1. Introduction
Sleep is an important disease-modifying factor in PD. Sleep disturbance can cause altered sleep neural circuits, neurodegeneration, inflammatory reactions, impaired nocturnal brain oxygenation, and irregular proteostasis, which can provoke the development of α-synucleinopathies, further increasing the risk of PD
[1]. The association between sleep disturbances and specific cognitive functions was evaluated in advanced PD patients. The results indicated that patients with sleep complaints performed worse than those without sleep complaints in terms of attention, reasoning, executive functions, and verbal fluency, but not memory. Also, PD-specific motor problems at night are correlated with neuropsychological dysfunctions in all studied cognitive domains, excluding memory. Additionally, no relationship was observed between daytime sleepiness and cognitive impairment
[2]. Sleep disturbances in PD can also be due to other contributing factors, such as the side effects of dopaminergic drugs, other medications, comorbidities, genetic factors, lifestyle, and impulse control disorders
[3]. In addition to disorders like bradykinesia, rigidity, tremors, and postural instability, and conditions like loss of dopaminergic neurons, sleep disturbances are seen in PD patients. The regulation and balance of sleep and wakefulness require the highly integrated functions of multiple brain regions and neurotransmitters. Parkinson’s disease-associated sleep disorders and their characteristics are illustrated in (
Figure 1).
Figure 1. The illustration describes various sleep disorders associated with Parkinson’s disease and their causes and neurological changes. EDS: excessive daytime sleeping; DA: dopamine; REM: rapid eye movement; RBD: REM sleep behavior disorder; OSA: obstructive sleep apnea; RLS: restless legs syndrome; SWD: sleep–wake disorder; CLOCK: circadian locomotor output cycles kaput. (Figure created using
BioRender.com; accessed on 27 June 2023).
A wide range of sleep disorders, e.g., insomnia, sleep fragmentation, excessive daytime sleepiness (EDS), sleep-related breathing disorders (SRBDs), restless legs syndrome (RLS), nightmares, circadian-rhythm-related sleep–wake disorders (CRSWDs), obstructive sleep apnea (OSA), rapid eye movement (REM), and REM sleep behavior disorder (RBD), were observed in PD patients
[4]. Visual hallucinations, psychosis, autonomic disturbances, dementia, and abnormal behaviors during sleep, such as dream enactments and excessive muscle twitching during REM sleep, are the characteristic features of RBD. PD patients with RBD also have visual hallucinations, psychosis, autonomic disturbances, and dementia
[5]. Sleep disturbances like RBD are commonly considered a prodromal stage of neurodegeneration diseases like PD, Lewy body dementia (LBD), and multi-system atrophy
[6]. The evaluation of early signs, such as sleep disturbances, especially RBD-like symptoms, and CSF α-Syn levels, provides an understanding of the central causes, biomarkers, and strategies to develop effective treatment for PD.
The relation between sleep disturbance and α-Syn levels in cerebrospinal fluid (CSF) was measured in PD, prodromal PD, and healthy subjects. The study showed that sleep disturbance was high in prodromal PD, followed by PD and healthy subjects. The CSF α-Syn levels were significantly lower in PD subjects with RBD than in subjects with only PD
[7]. Wang et al. investigated the associations of sleep disorders and CSF α-Syn levels among healthy controls, prodromal PD patients, and early PD patients. Their study demonstrated that sleep disorders lowered CSF α-Syn levels, with reduced sensorimotor function and impaired motor function. It has been hypothesized that PD-RBD subjects show increased neurophysiological abnormalities compared with PD patients without RBD
[8]. Mutation in
GBA1 variants has effects on CSF α-Syn profiles. Hence, CSF α-Syn acts as a biomarker depending on mutation severity. The results of a large PD cohort study revealed that CSF α-Syn levels were reduced with respect to
GBA1 mutation. In addition to
GBA1 mutation, age is an important factor, where older age is associated with increased CSF α-Syn levels
[9].
Depending upon the reported symptoms of sleep disorders, sleep behavior differs according to gender differences. Studying the importance of gender differences in sleep disorders can help improve the diagnosis, treatment, and prevention of sleep disorders and comorbid conditions
[10]. Various factors, including hormonal and physical changes in a woman’s lifespan, can influence her sleep health. Certain sleep disorders like OSA and insomnia are more prevalent in women during specific periods
[11]. On the other hand, narcolepsy, REM, and RBD are predominant in men, and the risk of RLS is double in women compared with men
[10].
