PEDIATRICS Vol. 120 No. 5 November 2007, pp. 1067-1073 (doi:10.1542/peds.2006-3024)
ARTICLE |
Ontogeny of Bilirubin-Binding Capacity and the Effect of Clinical Status in Premature Infants Born at Less Than 1300 Grams
Department of Pediatrics, Brown Medical School, Women and Infants’ Hospital of Rhode Island, Providence, Rhode Island
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
BACKGROUND. Bilirubin is toxic to the brain and enters the brain in unbound form. Serum unconjugated, unbound bilirubin may be a good predictor of bilirubin encephalopathy. Unbound bilirubin levels may depend on the bilirubin-binding capacity of albumin, which has not been described for neonates of <28 weeks’ gestation.
OBJECTIVE. The purpose of this work was to determine the ontogeny of bilirubin-binding capacity and the effect of clinical status in very preterm neonates.
METHODS. A total of 152 neonates (23–31 weeks’ gestational age; 440–1300 g) were enrolled prospectively. At 5 days of age, total serum bilirubin and unbound bilirubin were measured with the unbound bilirubin-A1 analyzer (Arrows Co, Osaka, Japan) and albumin with the Bromocresol-purple method. Scatchard plots were used to estimate bilirubin-binding affinity and capacity. Clinical status for each infant was rated as high, moderate, or low risk by using a modified Score for Neonatal Acute Physiology model. Low risk was considered clinically stable.
RESULTS. Unbound bilirubin has a significant, direct correlation to total bilirubin and is greater in unstable than in stable neonates. For the entire cohort, bilirubin-binding capacity had a direct relationship to gestational age. The bilirubin-binding capacities of infants in the low- and high-risk groups also had a direct relationship to gestational age. Bilirubin-binding capacity was greater in the low-risk group (20.8 ± 4.6 mg/dL; 356 ± 79 µmol/L) than in the moderate- (17.8 ± 3.5 mg/dL; 304 ± 60 µmol/L) or high- (17.3 ± 3.4 mg/dL; 296 ± 58 µmol/L) risk groups. Bilirubin-binding affinity did not differ by clinical risk status or gestational age.
CONCLUSIONS. In very preterm, very low birth weight infants, bilirubin-binding capacity is directly proportional to gestational age. Bilirubin-binding capacity is lower and unbound bilirubin higher in unstable than in stable neonates. These data may be useful in guiding the management of hyperbilirubinemia in very low birth weight infants.
Key Words: bilirubin encephalopathy • kernicterus • bilirubin-binding capacity • bilirubin-binding affinity
Abbreviations: TSB—total serum bilirubin • UB—unbound bilirubin • BBC—bilirubin-binding capacity • SNAP-PE—Score for Neonatal Acute Physiology Perinatal Extension
Although extreme hyperbilirubinemia has been associated with acute bilirubin encephalopathy and kernicterus, the long-term impact of modest hyperbilirubinemia in very low birth weight neonates is unknown. Treatment protocols for phototherapy and exchange transfusion use total serum bilirubin (TSB) as the parameter. Wennberg et al1 recently concluded that experimental and clinical data suggest that measurement of unbound bilirubin (UB) in newborns with hyperbilirubinemia may improve the risk assessment for neurotoxicity. Cellular uptake of UB varies in different tissues.2 UB may vary widely for a given TSB, because the binding capacity of albumin may be altered by many clinical and physiologic variables. Bilirubin-binding capacity (BBC) has been described for sick and well neonates of >28 weeks’ gestation3 but not for the very preterm and extremely preterm neonates. With increasing survival of these infants,4 particularly with the retrospective association between peak serum bilirubin and neurodevelopmental impairment,5 there is a need to clarify the risk of hyperbilirubinemia in this group.
The primary hypothesis of this study was that BBC is directly related to gestational age and is influenced by clinical status in a very preterm and very low birth weight population. The secondary hypothesis was that BBC is greater (and, hence, UB levels lower at a given TSB) in clinically stable than in unstable neonates at any gestational and postnatal age.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Subjects
Neonates born at or transferred to Women and Infants Hospital of Rhode Island, with birth weight between 401 and 1300 g and estimated gestational age from 23 weeks and 1 day through 30 weeks and 6 days, were eligible and enrolled when parental consent was obtained. Gestational age was determined by the best obstetrical estimate of first-trimester fetal ultrasound for fetal dating, last menstrual period, or Ballard scores. Infants were excluded if they had lethal chromosomal, craniofacial, or other congenital abnormalities or if they died within 48 hours of life. The study underwent Women and Infants’ Hospital Institutional Review Board review and approval, and informed consent was obtained.
