Cumulative exposure and dynamic trajectories of the C-reactive protein-triglyceride-glucose index (CTI) versus the cholesterol, high-density lipoprotein, and glucose index (CHG) for incident hypertension prediction: a national cohort study

The C-reactive protein-triglyceride-glucose index (CTI) and cholesterol-high-density lipoprotein-glucose index (CHG) are emerging composite biomarker indices, but their cumulative exposure and comparative value for predicting incident hypertension remain unclear. Leveraging a nationwide longitudinal cohort from the China Health and Retirement Longitudinal Study (CHARLS), the cumulative burden of CTI and CHG was modeled. These cumulative indices were defined as the average of the values measured at Wave 1 and Wave 3, multiplied by the time interval between these two assessments. Then, multivariable Cox proportional hazards models were used to assess associations and restricted cubic splines (RCS) for dose–response relationships. Group-based trajectory modeling (GBTM) was used to identify trajectory patterns. To gauge the predictive performance at 7 and 9 years, we analyzed time-dependent receiver operating characteristic (ROC) curves, alongside the C-index, net reclassification improvement (NRI), and integrated discrimination improvement (IDI), and cuCTI was further compared with triglyceride-glucose (TyG). Finally, the findings were further subjected to subgroup and sensitivity analyses to test their robustness. Over a median follow-up of 9 years, 408 of 2673 participants (15.3%) developed hypertension. Both elevated cuCTI and cuCHG significantly increased hypertension risk. In the fully adjusted model, participants in the highest quartile had a higher risk of hypertension for both cuCTI (HR = 2.16; 95% CI 1.63–2.87) and cuCHG (HR = 1.63; 95% CI 1.24–2.15). Crucially, cuCTI demonstrated superior predictive accuracy in time-dependent ROC analysis (9 years DeLong P = 0.025). In the fully adjusted model, adding cuCTI improved the 7 years C-index from 0.572 to 0.609 (P = 0.014) and the 9 years C-index from 0.582 to 0.618 (P = 0.003), compared with 0.590 and 0.597 for cuCHG (P = 0.110 and 0.094, respectively). Furthermore, cuCTI showed stronger gains in NRI and IDI compared to cuCHG (NRI: 30.31% vs. 12.61% at 7 years and 35.47% versus 15.77% at 9 years; IDI: 0.86% versus 0.47% at 7 years and 1.32% vs. 0.71% at 9 years; all P < 0.05). Further comparison with TyG also suggested superior predictive performance of cuCTI. Subgroup and sensitivity analyses also showed consistent results. Both indices were independently associated with incident hypertension, and cuCTI showed better long-term predictive performance than cuCHG. These findings support the value of monitoring cumulative inflammatory-metabolic burden for early hypertension risk stratification.