Homovanillic acid and 5-hydroxyindoleacetic acid in lumbar cerebrospinal fluid after total and REM sleep deprivation in humans.

Lumbar CSF HVA and 5-HIAA levels were assayed in 3 groups each of 10 subjects, which were respectively deprived of sleep for 30 h, deprived of REM sleep and disturbed with several awakenings during SW sleep for two consecutive nights. HVA levels after total sleep (39 +/- 20 ng/ml) or REM (35 +/- 11 ng/ml) deprivation as well as after SW sleep awakenings (32 +/- 26 ng/ml) were not different from controls (42 +/- 14 ng/ml). 5-HIAA levels after REM deprivation (32 +/- 15 ng/ml) appeared increased when compared with controls (21 +/- 7 ng/ml), total sleep-deprived subjects (21 +/- 10 ng/ml) or subjects with SW sleep awakenings (27 +/- 13 ng/ml). Possible increase in 5-HT turnover after REM deprivation and possible 5-HT role in REM sleep regulation in humans are discussed.     

Role of polysialylated neural cell adhesion molecule in rapid eye movement sleep regulation in rats

Recent evidence suggests that synaptic plasticity occurs during homeostatic processes, including sleep–wakefulness regulation, although the underlying mechanisms are not well understood. Polysialylated neural cell adhesion molecule (PSA NCAM) is a transmembrane protein that has been implicated in various forms of plasticity. To investigate whether PSA NCAM is involved in the neuronal plasticity associated with spontaneous sleep–wakefulness regulation and sleep homeostasis, four studies were conducted using rats. First, we showed that PSA NCAM immunoreactivity is present in close proximity to key neurons in several nuclei of the sleep–wakefulness system, including the tuberomammillary hypothalamic nucleus, dorsal raphe nucleus, and locus coeruleus. Second, using western blot analysis and densitometric image analysis of immunoreactivity, we found that 6 h of sleep deprivation changed neither the levels nor the general location of PSA NCAM in the sleep–wakefulness system. Finally, we injected endoneuraminidase (Endo N) intracerebroventricularly to examine the effects of polysialic acid removal on sleep–wakefulness states and electroencephalogram (EEG) slow waves at both baseline and during recovery from 6 h of sleep deprivation. Endo N‐treated rats showed a small but significant decrease in baseline rapid eye movement (REM) sleep selectively in the late light phase, and a facilitated REM sleep rebound after sleep deprivation, as compared with saline‐injected controls. Non‐REM sleep and wakefulness were unaffected by Endo N. These results suggest that PSA NCAM is not particularly involved in the regulation of wakefulness or non‐REM sleep, but plays a role in the diurnal pattern of REM sleep as well as in some aspects of REM sleep homeostasis.     

Increased hypocretin-1 levels in cerebrospinal fluid after REM sleep deprivation

Rat cisternal (CSF) hypocretin-1 in cerebrospinal fluid was measured after 6 or 96 h of REM sleep deprivation and following 24 h of REM sleep rebound. REM deprivation was found to increase CSF hypocretin-1 collected at zeitgeber time (ZT) 8 but not ZT0. Decreased CSF hypocretin levels were also observed at ZT8 after 24 h of REM sleep rebound. These results suggest that REM sleep deprivation activates and REM sleep rebound inhibits the hypocretin system. Increased hypocretin tone during REM deprivation may be important in mediating some of the effects of REM sleep deprivation such as antidepressant effects, hyperphagia and increased sympathetic activity.     

Neuromodulation of the prefrontal cortex during sleep: a microdialysis study in rats

To test the hypothesis that biogenic amines of the prefrontal cortex are involved in state-dependent cortical and behavioural activation, changes in extracellular levels of serotonin (5-HT), dopamine (DA), and noradrenaline (NA) were determined during the sleep-wake cycle in freely moving rats using microdialysis probes with parallel EEG recording. Serotonin gradually increased up to 450% during wakefulness (W) as compared to slow wave sleep (SWS), before decreasing toward stable levels during the next episode of SWS. Dopamine and its metabolite homovanillic acid (HVA) were reduced during W as compared to SWS. Although contradictory with the generally admitted enhancement of DA activity related to vigilance, this may be due to the particular role of DA neurons in the prefrontal cortex. However, DA and HVA showed dramatic changes announcing the transition between SWS and W. During paradoxical sleep (PS), DA and 5-HT showed complex changes, the direction of which depended on whether PS was followed by SWS or W. Biogenic amines of the prefrontal cortex are probably involved in cortical and behavioural activation.     

