Predicting the Severity of Acute Pancreatitis by Rapid Measurement of Trypsinogen-2 in Urine

Abstract

Abstract Background: Early identification of patients at risk of developing a severe attack of acute pancreatitis (AP) is of great importance because rapid therapeutic interventions improve outcome. At a cutoff of 50 μg/L, trypsinogen-2 measured by a rapid urinary dipstick is a sensitive and specific diagnostic test in AP. The trypsinogen-2 concentration correlates with the severity of the disease, and a test with a higher cutoff might therefore be useful for prediction of disease severity. Methods: We increased the detection limit of the urinary trypsinogen-2 test strip (Actim Pancreatitis) from 50 μg/L to 2000 μg/L and evaluated the prognostic value of this test. The results were compared with those obtained with serum C-reactive protein and the acute physiology and chronic health evaluation II (APACHE II) score. The study population consisted of 150 consecutive patients with AP (42 with severe disease). Results: The sensitivity of the rapid urinary test strip (detection limit, 2000 μg/L) for prediction of severe AP, both on admission and at 24 h, was 62%; specificities were 87% and 85%, respectively, positive predictive values were 65% and 62%, and negative predictive values were 85% and 85%. C-Reactive protein had a sensitivity of only 38% on admission, but at 24 h, it was 83%; specificities were 90% and 70%, respectively, whereas positive predictive values were 59% and 52%, and NPVs were 79% and 91%, respectively. On admission the positive-likelihood ratio for the urinary trypsinogen-2 test strip was 4.8, and at 24 h it was 4.2; for C-reactive protein, the values were 3.7 and 2.7, respectively. Conclusions: The urinary trypsinogen-2 dipstick is a simple and rapid method for prediction of severe acute pancreatitis.