A multi-stage validation of the C-reactive protein-triglyceride-glucose index for predicting mild cognitive impairment: evidence from clinical and nationwide prospective cohorts.

BACKGROUND

Mild cognitive impairment (MCI) is a common but under-recognized complication of type 2 diabetes mellitus (T2DM), and practical predictors are limited. The C-reactive protein-triglyceride-glucose index (CTI) integrates metabolic and inflammatory status and may better capture the pathways underlying cognitive decline.

METHODS

The study included 1, 580 patients with T2DM as the primary cohort and utilized the China Health and Retirement Longitudinal Study (CHARLS, n = 4, 575) for external validation. Variables were selected using LASSO and stepwise regression. The area under the curve (AUC), net reclassification improvement (NRI), and integrated discriminant improvement (IDI) were used to evaluate the model's performance, while restricted cubic splines (RCS) were used to investigate the dose-response connection.

RESULTS

In both the main cohort (OR = 1.16, 95% CI: 1.04-1.30) and the CHARLS cohort (HR: 1.245, 95% CI: 1.164-1.332), a high CTI score was independently associated with an increased risk of cognitive impairment. In the main cohort, the CTI significantly outperformed the TyG index in terms of risk reclassification (NRI = 0.132, P = 0.031). RSC analysis revealed a significant linear relationship in individuals with normal glucose levels and prediabetes. Furthermore, the ROC curve developed in this study demonstrated good calibration and clinical net benefit.

CONCLUSIONS

The CTI index is a robust composite marker for predicting the risk of MCI. It has strong predictive value in the early stages of impaired glucose metabolism, making it a useful tool for identifying high-risk individuals and delivering customized therapies.