Abstract
BACKGROUND: Insulin resistance and metabolic disorders are important mechanisms in heart failure (HF) development. Patients with coronary artery disease (CAD) often present metabolic abnormalities, which may further increase HF risk. Therefore, examining the relationship between obesity-metabolic indices and HF in CAD patients is clinically relevant. METHODS: This multicenter study included 3872 CAD patients. Multivariable logistic regression assessed associations of four obesity-metabolic indices [triglyceride glucose (TyG), TyG-body mass index (TyG-BMI), TyG-waist circumference (TyG-WC), metabolic score for insulin resistance (METS-IR)] with HF. Restricted cubic splines (RCS) evaluated nonlinearity and threshold effects. Diagnostic performance was compared using receiver operating characteristic (ROC), decision curve analysis (DCA), and variable importance measures, with robustness tested via subgroup and sensitivity analyses. Mediation analysis examined renal function as a potential mediator. RESULTS: All four indices were significantly associated with HF, showing a nonlinear dose-response relationship. Associations strengthened notably when TyG, TyG-BMI, TyG-WC, and METS-IR exceeded 8.27, 240, 520, and 38.5, respectively. Subgroup and sensitivity analyses confirmed robustness. ROC, DCA, and variable importance ranking analyses collectively demonstrated that METS-IR possessed the best diagnostic ability compared to the other three obesity-metabolic indices. Mediation analysis indicated that obesity-metabolic indices may influence HF risk partly through renal function. CONCLUSION: Obesity-metabolic indices, particularly METS-IR, are associated with HF in CAD patients and exhibit clear threshold effects. METS-IR demonstrated the strongest diagnostic ability for prevalent HF, and its association with HF was partially mediated by renal function.