Sleep and circadian rhythm disturbances are central features of many movement disorders, exacerbating motor and non-motor symptoms and impairing quality of life. Understanding these disturbances to sleep is clinically important and may further our understanding of the underlying movement disorder. This review evaluates the current anatomical and neurochemical understanding of normal sleep and the recognised primary sleep disorders. In addition, we undertook a systematic review of the evidence for disruption to sleep across multiple movement disorders. Rapid eye movement sleep behaviour disorder has emerged as the most reliable prodromal biomarker for the alpha synucleinopathies, including Parkinson’s disease and multiple system atrophy, often preceding motor symptom onset by several years. Abnormal sleep has also been described for many other movement disorders, but further evidence is needed to determine whether this is a primary or secondary phenotypic component of the underlying condition. Medication used in the treatment of motor symptoms also affects sleep and can aggravate or cause certain sleep disorders. Within the context of movement disorders, there is also some suggestion of a shared underlying mechanism for motor and sleep pathophysiology, with evidence implicating thalamic and brainstem structures and monoaminergic neurotransmission. This review highlights the need for an understanding of normal and abnormal sleep within the movement disorder clinic, an ability to screen for specific causes of poor sleep and to treat sleep disturbance to improve quality of life. Key sleep disorders also act as important biomarkers and have implications in diagnosis, prognosis and the development of future therapies.
- movement disorders
- sleep disorders
Statistics from Altmetric.com
Abnormal sleep and circadian rhythm disorders are seen in many patients with movement disorders, although they remain underdiagnosed and are a significant unmet need in clinical care. Sleep and circadian rhythms are generated from specific regulatory centres, including the forebrain, thalamus and midbrain dopaminergic neurons, regions also implicated in movement disorder pathogenesis.
The rate, nature and temporal pattern of these sleep disturbances vary across the different movement disorders. For example, in Parkinson’s disease (PD) the rate of sleep disorders has been estimated to be as high as 98%, with rapid eye movement (REM) sleep behaviour disorder well established as predating motor symptom onset.1 In contrast, those such as adult-onset primary, idiopathic dystonia, may have poor sleep with evidence suggesting a link to psychiatric symptom severity.2
This review provides an overview of the common sleep disorders and the evidence to date for the patterns and prevalence across a spectrum of movement disorders. We discuss the tools available for sleep assessment, the impact of medication used in motor symptom management on sleep, and how disturbed sleep can affect daytime motor function. Finally, we highlight future challenges to the field, and general considerations in the assessment and treatment of sleep disorders in patients diagnosed with a movement disorder.
Sleep structure, anatomy and physiology
Sleep cycles approximately every 90–120 min between an initial non-REM (NREM) stage (subdivided into N1–3 stages), followed by a REM/R phase with a relative muscle atonia. Figure 1 summarises the typical hypnogram, although with normal ageing, night awakenings increase, N3 decreases and total sleep time shortens.
Maintenance of the sleep/wake circuit is regulated by the fast-acting neurotransmitters glutamate and gamma aminobutyric acid (GABA), while an array of excitatory neurotransmitters are thought to have a modulatory role, including hypocretin/orexin (lateral hypothalamus, LH), histamine (tuberomammillary nucleus), acetylcholine (pedunculopontine (PPT) and laterodorsal tegmentum), norepinephrine (locus coeruleus), dopamine (ventral periaqueductal grey (vPAG) matter) and serotonin (raphe nuclei) (figure 2A–C). Sleep regulatory circuits are widespread throughout the brainstem and hypothalamus. GABAergic parvalbumin neurons in the basal forebrain arousal system and GABAergic neurons in the parafacial zone of the medulla are essential for driving wakefulness, with the latter promoting sleep via inhibition of glutamatergic neurons in the parabrachial nucleus (PB). More recently, evidence suggests that GABAergic neurons in the LH contribute to promoting wakefulness by inhibiting the thalamus and preoptic area. The ventrolateral preoptic area also contains GABAergic neurons, essential for promoting sleep via inhibition of the arousal system.
