The effect of bilevel ventilation or modified continuous positive airway pressure on treatment adherence in adults with obstructive sleep apnea: a systematic review and meta-analysis

Background Modified continuous positive airway pressure (CPAP) modalities providing pressure support (bilevel PAP (BPAP) or expiratory pressure relief algorithms (EPRAs)) are routinely used for obstructive sleep apnea (OSA) treatment, despite little evidence of increased adherence and improved downstream outcomes. An updated, comprehensive systematic review and meta-analysis of pressure support on adherence is needed. Methods MEDLINE, EMBASE, CENTRAL, Clinicaltrials.gov and major abstract repositories were searched from database inception until 4/7/2026 for randomized and non-randomized control trials of varying design assessing the effect of pressure support (BPAP, EPRA, or both) on adherence (primary outcome) vs. CPAP in adults with OSA and in those exhibiting CPAP intolerance (exploratory). Subgroup analyses and meta-regressions assessed potential causes of heterogeneity. Publication bias was examined via Eggers' test. Epworth Sleepiness Scale (ESS) changes from baseline, residual apnoea-hypopnea index (AHI) and leak (secondary outcomes) were also analysed when available. GRADE was used for certainty of evidence (CoE) rating. This study is registered with PROSPERO (CRD420251086824). Findings In twenty-one studies (N = 1462 participants), pressure support did not affect adherence vs. CPAP (mean difference 95% CI: 0.21 −0.01, 0.44 h/night, P=0.06; heterogeneity=45%; CoE: very-low-to-moderate). BPAP had a larger effect amongst those exhibiting PAP intolerance (0.68 0.14, 1.23 h/night, P=0.01), but CoE was very-low. Longer follow-up duration was associated with decreasing adherence to pressure support vs. CPAP (−0.18 −0.36, −0.03 h/night per additional trimester beyond 3 months, P=0.03). There was no difference in ESS and residual AHI between interventions, but leak was reduced on pressure support (CoE: low-to-moderate). No publication bias was detected for any outcome. Interpretation Pressure support yields similar adherence and residual AHI to CPAP. Given increased research and development demands and device cost, its use in OSA is not supported. Future work will need to clarify the exact role of pressure support in patients intolerant to CPAP. Funding None.