Sleep Medicine Reviews, Vol. 4, No.

3, pp 253–262, 2000
doi:10.1053/smrv.1999.0075,
available online at http://www.idealibrary.com on
SLEEP
MEDICINE
reviews
REVIEW ARTICLE

Depression and insomnia: questions of cause

and effect
Lisa Lustberg and Charles F. Reynolds III
Interventions Research Center for Late-Life Mood Disorders and the Clinical Neuroscience
Center, Department of Psychiatry, University of Pittsburgh School of Medicine, USA

Chronic insomnia is a risk factor for the development of psychiatric disorders, including depression, as
well as a prodrome of major depressive episodes, a consequence or complication of depression that often
persists beyond the clinical episode, and a prognostic indicator of long-term illness course and treatment
response. In addition, sleep is physiologically abnormal in persons at risk for depression; for example,
shortened REM sleep latency is present not only during clinical episodes of depression, but also before
the clinical episode in subjects at risk for depressive illness. Although insomnia usually disappears as
depression is treated, it may persist, indicating heightened vulnerability to depressive relapse or recurrence.
Physiological changes in sleep related to depression correlate with the likelihood of response to psychotherapy
alone and may also identify which patients are unlikely to do well with psychosocial treatment and,
therefore, to need somatic therapy in order to preserve recovery. Electroencephalographic (EEG) sleep
changes also correlate with the speed of response and with the brittleness or durability of response (i.e.,
probability of relapse or recurrence). These observations suggest a close relationship between the regulation
of sleep and the regulation of mood. The importance of this relationship is further underscored by recent
brain imaging studies of sleep and sleep deprivation in patients with major depression. For example,
therapeutic sleep deprivation (TSD) may serve as both a catalyst of rapid antidepressant activity and as
a probe of treatment resistance. TSD’s effects on brain metabolic rates, especially in limbic areas, may
correlate with a therapeutic response to a night of sleep loss and to antidepressant medication. Finally,
treating chronic insomnia with newer selective serotonin reuptake inhibitor (SSRI) antidepressant
medication may represent an opportunity for preventing complications of insomnia, including depressive
illness.  2000 Harcourt Publishers Ltd

Key words: chronic insomnia, depression, preventative treatment, therapeutic sleep deprivation.

The bidirectional relationship between chronic insomnia and depression

Chronic insomnia is a common complaint and a risk factor for the development of
depression. The NIMH Epidemiologic Catchment Area (ECA) study of sleep dis-
turbances and psychiatric disorders identified sleep disturbance as a highly significant
risk factor (odds ratio 39.8) for the subsequent development of major depression [1].

Correspondence to be addressed to: Dr Reynolds at Western Psychiatric Institute and Clinic,

3811 O’Hara Street, Pittsburgh, PA 15213. [email protected]; Telephone: 412-624-2246;
Fax: 412-624-2841.

1087–0792/00/030253+10 $35.00/0  2000 Harcourt Publishers Ltd

254 L. Lustberg and C. F. Reynolds III

As evidence of the pervasiveness of insomnia in the community, the study found a

12–15% 6-month prevalence rate of persistent sleep disturbance. In the study, out of
7954 respondents, 40.4% of those with insomnia and 46.5% of those with hypersomnia
had a psychiatric disorder as compared to 16.4% of those with no sleep complaints.
The risk for the development of mental disorders posed by chronic insomnia has been
confirmed in other studies (e.g., Breslau et al. [2]).
Various factors contribute to persistent insomnia, including psychiatric disorders,
behavioral problems interfering with sleep (such as trying too hard to sleep), disruption
of the sleep–wake rhythm by life events, fear of insomnia, medication and substance
use/abuse, medical problems associated with pain and limitation of mobility, circadian
dysynchrony and intrinsic sleep pathologies (such as sleep apnea and periodic limb
movements) [3,4]. Clinicians should consider psychiatric, medical, behavioral or phar-
macological factors underlying the complaint of insomnia [4]. The evaluation and
treatment of chronic insomnia (that is, insomnia lasting at least 3 weeks) should take
into account underlying psychiatric, behavioral, medical, and pharmacological causes.
Ten to 20% of respondents of epidemiological studies of insomnia report severe or
constant sleep problems [5]. These problems manifest themselves in decreased sleep
maintenance, difficulty falling asleep and early morning awakening. Insomnia is
more prevalent in women, the elderly and individuals with fewer socioeconomic
resources [1].
Many psychosocial factors can have an impact on sleep quality. These factors include
major life events, such as bereavement, ongoing strains, such as chronic illness,
psychological state, age and gender [6]. A study by Dew et al. [6] examined longitudinal
data on psychosocial status and polysomnographic sleep collected annually for 3 years
from 57 healthy subjects, aged 61–89. Decreased rhythmicity and volume of social
activities, increased levels of depressive symptoms, and increased burden of chronic
medical problems correlated significantly with poor sleep quality. A study by
Kennedy [7], using data from ECA elderly respondents, also showed that adverse
health changes, sleep disturbance, and diminished social support contributed to
persistence of depressive symptoms [7]. These finding underscore the bidirectional
relationship between chronic insomnia and depression.

