Gastrointestinal (GI) failure can be both a cause of sepsis and a consequence of the systemic pro-inflammatory response in sepsis. Changes in biomarkers of enterocyte damage, citrulline and I-FABP (intestinal fatty acid binding protein), may indicate altered intestinal permeability and damage. The study group consisted of patients with sepsis (N = 28) and septic shock (N = 30); the control group included patients without infection (N = 10). Blood samples were collected for citrulline and I-FABP and a 4-point AGI score (acute GI injury score) was calculated to monitor GI function on days 1, 3, 5, 7, and 10. Citrulline concentrations in the study group were lower than in the control. Lower values were also noted in septic patients with shock when compared to the non-shock group throughout the study period. I-FABP was higher in the septic shock group than in the sepsis group only on days 1 and 3. Citrulline was lower in patients with GI failure (AGI III) when compared to AGI I/II, reaching significance on days 7 (p = 0.034) and 10 (p = 0.015); moreover, a higher AGI score was associated with an increased 28 day mortality (p = 0.038). The results indicate that citrulline measurements, along with the AGI assessment, have clinical potential in monitoring GI function and integrity in sepsis.
Fibronectin (FN) plays an essential role in the host’s response to infection. In previous studies, a significant decrease in the FN level was observed in sepsis; however, it has not been clearly elucidated how this parameter affects the patient’s survival. To better understand the relationship between FN and survival, we utilized innovative approaches from the field of explainable machine learning, including local explanations (Break Down, Shapley Additive Values, Ceteris Paribus), to understand the contribution of FN to predicting individual patient survival. The methodology provides new opportunities to personalize informative predictions for patients. The results showed that the most important indicators for predicting survival in sepsis were INR, FN, age, and the APACHE II score. ROC curve analysis showed that the model’s successful classification rate was 0.92, its sensitivity was 0.92, its positive predictive value was 0.76, and its accuracy was 0.79. To illustrate these possibilities, we have developed and shared a web-based risk calculator for exploring individual patient risk. The web application can be continuously updated with new data in order to further improve the model.