Accuracy of hepatobiliary scintigraphy for differentiation of neonatal hepatitis from biliary atresia: systematic review and meta-analysis of the literature.

Hepatobiliary scintigraphy is an important diagnostic modality for work-up of neonatal cholestasis. Therefore, our objective was to evaluate the literature regarding the accuracy of hepatobiliary scintigraphy in differentiating biliary atresia from non-biliary atresia causes of cholestasis (collectively called neonatal hepatitis). Our search included Medline, SCOPUS and Google Scholar. Only studies using Tc-99 m-labeled immunodiacetic acid (IDA) derivatives were included. Overall, 81 studies were included in the meta-analysis. Pooled sensitivity and specificity were 98.7% (range 98.1-99.2%) and 70.4% (range 68.5-72.2%), respectively. Factors that increased specificity included the use of radiotracers with high hepatic extraction, administration of hepatic-inducing drugs (such as phenobarbital), use of a calculated dose/kg and administration of a booster dose in cases of non-excretion of the tracer in the bowel. SPECT imaging and duodenal fluid sampling also had high specificity; however, they need further validation because of the low number of studies. Semiquantitative imaging methods do not seem to have any incremental value. We conclude that hepatobiliary scintigraphy using IDA derivatives can be very useful for diagnostic work-up of neonatal cholestasis. To improve the specificity, several measures can be followed regarding type and dose of the radiotracer and imaging protocols. Non-imaging methods seem to be promising and warrant further validation.