C-reactive protein-triglyceride-glucose index versus triglyceride-glucose index in predicting cardiovascular metabolic multimorbidity risk: A cohort study

Abstract

Background Cardiovascular Metabolic Multimorbidity (CMM), a leading global cause of mortality, lacks evidence on the predictive utility of the novel C-reactive protein-triglyceride-glucose index (CTI), which integrates insulin resistance and inflammation. This study compared CTI with the established Triglyceride-Glucose (TyG) index in predicting CMM risk. Methods A cohort of 8,487 adults aged ≥45 from the CHARLS database (2011–2020) was analyzed. Over nine years, CMM events were tracked. Cox regression, restricted cubic spline (RCS), and ROC curve analyses assessed associations between TyG, CTI, and CMM risk. Subgroup analyses evaluated population-specific variations. Results Among participants, 1,030 (12.14%) developed CMM. Unadjusted Cox models showed TyG (HR = 1.89, 95%CI 1.73–2.07, P  < 0.001) and CTI (HR = 1.83, 95%CI 1.70–1.97, P  < 0.001) predicted CMM; adjusted models confirmed persistence. A dose–response association was observed for both CTI and TyG with CMM risk. In fully adjusted models, the overall dose–response trend remained similar. ROC analysis favored CTI (higher AUC). Subgroup analyses indicated TyG’s association varied by sex, smoking, and hypertension ( P  < 0.05), while CTI’s association differed by age, sex, and hypertension ( P  < 0.05). Conclusions Elevated CTI independently correlates with increased CMM risk, demonstrating superior predictive accuracy over TyG. Its linear association highlights potential clinical utility for early CMM risk stratification.