Sleep disorders in Parkinson’s disease: evaluating pharmacological treatments through systematic review and meta-analysis Authors: Jingqi Li, Runtang Zhou, Jialin Deng, Hong Zhou, Jun Hu, Xinzheng Wang Published: 2025 ABSTRACT Sleep disorders are the most common non-motor symptoms in patients with Parkinson’s disease (PD) and can significantly impact quality of life. However, the pharmacological treatments for sleep disorders in patients with PD remain controversial. Databases (PubMed, Embase, Web of Science, and the Cochrane Library) were searched up to 20 January 2025. Randomized controlled trials were included if they investigated the efficacy of pharmacological interventions for sleep disorders in populations with PD. The weighted mean difference (WMD) and standardized mean difference (SMD) were used as effect size. The random-effects model was applied due to the heterogeneity between studies. The certainty of evidence was assessed using the Grading of Recommendations Assessment, Development and Evaluation tool. A total of 23 studies were included in the systematic review and meta-analysis. Seven trials focused on the treatment of rapid eye movement sleep behavior disorder (RBD), seven on excessive daytime sleepiness (EDS), six on insomnia, and three on sleep disorders that were not specifically classified. The meta-analysis showed that pharmacological interventions were effective in improving EDS (SMD: -0.36; 95% CI: -0.63, -0.10) and insomnia in patients with PD (SMD: -0.53; 95% CI: -0.82, -0.24). However, pharmacological interventions did not show significant benefits for managing RBD (SMD: -0.19; 95% CI: -0.54, 0.16). The certainty of evidence across all studies ranged from very low to moderate. In population with PD, pharmacological interventions were effective in improving EDS and insomnia. However, the management of RBD with pharmacological treatments remains challenging. In the future, strong evidence is needed to address this challenge.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12883-025-04526-5. Introduction Sleep disorders are prevalent across neurodegenerative diseases and are increasingly recognized as both consequences of and contributors to disease progression [1–3]. Sleep disorders are heterogeneous in nature, comprising various clinical manifestations [4, 5]. For example, insomnia, often measured by the Pittsburgh Sleep Quality Index (PSQI) or the Insomnia Severity Index (ISI), refers to difficulty falling or staying asleep [6]. Excessive daytime sleepiness (EDS) describes a reduced ability to stay awake during the day [7]. In patients with multiple sclerosis, poor sleep quality negatively impacts cognitive function and daily living [8, 9]. Among individuals with Alzheimer’s disease, the prevalence of sleep problems exceeds affects more than half [10]. Likewise, for patients with Parkinson’s disease (PD), sleep disturbances are the most common non-motor symptoms, affecting up to 80% of patients and contributing to decreasing quality of life and accelerating PD progression [11, 12]. In particular, rapid eye movement (REM) sleep behavior disorder (RBD), a condition involving dream-enacting movements during sleep, is associated with early brain changes and the accumulation of α-synuclein in locus coeruleus and thalamus [13]. In summary, these studies highlight the necessity for the management of sleep disorders in patients with PD. Currently, therapeutic strategies for sleep disorders in patients with PD include pharmacotherapy, cell therapy, and long-term exercise therapy [13, 14]. Pharmacological interventions remain a primary approach for the management of sleep disturbance. A variety of agents, including dopaminergic medications, sedatives, and antidepressants, have shown positive effects in improving sleep quality and alleviating EDS [15]. However, due to the varied classifications of sleep disorders and the diversity of therapeutic drugs, the present evidence on the pharmacological management of sleep disorders, particularly in patients with PD, remains controversial. Specifically, in the study by Litvinenko et al., polysomnography (PSG) parameters were improved after eight weeks of melatonin administration among patients with PD [16]. In contrast, slow-release melatonin showed no significant efficacy in managing RBD in patients with PD [17]. Moreover, the therapeutic effects of different pharmacological agents can vary within the same type of sleep disorder. Among PD patients with EDS, modafinil showed significant improvement, whereas solriamfetol did not yield a statistically significant effect [18]. Similarly, rivastigmine was associated with a reduction in the frequency of RBD episodes [19], while nelotanserin had a limited effect on REM sleep events [20]. The previous research investigated pharmacological treatments for sleep disturbances in PD, including 10 RCTs and 338 participants [21]. However, their findings were limited by the small sample size and might not be up to date with the current evidence [21]. Although recent study in Alzheimer’s disease has explored pharmacological interventions for sleep disorders, their conclusions cannot be readily applied to PD [10]. For patients with PD, sleep disturbances are tied to α‑synuclein pathology, dopaminergic degeneration, and disruption of arousal‑regulating circuits [22, 23]. In contrast, sleep dysfunction in Alzheimer’s disease is closed to amyloid pathology, cortical degeneration, and cholinergic deficits [24, 25]. To bridge these gaps, we conducted the present study by incorporating a larger body of recent evidence. Through this systematic review and meta-analysis, we aim to provide practical insights into clinical practice and optimize therapeutic strategies for the management of sleep disorders in patients with PD. Methods Literature search This systematic review and