SE

M I N A R S I N

P

E R I N A T O L O G Y

38 (2014) 412–421

Available online at www.sciencedirect.com

www.elsevier.com/locate/semperi

Bilirubin–albumin binding, bilirubin/albumin ratios, and free bilirubin levels: Where do we stand? Christian V. Hulzebos, MD, PhDn, and Peter H. Dijk, MD, PhD Department of Pediatrics, Beatrix Children's Hospital, University of Groningen, University Medical Center Groningen, Hanzeplein 1, Groningen 9713 GZ, The Netherlands

article info

abstract

Keywords:

Treatment for unconjugated hyperbilirubinemia is predominantly based on one parameter, i.e.,

bilirubin-albumin binding

total serum bilirubin (TSB) levels. Yet, overt kernicterus has been reported in preterm infants at

bilirubin/ albumin ratio

relatively low TSB levels, and it has been repeatedly shown that free unconjugated bilirubin

free bilirubin

(freeUCB) levels, or bilirubin/albumin (B/A) ratios for that matter, are more closely associated

hyperbilirubinemia

with bilirubin neurotoxicity. In this article, we review bilirubin-albumin binding, UCBfree levels, and B/A ratios in addition to TSB levels to individualize and optimize treatment especially in preterm infants. Methods to measure bilirubin-albumin binding or UCBfree are neither routinely performed in Western clinical laboratories nor incorporated in current management guidelines on unconjugated hyperbilirubinemia. For bilirubin-albumin binding, this seems justified because several of these methods have been challenged, and sufficiently powered prospective trials on the clinical benefits are lacking. Technological advances in the measurement of UCBfree may provide a convenient means for integrating UCBfree measurements into routine clinical management of jaundiced infants. A point-of-care method, as well as determination of UCBfree levels in various newborn populations, is desirable to learn more about variations in time and how various clinical pathophysiological conditions affect UCBfree levels. This will improve the estimation of approximate UCBfree levels associated with neurotoxicity. To delineate the role of UCBfree in the management of jaundiced (preterm) infants, trials are needed using UCBfree as treatment parameter. The additional use of the B/A ratio in jaundiced preterms has been evaluated in the Bilirubin Albumin Ratio Trial (BARTrial; Clinical Trials: ISRCTN74465643) but failed to demonstrate better neurodevelopmental outcome in preterm infants o32 weeks assigned to the study group. Awaiting a study in which infants are assigned to be managed solely on the basis of their B/A ratio (with TSB excluded ) versus TSB levels alone—and determining which group does better—the additional use of the B/A ratio in the management of hyperbilirubinemia in preterms is not advised. In conjunction with TSB levels, other parameters possibly allow for more accurate prediction of bilirubin toxicity. Yet, different methodologies for estimating these parameters exist, and sufficiently powered, prospective clinical trials supporting their clinical benefit, i.e., reduced bilirubin neurotoxicity when using these parameters, are lacking. Their use in addition to TSB needs to be prospectively evaluated, especially in preterm neonates, and preferentially in randomized clinical trials, which include specific risk factors and assessment of clinical relevant outcome measures for detecting those infants at risk of bilirubin toxicity. & 2014 Elsevier Inc. All rights reserved.

n

Corresponding author. E-mail address: [email protected] (C.V. Hulzebos).

http://dx.doi.org/10.1053/j.semperi.2014.08.004 0146-0005/& 2014 Elsevier Inc. All rights reserved.

S

E M I N A R S I N

P

E R I N A T O L O G Y

Introduction Current management guidelines for neonates with unconjugated hyperbilirubinemia recommend measuring albumin levels together with total serum bilirubin (TSB) levels. The 2004 American Academy of Pediatrics (AAP) guidelines propose considering a more aggressive treatment should albumin levels drop below 30 g/L.1 According to some European guidelines, including the Dutch guideline of 2009, albumin infusion before exchange transfusion should be considered when albumin levels drop below 20 g/L.2 But why bother with albumin, and what is the relevance of knowledge on bilirubin-albumin binding in the management of hyperbilirubinemia? Are bilirubin/albumin (B/A) ratios more appropriate to optimize clinical management of unconjugated hyperbilirubinemia? Would it not be better to strive for guidelines that incorporate levels of free unconjugated bilirubin (UCBfree), which are more closely associated with bilirubin neurotoxicity? The purpose of this review is to provide some answers on the above questions with a view to individualizing and improving risk assessment of imminent bilirubin neurotoxicity in preterm and fullterm neonates.

The rationale for using more than one parameter in the management of unconjugated hyperbilirubinemia In 1959, Bowen et al.3 injected newborn puppies with a bilirubin solution and demonstrated marked clinical improvement when the pups were treated simultaneously with human albumin: jaundice resolved and survival improved, with a concomitant reduction of bilirubin tissue content. In the same year, Odell,4 after giving a preterm infant a simple albumin injection, interpreted the rise in total serum bilirubin (TSB) levels as a bilirubin shift from tissues to plasma “presumably attached to the administered albumin.” These insights provide an understanding of the sulfisoxazole tragedy in preterms more than half a century ago: Silverman described an increased mortality rate and incidence of kernicterus among premature infants treated with prophylactic sulfisoxazole, a potent bilirubin-displacer, which shifts bilirubin from albumin to tissues.5 In the 1970s, an apparent reduction of kernicterus in preterms at autopsy was observed, probably due to a combination of maternal Rhesus immunization, early use of phototherapy (PT), and better neonatal care in general.6 Consensus among pediatricians that there is no strong correlation between TSB levels and kernicterus in sick preterms grew. This ultimately resulted in the statement that TSB levels alone could no longer be considered a sufficient means of assessing the neurotoxicity risk of jaundiced infants.7 Today, 35 years on, this statement still holds. Kernicterus in extreme low-birth-weight (ELBW) infants was reported at relatively low bilirubin levels of 150 mmol/L (
9 mg/dL, 17.1 mmol/L ¼ 1 mg/dL bilirubin), i.e., between commonly recommended PT and exchange transfusion thresholds.8 Recently, this was confirmed by a report on ELBW infants who had developed

38 (2014) 412–421

413

kernicterus despite moderate, and treated, unconjugated hyperbilirubinemia.9 In line with the inconsistent correlation between TSB and kernicterus in preterms is the inconsistency between peak TSB and neurodevelopmental outcome in fullterms, which suggests that other factors also contribute to bilirubin neurotoxicity.10–13 Another argument for using more than one parameter to optimize management of hyperbilirubinemia is that PT, although generally considered safe, should not be given unnecessarily and only to infants at risk of bilirubin neurotoxicity—preterms being most vulnerable. Neurologic sequelae in near-fullterms with TSB levels maintained at o340 mmol/L, or even at o428 mmol/L, are scarce.12,14 Nevertheless, also in this particular group, PT requires hospitalization and limits parents' access to their infants. In addition, a few studies even suggested that PT may increase mortality, especially among ELBW infants.15–17 Finally, associations of PT and systemic diseases were also reported in near-fullterms.18,19

