Simple scoring system for the prediction of the prognosis of severe acute pancreatitis

Abstract

Background: In severe acute pancreatitis (SAP), it is important clinically to predict the prognosis at the time of admission. Most scoring systems for severity of acute pancreatitis consist of multiple factors and are complicated. This investigation aimed to propose a simple scoring system for the prediction of the prognosis of SAP.

Methods: Prognostic factors were evaluated by receiver operator characteristic curve analyses and multivariate analysis from data that were obtained on admission of 137 patients with SAP. A simple scoring system with 3 most useful factors was made, and its usefulness was investigated in comparison with conventional scoring systems.

Results: Three prognostic factors were selected: serum blood urea nitrogen > or = 25 mg/dL, serum lactate dehydrogenase > or = 900 IU/L, and contrast-enhanced computed tomography finding with pancreatic necrosis. On admission, 137 patients were classified from 0 to 3 by the number of positive items (simple prognostic score [SPS]). Mortality rates for patients whose SPS was 0, 1, 2, and 3 were 2% (1/42 patients), 18% (7/40 patients), 48% (12/25 patients), and 67% (20/30 patients), respectively. Furthermore, when usefulness of SPS was compared with conventional scoring systems, the area under the curve by receiver operator characteristic curve analyses in SPS was 0.83; the Ranson score was 0.83; the Japanese severity score was 0.83; the Acute Physiology and Chronic Health Evaluation II score was 0.81, and the Glasgow score was 0.75. After onset, SPS kept almost same levels from day 2 to day 6, and a significant difference was observed between survivors and nonsurvivors from day 1 to day 6.

Conclusion: This scoring system that comprised 3 items is simple, is feasible for the prediction of prognosis and conventional scoring systems, and is useful for the selection of the extremely severe patients with SAP on admission.