Objectives: To evaluate a community-based screening programme for detecting neonatal liver disease by the quantitation of conjugated bilirubin in blood.
Settings and methods: Prospective cohort/observational study using spare plasma from routinely collected liquid neonatal screening specimens from babies born in Birmingham over a two-year period. Babies with a conjugated bilirubin above 18 mumol/l and comprising more than 20% of the total bilirubin were followed up. A total of 27654 neonates were tested in the community, with a further 2425 samples from babies hospitalised at the time of the test.
Results: In the community-based series, 84.7% of the specimens received were analysed, the remainder being unusable mainly because of gross haemolysis (8.6%) or insufficient sample (5.8%). In 107 neonates the results were above the cut-off limits (0.46% of the number analysed). Of these, 12 had persistently abnormal results, 11 of whom had confirmed liver disease. The liver diseases detected included neonatal hepatitis (n=6), extra-hepatic biliary atresia (n=2), hypopituitarism (n=1), alpha-1-antitrypsin deficiency (n=1) and Alagille syndrome (n=1). The sensitivity and specificity of the test for babies in the community were 100% and 99.6%, respectively.
Conclusions: Conjugated bilirubin in plasma measured at 6-10 days is a reliable marker for neonatal liver disease, and a population screening programme based on this method has the potential to improve the survival and quality of life of infants born with liver disease. However, testing as part of the neonatal screening programme will prove practical only if the method can be adapted to use dried blood spots.