Causal association between antidiabetic drug targets and risk of Bell’s palsy: A 2-sample Mendelian randomization study

To investigate the causal effects of antidiabetic drug targets on Bell's palsy risk using Mendelian randomization (MR), addressing limitations of observational studies and informing therapeutic strategies for diabetic patients. A 2-sample MR analysis was conducted using genetic instruments for 5 major antidiabetic drug classes: sulfonylureas (ABCC8/KCNJ11), metformin (GPD1), thiazolidinediones (PPARG), GLP1-RAs (GLP1R), and SGLTi (SLC5A1/2). Instrumental variables were selected from cis-regions (±500 kb) of target genes using HbA1c-associated single nucleotide polymorphisms (GWAS ID: ebi-a-GCST90014006) with stringent clumping (P < 5 × 10-8, r2 < 0.2) and F-statistic > 10 to ensure robustness. Bell's palsy data were sourced from FinnGen (finn-b-G6_BELLPA) including. Causal effects were assessed via inverse variance weighted, MR-Egger, and weighted median methods, complemented by sensitivity analyses (MR-Pleiotropy RESidual Sum and Outlier, Cochran's Q). Genetically proxied sulfonylurea (via ABCC8/KCNJ11) and metformin (via GPD1) targets exhibited significant protective effects against Bell's palsy, with odds ratios of 0.19 (95% CI: 0.04-0.92, P = .039) and 0.16 (0.03-0.87, P = .034), respectively. Sensitivity analyses confirmed minimal pleiotropy (MR-Egger intercept P > .05) and heterogeneity (Cochran's Q P > .05). No associations were observed for thiazolidinediones, GLP1-RAs, or SGLTi. Scatter plots and inverse variance weighted median concordance supported robustness for ABCC8/KCNJ11, though limited single nucleotide polymorphisms for GPD1 warranted cautious interpretation. This MR analysis provides genetic evidence that sulfonylureas and metformin may confer neuroprotection against Bell's palsy, supporting potential repurposing of these agents to mitigate neurological complications in diabetic patients.