REM sleep deprivation during 5 hours leads to an immediate REM sleep rebound and to suppression of non-REM sleep intensity.

Nine healthy male subjects were deprived of REM sleep during the first 5 h after sleep onset. Afterwards recovery sleep was undisturbed. During the deprivation period the non-REM EEG power spectrum was reduced when compared to baseline for the frequencies up to 7 Hz, despite the fact that non-REM sleep was not experimentally disturbed. During the recovery interval a significant rebound of REM sleep was observed, which was only accompanied by a very slight increase of power in the lower non-REM EEG frequencies. In order to control for intermittent wakefulness, the same subjects were subjected to non-REM sleep interruption during the first 5 h after sleep onset 2 weeks later. Again subsequent recovery sleep was undisturbed. The interventions resulted in a similar amount of wakefulness in both conditions. During the intervention period, the non-REM EEG power spectrum was only marginally reduced in the delta frequency range. REM sleep duration was only slightly reduced. During the recovery interval, however, a substantial increase in EEG power in the delta frequency range was noted, without notable changes in REM time. It is concluded that an increased pressure for REM sleep results in longer REM episodes and a reduced intensity of non-REM sleep.     

Differential effects of persistent nociceptive stimulation on sleep stages

The purpose of this work was to investigate the sequence of modifications of sleep and pain parameters in a condition of persistent nociceptive stimulation. In freely moving cats carrying implanted electrodes, continuous polygraphic and behavioral recordings were collected 24 h a day for several consecutive days before and after treatment. Injection of formalin (2 ml, 37%) elicited continuous wakefulness (1-6 h) associated with behavioral manifestations of pain. This insomnia was followed by the delayed appearance of LS (light, slow wave sleep) DS (deep slow wave sleep) and REM (rapid eye movement sleep). On days 1 and 2 after injection, pain manifestations displayed a gradual decrease, while total sleep time (LS + DS + REM) slowly returned to normal levels. On day 1, the amount of LS was not modified, but DS and REM were greatly decreased. For 12 h after the first REM episode, REM was decreased while DS was already at the basal levels. Formalin elicited a long-lasting increase in EMG activity of the tibialis anterior muscle which was suppressed during REM and returned to higher levels afterwards. Prolonged wakefulness and delay in sleep stage appearance were also recorded when a 24-h sleep deprivation preceded formalin injection. In this condition, LS, DS and REM amount were at basal levels from their first reappearance, and a rebound in total sleep time and DS occurred on day 2 after the injection. After injection of smaller doses of formalin (0.5 ml, 8%), the amount of LS, DS and REM was at control levels since day 1. The results suggest that (1) the amount of sleep depends on sleep debt and on the level of pain intensity and (2) sleep stages are differentially sensitive to persistent pain.     

Effects of five to seven days of sleep deprivation produced by electrical stimulation of the midbrain reticular formation.

Most methods of sleep deprivation either induce stress or fail to totally eliminate sleep. In this study, sleep deprivation was produced by electrical stimulation of the midbrain reticular formation. Seven cats were prepared for chronic recordings: Four animals received continuous stimulation, and five were stimulated intermittently. Electrographic data were collected continuously during 5–7 days of stimulation and for 48 hr thereafter. Behavioral and electrographic correlates of sleep were abolished by both methods of stimulation for 3 days. Although stimulation continued, behavioral signs of sleep appeared along with electrographic evidence of brief epochs of slow-wave sleep. Maintenance of wakefulness beyond 3 days became increasingly difficult and required frequent changes in the stimulus values. After the cessation of stimulation, there was a significant increase in total sleep in both groups for 48 hr, primarily at the expense of the drowsy state. The termination of continuous stimulation resulted in an initial rebound in slow-wave sleep and a later rebound in rapid eye movement (REM) sleep. The cessation of intermittent stimulation resulted in a large REM sleep rebound followed by a modest rebound in slow-wave sleep. These results, corroborated by data from previous studies, show that total sleep deprivation is attained by continuous stimulation; whereas REM deprivation is the main effect of intermittent stimulation.     

