The harmless acute pancreatitis score: a clinical algorithm for rapid initial stratification of nonsevere disease.

Background & Aims: Only severe acute pancreatitis requires treatment, according to the principles of intensive care medicine in an intensive care or intermediate care unit. The aim of the study was to define and evaluate a simple clinical algorithm for rapid initial identification of patients with a first attack of acute pancreatitis who do not require intensive care.

Methods: This prospective study included 394 patients who were admitted to the Municipal Clinic of Lüneburg, Germany, between 1987 and 2003. From a number of parameters of disease severity on admission, 3 parameters that showed the strongest prediction of a nonsevere course (no rebound tenderness and/or guarding, normal hematocrit level, and normal serum creatinine level) were combined to form the harmless acute pancreatitis score (HAPS). The score then was validated in a German multicenter study including 452 patients between 2004 and 2006.

Results: In both the initial and the validation set, the HAPS correlated with a nonsevere course of the disease (P < .0001). The score correctly identified a harmless course in 200 (98%) of 204 patients.

Conclusions: The HAPS enables identification, within approximately 30 minutes after admission, of patients with acute pancreatitis whose disease will run a mild course. The high level of accuracy of this test (98%) will allow physicians to identify patients quickly who do not require intensive care, and potentially those who will not require inpatient treatment at all. Thus, the HAPS may save substantial hospital costs.