Cannabinoids Show Short-Term Promise for Sleep Apnea, But Evidence Remains Too Thin for Clinical Use

A new systematic review finds mostly positive signals from small, brief trials, yet median treatment duration of just three weeks leaves major questions about safety, tolerability, and durability of benefit entirely unanswered for a condition that demands lifelong management.

Why This Matters

Obstructive sleep apnea affects an estimated one billion people worldwide and carries serious cardiovascular, metabolic, and neurocognitive consequences when undertreated. With medical cannabis legalization expanding and at least one U.S. state now listing OSA as a qualifying condition, patients are increasingly asking whether cannabinoids represent a viable alternative to CPAP or oral appliances. The timing of this systematic review matters because it incorporates studies published after the American Academy of Sleep Medicine’s 2018 position statement recommending against cannabinoid use for OSA, offering the most current evidence synthesis available to clinicians navigating these conversations.

Clinical Summary

Obstructive sleep apnea is characterized by repetitive upper airway collapse during sleep, leading to intermittent hypoxia and sleep fragmentation. The endocannabinoid system modulates autonomic respiratory regulation and upper airway muscle tone through CB1 and CB2 receptor signaling, providing a plausible mechanistic basis for investigating cannabinoids as a pharmacological intervention. This 2024 PROSPERO-registered systematic review, conducted following PRISMA guidelines and published by Engstrom and colleagues, searched five databases through June 2024 and identified nine studies (five full-text articles and four conference abstracts) encompassing 485 total participants. Seven of the nine studies evaluated dronabinol, a synthetic form of THC approved in the United States for other indications, while two observational studies examined cannabis.

Eight of nine included studies reported statistically significant improvements in at least one OSA outcome, including reductions in the apnea-hypopnea index and improvements on patient-reported daytime sleepiness scales. However, between 70% and 80% of participants across studies experienced neuropsychiatric or gastrointestinal adverse events attributable to the cannabinoid intervention. The median treatment duration was only three weeks, with a range of one to six weeks, making it impossible to assess whether benefits persist or whether chronic-use safety is acceptable for a lifelong condition. The authors explicitly note that the evidence does not support clinical recommendations and call for adequately powered, longer-duration randomized controlled trials before cannabinoids can be considered for OSA management.

Dr. Caplan’s Take

This review does an honest job of laying out what is actually there, and what is actually there is not much. The mechanistic rationale connecting endocannabinoid signaling to upper airway tone is real and worth pursuing, but eight positive results out of nine tiny, weeks-long studies is not the kind of evidence that should change what we tell patients. When someone with moderate-to-severe OSA asks me whether cannabis might replace their CPAP, what they deserve is a straightforward answer: we do not have the data to say yes, and the adverse event rates in even these short trials suggest the risk-benefit calculus is far from settled.

In practice, I treat OSA as the serious cardiovascular and metabolic risk factor it is, and I prioritize interventions with robust long-term evidence. If a patient is already using cannabis and reports subjective sleep improvement, I document that but I do not withdraw proven therapy on that basis. I continue to recommend adherence to CPAP or validated alternatives, and I tell patients honestly that cannabinoid research for OSA is early-stage. Monitoring for interactions between cannabinoids and sedating medications remains essential in this population.

Clinical Perspective

This systematic review sits at an early point in the research arc for cannabinoids in OSA. It confirms that preliminary signals of efficacy exist, particularly for dronabinol’s effect on the apnea-hypopnea index, but it does not move the needle toward clinical adoption. The evidence base remains too small, too brief, and too methodologically heterogeneous to support patient-facing treatment recommendations. Notably, quality appraisal of the five full-text studies revealed only two with low risk of bias, one with some concerns, and two with high risk of bias. The four abstract-only studies could not be appraised at all. The AASM’s recommendation against cannabinoids for OSA remains well supported by this review’s own findings.

Clinicians should be aware that dronabinol carries known pharmacological effects including sedation, psychoactive effects, and gastrointestinal disturbance, all of which were reflected in the 70-to-80% adverse event rate reported here. In patients with OSA who often have comorbid obesity, cardiovascular disease, and polypharmacy, these effects warrant particular caution. Drug interactions with sedatives, opioids, and CNS depressants are relevant in this population. The one actionable step clinicians can implement now is to proactively raise this topic with patients who may be self-medicating with cannabis for sleep, document their use, ensure CPAP or other proven therapies are not being abandoned, and monitor for additive sedation.

Study at a Glance

Study Type

Systematic review (no meta-analysis)

Population

Adults with obstructive sleep apnea (485 participants across 9 studies)

Intervention

Dronabinol (7 studies); cannabis (2 studies)

Comparator

Placebo or no comparator (varied by study)

Primary Outcomes

Apnea-hypopnea index, daytime sleepiness, adverse events

Sample Size

9 studies, 485 total participants

Journal

Not specified in source data

Year

2024

DOI or PMID

PROSPERO CRD42024555089

Funding Source

Not specified in source data

What Kind of Evidence Is This

This is a PROSPERO-registered systematic review conducted under PRISMA guidelines, with dual independent screening and validated quality appraisal tools. In the evidence hierarchy, systematic reviews typically occupy a high position, but this review’s strength is constrained by the weakness of the underlying studies it synthesizes. No meta-analysis was performed, meaning findings are narrative rather than pooled quantitative estimates. The single most important inference constraint is that nearly half the included studies were available only as conference abstracts, precluding complete quality assessment and limiting the reliability of any conclusions drawn from the overall body of evidence.

How This Fits With the Broader Literature

This review extends the evidence base considered by the 2018 AASM position statement, which recommended against cannabinoid use for OSA based on even sparser data available at that time. The largest and most frequently cited trial in this space, the PACE study by Carley and colleagues (2018), demonstrated a dose-dependent reduction in AHI with dronabinol over six weeks in 73 participants. This systematic review incorporates the PACE trial along with subsequent smaller studies and reaches a conclusion entirely consistent with the AASM’s position: promising mechanistic signals exist, but the clinical evidence is insufficient for recommendation. The review does not challenge prior conclusions so much as it confirms that the field has not yet progressed meaningfully beyond proof-of-concept.

Common Misreadings

The most likely overinterpretation is reading “eight of nine studies showed positive results” as strong evidence that cannabinoids work for sleep apnea. This framing obscures critical context: the nine studies collectively enrolled only 485 participants with overlapping samples in at least three reports, the median treatment lasted just three weeks, and several studies carried high or unclear risk of bias. Statistical significance in small, short trials does not establish clinical significance, durability of benefit, or an acceptable safety profile for chronic use. The high proportion of positive results may also reflect publication bias, particularly among the abstract-only studies that could not be fully appraised.

Bottom Line

This systematic review confirms that cannabinoids, primarily dronabinol, produce short-term reductions in sleep apnea severity in small trials, but it does not establish that these benefits are clinically meaningful, durable, or safe over the long term. Adverse event rates of 70 to 80% in studies lasting only weeks underscore serious tolerability concerns. The evidence does not support clinical use of cannabinoids for OSA, and the AASM recommendation against their use remains appropriate pending larger, longer randomized controlled trials.

References

1. Engstrom E, et al. Cannabinoids for obstructive sleep apnea: a systematic review. PROSPERO registration CRD42024555089. 2024.
2. Carley DW, Prasad B, Reid KJ, et al. Pharmacotherapy of apnea by cannabimimetic enhancement, the PACE clinical trial: effects of dronabinol in obstructive sleep apnea. Sleep. 2018;41(1):zsx184. doi:10.1093/sleep/zsx184
3. Ramar K, Rosen IM, Kirsch DB, et al. Medical cannabis and the treatment of obstructive sleep apnea: an American Academy of Sleep Medicine position statement. J Clin Sleep Med. 2018;14(4):679-681. doi:10.5664/jcsm.7070