The rationale for using other parameters in addition to TSB levels The foregoing observations thus lend support to using other parameters in addition to TSB levels to quantify the risk of bilirubin neurotoxicity and to reduce unnecessary treatment, while causing no harm. In addition to, and not instead of TSB, because TSB reflects the total exchangeable body pool of bilirubin. The inconsistency of the univariate correlation between TSB and bilirubin neurotoxicity is most likely to apply to other parameters as well. This inconsistency can, at least in part, be explained by the contribution of “downstream” factors of bilirubin neurotoxicity.20–23 Downstream factors include integrity or leakiness of the blood–brain barrier; duration of contact between UCBfree and brain capillary endothelium; transport of UCBfree across the brain endothelium, which depends on blood flow, on contact time, and on the dissociation rate of bilirubin from albumin; and rates of regional bilirubin entry and clearance from the brain. Neuronal excitotoxicity, mitochondrial energy failure, changes in membrane functions of the cell together with neuronal susceptibility, and antiapoptosis factors of the newborn infant represent other downstream factors collectively involved in the mechanism of bilirubin neurotoxicity.24 The broad variety of neurological sequelae in response to severe unconjugated hyperbilirubinemia is thus conceived to be attributable to complex contemporaneous pathways involved in bilirubin neurotoxicity.25 Bilirubin-albumin binding, bilirubin/albumin (B/A) ratios, and UCBfree levels have been considered in predicting the risk of bilirubin neurotoxicity. But, what about their clinical relevance? In order to find evidence to adapt current guidelines by incorporating one or more of these parameters, one should obtain data on their relation with appropriate measures of bilirubin neurotoxicity from either experimental or clinical studies.

Bilirubin-albumin binding In the Gunn rat, the well-established animal model for unconjugated hyperbilirubinemia, infusion of albumin

414

SE

M I N A R S I N

P

E R I N A T O L O G Y

Fig – Postnatal course of bilirubin-albumin-binding affinity (K) in preterm infants. Data were obtained from a subgroup of preterm infants (GA o32 weeks) included in the BARTrial and admitted to a Dutch tertiary NICU.79 The boxes are limited by the 25th and 75th percentile. The horizontal line in the box represents the median. The whiskers (?) represent the lowest and highest K value measured on that particular day. Outliers (○) are depicted separately. significantly increases TSB levels; this reflects the powerful attractive effect of albumin on tissue bilirubin.26 If administered one to two hours before an exchange transfusion, albumin significantly enhances bilirubin removal.27 The association and dissociation of bilirubin with albumin has rapid kinetics, explaining the precedent finding of rebound TSB levels within the hour after an exchange transfusion.28 Albumin has distinct bilirubin-binding sites: a primary site with high bilirubin affinity and several secondary binding sites of weaker affinity.29 According to a biochemical principle, i.e., the mass action equation, “K” is referred to as an bilirubin-albumin equilibrium binding constant and reflects the ability of albumin to bind bilirubin. In physiological terms, “K” may reflect a compromise between the need to prevent excessive tissue– bilirubin binding and the need for efficient hepatic clearance; avid bilirubin-albumin binding may decrease hepatic clearance. Bilirubin-albumin binding fluctuates greatly in the first postnatal days (Fig). In healthy infants, the bilirubin binding fraction of albumin, measured with monoacetyldiaminodiphenyl sulfone (MADDS), a deputy ligand for bilirubin, increased to peak values after 60 h of age, which did correlate with maturity.30 Postnatal age positively affects bilirubinalbumin binding (Table 1), and binding is lower in the smallest and sickest infants.31 This appears also to be true for reserve albumin levels, i.e., albumin available for binding of unconjugated bilirubin. Reserve albumin was determined in the serum samples of 76 infants by dialysis with (14C) MADDS.32 A bilirubin toxicity index was calculated based on TSB levels and reserve albumin, the affinity of bilirubin for serum albumin, and the pH-dependent solubility of bilirubin. Reserve albumin and bilirubin toxicity index varied significantly with gestational age (GA), clinical condition, and TSB levels. Reserve albumin was decreased and the bilirubin toxicity

38 (2014) 412–421

index was increased in association with clinical risk factors (e.g., hypoxia, acidosis, or sepsis) recognized as increasing the risk of bilirubin neurotoxicity. Gestational age was inversely related to reserve albumin levels; the least mature and sickest infants had the lowest reserve albumin levels and the highest bilirubin toxicity index. Bilirubin-albumin binding is also influenced by endogenous and exogenous substances. At secondary sites, exogenous ligands, i.e., organic anions like salicylate and sulfisoxazole, displace bilirubin readily, even below molar equivalence. Table 2 lists displacement factors of various ligands.33 Nonesterified free fatty acids (FFAs) add to the displacing effect of anions when the molar ratio exceeds two: then both tight binding sites of albumin for FFAs are occupied and surplus FFAs displace bilirubin from its primary to the secondary site, where anions can again displace bilirubin. At a molar ratio above three, FFAs compete directly at secondary sites.34 Amin et al.35 showed a statistically significant decrease in binding affinity, increased UCBfree levels, and FFA levels with higher parenteral lipid intake in infants with a GA of 28 weeks or less, but not in infants 428 weeks' GA.36 Plasma bilirubin binding also contributes to the hourly rate of increase in TSB,37 although this increase is primarily the result of an imbalance between bilirubin production and elimination. The strength of plasma bilirubin binding depends on plasma albumin levels and the ability of albumin to bind bilirubin.37 This affinity of human serum albumin to bind bilirubin is inversely related to the albumin level.38 The type of albumin (neonatal versus adult) is another determinant of bilirubin binding. Cashore et al.39 demonstrated lower bilirubin-binding capacity of newborn albumin in comparison to adult albumin. Cashore and Oh40 reconfirmed the limited capacity of neonatal albumin to bind bilirubin: sick preterms who had kernicterus at autopsy had a theoretical limit of highaffinity binding of around 0.6 molar ratio, whereas those preterms without kernicterus could bind up to 0.8 mol/mol. In some sick, kernicterus-prone preterms, primary binding Table 1 – Factors affecting bilirubin-albumin binding. Factor

Bilirubin-albumin binding Low vs. high

Type of albumin Albumin levels Gestational age Postnatal age

Newborn vs. adult Low vs. normal Preterm vs. term Day of birth vs. end of 1st week ELBW vs. AGA Sick vs. healthy Acidosis vs. neutral pH For example: High vs. low levels of free fatty acids, i.v. vs. no sulfonamide antibiotics

Birthweight Illness pH Endogenous and exogenous substances including drugs

Note: Sampling site and phototherapy do not affect bilirubinalbumin binding. ELBW: extreme low birth weight; AGA: appropriate for gestational age.

