TNF-238 polymorphism may predict bronchopulmonary dysplasia among preterm infants in the Egyptian population

Bronchopulmonary dysplasia (BPD) remains as a major and increasing burden in Egypt.RATIONALE: To determine whether alleles of TNFα-238G > A affect the risk of BPD or the severity of BPD in preterm infants in Egypt.STUDY DESIGN: We prospectively
of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Pediatric Pulmonology 48:699–706 (2013) TNF-238 Polymorphism May PredictBronchopulmonary Dysplasia Among PretermInfants in the Egyptian Population Nasser A. Elhawary,  PhD , 1,2 * Mohammed T. Tayeb,  PhD , 1 Shereen Abdel-Ghafar,  MD , 3 Mona Rashad,  MD , 4 and Abdel-Aziz Alkhotani,  MD 5 Summary.  Bronchopulmonary dysplasia (BPD) remains as a major and increasing burden inEgypt. Rationale: To determine whether alleles of TNF a -238G  >  A affect the risk of BPD orthe severity of BPD in preterm infants in Egypt. Study Design: We prospectively genotyped 220premature neonates (birth weight  < 1,500 g and gestational age 26–32 weeks) for the   238polymorphism, and assessed the clinical risk factors for BPD in our study populations. Infantswith BPD were mechanically ventilated. Results: Infants who developed BPD (n  ¼  120) had ayounger gestational age (31.0    2.1 weeks vs. 34.3    1.5 weeks) and lower birth weight(1,490    360 g vs. 1,880    520 g) than infants who did not develop BPD (n  ¼  100). Resultsof antenatal steroid supplementation, surfactant therapy, or sepsis might affect the geneticmodulation of BPD. The   238G  >  A polymorphism was associated with a twofold risk of BPD(OR  ¼  2.86; 95% confidence interval, 1.35–3.83). Despite the dominance of the G allele in theEgyptian population, the   238A allele was more common among infants with BPD (23%)than among infants without BPD (15%). The A allele occurred less often in infants with mildBPD (9%) than in infants with severe (39%) or moderate (52%). The AA genotype occurred in15% of cases but in none of the controls. Conclusion: The TNF a   238G  >  A polymorphism—particularly the presence of an A allele—should be evaluated as a biomarker to predict theclinical outcome of preterm infants with BPD in Egypt. Even the presence of one copy of thismutant allele appears to be sufficient to influence the severity of disease.  Pediatr Pulmonol.2013; 48:699–706.   2012 Wiley Periodicals, Inc. Key words: bronchopulmonary dysplasia; low birth weight; TNF a   238G  >  A;polymorphism; Egyptian infants.Funding source: none reported. INTRODUCTION Bronchopulmonary dysplasia (BPD) was describedby Northway et al. 1 in 1967 as a lung injury in preterminfants resulting from oxygen supplementation and me-chanical ventilation. BPD is recognized as a systemicsyndrome associated with neurodevelopmental deficits,cerebral palsy, cognitive impairments, failure to thrive,corpulmonale, and pulmonary hypertension. 1 High rates of in utero and perinatal exposure to infec-tion may be causally related to preterm delivery andsubsequent lung injury. 2 In utero BPD may begin withinflammation, which may postnatally impair lung devel-opment and give rise to alveolar hypoplasia. 3 Althoughthe incidence of BPD has not declined substantiallyover time, improved neonatal intensive care—such asthe use of surfactant replacement—has led to enhancedsurvival and other clinical outcomes among infants with 1 Faculty of Medicine, Department of Medical Genetics, Umm Al-QuraUniversity, Makkah, Kingdom of Saudi Arabia. 2 Faculty of Medicine, Medical Genetics Center, Ain Shams University,Cairo, Egypt. 3 El–Gala´ Hospital, Neonatal Intensive Care Unit, Cairo, Egypt. 4 Faculty of Medicine, Department of Pediatrics, Ain Shams University,Cairo, Egypt. 