Abstract
Purpose: To investigate the associations between the estimated glucose disposal rate (eGDR), a composite indicator of insulin sensitivity, and the risks of incident age-related cataract (ARC) and AMD, with a focus on nonlinear dose-response patterns, age-specific heterogeneity, and genetic modification.</p>
Methods: This prospective cohort study included 444,137 UK Biobank participants free of ARC and AMD at baseline. eGDR was analyzed both as a continuous variable (per 1-SD increase) and by quartiles. Multivariable Cox proportional hazards models were used to estimate hazard ratios and 95% confidence intervals. Restricted cubic spline models assessed nonlinear associations and identified metabolic plateau ranges. Age-stratified analyses (<60 years vs. ≥60 years), landmark analyses, polygenic risk score-stratified analyses, and multiple sensitivity analyses were conducted to evaluate robustness.</p>
Results: A total of 29,215 incident ARC cases and 8992 incident AMD cases were documented during follow-up. A higher eGDR was consistently associated with lower risks of both ARC and AMD, corresponding with an approximately 7% to 8% risk reduction per 1-SD increase after full adjustment. Quartile-based analyses demonstrated clear dose-response relationships, with participants in the highest eGDR quartile exhibiting substantially lower risks compared with those in the lowest quartile. Restricted cubic spline analyses revealed significant nonlinear, L-shaped associations, characterized by a steep risk decline at lower eGDR levels followed by a risk-neutral metabolic plateau. The plateau range occurred at higher eGDR levels among participants aged <60 years than among those aged ≥60 years. Associations were stronger during extended follow-up periods and remained robust across landmark, competing-risk, and other sensitivity analyses. Moreover, higher eGDR levels attenuated genetic susceptibility to ARC and AMD across polygenic risk score strata.</p>
Conclusions: A higher eGDR is associated with lower risks of incident ARC and AMD in a nonlinear and age-dependent manner. These findings highlight insulin sensitivity as a modifiable metabolic factor in ocular aging and support the potential value of metabolic risk stratification for age-related eye diseases.</p>