ABC of Sleep Medicine (ABC Series)

Chapter 1

Normal versus Abnormal Sleep


·        Sleep almost certainly serves a vital function at the cellular level and is an absolute requirement for every animal

·        In large population studies across the globe, chronically poor or insufficient sleep appears to correlate with increased mortality, arterial disease, diabetes and possibly cancer rates

·        Sleep is highly orchestrated into discrete cycles of non-rapid eye movement (non-REM) and rapid eye movement (REM) stages

·        Vivid dreams most often occur from the REM sleep stage

·        Increasing age dramatically alters sleep quality and consolidation

·        If insufficient deep non-REM sleep is obtained during a night, subjects will generally awake unrefreshed

·        REM sleep is a very active brain state which has been proposed to facilitate memory consolidation and emotional processing although its true function ultimately remains obscure

·        At least 90% of the adult population benefit from 7–8 hours of good sleep per night

·        Around 5% of the population can be considered excessively sleepy during the day although increased sleepiness may not be recognised as such and may be expressed through other symptoms

·        During nocturnal sleep, a variety of bodily movements is experienced normally

The importance of sleep

Virtually everyone acknowledges that significantly disturbed sleep has profound and immediate adverse effects on mental, cognitive and even physical well-being. However, the true long-term importance of good sleep for optimal general health may not yet be fully recognised.

The fact that every animal has evolved to have an absolute need for regular sleep in order to survive clearly suggests that it performs some vital and, as yet, ill-defined function. Intuitively, sleep appears to facilitate restoration and repair. Almost certainly, however, it has much more than a simple passive or restful role. In many respects, sleep is an active brain state and is not merely the absence of wakefulness. Indeed, during rapid eye movement (REM) sleep, the brain is as metabolically active as during wakefulness. As a consequence, some authorities have termed REM sleep as ‘paradoxical sleep’.

In recent animal models, it appears that being awake for just a few hours vigorously activates metabolic ‘cell stress’ or adaptive biochemical pathways. This response to prolonged wakefulness is seen particularly in nerve cells that appear to be protecting themselves from damage and potential early death or apoptosis. It is therefore conceivable that any excess of wakefulness is the potentially damaging factor rather than a lack of sleep per se.

It is clearly difficult to study the long-term consequences of bad sleep in humans. Of interest, however, a large prospective four-year study of healthy elderly subjects suggested that the only reliable predictor of death or subsequent dependency from a large number of demographic details was a complaint of disordered sleep, particularly in males.

A renowned sleep researcher (William Dement) famously stated that ‘Sleep is of the brain, for the brain and by the brain’, emphasising the adverse effects of poor sleep predominantly on brain function and mental health. Conversely, virtually every central nervous system and mental health disorder has the potential to disturb the sleep–wake cycle. Furthermore, chronic poor quality or insufficient sleep may actively fuel many common conditions, such as generalised pain syndromes and affective disorders. This strongly suggests that sleep has a ‘bi-directional’ relationship with many common conditions (Figure 1.1).

Figure 1.1 Chronically poor quality or insufficient sleep is rarely an isolated problem. For example, pain and anxiety may well impede good sleep but sleep restriction or impairment can also increase sensitivity to pain and raise anxiety levels.


In many situations, it is possible that direct attention both to sleep quantity and quality may have indirect and positive effects on an unexpected range of health issues. In this respect, the established epidemiological links between chronic sleep deprivation (less than six hours a night) and diabetes, hypertension, vascular disease or even cancer are increasingly germane. The difficult and important question of whether increasing quantity or quality of sleep in ‘at risk’ populations will positively affect outcome remains to be established.

Defining sleep

The loose behavioural definition of sleep as a temporary and reversible state of altered consciousness and perceptual disengagement has been superseded by electrophysiological criteria.

Although simplistic, it is valid to consider three distinct and mutually exclusive brain states, namely wakefulness, rapid eye movement (REM) sleep and non-REM sleep (Figure 1.2). Switches between these states should occur automatically, quickly and relatively seamlessly. A large proportion of sleep disorders is associated with inefficient, faulty or incomplete states of transition.

