Abstract
ABSTRACT Background Although pediatric myopia is well researched, adult-onset myopia remains poorly understood. This study investigated the associations between serum lipid biomarkers and adult-onset myopia, and the potential mediating mechanisms. Methods This prospective population-based cohort study included 248,682 UK Biobank participants (aged 37-73 years) without myopia at baseline in 2006-2010. Seven lipid biomarkers - namely, apolipoprotein A (APOA), apolipoprotein B (APOB), total cholesterol, high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), lipoprotein A, and triglycerides - were analyzed. Hazard ratios (HRs) were estimated in Cox proportional hazard models adjusted for sociodemographics, lifestyle factors, and comorbidities. Nonlinear relationships were evaluated by restricted cubic spline (RCS) analysis. Robustness was evaluated by sensitivity analysis. The roles of inflammation (C-reactive protein), glucose homeostasis, insulin-like growth factor 1 (IGF-1), vitamin D, and testosterone were assessed by mediation analysis. Results Over a median 15.16 years of follow-up, 27,929 (11.2%) of the study participants developed myopia. Elevated APOA (HR: 1.28, 95% confidence interval [CI]: 1.21-1.35) and HDL-C (HR: 1.22, 95%CI: 1.17-1.27) were associated with a significant increase in the risk of myopia (both p < 0.001). Conversely, higher levels of APOB (HR: 0.91, 95%CI: 0.86-0.96), LDL-C (HR: 0.96, 95%CI: 0.95-0.98), total cholesterol (HR: 0.99, 95%CI: 0.97-1.00), and triglycerides (HR: 0.97, 95%CI: 0.95-0.98) correlated with a reduced risk of myopia (all p < 0.05). There was no significant association between the lipoprotein A level and myopia. RCS analysis revealed the following nonlinear relationships: APOA (inverted U-shaped, inflection at 1.8 g/L; p < 0.0001), HDL-C (threshold effect at 1.5 mmol/L; p = 0.0021), and triglycerides (inverted U-shaped, inflection at 1.2 mmol/L; p = 0.0356). The results remained consistent after excluding participants who had an eye infection or underwent ophthalmic surgery in the 4 weeks before the baseline assessment. Key mediators identified were vitamin D (explaining 16.21%-53.69% of associations), glucose (5.78%-13.32%), testosterone (0.68%-19.17%), C-reactive protein (0.49%-8.63%), and IGF-1 (0.55%-3.55%). Conclusions Specific lipid profiles demonstrated associations with the risk of adult-onset myopia, with nonlinear dose-response patterns. Vitamin D, systemic inflammation, and hormonal pathways may partially mediate these relationships. These findings advance our understanding of the metabolic factors contributing to the development of refractive error and suggest novel targets for preventive strategies, including a lifelong approach to the management of adult-onset myopia. </p>