Doctors

Introduction
Sleep disorders are commonly encountered in both paediatric and adult populations, encompassing a wide range of conditions which have been most recently comprehensively categorised in the International Classification of Sleep Disorders, third edition (ICSD-3) (AASM, 2014). A broad understanding of the major diagnostic categories of sleep disorders in the ICSD-3, the rationale for their classification and symptom definitions is an important first step in the accurate diagnosis of sleep disorders. Once the presenting sleep problem is defined and given a diagnostic label from the ICSD-3, often on clinical features alone and sometimes with the help of sleep studies, the management follows published practice parameters (eg. from the American Academy of Sleep Medicine) or suggested algorithms. The most frequently encountered sleep disorders are discussed in detail in each of the major sections focused on the three most common sleep related symptoms presenting to sleep disorders clinics: Insomnia, Excessive Daytime Sleepiness (EDS) and Abnormal Movements/Behaviours in Sleep. Circadian Rhythm Sleep Disorders are discussed separately because they can present as either insomnia or EDS.

International Classification of Sleep Disorders
The ICSD-3 lists over 80 distinct sleep disorders sorted into six major categories, including the Insomnias, Sleep Related Breathing Disorders, Hypersomnias of Central Origin, Circadian Rhythm Sleep Disorders, Parasomnias and Sleep Related Movement Disorders.

• Insomnia is defined as repeated difficulty initiating or maintaining sleep, or poor
  quality, “nonrestorative” sleep in spite of adequate opportunity for sleep, which results in daytime impairment. Insomnia may occur acutely in association with an identifiable stressor, or is chronic, often related to primary medical conditions (eg. disorders that cause pain or breathing problems), psychiatric disorders (eg. depression, anxiety) or other sleep disorders (e.g. sleep apnoea, restless legs syndrome).

• Sleep related breathing disorders (SRBD) are characterised by disturbed
  respiration in sleep, and include obstructive sleep apnoea (OSA), central sleep apnoea (CSA), and sleep related hypoventilation/hypoxaemic syndromes. In CSA there is diminished or absent respiration due to central neurologic or cardiac dysfunction. In OSA there is upper airway obstruction leading to reduced ventilation in the presence of continuing respiratory effort. Sleep related hypoventilation is characterised by reduced ventilation and tidal volume (due to neuromuscular, chest or brainstem disease), hypoxaemia and hypercapnia.

• Hypersomnias of central origin present with EDS and encompass the syndromes
  of narcolepsy, idiopathic hypersomnia and recurrent hypersomnia, each of which has unique defining characteristics and is believed to be caused by intrinsic brain dysfunction. All are rare, with narcolepsy being the least uncommon and is discussed further in the chapter on narcolepsy.

• Circadian rhythm sleep disorders are persistent sleep disturbances primarily due
  to abnormalities of the endogenous circadian timing system or misalignment between this internal circadian clock and the environment. Common examples are shift work and jet lag disorder. Circadian rhythm sleep disorders can lead to insomnia or EDS, and are associated with impaired social and occupational functioning.

• Parasomnias are undesirable events which occur in sleep, and are classified as
  Disorders of Arousal (from non-rapid eye movement or NREM sleep) (eg. confusional arousals, sleepwalking, sleep terrors), Parasomnias Usually Associated With Rapid Eye Movement (REM) Sleep (eg. REM sleep behaviour disorder, nightmare disorder), and Other Parasomnias. Parasomnias usually involve complex behaviours which may lead to sleep disruption and injury to the patient or bed partner, or both.

• Sleep related movement disorders are an important subset of separately classified
  parasomnias characterised by relatively simple movements that disturb sleep. Examples are restless legs syndrome (RLS), periodic limb movement disorder (PLMD), sleep related bruxism and sleep related leg cramps.

Neuroanatomy and Regulation of Normal Sleep
An understanding of what constitutes normal sleep is fundamental to appreciating what happens when sleep becomes abnormal. This is briefly summarised here for clinical correlation.

Sleep is a normal recurring state of loss of responsiveness to the external environment, now known to be an active physiologic state involving dynamic changes in neural, metabolic and cardiorespiratory function rather than a passive state which ensues in the absence of wakefulness. Normal sleep and wake states are generated by a complex neuronal network in the brain. The principal REM sleep generator is localised in the pons while NREM sleep is mediated by a widespread interconnecting network of structures including the brainstem, diencephalon and forebrain. Wakefulness involves the brainstem ascending reticular formation, thalamus, posterior hypothalamus and basal forebrain. A range of neurotransmitters are important in sleep-wake states including acetylcholine, catecholamines, serotonin, histamine and hypocretin, as well as other factors like adenosine and melatonin. Sleep is regulated by homeostatic and circadian mechanisms. Sleep homeostasis refers to the regulatory mechanism which maintains an overall constancy of sleep intensity and duration. Sleep deprivation creates a sleep “debt” which must be paid back, resulting in compensatory heightened pressure to sleep, and eventual increased sleep intensity and duration. Conversely excessive sleep reduces sleep propensity and amount of sleep. The circadian clock, which resides in the suprachiasmatic nucleus of the hypothalamus, determines the physiological level of alertness in an approximately 24-hour cycle, thereby regulating the timing of sleep. This “biological” clock is normally synchronised to the external environment, and can be reset when misaligned, such as with appropriately timed light exposure and administration of melatonin which has chronobiotic properties.

Normal sleep requires intact brain structures as detailed above, and can be disrupted in disease states such as stroke, trauma, inflammation, malignancy, infection (eg. meningitis, encephalitis) and neurodegenerative processes (eg. dementia, Parkinson’s disease). The presentation (which can be insomnia, EDS, abnormal behaviours in sleep or a circadian rhythm sleep disorder) depends on which structures (eg. sleep-wake centres, REM sleep generator, circadian clock) are damaged. For example insomnia may occur in neurodegenerative disorders with fragmented sleep patterns and disruption of sleep-wake cycles, classified as “Insomnia Due to Medical Condition” (AASM, 2005). Excessive daytime sleepiness (“Hypersomnia Due to Medical Condition”) of central origin can be seen in a range of neurological disorders including head injury, brain tumours, stroke, CNS infections, Parkinson’s disease and other neurodegenerative diseases. Abnormal behaviour in sleep associated with neurological disorders is typified by REM sleep behaviour disorder which is seen in 1/3 of patients with newly diagnosed Parkinson’s disease, and 90% of patients with multiple systems atrophy (AASM, 2005). Notwithstanding the importance of intact neural structures to normal sleep, the majority of patients presenting to sleep disorders clinics with the most common sleep related symptoms, which are insomnia or EDS, have normal brains. The most common pathological cause of insomnia seen in sleep disorders clinics is psychological hyperarousal (eg. psychophysiological insomnia, depression, anxiety), while that of EDS is OSA.

Symptoms and Signs
One suggested practical approach when assessing patients with sleep related disturbances would be to elicit symptoms and signs according to the 3 cardinal clinical presentations of sleep disorders ie. insomnia, excessive daytime sleepiness and abnormal movements or behaviours in sleep. The constellation of clinical features in each diagnostic label is usually sufficient to suggest the most likely diagnosis. These 3 symptom categories are discussed in detail below:

(1) Insomnia: Patients with insomnia most commonly describe difficulty with falling asleep, and less frequently, difficulty maintaining sleep or a perception of unrefreshing sleep. Regardless of the cause, insomnia often results in daytime fatigue, general malaise and in severe cases, cognitive and mood disturbances. Chronic insomnia often affects social and occupational functioning, and diminishes quality of life. In patients with insomnia related to medical and psychiatric conditions, associated symptoms include chronic pain or other physical discomfort (eg. restless legs, symptoms of menopause or pregnancy, gastroesophageal reflux, fibromyalgia), respiratory difficulty (eg. asthma, chronic obstructive pulmonary disease), depression, anxiety, and frequently, psychosocial stressors. Neurodegenerative disorders (eg. Parkinson’s disease, dementia) are another common category of disease which is associated with sleep disturbance. Patients with multiple medical problems may also be taking drugs which can cause insomnia, so a careful drug history is important (eg. steroids, bronchodilators, some antidepressants such as fluoxetine).

Chronic insomnia is frequently multifactorial, encompassing psychophysiological factors (eg. stress, depression, anxiety), and sometimes related to drug and underlying disease related components such as chronic pain, as well as maladaptive behavioural aspects. “Inadequate sleep hygiene” is a,common problem across all patients with chronic insomnia, This term refers to the range of well recognized recognised sleep-incompatible behaviours, which include excessive use of substances which disrupt sleep (eg. caffeine, nicotine, alcohol), mentally or physically arousing activities close to bedtime, excessive napping or time in bed, irregular sleep-wake times, and preoccupation with sleep difficulty.

The common causes of insomnia include: Psychophysiological Insomnia, Insomnia Due to Mental Disorder (usually depression and/or anxiety), Adjustment Insomnia, and Inadequate Sleep Hygiene. Restless Legs Syndrome (RLS) is another common cause of sleep-onset insomnia in the West but relatively uncommon locally. Uncommon causes of insomnia are Idiopathic Insomnia and Paradoxical Insomnia (or Sleep State Misperception). Other causes of insomnia which present less frequently in sleep clinics but are commonly experienced are the circadian rhythm disorders, Jet Lag and Shift Work Disorder. Uncommon causes of insomnia are the other circadian rhythm disorders (eg. Delayed Sleep Phase Syndrome, Irregular Sleep-Wake Rhythm), and SRBD (eg. OSA may occasionally present primarily with sleep maintenance insomnia and unrefreshing sleep, instead of snoring with daytime sleepiness which are more classical presentations). Circadian rhythm disorders may present with either insomnia or daytime sleepiness, depending on the level of circadian clock-determined alertness in relation to the external environment.

A detailed history is central to the evaluation of insomnia, and essential for an accurate diagnosis and planning treatment. This includes the habitual sleep-wake schedule on weekdays and weekends, as well as before the onset of insomnia and during its course. Information regarding the nature of activities before sleep and during awakenings in the night is extremely important in identifying maladaptive behaviours perpetuating insomnia. The patient’s beliefs about sleep and fears regarding the effects of insomnia are useful for treatment planning, and will need to be addressed as part of successful therapy. Documentation of all medical and psychiatric problems and medication use is needed. Past treatment attempts should be enquired into which can give insight into the cause of the insomnia and help plan treatment. The pertinent points in a sleep history for insomnia are shown in Table 1.

Besides clinical signs of an underlying medical condition, depression or anxiety, or obesity and craniofacial characteristics predisposing to upper airway obstruction (eg. retrognathia, enlarged tonsils) in OSA patients, the physical examination in patients with insomnia is generally unremarkable.

(2) Excessive Daytime Sleepiness: Excessive daytime sleepiness refers to the inability to stay alert during the major awake period of the day, resulting in falling asleep at inappropriate times. EDS is more likely to occur in monotonous situations when alerting stimuli are absent, and is associated with increased risk of accidents, such as when operating motor vehicles or other machinery. EDS needs to be distinguished from other similar symptoms such as fatigue. The severity of sleepiness can be quantified subjectively using scales such as the Epworth Sleepiness Scale (Johns, 1991) (see appendix 1 in the chapter on “An Approach to Excessive Daytime Sleepiness”), or measured objectively in the sleep laboratory using the multiple sleep latency test (MSLT) or maintenance of wakefulness test (MWT). The MSLT measures the physiological tendency to fall asleep in quiet situations, while the MWT measures the ability to stay awake in quiet situations. The MSLT and MWT are discussed in more detail in the next chapter.

The most common cause of EDS due to a medical condition is OSA. The typical patient is obese, with snoring, witnessed apnoeas, choking or gasping episodes in sleep, and unrefreshing sleep. OSA patients may be obese or have craniofacial or localised structural abnormalities which narrow the upper airway. Less common central causes of EDS are Narcolepsy, Recurrent Hypersomnia (eg. Klein-Levin Syndrome) and Idiopathic Hypersomnia. Circadian rhythm disorders are another uncommon cause of EDS, such as Advanced (EDS in the early evening) or Delayed Sleep Phase Disorder (EDS in the early morning), when the timing of sleepiness is inappropriately misaligned with that of the external environment. A common behavioural cause of EDS is volitional sleep deprivation, which should be excluded in every case of EDS by a careful history of sleep habits and daily sleep duration. Most normal adults require at least 6 to 8 hours of sleep to maintain normal alertness the following day. A habitual sleep period of less than 4 to 5 hours daily is likely to contribute to EDS.

(3) Abnormal Movements or Behaviours in Sleep: These encompass the NREM and REM parasomnias, sleep related movement disorders (mainly RLS and PLMD) and sleep related epilepsy. The NREM parasomnias are disorders of arousal seen usually in the paediatric population, and include confusional arousals, sleep walking and sleep terrors. The REM parasomnias include Nightmare Disorder and REM Sleep Behaviour Disorder (RBD). While the synchronised state of NREM sleep facilitates epileptic activity in general, some epileptic syndromes have a marked tendency to manifest predominantly during sleep, and must be distinguished from the parasomnias, usually requiring EEG documentation of epileptiform discharges. Examples are nocturnal frontal lobe epilepsy and benign epilepsy of childhood with centrotemporal spikes. Epileptic phenomena are characterised by repetitive stereotypic behaviour, but can be difficult to distinguish clinically from non-epileptic phenomena. This category of sleep disturbance usually requires polysomnography (PSG) with full EEG and simultaneous audiovisual recording in addition to history for definitive diagnosis.

The diagnosis of each specific cause of insomnia, EDS or abnormal behaviours in sleep can usually be discerned by recognising the unique characteristics of each specific disorder (ie. pattern recognition), which are discussed in the next section.

Diagnostic Clinical Features of Common Sleep Disorders
Most sleep disorders can be diagnosed by a comprehensive sleep history, which includes a detailed account of routine sleep related habits (eg. bedtime, wake time, number of awakenings), sleep duration, sleeping environment, daytime activities, psychosocial stressors, current drug use and abnormal behaviours in sleep. Important collateral information is often provided by a bed partner or other observer (eg. caregiver or parent) regarding behaviours that the patient may be unaware of, such as snoring and acting out dreams. Sleep questionnaires detailing pertinent sleep related information and sleep logs are often useful (Table 1). Sleep logs are especially helpful in documenting sleep-wake patterns in the circadian rhythm sleep disorders. The Epworth Sleepiness Scale is often used to assess the level of daytime sleepiness and in monitoring the response to therapeutic interventions. A score of 10 or more is considered sleepy. Diagnosis of most sleep disorders can be made on the medical history alone, based on pattern recognition of clinical characteristics determined from the comprehensive sleep history and a physical examination. PSG is needed for confirmation of diagnosis in some conditions like OSA, narcolepsy (with MSLT) and certain parasomnias.

(1) Sleep Disorders presenting with Insomnia: Patients with sleep onset difficulty may have one of the insomnias as classified in ICSD-3. RLS should be considered, as well as a careful history taken to rule out drugs and underlying medical problems which cause insomnia. The circadian rhythm disorders are less common causes of sleep onset insomnia. Also uncommon are sleep maintenance problems alone causing unrefreshing sleep (without snoring or marked EDS) due to OSA or abnormal behaviours in sleep. These strikingly abnormal sleep phenomena are usually evident from the history.

• “Insomnia” Category: Psychophysiological Insomnia, Insomnia Due to Mental Disorder (usually depression and/or anxiety), Adjustment Insomnia, Idiopathic Insomnia, Paradoxical Insomnia (“Sleep State Misperception”).

Psychophysiological Insomnia: Psychophysiological insomnia is also known as “learned insomnia”, “conditioned insomnia” or “primary insomnia”, and constitutes about 12-15% of patients seen at sleep disorders clinics, affecting 1-2% of the general population. It is characterised by a physiological heightened arousal state which predisposes to learned sleep-preventing associations, usually in the setting of social and environmental psychosocial stressors. Individuals with psychophysiological insomnia are typically light or poor sleepers, and may develop chronic insomnia after an initial episode of acute insomnia which failed to resolve following a precipitating stressful event. A counter-productive overconcern with sleep and the consequences of lack of sleep ensue, leading to a mental hyperarousal state (“racing mind”) and a form of conditioned insomnia associated with the individual’s habitual bedtime rituals and sleeping environment. Such individuals typically report sleeping better while away from home and their usual routines, eg. on vacation or during a business trip. This form of insomnia is often perpetuated because healthy sleep incompatible habits develop, such as excessive time in bed tossing and turning, watching the clock, intense preoccupation with sleep, and abuse of prescription sleep aids. This form of insomnia is associated with an increased risk of depression and hypnotic dependence. It may closely resemble insomnia associated with depression or anxiety, which is discussed in the next section.

Insomnia Due to Mental Disorder: Insomnia due to underlying psychopathology (usually depression or anxiety) is one of the most frequently encountered problems at sleep disorders clinics, affecting about 3% of the general population, more in the middle-aged and in women. Insomnia may be a presenting symptom in a variety of psychiatric conditions, including mood, anxiety, psychotic and personality disorders, for which diagnostic criteria are detailed in the Diagnostic and Statistical Manual of Mental Disorders (American Psychiatric Association, 1994). Depression is characterised by a depressed mood or diminished interest in activities, associated with other symptoms such as loss of appetite, fatigue, poor concentration, feelings of worthlessness or guilt, psychomotor retardation and thoughts of death. Insomnia is the most common sleep disturbance associated with major depression, seen in 80-85% of patients, usually manifesting as recurrent or early morning awakenings. In anxiety disorders, difficulty falling asleep is more typical, and accompanies excessive worrying about a range of activities or events. In contrast to psychophysiological insomnia where anxiety is typically focused on sleep difficulty alone, patients with anxiety disorder manifest more pervasive anxiety symptoms attributable to a broader range of reasons. In this diagnostic category which may closely resemble psychophysiological insomnia, the underlying mental disorder plays a key role in the insomnia, with greater persistence and severity of the mood or anxiety disorder.

Adjustment Insomnia: Adjustment or acute insomnia refers to sleep disturbance of relatively short duration (less than 3 months) caused by an identifiable stressor. The one-year prevalence of adjustment insomnia is estimated to be about 15-20%, and is more common in women and older adults. The sleep disturbance may occur after positive or negative events, such as getting a new job, an unexpected windfall, work stress, bereavement or relationship problems, and is expected to resolve once the acute stressor is removed, or when the individual has adapted to the triggering circumstances.

Idiopathic Insomnia: Idiopathic or childhood-onset insomnia has an estimated prevalence of about 1%, and accounts for less than 10% of patients seen at a sleep disorders clinic. The cause of this form of insomnia which manifests in infancy or childhood, and persists throughout adult life, is not well understood. Genetic factors or abnormalities in the normal sleep-wake mechanisms may be involved, though none have been identified. It is unrelated to other factors such as psychosocial stressors, physical or psychiatric disorders. The very early onset and absence of any identifiable cause or trigger distinguish idiopathic insomnia from other forms of insomnia.

Paradoxical Insomnia: Paradoxical insomnia is also known as sleep state misperception, and is probably found in less than 5% of patients presenting with sleep disorders. Patients complain of insomnia symptoms similar to other forms of insomnia, but do not have any objective sleep disturbance when their sleep is recorded in the laboratory. The diagnosis of paradoxical insomnia therefore requires some form of objective sleep recording, such as PSG or actigraphy. Patients may report exaggerated symptoms such as little or no sleep or staying awake all night for prolonged periods, suggesting extreme sleep deprivation which is physiologically improbable and incompatible with the degree of daytime functioning that they are able to manage.

• Sleep Related Movement Disorders: Restless Legs Syndrome (RLS)

Restless Legs Syndrome: RLS is a clinical diagnosis made based on fulfilling 4 essential criteria originally proposed in 1995 by the International RLS Study Group, and modified in 2002 (Walters, 1995). The 4 Essential Criteria for RLS are:

1. An urge to move the legs, usually accompanied or caused by uncomfortable and unpleasant sensations in the legs.
2. The urge to move or unpleasant sensations begin or worsen during periods of rest or inactivity such as lying or sitting.
3. The urge to move or unpleasant sensations are partially or totally relieved by movement, such as walking or stretching, at least as long as the activity continues.
4. The urge to move or unpleasant sensations are worse in the evening or at night than during the day or only occur in the evening or at night.

RLS affects 5-15% of Caucasian populations, more often women than men. It appears to be less common in Asian populations. Sensorimotor symptoms predominate at bedtime, and give rise to sleep onset insomnia. RLS is associated with periodic limb movements in sleep (PLMS) in 80-90% of cases, which may contribute to sleep maintenance problems if these are associated with arousals. RLS may be idiopathic, or related to iron deficiency, peripheral neuropathy, uraemia or pregnancy. The diagnosis is usually straightforward in patients with prominent sensory symptoms (usually in the legs) in wakefulness prior to sleep onset, which fulfill the clinical diagnostic criteria listed above. RLS is often confused with Periodic Limb Movement Disorder (PLMD) which is a controversial disorder of uncertain clinical significance manifesting in sleep rather than wake, causing sleep maintenance rather than sleep onset insomnia as a result of repetitive, stereotyped (triple flexion) leg movements which cause repeated arousals. While RLS is a clinical bedside diagnosis, PLMS requires documentation of frequent PLMS with associated arousals and sleep fragmentation on PSG.

• Circadian Rhythm Sleep Disorders: Jet Lag, Shift Work Disorder, Delayed Sleep Phase Disorder, Irregular Sleep-Wake Rhythm.

Jet Lag Disorder: Jet lag disorder is a temporary condition which occurs after air travel across at least 2 time zones. Symptoms of insomnia (or EDS) occur because the endogenous circadian clock is initially misaligned with the external environment. Besides insomnia, associated symptoms may include general malaise and gastrointestinal upset.

Shift Work Disorder: Shift workers who have to work during the body’s usual sleep period as determined by the endogenous circadian clock often complain of sleep onset insomnia in the morning after the night shift (and conversely, excessive sleepiness when working at night). Shift work disorder is usually evident from a careful review of the work schedule, and typically resolves when the sleep period is restored to a conventional time. Like most circadian rhythm disorders, the diagnosis can be made from history.

Delayed Sleep Phase Syndrome: Delayed sleep phase syndrome has an estimated prevalence of 7 to16% among adolescents and young adults. It is characterised by a delayed habitual sleep phase, which is usually more than 2 hours or more later than conventional sleep-wake times. Such patients have a “night owl” tendency, and report difficulty falling asleep at conventional bedtimes, or difficulty waking up at socially acceptable morning hours, or both. The late bedtime and wake-up times often disrupt work or school schedules. In the absence of social pressures to go to bed or wake up early (such as having to go to work or school at conventional times), people with delayed sleep phase exhibit stable sleep wake patterns with no sleep related disturbances.

Irregular Sleep-Wake Rhythm: This circadian rhythm sleep disorder is found usually in institutionalised elderly patients who have chaotic sleep wake rhythms as a result of poor sleep hygiene and a lack regular scheduled activity which would help to synchronise their circadian clocks to the external environment, such as sunlight exposure, physical or social activity. Such people may feel sleepy and nap at various times of day, and have difficulty falling or staying asleep at night. The sleep-wake pattern is very disorganised, with irregular sleep periods scattered throughout the day.

(2) Sleep Disorders presenting with Excessive Daytime Sleepiness: The most common non-pathological cause of daytime sleepiness is probably volitional lack of adequate sleep. Habitual sleep duration should be part of the sleep history, sleeping less than 4 to 5 hours is generally insufficient to maintain normal daytime alertness. The most common cause of EDS seen in sleep disorders clinics is OSA. Narcolepsy and the other hypersomnias of central origin are less common. Again, the clinical characteristics are often sufficient to make a clinical diagnosis, though sleep studies are often required to definitely rule out OSA in patients with EDS, which can also co-exist with narcolepsy for which PSG (with MSLT) is also needed for diagnosis.

• Sleep Related Breathing Disorders: OSA

Obstructive Sleep Apnoea: OSA is characterised by recurrent episodes of complete (apnoeas) or partial (hypopnoeas) upper airway obstruction during sleep, often associated with oxygen desaturation and recurrent arousals which are usually quantifiable and confirmed on PSG. The prevalence of OSA syndrome (PSG confirmation in the presence of EDS) reported from American studies has been estimated to be about 4% in men and 2% in women. One local study has suggested that OSA may be much more common in Singapore, affecting as many as 15% of the local population (Puvanendran, 1999). Risk factors are obesity and craniofacial abnormalities which narrow the upper airway, such as retrognathia or adenotonsillar enlargement. Other risk factors are a large neck circumference, menopause, smoking, and endocrine disorders like hypothyroidism and acromegaly. OSA has been associated with hypertension, ischaemic heart disease, stroke and diabetes. The classical history which often suggests the diagnosis of OSA includes snoring, EDS, witnessed apnoeas and choking/gasping episodes by a bed partner and unrefreshing sleep regardless of duration, typically (but not always) in an obese or overweight individual. Confirmation of diagnosis is by PSG showing at least 5 or more apnoeas or hypopnoeas per hour (ie. an apnoea-hypopnoea index or AHI > 5).

