Potential causal association between gut microbiota, inflammatory cytokines, and acute pancreatitis: A Mendelian randomization study

Abstract

Background

Acute pancreatitis (AP) ranks among the most frequently encountered gastrointestinal diseases in the emergency department. Recent studies have increasingly emphasized the substantial connection among gut microbiota, inflammatory cytokines, and AP.

Methods

A two-sample Mendelian randomization (MR) study was conducted using summary statistics of gut microbiota (GM) from the largest available meta-analysis of genome-wide association studies conducted by the MiBioGen consortium (n=18,340). For cytokines, the data were obtained from a study that investigated genome variant associations with 41 inflammatory cytokines and growth factors (n=8293). The summary statistics of AP were obtained from the FinnGen consortium version R5 data (3022 cases and 195,144 controls). The inverse variance weighted (IVW) method was used as the main analysis, with MR–Egger and weighted median as complementary analytical methods. Sensitivity analyses were performed using Cochran's Q-test, MR–Egger intercept test, leave-one-out analyses, and MR–PRESSO. In addition, we employed the reverse MR analysis and MR Steiger method to estimate the orientations of exposure and outcome.

Result

Among the 211 examined GM taxa, the IVW method revealed that Bacteroidales (odds ratio [OR]=1.412, 95% confidence interval [CI]:1.057 to 1.885, P=0.019), Eubacterium fissicatena group (OR=1.240, 95% CI:1.045 to 1.470, P=0.014), and Coprococcus3 (OR=1.481, 95 % CI:1.049 to 2.090, P=0.026) exhibited a positive association with AP. Conversely, Prevotella9 (OR=0.821, 95% CI:0.680 to 0.990, P=0.038), RuminococcaceaeUCG004 (OR=0.757, 95% CI:0.577 to 0.994, P=0.045), and Ruminiclostridium6 (OR=0.696, 95% CI:0.548 to 0.884, P=0.003) displayed a negative correlation with AP. Among the 41 inflammatory cytokines, only macrophage colony-stimulating factor (M_CSF, OR=0.894, 95% CI:0.847 to 0.943, P=0.037) exhibited a negative association with AP. Sensitivity analyses revealed no evidence of pleiotropy or heterogeneity. Nevertheless, the mediation analysis showed that M_CSF did not act as a mediating factor.

Conclusion

This two-sample MR study revealed causal associations between specific GM and inflammatory cytokines with AP, respectively. However, inflammatory cytokines did not appear to act as mediating factors in the pathway from GM to AP.