Causal Associations of Thyroid Function and Age-Related Macular Degeneration: A Two-Sample Mendelian Randomization Study

Abstract

To determine whether causal association lies between thyroid function and age-related macular degeneration (AMD) risk in human beings. Two-sample Mendelian randomization (MR) study. The single-nucleotide polymorphisms associated with free thyroxine (FT4) and thyroid-stimulating hormone (TSH) were selected from a genome-wide association study (GWAS) of 72,167 individuals of European descent. Summary-level data for AMD were obtained from a GWAS published by the International Age-related Macular Degeneration Genomics Consortium of 33,526 individuals (16,144 cases and 17,832 controls). An inverse-variance-weighted (IVW) method was the main MR analysis. Maximum likelihood, weighted median, MR-Egger, MR-pleiotropy residual sum outlier methods were used for the sensitivity analysis. An increase of 1 SD in genetically predicted FT4 levels was found to be significantly associated with an 18.9 % increase in the overall AMD risk (P=.005). In the multivariable MR analysis controlling for TSH level, the causal effect of FT4 level on the risk of AMD remained (odds ratio [OR]=1.207, P=.004). A 1-SD increase in TSH levels was nominally associated with a 10.0% decrease in the overall AMD risk (P=.032). After adjusting for FT4 level by multivariable MR analysis, no direct causal relationship was found between TSH level and AMD risk (95% CI=0.810, 1.125, P=.582). Genetic variants predisposing to higher FT4 levels within the normal range were associated with higher AMD risk. Further studies are required to understand the mechanism underlying this putative causal relationship.