Background: Previous studies suggest that prevalent heart failure (HF) differs based on left ventricular ejection fraction (LVEF) and left ventricular (LV) chamber size. Furthermore, the prevalence of HF with preserved ejection fraction (HFpEF) is often considered approaching, or exceeding that of HF with reduced ejection fraction in the community.
Aim: The aim of this study was to evaluate prevalent and incident HF based on LVEF and CMR-determined LV size within a large community-dwelling cohort.
Methods: Individuals from the United Kingdom Biobank (UKB) who underwent CMR and had available health record linkage to allow ascertainment of HF diagnosis were included. The cohort was analysed according to LVEF, LV end-diastolic volume (LVEDV) quartiles and LVEDV indexed to body surface area (LVEDVi).
Results: 38,129 individuals were included, comprising those with reduced LVEF (LVEF<50 %, n = 5096) and preserved LVEF (LVEF 50-60 %, n = 22,907, LVEF≥60 %, n = 10,126). Prevalent HF was highest in males with LVEF<50 %, and participants with reduced LVEF had higher rates of incident HF (p < 0.001) during the follow-up period (median = 2.46 years from CMR). Mean LVEDV and LVEDVi were largest in individuals with EF < 50 % (146.9 ± 36.2 ml and 76.8 ± 16.4 ml/m2 respectively). Compared to the smallest quartiles, the largest quartiles for LVEDV were associated with increased odds of HF (odds ratio 2.14 [95 % confidence interval 1.47-3.12], p < 0.001).
Conclusions: Over 50 % of HF cases occur in individuals with LVEF ≥50 %, however HF prevalence is highest in those with reduced LVEF, particularly in males. Larger LV size is associated with increased HF across the LVEF spectrum.
Keywords: Cardiac magnetic resonance imaging; Ejection fraction; Heart failure.
Copyright © 2024 Elsevier B.V. All rights reserved.