Objective To explore the causal relationship between body mass index (BMI) and vitiligo using the two-sample Mendelian randomization (MR) analysis method. Methods This study was a two-sample MR analysis based on summary data from a genome-wide association study. The exposure data (BMI, n=339,224) and outcome data (vitiligo, n=337,159) were both derived from the European population. The screening criteria for instrumental variables were as follows: (1) Single nucleotide polymorphisms (SNPs) significantly associated with BMI at the genome-wide level (P<5×10-8) were selected; (2) Linkage disequilibrium filtering was performed on the initially screened SNPs (r2<0.001, physical distance>10,000 kb); (3) The PhenoScanner database (v2) was used to retrieve the associated phenotypes of each SNP, and SNPs significantly associated with known confounding factors (such as neurological diseases, mood disorders, sleep status, etc.) were excluded. Five MR analysis methods, including random-effects inverse variance weighting (IVW), MR Egger regression, weighted median estimator (WME), simple mode, and weighted mode, were used to study the causal relationship between BMI and vitiligo. Cochran's Q test was used to assess heterogeneity, the intercept term test in MR Egger regression was used to evaluate horizontal pleiotropy, and a "leave-one-out" method was used for sensitivity analysis. Results The results of all five MR analysis methods showed no statistically significant causal relationship between BMI and vitiligo (all P>0.05). Specifically, the OR obtained by the random-effects IVW method was 1.000 1 (95% CI: 0.999 6-1.000 6); the OR from MR Egger regression was 1.000 5 (95% CI: 0.999 1-1.001 8); the OR via WME was 1.000 1 (95% CI: 0.999 5-1.000 8); the OR using the simple mode was 1.000 2 (95% CI: 0.998 7-1.001 8); and the OR from the weighted mode was 1.000 2 (95% CI: 0.999 0-1.001 4). The results of Cochran's Q test showed no statistically significant heterogeneity among SNPs strongly associated with BMI (P>0.05). The intercept term test results of MR Egger regression indicated no horizontal pleiotropy in SNPs strongly associated with BMI (P>0.05). The results of the "leave-one-out" sensitivity analysis showed that after excluding SNPs one by one, the IVW analysis results did not change significantly. Conclusion No genetic-level causal relationship between BMI and vitiligo was found in this study.

无