Clarifying the Role of Serum Albumin in Hyperbilirubinemia Management

In September 2022, the American Academy of Pediatrics published its revised clinical practice guideline on the management of hyperbilirubinemia in neonates 35 weeks’ gestation or more.¹ Since its publication, we, as authors of the clinical practice guideline, have received many questions about the value of routinely measuring the serum albumin concentration. This issue is partially addressed in a report by Ruangkit and colleagues.² 

The 2022 clinical practice guideline includes a serum albumin concentration of less than 3.0 g/dL as a hyperbilirubinemia neurotoxicity risk factor (hereafter referred to as a neurotoxicity risk factor). This specification has created some confusion because the clinical practice guideline does not recommend routine serum albumin concentration measurement for neonates below the escalation-of-care threshold, even though up to 50% of newborns have a serum albumin concentration of less than 3.0 g/dL.^3–8 Without routine serum albumin testing at the time of bilirubin measurement, there would be misclassification of the neurotoxicity risk if the serum albumin concentration was less than 3.0 g/dL and this was the only neurotoxicity risk factor other than lower gestational age. This misclassification would lead to the use of a higher phototherapy or escalation-of-care threshold than if the serum albumin concentration was known to be less than 3.0 g/dL.

The guideline writing committee did consider recommending routine serum albumin measurement when measuring the first total serum bilirubin concentration (TSB) but decided against the recommendation. Choosing the threshold of serum albumin concentration that represents a neurotoxicity risk factor is challenging because of a lack of data linking albumin levels to health outcomes in newborns with hyperbilirubinemia. The 2004 guideline states, “It is an option to measure the serum albumin level and consider an albumin level of less than 3.0 g/dL as one risk factor for lowering the threshold for phototherapy use.”⁹ This threshold was carried forward in the 2022 guideline but without specific guidance on when to measure the albumin level other than when the TSB reaches escalation-of-care levels.

Unlike isoimmune hemolytic disease and other hemolytic conditions, a serum albumin concentration of less than 3.0 g/dL is not a hyperbilirubinemia risk factor (ie, it does not predispose to subsequent hyperbilirubinemia). Therefore, nearly all newborns with hyperbilirubinemia whose only neurotoxicity risk factor is a serum albumin concentration of less than 3.0 g/dL will have a robust response to phototherapy.

Ruangkit and colleagues conducted a retrospective review of 935 neonates examined for early (<2 weeks) neonatal jaundice in 3 hospitals in Thailand by using a rigorous jaundice assessment protocol. The protocol included daily transcutaneous bilirubin levels in the hospital with TSB levels measured when the transcutaneous bilirubin was within 3 mg/dL of the phototherapy threshold or exceeded 15 mg/dL, which is consistent with the American Academy of Pediatrics 2022 guideline. Their protocol included the measurement of serum albumin concentrations along with glucose-6-phosphate dehydrogenase deficiency screening and other tests for hemolysis in all neonates who had TSB measured as part of that examination. The authors sought to determine the prevalence of serum albumin levels less than 3 g/dL and how often these levels affected management decisions.

Of the 935 newborns who had both TSB and albumin levels measured, they found that 250 (26.7%) had a serum albumin concentration of less than 3.0 g/dL. However, they reported that these low albumin levels seldom affected phototherapy decisions because most neonates with low albumin levels either had other neurotoxicity risk factors (n = 117) or had TSB levels that did not reach the phototherapy threshold for those with neurotoxicity risk factors (n = 113). An additional 4 neonates had TSB levels high enough that they also exceeded the phototherapy threshold for those without neurotoxicity risk factors, leaving 16 neonates (1.7%) for whom the consideration of serum albumin would lead to a change in the recommendation to start (n = 15) and or continue (n = 1) phototherapy.

The authors’ careful analysis supports their conclusion that serum albumin measurements in this setting seldom affect treatment decisions. However, their 1.7% estimate of the frequency at which this occurs is probably too low both for their hospitals and elsewhere. First, they looked only at whether consideration of the serum albumin caused newborns to exceed the phototherapy threshold, but most of the phototherapy in their hospitals (70%) and other hospitals where this topic has been studied¹⁰ was initiated at levels below the recommended thresholds. It is likely that low albumin levels and resulting lower treatment thresholds contributed to at least some of these newborns receiving subthreshold phototherapy. Second, the incidence of other neurotoxicity risk factors in their study population was high. They reported sepsis in 9.9%, glucose-6-phosphate dehydrogenase deficiency in 9.9%, and clinical instability in 9.8% of their study population. In settings with a lower prevalence of these other neurotoxicity risk factors, the serum albumin concentration would be more likely to determine the neurotoxicity risk factor status and, hence, affect management recommendations.

The data provided by Ruangkit and colleagues and other studies on the implementation of the current guideline will be helpful for future updates. These updates will include greater specificity about the circumstances that should lead to serum albumin measurement and the threshold that should be considered a neurotoxicity risk factor, which probably should be lower than 3 g/dL. We know that newborns can be at higher risk for kernicterus based on having serum albumin levels much lower than 3 g/dL. Significant hypoalbuminemia can be caused by albumin loss (eg, fetal perinatal hemorrhage due to fetal-maternal transfusion, donor twin in twin-twin transfusion syndrome, twin-anemia-polycythemia sequence, anemia due to malformations of the placenta or umbilical cord, blood loss after birth, congenital nephrotic syndrome) or leakage from the intravascular space in newborns with significant illness (eg, hydrops fetalis, congenital chylothorax, gastroschisis).^4,11 Fortunately, neonates with these conditions are often readily identified or there are other signs pointing to the need to measure the serum albumin concentration. These neonates are almost always hospitalized in a neonatal intensive care unit and may have significant clinical instability, which is listed in the guideline as a neurotoxicity risk factor. Clinicians should, therefore, measure the serum albumin concentration based on clinical judgment, regardless of the TSB.

The guideline does recommend measuring the serum albumin concentration as a part of escalation of care and considering the bilirubin to albumin ratio in determining the need for exchange transfusion. The latter may be particularly relevant to neonates with conditions in which significant hypoalbuminemia is present. Additional data on the bilirubin to albumin ratio and its relation to the development of bilirubin encephalopathy¹² and neurodevelopmental outcomes in newborns with hyperbilirubinemia will further inform the ratios at which exchange transfusion is indicated. The finding of Ruangkit and colleagues that 14 neonates met criteria for exchange based on the bilirubin to albumin ratio, of whom only 1 qualified based on the TSB level, is sobering in this regard.

We are grateful for the work of Ruangkit and colleagues and the many others who critically appraise the 2022 guideline. Although hyperbilirubinemia is common, its management can be complex, and there are still important gaps in the evidence base for treatment decisions. We know that future versions of the guideline will improve through the thoughtful evaluation of current practice and the insightful questions that clinicians pose.

TSB

total serum bilirubin