Body roundness index and other anthropometric indices in relation to incident myocardial infarction: a cohort study from the UK Biobank

Background Traditional anthropometric measures like body mass index (BMI) and waist circumference (WC) inadequately reflect body fat distribution. Body roundness index (BRI), a derived anthropometric index mathematically equivalent to waist-to-height ratio (WHtR) after removing non-functional constant terms, has been proposed as a simplified alternative for adiposity assessment. However, previous studies on the relationship between BRI and cardiovascular disease (CVD) have yielded inconsistent results, and few have directly compared BRI’s predictive value for myocardial infarction (MI), with other anthropometric measures. This study aims to conduct a prospective analysis of the associations between BRI, as well as other traditional and novel anthropometric indices, and MI. It also intends to compare their predictive values for MI. Methods A total of 395,831 participants were recruited from the UK Biobank cohort. The anthropometric indices—BRI, BMI, WC, Weight-Adjusted Waist Index (WWI), Body Surface Area (BSA), and A Body Shape Index (ABSI)—were categorized into quartiles. Cox proportional hazards regression and restricted cubic spline (RCS) models were employed to assess the associations between these indices and MI risk. Predictive performance was evaluated and compared using the area under the receiver operating characteristic (ROC) curve (AUC), Net Reclassification Improvement (NRI) and Integrated Discrimination Improvement (IDI). Results After a median follow-up of 13.3 years, 14,419 MI events were recorded. Compared to participants in the lowest quartile of each anthropometric index, multivariate-adjusted hazard ratios (HRs) and 95% CIs for the highest quartiles were as follows: BRI (1.52 [1.43–1.61]), WWI (1.46 [1.37–1.55]), ABSI (1.32 [1.24–1.41]), BSA (1.01 [0.95–1.07]), BMI (1.32 [1.25–1.39]), and WC (1.40 [1.31–1.50]). BRI demonstrated relatively higher predictive performance for MI (AUC: 0.710 [95% CI: 0.706–0.714]), outperforming all traditional and novel indices (all p < 0.001 by DeLong’s test). Compared with the other five anthropometric indicators, the NRI and IDI values of BRI in predicting MI risk were greater than 0, further underscoring its enhanced predictive capability. Conclusion BRI is positively associated with higher incident MI risk, showing statistically significant but modest predictive improvements over other anthropometric indices. Its clinical utility requires further independent validation before routine adoption.