<p>Visible jaundice, a common finding in newborn infants of all gestational ages, is an almost universal feature of prematurity. Preterm (PT) and full term (FT) infants become jaundiced by similar mechanisms. What makes the difference (if at all any) is the exaggeration of the processes that lead to jaundice. Early application of phototherapy is the standard approach to management in most PT nurseries, but the actual risk to the central nervous system (CNS) from plasma unconjugated bilirubin (UB) and the need for early preventive treatment of physiologic bilirubin levels in tiny PT infants are based more on anecdote than on sound epidemiology or adequate follow-up. Although unconjugatedhyperbilirubinemia is universally found in PT infants, evidence to show a causal relationship between UB levels and neurological, visual, speech, hearing and developmental disorders is inadequate. This article will review the evidences pertaining to this issue and suggest a management guideline in PT infants.</p></sec><sec id="sec002"><title>What are the controversies about this issue?

Journal of Neonatology

0971-2179

National Neonatology Forum

jn-21-3-005

REVIEW ARTICLE

Hyperbilirubinemia in preterm neonates — Management guidelines

Venkataseshan

Department of Pediatrics, Advanced Pediatrics Centre, PGIMER, Chandigarh. E-mail: sesh6@rediffmail.com

7–92007

213170174

<p>Visible jaundice, a common finding in newborn infants of all gestational ages, is an almost universal feature of prematurity. Preterm (PT) and full term (FT) infants become jaundiced by similar mechanisms. What makes the difference (if at all any) is the exaggeration of the processes that lead to jaundice. Early application of phototherapy is the standard approach to management in most PT nurseries, but the actual risk to the central nervous system (CNS) from plasma unconjugated bilirubin (UB) and the need for early preventive treatment of physiologic bilirubin levels in tiny PT infants are based more on anecdote than on sound epidemiology or adequate follow-up. Although unconjugatedhyperbilirubinemia is universally found in PT infants, evidence to show a causal relationship between UB levels and neurological, visual, speech, hearing and developmental disorders is inadequate. This article will review the evidences pertaining to this issue and suggest a management guideline in PT infants.</p></sec><sec id="sec002"><title>What are the controversies about this issue?

1) Are PT infants at higher risk of developing more severe hyperbilirubinemia in comparison to FT infants?

2) Are PT infants at a greater risk of developing bilirubin induced neurological damage (BIND)?

3) What are the risk factors that predispose a PT infant to develop BIND?

4) What is the tolerable level of total serum bilirubin (TSB) beyond which the risk of BIND increases significantly? When should the phototherapy be initiated?

5) Is there any association between peak TSB levels and later adverse outcomes in very low birth weight (VLBW) infants?

6) Can we predict the risk of kernicterus in these infants?

7) Current status of prophylactic therapies in hyperbilirubinemia of PT infants?

Do preterm infants develop more severe hyperbilirubinemia?

There is a strong physiological basis to suggest that in PT infants, hyperbilirubinemia is more prevalent, more severe and its course more protracted than its FT counterpart. This is due to certain deficiencies in bilirubin handling:

a) A shorter half life of red cells and an apparent tendency towards low-grade hemolysis in some VLBW infants.

b) Delayed feeding (due to illness/gut immaturity), poor/delayed bacterial intestinal colonization and an exaggerated entero-hepatic circulation

c) A lower serum albumin concentration¹ & reduced bilirubin binding capacity

d) Slower postnatal maturation of hepatic acceptor proteins (i.e., ligandins) and conjugating enzyme (i.e., glucuronyl transferase)

e) Bruising, physiologic instability, and risks related to treatments

Are preterm infants at an increased risk of BIND?

The clinical concept of ‘low bilirubin kernicterus’ suggests that PT infants are at greater risk for BIND. Autopsy findings of deceased newborns with bilirubin levels < 20 mg/dl made clinicians think that BIND can occur at levels of bilirubin once thought to be safe². Clinical and basic studies on bilirubin handling showed that plasma concentrations of albumin, the main protein carrier of bilirubin in the extracellular fluid, was lower in PT than in FT and in sick than in well newborns¹. Results of in vitro tests of albuminbinding capacity for bilirubin and free bilirubin (FB) concentrations in neonatal plasma suggested that distressed infants had decreased binding and carrying capacity for bilirubin in the plasma and extracellular space, increasing the risk that FB could enter and damage the CNS³. But this clinical concept has undergone extensive reviews and criticisms. Reviews of large case series and retrospective studies did not reveal a consistent pattern of relation between hazardous bilirubin concentrations and birth weight or gestation⁴. Postkernicteric bilirubin encephalopathy and central deafness related to TSB concentration have not emerged as important clinical outcomes in neonatal follow-up clinics for LBW high-risk infants suggesting that past autopsy findings of bilirubin staining may have been agonal or postmortem⁵.

