Kernicterus in the 21St Century: Frequently Asked Questions

Home , Kernicterus, Rh disease

VK Bhutani1 and L Johnson2 1Department of Pediatrics, Division of Neonatal and Developmental Medicine, Stanford University School of Medicine and Lucile Packard Children’s Hospital, Stanford, CA, USA and 2Pennsylvania Center for Kernicterus, Philadelphia, PA, USA

8 to 11% of infants.1,2 The etiological basis for progressive Acute kernicterus remains a clinical emergency and its delayed hyperbilirubinemia is multifactorial and due to excessive bilirubin management represents an easily preventable neonatal brain injury. Yet, production and/or decrease its elimination is influenced by both practitioners encounter recurrent questions regarding the risk and timing of constitutional and environmental factors. In 1994, the Provisional bilirubin-related neurotoxicity. These include the following: does bilirubin Committee for Quality Improvement and Subcommittee on damage the brain of healthy infants? Is there a re-emergence of kernicterus Hyperbilirubinemia of the American Academy of Pediatrics (AAP) in the United States? Was kernicterus previously prevented in the United published a practice parameter to manage hyperbilirubinemia in States? What was the public health impact of 1994 American Academy of the healthy term newborns.3 Updated 2004 clinical practice Pediatrics Guidelines? What is the current incidence of kernicterus and guidelines address concerns regarding the risk of kernicterus and severe neonatal hyperbilirubinemia? What is the estimated risk of represent a consensus of the committee charged by the AAP with kernicterus in infants with excessive hyperbilirubinemia? Is there a specific a careful review of evidence and literature.4,5 A number of bilirubin threshold total serum bilirubin (TSB) value for neurotoxicity? Are questions impact daily clinical practice. Governed by concerns for there sequelae of severe or prolonged moderate hyperbilirubinemia in the patient safety, clinical practice needs to range from reassurance, absence of recognized acute bilirubin encephalopathy? Can we define a alerts and concerns of alarm for target populations in the bilirubin level that is safe in newborns? We address these questions in the continuum of progressive hyperbilirubinemia. Some of the context of available data and evidence, and estimate the current risk of questions that clinicians have been asked are addressed below in chronic kernicterus is about one in seven in infants with TSB >30 mg per the context of available evidence. 100 ml (513 mmol lÀ1). Journal of Perinatology (2009) 29, S20–S24; doi:10.1038/jp.2008.212

