New onset atrial fibrillation and long-term risk of heart failure in the general population: the Rotterdam study

Abstract

Purpose: Chronic heart failure is often a manifestation of long-standing underlying cardiac conditions, such as coronary heart disease (CHD) and left ventricular hypertrophy (LVH). Atrial fibrillation (AF) is a well-recognized consequence of heart failure which can be prevented by timely pharmacological treatment. However, limited data exists on the role of AF in the etiology of heart failure in the general population. We sought to evaluate the long-term risk of heart failure associated with new onset AF; in addition, we evaluated the interplay between AF and other cardiac conditions (CHD and LVH) in the development of heart failure. Methods: Within the Rotterdam Study, a prospective population-based cohort, we followed 6178 persons (mean age 68.6 years, 41% men) free of heart disease at baseline (1990-1993) for the occurrence of AF, CHD (defined as myocardial infarction or revascularization), and heart failure. We constructed time-dependent Cox models adjusted for traditional cardiovascular risk factors to study the effect of newly-diagnosed AF (and CHD) during follow-up and subsequent heart failure risk. Since both AF and heart failure can have a substantial time-lag between the occurrence of first symptoms and clinical diagnosis, we modeled various time-lags of up to 5 years since first diagnosis of AF in order to ascertain that AF preceded heart failure symptoms. Results: During a median 14.2 years of follow-up 584 diagnoses of AF were made and 994 participants developed heart failure. New onset AF was associated with an increased risk of heart failure (adjusted HR 3.16 [95% CI: 2.63-3.79]), which persisted even 5 years after AF diagnosis (adjusted HR 2.60 [95% CI: 1.88-3.58]). Regarding the interplay between AF and CHD, the presence of AF only, CHD only, and concomitant AF and CHD showed a graded increase in heart failure risk (adjusted HRs 3.52 [95% CI: 2.85-4.34], 4.30 [95% CI: 3.40-5.45], and 6.23 [95% CI: 3.56-10.88], respectively). A similar graded increase in risk estimates was observed for the presence of ECG detected LVH and concurrent AF. Population attributable risks were 12.2% for AF, 6.3% for CHD, and 5.2% for LVH. Conclusions: New onset AF was a strong risk factor for incident heart failure in the general population, with long-term relative risks close to those of manifest CHD. Absolute and relative risk estimates were especially high for persons with concomitant AF and CHD. Persons with AF may well constitute an important part of the general population that could potentially benefit from timely introduced intensive preventive measures to reduce the growing burden of chronic heart failure.