Associations of triglyceride-glucose-obesity composite indices with all-cause and cardiovascular mortality in individuals with cardiovascular-kidney-metabolic syndrome stages 0–3 and their potential incremental value beyond a PREVENT-variable-based model

The PREVENT equations represent a paradigm shift in cardiovascular risk assessment by incorporating cardiovascular-kidney-metabolic (CKM) health metrics. However, in individuals with CKM stages 0–3, the associations of TyG and TyG-obesity composite indices with all-cause mortality (ACM) and cardiovascular mortality (CVM), and whether they provide incremental predictive value beyond a PREVENT-variable-based model, remain unclear. We analyzed data from 12,947 adults with CKM stages 0–3 in the National Health and Nutrition Examination Survey (NHANES), with a median follow-up of 10.25 years. Seven indices were calculated: TyG alone and six TyG-obesity composite indices. Weighted Cox proportional hazards models assessed associations with ACM and CVM. Kaplan-Meier curves were used for ACM, and Aalen-Johansen cumulative incidence functions with cause-specific Cox and Fine-Gray models were used for CVM. Restricted cubic splines (RCS) examined nonlinearity, and Cox-based quantile g-computation (qgcomp) assessed the joint effect of all seven indices. Incremental predictive value beyond the PREVENT-variable-based model was evaluated using changes in C-statistic (ΔC), time-dependent AUC at 10 years, survival-based continuous net reclassification improvement (NRI) and integrated discrimination improvement (IDI), bootstrap resampling, calibration, and decision curve analysis. Subgroup, stage-specific and sensitivity analyses were also performed. During follow-up, 1,284 ACM and 342 CVM occurred. Kaplan-Meier and Aalen-Johansen curves showed that cumulative incidence increased across TyG quartiles for both ACM and CVM. After full adjustment, all seven indices were independently associated with ACM; TyG-BMI showed the strongest association (HR = 1.99, 95% CI: 1.21–3.28), followed by TyG-BRI (HR = 1.89, 95% CI: 1.38–2.59). For CVM, none of the indices remained significant in Model 3, and competing-risk analyses using cause-specific Cox and Fine-Gray models also showed no significant associations. In quartile analyses, TyG-WWI was associated with increased CVM risk in Q3 (HR = 4.53, 95% CI: 1.41–14.58), but not in Q4. Stage-specific analyses showed that associations with ACM were mainly significant in CKM stage 2; for CVM, significant interactions with CKM stage were observed for TyG-BMI, TyG-WC, and TyG-WHtR (P < 0.05). RCS analysis suggested nonlinear associations of TyG-BMI and TyG-BRI with ACM. Qgcomp indicated a modest joint effect on CVM. Time-dependent AUC increases were all < 0.01. Bootstrap analyses showed small but significant improvements for TyG-ABSI and TyG-WWI in ACM (ΔC = 0.0024, 95% CI: 0.0006–0.0055, and 0.0017, 95% CI: 0.0002–0.0046, respectively), but not in CVM. Survival-based NRI/IDI showed modest improvement only for TyG-ABSI (NRI = 0.092; IDI = 0.003) and TyG-WWI (NRI = 0.094; IDI = 0.003) in ACM. Sex was a significant effect modifier for several indices (P < 0.001). Results from multiple imputation sensitivity analyses were consistent with the complete-case analysis. Findings were broadly consistent after excluding deaths within the first 2 years. In individuals with CKM stages 0–3, TyG and TyG-obesity composite indices were independently associated with ACM, with TyG-BMI and TyG-BRI showing the strongest associations. However, no significant associations were observed with CVM. Incremental predictive value beyond a PREVENT-variable-based model was limited, with modest improvement in risk reclassification for ACM only. Their clinical utility warrants further validation. What is currently known about this topic? TyG and TyG-obesity composite indices have been associated with the development and prognosis of cardiovascular disease. However, their associations with ACM and CVM, and whether they improve mortality risk prediction beyond a PREVENT-variable-based model, remain insufficiently supported by epidemiological evidence. What is the key research question? Beyond the traditional cardiometabolic risk factors already incorporated in a PREVENT-variable-based model, can TyG and TyG-obesity composite indices independently predict ACM and CVM and provide measurable improvements in risk prediction? What is new? This study systematically compared seven TyG and TyG-obesity composite indices in a population with CKM stages 0-3. All seven indices were independently associated with ACM, with TyG-BMI and TyG-BRI showing the strongest associations; however, none were significantly associated with CVM after full adjustment. Incremental predictive value beyond a PREVENT-variable-based model was limited (all ΔC and AUC <0.01); only TyG-ABSI and TyG-WWI showed modest improvement in risk reclassification for ACM. Associations with ACM were mainly observed in CKM stage 2, and significant interactions with CKM stage were found for several indices in CVM. How might this study influence clinical practice? As low-cost indicators that rely only on routine laboratory tests and simple anthropometric measurements, TyG and TyG-obesity composite indices may help identify individuals at underestimated risk for mortality when used alongside a PREVENT-variable-based model. However, their net clinical benefit still requires further validation in prospective studies.