Abstract
BACKGROUND: The link between uric acid (UA) and cardiovascular diseases is debated, with insulin resistance possibly affecting this relationship. The triglyceride-glucose (TyG) index is a recognized marker for insulin resistance. However, the combined effect of different levels of UA and TyG on the severity of coronary artery disease (CAD) remains unclear. METHODS: A cohort of 1,835 patients with newly diagnosed CAD was divided into single-vessel (743 patients) and multi-vessel (1,092 patients) CAD groups. The study utilized logistic regression, restricted cubic spline (RCS), and extreme gradient boosting (XGBoost) models to explore the associations between UA, TyG, and multi-vessel CAD. Interaction analysis assessed potential additive and multiplicative interactions. A mediation analysis was performed to assess the indirect effects of TyG and UA on the severity of CAD. RESULTS: The TyG > 9.33 and non-HUA group is linked to a higher risk of multi-vessel CAD (OR 1.41, 95% CI 1.08-1.85), while the TyG ≤ 9.33 and HUA group shows no significant association (OR 1.08, 95% CI 0.76-1.53). The feature importance analysis, using the XGBoost model, demonstrated that TyG has a higher predictive value for multi-vessel CAD. No nonlinear correlations were observed for RCS. No notable additive or multiplicative interactions were detected between TyG and UA. Mediation analysis revealed that TyG significantly mediated the relationship between UA and multi-vessel CAD, with a proportion mediated of 18.89% (p = 0.026). In contrast, UA did not significantly mediate the TyG-CAD relationship (p = 0.082). CONCLUSION: The TyG index correlated more strongly with multi-vessel CAD compared to UA. Hyperuricemia correlated with multi-vessel CAD exclusively at elevated TyG levels, with TyG mediating the link between uric acid and CAD severity.