Review

Nature Clinical Practice Neurology (2008) 4, 254-266
doi:10.1038/ncpneuro0775  
Received 5 November 2007 | Accepted 4 February 2008 | Published online: 8 April 2008

Sleep disturbances in patients with parkinsonism

Valérie Cochen De Cock, Marie Vidailhet and Isabelle Arnulf*  About the authors

Correspondence *Fédération des Pathologies du Sommeil, Hôpital Pitié-Salpêtrière, 47–83 Boulevard de l'Hôpital, 75651 Paris Cedex 13, France

Email
 isabelle.arnulf@psl.aphp.fr

Summary

Altered sleep and vigilance are among the most frequent symptoms, besides parkinsonism, in movement disorders. As many as 60% of patients with Parkinson's disease (PD) experience insomnia, 15–59% show rapid eye movement (REM) sleep behavior disorders (RBDs), and 30% show excessive daytime sleepiness. Insomnia is a distressing difficulty to maintain sleep, which is exacerbated by motor disability, painful dystonia, restless legs, dysuria, anxiety and depressed mood. Improving night-time motor control by overnight treatment with levodopa, transdermal or long-acting dopamine agonists, or bilateral subthalamus stimulation, can improve sleep continuity. RBDs are violent, enacted dreams that expose the patient or their sleeping partner to night-time injuries. A striking improvement of parkinsonism is observed during these behaviors in PD. RBDs are thought to be caused by lesions in the REM sleep atonia system, and can, in association with other early markers of neurodegenerative diseases, such as olfactory, cognitive and autonomic disturbances, precede parkinsonism by several years. Daytime sleepiness, often with a narcolepsy-like phenotype, is a common occurrence in PD, owing to lesions in the arousal systems of the brain. The use of dopamine agonists increases the risk of sleep attacks, especially when driving, suggesting a drug–disease interaction.

Review criteria

We performed a search of the PubMed database for articles on sleep disturbances in patients with movements disorders published between 1970 and October 2007. We used combinations of the following search terms: "sleep", "REM sleep", "daytime sleepiness", "RBD", "non-motor symptoms", "RLS", "PLMS", "Parkinson's disease", "parkinsonism", "multiple system atrophy" and "progressive supranuclear palsy". We selected 86 articles that we judged to be of high impact on the field. With two exceptions, articles not written in English were excluded.

Top

Introduction

Parkinsonism, which is defined by the association of slow movements, difficulty walking, muscle rigidity, and sometimes tremor, characterizes a group of neurodegenerative movement disorders that affect 1–2% of adults over 65 years of age. Idiopathic Parkinson's disease (PD) is the most common cause of parkinsonism, and dopaminergic agents provide important benefits in terms of relief of motor symptoms. By contrast, 5–10% of cases of 'atypical' parkinsonism are poorly levodopa-sensitive and include lesions in other motor systems ('Parkinson-plus' syndromes). For example, progressive supranuclear palsy (PSP) is characterized by parkinsonism, abnormal eye movement, and postural instability including falls in the early stages of the disease, and multiple system atrophy (MSA) is characterized by parkinsonism plus cerebellar or autonomic dysfunction.

Major advances have been made in the treatment of PD during the past few decades, including fine control of dopamine substitution, and functional neurosurgery. These advances have, however, unmasked nondopaminergic and extranigral symptoms (cognitive, mood, psychiatric, sleep and vegetative disturbances) that are not alleviated or only poorly alleviated by these treatments. Sleep disturbances are frequent symptoms that have recently been recognized in PD and other movement disorders. The disturbances include insomnia, abnormal movements during sleep (e.g. periodic leg movements, rapid eye movement [REM] sleep behavior disorders [RBDs]) and excessive daytime sleepiness. Insomnia has a serious impact on a patient's quality of life. RBDs are enacted dreams that can cause patient or bed-partner injuries. Daytime sleepiness exposes the patients to risk of a driving accident. The mechanisms of these sleep disturbances are not fully known, but they include a clear drug–disease interaction.

Top

Insomnia

Insomnia is defined as a recurrent difficulty to fall asleep or stay asleep, with daytime psychological or cognitive consequences.1 As there are large between-individual differences in sleep needs, there is no standard definition for insomnia on the basis of time asleep or number of awakenings, as determined by sleep or rest–activity monitoring. A community-based survey determined that 60.3% of patients with PD had a sleep problem, a substantially higher proportion than in patients with diabetes mellitus (45%), or in elderly controls (33%).2 There was no difference between groups regarding difficulty falling asleep, but frequent (38.9%) or early (23.4%) awakenings were twice as frequent in PD as in other groups, and felt more distressing to patients with PD than to other groups. In hospital surveys, 76% of patients with PD complained of 'broken sleep', and 18% complained of poor sleep.3 As many as 52.5% of patients with MSA have complained of sleep fragmentation, compared with 38.7% patients with PD.4 The most severe and specific insomnia is, however, observed in patients with PSP.5

Night-time motor problems as a cause of insomnia

Most patients with PD report two to five awakenings during the night (twice as many as are experienced by controls),6 with the patient being awake for 30–40% of the night (Figure 1).7, 8 Sleep monitoring does not provide additional information on the cause of insomnia in patients with parkinsonism, except for a specific, severe decrease in REM sleep in patients with PSP.5 Insomnia is a nonspecific symptom that does not imply the presence of a selective lesion in any sleep system, except in the case of PSP, in which cholinergic lesions in the brainstem can encompass the REM-sleep executive systems. Rather, any condition that increases arousal can cause insomnia. The frequency of insomnia increases with advanced motor stages of PD and a need for a higher daily dose of dopaminergic therapy, an indicator of dopamine denervation. Indeed, slowed movements during the night, difficulties turning in bed or adjusting blankets, pain, cramps, nocturnal and early morning dystonia, and frequent need to pass urine, are reported by patients with advanced PD as the main causes of their insomnia.3 The PD Sleep Scale (Box 1) can be helpful for identifying the causes of insomnia.9 In patients with advanced PD, the alleviation of nocturnal akinesia and dystonia using subthalamic nucleus stimulation at night resulted in a substantial decrease in time awake at night,10 and led to long-term improvement of insomnia in spite of a major reduction in drug dose.11, 12

