LATE-BREAKING ABSTRACT: Adaptive Servo-Ventilation (ASV) Pressures and Cardiovascular (CV) Mortality Risk in SERVE-HF

Introduction: The SERVE-HF study showed that treatment of predominant central sleep apnoea (CSA) with ASV increased the risk of all-cause and CV death in systolic heart failure patients. It has been suggested that higher ASV pressures might be a contributor to worse outcomes. Aim: To investigate whether CV mortality risk was modified by different ASV pressure levels in SERVE-HF. Methods: CV mortality rates were determined in patient subgroups based on delivered end-expiratory positive airway pressure (EPAP), pressure, peak inspiratory pressure (PIP) and leak (median, maximum, 95th percentile values). Results: There were no significant differences in the risk of CV death, CV death without prior hospitalisation or all-cause mortality between any patient subgroups based on ASV pressures (median, maximum, 95th percentile values for EPAP, pressure and PIP) or leak. Partial pressures of oxygen and carbon dioxide, base excess and pH remained stable throughout the study. Conclusion: This analysis did not show any relationship between ASV pressures and risk of CV death in SERVE-HF. In addition, blood gases were similar throughout the treatment period. This suggests that the increased risk of CV death associated with randomization to ASV in patients with chronic heart failure with left ventricular ejection fraction ≤45% and predominant CSA is independent of pressure settings.