PSD) procedures have been widely used to

investigate the eect of sleep on learning and memory. These PSD procedures are
based on the hypothesis that if sleep modulates memory consolidation, then learning
should be aected in sleep-deprived animals. The most commonly used technique
to deprive animals of REM sleep is the pedestal or ower pot technique.
This technique involves placing the animal on a small and narrow platform elevated
over a container of water. Animals are able to obtain NREM sleep while on
the pedestal, but at the onset of REM sleep atonia, the animal will fall into the
water surrounding the pedestal, thereby resulting in relatively selective suppression
of REM sleep. The majority of studies examining the eect of sleep deprivation
in the experimental animal have focused primarily on REM sleep, due to initial
prepositions that REM sleep alone was involved in learning and memory, as well as
the diculty of selectively depriving animals of NREM sleep (Graves et al., 2001).
However, it should be noted that REM PSD manipulations may also aect
NREM sleep. Also, REM sleep deprivation has been reported to alter NREM
sleep architecture (increasing sleep fragmentation and suppressing deeper stages
of NREM sleep) even when total amounts of NREM sleep are preserved (Beersma
et al., 1990; Brunner et al., 1993; Endo et al., 1997). Therefore, many of the eects
of PSD that are presumed to be specic to the stage of sleep that is disrupted
may also be due to disruption of other stages of sleep that are also aected by PSD.
The use of post-training procedures is based on the hypothesis that the processes
underlying the storage of information are initiated by training and continue for some
period following the completion of training (McGaugh 1972a,b; McGaugh et al.,
1972). According to this hypothesis, it should be possible to modulate memory
storage by administering treatments shortly after training, without aecting the
acquisition of the task. Conversely, treatments administered shortly after training
that fail to modulate memory storage may be working through non-mnemonic
processes.
Studies of REM sleep deprivation have been most widely investigated in rodents.
In many cases, REM sleep deprivation following learning has been shown to block
improved task performance on subsequent retesting, although it should be noted
that not all studies have shown impaired learning following sleep deprivation, and
some studies have even shown facilitation of performance (Gisquet-Verrier and
Smith, 1989; Smith, 1996a; Vertes and Eastman, 2000; Benington and Frank, 2003).
The eects of sleep on learning are highly task-dependent and vary considerably
within and across animal species, which makes comparisons dicult (Smith, 1985).
More recently, sleep deprivation studies in rodents have selectively implicated sleep
deprivation in hippocampal function. As mentioned earlier, Smith and Rose (1996)
found that 4 h of sleep deprivation selectively impaired subsequent performance
on a spatial version of the water maze. In contrast, no eect of sleep deprivation
was observed on a cued version (Smith and Rose, 1996). Given that the hidden
and cued versions of the water maze share similar sensory, motor, and motivational
requirements, the dierential eect of sleep deprivation on these tasks suggests
that hippocampal function was selectively disrupted. A similar dissociation of
post-learning REM sleep deprivation was observed by Graves et al. (2003), who
compared the eect of immediate (05 h post-training) and delayed (510 h) sleep
deprivation on contextual and cued fear conditioning, tasks which are hippocampal-
dependent and hippocampal-independent, respectively (Graves et al., 2003; Sanders
et al., 2003). Post-training sleep deprivation from 0 to 5 h signicantly impaired
contextual fear conditioning, while PSD from 5 to 10 h had no eect. In contrast, on
a non-hippocampal dependent cued fear conditioning task, no eect of PSD was
observed, providing further support that hippocampal tasks are especially sensitive
to the eects of REM sleep deprivation (Graves et al., 2003).
The REM sleep deprivation ndings should be interpreted with caution, however.
Traditional methods of sleep deprivation are highly stressful, and this stress has been
a source of signicant criticism owing to the dramatic eects of stress on learning
and synaptic plasticity (Vertes and Eastman, 2000; Garcia, 2001). There is general
consensus among sleep researchers that the small platform method of sleep
deprivation produces results that are nonspecic for SD. Rodents are by nature
active creatures, and the relative immobilization associated with this type of sleep
deprivation alone is highly stressful to the animal, in addition to any other stresses
that may be associated with the technique, such as water aversion (Lee Kavanau,
2002). Animals sleep-deprived via the pedestal technique show substantial
activation of the HPA axis, elevations of plasma corticosterone, and weight loss
(Youngblood et al., 1997; Suchecki et al., 2002) and it is well-established that such
() Learning and memory ()
() Hippocampal LTP (nc or )
(nc) Hippocampal LTD (nc or )

SD Stress
(nc) Membrane Excitability ()
(+) Membrane Adaptation (nc)
(+) [Corticosterone] (+)
Fig. 1. Summary sketch summarizing the dierential eects on of sleep deprivation versus stress on
behavior, stress responses and various synaptic and membrane properties of CA1 hippocampal cells. nc,
not changed, increase, decrease, Modied from McDermott et al. (2003).
stress is highly detrimental to synaptic plasticity and hippocampal functioning

(Garcia, 2001). More benign methods of sleep deprivation, such as gentle
handling or forced locomotion, are also associated with increases in glucocorticoid
release and often fail to show eects on learning and memory (Tobler, 1983; Siegel,
2001). Other techniques of PSD are now being developed and include a variety
of more or less stressful paradigms as well as variable reductions in the global
duration and intensity of NREM and REM sleep. Some of these techniques, which
include mechanical or sound-activated devices (Dematteis et al., 2004), should
undoubtedly allow for better tailored experiments testing the eect of REM and
NREM sleep deprivation after a learning paradigm.
More recent studies have attempted to account for the potential interactions of
stress on the eect of sleep on learning and memory. For example, McDermott et al.
(2003) have recently attempted to dierentiate the eects of sleep deprivation from
those associated with non-specic eects induced by sleep deprivation. To account
for the eects of immobilization stress, they used a single, large platform (LP), which
was big enough for an animal to sleep on, but too small to allow the animal to move
around. To account for the eects of sleep deprivation, they sleep deprived animals
using the multiple platform method (MP), which consists of multiple platforms
too small for the animal to sleep on, but in close enough proximity to allow the
animal to move from one platform to another, resulting in sleep deprivation, but
presumably reducing immobilization stress. Both methods selectively impaired
functioning on a hippocampal-dependent contextual memory task, but had no eect
on a hippocampal-independent cued task. However, the MP method produced a
much greater impairment of freezing behavior in the contextual memory task,
suggesting that stress eects alone are not sucient to account for the eects of
REM sleep deprivation, although they undoubtedly play a signicant role
(McDermott et al., 2003). Because the eects of standard REM PSD are clearly at
least partly due to stress, Wetzel et al. (2003) employed an alternative strategy to
investigate the eects of REM sleep on learning and memory. Using multiple
methods, they tested the hypothesis that increased REM sleep in the post-training
period would improve memory retention in rats. They found that selective post-
training REM sleep enhancement by injection of carbachol, CLIP (corticotropin-
like-intermediate-lobe peptide), or induction of REM sleep rebound, all produced
enhancement of hippocampal-dependent Y-maze discrimination learning in rats,
underscoring the fact that stress is insucient to fully account for the role of sleep in
learning memory and synaptic plasticity (Wetzel et al., 2003).
6. Reactivation of neural activity patterns in sleep
Observations that waking neuronal ring patterns are replayed during sleep
have also been used to support the hypothesis that sleep is involved in learning,
memory, and synaptic plasticity. Pavlides and Winson (1989) initially described
enhanced ring of hippocampal place cells activated during waking during
subsequent REM sleep, and it has been reported that in most cases the observed
ring patterns are temporally compressed or expanded (Pavlides and Winson, 1989;
Louie and Wilson, 2001). These ndings suggested that the neuronal activity of
hippocampal place cells in conscious states may inuence the ring characteristics of
these cells in the subsequent sleep episodes. Later studies have further shown that
pairs of hippocampal neurons whose activity is correlated during a learned behavior
are more likely to show correlated activity in NREM sleep subsequent to the prior
training period (Wilson and McNaughton, 1994; Kudrimoti et al., 1999) and that
the temporal sequence of neuronal ring is preserved in subsequent NREM sleep
(Skaggs and McNaughton, 1996; Lee and Wilson, 2002). This neuronal replay
phenomenon during sleep has also been reported to occur at an accelerated rate
(hence it is often referred to as being time-compressed), and may occur as long as
24 h after the task-specic activation of these neurons (Kudrimoti et al., 1999;
Nadasdy et al., 1999; Hirase et al., 2001; Lee and Wilson, 2002).
Although the replay phenomenon is intriguing, not all investigators have
found evidence for enhanced neuronal ring during REM sleep, or for temporal
alteration of activity patterns during sleep (Nadasdy et al., 1999; Poe et al., 2000).
Poe has attempted to explain these contradictory ndings by suggesting that
neuronal reactivation during REM sleep is modulated by the background theta
oscillation, which is essential for hippocampal-dependent learning, as well as for
cellular models of synaptic plasticity, such as long-term potentiation (Winson, 1978;
Pavlides et al., 1988). Poe et al. (2000) reported that hippocampal neurons previously
active during waking are reactivated during REM sleep in a manner that is highly
dependent upon their temporal phase in relation to underlying theta oscillations.
Neurons that were activated during a familiar maze-running task tended to re at the
troughs of the theta wave in subsequent REM sleep, while neurons that were
activated during a novel task tended to re at the peaks of the theta wave. Electrical
stimulation of the hippocampus at the peak of the theta frequency was predicted and
indeed associated with LTP in this structure, while stimulation at the trough was
associated with another cellular model of synaptic plasticity, namely long-term

depression or LTD (Larson et al., 1986; Greenstein et al., 1988; Huerta and Lisman,
1995; Holscher et al., 1997). These ndings suggest that neuronal connections
are modied during sleep in a manner that provides for the strengthening of
newly acquired memories and the erosion of older ones. However, these ndings
should be interpreted with caution, as it has been noted that the eect of sizes in
these studies are often small, and there is as yet no evidence that reactivation is
functionally involved in synaptic plasticity, and could simply represent an
epiphenomenon of synaptic changes that occurred prior learning (Benington and
Frank, 2003).
7. Sleep and learning potential mechanisms
Recent research has focused on the mechanisms whereby sleep could modulate
learning, memory, and synaptic plasticity. The synaptic plasticity underlying
learning and memory is thought to involve long-term changes in synaptic strength or
ecacy. These alterations of synaptic ecacy are generally thought to be mediated
by activity-induced modications of synapses that are established or consolidated
over time, and typically require new gene expression and protein synthesis
(Wittenberg and Tsien, 2002). The exact molecular mechanisms underlying such
long-lasting neural modications remain to be dened and likely occur via a number
of dierent mechanisms. Sleep has been hypothesized to aect the molecular
and biochemical events underlying memory consolidation by promoting the
synthesis of biologically active molecules that are necessary for the consolidation of
the waking experience. Furthermore, sleep could enhance the release of neuro-
humoral factors important for synaptic reorganization, and also lead to increased
transcription of genes underlying synaptic remodeling, as can be predicted from
the transient elevation of Ca2 concentrations typical of NREM sleep (Buzsaki,
1998; Gardi et al., 2002; Obal et al., 2003). Although this discussion is conned to
some of the cellular and molecular events whereby sleep may promote synaptic
plasticity, it is important to note that use-dependent plasticity has been proposed to
also occur at the network level, and likely participates in memory consolidation
during sleep (see Benington and Frank, 2003 for review).
Cholinergic modulation of the protein kinase A (PKA) signaling pathway and
protein synthesis have been proposed as a potential mechanism whereby sleep
could inuence learning, memory, and synaptic plasticity (Graves et al., 2001). Long-
term memory storage is acutely sensitive to inhibitors of both protein synthesis and
the protein kinase A (PKA) signaling pathway, when such inhibitors are admini-
stered at critical periods following training on learning tasks (Kesner et al., 1981;
Abel et al., 1997; Vazquez et al., 2000; Lattal and Abel, 2001; Bradley and
Finkbeiner, 2002; Naghdi et al., 2003). The ndings that molecular processes during
the post-training period are important for the consolidation of memory, and
observations that post-training sleep is important for memory consolidation, have
led to the suggestion that one of the eects of REM sleep may be to modulate
the PKA signaling cascade, ultimately leading to the induction of gene expression
and the synthesis of new proteins underlying synaptic plasticity (Bourtchouladze

et al., 1998; Graves et al., 2001; Ahi et al., 2004). Acetylcholine (Ach) facilitates
LTP in the hippocampus, and microdialysis studies have shown that during REM
sleep there is a marked increase in Ach release (Auerbach and Segal, 1994; Marrosu
et al., 1995; Matsuyama et al., 2000). Given that the late phase of LTP is dependent
on the PKA pathway, Graves has suggested that the enhanced release of Ach
occurring during REM sleep provides a permissive environment for the induction of
LTP, and presumably facilitates memory consolidation (Nguyen and Kandel, 1997;
Graves et al., 2001).
It is clear that long-term memory formation is dependent on the induction of
gene transcription via activity-dependent signaling pathways (West et al., 2002).
Although the initial ndings were inconclusive, recent studies have begun to
provide support for such a role in sleep (Pompeiano et al., 1997; Tononi and Cirelli,
2001; Cirelli, 2002). The plasticity related gene MMP-9 is up-regulated following
warm ambient temperatures that increase sleep and is decreased by gentle han-
dling under the same temperatures, indicating that sleep may be involved in its
up-regulation (Taishi, 2001). Hippocampal inactivation during REM sleep blocks
up-regulation of the activity-dependent gene zif-268, which integrates a major
calcium signal transduction pathway and is implicated in activity-dependent
synaptic plasticity, following exposure to enriched environments or LTP (Ribeiro
et al., 2002). Using microarray approaches, Cirelli and colleagues have proled the
gene expression patterns in the cortex and cerebellum of awake and sleeping
rats (Cirelli et al., 2004). Although similar numbers of transcripts were found to
be up-cregulated in both sleep and waking, a substantial number of genes involved
in synaptic plasticity were dierentially regulated by sleep and waking. Plasticity-
related genes involved in acquisition and potentiation of learned tasks such as
Arc, BDNF, and NGF1-A were up-regulated by wakefulness. In contrast, plasticity-
related genes involved in consolidation and depression, such as calcineurin and
CAMKIV, were selectively up-regulated by sleep. These ndings suggest that
both sleep and waking states are involved in synaptic plasticity and may favor
dierent cellular processes, both of which contribute to dierent aspects of synaptic
plasticity (Cirelli et al., 2004). In this study, no attempts were made to examine
the relative contributions of the dierent sleep stages to sleep-induced gene
regulation, given that REM and NREM sleep have been hypothesized to be
involved in dierent types of memory (Smith, 2001). However, it remains to be seen
to what degree sleep in general, and more specically REM and NREM sleep
states promote the translation of genes into their active proteins. Early studies
examining incorporation of radioactive amino acids into newly formed proteins
yielded somewhat contradictory results with studies reporting either no eect of
REM SD on in vivo protein synthesis (Bobillier et al., 1971) or only small increases
(Shapiro and Girdwood, 1981). Later studies in rodents and monkeys have demon-
strated positive correlations between cerebral protein synthesis and NREM sleep
(Ramm and Smith, 1990; Nakanishi et al., 1997). Additionally, Sei et al. (2000)
reported that 6 h of REM PSD signicantly decreased NGF levels in the
hippocampus, and BDNF levels in the cortex and cerebellum, indicating
Fig. 2. Rat Cortical and Cerebellar Genes whose mRNA Levels Are Modulated by Sleep and Wakefulness
Independent of Time of Day (A) State-dependent transcripts (GeneChip RGU34A) in cerebral cortex (Cx)
and cerebellum (Cb) and their overlap. (B) Biological functions associated with transcripts with higher
expression in wakefulness (red box) and sleep (blue box). The tree on the left (dots and connecting paths)
represents biological processes; annotations according to the gene ontology hierarchy. Used with
permission from Cirelli, C., Gutierrez, C. M. and Tononi, G. (2004).
that REM sleep may contribute to memory functions via increased secretion of
neurotrophic factors (Sei et al., 2000).
8. Summary
The putative association between sleep states and learning has been strengthened
over the years by a long series of studies supporting a critical role for both
global sleep and specic sleep states in the facilitation of synaptic plasticity. While
the exact mechanisms mediating the various roles of sleep on learning and
memory are for the most part still undened, the overwhelming evidence derived
from sleep deprivation and sleep modication paradigms points to sleep and its
component stages as an important modulator of the acquisition and retention of
learned behaviors. Thus, future studies examining the eects of biological aging
and aging-related disorders on memory and synaptic plasticity phenomena should
incorporate careful assessments of simultaneously occurring alterations in sleep
states, for the latter may markedly inuence the overall ndings of such studies.
Acknowledgments
BWR is supported by F32 HD42395 from NIH; DG is supported by NIH grants

HL69932, HL63912, HL65270, and 2P50HL60296, and The Commonwealth of
Kentucky Research Challenge Trust Fund.
References
Abel, T., Nguyen, P.V., Barad, M., Deuel, T.A., Kandel, E.R., Bourtchouladze, R., 1997. Genetic
demonstration of a role for PKA in the late phase of LTP and in hippocampus-based long-term
memory. Cell 88(5), 615626.
Ahi, J., Radulovic, J., Spiess, J., 2004. The role of hippocampal signaling cascades in consolidation of fear
memory. Behav. Brain Res. 149(1), 1731.
Allison, T., Cicchetti, D.V., 1976. Sleep in mammals: ecological and constitutional correlates. Science
194(4266), 732734.
Ambrosini, M.V., Giuditta, A., 2001. Learning and sleep: the sequential hypothesis. Sleep Med. Rev. 5(6),
477490.
Ambrosini, M.V., Mariucci, G., Bruschelli, G., Colarieti, L., Giuditta, A., 1995. Sequential hypothesis of
sleep function. V. Lengthening of post-trial SS episodes in reminiscent rats. Physiol. Behav. 58(5),
10431049.
Auerbach, J.M., Segal, M., 1994. A novel cholinergic induction of long-term potentiation in rat
hippocampus. J. Neurophysiol. 72(4), 20342040.
Beersma, D.G., Dijk, D.J., Blok, C.G., Everhardus, I., 1990. REM sleep deprivation during 5 hours
leads to an immediate REM sleep rebound and to suppression of non-REM sleep intensity.
Electroencephalogr. Clin. Neurophysiol. 76(2), 114122.
Benington, J.H., Frank, M.G., 2003. Cellular and molecular connections between sleep and synaptic
plasticity. Prog. Neurobiol. 69(2), 71101.
Bobillier, P., Sakai, F., Seguin, S., Jouvet, M., 1971. Deprivation of paradoxical sleep and in vitro cerebral
protein synthesis in the rat. Life Sci. II 10(23), 13491357.
Bourtchouladze, R., Abel, T., Berman, N., Gordon, R., Lapidus, K., Kandel, E.R., 1998. Dierent
training procedures recruit either one or two critical periods for contextual memory consolidation,
each of which requires protein synthesis and PKA. Learn Mem. 5(45), 365374.
Bradley, J., Finkbeiner, S., 2002. An evaluation of specicity in activity-dependent gene expression in
neurons. Prog. Neurobiol. 67(6), 469477.
Brunner, D.P., Dijk, D.J., Borbely, A.A., 1993. Repeated partial sleep deprivation progressively changes
in EEG during sleep and wakefulness. Sleep 16(2), 100113.
Buzsaki, G., 1998. Memory consolidation during sleep: a neurophysiological perspective. J. Sleep Res.
7(Suppl 1), 1723.
Cirelli, C., 2002. How sleep deprivation aects gene expression in the brain: a review of recent ndings.
J. Appl. Physiol. 92(1), 394400.
Cirelli, C., Gutierrez, C.M., Tononi, G., 2004. Extensive and divergent eects of sleep and wakefulness on
brain gene expression. Neuron 41(1), 3543.
Coenen, A.M., 1995. Neuronal activities underlying the electroencephalogram and evoked potentials of
sleeping and waking: implications for information processing. Neurosci. Biobehav. Rev. 19(3), 447463.
del, C.G.M.M., Debilly, G., Valatx, J.L., Jouvet, M., 1995. Sleep increase after immobilization stress: role
of the noradrenergic locus coeruleus system in the rat. Neurosci. Lett. 202(12), 58.
Dematteis, M., Gozal, E. G.D., 2004. A novel procedure for sleep deprivation and sleep fragmentation in
rodents. Sleep 26 (Suppl: AXX).
DHooge, R., De Deyn, P.P., 2001. Applications of the Morris water maze in the study of learning and
memory. Brain Res. Brain Res. Rev. 36(1), 6090.
Endo, T., Schwierin, B., Borbely, A.A., Tobler, I., 1997. Selective and total sleep deprivation: eect on the
sleep EEG in the rat. Psychiatry Res. 66(23), 97110.
Fishbein, W., Kastaniotis, C., Chattman, D., 1974. Paradoxical sleep: prolonged augmentation following
learning. Brain Res. 79(1), 6175.
Garcia, R., 2001. Stress, hippocampal plasticity, and spatial learning. Synapse 40(3), 180183.
Gardi, J., Taishi, P., Speth, R., Obal, F. Jr., Krueger, J.M., 2002. Sleep loss alters hypothalamic growth
hormone-releasing hormone receptors in rats. Neurosci. Lett. 329(1), 6972.
Gisquet-Verrier, P., Smith, C., 1989. Avoidance performance in rat enhanced by postlearning paradoxical
sleep deprivation. Behav. Neural. Biol. 52(2), 152169.
Giuditta, A., Ambrosini, M.V., Montagnese, P., Mandile, P., Cotugno, M., Grassi Zucconi, G., Vescia, S.,
1995. The sequential hypothesis of the function of sleep. Behav. Brain Res. 69(12), 157166.
Graves, L., Pack, A., Abel, T., 2001. Sleep and memory: a molecular perspective. Trends Neurosci. 24(4),
237243.
Graves, L.A., Heller, E.A., Pack, A.I., Abel, T., 2003. Sleep deprivation selectively impairs memory
consolidation for contextual fear conditioning. Learn Mem. 10(3), 168176.
Greenstein, Y.J., Pavlides, C., Winson, J., 1988. Long-term potentiation in the dentate gyrus is
preferentially induced at theta rhythm periodicity. Brain Res. 438(12), 331334.
Gutwein, B.M., Fishbein, W., 1980a. Paradoxical sleep and memory (I): Selective alterations following
enriched and impoverished environmental rearing. Brain Res. Bull. 5(1), 912.
Gutwein, B.M., Fishbein, W., 1980b. Paradoxical sleep and memory (II): sleep circadian rhythmicity
following enriched and impoverished environmental rearing. Brain Res. Bull. 5(2), 105109.
Hasselmo, M.E., 1999. Neuromodulation: acetylcholine and memory consolidation. Trends Cogn. Sci.
3(9), 351359.
Hebb, D.O., 1949. The Organization of Behavior: A neuropsychological theory. John Wiley & Sons,
New York.
Hennevin, E., Hars, B., Maho, C., Bloch, V., 1995. Processing of learned information in paradoxical sleep:
relevance for memory. Behav. Brain Res. 69(12), 125135.
Hirase, H., Leinekugel, X., Czurko, A., Csicsvari, J., Buzsaki, G., 2001. Firing rates of hippocampal
neurons are preserved during subsequent sleep episodes and modied by novel awake experience.
Proc. Natl. Acad. Sci. U S A. 98(16), 93869390.
Holscher, C., Anwyl, R., Rowan, M.J., 1997. Stimulation on the positive phase of hippocampal theta
rhythm induces long-term potentiation that can Be depotentiated by stimulation on the negative phase
in area CA1 in vivo. J. Neurosci. 17(16), 64706477.
Horne, J.A., 2000. REM sleep - by default? Neurosci. Biobehav. Rev. 24(8), 777797.
Huber, R., Ghilardi, M.F., Massimini, M., Tononi, G., 2004. Local sleep and learning. Nature 430(6995),
7881.
Huerta, P.T., Lisman, J.E., 1995. Bidirectional synaptic plasticity induced by a single burst during
cholinergic theta oscillation in CA1 in vitro. Neuron 15(5), 10531063.
Kesner, R.P., Partlow, L.M., Bush, L.G., Berman, R.F., 1981. A quantitative regional analysis of protein
synthesis inhibition in the rat brain following localized injection of cycloheximide. Brain Res. 209(1),
159176.
Kudrimoti, H.S., Barnes, C.A., McNaughton, B.L., 1999. Reactivation of hippocampal cell assemblies:
eects of behavioral state, experience, and EEG dynamics. J. Neurosci. 19(10), 40904101.
Larson, J., Wong, D., Lynch, G., 1986. Patterned stimulation at the theta frequency is optimal for the
induction of hippocampal long-term potentiation. Brain Res. 368(2), 347350.
Lattal, K.M., Abel, T., 2001. Dierent requirements for protein synthesis in acquisition and extinction of
spatial preferences and context-evoked fear. J. Neurosci. 21(15), 57735780.
Lee, A.K., Wilson, M.A., 2002. Memory of sequential experience in the hippocampus during slow wave
sleep. Neuron 36(6), 11831194.
Lee Kavanau, J., 2002. REM and NREM sleep as natural accompaniments of the evolution of warm-
bloodedness. Neurosci. Biobehav. Rev. 26(8), 889906.
Llinas, R.R., 2003. The contribution of Santiago Ramon y Cajal to functional neuroscience. Nat. Rev.
Neurosci. 4(1), 7780.
Louie, K., Wilson, M.A., 2001. Temporally structured replay of awake hippocampal ensemble activity
during rapid eye movement sleep. Neuron 29(1), 145156.
Lucero, M.A., 1970. Lengthening of REM sleep duration consecutive to learning in the rat. Brain Res.
20(2), 319322.
Marrosu, F., Portas, C., Mascia, M.S., Casu, M.A., Fa, M., Giagheddu, M., Imperato, A., Gessa, G.L.,
1995. Microdialysis measurement of cortical and hippocampal acetylcholine release during sleep-wake
cycle in freely moving cats. Brain Res. 671(2), 329332.
Martin, S.J., Grimwood, P.D., Morris, R.G., 2000. Synaptic plasticity and memory: an evaluation of the
hypothesis. Annu. Rev. Neurosci. 23, 649711.
Matsuyama, S., Matsumoto, A., Enomoto, T., Nishizaki, T., 2000. Activation of nicotinic acetylcholine
receptors induces long-term potentiation in vivo in the intact mouse dentate gyrus. Eur. J. Neurosci.
12(10), 37413747.
McCormick, D.A., Bal, T., 1997. Sleep and arousal: thalamocortical mechanisms. Annu. Rev. Neurosci.
20, 185215.
McDermott, C.M., LaHoste, G.J., Chen, C., Musto, A., Bazan, N.G., Magee, J.C., 2003. Sleep
deprivation causes behavioral, synaptic, and membrane excitability alterations in hippocampal
neurons. J. Neurosci. 23(29), 96879695.
McGaugh, J.L., 1972a. Impairment and facilitation of memory consolidation. Act. Nerv. Super (Praha)
14(1), 6474.
McGaugh, J.L., 1972b. The search for the memory trace. Ann. N. Y. Acad. Sci. 193, 112123.
McGaugh, J.L., Zornetzer, S.F., Gold, P.E., Landeld, P.W., 1972. Modication of memory systems:
some neurobiological aspects. Q. Rev. Biophys. 5(2), 163186.
Mirmiran, M., van den Dungen, H., Uylings, H.B., 1982. Sleep patterns during rearing under dierent
environmental conditions in juvenile rats. Brain Res. 233(2), 287298.
Myhrer, T., 2003. Neurotransmitter systems involved in learning and memory in the rat: a meta-analysis
based on studies of four behavioral tasks. Brain Res. Brain Res. Rev. 41(23), 268287.
Nadasdy, Z., Hirase, H., Czurko, A., Csicsvari, J., Buzsaki, G., 1999. Replay and time compression of
recurring spike sequences in the hippocampus. J. Neurosci. 19(21), 94979507.
Naghdi, N., Majlessi, N., Bozorgmehr, T., 2003. The eects of anisomycin (a protein synthesis inhibitor)
on spatial learning and memory in CA1 region of rats hippocampus. Behav. Brain Res. 139(12), 6973.
Nakanishi, H., Sun, Y., Nakamura, R.K., Mori, K., Ito, M., Suda, S., Namba, H., Storch, F.I.,
Dang, T.P., Mendelson, W., Mishkin, M., Kennedy, C., Gillin, J.C., Smith, C.B., Sokolo, L., 1997.
Positive correlations between cerebral protein synthesis rates and deep sleep in Macaca mulatta. Eur. J.
Neurosci. 9(2), 271279.
Nguyen, P.V., Kandel, E.R., 1997. Brief theta-burst stimulation induces a transcription-dependent late
phase of LTP requiring cAMP in area CA1 of the mouse hippocampus. Learn Mem. 4(2), 230243.
Obal, F. Jr., Alt, J., Taishi, P., Gardi, J., Krueger, J.M., 2003. Sleep in mice with nonfunctional
growth hormone-releasing hormone receptors. Am. J. Physiol. Regul. Integr. Comp. Physiol. 284(1),
R131R139.
Oda, Y., 1999. Choline acetyltransferase: the structure, distribution and pathologic changes in the central
nervous system. Pathol. Int. 49(11), 921937.
Pagel, J., Pegram, V., Vaughn, S., Donaldson, P., Bridgers, W., 1973. The relationship of REM sleep with
learning and memory in mice. Behav. Biol. 9(3), 383388.
Pavlides, C., Greenstein, Y.J., Grudman, M., Winson, J., 1988. Long-term potentiation in the
dentate gyrus is induced preferentially on the positive phase of theta-rhythm. Brain Res. 439(12),
383387.
Pavlides, C., Winson, J., 1989. Inuences of hippocampal place cell ring in the awake state on the activity
of these cells during subsequent sleep episodes. J. Neurosci. 9(8), 29072918.
Pearlman, C., Becker, M., 1974. REM sleep deprivation impairs bar-press acquisition in rats. Physiol.
Behav. 13(6), 813817.
Peigneux, P., Laureys, S., Delbeuck, X., Maquet, P., 2001. Sleeping brain, learning brain. The role of sleep
for memory systems. Neuroreport 12(18), A111A124.
Piscopo, S., Mandile, P., Montagnese, P., Cotugno, M., Giuditta, A., Vescia, S., 2001. Trains of sleep
sequences are indices of learning capacity in rats. Behav. Brain Res. 120(1), 1321.
Poe, G.R., Nitz, D.A., McNaughton, B.L., Barnes, C.A., 2000. Experience-dependent phase-reversal of
hippocampal neuron ring during REM sleep. Brain Res. 855(1), 176180.
Pompeiano, M., Cirelli, C., Ronca-Testoni, S., Tononi, G., 1997. NGFI-A expression in the rat brain after
sleep deprivation. Brain Res. Mol. Brain Res. 46(12), 143153.
Ramm, P., Smith, C.T., 1990. Rates of cerebral protein synthesis are linked to slow wave sleep in the rat.
Physiol. Behav. 48(5), 749753.
Rampin, C., Cespuglio, R., Chastrette, N., Jouvet, M., 1991. Immobilisation stress induces a paradoxical
sleep rebound in rat. Neurosci. Lett. 126(2), 113118.
Ribeiro, S., Mello, C.V., Velho, T., Gardner, T.J., Jarvis, E.D., Pavlides, C., 2002. Induction of
hippocampal long-term potentiation during waking leads to increased extrahippocampal zif-268
expression during ensuing rapid-eye-movement sleep. J. Neurosci. 22(24), 1091410923.
Sanders, M.J., Wiltgen, B.J., Fanselow, M.S., 2003. The place of the hippocampus in fear conditioning.
Eur. J. Pharmacol. 463(13), 217223.
Sanford, L.D., Fang, J., Tang, X., 2003a. Sleep after diering amounts of conditioned fear training in
BALB/cJ mice. Behav. Brain Res. 147(12), 193202.
Sanford, L.D., Yang, L., Tang, X., 2003b. Inuence of contextual fear on sleep in mice: a strain
comparison. Sleep 26(5), 527540.
Schwab, C., Bruckner, G., Rothe, T., Castellano, C., Oliverio, A., 1992. Autoradiography of muscarinic
cholinergic receptors in cortical and subcortical brain regions of C57BL/6 and DBA/2 mice.
Neurochem. Res. 17(11), 10571062.
Schwegler, H., Boldyreva, M., Linke, R., Wu, J., Zilles, K., Crusio, W.E., 1996. Genetic variation in the
morphology of the septo-hippocampal cholinergic and GABAergic systems in mice: II. Morpho-
behavioral correlations. Hippocampus 6(5), 535545.
Sei, H., Saitoh, D., Yamamoto, K., Morita, K., Morita, Y., 2000. Dierential eect of short-term
REM sleep deprivation on NGF and BDNF protein levels in the rat brain. Brain Res. 877(2),
387390.
Shapiro, C., Girdwood, P., 1981. Protein synthesis in rat brain during sleep. Neuropharmacology 20(5),
457460.
Shiromani, P.J., Velazquez-Moctezuma, J., Overstreet, D., Shalauta, M., Lucero, S., Floyd, C., 1991.
Eects of sleep deprivation on sleepiness and increased REM sleep in rats selectively bred for
cholinergic hyperactivity. Sleep 14(2), 116120.
Siegel, J.M., 2001. The REM sleep-memory consolidation hypothesis. Science 294(5544), 10581063.
Skaggs, W.E., McNaughton, B.L., 1996. Replay of neuronal ring sequences in rat hippocampus during
sleep following spatial experience. Science 271(5257), 18701873.
Smith, C., 1985. Sleep states and learning: a review of the animal literature. Neurosci. Biobehav. Rev. 9(2),
157168.
Smith, C., 1996a. Paradoxical sleep deprivation and sleep recording following training in a brightness
discrimination avoidance task in Sprague-Dawley rats: paradoxical eects. Neurobiol. Learn Mem.
66(3), 283294.
Smith, C., 1996b. Sleep states, memory processes and synaptic plasticity. Behav. Brain Res. 78(1),
4956.
Smith, C., 2001. Sleep states and memory processes in humans: procedural versus declarative memory
systems. Sleep Med. Rev. 5(6), 491506.
Smith, C., Lapp, L., 1986. Prolonged increases in both PS and number of REMS following a shuttle
avoidance task. Physiol. Behav. 36(6), 10531057.
Smith, C., Rose, G.M., 1996. Evidence for a paradoxical sleep window for place learning in the Morris
water maze. Physiol. Behav. 59(1), 9397.
Smith, C., Rose, G.M., 1997. Posttraining paradoxical sleep in rats is increased after spatial learning in the
Morris water maze. Behav. Neurosci. 111(6), 11971204.
Squire, L.R., 1992. Memory and the hippocampus: a synthesis from ndings with rats, monkeys, and
humans. Psychol. Rev. 99(2), 195231.
Squire, L.R., Zola, S.M., 1996. Structure and function of declarative and nondeclarative memory systems.
Proc. Natl. Acad. Sci. U S A 93(24), 1351513522.
Steriade, M., 1994. Sleep oscillations and their blockage by activating systems. J. Psychiatry. Neurosci.
19(5), 354358.
Steriade, M., 2003. The corticothalamic system in sleep. Front Biosci. 8, d878d899.
Steriade, M., McCormick, D.A., Sejnowski, T.J., 1993. Thalamocortical oscillations in the sleeping and
aroused brain. Science 262(5134), 679685.
Suchecki, D., Tiba, P.A., Tuk, S., 2002. Hormonal and behavioural responses of paradoxical sleep-
deprived rats to the elevated plus maze. J. Neuroendocrinol. 14(7), 549554.
Taishi, P., Sanchez, C., Wang, Y., Fang, J., Harding, J.W., Krueger, J.M., 2001. Conditions that aect
sleep alter the expression of molecules associated with synaptic plasticity. Am. J. Physiol. Regul.
Integr. Comp. Physiol. 281(3), R839R845.
Tobler, I., 1983. Eect of forced locomotion on the rest-activity cycle of the cockroach. Behav. Brain Res.
8(3), 351360.
Tobler, I., 1995. Is sleep fundamentally dierent between mammalian species? Behav. Brain Res. 69(12),
3541.
Tononi, G., Cirelli, C., 2001. Modulation of brain gene expression during sleep and wakefulness: a review
of recent ndings. Neuropsychopharmacology 25(Suppl 5), S28S35.
Tononi, G., Cirelli, C., 2003. Sleep and synaptic homeostasis: a hypothesis. Brain Res. Bull 62(2), 143150.
van Gool, W.A., Mirmiran, M., 1986. Eects of aging and housing in an enriched environment on sleep-
wake patterns in rats. Sleep 9(2), 335347.
Vazquez, J., Baghdoyan, H.A., 2001. Basal forebrain acetylcholine release during REM sleep is
signicantly greater than during waking. Am. J. Physiol. Regul. Integr. Comp. Physiol. 280(2),
R598R601.
Vazquez, S.I., Vazquez, A., Pena de Ortiz, S., 2000. Dierent hippocampal activity proles for PKA and
PKC in spatial discrimination learning. Behav. Neurosci. 114(6), 11091118.
Vertes, R.P., Eastman, K.E., 2000. The case against memory consolidation in REM sleep. Behav. Brain
Sci. 23(6), 867876 discussion 9041121.
West, A.E., Grith, E.C., Greenberg, M.E., 2002. Regulation of transcription factors by neuronal
activity. Nat. Rev. Neurosci. 3(12), 921931.
Wetzel, W., Wagner, T., Balschun, D., 2003. REM sleep enhancement induced by dierent procedures
improves memory retention in rats. Eur. J. Neurosci. 18(9), 26112617.
White, N.M., McDonald, R.J., 2002. Multiple parallel memory systems in the brain of the rat. Neurobiol.
Learn Mem. 77(2), 125184.
Wilson, M.A., McNaughton, B.L., 1994. Reactivation of hippocampal ensemble memories during sleep.
Science 265(5172), 676679.
Winson, J., 1978. Loss of hippocampal theta rhythm results in spatial memory decit in the rat. Science
201(4351), 160163.
Wittenberg, G.M., Tsien, J.Z., 2002. An emerging molecular and cellular framework for memory
processing by the hippocampus. Trends Neurosci. 25(10), 501505.
Youngblood, B.D., Zhou, J., Smagin, G.N., Ryan, D.H., Harris, R.B., 1997. Sleep deprivation by the
ower pot technique and spatial reference memory. Physiol. Behav. 61(2), 249256.
Advances in
Cell Aging and
Gerontology
Sleep disordered breathing in the geriatric

patient population
Alon Y. Avidan
Department of Neurology, Michael S. Aldrich Sleep Disorders Laboratory,
University of Michigan Health System, 8D-8702 University Hospital, Box 0117,
1500 East Medical Center Drive, Ann Arbor, MI 48109-0117, USA
Tel: 1-734-647-9064; fax:+1-734-647-9065.
E-mail address: avidana@umich.edu
Contents
1. Introduction
2. Age-related sleep changes
2.1. Sleep architectural changes
2.2. Sleepiness and aging
3. Age-related changes in respiratory physiology
4. Obstructive sleep apnea in the elderly
4.1. Epidemiology
5. Pathophysiology of sleep disordered breathing in aging
5.1. Obstructive sleep apnea
5.2. Obstructive sleep apnea and clinical consequences
5.3. Central sleep apnea
5.4. Risk factors for obstructive sleep apnea
5.5. Obstructive sleep patients with chronic obstructive
pulmonary disease (COPD)
5.6. Consequences of obstructive sleep apnea in the elderly
5.7. Diagnosis of sleep disordered breathing in the elderly
6. Management of sleep-disordered breathing
6.1. Conservative non-surgical medical management
6.2. Surgical management
7. Conclusion
Abstract. Sleep disordered breathing (SDB) is a spectrum of nocturnal breathing events. It consists
of obstructive, central, and mixed sleep apnea as well as upper airway resistance syndrome and
snoring. Obstructive sleep apnea is characterized by repetitive upper airway obstructive events.
The primary symptoms include excessive daytime sleepiness, snoring, and impairment of cognition.
In some cases, it can be associated with increased risk of motor vehicle accidents. Obstructive sleep
apnea (OSA) varies considerably in its distribution in age and its severity. The exact mechanism of

DOI: 10.1016/S1566-3124(05)17004-7
80 A. Y. Avidan
OSA is not yet fully delineated but multiple factors are generally associated with the development of
the sleep apnea (SA) syndrome. These factors include upper airway anatomic obstruction, as well
as mechanical, neurological, and possibly inammatory changes in the pharyngeal airspace. The
prevalence of the OSA increases with aging. Most studies report a prevalence rate ranging between
11 and 62%. It is further associated with increased mortality in the older patient population.
Central sleep apnea (CSA) occurs in higher frequency among patients with cerebral vascular
accidents, central nervous system (CNS) tumors, CNS infections, encephalopathies, and congestive
heart failure. Patients with cerebral vascular accidents have a markedly elevated prevalence of sleep
apnea mainly in the form of OSA and also have a poorer prognosis after an acute infarct. Patients
with Alzheimers disease also have more frequent SDcB. This SDB should certainly be considered
in the dierential diagnosis of reversible dementia as treatment of the underlying SDB improves
cognitive dysfunction. Medical management of SA relies on treatment with positive pressure
therapy and rarely with use of medications, or with surgery. The role of oral appliances has not
been clearly determined in elderly cohorts. The aim of this chapter is to help shed light on
the epidemiology, pathophysiology, diagnosis, and management of SDB in the geriatric patient
population.
1. Introduction
The older patient population is growing very fast in the United States and around
the world. In the year 2000, 34 million people in the US were older than 65 years. By
the year 2025, this number is expected to almost double to 62 million (Census 2000).
In light of this fact, geriatricians and other health care providers need to manage an
increased number of conditions which are exacerbated with aging. Sleep disordered
breathing is one of these conditions.
2. Age-related sleep changes
2.1. Sleep architectural changes
There are two distinct EEG states that occur during sleep: Rapid eye movement
sleep (REM) and the non-rapid eye movement (non-REM). The non-REM sleep is
subsequently divided into four stages, stages 1, 2, 3, and 4. The normal sleep cycle
consists of cycling from non-REM into REM sleep with a periodicity lasting on the
average for about 90 min. As the night goes on, stages 3 and 4 become shorter and
stage REM sleep becomes longer. The elderly patient experiences increased
fragmentation, decreased sleep eciency and quality of sleep, as well as a decrease
in the amplitude of delta waves which make up stages 3 and 4. There is also a total
reduction in the amount of slow-wave sleep, decreased density of REM sleep,
increasing sleep latency and increased fragmentations of the entire cycle (Avidan,
2002). These changes may be related to the underlying age-related neuronal loss as
well as a disruption of the suprachiasmatic circadian generator.
Sleep Disordered Breathing in the Geriatric Patient Population 81
2.2. Sleepiness and aging
Aging is associated with an increased dissatisfaction with the quality of sleep

