Abstract
BACKGROUND: Combining biomarkers and anthropometric indicators is a common way to improve predictive efficacy. Yet the effect of the C-reactive protein-triglycerides-glucose index (CTI) and its derivatives on stroke is unknown. This study aims to explore their association with stroke and compare their predictive value. METHODS: A total of 10,070 participants from the China Health and Retirement Longitudinal Study (CHARLS) were included. Covariate selection was performed using the Boruta algorithm and complementary methods. The predictive performance of various indicators was compared via metrics including area under the curve (AUC), net reclassification improvement (NRI), and integrated discrimination improvement (IDI) to determine the optimal predictive indicator. Subsequently, Cox proportional hazards models, trajectory pattern analysis, restricted cubic spline (RCS) analysis, Kaplan-Meier curves were used to investigate its association with stroke. RESULTS: From 2011 to 2020, 950 participants (9.43%) experienced an incident stroke during follow-up. CTI derivatives outperformed standalone CTI in prediction, with CTI-waist-to-height ratio (CTI-WHtR) exhibiting the strongest association with stroke risk. In the fully adjusted model, each one-unit increase in CTI-WHtR linked to a 73% higher risk of stroke [odds ratio (OR) = 1.73, 95% confidence interval (CI) 1.45-2.07], and participants in the highest quartile of CTI-WHtR had more than double the risk compared with those in the lowest quartile (OR = 2.20, 95% CI 1.71-2.83). There were three distinct trajectories of CTI-WHtR over time. Compared with Cluster 1 (low value rising), Cluster 2 (stable high value) was associated with a 69% higher risk of stroke (OR = 1.69, 95% CI 1.34-2.13), and Cluster 3 (moderate value rising) with a 26% higher risk (OR = 1.26, 95% CI 1.06-1.52). A clear dose-response relationship was observed between CTI-WHtR and stroke risk, with risk increasing sharply when CTI-WHtR exceeded 2.52. CONCLUSION: The consistent positive linear association between CTI-WHtR and stroke risk highlights its potential utility as a clinical and public health indicator. Monitoring and maintaining optimal CTI-WHtR levels may aid in identification of individuals at elevated stroke risk. Graphical Abstract.