[Effect analysis of trimethylamine N-oxide and its precursors on susceptibility to pancreatic diseases].

Objective: To investigate the causal relationship between trimethylamine N-oxide (TMAO) and its precursors (betaine, carnitine, and choline) and pancreatic diseases based on the Mendelian randomization (MR) method.

Methods: Genome-wide association study data of TMAO, betaine, carnitine, choline, acute pancreatitis (AP), chronic pancreatitis (CP), pancreatic cancer (PC), and circulating immune cell characteristics (white blood cell, lymphocyte, monocyte, neutrophil, eosinophil and basophil) were collected. According to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE)-MR reporting guidelines, the available genetic variants [single nucleotide polymorphism (SNP)] were strictly screened. The causal relationship between exposure (TMAO and its precursors) and outcomes (pancreatic diseases and circulating immune cell characteristics) was evaluated using inverse variance weighting (IVW), MR-Egger regression and weighted median. The reliability of the results was evaluated by sensitivity analysis based on MR-Egger regression, MR-PRESSO, Cochrane's Q test and leave-one-out method.

Results: A total of 36 SNP associated with TMAO and its precursors were included. Five of these were associated with TMAO, 13 with betaine, 12 with carnitine, and 6 with choline. (1) MR analysis showed that TMAO may increase the risk of AP [odds ratio (OR) = 1.100, 95% confidence interval (95%CI) was 1.008-1.200, P = 0.032], and choline may reduce the risk of alcoholic acute pancreatitis (AAP; OR = 0.743, 95%CI was 0.585-0.944, P = 0.015). The analysis results of MR-Egger regression and weighted median were consistent with the IVW results. There is no evidence to support a causal relationship between TMAO and its precursors and the risk of CP and PC. Sensitivity analysis indicated that SNP analyzed by MR showed no heterogeneity and low pleiotropy. The leave-one-out method analysis determined that after excluding any SNP, the effect intervals of the remaining SNP on the results were similar to the overall effect intervals, which suggested the robustness of MR results. (2) There was a positive causal relationship between plasma TMAO level and circulating monocyte count (OR = 1.017, 95%CI was 1.000*-1.034, P = 0.048, * represented that the data was obtained by correcting to 3 decimal places from 1.000 1). The causal effect obtained by MR-Egger regression and weighted median analysis was consistent with the results of IVW. Sensitivity analysis illustrated SNP analyzed by MR showed no heterogeneity and pleiotropy. The leave-one-out method analysis determined that after excluding any SNP, the effect intervals of the remaining SNP on the results were similar to the overall effect intervals, which suggested the robustness of MR results.

Conclusions: TMAO and choline may change the risk of AP, and TMAO may contribute to the increase of circulating monocyte count in AP.