Abstract
This study aimed to investigate the bidirectional causal relationship between lipidomes and ulcerative colitis (UC) using a 2-step, 2-sample Mendelian randomization (MR) analysis. We also examined the mediating role of inflammatory proteins as intermediaries. We screened relevant summary statistics from European populations and conducted a 2-sample, bidirectional MR analysis on 179 lipids and UC using the inverse variance-weighted method. Causal effects were further validated through Bayesian-weighted MR. A 2-step MR analysis was performed to assess the mediating role of inflammatory proteins. Sensitivity, heterogeneity, and horizontal pleiotropy analyses were conducted to confirm the robustness of the causal relationships. The MR analysis demonstrated a significant positive association between PC (17:0_18:2) and UC (odds ratio [OR] = 1.159, 95% confidence interval [CI] = 1.062-1.266, P = 4.01e-5). TNFSF12 was also positively associated with UC (OR = 1.096, 95% CI = 1.011-1.189, P = .027). However, PC (17:0_18:2) and TNFSF12 were negatively correlated (OR = 0.943, 95% CI = 0.898-0.990, P = .018), with a mediation proportion of 3.67%. Our findings identified causal relationships between 9 lipidomes and 2 inflammatory proteins with UC. TNFSF12 showed a masking effect on UC for PC (17:0_18:2).