SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Aim  The aim of our meta-analysis was to summarize quantitatively the association of genetic polymorphisms with cerebral palsy (CP).

Method  We identified 16 studies on the association of genetic polymorphisms with CP in Pubmed, Elsevier Science Direct, Chinese Biomedical Literature Database, Chinese National Knowledge Infrastructure, and Wanfang. Eleven of these studies (involving a total of 2533 cases and 4432 controls) were used in the current meta-analysis. A study was included if (1) it was published up to September 2010 and (2) it was a case–control study. We excluded one study of family members because the analysis was based on linkage considerations. Meta odds ratios and 95% confidence intervals based on fixed-effects models or random-effects models were dependent on Cochran’s Q statistic. We examined the relationship between alleles, as well as genotypes and susceptibility to CP.

Results  Meta-analysis was performed for 17 genetic polymorphisms: apolipoprotein E (ε2,ε3,ε4), methylenetetrahydrofolate reductase (MTHFR) (rs1801133), coagulation factor II (rs1799963]), coagulation factor V (rs6025), coagulation factor VII (rs5742910/rs6046), interleukin-6 (IL-6) (rs1800795), endothelial nitric oxide (rs1800779/rs1799983/rs3918226), fibrinogen β-polypeptide (rs1800790), plasminogen activator inhibitor 1 (rs1799768/rs7242), TNF-β lymphotoxin α precursor (rs1041981), adducin 1 (α) (rs4961), ADRB2 (rs1042714), and tumour necrosis factor α (rs1800629). We found a significant association between CP and IL-6 (rs1800795) [C vs G: odds ratio (OR) 1.79, 95% confidence interval (CI) 1.44–2.22, p<0.001; CC+GC vs GG: OR 1.72, 95% CI 1.29–2.29, p=0.002; CC vs GG+GC: OR 2.17, 95% CI 1.52–3.09, p<0.001], but no other genetic polymorphisms.

Interpretation  This meta-analysis demonstrated that CP is associated with the genetic polymorphism IL-6 (rs1800795).


Abbreviations
ADD1

Adducin 1 (α)

ADRB2

β2-adrenergic receptor

APOE

Apolipoprotein E

eNOS3

Endothelial nitric oxide

FGB

Fibrinogen β-polypeptide

IL-6

Interleukin-6

LTA

TNF-β lymphotoxin α precursor

MTHFR

Methylenetetrahydrofolate reductase

PAI1

Plasminogen activator inhibitor 1

TNF-α

Tumour necrosis factor-α

What this paper adds

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information
  • • 
    This is the first paper, to our knowledge, to use meta-analysis to investigate the association of genetic polymorphisms with CP.
  • • 
    The meta-analysis demonstrates that CP is associated with the genetic polymorphism interleukin-6 (rs1800795).

Cerebral palsy (CP) is defined as ‘a group of permanent disorders of the development of movement and posture, causing activity limitation that is attributed to non-progressive disturbances that occurred in the developing fetal or infant brain, and the motor disorders of CP are often accompanied by disturbances of sensation, perception, cognition, communication, and behaviour, by epilepsy, and by secondary musculoskeletal problems’.1 CP is diagnosed in approximately two in every 1000 infants, and this incidence has remained constant worldwide despite increasing obstetric intervention.2 The detailed mechanisms leading to CP are not known but, in the past few years, enormous effort has been invested in the attempt to identify risk factors for developing CP. Asphyxia, neuronal disorders, infection in utero, amnionitis, maternal autoimmune disease, metabolic disease, vascular lesions, as well as coagulation disorders all have been suggested as risk factors for CP.3–8

How the fetus responds in the presence of these risk factors is influenced by its genetic makeup, with some genotypes creating susceptibility to cerebral damage. Recently, a number of case–control studies have been conducted to investigate the association of genetic polymorphisms with CP.9–24 These studies have explored many genes, including those coding for apolipoprotein E (APOE), interleukin-4 (IL-4), interleukin-6 (IL-6), interleukin-8 (IL-8), mannose-binding lectin, toll-like receptor 4, coagulation factor II (F2), coagulation factor V (F5), coagulation factor VII (F7), tissue factor pathway inhibitor, TNF-β lymphotoxin α precursor (LTA), MTHFR, cystathionine β-synthase, annexin 5A, adducin 1 (α) (ADD1), nitric oxide synthase 2A isoform 1, endothelial nitric oxide synthase 3 (eNOS3), endothelial protein C receptor precursor, plasminogen activator inhibitor 1 (PAI1), plasminogen activator inhibitor 2 precursor, phosphodiesterase 4D, tissue plasminogen activator, β2-adrenergic receptor (ADRB2), fibrinogen β polypeptide (FGB), arachidonate 5-lipoxygenase-activating protein, and tumour necrosis factor-α (TNF-α). However, these studies have reported conflicting results. There are several possible factors that could explain the conflicting results, such as small sample sizes, participants’ ethnic background, uncorrected multiple hypothesis testing, and publication bias.

Meta-analysis is a method of increasing the effective sample size by the pooling of the data from individual associated studies, thus enhancing the statistical power of the analysis.25 The aim of the present study was to use a meta-analysis to investigate whether genetic polymorphism is associated with CP.

Method

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Identification of eligible studies

All studies that examined the association of genetic polymorphism with CP were carefully selected. Data were collected from the following electronic databases: Pubmed, Elsevier Science Direct, Chinese Biomedical Literature Database, Chinese National Knowledge Infrastructure, and the Chinese database, Wanfang. The following keywords were used for searching: ‘cerebral palsy’ and (‘gene’ OR ‘genotype’ OR ‘polymorphism’ OR ‘mutation’ OR ‘variation’). In the Chinese Biomedical Literature Database, the Chinese National Knowledge Infrastructure, and Wanfang, Chinese technical terms of the keywords were used for searching. Additional literature was collected from cross-references within both original and review articles. No restriction was placed on language. A study was included in the current meta-analysis if (1) it was published up to September 2010 and (2) it was a case–control study. We excluded one study involving family members because the analysis was based on linkage considerations. When a study reported the results on different subpopulations, we treated them as separate studies. When there were multiple publications from the same population, only the largest study was included.

In addition, an independent Pubmed search (by WD and XXY), an independent Elsevier Science Direct search (by ZGC and YL), and independent Chinese Biomedical Literature Database, Chinese National Knowledge Infrastructure, and Wanfang searches (by FXL and BXS) were performed using the same criteria. The contents of abstracts were reviewed independently by two investigators (WD and ZYF) to determine if they met the eligibility criteria. Where discrepancies occurred, a third investigator (TJL) performed an additional assessment. References in the studies were reviewed (by FXL) to identify additional studies.

Data extraction

If original genotype frequency data were unavailable in relevant articles, a request for additional data was sent to the corresponding author. Additionally, two investigators (WD and ZYF) independently extracted the data with the standard protocol and the result was reviewed by a third investigator (TJL). Discrepancies were resolved by discussion with our research team. From each study, we extracted the first author’s name, year of publication, racial ancestry of the study participants, source of publication, genes, polymorphisms, the number of cases and controls, and the available genotype and allele frequency information from genetic polymorphisms.

Meta-analysis methods

Allele frequencies at genetic polymorphisms from each study were determined by the allele counting method. A χ2 test was used to determine if observed frequencies of genotypes conformed to Hardy–Weinberg equilibrium expectations.

We examined the relationship between the alleles, as well as the genotypes and susceptibility to CP. The odds ratio (OR) and its 95% confidence interval (CI) were estimated for each study. Between-study heterogeneity was assessed by the χ2 test-based Q statistic.26 In the case of heterogeneity between studies, indicated by a significant Q statistic (p<0.10), the result of the random-effects model was selected. Otherwise, the result of the fixed-effects model was selected. We also measured the effect of heterogeneity by another measure:27

  • image

The pooled OR was obtained by the Mantel–Haenszel method in the fixed-effects model and by the DerSimonian and Laird method in the random-effects model.28,29 The pooled OR was performed by weighting individual ORs by the inverse of their variance, and the significance of the pooled OR was determined by the z test.

Evaluation of publication bias

Publication bias was investigated with the funnel plot, and funnel plot asymmetry was further assessed using Egger’s linear regression test.25,30

Statistical analyses were performed using Review Manager 4.2 software (Cochrane Collaboration, http://www.cc-ims.net/RevMan/relnotes.htm/) and Stata version 10 (StataCorp LP, College Station, TX, USA).

The Bonferroni method was used to adjust the significance alpha level to correct for the problem of multiple comparisons. Specifically, the usual significance level (α=0.05) was divided by 57 to account for 57 comparisons. Thus, a p value less than 0.009 was considered statistically significant in the study, and all the p values were two-sided.

Results

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Characteristics of studies

The characteristics of the studies that investigated the association of genetic polymorphisms with CP are presented in Table I.9–22 There were 14 443 papers relevant to the search words (Pubmed, 605; Elsevier Science Direct, 13 810; Chinese Biomedical Literature Database, 11; Chinese National Knowledge Infrastructure, 8; Wanfang, 9; Fig. S1 [published online]), of which 13 830 were potentially relevant studies after duplicates were removed. By screening the abstract, 13 814 of these articles were excluded (1763 were reviews; 1361 were not conducted in humans; 8011 did not explore genetic polymorphisms; 2679 were not case–control studies), leaving 16 studies9–24 for full publication review. Of these, two studies12,13 were excluded owing to unavailable data and three studies19,23,24 were excluded as they were duplicate reports. Thus, 11 studies9–11,14–18,20–22 were included in the current meta-analysis. One of these studies contained data on four different subpopulations, which we treated independently.10 Finally, a total of 15 separate comparisons were considered in the current meta-analysis. Meta-analysis was performed for 17 genetic polymorphisms: APOE (ε2,ε3,ε4),9,14,15,22 MTHFR (rs1801133),16,20,21 coagulation factor II (rs1799963),16,18,20,21 coagulation factor V (rs6025),16,18,20,21 coagulation factor VII (rs5742910/rs6046),16,21 IL-6 (rs1800795),10,11 eNOS3 (rs1800779/rs1799983/rs3918226), 16,21 FGB (rs1800790),16,21 PAI1 (rs1799768/rs7242),16,21 LTA (rs1041981),16,21 ADD1 (rs4961),16,21 ADRB2 (rs1042714),16,21 and TNF-α (rs1800629).17,21

Table I.   Characteristics of studies investigating the association of genetic polymorphism with cerebral palsya
IDStudyYearPopulationGenes (polymorphisms)Sample sizeFrequencies of allelesFrequencies of genotypes
CaseControl
  1. aNot including overlapping data. NA, not available; APOE, apolipoprotein E; IL-6, interleukin-6; MBL, mannose-binding lectin; TLR-4, toll-like receptor 4; F2, coagulation factor II; F5, coagulation factor V; F7, coagulation factor VII; TFPI, tissue factor pathway inhibitor; LTA, TNF-β lymphotoxin α precursor; MTHFR, methylenetetrahydrofolate reductase; CBS, cystathionine β-synthase; ANX5, annexin 5A; ADD1, adducin 1 (α); NOS2A, nitric oxide synthase 2a isoform 1; eNOS3, endothelial nitric oxide synthase 3; EPCR, endothelial protein C receptor precursor; THBD, thrombomodulin; PAI1, plasminogen activator inhibitor 1; PAI2, plasminogen activator inhibitor 2 precursor; PDE4D, phosphodiesterase 4D; PLAT, tissue plasminogen activator; ADRB2, β2-adrenergic receptor; FGB, fibrinogen β polypeptide; ALOX5AP, arachidonate 5-lipoxygenase activating protein; TNF-α, tumour necrosis factor α; AGT, angiotensin 1; AGTR1, angiotensin receptor 1; GNB3, guanine nucleotide-binding protein, beta-3; ICAM1, intercellular adhesion molecule 1; ITGA2, integrin, alpha 2; ITGB3, integrin, beta 3; MMP3, matrix metalloproteinase 3; NPPA, natriuretic peptide precursor A; SCNN1A, sodium channel, nonvoltage gated 1, alpha subunit; SELE, selectin E.

 1Braga et al.92010BrazilianAPOE (ε2,ε3,ε4)243243AvailableAvailable
 2Wu et al.102009WhiteIL-6 (rs1800795)131152AvailableAvailable
 3Wu et al.102009HispanicIL-6 (rs1800795)4451AvailableAvailable
 4Wu et al.102009BlackIL-6 (rs1800795)2322AvailableAvailable
 5Wu et al.102009AsianIL-6 (rs1800795)2446AvailableAvailable
 6Djukic et al.112009AustralianTLR-4 (Asp299Gly), IL-6 (rs1800795), IL-4 (−589C/T)413857AvailableAvailable
 7Clark et al.122009AmericanIL-6 (−7227)7067NANA
 8Gibson et al.132008AustralianMBL (−550/−221/+4/−52/−54/−57)443883NANA
 9McMichael et al.142008AustralianAPOE (ε2,ε3,ε4)343774AvailableAvailable
10Kuroda et al.152007AmericanAPOE (ε2,ε3,ε4)209209AvailableAvailable
11Gibson et al.162007AustralianF2 (rs1799963), F5 (rs6025), F7 (rs5742910/rs6046), TFPI (rs11896231), IL-8 (rs4073), LTA (rs1041981), MTHFR (rs1801133), CBS (rs12329790), ANX5 (rs8145113), ADD1 (rs4961), NOS2A (rs1137933), eNOS3 (rs1800779/rs1799983/rs3918226), EPCR (rs867186), THBD(rs1800576), PAI1 (rs1799768/rs7242), PAI2 (rs6098/rs6103/rs6104), PDE4D (rs12188950), PLAT (rs2020918), ADRB2 (rs1042714), FGB (rs4220/rs1800790), ALOX5AP (rs1769874/SG13S32/SG13S35)443883AvailableAvailable
12Gibson et al.172006AustralianTNF-α (−308), MBL (−221/−52/−54/−57)414856AvailableAvailable
13Yehezkely-Schildkraut et al.182005IsraeliF5 (G1691A)6162AvailableAvailable
F2 (G20210A)NASome available
MTHFR (C677T)NANA
14Gibson et al.192005AustralianMTHFR (A1298C)443883AvailableAvailable
15Fattal-Valevski et al.202005IsraeliF5 (G1691A), F2 (G20210A), MTHFR (C677T)49118NASome available
16Nelson et al.212005AmericanF2 (rs1799963), F5 (rs6025), F7 (rs5742910/rs6046), LTA (rs1041981), MTHFR (rs1801133), ADD1 (rs4961), eNOS3 (rs1800779/rs1799983/rs3918226), PAI1 (rs1799768/rs7242), TNF-α(−308/−238), ADRB2 (gln27glu/arg16gly/met235thr), FGB (rs1800790), AGT (met235thr), AGTR1 (1166A/C), GNB3 (825C/T), ICAM1 (gly214arg), ITGA2 (873G/A), ITGB3 (leu33pro), MMP3 (−1171 5A/6A), NPPA (664G/A,2238T/C), SCNN1A(trp493ag/ala663thr), SELE (ser128arg/leu554phe)96119AvailableAvailable
17Meirelles et al.222000BrazilianAPOE (ε2,ε3,ε4)4040AvailableAvailable

The distribution of the genotype in the control population was not in Hardy–Weinberg equilibrium for IL-6 (rs1800795),11 coagulation factor II (rs1799963),16 eNOS3 (rs3918226),16 PAI1 (rs1799768),16 coagulation factor V (rs6025),18 eNOS3 (rs1800779)21, and TNF-α (rs1800629)17 (all p<0.05). In addition, it was not known whether the distribution of the genotype in the control population was in Hardy–Weinberg equilibrium for coagulation factor II (rs1799963),18,20 coagulation factor V (rs6025),20 and MTHFR (rs1801133),20 owing to the lack of genotype data. The distribution of the genotype in control population was in Hardy–Weinberg equilibrium for the other genetic polymorphisms (p>0.05).

Meta-analysis

A summary of the meta-analyses of the association of genetic polymorphisms with CP is shown in Tables II and Table SI (published online).

Table II.   Meta-analysis of the association of genetic polymorphisms and cerebral palsy
GenePolymorphismsComparisonsSample sizeNumber of studiesTest of associationTest of heterogeneity
CaseControlOR (95% CI)zp valueModelχ2p valueI2 (%)
  1. #, allele 3 or 4; , allele 2 or 4; , allele 2 or 3; OR, odds ratio; R, random-effects model; F, fixed-effects model; NA, not applicable; APOE, apolipoprotein E; MTHFR, methylenetetrahydrofolate reductase; F2, coagulation factor II; F5, coagulation factor V; F7, coagulation factor VII; d, del; i, ins; IL-6, interleukin-6.

