A Comparative Evaluation Of Three Scoring Systems In Early Prediction Of Severity In Acute Pancreatitis

ObjectivesThis study compares the evaluation of the bedside index for severity in acute pancreatitis (BISAP), acute physiology and chronic health evaluation II (APACHE II) and computed tomography severity index (CTSI) three scoring systems in early prediction of severity in acute pancreatitis (AP). By discussing the relationship of BISAP scores among different etiological factors and advantages and disadvantages of BISAP, APACHE II and CTSI scoring systems in early prediction of severity in AP, it tried to provide evidence for clinical application of BISAP in early prediction of severity in AP and guide treatment of AP patients.MethodsThis study retrospectively analysis data from182consecutive AP cases in general surgery department of Xiangya Hospital of Central South University collected during May2010and September2011. Scores for all the AP patients are computed using BISAP, APACHE II and CTSI methods. Analysis is performed for the relationship between BISAP scores and AP caused by different etiological factors. By analyzing the receiver operating characteristic curve (ROC), the abilities to predict the severity of AP are compared among the three scoring systems.Results Of182AP cases, there are67diagnosed SAP,54with organ failures,60appear local complications. Distributions of BIS AP scores conditioned on different AP etiological factors are statistically the same (P>0.05). The differences among three types of scores are statistically significant across MAP and SAP, with/without organ failures, with/without local complications. The area under receiver operating characteristic curve (AUC) for BISAP, APACHEⅡ and CTSI are0.847,0.889and0.886respectively in SAP prediction (pairwise differences insignificant);0.859,0.861and0.860respectively in organ failures prediction (pairwise differences insignificant);0.794,0.763and0.962respectively in local complications prediction (CTSI significantly better). When BISAP score≥2, sensitivities in predicting SAP, organ failures and local complications are80.6%,85,2%,76.7%respectively, CTSI has higher sensitivity for local complications prediction when its score≥3, but lower sensitivity for organ failure prediction when its score≥4.ConclusionsBISAP scoring system shows no significant difference among different etiological factors, BISAP scores differs in MAP vs. SAP, with/without organ failure and with/without local complications and BISAP has certain applicability in early prediction of severity in AP. All three scoring systems have good performance in predicting SAP and APACHE II seems the best. CTSI is less powerful in predicting organ failures, but much more powerful in predicting local complications than BISAP and APACHE Ⅱ.