Prediction of severe hyperbilirubinaemia using the Bilicheck transcutaneous bilirubinometer

OBJECTIVES: To determine the sensitivity and specificity of different levels of bilirubin measured by the transcutaneous bilirubinometer Bilicheck on forehead and sternum for predicting severe hyperbilirubinaemia of total serum bilirubin (TSB)>or=300 micromol/L in Malay, Chinese and Indian infants. DESIGN: A prospective observational study. SETTING: A tertiary care University hospital. METHODS: A total of 345 healthy jaundiced term infants were recruited prior to commencement of phototherapy or exchange transfusion. Transcutaneous bilirubin (TcB) level was measured with the Bilicheck from infants' foreheads (TcBh) and sternums (TcBs) within 30 min of serum bilirubin measurement by the diazo method in the hospital laboratory. RESULTS: The median serum TSB level of these infants was 233.0 micromol/L (range: 108.0-589.0). Ninety-five (27.5%) infants had TSB>or=300 micromol/L. There was good correlation between log10TSB and TcB measured from the forehead (r=0.80, P<0.0001) and the sternum (r=0.86, P<0.0001). At TcBh cut-off of 250 micromol/L, the Bilicheck detected TSB>or=300 micromol/L with a sensitivity of 100% and a specificity of 39.2%, the area under the receiver operative characteristic curve being 0.89 (95% confidence interval 0.85, 0.92). At TcBs cut-off of 200 micromol/L, the Bilicheck detected TSB>or=300 micromol/L with a sensitivity of 100% and a specificity of 33.6%, the area under receiver operative characteristic curve being 0.93 (95% confidence interval 0.90, 0.96). CONCLUSION: The Bilicheck is not a substitute for measuring serum bilirubin. However, using predetermined TcB cut-off values with reasonable sensitivity and specificity, it is a useful screening tool to identify infants with TSB>or=300 micromol/L requiring blood sampling, hospital admission and treatment.