Between February 2003 and November 2004, 152 neonates were enrolled. One pair of twins was subsequently withdrawn per parental request. Blood samples were collected, spun, and stored according to protocol with sufficient serum for serial binding capacity measurements from 137 neonates. Twenty-four samples did not exhibit 2-site binding, having linear low-affinity binding even at low bilirubin levels, hence making BBC estimation unreliable. Eight charts did not have enough data for risk assessment, leaving 105 risk-classed patient sera with BBC estimates.
Study Design
Only 1 serum sample was collected per infant to minimize the physiologic impact of blood withdrawal. This was collected on day 5 ± 1 (SD) of life, at the expected peak of TSB, taken from routine blood draws during the evaluation of hyperbilirubinemia. Each sample was 0.3 to 0.5 mL of whole blood, shielded from light, and centrifuged, and then the serum was separated and frozen at –20°C within 2 hours of extraction for later analysis. Samples were labeled with a unique identifier for storage, with the code correlating to patient identification kept in a protected logbook.
Identity was blinded during total bilirubin, UB, and albumin measurement and analysis. Physiologic data were collected, and clinical risk status was assigned, without knowledge of UB measurement or BBC calculations. Care decisions for the infant were made according to standard management with indirect/total bilirubin and without knowledge of UB, BBC, or assigned clinical risk status.
Samples were thawed within 1 hour of UB determination. UB and TSB were measured concurrently using the UB-A1 analyzer (Arrows Co, Osaka, Japan) at the recommended sample dilution (1:43), at 29°C, with buffer and UB and TSB standards from the Arrows Reagent kit.6 Measurement error was estimated by repeat testing in triplicate when enough serum was available; SE was 0.086 mg/dL for TSB and 0.017 µg/dL for UBC. Concurrent serum albumin concentration was measured with the Bromocresol-purple method as described by Pinnell and Northam.7 UB was measured at increasing TSB concentrations, with 2 µL of solution (0.001 g of bilirubin-9
from Frontier Scientific [Logan, UT] suspended in 25 µL of 0.1 M EDTA and 10 µL of 10 N NaOH, then diluted with 1 mL of deionized H2O, set >1 hour, split, and measured at 3 concentrations) per 25 µL of patient sera. For sera with clearly distinct high-affinity binding affinities represented by the Scatchard plot (5–7 points), an estimate of primary site binding affinity (slope of K1) was done, and binding capacity (K1 intercept of the x-axis) was calculated from the raw data. Thus, BBC was an estimate of the total bilirubin at which saturation of the primary binding site on albumin would occur in the absence of secondary and nonspecific binding sites.
The presence of multiple binding sites was determined by Scatchard transformation of UB versus incremental total bilirubin:albumin molar ratios. Typically, a very high affinity primary binding (K1) was noted, followed by lower affinity sites (K2 and K3) at increasing total bilirubin levels. For these samples, BBC and affinity calculations were made. A subset of these sera (n = 15) had sufficient data points for confirmation of the K1 estimate with nonlinear regression, using GraphPad Prism (GraphPad, San Diego, CA).8 For some sera with high TSB at baseline (n = 14), K1 estimation was not possible, because primary binding sites were saturated. Other sera (n = 10) exhibited low binding affinity even at low total bilirubin levels. This "K2 analysis," segregating the sera into those demonstrating primary versus secondary site binding at the onset, was done independently by 2 investigators, using assessment of the Scatchard plot linearity, initial total bilirubin:albumin molar ratio, deviation from population mean affinities (K1 vs K2), and deflection point between dominant site regression lines.
Clinical Data Collection
Perinatal and neonatal clinical data included gestational age derived by best obstetrical estimate, Apgar scores at 5 minutes, gender, race, and birth weight. Physiologic variables for risk assessment during the first and second 24-hour periods after admission to the NICU were recorded, including (1) axillary temperature, (2) lowest mean arterial blood pressure by transducer connected to an arterial catheter or Dynamap, (3) blood gas data with the lowest sustained PO2/fraction of inspired oxygen ratio, with maximal mode of ventilation support, and the lowest recorded arterial pH or venous pH adjusted +0.1, and (4) average urine output in milliliters per kilogram per hour as measured by nursing staff during routine care. The presence of seizures observed by a caretaker in the first 24 hours was also noted. Small-for-gestational-age status was considered as birth weight less than the 3rd percentile for gestational age.