What are the risk factors that predispose them to BIND?

During the low bilirubin kernicterus era (mid 1960s and early 1970s), in response to a series of autopsy observations ^6,7,8,9 (Table 1), Jerold Lucey suggested that it is possible to identify the premature infant who is particularly susceptible to death from kernicterus and gave a list of putative risk factors (Table 2)¹⁰. He suggested that pediatricians use phototherapy aggressively to prevent hyperbilirubinemia (>10 mg/dl). But later in the 1980s, two important studies casted a serious doubt on the validity of the suggested risk factors^11,12. These studies were unable to identify any risk factor or group of factors (including those given by Lucey) that was associated with development of kernicterus in the premature neonate. Various factors like hypoxemia, acidosis, hypercarbia, drugs displacing bilirubin from albumin binding sites have been shown to increase the risk of bilirubin entry in to the brain either by increasing cerebral blood flow or by damaging the blood — brain barrier. These risk factors are more commonly encountered in PT sick infants. Hence, until further information is available, it makes sense to take a more aggressive approach to maintain low bilirubin levels in this population Table 1

Low bilirubin kernicterus studies: Serum bilirubin concentrations at which kernicterus was noted and mean birth weight of newborns.

Study (year)	Concentrations (mg/dl)	Mean Birth weight (g)

Stern and Denton⁶ (1965)	17.1–18.8	1170
Gartner et al⁷ (1970)	9.4–15.6	998
Ackerman et al⁸ (1970)	18.5–18.8	1004
Kennan et al⁹ (1972)	10.5–17.8	850

Table 2

Putative risk factors to identify premature newborns at risk of developing Kernicterus¹¹

Birth weight <1500 g

Hypothermia

Asphyxia

Acidosis

Hypoalbuminemia

Sepsis

Meningitis

Drugs that affect albumin binding

Serum bilirubin >10 mg/dl

Can serum bilirubin predict long-term neurodevelopment outcome?

The total serum bilirubin (TSB) concentration is an unreliable predictor of BIND in premature infants^13,14,15. One earlier study had shown a relation between maximal TSB concentrations in the neonatal period and cerebral palsy at a corrected age of 2 years but there was subsequently no difference at 5 years of age. More recently, peak TSB levels were associated with increased risk of death, hearing loss, and neurodevelopmental impairment in ELBW infants born between 1994 and 1997 but the effect sizes were small to make any definitive conclusions¹⁶. Significant associations between peak TSB levels and hearing loss has been documented in LBW infants, and confirmed recently^16,17. Similarly, Intelligence Quotient (IQ) was not associated with mean bilirubin level, time and duration of exposure to bilirubin, or measures of bilirubin-albumin binding¹⁸. Several investigators have appropriately questioned the role of TSB value as an accurate index for risk of bilirubin toxicity in this group of infants. Rises in FB have been associated with kernicterus in sick, PT infants¹⁹. In addition, raised FB concentrations are more closely associated than TSB levels with transient abnormalities in the audiometric brain stem response in both term and preterm babies²⁰. However, there are no contemporary long-term studies relating FB concentrations in LBW infants to developmental outcome. Nevertheless, from a clinical perspective, in the absence of better predictive indices, TSB levels, define the thresholds of concern.

When should treatment be initiated?

The aim of treatment is to prevent bilirubin related neurological complications while not causing harm. But framing guidelines for phototherapy and exchange transfusion (ET) in these infants is a capricious exercise and no claim of an evidence base can be made. Various guidelines have been suggested by different experts stratified for birth weight²¹ and gestational age²² and none of them can claim for a greater validity of one approach over the other. Most of the algorithms do not take into account the lower concentration of albumin and its diminished ability to bind bilirubin especially in the first 72 hours after birth in the sick and preterm infants.

