BMJ Public Health. 2026 Oct 12;4(4):e004508. doi: 10.1136/bmjph-2025-004508. eCollection 2026.
ABSTRACT
INTRODUCTION: The C-reactive protein-triglyceride-glucose index (CTI) has emerged as an innovative biomarker for insulin resistance and inflammatory processes. Nevertheless, the association between CTI and diabetic vascular complications (DVCs) and its subtypes in individuals with diabetes remains unclear.
METHODS: This cohort study included 23 068 participants with diabetes from the UK Biobank at baseline (2006-2010). Cox proportional hazards models and restricted cubic splines (RCS) were employed to evaluate the associations of CTI and triglyceride-glucose-related indices with the risk of DVCs and its subtypes. Separate analyses were conducted for individuals with type 2 diabetes (T2D). The predictive performance of these indices was assessed using the net reclassification index (NRI) and Harrell's concordance index.
RESULTS: During a mean follow-up period of 13.3 years, 4089 (26.8%) individuals developed DVCs. A significant positive linear relationship was observed between CTI and the risk of incident DVCs. For 1 SD increase in CTI, the corresponding HRs (95% CIs) were 1.05 (1.02 to 1.11) for overall DVCs, 1.21 (1.16 to 1.27) for diabetic kidney disease, 1.19 (1.13 to 1.26) for diabetic retinopathy (DR), 1.34 (1.26 to 1.43) for diabetic neuropathy (DN), 1.22 (1.14 to 1.31) for peripheral artery disease (PAD) and 1.05 (1.00 to 1.10) for cardiovascular disease. RCS revealed approximately an inverse L-shaped relationship for DR, DN and PAD. The addition of CTI to the model significantly improved NRI and Harrell's concordance index for DVCs incidence.
CONCLUSIONS: Our findings demonstrated a significant positive association between CTI and the incidence of DVCs in individuals with T1D and T2D. Furthermore, our study revealed dose-response relationships between CTI and all subtypes of DVCs, providing more comprehensive risk stratification information for individuals with diabetes.
PMID:42876706 | PMC:PMC13662079 | DOI:10.1136/bmjph-2025-004508