GLOBAL SIGNATURES OF THE MICROBIOME AND METABOLOME DURING HOSPITALIZATION OF SEPTIC PATIENTS

Background: The gut plays an important role in the development of sepsis and acts as one of the possible drivers of multiple-organ dysfunction syndrome. This study aimed to explore the dynamic alterations in the gut microbiota and its metabolites in septic patients at different stages of intensive care unit (ICU) admission. Methods: In this prospective observational study, a total of 109 fecal samples from 23 septic patients, 16 nonseptic ICU patients and 10 healthy controls were analyzed. 16S rRNA gene sequencing and ultra-performance liquid chromatography coupled to tandem mass spectrometry targeted metabolomics were used for microbiota and metabolome analysis. A prediction model combining the Sequential Organ Failure Assessment score, Klebsiella, taurocholic acid, and butyric acid was used to predict the prognosis of sepsis. Results: The diversity and dominant species of the gut microbiota of septic patients were significantly disturbed. The proportions of normal gut microbiota, such as Firmicutes on the phylum level, as well as Faecalibacterium, Subdoligranulum, Ruminococcus, Agathobacter, and Blautia on the genus level, were decreased at different stages of ICU admission, while the proportions of potential pathogenic bacteria, such as Proteobacteria on the phylum level, and Enterococcus and Stenotrophomonas on the genus level were significantly increased. In addition, the amount of short-chain fatty acids and secondary bile acids decreased in septic patients, while that of the primary bile acids increased markedly. Bacterial richness and diversity were lower in the nonsurviving patients than those in the surviving patients in the later stage of ICU admission. In the nomogram model, the higher abundance of Klebsiella, concentration of taurocholic acid, and Sequential Organ Failure Assessment score, combined with a lower butyric acid concentration, could predict a higher probability of death from sepsis. Conclusions: Our study indicated that the dynamical alterations of gut microbiota and its metabolites were associated with the prognosis of the sepsis. Based on these alterations and clinical indicators, a nomogram model to predict the prognosis of septic patients was performed.