The more widely accepted “free bilirubin theory” demands an important role for serum albumin. FB concentration is a function of serum albumin concentration and total bilirubin concentration. FB concentration increases as the B/A ratio increases. Hence it makes sense to include serum albumin concentration in the criteria for treating hyperbilirubinemia especially in PT infants. In addition, serum albumin concentration may contribute to the limited predictive ability of TSB for BIND as higher concentrations of albumin draw more of the miscible pool of bilirubin into the vascular space, raising the TSB concentration but reducing the risk of BIND^23,24. Given the unreliability of TSB to predict the risk of BIND and long term outcome, and the lack of easily available assays to measure FB levels, bilirubin/albumin ratio (B:A ratio) can be a reliable surrogate for FB. Previous guidelines in this respect have included a single value of serum albumin as cut off in the management criteria²⁵. But this introduces an element of ambiguity in management which can effectively be eliminated while retaining an albumin criterion by using bilirubin: albumin ratio (B/A ratio) ²⁶. The proposed guideline based on the above evidence is given in Table 3 This guideline is sickness and weight specific. Short of well conducted, prospective studies in PT infants to identify the safe level of bilirubin, this guideline should be an acceptable alternative inspite of the drawback of not being hour specific. Table 3

Suggested guidelines for exchange transfusion in low birth weight infants.

	<1250 g	1250–1499 g	1500–1999 g	2000–2499 g

Standard risk
TSB	13	15	17	18
B/A ratio	5.2	6.0	6.8	7.2
High risk^*
TSB	10	13	15	17
B/A ratio	4.0	5.2	6.0	6.8

* Risk factors: Apgar <3 at five minutes; PaO2 <40 mmHg for ≥ 2hours; pH<7.15 for ≥ 1 hour; birth weight <1000 g; evidence of hemolysis; clinical or central nervous system deterioration; total protein ≤ 4 g/dl or albumin ≤ 2.5 g/dl.
• TSB: Total serum bilirubin; B/A ratio: Bilirubin/Albumin ratio.
• Consider starting phototherapy at TSB 5 mg/dl less than the cut off for exchange transfusion; “Failed phototherapy” — No reduction or reduction <1 mg/dl in 4–6 hours.

How to predict the risk of BIND?

The risk of bilirubin encephalopathy is not just a function of TSB level or FB concentration alone but is dependent on total amount of bilirubin available (miscible pool of bilirubin) as well as the tendency of bilirubin to enter the tissues (FB concentration). Additionally, susceptibility of the cells of the CNS to damage by bilirubin (genetic polymorphisms in neuronal bilirubin oxidase function) is also responsible for the differential vulnerability.

Prediction of this risk is a tremendous task and has been attempted by various authors in the past.

a) Early clinical features: Clinical signs progress through three identifiable stages (early reversible, intermediate, and advanced)²⁷. But in VLBW and ELBW infants, progression of these signs are usually masked by the concurrent illnesses and hence are of little clinical use. However, a few studies have shown an association between apnea in the first 2 weeks of life and brain stem dysfunction (by Auditory brain stem evoked response - ABR)^28,29

b) Total serum bilirubin: Though imprecise, TSB levels still remains the only currently available clinical risk factor test for kernicterus. But most of the algorithms do not take into consideration, the rate of rise in TSB. The TSB percentile tracks (at the 95th, 75th, 40th and 25th) can be used as a preliminary surrogate to define the peak TSB and its rate of rise (Figure 1). For example, at about 36 weeks gestation, a rate of TSB rise at 0.20 mg/dl, represented by the 95th percentile track could be considered excessive hyperbilirubinemia. At 32 weeks gestation a rate of TSB rise at 0.10 mg/dl, represented by the 40th percentile track may be considered excessive³⁰.