Glutamatergic REM-promoting neurons in the sublaterodorsal nucleus (SLD) are inhibited by GABAergic NREM-promoting neurons in the vPAG, while activation of GABAergic neurons in the ventral medulla which innervate vPAG promotes transition to REM sleep by disinhibition of the SLD. NREM sleep is driven by GABAergic neurons in the vPAG and glutamatergic neurons located ventromedial to the superior cerebellar peduncle. These neurons suppress both wakefulness and REM sleep by inhibiting the SLD, with additional inhibition of the medial PB by GABAergic neurons also promoting NREM sleep. Newer models of the sleep/wake circuit recognise that these interactions are modulated by several brain regions including the PPT, laterodorsal tegmental nuclei (LDT), projections from the medulla and the melanin-concentrating neuron system (figure 2A,B).3
Types of sleep disorders
The different forms of sleep disorders, as defined by the International Classification of Sleep Disorders, their clinical features, diagnostic criteria, proposed underlying pathophysiology and currently suggested treatments are summarised in table 1.
Diagnosing sleep disorders
Online supplemental table 1 summarises the available sleep questionnaires. Sleep diaries and wrist actigraphy are often used for the diagnosis of insomnia and circadian rhythm disorders as well as population screening in research cohorts. The gold standard for objective sleep staging remains in-patient video polysomnography (PSG), involving video, electroencephalography, electromyography and electrooculography. PSG is required for confirming the diagnosis of REM Sleep Behavioural Disorder (RBD), while PSG and multiple sleep latency test (MSLT) are used in the diagnoses of narcolepsy and other central hypersomnias.
Technology to aid diagnostic and therapeutic management
Development of consumer wearables is becoming recognised as a prospective clinical tool, aiding early diagnosis as well as the opportunity to monitor disease progression particularly, for example, in response to therapeutics. Actigraphy, an accelerometer validated to detect sleep/wake activity patterns, is particularly useful for documenting sleep-wake patterns for periods of days or weeks in the patient’s own environment, as well as measuring physical activity during periods of wakefulness. Use of these wearable devices extends far beyond disease alone, potentially providing avenues to better understand the association between sleep and health, as well as opportunities to optimise interventions and reduce the adverse health impacts of poor sleep.
Review of the evidence for sleep disturbance across degenerative and non-degenerative movement disorders
The literature search included the key words “sleep disturbances”, “sleep disorders”; “RLS”; “PLMS”; “sleep related breathing disorders”; “RBD”; “excessive daytime sleepiness”; “insomnia” in combination with degenerative movement disorders: PD, Dementia with Lewy bodies (DLB), Multiple System Atrophy (MSA), Progressive Supranuclear Palsy (PSP), Corticobasal Syndrome (CBS), Huntington’s Disease (HD), Dentatorubral-pallidoluysian Atrophy, spinocerebellar ataxia (SCA), Anti-immunoglobulin-like cell adhesion molecule 5 (IgLON5) disease, Wilson’s Disease (WD), Niemann-Pick disease type C (NPC), Neurodegeneration with Brain Iron Accumulation (NBIA) including: pantothenate kinase-associated neurodegeneration, PLA2G6-associated neurodegeneration, beta-propeller protein-associated neurodegeneration, mitochondrial membrane protein-associated neurodegeneration and fatty acid hydroxylase-associated neurodegeneration, neuroacanthocytosis including: McLeod syndrome and Chorea-acanthocytosis and non-degenerative movement disorders: Dystonia, Essential Tremor (ET), Dystonic Tremor, Myoclonus (excluding myoclonus in relation to epilepsy), Tourette’s syndrome (TS), Tic Disorders and Benign Hereditary Chorea (BHC). Studies identified were divided according to cohort size and methods of assessment (online supplemental tables 2–6). Only those involving PSG studies undertaken in case–control cohorts are discussed below and summarised in tables 2 and 3. Where there is evidence, we have also sought to discuss potential underlying pathophysiological mechanisms.
