Human slow wave sleep: a review and appraisal of recent findings, with implications for sleep functions, and psychiatric illness.

Recent findings concerning human slow wave sleep (hSWS-stages 3 + 4; delta EEG activity) are critically reviewed. Areas covered include the significance of the first hSWS cycle; hSWS in extended sleep; relationship between hSWS, prior wakefulness and sleep loss; hSWS influence on sleep length; problems with hSWS deprivation; influence of the circadian rhythm; individual differences in hSWS, especially, age, gender and constitutional variables such as physical fitness and body composition. Transient increases in hSWS can be produced by increasing both the quality and quantity of prior wakefulness, with an underlying mechanism perhaps relating to the waking level of brain metabolism. Whilst there may also be thermoregulatory influences on hSWS, hypotheses that energy conservation and brain cooling are major roles for hSWS are debatable. hSWS seems to offer some form of cerebral recovery, with the prefrontal cortex being particularly implicated. The hSWS characteristics of certain forms of major psychiatric disorders may well endorse this prefrontal link.     

Genetics of sleep and sleep disorders

Genetic factors affect sleep. Studies in twin pairs demonstrate that the strong hereditary influences on sleep architecture and some sleep disorders are transmitted through families. Evidence like this strongly suggests that sleep regulation receives significant influence from genetic factors. Although recent molecular technologies have revealed evidence that genetic traits or gene products trigger particular changes in sleep electroencephalogram activity, we are still far from finding candidate genes or multiple mutations responsible for individual sleep disorders. Sleep is a very complex phenotype. Genetic susceptibility and environmental factors should be also considered as contributors to sleep phenotype. The aim of this review is to present a current summary and future prospects for genetic studies on sleep and selected sleep-associated disorders. An erratum to this article is available at .     

Experimental dissociation between hippocampal theta activity in wakefulness and hippocampal theta activity in paradoxal sleep in the rat]

The effects of electrolytic lesion of the septum on the theta activity of the dorsal hippocampus were studied in the chronically implanted Rat during wakefulness and paradoxical sleep. The experimental results show that depending on its localization, septal lesion can either: (1) eliminate the wakefulness theta rhythm without suppressing that of paradoxical sleep; (2) eliminate the paradoxical sleep theta rhythm without suppressing that of wakefulness. These results suggest that there are two kinds of theta activity having different anatomophysiological bases and a different functional significance: one associated with wakefulness and the other with paradoxical sleep.     

Sleep in the tortoiseKinosternon sp.

Summary Individuals ofKinosternon sp., previously confined to laboratory conditions, were chronically implanted with electrodes for electroencephalogram, electro-oculogram and electrocardiogram recording. Behavioral states of waking and sleep were clearly observed. Two sleep stages were present: quiet sleep and REM or active sleep. Electrical cerebral activity was polymorphic and irregular. EEG frequencies declined and amplitudes diminished with sleep. Arrhythmic spikes occurred during behavioral sleep and declined with waking. Heart rate decreased when passing from wakefulness to quiet sleep. It was slightly but consistently higher during active sleep compared with quiet sleep.     

Sleep and body restitution

Summary Although human non-REM sleep is usually associated with body restitution, such an hypothesis is debatable. This sleep, like REM sleep, may have a beneficial role for the brain. Because man demonstrates relaxed wakefulness, body restitution may not be confined to human sleep. However, for active mammals, sleep may be an enforced immobiliser facilitating this restitution.     

Seasonality in human sleep

The timing of sleep and sleep EEG parameters in 10 healthy male subjects were investigated in four seasons under controlled conditions. The phase of nocturnal sleep was delayed about one and a half hours in winter as compared to that in summer. The duration of stage 4 sleep decreased and REM sleep increased significantly in winter compared with summer. The seasonality in the timing of sleep can be explained by photoperiodic time cues, but the changes in sleep EEG parameters are diffucult to explain in terms of photoperiod.     

Neuroanatomy and neurochemistry of sleep

Sleep is regulated by homeostatic and circadian factors, and the regulation of sleep of mammals shares many molecular properties with the rest state of submammalian species. Several brain structures take part in waking: the basal forebrain, posterior and lateral hypothalamus, and nuclei in the tegmentum and pons. Active sleep mechanisms are located to the preoptic/anterior hypothalamic area. In addition to acetylcholine and monoamines, glutamate and hypocretin/orexin are important waking factors. Gamma-aminobutyric acid and several peptide factors, including cytokines, growth hormone-releasing hormone and prolactin, are related to sleep promotion. Adenosine is an important homeostatic sleep factor acting in basal forebrain and preoptic areas through A1 and A2A receptors. Prolonged waking activates inducible nitric oxide synthase in the basal forebrain, which through energy depletion causes adenosine release and recovery sleep. Numerous genes have been found differentially displayed in waking compared with sleep, and they relate to neural transmission, synaptic plasticity, energy metabolism and stress protection. The genetic background of a few sleep disorders has been solved.     

