Tumor necrosis factor-alpha G-308A gene polymorphism and coronary heart disease susceptibility: An updated meta-analysis

Thrombosis Research 127 (2011) 400–405

Contents lists available at ScienceDirect

Thrombosis Research j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / t h r o m r e s

Regular Article

Tumor necrosis factor-alpha G-308A gene polymorphism and coronary heart disease susceptibility: An updated meta-analysis Hai-Feng Zhang a, Shuang-Lun Xie a, Jing-Feng Wang a,⁎, Yang-Xin Chen a, Yan Wang b, Tu-Cheng Huang a a b

Department of Cardiology, Sun Yat-sen Memorial Hospital, Sun Yat-sen University, Guangzhou, China Department of Cardiology, Tongji Hospital, Huazhong University of Science and Technology, Wuhan, China

a r t i c l e

i n f o

Article history: Received 8 July 2010 Received in revised form 28 November 2010 Accepted 29 December 2010 Keywords: Tumor necrosis factor-alpha Coronary heart disease Polymorphism Susceptibility Meta-analysis

a b s t r a c t Purpose: Several studies have reported apparently conflicting findings for the effects of tumor necrosis factor-alpha (TNF-α) G-308A polymorphism on coronary heart disease (CHD) susceptibility. We undertook a systematic review and meta-analysis to investigate the association between this gene variant and CHD predisposition. Methods: We systematically searched electronic databases (Medline, EMbase, Chinese BioMedical, BIOSIS, Global Health, PsycINFO, Allied and Complementary Medicine Database, Cochrane Library, HuGE Navigator, and British Nursing) for relevant studies published between 1947 and October, 2010. Summarized estimation of odds ratio (OR) and 95% confidence interval (CI) were calculated. Publication bias and heterogeneity among studies were explored. Results: We identified 24 studies providing data for 9 921 cases and 7 944 controls. Pooled analysis based on ORs adjusted by CHD risk factors showed that carrying the TNF-α gene A variant conferred a 1.5-fold increased risk of developing CHD (AG + AA vs. GG, OR= 1.50, 95% CI: 1.23-1.77) in Caucasian population. No significant association between the gene polymorphism and CHD risk could be found in other ethnic groups. Conclusions: It is probable that carrying the A variant is associated with CHD risk in Caucasians but not in Asians, Indians, or Africans. Further studies are merited to assess the association in greater details, especially in Asians, Indians and Africans. © 2011 Elsevier Ltd. All rights reserved.

1. Introduction It is widely accepted that coronary heart disease (CHD) is a chronic inflammatory disease [1]. Proinflammatory and inflammatory factors contribute from the initiative process of an atherosclerotic lesion to its aggravation and finally to the plaque rupture which is thought to be a major step in triggering the occurrence of clinical coronary events [2]. Inflammatory process impairs endothelium function resulting in an alteration of the anti-atherosclerotic property and fatty streak formation [3]. Moreover, in the later phrase, it degrades the fibrous cap tissue, thus predisposes plaque rupture and leads to acute coronary events [3,4]. Atherosclerosis is a partly heritable disorder [5]. Certain genes related to the regulations of proinflammatory cytokine productions have been postulated to influence CHD susceptibility. For example, the tumor necrosis factor-alpha (TNF-α) gene, which is located within the class III region of the major Abbreviations: CHD, coronary heart disease; TNF-α, tumor necrosis factor-alpha; MHC, major histocompatibility complex; MI, myocardial infarction; HWE, HardyWeinberg equilibrium; PCR, polymerase chain reaction; RFLP, restriction fragment length polymorphism; CS, coronary stenosis; MAF, minor allele frequency; UA, unstable angina; SA, stable angina; SCD, sudden cardiac death. ⁎ Corresponding author. Department of Cardiology, Sun Yat-sen Memorial Hospital, Sun Yat-sen University, Guangzhou, Guangdong 510120, China. Tel.: +86 20 813 323 99. E-mail address: [email protected] (J.-F. Wang). 0049-3848/$ – see front matter © 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.thromres.2010.12.018