S

E M I N A R S I N

P

E R I N A T O L O G Y

sites may be saturated at TSB levels of 190 mmol/L (11.1 mg/dL). Meisel et al.41 evaluated the influence of acid–base state on the bilirubin-albumin-binding state in the blood of newborn infants. In 171 appropriate-for-gestational-age (AGA) and 83 small-for-gestational-age (SGA) infants (with birth weights o2500 g), they measured the acid–base state and bilirubinbinding state in serum at the ages of 3, 4, 5, and 8 days. They found a weak but significant correlation between standard base deficit and the ratio bilirubin/reserve albumin as well as the bilirubin toxicity index: a decrease in bilirubin binding and a doubling of the toxic potential of bilirubin occurred on shifting from base deficit 0 to 10. In 11 near-fullterms, the correction on day 1 (between 9 and 18 h of life on average) of a metabolic acidosis (from pH 7.12 to 7.34 on average) resulted in an improvement of apparent bilirubin-albumin-binding affinity and a drop of UCBfree levels with
25%.42 The initial mean (7SD) albumin levels were 33.5 g/L (72) and dropped to 30.5 g/L (71.5) at the time of the second measurement. In line with the important role of albumin, Miwa et al.43 demonstrated reduced albumin levels and concomitantly higher UCBfree levels in 15 newborns (434 weeks' GA) following surgery for congenital abnormalities of the gastrointestinal tract. TSB levels were comparable despite a delay in enteral feeding in the surgery group. Unfortunately, levels of FFA were not measured. Fasting may increase FFA levels, which could have contributed to higher UCBfree levels (Table 1). Many studies have evaluated bilirubin-albumin binding using various methods, but few studies have assessed the use of bilirubin-albumin binding to neurodevelopmental outcome. Kapitulnik et al.44 assessed the binding status using the Sephadex gel filtration method. In short, a mix of infants' serum and phosphate buffer was run through the column filled with 2-ml Sephadex G-25, followed by a buffer to wash the serum out of the column. Next, another solution ran through the column. Bilirubin, adsorbed by the Sephadex gel from serum, reacted with this reagent solution to give a blue band (positive test result) which increased in intensity as it passed down the column. Table 2 – In vitro displacement factors for various parenterally administered drugs in NICUs. Type of drug

Displacement factor

Ibuprofena Ceftriaxone Sulfisoxazole Sulfadiazine Aminophylline Methicillin Furosemide Oxacillin Penicillin G Cefotaxime Phenobarbital Hydrochlorothiazide Vancomycin Diazepam Indomethacin Doxapram

4.00 3.00 2.43 1.69 1.24 1.17 1.07 1.07 1.06 1.05 1.04 1.03 1.01 1.00 1.00 1.00

A displacement factor of 1.00 means no effect. a At therapeutic dosages, ibuprofen does not have any effect on UCBfree levels in vivo.

38 (2014) 412–421

415

Thus, a positive test reflects poor bilirubin-albumin binding. In 166 preterms with neonatal jaundice who underwent an exchange transfusion, those infants who consistently showed a negative Sephadex test, i.e., good bilirubin-albumin binding, never showed evidence of bilirubin neurotoxicity, even with TSB levels 4510 mmol/L (29.8 mg/dL). A negative Sephadex test thus discriminated jaundiced infants not in immediate danger of bilirubin encephalopathy fairly well. Unfortunately, no data on long-term evidence of brain damage are available to determine whether using the Sephadex test as a major criterion for exchange transfusion will improve outcome. Data from studies by Zamet et al.45 and Valaes et al.46 showed good correlation of the Sephadex test, UCBfree levels, and kernicterus. Another study demonstrated that UCBfree levels obtained by the Sephadex test showed good agreement with UCBfree levels obtained with the peroxidase method.47 Odell et al.48 described the relationship of “bilirubin-saturated” serum proteins in the postnatal period and cognitive dysfunction in 32 infants (BW ranging from 1400 to 3800 g) at four to seven years of age. They used the salicylate saturation index: a measure for the amount of bilirubin displacement from secondary binding sites and, therefore, an indicator of overload of primary and secondary sites. Unlike TSB levels, the saturation index correlated with isolated cognitive dysfunction. The duration of exposure to TSB 4 257 mmol/L was statistically significant longer in children who later appeared to be abnormal. Although statistically insignificant, exchange transfusions were performed in 90% of abnormal children versus 60% of normal children [the odds ratio was 6.0 (95 CI: 0.98–36.7; P ¼ 0.096)]. In addition to a higher salicylate saturation index, a lower reserve bilirubin–albumin binding capacity was reported in jaundiced newborn infants with neurological sequelae.49,50 In these studies, the non-specific, competitive binding of 2-(40 hydroxybenzeneazo) benzoic acid (HBABA) or phenolsulfonphthalein (PSP) to albumin was used to estimate the number of available binding sites on the albumin molecule. This “dye-binding reserve” is inversely related to TSB levels, but unfortunately also to levels of other competing anions for other dye-binding sites not specific for bilirubin.7 In 74 children of 9–11 years of age, Hansen et al. investigated the relation between neonatal hyperbilirubinemia and several measures of psychoeducational outcome, including the Kaufman Mental Processing Composite (KMPC). Albumindetermined binding capacity [calculated as 5 (albumin) þ 2] was statistically significant and correlated positively with the KMPC, in contrast to direct measures of bilirubin binding.51 Kim et al.52 documented lower binding capacity and lower serum albumin levels (20 7 8 g/L versus 26 7 5 g/L) and acidosis in a very low-birth-weight (VLBW) kernicterus cohort at autopsy compared with a case–control population, but none of these associations was statistically significant. In spite of the possibility that data on bilirubin-albumin binding may allow for more appropriate clinical management of unconjugated hyperbilirubinemia, the clinical relevance of bilirubin-albumin binding was disputed by others.20,53,54 Currently, methods to measure bilirubin-albumin binding are neither routinely performed in clinical laboratories nor incorporated in current management guidelines on unconjugated hyperbilirubinemia. This seems justified because several of these methods have been challenged, and sufficiently

416

SE

M I N A R S I N

P

E R I N A T O L O G Y

powered prospective trials on the clinical benefits of using bilirubin-albumin-binding methods are lacking.7

Unconjugated free bilirubin Excellent original and review articles have delineated the role of UCBfree in bilirubin toxicity.55,56 Unconjugated bilirubin (UCB) can impair mitochondrial function and viability of astrocytes and induce apoptosis in brain neurons. The amount of UCBfree that is able to cross the blood–brain barrier depends merely on concentrations of UCBfree and not on TSB concentrations.57 This is in line with data showing that UCBfree correlates better with kernicterus and other signs of bilirubin neurotoxicity than the TSB. Independent of one another, Zamet et al.45 and Cashore40 correlated UCBfree levels with autopsy-confirmed kernicterus in preterms: kernicteric LBW infants had approximately two-fold higher UCBfree levels versus controls without kernicterus. Kernicterus, however, also occurred in sick preterms with rather low UCBfree levels, once more underlining the importance of other factors in kernicterus.53 Nakamura et al. were amongst the first to construct receiver-operating characteristic (ROC) curves for UCBfree on acute and chronic bilirubin-associated neurological signs, and autopsy-conformed kernicterus in 50 VLBW and 88 LBW infants. ROC curves for predicting bilirubin encephalopathy using UCBfree were shifted left-upward, compatible with markedly improved sensitivity and specificity of UCBfree when compared to TSB.58 Although kernicterus is a very relevant outcome measure of bilirubin neurotoxicity, the relatively low frequency does not allow for estimating “dangerous” UCBfree levels. For that matter, studies on bilirubin-induced auditory dysfunction, which starts at brainstem level and later proceeds into the auditory nerve and spiral ganglion, may be more sensitive and consequently, more relevant. Nakamura et al. found that UCBfree predicted early bilirubin neurotoxicity, as assessed by auditory brainstem responses (ABRs), better than TSB in 56 hyperbilirubinemic fullterms with peak TSB levels Z257 mmol/L (15 mg/dL): changes in ABRs were more profound in infants with the highest UCBfree levels. Irreversible hearing problems were likely with UCBfree 417.1 nmol/L (1 μg/dL).59 This “toxic threshold,” albeit influenced by the methodology used, was also found in a historical data analysis of kernicterus case clusters and somewhat higher in studies on ABR changes in fullterms.60 Funato et al.61 confirmed that changes in ABRs were more closely associated with UCBfree than with TSB: bilirubin-induced changes in ABR occurred at mean peak UCBfree levels of 23 nmol/L in 37 fullterms with TSB 4 340 mmol/L. ABR changes in preterms with GAs between 28 and 32 weeks occurred at lower UCBfree levels; the relative risk of abnormal ABR maturation with UCBfree levels 48.6 nmol/L proved to be 2.5 (95% CI: 1.3–4.5).62 In summary, UCBfree proved to be the most sensitive predictor of transient bilirubin encephalopathy reflected by maturation ABRs in preterm and fullterm infants with unconjugated hyperbilirubinemia. Technically, ABR testing may be a feasible method to assess acute bilirubin encephalopathy objectively, but it remains a rather laborious procedure (especially in preterms), is not easily obtained, and even