Figure 1 Sleep histograms from patients with parkinsonism.
Figure 1 : Sleep histograms from patients with parkinsonism. Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, or to obtain a text description, please contact npg@nature.com

(A) Healthy 60-year-old woman with normal sleep. (B) 54-year-old woman with Parkinson's disease (PD) treated with levodopa 300 mg/day and bromocriptine 30 mg/day, who reported frequent night-time awakenings, daytime sleepiness, and hallucinations (arrow indicates a hallucination in which she saw a stranger in the room) that were synchronous with abnormal irruption of rapid eye movement (REM) sleep during the daytime (narcolepsy-like phenotype). (C) 80-year-old man with mild PD, complaining of falling asleep 2–3 hours after each levodopa intake (arrows), who displayed severe hypersomnia. (D) 62-year-old woman with de novo untreated PD, who complained of disturbing night-time hallucinations (arrow indicates a hallucination in which she saw her daughter in the hospital room) that emerged from REM sleep. The x-axis displays the time of night and day (clock hours), the y-axis displays the stages of sleep and wakefulness, with awakening (A), non-REM sleep stage 1 (1), stage 2 (2), stage 3 (3) and stage 4 (4), and REM sleep (R).

Full figure and legend (39K)Figures & Tables indexDownload PowerPoint slide (136K)

Box 1 Parkinson's Disease Sleep Scale.a

Sleep disturbances in patients with parkinsonism

APermission obtained from the BMJ Publishing Group Ltd © Chaudhuri KR et al. (2002) J Neurol Neurosurg Psychiatry 73: 629–635. bThe adjectives are placed at each extremity of a 10 cm visual-analog scale. The patient places a cross mark on this line.

Restless legs syndrome (RLS) is more prevalent (12–20.8%) in PD than in the general population.13, 14 It is mostly of mild severity, but it worsens insomnia. PD patients with RLS seem to have lower ferritin levels than those without.13 Poor sleep quality also correlates with depression and anxiety scores.15 Dopaminergic agents, which have long been considered to be stimulants, have been shown to have a deleterious effect on sleep when given at bedtime.16

Treatment of insomnia and nocturnal motor symptoms

Despite the frequency of insomnia in PD, treatments for this condition have been poorly studied in this context. There have, however, been a few trials of sleep-inducing drugs, and of treatments aimed at improving nocturnal movements. High doses of melatonin (50 mg) provide a modest beneficial effect in PD insomnia, and are very well tolerated.17 Single, small, nonblinded trials of zolpidem,18 clozapine for nocturnal akathisia,19 and quetiapine have shown a beneficial effect on sleep in patients with PD.20 Hypnotics and small doses of sedative antidepressants are frequently used in clinical PD practice. Most antidepressants carry a risk of worsening pre-existing RLS, periodic leg movements and RBDs, although bupropion, which lacks these side effects, can be an appropriate first-line treatment for insomnia.

Re-establishing continuous dopaminergic stimulation during the day and night can be a strategy to improve night-time motor disability. The benefits of dopaminergic agents in the evening and night-time need, however, to be weighed against their alerting effect. For example, pergolide 1 mg in the evening worsened sleep efficiency and fragmentation in patients with repeated nocturnal awakenings caused by PD symptoms.16 By contrast, the use of transdermal rotigotine during day and night improved most aspects of sleep and night quality in patients with advanced PD, compared with placebo, but it was not superior to the use of oral pramipexole three times a day.21 The use of levodopa 200 mg in the evening improves subjective sleep quality, and decreases night-time movements.22 Sustained-release forms of levodopa given in the evening have not yet been compared with normal-release forms of levodopa. Sustained-release levodopa does not change subjective sleep (general quality, sleep onset latency, total sleep time, number of awakenings), but it has a mild effect on nocturnal akinesia in small, nonblinded trials. By contrast, subthalamic nucleus stimulation consistently improves sleep duration and reduces nocturnal awakenings in patients with advanced PD.10, 11, 12 This benefit might be attributable to the specific effect of stimulation on a single dopaminergic loop, whereas oral or transdermal dopaminergic therapy might, in addition, stimulate arousal systems.23

The treatment of nocturnal RLS in PD is likely to be complex, and could include iron supplementation in cases of deficiency, evening use of gabapentine and derivatives, and, more rarely, use of opiates (in patients without dementia or hallucinations).14 No guidelines are yet available for treating this condition.

Future prospects

Despite its frequency, insomnia in PD has not been a frequently investigated topic, either in terms of understanding its multifactorial mechanisms or finding an adequate therapy. Studying sleep abnormalities in various animal models of PD should help to fill this gap, including attempts to determine the respective contribution of dopaminergic and nondopaminergic lesions, changes in the circadian clock, motor symptoms (e.g. akinesia, dysuria), and nonmotor symptoms (e.g. depression, anxiety).

Whether the relative reduction of dopamine stimulation during the night is a good thing (i.e. would help to eliminate inactive metabolites) or a bad thing (i.e. would promote dopamine receptor desensitization) for patients is an open question. The pathophysiology of RLS in PD presents a similar dilemma. RLS might either correspond to a deficit of dopamine stimulation at night (and would benefit from an evening additional dose of dopamine agonist) or, by contrast, to an excess of dopamine stimulation during the daytime with rebound restless legs during the night, a condition named 'augmentation' (in which case it would benefit from decreasing the daily dopamine dose). The occurrence of restless legs years after (and not before) the onset of parkinsonism supports the latter hypothesis. The emergence of RLS (and periodic leg movements) during subthalamic stimulation suggests that these phenomena are not controlled by the basal ganglia.10, 24 The development of 24-hour transdermal forms of dopamine agonist, 24-hour sustained release dopamine agonists, and continuous subthalamic stimulation should shed light on these phenomena.

