- Email: [email protected]
Bilirubin Uridine Diphosphate-glucuronosyltransferase Polymorphism as a Risk Factor for Prolonged Hyperbilirubinemia in Japanese Preterm Infants
ARTICLE IN PRESS THE JOURNAL OF PEDIATRICS • www.jpeds.com
ORIGINAL ARTICLES
Bilirubin Uridine Diphosphate-glucuronosyltransferase Polymorphism as a Risk Factor for Prolonged Hyperbilirubinemia in Japanese Preterm Infants Takahide Yanagi, MD, Sayuri Nakahara, MD, and Yoshihiro Maruo, MD, PhD Objective To determine whether a variant of the bilirubin uridine diphosphate-glucuronosyltransferase gene (UGT1A1*6) is a risk factor for prolonged hyperbilirubinemia in preterm infants. Study design UGT1A1 genotypes in 46 Japanese preterm infants (<37 weeks of gestation) were compared with UGT1A1 genotypes in 38 control infants, using polymerase chain reaction-direct sequencing. Prolonged unconjugated hyperbilirubinemia was defined as serum total bilirubin concentration of >150 µmol/L (8.77 mg/dL) beyond 14 days of life. Results In the case group, 41 of 46 infants (89.1%) had a polymorphic variant, c.211G>A, p.G71R (UGT1A1*6). In the control group, 7 of 38 (18.4%) had UGT1A1*6. The allele frequency of UGT1A1*6 was 0.641 in the prolonged hyperbilirubinemia group, which was significantly higher than in the control group (0.092; P < .001). In total, 39 of 46 infants in the case group were breast fed, and only 10 infants in the control group were breast fed. Conclusions These data suggest that UGT1A1*6 is a risk factor for prolonged unconjugated hyperbilirubinemia in preterm infants in Japan. Given the different rate of breast feeding in this study, additional data are necessary for drawing a definitive conclusion. (J Pediatr 2017;■■:■■-■■). yperbilirubinemia in the neonatal period is a common clinical problem that may be associated with genetic factors.1,2 Bilirubin UDP-glucuronosyltransferase (UGT1A1, EC2.4.1.17) is the only enzyme shown to be responsible for bilirubin glucuronidation. Mutations of the gene encoding UGT1A1 (UGT1A1) are known to cause hereditary unconjugated hyperbilirubinemia, Crigler-Najjar syndrome type I (MIM #21880) and type II (MIM #606785), and Gilbert syndrome (MIM #143500).3-5 A missense mutation at nucleotide 211 of UGT1A1 (c.211G>A: UGT1A1*6) is one of the most common causes of Gilbert syndrome in East Asians.6 In previous studies in term infants, we found that UGT1A1*6 is a risk factor for neonatal hyperbilirubinemia in the early neonatal period7 and a genetic cause of prolonged unconjugated hyperbilirubinemia associated with breast feeding in the late neonatal period.8,9 In preterm infants, neonatal hyperbilirubinemia is more prevalent, more severe, and has a more protracted course than in term infants owing to differences in factors influencing bilirubin metabolism, including hepatic and gastrointestinal immaturity, slow postnatal maturation of hepatic bilirubin uptake and conjugation, and delayed initiation of enteral feeding.10,11 It has not been clarified whether genetic background also affects the prevalence of hyperbilirubinemia in preterm infants. The purpose of this study was to determine whether a variant of the bilirubin uridine diphosphate-glucuronosyltransferase gene (UGT1A1*6) is a risk factor for prolonged hyperbilirubinemia in preterm infants.
H
Methods This retrospective case-control study used convenience samples from 48 peripheral blood samples that were sent for UGT1A1 genotype analysis from several institutions across Japan to our university for clinical purposes in the course of diagnosis of prolonged jaundice. Eligible subjects included Japanese preterm infants (<37 weeks of gestation) with prolonged unconjugated hyperbilirubinemia, with written informed parental consent for enrollment. Exclusion criteria were other causes of hyperbilirubinemia, such as hemolytic anemia, liver dysfunction, cholestasis, or hypothyroidism, which were confirmed on declaration by each institution (specific numerical values were not available for all cases). Prolonged hyperbilirubinemia was defined as serum total bilirubin concentration of >150 µmol/L (8.77 mg/dL) beyond 14 days of life.12 The control group comprised 38 Japanese preterm infants who were admitted to our neonatal unit and did not show prolonged hyperbilirubinemia beyond 14 From the Department of Pediatrics, Shiga University of days of life. Medical Science, Otsu, Shiga, Japan
The authors declare no conflicts of interest. 0022-3476/$ - see front matter. © 2017 Elsevier Inc. All rights
UGT1A1
Bilirubin uridine diphosphate-glucuronosyl transferase
reserved. http://dx.doi.org10.1016/j.jpeds.2017.07.014
1 FLA 5.5.0 DTD ■ YMPD9332_proof ■ September 6, 2017
THE JOURNAL OF PEDIATRICS • www.jpeds.com All of the parents provided written informed consent. This study was approved by the ethics committee of Shiga University of Medical Science, Otsu, Shiga, Japan. Based on our previous study, the UGT1A1*6 allele frequency in the Japanese general population is 0.16.9 We consider a 2-fold higher allele frequency (0.32) in the case group as clinically meaningful compared with that in the control group. We calculated that with a sample of 40 patients (80 alleles), the study would have 80% power to detect 0.16 increase of allele frequency in the control group, with a type I error of 5%. For sequencing analysis of UGT1A1 polymorphism, genomic DNA was extracted from the peripheral blood leukocytes of infants using a blood isolation kit (DNA Quick 2; DS Pharma Biomedical, Osaka, Japan). Exons, the promoter region, and the phenobarbital-responsive enhancer module of UGT1A1 were amplified from genomic DNA by polymerase chain reaction. Approximately 100 ng of total genomic DNA was amplified using pairs of oligonucleotide primers as listed in Table I (available at www.jpeds.com). Conditions for polymerase chain reaction were initial denaturation for 2 minutes at 94°C, followed by 30 cycles of 1 minute at 94°C, 1 minute at 60°C, and 2 minutes at 72°C with a Minicycler (MJ Research, Inc, Watertown, Massachusetts). A final extension for 10 minutes at 72°C was performed to ensure complete extension of the polymerase chain reaction products. The sequences of the amplified DNA fragments were determined directly using a Big Dye Terminator v1.1 Cycle Sequencing Kit and Genetic Analyzer ABI Prism 3130xl (Applied Biosystems, Carlsbad, California). The used primers are listed in Table I. Statistical Analyses Genotype results were analyzed as the prolonged hyperbilirubinemia group vs the control group. A c2 analysis was conducted on the allele frequency. The highest values of serum bilirubin concentrations after 14 days of life in the case group for the different genotypes detected were examined by ANOVA, and the Tukey’s honestly significant difference testing was used for pairwise comparison. We performed all statistical analyses using SPSS (version 22; SPSS, Inc, Chicago, Illinois).
