Neonatal cholestasis: differentiation of biliary atresia from neonatal hepatitis in a developing country

Abstract

Aim: To study the accuracy of various clinical and investigational parameters to differentiate biliary atresia (BA) from neonatal hepatitis (NH).

Methods: It was a prospective study, conducted in a tertiary care hospital. A total 101 infants with neonatal cholestasis (NCS) were included in this study. Following a baseline hepatobiliary scintigraphic study (HBS), it was repeated after giving UDCA (40 mg/kg/day for 48-72 h). The sensitivity and specificity of clinical and investigational parameters were calculated with peroperative cholangiogram as gold standard.

Results: The mean age was 2.8 +/- 1.7 months and 82 were male. Of these, 35 were diagnosed to have BA and 66 had NH (idiopathic 25, sepsis/UTI 20, galactosaemia 11, TORCH 2 and others 8). Persistently clay stool was found to have modest accuracy (79%) and the accuracy of HBS improved significantly following UDCA therapy (91% from 77%, p < 0.01) whereas liver biopsy was 100% accurate in differentiating BA from NH. The outcome of BA cases with ductal plate malformation (DPM) was worse.

Conclusion: One-third of all NCS in India is due to BA and among the intrahepatic causes acquired infection and galactosaemia are common. Liver biopsy is the best method to differentiate NH from BA.