BACKGROUND: Identification of patients at risk for severe disease early in the course of acute pancreatitis (AP) is essential to optimize management and to improve outcomes. AIM: To assess BISAP score as a predictor of severity of AP. PATIENTS AND METHODS: Retrospective review of AP patients between January 2009 and December 2010. BISAP, APACHE II and Balthazar scores were calculated. Length of stay, local complications, organ failure and mortality were registered. Accuracy of the scoring system for predicting severity was measured by the area under the receiver operating curve (AUC). RESULTS: The medical records of 128 patients, median age 46.5 years (55.5% men), were reviewed. Mean hospital stay was 15 days, 18 patients (14%) had local complications, 7 patients (5.4%) developed organ failure and 2 patients died (1.6%). The AUC for BISAP score to detect organ failure was 0.977 (95% IC 0.947-1.000). A BISAP score > 3 had a sensitivity, specificity, positive and negative predictive value of 71.4, 99.1, 83.3 and 98.3% respectively. An APACHE II score > 8 had a sensitivity and specificity of 71.5 and 86.8% respectively. The figures for a Balthazar score > 6 were 42.8 and 98.3% respectively. There was a significant correlation between BISAP score and length of hospital stay. CONCLUSIONS: BISAP score was a useful method for predicting the severity of PA, with the advantage of being simple and based on parameters obtained on the first day of hospitalization. Its sensitivity and specificity were superior to APACHE II and Balthazar score in this cohort.
BACKGROUND: Identification of patients at risk for severe disease early in the course of acute pancreatitis (AP) is essential to optimize management and to improve outcomes. AIM: To assess BISAP score as a predictor of severity of AP. PATIENTS AND METHODS: Retrospective review of AP patients between January 2009 and December 2010. BISAP, APACHE II and Balthazar scores were calculated. Length of stay, local complications, organ failure and mortality were registered. Accuracy of the scoring system for predicting severity was measured by the area under the receiver operating curve (AUC). RESULTS: The medical records of 128 patients, median age 46.5 years (55.5% men), were reviewed. Mean hospital stay was 15 days, 18 patients (14%) had local complications, 7 patients (5.4%) developed organ failure and 2 patients died (1.6%). The AUC for BISAP score to detect organ failure was 0.977 (95% IC 0.947-1.000). A BISAP score > 3 had a sensitivity, specificity, positive and negative predictive value of 71.4, 99.1, 83.3 and 98.3% respectively. An APACHE II score > 8 had a sensitivity and specificity of 71.5 and 86.8% respectively. The figures for a Balthazar score > 6 were 42.8 and 98.3% respectively. There was a significant correlation between BISAP score and length of hospital stay. CONCLUSIONS: BISAP score was a useful method for predicting the severity of PA, with the advantage of being simple and based on parameters obtained on the first day of hospitalization. Its sensitivity and specificity were superior to APACHE II and Balthazar score in this cohort.