Hypersomnias of Central Origin: Narcolepsy, Recurrent Hypersomnia (eg. Klein-Levin Syndrome), Idiopathic Hypersomnia

Narcolepsy: Narcolepsy is a very rare hypersomnia of central origin affecting 0.013% to 0.18% of Western populations, with onset usually in adolescence or young adulthood (ages 15-25 years). It is characterised by EDS, cataplexy, often irresistible sleep attacks, sleep paralysis (transient inability to move or speak) and hypnagogic/hypnopompic hallucinations (usually vivid audiovisual phenomena which occur upon falling to sleep/waking up). Classically, patients with narcolepsy report short, refreshing naps, usually followed by 2-3 hours of alertness. Not all narcoleptic patients report cataplexy, the hallmark of narcolepsy, which refers to a sudden loss of muscle tone provoked by strong emotions such as laughter or anger, with preserved consciousness. These patients form a subgroup listed in ICSD-2 as “narcolepsy without cataplexy”. Although a diagnosis can be made by clinical history in typical cases with all the classic features, PSG with MSLT is usually performed to rule out other common causes of EDS like OSA. The MSLT usually shows a very short sleep onset latency, and sleep onset REM periods (SOREMPs) which refers to the abnormal appearance of REM sleep (usually seen 90-120 minutes into sleep) in a 20-minute nap.

Klein-Levin Syndrome: Recurrent or periodic hypersomnia is another rare condition, of which the most well known is Klein-Levin Syndrome. This disorder which usually presents in adolescence is characterised by recurring prolonged bouts of sleeping up to 18 hours a day, lasting days to weeks, associated with behavioural abnormalities such as hypersexuality, hyperphagia, cognitive disturbances and aggressiveness. In between these periodic somnolent phases, sleep and behaviour are normal. The diagnosis is usually suggested by the bizarre and striking clinical features. Important differential diagnoses are brain tumors, head injury or encephalitis which can also cause EDS associated with cognitive and/or behavioural abnormalities and should be excluded.

Idiopathic Hypersomnia: This form of hypersomnia is characterised by EDS which presents at a young age, usually before 25 years, and is distinguished from narcolepsy by the absence of cataplexy and other narcolepsy-associated phenomena (eg. sleep paralysis, hypnagogic hallucinations). Also unlike narcoleptics who take short refreshing naps, idiopathic hypersomnia patients have long, unrefreshing daytime sleep episodes. On MSLT, such patients also do not have SOREMPs which are seen in narcolepsy.

• Circadian Rhythm Sleep Disorders: Jet Lag, Shift Work, Delayed (as above) and Advanced Sleep Phase Disorders can all present with EDS.

Advanced Sleep Phase Disorder: Advanced sleep phase disorders are seen in about 1% of middle-aged and older adults, and characterised by sleep-wake times that are several hours earlier than conventional or desired times. There is a stable advance of the habitual sleep period, eg. sleeping at 6pm and waking at 2am. Such patients complain of EDS in the late afternoon and early evening, and spontaneous early morning awakenings.

(3) Sleep Disorders presenting with Abnormal Movements or Behaviours in Sleep: The parasomnias, sleep related movement disorders and sleep related epilepsy often have unique clinical characteristics which suggest the correct diagnosis. In doubtful cases (or when epilepsy is suspected), PSG (with expanded EEG montage) is needed.

NREM parasomnias: Confusional arousals, sleepwalking and sleep terrors

Confusional arousals: Confusional arousals occur in children and young adults, consisting of episodes of confusion, disorientation and inappropriate behaviour when partially awoken from deep NREM sleep, which can either be spontaneous or forced. Predisposing factors include rebound deep sleep when recovering from sleep deprivation, and sleep disorders which cause arousals such as OSA and PLMD. While confusional episodes occurring in the first part of the night when deep NREM sleep predominates may be characteristic, occasionally the bizarre behaviour seen in confusional arousals may need to be distinguished from other causes of abnormal behaviour in sleep, such as nocturnal epilepsy or REM parasomnias, usually with PSG.

Sleepwalking: Sleepwalking or somnambulism consists of complex behaviours while in NREM sleep, comprising walking around in an impaired state of consciousness, with or without other actions ranging from violent behaviour, driving a car, to climbing out a window. The prevalence of sleepwalking is higher in children (up to 40%) than adults (up to 4%). Precipitating factors include sleep deprivation, sleep disorders which precipitate arousals, febrile illness in children, and physical or emotional stress in adults. The main concern of sleepwalking is the risk of self-injury eg. when someone climbs out a window. Any underlying triggers should be identified and treated appropriately.

Sleep Terrors: Sleep terrors or night terrors occur in 2-3% of children and adults, consisting of arousals from deep NREM sleep and characterised by intense behavioural manifestations of fear and autonomic hyperactivity (eg. tachycardia, tachypnoea, pupillary dilatation, diaphoresis). The episodes are associated with frightening dreams, confusion, disorientation and amnesia following each episode. They may be difficult to distinguish from other parasomnias or sleep related epilepsy, which may require PSG with additional EEG channels for diagnosis.

REM parasomnias: Nightmare Disorder, REM Sleep Behaviour Disorder

Nightmare Disorder: In nightmare disorder which affects 2-8% of the general population, and up to 50% of young children, recurrent frightening dreams occur in REM sleep, which often result in awakenings and sleep disruption. Patients usually can recall details of their disturbing dream on awakening. In adults frequent nightmares have been associated with physical or emotional trauma, stress and psychopathology. Nightmares are characterised by detailed recollection of bad dreams (unlike night terrors when there is usually amnesia for the event), and should be differentiated from other parasomnias and nocturnal panic attacks. Underlying psychological disturbances should be identified.

REM Sleep Behaviour Disorder (RBD): In RBD, which affects mainly older men (prevalence of 0.3-0.5% of the general population), there is dream enactment behaviour which causes injury (to self or bed partner) and sleep disruption. Typically such episodes consist of acting out unpleasant or violent dreams with behaviours such as shouting, swearing, punching, kicking, running and jumping, and are reported because of sleep related injury, usually occurring in the last one-third of the sleep period. RBD is associated with neurodegenerative disorders, such as Parkinson’s disease, and can also be acutely triggered by psychotropic medications, or withdrawal from alcohol and sedative-hypnotic agents. As with the other parasomnias, PSG may be required to rule out seizures. Specific treatment is needed to prevent injury.

• Sleep Related Movement Disorders: RLS, PLMD Restless Legs Syndrome and Periodic Limb Movement Disorder have been discussed in detail in a separate article.

Summary
There are a large number of sleep disorders, but only a handful
of common ones, most of which can be diagnosed from history by clinical pattern recognition. Sleep disorders are common, and have a limited range of symptomatology. A practical approach to diagnosis – which is mainly from clinical history – would be to consider the various differential diagnoses from the 3 main symptom categories of insomnia, excessive daytime sleepiness and abnormal movements in sleep. Insomnia alone is a symptom and not a diagnosis. A sleep questionnaire is helpful in collecting the information which would be useful for diagnosis. There are a range of causes of insomnia which is often psychologically based and multifactorial, each with different management approaches. An accurate diagnosis is the first step, with specific therapy directed to the cause identified, eg. antidepressants for depression, dopaminergic agents for RLS, behavioural strategies for psychophysiological insomnia. Chronic partial sleep deprivation and OSA are the most common causes of excessive daytime sleepiness in practice. Parasomnias are much less frequently encountered sleep disorders. While a typical history is often suggestive of the diagnosis, PSG is sometimes needed. Sleep studies are indicated in the diagnosis of certain conditions such as suspected sleep related breathing disorders, narcolepsy (PSG with MSLT), injurious parasomnias and sleep related epilepsy. Many sleep disorders are multifactorial and respond to a range of treatment modalities, from lifestyle (eg. diet, exercise), behavioural (eg. planned sleep schedules, scheduled naps, good sleep hygiene techniques), pharmacotherapy (eg. sedative-hypnotics, CNS stimulants, psychotropic drugs), timed bright light therapy, positive airway pressure therapy and surgery (for OSA). These are discussed in subsequent chapters in greater detail. Once an accurate diagnosis is made, a multifaceted and holistic approach achieves the best long term outcomes.

Table 1: Sleep Questionnaire

References
American Academy of Sleep Medicine. International Classification of Sleep Disorders 3rd Ed.: American Academy of Sleep Medicine; 2014.

Michael J Sateia. International classification of sleep disorders-third edition: highlights and modifications. Chest. 2014 Nov;146(5):1387-1394. doi: 10.1378/chest.14-0970. (https://pubmed.ncbi.nlm.nih.gov/25367475/)

American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 4th ed. Arlington: American Psychiatric Publishing Inc.; 1994.

Johns MW. A new method for measuring daytime sleepiness: the Epworth Sleepiness Scale. Sleep. 1991; 14: 540-5.

Puvanendran K, Goh KL. From snoring to sleep apnea in a Singapore population. Sleep Res Online. 1999; 2(1):11-4.

Walters, AS. “Toward a better definition of the restless legs syndrome. The International Restless Legs Syndrome Study Group.” Mov Disord. 1995; 10(5): 634-42.

INTRODUCTION
Polysomnography (PSG) is the recording of multiple physiological parameters during sleep, including electroencephalography (EEG), electromyography (EMG), electro-oculography (EOG), electrocardiography (ECG) and respiration used in the diagnosis and treatment of sleep disorders. Modern day sleep studies have evolved from the earliest electrophysiologic techniques (beginning with the discovery of EEG in 1929 by Hans Berger) and have led to the identification of two distinct sleep states, non-rapid eye movement (NREM) and rapid eye movement (REM) sleep, and the normal sleep cycle.

NORMAL SLEEP CYCLE
The description of normal sleep architecture in 1957, dividing sleep into NREM and REM sleep, was an important milestone in the history of human sleep research (Dement and Kleitman, 1957). The normal adult sleep cycle comprises 5 to 7 successive cycles of NREM and REM sleep, each lasting about 90 minutes, progressing through stages N1, N2, N3 and R, with stage N2 being the transitional stage between NREM and REM stage. In adults, the majority of sleep comprises stage N2 (about 50% of normal sleep), with N1 being more transient (2 to 5% of sleep time), and deep sleep (stage N3) and REM sleep (stage R) making up the remainder of total sleep time (15 to 20% and 20 to 25% respectively). Sleep physiology changes with age as the brain matures and eventually degenerates. Neonates have less well defined sleep EEG features, longer sleep duration (at least 16 hours), shorter sleep cycles (50 to 60 minutes) and polyphasic sleep spread throughout the day. In this age group, sleep is divided into quiet (NREM) and active (REM) sleep, with active sleep comprising up to 50% of total sleep time, and frequently being the first stage of sleep entered into from wakefulness (“sleep onset REM”). Slow wave sleep is prominent in young children, accounting for 30-40% of sleep time. With maturity, sleep becomes consolidated into a single nocturnal period, with reduced amounts of REM sleep, and sleep architecture approaching that of adulthood. With advancing age, there is decline in the percentage of deep sleep, more frequent awakenings, sleep fragmentation and periodic limb movements in sleep (PLMS). In the elderly sleep disorders like obstructive sleep apnoea (OSA) also occur more frequently.

NREM sleep is characterised by synchronised EEG activity, muscle relaxation and decreased heart rate, blood pressure and tidal volume. It is divided into light NREM sleep stages (stages N1, N2) and deep sleep (stage N3), with defining PSG features originally outlined in the classic 1968 Rechtschaffen and Kales scoring manual, and most recently updated in 2007 in the AASM Manual for the Scoring of Sleep and Associated Events (Rechtschaffen and Kales, 1968; Iber et al., 2007) .

REM sleep (stage R) is also known as paradoxical sleep because it resembles wakefulness with desynchronized EEG activity, phasic events such as rapid eye movements and bursts of muscle activity. REM sleep is characterized by dreaming, with subjects reporting dreaming in about 85% of awakenings when roused from REM sleep (although dreaming also occurs in NREM sleep).

POLYSOMNOGRAPHY
PSG involves monitoring and recording EEG, EMG, EOG, ECG and other physiologic data used to analyse sleep architecture, cardiopulmonary function and movements in sleep. Different NREM and REM sleep stages can be identified based on specific EEG, EMG and EOG characteristics. Monitoring respiratory parameters and ECG allows simultaneous documentation of sleep related cardiorespiratory disturbances in conditions such as OSA. Additional EEG channels and video-recording may be required for the assessment of abnormal behaviours in sleep, mainly to distinguish between NREM and REM parasomnias, and nocturnal sleep related epilepsy.

The International Classification of Sleep Disorders lists over 80 distinct sleep disorders sorted into eight categories, including the Insomnias, Sleep Related Breathing Disorders, Hypersomnias of Central Origin, Circadian Rhythm Sleep Disorders, Parasomnias and Sleep Related Movement Disorders (AASM, 2014). Many of these conditions are discussed elsewhere in this book. Not all sleep disorders require a PSG for diagnosis, each patient should have a detailed clinical evaluation first.

PSG is routinely indicated for the diagnosis of sleep related breathing disorders (eg. OSA, central sleep apnoea), for continuous positive airway pressure (CPAP) titration in sleep apnoea, sleep related movement disorders (eg. periodic limb movements in sleep) and abnormal behaviours in sleep (eg. parasomnias, sleep related epilepsy) (Kushida et al., 2005). PSG is also used with the multiple sleep latency test (MSLT) in the evaluation of narcolepsy. Occasionally PSG is performed to evaluate sleep maintenance insomnias presenting with recurrent, unexplained awakenings. However PSG and MSLT should not be routinely used to screen or diagnose patients with insomnia complaints (Chesson et al, 2000). There are no absolute contraindications to PSG, nor any serious safety issues.

MULTIPLE SLEEP LATENCY TEST & MAINTENANCE OF WAKEFULNESS TEST
The objective measurement of sleepiness in an important part of the assessment pathlogical sleepiness in certain sleep disorders which present with excessive daytime sleepiness (EDS) such as narcolepsy. The most commonly used diagnostic tool for measuring EDS in the sleep laboratory is the MSLT, which complements sleep questionnaires such as the Epworth Sleepiness Scale (ESS) (Johns, 1991). The ESS gives is a rough estimate of the degree of daytime sleepiness, and is discussed in more detail in the next section on Excessive Daytime Sleepiness.

The MSLT is a validated objective measure of the ability or tendency to fall asleep (Littner et al., 2005). The MSLT was first described in 1976 as a research tool (Carskadon and Dement, 1997). The MSLT is considered to be the gold standard in the objective evaluation of EDS. Four to five naps are recorded at two hour intervals, after which the average time to fall asleep (sleep onset latency) over the four or five naps is calculated, the mean sleep latency (MSL). A MSL of less than 5 minutes is considered to be pathological and correlates with severe sleepiness. There are 2 general indications for the MSLT. The first is to evaluate patients for a diagnosis of narcolepsy. In narcolepsy there is a very short MSL and at least 2 sleep onset REM periods (SOREMPs) on MSLT. REM sleep normally occurs 90 to 120 minutes after sleep onset, SOREMPs refer to the appearance of REM sleep after less than 15 to 20 minutes of sleep onset. SOREMPs are not pathognomonic of narcolepsy, and can occur in sleep deprivation and other sleep disorders like OSA. The MSLT may also useful in the evaluation of patients with suspected idiopathic hypersomnia. Narcolepsy and idiopathic hypersomnia are discussed in more detail in the section on Excessive Daytime Sleepiness. The second indication for MSLT is to document the degree of sleepiness in an objective manner. For example, it may be important in the setting of OSA to assess the degree of EDS for safety reasons, or to document treatment efficacy eg. of CPAP.

The Maintenance of Wakefulness Test (MWT) is a variation of the MSLT and measures the ability to stay awake for a defined time rather than the tendency to fall asleep. The MWT was originally described as a measure of treatment efficacy in patients with EDS (Mitler et al., 1982). Similar to the MSLT, there are a series of nap trials, which may be 20 or 40 minutes in duration. A normal MSL in the 20 minute protocol is 18 minutes (representing 1 standard deviation below normals), while a MSL of less than 11 minutes represents impaired wakefulness. In the 40 minute protocol, a normal MSL is 35 minutes or greater. A MSL of less than 19.4 minutes is considered to be abnormal (Doghramji et al., 1997). The MWT may be indicated to assess the treatment response in narcolepsy and idiopathic hypersomnia, and to assess wakefulness in patients in whom the inability to remain awake represents a public or personal safety concern (Littner et al., 2005)

ROLE OF SLEEP STUDIES IN OSA
A range of sleep studies are commercially available in Singapore for the evaluation of sleep related breathing disorders, the most common condition being OSA. Ideally all such patients should be formally evaluated with PSG in a Sleep Laboratory, with a trained sleep technologist in attendance. When such laboratory facilities are unavailable, home sleep studies are performed which can be comprehensive (ie. similar recording parameters as a full PSG in a sleep laboratory, except done in a home setting) or partial (usually only cardiorespiratory parameters ie. respiratory effort and airflow, pulse oximetry and ECG). Practice parameters set forth by the American Academy of Sleep Medicine have set out clear guidelines for the appropriate use of each of these types of studies (Kushida et al., 2005). The appropriate diagnostic test should be determined by a sleep trained physician or surgeon based upon the presenting features and clinical judgment.

Polysomnography
PSG is the current reference or “gold standard” for the evaluation of sleep and sleep related breathing disorders (Kushida et al., 2005). For sleep related breathing disorders, the number of apnoeas and hypopnoeas per hour of sleep is expressed as the “apnoea-hyponoea index” (AHI). When a portable monitor is used which does not measure sleep stages (ie. a partial study), the term “respiratory disturbance index” (RDI) is used to refer to the apnoeas or hypopnoeas per hour of recording.

Portable Monitoring Devices
Portable monitors are commonly used locally for the diagnosis of OSA. The American Academy of Sleep Medicine has published guidelines on the appropriate use of such devices (Chesson et al., 2003). Sleep monitoring procedures are categorized as Type 1 to 4, in which Type 1 refers to the standard attended (ie. with a sleep technologist in attendance throughout the sleep recording) PSG in a sleep laboratory, Type 2 refers to a comprehensive but unattended PSG usually done at home, Type 3 refers to modified portable cardiorespiratory sleep studies which do not record EEG or EMG (and therefore cannot monitor sleep stages), and Type 4 refers to a continuous single or dual bioparameter recording such as ambulatory overnight pulse oximetry.

The high cost and limited availability of sleep laboratory PSG has led to the development of simpler and less expensive diagnostic techniques based on a smaller selection of signals than is traditionally recorded by PSG. Portable wrist worn devices which record peripheral arterial tone, heart rate, oxygen saturation and actigraphy, and provide an automatic analysis of the RDI, AHI, oxygen desaturation index (ODI) and sleep wake state have been studied (Zou et al., 2006). The peripheral arterial tone signal measures finger pulsatile volume, which changes during sympathetic nervous system activation associated with respiratory events. Sleep (with decreased movement) and wake (increased movement) are assessed by actigraphy which provides an estimate of sleep and wake by analysing movement data. Validation studies in patients with OSA and those not preselected for OSA symptoms have suggested that such devices may provide sleep data which correlates closely with that recorded from home or laboratory PSG (Zou et al., 2006; Pittman et al., 2004). Certain conditions may affect the sensor function of such devices, such as peripheral vascular disease, peripheral neuropathy, cardiac arrhythmias, status post cervical or thoracic sympathectomy, the presence of a permanent pacemaker, and the use of -adrenergic receptor-blocking agents. No formal guidelines have been issued regarding this emerging technology on its appropriate use in clinical practice.

Overall the use of portable monitoring devices is not recommended for general screening purposes, in patients with comorbid conditions, and to diagnose sleep disorders other than OSA. Such studies should be interpreted by physicians with sleep training and familiarity with the devices and their limitations. Careful scrutiny of the raw data is needed by both the sleep technologist and physician.

Diagnosis of OSA
Full-night (Type 1) PSG is recommended for the diagnosis of sleep related breathing disorders. The 2003 AASM guidelines on the use of portable monitoring devices indicate that there is insufficient evidence to recommend the routine use of Type 2 devices (which are also reported to have a high rate of data loss in unattended settings), and that Type 3 cardiorespiratory sleep studies should only be used in an attended setting in appropriate patients suspected of OSA. Such patients should be carefully selected with a high pre-test probability for OSA, and free from significant co-morbid conditions such as congestive heart failure, coronary artery disease, significant arrhythmias and stroke. Symptomatic patients with negative partial studies using portable monitoring devices should undergo full (Type 1) PSG to truly exclude OSA. The routine use of Type 4 devices such as ambulatory overnight pulse oximetry for the diagnosis of OSA is not recommended, either in attended or unattended settings.

CPAP Titration Studies for OSA
PSG is needed for continuous positive airway pressure (CPAP) titration in patients diagnosed to have a sleep related breathing disorder for which positive airway pressure therapy is indicated. Previously, an initial full-night PSG was followed by a second night CPAP titration study. A more cost and time efficient alternative which is commonly performed locally is the “split-night study” in which the first half of the PSG is diagnostic, and is followed by CPAP titration in the second half of the study on the same night. Split-night studies are an acceptable alternative to the full-night PSG and CPAP titration PSG over 2 nights provided the following criteria are met:

1. AHI > 30 during a minimum of 2 hours of a diagnostic PSG; or an AHI in the 20-40 range based on clinical judgment (eg. the presence of long respiratory events with severe oxygen desaturation).
2. CPAP titration is carried out for a minimum of 3 hours
3. PSG documents elimination or near elimination of respiratory events during NREM and REM sleep, including REM sleep in the supine position.

A second full night of PSG for a full CPAP titration study is recommended if criteria 2 and 3 are not fulfilled.

Pre and Post-Operative Sleep Studies
It is recommended that patients be formally evaluated for OSA using PSG or attended (Type 3) cardiorespiratory sleep studies before they undergo upper airway surgery for snoring or OSA. Such studies are also routinely indicated post-operatively in patients with moderate to severe OSA to document satisfactory outcome.

Assessment of OSA Treatment Outcomes
Besides assessing surgical treatment outcomes, PSG or attended (Type 3) cardiorespiratory sleep studies are also routinely indicated for the evaluation of treatment results in moderate to severe OSA patients who have had good clinical response to oral appliance treatment, in patients who have lost a substantial amount of weight (eg. 10% of body weight) to ascertain if CPAP is still needed at previously titrated pressures, in patients who have gained a substantial amount of weight (eg. 10% of body weight) while on CPAP treatment and experience recurrent symptoms to ascertain if pressure adjustments are needed, when symptoms recur in spite of good initial response to any treatment, or when clinical response is insufficient. When there is suboptimal response to adequate treatment for OSA, a co-existing sleep disorder may be present eg. narcolepsy, which may require additional studies, such as repeat PSG followed by MSLT.

MSLT IN OSA
MSLT is not routinely indicated for the diagnosis of OSA or in assessing response to treatment (Littner et al., 2005). As discussed above, MSLT (following an overnight PSG) may be indicated to rule out narcolepsy if there is persisting daytime sleepiness even after optimal treatment for OSA.

TECHNICAL CONSIDERATIONS IN POLYSOMNOGRAPHY

Preparation by the Sleep Technologist
The standard attended overnight PSG requires the presence of a qualified sleep technologist to ensure patient compliance and quality of the study. The available clinical information is first reviewed by the sleep technologist to verify the indication for the study and the appropriate study protocol to follow eg. routine overnight PSG for diagnosis of OSA, CPAP titration study, split-night study, overnight PSG with audiovisual and additional EEG electrodes for sleep related epilepsy. Testing procedures are explained to the patient to allay anxiety and encourage co-operation, enhancing the likelihood of performing a good quality study. If a CPAP titration study is to be performed, this is explained to the patient beforehand, known as “CPAP education”. The PSG is typically scheduled at the patient’s usual sleeping time.