Sleep disorders in PD
Sleep disturbance, which is typically multifactorial, is as high as 74%–98% in individuals diagnosed with PD, with well documented evidence of parasomnias including RBD, Periodic Limb Movements During Sleep (PLMS) and sleep related breathing disorders. Many sleep symptoms are related to disease duration, for example, there are increased disturbances to sleep architecture, insomnia and somnolence in later disease stages compared with earlier timepoints. Two recent review articles provide detailed oversight of the PSG findings and therapeutic management of sleep disturbance in PD, and therefore we have only provided a brief overview below.4 5
RBD is well characterised in PD populations with longitudinal case–control studies identifying 73.5% of patients with idiopathic RBD (iRBD) develop a neurodegenerative disorder after 12 years.6 Those patients diagnosed with PD and RBD tend towards more severe clinical features, including increased motor symptom severity, higher rates of axial symptoms, psychiatric symptoms, hallucinations, autonomic dysfunction, cognitive impairment and impulse control disorders, as well as a generally longer motor disease duration, when compared with PD patients without RBD.7–9 Recent imaging studies of those with comorbid PD and RBD demonstrate prominent loss of volume in the pontomesencephalic tegmentum, PPN network dysfunction and reduced signalling in the locus coeruleus compared with those with PD alone.10–12
Although often reported, systematic studies of RLS and PLMS have found conflicting results, but with close matching of control populations a general absence of RLS was observed, while use of actigraphy demonstrated high rates of PLMS. Evidence for the pathogenesis of PD and RLS is conflicting, some studies have reported reduced nigrostriatal dopaminergic activity in idiopathic RLS,13 while one longitudinal imaging study demonstrated increased dopaminergic transporter availability in both the caudate and putamen to be associated with the presence of RLS.14 It is also possible that RLS in PD is related to dopaminergic treatment rather than the disease itself. The prevalence of RLS in de novo PD patients is comparable to controls, with a longitudinal study showing significantly higher rates of RLS at 3 year follow-up than that at baseline in drug naïve patients.15 Stimulation of the subthalamic nuclei (STN) has also been shown to increase rates of RLS, with overstimulation of the dopaminergic system thought to resemble the clinical features of augmentation.7
Conflicting results have been found for rates of sleep apnoea syndromes (SAS), with some studies identifying comparable apnoea-hypopnea indices (AHI) levels and a comparable risk of OSA and central sleep apnoea (CSA), while elevated AHI have also been reported at higher rates in those with PD compared with controls.16–19 However, several of these studies have not identified the typical oxygen desaturation profiles observed with obstructive SAS, as well as the body mass indices of those with PD being significantly lower compared with that of controls. These findings suggest that the sleep disordered breathing identified may represent nocturnal motor involvement of respiratory muscles or a precursor to later stage dysautonomia, rather than the more typical OSA observed in the general population.20 Extensive loss of neurons, in particular within respiratory control structures such as the brainstem, may also contribute to respiratory disorders. PD patients show abnormal ventilatory response to hypoxia and hypercapnia suggesting involvement of the medulla, with neuronal and astrocyte reduction in the retrotrapezoid nucleus, nucleus of the solitary tract and pre-Bötzinger complex resulting in abnormal respiratory function in murine models of PD.21
Several studies have also reported reduced melatonin secretion in patients, with a case–control study showing a fourfold decrease in circulating melatonin levels in those with PD, and reduced melatonin rhythm amplitude in those with excessive daytime sleepiness.22 These circadian disturbances are present in early disease stages and are often associated with cognitive decline and psychiatric disorders.23 In addition, reduced expression of Bmal1 (a clock gene) has been found to correlate positively with PD severity and PSQI scores, providing further evidence of a link between circadian rhythm and sleep disturbances.24
Sleep disorders related to motor symptoms in PD
Studies involving PD cohorts (n=45) have shown a positive correlation between PLMS, daytime sleepiness and more severe motor rating scores and reduced mobility.25 A number of studies have also established a strong link between iRBD and subsequent neurodegenerative disorders, with retrospective analysis of PSG data (n=59 PD patients) identifying faster motor progression in those with RBD (n=15) and REM sleep without atonia (RWA) (n=22) compared with those with preserved REM sleep atonia (n=22, p=0.015). However, this could also be due to increased disease severity and degeneration associated with RBD in PD patients.26 Several studies also implicate cognitive impairment in the exacerbation of sleep disorders and altered sleep structure.27 Patients with PD with primary sleep disorders (RBD and insomnia) had significantly poorer global cognition, visuospatial and executive function in comparison to both unaffected controls and patients with PD without sleep disorders.28
Impact of treatment used in the management of motor symptoms on sleep
Night-time akinesia can worsen underlying disturbances to sleep. A placebo-controlled study of controlled release levodopa/carbidopa in patients with PD (n=40) found improved nocturnal akinesia and therefore increased total hours slept, however, immediate release preparations result in reduced total sleep time and lower rates of RBD-related symptoms.17 In contrast, several studies have shown that the use of dopamine agonists results in increased levels of nocturnal activity and higher levels of daytime sleepiness compared with levodopa therapy.29 Sudden-onset sleep is sometimes described in PD and attributed to the sedative effects of dopamine agonists, with these ‘sleep attacks’ prompting safety concerns, particularly in relation to driving.30 However, transdermal rotigotine patches have been shown to effectively aid sleep continuity, sleep stability and REM sleep, with comparable outcomes to pramipexole and superior to treatment with ropinirole.31 32 Dopamine agonists are also licensed for use in the treatment of RLS and PLMS, however, increasing recognition of impulse control disorder and augmentation as common side effects have limited their use, with pregabalin or gabapentin frequently used as alternatives.
