Background: Although several conditions have been proposed as risk factors contributing to the incidence of atrial fibrillation, many individuals without such ‘risk factors’ also suffer from atrial fibrillation. The present study tested the hypothesis that the risk of new-onset atrial fibrillation increases with increasing circulating levels of B-type natriuretic peptide in the general population.

Methods: Participants in our health checkup program without atrial fibrillation or a history of atrial fibrillation were enrolled (n=10,058, 54.3±11.3 years old). After baseline evaluation, subjects were followed up for the median of 1,791 days with the endpoint being the new onset of atrial fibrillation.

Results: Atrial fibrillation occurred in 53 subjects during the follow-up period (1.16 per 1,000 person-year). The risk of new-onset atrial fibrillation increased across the gender-specific quartiles of B-type natriuretic peptide levels at baseline. In multivariate Cox proportional hazard regression analysis where B-type natriuretic peptide concentrations were taken as a continuous variable, B-type natriuretic peptide was a significant predictor of new onset of atrial fibrillation after adjustment for possible factors (hazard ratio 5.65 [95% CI 2.63–12.41]).

Conclusions: The risk of new onset of atrial fibrillation increases with increasing B-type natriuretic peptide levels in the general population. Measurement of B-type natriuretic peptide may improve the prediction of incident atrial fibrillation.