38 (2014) 412–421

unavailable in many hospitals. To this end, Ahlfors et al. described the usefulness of a bedside automated auditory brainstem response (AABR) in a retrospective study. In contrast to a formal ABR test, the AABR is a hearing screening test (Natus Medical, San Carlos, CA) based on the presence (pass) or absence (refer) of wave 5 (which originates from neuronal activity at the inferior colliculus) at 35 dB. In 191 newborn infants with BWs ranging from
400 to 4500 g and GAs between 24 and 42 weeks, UCBfree levels (between 1 and 30 nmol/L) doubled when AABR was abnormal. As such it proved to be a better predictor of bilirubin-related auditory dysfunction, defined as a “refer” AABR test, than TSB.63 Regarding acute clinical neurologic signs related to hyperbilirubinemia, Murki et al. reported that UCBfree levels were statistically significant higher (between 13 and 46 nmol/L) in 14 symptomatic hyperbilirubinemic fullterms compared to their asymptomatic fullterm counterparts.64 Acute bilirubin encephalopathy was found in hyperbilirubinemic fullterms with UCBfree levels of
75 nmol/L (“refer” AABR) and of
130 nmol/L (signs of classic kernicterus) as measured by a modified peroxidase method.65 Unfortunately, long-term follow-up data of these fullterms (with TSB levels up to
800 mmol/L) are unknown. Few studies relate UCBfree to long-term neurodevelopmental data in preterms. Oh et al. found that higher UCBfree levels in 1101 ELBW infants (
26 weeks’ GA) at 5 7 1 days of age were associated with a higher risk of death or adverse neurodevelopmental outcome at 18 to 22 months corrected age regardless of clinical status.66 An inverse relationship was found between TSB and death or cerebral palsy in stable infants. Although death and developmental status depend on many neonatal factors, and not on bilirubin alone, the authors suggested using UCBfree levels in the management of jaundiced ELBW infants to determine when to initiate or stop treatment since its correlation was independent of clinical status. Clinical status, however, may affect UCBfree levels. Bender et al.67 prospectively studied 152 preterms (23–31 weeks’ GA) and demonstrated higher UCBfree levels in infants with the highest risk scores according to a simplified Score for Neonatal Acute Physiological Perinatal Extension (SNAP-PE)-II. In addition, Lee et al. measured higher UCBfree levels in infants with lower BW, and highest in VLBW infants (BW o 1500 g) with complications such as sepsis and intraventricular hemorrhage even in the presence of lower TSB levels.68 Although direct measurement of UCBfree levels with the peroxidase method, in which UCBfree is oxidized and changes to a colorless compound, appeared promising,69 some authors argued that UCBfree levels would change too rapidly and equilibrate too swiftly with tissue bilirubin stores to be used as a marker for the risk of kernicterus.70 Within minutes after using a displacer, the UCBfree levels, which are a magnitude smaller than TSB levels, return to predisplacement values. It is conceivable, therefore, that similar UCBfree levels may occur in infants with and without kernicterus. Early peroxidase techniques required a 41-sample size dilution. Dilution affects intrinsic bilirubinalbumin binding, and effects of bilirubin-displacing drugs on UCBfree levels are underestimated.38 The increase in UCBfree level was significantly attenuated after dilution of plasma samples containing benzoate, a metabolite of benzyl alcohol:

S

E M I N A R S I N

P

E R I N A T O L O G Y

actual levels of UCBfree around 10 nmol/L were in fact measured with dilution at around 2 nmol/L.71 For this and other methodological and technical reasons, some experts consider UCBfree measurements unreliable.72 Technological advances in the measurement of UCBfree with a modified peroxidase method (using more than a single peroxidase concentration and minimal sample size dilution) or with a fatty acid binding protein labeled with a fluorescent sensor (whose fluorescence is quenched upon binding bilirubin) may provide a convenient means for integrating UCBfree measurements into routine clinical practice.73,74 To this end, determination of UCBfree levels in various newborn populations, e.g., sick (acidotic) and healthy preterms, post-surgery infants, infants on specific medications, and infants with signs of bilirubin toxicity, is desirable to learn more about variations in time and how various clinical pathophysiological conditions affect UCBfree levels. Improvement of UCBfree standardization, using one specific (and preferably point of care) method for the quantitative analysis of UCBfree, will probably reduce the large range of reported peak UCBfree levels in different studies (Table 3) and will improve the estimation of approximate UCBfree levels associated with neurotoxicity. Finally, trials are needed using UCBfree as treatment parameter to delineate its exact role in the management of jaundiced (preterm) infants. Again, a mutual role exists for UCBfree and the miscible bilirubin pool, i.e., TSB, as was recently demonstrated: UCBfree/TSB ratios predicted bilirubin neurotoxicity better than UCBfree only.63 Table 3 – Overview of UCBfree levels assessed with different peroxidase methodologies in a diverse population. Methodology

Study population (number)

Peak UCBfree (nmol/L)

Arrows UB Analyzer58

VLBW (n ¼ 45) LBW (n ¼ 81) Infants at risk of kernicterus (n ¼ 12) Fullterms (n ¼ 37) Normal ABR, n ¼ 18 Mild abnormal ABR, n¼8 Definite abnormal ABR, n ¼ 11 Preterms (n ¼ 45) Normal ABR, n ¼ 20 Abnormal ABR, n ¼ 25 Preterms and fullterms (n ¼ 191) Normal AABR, n ¼ 175 Abnormal AABR, n ¼ 16 VLBW (n ¼ 86) LBW (n ¼ 183) AGA (n ¼ 119) Preterms (n ¼ 72)

7.9 7 2.9 9.2 7 3.4 20.6 7 4.8

Arrows UB Analyzer61

Arrows UB Analyzer62

Arrows UB Analyzer63 (2 peroxidase concentrations)

UB-Analyzer UA-268

Zone Fluidics system (Global Flopro)79

13.3 7 4.4 21.7 7 12 22.9 7 8.9

6.8 7 2.6 10.6 7 3.4

15.9 7 12 30.1 7 22.4 8 (1.4–30.6) 10.3 (4.1–29.6) 12.7 (6.2–33.5) 50 (13–197)

Data are expressed as mean (7SD) and/or median (range) values.

