Background: Pancreatitis is a serious and complex inflammatory disease that imposes a severe effect on quality of life. Links between plasma lipidome and pancreatitis have been reported, some of which have not yet been clearly elucidated.
Methods: Therefore, our study aimed to investigate the causal relationships between plasma lipidome and four types of pancreatitis by conducting a bidirectional, two-sample Mendelian randomization (MR) analysis. We obtained genetic variants associated with 179 lipid species from a Genome-wide association analysis of plasma lipidome. The aggregated statistical data of acute pancreatitis (AP), alcohol-induced acute pancreatitis (AAP), chronic pancreatitis (CP), and alcohol-induced chronic pancreatitis (ACP) from the FinnGen consortium were exploited as the outcome. The inverse variance weighted (IVW) technique as the main method was used for MR analysis and sensitivity analyses were used to evaluate heterogeneity and pleiotropy.
Results: After FDR correction, SE (27:1/20:4) (OR = 0.938, 95%CI = 0.906-0.972, P = 4.38 × 10-4, PFDR = 0.039) was identified to be significantly associated with AP risk. Eight lipid species were identified to be significantly associated with CP risk: SE (27:1/20:4) (OR = 0.911, 95%CI = 0.869-0.954, P = 8.89 × 10-5, PFDR = 0.016), LPC (20:4) (OR = 0.892, 95%CI = 0.843-0.945, P = 9.74 × 10-5, PFDR = 0.009), PC (16:0_22:5) (OR = 0.880, 95%CI = 0.818-0.947, P = 6.29 × 10-4, PFDR = 0.028), PC (17:0_20:4) (OR = 0.893, 95%CI = 0.842-0.948, P = 1.76 × 10-4, PFDR = 0.010), PC (18:0_20:4) (OR = 0.920, 95%CI = 0.874-0.969, P = 1.70 × 10-3, PFDR = 0.038), PC (O-16:0/20:4) (OR = 0.871, 95%CI = 0.804-0.943, P = 6.95 × 10-4, PFDR = 0.025), PC (O-16:1/20:4) (OR = 0.890, 95%CI = 0.832-0.953, P = 7.85 × 10-4, PFDR = 0.023), and PE (O-18:1/20:4) (OR = 0.866, 95%CI = 0.791-0.947, P = 1.61 × 10-3, PFDR = 0.041). Furthermore, genetically predicted increased LPC (20:4) (OR = 0.862, 95%CI = 0.796-0.934, P = 3.00 × 10-4, PFDR = 0.027) and SM (34:2;O2) (OR = 0.753, 95%CI = 0.659-0.860, P = 2.97 × 10-5, PFDR = 0.005) levels were associated with decreased risk of ACP.
Conclusions: Our findings provide evidence of causal associations between the specific types of lipidome and pancreatitis, offering new insights into future clinical research.
Keywords: bidirectional Mendelian randomization; causal inference; lipidome; pancreatitis; sensitivity.
Copyright © 2024 Ma, Chen, Wang, Guo and Zhang.