5-HT1A receptor-responsive pedunculopontine tegmental neurons suppress REM sleep and respiratory motor activity

Serotonin type 1A (5-HT(1A)) receptor-responsive neurons in the pedunculopontine tegmental nucleus (PPTn) become maximally active immediately before and during rapid eye movement (REM) sleep. A prevailing model of REM sleep generation indicates that activation of such neurons contributes significantly to the generation of REM sleep, and if correct then inactivation of such neurons ought to suppress REM sleep. We test this hypothesis using bilateral microperfusion of the 5-HT(1A) receptor agonist 8-hydroxy-2-(di-n-propylamino)tetralin (8-OH-DPAT, 10 μm) into the PPTn; this tool has been shown to selectively silence REM sleep-active PPTn neurons while the activity of wake/REM sleep-active PPTn neurons is unaffected. Contrary to the prevailing model, bilateral microperfusion of 8-OH-DPAT into the PPTn (n = 23 rats) significantly increased REM sleep both as a percentage of the total recording time and sleep time, compared with both within-animal vehicle controls and between-animal time-controls. This increased REM sleep resulted from an increased frequency of REM sleep bouts but not their duration, indicating an effect on mechanisms of REM sleep initiation but not maintenance. Furthermore, an increased proportion of the REM sleep bouts stemmed from periods of low REM sleep drive quantified electrographically. Targeted suppression of 5-HT(1A) receptor-responsive PPTn neurons also increased respiratory rate and respiratory-related genioglossus activity, and increased the frequency and amplitude of the sporadic genioglossus activations occurring during REM sleep. These data indicate that 5-HT(1A) receptor-responsive PPTn neurons normally function to restrain REM sleep by elevating the drive threshold for REM sleep induction, and restrain the expression of respiratory rate and motor activities.     

Effect of sleep and sleep deprivation on serotonergic neurotransmission in the hippocampus: a combined in vivo microdialysis/EEG study in rats

Brainstem serotonergic neurotransmission is implicated in sleep regulation. However, the role of serotonin (5-HT) in forebrain regions in sleep-wake mechanisms is still unclear. Here, we have investigated, using a combined in vivo microdialysis/electroencephalogram method, the relationship between hippocampal 5-HT levels and sleep-wake behaviour in the rat. A clear-cut relationship was found between hippocampal 5-HT levels and vigilance state. The highest levels of 5-HT were observed during wakefulness, whereas a progressive decrease of 5-HT going from nonrapid eye movement sleep to rapid eye movement sleep was found. Sleep deprivation (SD) causes a transient enhancement of mood in depressed patients. Given the putative role of 5-HT in the aetiology of depression and the therapeutical efficacy of selective serotonin reuptake inhibitors in this illness, we also studied hippocampal 5-HT during 4 h of SD and during the subsequent recovery period. During the whole SD period, 5-HT levels were elevated substantially when compared to 5-HT levels during basal wakefulness. However, no changes in 5-HT levels and the relationship between hippocampal 5-HT and vigilance state were found during the subsequent recovery period. As SD is a potentially stressful experience and glucocorticoids are involved in the regulation of serotonergic neurotransmission and sleep, we investigated the effects of SD on free corticosterone levels. SD caused a marked rise in free corticosterone levels. However, the effects of SD on 5-HT seem not to be mediated by this hormone, because adrenalectomy did not affect the rise in hippocampal 5-HT during SD. We hypothesize that the elevated hippocampal 5-HT levels during SD may participate in the transient mood enhancing properties of forced wakefulness observed in depressed patients.     

Pineal vasotocin and sleep: Involvement of serotonin containing neurons

Extremely small amounts (10^?4 pg) of arginine vasotocin (AVT) injected into the third ventricle of unanesthetized cats induce non-rapid eye movement (NREM) sleep and suppress rapid eye movement (REM) sleep. The same amount (10^?4 pg) of intraventricularly injected AVT increased at 15 and 30 min 5-hydroxytryptamine (5-HT) and decreased at 30 and 60 min 5-hydroxyindole acetic acid (5-HIAA) levels of the brain. Fluoxetine, a specific 5-HT uptake inhibitor, greatly enhanced the NREM sleep induced by AVT. Neither arginine vasopressin, nor oxytocin at the doses used (10^?4 pg), was able to affect indole levels of the brain or to enhance NREM sleep and to suppress REM sleep after fluoxetine, demonstrating the high specificity of AVT effects. Methergoline, a selective central 5-HT receptor blocker, completely prevented AVT induction of NREM sleep. Extremely small amounts of AVT (10^?6 pg), which in some cats were unable to induce NREM sleep and to suppress REM sleep, become effective after pretreatment with small amounts of 5-hydroxytryptophan (5-HTP). It is suggested that AVT induces NREM sleep and suppresses REM sleep by interfering with 5-HT release at postsynaptic receptor sites. The present results strongly support the monoaminergic theory of sleep providing the first evidence that a peptide synthesized by the brain induces sleep by a serotoninergic mechanism.     