2. Excessive Daytime Sleepiness (EDS)
Sleep disturbances and wakefulness are the most common non-motor symptoms of PD. EDS affects 16 to 55% of PD patients, and the severity of EDS increases with disease duration and severity
[12]. EDS is the second most prevalent, troublesome sleep-disorder symptom in PD, and it could be the preclinical marker for the development of PD
[13]. EDS can be defined as the sleep trend or falling asleep excessively during various activities like reading, eating, and other circumstances. The progression of EDS equals the rate of progression of neurodegeneration
[14]. EDS can be a prodromal risk factor for further neurodegeneration and increased risk of PD. EDS in PD showed a correlation with alterations in cerebral regions, such as the hypothalamus and brainstem regions; damage to the ascending arousal system; and changes in neurotransmitter and neuropeptide balances, especially GABAergic, orexinergic, and serotonergic systems
[15][16].
In a study, EDS was measured in baseline PD patients using the Epworth Sleepiness Scale (ESS) for up to 3 years. The results indicated that the ESS score was increased from baseline to the third year in the PD group, with no changes in healthy controls. Conclusively, it was found that EDS significantly increased over time in PD relevant to the dosage of dopaminergic therapy but remained unchanged in healthy controls. A 123I ioflupane dopamine transporter imaging (DaTscan) study showed that the biological correlates of PD and EDS exhibited major dopaminergic dysfunction in brain regions like contralateral and ipsilateral caudate, and contralateral putamen compared with PD patients without EDS
[17].
IPD patients were evaluated for nocturnal disturbance, EDS, and RBD symptoms with neuropsychological testing and self-report questionnaires. The study results demonstrated that patients with EDS showed significantly poor working memory, and RBD patients showed poor working memory and verbal fluency. Brain regions like medial temporal regions and subcortical regions were found to be associated with nocturnal disturbances, memory consolidation, and slow processing speed
[18].
3. Insomnia
Insomnia was reported in 80% of PD patients, with difficulty in falling and staying asleep, and poor sleep quality. The frequency of insomnia is directly proportional to the advancement of motor stages in PD
[19][20]. Primary insomnia and secondary insomnia develop due to depression, nocturnal worsening, and motor and non-motor dysfunctions
[1]. PD insomnia and hyposomnia pathophysiology involves tremors, RLS, night-time cramps, dystonia, dyskinesia, and non-motor symptoms like psychiatric and autonomic dysfunctions
[16]. The studied pathophysiological factors of insomnia include circadian-rhythm disruption, mutation in circadian locomotor output cycles kaput (
CLOCK) genes, and neurochemical imbalances. In addition, disturbances in cortisol secretion and lesions in the sleep regulatory systems of the brain also cause insomnia
[21]. Insomnia in PD might occur for various reasons, including the neurodegeneration of sleep regulation centers like the hypothalamus and brain stem, and continuous medications like dopaminergic drugs
[22]. A longitudinal follow-up study suggested that the frequency of insomnia subtypes was changed in early PD patients. Also, the frequency of sleep-maintenance problems increased with dopamine agonists
[23].
Insomnia in PD has been shown to increase cognitive decline and mental illness and to exert negative impacts on the health of individuals. Basal ganglia neural circuits and dopaminergic neurons in the SN and VTA are involved in sleep regulation. The effects of lesions in the basal ganglia and SN on sleep were evaluated in PD patients and animal models of PD
[24]. Given the involvement of the basal ganglia in sleep maintenance, recent techniques use basal ganglia neuromodulation to ameliorate PD insomnia
[24].
4. Rapid-Eye-Movement (REM) Sleep Behavior Disorder (RBD)
RBD is one of the prodromal symptoms of PD. RBD showed visual hallucinations, dream enactments, muscle twitching during REM sleep, psychosis, autonomic disturbances, and dementia
[4]. Abnormal dream enactments characterize RBD during REM sleep with activities such as punching, waving, swinging wildly, or jumping out of bed
[25]. Besides dream-enactment behaviors, clinical characteristics like severe cognitive and motor impairment, higher sleeplessness, and hallucinations were also observed in RBD
[26]. Any changes in the brain stem regions that control motor inhibitions during REM sleep could result in RBD. Mesencephalic, pontine, or medullary reticular lesions were observed in animal models in REM sleep without atonia
[25].
