ABSTRACT
BACKGROUND AND STUDY AIMS: The ability to predict severe acute pancreatitis (SAP) at an early stage is crucial for reducing the associated complications and mortality. In this study, we compared the ratio of red cell distribution width to albumin (RDW-to-ALB) using predictive scoring systems, such as the Ranson score, BISAP, and MCTSI, to develop a simple and accurate method of predicting SAP. PATIENTS AND METHODS: We included 212 patients with mild acute pancreatitis (MAP) and 89 with SAP between January 2013 and December 2018. The differences in the general characteristics and biochemical analysis as well as the various predictive scores were compared between the two groups. We evaluated the sensitivity and specificity between the RDW-to-ALB ratio, RDW, ALB, and multiple predictive scores in patients with early acute pancreatitis (AP) by using the receiver operating characteristic (ROC) curve. RESULTS: The RDW-to-ALB ratio (%) of patients with SAP was higher than that of patients with MAP (0.43 ± 0.08 vs. 0.32 ± 0.04, p < 0.001). Patients with SAP had higher Ranson, BISAP, and MCTSI scores than those with MAP. The ROC curve revealed that, when the RDW-to-ALB ratio (%) was >0.36, the sensitivity and specificity of the predicted SAP were 80.0% and 80.7%, respectively. Further statistical analysis found that the RDW-to-ALB ratio and Ranson, BISAP, and MCTSI scores were consistent in predicting SAP effectiveness (P > 0.05). CONCLUSIONS: The RDW-to-ALB ratio has a promising predictive power for SAP, and its effectiveness is comparable with those of Ranson, BISAP, and MCTSI scores.