(Janssens et al., 2000). Almost 50% of the patients over the age of 65 who live at
home are dissatised with their sleep and two-thirds of those residing in nursing
home facilities suer from sleep disorders (Burger and Shephard, 1993). Among the
secondary causes of sleep disruption, sleep disordered breathing (SDB) has been
increasingly recognized as signicant in the elderly patient population (Hoch et al.,
1990; Ancoli-Israel et al., 1991a; Fleury, 1992). A signicant portion of the elderly
patient population have SDB, but currently there is a great controversy regarding
the clinical signicance of obstructive sleep apnea in this patient population
especially in regard to cognitive impairment, increased morbidity, and mortality
(Phillips et al., 1992; Janssens et al., 2000).
The diagnosis of SDB in the elderly patient population is often delayed or does
not occur because many physicians believe that the elderly patient population is
normally sleepy during the daytime. These patients often present with atypical
clinical presentation, unlike younger patients and the frequency of sleep-related
complaints are often considered to be inherent to the normal aging process.
From a technical standpoint it is often dicult to perform polysomnography in
this patient population, especially in those who have cognitive impairment
(Janssens et al., 2000).
A large proportion of the elderly patient population suers from sleep
disturbances caused by numerous factors ranging from polypharmacy, changes in
sleep-wake pattern, retirement, the loss of spouse, and prolonged bed rest. Recently,
however, new ndings indicate that sleep disturbances in elderly should not
be viewed, or accepted as a part of the normal aging process. It is now well
recognized that many seniors suer from sleep disorders that are pathologic and
treatment of these disturbances may signicantly improve complaints such as
daytime sleepiness, napping, and potentially control hypertension and reverse
cognitive impairments. The most important sleep disorder to consider in the elderly
is sleep apnea. Before providing an account of the pattern of sleep disordered
breathing, a consideration of the age-related changes in respiratory physiology and
anatomy deserve mention.
3. Age-related changes in respiratory physiology
As the lungs age, they undergo various predictable anatomic and physiologic
changes (Fig. 1). The major anatomic change is smaller airway size due to alterations
in the supporting connective tissue (Chan and Welsh, 1998). Additional anatomic
changes consist of decreased bronchiolar diameter, increase in the diameter of the
alveolar ducts and shallower alveolar sacs owing to changes in the proportion of
decreased elastic tissues and increased collagen that occur with aging. This
histological change translates physiologically to decreased elastic recoil of the
lung, decrease in oxygen diusion capacity, premature airway closure leading to
82
Diminishing of Smaller airway size-
ventilatory response Due to alteration in
to hypoxemia supporting
and hypercapnia connective tissue:
Increase diameter of
alveolar ducts-Alveolar
sac becomes shallower
Bronchiolar
Collagen
Narrowing Elastic
recoil
Elastic tissue
A. Y. Avidan
O2 diffusion capacity
Diaphragmatic and Chest Wall compliance

abdominal breathing decreases due to age-
become more critical associated kyphosis, &
than that generated calcification of the
by the thoracic muscles intercostals cartilage,
Decreased ability to generate

inspiratory effort during the
obstructive apneic spell
Fig. 1. Physiologic and anatomic changes that occur in pulmonary function with aging.
Age-Related changes in LungVolume
Normal Lung Aged Lung
INSPIRATORY INSPIRATORY
RESERVE VOLUME RESERVE VOLUME
TIDAL VOLUME
TIDAL VOLUME EXPIRATORY
RESERVE
EXPIRATORY VOLUME
RESERVE
VOLUME
RESIDUAL
VOLUME
RESIDUAL VOLUME
VITAL CAPACITY
FUNCTIONAL RESIDUAL CAPACITY
Fig. 2. Age related changes in pulmonary function.
ventilation-perfusion mismatching, and increased alveolar-arterial oxygen gradient,

and small airway closure resulting in air trapping (Chan and Welsh, 1998).
In addition to these histological changes, the chest wall and muscles of respiration
also undergo changes which may impact SDB in aging (Fig. 1). Aging is often
associated with kyphoscoliosis, calcication of intercostal cartilage, and arthritis of
the costovertebral joints. When the thoracic cage becomes more rigid, the patient
starts relying on diaphragmatic and abdominal musculature for breathing. Added to
this is the prevalence of age-related diaphragmatic defects relating to diminution of
diaphragmatic strength (Chan and Welsh, 1998). The elderly undergo predictable
changes in pulmonary function (Fig. 2). Vital capacity decreases due to increased
chest wall stiness, loss of the elastic recoil of the lung, and decreased force
generated by the respiratory muscles. This leads to increased residual volume,
increased functional residual capacity (FRC), decrease in vital capacity, and decrease
in inspiratory reserve volume (Chan and Welsh, 1998).
The aging process is associated with a diminishing of ventilatory response to
hypoxemia and hypercapnia (Kronenberg and Drage, 1973; Peterson et al., 1981;
Brischetto et al., 1984; Janssens et al., 2000). Kronenberg and Drage studied the
response to hypercapnia and hypoxemia in younger men (ages 2230) compared to
older men (ages 6473). They found that in the older cohort, the ventilatory response
84 A. Y. Avidan
to hypercapnia was substantially lower (decreased to 60%) and that the response to
hypoxemia was also lowered (by as much as 75%) when compared to younger
patients (Kronenberg and Drage, 1973). This observation suggests an underlying
age-related decline in the ability to interpret and integrate information derived from
peripheral and central chemoreceptors and mechanoreceptors to generate an
appropriate neuronal feedback response (Kronenberg and Drage, 1973; Janssens
et al., 2000). During sleep, the hypoxic ventilatory response is diminished more
profoundly during non-REM than during REM sleep (Burger and Shephard,
1993). In addition, sleep deprivation further diminishes the nocturnal ventilatory
response to hypoxemia and hypercapnia (Grassino and Begin, 1990; Janssens et al.,
2000). The arousal response from sleep is often preserved during hypercapnia but is
often markedly impaired during hypoxemia. With aging, impaired ventilatory
response may lengthen the duration of the respiratory disturbances (apneas and
hypopneas) and may result in more profound hypoxemia, particularly during REM
sleep. Aging is also associated with diminished perception of added resistive loads
such as generated by upper airway collapse (Tack et al., 1981; Manning, 1993). Older
patients also have decreased ability to generate inspiratory eort during the
obstructive apneic spell compared to younger people (Krieger et al., 1997; Janssens
et al., 2000).
4. Obstructive sleep apnea in the elderly
There are several dierent types of SDB. Obstructive sleep apnea (OSA), the
most ubiquitous type of SDB was rst described in 1976 (Guilleminault et al., 1976).
It is dened as a complete obstruction in the upper airway despite the subjects
attempt to resume normal respiration (Peppard et al., 2000). Obstructive sleep
apnea consists of repeated cessation in respiration during sleep. The episodes
which by denition last for ten seconds or longer can be complete (apneas) or there
may be a partial impairment in ventilation (hypopneas) in spite of continuous
respiratory eort (Fig. 3). Obstructive sleep apnea can result in a drop in oxyhe-
moglobin concentration and are often proceeded by a brief electroencephalogram
(EEG) arousal pattern as can be seen by a shift to a higher EEG frequency pattern
(Fig. 3). The gure also illustrates a delay (of about 15 s) in the desaturation
event from the beginning of the apnea. This lag between onset of apnea and
the drop in oxygen saturation is a reection of the lag due to circulatory delay during
which the physiologic response to hypoxemia takes places. Some patients
may experience cardiac arrhythmias in association with the desaturation event
which can some-times be fatal. There is also a surge of sympathetic activity
associated with elevation of blood pressure and tachycardia (Somers et al., 1995).
This increase in sympathetic tone has been proposed as a mechanism of action
implicating increased cardiovascular morbidity with OSA (Narkiewicz and Somers,
1997; Morgan et al., 1998; Young et al., 2002). Patients with OSA may also
experience symptoms such as hypersomnolence, concentration problems, memory
disturbances, nocturnal hypertension, nighttime arousals, confusion, and impair-
ment in the neuropsychologic functioning. Many of these problems can be found in
Sleep Disordered Breathing in the Geriatric Patient Population
1
Fig. 3. Obstructive sleep apnea. Illustrated in this gure is a 1 min epoch from a diagnostic polysomnogram of a 73 year old man with a history of nocturnal
breathing cessation, snoring, and daytime somnolence. Obstructive sleep apnea characterized by nasal oral (N/O) breathing cessation (1) in the presence of
persistent respiratory eort (2), and hypoxemia (3). Snoring was note electrographically and heard by the technicians monitoring the patients. (Snore channel,
prior to and subsequent to the apneic event). Channels are as follows: Electrooculogram (left: LOC-A2, right: ROC-A1), chin EMG (Chin-Chin), EEG [Left
85
central (C3-A2), right central (C4-A1), left occipital (O1-A2), right occipital (O2-A1)], electrocardiogram (ECG), limb EMG [left leg (LAT), right leg (RAT)],
snoring (SNORE), nasal-oral airow (N/O), respiratory eort [Thoracic (THOR), Abdominal (ABD)], nasal pressure (NPRE), and oxygen saturation (SpO2).
86 A. Y. Avidan
the elderly population cand many times are incorrectly attributed to the aging
process (Quinnell and Smith, 2004).
Bliwise has hypothesized that sleep apnea in the elderly may represent two
dierent conditions (Bliwise, 1993a). One form is age-related and the other is age-
dependent. The age-related sleep apnea develops in middle age. The age-dependent
sleep apnea is likely to develop with advancing age and may be secondary to other
age-dependent conditions such as increasing weight and increased upper airway
collapsibility. It is also related to a decreased pharyngeal muscle response to
negative pressure as well as to decreased airway size. This may be due to increased
parapharyngeal fat deposits as well as such other comorbidities such as endo-
crinopathies consisting of hypothyroidism as will be discussed later in the text. The
concept is that age-dependent sleep apnea is really a product of age itself. The longer
one lives, the more likely he or she is to develop the condition, which is in fact
the case based on observations of older community volunteers who are more
likely to develop OSA compared to their younger cohorts (Bliwise, 1993b, 2000).
Figure 4A illustrates the typical hypnogram of a patient with OSA. The x-axis is
the hours of recording during the night and the y-axis is the oximetry curve and sleep
stages. The rst half of the gure demarcates the baseline untreated (before positive
pressure therapy) portion of the night. The overall pattern is that multiple
awakenings and multiple episodes of hypoxemia produce disrupted nocturnal sleep
which can lead to excessive daytime somnolence, increased napping, and sleepiness.
Figure 4B illustrates the typical morphology of an apnea; breathing cessation,
terminating in a drop in the blood oxygen level leading to an EEG arousal response
permitting the subject to resume eective breathing. Once positive pressure therapy
begins (Fig. 4A), oximetry readings normalize, and the patient has consolidated
sleep. Figure 4C depicts resumption of normal breathing with positive pressure
therapy. The next set of questions that need to be addressed are how frequent is
sleep apnea in the elderly, and what is its epidemiology distribution among various
types of geriatric communities.
4.1. Epidemiology
The prevalence of sleep apnea in healthy adult population is based on minimal

diagnostic criteria of more than ve apneas or hypopneas per hour of sleep
(Young et al., 1993). The prevalence of undiagnosed SDB is high among men
(4%) and is much higher than previously suspected among women (2%) (Young
et al., 1993). For patients over the age of 65 the prevalence of sleep apnea is increased
(Shochat and Pillar, 2003). Subjective reports from a large epidemiologic study in
the United States have found that the prevalence of reported sleep apnea to be
as high as 4% for women and 13% for men while the prevalence of reported snoring
was much higher, 19% for women and 33% for men (Enright et al., 1996; Shochat
and Pillar, 2003).
Although there is a general agreement that sleep apnea is common in the aging
population there is a lack of agreement about the actual prevalence of sleep apnea in
Baseline Portion (Without CPAP) Begin CPAP therapy Resolution of OSA with higher CPAP pressures
PANEL A PANEL B
Sleep fragmentation, hypoxemia 5cm7cm9cm11cm water
REM

MOV AWK
1
2
3
4
30
PRES
0
100
Sp O2
70
Time 22:44:00 06:40:30
Hrs
Epoch 0 1 2 3 4 5 6 7 8
8 128 248 368 488 608 728 848 968
Fig. 4A. Sleep hypnogram depicting the baseline portion of a spilt night polysomnogram. Panel A demonstrates repeated hypoxemia (upper 70s) associated
with the apneic episodes which resolve once positive pressure (CPAP) is introduced in panel B. As higher CPAP levels are used (5 cm !7 cm !9 cm !11 cm
water), the sleep becomes less fragmented and hypoxemia is normalized. Channels are as follows: Electrooculogram (left: LOC-A2, right: ROC-A1), chin EMG,
EEG (left central, right central, left occipital, right occipital), electrocardiogram, limb EMG (left leg, right leg), snoring, nasal-oral airow, respiratory eort
87
(thoracic, abdominal), nasal pressure, and oxygen saturation.

88

A. Y. Avidan
#
*
** **
Fig. 4B. A 1 min epoch from the baseline portion (Panel-A) of the study demonstrating obstructive sleep apnea characterized by nasal oral (N/O) breathing
cessation (*) in the presence of persistent respiratory eort (**) EEG arousals () and hypoxemia ( ). An example of an hypopnea (#) is shown towards the end
of the epoch (#) demonstrating partial as opposed to complete (*) airway obstruction. Channels are as follows: Electrooculogram (left: LOC-A2, right: ROC-
A1), chin EMG (Chin-Chin), EEG [Left central (C3-A2), right central (C4-A1), left occipital (O1-A2), right occipital (O2-A1)], electrocardiogram (ECG), limb
EMG [left leg (LAT), right leg (RAT)], snoring (SNORE), nasal-oral airow (N/O), respiratory eort [Thoracic (THOR), Abdominal (ABD)], nasal pressure
(NPRE), and oxygen saturation (SpO2).
Fig. 4C. Treatment with CPAP (Panel-B) showing resolution of the apneas, hypopneas, hypoxemia, and absence of sleep disruption and arousals. Channels are
as follows: Electrooculogram (left: LOC-A2, right: ROC-A1), chin EMG (Chin-Chin), EEG [Left central (C3-A2), right central (C4-A1), left occipital (O1-A2),
89
right occipital (O2-A1)], electrocardiogram (ECG), limb EMG [left leg (LAT), right leg (RAT)], snoring (SNORE), mask ow signal (MFLO), nasal-oral
airow (N/O), respiratory eort [Thoracic (THOR), Abdominal (ABD)], nasal pressure (NPRE), and oxygen saturation (SpO2).
90 A. Y. Avidan
Table 1. Prevalence of sleep apnea in the elderly

Reference N Age (Years) OSA (%)
Carskadon and Dement, 1981 40 6285 38

Coleman et al., 1981 83 6085 39
Roehrs et al., 1983 97 6181 27
Yesavage et al., 1985 41 69.5 6.5 73
Ancoli-Israel et al., 1985 145 6595 18
Mosko et al., 1988 46 6095 26
Ancoli-Israel et al., 1991 427 6595 24
this patient population. One of the earliest prevalence studies of sleep apnea in the
elderly population was performed in 1974 by Webb who found that nearly 75% of
healthy volunteer men had periodic breathing with apneas (Webb, 1974). Since then,
there have been other studies documenting the prevalence of sleep apnea in the
elderly patient population. These are summarized in Table 1 (Carskadon and
Dement, 1981; Coleman et al., 1981; Roehrs et al., 1983; Yesavage et al., 1985;
Ancoli-Israel et al., 1985, 1991a; Mosko et al., 1988). The studies show a wide
variability in the reported prevalence of OSA. This discrepancy in studies may be
due to analysis of small sample size and lack of studies evaluating randomly selected
patient population. Discrepancies may also be related to how we conduct
measurements in these studies; how we screen patients for sleep apnea, what criteria
of denition and quantication of respiratory events we use, and what tools we use
in evaluating patients (Janssens et al., 2000). A study by Hoch et al. (1990) looking
at the prevalence and severity of SDB in healthy 80-year-old subjects as compared
with 70 and 60-year-old subjects found that the Apnea Hypopnea Index (AHI)
increased signicantly across decades: 39.5% of 80-year olds, 33.3% of 70-year olds,
and 2.9% of 60-year olds had an AHI greater than or equal to 5 (Fig. 5). Signicant
gender dierences were noted in the proportion of subjects with AHI greater than
or equal to 10: 22.4% of men versus 5.4% of women. These data suggest that SDB
increases with advancing age even in the healthy elderly and may be more marked
in healthy men than women (Hoch et al., 1990).
The prevalence in women increases after menopause possibly due to declining
levels of estrogen and progesterone. The higher male prevalence and the severity of
OSA disappear in men older than 55 years, and menopause seems to play a pivotal
role in modulating both the presence and the degree of OSA (Resta et al., 2003).
In fact, data from a population-based sample of 589 women enrolled in the
Wisconsin Sleep Cohort Study demonstrated that the menopausal transition is
signicantly associated with an increased likelihood of having SDB, independent of
known confounding factors (Young et al., 2003). The authors proceed to suggest
that evaluation for SDB should be a priority for menopausal women with complaints
of snoring, daytime sleepiness, or unsatisfactory sleep (Young et al., 2003). The
role of hormone replacement therapy in SDB was examined in a cohort consisting
of 2852 women, 50 years of age or older, who participated in the Sleep Heart
Health Study (Shahar et al., 2003). The authors found that the prevalence of SDB
OSA Severity and Age
40
35
30
25
% of
20
Subjects
AHI>5
15
AHI>10
10
5
0
60-69 70-70 80-89 total
Years of age
AHI = Apnea-hypopnea index (number of apneas and hypopneas per hour of sleep)
Fig. 5. OSA severity and age (from Hoch et al. (1990)).
(as dened by an AHI of 15 or more) among hormone users was approximately

half the prevalence among non-users. The inverse association between hormone use
and SDB was evident in various subgroups and was particularly strong among
women 50 to 59 years old (Shahar et al., 2003). An earlier study from 2001 found
that postmenopausal women had a signicantly higher mean AHI compared
to premenopausal women and this signicant dierence persisted even after
adjusting for BMI (body mass index) and neck circumference. It may be functional
or physiologic, rather than anatomic, dierences in the upper airway between
premenopausal and post-menopausal women, which may account for the observed
dierences in apnea prevalence and severity (Dancey et al., 2001). There is a
general lack of data as to why OSA becomes more prevalent after menopause but it
may theoretically be related to progesterone and estrogen dysfunction (progesterone
function as a respiratory stimulant, and estrogen eects body fat distribution).
5. Pathophysiology of sleep disordered breathing in aging
5.1. Obstructive sleep apnea
Studies indicate that OSA may occur at multiple levels of obstruction including
the nasopharynx, oropharynx, and hypopharynx (Fig. 6). Predisposed individuals
include patients with reduced oral pharyngeal airway including patients with
macroglossia, hypertrophy of the tonsils, and patients with long uvula (Shochat
and Pillar, 2003). This has led to further implications that the underlying
pathophysiology of OSA is due to abnormal anatomy. It is hypothesized that
with the onset of sleep, patients with obstructive sleep apnea lose the neuromus-
cular compensation present during wakefulness leading to airway collapse and
92
Progesterone ?
+
Nasopharynx Nasal-septopalsty
Positive pressure UPPP
therapy: CPAP, BiPAP Oropharynx
Maxillomandibular Advancement
SSRI CN XII
+ Hypopharynx
Genioglossus Advancement
Hyoid Myotomy
TCA
Trachea Tracheaostomy
A. Y. Avidan
REM
Oral Appliance
Treatment Options for OSA in the Elderly

Fig. 6. Treatment options for obstructive sleep apnea (OSA).
subsequently to OSA (Mezzanotte et al., 1992). A faulty upper airway anatomy

and loss of pharyngeal dilator muscle activation results in collapse of the pharynx
which leads to the obstructive event. There are also theories to explain the reasons
behind the age-related OSA, looking at the both the pharyngeal anatomy and
physiology which becomes even more collapsible with increasing age (Shochat
and Pillar, 2003). In the elderly, OSA may become more prominent due to the loss
of genioglossus muscle response to negative pressure and due to proliferation of
prarapharyngeal fat deposits which compromise the upper airway anatomy
(Malhotra et al., 2000).
5.2. Obstructive sleep apnea and clinical consequences
OSA is dened as repetitive episodes of upper airway obstruction occurring

during sleep cessation of airways for at least 10 s (apnea) or a reduction by 50% in
airow for about 10 s (hypopnea) occurring despite continuing respiratory eort.
The absence of a respiratory eort with a loss of airow constitutes a central apneic
episode. The upper airway obstruction is a consequence of hypotonic muscular tone
of the dilator muscles of the pharynx which increase the collapsibility of the upper
airway and the negative airway pressure generated by the inspiratory muscles; the
gravitational eects of supine posture usually adopted for sleep tend to further
promote the upper airway obstruction. The posterior movement of the tongue
especially in the supine position further compromises the upper airway space. The
principal symptoms of obstructive sleep apnea include the following: daytime
sleepiness, moderate neuropsychiatric impairment which results from repeated
arousals required to re-establish sucient muscular tone in the pharynx to reopen
the upper airways, and reduced sleep eciency and quality which is reected by a
decrease in stage 3, stage 4, and REM sleep.
5.3. Central sleep apnea
Central sleep apnea, a less frequent sleep disordered breathing syndrome, results
from a complete or partial impairment of airow in the absence of a respiratory
eort (Shochat and Pillar, 2003). Polysomnographic recording in CSA demonstrates
central breathing cessation for 10 s or longer and may be associated with frequent
arousals from sleep, bradytachycardia, and arterial oxygen desaturation.
Central sleep apnea can occur in the form of CheyneStokes respirations which
is a unique form of central sleep apnea whereby the respiratory pattern consists
of crescendodecrescendo pattern followed by a central apnea which cycles for
about a minute (Shochat and Pillar, 2003). This is a common and an important
form of SDB to consider in aging and especially in patients with an underlying
cerebrovascular accident, and congestive heart failure. Its presence is considered
as a poor prognostic indicator (Naughton and Bradlery, 1998).
94 A. Y. Avidan
5.4. Risk factors for obstructive sleep apnea
Obstructive sleep apnea is associated with multiple risk factors. The main ones are
obesity, increased neck circumference, family history of OSA, alcohol consumption,
and endocrine disorders. Other risk factors or aggravating factors include tobacco,
sedative or hypnotic use, sleeping in a supine position, and sleep deprivation.
5.4.1. Obesity
A major risk factor for the development of OSA is obesity. The airways may
become compressed extrinsically by excessive deposition of parapharyngeal adipose
tissue which restricts the airway size making it prone to collapse (Quinnell and Smith,
2004). Upper airway morphology that is associated with sleep apnea includes reduced
pharyngeal cross sectional area and retrognathic mandible. While approximately
24% of the population is estimated to have OSA, the prevalence increases to
2040% in the obese population (Kyzer and Charuzi, 1998). In a longitudinal study
of OSA in obesity in an adult population over a 10-year period, even moderate gains
and reductions of weight were statistically signicant in their association with
signicant increase and decrease in respiratory disturbance index (RDI). Relative to
stable weight, a 10% weight gain predicted an approximate 32% increase in the RDI
and predicted a six fold increase in the odds of developing moderate-to-severe SDB.
A 10% weight loss predicted a 26% decrease in the RDI. (Peppard et al., 2000).
Longitudinal studies involving elderly patients ages 65 and above, demonstrated
changes of the respiratory disturbance index as a function of body mass index, rather
than age (Ancoli-Israel et al., 2001). In fact, the body mass index was found to be the
strongest predicator of sleep apnea in elderly patient populations (Ancoli-Israel et al.,
1991a; Shochat and Pillar, 2003).
5.4.2. Neck circumference

Increased neck circumference is also an important a risk factor for OSA. Patients
with a neck circumference greater than 48 cm have a twentyfold increased risk of
having OSA (Flemons et al., 1994).
5.4.3. Family history

An inherited basis for OSA has been suggested by reports of families with
multiple aected members and by a previous study of the familial aggregation of
symptoms of OSA. The rst degree relatives had higher prevalence of SDB
compared to non-related controls (Redline et al., 1995). In an overnight poly-
somnography study of adult ospring of sleep apnea patients, 47% were found to
have sleep apnea and another 22% were found to have simple snoring without
apneas (Pillar and Lavie, 1995). Considering the well-established prevalence of
sleep apnea in the general population, these results may suggest that OSA can be
an inherited syndrome.
5.4.4. Alcohol
Alcohol consumption can exacerbate SDB. It is well-established that alcohol
usage can increase upper airway resistance. Alcohols ability to worsen obstructive
sleep apnea may be accounted for by its ability to decrease pharyngeal airway size
and increase nasal resistance (Robinson et al., 1985). Alcohol may also cause
mucosal congestion, depress central respiratory drive, and enhance muscle
relaxation, all of which would exacerbate OSA (Quinnell and Smith, 2004).
5.4.5. Endocrinopathies
Sleep apnea is associated with endocrinopathies such as hypothyroidism
(Grunstein and Sullivan, 1988; Rosenow et al., 1998) and acromegaly (Weiss et al.,
2000). There is a high rate of clinically undetected thyroid dysfunction among the
healthy institutionalized geriatric population (Ayala et al., 2001). Hypothyroidism
is a known cause of secondary OSA causing oropharyngeal airway myopathy,
edema, and obesity all which predispose patients to upper airway collapse and
obstruction (Quinnell and Smith, 2004). Thyroid hormone replacement therapy
can improve and sometimes even resolve OSA in hypothyroid patients (Hattori et al.,
2003). An excess of growth hormone in adults causes acromegaly which is yet
another endocrinopathy associated with OSA. Excessive growth of the craniofacial
bones, enlargement of the tongue (macroglossia), and thickening and enlargement of
laryngeal region leads to further reduction of the upper airway anatomy. The upper
airway obstruction in acromegaly may result in severe hypoxemia and disruption
of normal sleep (Mezon et al., 1980). Treatment of OSA in acromegaly by
tracheostomy resulted in disappearance of the OSA-associated somnolence.
When compared to premenopausal women, and post-menopausal women who are
maintained on hormone replacement therapy, post-menopausal women are at risk
for developing OSA (Bixler et al., 2001).
5.5. Obstructive sleep patients with chronic obstructive pulmonary disease (COPD)
The frequency of OSA in COPD has been studied previously. Indeed COPD
is highly prevalent in elderly patients; 1015% of patients with COPD may have
obstructive sleep apnea. Those patients appear to have a lower baseline oxygen
saturation and higher partial pressure of carbon dioxide than patients with OSA
syndrome itself. These patients are also at a high risk for developing episodes of
prolonged hypoxemia during sleep, especially during REM sleep than patients
without COPD, and they are therefore more prone to suer from complications of
hypoxemia such as cor pulmanale, pulmonary hypertension, and polycythemia. The
recurring hypoxemic episodes that characterize OSA, particularly when associated
with chronic obstructive pulmonary disease and/or daytime hypoxemia, may lead
to both pulmonary and systemic hypertension, cardiac arrhythmias, and possibly
premature death.
96 A. Y. Avidan
5.6. Consequences of obstructive sleep apnea in the elderly
Several large epidemiologic studies have reported strong associations between

sleep apnea and cardiovascular disease, systemic and pulmonary hypertension,
myocardial infarction, cerebrovascular accidents, nocturnal cardiac arrhythmias,
and congestive heart failure. Sleep apnea is also associated with neurocognitive
dysfunction, and may be viewed as a reversible form of dementia.
5.6.1. Cardiovascular disease, hypertension, and sleep apnea

In a study of 1988 non-institutionalized elderly aged 67 8 years, Bliwise et al.
showed that an apnea/hypopnea index greater than 10 per hour was associated with
an increased risk of cardiovascular disease (Bliwise et al., 1988; Janssens et al., 2000).
In another study of 233 seniors Ancoli-Israel et al. found that patients with OSA
have a greater tendency to die during sleep and there was a strong association
between AHI and mortality in women (Ancoli-Israel et al., 1989). In a laboratory
based perspective study on sleep apnea and hypertension, OSA was found to be an
independent risk factor for the development of hypertension in adults, even after
adjusting for age, sex, and body mass index (Lavie et al., 2000). These ndings were
supported by a general population-based study in which the association between
OSA and hypertension was found to be strongest in the younger patients (Bixler
et al., 2000). A study which was performed as part of the Sleep-Heart Health Study,
a nationwide study on the cardiovascular consequences of OSA, similar results were
found in middle age and elderly adults (Nieto et al., 2000). Other ndings from these
studies have shown that even mild levels of signicant sleep apnea, reected by the
respiratory disturbance index, may have substantial impact on coronary artery
disease, congestive heart failure, and cerebral vascular accidents.
5.6.2. Obstructive sleep apnea in neurocognitive decline

Elderly patients with OSA are at risk for cognitive decline, impaired daily
functioning and excessive daytime sleepiness. Some have related this impairment to
the hypoxemia that accompanies the cessation of airow in sleep apnea (Findley
et al., 1986; Greenberg et al., 1987; Shochat and Pillar, 2003). Elderly patients who
have sleep apnea with hypoxemia are also at risk for having more severe cognitive
impairment than those with sleep apnea without hypoxemia. In addition, the
patients who had sleep apnea with hypoxemia had mean performance scores in the
impaired range on measures of attention, concentration, complex problem-solving,
and short-term recall of verbal and spatial information (Findley et al., 1986). While
some have concluded that neurocognitive decline may be due to sleep disruption
and subsequent excessive daytime somnolence, others have related the cognitive
impairment to hypoxemia (Shochat and Pillar, 2003). The apparent contradicting
reports may be due to dierent types of neuro psychrometric types of testing
performed, as well as to dierent denitions of sleep apnea severity and signicance
(Shochat and Pillar, 2003). Thus it may be suggested that cognitive impairment
associated with psychomotor and executive function may be related to the
hypoxemia while memory and attentional impairment may be related to decreased
vigilance due to the underlying somnolence (Bedard et al., 1991). Some have
suggested that cognitive impairment in patients with severe sleep apnea may be
associated with daytime sleepiness in patients with mild to or moderate sleep apnea.
Finally, mood instability and depression have also been associated with sleep apnea,
although not all studies have reported this relationship (Cohen-Zion et al., 2001).
The declining cognitive function associated with sleep apnea in the elderly is
associated primarily with increases in daytime sleepiness. One theoretical model
suggests that any relationship between sleep apnea and cognitive function may be
mediated by the eect of sleep apnea and resultant sleep fragmentation on daytime
sleepiness (Cohen-Zion et al., 2001).
5.6.3. Sleep disordered breathing and cognitive impairment

Chronic hypoxic episodes in patients with obstructive sleep apnea have been
associated with moderate to severe neuropsychologic impairment. Grant et al.
studied 203 patients with chronic obstructive pulmonary disease. The results of their
studies show that 77% of patients were found to have neuropsychologic decits and
42% of the patients decits were moderate to severe in magnitude, a rate triple that
of the control group of all the patients with no evidence of COPD. Higher cognitive
function was most severely aected, and this has also been noted by Krop et al.
(1973). Findley et al. (1986) studied younger patients with OSA and found that
cognitive impairment associated with OSA was correlated with a mean oxygen
saturation during the night and awake PO2 levels.
5.6.4. Obstructive sleep apnea and dementia

Dementia occurs in about 1% of patients 60 and older and doubles every 5 years
to reach between 30 and 50% by the age of 85. Sleep apnea is thought to alter mental
functioning. Accordingly, research has focused on relationships between sleep apnea
and dementia. Ancoli-Israel et al. (1991b) reported the relationship between sleep
apnea and dementia. The authors used the Mattis Dementia Rating Scale (DRS) and
the Geriatric Depression Scale and determined that sleep apnea was signicantly
correlated with all subscales on the dementia rating scale. In particular, items
reecting attention, initiation and perseveration, conceptualization, and memory
tasks on the DRS distinguished between those with and without severe sleep apnea
(Ancoli-Israel et al., 1991b). Among those patients with no depression, all patients
with severe sleep apnea were also severely demented. The authors hypothesized that,
although causality could not be inferred from associations, sleep apnea could cause
decits in brain function, possibly due to global eects rather than any particular
cortical or subcortical structure (Ancoli-Israel et al., 1991b).
5.6.5. Sleep apnea and alzheimers disease

Patients with Alzheimers disease account for approximately 70% of all cases of
dementia (Janssens et al., 2000). In regards to the association between sleep apnea
and dementia of the Alzheimer type, several investigators have noted a higher
prevalence of sleep disordered breathing in patients with Alzheimers disease when
compared to control groups (Hoch et al., 1986; Erkinjuntti et al., 1987), while others
98 A. Y. Avidan
have not (Bliwise et al., 1989). Furthermore, anecdotal reports of dementia-

like symptoms (memory impairment, concentration diculties, and confusion)
associated with sleep apnea suggesting that there may be a causal relationship
between sleep disordered breathing and Alzheimers disease (Schletens et al., 1991).
In Alzheimers dementia, sleep apnea could be a consequence of cell loss in the
brainstem respiratory center. Conversely neuronal degradation in Alzheimers
disease could be hastened by nightly insults of intermittent cerebral hypoxemia
related to the underlying sleep disordered breathing (Janssens et al., 2000).
5.7. Diagnosis of sleep disordered breathing in the elderly
5.7.1. Clinical diagnosis

The diagnosis of obstructive sleep apnea can be suspected based on the clinical
history of excessive daytime sleepiness (the most frequent presenting symptom),
snoring, and nocturnal apneic episodes (reported by the bed partner). Many patients
will also present with memory problems, and some will describe impotence and
nocturia. The clinical encounter serves as a way not only to establish the possibility
of SDB, but it also enables the clinician to ask questions the may lead to other
possible causes of daytime sleepiness such as insucient sleep time, depression and
other sleep disorders including disorders of motor control during sleep (nocturnal
myoclonus, parasomnias), circadian rhythm disorders (advanced sleep phase
syndrome, which is common in the elderly) and rarely narcolepsy. During the
clinical encounter, patients are asked about their bed time, rise time, sleep latency,
number of awakenings during the night, length of naps, and sleep pattern on the
weekends. Inquiries regarding medical conditions such as hypothyroidism,
acromegaly, and oropahryngeal carcinoma are important as they put patients at
risk for OSA. Patients need to bring a complete medication list showing all
prescription and over the counter medications and should be asked about caeine,
alcohol, and nicotine usage. While in the waiting room, they may complete the
Epworth Sleepiness Scale (ESS) which is an eight point questionnaire asking patients
about the propensity to fall asleep in various situations on a scale from 0 to 3, where
24 is the maximum score. Patients with a total score of 10 or higher should be
considered to have signicant daytime sleepiness, and those over 15 have severe
daytime sleepiness (Johns, 1991).
Obesity, thick neck, crowded oropharyngeal inlet, and the presence of
retrognathia or micrognathia are consistent physical examination ndings associated
with the presence of sleep apnea. At our Sleep Center we often rely on the
Mallapmati Classication to rate the severity of oropharyngeal occlusion (Friedman
et al., 1999). Intranasal examination looking for the possibility of nasoseptal
deviation and signs of allergic rhinitis is important in anticipation of positive
pressure therapy. If patients have signicant stenosis of the nasal passages due to an
anatomic occlusion, a referral to an ear, nose, and throat specialist may be advisable.
One should also screen for endocrinopathies such as acromegally and hypothyroid-
ism by looking for evidence of macroglossia and macrognathia. Patients at risk for
having sleep apnea should also be evaluated for the systemic consequences of OSA,
namely have their pulse rate and blood pressure checked and have a good
cardiopulmonary examination. A neurologic exam is warranted with particular
emphasis on the mini mental state exam (MMSE) in demented patients with possible
OSA. The MMSE score may help when following these patients once they have been
treated for OSA.
5.7.2. Laboratory diagnosis

Patients with the clinical suspicion of OSA are generally brought in for a full-
night polysomnography (PSG) which is considered the gold standard and study
of choice for diagnosing sleep apnea. It is a technique which involves the
simultaneous recording of several physiologic variables such as electrical potentials
from the brain (electroencephalography EEG), eye movements (EOG electro-
oculography) muscle activity (EMG electromyography), heart rhythm (EKG),
body position (supine, left, right), oxymetry, and respiratory activity (airow,
thoracic, and abdominal excursion). The downside of polysomnography is that it is
expensive, time consuming (in laboratory monitoring and interpretation), and is not
readily available in many health care facilities, where the waiting time can be quite
long. This problem on poor access has made the case for other measurements such as
in house monitoring and oxymetery readings, which is currently a cause of debate
amongst sleep specialists.
Many patients with ndings suggestive of OSA are brought in for split-night
PSG. This consists of diagnostic recording during the rst half of the night, with
the second half of the PSG utilized for positive pressure therapy (CPAP) titration.
Recurrence or persistence of daytime sleepiness or fatigability despite optimum
therapy of sleep apnea can be assessed objectively through the multiple sleep latency
test (MSLT) which provides an objective measurement of sleepiness and propensity
to fall asleep.
6. Management of sleep disordered breathing
Management of OSA in the elderly can be divided into conservative and non-
conservative approaches (Fig. 6). The question of the management of sleep apnea
in the elderly is currently under close investigation. At this point, there are limited
data showing the true impact of the treatment of the underlying sleep apnea as
a therapeutic intervention in the elderly on subjective complaints. It is possible that
the severity of the disorder rather than the advancing age should be the leading
factor when considering treatment of these disorders. It is in the authors experience
that before contemplating denitive therapy, clinicians need to thoroughly evaluate
the impact of sleep apnea on the patients lives and impact in the setting of other
comorbidities rather than treating the sleep study or the respiratory disturbance
index. For example, an elderly patient with an apneahypopnea index of 7 without
any clear symptoms of daytime sleepiness, or underlying cardiovascular disease, may
be followed clinically over time and not be subjected to therapy. On the other hand,
a patient with an AHI of 4 with history of witnessed apnea, daytime sleepiness in
the context of ischemic heart disease may be treated more aggressively.
100 A. Y. Avidan
Conservative approaches include modication of behavioral factors such as

weight loss, avoidance of alcohol, limiting the use of sedating compounds, and
smoking cessation. The most commonly accepted conservative mode of therapy
includes the use of positive pressure therapy in the form of nasal continuous positive
airway pressure (nasal CPAP). Some selected patients may be treated successfully
with oral appliances. Non-conservative treatment approaches include the use
of surgical management such as uvulopalatopharyngoplasty (UPPP), maxillo
mandibular advancement, genioglossus advancement, glossectomy, and tracheo-
tomy. Figure 6 outlines the various diagnostic schemes.
6.1. Conservative non-surgical medical management
As a general principle, patients with sleep apnea need to avoid situations that
may put them at risk for sleep apnea. These include avoidance of hypnotics and
psychotropic agents as these suppress respiratory drive and place patients at greater
risk of OSA due to increased frequency and length of apneas and hypopneas.
6.1.1. Weight loss

Obesity is associated with increased prarapharyngeal fat pad size in the oral
pharynx, a contributing risk factor for sleep apnea. Weight loss is therefore indicated
for the treatment of sleep apnea. Weight reduction in obese patients with OSA
improves the upper airway cross-sectional area (Rubinstein et al., 1988). This may
decrease the number of respiratory disturbances and sleep disruption and improve
nocturnal oxygen saturation. Although signicant weight loss is dicult to obtain,
particularly in the elderly, even moderate weight loss may markedly reduce the
apnea/hypopnea index. Weight loss in the elderly may be more challenging due
to the existence of other comorbidities that my limit exercise capacity. Weight
reduction should nevertheless be encouraged due to its added benet when used
in conjunction with other therapies for OSA (Quinnell and Smith, 2004). The use
of pharmacotherapy to achieve weight loss is discouraged in the elderly due
to insucient data and greater risk associated with using these drugs with
advanced age.
6.1.2. Body position

Assuming the supine position during sleep places the patient at increased risk of
OSA and sleep position adjustment may be a viable treatment for some non-obese
sleep apnea patients (Cartwright, 1984; Oksenberg et al., 1997). Patients may be
asked to use tennis balls which may be sewn into the back part of the pajama and
make it uncomfortable for the patient to lay in the supine position. Sometimes
special mattresses can be used to avoid the patient sleeping on his or her back.
Positional therapy may be particularly benecial for those patients with mild strictly
positional sleep apnea compared to those patients with more severe non-positional
sleep apnea. In the authors experience, the use of tennis balls to prevent supine sleep
may sometimes be associated with exacerbation of low back pain in predisposed
individuals and its use should therefore be limited in individuals predisposed to

lumbosacral disease and other rheumatologic condition.
6.1.3. Positive pressure therapy

Positive pressure therapy in the form of either continuous positive pressure
therapy (CPAP) or bi-level positive airway pressure is probably the most eective
option for treating OSA in the elderly. Positive pressure therapy should also be
considered as rst line treatment option. The CPAP is the most widely used
treatment for obstructive sleep apnea and has shown to improve the nocturnal
oxygen saturation and quality of sleep in younger adults. To date, there are no
studies of the use and impact of long-term treatment with nasal CPAP in elderly
subjects. Nasal CPAP consists of an air pressure generating device (used at a
predetermined level) that is applied over the nose (and sometimes the mouth) of the
patient utilizing a securely tting mask (Strollo et al., 1998). The positive air pressure
acts as a pneumatic splint in maintaining upper airway patency while the patient is
asleep (Quinnell and Smith, 2004). The device is used every night for as long as the
patient has obstructive sleep apnea. The positive airway pressure in a way acts like a
pneumatic splint to maintain airway patency. The patient is generally brought in for
a CPAP titration trial in the sleep laboratory. The trial utilizes multiple CPAP
pressures and the patient is prescribed an ideal pressure that prevents apneic
episodes, snoring, and maintains adequate oxygen saturation while patient is in the
supine position and during REM sleep. Variations of nasal CPAP are in the form of
bi-level positive airway pressure (Bi-PAP) and automated CPAP (auto CPAP).
Bi-Pap allows two separate pressures; an inspiratory and an expiratory pressure,
to maintain upper airway patency. This modality may improve patient comfort,
particularly on expiration and may be required in certain conditions such as chronic
obstructive pulmonary disease, in patients with neuromuscular weakness such as
myasthenia gravis and muscular dystrophy, and in patients with amyotrophic lateral
sclerosis (Shochat and Pillar, 2003). Auto CPAP is a recently introduced treatment
option which changes the level of pressure requirement automatically throughout the
night based on patients airow resistance. This may be indicated in patients who
may have diculties tolerating other modes of therapy.
The CPAP and Bi-Pap options are non-invasive and probably the rst choice for
therapy, but compliance rates are from 46 to 80% (Kribbs et al., 1993; Pepin et al.
1999). Some of the compliance problems may stem from side eects attributable to
this treatment modality and include claustrophobia, nasal congestion, and dryness.
In the authors clinical experience some of the adherence problems are unique to the
elderly and may include problems with dexterity (putting the mask on and o when
getting up to use the bathroom) and age-related neurocognitive diculties limiting
comprehension when learning how to rst use the mask leading to non-compliance.
Some of these can be overcome when the patients spouse becomes involved and
participates in the treatment plan. When monitoring compliance, many of the new
CPAP machines on the market have an internal compliance meter which may allow
for objective assessment and verication of use.
102 A. Y. Avidan
Continuous positive pressure therapy should be viewed as symptomatic rather

than a curative therapy, patients have to use it on a nightly basis as long as they
have sleep apnea, which for many patients means indenitely (Shochat and Pillar,
2003). Treatment adherence can be a problem which requires the practitioner to
maintain a close therapeutic alliance with the patient. The patients spouse and
family are often very helpful in assisting the patient and they should be encouraged
to help educate the patient about improving compliance and looking for any
challenges in therapy. In the elderly, close therapeutic relationship and education
during the initial period of therapy can improve adherence. A modest cognitive-
behavioral intervention can also be eective in increasing CPAP compliance and
improving vigilance in older adults (Aloia et al., 2001). The CPAP may also be
applied for elderly patients with dementia and sleep apnea. Pilot studies showed that
mild to early stage dementia patients with obstructive sleep apnea were able to
comply with a CPAP machine nightly for a two week period and showed some
signs of improved cognitive performance (Shochat et al., 2000).
It is common to encounter patients who despite using positive pressure therapy
may still complain of excessive daytime sleepiness. Figure 7 outlines some of the
reasons for residual excessive daytime sleepiness (EDS) in treated OSA patients.
It also identies some of the solutions that can be taken to improve adherence. It is
always a good idea to review the patients history, medication usage, sleep diary, and
look for other underlying medical or sleep disorders that may be causing EDS. In the
elderly, it is especially important to review medication list. The elderly as a group
are often seen by multiple physicians and may be prescribed multiple medications,
many which cause daytime sleepiness.
Wake promoting agents may be used in selected patients where the clinician is
certain that there are no underlying medical, sleep, or psychiatric issues causing
EDS. In these patients, modanil, which is a relatively new wake promoting agent
may be used to reduce subjective sleepiness. Modanil is believed to act in specic
regions of the anterior hypothalamus but has unknown mechanism of action. Its use
in the elderly needs to be monitored with scrutiny. First, there are very limited data
regarding its safety and ecacy in elderly patients as much of the available data has
looked at patients younger than 65 years of age (Quinnell and Smith, 2004). Second,
its pharmacodynamic and pharmacokinetic properties in the elderly who are more
prone to renal and hepatic impairment are unknown and dosage recommendations
have not been established. In the authors view, modanil needs to be used with great
care in the elderly and until safety and ecacy trials are available in this cohort, its
use should be limited.
6.1.4. Oral appliances

Oral appliances (Fig. 6) work by repositioning the mandible and improving the
upper airway space at the hypopharyngeal level. There are several types of oral
appliances available for use. They may take the form of mandibular advancement
devices (MAD) which work by advancing the mandible anteriorly. They may also
take the form of a tongue retraining device (TRD) which repositions the tongue
and thus prevents its collapse. A TRD is a bubble shaped device made of soft
EDS in Treated OSA
Are there other

Is the patient using CPAP? Is the patient sleeping co-existing medical diagnosis?

enough? i.e. Chronic pain?
Check adherence Sleep Diary Are there other co-existing

Actigraphy psychiatric diagnosis?
Worsening of OSA? i.e. Depression?
CPAP retitration
Residual Is the patient taking hypnotics,
Snoring? Poor mask fit? antidepressants, alcohol?
Assess mask fit
Dry mouth,
Check mask fit
mouth leak
Add a chinstrap
Are there other co-existing sleep disorders?
Nasal Prescribe nasal Steroids

Congestion? Nasal Saline Undiagnosed Narcolepsy?
Humidification
PLMS, RLS?
Smaller Mask
Claustrophobia? Insomnia?
Desensitization
Check for air leaks-Pressure Parasomnias (i.e. RBD) causing sleep disruption?
Excessive noise
may be too high. If not,
Limiting use?
use earplugs
103
Fig. 7. Approach to the sleep apnea patient with residual daytime sleepiness despite appropriate therapy.
104 A. Y. Avidan
polyvinyl. The patient positions his/her teeth in the grooves, sticks his tongue
forward into the bubble until suction grabs and holds the tongue in place. The TRD
is especially useful in edentulous patients, patients with macroglossia, and poor
dental health.
Oral appliances are considered by many as very useful conservative non-surgical
approach for the treatment of mild to moderate obstructive sleep apnea and for
snoring. They are particularly advantageous as they require minimal upkeep and
are relatively simple to use which becomes an attractive option in elderly patients
with underlying neurocognitive impairment. The oral appliances are fabricated
and tted by dentists. It is recommended that a follow-up sleep study takes place
with the appliances in place to look for evidence of improvement. The oral
appliances are attractive for many patients because they are also much more mobile
compared to CPAP, and they are small and not expensive (Quinnell and Smith,
2004). They are a nice alternative to patients who are very claustrophobic with the
CPAP masks.
Adverse side eects attributed to the appliances may include temporal mandibular
joint (TMJ) dysfunction (which is also a contraindication for use), tooth and gum
pain, proprioceptive malocclusion, excessive/reduced salivation and at times, gum
disease. Elderly patients also have dental problems including the fact that many are
edentulous and the oral appliance (MAD) has to t over intact healthy teeth. In
some selected edentulous patients, TRD may be an option. However, in the authors
experience TRD devices are rarely tolerated beyond 34 h of use and compliance
is an obvious drawback. The principal challenge with prescribing either the MAD
or TRD is that there have been no studies looking at the ecacy of these therapies
in the elderly patient population.
6.1.5. Medications
Generally speaking, drug therapy for obstructive sleep apnea has been
unsuccessful and its use, especially in elderly patients may be discouraged due to
the higher likelihood of side eects in this patient group ranging from tolerance to
serious adverse side eects (Shochat and Pillar, 2003). Mechanisms of action of the
principal pharmacologic agents include reduction of OSA via respiratory drive
stimulation, REM sleep suppression and stimulation of upper airway muscle tone
(Quinnell and Smith, 2004).
6.1.5.1. Tricyclic antidepressants (TCAs)

The REM suppressants such as protriptyline act by reducing REM sleep (Hanzel
et al., 1991). These REM suppressants may improve OSA during REM sleep since
respiratory events during this sleep stage are more severe both in duration and
oxygen saturation These agents decrease REM stage time and REM apnea duration
and produce improvement in oxygenations which may last beyond six months of
treatment (Brownell et al., 1982). The diculties with using TCAs, especially in the
elderly, are that they may produce cardiovascular complications owing to their
strong anticholinergic side eects. Tricyclic antidepressants may therefore be less

attractive in this patient population.
6.1.5.2. Selective serotonin reuptake inhibitors (SSRIs)

Selective serotonin reuptake inhibitors increase genioglossal muscle activity in
sleep apnea patients and may therefore prevent the apneas from occurring (Shochat
and Pillar, 2003). Fluoxetine has been well tolerated in the treatment of OSA (Hanzel
et al., 1991). Serotonin may be a central respiratory stimulant and the serotonin-
uptake inhibitor, uoxetine, was better tolerated than protriptyline in the treatment
of OSA. When comparing protriptyline to uoxetine, in patients with OSA, both
drugs signicantly decreased the proportion of REM sleep time and decreased the
number of apneas or hypopneas in NREM sleep. However, there was no signicant
improvement in the number of arterial oxygen desaturation events, the level of
arterial oxygen desaturation, or the number of arousals with either agent. Half of
the patients had good responses, including improvement in oxygenation, to either
uoxetine or protriptyline (Hanzel et al., 1991).
6.1.5.3. Respiratory stimulants

Progesterone is believed to stimulate the central respiratory center and has been
suggested for the treatment of sleep apnea in the form of Medroxyprogesterone
acetate (MPA). It probably acts by altering pharyngeal muscle function relative to
diaphragm and external intercostal function. However, a number of trials using
MPA for the therapy of OSA has been disappointing (Cook et al., 1989). Even
though MPA increased the minute ventilation and occlusion pressure responses
to hypercapnia measured in the awake state; it does not improve the breathing
disorders during sleep in the non-hypercapnic patients with obstructive sleep apnea.
(Rajagopal et al., 1986).
Acetazolamide, a carbonic anhydrase inhibitor, induces metabolic acidosis and
causes increased respiratory drive. It has been investigated in the treatment of the
central events of CheyneStokes respiration and the obstructive sleep events in OSA
(Quinnell and Smith, 2004). However, it does not improve daytime somnolence in
these patients and it has little clinical merits in the treatment of OSA.
6.2. Surgical management
Surgical treatment options for OSA target potential sites of airway obstruction.
Surgical techniques involve extirpation of soft tissue, secondary soft tissue
repositioning through primary skeletal mobilization, or bypass of the pharyngeal
airway. While surgery, especially in the form of tracheostomy, was commonly used
in the past for the management of OSA, it is increasingly underutilized owing to
the favorable responses with positive pressure therapy. There are three sites of
obstruction which include the nasopharynx, oropharynx, and the hypopharynx
(Fig. 6). Surgical procedures for the management of obstructive sleep apnea
have been chosen when there is a clear anatomical airway obstruction or when
106 A. Y. Avidan
medical management with nasal CPAP is not achieved. Patients need to be evaluated
by otolaryngologists and possibly, an oral-maxillofacial surgeon and undergo
extensive presurgical evaluation with cephalometric X-rays, and ber optic
endoscopy. Surgery may be targeted at the level of the upper airways preventing
their collapse.
6.2.1. Palatopharyngeal reconstruction