c) Free bilirubin: Most of the published studies have suggested FB as the most sensitive and specific biochemical measure to evaluate PT infants with hyperbilirubinemia^20,31,32. Concentrations > 0.5 μg/dl have the maximum sensitivity and specificity (70% and 75% respectively) for bilirubin induced auditory dysfunction (RR 2.45, 95% CI 1.33–4.49)²⁰. Studies quoted by Poland have suggested that unbound bilirubin concentrations greater than 0.8 ìg/dL are associated with an increasing risk of BIND. A study from Northern India in term neonates with severe non-hemolytic jaundice (>18 mg/dl) has shown a significant association between high FB concentrations and kernicterus (p=0.006)³³. But in this study no difference was observed in albumin levels between the two groups, which probably in turn led to a borderline significance of bilirubin/albumin ratio (p=0.05). Short of FB assays, B/A ratio seem to be an acceptable surrogate marker for risk assessment.

d) Auditory brainstem evoked response: ABR has been shown to vary as a function of bilirubin and a sensitive technique for bilirubin induced auditory neurotoxicity. A temporal relation between the peak TSB levels and abnormal progression in ABR has been suggested³⁴

Bilirubin toxicity in the late preterm infant: Late preterm gestation is one of the most prevalent identified risk factors for the development of severe hyperbilirubinemia (TSB ≥ 20mg/dl) and kernicterus^35,36. These infants are relatively mature enough to maintain their body temperature, have a good suck-swallow reflex and respiratory drive and hence are often managed like their term counterparts. But in contrast, late preterm infants are relatively immature in their capacity to handle unconjugated bilirubin. Risk of developing a TSB >20mg/dl is eight fold increased in infants born at 36 weeks gestation³⁵. Late PT infants have a similar degree of heme production as their term counterparts but suffer from hepatic immaturity which leads to immature uptake and conjugation of bilirubin. A unitless index (bilirubin production-conjugation index) created by Kaplan and colleagues shows this imbalance. This index is given by the numeric ratio of the blood carboxyhemoglobin corrected for ambient carbon monoxide and the serum total conjugated bilirubin³⁷. Kaplan and colleagues have shown that the production-conjugation index was significantly high in ≤36 weeks gestation in comparison with normal term newborns (2.31 vs. 1.05, p=0.003)³⁸. Late PT infants are at a higher risk for kernicterus compared with their term counterparts³⁹. They also show signs of bilirubin encephalopathy at an earlier postnatal age than term newborns suggesting a greater vulnerability³⁶. The mechanisms that could potentially account for an increased susceptibility to BIND in late PT infants have not been defined.

Various clinical risk factors for marked hyperbilirubinemia and kernicterus have been identified in late PT neonates (Table 4). Of these, inadequate breast milk feeding has been identified most consistently. Inadequate feeding, in addition to contributing to varying degrees of dehydration, can enhance hyperbilirubinemia by increasing the enterohepatic circulation. Table 4

Clinical risk factors for marked hyperbilirubinemia and/or kernicterus in late preterm neonates

1	Breast milk feeding^36,39
2	Large for gestational age status³⁶
3	Male sex^36,39
4	Glucose-6-phosphate dehydrogenase deficiency and
5	Inadequate breast feeding

Treatment guideline in late PT neonates: The current guidelines of AAP subcommittee on hyperbilirubinemia has given hour, gestational age and risk specific guidelines for phototherapy and exchange transfusion treatment thresholds stratified in to three groups⁴⁰. Late PT infants are grouped under medium risk or high risk based on the presence or absence of risk factors. But the management of late PT infants between 34^0/7 and 34^6/7 weeks is not addressed in this guideline. Management of this group is akin to guidelines for PT infants (Table 3).

Role of Phenobarbitone prophylaxis: Phenobarbitone has been evaluated as a prophylactic agent for neonatal jaundice. Most of these studies were done when use of phototherapy was not common^41,42. The only RCT from our country in this regard compared phenobarbitone in different dosage regimens with no phenobarbitone in preterm infants with birth weight 1000–1499 grams⁴³. Phenobarbitone in the dose of 10mg/kg given within 6 hours of life followed by 5mg/kg/day for 5 days by intravenous route significantly decreased the need for exchange transfusion and duration of phototherapy in this group of babies. The authors quoted that the amount of phenobarbitone given as first dose within 6 hours of life seemed to be crucial for its action. Further studies are warranted specially targeting ELBW infants before recommending routine phenobarbitone prophylaxis.

