Evaluation of the Apnea-Hypopnea Index Determined by Adaptive Servo-Ventilation Devices in Patients With Heart Failure and Sleep-Disordered Breathing

Abstract

Introduction: Adaptive servo-ventilation (ASV) devices are designed to suppress central respiratory events, and therefore effective for sleep-disordered breathing (SDB) in patients with heart failure (HF) and provide information about their residual respiratory events. However, whether the apnea-hypopnea index (AHI), determined by the ASV device AutoSet CS (ASC), correlates with the AHI calculated by polysomnography (PSG) in patients with HF and SDB remains to be evaluated. Methods: Consecutive patients with SDB titrated on ASC were included in the study. We assessed the correlation between AHI determined by manual scoring during PSG (AHI-PSG) and that determined by the ASC device (AHI-ASC) during an overnight session. Results: Thirty patients with HF and SDB (age, 68.8 ± 15.4 years; two women; left ventricular ejection fraction, 53.8 ± 17.9%) were included. The median AHI in the diagnostic study was 28.4 events/h, including both obstructive and central respiratory events. During the titration, ASC markedly suppressed the respiratory events (AHI-PSG, 3.3 events/h), while the median AHI-ASC was 12.8 events/h. We identified a modest correlation between AHI-PSG and AHI-ASC (r = 0.36, p = 0.048). The Brand-Altman plot indicated that the ASC device overestimated the AHI, and a moderate agreement was observed with PSG. Conclusions: There was only a modest correlation between AHI-PSG and AHI-ASC. The discrepancy may be explained by either the central respiratory events that occur during wakefulness or the other differences between PSG and ASC in the detected respiratory events. Therefore, clinicians should consider this divergence when assessing residual respiratory events using ASC.