Cross-sectional study finds longer axial length is associated with lower odds of diabetic retinopathy in adults with diabetes, suggesting an ocular structural correlate rather than causal protection.
Longer axial length has been associated with lower odds of diabetic retinopathy (DR), but population-based evidence using directly measured axial length remains limited. We evaluated this association among Korean adults with diabetes using Korea National Health and Nutrition Examination Survey (KNHANES) 2017–2021 data. This cross-sectional study restructured right- and left-eye data into an eye-level dataset matching axial length with eye-specific DR status. The primary analysis used rescaled examination-weighted generalized estimating equations (GEE) accounting for inter-eye correlation and adjusted for age, sex, hypertension, body mass index, HbA1c, diabetes duration, and serum creatinine. Secondary analyses included unweighted GEE, axial length quartiles, ocular covariate adjustment, participant-level survey logistic regression incorporating KNHANES strata, primary sampling units, and examination weights, restricted cubic spline analyses, and an exploratory analysis of moderate-or-worse DR. Diabetic macular edema (DME) was summarized descriptively because of sparse events. Among 2,773 eyes from 1,450 participants, each 1-mm longer axial length was associated with lower odds of prevalent DR in the primary examination-weighted GEE model (OR, 0.881; 95% CI, 0.786–0.988; P = 0.031). Q4 versus Q1 showed an OR of 0.687 (95% CI, 0.486–0.972; P = 0.034), with a significant quartile trend ( P = 0.027). In participant-level design-based analyses incorporating KNHANES strata, primary sampling units, and examination weights, the association remained inverse but was not significant for right eyes (OR, 0.855; 95% CI, 0.723–1.011; P = 0.067), whereas it was significant for left eyes (OR, 0.832; 95% CI, 0.702–0.987; P = 0.035). No significant non-linearity or axial length-by-sex interaction was detected. The association was attenuated after spherical equivalent adjustment and was not significant for moderate-or-worse DR. DME was uncommon, and was evaluated descriptively. Longer axial length was modestly associated with lower odds of prevalent DR, although statistical significance varied across analytic approaches. These cross-sectional findings support axial length as a contextual ocular structural correlate rather than a causal protective factor. Not applicable.
No takes yet. Share an insight, caveat, or question.
Choi et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: