Persistent atrial fibrillation as a prognostic factor of outcome in patients with advanced heart failure

INTRODUCTION: Chronic heart failure (CHF) is associated with high morbidity and mortality and is diagnosed more and more frequently. Fifteen to 30% of patients with systolic CHF develop atrial fibrillation (AF). AIM: To establish whether persistent AF was an independent predictor of mortality, and had a predictive value with respect to late clinical outcomes in patients with systolic CHF. METHODS: Analysis comprised 120 men with systolic CHF. In 35 (58%) patients CHF was the result of ischaemic heart disease and in 25 (42%)--idiopathic dilated cardiomyopathy (DCM). Presence or absence of AF was a criterion of patients' subsequent division into two subgroups. Sixty patients with AF were assigned to the AF group. The control group involved 60 individuals with CHF and sinus rhythm (SR) on enrollment. Mean follow-up time was 36 months. RESULTS: Overall 59 (49%) patients died during 3-year follow-up, including 33 (56%) in the AF group. Deaths were noted more often in CHF patients with underlying ischaemic heart disease than DCM (66% vs 34%). This difference reached statistical significance in the AF group (72% vs 28%, p<0.001). Moreover, patients with AF more often complained of palpitations (p<0.01), had worse exercise capacity (p<0.01) as well as more frequently presented complex ventricular arrhythmia (p<0.01). The rate of hospital readmission was also higher (p<0.02). In univariate as well as multivariate analysis, AF was not found to be an independent predictor of mortality. Factors with a potential impact on adverse prognosis were concomitant complex ventricular arrhythmias (p=0.01), diabetes (0.04) and reduced exercise capacity (p<0.01). CONCLUSIONS: Persistent AF is not an independent risk factor of death in patients with advanced systolic CHF. However, it has an unfavourable impact on functional status. Concomitant complex ventricular arrhythmias and reduced exercise capacity worsen prognosis in this group of patients.