Uvulopalatopharyngoplasty (UPPP) is a technique which utilizes surgery, laser or
radiofrequency ablation. So, UPPP and its modication (uvulopalatal ap) is the
most commonly used technique used to enlarge the retropalatal airway (Sher, 2002).
Some of the trials looking at UPPP utilized an apneahypopnea index (AHI) of less
than 10, or a reduction of the AHI by 50% as a measure of clinical success (Janson
et al., 1997). It remains clinically debatable whether these criteria are valid as
patients may remain clinically symptomatic even after these improvements are
achieved (Quinnell and Smith, 2004). The long term success of UPPP using these
criteria was less than 50%. In about a third of the patients, the underlying SDB
returned to its baseline morbid condition after an initial level of success (Larsson
et al., 1994). Postoperative complications of UPPP, which may be more common
with aging, include postoperative bleeding, infection, velopharyngeal insuciency,
nasopharyngeal stenosis, voice change, vague foreign body sensation, and rarely,
death within 48 h of surgery secondary to upper airway obstruction (Sher, 2002).
6.2.2. Reduction glossectomy

Reduction glossectomy is a procedure performed to reduce tongue volume but is
commonly associated with pain and greater postoperative complications.
6.2.3. Maxillomandibular advancement (MMA)

In younger patients, the main surgical technique for maximally enlarging the
retrolingual airway and partially enlarging the retropalatal airway is maxillo
mandibular advancement (MMA), which has a reported success rate of 90% (Riley
et al., 2000; Sher, 2002). In this procedure, both the maxilla and mandible are
advanced utilizing bilateral osteotomies enlarging the posterior airway space. For
younger patients, it may be eective for moderate to severe OSA. This is an extensive
complex surgery with multiple risks and should probably be avoided as rst line
therapy in the elderly.
6.2.4. Genioglossus advancement and hyoid myotomy (GAHM)

This surgical option utilizes anterior advancement of the tongue producing
improvement in hypopharyngeal airway space. This is a more complex surgery than
UPPP but generally produces a more favorable outcome.
6.2.5. Tracheostomy
Tracheostomy, which is designed to bypass the collapsible upper airway is used
in very refractory cases, and/or in severe, life threatening cases of OSA (He et al.,
1988). Tracheostomy may be used to bypass the oropharyngeal airway in sleep
apnea patients with underlying morbid obesity, considerable facial skeletal

deformities with hypersomnolance, severe hypoxemia (SaO2 < 70%), or signicant
cardiac arrhythmias (Riley, 2000).The tracheotomy tube is plugged when the
patient is awake to allow for normal speech and swallowing. It used to be the gold
standard of therapy for OSA until CPAP became available. It is invariably a curable
mode of therapy with a major disadvantage in that it alters the patients lifestyle
forever.
6.2.6. Alternatives to CPAP clinic

At our institution, we have implemented the Alternatives to CPAP Clinic which
utilized the expertise of the sleep neurologist, an otolaryngologist, a maxillofacial
surgeon, and a dentist. We are very careful when oering surgical therapies to elderly
patients given the risk of respiratory depression eects of anesthesia, and higher
cardiovascular morbidity in this cohort. Surgical management may be an attractive
treatment option in selected younger patients given the life long need for treatment;
many younger single patients with OSA may be embarrassed to use CPAP due to
psychosocial issues. The elderly need to be evaluated with great scrutiny before
surgery is carried out, and during the pre- and postoperative procedure.
7. Conclusion
Obstructive sleep apnea is very common in elderly patients. It impacts this patient
population in that it may lower quality of life, and have adverse eects on almost
every organ system ranging from neurocognitive impairment to cardiovascular heart
disease. Elderly patients do not present with the typical symptoms of their younger
cohorts. Therefore lack of snoring and witnessed apneas per se is not a criterion for
dismissing the possibility of OSA in this age group. It is often dicult to establish the
diagnosis of OSA in the elderly as many of the studies conducted so far show no
uniform distribution of the apneahypopnea index. Treatment of this condition is
extremely important as it may improve neurocognitive disturbances, improve the
patients quality of life and optimize therapies for other chronic conditions such as
hypertension. The treatment of OSA can be extremely rewarding. Most patients
respond very well with positive pressure therapy. For others, dierent options may
be available.
Acknowledgments
The author wishes to thank Mrs. Karen Gowen for her assistance in the
preparation of the manuscript.
References
Aloia, M.S., Di Dio, L., Ilniczky, N., Perlis, M.L., Greenblatt, D.W., Giles, D.E., 2001. Improving
compliance with nasal CPAP and vigilance in older adults with OAHS. Sleep Breath 5(1), 1321.
108 A. Y. Avidan
Ancoli-Israel, S., Gehrman, P., Kripke, D.F., Stepnowsky, C., Mason, W., Cohen-Zion, M., Marler, M.,
2001. Long-term follow-up of sleep disordered breathing in older adults. Sleep Med. 2(6), 511516.
Ancoli-Israel, S., Kripke, D.F., Klauber, M.R., Mason, W.J., Fell, R., Kaplan, O., 1991a. Sleep
disordered breathing in community-dwelling elderly. Sleep 14(6), 486495.
Ancoli-Israel, S., Klauber, M.R., Butters, N., Parker, L., Kripke, D.F., 1991b. Dementia in
institutionalized elderly: relation to sleep apnea. J. Am. Geriatr. Soc. 39(3), 258263.
Ancoli-Israel, S., Klauber, M.R., Kripke, D.F., Parker, L., Cobarrubias, M., 1989. Sleep apnea in female
patients in a nursing home. Increased risk of mortality. Chest 96(5), 10541058.
Ancoli-Israel, S., Kripke, D.F., Mason, W., Kaplan, O.J., 1985. Sleep apnea and periodic movements in
an aging sample. J. Gerontol. 40(4), 419425.
Avidan, A.Y., 2002. Sleep changes and disorders in the elderly patient. Curr. Neurol. Neurosci. Rep. 2(2),
178185.
Ayala, C., Cozar, M.V., Rodriguez, J.R., Silva, H., Pereira, J.L., Garcia-Luna, P.P., 2001. [Subclinical
thyroid disease in institutionalised healthy geriatric population]. Med. Clin. (Barc) 117(14), 534535.
Bedard, M.A., Montplaisir, J., Richer, F., Rouleau, I., Malo, J., 1991. Obstructive sleep apnea syndrome:
pathogenesis of neuropsychological decits. J. Clin. Exp. Neuropsychol. 13(6), 950964.
Bixler, E.O., Vgontzas, A.N., Lin, H.M., Ten Have, T., Rein, J., Vela-Bueno, A., Kales, A., 2001.
Prevalence of sleep-disordered breathing in women: eects of gender. Am. J. Respir. Crit. Care. Med.
163(3 Pt 1), 608613.
Bliwise, D., 2000. Normal aging. In: M.R.T. Kryger, W. Dement (Eds.), Principles and Practice of Sleep
Medicine, 3rd ed. Saunders, Philadelphia, pp. 2642.
Bliwise, D., Bliwise, N., Partinen, M., Pursley, A., Dement, W., 1988. Sleep apnea and mortality in an
aged cohort. Am. J. Public Health 78, 544547.
Bliwise, D.L., 1993a. Sleep apnea and cognitive function: where do we stand now? Sleep 16(Suppl 8),
S72S73.
Bliwise, D.L., 1993b. Sleep in normal aging and dementia. Sleep 16(1), 140181.
Bliwise, D.L., Yesavage, J.A., Tinklenberg, J.R., Dement, W.C., 1989. Sleep apnea in Alzheimers disease.
Neurobiol. Aging 10(4), 343346.
Brischetto, M.J., Millman, R.P., Peterson, D.D., Silage, D.A., Pack, A.I., 1984. Eect of aging on
ventilatory response to exercise and CO2. J. Appl. Physiol. 56(5), 11431150.
Brownell, L.G., West, P., Sweatman, P., Acres, J.C., Kryger, M.H., 1982. Protriptyline in obstructive sleep
apnea: a double-blind trial. N. Engl. J. Med. 307(17), 10371042.
Burger, C., Shephard, J., 1993. Sleep-disordered breathing and aging. In: D. Mahler (Ed.), Pulmonary
Disease in the Patient. Marcel Dekker, New York Inc., pp. 6180.
Carskadon, M.A., Dement, W.C., 1981. Respiration during sleep in the aged human. J. Gerontol. 36(4),
420423.
Cartwright, R.D., 1984. Eect of sleep position on sleep apnea severity. Sleep 7(2), 110114.
Census USBot., 2000. Population Project Program, Washington, DC.
Chan, E.D., Welsh, C.H., 1998. Geriatric respiratory medicine. Chest 114(6), 17041733.
Cohen-Zion, M., Stepnowsky, C., Marler, Shochat, T., Kripke, D.F., Ancoli-Israel, S., 2001. Changes in
cognitive function associated with sleep disordered breathing in older people. J. Am. Geriatr. Soc.
49(12), 16221627.
Coleman, R.M., Miles, L.E., Guilleminault, C.C., Zarcone, V.P., Jr., van den Hoed, J., Dement, W.C.,
1981. Sleep-wake disorders in the elderly: polysomnographic analysis. J. Am. Geriatr. Soc. 29(7),
289296.
Cook, W.R., Benich, J.J., Wooten, S.A., 1989. Indices of severity of obstructive sleep apnea syndrome do
not change during medroxyprogesterone acetate therapy. Chest 96(2), 262266.
Dancey, D.R., Hanly, P.J., Soong, C., Lee, B., Hostein, V., 2001. Impact of menopause on the prevalence
and severity of sleep apnea. Chest 120(1), 151155.
Enright, P.L., Newman, A.B., Wahl, P.W., Manolio, T.A., Haponik, E.F., Boyle, P.J., 1996. Prevalence
and correlates of snoring and observed apneas in 5,201 older adults. Sleep 19(7), 531538.
Erkinjuntti, T., Partinen, M., Sulkava, R., Telakivi, T., Salmi, T., Tilvis, R., 1987. Sleep apnea in
multiinfarct dementia and Alzheimers disease. Sleep 10(5), 419425.
Findley, L.J., Barth, J.T., Powers, D.C., Wilhoit, S.C., Boyd, D.G., Suratt, P.M., 1986. Cognitive
impairment in patients with obstructive sleep apnea and associated hypoxemia. Chest 90(5),
686690.
Flemons, W.W., Whitelaw, W.A., Brant, R., Remmers, J.E., 1994. Likelihood ratios for a sleep apnea
clinical prediction rule. Am. J. Respir. Crit. Care Med. 150(5 Pt 1), 12791285.
Fleury, B., 1992. Sleep apnea syndrome in the elderly. Sleep 15, S39S41.
Friedman, M., Tanyeri, H., La Rosa, M., Landsberg, R., Vaidyanathan, K., Pieri, S., Caldarelli, D., 1999.
Clinical predictors of obstructive sleep apnea. Laryngoscope 109(12), 19011907.
Grant, I., Heaton, R.K., McSweeny, A.J., Adams, K.M., Timms, R.M. 1982. Neuropsychologic ndings
in hypoxemic chronic obstructive pulmonary disease. Arch. Intern. Med. 142, 14701476.
Grassino, A.E., Begin, P., 1990. [The role of respiratory muscles in the interruption of the obstructive sleep
apnea]. Rev. Mal. Respir. 7(5), 403410.
Greenberg, G.D., Watson, R.K., Deptula, D., 1987. Neuropsychological dysfunction in sleep apnea. Sleep
10(3), 254262.
Grunstein, R.R., Sullivan, C.E., 1988. Sleep apnea and hypothyroidism: mechanisms and management.
Am. J. Med. 85(6), 775779.
Guilleminault, C., Tilkian, A., Dement, W.C., 1976. The sleep apnea syndromes. Annu. Rev. Med. 27,
465484.
Hanzel, D.A., Proia, N.G., Hudgel, D.W., 1991. Response of obstructive sleep apnea to uoxetine and
protriptyline. Chest 100(2), 416421.
Hattori, H., Hattori, C., Yonekura, A., Nishimura, T., 2003. Two cases of sleep apnea syndrome caused
by primary hypothyroidism. Acta Otolaryngol. (Suppl 550), 5964.
He, J., Kryger, M.H., Zorick, F.J., Conway, W., Roth, T., 1988. Mortality and apnea index in obstructive
sleep apnea. Experience in 385 male patients. Chest 94(1), 914.
Hoch, C.C., Reynolds, C.F., 3rd, Monk, T.H., Buysse, D.J., Yeager, A.L., Houck, P.R., Kupfer, D.J.,
1990. Comparison of sleep-disordered breathing among healthy elderly in the seventh, eighth, and
ninth decades of life. Sleep 13(6), 502511.
Hoch, C.C., Reynolds, C.F., 3rd, Kupfer, D.J., Houck, P.R., Berman, S.R., Stack, J.A., 1986. Sleep-
disordered breathing in normal and pathologic aging. J. Clin. Psychiatry 47(10), 499503.
Janson, C., Gislason, T., Bengtsson, H., Eriksson, G., Lindberg, E., Lindholm, C.E., Hultcrantz, E.,
Hetta, J., Boman, G., 1997. Long-term follow-up of patients with obstructive sleep apnea treated with
uvulopalatopharyngoplasty. Arch. Otolaryngol. Head Neck Surg. 123(3), 257262.
Janssens, J.P., Pautex, S., Hilleret, H., Michel, J.P., 2000. Sleep-disordered breathing in the elderly. Aging
(Milano) 12(6), 417429.
Johns, M.W., 1991. A new method for measuring daytime sleepiness, the Epworth sleepiness scale. Sleep
14(6), 540545.
Kribbs, N.B., Pack, A.I., Kline, L.R., Smith, P.L., Schwartz, A.R., Schubert, N.M., Redline, S.,
Henry, J.N., Getsy, J.E., Dinges, D.F., 1993. Objective measurement of patterns of nasal CPAP use by
patients with obstructive sleep apnea. Am. Rev. Respir. Dis. 147(4), 887895.
Krieger, J., Sforza, E., Boudewijns, A.N., Zamagni, M., Petiau, C., 1997. Respiratory eort during
obstructive sleep apnea: role of age and sleep state. Chest 112(4), 875884.
Kronenberg, R.S., Drage, C.W., 1973. Attenuation of the ventilatory and heart rate responses to hypoxia
and hypercapnia with aging in normal men. J. Clin. Invest. 52(8), 18121819.
Krop, H.D., Block, A.J., Cohen, E., 1973. Neuropsychologic eects of continuous oxygen therapy in
chronic obstructive pulmonary disease. Chest 64, 317322.
Kyzer, S., Charuzi, I., 1998. Obstructive sleep apnea in the obese. World J. Surg. 22(9), 9981001.
Larsson, L.H., Carlsson-Nordlander, B., Svanborg, E., 1994. Four-year follow-up after uvulopalatophar-
yngoplasty in 50 unselected patients with obstructive sleep apnea syndrome. Laryngoscope 104(11 Pt 1),
13621368.
Lavie, P., Herer, P., Hostein, V., 2000. Obstructive sleep apnoea syndrome as a risk factor for
hypertension: population study. BMJ 320(7233), 479482.
Malhotra, A., Crowley, S., Pillar, G., et al., 2000. Age-related changes in pharyngeal structure and
function in normal subjects (abstract). Sleep 151(Suppl 2), A42.
110 A. Y. Avidan
Manning, H., 1993. Dyspnea in the elderly. In: Mahler, D.A. (Ed.), Marcel Dekker, New York Inc., pp.
81111.
Mezon, B.J., West, P., Ma Clean, J.P., Kryger, M.H., 1980. Sleep apnea in acromegaly. Am. J. Med. 69(4),
615618.
Mezzanotte, W.S., Tangel, D.J., White, D.P., 1992. Waking genioglossal electromyogram in sleep apnea
patients versus normal controls (a neuromuscular compensatory mechanism). J. Clin. Invest. 89(5),
15711579.
Morgan, B.J., Dempsey, J.A., Pegelow, D.F., Jacques, A., Finn, L., Palta, M., Skatrud, J.B., Young, T.B.,
1998. Blood pressure perturbations caused by subclinical sleep-disordered breathing. Sleep 21(7),
737746.
Mosko, S.S., Dickel, M.J., Paul, T., La Tour, T., Dhillon, S., Ghanim, A., Sassin, J.F., 1988. Sleep apnea
and sleep-related periodic leg movements in community resident seniors. J. Am. Geriatr. Soc. 36(6),
502508.
Narkiewicz, K., Somers, V.K., 1997. The sympathetic nervous system and obstructive sleep apnea,
implications for hypertension. J. Hypertens. 15(12 Pt 2), 16131619.
Naughton, M.T., Bradley, T.D., 1998. Sleep apnea in congestive heart failure. Clin. Chest Med. 19(1),
99113.
Nieto, F.J., Young, T.B., Lind, B.K., Shahar, E., Samet, J.M., Redline, S., DAgostino, R.B.,
Newman, A.B., Lebowitz, M.D., Pickering, T.G., 2000. Association of sleep-disordered breathing,
sleep apnea, and hypertension in a large community-based study. Sleep Heart Health Study. Jama
283(14), 18291836.
Oksenberg, A., Silverberg, D.S., Arons, E., Radwan, H., 1997. Positional vs nonpositional obstructive
sleep apnea patients: anthropomorphic, nocturnal polysomnographic, and multiple sleep latency test
data. Chest 112(3), 629639.
Pepin, J.L., Krieger, J., Rodenstein, D., Cornette, A., Sforza, E., Delguste, P., Deschaux, C., Grillier, V.,
Levy, P., 1999. Eective compliance during the rst 3 months of continuous positive airway pressure.
A European prospective study of 121 patients. Am. J. Respir. Crit. Care Med. 160(4), 11241129.
Peppard, P.E., Young, T., Palta, M., Dempsey, J., Skatrud, J., 2000. Longitudinal study of moderate
weight change and sleep-disordered breathing. Jama 284(23), 30153021.
Peterson, D.D., Pack, A.I., Silage, D.A., Fishman, A.P., 1981. Eects of aging on ventilatory and
occlusion pressure responses to hypoxia and hypercapnia. Am. Rev. Respir. Dis. 124(4), 387391.
Phillips, B.A., Berry, D.T., Schmitt, F.A., Magan, L.K., Gerhardstein, D.C., Cook, Y.R., 1992. Sleep-
disordered breathing in the healthy elderly. Clinically signicant? Chest 101(2), 345349.
Pillar, G., Lavie, P., 1995. Assessment of the role of inheritance in sleep apnea syndrome. Am. J. Respir.
Crit. Care Med. 151(3 Pt 1), 688691.
Quinnell, T.G., Smith, I.E., 2004. Obstructive sleep apnoea in the elderly: recognition and management
considerations. Drugs Aging 21(5), 307322.
Rajagopal, K.R., Abbrecht, P.H., Jabbari, B., 1986. Eects of medroxyprogesterone acetate in obstructive
sleep apnea. Chest 90(6), 815821.
Redline, S., Tishler, P.V., Tosteson, T.D., Williamson, J., Kump, K., Browner, I., Ferrette, V., Krejci, P.,
1995. The familial aggregation of obstructive sleep apnea. Am. J. Respir. Crit. Care Med. 151(3 Pt 1),
682687.
Resta, O., Caratozzolo, G., Pannacciulli, N., Stefano, A., Giliberti, T., Carpagnano, G.E., De Pergola, G.,
2003. Gender, age and menopause eects on the prevalence and the characteristics of obstructive sleep
apnea in obesity. Eur. J. Clin. Invest. 33(12), 10841089.
Riley, R., Powell, N., Li, K., Guilleminault, C., 2000. Surgical therapy for obstructive sleep apnea-
hypopnea syndrome. In: M.R.T. Kryger, W. Dement (Eds.), WB Saunders Company,
Philadelphiapp, pp. 913928.
Robinson, R.W., White, D.P., Zwillich, C.W., 1985. Moderate alcohol ingestion increases upper airway
resistance in normal subjects. Am. Rev. Respir. Dis. 132(6), 12381241.
Roehrs, T., Zorick, F., Sicklesteel, J., Wittig, R., Roth, T., 1983. Age-related sleep-wake disorders at a
sleep disorder center. J. Am. Geriatr. Soc. 31(6), 364370.
Rosenow, F., Mc Carthy, V., Caruso, A.C., 1998. Sleep apnoea in endocrine diseases. J. Sleep. Res. 7(1),
311.
Rubinstein, I., Colapinto, N., Rotstein, L.E., Brown, I.G., Hostein, V., 1988. Improvement in upper
airway function after weight loss in patients with obstructive sleep apnea. Am. Rev. Respir. Dis.
138(5), 11921195.
Scheltens, P., Visscher, F., Van Keimpema, A.R., Lindeboom, J., Taphoorn, M.J., Wolters, E.C., 1991.
Sleep apnea syndrome presenting with cognitive impairment. Neurology 41(1), 155156.
Shahar, E., Redline, S., Young, T., Boland, L.L., Baldwin, C.M., Nieto, F.J, OConnor, G.T.,
Rapoport, D.M., Robbins, J.A., 2003. Hormone replacement therapy and sleep-disordered breathing.
Am. J. Respir. Crit. Care Med. 167(9), 11861192.
Sher, A., 2002. Upper airway surgery for obstructive sleep apnea. In: A. Pack (Ed.), Marcel Dekker,
Shochat, T., Cohen-Zion, M., Ancoli-Israel, S., 2000. The eects of CPAP treatment on cognitive function
in dementia: a pilot study (abstract). Sleep 23(Suppl 2), A218.
Shochat, T., Pillar, G., 2003. Sleep apnoea in the older adult: pathophysiology, epidemiology,
consequences and management. Drugs Aging 20(8), 551560.
Somers, V.K., Dyken, M.E., Clary, M.P., Abboud, F.M., 1995. Sympathetic neural mechanisms in
obstructive sleep apnea. J. Clin. Invest. 96(4), 18971904.
Strollo, Jr P.J., Sanders, M.H., Atwood, C.W., 1998. Positive pressure therapy. Clin. Chest Med. 19(1),
5568.
Tack, M., Altose, M.D., Cherniack, N.S., 1981. Eect of aging on respiratory sensations produced by
elastic loads. J. Appl. Physiol. 50(4), 844850.
Webb, W., 1974. Periodic breathing during sleep. J. Appl. Physiol. 37, 899903.
Weiss, V., Sonka, K., Pretl, M., Dostalova, S., Klozar, J., Rambousek, P., Marek, J., Haas, T., 2000.
Prevalence of the sleep apnea syndrome in acromegaly population. J. Endocrinol. Invest. 23(8),
515519.
Yesavage, J., Bliwise, D., Guilleminault, C., Carskadon, M., Dement, W., 1985. Preliminary
communication: intellectual decit and sleep-related respiratory disturbance in the elderly. Sleep
8(1), 3033.
Young, T., Finn, L., Austin, D., Peterson, A., 2003. Menopausal status and sleep-disordered breathing in
the Wisconsin Sleep Cohort Study. Am. J. Respir. Crit. Care. Med. 167(9), 11811185.
Young, T., Palta, M., Dempsey, J., Skatrud, J., Weber, S., Badr, S., 1993. The occurrence of sleep-
disordered breathing among middle-aged adults. N. Engl. J. Med. 328(17), 12301235.
Young, T., Peppard, P.E., Gottlieb, D.J., 2002. Epidemiology of obstructive sleep apnea: a population
health perspective. Am. J. Respir. Crit. Care. Med. 165(9), 12171239.
Advances in
Cell Aging and
Gerontology
Sleep associated endocrine and immune

changes in the elderly
Boris Perrasa,b and Jan Borna,*
a
Institute of Neuroendocrinology, University of Lubeck, Germany
b
Department of Internal Medicine I, University of Lubeck, Germany
*Correspondence address: J. Born, University of Lubeck, Institute of Neuroendocrinology,
Haus 23a Ratzeburger Allee 160, 23538 Lubeck, Germany. Tel: 49-4515003641;
fax: 49-4515003640.
E-mail address: born@kfg.mu-luebeck.de
Contents
1. Introduction
2. Regulation of sleep-related endocrine systems in young humans
2.1. The HPA and somatotropic system
2.2. Vasopressin (VP)
2.3. Prolactin, melatonin, and other hormones
3. Regulation of sleep-related immune function in young humans
3.1. Undisturbed sleep
3.2. Deprivation of sleep
3.3. Cytokine effects on sleep
4. Sleep and endocrine regulation in the aged
4.1. Sleep-endocrine regulation of the HPA and the somatotropic system
4.2. Vasopressin in the elderly
4.3. Effects of hormone administration on sleep in the aged
5. Immune function during sleep in the aged
5.1. Immune system function in the aged
5.2. Sleep-associated immune function in the aged
5.3. Endocrine mediation of sleep-immune changes in the aged
6. Summary and conclusion
1. Introduction
Wakefulness and sleep show remarkable dierences in electrical brain activity,

which are paralleled by characteristic changes in the pattern of endocrine secre-
tion and immune activity. These uctuations in hormonal and immune activity,
however, reect not only an interaction with mechanisms regulating sleep and

DOI: 10.1016/S1566-3124(05)17005-9
114 B. Perras and J. Born
wakefulness, but also inuences from various other factors modulating sleep
endocrine and sleepimmune interactions. Most important in this respect are
circadian oscillators playing a supra-ordinate role in sleepwake regulation.
In addition, environmental (e.g., light exposure) and behavioral conditions (e.g.,
stress), and factors such as gender and age are to be considered in this context as
potent modulators.
During the wake phase, neuroendocrine and neuroimmune mechanisms assist
the body in coping with stressful environmental and internal events. However
during sleep, stressors and stress responses are absent, and activity in the
neuroendocrine and immune systems becomes subjected to a spontaneous and basal
mode of self-regulation whose mechanisms are not yet fully understood. In the
course of aging, prominent sleep-related changes of endocrine secretion and immune
function occur and contribute to the morbidity and mortality in the elderly.
Here we review the interaction of neuro-endocrine as well as neuro-endocrine
immune regulation with sleep, rst in young and then in aged humans. The focus will
be on the hypothalamo-pituitary-adrenal (HPA) system, the somatotropic system,
and the vasopressinergic system. Apart from eerent inuences of the sleeping
brain on the release of hormones into the blood stream, the aerent eects of
circulating hormones and cytokines of the immune system on the sleeping brain will
also be discussed.
2. Regulation of sleep-related endocrine systems in young humans
2.1. The HPA and somatotropic system
Together with the sympathetic nervous system, the bodys response to stressful
events during wakefulness is mediated through the secretory activity of both the
HPA axis and the somatotropic axis. Stress stimulates corticotropin-releasing
hormone (CRH), which is secreted from the hypothalamus with highest concen-
trations of the hormone found in the nucleus paraventricularis. This CRH induces
release of adrenocorticotropic hormone (ACTH) from the anterior pituitary into
the circulating blood, which subsequently induces the release of cortisol from
the adrenals. Stimulation of the HPA axis by physical, metabolic, and mental
stressors is in many cases paralleled by the secretion of growth hormone-releasing
hormone (GHRH) mainly from the basomedial hypothalamus, i.e., the arcuate
nucleus. This GHRH is the secretagogue of growth hormone (GH), which is released
from the anterior pituitary to regulate metabolic requirements at a cellular level,
mainly via stimulating the hepatic release of insulin-like growth factor I (IGF-I).
With the ending of the stress stimulus, the secretion of CRH and GHRH and all
subsequent hormones is terminated by negative feedback regulation. Since sleep is
generally regarded as a state devoid of stress, and in the light of the predominance of
stress-related activity in the wake phase, activity of the HPA and somatotropic
system during sleep can be expected to be low. However, examination of the
temporal patterns of ACTH/cortisol and GH secretion during sleep reveals a
complex and surprising relation between sleep, sleep stages, and hormonal activity.
Sleep Associated Endocrine and Immune Changes in the Elderly 115
Figure 1 illustrates the typical time course of HPA and somatotropic secretory
activity during nocturnal sleep in a healthy young man: the cyclic appearance of
nonREM sleep or slow wave sleep (SWS) and REM sleep over the night changes
in that the rst half of the night is characterized by a predominance of SWS while
the latter part of the night is dominated by REM sleep. In parallel, the changing
hormonal secretion shows very low ACTH and cortisol plasma concentrations
and concurrent bursts of plasma GH levels during the SWS-rich early part of sleep.
In contrast, in the second, late part of nocturnal sleep, the secretory activity of
W
REM
Sleep S1
S2
S3
S4
20.00 22.00 0.00 2.00 4.00 6.00 8.00 10.00 12.00 time
Cortisol ACTH 24 60
g/dl pg/ml
16 40
8 20
0 0
20.00 22.00 0.00 2.00 4.00 6.00 8.00 10.00 12.00 time
10
8
GH
ng/ml 6
4
2
0
20.00 22.00 0.00 2.00 4.00 6.00 8.00 10.00 12.00 time
Fig. 1. Time course of HPA and somatotropic secretory activity during nocturnal sleep in a healthy young
man. The upper panel shows the sleep prole characterized by a predominance of slow wave sleep (SWS)
during the early part of sleep, and a predominance of REM sleep during the late part. The middle panel
indicates associated plasma ACTH (dashed line) and cortisol (solid) levels indicating activity of the HPA
system. The lower panel shows the associated plasma growth hormone (GH) prole.
the HPA system is distinctly increasing, while activity of the somatotropic system
is minimal.
2.1.1. The HPA-system

Plasma levels of ACTH and cortisol are subject to a circadian rhythm (Weitzman
et al., 1971; Gallagher et al., 1973). This rhythm can be observed independent of
sleep and wakefulness. However, sleep plays an important synergistic role in this
regulation of HPA activity during the 24-h cycle. During normal nocturnal sleep,
both ACTH and cortisol concentrations reach a 24-h minimum (nadir) during the
early SWS-rich part of sleep and increase during the late part when REM-sleep is
predominant (Kupfer et al., 1983). Several studies have elucidated the relation
between sleep and release of ACTH and cortisol by manipulating sleep. For
example, Spath-Schwalbe et al. (1993) stimulated ACTH and cortisol secretion
by intravenous bolus injection of CRH during dierent periods of sleep and
wakefulness. In one condition CRH was injected during early sleep while the subject
was in a period of SWS. In another condition CRH was injected at the same time of
the night with the subject kept awake this time. Compared to the wake state, the
ACTH/cortisol secretory response to CRH was distinctly suppressed during early
SWS. No such suppression was found during late sleep when REM sleep prevails.
These and related observations indicate that early sleep and in particular, SWS
inhibits pituitaryadrenal activity.
The contribution of late sleep to the regulation of the HPA-axis is illustrated in
another experiment indicating that in comparison with wakefulness during the REM
sleep-rich late part of the night, the HPA system is disinhibited, so that plasma
concentrations of ACTH and cortisol are relatively increased during this period of
sleep (Spath-Schwalbe et al., 1991). Also, shortly after awakening in the morning
ACTH/cortisol concentrations decrease again which has led to the hypothesis that
the increase in HPA activity during late sleep prepares the organism for the stress
anticipated in the upcoming wake phase (Born et al., 1999; Pruessner et al., 2003).
Additionally, dierential eects of sleep stages and the ultradian structure of sleep
on the HPA system have been revealed. Especially during the late part of nocturnal
sleep the cyclic nature of sleep with the alternate appearance of nonREM and REM-
sleep is closely linked to respectively, global increases and decreases in secretory
activity of the anterior pituitary (Follenius et al., 1988). In conditions of free running
circadian rhythms and also when the sleep phase is experimentally shifted (by up to
12 h), it becomes evident that the suppressing inuence of SWS on secretion of
ACTH/cortisol is most prominent during a limited period around the habitual
bedtime, indicating an entrainment of this sleep-related inhibition of pituitary
adrenal activity to the circadian oscillation (Weitzman et al., 1971; Pietrowsky et al.,
1994; Weibel et al., 1995). Thus, when SWS is shifted by more than about 3 h into
the late part of the night or is advanced by the same time into the evening, the
suppressive inuence of SWS on pituitary adrenal function vanishes. However, if
occurring in phase with the circadian rhythm this inhibitory eect of early SWS is
quite eective and cannot be overridden even by stimulation of the HPA system with
a combined application of CRH and vasopressin (Bierwolf et al., 1997).
In sum, HPA activity is subject to a circadian inuence, with nadir values

during the early night and peak activity during the early morning hours. Nocturnal
sleep enhances this temporal pattern. Early sleep and, particularly, SWS suppresses
HPA secretory activity whereas late sleep has an overall activating inuence on
HPA-system which, however, is not mediated by REM sleep.
2.1.2. The somatotropic system

GH levels show a pattern reciprocal to that of HPA activity, reaching a maximum
during early SWS which coincides with nadir plasma concentrations of cortisol
(Takahashi et al., 1968). Minimum levels are observed during late sleep. Contrary
to the HPA system, secretory activity of the somatotropic system displays no
clear circadian rhythm and also shows remarkable dierences between males and
females. Whereas in men 70% of the total amount of GH secreted in a 24-h period is
released during the rst half of night-time sleep, in women, secretory patterns appear
to be more variable, with GH pulses more frequently occurring also spontaneously
during daytime before sleep and during the late half of the night (Quabbe et al., 1966;
Antonijevic et al., 1999). The reason for this sexual dimorphism is not known at
present. Nevertheless, as shown for men, the secretory activity of the GHRHGH
system is linked closely to the onset of sleep (Born et al., 1988). In this study, delayed
sleep onset (by 3 h) resulted in a parallel delay of the nocturnal rise in GH plasma
concentrations. To investigate the dependence of GH secretion on SWS, in a further
condition the men were selectively deprived of SWS during the early part of the night
by gently arousing them at the occurrence of rst signs of SWS without awakening
them. The absence of SWS did not inuence the nocturnal peak in plasma GH
concentration which remained time-locked to sleep onset and did not signicantly
decrease in amplitude. Another study expanded these observations to daytime
sleep, examining two groups of subjects who, after they had stayed awake the
previous night, were allowed to sleep for 8 h either after 11.00 am or 3.00 pm.
(Pietrowsky et al., 1994). Regardless of the time of sleep, the rise in GH secretion
was always closely linked to sleep onset showing a maximum within about 1 h
after sleep onset. Corroborative evidence for the strong dependence of GH release
on sleep stems from studies employing the constant routine protocol. In periods
of this protocol where sleep is absent, GH secretion is largely reduced and a distinct
GH peak near habitual bedtime is no longer observed (Czeisler et al., 1999).
The striking temporal coincidence of the rise of GH concentrations and the
rst periods of SWS led to further investigations indicating that the amount of
GH released is proportional to the amount of SWS (Van Cauter et al., 2000a).
Also, GHRH exerts a promoting inuence on SWS which is, however, independent
of its eect on GH release. A commonly accepted view is that GHRH although
originating from dierent hypothalamic structures is initiating both GH
secretion as well as SWS during the early part of sleep (Ehlers et al., 1986;
Obal et al., 1992; Spath-Schwalbe et al., 1995; Zhang et al., 1996; Toppila et al.,
1997). However, GH secretion as well as SWS do not solely depend on GHRH,
and are each under the control of separate factors. Thus, GH secretion is
inhibited by somatostatin (Muller et al., 1999) and strongly stimulated by ghrelin
(Kojima et al., 1999), a gut hormone recently discovered as an endogenous

ligand of the growth hormone secretagogue receptor. An interaction between
sleep and secretion of somatostatin and ghrelin has been suggested, although this
remains to be further claried in humans (Steiger et al., 1992; Weikel et al., 2003;
Dzaja et al., 2004).
In sum, somatotropic activity is linked to sleep onset showing a maximum during
the rst epochs of SWS. It is not clearly inuenced by circadian mechanisms. The
GHRH activity is probably a stimulus for both SWS and GH release coinciding
during the early part of nocturnal sleep.
2.2. Vasopressin (VP)
Vasopressin is produced in the nucleus supraopticus and nucleus paraventricularis

of the hypothalamus. Both nuclei project to the posterior pituitary, i.e., the
neurohypophysis, where VP is stored till release. Another source of VP is the nucleus
suprachiasmaticus of the hypothalamus which is the structure representing the
brains major circadian oscillator (Klein et al., 1991; Moore et al., 2002). VP plays an
important role in the regulation of uid homeostasis and blood pressure. It is
released from the posterior pituitary following osmotic stimuli and after
the stimulation of volume receptors in the systemic circulation. Also, VP is known
to act in addition as a potent stimulus of pituitaryadrenal activity. Naturally,
it acts synergistically with CRH to regulate the response to various stressors
(Spath-Schwalbe et al., 1987; Sugimoto et al., 1994). During sleep, VP plasma
concentrations are elevated with peak concentrations during the early part of the
night (Forsling, 1993). This temporal pattern seems to be linked to the nocturnal rise
of melatonin (Forsling, 2000). An association of VP secretion specically to sleep
stages has not been thoroughly examined although in one study such relationship
was not found (Rubin et al., 1978). Based on the high proportion of VP reactive
neurons in the nucleus suprachiasmaticus (Moore et al., 2002), a role of VP in
circadian regulation has been proposed. This view has been corroborated by ndings
of a marked 24-h cycle in the VP-producing activity of neurons in human nucleus
suprachiasmaticus (Hofman and Swaab, 1994; Hofman, 2003).
In combination with the lack of evidence for a sleep stage associated modulation
of VP release, the presence of large numbers of VP-expressing neurons in the
nucleus suprachiasmaticus suggest that potential interactions of VP with sleep
arise from its role in the circadian regulation of the sleepwake cycle.
2.3. Prolactin, melatonin, and other hormones
For the secretion of various other hormones, rhythms have been identied that
appear to be linked to sleep, sleep stages, as well as to circadian oscillations. For
example, the secretion of prolactin is enhanced during sleep and appears to be
secreted in particular during the early periods of SWS (Haus et al., 1980; Follenius
et al., 1988; Spiegel et al., 1995). During REM sleep, prolactin secretion is
inhibited. Plasma concentrations of thyroid stimulating hormone display a circadian
variation with a peak in the evening before nocturnal sleep exerting an inhibitory
inuence on the release of this hormone (Brabant et al., 1990; Goichot et al., 1992;
Allan and Czeisler, 1994). A hormone that has attracted much interest in sleep
research is melatonin, a product of the pineal gland. Diurnal melatonin plasma
concentrations are low and secretion of the hormone can be totally inhibited by light.
With the onset of darkness, melatonin levels rise sharply to reach a peak in the middle
of the night, thus representing an excellent indicator of the circadian clock. Sleep per
se does not substantially change melatonin secretion. However, melatonin has been
consistently found to facilitate sleep, thus reecting the drive of circadian rhythm on
sleep propensity. Probably this eect is mediated by an inuence on sleep inducing
thermoregulatory mechanisms (Cajochen et al., 2003).
Further hormones for which nonREM-REM sleep and sleep-dependent rhythms
of secretion have been described, are luteinizing hormone (Fehm et al., 1991),
oxytocin (Forsling, 2000), renin (Brandenberger et al., 1998), and aldosterone
(Charloux et al., 1999; see Czeisler and Klerman, 1999; Steiger, 2003; Touitou and
Haus, 2000 for reviews). Also, there are hormones, such as atrial natriuretic
peptide, the secretion of which does not follow a clear nycthemeral pattern
(Follenius et al., 1992).
3. Regulation of sleep-related immune function in young humans
Many of the endocrine signals discussed above exert distinct inuences on

immune functions. Accordingly, the sleep specic regulation of endocrine activity
goes along with a specic regulation of immune functions during sleep. The
characteristics of immune regulation specic to sleep have been the object of a
growing number of studies in humans (Moldofsky, 1995; Born, 1998; Sothern and
Roitmann-Johnson, 2001; Irwin, 2002; Marshall and Born, 2002). Most of these
studies examined variations in immune functions during undisturbed sleep or after
deprivation of sleep. To describe various aspects of the acute innate and adaptive
immune function in these human studies white blood cell dierential counts,
cytokine activity, proliferative capabilities of lymphocyte, and antibody responses
were measured. The principal eector cells of innate immunity are mononuclear
phagocytic cells (monocytes, macrophages), polymorphonuclear neutrophils,
dendritic cells and natural killer (NK) cells. Some of the cells of innate immunity,
notably macrophages, after infection secrete cytokines that produce inammation.
These so called proinammatory cytokines include interleukin-1 (IL-1), tumor
necrose factor- (TNF-) and IL-6. Development of an adaptive immune response
relies on an interplay of T and B lymphocytes mainly with antigen presenting
macrophages and dendritic cells. The adaptive immune response leads eventually to
the dierentiation of antigen specic eector and memory T and B cells, the latter
producing antigen specic antibodies. The T cell derived cytokine IL-2 plays a key
role in the formation of an adaptive immune response mainly by stimulating
proliferation and dierentiation of T cells. Overall, the presently available data
in humans point to an enhancing eect of sleep specically on aspects of
adaptive immunity, while acute innate immune functions appear to be less aected
or even diminished by regular sleep. Some evidence supports the notion that
sleep may play a facilitating role in establishing a primary adaptive immune
response to an antigen.
3.1. Undisturbed sleep
Undisturbed nocturnal sleep, embedded in a regular sleepwake schedule, is

associated with most prominent variations in the numbers of circulating white
blood cells. Numbers of granulocytes, monocytes as well as the major lymphocyte
subsets, including T-helper cells (CD4), cytotoxic T cells (CD8), activated
T cells (HLA-DR) and B cells (CD19 ), all appear to reach a maximum in the
evening hours and to decline most of the night-time in sleep to reach a minimum
in the morning hours (Ritchie et al., 1983; Miyawaki et al., 1984; Haus, 1992;
Palm et al., 1996; Born et al., 1997b; Haus and Smolensky, 1999). In contrast,
NK cell counts as well as the cytotoxic activity of these cells reach a minimum
during the rst part of nocturnal sleep, and thereafter increase till the afternoon
hours. While this pattern points to a dierential regulation of cell counts within
the dierent leukocyte subsets, their physiological meaning remains obscure.
A decrease in subset counts could reect increased migration of respective
cells into extravascular and lymphoid tissues, but could likewise be due to a
margination of the cells sticking to the walls of blood vessels or to an accumulation
in the spleen.
Nocturnal sleep appears to be also accompanied by specic changes in cytokine
activity although overall results are less consistent. This may in part be due to the
fact that, with the exception of IL-6, most cytokines of interest do not have a
hormone-like function. Thus, under normal non-pathological conditions the
endogenous levels of these cytokines in blood are close to the lower detection limit
of most assays. With this background, rather than determining cytokine levels
in serum some more recent studies measured the production of cytokines after
stimulating respective blood cells in vitro with mitogens, and with this approach
appeared to reveal more consistent changes in sleep-dependent cytokine activity.
Lipopolysaccharides (LPS) were used in most of these studies as mitogen stimulant
for cytokines released from monocytes and macrophages (IL-1, TNF-), whereas
phytohemagglutinin and concavalin A were used for stimulating release of T cell
cytokines, like IL-2 and interferon- (IFN- ).
Several reports suggested that the production of IL-1 and TNF- during
undisturbed nocturnal sleep is most of the time decreasing (Zabel et al., 1990, 1993;
Hohagen et al., 1993; Petrovsky and Harrison, 1998; Petrovsky et al., 1998).
However, these ndings were based on absolute cytokine production determined for
the sampled blood. When the production of IL-1 and TNF- was put into
relationship with the respective numbers of monocytes, the main cellular source of
this cytokine in the blood samples, the sleep-related temporal dynamics in cytokine
activity disappeared (Born et al., 1997b). This suggests that the sleep-related
variations in IL-1 and TNF- do not primarily reect changes in cytokine
production per se, but rather a decrease in the number of circulating cells producing
these cytokines.
In contrast with IL-1 and TNF-, concentrations of IL-6 in blood are
much higher, and this cytokine also acts as an important stimulus of pituitary
adrenal secretory activity (Spath-Schwalbe et al., 1996, 1998). However, reports
vary largely regarding a possible association of variations in IL-6 concentrations
with sleep. While some studies failed to nd any systematic changes in IL-6
concentrations across the 24-h cycle (Born et al., 1997b; Lissoni et al., 1998),
ndings from others suggest signicant maxima at dierent times of the day
(Gudewill et al., 1992; Zabel et al., 1993; Bauer et al., 1994; Sothern et al., 1995;
Vgontzas et al., 1999; Redwine et al., 2000). This divergence reects in part a
methodological problem. The IL-6 measurements obtained by frequent blood
sampling using an indwelling venous catheter over extended time periods have
been shown to produce spurious results, most likely due to local proinammatory
activity resulting from mechanical irritation of venous walls by the catheter
(Haack et al., 2002).
Distinct increases during nocturnal sleep have been observed for the production
of the T cell derived cytokine IL-2 (Born et al., 1997b; Lissoni et al., 1998). The
T-cellular production of IFN- seemed likewise to be enhanced during nighttime
sleep (Hohagen et al., 1993; Petrovsky et al., 1994, 1998; Petrovsky and Harrison,
1998). The increase in production of these T cell cytokines in these studies started
already before the sleep period and are independent of changes in the number of
circulating T cells.
Petrovsky and Harrison (1997) examined 24-h variations in the ratio of IFN-
to IL-10 production as an indicator of the balance between T helper 1/T helper 2
(Th1/Th2) cytokine activity, considered an essential determinant for the selection
of the eector mechanisms of immune defense. The Th1 cells releasing mainly
IFN- , aside from other cytokines including IL-2, become activated in response
to intracellular viral and bacterial challenges and support various cellular (type 1)
responses, including macrophage activation and antigen presentation. In contrast,
the cytokines characteristic of Th2 immunity, such as IL-4, IL-5, and IL-10, tend to
drive humoral (type 2) defense via stimulating mast cells, eosinophils, and B cells
against extracellular pathogens. Predominance of type 2 cytokines, as observed for
example in aged humans, has been associated with an inferior response to
vaccination (Ginaldi et al., 1999a). The work of Petrovsky and Harrison (1997, 1998)
suggests that nocturnal sleep favors a shift towards Th1 (predominance of IFN- )
mediated immune defense, peaking around 03:00 h.
3.2. Deprivation of sleep
While changes in immune functions observed in conditions of undisturbed

night-time sleep have apparent clinical relevance, their origin is ambiguous, since
they can reect an inuence of a circadian oscillator rather than of sleep.
To dissociate inuences of sleep from those of circadian rhythm, typically eects of
nocturnal sleep deprivation are compared with those of undisturbed nocturnal sleep.
Sleep deprivation, especially if extended for more than one night represents a stressful
experience that via activation of the major stress hormonal systems can per se
inuence immune functions. However, in humans, when staying awake for a single
night voluntarily such side eects appear to be limited. In fact, assessment in humans
indicates that plasma levels of cortisol and catecholamines during deprivation
of nocturnal sleep on a single night, although signicantly enhanced in comparison
with sleep, do not reach the levels seen during daytime wakefulness or stress (e.g.,
Dodt et al., 1994a; Lange et al., 2003). Overall data from studies employing total
or partial deprivation of sleep in a single night conrm that sleep per se enhances
signs of adaptive immunity. For some parameters, the eect of sleep appears
to add synergistically to a circadian inuence. Thus, compared with a night awake
or partial sleep deprivation, regular nocturnal sleep was consistently found
to induce reductions in the number of granulocytes, monocytes, and
the major lymphocyte subsets, but to increase NK cell counts (McClintick et al.,
1994; Irwin et al., 1996; Born et al., 1997b). In parallel, production of T cell derived
IL-2 has been demonstrated to be enhanced during regular sleep as compared
to nocturnal wake conditions (Uthgenannt et al., 1995; Irwin et al., 1996). Also,
nocturnal sleep was shown to induce a shift in the Th1/Th2 cytokine balance
towards increased Th1 activity, as indicated by an increased ratio of IFN- /IL-4
producing T helper cells. However, the Th1 shift was only of moderate size and
replaced by Th2 dominance during late sleep (Dimitrov et al., 2004a). Moreover,
recent studies provided the rst evidence for a facilitating eect of sleep on a
primary adaptive immune response to vaccination (Spiegel et al., 2002; Lange et al.,
2003). In the latter study healthy young humans either regularly slept or stayed
awake on the night following primary vaccination with inactivated hepatitis
A virus (at 9.00 am). Hepatitis A virus antibody titers measured 28 days
after vaccination, a time when antibody titers have reached a plateau, were
nearly two-fold higher in the subjects who had regular sleep following vaccination,
than in the subjects who stayed awake on this rst night.
Proinammatory cytokines of acute innate immunity, like IL-1 and TNF-
showed sleep associated decreases in comparisons with conditions of sleep
deprivation in some studies, although this again could reect mainly a sleep-related
decrease in the number of monocytes as a main source for these cytokines
(Uthgenannt et al., 1995; Born et al., 1997b).
In combination, results speak for an enhancing eect of sleep on processes
of adaptive immunity establishing antigen-specic defense, whereas acute innate
responses are less inuenced by sleep or even superior during times of
continued waking than during sleep. Indeed, periods of sleep deprivation
extended over 34 days have been found to be associated not only with distinctly
increased counts of neutrophils and monocytes as well as NK cell activity but
also with increased plasma IL-6 concentration (Dinges et al., 1994; Shearer et al.,
2001). Lymphocyte counts decreased under such conditions. Together this picture
speaks for a stress-like eect of extended periods of sleep deprivation, enhancing
in particular signs of non-specic phagocyte-related immunity (Boyum et al.,
1996).
3.3. Cytokine effects on sleep
Observations of increased tiredness that can accompany acute infection have

stimulated the hypothesis that some cytokines especially those involved in acute
innate immunity act as signals to the brain to promote sleep. In animals, this view
was elaborated by the group of Krueger (Krueger et al., 1999, 2003) who showed that
proinammatory cytokines such as IL-1 and TNF- stimulate two major cascades
of reactions that promote SWS via activation of GHRH and via activation of nuclear
factor-B (NF-B) followed by induction of prostaglandin 2 or IL-2, respectively.
In humans, sleep promoting eects of proinammatory cytokines are
overall less well established. The group of Pollmacher (Pollmacher et al., 1993,
1995; Korth et al., 1996; Mullington et al., 2000) administered healthy humans
Salmonella abortus endotoxin inducing acute increases in plasma concentrations of
TNF-, IL-6, GH, ACTH, and cortisol. After doses of 0.8 and 0.4 ng/kg body
weight of endotoxin, there was a signicant increase in time spent in sleep stage
2, whereas SWS remained unaected. However, when an even lower dose of 0.2 ng/
kg body weight of endotoxin was administered which unlike the higher doses did
not induce fever or cortisol release, time spent in SWS was acutely enhanced. On
the other hand, endotoxin at higher doses decreased time in REM sleep. This eect
is probably due to the increased release of cortisol known to acutely suppress
REM sleep (Born et al., 1989).
Two studies investigated changes in sleep in healthy humans after subcutaneous
administration of low doses of IL-6 and interferon- (IFN-) representing key
mediators of acute innate immunity to bacterial and viral infection, respectively
(Spath-Schwalbe et al., 1998, 2000). Both cytokines acutely suppressed SWS,
which dominates the early part of nocturnal sleep. Following IL-6, the acute
suppression was followed by enhanced rebound of SWS during the latter part of
the night, suggesting that via secondary mechanisms IL-6 might have delayed
the enhancing eects on SWS. Interestingly, the suppression of SWS following
both cytokines was associated with substantially increased feelings of tiredness
measured before sleep, a pattern observed similarly in many depressed patients
who complain of fatigue but simultaneously show reduced SWS (Born and
Spath-Schwalbe, 1999). Indirect evidence that proinammatory cytokines such as
IL-1 and TNF- exert enhancing inuences on SWS derives from a study by
Schuld et al. (1999) on the eects of granulocyte-colony stimulating factor (G-CSF).
The G-CSF administered subcutaneously before sleep reduced SWS within the rst
2 h of the sleep period. The decrease occurred in parallel to steep increases in the
plasma levels of IL-1 receptor antagonist (IL-1ra) and soluble TNF- receptors
(p55, p75) known to antagonize eects of IL-1 and TNF-, respectively. Moreover,
an increase in TNF- during the late night after G-CSF was paralleled by an increase
in SWS. The T cell cytokine IL-2 proved ineective in modulating sleep after
administration of low doses in humans (Lange et al., 2002).
Overall these data from human studies suggest that proinammatory cytokines
of acute innate immunity at concentrations inducing fever acutely suppress SWS
and enhance feelings of tiredness. However, very slight increases in plasma
concentration of these cytokines as they may occur as part of non-pathological

uctuations or follow acute innate responses could support the expression of sleep
and SWS.
4. Sleep and endocrine regulation in the aged
4.1. Sleep-endocrine regulation of the HPA and the somatotropic system
Disturbed sleep is a common complaint in the elderly. Assessment in the sleep

laboratory reveals that the elderly at night, typically display increased time in bed, a
fragmentation of sleep by frequent arousals, and increased time spent awake and
in stage 1 sleep after sleep onset. Also, a reduction in total sleep time, SWS and in
advanced age in REM sleep is observed (Prinz et al., 1990). Furthermore, SWS
in the elderly is characterized by a reduced EEG power (Ehlers and Kupfer, 1989).
However, subjective complaints of sleep loss do not necessarily predict sleep
laboratory results, although correlations between subjective and polysomnographic
indicators of sleep quality appear to be higher in aged than in young subjects
(Carskadon et al., 1976; Roehrs et al., 1983; Mendelson et al., 1986).
Aging is also associated with distinct alterations of the neuroendocrine pattern
of sleep, which appear to be even more pronounced than those of central nervous
sleep architecture (Kern et al., 1996; Deuschle et al., 1997; Van Cauter et al., 2000b;
Steiger, 2003). The most prominent feature of these alterations pertains to the
regulation of the HPA and the somatotropic systems (Fig. 2). In parallel with the
well-known and distinct decrease in SWS with increasing age, a gradual and
highly signicant increase in cortisol secretion in the course of aging can be observed
which is particularly consistent around the time of nadir concentrations during
early sleep. This basal hypercortisolism indicates a weakening of inhibitory control
over HPA activity during early sleep. Concomitantly, the GH peak amplitude
during early sleep distinctly declines with age (Kern et al., 1996). These hallmarks
of the age-related decrease in endocrine function are of considerable clinical
relevance since they contribute essentially to the morbidity in elderly people.
Disorders common in the aged which are associated with a more or less pronounced
hypercortisolism are obesity, the metabolic syndrome, depression, and memory loss
(Seeman and Robbins, 1994; Lupien et al., 1999; Sapolsky, 1999; McEwen, 2000).
Additionally, characteristic changes in the circadian timing system develop in the
course of aging. The most obvious of these circadian changes are a phase advance of
the sleepwake cycle and an attenuation of the amplitude of circadian rhythms. The
disturbances are linked to obvious functional changes in brain structures like
the nucleus suprachiasmaticus representing the circadian clock. For example, the
number and circadian oscillation of VP-reactive neurons in this nucleus have been
found to deteriorate with aging which results in a dampening of the circadian rhythm
and probably also of circannual rhythms (Hofman et al., 1994; Swaab, 1995). Also,
factors arising from age-related behavioral changes in environmental adaptation,
such as a decreased exposure to light and social clues, as well as general physical and
GH max (g/l) Cort nadir (nmol/l)
20 140
r = 0.80 120
15 p< 0.001
100
10 80
60
5 40 r = 0.79
20 p< 0.001
0 0
15 25 35 45 55 65 75 85 95 15 25 35 45 55 65 75 85 95
age (years) age (years)
SWS (min) REM (min)

