Efficacy of pharmacotherapy for OSA in adults: A systematic review and network meta-analysis. (August 2019)

• Record Type: Journal Article
• Authors: Gaisl, Thomas; Haile, Sarah R.; Thiel, Sira; Osswald, Martin; Kohler, Malcolm
• Is Part Of: Sleep medicine reviews. Volume 46(2019)
• Page Start: 74
• Page End: 86
• Publication Date: 2019-08
• Subjects: Obstructive sleep apnoea -- Pharmacotherapy -- Drug -- Randomised controlled trial -- Meta-analysis; Sleep disorders -- Periodicals; Sleep Disorders -- Periodicals; Sommeil, Troubles du -- Périodiques; Electronic journals; 616.8498
• DOI: 10.1016/j.smrv.2019.04.009
• Languages: English
• ISSNs: 1087-0792

Abstract

Summary: Pharmacotherapy represents a desirable potential therapeutic alternative for patients with obstructive sleep apnoea (OSA). We aimed to summarize evidence on the efficacy of pharmacotherapy in adults with OSA and delineate the underlying mechanisms. Seven databases were systematically screened for randomised controlled trials (RCTs) from their inception to September 2018. According to a pre-registered study protocol (PROSPERO-ID-CRD42018086446) network meta-analysis was performed to obtain intervention effects on the apnoea-hypopnoea-index (AHI) based on data extracted from published reports. We identified 58 RCTs (n = 1710 patients) investigating 44 different drugs or drug-combinations. Interventions were classified into seven pathomechanism-groups and summarized narratively. A meta-analysis of 17 trials for seven drugs (acetazolamide, donepezil, mirtazapine, ondansetron, paroxetine, protriptyline, theophylline) indicated a small effect for acetazolamide (mean difference in AHI −9.6/h [−17.7; −1.4]; p = 0.02). In the network meta-analysis (I 2 = 50%) nine drugs (tramazoline, liraglutide, spironolactone/furosemide, acetazolamide, dronabinol, zonisamide, phentermine, spironolactone, and ondansetron/fluoxetine) significantly lowered the AHI compared to placebo. Although some trials indicate favorable outcomes, these results are only valid for distinctive OSA-phenotypes or were not clinically significant. The effect sizes were small, the majority of trials were not adequately powered. There is currently insufficient evidence to recommend any pharmacotherapy for OSA and no phase-III trials are available.