Abstract
Starting in the 1950s through the 1980s, clinical practice has evolved to essentially eliminate classic kernicterus. Great strides have been made in the management of neonatal jaundice, hyperbilirubinemia, and prevention of kernicterus over the past 25 years. The American Academy of Pediatrics practice parameter on the management of jaundice, published in 2004 and updated in 2009, marked a shift in clinical practice that initially resulted in a large decrease in the number of children developing kernicterus. However, a reemergence of kernicterus occurred in recent years and continues to the present. This has been associated with changes in medical practice and health care delivery. Importantly, universal screening of infants combined with close follow-up, monitoring, and aggressive phototherapy treatment at relatively low bilirubin levels could eliminate most of the new cases of kernicterus, but the prevention of a devastating but very rare disorder must be balanced against the costs and possible risks of overtreating a very large numbers of infants. Determining the bilirubin level at which toxicity is increased is a difficult issue for physicians in everyday practice. This must be balanced against the gains of an early visit versus the concerns that increased surveillance will result in a greater use of resources (phototherapy), more parental anxiety, and reduced breastfeeding. Gaps in knowledge persist, including the neurotoxic impact of hyperbilirubinemia in preterm infants, and the potential importance of genomic susceptibility to bilirubin neurotoxicity.
Keywords
acute bilirubin encephalopathy, kernicterus, hearing loss, phototherapy light intensity
- •
Following AAP guidelines prevents most cases of kernicterus.
- •
Extreme hyperbilirubinemia requires emergent treatment with double volume exchange transfusion.
- •
Early auditory assessment is important because both ANSD and SNHL respond better to early treatments including cochlear implantation.
- •
Acute MRI findings are T1 hyperintensity of bilateral globus pallidus (GP), which then resolves to be later replaced by T2 hyperintensity.
- •
A novel term Kernicterus Spectrum Disorders (KSDs) is used to clinically define and categorize kernicterus subtypes and severity.
BB is a white male infant, born at term gestation weighing 2980 g to a 25-year-old G1P0 mother, blood type B + , via spontaneous vaginal delivery. Apgar scores were 8 and 9 at 1 and 5 minutes, respectively. The infant was blood type A + , had a large cephalohematoma, and appeared jaundiced at 24 hours of life. He passed an automated auditory brainstem response (ABR) screening and was discharged at 58 hours with a transcutaneous bilirubin level of 13.2 mg/dL and a total serum bilirubin level of 14 mg/dL. One day later at follow-up, the pediatrician said he looked fine. He returned at 6 days of age with a history of lethargy but had regained his birth weight. The pediatrician estimated visually that the bilirubin level was about 5 mg/dL. On day 7 the infant was more lethargic and had feeding difficulty. On day 8 he was noted to have a high-pitched cry, downward deviation of the eyes, and episodic and then continual extension of arms, legs, neck, and trunk. He was evaluated in an emergency room. Cerebral spinal fluid (CSF) analysis was normal except for yellow color of the fluid. Total bilirubin level at 211 hours (8.8 days) of age was 45.6 mg/dL. He was placed under phototherapy lights with a blanket underneath, was intubated because of low oxygen (O 2 ) saturation, and received a double-volume exchange transfusion at 214 hours of age. Preexchange and postexchange total bilirubin values were 35.8 and 20.9 mg/dL, respectively. BB was intubated and sedated. Five days later, emerging from sedation, he was hypotonic with a setting sun sign and with episodic O 2 desaturation. He was treated with phenobarbital for presumed seizures. However, an electroencephalogram (EEG) was reported as normal except for some sharp waves over the left temporal region. He was discharged after a 10-day hospitalization. He has subsequently demonstrated feeding issues, incoordination of suck and swallow, hypertonia and dystonia, and delayed motor development. ABRs at 2.5 weeks and 2 months of age were absent, a finding consistent with auditory neuropathy/dyssynchrony, despite normal distortion product otoacoustic emissions (OAEs). Magnetic resonance imaging (MRI) done at 11 days of age showed increased signal intensity in the globus pallidus bilaterally on both T1- and T2-weighted images and a resolving cephalohematoma. Because of reflux and failure to thrive, a gastrostomy tube was placed with Nissen fundoplication. Treatment with diazepam was instituted to treat increased tone. Phenobarbital dosage was tapered and then discontinued. Intermittent clonic activity was noted, but video EEGs repeatedly failed to demonstrate an electrographic correlate. At 1.5 years of age, BB had not developed speech or language, had not responded to sound amplification, and could not sit or crawl.
