Causal Relationship between Adiponectin and Diabetic Retinopathy: A Mendelian Randomization Study in an Asian Population.

Adiponectin (APN) is suggested to be a potential biomarker for predicting diabetic retinopathy (DR) risk, but the association between APN and DR has been inconsistent in observational studies. We used a Mendelian randomization (MR) analysis to evaluate if circulating APN levels result in DR. We applied three different genetic risk scores (GRS): GRSAll combined all 47 single nucleotide polymorphisms (SNPs), which from a genome-wide association study (GWAS) database-catalog reach significance level; GRSLimited comprised 16 GRSAll-SNPs with a rigorous threshold (p < 5.0 × 10-8 for GWAS), and GRSAPN combined 5 SNPs significantly associated with APN level. The MR-inverse-variance weighted method analysis showed that for each 1-SD increase in genetically induced increase in plasma APN, the OR of having DR was β = 0.20 (95% CI: -0.46-0.85, p = 0.553) for GRSAPN, 0.61 (95% CI: 0.10-1.13, p = 0.020) for GRSAll, and 0.57 (95% CI: -0.06 to 1.20, p = 0.078) for GRSLimited. Sensitivity analysis, including MR-egger regression and the weighted-median approach, did not provide evidence of the pleiotropic effect of IVs. Limited evidence for the causal role of APN in DR risk among Taiwanese diabetic patients was shown based on MR analysis in the present study.