Practice Point
Chronic insomnia is a risk factor for the development of clinical depression. Treating
chronic insomnia effectively may reduce the incidence of depression and other
mental disorders.

Pathophysiological changes of sleep in depression

Sleep changes physiologically in chronic insomnia and in depression. Poly-

somnographic sleep characteristics of depressed patients include decreased sleep
continuity, decreased slow-wave sleep (particularly in the first NREM sleep period),
enhanced REM sleep percent and alterations in the temporal distribution of REM
sleep [8]. Depressed patients suffer disturbed sleep continuity associated with decreased
sleep efficiency, increased wakefulness and difficulty maintaining sleep. Older de-
pressed patients typically have early morning awakening, while younger depressed
 Depression and Insomnia 255

patients complain of difficulty falling asleep [9]. Polysomnographic findings in young

adult and middle aged depressives show a reduction in the absolute number of EEG
delta waves during the first non-rapid eye movement (NREM) sleep period and a
concomitant decrease in density, or rate of production, of delta activity. Similar findings
have been demonstrated in elderly depressives [8,9]. The precise nature of delta sleep
disturbances in depression and the mechanism responsible for them is somewhat
controversial, but most research would agree that the temporal distribution of delta
wave activity is altered in depression. Coupled with the altered temporal distribution
of REM sleep, the data suggest that the timing of the REM/NREM sleep cycle is
impaired in depression.
The most specific characterization of pathophysiological changes is altered intranight
temporal distribution of REM sleep, with increased amounts of early REM sleep and
reductions in REM sleep latency. REM latency is defined as the time from sleep onset
to the first REM period. Decreased REM latency is common in severe or endogenous
depression, and often persists into post-treatment periods of remission. As discussed
below, decreased REM latency also is associated with increased incidence of recurrence,
and decreased time to recurrence in patients who displayed decreased REM latency
during a depressive episode [10]. Reductions of REM latency in depression may be
due to reduction of NREM sleep, especially slow-wave sleep (SWS), following sleep
onset. Factors such as family history of depression and aging reduce the intensity of
SWS. REM latency reduction may also reflect heightened REM activity and pressure,
causing REM sleep to occur earlier [11]. This type of REM latency alteration may be
episode-related and associated with severity of depression and acute stress.
The relationships between REM latency, prognosis and treatment response in de-
pressed patients has being extensively investigated. A study by Giles [10] evaluated how
polysomnographic sleep assessment, in particular decreased REM latency, provides
prognostic information in depression. Patients with decreased REM latency prior to
treatment had an increased likelihood of developing subsequent episodes of depression
and experienced more rapid relapse after remission than those with normal REM
latency. A later study by Giles [12] found the abnormalities observed in depressed
patients are also present in never-ill first-degree relatives, suggesting that abnormal
sleep may be a marker of those at increased risk for depression. Interestingly, this
study also found that relatives of decreased REM latency probands exhibited negative
cognitions pertaining to areas of beliefs, attributional style and moment to moment
thinking, and confirmed that decreased REM latency in probands correlated with
approximately a two-fold increase in lifetime prevalence of abnormal sleep in first-
degree relatives. These data illustrate the interplay between ‘‘cognitive vulnerability
and physiologic liability’’ to depression [12]. Further exploration of the specificity of
REM latency in depression led to a study by Kupfer and Frank, which illustrated
how decreased REM latency indicates greater liability to more rapid recurrence of
depression [13,14].
Although epidemiological and sleep electroencephalographic (EEG) studies clearly
implicate a role for sleep disturbances in the pathogenesis of depression, controversies
exist regarding the specificity of physiological sleep changes as markers of depression.
A study by Benca attempted to clarify this issue in a meta-analysis of polysomnographic
findings from 177 studies of 7151 patients, testing for the specificity of sleep changes
to specific psychiatric illnesses [15]. No single variable was found to be a sensitive or
specific marker for depressive disorder. Critics of this study believe that a constellation
of findings such as a shortened REM latency and prolonged first REM period may be
256 L. Lustberg and C. F. Reynolds III