Top

Rapid eye movement sleep behavior disorders

RBDs are dream-enacting behaviors—for example, laughing, talking, shouting, kicking, or fighting invisible enemies—during sleep.1 Normal REM sleep is associated with an active blockade of motor output (a transient paralysis termed muscle atonia), which is unblocked during RBD, allowing prominent motor activity. RBD behaviors can be violent enough to disrupt sleep and cause injuries to the patient or their sleeping partner, but RBD does not induce daytime sleepiness. Sleep monitoring demonstrates enhanced tonic muscle activity (REM sleep without atonia) and increased phasic muscle twitches during REM sleep in patients with RBD.1

Association with synucleopathies

RBD can occur without concomitant medical disorders ('idiopathic' RBD), but is mostly reported in patients with neurodegenerative disorders (Table 1). RBD affects 30–90% of patients with synucleopathies (conditions characterized by aggregation of alpha-synuclein, forming Lewy bodies within neurons), but is rare in tauopathies.25 However, in a toxic form of parkinsonism with tau protein deposits, observed in the French West Indies (and possibly caused by eating soursop, a tropical fruit), we found that 7 out of 10 patients had severe RBD.26 In addition, RBD might be triggered or exacerbated by antidepressants.27 Most,28, 29 but not all,30 series of RBD patients with parkinsonism contain more men than women, a ratio that is consistently observed in idiopathic RBD.31 Whether men experience or enact more violent dreams than women, or whether wives are more attentive to nocturnal patient symptoms than husbands, is unknown.

Table 1 Prevalence of rapid eye movement sleep behavior disorders and rapid eye movement sleep without atonia in neurodegenerative diseases.
Table 1 - Prevalence of rapid eye movement sleep behavior disorders and rapid eye movement sleep without atonia in neurodegenerative diseases.
Full tableFigures & Tables indexDownload PowerPoint slide (110K)

Complex behaviors during dreams suggest cortical involvement

The frequency of RBD activity varies between individuals and from night to night in the same individual. One third of patients with PD have at least one episode per week.29, 30 Typical behaviors during RBD include talking, laughing, yelling, gesturing, punching and kicking, all conditions that can be disturbing and dangerous for a sleeping partner (Figure 2 and Supplementary Movie 1 online). Compared with patients with idiopathic RBD, who are usually selectively referred for violent behaviors, systematic investigations of sleep have revealed that the RBD behaviors are less frequent and violent in parkinsonism.28 Indeed, we also observed quiet behaviors, including drinking soup, singing a song, cutting beans or giving a lecture, which were mentioned incidentally by the bed partner as being nondisturbing.26, 30 Interestingly, the complexity of these behaviors and the presence of intelligible speech suggest that not only the brainstem, but also the cortex (and especially Broca's area), are activated during RBD.30

Figure 2 Rapid eye movement sleep behavior disorder.
Figure 2 : Rapid eye movement sleep behavior disorder. Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, or to obtain a text description, please contact npg@nature.com

Abnormal, violent behavior during rapid eye movement sleep behavior disorder in a 68-year-old man with advanced Parkinson's disease. The patient dreams that he is arguing and fighting with a stranger. He has imperfect muscle atonia during rapid eye movement sleep and is therefore able to move, but he does not stand up. Note that his movements are extremely fast and coordinated, and his face is expressive and does not show parkinsonism, despite is the fact that he has been weaned from levodopa for 12 hours.30 See also Supplementary Movie 1 online.

Full figure and legend (14K)Figures & Tables indexDownload PowerPoint slide (77K)

Arm and leg movements, speech, laughter and tears are frequent in PD patients with RBD, but behaviors that are common in sleepwalking1 (a condition associated with non-REM sleep), such as standing up, walking and appropriate use of objects in the vicinity, are rare.29 The movements during RBD are purposeful, nonstereotyped and complex, in contrast to nocturnal frontal lobe epilepsy and periodic leg movements. Some patients might, however, display gross, startle-like body jerks that resemble increased normal REM sleep twitching or aborted complex behaviors, but do not share the focal location, distribution and propagation of myoclonus.30

In addition to enacted dreams, patients with RBD experience abnormal dreaming phenomena. Compared with dreams of matched controls, patients' dreams contain more aggressive content (the dreamer being an aggressor, despite normal and even decreased levels of daytime aggressiveness), more frequent presence of animals, and less sexual content.32 Interestingly, the dreamers are usually not the primary aggressors during these nightmares, but rather try, by fighting back, to protect themselves or their loved ones from an external aggressor.

Extranigral, nondopaminergic pathology

The exact cause of RBD in parkinsonism is largely unknown, but a nondopaminergic lesion of the system controlling atonia during REM sleep has been strongly implicated (Figure 3).33 A cat model of RBD has been generated by creating lesions of the perilocus ceruleus alpha in the pons. The animals displayed complex behaviors during REM sleep, including grooming, licking, being on watch, and hunting invisible prey.34 Lesions of the pathway descending from this nucleus through the pons and the medial medulla (at the level of the magnocellularis nucleus) also result in REM sleep without atonia and with complex movements.34, 35 Muscle atonia is imperfect in rats after lesions are created in the sublaterodorsalis nucleus.36, 37 The equivalent of these nuclei in humans is the subceruleus nucleus in the pons. In two PD patients with RBD and hallucinations, this nucleus was found to contain Lewy bodies.38 Interestingly, the subceruleus–ceruleus complex is affected from the early stages of PD.39 RBD persists during subthalamic stimulation, suggesting that its pathophysiology does not involve the basal ganglia loop.

Figure 3 Brain pathology in rapid eye movement sleep behavior disorders.
Figure 3 : Brain pathology in rapid eye movement sleep behavior disorders. Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, or to obtain a text description, please contact npg@nature.com

