The Case

A full-term neonate was delivered uneventfully to an Rh-negative woman who had received RhoGAM (Rho[D] immunoglobulin treatment) at 28 weeks. The neonate developed hyperbilirubinemia within 24 hours of delivery, which prompted initiation of aggressive phototherapy. He also received two blood transfusions and two doses of intravenous immunoglobulin in the neonatal intensive care unit. These therapies were given to prevent the need for a more risky exchange transfusion. On discharge at day-of-life (DOL) 10, hemoglobin and bilirubin levels were stable. The treatment plan was for regular blood draws to assess the ongoing stability of hemoglobin and bilirubin levels.

Over 3 visits in the subsequent 10 days, the patient was seen by different providers who all noted that the hemoglobin remained stable at 12 g/dL, and the bilirubin continued to fall. These reassuring trends were noted in the record. On DOL 25, the bilirubin maintained its downward trend, but the hemoglobin was also noted to have dropped to 10 g/dL. Given the bilirubin was decreasing, the providers were reassured and no additional lab draws were scheduled.

Two days later, the family brought the infant back to the urgent care clinic with unrelated concerns about redness around the umbilicus. Noting the reduction in hemoglobin from the last visit, a repeat hemoglobin test was ordered that came back even lower: 7.7 g/dL. The clinicians now recognized that the infant had active Rh hemolytic disease, and he was admitted to the hospital for blood transfusion and close monitoring. Due to the readmission and the family's frustration with the inconsistent care provided since the first discharge, the case was reviewed by the pediatric quality committee. While many communication concerns were identified, the key issue was the management of an uncommon disease using markers that are frequently trended for common conditions. The patient improved and he ultimately did well with appropriate therapies.

The Commentary

Causes and importance of neonatal jaundice

About 80% of normal term newborns become jaundiced.(1,2) Most of the time this is a benign occurrence, but rarely (about 1/100,000 births [3]) it can cause irreversible brain damage, known as kernicterus. Therefore, ensuring sufficient follow-up and treatment to prevent kernicterus without excessive visits, blood tests, and treatments is challenging.

Jaundice is the visible manifestation of hyperbilirubinemia, which can be caused by higher bilirubin production, decreased excretion, increased enterohepatic circulation, or some combination of the three. Newborns have raised bilirubin production compared with adults because of higher hematocrits and increased red cell destruction. This can be severe if there is a pathologic cause of hemolysis, like Rh isoimmunization in which maternal antibodies to fetal red cell antigens cross the placenta and cause hemolysis. Isoimmunization tends to present early because of red cell destruction and increased bilirubin production before birth. In contrast, hyperbilirubinemia due to hemolysis from glucose-6-phosphate dehydrogenase (G6PD) deficiency or from bruising or cephalhematoma is more likely to present later.

Decreased excretion of bilirubin in newborns occurs because it takes time to increase the activity of uridine diphosphate (UDP) glucuronosyltransferase, the enzyme responsible for conjugating bilirubin in the liver. Prenatally, fetal bilirubin crosses the placenta and is conjugated by the mother's liver and excreted in her stool. Thus, newborn bilirubin levels typically rise for a few days until the liver's conjugating ability catches up with and then exceeds bilirubin production, at which point levels typically decline. The process takes longer in less mature infants, leading to higher and later peak bilirubin levels with decreasing gestational age. Generally, once the bilirubin peaks and begins to decline on its own, it continues to decline unless there is some new source of bilirubin production. Genetic factors can also contribute to decreased bilirubin excretion. Notable among these is Gilbert syndrome, a common genetic variant in UDP glucuronosyltransferase activity, which causes benign, mild jaundice in adults and can contribute to elevated bilirubin levels in newborns, particularly if they have G6PD deficiency.(4)

Finally, hyperbilirubinemia may be due in part to raised enterohepatic circulation, in which bilirubin that has been conjugated in the liver is unconjugated by beta-glucuronidase in the newborn intestine and, now lipid soluble again, is reabsorbed into the portal circulation, requiring it to be conjugated once more. Anything that leads to decreased stooling will tend to increase this enterohepatic circulation; most often we see this in breastfed infants who are not yet getting much milk.