5 Department of Pediatrics, Faculty of Medicine, Umm Al-Qura University,Makkah, Kingdom of Saudi Arabia.The authors have reported no conflicts of interest.*Correspondence to: Nasser Attia Elhawary, PhD, Faculty of Medicine,Department of Medical Genetics, Umm Al-Qura University, P.O. 57543,21955 Makkah, Kingdom of Saudi Arabia.E-mail:, nasgenet@hotmail.comReceived 11 February 2012; Accepted 12 November 2012.DOI 10.1002/ppul.22748Published online 28 January 2013 in Wiley Online Library(  2012 Wiley Periodicals, Inc.  BPD. 4 Structural changes in the lung and impairment of lung function persist in patients with BPD, so survivorsare at high risk for chronic obstructive pulmonary dis-ease (COPD), pulmonary hypertension, and perhapsother adult lung diseases. 5 Several research studies suggest that siblings, mono-zygotic twins, and dizygotic twins might share geneticsusceptibility to BPD. 6–9 To date, the literature includesfew genome-wide association studies on this topic.However, SPOCK2 (MIM 607988, located on 10q21–q23)has recently been explored as a new candidate suscepti-bility gene for BPD. 10 Other candidate genes that havebeen studied include  TNF  a  (MIM 191160), mannosebinding lectin (  MBL2 , MIM 154545), vascular endothe-lial growth factor ( VEGF  , MIM 192240), matrix metal-loproteinase16 (  MMP16  , MIM 602262), pulmonarysurfactant-associated protein B ( SFTPB , MIM 178640), 11 and ATP-binding cassette A3 (  ABCA3 , MIM 601615). 12 TNF is located at 6p21.1–6p21.3, with a physical dis-tance of about 13 kb between the  TNF  a  and  TNF  b genes. 13 Growing evidence suggests that TNF a  is im-plicated in the pathogenesis of several lung diseasesthrough a multifunctional cytokine T helper 1 role.Conditions affected include asthma, 14,15 COPD 5 (MIM606963), acute lung injury, 16 and respiratory distresssyndrome (RDS, MIM 267450).Single nucleotide polymorphisms (SNPs) in  TNF  a have been found to modulate differential expression of the gene in vitro and in vivo. 17 It has been reported thatthe TNF a -308 SNP (in the promoter region of the gene)is associated with a fivefold increase in transcription. 18 The  1,031,  857, and  863 SNPs have also been asso-ciated with higher TNF a  levels. 19 Another polymorphismarising from the substitution of an adenine for a guaninemoiety occurs at position  238 within the TNF a  promot-er region. 20 The TNF a   238A allele has been associatedwith decreased TNF a , 21 reduced severity of nonpulmo-nary diseases, and improved outcomes among adults withrheumatic arthritis and rheumatoid heart diseases. 22,23 However, a linkage between the TNF a   238 SNP andprotection from BPD is still disputed. 24,25 If TNF a  is involved in the development of BPD,better understanding of this involvement could helphealthcare providers develop strategies to prevent andtreat inflammation that leads to the disease. Biochemi-cal studies of RDS have been conducted in Egyptianinfants, but no molecular analyses of BPD have beenexecuted. 26 Our hypothesis is that the frequency of theA allele or the AA genotype will be higher in Egyptianpremature infants with BPD than in healthy infants.To our knowledge, the present study is the first toexplore the   238G  >  A SNP within the promoter of the  TNF  a  gene in preterm Egyptian infants with lowbirth weight. The possible impact of antenatal, natal,and postnatal clinical risk factors is also explored. PATIENTS AND METHODSStudy Population Preterm infants (26–32 weeks of gestation) withvery low birth weight ( < 1,500 g) and BPD who wereadmitted to the neonatal intensive care unit (NICU) atEl-Gala` teaching hospital-Remises in Cairo betweenMarch 2008 