Figure 1.2 The brain can be considered as normally existing in the three mutually exclusive states of WAKE, non-REM or REM sleep. Orchestrated transitions between these states occur automatically and relatively quickly over the 24-hour period. In many sleep disorders, particularly the parasomnias, the switch between states may be inefficient or incomplete.


Non-REM sleep can be divided into light (stages 1 and 2) or deep phases (stages 3 and 4) on the basis of the surface cerebral electroencephalogram (EEG) (Figure 1.3). Around 20% of the night is spent in the curious state of REM sleep, in which most of the cerebral cortex and limbic system is extremely active, as confirmed by recent functional brain imaging studies. In contrast to this enhanced metabolic activity, there are descending inhibitory neural impulses from the brainstem that innervate the vast majority of peripheral muscles during REM sleep, rendering the subject floppy (atonic) and areflexic.

Figure 1.3 Representative electroencephalographic (EEG) traces of the 4 stages of non-REM sleep. The arrow indicates a K-complex, one of the hallmarks of light non-REM sleep. Large amplitude delta (slow) waves dominate the trace in deep non-REM sleep (stages 3 and 4).


In recently revised criteria for sleep staging, light non-REM sleep (stages 1 and 2) has been designated as N1 whereas deep non-REM (stages 3 and 4) is N2. This revision has not been completely accepted internationally at the time of writing.

REM sleep loosely correlates with the normal phenomena of dreams or nightmares, usually recalled briefly if a subject is awoken from this stage of sleep. However, less vivid or bizarre ‘sleep mentation’, usually without a narrative thread, is also frequently reported if arousals from sleep occur from non-REM sleep stages.

An ideal nocturnal sleep in young adults consists of four or five cycles of REM/non-REM sleep with deep non-REM sleep dominating the first third of the night and REM sleep the last third (Figure 1.4). Minor arousals from sleep are common, especially with increasing age, and often not registered or recalled.

Figure 1.4 A typical hypnogram of a young adult showing discrete cycles of non-REM and REM sleep through the night. Deep non-REM sleep predominates in the first third whereas REM sleep is most dense in the last third. Brief awakenings are often not recognised. If they occur during REM sleep, a vivid dream would be expected.


How much sleep is needed?

There is clearly a degree of individual variation in the optimum length of the nocturnal sleep period but probably 90% of adults require at least seven hours of good quality sleep. For most people, sleeping regularly for six hours or less produces objective signs of reduced vigilance, even if subjective sleepiness is minimal.

Although sleepiness is the obvious consequence of acute sleep deprivation, increasingly, a number of studies have demonstrated neuropsychological effects (Table 1.1). The majority of these can be interpreted in terms of temporary frontal lobe dysfunction. Many have commented that the sleep-deprived brain of a young adult functions in a similar way to the brain in extreme old age.

Table 1.1 A selected list of the neuropsychological effects secondary to acute sleep deprivation.

Main neuropsychological effect of acute

Year of study

sleep deprivation


Increased reaction times


Perseveration and reduced flexibility


Impaired sense of humour


Increased risk taking


Impaired moral judgement


Reduced emotional intelligence


Increased ‘negativity’ with enhanced memory for adverse events


Increased distractibility


Although underlying mechanisms are unclear and precise interpretation is difficult, several enormous population studies have consistently reported increased mortality, vascular disease and diabetes in those reporting less than five hours of sleep per night when followed up for several years. Intriguingly, those sleeping for more than 9.5 hours a night similarly have reduced longevity.

Increasingly, it is realised that the nature of sleep may be as crucially important as its quantity, although precise definitions of sleep quality are poorly defined. The absolute amount of deep non-REM (slow wave) sleep through the night may predict how refreshed a subject feels in the morning but other measures, such as the time spent awake after sleep onset, may also be used as surrogate markers of quality.

The true function of REM sleep remains a mystery. Most awakenings from REM sleep are associated with vivid dreams that often have a bizarre narrative, incorporating elements of recent events or more distant memories. Psychoanalytic explanations of dreams in terms of ‘wish fulfilment’ have largely been superseded by more biological explanations. However, it remains unclear whether it is the dream itself or the underlying neurobiological processes behind REM sleep that are more important.