Studies in PD patients with RBD and cognitive deficits showed functional disturbances in the dorsolateral prefrontal cortex and posterior cortical regions
[27][28]. RBD with non-motor symptoms and constipation are the predictors of the conversion of RBD into parkinsonism
[29]. REM without atonia, with increased sustained and intermittent electromyographic (EMG) activity, is the hallmark neurophysiological symptom of RBD. Any impairment or imbalance in the neural circuits that control the excitatory and inhibitory signals results in episodic sleep disturbances in RBD. REM without muscle atonia can be differentiated into iRBD and secondary RBD, which occurs in PD patients
[1]. PD patients with the RBD phenotype were predominantly older males. They possessed akinetic–rigid dominant motor disease, autonomic dysfunction, increased falls, EDS, and increased risk of developing future dementia and visual hallucinations
[30]. An analysis using psychiatric/clinical questionnaires and neuropsychological assessment in PD patients with probable RBD and healthy controls revealed that RBD affects 33–46% of PD patients and poses the risk of neuropsychological deficits such as poorer cognitive, functional, and emotional outcomes
[31].
5. Obstructive Sleep Apnea (OSA)
OSA and PD coincide with one another. The incidence of OSA in PD or the PD-predisposing condition of OSA is high. Large-scale population follow-up studies describe the increased incidence of OSA in PD
[32]. SRBDs are the least commonly studied sleep disturbances associated with PD. Patients with postencephalitic parkinsonism showed changes like irregular respiratory patterns, hypoventilation, and nocturnal respiration worsening. OSA is a common comorbidity, and obstructive, central, and mixed apnea types have been documented in PD
[33]. OSA is characterized by episodic cessation of breathing due to partial (hypopnea) or complete (apnea) recurrent obstructions in the upper airway, resulting in periodic arrests in breathing during sleep (
Figure 2A). These disturbances in breathing consequentially cause intermittent hypoxia and frequent arousal during OSA
[34]. The symptoms of OSA are commonly associated with sleep apnea, such as cognitive impairment, sleepiness, nocturia, and snoring
[3].
Figure 2. (
A) Obstructive sleep apnea (OSA): The episodic breathing session and repeated obstructions in the upper airway worsen nocturnal respiration and sleep. OSA produces irregular respiratory patterns, hypoventilation, nocturnal worsened respiration, and oxidative stress due to the resaturation and desaturation of oxygen levels, and produces damage to dopaminergic neurons. (
B) Restless legs syndrome (RLS): Hypo-functioning of dopamine signaling due to reduced dopamine subtype 2 receptor (D2R) expression in the CNS. Reduced peripheral blood flow causes peripheral hypoxia, which leads to urges to move legs and causes defects in neurological sensorimotor functions. (Figure created using
BioRender.com; accessed on 27 June 2023).
Repeated oxygen desaturation and resaturation that occur during sleep can result in the production of reactive oxygen species that initiate oxidative stress and certain molecular events that interfere with the cellular proteins, lipids, and mitochondrial functions that damage the dopaminergic neurons in the brain and produce neurodegeneration in PD
[35]. Age is the major risk factor for the development of OSA. Other than age, infection in the upper airway, pulmonary dysfunction, and some PD-associated symptoms (including restrictive lungs due to chest-wall rigidity, postural instability, autonomic dysfunction, and loss of neurons in the brain sites responsible for sleep physiology) contribute to OSA
[36].
Clinically, OSA could cause other sleep-related issues, like EDS, nocturia, non-refreshing sleep, and memory problems
[34]. A recent meta-analysis stated the role of the severity of OSA in establishing cognitive disturbance in PD patients. PD patients with OSA scored significantly lower on the Montreal Cognitive Assessment (MoCA) and Mini-Mental State Examination (MMSE). The results suggest that OSA can worsen cognitive abilities like working memory, attention, and executive functions independently of PD-associated cognitive decline due to other factors, like sleep fragmentation, hypoxemia, neuroinflammation in the brain stem nuclei, and malfunction in certain brain regions
[37]. Another meta-analysis revealed that OSA acts as a risk factor for PD. Chronic intermittent hypoxia due to OSA induces oxidative stress and inflammatory pathways, which result in PD pathophysiology
[38].
