Association of Positive airway pressure adherence with mortality and cardiovascular events in high vs low risk sleep apnoea patients.
BACKGROUND Positive airway pressure (PAP) has shown inconsistent effects on cardiovascular (CV) outcomes in obstructive sleep apnoea (OSA). Physiological biomarkers derived from sleep studies such as sleep apnoea-specific hypoxic burden (SASHB) and event-related heart rate response (ΔHR) may help identify OSA patients with modifiable CV risk. METHODS 3370 PAP-treated moderate-to-severe OSA patients, from the Pays de la Loire sleep-clinic cohort linked to the French health database (SNDS), were stratified by a high-risk status defined by SASHB >40.6%·min·h⁻1 or ΔHR >21.7 bpm, using percentile-based approach for threshold estimation. The primary composite outcome was defined using the first occurrence in SNDS of major adverse CV event (MACE). Cox models assessed the association between PAP adherence (mean PAP use ≥4 h/night) and MACE occurrence. RESULTS Over a median follow-up of 9 years, 740 patients experienced a MACE. PAP adherence versus non-adherence was associated with a reduced risk of MACE (adjusted hazard ratio [HR] 0.53, 95% CI [0.46-0.62], p<0.001), but the association was greater in patients with (71.7%) versus without (28.3%) high risk status (interaction HR 0.61 [0.43-0.87]; interaction p value =0.006). Similar findings were obtained using a simplified version of SASHB and ΔHR automatically derived from the single oximetry signal. The interaction of high-risk status between PAP adherence and MACE appeared stronger in non-sleepy patients. CONCLUSIONS In sleep-clinic, the association of PAP adherence with reduced MACE risk was stronger in high risk OSA. These findings support the integration of hypoxic and autonomic biomarkers into clinical decision pathways for CV risk reduction in OSA.