Role of hydration/supplemental fluids: Fluid supplementation may play a role in jaundiced babies with evidence of dehydration. Dilution of a well hydrated infant's serum to reduce total bilirubin level is not supported by any data as of now. Recently, a RCT, conducted in term infants with severe non-hemolytic hyperbilirubinemia (TSB >18mg/dl) to determine the efficacy of fluid supplementation has shown a significant decrease in the need for exchange transfusion (p=0.001) and a shorter duration of phototherapy (p=0.004) in the extra fluids group⁴⁴. No recommendation can be made in PT infants as no studies are available in this group.

Conclusion

Preterm infants are at a higher risk of suffering from severe hyperbilirubinemia

The safe level of bilirubin in this group of infants is yet to be defined

Bilirubin/Albumin ratio is a useful surrogate marker for free bilirubin and hence should be included in management decisions

References1Cashore

Horwich

Karotkin

: Influence of gestational age and clinical status on bilirubin binding capacity in newborn infants determined by Sephadex G-25 gel filtration technique. Am J Dis Child. 1977; 131: 898.2Ahdab-Barmada

Moosy

: The neuropathology of kernicterus in the premature neonate: Diagnostic problems. J Neuropathol Exp Neurol. 1984; 43: 45.3Cashore

Brodersen

: Reserve albumin and bilirubin toxicity index in infant serum. Acta Pediatr. 1979; 72: 415.4Watchko

Claasen

: Kernicterus in premature infants: Current prevalence and relationship to NICHD phototherapy study exchange criteria. Pediatrics. 1994; 93: 996.5Turkel

Miller

Guttenberg

: A clinical pathologic re-appraisal of kernicterus. Pediatrics. 1982; 69: 267.6

Stern

Denton

Kernicterus in small premature infants. Pediatrics. 1965; 35: 483–485.7

Gartner

Snyder

Chabon

Bernstein

Kernicterus: high incidence in premature infants with low serum bilirubin concentrations. Pediatrics. 1970; 45: 906–917.8

Ackerman

Dyer

Lejeydorf

Hyperbilirubinemia and kernicterus in small premature infants. Pediatrics. 1970; 45: 918–925.9

Kennan

Perlstein

Light

Sutherland

Kernicterus in small, sick, premature infants receiving phototherapy. Pediatrics. 1972; 49: 652–655.10

Lucey

The unsolved problem of kernicterus in the susceptible low birth weight infant. Pediatrics. 1972; 49: 646–651.11

Turkel

Guttenberg

Moynes

Hodgman

Lack of identifiable risk factors for kernicterus. Pediatrics. 1980; 66: 502–506.12

Kim

Yoon

Sher

Brown

Lack of predictive indices in kernicterus: a comparison of clinical and pathological factors in infants with or without kernicterus. Pediatrics. 1980; 66: 852–858.13

van de Bor

Ens-Dokkum

Schreuder

Hyperbilirubinemia in low birth weight infants and outcome at 5 years of age. Pediatrics. 1992; 89: 359–364.14

Yeo

Perlman

Hao

Outcomes of extremely premature infants related to their peak serum bilirubin concentrations and exposure to phototherapy. Pediatrics. 1998; 102: 1426–1431.15

O’Shea

Dillard

Klinepeter

Serum bilirubin levels, intracranial hemorrhage, and the risk of developmental problems in very low birth weight infants. Pediatrics. 1992; 90: 888–892.16

Tyson

Fanaroff

Association between peak serum bilirubin and neurodevelopmental outcomes in extremely low birth weight infants. Pediatrics. 2003; 112: 773–779.17

Hack

Wilson-Costello

Friedman

Neurodevelopment and predictors of outcomes of children with birth weights less than 1000g, 1992–1995, Arch Pediatr Adolesc Med. 2000; 154: 725–731.18

Scheidt

Graubard

Nelson

Intelligence at six years in relation to neonatal bilirubin level: follow up of the National Institute of Child Health and Human Development Clinical Trial of Phototherapy. Pediatrics. 1991; 87: 797–805.19

Nakamura

Yanetani

Uetani

Determination of serum unbound bilirubin for prediction of kernicterus in low birth weight infants. Acta Paediatr Jpn. 1992; 54: 642–647.20

Amin

Ahlfors

Orlando

Bilirubin and serial auditory brainstem responses in premature infants, Pediatrics. 2001; 107: 664–670.21Maisels