specific to depression more than any single marker of depression [11]. It is unclear
why some depressed patients fail to exhibit these characteristic sleep changes, or why
patients with other psychiatric disorders exhibit sleep manifestations linked with
depression. However, it is clear that the combination of changes, such as decrease in
slow-wave activity during the first NREM period and corresponding increases in the
rate of production of REMs in the first REM period, is more pronounced earlier in a
clinical episode of depression than later [16,17]. While patients with chronic primary
insomnia typically show sleep continuity disturbance, they lack these more specific
macro and micro-architectural features of depression [5].
Increased cholinergic activity relative to monoaminergic neurotransmission occurs
in depression and may be the basis for the observed alterations in EEG sleep [18].
Increased cholinergic activity is associated with sleep changes linked to depression
such as increases in REM sleep, early REM induction and frequent arousals. Many
effective antidepressants increase serotonergic and or noradrenergic neurotransmission.
Some have anticholinergic effects. Antidepressants follow different mechanisms with
varying effects on sleep. Some antidepressants inhibit serotonin reuptake into pre-
synaptic neurons, antagonize postsynaptic 5HT2 receptors, or function and 5HT1A
receptor agonists [8].
Functional neuroimaging techniques, such as positron emission tomography, are
beginning to enhance our understanding of the relationship between insomnia and
depression. Patients with major depressive disorder have higher rates of glucose
metabolism during the first NREM sleep period as compared with non-depressed
control subjects [19,20]. This activation is also associated with decreased slow wave
sleep. REM sleep is associated with metabolic rate similar to wakefulness and increased
metabolism in limbic and paralimbic structures whereas REM sleep in depressed
patients correlates with overactivity and decreased response in structures which
respond to physiologic challenge [21]. To date, we are not aware of any similar PET
studies on chronic primary insomnia.

Practice Point
Electroencephalographic (EEG) sleep measures are abnormal in those at risk for
depression.

Sleep and the longitudinal course of depression

Although the influence of neurobiological and psychosocial factors on the occurrence

and extent of depression has been studied, the course of depression during treatment
is of increasing interest, particularly the temporal trajectories toward recovery, the
maintenance of recovery and the factors associated with speed and durability of
recovery [22]. Focus on the ability of patients to meet a certain criterion of response
and the time required to meet this response criterion have been used to understand
the constellation of demographic, clinical and biological variables associated with
speed and durability of response to treatment (as well as treatment resistance) [23].
The various temporal pathways towards recovery and the ability to predict these
temporal profiles of recovery has been the focus of an investigation by Dew et al. [22].
These investigations examined the temporal profiles of depressive illness course and
 Depression and Insomnia 257

treatment response in 95 patients aged 60 years or older who received standardized