Keywords: newborn jaundice; kernicterus; bilirubin; hyperbilirubinemia; well babies; bilirubin-induced neurological dysfunction Frequently asked questions Question I: Does bilirubin damage the brain of healthy infants? Yes. Prior to publication of the 1994 AAP recommendations, the ‘lack of evidence’ of bilirubin neurotoxicity in term apparently Introduction healthy babies in the pediatric literature promoted a kinder, An American clinical emergency is the occurrence of acute gentler, demedicalized management of neonatal kernicterus in the 21st century. Unmonitored common signs of hyperbilirubinemia.6–8 These assertions excluded infants with newborn jaundice and hyperbilirubinemia (otherwise easily treated Crigler–Najjar syndrome who were healthy at birth but with phototherapy), unrecognized or untreated in a timely subsequently progressed to severe hyperbilirubinemia and some manner, can cause serious and often irreversible post-icteric sustained kernicterus. At present, documented cases of classic outcomes. Neonatal hyperbilirubinemia occurs in all newborn kernicterus in breast-fed infants discharged as healthy from their infants and may be recognized as jaundice in about 60% of healthy birth hospitals confirm that bilirubin can be neurotoxic in the newborns in the United States. Most infants are cared for at home absence of hemolytic disease.9,10 at the age of >72 h during the usual peak hyperbilirubinemia (age: 72 to 120 h). Jaundice and hyperbilirubinemia should be Question 2: Is there a re-emergence of kernicterus in the benign and resolved by the age of 2 to 3. Significant United States? 7,8 hyperbilirubinemia (>95th percentile for age in hours) occurs in Possibly, yes. Soon after publications of articles questioning the occurrence of kernicterus in term infants without hemolytic disease Correspondence: Dr VK Bhutani, Division of Neonatal and Developmental Medicine, (as discussed above), reports of kernicterus in term healthy Department of Pediatrics, Stanford University School of Medicine, Lucile Packard Children’s 9 Hospital, 750 Welch Ave no. 315; Stanford, CA 94304, USA. newborns appeared. Due to the absence of formal reporting for E-mail: [email protected] incidence of kernicterus or of its surrogate, extreme Kernicterus FAQ VK Bhutani and L Johnson S21 hyperbilirubinemia, it has been a challenge to gauge the extensively on exchange transfusion for total serum bilirubin magnitude of its reemergence. Clinical reports from the Pilot USA (TSB) levels >20 mg per 100 ml. Changing demographics, health- Kernicterus Registry presented in this supplement provide evidence care practices and waning experiences of infants with severe for the ‘tip of the iceberg’ of the adverse outcomes of unrecognized hemolytic diseases in the newborn characterized the 1990s. A and unmonitored neonatal hyperbilirubinemia.10 resurgence of kernicterus was reported in infants that were being cared for in a home environment, often with limited medical Question 3: Was kernicterus previously prevented in the supervision, during the first week after birth.2,4,10–12 United States? Probably, yes. Ip et al.5 compiled voluntary reports of kernicterus Question 4: What was the public health impact on the in the pediatric literature for the past 5 decades. We juxtaposed the 1994 AAP Guidelines? date of births of these infants to the prevalent and changing Implementation and surveillance strategies were not planned. No clinical practices in bilirubin management (Table 1). Prior to the formal public health evaluation of the impact of the 1994 practice introduction of exchange transfusion, severe neonatal parameters was initiated to track national incidence of kernicterus hyperbilirubinemia was generally due to Rh hemolytic disease and or its surrogate, extreme hyperbilirubinemia. To track the cases of perinatal sepsis, and often complicated with birth asphyxia. kernicterus encountered by colleagues, Dr Audrey K Brown initiated Kernicterus was one of the major pediatric problems in the early an informal voluntary registry of cases at the 1992 Kernicterus 20th century (prior to the 1950s). Over the ensuing five decades symposium (Pediatric Academic Societies). Evidence from this due to prenatal diagnosis of Rh isoimmunization and use of convenient sample led to a systems approach to manage newborn 10,12 exchange transfusions as neonatal intensive care was refined, jaundice. This inquiry of ‘numerators’ does not provide any evidence-based management of Rh disease was delineated. incidence or prevalence of kernicterus in the US population, rather Concurrently, the availability of Rhogam reduced the incidence of it provides insight into the occurrence for a condition where Rh isoimmunization and hemolysis. Advent of phototherapy and its nationally recommended guidelines either were not implemented liberal use in the United States in the late 1960s and early 1970s or failed to identify an infant with progressive severe led to the effective management of severe neonatal hyperbilirubinemia. hyperbilirubinemia.11 Unfortunately, evidence-based studies were not conducted to show its effectiveness in preventing kernicterus. Question 5: What is the current incidence of kernicterus The inherent ability of phototherapy to reduce bilirubin and severe neonatal hyperbilirubinemia? concentrations led to a widespread clinical practice for a staged Approximately 1 in 650 to 1000 infants, >35 weeks of gestation approach of using phototherapy prior to exchange transfusion. As can develop serum bilirubin values of 427 mmol lÀ1 (>25 mg per shown in Table 1, the least number of cases of kernicterus were 100 ml) and approximately 1 in 10 000 have levels of 513 mmol lÀ1 reported in the pediatric literature from 1973 to 1982. During this (>30 mg per 100 ml).13 There is a varying occurrence of time period, liberal phototherapy with white light devices was infants with total bilirubin levels >25 mg per 100 ml from readily adopted and implemented by pediatricians who relied diverse communities and practices (Table 2). There are few

Table 1 Reports of 150 cases of kernicterus in the US literature (1953–2001)

Years of reports Cases reported Average cases/year Prevalent health care practices for management of jaundice

