ABSTRACT
PURPOSE: The aim of the study was to assess and compare the efficacy of various scoring systems in predicting the severity and outcome of patients with acute pancreatitis (AP) admitted in intensive care unit (ICU). METHODS: Prospective, single institution review of 55 consecutive AP patients admitted in ICU during a 2-year period. Disease severity scores and mortality predictions were calculated using the collected data in the first 48 hours of ICU admission for Ranson and Glasgow scores and in the first 24 hours for other scores. RESULTS: Forty-two patients (76.4%) developed severe pancreatitis. Intensive care unit and 30-day mortality was 18.2% and 27.3%, respectively. Use of mechanical ventilation (MV) was an independent predictor of outcome on multivariate analysis with lack of MV being protective (adjusted odds ratio, 0.003; 95% confidence interval [CI], 0.00001-0.67; P = .04). All scoring systems had comparable accuracy in predicting severity and 30-day mortality, but sequential organ failure assessment (SOFA) score had greater efficacy with its area under curve for predicting severity and 30-day mortality being 0.81 (95% CI, 0.69-0.92) and 0.93 (95% CI, 0.85-0.99), respectively. Sensitivity and specificity (SOFA score, >4) was 76.2% and 69.2%, respectively, for predicting severity, and sensitivity and specificity (SOFA score, >8) was 86.7% and 90%, respectively, for predicting 30-day mortality. CONCLUSIONS: Use of MV is an independent predictor of outcome in AP patients admitted to ICU. Although all scoring systems had reliable accuracy in predicting severity and outcome, SOFA score performed better with additional advantages of easy applicability and timely assessment.