APOEε2ε2 vs ε#1670253241.84 (0.82–4.71)1.470.14R12.180.00775.4
ε2ε2+ε2ε# vs ε#ε#835126642.12 (0.80–5.62)1.510.13R12.590.00672.6
ε2ε2 vs ε#ε#+ε2ε#835126640.71 (0.24–2.11)0.610.54R3.070.0867.4
ε3ε3 vs ε1670253240.63 (0.38–1.06)1.750.08R18.650.00383.9
ε3ε3+ε3ε vs εε835126640.91 (0.61–1.35)0.480.63F2.520.470.0
ε3ε3 vs εεε3ε835126640.70 (0.40–1.21)1.280.20R15.830.00181.0
ε4ε4 vs ε1670253241.37 (0.82–2.26)1.220.22R13.050.00577.0
ε4ε4+ε4ε vs εε835126641.47 (0.83–2.61)1.330.18R13.340.00477.5
ε4ε4 vs εε▪+ε4ε835126640.85 (0.45–1.63)0.480.63F2.290.520.0
MTHFRrs1801133T vs C642130820.97 (0.80–1.18)0.300.77F0.630.430.0
TT+CT vs CC37077231.02 (0.79–1.33)0.160.87F1.120.570.0
TT vs CC+CT32165420.87 (0.57–1.30)0.690.49F0.010.920.0
F2rs1799963A vs G628126821.37 (0.63–3.01)0.790.43F1.690.1941.0
AA+GA vs GG42481441.17 (0.60–2.28)0.470.64F2.030.570.0
AA vs GG+GA3146342NANANANANANANA
F5rs6025A vs G756141431.04 (0.65–1.66)0.180.86F1.330.510.0
AA+GA vs GG42782641.04 (0.45–2.45)0.100.92R7.140.0758.0
AA vs GG+GA37870730.44 (0.07–2.81)0.870.38F0.190.660.0
F7rs5742910i vs d632130021.44 (0.68–3.04)0.940.35R5.440.0281.6
ii+di vs dd31665021.48 (0.67–3.27)0.960.34R4.770.0379.0
ii vs dd+di31665021.55 (0.57–4.20)0.860.39F1.140.2912.4
rs6046A vs G652132020.99 (0.72–1.34)0.080.93F2.250.1355.6
AA+GA vs GG32666021.28 (0.54–3.00)0.560.58R4.890.0379.5
AA vs GG+GA32666021.51 (0.06–35.66)0.260.80R3.040.0867.1
IL-6rs1800795C vs G612176851.79 (1.44–2.22)5.31<0.001F2.220.700.0
CC+GC vs GG30687951.72 (1.29–2.29)3.660.0002F2.230.690.0
CC vs GG+GC30687952.17 (1.52–3.09)4.29<0.001F1.530.820.0
Analysis for APOE (ε2,ε3,ε4)

The Q test of heterogeneity was significant and we conducted analyses using the random-effects models except in the contrasts of ε3ε3 versus εbsl00066εbsl00066+ε3εbsl00066 and ε4ε4 versus εε+ε4ε (# represents allele 3 or 4; bsl00066 represents allele 2 or 4; represents alleles 2 or 3). We did not detect an association of the APOE ε2, ε3, and ε4 genotypes with CP (ε2 vs ε#: OR 1.84, 95% CI 0.82–4.71, p=0.14; ε2ε2+ε2ε# vs ε#ε#: OR 2.12, 95% CI 0.80–5.62, p=0.13; ε2ε2 vs ε#ε#+ε2ε#: OR 0.71, 95% CI 0.24–2.11, p=0.54; ε3 vs εbsl00066: OR 0.63, 95% CI 0.38–1.06, p=0.08; ε3ε3+ε3εbsl00066 vs εbsl00066εbsl00066: OR 0.91, 95% CI 0.61–1.35, p=0.63; ε3ε3 vs εbsl00066εbsl00066+ε3εbsl00066: OR 0.70, 95% CI 0.40–1.21, p=0.20; ε4 vs ε: OR 1.37, 95% CI 0.82–2.26, p=0.22; ε4ε4+ε4ε vs εε: OR 1.47, 95% CI 0.83–2.61, p=0.18; ε4ε4 vs εε+ε4ε: OR 0.85, 95% CI 0.45–1.63, p=0.63).

Analysis for MTHFR (rs1801133)

The Q test of heterogeneity was not significant for MTHFR and we conducted analyses using the fixed-effects models. We did not detect an association of the MTHFR gene rs1801133 polymorphism with CP (Thymine [T] vs Cytosine [C]: OR 0.97, 95% CI 0.80–1.18, p=0.77; TT+CT vs CC: OR 1.02, 95% CI 0.79–1.33, p=0.87; TT vs CC+CT: OR 0.87, 95% CI 0.57–1.30, p=0.49).

Analysis for coagulation factor II (rs1799963)

The meta-analysis of coagulation factor II was not conducted because of the low frequency of AdenineAdenine (AA) when examining the contrast of AA versus GuanineGuanine (GG) +GuanineAdenin (GA). The Q test of heterogeneity was not significant, and we conducted analyses using the fixed-effects models. We did not detect an association between the coagulation factor II gene rs1799963 polymorphism and CP (A vs G: OR 1.37, 95% CI 0.63–3.01, p=0.43; AA+GA vs GG: OR 1.17, 95% CI 0.60–2.28, p=0.64).

Analysis for coagulation factor V (rs6025)

The Q test of heterogeneity was not significant for coagulation factor V, and we conducted analyses using the fixed-effects model, except in the contrast of AA+GA versus GG. We did not detect an association between the coagulation factor V gene rs6025 polymorphism and CP (A vs G: OR 1.04, 95% CI 0.65–1.66, p=0.86; AA+GA vs GG: OR 1.04, 95% CI 0.45–2.45, p=0.92; AA vs GG+GA: OR 0.44, 95% CI 0.07–2.81, p=0.38).