Preparation by the Patient
Patients are encouraged to visit the sleep laboratory to familiarize themselves with the environment prior to testing and to bring any personal items which may facilitate relaxation and sleep. Brochures containing information related to the PSG are given to the patient and any questions addressed prior to the appointment so that they will better understand the procedures and be able to follow instructions. Patients may be asked to discontinue medication which can affect sleep architecture if there are no contraindications to doing this.

Setting
The sleep laboratory is ideally a quiet place conducive to sleep, simulating the bedroom environment at home, situated away from the main hospital. Bathroom facilities, light and temperature control, television, a nightstand and other amenities are made available to enhance patient comfort. Recording equipment is stored in a separate room to minimize disturbance and audiovisual communication maintained at all times via closed circuit television and 2-way intercom. Emergency medical coverage is mandatory throughout the study, provided by sleep physicians on call or other physicians within the hospital.

Portable systems are available for use outside the sleep laboratory and have several advantages. These include lower cost, less first night effect when studied at home, convenience for the patient and availability in places where there are no sleep laboratory facilities. These advantages should be weighed against factors such as poorer quality data collection if the study is unattended and lack of control of the sleeping environment. Portable sleep studies may be indicated where standard sleep laboratory PSG is not available, when patients cannot be studied in the sleep laboratory, and for follow-up after therapeutic intervention based on findings from a diagnostic standard PSG.

PSG Recording
During polysomnography data is recorded, which is amplified and converted to a digital signal. Data recorded includes EEG, EMG, EOG, ECG, the airflow channel, respiratory effort channels, the snore microphone and pulse oximeter. Audiovisual (including video-EEG) recording is added when parasomnias, seizures or other paroxysmal events are suspected.

EEG records brain waveforms used to identify different sleep stages. EMG recording of muscle activity is useful in several ways. The most important is the chin EMG which monitors the axial muscle tone, useful in distinguishing REM (when muscle tone is the lowest of the study) from NREM sleep stages (when muscle tone is relatively elevated) and identifying movement arousals in REM sleep. EMG activity in the leg (tibialis anterior) and forearm (extensor digitorum) muscles reflects excessive leg and arm movements in wakefulness and in sleep. This may be useful in the evaluation of abnormal movements, such as in the sleep related movement disorders and parasomnias.
ECG recording is used to record heart rhythm irregularities associated respiratory events.

Normal respiration can be evaluated by measuring airflow into the lungs and the degree of respiratory effort, and monitoring oxygen saturation and carbon dioxide (CO2) levels during the study. Airflow is frequently measured using nasal pressure transducers or nasal thermistors. Respiratory effort can be assessed using respiratory belts and intercostal muscle EMG.

In adults an apnoea refers to complete cessation of airflow for ten seconds or more. A hypopnoea refers to a drop in airflow by 30-50% for ten seconds or more. Apnoeas are classified as either central, obstructive or mixed. A central apnoea is scored when there is no evidence of respiratory effort in the thoracoabdominal respiratory belts and therefore no airflow lasting > 10 seconds. An obstructive apnoea is scored when there is evidence of ongoing respiratory effort with reduced or absent airflow lasting > 10 seconds. A mixed apnoea is scored when a central event is followed by a return of respiratory effort. The 10-second duration for the criteria of apnoeas and hypopnoeas is based on normal adult respiratory rates, during which approximately two breaths are lost. Children have higher respiratory rates, correspondingly the duration of clinically significant respiratory events is shorter.

Oesophageal balloon manometry is the most accurate measure of respiratory effort, using a nasogastric tube with a balloon tip placed in the distal oesophagus. A pressure transducer detects changes in intrathoracic pressure, which gives an estimate of the respiratory effort. This technique is a sensitive measure of upper airway obstruction but is not widely available and may be poorly tolerated and disturb sleep because it is invasive. PSG with esophageal balloon manometry is indicated when upper airway resistance syndrome (UARS) is suspected clinically. UARS is discussed more in detail in the chapters on OSA. Typically the pleural pressure becomes progressively more negative as respiratory effort increases, leading to an arousal.

Oxygen saturation is monitored by a pulse oximeter. Carbon dioxide (CO2) levels can be measured using transcutaneous CO2 electrodes. CO2 retention occurs in hypoventilation which is seen in obesity, restrictive or obstructive pulmonary disease and neuromuscular disease, especially in REM sleep. Snoring and snorting are detected by small microphones attached to the patient’s neck.

Oesophageal pH studies are used in the evaluation of gastroesophageal reflux (GER). GER can cause frequent arousals during sleep, leading to insomnia or excessive daytime somnolence. Polysomnography with oesophageal pH monitoring is useful in evaluating frequent unexplained arousals or suspected sleep-related GER with aspiration. A pH sensitive probe is placed in the distal esophagus at the level of the lower oesophageal sphincter and referenced to the forearm or the forehead. Reflux is defined as a fall in the distal oesophageal pH below 4, the normal value being about 6.

If CPAP titration or a split-night study is performed, the patient is given CPAP education beforehand, which usually includes watching a video which describes OSA and CPAP therapy. The actual CPAP machine and mask are shown to the patient. A CPAP mask is fitted for the patient who can then try on the mask and feel the air pressure. During the PSG, the premise of CPAP titration is to gradually adjust the CPAP setting according to standard protocols until apnoeas, hypopnoeas, oxygen desaturation, snoring and arousals are eliminated. CPAP, bilevel autotitrating positive airway pressure (BiPAP) or autotitrating positive airway pressure (APAP) devices can all be used. These are discussed in more detail in the chapter on Medical & Positive Airway Pressure Therapy for OSA.

Interpretation Of The Polysomnogram
The interpretation of the PSG is based on analysis of sleep architecture,
respiratory events and abnormal limb movements. The sleep indices, physiologic and pathological events which are usually included in a PSG report are listed in table 1 (Iber et al., 2007).

Table 1. Parameters to be Reported in PSG

Scoring Guidelines for Sleep Stages
Staging of sleep and “scoring” of abnormal events like arousals, apnoeas and limb movements are usually done by a sleep technologist, and reviewed by a sleep trained physician. Criteria for scoring arousals, respiratory events and abnormal movements have been described (Rechtschaffen and Kales, 1968; Iber et al., 2007). A separate scoring system is used for neonates because sleep stages are not as well defined as in the adult.

Polysomnographic features of normal wakefulness and the different stages of sleep (based on the EEG, EOG and EMG) are summarized in table 2. On MSLT or MWT, only the EEG, EOG and EMG channels are used for the staging of sleep.

Table 2. PSG Features of Wakefulness and Stages of Sleep in Adults

Sleep Related Breathing Disorders
Sleep apnoea in adults is generally diagnosed when there are 5 or more apnoeas or hypopnoeas per sleep hour (ie AHI > 5) (Guilleminault et al., 1978). These criteria apply to adults < 60 years old only and cannot be used the elderly or in children because of age-related variations in the frequency of respiratory events. Respiratory events increase with age and are not invariably associated with long term cardiopulmonary complications. Therefore using the standard diagnostic criteria may overdiagnose sleep apnoea in those > 60 years old (Berry et al., 1984). In children, apnoeas are less frequent and less PSG data is available for establishing normal control values. Based on a series of 50 normal children, Marcus et al. suggested a normal AHI in children of 92% (Marcus et al., 1992). Rosen et al. studied 20 children with clinical features of obstructive sleep apnoea syndrome and found that though the mean AHI was only 2, there were frequent episodes of severe oxygen desaturation (Rosen et al., 1992). Therefore, in a child with suggestive symptoms like snoring, an AHI > 2 or frequent oxygen desaturation is generally considered to be clinically significant.

The severity of OSA is classified as mild (AHI 5-15), moderate (AHI >15-30) or severe (AHI > 40) based on the AHI. More severe OSA is also associated with longer duration of apnoeas or hypopnoeas, more severe oxygen desaturation or hypercapnia and the presence of associated cardiac arrhythmias. Patients with an AHI > 20 have been shown to have significantly increased mortality during long term follow up (He et al., 1988).

Sleep Related Movement Disorders & Parasomnias
EMG recordings from the leg (and sometimes the arm) are used as a measure of abnormal periodic limb movements in sleep (PLMS) and while awake. PLMS are characterized by big toe extension, ankle dorsiflexion and knee flexion. They occur in clusters, intermittently throughout the night, especially in sleep stages 1 and 2. PLMS may be excessive in restless legs syndrome (while awake), periodic limb movement disorder (during sleep) and parasomnias like REM behaviour disorder (Figs 19-20). Periodic limb movements with arousal may cause significant sleep disruption and excessive daytime sleepiness. The PLM index refers to the number of limb jerks per sleep hour, while the PLM-arousal index is the number of jerk-arousals per sleep hour.

An 11% incidence of frequent PLMS (ie. those with a PLM index > 5) has been reported in a normal population. The frequency of PLMS was found to increase markedly with age, but symptoms were present only in those with a PLM-arousal indices > 5. Therefore a PLM-arousal index > 5 is generally considered abnormal. As PLMS increase with age, the cut-off for abnormality is higher for elderly patients (PLM-arousal index >10-15). In isolation, frequent PLMS are of uncertain clinical significance and should be interpreted in each clinical context.

The Polysomnography Report
The PSG report is a summary statement of the sleep architecture, EEG abnormalities, cardiopulmonary events and abnormal movements in sleep. The final impressions and recommendations are based on both clinical features and PSG recorded data.

CONCLUSIONS
Sleep studies (including PSG, MSLT and MWT) are a helpful tool in the evaluation of sleep disorders such as OSA, narcolepsy and abnormal behaviours in sleep. PSG in a sleep laboratory attended by qualified sleep technologists is the current gold standard for the evaluation of sleep and sleep related breathing disorders. Technological advances will likely give rise to more cost effective, widely available and less labour intensive techniques to meet the increasing demand for diagnostic sleep studies as sleep disorders are increasingly recognised. Appropriate patient selection by a trained sleep specialist when such portable monitoring devices are used for the diagnosis and to guide treatment of OSA is necessary, and data recorded from such devices should be analysed and interpreted by trained sleep techologists and sleep physicians. When such studies are negative or equivocal, a full PSG remains the standard of care to definitively rule out OSA in patients with suggestive symptoms.

REFERENCES
Anders T, Emde R, Parmelee A, eds. A Manual of Standardized Terminology: Techniques and Criteria for Scoring of States of Sleep and Wakefulness in Newborn Infants. Los Angeles, Calif: UCLA Brain Information Service/Brain Research Institute; 1971.

American Academy of Sleep Medicine. The International Classification of Sleep Disorders 2nd Ed. : Diagnostic and Coding Manual. Westchester, Illinois (US): American Academy of Sleep Medicine; 2005.

Berry DT, Webb WB, Block AJ. Sleep Apnea Syndrome: A Critical Review of the Apnea Index as a Diagnostic Criterion. Chest 1984; 86 (4): 529-531.

Carskadon M, Dement W. Sleep tendency: an objective measure of sleep loss. Sleep Res 1977; 6:200.

Chesson A Jr, Hartse K, Anderson WM, et al. Practice parameters for the evaluation of chronic insomnia. An American Academy of Sleep Medicine report. Standards of Practice Committee of the American Academy of Sleep Medicine. Sleep 2000; 23(2): 237-241.

Chesson AL Jr, Berry RB, Pack A; American Academy of Sleep Medicine; American Thoracic Society; American College of Chest Physicians. Practice parameters for the use of portable monitoring devices in the investigation of suspected obstructive sleep apnea in adults. Sleep. 2003 Nov 1;26(7):907-13

Dement WC, Kleitman N. Cyclic variations in EEG during sleep and their relation to eye movements, body motility and dreaming. Electroencephalogr Clin Neurophysiol 1957; 9: 673-690.

Doghramji K, Mitler M, Sangal R, Shapiro C, al. e. A normative study of the maintenance of wakefulness test (MWT). Electroencephogr and Clin Neurophysiol 1997; 103:554-562.

Guilleminault C, van den Hoed J, Mitler M. Clinical overview of the sleep apnea syndrome. Sleep Apnea Syndromes (Guilleminault C, Dement WC, eds.). New York: Alan R. Liss, 1978: 1-12.

He J, Kryger MH, Zorick FJ, Conway W, Roth T. Mortality and apnea index in obstructive sleep apnea. Experience in 385 male patients. Chest. 1988 Jul;94(1):9-14.
Iber C, Ancoli-Israel S, Chesson AL, Quan SF. The AASM Manual for the Scoring of Sleep and Associated Events: Rules, Terminology, and Technical Specifications. Westchester, Ill: American Academy of Sleep Medicine; 2007.

Kushida CA, Littner MR, Morgenthaler T, Alessi CA, Bailey D, Coleman J Jr, Friedman L, Hirshkowitz M, Kapen S, Kramer M, Lee-Chiong T, Loube DL, Owens J, Pancer JP, Wise M. Practice parameters for the indications for polysomnography and related procedures: an update for 2005. Sleep. 2005 Apr 1;28(4):499-521.

Johns MW. A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep. 1991 Dec;14 (6):540-5.

Littner MR, Kushida C, Wise M, Davila DG, Morgenthaler T, Lee-Chiong T, Hirshkowitz M, Daniel LL, Bailey D, Berry RB, Kapen S, Kramer M; Standards of Practice Committee of the American Academy of Sleep Medicine. Practice parameters for clinical use of the multiple sleep latency test and the maintenance of wakefulness test. Sleep. 2005 Jan 1;28(1):113-21.

Marcus CL, Omlin KJ, Basinski DJ, Bailey SL, et al. Normal Polysomnographic Values for Children and Adolescents. Am Rev Respir Dis 1992; 146: 1235-1239.

Mitler M, Gujavarty K, Browman C. Maintenance of Wakefulness Test: A Polysomnographic Technique for Evaluating Treatment Efficacy in Patients with Excessive Somnolence. Electroencephogr and Clin Neurophysiol 1982; 53:658-661.

Pittman SD, Ayas NT, MacDonald MM, Malhotra A, Fogel RB, White DP. Using a wrist-worn device based on peripheral arterial tonometry to diagnose obstructive sleep apnea: in-laboratory and ambulatory validation. Sleep. 2004 Aug 1;27(5):923-33.

Rechtschaffen A, Kales A, eds. A Manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages of Human Subjects. Los Angeles: UCLA Brain Information Service, NINDS Neurological Information Network; 1968.

Rosen CL, D’Andrea L, Haddad GG. Adult Criteria for Obstructive Sleep Apnea Do Not Identify Children with Serious Obstruction. Am Rev Respir Dis 1992; 146: 1231-1234.

Zou D, Grote L, Peker Y, Lindblad U, Hedner J. Validation a portable monitoring device for sleep apnea diagnosis in a population based cohort using synchronized home polysomnography. Sleep. 2006 Mar 1;29(3):367-74.

MULTIPLE SLEEP LATENCY TEST

Introduction
Excessive daytime sleepiness (EDS) is an extremely common reason for patients to be seen in a sleep disorders clinic. The multiple sleep latency test (MSLT) provides an objective measure of sleepiness. A MSLT may be helpful in diagnosing narcolepsy if there are at least 2 sleep onset REM periods (SOREMPs).
This test is the most reliable objective test used to assess sleepiness.1 The MSLT was first described by Dr. Carskadon in 1976 as a research tool.2 In 1978, the Sleep Disorder Center at Stanford described the first use of the MSLT to differentiate patients with narcolepsy and normal controls using sleep latencies.3 In 1979, the same group reported that patients with narcolepsy had 2 or more SOREMPs, while control subjects had no SOREMPs.4
There are 2 general indications for the MSLT. The first is to document sleepiness in an objective manner. For example, it may be important in the setting of obstuctive sleep apnea syndrome or even insomnia to assess the degree of daytime somnolence. In addition, treatment effects of different therapies can be assessed.1 The second indication is to evaluate patients for a diagnosis of narcolepsy, using the sleep latency in addition to the presence of SOREMPs.
There are advantages and disadvantages of the MSLT described in the literature.5
The first advantage is that test results are not easy to falsify, either being more sleepy or more alert. Secondly, there is no practice effect, and day-to-day reliability of this test is high. Thirdly, the test can be administered to most patients, including children. There are few exceptions to this such as young children who may not be able to follow directions. Lastly, motivation to perform does not affect the results of the MSLT. There are certain disadvantages to the MSLT. First, the test takes a full day to perform and is time-consuming for both the patient and technician. Secondly, the test needs to be performed in a sleep laboratory due to the specific requirements needed. Finally, results are not valid if the patient is sick or in pain.

Technique
Patient preparation
In order to ensure valid MSLT results, there are general guidelines that all patients should follow.6 Prior to the MSLT, patients are asked to complete a 1 to 2 week sleep diary, since results can be affected by the presence of sleep restriction up to one week prior to testing.7
The MSLT should be performed on the day after an overnight polysomnogram in order to accurately document aspects of the prior night’s sleep, such as REM percentage and that the patient has had an adequate night of sleep. It is also important to document that patients do not have obstructive sleep apnea syndrome or periodic limb movements in sleep which may account for the presence of EDS. In our laboratory, if patients have an apnea-hypopnea index of greater than 20 and a sleep efficiency less than 65%, the MSLT is cancelled. One study showed that MSLT parameters were not significantly different in patients who had an overnight polysomnogram prior to the MSLT compared to those who did not have a polysomnogram.8 In our laboratory, a MSLT will be performed even though patients may not have undergone an overnight polysomnogram, provided that they report having slept for at least 80% of their usual sleep time.
Medication use is an important consideration. Medications that can affect sleep latency (i.e. sedatives, hypnotics, antihistamines, stimulants) or REM latency (i.e. tricyclic antidepressants, MAO inhibitors, amphetamines) will affect MSLT results and should be discontinued for 2 weeks prior to the MSLT.

General considerations
Patients should be dressed in comfortable clothes for the MSLT. Nap trials should occur at 2 hour intervals, starting 1.5-3 hours after awakening. Since the purpose of the MSLT is to document how quickly a patient can fall asleep in a sleep promoting environment without other stimuli, there are a number of precautions that need to be taken. The bedroom should be quiet, dark and temperature controlled. If there are noises that occur during nap trials, such as alarms or sirens, they should be annotated in the MSLT recording. Patients should be instructed that during the MSLT, they should avoid alcohol, caffeine, smoking (30 minutes before each trial) and vigorous exercise (15 minutes before each trial). Patients should be out of the bed and should be not be allowed to sleep between nap trials.
On the morning of the MSLT, obtain a medical and sleep history including a list of current medications. Note the type and doses of medications taken the morning of the study.

Montage and filter settings
The montage is based on leads of the 10-20 international electrode placement system. At a minimum, there should be 2 central EEG channels, 2 occipital channels, 2 eye channels, and an EMG channel.
of sleep latencies.

Biocalibration
Once all electrodes are attached, adjust sensitivities as needed. Then ask patients to perform the following tasks while keeping their head still. Annotate the following actions in the MSLT recording:
1. Open eyes for 60 seconds
2. Close eyes for 60 seconds to identify an alpha rhythm
3. Open eyes and look left, right, left, right, up, down, up, down
4. Blink 5 times
5. Grit teethIt is important to record with the eyes open to be familiar with the awake polysomnographic features such as eye movements, background activity and level of EMG activity. When the eyes are closed, an alpha rhythm is seen best in the occipital channels, since eye opening attenuates the alpha rhythm. When the patient looks in different directions, this allows the recording of typical eye movments for that patient. Gritting teeth causes an increase in muscle tone that can be seen in the chin EMG channel, and may also produce EMG artifact in other channels.

Recording
After biocalibration, the patient may be asked to rate their level of sleepiness. Have the patient loosen any restrictive clothing and remove their shoes before lying in bed. Annotate the trial number and body position. Each trial should be started the same way, and the patient should be given the same set of instructions. For example, “lie quietly, keep your eyes closed and try to fall asleep.” After the instructions are given, the lights should be turned off and the door should be closed.In between nap trials, the patient should be out of the bed and should not sleep. Additionally, the patient should not engage in any type of exercise. This may need careful monitoring by the sleep technician.
If the patient does not fall asleep within 20 minutes, end the trial. If the patient falls asleep within 20 minutes and remains asleep, end the trial 15 minutes after sleep onset. Sleep onset is defined as one epoch of any stage of sleep. Although not in the guidelines, if REM sleep is recorded, it may be helpful to document that the eyes are closed by viewing the video monitor.
Four to five nap trials should be recorded during the same session, with the first nap 1.5-3 hours after awakening. Typically, if there are no SOREMPs or if there are 2 or more SOREMPs in the first 4 trials, the MSLT is ended. In the study by Mitler et al., over 12% of the 40 narcoleptic patients that were studied did not have a second SOREMP until the 4th nap trial.4 However, if only a single SOREMP is recorded in the first 4 trials, a fifth nap trial should be recorded to see if a second SOREMP can be recorded. A recent study looked at 588 consecutive MSLTs and found that 152 MSLTs (26%) had one SOREMP and 103 MSLTs (18%) had 2 or more SOREMPs. In 56 MSLTs (9.5%), a fifth nap was required to record a second SOREMP.9

Scoring
Sleep scoring is performed using the criteria set forth by Rechtschaffen and Kales. 10 Sleep latency is the time from trial onset to the first epoch of any stage of sleep. If a 30 second epoch is used, sleep onset occurs when there is 15 seconds or more of sleep. REM sleep latency is the time from sleep onset to the first epoch of REM sleep.The report should include the following:
1. Results of prior night’s polysomnogram, if available
2. Sleep latency values in all nap trials
3. Mean (or median) sleep latency for all nap trials
4. Number of SOREMPs
5. REM sleep latency values

Interpretation
The MSLT is considered to be the gold standard in the objective evaluation of EDS. A mean sleep latency (MSL) less than 5 minutes is considered to be pathological and correlates with severe sleepiness. This is based from the initial report which found that 27 patients with narcolepsy had a MSL of 1-3 minutes, while 14 control subjects had a MSL of 7-14 minutes. 3 A MSL of 5-10 minutes is thought to correlate with moderate sleepiness and is considered to be a diagnostic “gray area.”1, 11 In one study, 46 patients with narcolepsy had a MSL of 3.3 minutes, 17 patients with idiopathic CNS hypersomnia had a MSL of 6.5 mintues, and 16 patients with psychological disturbances had a MSL of 10.6 minutes.11
There is high interrater, intrarater and test-retest reliability when using the MSLT to evaluate EDS. The MSLT has been shown to be reliable over a test-retest interval of a year regardless of the level of sleepiness.12 One study looked at 192 MSLTs to determine interrater and intrarater reliability.13 The interrater reliability coefficient for MSL was 0.90 and the intrarater reliability was 0.87. The interrater reliability coefficient for the number of SOREMPs was 0.88 and the intrarater reliability was 0.81. The kappa for interrater agreement for the presence of more than one SOREMP was 0.91 compared to a kappa score of 0.78 for intrarater agreement.

The MSLT is helpful in diagnosing narcolepsy. Two or more SOREMPs in association with a MSL less than 5 minutes is considered to be highly suggestive for the diagnosis of narcolepsy. This was first reported in a group of 40 patients with narcolepsy and 14 control subjects.4 Another study found that in 144 patients with EDS, 2 or more SOREMPs were found in 52 out of 61 patients with a clinical diagnosis of narcolepsy and in only 1 patient with a diagnosis of severe sleep apnea resulting in a sensitivity of 84% and a specificity of 99%.14 Additionally, when a history of cataplexy was present in the same group of patients, narcoleptics with cataplexy tended to have a shorter sleep latency and more SOREMPs than narcoleptics without cataplexy.

The ability of the MSLT to help diagnose narcolepsy is probably best in patients in whom the suspicion of narcolepsy is high based on clinical criteria.15 Using the MSLT to assess EDS independent of other clinical findings may overdiagnose or even misdiagnose narcolepsy. Clinical correlation is essential in patients who may have only one SOREMP during a MSLT.In our laboratory, the following guidelines are used when interpreting a MSLT:

MSLT intrepretation guidelines

• MSL10 minutes normal
• 1 SOREMP unclear significance without clinical information
• 2 or more SOREMPs highly suggestive of narcolepsy given the clinical situation

In patients with OSAS, MSLs are usually less than 10 minutes.1 SOREMPs can be seen during MSLTs in patients with OSAS. OSAS and narcolepsy have been found to coexist in up to 6% of patients with excessive daytime sleepiness. One study showed that in 1145 patients diagnosed with OSAS, 4.7% of the patients had 2 or more SOREMPs and 9.7% had one SOREMP on a MSLT the day after a polysomnogram. When a sleep disorder such as OSAS is present in a patient with suspected narcolepsy, the underlying sleep disorder needs to be adequately treated before using the MSLT as a helpful tool in diagnosing narcolepsy. Upper airway resistance syndrome(UARS) has been associated with sleepiness. When MSLs of patients with UARS are compared to patients with OSAS, they are similar. In a study that compared 12 control subjects to 12 patients with UARS and OSAS, the MSLs were 17 minutes, 8 minutes and 8 minutes respectively.

