0430 Central Sleep Apnea and Incident Atrial Fibrillation in a Sleep Disorders Clinic
Amanda Tompkins,Sanjay Patel,Michael Genuardi,Scott McKane,Marat Fudim,Lars Lund,Reena Mehra,Nan Song,Robin Germany,Rebecca DeSensi
DOI: https://doi.org/10.1093/sleep/zsad077.0430
IF: 6.313
2023-05-01
SLEEP
Abstract:Abstract Introduction Given conflicting evidence in the literature, we sought to evaluate the relative importance of central sleep apnea (CSA) and obstructive sleep apnea (OSA) on incident atrial fibrillation (AF) risk in a cohort of patients referred for sleep evaluation. Methods We reviewed the medical records of patients without AF at baseline who underwent in-lab diagnostic polysomnography (PSG) at six sleep laboratories within the University of Pittsburgh Medical Center (UPMC) between 2004 and 2018. Baseline characteristics including apnea hypopnea index (AHI), central apnea index (CAI), and obstructive apnea index (OAI) were extracted from the electronic health record. Among those with AHI≥15, patients were categorized as CSA if CAI≥5 and CAI>OAI or OSA if OAI≥5 and OAI>CAI. Patients with AHI< 15 were included as the reference group. Incident AF was identified based on AF as a primary or secondary diagnosis on either inpatient or outpatient UPMC visits. Results We identified 24,199 patients (425 CSA, 7559 OSA, 16,215 no sleep-disordered breathing [reference]) for analysis who had a mean age of 49, mean body mass index (BMI) 34, and 48% male. The incidence rates for AF ranged from 1.07 per 100 person-years in the reference group to 1.91 and 2.60 in the OSA and CSA groups (p< 0.01). Using Cox proportional hazards models, the hazard ratio for incident AF in the OSA group compared to reference was 1.78 (95% CI [1.63-1.94], p< 0.01) and in the CSA group was 2.48 (95% CI [1.92-3.21], p< 0.01). After adjusting for age, sex, race, BMI, and baseline diagnoses of diabetes, heart failure, prior heart attack, and prior stroke, OSA was no longer associated with incident AF risk (HR=1.01, 95% CI [0.92-1.12], p=0.76). However, a potentially important association between CSA and incident AF persisted (HR=1.28, 95% CI [0.99-1.67], p=0.06). Conclusion We found that CSA was associated with a 28% increased rate of incident AF independent of traditional risk factors, which was nearly statistically significant. This suggests CSA may be a risk factor for AF. In contrast, OSA was a risk marker but not independent risk factor for AF. Support (if any) ZOLL Respicardia, Inc.
neurosciences,clinical neurology