treatment with nortriptlyine and interpersonal psychotherapy over an 18-week period.
Pre-treatment psychosocial, clinical and EEG sleep factors were evaluated for their
predictive effects on recovery profile. The study found that more severe depressive
symptoms, greater sleep continuity disturbance and amounts of REM sleep, severe
life events, depletion of psychosocial resources and earlier age at lifetime onset of
depressive illness predicted slower, more variable patterns of response. Utilization of
cluster analysis established stable and replicable temporal profiles of the course of
recovery from depression: (1) rapid decline of depression symptoms with maintenance
of decreased levels of symptoms; (2) slower decrease in depressive symptoms with
maintained decreased levels of symptoms; (3) partial or mixed response and (4) little
decline of symptoms [22].
The reciprocal relationship between insomnia and depression is further underscored
in an examination of long-term maintenance treatment response. If good subjective
sleep quality is recovered during the acute treatment of depression, patients may be
able to be successfully treated with interpersonal psychotherapy (IPT) alone, without
antidepressant medication. A study by Reynolds et al. [24] examined the outcome of
maintenance therapy over 1 year in 47 elderly depressed patients. Patients received
maintenance interpersonal psychotherapy plus placebo or placebo plus clinical man-
agement with no interpersonal therapy. The study found that patients receiving
monthly IPT who had recovered good subjective sleep quality remained well for longer
than patients who had failed to recover good sleep quality and were still complaining
of insomnia. The authors speculated that the persistence of ruminative thoughts and
dysphoric arousal might render patients with persistent sleep disturbances less able
to engage in the cognitive strategies required by IPT. In other words, subjective sleep
quality has an association with cognitive factors. Intrusive or dysphoric ruminations
leading to diminished sleep quality, attention, memory and ability to cope is associated
with reduced effectiveness of IPT or other psychosocial treatments. In this study of
geriatric subjects, objective polysomnographic sleep measures (sleep efficiency, REM
sleep latency, REM activity, REM density, REM period 1 count, delta sleep ratio) did
not predict the duration of remission of symptoms during the first year of maintenance
therapy, while poor subjective sleep quality did. By contrast, in midlife studies, patients
who showed increased pre-treatment delta sleep ratio (that is, more normal temporal
distribution of slow-wave sleep) exhibited a five times greater chance of remaining
well on once-monthly maintenance IPT [25].

Practice Point
• Subjective sleep quality can be reliably measured by the Pittsburgh Sleep Quality
Index [26], where scores above 5 suggest clinically significant insomnia, either
primary psychophysiological insomnia or insomnia secondary to mental disorders
such as depression.
• Clinicians should pay attention to self-reported sleep quality. Return of normal
sleep quality is a good indicator of continued recovery from depression in patients
treated with maintenance interpersonal psychotherapy on a once monthly basis.
258 L. Lustberg and C. F. Reynolds III