Patients without EDS can have SOREMPs as well. A study with 139 healthy controls found that 24 (17%) of the controls had 2 or more SOREMPs, 8 (6%) had 1 SOREMP, and 107 (77%) had no SOREMPs.18 The individuals with SOREMPs did not have evidence of sleep disordered breathing or periodic leg movements during an overnight polysomnogram prior to the MSLT.

The MSLT can be helpful in evaluating excessive sleepiness of unknown cause. Idiopathic hypersomnia is associated with a short sleep latency and 80% of patients do not have a SOREMP. The MSLT can provide useful information when evaluating patients with hypersomnolence who complain of insomnia but actually overestimate their sleep latency. In a group of 147 patients, the mean objective sleep latency was 8.9 minutes compared to the mean subjective sleep latency of 11.9 minutes.

One study found that in 1124 patients, periodic limb movements during sleep (PLMS), with or without arousals, were not associated with EDS. Patients with 5 or more leg movements per hour of sleep were included in this study. Two hundred and seventy of the 1124 patients with an apnea-hypopnea index (AHI) of greater than 45 (mean AHI=79) had a MSL of 5 minutes, whereas 854 of the 1124 patients with an AHI of less than 45 (mean AHI=16) had a MSL of 8 minutes. Interestingly, a higher PLM-arousal index was associated with less sleepiness.

MSLT findings need to be interpreted with other factors in consideration such as the amount of sleep the night before and during the week before the test, possible drug effects, circadian phase and other sleep disorders that can cause sleep fragmentation. One important factor is the amount of physical activity undertaken by the patient between naps. It has been shown that MSLs can be reduced by 5.8 minutes when patients relaxed in bed watching television compared to walking prior to naps.22 This can affect the interpretation of the MSLT significantly. Increased MSLs can be seen in patients who are in a significant amount of pain or those who are overly anxious. The MSLT is performed in a setting that is conducive to sleep, and may not be representative of conditions of situations during which patients are likely to fall asleep. Patients need to cooperate for accurate results. The MSLT can be difficult to administer to cognitively impaired patients or young children.

A repeat MSLT may be indicated when the initial test is thought not to be representative of a patient’s status, when results are ambiguous, to document treatment effects and when more than one sleep disorder is suspected. An important consideration when interpreting the significance of a single SOREMP is whether it is the result of REM rebound that can be seen in the setting of REM suppressant withdrawal or alcohol withdrawal.

MAINTENANCE OF WAKEFULNESS TEST

Introduction
The Maintenance of Wakefulness Test (MWT) is a variation of the MSLT. The MWT was originally described by Mitler et al.23 to measure treatment efficacy in patients with excessive sleepiness. The MWT provides an objective measure of the ability of an individual to stay awake. Most sleep laboratories only use the MSLT routinely to document sleepiness in an objective manner and to help with diagnosis. There is extensive literature showing good correlation with MSLs and the degree of somnolence. However, the MWT is not widely used as an objective measure of treatment efficacy because in most cases, the clinical response is used more.
A limitation of the MWT is the variation in nap trial length (20 vs. 40 minutes) and definition of sleep onset (3 consecutive epochs of stage 1 sleep or one epoch of any other stage of sleep vs. the first epoch of any stage of sleep) . Specific recommendations.23, 24 and normative data 24, 25 have been published. Doghramji et al. recommend using the 20 minute protocol (with sleep onset defined as the first epoch of sleep) for routine clinical use.24

Indications for the MWT include the evaluation of the ability to stay awake in the setting of disorders associated with excessive somnolence, such as narcolepsy and OSAS. More importantly, the MWT can be used to evaluate the efficiacy of various treatments, such as continuous positive airway pressure (CPAP) or medications. The Federal Aviation Administration uses the MWT to determine wheter noncommercial pilots can be licensed after treatment for OSAS.26 Depending on the clinical situation or question, the ability to stay awake may be more important than the ability to fall asleep.

Technique
General guidelines have been described23, which are similar to those for the MSLT. Patient preparation, general considerations and technical calibrations are similar to those used for the MSLT. Seven channels as described in table 1 are sufficient. The main difference between the MSLT and the MWT are the instructions given to the patient. During a MWT, patients are asked to stay awake, instead trying to fall asleep.

Recording
The room should be dark with one light source positioned behind the patient’s head, out of the field of vision. The light should deliver an illuminance of 0.10-0.13 lux at the corneal level. The room temperature should be as close to 22 C as possible. The trial number, body position and room temperature at the beginning of each trial should be annotated. After biocalibrations, have the patient sit in bed, with the back and head supported by a bedrest so that the neck is not in an awkward position. Each trial should be started the same way, and the patient should be given the same set of instructions. For example, “Sit still and try to stay awake for as long as possible. Look ahead of you and do not look directly at the light.” Patients should be discouraged from using extraordinary measures to stay awake such as singing, slapping the face or exercising in bed. In between nap trials, the patient should be out of the bed and should not sleep. This may need careful monitoring by the sleep technologist.

If the patient does not fall asleep within 20 minutes, end the trial. If the patient falls asleep and stays asleep, end the trial 10 minutes after sleep onset. Sleep onset is defined as one epoch of any stage of sleep. If the patient falls asleep but arouses, end the trial. Four nap trials should be recorded during the same session, with the first nap starting at 10 AM. If labor and cost of the MWT is an issue, only 2 nap trials can be recorded with minimal loss of variance.

Scoring
Sleep scoring is performed using the criteria set forth by Rechtschaffen and Kales. Sleep latency is the time from trial onset to the first epoch of any stage of sleep. Using a 30 second epoch, sleep onset occurs when there is 15 seconds or more of sleep.
The report should include the following:
1. Sleep latency
2. Total sleep time
3. Total wake time
4. Sleep stages reached in each trial

Interpretation
The MWT is used to provide an objective measure of the ability to maintain wakefulness. During nap trials of 20 minutes in which sleep onset is defined as the first epoch of sleep, a normal MSL is 18 minutes. A MSL of less than 11 minutes is considered to represent an impairment in wake tendency. This is based on a study with 64 patients in whom it was determined that the lower end of normal would be 2 standard deviations below the mean. Eight percent of the 64 patients had MSLs less than 11 minutes.

During nap trials of 40 minutes in which sleep onset is defined as the first epoch of sleep, a normal MSL is 35 minutes. A MSL of less than 19.4 minutes is considered to be abnormal. 24 In the setting of OSAS, a study of 322 patients that had 40 minute nap trials found that MSLs correlate best with the respiratory arousal index and mean oxygen saturation. When the same group was evaluated for survival after 7.5 years, there was follow-up data for 142 patients. Of the 142 patients, 120 were alive and 22 were dead. Those who died had a shorter MSL (21 minutes) compared to those who lived (28 minutes).

The significance of SOREMPs during the 10 minutes of sleep during a nap trial is unclear. In a study of 12 patients with sleep apnea, 12 patients with narcolepsy and 10 control subjects, it was found that patients with narcolepsy had the most SOREMPs. Eight of 12 narcoleptics had SOREMPs, 6 of 12 patients with sleep apnea had SOREMPs and none of the controls had SOREMPs.

The MWT is different from the MSLT in that some patients who have short sleep latencies on the MSLT are able to stay awake on the MWT, while some patients who have long sleep latencies on the MSLT are not able to stay awake on the MWT. A study of 47 patients with various sleep disorders showed that the MWT was better than the MSLT at assessing treatment efficacy. In another study, 30 patients with OSAS who were treated with CPAP but still complained of sleepiness were given modafinil, and found that sleep latencies significantly improved when measured by the MWT but not the MSLT.

Since the ability to stay awake is an important factor for meaningful employment, the MWT may better assess the impact of sleep disorders on the ability to drive or work. The MWT is limited since testing conditions do not simulate real life conditions in which an individual is required to stay awake, such as driving a truck or sitting behind a desk. Nonetheless, the MWT remains a good tool to demonstrate that patients can stay awake and that a particular treatment is effective.

REFERENCES
1. American Sleep Disorders Association. The Clinical Use of the Multiple Sleep Latency Test. Sleep 1992; 15:268-276.
2. Carskadon M, Dement W. Sleep tendency: an objective measure of sleep loss. Sleep Res 1977; 6:200.
3. Richardson G, Carskadon M, Flagg W, van den Hoed J, Dement W, Mitler M. Excessive daytime sleepiness in man: multiple sleep latency measurements in narcoleptic and control subjects. Electroencephogr and Clin Neurophysiol 1978; 45:621-627.
4. Mitler M, van den Hoed J, Carskadon M, et al. REM sleep periods during the multiple sleep latency test in narcoleptic patients. Electroencephogr and Clin Neurophysiol 1979; 46:479-481.
5. Carskadon M. Evaluation of excessive daytime sleepiness. Neurophysiol Clin 1993; 23:91-100.
6. Carskadon M, Dement W, Mitler M, Roth T, Westbrook P, Keenan S. Guidelines for the Multiple Sleep Latency Test (MSLT): A Standard Measure of Sleepiness. Sleep 1986; 9:519-524.
7. Carskadon M, Dement W. Cumulative effects of sleep restriction on daytime sleepiness. Psychophysiology 1981; 18:107-13.
8. Wichniak A, Geisler P, Tracik F, Cronlein T, Morrissey S, Zulley J. The influence of polysomnography on the Multiple Sleep Latency Test and other measures of daytime sleepiness. Physiol Behav 2002; 75:183-8.
9. Golish J, Sarodia B, Blanchard A, Dinner D, Foldvary N, Perry M. Prediction of the final MSLT result from the results of the first three naps. Sleep Medicine 2002; 3:249-253.
10. Rechtschaffen A, Kales A. A Manual of standardized terminiology, techniques and scoring system for sleep stages of human subjects. Los Angeles, UCLA: BIS/BRI, 1968.
11. van den Hoed J, Kraemer H, Guilleminault C, et al. Disorders of Excessive Daytime Somnolence: Polygraphic and Clinical Data for 100 Patients. Sleep 1981; 4:23-37.
12. Zwyghuizen-Doorenbos A, Roehrs T, Schaefer M, Roth T. Test-Retest Reliability of the MSLT. Sleep 1988; 11:562-565.
13. Drake C, Rice M, Roehrs T, Rosenthal L, Guido P, Roth T. Scoring Reliability of te Multiple Sleep Latency Test in a Clinical Population. Sleep 2000; 23:911-913.
14. Amira S, Johnson T, Logowitz N. Diagnosis of Narcolepsy Using the Multiple Sleep Latency Test: Analysis of Current Laboratory Criteria. Sleep 1985; 8:325-331.
15. Aldrich M, Chervin R, Malow B. Value of the Multiple Sleep Latency Test (MSLT) for the Diagnosis of Narcolepsy. Sleep 1997; 20:620-629.
16. Chervin R, Aldrich M. Sleep Onset REM Periods during Multiple Sleep Latency Tests in Patients Evaluated for Sleep Apnea. Am J Respir Crit Care Med 2000; 161.
17. Guilleminault C, Kim Y, Chowdhuri S, Horita M, Ohayon M, Kushida C. Sleep and daytime sleepiness in upper airway resistance syndrome compared to obstructive sleep apnoea syndrome. Eur Respir J 2001; 17:838-847.
18. Bishop C, Rosenthal L, Helmus T, Roehrs T, Roth T. The Frequency of Multiple Sleep Onset REM Periods Among Subjects With No Excessive Daytime Sleepiness. Sleep 1996; 19:727-730.
19. Baker T, Guilleminault C, Nino-Mucia G, Dement W. Comparative polysomnographic study of narcolepsy and idiopathic central nervous system hypersomnia. Sleep 1986; 9:232-42.
20. Chervin R, Guilleminault C. Overestimation of Sleep Latency by Patients With Suspected Hypersomnolence. Sleep 1996; 19:94-100.
21. Chervin R. Periodic Leg Movements and Sleepiness in Patients Evaluated for Sleep-disordered Breathing. Am J Respir Crit Care Med 2001; 164:1454-1458.
22. Bonnet M, Arand D. Sleepiness as Measured by Modified Multiple Sleep Latency Testing Varies as a Function of Preceding Activity. Sleep 1998; 21:477-483.
23. Mitler M, Gujavarty K, Browman C. Maintenance of Wakefulness Test: A Polysomnographic Technique for Evaluating Treatment Efficacy in Patients with Excessive Somnolence. Electroencephogr and Clin Neurophysiol 1982; 53:658-661.
24. Doghramji K, Mitler M, Sangal R, Shapiro C, al. e. A normative study of the maintenance of wakefulness test (MWT). Electroencephogr and Clin Neurophysiol 1997; 103:554-562.
25. Mitler M, Doghramji K, Shapiro C. The Maintenance of Wakefulness Test: Normative data by age. Journal of Psychosomatic Research 2000; 49:363-365.
26. Mitler M, Carskadon M, Hirshkowitz M. Evaluating Sleepiness. In: Dement W, ed. Principles and Practice of Sleep Medicine, 3rd edition. Philadelphia: W.B. Saunders Company, 2000:1251-1257.
27. Poceta J, Timms R, Jeong D, Ho S, Erman M, Mitler M. Maintenance of Wakefulness Test in Obstructive Sleep Apnea Syndrome. Chest 1992; 101:893-97.
28. Poceta J, Loube D, Kellgren E, Bizik K, Mitler M. Mortality in Obstructive Sleep Apnea: Association with Impaired Wakefulness. Sleep and Breathing 1999; 3:3-8.
29. Browman C, Gujavarty K, Sampson M, Mitler M. REM Sleep Episodes During the Maintenance of Wakefulness Test in Patients with Sleep Apnea Syndrome and Patients with Narcolepsy. Sleep 1983; 6:23-28.
30. Sangal R, Thomas L, Mitler M. Maintenance of Wakefulness Test and Multiple Sleep Latency Test. Chest 1992; 101:898-902.
31. Sangal R, Thomas L, Mitler M. Disorders of Excessive Sleepiness: Treatment Improves Ability to Stay Awake But Does Not Reduce Sleepiness. Chest 1992; 102:699-703.
32. Kingshott R, Vennelle M, Coleman E, Engelman H, Mackay T, Douglas N. Randomized, Double-blind, Placebo-controlled Crossover Trial of Modafinil in the Treatment of Residual Excessive Daytime Sleepiness in the Sleep Apnea/Hypopnea Syndrome. Am J Respir Crit Care Med 2001; 163:918-923.

Introduction
Abnormal behaviours or movements which disrupt sleep are less common presentations to sleep disorders clinics than insomnia or excessive daytime sleepiness. In the International Classification of Sleep Disorders, these disorders are classified as Parasomnias and Sleep Related Movement Disorders (AASM, 2005). These conditions can lead to sleep disturbance, physical injury and psychosocial problems which can affect the patient or the bed partner.

• Parasomnias are undesirable sleep related behaviours, movements or experiences which can cause sleep disruption and other adverse effects on health. They are classified as Disorders of Arousal (from NREM Sleep) (eg. confusional arousals, sleepwalking and sleep terrors), Parasomnias usually associated with REM Sleep (eg. REM Sleep Behaviour Disorder, Nightmare Disorder), and Other Parasomnias (eg. Sleep Enuresis, Sleep Related Eating Disorder). Most of the parasomnias occur more commonly in childhood and are discussed in greater detail in the Paediatric section in the chapter on Parasomnias. The more common parasomnias referred to adult sleep clinics are sleepwalking and REM Sleep Behaviour Disorder (RBD).
• Sleep Related Movement Disorders are an important subset of abnormal behaviours which are classified separately. In contrast to the parasomnias which often comprise complex and seemingly purposeful behaviours, they manifest as relatively simpler stereotyped movements. The main sleep related movement disorders seen in adults are Restless Legs Syndrome (RLS), Periodic Limb Movement Disorder (PLMD) and Sleep Related Bruxism.

Parasomnias and sleep related movements disorders must be distinguished from epileptic seizures which also present with stereotypic or complex behaviours. In general the synchronised state of NREM sleep facilitates epileptic activity, with some epileptic syndromes having a marked tendency to manifest predominantly during sleep e.g. nocturnal frontal lobe epilepsy. Epileptic phenomena are characterised by repetitive stereotypic behaviour, and can be difficult to distinguish clinically from non-epileptic phenomena such as the parasomnias. This category of sleep disturbance usually requires PSG with full EEG (for documentation of epileptiform discharges) and simultaneous audiovisual recording in addition to detailed history for a definitive diagnosis.

The NREM parasomnias or disorders of arousal are discussed in the next section. The other commonly encountered adult parasomnias, seizures, and sleep related movement disorders (ie RBD, sleep related epilepsy, RLS and PLMD).

Disorders of Arousal (from NREM Sleep)
Disorders of arousal (DA) from NREM sleep include confusional arousals, sleepwalking and sleep terrors. These usually occur in NREM slow wave (deep) sleep, but can also occur in stage N2 NREM (light) sleep.

Epidemiology & Risk Factors
DA are more common in children than in adults. Prevalence estimates in an adult general population are – confusional arousals: 2.9-4.2%, sleepwalking: 2-4% and sleep terrors: 2.2% (Mahowald et al., 2005; Ohayon et al.,1999). In adults, sleepwalking associated with violence is more commonly reported in men. Sleepwalking peaks by age 8-12 years. Sleep terrors peak at about 5-7 years and usually resolve by adolescence. DA can be triggered by sleep deprivation, forced awakenings, physical or emotional stress, anxiety, fever, psychotropic drugs (eg. stimulants, neuroleptics, sedative-hypnotics), antihistamines, alcohol, environmental stimuli (eg. noise or light), and primary sleep disorders such as obstructive sleep apnoea (OSA). In adults, associated psychiatric problems (depression, anxiety, or bipolar disorders) have been reported, though significant psychopathology is usually not present (Ohayon et al., 1999).

Pathophysiology
DA tend to run in families and are believed to be a result of faulty transitions between deep and lighter stages of sleep, with sleep-wake state dissociation reflected in the EEG activity during episodes which comprise an admixture of slower sleep and faster wake-like frequencies. Other factors contributing to the complex behaviours include the inherent instability of slow wave sleep in patients with DA. Structural lesions in the brain’s normal wake centre (eg. posterior hypothalamus, reticular activating system) have been reported to cause DA, but the majority of patients have normal brains.

Of all the factors contributing to DA, genetic predisposition is probably the most important. A positive family history in a first-degree relative is found in 60% of children with DA. One study reported the prevalence of sleepwalking and sleep terrors being 10 times higher in first degree relatives of sleep terror patients than the general population (Kales et al., 1980). The rate of sleepwalking in children with a family history increases to 45% with one parent, and 60% with both parents affected.

Clinical Features & Differential Diagnoses
Patients with DA act out complex behaviours while in deep sleep, and typically remain unaware of their actions, and amnestic of the event. Episodes tend to occur in the first third of the night when slow wave sleep predominates. Forced arousals from sleep can also induce episodes. Confusional arousals occur most frequently in infants and toddlers, but are also seen in young adults (age 15-24 years), decreasing with age. A typical episode is characterised by disorientation, slowed mentation, agitation, crying, thrashing and combative behaviour, usually lasting 5-15 minutes (up to as long as 30 to 40 minutes), before returning to sleep.

Sleepwalking consists of walking around in a state of altered consciousness, either calm or agitated, after a partial arousal from deep sleep. It can vary in duration and complexity. Safety is a main concern as injury may occur during wandering, falls and environmental exposure.

In contrast to sleepwalking in which the subject usually remains calm, sleep terrors present dramatically, usually beginning with a piercing cry or blood-curdling scream, with associated autonomic arousal (tachypnoea, mydriasis, tachycardia and diaphoresis), behavioural manifestations of intense fear and prominent motor activity (eg. running, hitting). The subject is typically inconsolable and difficult to arouse, and may later recall feeling threatened and scared.

Conditions which can trigger DA include sleep disorders which cause recurrent arousals such as OSA and PLMD. Physical examination is usually normal. The diagnosis of these parasomnias can usually be made from the witnessed (usually a parent or bed partner) description of events. Movements and behaviours can be recorded in sleep diaries and on home videos. Conditions which mimic DA include nocturnal seizures (eg. frontal lobe epilepsy, complex partial seizures), sleep related movement disorders, panic attacks and nightmares.

Investigations
Polysomnography (PSG) is not routinely required for diagnosis if the history is suggestive, but is useful in excluding associated primary sleep disorders like OSA and PLMD. If epilepsy is suspected, an expanded EEG montage is required, ideally with time-synchronised video-EEG recording. Frequent arousals from slow wave sleep are a classic PSG finding in DA. EEG may show a mixture of faster and slower (ie. alpha, theta and delta) waves, reflecting the sleep-wake state dissociation characteristic of DA.

Management
DA are common (in children) and generally benign. Often no specific therapy is required, but sensible safety precautions should be instituted, such as padding the bedroom environment, securing doors and windows and installing alarms or other monitoring systems. Good sleep hygiene, avoiding sleep deprivation, discontinuing stimulants such as caffeine and triggering medications should be discussed. If the episodes are predictably recurrent, scheduled awakenings just before the typical time of a sleepwalking episode has been reported to successfully eliminate sleepwalking ( Frank et al., 1997). Relaxation therapy may also be useful.

Parasomnias that pose a risk of injury to the patient or bed partner, and those which are triggered by treatable conditions (eg. OSA, PLMD) will require specific treatment of the specific underlying disorder. For frequent or potentially injurious arousal parasomnias, benzodiazepines and tricyclic antidepressants may be helpful. Clonazepam has been used successfully, starting at low doses (eg. 0.25 mg at bedtime) and titrating according to effect and tolerability.

Clinical Course

The arousal parasomnias are generally benign and self-limited, and generally do not persist into late adolescence or adulthood. Confusional arousals decrease after the age of 5 years, but may progress to sleep walking in adolescence. The adult variant may persist and is associated with sleep related injury and impaired performance. Childhood sleep walking and sleep terrors typically resolve around the time of puberty.

Summary
Parasomnias, sleep related movement disorders and sleep related epilepsy can all present with sleep disturbance, physical injury and abnormal behaviours in sleep. While clinical features can be striking and suggestive enough for a diagnosis to be made on history alone, PSG with full EEG recording may be needed to make a definitive diagnosis of nocturnal sleep related epilepsy.

The classic arousal (NREM) parasomnias form part of a spectrum, with features in common including abnormal transition from slow wave sleep, complex automatic behaviours and amnesia following the episodes. As with all abnormal behaviours in sleep, management should focus on accurate diagnosis, exclusion of treatable associated conditions and simple behavioural interventions to reduce the risk of physical injury and management of associated psychosocial problems.

References
American Academy of Sleep Medicine. The International Classification of Sleep Disorders: Diagnostic and Coding Manual. 2nd ed. Westchester, Illinois (US): American Academy of Sleep Medicine; 2005.

Frank NC, Spirito A, Stark L, Owens-Stively J. The use of scheduled awakenings to eliminate childhood sleepwalking. J Pediatr Psychol. 1997; 22(3): 345-353.

Kales A, Soldatos CR, Bixler EO et al. Hereditary factors in sleepwalking and night terrors. Br J Psychiatry. 1980; 137: 111-118.

Mahowald M, Schenck C, Basetti C, et al. Parasomnias. Sateia M, editor. In: The International Classification of Sleep Disorders. 2nd ed. Westchester, IL: American Academy of Sleep Medicine, 2005. p. 137-147.

Ohayon MM, Guilleminault C, Priest RG. Night terrors, sleepwalking, and confusional arousals in the general population: their frequency and relationship to other sleep and mental disorders. J Clin Psychiatry. 1999; 60(4): 268-276.

Introduction
Rapid eye movement (REM) sleep behaviour disorder (RBD) is the most common REM sleep parasomnia. RBD is characterized by the absence of muscle atonia, which permits acting out of dreams, often resulting in physical injury. RBD was first described as a distinct clinical entity in a series of adults manifesting violent behaviors while acting out dreams during sleep and injuring themselves or their spouses in the process (1). RBD is characterized by a loss of normal REM atonia, loss of chin muscle atonia, and excessive muscle twitching on polysomnography (PSG). RBD is often associated with neurodegenerative disease and is responsive to clonazepam.