38 (2014) 412–421

417

Bilirubin–albumin ratio Studies demonstrating that UCBfree is more closely associated with bilirubin neurotoxicity also support the potential benefits of using B/A ratios in addition to TSB levels: the B/A ratio seems a more useful indicator of UCBfree levels than using TSB levels alone.75 The B/A ratio can be calculated on the basis of standard laboratory assays. Moreover, via serum albumin levels, it has already been incorporated indirectly into exchange transfusion criteria. However, due to the presence of drugs that interfere with bilirubin–albumin binding, UCBfree levels may be much higher than suggested by the calculated B/A ratio and the intrinsic affinity of albumin for bilirubin. Furthermore, other plasma constituents such as apolipoproteins, and α-fetoprotein, bind unconjugated bilirubin (up to
10%). Consequently, the B/A ratio seems an imperfect surrogate for estimating UCBfree. But even UCBfree is likely to be an imperfect predictor of neurotoxicity since many (downstream) factors affect the development of neurotoxicity at any given UCBfree level.23 The theoretical considerations and clinical evidence for the concept that using the B/A ratio in addition to other parameters in jaundiced preterms might improve the prediction of bilirubin neurotoxicity were reviewed in 2008.76 Table 4 summarizes these data (adapted with permission from Archives in Diseases of Childhood). Overall, the B/A ratio was not significantly better than TSB in predicting acute bilirubin encephalopathy as assessed by abnormal maturation of ABRs in a prospective study in 143 preterms (28–32 weeks' GA).62 In a subset of 45 infants, higher B/A ratios of 0.39 versus 0.33 molar ratios were related to abnormal versus normal ABRs, and preceded abnormal ABR maturation, as did higher UCBfree levels, whereas TSB did not. Scheidt et al.77 related neurodevelopmental outcomes to TSB levels in 224 preterm and low-birth-weight (LBW) infants who were allocated to the control group of the randomized controlled PT trial of the US National Institute of Child Health and Human Development (n ¼ 1339). The aim of this randomized controlled trial (RCT) was to compare the efficacy of prophylactic or therapeutic PT and/or exchange transfusion. Exchange transfusions maintained TSB levels below specified levels in the control group that did not receive PT. Neurodevelopmental outcome at six years of age was similar in both treatment groups. Neither cerebral palsy nor IQ was significantly associated with TSB levels, duration, or time of exposure to bilirubin. IQ was inversely related to B/A ratios, but not after correcting for neonatal risk factors. Clinical and pathologic factors in 27 kernicteric (witnessed by yellowish staining of cerebral gray matter) infants (with a mean GA of 32 weeks and a mean BW of
1400 g,) were compared retrospectively to a matched control group of 103 autopsied infants without kernicterus. In analogy to TSB peak levels (
200 mmol/L), molar B/A ratios were similar in kernicteric versus non-kernicteric infants.52 Cashore et al. determined B/A ratios in 13 deceased preterms (BW o 1500 g) with hyperbilirubinemia; five had kernicterus. Compared to non-kernicteric infants, bilirubinbinding capacities expressed as molar B/A ratios, were
20% lower (P o 0.05) in infants with kernicterus.40

418

SE

M I N A R S I N

P

E R I N A T O L O G Y

38 (2014) 412–421

Table 4 – Summarized evidence on the use of B/A ratios and outcome in preterm infants with unconjugated hyperbilirubinemia. References

Study design

Study population

Method

Relation B/A ratioa and outcome

Amin et al.62

Prospective cohort study

143 Infants (GA: 28–32 weeks)

Higher B/A ratios preceded abnormal ABR maturation

Scheidt et al.77

Retrospective analysis of a RCT

224 Premature/LBW infants (o2000 g)

Kim et al.52

Case–control post-mortem study

27 Kernicteric infants (GA: 27–38 weeks) and 103 matched controls

Maturation of the auditory brainstem responses (ABR) Neurological examination, Wechsler Intelligence Scale for ChildrenRevised at six years Comparison of clinical, laboratory, and histopathological data

Cashore and Oh40

Case–control post-mortem study Prospective cohort study

13 Premature infants (o1500 g); five with kernicterus 91 Premature VLBW infants (o1500 g)

Govaert et al.8

Retrospective case study

Iskander et al.78

Prospective cohort study Randomized controlled trial (B/A ratio þ TSB vs. TSB)

5 Premature (GA: 25–29 weeks) and three term infants with clinical signs of kernicterus 193 Near-fullterm and fullterm infants 614 Preterm infants (GA o 32 weeks)

Ritter et al.53

Hulzebos et al.79

Comparison of laboratory data Assessment of risk factors in the development of kernicterus ABR, magnetic resonance imaging (MRI), and/or ultrasound BIND score, neurological examination, and AABRc Neurodevelopmental impairment; Bayleys Scale Infant Development 3rd Edition

IQ decreased at higher B/A ratios, no relationship after correction for neonatal risk factors Lower albumin levels in kernicteric infants, but available B/A ratios (6 of 27 and 15 of 103) appeared similar Binding capacities expressed as molar B/A ratios were lower in infants with kernicterus B/A ratios appeared higher in infants with kernicterusb Abnormal ABR, MRI, and/or ultrasound in all premature infants with elevated B/A ratios Specificity B/A ratio 4 TSB in predicting all outcomes Additional use of B/A ratio did not confer beneficial effects

Adapted with permission from BMJ Publishing Group Limited. Usefulness of the bilirubin/albumin ratio for predicting bilirubin-induced neurotoxicity in premature infants. Hulzebos CV, van Imhoff DE, Bos AF, Ahlfors CE, Verkade HJ, Dijk PH. Arch Dis Child Fetal Neonatal Ed. 2008 Sep;93(5):F384-8. Copyright notice year 2014. a B/A ratios were calculated from serum bilirubin and albumin concentrations, which were measured by routine laboratory techniques. In the study of Cashore, bilirubin-binding capacities were estimated from bilirubin titration curves and expressed as moles of bilirubin per mole of albumin, i.e., molar B/A ratios at apparent saturation of first binding site of albumin. Accordingly, low B/A ratios reflect low bilirubin-binding capacities in the latter study. b Data not provided, but calculated by presenting authors. c BIND: bilirubin-induced neurological dysfunction on admission day, neurological examination at three to five months, and automated auditory brainstem response (AABR) on admission, at discharge, and at the age of three to five months.

