OBJECTIVE

We evaluated the diagnostic value of changes in transcutaneous bilirubin (TcB) levels for prediction of subsequent hyperbilirubinemia in healthy term and late-preterm neonates.

METHODS

Neonates at 35 weeks of gestation were enrolled in a prospective study. Two TcB determinations were performed for all enrolled neonates (N = 358). The first assessment (TcB(1)) was performed at 24 +/- 6 hours of age, followed by a second (TcB(2)) >or=12 hours later. Changes in TcB levels were calculated. TcB values were plotted on an hour-specific serum bilirubin nomogram, and risk zones were recorded. Of the 358 neonates enrolled, 325 neonates (91%) were monitored for hyperbilirubinemia until 5 days of age.

RESULTS

The mean ages of TcB(1) and TcB(2) estimations were 23 +/- 4 hours and 42 +/- 4 hours, respectively. A total of 14.9% of neonates (48 of 325 neonates) developed hyperbilirubinemia by 5 days of age. The sensitivity, specificity, and positive and negative likelihood ratios for prediction of subsequent hyperbilirubinemia for TcB(1) (zone >2, >75th percentile) were 80.4%, 58.0%, 1.9, and 0.34; those for TcB(2) (zone >2, >75th percentile) were 82.6%, 79.0%, 4.0, and 0.22; and those for the change in TcB levels (>0.18 mg/dL per hour, >75th percentile) were 82.5%, 82.9%, 4.8, and 0.21, respectively. Gestational age, TcB risk zone, and change in TcB levels were found to be independent predictors of subsequent hyperbilirubinemia.

CONCLUSIONS

Single TcB measurements at 30 to 48 hours predict hyperbilirubinemia with a reasonably high degree of accuracy. Changes in TcB levels do not offer any added clinical benefit.