Unbound Bilirubin and Risk Assessment in the Jaundiced Newborn: Possibilities and Limitations

In an important article in this issue of Pediatrics, Wennberg et al1 review the literature concerning the physiology of bilirubin transport and brain uptake and evaluate the sensitivity and specificity of total serum bilirubin (TSB) and unbound (free) bilirubin (Bf) in assessing the risk of kernicterus. Theoretically, the level of Bf should be the most direct and accurate indicator of the risk of kernicterus because, generally, only bilirubin that is not complexed with albumin enters the brain. However, the level of Bf can change rapidly because of changes in a variety of parameters such as drug concentrations, free fatty acid concentrations, and the binding constant of bilirubin to albumin.1,2 The authors state that both in vitro and in vivo studies show that Bf correlates more accurately than TSB with bilirubin-induced effects in cultured cells and direct indicators of neurotoxicity such as electroencephalographic effects or changes in auditory brainstem-evoked responses.1 Also, Bf has been found to correlate better with kernicterus in jaundiced neonates than TSB.3,4 This is important, because measurements of TSB alone have a low specificity in identifying infants who are at risk of developing kernicterus.5The authors also state that if Bf rather than TSB were used as an indicator for aggressive interventions, including exchange transfusions, the number of such interventions would be reduced greatly. However, this statement is based on the converse assumption, that low levels of Bf indicate a low risk of toxicity, which is not necessarily correct. If low Bf values should reduce the frequency of aggressive interventions according to existing guidelines that are based on TSB values, it implies that a low Bf value should decide the level of therapeutic intervention either alone or in the presence of a high TSB. In practical terms, in a jaundiced infant admitted from home to the pediatric ward with a TSB of, for example, 560 μmol/L (33 mg/dL), the decision to reduce treatment intensity, in most instances, would have to be based on the first measurement of Bf given the possible adverse consequences of a treatment delay while awaiting the results of repeated Bf measurements.As discussed by the authors, the potential impact of Bf measurements on treatment guidelines can be discerned only through well-designed clinical trials. It should be kept in mind, however, that the concept that only the bilirubin that is not bound to albumin in the brain capillaries crosses the blood-brain barrier is a theory, not an established fact. What is certain is that the bilirubin that actually crosses an intact blood-brain barrier is not complexed with albumin. Given our lack of knowledge regarding this crucial aspect of bilirubin toxicity, increased research efforts are warranted.