meta-analysis followed the PRISMA guideline [26], and its protocol was registered with PROSPERO (CRD42025583494). Two researchers independently searched PubMed, Embase, Web of Science, and the Cochrane Library databases to identify relevant literature. The third researcher was consulted to solve any disagreement. Searches were limited to January 20, 2025. The following search terms and Boolean operators were used: (“Sleep Disorder” OR “Sleeper Syndrome” OR “Sleep Phenotype” OR “Sleep-Related Neurogenic Tachypnea”) AND (“Parkinson” OR “Parkinson’s disease”) AND (randomized controlled trial). The detailed retrieval strategies for each database can be found in Table S1. Inclusion and exclusion criteria Following duplicate removal, two investigators independently performed title and abstract screening, with a third reviewer responsible for resolving discrepancies. We retrieved the full texts of eligible studies for further evaluation. The following criteria were used to select studies: (1) Patients with PD were diagnosed according to the validated criteria, such as the International Parkinson and Movement Disorder Society and UK Parkinson’s Disease Society Brain Bank Criteria. (2) Patients with sleep disorders or poor sleep quality were defined by validated sleep rating scales, such as Epworth Sleepiness Scale (ESS) or PSQI. (3) Study outcomes included the efficacy of pharmacological interventions for sleep disorders in populations with PD. (4) Studies were randomized controlled trials. Studies were excluded according to the following criteria: (1) Patients suffered from other neurological diseases, such as Alzheimer’s disease, multiple sclerosis, and Lewy body dementia. (2) Patients with psychiatric conditions were excluded, such as major depressive disorder and bipolar disorder. (3) Studies that used multiple drugs for the treatment of sleep disturbances in patients with PD. (4) Studies involving a very small number of samples (less than 10). The details of the excluded studies and specific reason were presented in Table S2. Data extraction and quality assessment The following data were extracted: first author, publication year, country, study design, sample size, the method of diagnosing sleep disorder and PD, sleep assessment, improvement score for sleep disturbance in the control and experimental groups. Two independent researchers evaluated the quality of included studies using the Cochrane Risk of Bias Tool. In addition, we assessed the certainty of evidence using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach [27]. Certainty ratings were determined based on five domains: risk of bias, inconsistency, indirectness, imprecision and other relevant considerations. The certainty of evidence from randomized controlled trials was initially rated as high and could be downgraded due to risk of bias, inconsistency, indirectness, imprecision or small sample size. Statistical analysis All analyses were conducted using Stata 14. The mean and standard deviation (SD) from each study were pooled to calculate effect sizes and 95% confidence intervals (CIs). Following previous research [28], the use of the standardized mean difference (SMD) allows the combination of different validated sleep scales that assess the same category of sleep disorder. When the same scale was used across studies, the weighted mean difference (WMD) was employed as the effect measure; if different scales were used, the SMD was applied instead. Given the heterogeneity in sleep disorder types among people with PD, a random-effects model was adopted to derive more conservative estimates. Heterogeneity was assessed using the I2 statistic, with thresholds interpreted as follows: 25% (low), 50% (moderate), and 75% (high) heterogeneity. Sensitivity analysis was conducted to evaluate the robustness of results by sequentially removing each study. Publication bias was assessed with funnel plots and Egger’s test. When the outcomes of studies were not suitable for meta-analysis, a qualitative evaluation of the study results was conducted. Results Characteristics of included studies A total of 1,686 studies were identified through database searching. Of these, we read 46 articles in full and excluded 23 studies. Finally, we included 23 studies for systematic review and 20 for meta-analyses (Fig. 1). These 23 trials included 1073 patients, with seven trials focusing on the treatment of RBD, seven on the treatment of EDS, six on the treatment of insomnia, and three on the treatment of sleep disorders that were not specifically classified. Table 1 presents the characteristics of included studies. Fig. 1The PRISMA flow chart The PRISMA flow chart Table 1Description of included studiesStudy/YearCountryStudy designParticipantsPrimary outcomeControl group vs. Experimental groupDose, Route and DurationMain findingsShin 2019 [45]South KoreaRandomized, double-blind, placebo-controlled trial40CGI-ImprovementPlacebo vs. Clonazepam0.5 mg/day, take at bedtime, and 4 weeks.No significant differences were observed between the clonazepam and placebo groups.de Almeida 2021 [44]BrazilRandomized, double-blind, placebo-controlled trial36RBD events, CGI-Improvement and CGI-SeverityPlacebo vs. Cannabidiol75–300 mg/day, take after dinner, and 12 weeks.Cannabidiol did not significantly reduce RBD symptoms compared to placebo, but a temporary improvement in sleep satisfaction was observed in the Cannabidiol group during the early weeks of the trial.Kataoka 2024 [32]JapanRandomized, single-blind, placebo-controlled trial69Sleep efficiencyPlacebo vs. Zonisamide25 mg/day, take at bedtime, and 4 weeks.No significant differences were observed between the zonisamide and placebo groups, but improvements in WASO and SOL were observed in the zonisamide group.Meloni 2022 [31]ItalyRandomized, double-blind, placebo-controlled, crossover trial18CGI, RBD