Infants were classified by clinical risk, stratified into a categorical, ordinal variable with 3 possible values: high, moderate, or low risk. The risk class was determined by clinical and physiologic data, augmented by the concurrent degree of intervention provided by the clinicians. Measures of hemodynamics, oxygenation, temperature, metabolism, acid/base, and renal status were recorded and then processed through a weighted risk algorithm designed to balance the importance of each variable. The weighting is a simplification from the validated Score for Neonatal Acute Physiology Perinatal Extension (SNAP-PE)-II logistic model for mortality risk in the perinatal period.9 The same 5 physiologic variables and cutoff parameters were used in the current risk assignment as in the SNAP-PE model. On an a priori basis, a score from 0 to 5 was assigned for each variable, in proportion to the weights in the original multiple regression model. Additional points were added based on birth weight, small-for-gestational-age status, low 5-minute Apgar score suggesting difficult transition, or persistent seizures in the first 24 hours life consistent with hypoxic-ischemic encephalopathy. The weighted risk scoring is summarized in Table 1. The net risk score was the sum of all of the risk points, which was labeled "high risk" for a score of 
8, "moderate risk" for a score of 4 to 7, and "low risk" for a score of 
3. These risk classifications correlate with a 10%, 1% to 10%, and <1% mortality risk in the SNAP-PE study, respectively.
|
Sample Size
Cashore et al3 found that the difference in BBC between sick versus well was larger in the term than in preterm neonates. The projected effect size in this study, 2.1 mg/dL, was divided among 3 risk groups, with assumed uneven distribution (high risk: 15.5; moderate risk: 16.2; low risk: 17.6), with SDs at 0.6 and 0.5 from sick and well low birth weight neonates, respectively. A priori grouping of infants into 3 gestational age categories was done at 23 to 25, 26 to 27, and 28 to 30 weeks. Three gestational age clusters for 3 risk groups (9 cells), an
value of .05, and a power of 90% result in a net sample size of 135 patients. As noted above, the actual number of risk-adjusted BBC data points is 105, which precluded the intended cluster analysis but was sufficient for logistic regression.
Data Analysis
Risk-class grouping differed according to the time interval of assessment. Among the 105 neonates with sufficient serum and risk-assessment data, 30%, 36%, and 34% were at low, moderate, and high risk, respectively, when classified by using the first 24 hours of physiologic data. Results were the same whether the first 24-hour data or average 48-hour data were used: the tables and figures, therefore, present only the first 24-hour data.
Linear regression analyses (Pearson) were performed to compare UB and BBC among the risk groups, with means compared by analysis of variance and Bonferroni multiple comparisons. Statistical analysis was done using Graphpad Prism. Risk-group analysis within gestational age clusters was not significant, in part because of inadequate numbers of low-risk/low-gestation neonates.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Demographics and Clinical Data
As shown in Table 2, the neonates at high risk are of significantly lower gestational age and birth weight and more likely to be boys. There is no difference in the distribution of the Hispanic and black population among the risk groups.
|
Outcomes that proxy for severity of illness separate into groups as expected. Neonates at high risk are more likely to receive surfactant, be ventilated at 48 hours, and require oxygen at 36 weeks’ after conception age. The raw SNAP-PE scores differ significantly between the risk groups. No patient-specific outcomes are measured.
Total and Unbound Serum Bilirubin
The study infants’ actual serum total and UB data, before titration, are summarized in Table 3. Total bilirubin is not different between the groups, but UB shows a significant (analysis of variance: P < .01) increase with increasing risk. Albumin is statistically greater (analysis of variance: P < .001) in the low-risk group than in the other 2 risk groups but relatively equivalent from a clinical standpoint. Gestational age has no significant effect on albumin, total bilirubin, or UB in any of the risk groups (data not shown). As shown in Fig 1, the direct correlation between UB and TSB is significant (r2 = 0.376; P < .001), as well as within each of the risk groups: low (r2 = 0.230; P < .001), moderate (r2 = 0.389; P < .001) and high (r2 = 0.486; P < .001). The linear regression between unbound and total bilirubin has a different slope in each risk group (F test, P = .03), and each slope is significantly nonzero (F test; P < .001). The SEs of replicate measurements were 0.003 ± 0.002 g/dL of albumin, 0.21 ± 0.14 mg/dL of TSB, and 0.04 ± 0.04 µg/dL of UB.
