Abstract
BACKGROUND: Retinal sublayer thickness is associated with both ophthalmic and neurodegenerative diseases, serving as a valuable biomarker. Despite the established role of diet in chronic disease prevention, the relationship between healthy dietary patterns and retinal sublayer thickness remains underexplored. OBJECTIVE: This study aims to investigate the associations between four commonly used diet assessment scores and retinal sublayer thickness in a large-scale population. METHODS: This study utilized data from the UK Biobank, including 13,993 participants with complete optical coherence tomography (OCT) measurements and dietary assessments. The dietary patterns analyzed were the Alternate Mediterranean Diet (AMED), Anti-Empirical Dietary Inflammatory Index (AEDII), Alternate Healthy Eating Index-2010 (AHEI-2010), and Healthful Plant-Based Diet Index (HPDI). Retinal sublayers measured included macular retinal nerve fiber layer (mRNFL), macular ganglion cell-inner plexiform layer (mGCIPL), macular ganglion cell complex (mGCC), retinal pigment epithelium (mRPE), the outer nuclear layer of the central subfield (ONL), photoreceptor inner segment of the central subfield (IS) and total macular thickness. Multivariable linear regression models adjusted for demographics, ophthalmic measurements, and lifestyle factors were employed to examine associations. Mediation analysis was applied to explore the potential mediation effect of several blood biochemical indicators in simple mediation models. RESULTS: Higher HPDI scores were significantly associated with increased central retinal macular thickness (β = 0.106, p = 0.001), mGCIPL thickness (β = 0.017, p = 0.033), mRPE (β = -0.012, p = 0.161), ONL (β = 0.044, p = 0.003), IS (β = 0.003, p = 0.005) and mGCC thickness (β = 0.027, p = 0.025). In contrast, the AMED, AEDII, and AHEI-2010 scores did not show consistent associations with retinal sublayer thickness except ONL and IS. The mediation analysis revealed that cholesterol acted as a suppressor variable, partially mediating the relationship between HPDI and retinal sublayer thickness. Sensitivity analyses indicated that the associations between HPDI and retinal sublayer thickness were more pronounced in younger individuals and females. Additionally, the variability in associations across different age and sex subgroups highlighted the potential influence of demographic factors on dietary impacts. CONCLUSION: The study findings suggest that adherence to a healthful plant-based diet may confer neuroprotective benefits, particularly for retinal ganglion cell health. This large-scale population-based study underscores the potential role of diet in mitigating early neurodegenerative changes detectable through non-invasive retinal imaging. Further longitudinal research is needed to establish causal relationships and elucidate the underlying mechanisms between diet and retinal health.