Abstract
We investigated the association between lung function and atrial fibrillation (AF) in 21,349 adults without AF aged ≥ 40 years who underwent spirometry. The study participants were enrolled from the Korean National Health and Nutritional Examination Survey between 2008 and 2016. The primary outcome was new-onset non-valvular AF identified from the National Health Insurance Service database. During the median follow-up of 6.5 years, 2.15% of participants developed new-onset AF. The incidence rate of AF per 1000 person-years was inversely related to the forced expiratory volume in 1 s (FEV(1)), forced vital capacity (FVC), and FEV(1)/FVC quartile. After adjustment for multiple variables, the AF risk in the lowest FEV(1) quartile was 1.64-fold higher than that in the highest quartile (hazard ratio (HR) 1.64 (95% confidence interval (CI) 1.26-2.12) for lowest FEV(1) quartile). The lowest quartile of FVC had 1.56-fold higher AF risk than the highest quartile (HR 1.56 (95% CI 1.18-2.08) for lowest FVC quartile). Although the lowest FEV(1)/FVC quartile was associated with an increased risk of AF in the unadjusted model, this increased risk was not statistically significant in the multivariable analysis. Compared to those with normal lung function, participants with restrictive or obstructive lung function had 1.49 and 1.42-fold higher AF risks, respectively. In this large nationwide cohort study, both obstructive and restrictive patterns of reduced lung function were significantly associated with increased AF risk.