Phenotyping of acute heart failure with preserved ejection fraction: real‐world outcomes in a cohort of older patients

Abstract Background Previous studies have identified distinct subgroups (phenogroups) of patients with heart failure with preserved ejection fraction (HFpEF). Aims This study aims to identify distinct phenotypes in older patients with HFpEF hospitalised for acute heart failure (AHF) and investigate the relationship between subgroups and outcomes. Methods Retrospective, single‐center study, including patients ≥65 years hospitalised for AHF over a 4‐year period. We used electronic medical records to collect clinical data, including hospital outcomes. Latent class analysis (LCA) was performed to identify clusters of clinical phenogroups. The primary outcome was all‐cause in‐hospital mortality. Results Overall, 770 patients were included. Based on LCA, three phenogroups were identified. Phenogroup 1 ( n = 323) had both the lowest burden of comorbidities and N‐terminal pro‐brain natriuretic peptides (NT proBNP) values. Phenogroup 2 ( n = 224) had the oldest patients (median age 82 years), the highest prevalence of women (62%) and atrial fibrillation and the worst right ventricular function. Phenogroup 3 ( n = 223) consisted mainly of men (57%) and had a higher prevalence of diabetes, obesity and established cardiovascular disease and the worst renal function. Phenogroups 2 and 3 showed a significantly higher risk of primary outcome than phenogroup 1. In addition, survival analysis showed that phenogroup 2 had the worst prognosis, with more than double the risk of in‐hospital death. Conclusions In this real‐world cohort of older patients with HFpEF hospitalised for AHF, we identified three subgroups with significantly different features and prognoses. Phenomapping may be an effective tool to identify individuals most likely to experience adverse outcomes, providing a basis for phenotype‐specific treatment strategies.