Analysis for coagulation factor VII (rs5742910/rs6046)

The Q test of heterogeneity was significant for coagulation factor VII, and we conducted analyses using the random-effect model, except in the contrasts of ins/ins versus del/del+del/ins and A versus G. We did not detect an association between the coagulation factor VII gene rs5742910, as well as rs6046 polymorphisms and CP (rs5742910: ins vs del: OR 1.44, 95% CI 0.68–3.04, p=0.35; ins/ins+del/ins vs del/del: OR 1.48, 95% CI 0.67–3.27, p=0.34; ins/ins vs del/del+del/ins: OR 1.55, 95% CI 0.57–4.20, p=0.39; rs6046: A vs G: OR 0.99, 95% CI 0.72–1.34, p=0.93; AA+GA vs GG: OR 1.28, 95% CI 0.54–3.00, p=0.58; AA vs GG+GA: OR 1.51, 95% CI 0.06–35.66, p=0.80).

Analysis for IL-6 (rs1800795)

The Q test of heterogeneity was not significant for IL-6, and we conducted analyses using the fixed-effects model. An association of IL-6 gene rs1800795 polymorphisms with CP was found (C vs G: OR 1.79, 95% CI 1.44–2.22, p<0.001; CC+GC vs GG: OR 1.72, 95% CI 1.29–2.29, p=0.002; CC vs GG+GC: OR 2.17, 95% CI 1.52–3.09, p<0.001), and the forest plots are shown in Fig. 1.

image

Figure 1.  Forest plots for meta-analysis of positive results. IL-6, interleukin-6.

Download figure to PowerPoint

eNOS3 (rs1800779/rs1799983/rs3918226)

The Q test of heterogeneity was not significant for eNOS3, and we conducted analyses using the fixed-effects model, except in the contrast of AG+GG versus AA. We did not detect an association of eNOS3 gene rs1800779, rs1799983, and rs3918226 polymorphisms with CP (rs1800779: G vs A: OR 0.86, 95% CI 0.71–1.05, p=0.14; AG+GG vs AA: OR 1.09, 95% CI 0.49–2.40, p=0.83; GG vs AA+AG: OR 0.66, 95% CI 0.44–0.99, p=0.04; rs1799983: T vs G: OR 1.02, 95% CI 0.83–1.26, p=0.86; TT+GT vs GG: OR 1.10, 95% CI 0.83–1.46, p=0.51; TT vs GG+GT: OR 0.85, 95% CI 0.54–1.35, p=0.49; rs3918226: T vs C: OR 1.02, 95% CI 0.70–1.48, p=0.91; TT+CT vs CC: OR 1.02, 95% CI 0.68–1.53, p=0.91; TT vs CC+CT: OR 1.03, 95% CI 0.33–3.19, p=0.96).

FGB (rs1800790)

The Q test of heterogeneity was not significant for FGB, and we conducted analyses using the fixed-effects model. We did not detect an association of the FGB gene rs1800790 polymorphism with CP (A vs G: OR 0.88, 95% CI 0.69–1.13, p=0.33; AA+GA vs GG: OR 0.89, 95% CI 0.66–1.19, p=0.41; AA vs GG+GA: OR 0.76, 95% CI 0.38–1.54, p=0.45).

PAI1 (rs1799768/rs7242)

The Q test of heterogeneity was not significant for PAI1, and we conducted analyses using the fixed-effects model. We did not detect an association of PAI1 gene rs1799768 and rs7242 polymorphisms with CP (rs1799768: G4 vs G5: OR 1.10, 95% CI 0.91–1.34, p=0.32; G4G4+G5G4 vs G5G5: OR 1.20, 95% CI 0.91–1.58, p=0.20; G4G4 vs G5G5+G5G4: OR 1.03, 95% CI 0.76–1.38, p=0.87; rs7242: T vs G: OR 1.01, 95% CI 0.83–1.22, p=0.92; TT+GT vs GG: OR 1.01, 95% CI 0.75–1.36, p=0.94; TT vs GG+GT: OR 1.01, 95% CI 0.73–1.40, p=0.94).

LTA (rs1041981)

The Q test of heterogeneity was not significant for LTA, and we conducted analyses using the fixed-effects model, except in the contrast of AA+CA versus CC. We did not detect an association of LTA gene rs1041981 polymorphisms with CP (A vs C: OR 1.07, 95% CI 0.88–1.31, p=0.48; AA+CA vs CC: OR 0.69, 95% CI 0.19–2.51, p=0.58; AA vs CC+CA: OR 1.16, 95% CI 0.89–1.52, p=0.28).

