Design and validation of a diagnostic score for biliary atresia

Abstract

The dilemma of early diagnosis of biliary Atresia (BA), particularly distinguishing it from other causes of neonatal cholestasis is challenging. The aim was to design and validate a scoring system for early discrimination of BA from other causes of neonatal cholestasis. A twelve-point scoring system was proposed according to clinical, laboratory, ultrasonographic, and histopathological parameters. A total of 135 patients with neonatal cholestasis in two sets were recruited to design (n=60) and validate (n=75) a scoring system. Parameters with significant statistical difference between BA (n=30) and non-BA (n=30) patients in the design set were analyzed by logistic regression to predict the presence or absence of BA then a scoring system was designed and validated. The total score ranged from 0 to 37.18 and a cut-off value of >23.927 could discriminate BA from other causes of neonatal cholestasis with sensitivity and specificity of 100% each. By applying this score in the validation set, the accuracy was 98.83% in predicting BA. The diagnosis of BA was proposed correctly in 100% and the diagnosis of non-BA was proposed correctly in 97.67% of patients. By applying this model, unnecessary intraoperative cholangiography would be avoided in non-BA patients. This scoring system accurately separates infants with BA and those with non-BA, rendering intraoperative cholangiography for confirming or excluding BA unnecessary in a substantial proportion of patients.