Treating insomnia in depression

The goal of treating concomitant insomnia and depression is to resolve the underlying
depression and prevent the onset and perpetuation of insomnia. The clinical man-
agement of insomnia entails an awareness of the reciprocal relationship between
insomnia and depression. Treating depression usually, but not always, mitigates
the symptomatic sleep disturbances. Prevention of persistent sleep disturbance may
decrease the risk of relapse or recurrence of depression.
Coprescription of sedative hypnotic agents in the management of insomnia associated
with depression assumes that the patient has been instructed in rules of good sleep
hygiene and entails five main principles: (1) administrating the lowest effective dose;
(2) following an intermittent dosing schedule of 2–4 times per week; (3) prescribing
medication for short-term use; (4) making an effort to gradually discontinue medication;
and (5) preventing rebound insomnia. In addition, serotonin specific antidepressants
such as paroxetine and nefazodone may relieve both sleep disturbances and depression,
potentially obviating the need for coprescription sedatives. SSRIs have fewer side
effects than tertiary tricyclic amines. However, no systematic evaluation has been
performed regarding the benefit of serotonin specific treatment for chronic primary
insomnia. Unlike benzodiazepines and imidazopyridines, which work on GABA
receptors, they have the advantage of decreases in dependence, withdrawal, and
rebound insomnia. We have recently reviewed this literature elsewhere [27].
Physicians prescribe tricyclic antidepressants (TCAs) to treat major depressive dis-
orders. Advantages of TCAs include low cost and efficacy [28]. However, TCAs produce
significant adverse side effects. TCAs induce anticholinergic symptoms with autonomic
and cardiovascular manifestations. They also pose a high risk in overdose. Considering
the possibility of lethal effects of high doses, TCAs should not be considered for
patients who present with increased risk of suicide. Physicians have shifted to selective
serotonin re-uptake inhibitors (SSRIs), notably in the elderly, for the treatment of
depression. These agents have shared a similar degree of success as TCAs but have
fewer side effects and are much safer in overdose. SSRIs such as paroxetine, and 5-
HT2 antagonists like nefazodone relieve sleep disturbances and depression with
decreases in dependency, withdrawal and rebound. However, no systematic evaluation
has been performed regarding the benefit of serotonin specific treatment of chronic
primary insomnia. Some SSRIs are associated with negative side effects subsequent to
overstimulation of the serotonin system such as agitation, headaches, nervousness,
insomnia, GI distress and sexual dysfunction. To decrease the severity of the effects,
physicians may prescribe hypnotics with SSRIs. The hypnotic may improve the
insomnia as the antidepressant treats the underlying psychiatric condition. As a class,
SSRIs tend to be alerting to sleep and increase the amount of wakefulness or light
sleep in those with depression. Subjectively, however, patients and clinicians rate sleep
as improved on SSRIs. Recent studies have contributed on this issue: Trivedi et al. [29],
and Rush et al. [30]. Treatment-emergent sleep disturbances may be successfully
managed with coprescription of agents such as trazadone in low doses or zolpidem [27].
Benzodiazepine sedatives exhibit CNS depressant effects and induce sleep, but also
cause muscle relation and anxiolytic and anticonvulsant activity as well. Negative
effects of benzodiazepines include residual sedative effects, dependency and increased
withdrawal effects such as rebound insomnia. Zolpidem, a non-benzodiazepine hyp-
notic, exhibits sedative effects without associated properties inherent in treatment with
 Depression and Insomnia 259

benzodiazepine, due to the receptor specificity. Lack of withdrawal effects, rebound

insomnia and tolerance make it a rational, short-term treatment option for managing
insomnia in depression [27]. Success in cotherapy with zolpidem and the SSRI fluoxetine
demonstrates the value of treating insomnia and depression [31]. Minimal clinically
significant interactions occur between the two agents. Results of the study indicate
improvements in sleep time, quality, cognitive function, and status of depression.
The relationship between insomnia and depression also provides a rationale studying
whether non-sedating antidepressants can effectively treat primary insomnia. Pa-
roxetine, a non-sedating SSRI which allows for once-a-day administration, may be
effective in treating chronic primary insomnia [32]. Indeed, 11 of 14 patients with
primary insomnia showed improvements after 6 weeks of treatment in an open pilot
study. Improvement was assessed via subjective sleep quality, daytime well being,
and polysomnographic data. This study suggested that mechanisms of treatment other
than sedation may account for improvements in sleep quality in depressed patients
with resulting insomnia. However, further work including double-blind placebo-
controlled evaluation, longitudinal studies and increasing the number of subjects may
shed further light on the mechanisms by which SSRI treatment may affect chronic
primary insomnia.

Practice Point
SSRI pharmacotherapy may be effective in the management of chronic primary
insomnia, when combined with basic behavioral intervention such as instruction in
good sleep hygiene.

Sleep deprivation as a catalyst of antidepressant treatment response

and probe of treatment resistance

Antidepressant treatment response is often variable; for example, it takes an average

of 12 weeks of pharmacotherapy before attaining robust remission in elderly patients
with major depression [33]. It is therefore desirable to find ways of accelerating the
response to treatment in depression and of recognizing patients likely to be resistant
to first-time treatment. In an effort to hasten the effect of antidepressant activity,
combined use of total sleep deprivation and SSRI pharmacotherapy may be useful. In
a pilot study, eight out of 13 elderly depressed patients who were sleep deprived for
one night and started on paroxetine at the start of recovery sleep exhibited enduring
improvements in their symptoms of depression, about twice the rate of rapid response
observed with drug monotherapy [34]. Sleep deprivation may be an attractive choice
for accelerated response to treatment, since it is well tolerated, devoid of potential
drug interaction and lacks side effects. It is also an informative probe of the relationship
between sleep and antidepressant response.
Sleep deprivation elicits polysomnographic changes such as decreased sleep latency,
increased sleep efficiency and increased total accumulated delta counts. TSD and
paroxetine appear to exhibit a synergistic effect with subsequent rapid onset of clinical
antidepressant action. Its mechanism of action may illuminate the relationship of sleep
changes and depression, as suggested by functional brain imaging studies [20]. For
example, in a study by Smith et al. [35], positron emission tomography (PET) was
employed to understand the effects of combined TSD and paroxetine treatment on
260 L. Lustberg and C. F. Reynolds III

