Estimated glucose disposal rate outperforms other insulin resistance indices for cardiometabolic multimorbidity prediction in CKM syndrome stages 0–3

Abstract

Cardiovascular-kidney-metabolic (CKM) syndrome is a potential public health challenge characterized by overlapping cardiovascular, kidney, and metabolic disorders. Although insulin resistance-related surrogate indices have been linked to cardiometabolic outcomes, their prognostic value for incidence of cardiometabolic multimorbidity (CMM) among individuals with CKM syndrome remains unclear. This prospective cohort study used data from the China Health and Retirement Longitudinal Study between 2011 and 2020. A total of 7,237 participants with CKM stages 0–3 were included. Associations of five insulin resistance-related indices with incident cardiometabolic multimorbidity were assessed using Cox proportional hazards models, restricted cubic spline analyses, and Kaplan–Meier methods. The predictive performance of the five indices was compared using time-dependent receiver operating characteristic curves, net reclassification improvement, and integrated discrimination improvement. Additionally, joint and mediation analyses were performed to evaluate the mediating role of C-reactive protein. Over a median follow-up of 9 years, 703 participants (9.7%) developed CMM. In fully adjusted models, eGDR was inversely associated with incident CMM, whereas METS-IR, TyG, LAP, and TG/HDL-C were positively associated, with corresponding HRs for the highest versus lowest quartiles of 0.13, 2.86, 2.81, 2.99, and 1.99, respectively (all P  < 0.001). RCS analyses demonstrated linear associations for eGDR and TyG and nonlinear associations for METS-IR, LAP, and TG/HDL-C. Among the five indices, eGDR showed the strongest incremental predictive value for CMM beyond the baseline model, with significant improvements in NRI and IDI (all P  < 0.01). Joint analyses showed that low eGDR combined with elevated CRP levels was associated with the highest risk of CMM, whereas mediation analyses suggested a statistically significant but quantitatively modest mediating role of CRP (2.6%). Among individuals with CKM stages 0–3, all five insulin resistance indices were independently associated with incident CMM, with eGDR demonstrating superior incremental predictive value in this cohort. The coexistence of low eGDR and elevated CRP identified individuals at particularly high risk. Our findings suggest that eGDR may serve as a promising surrogate marker for risk stratification in this population, pending further validation.