Design and validation of an early scoring system for predicting early outcomes of type III biliary atresia after Kasai’s operation

Background/purpose The purpose of this study was to determine reliable predictors of outcome of biliary atresia (BA) after Kasai’s operation. Patients and methods This study included two consecutive stages of infants with biliary atresia. First stage, 200 BA cases who underwent Kasai’s operation at our institution between May 2006 and May 2010 were retrospectively reviewed. In the first stage, we wanted to calculate a clinical scoring system by logistic regression. For cut-off determination, receiver operating characteristic (ROC) analysis was employed. Second stage (validation of scoring system), we used our scoring system to prospectively predict the outcomes of patients underwent a Kasai operation from inspecting the accuracy of our system. Results Early cholangitis, age at operation, JC time, post-operative TB, DB, AST, ALT and surgical method entered into our scoring system. The most reliable cut-offs determined by ROC analysis were 7.71 (sensitivity: 86.0 %, specificity: 98.0 %). We used our scoring system to predict the prognosis of the 15 BA patients and found that 13 of 15 patients were correctly predicted at the cut-off value of 7.71. Conclusion Our scoring system is considered to be a reliable and useful predictor of the prognosis of biliary atresia.