ADD1 (rs4961)

The Q test of heterogeneity was not significant for ADD1, and we conducted analyses using the fixed-effects model, except in the contrast of TT versus GG+GT. We did not detect an association of ADD1 gene rs4961 polymorphisms with CP (T vs G: OR 0.83, 95% CI 0.64–1.08, p=0.16; TT+GT vs GG: OR 0.82, 95% CI 0.60–1.11, p=0.20; TT vs GG+GT: OR 0.59, 95% CI 0.12–2.86, p=0.51).

ADRB2 (rs1042714)

The Q test of heterogeneity was not significant for ADRB2, and we conducted analyses using the fixed-effects model. We did not detect an association of ADRB2 gene rs1042714 polymorphisms with CP (G vs C: OR 1.09, 95% CI 0.90–1.32, p=0.39; GG+CG vs CC: OR 1.08, 95% CI 0.81–1.43, p=0.61; GG vs CC+CG: OR 1.18, 95% CI 0.83–1.69, p=0.35).

TNF-α (rs1800629)

The Q test of heterogeneity was not significant for TNF-α, and we conducted analyses using the fixed-effects models. We did not detect an association of TNF-α gene rs1800629 polymorphisms with CP (A vs G: OR 1.06, 95% CI 0.87–1.30, p=0.56; AA+GA vs GG: OR 1.16, 95% CI 0.92–1.47, p=0.20; AA vs GG+GA: OR 0.60, 95% CI 0.31–1.16, p=0.13).

Evaluation of publication bias

Funnel plot asymmetry was assessed using the Egger’s linear regression test. If there is asymmetry, the regression line will not run through the origin. The intercept a provides a measure of asymmetry – the larger its deviation from zero, the more pronounced the asymmetry. The results of Egger’s linear regression test are shown in Table III. It was shown that there was no publication bias in some comparisons (APOE [ε2,ε3,ε4]: ε2 vs ε#, ε2ε2+ε2ε# vs ε#ε#, ε3 versus εbsl00066, ε3ε3+ε3εbsl00066 vs εbsl00066εbsl00066, ε3ε3 vs εbsl00066εbsl00066ε3εbsl00066, ε4 vs ε, ε4ε4+ε4ε vs εε, ε4ε4 vs εε▪+ε4ε; MTHFR(rs1801133): TT+CT vs CC; coagulation factor II (rs1799963): AA+GA vs GG; coagulation factor V (rs6025): A vs G, AA+GA vs GG; IL-6 (rs1800795): C vs G, CC+GC vs GG, CC vs GG+GC). The funnel plots are shown in Figure 2. However, Egger’s test was not applied in some comparisons because of the small number of studies or rare occurrence of homozygous mutations.

Table III.   Egger’s linear regression test to measure funnel plot asymmetrya
PolymorphismY axis intercept a (95% CI)
  1. aAll p>0.05. NA, not applicable; APOE, apolipoprotein E; MTHFR, methylenetetrahydrofolate reductase; F2, coagulation factor II; F5, coagulation factor V; IL-6, interleukin-6; #, allele 3 or 4; bsl00066, allele 2 or 4; , allele 2 or 3.

APOE (ε2)ε2 vs ε#ε2ε2+ε2ε# vs ε#ε#ε2ε2 vs ε#ε#+ε2ε#
2.10 (−4.48 to 8.68)2.17 (−5.07 to 9.41)NA
ε3 vs εε3ε3+ε3ε vs εεε3ε3 vs εε+ε3ε
APOE (ε3)−4.09 (−10.47 to 2.28)−0.96 (−4.28 to 2.36)−4.31 (−10.50 to 1.87)
ε4 vs εε4ε4+ε4ε vs εεε4ε4 vs εε+ε4ε
APOE (ε4)3.69 (−0.61 to 7.99)3.82 (−0.433 to 7.21)2.94 (−17.56 to 23.45)
T vs CTT+CT vs CCTT vs CC+CT
MTHFR (rs1801133)NA−0.43 (−20.33 to 19.47)NA
A vs GAA+GA vs GGAA vs GG+GA
F2 (rs1799963)NA1.14 (−4.0 to 6.31)NA
A vs GAA+GA vs GGAA vs GG+GA
F5 (rs6025)−0.60 (−29.63 to 28.42)0.77 (−17.12 to 18.67)NA
C vs GCC+GC vs GGCC vs GG+GC
IL-6 (rs1800795)−0.15 (−2.86 to 2.56)−0.81 (−3.46 to 1.83)0.61 (−0.67 to 1.91)
image