Epidemiology and risk factors
The estimated overall prevalence of RBD is about 0.5%, with a reported range from 0.38% to as high as 0.8%. The majority (approximately 90%) of RBD patients are older men, although any age group or gender can be affected.

RBD has been associated with a range of neurologic conditions, most notably parkinsonism and degenerative dementia, often reflecting an underlying synucleinopathy (2). RBD may precede the onset of parkinsonism or dementia in patients with Parkinson’s disease (PD), multiple system atrophy (MSA), or dementia with Lewy bodies (DLB) by years or decades. Thus “idiopathic” RBD may represent the initial manifestation of an evolving neurodegenerative disorder. A higher incidence of RBD is also seen in narcolepsy. This probably reflects the REM sleep-related dyscontrol common to both conditions in which the elements normally regulating REM sleep are not present.
Many commonly used drugs (e.g., selective serotonin reuptake inhibitors (SSRIs), monoamine oxidase inhibitors, and tricyclic antidepressants) can induce or aggravate RBD symptoms in patients at risk.

Pathophysiology
The pathophysiology of RBD is believed to be analogous to that described in animal models in which damage to pontine tegmental pathways mediating REM-atonia (and those structures that normally suppress the phasic locomotor drive in REM sleep) result in complex behaviors as seen in RBD. Positron emission tomography (PET) and single photon emission computed tomography (SPECT) studies have shown dysfunction in the nigrostriatal dopaminergic pathways in patients with idiopathic RBD. In humans, RBD may precede the onset of the motor symptoms of parkinsonism. This suggests that the first clinical symptoms to appear may correspond to the brainstem regions first affected by neuronal degeneration (RBD when beginning in the mesopontine junction or parkinsonism when beginning in the midbrain).

Clinical features and differential diagnoses
RBD typically manifests as complex dream-enacting behaviors that are often vivid, unpleasant, or violent, such as being attacked or chased. Typically, the abnormal movements include vocalizations, flailing, punching, kicking, swearing, gesturing, leaping, and running. These movements lead to sleep disruption and sometimes injuries. The episodes typically occur approximately 90 minutes after sleep onset, coinciding with the timing of the first REM cycle, and may recur with subsequent REM sleep cycles.
RBD is usually a chronic and progressive disorder that is either idiopathic or associated with a range of neurological disorders – including the synucleinopathies such as PD, MSA, and DLB. In PD, RBD is associated with more non-motor symptoms, such as increased depressive symptoms, sleep disturbances, and fatigue (3). RBD can also occur with cerebrovascular disease, neoplasm (such as brainstem and cerebellopontine angle tumors), and inflammatory conditions (such as Guillain–Barré syndrome and multiple sclerosis). RBD has also been noted in patients with narcolepsy, mitochondrial disorders, Tourette syndrome, autism, normal pressure hydrocephalus, and spinocerebellar ataxia. An acute form of RBD can occur during REM sleep rebound states, such as withdrawal from alcohol or sedative-hypnotic agents, and may be triggered by medications, including serotonergic antidepressants.
The diagnosis of RBD is based on a clinical history of injurious or potentially injurious dream-enacting behavior that disrupts REM sleep. PSG shows REM sleep without atonia, excessive tonic and phasic electromyogram (EMG) activity recorded from the chin, excessive phasic EMG activity in the limbs, and abnormal REM sleep behaviors. This occurs in the absence of EEG epileptiform activity during REM sleep.
Differential diagnoses encompass other causes of abnormal behavior in sleep, such as NREM parasomnias (sleepwalking, sleep terrors), nocturnal epilepsy, and obstructive sleep apnea (OSA), which may mimic RBD.

Investigations
PSG with time-synchronized video-EEG recording is needed to document the typical PSG features of RBD and to exclude disorders that may mimic RBD. Periodic limb movements of sleep (PLMS) can be seen in about 75% of RBD patients during NREM sleep.

Management
Clonazepam (beginning dose of 0.5 mg at night, increasing to 1 or 2 mg) is very effective in treating RBD and is considered first-line treatment. It is generally well tolerated and produces rapid (within the first week) and sustained (up to several years) improvement in the majority of patients, with little evidence of tolerance or abuse (4). Beneficial effects may be related to suppression of motor manifestations and partly to clonazepam’s serotonergic properties.

Alternatively, melatonin (dosing between 3 and 12 mg at night) works as monotherapy for patients who may not tolerate long-acting benzodiazepines, or who have OSA. The mechanism of melatonin is not well understood, but has been reported to be effective for RBD, especially in patients with low melatonin levels (5,6). Studies have reported that melatonin reduces motor activity during sleep and partially restores REM sleep muscle atonia. Postulated mechanisms of melatonin include restoration of RBD-related desynchronization of the circadian rhythm and mechanisms producing REM sleep muscle atonia.
The effectiveness of other drugs including dopaminergic agents (pramipexole or levodopa), acetylcholinesterase inhibitors (donepezil or rivastigmine), and antiepileptic agents (carbamazepine or gabapentin) remains unclear. Dopamine receptor agonist therapy using pramipexole, while improving symptoms of parkinsonism in PD, has not been shown in a small, prospective, uncontrolled study to reduce RBD symptoms, suggesting that dopamine mechanisms do not play a central role in the pathogenesis of RBD (7). A recently published placebo-controlled, cross-over pilot study using rivastigmine in 25 consecutive patients suggested that in patients with mild cognitive impairment and RBD resistant to conventional therapies (including benzodiazepines or melatonin) treatment with rivastigmine may induce a reduction in the frequency of RBD episodes compared to placebo (8).
Improving sleeping environment safety, removing potentially dangerous objects, and allotting separate sleeping arrangements for bed partners are also useful measures.

Clinical course
RBD is slowly progressive and is rarely associated with spontaneous remissions, though symptoms may subside in the advanced stages of an underlying neurodegenerative condition. Drug-induced RBD should improve upon withdrawal of the offending medication.

Summary
RBD, the most common REM parasomnia, is a striking clinical entity associated with neurodegeneration, affecting primarily older men. Evaluation should include a comprehensive clinical history detailing sleep behaviors and medication use, neurologic examination, and PSG. Prompt recognition is important to reduce potential complications including physical injury and marital discord arising from trauma and sleep disruption. Treatment with clonazepam is safe and effective in the majority of cases.

Further reading
1.Schenck CH, Bundlie SR, Ettinger MG, Mahowald MW. Chronic behavioral disorders of human REM sleep: a new category of parasomnia. Sleep 1986; 9(2): 293–308.

2.Boeve BF, Silber MH, Parisi JE, et al. Synucleinopathy pathology and REM sleep behavior disorder plus dementia or parkin-sonism. Neurology 2003; 61(1): 40–5.

3.Neikrug AB, Avanzino JA, Liu L, Maglione JE, Natarajan L, Corey-Bloom J, Palmer BW, Loredo JS, Ancoli-Israel S. Parkinson’s disease and REM sleep behavior disorder result in increased non-motor symptoms. Sleep Med. 2014 May 10. pii: S1389-9457(14)00170-1. doi: 10.1016/j.sleep.2014.04.009.

4.Schenck CH, Mahowald MW. Polysomnographic, neurologic, psychiatric, and clinical outcome report on 70 consecutive cases with REM sleep behavior disorder (RBD): sustained clonazepam efficacy in 89.5% of 57 treated patients. Cleve Clin J Med 1990; 57(Suppl): S9–23.

5.Boeve BF, Silber MH, Ferman TJ. Melatonin for treatment of REM sleep behavior disorder in neurologic disorders: results in 14 patients. Sleep Med 2003; 4(4): 281–4.

6.Gagnon JF, Postuma RB, Montplaisir J. Update on the pharmacology of REM sleep behavior disorder. Neurology 2006; 67(5): 742–7.

7.Kumru H; Iranzo A; Carrasco E; Valldeoriola F; Martí MJ; Santamaria J; Tolosa E. Lack of effects of pramipexole on REM sleep behavior disorder in Parkinson disease. SLEEP 2008;31(10):1418-1421.

8.Brunetti V, Losurdo A, Testani E, Lapenta L, Mariotti P, Marra C, Rossini PM, Della Marca G. Rivastigmine for refractory REM behavior disorder in mild cognitive impairment. Curr Alzheimer Res. 2014 Mar;11(3):267-73.

INTRODUCTION
Symptoms of Restless Legs Syndrome (RLS) were initially described in 1685 by Sir Thomas Willis, but the term “Restless Legs Syndrome” was only introduced later in 1945 by Karl Ekbom, a Swedish neurologist. The reported prevalence of RLS ranges widely from less than 1% to over 20%, increasing with age and about twice as common in women (Phillips et al., 2000; Berger et al., 2004). There is an ethnic difference, RLS being much less frequently reported in Asians (Tan et al., 2001). In one questionnaire-based survey of a primary care population, the reported prevalence of RLS was as high as 24%, with 15% of these patients experiencing weekly symptoms (Nichols et al., 2003). Although RLS causes distressing symptoms for which effective treatment is available, these are commonly under-reported by patients and overlooked by physicians (Walters et al., 1996). In a large survey of more than 23, 000 patients in primary care settings in 5 countries, 10% were found to have RLS (Hening et al., 2004). Although nearly 65% of the chronic sufferers had seen a doctor about their symptoms, fewer than 13% of patients were correctly diagnosed. It is likely that the prevalence of RLS is underestimated and the scope of the problem larger than is generally recognised.

DIAGNOSIS OF RLS: ESSENTIAL CRITERIA
RLS is a clinical diagnosis made based on fulfilling 4 essential criteria originally proposed in 1995 by the International RLS Study Group (Walters, 1995). These criteria were subsequently modified with additional criteria also developed for the cognitively impaired elderly and children (Allen et al., 2003). The 4 Essential Criteria for RLS are:

1. An urge to move the legs, usually accompanied or caused by uncomfortable and unpleasant sensations in the legs.

2. The urge to move or unpleasant sensations begin or worsen during periods of rest or inactivity such as lying or sitting.

3. The urge to move or unpleasant sensations are partially or totally relieved by movement, such as walking or stretching, at least as long as the activity continues.

4. The urge to move or unpleasant sensations are worse in the evening or at night than during the day or only occur in the evening or at night.

Although the syndrome refers to restless “legs”, other body parts (arms, hips, trunk) can be involved, usually with disease progression. However the legs are almost always affected first, and most severely. The unpleasant sensations are often difficult to describe, sometimes characterised simply as “annoying” or “irritating”. A variety of adjectives such as “cramping”, “aching”, “electric-like shocks” and “muscle soreness” have been used.

Patients most often notice their symptoms at the end of the day while in bed, or during provocative situations, usually periods of prolonged inactivity such as sitting still on long car rides. Often patients would have found ways to alleviate their own symptoms, such as by rubbing the legs, taking walks or stretching. Symptoms are usually most severe in the early morning hours. Most patients do not notice symptoms in the daytime, but will regularly experience limb discomfort about the same time at night upon retiring to bed.

DIAGNOSIS OF RLS: SUPPORTIVE CLINICAL FEATURES
In addition to the 4 Essential Criteria listed above, there are other distinctive clinical features which support but are not needed for the diagnosis of RLS. The 3 supportive clinical features are:

1. Positive family history of RLS

2. Response (at least initially) to low doses of dopaminergic agents

3. Periodic limb movements during wakefulness or sleep

A positive family history of RLS is frequently reported, in the range of 60-90%. This is attributed to genetic factors thought to be important in the pathogenesis of RLS. Responsiveness to low doses of dopaminergic agents is so characteristic in primary RLS that a lack of drug effectiveness is a red flag for the possibility of an incorrect diagnosis or an underlying untreated secondary cause of RLS, such as iron deficiency or peripheral neuropathy. Numerous studies have shown that dopaminergic therapy (mainly levodopa and the dopamine agonists) is effective in reducing sensory and motor symptoms of RLS. The majority of RLS patients will improve at least partially with dopaminergic therapy, generally at much lower doses than are used for Parkinson’s disease. In addition to leg restlessness while awake, 80-90% of patients with RLS also have periodic limb movements in sleep (PLMS) (Montplaisir et al.,1997). These are periodic, involuntary limb movements comprising extension of the big toe, and triple flexion at the ankle, knee and sometimes the hip. The presence of PLMS can be inferred by a history of excessive leg movements (kicking, jerking) in sleep from the patient or reported by a bed partner. Objectively PLMS can be quantified on polysomnography (PSG), though PSG is not routinely indicated for the diagnosis of RLS, which is based on clinical criteria.

The terms RLS and PLMS are sometimes confused. RLS is a specific clinical diagnosis made based on the presence of symptoms experienced during wakefulness, frequently but not always associated with PLMS. PLMS are a polysomnographic finding, common especially in the elderly, rather nonspecific and often incidental findings on a sleep study. While most RLS patients have PLMS, the majority of patients with PLMS do not have RLS or any other sleep disorder.

RLS vs. PERIODIC LIMB MOVEMENT DISORDER
Periodic Limb Movement Disorder (PLMD) is another diagnostic term which is often used interchangeably, and incorrectly, with RLS probably because both PLMD and RLS are associated with PLMS. PLMD is diagnosed when there is an elevated PLM-index (15 or more PLMS per hour) associated with a sleep-related complaint (usually insomnia or excessive daytime sleepiness) which cannot be explained by another sleep, medical or psychiatric disorder, or medication/substance use (AASM, 2005). Therefore PLMD is a diagnosis of exclusion.

While the terminology for RLS and PLMD is often a source of confusion, the therapeutic approach to both disorders is similar because they are thought to have similar (albeit poorly understood) underlying pathophysiology. In contrast to RLS, the impact of PLMD on sleep and daytime sleepiness is unclear (Chervin, 2001). As such there is no consensus on the treatment of PLMD, which remains of uncertain clinical significance.

CLINICAL FEATURES OF RLS
The mean age of onset of RLS symptoms is 27 years, about 1/3 will have symptoms before the age of 20 years (Montplaisir et al., 1997). Although the diagnostic clinical features are characteristic, the diagnosis of RLS is often missed because patients with RLS tend to under-report symptoms, and physicians often fail to recognise the symptom complex. The leg discomfort is often incorrectly attributed to psychogenic causes, nocturnal leg cramps, arthritis or nonspecific musculoskeletal pain (Walters et al., 1996).

The most common symptoms are difficulty initiating sleep (sleep onset insomnia) due to restless leg sensations or difficulty maintaining sleep (sleep maintenance insomnia) due to PLMS. Sleep disruption can cause excessive daytime sleepiness. The patient or bed partner may report frequent leg jerks or restless sleep, and bed sheets strewn all over in the morning. In primary

RLS, the rest of the clinical evaluation is usually normal. Secondary causes of RLS like renal failure, peripheral neuropathy or pregnancy are usually self-evident.

DIFFERENTIAL DIAGNOSIS
Many other conditions can present with restless legs or pain in the legs at rest. For example, restless limb movements can be seen in habitual foot-tapping and akathisia, while leg discomfort is a feature of painful neuropathy, nocturnal leg cramps, peripheral vascular disease and arthritis. Although the symptoms can be distressing and more noticeable at rest in some of these other conditions, they are usually not rapidly relieved by movement, and do not invariably worsen in the evening or night. Sleep related leg cramps are a phenomena of sleep, not wakefulness, characterised by recurrent awakenings associated with painful sensations in the legs and palpable muscle hardness, relieved by massage, movement or application of heat (AASM, 2005). RLS can be readily distinguished with a careful history because the symptoms occur during wakefulness prior to sleep onset and are quite distinctive in their trademark circadian rhythmicity and rapid relief with movement alone.

ASSOCIATED DISORDERS: PRIMARY vs. SECONDARY RLS
In primary RLS there is no apparent cause other than a genetic predisposition. These patients may have an earlier age of onset with an autosomal dominant inheritance pattern in affected families; women may experience exacerbation of symptoms in pregnancy.

Secondary RLS occurs in association with other conditions, most commonly iron deficiency, pregnancy, renal failure and peripheral nerve disease. It is important to identify and treat RLS with associated disorders because symptoms generally improve when the underlying condition is resolved, such as correction of iron deficiency, after renal transplant and post-partum. Secondary RLS may present acutely and progress rapidly (such as after acute blood loss), and may not respond to conventional therapy if the underlying disease is not treated. Iron deficiency is an easily treatable problem, and should be considered in patients with risk factors such as a history of acute or chronic blood loss (gastrointestinal disease, menorrhagia, regular blood donation). Co-morbid conditions can also affect the management of RLS. For example, depression is common, treatable and can exacerbate insomnia caused by RLS or PLMD. Knowing that many common antidepressants such as the selective serotonin reuptake inhibitors (SSRIs) and tricyclic antidepressants (TCAs) can exacerbate RLS will influence the choice of drug therapy in RLS patients with depression.

DRUGS WHICH MAY WORSEN RLS
Many common drugs have been reported to induce or worsen RLS (Table 1). These include SSRIs (eg. fluoxetine, paroxetine, sertraline, mirtazapine), TCAs and neuroleptics (eg. risperidone, olanzapine). A careful review of the drug history is an essential part of the RLS evaluation. If possible, drugs which exacerbate RLS should be avoided. Patients with seemingly intractable RLS may improve after withdrawal of offending drugs alone.

PATHOPHYSIOLOGY OF RLS
The pathophysiology of RLS is unknown. However various lines of evidence have suggested that it is a multifactorial disorder of the central nervous system, including genetic factors, dysfunction of iron metabolism, dopamine and opioid neurotransmission.

Iron deficiency is one of the major associations of primary RLS. It is also a prominent feature of two of the conditions commonly associated with RLS: pregnancy and renal failure. Iron is thought to play a part in the pathogenesis of RLS via its role as a co-factor for tyrosine hydroxylase, the rate-limiting enzyme in dopamine synthesis. Low serum ferritin levels are associated with more severe RLS symptoms and PLM-associated arousals. Oral iron replacement has been reported to improve symptoms in iron deficient RLS patients, with greater benefits the lower the serum ferritin level (

The importance of dopamine neurotransmission in RLS is also suggested by the numerous pharmacological studies which have clearly demonstrated the effectiveness of dopaminergic agents in the treatment of RLS. Conversely dopamine antagonists can induce or worsen RLS.

As with the dopaminergic agents, numerous studies have demonstrated the effectiveness of opiates in the treatment of RLS while opioid antagonists can induce or worsen RLS. Overall, the involvement of dopamine neurotransmitter system seems probable in the pathogenesis of RLS However the role of the opioid system is less clear.

DIAGNOSTIC EVALUATION
Surprisingly many symptomatic patients with difficulty sleeping do not mention their restless legs unless specifically asked. Some patients may have had their vague and ill-defined symptoms ignored in the past, or incorrectly diagnosed. Others may have despaired of having their discomfort relieved, or fail to make a connection between their restless leg symptoms and sleep disturbance. Needless to say, all patients with sleep related complaints should be screened for RLS, which is a straightforward diagnosis once one becomes aware of its existence and characteristic features.

Once the diagnosis of RLS is made the clinical evaluation should be focussed on excluding exacerbating factors (especially drugs, listed in Table 1), co-existing depression which is common in RLS, and treatable secondary causes of RLS (especially iron deficiency and peripheral nerve disease).

Routine blood testing may include a complete blood count, renal function, serum ferritin, electrolyte and glucose levels. Folate levels should be checked in pregnant women. Additional testing if peripheral neuropathy is suspected would include vitamin B12 and thyroid stimulating hormone (TSH) levels, and oral glucose tolerance testing. If iron deficiency is detected, further investigation is indicated to determine the underlying cause, especially if there is associated anaemia and if the source of blood loss is unclear. It is important to exclude occult blood loss from gastrointestinal disease or malignancy.

Nerve Conduction Studies (NCS), Electromyography (EMG) and small fibre studies are indicated if peripheral nerve disease is suspected.

PSG is not required for the routine diagnosis of RLS which is based on clinical criteria alone. PSG is usually performed to rule out sleep disordered breathing as a cause of sleep disruption and daytime sleepiness. Besides ruling out OSA, quantification of PLMS can give an indication of disease severity and may influence the choice of drug therapy (eg. dopaminergic agents and the opioids are particularly effective at suppressing PLMS, more so than the anticonvulsants or the benzodiazepines). Typical findings on PSG in RLS are prolonged sleep onset latency, frequent leg movements before sleep onset, increased light sleep, decreased deep and REM sleep, frequent arousals and an elevated periodic limb movement index (number per hour of PLMS).

BEHAVIOURAL MODIFICATION THERAPY FOR RLS
Patients with mild or infrequent RLS may respond to non-drug therapies. In general all patients with sleep related complaints will benefit from good sleep hygiene instructions. Sleep deprivation can worsen RLS, so a regular sleep-wake schedule is important. Drugs which

exacerbate RLS (listed in Table 1) should be eliminated or avoided if possible. Most patients try walking around and counterstimulation techniques to alleviate their symptoms, such as rubbing their legs, hot or cold baths or ice packs. Other methods include distracting mental activities (games, hobbies), regular exercise (though not late in the evening or at night) and avoiding provocative situations (long periods of sitting still). Reduction in caffeine intake may be helpful because RLS can be exacerbated by caffeine. Alcohol can also exacerbate PLMS in susceptible individuals who take more than 2 alcoholic drinks in a day. Smoking cessation has been reported to relieve RLS. There are virtually no controlled studies on, nor any established effective non-pharmacological interventions for RLS.

DRUG THERAPY FOR RLS
Practice parameters for the dopaminergic treatment of RLS and PLMD were published in 2004 (Littner et al., 2004). Levodopa with decarboxylase inhibitor and the dopaminergic agents pergolide, pramipexole and ropinirole were listed as effective in the treatment of RLS and PLMD. Currently, the four major categories of drugs for RLS are the dopaminergic agents (dopamine precursor levodopa and the various dopamine agonists), anticonvulsants (mainly gabapentin), the sedative-hypnotics (benzodiazepines and the non-benzodiazepine sedative-hypnotic agents) and the opioids. The dopamine agonists are the most widely used first line agents, replacing levodopa because of a better side effect profile. Gabapentin has emerged as a relatively safe and effective drug for mild cases of RLS, and also for painful RLS. Generally the benzodiazepines and opioids are reserved for second or third line use because of the potential for dependence and abuse.

A suggested approach (including drug dosages and a checklist) to RLS management based on a review of literature and clinical experience is summarised in Tables 2 and 3. The medications are usually needed only as a single dose taken at night when RLS symptoms are troubling. Initial dosing should be low, with gradual titration upward to the lowest effective dose (“start low-go slow”).

1. DOPAMINERGIC AGENTS
While levodopa improves RLS symptoms, PLMS and sleep quality, its long term use is limited by the high risk of augmentation. With the introduction of dopamine agonists which have a better side effect profile, levodopa is no longer recommended for daily use because of the significant risk of augmentation (ie it may actually cause worsening of the symptoms after prolonged usage – see below). Levodopa should be reserved for intermittent use at low doses for rapid relief of breakthrough or infrequent symptoms.

The dopamine agonists are divided into the ergot-derived and the non-ergot-derived ones. The ergot-derived dopamine agonists have been associated with more side effects. In particular, isolated cases of pleuropulmonary fibrosis and cardiac valvulopathy have been described with pergolide. For this reason pergolide has been replaced by pramipexole and ropinirole, both non-ergot derived dopamine agonists, as first line agents for primary RLS.

Besides the well-known side effects (nausea, nasal congestion, orthostatic hypotension, fluid retention, sleepiness, insomnia, hallucinations and dyskinesias), treatment of RLS with dopaminergic agents is complicated by augmentation and rebound, both discussed in more details below. Although sudden, unexpected sleep attacks are rare in RLS patients taking dopamine agonists, patients should be warned of the possibility of sleepiness as a potential side effect of treatment.

AUGMENTATION WITH LEVODOPA & DOPAMINE AGONISTS

Augmentation refers to the worsening of RLS symptoms caused by a specific drug therapy, usually the dopaminergic agents. It has not been reported to occur with the anticonvulsants, benzodiazepines or opioid drugs. Augmentation usually occurs within 6 months of initiation of drug therapy or after a dose increase. Typically there is progressively earlier onset of symptoms, with symptom intensity increasing with an increase in the drug dosage and vice versa.

* Augmentation with dopamine precursors

Augmentation has been reported to occur in 82% of patients taking carbidopa/levodopa nightly, worse in patients with more severe symptoms and on higher doses exceeding 50/200 mg of carbidopa/levodopa daily. Augmentation may resolve with cessation of the medication and can be minimised by keeping the levodopa dose low.

