BACKGROUND: Sleep apnea (SA) is a relevant issue in the management of patients with heart failure for risk stratification and for implementing treatment strategies.
OBJECTIVE: The purpose of this study was to evaluate in patients with implantable cardioverter-defibrillators (ICDs) the performance of the respiratory disturbance index (RDI) computed by the ApneaScan algorithm (Boston Scientific Inc., Natick, MA) as a discriminator of severe SA.
METHODS: ICD-indicated patients with left ventricular ejection fraction ≤35% were enrolled. One month after implantation, patients underwent a polysomnographic study. We evaluated the accuracy of the RDI for the prediction of severe SA (apnea-hypopnea index [AHI] ≥30 episodes/h) and the agreement between RDI and AHI during the sleep study night.
RESULTS: Two hundred sixty-five patients were enrolled to obtain the required sample of 173 patients with AHI and RDI data for analysis. The mean AHI was 21 ± 15 episodes/h and severe SA was diagnosed in 38 patients (22%), while the mean RDI was 33 ± 13 episodes/h. On the basis of the receiver operating characteristic curve analysis of RDI values, the area under the curve was 0.77 (95% confidence interval [CI] 0.70-0.83; P < .001). At an RDI value of 31 episodes/h, severe SA was detected with 87% (95% CI 72%-96%) sensitivity and 56% (95% CI 48%-66%) specificity. RDI closely correlated with AHI recorded during the same night (r = 0.74; 95% CI 0.57-0.84; P < .001), and the Bland-Altman agreement analysis revealed a bias of 11 episodes/h, with limits of agreement being -10 to 32 episodes/h.
CONCLUSION: The RDI accurately identified severe SA and demonstrated good agreement with AHI. Therefore, it may serve as an efficient tool for screening patients at risk of SA.