35 35
30 30 r = 0.55
r = 0.56 p< 0.01
25 p< 0.01 25
20 20
15 15
10 10
5 5
0 0
15 25 35 45 55 65 75 85 95 15 25 35 45 55 65 75 85 95
age (years) age (years)
Fig. 2. Age-related impairment in the sleep associated regulation of HPA and somatotropic secretory
activity. The GH peak amplitude during early sleep distinctly declines with increasing age (top left).
Concomitantly, cortisol nadir values steadily increase during early sleep with increasing age, indicating
a disinhibiting inuence of age on basal HPA secretory activity during early sleep (top right). SWS and
REM sleep (bottom panels) show the well-known and distinct decrease in SWS and REM sleep with
increasing age. (From Kern et al., 1996).
mental impairments can adversely aect the circadian timing system thereby
aggravating the circadian maladaptation in the aged.
The mechanisms accounting for the age-related sleep-endocrine changes are
presently unclear. A promising approach to this issue probably relies on a detailed
analysis of the mechanisms mediating the inhibition of HPA activity in conjunction
with the increase in somatotropic activity during early sleep. In addition, contri-
butions of impaired circadian control are to be considered. It has been proposed,
that in combination the phase advance and reduced amplitude of circadian rhythms
in aged lead to a desynchronization of major endogenous circadian rhythms, so
that rhythms like those of cortisol, melatonin, and temperature are not properly
coupled to the sleepwake cycle. This desynchronization may indeed account for
some of the sleep disturbances in the aged (Czeisler et al., 1992; Touitou and Haus,
2000; Pandi-Perumal et al., 2002).
4.1.1. Hippocampal control of HPA activity

Hormonal secretory activity of the HPA axis is controlled via supra-ordinate
brain structures. Most important in this context is the hippocampus, a structure of
the limbic system, that is believed to regulate the HPA system to achieve homeostatic
balance and adaptation to stress. Hippocampal inuence aects both circadian

rhythm and stress regulation of pituitaryadrenal activity. But the hippocampus is
also the target of feedback signals from the periphery within the HPA system and
mediates inhibition of the system by negative feedback (Jacobson and Sapolsky,
1991). Thus, the hippocampus is responsible for timely and adequate termination of
the stress response during wakefulness, while during sleep, it mediates tonic
inhibition of pituitaryadrenal activity (De Kloet et al., 1998).
To understand disturbances of HPA activity in the aged, the dysregulation of
feedback signals in the course of aging is of interest, especially of those signals
regulating the basal activity of this system during sleep. At a cellular basis
glucocorticoid receptors (GR) and mineralocorticoid receptors (MR) participate
in the regulation of feedback. The GR are widely distributed in the brain,
whereas MR are expressed at a high density selectively in the hippocampus and
some associated regions. These receptors play a most important role in the tonic
control of HPA activity during sleep, but to a lesser degree are also involved in
the regulation of the HPA system during wakefulness (Deuschle et al., 1998;
Young et al., 1998). Binding anity of cortisol to MR is higher than to GR.
Accordingly, the relative occupation of the receptors by cortisol depends critically
on the plasma concentrations of the steroid. The balance between MR and
GR activation in turn determines the eect of the hippocampus on the HPA system
(De Kloet et al., 1998). In humans, at the time of the cortisol nadir during early
sleep there is little occupation of the classical GR in the brain while MR in
the hippocampus remain occupied by more than 80%. With the rise in cortisol
during late sleep, in addition, GR become increasingly occupied. Blocking of MR
by (pre-) treating healthy humans with canrenoate almost completely abolishes
sleep-dependent suppression of pituitaryadrenal activity during the early night
and also reduces SWS substantially, a pattern resembling that of early sleep in the
aged (Born et al., 1997a). Moreover, experiments showed that increased activation
of MR after administration of lower doses of cortisol increased the time in SWS
(Born et al., 1991; Friess et al., 1994, 2004). On the other hand, a reduction in SWS
is observed following administration of GR agonists like dexamethasone or when
cortisol concentrations are very high. Also, activation of GR by dexamethasone
decreases REM sleep in the late part of the night. A similar REM sleep reduction
can be induced by cortisol administration adding to the high endogenous cortisol
levels during late nocturnal sleep (Born et al., 1991).
Thus hypercortisolism in conjunction with reduced limbic MR expression could
be a mechanism relevant for the age-associated decreases in both SWS and REM
sleep.
4.1.2. Disturbance of hippocampal function during aging

A commonly held view is that protective and adaptive eects of HPA hormones
turn out to be damaging if their action prevails inadequately long. Hyperactivity of
the HPA-system has been detected in a number of clinically relevant conditions not
only in patients with obesity, diabetes, and the metabolic syndrome exhibiting a
predominant increase of abdominal fat, but also in patients with insomnia
(Bjorntorp, 2001; Vgontzas et al., 2001; Rosmond, 2003) . In the case of aging, it
has been hypothesized that repetitive stressful events during lifetime result in an
elevation of steroid plasma concentrations due to an increasing impairment of
feedback inhibition. Compelling evidence has been provided that elevated steroid
levels aect brain function adversely with the hippocampus being the structure in the
brain most susceptible to these eects (McEwen, 1999; Sousa et al., 2000). Following
repeated stress, reduced excitability of hippocampal neurons, suppression of
neurogenesis in the dentate gyrus and atrophy of the dendrites of the CA 3 region
of the hippocampus have been reported (Pavlides et al., 1995; Gould et al., 1997;
De Kloet et al., 1998; McEwen, 1999). As a consequence, the loss of MR leads in
the hippocampus to a weakening of the tonic basal inhibition of the HPA axis
(Sapolsky et al., 1986). Additionally, overactivity of the HPA system negatively
interferes with other neuronal functions in the hippocampus. Hippocampal
atrophy in the elderly is associated with mild cognitive impairment (Golomb et al.,
1994; Convit et al., 1995). Furthermore, enhanced cortisol plasma concentrations
seem to be correlated with impaired memory in parallel with reduced hippocampal
volume in the elderly humans (Lupien et al., 1998). Apart from aging, hippocampal
nerve atrophy has also been found in dementia, depressive illness, and Cushings
disease (De Leon et al., 1993; Sheline et al., 1996; Starkman et al., 1999).
Overall, the age-related dysfunction of the hippocampus primarily via loss of MR
expressing neurons gives rise to the overactivity of the pituitaryadrenal system seen
most consistently during the basal nadir activity of early sleep. The release of tonic
inhibition manifests in an augmented hypothalamic release and action of CRH at
the pituitary level. The CRH itself has a disturbing eect on sleep. In humans, the
intravenous administration of four repetitive boluses of 50 mg of CRH during the
early part of nocturnal sleep signicantly reduced the amount of SWS and REM
sleep (Holsboer et al., 1988). Furthermore, there is evidence suggesting a role of
CRH in the development of depression which is also often associated with age-like
disturbances of sleep, i.e., a reduction in sleep continuity and amounts of SWS
(Benca et al., 1992). In light of the similarity of the sleep disturbances in the elderly
and depressed patients to those resulting from administration of CRH, it has been
proposed that an increased release of CRH is the principal neuroendocrine factor
mediating the changes in sleep architecture in aging (Steiger, 2002).
4.1.3. Corticotropin release-inhibiting factor

Eective inhibition of HPA secretory activity during early nocturnal sleep is
essential for normal neuroendocrine sleep regulation and implies an active
mechanism of control by limbic-hippocampal structures (Bierwolf et al., 1997;
Born and Fehm, 1998). This inhibition can be established via MR mediated neural
inhibition of the CRH releasing cells, but also by stimulating the hypothalamic
release of a corticotropin release inhibiting factor (CRIF), that could serve to
counteract the stimulating inuence of CRH on pituitary release of ACTH (Engler
et al., 1999). Up to now, a substance that selectively acts as CRIF has not been
identied. However, there are some candidates which inhibit HPA secretory activity
in humans and might exert this action naturally during early nocturnal sleep (Engler
et al., 1999; Jessop, 1999). Among them is atrial natriuretic peptide (ANP). Bierwolf
et al. (1998) showed in humans, that nocturnal ACTH and cortisol responses to
intravenous injections of a combination of CRH and VP were distinctly diminished
when the subjects were simultaneously infused with a low dose of ANP over a period
of 90 min, starting 15 min before CRH/VP injection. Plasma concentrations of ANP
are known to increase with age (e.g., Kato et al., 2002). However, whether ANP or
one of the related brain peptides functions as an inhibitor of HPA activity under
natural conditions during early humans sleep is unknown.
In contrast to ANP, GHRH is a substance whose inhibiting inuence on
pituitaryadrenal activity under natural conditions during early sleep is well
documented. The somatotropic axis exhibits distinctly enhanced activity exactly
at the time of the nadir of HPA activity. Repetitive intravenous injections of
GHRH during early sleep reduced plasma cortisol concentrations during sleep
(Steiger et al., 1992). This nding was conrmed using an intranasal route of GHRH
administration known to enable a direct access of the peptide to the cerebrospinal
uid compartment (Perras et al., 1999a, Born et al., 2002). Notably, in contrast to
the intravenous injection, intranasal GHRH simultaneously reduced secretion of
GH during early sleep supporting the view that the peptide acted mainly at the
hypothalamic level to induce inhibition of GHRH release. This hypothalamic
action obviously overrides stimulating eects on pituitary GH release resulting
from GHRH partly absorbed into the bloodstream. Together, the data indicate
that GHRH suppresses HPA activity probably mainly at the hypothalamic level.
The data lend themselves to suggest that decline in GHRH activity in the course
of aging signicantly contributes to the tonic overactivity of the HPA system during
sleep in elderly.
Another factor exerting inhibitory actions on HPA secretory activity during
nocturnal sleep is melatonin the secretion of which depends on darkness (Leproult
et al., 2001). Evidence for an inhibiting eect of melatonin on cortisol release
during early sleep has been provided by a recent study in totally blind persons
(Fischer et al., 2003). Aside from disturbed sleep patterns the blind in these
experiments showed temporal patterns of ACTH and cortisol secretion which did
not appear to be coupled to the sleep process. Specically, they did not show
minimum and maximum secretion of these hormones during early and late sleep,
respectively, but the average concentration of these hormones did not at all dier
between early and late halves of nocturnal sleep. The administration of melatonin
before sleep normalized this pattern leading to distinctly suppressed levels of
ACTH/cortisol during early sleep and increased levels during late sleep. The eects
of melatonin were accompanied by a slight enhancement of SWS. The gradual
decrease in nocturnal melatonin secretion occurring in the course of aging (e.g.,
Toutiou, 1995; Skene and Swaab, 2003) could thus well cause a desynchronisation
between the sleepwake rhythm and pituitaryadrenal activity during sleep in
the aged, expressing in relatively enhanced cortisol levels during early sleep. Of
note, since pineal melatonin secretion is triggered by darkness rather than sleep,
the decrease in melatonin represents a contribution of a circadian impairment to
the disturbed neuroendocrine pattern of sleep in the aged.
There are several further candidate substances that may act as CRIF, such as
preproTRH158183 whose analog in the rat has been shown to inhibit stimulated
cortisol release (Redei et al., 1995, 1998; McGivern et al., 1997), -endorphin
and substance P (Jessop, 1999). However, the central nervous eects of these
substances on HPA activity in humans are not clear, as well as their role during
sleep (Lieb et al., 2002).
Overall the available data suggest that the inhibition of HPA activity during
early sleep involves multiple mechanisms acting not only at the pituitary but also
at the hypothalamic (like GHRH, melatonin) level of this axis. Such signals
mediate a supra-ordinate feedback control exerted by the hippocampus as well
as a circadian control probably originating from the nucleus suprachiasmaticus.
Inecient inhibition of pituitaryadrenal activity during sleep in the aged could
thus reect a decline in production and release of substances acting as CRIFs
at the hypothalamic and pituitary level in this system. Impaired release of these
factors would add to the diminished hippocampal negative feedback sensitivity to
cortisol in the aged, and thereby further weaken the tonic inhibitory control of
ACTH/cortisol release.
4.1.4. The somatotropic system in the elderly

For the somatotropic system an age associated deterioration in function is well
known (Corpas et al., 1993, Rusell-Aulett et al., 1999). The amount of sleep
associated release as well as stimulated release of GH is reduced. Consequently,
insulin-like growth factor (IGF)-1 levels are also diminished. Several mechanisms
may account for this impairment. The release of peptide in GHRH producing
neurons is diminished in relation to age with a marked decrease of GH secretory
pulses (De Gennaro Colonna et al., 1989). Also, the GH response to challenge of the
somatotropic system by GH secretagogues is reduced (Shibasaki et al., 1984; Iovino
et al., 1989; Deslauriers et al., 1991). Increased production of somatostatin probably
adds to the age-related impairment of the GHRH-GH system (Cocchi et al., 1986).
Whether ghrelin, a major stimulant of GH secretion, is also involved in the age
associated decrease in GH release, is presently not clear.
The somatotropic dysfunction in the aged is closely linked to the sleep dis-
turbances. In particular, the amount of SWS declines in the course of aging, and is
strongly correlated to the reduction of GH secretory activity (Prinz et al., 1995; Kern
et al., 1996, Van Cauter et al., 2000a). Both reduced amounts of SWS and secreted
GH reect a decrease in GHRH secretory activity. This probably has several
reasons, including decreased functioning of GHRH expressing neurons, increased
release of somatostatin inhibiting GHRH release, and perhaps also a reduced
sensitivity to GH secretagogues such as ghrelin.
In the light of the unique constellation during the early SWS-rich part of sleep of
stimulated somatotropic activity and a concomitantly inhibited HPA system it has
been proposed, that the age-related disturbance of the neuroendocrine regulation
is a consequence of an altered balance between the activity of the two systems.
Higher nocturnal cortisol concentrations with lower GH levels in the aged as
compared to young subjects thus indicates a shift in this balance in favor of the
HPA system. The view of a balance between the two systems is supported by a
body of evidence indicating opposite eects of these systems on sleep as well as
mutually inhibiting eects between the systems (Ehlers et al., 1986; Ehlers and
Kupfer, 1987; Steiger, 2002). Not only does GHRH promote SWS but also
suppresses pituitaryadrenal activity during early sleep. In contrast, overactivity of
CRH aside from stimulating ACTH/cortisol release results in sleep suppression.
Also, increased HPA activity exerts some inhibitory inuence on somatotropic
activity. In humans repeated intravenous administration of CRH in parallel
with reductions in SWS and REM sleep signicantly reduced the amount of
GH secreted during sleep (Holsboer et al., 1988).
4.2. Vasopressin in the elderly
In the aged, ndings on blood concentrations of VP during night-time and sleep

are inconsistent. Compared to young subjects, elevated, reduced, and unchanged
levels of VP have been observed depending on sex and the presence of the nocturnal
polyuria syndrome (ONeill and McLean, 1992; Johnson et al., 1994; Nadal et al.,
1994; Asplund, 1995; Forsling et al., 1998). Also VP concentrations in the elderly are
not related to serum osmolality as in young subjects, although a decreased ability of
the kidneys to concentrate urine has been described (Johnson et al., 1994; Forsling
et al., 1998). However, VP plasma levels do not necessarily reect changes in VP
related to the circadian alterations in the aged, since it does not stem from the
nucleus suprachiasmaticus.
In fact, aging dierentially aects vasopressinergic transmission in the brain.
While the VP content of the nucleus supraopticus and paraventricularis is constant
or even increased in the aged, the number and volume of VP-producing cells in
the nucleus suprachiasmaticus is reduced (Swaab, 1995). The selective loss of
vasopressinergic neurons in a brain structure known as the circadian pacemaker of
the body playing a key role in the regulation of the sleepwake cycle led to the
assumption, that disturbances of sleep and sleepwake regulation can be explained
by a dysfunction of vasopressinergic signaling in the nucleus suprachiasmaticus
(Czeisler et al., 1992).
4.3. Effects of hormone administration on sleep in the aged
In view of the pronounced changes in neuroendocrine activity developing in the

course of aging, in a small but increasing number of studies, hormones have been
administered in order to test them for possible ameliorating inuences on sleep in
elderly. Two dierent strategies have been adopted in these studies: In most cases the
exogenous administration of a substance (like GHRH) aimed at compensating
directly for a presumed age-related decline in endogenous activity of the respective
hormone. However, hormones (like cortisol) were also administered in a few cases
to enhance negative feedback eects in order to reduce a presumed age-related
overactivity of a certain hypothalamic factor.
4.3.1. Hormones of the HPA system

The administration of cortisol at lower doses in young humans increased SWS,
reduced REM sleep and in some studies increased also GH plasma concentrations
(Born et al., 1989; Friess et al., 1994, 2004). A decrease in REM sleep is also seen
following infusion of ACTH, presumably mediated via ACTH-induced cortisol
release. Blocking MR by administration of canrenoate induced a decrease in SWS in
combination with elevated plasma cortisol concentration, while activation of GR
after administration of dexamethasone reduced REM sleep (Fehm et al., 1986;
Born et al., 1991, 1997a). This pattern was taken to suggest that SWS can be
enhanced by increasing activation of MR whereas predominant GR activation
results mainly in reduced REM sleep.
On this background the group of Steiger (Bohlhalter et al., 1997) conducted a
remarkable experiment in aged humans, in which they assessed the inuence of
repeated intravenous injections of cortisol on the neuroendocrine architecture of
sleep. Thus, the exogenous cortisol was added to already enhanced endogenous
cortisol concentration in the aged. Cortisol injections started at 5.00 pm with a dose
of 1 mg/kg body weight and were repeated hourly until 6 am the next morning, but
with a reduced dose of 0.3 mg/kg body weight. In a parallel study in young subjects
(Friess et al., 1994), it was shown that cortisol distinctly increased SWS, decreased
REM sleep, and strongly increased the secretion of GH. Note these eects represent
an immediate feedback inuence of cortisol and are not expected with long-term
administration of cortisol known to be associated with restlessness and insomnia
(e.g., Krieger and Glick, 1974). The increase in SWS might reect increased MR
activity while the decrease in REM sleep could stem from enhanced GR activation
in this experiment. However, experiments in dogs provided hints that aging is
accompanied by a selective decrease in MR capacity while numbers of GR in the
pituitary even increased with age (Reul et al., 1991; Rothuizen et al., 1993).
Bohlhalter et al. (1997) took these data to argue that a mediation of the cortisol-
induced increase in SWS in the aged is unlikely to be mediated via MR activation.
They propose that the eects of cortisol on SWS and GH secretion rather originate
from immediate eects on the somatotropic axis, enhancing sensitivity to GHRH
(e.g., Seifert et al., 1985) and possibly reducing in parallel the inhibiting inuence of
somatostatin. A strengthened feedback suppression of hypothalamic CRH probably
adds to the eects, given that in young humans CRH appears to disturb sleep
(Holsboer et al., 1988; Chang and Opp, 2001). Together these data support the view
of a disturbed balance between activation of the HPA and somatotropic system
characterizing sleep in the aged, which acutely can be eectively shifted towards
increased somatotropic activity by enhancing the cortisol feedback signal to the
brain (Ehlers et al., 1986; Holsboer et al., 1988).
4.3.2. Hormones of the somatotropic system

In order to compensate for the obvious decline in somatotropic activity in the
course of aging, primarily treatments with GHRH have been considered. This is
partly due to the fact that trials examining the eects of GH administration in
healthy young subjects as well as in dierent groups of patients with GH deciency

failed to reveal consistently positive eects on sleep (Mendelsohn et al., 1980; Wu
and Thorpy, 1988; Kern et al., 1993). Also, GHRH is preferred since the GH decit
during sleep in aged is considered to be caused by reduced hypothalamic GHRH
activity. In fact, in a number of studies intravenous administration of GHRH in
healthy controls not only increased GH release but also improved sleep, resulting in
consistent though moderate increase in the time spent in SWS (Steiger et al., 1992;
Kerkhofs et al., 1993; Marshall et al., 1996, 1999). However, also negative results
were reported (Kupfer et al., 1991). Notably, the improving eects of GHRH on
sleep appeared to depend critically on the way of administering the substance.
Steiger et al. (1992) used the repetitive intravenous administration of four boli of
50 mg GHRH each (total dose 200 mg of GHRH) and observed signicant increases
in SWS and GH concentrations while plasma cortisol concentrations during sleep
were decreased. Marshall et al. (1996, 1999) observed essentially the same changes
when using the same repetitive mode of intravenous GHRH administration.
However, no or only minor changes in sleep were obtained when comparable doses
of GHRH were administered as continuous infusion or in repeated boli of smaller
size. From these and related results Marshall et al. (1999) concluded that the short-
term induction of very high GHRH concentration in plasma is the factor critical for
producing central nervous eects on sleep. The GHRH does not easily pass the
BBB. Extreme peak concentrations as reached after a bolus of 50 mg GHRH may
represent conditions in which a sucient amount of the substance can enter the
brain compartment.
Based on the successful trials in young subjects, the group of Steiger conducted
two trails in the aged with GHRH, relying on the same mode of administering
repeated intravenous boli as used in the studies in young humans. Guldner et al.
(1997) investigated thirteen healthy elderly with a mean age of 69 years, and found
that GHRH acutely reduced the time awake and increased REM latency. Also GH
secretion was enhanced, although not consistently. However, there was neither an
eect on SWS nor on plasma cortisol concentrations. Murck et al. (1997) explored
the eects of a long-term treatment with GHRH in two aged subjects. The two
elderly were treated on a rst night with four intravenous boli (50 mg each). Then,
treatment was continued such that they received another GHRH bolus of 100 mg at
9.00 a.m. every second day for the 12 following days. However, subsequent testing in
the sleep laboratory did not reveal substantial changes. Sleep period time and time in
SWS in the two elderly were even reduced as compared with the baseline nights
before treatment. Also, there was no eect of GHRH on GH release. From these
overall less promising results the authors concluded that the process of aging leads to
a reduced ecacy of GHRH which cannot be compensated by administering the
substance. A possible explanation for this is an increased activity of or sensitivity to
somatostatin in the aged. Consistent with this view, Frieboes et al. (1997) showed
that in the elderly, administration of somatostatin further impaired sleep mainly by
reducing total sleep time and REM sleep. This response contrasted with those in
young humans who did not show notable changes in sleep structure following
administration of somatostatin.
Another reason for the reduced ecacy of GHRH in these studies could be, that
even with the administration of repeated boli in the elderly the amount of substance
reaching the brain is not sucient for inducing notable eects on sleep. Based on this
rationale Perras et al. (1999a) performed a study in the young and the elderly using
an intranasal administration of GHRH. This route of administration has been
proven in human studies to provide direct access of neuropeptides to the
cerebrospinal uid compartment (Born et al., 2002). After intranasal administration
of GHRH in the elderly before the night in the sleep laboratory substantially
improved several sleep-endocrine markers of aging (Perras et al., 1999a). The time in
SWS and also in REM sleep was enhanced, with these changes concentrating in the
late night. Remarkably, cortisol nadir concentrations in the aged after intranasal
GHRH were on average even lower than those in the young men after placebo. Peak
concentrations of cortisol were reached later in the night suggesting a phase delay of
circadian rhythm. However, as noted earlier, nocturnal GH concentrations were
signicantly reduced after GHRH in both young and aged indicating that intranasal
GHRH activates a short-loop negative feedback inhibition at the hypothalamic level
of this system. Also, increases in SWS and REM sleep were comparable for the
young and aged subjects, indicating that GHRH does not specically compensate for
the age related decits in sleep, although the brains sensitivity to GHRH appeared
to be preserved in the elderly.
4.3.3. Vasopressin
Vasopressin was among the rst neuropeptides whose eects on brain functions
were studied in detail in animals. From these studies an important role of VP for
memory and learning was inferred, although in humans the eects of vasopressin
administration on these functions appeared to be generally less consistent (Nebes
et al., 1984; Van Ree et al., 1985; De Wied et al., 1989; Dodt et al., 1994b; Born et al.,
1998c). The trials in humans also included aged subjects with and without memory
decits. Despite the well-known presence of VP in the nucleus suprachiasmaticus as
a circadian regulator of the sleepwake cycle, surprisingly few studies in animals
looked at the eects of VP on sleep (Urban et al., 1978; Danguir, 1983; Kruisbrink
et al., 1987; Arnauld et al., 1989; Brown et al., 1989). The results were interpreted in
terms of increased arousal induced by acute administration of VP.
Arousing inuences on sleep were observed also following acute administration of
VP in humans. In young subjects, VP consistently increased time spent in stage 2
sleep and awake regardless of whether administered intranasally or intravenously
(Timsit-Berthier et al., 1982; Born et al., 1992). Intravenous VP administration
also reduced REM sleep. Subjective sleep quality was not found to be changed
in one study following intranasal administration of an analog of VP (DGAVP,
Snel et al., 1987).
The failure to nd consistently improving eects on brain functions in the elderly
after single or short-term administration of VP led to the assumption that more
encouraging benets result perhaps from a more prolonged treatment with the
peptide. Perras et al. (1996) tested this assumption rst in a pilot study in two healthy
elderly persons. The focus of this study was on brain indicators of cognitive
functions (event-related brain potentials), but sleep was also tested. The two
elderly persons were treated with VP twice daily (20 IU) for a period of three
months. An intranasal route of VP administration was used to assure direct brain
access of the peptide (Born et al., 2002). Surprisingly, results indicated most
pronounced changes in sleep rather than on neurocognitive functions. The time
spent in SWS was increased by about 100% with this enhancement emerging after six
weeks of treatment. Interestingly, this long-term eect was opposite to the arousing
eect of VP on sleep seen acutely after administration of a single dose of the peptide,
pointing to a dierent mechanism activated with prolonged VP treatment.
These results prompted more elaborate investigations of long-term intranasal
treatment with VP in the elderly (Perras et al., 1999b, 2003). In the rst of these
studies, 26 healthy elderly (14 women and 12 men >70 years) were tested, who did
not complain of poor sleep but, at the sleep laboratory, showed the typical
fragmentation of sleep and predominance of light sleep. After three placebo nights,
one group of subjects continued to take a placebo while the other group was
treated with VP according to the same protocol as used in the foregoing pilot study
(2 20 IU per day, Perras et al., 1996). The treatments were administered each
day in the morning and in the evening, for three months. Revaluation of sleep at
the end of the treatment epoch conrmed a pronounced increase in SWS in the
VP group averaging 20 min. Moreover, compared to the controls the VP treated
elderly slept longer (for 45 min) and displayed increased REM sleep (10 min,
Fig. 3). No cardiovascular side eects or uid retention were observed. The
improvement in polysomnographical sleep was not correlated with an improve-
ment of sleep quality ratings. However, this was probably a consequence of the fact
that the elderly selected for this trial perceived a good sleep quality already before
the study.
Considering this remarkable improvement of sleep after prolonged VP treatment
in the elderly, a further study (Perras et al., 2003) aimed to explore whether the
benecial eect is accompanied by changes in the neuroendocrine pattern of sleep in
the aged. In particular, pituitaryadrenal activity was of interest since VP is known
to be a potent co-stimulator of the pituitary release of ACTH following acute
administration (Spath-Schwalbe et al., 1987). The 26 healthy elderly (mean age 72.9
years) of this study, with only mild sleep complaints were treated with placebo or
intranasal VP for a period of 10 weeks, according to the same study protocol and
dosage of daily treatment as in the two previous studies with subchronic VP.
However, this time polysomnographical recordings in the beginning and end of
the treatment period were complemented by repeated blood sampling during the
nights. Again, the prolonged intranasal treatment with VP led to a profound increase
in SWS averaging 21.5 min. The VP-induced increase in SWS was persistent and
was also observed on a night following the last treatment, i.e., 24 h after the last
intake of VP, excluding any contributions of immediate eects of the peptide.
Notably, rather than increasing pituitaryadrenal activity, VP signicantly decreased
the cortisol nadir during early sleep on average by 0.5 mg/dl. Changes in GH
concentrations were not signicant. Again, no side eects regarding uid balance
or cardiovascular activity were observed. Overall, these results indicate a promoting
W W
Sleep Associated Endocrine and Immune Changes in the Elderly

REM REM
S1 S1
S2 S2
S3 S3
S4 S4
23.00 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 time 23.00 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 time
W W
REM REM
S1 S1
S2 S2
S3 S3
S4 S4
23.00 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 time 23.00 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 time
Fig. 3. Sleep proles from two selected subjects (70 and 76 years old) before (left panel) and after (right panel) daily intranasal treatment with vasopressin
during a period of 3 months. Note, sleep time, time in SWS and in REM sleep are distinctly increased in both subjects. Lights were turned o at 11.00 pm
(23.00 h). W, awake time; S1, stage 1 sleep; S2, stage 2 sleep; S3, stage 3 sleep; S4, stage 4 sleep; S3+S4, SWS. (From Perras et al., 1999b).
135
eect on sleep of prolonged intranasal treatment with VP in the aged, accompanied

by a benecial rather than impairing inuence on sleep-associated HPA activity.
The experiments show that the improving inuence of VP on sleep requires
a prolonged period of treatment. While the arousing inuence of VP seen
after short-term administration of the peptide probably reects acute changes in
neurotransmitter systems involved in arousal regulation, the eects of long-term VP
administration obviously originate from slower processes, possibly inducing some
kind of plastic changes in the respective neuronal networks. In rats, prolonged
treatment with VP increased the capacity of MR in the hippocampus (Veldhuis
et al., 1982). This change would be expected to strengthen the inhibiting inuence
of the hippocampus on basal HPA secretory and, thus, could well explain the
reduction in cortisol nadir values during sleep in elderly developing after prolonged
VP treatment. Given that blocking of MR substantially reduced SWS in young
humans (Born et al., 1991), increased central nervous MR capacity after prolonged
VP treatment could also explain the profound increase in SWS in the elderly of the
VP group.
Alternatively, prolonged VP administration may compensate for the age-related
reduction of VP in the nucleus suprachiasmaticus (Swaab, 1995), thus improving
VP-dependent output of the circadian pacemaker. This explanation would implicate
changes towards enhanced activity during daytime, which remains to be examined.
5. Immune function during sleep in the aged
While a considerable number of studies in animals and humans have provided

convergent evidence for mutual and specic interactions between sleep and
immune functions, surprisingly few studies have directly addressed this relationship
in the aged organism, and of these studies, a minority have been conducted in
humans. The paucity of studies regarding this issue is in contrast with its obvious
clinical relevance. Aging is not only characterized by a specic impairment of sleep
but also there is a general consensus that immune system function declines with
age (Miller et al., 1996; Ginaldi et al., 1999 a,b,c; Franceschi et al., 2000; Linton
and Dorshkind, 2004). Infections such as pneumonia and inuenza are among
the leading causes of death in persons aged 65 and older. The two principal questions
to be answered in this context are: (i) Are the changes in immune system function
in elderly persons in part a consequence of the altered sleep? (ii) To what extent
does the age-related decline in immune function contribute to some of the
disturbances characterizing the architecture of sleep in the elderly? Both questions
cannot be answered on the basis of available data, although overall the data suggest
that the disturbed sleep has at least an aggravating inuence on the age-related
decline in immune function.
5.1. Immune system function in the aged
Aging is associated with multiple changes of immunity that have been collectively
termed immunosenescence. Its most prominent feature is a general decline in the
number of T cells, mainly of naive T cells, which is commonly linked to a progressive

involution of the thymus (Murasko and Goonewardene, 1990; Ginaldi et al., 1999a).
The decline of naive cells is apparent for the CD4 T helper cell subset and
associated with relatively increased numbers of CD4 cells with memory phenotype
coexpressing CD45RO (Sansoni et al., 1993; Rea et al., 1996; Rink and Seyfarth,
1997; Ginaldi et al., 2000). For CD8 cytotoxic T cells the decline appears to be less
dramatic. Also, T cell responsiveness is decreased in the elderly. Proliferation to
stimulation with dierent mitogens is reduced as well as the diversity of the T cell
receptor repertoire for dierent antigens (e.g., Pisciotta et al., 1967; Lehtonen et al.,
1990; Posnett et al., 1994; Schwab et al., 1997). There has been a number of reports
indicating that the production of T cell derived cytokines like IL-2 and IFN- is
decreased in elderly humans (Gillis et al., 1981; Nagel et al., 1989), although with
regard to IL-2 this was not conrmed in all studies especially in those using whole
blood stimulation techniques (Sindermann et al., 1993; Ahluwalia et al., 2001).
A further prominent feature of T cell related immunosenescence is a shift in the
Th1/Th2 balance towards predominant production of Th2 cytokines including
IL-4 and IL-10 (Cakman et al., 1996; Shearer, 1997). The B cell numbers and
antibody production are likewise distinctly decreased in the elderly. This, in
combination with the marked decits mainly regarding the naive CD4 subset of
the T cell compartment can explain the widely reported clinical observation of
a decline in the response to vaccination also, for example to inuenza vaccine
and hepatitis B vaccine (Remarque, 1999; Looney et al., 2001). However, this
decline has not been universally conrmed, highlighting the possibility of interve-
ning variables including sleep that could act to modulate this relationship (Carson
et al., 2000).
The T cell related decits contrast with signs of enhanced innate immunity with
aging. While the number of monocytes/macrophages appear to remain stable with
age the production of monocyte/macrophage-related proinammatory cytokines
such as IL-1, TNF-, and IL-6 has been consistently found to be enhanced (e.g.,
Fagiolo et al., 1993; Rink et al., 1998). Also, the number of circulating NK cells is
increased with age, although their proliferative responses and cytotoxic activity are
decreased (Facchini et al., 1987; Solana and Mariani, 2000). The nding of an
increased number of circulating activated T cells (Sansoni et al., 1993; Hulstaert
et al., 1994) may likewise t the picture of an enhanced state of innate immune
activation that may be compensatory for the decline in T cell related immune
function.
5.2. Sleep-associated immune function in the aged
The characteristics of immunosenescence in some aspects bear distinct similarities

with the changes observed in young subjects following deprivation of sleep. In both
states, marked changes towards diminished T cell mediated immune functioning
appear to be paralleled overall by selected signs of increased innate immune
activation. Notably, both immunosenescence and sleep deprivation are associated
with reduced production of the T cell cytokines IL-2 and IFN- , and a shift of the
Th1/Th2 balance towards increased Th2 cytokine activity (Nagel et al., 1989; Born
et al., 1997b; Shearer, 1997; Dimitrov et al., 2004a). Moreover, both states have been
found to impair the formation of a primary adaptive immune response to vaccines
(Remarque, 1999; Looney et al., 2001; Lange et al., 2003). Regarding innate immune
functions, sleep deprivation like aging leads to increased numbers of circulating
NK cells but, a diminished NK cell activity when considered on a per cell basis
(Irwin et al., 1994; Solana and Mariani, 2000). Blood counts of activated T cells
(HLA-DR) have likewise been found to be increased after sleep deprivation and
in the aged (Sansoni et al., 1993; Born et al., 1997b). Also, the production of
proinammatory cytokines IL-1, TNF- and IL-6 has been found to be enhanced
in the aged as well as after sleep deprivation, although the eects of sleep deprivation
were in general less pronounced and, if cytokines were assessed in mitogen-
stimulated whole blood samples, appeared to primarily reect eects on circulating
monocytes (Born et al., 1997b; Rink et al., 1998).
Collectively these data suggest that the disturbances of sleep commonly present
in aged persons induce a state of chronic sleep decit that acts to suppress T cell
mediated functioning and, in turn, upregulates certain aspects of innate immune
functioning. However, there is currently no study providing a direct test of this
hypothesis. Sakami et al. (2003) examined relationships between insomnia
determined mainly as perception of insucient sleep and T cell cytokine activity
as well as NK cell activity in a larger sample of 254 men aged between 20 and
64 years. While age was inversely, but only moderately correlated with NK cell
activity as well as with the IFN- to IL-4 ratio (r<0.19), the men suering from
insomnia had a signicantly lower production of IFN- and, in fact, a shift towards
prevailing Th2 cytokine activity. The NK cell activity was found to be independent
of insomnia. On the background of the strong association between aging and
disturbed sleep, this result provides rst evidence that poor sleep, if present in an
elderly person, probably contributes to a decline in specic T cell functioning by
shifting the Th1/Th2 balance towards Th2. However, even though insomnia was
examined in that study, the change in the Th1/Th2 balance may not be specic to the
impairment of sleep but could likewise reect the non-specic impact of a stressor.
Stressors other than insomnia and chronic sleep decit have also the capability to
induce a dominance of Th2 cytokine activity, and such factors are dicult to
dissociate from those of altered sleep, particularly so in the aged (Decker et al., 1996;
Marshall and Agrarwal, 2000).
There are also clear dierences between the eects of sleep deprivation and
aging mainly regarding blood cell counts. Most obvious are the opposite eects
of both conditions on lymphocyte counts which increase during sleep deprivation
(e.g., Dinges et al., 1994) but decrease with aging due to thymic involution. The
increase in T and B cell counts in blood during sleep deprivation probably reects
a diminished emigration of these cells into extravascular and lymphoid tissues
(Dickstein et al., 2000). Such dierences underline that immunosenescence cannot
be simply reduced to disturbances of sleep playing in this context the role of a
modulating factor. Against this background, dierences in immune regulation
between old and young humans during sleep are expected to be a composite, on
the one hand, of changes due to biological aging and, on the other hand, due to
impairments of sleep.
In one study in humans, white blood cell dierential counts and cytokine
production was compared between young and elderly subjects before and
during regular nocturnal sleep (Born et al., 1995). All subjects were healthy
and did not suer from pathological sleep disturbances. The aged subjects
(mean SEM: 79.6 7.5 year) were recruited according to the strict criteria of the
SENIEUR protocol. As expected, sleep in the elderly persons was worse than in the
young persons, although total sleep time was roughly the same as in the young
sleepers. Compared to the young subjects the elderly spent much more time awake
(24.0 vs. 5.2%), and distinctly less time in SWS (7.7 vs. 20.0%) and REM sleep (10.0
vs. 19.9%). In parallel with this, the aged showed a general reduction in the number
of circulating white blood cells. Monocyte counts were unchanged, while the
circadian decrease in neutrophil counts in the late night appeared to be somewhat
more pronounced in the aged. The T and B cell counts as well as the number of
CD4 and CD8 were distinctly reduced at all times. The number of activated
T cells were more than twofold enhanced in the aged throughout the recordings.
Notably, only changes in NK cells appeared to depend on sleep. The NK cell
numbers were comparable before sleep in both groups but, in young subjects
decreased across nocturnal sleep, while in the aged, on an average, even a slight
increase was observed.
Contrary to expectations, there was no clear hint from this study at a decline in
the production of IL-2 and IFN- by T cells in the aged. Although the production of
these cytokines per T cell before sleep was on average lower in the aged than in the
young, this dierence was not signicant, and this held true also during the time of
sleep. However, T cells in the aged expressed a higher number of IL-2 receptors. In
contrast with the T cell derived cytokines, the production of acute proinammatory
cytokines of innate immunity, IL-1 and TNF-, exhibited an increase in the aged
which was most pronounced during sleep. This was partly due to the fact that
average production of these cytokines tended to decrease across the sleep period only
in the young subjects which conrmed previous studies (Hohagen et al., 1993, Zabel
et al., 1993; Petrovsky et al., 1998). The sleep-associated enhancement in the
production of IL-1 and TNF- in the aged remained signicant also, when the
production of the cytokines were determined per monocyte representing the major
source of these cytokines in the stimulated whole blood samples. Together with the
increase in NK cell numbers, the enhanced proinammatory cytokine activity
during sleep may indeed reect acute consequences of the impaired sleep regulation
in the aged.
Another study in mice likewise failed to detect any sleep-related decline in signs of
adaptive immune function with age (Renegar et al., 1998). This study aimed at
dissociating eects of sleep deprivation on IgA and IgG levels in the upper and lower
respiratory tract mucosa and in serum in young and senescent mice 72 h after a viral
challenge test. The inuenza immune mice were sleep deprived once or twice for up
to 6 h on the day(s) before and after the viral challenge. While sleep deprivation
remained without eect in the young mice, the old mice showed even increased
serum levels of inuenza specic IgG after viral challenge under one of the
deprivation conditions. The nding may t the predominance of Th2 cytokine
activity supporting specic antibody secretion from B cells. However, it would be
also consistent with an increased response in the old mice to the stress of sleep
deprivation. Whether sleep does at all exert regulatory inuences on a secondary
response in the solidly immune organism is questionable. In humans, an enhancing
eect of sleep has been found on antibody titers only in the context of a primary
immunization (Spiegel et al., 2002; Lange et al., 2003). In any case, it would be
premature to generalize these ndings in rodents to the human being, in light of the
fact that eects of age on immune reactions can widely dier among species.
In another animal study in rats, LPS was administered to examine eects of the
acute phase proinammatory cytokines on sleep in young and middle-aged rats
(Schielholz and Lancel, 2001). The LPS is a robust stimulus of the secretion of IL-
1, TNF- and IL-6 from monocytes in rats and humans. Sleep in the young and
middle-aged rats diered in this study mainly regarding REM sleep and pre-REM
sleep (a type of nonREM sleep at the transition into REM sleep) which were both
reduced in the middle aged rats. The intraperitoneal administration of LPS induced a
comparable increase in body temperature in both groups. However, while the young
rats showed increased amounts of SWS in response to LPS, in the middle-aged rats
EEG power decreased in most frequency bands pointing to attened sleep. This
response in the middle-aged rats resembles that seen in adult humans following
administration of higher pyrogenic doses of LPS (Pollmacher et al., 1995). This
result well agrees with the notion that the sensitivity of innate immunity to LPS is
enhanced with aging (e.g., Gabriel et al., 2002), and extends it to the acute phase
sleep response to immune challenge.
In sum these data indicate that the impaired sleep in aged is acutely associated
with a state of increased responsiveness of some innate immune functions. In the
case of acute infection increased amounts of circulating proinammatory cytokines
could contribute to a further attening of central nervous sleep processes in the
aged. Also, the reduced depth of sleep may support the predominance of Th2
cytokine activity in the elderly. However, although primary T cell cytokines like IL-2
and IFN- have been shown to be acutely enhanced by sleep and to be generally
reduced with aging, so far no evidence has been provided that these cytokines
are particularly suppressed during the sleep period in aged. Thus, the decline
in lymphocyte generation and function representing one of the most prominent sign
of immunosenescence, appears to go in parallel with an insensitivity to the acute
consequences of poor sleep.
5.3. Endocrine mediation of sleep-immune changes in the aged
Sleep in the aged is associated with distinct endocrine changes. Many of the
hormones, the release of which is specically regulated during sleep, exert specic
immune-regulatory functions and, thus, could play an important role for the
manifestation of age-related changes in the immune system function, i.e., for the
acute changes emerging during the sleep period as well as for the persistent changes
of immunosenescence. Research in this regard has focussed on GH and prolactin,