histocompatibility complex (MHC) on chromosome 6p21.3 [6,7], was reported to be associated with myocardial infarction (MI) risk in patients with a parental history [8]. It encodes TNF-α, an important proinflammatory cytokine highly expressed in the atherosclerosis plaques, which is produced and secreted by inflammatory cells (e.g. macrophages and monocytes) [3,9]. Wilson and his colleagues first reported the bi-allelic polymorphism within the 5' genomic region of the TNF-α gene promoter at position -308 (rs1800629) in 1993. This polymorphism involves a substitution of guanine (G, wild-type, the common type) by adenosine (A, the variant-type) [10]. Carrying the A allele enhances transcriptional activity and is reported to be associated with higher levels of circulating TNF-α [11,12]. Serum levels of TNF-α is elevated in CHD patients and may modify the risk for developing coronary events since it affects endothelial cell hemostatic function [13]. Therefore, it has been hypothesized that the TNF-α G-308A gene polymorphism could change CHD susceptibility. This postulate has been investigated in a number studies since it was firstly suggested in 1998 that the carriers of -308A variant conferred a 1.43-fold risk of MI [8]. However, many later studies failed to demonstrate this association and current results were obviously conflicting. Pereira et al. once performed a meta-analysis and reported a nonsignificant result [14]. Many new studies have now been presented and the previous metaanalysis paid little attention to the gene effects adjusted by CHD risk

H.-F. Zhang et al. / Thrombosis Research 127 (2011) 400–405

factors [14]. In addition, few Asian studies were included in the previous analysis [14]. Therefore, we re-evaluate the evidence for the association between the TNF-α G-308A gene polymorphism and CHD using methods developed by the Human Genome Epidemiology Network [15], paying more attention to the pooled analysis based on adjusted gene effects, supplementing information on Asian studies. 2. Method 2.1. Selection criteria All population based studies reporting the association of TNF-α G308A polymorphism and CHD risk were considered for inclusion in this meta-analysis. The inclusion criteria were as follows: 1) casecontrol studies or cohort studies; 2) proper method for identifying CHD patients and controls (e.g., angiographically and cardiac enzymes confirmed; a document history of percutaneous coronary intervention, percutaneous transluminal coronary angioplasty, or coronary artery bypass graft). If essential information was not presented, authors were contacted for details. The study would be excluded if enough information could not be obtained. 2.2. Search strategy We followed the HuGENet™ HuGE review handbook and metaanalysis of observational studies in epidemiology protocols (including the design, accomplishment, analysis and report) [15,16]. A comprehensive computer based search strategy was used to identify eligible studies published before October 2010. The following electronic databases were used: Medline, EMBase, BIOSIS, Global Health, PsycINFO, Allied and Complementary Medicine Database, Cochrane Library, British Nursing, HuGE Navigator (http://hugenavigator.net) [17] and Chinese BioMedical Web (http://sinomed.imicams.ac.cn). Key words were polymorphism, coronary heart disease, and tumor necrosis factor-α. Hand search was also performed. Details of search strategy were described in the Supplemental Materials. 2.3. Data extraction Information was carefully extracted from all included studies independently by two investigators (Zhang HF & Xie SL). Disagreement was dealt with by a discussion between the two. If a consensus could not be established, a third investigator (Wang JF) was consulted to resolve the dispute and a final decision was made by the majority of the votes. The following data were collected: authors, publication date, study designs, ethnicities, CHD phenotypes, characteristics of the controls, genotyping method, total number of cases and controls, and genotypes distributions. Ethnicities were categorized as Caucasian, Asian, African and others [18].

401

under the dominant model (AG + AA vs. GG). Results for recessive model (GG + AG vs. AA), homogeneous comparison (AA vs. GG), and additive model (A vs. G) were also calculated. To derive adjusted gene effetcs, further pooled analysis was carried out based on ORs adjusted by CHD risk factors (age, gender, body mass index, smoking habit, diabetes, hypertension, high-density lipoprotein cholesterol level and family history of coronary heart disease). The Mantel-Haenszel fixedeffects model was used if no heterogeneity existed; otherwise, the DerSimonian-Laird random-effects model was used. Publication bias was assessed using Egger's test and Begg's funnel plot was produced [21]. All the statistical analyses were performed with STATA version 9.2 (Stata Corporation, College Station, TX, USA). 3. Results 3.1. Study inclusion and characteristics A total of 24 studies were identified [8,22–28], [29–38] [39–44] including 19 English studies (16 Caucasian studies, 1 Asian study, 1 Indian study, and 1 African study) [8,22,23,25,29–32,39,40] [24,32,33,36,38–41,43] and 5 Chinese ones (Asian studies) [27,28,37,42,44]. The flow chat of study inclusion was shown in Fig. 1. A list of the involved phenotypes, genotypes distributions and quality scores of included studies is provided in the Table 1 and more details were shown in the Supplemental Table. Studies were performed in a wide range of geographical settings, with 75.19% (13 433 of 17 865) cases being Caucasians, 17.72% (3 166 of 17 865) being Asians, 2.47% (442 of 17 865) being Indians, and 4.61% (824 of 17 865) being Africans. All studies were case-control in design. Controls in 9 studies were drawn at random from approximately general population [8,25,30,31,33, 38,39,41,43]. Eleven studies included controls from healthy volunteers, healthy blood donors, health check visits or outpatient departments [23,24,27–29,32,35–37,42,44], and 4 recruited controls from other