Calculated mean B/A ratios appeared higher in VLBW infants with kernicterus in a prospective study of 91 VLBW infants with and without kernicterus (2.6 versus 2.2 mg/g, kernicteric versus non-kernicteric VLBW infants, respectively).53 Govaert et al. reported on five preterms (25–29 weeks’ GA) with pallidal cell death and gliosis together with hearing loss attributed to bilirubin, despite TSB levels below exchange transfusion thresholds. Respiratory and metabolic acidosis was present in three infants around the TSB peak. The B/A molar ratio was above 0.5 in all five infants in the absence of bilirubin-displacer drugs.8 Iskander et al.78 studied whether the B/A ratio is more precise than TSB in predicting the risk of bilirubin encephalopathy in 193 near-fullterm and fullterm hyperbilirubinemic infants (TSB ranged from
320 to 4 1000 mmol/L, and B/A ranged from 6.3 to 21 mg/g). Among others, analyses included specificities of B/A ratio and TSB in identifying pretreatment and posttreatment outcomes (highest bilirubin-induced neurological dysfunction (BIND) score obtained on admission day, AABR on admission and at discharge, and-together with

a neurological examination-at the age of three to five months). A total of 62 subjects had moderate to severe BIND (scores 4–9) on admission, including 32 cases of kernicterus confirmed by death or follow-up examination. “Refer” AABRs occurred in 83 out of 161 (52%) patients tested on admission, in 59 out of 154 (38%) at discharge, and in 21 out of 101 (21%) when examined at three to five months. “Refer” AABRs resolved to normal in all infants with TSB o 530 mmol/L (31 mg/dL) or B/A ratio o 8.6. Using the current AAP guidelines, B/A ratio had higher specificity than TSB in predicting all outcomes. Until recently, prospective clinical trials on the use of B/A ratio in jaundiced preterms supporting or refuting its clinical benefit, i.e., prevention or reduction of long-term bilirubininduced neurotoxicity, were lacking. To the best of our knowledge, one prospective, RCT was conducted to evaluate the additional use of the B/A ratio in jaundiced preterms, the Bilirubin Albumin Ratio Trial (BARTrial; Clinical Trials: ISRCTN74465643).79 In the BARTrial, 615 preterms of 32 weeks’ gestation or less were randomly assigned to treatment

S

E M I N A R S I N

P

E R I N A T O L O G Y

based on either B/A ratio or TSB thresholds, whichever threshold was crossed first, or on TSB thresholds only. The primary outcome was neurodevelopment at 18–24 months’ corrected age as assessed with the Bayley Scales of Infant Development III by investigators unaware of treatment allocation. Secondary outcomes included complications of preterm birth and death. Composite motor and cognitive scores did not differ between infants allocated to either the B/A ratio or TSB groups. Demographic characteristics, maximum TSB levels, B/A ratios, and other secondary outcomes were similar. The rates of a composite outcome, i.e., death and/or severe neurodevelopmental impairment, for the two groups were similar for infants with birth weights r1000 g. Fewer infants died in the prespecified subgroup of infants 41000 g, who were treated on the basis of the B/A ratio and TSB—although this could have been by chance—whereas neurodevelopmental outcome was not different. In retrospect, the B/A ratio threshold may not have been a robust discriminator of bilirubin neurotoxicity relative to TSB alone:
30% of the infants in the B/A ratio group were treated on the basis of B/A as opposed to TSB, in fact under-powering this study to assess the clinical value of the B/A ratio. Nevertheless, this methodology was chosen because no ethical justification could be found for safe B/A thresholds. We await a study in which infants are assigned to be managed solely on the basis of their B/A ratio (with TSB excluded) versus TSB levels alone—and are subsequently evaluated for neurodevelopmental outcome. When the B/A ratio is used in addition to the TSB in the management of hyperbilirubinemia in preterms, it does not improve neurodevelopmental outcome.

Conclusion Currently, the published data appear to favor the use of bilirubin-albumin-binding data, UCBfree levels, and B/A ratios in addition to TSB levels in the management of the jaundiced newborn. In conjunction with TSB levels, these other parameters should allow for more accurate prediction of bilirubin toxicity. Yet, different methodologies for estimating these parameters exist, and the number of publications on these parameters and outcome is relatively small. Moreover, we lack sufficiently powered, prospective clinical trials that demonstrate a reduction in bilirubin neurotoxicity when these are used. Use of B/A ratios and UCBfree levels in addition to TSB needs to be evaluated prospectively, especially in preterm neonates, and preferentially in randomized clinical trials that include specific risk factors and assessment of clinically relevant outcome measures for detecting those infants at risk of bilirubin toxicity.

Implications Prevention of severe hyperbilirubinemia and bilirubininduced neurotoxicity by using the current management guidelines remains paramount and has been achieved with success in many countries. The presence of rather effective and high-intensity PT together with the decreased need for

38 (2014) 412–421

419

exchange transfusion might hinder interest in additional treatment parameters to individualize treatment. How do we rekindle interest in “bilirubin matters”? Awareness of many factors involved in bilirubin toxicity, potential hazards of prophylactic PT in ELBW infants, and limited evidence of current national and international TSB-based guidelines, together with data on (preterm) patients with kernicterus despite treatment, should be sufficient to reinforce efforts at home and abroad to elaborate on this specific challenge: (i) to identify the preterm and fullterm infants at risk for bilirubin neurotoxicity, (ii) to install timely and individualized treatment, and (iii) to reduce unnecessary treatment. As a first step, we propose to establish an international databank containing serial blood samples from a diverse newborn population, and measurements of bilirubin-albumin-binding data (i.e., B/A ratios and UCBfree levels, preferably using one robust method) of infants with signs of acute or chronic bilirubin neurotoxicity (as witnessed by ABR, MRI, and neurodevelopmental tests). These data should not only be obtained in Western countries but also in developing countries where severe hyperbilirubinemia still constitutes a major health problem. In addition to TSB, normative, daily reference levels of these treatment parameters should ultimately result in prospective trials assessing the clinical benefits of these parameters and eventually, depending on the results, in “TSB-plus” treatment thresholds (e.g., plus UCBfree) in order to optimize and individualize management of unconjugated hyperbilirubinemia. Potential hazards, be they methodological, logistic, and/or financial, and complaints may be foreseen, but it is worth trying so that in 40 years from now one might read, “measuring a variety of bilirubin parameters is considered an appropriate way of assessing the hazard of hyperbilirubinemia and of installing timely treatment in all infants with hyperbilirubinemia.”

Acknowledgments We greatly appreciate the help of Titia Brantsma–van Wulfften Palthe in correcting the English manuscript, and Mrs. Andrea Schreuder (MD, PhD) for data of Fig and Table 3. We are also indebted to Paul Govaert (MD, PhD) for his view on the role of albumin.

refere nces

1. American Academy of Pediatrics Subcommittee on Hyperbilirubinemia. Management of hyperbilirubinemia in the newborn infant 35 or more weeks of gestation. Pediatrics. 2004;114(1):297–316. 2. van Imhoff DE, Dijk PH, Hulzebos CVBARTrial study group, Netherlands Neonatal Research NetworkUniform treatment thresholds for hyperbilirubinemia in preterm infants: background and synopsis of a national guideline. Early Hum Dev. 2011;87(8):521–525. 3. Bowen WR, Porter E, Waters WJ. The protective action of albumin in bilirubin toxicity in newborn puppies. A.M.A. J Dis Child. 1959;48:568. 4. Odell GB. The dissociation of bilirubin from albumin and its clinical implications. J Pediatr. 1959;55:268–279.

