Abstract 13171: Complex Heart Failure Phenotypes May Occur Across the Spectrum of Lvef With Important Implications for Etiology and Clinical Outcomes

BackgroundHeart failure (HF) is typically classified according to ejection fraction (EF). The underlying pathophysiology is a heterogeneous process that may be better characterized by clinical phenotypes.MethodsIndividual patient data from 2130 patients enrolled in HF-ACTION (LVEF ≤ 35%) and 1767 patients enrolled from the Americas in TOPCAT (LVEF ≥ 45%) were harmonized and aggregated. Cluster-based HF phenotypes were identified using latent class analysis based on 13 clinical characteristics (Table) but excluding LVEF. LVEF distribution within phenotypes was then quantified following phenotyping. The primary outcome was a composite of cardiovascular mortality (CVM) or HF hospitalization (HFH). Secondary outcomes included CVM, HFH, and all-cause mortality (ACM). Phenotype characteristics were analyzed in aggregate and outcomes were stratified by LVEF (< 45% = HFrEF vs. ≥ 45% = HFpEF).ResultsSix phenotypes were identified with distinct clinical profiles (Table)nonischemic CM in young patients without significant comorbidities (A), diabetic/insulin resistant CM (B), premature ischemic CM (C), older patients with CAD and multiple risk factors (D), older patients with AF and hypertension but little CAD (E), and hypertensive, non-ischemic, non-white subjects (F). Rates of all outcomes varied significantly with phenotype (p