Causal association of circulating metabolites with diabetic retinopathy: a bidirectional Mendelian randomization analysis.

Abstract

The retina is a highly metabolically active tissue, and there is a lack of clarity about the relationship between metabolites and diabetic retinopathy (DR). This study used two-sample bidirectional Mendelian randomization (MR) analyses to identify causal relationships between metabolites and DR. Genetic variants were selected from the open-access Genome-Wide Association Studies (GWAS) summary database as proxies for the 1400 most recently published metabolites. MR analysis was performed to examine associations between these metabolite traits and DR. Single nucleotide polymorphism (SNP) data that were significantly associated with exposure were screened through association analysis. Validated instrumental variables (IVs) were obtained by removing SNPs with linkage disequilibrium (LD) and F-statistic values below 10. MR analyses were performed using the inverse variance weighted (IVW) method as the primary approach. The robustness of the results was verified by sensitivity analyses, including assessments of heterogeneity, horizontal pleiotropy, and the leave-one-out method. In the IVW approach and in the primary analysis of several sensitivity analyses, genetically determined glycolithocholate sulfate levels, androstenediol (3 beta, 17 beta) monosulfate (1) levels, 1-stearoyl-2-arachidonoyl-GPE (18:0/20:4) levels, 1-oleoyl-2-arachidonoyl-GPE (18:1/20:4) levels, 1-oleoyl-2-linoleoyl-GPE (18:1/18:2) levels, X-26109 levels, N6-methyllysine levels, (N6,N6-dimethyllysine levels), and (N2-acetyl,N6,N6-dimethyllysine levels) were negatively associated with the risk of DR. 5-hydroxymethyl-2-furoylcarnitine levels and the glutamate-to-alanine ratio were positively associated with the risk of DR. No reverse causal association was found between DR and metabolites. This MR study suggests that nine metabolites may have a protective effect in DR, while two metabolites may be associated with an increased risk of DR. However, further research is needed to confirm these findings. Supplementation with beneficial metabolites may reduce DR risk and could potentially be a novel therapeutic approach to DR treatment.