1953–1962 15 1.5 Use of exchange transfusion for TSB levels X20 mg per 100 ml 1963–1972 17 1.7 Phototherapy introduced 1973–1982 (Index decade) 1 0.1 Phototherapy use becomes a clinical standard of care to prevent progression of TSB from 15 to 20 mg per 100 ml so as to prevent the need for an exchange transfusion 1983–1985 5 1.7 Vigintiphobia questioned 1986–1988 15 5 Evidence for bilirubin toxicity sought in reports of kernicterus in healthy babies 1989–1991 18 6 Kinder, gentler approach recommended 1992–1994 29 10 AAP Practice Guidelines in development 1995–1997 32 11 Ongoing implementation and distribution of AAP Practice Guideline (based on visual recognition of jaundice) 1998–2001 28 7 Increasing awareness of the re-emergence of kernicterus in the United States

Abbreviation: TSB, total serum bilirubin.

Journal of Perinatology Kernicterus FAQ VK Bhutani and L Johnson S22 population-based studies on the incidence of either acute bilirubin 100 ml was 1 in 4320 for a concurrent incidence of kernicterus as 1 encephalopathy and/or chronic bilirubin encephalopathy in 70 000. Thus, risk of kernicterus in infants with TSB >25 mg (kernicterus). Incidence of chronic kernicterus in North America per 100 ml in Denmark was estimated as 1 in 16.2 (70 000/4320). and Europe is listed in Table 3. In Denmark, there were no case The British and Danish studies define incidences of TSB X30 mg reports of kernicterus in the 20 years prior to Ebbesen’s report14 of per 100 ml as 1 in 14 084 and 1 in 10 000 live births and 8 cases between 1994 and 2002 (incidence of 1.4/100 000 live kernicterus incidence of 1 and 1.4/100 00 live births, respectively. births). Between 2002 and 2005, with a more vigilant approach to Thus, the risk of kernicterus in infants with TSB >30 mg per newborn jaundice management, no more cases have been seen in 100 ml is about one in seven.15–17 The risk is 1 in 5.5, assuming 2 Denmark, reducing the overall incidence to 1.1/100 000 live births of the 11 infants with TSB >30 mg per 100 ml ‘lost to follow-up’ in between 1994 and 2005.15 A 2004 UK surveillance study has the California data have kernicterus.24 None of these retrospective reported an occurrence of kernicterus in 1 in 100 000 live births.16 data provide any insight to signs of acute kernicterus in infants The frequency of catastrophic hyperbilirubinemia (serum bilirubin with TSB <25 mg per 100 ml. 513 mmol lÀ1) was 1 in 14 084 live births in the United Kingdom.16 A 2004 Canadian survey assessed the frequency of extreme Question 7: Is there a specific threshold TSB value hyperbilirubinemia (serum bilirubin >427 mmol lÀ1) as 1 in 2840 for bilirubin related neurotoxicity? live births, of which 13 (2/100 000 live births) had abnormal As discussed in a recent review, previous investigators have neurological outcomes at the time of discharge.17 Use of unsuccessfully attempted to correlate the occurrence of kernicterus 25 phototherapy and exchange transfusion during the development to a specific or threshold TSB level. In this supplement, the report and implementation of an institutional systems approach for from the Pilot USA Kernicterus Registry also fails to identify a newborn jaundice management is shown in Figure 1.18 Similarly, specific TSB level that is associated with the onset of bilirubin the use of phototherapy and exchange transfusion at a single- neurotoxicity. It appears that bilirubin levels of >35 mg per 100 ml À1 payor, regional health maintenance organization study is shown in (598 mmol l ) can have profoundly deleterious effect. Data are Table 4.19,20 These data from the United States compare to the needed to document clinical signs of acute kernicterus to a range Jerusalem experience with 1 in 4520 infants requiring an exchange of TSB levels and unbound bilirubin levels. transfusion, 1.9% requiring hospital-based intensive phototherapy and 0.41% being readmitted for phototherapy.21