cortisol, and melatonin. Also functions of other hormones like TSH showing a sleep-
associated decrease in aged (Van Coevorden et al., 1991), and catecholamines, which
are secreted at a reduced rate during sleep and at an increased rate in aged, and
counteract proinammatory cytokines and NK cell activity, might be relevant in this
context but will not be considered here (Maess et al., 2000; Elenkov and Chrousos,
2002; Irwin et al., 2003).
As noted before, the profound decrease in GH secretion is a landmark of sleep in
the aged. There is now a body of evidence indicating that subchronic treatment with
GH and also with IGF-1 can restore the architecture of the involuted thymus gland
by reversing the loss of immature cortical thymocytes and preventing the decline in
thymulin synthesis that occurs in old and GH-decient animals and probably also in
humans (Weigent and Blalock, 1995; Burgess et al., 1999). This implicates that the
persisting decline in sleep-associated GH secretion is one factor playing a causative
role for the age-dependent thymic involution explaining the enduring decrease in
T cell numbers and generation in elderly.
Reduced secretion of GH at the same time could also contribute to some of the
changes in T cell function characterizing immunosenescence. The GH has been
found to enhance mitogen-stimulated T cell dierentiation, and to increase Th1
related cytokine activity but to reduce Th2 activity (Postel-Vinay et al., 1997;
Mellado et al., 1998; Takagi et al., 1998; Dimitrov et al., 2004b). In children,
repeated GH administration increased serum levels of IL-2 and IFN- as well as of
proinammatory cytokines (IL-1 and TNF-; Bozzola et al., 2003). A stimulating
inuence of GH on NK cell activity is less well established (Bidlingmaier et al., 1997).
However, GH appears to stimulate macrophage activity also (Edwards et al., 1992).
The decline in GH in the elderly probably acts in synchrony with a similar sleep-
associated decline in the release of prolactin with aging (Van Coevorden et al., 1991).
Like GH, prolactin is preferentially released during SWS and acts via receptors of
the same cytokine/hemopoetin receptor superfamily, sharing binding anity and
a similar intracellular protein cascade during transduction (Matera et al., 2000;
Yu-Lee, 2002).
A supportive action on Th1 cytokine activity has been demonstrated also for
melatonin (Garcia-Maurino et al., 1997; Lissoni et al., 1998; Kuhlwein and Irwin,
2001). The age-related decrease in the secretion of melatonin during nocturnal
sleep could thus further add to the T cell related decits characterizing immuno-
senescence. However, at the same time melatonin appears to stimulate proin-
ammatory cytokines such as TNF- and IL-6 (e.g., Sutherland et al., 2002).
The corticosteroids are, of course, the best studied hormonal regulators of
immune functions, which at the same time are subject to a sleep specic regulation,
and the release of which during sleep increase with aging. Cortisol is known
to strongly aect the distribution of white blood cells inducing a pronounced
decrease in the numbers of circulating lymphocytes. Also, it inhibits production of
proinammatory cytokines as well as T cell cytokines (Fauci et al., 1976, Angeli et al.,
1999). Petrovsky and Harrison (1997) showed that the shift in the Th1/Th2 balance
towards increased Th1 cytokine activity during early nocturnal sleep can be
eectively inhibited by prior administration of cortisol. However, the concentrations

of corticosteroids examined in these experiments were in general rather high in
comparison with the small size of the age-related increased plasma cortisol
concentrations seen during sleep. This should caution against a premature
generalization from these studies to the conditions in the elderly. It can also not
be excluded that small increases in cortisol exert eects opposite to those of
higher doses (Calandra et al., 1995). Though, it is more likely that slight increases in
cortisol as seen during sleep in the aged in the long run act in the same way as the
acute but much higher corticosteroid concentrations, typically used for experimental
purposes.
The increased activity of proinammatory cytokines during sleep in the aged can
conversely act to modulate the neuroendocrine architecture of sleep. Cytokines such
as IL-1, TNF- and IL-6 have strong inuences on the pituitary, where they
stimulate pituitaryadrenal activity and at the same time suppress GH release (e.g.,
Jones and Kennedy, 1993, Spath-Schwalbe et al., 1996). These inuences suggest that
the enhanced state of innate immune responsiveness in the aged could contribute to
the sleep impairment seen in the aged, at least with respect to its neuroendocrine
features. This contrasts, however, with a concept derived from animal studies of a
promoting eect of such proinammatory cytokines on sleep which in part is
mediated via induction of GHRH (e.g., Kruger and Majde, 2003). Overall these
obvious discrepancies call for further research directly addressing the conditions of
sleep in aged humans.
6. Summary and conclusion
Early sleep is characterized by a trias of neuroendocrine phenomena, comprised

of a predominance of SWS, a maximum inhibition of HPA activity, and a strong
activation of somatotropic activity. During aging, there are distinct changes of
this trias. It decreases SWS, disinhibits pituitaryadrenal activity, and reduces
somatotropic activity during this time. In addition, sleep in the aged is less
deep, fragmented by frequent awakenings and inuenced by a phase advance
of circadian sleepwake regulation. Immunological measures reect an acute
sleep-related enhancement of signs of innate immune activity in the aged, beyond
a general decline in T cell related immune function, which is not restricted to the
sleep period.
The mechanisms of these changes are presently not clear. A key role is played by
the hypothalamic structures balancing the activity between CRH and GHRH
secreting neurons, which in aged enable a shift towards increased activity of
CRH and the pituitary adrenal system. This shift in neuroendocrine balance
during sleep is likely responsible for some of the enduring changes character-
izing immunosenescence, in particular, the decline of T cell numbers and function.
However, it cannot explain acute increases of signs of innate immune function in
the aged.
The balance between the activity of the HPA and the somatotropic system is
under the control of several supra-ordinate structures. Two of these structures are
the hippocampus and the nucleus suprachiasmaticus. Aging is associated with a

deterioration of hippocampal functioning including a reduction of MR receptor
expressing cells. The resulting decrease in inhibitory feedback regulation of the HPA
system probably adds to the overactivity of the HPA system seen during sleep in the
elderly. Aging is also associated with a decrease in vasopressin-expressing cells
selectively in the nucleus suprachiasmaticus representing the brains major circadian
clock. Such a decrease in combination with decreases in other circadian signals,
such as pineal melatonin, might be responsible for the circadian disturbances of sleep
manifesting a generally more shallow and phase advanced sleep.
The neuroendocrine changes of sleep in the aged are pronounced and clinically
relevant. With this background, an increasing number of trials have been started that
are aimed at ameliorating sleep in the aged by the administration of hormones
centrally involved in neuroendocrine sleep regulation. Promising approaches pertain
to the administration of substances counteracting HPA overactivity during sleep and
to the replacement of factors assumed to improve circadian functioning in the aged,
such as vasopressin and melatonin.
References
Ahluwalia, N., Mastro, A.M., Ball, R., Miles, M.P., Rajendra, R., Handte, G., 2001. Cytokine production
by stimulated mononuclear cells did not change with aging in apparently healthy, well-nourished
women. Mech. Ageing Dev. 122, 12691279.
Allan, J.S., Czeisler, C.A., 1994. Persistence of the circadian thyrotropin rhythm under constant
conditions and after light-induced shifts of circadian phase. J. Clin. Endocrinol. Metab. 79, 508512.
Angeli, A., Masera, R.G., Sartori, M.L., Fortunati, N., Racca, S., Dovio, A., 1999. Modulation by
cytokines of glucocorticoid action. Ann. N. Y. Acad. Sci. 876, 210220.
Antonijevic, I.A., Murck, H., Frieboes, R., Holsboer, F., Steiger, A., 1999. On the gender dierences in
sleep-endocrine regulation in young normal humans. Neuroendocrinology 70, 280287.
Arnauld, E., Bibene, V., Meynard, J., Rodriguez, F., Vincent, J.D., 1989. Eects of chronic icv infusion of
vasopressin on sleep-waking cycle of rats. Am. J. Physiol. 256, R674R684.
Asplund, R., 1995. The nocturnal polyuria syndrome (NPS). Gen. Pharmaco. 26, 12031209.
Bauer, J., Hohagen, F., Ebert, T., Timmer, J., Ganter, U., Krieger, S., 1994. Interleukin-6 serum levels
in healthy persons correspond to the sleep-wake cycle. Clin. Investig. 72, 315.
Benca, R.M., Obermeyer, W.H., Thisted, R.A., Gillin, J.C., 1992. Sleep and psychiatric disorders.
A meta-analysis. Arch. Gen. Psychiatry 49, 651668.
Bidlingmaier, M., Auernhammer, C.J., Feldmeier, H., Strasburger, C.J., 1997. Eects of growth hormone
and insulin-like growth factor I binding to NK cells. Acta. Paediatr. Suppl. 423, 8081.
Bierwolf, C., Struve, K., Marshall, L., Born, J., Fehm, H.L., 1997. Slow wave sleep drives inhibition of
pituitary-adrenal secretion in humans. J. Neuroendocrinol. 9, 479484.
Bierwolf, C., Burgemeister, A., Luthke, K., Born, J., Fehm, H.L., 1998. Inuence of exogenous atrial
natriuretic peptide on the pituitary-adrenal response to corticotropin-releasing hormone and
vasopressin in healthy men. J. Clin. Endocrinol. Metab. 83, 11511157.
Bjorntorp, P., 2001. Do stress reactions cause abdominal obesity and comorbidities? Obes. Rev. 2, 7386.
Bohlhalter, S., Murck, H., Holsboer, F., Steiger, A., 1997. Cortisol enhances non-REM sleep and growth
hormone secretion in elderly subjects. Neurobiol. Aging 18, 423429.
Born, J., 1998. Sleep and immune functions. In: M. Schedlowski, U. Tewes (Eds.), Psychoneuro-
immunology: A Textbook. Plenum Press, New York, pp. 417442.
Born, J., Fehm, H.L., 1998. Hypothalamus-pituitary-adrenal activity during human sleep: a coordinating
role for the limbic hippocampal system. Exp. Clin. Endocrinol. Diabetes 106, 153163.
Born, J., Spath-Schwalbe, E., 1999. The role of interferon-alpha in the regulation of sleep. In: N. Muller
(Ed.), Psychiatry, Psychoimmunology, and Viruses (Key Topics in Brain Research). Springer-Verlag,
Wien, pp. 131144.
Born, J., Muth, S., Fehm, H.L., 1988. The signicance of sleep onset and slow wave sleep for nocturnal
release of growth hormone (GH) and cortisol. Psychoneuroendocrinology 13, 233243.
Born, J., Spath-Schwalbe, E., Schwakenhofer, H., Kern, W., Fehm, H.L., 1989. Inuences of
corticotropin-releasing hormone, adrenocorticotropin, and cortisol on sleep in normal man. J. Clin.
Endocrinol. Metab. 68, 904911.
Born, J., de Kloet, E.R., Wenz, H., Kern, W., Fehm, H.L., 1991. Gluco- and mineralocorticoid eects on
human sleep: a role of central corticosteroid receptors. Am. J. Physiol. 260, E183E188.
Born, J., Kellner, C., Uthgenannt, D., Kern, W., Fehm, H.L., 1992. Vasopressin regulates human sleep by
reducing rapid-eye-movement sleep. Am. J. Physiol. 262, E295E300.
Born, J., Uthgenannt, D., Dodt, C., Nunningho, D., Ringvolt, E., Wagner, T., Fehm, H.L., 1995.
Cytokine production and lymphocyte subpopulations in aged humans. An assessment during
nocturnal sleep. Mech. Ageing Dev. 84, 113126.
Born, J., Steinbach, D., Dodt, C., Fehm, H.L., 1997a. Blocking of central nervous mineralocorticoid
receptors counteracts inhibition of pituitary-adrenal activity in human sleep. J. Clin. Endocrinol.
Metab. 82, 11061110.
Born, J., Lange, T., Hansen, K., Molle, M., Fehm, H.L., 1997b. Eects of sleep and circadian rhythm on
human circulating immune cells. J. Immunol. 158, 44544464.
Born, J., Pietrowsky, R., Fehm, H.L., 1998. Neuropsychological eects of vasopressin in healthy humans.
Prog. Brain Res. 119, 619643.
Born, J., Hansen, K., Marshall, L., Molle, M., Fehm, H.L., 1999. Timing the end of nocturnal sleep.
Nature 397, 2930.
Born, J., Lange, T., Kern, W., McGregor, G.P., Bickel, U., Fehm, H.L., 2002. Sning neuropeptides: a
transnasal approach to the human brain. Nat. Neurosc. 5, 514516.
Boyum, A., Wiik, P., Gustavsson, E., Veiby, O.P., Reseland, J., Haugen, A.H., 1996. The eect of
strenuous exercise, calorie deciency and sleep deprivation on white blood cells, plasma
immunoglobulins and cytokines. Scand. J. Immunol. 43, 228235.
Bozzola, M., De Benedetti, F., De Amici, M., Jouret, B., Travaglino, P., Pagani, S., Conte, F.,
Tauber, M., 2003. Stimulating eect of GH on cytokine release in children. Eur. J. Endocrinol. 149,
397401.
Brabant, G., Prank, K., Ranft, U., Schuermeyer, T., Wagner, T.O., Hauser, H., Kummer, B., Feistner, H.,
Hesch, R.D., von zur, M.A., 1990. Physiological regulation of circadian and pulsatile thyrotropin
secretion in normal man and woman. J. Clin. Endocrinol. Metab. 70, 403409.
Brandenberger, G., Charloux, A., Groner, C., Otzenberger, H., 1998. Ultradian rhythms in hydromineral
hormones. Horm. Res. 49, 131135.
Brown, M.H., Nunez, A.A., 1989. Vasopressin-decient rats show a reduced amplitude of the circadian
sleep rhythm. Physiol. Behav. 46, 759762.
Burgess, W., Liu, Q., Zhou, J., Tang, Q., Ozawa, A., Van Hoy, R., Arkins, S., Dantzer, R., Kelley, K.W.,
1999. The immune-endocrine loop during aging: role of GH and insulin-like growth factor-I.
Neuroimmunomodulation 6, 5668.
Cajochen, C., Krauchi, K., Wirz-Justice, A., 2003. Role of melatonin in the regulation of human circadian
rhythms and sleep. J. Neuroendocrinol. 15, 432437.
Cakman, I., Rohwer, J., Schutz, R.M., Kirchner, H., Rink, L., 1996. Dysregulation between TH1 and
TH2 T cell subpopulations in the elderly. Mech. Ageing Dev. 87, 197209.
Calandra, T., Bernhagen, J., Metz, C.N., Spiegel, L.A., Bacher, M., Donnelly, T., Cerami, A.,
Bucala, R., 1995. MIF as a glucocorticoid-induced modulator of cytokine production. Nature 377,
6871.
Carskadon, M.A., Dement, W.C., Mitler, M.M., Guilleminault, C., Zarcone, V.P., Spiegel, R., 1976.
Self-reports versus sleep laboratory ndings in 122 drug-free subjects with complaints of chronic
insomnia. Am. J. Psychiatry 133, 13821388.
Carson, P.J., Nichol, K.L., OBrien, J., Hilo, P., Jano, E.N., 2000. Immune function and vaccine
responses in healthy advanced elderly patients. Arch. Intern. Med. 160, 20172024.
Chang, F.C., Opp, M.R., 2001. Corticotropin-releasing hormone (CRH) as a regulator of waking.
Neurosci. Biobehav. Rev. 25, 445453.
Charloux, A., Groner, C., Lonsdorfer-Wolf, E., Piquard, F., Brandenberger, G., 1999. Aldosterone
release during the sleep-wake cycle in humans. Am. J. Physiol. 276, E43E49.
Cocchi, D., Calderini, G., Ganzetti, I., Galbiati, E., Parenti, M., Muller, E.E., 1986. Aspects of the control
of prolactin and growth hormone secretion in aging. In: E.E. Muller, R.M. McLeod (Eds.),
Neuroendocrine Perspectives, Elsevier, Amsterdam, pp. 191204.
Convit, A., de Leon, M.J., Tarshish, C., De Santi, S., Kluger, A., Rusinek, H., George, A.E., 1995.
Hippocampal volume losses in minimally impaired elderly. Lancet. 345, 266.
Corpas, E., Harman, S.M., Blackman, M.R., 1993. Human growth hormone and human aging. Endocr.
Rev. 14, 2039.
Czeisler, C.A., Dumont, M., Duy, J.F., Steinberg, J.D., Richardson, G.S., Brown, E.N., Sanchez, R.,
Rios, C.D., Ronda, J.M., 1992. Association of sleep-wake habits in older people with changes in
output of circadian pacemaker. Lancet. 340, 933936.
Czeisler, C.A., Klerman, E.B., 1999. Circadian and sleep-dependent regulation of hormone release in
humans. Recent Prog. Horm. Res. 54, 97130.
Danguir, J., 1983. Sleep decits in rats with hereditary diabetes insipidus. Nature 304, 163164.
De Gennaro Colonna, V., Zoli, M., Cocchi, D., Maggi, A., Marrama, P., Agnati, L.F., Muller, E.E., 1989.
Reduced growth hormone releasing factor (GHRF)-like immunoreactivity and GHRF gene expression
in the hypothalamus of aged rats. Peptides 10, 705708.
De Kloet, E.R., Vreugdenhil, E., Oitzl, M.S., Joels, M., 1998. Brain corticosteroid receptor balance in
health and disease. Endocr. Rev. 19, 269301.
De Leon, M.J., Golomb, J., George, A.E., Convit, A., Tarshish, C.Y., Mc Rae, T., De Santi, S., Smith, G.,
Ferris, S.H., Noz, M., 1993. The radiologic prediction of Alzheimer disease: the atrophic hippocampal
formation. AJNR Am. J. Neuroradiol. 14, 897906.
De Wied, D., Van Ree, J.M., 1989. Neuropeptides: animal behaviour and human psychopathology. Eur.
Arch. Psychiatry Neurol. Sci. 238, 323331.
Decker, D., Schondorf, M., Bidlingmaier, F., Hirner, A., von Ruecker, A.A., 1996. Surgical stress induces
a shift in the type-1/type-2 T-helper cell balance, suggesting down-regulation of cell-mediated and
up-regulation of antibody-mediated immunity. Surgery 119, 316325.
Deslauriers, N., Gaudreau, P., Abribat, T., Renier, G., Peticlerc, D., Brazeau, P., 1991. Dynamics of
growth hormone responsiveness to growth hormone releasing factor in aging rats: peripheral and
central inuences. Neuroendocrinology 53, 439446.
Deuschle, M., Gotthardt, U., Schweiger, U., Weber, B., Korner, A., Schmider, J., Standhardt, H.,
Lammers, C.H., Heuser, I., 1997. With aging in humans the activity of the hypothalamus-pituitary-
adrenal system increases and its diurnal amplitude attens. Life Sci. 61, 22392246.
Deuschle, M., Weber, B., Colla, M., Muller, M., Kniest, A., Heuser, I., 1998. Mineralocorticoid receptor
also modulates basal activity of hypothalamus-pituitary-adrenocortical system in humans.
Neuroendocrinology 68, 355360.
Dickstein, J.B., Hay, J.B., Lue, F.A., Moldofsky, H., 2000. The relationship of lymphocytes in blood and
in lymph to sleep/wake states in sheep. Sleep 23, 185190.
Dimitrov, S., Lange, T., Tieken, S., Fehm, H.L., Born, J., 2004a. Sleep associated regulation of T helper
1/ T helper 2 cytokine balance in humans. Brain Behav. Immun., 18(4), 341348.
Dimitrov, S., Lange, T., Fehm, H.L., Born, J., 2004b. A regulatory role of prolactin, growth hormone and
corticosteroids for human T-cell production of cytokines. Brain Behav. Immun., 18(4), 368374.
Dinges, D.F., Douglas, S.D., Zaugg, L., Campbell, D.E., Mc Mann, J.M., Whitehouse, W.G., 1994.
Leukocytosis and natural killer cell function parallel neurobehavioral fatigue induced by 64 h of sleep
deprivation. J. Clin. Invest. 93, 19301939.
Dodt, C., Theine, K.J., Uthgenannt, D., Born, J., Fehm, H.L., 1994a. Basal secretory activity of the
hypothalamo-pituitary-adrenocortical axis is enhanced in healthy elderly. An assessment during
undisturbed night-time sleep. Eur. J. Endocrinol. 131, 443450.
Dodt, C., Pietrowsky, R., Sewing, A., Zabel, A., Fehm, H.L., Born, J., 1994b. Eects of vasopressin on
event-related potential indicators of cognitive stimulus processing in young and old humans.
J. Gerontol. 49, M183M188.
Dzaja, A., Dalal, M.A., Himmerich, H., Uhr, M., Pollmacher, T., Schuld, A., 2004. Sleep enhances
nocturnal plasma ghrelin levels in healthy subjects. Am. J. Physiol. Endocrinol. Metab., 286(6), E963
E967. Epub 2004 Feb 10.
Edwards, C.K., Arkins, S., Yunger, L.M., Blum, A., Dantzer, R., Kelley, K.W., 1992. The macrophage-
activating properties of growth hormone. Cell Mol. Neurobiol. 12, 499510.
Ehlers, C.L., Kupfer, D.J., 1987. Hypothalamic peptide modulation of EEG sleep in depression: A further
application of the S-process hypothesis. Biol. Psychiatry 22, 513517.
Ehlers, C.L., Kupfer, D.J., 1989. Eects of age on delta and REM sleep parameters. Electroencephalogr.
Clin. Neurophysiol. 72, 118125.
Ehlers, C.L., Reed, T.K., Hendriksen, S.J., 1986. Eects of corticotropin-releasing factor and growth
hormone-releasing factor on sleep and activity in rats. Neuroendocrinology 42, 467474.
Elenkov, I.J., Chrousos, G.P., 2002. Stress hormones, proinammatory and anti-inammatory cytokines,
and autoimmunity. Ann. N. Y. Acad. Sci. 966, 290303.
Engler, D., Redei, E., Kola, I., 1999. The corticotropin-release inhibitory factor hypothesis: a review of
the evidence for the existence of inhibitory as well as stimulatory hypophysiotropic regulation
of adrenocorticotropin secretion and biosynthesis. Endocr. Rev. 20, 460500.
Facchini, A., Mariani, E., Mariani, A.R., Papa, S., Vitale, M., Manzoli, F.A., 1987. Increased number of
circulating Leu 11 (CD 16) large granular lymphocytes and decreased NK activity during human
ageing. Clin. Exp. Immunol. 68, 340347.
Fagiolo, U., Amadori, A., Cozzi, E., Bendo, R., Lama, M., Douglas, A., Palu, G., 1993. Humoral and
cellular immune response to inuenza virus vaccination in aged humans. Aging 5, 451458.
Fauci, A.S., Dale, D.C., Balow, J.E., 1976. Glucocorticosteroid therapy: mechanisms of action and clinical
considerations. Ann. Intern. Med. 84, 304315.
Fehm, H.L., Benkowitsch, R., Kern, W., Fehm-Wolfsdorf, G., Pauschinger, P., Born, J., 1986. Inuences
of corticosteroids, dexamethasone and hydrocortisone on sleep in humans. Neuropsychobiology 16,
198204.
Fehm, H.L., Clausing, J., Kern, W., Pietrowsky, R., Born, J., 1991. Sleep-associated augmentation and
synchronization of luteinizing hormone pulses in adult men. Neuroendocrinology 54, 192195.
Fischer, S., Smolnik, R., Herms, M., Born, J., Fehm, H.L., 2003. Melatonin acutely improves the
neuroendocrine architecture of sleep in blind individuals. J. Clin. Endocrinol. Metab. 88, 53155320.
Follenius, M., Brandenberger, G., Simon, C., Schlienger, J.L., 1988. REM sleep in humans begins during
decreased secretory activity of the anterior pituitary. Sleep 11, 546555.
Follenius, M., Brandenberger, G., Saini, J., 1992. Lack of diurnal rhythm in plasma atrial natriuretic
peptide. Life Sci. 51, 143149.
Forsling, M.L., 1993. Neurohypophysial hormones and circadian rhythm. Ann. N.Y. Acad. Sci. 689,
382395.
Forsling, M.L., 2000. Diurnal rhythms in neurohypophysial function. Exp. Physiol. 85, 179S186S.
Forsling, M.L., Montgomery, H., Halpin, D., Windle, R.J., Treacher, D.F., 1998. Daily patterns of
secretion of neurohypophysial hormones in man: eect of age. Exp. Physiol. 83, 409418.
Franceschi, C., Bonafe, M., Valensin, S., Olivieri, F., De Luca, M., Ottaviani, E., De Benedictis, G., 2000.
Inamm-aging. An evolutionary perspective on immunosenescence. Ann. N. Y. Acad. Sci. 908,
244254.
Frieboes, R.M., Murck, H., Schier, T., Holsboer, F., Steiger, A., 1997. Somatostatin impairs sleep in
elderly human subjects. Neuropsychopharmacol. 16, 339345.
Friess, E., Bardeleben, V., Wiedemann, K., Lauer, C.J., Holsboer, F., 1994. Eects of pulsatile cortisol
infusion on sleep-EEG and nocturnal growth hormone release in healthy men. J. Sleep Res. 3, 7379.
Friess, E., Tagaya, H., Grethe, C., Trachsel, L., Holsboer, F., 2004. Acute cortisol administration
promotes sleep intensity in man. Neuropsychopharmacology 29, 598604.
Gabriel, P., Cakman, I., Rink, L., 2002. Overproduction of monokines by leukocytes after stimulation
with lipopolysaccharide in the elderly. Exp. Gerontol. 37, 235247.
Gallagher, T.F., Yoshida, K., Rowarg, H.D., Fukushima, D.K., Weitzman, E.D., Hellman, L., 1973.
ACTH and cortisol secretory patterns in man. J. Clin. Endocrinol. Metab. 36, 10581068.
Garcia-Maurino, S., Gonzalez-Haba, M.G., Calvo, J.R., Rai-El-Idrissi, M., Sanchez-Margalet, V.,
Goberna, R., Guerrero, J.M., 1997. Melatonin enhances IL-2, IL-6, and IFN-gamma production by
human circulating CD4 cells: a possible nuclear receptor-mediated mechanism involving T helper
type 1 lymphocytes and monocytes. J. Immunol. 159, 574581.
Gillis, S., Kozak, R., Durante, M., Weksler, M.E., 1981. Immunological studies of aging. Decreased
production of and response to T cell growth factor by lymphocytes from aged humans. J. Clin. Invest.
67, 937942.
Ginaldi, L., De Martinis, M., DOstilio, A., Marini, L., Loreto, M.F., Corsi, M.P., 1999a. The immune
system in the elderly: I. Specic humoral immunity. Immunol. Res. 20, 101108.
Ginaldi, L., De Martinis, M., DOstilio, A., Marini, L., Loreto, M.F., Martorelli, V., 1999b. The immune
system in the elderly: II. Specic cellular immunity. Immunol. Res. 20, 109115.
Ginaldi, L., De Martinis, M., DOstilio, A., Marini, L., Loreto, M.F., Quaglino, D., 1999c. The immune
system in the elderly: III. Innate immunity Immunol. Res. 20, 117126.
Ginaldi, L., De Martinis, M., DOstilio, A., Marini, L., Loreto, M.F., Corsi, M.P., Quaglino, D., 2000.
Cell proliferation and apoptosis in the immune system in the elderly. Immunol. Res. 21, 3138.
Goichot, B., Brandenberger, G., Saini, J., Wittersheim, G., Follenius, M., 1992. Nocturnal plasma
thyrotropin variations are related to slow-wave sleep. J. Sleep Res. 1, 186190.
Golomb, J., Kluger, A., de Leon, M.J., Ferris, S.H., Convit, A., Mittelman, M.S., Cohen, J., Rusinek, H.,
De Santi, S., George, A.E., 1994. Hippocampal formation size in normal human aging: a correlate of
delayed secondary memory performance. Learn Mem. 1, 4554.
Gould, E., McEwen, B.S., Tanapat, P., Galea, L.A., Fuchs, E., 1997. Neurogenesis in the dentate gyrus of
the adult tree shrew is regulated by psychosocial stress and NMDA receptor activation. J. Neurosci.
17, 24922498.
Gudewill, S., Pollmacher, T., Vedder, H., Schreiber, W., Fassbender, K., Holsboer, F., 1992. Nocturnal
plasma levels of cytokines in healthy men. Eur. Arch. Psychiatry Clin. Neurosci. 242, 5356.
Guldner, J., Schier, T., Friess, E., Colla, M., Holsboer, F., Steiger, A., 1997. Reduced ecacy of growth
hormone-releasing hormone in modulating sleep endocrine activity in the elderly. Neurobiol. Aging 18,
491495.
Haack, M., Kraus, T., Schuld, A., Dalal, M., Koethe, D., Pollmacher, T., 2002. Diurnal variations of
interleukin-6 plasma levels are confounded by blood drawing procedures. Psychoneuroendocrinology
27, 921931.
Haus, E., 1992. Chronobiology of circulating blood cells and platelets. In: Y. Touitou, E. Haus (Eds.),
Biologic Rhythms in Clinical and Laboratory Medicine Springer, New York, Vol. , pp. 504526.
Haus, E., Smolensky, M.H., 1999. Biologic rhythms in the immune system. Chronobiol. Int. 16, 581622.
Haus, E., Lakatua, D.J., Halberg, F., Halberg, E., Cornelissen, G., Sackett, L.L., Berg, H.G.,
Kawasaki, T., Ueno, M., Uezono, K., Matsuoka, M., Omae, T., 1980. Chrono-biological studies
of plasma prolactin in women in Kyushu, Japan, and Minnesota, USA. J. Clin. Endocrinol.
Metab. 51, 632640.
Hofman, M.A., 2003. Circadian oscillations of neuropeptide expression in the human biological clock.
J. Comp. Physiol. A Neuroethol. Sens. Neural Behav. Physiol. 189, 823831.
Hofman, M.A., Swaab, D.F., 1994. Alterations in circadian rhythmicity of the vasopressin-producing
neurons of the human suprachiasmatic nucleus (SCN) with aging. Brain Res. 651, 134142.
Hohagen, F., Timmer, J., Weyerbrock, A., Fritsch-Montero, R., Ganter, U., Krieger, S., 1993. Cytokine
production during sleep and wakefulness and its relationship to cortisol in healthy humans.
Neuropsychobiology 28, 916.
Holsboer, F., von Bardeleben, U., Steiger, A., 1988. Eects of intravenous corticotropin-releasing
hormone upon sleep-related growth hormone surge and sleep EEG in man. Neuroendocrinology
48, 3238.
Hulstaert, F., Hannet, I., Deneys, V., Munhyeshuli, V., Reichert, T., De Bruyere, M., Strauss, K., 1994.
Age-related changes in human blood lymphocyte subpopulations. II. Varying kinetics of percentage
and absolute count measurements. Clin. Immunol. Immunopathol. 70, 152158.
Iovino, M., Monteleone, P., Steardo, L., 1989. Repetitive growth hormone-releasing hormone
administration restores the attenuated growth hormone (GH) response to GH-releasing hormone
testing in normal aging. J. Clin. Endocrinol. Metab. 69, 910913.
Irwin, M., 2002. Eects of sleep and sleep loss on immunity and cytokines. Brain Behav. Immun. 16,
503512.
Irwin, M., Mascovich, A., Gillin, J.C., Willoughby, R., Pike, J., Smith, T.L., 1994. Partial sleep
deprivation reduces natural killer cell activity in humans. Psychosom. Med. 56, 493498.
Irwin, M., McClintick, J., Costlow, C., Fortner, M., White, J., Gillin, J.C., 1996. Partial night
sleep deprivation reduces natural killer and cellular immune responses in humans. FASEB J. 10,
643653.
Irwin, M., Clark, C., Kennedy, B., Gillin, J., Ziegler, M., 2003. Nocturnal catecholamines and
immune function in insomnia depressed patients, and control subjects. Brain Behav. Immun. 17,
365372.
Jacobson, L., Sapolsky, R., 1991. The role of the hippocampus in feedback regulation of the
hypothalamic-pituitary-adrenocortical axis. Endocr. Rev. 12, 118134.
Jessop, D.S., 1999. Review: Central non-glucocorticoid inhibitors of the hypothalamo-pituitary-adrenal
axis. J. Endocrinol. 160, 169180.
Johnson, A.G., Crawford, G.A., Kelly, D., Nguyen, T.V., Gyory, A.Z., 1994. Arginine vasopressin and
osmolality in the elderly. J. Am. Geriatr. Soc. 42, 399404.
Jones, T.H., Kennedy, R.L., 1993. Cytokines and hypothalamic-pituitary function. Cytokine 5,
531538.
Kato, J., Kitamura, K., Uemura, T., Kuwasako, K., Kita, T., Kangawa, K., Eto, T., 2002. Plasma levels
of adrenomedullin and atrial and brain natriuretic peptides in the general population: their relations
to age and pulse pressure. Hypertens Res. 25, 887892.
Kerkhofs, M., van Cauter, E., van Ondersbergen, A., Caufriez, A., Thorner, M.O., Copinschi, G., 1993.
Sleep-promoting eects of growth hormone-releasing hormone in normal men. Am. J. Physiol. 264,
E594E598.
Kern, W., Halder, R., al Reda, S., Spath-Schwalbe, E., Fehm, H.L., Born, J., 1993. Systemic growth
hormone does not aect human sleep. J. Clin. Endocrinol. Metab. 76, 14281432.
Kern, W., Dodt, C., Born, J., Fehm, H.L., 1996. Changes in cortisol and growth hormone secretion during
nocturnal sleep in the course of aging. J. Gerontol. A Biol. Sci. Med. Sci. 51, M3M9.
Klein, D.C., Moore, R.Y., Reppert, S.M., 1991. Suprachiasmatic nucleus: the minds clock, Oxford
University Press, New York.
Kojima, M., Hosoda, H., Date, Y., Nakazato, M., Matsuo, H., Kangawa, K., 1999. Ghrelin is a growth-
hormone-releasing acylated peptide from stomach. Nature 402, 656660.
Korth, C., Mullington, J., Schreiber, W., Pollmacher, T., 1996. Inuence of endotoxin on daytime sleep
in humans. Infect Immun. 64, 11101115.
Krieger, D.T., Glick, S.M., 1974. Sleep EEG stages and plasma growth hormone in states of endogenous
and exogenous hypercortisolemia or ACTH elevation. J. Clin. Endocrinol. Metab. 11, 409416.
Krueger, J.M., Majde, J.A., 2003. Humoral links between sleep and the immune system: research issues.
Ann. N. Y. Acad. Sci. 992, 920.
Krueger, J.M., Obal, F., Fang, J., 1999. Humoral regulation of physiological sleep: cytokines and GHRH.
J. Sleep Res. 8, 5359.
Kruisbrink, J., Mirmiran, M., Van der Woude, T.P., Boer, G.J., 1987. Eects of enhanced cerebrospinal
uid levels of vasopressin, vasopressin antagonist or vasoactive intestinal polypeptide on circadian
sleep-wake rhythm in the rat. Brain Res. 419, 7686.
Kuhlwein, E., Irwin, M., 2001. Melatonin modulation of lymphocyte proliferation and Th1/Th2 cytokine
expression. J. Neuroimmunol. 117, 5157.
Kupfer, D.J., Bulik, C.M., Jarrett, D.B., 1983. Nighttime plasma cortisol secretion and EEG sleep are
they associated? Psychiatry Res. 10, 191199.
Kupfer, D.J., Jarrett, D.B., Ehlers, C.L., 1991. The eect of GRF on the EEG sleep of normal males. Sleep
14, 8788.
Lange, T., Marshall, L., Spath-Schwalbe, E., Fehm, H.L., Born, J., 2002. Systemic immune param-
eters and sleep after ultra-low dose administration of IL-2 in healthy men. Brain Behav. Immun. 16,
663674.
Lange, T., Perras, B., Fehm, H.L., Born, J., 2003. Sleep enhances the human antibody response to
hepatitis A vaccination. Psychosom. Med. 65, 831835.
Lehtonen, L., Eskola, J., Vainio, O., Lehtonen, A., 1990. Changes in lymphocyte subsets and immune
competence in very advanced age. J. Gerontol. 45, M108M112.
Leproult, R., Colecchia, E.F., LHermite-Baleriaux, M., Van Cauter, E., 2001. Transition from dim to
bright light in the morning induces an immediate elevation of cortisol levels. J. Clin. Endocrinol.
Metab. 86, 151157.
Lieb, K., Ahlvers, K., Dancker, K., Strohbusch, S., Reincke, M., Feige, B., Berger, M., Riemann, D.,
Voderholzer, U., 2002. Eects of the neuropeptide substance P on sleep, mood, and neuroendocrine
measures in healthy young men. Neuropsychopharmacology 27, 10411049.
Linton, P.J., Dorshkind, K., 2004. Age-related changes in lymphocyte development and function. Nature
Immunology 5, 133139.
Lissoni, P., Rovelli, F., Brivio, F., Brivio, O., Fumagalli, L., 1998. Circadian secretions of IL-2., IL-12,
IL-6 and IL-10 in relation to the light/dark rhythm of the pineal hormone melatonin in healthy
humans. Nat. Immun. 16, 15.
Looney, R.J., Hasan, M.S., Con, D., Campbell, D., Falsey, A.R., Kolassa, J., Agosti, J.M.,
Abraham, G.N., Evans, T.G., 2001. Hepatitis B immunization of healthy elderly adults: relationship
between naive CD4 T cells and primary immune response and evaluation of GM-CSF as an
adjuvant. J. Clin. Immunol. 21, 3036.
Lupien, S.J., de Leon, M., De Santi, S., Convit, A., Tarshish, C., Nair, N.P., Thakur, M., McEwen, B.S.,
Hauger, R.L., Meaney, M.J., 1998. Cortisol levels during human aging predict hippocampal atrophy
and memory decits. Nat. Neurosci. 1, 6973.
Lupien, S.J., Nair, N.P., Briere, S., Maheu, F., Tu, M.T., Lemay, M., McEwen, B.S., Meaney, M.J., 1999.
Increased cortisol levels and impaired cognition in human aging: implication for depression and
dementia in later life. Rev. Neurosci. 10, 117139.
Maes, M., Lin, A., Kenis, G., Egyed, B., Bosmans, E., 2000. The eects of noradrenaline and alpha-2
adrenoceptor agent on the production of monocytic products. Psychiatry Res. 96, 245253.
Marshall, G.D., Agrarwal, S.K., 2000. Stress, immune regulation, and immunity: applications for asthma.
Allergy Asthma Proc. 21, 241246.
Marshall, L., Born, J., 2002. Brain-immune interactions in sleep. In: Neurobiology of the immune system.
Int. Rev. Neurobiol. 52, 93131.
Marshall, L., Molle, M., Boschen, G., Steiger, A., Fehm, H.L., Born, J., 1996. Greater ecacy of episodic
than continuous growth hormone-releasing hormone (GHRH) administration in promoting slow-wave
sleep (SWS). J. Clin. Endocrinol. Metab. 81, 10091013.
Marshall, L., Derad, I., Strasburger, C.J., Fehm, H.L., Born, J., 1999. A determinant factor in the ecacy
of GHRH administration in promoting sleep: high peak concentration versus recurrent increasing
slopes. Psychoneuroendocrinology 24, 363370.
Matera, L., Mori, M., Geuna, M., Buttiglieri, S., Palestro, G., 2000. Prolactin in autoimmunity and
antitumor defence. J. Neuroimmunol. 109, 4755.
McClintick, J., Costlow, C., Fortner, M., White, J., Gillin, C., Irwin, M., 1994. Partial sleep deprivation
reduces natural killer cell activity, interleukin-2 production and lymphokine-activated killer cell
activity. Proceedings of the 25th Congress of the International Society of Psychoneuroendocrinology.
Seattle, WA, 160 (Abstract).
McEwen, B.S., 1999. Stress and hippocampal plasticity. Annu. Rev. Neurosci. 22, 105122.
McEwen, B.S., 2000. The neurobiology of stress: from serendipity to clinical relevance(1). Brain Res. 886,
172189.
McGivern, R.F., Rittenhouse, P., Aird, F., Van de Kar, L.D., Redei, E., 1997. Inhibition of stress-induced
neuroendocrine and behavioral responses in the rat by prepro-thyrotropin-releasing hormone178-199.
J. Neurosci. 17, 48864894.
Mellado, M., Llorente, M., Rodriguez-Frade, J.M., Lucas, P., Martinez, C., del Real, G., 1998. HIV-1
envelope protein gp120 triggers a Th2 response in mice that shifts to Th1 in the presence of human
growth hormone. Vaccine 16, 11111115.
Mendelson, W.B., Slater, S., Gold, P., Gillin, J.C., 1980. The eect of growth hormone administration
on human sleep: a dose-response study. Biol. Psychiatry 15, 613618.
Mendelson, W.B., James, S.P., Garnett, D., Sack, D.A., Rosenthal, N.E., 1986. A psychophysiological
study of insomnia. Psychiatry Res. 19, 267284.
Miller, A.H., Spencer, R.L., Pearce, B.D., Pisell, T.L., Tanapat, P., Leung, J.J., Dhabhar, F.S.,
McEwen, B.S., Biron, C.A., 1996. Eects of viral infection on corticosterone secretion and
glucocorticoid receptor binding in immune tissues. Psychoneuroendocrinology 22, 455474.
Miyawaki, T., Taga, K., Nagaoki, T., Seki, H., Suzuki, Y., Taniguchi, N., 1984. Circadian changes of
T lymphocyte subsets in human peripheral blood. Clin. Exp. Immunol. 55, 618622.
Moldofsky, H., 1995. Sleep, neuroimmune and neuroendocrine functions in bromyalgia and chronic
fatigue syndrome. Adv. Neuroimmunol. 5, 3956.
Moore, R.Y., Speh, J.C., Leak, R.K., 2002. Suprachiasmatic nucleus organization. Cell Tissue Res. 309,
8998.
Muller, E.E., Locatelli, V., Cocchi, D., 1999. Neuroendocrine control of growth hormone secretion.
Physiol. Rev. 79, 511607.
Mullington, J., Korth, C., Hermann, D.M., Orth, A., Galanos, C., Holsboer, F., 2000. Dose-dependent
eects of endotoxin on human sleep. Am. J. Physiol. Regul. Integr. Comp. Physiol. 278, R947R955.
Murasko, D.M., Goonewardene, I.M., 1990. T-cell function in aging: mechanisms of decline. Annu. Rev.
Gerontol. Geriatr. 10, 7196.
Murck, H., Frieboes, R.M., Schier, T., Steiger, A., 1997. Longtime administration of growth hormone-
releasing hormone (GHRH) does not restore the reduced eciency of GHRH on sleep endocrine
activity in 2 old-aged subjects a preliminary study. Pharmacopsychiatry 30, 122124.
Nadal, M., Wikstrom, L., Ruthstrom, L., 1994. Secretory pattern of vasopressin in plasma and
cerebrospinal uid of patients with dementia and of two control groups. Eur. J. Endocrinol. 130,
346349.
Nagel, J.E., Chopra, R.K., Powers, D.C., Adler, W.H., 1989. Eect of age on the human high anity
interleukin 2 receptor of phytohaemagglutinin stimulated peripheral blood lymphocytes. Clin. Exp.
Immunol. 75, 286291.
Nebes, R.D., Reynolds, C.F., III, Horn, L.C., 1984. The eect of vasopressin on memory in the healthy
elderly. Psychiatry Res. 11, 4959.
ONeill, P.A., McLean, K.A., 1992. Water homeostasis and ageing. Med. Lab. Sci. 49, 291298.
Obal, F., Payne, L., Opp, M., Alfoldi, P., Kapas, L., Krueger, J.M., 1992. Growth hormone-releasing
hormone antibodies suppress sleep and prevent enhancement of sleep after sleep deprivation. Am. J.
Physiol. 263, R1078R1085.
Pandi-Perumal, S.R., Seils, L.K., Kayumov, L., Ralph, M.R., Lowe, A., Moller, H., Swaab, D.F., 2002.
Senescence, sleep, and circadian rhythms. Ageing Res. Rev. 1, 559604.
Palm, S., Postler, E., Hinrichsen, H., Maier, H., Zabel, P., Kirch, W., 1996. Twenty-four-hour analysis of
lymphocyte subpopulations and cytokines in healthy subjects. Chronobiol. Int. 13, 423434.
Pavlides, C., Kimura, A., Magarinos, A.M., McEwen, B.S., 1995. Hippocampal homosynaptic long-term
depression/depotentiation induced by adrenal steroids. Neuroscience 68, 379385.
Perras, B., Molle, M., Born, J., Fehm, H.L., 1996. Sleep and signs of attention during 3 months of
intranasal vasopressin: a pilot study in two elderly subjects. Peptides 17, 12531255.
Perras, B., Marshall, L., Kohler, G., Born, J., Fehm, H.L., 1999a. Sleep and endocrine changes after
intranasal administration of growth hormone-releasing hormone in young and aged humans.
Psychoneuroendocrinology 24, 743757.
Perras, B., Pannenborg, H., Marshall, L., Pietrowsky, R., Born, J., Fehm, H.L., 1999b. Benecial
treatment of age-related sleep disturbances with prolonged intranasal vasopressin. J. Clin.
Psychopharmacol. 19, 2836.
Perras, B., Wagner, U., Born, J., Fehm, H.L., 2003. Improvement of sleep and pituitary-
adrenal inhibition after subchronic intranasal vasopressin treatment in elderly humans. J. Clin.
Psychopharmacol. 23, 3544.
Petrovsky, N., Harrison, L.C., 1997. Diurnal rhythmicity of human cytokine production: a dynamic
disequilibrium in T helper cell type 1/T helper cell type 2 balance? J. Immunol. 158, 51635168.
Petrovsky, N., Harrison, L.C., 1998. The chronobiology of human cytokine production. Int. Rev.
Immunol. 16, 635649.
Petrovsky, N., McNair, P., Harrison, L.C., 1994. Circadian rhythmicity of interferon-gamma production
in antigen-stimulated whole blood. Chronobiologia 21, 293300.
Petrovsky, N., McNair, P., Harrison, L.C., 1998. Diurnal rhythms of pro-inammatory cytokines:
regulation by plasma cortisol and therapeutic implications. Cytokine 10, 307312.
Pietrowsky, R., Meyrer, R., Kern, W., Born, J., Fehm, H.L., 1994. Eects of diurnal sleep on secretion
of cortisol, luteinizing hormone, and growth hormone in man. J. Clin. Endocrinol. Metab. 78,
683687.
Pisciotta, A.V., Westring, D.W., DePrey, C., Walsh, B., 1967. Mitogenic eect of phytohaemagglutinin at
dierent ages. Nature 215, 193194.
Pollmacher, T., Schreiber, W., Gudewill, S., Vedder, H., Fassbender, K., Wiedemann, K., 1993. Inuence
of endotoxin on nocturnal sleep in humans. Am. J. Physiol. 264, R1077R1083.
Pollmacher, T., Mullington, J., Korth, C., Hinze-Selch, D., 1995. Inuence of host defense activation
on sleep in humans. Adv. Neuroimmunol. 5, 155169.
Posnett, D.N., Sinha, R., Kabak, S., Russo, C., 1994. Clonal populations of T cells in normal elderly
humans: the T cell equivalent to benign monoclonal gammapathy. J. Exp. Med. 179, 609618.
Postel-Vinay, M.C., de Mello Coelho, V., Gagnerault, M.C., Dardenne, M., 1997. Growth hormone
stimulates the proliferation of activated mouse T lymphocytes. Endocrinology 138, 18161820.
Prinz, P.N., Vitiello, M.V., Raskind, M.A., Thorpy, M.J., 1990. Geriatrics: sleep disorders and aging.
N. Engl. J. Med. 323, 520526.
Prinz, P.N., Moe, K.E., Dulberg, E.M., Larsen, L.H., Vitiello, M.V., Toivola, B., Merriam, G.R., 1995.
Higher plasma IGF-1 levels are associated with increased delta sleep in healthy older men. J. Gerontol.
A Biol. Sci. Med. Sci. 50, M222M226.
Pruessner, M., Hellhammer, D.H., Pruessner, J.C., Lupien, S.J., 2003. Self-reported depressive symptoms
and stress levels in healthy young men: associations with the cortisol response to awakening.
Psychosom. Med. 65, 9299.
Quabbe, H.J., Schilling, E., Helge, H., 1966. Pattern of growth hormone secretion during a 24-h fast in
normal adults. J. Clin. Endocrinol. Metab. 26, 11731177.
Rea, I.M., Alexander, H.D., Crockard, A.D., Morris, T.C., Rae, I.M., 1996. CD4 lymphopenia in very
elderly people. Lancet. 347, 328329.
Remarque, E.J., 1999. Inuenza vaccination in elderly people. Exp. Gerontol. 34, 445452.
Renegar, K.B., Floyd, R.A., Krueger, J.M., 1998. Eects of short-term sleep deprivation on murine
immunity to inuenza virus in young adult and senescent mice. Sleep 21, 241248.
Redei, E., Hilderbrand, H., Aird, F., 1995. Corticotropin release-inhibiting factor is preprothyrotropin-
releasing hormone-(178-199). Endocrinology 136, 35573563.
Redei, E., Rittenhouse, P.A., Revskoy, S., McGivern, R.F., Aird, F., 1998. A novel endogenous
corticotropin release inhibiting factor. Ann. N. Y. Acad. Sci. 840, 456469.
Redwine, L., Hauger, R.L., Gillin, J.C., Irwin, M., 2000. Eects of sleep and sleep deprivation on
interleukin-6, growth hormone, cortisol, and melatonin levels in humans. J. Clin. Endocrinol. Metab.
85, 35973603.
Reul, J.M.H.M., Rothuizen, J., de Kloet, E.R., 1991. Age-related changes in the dog hypothalamic-
pituitary-adrenocortical system: Neuroendocrine activity and corticosteroid receptors. J. Steroid
Biochem. Mol. Biol. 40, 6369.
Rink, L., Seyfarth, M., 1997. Characteristics of immunologic test values in the elderly. Z. Gerontol.
Geriatr. 30, 220225.
Rink, L., Cakman, I., Kirchner, H., 1998. Altered cytokine production in the elderly. Mech. Ageing Dev.
102, 199209.
Ritchie, A.W., Oswald, I., Micklem, H.S., Boyd, J.E., Elton, R.A., Jazwinska, E., 1983. Circadian
variation of lymphocyte subpopulations: a study with monoclonal antibodies. Br. Med. J. (Clin. Res.
Ed.) 286, 17731775.
Roehrs, T., Zorick, F., Sicklesteel, J., Wittig, R., Roth, T., 1983. Age-related sleep-wake disorders at a
sleep disorder center. J. Am. Geriatr. Soc. 31, 364370.
Rosmond, R., 2003. Stress induced disturbances of the HPA axis: a pathway to Type 2 diabetes? Med. Sci.
Monit. 9, RA35RA39.
Rothuizen, J., Reul, J.M.H.M., van Sluijs, F.J., Mol, J.A., Rijnbek, A., de Kloet, E.R., 1993. Increased
neuroendocrine reactivity and decreased brain mineralocorticoid receptor-binding capacity in aged
dogs. Endocrinology 132, 161168.
Rubin, R.T., Poland, R.E., Gouin, P.R., Tower, B.B., 1978. Secretion of hormones inuencing water and
electrolyte balance (antidiuretic hormone, aldosterone, prolactin) during sleep in normal adult men.
Psychosom. Med. 40, 4459.
Russell-Aulet, M., Jae, C.A., Demott-Friberg, R., Barkan, A.L., 1999. In vivo semiquantication of
hypothalamic growth hormone-releasing hormone (GHRH) output in humans: evidence for relative
GHRH deciency in aging. J. Clin. Endocrinol. Metab. 84, 34903497.
Sakami, S., Ishikawa, T., Kawakami, N., Haratani, T., Fukui, A., Kobayashi, F., Fujita, O., Araki, S.,
Kawamura, N., 2003. Coemergence of insomnia and a shift in the Th1/Th2 balance toward Th2
dominance. Neuroimmunomodulation 10, 337343.
Sansoni, P., Cossarizza, A., Brianti, V., et al., 1993. Lymphocyte subsets and NK-cell activity in healthy
old people and centenarians. Blood 82, 27672773.
Sapolsky, R.M., 1999. Glucocorticoids, stress, and their adverse neurological eects: relevance to aging.
Exp. Gerontol. 34, 721732.
Sapolsky, R.M., Krey, L.C., McEwen, B.S., 1986. The neuroendocrinology of stress and aging: the
glucocorticoid cascade hypothesis. Endocr. Rev. 7, 284301.
Schuld, A., Mullington, J., Hermann, D., Hinze-Selch, D., Fenzel, T., Holsboer, F., 1999. Eects
of granulocyte colony-stimulating factor on night sleep in humans. Am. J. Physiol. 276,
R1149R1155.
Schielholz, T., Lancel, M., 2001. Sleep changes induced by lipopolysaccaride in the rat are inuenced by
age. Am. J. Physiol. Regul. Integr. Comp. Physiol. 280, R398403.
Schwab, R., Szabo, P., Manavalan, J.S., Weksler, M.E., Posnett, D.N., Pannetier, C., Kourilsky, P., Even, J.,
1997. Expanded CD4 and CD8 T cell clones in elderly humans. J. Immunol. 158, 44934499.
Seeman, T.E., Robbins, R.J., 1994. Aging and hypothalamic-pituitary-adrenal response to challenge in
humans. Endocr. Rev. 15, 233260.
Seifert, H., Perrin, M., Rivier, J., Vale, W., 1985. Growth hormone-releasing factor binding sites in rat
anterior pituitary membrane homogenates: Modulation by glucocorticoids. Endocrinology 117,
424429.
Shearer, G.M., 1997. Th1/Th2 changes in aging. Mech. Ageing Dev. 94, 15.
Shearer, W.T., Reuben, J.M., Mullington, J.M., Price, N.J., Lee, B.N., Smith, E.O., Szuba, M.P.,
Dinges, D.F., 2001. Soluble TNF-alpha receptor 1 and IL-6 plasma levels in humans subjected to the
sleep deprivation model of spaceight. J. Allergy Clin. Immunol. 107, 1920.
Sheline, Y.I., Wang, P.W., Gado, M.H., Csernansky, J.G., Vannier, M.W., 1996. Hippocampal atrophy in
recurrent major depression. Proc. Natl. Acad. Sci. USA 93, 39083913.
Shibasaki, T., Shizume, K., Nakahara, M., Masuda, A., Jibiki, K., Demura, H., Wakabayashi, T.,
Ling, N., 1984. Age related changes in plasma growth hormone response to growth hormone releasing
factor in man. J. Clin. Endocrinol. Metab. 58, 212214.
Sindermann, J., Kruse, A., Frercks, H.J., Schutz, R.M., Kirchner, H., 1993. Investigations of the
lymphokine system in elderly individuals. Mech. Ageing Dev. 70, 149159.
Skene, D.J., Swaab, D.F., 2003. Melatonin rhythmicity: eect of age and Alzheimers disease. Exp.
Gerontol. 38, 199206.
Snel, J., Taylor, J., Wegman, M., 1987. Does DGAVP inuence memory, attention and mood in young
healthy men? Psychopharmacology (Berl) 92, 224228.
Solana, R., Mariani, E., 2000. NK and NK/T cells in human senescence. Vaccine 18, 16131620.
Sothern, R.B., Roitman-Johnson, B., 2001. Biological rhythms and immune function. In: R. Ader,
D.L. Felten, N. Cohen (Eds.), Psychoneuroimmunology. Academic Press, London, Vol. 1, pp. 445479.
Sothern, R.B., Roitman-Johnson, B., Kanabrocki, E.L., Yager, J.G., Roodell, M.M., Weatherbee, J.A.,
1995. Circadian characteristics of circulating interleukin-6 in men. J. Allergy Clin. Immunol. 95,
10291035.
Sousa, N., Lukoyanov, N.V., Madeira, M.D., Almeida, O.F., Paula-Barbosa, M.M., 2000.
Reorganization of the morphology of hippocampal neurites and synapses after stress-induced damage
correlates with behavioral improvement. Neuroscience 97, 253266.
Spath-Schwalbe, E., Born, J., Fehm, H.L., Pfeier, E.F., 1987. Combined corticotropin-releasing
hormone-vasopressin test: A new test for the evaluation of the pituitary adrenal system. Horm. Metab.
Res. 19, 665666.
Spath-Schwalbe, E., Goerje, M., Kern, W., Born, J., Fehm, H.L., 1991. Sleep disruption alters nocturnal
ACTH and cortisol secretory patterns. Biol. Psychiatry 29, 575584.
Spath-Schwalbe, E., Uthgenannt, D., Voget, G., Kern, W., Born, J., Fehm, H.L., 1993. Corticotropin-
releasing hormone-induced adrenocorticotropin and cortisol secretion depends on sleep and
wakefulness. J. Clin. Endocrinol. Metab. 77, 11701173.
Spath-Schwalbe, E., Hundenborn, C., Kern, W., Fehm, H.L., Born, J., 1995. Nocturnal wakefulness
inhibits growth hormone (GH)-releasing hormone- induced GH secretion. J. Clin. Endocrinol. Metab.
80, 214219.
Spath-Schwalbe, E., Schrezenmeier, H., Bornstein, S., Burger, K., Porzsolt, F., Born, J., 1996. Endocrine
eects of recombinant IL-6 in man. Neuroendocrinology 63, 237243.
Spath-Schwalbe, E., Hansen, K., Schmidt, F., Schrezenmeier, H., Marshall, L., Burger, K., 1998. Acute
eects of recombinant human interleukin-6 on endocrine and central nervous sleep functions in healthy
men. J. Clin. Endocrinol. Metab. 83, 15731579.
Spath-Schwalbe, E., Lange, T., Perras, B., Fehm, H.L., Born, J., 2000. Interferon-alpha acutely impairs
sleep in healthy humans. Cytokine 12, 518521.
Spiegel, K., Luthringer, R., Follenius, M., Schlatenbrand, N., Macher, J.P., Muzet, A., Brandenberger, G.,
1995. Temporal relationship between prolactin secretion and slow-wave electroencephalographic
activity during sleep. Sleep 18, 543548.
Spiegel, K., Sheridan, J.F., Van Cauter, E., 2002. Eect of sleep deprivation on response to immunization.
JAMA 288, 14711472.
Starkman, M.N., Giordani, B., Gebarski, S.S., Berent, S., Schork, M.A., Schteingart, D.E., 1999.
Decrease in cortisol reverses human hippocampal atrophy following treatment of Cushings disease.
Biol. Psychiatry 46, 15951602.
Steiger, A., 2002. Sleep and the hypothalamo-pituitary-adrenocortical system. Sleep Med. Rev. 6,
125138.
Steiger, A., 2003. Sleep and endocrinology. J. Intern. Med. 254, 1322.
Steiger, A., Guldner, J., Hemmeter, U., Rothe, B., Wiedemann, K., Holsboer, F., 1992. Eects of growth
hormone-releasing hormone and somatostatin on sleep EEG and nocturnal hormone secretion in
male controls. Neuroendocrinology 56, 566573.
Sugimoto, T., Saito, M., Mochizuki, S., Watanabe, Y., Hashimoto, S., Kawashima, H., 1994. Molecular
cloning and functional expression of a cDNA encoding the human V1b vasopressin receptor. J. Biol.
Chem. 269, 2708827092.
Sutherland, E.R., Martin, R.J., Ellison, M.C., Kraft, M., 2002. Immunomodulatory eects of melatonin
in asthma. Am. J. Respir. Crit. Care Med. 166, 10551061.
Swaab, D.F., 1995. Ageing of the human hypothalamus. Horm. Res. 43, 811.
Takagi, K., Suzuki, F., Barrow, R.W., Wolf, S.E., Herndon, D.N., 1998. Recombinant human growth
hormone modulates Th1 and Th2 cytokine response in burned mice. Ann. Surg. 228, 106111.
Takahashi, Y., Kipnis, D.M., Daughaday, W.H., 1968. Growth hormone secretion during sleep. J. Clin.
Invest. 47, 20792090.
Timsit-Berthier, M., Mantanus, H., Devos, J.E., Spiegel, R., 1982. Action of lysine-vasopressin on human
electroencephalographic activity. Night sleep pattern, auditory evoked potential, contingent negative
variation. Neuropsychobiology 8, 248258.
Toppila, J., Alanko, L., Asikainen, M., Tobler, I., Stenberg, D., Porkka-Heiskanen, T., 1997. Sleep
deprivation increases somatostatin and growth hormone-releasing hormone messenger RNA in the
rat hypothalamus. J. Sleep Res. 6, 171178.
Touitou, Y., 1995. Eects of ageing on endocrine and neuroendocrine rhythms in humans. Horm. Res.
43, 1219.
Touitou, Y., Haus, E., 2000. Alterations with aging of the endocrine and neuroendocrine circadian system
in humans. Chronobiol. Int. 17, 369390.
Urban, I., Wied, D., 1978. Neuropeptides: eects on paradoxical sleep and theta rhythm in rats.
Pharmacol. Biochem. Behav. 8, 5159.
Uthgenannt, D., Schoolmann, D., Pietrowsky, R., Fehm, H.L., Born, J., 1995. Eects of sleep on the
production of cytokines in humans. Psychosom. Med. 57, 97104.
Van Cauter, E., Copinschi, G., 2000a. Interrelationships between growth hormone and sleep. Growth
Horm. IGF Res. 10 Suppl B, S57S62.
Van Cauter, E., Leproult, R., Plat, L., 2000b. Age-related changes in slow wave sleep and REM sleep
and relationship with growth hormone and cortisol levels in healthy men. JAMA 284, 861868.
Van Coevorden, A., Mockel, J., Laurent, E., Kerkhofs, M., LHermite-Baleriaux, M., Decoster, C.,
Neve, P., Van Cauter, E., 1991. Neuroendocrine rhythms and sleep in aging men. Am. J. Physiol. 260,
E651E661.
Van Ree, J.M., Hijman, R., Jolles, J., De Wied, D., 1985. Vasopressin and related peptides: animal and
human studies. Prog. Neuropsychopharmacol. Biol. Psychiatry 9, 551559.
Veldhuis, H.D., de Kloet, E.R., 1982. Vasopressin-related peptides increase the hippocampal
corticosterone receptor capacity of diabetes insipidus (Brattleboro) rats. Endocrinology 110, 153157.
Vgontzas, A.N., Papanicolaou, D.A., Bixler, E.O., Lotsikas, A., Zachman, K., Kales, A., 1999. Circadian
interleukin-6 secretion and quantity and depth of sleep. J. Clin. Endocrinol. Metab. 84, 26032607.
Vgontzas, A.N., Bixler, E.O., Lin, H.M., Prolo, P., Mastorakos, G., Vela-Bueno, A., Kales, A.,
Chrousos, G.P., 2001. Chronic insomnia is associated with nyctohemeral activation of the
hypothalamic-pituitary-adrenal axis: clinical implications. J. Clin. Endocrinol. Metab. 86, 37873794.
Weibel, L., Follenius, M., Spiegel, K., Ehrhart, J., Brandenberger, G., 1995. Comparative eect of night
and daytime sleep on the 24-h cortisol secretory prole. Sleep 18, 549556.
Weigent, D.A., Blalock, J.E., 1995. Associations between the neuroendocrine and immune systems.
J. Leukoc. Biol. 58, 137150.
Weikel, J.C., Wichniak, A., Ising, M., Brunner, H., Friess, E., Held, K., Mathias, S., Schmid, D.A.,
Uhr, M., Steiger, A., 2003. Ghrelin promotes slow-wave sleep in humans. Am. J. Physiol. Endocrinol.
Metab. 284, E407E415.
Weitzman, E.D., Fukushima, D., Nogeire, C., Rowarg, H., Gallagher, T.F., Hellman, L., 1971. Twenty-
four hour pattern of the episodic secretion of cortisol in normal subjects. J. Clin. Endocrinol. Metab.
33, 1422.
Wu, R.H., Thorpy, M.J., 1988. Eect of growth hormone treatment on sleep EEG in growth hormone-
decient children. Sleep 11, 425429.
Young, E.A., Lopez, J.F., Murphy-Weinberg, V., Watson, S.J., Akil, H., 1998. The role of
mineralocorticoid receptors in hypothalamic-pituitary-adrenal axis regulation in humans. J. Clin.
Endocrinol. Metab. 83, 33393345.
Yu-Lee, L.Y., 2002. Prolactin modulation of immune and inammatory responses. Recent Prog. Hom.
Res. 57, 435455.
Zabel, P., Horst, H.J., Kreiker, C., Schlaak, M., 1990. Circadian rhythm of interleukin-1 production of
monocytes and the inuence of endogenous and exogenous glucocorticoids in man. Klin. Wochenschr.
68, 12171221.
Zabel, P., Linnemann, K., Schlaak, M., 1993. Circadian rhythm in cytokines. Immun. Infekt. 21, 3840.
Zhang, J., Obal, F., Fang, J., Collins, B.J., Krueger, J.M., 1996. Non-rapid eye movement sleep is
suppressed in transgenic mice with a deciency in the somatotropic system. Neurosci. Lett. 220,
97100.
Advances in
Cell Aging and
Gerontology
Neurotrophic factors and sleep

