Comparison of performance of four adaptive servo ventilation devices in patients with complex sleep apnea

Introduction: Adaptive Servo Ventilation (ASV) increased the risk for mortality in the SERVE-HF trial. The mechanisms for such a finding remain unclear. Conceivably, device algorithms that control respiratory rate and pressure support may have led to high minute ventilation, which, in turn, may have contributed to hypocapnia and life-threatening arrhythmias. Whether such a finding was due to a device algorithm-based effect (?device-effect?) or applies to all devices with servo-algorithm (?class-effect?) is uncertain. Our objective was to compare the performance of various ASV devices on indices of sleep-disordered breathing and minute ventilation. Methods: We performed a randomized controlled cross-over trial of patients with complex sleep apnea with preserved cardiac contractility who were adherent with ASV therapy. Patients underwent four nights of laboratory-based polysomnography while receiving therapy from a Philips ASV device (System One), updated Philips ASV device (Dreamstation), ResMed S7 VPAP Adapt (used in the SERVE-HF trial), and ResMed S9 VPAP Adapt. Results: During polysomnography, apnea-hypopnea index was not different across the different devices (P=0.9). Minute ventilation (VE) was greater during S7 device (9.8 ± 0.4 Lpm) when compared to all other devices for wakefulness state: S9 (4.9 ± 1.0 Lpm), Dream Station (8.4 ± 0.2 Lpm), and System One (8.6 ± 0.3 Lpm; P<0.0001). VE was greater during S7 device therapy when compared to both Philips devices for stage N1 and REM sleep (P<0.01). Respiratory rate was greater during S7 device therapy when compared to other devices for all sleep-wakefulness states (P<0.0001). Pressure support level was greater during S7 device therapy when compared to all other devices during wakefulness (P<0.0001). Conclusion: There were significant differences in minute ventilation delivered by various ASV devices. We speculate that the mechanisms underlying the adverse effects of ASV in previous trials could be secondary to excessive ventilation due to device-effect and not due to class-effect.