Kernicterus Workup

Hematologic studies

Hematologic laboratory evaluation is the cornerstone of evaluation of the baby with hyperbilirubinemia. Although jaundice can be appreciated clinically, observation alone is not a reliable method to assess the severity or estimate risk factors for the infant.

Total and direct bilirubin

Quantitative measurement of total and direct bilirubin levels should be undertaken in every baby at risk for significant hyperbilirubinemia or kernicterus.

Total bilirubin measures the aggregate of all forms of bilirubin in the serum. The direct fraction measures the amount of conjugated bilirubin. Subtraction of the direct fraction from the total yields the calculated indirect bilirubin, or the unconjugated form. Remember that the indirect fraction is composed of bound bilirubin, free bilirubin, and lumirubin, as well as many other clinically unidentifiable isomers if the baby is receiving phototherapy.^[20] Only the free bilirubin is available to cross the blood-brain barrier and has the potential to cause neurotoxicity; if and how the presence of other isomers modulates this process is unclear.

Attempts to measure the amount of bound albumin or to estimate the bound fraction from measures of serum albumin have not proven to be clinically useful, although there has been some renewed interest in the clinical use of this tool.^{[21, 22, 18]}

Serial measurements may be necessary to track the evolution of hyperbilirubinemia; frequency of measurements depends on the baby's gestational age, chronologic age, risk factors, and other clinical characteristics.

In infants who were reported to the US kernicterus registry, total serum bilirubin levels ranged from 20.7-59.9 mg/dL at the time of presentation with the classic physical signs of kernicterus.^[6] A similar series of 249 infants in Cairo reported by Gamaleldin et al presented with total serum bilirubin levels ranging from 25-76.4 mg/dL; at presentation, 44 infants (18%) had moderate or severe acute bilirubin encephalopathy.^[14]

Every baby with hyperbilirubinemia should have a direct fraction measured at least once to rule out direct hyperbilirubinemia. Direct hyperbilirubinemia in the neonate is defined as a direct fraction of more than 2 mg/dL or more than one third of the total bilirubin concentration and is always pathologic. Subsequently, if the hyperbilirubinemia is established as the indirect type, obtaining a direct fraction with every measurement is unnecessary unless the hyperbilirubinemia develops after the expected time frame for typical neonatal hyperbilirubinemia.

With the advent of early discharge (before the physiologic peak of serum bilirubin) some clinicians are advocating universal bilirubin measurements in all babies prior to discharge. Nomograms have been published that estimate a baby's risk of disease based on measured levels of bilirubin. One nomogram that assessed the risk of critical hyperbilirubinemia in babies leaving the hospital within 24-48 hours of birth adjusted for gestational age and postnatal age in hours (see following image).^[23] In 2011, Yu et al published a similar nomogram using transcutaneous measurements in healthy term and late-preterm Chinese infants.^[24] However, after reviewing available experimental and observational studies that included comparison groups, the US Preventive Services Task Force concluded there was insufficient evidence to assess the balance of benefits and harms of universal screening for hyperbilirubinemia to prevent bilirubin encephalopathy.^[25]

Blood type

The baby's blood type should be determined and compared with that of the mother. Mothers with type O blood may have circulating antibodies to other red cell antigens that can cross the placenta and cause hemolytic disease in a baby with a different blood type, such as blood type A or B. Similarly, mothers who are Rh negative may have antibody to the Rh antigen if they have not been treated with RhoGAM. Antibody to the Rh antigen causes the most fulminant type of hemolytic hyperbilirubinemia, termed erythroblastosis fetalis in its most severe form. ABO incompatibility can cause significant hemolysis as well. Minor antigens on the baby's RBC are also susceptible to immune-mediated hemolysis from maternally acquired antibody but usually to a lesser extent than the major antigens. Documentation of maternal antibody status during the pregnancy should alert the caretaker about potential risk for hemolytic disease and hyperbilirubinemia.

Reticulocyte count

Babies typically have reticulocyte counts higher than older infants and adults. However, significant elevation in the neonate's reticulocyte count (>7 mg/dL) can indicate the presence of an ongoing hemolytic process.

Direct Coombs test

This test assays for antibody on the RBC membrane. A positive result indicates that antibodies are attached to the RBC, placing it at risk for immune-mediated destruction. This is a qualitative test, so a positive result does not suggest the amount of antibody or the degree of hemolysis. However, pairing these results with the reticulocyte count can provide some idea of the severity of the process. This test, although reliable, does not have 100% sensitivity. Because false-negative results do occur, repeating a test with an initial negative result is not unreasonable if the clinical course supports an ongoing hemolytic process. In a many cases of ABO incompatibility, direct Coombs test findings may be negative. Therefore, an elution test should be performed to demonstrate anti-A or anti-B antibodies in the serum.

CBC count

A CBC with manual differential should always be included in the evaluation of a newborn with jaundice. Measurement of the hemoglobin and hematocrit can be helpful to determine if ongoing hemolysis severe enough to cause anemia is present. The peripheral smear inspection is particularly valuable because it may reveal large amounts of nucleated RBCs, suggesting active reticulocytosis; it may reveal abnormally shaped RBCs in the case of hereditary membrane defects such as spherocytosis and elliptocytosis or marked ovalocytosis in the case of hemolytic disease of the newborn. Babies with sepsis can develop hyperbilirubinemia, and, although not conclusive, normal total WBC count and manual differential can be reassuring in a healthy-appearing baby with hyperbilirubinemia.