* Augmentation with dopamine agonists

Dopamine agonists have replaced levodopa as first line agents for RLS because of the significantly lower risk of augmentation. Augmentation has been reported to occur in 33% of patients taking pramipexole, treatable with small dose increases or earlier dosing.

REBOUND WITH LEVODOPA

Rebound refers to the recurrence of symptoms at the end of a dosing period, usually with the shorter acting dopaminergic agents like levodopa (half-life 1.5 to 2 hours). This results in early morning exacerbation, which wakes the patient up, necessitating additional medication. Rebound should not be confused with augmentation. In rebound, recurrent symptoms occur in the early morning after an initial period of relief from medication. In augmentation, symptoms occur progressively earlier in the evening or afternoon and may progress in severity or spread to previously unaffected limbs.

End-of-dose rebound has been reported to occur in 25% of patients taking carbidopa/levodopa nightly (Allen and Earley, 1996). Rebound is less of a problem with the dopamine agonists with longer half-lives (6 to 8 hours or more) which have replaced levodopa as first line agents for RLS.

2. ANTICONVULSANTS
Anticonvulsants such as gabapentin, carbamazepine, valproic acid and topiramate have been reported to be effective. A mean effective dose of 1855mg of gabapentin was shown to improve RLS symptoms, sleep architecture and PLMS, with the most benefit in patients with painful RLS (Garcia-Borreguero et al., 2002). Gabapentin (dose range 300-1200 mg) has been reported to be well tolerated and as effective as ropinirole (dose range 0.25-1.5 mg) in reducing RLS symptoms and PLMS.

3. BENZODIAZEPINES AND NON-BENZODIAZEPINE SEDATIVE-HYPNOTICS
The use of benzodiazepines is limited by side effects, including sedation, respiratory depression, dependency and withdrawal. They should be avoided in patients with obstructive sleep apnoea (OSA), and may cause falls in the elderly at night. Generally they help induce and maintain sleep, reducing movement arousals but not PLMS. Although frequently prescribed for RLS, clonazepam has a long duration of action that can cause mental clouding in the daytime and should be avoided unless there is comorbidity for which it is also effective, such as anxiety, epilepsy or REM sleep behaviour disorder. The newer non-benzodiazepine sedative-hypnotic zolpidem has a shorter half-life (about 1-4 hours) and is thought to have a better side effect profile and preserve normal sleep architecture (unlike the benzodiazepines which suppress slow wave sleep). It may be a useful drug for patients with mild RLS who have sleep onset insomnia.

4. OPIOIDS
Opioid drugs have been used for RLS since the time of its original description in the 1600s. A retrospective study of the use of opioids in RLS showed sustained improvement in PLMS and sleep efficiency, with very low incidence of addiction or tolerance (Walters et al., 2001). Opioid therapy has also been reported to be effective in reducing PLMS in patients with PLMD.

The opioids are generally reserved for intractable cases of RLS because of the potential for dependence, even though studies have suggested that there is low potential for addiction and tolerance when used for RLS. Nonetheless, the opioids (and the benzodiazepines) should probably be avoided in patients with a history of substance abuse. Side effects are well known, including nausea, sedation, dizziness, constipation and respiratory suppression. Like the benzodiazepines, opioids should be avoided in patients with OSA.

IRON DEFICIENCY AND RLS
Oral iron therapy is safe, cheap and effective in restoring iron balance. Generally iron supplements are recommended for RLS patients with serum ferritin below 50 ug/L. A relatively inexpensive preparation is iron sulfate, containing 300 mg of iron salts, of which 60 mg is elemental iron. Iron is best absorbed in a mildly acidic medium: Vitamin C (250 mg ascorbic acid) is given at the same time to enhance iron absorption. Iron should be given on an empty stomach, and two hours before or four hours after ingestion of antacids. Improvement may take weeks to months.

In addition to oral iron supplements, the effectiveness of parenteral iron therapy has been reported in patients with RLS secondary to renal failure but is not routinely available (Sloand et al., 2004). Long term efficacy and safety of IV iron treatment for RLS have not been established.

FOLATE DEFICIENCY AND RLS
Folate deficiency is associated with several neurologic disorders including RLS. Folate replacement can improve RLS symptoms in these patients. Serum folate levels are not routinely studied, but should be checked in pregnant women, patients with neuropathy or with intractable symptoms unresponsive to low doses of the usual drugs for RLS.

PREGNANCY AND RLS
RLS has been reported to occur in 23% of women in the third trimester or to worsen during pregnancy in familial cases. RLS in pregnancy is associated with low serum ferritin and folate levels, depression and sleep onset insomnia. Most cases resolve after delivery.

Generally behavioural approaches and correction of iron and folate deficiency are all that is recommended in pregnant patients with RLS. There are few category A drugs for RLS in pregnancy: Only iron, folate and vitamin B12 may be used safely. Category B drugs (no evidence of risk by controlled studies) include pergolide, oxycodone, methadone and magnesium. Category C drugs (toxic to animals; human studies inadequate) include levodopa, gabapentin, carbamazepine, clonazepam, codeine, propoxyphene and clonidine. Temazepam is of proven teratogenicity.

RENAL FAILURE AND RLS
RLS has been reported in up to 60% of patients with renal failure on haemodialysis, and is associated with increased mortality. Secondary causes of RLS like iron deficiency and peripheral neuropathy both occur commonly in renal failure and should be excluded. Renal transplant but not haemodialysis is usually effective in resolving symptoms of RLS. RLS associated with renal failure responds to the same drugs used for primary RLS. Lower doses in general should be used with the renally excreted drugs such as gabapentin and the opioids.

Certain drugs such as gabapentin are dialysable while others such as carbamazepine and valproic acid are not.

PERIPHERAL NERVE DISEASE AND RLS
RLS has been reported in association with virtually every form of peripheral nerve disease, both axonal and demyelinating, hereditary and acquired, small and large fibre neuropathy.

Peripheral nerve studies (EMG, small fibre studies) are not routinely indicated in patients with RLS unless there are clinical features to suggest underlying peripheral nerve disease, such as sensorimotor deficits, underlying metabolic diseases or abnormalities on neurological examination. Patients complaining of painful feet may have small fibre neuropathy (Polydefkis et al., 2000). Treatable causes of neuropathy should be identified, including impaired glucose tolerance, electrolyte disturbances, hypothyroidism, vitamin B12 deficiency, infection (eg. HIV), vasculitis and the demyelinating neuropathies.

Patients with painful or neuropathic RLS may benefit from drugs used for neuropathic pain, especially gabapentin (Garcia-Borreguero et al., 2002).

DEPRESSION AND RLS
Psychiatric disorders (most often depression) have been reported in over 40% of men and women with RLS. While depression is common in RLS, many antidepressants (eg. SSRIs, TCAs) can actually exacerbate RLS symptoms. So whenever possible, alternative drugs which improve mood without exacerbating periodic leg movements and affecting normal sleep architecture should be considered. Bupropion (Wellbutrin) enhances central dopaminergic activity and has been shown to improve PLMS and depression in patients with PLMD (Nofzinger et al., 2000). Trazodone is another antidepressant which does not worsen RLS or PLMS. It has been shown to significantly improve quantity, quality and continuity of sleep in depressed patients with insomnia and on SSRIs (Kaynak et al., 2004). Of the dopamine agonists, pramipexole (which is probably the most widely used dopamine agonist for RLS) has been shown to have an antidepressant effect in addition to improving RLS symptoms (Saletu et al., 2002).

PROGNOSIS
Primary RLS tends to be a chronic condition with disease progression over time (Walters et al., 1996). Patients with young onset familial RLS may present only in middle age when symptoms become more frequent and disabling. Very mild cases may remit spontaneously or become symptomatic only during very prolonged periods of immobilisation. In secondary RLS, onset and progression can be rapid, with symptoms often resolving with treatment of the underlying condition. Generally the vast majority are responsive to conventional therapy. Most will experience at least some relief of symptoms, and only about 5% will remain refractory to treatment.

SUMMARY
RLS is a common, distressing and treatable disorder which should not be overlooked. It is readily diagnosed based on clinical criteria and responsive to relatively low doses of dopaminergic agents and anticonvulsants. Clinical practitioners should become familiar with the essential criteria, supportive clinical features, treatable associated conditions, common drugs which worsen symptoms and the first line drugs for RLS. Greater awareness of this distressing disorder for which effective treatment is available will improve the quality of lives of those who are affected by this frequently misdiagnosed problem.

Table 1. Drugs Which Worsen RLS

*Instructions for patients: Pergolide 0.05 mg or Pramipexole 0.125 mg or Ropinirole 0.25 mg:

Take ½ tablet 2 hours before bedtime. Increase to 1 tablet after 3 days if no side effects or benefit. Continue to increase by ½ tablet every 3 days until there is benefit or side effects develop.

ABSTRACT
Restless Legs Syndrome (RLS) is a commonly misdiagnosed disorder, which can be primary or associated with conditions like iron deficiency and renal failure. Diagnosis is based on clinical characteristics: a distressing urge to move the limbs, provocation by inactivity, relief by movement and nocturnal exacerbation. RLS can be managed with dopaminergic, anticonvulsant, sedative-hypnotic and opioid drugs.

INTRODUCTION
The term “Restless Legs Syndrome” was introduced in 1945 by Karl Ekbom, a Swedish neurologist. Although the syndrome refers to restless “legs”, other body parts (arms, hips, trunk) can be involved with disease progression, with the legs usually affected first, and most severely. The unpleasant sensations are difficult to describe, with adjectives such as “jumpy”, “cramping”, “aching”, “bugs crawling”, “electric-like shocks”. These sensations can be unilateral or bilateral, usually deep-seated and may be painful. Patients often notice their symptoms at the end of the day in bed, or during provocative situations like periods of prolonged inactivity. They may alleviate their symptoms by rubbing their legs, taking walks or stretching. Symptoms are most severe in the early morning hours (12MN-1am), and improve in the late morning (9-11am). Most patients do not notice symptoms in the daytime, but regularly experience limb discomfort about the same time at night upon retiring to bed. This distinctive circadian variation may be lost with disease progression (when symptoms may occur all day) or augmentation (when symptoms occur progressively earlier in the day) but is always present at disease onset.

The mean age of onset of RLS symptoms is 27 years, about 1/3 will have symptoms before the age of 20 years (Montplaisir et al. 61-65). Patients usually present years after initial symptom onset as disease severity progresses with age. The leg discomfort is often misdiagnosed as psychogenic, nocturnal leg cramps, arthritis or nonspecific musculoskeletal pain. Common symptoms are difficulty initiating sleep due to leg restlessness or difficulty maintaining sleep due to periodic limb movements in sleep (PLMS). Sleep disruption can cause excessive daytime sleepiness. The bed partner may report frequent leg jerks or restless sleep.

The reported prevalence of RLS ranges widely from less than 1% to over 20%, increasing with age and about twice as common in women (Phillips et al. 2137-41). There is an ethnic difference, RLS being less frequently reported in Asian populations (Tan et al. 577-79). RLS symptoms are commonly under-reported by patients and overlooked by physicians. In a survey of more than 23, 000 primary care patients, although 65% of the chronic sufferers had seen a doctor about their symptoms, fewer than 13% of patients were correctly diagnosed (Hening et al. 237-46). It is likely that the prevalence of RLS is underestimated and the scope of the problem larger than is generally recognized.

DIAGNOSIS OF RLS: ESSENTIAL CRITERIA & SUPPORTIVE CLINICAL FEATURES
RLS is a clinical diagnosis made based on fulfilling 4 essential criteria originally proposed in 1995 by the International RLS Study Group, and subsequently modified in 2002. (Walters 634-42;Allen et al. 101-19). The 4 Essential Criteria for RLS are:

1. An urge to move the legs, usually accompanied or caused by unpleasant sensations in the legs.
2. The urge to move or unpleasant sensations begin or worsen during periods of inactivity such as lying or sitting.
3. The urge to move or unpleasant sensations are partially or totally relieved by movement, such as walking or stretching, at least as long as the activity continues.
4. The urge to move or unpleasant sensations are worse in the evening or at night than during the day or only occur in the evening or at night.

Besides the 4 Essential Criteria, there are other additional clinical features which support but are not needed for the diagnosis of RLS:

1. Positive family history of RLS
2. Response (at least initially) to low doses of dopaminergic agents
3. Periodic limb movements during wakefulness or sleep

A positive family history of RLS is frequently reported, in the range of 60-90% (Ondo and Jankovic 1435-41). This is attributed to genetic factors thought to be important in the pathogenesis of RLS.

Responsiveness to low doses of dopaminergic agents is so characteristic in primary RLS that a lack of drug effectiveness is a red flag for the possibility of an incorrect diagnosis or an underlying untreated secondary cause like iron deficiency. Numerous studies have shown that dopaminergic therapy is effective in reducing sensory and motor symptoms of RLS (Benes et al. 1073-81). Most RLS patients will improve at least partially with dopaminergic therapy.

In addition to leg restlessness while awake, 80-90% of patients with RLS also have PLMS (Montplaisir et al. 61-65). These are periodic, involuntary limb movements comprising extension of the big toe, and triple flexion at the ankle, knee and hip. Although nonspecific, PLMS are so characteristic of RLS that their absence is another red flag for possibly an incorrect diagnosis.

The terms RLS and PLMS are sometimes confused. RLS is a specific clinical diagnosis based on symptoms experienced during wakefulness, frequently but not always associated with PLMS. PLMS are a polysomnographic finding in sleep, common in the elderly, rather nonspecific and often incidental. While most RLS patients have PLMS, the majority of patients with PLMS do not have RLS or any other sleep disorder (Trenkwalder, Walters, and Hening 629-50).

RLS IN SPECIAL POPULATIONS: COGNITIVELY IMPAIRED ELDERLY AND CHILDREN
RLS may be difficult to diagnose in the cognitively impaired elderly and in children because of their inability to describe their subjective symptoms. Observations made by family members and caregivers, and supportive clinical features are helpful. The following criteria have been proposed for the diagnosis of probable RLS in these special populations (Allen et al. 101-19).

Essential criteria for probable RLS in the cognitively impaired elderly (all 5 must be present) are:
1. Signs of leg discomfort (rubbing legs, groaning while holding the legs).
2. Excessive motor activity in the lower extremities (pacing, fidgeting, repetitive kicking, tossing and turning, cycling movements, repetitive foot tapping, rubbing the feet, inability to remain seated).
3. Signs of leg discomfort are exclusively present or worsen during periods of rest or inactivity.
4. Signs of leg discomfort are diminished with activity.
5. Criteria 1 and 2 occur only in the evening or at night or are worse at those times than during the day.

In addition to the usual supportive clinical criteria, suggestive features include the use of restraints at night in institutionalized patients and evidence of iron deficiency, end stage renal disease (ESRD), diabetes or peripheral nerve disease.

In children, the leg discomfort of RLS may be attributed to “growing pains”, musculoskeletal or joint problems and hereditary neuropathy (Walters et al. 241-45; Walters 93-98). Besides causing sleep disturbance as is typical in adults, RLS (like sleep apnoea) can be a cause of neurobehavioral problems in children. Childhood RLS has been associated with iron deficiency and attention-deficit hyperactivity disorder (ADHD) (Kryger, Otake, and Foerster 127-32;Picchietti et al. 1000-07;Picchietti et al. 588-94). While the diagnostic criteria are stricter in children because of their overall higher level of motor activity, adolescents and teenagers over the age of 12 years follow the same guidelines as for adults.

Essential criteria for definite RLS in children (age 2 through 12 years) include all 4 essential adult criteria and either one of the following:
1. The child relates a description in his/her own words consistent with leg discomfort
2. The presence of at least 2 of the following 3 supportive criteria:
a. Sleep disturbance for age
b. Biologic parent or sibling with definite RLS
c. Polysomnographically documented PLM-index of > 5

DIFFERENTIAL DIAGNOSIS
Many other conditions can present with restless legs or pain in the legs at rest. For example, restless limb movements can be seen in habitual foot-tapping and akathisia, while leg discomfort is a feature of painful neuropathy, nocturnal leg cramps, peripheral vascular disease and arthritis. Although the symptoms can be distressing and more noticeable at rest in some of these other conditions, they are usually not rapidly relieved by movement, and do not invariably worsen in the evening or night. Nocturnal leg cramps are a phenomena of sleep not wakefulness, characterized by recurrent awakenings associated with painful sensations in the leg and palpable muscle hardness, relieved by massage, movement or application of heat (American Academy of Sleep Medicine 31-33). RLS can be readily distinguished with a careful history because the symptoms occur during wakefulness prior to sleep onset and are quite distinctive in their trademark circadian rhythmicity and rapid relief with movement alone.

ASSOCIATED DISORDERS: PRIMARY vs SECONDARY RLS
In primary RLS there is no apparent cause other than a genetic predisposition. Patients have earlier symptom onset with an autosomal dominant inheritance pattern, women may experience exacerbation of symptoms in pregnancy (Winkelmann et al. 597-602).

Secondary RLS occurs in association with conditions like iron deficiency (O’Keeffe, Gavin, and Lavan 200-03), pregnancy (Lee, Zaffke, and Baratte-Beebe 335-41), chronic renal failure (CRF) (Winkelman, Chertow, and Lazarus 372-78) and peripheral nerve disease (Iannaccone et al. 2-9; Rutkove, Matheson, and Logigian 670-72; Polydefkis et al. 1115-21). Other reported associations include anxiety/depression (Sevim et al. 226-30), rheumatoid arthritis, diabetes and magnesium deficiency. Secondary RLS symptoms generally improve when the underlying condition is resolved, such as correction of iron deficiency, after renal transplant and post-partum (O’Keeffe, Noel, and Lavan 701-03;Winkelmann et al. 1072-76; Lee, Zaffke, and Baratte-Beebe 335-41).

While motor restlessness has been associated with Parkinson’s Disease (PD) with both conditions sharing common characteristics of dopaminergic dysfunction and response to dopaminergic therapy, the prevalence of RLS in PD patients has not been consistently found to be different than the general population (Tan, Lum, and Wong, 33-36). Though sharing similar features, the link between RLS and PD remains unclear: There is not enough evidence to suggest that the actual pathophysiologic mechanism in both diseases is identical (Tan, 33-37).

Many common drugs may exacerbate or induce RLS symptoms, including selective serotonin reuptake inhibitors (SSRIs eg. fluoxetine, paroxetine), tricyclic antidepressants, neuroleptics (eg. risperidone, olanzapine), anticonvulsants (eg. phenytoin, zonisamide), H1-antihistamines, antiemetics (eg. metoclopramide, prochlorperazine) and lithium. A review of the drug history is part of RLS evaluation, with offending drugs avoided if possible.

PATHOPHYSIOLOGY OF RLS
The pathophysiology of RLS is poorly understood, with multiple factors implicated, including genetics, dysfunction of the central and peripheral nervous system, iron metabolism, dopamine and opioid neurotransmission.

1. Anatomic localization
All levels of the central and peripheral neuraxis have been implicated in the pathogenesis of RLS. One hypothesis is that there is subcortical dysfunction causing increased activity of lumbosacral generators of periodic leg movements. There may be decreased supraspinal inhibition leading to hyperexcitability of motor pathways (Entezari-Taher et al. 1201-05; Tergau, Wischer, and Paulus 1060-63). Functional MRI studies have shown activation of the cerebellum and thalamus during sensory symptoms and activation of the red nucleus and part of the brainstem near the reticular formation during PLMS (Bucher et al. 639-45). Acute onset RLS and PLMS have been reported to occur with transverse myelitis, spinal cord injury and after spinal anaesthesia. Increased spinal cord excitability may be seen in patients with PLMS (Bara-Jimenez et al. 1609-16). The prominent sensory symptoms of RLS suggest that abnormal afferent input may be involved in the pathogenesis of RLS, possibly by modulating central motor pathways. RLS is associated with peripheral nerve disorders like peripheral polyneuropathy, lumbosacral radiculopathy and small fibre neuropathy (Iannaccone et al. 2-9; Rutkove, Matheson, and Logigian 670-72;Polydefkis et al. 1115-21).

2. Genetic Factors
There is a strong positive family history (60 to 90%) in RLS, inheritance occurring in an autosomal dominant pattern with variable penetrance and anticipation (Winkelmann et al. 597-602). Clinical features are similar to non-familial RLS, though with a younger onset. High concordance rate has been reported identical twins (Ondo, Vuong, and Wang 1404-06). Susceptibility loci have been identified on chromosomes 9, 12 and 14 (Bonati et al. 1485-92;Desautels et al. 1266-70). More recently, genome-wide association studies have identified variants within intronic or intergenic regions of BTBD9, MEIS1, and MAP2K5/LBOXCOR1. In an Icelandic population of RLS patients with PLMS, a significant association was found with a common variant in an intron of BTBD9 on chromosome 6p21.2. The population attributable risk with this variant of RLS with PLMS was approximately 50% (Stefansson et al. 639-647). Another study of German and Canadian RLS patients found highly significant associations between RLS and intronic variants in the homeobox gene MEIS1, the BTBD9 gene encoding a BTB (POZ) domain as well as variants in a third locus containing the genes encoding mitogen-activated protein kinase MAP2K5 and the transcription factor LBXCOR1 on chromosomes 2p, 6p and 15q, respectively (Winkelmann et al. 1000-6). Each genetic variant was found to be associated with a more than 50% increase in risk for RLS.

3. Iron Metabolism
Iron is thought to play a part in the pathogenesis of RLS via its role as a co-factor for tyrosine hydroxylase, the rate-limiting enzyme in dopamine synthesis. MRI studies have shown low levels of iron within the substantia nigra and putamen, correlating with RLS severity (Allen et al. 263-65). RLS is associated with low CSF ferritin and high transferrin, another reflection of low brain iron stores (Earley et al. 1698-700). Neuropathological studies have shown decreased iron and transferrin receptor staining in the brains of RLS patients (Connor et al. 304-09).

Low serum ferritin levels are associated with more severe RLS symptoms and PLM-associated arousals (O’Keeffe, Gavin, and Lavan 200-03). Oral iron replacement has been reported to improve symptoms in iron deficient RLS patients, with greater benefits the lower the serum ferritin level.

4. Neurotransmitter Systems
Functional neuroimaging studies in RLS have demonstrated dopaminergic dysfunction in the basal ganglia (Turjanski, Lees, and Brooks 932-37). The circadian variation in dopamine levels may explain the cyclical pattern of RLS (Sowers and Vlachakis 341-45). The role of dopamine neurotransmission in RLS is suggested by the demonstrated effectiveness of dopaminergic agents in RLS treatment. Conversely RLS is worsened by dopamine antagonists that cross the blood-brain-barrier but not peripheral ones like domperidone. Similarly, studies have shown the effectiveness of opiate therapy, while opioid antagonists can induce or worsen RLS (Walters et al. 1105-09). Overall, the involvement of dopamine neurotransmitter system seems probable in the pathogenesis of RLS, with the role of the opioid system being less clear.

DIAGNOSTIC EVALUATION
Many symptomatic patients do not mention their restless legs unless specifically asked. All patients with sleep related complaints should be screened for RLS, which is straightforward once one becomes aware of its existence and diagnostic features. Exacerbating factors and treatable secondary causes of RLS should be excluded. Physical examination is usually normal in primary RLS.

Routine blood testing should include a complete blood count, renal function, serum ferritin, electrolyte and glucose levels. Folate levels should be checked in pregnant women. If iron deficiency is detected, further investigation is indicated to determine the underlying cause eg. occult gastrointestinal blood loss or malignancy. Electrophysiology studies are indicated if peripheral nerve disease is suspected.

Polysomnography (PSG) is not required for the routine diagnosis of RLS which is based on clinical criteria. PSG is usually performed to rule out sleep apnoea and identify PLMS. Typical findings are a prolonged sleep onset latency, frequent leg movements before sleep onset, increased stages 1, 2 non-REM sleep, decreased slow wave and REM sleep, frequent arousals and elevated periodic limb movement index.

BEHAVIOURAL MODIFICATION THERAPY FOR RLS

Patients with mild or infrequent RLS may respond to non-drug therapies. In general all patients with sleep related complaints will benefit from good sleep hygiene instructions. Sleep deprivation can worsen RLS, so a regular sleep-wake schedule is important. Drugs which exacerbate RLS (listed in Table 1) should be eliminated or avoided if possible. Most patients try walking around and counter stimulation techniques to alleviate their symptoms, such as rubbing their legs, hot or cold baths or ice packs. Other methods include distracting mental activities (games, hobbies), regular exercise (though not late in the evening or at night) and avoiding provocative situations (long periods of sitting still). Reduction in caffeine intake may be helpful because RLS can be exacerbated by caffeine. Alcohol can also exacerbate PLMS in susceptible individuals who take more than 2 alcoholic drinks in a day. Smoking cessation has been reported to relieve RLS. There are virtually no controlled studies on, nor any established effective non-pharmacological interventions for RLS.