brain metabolic activity, as reflected through the use of 18-flurodeoxyglucose (18-

FDG). The study found that this treatment approach could produce a decrease in
glucose metabolism in the right anterior cingulate cortex and the right medial frontal
cortex. Specifically, all but one of six elderly patients who underwent PET studies after
TSD, recovery and 2 weeks of paroxetine, exhibited a decrease in the anterior cingulate
glucose metabolism after TSD. The patient who failed to demonstrate a metabolic shift
did not respond rapidly. These studies may have clinical value, as they serve as a
predictor of response to antidepressant treatment and as an early warning of treatment
resistance. If TSD accelerates neurochemical systems and paroxetine increases mon-
amine function, treatment response may be accelerated. Further controlled studies
contrasting rapid response rates under TSD+paroxetine, TSD+placebo, and paroxetine
monotherapy are being pursued to determine whether combined TSD and paroxetine
accelerate treatment response. Further work is also underway to understand the
functional neuroanatomy in patients who respond to this treatment approach, as well
as those who respond only partially or not at all. In this fashion, the neurobiological
substrates of insomnia, of sleep changes in depression and of sleep deprivation’s
antidepressant effect are being understood and harnessed for the treatment of patients.
In the review of sleep deprivation effects, it is important to note studies which have
suggested that partial sleep deprivation may also be effective, e.g., Giedke et al. [36],
Parry et al. [37] and Volk et al. [38], among others. While this work is not without
controversy (e.g., some investigators believe that late partial sleep deprivation may be
more effective than early partial sleep deprivation in relieving depression), it is useful
to mention here since some clinicians and patients may be more favorably disposed
to using partial sleep deprivation rather than total sleep deprivation as an augmentation
strategy to pharmacotherapy. Similarly, it is useful to recall Vogel’s pioneering studies
of REM sleep deprivation in endogenous depression, finding that REM sleep dep-
rivation, but not non-REM sleep deprivation has specific, sustained antidepressant
effect [39]. This observation led to the hypothesis that REM sleep deprivation may be
neurobiologically necessary to antidepressant efficacy. Although there are certainly
exceptions, (e.g., bupropion, nefazadone and trimipramine), most antidepressants
suppress REM sleep. Vogel’s work further suggests that some form of sleep deprivation
(total sleep deprivation, late partial sleep deprivation, or REM sleep deprivation) may
be clinically beneficial either in mediating or augmenting antidepressant response.

Practice Point
Combining one night of sleep deprivation with the start of antidepressant medication
may accelerate antidepressant response and improve early recognition of treatment
resistance.

Summary

The relationship between insomnia and depression is bidirectional. Insomnia is both

a risk factor for depression and a consequence of depression. Effective management
of chronic primary insomnia may be a means of effectively reducing the incidence of
depression, and treating insomnia in the wake of depression may modify the risk for
relapsing depression. Measures of sleep provide useful information about variability
 Depression and Insomnia 261

in antidepressant treatment response, including speed of response and durability

of response. Similarly, brief total sleep deprivation provides a useful catalyst of
antidepressant treatment response and probe of antidepressant treatment resistance.
Finally, SSRI pharmacotherapy may be a viable approach to managing chronic insomnia.