Neurobiological bases for the relation between sleep and depression

The serotoninergic system is involved in the regulation of sleep and wakefulness, its activity being at maximum during the awake state and minimum during sleep. In particular, the production of rapid eye movement (REM) sleep depends on the decrease of serotoninergic tone in brain stem structures. Thus, serotoninergic compounds which increase this tone (such as antidepressants) induce inhibition of REM sleep.Depression is associated with a functional decrease of serotoninergic neurotransmission and with specific alterations of sleep, notably insomnia. Paradoxically, even though they complain of sleep loss, depressed patients exhibit significant mood improvement after one night of sleep deprivation. This antidepressant effect can be accounted for by the same serotoninergic mechanisms as those described for pharmacological treatments. Indeed, the therapeutic action of antidepressants such as selective serotonin reuptake inhibitors is thought to depend directly on the enhancement of central serotoninergic neurotransmission. Such enhancement is achieved through desensitization of serotoninergic autoreceptors, which results from chronic treatment with these compounds. Sleep deprivation also induces an activation of serotoninergic neurons due to prolonged wakefulness, and leads to similar serotoninergic adaptive processes.The common neurobiological mechanisms resulting from pharmacological antidepressant treatment and sleep deprivation suggest that sleep loss in some insomniac or in depressed patients might be an endogenous compensatory process which would be therapeutical rather than pathological. This proposal should open the way to new strategies in the treatment of depression.     

Estradiol treatment modulates spontaneous sleep and recovery after sleep deprivation in castrated male rats

Exogenous estradiol (E) is used occasionally to treat the side effects associated with androgen-deprivation in men, but its effects on sleep patterns have received little attention. We examined whether E modulates sleep patterns and recovery from sleep loss in castrated male rats. Adult male rats were castrated and implanted subcutaneously with Silastic tubes containing either oil (Cast + Oil) or E (Cast + E). Sham-operated male rats (Intact) were implanted with oil-filled tubes. All rats were also implanted with EEG and EMG electrodes for sleep/wake recordings. After two weeks, polysomnographic recordings were made before, during, and following 6 h of sleep deprivation (SD). At baseline, the Cast + Oil group showed sleep and EEG patterns similar to those in the Intact group. Compared to these groups, the Cast + E group spent more time awake during the dark (active) phase, and showed higher EEG theta power (a measure of cortical activation) during wake and rapid eye movement (REM) sleep in both the light and dark phases. Following SD, the Cast + E group showed a larger increase from baseline in REM sleep amount, compared to the Cast + Oil group. The Cast + Oil group showed prolonged rebound in non-REM sleep and EEG delta power, and reduced REM sleep rebound, compared to the other two groups. These results indicate that E treatment in castrated male rats promotes baseline wakefulness during the active phase, and facilitates recovery of REM sleep after acute sleep loss. The possible benefit of E treatment for improving sleep quality in androgen-deprived men remains to be investigated.     

Effects of the 5-HT(7) receptor antagonist SB-269970 microinjected into the dorsal raphe nucleus on REM sleep in the rat

The effects of SB-269970, a selective 5-HT(7) receptor antagonist, on spontaneous sleep were studied in adult rats implanted for chronic sleep recordings. SB-269970 was infused directly into the dorsal raphe nucleus (DRN) during the light phase. The 5-HT(7) receptor antagonist (0.25-1.0 mM) induced a significant reduction of REM sleep and of the number of REM periods whereas REM sleep latency was augmented. Pretreatment with the GABA(A) receptor agonist muscimol (1.0-2.0 mM), which by itself did not affect sleep variables, prevented the decrease of REM sleep induced by SB-269970 (1.0 mM). Our results indicate that the 5-HT(7) receptor is involved in the effect of DRN serotonergic (5-HT) neurons on brainstem structures that act to promote and induce REM sleep. We propose that the SB-269970-induced suppression of REM sleep is dependent upon the inhibition of GABA release in the DRN.     

