Objective To elucidate the causal relationship between specific immune cells and autoimmune hepatitis (AIH) using a two-sample Mendelian randomization (MR) approach.

Methods A bidirectional MR analysis was conducted using data from large publicly accessible Genome-Wide Association Study (GWAS) databases. The inverse variance weighted (IVW) method was employed as the primary method to evaluate the relationship between 731 immune cell traits and AIH. The false discovery rate (FDR) was controlled using the Benjamini-Hochberg correction. Additionally, pleiotropy and heterogeneity tests were performed, and a leave-one-out sensitivity analysis was conducted to further validate the robustness of the results.

Results At a significance level of 0.20, it was found that the absolute count of CD28⁻CD8⁺ regulatory T-cells (IVW: odds ratio OR = 1.486; 95% confidence interval CI, 1.189-1.859; P < 0.001; P_FDR = 0.185), the level of CD28 on CD39⁺ secreting regulatory T-cells (IVW: OR = 1.194; 95% CI, 1.074-1.328; P = 0.001; P_FDR = 0.185), and the level of CD45 on mononuclear myeloid-derived suppressor cells (IVW: OR = 1.243; 95% CI, 1.108-1.394; P < 0.001; P_FDR = 0.143) were associated with an increased risk of AIH. The level of programmed death-ligand 1 on CD14⁺CD16⁺ monocytes (IVW: OR = 0.849; 95% CI, 0.771-0.935; P < 0.001; P_FDR = 0.185) was associated with a reduced risk of AIH.

Conclusion Four immune cell phenotypes associated with AIH risk are identified. Further investigation is needed to validate these findings and explore new therapeutic avenues.