Characteristics of metagenome and metabolomics of biliary flora in biliary acute pancreatitis for severity prediction.

Abstract

Objective: To investigate the characteristics of the metagenome and metabolomics of biliary flora in patients with acute biliary pancreatitis and its predictive value for the severity of the disease.

Methods: This study included 80 ABP patients (Jan 2022 - Dec 2024) and 30 healthy controls. ABP patients were divided into mild (n = 50) and moderate-severe (n = 30) groups using the revised Atlanta criteria. Chao1 index, Shannon index, relative abundances of Klebsiella, Enterococcus, and Escherichia coli, serum amylase, and C-reactive protein (CRP) were compared. ROC analysis assessed predictive value of Shannon index, Klebsiella abundance, and CRP for severity. Correlations between these parameters and severity were examined.

Results: Significant differences were observed among the three groups in Chao1 and Shannon indices, relative abundances of the three bacterial genera, and serum amylase and CRP levels (all P < 0.05). ROC analysis showed that Shannon index, Klebsiella relative abundance, and CRP level predicted disease severity with AUCs of 0.804, 0.804, and 0.850, respectively (P < 0.05). Shannon index was negatively correlated with ABP severity, while Klebsiella abundance and CRP level were positively correlated with severity (P < 0.05).

Conclusion: In this cohort, biliary microbiome characteristics and serum inflammatory markers were associated with the severity of biliary acute pancreatitis. The Shannon index, Klebsiella relative abundance, and CRP showed exploratory discriminatory performance. These findings provide a basis for further prospective validation; they do not establish a causal driving factor, a clinically validated early-warning system, or an individualized treatment strategy.