**Background:** Coffee and tea are widely consumed beverages containing caffeine and antioxidants (e.g., flavonoids, polyphenols) that may have neuroprotective effects. Epidemiological studies have linked their intake to reduced risk of Parkinson's disease and dementia, but MRI findings on brain volume are inconsistent. The retina, as an extension of the central nervous system, can be non-invasively imaged via optical coherence tomography (OCT). Retinal nerve fiber layer (RNFL) thinning is associated with cognitive decline and neurodegenerative diseases such as glaucoma, Alzheimer's, and Parkinson's. This study aimed to investigate associations between self-reported coffee and tea consumption and macular RNFL (mRNFL) thickness in a large UK Biobank subgroup.
**Methods:** This cross-sectional analysis included 35,557 participants from the UK Biobank (aged 40–69) who completed baseline OCT measurements (2009–2010) and had no history of glaucoma, retinal disorders, multiple sclerosis, dementia, or Parkinson's disease. Coffee and tea consumption were self-reported as cups per day and categorized into 0, 0.5–1, 2–3, and ≥4 cups/day. Instant coffee intake was also recorded. Macular OCT scans (6×6 mm²) were obtained using spectral-domain OCT (Topcon 3D OCT-1000 Mark II), and average mRNFL thickness across six subfields was analyzed. Multivariable linear regression models adjusted for demographics (age, sex, assessment center), socioeconomic factors (Townsend deprivation index, income, ethnicity, education), lifestyle (BMI, physical activity, sleep, smoking, alcohol, diet), health-related covariates (diabetes, cardiovascular disease, hypertension, HDL/LDL cholesterol), and ocular parameters (spherical equivalent, intraocular pressure). Restricted cubic spline (RCS) models explored non-linear associations. Sensitivity analyses were performed by age (≤60 vs. >60 years) and sex.
**Key Results:** Coffee was consumed by 78% and tea by 86% of participants. In fully adjusted models (Model 3, including instant coffee adjustment), coffee consumption was significantly associated with increased mRNFL thickness (β = 0.13, 95% CI = 0.01–0.25), with the strongest effect in those drinking 2–3 cups/day (β = 0.16, 95% CI = 0.03–0.30). RCS analysis revealed an inverted U-shaped association (p for non-linear = 0.01). Tea consumption was also significantly associated with thicker mRNFL (β = 0.13, 95% CI = 0.01–0.26), particularly for ≥4 cups/day (β = 0.15, 95% CI = 0.01–0.29; p for trend = 0.03), with a linear dose-response (p for non-linear = 0.29). Instant coffee intake was independently associated with thinner mRNFL (β = −0.19, 95% CI = −0.29 to −0.10). No significant interaction was found between coffee and tea (p = 0.84). Sensitivity analyses showed no effect modification by age or sex (all p for interaction > 0.05).
**Clinical Implications:** This study provides novel evidence linking moderate coffee (2–3 cups/day) and tea (≥4 cups/day) consumption to thicker mRNFL, a retinal marker of neurodegeneration. These findings support the hypothesis that these beverages may have neuroprotective properties, potentially through anti-inflammatory and antioxidant mechanisms. However, the cross-sectional design precludes causal inference, and the negative association with instant coffee warrants caution. The results highlight the need for longitudinal studies and clinical trials to validate these associations and to consider coffee type in future research. Given the widespread consumption of coffee and tea, even modest neuroprotective effects could have significant public health implications for preventing neurodegenerative diseases.