Gut microbiota, circulating inflammatory proteins, and  cirrhosis: a multivariable Mendelian randomization study

Background The liver-gut axis is the focal point of cirrhosis research, suggesting a close association between the gut microbiota (GM) and cirrhosis. Previous studies have shown a significant correlation between cirrhosis and changes in gut microbial composition. There was a significant correlation between the severity of cirrhosis compared to healthy individuals, the displacement of specific GM, and the number of invading microorganisms. However, the causal relationship between GM and cirrhosis and whether inflammatory proteins play a mediating role remain unclear. Therefore, it is necessary to explore the specificity of specific GMs associated with cirrhosis and their underlying inflammatory mechanisms for subsequent risk prediction, treatment, and prognosis of patients with cirrhosis. Methods We identified genetic variants closely associated with GM, circulating inflammatory proteins, and cirrhosis from large-scale genome-wide association studies (GWAS) summary data and explored the causal relationship between the three and whether circulating inflammatory proteins mediate the GM-to-cirrhosis pathway using multivariate Mendelian randomization. This study used the inverse variance weighting (IVW) method and MR-Egger as the primary methods, supplemented by the weighted median estimator (WME), the Weighted model, and the Simple model. Results There were four positive and three negative results between GM and cirrhosis and five positive and five negative results between circulating inflammatory proteins and cirrhosis. In addition, Tumor necrosis factor ligand superfamily member 12 (TNFSF12) may mediate the Genus Ruminococcus torques-cirrhosis pathway.