frequency, REM sleep without atonia percentagePlacebo vs. 5-hydroxytryptophan50 mg/day, take daily, and 4 weeks.5-HTP increased REM sleep but did not significantly reduce RBD episodes.Gilat 2020 [17]AustraliaRandomized, double-blind, placebo-controlled, parallel-group trial30RBD incidentsPlacebo vs. Prolonged-release melatonin4 mg/day, take orally once daily before bedtime, and 8 weeks.Prolonged-release melatonin 4 mg did not reduce rapid eye movement sleep behavior disorder in PD.Stefani 2021 [20]United StatesRandomized, double-blind, placebo-controlled, multicenter study.34RBD manifestations during REM sleepPlacebo vs. Nelotanserin80 mg/day, take orally on an empty stomach (at least 1 h before or 2 h after a meal), and 4 weeks.No significant differences between Nelotanserin and placebo in terms of RBD behaviors.Di Giacopo 2012 [19]ItalyRandomized, double-blind, placebo-controlled, crossover trial12RBD frequencyPlacebo vs. Rivastigmine4.6 mg/day, patch, and 3 weeks.Rivastigmine was significantly reduced the mean frequency of RBD episodes during the observation time.Högl 2002 [18]AustriaRandomized, double-blind, placebo-controlled, crossover trial15Not statedPlacebo vs. Modafinil100–200 mg/day, and 2 weeks.Modafinil improved daytime sleepiness in PD patients.Adler 2003 [36]United StatesRandomized, double-blind, placebo-controlled crossover trial21ESSPlacebo vs. Modafinil200 mg/day, take the dose 30 to 45 min after breakfast, and 3 weeks.200 mg/day of modafinil was associated with few side effects and was modestly effective for the treatment of EDS in PD.Ondo 2005 [37]United StatesRandomized, double-blind, placebo-controlled, parallel-group trial40ESSPlacebo vs. Modafinil200–400 mg/day, take upon waking and at lunch, and 4 weeks.No significant differences were observed between the Modafinil and placebo. The drug did not alter motor symptoms in PD and was well tolerated.Videnovic 2021 [40]United StatesRandomized, double-blind, placebo-controlled, crossover trial64Safety and tolerabilityPlacebo vs. Solriamfetol75–300 mg, take orally once daily, and 4 weeks.No significant improvements were observed on ESS; The results by maintenance of wakefulness test suggest possible benefit with solriamfetol in PD.Büchele 2018 [39]SwitzerlandRandomized, double-blind, placebo-controlled, crossover trial12Sleep latencyPlacebo vs. Sodium Oxybate3–9 g/day, taken at bedtime and 2.5 to 4.0 h later, and 6 weeks.Sodium oxybate significantly improved EDS and enhanced subjective sleep quality.Mehta 2021 [38]United StatesRandomized, single-blind, placebo-controlled trial203The Unified Dyskinesia Rating ScalePlacebo vs. Amantadine137–274 mg/day, take at bedtime, and 12 weeks.Amantadine treatment significantly improved the daytime sleepiness.Postuma 2012 [46]Canada, BrazilRandomized controlled trial61ESSPlacebo vs. Caffeine100–200 mg twice daily, take upon awakening and after lunch, and 6 weeks.Caffeine provided only equivocal borderline improvement in excessive somnolence in PD.De Cock 2022 [41]FranceRandomized, double-blind, placebo-controlled, crossover trial46PDSSPlacebo vs. Apomorphine0.5–5.5 mg/h, delivered subcutaneously by a pump, and 17 nights.Mean change in PDSS score was significantly greater with night-time apomorphine infusion compared with placebo.Rios Romenets 2013 [30]CanadaRandomized, three-arm trial18Scales for Outcomes in Parkinson’s Disease-Sleep, insomnia severity indexPlacebo vs. Doxepin and non-pharmacologic treatmentDoxepin 10 mg at bedtime and CBT, and 6 weeks.Doxepin and non-pharmacologic treatment substantially improved insomnia in PD.Menza 2010 [42]United StatesRandomized, double-blind, placebo-controlled trial30Total sleep timePlacebo vs. Eszopiclone3 mg/day under 65 years old and 2 mg nightly 65 years or older, administered orally, and 6 weeks.Eszopiclone was superior to placebo in improving sleep quality and reducing the number of awakenings in patients with Parkinson’s disease, but it did not significantly affect total sleep time or daytime functioning.Pan 2013 [35]ChinaRandomized, double-blind, placebo-controlled trial96Not statedPlacebo vs. Yang-Xue-Qing-Nao granules4 g, 3 times/day, and 12 weeks.Yang-Xue-Qing-Nao granules improved the DFA parameters, and PDSS and the Unified Parkinson’s Disease Rating Scale scores in PD.Comella 2005 [29]United StatesRandomized, double-blind, placebo-controlled, parallel trial26Not statedPlacebo vs. Pergolide0.05–1.05 mg/day, take 1 h prior to bedtime, and 6 weeks.The pergolide group worsened in actigraphic measures of sleep efficiency and sleep fragmentation vs. the placebo group.Dowling 2005 [47]United StatesRandomized, double-blind, placebo-controlled, crossover trial40Not statedPlacebo vs. Melatonin5/50 mg, take 30 min before bedtime, and 2 weeks.Significant improvement in actigraphically measured total sleep time on 50 mg melatonin compared to 5 mg or placebo, this small improvement (10 min) may not be clinically significant.Ahn 2020 [34]KoreaRandomized, double-blind, placebo-controlled trial34PSQIPlacebo vs. Melatonin2 mg/day, take 1 h before bedtime, and 4 weeks.PRM is an effective treatment option for sleep quality in PD patients. Beneficial effects on sleep quality are associated with improved non-motor symptoms and quality of life in PD patients.Pierantozzi 2016 [43]ItalyRandomized, double-blind, placebo-controlled, parallel-group trial42Objective sleep assessment (time in bed, sleep efficiency, sleep latency, wakefulness after sleep onset.)Placebo vs. Rotigotine2–16 mg/day, patches from 18:00 h to 08:00 h, and 6–10 weeks.Rotigotine significantly improved sleep quality and continuity in PD.Sugumaran 2024 [33]IndiaRandomized, double-blind, placebo-controlled trial86PSQIPlacebo vs. Melatonin3 mg/day, take 1 h before bedtime, and 8 weeks.Melatonin is an effective and safe treatment option for sleep problems in PD.CGI Clinical Global Impression, REM Rapid Eye Movement, RBD Rapid Eye Movement Sleep Behavior Disorder, PSG Polysomnography, ICSD-3 International Classification of Sleep Disorders-Third Edition, v-PSG Video-Polysomnography, EDS Excessive Daytime Sleepiness, ESS Epworth Sleepiness Scale, PDSS Parkinson’s Disease Sleep Scale, PSQI Pittsburgh Sleep Quality Index, PD Parkinson’s Disease Table 1Description of included studiesStudy/YearCountryStudy designParticipantsPrimary outcomeControl group vs. Experimental groupDose, Route and DurationMain findingsShin 2019 [45]South KoreaRandomized, double-blind, placebo-controlled trial40CGI-ImprovementPlacebo vs. Clonazepam0.5 mg/day, take at bedtime, and 4 weeks.No significant differences were observed between the clonazepam and placebo groups.de Almeida 2021 [44]BrazilRandomized, double-blind, placebo-controlled trial36RBD events, CGI-Improvement and CGI-SeverityPlacebo vs. Cannabidiol75–300 mg/day, take after dinner, and 12 weeks.Cannabidiol did not significantly reduce RBD symptoms compared to placebo, but a temporary improvement in sleep satisfaction was observed in the Cannabidiol group during the early weeks of the trial.Kataoka 2024 [32]JapanRandomized, single-blind, placebo-controlled trial69Sleep efficiencyPlacebo vs. Zonisamide25 mg/day, take at bedtime, and 4 weeks.No significant differences were observed between the zonisamide and placebo groups, but improvements in WASO and SOL were observed in the zonisamide group.Meloni 2022 [31]ItalyRandomized, double-blind, placebo-controlled, crossover trial18CGI, RBD frequency, REM sleep without atonia percentagePlacebo vs. 5-hydroxytryptophan50 mg/day, take daily, and 4 weeks.5-HTP increased REM sleep but did not significantly reduce RBD episodes.Gilat 2020 [17]AustraliaRandomized, double-blind, placebo-controlled, parallel-group trial30RBD incidentsPlacebo vs. Prolonged-release melatonin4 mg/day, take orally once daily before bedtime, and 8 weeks.Prolonged-release melatonin 4 mg did not reduce rapid eye movement sleep behavior disorder in PD.Stefani 2021 [20]United StatesRandomized, double-blind, placebo-controlled, multicenter study.34RBD manifestations during REM sleepPlacebo vs. Nelotanserin80 mg/day, take orally on an empty stomach (at least 1 h before or 2 h after a meal), and 4 weeks.No significant differences between Nelotanserin and placebo in terms of RBD behaviors.Di Giacopo 2012 [19]ItalyRandomized, double-blind, placebo-controlled, crossover trial12RBD frequencyPlacebo vs. Rivastigmine4.6 mg/day, patch, and 3 weeks.Rivastigmine was significantly reduced the mean frequency of RBD episodes during the observation time.Högl 2002 [18]AustriaRandomized, double-blind, placebo-controlled, crossover trial15Not statedPlacebo vs. Modafinil100–200 mg/day, and 2 weeks.Modafinil improved daytime sleepiness in PD patients.Adler 2003 [36]United StatesRandomized, double-blind, placebo-controlled crossover trial21ESSPlacebo vs. Modafinil200 mg/day, take the dose 30 to 45 min after breakfast, and 3 weeks.200 mg/day of modafinil was associated with few side effects and was modestly effective for the treatment of EDS in PD.Ondo 2005 [37]United StatesRandomized, double-blind, placebo-controlled, parallel-group trial40ESSPlacebo vs. Modafinil200–400 mg/day, take upon waking and at lunch, and 4 weeks.No significant differences were observed between the Modafinil and placebo. The drug did not alter motor symptoms in PD and was well tolerated.Videnovic 2021 [40]United StatesRandomized, double-blind, placebo-controlled, crossover trial64Safety and tolerabilityPlacebo vs. Solriamfetol75–300 mg, take orally once daily, and 4 weeks.No significant improvements were observed on ESS; The results by maintenance of wakefulness test suggest possible benefit with solriamfetol in PD.Büchele 2018 [39]SwitzerlandRandomized, double-blind, placebo-controlled, crossover trial12Sleep latencyPlacebo vs. Sodium Oxybate3–9 g/day, taken at bedtime and 2.5 to 4.0 h later, and 6 weeks.Sodium oxybate significantly improved EDS and enhanced subjective sleep quality.Mehta 2021 [38]United StatesRandomized, single-blind, placebo-controlled trial203The Unified Dyskinesia Rating ScalePlacebo vs. Amantadine137–274 mg/day, take at bedtime, and 12 weeks.Amantadine treatment significantly improved the daytime sleepiness.Postuma 2012 [46]Canada, BrazilRandomized controlled trial61ESSPlacebo vs. Caffeine100–200 mg twice daily, take upon awakening and after lunch, and 6 weeks.Caffeine provided only equivocal borderline improvement in excessive somnolence in PD.De Cock 2022 [41]FranceRandomized, double-blind, placebo-controlled, crossover trial46PDSSPlacebo vs. Apomorphine0.5–5.5 mg/h, delivered subcutaneously by a pump, and 17 nights.Mean change in PDSS score was significantly greater with night-time apomorphine infusion compared with placebo.Rios Romenets 2013 [30]CanadaRandomized, three-arm trial18Scales for Outcomes in Parkinson’s Disease-Sleep, insomnia severity indexPlacebo vs. Doxepin and non-pharmacologic treatmentDoxepin 10 mg at bedtime and CBT, and 6 weeks.Doxepin and non-pharmacologic treatment substantially improved insomnia in PD.Menza 2010 [42]United StatesRandomized, double-blind, placebo-controlled trial30Total sleep timePlacebo vs. Eszopiclone3 mg/day under 65 years old and 2 mg nightly 65 years or older, administered orally, and 6 weeks.Eszopiclone was superior to placebo in improving sleep quality and reducing the number of awakenings in patients with Parkinson’s disease, but it did not significantly affect total sleep time or daytime functioning.Pan 2013 [35]ChinaRandomized, double-blind, placebo-controlled trial96Not statedPlacebo vs. Yang-Xue-Qing-Nao granules4 g, 3 times/day, and 12 weeks.Yang-Xue-Qing-Nao granules improved the DFA parameters, and PDSS and the Unified Parkinson’s Disease Rating Scale