Association between triglyceride-glucose indices and incident glaucoma: integrated UK Biobank and Mendelian randomization study

Abstract Background To evaluate the associations of the triglyceride-glucose (TyG) index and related adiposity-integrated indices with incident glaucoma and explore potential causal evidence using Mendelian randomization (MR). Methods This prospective cohort study included 396,213 participants without glaucoma of European ancestry from the UK. The TyG index and four related indices, including TyG-body mass index (TyG-BMI), TyG-waist circumference (TyG-WC), TyG-waist-to-height ratio (TyG-WHtR), and TyG-waist-to-hip ratio (TyG-WHR), were assessed. Multivariate Cox models and restricted cubic splines were used to examine the association with incident glaucoma. Associations between the corneal-compensated intraocular pressure (IOPcc) and retinal structures were assessed using regression analyses. The predictive performance was evaluated using extreme gradient boosting (XGBoost). Two-sample MR was conducted to assess the genetic association between the TyG index and glaucoma. Results During a mean follow-up of 8.06 years, 6957 cases of incident glaucoma occurred. In fully adjusted models, higher TyG and related indices were associated with increased glaucoma risk, including TyG (hazard ratio [HR] = 1.05, 95% confidence interval [CI]: 1.02–1.08), TyG-BMI (HR = 1.20, 95% CI: 1.14–1.27), TyG-WC (HR = 1.30, 95% CI: 1.23–1.37), TyG-WHtR (HR = 1.26, 95% CI: 1.20–1.31), and TyG-WHR (HR = 1.16, 95% CI: 1.12–1.21). The TyG showed a significant non-linear U-shaped association with incident glaucoma. TyG-related indices were positively associated with IOPcc, although the effect sizes were small and inversely associated with several measurements of retinal layer thickness. The XGBoost model showed good discrimination, with TyG-WHtR contributing the strongest among the TyG-related indices (C-index = 0.851, area under the curve = 0.823). MR provided suggestive evidence for an association between genetically predicted TyG and glaucoma risk (odds ratio = 1.27, 95% CI: 1.00–1.62, P = 0.046), although sensitivity estimates were not statistically significant. Conclusions TyG-related indices, particularly adiposity-integrated measures, were associated with a higher glaucoma risk. Systemic glucose-lipid metabolic dysfunction and central adiposity may contribute to glaucoma susceptibility. Thus, TyG-related indices may serve as low-cost adjunct indicators for glaucoma risk stratification and external validation.