Volume ■■
Gestational age Range Median Interquartile range Birth weight (g) Range Mean ± SD Sex (male/female) Nutrition Breast fed Formula fed Mixture Direct bilirubin (mg/dL) Range Mean ± SD
Case group n = 46
Control group n = 38
P value
28w0d-36w6d 34 w 0 d 3.61 w
28w3d-36w6d 34 w 0 d 3.75 w
.390 *
914-2815 2032 ± 462 26/19 (NA 1)
916-2880 1843 ± 456 19/19
39 0 2 (NA 5) (NA 8) 0.00-1.40 0.67 ± 0.42
10 14 14
.064†
NA, data not available. *Mann–Whitney test. †Two-sample t test.
there were no cases of homozygosity for UGT1A1*6 in the control group. The allele frequencies of the UGT1A1*6 was 0.641 in the case group and 0.092 in the control group (c2 = 52.6; P < .001; Tables III and IV). The distribution of UGT1A1 genotypes was evaluated in subgroups consisting of preterm (28-33 weeks of gestation) and late preterm (34-36 weeks of gestation) infants. In the case group, the allele frequencies of UGT1A1*6 were 0.575 and 0.692 in the preterm and late preterm groups, respectively (Table V; available at www.jpeds.com). In the control group, the allele frequencies of UGT1A1*6 were 0.138 and 0.050 in the preterm and late preterm groups, respectively. Maximum serum bilirubin concentrations after 14 days of age in the infants with prolonged hyperbilirubinemia were
Table III. Distribution of UGT1A1 genotypes in the case group and the control group Allele 1
The clinical characteristics of the infants in the prolonged hyperbilirubinemia group and the control group are shown in Table II. Gestational age and body weight were not different between the 2 groups. In total, 39 of 46 infants in the case group were fed with breast milk, and only 10 infants in the control group were breast fed. The results of UGT1A1 analysis of all infants are shown in Table III. In the case group, 41 of 46 infants (89.1%) had UGT1A1*6. A total of 18 cases were homozygous and the other 23 were heterozygous. In the control group, 7 of 38 (18.4%) had UGT1A1*6. All of these 7 infants were heterozygous and
■■ 2017
Table II. Clinical characteristics of infants in the case group and the control group
Genotypes
Results
•
Case group Allele 2
UGT1A1*6 UGT1A1*6 UGT1A1*6 UGT1A1*1 UGT1A1*6 UGT1A1*60 UGT1A1*6 UGT1A1*28 UGT1A1*6 UGT1A1*7 UGT1A1*60 UGT1A1*63 UGT1A1*60 UGT1A1*1 UGT1A1*60 UGT1A1*60 UGT1A1*28 UGT1A1*1 UGT1A1*1 UGT1A1*1 Allele frequency of UGT1A1*6
Control group
(n = 46)
%
(n = 38)
%
18 16 5 1 1 1 1 0 0 3
39.3 34.7 10.8 2.1 2.1 2.1 2.1 0.0 0.0 6.5
0 5 1 1 0 0 13 1 1 16
0.0 13.2 2.6 2.6 0.0 0.0 34.2 2.6 2.6 42.1
0.641* (P < .001)
0.092
UGT1A1*1, wild-type allele; UGT1A1*6, p.G71R; UGT1A1*7, p.Y486D; UGT1A1*28, c.−3279T>G + A (TA) 7TAA; UGT1A1*60, c.−3279T>G; UGT1A1*63, p.P364L. *Significant difference from control group: c2 = 52.6, P < .001.
2
Yanagi, Nakahara, and Maruo FLA 5.5.0 DTD ■ YMPD9332_proof ■ September 6, 2017
■■ 2017
ORIGINAL ARTICLES
Table IV. Allele frequency of UGT1A1 polymorphism in prolonged jaundice group, control group, and Japanese general population
Case group
Control group
Number Number of alleles of alleles (total 92) Frequency (total 76) Frequency UGT1A1*1 UGT1A1*6 UGT1A1*28 UGT1A1*60
23 58 1 7
0.250 0.641 0.010 0.076
51 7 2 16
0.671 0.092 0.026 0.210
General Japanese population*
Frequency 0.151-0.16 0.121-0.15 0.115
*Allelic frequency in general Japanese population as determined in a previous study.9
compared among the 5 UGT1A1 genotypes (Figure). ANOVA and post hoc testing showed a significant difference between those homozygous for UGT1A1*6 and those heterozygous UGT1A1*6 (P = .018).
Figure. Differences in serum bilirubin concentration (maximum value after 14 days of life) depending on genotype in the case group. ANOVA showed a significant difference among genotypes (degrees of freedom: 4, 41, F = 3.094, P < .05). Tukey’s honestly significant difference testing showed significant difference between those homozygous for UGT1A1*6 and those heterozygous for it (P = .018). The distribution of bilirubin concentration in each genotype is as follows: homozygous for UGT1A1*6 (n = 18), 19.3 ± 3.5 mg/dL (mean ± SD); heterozygous for UGT1A1*6 and UGT1A1*1 (n = 16), 15.3 ± 3.1; heterozygous for UGT1A1*6 and UGT1A1*60 (n = 5), 17.5 ± 2.8; other genotypes (n = 4, including compound heterozygous for UGT1A1*6 and UGT1A1*28; UGT1A1*6 and UGT1A1*7; UGT1A1*60 and UGT1A1*63; and UGT1A1*60 and UGT1A1*1), 16.6 ± 4.5; and wild type (n = 3), 14.7 ± 5.6.