scores in PD.Comella 2005 [29]United StatesRandomized, double-blind, placebo-controlled, parallel trial26Not statedPlacebo vs. Pergolide0.05–1.05 mg/day, take 1 h prior to bedtime, and 6 weeks.The pergolide group worsened in actigraphic measures of sleep efficiency and sleep fragmentation vs. the placebo group.Dowling 2005 [47]United StatesRandomized, double-blind, placebo-controlled, crossover trial40Not statedPlacebo vs. Melatonin5/50 mg, take 30 min before bedtime, and 2 weeks.Significant improvement in actigraphically measured total sleep time on 50 mg melatonin compared to 5 mg or placebo, this small improvement (10 min) may not be clinically significant.Ahn 2020 [34]KoreaRandomized, double-blind, placebo-controlled trial34PSQIPlacebo vs. Melatonin2 mg/day, take 1 h before bedtime, and 4 weeks.PRM is an effective treatment option for sleep quality in PD patients. Beneficial effects on sleep quality are associated with improved non-motor symptoms and quality of life in PD patients.Pierantozzi 2016 [43]ItalyRandomized, double-blind, placebo-controlled, parallel-group trial42Objective sleep assessment (time in bed, sleep efficiency, sleep latency, wakefulness after sleep onset.)Placebo vs. Rotigotine2–16 mg/day, patches from 18:00 h to 08:00 h, and 6–10 weeks.Rotigotine significantly improved sleep quality and continuity in PD.Sugumaran 2024 [33]IndiaRandomized, double-blind, placebo-controlled trial86PSQIPlacebo vs. Melatonin3 mg/day, take 1 h before bedtime, and 8 weeks.Melatonin is an effective and safe treatment option for sleep problems in PD.CGI Clinical Global Impression, REM Rapid Eye Movement, RBD Rapid Eye Movement Sleep Behavior Disorder, PSG Polysomnography, ICSD-3 International Classification of Sleep Disorders-Third Edition, v-PSG Video-Polysomnography, EDS Excessive Daytime Sleepiness, ESS Epworth Sleepiness Scale, PDSS Parkinson’s Disease Sleep Scale, PSQI Pittsburgh Sleep Quality Index, PD Parkinson’s Disease Description of included studies CGI Clinical Global Impression, REM Rapid Eye Movement, RBD Rapid Eye Movement Sleep Behavior Disorder, PSG Polysomnography, ICSD-3 International Classification of Sleep Disorders-Third Edition, v-PSG Video-Polysomnography, EDS Excessive Daytime Sleepiness, ESS Epworth Sleepiness Scale, PDSS Parkinson’s Disease Sleep Scale, PSQI Pittsburgh Sleep Quality Index, PD Parkinson’s Disease As shown in Fig. 2, most included studies were rated as low or unclear risk in selection bias, performance bias, detection bias, attrition bias, and reporting bias, while some studies were judged as high risk. Specifically, in the “Random sequence generation” domain, the high-risk score was given to Comella [29]. In the “Allocation concealment” section, the studies with high-risk scores were Rios Romenets [30] and Meloni [31]. In the “Blinding of participants and personnel” domain, the one with the high risk was Rios Romenets [30]. In the “Blinding of outcome assessment” domain, the high-risk scores were given to Kataoka [32] and Di Giacopo [19]. In the “Incomplete outcome data” domain, the high risk were assigned to Sugumaran [33], Kataoka [32], and Ahn [34]. In the “Selective Reporting” domain, the high-risk scores were assigned to Pan [35] and Adler [36]. Additionally, nine studies were rated as high risk for the “Other bias” domain. For example, in Comella, patients had inconsistent baseline levodopa dosages. Fig. 2Risk of bias assessment for included randomized. The evaluation primarily focused on seven key domains: (1) Random sequence generation (selection bias), (2) Allocation concealment (selection bias), (3) Blinding of participants and personnel (performance bias), (4) Blinding of outcome assessment (detection bias), (5) Incomplete outcome data (attrition bias), (6) Selective reporting (reporting bias), and (7) Other potential sources of bias Risk of bias assessment for included randomized. The evaluation primarily focused on seven key domains: (1) Random sequence generation (selection bias), (2) Allocation concealment (selection bias), (3) Blinding of participants and personnel (performance bias), (4) Blinding of outcome assessment (detection bias), (5) Incomplete outcome data (attrition bias), (6) Selective reporting (reporting bias), and (7) Other potential sources of bias Furthermore, according to the GRADE assessment [27], the certainty of evidence across all studies ranged from very low to moderate. The certainty of evidence for RBD ranged from very low to low. Studies evaluating melatonin [17], nelotanserin [20] and 5-hydroxytryptophan (5-HTP) [31] was rated as low certainty. For EDS, the overall certainty of evidence ranged from low to moderate. Studies assessing modafinil [18, 36, 37], amantadine [38] and sodium oxybate [39] were moderate certainty, while solriamfetol [40] was rated as low certainty. For insomnia, the certainty ranged from very low to moderate [30, 35, 41, 42], and for sleep quality, both melatonin [33, 34] and rotigotine [43] demonstrated moderate certainty (Tables S3-S6). Systematic review of RBD Seven studies evaluated the treatment effect of RBD, which enrolled a total of 231 patients [17, 19, 20, 31, 32, 44, 45]. These studies covered seven pharmacological interventions: cannabidiol [44], melatonin [17], zonisamide [32], 5-HTP [31], clonazepam [45], nelotanserin [20], rivastigmine [19]. Except for rivastigmine, all other interventions were administered orally. Kataoka [32] was single-blind randomized placebo-controlled, and the remaining six studies were all double-blind randomized placebo-controlled. In seven trials, only two [19, 31] had a crossover design and the other five trials [17, 20, 32, 44, 45] had a parallel design. The included studies utilized diverse scales to assess RBD, including the PSG, the RBD screening questionnaire, and the International Classification of Sleep Disorders, Third