HEART failure risk was accurately predicted years before clinical onset using routine cardiac CT scans, according to a large multicentre study analysing the biological properties of epicardial fat.
Heart failure remains a leading cause of morbidity and mortality worldwide, often diagnosed late when symptoms have already developed. Identifying individuals at risk earlier is therefore a major priority.
Epicardial adipose tissue, a metabolically active fat depot surrounding the heart, has increasingly been recognised as a key player in cardiovascular disease due to its interaction with the myocardium.
Heart Failure Prediction Enhanced by Radiomic Profiling
In this study of 72,751 adults without prior heart failure or myocardial infarction, researchers used automated imaging analysis to extract over 1,600 radiomic features from epicardial fat captured during coronary CT angiography.
These features were integrated into a novel fat radiomic profile for heart failure (FRPHF), designed to detect subtle biological changes not visible through conventional imaging.
Over a median follow-up of approximately 4–5 years, heart failure developed in 2.9% of the internal cohort and 2.7% of the external validation cohort. FRPHF demonstrated strong predictive performance (C-statistics: 0.869 internal; 0.850 external).
Notably, each 25-percentile increase in the radiomic score was associated with nearly a four-fold increase in heart failure risk (hazard ratio: 3.90; 95% CI: 3.13–4.84), with individuals in the highest decile facing almost 20 times greater risk compared with those in the lowest.
Importantly, adding this radiomic profile to standard risk models, including coronary artery disease severity and fat volume, significantly improved prediction accuracy and patient risk classification.
Implications for Early Heart Failure Prevention
The findings suggested that routine cardiac CT scans could be repurposed to provide clinically meaningful risk stratification without additional testing.
By capturing early biological remodelling within epicardial fat, this approach may enable earlier identification of individuals at high risk of heart failure, supporting more targeted prevention strategies.
Further research is needed to determine how best to integrate this technology into clinical workflows and whether intervention based on these findings improves outcomes.
Overall, radiomic phenotyping of cardiac fat represents a promising step towards precision prevention in cardiovascular medicine, offering a scalable and non-invasive tool for early heart failure prediction.
Reference
Oikonomou E et al.; the ORFAN Consortium. Early prediction of heart failure from routine cardiac CT using radiomic phenotyping of epicardial fat. JACC. 2026;DOI:10.1016/j.jacc.2026.02.5116.
Featured image: doraclub on Adobe Stock