This case highlights the risk of visual assessment of jaundice. The serum bilirubin level at discharge, 14 mg/dL, was at the 95th percentile of the hour-specific nomogram developed by Bhutani and colleagues. The bilirubin at the 95th percentile, the significant cephalohematoma, and an ABO blood group abnormality were predictive of a high subsequent bilirubin level. The infant had symptoms of acute bilirubin encephalopathy (ABE) that were not recognized until it was too late to prevent extreme hyperbilirubinemia, and as a result the child developed all the clinical signs, symptoms, and laboratory findings of kernicterus.
Definition and Epidemiology
The classic clinical syndrome caused by brain injury as the result of hyperbilirubinemia is called kernicterus . Kernicterus was originally a pathologic term referring to the yellow staining ( icterus ) of the deep nuclei of the brain ( kern, relating to the basal ganglia). The term chronic bilirubin encephalopathy has been used interchangeably with kernicterus. Acute bilirubin encephalopathy (ABE) refers to the acute injury associated with bilirubin neurotoxicity in the neonatal period. Historically the terms acute bilirubin encephalopathy and chronic bilirubin encephalopathy or kernicterus have been used to describe the clinical symptoms associated with the neuropathology. With the growth of modern medicine clinical studies such as magnetic resonance imaging (MRI) and auditory brainstem responses (ABRs), also known as brainstem auditory evoked potentials (BAEPs) or responses (BAERs), have shown objective evidence of bilirubin encephalopathy as evidenced by characteristic neuropathologic lesions (MRI) and abnormal neurologic function (ABRs).
Classic kernicterus is a well-described clinical syndrome that involves (1) a dystonic or athetoid movement disorder, (2) an auditory processing disturbance now called auditory neuropathy spectrum disorder (ANSD) with or without hearing loss, (3) oculomotor impairment, especially impairment of upward gaze, (4) dysplasia of the enamel of deciduous (baby) teeth, and, perhaps less well-known, (5) hypotonia and ataxia owing to cerebellar involvement. It is important to emphasize that more subtle forms of brain injury caused by an excess of hyperbilirubinemia have also been described. These have been referred to by different terms, including subtle kernicterus, chronic bilirubin encephalopathy of the subtle type, and bilirubin-induced neurologic dysfunction (BIND). We now advocate use of the term kernicterus spectrum disorders (KSDs) to encompass the entire spectrum of bilirubin-induced neurologic disorders, ranging from subtle neurodevelopmental disorders in children with less than all of the clinical features of classic kernicterus, up to and including the most severe forms of classic kernicterus.
Classic kernicterus is a rare disorder. However, the severity of the lifelong disability, especially deafness, abnormal tone, and the loss of motor control, can be profound. Studies in Canada find an incidence of 1 case per 44,000, whereas in Denmark the incidence of kernicterus is estimated to be 1 case per 110,000 live births. The incidence of extreme hyperbilirubinemia, defined as at or above the criteria for performing exchange transfusion (median 28.8 mg/dL, range 22.5–40.3 mg/dL), was 25 cases per 100,000. A population-based survey in northern California found that 150 per 100,000 infants had total serum bilirubin (TSB) levels of 25 mg/dL or higher and 10 per 100,000 had levels of 30 mg/dL or higher. In the United States, based on the Canadian estimate of 1 per 44,000 live births, at least 90 cases of severe kernicterus per year are predicted. The incidence in underdeveloped low- and middle-income countries is much higher. A study in Nigeria revealed an incidence of 2.7%, equivalent to 1 in 37 live births. A review by Bhutani and Johnson estimated the risk of kernicterus in infants with TSB levels above 25 mg/dL to be about 1 in 17.6 cases in Canada and 1 in 16.2 cases in Denmark, and for TSB levels above 30 mg/dL, the risk of kernicterus was estimated to be as high as 1 in 7 cases.