417 Itemes identified by searches 382 Excluded after abstract review 191 TNF-α G-308A not involved 161 Duplications 30 Reviews or editorials 35 Articles fully reviwed 9 Excluded 5 No control 2 No CHD patient included 2 TNF-α G-308A not involved

2.4. Quality score assessment The quality of included studies were assessed independently by the same two investigators using modified quality assessment scores described previously [19]. Disagreement was settled as above. Scores were ranged from 0 (worst) to 13 (best). 2.5. Statistical analysis The genotypes distributions among controls were tested for Hardy-Weinberg equilibrium (HWE) using Fisher's exact test. The frequency of putative risk allele (-308A) in the controls was estimated by the inverse-variance method described elsewhere [19,20]. The Cochrane Q-test was used for heterogeneity test and P b 0.1 was considered an existence of heterogeneity among results. Pooled odds ratio (OR) with 95% confidence interval (CI) were estimated mainly

26 Articles considered for inclusion 2 Excluded No data obtained 24 Studies 16 Caucasian studies (7 716 cases/5 717 controls) 6 Asian study (1 577 cases/1 589 controls) 1 Indian study (210 cases/232 controls) 1 African study (418 cases/406 controls) Fig. 1. Flowchart of selection of studies for the meta-analysis. Abbreviations: CHD, coronary heart disease; TNF-α, tumor necrosis factor-alpha.

402

H.-F. Zhang et al. / Thrombosis Research 127 (2011) 400–405

Table 1 List of the included studies. Cases Study

Phenotype

Controls Genotypes

N

GG

GA

AA

445 196 148 180 793 998 318 341 95 204 849 306 149 547 293 237 105 1167 248 97 7716

325 117 120 127 565 701 229 231 73 146 613 242 98 365 224 206 82 799 175 59 5497

113 63 26 46 215 268 80 92 20 48 212 61 51 163 69§ 29 20 340 63 36 2015

7 16 2 7 13 29 9 18 2 10 24 3 0 19 § 2 3 28 10 2 204

300 504 286 100 73 74 38 40 162 1577

268 439 234 79 54 66 36 29 148 1353

32 62 22 14 17 8 2 9 14 180

UA + MI

210

181

CS + MI

418

265

Caucasian Herrmann et al.† Herrmann et al.‡ Padovani et al. Allen et al. Koch et al. Koch et al. Szalai et al. Verdrell et al. Bernard et al. Bernard et al. Georges et al. Tulyakova et al. Tulyakova et al. Tobin et al. Antonicelli et al. Dedoussis et al. Giacconi et al.* Bennet et al. Sbarsi et al. Elahi et al. Subtotal

MI MI MI CS MI CS CS CS SA UA + MI CS MI CS (SCD) MI MI MI SA + MI MI CS CS

Asian Hou et al. Hou et al. Liu et al. Liu et al. Shun et al. Li et al. Li et al. Chen et al. Xiang et al. Subtotal

CS MI CS MI CS CS MI CS CS + UA + MI

Indian Banerjee et al. African Ghazouani et al.

N

Genotypes

HWE P value

MAF

Score

GG

GA

AA

534 176 148 329 340 268 207 80 314 246 505 310 237 190 1497 241 95 5717

376 97 114 222 244 181 159 61 222 177 337 246 227 160 1037 185 41 4086

148 73 33 98 83 65 46 18 80 64 146 64§ 10 28 412 53 40 1461

10 6 1 9 13 22 2 1 12 5 22 § 0 2 48 3 14 170

0.29 0.08 0.40 0.64 0.09 b 0.01 0.51 0.80 0.17 0.17 0.23 § 0.74 0.50 0.37 0.71 0.41