Management approach

Because neonatal bilirubin levels typically peak at 3–7 days, when most term newborns are at home, ensuring that the levels stay in a safe range is a challenge. The first step is pre-discharge risk assessment.(2,5) By assessing risk factors (especially prematurity and breastfeeding) and a pre-discharge bilirubin level (obtained from blood or transcutaneously), the second step, appropriate follow-up, can be tailored to the infant's risk of developing significant hyperbilirubinemia. Most infants should be seen 1–2 days after discharge; an online tool (www.bilitool.org) can help with interpretation of age-specific bilirubin levels. Follow-up should include assessment of bilirubin levels, breastfeeding, and hydration and should be with providers able to support breastfeeding. The third step is judicious treatment with phototherapy (or, in rare cases, exchange transfusion) when indicated. The American Academy of Pediatrics (AAP) guidelines are available on the Web. It is worth appreciating that phototherapy thresholds provided by the AAP are based on expert opinion rather than good data that the treatment thresholds represent bilirubin levels at which the benefits of treatment exceed risks and costs. In fact, the AAP guidelines are internally inconsistent (6) and probably err on the side of recommending too much treatment, a result of being written by people (including us) with extensive contact with kernicterus cases.(7,8)

Common pitfalls

Common pitfalls in the management of jaundice involve failure at any of the steps above: risk assessment, follow-up, and treatment. In addition, although it is common in patient safety discussions to focus on errors of omission (9), a common error in managing neonatal jaundice is overtreatment—i.e., treating newborns at bilirubin levels below those at which it is recommended. This leads to unnecessary separation of mother and infant and extra costs. Also, phototherapy has been associated with delayed adverse effects in some studies.(10-14) Thus, for many infants close to the AAP phototherapy threshold, a follow-up bilirubin level in 4–24 hours should be considered as an alternative to immediate phototherapy.

Current case

The current case is unusual for many reasons. First, maternal blood typing and use of Rho(D) immunoglobulin have made maternal Rh isoimmunization rare. One wonders whether the first (and arguably most important) error in this case was failure to provide that treatment at the time of an earlier pregnancy or miscarriage. Second, there appears to have been a sudden increase in hemolysis or decrease in bone marrow compensation around day 25. With Rh isoimmunization, maternal antibody can persist for months, and can continue to cause hemolysis, sometimes requiring additional transfusions, as was the case here. However, when these infants receive either exchange transfusion or simple transfusion, they are transfused with Rh-negative blood. Thus, the late anemia occurs due to ongoing hemolysis of the remaining Rh+ cells and because the transfused Rh-negative red cells gradually die and the Rh+ reticulocytes with which the bone marrow tries to replace them may be destroyed by maternal antibody. This typically leads to a gradually worsening anemia, rather than a stable hemoglobin for 10 days followed by a more sudden decrease in hemoglobin of 4.3 g/dL (from 12 to 7.7 g/dL) over 7 days, as in this case. We would not expect clinicians to be alarmed at a drop from a hemoglobin of 12 g/dL from 10–20 days to 10 g/dL at 25 days because hemoglobin levels in newborns commonly drop by about 5 g/dL over the first 4 weeks.(15) Although it was an error to send the baby home on day 25 with no additional laboratory follow-up, we do not believe a follow-up hemoglobin within 2 days (which would have been necessary to prevent the outcome in this case) was clearly indicated. Families of infants at risk for late anemia should be counseled about symptoms of anemia (pallor, poor feeding, lethargy, respiratory distress), and weekly follow-up is generally sufficient. We wonder if part of the family's dissatisfaction in this case (which had a fine outcome) was related to mixed messages about the need for follow-up and the significance of the hemoglobin of 10 g/dL.

Take-Home Points

• Rh-typing of all pregnant women and administration of Rh(D) immunoglobulin to Rh-negative women prevents most cases of severe Rh disease.
• The risk of developing significant jaundice should be assessed in every newborn before discharge. A transcutaneous or blood bilirubin level can help; a follow-up visit 1–2 days after discharge will generally be indicated.
• Available guidelines for phototherapy and exchange transfusion are conservative; treating below guideline levels is not recommended.
• Isoimmunization requires education of families and continued follow-up for late anemia.