Question 6: What is the estimated risk of kernicterus in infants Table 3 Reported incidence for chronic kernicterus with excessive hyperbilirubinemia? Study site Number of cases Incidence (live births) There has been an historic concern that untreated TSB X30 mg 16 per 100 ml was more likely to be associated with irreversible United Kingdom 7 1 in 100 000 Denmark14 8 1 in 70 000 kernicterus. We estimate the risk of kernicterus in infants with 17 extreme hyperbilirubinemia using contemporary population-based Canada 13 1 in 50 000 United States: health system23 6 1 in 27 000 data from Canada and Denmark. In Canada, the incidence of TSB 19 a United States: regional health 2 1 in 55 000 >25 mg per 100 ml is 1 in 2840 and the concurrent incidence of chronic kernicterus is 1 in 50 000. Thus, risk of kernicterus in The incidence of acute kernicterus (ABE) in the United States is estimated at 1 in 40 000 live births based on these data and the ability to prevent post-icteric sequelae with crash- infants with TSB >25 mg per 100 ml may be calculated as 50 000/ cart approach. These data need to be veriﬁed by prospective surveillance. 2840 or 1 in 17.6. In Denmark, the incidence of TSB >25 mg per aTwo cases reported as ‘lost to follow-up’ and assumed to have kernicterus.

Table 2 Frequency of TSB levels >25 mg per 100 ml in diverse communities with varying clinical practices

Regions Health practice style Years Frequency

United States (CA) Health system data;19 selective bilirubin screen by MD based on clinical observations 1994–1998 1 in 700 United States (UT) Health system data;22 universal bilirubin screen and clinician-based follow-up 2002 1 in 1522 United States (HCA) Health system data;23 universal bilirubin screen and clinician-based follow-up 2003 1 in 1878 Canada National survey data;17 selective bilirubin screen and national follow-up program 2002–2004 1 in 2840 Denmark National survey data;14 screening program and a national home follow-up. 1994–2002 1 in 4320 United States (PA) System-based; universal bilirubin screen and targeted follow-up18 1990–2003 1 in 15 000 Israel Universal screening program; clinical risk factors+selective bilirubin screening21 2001–2002 1 in 18 079

Abbreviation: TSB, total serum bilirubin.

Journal of Perinatology Kernicterus FAQ VK Bhutani and L Johnson S23

Exchange Exchange Exchange Exchange Exchange Exchange 6.00% 1: 1827 1: 11 995 1: 2317 1: 1322 1: 1632 1: 3198

5.00% Comparison

4.00%

3.00%

2.00% Rate of phototherapy use (%)

1.00%

3.65% 4.49% 5.44% 2.49% 4.01% 1.30% 0.00% 1990–1992 1993–1995 1996–1998 1999–2000 1990–2000 2001–2003 Systems Selective TSB Program development Study cohort Current approach

Figure 1 Rates of readmission for well babies pre-, during and subsequent to implementation of systems-based approach for newborn jaundice (from Bhutani et al.18).

Table 4 Comparison of two reports (Newman et al.) from the same population Soorani-Lunsing et al.28 prospectively observed a dose–response cohort on outcome of infants with extreme hyperbilirubinemia (TSB X25 and relationship increase in minor neurological dysfunction X 30 mg per 100 ml) throughout the first year of life to the degree of hyperbilirubinemia Reported 200620 Reported 200319 (13.6 to 25 mg per 100 ml; 233 to 444 mmol lÀ1). Are moderate degrees of hyperbilirubinemia really safe for the brain? This Sample population Infants with Infants with question remains unanswered. Minor neurological abnormalities TSBX25 mg per TSB X30 mg per 100 ml 100 ml have been described as abnormal postural behavior, high- Study period 1995–1998 1995–1998 frequency tremors and deviancies in muscle tone regulation at ages Cohort size 106 627 111 009 3 and 12 months. These may continue beyond puberty as muscle Study infants with TSB >25 mg per 140 F tone dysregulation, choreiform dyskinesia, manipulative disability, 100 ml coordination problems, gait abnormalities, awkwardness, tremors Study infants with TSB >30 mg per 10 11 or exaggerated extrapyramidal reflexes. Their correlation to 100 ml moderate or severe hyperbilirubinemia without concurrent signs of Use of exchange and phototherapy 5 5 acute bilirubin encephalopathy has been sought but not proven. Use of phototherapy only 131 8 Prospective long-term clinical studies are needed to allay these a No reported interventions 4 0 concerns. Signs of ABE 0 0 Number of infants followed a the 132/140 SIDS ¼ 1; Question 9: Can we define a bilirubin level that is safe? age of 2 years (94%) lost ¼ 2, speech Evidence needs to be determined. To assure a safe approach to therapy 1 ¼ manage newborn hyperbilirubinemia in a changing health-care Number of infants followed at the 82 (54%) None age of 5 years environment, optimal operational thresholds for intervention need to be delineated. Consensus and evidence-based guidelines have a Details of outcome or follow-up were not reported for this high-risk group. been established and need to be implemented.4 Clinical experience and practices indicate that kernicterus can almost always be prevented in infants >35 weeks of gestation. Our experience with Question 8: Are there sequelae of severe or prolonged moderate universal pre-discharge bilirubin screening in the well-baby hyperbilirubinemia in the absence of recognized acute nursery indicates that hour-specific TSB levels <40th percentile bilirubin encephalopathy? appear to be safe in terms of the magnitude of hyperbilirubinemia To date, no convincing evidence has been provided. Clinical and the potential risk for subsequent adverse outcome during the concerns remain but are unproven.26,27 In their article, first week after birth.18

