Cardiovascular disease (CVD) has become an increasingly severe global public health issue, with its disease burden continuing to rise worldwide. Insulin resistance (IR) is a key metabolic process underlying CVD, but direct measurement is challenging in large populations.Various composite cardiometabolic indices are used to reflect IR-related metabolic alterations.Prediabetes, prehypertension, and predyslipidemia are early manifestations of subclinical metabolic dysfunction associated with CVD. However, their contributions to CVD risk and their relationship with dynamic changes in IR remain unclear.Therefore, this study aimed to comprehensively evaluate multiple insulin resistance–related surrogate markers and explore the associations of baseline IR, cumulative IR (cuIR), and IR trajectories with CVD risk in middle-aged and older adults. Data were obtained from the China Health and Retirement Longitudinal Study (CHARLS, 2011–2020), with relevant information collected via standardized questionnaires during follow-up.Eleven IR-related surrogate markers (TyG, TyG-WC, TyG-WWI, TyG-BMI, TyG-ABSI, TyG-BRI, TyG-WHtR, METS-IR, AIP, CTI, CHG) were analyzed, with covariates pre-specified based on established cardiovascular risk factors. The primary outcome was the incidence of newly diagnosed CVD. In the CHARLS cohort, IR-related indicators were available at two time points (Wave 1: 2012 and Wave 3: 2015). Therefore, cumulative insulin resistance exposure (cuIR) was calculated as: cuIR = (IR2012 + IR2015) / 2 × time. Due to the limited number of repeated measurements, clustering analysis (K-means) was applied to classify cumulative exposure patterns of IR-related indicators.Multivariate Cox proportional hazards regression models estimated hazard ratios (HR) with 95% confidence intervals (95% CI), and restricted cubic splines (RCS) examined nonlinear associations. Predictive performance of IR-related surrogate indicators was assessed using C-statistic, net reclassification improvement (NRI), and integrated discrimination improvement (IDI). Subgroup analyses further tested robustness of findings. Additionally, external validation of baseline IR findings was conducted using the English Longitudinal Study of Ageing (ELSA, 2004–2016). In the ELSA cohort, longitudinal trajectories of IR-related surrogate markers were constructed using group-based trajectory modeling (GBTM) based on repeated measurements (Waves 2, 4, and 6). In the study examining the association between baseline IR-related surrogate markers and the risk of CVD occurrence, 7,299 participants were enrolled, with 1,870 new CVD events occurring during an 8-year follow-up period. Across three Cox proportional hazards models, all 11 surrogate markers were significantly associated with increased CVD risk. In Model 3, fully adjusted for confounders, TyG-ABSI demonstrated the strongest association with CVD risk (HR = 4.92, 95% CI 3.13–7.74). After stratification by quantiles, higher quantiles were associated with increased CVD risk (P for trend < 0.05). In the external validation cohort ELSA, all IR indicators except TyG showed positive correlations with CVD risk in unadjusted models; After multivariable adjustment TyG-WC, TyG-BMI, TyG-BRI, TyG-WHtR, METS-IR, and and CTI remained independently associated with CVD risk. Participants in the highest quantile showed significantly increased risk ( P for trend < 0.05). In the study of cuIR and CVD risk, 773 out of 3,847 participants ultimately developed CVD. In the multivariable Cox proportional hazards model, all 11 cuIR-related surrogate markers significantly increased CVD risk. Compared to Q1, both Q2 and Q3 groups showed elevated risk, with a significant linear trend ( P for trend < 0.05). Clustering pattern analysis based on cumulative exposure showed that individuals in the high-level groups had significantly higher CVD risk than those in the low-level groups. Trajectory analysis of the ELSA cohort revealed unique longitudinal patterns of insulin resistance-related indicators. In the fully adjusted model, compared with the low-risk trajectory group, individuals in the high-risk trajectory groups for TyG-WC, TyG-BMI, TyG-BRI, METS-IR, and CTI had a significantly increased risk of developing CVD. Among these, CTI showed the strongest association with CVD risk (HR = 1.90, 95% CI 1.25–2.89).In contrast, associations for TyG, TyG-WHtR, TyG-WWI, and TyG-ABSI were not significant after multivariable adjustment. No significant associations were observed for the trajectory groups of AIP and CHG. This study demonstrates that baseline IR, cuIR, and longitudinal trajectory patterns are significantly associated with CVD risk in middle-aged and older populations. These measures provide incremental predictive value beyond traditional risk factors. Early identification and long-term monitoring of IR-related surrogate markers during the subclinical metabolic stage may enhance CVD risk stratification and support precision prevention and intervention strategies.
Zhao et al. (Tue,) studied this question.