Circ Res. 2026 Jul 24. doi: 10.1161/CIRCRESAHA.126.328542. Online ahead of print.
ABSTRACT
BACKGROUND: Longitudinal metabolomic studies can refine understanding of heart failure (HF) progression and enable precision prevention. This study aims to identify serum metabolites associated with HF risk via longitudinal metabolomic analysis, delineate their dynamic trajectories, and explore metabolite profiles in populations with different metabolic disorders.
METHODS: This study analyzed longitudinal serum metabolomic data from 4774 serum samples from 1728 HF-free participants in the Chinese Multi-Provincial Cohort Study Metabolomics Project at 4 time points over a 20-year follow-up. Intensity models and Cox proportional-hazards models identified metabolites associated with HF risk. Latent variable mixed-effects models evaluated metabolite trajectories.
RESULTS: Of the 784 detected metabolites, 23 were associated with HF risk at a false discovery rate-adjusted P<0.05, including 9 not previously reported in relation to HF. The HF risk-associated metabolites exhibited 4 distinct trajectory clusters and corresponding biological trends. Most metabolites that showed positive associations with HF risk remained relatively stable throughout the 20-year follow-up period, whereas metabolites that were negatively associated with HF risk generally exhibited a declining trend. The levels of these 23 metabolites in the group who developed HF began to diverge from the levels in the non-HF group >5 years before clinical HF diagnosis, with most changes initiating 15 to 20 years before clinical manifestation. Populations with different metabolic disorders exhibited distinct metabolite profiles related to HF. The HF-associated metabolites were primarily involved in energy metabolism and the vasodilatory response among individuals with hypertension, lipotoxic effects and oxidative stress among those with obesity, and inflammatory processes and glucotoxic mechanisms among individuals with dysglycemia.
CONCLUSIONS: This longitudinal metabolomic study identifies HF-associated metabolite profiles, characterizes their changes during the 20 years preceding clinical diagnosis, and reveals heterogeneity across individuals with different metabolic disorders, thereby informing future biomarker and intervention research.
PMID:42495742 | DOI:10.1161/CIRCRESAHA.126.328542