Universal screening for newborn hyperbilirubinemia with either total serum bilirubin (TSB) or transcutaneous bilirubin (TcB) has been adopted by many countries1. Currently, TcB is most common as it is non-invasive and can be performed at the bedside. However, it is also known to be less accurate at high bilirubin concentrations and may be affected by skin pigmentation (improved with newer models) and skin thickening, making it unsuitable for older babies. Most importantly, TcB is a physiologically different parameter from the TSB because TcB assesses mainly extravascular bilirubin2 whereas TSB reflects intravascular bilirubin concentration. Although it may be debated which better predicts kernicterus, TSB is the current clinical standard for determining risk for kernicterus.

Whole blood bilirubin (TwB) analyzed on a blood gas instrument is a promising alternate method for neonatal hyperbilirubinemia screening. Compared to TSB it requires smaller sample volume, has faster turn around time, and offers concurrent measurement of a full range of analytes (blood gas, electrolytes, glucose, lactate and co-oximetry), allowing efficient comprehensive assessment of newborn status especially in critically ill neonates whose blood volume is small.

Given these potential advantages of TwB, we conducted a clinical study to compare TwB to TSB in the lab. Heel prick whole blood samples were collected in plain heparinized microtainers from healthy babies < 14 days postnatal age who had bilirubin testing as per usual care. If samples visually had sufficient volume for both tests, TwB was measured first and the rest of the blood was spun for TSB measurement (Bu/Bc on Vitros 5600, Ortho Clinical Diagnostics). TSB results were reported as usual.

We found an imperfect correlation between TwB and TSB. Many reasons may have contributed to the observed differences in results. First, although TwB is measured, the instrument reports plasma equivalent bilirubin using a calculation that relies on the measurement of tHb measurement. While this conversion facilitates comparison with the Bhutani nomogram for risk stratification, inaccurate tHb (analyzer related or not) causes deviation in plasma equivalent bilirubin results (The calculation is: plasma bilirubin = TwB/ (1-hematocrit) where hematocrit = 0.03 x tHb). Additionally, TwB was unreportable in 14% of samples because tHb was out of the manufacturer’s claimed measurable range (30-230 g/L). Second, we collected capillary samples in plain heparin tubes. Microclot formation may have affected the accuracy of the co-oximetry results if specimens were poorly mixed. Third, commercial calibrators used by field methods are known to affect the accuracy of bilirubin measurements on chemistry analyzers3. Standardization is needed for all bilirubin methods. All these factors raise challenges for design of whole blood bilirubin methods.

At this point, TwB is not ready yet for neonatal hyperbilirubinemia screening. Handling the difficult matrix of whole blood requires technological improvements to control preanalytical and analytical variables. From the sources of variability we identified, a follow up evaluation study against a reference procedure is planned to further assess the bias between TwB and TSB.


  1. Raimondi F, Ferrara T, Borrelli AC, Schettino D, Parrella C, Capasso L. Neonatal hyperbilirubinemia: a critical appraisal of current guidelines and evidence. Journal of Pediatric and neonatal individualized medicine 2012; 1 (1): 25-32.
  2. Bosschaart N, Kok JH, Newsum AM, Quweneel DM, Mentink R, van Leeuwen TG, Aalders MC. Limitations and opportunities of transcutaneous bilirubin measurements. Pediatrics. 2012 Apr; 129 (4): 689-94.
  3. Lo SF, Doumas BT. The status of bilirubin measurements in U.S. laboratories: why is accuracy elusive? Semin Perinatol. 2011 Jun; 35 (3):141-7.