Edition (ICSD-3). Systematic review of EDS A total of seven studies evaluated the treatment of EDS, enrolling 400 participants [18, 36–40, 46]. These studies covered five pharmacological interventions: modafinil [18, 36, 37], amantadine [38], sodium oxybate [39], solriamfetol [40], and caffeine [46]. All drugs were administered orally. Four of the trials had a crossover design [18, 36, 39, 40], whereas the remaining three had a parallel design [37, 38, 46]. EDS was diagnosed using the ESS in most included studies, except Mehta [38], which used the Movement Disorder Society-sponsored revision of the Unified Parkinson’s Disease Rating Scale (MDS-UPDRS) Part I. Systematic review of insomnia For insomnia, six studies were included with a total of 213 patients. These studies addressed six distinct pharmacological interventions: apomorphine [41], eszopiclone [42], doxepin [30], Yang-Xue-Qing-Nao granules [35], pergolide [29] and melatonin [47]. In these trials, all drugs or placebo were administered orally, except for apomorphine [41], which was delivered via subcutaneous infusion. All studies were randomized placebo-controlled trials. Rios Romenets [30] was a partially blinded, while others were double-blind studies. Definitions of insomnia varied considerably: Menza [42] used total sleep time, Cochen De Cock [41] used Insomnia Severity Index (ISI) scores, and Rios Romenets [30] used scales for outcomes in Parkinson’s disease-sleep nocturnal subscores. Systematic review of sleep quality In the remaining three trials, although the specific type of sleep disorder was not reported, participants consistently reported the overall sleep quality. We included these studies with 149 patients. The drug interventions used varied: Ahn [34] and Sugumaran [33] studied melatonin, while Pierantozzi [43] examined Rotigotine. The routes of drug administration differed: Rotigotine was administered via transdermal patch, while the others were administered orally. All studies were double-blind, randomized, placebo-controlled trials and parallel-group designs. PSQI was used in Sugumaran and Ahn to assess poor sleep quality [33, 34], while Pierantozzi combined PSG with PSQI and Parkinson’s Disease Sleep Scale (PDSS) for evaluation [43]. Meta analysis of the overall results The meta-analysis suggested that pharmacological interventions were effective in improving sleep disorders for patients with PD (SMD: −0.45; 95% CI: −0.64, −0.26) (Fig. 3). Specifically, pharmacological interventions showed significant efficacy in addressing insomnia (SMD: −0.53; 95% CI: −0.82, −0.24) and EDS (SMD: −0.36; 95% CI: −0.63, −0.10), as well as improving overall sleep quality (SMD: −0.96; 95% CI: −1.67, −0.25). The overall result showed moderate heterogeneity (I2 = 53.3%). Additionally, no significant publication bias was detected through funnel plot for all included studies (Fig. S1). Egger’s test showed no significant publication bias for sleep disorder subgroups (Fig. S2). The sensitivity analysis further supported the robustness of meta-analysis results (Fig. S3). Fig. 3Subgroup analysis of the effects of pharmacological interventions in people with Parkinson’s disease by different sleep disorders Subgroup analysis of the effects of pharmacological interventions in people with Parkinson’s disease by different sleep disorders Meta-analysis of RBD For RBD, although the pharmacological interventions differed among Shin [45], de Almeida [44], and Kataoka [32] (clonazepam, cannabidiol, and zonisamide, respectively), all could be classified as anticonvulsants. The meta-analysis indicated that anticonvulsants did not show significant efficacy in treating RBD (SMD: −0.29; 95% CI: −0.69, 0.10) (Fig. 4A). Three studies reported that 5-HTP, nelotanserin, and melatonin have no significant effect in improving RBD [17, 20, 31]. In contrast, rivastigmine could significantly reduce RBD events [19]. Fig. 4Subgroup analysis of the effects of pharmacological interventions in people with Parkinson’s disease by different pharmacological interventions. A rapid eye movement behavior disorder, (B) excessive daytime sleepiness, (C) insomnia, (D) sleep quality Subgroup analysis of the effects of pharmacological interventions in people with Parkinson’s disease by different pharmacological interventions. A rapid eye movement behavior disorder, (B) excessive daytime sleepiness, (C) insomnia, (D) sleep quality Meta-analysis of EDS The meta-analysis suggested that modafinil significantly reduced EDS. As showed in Fig. 4B, the overall SMD was − 0.58 (95% CI: −1.11, −0.05), with low heterogeneity (I2 = 25.5%). However, the study by Videnovic suggested that solriamfetol, regardless of dosage (75 mg, 150 mg, or 300 mg), did not show any improvement [40]. Both sodium oxybate and amantadine demonstrated favorable effects on EDS [38, 39]. Interestingly, although caffeine did not significantly improve ESS scores in the intention-to-treat analysis, it showed a significant reduction in the per-protocol analysis [46]. Meta-analysis of insomnia As shown in Fig. 4C, our meta-analysis revealed that pharmacological treatment was overall effective in improving insomnia (SMD: −0.53; 95% CI: −0.82, −0.24). Specifically, doxepin showed significant improvement in ISI, while eszopiclone failed to show benefit in total sleep time [30, 42]. The study by Cochen De Cock demonstrated that nocturnal apomorphine infusion therapy showed significant improvements in the ISI, PDSS, and clinical global impression [41]. Notably, the continuous dopaminergic stimulation provided by this therapy also exhibited beneficial effects on patients’ motor symptoms. Additionally, Yang-Xue-Qing-Nao granules significantly improved PDSS scores [35]. Meta-analysis of sleep quality The meta-analysis results suggested that melatonin significantly improved sleep quality, as assessed by