Purpose: No consensus has been reached to define gastrointestinal failure (GIF) associated with severe acute pancreatitis (SAP). Reintam and colleagues proposed a scoring system of GIF for critically ill patients, but its suitability for patients with SAP is questionable. The present study evaluates a modified GIF score we developed to assess the GIF of patients with SAP.
Methods: The subjects of this study were 52 patients with SAP treated between September 2010 and July 2011. We recorded the Reintam's GIF score, our modified GIF score, the acute physiology and chronic health evaluation (APACHE) II score, the sequential organ failure assessment (SOFA) score, and other clinical values during the first 3 days after admission. The prognostic value of the modified GIF score, for evaluating the severity and outcomes of SAP, was also assessed.
Results: Compared with the Reintam's GIF score, the modified GIF score seemed to be more valuable for predicting hospital mortality (the area under curve, AUC 0.915 vs. 0.850), multiple organ dysfunction syndrome (MODS) (AUC 0.829 vs. 0.766), and pancreatic infection (AUC 0.796 vs. 0.776). Moreover, combining the modified GIF score and the SOFA or APACHEII scores resulted in more accurate prediction of the prognosis of SAP than either score alone.
Conclusion: The modified GIF score is useful for assessing gastrointestinal system function, which may serve as an early prognostic tool to evaluate the severity and predict the outcomes of SAP.