Journal of Perinatology Kernicterus FAQ VK Bhutani and L Johnson S24

Question 10: Can we prevent kernicterus by a practical 12 Bhutani VK, Johnson L. Kernicterus: lessons for the future from a current tragedy. approach? NeoReviews 2003; 4: 30–32. In infants >35 weeks GA, kernicterus should be unacceptable. 13 Bhutani VK, Maisels MJ, Stark AR, Buonocore G. Management of jaundice and Prevention of kernicterus should be a public health goal. As the prevention of severe neonatal hyperbilirubinemia in infants X35 weeks gestation. 2004 guidelines12 is implemented, success for reduction in the Neonatology 2008; 94(1): 63–67. 14 Ebbesen F. Recurrence of kernicterus in term and near-term infants in Denmark. Acta incidence of severe hyperbilirubinemia and possibly kernicterus has Paediatr 2000; 89: 1213–1217. yet to be shown through national surveillance. 15 Ebbesen F, Andersson C, Verder H, Grytter C, Pedersen-Bjergaard L, Petersen JR et al. The ensuing articles and a report from the Pilot USA Kernicterus Extreme hyperbilirubinaemia in term and near-term infants in Denmark. Acta Registry, presented in this supplement, provide a current Paediatr 2005; 94: 59–64. perspective on these questions and inform strategies to transform 16 Manning DJ, Maxwell MJ, Todd PJ, Platt MJ. Prospective surveillance study of severe management of newborn jaundice. hyperbilirubinaemia in the newborn in the United Kingdom and Ireland. Arch Dis Child Fetal Neonatal Ed. Online publication. 0.1136/adc.2006.105361. Available at http://fn.bmj.com/cgi/rapidpdf/ adc.2006.105361v1Ebb. 17 Sgro M, Campbell D, Shah V. Incidence and causes of severe neonatal Disclosure hyperbilirubinemia in Canada. CMAJ 2006; 175: 587–590. The authors have declared no ﬁnancial interests. 18 Bhutani VK, Johnson LH, Schwoebel A, Gennaro S. A systems approach for neonatal hyperbilirubinemia in term and near-term newborns. J Obstet Gynecol Neonatal Nurs 2006; 35: 444–455. References 19 Newman TB, Liljestrand P, Jeremy RJ, Ferriero DM, Wu YW, Hudes ES et al. Jaundice and Infant Feeding Study Team: outcomes among newborns with total serum bilirubin 1 Bhutani VK, Johnson LH, Keren R. Diagnosis and management of hyperbilirubinemia levels of 25 mg per deciliter or more. N Engl J Med 2006; 354: 1889–1900. in the term neonate: for a safer ﬁrst week. Pediatr Clin North Am 2004; 51(4): 20 Newman TB, Xiong B, Gonzales VM, Escobar GJ. Prediction and prevention of extreme 843–861. 2 Bhutani VK, Johnson LH, Maisels MJ, Newman TB, Phibbs C, Stark AR et al. neonatal hyperbilirubinemia in a mature health maintenance organization. Arch Kernicterus: epidemiological strategies for its prevention through systems-based Pediatr Adolesc Med 2000; 154: 1140–1147. approaches. J Perinatol 2004; 24(10): 650–662. 