Meng and colleagues studied the association between OSA and motor dysfunction, and the effect of OSA treatment. PD patients with OSA were treated with continuous positive airway pressure (CPAP), and motor symptoms were assessed using the Movement Disorder Society-Sponsored Unified Parkinson’s Disease Rating Scale (MDS-UPDRS) and Timed Up and Go (TUG) at 3, 6, and 12 months of follow-up. The results showed that PD-OSA individuals showed higher MDS-UPDRS scores at baseline and CPAP treatment stabilized the motor function over 12 months
[32]. Upper-airway dysfunction was reported in some PD patients, which shows that laryngopharyngeal motor dysfunction is one of the factors that cause obstructive phenomena of upper-airway dysfunction in OSA-PD patients
[39]. An observational study in 239 Chinese PD patients with and without OSA revealed certain characteristic features of the disease, such as age and male gender, which are the risk factors for OSA in PD. PD patients with RBD and higher levodopa equivalent doses showed a lower risk of developing OSA
[40]. Certain studies in relation to sleep disorders in PD are given in
Table 1.
Table 1. Representative studies on sleep disorders associated with Parkinson’s disease.
No. |
No. of Subjects |
Age |
Study Conditions |
Study Findings |
Ref. |
Rapid-eye-movement sleep behavior disorder (RBD) |
1 |
PD-RBD (n = 20); PD without RBD (n = 20) |
Age- and gender-matched with controls |
PD patients with and without RBD were evaluated for neurophysiological abnormalities with single- and paired-pulse TMS, and RMT, CMCT, SP, SICI, and ICF were recorded. |
PD-RBD patients showed reduced intracortical facilitation, reduced glutaminergic transmission, and enhanced GABAergic transmission. |
[4] |
2 |
PD subjects (n = 360); prodromal PD subjects ((n = 46); subjects displaying RBD behaviors); controls (n = 169) |
Mean age: 61.24 years for controls, 61.31 years for PD patients, 68.20 years for prodromal PD subjects |
The association of RBD and the level of CSF alpha-synuclein was evaluated. |
PD individuals with probable RBD had significantly lower alpha-synuclein levels in CSF. No significant association between daytime sleepiness and CSF alpha-synuclein levels was found. |
[8] |
3 |
PDGBA (n = 80); PDGBA-wildtype (n = 80); controls (n = 39) |
59 ± 12 years for controls, 64 ± 10 years for PD-GBA patients, 66 ± 10 years for PD-GBA risk-variant patients |
PD patients with and without GBA1 mutation were screened for total CSF alpha-synuclein. |
PDGBA patients showed early-onset cognitive decline, high chance of RBD development, and reduced total CSF alpha-synuclein. |
[9] |
4 |
Idiopathic RBD patients (n = 1061); controls (n = 3086) |
- |
The role of GBA variants in the risk of developing idiopathic RBD and development of neurodegeneration was studied. |
Individuals with GBA variants had increased risk of idiopathic RBD, and the rate of neurodegeneration also increased in GBA-variant individuals. |
[41] |
5 |
RBD patients (n = 261); controls (n = 379) |
67.2 ± 9.2 years for RBD patients, 58.9 ± 12.3 years for controls |
RBD patients and controls were screened for PD-associated SNPs and their effects on RBD and progression of synucleinopathies. |
Data from 56 RBD patients showed that 19 developed neurodegeneration during the follow-up period, 9 were diagnosed with PD, and 10 had DLB. The SCARB2 rs6812193 SNP and the MAPT rs12185268 SNP were associated with RBD, and the carriers of these SNPs progressed to synucleinopathies. A few patients with the USP25 rs2823357 SNP demonstrated faster progression to synucleinopathy from RBD. |
[42] |
Excessive daytime sleepiness (EDS) |
6 |
PD patients (n = 400) |
- |
Five-year hospital-based cohort study to analyze the risk factors of EDS in PD using SCOPA-SLEEP-DS scores. |
The proportion of EDS in PD increased with longer follow-up. In total, 43% of PD patients had EDS at baseline. A total of 46% of patients without EDS at baseline developed EDS during follow-up. |
[12] |
7 |
Unmedicated PD patients (n = 423); Controls (n = 196) |
- |