0.16 0.24 0.12 0.18 0.16 0.20 0.12 0.13 0.17 0.15 0.19 N 0.05§ 0.02 0.08 0.17 0.12 0.36

11 11 11 11 12 12 11 10 9 9 10 10 10 12 12 11 9 12 7 8

0 3 30 7 2 0 0 2 0 44

905 176 138 158 30 182 1589

802 142 118 138 21 163 1384

101 11 20 20 8 19 179

2 23 0 0 1 0 26

0.53 b 0.01 0.36 0.40 0.83 0.46

0.06 0.16 0.07 0.06 0.17 0.05

12 12 11 11 10 9 9 7 9

28

1

232

201

31

0

0.28

0.07

7

142

11

406

267

124

15

0.90

0.19

8

Abbreviations: N, sample size; HWE, Hardy-Weinberg equilibrium; MAF, minor allele frequency; MI, myocardial infarction; CS, coronary stenosis; SA, stable angina; UA, unstable angina; SCD, sudden cardiac death. † France cohort of the study; ‡ Irlandais cohort of the study; - using the same controls as another cohort in a study; * this study was not considered as myocardial infarction subgroup since stable angina is more prevalent; § AA genotype was reported together with GA genotypes and thus proportion of MAF could not be calculated, and HWE test result was obatined directly from the text; ¶ articles were published in Chinese language.

groups of patients or identified after negative cardiac assessment [22,26,34,40]. Twelve studies matched controls to cases on age and gender [8,22–26,33–35,39,40,43]. All but 2 studies used polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) with standard restriction enzymes (NcoI). Of the remaining studies, used oligonucleotide probes [43] and mass spectrometry analysis platform [42], respectively.

analysis including the study deviated from HWE showed a similar result (data not shown) [33]. The pooled -308A allele frequencies were 8.20% (random-effects model: 95% CI: 5.72%-10.69%; χ215 = 25.59, P b 0.01), 18.96% (95% CI: 16.27%-21.66%), and 6.68% (95% CI: 4.41%-8.95%), respectively for Asians, Africans and Indians. 4. Association between TNF-α G-308A gene polymorphism and CHD risk

3.2. Pooled prevalence of TNF-α -308A in the controls 4.1. Meta-analysis of caucasian studies Genotypes distributions in the controls deviated from HWE in 2 studies (1 Caucasian study and 1 Asian study, the Table 1) [33,42]. One Caucasian study investigated 2 regimens and each was considered separately in the pooled analysis [8]. The -308A allele frequency could not be obtained in 1 Caucasian study [41]. Thus, data in a total of 16 cohorts comprising of 7 716 cases and 5 717 controls were included in the pooled analysis of the TNF-α -308A variant frequency in Caucasians. For Asian population, a total of 6 studies comprising of 1 577 case and 1 589 controls were involved. There was a great heterogeneity among Caucasian studies (χ215 = 319.26, P b 0.01). The pooled -308A frequency using the randomeffects model was 15.70% (95% CI: 12.70%-18.70%); a sensitivity

The meta-analysis results of Caucasian studies were summarized in Fig. 2. The overall-analysis under the dominant model combining 16 studies (20 cohorts of participants; 7 716 cases and 5 717 controls, the Table 1) yielded a significant heterogeneity (χ219 = 37.55, P b 0.01). Pooled OR by the random-effects model indicated no association between the gene polymorphism and CHD predisposition (OR = 1.02, 95% CI: 0.91-1.15). Subgroup analysis by CHD phenotype (MI or coronary stenosis (CS)) also showed heterogeneous outcomes (MI: χ29 = 18.39, P = 0.03; CS: χ28 = 18.85, P = 0.03) and pooled results in neither subgroup exhibited a significant association (MI: OR = 1.00, 95% CI: 0.86-1.17; CS: OR = 1.05, 95% CI: 0.87-1.26, Fig. 2). All heterogeneities

H.-F. Zhang et al. / Thrombosis Research 127 (2011) 400–405 Groups of studies

Number of studies

P (Q-test)

Cases/controls

403

Odds ratio (95%CI)

Caucasian 4336/4073

1.00 (0.86-1.17)

0.03

3380/2310

1.05 (0.87-1.26)

18

<0.01

7716/5717

1.02 (0.91-1.15)

Myocardial infarction

3

0.37

642/1239

1.08 (0.81-1.44)

Coronary stenosis

6

0.43

935/1589

1.00 (0.78-1.28)

All cases/controls

6

0.53

1577/1589

1.03 (0.86-1.24)

Myocardial infarction

9¶

Coronary stenosis

10

All cases/controls

0.02

Asian

.8

1

1.5

Fig. 2. Pooled OR (AA + AG vs. AA) by DerSimonian-Laird random-effects model, group by ethnicities (Caucasian or Asian) and phenotypes (coronary stenosis or myocardial infarction). The horizontal axis is plotted as OR. ¶, ten cohorts were involved since Herrmann et al. contains two cohorts of participants and each was considered as separately for pooled analysis.