(A) gamma-Aminobutyric-acid (GABA)-immunoreactive neurons in the rostral part of the rat SLD).36 The SLD is equivalent to the cat perilocus ceruleus alpha (peri-LCalpha). SLD blockade with GABA antagonists induces muscle atonia and a rapid eye movement (REM)-sleep-like state. Lesioning of the ventral SLD causes REM sleep episodes without atonia.37 Permission obtained from Federation of European Neuroscience Societies © Boissard R et al. (2003) Eur J Neurosci 18: 1627–1639. (B) Frontal section at the P4 level of the cat pons. Bilateral coagulation of or injections of ibotenic acid into the peri-LCalpha (red circles) induced REM sleep without atonia and REM sleep behavior disorders (RBDs).34 (C) Schematized section cut perpendicular to Meynert's brainstem axis in humans, showing the distribution pattern of Parkinson's disease (PD)-related pathology in the lower brainstem.39 Early neuronal loss and Lewy body pathology are observed in the lower brainstem, possibly causing the RBDs that precede PD onset. The blue shading in the ceruleus–subceruleus complex indicates that 17 out of 30 incidental cases had PD-related pathology in this area.39 Permission obtained from American Association of Neuropathologists, Inc. © Del Tredici K et al. (2002) J Neuropathol Exp Neurol 61: 413–426. (D) Lewy bodies stained with anti-alpha-synuclein antibodies, observed in large numbers in melanized and nonmelanized neurons in the subceruleus nucleus of a 69-year-old man with idiopathic PD, RBD, and sudden daytime REM sleep episodes temporally associated with severe visual hallucinations. PD-related pathology was sparse in the upper brainstem and cortex.38 Abbrevations: BC, brachium conjunctivum; CS, ceruleus–subceruleus complex; DTN, dorsal tegmentum nucleus; LDT, laterodorsal tegmentum nucleus; MEV, mesencephalic trigeminal nucleus; MLE, medial lemniscus; MLF, medial longitudinal fascicle; MPB, medial parabrachial nucleus; PNC, central pontine nucleus; PPN, pedonculopontine nucleus; PRNr, rostral pontine reticular nucleus; SCP, superior cerebellar peduncle; SLD, sublaterodorsal nucleus; V4, fourth ventricle; VTN, ventral tegmental nucleus.

Full figure and legend (58K)Figures & Tables indexDownload PowerPoint slide (141K)

Safety measures and treatment

Patients with RBD can adopt various measures to safeguard their bedroom environment and partner, including use of twin beds, placing the mattress directly on the floor, having a wall light instead of a bedside lamp, and having no night table. In severe cases, antidepressants might need to be reduced or withdrawn to alleviate RBD. No randomized, double-blind, placebo controlled study has been reported for any drug treatment for RBD, although clonazepam (0.5–2 mg before bedtime) is highly effective and well tolerated in most cases.27, 40 Melatonin, which is also effective in RBD, partially restores normal muscle tone during REM sleep.41 Dopaminergic agents, on the other hand, have no effect on the expression of RBD in patients with PD.28, 42

An early marker of neurodegenerative disorders

RBD can precede, by several years, the first symptoms of parkinsonism or dementia. In a series of 29 patients with idiopathic RBD, 11 (39%) developed parkinsonism 5 years later and 19 (65%) developed parkinsonism 7 years later.43, 44 In 44 consecutive patients followed for 2–15 years after a diagnosis of idiopathic RBD, nine had developed Parkinson's disease, six had developed dementia with Lewy bodies, one had developed MSA, and four had developed mild cognitive impairment.45 Functional imaging has shown that patients with idiopathic RBD have reduced numbers of striatal dopaminergic transporters.46 Retrospective studies showed that RBD preceded parkinsonism by a median of 3–11 years in 27–50% patients with Parkinson's disease.29, 30, 42

The association between idiopathic RBD and subsequent neurodegenerative disease has prompted research into other early markers of neurodegenerative disease in patients with idiopathic RBD, in an attempt to identify a subgroup of patients who show increased vulnerability to parkinsonism or dementia. It was found that patients with these conditions display more-frequent visuospatial constructional dysfunction,47 autonomic dysfunction,48 olfactory impairment49 and slowed electroencephalographic (EEG) activity50 than do controls. The presence of idiopathic RBD, especially when associated with one of these signs or symptoms, is, therefore, highly predictive of emerging neurodegenerative disease. The pattern of dysfunction suggests an early, patchy loss or impairment of both dopaminergic and nondopaminergic neurons.

Several ideas for future investigations arise from these recent findings. For example, trials lasting 5–10 years could be set up of neuroprotective agents in patients with idiopathic RBD. Such trials might help to solve the ethical dilemma of whether to disclose the high risk of emergent neurodegenerative disease to otherwise healthy individuals: it is easier to tell a 55-year-old patient that he has a 70% risk of developing PD or dementia within 5–10 years if, as a doctor, you can suggest that he takes a neuroprotective agent in a trial. So far, we know of no drugs that would definitely be effective, but some drugs (e.g. melatonin, modafinil) are potential candidates. Cohorts of patients with idiopathic RBD could also be followed to provide insights into preclinical parkinsonism or dementia.

A marker of severity in parkinsonism?

The question of whether RBD represents a 'red flag' for developing other disturbances in PD has only recently been investigated. Interestingly, PD patients with RBDs have a threefold higher frequency of hallucinations than do PD patients without RBDs.51 As hallucinations in PD might also correspond to wakeful dreams (associated with abnormal irruption of REM sleep during daytime),38 it is possible to envisage a common, wider dysfunction of phasic and tonic REM sleep mechanisms in PD. There were no differences in motor disability, simple cognitive tests, depression score and daytime sleepiness between patients with and without RBD.30 In a small group of PD patients without dementia, however, patients with RBD had lower EEG frequencies than did patients without RBD, suggesting either 'bottom-up' consequences of brainstem lesions or early loss of cortical neurons in patients with RBD.52 The question of whether these isolated EEG findings are associated with early cognitive impairment in PD patients with RBD, in transversal or longitudinal cohort studies, is a current area of investigation.53

Top

Disappearance of parkinsonism during enacted dreams

Interestingly, several spouses sleeping with PD patients reported that they observed a sharp contrast between the slow limited movements and unintelligible low voice of their affected spouse when awake, and the fast, vigorous movements with loud speech that the same patient exhibited during enacted dreams. We recently conducted a study on the quality of movements during RBD in 100 consecutive PD patients without dementia.30 During the interview, all spouses of RBD patients reported an improvement of motor control during the dreams. Movements were improved in 87% of patients (faster in 87%; stronger in 87%; smoother in 51%), speech was better in 77% of patients (more intelligible in 77%; louder in 38%; better articulated in 57%), and facial expression was normalized (loss of parkinsonian amimia) in 47% of patients. This clinical improvement was confirmed on sleep and video monitoring, which was performed after a 12-hour withdrawal of dopaminergic drugs. During REM sleep, there was no evidence of bradykinesia, tremor or dystonia. Movements were surprisingly fast, wide, coordinated, and symmetrical. The restored motor control during REM sleep suggests a transient 'levodopa-like' re-establishment of the basal ganglion loop. Alternatively, parkinsonism might disappear through REM-sleep-related disjunction between pyramidal and extrapyramidal systems. Analyzing the quality of movements during RBD and finding the cause of this spontaneous improvement could both help in the treatment of patients and provide insights into motor control during sleep.