Sleep and sleep disturbances: biological basis and clinical implications

Sleep is a neurochemical process involving sleep promoting and arousal centers in the brain. Sleep performs an essential restorative function and facilitates memory consolidation in humans. The remarkably standardized bouts of consolidated sleep at night and daytime wakefulness reflect an interaction between the homeostatic sleep need that is manifested by increase in sleep propensity after sleep deprivation and decrease during sleep and the circadian pacemaker. Melatonin, the hormone produced nocturnally by the pineal gland, serves as a time cue and sleep-anticipating signal. A close interaction exists between the sleep-wake, melatonin, core temperature, blood pressure, immune and hormonal rhythms leading to optimization of the internal temporal order. With age the robustness of the circadian system decreases and the prevalence of sleep disorders, particularly insomnia, increases. Deviant sleep patterns are associated with increased risks of morbidity, poor quality of life and mortality. Current sleep pharmacotherapies treat insufficient sleep quantity, but fail to improve daytime functioning. New treatment modalities for sleep disorders that will also improve daytime functioning remain a scientific and medical challenge.     

Technologies of sleep research

Sleep is investigated in many different ways, many different species and under many different circumstances. Modern sleep research is a multidisciplinary venture. Therefore, this review cannot give a complete overview of all techniques used in sleep research and sleep medicine. What it will try to do is to give an overview of widely applied techniques and exciting new developments. Electroencephalography has been the backbone of sleep research and sleep medicine since its first application in the 1930s. The electroencephalogram is still used but now combined with many different techniques monitoring body and brain temperature, changes in brain and blood chemistry, or changes in brain functioning. Animal research has been very important for progress in sleep research and sleep medicine. It provides opportunities to investigate the sleeping brain in ways not possible in healthy volunteers. Progress in genomics has brought new insights in sleep regulation, the best example being the discovery of hypocretin/orexin deficiency as the cause of narcolepsy. Gene manipulation holds great promise for the future since it is possible not only to investigate the functions of different genes under normal conditions, but also to mimic human pathology in much greater detail.     

Chronobiology of sleep in humans

Periodic circadian (24-h) cycles play an important role in daily hormonal and behavioural rhythms. Usually our sleep/wake cycle, temperature and melatonin rhythms are internally synchronized with a stable phase relationship. When there is a desynchrony between the sleep/wake cycle and circadian rhythm, sleep disorders such as advanced and delayed sleep phase syndrome can arise as well as transient chronobiologic disturbances, for example from jet lag and shift work. Appropriately timed bright light is effective in re-timing the circadian rhythm and sleep pattern to a more desired time, ameliorating these disturbances. Other less potent retiming effects may also be obtained from the judicious use of melatonin and exercise.     

Sex differences in human circadian rhythms: Intrinsic periods and sleep fractions

Summary The period of freerunning circadian rhythms is significantly shorter and the fraction of sleep is significantly larger in human females than in males, as long as the rhythms run internally synchronized. The sex difference in the period could be a property either of the whole circadian system or of only one of the oscillators in a multi-oscillator system. The sex difference in the sleep fraction could be a fixed property of the sleep-wake rhythm or could depend on interactions in the multi-oscillator system. To investigate these questions, a sample of 33 long-term experiments, in which the rhythms ran internally synchronized in one section and internally desynchronized in another section, were analyzed. The periods of rhythms in rectal temperature were different in females and males during internal synchronization, but became identical during internal desynchronization. In contrast, sex differences in sleep-wake periods were more pronounced when the rhythms were desynchronized than when they were internally synchronized. This result provides evidence that the sex difference in periodicity is a property only of the sleep-wake rhythm; the intrinsic periods of temperature rhythms are identical in females and males, whereas those of sleep-wake rhythms are distinctly shorter in females than in males. In the state of internal synchronization, the joint period is a compromise between the intrinsic periods of the rhythms involved, and therefore it shows a small but significant sex difference. Moreover, the transition from internally synchronized to desynchronized rhythms is combined with a highly significant reduction in the sleep fraction, which is considerably greater in females than in males. These results suggest that the occurrence of internal desynchronization strongly affects the sleep-wake rhythm, and that the influence of rhythm disorders is considerably greater in females than in males.     

EEG and sleep in aged hospitalized patients with senile dementia: 24-h recordings

Polygraphic recordings of wake and sleep were performed on 10 partly bed-ridden, severely deteriorated patients with senile dementia. Compared with healthy elderly persons these subjects showed less SWS (slow wave sleep, characterized by high amplitude, slow EEG waves), less REM sleep (rapid eye movement sleep, usually accompanied by dream activity) and poorly organized stage 2 sleep (no sleep spindles, i.e. phasic EEG activity with a frequency of 12-14 Hz). Six of the 10 patients had no dominant alpha rhythm during wakefulness; this seemed to be related to their more deteriorated clinical state, to still less SWS and REM sleep and more time spent in stage 2. The basic NREM-REM cycle of sleep, i.e. the regular alternation between non-REM- and REM-periods, could still be distinguished, however, and showed similar average temporal characteristics as in healthy old and younger people. Similarly, although sleep was severely fragmented in most patients and many sleep episodes occurred during the day, the day-night alternation of wakefulness and sleep was maintained in the sample as a whole.     