Platform sleep deprivation affects deep slow wave sleep in addition to REM sleep

Rats were sleep deprived by the platform method to look for differential effects on light and deep slow wave sleep depending on platform size. Diameters of large and small platforms were 15 cm and 5.1 cm respectively. Sleep was recorded during a baseline light period (09.00-19.00 h), continuously during 48 h of sleep deprivation and during the first lights on recovery period (09.00-19.00 h). In both platform conditions REM sleep was virtually abolished during the first light period (hours 0-10 of sleep deprivation), while NREM sleep was reduced to approximately half of control values. During the second light period (hours 22-34 of sleep deprivation) REM sleep recovered somewhat in the large platform group. Light slow wave sleep (SWS-1) was comparable to baseline while deep slow wave sleep (SWS-2) was still significantly reduced. In the small platform group both SWS-2 and REM sleep was considerably reduced on day 2. Over the whole deprivation period there was an effect of platform size on SWS-1 (higher in the small platform group), and on SWS-2 and REM sleep (lower in the small platform group). During the 9 h light-time recovery sleep there was an REM sleep rebound in both groups. SWS-1 was reduced in both groups while SWS-2 was not significantly increased. The ratio SWS-2/SWS-1 was, however, significantly increased only in the small platform group recovery sleep. The results suggest that platform sleep deprivation deprives the animals of deep slow wave sleep in addition to REM sleep. This has implications for conclusions on REM sleep function based upon REM sleep deprivation.     

Hippocampal Noradrenaline and Serotonin Release over 24 Hours as Measured by the Dialysis Technique in Freely Moving Rats: Correlation to Behavioural Activity State,Effect of Handling and Tail-Pinch

Hippocampal extracellular levels of noradrenaline (NA), 5-hydroxytryptamine (5-HT), and 5-hydroxyindoleacetic acid (5-HIAA) were monitored with the microdialysis technique in freely moving rats. In one experiment 30 min samples were collected during 24 h of continuous perfusion, and the monoamine output was compared to the behavioural activity state, as arbitrarily classified in three categories: sleep/rest, drowsiness and full alertness associated with complex behaviours. In the individual animal the hippocampal NA and 5-HT output showed pronounced fluctuations during the 24 h period, but the 30 min sampling times did not allow for a clear-cut correlation to behavioural activity state. However, the mean NA and 5-HT output for all animals during the dark period of the day was 43 and 38% higher, respectively, than during the light period, and the average NA and 5-HT levels in samples collected during periods of high behavioural activity was 34 and 45% higher, respectively, than during periods of rest or sleep. In contrast, there were no detectable changes in extracellular 5-HIAA. The selective serotonin uptake blocker indalpine, added to the perfusion fluid at 1 microM, increased the extracellular 5-HT levels 6-fold, with a similar correlation to behavioural activity state as without indalpine. In a second experiment the effect of handling and tail-pinch was studied in 15 min sample fractions. Gentle handling of the animals during the sampling period increased the hippocampal NA and 5-HT output by 32 and 72%, respectively, and a similar increase (63 and 48%) was obtained by application of tail-pinch. Maximum NA output was reached during the handling or tail-pinch period, whereas maximal 5-HT levels were detected in the subsequent 15 min sample fraction. No changes in extracellular 5-HIAA was observed. It is concluded (1) that intracerebral microdialysis provides a useful method for the study of extracellular NA and 5-HT in the hippocampal formation of conscious rats during active behaviour; (2) that there are substantial fluctuations in hippocampal NA and 5-HT output in freely moving rats which correlate with the light - dark cycle as well as with the activity state of the animals; (3) that the spontaneous variations in 5-HT output are maintained during reuptake blockade; and (4) that behavioural activation through gentle handling or tail-pinch elicits NA and 5-HT release. The present data support a role of the forebrain NA and 5-HT systems in behavioural state control and highlights the necessity of experimental designs in which the spontaneous fluctuations in transmitter release are controlled for in studies of, for example, drug effects on NA and 5-HT release in conscious animals.     

Effect of partial sleep deprivation on sleep stages and EEG power spectra: evidence for non-REM and REM sleep homeostasis.