and November 2010 were recruited intothe study. Antenatal history, birth weight, gestationalage (assessed by two physicians using the Ballardscale), steroid supplementation (partial or completecourse), and Apgar score after birth were recorded foreach infant. In most cases, the history of antenatal corti-coid therapy was not known. The treatment history of infants with BPD, including mechanical ventilation,oxygen therapy, and surfactant treatment, was alsorecorded. Children with major congenital or chromo-somal anomalies were excluded.A diagnosis of BPD was made if supplemental oxy-gen was administered for at least 28 days to maintainoxygen saturation   88%. Preterm infants were consid-ered to have  mild   BPD if they needed supplementaloxygen   28 days, but not at 36 weeks postmenstrualage (PMA). They were considered to have  moderate BPD if they needed supplemental oxygen  28 days andhad  < 30% FiO 2  at 36 weeks PMA. They were consid-ered to have  severe  BPD if they needed supplementaloxygen   28 days and had either  > 30% FiO 2  at36 weeks PMA and/or positive pressure at 36 weeksPMA. 3,27 Our study also contained a group of healthy controls.Participants in our control group had been enrolled atthe same time as the study infants and were retrospec-tively classified as not having BPD. The study was con-ducted Medical Genetics Center, Ain Shams University,Cairo. The study protocol was approved by the ethicscommittee of the same institute, and parental informedconsent was obtained. DNA Isolation Genomic DNA was extracted from peripheral blood(200  m l) using the QIAamp DNA blood kit (Qiagen,Hilden, Germany). In some cases, DNA was preparedin situ by gently scraping the buccal mucosa for 30 secusing a cytobrush. 28 The cells obtained were treateddirectly with diluted NaOH solution, heated, andneutralized with Tris–Cl, pH 8.0. A 2.5  m l volume of buccal cells typically sufficed for amplification by poly-merase chain reaction (PCR). Genotyping Analysis Genomic DNA was added to a 25  m l reaction volumeof 0.5  m M of each primer, 200  m M of each dNTP,67 mM Tris–HCl, 16 mM (NH 4 ) 2 SO 4 , 0.01% Tween-20, 700 Elhawary et al. Pediatric Pulmonology  1 mM MgCl 2 , and 0.15 units  Taq  DNA polymerase.The samples were then subjected to 35 cycles on PCREngine Dyad (Bio-Rad Laboratories Inc., Hercules,CA) with annealing at 59 8 C for 1 min. To identify theTNF a   238G  >  A polymorphism, DNA was amplifiedusing specific primers F: 5 0 -AAA CAG ACC ACAGAC CTG GTC-3 0 and R: 5 0 -CTC ACA CTC CCCATC CTC CCG GAT C-3 0 . The reverse primer con-tained two amplification creation restriction sites(underlined) to incorporate a  Bam HI restriction site forthe G allele. 29 The PCR product was incubated with the  Bam HI enzyme at 37 8 C for 2 hr, and the fragmentswere separated on a 3% MetaPhor agarose gel (BMA,Rockland, ME). The product was then identified byethidium bromide staining and was photographed usinga gel-documentation system (G-Box, SynGene, Freder-ick, MD) (Fig. 1). The genotypes of all samples werereassessed twice to ensure reproducibility. Positive con-trols for wild and variant genotypes were comparedwith the study samples. Data Analysis Statistical analysis was performed using SPSS 16.0(SPSS Inc, Chicago, IL). The data were presented asmeans    standard deviations. Student’s  t  -tests,  x 2 -tests,and  F  -test were used to compare continuous and cate-gorical variables. Multivariate logistic regression analy-sis was performed to assess the contributions of TNF a alleles and other  independent   risk factors to the studyoutcomes. A probability  < 0.05 was considered statisti-cally significant. Odds ratios with 95% confidence inter-vals (CIs) were also calculated.We used G  Power software (Germany, version 3.1.5, gpower3/) to perform  priori  power analysis to estimatesufficient sample sizes to achieve adequate power for  z -testing of two independent proportions.  