A Comparison of Serum Trypsinogen-2 and Trypsin-2-1-Antitrypsin Complex With Lipase and Amylase in the Diagnosis and Assessment of Severity in the Early Phase of Acute Pancreatitis Johan Hedstro ¨m, M.D., Esko Kemppainen, M.D., Jan Anderse ´n, M.D., Hannu Jokela, Ph.D., Pauli Puolakkainen, M.D., and Ulf-Håkan Stenman, M.D. Departments of Clinical Chemistry and Otorhinolaryngology, Second Department of Surgery, University of Helsinki, Helsinki, and Department of Clinical Chemistry, University of Tampere, Tampere, Finland OBJECTIVE: The aim of the study was to compare the re- cently introduced laboratory markers trypsinogen-2 and trypsin-2-1antitrypsin complex (trypsin-2-AAT) in serum with lipase and amylase in the diagnostic and prognostic evaluation of patients with acute pancreatitis (AP). METHODS: The analytes were measured on admission in 64 consecutive patients with AP and in 30 controls with acute abdominal disease of extrapancreatic origin. Twenty-one patients had severe and 43 mild AP. As reference methods we used serum amylase and C-reactive protein. RESULTS: In subjects with AP, elevated trypsinogen-2 val- ues (90 g/L) were observed in 63 patients (98%), tryp- sin-2-AAT values (12 g/L) in 64 patients (100%), lipase values (200 U/L) in 64 patients (100%), and amylase values (300 IU/L) in 62 patients (97%). The diagnostic accuracy of the markers was evaluated by receiver operating characteristic (ROC) analysis. On admission, trypsinogen-2, trypsin-2-AAT, lipase, and amylase differentiated patients with AP from controls with high accuracy and ROC anal- yses showed similar areas under the ROC curves (AUC) for trypsinogen-2 (AUC 0.960), trypsin-2-AAT (0.948), lipase (AUC 0.947), and amylase (AUC 0.930). For differentiation between severe and mild AP, trypsin-2-AAT (AUC 0.805) was slightly better than trypsinogen-2 (AUC 0.792), and they were both clearly better than lipase (AUC 0.583), C-reactive protein (AUC 0.519), or amylase (AUC 0.632) (p  0.05). CONCLUSIONS: All the markers studied showed high accu- racy for differentiating between AP and extrapancreatic diseases. However, trypsinogen-2 and trypsin-2-AAT dis- played the best accuracy for predicting a severe AP already at admission, which makes these markers superior for clin- ical purposes. (Am J Gastroenterol 2001;96:424 – 430. © 2001 by Am. Coll. of Gastroenterology) INTRODUCTION Most cases of acute pancreatitis (AP) run a mild course that resolves spontaneously. However, about 20 –30% of the patients develop a severe form of the disease, which is associated with a variety of potentially fatal local and sys- temic complications (1, 2). Mortality in severe AP can reach 40%, especially when bacterial contamination of the pan- creatic necrosis develops (3). Rapid diagnosis and establish- ment of the disease severity facilitates prompt institution of early intensive therapy that is likely to be beneficial (2) and is a prerequisite for evaluation of new early phase therapies (4). The diagnosis of AP and its differentiation from other abdominal diseases depend on both careful clinical assess- ment and the use of supportive laboratory techniques. De- termination of serum amylase and lipase levels are the most widely used tests to support the diagnosis of AP (1). How- ever, their sensitivity and specificity are considered unsat- isfactory and the need for more sensitive tests has been pointed out (5). Trypsinogen is a 25-kD pancreatic proteinase produced in the exocrine pancreas. The two main isoenzymes, trypsino- gen-1 (cationic) and trypsinogen-2 (anionic), are secreted at high concentrations into pancreatic fluid and are normally activated in the duodenum by enterokinase to active trypsins (6). However, a small proportion escapes into the circulation where trypsinogen remains free but active trypsin is rapidly inactivated by complexation with  2 -macroglobulin and  1 - antitrypsin (AAT) (7) (Fig. 1). Intrapancreatic activation of trypsin is believed to play an essential role in the development of especially the necrotiz- ing form of AP (7). Markedly increased levels of trypsino- gen-2 and trypsin-2-AAT are associated with severe disease (8). Recently, evidence supporting a crucial role of trypsino- gen activation in the pathogenesis of AP has been provided by the finding that hereditary pancreatitis may be caused by mutations in the trypsinogen-1 gene (9). Because of the pivotal role of trypsin activation in the pathophysiology of AP assays based on this phenomenon may be expected to be especially useful in the diagnosis of AP. Trypsinogen-2 and trypsin-2-AAT, which are recently introduced laboratory THE AMERICAN JOURNAL OF GASTROENTEROLOGY Vol. 96, No. 2, 2001 © 2001 by Am. Coll. of Gastroenterology ISSN 0002-9270/01/$20.00 Published by Elsevier Science Inc. PII S0002-9270(00)02250-4