Fundus-Derived Predicted Age Acceleration in Glaucoma Patients Using Deep Learning and Propensity Score-Matched Controls.

Glaucoma, a leading cause of irreversible blindness, has been associated with systemic and ocular aging processes. This study aimed to investigate the relationship between glaucoma and accelerated biological aging using fundus-derived age prediction. Additionally, the role of systemic factors and retinal vascular changes in this association was explored. A total of 6023 participants, including 547 glaucoma patients and 547 matched controls, were analyzed. Fundus-derived predicted age was assessed using a deep learning model (EfficientNet). Systemic factors such as BMI, blood pressure, lipid profiles, liver function markers, glucose levels, and retinal vascular changes (Scheie classifications) were analyzed. Statistical comparisons and multivariate regression analyses were performed to evaluate the impact of glaucoma on predicted age acceleration, adjusting for age, gender, and systemic factors. Glaucoma was significantly associated with higher predicted age acceleration (prediction difference: -1.5 ± 4.5 vs. -2.1 ± 4.5 years; = 0.040). Multivariate regression confirmed that glaucoma independently influenced predicted age ( = 0.021) and prediction difference ( = 0.021). Among systemic factors, γ-GTP was positively associated with prediction difference ( = 0.036), while other factors, such as BMI, blood pressure, and glucose levels, showed no significant association. Retinal vascular changes, including hypertensive and sclerotic changes (Scheie classifications), were significantly more prevalent in glaucoma patients and correlated with predicted age acceleration. Glaucoma is associated with accelerated biological aging, as indicated by fundus-derived predicted age. Systemic factors such as γ-GTP and retinal vascular changes may play contributory roles. Fundus-derived predicted age holds promise as a non-invasive biomarker for monitoring systemic aging. Further longitudinal studies are warranted to establish causal relationships and enhance clinical applications.