Comparison of triglyceride-glucose index and HOMA-IR for predicting prevalence and incidence of metabolic syndrome

Background and aims: Insulin resistance is related closely to metabolic syndrome (MetS). The homeostasis model assessment of insulin resistance (HOMA-IR) is the most commonly used insulin resistance index, but the triglyceride-glucose (TyG) index has been suggested as a reliable alternative insulin resistance index. This study aims to compare the predictive powers of TyG index and HOMA-IR for the prevalence and incidence of MetS in a large, community-based, prospective cohort over 12 years of follow-up. Methods and results: Data from 9730 adults with or without MetS at baseline, 6091 adults without MetS who were followed as part of the Korean Genome and Epidemiology Study were analyzed. Receiver-operating-characteristic (ROC) curves and time-dependent ROC curves were performed to compare the areas under the ROC curve (AUROC) of the TyG index and HOMA-IR for predicting the prevalence and incidence of MetS. The optimal cut-off points were calculated. Cox proportional hazard spline curves were used to verify dose-response relationship between TyG index/HOMA-IR and incident MetS. TyG index showed higher predictive power for prevalent MetS than HOMA-IR (0.837 vs. 0.680, p < 0.001). The AUROC for incident MetS of TyG index and HOMA-IR was 0.654 (0.644-0.664) and 0.556 (0.531-0.581), respectively (p < 0.001). Cut-off points of TyG index and HOMA-IR for predicting the prevalence of MetS were 8.718 and 1.8 and for predicting incident MetS were 8.518 and 1.5, respectively. Both TyG index and HOMA-IR had a linear relationship with incident MetS. Conclusions: TyG index is superior to HOMA-IR for predicting MetS.restrictio