Monoamine oxidase inhibitors
Rasagiline can be alerting, with fatigue-related questionnaires completed by those with PD showing significant improvements to baseline scores at 12 weeks (n=16, p=0.003). Dopamine-naïve PD patients (n=1176) demonstrated dose-related improvements at week 36 compared with placebo (1 mg p=0.003; 2 mg p<0.0001), while PSG studies show increased sleep maintenance and lower wake after sleep onset (WASO) compared with controls.33 Safinamide also improves sleep disorders in PD, likely due to its dual dopaminergic and glutaminergic mechanisms, while amphetamine-like derivatives of selegiline can worsen sleep quality by promoting the onset of insomnia.
Deep brain stimulation
PSG studies have noted improvements following STN stimulation, with increased total sleep time, decreased WASO (p<0.05) and longer periods of continuous sleep (p<0.05), while PPN stimulation—although experimental and performed in few patients—resulted in increased percentage REM sleep in both PD and PSP populations.34 The efficacy of DBS in the management of sleep disorders in PD is conflicting, the majority of studies suggest that DBS has no impact on RBD,35 although restoration of RWA has been noted.36 Several studies have shown improvement in RLS symptoms following DBS,37 38 while one study found that STN-DBS resulted in the emergence of RLS in 19% of those with PD.7
Therapeutics used in the management of sleep disorders in PD
These include stimulants such as modafinil and methylphenidate, used to promote wakefulness in hypersomnolence. A recent meta-analysis demonstrated improvement in subjective sleepiness with modafinil, although no subjective improvement in fatigue or objective measures of sleepiness (MSLT or MWT).39 Studies of other agents have also shown promising results with sodium oxybate (GABAB receptor agonist) decreasing ESS scores and increasing mean sleep latency, while methylphenidate resulted in a significant reduction to fatigue compared with placebo in PD populations.40
Melatonin is produced by the pineal gland with levels rising in darkness and acting as a key regulator of the sleep wake cycle. It has a short half-life at 30–60 min and is licensed to treat sleep onset insomnia as well as being used off licence to treat RBD with PSG studies showing restoration of REM atonia, with case–control studies suggesting fewer side effects compared with clonazepam.41 However, a recent placebo-controlled randomised controlled trial (RCT) found prolonged-release melatonin to have no significant impact on the subjective reporting of RBD symptoms.42 Agomelatine, a synthetic melatonin analogue, initially developed for treatment of depression, has also been shown to alleviate sleep disturbances (PLMS and awakenings) in PD.43
Antidepressants and antipsychotics
Trazodone, a serotonin antagonist and reuptake inhibitor, is licensed for use in the treatment of insomnia in some countries. While there are no studies which evaluate its therapeutic use for sleep disorders within the PD population, effective doses within animal models alleviate dyskinesia and psychosis, but appear to cause sedation.44 Other atypical antidepressants including mirtazapine can be effective for insomnia due to potent antihistamine effects but have also been reported to induce or aggravate RBD symptoms in PD cohorts, while use of doxepin in those with PD demonstrated substantial improvements to insomnia, sleep quality and fatigue. Commonly used for the management of hallucinations and delusions, quetiapine, clozapine and pimavenserin are not typically associated with the worsening of motor symptoms in movement disorders. While quetiapine leads to improvements in PD-related psychosis, it has not been shown to impact disturbances to sleep architecture. In contrast, preliminary data relating to pimavanserin (a novel 5-HT2A agonist) suggests improvements to nocturnal symptoms without worsening daytime sedation.45 Finally, while there are no RCTs, clonazepam remains widely used in the treatment of RBD, in doses from 0.5 to 2 mg, although side effects of sedation at higher doses can be problematic.
