Association of body composition indices with cardiovascular outcomes: a nationwide cohort study

Previous studies regarding BMI (kg/m2) and associated cardiovascular outcomes yield inconsistent results. We aimed to investigate the association between body composition and cardiovascular outcomes according to BMI categories in the Korean general population. A total of 2,604,401 participants were enrolled in this nationwide cohort study using the National Health Insurance Service-Health Checkup data set. Predicted lean BMI (pLBMI), body fat mass index (pBFMI), and appendicular skeletal muscle mass index (pASMMI) were calculated using validated anthropometric prediction equations. A multivariable time-dependent Cox regression analysis was conducted to assess the association with cardiovascular outcomes. The results were presented with adjusted hazard ratios (HRs) with 95% confidence intervals (CIs), considering BMI categories (BMI < 18.5, BMI 18.5–24.9, BMI 25–29.9, and BMI ≥ 30). Higher pLBMI and pASMMI were correlated with a reduced risk of composite cardiovascular outcomes. For pLBMI, HR was 0.910 (95% CI: 0.908, 0.913, P < 0.001) for males and 0.905 (95% CI: 0.899, 0.910, P < 0.001) for females. For pASMMI, HR was 0.825 (95% CI: 0.820, 0.829, P < 0.001) for males and 0.788 (95% CI: 0.777, 0.800, P < 0.001) for females. Conversely, a higher pBFMI was associated with an increased risk, with HR of 1.082 (95% CI: 1.071, 1.093, P < 0.001) for males and 1.181 (95% CI: 1.170, 1.192, P < 0.001) for females. Subgroup analysis based on BMI categories revealed no significant risk association for pBFMI in the BMI < 18.5 group. In the group with BMI ≥ 30, neither pLBMI nor pASMMI demonstrated a significant risk association. Our results highlight the value of pLBMI, pBFMI, and pASMMI as variables for assessing risk of composite cardiovascular outcomes. The significance of indicators may vary depending on BMI categories.