could be eliminated by excluding studies by Dedoussis et al. [40]. and Elahi et al [26]. However, such exclusions did not change the results significantly (data not shown). Other assumed modes of inheritance showed similar results (Supplemental Figs. 1–3). Egger's test and Begg's plot did not show any evidence of publication bias (P = 0.55, Fig. 3). 4.2. Meta-analysis of asian studies The main results of the pooled analysis on Asian studies were shown in Fig. 2. Overall, no significant heterogeneity was found and fixed-effects model showed no association between the polymorphism and CHD susceptibility either in the overall analysis or subgroup analyses according to CHD phenotypes (MI or CS) under any assumed mode of inheritance (Fig. 2, Supplemental Figs. 4–6). No publication bias could be observed (P = 0.39). 4.3. Meta-analysis of adjusted ORs Considering that both genetics and environmental factors could contribute to CHD susceptibility and taking these risk factors into account in a meta-analysis may produce more convincible results, we performed a meta-analysis based on adjusted ORs. Adjusted ORs and

logOR

1

0

−1 0

.2

.4

SE of logOR Fig. 3. Begg's funnel plot with pseudo 95% confidence limits under dominant model (AG + AA vs. GG) for Caucasian studies. The horizontal axis is plotted as standard error of logOR and the vertical axis is plotted as logOR. The sizes of the circles are corresponds to sample sizes.

corresponding 95% CIs were available in 5 Caucasian studies comprising of 2 125 cases and 1 761 controls [8,30,39–41]. A combined analysis of these studies showed a significant association between TNF-α -308A allele carriers and CHD risk, showing that carrying the -308A allele conferred about 1.5-fold increased risk for CHD in Caucasians (OR = 1.50, 95% CI: 1.23-1.77, Fig. 4), with no heterogeneity (χ24 = 2.49, P = 0.65) among studies. 5. Discussion The association between the TNF-α G-308A polymorphism and CHD risk has been highly controversial. Much evidence has been added since the last meta-analysis was published [14]. What's different from the previous work [14], we enlarged the sample size to 17 865 participants (9 921 cases and 7 944 controls); moreover, we performed a pooled analysis based on the ORs adjusted by potential CHD risk factor. Results indicated that the -308A allele was a risk factor for CHD in Caucasians. CHD is a typical complex disease with multifactorial etiology. Both genetics and environmental factors could contribute to its incidence and development. The present meta-analysis revealed that significant association was observed only when taking potential CHD risk factors into consideration, though with smaller sample size. The different CHD risk factors status between the cases and controls may be an explanation for the discrepancy. In many of the included studies, notwithstanding selecting the controls randomly from community and matching the cases by sex and age, the prevalences of various CHD risk factors (e.g., smoking habit, concurrence of hypertension and diabetes mellitus) in the cases and controls were distinct. Therefore, the genetics effects may be masked by the strong concurrent relevant factors. This was supported in a recent study by Elahi et al., in which authors concluded that the TNF-α -308A allele was associated with a greater risk of CHD [26]. On the other hand, great heterogeneity was observed in the pooled analysis including all Caucasian studies. This could also, at least partly, be explained by distinct risk factors status within a study (between cases and controls) and among studies. Therefore, the negative result yielded from the pooled analysis of all Caucasian might actually be confounded by the co-existing CHD risk factors. Co-variates used for adjusting the gene effects were largely the same, though not completely identical. Therefore, results from the pooled analysis of these studies were convincible. Relatively less studies concerning Asians. Hou et al. carried out a case-control study and reported a nonsignificant association despite

404

H.-F. Zhang et al. / Thrombosis Research 127 (2011) 400–405

Odds ratio (95% CI)

Case

Control

Confounders for adjustment

Herrmann,1998

1.43 (1.08, 1.90)

641

710

1, 2, 8

Georges,2003

3.30 (1.60, 6.90)

849

314

1-7

Antonicelli,2005

1.64 (1.03, 2.64)

293

310

1, 2, 4-7

Dedoussis,2005

1.94 (1.06, 3.68)

237

237

1-6, 8

Giacconi,2006

1.45 (1.09, 1.93)

105

190

1-4

Overall (Q-test, P = 0.65)

1.50 (1.23, 1.77)

2125

1761

Study

1

1.5

3.69

Fig. 4. Pooled adjusted OR (AA + AG vs. AA) by Mantel-Haenszel fixed-effects model for Caucasian studies. The sizes of the data markers are proportional to sample sizes. The horizontal axis is plotted as ORs. Cofounders of adjustment: 1, age; 2, gender; 3, body mass index; 4, smoking habit; 5, diabetes; 6, hypertension; 7, high-density lipoprotein cholesterol level; 8, family history of coronary heart disease.