The incidence of extreme hyperbilirubinemia and kernicterus depends on the care infants receive in the first few weeks of life, the vigilance of screening, and the ability to treat excessive hyperbilirubinemia should it occur. Incidence and prevalence studies are complicated by the lack of specific objective definitions of kernicterus or KSDs and the inconsistency of reporting kernicterus or hyperbilirubinemia in most countries. Further, the incidence and prevalence of subtle neurodevelopmental disorders, such as auditory processing disorders and visual motor disabilities resulting from hyperbilirubinemia, remain unknown, largely owing to a lack of recognition of the syndrome.
Precisely how to determine the risk of brain injury in a hyperbilirubinemic newborn is important not only for choosing the level of care but also for optimally allocating health care resources. The precise determination of risk of brain injury will allow not only better guidelines for treatment to prevent brain damage but also the reduction or elimination of unnecessary treatments. This chapter discusses new concepts of pathogenesis, approaches or strategies for diagnosis and treatment, gaps in knowledge, and recommendations for treatment.
History of KSD Prevalence
Through the decades from the 1950s to the 1980s, clinical practice evolved to essentially eliminate classic kernicterus; however, a reemergence of kernicterus occurred in the 1990s and continues to the present and is associated with changes in medical practice and health care delivery. Clearly, universal screening of infants combined with close follow-up, monitoring, and aggressive phototherapy treatment at relatively low bilirubin levels could eliminate most of the new cases of kernicterus, but the prevention of a devastating but very rare disorder must be balanced against the costs and possible risks of overtreating very large numbers of infants. Currently a postdischarge bilirubin level is now standard practice within 2 days of discharge or earlier, depending on the predischarge level to identify patients with spikes in bilirubin after discharge. Determining the bilirubin level at which toxicity is increased is a difficult issue for physicians in everyday practice. They must balance the gains of an early visit versus the concerns that increased surveillance will result in a greater use of resources (phototherapy), more parental anxiety, and reduced breastfeeding.
Pathogenesis of Bilirubin Neurotoxicity
Bilirubin neurotoxicity is highly selective, targeting specific neurons in the central nervous system (CNS). The clinical expression and neuropathology of kernicterus are likewise highly selective. The movement disorders, dystonia and athetosis, are likely the result of lesions in the basal ganglia (globus pallidus and subthalamic nucleus) and cerebellum. Lesions to brainstem nuclei result in auditory, vestibular, and oculomotor function impairment and poor truncal tone. Any explanation of pathogenesis must account for the selective neurotoxicity of bilirubin.
Total bilirubin in the blood, called total plasma or TSB, is composed of unconjugated bilirubin (UCB) plus conjugated bilirubin, also known as indirect and direct bilirubin, respectively. Neurotoxic UCB is largely bound to blood proteins, especially albumin, and except in unusual circumstances (e.g., disruption of the blood-brain barrier [BBB]) albumin-bound UCB does not move out of blood into brain. However, unbound or “free” bilirubin (Bf) is capable of crossing the BBB and moving into the central nervous system (CNS). Bf formation is favored when bilirubin binding sites become saturated and with low pH.
It is useful to compartmentalize the concept of bilirubin neurotoxicity into (1) processes that cause the cells to be exposed to excess bilirubin (production and elimination), (2) those that interfere with the cell’s ability to handle the excess bilirubin, and (3) the molecular mechanisms responsible for the cell’s response to the stress caused by bilirubin neurotoxicity.
Overall bilirubin exposure is a combination of bilirubin production, binding, and excretion. Bilirubin is actively excreted from the cells in the CNS. The ability of cells to handle excess bilirubin may vary. Risk factors can affect one or more of the compartments defined previously. For example, acidosis decreases binding and increases unbound Bf, thus exposing the cell to more bilirubin. However, pH might also affect cellular enzymes, energy production, and transport. Other risk factors, such as prematurity, inflammation, and isoimmunization, may act in one or more specific compartments.
In the following discussion, we examine individually the contributions from each of the compartments defined earlier, starting with production and elimination and concluding with cellular and molecular mechanisms underlying bilirubin neurotoxicity. An essential concept for clinicians caring for children with hyperbilirubinemia is the importance of thinking in terms of Bf exposure rather than total bilirubin. The BBB excludes large molecules, including albumin and albumin-bound bilirubin from the CNS, and transports substances into and out of the CNS. The endothelial cells of the BBB restrict diffusion of many toxic compounds to adjacent neurons and astrocytes but do not restrict the free diffusion of free (Bf), unbound, unconjugated bilirubin (UCB), which is permeable with single-pass uptake estimated to be as high as 28% in rats. Therefore it is the free fraction of bilirubin that matters as it relates to bilirubin neurotoxicity.