A common thread with much recent research on REM sleep function concerns memory processing or consolidation. In brief, one influential view is that the brain restructures or consolidates certain forms of memory through a form of rehearsal when ‘off-line’ during REM sleep. It is likely that processing of emotional memories is a particularly important function. Many cognitive tasks seem to improve after a period of sleep and some evidence even supports the notion that unexpected insights into mathematical problems occur during the unconscious state of sleep.

The study of dreams, oneirology, has a rich history. Many theories regarding dream function and REM sleep have been generated. Some interesting facts concerning REM sleep are listed in Box 1.1.

Box 1.1 Interesting facts about REM or ‘paradoxical’ sleep

·        Almost one third of the first few months of life are spent in a form of REM sleep.

·        In animal experiments, preventing rats from entering REM sleep for around four weeks has fatal consequences.

·        There are profound autonomic changes during REM sleep causing ‘instability’ of pulse rate and blood pressure. In theory, this might predispose certain individuals to cardiac arrhythmias during REM sleep.

·        Thermoregulation is absent during REM sleep. If the environment is too hot or cold, arousals from sleep are particularly likely during the REM stage.

·        Many drugs, including alcohol, suppress REM sleep. If stopped abruptly after prolonged intake, such drugs lead to rebound symptoms, including vivid dreaming or even visual hallucinations.

·        Depressed patients enter REM sleep more quickly.

·        The majority of antidepressant drugs suppress or delay REM sleep.

·        Domestic animals have significantly more REM sleep than their feral counterparts.

It is often overlooked that sleepiness, like thirst or hunger, is a true drive state which builds up during prolonged wakefulness and will only be satiated by sleep itself. The neurobiological substrates of sleepiness and the underlying homeostatic mechanisms remain poorly understood although accumulation of the neurochemical adenosine in certain key areas, such as the basal forebrain, may be crucial. Indeed, caffeine works to offset the sleep drive by inhibiting adenosine receptors in these areas.

Superimposed on the homeostatic sleep drive are circadian factors which confer additional levels of somnolence at certain times of the day, irrespective of the immediate sleep history. These ‘dips’ in alertness, seen particularly at 3.00 p.m. and 4.00 a.m., may have consequences for behaviour and explain the rise in sleep-related traffic accidents at these times.

The fascinating field of chronobiology has provided a sound scientific basis for understanding the mechanisms of our internal clocks. Remarkably, the molecular machinery at a subcellular level has been largely established and hardly differs across all species studied from fruit flies to humans. The suprachiasmatic nucleus, a small area in the hypothalamus comprising around 25 000 neurons, has been established as the ‘master clock’, with the capacity to influence all circadian rhythms throughout the body (Figure 1.5).

Figure 1.5 The suprachiasmatic nucleus in the hypothalamus is circled with magnification of the cellular component. This small area of the brain contains around 25 000 neurons but is the ‘master’ timekeeper of all circadian rhythms in the body.


The effects of age

The quantity and, particularly, the quality of sleep change enormously across the normal lifespan. Newborns typically spend over half of the 24-hour period in a state resembling REM sleep. They have wakeful episodes lasting two or three hours interspersed with similar relatively short sleep periods across the day–night period. In the first two years, a prolonged nocturnal sleep usually becomes established, invariably encouraged by parental input, with afternoon naps remaining the norm until around five years of age.

The commonest pattern through teenage life is for sleep onset to become progressively later. Although lifestyle factors and habits may fuel this tendency, there is also evidence that most teenagers have internal clock mechanisms predisposing them to become ‘night owls’. The most obvious practical consequence of this pattern is extreme difficulty arising at a conventional hour for educational purposes.

Beyond adolescence, however, the brain's internal clock tends to ‘advance’. With each subsequent decade, the natural desire to sleep typically occurs around 30 minutes earlier.

The most striking age-related change in sleep patterns relates to a progressive deterioration in sleep consolidation. Several minor arousals, usually later in the night, might be considered normal in early middle-age. In the healthy elderly, frank sleep fragmentation is extremely common. The depth of slow wave sleep (non-REM stages 3 and 4) also reduces dramatically with age, with the earliest changes evident in males as young as 25.