adjusted by a wide range of co-variates. The most plausible explanations lie in (1) insufficient statistical power due to small sample size; and (2) population stratification, as indicated by the distinct prevalences of the -308A allele between Asians and Caucasians (15.70% vs. 8.20%), which may lead to inconsistency, especially the case in which both allelic frequencies and incidence rates of the disease varies across ethnic groups [20,45]. Similar considerations should be given in interpreting results from Indian and African studies. Some limitations of this meta-analysis should be acknowledged. First, as mentioned above, few studies and limited participants were included in Asians, Indians, and Africans, which may lead to a falsenegative association due to a lack of statistical power. Second, lack of individual participants' data has restricted further adjustments of the results by potential valuable co-variables. Third, the TNF-α gene is located within the MHC region which contains a number of genes that are crucial for immune regulations and are in linkage disequilibrium with TNF-α [46–48]. Thus, the genetics effects by genes in linkage disequilibrium can not be ruled out. Despite limitations, this meta-analysis suggests that the TNF-α -308A allele is a risk factor for developing CHD in Caucasians but evidence is not sufficient to confirm this association in other ethnic groups. It is needed, on one hand, to conduct larger sample studies, especially those on Asians, Indians, and Africans, using rigorous study designs, homogeneous CHD patients, well matched (especially in risk factors status) and representativeness controls, paying more attention to gene-gene and gene-environment interactions; on the other hand, there is a greater need in genetics epidemiology for quantitative systematic reviews to help conclude more conclusive results [49]. Such strategies may eventually lead to better and comprehensive understandings of the association between the TNF-α G-308A polymorphism and CHD risk.

Conflict of interest statement My co-authors and I declare that we have no financial and personal relationships with other people or organizations that can inappropriately influence our work, there is no professional or other personal interest of any nature or kind in any product, service and/or company that could be construed as influencing the position presented in, or the review of, the manuscript entitled “Tumor necrosis factor-alpha G308A gene polymorphism and coronary heart disease susceptibility: an updated meta-analysis”.

Appendix A. Supplementary data Supplementary data to this article can be found online at doi:10.1016/j.thromres.2010.12.018.

References [1] Hansson GK. Inflammation, atherosclerosis, and coronary artery disease. N Engl J Med 2005;352:1685–95. [2] Libby P, Ridker PM, Maseri A. Inflammation and atherosclerosis. Circulation 2002;105:1135–43. [3] Hansson GK, Libby P. The immune response in atherosclerosis: a double-edged sword. Nat Rev Immunol 2006;6:508–19. [4] Stone PH. Triggering myocardial infarction. N Engl J Med 2004;351:1716–8. [5] Chen Y, Rollins J, Paigen B, Wang X. Genetic and genomic insights into the molecular basis of atherosclerosis. Cell Metab 2007;6:164–79. [6] Hajeer AH, Hutchinson IV. Influence of TNF-alpha gene polymorphisms on TNFalpha production and disease. Hum Immunol 2001;62:1191–9. [7] Kamizono S, Hiromatsu Y, Seki N, Bednarczuk T, Matsumoto H, Kimura A, et al. A polymorphism of the 5' flanking region of tumour necrosis factor-alpha gene is associated with thyroid-associated ophthalmopathy in Japanese. Clin Endocrinol (Oxf) 2000;52:759–64. [8] Herrmann SM, Ricard S, Nicaud V, Mallet C, Arveiler D, Evans A, et al. Polymorphisms of the tumour necrosis factor-alpha gene, coronary heart disease and obesity. Eur J Clin Invest 1998;28:59–66. [9] Yudkin JS, Kumari M, Humphries SE, Mohamed-Ali V. Inflammation, obesity, stress and coronary heart disease: is interleukin-6 the link? Atherosclerosis 2000;148: 209–14. [10] Wilson AG, de Vries N, Pociot F, di Giovine FS, van der Putte LB, Duff GW. An allelic polymorphism within the human tumor necrosis factor-alpha promoter region is strongly associated with HLA A1, B8, and DR3 alleles. J Exp Med 1993;177:557–60. [11] Wilson AG, Symons JA, McDowell TL, McDevitt HO, Duff GW. Effects of a polymorphism in the human tumor necrosis factor-alpha promoter on transcriptional activation. Proc Natl Acad Sci 1997;94:3195–9. [12] Allen RD. Polymorphism of the human TNF-alpha promoter-random variation or functional diversity. Mol Immunol 1999;36:1017–27. [13] Plutzky J. Inflammatory pathways in atherosclerosis and acute coronary syndromes. Am J Cardiol 2001;88:10–5. [14] Pereira TV, Rudnicki M, Franco RF, Pereira AC, Krieger JE. Effect of the G-308A polymorphism of the tumor necrosis factor-alpha gene on the risk of ischemic heart disease and ischemic stroke: a meta-analysis. Am Heart J 2007;153:821–30. [15] Little J, Higgins JPT. The HuGENet™ HuGE review handbook; 2010. version 1.0. Available at: http://www.hugenet.org.uk. Accessed October 10, 2010. [16] Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD, Rennie D, et al. Metaanalysis of observational studies in epidemiology: a proposal for reporting. Metaanalysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA 2000;283:2008–12. [17] Yu W, Gwinn M, Clyne M, Yesupriya A, Khoury MJ. A navigator for human genome epidemiology. Nat Genet 2008;40:124–5. [18] Ioannidis JP, Ntzani EE, Trikalinos TA. 'Racial' differences in genetic effects for complex diseases. Nat Genet 2004;36:1312–8. [19] Thakkinstian A, McEvoy M, Minelli C, Gibson P, Hancox B, Duffy D, et al. Systematic review and meta-analysis of the association between β2-adrenoceptor polymorphisms and asthma: a HuGE review. Am J Epidemiol 2005;162:201–11.