21 Kaplan M, Bromiker R, Schimmel MS, Algur N, Hammerman C. Evaluation of 3 American Academy of Pediatrics, Provisional Committee for Quality Improvement. discharge management in the prediction of hyperbilirubinemia: the Jerusalem Practice parameter: management of hyperbilirubinemia in the healthy term newborn. experience. J Pediatr 2007; 150: 412–417. Pediatrics 1994; 94: 558–565. 22 Eggert LD, Wiedmeier SE, Wilson J, Christensen RD. The effect of instituting a 4 American Academy of Pediatrics, Subcommittee on Hyperbilirubinemia, Maisels MJ, prehospital-discharge newborn bilirubin screening program in an 18-hospital health Baltz RD, Bhutani VK, Newman TB, Palmer H, Rosenfeld W et al. Clinical Practice system. Pediatrics 2006; 117: e855–e862. Guideline: management of hyperbilirubinemia in the newborn infant 35 weeks of 23 Bhutani VK, Rodriguez M. HCA Perinatal Safety Initiative Kernicterus Taskforce. Recent gestation. Pediatrics 2004; 114: 297–316. Available at http://www.aappolicy. epidemiologic indices of severe hyperbilirubinemia and kernicterus for a diverse multi- aappublications.org/cgi/content/full/pediatrics; 114/1/297. state newborn population in USA. http://www.abstracts2view.com/pasall/.E-PAS 5 Ip S, Chung M, Kulig J, O’Brien R, Sege R, Glicken S, et al., Subcommittee on 2005:1225. Hyperbilirubinemia. An evidence-based review of important issues concerning neonatal 24 Newman TB, Liljestrand P, Escobar GJ. Infants with bilirubin levels of 30 mg/dl or hyperbilirubinemia. Pediatrics 2004; 114: e130–e153. more in a large managed care organization. Pediatrics 2003; 111(6 Part 1): 6 Newman TB, Maisels MJ. Bilirubin and brain damage: what do we do now? Pediatrics 1303–1311. 1989; 83(6): 1062–1065. 25 Wennberg RP, Ahlfors CE, Bhutani VK, Johnson LH, Shapiro SM. Toward 7 Newman TB, Maisels MJ. Does hyperbilirubinemia damage the brain of healthy understanding kernicterus: a challenge to improve the management of jaundiced full-term infants? Clin Perinatol 1990; 17(2): 331–358. newborns. Pediatrics 2006; 117: 474–485. 8 Newman TB, Maisels MJ. Evaluation and treatment of jaundice in the term newborn: a 26 Newman TB, Klebanoff MA. Neonatal hyperbilirubinemia and long-term outcome: kinder, gentler approach. Pediatrics 1992; 89(5 Part 1): 809–818. another look at the Collaborative Perinatal Project. Pediatrics 1993; 92(5): 651–657. 9 Penn AA, Enzmann DR, Hahn JS, Stevenson DK. Kernicterus in a full term infant. 27 Bhutani VK. Neonatal hyperbilirubinemia and the potential risk of subtle neurological Pediatrics 1994; 93: 1003–1006. dysfunction. Pediatr Res 2001; 50(6): 679–680. 10 Johnson LH, Brown AK, Bhutani VK. System-based approach to management of 28 Soorani-Lunsing I, Woltil HA, Hadders-Algra M. Are moderate degrees of neonatal jaundice and prevention of kernicterus. J Pediatr 2002; 140: 396–403. hyperbilirubinemia in healthy term neonates really safe for the brain? Pediatr Res 11 Philip AG. The evolution of neonatology. Pediatr Res 2005; 58: 799–815. 2001; 50(6): 701–705.

Journal of Perinatology