Serum electrolytes

Breastfed babies are known to normally develop higher levels of serum bilirubin than their formula-fed counterparts. However, with the trend toward earlier discharge, most breastfed babies are being discharged home before breastfeeding is well established, and a concomitant increase in the number of infants readmitted to the hospital in the first week of life with hypernatremic dehydration has occurred. Many of these babies are also significantly hyperbilirubinemic. Therefore, assessing serum sodium, potassium, chloride, bicarbonate, BUN, and creatinine levels is essential; initiate treatment as appropriate. In regions where the cultural practice of newborn salting exists, assessing serum electrolytes is especially important as reports of severe hyperbilirubinemia, kernicterus, and death have been reported with serum sodium levels as high as 194 mEq/dL.^[16]

Lumbar puncture

In the initial evaluation of hyperbilirubinemia, sepsis should be included in the differential diagnosis (odds ratio 20.6^[14] ). If so, collection of spinal fluid for culture and cell count should be considered to rule out meningitis. If the baby is having neurologic symptoms, cerebral spinal fluid (CSF) evaluation is imperative; depending on the baby's symptoms, expanding the evaluation beyond the normal aerobic bacterial culture may be prudent. If, on the other hand, the baby is vigorous and well-appearing with isolated hyperbilirubinemia as the only symptom (ie, sepsis is not likely to be the cause of the hyperbilirubinemia), a spinal tap may not be necessary.

Transcutaneous bilirubin measurement

Numerous devices that transcutaneously measure total serum bilirubin levels have become commercially available. These devices have been tested in babies of varying ethnicities, skin pigmentations, and gestational ages Correlation with serum bilirubin measurements is generally good. Although this approach is not recommended to replace the criterion standard of serum measurement, it may be a useful adjunct to the clinical management of hyperbilirubinemia in the term and/or preterm infant.

In 2009, Bental et al reported on the correlation of transcutaneous measurements obtained with the Jaundice Meter Minolta/Draeger JM-103 and serum measurements in 1091 paired measurements obtained on 628 infants. Linear regression analysis of the results yielded a correlation coefficient (R²) of 0.846; the group used these data to create a local Bhutani-type TcB nomogram for universal predischarge screening in US populations.^[26] Yu et al performed 36,921 measurements in 6,035 healthy term and late-preterm Chinese infants to develop a comprehensive nomogram for this population.^[24]

In the acute phase of bilirubin encephalopathy, neuroimaging has no major diagnostic benefit. However, it can help rule out other diagnoses, particularly in the absence of profound hyperbilirubinemia.

Head ultrasonography (HUS)

This modality is particularly well suited to the neonate because it is painless, portable, and noninvasive; also, the neonatal brain is easily imaged through the fontanelles. Ultrasonography is not helpful in diagnosing acute bacterial encephalopathy; however, other entities, such as intraventricular hemorrhage or parenchymal abnormalities, can be ruled out.

CT scanning

CT scanning has little place in the evaluation of the neonatal brain. The subtle abnormalities often present in the neonatal period are not well visualized by CT scanning, and false-negative findings are not uncommon.

MRI

Previously, the neuronal damage characteristic of kernicterus was thought to only be identifiable on histologic examination postmortem. However, experience has revealed that MRI can be used to depict characteristic bilateral symmetric high-intensity signals in the globus pallidus on both T1-weighted and T2-weighted images in patients surviving with chronic bilirubin encephalopathy (see image below). The usefulness and cost-effectiveness of this modality in the diagnosis of more subtle forms of bilirubin-induced neurologic dysfunction (BIND) remains to be fully elucidated.

Brainstem auditory evoked response (BAER): Hearing impediment is the most common sequela of bilirubin toxicity. Impairment may be subtle and may not be clinically apparent until the baby manifests delayed language acquisition. To maximize the baby's long-term neurologic functioning, early identification of any degree of hearing loss is important so that early developmental assessment and intervention can be initiated in a timely fashion. Serial assessments of hearing function may be necessary. With the advent of mandated universal newborn hearing screening, many newborn hearing screening programs contain specific protocols for infants with significant hyperbilirubinemia designed to identify early hearing loss.

On macroscopic examinations, characteristic yellow staining can be readily observed in fresh or frozen sections of the brain obtained within 7-10 days after the initial bilirubin insult. The regions most commonly involved include the basal ganglia, particularly the globus pallidus and subthalamic nucleus; the hippocampus; the substantia nigra; cranial nerve nuclei, including the oculomotor, cochlear, and facial nerve nuclei; other brainstem nuclei, including the reticular formation and the inferior olivary nuclei; cerebellar nuclei, particularly the dentate; and the anterior horn cells of the spinal cord.

Neuronal necrosis occurs later and results in the clinical findings consistent with chronic bilirubin encephalopathy. Histologically, this appears as cytoplasmic vacuolation, loss of Nissl substance, increased nuclear density with haziness to the nuclear membrane, and pyknotic nuclei (see image below).