Objectives To evaluate transcutaneous bilirubin measurement and skin bilirubin kinetics during phototherapy in late preterm infants.

Methods We enrolled in this study 313 neonates with gestational ages between 34 and 36 weeks. Skin bilirubin was measured with a transcutaneous device in covered and uncovered areas during phototherapy. Serum bilirubin was contemporarily assayed and the timing of measurement was randomly selected. Results were analysed by modified Bland–Altman test, multiple linear, non-linear and smoother regression analysis.

Results Measurement on uncovered areas (but not on covered areas) is negatively affected by the cumulative phototherapy hours (p<0.001) and always underestimates serum bilirubin. Skin bilirubin measurement on covered areas seems to agree better with serum values after ≈72 h of phototherapy. The difference between covered and uncovered areas is directly associated with pH (p = 0.01) and inversely associated with the cumulative hours of phototherapy (p = 0.037). This difference decreases over the therapy hours following a quadratic relationship (p = 0.03).

Conclusion Skin bilirubin measurement in covered areas is quite accurate to monitor jaundice. The difference with uncovered areas tends to 0 over time, whereas serum and skin bilirubin agree more strictly with the hours of phototherapy. This suggests a re-equilibration between tissue and circulating bilirubin after some hours of therapy.