A Multiple Factor Analysis Of The Clinical Data Anticipating The Result Of Endoscopic Retrograde Cholangiopancreatography

Objective: To evalute the value of the clinical routine tests in predicting positive endoscopic retrograde cholangiopancreatography (ERCP) and to define the optimal indications for ERCP, so as to avoid negative ERCP.Methods: 132 endoscopic retrograde cholangiopancreatographies performed were employed, of the total 17 outcomes were normal. Retrospective collection of historical, biochemical and ultrasonographic data was used. Univariate analysis using the chi-square test was performed to determine the factors significantly related to positive ERCP. Multivariate analysis using logistic regression with generation of the best model identifying independent predictors was employed.Results: The seven significant factors, including sex, obvious jaundice, abnormal gamma-glutamyltransferase, abnormal alkaline phosphatase, abnormal glutamic-pyruvic transaminase, abnormal glutamic-oxalacetic transaminase and elevated direct bilirubin were identified by the univariate analysis .The multivariate logistic regression analysis identified obvious jaundice and abnormal gamma-glutamyltransferase as the two significant independent factors and a statistical model was developed with a formula. A curve of receiver operating characteristics was constructed to identify a gamma-glutamyltransferase level greater than 100 U/L which would have both higher sensitivity and specificity.Conclusions: Obvious jaundice and abnormal gamma-glutamyltransferase should be taked as the major indication of ERCP ,which can avoid the negative ERCP effectively.