[Assessment of severity of acute pancreatitis in early stage]

To assess the accuracy of indexes for predicting severe acute pancreatitis (SAP).

Thirty-nine patients suffering from acute pancreatitis (AP) were randomly selected, including 20 SAP cases and 19 mild acute pancreatitis (MAP) cases. The levels of polymorphonuclear leucocyte-elastase (PMN-E), serum phospholipase A2 (PLA2), pancreatic PLA2 (Pan-PLA2), PLA2 catalytic activity (CA-PLA2), amylase, as well as C-reactive protein (CRP) were determined Acute Physiology and chronic health evaluation II (APACHE II) was scored in every patient. Sensitivity and specificity of all the parameters were assessed, and receiver operator characteristic curve (ROC) was plotted. Positive predictive value, negative predictive value, and overall accuracy were then analyzed.

PMN-E, CRP and CA-PLA2 were obviously higher in SAP than in MAP, and were indicative of the severity of the disease (all P < 0.01). Pan-PLA2 and amylase of AP patients raised at the onset of the disease, and they showed no difference between the SAP groups and MAP groups. When SAP was predicted by PMN-E, sensitivity was 94.5%, specificity was 99.4%, positive predictive value was 97.8%, negative predictive value was 99.4%, overall accuracy was 98.7%, higher than other indexes. When SAP was Predicted by CRP, the overall accuracy was also high and reached 84.0%.

PMN-E, CA-PLA2, CRP, and APACHE II are all indexes for the diagnosis of SAP. PMN-E is found to be the best index in predicting SAP.