Discussion These data suggest that UGT1A1*6 is a risk factor for prolonged unconjugated hyperbilirubinemia in preterm infants. In preterm infants, there are many factors influencing bilirubin metabolism, for example, excessive neonatal red cells, hepatic and gastrointestinal immaturity, slow postnatal maturation of hepatic bilirubin uptake and conjugation, and delayed initiation of enteral feeding.10 These factors may attenuate the impact of genetics on hyperbilirubinemia and genetic background may not be as important in preterm infants as in term infants. In the present study of Japanese preterm infants, the allele frequency of UGT1A1*6 was 0.641 in the case group, which was significantly higher than in the control group (0.092). It was also higher than in the general Japanese population (range, 0.151-0.160; Table IV).9 Similar to what has been described in Japanese term infants,9 the most common genotype in the case group was homozygous for UGT1A1*6 (39.3%), and the second most common was heterozygous for UGT1A1*6 (34.7%). These findings show that the UGT1A1*6 allele is an important causative factor for prolonged hyperbilirubinemia in Japanese preterm infants. UGT1A1*6 is a prevalent polymorphism in East Asia, but is not found among white, black, or West Asian subjects.5,13 Instead, UGT1A1*28 (c.−3279T>G in the promotor region that is linked to A[TA]7TAA in the TATA box) is the most common variant in white, black, and West Asian patients with Gilbert syndrome. In the present study, we detected only 1 infant with the UGT1A1*28 allele in the jaundiced group, and this patient was a compound heterozygote for UGT1A1*6. The allele frequency of UGT1A1*28 in the case group was 0.010, lower than that in the control group (0.026). In our previous study on breast milk jaundice in term infants, we did not detect the UGT1A1*28 allele in the jaundiced group. These findings suggest that the UGT1A1*28 allele causes hyperbilirubinemia in adulthood (Gilbert syndrome), but not during the neonatal period. This finding may explain the difference in the incidence of neonatal hyperbilirubinemia among different ethnic groups.1 Recent studies have reported that preterm infants are at increased risk for kernicterus (bilirubin encephalopathy).10,14,15 Not only acute but also prolonged unconjugated hyperbilirubinemia can cause kernicterus in preterm infants. This problem is becoming increasingly important, but evidence-based guidelines for the management of hyperbilirubinemia in preterm infants have not yet been optimized.16 It is not known whether UGT1A1*6 is a risk factor for kernicterus, but if it is, it might be necessary to take genetics into account for the management of jaundice in preterm infants, for example, by applying a lower threshold for phototherapy and exchange transfusion for East Asians than for other ethnic groups. The findings in this study suggest that future studies should evaluate whether UGT1A1*6 is a risk factor for kernicterus in East Asian preterm infants. We acknowledge that this study has several limitations. First, the case and the control groups were recruited from different
Bilirubin Uridine Diphosphate-glucuronosyltransferase Polymorphism as a Risk Factor for Prolonged Hyperbilirubinemia in Japanese Preterm Infants FLA 5.5.0 DTD ■ YMPD9332_proof ■ September 6, 2017
3
THE JOURNAL OF PEDIATRICS • www.jpeds.com institutions. The subjects in the case group were from several institutions across Japan, and those in the control group were from our neonatal units. Different policies on phototherapy, blood transfusion, and nutrition management between institutions may result in a bias. Information on detailed management of each infant was not obtained in this study, and we should take this into account when evaluating the result of this study. The second limitation is the different rate of breast feeding between the 2 groups, which may be a considerable source of a bias in a case-control study. It is possible that only breast feeding is a risk factor and that genetic differences are not associated with prolonged jaundice in preterm infants. It is not reasonable, however, to assert that compared with the general Japanese population, the 4-fold higher allele frequency of UGT1A1*6 in the case group bears no relation to jaundice at all. We speculate that preterm infants are more likely to develop prolonged jaundice when breast feeding, an external factor, and UGT1A1*6, an internal factor, exist together. Formula-fed infants rarely develop jaundice, even if they have UGT1A1*6. This speculation would explain the result of our investigation. Recent research using humanized UGT1 mice showed that the expression of intestinal UGT1A1, but not hepatic UGT1A1, correlated with glucuronidation of bilirubin in the neonatal period.17 Thus, neonatal jaundice and breast milk play an important role in the suppression of intestinal UGT1A1 expression. The UGT1A1*6 genotype was not evaluated in these models, although these findings imply that breast milk suppresses intestinal UGT1A1 expression in UGT1A1*6 genotype, leading to prolonged jaundice. Third, our investigation included only a small number of extremely premature babies. There were no infants with a gestational age of less than 28 weeks in either group. It seems likely that, with decreasing gestational age, other factors related to immature metabolism of bilirubin may predominate and the importance of genetic factors is reduced. In our study, infants of 28-33 weeks of gestational age in the case group had a lower UGT1A1*6 allele frequency than infants of 3436 weeks of gestational age (0.575 and 0.692, respectively). The difference in allele frequency between the case and the control groups is smaller in 28-33 weeks of gestational age than in 34-36 weeks of gestational age (0.437 and 0.643, respectively). The allele frequency in the case group is, however, still considerably higher (4-fold) than that in the control group in infants 28-33 weeks of gestational age (0.575 and 0.138, respectively), which suggests that genetics has a great impact even in those groups. The association between genetic factors and hyperbilirubinemia in more premature infants should be investigated in the future. ■ Submitted for publication Jan 21, 2017; last revision received May 29, 2017; accepted Jul 7, 2017
