Hepatic technetium-99m-mebrofenin iminodiacetate (99mTc-mebrofenin IDA) scans and serum gamma-glutamyl transpeptidase (GGTP) have high sensitivity for extrahepatic biliary atresia (EHBA). This study was based on the hypothesis that the interpretation of results of 99mTc-mebrofenin IDA scans and serum GGTP levels in series would result in a reduction of the false positivity observed with these tests individually. The aetiology of neonatal cholestasis in 132 study patients was: 25% (33/132) EHBA, 45.5% (60/132) neonatal hepatitis (NH) with an identifiable cause and 19.7% (26/132) idiopathic NH. Of the various clinical, biochemical and imaging parameters that were significantly different between patient groups, sensitivity for EHBA was: serum GGTP > or = 150 IU l(-1) (100%), 99mTc-mebrofenin IDA scans (100%), pale stools (82.8%) and total serum bilirubin > or = 12 mg dl(-1) (66%). However, specificity ranged from 48.5 to 79%. Of the 63 patients who had non-excreting IDA scans, operative cholangiograms could be avoided on the basis of a specific aetiological diagnosis of NH, made concurrently, in only 9 infants. The rest (54) underwent operative cholangiograms; 21 (39%) of these had patent biliary trees and therefore underwent the procedure unnecessarily. If serum GGTP (< 150 IU l(-1)) had been used as a screen after IDA scanning in these 54 patients, operative cholangiograms could have been avoided in another 12 patients and thereafter only 9/42 (21%) of the operative cholangiograms would have been considered unnecessary.
Conclusion: A diagnostic algorithm is proposed wherein serum GGTP level (at a cut-off level that maintains 100% sensitivity for EHBA) is used in series with non-excreting 99mTc-mebrofenin IDA scans (for patients with no specific aetiological label). This strategy reduces the false positivity of individual tests.