Priori  samplesize estimations were performed using known informa-tion on the common allele frequencies in preterminfants, a criterion probability of   a  ¼  0.05, and a powersensitivity of 80%. The prevalence of BPD in thestudied population was assumed to be 50%, with acase–control ratio of 1. RESULTS For the study, 120 of 160 eligible infants with BPD(54 boys and 66 girls) and 100 healthy ‘‘non-BPD’’controls were enrolled. The non-respondents (n  ¼  40)either could not be traced or had parents who did notwant their infants involved in this work. The BPDinfants had a mean birth weight of 1,490    360 g and amean gestational age of 31.0    5.1 weeks. The non-BPD controls had a mean birth weight of 1,880    520 g and a mean gestational age of 34.3    1.5 weeks. No significant differences werefound between the BPD and non-BPD groups regardinggender ratio ( P  ¼  0.71), mode of delivery ( P  ¼  0.75),or use of prenatal corticosteroids ( P  ¼  0.19) (Table 1).Prenatal maternal risk factors, such as diabetes type2, hypertension, and prolonged rupture of membranes(pROM), were significantly more likely among BPDinfants than among non-BPD infants ( P  ¼  0.04). Themean Apgar score of the BPD group was significantlylower than that of the non-BPD group ( P  <  0.0001).Table 2 reveals the postnatal clinical characteristics of BPD infants. As shown, a proportion of infants in theBPD group had surfactant treatment (5%), persistentductus arteriosus (15%), necrotizing enterocolitis(12.5%), or sepsis (82.5%). Pulmonary infections, in-cluding pneumonia, recurrent respiratory infection, andhyperactivity airway inflammation, were present in 55%of the BPD infants (Table 2).Of the 120 infants with BPD, 20 (16.7%) had milddisease, 43 (35.8%) had moderate disease, and 57(47.5%) had severe disease. Mechanical ventilation wassignificantly associated with the phenotypes of BPDinfants ( P  ¼  0.007), although nasal continuous positiveairway pressure (nCPAP) was not ( P  ¼  0.56). Rates of postnatal complications in BPD infants were 12.5% for Fig. 1. Electrophoretic gel pattern of   238G > A polymorphismin the  TNF  a  gene. The G allele, having a  Bam  HI restriction site,appears in two fragments (123 and 42 bp). The A allele givesthe uncut PCR product (165 bp). The GA genotype isheterozygous. TNF-238 Polymorphism in Preterm BPD Infants 701 Pediatric Pulmonology  intraventricular hemorrhage (IVH), 12.5% for posthe-morrhage hydrocephalus, and 7.5% for retinopathy of prematurity (Table 2). The mortality rate among the120 BPD infants was 45%, with a higher frequency inmale infants (67%) than in female infants (33%). Of the 54 BPD infants who died, 48 (89%) had severedisease (Table 3).The allele frequencies of the TNF a   238G  >  Apolymorphism in our study population remained withinthe Hardy–Weinberg equilibrium. The   238G  >  Apolymorphism was associated with a twofold risk of BPD (OR  ¼  2.86; 95% CI, 1.35–3.83). None of the non-BPD infants carried the   238AA genotype, comparedwith 15% of infants with BPD. In comparison, 70% of the non-BPD infants and 68% of the BPD infants carriedthe   238GG genotype ( x 2 ¼  4.17,  P  <  0.0001). Theheterozygous status of the   238G  >  A polymporphismwas 17% (Table 4). The frequency of the A allele washigher in BPD infants (23%) than in non-BPD infants(15%), but the difference was not statistically signifi-cant ( P  ¼  0.132) (Table 4). The frequencies of the Gand A alleles in the