Purchase this article with an account.
Ashley Li, Gayatri Susarla, Weilin Chan, Samaneh Davoudi, Tina Ahmadi, Shaleen Sathe, Lisa Michelle Tom, George Papaliodis, Josep Mercader, Aaron Leong, Lucia Sobrin; Mendelian Randomization Supports a Casual Effect of Low Vitamin D on Non-infectious Uveitis Risk. Invest. Ophthalmol. Vis. Sci. 2021;62(8):3478.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Infectious and non-infectious uveitis (NIU) are significant causes of vision loss. Several studies have reported an association between NIU and low Vitamin D levels. However, a causal relationship between low Vitamin D level and NIU has not been established. We employed Mendelian randomization (MR) to investigate a possible causal association between Vitamin D levels and NIU.
Cases and controls were identified from the Massachusetts General Brigham Biobank. NIU cases were found by diagnostic codes and verified with an electronic medical record review. Controls were identified by the absence of uveitis diagnostic codes. Systemic diseases associated with NIU and diseases associated with Vitamin D metabolism were recorded for cases and controls. Genome-wide genotyping was performed on all patients using the Illumina Multi-Ethnic Global Array followed by imputation based on the European Haplotype Reference Consortium Panel. Then, 25-hydroxy Vitamin D (25OHD) candidate instruments (genetic variants) were selected from the summary statistics of a large genome-wide association study for Vitamin D level. Two-sample MR was performed where the causal effect was evaluated by the random-effects inverse-variance weighted (IVW) method. Individual 25OHD loci were evaluated for association with NIU.
375 cases and 4,167 controls were identified. 75 genetic variants were used in the primary MR analysis. We found a suggestive association of genetically decreased 25OHD with increased NIU risk (OR=2.17, 95%CI=0.99-4.73, P=.05; Fig. 1). Other MR methods and the leave-one-out analysis yielded consistent results. Excluding variants in the vitamin D binding gene (GC) showed a significant association (OR=2.40, 95%CI =1.06-5.46, P=.04; Fig. 1). In analyzing individual loci, genetically decreased 25OHD was associated with increased NIU risk using 6 variants within the CYP2R1 locus (r2<.01 and >1 kb apart) (OR=7.09, 95%CI =2.76-18.25, P=4.8x10-5; Fig. 2). We identified no heterogeneity of effects or outlying genetic variants in the primary MR or individual locus analyses.
Our findings support a causal association between low Vitamin D levels and higher risk of NIU. This provides further support for a clinical study of Vitamin D supplementation as an adjunct intervention for preventing NIU recurrences or inducing remission.
This is a 2021 ARVO Annual Meeting abstract.
Figure 1. MR of 75 genetic variants
Figure 2. MR of variants in CYP2R1
This PDF is available to Subscribers Only