Newborn jaundice, regardless of its etiology, is a matter of newborn safety, and when it is unrecognized or unmonitored or progresses untreated, it can lead to severe hyperbilirubinemia.1–5 Because kernicterus is preventable, but not treatable, public health policies need to be focused and rooted in a preventive approach.6 Thus, neonatal hyperbilirubinemia not only is an important public health issue but also has significant clinical, societal, and economic consequences for both maternal–child health care and educational systems in the United States. Of the approximately 4 million live births each year, over 80% of term infants, and most preterm infants, manifest jaundice during the first week after birth.6–8 Progression to severe hyperbilirubinemia is due to either increased bilirubin production or impaired bilirubin elimination.9–11 From 1967 to 2000, about 0.14–0.16% of term infants without known Rh disease (140–160/100,000 live births) developed severe hyperbilirubinemia (total serum bilirubin [TSB] levels >20 mg/dL) that required emergency treatment, such as an exchange transfusion, and were at risk for adverse neurologic outcomes.10,12–14 Kernicterus, the ultimate manifestation of irreversible bilirubin-induced neurologic dysfunction (BIND), is more evident among infants with concurrent hemolytic disorders, prematurity, sepsis, and glucose-6-phosphate dehydrogenase (G6PD) deficiency. In about half of the reported cases of kernicterus in the United States, the cause is not diagnosed or investigated.3 Estimates of the incidence of kernicterus range from 1 to 5.8 per 100,000.15 The incidence of hazardous hyperbilirubinemia (TSB >30 mg/dL) ranges from 25 to 60 per 100,000, whereas the risk of kernicterus in this subpopulation is estimated at 1 in 4.12 In the context of prevailing clinical practices, both the use of exchange transfusion and the occurrence of kernicterus are unusual (Figure 12-1), based on reports from health care systems that have accessible service infrastructures and the ability to respond in a timely manner.16–18 Another public health issue is the association of the intent to breastfeed and the severity of hyperbilirubinemia that has often led to the inappropriate withholding of breast milk intake.2,5 The societal costs of neonatal hyperbilirubinemia include the cost of predischarge risk assessment in the context of routine newborn care, interrupted breastfeeding, readmission for treatment with phototherapy and/or an urgent need for an exchange transfusion, parental–infant separation, and the onset and persistence of neurologic injury with lifelong learning and movement disorders.19

This chapter summarizes key and pertinent maternal–child health and other related societal issues pertaining to neonatal jaundice. It also raises key policy issues with respect to maternal–child health care practices during the first week after birth, and highlights the importance of family/ societal education in order to reduce the risk of permanent neurologic sequelae caused by hyperbilirubinemia.

Statement of the Problem

Over 3 million live births in the United States occur after 35 weeks gestation and should have benign outcomes with little or no threat of neurologic compromise from medical conditions during the first year after birth. Proven preventive health measures provided in well baby nurseries have been effective in reducing infant mortality and morbidity in this group of infants. Among the leading causes of neurologic injuries during infancy, kernicterus is one of those conditions that might be prevented. Whether preventive measures are implemented through best clinical practice models or mandated through patient safety strategies has been a matter of recent debate.2,19–24

Standard definitions for the severity of neonatal hyperbilirubinemia at age >72 hours are listed in Table 12-1.3 Confirmation of the diagnosis is made by a transcutaneous bilirubin (TcB) and/or a TSB measurement. Progression of newborn jaundice to severe hyperbilirubinemia and suggested clinical actions are shown in Table 12-2. A range of TSB levels, rather than a singular threshold level, actually represents a range of risk for developing BIND in otherwise healthy term infants.2,6,25–27 A spectrum of BIND further defines more subtle neurologic manifestations in vulnerable infants who have experienced an exposure to bilirubin of lesser degree than generally described in previous publications.28 Such neurologic manifestations (kernicterus spectrum disorders) include various neurologic processing problems, with disturbances of visual motor, auditory, speech, language, and cognition among infants with a prior history of moderate to severe hyperbilirubinemia of variable duration.29 It is in this context that predischarge screening combined with assessment of gestational age is most predictive of subsequent development of severe hyperbilirubinemia at age >72 hours and can direct timely interventions to prevent an infant from progressing to kernicterus.2,21,24,30–34

