Defining the limitations of transcutaneous bilirubin measurement in late preterm newborns

The Accuracy of Transcutaneous Bilirubin as a Screening Test in Preterm Infants

OBJECTIVE

To determine the accuracy of transcutaneous bilirubin (TcB) to predict total serum bilirubin (TSB) in preterm infants across gestational age (GA) ranges and to calculate the cost-effectiveness of TcB as the primary screening test of choice for neonatal jaundice in neonatal intensive care unit (NICU) settings.

METHODS

Single-center retrospective study of infants aged ≤ seven days admitted to the NICU over a six-month period with a paired TSB and TcB, with or without phototherapy as part of their routine clinical care. Infants were divided into GA-specific groups as term, late preterm, moderate preterm, and very preterm. Measurement bias (bias=TSB-TcB) was calculated on the paired TSB and TcB values, and a Bland-Altman analysis was carried out. The impacts of additional infant-specific variables on the bias were assessed with univariate and multivariate linear regression techniques. The potential direct cost savings associated with the use of TcB as the primary screening test were calculated.

RESULTS

A total of 263 paired TSB and TcB samples from 95 patients were included (130 paired samples from term (n=60), 75 from late preterm (n=21), 27 from moderate preterm (n=7), and 31 from very preterm (n=7)). The mean paired measurement bias across all the GA groups was -0.9 ± 2.9 mg/dL. The sensitivity and specificity of TcB in GA < 35 weeks were 92% and 62%, respectively. A conservative estimate of a one-third reduction in TSB measurement by using TcB as the primary screening test will have a direct cost saving of $3,148 over a six-month period.

CONCLUSION

Our data suggest that TcB is a safe and potentially cost-effective screening test for jaundice across GA groups.

Refining interpretation of transcutaneous bilirubin measurement in newborns born late preterm

BACKGROUND

Transcutaneous bilirubin (TCB) monitoring is widely used for jaundice screening in the newborn period. Limited data exists on adjusting TCB for bias in late preterm infants. The objective of this study was to determine the median bias between transcutaneous bilirubin and total serum bilirubin levels in newborns born at 35-36 weeks' gestation.

METHODS

This was a retrospective cohort study of late preterm infants born at 35-0/7 to 36-6/7 weeks' gestation who were admitted to a level III neonatal intensive care unit from May 2018 to February 2020. Transcutaneous and total serum bilirubin levels were assessed within 2 h of each other during the first 60 h of life. Bland-Altman plots were used to evaluate transcutaneous bilirubin bias. Bilirubin risk stratification based on age (in hours) was done using an adaptation of the Bhutani nomogram for transcutaneous, adjusted transcutaneous, and total serum bilirubin measurements.

RESULTS

The median bias between transcutaneous and total serum bilirubin bias was 2.4 mg/dL (IQR 1.7-3.4, 95% CI 2.2-2.7). The kappa statistic demonstrated slight agreement between the unadjusted transcutaneous bilirubin and total serum bilirubin (k = 0.033, p = 0.194. The kappa statistic demonstrated fair agreement between an adjusted transcutaneous bilirubin (subtract 1 mg/dL) and total serum bilirubin (k = 0.298, p < 0.0001) and moderate agreement between another adjusted transcutaneous bilirubin (subtract 2 mg/dL) and total serum bilirubin (k = 0.430, p < 0.0001).

CONCLUSION

In a single center study of late preterm infants, transcutaneous bilirubin systematically overestimated the total serum bilirubin level. Subtracting 1 mg/dL from the transcutaneous bilirubin identified infants with total serum bilirubin levels in the high or high intermediate risk range. Adjusting the transcutaneous bilirubin prior to risk stratification may reduce unnecessary blood draws for total serum bilirubin. Studies of racially and ethnically diverse newborns using various transcutaneous bilirubin meters are needed prior to broad application of the adjusted transcutaneous bilirubin approach.