Heart Failure Phenotypes Across Age Groups in the Danish Administrative Registers

Caroline Hartwell Garred,1 Clara Friis,1 Mariam Elmegaard,1 Daniel Mølager Christensen,1 Emil Wolsk,1 Lars Køber,1,2 Morten Schou1,21Department of Cardiology, Herlev and Gentofte University Hospital, Copenhagen, Denmark; 2Department of Clinical Medicine, Faculty of Health and Science, University of Copenhagen, Copenhagen, DenmarkCorrespondence: Caroline Hartwell Garred, Department of Cardiology, Herlev and Gentofte University Hospital, Copenhagen, Denmark, Tel +45 23 60 16 06, Email caroline.amalie.hartwell.garred.01@regionh.dkBackground: The Danish administrative registers are a powerful tool for heart failure (HF) research. However, the coding system has historically does not distinguished between HF phenotypes. We aimed to identify the distribution of HF with reduced ejection fraction (HFrEF), mildly reduced EF (HFmrEF), and preserved EF (HFpEF) within the Danish National Patient Register overall and across age groups.Methods: We manually reviewed electronic health records of all patients with an incident primary diagnosis of HF (ICD-10: I50) at two large university hospitals in the Capital Region of Denmark between June 2021 and May 2024. HFrEF was defined as left ventricular ejection fraction (LVEF) ≤ 40%, HFmrEF as 41– 49%, and HFpEF as ≥ 50%. Patients without available echocardiographic measurements were classified together with the HFpEF group. The phenotypic distribution was stratified by age groups (< 65, 65– 79, and ≥ 80 years) and six clinical subgroups (female sex, obesity, chronic kidney disease, atrial fibrillation/flutter, diabetes, and ischemic heart disease).Results: Following manual review of 1888 patients with a primary diagnosis of HF (median age 74, 67% male), 1687 (89%) had HFrEF, 72 (4%) HFmrEF, and 129 (7%) HFpEF. HFrEF was dominant across all ages but was lower with advancing age: 95% in patients aged < 65 years, 90% in those aged 65– 79, and 83% among those aged ≥ 80. In all clinical subgroups, the proportion of HFrEF remained high, ranging from 83 to 90%.Conclusion: Patients with an incident hospital diagnosis of HF at two Danish university hospitals predominantly presented with HFrEF. This phenotype distribution may reflect that of patients diagnosed with HF in the Danish National Patient Register. Even among those aged ≥ 80 years, fewer than one in five patients presented with HFmrEF or HFpEF. These findings provide a necessary reference for interpreting Danish registry-based HF data and inform the design of future registry-based clinical trials.Keywords: heart failure, phenotype, register data, epidemiology