BPD infants were significantly dif-ferent ( x 2 ¼  19.57,  P  <  0.0001).As for genotype–phenotype correlation in infantswith BPD, the   238GG genotype was lowest amonginfants with mild BPD (18%) and highest among thosewith severe disease (50%). A similar trend was seen forthe   238GA genotype in infants with BPD. However,the trend was different for the   238AA genotype,which was more common among those with moderateBPD (67%) than among those with severe BPD (33%).Of note, none of the differences at the   238 loci werestatistically significant when the subgroups of BPD se-verity were compared ( F   ¼  0.188,  P  ¼  0.83). DISCUSSION In our study population, we showed that gestationalage  < 26–32 weeks, birth weight  < 1,500 g, and lowApgar score were significantly associated with BPD.Mode of delivery and history of antenatal steroid treat-ment had no effect on the risk of BPD. We also foundthat the TNF a   238G  >  A polymorphism was associ-ated with the development of BPD in preterm infants.Although   238A is not the dominant allele in theEgyptian population, it could potentially be used as apredictive marker of BPD in Egyptian preterm infants.The incidence of BPD reported in the literature varieswidely. In Egypt, like in most populations, the inci-dence is still under investigation. The estimated numberof preterm infants with BPD in Egypt is 1,496 amongthe total population of 81,400,000. In the United States,Stroustrup and Trasande 30 concluded that the incidenceof BPD in American preterm infants decreased by 4.3%per year between 1993 and 2006. However, a trial bythe National Institute of Child Health and Human De-velopment (NICHD) reported no decrease between2003 and 2007. 31 Although the prevalence of BPD hasbeen shown to differ between centers in the UnitedStates, about one-quarter of infants of very low birth TABLE 1—Clinical Characteristics of Antenatal and Perinatal Data Antenatal data Non-BPD (n  ¼  100) BPD infants (n  ¼  120)  P -valueBirth weight (g) 1 1,880    520 1,490    360  < 0.0001  Gestational age (weeks) 1 34.3    1.5 31.0    2.1 0.007  Female:male ratio 1:1 1.2:1 0.71Caesarean section 2 57 (57.0) 72 (60.0) 0.75Prenatal steroid use 2 32 (32.0) 42 (35.0) 0.19Maternal risk factors 2 25 (25.0) 63 (52.5) 0.04  Apgar score  5 at 5th min 0 (0.0) 69 (57.5)  < 0.0001  1 Student’s  t  -test (mean    SD). 2 x 2 -test. Numbers in parenthesis represent percentages of infants.  P  <  0.05. TABLE 2—Postnatal Clinical Characteristics of EgyptianBPD Infants CharacterNumber of BPD infants(n  ¼  120)Frequency(%)Respiratory distress syndrome (RDS) 96 80.0Barotraumas 69 57.5Pulmonary infection 1 66 55.0Surfactant treatment 6 5.0Pulmonary hemorrhage 9 7.5Patent ductus arteriosus (PDA)(med/sur rx)18 15.0Necrotizing enterocolitis (NEC)(  stage II)15 12.5Sepsis (only Gram’s positive) 99 82.5Intraventricular hemorrhage (IVH)(grade  3)15 12.5Posthemorrhagic hydrocephalus 15 12.5Retinopathy of prematurity(ROP) (stage  3)9 7.5 1 Such as pneumonia, recurrent respiratory infection, and hyper-reactivityairway inflammation. 702 Elhawary et al. Pediatric Pulmonology  weight ( < 1,500 g) have been estimated to go on to de-velop BPD. 32 BPD has become more common amongEgyptian infants because of advances in care andchanges in the gestational age of premature infants. Thelack of uniformity in the diagnostic criteria of BPDmay be responsible for some of the reported variation.Northway et al. 1 definition of BPD has changed overthe past three decades. 33,34 The old definitions have thelimitation that they do not take into account the varyinguse of supplemental oxygen and positive pressureamong different centers. Given the importance of a con-sistent definition of BPD, a more extensive definitionwas proposed at a National Institutes of Health (NIH)BPD workshop in 2000. 