Relationship between QRS duration and incident atrial fibrillation

BACKGROUND: QRS duration (QRSd), a measure of ventricular conduction, has been associated with adverse cardiovascular outcomes, but its relationship with incident atrial fibrillation (AF) is poorly understood. METHODS AND RESULTS: This study included 15,314 participants from the Atherosclerosis Risk in Communities (ARIC) study who were free of AF at baseline. QRSd was automatically measured from resting 12-lead electrocardiograms (ECGs) at baseline. Incident AF cases were systematically ascertained using ECGs, hospital discharge diagnoses and death certificates. Multivariable adjusted Cox regression analyses were performed to investigate the relationship between QRSd and incident AF. Mean age of our population was 54 ± 6 years (55% females). During a median follow-up of 21.2 years, 2041 confirmed incident AF cases occurred. In multivariable adjusted Cox models, a 1-SD increase in QRSd was associated with a hazard ratio (HR) (95% CI) for AF of 1.05 (1.01; 1.10), p = 0.01. This relationship was significant among women (HR per 1-SD increase in QRSd (95% CI) 1.13 (1.06; 1.20), p < 0.001), but not among men (1.00 (0.95; 1.06), p = 0.97) (p for interaction 0.005). Compared to individuals with a QRSd <100 ms, the HRs for incident AF in individuals with a QRSd of 100-119 and ≥120 ms were 1.13 (1.02; 1.26) and 1.35 (1.08; 1.68), respectively (p for trend 0.002). Again, this relationship was significant among women (p for trend <0.001) but not among men (p for trend 0.23). CONCLUSION: In this large population-based study, QRSd was an independent predictor of incident AF among women, but not in men. Further studies are needed to better understand the underlying mechanisms.