The increased depth of non-REM sleep in younger subjects accounts for the extreme difficulty often encountered if children need to be woken from deep sleep in the first third of the night. Forced arousals frequently produce apparent confusion or ‘sleep drunkenness’, a phenomenon also frequently observed in sleep-deprived adults.

When is daytime sleepiness abnormal?

Although napping most afternoons is clearly a normal phenomenon, in many Mediterranean cultures, for example, the propensity to fall asleep through the day if rested or unoccupied may well indicate an underlying disorder. Increased somnolence is usually different to simple tiredness or fatigue and particular care is needed in nomenclature and obtaining an accurate history. Some useful leading questions are given in Table 1.2.

Table 1.2 Questions for assessing increased somnolence.

Question or probe

Implications for positive answer

Are you waking up from naps without realising you had been sleep?

Indicates ‘sleep attacks’; can be seen in any cause of severe excessive sleepiness such as narcolepsy

Given the chance, could you nap more than once a day?

Implies excessive sleepiness rather than, for example, simple fatigue

Have you fallen asleep in unusual situations, such as in public (e.g. on buses, in shops)?

Implies significantly excessive sleepiness

Do you regularly fall asleep within a few minutes when a passenger in a car?

May indicate significant sleepiness

Do you routinely fall asleep when watching films on television?

May indicate significant sleepiness

Have friends or family commented on you dropping off to sleep inappropriately?

Usually implies excessive sleepiness, particularly in some situations where it may not be appreciated by the subject (e.g. Parkinson's disease or dementia)

Are you prone to frequent lapses or ‘automatic’ behaviours (e.g. placing objects in inappropriate places or losing items around the house)?

May imply ‘microsleeps’ or brief periods of inattention as an indication of significant sleepiness

At times do you find it very hard to concentrate or take in new information?

Particularly in those who fight the urge to sleep, it might not be appreciated that impaired vigilance and concentration are early markers of significant sleepiness

Do you feel hyperactive at times and unable to focus or attend to a simple task?

Distractibility resembling attention deficit and hyperactivity disorder (ADHD) may reflect underlying sleepiness, particularly in children

Probably 5% of the population can be considered to have excessive daytime somnolence that potentially interferes with daily activities. The dangers of impaired wakefulness or alertness when engaged in tasks such as driving or operating machinery are obvious.

One of the first signs of excessive sleepiness is the presence of minor lapses or ‘microsleeps’. These last up to three seconds, during which time there is incomplete awareness or attention to events in the external world. A further consequence of reduced alertness is impaired cognitive processing, particularly in tasks requiring vigilance and short-term memory. Indeed, it is not uncommon for severe hypersomnolence to masquerade clinically as a dementing illness.

With increasing levels of daytime somnolence, subjects are prone to automatic behaviours, performing tasks on ‘auto pilot’ with no clear subsequent recollection. Typically, objects are either lost around the house or placed in inappropriate locations. A common situation in severe sleepiness is for the subject to carrying on writing or texting whilst half asleep, producing nonsense prose or unintelligible handwriting.

Normal nocturnal motor phenomena during sleep

It is normal to exhibit some movement even during deep sleep. Involuntary simple shifts of body position every 20 minutes or so reflect the commonest phenomenon. However, vigorous so-called hypnic jerks at the point of sleep onset are also common, particularly in young or sleep deprived subjects. Occasionally, such jerks occur in association with alarming sensory symptoms, such as ‘flashes’ or ‘bangs’. The extreme variety of this generally benign phenomenon has been called ‘exploding head syndrome’ and can be a potent cause for sleep-onset insomnia.

During REM sleep, all voluntary movements except those involving the diaphragm and oculomotor muscles are actively inhibited by descending pathways from the brainstem to the motor neurones. However, minor intermittent (myoclonic) jerks of limbs or facial muscles are frequently seen. Arousals to full wakefulness from REM sleep are also very common, especially late in the night, usually in association with vivid dreams or even disturbing nightmares.

Leg movements resembling a slow withdrawal reflex may recur in bursts every 20 seconds or so during all stages of sleep. Such periodic limb movements (PLMs) may be observed in up to a third of the normal elderly population and are most often a benign epiphenomenon. However, if PLMs are particularly violent or observed in association with significant restless legs syndrome, sleep may be disturbed. Drug therapy may be warranted, if only to improve the sleep quality of the bed partner.


