Respiratory disturbance index as a predictor of atrial fibrillation

Abstract Background Obstructive sleep apnea (OSA) is a risk factor for development of atrial fibrillation (AF). Recent pacemakers have specific algorithms for detecting of OSA based on impedance measurement. Purpose The aim was to determine the burden of AF in patients with high respiratory disturbance index (RDI). It was also intended to determine if the onset of continuous positive airway pressure (CPAP) reduced the AF burden. Methods Retrospective observational study carried out through the analysis of the patient records. It included patients with pacemakers evaluated in an outpatient clinic during the first 8 months of 2018. All had the RDI algorithm. AF burden was assessed according to automatic mode switch (AMS) duration. The percentage of nights with an RDI>20/h was analysed to assess the risk of OSA. Results 108 patients were included (mean age=70.3±9.0 years; 56.5% men). The most prevalent indication for cardiac pacing was atrioventricular node dysfunction (59.3%). The average percentage of nights with an RDI>20/h was 44.2%. The average follow-up (FU) period was 4.1 years. Of all cardiovascular risk factors, diabetes mellitus was the only one associated with a higher incidence of AF at FU (p=0.044). Considering just the patients with an RDI >20/h in more than 20% of nights (RDI20%), male gender was associated with higher rate of major adverse cardiovascular events [(acute myocardial infarction, stroke, acute heart failure and death) (p=0.019)]. Considering patients with AF prior to pacemaker implantation (n=34), 76% of patients had an RDI >20/h in more than 20% of nights (mean of 58.1%). 23 of these patients had long standing persistent AF at the end of FU (vs 17 at the beginnig of FU). Considering patients who developed AF after pacemaker implantation (n=24), 58% of patients had an RDI>20/h in more than 20% of nights (mean of 45.3%). Most patients had paroxismal AF (83%). Patients with prior paroxysmal AF or those who developed AF during FU (n=37) had a higher burden of AF at the end of FU when they had RDI >20/h in more than 50% of nights (mean burden of 24.66% vs 14.69%, p=0.005). There was no statistically significant correlation between the percentage of nights with RDI >20/h and the Apnea Hypopnea Index (AHI) value. Patients who underwent polysomnography (n=30) had a mean AHI of 16/h (which corresponds to moderately severe OSA). 63% of patients who were referred from the pacemaker consultation to pneumology and performed polysomnography started CPAP. CPAP use was not associated with a statistically significant reduction in any endpoints or AF burden. Conclusion A high RDI was associated with onset of de novo AF and higher AF burden in patients with paroxysmal AF. Thus, there is a rationale for early diagnosis and treatment of OSAS to prevent and reduce AF progression. In this study, CPAP was not shown to significantly reduce this endpoint, probably due to reduced polysomnography and CPAP treatment. Funding Acknowledgement Type of funding sources: None.