Association of coffee, tea, and caffeine consumption with intraocular pressure and interaction with genetic risk: findings from the UK Biobank

Abstract

Purpose : Coffee and tea are commonly consumed caffeinated beverages that may affect ocular health. Hence, it is of interest whether caffeine intake affects intraocular pressure (IOP). Some studies reported that coffee and caffeine consumption is associated with higher IOP. However, evidence from large-scale general population studies is limited.

Methods : We included 121,755 UK Biobank participants (aged 39-73y) who had dietary data and cornea-compensated IOP measurements in 2006-2010. In a subset (n=78,017), data were available from up to five web-based 24-hour-recall food frequency questionnaires (2009-2012); we derived participants’ total caffeine consumption based on caffeine content and frequency of caffeine-containing food intake. Using multivariable linear regression, we assessed the cross-sectional relationships between coffee or tea intake (cups/day) and IOP (mmHg) as primary analyses and the associations between total caffeine intake (80 mg/day) and IOP (mmHg) as secondary analyses. Analyses were adjusted for age, sex, ethnicity, smoking, alcohol intake, physical activity, deprivation, body mass index, blood pressure, and diabetes (and total energy intake in the subset). To explore gene-diet interactions, we examined interactions between coffee, tea, caffeine intake and an IOP polygenic risk score (PRS) combining effects of 111 genetic variants associated with IOP identified by genome-wide association studies.

Results : Mean (SD) IOP was 16.0 (3.8) mmHg. Higher coffee, tea, and total caffeine intakes were associated with lower IOP (-0.049, -0.045 and -0.039 mmHg, respectively; P ≤ 0.001). When evaluating PRS-diet interactions, we found positive and significant interactions with tea and caffeine intake (both P-interactions < 0.05), where higher tea and caffeine intake was associated with lower IOP among individuals with IOP PRS in the lower 75^th percentile, while among individuals with high IOP PRS (top 25^th percentile), higher intakes were associated with modestly higher IOP.

Conclusions : We found evidence for very weak associations between higher coffee, tea and caffeine intake and lower IOP. However, our finding on gene-diet interactions suggest that genetic predisposition to higher IOP may influence the association between caffeinated beverage consumption and IOP.

This is a 2020 ARVO Annual Meeting abstract.