Comprehensive reviews have been published about the molecular mechanism of bilirubin neurotoxicity. Recent findings have shed new light on the molecular basis of jaundice, possible molecular mechanisms underlying the neurologic selectivity of bilirubin neurotoxicity, the role of unconjugated bilirubin in health and disease, and the possibility of genetic factors leading to susceptibility or resistance to bilirubin neurotoxicity.
Active Transport of Bilirubin
Specific transporters for bilirubin, such as multidrug resistance protein 1 (MRP1), a member of the multidrug resistance–associated protein subfamily of adenosine triphosphate (ATP)-binding cassette transporters, may protect the CNS from exposure to excessive levels of bilirubin. Accumulating evidence indicates that bilirubin is removed from cells via MRP1. MRP1 transports bilirubin with an affinity that is 10 times greater than that of other substrates and may represent a mechanism by which bilirubin is transported out of the CNS and excreted into the circulation. It is hypothesized that when this transport system is overwhelmed, toxic levels of bilirubin can accumulate in the cell.
Mechanisms of Bilirubin Damage
It appears that bilirubin damages cells by both apoptotic and necrotic mechanisms and affects mitochondrial energy metabolism. Studies in cultured cells are starting to clarify the roles of apoptosis, mitochondria, and other molecular mechanisms contributing to bilirubin neurotoxicity. Purified bilirubin can induce apoptosis in cultured rat brain neurons by triggering the release of cytochrome c from mitochondria with caspase-3 activation and cleavage of poly(adenosine diphosphate [ADP]-ribose) polymerase, confirming the role of one of the pathways that underlies induction of apoptosis. Studies have shown that bilirubin facilitates transmitter release in cochlear nucleus neurons via presynaptic protein kinase A activation, which might provide insight into the cellular mechanism underlying bilirubin-induced hearing dysfunction. Apoptotic changes are also reported in the cerebellum and brainstem of jaundiced Gunn rats (an animal model of hyperbilirubinemia) and in the basal ganglia of kernicteric human infants, accounting for prominent signs of bilirubin neurotoxicity in each. Furthermore, Falcao and colleagues have raised the possibility that neuroinflammation may be a significant contributor to bilirubin neurotoxicity.
Bilirubin and Calcium Homeostasis
Braun and Schulman have proposed that bilirubin interferes with intracellular calcium homeostasis and have demonstrated decreased activity of calcium and calmodulin-dependent protein kinase II (CaMKII) in the Gunn rat model of kernicterus. Bilirubin inhibits CaMKII in vitro, and its developmental expression is impaired in jaundiced (jj) Gunn rats. Further, there is a selective decrease in expression of calcium-binding proteins (CBPs) in specific brainstem areas susceptible to bilirubin neurotoxicity in the CNS of kernicteric Gunn rats. CaMKII regulates neurotransmitter release, calcium-regulated ion conductance, and neuroskeletal dynamics and can trigger programmed cell death (apoptosis). It has also been proposed that bilirubin and increased calcium cause damage through an excitotoxic, N -methyl- d -aspartate (NMDA)-dependent mechanism, although there is also evidence against this hypothesis. My colleagues and I have found that MK-801 (dizocilpine), an NMDA channel blocker, does not protect against bilirubin neurotoxicity in vitro in hippocampal neurons, nor does it protect against auditory dysfunction in vivo in the Gunn rat model of ABE.
Mechanisms of Auditory Dysfunction
Another study used the Gunn rat to investigate bilirubin-induced auditory deficits. In vivo ABRs revealed severe auditory deficits within 18 hours of exposure to high bilirubin levels. Extracellular multielectrode array recordings following hyperbilirubinemia in an in vitro preparation of the auditory brainstem demonstrated transmission failure indicative of damage at a presynaptic site in the medial nucleus of the trapezoid body. Similarly, multiphoton imaging demonstrated that giant synapses in this nucleus were destroyed. These neurons express high levels of nitric oxide synthase (NOS); nitric oxide has been implicated in mechanisms of bilirubin toxicity elsewhere in the brain, and antagonism of neuronal NOS by 7-nitroindazole was found to protect hearing during bilirubin exposure.