Mark P. Mattson
Correspondence address: Laboratory of Neurosciences, National Institute on Aging Gerontology
Research Center 4F01, 5600 Nathan Shock Drive, Baltimore, MD 21224, USA.
Tel: 1-410-558-8463; fax: 1-410-558-8465.
E-mail address: mattsonm@grc.nia.nih.gov
Contents
1. Introduction
2. Circadian variations in neurotrophic factor expression
3. Effects of phase-shifts and sleep deprivation on neurotrophic factor signaling
4. Functions of neurotrophic factors in regulating sleep
5. Depression, neurotrophic factors, and sleep
6. Neurotrophic factors and normal aging
7. Neurotrophic factors and neurodegenerative disorders
7.1. Alzheimers disease
7.2. Parkinsons disease
7.3. Huntingtons disease
8. Conclusions
Abstract. Well known for their ability to promote the survival and outgrowth of neurons during
development of the brain, neurotrophic factors also serve multiple functions in the adult brain.
For example, the modulation of neurotransmitter signaling and neurogenesis by neurotrophic
factors may play a role in learning and memory. Important roles for neurotrophic factors in sleep
regulation and sleep disorders are suggested by circadian variations in their expression and
alterations in their expression in animal models of phase shift and sleep deprivation. Brain-derived
neurotrophic factor (BDNF) may play particularly important roles in sleep regulation and sleep
disorders because of its reciprocal interactions with serotonin, and its involvement in depression and
anxiety disorders. Perturbed neurotrophic factor signaling occurs in neurodegenerative disorders in
which sleep disturbance is prominent including Alzheimers, Parkinsons, and Huntingtons
diseases. A better understanding of the roles of neurotrophic factor signaling in sleep regulation
may lead to novel approaches for preventing and treating sleep disorders.

DOI: 10.1016/S1566-3124(05)17006-0
156 M. P. Mattson
1. Introduction
During the past several decades, an increasing number of proteins that promote
the survival and growth of neurons have been identied. These include growth
factors originally identied because of their actions on non-neuronal cells such as
basic broblast growth factor (bFGF), epidermal growth factor (EGF) and
transforming growth factor-beta (TGF) (Plata-Salaman, 1991; Abe and Saito, 2001;
Unsicker and Krieglstein, 2002). Beginning with the discovery of nerve growth factor
(NGF), a novel family of neurotrophins was identied which includes brain-
derived neurotrophic factor (BDNF), neurotrophin-3 (NT-3) and neurotrophin-4/5
(NT-4/5) (Thoenen, 2000; Dechant and Neumann, 2002). Each of these neurotrophic
factors and their receptors is expressed in one or more populations of neurons
involved in the regulation of sleep, including those in structures such as the
hypothalamus, brainstem, hippocampus and various regions of the cerebral cortex
(Hobson and Pace-Schott, 2002; Huang and Reichardt, 2003).
The expression of many neurotrophic factors is stimulated by activity in neuronal
circuits and by cellular stress (Hughes et al., 1999). For example, the production of
BDNF is increased in response to neurotransmitters that stimulate cyclic AMP
production and/or calcium inux including serotnin and glutamate (Jiang et al.,
2003; Russo-Neustadt et al., 2003). Neurotrophic factors exert their actions by
binding to cell surface receptors which, in turn, initiate kinase cascades resulting in
the activation of transcription factors. Some of the gene targets for neurotrophic
factors include those encoding enzymes involved in neurotransmitter synthesis or
transport, cell survival-promoting proteins, and proteins the regulate neurite
outgrowth and synaptic plasticity (Lu and Chow, 1999). The present article reviews
the evidence that neurotrophic factors play roles in regulating various aspects of
sleep, and discuss the possible roles of perturbed neurotrophic factor signaling in
sleep disorders.
2. Circadian variations in neurotrophic factor expression
The expression of several dierent neurotrophic factors changes during the light
dark cycle in brain regions involved in sleep regulation. The BDNF mRNA levels in
the suprachiasmatic nucleus (SCN) of rats maintained in constant darkness were
high during the early subjective day (Liang et al., 1998). Levels of BDNF and its
receptor trkB in the hippocampus were reported to oscillate during the lightdark
cycle in rats (Bova et al., 1998). Cirelli and Tononi (2000) reported that the amounts
of BDNF, phosphorylated CRE-binding protein (a transcriptional regulator of
BDNF expression) and Arc (an immediate early gene target of BDNF) are high
during waking and low during sleep in rats; and that the higher levels of these
proteins during waking depends on noradrenergic signaling. Levels of BDNF are
elevated during the light phase of the cycle in the retina, superior colliculus, and
primary visual cortex, but are decreased during the light phase in the hippocampus
and cerebellum (Pollock et al., 2001). Signicant circadian variations in trkB
(the high anity receptor for BDNF) expression were observed in dentate gyrus,
Neurotrophins and Sleep 157
entorhinal cortex, and the CA3 and hilar regions of the hippocampus; for each
region the highest expression occurred during the rst half of the dark period (Dolci
et al., 2003).
3. Effects of phase-shifts and sleep deprivation on neurotrophic factor signaling
Forced changes in the lightdark cycle can change the levels of BDNF in brain
region-specic manner. For example an 8 h advance in the lightdark cycle in rats
resulted in increased levels of BDNF protein in the hippocampus, but not in the
cerebellum and brainstem (Sei et al., 2003). In another study rats were subjected to
either sleep deprivation for 8 h or a mild increase in ambient temperature (which
induces sleep), and levels of BDNF mRNA were measured in the cerebral cortex and
hippocampus (Taishi et al., 2001). Levels of BDNF mRNA were increased in the
cerebral cortex (but not in the hippocampus) of sleep-deprived rats, but were
unchanged by the increased temperature environment.
Infusion of NGF into the suprachiasmatic nucleus resulted in phase advances in
hamsters (Bina and Rusak, 1996). The REM sleep deprivation resulted in decreased
levels of BDNF protein levels in the cerebellum and brainstem, and decreased levels
of NGF protein in the hippocampus (Sei et al., 2000). Transforming growth factor-
alpha (TGFalpha) is expressed rhythmically in the suprachiasmatic nucleus, and
infusion of TGFalpha into the third ventricle suppresses locomotor activity and
disrupts sleepwake cycles (Kramer et al., 2001). Those actions of TGFalpha are
mediated by EGF receptors as indicated by increased daytime locomotor activity in
mice with a hypomorphic EGF receptor mutation. Changes in neurotrophic factor
levels in association with sleepwake cycles, particular stages of sleep, and sleep
deprivation suggest roles for these factors in sleep regulation.
4. Functions of neurotrophic factors in regulating sleep
The ability of neurotrophic factors to alter aspects of sleep has been documented
in several studies. Intracerebroventricular infusion of BDNF increased the amount
of time spent in non-REM sleep in adult rats and rabbits (Kushikata et al., 1999).
The latter investigators also found that intracerebroventricular infusion of NT-3 and
NT-4 increased the time spent in non-REM sleep when infused at dark onset
(Kushikata et al., 2003). In other studies, manipulations of BDNF signaling have
been shown to modify the circadian regulation of the SCN. For example, when
BDNF was infused into the SCN of rats their free-running rhythm in constant
darkness exhibited large phase advances in response to light, which contrasted
with saline-infused control rats whose rhythm did not exhibit a response to light
(Liang et al., 2000). Also, BDNF did not aect either the phase-delaying or phase-
advancing eects of light on the activity rhythm during the day. In BDNF
heterozygous mice with reduced BDNF levels there was a decrease in the amplitude
of light-induced phase shifts during the light period (Liang et al., 2000). The latter
study further showed that an inhibitor of trkB can inhibit both the phase-advancing
and phase-delaying eects of light during the light phase.
158 M. P. Mattson
Neurons in the SCN contain considerable amounts of the low anity

neurotrophin receptor p75. However, using a toxin (192 IgG-aporin) that selectively
kills neurons expressing p75, it was shown that destruction of these neurons in
the SCN does not aect photic entrainment or free-running rhythms (Beaule and
Amir, 2002). Infusion of NT-3 into the nucleus reticularis pontis oralis (NRPO)
induced long duration episodes (up to 30 min) of REM-like brain wave activity
in cats (Yamuy et al., 2002). Additional data in the later study suggested that
cholinergic neurons in the latero-dorsal and pedunculo-pontine tegmental nuclei,
which project to the NRPO may be a source of endogenous NT-3 that regulates
sleep. Intracerebroventricular infusion of EGF suppressed REM sleep and increased
brain temperature in rabbits (Kushikata et al., 1998), consistent with a role EGF in
sleep regulation. Another neurotrophic factor which may regulate sleep is glial cell
line-derived neurotrophic factor (GDNF). When GDNF was infused into the
cerebroventricular uid of adult rats and rabbits, the amount of time the animals
spent in non-REM sleep was increased (Kushikata et al., 2001). While such ndings
demonstrate the ability of several dierent neurotrophic factors to modify sleep,
the specic roles of each factor in sleep regulation and the underlying mechanisms
are unknown.
Indirect evidence for a role for BDNF in improving sleep comes from studies
demonstrating that treatments that improve sleep increase BDNF expression. For
example, physical exercise during daylight hours is eective in improving sleep in
humans with various types of sleep disorders (Zisapel, 2001), and it has been shown
that exercise stimulates BDNF production in several dierent brain regions (Cotman
and Berchtold, 2002). Patients with insomnia benet from light therapy in the
morning (Guilleminault et al., 1995); because light exposure increases BDNF
production in the brain (Castren et al., 1992), it is reasonable to consider that BDNF
plays a role in the benecial eects of light therapy. In addition, BDNF production
is stimulated by serotonin and, conversely, BDNF enhances serotonergic signaling
(Russo-Neustadt, 2003). As described in other articles in this volume, serotonin is
a neurotransmitter that plays prominent roles in sleep regulation.
5. Depression, neurotrophic factors, and sleep
Serotonin plays major roles in both depression and sleep. Most patients with
clinical depression have severely disturbed sleep and exhibit reduced levels of
serotonin; the most eective antidepressants including serotonin-selective reuptake
inhibitors and monoamine oxidase inhibitors act, in large part, by increasing
serotonergic signaling (Vaswani et al., 2003). Levels of serotonergic tone decrease in
the brainstem during REM sleep and patients taking antidepressants have reduced
amounts of REM sleep. Interestingly, sleep deprivation increases serotonin levels
and can have antidepressant eects, suggesting that insomnia in depressed patients
may actually be an adaptive response (Adrien, 2002). Antidepressants that increase
serotonergic signaling are very eective in upregulating the expression of BDNF,
and increasing evidence suggests that BDNF plays a major role in the improvement
of symptoms in depressed patients (Russo-Neustadt, 2003). Treatments that have
proven eective in sleep disorders in depressive patients include electroconvulsive

therapy (ECT), sleep deprivation, sleep schedule shifts and bright light therapy
(Saletu, 1986). Each of the latter treatments are associated with increased BDNF
production in the brain (Altar et al., 2004; Hairston et al., 2004). In addition to
aecting the brain circuits involved in the regulation of mood and sleep, BDNF
stimulates neurogenesis, the production of new neurons from stem cells (Lee et al.,
2002; Cheng et al., 2003). Neurogenesis is increasingly recognized as being important
for adaptive responses of the brain to environmental stimuli, and recent ndings
suggest that increased neurogenesis may play a role in recovery from depression
(Benningho et al., 2002).
6. Neurotrophic factors and normal aging
Age-related changes in levels of several dierent neurotrophic factors in the

brains of rodents have been reported. Levels of BDNF were increased in the
hippocampus, but were decreased in the cerebral cortex, striatum, and septum in
aged rats compared to young adult rats (Katoh-Semba et al., 1998). Analysis of
BDNF mRNA levels in samples of prefrontal cortex from humans of dierent ages
demonstrated an increase from youth to adulthood and then stable expression
during aging (Webster et al., 2002). The latter study also documented an age-related
decrease in NGF levels in the cerebral cortex. In mice, BDNF expression levels
decrease with age and this decrease is suppressed by caloric restriction, a dietary
manipulation that extends lifespan (Prolla and Mattson, 2001). Levels of GDNF in
the hippocampus of senescence-accelerated mice were decreased compared to normal
mice of the same genetic background (Miyazaki et al., 2003). Levels of GDNF
decrease in the striatum and ventral midbrain during aging in rats (Yurek and
Fletcher-Turner, 2001). Collectively, the available data suggest that there are brain
region-specic changes in several neurotrophic factors that occur during normal
aging. However, the role of such alterations in age-related sleep disturbances remains
to be determined.
7. Neurotrophic factors and neurodegenerative disorders
7.1. Alzheimers disease
Alzheimers disease (AD) is characterized by dysfunction and degeneration of

neurons in brain regions involved in learning and memory processes (hippocampus,
basal forebrain, and cerebrocortical association areas) and regulation of mood
(amygdala, raphe nucleus, and locus ceruleus) (Engelborghs and De Deyn, 1997).
In addition to their memory and mood problems, AD patients suer from disruption
of circadian rythms and sleep disturbances with many patients sleeping very little
during the night (Vitiello and Borson, 2001). Neuronal damage and neurochemical
abnormalities in the SCN are present in AD, and presumably play a role in sleep
disturbance in these patients (Stopa et al., 1999). The brains of AD patients contain
abnormally high amounts of a protein called amyloid beta-peptide (A) which forms
160 M. P. Mattson
extracellular plaques that are often associated with degenerated neurons. A wealth of
genetic and experimental evidence suggests that A plays a major role in promoting
the dysfunction and degeneration of neurons in AD (Mattson, 2004).
In addition to its adverse eects on learning and memory, the increased
production of A may contribute to sleep disturbances in AD because transgenic
mice with increased levels of A in their brains exhibit abnormalities in the sleep-
wake cycle (Huitron-Resendiz et al., 2002). The mechanism by which A disturbs
circadian rhythms may involve impairment of neurotrophic factor signaling.
As evidence, levels of BDNF are decreased in several brain regions of AD patients
(Hock et al., 2000), and exposure of neurons to A results in impairment of CREB
activation (Tong et al., 2001). It will be of considerable interest to determine the
mechanisms responsible for perturbed sleep regulation in mouse models of AD,
and to elucidate the roles of neurotrophic factors. Sleep disturbances can occur
relatively early in the course of AD, perhaps even prior to cognitive dysfunction,
and may therefore not simply be the consequence of neuronal death.
7.2. Parkinsons disease
Parkinsons disease (PD) is a movement disorder that results from the

degeneration of dopaminergic neurons in the substantia nigra (Santens et al.,
2003). While the substantia nigra is the most severely aected brain region,
certain regions of the cerebral cortex and brainstem also exhibit abnormalities.
Patients with PD typically exhibit insomnia and daytime sleeping (Larsen, 2003).
Some Patients with PD may develop an REM sleep behavior disorder, which has
been suggested to result from degeneration of neurons in brainstem nuclei (Lai and
Siegel, 2003). There is strong evidence that alterations in neurotrophic factors
occur in PD, and emerging ndings suggest that neurotrophic factor therapy can be
eective in ameliorating symptoms in PD patients. Levels of GDNF and BDNF are
decreased in vulnerable brain regions of PD patients (Hunot et al., 1996; Parain et al.,
1999). The GDNF therapy is an eective treatment in animal models of PD, and
recent clinical studies suggest that GDNF therapy will benet humans with PD
(Kirik et al., 2004). The BDNF levels are also decreased in the basal ganglia of PD
patients (Howells et al., 2000), and BDNF therapy can protect dopaminergic
neurons and improve functional outcome in rodent models of PD (Levivier et al.,
1995). Because both GDNF and BDNF have been shown to modify sleep, it may
be the case that depletion of these neurotrophic factors contributes to sleep
disturbances in PD.
7.3. Huntingtons disease
Huntingtons disease (HD) is an inherited disorder that is caused by trinucleotide

(CAG) expansions in the huntingtin gene resulting in polyglutamine expansions
in the huntingtin protein (MacDonald et al., 2003). Mutant forms of huntingtin gene
cause the degeneration of neurons in the striatum (caudate nucleus and putamen)
and cerebral cortex. Other brain regions aected in HD include the hypothalamus
and locus ceruleus (Kremer et al., 1990; Zweig et al., 1992), and HD patients exhibit
sleep disturbances including increased sleep onset latency, reduced sleep eciency,
frequent nocturnal wakings, and reduced amounts of slow wave sleep (Wiegand
et al., 1991). Both HD patients and the transgenic huntingtin mutant mice exhibit
reduced levels of BDNF in basal ganglia (Ferrer et al., 2000; Zucatto et al., 2001;
Duan et al., 2003). Dietary restriction and administration of the antidepressant
paroxetine can increase BDNF levels, suppress the neurodegenerative process and
extend the life of huntingtin mutant mice (Duan et al., 2003, 2004). Analyses of sleep
have not been performed in mouse models of HD, and it therefore remains to be
determined if such mice can be used to elucidate the cellular and molecular
mechanisms responsible for sleep disturbances in HD patients.
8. Conclusions
At this point in time our knowledge of the roles of neurotrophic factors in sleep
regulation and age- and disease-related sleep disturbances is sparse. However,
ndings described earlier and enumerated here are consistent with prominent roles
of neurotrophic factors in sleep regulation. (1) Neurotrophic factor expression
changes during the circadian cycle. (2) Neurons in brain regions involved in sleep
regulation are responsive to one or more neurotrophic factors. (3) Various aspects
of sleep are aected by infusion of neurotrophic factors into the brain.
(4) Neurotrophic factor levels are decreased in humans with, and rodent models
of, neurodegenerative disorders that are characterized by sleep disturbances.
(5) Treatments used to treat sleep disorders increase the expression of BDNF.
Considerable further work will be required to understand the inter-relationships
between neurotrophic factor signaling and the neurotransmitter and neuroendo-
crine systems that regulate sleep. Neurotrophic factors involved in sleep regulation
and components of their signal transduction pathways may prove to be useful
targets for the development of novel approaches for preventing and treating sleep
disorders.
References
Abe, K., Saito, H., 2001. Eects of basic broblast growth factor on central nervous system functions.
Pharmacol. Res. 43, 307312.
Adrien, J., 2002. Neurobiological bases for the relation between sleep and depression. Sleep Med. Rev.
6, 341351.
Altar, C.A., Laeng, P., Jurata, L.W., Brockman, J.A., Lemire, A., Bullard, J., Bukhman, Y.V.,
Toung, T.A., Charles, V., Palfreyman, M.G., 2004. J. Neurosci. 24, 26672677.
Beaule, C., Amir, S., 2002. Eect of 192 IgG-saporin on circadian activity rhythms, expression of P75
neurotrophin receptors, calbindin-D28K, and light-induced Fos in the suprachiasmatic nucleus in rats.
Exp. Neurol. 176, 377389.
Benningho, J., Schmitt, A., Mossner, R., Lesch, K.P., 2002. When cells become depressed; focus on
neural stem cells in novel treatment strategies against depression. J. Neural. Transm. 109, 947962.
Bina, K.G., Rusak, B., 1996. Nerve growth factor phase shifts circadian activity rhythms in Syrian
hamsters. Neurosci. Lett. 206, 97100.
162 M. P. Mattson
Bova, R., Micheli, M.R., Qualadrucci, P., Zucconi, G.G., 1998. BDNF and trkB mRNAs oscillate in rat
brain during the lightdark cycle. Mol. Brain Res. 57, 321324.
Castren, E., Zafra, F., Thoenen, H., Lindholm, D., 1992. Light regulates expression of brain-derived
neurotrophic factor mRNA in rat visual cortex. Proc. Natl. Acad. Sci. USA 89, 94449448.
Cheng, A., Wang, S., Cai, J., Rao, M.S., Mattson, M.P., Cheng, A., Wang, S., Cai, J., Rao, M.S.,
Mattson, M.P., 2003. Nitric oxide acts in a positive feedback loop with BDNF to regulate
neural progenitor cell proliferation and dierentiation in the mammalian brain. Dev. Biol. 258,
319333.
Cirelli, C., Tononi, G., 2000. Dierential expression of plasticity-related genes in waking and sleep and
their regulation by the noradrenergic system. J. Neurosci. 20, 91879194.
Cotman, C.W., Berchtold, N.C., 2002. Exercise: a behavioral intervention to enhance brain health and
plasticity. Trends Neurosci. 25, 295301.
Dechant, G., Neumann, H., 2002. Neurotrophins. Adv. Exp. Med. Biol. 513, 303334.
Dolci, C., Montaruli, A., Roveda, E., Barajon, I., Vizzotto, L., Grassi Zucconi, G., Carandente, F., 2003.
Circadian variations in expression of the trkB receptor in adult rat hippocampus. Brain Res. 994,
6772.
Duan, W., Guo, Z., Jiang, H., Ware, M., Li, X.J., Mattson, M.P., 2003. Dietary restriction normalizes
glucose metabolism and BDNF levels, slows disease progression, and increases survival in huntingtin
mutant mice. Proc. Natl. Acad. Sci. USA 100, 29112916.
Duan, W., Guo, Z., Jiang, H., Ladenheim, B., Xu, X., Cadet, J.L., Mattson, M.P., 2004. Paroxetine
retards disease onset and progression in Huntingtin mutant mice. Ann. Neurol. 55, 590594.
Engelborghs, S., De Deyn, P.P., 1997. The neurochemistry of Alzheimers disease. Acta. Neurol. Belg. 97,
6784.
Ferrer, I., Goutan, E., Marin, C., Rey, M.J., Ribalta, T., 2000. Brain-derived neurotrophic factor in
Huntington disease. Brain Res. 866, 257261.
Guilleminault, C., Clerk, A., Black, J., Labanowski, M., Pelayo, R., Claman, D., 1995. Nondrug
treatment trials in psychophysiologic insomnia. Arch. Intern. Med. 155, 838844.
Hairston, I.S., Peyron, C., Denning, D.P., Ruby, N.F., Flores, J., Sapolsky, R.M., Heller, H.C.,
OHara, B.F., 2004. Sleep deprivation eects on growth factor expression in neonatal rats; a potential
role for BDNF in the mediation of delta power. J. Neurophysiol. 91, 15861595.
Hobson, J.A., Pace-Schott, E.F., 2002. The cognitive neuroscience of sleep: neuronal systems,
consciousness and learning. Nat. Rev. Neurosci. 3, 679693.
Hock, C., Heese, K., Hulette, C., Rosenberg, C., Otten, U., 2000. Region-specic neurotrophin
imbalances in Alzheimer disease: decreased levels of brain-derived neurotrophic factor and increased
levels of nerve growth factor in hippocampus and cortical areas. Arch. Neurol. 57, 846851.
Howells, D.W., Porritt, M.J., Wong, J.Y., Batchelor, P.E., Kalnins, R., Hughes, A.J., Donnan, G.A.,
2000. Reduced BDNF mRNA expression in the Parkinsons disease substantia nigra. Exp. Neurol.
166, 127135.
Huang, E.J., Reichardt, L.F., 2003. Trk receptors: roles in neuronal signal transduction. Annu. Rev.
Biochem. 72, 609642.
Hughes, P.E., Alexi, T., Walton, M., Williams, C.E., Dragunow, M., Clark, R.G., Gluckman, P.D., 1999.
Activity and injury-dependent expression of inducible transcription factors, growth factors and
apoptosis-related genes within the central nervous system. Prog. Neurobiol. 57, 421450.
Huitron-Resendiz, S., Sanchez-Alavez, M., Gallegos, R., Berg, G., Crawford, E., Giacchino, J.L., Games,
D., Henriksen, S.J., Criado, J.R., 2002. Age-independent and age-related decits in visuospatial
learning, sleep-wake states, thermoregulation and motor activity in PDAPP mice. Brain Res. 929,
126137. Glial cell line-derived neurotrophic factor (GDNF) gene expression in the human brain: a
post mortem in situ hybridization study with special reference to Parkinsons disease. J. Neural.
Transm. 103, 10431052.
Hunot, S., Bernard, V., Faucheux, B., Boissiere, F., Leguern, E., Brana, C., Gautris, P.P., Guerin, J.,
Block, B., Agid, Y., Hirsch, E.C., 1996. Glial cell line-derived neutrophic factor (GDNF) gene
expression in the human brain: a post mortem in situ hybridization study with special reference to
Parkinsons disease. J. Neural. Transm. 103, 10431052.
Jiang, X., Zhu, D., Okagaki, P., Lipsky, R., Wu, X., Banaudha, K., Mearow, K., Strauss, K.I.,
Marini, A.M., 2003. N-methyl-D-aspartate and TrkB receptor activation in cerebellar granule cells: an
in vitro model of preconditioning to stimulate intrinsic survival pathways in neurons. Ann. N.Y.
Acad. Sci. 993, 134145.
Katoh-Semba, R., Semba, R., Takeuchi, I.K., Kato, K., 1998. Age-related changes in levels of
brain-derived neurotrophic factor in selected brain regions of rats, normal mice and senescene-
accelerated mice: a comparison to those of nerve growth factor and neurotrophin-3. Neurosci.
Res. 31, 227234.
Kirik, D., Georgievska, B., Bjorklund, A., 2004. Localized striatal delivery of GDNF as a treatment for
Parkinsons disease. Nat. Neurosci. 7, 105110.
Kramer, A., Yang, F.C., Snodgrass, P., Li, X., Scammell, T.E., Davis, F.C., Weitz, C.J., 2001.
Regulation of daily locomotor activity and sleep by hypothalamic EGF receptor signaling. Science
294, 25112515.
Kremer, H.P., Roos, R.A., Dingjan, G., Marani, E., Bots, G.T., 1990. J. Neuropathol. Exp. Neurol. 49,
371382.
Kushikata, T., Fang, J., Chen, Z., Wang, Y., Krueger, J.M., 1998. Epidermal growth factor enhances
spontaneous sleep in rabbits. Am. J. Physiol. 275, R509R514.
Kushikata, T., Fang, J., Krueger, J.M., 1999. Brain-derived neurotrophic factor enhances spontaneous
sleep in rats and rabbits. Am. J. Physiol. 276, R13341338.
Kushikata, T., Kubota, T., Fang, J., Krueger, J.M., 2001. Glial cell line-derived neurotrophic factor
promotes sleep in rats and rabbits. Am. J. Physiol. Regul. Integr. Comp. Physiol. 280, R1001R1006.
Kushikata, T., Kubota, T., Fang, J., Krueger, J.M., 2003. Neurotrophins 3 and 4 enhance non-rapid eye
movement sleep in rabbits. Neurosci. Lett. 346, 161164.
Lai, Y.Y., Siegel, J.M., 2003. Physiological and anatomical link between Parkinson-like disease and REM
sleep behavior disorder. Mol. Neurobiol. 27, 137152.
Larsen, J.P., 2003. Sleep disorders in Parkinsons disease. Adv. Neurol. 91, 329334.
Lee, J., Duan, W., Mattson, M.P., 2002. Evidence that brain-derived neurotrophic factor is required for
basal neurogenesis and mediates, in part, the enhancement of neurogenesis by dietary restriction in the
hippocampus of adult mice. J. Neurochem. 82, 13671375.
Levivier, M., Przedborski, S., Bencsics, C., Kang, U.J., 1995. Intrastriatal implantation of broblasts
genetically engineered to produce brain-derived neurotrophic factor prevents degeneration of
dopaminergic neurons in a rat model of Parkinsons disease. J. Neurosci. 15, 78107820.
Liang, F.Q., Walline, R., Earnest, D.J., 1998. Circadian rhythm of brain-derived neurotrophic factor in
the rat suprachiasmatic nucleus. Neurosc. Lett. 242, 8992.
Liang, F.Q., Allen, G., Earnest, D., 2000. Role of brain-derived neurotrophic factor in the circadian
regulation of the suprachiasmatic pacemaker by light. J. Neurosci. 20, 29782987.
Lu, B., Chow, A., 1999. Neurotrophins and hippocampal synaptic transmission and plasticity. J. Neurosci.
Res. 58, 7687.
MacDonald, M.E., Gines, S., Gusella, J.F., Wheeler, V.C., 2003. Huntingtons disease. Neuromolecular
Med. 4, 720.
Mattson, M.P., 2004. Pathways towards and away from Alzheimers disease. Nature 430, 631639.
Miyazaki, H., Okuma, Y., Nomura, J., Nagashima, K., Nomura, Y., 2003. Age-related alterations in the
expression of glial cell line-derived neurotrophic factor in senescene-accelerated mouse brain.
J. Pharmacol. Sci. 92, 2834.
Parain, K., Murer, M.G., Yan, Q., Faucheux, B., Agid, Y., Hirsch, E., Raisman-Vozari, R., 1999.
Reduced expression of brain-derived neurotrophic factor protein in Parkinsons disease substantia
nigra. Neuroreport 10, 557561.
Plata-Salaman, C.R., 1991. Epidermal growth factor and the nervous system. Peptides 12, 653663.
Pollock, G.S., Vernon, E., Forbes, M.E., Yan, Q., Ma, Y.T., Hsieh, T., Robichon, R., Frost, D.O.,
Johnson, J.E., 2001. Eects of early visual experience and diurnal rhythms on BDNF mRNA and
protein levels in the visual system, hippocampus, and cerebellum. J. Neurosci. 21, 39233931.
Prolla, T.A., Mattson, M.P., 2001. Molecular mechanisms of brain agingand neurodegenerative disorders:
lessons from dietary restriction. Trends Neurosci. 23, S21S31.
164 M. P. Mattson
Russo-Neustadt, A., 2003. Brain-derived neurotrophic factor, behavior, and new directions for the
treatment of mental disorders. Semin. Clin. Neuropsychiatry 8, 109118.
Saletu, B., 1986. Therapy for sleep disorders in depressives. Psychopathology 19, S239S262.
Santens, P., Boon, P., Van Roost, D., Caemaert, J., 2003. The pathophysiology of motor symptoms in
Parkinsons disease. Acta Neurol. Belg. 103, 129134.
Sei, H., Saitoh, D., Yamamoto, K., Morita, K., Morita, Y., 2000. Dierential eect of short-term REM
sleep deprivation on NGF and BDNF protein levels in the rat brain. Brain Res. 877, 387390.
Sei, H., Fujihara, H., Ueta, Y., Morita, K., Kitahama, K., Morita, Y., 2003. Single eight-hour shift
of light-dark cycle increases brain-derived neurotrophic factor protein levels in the rat hippocampus.
Life Sci. 73, 5359.
Stopa, E.G., Volicer, L., Kuo-Leblanc, V., Harper, D., Lathi, D., Tate, B., Satlin, A., 1999. Pathologic
evaluation of the human suprachiasmatic nucleus in severe dementia. J. Neuropathol. Exp. Neurol.
58, 2939.
Taishi, P., Sanchez, C., Wang, Y., Fang, J., Harding, J.W., Krueger, J.M., 2001. Conditions that aect
sleep alter the expression of molecules associated with synaptic plasticity. Am. J. Physiol. Regul.
Integr. Comp. Physiol. 281, R839R845.
Thoenen, H., 2000. Neurotrophins and activity-dependent plasticity. Prog. Brain Res. 128, 183191.
Tong, L., Thornton, P.L., Balazs, R., Cotman, C.W., 2001. Beta-amyloid-(1-42) impairs activity-
dependent cAMP-response element-binding protein signaling in neurons at concentrations in which
cell survival is not compromised. J. Biol. Chem. 276, 1730117306.
Unsicker, K., Krieglstein, K., 2002. TGF-betas and their roles in the regulation of neuron survival. Adv.
Exp. Med. Biol. 513, 353374.
Vaswani, M., Linda, F.K., Ramesh, S., 2003. Role of serotonin reuptake inhibitors inpsychiatric
disorders: a comprehensive review. Prog. Neuropsychopharmacol. Biol. Psychiatry 27, 85102.
Vitiello, M.V., Borson, S., 2001. Sleep disturbances in patients with Alzheimers disease: epidemiology,
pathophysiology and treatment. CNS Drugs 15, 777796.
Webster, M.J., Wickert, C.S., Herman, M.M., Kleinman, J.E., 2002. BDNF mRNA expression during
postnatal development, maturation and aging of the human prefrontal cortex. Dev. Brain Res. 139,
139150.
Wiegand, M., Moller, A.A., Lauer, C.J., Stolz, S., Schreiber, W., Dose, M., Krieg, J.C., 1991. Nocturnal
sleep in Huntingtons disease. J. Neurol. 238, 203208.
Yamuy, J., Rojas, M.J., Torterolo, P., Sampogna, S., Chase, M.H., 2002. Induction of rapid eye
movement sleep by neurotrophin-3 and its co-localization with choline acetyltransferase in
mesopontine neurons. Neuroscience 115, 8595.
Yurek, D.M., Fletcher-Turner, A., 2001. Dierential expression of GDNF, BDNF and NT-3 in the aging
nigrostriatal system following a neurotoxic lesion. Brain Res. 891, 228235.
Zisapel, N., 2001. Circadian rhythm sleep disorders: pathophysiology and potential approaches to
management. CNS Drugs 15, 311328.
Zuccato, C., Ciammola, A., Rigamonti, D., Leavitt, B.R., Goredo, D., Conti, L., MacDonald, M.E.,
Friedlander, R.M., Silani, V., Hayden, M.R., Timmusk, T., Sipione, S., Cattaneo, E., 2001. Loss of
huntingtin-mediated BDNF gene transcription in Huntingtons disease. Science 293, 493498.
Zweig, R.M., Ross, C.A., Hedreen, J.C., Peyser, C., Cardillo, J.E., Folstein, S.E., Price, D.L., 1992. Locus
coeruleus involvement in Huntingtons disease. Arch. Neurol. 49, 152156.
Advances in
Cell Aging and
Gerontology
Sleep and aging: molecular approaches within

a systems neurobiology context
Akira Terao and Thomas S. Kilduff
Molecular Neurobiology Laboratory, Biosciences Division, SRI International Menlo Park, CA 94025, USA.
Correspondence address: T. S. Kilduff, Biosciences Division, SRI International, 333 Ravenswood Avenue,
Menlo Park, CA 94025, USA. Tel: 1-650-859-5509; fax: 1-650-859-5535.
E-mail address: thomas.kilduff@sri.com
Contents
1. Introduction
2. Sleep and aging
2.1. Sleep and the elderly
2.2. Sleep in aged rodents
3. Neurotransmitter systems involved in sleep and wakefulness
3.1. Discovery of the hypocretin/orexin system
3.2. The H/O system and the sleep disorder narcolepsy
3.3. A model for the role of the H/O system in sleep and wakefulness
3.4. H/O system and aging
3.5. Histaminergic system and aging
4. Sleep deprivation, recovery sleep, and gene expression
4.1. Homeostatic regulation of sleep
4.2. Gene expression studies of sleep deprivation and recovery sleep
5. Microarray studies of gene expression in the aged brain
6. Conclusions and future directions
1. Introduction
As the survivorship curve for humans in the developed world approaches the
rectangular ideal, improving the quality of life in advanced years is an increasing
concern. A major complaint of the elderly is poor sleep (Ballinger, 1976; Karacan
et al., 1976; Bliwise et al., 1992). Despite the fact that the elderly tend to spend
more hours in bed, they have less total sleep time (Feinberg et al., 1967; Kahn, 1970;