H.-F. Zhang et al. / Thrombosis Research 127 (2011) 400–405 [20] Zhang HF, Zhong BL, Zhu WL, Xie SL, Qiu LX, Zhu LG, et al. CD14 C-260 T gene polymorphism and ischemic heart disease susceptibility: a HuGE review and meta-analysis. Genet Med 2009;11:403–8. [21] Egger M, Davey SG, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997;315:629–34. [22] Banerjee I, Pandey U, Hasan OM, Parihar R, Tripathi V, Ganesh S. Association between inflammatory gene polymorphisms and coronary artery disease in an Indian population. J Thromb Thrombolysis 2009;27:88–94. [23] Ghazouani L, Ben HKS, Abboud N, Hamda KB, Khalfallah AB, Brahim N, et al. TNFalpha -308 G N A and IL-6–174 G N C polymorphisms in the Tunisian patients with coronary artery disease. Clin Biochem 2010;43:1085–9. [24] Sbarsi I, Falcone C, Boiocchi C, Campo I, Zorzetto M, De Silvestri A, et al. Inflammation and atherosclerosis: the role of TNF and TNF receptors polymorphisms in coronary artery disease. Int J Immunopathol Pharmacol 2007;20:145–54. [25] Hou L, Huang J, Lu X, Wang L, Fan Z, Gu D. Polymorphisms of tumor necrosis factoralpha gene and coronary heart disease in a Chinese Han population: interaction with cigarette smoking. Thromb Res 2009;123:822–6. [26] Elahi MM, Gilmour A, Matata BM, Mastana SS. A variant of position -308 of the tumour necrosis factor-alpha gene promoter and the risk of coronary heart disease. Heart Lung Circ 2008;17:14–8. [27] Shun SY, Zeng XQ, Qi AM, Fan WH, Zhang JC. Association of tumor necrosis factoralpha gene polymorphisms in the promoter region with chronic periodontitis and coronary heart disease. J Clin Stomatol 2009;5:279–82. [28] Chen ZQ, Zheng ZF, Ma JF, Qiu FY, Shi SL, Chen H, et al. Plasma levels of TNF-α, but not TNF-α G-308A gene polymorphism, is associated with coronary heart disease. Chin J Geriatr Cardiovasc Cerebrovasc Dis 2001;06(15;3):156–8. [29] Allen RA, Lee EM, Roberts DH, Park BK, Pirmohamed M. Polymorphisms in the TNF-alpha and TNF-receptor genes in patients with coronary artery disease. Eur J Clin Invest 2001;31:843–51. [30] Georges JL, Rupprecht HJ, Blankenberg S, Poirier O, Bickel C, Hafner G, et al. Impact of pathogen burden in patients with coronary artery disease in relation to systemic inflammation and variation in genes encoding cytokines. Am J Cardiol 2003;92:515–21. [31] Vendrell J, Fernandez-Real JM, Gutierrez C, Zamora A, Simon I, Bardaji A, et al. A polymorphism in the promoter of the tumor necrosis factor-alpha gene (-308) is associated with coronary heart disease in type 2 diabetic patients. Atherosclerosis 2003;167:257–64. [32] Bernard V, Pillois X, Dubus I, Benchimol D, Labouyrie JP, Couffinhal T, et al. The -308 G/A tumor necrosis factor-alpha gene dimorphism: a risk factor for unstable angina. Clin Chem Lab Med 2003;41:511–6. [33] Szalai C, Fust G, Duba J, Kramer J, Romics L, Prohaszka Z, et al. Association of polymorphisms and allelic combinations in the tumour necrosis factor-alphacomplement MHC region with coronary artery disease. J Med Genet 2002;39:46–51. [34] Koch W, Kastrati A, Bottiger C, Mehilli J, von BN, Schomig A. Interleukin-10 and tumor necrosis factor gene polymorphisms and risk of coronary artery disease and myocardial infarction. Atherosclerosis 2001;159:137–44. [35] Padovani JC, Pazin-Filho A, Simoes MV, Marin-Neto JA, Zago MA, Franco RF. Gene polymorphisms in the TNF locus and the risk of myocardial infarction. Thromb Res 2000;100:263–9.