Traditional epidemiological investigations have failed to track accurately the national incidence of kernicterus.3,17,19,24 Strategies to prevent its surrogate, severe hyperbilirubinemia, are amenable to setting an “aviation safety standard” for newborn care services during the first week after birth. Now, a better understanding of system failures2,5 has led to considerable progress in quality improvement and should encourage the development of a national strategy to prevent kernicterus.5,19,21,24 The cause of kernicterus could be considered as largely a system failure in neonatal care in the first week after birth, probably related to the fact that multiple providers deliver health care services at many different sites to the same infant. A contributing root cause might also be an insufficient understanding of the potential for BIND in the professional community of physicians, nurses, maternal and child health care providers, child health advocates, lactation consultants, as well as third-party payors and the lay society in general. Thus, compounding factors could have contributed to the following: (a) early hospital discharge at age <72 hours (before the risk of severe hyperbilirubinemia is appreciated); (b) insufficient evidence to identify risks of severe jaundice in healthy term and late preterm newborns; (c) an increase in breastfeeding but without counseling and sufficient support to ensure optimal lactation; (d) perceived medical cost constraints complicating reimbursement for follow-up at age 3–5 days; (e) a paucity of educational materials to help families learn about the problem; and (f) a failure of the health care system to provide predischarge screening and identify infants at increased risk for severe hyperbilirubinemia and ensure postdischarge follow-up.19 In response to such deficiencies, the American Academy of Pediatrics (AAP) guideline for the management of newborn jaundice now recommends updated approaches and practices to prevent severe neonatal hyperbilirubinemia as well as avoid the development of acute bilirubin encephalopathy (ABE) and kernicterus.21

A public health question is whether the AAP guideline might be implemented as a national “aviation safety” measure of quality and safety. Using occurrence of defect per critical encounters of adverse icteric experiences (based on critical to quality [CTQ] characteristics), we calculated the defect rate per encounter as the CTQ index, converted to defect rate per million well baby discharges, and determined specific sigma values. Successful implementation of “aviation” safety and quality standards could allow a defect-free experience of 99.99966% to achieve a 6 sigma level of care. For comparison, a sigma level of 4–4.5 represents safety standards prevalent in the US steel industry in the 1960s. Using a TSB level >25 mg/dL as a surrogate, we have previously calculated a 4.5 sigma level in the prephototherapy era of the 1960s and a sigma level of 4.45 in the 1990s.3,4,19 Rates for readmission (1988–1998) show a 4 sigma level performance. In a Danish population-based report of kernicterus,35 6 cases over a 5-year period (estimated at 3 per 100,000 well babies) suggest a 5.5 sigma level. An unproven hypothesis is whether the coordination of current multiorganizational strategies would limit nationwide adverse experiences of TSB >30 mg/dL to 3–4 well babies per million live births and achievement of a 6 sigma level of care.

Currently, the national incidence of kernicterus is not known, but is likely to be low, and the incidence of other more subtle posticteric sequelae is unknown. The US Preventive Services Task Force (USPSTF) has recognized efforts made by clinicians to eliminate this rare, but devastating, condition by instituting system-level measures to screen for hyperbilirubinemia and aggressively manage infants with high bilirubin levels.24 The USPSTF also has noted that there is adequate evidence to support screening using risk factors and/or hour-specific bilirubin measurements for the identification of infants at risk for developing severe hyperbilirubinemia. Because of the remote relationship between a history of hyperbilirubinemia and the subsequent (often delayed) diagnosis of kernicterus later in childhood, it remains unproven whether or not a neonatal screening test will reliably identify all infants who will develop the condition.

Significant advances to prevent severe hyperbilirubinemia during the first week after birth can be attributed to system-based approaches to the management of neonatal jaundice. Clinicians are now less likely to rely on visual assessment of jaundice alone to predict bilirubin levels. They also recognize that there is little evidence to support reliance on clinical judgment alone. They know that bilirubin concentrations should be interpreted in the context of postnatal age in hours. Moreover, there have been technical advances in TcB measurements.7,36–41 A community-based toolkit to track progression of jaundice complemented by the selective use of a TcB monitor or other objective measure of skin color is presented in Table 12-3. This toolkit has been adapted by several state and national professional organizations, as well as health care networks and birthing facilities. Such advances are likely to balance the risk of early discharge after birth (age <72 hours) and the absence of direct medical supervision for the infant cared for at home.

An analysis of existing literature on testing TSB levels in the first 24 hours of life indicates that a TSB ≥17 mg/dL (range 2.9–12.0% of all risk levels) can be predicted with a sensitivity of 94% (95% confidence interval [CI]: 88–97%).36–38,40 On the other hand, the specificity of prediction is 62% (95% CI: 59–65%). These data also indicate that a TSB ≥6 mg/dL at about 24 hours of age is a sensitive predictor of later severe hyperbilirubinemia. In another study, combining a TSB >5 mg/dL at <24 hours with clinical variables (maternal blood group O, maternal age, intent for exclusive breastfeeding, and maternal education) was also predictive.42 This evidence confirms a long-standing clinical “pearl” that visible jaundice in the first 24 hours of life is a risk factor for the later development of severe hyperbilirubinemia.43 Thus, any visible or suspected jaundice in the first 24 hours requires urgent medical attention (within a couple of hours), which must include TSB measurement and an investigation of potential underlying causes.1,2,7,40 Current evidence also suggests that it is possible to identify babies who are likely to develop significant hyperbilirubinemia using a predischarge TSB (age 24 to <60 hours) plotted on an hour-specific nomogram.2,41 The evidence for this recommendation is listed in Table 12-4.

The AAP has published a guideline for the management of hyperbilirubinemia in the newborn infant ≥35 weeks gestation in 20042 that followed a Joint Commission on Accreditation of Healthcare Organizations (JCAHO) Sentinel Alert20 and the Centers for Disease Control and Prevention (CDC) Morbidity and Mortality Weekly Report (MMWR).44 Subsequently, similar consensus-based guidelines were developed and implemented in Denmark,45,46 Canada,47 Israel,43,48 Brazil,42 Europe,49 Norway,50 and the United Kingdom.51 Over the past several years, the AAP guideline has been widely implemented in the United States. Recent publications have also reported the beneficial effect of this guideline in reducing the acute adverse events attributed to severe hyperbilirubinemia, including reduced rates of readmission, TSB levels >25 and >30 mg/dL, and the need for exchange transfusion.16–18,52 Some concerns have been raised about a marginal increase in the use of phototherapy.16,17,52 This may be due to an improved early recognition of severe hyperbilirubinemia when TSB levels are actually measured rather than relying on the visual assessment of jaundice alone.

Promotion of Breastfeeding

Suboptimal breastfeeding among preterm infants is a public health problem. Increasing the rates of breastfeeding initiation, duration, and exclusivity is a national health objective and one of the goals of Healthy People 2010 and 2020. The Department of Health and Human Services,53 AAP,54 and the American College of Obstetrics and Gynecology (ACOG)55 have supported vigorously this initiative. The rates of intent to breastfeed are routinely reported to the CDC.56 More recently, much evidence has focused on the prenatal and intrapartum period as critical for the success of exclusive (or any) breastfeeding.57 Many states, including California, also report the use of breastfeeding at the hospital level.58 Exclusive breastfeeding rates during birth hospital stay have been recorded by the California Department of Public Health using the framework of newborn genetic disease screening program. These birth hospitalization breastfeeding data range from 8% to over 90%. Prevalence of the initiation of breastfeeding and continued breastfeeding at 6 months of age among mothers of term infants in the United States have reached their highest levels recorded to date, 73.8% and 41.5%, respectively; exclusive breastfeeding through 3 months was 30.5%. The AAP recommends that, as a primary mode of prevention for severe hyperbilirubinemia, clinicians should advise mothers to nurse their infants at least 8–12 times per day for the first several days. Poor caloric intake and/or dehydration associated with inadequate breastfeeding may contribute to the development of hyperbilirubinemia. Providing appropriate support and advice to breastfeeding mothers increases the likelihood that breastfeeding will be successful. Clinicians also need to assess the adequacy of intake in breastfeeding infants during the entire neonatal period. Data from a number of studies indicate that unsupplemented breastfed infants experience their maximum weight loss by day 3 and, on average, lose 6.1 ± 2.5% of their birth weight. It is estimated that 5–10% of exclusively breastfed infants lose 10% or more of their birth weight by day 3, suggesting that adequacy of intake should be followed closely. Evidence of adequate intake in breastfed infants includes four to six thoroughly wet diapers in 24 hours and the passage of three to four stools per day by the fourth day. By the third to fourth day, the stools in adequately breastfed infants should have changed from meconium to a mustard yellow, mushy stool. Nationwide policies for newborn jaundice should promote successful breastfeeding and provide toolkits to identify breastfed infants who are at risk for dehydration because of inadequate intake.

