Prognostic impact of obstructive sleep apnea on major cardiovascular outcomes in patients with atrial fibrillation

Abstract

There is a paucity of data examining the prognostic significance of obstructive sleep apnea (OSA) in patients with atrial fibrillation (AF). This was a prospective, multicenter, study of patients with AF and high cardiovascular risk (at least 1 of the followings: age ≥ 75 years, hypertension, diabetes mellitus, stroke, chronic kidney disease, or coronary artery disease) were recruited for a home-based overnight sleep study using a FDA approved sleep diagnostic device. All the sleep tracings were manually scored by a sleep technologist using the latest American Academy of Sleep Medicine scoring guideline. The recruited patients were categorized based on the apnea-hypopnea index (AHI), into OSA (AHI ≥15 events/hour) versus non-OSA (AHI <15 events/hour) group. The primary endpoint was Major Adverse Cardiac and Cerebrovascular Event (MACCE), defined as time to first occurrence of all-cause mortality, myocardial infarction, stroke, or heart failure hospitalization. A total of 1,053 patients were recruited and underwent an overnight sleep study between 2019 and 2025. OSA was diagnosed in 55.2% (n=581) of the patients. Comparing the OSA and non-OSA group, the AF was persistent or permanent in 86.1% vs 86.4% of the OSA and non-OSA group, respectively (p=0.858). Prior AF ablation was performed in 6.2% vs 9.3% of the OSA and non-OSA group, respectively (p=0.060). The CHADS-Vasc score for the OSA and the non-OSA group was 3.9±1.6 vs 3.5±1.6, respectively (p=0.015). Anticoagulation usage on 92.3% vs 88.6% of the OSA and non-OSA group, respectively (p=0.060). Of the 581 patients in the OSA group, 37 (6.4%) used OSA treatment (CPAP, n=28; oral appliance, n=9) for over 6 months. Over a mean follow-up of 3.1±1.7 years, the crude incidence of MACCE was higher in the OSA than the non-OSA group (24.3% vs 17.2%, p=0.005). Among the 4 components, the incidences of all-cause mortality for the OSA vs non-OSA group was 9.5% vs 7.2% (p=0.189), myocardial infarction was 2.4% vs 3.0% (p=0.577), stroke was 6.7% vs 3.6%, p=0.025), and heart failure hospitalization was 10.0% vs 7.0% (p=0.086). Multivariate Cox regression analysis indicated that OSA was an independent predictor of MACCEs, with an adjusted hazard ratio of 1.43 (95% confidence interval 1.07–1.92; p=0.016), independently of age, sex, body mass index, persistent or permanent AF, prior AF ablation, hypertension, chronic kidney disease, and systolic blood pressure. In addition, OSA was an independent predictor of stroke (HR 1.94, 95% CI 1.06–3.56; p=0.033). Sensitivity analyses were performed using a more parsimonious adjustment model, with results remaining consistent. OSA is independently associated with an increased risk of MACCE in patients with AF and high cardiovascular risk. OSA is also an independent predictor of stroke. These findings support consideration of OSA status in cardiovascular risk assessment and follow-up planning in AF.Demographic and Clinical CharacteristicsMACCE analysis (Kaplan–Meier curve)