Top

Excessive daytime sleepiness

Excessive daytime sleepiness is defined as a disabling trend to doze or fall asleep in various circumstances (e.g. reading, as a passenger in a car, during a meeting) that interferes with family, professional and social life. In view of the high risk of accidents for sleepy drivers, the level of daytime sleepiness must be regularly checked in patients with PD, especially when the dopaminergic treatment is changed. Sleepiness can be easily self-assessed using the Epworth sleepiness scale (Box 2).54 A score greater than 10 indicates abnormal sleepiness. A score greater than 7 has a 75% sensitivity (and a 50% specificity) for predicting the risk of a driving accident.55 The scale has been validated in large populations of patients with PD. Case–control epidemiological studies performed in various countries consistently found higher sleepiness scores and a higher incidence (range 16–74%, but usually around 33%) of abnormal somnolence in patients with PD than in age-matched and sex-matched controls.6, 56, 57 This range of values probably reflects differences in referral patterns between centers, as well as geographical variations in the probabilities of detecting various pathologies. The incidence of sleepiness was 6% per year in a prospective series of patients with PD.58 It is possible that sleepiness precedes PD onset, as sleepy adults in a large Asian longitudinal study were 3.3 times more likely than nonsleepy adults to develop PD later in life.59

Box 2 Epworth Sleepiness Scale.a

 

How likely are you to doze off or fall asleep in the following situations, in contrast to feeling just tired? This refers to your usual way of life in recent times. Even if you have not done some of these things recently, try to work out how they would have affected you. Use the following scale to choose the most appropriate number for each situation:

0 = would never doze

1 = slight chance of dozing

2 = moderate chance of dozing

3 = high chance of dozing

Sleep disturbances in patients with parkinsonism

APermission obtained from American Sleep Disorders Association and Sleep Research Society © Johns MW (1991) Sleep  14: 540–545.

The most worrying aspect of sleepiness in PD is the sleep attack, defined as 'sudden onset of sleep, without prodroma', which has been described in patients who use the new nonergot dopamine agonist drugs pramipexole and ropinirole.60 Examples of sleep attacks include patients falling asleep during stimulating life conditions, such as eating a meal (the head dropping onto the plate), walking, attending work, carrying a child on an escalator, and, in the most dangerous situation, while driving a car. Estimates of the percentage of patients with PD experiencing sleep attacks vary from 1–14%,55, 61 with 1–4% of patients reporting sleep attacks while driving.

A narcolepsy-like intrinsic phenotype in Parkinson's disease

We monitored daytime sleep in patients with excessive sleepiness using five sequential laboratory-based nap opportunities scheduled every 2 hours (multiple sleep latency tests). In a series of 54 patients with PD, we found that more than half would fall asleep within, on average, 5 minutes, indicating pathological sleepiness.62 Moreover, 41% of sleepy patients fell directly into REM sleep at least twice, an abnormal pattern mostly observed in primary narcolepsy (Figure 1). The 'narcolepsy-like' pattern of sleepiness (multiple sleep-onset REM periods) was also found, although less frequently, in some cases of MSA, dementia with Lewy bodies, and patients with parkin (PARK2) mutations,63 but not in PSP.64 In PD patients with severe hallucinations, visual hallucinations were temporally associated with abnormal irruptions of REM sleep episodes during the daytime, as are the hypnagogic hallucinations in primary narcolepsy.38 Sleepiness is likely to be more frequent in patients with advanced disease.56 Sleep deprivation, a condition observed when patients with PD are kept awake at night by restless legs, disabling motor disability or painful cramps, is a classic cause of somnolence. In large groups of patients with PD, however, a longer time spent asleep at night is associated with more-severe daytime sleepiness,62, 65, 66 an association that is suggestive of central hypersomnia. Sleep apnea (observed in 20–30% of patients with PD), periodic leg movement during sleep and sleep fragmentation do not correlate with the severity of daytime sleepiness,62, 65, 66 suggesting that they do not contribute substantially to the mechanisms of sleepiness.

Mechanisms of sleepiness

The mechanisms of sleepiness and sleep attacks in PD might include a complex drug–disease interaction. Most arousal systems (Figure 4 and Table 2), including the noradrenergic neurons in the locus ceruleus,67 the serotonergic neurons in the raphe,67 the cholinergic neurons in the basal forebrain,67 and the orexinergic neurons (also affected in primary narcolepsy) in the hypothalamus,68, 69 are affected by neuronal loss and Lewy bodies in the brains of individuals with PD. Despite marked cell loss, however, the cerebrospinal fluid levels of orexin are mostly normal in PD and MSA. The wake–active dopaminergic neurons in the ventral periaqueducal gray matter and the histamine neurons in the hypothalamus are intact in the brains of individuals with PD.70, 71

Figure 4 Key components of the ascending arousal systems in the human brain.
Figure 4 : Key components of the ascending arousal systems in the human brain. Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, or to obtain a text description, please contact npg@nature.com

Abbreviations: 5-HT, 5-hydroxytryptamine (serotonin); ACh, acetylcholine; BF, basal forebrain; DA, dopamine; GABA, gamma-aminobutyric acid; His, histamine; LC, locus ceruleus; LH, lateral hypothalamus; NE, norepinephrine; ORX, orexin; Raphe, median raphe nucleus; TMN, tuberomamillary nucleus; vPAG, ventral periaqueductal gray matter.

