Study on the characteristics and correlation of fecal microbiota and metabolites in patients with acute lung injury after cardiopulmonary bypass based on 16S rRNA sequencing and non-targeted metabolomics analysis.

Acute lung injury (ALI) is a severe complication following cardiopulmonary bypass (CPB), associated with high mortality and impaired patient prognosis. At present, there is no effective therapeutic strategy for ALI after CPB. Although the gut microbiota has been implicated in ALI, the biological significance of these associations remains largely elusive. A prospective, single-center, case-control design was adopted. A total of 53 post-CPB patients were enrolled, including 21 in the ALI group and 32 in the non-ALI (NALI) group. Postoperative fecal samples were collected for microbiome and metabolomic analyses, which were subsequently correlated with clinical data. Results revealed that β diversity analysis indicated distinct differences in microbial community structure (Anosim: R = 0.14, P = 0.004; Permanova: R2 = 0.058, P = 0.008). ALI patients exhibited a significant increase in the Bacillota, alongside reductions in Bacteroidota and Actinomycetota. At the genus level, Streptococcus and Enterococcus were enriched in the ALI group, while Bacteroides and Akkermansia were diminished. Metabolomics analysis identified 130 differentially expressed metabolites, 109 of which were significantly reduced in the ALI group, primarily involving amino acid metabolic pathways such as phenylalanine, tryptophan, and tyrosine. A random forest model identified genera such as Bacteroides, Corynebacterium, and Lactobacillus as having high predictive value for ALI (AUC > 0.7). Combined microbiota-metabolite analysis revealed significant correlations between specific genera and differentially expressed metabolites, suggesting a potential role for the gut-lung axis in the development of ALI following CPB. Patients with postoperative ALI following CPB exhibit marked gut microbiota structural disruption and metabolic dysfunction, both closely associated with adverse clinical outcomes. Genera such as Bacteroides and their associated metabolites may serve as early predictive biomarkers, offering novel therapeutic targets for the prevention and management of ALI.