Background: The effect of gut microbiota on enteral nutrition tolerance in critically ill patients is unclear. Methods: Non-abdominal sepsis patients in an ICU, sorted by whether they reached 20 Kcal/kg/day on the 3rd day of EN, were divided into tolerance and intolerance groups. Their feces on day 1 and day 3 of EN initiation were collected for 16s rDNA and short-chain fatty acid (SCFA) testing. Results: There were 14 patients included in the tolerance group and 10 in the intolerance group. On EN day 1, the OTUs and microbiota diversity were higher in the tolerance group than in the intolerance group. The ratio of Firmicutes to Bacteroidetes was higher in the intolerance group on EN day 1. The genus Parabacteroides were the most significantly elevated in the tolerance group. On EN day 3, the genus Escherichia-Shigella was the most significantly elevated in the tolerance group. On EN day 3, the levels of SCFA decreased more significantly in the intolerance group. Conclusion: Enteral nutrition tolerance is associated with microbiota features and short-chain fatty acid levels. A higher ratio of Firmicutes to Bacteroidetes and microbiota diversity on EN day 1 may help in the early prediction of EN tolerance.

Sepsis has heterogeneous clinical trajectories, but conventional severity scores offer only static risk estimates. Timely, dynamic prediction could enable personalized intervention. In this multicenter retrospective study of 47,936 ICU patients meeting Sepsis-3 criteria from one institutional and two public datasets (MIMIC-III, eICU; sensitivity in MIMIC-IV), group-based trajectory modeling identified latent recovery patterns. An ensemble machine-learning model incorporating dynamic physiological variability was trained, temporally validated, and externally tested; clinical impact was assessed following implementation. Three trajectories emerged: rapid recovery (41.5%), slow recovery (36.4%), and clinical deterioration (22.1%). In the final binary classification task, AUROC was 0.92 (development), 0.89 (internal), 0.84 (MIMIC-III) and 0.77 (eICU); median warning time before deterioration was 17.6 h (Overall pooled across all cohorts). Reduced heart rate variability (SD < 10 bpm) predicted mortality (adjusted HR 2.17). Implementation reduced ICU stay by 1.8 days, machanical ventilation by 2.3 days, and 28-day mortality by 5.7%. This externally validated trajectory-based model offers accurate, early risk stratification for sepsis, supporting proactive, individualized critical care.