Abstract
BACKGROUND: Regarding the impact of metformin on COVID-19, there are currently varying opinions from multiple studies. Growth differentiation factor 15 (GDF-15) is a biomarker of metformin use and dosage, and we used two-sample Mendelian randomization (MR) to assess the causal effect of GDF-15 (metformin) on COVID-19 susceptibility, hospitalization, and severe COVID-19, thereby guiding the selection of glucose-lowering agents for diabetic patients during the COVID-19 pandemic. METHODS: Two sets of genetic tools were utilized for MR analysis, derived from publicly available genetic data. The first set was GDF-15 genome-wide association study (GWAS) data from a study with 5440 participants, while the second set was COVID-19 GWAS data from the Host Genetics Initiative (HGI) GWAS meta-analysis. The primary method used to assess causal effects was random effects inverse variance weighted estimation. Complementary methods included weighted median and MR-Egger analyses. Sensitivity analysis was performed using Cochran Q tests, MR-Egger intercept tests, MR-PRESSO, leave-one-out analyses, and funnel plots. RESULTS: GDF-15 increased the risk of severe COVID-19 in patients (OR = 1.10, 95% CI 1.03-1.19; P = .006); there was no causal effect of GDF-15 on hospitalization for COVID-19 (OR = 1.02, 95% CI 0.96-1.07; P = .47) or susceptibility to COVID-19 in the general population (OR = 1.010, 95% CI 0.988-1.034; P = .354). CONCLUSIONS: Our study supports the notion that GDF-15 increases the risk of severe COVID-19 in patients. However, there is no causal relationship between GDF-15 and hospitalization or susceptibility to COVID-19.