Electrophysiological semeiology of sleep

Summary The main characteristics of electroencephalograms, electro-oculograms and electromyograms in human sleep are described. This electrophysiological semeiology permits the identification of the different stages in normal sleep. In animals, sleep is generally less differentiated; the possibility of recording subcortical structures allows the observation of additional phenomena such as hippocampal theta, activity and PGO spikes. Evoked brain electrical activity is less well known than the spontaneous activity in sleep. Recent technological developments offer many interesting possibilities in the processing of the EEG and other physiological signals.     

Shivering during sleep: Relationship between muscle blood flow and fiber type composition

The present study considers in rabbit: i) the relationship between muscle blood flow (BF) increase and fiber-type composition during shivering; ii) the influence of the vigilance states (Quiet Wakefulness, QW; Synchronized Sleep, SS; Desynchronized Sleep, DS) on this relationship. The results show that muscle BF increase during shivering is proportional to the slow-twitch oxidative (SO) fiber component in QW and SS; in DS the proportionality is lost. This is in accordance with the disappearance of shivering, together with all thermoregulatory effector responses, in this sleep state. Another muscle circulation pattern occurring at low ambient temperature, the relationship between BF increase and muscle depth, also disappears in DS. This confirms that the integrative control of muscle circulation, like other integrative mechanisms, is impaired during DS.     

Changes in cAMP concentration in the rat preoptic area during synchronized and desynchronized sleep

Summary cAMP concentration was found to be significantly lower during desynchronized sleep than during synchronized sleep in the preoptic area of rats kept at normal laboratory temperature. No significant changes in cerebral cortex cAMP concentraion were observed in the same experimental conditions.This work was supported by grants from Consiglio Nazionale delle Ricerche and Ministero della Pubblica Istruzione.The authors wish to thank G. Mancinelli and L. Sabattini for technical assistance and M. Luppi for secretarial assistance.     

Behavioral phenomenology of sleep (somatic and vegetative)

Conclusions The foregoing analysis of behavioral sleep phenomenology shows that the most significant factual and theoretical aspects of sleep can be logically organized only according to several criteria, it being impossible to choose a singli one as truly paradigmatic. For this reason an ordinal classification of sleep phases was preferred. This fact does not detract from the usefulness of classifications based consistently on 1 criterion at a time (e.g.: synchronized-desynchronized; quiet-active; orthodoxical-paradoxical; NREM-REM; homeostatic-poikilostatic; spindle wave-slow wave-fast wave; external appetitive-internal appetitive-internal consummatory; and so on). In this respect, the bioelectrical classification is surely the best as it allows an analytical subdivision of the evolution of sleep with high resolving power^137–139. In particular, the electroencephalographic activity of late phase II (stage 4 in man¹³⁹ and slow wave¹¹ or deep slow wave¹⁴⁰ sleep in the cat) appears to be related to the triggering mechanisms and to the quantitative regulation of the circadian amount of phase III^3,5,11,140. However, in extending the field of functional implications of sleep phenomenology other criteria may be more significant. In fact, the somatic and vegetative events of sleep also lend themselves to an analysis according to the behavioral model of ethology^6,141–144 and the theory of homeostasis^{3–5, 145}, respectively. As an example, a number of classifying criteria are indicated in the table, where others, particularly neurochemical ones^146,147, could be added. At any rate, the difficulty of organizing sleep events into a satisfactory operational scheme is due to the fact that sleep is still an open problem as far as its mechanisms and functional significance are concerned.     

Long-term motor activity recording of dogs and the effect of sleep deprivation

Summary Motor activity of laboratory dogs was recorded for several weeks with an ambulatory monitoring device. The effect of 24 h sleep deprivation (SD) on motor activity during recovery was investigated. A clear rest-activity rhythm was established. The dogs exhibited a similar mean daily rest-activity pattern: 1) rest occurred mainly in the dark; 2) the amimals were most active after light onset; activity increased during the last two dark hours; 3) a rest period was found at noon and reduced activity during afternoon hours. There was a marked difference in total activity between individual dogs. Activity patterns varied as a function of the day of the week; this may have been a reflection of variations in the level of human activities in the laboratory. There was a significant reduction of motor activity during the 24-h period following SD. This was particularly evident in the first 6 h of the light period immediately following the deprivation.In addition, there was a significant increase in the number of episodes with activity 5 counts during recovery. The study confirms the possibility of measuring motor activity to assess compensatory mechanisms during recovery after SD. Sleep regulation, therefore, does not necessarily need to be exclusively examined by the invasive technique of EEG registration.     

Proposed effects of brain noradrenaline on neuronal activity and cerebral blood flow during REM sleep

Summary We propose that the observed increases of both neuronal activity and cerebral blood flow seen throughout the brain during REM sleep may be effects of decreased central noradrenaline release.