Thomas B. Newman, MD, MPH Professor of Epidemiology and Biostatistics and Pediatrics University of California, San Francisco

M. Jeffrey Maisels, MB, BCh, DSc Professor, Oakland University, William Beaumont School of Medicine Royal Oak, MI

References

1. Keren R, Luan X, Friedman S, Saddlemire S, Cnaan A, Bhutani VK. A comparison of alternative risk-assessment strategies for predicting significant neonatal hyperbilirubinemia in term and near-term infants. Pediatrics. 2008;121:e170-e179. [go to PubMed]

2. Maisels MJ, Bhutani VK, Bogen D, Newman TB, Stark AR, Watchko JF. Hyperbilirubinemia in the newborn infant > or =35 weeks' gestation: an update with clarifications. Pediatrics. 2009;124:1193-1198. [go to PubMed]

3. Burgos AE, Flaherman VJ, Newman TB. Screening and follow-up for neonatal hyperbilirubinemia: a review. Clin Pediatr (Phila). 2012;51:7-16. [go to PubMed]

4. Kaplan M, Renbaum P, Levy-Lahad E, Hammerman C, Lahad A, Beutler E. Gilbert syndrome and glucose-6-phosphate dehydrogenase deficiency: a dose-dependent genetic interaction crucial to neonatal hyperbilirubinemia. Proc Natl Acad Sci U S A. 1997;94:12128-12132. [go to PubMed]

5. Maisels MJ, Baltz RD, Bhutani VK, et al. Management of hyperbilirubinemia in the newborn infant 35 or more weeks of gestation. Pediatrics. 2004;114:297-316. [go to PubMed]

6. Newman TB, Kuzniewicz MW, Liljestrand P, Wi S, McCulloch C, Escobar GJ. Numbers needed to treat with phototherapy according to American Academy of Pediatrics guidelines. Pediatrics. 2009;123:1352-1359. [go to PubMed]

7. Newman TB, Maisels MJ. Less aggressive treatment of neonatal jaundice and reports of kernicterus: lessons about practice guidelines. Pediatrics. 2000;105:242-245. [go to PubMed]

8. Newman TB. Universal bilirubin screening, guidelines, and evidence. Pediatrics. 2009;124:1199-1202. [go to PubMed]

9. Schroeder AR, Harris SJ, Newman TB. Safely doing less: a missing component of the patient safety dialogue. Pediatrics. 2011;128:e1596-e1597. [go to PubMed]

10. Aspberg S, Dahlquist G, Kahan T, Källén B. Is neonatal phototherapy associated with an increased risk for hospitalized childhood bronchial asthma? Pediatr Allergy Immunol. 2007;18:313-319. [go to PubMed]

11. Aspberg S, Dahlquist G, Kahan T, Källén B. Confirmed association between neonatal phototherapy or neonatal icterus and risk of childhood asthma. Pediatr Allergy Immunol. 2010;21:e733-e739. [go to PubMed]

12. Dahlquist GG, Patterson C, Soltesz G. Perinatal risk factors for childhood type 1 diabetes in Europe. The EURODIAB Substudy 2 Study Group. Diabetes Care. 1999;22:1698-1702. [go to PubMed]

13. Dahlquist G, Källén B. Indications that phototherapy is a risk factor for insulin-dependent diabetes. Diabetes Care. 2003;26:247-248. [go to PubMed]

14. Cnattingius S, Zack M, Ekbom A, Gunnarskog J, Linet M, Adami HO. Prenatal and neonatal risk factors for childhood myeloid leukemia. Cancer Epidemiol Biomarkers Prev. 1995;4:441-445. [go to PubMed]

15. Jopling J, Henry E, Wiedmeier SE, Christensen RD. Reference ranges for hematocrit and blood hemoglobin concentration during the neonatal period: data from a multihospital health care system. Pediatrics. 2009;123:e333-e337. [go to PubMed]