DRUG THERAPY FOR RLS
In 1999 the American Academy of Sleep Medicine published guidelines for RLS management (Chesson, Jr. et al. 961-68). Recommended agents were levodopa, pergolide, carbamazepine, oxycodone and propoxyphene. Iron supplementation, gabapentin, clonazepam and clonidine were listed as options.

Practice parameters for the dopaminergic treatment of RLS were updated in 2004 (Littner et al. 557-59). Levodopa with decarboxylase inhibitor and the dopaminergic agents pergolide, pramipexole and ropinirole were listed as effective in the treatment of RLS. Currently, the four major categories of RLS drugs are the dopaminergic agents, anticonvulsants, the sedative-hypnotics and the opioids. The dopamine agonists are the most widely used 1st line agents, replacing levodopa because of a better safety profile. Gabapentin has emerged as a relatively safe and effective drug for mild or painful RLS. Generally the benzodiazepines and opioids are reserved for 2nd or 3rd line use because of the potential for dependence and abuse. Drug therapy for RLS is summarized in table 1.

1. DOPAMINERGIC AGENTS
Low doses of levodopa (mean effective dose 31mg) have been shown to improve PLMS and sleep quality with a rapid onset of action within the first few days of therapy. While levodopa improves RLS symptoms, PLMS and sleep quality, its long term use is limited by the high risk of augmentation (Trenkwalder et al. 1184-89;Saletu et al. 611-26). Daily use is still widespread because of its effectiveness and a general lack of awareness of the potential for augmentation.

With the introduction of dopamine agonists which have a better side effect profile, levodopa is no longer recommended for daily use because of the significant risk of augmentation. Levodopa should be reserved for intermittent use at low doses for rapid relief of breakthrough or infrequent symptoms. In general much lower doses are used than for treatment of Parkinson’s disease.

The dopamine agonists are divided into ergot-derived (eg. pergolide, bromocriptine, cabergoline, apomorphine) and non-ergot-derived (eg. pramipexole, ropinirole). Ergot-derived dopamine agonists are associated with more side effects. Pleuropulmonary fibrosis and cardiac valvulopathy have been described with pergolide (Danoff et al. 313-16). As such pergolide has been replaced by pramipexole and ropinirole, non-ergot derived dopamine agonists, as 1st line agents for primary RLS. Pramipexole has been shown to significantly reduce periodic limb movements in wakefulness and in sleep, with sustained effectiveness over time without tolerance (Montplaisir et al. 938-43;Montplaisir, Denesle, and Petit 27-31). Effective doses ranged from 0.25 to 0.75mg, with most patients requiring a low optimal dose of 0.25mg. Ropinirole (dose range 0.25-4mg) has been shown to significantly improve RLS symptoms, quality of life, quality of sleep and reduce PLMS (Trenkwalder et al. 92-97). Other non-ergot derived dopamine agonists shown to be effective include piribedil, talipexole and terguride.

Although sudden, unexpected sleep attacks are rare with dopamine agonists, patients should be warned of the possibility of sleepiness as a potential side effect of treatment. Other well-known side effects include nausea, nasal congestion, orthostatic hypotension, fluid retention, insomnia, hallucinations, dyskinesias, augmentation and rebound.

AUGMENTATION WITH LEVODOPA & DOPAMINE AGONISTS
Augmentation refers to the worsening of RLS symptoms caused by a specific drug therapy, usually dopaminergic agents. Augmentation usually occurs within 6 months of drug initiation or after a dose increase. Typically there is progressively earlier symptom onset, increasing symptom severity with intensity increasing with drug dosage, shorter latency to symptoms at rest, extension of symptoms to previously unaffected body parts and decreasing duration of treatment effect.

Augmentation has been reported to occur in 82% of patients taking carbidopa/levodopa nightly, worse in patients with more severe symptoms and on higher doses exceeding 50/200 mg of carbidopa/levodopa daily (Allen and Earley 205-13). Augmentation resolved with cessation of the medication and could be minimized by keeping the levodopa dose low. Dopamine agonists have replaced levodopa as 1st line agents because of the significantly lower risk of augmentation.

REBOUND WITH LEVODOPA
Rebound refers to the recurrence of symptoms at the end of a dosing period, usually with the shorter acting dopaminergic agents like levodopa (half-life 1.5 to 2 hours), causing early morning exacerbation which wakes the patient up, necessitating additional medication. Rebound should not be confused with augmentation. In rebound, recurrent symptoms occur in the early morning after an initial period of relief from medication. In augmentation symptoms occur progressively earlier in the evening or afternoon and may progress in severity or spread to previously unaffected limbs. End-of-dose rebound has been reported to occur in 25% of patients taking carbidopa/levodopa nightly (Allen and Earley 205-13). The addition of sustained release formulations of levodopa at bedtime can reduce PLMS and improve sleep quality in the 2nd half of the night (Collado-Seidel et al. 285-90). Duration of action of levodopa may be increased using entacapone, a catechol O-methyltransferase (COMT) inhibitor which produces longer symptomatic relief (Sharif 421). Rebound is less of a problem with the dopamine agonists with longer half-lives (6 to 8 hours or more) which have replaced levodopa as 1st line agents.

2. ANTICONVULSANTS
a. Gabapentin: A mean effective dose of 1855mg was shown to improve RLS symptoms, sleep architecture and PLMS, with the most benefit in patients with painful RLS. (Garcia-Borreguero et al. 1573-79). Gabapentin (dose range 300-1200 mg) has been reported to be well tolerated and as effective as ropinirole in reducing RLS symptoms and PLMS (dose range 0.25-1.5 mg) (Happe et al. 82-86).
b. Carbamazepine: This has been reported to improve RLS symptoms but is less used because of potential complications including aplastic anemia and agranulocytosis (Telstad et al. 444-46).
c. Valproic acid: Slow-release valproic acid (600 mg) has been shown to be as effective as slow-release levodopa (200 mg, with 50 mg benserazide) in a randomized, placebo-controlled, cross-over, double-blind study (Eisensehr et al. 579-83).
d. Topiramate: Topiramate has been reported to improve sensory and motor RLS symptoms (mean effective dose 42 +/- 19 mg) in an observational study. Significant weight loss was among the side effects, with good tolerability and few adverse reactions overall (Perez 132-37).

3. BENZODIAZEPINES AND NON-BENZODIAZEPINE SEDATIVE-HYPNOTICS
The use of benzodiazepines is limited by side effects, including sedation, respiratory depression, dependency and withdrawal. They should be avoided in patients with sleep apnoea, and may cause falls in the elderly at night. Generally they help induce and maintain sleep, reducing movement arousals but not PLMS. Although frequently prescribed for RLS, clonazepam has a long duration of action that can cause mental clouding in the daytime and should be avoided unless there is comorbidity for which it is also effective, such as anxiety or epilepsy. The newer non-benzodiazepine sedative-hypnotic zolpidem has a shorter half-life (1.4-4.5 hours) compared to the benzodiazepines like clonazepam (30 to 40 hours), and is thought to have a better side effect profile and preserve normal sleep architecture (unlike the benzodiazepines which suppress slow wave sleep). It may be a useful drug for patients with mild RLS who have sleep onset insomnia. Zolpidem 10 mg daily in a prospective open-label study produced sustained relief of RLS symptoms without side effects. Symptoms recurred on stopping and resolved on resuming zolpidem (Bezerra and Martinez 180-81).

4. OPIOIDS
The opioids are generally reserved for intractable cases of RLS because of the potential for dependence, even though studies have suggested that there is low potential for addiction and tolerance when used for RLS. Nonetheless, the opioids (and the benzodiazepines) should probably be avoided in patients with a history of substance abuse. Side effects include nausea, sedation, dizziness, constipation and respiratory suppression. Like the benzodiazepines, opioids should be avoided in patients with sleep apnoea. Opioids may be considered in patients with pain as a major symptom, when other conventional drugs have failed, or in the post-operative setting when oral medication is restricted (eg. parenteral morphine, fentanyl). Oral opioids include the regular and sustained release formulations, and the low potency (codeine, propoxyphene) and high potency (oxycodone, hydrocodone, methadone) forms.

SECONDARY RLS
Iron studies should be performed on all patients with RLS, and especially those with a history to suggest acute or chronic blood loss. Iron supplementation with ferrous sulphate has been reported to be effective in relieving RLS symptoms in elderly patients with low ferritin levels (below 45 ug/L) (O’Keeffe, Gavin, and Lavan 200-03). Oral iron therapy is safe, inexpensive and effective in restoring iron balance. Generally iron supplements are recommended for RLS patients with serum ferritin below 50 ug/L. The cheapest preparation is iron sulphate, containing 300 mg of iron salts, of which 60 mg is elemental iron. Iron is best absorbed in a mildly acidic medium: Vitamin C (250 mg ascorbic acid) is given at the same time to enhance iron absorption. Iron should be given on an empty stomach, and two hours before or four hours after ingestion of antacids. Improvement may take weeks to months.

Folate deficiency is associated with many neurologic disorders including RLS. Folate replacement can improve RLS symptoms. Serum folate levels should be checked in pregnant women, patients with neuropathy or with intractable symptoms unresponsive to low doses of the usual drugs for RLS.

RLS has been reported to occur in 23% of women in the 3rd trimester or to worsen during pregnancy in familial cases (Winkelmann et al. 597-602; Lee, Zaffke, and Baratte-Beebe 335-41). RLS in pregnancy is associated with low serum ferritin and folate levels, depression and sleep onset insomnia. Most cases resolve after delivery. Generally behavioural approaches and correction of iron and folate deficiency are recommended. There are few category A drugs for RLS in pregnancy: Only iron, folate and vitamin B12 may be used safely. Category B drugs (no evidence of risk by controlled studies) include pergolide, oxycodone, methadone and magnesium. Category C drugs (toxic to animals, human studies inadequate) include levodopa, gabapentin, carbamazepine, clonazepam, codeine, propoxyphene and clonidine.

RLS has been reported in up to 60% (more frequently in Caucasians) of CRF patients on hemodialysis, and is associated with increased mortality. Renal transplant but not haemodialysis is effective in resolving RLS symptoms. A study of hemodialysis patients showed that RLS symptoms resolved within 3 weeks of successful renal transplantation. RLS recurred with transplant failure and improved again with a second transplant in 1 patient (Winkelmann et al. 1072-76). RLS associated with CRF responds to the same drugs used for primary RLS. Lower doses in general should be used with the renally excreted drugs such as gabapentin and the opioids. Certain drugs such as gabapentin are dialyzable while others such as carbamazepine and valproic acid are not.

RLS has been reported in association with virtually every form of peripheral nerve disease.
Treatable causes of neuropathy should be identified. Patients with painful or neuropathic RLS may benefit from drugs used for neuropathic pain, especially gabapentin (Garcia-Borreguero et al. 1573-79).

Psychiatric disorders (most often depression) have been reported in over 40% of men and women with RLS. RLS severity has been shown to correlate with the severity of depression. While depression is common in RLS, many antidepressants (eg. SSRIs, TCAs) can actually exacerbate RLS symptoms. Whenever possible, alternative drugs which improve mood without exacerbating RLS should be considered. Bupropion (Wellbutrin) enhances central dopaminergic activity and has been shown to improve PLMS and depression in patients with periodic limb movement disorder (PLMD) (Nofzinger et al. 858-62). Bupropion also has an alerting effect which may be useful in RLS patients with daytime sleepiness. Extended formulations (Wellbutrin SR, Wellbutrin XL) allow once a day dosing. Trazodone is another antidepressant which does not worsen RLS or PLMS. It has been shown to significantly improve quantity, quality and continuity of sleep in depressed patients with insomnia and on SSRIs (Kaynak et al. 15-20). Trazodone may be considered in depressed patients who have RLS and sleep onset insomnia. Of the dopamine agonists, pramipexole (has been shown to have an antidepressant effect in addition to improving RLS (Saletu et al. 185-94).

PROGNOSIS
Primary RLS tends to be a chronic condition with disease progression over time. Patients with young onset familial RLS may present only in middle age when symptoms become more frequent and disabling. Very mild cases may remit spontaneously or become symptomatic only during prolonged periods of immobilization. In secondary RLS, onset and progression can be rapid, with symptoms often resolving with treatment of the underlying condition. Generally the vast majority are responsive to conventional therapy. Most will experience at least some relief of symptoms, and only about 5% will remain refractory to treatment.

CONCLUSION
RLS is a common, distressing and treatable disorder which should not be overlooked. It is readily diagnosed based on clinical criteria and responsive to relatively low doses of dopaminergic agents or anticonvulsants. Clinical practitioners should become familiar with the essential diagnostic criteria, supportive clinical features, treatable associated conditions, common drugs which worsen symptoms and the 1st line drugs for RLS. Accurate diagnosis and treatment will improve the quality of life of the majority of patients with RLS, one of the most frequently overlooked causes of disturbed sleep.

Table 1: Drugs Used In Restless Legs Syndrome

*Instructions for patients: Pramipexole 0.125 mg or Ropinirole 0.25 mg:
Take ½ tablet 2 hours before bedtime. Increase to 1 tablet after 3 days if no side effects or benefit. Continue to increase by ½ tablet every 3 days until there is benefit or side effects develop.

Table 2: Drug Therapy for Restless Legs Syndrome

Evaluation and Treatment of Circadian Rhythm Sleep Disorders

Introduction
The circadian clock, which resides in the suprachiasmatic nucleus of the hypothalamus, determines the physiological level of alertness in an approximately 24-hour cycle, thereby regulating the timing of sleep. This “biological” clock is normally synchronised to the external environment, resulting in alertness and peak performance during daylight hours and consolidated sleep during the night. For optimal sleep, the habitual sleep time should match that of the circadian clock determined level of least alertness, and when there is a heightened pressure to sleep, usually at the end of the day when a significant sleep debt has also accumulated. Such ideal conditions which are conducive to sleep occur routinely in normal people who keep regular sleep-wake schedules. This is the basis of the sleep hygiene “rule” of keeping to a fixed sleep-wake schedule. A fixed wake time is especially important to anchoring our endogenous clock. Sleep disturbances occur when the internal circadian timing system is out-of-synch with the external 24-hour physical and social environment. This is commonly seen in shift workers who do not have regular work (and sleep) hours (“shift work disorder”), long haul travellers who cross multiple time zones and have to function in an external environment which is misaligned with their internal clocks (“jet lag disorder”) and in institutionalised elderly patients who have disorganised sleep-wake patterns with sleep-wake periods scattered throughout the day usually due to lack of light exposure or physical activity (“irregular sleep wake rhythm”). Irregular sleep wake rhythm is believed to be due to neurologic disease (eg. dementia, brain trauma), and/or from lack of exposure to periodic synchronisers such as light, physical and social activity (Reid et al., 2004). Circadian rhythm sleep disorders (CRSD) are persistent sleep disturbances primarily due to abnormalities of the endogenous circadian timing system or misalignment between this internal circadian clock and the environment (AASM, 2005). In CRSD, the pattern of sleep-wake is out-of-synch with the patient’s internal circadian system, resulting in insomnia, fatigue, excessive daytime sleepiness (EDS) and deterioration in performance (Reid et al, 2004). The clock can be reset when misaligned, such as with appropriately timed light exposure and administration of melatonin which has chronobiotic (phase shifting) properties. This forms the basis of light and melatonin therapy for the treatment of CRSD. The process by which this optimal synchronisation occurs is called “entrainment”. The American Academy of Sleep Medicine (AASM) has recently published guidelines for the clinical evaluation and treatment of CRSD (Morgenthaler et al., 2007). These will be discussed in this chapter, including appropriate and safe use of timed light therapy, timed melatonin administration and chronotherapy.

Entrainment of Circadian Rhythms

Light is the strongest agent which can synchronise the circadian clock, advancing or delaying circadian rhythms depending on the time of exposure (Czeisler et al., 1986). Appropriately timed exposure to bright light (and darkness) can reset the internal clock to become synchronised with one’s desired sleep and wake times, thereby improving sleep quality and daytime alertness (Terman, 2005). Exposure to light in the first half of the night delays circadian rhythms, while light exposure in the second half of the night or early morning will advance the circadian rhythms.Entrainment of circadian rhythms can also be achieved with melatonin, which gives a less potent phase shifting response which is generally in the opposite direction to light (Lewy et al., 1998). Melatonin given in the early evening advances circadian rhythms, while melatonin given in the early morning delays circadian rhythms.Besides bright light and melatonin, other agents which can entrain human circadian rhythms are social and physical activities, including eating and exercise.

Evaluation of Circadian Rhythm Sleep Disorders (CRSD)

The diagnosis of CRSD can usually be made from a detailed history, as with the insomnias and sleep disorders presenting with excessive daytime sleepiness (EDS), based on pattern recognition of the distinctive clinical syndromes.

The more common CRSD are the extrinsic ones, shift work disorder and jet lag disorder. The intrinsic circadian rhythm abnormalities such as Advanced Sleep Phase Syndrome (ASPS), Delayed Sleep Phase Syndrome (DSPS) and irregular sleep wake rhythm are generally less common than the extrinsic ones. While precise mechanisms are not well understood in intrinsic CRSD, the endogenous circadian rhythm and sleep homeostatic processes are believed to interact abnormally (AASM, 2005).

ASPS typically begins in middle age, characterised by advanced habitual sleep onset and wake times, typically presenting with EDS in the late afternoon or early evening and spontaneous early morning awakening. Conversely, DSPS occurs in adolescents and young adults who have delayed habitual sleep onset and wake times, typically presenting with sleep onset insomnia (at conventional bedtimes), difficulty waking (at conventional wake times) and EDS. All of these patterns can be discerned from a detailed sleep history and sleep logs.

CRSD should always be considered in the differential diagnosis of patients presenting with insomnia or EDS. For example, in a teenager with sleep onset insomnia and EDS in the early morning hours who is unable to get up for school may have DSPS. Sleep maintenance insomnia in an older person with early morning awakenings may indicate ASPS. Depression is a differential diagnosis in both of these examples, and should always be excluded. Nocturnal insomnia with frequent daytime naps in an elderly patient in a nursing home suggests irregular sleep-wake rhythm. The appropriate intervention is dependent on making an accurate diagnosis. Other medical and sleep disorders which could mimic or exacerbate an underlying CRSD should be considered. For CRSD there is a range of therapeutic interventions, including planned sleep and activity schedules, timed light administration, timed melatonin, hypnotics and stimulants.

In CRSD, the diagnostic sleep pattern can be determined from the history provided by the patient. A sleep log or diary is helpful in the assessment of the sleep pattern. Actigraphy, which provides an estimate of sleep and wake by analysing movement data, can be used in both diagnosis as well as evaluating treatment response for CRSD. Polysomnography (PSG) is not routinely indicated for the diagnosis of CRSDs but may be indicated to rule out another primary sleep disorder eg. obstructive sleep apnoea (OSA) (Morgenthaler et al., 2007).

Management of Circadian Rhythm Sleep Disorders
A range of therapeutic interventions can be considered in the management of CRSD. Foremost is keeping to a fixed sleep-wake schedule whenever possible, which is most physiological. This however would represent difficulty in individuals who are engaged in shift work or travel frequently across multiple time zones. The AASM has published guidelines on the following: planned sleep schedules, timed light exposure, timed melatonin doses, hypnotics, stimulants and alerting agents (Morgenthaler et al., 2007). These are discussed in this section in greater detail. Planned sleep schedules, timed light exposure and timed melatonin doses are the usual initial interventions used in CRSD, with varying strengths of recommendation, ranging from “Standard” (Level 1 to 2 evidence), “Guideline” (Level 2 to 3 evidence), to “Option” (Inconclusive or conflicting evidence). Hypnotic medications may be indicated in shift work disorder to improve daytime sleep in night shift workers with insomnia (Guideline) or for insomnia in jet lag disorder (Option). Stimulants may be used to improve alertness in shift work and jet lag disorder (Option). Specifically modafinil may be used to improve alertness during the night shift in patients with shift work disorder (Guideline).

In addition to all of the above interventions, good sleep hygiene and social factors need to be considered in patients with CRSD. Adequate sleep on a regular basis and fixed sleep-wake schedules are helpful whenever feasible. The sleep environment needs to be optimised, eg. for shift workers who sleep in the daytime, including decreasing daytime noise and a darkened bedroom conducive to sleep. Appropriately timed exposure to synchronising agents such as bright light, structured social and physical activities, and eating meals at the proper times can all help to consolidate sleep-wake cycles. Educating the patient, family members, caregivers and employers is an important part of effectively managing CRSD.

1. Planned Sleep Schedules
In shift work disorder napping before shifts helps to improve alertness and vigilance, and reduce accidents during night shift work without affecting post-shift daytime sleep (Standard). In jet lag disorder, one suggested sleep schedule is to keep to home-base sleep hours when the time at destination is expected to be brief (ie. two days or less) (Option). Another suggested method is to shift the sleep schedule one hour earlier each day for three days prior to eastward travel to lessen the symptoms of jet lag, in combination with morning exposure to bright light (> 3, 000 lux for 3.5 hours) (Option) (Morgenthaler et al., 2007). Such techniques alone or in combination may be impractical because of the significant behavioural changes required by the busy patient.

Prescribed sleep-wake scheduling, together with timed bright light (evening light exposure) and melatonin are indicated as treatments for ASPS (Option). Although the evidence for efficacy for these interventions is weak or conflicting, their use can be considered in ASPS given that the risks and costs involved are low, and there are few treatment alternatives. In ASPS, advancing the sleep phase by 3 hours every 2 days may allow the patient to achieve the desired later sleep schedule. This is referred to as “chronotherapy” but has had limited success because of difficulty with compliance with behavioural restrictions.

Chronotherapy has also been used in DSPS. The prescribed progressive delay in the scheduled sleep time until the desired sleep schedule is reached may be useful for DSPS (Option). However compliance is difficult and lasting benefit has not been shown. This form of therapy involves progressively delaying sleep times by approximately 3 hours a day until the desired earlier sleep schedule is achieved, and should be combined with enforcement thereafter of regular sleep-wake times. Although effective, its practicality is limited in the clinical setting because of its impact on social, professional and school schedules.

2. Timed Light Exposure
Bright light therapy for CRSD is mediated by ocular photoreceptors (specialised retinal ganglion cells) which project to the circadian clock in the suprachiasmatic nucleus. The efficacy of light therapy in the treatment of CRSD is highly dependent on the time of day that light is administered. The magnitude and direction of phase-shifting is dependent on the circadian phase at which the light exposure occurs. Exposure to bright light for 1 to 2 hours in the morning results in advance of the phase of circadian rhythms, while evening bright light exposure causes phase delays. Studies have shown that exposure to bright artificial light (typically 2,500-10,000 lux) improves sleep-wake quality and mood (Terman, 2005). Bright light also increases subjective ratings of alertness, as well as objective measures of arousal and performance (Cajochen et al., 2000).

In shift work disorder timed light exposure of various intensities (from 2, 350 to 12,000 lux) for as long as possible during the night shift has been reported to give subjective improvements in performance, alertness and mood, compared to ordinary light exposure. Timed light exposure at night in the work environment and light restriction in the morning is indicated to decrease sleepiness and improve alertness during night shift work (Guideline). It is important to note that avoidance of light in the early morning is just as important as bright light exposure during the night in the work environment to ensure alignment of circadian clock determined sleep propensity with the desired daytime sleep period. Avoidance of bright light in the early morning on the way home after the night shift can be achieved by wearing dark glasses.

In jet lag disorder light therapy can be used in combination with planned sleep schedules (as discussed above). The precise timing of exposure to light and darkness in jet lag disorder is dependent on the individual’s circadian phase, the direction of travel (eg. eastward or westward) and the number of time zones to be crossed. To address this complex multitude of variables, computer and website based “calculators” have been developed with algorithms for light treatment of jet lag (Arendt et al., 2005; Houpt et al., 1996).