Volume ■■ Reprint requests: Takahide Yanagi, MD, Department of Pediatrics, Shiga University of Medical Science, Tsukinowa-cho, Seta, Otsu, Shiga 520-2192, Japan. E-mail: [email protected]
References 1. Newman TB, Easterling MJ, Goldman ES, Stevenson DK. Laboratory evaluation of jaundice in newborns. Frequency, cost, and yield. Am J Dis Child 1990;144:364-8. 2. Fischer AF, Nakamura H, Uetani Y, Vreman HJ, Stevenson DK. Comparison of bilirubin production in Japanese and Caucasian infants. J Pediatr Gastroenterol Nutr 1988;7:27-9. 3. Bosma PJ, Chowdhury NR, Goldhoorn BG, Hofker MH, Oude Elferink RP, Jansen PL, et al. Sequence of exons and the flanking regions of human bilirubin-UDP-glucuronosyltransferase gene complex and identification of a genetic mutation in a patient with Crigler-Najjar syndrome, type I. Hepatology 1992;15:941-7. 4. Aono S, Yamada Y, Keino H, Hanada N, Nakagawa T, Sasaoka Y, et al. Identification of defect in the genes for bilirubin UDP-glucuronosyltransferase in a patient with Crigler-Najjar syndrome type II. Biochem Biophys Res Commun 1993;197:1239-44. 5. Bosma PJ, Chowdhury JR, Bakker C, Gantla S, de Boer A, Oostra BA, et al. The genetic basis of the reduced expression of bilirubin UDPglucuronosyltransferase 1 in Gilbert’s syndrome. N Engl J Med 1995;333:1171-5. 6. Sato H, Adachi Y, Koiwai O. The genetic basis of Gilbert’s syndrome. Lancet 1996;347:557-8. 7. Maruo Y, Nishizawa K, Sato H, Doida Y, Shimada M. Association of neonatal hyperbilirubinemia with bilirubin UDP-glucuronosyltransferase polymorphism. Pediatrics 1999;103:1224-7. 8. Maruo Y, Nishizawa K, Sato H, Sawa H, Shimada M. Prolonged unconjugated hyperbilirubinemia associated with breast milk and mutations of the bilirubin uridine diphosphate- glucuronosyltransferase gene. Pediatrics 2000;106:E59. 9. Maruo Y, Morioka Y, Fujito H, Nakahara S, Yanagi T, Matsui K, et al. Bilirubin uridine diphosphate-glucuronosyltransferase variation is a genetic basis of breast milk jaundice. J Pediatr 2014;165:36-41, e1. 10. Watchko JF, Maisels MJ. Jaundice in low birthweight infants: pathobiology and outcome. Arch Dis Child Fetal Neonatal Ed 2003;88:F455-8. 11. Onishi S, Kawade N, Itoh S, Isobe K, Sugiyama S. Postnatal development of uridine diphosphate glucuronyltransferase activity towards bilirubin and 2-aminophenol in human liver. Biochem J 1979;184:7057. 12. Monaghan G, McLellan A, McGeehan A, Li Volti S, Mollica F, Salemi I, et al. Gilbert’s syndrome is a contributory factor in prolonged unconjugated hyperbilirubinemia of the newborn. J Pediatr 1999;134:441-6. 13. Monaghan G, Ryan M, Seddon R, Hume R, Burchell B. Genetic variation in bilirubin UPD-glucuronosyltransferase gene promoter and Gilbert’s syndrome. Lancet 1996;347:578-81. 14. Bhutani VK, Johnson LH, Shapiro SM. Kernicterus in sick and preterm infants (1999-2002): a need for an effective preventive approach. Semin Perinatol 2004;28:319-25. 15. Okumura A, Kidokoro H, Shoji H, Nakazawa T, Mimaki M, Fujii K, et al. Kernicterus in preterm infants. Pediatrics 2009;123:e1052-8. 16. Maisels MJ, Watchko JF, Bhutani VK, Stevenson DK. An approach to the management of hyperbilirubinemia in the preterm infant less than 35 weeks of gestation. J Perinatol 2012;32:660-4. 17. Fujiwara R, Maruo Y, Chen S, Tukey RH. Role of extrahepatic UDPglucuronosyltransferase 1A1: advances in understanding breast milkinduced neonatal hyperbilirubinemia. Toxicol Appl Pharmacol 2015;289:124-32.
4
Yanagi, Nakahara, and Maruo FLA 5.5.0 DTD ■ YMPD9332_proof ■ September 6, 2017
■■ 2017
ORIGINAL ARTICLES
Table I. Lists of primers used for sequence analysis of UGT1A1 Target regions
Sense primer
Anti-sense primer
Tm (°C)
PCR product (bps)
gtPBREM TATA box– exon 1
5′-CTGGGGATAAACATGGGATG-3′ 5′-AAGTGAACTCCCTGCTACCTT-3′
5′-CACCACCACTTCTGGAACCT-3′ 5′-GCTTGCTCAGCATATATCTGGG-3′
60 60
605 1104
Exon 2– exon 4
5′-CTCTATCTCAAACACGCATGCC-3′
5′-TTTTATCATGAATGCCATGACC-3′
60
1584
Exon 5
5′-GAGGATTGTTCATACCACAGG-3′
5′-GCACTCTGGGGCTGATTAAT-3′
60
488
Region
Primers
gtPBREM TATAbox–exon 1 Exon 1 Exon 1 Exon 1 Exon 1 Exon 2 Exon 3 Exon 4 Exon 5 Exon 5
5′-TGAGTTTATATAACCTC-3′ 5′-CTATTTCATGTCCCCTCTGC-3′ 5′-GTCTTTTGTTAGTCTCGGGC-3′ 5′-TTGTTGTGCAGTAAGTGGGA-3′ 5′-CCATTCTCCTACGTGCCCAG-3′ 5′-AAGGGTTGCATACGGGGAATA-3′ 5′-GGAAGCTGGAAGTCTGGG-3′ 5′-CTAGTTAGTATAGCAGAT-3′ 5′-CAGCTGTGAAACTCAGAG-3′ 5′-TGCTGACAGTGGCCTTCATC-3′ 5′-GGTAGCCATAAGCACAACAT-3′
gtPBREM, phenobarbital responsive enhancer module; PCR, polymerase chain reaction.
Table V. Distribution of UGT1A1 genotypes among infants in the case group (preterm vs late preterm) Preterm (28-33 weeks)
Genotypes Allele 1
Allele 2
UGT1A1*6 UGT1A1*6 UGT1A1*6 UGT1A1*1 UGT1A1*6 UGT1A1*60 UGT1A1*6 UGT1A1*28 UGT1A1*6 UGT1A1*7 UGT1A1*60 UGT1A1*63 UGT1A1*60 UGT1A1*1 UGT1A1*1 UGT1A1*1 Allele frequency of UGT1A1*6
Late preterm (34-36 weeks)
(n = 20)
%
(n = 26)
%
6 9 2 0 0 0 0 3
30.0 45.0 10.0 0.0 0.0 0.0 0.0 15.0
12 7 3 1 1 1 1 0
46.1 26.9 11.5 3.8 3.8 3.8 3.8 0.0
0.575
0.692
UGT1A1*1, wild-type allele; UGT1A1*6, p.G71R; UGT1A1*7, p.Y486D; UGT1A1*28, c. –3279T>G+A (TA) 7TAA; UGT1A1*60, c.–3279T>G; UGT1A1*63, p.P364L. In the control group, the allele frequencies of UGT1A1*6 were 0.138 and 0.050 in the preterm and late preterm groups, respectively.