Sleep disorders in DLB and the impact of motor disease treatment on sleep
Fluctuating levels of alertness and REM sleep behaviour disorder form core diagnostic criteria for DLB, while PSG evidence of loss of REM atonia also forms a supportive diagnostic biomarker. RBD in DLB, is often experienced several years before the onset of cognitive decline with PSG studies indicating rates of RBD in DLB to be as high as 71%.46 Case–control studies have also identified higher levels of PLMS, with sleep architecture abnormalities evident in 75%, predominantly in the form of RWA (44%).46 Cholinesterase inhibitors (galantamine, donepezil and rivastigmine) used to treat cognitive decline in DLB have been shown to improve sleep quality, reduce fragmented sleep and reduce nocturnal activity.47 48
Sleep disorders in MSA
Sleep disorders are also well documented for MSA. In brief, particularly high rates of RBD are reported (64%–100%), with 76% (n=21) of one study reporting onset prior to their motor symptoms.49 Respiratory stridor is also well recognised, as well as being associated with more severe disease progression and sudden death. Case–control studies have identified increased rates of OSA and AHI levels >5, compared with controls with continuous positive airways pressure (CPAP) improving stridor and sleep apnoea.50 Disruption to subcortical serotonergic neurotransmission is considered to be a key pathophysiological contributor, with depletion of the serotonergic innervation of the raphe nuclei and neurodegeneration of regions involved in respiratory regulation—the brainstem ventral arcuate nucleus and pre-Bötzinger complex cells—considered central in pathogenesis.
REM sleep behaviour disorder in α-synuclein disorders
As outlined above, RBD is a recognised early clinical manifestation among the α-synucleinopathies; PD, MSA and DLB, often preceding disease onset by several years and representing a key prodromal marker.51 Several imaging studies of those with iRBD have identified changes in multiple brainstem structures, namely the PPN, SLD and LDT region, with those without changes in these regions not observed to later develop PD.52
Sleep disorders in PSP
Studies to date suggest at least 60% of patients diagnosed with PSP report sleep disturbance, with insomnia being the dominant symptom. RBD and RWA also occur, but at lower rates (5%–13%) than that seen in PD, with this not seeming to antedate motor disease onset.53 Imaging studies show degeneration of REM-promoting structures, including the PPN and pontine tegmentum, as the likely cause.54 A recent study also reported that patients diagnosed with PSP with Lewy type alpha-synucleinopathy were more frequently diagnosed with probable RBD compared with those with PSP without Lewy pathology.55
Sleep disorders in CBS
Two small PSG studies reported sleep disorders in all of the patients involved, these included PLMS and/or RLS and sleep related breathing disorders including OSA and CSA. Alterations to sleep architecture were also observed, with patients showing decreases in sleep maintenance, total sleep time and RWA.56 57Although no studies have reported narcolepsy in CBS, there is evidence of reduced hypocretin in cerebral spinal fluid (CSF) compared with those with PD.58
Neurodegenerative trinucleotide repeat disorders
Sleep disorders in HD
While sleep disturbance is a frequent complaint, with at least 90% of carers describing poor sleep, it remains less well characterised due to motor and psychiatric symptoms often dominating, as well as a lack of patient insight. Approximately two-thirds of patient cohorts report insomnia, describing difficulty falling asleep and maintaining sleep, with PSG showing increased sleep fragmentation, decreased N3 and REM compared with controls. Insomnia is often present early in disease, as well as being reported in premotor manifesting HTT carriers, and increasing with disease progression.59 Reduced REM sleep (n=30) has also been shown to be inversely correlated with motor symptom severity (p<0.05), with cognitive impairment also implicated in the exacerbation of sleep disorders and altered sleep structure.27
Patients with HD also demonstrate strikingly abnormal circadian rhythms with both a delayed and irregular phase to sleep, with evidence from actigraphy monitoring, as well as abnormal melatonin and cortisol profiles. These changes are also consistently observed in animal models of disease, and are attributed to degeneration of the suprachiasmatic nucleus, hypothalamus and striatum.60 61 They can also occur early in the condition and are associated with higher levels of cognitive impairment and anxiety.62 Human postmortem studies have shown a loss of orexin-releasing neurons in individuals diagnosed with HD, while transgenic HD mouse models suggest that changes to SCN circuitry are central in driving circadian disruption.60
Sleep in relation to the treatment of HD