the PSQI [33, 34]. As shown in Fig. 4D, the overall WMD was − 1.75 (95% CI: −2.87, −0.64). In Pierantozzi [43], rotigotine demonstrated significant improvements in sleep efficiency and sleep latency. Investigating heterogeneity As illustrated in Fig. 3, pharmacological interventions for sleep disorders in patients with PD showed moderate heterogeneity across all included studies (I2 = 53.3%). When grouped by the type of sleep disorder, low heterogeneity was observed within the subgroups for RBD and insomnia (I2 = 21.0% and 0.0%, respectively). In contrast, the subgroups for EDS and sleep quality exhibited moderate to substantial heterogeneity, (I2 = 56.6% and 74.3%, respectively). These results suggested that heterogeneity might be attributed to the diversity of sleep disorder phenotypes. To further investigate potential sources of heterogeneity, subgroup analyses were conducted based on drug classes within each sleep disorder type, as presented in Fig. 4. In subgroup analysis for EDS (Fig. 4B), significant between-group heterogeneity was observed across different pharmacological interventions (p = 0.005). Similarly, in subgroup analysis for sleep quality (Fig. 4D), notable between-group differences were found between melatonin and rotigotine (p = 0.009). These results indicate that variation in drug class may be a contributor to heterogeneity, particularly in the EDS and sleep quality subgroups. Discussion In this study, we systematically evaluated the pharmacological interventions for managing sleep disorders in patients with PD by including 23 RCTs. Current evidence indicated that pharmacotherapy for EDS and insomnia had acceptable efficacy. However, pharmacological interventions had a limited efficacy in improving RBD among people with PD. Interestingly, some recent studies have shown that melatonin has an effective performance in improving sleep quality in people with PD. The studies included in our review indicated that most drugs were effective in treating EDS, including modafinil, sodium oxybate, and amantadine. Among these drugs, modafinil is one of the most commonly used for EDS management in clinical practice. Our meta-analysis further confirmed its efficacy in people with PD. Unlike traditional dopaminergic agents, such as amphetamines, modafinil has a slower, more sustained, and lower-efficacy stimulatory effect on the dopamine system, which reduces the risks of rapid tolerance and addiction [48–50]. By inhibiting the dopamine transporter, modafinil reduces the reuptake of dopamine and enhances the dopaminergic neurotransmission in the central nervous system of patients with PD, ultimately promoting wakefulness [51]. Moreover, it is worth noting that modafinil is a neuroenhancer [52] and it could reduce dyskinesia, tremor, and urinary frequency in patients with PD [53]. The neuroprotective properties of modafinil may involve modulation of cellular energy dynamics, neurotrophic signaling, calcium regulation, dopamine transporter activity, and apoptosis pathways [54]. In summary, for patients with PD, modafinil could offer long-term clinical benefits by improving sleep quality and providing neuroprotective effects. In addition, we reviewed the efficacy of pharmacological interventions for the RBD treatment in people with PD. Unfortunately, the meta-analysis results suggested that anticonvulsants (clonazepam, cannabidiol, and zonisamide), nelotanserin, 5-HTP, and melatonin have limited efficacy in the treatment of RBD. Only the study by Di Giacopo suggested that rivastigmine could reduce RBD event [19]. Rivastigmine may improve RBD by inhibiting cholinergic pathways [55]. Although pharmacological treatments showed limited efficacy in alleviating RBD symptoms, the potential benefits of non-pharmacological interventions should not be overlooked. For instance, cognitive behavioral therapy may help reduce the frequency of distressing dreams, thereby alleviating RBD symptoms [56]. Additionally, approaches such as eye movement desensitization and lucid dream training are effective in improving nightmares and may offer therapeutic value for individuals with RBD [56]. In summary, the management of RBD in patients with PD remains challenging, and more strong evidence is needed to confirm these findings. Furthermore, we evaluated the effects of pharmacological interventions on insomnia symptoms in patients with PD. Sedative-hypnotic agents, including doxepin and eszopiclone, were not significantly effective in relieving insomnia symptoms [30, 42]. However, nocturnal apomorphine infusion therapy showed benefits in improve insomnia without causing daytime sleepiness [41]. As a potent dopamine agonist, apomorphine acts through multiple mechanisms, including promoting the degradation of β-amyloid protein and reducing oxidative stress in neuronal cells [57]. In addition, apomorphine could alleviate motor symptoms and enhance well-being of patients [58]. These benefits suggested that apomorphine might offer clinical advantages in managing both motor and non-motor symptoms, particularly for PD patients with insomnia. Moreover, two studies indicated that melatonin can improve sleep efficiency, reduce nocturnal activity, and shorten sleep latency in patients with PD [33, 34]. Melatonin exerts its effects primarily by stabilizing circadian rhythms, correcting abnormal expression of circadian clock proteins, and mitigating disruptions in the sleep-wake cycle [59]. Melatonin acts through two main receptor subtypes: the high-affinity melatonin type 1 (MT1) receptor and the low-affinity melatonin type 2 (MT2) receptor [60]. Activation of the MT1 receptor inhibits adenylate cyclase activity, reducing cyclic AMP (cAMP) levels and promoting sleep [61]. In contrast, activation of the MT2 receptor triggers phosphoinositide hydrolysis, which plays a key role in sleep patterns [62]. Through this dual MT receptor signaling, melatonin can modulate multiple physiological processes, supporting overall circadian stability and improved sleep quality in patients with PD. Nevertheless, the limitations of melatonin in clinical settings should be noted. Melatonin may cause adverse effects, such as headache, dizziness, nausea, and somnolence [63]. The co-administration of melatonin with antidepressants or sedatives may increase the risks of excessive sedation, orthostatic hypotension, and cognitive impairment [64]. Additionally, there is growing interest in dual orexin receptor antagonists (DORAs). Although we attempted to include available evidence on the efficacy of DORAs for population with PD, the current research remains limited. Currently, the most widely studied DORAs include lemborexant and suvorexant. Orexin A and orexin B are endogenous neuropeptides that regulate wakefulness and arousal by activating the orexin-1 receptor (OX1R) and orexin-2 receptor (OX2R), respectively [65]. DORAs exert their therapeutic effects by competitively blocking both OX1R and OX2R, thereby inhibiting orexin signaling, reducing wakefulness drive, and promoting physiological sleep [66, 67]. Compared to traditional hypnotic agents, such as benzodiazepines, DORAs have a more targeted mechanism of action, selectively inhibiting the wake-promoting pathways while preserving natural sleep architecture, which may lead to a better safety profile [68]. In addition, DORAs are being explored in other neurodegenerative disorders such as Alzheimer’s disease [69]. Recent research indicates that suvorexant may improve total sleep time and sleep efficiency in Alzheimer’s disease patients with insomnia and lemborexant showed potential in improving circadian rhythm parameters [70]. In this study, we assessed the bias risks of the included studies using the Cochrane Risk of Bias tool, and evaluated the certainty of evidence using the GRADE tool. The consistent use of the ESS in these studies allowed for a reliable assessment of the modafinil efficacy. For RBD, several studies had short durations, which may influence the evaluation of treatment effect. In addition, Comella’s study reported that the placebo group had a significantly higher baseline dopamine dosage compared to the experimental group [29]. The experimental group in Comella’s study also had a dropout rate of over 30%, potentially introducing bias into the study outcomes [29]. Our subgroup analyses suggest that the diversity in sleep disorders and pharmacological interventions may contribute to the observed heterogeneity. Different sleep disorders in PD are underpinned by distinct pathophysiological mechanisms. Specifically, RBD is associated with dysfunction in brainstem circuits that regulate muscle atonia during REM sleep [71]. These circuits include inhibitory and excitatory pathways across the hypothalamus, pons, and medullary reticular formation [72]. Insomnia in PD is a multifactorial condition that may arise from neurodegeneration, circadian rhythm disruption, altered arousal regulation, and comorbid mood disturbances [73, 74]. In addition, damage to brain regions responsible for maintaining wakefulness may underlie the pathogenesis of EDS [75]. EDS may be a primary symptom of PD-related degeneration in arousal systems [76]. Furthermore, the diversity in pharmacological mechanisms of action may also contribute to heterogeneity. For example, modafinil is thought to enhance wakefulness by stimulating multiple neurotransmitter systems, including histamine, norepinephrine, and dopamine [77, 78]. In contrast, solriamfetol acts as a dopamine and norepinephrine reuptake inhibitor, selectively increasing the extracellular concentrations of these neurotransmitters [79]. In summary, our findings underscore the importance of tailoring pharmacological interventions to the specific type of sleep disorders and the individual characteristics of patients with PD. This highlights the critical role of precision medicine in the effective management of sleep disturbances in population with PD. The main strength of this study was the inclusion of a total of 23 RCTs, which provided a comprehensive evaluation of the pharmacological treatments of sleep disorders in populations with PD. However, several limitations should be acknowledged. First, we did not include RCTs on DORAs due to the limited evidence available in this area. Second, our analysis included crossover trials, which might introduce carry-over bias. Third, although we included studies from India and China, the majority of the included studies were conducted in developed countries. Such geographic imbalance might limit the generalizability of our findings to global population. Fourth, subgroup analyses suggested that heterogeneity might result from variations in pharmacological interventions and the diversity of sleep disorder phenotypes. However, given the small number of studies within each subgroup, these findings should be interpreted with discretion. Fifth, the Egger’s tests may be underpowered for subgroups with fewer than 10 included studies, and the results should be interpreted with caution. In summary, in the setting of PD, strong evidence from multicenter RCTs is needed to address the challenges of RBD management and to explore the potential clinical benefits of newly therapies, such as DORAs. Conclusion Based on evidence from 23 RCTs, pharmacological interventions are effective in improving EDS and insomnia in populations with PD. However, the management of RBD with pharmacological treatments remains challenging. Strong evidence from multicenter RCTs is needed to address this challenge. Novel agents such as DORAs may be explored as potential therapeutic options for sleep disturbances in PD. Supplementary Information Supplementary Material 1. Supplementary Material 1.