Bilirubin Uridine Diphosphate-glucuronosyltransferase Polymorphism as a Risk Factor for Prolonged Hyperbilirubinemia in Japanese Preterm Infants FLA 5.5.0 DTD ■ YMPD9332_proof ■ September 6, 2017
4.e1
ORIGINAL ARTICLES
Bilirubin Uridine Diphosphate-glucuronosyltransferase Polymorphism as a Risk Factor for Prolonged Hyperbilirubinemia in Japanese Preterm Infants Takahide Yanagi, MD, Sayuri Nakahara, MD, and Yoshihiro Maruo, MD, PhD Objective To determine whether a variant of the bilirubin uridine diphosphate-glucuronosyltransferase gene (UGT1A1*6) is a risk factor for prolonged hyperbilirubinemia in preterm infants. Study design UGT1A1 genotypes in 46 Japanese preterm infants (<37 weeks of gestation) were compared with UGT1A1 genotypes in 38 control infants, using polymerase chain reaction-direct sequencing. Prolonged unconjugated hyperbilirubinemia was defined as serum total bilirubin concentration of >150 µmol/L (8.77 mg/dL) beyond 14 days of life. Results In the case group, 41 of 46 infants (89.1%) had a polymorphic variant, c.211G>A, p.G71R (UGT1A1*6). In the control group, 7 of 38 (18.4%) had UGT1A1*6. The allele frequency of UGT1A1*6 was 0.641 in the prolonged hyperbilirubinemia group, which was significantly higher than in the control group (0.092; P < .001). In total, 39 of 46 infants in the case group were breast fed, and only 10 infants in the control group were breast fed. Conclusions These data suggest that UGT1A1*6 is a risk factor for prolonged unconjugated hyperbilirubinemia in preterm infants in Japan. Given the different rate of breast feeding in this study, additional data are necessary for drawing a definitive conclusion. (J Pediatr 2017;■■:■■-■■). yperbilirubinemia in the neonatal period is a common clinical problem that may be associated with genetic factors.1,2 Bilirubin UDP-glucuronosyltransferase (UGT1A1, EC2.4.1.17) is the only enzyme shown to be responsible for bilirubin glucuronidation. Mutations of the gene encoding UGT1A1 (UGT1A1) are known to cause hereditary unconjugated hyperbilirubinemia, Crigler-Najjar syndrome type I (MIM #21880) and type II (MIM #606785), and Gilbert syndrome (MIM #143500).3-5 A missense mutation at nucleotide 211 of UGT1A1 (c.211G>A: UGT1A1*6) is one of the most common causes of Gilbert syndrome in East Asians.6 In previous studies in term infants, we found that UGT1A1*6 is a risk factor for neonatal hyperbilirubinemia in the early neonatal period7 and a genetic cause of prolonged unconjugated hyperbilirubinemia associated with breast feeding in the late neonatal period.8,9 In preterm infants, neonatal hyperbilirubinemia is more prevalent, more severe, and has a more protracted course than in term infants owing to differences in factors influencing bilirubin metabolism, including hepatic and gastrointestinal immaturity, slow postnatal maturation of hepatic bilirubin uptake and conjugation, and delayed initiation of enteral feeding.10,11 It has not been clarified whether genetic background also affects the prevalence of hyperbilirubinemia in preterm infants. The purpose of this study was to determine whether a variant of the bilirubin uridine diphosphate-glucuronosyltransferase gene (UGT1A1*6) is a risk factor for prolonged hyperbilirubinemia in preterm infants.
H
Methods This retrospective case-control study used convenience samples from 48 peripheral blood samples that were sent for UGT1A1 genotype analysis from several institutions across Japan to our university for clinical purposes in the course of diagnosis of prolonged jaundice. Eligible subjects included Japanese preterm infants (<37 weeks of gestation) with prolonged unconjugated hyperbilirubinemia, with written informed parental consent for enrollment. Exclusion criteria were other causes of hyperbilirubinemia, such as hemolytic anemia, liver dysfunction, cholestasis, or hypothyroidism, which were confirmed on declaration by each institution (specific numerical values were not available for all cases). Prolonged hyperbilirubinemia was defined as serum total bilirubin concentration of >150 µmol/L (8.77 mg/dL) beyond 14 days of life.12 The control group comprised 38 Japanese preterm infants who were admitted to our neonatal unit and did not show prolonged hyperbilirubinemia beyond 14 From the Department of Pediatrics, Shiga University of days of life. Medical Science, Otsu, Shiga, Japan
The authors declare no conflicts of interest. 0022-3476/$ - see front matter. © 2017 Elsevier Inc. All rights
UGT1A1
Bilirubin uridine diphosphate-glucuronosyl transferase
reserved. http://dx.doi.org10.1016/j.jpeds.2017.07.014
1 FLA 5.5.0 DTD ■ YMPD9332_proof ■ September 6, 2017
THE JOURNAL OF PEDIATRICS • www.jpeds.com All of the parents provided written informed consent. This study was approved by the ethics committee of Shiga University of Medical Science, Otsu, Shiga, Japan. Based on our previous study, the UGT1A1*6 allele frequency in the Japanese general population is 0.16.9 We consider a 2-fold higher allele frequency (0.32) in the case group as clinically meaningful compared with that in the control group. We calculated that with a sample of 40 patients (80 alleles), the study would have 80% power to detect 0.16 increase of allele frequency in the control group, with a type I error of 5%. For sequencing analysis of UGT1A1 polymorphism, genomic DNA was extracted from the peripheral blood leukocytes of infants using a blood isolation kit (DNA Quick 2; DS Pharma Biomedical, Osaka, Japan). Exons, the promoter region, and the phenobarbital-responsive enhancer module of UGT1A1 were amplified from genomic DNA by polymerase chain reaction. Approximately 100 ng of total genomic DNA was amplified using pairs of oligonucleotide primers as listed in Table I (available at www.jpeds.com). Conditions for polymerase chain reaction were initial denaturation for 2 minutes at 94°C, followed by 30 cycles of 1 minute at 94°C, 1 minute at 60°C, and 2 minutes at 72°C with a Minicycler (MJ Research, Inc, Watertown, Massachusetts). A final extension for 10 minutes at 72°C was performed to ensure complete extension of the polymerase chain reaction products. The sequences of the amplified DNA fragments were determined directly using a Big Dye Terminator v1.1 Cycle Sequencing Kit and Genetic Analyzer ABI Prism 3130xl (Applied Biosystems, Carlsbad, California). The used primers are listed in Table I. Statistical Analyses Genotype results were analyzed as the prolonged hyperbilirubinemia group vs the control group. A c2 analysis was conducted on the allele frequency. The highest values of serum bilirubin concentrations after 14 days of life in the case group for the different genotypes detected were examined by ANOVA, and the Tukey’s honestly significant difference testing was used for pairwise comparison. We performed all statistical analyses using SPSS (version 22; SPSS, Inc, Chicago, Illinois).