Full figure and legend (41K)Figures & Tables indexDownload PowerPoint slide (101K)

Table 2 Severity of neuronal loss of the ascending arousal systems in the brains of patients with Parkinson's disease.
Table 2 - Severity of neuronal loss of the ascending arousal systems in the brains of patients with Parkinson's disease.
Full tableFigures & Tables indexDownload PowerPoint slide (100K)

The sedative effects of dopamine agonists have recently been recognized. The use of these agents exposes patients with PD to a twofold to threefold higher risk of sleep attacks.55, 56, 61 The risk of sleep attack is similar for ergot agonists (e.g. bromocriptine, pergolide) and nonergot agonists (e.g. pramipexole, ropinirole). The risk increases with increasing daily dosage and decreases with drug withdrawal. By contrast, levodopa rarely has sedative effects.62 It is currently unknown why drugs that are supposed to stimulate the alerting brain system (and do so when given at bedtime)16 have sedative effects during the daytime. This effect cannot be explained, at these high doses, by the biphasic effect (i.e. presynaptic sedative effect at low dose, postsynaptic alerting effect at high dose) of dopamine agonists that has been described in animals.72

Treatment

Finding and treating the cause of daytime sleepiness in PD usually requires a minimal interview on nocturnal disturbances, hallucinations, recent changes in dopaminergic and psychotropic treatment, and, usually, night-time sleep monitoring, if possible followed by multiple sleep latency tests. These latter tests can recognize severe sleep apnea, or a central disorder of arousal. If efforts to reduce the dose of any sedative drug (e.g. clonazepam, other benzodiazepines, dopamine agonists, sedative antidepressants or opioids) have no effect or worsen the motor symptoms, the solution could be to add a psychostimulant during the daytime. Modafinil, a drug routinely used in primary narcolepsy and sleepiness of various causes, is well tolerated in PD patients,73 and has intriguing neuroprotective effects in animal models of dopamine depletion.74 The alerting effect is limited, however, with fewer than one third of individuals showing a response.73

Future developments

Understanding the mechanisms of sleepiness in PD and the sedative effects of dopamine agonists is a key issue that is now being explored by motor and sleep monitoring and pharmacological challenges in various animal models of PD.75 Increasing the levels of brain histamine, by use of selective reuptake inhibitors (anti-H3 receptor drugs) is a new research approach that is being explored in various models of human somnolence, including parkinsonism. Drugs that have been developed to treat primary narcolepsy (e.g. sodium oxybate given at night, or adenosine receptor antagonists) might benefit patients with PD-associated sleepiness.

Top

Stridor

Stridor is a partial obstruction of the larynx, which results in a harsh, high-pitched inspiratory noise. In contrast to obstructive sleep apnea (a pharynx collapse), which does not expose the individual to an immediate risk of death, stridor is a life-threatening condition. The larynx obstruction usually begins during the night, and was observed in 42% of unselected patients with MSA.76 It can be recognized quite easily by mimicking the associated noise to the caregiver, or by conducting an audio recording during the night, but it is not detected on standard apnea monitoring devices. Interestingly, night-time stridor is alleviated by the application of nasal continuous positive airway pressure,77 which can avoid tracheotomy and provide long-term benefits with regard to quality of sleep and median survival time.78

Top

Conclusions

The recent resurgence of interest in sleep disturbances in parkinsonism benefits both the movement disorder and sleep fields. The mechanisms of sleep disorders in PD point to a complex interaction between lesions in sleep–wake systems, movement disorders and adverse effects of dopaminergic agents. A specific, complex REM sleep disorder contributes to RBD, hallucinations and excessive daytime sleepiness, and cohorts of patients with idiopathic RBD can provide insights into preclinical PD and dementia. The range of behaviors observed during RBD is a precious, objective insight into dream content and motor control during sleep, as shown by the recently recognized alleviation of parkinsonism during RBD.

From the clinical point of view, physicians now take a more active interest in the patient's night-time suffering, and are increasingly aware of public health issues relative to alertness while driving with PD. The management of sleep and alertness problems benefits from thorough analysis of their potential causes, through clinical interviews, standardized scales, and night-time and daytime sleep monitoring. Each treatment can be assessed with regard to its sleep–wakefulness effects. For example, small doses of sedative antidepressant can improve sleep continuity but worsen an underlying RBD. By contrast, clonazepam alleviates RBD but might worsen sleep apnea. Dopamine agonists given in the evening can improve RLS. Levodopa at night might dramatically alleviate akinesia and dystonia, but higher doses might increase wakefulness. As in the case of other motor and nonmotor symptoms in PD, subtle adjustments in drug regimens could lead to major changes in night-time sleep quality and daytime alertness of patients.

Key points

  • As many as 60% of patients with Parkinson's disease (PD) experience insomnia, 15–60% exhibit rapid eye movement (REM) sleep behavior disorders (RBDs), and 30% show excessive daytime sleepiness
  • Insomnia in PD is a distressing symptom that can be improved by better control of night-time disability, restless legs syndrome and dystonia
  • RBDs are characterized by violent enacted dreams; this condition might act as a harbinger of parkinsonism several years before disease onset
  • A striking disappearance of parkinsonism is observed during RBD in PD, suggesting a restoration of motor control during sleep
  • In patients with PD, daytime sleepiness can culminate in sleep attacks that can impair driving ability; a narcolepsy-like phenotype is observed in one-third of patients
  • Sleepiness is caused by a disease–drug interaction, with specific lesions in arousal (noradrenergic, serotonergic and orexinergic) systems; dopamine agonists expose patients to a twofold increase in the frequency of sleep attacks
  • Stridor is the main nocturnal complication in multiple system atrophy, and patients with this condition can benefit from continuous positive airway pressure

Acknowledgments

The sleep studies performed by the authors in patients with parkinsonism were in part financed by grants from Inserm, Fondation Lilly, Fédération pour la Recherche sur le Cerveau, and France-Parkinson.