References

1 Ford DE, Kamerow DB. Epidemiologic study of sleep disturbances and psychiatric disorders.
JAMA 1989; 262: 1479–1484.
2 Breslau N, Roth T, Rosenthal L, Andreski P. Sleep disturbance and psychiatric disorders: a
longitudinal epidemiological study of young adults. Biol Psychiatry 1996; 39: 411–418.
3 Coleman RM, Roffwarg HP, Kennedy SJ et al. Sleep–wake disorders based on a poly-
somnographic diagnosis; a national cooperative study. JAMA 1982; 247: 997–1033.
4 Gillin JC, Byerley W. The diagnosis and management of insomnia. N Engl J Med 1990; 322:
239–248.
5 Reynolds CF, Buysse DJ, Kupfer DJ. Disordered sleep: Developmental and biopsychosocial
perspectives on the diagnosis and treatment of persistent insomnia. In: Kupfer DJ, Bloom F
(eds) American College of Neuropsychopharmacology: Fourth Generation of Progress. New York:
Raven Press, Ltd. 1995; p. 1617–1629.
6 Dew MA, Reynolds CF, Monk TH, Buysse DJ, Hoch CC, Jennings HR, Kupfer DJ. Psychosocial
correlates and sequelae of EEG sleep in healthy elders. J Gerontol 1994; 49: P8–P18.
7 Kennedy GJ, Kelman HR, Thomas C. Persistence and remission of depressive symptoms in
late life. Am J Psychiatry 1991; 148: 174–178.
8 Buysse DJ, Nofzinger EA, Keshavan MS et al. Psychiatric disorders associated with disturbed
sleep and circadian rhythms. In: Turek F, Zee P (eds) Regulation of Sleep and Circadian Rhythms.
New York: Marcel Dekker, Inc. 1999; p. 597–641.
9 Prinz PN, Vitiello MV, Raskind MA, Thorpy MJ. Geriatrics: sleep disorders and aging. N
Engl J Med 1990; 323: 520–526.
10 Giles DE, Jarrett RB, Roffwarg HP, Rush AJ. Reduced rapid eye movement latency: a predictor
of recurrence in depression. Neuropsychopharmacology 1987; 1: 33–39.
11 Kupfer DJ, Reynolds CF. Sleep and affective disorders. In: Paykel ES (ed) Handbook of Affective
Disorders. Second ed. London: Churchhill Livingstone 1992; p. 311–323.
12 Giles DE, Etzel BA, Biggs MM. Risk factors in unipolar depression: II. Relation between
proband REM latency and cognitions of relatives. Psychiatry Res 1990; 33: 39–49.
13 Kupfer DJ, Ehlers CL, Frank E, Grochocinski VJ, McEachran AB. EEG sleep profiles and
recurrent depression. Biol Psychiatry 1991; 30: 641–655.
14 Kupfer DJ, Frank E, Grochocinski VJ, Gregor M, McEachran AB. Electroencephalographic
sleep profiles in recurrent depression: a longitudinal investigation. Arch Gen Psychiatry 1988;
45: 678–681.
15 Benca RM, Obermeyer WH, Thisted RA, Gillin JC. Sleep and psychiatric disorders: a meta-
analysis. Arch Gen Psychiatry 1992; 49: 651–668.
16 Kupfer DJ, Ehlers CL. Two roads to rapid eye movement latency. Arch Gen Psychiatry 1989;
46: 945–948.
17 Dew MA, Reynolds CF, Buysse DJ et al. EEG sleep profiles during depression: effects of
episode duration and other clinical and psychosocial factors in older adults. Arch Gen
Psychiatry 1996; 53: 148–156.
18 McCarley RW. REM sleep and depression: common neurobiological control mechanisms. Am
J Psychiatry 1982; 139: 565–570.
19 Gillin JC, Buchsbaum MS, Valladares-Neto DC et al. Effects of zolpidem on local cerebral
glucose metabolism during non-REM sleep in normal volunteers: a positron emission
tomography study. Neuropsychopharmacology 1996; 15: 302–313.
20 Wu JC, Gillin JC, Buchsbaum MS et al. The effect of sleep deprivation on cerebral glucose
metabolic rate in normal humans assessed with positron emission tomography. Sleep 1991;
14: 155–162.
21 Nofzinger EA, Keshavan MS, Buysse DJ et al. Neurobiology of sleep in relation to mental
262 L. Lustberg and C. F. Reynolds III

