Association between BMI-discordances in cardiometabolic biomarkers and risk of incident heart failure: a prospective cohort study

Background: Obesity is a major risk factor for heart failure (HF). However, individuals with similar BMI, the main diagnostic measure of obesity, exhibit considerable heterogeneity in developing HF. Objectives: This study aimed to investigate the association between metabolically discordant subgroups characterized by cardiometabolic biomarkers deviated from those predicted by BMI and risk of incident HF. Methods: Data from 394,198 participants in the UK Biobank were analyzed. A data-driven cluster approach was used to classify BMI subgroups according to cardiometabolic biomarker profiles. Incident HF during the follow-up was ascertained using the International Classification of Diseases, Tenth Revision codes. Cox proportional hazards models were applied to assess the association of the subclassified BMI with HF risk. Results: Over a median follow-up of 12.3 y, 5176 (2.94%) males and 3602 (1.65%) females developed the first incidence of HF. Compared with participants in BMI-concordant subgroups, individuals with C-reactive protein or blood glucose deviating from the expected risk based on their BMI showed a higher risk of HF. The fully adjusted hazard ratios (95% confidence intervals) for HF in relation to discordant inflammatory state and discordant hyperglycemic status were 1.59 (1.40, 1.80) and 1.39 (1.22, 1.60) in males and 2.03 (1.77, 2.34) and 1.80 (1.53, 2.12) in females, respectively. In contrast, individuals with discordantly high blood pressure (only in females) or adverse blood lipid levels demonstrated a lower risk of HF than those in BMI-concordant subgroups. No significant association with HF was observed for discordant liver transaminase subgroups. Conclusions: Metabolically distinct BMI subgroups exhibit varying risks of HF, suggesting that subclassifying BMI based on cardiometabolic biomarkers may facilitate the precision prevention of HF.