Impact of atrial fibrillation on long-term clinical outcomes in heart failure outpatients

Abstract

Purpose: Atrial fibrillation (AF) is a common arrhythmia in patients with heart failure (HF). Recently, several reports revealed that AF was not associated with adverse long-term clinical outcomes of hospitalized HF patients. However, the impact of AF in unselected HF outpatients on long-term clinical outcomes remained unclear. In the present study, we aimed to clarify the impact of AF on clinical outcomes among Japanese HF outpatients in real-world clinical practice. Methods: With a single hospital-based cohort in the Shinken Database 2004-2011, comprising all the new patients (n=17517) visiting the Cardiovascular Institute Hospital, we followed 2024 outpatients who were diagnosed as having symptomatic HF at the initial visit. Results: AF was observed in 310 patients (15%). Patients with AF were older, and more likely to be female, more likely to have a lower rate of obesity, hypertension, dyslipidemia, diabetes mellitus, ischemic heart disease, whereas, had a higher rate of anemia, chronic kidney disease, valvular heart disease, dilated cardiomyopathy and prior history of stroke. Averaged BNP in patients with AF was significantly higher than that in patients without AF. AF group had a lower prevalence of NYHA II, whereas a higher prevalence of NYHA III and IV heart failure than Non-AF group. Patients using beta-blockers, renin-angiotensin-system inhibitors, diuretics, digitalis, anti-arrhythmic drugs, and warfarin were more common in patients with AF. Left ventricular ejection fraction was lower in patients with AF than those without. Cox regression analysis showed that AF was associated with higher incidence of all-cause death (p=0.015, hazard ratio (HR) 1.631, 95% confidence interval (CI) 1.099-2.420), cardiovascular death (p=0.009, HR 2.022, 95% CI 1.196-3.420), and HF admission (p<0.001, HR 3.107, 95% CI 2.286-4.224). The Cox regression model used in the analysis adjusted for the covariates showed that patients with AF had a comparable risk for all-cause death (p=0.115, HR 0.573, 95% CI 0.286-1.146), HF death (p=0.608, HR 0.690, 95% CI 0.167-2.844), and cardiovascular death (p=0.564, HR 0.772, 95% CI 0.320-1.861), whereas had significantly higher risk for HF admission (p=0.007, HR 1.781, 95% CI 1.172-2.704). Sub analysis showed that the clinical impact of AF on HF outpatients might be stronger in patients with female, younger age, ischemic etiology, preserved EF, and mild HF symptom. Conclusions; Among unselected HF outpatients, AF was commonly observed. AF was not associated with long-term mortality, but independently associated with HF admission in HF outpatients.