Non-Invasive Diagnostic Methods for Differentiation of Biliary Atresia from Neonatal Hepatitis in Upper Egypt

Authors

1 Department of Pediatrics, Faculty of Medicine, Sohag University, Egypt.

2 Nuclear Medicine Department, Faculty of Medicine, Sohag University, Egypt.

3 Department of Pathology, Faculty of Medicine, Sohag University, Egypt.

4 Department of Pathology, Faculty of Medicine, Cairo University, Egypt.

Abstract

Background
Cholestatic jaundice in infancy is always pathologic and mainly caused by biliary atresia (BA), and neonatal hepatitis (NH). The early discrimination of both conditions is critical for the outcome of BA. We aimed to assess different non-invasive diagnostic tools in differentiating BA from NH.
Materials and Methods
Forty infants (25 boys, 15 girls) with cholestatic jaundice and final diagnosis of BA (n=17), and NH (n=23) were studied retrospectively from January 2015 to December 2017. All patients were subjected to thorough history and complete physical examination. Liver function tests, abdominal ultrasonography, hepatobiliary scintigraphy using (hepatobiliary iminodiacetic acid [HIDA]), and percutaneous liver biopsy were performed for all patients. Finally the accuracy of HIDA scan and liver function tests for differentiating BA from NH, in comparison with histopathological diagnosis, was evaluated.
Results
Acholic stool, absence of gall bladder visualization by ultrasonography and high level of serum gamma-glutamyl transpeptidase (GGT), and positive HIDA scan findings were strong indicators of BA. The accuracy of GGT > 250 IU/L for diagnosis of BA was 92.7 % and that of positive HIDA scan findings was 82.7 %. The diagnostic accuracy for using both parameters was 98.1% (95% CI: 94.9 - 100.0%, p<0.001).
Conclusion
According to the results, BA can be differentiated from NH by non-invasive methods as presence of acholic stool, absent gall bladder by ultrasonography, elevated GGT, positive HIDA scan findings. GGT > 250 IU/L and positive HIDA scan had high accuracy to differentiate BA from NH and combination of both parameters has increased the accuracy to 98.1%.

Keywords