|
|
Bilirubin-Binding Capacity
Figure 2 shows the relationship between gestational age and BBC expressed as milligrams per deciliter of bilirubin with different risk groups. As a group, irrespective of risk category, BBC has a direct relationship to gestational age (r2 = 0.187; P < .001). This relationship to gestational age holds in the low-risk (r2 = 0.179; P < .05) and high-risk (r2 = 0.139; P < .05) groups but not the moderate-risk group. The slopes of the regression lines are not significantly different. BBC is greater (analysis of variance, P < .001) in the low-risk group (20.8 ± 4.6 mg/dL; 356 ± 79 µmol/L) than in the moderate- (17.8 ± 3.5 mg/dL; 304 ± 60 µmol/L) or high- (17.3 ± 3.4 mg/dL; 296 ± 58 µmol/L) risk groups.
|
Bilirubin-Binding Affinity
Binding affinity (mean K1 slope) is not significantly different by analysis of variance between clinical risk groups and gestational age (data not shown). The 3 K values are distinct from each other (K1: 11.4 ± 5.9 [95% confidence interval: 10.3–12.4]; K2: 2.6 ± 1.4 [95% confidence interval: 2.3–2.8]; and K3: 1.2 ± 0.88 [95% confidence interval: 0.9–1.4]; all values are x107 L·mol–1).
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Although the incidence of kernicterus has decreased significantly with the widespread use of phototherapy and exchange transfusions over recent decades, an increase has been noted particularly in the near-term and term infants over the past few years.10 Guidelines for appropriate bilirubin levels at which to initiate such interventions have been updated periodically for near-term or term infants.11,12 Lack of good scientific evidence for very preterm neonates resulted in publication of guidelines projected from term TSB data,13,14 although kernicterus has been reported in premature neonates at low TSB levels.15 Comparative neuropathy study has shown that the localized patterns in term kernicterus may differ from the more diffuse uptake in premature neonates with hyperbilirubinemia.16 Similar patterns have been shown by MRI for term17 and preterm infants with kernicterus.18 Peak serum bilirubin has been retrospectively correlated with hearing impairment and Bayley Psychomotor Development Index of <70.4
Our data support the relationship between UB and TSB that has been historically assumed in the clinical management of very low birth weight infants with mild hyperbilirubinemia. Neonates at high risk have a greater UB at a given TSB than neonates at either low or moderate risk, which justifies the use of clinical status to tailor interventions. However, the difference in mean UB between the risk groups is statistically but not clinically impressive, which, along with a high UB-TSB coefficient of variation, suggest that measurement of UB itself may be of even more use. Using the same methodology in a general population of newborns with jaundice, a UB between 0.86 and 1.19 µg/dL correlated with an increased likelihood of kernicterus,19 consistent with previous studies showing reversible acute bilirubin encephalopathy at UB levels of >1.0 µg/dL,20 as well as increased risk of kernicterus.21,22 Although our population mean UB is in the nontoxic range, 31%, 16%, and 7%, respectively, of the patients at high, moderate, and low risk had a UB level of >1.0 µg/dL.
For unclear reasons, a subset (14 with high initial TSB and 10 with low TSB) of our patients have no identifiable K1 binding, because only a single shallow slope was observed on Scatchard plot. Both the mean TSB (10.2 mg/dL) and UB (0.89µg/dL) are higher for these patients than for the other patients. Although binding affinity and capacity measurements could, therefore, not be made for these sera, these neonates are potentially at the greatest risk for bilirubin encephalopathy. Fully 37% of these sera with no BBC measurement have an initial UB level of >1.0 µg/dL.
