Predictors and prognostic implications of incident heart failure following the first diagnosis of atrial fibrillation in patients with structurally normal hearts: the Belgrade Atrial Fibrillation Study

Atrial fibrillation (AF) commonly co-exists with heart failure (HF). The risk factors for and prognostic implications of incident HF development in patients with first-diagnosed AF and structurally normal hearts are poorly defined. In a cohort of patients with first-diagnosed AF and structurally normal hearts on baseline echocardiography, we investigated baseline risk factors for the development of incident HF and tested the hypothesis that incident HF was an independent predictor of adverse outcomes during a mean 10-year follow-up period. This was a registry-based, observational cohort study of 842 patients initially diagnosed between 1992 and 2007 (mean age 51.6 12.4 years), whereby 83 (9.9) developed HF. The linearized rate of incident HF was 0.97 [95 confidence interval (CI) 0.781.19] per 100 patient-years. Baseline history of hypertension, diabetes mellitus, dilated left atrium, and low-normal LVEF (5054) were multivariable predictors of subsequent HF (all P 0.05). HF development was significantly associated with increased number of hospitalizations, AF progression, any stroke/peripheral thrombo-embolism, ischaemic stroke, cardiovascular death, and all-cause mortality (all P 0.001). KaplanMeier 10-year estimates of survival free of the composite endpoint of AF progression, thrombo-embolism, and mortality were significantly worse for AF patients with incident HF compared with those without HF (68.8; 95 CI 64.772.9 vs. 25.9 95 CI 15.736.1, P 0.001). Underlying co-morbidities or subtle alterations such as mild left atrial dilatation or low-normal LVEF in the absence of overt underlying heart disease are baseline independent risk factors for incident HF during a long-term follow-up. Incident HF was an independent predictor of adverse outcomes in patients initially diagnosed with first-diagnosed AF and structurally normal hearts. These findings could facilitate the identification of AF patients at increased risk for adverse outcomes within the cohort perceived as being at low risk given a structurally normal heart on echocardiography.