Volume ■■
Gestational age Range Median Interquartile range Birth weight (g) Range Mean ± SD Sex (male/female) Nutrition Breast fed Formula fed Mixture Direct bilirubin (mg/dL) Range Mean ± SD
Case group n = 46
Control group n = 38
P value
28w0d-36w6d 34 w 0 d 3.61 w
28w3d-36w6d 34 w 0 d 3.75 w
.390 *
914-2815 2032 ± 462 26/19 (NA 1)
916-2880 1843 ± 456 19/19
39 0 2 (NA 5) (NA 8) 0.00-1.40 0.67 ± 0.42
10 14 14
.064†
NA, data not available. *Mann–Whitney test. †Two-sample t test.
there were no cases of homozygosity for UGT1A1*6 in the control group. The allele frequencies of the UGT1A1*6 was 0.641 in the case group and 0.092 in the control group (c2 = 52.6; P < .001; Tables III and IV). The distribution of UGT1A1 genotypes was evaluated in subgroups consisting of preterm (28-33 weeks of gestation) and late preterm (34-36 weeks of gestation) infants. In the case group, the allele frequencies of UGT1A1*6 were 0.575 and 0.692 in the preterm and late preterm groups, respectively (Table V; available at www.jpeds.com). In the control group, the allele frequencies of UGT1A1*6 were 0.138 and 0.050 in the preterm and late preterm groups, respectively. Maximum serum bilirubin concentrations after 14 days of age in the infants with prolonged hyperbilirubinemia were
Table III. Distribution of UGT1A1 genotypes in the case group and the control group Allele 1
The clinical characteristics of the infants in the prolonged hyperbilirubinemia group and the control group are shown in Table II. Gestational age and body weight were not different between the 2 groups. In total, 39 of 46 infants in the case group were fed with breast milk, and only 10 infants in the control group were breast fed. The results of UGT1A1 analysis of all infants are shown in Table III. In the case group, 41 of 46 infants (89.1%) had UGT1A1*6. A total of 18 cases were homozygous and the other 23 were heterozygous. In the control group, 7 of 38 (18.4%) had UGT1A1*6. All of these 7 infants were heterozygous and
■■ 2017
Table II. Clinical characteristics of infants in the case group and the control group
Genotypes
Results
•
Case group Allele 2
UGT1A1*6 UGT1A1*6 UGT1A1*6 UGT1A1*1 UGT1A1*6 UGT1A1*60 UGT1A1*6 UGT1A1*28 UGT1A1*6 UGT1A1*7 UGT1A1*60 UGT1A1*63 UGT1A1*60 UGT1A1*1 UGT1A1*60 UGT1A1*60 UGT1A1*28 UGT1A1*1 UGT1A1*1 UGT1A1*1 Allele frequency of UGT1A1*6
Control group
(n = 46)
%
(n = 38)
%
18 16 5 1 1 1 1 0 0 3
39.3 34.7 10.8 2.1 2.1 2.1 2.1 0.0 0.0 6.5
0 5 1 1 0 0 13 1 1 16
0.0 13.2 2.6 2.6 0.0 0.0 34.2 2.6 2.6 42.1
0.641* (P < .001)
0.092
UGT1A1*1, wild-type allele; UGT1A1*6, p.G71R; UGT1A1*7, p.Y486D; UGT1A1*28, c.−3279T>G + A (TA) 7TAA; UGT1A1*60, c.−3279T>G; UGT1A1*63, p.P364L. *Significant difference from control group: c2 = 52.6, P < .001.
2
Yanagi, Nakahara, and Maruo FLA 5.5.0 DTD ■ YMPD9332_proof ■ September 6, 2017
■■ 2017
ORIGINAL ARTICLES
Table IV. Allele frequency of UGT1A1 polymorphism in prolonged jaundice group, control group, and Japanese general population
Case group
Control group
Number Number of alleles of alleles (total 92) Frequency (total 76) Frequency UGT1A1*1 UGT1A1*6 UGT1A1*28 UGT1A1*60
23 58 1 7
0.250 0.641 0.010 0.076
51 7 2 16
0.671 0.092 0.026 0.210
General Japanese population*
Frequency 0.151-0.16 0.121-0.15 0.115
*Allelic frequency in general Japanese population as determined in a previous study.9
compared among the 5 UGT1A1 genotypes (Figure). ANOVA and post hoc testing showed a significant difference between those homozygous for UGT1A1*6 and those heterozygous UGT1A1*6 (P = .018).
Figure. Differences in serum bilirubin concentration (maximum value after 14 days of life) depending on genotype in the case group. ANOVA showed a significant difference among genotypes (degrees of freedom: 4, 41, F = 3.094, P < .05). Tukey’s honestly significant difference testing showed significant difference between those homozygous for UGT1A1*6 and those heterozygous for it (P = .018). The distribution of bilirubin concentration in each genotype is as follows: homozygous for UGT1A1*6 (n = 18), 19.3 ± 3.5 mg/dL (mean ± SD); heterozygous for UGT1A1*6 and UGT1A1*1 (n = 16), 15.3 ± 3.1; heterozygous for UGT1A1*6 and UGT1A1*60 (n = 5), 17.5 ± 2.8; other genotypes (n = 4, including compound heterozygous for UGT1A1*6 and UGT1A1*28; UGT1A1*6 and UGT1A1*7; UGT1A1*60 and UGT1A1*63; and UGT1A1*60 and UGT1A1*1), 16.6 ± 4.5; and wild type (n = 3), 14.7 ± 5.6.