420

SE

M I N A R S I N

P

E R I N A T O L O G Y

5. Sinclair CJ. A difference in mortality rate and incidence of kernicterus among premature infants allotted to two prophylactic antibacterial regimens, by William A. Silverman, et al., Pediatrics. 1956;18:614-624. Pediatrics. 1998;102(1 Pt 2):225–227. 6. Pearlman MA, Gartner LM, Lee K, Morecki R, Horoupian DS. Absence of kernicterus in low-birth-weight infants from 1971 through 1976: comparison with findings in 1966 and 1967. Pediatrics. 1978;62(4):460–464. 7. Cashore WJ, Gartner LM, Oh W, Stern L. Clinical application of neonatal bilirubin-binding determinations: current status. J Pediatr. 1978;93(5):827–833. 8. Govaert P, Lequin M, Swarte R, et al. Changes in globus pallidus with (pre)term kernicterus. Pediatrics. 2003;112(6 Pt 1): 1256–1263. 9. Moll M, Goelz R, Naegele T, Wilke M, Poets CF. Are recommended phototherapy thresholds safe enough for extremely low birth weight (ELBW) infants? A report on 2 ELBW infants with kernicterus despite only moderate hyperbilirubinemia. Neonatology. 2011;99(2):90–94. 10. van der Bor M, Ens-Dokkum M, Schreuder AM, Veen S, Brand R, Verloove-Vanhorick SP. Hyperbilirubinemia in low birth weight infants and outcome at 5 years of age. Pediatrics. 1992;89(3):359–364. 11. Oh W, Tyson JE, Fanaroff AA, et al. Association between peak serum bilirubin and neurodevelopmental outcomes in extremely low birth weight infants. Pediatrics. 2003;112(4):773–779. 12. Newman TB, Liljestrand P, Jeremy RJ, et al. Outcomes among newborns with total serum bilirubin levels of 25 mg per deciliter or more. N Engl J Med. 2006;354(18):1889–1900. 13. Lunsing RJ, Pardoen WF, Hadders-Algra M. Neurodevelopment after moderate hyperbilirubinemia at term. Pediatr Res. 2013;73(5):655–660. 14. Watchko JF. Vigintiphobia revisited. Pediatrics. 2005;115(6): 1747–1753. 15. Brown AK, Kim MH, Wu PYK, Bryla DA. Efficacy of phototherapy in prevention and management of neonatal hyperbilirubinemia. Pediatrics. 1985;75(2 Pt 2):393–400. 16. Morris BH, Oh W, Tyson JE, et al. Aggressive vs. conservative phototherapy for infants with extremely low birth weight. N Engl J Med. 2008;359(18):1885–1896. 17. Tyson JE, Pedroza C, Langer J, et al. Does aggressive phototherapy increase mortality while decreasing profound impairment among the smallest and sickest newborns? J Perinatol. 2012;32(9):677–684. 18. 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(2):733–739. 19. Beken S, Aydin B, Zenciroğ ğ lu A, et al. The effects of phototherapy on eosinophil and eosinophilic cationic protein in newborns with hyperbilirubinemia. Fetal Pediatr Pathol. 2014; 33(3):151–156. 20. Levine RL, Fredericks WR, Rapoport SI. Entry of bilirubin into the brain due to opening of the blood–brain barrier. Pediatrics. 1982;69(3):255–259. 21. Wennberg RP, Hance AJ. Experimental bilirubin encephalopathy: importance of total bilirubin, protein binding, and blood–brain barrier. Pediatr Res. 1986;20(8):789–792. 22. Brodersen R, Stern L. Deposition of bilirubin acid in the central nervous system—a hypothesis for the development of kernicterus. Acta Paediatr Scand. 1990;79(1):12–19. 23. Wennberg RP. The blood–brain barrier and bilirubin encephalopathy. Cell Mol Neurobiol. 2000;20(1):97–109. 24. Ostrow JD, Pascolo L, Tiribelli C. Reassessment of the unbound concentrations of unconjugated bilirubin in relation to neurotoxicity in vitro. Pediatr Res. 2003;54(1):98–104. 25. Watchko JF. Kernicterus and the molecular mechanisms of bilirubin-induced CNS injury in newborns. Neuromol Med. 2006;8(4):513–529.

38 (2014) 412–421

26. Cuperus FJ, Schreuder AB, van Imhoff DE, et al. Beyond plasma bilirubin: the effects of phototherapy and albumin on brain bilirubin levels in Gunn rats. J Hepatol. 2013;58 (1):134–140. 27. Odell GB, Cohen SN, Gordes EH. Administration of albumin in the management of hyperbilirubinemia by exchange transfusions. Pediatrics. 1962;30:613–621. 28. Brown AK, Zuelzer WW, Robinson AR. Studies in hyperbilirubinemia. II. Clearance of bilirubin from plasma and intravascular space in newborn infants during exchange transfusion. A.M.A. J Dis Child. 1957;93(3):274–286. 29. Odell GB. Studies in kernicterus I. The protein binding of bilirubin. J Clin Invest. 1959;38(5):823–833. 30. Ebbesen F, Nyboe J. Postnatal changes in the ability of plasma albumin to bind bilirubin. Acta Paediatr Scand. 1983;72(5):665–670. 31. Watchko JF, Maisels MJ. Enduring controversies in the management of hyperbilirubinemia in preterm neonates. Semin Fetal Neonatal Med. 2010;15(3):136–140. 32. Cashore WJ, Oh W, Brodersen R. Reserve albumin and bilirubin toxicity index in infant serum. Acta Paediatr Scand. 1983;72(3):415–419. 33. Robertson A, Karp W, Brodersen R. Bilirubin displacing effect of drugs used in neonatology. Acta Paediatr Scand. 1991;80 (12):1119–1127. 34. Spear ML, Stahl GE, Paul MH, Egler JM, Pereira GR, Polin RA. The effect of 15-hour fat infusions of varying dosage on bilirubin binding to albumin. J Parenter Enteral Nutr. 1985;9 (2):144–147. 35. Amin SB, Harte T, Scholer L, Wang H. Intravenous lipid and bilirubin–albumin binding variables in premature infants. Pediatrics. 2009;124(1):211–217. 36. Amin SB. Effect of free fatty acids on bilirubin–albumin binding affinity and unbound bilirubin in premature infants. J Parenter Enteral Nutr. 2010;34(4):414–420. 37. Ahlfors CE, Parker AE. Bilirubin binding contributes to the increase in total bilirubin concentration in newborns with jaundice. Pediatrics. 2010;126(3):e639–e643. 38. Weisiger RA, Ostrow JD, Koehler RK, et al. Affinity of human serum albumin for bilirubin varies with albumin concentration and buffer composition. J Biol Chem. 2001;276(32): 29953–29960. 39. Cashore WJ, Horwich A, Laterra J, Oh W. Effect of postnatal age and clinical status of newborn infants on bilirubinbinding capacity. Biol Neonate. 1977;32(5–6):304–309. 40. Cashore WJ, Oh W. Unbound bilirubin and kernicterus in lowbirth-weight infants. Pediatrics. 1982;69(4):481–485. 41. Meisel P, Jährig D, Beyersdorff E, Jährig K. Bilirubin binding and acid–base equilibrium in newborn infants with low birthweight. Acta Paediatr Scand. 1988;77(4):496–501. 42. Kozuki K, Oh W, Widness J, Cashore WJ. Increase in bilirubin binding to albumin with correction of neonatal acidosis. Acta Paediatr Scand. 1979;68(2):213–217. 43. Miwa A, Morioka I, Hisamatsu C, et al. Hypoalbuminemia following abdominal surgery leads to high serum unbound bilirubin concentrations in newborns soon after birth. Neonatology. 2011;99(3):202–207. 44. Kapitulnik J, Valaes T, Kaufmann NA, Blondheim SH. Clinical evaluation of Sephadex gel filtration in estimation of bilirubin binding in serum in neonatal jaundice. Arch Dis Child. 1974;49 (11):886–894. 45. Zamet P, Nakamura H, Perez-Robles S, Larroche JC, Minkowski A. The use of critical levels of birth weight and “free bilirubin” as an approach for prevention of kernicterus. Biol Neonate. 1975;26(3–4):274–282. 46. Valaes T, Kapitulnik J, Kaufmann NA, Blondheim SH. Experience with Sephadex gel filtration in assessing the risk of bilirubin encephalopathy in neonatal jaundice. Birth Defects Orig Artic Ser. 1976;12(2):215–228.

