Abstract
BACKGROUND: The triglyceride-glucose (TyG) index, estimated glucose disposal rate (eGDR), and metabolic score for insulin resistance (METS-IR) are well-established surrogate indices of insulin resistance (IR). Although both IR and elevated basal metabolic rate (BMR) are recognized risk factors for cardiometabolic diseases, their joint effects on the risk of cardiometabolic multimorbidity (CMM) remain unclear. This study aimed to investigate the separate and combined associations of these IR surrogates and BMR with incident CMM. METHODS: We included 7204 eligible participants from the 2011-2020 survey waves of the China Health and Retirement Longitudinal Study (CHARLS). Participants were stratified by median values of IR surrogate indices and BMR. The associations with CMM were assessed using Kaplan-Meier curves, multivariable Cox regression, and restricted cubic splines (RCS). Predictive performance was evaluated using receiver operating characteristic (ROC) curves, net reclassification improvement (NRI), and integrated discrimination improvement (IDI). Mediation and interaction analyses were performed to explore potential underlying relationships. RESULTS: Over a median follow-up of 9 years, 1103 participants (15.31%) developed CMM. Compared to those with low IR and low BMR, participants with high levels of both exhibited the highest risk of CMM, with hazard ratios of 2.13 (95% CI 1.73-2.63) for TyG, 1.93 (95% CI 1.57-2.36) for eGDR, and 1.92 (95% CI 1.61-2.29) for METS-IR. These associations remained consistent in subgroup and sensitivity analyses. Adding IR indices and BMR to the baseline model significantly improved CMM prediction: TyG (AUC 0.759, NRI 0.371, IDI 0.017; all P < 0.001), eGDR (AUC 0.753, NRI 0.330, IDI 0.012; all P < 0.01), and METS-IR (AUC 0.747, NRI 0.170, IDI 0.004; all P < 0.01). Mediation analysis demonstrated that all IR indices significantly mediated the association between BMR and CMM, and a bidirectional mediation relationship was specifically observed between BMR and the TyG index. Notably, no significant additive or multiplicative interactions were detected. CONCLUSION: IR surrogate indices and BMR independently and jointly predicted the risk of CMM, with IR pathways substantially mediating the effect of BMR. The combined assessment of these parameters may improve CMM risk stratification and guide primary prevention strategies.