Utility of new-generation pacemakers in sleep apnea screening

Abstract

Objective: Patients with cardiac pacemakers present a high prevalence of undiagnosed sleep apnea syndrome (SAS). New-generation pacemakers have algorithms that identify sleep respiratory events. Our aim was to evaluate their accuracy in the diagnosis of SAS. Methods: We performed a prospective study that included patients with new-generation pacemakers (Reply 200 pacemakers). All patients underwent a polysomnography (PSG). On the same night, the respiratory disturbance index of the PSG (RDI-PSG) and of the pacemaker (RDI-PM) were recorded. The agreement between methods was assessed using the kappa coefficient, Bland and Altman statistics and receiver operating characteristic (ROC) curves. Results: Sixty patients were recruited but the RDI-PM for the PSG night was not available in six patients. PSG diagnosed SAS in 74% of patients (20% severe, 19% moderate, 35% mild). Besides snoring (63%), most patients had no SAS symptoms. There was a strong positive correlation between RDI-PSG and RDI-PM (r ¼ 0.522, p < 0.001), but the level of agreement between methods regarding SA diagnosis/severity was poor (k ¼ 0.167). ROC curves identified a RDI-PM of 10 events/h as the optimal cut-off point for diagnosing SAS (area under the curve (AUC): 0.81, sensitivity: 80%, specificity: 79%, positive predictive value: 91%, negative predictive value: 58%). The best cut-off for identifying moderate/severe SAS was at 13 events/h (AUC: 0.86, sensitivity: 100%, specificity: 70%, positive predictive value: 68%, negative predictive value: 100%). Conclusions: SAS prevalence in patients with pacemakers is high (74%). Most are asymptomatic, which could delay the diagnosis. Patients with clinical indication for a pacemaker may benefit from a device with sleep apnea monitoring.