Article ID: CJ-20-0864
Background:The associations between body mass index (BMI) and incidence of atrial fibrillation (AF) in young men are scarce, especially in Asian countries, given the differences in BMI between Asians and Western populations.
Methods and Results:This study analyzed 17,865 middle-aged Japanese men without AF from a cohort of employees undergoing annual health examinations. AF incidence was evaluated during a follow-up period (median 4.0 years, interquartile range 2.0–7.1 years). Among young men aged 30–49 years, AF incidence was 0.64/1,000 person-years, whereas it was 2.54/1,000 and 7.60/1,000 person-years among men aged 50–59 and ≥60 years, respectively. Multivariable Cox regression analysis among young men revealed age (hazard ratio [HR] 3.28 by 10-years’ increase, 95% confidence interval [CI] 1.72–6.25, P<0.001), BMI (BMI-quadratic, HR 1.01, 95% CI 1.00–1.01, P<0.001, BMI-linear, HR 0.95, 95% CI 0.86–1.05, P=0.33), and electrocardiogram (ECG) abnormalities, such as PQ prolongation, supraventricular beat, and p wave abnormality (HR 8.79, 95% CI 3.05–25.32, P<0.001), were significantly associated with AF incidence. There was a reverse J-shaped association between BMI and AF incidence in young men, whereby the presence of ECG abnormality inversely influenced the BMI-incident AF relationship. A linear association between BMI and AF incidence in men aged 50–59 and ≥60 years was present.
Conclusions:AF incidence displays a reverse J-shaped relationship with BMI in young men, but a linear association in men aged ≥50 years. The paradoxical relationship seen in young men only may reflect atrial electrical or structural abnormalities.