EDS was assessed using the ESS. Clinical, biological, and imaging variables were assessed. |
EDS was developed during the follow-up. EDS in PD was associated with autonomic dysfunction, depression, and anxiety. EDS was also associated with presynaptic dopaminergic dysfunction. |
[17] |
8 |
Idiopathic PD patients (n = 101); unmedicated (n = 12); Patients with levodopa monotherapy (n = 29); Patients with dopamine agonist monotherapy (n = 5); Patients with levodopa plus adjuvant agent therapy (n = 55); Patients, who taking anti-depressants (n = 26), Patients, who taking benzodiazepines (n = 15) |
67.3 ± 8.0 years for controls, 65.9 ± 9.5 years for all PD patients, 67.9 ± 9.0 years for PD-RBD patients, and 62.8 ± 9.6 years for PD-non-RBD patients |
All patients’ neuropsychological functioning was assessed with standard tests using Wechsler Adult Intelligent Scale-III, Cambridge Neuropsychological Test Automated Battery, and Wechsler Memory Scale-III; daytime sleepiness was assessed with the SCOPA-day, and EDS, with the ESS. |
Patients with greater levodopa dose equivalents showed greater nocturnal disturbances and daytime sleepiness but not RBD symptoms. EDS was a significant predictor of slow processing speed, working memory, and verbal frequency performance. |
[18] |
9 |
Patients with EDS receiving stable dopaminergic therapy without cognitive impairment or primary sleep disorder (n = 31) |
- |
Safety and efficacy of light therapy on EDS were evaluated. Participants were randomly assigned in a 1:1 ratio to receive bright light and dim light (as control) twice daily in 1-hour intervals for 14 days. |
Bright-light therapy significantly improved EDS scores. Bright and dim light improved sleep quality based on the Pittsburg Sleep Quality Index. Bright-light therapy improved mean sleep metrics and sleep quality. |
[43] |
Insomnia |
|
10 |
Drug naïve PD patients (n = 182); Controls (n = 202). |
67.5 ± 9.2 years for patients and 66.2± 9.6 for controls |
Participants were assessed for insomnia with the Stavanger Sleepiness Questionnaire and Parkinson’s Disease Sleep Scale before treatment initiation and after 1, 3, and 5 years. |
Insomnia prevalence was not higher in PD patients at the 5-year follow-up. Sleep-maintenance problems increased, and solitary-sleep-initiation problems decreased after 5 years. |
[23] |
11 |
PD Patients with insomnia randomized for three-arm six-week randomized pilot study (n = 18); Placebo (n = 6); CBT with BLT (n = 6); Doxepin (10 mg/daily) (n = 6). |
- |
This three-arm, six-week randomized pilot study assessed non-pharmacological and pharmacological treatment outputs in PD patients with insomnia. Sleep outcomes were measured using insomnia scales, sleep diaries, actigraphy, and clinical global impression. |
Doxepin improved the scores in Insomnia Severity Index, SCOPA-night, and Pittsburgh Sleep Quality Index-Sleep Disturbances Subscale. Doxepin reduced the score on the Fatigue Severity Scale and improved the scores on the Montreal Cognitive Assessment. Non-pharmacological treatment reduced the Insomnia Severity Index. |
[44] |
12 |
Patients under 65 received 3 mg eszopiclone or matching placebo at night. Patients 65 or older received 2 mg of eszopiclone or placebo at night (n = 30). |
35 to 85 years |
Patients were equally randomized to eszopiclone and placebo for 6 weeks. Patients with other primary sleep disorders were excluded. Total sleep time, wake after sleep onset, and number of awakenings were measured. |
Significant differences were found in the number of awakenings, sleep quality, and wake after sleep onset, favoring eszopiclone. Eszopiclone did not increase the total sleep time but improved the sleep quality compared with the placebo group. |
[45] |
Obstructive sleep apnea (OSA) |
13 |
PD patients with OSA (n = 67). |
64.7 years |
Patients were treated with CPAP, and motor symptoms were assessed using the MDS-UPDRS and TUG with a follow-up time of 3, 6, and 12 months. |
CPAP treatment stabilized the motor function over 12 months of follow-up treatment. |
[32] |
14 |