Neuroprotective Action of Bilirubin
Somewhat paradoxically, bilirubin and heme oxygenase 2, the enzyme that catalyzes its formation, have been shown to have neuroprotective antioxidant properties. Neurons are susceptible to apoptotic death from oxidative stress. Work from the Snyder laboratory has begun to define the role of bilirubin and its precursor, biliverdin, in normal cells. Depleting cultured HeLA cells and cortical neurons of biliverdin reductase by ribonucleic acid (RNA) interference leads to increased oxidative activity and renders cells more susceptible to caspase-dependent death from hyperoxia and hydrogen peroxide toxicity. The antioxidant activity of these compounds is comparable to that of glutathione, long assumed to be the principal cellular antioxidant. Hippocampal cultures from heme oxygenase 2 (but not 1) knockout mice are more susceptible to hydrogen peroxide toxicity, and heme oxygenase 2 knockout mice are more susceptible to focal cerebral ischemia and excitotoxic injury. Despite its extremely low concentration in the cell, bilirubin is recycled to greatly increase its effect. Bararano and colleagues have proposed that the potent physiologic antioxidant action of bilirubin is amplified when the detoxification of reactive oxygen species oxidizes bilirubin back to biliverdin, which is then reduced again by biliverdin reductase to form bilirubin. As this redox-amplification cycle is repeated, the antioxidant effect of bilirubin is multiplied. Although this redox-amplification cycle has been proposed to constitute the principal physiologic function of bilirubin, it should be noted that a protective action of modest levels of bilirubin does not alter the well-established dangers of kernicterus associated with major elevations of serum bilirubin.
Reinforcing the concept that bilirubin has a protective/beneficial effect, numerous studies in adults have shown the beneficial effects of slight elevations of UCB throughout the life span—for example, the lower incidence of heart disease and stroke in individuals with Gilbert syndrome and a mild unconjugated hyperbilirubinemia. The evidence for molecular and clinical effects of bilirubin in health and disease has recently been reviewed and summarized by Gazzin and colleagues.
Molecular Response to Hyperbilirubinemia
As just discussed, bilirubin is on the one hand a powerful antioxidant that has protective properties and modulates cell growth; however, in excess, it has deleterious properties that cause cell death and apoptosis, in part owing to oxidative stress. Cells exposed to bilirubin increase expression of a multifunctional neuroprotective protein DJ-1, an adaptive response that maintains cellular oxidative stress homeostasis. Dogan and coworkers have demonstrated a relationship between serum total bilirubin and antioxidants and oxidative stress by studying oxidant and antioxidant status in 36 neonates with hyperbilirubinemia requiring phototherapy (per the 2004 American Academy of Pediatrics [AAP] guidelines), 33 with kernicterus (ABE on examination and absence of ABRs), and 25 age-matched healthy controls. Plasma total antioxidant capacity and serum total oxidant status were significantly higher in the hyperbilirubinemia and kernicterus groups than in the controls, and total free sulfhydryl values were significantly elevated in the hyperbilirubinemia groups. The investigators found a relationship between serum total bilirubin, antioxidants, and oxidative stress that could be expressed by a quadratic correlation curve. They suggested that that after a certain point, kernicterus occurs with a direct toxic effect on the cell. Certainly, more studies on the role of oxidant and antioxidant status in hyperbilirubinemic and kernicteric infants will be of great interest.
Developmental Susceptibility
Windows of developmental susceptibility of the CNS to bilirubin toxicity may exist. Cerebellar neurons undergoing early differentiation at the time of bilirubin exposure are highly susceptible to bilirubin neurotoxicity, whereas slightly more or slightly less mature neurons may show only transient changes. Bilirubin-induced cell death, glutamate release, cytokine production, and activation of transcription factors involved in inflammation are enhanced in undifferentiated astrocytes and neurons compared with more mature cells. Falcao and associates have shown that both messenger RNA and protein levels of MRP1 increase with cell differentiation, and suggest that the developmental expression of MRP1 may explain the increased susceptibility of premature infants to bilirubin neurotoxicity.
It is likely that prolonged exposure of neurons to bilirubin causes cell death and the developmental stage of the neuron determine whether the cell can handle the bilirubin load. The hyperbilirubinemia must be sufficient to permit Bf to enter the neuron and exceed the neuron’s capacity to handle it. Neuronal exposure to extremely high bilirubin levels for relatively short periods probably affects the cell differently than exposure to relatively lower but still excessive levels for a prolonged period. The developmental stage of the neuron, for reasons that are not completely clear, probably determines in part whether or not the cell can handle the bilirubin load.