Jaundice. In: MacDonald

Seshia

MMK

Mullett

eds. Neonatology: Pathophysiology and management of the newborn. Philadelphia: JB Lippincott & Co. 2005: 768–846.22

Ives

Neonatal jaundice. In: Rennie

Roberton

NRC

Eds. Textbook of Neonatology. New York: Churchill Livingston, 1999: 715–732.23

Odell

The dissociation of bilirubin form albumin and its clinical implications. J Pediatr. 1959; 55: 268–79.24

Bratlid

How bilirubin gets into the brain. Clin Perinatol. 1990; 17: 449–465.25

Pearlman

Gartner

Lee

Morecki

Horoupian

Absence of kernicterus in low birth weight infants from 1971 through 1976: comparison with findings in 1966 and 1967 Pediatrics. 1978; 62: 460–464.26

Ahlfors

Criteria for exchange transfusion in jaundiced newborns. Pediatrics. 1994; 93: 488–494.27

Volpe

Bilirubin and brain injury. In: Volpe

Ed. Neurology of the newborn. Philadelphia: WB Saunders, 2001: 521–46.28Henderson-Smart

Pettigrew

Campbell

: Clinical apnea and brainstem neural function in preterm infants. N Engl J Med. 1983; 308: 353–357.29Amin

Charafeddine

Guillet

: Transient bilirubin encephalopathy can manifest as apnea of prematurity. Pediatr Res. 1999; 45: 181A (abstract).30

Bhutani

Johnson

Shapiro

Kernicterus in sick and preterm infants (1999–2002): A need for an effective preventive approach. Semin Perinatol. 2004; 28: 319–325.31

Cashore

Unbound bilirubin and kernicterus in low birth weight infants. Pediatrics. 1982; 69: 481–485.32

Ritter

Kenney

Norton

A prospective study of free bilirubin and other risk factors in the development of kernicterus in premature infants. Pediatrics. 1982; 69: 260–266.33

Murki

Kumar

Majumdar

Marwaha

Narang

Risk factors for kernicterus in term babies with non-hemolytic jaundice. Indian Pediatrics. 2001; 38: 757–762.34

Amin

Orlando

Merle

Sequential changes in BAER and their relation to bilirubin levels in neonates less than 32 weeks gestational age. Pediatr Res. 1997; 41: 135A (abstract 798).35

Newman

Escobar

Gonzales

Frequency of neonatal bilirubin testing and hyperbilirubinemia in a large health maintenance organization. Pediatrics. 1999; 104: 1198–1203.36

Bhutani

Johnson

Kernicterus in the late preterm infants cared for as term healthy infants. Semin Perinatol. 2006; 30: 89–97.37

Kaplan

Muraca

Hammerman

Imbalance between production and conjugation of bilirubin: a fundamental concept in the mechanism of neonatal jaundice. Pediatrics. 2002; 110: e47.38

Kaplan

Muraca

Vreman

Neonatal bilirubin production-conjugation imbalance: effect of glucose-6-phosphate dehydrogenase deficiency and borderline prematurity. Arch Dis Child Fetal Neonatal. Ed 2005; 90: F125.39

Bhutani

Johnson

Maisels

Kernicterus: epidemiological strategies for its prevention through system based approaches. J Perinatol. 2004; 24: 650–662.40American Academy of Pediatrics Subcommittee on Hyperbilirubinemia. Management of hyperbilirubinemia in the newborn infant 35 or more weeks of gestation. Pediatrics. 2004; 114: 297–316.41

Meloni

Cagnazzo

Dose

Cutillo

Phenobarbital for prevention of hyperbilirubinemia in glucose-6-phophate dehydrogenase deficient newborn infants. J Pediatr. 1973; 79: 1078–1081.42

Singer

Vest

Weisser

Olafsson

A double blind study on the effect of phenobarbitone on neonatal jaundice and frequency of exchange transfusion. Acta Paediatr Scand. 1970; 59: 681–684.43

Kumar

Narang

Kumar

Garewal

Phenobarbitone prophylaxis for neonatal jaundice in babies with birth weight 1000–1499 grams. Indian Pediatrics. 2002; 39: 945–951.44

Mehta

Kumar

Narang

A Randomized controlled trial of fluid supplementation in term neonates with severe hyperbilirubinemia. J Pediatr. 2005; 147: 781–785.