A comparative study of the urinary trypsinogen-2, trypsinogen activation peptide, and the computed tomography severity index as early predictors of the severity of acute pancreatitis.

Background/aims: The aim of the study was to establish the value of urinary trypsinogen-2 in predicting the severity of acute pancreatitis (AP) and to compare it with the accuracy of the urinary trypsinogen activation peptide (TAP) and the computed tomography severity index (CTSI).
Methodology: The study population consisted of 187 consecutive patients with AP, of whom 38 had severe disease. The predictive values of urinary trypsinogen-2, TAP and CTSI were assessed within 24 h of the onset of symptoms.
Results: The mean values of predictive markers in the mild and severe pancreatitis groups were: urinary trypsinogen-2, 59/90 and 25/13 (p < 0.001); urinary TAP, 13.2 +/- 3.3nmol/l and 66.2 +/- 19.3 nmol/l (p < 0.001); and computed tomography severity index, 1.42 +/- 1.1 and 5.31 +/- 2.6 (p < 0.001). The sensitivity, specificity, positive predictive value, negative predictive value, and positive and negative likelihood ratios were calculated for the urinary trypsinogen-2 (65.7%, 66.4%, 33.3%, 88.4%, 1.9, and 0.51), for TAP (greater than 35 nmol/l: 63.2%, 65.8%, 32.0%, 87.5%, 1.9, and 0.58) and for CTSI (greater than 3: 47.4%, 95.3%, 69.2%, 87.7%, 9.0 and 0.55). To differentiate between severe and mild AP, urinary trypsinogen-2 (AUC 0.724) was slightly better than TAP (AUC 0.722), and they were both clearly better than CTSI (AUC 0.597) (p < 0.05). Urinary trypsinogen-2 had significantly lower cost (p < 0.001) than TAP and computed tomography.
Conclusion: Urinary trypsinogen-2 was superior to CTSI and was as good as or even better than urinary TAP in the early prediction of severity in AP. This suggests that this simple and quick method deserves routine clinical application.