At a personal level, everyone appreciates how important good sleep is for optimum health and vitality. However, determining when a reportedly disturbed sleep–wake cycle reflects a defined sleep disorder can be difficult, since many of the associated symptoms form part of a normal spectrum. However, chronically poor sleep, particularly in association with the inevitable adverse effects of natural ageing, may actively fuel many common complaints, such as headaches, generalised pain and depressed mood.

Sleep is also an important and often overlooked consideration in the majority of general medical conditions. Because sleep medicine is a young and poorly defined discipline in most countries, education given to medical students is generally limited. As a likely consequence, the quality of clinical practice varies enormously. The increasingly recognised relevance of sleep medicine to public health will hopefully improve future resources.

Treatments for the whole spectrum of sleep disorders are established and are becoming both evidence based and increasingly sophisticated. A common dilemma in general practice is when to refer to a sleep centre for specialist advice. Severity of symptoms is often a sufficient prompt. However, Table 1.3 provides a selection of ‘red flags’ which, if they accompany a sleep-related symptom, might merit referral for further assessment and possible investigation.

Table 1.3 ‘Red flags’ which might merit referral for further assessment.

Sleep-related symptom

Selected ‘red flags’ from history that might indicate need for further specialist assessment

Excessive daytime sleepiness (not simple tiredness or fatigue)

Feeling the need to sleep in potentially dangerous situations, such as behind the steering wheel

Sudden episodes of sleep with little or no warning of the imperative to sleep (‘sleep attacks’)

Historical evidence for overnight sleep apnoea or waking unrefreshed with severe dryness of mouth

Other symptoms, such as cataplexy, that suggest a primary (central) sleep disorder

Sleep-onset insomnia

Evidence for significant restless legs syndrome before sleep interfering with sleep onset

Inability to rise at a conventional hour might suggest a ‘clock’ problem, such as delayed sleep phase syndrome

Sleep-maintenance insomnia

If there is accompanying frank daytime sleepiness, it is likely there is a defined medical reason for poor quality sleep

The younger the patient, the more likely there is a defined cause for interrupted overnight sleep

Nocturnal motor activity suggesting a parasomnia

Potentially dangerous or frequent behaviours

Accompanying significant daytime sleepiness might indicate presence of another sleep disorder fuelling parasomnia

New onset of parasomnia activity is unusual beyond 40 years of age and might suggest REM sleep abnormalities associated with other disorders such as Parkinson's disease

Features that might suggest an epileptic cause for nocturnal motor activity (Chapter 7)

We are still some way from understanding the true function of sleep but the adverse consequences of its dysfunction are becoming clearer across all areas of general medicine and psychiatry.

Further reading

Altena. E, Ramautar, J.R., Van der Werf, Y.D. and Van Someren, E.J. (2010) Do sleep complaints contribute to age-related cognitive decline? Prog Brain Res185, 181–205.

Axelsson, J., Sundelin, T., Ingre, M. et al. (2010) Beauty sleep: experimental study on the perceived health and attractiveness of sleep deprived people. BMJ341, c6614

Cappuccio, F.P., D'Elia, L., Strazzullo, P. and Miller, M.A. (2010) Sleep duration and all-cause mortality: a systematic review and meta-analysis of prospective studies. Sleep , 33, 582–595.

Frank, M.G. (2006) The mystery of sleep function: current perspectives and future directions. Rev Neurosciences17, 375–392.

Knutson, K.L., Spiegel, K., Penev, P. and van Cauter, E. (2007) The metabolic consequences of sleep deprivation. Sleep Med11, 163–178.

Pollak, C.P., Derlick, P., Linsner, J.P., Wenston, J. and Hsieh F. (1990) Sleep problems in the community elderly as predictors of death and nursing home placement. J Community Health15, 123–135.

Reynolds, A.C. and Banks, S. (2010) Total sleep deprivation, chronic sleep restriction and sleep disruption. Prog Brain Res185, 91–103.



If you find an error or have any questions, please email us at Thank you!