Triglyceride glucose‑derived indices combined with high‑sensitivity C‑reactive protein for predicting glycemic progression and regression in prediabetes: findings from a prospective national cohort study based on baseline and cumulative change analyses

Recent studies have identified triglyceride-glucose (TyG)‑derived indices as important novel predictors of progression from prediabetes to diabetes and regression to normoglycemia, with chronic inflammatory burden serving as a key mediator in this association. Building on these observations, this study aimed to evaluate the clinical utility of a series of C‑reactive protein-triglyceride-glucose (CTI) indices, which integrate TyG-derived indices with the inflammatory marker high‑sensitivity C‑reactive protein, in predicting glycemic progression and regression among individuals with prediabetes. This analysis used data from the nationally representative China Health and Retirement Longitudinal Study (CHARLS) cohort and included 2,003 participants with prediabetes. Using baseline survey data (2011–2012) and follow-up records from 2015, we systematically quantified the baseline levels, cumulative exposure, and longitudinal trajectories of the CTI series indices. To evaluate their associations with glycemic transitions, incremental predictive performance, and relative component contributions, we applied multivariable Cox regression, restricted cubic spline models, trajectory analysis, receiver operating characteristic curves, incremental prediction analysis, and quantile g-computation. Finally, external validation of the CTI series indices was performed using data from the English Longitudinal Study of Ageing (ELSA, 2004–2016). Over a median observation period of 3 years, 306 participants (15.28%) progressed to diabetes and 452 (22.57%) regressed to normoglycemia. Higher baseline levels and cumulative exposure of the CTI series indices were associated with an increased risk of diabetes and a lower likelihood of regression to normoglycemia, with cumulative exposure showing stronger associations than static baseline values. Trajectory analyses further showed that participants with persistently high trajectories of these indices had a greater risk of diabetes and a lower probability of regression than those with consistently low trajectories. Compared with the conventional TyG index, the CTI series indices demonstrated higher areas under the curve and improved risk reclassification for both progression to diabetes and regression to normoglycemia. Among these indices, CTI-WHtR exhibited the best overall predictive performance. Quantile g-computation further suggested distinct relative contributions to the two glycemic outcomes: glucose levels, adiposity, and inflammatory burden may be more important for preventing progression to diabetes, whereas glycemic control and adiposity reduction may be more critical for facilitating regression to normoglycemia. Finally, findings from the ELSA cohort validated those from the CHARLS cohort regarding progression from prediabetes, showing that the CTI series indices were closely associated with glycemic progression, with CTI-WHtR exhibiting the strongest association. The CTI series indices, which combine TyG-derived indices with the inflammatory marker high-sensitivity C-reactive protein, represent a novel and promising strategy for predicting glycemic progression and regression among individuals with prediabetes. Moreover, evaluating cumulative exposure and dynamic trajectories using repeated‑measures data may enable better prediction of glycemic outcomes. Among these indices, CTI-WHtR demonstrated the strongest overall association and predictive performance, suggesting its potential as a preferred indicator for risk stratification and individualized management of prediabetes.