Tetrabenazine, a monoamine reuptake inhibitor, is often used in the management of chorea associated with HD although daytime sleepiness is a commonly reported side effect. A study examining the effect of quetiapine (a partial D2 receptor antagonist) in those with HD found amelioration of their symptoms of insomnia, without any worsening of motor function.63
Sleep disorders in the SCA
A broad spectrum of sleep disorders is described among the SCAs, potentially representing widespread brain degeneration beyond the cerebellum. However, technical difficulties exist in measuring REM in those who have developed impaired eye movements and saccades, with this likely accounting for some of the previous reports of abnormal or absent REM sleep. Individuals with SCA2 mutations often report good sleep quality but have abnormal PSG, including reduced N3 sleep and REM sleep.64 NREM sleep appears to be the stage most severely affected in those with SCA3 and SCA6 mutations, with increased arousals from slow wave sleep and increased WASO.65 66 In contrast, a study of SCA10 patients found changes predominantly in REM sleep, including an increased number of REM sleep arousals.67
Overall, low rates (0%–10%) of PLMS have been noted in those with SCA1 and SCA6 mutations, while higher rates and symptomatic RLS are more often seen in SCA3 cohorts (77%).66 RLS is thought to reflect the basal ganglia involvement, with functional imaging supporting reduced striatal dopamine transporter activity in this region. Further investigation of the postsynaptic dopaminergic system has shown progressive loss of D2 receptors in the caudate, dorsal putamen and ventral striatum.68
Sleep-related breathing disorders have been identified in those with SCA2 and SCA6 mutations, including an increased CSA index (events/h=0.97) and AHI (p=0.022) in those with SCA2 mutations, and increased CSA events during sleep (p=0.024) and oxygen desaturations (p=0.03) in those with SCA6 mutations.66 The underlying causative mechanisms remain unknown, although vocal paralysis and dysphonia in combination with neurodegeneration of the supramedullary pathway and brainstem neurons have been proposed.69
Sleep disorders in WD and the impact of motor symptom treatment on sleep
Two video-PSG case–control studies have reported abnormal sleep architecture, with REM stages predominantly affected and a third demonstrating objective daytime sleepiness on MSLT and cataplexy-like epsiodes.70 71 There is some limited evidence for higher rates of RLS among those with WD compared with the general population, with 31% of a single cohort (n=42) meeting diagnostic criteria, and subsequent PSG demonstrating increased PLMS compared with controls (p=0.009).72 Although in a single case report, D-penicillamine resolved complaints of hypersomnia, while a case–control study identified higher rates of probable RBD found in those treated with D-penicillamine.71 73
Sleep disorders in NPC
The heterogeneous and non-specific clinical presentation of NPC challenges its diagnostic process, although secondary cataplexy is an important and discriminatory clinical feature. A single, small (n=5) PSG study has been performed to date, with evidence of reduced total sleep time (p<0.01), sleep efficiency (p<0.001), shorter sleep latency (p<0.01), increased (WASO, p<0.05) and decreased REM sleep (p<0.05).74 Reduced CSF hypocretin levels were also observed compared with controls, a feature also seen in narcolepsy and cataplexy, and consistent with hypothalamic dysfunction (figure 2C).74
Sleep disorders in anti-IgLON5 and the impact of disease treatment on sleep
IgLON5 is a recently described disorder in which sleep disturbance is pronounced, as well as gait instability and movement disorders. The underlying aetiology of this disorder is thought to be a combination of neuroinflammation and neurodegeneration, with postmortem evidence of hyperphosphorylated tau deposition.75 The sleep disorders described include REM and NREM sleep parasomnias, sleep-disordered breathing, namely stridor and obstructive apnoea, insomnia and hypersomnolence. Three PSG studies have identified altered sleep architecture, with evidence of undifferentiated NREM sleep and poorly structured N2 associated with abnormal behavioural manifestations.76–78 Treatment of IgLON5 with intravenous steroids show normalised sleep onset and N1 and N2 stages, although marked abnormal sleep reremerged 3 months later in spite of additional immunotherapy.78
Sleep disorders in dystonia
Disturbances to sleep have been reported in 40%–70% of dystonia cohorts, with insomnia and abnormal movements during the sleep (eg, RLS) most commonly described, although several studies have also reported few or no disturbances to sleep.79 80 A probable contributory role of the common psychiatric comorbidities (depression and anxiety in particular) to these sleep-related symptoms, is also likely and important to bear in mind. As such, varying changes in sleep architecture have been observed in distinct types of isolated idiopathic dystonia with PSG recordings of individuals with cervical dystonia (n=22), blepharospasm (n=3), Meige Syndrome (n=7) and in combined genetic forms, such as GCH1 mutation positive dopa-responsive dystonia (n=6), showing increased N1 phase (NREM) duration and reduced REM periods, compared with controls.79 81
