Objective: 1) To evaluate the accuracy of transcutaneous bilirubin (TCB) measurement compared to total serum bilirubin level (TSB) in the pre-term infants and 2) To establish cut-off values of TCB that indicated need for TSB.
Material and Method: Premature infants whose birth weight was more than 1,000 grams and gestational age less than 36 weeks had paired TCB-TSB assessment when jaundice was observed. TCB was done using JM 103 (Minolta AirShields Jaundice Meter) on the forehead. TCB, which corresponded to TSB level that required phototherapy, was chosen as the cut-off point that indicated blood sampling for TSB.
Results: Two hundred and forty-nine paired TCB TSB measurements from 196 premature neonates were obtained. Birth weight was 1,887 + 344.4 grams. TSB ranged from 4.5-17.6 mg/dL (mean 9.4, SD 2.2mg/dL), TCB 4.1-17.7 mg/dL (mean 9.7, SD 2.4 mg/dL). The correlation coefficient between TCB and TSB was significant (r 0.79, p < 0.0001). TCB had a tendency to overestimate TSB with the mean difference of TSB- TCB = -0.3 + 1.5 mg/dL and 95% confidence interval of the mean -0.1 to -0.5 mg/dL. Of all the variables of birth weight, gestational age, and postnatal age, only postnatal age significantly influenced the correlation of TCB-TSB. In the early postnatal age of 1-4 days, the number of TCB reading overestimated TSB more than underestimated. However, at > 5 days of age, the number of underestimation was more than those of overestimation. The cut-off points of TCB that indicated the need for blood sampling for TSB were chosen as the same level of TSB of 6, 8, 10, 11, and 12 mg/dL when phototherapy was recommended. Screening with TCB would eliminate painful procedure of blood taking by 40%.
Conclusion: Noninvasive TCB assessment demonstrated significant accuracy when compared to TSB. TCB can be adopted as a screening test to identify the need for blood sampling of serum bilirubin in premature infants.

Keywords: Transcutaneous bilirubin, jaundice, hyperbilirubinemia, pre-term infants