0453 Sleep Apnea-Specific Hypoxic Burden and Delta Heart Rate in Relation to Atrial Fibrillation in a Clinical Cohort

Abstract Introduction Mechanistic animal studies implicate intermittent hypoxia and autonomic dysfunction in the pathogenesis of obstructive sleep apnea (OSA) and atrial fibrillation (AF). Polysomnographic correlates of hypoxia and autonomic fluctuations in OSA may be useful biomarkers of AF development. However, to date, novel, reflective physiologic sleep apnea-specific hypoxic burden (SAHB) and delta heart rate (DHR) have yet to be examined in AF. We hypothesize these novel measures are associated with incident AF. Methods Cleveland Clinic patients (age≥18) without AF who underwent polysomnography 1/2/2000-12/30/2017 were retrospectively examined. Cox proportional hazards models evaluated time from sleep study to AF by diagnosis code. SAHB (area under the respiratory event-related desaturation curve), DHR (respiratory event-related oximetry-based heart rate difference), apnea hypopnea index (AHI), and percent time oxygen saturation< 90% (T90) were examined as predictors adjusting for age, sex, race, body mass index(BMI), tobacco use, cardiopulmonary disease, anti-arrhythmic medications, and positive airway pressure. Signal analysis used Python. Statistical analysis used R and SAS. Results The sample included n=15,712 patients [age 50(40,60) years, 54% female, 74% White, BMI 32(28,38) kg/m2, AHI 12.8(5.6,26.6)] over 7.3±3.2-year follow-up. In unadjusted analyses, SAHB showed monotonic increases in incident AF across quartiles(p< 0.001). In adjusted analyses, 10-unit increased SAHB was associated with 2% increased incident AF (HR=1.02,95%CI=1.01-1.03). Similarly, 10-unit increased AHI and T90 were associated with 3% (HR=1.03,95%CI=1.01-1.06) and 8% (HR=1.08,95%CI=1.05-1.11) increased AF, respectively. DHR was analyzed in a subset with available data [n=5,930, age 46(36,56) years, 52% female, 76% White, BMI 33(28,39) kg/m2, AHI 11.6(5.2,24.7)] over 6.7±.3.2-year follow-up. Low vs. mid-range DHR was associated with 39% increased incident AF (HR=1.39,95%CI=1.08-1.80), attenuated with covariate adjustment; high-range DHR did not differ. Conclusion In this clinical cohort, signal-based measures of OSA-specific hypoxic burden and autonomic responses were associated with increased incident AF after accounting for confounders. Findings suggest that degree of hypoxia and a dampened heart rate response are important physiologic OSA-specific AF risk factors and precision medicine phenotypes to consider for treatment responsiveness and inclusion in clinical trials. Support (if any) American Academy of Sleep Medicine Foundation Physician Scientist Training Grant, Cleveland Clinic Neurological Institute Center for Outcomes Research & Evaluation Pilot Grant, Neuroscience Transformative Research Resource Development Award