The majority of patients in the current study have BBC estimates higher than those retrospectively associated with kernicterus using the Sephadex method.23 Clinical or postmortem evidences of kernicterus were not observed in our patients. BBC is directly proportional to gestational age, which is not accounted for by significant relationships of albumin, TSB, or UB with gestational age. As expected, UB has a much closer relationship with TSB than either gestational age or BBC. Second, the significance of BBC versus gestational age in the high- and low-risk groups is lost in the moderate-risk group, which may represent a gray zone in the risk stratification. Third, because >80% of the variance in BBC is not related to gestational age, other serum components or clinical factors must have an important role. Competitors for albumin binding include many drugs and molecules, such as fatty acids.24–26 Our patients received exogenous free fatty acids via parenteral nutrition. However, we did not measure serum fatty acid concentrations. Analysis of BBC versus raw risk score does not identify which clinical factors contribute most (data not shown).
Bilirubin-binding affinity is highly variable within our population and does not correlate with clinical risk severity or gestational age. The mean affinity is >1 order of magnitude higher than that reported for term newborns with other methods.2 The UB-A1 analyzer uses a kinetic peroxidase method that may underestimate the actual serum UB concentration. UB measurement can be complicated by dilution of competitors, interference by conjugated bilirubin, and failure to correct for the rate-limiting albumin-bilirubin dissociation step. Interference with UB measurement by bilirubin photoisomers must also be considered given the prevalence of phototherapy in this population, although the effect in the clinical range is expected to be minimal.27 A combined peroxidase-diazo assay adjusting for these factors found UB to be poorly correlated with, and consistently greater than, that measured with the kinetic peroxidase method used in this study.28 A fivefold underestimation of the mean UB using the standard peroxidase method has been shown by using minimal dilution methods.29 Thus, the next generation of UB measurement may be more precise to bring UB to the forefront of hyperbilirubinemia management.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Within the gestational age range of 23 to 31 weeks and birth weight between 401 and 1300 g, the total BBC of the primary binding sites of albumin is directly proportional to maturity. BBC is greater in stable than in unstable neonates. UB is lower in stable than in unstable neonates. UB correlates strongly with total bilirubin and is affected by clinical status. No consistent relationship was seen between binding affinity and gestational age or physiologic risk. These data may be useful in guiding the management of hyperbilirubinemia in these infants, particularly if the National Institute of Child Health and Human Development Neonatal Network Phototherapy Trial identifies significant relationships between developmental outcome and either UB or BBC.
|
ACKNOWLEDGMENTS |
|---|
Dr Oh was supported by grant 2 U10 HD27904 from the National Institute of Child Health and Human Development Neonatal Research Network.
|
FOOTNOTES |
|---|
Accepted May 30, 2007.
Address correspondence to George Jesse Bender, MD, Department of Pediatrics, Brown Medical School, Women and Infants’ Hospital of Rhode Island, 100 Dudley St, Providence, RI 02860. E-mail: gbender{at}wihri.org
The authors have indicated they have no financial relationships relevant to this article to disclose.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
1. Wennberg RP, Ahlfors CE, Bhutani VK, Johnson LH, Shapiro SM. Toward understanding kernicterus: a challenge to improve the management of jaundiced newborns [published correction appears in Pediatrics. 2006;117:1467]. Pediatrics. 2006;117 :474 –485
2. Ostrow JD, Pascolo L, Shapiro SM, Tiribelli C. New concepts in bilirubin encephalopathy. Eur J Clin Invest. 2003;33 :988 –997[CrossRef][Web of Science][Medline]
3. Cashore WJ, Horwich A, Karotkin EH, Oh W. Influence of gestational age and clinical status on bilirubin-binding capacity in newborn infants.
Am J Dis Child. 1977;131
:898
–901
4. Lemons JA, Bauer CR, Oh W, et al. Very low birth weight outcomes of the National Institute of Child Health and human development neonatal research network, January 1995 through December 1996. NICHD Neonatal Research Network. Pediatrics. 2001;107(1) . Available at: www.pediatrics.org/cgi/content/full/107/1/e1
5. 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
:773
–779
6. Shimabuku R, Nakamura H. Total and unbound bilirubin determination using an automated peroxidase micromethod. Kobe J Med Sci. 1982;28 :91 –104[Medline]
7. Pinnell AE, Northam BE. New automated dye-binding method for serum albumin determination with bromcresol purple.
Clin Chem. 1978;24
:80
–86
8. Graphpad Prism 4 for Windows [computer program]. San Diego, CA: GraphPad Software, Inc; 2005
9. Richardson DK, Corcoran JD, Escobar GJ, Lee SK. SNAP-II and SNAPPE-II: Simplified newborn illness severity and mortality risk scores. J Pediatr. 2001;138 :92 –100[CrossRef][Web of Science][Medline]
10. Bhutani VK, Johnson LH, Maisels J, Newman TB. Kernicterus: epidemiological strategies for its prevention through systems-based approaches. J Perinatol. 2004;24 :650 –662[CrossRef][Medline]
11. American Academy of Pediatrics, Provisional Committee for Quality Improvement and Subcommittee on Hyperbilirubinemia. Practice parameter: management of hyperbilirubinemia in the healthy term newborn [published correction appears in Pediatrics. 1995;9:458–461].