DOI: 10.1016/S1566-3124(05)17007-2
166 A. Terao and T. S. Kilduff
Webb and Swinburne, 1971; Prinz, 1995). In general, sleep eciency decreases with
age, mostly due to more and lengthier arousals, in other words, increased sleep
fragmentation (Feinberg et al., 1967; Kahn and Fisher, 1969; Prinz, 1977; Bliwise,
1993). In this chapter, we will review sleep characteristics of the elderly and introduce
the homeostatic regulation of sleep, discuss some of the neurotransmitter systems
known to play a role in sleep and wakefulness and how these systems change with
age, and describe gene expression studies of sleep and wakefulness undertaken
to identify the molecular mechanisms that underlie sleep homeostasis. We will
conclude with a discussion of the potential application of these latter approaches
to understand what goes awry with sleep during aging.
2. Sleep and aging
2.1. Sleep and the elderly
A variety of studies have established that sleep in elderly humans is characterized

by sleep fragmentation, a reduced amplitude of the diurnal rhythm of sleep and
wakefulness, and reduced intensity of delta range of the EEG (Bliwise, 1993). Sleep
fragmentation is evident from the common observation of frequent nocturnal
awakenings in older adults. Such awakenings, coupled with the increased tendency
of the elderly to nap during the daytime, combine to result in a reduced amplitude of
the distribution or rhythm of sleep and wakefulness in comparison with that found
in younger individuals. Consequently, it is thought that at least some age-related
sleepwake problems may reect a dampening of the amplitude of the circadian
rhythm pacemaker or its coupling to the sleepwake system (van Someren et al.,
1993). However, evidence from both human (Czeisler et al., 1999) and animal (Davis
and Viswanathan, 1998; Duy et al., 1999) studies indicates that the period of the
circadian pacemaker in the suprachiasmatic nucleus (SCN) does not change with
age, suggesting that the reduced amplitude of the sleepwake rhythm occurs in one
of the output targets downstream from the SCN.
Along with sleep fragmentation, polysomnographic analyses of sleep in the elderly
reveal substantial changes in the architecture of both rapid eye movement (REM)
and the four stages of non-REM (NREM) sleep. Since the earliest reports (Agnew
et al., 1967; Kales et al., 1967; Kahn, 1970), virtually all studies agree on an increase
in stage 1, increased fragmentation of stage 2, very large decrements in stages 3 and 4,
and a decline in total REM sleep in older humans, although the proportion of REM
sleep as percent of total sleep time does not change. Given the large and consistent
reports of diminished stages 3 and 4 sleep, the decline of delta frequencies in the
electroencephalogram (EEG) of the elderly has been of particular interest because
of the potential implications for models of the homeostatic regulation of sleep.
The brain mechanisms underlying the increased sleep fragmentation and
decreased diurnal rhythms characteristic of aging are unknown. Sleep and wake-
fulness rhythms appear to be under the control of at least two distinct systems, a
homeostatic process that increases sleep tendency as a function of prior time awake
and an opponent process that promotes wakefulness at particular times of day
Sleep and Aging: Molecular Approaches within a Systems Neurobiology Context 167
(Borbely, 1982; Edgar et al., 1993). Aging could alter either the wakefulness
promotion process, thought to be mediated via the circadian pacemaker in the SCN,
and/or the sleep homeostatic process (Edgar, 1994). The sleep homeostatic process is
measured by spectral analyses of the delta band of the EEG during sleep because of
the relationship between the duration of prior wakefulness and subsequent power in
this band during sleep (Borbely et al., 1981). As formulated in the two-process
model of sleep regulation (Borbely, 1982), EEG delta power (mostly occurring
during stages 3 and 4) reects the homeostatic drive to sleep and the restorative
process that occurs during sleep.
Because of the tantalizing possibility that delta power is directly related to a
function of sleep, it is of interest to know how it changes with aging. A considerable
change in the dynamics of delta power has been observed as early as middle age in a
study comparing young (2028 years) and middle-age (4256 years) men. Although
delta power was much higher at the beginning of the sleep period in the young group,
delta power declined throughout the sleep period to reach very similar levels at the
end of the night in both groups (Dijk et al., 1989). So, if delta power is a measure of a
function of sleep, it already undergoes considerable decline by middle-age. As
indicated above, elderly humans are well-known to exhibit a dramatic reduction in
delta power, particularly in the rst half of the night (Ehlers and Kupfer, 1989),
which suggests that their sleep may be less restorative.
Aside from these objective characteristics of sleep in the elderly, a key element is
the relationship between nocturnal sleep and alertness during the subsequent
daytime. Sleep quality in the elderly and its impact on subsequent wakefulness has
been assessed by the multiple sleep latency test (MSLT) (Carskadon et al., 1982).
Low MSLT scores, indicative of a low level of daytime alertness, correlated less with
total amount of prior sleep than with the extent to which sleep was fragmented.
Thus, insights into causes of poor sleep quality in the elderly are likely to come from
investigations of sleep structure.
2.2. Sleep in aged rodents
As summarized earlier, the three major characteristics of sleep in elderly humans

are sleep fragmentation, reduced amplitude of the diurnal rhythm of sleep and
wakefulness, and reduced intensity of delta range of the EEG. By and large, these
characteristics are also present in a number of animal models. Studies of sleep in
aged rats have varied with regard to animal age, sex, and strain (Sprague-Dawley,
Wistar, Long Evans, Brown Norway, Fischer 344) and have utilized dierent
sleepwake analyses, but researchers have reported remarkably similar results.
Unlike humans, laboratory animal species do not show a single consolidated sleep
period. Therefore, evidence for sleep fragmentation has been indicated by decreased
wake, NREM, and/or REM bout lengths in a number of studies of aged rats
(Rosenberg et al., 1979; Van Gool and Mirmiran, 1983; van Gool et al., 1987;
Shiromani et al., 2000). Studies in other species, including mice (Welsh et al., 1986),
hamsters (Naylor et al., 1998), and cats (Bowersox et al., 1984) have also found sleep
fragmentation to be a characteristic of aged animals. It is encouraging that both
human and animal studies are consistent with regard to the dominant change in sleep
that correlates best with decreased daytime alertness in the elderly.
A reduction in the amplitude of the diurnal rhythms of sleep and wakefulness
(sometimes reported as a reduced light/dark ratio) is also a consistently observed
characteristic of sleep in aged rats (Zepelin et al., 1972; Rosenberg et al., 1979; Van
Gool and Mirmiran, 1983; Witting et al., 1993; Li and Satino, 1995). However, this
reduction may be dependent upon environmental conditions since such a diminution
was not evident in sleep studies conducted in constant dim light (van Gool et al.,
1987; Mendelson and Bergmann, 1999a).
The third characteristic of sleep of elderly people, the reduction in EEG delta
activity, has been more controversial in animal studies. Although reduced delta in
NREM sleep is clearly present in the cat (Bowersox et al., 1984), early studies of the
rat were unable to document such changes (Zepelin et al., 1972; Rosenberg et al.,
1979; Van Gool and Mirmiran, 1983). However, it is likely that the failure to
identify changes in delta activity in these studies was due to methodological
problems. For example, the studies of Rosenberg et al. (Rosenberg et al., 1979)
ltered out EEG frequencies below 2 Hz and above 4 Hz and then quantied delta
activity by using zero-crossing and accumulated voltage over time. This method
misses a signicant part of the delta band (0.52 Hz) that is of considerable
signicance for studies of delta power. Furthermore, the quantitation of delta power
used in these early studies is not as sensitive as spectral analysis using a fast
Fourier transform (FFT). Using FFT analysis of the 2.03.6 Hz range, a reduction
in delta activity was documented during slow wave sleep (SWS) in F344 rats
(Tani and Ishihara, 1988). Subsequently, reduced delta activity during NREM sleep
was observed in aged hamsters (Naylor et al., 1998) and Fischer rats (Mendelson
and Bergmann, 1999a,b; Shiromani et al., 2000) but not in Sprague-Dawley rats
(Shiromani et al., 2000). High amplitude NREM sleep has also been reported to
decrease (Mendelson and Bergmann, 1999a) in Fischer rats, suggesting a similarity
with aged human sleep in which decits in stages 3 and 4 of NREM sleep are
observed. Thus, it is likely that the failure to report age-related changes in delta
activity in older studies was due to (1) inappropriate ltering and analyses and (2) a
lack of distinction between light and deep NREM states.
Heterogeneity of the aging process has become a dominant theme in the clinical
literature. A number of researchers studying processes such as cognitive function
(Ylikoski et al., 1999), molecular genetics (De Benedictis et al., 1999), and sleep
(Reynolds et al., 1993) have identied distinct clusters of the aged population that
represent successful, usual, and pathological aging. This concept has also inuenced
the design of animal studies. For example, in aged female Long-Evans rats, some
rats have been found to have severely disrupted circadian temperature rhythms
(CTRs) while other rats display sleep patterns and CTRs identical to those of young
animals (Li and Satino, 1995; Satino, 1998). Importantly, disruption of CTRs was
a good predictor of sleep pattern changes such as reduced amplitude of the
sleepwake rhythms and sleep fragmentation, particularly REM sleep (Li and
Satino, 1995). On the other hand, a study of male F344 rats did not identify
changes in CTRs despite signicant decreases in both sleep bout lengths and high
voltage NREM sleep (Mendelson and Bergmann, 1999a). While strain, sex, and age
dierences may account for some of this dierence between these studies, the lack of
agreement may also be related to the heterogeneity of the aged population under
study and whether the experimental design focused on group versus individual
dierences.
3. Neurotransmitter systems involved in sleep and wakefulness
As indicated before, the brain mechanisms underlying the increased sleep

fragmentation and decreased diurnal rhythms characteristic of aging are currently
unknown. Recent studies have pointed to the importance of the hypocretin/orexin
(H/O) system as a novel neurotransmitter system involved in the maintenance of
alertness. Although the primary role of the H/O system was originally believed to be
appetite regulation, research on this system has become focused on its role in the
regulation of sleep, arousal, and locomotor activity and it is in this context that the
H/O system may be related to sleep problems in the elderly. Intriguingly, this
system may also be a downstream target of the SCN (Abrahamson et al., 2001;
Gerashchenko et al., 2001).
3.1. Discovery of the hypocretin/orexin system
The hypocretinsare a pair of peptides that were rst described early in 1998
(de Lecea et al., 1998; Sakurai et al., 1998) and were given this name because the
neurons expressing this gene were restricted to an area centered around the
perifornical nucleus of the hypothalamus (PFH) and because of a weak homology to
the gut peptide secretin. The name orexins stemmed from early studies which
indicated that intracerebroventricular (ICV) injections of either of these peptides into
rats stimulated food intake (Sakurai et al., 1998). The H/O system includes two
G protein-coupled receptors: the orexin-1 receptor (OX1R) was shown to
preferentially bind orexin-A (hypocretin-1) over orexin-B (hypocretin-2), whereas
the orexin-2 receptor (OX2R) binds both peptides at high anity (Sakurai et al.,
1998). In the rodent hypothalamus, there are a few thousand cell bodies that make
these peptides but these neurons project widely throughout the brain except for the
cerebellum (Peyron et al., 1998). The monoaminergic and cholinergic cell groups
classically implicated in arousal state regulation receive H/O input and cells in the
locus coeruleus (LC) are particularly highly innervated.
3.2. The H/O system and the sleep disorder narcolepsy
The widespread anatomical projections of the H/O neurons immediately suggested

that they may play a role in functions other than food intake regulation (Peyron
et al., 1998). Indeed, subsequent studies implicated the H/O system in neuro-
endocrine (Pu et al., 1998; van den Pol et al., 1998; Lopez et al., 1999; Mitsuma et al.,
1999; Tamura et al., 1999), cardiovascular (Samson et al., 1999; Shirasaka et al.,
1999), water balance (Kunii et al., 1999), and gastrointestinal (Takahashi et al., 1999)
control. Of particular relevance to sleep and aging are two landmark papers
(Chemelli et al., 1999; Lin et al., 1999) which established that dysfunction of the
H/O system can result in the sleep disorder narcolepsy.
Narcolepsy is characterized by excessive daytime sleepiness (EDS), episodes of
muscle weakness (cataplexy) triggered by emotional stimulation, and abnormalities
of REM sleep. A genetic component of this disorder has been established in
both humans and dogs. A strong link with the HLA Class II antigens exists in
human narcoleptics (Juji et al., 1984): across ethnic groups, HLA DQB1*0602 is
presented in more than 85% of narcoleptic patients with denite cataplexy but
only 1238% of the general population (Mignot, 1998). Such close association with
the HLA system has led to the suggestion that narcolepsy may be an autoimmune
disease (Mignot et al., 1995).
In studies in narcoleptic dogs, the canarc-1 gene, transmitted in Doberman
pinschers and Labrador retrievers as an autosomal recessive trait with full
penetrance, was identied as a deletion mutation in the hypocretin receptor 2
(hcrtr2 OX2R) gene, resulting in a truncated, nonfunctional protein (Lin et al.,
1999). In a remarkable convergence, H/O null mutant mice were found to exhibit
periods of behavioral arrest that strongly resemble the cataplectic attacks and
sleep-onset REM periods characteristic of narcolepsy in other species (Chemelli et al.,
1999). These mice also have an altered sleep architecture, as evidenced by increased
levels of both REM and NREM sleep, short latency REM periods, and decreased
sleep bout lengths, primarily during the dark (active) period. Thus, these two articles
indicate that dysfunction of either the H/O ligand or one of its receptors can result in
narcolepsy.
Although these studies in narcoleptic dog and mouse models implicated the
H/O system, the cause of narcolepsy in humans was unknown until recently.
An abnormality in H/O neurotransmission in narcoleptic humans was rst
suggested by undetectable levels of Hcrt-1 in cerebrospinal uid (CSF) from 7 of 9
narcoleptic patients (Nishino et al., 2000). In postmortem analyses, H/O mRNA
was undetectable in two narcoleptic brains, although mRNA for MCH was
readily detectable in both controls and narcoleptics (Peyron et al., 2000).
Immunohistochemistry showed an 8595% reduction in the number of H/O-
containing cells in narcoleptic brains with no evident change in the number of MCH
cells (Thannickal et al., 2000). These studies suggest that degeneration of the H/O
cells may be the cause of human narcolepsy. Increased staining for glial brillary
acid protein in the PFH suggests that the H/O cells may degenerate by an
autoimmune mechanism(Thannickal et al., 2000). In animal models, targeted lesion
of the H/O neurons by either genetic (Hara et al., 2001) or pharmacological
(Gerashchenko et al., 2001) means results in sleep fragmentation and decreased
amplitude of the diurnal rhythm of sleep and wakefulness.
3.3. A model for the role of the H/O system in sleep and wakefulness
The region of the posterior hypothalamus (PH) containing the H/O cells has long
been implicated in arousal state control (Nauta, 1946). Since narcolepsy is
characterized by both EDS and abnormal REM sleep, dysfunction of the H/O
system in narcoleptic dogs, mice, and humans suggests that this system plays an
important role in both waking and REM sleep regulation and raises the question of
the role of these peptides in normal sleep regulation. The ICV injections of
hypocretin-1 (Hcrt-1) into rats at light onset (the major sleep period) increases
arousal and locomotor activity and decreases REM sleep without aecting NREM
sleep (Hagan et al., 1999). The wakefulness induced by Hcrt-1 injections is associated
with an intense grooming response (Ida et al., 1999) and increased plasma
corticosterone (Hagan et al., 1999). These eects may be related to projections from
the H/O cells to monoaminergic cell groups classically implicated in arousal state
regulation, including histaminergic cells of the tuberomammillary nucleus (TMN)
(Chemelli et al., 1999), the serotonergic cells of the DRN (Chemelli et al., 1999), and
the noradrenergic cells of the LC (Horvath et al., 1999). The H/O terminals are also
in apposition with cholinergic cells in brain regions involved in the EEG
desynchronization characteristic of waking and REM sleep, including the
laterodorsal tegmental nucleus (LDT), the pedunculopontine nucleus (PPT), and
in basal forebrain (BF) regions, e.g., the horizontal and vertical limbs of the diagonal
band of Broca (Chemelli et al., 1999; Scammell et al., 2000). The preoptic area (POA)
also contains H/O terminals and is a sleep/arousal regulatory site; microinjection of
Hcrt-1 increased wakefulness and suppressed sleep for at least 70 min post-injection
(Methippara et al., 2000).
We have proposed a model whereby the H/O cells might be involved in arousal
state control and the EDS of narcolepsy (Kildu and Peyron, 2000) (Fig. 1).
A complementary model has been proposed to account for the role of this
system in descending motor control and cataplexy (Siegel, 1999). Our model
assumes that H/O cells have a wake/REM-related activity. Three types of neurons
have been electrophysiologically identied in the PH: wake-related neurons, wake/
REM-related neurons, and REM-related neurons (Vanni-Mercier et al., 1984;
Steininger et al., 1999). Although the neuronal populations recorded were not
Waking
LDT, PPT (Ach)

Basal Forebrain
(Ach, GABA) Hcrt
LC (NE)
DRN (5-HT)
TMN (Histamine)
VTA (Dopamine)
Fig. 1. Model illustrating major connections from the H/O cells to some of the brain regions implicated in
arousal state control. See text for details and Kildu and Peyron (2000) for full explanation of model.
Abbreviations: 5-HT, serotonin; Ach, acetylcholine; DRN, dorsal raphe nucleus; LC, locus coeruleus;
LDT, laterodorsal tegmental nucleus; NE, norepinephrine; PPT, pedunculopontine nucleus; TMN,
tuberomammillary nucleus. Symbols: excitatory inputs j; inhibitory inputs 3.
chemically characterized, these studies were conducted in the general region

containing H/O cells. The presence of Fos labeling in H/O cells after treatment with
the wake-promoting drug modanil suggests that these cells may indeed be active
during waking (Chemelli et al., 1999). During waking, these cells might promote
arousal through an excitation of the wake-active monoaminergic populations in
the TMN, LC, and DRN. The H/O cells could also promote EEG desynchronization
via excitatory inputs onto the cholinergic cell groups of the BF and the pons (LDT,
PPT); however, the LDT and PPT are also subject to inhibitory input from the
monoaminergic groups in the awake state. During NREM sleep, a decline in the
activity of the H/O cells, perhaps due to release of GABA (Nitz and Siegel, 1996)
from the POA (Gritti et al., 1994), would result in decreased excitation to these
monoaminergic and cholinergic groups, facilitating cortical synchronization. During
REM sleep, excitatory input from the H/O cells onto cholinergic cells would
facilitate cortical desynchronization, but excitatory eects onto the monoaminergic
cell groups might be blocked by presynaptic inhibition (Gervasoni et al., 1998), since
GABA levels are high in both the LC (Nitz and Siegel, 1997b) and DRN (Nitz and
Siegel, 1997a) during this state. The absence of excitatory input from the H/O cells
would result in a net excitation of the REM on groups in the LDT and PPT
through disinhibition. Characterization of the discharge pattern of the H/O cells
across arousal states will be essential to understand the involvement of these cells in
normal sleepwake regulation. Because the H/O system appears to play a crucial
role innervating the monoaminergic and cholinergic systems classically implicated in
arousal state control, an age-related dysfunction of the H/O system could have
signicant implications for the quality of sleep in the elderly.
3.4. H/O system and aging
As indicated above, the H/O system has been implicated in arousal state
regulation and energy metabolism. We hypothesize that the changes in the sleep of
elderly humans, particularly decreased amplitude of the diurnal sleepwake rhythms
and increased sleep fragmentation, may be due to an age-related dysfunction of the
H/O system. As an initial test of this hypothesis, we measured mRNA levels of
preprohcrt and the colocalized dynorphin in the hypothalamus, and hcrt receptor 1
(hcrtr1) and hcrtr2 in 8 brain regions of young (3 months), middle-aged (12 and 18
months) and old (24 months) male C57BL/6 mice (Terao et al., 2002b). The cDNA
synthesis and mRNA quantitation using a real-time uorescence detection method
was conducted as described previously (Terao et al., 2000). Expression of the
preprohcrt and dynorphin genes did not change with age. An age-related change in
the expression of hcrtr1 mRNA was observed only in the hippocampus (p 0.029).
In contrast, hcrtr2 mRNA levels showed an age-related variation in the
hippocampus (p 0.035), thalamus (p 0.002), pons (p 0.022), and medulla
(p 0.0001); these reductions in mRNA levels ranged from 33 to 44% (Fig. 2A).
Trends (p 0.1) toward age-related declines in hcrtr2 mRNA levels were also
observed in the basal forebrain and hypothalamus. Narcoleptic dogs have a non-
functional hcrtr2 gene and exhibit excessive sleepiness (Lin et al., 1999) and hcrtr2
A B
R1 Cx R1 Cx
Hp 27% Hp
Cb Cb
BF BF
Th Th
Sleep and Aging: Molecular Approaches within a Systems Neurobiology Context

Hy Pn Hy Pn
Md 38% Md
33%
R2 Cx R2 Cx
Hp Hp
Cb Cb
BF 33% BF
Th Th 91%
44%
Hy Pn Hy Pn
Md Md
38%
R3 Cx
Hp
Cb
BF
Th
Hy Pn
Md
32%
173
Fig. 2. Summary of hypocretin/orexin (A) and histamine (B) receptor family gene mRNA expression changes across eight brain areas. Number indicates
percent change compared to 3 month old group, as determined by real-time RTPCR analysis. Abbreviations: Cx, cerebral cortex; Hp, hippocampus; Th,
thalamus; BF, basal forebrain; Hy, hypothalamus; Pn, pons; Md, medulla; Cb, cerebellum; NS no signicant dierence.
knockout mice (Willie et al., 2003) also exhibit disrupted sleep. Thus, these results
are consistent with the hypothesis that an age-related deterioration occurs in the H/O
system that may underlie age-related sleep disorders.
At the present time, the idea that such age-related changes in sleep parameters are
due, at least in part, to dysfunction of the H/O system is simply a hypothesis.
From both in vivo microinjection studies of the peptides (Hagan et al., 1999; Bourgin
et al., 2000; Methippara et al., 2000; Piper et al., 2000) and in vivo measurements of
Hcrt-1 (Fujiki et al., 2001; Yoshida et al., 2001), the H/O system appears to
potentiate wakefulness. A wakefulness-promoting role for the H/O system is also
supported by both animal and human studies in which EDS (and cataplexy)
results when this system is dysfunctional. This EDS contributes to an overall
dimunition of the diurnal rhythms of sleep and wakefulness in human narcoleptics.
In rats, neurotoxic destruction of the Hcrt receptor-bearing cells in the hypo-
thalamus attens the diurnal rhythm of sleepwake by increasing the amount of
sleep that occurs during the normal active period (Gerashchenko et al., 2001).
Thus, an age-related dysfunction of the H/O system could result in the reduced
amplitude of the diurnal rhythms of sleep and wakefulness characteristic of
both elderly humans and aged animals by blunting the normal wakefulness-
promoting activity of the H/O system. Age-related dysfunction of this system
may also have considerable impact on the regulation of energy metabolism (Willie
et al., 2001).
3.5. Histaminergic system and aging
The histamine (HA) system is known to be involved in the generation and/or

maintenance of wakefulness and a modulator of sleepwakefulness rhythms (Brown
et al., 2001; Haas and Panula, 2003). Although well-known to be produced in mast
cells in the periphery, HA is also synthesized in the CNS from the amino acid
L-histidine by the enzyme histidine decarboxylase (HDC) within neurons of
the tuberomammillary nucleus (TMN) of the hypothalamus. The wakefulness-
promoting eects of Hcrt-1 described above are likely mediated through the HA
system since microinjection of Hcrt-1 directly into the TMN elicits wakefulness,
an eect that is not seen in the H1 receptor knockout mouse (Huang et al., 2001),
and central administration of the H1 antagonist pyrilamine signicantly attenuates
Hcrt-1-induced wakefulness (Yamanaka et al., 2002).
The HA neurons send widespread, diuse axons throughout virtually the entire
brain and four subtypes of HA receptors are known (Brown et al., 2001; Haas and
Panula, 2003). Histamine H1 receptors (H1R) are located postsynaptically; high
densities of these receptors are seen in the hypothalamus and other limbic regions.
Histamine H2 receptors (H2R) are also mainly located postsynaptically and high
densities H2R are found in hippocampus, amygdala, and basal ganglia. Histamine
H3 receptors (H3R) are located on the somata and axon terminals of HA neurons
where they serve as autoreceptors to modulate HA synthesis and release, and are
also located pre- and post-synaptically in other brain regions. Histamine H4
receptors (H4R) are expressed primarily in bone marrow and eosinophils and are
found at very low levels in the brain. To determine whether age-related changes
occur in the HA system in aging, we measured the expression of HDC, H1R, H2R,
and H3R mRNAs in eight brain regions of young (3 months), middle-aged (12 and
18 months) and old (24 months) male C57BL/6 mice (Terao et al., 2004).
Although HDC mRNA levels did not change with age in C57BL/6 mice,
signicant dierences were found in H1R, H2R, and H3R mRNA levels in several
brain regions (Fig. 2B). The most widespread changes were observed in H1R mRNA
levels, which were signicantly lower (2738%) in the cortex, hypothalamus,
hippocampus, and medulla of 24-month-old mice relative to 3-month-old animals.
Age-related changes in H2R mRNA levels were restricted to the pons and cerebellum
and decreased H3R mRNA was found only in the medulla.
Although there have been few studies of the HA system in aging, altered HA
turnover has been found in the cerebrospinal uid (CSF) of aged humans (Prell et al.,
1988; Prell and Green, 1994), and PET scan studies have revealed decreased H1
receptors in the aging human brain (Yanai et al., 1992). Functional studies indicate
that HA-stimulated adenylate cyclase activity is decreased in aged rabbits (Makman
et al., 1978) and K -stimulated HA release is lowered in in vitro experiments in aged
rat hypothalamus (Ferretti et al., 1998).
The age-related changes in HA receptor mRNA reported here may be related to
the decreased alertness seen in aging. The link between the HA system and
wakefulness originates from observations that sedation and drowsiness occur with
those antihistamines (H1 receptor antagonists) that cross the blood-brain barrier and
aect central nervous system activity whereas newer antihistamines that do not cross
the blood-brain barrier do not cause drowsiness (Simons, 1990). Depletion of brain
HA with the synthesis inhibitor alpha-uoromethylhistamine resulted in a reduction
of wakefulness and increased NREM sleep (Kiyono et al., 1985; Lin et al., 1988;
Monti et al., 1988). Central injections of HA or H1 agonists elicited behavioral
arousal (Kalivas, 1982), with a concomitant desynchronization of the electro-
encephalogram (EEG) (Monnier et al., 1970). Activation of the EEG was
accompanied by increased wakefulness and decreased NREM sleep, which was
blocked by the H1 receptor antagonist mepyramine (Tasaka et al., 1989; Monti,
1993; Lin et al., 1996). Increased wakefulness and decreased NREM sleep is also seen
following administration of the H3 autoreceptor antagonist thioperamide, which acts
to block autoinhibition of the HA neurons, while enhanced NREM sleep was seen
after administration of the H3 agonist R-alpha-methylhistamine (Lin et al., 1990;
Monti et al., 1991). The wakefulness-promoting eects of thioperamide were blocked
by the H1 antagonist mepyramine (Lin et al., 1990). In contrast, pharmacological
manipulation of the H2 receptor does not aect sleepwakefulness (Monti et al.,
1990). Together, this evidence indicates that the wakefulness eects of HA are
mediated through H1 and/or H3 receptors.
As indicated above, it is thought that the wakefulness-promoting eects of Hcrt-1
are mediated through the HA system (Huang et al., 2001). Evidence for direct
interactions between the H/O and HA neurotransmitter systems include the
presence of H/O peptidergic bers in the TMN that make synaptic contact with HA
neurons (Peyron et al., 1998; Eriksson et al., 2001), hcrtr2 mRNA in the TMN
(Marcus et al., 2001), excitatory eects of Hcrt-1 on HA neurons in the TMN

(Eriksson et al., 2001; Yamanaka et al., 2002) and reduced levels of HA in the
hcrtr2-mutant canine narcoleptics (Nishino et al., 2001). Since we have previously
shown that hcrtr2 mRNA expression declines in the aged mouse (Terao et al.,
2002b), age-related changes in these two neurotransmitter systems may compound
to result in a disruption of alertness, sleep continuity, and diurnal rhythms of
sleep and wakefulness in the aged.
4. Sleep deprivation, recovery sleep, and gene expression
4.1. Homeostatic regulation of sleep
Numerous studies of sleep deprivation (SD) in both man and animals indicate
that sleep is homeostatically regulated. Sleep loss produces proportional increases in
the drive to sleep, in subsequent sleep expression, and in slow wave activity (SWA)
recorded in the EEG during NREM sleep. This property of homeostatic regulation,
along with a circadian component, has been incorporated into the two-process
model of sleep regulation which addresses the timing of sleep (Borbely, 1982; Daan
et al., 1984). In this model, the homeostatic sleep-related Process S is proposed to
interact with input from the circadian system (Process C) to gate the occurrence of
sleep and wakefulness. Process S is proposed to be a neurochemical process(es)
which begins to build up at the onset of wakefulness. Once a threshold value is
reached, sleep will ensue only if Process C is in the appropriate circadian phase. This
model, although seemingly simplistic, accounts remarkably well for the timing of
sleep in humans and in other mammalian species (Borbely et al., 1981, 1984; Tobler
and Jaggi, 1987; Dijk and Daan, 1989; Deboer et al., 1994; Huber et al., 2000;
Franken et al., 2001).
A number of studies have suggested that EEG SWA, i.e., activity in the 0.54.0 Hz
(delta) range of the EEG during NREM sleep, is related to the underlying Process S.
When SWA in the delta range during sleep is subjected to Fourier analysis, the
spectral power measured has been found to increase in proportion to prior wake
duration in humans and other mammals. This EEG delta power also declines as
sleep time progresses, since Process S presumably diminishes during sleep. Thus,
EEG delta power has been interpreted to represent the cortical manifestation of the
recovery processes from prior waking activities that occurs during sleep (Borbely,
1982; Borbely and Tobler, 1989).
4.2. Gene expression studies of sleep deprivation and recovery sleep
Using cDNA array technology, we investigated a prediction of the two-process

model that sleep after prolonged wakefulness involves a change in macromolecular
synthesis that facilitates neuronal recovery or restoration (Terao et al., 2003b). Since
the SWA measured in the EEG during SWS is generally accepted as being due to
thalamocortical activity, we focused our initial eorts on gene expression in the
cerebral cortex. To measure gene expression, we used two technologies: cDNA
arrays and real-time quantitative RTPCR (TaqMan). The cDNA arrays were used
for initial screening to identify candidate genes since this method enables
measurement of the expression of hundreds to thousands of genes simultaneously.
TaqMan analysis was used to conrm the candidate genes obtained from the arrays
and to evaluate the expression of these genes in other brain regions. In the course of
screening DNA arrays with mRNAs isolated from the cerebral cortex of mice that
had undergone 6 h SD or 6 h SD followed by 4 h recovery sleep (RS; see Fig. 3), two
groups of genes were found to be upregulated: immediate early genes (IEGs) and
heat shock proteins (HSPs).
4.2.1. Expression of immediate early genes (IEGs) during sleep deprivation and
recovery sleep
Since our DNA array studies revealed widespread activation of IEG expression in
the cortex after SD, we examined the expression of 11 IEG family members by
Taqman in seven brain regions to thoroughly evaluate the expression of the IEG/
transcription factor family members in response to SD and RS (Terao et al., 2003a).
During SD, a widespread activation of IEG expression was observed (Fig. 4A and
4B) and, in particular, coordinate expression of c-fos, fosB and nur77 mRNAs at a
regional level across the experimental conditions. In contrast, increased junB and
egr-3 expression was restricted to the cerebral cortex, increased fra-1 expression was
restricted to the basal forebrain, increased fra-2 expression occurred in both the
basal forebrain and cerebellum, and increased egr-1 expression occurred in both the
A B
100 200
EEG Delta Power (%)
Total Sleep Time (%)
p<0.05
80
160
60
40
120
20
0 80
0 12 24 36 48 30 36
Time (h) Time (h)
Fig. 3. A. Hourly total sleep time values (mean S.E.M.) in male C57BL/6 mice (n 10) over a 24 h
period followed by 6 h of sleep deprivation (SD) and a subsequent 18 h recovery period. Black horizontal
bars indicate dark periods, and white bars indicate light period. B. EEG delta power values (mean
S.E.M.) in male C57BL/6 mice (n 10) during hours 3036, the rst 6 h of the recovery period subsequent
to the SD period illustrated in A. Baseline values (mean S.E.M.) are EEG delta power during the same
period on the prior baseline day (ZT 612). Note that sleep intensity, as measured by EEG delta power
during NREM sleep, is signicantly increased for the rst 5.5 h of the recovery period. From Terao et al.
(2003b) with permission.
Sleep Deprivation Recovery Sleep
178
A Cx Cx
c-fos fra-2
fosB Cb
junB Cb
BF c-fos BF
c-fos Th fosB Th
fosB c-fos fra-2
fra-1 fosB
fra-2 Pn Hy Pn
Hy
Md Md
B Cx Cx
egr-1
egr-3 egr-3
A. Terao and T. S. Kilduff

Cb Cb
BF nur77 BF
nur77
egr-1 Th Th
nur77 nur77
Hy Pn Hy Pn
Md Md
C ERp72 GRp78
Cx Cx GRp94
GRp78
Cb Cb
BF
Th GRp78 BF
GRp78 Th
HSP27
GRp94
Hy Pn Hy Pn
HSP27 Md Md
GRp78 GRp78
HSP70-1 GRp94
Fig. 4. Expression of (A) Fos/Jun family genes, (B) other IEGs, and (C) HSP family genes in seven brain areas during sleep deprivation and recovery sleep.
Number of arrows indicates x-fold change as determined by Taqman analysis. " indicates mRNA is signicantly upregulated but the magnitude is 1.5-fold or
less; "" indicates mRNA is signicantly upregulated and the magnitude is 1.5- to 2.5- fold; """ indicates mRNA is signicantly upregulated and the magnitude
is 2.5- to 3.5-fold, etc. Abbreviations: BF, basal forebrain; Cx,cortex; Th; thalamus, Hy; hypothalamus, Pn; pons, Cb; cerebellum, Md; medulla.
cortex and basal forebrain. The IEGs c-jun and junD were invariant across
experimental conditions.
In contrast to the widespread activation of IEGs during SD, very limited changes
in IEG expression occurred during RS with increased expression of fra-2 and egr-3
mRNAs in the cortex being the only signicant changes. In the case of egr-3,
increased expression was observed in the cortex both after 6 h SD and RS (Fig. 4B).
Thus, it is conceivable that the increased levels of egr-3 mRNA observed in the RS
condition could reect persistence of a long-lived mRNA that was induced during
the preceding SD period. On the other hand, egr-3 has been reported to be a late
response gene and to be expressed after some delay in the expression of egr-1
(ODonovan et al., 1998). Therefore, the increased egr-3 expression observed during
RS could also be a consequence of increased Egr-1 expression during SD. However,
egr-1 levels were increased in the basal forebrain during SD and no subsequent
increase in egr-3 mRNA was observed during RS in this brain area (Fig. 4B). A time
course study in which mice are allowed dierent durations of RS after a xed time of
SD would help resolve these possibilities.
In the case of fra-2 mRNA, a signicant increase was only observed during RS
(Fig. 4A). There are many examples of increased fra-2 mRNA and/or Fra-2 protein
expression in the CNS including rhythmic expression in the pineal gland (Baler and
Klein, 1995) or induction in specic brain regions after photic (Schwartz et al., 2000),
osmotic (Miyata et al., 2001) or pharmacological (Nye and Nestler, 1996)
stimulation or after brain injury (Beer et al., 1998). The time course of fra-2
induction has been studied in the SCN after photic stimulation either early or late in
the subjective night. In both cases, fra-2 mRNA is induced as rapidly as c-fos mRNA
but is more long-lived (Schwartz et al., 2000). Similarly, Fra-2 protein is induced in
the SCN with the same time course as Fos. This Fra-2 can either activate or repress
gene transcription depending on its heterodimerization partner (Suzuki et al., 1991;
McCabe et al., 1996; Rutberg et al., 1997) and the extent of its phosphorylation
(Gruda et al., 1994; Murakami et al., 1999). Taken together, these observations
indicate that the increased fra-2 levels observed in the cortex during RS may have
considerable physiological signicance, particularly since fra-2 mRNA is induced
during RS in the apparent absence of c-fos mRNA.
Among the brain regions examined, the cerebral cortex, basal forebrain, and
cerebellum appear to be the most sensitive regions to SD, in terms of both the
number of IEGs induced and the magnitude of the responses. Induction of c-fos
mRNA was greatest in the cerebellum followed by the cortex and then thalamus,
consistent with previous observations in the rat brain after 6 h SD (OHara et al.,
1993). Expression of the transcription factors examined did not change signicantly
in the hypothalamus, pons and medulla; the pons and hypothalamus were reported
to undergo the smallest change in c-fos mRNA in the rat study (medulla was not
measured), although statistics were not used in that study (OHara et al., 1993).
The absence of a signicant change in gene expression in the hypothalamus is
somewhat surprising, given the implication of H/O cells in sleepwake regulation
(Kildu and Peyron, 2000; Willie et al., 2001; Taheri et al., 2002). However, the
hypothalamus is a heterogeneous tissue and there are only about 3000 hypocretin
neurons in this region (Peyron et al., 1998; Kildu and Peyron, 2000), so a change in
IEG expression in these cells may be dicult to detect by Taqman analysis within the
larger mass of the hypothalamus. Furthermore, our previous work showed no
signicant change in preprohypocretin expression in either the mouse or the rat in
response to 6 h SD (Terao et al., 2000). In our study of heat shock proteins
(discussed next), the cortex and basal forebrain were among the most responsive
brain regions to SD in terms of the number of mRNA species that changed whereas
the thalamus and pons were the least responsive. In both studies, the cerebral cortex
was the only region in which elevated mRNA expression was found during RS.
These observations are consistent with the suggestion that the cerebral cortex, a site
of neural plasticity, may undergo important molecular changes during sleep
(Cirelli and Tononi, 2000a).
As is well known, Fos proteins must dimerize to form AP-1 and aect
transcriptional activity. As indicated in Fig. 4A and 4B, a dierent set of mRNAs are
co-induced with c-fos in dierent brain areas and, thus, dimerization pairings may
show regional specicity within the brain. Neurons and glia exhibit basal levels of
both c-Jun and JunD whereas c-Fos and FosB are found at very low levels in
untreated animals. Promoter regions bearing a single AP-1 site are eciently
activated by c-Jun and this activation is inhibited by JunB (Deng and Karin, 1993).
Consequently, to the extent that mRNA induction is reected in protein levels,
the coordinate expression of c-fos, fosB and junB mRNAs in the cortex during SD,
for example, may have considerable functional consequences for the regulation
of target genes bearing AP-1 binding sites. If the restorative function of sleep is at
all related to transcriptional activity, the current results indicate that pathways
other than the Fos/Jun transcriptional regulation are involved. Furthermore, the
region-specic nature of the IEG induction suggests that sleep subserves dierent
functions among brain regions by aecting the synthesis of dierent sets of
macromolecules.
4.2.2. Expression of heat shock proteins (HSPs) during sleep deprivation and recovery
sleep
The other class of genes whose expression was elevated in our DNA array studies
in the mouse cortex after SD were the chaperone/heat shock protein family
members (Fig. 4C). To thoroughly evaluate the expression of the heat shock protein
(HSP) family members in response to SD and RS, we examined the expression of
six HSP family members(erp72, grp78, grp94, hsp27, hsp-70-1, hsp84 ) by Taqman in
seven brain regions (Terao et al., 2003b). Our studies revealed increased mRNA
levels for several HSPs in cortex, basal forebrain, hypothalamus, cerebellum, and
medulla during SD, whereas increased mRNA levels during RS were limited to grp78
and grp94 and were anatomically restricted to the cortex and medulla (Fig. 4C).
Immunohistochemical studies conrmed these results by identifying increased
numbers of GRp78-, GRp94-, and ERp72-immunoreactive cells in the dorsal and
lateral cortex during SD. During RS, however, the number of these cells increased
only in the lateral cortex. In the medulla, increased numbers of GRp94-ir cells were
observed in the nucleus tractus solitarius, dorsal motor nucleus of the vagus, and
the rostroventrolateral medulla during RS. The widespread increase of HSP family
mRNAs in brain during SD may be a neuroprotective response to prolonged
wakefulness. In contrast, the relatively limited HSP family mRNA expression during
RS may be related to the role of HSPs in protein biogenesis and thus to the
restorative function of sleep.
As is evident in Fig. 4C, a dierent combination of HSP mRNAs was upregulated
during SD in each of the brain regions examined. In terms of the number of mRNAs
that changed, the cortex and basal forebrain were among the most responsive brain
regions to SD whereas the thalamus and pons were the least responsive. Dierential
expression of HSP mRNAs was also evident during RS (Fig. 4C). In a previous
study of the response of the rat brain to SD, we reported dierential IEG expression
among brain regions (OHara et al., 1993) and found the cerebellum and cortex to be
the most sensitive brain regions to SD. Future studies using tools that provide a
more encyclopedic view of gene expression among brain regions should allow
denitive determination of dierential sensitivity of brain regions to SD and
localization of where restorative processes occur in the brain during sleep.
What is the physiological signicance of these changes in HSP and GRP
expression in brain during SD and RS? Other than in response to SD (Cirelli and
Tononi, 2000b), the only examples of induction of the GRPs in brain are in
pathological conditions: GRp75, GRp78, GRp94 are induced in brain by ischemia
(Massa et al., 1995; Sharp et al., 1999) and status epilepticus (Little et al., 1996). The
GRp78 and HSP70 are also increased in several brain regions in response to dietary
restriction (Yu and Mattson, 1999; Guo et al., 2000). These GRp78 (BiP, a HSP70
homologue), GRp94 (ERp99, a HSP90 homologue), and ERp72 are glucose-
regulated proteins localized to the endoplasmic reticulum (ER). In other tissues,
GRp78 and GRp94, probably along with ERp72 and GRp170, have been shown to
act as molecular chaperones, playing a role in the proper folding of proteins that
pass through the ER such as thyroglobulin (Kuznetsov et al., 1997), the light and
heavy chains of immunoglobulins (Melnick et al., 1992, 1994), and procollagen
(Ferreira et al., 1994).
There appear to be several pathways for induction of GRPs. Misfolding of large
oligomeric glycoproteins in the ER is a potent stimulus for all of the ER-associated
GRPs, including GRp78, GRp94, GRp170 and ERp72 (Kuznetsov et al., 1997). The
GRPs also share a biochemical feature of being inducible by various compounds that
stress the ER and appear to protect cells against a variety of stimuli. Stimuli known
to induce GRPs include low glucose, agents that aect ER calcium levels (such as the
calcium ionophore A23187 or thapsigargin), and agents that inhibit protein
glycosylation in the ER (such as tunicamycin) (Sharp et al., 1999). Interestingly, the
calcium ionophore A23187 induces the GRPs but inhibits HSP70 expression (Elia
et al., 1996). Another route for induction of GRPs is related to mitogenesis, a
pathway distinct from the misfolding response. This pathway has been proposed to
control expression of the ER chaperones and folding enzymes needed to assist
protein biogenesis in the ER of normal, non-stressed cells (Brewer et al., 1997).
Based on these observations, we suggest that the widespread induction of HSPs and
GRPs in the brain during SD (Fig. 4C) may be a neuroprotective response to cellular
stress accumulating during prolonged wakefulness. In contrast, the relatively

