The atrial fibrillation paradox of heart failure.

The prevalence of atrial fibrillation (AF) in patients with heart failure (HF) is high, but longitudinal studies suggest that the incidence of AF is relatively low. The authors investigated this paradox prospectively in an epidemiologically representative population of patients with HF and persistent AF. In all, 891 consecutive patients with HF [mean age, 70+/-10 years; 70% male; left ventricular ejection fraction, 32%+/-9%] were enrolled. The prevalence of persistent AF at baseline was 22%. The incidence of persistent AF at 1 year was 26 per 1000 person-years, ranging from 15 in New York Heart Association class I/II to 44 in class III/IV. AF occurred either at the same time or prior to HF in 76% of patients and following HF in 24%. A risk score was developed to predict the occurrence of persistent AF. The annual risk of persistent AF developing was 0.5% (0%-1.3%) for those in the low-risk group compared with 15% (3.4%-26.6%) in the high-risk group. Despite a high prevalence of persistent AF in patients with HF, the incidence of persistent AF is relatively low. This is predominantly due to AF coinciding with or preceding the development of HF. The annual risk of persistent AF developing can be estimated from clinical variables.