Predischarge Risk Assessment

The bilirubin load for an infant represents the net accumulated bilirubin in circulation—the balance of bilirubin production, conjugation, elimination, and enterohepatic circulation. The hour-specific bilirubin nomogram has been recommended as one approach for predicting severe neonatal hyperbilirubinemia and was developed based on the hypothesis that TSB levels measured early in the postnatal course, expressed as a percentile with respect to the infant’s age in hours, are predictive of an infant’s later TSB levels.30

The Agency for Healthcare Research and Quality (AHRQ) sponsored the development of Evidence Syntheses through its Evidence-Based Practice Centers. With guidance from the USPSTF, evidence for clinical preventive services in the primary care setting was reviewed. Such review serves as the foundation for the recommendations of the USPSTF, which provides age- and risk-factor-specific recommendations for the delivery of these services in the primary care setting. The evidence regarding the benefits, limitations, and cost-effectiveness of a broad range of clinical preventive services would help further awareness, delivery, and coverage of preventive care as an integral part of quality primary health care. In a recent systematic review for evidence21,41 of the effectiveness of strategies to prevent kernicterus, there was no study that addressed directly the effectiveness of using risk factor assessment and/or bilirubin testing to reduce the incidence of the condition. The reviewed studies used different threshold values and definitions of hyperbilirubinemia. In one study, two consecutive TSB readings plotted on the nomogram had greater predictive accuracy than a single TSB measurement. Another study indicated that predischarge TcB plotted on an hour-specific nomogram of bilirubin levels derived from healthy babies could predict hyperbilirubinemia with 100% sensitivity and 88% specificity. The threshold values for defining severe hyperbilirubinemia were different for the TcB (≥75th percentile) and TSB (≥95th percentile) levels. Current data suggest comparable predictability of the risk score (newborn and familial jaundice history and clinical characteristics) and the modified risk index (i.e., the risk score without family history of jaundice in a newborn) in predicting later severe hyperbilirubinemia (TSB >95th percentile for age in hours) or use of phototherapy. Studies also suggest that the combination of using the modified risk index with early TSB levels significantly enhanced prediction when compared with using either one of the strategies alone. None of the studies in this review assessed potential harms of screening. Ip et al.4 have previously reported that one needs to treat 6–10 otherwise healthy jaundiced neonates with TSB ≥15 mg/dL by phototherapy in order to prevent the TSB in one infant from rising above 20 mg/dL. However, no substantive harm could be attributed to phototherapy by either of the systematic reviews reported by AHRQ in 2004 and 2009.4,41 Based on these reviews, the USPSTF concluded that a study to evaluate directly the effectiveness of different strategies to reduce the incidence of kernicterus would not be feasible given the rare occurrence of the condition. For practical reasons, studies on the effectiveness of different strategies to reduce the incidence of BIND could only rely on a surrogate outcome such as the TSB level. Outcomes of multiple recent international reviews described above and the most recent AHRQ review41 also confirm the earlier report30 that predischarge TSB plotted on hour-specific bilirubin nomograms show good accuracy in predicting subsequent hyperbilirubinemia (Figure 12-2).

Effective Use of Phototherapy

Clinical trials have validated the efficacy of phototherapy in reducing excessive unconjugated hyperbilirubinemia, and the use of phototherapy has drastically curtailed the use of exchange transfusions.1,2,59–61 The initiation and duration of phototherapy is defined by a specific range of TSB levels on the basis of an infant’s postnatal age and the potential risk for developing BIND (Table 12-5). Aggressive implementation of phototherapy for excessive hyperbilirubinemia, sometimes referred to as the “crash-cart” approach, has been reported to reduce the need for exchange transfusions, and, possibly, reduce the severity of BIND. Barriers to effective use of phototherapy are described in Table 12-6. Affordable access to effective phototherapy devices (see checklist in Table 12-7) is an urgent global public health need.