Growing evidence indicates that immune and metabolic dysregulation contribute to the development of appendicitis; however, the causal pathways remain unclear. This study employed Mendelian randomization (MR) to assess the effect of immune cell traits on appendicitis risk, with an emphasis on the mediating role of plasma metabolites. We utilized publicly available summary statistics from genome-wide association studies based on European cohorts, encompassing 731 immune cell traits, 1400 plasma metabolites, and appendicitis. The inverse variance weighted method served as the primary analytical strategy. Mediation analysis was conducted to estimate indirect effects, and reverse MR was performed to evaluate reverse causality. To address multiple testing in these high-dimensional screens, Benjamini–Hochberg false discovery rate adjustment was additionally applied within the immune-cell and plasma-metabolite discovery stages. Three immune cell traits and 12 plasma metabolites were significantly associated with appendicitis. Specifically, CD86+ plasmacytoid DC %DC exhibited a protective effect (odds ratio [OR]: 0.910,
P
= .003), while CD28+ DN (CD4-CD8-) %T cell (OR: 1.090,
P
= .001) and CD27 on IgD+ CD38- unsw mem (OR: 1.071,
P
= .002) were associated with increased risk. Among the identified metabolites, 7 exhibited protective effects, whereas 5 were linked to elevated risk. Notably, 13-HODE + 9-HODE levels partially mediated the effect of CD28+ DN (CD4-CD8-) %T cell on appendicitis (mediation proportion: 17.7%,
P
< .001). No evidence of heterogeneity, horizontal pleiotropy, or reverse causation was detected in sensitivity analyses and reverse MR. However, after false discovery rate adjustment, none of the immune-cell or metabolite associations remained significant; these signals should therefore be interpreted as exploratory candidates rather than definitive findings. This study provides evidence supporting a role for the immune–metabolic axis in the pathogenesis of appendicitis, providing novel insights into causal mechanisms and informing potential strategies for targeted prevention. Accordingly, the present results are best viewed as hypothesis-generating and require validation in stratified and prospective datasets before clinical translation.