Sleep in relation to the treatment of dystonia
Levodopa has been shown to benefit sleep in those with GCH1 mutation positive dopa-responsive dystonia, reducing rates of excessive daytime sleepiness, although also reducing total REM sleep.82 Trihexyphenidyl has been reported to increase wakefulness and decrease REM sleep, however, the drowsiness experienced at lower doses has limited its clinical use. Benzodiazepines are also often used in the management of dystonia, and as traditional hypnotics they decrease sleep latency but also suppress slow-wave sleep, lengthen REM sleep latency and can cause habituation with rebound insomnia. Also sometimes used, is gabapentin, an alpha 2 delta ligand that increases slow wave sleep, although when used during the day can be associated with drowsiness, fatigue and sedation. However, in children diagnosed with dystonia, gabapentin has been shown to reduce motor symptoms severity (n=69), as well as improve sleep quality and duration (p<0.01).83
Sleep disorders in ET
Sleep disturbances in ET are milder than those seen in PD, with a single PSG-based study showing few objective changes compared with controls, in contrast to those with PD, although examination of the phenotypic heterogeneity among the ET cohort found that those PSG changes observed were most prominent in those with mild cognitive impairment.84 A recent review, although using questionnaire only data, describes excessive daytime somnolence, RLS, insomnia and nocturia to be frequently reported by patients diagnosed with ET.85 Once again, psychiatric comorbidities (anxiety in particular) likely play some role in contributing to the symptoms described.
Sleep in relation to the treatment of ET
Commonly used in the treatment of ET, beta-blockers often cause or worsen nightmares and restless legs, with 12 studies to date reporting rates of sleep disturbance between 12% and 60%.86 Primidone has also been reported to have sedative side effects, including daytime sleepiness and tiredness (50%), often resulting in discontinuation of treatment (27%). As described above, gabapentin, also sometimes used in the management of tremor, carries with it side effects that include drowsiness, fatigue and sedation.
Sleep in TS
Many with TS have both insomnia and increased movements at night, during both REM and NREM stages, with studies of those with comorbid attention deficit hyperactivity disorder (ADHD) reporting rates of sleep disorders to be as high as 80%.87 Comparison with healthy controls has shown significant reductions in sleep efficiency (p<0.05) and total sleep time (p<0.01), while sleep latency (p<0.05) and WASO (p<0.001) were increased. In contrast, changes to the REM portion of sleep has varied between studies, with some reporting reduced periods (n=18/34), while others show increased or comparable levels to controls.88 Other studies have sought to separate TS patients with and without symptoms of ADHD, with reduced REM sleep latency and increased REM sleep observed in both groups, while those with ADHD symptoms also had evidence of insomnia, RLS/PLMS and NREM parasomnias.89 Interestingly, variants in the orexin-2/hypocretin-2 (OX2R) receptor gene, mutated in canine narcolepsy, have also been found to be present in those diagnosed with TS.90
Sleep in relation to the treatment of TS
Although in only a small cohort of patients with tics (n=3), abnormal NREM sleep and sleep apnoea resolved in two when treated with tetrabenazine.91 A recent review article also noted that those treated with antipsychotics (risperidone, aripiprazole and tiapride) were more likely to experience somnolence, sedation and sleep disturbance.92
Challenges in the field
As outlined above, increasing data suggests sleep disorders to be prevalent across the spectrum of movement disorders, although much of this work is limited by small, heterogenous cohorts. Larger studies often involve questionnaire-based assessments alone, while those that do involve PSG measurements vary in their approach with differences in single night versus multinight assessments. The night-to-night variability typical of all sleep patterns questions the validity of single-night studies, particularly those undertaken in unfamiliar environments. To counter this, portable devices are now widely used in the home setting, allowing longer periods of monitoring. Further work with these non-laboratory-based techniques is needed in order to establish more robust normative values on an individual and disorder-specific level.