3 The ‘‘new’’ definition usesoxygen dependency at 36 weeks PMA, total duration of oxygen supplementation, positive pressure require-ments, and gestational age of the infant to delineate thethree degrees of disease severity. 27,35 In our study, wecategorized BPD outcomes for each gestational ageaccording to the 2000 classification of BPD. 3,27,35,36 These classifications were mild BPD (oxygen use for atleast 28 days), moderate BPD (oxygen still needed at36 weeks), or severe BPD (ventilator support needed at36 weeks).Results for some clinical risk factors, such as birthweight, gestational age, Apgar score, mode of delivery,and the use of prenatal steroids, were consistent withprevious studies, but results for others were not. 31 Outcomes of most neonatal disorders, including BPD,vary across neonatal centers even after adjusting for de-mographic and antenatal characteristics. Recently, the useof progesterone supplementation (17- a -hydroxyprogesterone caproate) during pregnancy has been approved bythe US Food and Drug Administration (FDA) to reducethe risk of recurrent preterm birth in women with ahistory of spontaneous preterm delivery. 37 Unfortunately,the records on progesterone supplementation were miss-ing for most women in our study. Not having progester-one supplementation could greatly increase secondaryproinflammatory, which might explain part of the con-flict between our results and those of other studies.Because of a lack of financial support or healthcareresources, some Egyptian parents could not obtain sur-factant medication. Others had to delay the start of  TABLE 3—Major Complications and Outcomes With BPD Phenotypes Variable Subjects (%)BPD phenotypes P -valueMild (n  ¼  20) Moderate (n  ¼  43) Severe (n  ¼  57)Mechanical ventilation 1 120 (100) 18.8    9.8 24.5    9.1 36.3    14.7 0.007  nCPAP 1 120 (100) 7.7    0.6 10.0    7.2 13.1    10.5 0.560Discharge without complications 2 42 (35.0) 9 (21.4) 24 (57.1) 9 (21.4) 0.002  Discharge with complications 2,3 24 (20.0) 6 (25.0) 9 (37.5) 9 (37.5)Mortality 2 54 (45.0) 0 (0.0) 6 (11.1) 48 (88.9) 1 Student’s  t  -test. Values are mean    SD. 2 x 2 -test. Numbers in parenthesis represent percentages of infants. 3 Complications being intraventicular hemorrhage (IVH), posthemorrhage hydrocephalus (PHH), and retinopathy of prematurity (ROP).  P  <  0.05. TABLE 4—Genotype and Allele Distributions in BPD Infants Genotype  238G  >  A Non-BPD n (%) BPD n (%)BPD phenotype n (%)Mild (n  ¼  20) Moderate (n  ¼  43) Severe (n  ¼  57)GG 70 (70.0) 82 (68.3) 15 (18.3) 26 (31.7) 41 (50.0)GA 30 (30.0) 20 (16.7) 5 (25.0) 5 (25.0) 10 (50.0)AA 0 (0.0) 18 (15.0) 0 (0.0) 12 (66.7) 6 (33.3) x 2 ( P -value) 17.3 ( < 0.0001) 1 4.17 (0.041) 3 F   ¼  0.188 (0.83) 5 Allele frequencyG 170 (85.0) 184 (76.7) 35 (19.0) 57 (31.0) 92 (50.0)A 30 (15.0) 56 (23.3) 5 (8.9) 29 (51.8) 22 (39.3) x 2 ( P -value) 8.02 (0.005) 2 19.57 ( < 0.0001) 4 2.27 (0.132) 61 Significant differences in distribution of GG, GA, and AA genotypes among BPD infants. 2 Significant difference of the G allele in BPD infants compared with non-BPD infants. 3 Significant differences among genotypes in both BPD and non-BPD infants. 4 Significant difference between the G allele versus the A allele in BPD infants. 5 F  -test value compared the non-significance among the GG, GA, and AA genotypes with the severity forms. 6 Significant difference in the A alleles in BPD versus non-BPD infants. TNF-238 Polymorphism in Preterm BPD Infants 703 Pediatric Pulmonology
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks