Clinical question
Are orexin antagonists safe and effective for treatment of primary insomnia?
Bottom line
Orexin antagonists help people fall asleep about 9 minutes faster and increase total sleep time by about 19 minutes versus placebo over the course of 1 to 3 months. About 8% of those taking orexin antagonists will experience next-day somnolence compared with 2% taking placebo.
Evidence
Seven systematic reviews of RCTs were found1-7; we report mainly on the most recent and comprehensive review. Differences were statistically significant unless indicated.
Versus placebo
- In a systematic review1 (13 RCTs, 7875 patients, mean age about 55 years) of sleep diary outcomes, orexin antagonists changed the following after 1 to 3 months:
— Sleep onset (65 min at baseline) was about 9 minutes faster with orexin antagonists (47 min until sleep onset) than with placebo (56 min).
— Total sleep time increased by about 19 minutes.
— Time awake after falling asleep was about 9 minutes less.
— Sleep quality improved about 5% (ie, improved 0.2 points on 4-point scale, which is not likely clinically meaningful).8
— No clinical differences between number of awakenings or feeling refreshed on awakening.
- Other reviews had similar findings.2-6
— Insomnia score response6: 55% versus 42% (placebo), number needed to treat=8.
- An RCT with 12 months’ follow-up had similar results.9
- Adverse events
— There was no difference in the rate of patients stopping treatment owing to adverse effects.2,3,7,10
— The most common adverse events1 reported included somnolence (8.3% vs 2.2% [placebo], number needed to harm [NNH]=16) and fatigue, dry mouth, and abnormal dreams (each about 2% to 3% vs 1% with placebo).
— Rates of excessive daytime sleepiness were 0.6% versus 0.3% (placebo), NNH=290; rates of sleep paralysis10 were 0.6% versus 0% (placebo), NNH=155.
— Effects on risk of falling are unclear (4 small observational studies in hospitals) and range from increased to decreased association.11-14
— One observational study suggests fracture risk is similar between suvorexant and z drugs.15
Versus active comparator
- An RCT comparing 5 mg or 10 mg lemborexant versus zolpidem extended release16 found the following:
— Sleep onset with lemborexant was 5 to 7 minutes better (dose dependent).
— Time awake after falling asleep ranged from no difference to being about 15 minutes better with zolpidem.
— No difference was noted in time asleep.
— Dropout owing to adverse events: 0.9% versus 2.7% with zolpidem.
- Limitations: Outcome reporting was incomplete, runins were used, and the RCTs were industry sponsored.
Context
There was similar efficacy in those older than age 65.17
Studies suggest minimal withdrawal symptoms, but there is limited evidence.2,14,16,18
- Abuse potential not formally assessed in insomnia RCTs.
Nonpharmacologic sleep restriction therapy is effective.19
Lemborexant (available in Canada) costs approximately $50 for 30 tablets.20
Implementation
Insomnia affects around 30% of patients in primary care populations.21 Management requires addressing contributing factors. In a primary care cohort,21 primary insomnia constituted 12% of insomnia cases but other overlapping factors included depression and anxiety (50%), general health problems (43%), restless leg syndrome (22%), sleep apnea (9%), and alcohol or other substance use problems (12%). Guidelines recommend treating insomnia with cognitive behavioural therapy, noting that sleep consolidation and dysfunctional beliefs about sleep should be incorporated into treatment.22 Validated online tools for practitioners and patients are available.23
Notes
Tools for Practice articles in CFP are adapted from peer-reviewed articles at http://www.toolsforpractice.ca and summarize practice-changing medical evidence for primary care. Coordinated by Dr G. Michael Allan and Dr Adrienne J. Lindblad, articles are developed by the Patients, Experience, Evidence, Research (PEER) team and supported by the College of Family Physicians of Canada and its Alberta, Ontario, and Saskatchewan Chapters. Feedback is welcome at toolsforpractice{at}cfpc.ca.
Footnotes
Competing interests
None declared
This article is eligible for Mainpro+ certified Self-Learning credits. To earn credits, go to https://www.cfp.ca and click on the Mainpro+ link.
Cet article se trouve aussi en français à la page 185.
- Copyright © 2024 the College of Family Physicians of Canada
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.