While motor symptoms are usually considered as the first cause of sleep disturbance (eg, dystonia in PD), it is important to recognise that comorbidities such as mood disorders and pain also have the potential to impact sleep. The relationship between depression and disturbed sleep is complex and bidirectional, for example, insomnia may be a direct result of depression or secondary to the movement disorder. Although few studies have addressed this issue, some have reported more impaired sleep (sleep maintenance insomnia) in patients with PD and depression than those without depression. Treatment of anxiety and depression should be managed as effectively as possible and may help consolidate the sleep/wake cycle.
Another important area for future understanding, and a potential avenue for therapeutic intervention is the glial-lymphatic glymphatic) pathway, a highly organised fluid transport system moving cerebrospinal fluid (CSF) from the subarachnoid space to the periparenchymal interstitial fluid (ISF).93 Murine studies have shown that glymphatic function increases during the sleep, allowing for clearance of metabolic by products. Several studies have demonstrated that during disrupted or impaired sleep, these processes do not have sufficient time for normal clearance, leading to the accumulation of metabolites with subsequent impact on cognition and behaviour. Protein aggregates found in the glymphatic system include alpha-synuclein, as well as amyloid β observed in Alzheimer’s disease. Dopamine and norepinephrine also appear to contribute to regulating ISF flow within the glymphatic system, resulting in feed-forward enhanced neurotransmission increasing dopamine and norepinephrine neurotransmission during wake states.94
Assessment and treatment considerations
A detailed discussion of the management of different sleep disorders is beyond the scope of this review, and only general guidelines are provided. An initial clinical history from both patient and bed partner wherever possible, alongside simple questionnaires, often allows good differentiation of many sleep disorders. Consideration of the impact of motor symptom onset, severity and prescribed medication is also important. Comorbid medical conditions that may cause or contribute to sleep disorders should be considered, for example, primary treatment of psychiatric disorders may improve sleep-related symptoms. Sleep hygiene should be reviewed (eg, caffeine use) and simple behavioural interventions considered in the first instance. Cognitive Behavioural Therapy (CBT) is recommended as standard therapy for the long-term management of insomnia, with digital forms of this treatment becoming increasingly available to broaden access and availability. Pharmacotherapy is effective in the treatment of moderate or severe RLS, with several clear treatment algorithms available, while extensive evidence supports the use of CPAP in the management of OSA, again aided by standardised therapy guidelines.
This review demonstrates that many patients with movement disorders also experience symptoms related to sleep disorders. Dysfunction of the medulla, raphe nuclei, pons and basal ganglia potentially form a shared underlying mechanism for both motor and sleep pathophysiology. For some, there are specific patterns to the sleep problems but for many, changes to sleep are multifactorial, involving medication side effects, motor and non-motor symptoms, suggesting the need for a systematic approach to assessment. Further work is needed to better understand the aetiology of sleep disturbances across distinct movement disorders in order to facilitate more targeted and effective therapeutic strategies, and to understand whether targeting poor sleep leads to clear improvements in motor outcomes. Future studies should be directed towards combining limited PSG assessment with far wider application of wearable or mobile home devices in order to identify robust digital biomarkers of utility in diagnosis and symptomatic monitoring.
Servier Medical Art Images were used in the generation of Figure 2.
Contributors GAB: Conception, acquisition, analysis and interpretation of the data, drafting and revising critically for important intellectual content. EKH: acquisition, analysis and interpretation of the data, drafting and revising critically for important intellectual content. AF: drafting and revising critically for important intellectual content. MAT: drafting and revising critically for important intellectual content. TL: drafting and revising critically for important intellectual content. KNA: Conception, acquisition, analysis and interpretation of the data, drafting and revising critically for important intellectual content. KJP: Conception, acquisition, analysis and interpretation of the data, drafting and revising critically for important intellectual content.
Funding GAB is funded by a KESS2, European Social Fund and Cardiff University PhD Studentship. KJP is funded by an MRC Clinician-Scientist Fellowship (MR/P008593/1).
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.