Discussion These data suggest that UGT1A1*6 is a risk factor for prolonged unconjugated hyperbilirubinemia in preterm infants. In preterm infants, there are many factors influencing bilirubin metabolism, for example, excessive neonatal red cells, hepatic and gastrointestinal immaturity, slow postnatal maturation of hepatic bilirubin uptake and conjugation, and delayed initiation of enteral feeding.10 These factors may attenuate the impact of genetics on hyperbilirubinemia and genetic background may not be as important in preterm infants as in term infants. In the present study of Japanese preterm infants, the allele frequency of UGT1A1*6 was 0.641 in the case group, which was significantly higher than in the control group (0.092). It was also higher than in the general Japanese population (range, 0.151-0.160; Table IV).9 Similar to what has been described in Japanese term infants,9 the most common genotype in the case group was homozygous for UGT1A1*6 (39.3%), and the second most common was heterozygous for UGT1A1*6 (34.7%). These findings show that the UGT1A1*6 allele is an important causative factor for prolonged hyperbilirubinemia in Japanese preterm infants. UGT1A1*6 is a prevalent polymorphism in East Asia, but is not found among white, black, or West Asian subjects.5,13 Instead, UGT1A1*28 (c.−3279T>G in the promotor region that is linked to A[TA]7TAA in the TATA box) is the most common variant in white, black, and West Asian patients with Gilbert syndrome. In the present study, we detected only 1 infant with the UGT1A1*28 allele in the jaundiced group, and this patient was a compound heterozygote for UGT1A1*6. The allele frequency of UGT1A1*28 in the case group was 0.010, lower than that in the control group (0.026). In our previous study on breast milk jaundice in term infants, we did not detect the UGT1A1*28 allele in the jaundiced group. These findings suggest that the UGT1A1*28 allele causes hyperbilirubinemia in adulthood (Gilbert syndrome), but not during the neonatal period. This finding may explain the difference in the incidence of neonatal hyperbilirubinemia among different ethnic groups.1 Recent studies have reported that preterm infants are at increased risk for kernicterus (bilirubin encephalopathy).10,14,15 Not only acute but also prolonged unconjugated hyperbilirubinemia can cause kernicterus in preterm infants. This problem is becoming increasingly important, but evidence-based guidelines for the management of hyperbilirubinemia in preterm infants have not yet been optimized.16 It is not known whether UGT1A1*6 is a risk factor for kernicterus, but if it is, it might be necessary to take genetics into account for the management of jaundice in preterm infants, for example, by applying a lower threshold for phototherapy and exchange transfusion for East Asians than for other ethnic groups. The findings in this study suggest that future studies should evaluate whether UGT1A1*6 is a risk factor for kernicterus in East Asian preterm infants. We acknowledge that this study has several limitations. First, the case and the control groups were recruited from different
Bilirubin Uridine Diphosphate-glucuronosyltransferase Polymorphism as a Risk Factor for Prolonged Hyperbilirubinemia in Japanese Preterm Infants FLA 5.5.0 DTD ■ YMPD9332_proof ■ September 6, 2017
3
THE JOURNAL OF PEDIATRICS • www.jpeds.com institutions. The subjects in the case group were from several institutions across Japan, and those in the control group were from our neonatal units. Different policies on phototherapy, blood transfusion, and nutrition management between institutions may result in a bias. Information on detailed management of each infant was not obtained in this study, and we should take this into account when evaluating the result of this study. The second limitation is the different rate of breast feeding between the 2 groups, which may be a considerable source of a bias in a case-control study. It is possible that only breast feeding is a risk factor and that genetic differences are not associated with prolonged jaundice in preterm infants. It is not reasonable, however, to assert that compared with the general Japanese population, the 4-fold higher allele frequency of UGT1A1*6 in the case group bears no relation to jaundice at all. We speculate that preterm infants are more likely to develop prolonged jaundice when breast feeding, an external factor, and UGT1A1*6, an internal factor, exist together. Formula-fed infants rarely develop jaundice, even if they have UGT1A1*6. This speculation would explain the result of our investigation. Recent research using humanized UGT1 mice showed that the expression of intestinal UGT1A1, but not hepatic UGT1A1, correlated with glucuronidation of bilirubin in the neonatal period.17 Thus, neonatal jaundice and breast milk play an important role in the suppression of intestinal UGT1A1 expression. The UGT1A1*6 genotype was not evaluated in these models, although these findings imply that breast milk suppresses intestinal UGT1A1 expression in UGT1A1*6 genotype, leading to prolonged jaundice. Third, our investigation included only a small number of extremely premature babies. There were no infants with a gestational age of less than 28 weeks in either group. It seems likely that, with decreasing gestational age, other factors related to immature metabolism of bilirubin may predominate and the importance of genetic factors is reduced. In our study, infants of 28-33 weeks of gestational age in the case group had a lower UGT1A1*6 allele frequency than infants of 3436 weeks of gestational age (0.575 and 0.692, respectively). The difference in allele frequency between the case and the control groups is smaller in 28-33 weeks of gestational age than in 34-36 weeks of gestational age (0.437 and 0.643, respectively). The allele frequency in the case group is, however, still considerably higher (4-fold) than that in the control group in infants 28-33 weeks of gestational age (0.575 and 0.138, respectively), which suggests that genetics has a great impact even in those groups. The association between genetic factors and hyperbilirubinemia in more premature infants should be investigated in the future. ■ Submitted for publication Jan 21, 2017; last revision received May 29, 2017; accepted Jul 7, 2017
