Association of High Cardiovascular Disease Risk OSA With Incident Atrial Fibrillation: The Multi-Ethnic Study of Atherosclerosis

Jing Xu; Younghoon Kwon; Neda Esmaeili; Scott A Sands; Mohammadreza Hajipour; Daniel Vena; Susan R Heckbert; Elsayed Z Soliman; Andrew Wellman; Erin D Michos; Daniel J Gottlieb; Susan Redline; Ali Azarbarzin

doi:10.1016/j.chest.2026.01.027

Association of High Cardiovascular Disease Risk OSA With Incident Atrial Fibrillation: The Multi-Ethnic Study of Atherosclerosis

Chest. 2026 Mar 5:S0012-3692(26)00269-2. doi: 10.1016/j.chest.2026.01.027. Online ahead of print.

Authors

Affiliations

¹ Division of Sleep and Circadian Disorders, Brigham and Women's Hospital and Harvard Medical School, Boston, MA; Department of Respiratory and Critical Care Medicine, The Affiliated Huai'an No.1 People's Hospital of Nanjing Medical University, Huaian, Jiangsu, China.
² Division of Cardiology, Department of Medicine, University of Washington, Seattle, WA.
³ Division of Sleep and Circadian Disorders, Brigham and Women's Hospital and Harvard Medical School, Boston, MA.
⁴ Division of Experimental Medicine, Department of Medicine, University of British Columbia, Vancouver, BC, Canada.
⁵ Department of Epidemiology, University of Washington, Seattle, WA.
⁶ Epidemiological Cardiology Research Center, Wake Forest School of Medicine, Winston-Salem, NC.
⁷ Division of Cardiology, Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, MD.
⁸ Division of Sleep and Circadian Disorders, Brigham and Women's Hospital and Harvard Medical School, Boston, MA. Electronic address: aazarbarzin@bwh.harvard.edu.

Abstract

Background: OSA is associated with atrial fibrillation (AF), but this association varies by population and definitions of OSA and AF.

Research question: Is high cardiovascular disease risk (High-CVD-Risk) OSA, defined by elevated hypoxic burden (HB) or heart rate response to respiratory events (ΔHR), associated with incident AF?

Study design and methods: Using data from the Multi-Ethnic Study of Atherosclerosis, HB was quantified as the cumulative area under the desaturation curve after respiratory events, and ΔHR was quantified as the increase in heart rate on event termination. Within OSA (apnea-hypopnea index [AHI] ≥15 events/h), High-CVD-Risk OSA was defined as having either a high ΔHR or high HB (both in the highest tertile), whereas individuals with OSA who did not meet these criteria were considered to have Low-CVD-Risk. Cox proportional hazards models were used to estimate the adjusted hazard ratios (HRs) of incident AF for High-CVD-Risk OSA and Low-CVD-Risk OSA (vs non-OSA, defined as AHI < 15 events/h) after adjusting for confounders.

Results: A total of 1,679 participants (45.4% male) were included, with a median age of 66.0 years (interquartile range [IQR], 60.0-74.0 years), HB of 35.3%min/h (IQR, 18.0-69.2%min/h), and ΔHR of 7.7 beats/min (IQR, 5.9-9.9 beats/min). During a median follow-up of 6.7 years, AF was identified in 14.1% of the patients with High-CVD-Risk OSA (n = 687), 10.1% of the patients with Low-CVD-Risk OSA (n = 278), and 8.4% of the patients with non-OSA (n = 714). Compared with the non-OSA group, an increased HR for AF was found in the High-CVD-Risk OSA group (HR, 1.68; 95% CI, 1.17-2.41), but not in the Low-CVD-Risk OSA group (HR, 1.20; 95% CI, 0.76-1.92). The adjusted HR was similar in female and male participants (HR, 1.71 [95% CI, 1.01-2.91] vs 1.64 [95% CI, 0.99-2.72]).

Interpretation: Our results show that OSA-related hypoxemia or heart rate surge may be useful for predicting which patients with OSA are at increased risk of incident AF.

Keywords: OSA; atrial fibrillation; heart rate; high risk; hypoxic burden.

Abstract

Grants and funding