Figure 2.  Funnel plots with pseudo 95% confidence limits for meta-analysis. SE, standard error; APOE, apolipoprotein E; MTHFR, methylenetetrahydrofolate reductase; F2, coagulation factor II; F5, coagulation factor V; IL-6, interleukin-6.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

The association of genetic polymorphisms with CP has been reported by a number of investigators; however, conclusions remain controversial. In the current meta-analysis, we retrieved 11 studies (2533 cases and 4432 controls) to evaluate the association of genetic polymorphism with CP. Meta-analysis was performed for 17 genetic polymorphisms: APOE (ε2,ε3,ε4), MTHFR (rs1801133), coagulation factor II (rs1799963), coagulation factor V (rs6025), coagulation factor VII (rs5742910/rs6046), IL-6 (rs1800795), eNOS3 (rs1800779/rs1799983/rs3918226), FGB (rs1800790), PAI1 (rs1799768/rs7242), LTA (rs1041981), ADD1 (rs4961), ADRB2 (rs1042714), and TNF-α (rs1800629). A significant association was seen for IL-6 (rs1800795), but not for the other genetic polymorphisms. As far as we know, this is the first meta-analysis carried out with the aim of investigating the relationship between genetic polymorphisms and CP.

Enormous effort has been invested in the attempt to identify the causes of CP, but the underlying mechanisms of CP remain obscure and efforts to reduce the incidence of perinatal brain injury in term-born infants have failed. Evidence indicates that inflammatory processes play an important and under-recognized role in the pathogenesis of CP.10 In this study, we detected a significant association of IL-6 gene rs1800795 polymorphisms with CP, and found that the C allele was a risk allele for susceptibility to CP. The IL-6 gene is located on chromosome 7, and the rs1800795 polymorphism, located at position −174 in the promoter region, has been implicated in altered IL-6 production.31–34 Meanwhile, the C allele has been associated with increased risk for newborn brain injury in preterm infants.35 Therefore, the IL-6 gene may play an important role in the pathogenesis of CP. However, the detailed mechanisms need further study. Of course, the association may also result from linkage disequilibrium with another functional polymorphism in the structural part of the gene or in the regulatory region. Additionally, there is an unexplained excess of CP among male infants.36 The sex-associated difference in susceptibility indicates that the cause of CP may be different in males and females. Thus, subgroup analysis by sex may be helpful to identify the cause of CP. However, this approach requires the authors of all of the published studies to share their data.

Some limitations of this study should be discussed. First, significant between-study heterogeneity was detected in some comparisons, especially for the APOE gene, which may distort the meta-analysis. However, this was not a major problem because CP itself is heterogeneous, and different patient populations may contribute to the heterogeneity. Second, we could not construct a funnel plot and conduct an Egger’s test for each meta-analysis because of the small number of studies in some cases or low prevalence of the homozygous mutation. Thus, publication bias may be present. Third, although we did not detect an association of other genetic polymorphisms (APOE [ε2,ε3,ε4], MTHFR [rs1801133], coagulation factor II [rs1799963], coagulation factor V [rs6025], coagulation factor VII [rs5742910/rs6046], eNOS3 [rs1800779/rs1799983/rs3918226], FGB [rs1800790], PAI1 [rs1799768/rs7242], LTA [rs1041981], ADD1 [rs4961], ADRB2 [rs1042714], TNF-α [rs1800629]) with CP, the result should be interpreted with caution because the number of studies and the number of participants were small. Fourth, some genetic polymorphisms were not in Hardy–Weinberg equilibrium in the current meta-analysis, which may affect the validity of the conclusion. Finally, meta-analysis is still retrospective research that is subject to the methodological deficiencies of the included studies.37 Thus, we minimized the likelihood of bias by developing a detailed protocol before initiating the study.

Conclusion

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Taking into account these findings, our study demonstrates that CP is associated with the genetic polymorphism of IL-6 (rs1800795). To reach a definitive conclusion, further gene–gene and gene–environment interaction studies based on larger sample sizes and a case–control design stratified by ethnicity are still needed.

Acknowledgements

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

We thank all the people who helped in this study. We are grateful for the kind help given by Dr Gai McMichael, Dr Alastair H MacLennan, Dr Jenny Ottley, Dr Catherine Gibson, and Dr Sue Reid. This work was supported by grants from the National Natural Science Foundation of China (30672270).

References

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. What this paper adds
  4. Method
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information
FilenameFormatSizeDescription
DMCN_3884_sm_figs1.pdf86KSupporting info item
DMCN_3884_sm_tables1.pdf141KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.