405

[36] Tulyakova G, Nasibullin T, Salmanov A, Avzaletdinova D, Khusnutdinova E, Zakirova A, et al. Association of the -308 (G/A) polymorphism of tumor necrosis factor-γ with myocardial infarction and sudden cardiac death. Balk J Med Genet 2004;8:31–5. [37] Li Y, Xu P, Chen H, Zhang PA, Huang CX. Association between tumor necrosis factor-α G-308A gene polymorphism and coronary heart disease: a case-control study. Chin J Geriatr 2003;9:568–9. [38] Tobin MD, Braund PS, Burton PR, Thompson JR, Steeds R, Channer K, et al. Genotypes and haplotypes predisposing to myocardial infarction: a multilocus case-control study. Eur Heart J 2004;25:459–67. [39] Giacconi R, Cipriano C, Muti E, Costarelli L, Malavolta M, Caruso C, et al. Involvement of -308 TNF-alpha and 1267 Hsp70-2 polymorphisms and zinc status in the susceptibility of coronary artery disease (CAD) in old patients. Biogerontology 2006;7:347–56. [40] Dedoussis GV, Panagiotakos DB, Vidra NV, Louizou E, Chrysohoou C, Germanos A, et al. Association between TNF-alpha -308 G N A polymorphism and the development of acute coronary syndromes in Greek subjects: the CARDIO2000-GENE Study. Genet Med 2005;7:411–6. [41] Antonicelli R, Olivieri F, Cavallone L, Spazzafumo L, Bonafe M, Marchegiani F, et al. Tumor necrosis factor-alpha gene -308 G N A polymorphism is associated with STelevation myocardial infarction and with high plasma levels of biochemical ischemia markers. Coron Artery Dis 2005;16:489–93. [42] Liu Y, Jin W, Lu L, Chen QJ, Shen WF. Association between single nucleotide polymorphism in the promoter of tumor necrosis factor-α gene and coronary heart disease. J Diagn Concepts Pract 2009;5:506–9. [43] Bennet AM, van MMC, Hallqvist J, Morgenstern R, Frostegard J, Wiman B, et al. Association of TNF-alpha serum levels and TNFA promoter polymorphisms with risk of myocardial infarction. Atherosclerosis 2006;187:408–14. [44] Xiang PX, Li Y, Zhang PA, Chen H. Plasma TNF-α level and TNF-α gene polymorphism in patients with coronary heart disease: a case-control study. Lab Med 2004;5:434–7. [45] McAteer JB, Prudente S, Bacci S, Lyon HN, Hirschhorn JN, Trischitta V, et al. The ENPP1 K121Q polymorphism is associated with type 2 diabetes in European populations: evidence from an updated meta-analysis in 42 042 subjects. Diabetes 2008;57:1125–30. [46] Bis JC, Heckbert SR, Smith NL, Reiner AP, Rice K, Lumley T, et al. Variation in inflammation-related genes and risk of incident nonfatal myocardial infarction or ischemic stroke. Atherosclerosis 2008;198:166–73. [47] McGuire W, Hill AV, Allsopp CE, Greenwood BM, Kwiatkowski D. Variation in the TNF-alpha promoter region associated with susceptibility to cerebral malaria. Nature 1994;371:508–10. [48] Hajeer AH, Hutchinson IV. Influence of TNF-alpha gene polymorphisms on TNFalpha production and disease. Hum Immunol 2001;62:1191–9. [49] Wheeler JG, Keavney BD, Watkins H, Collins R, Danesh J. Four paraoxonase gene polymorphisms in 11 212 cases of coronary heart disease and 12 786 controls: meta-analysis of 43 studies. Lancet 2004;363:689–95.