Evening bright light therapy (usually between 7 to 9 pm) has been used in ASPS to delay the phase of circadian rhythms and improve sleep efficiency. Daily light therapy should be administered in the evening, just prior to bedtime in order to delay the timing of the circadian rhythm of sleep-wake in such patients. In addition, the patient should avoid bright light exposure in the morning hours after awakening, to avoid an undesirable phase advance shift of the sleep-wake cycle. While potentially useful, the efficacy of evening morning light therapy in the treatment of ASPS has not been well studied in clinical practice. Light therapy can be combined with melatonin taken in the early hours of the morning, though the evidence for melatonin in ASPS is also lacking.

Morning light exposure is indicated in the treatment of DSPS, but optimal timing, duration and dosing of morning light treatment have yet to be determined (Guideline). Studies have used light intensities of 2, 500 lux for 2 to 3 hours prior to or at rise time. Bright light exposure during the early morning hours and avoidance of bright light in the evening have been used in the treatment of DSPS, resulting in earlier sleep times and improved morning alertness. Exposure to broad-spectrum light of 2, 000 to 10, 000 lux for about 1 to 2 hours is generally recommended in clinical practice.

In irregular sleep wake rhythm daytime bright light exposure may improve circadian rest-activity rhythms and consolidation of sleep and wake in nursing home residents with dementia (Option). A combination of approaches including bright light exposure, physical activity and other behavioural/environmental modalities (eg. structured bedtime routine, sleep scheduling, decreased night-time noise and light, sleep hygiene) are indicated in the treatment of irregular sleep wake rhythm in older people with dementia (Guideline).

Safety Concerns of Light Therapy
There are no standard guidelines governing the application of light therapy for the treatment of circadian rhythm sleep disorders, though attempts have been made to establish basic safety standards and recommendations for the use of light therapy devices (Terman, 2005). Recent studies have suggested that blue light (rather than white light) is most effective at resetting circadian rhythms, suppressing melatonin at night and enhancing performance (Lockley et al., 2006). As such white light boxes that are currently used for treatment of CRSD may soon be replaced by devices which emit blue-enriched light. There is a range of commercially available light therapy devices such as white (and coloured eg. blue LED) light boxes and light visors, none of which have been clinically tested for efficacy or safety. Light therapy devices should provide adequate filtering of ultraviolet and infrared light. A useful resource for the selection of light therapy devices can be found on the Center for Environmental Therapeutics website (www.cet.org).

In the clinical setting, caution is needed in patients taking photosensitising medications (eg. antipsychotics, antidepressants and antiarrhythmic agents). Patients with glaucoma and cataracts may be at risk for ocular damage, while retinopathy is considered an absolute contraindication for bright light therapy (Terman, 2005). Adverse effects of bright light therapy include irritability, headaches, nausea, blurred vision, eyestrain and photophobia. An ophthalmology or dermatology consultation is suggested if there is any doubt regarding the safety of administering light therapy. All new patients should undergo an eye examination prior to starting light therapy.

Since light therapy devices may not be readily available in Singapore, natural sunlight may be considered an alternative, though natural lighting is unpredictable eg. natural light outdoors at dusk or dawn on a clear day may provide illuminance similar to that of artificial light from a commercial light box (about 10,000 lux).

3. Melatonin
Although melatonin is widely sold as a nutritional supplement, little is known about the safety and efficacy of chronic administration. Melatonin preparations in the U.S. are not regulated by the Food and Drug Administration. While melatonin has not been proven safe in large scale studies, no major adverse effects have been associated to its use. In general, the benefits of melatonin in CRSD seem to be well supported, with relatively low risks. Melatonin has been used as an over-the-counter sleeping aid in sleep disorders because of its chronobiotic properties and a direct hypnotic effect (Sack et al., 1998).

In shift work disorder, melatonin at doses ranging from 0.5 to 10 mg given prior to daytime sleep improves daytime sleep quality and duration in night shift workers (Guideline). Level 1 studies have shown beneficial effects on sleep quality using melatonin at 1.8 to 3 mg.

In jet jag disorder, melatonin has been studied in doses ranging from 0.5 to 10 mg, given at bedtime, for up to 3 days prior to departure and up to 5 days upon arrival at the destination. Melatonin administered at the appropriate time is indicated in jet jag disorder to reduce symptoms of jet lag and improve sleep following travel across multiple time zones (Standard). The most effective dose of melatonin for jet jag disorder is unclear, but studies suggest that immediate-release melatonin formulations at 0.5 to 5 mg may be effective.

There is no reported evidence in support of the use of melatonin in ASPS. In DSPS, properly timed melatonin is indicated as therapy for DSPS (Guideline). Melatonin has been studied to be effective at doses ranging from 0.3 to 5 mg, administered at 1.5 to 6 hours prior to habitual bedtime. Melatonin given in this way can shift circadian rhythms to an earlier time, and reduce sleep onset latency. However the optimal timing and dosing of melatonin in DSPS are not established.

Melatonin is not indicated for treatment of irregular sleep wake rhythm in older people with dementia (Option). Studies on patients with Alzheimer’s disease and irregular sleep wake rhythm have shown no evidence of improved sleep as measured by actigraphy.

4. Sedative-Hypnotics
The appropriate use of sedative-hypnotics for insomnia in general of various aetiologies has been discussed in the chapter on Treatment of Insomnia. Such drugs should be used very cautiously at the lowest effective doses, preferably intermittently (rather than nightly) to reduce the risk of tolerance and dependence. Drugs should be chosen with the appropriate pharmacokinetic profile, and only after a correct diagnosis of the specific cause of the insomnia, and exclusion of other primary sleep disorders such as OSA, as well as organic mental disorders and substance abuse.

In shift work disorder sedative-hypnotics may be used to promote daytime sleep in night shift workers (Guideline), though this benefit should be weighed against residual sedation affecting night-time performance, and other potential adverse effects such as worsening OSA, tolerance and dependence. Treatment with sedative-hypnotics should always be individualised and closely monitored; specialist input may be needed in difficult cases.

In jet lag disorder, short-term use of sedative-hypnotics is indicated for treatment of insomnia (Option). Different agents have been studied and shown to improve sleep quality and duration, including zolpidem (10 mg) and zopiclone (7.5 mg). Again, potential benefits must be considered carefully and weighed against adverse effects.

There is insufficient evidence to support the use of sedative-hypnotic medication to promote sleep (or the use of stimulants to promote alertness) in DSPS (Option).

5. Stimulants and Alerting Agents
Wake promoting agents which have been used in sleep disorders include modafinil, methylphenidate, amphetamines and caffeine. Modafinil (Provigil) can be used to enhance alertness in OSA, narcolepsy and shift work disorder. Modafinil promotes wakefulness by an unknown mechanism and is usually given in doses of 100 to 200 mg daily. Amphetamines are non-catecholamine sympathomimetic amines with CNS stimulant activity which have a high potential for abuse. They may cause elevations in blood pressure and heart rate, and have been associated with arrhythmias and sudden death.

In shift work disorder stimulant drugs (eg. modafinil, methamphetamine) are effective in reducing sleepiness and improving performance during the night shift. Modafinil is indicated to enhance alertness during the night shift for SWD (Guideline). Caffeine is also indicated to enhance alertness during the night shift in shift work disorder (Option). Modafinil and caffeine have established safety records, so generally the benefits outweigh the risks when used to enhance alertness during the night shift in shift work disorder. Modafinil is not registered in Singapore, however user-specific approval may be granted at the discretion of the Health Sciences Authority (HSA).

In jet lag disorder caffeine is indicated to counteract sleepiness, but may also disrupt night-time sleep (Option).

Summary
A good understanding of the pathophysiology of CRSD and recognition of the distinctive sleep patterns of the more common ones are needed to correctly manage these infrequently encountered conditions. The CRSD are probably under-recognised, and should be considered in every patient who presents with insomnia or EDS. As with all patients with sleep disorders, good sleep hygiene (including ensuring adequate sleep regularly) and sleep education are an important part of managing these patients. Appropriate use of planned sleep schedules, timed light exposure and timed melatonin administration require a basic understanding of circadian biology. Sedative-hypnotic and stimulant use should be individualised, preferably intermittent and at lowest effective doses, and closely monitored.

References
American Academy of Sleep Medicine (AASM). The International Classification of Sleep Disorders: Diagnostic and Coding Manual. 2nd ed. Westchester, Illinois (US): American Academy of Sleep Medicine; 2005.

Arendt J, Stone B, Skene DJ. Sleep disruption in jet lag and other circadian rhythm-related disorders. In: Principles and Practice of Sleep Medicine. Kryger MJ, Roth T, Dement WC, editors. Philadelphia (US): Elsevier Saunders; 2005. p. 659-672.

Cajochen C, Zeitzer JM, Czeisler CA, Dijk DJ. Dose-response relationship for light intensity and ocular and electroencephalographic correlates of human alertness. Behav Brain Res. 2000; 115: 75-83.

Czeisler CA, Allan JS, Strogatz SH, Ronda JM, Sánchez R, Ríos CD et al. Bright light resets the human circadian pacemaker independent of the timing of the sleep-wake cycle. Science. 1986 Aug 8; 233(4764): 667-671.

Houpt TA, Boulos Z, Moore-Ede MC. MidnightSun: Software for determining light exposure and phase-shifting schedules during global travel. Physiol Behav. 1996; 59: 561-568.

Lewy AJ, Bauer VK, Ahmed S, Thomas KH, Cutler NL, Singer CM et al. The human phase response curve (PRC) to melatonin is about 12 hours out of phase with the PRC to light. Chronobiol Int. 1998 Jan; 15(1): 71-83.

Lockley SW, Evans EE, Scheer FAJL, Brainard GC, Czeisler CA, Aeschbach D. Short-wavelength sensitivity for the direct effects of light on alertness, vigilance, and the waking electroencephalogram in humans. Sleep. 2006; 29: 161-168.

Morgenthaler TI, Lee-Chiong T, Alessi C, Friedman L, Aurora RN, Boehlecke B et al.; Standards of Practice Committee of the American Academy of Sleep Medicine. Practice parameters for the clinical evaluation and treatment of circadian rhythm sleep disorders. An American Academy of Sleep Medicine report. Sleep. 2007 Nov 1; 30(11): 1445-1459.

Reid KJ, Chang AM, Zee PC. Circadian rhythm sleep disorders. Med Clin North Am. 2004; 88: 631-651.

Sack RL, Lewy AJ, Hughes RJ. Use of melatonin for sleep and circadian rhythm disorders. Ann Med. 1998 Feb; 30(1): 115-21. Review.

Terman M. Light Therapy. In: Principles and Practice of Sleep Medicine. Kryger MJ, Roth T, Dement WC, editors. Philadelphia (US): Elsevier Saunders; 2005. p.1424-1442.

Abstract: This article will discuss migraines, tension type headaches, medication overuse and touch upon red flags for important secondary headache disorders.

Introduction
There are many causes of headache which have been described in the medical literature and classified in The International Classification of Headache Disorders published by the International Headache Society. For practical clinical purposes, all headaches can be classified as one of the primary headache syndromes or as a headaches secondary to an underlying disease process or medical condition. This discussion will focus on the common primary headache syndromes – migraines, tension type and medication overuse headaches. Red flags for important secondary disorders will be highlighted.

Epidemiology
Migraines are common, affecting approximately 5% of males and 15% of females in population surveys around the world. Migraines may begin in childhood or early adulthood, reaching peak prevalence in middle-aged women. Migraines rarely occur for the first time after the age of 50 years.

Aetiology
The pathophysiology of migraine and tension type headaches is not well understood. In migraines it is believed that the brain is “hyper-reactive” to triggers such as stress, leading to the generation of a wave of cortical excitation. This is turn activates the trigeminovascular system, causing release of neurotransmitters, vasodilatation and neurogenic inflammation. Nociceptive impulses travel via the trigeminal nerve to the brainstem pain centres, producing headaches. At least some forms of migraines may be genetically determined (eg. familial hemiplegic migraine).

Clinical Features & Differential Diagnosis

Migraines have been described as “sick” headaches, comprising episodic headaches often unilateral, associated variably with nausea, vomiting, light and noise sensitivity, and dizziness. Often there may be a prodrome in the hours or days preceding a headache, though not always apparent, perhaps comprising fatigue, nausea, depression or food cravings. One in eight migraineurs experiences auras, usually visual (eg. flashes of light, zigzag lines, visual blurring) or somatosensory (eg. numbness, tingling in the face or limbs). Motor phenomena such as hemiplegia are rare. Prominent negative visual symptoms (ie. loss or obscuration) should raise an index of suspicion of mimics such as amaurosis fugax (transient monocular blindness from carotid atherosclerosis) or optic nerve ischaemia from giant cell arteritis. Transient sensory or motor deficits, especially when auras occur without headaches (“aura without headache”), should raise the possibility of transient ischaemic attack, stroke or seizures. These red flags should be heeded in patients who are atypical in presentation (eg. older patients with cardiovascular or epilepsy risk factors), especially if they do not have a past history of recurring headaches with similar accompanying neurological phenomena. Such patients may require further evaluation, including specialist referral for tertiary care.Headaches may last hours or days, and the mean frequency in population studies is about once a month. Migraine triggers include psychological, hormonal, environmental and dietary factors. Common factors include stress (often release from stress in the weekends) and sleep (often too little sleep, or sometimes oversleeping). Some describe an association with menses (menstrual migraine), or changes in pregnancy (eg. may improve in the second and third trimesters) and menopause (eg. worsening during the perimenopausal period). Environmental factors such as cigarette smoke and strong perfume scents can trigger an attack. Certain foods, alcohol, caffeine withdrawal and missing meals can all set off a migraine attack. Migraines which conform to the International Headache Society criteria for migraine diagnosis and which occur regularly should be straightforward, eg. unilateral, throbbing headaches associated with nausea/vomiting, photophobia and/or sonophobia, perhaps worsened by physical activity, with or without auras. A red flag for secondary headaches is the migraine patient presenting with “the worse headache ever”, usually presenting to the A&E department. Up to 12% of patients presenting in this way have a subarachnoid haemorrhage. Neuroimaging should be considered in every patient describing atypical headaches which do not conform to their usual headache pattern, which are progressive, unresponsive to the usual medications, unusually severe, related to head injury or which are associated with neurological deficits. Besides intracranial haemorrhage, transient ischaemic attack/stroke, seizures and giant cell arteritis, other important diagnostic considerations in patients presenting with headaches are hypertension, meningitis, raised intracranial pressure (eg. brain tumour, pseudotumour cerebri) and local ocular conditions like glaucoma. All patient should have vital signs routinely checked (ie. including temperature, blood pressure) and a thorough neurological examination for signs of meningism (eg. neck stiffness), papilloedema and focal deficits. Neurological or ophthalmology consultation may be required in complex cases. Another important consideration is medication overuse headaches, which should be considered in patients who have very frequent headaches and take analgesics, ergots or triptans very frequently.

Investigations
Primary headaches can often be diagnosed on clinical grounds. In older patients presenting with new headaches after the age of 50 years, especially those with a history of temporal headaches and visual symptoms (eg, transient visual blurring, diplopia, eye pain, sudden loss of vision), giant cell arteritis should be excluded. ESR elevation (moderate to >100 mm/h) is common and is rarely (~ 3%) normal. Other acute phase reactants like C-reactive protein (CRP) may also be elevated. The ESR can be followed serially in monitoring for treatment response.

Neuroimaging is indicated when intracranial haemorrhage, ischaemic stroke or space-occupying mass lesions are suspected. CT scans are useful in the detection of fresh blood and skull fractures. MRI is the preferred modality of imaging for acute stroke and more detailed delineation of the brain parenchyma and mass lesions in the brain, particularly in the posterior fossa which is not well visualized on routine CT scanning of the brain. Contrast enhancement is indicated to better outline infection and neoplasm, if not contraindicated.
Lumbar puncture is indicated if there are no focal mass lesions on neuroimaging or clinical evidence to suggest raised intracranial pressure, to evaluate suspected CNS infection or malignancy and to document opening pressures. Both low and high CSF pressures can cause headaches.

Treatment

Treatment of headache depends on the cause, many will require medication; however, other management methods also may be useful – such as getting enough sleep and managing underlying psychological stress. Secondary headache usually resolves when the underlying problem is treated.

Migraine
Many patients will require medication to abort acute headaches, those with frequent migraines may benefit daily preventive medication. Educating the migraine patient to recognize and avoid headache triggers helps to reduce the frequency of attacks. Common migraine triggers include bright lights, stress, and skipping meals. Migraine sufferers generally improve if they have regular eating and sleeping patterns.

Preventive Medications
Daily prophylactic medication may help if migraine attacks occur several times a month. Commonly used drugs for migraine prophylaxis include: beta blockers, calcium channel blockers, antidepressants, and antiepileptic drugs. The beta blockers propranolol, and the anticonvulsants sodium valproate and topiramate, are approved by the U.S. Food and Drug Administration (FDA) for migraine prevention. Other anticonvulsants such as gabapentin and pregabalin are also useful for neuropathic pain. They have been used off label clinically for migraine prevention. Botulinum toxin injected into the scalp muscles has also been found to decrease the frequency and severity of migraine.

Abortive Medications
For mild attacks, over-the-counter analgesics (especially those containing caffeine) can be used. A rapid-acting NSAID such as ibuprofen, or naproxen sodium can also be helpful in mild attacks. The triptan drugs are agonists that affect the serotonin receptors located in neurons and cerebral vessels. Several triptan agents are currently available, including: sumatriptan succinate (oral, nasal, and injectable), zolmitriptan (oral and nasal), and eletriptan (oral). All are well tolerated by patients; however, like ergotamine, the triptans should be avoided in patients with coronary artery disease, peripheral vascular disease, or severe hypertension. Triptan drugs are effective in 65% to 70% of patients, relieving migraine pain and associated symptoms within 2 hours of administration. The earlier they are taken acutely, the more effective they can be. Side effects are generally mild, including dizziness, sedation, or mild chest tightness. Triptans and other abortive agents are more effective when used early in the migraine attack. Combined use of an NSAID with a triptan can offer better headache relief and may be associated with less-frequent recurrence of the migraine. Oral abortive agents, especially the ergotamines, often cause nausea and vomiting, which can limit their use. However, taking metoclopramide before taking an abortive agent can control the nausea and enhance the effectiveness of the abortive drug.

Tension type Headache
Tension type headaches may be relieved by sleep, and stress management. Episodic tension-type headache can be managed with analgesics such as aspirin, acetaminophen, and NSAIDs. Chronic tension-type headache may be complicated by overuse of analgesics (medication overuse headaches). Treatment of medication overuse headaches will entail discontinuation of painkillers for a time. Patients suffering from chronic tension-type headache may require psychological therapy to manage longstanding psychosocial stressors.Tricyclic antidepressants are effective for the treatment of chronic tension-type headaches is the. Amitriptyline, doxepin and nortriptyline are helpful, with amitriptyline being the most commonly used. They are usually taken at bedtime because of their sedating effects. Side effects include daytime sedation, weight gain and dry mouth. Selective serotonin reuptake inhibitors, such as fluoxetine, sertraline, paroxetine, and citalopram, are better tolerated and have fewer side effects than the tricyclic antidepressants, and can be especially effective in easing headache if there is an underlying depression. Venlafaxine and duloxetine are both serotonin and norepinephrine uptake inhibitors are generally helpful in chronic pain conditions, including headache. Muscle relaxants such as orphenadrine citrate, can be helpful if there is increased muscle spasm or muscle tension present. Anarex is a combination of acetaminophen and orphenadrine citrate, commonly used for the treatment of tension type headaches.

What is Dementia?
Dementia refers to a condition in which there is gradual loss of memory and decline in thinking ability, typically progressive, initially affecting higher levels of functioning and eventually severe enough to impair many activities of daily living. Alzheimer’s disease (AD) is the most common form of dementia, accounting for 60-80% of all cases. The second most form of dementia known as vascular dementia, a condition in which multiple strokes affect different parts of the brain over time, eventually leading to a state of impaired brain function which looks very much like AD.

What Other Conditions Mimic Dementia?
Other important causes of memory and thinking disturbance which can look very much like dementia include depression, metabolic disturbances and infections (such as sexually transmitted diseases like HIV or syphilis). Examples of metabolic abnormalities are high calcium and low sugar (glucose) levels, deficiency in thyroid hormones or vitamin B12 which can all impair brain function. As such, simple tests of mineral, hormone and vitamin levels are a routine part of the neurologist’s assessment of patients who present with memory and thinking problems.

A more recent development in medical science is the discovery that vitamin D deficiency is associated with an increased risk in dementia later in life. Research participants who had deficient vitamin D levels have been shown to have greater decline in the ability to think, organize thoughts, make decisions and plan ahead than those who had normal vitamin D levels. While the role of vitamin D in the brain is not well understood, such research findings suggest that low vitamin D levels are harmful to the brain, and maintaining normal levels may actually be beneficial.

Depression is the other common condition which can mimic dementia. Depressed people may appear to have poor memory and thinking when they are actually experiencing low mood and energy, and have concentrating difficulty. The lack of interest in others, apathy, inattentiveness and general slowing down may suggest a disorder of the brain like dementia, when the real underlying cause is psychological.

The assessment of dementia, and making the diagnosis of AD is therefore best done by a doctor who is familiar with the various possible causes of dementia and its mimics. Typically there is a detailed history taking during which all the past medical problems are evaluated, followed by a physical examination, and a bedside test of memory and thinking function. Often some blood tests are done to rule out treatable biochemical problems or infections. A brain scan (usually an MRI scan) is helpful because it can help to assess if there are any structural problems which have damaged the brain – such as multiple strokes, and much less commonly, brain tumours, shrinkage (“atrophy”) of the brain, or abnormal fluid accumulation in the brain (“hydrocephalus”). PET scanning is a form of brain imaging in which the sugar uptake in the brain can be determined, giving very useful diagnostic information in people who have early dementia – when low sugar uptake in the brain may be the only sign of loss of brain cell function, and the MRI and blood tests appear normal.

Signs of Alzheimer’s Disease

Some early signs of AD include:

• Poor memory for recent events: While the recall of remote events appears intact, short term memory is poor. This includes eg. someone forgetting his appointments, where he has left keys, what he had for lunch, or asking the same questions over and over, and increasing reliance on memory aids
• Difficulty handling everyday matters which used to be routine: Such as keeping track of monthly bills, driving to familiar places, operating technical devices like a video recorder, taking a longer time getting errands done because of difficulty concentrating, or needing more help than before.
• Confusion about time and location – losing track of the dates and the passage of time, or feeling disoriented and easily lost especially in unfamiliar places.
• Difficulty with following a conversation, and sometimes difficulty finding words.
• Poor judgment: Such as making poor decisions with money.
• Inattention to grooming or reduced levels of personal hygiene.
• Social withdrawal: Becoming less engaged with others, decreased participation in sports, social activities, hobbies or work projects.
• Mood and personality changes: Becoming more fearful, confused, anxious or depressed; more easily upset especially when in unfamiliar settings.

The occasional memory lapse (eg. forgetting someone’s name, missing a monthly credit card payment), a poor judgment call or losing possessions once in awhile can occur in normal persons, especially as we get older. By comparison, in patients who suffer actual dementia like AD will experience progressively serious disability which eventually significantly impairs their ability to function at whatever level they were previously accustomed to, such as the inability to utilize information to judge and make a sensible decision, plan and execute a monthly budget, keep track of time and events in everyday life, or follow and participate in conversations.

Diagnosis & Treatment
Diagnosis of AD is made based on the typical symptoms and signs, and findings on brain imaging, as well as the exclusion of the various conditions mentioned above which can cause similar problems of memory and thinking.

AD is a result of progressive brain cell death and loss of the connections between brain cells, for which there is no cure. However there are drugs which can be used to help slow down the progressive loss of brain function and enhance the ability of the brain to carry messages by altering the biochemical environment. For example, one class of drug commonly used for AD is the cholinesterase inhibitors – which prevent the breakdown of the acetylcholine, a chemical messenger important for memory and learning. Another drug known as memantine alters the activity of a different messenger chemical involved in learning and memory known as glutamate. These drugs are known to delay the worsening of brain function in AD.

Besides these dementia specific drugs, it is also common practice to correct any identified biochemical deficiency (such as vitamin B12 or vitamin D deficiency).

Conclusion
AD is a slowly progressive condition for while early diagnosis and symptomatic treatment is available. A comprehensive evaluation will identify any easily treatable associated conditions. Drug treatments are available for AD which can help slow down the symptoms of brain decline. Earlier diagnosis allows for a better chance to benefit from treatment, the opportunity to plan ahead and make important decisions (eg. relating to advance medical directives, financial and legal matters) before severe mental deterioration has set in, and to explore options for the management of this chronic disabling condition which can reduce anxiety for the patient and caregivers.