Sleep Drugs Help Parkinson’s Patients Rest Better, But Only One Symptom Shows Solid Benefit

Sleep problems are among the most burdensome features of Parkinson’s disease, and a sweeping new analysis of randomised clinical trials has now mapped, with unusual honesty, exactly how much the available drugs can actually deliver. The systematic review and meta-analysis, published in the Journal of Neurology, pooled data from 29 randomised controlled trials involving 1,913 participants and found that pharmacological treatment produces a statistically significant improvement in overall sleep quality, while the evidence for excessive daytime sleepiness, insomnia symptoms, and REM sleep behaviour disorder remains too fragile to support firm clinical recommendations.

The scale of the problem the researchers set out to address is enormous. Parkinson’s disease is the second most common neurodegenerative disorder worldwide and the fastest-growing neurological condition in terms of prevalence, disability, and mortality. Driven by global population ageing, the number of people living with the disease has more than doubled over the past two decades and is projected to exceed 12 million by 2040. Sleep disturbances, including insomnia, excessive daytime sleepiness, REM sleep behaviour disorder, restless legs syndrome, and poor sleep quality, are consistently among the most prevalent and disabling of the disease’s non-motor symptoms, yet the pharmacological evidence base has remained fragmented across sleep phenotypes, drug classes, and outcome instruments.

The biology underlying these sleep problems is complex and multifactorial. Degeneration of dopaminergic and non-dopaminergic pathways, deposition of alpha-synuclein in brainstem and hypothalamic nuclei, disruption of circadian regulation, nocturnal motor symptoms, psychiatric comorbidities, autonomic dysfunction, and the adverse effects of antiparkinsonian medications may all contribute. This heterogeneity matters clinically, because different sleep phenotypes may respond to entirely different drugs: wake-promoting agents for daytime sleepiness, melatonergic compounds for circadian disruption, dopaminergic drugs for nocturnal motor interference with sleep, and sedative-hypnotics for insomnia. The new review, led by Jaquelini Betta Canever and colleagues at the Federal University of Santa Catarina in Brazil, is the first to synthesise the randomised trial evidence systematically across all of these phenotypes at once.

The trials included in the analysis spanned more than two decades, from 2002 to 2025, and were conducted in 14 countries across five continents. The pharmacological landscape they covered was remarkably broad. Chronobiotic agents, meaning melatonin and prolonged-release melatonin, were the most frequently evaluated class, appearing in seven trials at doses ranging from 2 to 10 milligrams. Wake-promoting agents and psychostimulants featured in seven trials, including modafinil, methylphenidate, the experimental modafinil-flecainide combination THN102, solriamfetol, and sodium oxybate. Dopaminergic receptor agonists were examined in five trials, including the rotigotine transdermal patch, apomorphine infusion, piribedil, and prolonged-release ropinirole. Single trials evaluated cannabidiol, zonisamide, clonazepam, eszopiclone, trazodone, 5-hydroxytryptophan, caffeine, homotaurine, the MAO-B inhibitors safinamide and rasagiline, and even a traditional nyctinastic herbs decoction.

The clearest signal emerged for overall sleep quality measured with the Pittsburgh Sleep Quality Index. Across six trials testing melatonin, modafinil, and rotigotine, pharmacological treatment produced a pooled mean difference of 2.66 points, with a 95 percent confidence interval of 1.63 to 3.69 and a p value below 0.001. The direction of benefit was consistent across all six contributing trials, and the magnitude approximates the minimal clinically important difference reported in comparable populations. Leave-one-out sensitivity analysis confirmed the result was highly robust: removing any single trial produced pooled estimates ranging from 2.20 to 3.11, all with confidence intervals entirely excluding the null. Subgroup analysis showed that melatonin preparations as a class exerted a consistent moderate benefit, with only moderate residual heterogeneity, while the rotigotine patch produced the single largest effect of any agent.

That rotigotine finding, however, comes with an important caveat that the authors flag explicitly. Rotigotine is primarily prescribed to treat the motor symptoms of Parkinson’s disease rather than sleep disturbance specifically, so part of the observed improvement in sleep quality scores may reflect secondary benefits of better nocturnal motor control, such as reduced nighttime akinesia, rigidity, and difficulty repositioning, rather than a direct sleep-specific pharmacological effect. The biologically plausible mechanism, involving D3 receptor activity in the basal ganglia and hypothalamus that could improve nocturnal sleep architecture, remains attractive, but the confound cannot be excluded with the available data.

For excessive daytime sleepiness, the picture was far less convincing. Six study arms from five trials, testing modafinil, methylphenidate, and melatonin, produced a pooled estimate on the Epworth Sleepiness Scale that pointed toward benefit but failed to reach statistical significance, with a mean difference of 1.75 and a confidence interval spanning zero. Heterogeneity was extreme at 91 percent, driven predominantly by a single small crossover trial from 2002 that was the only arm favouring placebo. When that influential outlier was removed in sensitivity analysis, the pooled estimate became statistically significant for the first time, suggesting the aggregate result was masked by a trial whose anomalous direction is consistent with carry-over effects from inadequate washout. Intriguingly, a subgroup analysis by treatment duration found that trials lasting more than four weeks showed a significant pooled benefit, while shorter trials did not, though the authors stress this rests on very few studies and should be regarded as hypothesis-generating rather than a dosing recommendation.

For insomnia symptoms and REM sleep behaviour disorder, the evidence collapsed under its own inconsistency. Three trials assessing insomnia with the Parkinson’s Disease Sleep Scale produced a directionally favourable but non-significant pooled estimate with 96 percent heterogeneity, reflecting pharmacologically distinct agents tested in substantially different populations. The REM sleep behaviour disorder analysis was limited to just two trials, zonisamide and safinamide, which produced directionally opposite results that nearly cancelled each other out, yielding a clinically uninformative pooled estimate with 99 percent heterogeneity. Compounding the problem, neither trial diagnosed the disorder using international classification criteria, and severity was inferred from a global Parkinson’s sleep scale rather than a dedicated, validated instrument. The authors therefore present these pooled estimates as exploratory only, emphasising narrative synthesis over the point estimates.

The methodological quality of the underlying trials emerged as a sobering thread throughout the analysis. Using the Cochrane Risk of Bias 2 tool, the reviewers judged only 5 of the 29 trials, or 17.2 percent, to be at low overall risk of bias, while 18 raised some concerns and 6 were classified as high risk. The main sources of concern were the randomisation process and missing outcome data. Applying the GRADE framework, the certainty of evidence ranged from low to very low across all outcome domains, meaning that even the statistically significant sleep quality finding should be interpreted with appropriate caution. Publication bias could not be formally excluded for any domain, since all meta-analyses contained fewer than the ten studies required for formal statistical testing, although the funnel plot for sleep quality offered tentative reassurance.

On safety, the review found no unexpected or severe signals beyond those already characterised in the broader pharmacological literature, though this reassurance is bounded by limited trial durations of two to 24 weeks and 1,913 total participants. Dopaminergic receptor agonists carried the highest rates of any adverse event, with ropinirole reaching 74.3 percent in one trial, consistent with known class effects such as nausea, orthostatic hypotension, and impulse control disorders. Adverse event rates for melatonin ranged from zero to 88.6 percent across studies, a striking spread the authors attribute to differences in how events were ascertained rather than to genuine toxicity. Treatment discontinuation rates were generally low, with cannabidiol showing the highest rate among agents with multiple contributing studies at 15 percent. The authors call for future trials with prospective phenotypic stratification, harmonised primary outcome instruments, head-to-head comparisons between drug classes, polysomnographic endpoints alongside patient-reported outcomes, and standardised safety reporting frameworks, so that the next generation of evidence can finally translate into reliable, clinically actionable guidance for the millions of Parkinson’s patients whose nights are currently governed by trial and error.

Subject of Research: Pharmacological treatment of sleep disorder phenotypes in Parkinson's disease

Article Title: Pharmacological treatment of sleep disorder phenotypes in Parkinson’s disease: a systematic review and meta-analysis of randomised controlled trials

Article References: Pharmacological treatment of sleep disorder phenotypes in Parkinson’s disease: a systematic review and meta-analysis of randomised controlled trials. (n.d.). https://doi.org/10.1007/s00415-026-14158-5

Image Credits: AI Generated

DOI: 10.1007/s00415-026-14158-5

Keywords: Parkinson's disease, sleep disorders, meta-analysis, melatonin, excessive daytime sleepiness, REM sleep behaviour disorder, insomnia, rotigotine, randomised controlled trials, sleep quality, neuropharmacology, systematic review