Optimizing early triage for subclinical cardiovascular-kidney-metabolic syndrome: longitudinal associations, incremental value, and clinical utility of nine insulin resistance surrogates

Cardiovascular-kidney-metabolic (CKM) syndrome stage 3 represents a critical, yet highly insidious, window of subclinical target organ damage. We aimed to comprehensively compare nine baseline non-insulin-based surrogate indices of insulin resistance to predict progression to CKM stage 3 and determine their scenario-specific clinical utility. This longitudinal study included 2958 adults initially at CKM stages 0–2 from the China Health and Retirement Longitudinal Study. Independent associations were assessed using multivariable logistic regression and restricted cubic splines. Discriminative ability was evaluated using the area under the receiver operating characteristic curve (AUC). Incremental predictive values over a basic clinical model were quantified via continuous net reclassification improvement (NRI) and integrated discrimination improvement (IDI). The net clinical benefit was further evaluated through decision curve analysis (DCA). During follow-up, 497 (16.8%) participants progressed to CKM stage 3. After adjusting for comprehensive covariates, all nine indices were independently associated with the outcome. Notably, composite indices coupling systemic glucolipotoxicity with central adiposity exhibited preferable predictive superiority. Triglyceride-glucose index combined with waist-to-height ratio (TyG-WHtR) and triglyceride-glucose index combined with waist circumference (TyG-WC) yielded the highest odds ratios per 1-SD increment (1.86 and 1.82, respectively). Incorporating them into the basic model significantly increased the AUC and provided the most substantial improvements in patient reclassification (NRI: 0.524 and 0.512, respectively; both P < 0.001) and discrimination (IDI). Furthermore, DCA confirmed that models augmented with TyG-WC or TyG-WHtR provided the highest net clinical benefit across varying threshold probabilities. This prognostic dominance remained robust across four stringent sensitivity analyses. TyG-WC and TyG-WHtR are the most robust, independent predictors of incident CKM stage 3. They offer a highly actionable, scenario-adaptable screening strategy: TyG-WC serves as an ultra-accessible “first-pass” filter for large-scale public health triage, while TyG-WHtR provides a calibrated prognostic anchor for high-threshold, individualized clinical decision-making.