Comparative efficacy of non-pharmacological interventions for adults with sleep apnea: A systematic review and network meta-analysis
Spyridon N. Papageorgiou, Ioannis Konstantinidis, Alexandra K. Papadopoulou, Fani Apostolidou-Kiouti, Ioannis Avgerinos, Athanasia Pataka, Theodore Eliades, Απόστολος Τσάπας, Anna‐Bettina Haidich
Abstract
BACKGROUND: Sleep apnea is associated with cardiovascular risk, work productivity, occupational/traffic accidents, and quality-of-life (QoL); however uncertainty exists regarding optimal treatment. We performed a systematic review on the efficacy of non-pharmacological interventions for adults with sleep apnea. METHOD: We searched MEDLINE, Scopus, Virtual-Health-Library and Web-of-Science through June 2023 for parallel/cross-over randomized trials on adults with sleep apnea (apnea-hypopnea-index>5 events/hour). Study selection, data extraction and risk-of-bias assessment were performed in duplicate, followed by frequentist network meta-analyses. RESULTS: Ultimately, 197 unique trials were included (15,931 patients; mean age 51.4 years; 78.9 % male) assessing 25 treatments. Positive Airway Pressure (PAP) (alone or combined with health behaviour modification) consistently improved more apnea-hypopnea-index or daytime sleepiness and physical/mental QoL in obstructive sleep apnea (OSA) patients compared to all other interventions but was not always well-tolerated. Mandibular advancement devices (MAD) yielded the greatest improvement in depression, while also improving objective/subjective apnea-outcomes, and physical/mental QoL-albeit less than PAP and less for moderate/severe cases. Acupuncture, health behaviour modifications, surgical maxillomandibular advancement, minor oral surgery, oropharyngeal training, oxygen supplementation, or electrical neurostimulation might improve apnea-related outcomes, but weak evidence exists. Finally, electrical neurostimulation performed best for central sleep apnea and PAP performed best for positional OSA. Confidence in the network meta-analysis estimates was low due to non-adherence issues that was rarely directly assessed in included trials with objective measures. CONCLUSION: PAP (alone or with co-interventions) performed best for the treatment of adult OSA patients regardless of disease severity. For patients not tolerating PAP, MADs might be a good alternative, but confer smaller improvements overall. However, adherence issues and the heterogenous response increase the complexity of OSA treatment.