Long-term ozone exposure and adverse outcomes in atrial fibrillation: the population-based ARS-PARTICLES cohort

INTRODUCTION: Patients with atrial fibrillation (AF) remain at high risk of cerebrovascular and cardiovascular events, but evidence on long-term ozone (O3) exposure in this population is limited. OBJECTIVES: We examined whether long-term O3 exposure was associated with adverse cerebrovascular and cardiovascular outcomes. PATIENTS AND METHODS: We analyzed the population-based ARS-PARTICLES cohort of 20757 adults with prevalent AF residing in Podlaskie Voivodeship, northeastern Poland, in 2014, with follow-up through December 31, 2019. O3, nitrogen dioxide and fine particulate matter concentrations were estimated using the GEM-AQ model and assigned by residential municipality. The primary endpoint was a composite of ischemic stroke, transient ischemic attack or ischemic stroke-related death; cardiovascular mortality was the key secondary endpoint. Associations were estimated using Fine-Gray competing-risk models with annually updated previous-year O3 exposure, adjusted for clinical characteristics, meteorology, area-level socioeconomic status, co-pollutants and outdoor artificial light at night. RESULTS: The mean (SD) age was 72.6 (11.5) years, and 52.9% were women. The median follow-up was 5.6 years (interquartile range [IQR], 2.8-6), during which 3,139 primary composite events and 5,400 cardiovascular deaths occurred. The median annual O₃ concentration was 43.48 μg/m³ (IQR, 42.6-46.63). Each IQR increase in annually updated O₃ exposure (4.03 μg/m³) was associated with higher risks of the primary composite endpoint (subdistribution hazard ratio [sHR], 1.27; 95% CI, 1.02-1.59), cardiovascular mortality (sHR, 1.14; 95% CI, 1.01-1.29), and ischemic stroke (sHR, 1.38; 95% CI, 1.08-1.77). Associations were directionally consistent across adjustment strategies and sensitivity analyses. CONCLUSIONS: Long-term O3 exposure was associated with higher cerebrovascular and cardiovascular risks in patients with AF. O3 may contribute to residual risk in this clinically vulnerable population and warrants evaluation in other high-risk cardiovascular cohorts.