limited HSP mRNA induction during RS (Fig. 4C) may be related to the HSP/GRP
role in protein biogenesis in unstressed cells and thus perhaps to the restorative
function of sleep.
5. Microarray studies of gene expression in the aged brain
Microarrays have become a powerful approach to measure the expression of

thousands of genes in parallel. Gene chip studies related to aging have examined
mRNA levels in the cortex and cerebellum (Lee et al., 2000), subregional expression
of mRNAs in the hippocampus (Zhao et al., 2001, Cho et al., 2002), and the eect of
exercise on gene expression (Tong et al., 2001). Using GeneChips , we measured the
expression of 12,488 genetic elements in the hippocampus of 3, 12, 18, and 24-
month-old male C57BL/6 mice to identify genes whose altered expression could
inuence hippocampal function in advanced age (Terao et al., 2002a). A range of
ages was included to determine whether the gene changes that were observed
occurred gradually or appeared abruptly at a specic time during the aging process.
The results obtained in these basal aging experiments were further validated by
intraperitoneal administration of lipopolysaccharide (LPS) from Pseudomonas
aeruginosa at each of the four ages.
Of the 12,488 gene elements on the GeneChip , 128 elements or 1% of the gene
elements showed age-related changes in 8 of 8 basal samples tested; of these, 14 gene
elements showed increased expression with age. After mice were stimulated with
LPS, 78 gene elements (0.6%) showed age-related changes in 4 of 4 samples tested.
The list of genes that were reproducibly observed to change with age in two basal
experiments as well as after LPS induction includes the major histocompatibility
complex TL region and thymic shared antigen, suggesting T cell activation in the
hippocampus with age. Expression patterns for several of these genes were conrmed
using real-time quantitative PCR. Although the array hybridization was performed
with pooled samples, PCR analysis was performed on individual samples which
allowed statistical comparison by ANOVA. Cytokine (interleukin-1 beta, tumor
necrosis factor) and chemokine (macrophage chemotactic protein-1) expression
increased sharply in 24-month-old mice (Fig. 5). As indicated in Fig. 5, there was
good correlation between the two methods.
6. Conclusions and future directions
In this chapter, we have described the sleep characteristics of the elderly and
introduced the concept of homeostatic sleep regulation. The great diminution of
stage 3 and 4 sleep suggests a problem with generation of the delta frequencies in
the EEG and potentially a problem with sleep homeostasis in the elderly, although
SWA in the delta range can be observed in senior individuals after one nights sleep
deprivation. Animal studies have demonstrated a reduction in mRNA levels for
both hcrtr2 and for HA receptors in selected brain areas. Since the alerting eects
of the H/O neuropeptides are thought to be mediated through the HA system,
A B
2 2
Expression Level
H-2T17
1 1
0 0
60 30
Expression Level
IL-1b
40 20
20 10
0 0
2 20
Expression Level
TNFa
1 10
0 0
150 30
Expression Level
MCP-1
100 20
50 10
0 0
4 1.5
Expression Level
c-fos
1.0
2
0.5
0 0
3 12 18 24 3 12 18 24
Age (Month) Age (Month)
Fig. 5. Comparison of expression patterns with age for ve genes as measured by GeneChip (A) and
real-timequantitative PCR (B). Open squares in both the panels represent data from untreated mice
whereas dark squares represent data from LPS-treated mice. For the GeneChip analysis 3 samples were
pooled per age. Basal data represent an average of two replicate experiments. Data from LPS-treated mice
are from a single experiment. For PCR analysis data represent means SEM of individual samples at
each age. N 5 or 6 for basal samples and 2 or 3 for samples from LPS-treated mice. y statistically
signicant (p < 0.05) compared to 3 months for all basal samples, by TukeyKramer post hoc test.
x statistically signicant (p < 0.05) compared to 3 months in samples from LPS-treated mice, by Tukey
Kramer post hoc test. * statistically signicant (p < 0.05) when samples from untreated and LPS-treated
mice were compared by TukeyKramer post hoc test. From Terao et al. (2002a) with permission.
age-related changes in these two neurotransmitter systems may compound to result

in a disruption of alertness, sleep continuity, and diurnal rhythms of sleep and
wakefulness in the aged. Gene expression studies have been undertaken to identify
the molecular basis of the restorative process that occurs during sleep and to
characterize changes in the brain with aging. Given the dierences in sleep patterns
between young and aged humans and animals, the application of these approaches
to compare the process of homeostatic sleep regulation across the lifespan is likely to
be an area of fruitful investigation.
Acknowledgments
Research supported by NIH grants AG20584, HL59658, and MH61755.
References
Abrahamson, E.E., Leak, R.K., Moore, R.Y., 2001. The suprachiasmatic nucleus projects to posterior
hypothalamic arousal systems. Neuroreport 12, 435440.
Agnew, H.W., Jr., Webb, W.W., Williams, R.L., 1967. Sleep patterns in late middle age males: an EEG
study. Electroencephalogr. Clin. Neurophysiol. 23, 168171.
Baler, R., Klein, D.C., 1995. Circadian expression of transcription factor Fra-2 in the rat pineal gland.
J. Biol. Chem. 270, 2731927325.
Ballinger, C.B., 1976. Subjective sleep disturbance at the menopause. J. Psychosom. Res. 20, 509513.
Beer, J., Mielke, K., Zipp, M., Zimmermann, M., Herdegen, T., 1998. Expression of c-jun, junB, c-fos,
fra-1 and fra-2 mRNA in the rat brain following seizure activity and axotomy. Brain Res. 794,
255266.
Bliwise, D.L., 1993. Sleep in normal aging and dementia. Sleep 16, 4081.
Bliwise, D.L., King, A.C., Harris, R.B., Haskell, W.L., 1992. Prevalence of self-reported poor sleep in a
healthy population aged 5065. Soc. Sci. Med. 34, 4955.
Borbely, A.A., 1982. A two process model of sleep regulation. Hum. Neurobiol. 1, 195204.
Borbely, A.A., Baumann, F., Brandeis, D., Strauch, I., Lehmann, D., 1981. Sleep deprivation: eect on
sleep stages and EEG power density in man. Electroencephalogr. Clin. Neurophysiol. 51, 483495.
Borbely, A.A., Tobler, I., 1989. Endogenous sleep-promoting substances and sleep regulation. Physiol.
Rev. 69, 605670.
Borbely, A.A., Tobler, I., Hanagasioglu, M., 1984. Eect of sleep deprivation on sleep and EEG power
spectra in the rat. Behav. Brain Res. 14, 171182.
Bourgin, P., Huitron-Resendiz, S., Spier, A.D., Fabre, V., Morte, B., Criado, J.R., Sutclie, J.G.,
Henriksen, S.J., de Lecea, L., 2000. Hypocretin-1 modulates REM sleep through activation of locus
coeruleus neurons. J. Neurosci. 20, 77607765.
Bowersox, S.S., Floyd, T., Dement, W.C., 1984. Electroencephalogram during sleep in the cat: age eects
on slow-wave activity. Sleep 7, 380384.
Brewer, J.W., Cleveland, J.L., Hendershot, L.M., 1997. A pathway distinct from the mammalian unfolded
protein response regulates expression of endoplasmic reticulum chaperones in non- stressed cells.
Embo. J. 16, 72077216.
Brown, R.E., Stevens, D.R., Haas, H.L., 2001. The physiology of brain histamine. Prog. Neurobiol. 63,
637672.
Carskadon, M.A., Brown, E.D., Dement, W.C., 1982. Sleep fragmentation in the elderly: relationship to
daytime sleep tendency. Neurobiol. Aging 3, 321327.
Chemelli, R.M., Willie, J.T., Sinton, C.M., Elmquist, J.K., Scammell, T., Lee, C., Richardson, J.A.,
Williams, S.C., Xiong, Y., Kisanuki, Y., Fitch, T.E., Nakazato, M., Hammer, R.E., Saper, C.B.,
Yanagisawa, M., 1999. Narcolepsy in orexin knockout mice: molecular genetics of sleep regulation.
Cell 98, 437451.
Cho, K.S., Choi, J., Ha, C.M., Son, Y.J., Choi, W.S., Lee, B.J., 2002. Comparison of gene expression in
old versus young rat hippocampus by cDNA array. Neuroreport 13, 285289.
Cirelli, C., Tononi, G., 2000a. Dierential expression of plasticity-related genes in waking and sleep and
their regulation by the noradrenergic system. J. Neurosci. 20, 91879194.
Cirelli, C., Tononi, G., 2000b. Gene expression in the brain across the sleep-waking cycle(1). Brain Res.
885, 303321.
Czeisler, C.A., Duy, J.F., Shanahan, T.L., Brown, E.N., Mitchell, J.F., Rimmer, D.W., Ronda, J.M.,
Silva, E.J., Allan, J.S., Emens, J.S., Dijk, D.J., Kronauer, R.E., 1999. Stability, precision, and
near-24-hour period of the human circadian pacemaker. Science 284, 21772181.
Daan, S., Beersma, D.G., Borbely, A.A., 1984. Timing of human sleep: recovery process gated by a
circadian pacemaker. Am. J. Physiol. 246, R161R183.
Davis, F.C., Viswanathan, N., 1998. Stability of circadian timing with age in Syrian hamsters. Am. J.
Physiol. 275, R960R968.
De Benedictis, G., Rose, G., Carrieri, G., De Luca, M., Falcone, E., Passarino, G., Bonafe, M., Monti, D.,
Baggio, G., Bertolini, S., Mari, D., Mattace, R., Franceschi, C., 1999. Mitochondrial DNA inherited
variants are associated with successful aging and longevity in humans. FASEB J. 13, 15321536.
de Lecea, L., Kildu, T.S., Peyron, C., Gao, X.-B., Foye, P.E., Danielson, P.E., Fukuhara, C.,
Battenberg, E.L.F., Gautvik, V.T., Bartlett, F.S., II, Frankel, W.N., van den Pol, A.N., Bloom, F.E.,
Gautvik, K.M., Sutclie, J.G., 1998. The hypocretins: hypothalamus-specic peptides with
neuroexcitatory activity. Proc. Natl. Acad. Sci. USA. 95, 322327.
Deboer, T., Franken, P., Tobler, I., 1994. Sleep and cortical temperature in the Djungarian hamster under
baseline conditions and after sleep deprivation. J. Comp. Physiol. [A]. 174, 145155.
Deng, T., Karin, M., 1993. JunB diers from c-Jun in its DNA-binding and dimerization domains, and
represses c-Jun by formation of inactive heterodimers. Genes Dev. 7, 479490.
Dijk, D.J., Beersma, D.G., van den Hoofdakker, R.H., 1989. All night spectral analysis of EEG sleep in
young adult and middle-aged male subjects. Neurobiol. Aging 10, 677682.
Dijk, D.J., Daan, S., 1989. Sleep EEG spectral analysis in a diurnal rodent: Eutamias sibiricus. J. Comp.
Physiol. [A]. 165, 205215.
Duy, J.F., Viswanathan, N., Davis, F.C., 1999. Free-running circadian period does not shorten with age
in female Syrian hamsters. Neurosci. Lett. 271, 7780.
Edgar, D.M., 1994. Sleep-wake circadian rhythms and aging: potential etiologies and relevance to
age-related changes in integrated physiological systems. Neurobiol. Aging 15, 499501.
Edgar, D.M., Dement, W.C., Fuller, C.A., 1993. Eect of SCN lesions on sleep in squirrel monkeys:
evidence for opponent processes in sleep-wake regulation. J. Neurosci. 13, 10651079.
Ehlers, C.L., Kupfer, D.J., 1989. Eects of age on delta and REM sleep parameters. Electroencephalogr.
Clin. Neurophysiol. 72, 118125.
Elia, G., De Marco, A., Rossi, A., Santoro, M.G., 1996. Inhibition of HSP70 expression by calcium
ionophore A23187 in human cells. An eect independent of the acquisition of DNA-binding activity by
the heat shock transcription factor. J. Biol. Chem. 271, 1611116118.
Eriksson, K.S., Sergeeva, O., Brown, R.S., Haas, H.L., 2001. Orexin/hypocretin excites the histaminergic
neurons of the tuberomammillary nucleus. J. Neurosci. 21, 92739279.
Feinberg, I., Koresko, R.L., Heller, N., 1967. EEG sleep patterns as a function of normal and pathological
aging in man. J. Psychiatr. Res. 5, 107144.
Ferreira, L.R., Norris, K., Smith, T., Hebert, C., Sauk, J.J., 1994. Association of Hsp47, Grp78, and
Grp94 with procollagen supports the successive or coupled action of molecular chaperones. J. Cell
Biochem. 56, 518526.
Ferretti, C., Blengio, M., Ghi, P., Adage, T., Portaleone, P., Ricci Gamalero, S., 1998. Hypothalamic
histamine release in normal and stressed rats is aected by sex and aging. Pharmacol. Biochem. Behav.
59, 255260.
Franken, P., Chollet, D., Tafti, M., 2001. The homeostatic regulation of sleep need is under genetic
control. J. Neurosci. 21, 26102621.
Fujiki, N., Yoshida, Y., Ripley, B., Honda, K., Mignot, E., Nishino, S., 2001. Changes in CSF hypocretin-1
(orexin A) levels in rats across 24 hours and in response to food deprivation. Neuroreport 12, 993997.
Gerashchenko, D., Kohls, M.D., Greco, M.A., Waleh, N.S., Salin-Pascual, R., Kildu, T.S., Lappi, D.A.,
Shiromani, P.J., 2001. Hypocretin-saporin induced lesion of the lateral hypothalamus produces
narcoleptic-like sleep behavior in the rat. J. Neurosci. 21, 72737283.
Gervasoni, D., Darracq, L., Fort, P., Souliere, F., Chouvet, G., Luppi, P.H., 1998. Electrophysiological
evidence that noradrenergic neurons of the rat locus coeruleus are tonically inhibited by GABA during
sleep. Eur. J. Neurosci. 10, 964970.
Gritti, I., Mainville, L., Jones, B.E., 1994. Projections of GABAergic and cholinergic basal forebrain and
GABAergic preoptic-anterior hypothalamic neurons to the posterior lateral hypothalamus of the rat.
J. Comp. Neurol. 339, 251268.
Gruda, M.C., Kovary, K., Metz, R., Bravo, R., 1994. Regulation of Fra-1 and Fra-2 phosphorylation
diers during the cell cycle of broblasts and phosphorylation in vitro by MAP kinase aects DNA
binding activity. Oncogene 9, 25372547.
Guo, Z., Ersoz, A., Buttereld, D.A., Mattson, M.P., 2000. Benecial eects of dietary restriction
on cerebral cortical synaptic terminals: preservation of glucose transport and mitochondrial
function after exposure to amyloid -peptide and oxidative and metabolic insults. J. Neurochem. 75,
314320.
Haas, H., Panula, P., 2003. The role of histamine and the tuberomamillary nucleus in the nervous system.
Nat. Rev. Neurosci. 4, 121130.
Hagan, J.J., Leslie, R.A., Patel, S., Evans, M.L., Wattam, T.A., Holmes, S., Benham, C.D., Taylor, S.G.,
Routledge, C., Hemmati, P., Munton, R.P., Ashmeade, T.E., Shah, A.S., Hatcher, J.P., Hatcher, P.D.,
Jones, D.N., Smith, M.I., Piper, D.C., Hunter, A.J., Porter, R.A., Upton, N., 1999. Orexin A activates
locus coeruleus cell ring and increases arousal in the rat. Proc. Natl. Acad. Sci. USA. 96,
1091110916.
Hara, J., Beuckmann, C.T., Nambu, T., Willie, J.T., Chemelli, R.M., Sinton, C.M., Sugiyama, F.,
Yagami, K., Goto, K., Yanagisawa, M., Sakurai, T., 2001. Genetic ablation of orexin neurons in mice
results in narcolepsy, hypophagia, and obesity. Neuron. 30, 345354.
Horvath, T.L., Peyron, C., Diano, S., Ivanov, A., Aston-Jones, G., Kildu, T.S., van Den Pol, A.N., 1999.
Hypocretin (orexin) activation and synaptic innervation of the locus coeruleus noradrenergic system. J.
Comp. Neurol. 415, 145159.
Huang, Z.L., Qu, W.M., Li, W.D., Mochizuki, T., Eguchi, N., Watanabe, T., Urade, Y., Hayaishi, O.,
2001. Arousal eect of orexin A depends on activation of the histaminergic system. Proc. Natl. Acad.
Sci. USA. 98, 99659970.
Huber, R., Deboer, T., Tobler, I., 2000. Eects of sleep deprivation on sleep and sleep EEG in three mouse
strains: empirical data and simulations. Brain Res. 857, 819.
Ida, T., Nakahara, K., Katayama, T., Murakami, N., Nakazato, M., 1999. Eect of lateral cerebroven-
tricular injection of the appetite- stimulating neuropeptide, orexin and neuropeptide Y, on the various
behavioral activities of rats. Brain Res. 821, 526529.
Juji, T., Satake, M., Honda, Y., Doi, Y., 1984. HLA antigens in Japanese patients with narcolepsy. All the
patients were DR2 positive. Tissue Antigens. 24, 316319.
Kahn, E., 1970. Age-related changes in sleep characteristics. Int. Psychiatry Clin. 7, 2529.
Kahn, E., Fisher, C., 1969. The sleep characteristics of the normal aged male. J. Nerv. Ment. Dis. 148,
477494.
Kales, A., Wilson, T., Kales, J.D., Jacobson, A., Paulson, M.J., Kollar, E., Walter, R.D., 1967.
Measurements of all-night sleep in normal elderly persons: eects of aging. J. Am. Geriatr. Soc. 15,
405414.
Kalivas, P.W., 1982. Histamine-induced arousal in the conscious and pentobarbital-pretreated rat. J.
Pharmacol. Exp. Ther. 222, 3742.
Karacan, I., Thornby, J.I., Anch, M., Holzer, C.E., Warheit, G.J., Schwab, J.J., Williams, R.L., 1976.
Prevalence of sleep disturbance in a primarily urban Florida County. Soc. Sci. Med. 10, 239244.
Kildu, T.S., Peyron, C., 2000. The hypocretin/orexin ligand-receptor system: Implications for sleep and
sleep disorders. Trends Neurosci. 23, 359365.
Kiyono, S., Seo, M.L., Shibagaki, M., Watanabe, T., Maeyama, K., Wada, H., 1985. Eects of alpha-
uoromethylhistidine on sleep-waking parameters in rats. Physiol. Behav. 34, 615617.
Kunii, K., Yamanaka, A., Nambu, T., Matsuzaki, I., Goto, K., Sakurai, T., 1999. Orexins/hypocretins
regulate drinking behaviour. Brain Res. 842, 256261.
Kuznetsov, G., Chen, L.B., Nigam, S.K., 1997. Multiple molecular chaperones complex with misfolded
large oligomeric glycoproteins in the endoplasmic reticulum. J. Biol. Chem. 272, 30573063.
Lee, C.K., Weindruch, R., Prolla, T.A., 2000. Gene-expression prole of the ageing brain in mice. Nat.
Genet. 25, 294297.
Li, H., Satino, E., 1995. Changes in circadian rhythms of body temperature and sleep in old rats. Am. J.
Physiol. 269, R208R214.
Lin, J.S., Hou, Y., Sakai, K., Jouvet, M., 1996. Histaminergic descending inputs to the mesopontine
tegmentum and their role in the control of cortical activation and wakefulness in the cat. J. Neurosci.
16, 15231537.
Lin, J.S., Sakai, K., Jouvet, M., 1988. Evidence for histaminergic arousal mechanisms in the hypo-
thalamus of cat. Neuropharmacology 27, 111122.
Lin, J.S., Sakai, K., Vanni-Mercier, G., Arrang, J.M., Garbarg, M., Schwartz, J.C., Jouvet, M., 1990.
Involvement of histaminergic neurons in arousal mechanisms demonstrated with H3-receptor
ligands in the cat. Brain Res. 523, 325330.
Lin, L., Faraco, J., Li, R., Kadotani, H., Rogers, W., Lin, X., Qiu, X., de Jong, P.J., Nishino, S.,
Mignot, E., 1999. The sleep disorder canine narcolepsy is caused by a mutation in the hypocretin
(orexin) receptor 2 gene. Cell 98, 365376.
Little, E., Tocco, G., Baudry, M., Lee, A.S., Schreiber, S.S., 1996. Induction of glucose-regulated
protein (glucose-regulated protein 78/BiP and glucose-regulated protein 94) and heat shock
protein 70 transcripts in the immature rat brain following status epilepticus. Neuroscience 75,
209219.
Lopez, M., Senaris, R., Gallego, R., Garcia-Caballero, T., Lago, F., Seoane, L., Casanueva, F.,
Dieguez, C., 1999. Orexin receptors are expressed in the adrenal medulla of the rat. Endocrinology 140,
59915994.
Makman, M.H., Ahn, H.S., Thal, L.J., Dvorkin, B., Horowitz, S.G., Sharpless, N.S., Rosenfeld, M., 1978.
Biogenic amine-stimulated adenylate cyclase and spiroperidol-binding sites in rabbit brain: evidence
for selective loss of receptors with aging. Adv. Exp. Med. Biol. 113, 211230.
Marcus, J.N., Aschkenasi, C.J., Lee, C.E., Chemelli, R.M., Saper, C.B., Yanagisawa, M., Elmquist, J.K.,
2001. Dierential expression of orexin receptors 1 and 2 in the rat brain. J. Comp. Neurol. 435, 625.
Massa, S.M., Longo, F.M., Zuo, J., Wang, S., Chen, J., Sharp, F.R., 1995. Cloning of rat grp75,
an hsp70-family member, and its expression in normal and ischemic brain. J. Neurosci. Res. 40,
807819.
McCabe, L.R., Banerjee, C., Kundu, R., Harrison, R.J., Dobner, P.R., Stein, J.L., Lian, J.B., Stein, G.S.,
1996. Developmental expression and activities of specic fos and jun proteins are functionally related
to osteoblast maturation: role of Fra- 2 and Jun D during dierentiation. Endocrinology 137,
43984408.
Melnick, J., Aviel, S., Argon, Y., 1992. The endoplasmic reticulum stress protein GRP94, in addition to
BiP, associates with unassembled immunoglobulin chains. J. Biol. Chem. 267, 2130321306.
Melnick, J., Dul, J.L., Argon, Y., 1994. Sequential interaction of the chaperones BiP and GRP94 with
immunoglobulin chains in the endoplasmic reticulum. Nature 370, 373375.
Mendelson, W.B., Bergmann, B.M., 1999a. Age-related changes in sleep in the rat. Sleep. 22, 145150.
Mendelson, W.B., Bergmann, B.M., 1999b. EEG delta power during sleep in young and old rats.
Neurobiol. Aging 20, 669673.
Methippara, M.M., Alam, M.N., Szymusiak, R., McGinty, D., 2000. Eects of lateral preoptic area
application of orexin-A on sleep-wakefulness. Neuroreport 11, 34233426.
Mignot, E., 1998. Genetic and familial aspects of narcolepsy. Neurology. 50, S16S22.
Mignot, E., Tafti, M., Dement, W.C., Grumet, F.C., 1995. Narcolepsy and immunity. Adv.
Neuroimmunol. 5, 2337.
Mitsuma, T., Hirooka, Y., Mori, Y., Kayama, M., Adachi, K., Rhue, N., Ping, J., Nogimori, T., 1999.
Eects of orexin A on thyrotropin-releasing hormone and thyrotropin secretion in rats. Horm. Metab.
Res. 31, 606609.
Miyata, S., Tsujioka, H., Itoh, M., Matsunaga, W., Kuramoto, H., Kiyohara, T., 2001. Time course of
Fos and Fras expression in the hypothalamic supraoptic neurons during chronic osmotic stimulation.
Brain Res. Mol. Brain Res. 90, 3947.
Monnier, M., Sauer, R., Hatt, A.M., 1970. The activating eect of histamine on the central nervous
system. Int. Rev. Neurobiol. 12, 265305.
Monti, J.M., 1993. Involvement of histamine in the control of the waking state. Life Sci. 53, 13311338.
Monti, J.M., D Angelo, L., Jantos, H., Pazos, S., 1988. Eects of a-uoromethylhistidine on sleep and
wakefulness in the rat. Short note. J. Neural. Transm. 72, 141145.
Monti, J.M., Jantos, H., Boussard, M., Altier, H., Orellana, C., Olivera, S., 1991. Eects of selective
activation or blockade of the histamine H3 receptor on sleep and wakefulness. Eur. J. Pharmacol. 205,
283287.
Monti, J.M., Orellana, C., Boussard, M., Jantos, H., Olivera, S., 1990. Sleep variables are unaltered by
zolantidine in rats: are histamine H2-receptors not involved in sleep regulation? Brain Res. Bull. 25,
229231.
Murakami, M., Ui, M., Iba, H., 1999. Fra-2-positive autoregulatory loop triggered by mitogen-activated
protein kinase (MAPK) and Fra-2 phosphorylation sites by MAPK. Cell Growth Dier. 10, 333342.
Nauta, W.J.H., 1946. Hypothalamic regulation of sleep in rats. An experimental study. J. Neurophysiol. 9,
285316.
Naylor, E., Buxton, O.M., Bergmann, B.M., Easton, A., Zee, P.C., Turek, F.W., 1998. Eects of aging on
sleep in the golden hamster. Sleep 21, 687693.
Nishino, S., Fujiki, N., Ripley, B., Sakurai, E., Kato, M., Watanabe, T., Mignot, E., Yanai, K., 2001.
Decreased brain histamine content in hypocretin/orexin receptor-2 mutated narcoleptic dogs.
Neurosci. Lett. 313, 125128.
Nishino, S., Ripley, B., Overeem, S., Lammers, G.J., Mignot, E., 2000. Hypocretin (orexin) deciency in
human narcolepsy. Lancet. 355, 3940.
Nitz, D., Siegel, J., 1997a. GABA release in the dorsal raphe nucleus: role in the control of REM sleep.
Am. J. Physiol. 273, R451R455.
Nitz, D., Siegel, J.M., 1996. GABA release in posterior hypothalamus across sleep-wake cycle. Am. J.
Physiol. 271, R1707R1712.
Nitz, D., Siegel, J.M., 1997b. GABA release in the locus coeruleus as a function of sleep/wake state.
Neuroscience 78, 795801.
Nye, H.E., Nestler, E.J., 1996. Induction of chronic Fos-related antigens in rat brain by chronic morphine
administration. Mol. Pharmacol. 49, 636645.
ODonovan, K.J., Wilkens, E.P., Baraban, J.M., 1998. Sequential expression of Egr-1 and Egr-3 in
hippocampal granule cells following electroconvulsive stimulation. J. Neurochem. 70, 12411248.
OHara, B.F., Young, K.A., Watson, F.L., Heller, H.C., Kildu, T.S., 1993. Immediate early gene
expression in brain during sleep deprivation: preliminary observations. Sleep 16, 17.
Peyron, C., Faraco, J., Rogers, W., Ripley, B., Overeem, S., Charnay, Y., Nevsimalova, S., Aldrich, M.,
Reynolds, D., Albin, R., Li, R., Hungs, M., Pedrazzoli, M., Padigaru, M., Kucherlapati, M., Fan, J.,
Maki, R., Lammers, G.J., Bouras, C., Kucherlapati, R., Nishino, S., Mignot, E., 2000. A mutation in a
case of early onset narcolepsy and a generalized absence of hypocretin peptides in human narcoleptic
brains. Nat. Med. 6, 991997.
Peyron, C., Tighe, D.K., van den Pol, A.N., de Lecea, L., Heller, H.C., Sutclie, J.G., Kildu, T.S., 1998.
Neurons containing hypocretin (orexin) project to multiple neuronal systems. J. Neurosci. 18,
999610015.
Piper, D.C., Upton, N., Smith, M.I., Hunter, A.J., 2000. The novel brain neuropeptide, orexin-A,
modulates the sleep-wake cycle of rats. Eur. J. Neurosci. 12, 726730.
Prell, G.D., Green, J.P., 1994. Measurement of histamine metabolites in brain and cerebrospinal uid
provides insights into histaminergic activity. Agents Actions. 41 Spec. No, C5C8.
Prell, G.D., Khandelwal, J.K., Burns, R.S., LeWitt, P.A., Green, J.P., 1988. Elevated levels of histamine
metabolites in cerebrospinal uid of aging, healthy humans. Compr. Gerontol. [A]. 2, 114119.
Prinz, P., 1977. Sleep patterns in the healthy aged: relationship with intellectual function. J. Gerontol. 32,
179186.
Prinz, P.N., 1995. Sleep and sleep disorders in older adults. J. Clin. Neurophysiol. 12, 139146.
Pu, S., Jain, M.R., Kalra, P.S., Kalra, S.P., 1998. Orexins, a novel family of hypothalamic neuropeptides,
modulate pituitary luteinizing hormone secretion in an ovarian steroid-dependent manner. Regul.
Pept. 78, 133136.
Reynolds, C.F., Hoch, C.C., Buysse, D.J., Monk, T.H., Houck, P.R., Kupfer, D.J., 1993. Symposium:
Normal and abnormal REM sleep regulation: REM sleep in successful, usual, and pathological aging:
the Pittsburgh experience 19801993. J. Sleep Res. 2, 203210.
Rosenberg, R.S., Zepelin, H., Rechtschaen, A., 1979. Sleep in young and old rats. J. Gerontol. 34,
525532.
Rutberg, S.E., Saez, E., Lo, S., Jang, S.I., Markova, N., Spiegelman, B.M., Yuspa, S.H., 1997. Opposing
activities of c-Fos and Fra-2 on AP-1 regulated transcriptional activity in mouse keratinocytes induced
to dierentiate by calcium and phorbol esters. Oncogene 15, 13371346.
Sakurai, T., Amemiya, A., Ishii, M., Matsuzaki, I., Chemelli, R.M., Tanaka, H., Williams, S.C.,
Richardson, J.A., Kozlowski, G.P., Wilson, S., Arch, J.R., Buckingham, R.E., Haynes, A.C.,
Carr, S.A., Annan, R.S., McNulty, D.E., Liu, W.S., Terrett, J.A., Elshourbagy, N.A., Bergsma, D.J.,
Yanagisawa, M., 1998. Orexins and orexin receptors: a family of hypothalamic neuropeptides and
G protein-coupled receptors that regulate feeding behavior. Cell 92, 573585.
Samson, W.K., Gosnell, B., Chang, J.K., Resch, Z.T., Murphy, T.C., 1999. Cardiovascular regulatory
actions of the hypocretins in brain. Brain Res. 831, 248253.
Satino, E., 1998. Patterns of circadian body temperature rhythms in aged rats. Clin. Exp. Pharmacol.
Physiol. 25, 135140.
Scammell, T.E., Estabrooke, V., I, McCarthy, M.T., Chemelli, R.M., Yanagisawa, M., Miller, M.S.,
Saper, C.B., 2000. Hypothalamic arousal regions are activated during modanil-induced wakefulness.
J. Neurosci. 20, 86208628.
Schwartz, W.J., Carpino, A., Jr., de la Iglesia, H.O., Baler, R., Klein, D.C., Nakabeppu, Y., Aronin, N.,
2000. Dierential regulation of fos family genes in the ventrolateral and dorsomedial subdivisions of
the rat suprachiasmatic nucleus. Neuroscience 98, 535547.
Sharp, F.R., Massa, S.M., Swanson, R.A., 1999. Heat-shock protein protection. Trends Neurosci. 22,
9799.
Shirasaka, T., Nakazato, M., Matsukura, S., Takasaki, M., Kannan, H., 1999. Sympathetic and
cardiovascular actions of orexins in conscious rats. Am. J. Physiol. 277, R1780R1785.
Shiromani, P.J., Lu, J., Wagner, D., Thakkar, J., Greco, M.A., Basheer, R., Thakkar, M., 2000.
Compensatory sleep response to 12 h wakefulness in young and old rats. Am. J. Physiol. Regul. Integr.
Comp. Physiol. 278, R125R133.
Siegel, J.M., 1999. Narcolepsy: a key role for hypocretins (orexins). Cell 98, 409412.
Simons, F.E., 1990. New H1-receptor antagonists: clinical pharmacology. Clin. Exp. Allergy 20 Suppl. 2,
1924.
Steininger, T.L., Alam, M.N., Gong, H., Szymusiak, R., McGinty, D., 1999. Sleep-waking discharge of
neurons in the posterior lateral hypothalamus of the albino rat. Brain Res. 840, 138147.
Suzuki, T., Okuno, H., Yoshida, T., Endo, T., Nishina, H., Iba, H., 1991. Dierence in transcriptional
regulatory function between c-Fos and Fra- 2. Nucleic Acids Res. 19, 55375542.
Taheri, S., Zeitzer, J.M., Mignot, E., 2002. The role of hypocretins (orexins) in sleep regulation and
narcolepsy. Annu. Rev. Neurosci. 25, 283313.
Takahashi, N., Okumura, T., Yamada, H., Kohgo, Y., 1999. Stimulation of gastric acid secretion by
centrally administered orexin- A in conscious rats. Biochem. Biophys. Res. Commun. 254,
623627.
Tamura, T., Irahara, M., Tezuka, M., Kiyokawa, M., Aono, T., 1999. Orexins, orexigenic hypothalamic
neuropeptides, suppress the pulsatile secretion of luteinizing hormone in ovariectomized female rats.
Biochem. Biophys. Res. Commun. 264, 759762.
Tani, Y., Ishihara, T., 1988. Changes in EEG associated with sleep-awake behavior in young adult versus
aged adult Fischer-344 rats. Physiol. Behav. 44, 389392.
Tasaka, K., Chung, Y.H., Sawada, K., 1989. Excitatory eect of histamine on EEGs of the cortex and
thalamus in rats. Agents Actions 27, 127130.
Terao, A., Apte-Deshpande, A., Dousman, L., Morairty, S., Eynon, B.P., Kildu, T.S., Freund, Y.R.,
2002a. Immune response gene expression increases in the aging murine hippocampus.
J. Neuroimmunol. 132, 99112.
Terao, A., Apte-Deshpande, A., Morairty, S., Freund, Y.R., Kildu, T.S., 2002b. Age-related decline in
hypocretin (orexin) receptor 2 mRNA levels in the mouse brain. Neurosci. Lett. 332, 190194.
Terao, A., Greco, M.A., Davis, R.W., Heller, H.C., Kildu, T.S., 2003a. Region-specic changes in
immediate early gene expression in response to sleep deprivation and recovery sleep in the mouse brain.
Neuroscience 120, 11151124.
Terao, A., Peyron, C., Ding, J., Wurts, S.W., Edgar, D.M., Heller, H.C., Kildu, T.S., 2000. Prepro-
hypocretin (prepro-orexin) expression is unaected by short-term sleep deprivation in rats and mice.
Sleep 23, 867874.
Terao, A., Steininger, T.L., Hyder, K., Apte-Deshpande, A., Ding, J., Rishipathak, D., Davis, R.W.,
Heller, H.C., Kildu, T.S., 2003b. Dierential increase in the expression of heat shock protein family
members during sleep deprivation and during sleep. Neuroscience 116, 187200.
Terao, A., Steininger, T.L., Morairty, S.R., Kildu, T.S., 2004. Age-related changes in histamine receptor
mRNA levels in the mouse brain. Neurosci. Lett. 355, 8184.
Thannickal, T., Moore, R.Y., Nienhuis, R., Ramanathan, L., Gulyani, S., Aldrich, M., Cornford, M.,
Siegel, J.M., 2000. Reduced number of hypocretin neurons in human narcolepsy. Neuron 27, 469474.
Tobler, I., Jaggi, K., 1987. Sleep and EEG spectra in the Syrian hamster (Mesocricetus auratus) under
baseline conditions and following sleep deprivation. J. Comp. Physiol. [A]. 161, 449459.
Tong, L., Shen, H., Perreau, V.M., Balazs, R., Cotman, C.W., 2001. Eects of exercise on gene-expression
prole in the rat hippocampus. Neurobiol. Dis. 8, 10461056.
van den Pol, A.N., Gao, X.B., Obrietan, K., Kildu, T.S., Belousov, A.B., 1998. Presynaptic and
postsynaptic actions and modulation of neuroendocrine neurons by a new hypothalamic peptide,
hypocretin/orexin. J. Neurosci. 18, 79627971.
Van Gool, W.A., Mirmiran, M., 1983. Age-related changes in the sleep pattern of male adult rats. Brain
Res. 279, 394398.
van Gool, W.A., Witting, W., Mirmiran, M., 1987. Age-related changes in circadian sleep-wakefulness
rhythms in male rats isolated from time cues. Brain Res. 413, 384387.
van Someren, E.J., Mirmiran, M., Swaab, D.F., 1993. Non-pharmacological treatment of sleep and wake
disturbances in aging and Alzheimers disease: chronobiological perspectives. Behav. Brain Res. 57,
235253.
Vanni-Mercier, G., Sakai, K., Jouvet, M., 1984. Specic neurons for wakefulness in the posterior
hypothalamus in the cat. C R Acad. Sci. III. 298, 195200.
Webb, W.B., Swinburne, H., 1971. An observational study of sleep of the aged. Percept Mot. Skills. 32,
895898.
Welsh, D.K., Richardson, G.S., Dement, W.C., 1986. Eect of age on the circadian pattern of sleep and
wakefulness in the mouse. J. Gerontol. 41, 579586.
Willie, J.T., Chemelli, R.M., Sinton, C.M., Tokita, S., Williams, S.C., Kisanuki, Y.Y., Marcus, J.N.,
Lee, C., Elmquist, J.K., Kohlmeier, K.A., Leonard, C.S., Richardson, J.A., Hammer, R.E.,
Yanagisawa, M., 2003. Distinct narcolepsy syndromes in Orexin receptor-2 and Orexin null
mice: molecular genetic dissection of Non-REM and REM sleep regulatory processes. Neuron 38,
715730.
Willie, J.T., Chemelli, R.M., Sinton, C.M., Yanagisawa, M., 2001. To eat or to sleep? Orexin in the
regulation of feeding and wakefulness. Annu. Rev. Neurosci. 24, 429458.
Witting, W., Mirmiran, M., Bos, N.P., Swaab, D.F., 1993. Eect of light intensity on diurnal sleep-wake
distribution in young and old rats. Brain Res. Bull. 30, 157162.
Yamanaka, A., Tsujino, N., Funahashi, H., Honda, K., Guan, J.L., Wang, Q.P., Tominaga, M., Goto, K.,
Shioda, S., Sakurai, T., 2002. Orexins activate histaminergic neurons via the orexin 2 receptor.
Biochem. Biophys. Res. Commun. 290, 12371245.
Yanai, K., Watanabe, T., Meguro, K., Yokoyama, H., Sato, I., Sasano, H., Itoh, M., Iwata, R.,
Takahashi, T., Ido, T., 1992. Age-dependent decrease in histamine H1 receptor in human brains
revealed by PET. Neuroreport 3, 433436.
Ylikoski, R., Ylikoski, A., Keskivaara, P., Tilvis, R., Sulkava, R., Erkinjuntti, T., 1999. Heterogeneity of
cognitive proles in aging: successful aging, normal aging, and individuals at risk for cognitive decline.
Eur. J. Neurol. 6, 645652.
Yoshida, Y., Fujiki, N., Nakajima, T., Ripley, B., Matsumura, H., Yoneda, H., Mignot, E., Nishino, S.,
2001. Fluctuation of extracellular hypocretin-1 (orexin A) levels in the rat in relation to the lightdark
cycle and sleep-wake activities. Eur. J. Neurosci. 14, 10751081.
Yu, Z.F., Mattson, M.P., 1999. Dietary restriction and 2-deoxyglucose administration reduce focal
ischemic brain damage and improve behavioral outcome: evidence for a preconditioning mechanism.
J. Neurosci. Res. 57:830839.
Zepelin, H., Whitehead, W.E., Rechtschaen, A., 1972. Aging and sleep in the albino rat. Behav. Biol. 7,
6574.
Zhao, X., Lein, E.S., He, A., Smith, S.C., Aston, C., Gage, F.H., 2001. Transcriptional proling reveals
strict boundaries between hippocampal subregions. J. Comp. Neurol. 441, 187196.
List of Contributors
J. Lee Kavanau University of California

Department of Biology
Los Angeles, CA 90095-1606, USA
Phone: 1-310-825-3474 Fax: 1-310-206-3987
Email: lkavanau@biology.ucla.edu
Jack D. Edinger Psychology Service

VA Medical Center
508 Fulton Street
Durham, NC 27705, USA
Phone: 1-919-286-0411 Fax: 1-919-416-5832
Email: jack.edinger@duke.edu
David Gozal Kosiar Childrens Hospital Research Institute

Department of Pediatrics
570 S. Preston St. Ste. 204
Louisville, KY 40202-3830, USA
Phone: 1-502-852-2323 Fax: 1-502-852-2215
Email: david.gozal@louisville.edu
Alon Y. Avidan Department of Neurology

Michael S. Aldrich Sleep Disorders Laboratory
University of Michigan Health Systems
8D-8702 University Hopsital, Box 0117
1500 East Medical Center Drive
Ann Arbor, MI 48109-0117, USA
Phone: 1-734-647-9064 Fax: 1-734-647-9065
Email: avidana@umich.edu
Jan Born University of Luebeck

Institute of Neuroendocrinology
Haus 23a Ratzeburger Allee 160
23538 Luebeck, Germany
Phone: 49-4515 003641 Fax: 49-4515 003640
Email: born@kfg.mu-luebeck.de
193
194 List of Contributors
Mark P. Mattson Laboratory of Neurosciences

National Institute on Aging
Gerontology Research Center 4F01
5600 Nathan Shock Drive
Baltimore, MD 21224, USA
Phone: 1-410-558-8463 Fax: 1-410-558-8465
Email: mattsonm@grc.nia.nih.gov
Thomas S. Kilduff Biosciences Division

SRI International
Room LA 229
333 Ravenswood Avenue
Menlo Park, CA 94025, USA
Phone: 1-650-859-5509 Fax: 1-650-859-5535
Email: thomas.kilduff@sri.com

(Advances in Cell Aging and Gerontology 17) Mark P. Mattson (Eds.) - Sleep and Aging-Academic Press, Elsevier (2005)

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

(Advances in Cell Aging and Gerontology 17) Mark P. Mattson (Eds.) - Sleep and Aging-Academic Press, Elsevier (2005)

Uploaded by

Copyright:

Available Formats

TABLE OF CONTENTS

List of Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193

Abnormalities in sleep can be symptoms or causes of a variety of disorders, many

information that will benet their research. In addition, neuroscientists, neurologists,

MARK P. MATTSON, PhD

Evolutionary aspects of sleep and

Advances in Cell Aging and Gerontology, vol. 17, 132

2. Evolutionary origin of sleep

2.1. Schwartzs admonition and Rauscheckers Fundamental Dogma

In addressing the basic function of sleep referred to by some as the supreme

sleeps origin, it is essential to heed Schwartzs (1997) admonition to sleep

Restful waking, including readiness, undoubtedly was the phylogenetically

number of incredibly specic bits and pieces of visual features, formed by

2.3. Primitive sleep obviated potentially conflicting brain activities

2.4. Vertebrates that never sleep

2.5. Need for sleep in congenitally and adventitiously blind mammals

2.6. Earliest sleep

2.7. Complex retinal processing

As with most biological phenomena, there are complicating circumstances.

3. Warm-bloodedness and selective pressures for REM and NREM sleep

An early attempt to correlate the evolution of sleep states with warm-bloodedness

3.3. Presence and absence of thermoregulation during sleep states

Ambient temperature probably was responsible for an adaptively non-uniform

3.4. Implications for the evolution of sleep states

These thermoregulatory relationships have signicant implications for the

Warm-bloodedness, coupled with great agility in terrestrial-arboreal habitats,

3.5. Influences of ambient temperature

This brings us to a signicant juncture for the primitive ancestors of mammals

3.6. Actions of brain waves during waking

Fig. 3. Representative electroencephalographic records (auricularoccipital electrodes) during 30-s

the temporally-correlated ring (mobilization and synchronous activation) of

meaningless fragments of memories) because the components do not become

3.7. Actions of brain waves during sleep

3.8. Long-term memory maintenance

Although reptilian sleep closely resembles NREM sleep of warm-blooded

3.9. Basic links between muscle-controlling and visual circuits

skeletal muscle controlling circuits automatically become reinforced during

3.11. Origin of fast waves of REM sleep

4. Evolutionary origin of sleep within a brain-wave paradigm

asleep. Our vigilance declines monotonically in progressing from NREM stages 1 to 4

5. Sequential cycling of NREM and REM sleep

In the earlier discussion, it was proposed, that individual component circuits of

In brief recapitulation, the neural adaptation of circuit multifunctionality

Sleep disturbances in aging

Advances in Cell Aging and Gerontology, vol. 17, 3360

6.5. Circadian rhythm disorders

For a variety of reasons, aging is accompanied by an increasing susceptibility to

2. Normal sleep and the effects of the aging process

2.1. Age-related sleep and circadian changes

Aging is associated with reliable changes in sleep/wake architecture that are

2.2. Gender differences in sleep and aging

Complaints about sleep disturbances and associated waking symptoms are

3.2. Risk factors

Considerable morbidity accompanies the sleep/wake disturbances in older

risk factor for mortality among older, chronically institutionalized patients.

4. Common sleep disorders in older age groups: nature and etiology

Sleep disturbances in older adults are often multifactorial in nature. Contributing

4.1. Sleep-disordered breathing (SDB)

Studies show that SDB (see Chapter 6) encompasses a variety of nocturnal

4.2. Primary insomnia

Primary insomnia is a multifaceted disorder characterized by diculty initiating

4.3. Circadian rhythm disorders

Individuals with circadian rhythm disturbances experience a disparity between

Last night I took mg of __ or