Diagnosis: Acute Bilirubin Encephalopathy and Kernicterus Spectrum Disorders
Strategies for diagnosis and treatment of hyperbilirubinemia are well known. The AAP guidelines for the treatment of hyperbilirubinemia are a significant improvement over previous guidelines. Following these guidelines has undoubtedly prevented many cases of kernicterus, although a study using data derived from diagnosis coding showed no change in reported cases of kernicterus, casting some doubt on the effect of the guidelines on the incidence of kernicterus. Furthermore, the actual number of cases of KSDs remains unclear. In addition, the number of unnecessary treatments and associated cost to the health care system is also unknown. Much of this gap in our knowledge regarding the incidence of KSD and its response to treatment is a result of the lack of objective methods to diagnose ABE and kernicterus. These tools are needed in any study that purports to determine the incidence of kernicterus; accepting diagnosis codes as the dependent measure in incidence studies fails to account for the large number of cases either not recognized or reported.
Suggested Improvements to the AAP Guidelines for the Treatment of Hyperbilirubinemia
Despite the improvements provided by the current AAP guidelines, some concerns should be mentioned. First is the failure to strongly recommend a predischarge bilirubin measurement as an indispensable part of risk assessment (rather than saying that this is one of two recommended options, the other being risk factor summation). Second is the lack of clear guidelines for bilirubin/jaundice assessment at the first and subsequent follow-up visits. As shown by the pilot kernicterus registry and personal experience, infants who have severe hyperbilirubinemia and kernicterus are often those whose jaundice at a follow-up visit is not considered severe or extensive enough to warrant a serum bilirubin measurement. This failure can be the result of forgetting that before an infant becomes pumpkin orange, full-body jaundice usually is associated with a clearly deeper yellow in the face and belly than in the legs and feet (Johnson, personal communication, 2007). The observer may therefore assume that full-body jaundice is not yet present when the level may indeed be elevated. Third, an infant with a serum bilirubin level of 11 to 12 mg/dL at early discharge (i.e., 48 to 72 hours) should have close follow-up and the bilirubin level confirmed by a TSB measurement or transcutaneous bilirubin evaluation. Finally, the guidelines do not yet take into account potential new methods of measuring unbound (free) bilirubin, not yet clinically available, that in the future may lead to better, more effective preventive treatment strategies. Basing screening and treatment decisions on the TSB value alone is not likely to improve the sensitivity and specificity of detecting and preventing kernicterus. Using new strategies to assess individual binding—perhaps a two-step method using TSB and clinical criteria to determine risk category and then using Bf to determine binding and, more precisely, TSB action levels for a particular infant—might improve sensitivity and specificity.
Acute Bilirubin Encephalopathy
In newborns with ABE, symptoms progress from lethargy and decreased feeding to variable or fluctuating tone (hypotonia and hypertonia), high-pitched cry, retrocollis and opisthotonus, impairment of upward gaze (setting sun sign), fever, seizures, and death. ABRs are absent or abnormal, and these findings are potentially reversible with double-volume exchange transfusion if identified early. MRI reveals abnormal signal bilaterally in specific subdivisions of the basal ganglia, globus pallidus, and subthalamic nucleus. Abnormalities are found within days after the peak TSB level and are initially hyperintense on T1-weighted images; they later become normal or hypointense on T1- and hyperintense on T2-weighted images, and the ABR is abnormal or absent bilaterally.
The diagnosis of acute encephalopathy can usually be made in term or near-term newborns through clinical examination (mental status, tone, posture, cry, setting sun sign, and late signs of opisthotonus, seizures, and fever) and laboratory values (measurements of TSB, UCB, and conjugated bilirubin, pH, and albumin). Some clinicians believe that hyperbilirubinemia from Rh disease (or hemolysis) is more likely to be damaging than that from other causes (perhaps because of a more rapid rate of rise or rapid rate of bilirubin production). Concurrent illnesses such as sepsis may increase the risk of ABE. Confirmatory studies include ABR and MRI, both of which provide specific diagnostic information in newborns and can usually distinguish encephalopathy owing to bilirubin toxicity from encephalopathy owing to other causes.