Pediatrics. 1994;94
:558
–565
12. American Academy of Pediatrics, Subcommittee on Hyperbilirubinemia. Management of hyperbilirubinemia in the newborn infant 35 or more weeks of gestation [published correction appears in Pediatrics. 2004;114:1138].
Pediatrics. 2004;114
:297
–316
13. Bratlid D. Criteria for treatment of neonatal jaundice. J Perinatol. 2001;21 :S88 –S92[CrossRef][Medline]
14. Cashore WJ. Bilirubin and jaundice in the micropremie. Clin Perinatol. 2000;27 :171 –179[CrossRef][Web of Science][Medline]
15. Gartner LM, Snyder RN, Chabon RS. Kernicterus: high incidence in premature infants with low serum bilirubin concentrations.
Pediatrics. 1970;45
:906
–917
16. Ahdab-Barmada M, Moossy J. The neuropathology of kernicterus in the premature neonate: diagnostic problems. J Neuropathol Exp Neurol. 1984;43 :45 –56[Web of Science][Medline]
17. Penn AA, Enzmann DR, Hahn JS, Stevenson DK. Kernicterus in a full term infant.
Pediatrics. 1994;93
:1003
–1006
18. Govaert P, Lequin M, Swarte R, et al. Changes in globus pallidus with (pre)term kernicterus.
Pediatrics. 2003;112
:1256
–1263
19. Ahlfors C. Unbound bilirubin associated with kernicterus: a historical approach. J Pediatr. 2000;137 :540 –544[CrossRef][Web of Science][Medline]
20. 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
:703
–708
21. 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
:260
–266
22. Nakamura H, Yonetani M, Uetani Y, Funato M, Lee Y. Determination of serum unbound bilirubin for prediction of kernicterus in low birth weight infants. Acta Paediatr Jpn. 1992;34 :642 –647[Medline]
23. Cashore WJ, Oh W. Unbound bilirubin and kernicterus in low-birth-weight infants.
Pediatrics. 1982;69
:481
–485
24. Robertson A, Karp W, Brodersen R. Bilirubin displacing effect of drugs used in neonatology. Acta Paediatr Scand. 1991;80 :1119 –1127[Web of Science][Medline]
25. Robertson A, Brodersen R. Effect of drug combinations on bilirubin-albumin binding. Dev Pharmacol Ther. 1991;17 :95 –99[Web of Science][Medline]
26. Rubin M, Harell D, Naor N, et al. Lipid infusion with different triglyceride cores: effect on plasma lipids and bilirubin binding in premature infants.
J Parenter Enteral Nutr. 1991;15
:642
–646
27. Itoh S, Yamakawa T, Onishi S, Isobe K, Manabe M, Sasaki K. The effect of bilirubin photoisomers on unbound-bilirubin concentrations estimated by the peroxidase method. Biochem J. 1986;239 :417 –421[Web of Science][Medline]
28. Ahlfors CE. Measurement of plasma unbound unconjugated bilirubin. Anal Biochem. 2000;279 :130 –135[CrossRef][Web of Science][Medline]
29. Ahlfors CE, Marshall GD, Wolcott DK, Olson DC, VanOvermeire B. Measurement of unbound bilirubin by the peroxidase test using Zone Fluidics. Clin Chim Acta. 2006;365 :78 –85[CrossRef][Web of Science][Medline]
PEDIATRICS (ISSN 1098-4275). ©2007 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  S. B. Amin and C. E. Ahlfors Bilirubin-binding Capacity in Premature Infants Pediatrics, April 1, 2008; 121(4): 872 - 873. [Full Text] [PDF]
|
|
|
|
|
|
A. F. McDonagh Bilirubin-binding Capacity in Low Birthweight Premature Infants Pediatrics, April 1, 2008; 121(4): 866 - 867. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
