Background: Estimates of the prevalence of atrial fibrillation (AF) in heart failure (HF) originate from patients enrolled in clinical trials.
Aims: To assess the prevalence and clinical correlates of AF among HF patients in everyday clinical practice from HF patients screened for the T-wave ALternans in Patients with Heart fAilure (ALPHA) study; to investigate the correlation between AF and functional status.
Methods and results: Consecutive patients (N=3513) seen at nine Heart Failure Clinics were studied; 21.4% were in AF. AF prevalence was greater with increasing age (OR 1.04/year, p<0.001) in non-ischaemic cardiomyopathy (OR 2.34, p<0.001) and with increasing NYHA class (p<0.0001). Multiple logistic regression predictors of AF were age >70 years (OR 2.35), NYHA class II III or IV vs class I (OR 1.8, 4.4 and 3.1) and non-ischaemic cardiomyopathy (OR 3.2). A logistic model indicated that AF was associated with a 2.5 OR of being in NYHA class III-IV vs I-II while accounting for age, gender, left ventricular ejection fraction (LVEF), and aetiology of HF.
Conclusions: The prevalence of AF in HF patients exceeds 20%, and increases with age and functional class. The presence of AF leads to a more severe NYHA class, indicating that AF contributes to the severity of heart failure.