The impact of atrial fibrillation on clinical outcomes in heart failure with mid-range and preserved ejection fraction patients

BACKGROUND The combined effect of left ventricular ejection fraction (LVEF) and atrial fibrillation (AF) on clinical outcomes in heart failure (HF) remains complex. OBJECTIVE In this post hoc analysis of the TOPCAT trial, we aimed to evaluate the impact of AF on clinical outcomes in patients with HF stratified by LVEF range. METHODS A total of 3442 patients were included, stratified into 3 groups according to LVEF range-HF with mid-range ejection fraction (HFmrEF), LVEF of 45%-50% (n = 823); HF with preserved ejection fraction (HFpEF), LVEF of 51%-60% (n = 1682); and HF with normal ejection fraction (HFnEF), LVEF >60% (n = 937)-and subdivided according to the presence of AF at enrollment. Cox regression analysis was used to define independent associations between AF and clinical outcomes. RESULTS AF was prevalent in 38.6% in HFmrEF, 34.6% in HFpEF, and 33.7% in HFnEF (P = .07). AF was associated with worse primary outcome in each subgroup and with HF hospitalizations and worse cardiovascular mortality in HFpEF and HFnEF. The hazard ratio for the primary outcome in those with AF compared with sinus rhythm (SR) was 1.11 (1.01-1.22; P = .03) in HFmrEF, 1.20 (1.11-1.29; P < .001) in HFpEF, and 1.16 (1.05-1.28; P = .004) in HFnEF. When LVEF was treated as a continuous variable, there was a linear negative association between LVEF and the effect of AF vs SR for the primary end point and HF hospitalizations and a linear positive association for cardiovascular mortality. CONCLUSION Compared with SR, AF was independently associated with worse outcomes across all LVEF ranges.