ABE and Auditory Nervous System
Because the auditory system is quite sensitive to bilirubin neurotoxicity, the ABR is a sensitive and objective measure of early CNS dysfunction owing to bilirubin. Early changes, such as increased latency and decreased amplitude of waves III and V, herald the onset of bilirubin neurotoxicity, probably at a reversible stage. Wave III arises from the cochlear nuclei in the pons, and wave V arises from the lateral lemniscus in the midbrain, a pathway terminating in the inferior colliculus. The use of interwave intervals (IWIs), such as I–III, III–V, and I–V, reflects brainstem conduction time. The earliest ABR indicators of bilirubin neurotoxicity are abnormality increases in the I–III and I–V IWIs. As bilirubin neurotoxicity progresses, ABR abnormalities progress to absence of waves III and V and, finally, to complete absence of all waves, including I. Clinically, ABR wave amplitudes (from peak to subsequent trough) are usually too variable to serve as criteria for ABR normality, although they are useful when serial studies can be compared with a baseline, and the use of amplitude ratios III:I and V:I helps decrease variability ( Fig. 10.1 ).
The ABR may be absent or abnormal with an increase of I–III and I–V IWIs. The cochlear microphonic (CM) response, which can be easily obtained at the time of ABR testing, should be present even in the absence of ABR neural waves I, III, and V. Occasionally, giant CM responses may occur and may be mistaken for ABR waves by inexperienced readers. CM waves are distinguished from ABR waves by the fact that the former do not change their latencies with intensity (whereas ABR waves do) and phase-reverse with phase reversal of the click stimulus (whereas ABR waves do not). Abnormal ABRs may or may not improve with time in infancy. OAEs are initially normal but may disappear with time for unknown reasons. A full audiology evaluation is usually advisable. As with MRI, an abnormal ABR result in the presence of ANSD is consistent with bilirubin neurotoxicity, but a normal ABR result does not rule out kernicterus.
The automated ABR (AABR) device has been used to assess auditory function in hyperbilirubinemic newborns in the neonatal intensive care unit (NICU). The AABR matches the ABR response to a predesigned template and is reported as either present (PASS) or absent (REFER). REFER AABRs in the presence of hyperbilirubinemia, especially in a neonate who has become hyperbilirubinemic and whose AABR result has changed from PASS to REFER, are evidence of abnormal auditory function and may reflect significant bilirubin neurotoxicity. Unfortunately, the AABR as opposed to the ABR does not distinguish CNS dysfunction from conditions that are not caused by bilirubin toxicity (e.g., middle or inner ear dysfunction) and current AABRs may not detect the early changes of bilirubin neurotoxicity. However, the ease of use and availability of AABR at all times in the NICU make it a clinically useful device to decide when and how aggressively to treat sick neonates in unclear or borderline clinical situations. Although the risks of overtreatment should be avoided, the risk of undertreatment causing lifelong severe kernicterus is potentially devastating .
ANSD may occur either with or without hearing loss. Mild ANSD may coexist with normal or mildly abnormal “hearing” and audiogram but with auditory dyssynchrony, difficulty distinguishing sound from background noise, and difficulty in sound localization. Severe ANSD may manifest as absent ABR with profound deafness. A prospective observational study in late preterm and term neonates with severe jaundice concluded that ANSD is a common manifestation of acute bilirubin neurotoxicity. In infants with bilirubin levels meeting the AAP requirements for exchange transfusion, a comprehensive auditory evaluation was performed before discharge. Six of 13 neonates (46%) had audiologic findings of acute ANSD. There were no detectable differences in clinical variables, including hemolysis, peak total bilirubin, and peak bilirubin-to-albumin molar ratio between the six neonates who had ANSD and the seven who had normal audiologic findings. Furthermore, only two of the six with ANSD had clinical signs and symptoms of ABE. Thus ANSD is a common manifestation of acute bilirubin neurotoxicity in late preterm and term infants and may occur without clinical signs and symptoms of ABE.