illness. In: Charney DS, Nestler EJ, Bunney BS (eds) Neurobiology of Mental Illness. New York:
Oxford University Press 1999; p. 915–929.
22 Dew MA, Reynolds CF, Houck PR et al. Temporal profiles of the course of depression during
treatment: predictors of pathways toward recovery in the elderly. Arch Gen Psychiatry 1997;
54: 1016–1024.
23 Frank E, Jarrett DB, Kupfer DJ, Grochocinski VJ. Biological and clinical predictors of response
in recurrent depression: a preliminary report. Psychiatry Res 1984; 13: 315–324.
24 Reynolds CF, Frank E, Houck PR et al. Which elderly patients with remitted depression remain
well with continued interpersonal psychotherapy after discontinuation of antidepressant
medication? Am J Psychiatry 1997; 154: 958–962.
25 Kupfer DJ, Frank E, McEachran AB, Grochocinski VJ. Delta sleep ratio: A biological correlate
of early recurrence in unipolar affective disorder. Arch Gen Psychiatry 1990; 47: 1100–1105.
26 Buysse DJ, Reynolds CF, Monk TH, Berman SR, Kupfer DJ. The Pittsburgh Sleep Quality
Index (PSQI): a new instrument for psychiatric research and practice. Psychiatry Res 1989;
28: 193–213.
27 Reynolds CF. Pharmacotherapy for concomitant insomnia and depression. Primary Care
Psychiatry (in press).
28 Paykel ES, Priest R. Recognition and management of depression in general practice: consensus
statement. Br Med J 1992; 305: 1198–1202.
29 Trivedi MH, Rush AJ, Armitage R et al. Effects of fluoxetine on the polysomnogram in
outpatients with major depression. Neuropsychopharmacology 1999; 20: 447–459.
30 Rush AJ, Armitage R, Gillin JC, Yonkers K, Winokur A, Moldofsky H, Vogel GW, Kaplita
SB, Fleming JB, Montplaisir J, et al. Comparative effects of nefazodone and fluoxetine on
sleep in outpatients with major depressive disorder. Biol Psychiatry 1998; 44: 14.
31 Piergies AA, Sweet J, Johnson M, Roth-Shechter BF, Allard S. The effect of co-administration
of zolpidem with fluoxetine: pharmacokinetics and pharmacodynamics. int J Pharmacol Ther
1996; 34; 178–183.
32 Nowell PD, Reynolds CF, Buysse DJ, Dew MA, Kupfer DJ. Paroxetine in the treatment of
primary insomnia: preliminary clinical and EEG sleep data. J Clin Psychiatry 1999; 60(2):
89–95.
33 Bump GM, Reynolds CF, Smith G et al. Accelerating response in geriatric depression: a pilot
study combining sleep deprivation and paroxetine. Depress Anxiety 1997; 6: 113–118.
34 Green TD, Reynolds CF, Mulsant BH et al. Accelerating anti-depressant response in geriatric
depression: a post-hoc comparison of combined sleep deprivation and paroxetine versus
monotherapy with paroxetine, nortriptyline, or placebo. J Geriatr Psychiatry Neurol (in press).
35 Smith G, Price JC, Lopresti B et al. Test–retest variability of positron emission tomography
(PET) and [18F]-altanserin for the in vivo imaging of serotonin 5-HT2A receptors in the
human brain. Synapse 1998; 30: 380–392.
36 Giedke H, Geilenkirchen R, Hauser M. The timing of partial sleep deprivation in depression.
J Affect Disord 1992; 25: 117–128.
37 Parry BL, Cover H, Mostofi N et al. Early versus late partial sleep deprivation in patients
with premenstrual dysphoric disorder and normal comparison subjects. Am J Psychiatry 1995;
152: 404–412.
38 Volk SA, Kaendler SH, Hertel A et al. Can response to partial sleep deprivation in depressed
patients be predicted by regional changes of cerebral blood flow? Psychiatry Res 1997; 75:
67–74.
39 Vogel GW, Vogel F, McAbee RS, Thurmond AJ. Improvement of depression by REM sleep
deprivation. Arch Gen Psychiatry 1980; 37: 247–253.