S

E M I N A R S I N

P

E R I N A T O L O G Y

47. Cashore WJ, Monin PJ, Oh W. Serum bilirubin binding capacity and free bilirubin concentration: a comparison between sephadex G-25 filtration and peroxidase oxidation techniques. Pediatr Res. 1978;12(3):195–198. 48. Odell GB, Storey GN, Rosenberg LA. Studies in kernicterus: 3. The saturation of serum proteins with bilirubin during neonatal life and its relationship to brain damage at five years. J Pediatr. 1970;76(1):12–21. 49. Waters WJ, Porter E. Indications for exchange transfusion based upon the role of albumin in the treatment of hemolytic disease of the newborn. Pediatrics. 1964;33(5):749–757. 50. Johnson L, Boggs TR Jr. Bilirubin-dependent brain damage: incidence and indications for treatment. Phototherapy of the Newborn: An Overview. Washington DC: National Academy of Sciences; 1974;122–149. 51. Hansen RL, Hughes GG, Ahlfors CE. Neonatal bilirubin exposure and psychoeducational outcome. J Dev Behav Pediatr. 1991;12(5):287–293. 52. Kim MH, Yoon JJ, Sher J, Brown AK. Lack of predictive indices in kernicterus: a comparison of clinical and pathologic factors in infants with or without kernicterus. Pediatrics. 1980;66(6): 852–858. 53. Ritter DA, Kenny JD, Norton HJ, Rudolph AJ. A prospective study of free bilirubin and other risk factors in the development of kernicterus in premature infants. Pediatrics. 1982;69(3): 260–266. 54. Robinson PJ, Rapoport SI. Binding effect of albumin on uptake of bilirubin by brain. Pediatrics. 1987;79(4):553–558. 55. Ostrow JD, Pascolo L, Brites D, Tiribelli C. Molecular basis of bilirubin-induced neurotoxicity. Trends Mol Med. 2004;10(2): 65–70. 56. Falcão AS, Silva RF, Vaz AR, et al. Cross-talk between neurons and astrocytes in response to bilirubin: adverse secondary impacts. Neurotox Res. 2014;26(1):1–15. 57. Odell GB. The distribution and toxicity of bilirubin. Pediatrics. 1970;46(1):16–24. 58. Nakamura H, Yonetani M, Uetani Y, Funato M, Lee Y. Determination of serum unbound bilirubin for prediction of kernicterus in low birthweight infants. Acta Paediatr Jpn. 1992;34(6):642–647. 59. Nakamura H, Takada S, Shimabuku R, Matsuo M, Matsuo T, Negishi H. Auditory nerve and brainstem responses in newborn infants with hyperbilirubinemia. Pediatrics. 1985;75(4): 703–708. 60. Ahlfors CE. Unbound bilirubin associated with kernicterus: a historical approach. J Pediatr. 2000;137(4):540–544. 61. Funato M, Tamai H, Shimada S, Nakamura H. Vigintiphobia, unbound bilirubin, and auditory brainstem responses. Pediatrics. 1994;93(1):50–53. 62. Amin SB, Ahlfors C, Orlando MS, Dalzell LE, Merle KS, Guillet R. Bilirubin and serial auditory brainstem responses in premature infants. Pediatrics. 2001;107(4):664–670. 63. Ahlfors CE, Amin SB, Parker AE. Unbound bilirubin predicts abnormal automated auditory brainstem response in a diverse newborn population. J Perinatol. 2009;29(4):305–309.

38 (2014) 412–421

421

64. Murki S, Kumar P, Majumdar S, Marwaha N, Narang A. Risk factors for kernicterus in term babies with non-hemolytic jaundice. Indian Pediatr. 2001;38(7):757–762. 65. Ahlfors CE, Herbsman O. Unbound bilirubin in a term newborn with kernicterus. Pediatrics. 2003;111(5 Pt 1):1110–1112. 66. Oh W, Stevenson DK, Tyson JE, et al. Influence of clinical status on the association between plasma total and unbound bilirubin and death or adverse neurodevelopmental outcomes in extremely low birth weight infants. Acta Paediatr. 2010;99 (5):673–678. 67. Bender GJ, Cashore WJ, Oh W. Ontogeny of bilirubin-binding capacity and the effect of clinical status in premature infants born at less than 1300 grams. Pediatrics. 2007;120(5): 1067–1073. 68. Lee YK, Daito Y, Katayama Y, Minami H, Negishi H. The significance of measurement of serum unbound bilirubin concentrations in high-risk infants. Pediatr Int. 2009;51(6): 795–799. 69. Jacobsen J, Wennberg RP. Determination of unbound bilirubin in the serum of newborns. Clin Chem. 1974;20(7):783. 70. Karp WB. Biochemical alterations in neonatal hyperbilirubinemia and bilirubin encephalopathy: a review. Pediatrics. 1979;64(3):361–368. 71. Ahlfors CE. Benzyl alcohol, kernicterus, and unbound bilirubin. J Pediatr. 2001;139(2):317–319. 72. McDonagh AF, Maisels MJ. Bilirubin unbound: déjà vu all over again? Pediatrics. 2006;117(2):523–525. 73. Ahlfors CE, Marshall GD, Wolcott DK, Olson DC, Van Overmeire B. Measurement of unbound bilirubin by the peroxidase test using Zone Fluidics. Clin Chim Acta. 2006;365(1–2): 78–85. 74. Huber AH, Zhu B, Kwan T, Kampf JP, Hegyi T, Kleinfeld AM. Fluorescence sensor for the quantification of unbound bilirubin concentrations. Clin Chem. 2012;58(5):869–876. 75. Ahlfors CE. Criteria for exchange transfusion in jaundiced newborns. Pediatrics. 1994;93(3):488–494. 76. Hulzebos CV, van Imhoff DE, Bos AF, Ahlfors CE, Verkade HJ, Dijk PH. Usefulness of the bilirubin/albumin ratio for predicting bilirubin-induced neurotoxicity in premature infants. Arch Dis Child Fetal Neonatal Ed. 2008;93(5):F384–F388. 77. Scheidt PC, Graubard BI, Nelson KB, et al. Intelligence at six years in relation to neonatal bilirubin levels: follow-up of the National Institute of Child Health and Human Development Clinical Trial of Phototherapy. Pediatrics. 1991;87(6): 797–805. 78. Iskander I, El Houchi S, El Shennawy A, et al. Validation of bilirubin/albumin ratio as a predictor of bilirubin induced neurological dysfunction (BIND). PAS Original Science Abstracts–Poster Session. 2013; 4500, 5. 〈http://www.abstracts2 view.com/pas/view.php?nu=PAS13L1_4500.5〉. 79. Hulzebos CV, Dijk PH, van Imhoff DE, et al. together with the BARTrial Study Group. The bilirubin albumin ratio in the management of hyperbilirubinemia in preterm infants to improve neurodevelopmental outcome: a randomized controlled trial—BARTrial. PLoS One. 2014;9(6):e99466.