ABE and Neuroimaging
As mentioned earlier, the findings of ABE on MRI imaging of the brain include characteristic changes in the basal ganglia. These changes occur very rapidly after peak bilirubin exposure and can usually be seen within days of the injury. The observed changes are initially hyperintense on T1-weighted images and later normalize or become hypointense. T2-weighted images show a hyperintense signal. While the subthalamic nucleus is also usually affected, this structure is small enough that it can be a challenge to identify on imaging and is frequently overlooked. In some patients, abnormalities may normalize. New findings are emerging from MRI and magnetic resonance spectroscopy (MRS) scans of term and preterm neonates with severe hyperbilirubinemia, including pathways from the dentate nucleus of the cerebellum to thalamus and cortex, with hopes that more advanced techniques (e.g., diffusion-weighted imaging and diffusion tensor imaging with tractography) will provide new insights into the pathogenesis of KSDs.
Difficulties in obtaining MRI in sick neonates in most centers preclude its use for making clinical decisions such as when to intervene (e.g., with an exchange transfusion). However, MRI is useful to establish the diagnosis of kernicterus, and in many cases MRI findings are abnormal almost immediately after the fact. If in the future neuroprotective rescue treatments after severe hyperbilirubinemia are proposed, then MRI and ABR testing may be critically important to objectively identify patients at highest risk of KSDs to determine their response to treatment.
Findings of other neuroimaging techniques, such as head computed tomography (CT) and head ultrasound scans (HUS), are generally normal in kernicterus, although hyperechogenicity is occasionally seen. In one study an abnormal HUS was reported in one of eight preterm and term infants with abnormal MRI findings. Certainly, CT and HUS are useful when intracranial bleeds, hydrocephalus, and congenital anomalies need to be excluded. MRS has currently unrealized potential to reveal abnormalities of brain energy metabolism, although recent studies by Wisnowski et al. have used advance techniques of MRI and MRS to shed new light on the role of the cerebellum in bilirubin neurotoxicity.
New Strategies in the Diagnosis of ABE
New strategies for the diagnosis of ABE have been proposed. The first is a bilirubin-induced neurologic dysfunction (BIND) score, similar to an Apgar score, to be used in infants with significant hyperbilirubinemia. ∗
∗ It is worth noting that the BIND acronym has been used in some cases to refer to all bilirubin-induced neurologic disorders from mild to severe, and in some cases to refer to only subtle bilirubin-induced neurologic disorders referring to subtle KSD. It is in part to avoid this confusion that we recommend replacing this terminology by the more explicit subtle KSD .
The BIND score, proposed by an ad hoc BIND study group, rates neurologic symptoms such as muscle tone, posture, cry, and mental status (based on Volpe’s description of ABE ) on a scale of 0 to 9 in infants with high TSB values ( Table 10.1 ) and can be used much like an Apgar score. Although the BIND score is simple to use, it has some limitations. For example, crying, a major component of the score, is difficult to assess objectively and cannot be assessed when the infant is intubated. The BIND score still needs to be validated. However, it has been used retrospectively to track the clinical progression of ABE from nonspecific, subtle encephalopathy (score 1–3) through progressive toxicity (score 4–6) to advanced toxicity (score 7–9), and it has been modified to be used in low- and middle-income countries. Finally, it is important to keep in mind that ABE in premature infants may occur with different or no clinical signs (see “ Bilirubin Neurotoxicity in the Very Premature Infant ”).| Category | Score |
|---|---|
| Mental status (0–3) | 0 = Normal |
| 1 = Sleepy, poor feeding | |
| 2 = Lethargic, irritable | |
| 3 = Semicoma, coma, seizures | |
| Muscle tone (0–3) | 0 = Normal |
| 1 = Neck stiffness, mild hypertonia or hypotonia | |
| 2 = Arching neck and/or trunk | |
| 3 = Opisthotonus | |
| Cry (0–3) | 0 = Normal |
| 1 = High-pitched | |
| 2 = Shrill | |
| 3 = Inconsolable |
Related posts:
Cerebral Circulation and Hypotension in the Premature Infant: Diagnosis and Treatment
Hypothermia for Neonatal Hypoxic-Ischemic Encephalopathy: Different Cooling Regimens and Infants Not Included in Prior Trials
General Supportive Management of the Term Infant With Neonatal Encephalopathy Following Intrapartum Hypoxia-Ischemia
Neonatal-Onset Epilepsies: Early Diagnosis and Targeted Treatment
Neonatal Meningitis: Current Treatment Options
Amplitude-Integrated EEG and Its Potential Role in Augmenting Management Within the NICU
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