The effect of repeated partial sleep deprivation on sleep stages and sleep EEG parameters was investigated in young subjects. After 2 baseline nights (B1, B2) of 7.5 h, sleep was restricted for 2 nights (D1, D2) to the first 4 h of the habitual bedtime period. Two recovery nights (R1, R2) with 7.5 h sleep followed. During the deprivation nights, stages 1 and 2 and REM sleep were reduced, while slow wave sleep (SWS; stages 3 and 4) was not significantly affected. However, the time integral of EEG power density in the range of 0.75-4.5 Hz (slow wave energy) was reduced. In the recovery period, SWS showed an enhancement in R1, and REM sleep showed a rebound in R1 and R2. An increase of REM sleep in the early part of the sleep period was evident in R1. Sleep latency was reduced in D2, R1 and R2. In accordance with the 2-process model of sleep regulation, EEG power density in non-REM sleep in the range of 0.75-4.5 Hz (slow wave activity) was only slightly higher in D2 and R1 than in baseline. An enhancement of slow wave activity in REM sleep was present in D2. Power density in the frequency range of 13-16 Hz was reduced in non-REM sleep (R1), SWS (R2) and stage 2 (R1). The results show (1) that the moderate reduction of slow wave energy in the deprivation nights induces only a minor enhancement of slow wave activity during recovery sleep; and (2) that a REM sleep deficit gives rise to an immediate rebound when 'slow wave pressure' is low.(ABSTRACT TRUNCATED AT 250 WORDS)     

Recovery after prolonged sleep deprivation: residual effects of slow-release caffeine on recovery sleep, sleepiness and cognitive functions

A long work schedule often results in sleep deprivation, sleepiness, impaired performance and fatigue. We investigated the residual effects of slow-release caffeine (SRC) on sleep, sleepiness and cognitive performance during a 42-hour recovery period following a 64-hour continuous wakefulness period in 16 healthy males, according to a double-blind, randomised, placebo-controlled, crossover study. Three hundred milligrams of SRC or placebo was given twice a day at 21:00 and 9:00 during the first 48 h of wakefulness. Recovery sleep was analysed with electroencephalography (EEG) and wrist actigraphy, daytime sleepiness with continuous EEG, sleep latency tests and actigraphy and cognitive functions with computerized tests from the NATO AGARD STRES battery. Both drug groups exhibited almost the same sleep architecture with a rebound of slow-wave sleep during both recovery nights and of REM sleep during the second night. Wakefulness level and cognitive functions were similarly impaired in both groups on the first day of recovery and partially returned to baseline on the second. To conclude, SRC appears to have no unwanted side-effects on recovery sleep, wakefulness and cognitive performance after a long period of sleep deprivation and might therefore be a useful choice over other psychostimulants for a long work schedule.     

Electroencephalographic sleep in depressive pseudodementia

Twenty-six patients with mixed symptoms of depression and cognitive impairment were studied with serial clinical ratings and sleep electroencephalograms during a one-night sleep-deprivation procedure. A subgroup of these patients with depressive pseudodementia (n = 8) had less severe symptoms of dementia at baseline and showed significant improvements in both Hamilton Depression Rating Scale scores and Profile of Mood States tension scores following sleep deprivation, while another subgroup of patients having primary degenerative dementia with depression (n = 18) showed no change or worsening in Hamilton depression and Profile of Mood States tension ratings. Baseline sleep measures demonstrated significantly higher rapid eye movement (REM) percent and phasic REM activity/intensity in pseudodemented compared with demented patients. While both groups had increases in sleep efficiency, sleep maintenance, and slow-wave sleep following sleep deprivation, recovery night 2 was characterized by greater first REM period duration in depressive pseudodementia than in dementia. These differences in REM sleep rebound (using an REM period 1 cutoff of greater than or equal to 25 minutes) permitted correct identification of 88.5% of patients. Implications of these data for current theories regarding sleep, aging, and psychopathology are discussed.     

Rapid eye movement sleep deprivation as a probe in elderly subjects

The effects of a 2-night rapid eye movement (REM) sleep deprivation (RSD) procedure on electroencephalographic sleep and mood were examined in 15 healthy elderly control subjects, 14 elderly patients with endogenous depression, and 15 patients with primary degenerative dementia. Compared with control subjects, both patient groups maintained a higher amount of REM sleep time and REM activity during RSD. Unexpectedly, depressed patients showed little rebound in visually scored or automated REM sleep measures following RSD, and they showed stability of REM activity temporal distribution from baseline to recovery conditions. This contrasted with the rebound in REM sleep activity seen in control subjects, and the more modest increase in demented patients. The RSD was fairly specific, with some impact on delta sleep during the procedure but not during recovery sleep. Mood ratings were unaffected by RSD. These findings demonstrated a greater plasticity of REM sleep regulation in the healthy elderly control subjects and suggested a higher REM "pressure" with a "ceiling effect" in depressed patients. Patients with dementia appeared to have an impaired capacity to respond to the challenge of RSD.