PD patients (n = 239); PD (n = 66) with OSA including mild (n = 34), moderate (n = 16), severe sleep apnea (n = 16); PD without OSA (n = 173). |
n = 66 PD patients with OSA had a mean age of 45 years; n = 173 PD patients without OSA had a mean age of 81 years |
Participants underwent assessments to examine disease severity, polysomnography characteristics, and non-motor symptoms. |
Binary logistic regression analysis showed that age and male gender were risk factors for OSA. RBD and higher levodopa equivalent dose were protective factors against OSA. Thus, OSA in PD was lower in PD patients with RBD. And OSA could increase excessive day sleeping in PD patients. |
[40] |
15 |
Subjects were divided into OSA and non-OSA groups (n = 95). |
69.1 ± 3.4 years |
Subjects were evaluated with protocols that included polysomnography, BMR, and body composition. BMR was evaluated in the morning after polysomnography. |
Patients with OSA had higher values in weight, fat mass, arousal, and AHI. The OSA group had lower REM sleep. |
[46] |
16 |
Idiopathic PD patients (n = 67) |
Mean age of 64.4 years |
Idiopathic PD patients were recruited from a movement-disorder clinic. OSA was defined using the AHI. The H&Y Scale and MDS-UPDRS were used to assess PD severity. And NMSs were assessed with the MoCA, ESS, Fatigue Severity Scale, Apathy Scale, BDI, HDAS, and PDSS. |
OSA in PD was associated with sleepiness and cognitive dysfunction. Treatment for OSA could improve excessive sleepiness and cognitive dysfunction in PD. |
[47] |
Restless legs syndrome (RLS) |
17 |
PD patients with parkin mutations (n = 11); Sex matched IPD patients (n = 11) |
PD patients with parkin mutations were aged 35–60 years and were from seven families; IPD patients were aged 51–65 years. |
Patients with parkin mutations and IPD patients were compared to evaluate the sleep–wake phenotype using the UPDRS, ESS, MMSE, and RLS Study Group Rating Scale; a sleep specialist interview; and video-polysomnography. |
Parkin patients showed sleep phenotypes like insomnia and RLS, and neuronal loss. Parkin-mutation patients had all polygraphical abnormalities reported in IPD. Two Parkin siblings had central hypersomnia and normal night-time sleep. |
[48] |
18 |
PD patients (n = 74); Drug-I patients (n = 16); Patients treated with levodopa/aromatic L-amino acid decarboxylase inhibitor, monoamine oxidase B inhibitor and amantadine (n = 58) |
65.5 ± 9.1 years |
Patients were assessed for RLS based on the diagnostic criteria of the International RLS Study Group revised in 2003. |
The frequency of RLS in PD patients was higher than the general RLS population. PD patients with RLS had worse sleep quality, anxiety, depression, and autonomic disturbances. |
[49] |
19 |
Idiopathic PD patients (n = 108); Matched controls (n = 424) |
≥35 years |
This comparative study analyzed the prevalence of RLS in PD patients and investigated the quality of life, nutritional status, and clinical characteristics using IRLSSG, PD severity scales, psychiatric features, nutritional status, and quality of life. |
RLS was significantly more common in IPD patients than controls. PD patients with RLS suffered from more anxiety, and worse nutritional status and quality of life. RLS was found to be correlated with psychiatric problems and cognitive impairment. |
[50] |
20 |
PD patients (n = 225) |
- |
RLS was diagnosed using IRLSSG criteria. Orthostatic vital signs and blood pressure were monitored. |
PD patients with RLS showed nocturnal/supine hypertension and fluctuations in blood pressure and some sleep dysfunctions. RLS could be a determinant for neurocirculatory abnormalities. |
[51] |
21 |
Drug naïve early, unmedicated PD patients (n = 200); Controls (n = 173) |
Age- and gender-matched controls |
Subjects were assessed for RLS with structured interviews, clinical examinations, and blood samples. RLS was diagnosed using IRLSSG criteria. |
PD patients reported leg restlessness, which was 3-fold greater in patients than in controls, which could indicate a relative risk for RLS. |
[52] |