Volume ■■ Reprint requests: Takahide Yanagi, MD, Department of Pediatrics, Shiga University of Medical Science, Tsukinowa-cho, Seta, Otsu, Shiga 520-2192, Japan. E-mail: [email protected]
References 1. Newman TB, Easterling MJ, Goldman ES, Stevenson DK. Laboratory evaluation of jaundice in newborns. Frequency, cost, and yield. Am J Dis Child 1990;144:364-8. 2. Fischer AF, Nakamura H, Uetani Y, Vreman HJ, Stevenson DK. Comparison of bilirubin production in Japanese and Caucasian infants. J Pediatr Gastroenterol Nutr 1988;7:27-9. 3. Bosma PJ, Chowdhury NR, Goldhoorn BG, Hofker MH, Oude Elferink RP, Jansen PL, et al. Sequence of exons and the flanking regions of human bilirubin-UDP-glucuronosyltransferase gene complex and identification of a genetic mutation in a patient with Crigler-Najjar syndrome, type I. Hepatology 1992;15:941-7. 4. Aono S, Yamada Y, Keino H, Hanada N, Nakagawa T, Sasaoka Y, et al. Identification of defect in the genes for bilirubin UDP-glucuronosyltransferase in a patient with Crigler-Najjar syndrome type II. Biochem Biophys Res Commun 1993;197:1239-44. 5. Bosma PJ, Chowdhury JR, Bakker C, Gantla S, de Boer A, Oostra BA, et al. The genetic basis of the reduced expression of bilirubin UDPglucuronosyltransferase 1 in Gilbert’s syndrome. N Engl J Med 1995;333:1171-5. 6. Sato H, Adachi Y, Koiwai O. The genetic basis of Gilbert’s syndrome. Lancet 1996;347:557-8. 7. Maruo Y, Nishizawa K, Sato H, Doida Y, Shimada M. Association of neonatal hyperbilirubinemia with bilirubin UDP-glucuronosyltransferase polymorphism. Pediatrics 1999;103:1224-7. 8. Maruo Y, Nishizawa K, Sato H, Sawa H, Shimada M. Prolonged unconjugated hyperbilirubinemia associated with breast milk and mutations of the bilirubin uridine diphosphate- glucuronosyltransferase gene. Pediatrics 2000;106:E59. 9. Maruo Y, Morioka Y, Fujito H, Nakahara S, Yanagi T, Matsui K, et al. Bilirubin uridine diphosphate-glucuronosyltransferase variation is a genetic basis of breast milk jaundice. J Pediatr 2014;165:36-41, e1. 10. Watchko JF, Maisels MJ. Jaundice in low birthweight infants: pathobiology and outcome. Arch Dis Child Fetal Neonatal Ed 2003;88:F455-8. 11. Onishi S, Kawade N, Itoh S, Isobe K, Sugiyama S. Postnatal development of uridine diphosphate glucuronyltransferase activity towards bilirubin and 2-aminophenol in human liver. Biochem J 1979;184:7057. 12. Monaghan G, McLellan A, McGeehan A, Li Volti S, Mollica F, Salemi I, et al. Gilbert’s syndrome is a contributory factor in prolonged unconjugated hyperbilirubinemia of the newborn. J Pediatr 1999;134:441-6. 13. Monaghan G, Ryan M, Seddon R, Hume R, Burchell B. Genetic variation in bilirubin UPD-glucuronosyltransferase gene promoter and Gilbert’s syndrome. Lancet 1996;347:578-81. 14. Bhutani VK, Johnson LH, Shapiro SM. Kernicterus in sick and preterm infants (1999-2002): a need for an effective preventive approach. Semin Perinatol 2004;28:319-25. 15. Okumura A, Kidokoro H, Shoji H, Nakazawa T, Mimaki M, Fujii K, et al. Kernicterus in preterm infants. Pediatrics 2009;123:e1052-8. 16. Maisels MJ, Watchko JF, Bhutani VK, Stevenson DK. An approach to the management of hyperbilirubinemia in the preterm infant less than 35 weeks of gestation. J Perinatol 2012;32:660-4. 17. Fujiwara R, Maruo Y, Chen S, Tukey RH. Role of extrahepatic UDPglucuronosyltransferase 1A1: advances in understanding breast milkinduced neonatal hyperbilirubinemia. Toxicol Appl Pharmacol 2015;289:124-32.
4
Yanagi, Nakahara, and Maruo FLA 5.5.0 DTD ■ YMPD9332_proof ■ September 6, 2017
■■ 2017
ORIGINAL ARTICLES
Table I. Lists of primers used for sequence analysis of UGT1A1 Target regions
Sense primer
Anti-sense primer
Tm (°C)
PCR product (bps)
gtPBREM TATA box– exon 1
5′-CTGGGGATAAACATGGGATG-3′ 5′-AAGTGAACTCCCTGCTACCTT-3′
5′-CACCACCACTTCTGGAACCT-3′ 5′-GCTTGCTCAGCATATATCTGGG-3′
60 60
605 1104
Exon 2– exon 4
5′-CTCTATCTCAAACACGCATGCC-3′
5′-TTTTATCATGAATGCCATGACC-3′
60
1584
Exon 5
5′-GAGGATTGTTCATACCACAGG-3′
5′-GCACTCTGGGGCTGATTAAT-3′
60
488
Region
Primers
gtPBREM TATAbox–exon 1 Exon 1 Exon 1 Exon 1 Exon 1 Exon 2 Exon 3 Exon 4 Exon 5 Exon 5
5′-TGAGTTTATATAACCTC-3′ 5′-CTATTTCATGTCCCCTCTGC-3′ 5′-GTCTTTTGTTAGTCTCGGGC-3′ 5′-TTGTTGTGCAGTAAGTGGGA-3′ 5′-CCATTCTCCTACGTGCCCAG-3′ 5′-AAGGGTTGCATACGGGGAATA-3′ 5′-GGAAGCTGGAAGTCTGGG-3′ 5′-CTAGTTAGTATAGCAGAT-3′ 5′-CAGCTGTGAAACTCAGAG-3′ 5′-TGCTGACAGTGGCCTTCATC-3′ 5′-GGTAGCCATAAGCACAACAT-3′
gtPBREM, phenobarbital responsive enhancer module; PCR, polymerase chain reaction.
Table V. Distribution of UGT1A1 genotypes among infants in the case group (preterm vs late preterm) Preterm (28-33 weeks)
Genotypes Allele 1
Allele 2
UGT1A1*6 UGT1A1*6 UGT1A1*6 UGT1A1*1 UGT1A1*6 UGT1A1*60 UGT1A1*6 UGT1A1*28 UGT1A1*6 UGT1A1*7 UGT1A1*60 UGT1A1*63 UGT1A1*60 UGT1A1*1 UGT1A1*1 UGT1A1*1 Allele frequency of UGT1A1*6
Late preterm (34-36 weeks)
(n = 20)
%
(n = 26)
%
6 9 2 0 0 0 0 3
30.0 45.0 10.0 0.0 0.0 0.0 0.0 15.0
12 7 3 1 1 1 1 0
46.1 26.9 11.5 3.8 3.8 3.8 3.8 0.0
0.575
0.692
UGT1A1*1, wild-type allele; UGT1A1*6, p.G71R; UGT1A1*7, p.Y486D; UGT1A1*28, c. –3279T>G+A (TA) 7TAA; UGT1A1*60, c.–3279T>G; UGT1A1*63, p.P364L. In the control group, the allele frequencies of UGT1A1*6 were 0.138 and 0.050 in the preterm and late preterm groups, respectively.
Bilirubin Uridine Diphosphate-glucuronosyltransferase Polymorphism as a Risk Factor for Prolonged Hyperbilirubinemia in Japanese Preterm Infants FLA 5.5.0 DTD ■ YMPD9332_proof ■ September 6, 2017
4.e1