References

  1. American Academy of Sleep Medicine (2005) The International Classification of Sleep Disorders—Revised. Chicago: American Academy of Sleep Medicine
  2. Tandberg E et al. (1998) A community-based study of sleep disorders in patients with Parkinson's disease. Mov Disord 13: 895–899 | Article | PubMed | ChemPort |
  3. Lees AJ et al. (1988) The night-time problems of Parkinson's disease. Clin Neuropharmacol 11: 512–519 | Article | PubMed | ChemPort |
  4. Ghorayeb I et al. (2002) Sleep disorders and their determinants in multiple system atrophy. J Neurol Neurosurg Psychiatry 72: 798–800 | Article | PubMed | ChemPort |
  5. Montplaisir J et al. (1997) Sleep and quantitative EEG in patients with progressive supranuclear palsy. Neurology 49: 999–1003 | PubMed | ChemPort |
  6. Tan E et al. (2002) Evaluation of somnolence in Parkinson's disease: comparison with age- and sex-matched controls. Neurology 58: 465–468 | PubMed | ChemPort |
  7. Factor SA et al. (1990) Sleep disorders and sleep effect in Parkinson's disease. Mov Disord 5: 280–285 | Article | PubMed | ChemPort |
  8. Van Hilten B et al. (1994) Sleep disruption in Parkinson's disease. Assessment by continuous activity monitoring. Arch Neurol 51: 922–928 | PubMed | ChemPort |
  9. Chaudhuri KR et al. (2002) The Parkinson's disease sleep scale: a new instrument for assessing sleep and nocturnal disability in Parkinson's disease. J Neurol Neurosurg Psychiatry 73: 629–635 | Article | PubMed | ChemPort |
  10. Arnulf I et al. (2000) Improvement of sleep architecture in PD with subthalamic nucleus stimulation. Neurology 55: 1732–1735 | PubMed | ChemPort |
  11. Iranzo A et al. (2002) Sleep symptoms and polysomnographic architecture in advanced Parkinson's disease after chronic bilateral subthalamic stimulation. J Neurol Neurosurg Psychiatry 72: 661–664 | Article | PubMed | ChemPort |
  12. Monaca C et al. (2004) Effects of bilateral stimulation on sleep in Parkinson's disease. J Neurol 251: 214–218 | Article | PubMed | ChemPort |
  13. Ondo W et al. (2002) Exploring the relationship between Parkinson disease and restless legs syndrome. Arch Neurol 59: 421–424 | Article | PubMed |
  14. Nomura T et al. (2006) Prevalence and clinical characteristics of restless legs syndrome in Japanese patients with Parkinson's disease. Mov Dis 21: 380–384 | Article |
  15. Borek LL et al. (2006) Mood and sleep in Parkinson's disease. J Clin Psychiatry 67: 958–963 | PubMed |
  16. Comella CL et al. (2005) Nocturnal activity with nighttime pergolide in Parkinson disease: a controlled study using actigraphy. Neurology 64: 1450–1451 | PubMed |
  17. Dowling GA et al. (2005) Melatonin for sleep disturbances in Parkinson's disease. Sleep Med 6: 459–466 | Article | PubMed |
  18. Abe K et al. (2005) A hypnotic drug for sleep disturbances in patients with Parkinson's disease [Japanese]. No To Shinkei 57: 301–305 | PubMed |
  19. Linazasoro G et al. (1993) Nocturnal akathisia in Parkinson's disease: treatment with clozapine. Mov Disord 8: 171–174 | Article | PubMed | ChemPort |
  20. Juri C et al. (2005) Quetiapine for insomnia in Parkinson disease: results from an open-label trial. Clin Neuropharmacol 28: 185–187 | Article | PubMed | ChemPort |
  21. Poewe WH et al. (2007) Efficacy of pramipexole and transdermal rotigotine in advanced Parkinson's disease: a double-blind, double-dummy, randomised controlled trial. Lancet Neurol 6: 513–520 | Article | PubMed | ISI | ChemPort |
  22. Leeman AL et al. (1987) Parkinson's disease in the elderly: response to and optimal spacing of night time dosing with levodopa. Br J Clin Pharmacol 24: 637–643 | PubMed | ChemPort |
  23. Lu J et al. (2006) Identification of wake-active neurones in the ventral periaqueducal gray matter. J Neurosci 26: 193–202 | Article | PubMed | ISI | ChemPort |
  24. Kedia S et al. (2004) Emergence of restless legs syndrome during subthalamic stimulation for Parkinson disease. Neurology 63: 2410–2412 | PubMed | ChemPort |
  25. Boeve B et al. (2001) Association of REM sleep behavior disorder and neurodegenerative disease may reflect an underlying synucleinopathy. Mov Disord 16: 622–630 | Article | PubMed | ChemPort |
  26. De Cock VC et al. (2007) REM sleep behavior disorder in patients with guadeloupean parkinsonism, a tauopathy. Sleep 30: 1026–1032 | PubMed |
  27. Gagnon JF et al. (2006) Update on the pharmacology of REM sleep behavior disorder. Neurology 67: 742–747 | Article | PubMed |
  28. Iranzo A et al. (2005) Characteristics of idiopathic REM sleep behavior disorder and that associated with MSA and PD. Neurology 65: 247–252 | Article | PubMed | ChemPort |
  29. Scaglione C et al. (2005) REM sleep behaviour disorder in Parkinson's disease: a questionnaire-based study. Neurol Sci 25: 316–321 | Article | PubMed | ChemPort |
  30. De Cock VC et al. (2007) Restoration of normal motor control in Parkinson's disease during REM sleep. Brain 130: 450–456 | Article | PubMed |
  31. Schenck CH and Mahowald MW (2002) REM sleep behavior disorder: clinical, developmental, and neuroscience perspectives 16 years after its formal identification in SLEEP. Sleep 25: 120–138 | PubMed |
  32. Fantini ML et al. (2005) Aggressive dream content without daytime aggressiveness in REM sleep behavior disorder. Neurology 65: 1010–1015 | Article | PubMed | ChemPort |
  33. Boeve BF et al. (2007) Pathophysiology of REM sleep behaviour disorder and relevance to neurodegenerative disease. Brain 130: 2770–2788 | Article | PubMed | ChemPort |
  34. Sastre J and Jouvet M (1979) Oneiric behavior in cats [French]. Physiol Behav 22: 979–989 | Article | PubMed | ISI | ChemPort |
  35. Schenkel E and Siegel J (1989) REM sleep without atonia after lesions of the medial medulla. Neurosci Lett 98: 159–165 | Article | PubMed | ChemPort |
  36. Boissard R et al. (2003) Localization of the GABAergic and non-GABAergic neurons projecting to the sublaterodorsal nucleus and potentially gating paradoxical sleep onset. Eur J Neurosci 18: 1627–1639 | Article | PubMed | ISI |
  37. Lu J et al. (2006) A putative flip–flop switch for control of REM sleep. Nature 441: 589–594 | Article | PubMed | ChemPort |
  38. Arnulf I et al. (2000) Hallucinations, REM sleep and Parkinson's disease. Neurology 55: 281–288 | PubMed | ChemPort |
  39. Del Tredici K et al. (2002) Where does Parkinson disease pathology begin in the brain? J Neuropathol Exp Neurol 61: 413–426 | PubMed |
  40. Schenck CH et al. (1986) Chronic behavioral disorders of human REM sleep: a new category of parasomnia. Sleep 9: 293–308 | PubMed | ChemPort |
  41. Kunz D and Bes F (1999) Melatonin as a therapy in REM sleep behavior disorder patients: an open-labeled pilot study on the possible influence of melatonin on REM-sleep regulation. Mov Disord 14: 507–511 | Article | PubMed | ChemPort |
  42. Olson EJ et al. (2000) Rapid eye movement sleep behaviour disorder: demographic, clinical and laboratory findings in 93 cases. Brain 123: 331–339 | Article | PubMed |
  43. Schenck CH et al. (2003) REM behavior disorder: delayed emergence of parkinsonism and/or dementia in 65 % of older men initially diagnosed with idiopathic RBD, and an analysis of the minimum and maximum tonic and/or phasic electromyographic abnormalities found during REM sleep [abstract]. Sleep 26 (Suppl): A316
  44. Schenck CH et al. (1996) Delayed emergence of a parkinsonian disorder in 38% of 29 older men initially diagnosed with idiopathic rapid eye movement sleep behaviour disorder. Neurology 46: 388–393 | PubMed | ChemPort |
  45. Iranzo A et al. (2006) Rapid-eye-movement sleep behaviour disorder as an early marker for a neurodegenerative disorder: a descriptive study. Lancet Neurol 5: 572–577 | Article | PubMed |
  46. Eisensehr I et al. (2000) Reduced striatal dopamine transporters in idiopathic rapid eye movement sleep behavior disorder. Comparison with Parkinson's disease and controls. Brain 123: 1155–1160 | Article | PubMed | ISI |
  47. Ferini-Strambi L et al. (2004) Neuropsychological assessment in idiopathic REM sleep behavior disorder (RBD): does the idiopathic form of RBD really exist? Neurology 62: 41–45 | PubMed | ChemPort |
  48. Ferini-Strambi L et al. (1996) Cardiac autonomic activity during wakefulness and sleep in REM sleep behavior disorder. Sleep 19: 367–369 | PubMed | ChemPort |
  49. Stiasny-Kolster K et al. (2005) Combination of 'idiopathic' REM sleep behaviour disorder and olfactory dysfunction as possible indicator for alpha-synucleinopathy demonstrated by dopamine transporter FP-CIT-SPECT. Brain 128: 126–137 | Article | PubMed | ISI | ChemPort |
  50. Fantini ML et al. (2003) Slowing of electroencephalogram in rapid eye movement sleep behavior disorder. Ann Neurol 53: 774–780 | Article | PubMed | ISI |
  51. Pacchetti C et al. (2005) Relationship between hallucinations, delusions, and rapid eye movement sleep behavior disorder in Parkinson's disease. Mov Disord 20: 1439–1448 | Article | PubMed |
  52. Gagnon JF et al. (2004) Association between waking EEG slowing and REM sleep behavior disorder in PD without dementia. Neurology 62: 401–406 | PubMed |
  53. Vendette M et al. (2007) REM sleep behavior disorder predicts cognitive impairment in Parkinson disease without dementia. Neurology 69: 1843–1849 | Article | PubMed | ChemPort |
  54. Johns MH (1991) A new method for measuring daytime sleepiness: the Epworth Sleepiness Scale. Sleep 14: 540–545 | PubMed | ISI | ChemPort |
  55. Hobson D et al. (2002) Excessive daytime sleepiness and sudden-onset sleep in Parkinson disease. A survey by the Canadian Movement Disorder Group. JAMA 287: 455–463 | Article | PubMed |
  56. Ondo WG et al. (2001) Daytime sleepiness and other sleep disorders in Parkinson's disease. Neurology 57: 1392–1396 | PubMed | ChemPort |
  57. Ghorayeb I et al. (2007) A nationwide survey of excessive daytime sleepiness in Parkinson's disease in France. Mov Disord 22: 1567–1572 | Article | PubMed |
  58. Gjerstad MD et al. (2002) Development of daytime somnolence over time in Parkinson's disease. Neurology 58: 1544–1546 | PubMed | ChemPort |
  59. Abbott RD et al. (2005) Excessive daytime sleepiness and subsequent development of Parkinson disease. Neurology 65: 1442–1446 | Article | PubMed | ChemPort |
  60. Frucht S et al. (1999) Falling asleep at the wheel: motor vehicle mishaps in persons taking pramipexole and ropinirole. Neurology 52: 1908–1910 | PubMed | ChemPort |
  61. Korner Y et al. (2004) Predictors of sudden onset of sleep in Parkinson's disease. Mov Disord 19: 1298–1305 | Article | PubMed |
  62. Arnulf I et al. (2002) Parkinson's disease and sleepiness: an integral part of PD. Neurology 58: 1019–1024 | PubMed | ChemPort |
  63. Arnulf I (2005) Excessive daytime sleepiness and parkinsonism. Sleep Med Rev 9: 185–200 | Article | PubMed |
  64. Arnulf I et al. (2005) REM sleep behavior disorder and REM sleep without atonia in patients with progressive supranuclear palsy. Sleep 28: 349–354 | PubMed |
  65. Razmy A et al. (2004) Predictors of impaired daytime sleep and wakefulness in patients with Parkinson's disease treated with older (ergot) vs newer (nonergot) dopamine agonists. Arch Neurol 61: 97–102 | Article | PubMed |
  66. Rye DB et al. (2000) FAST TRACK: daytime sleepiness in Parkinson's disease. J Sleep Res 9: 63–69 | Article | PubMed | ChemPort |