Association of tissue plasminogen activator and plasminogen activator inhibitor polymorphism with myocardial infarction: A meta-analysis

Association of tissue plasminogen activator and plasminogen activator inhibitor polymorphism with myocardial infarction: A meta-analysis

Thrombosis Research 130 (2012) e43–e51 Contents lists available at SciVerse ScienceDirect Thrombosis Research journal homepage: www.elsevier.com/loc...

1MB Sizes 0 Downloads 59 Views

Thrombosis Research 130 (2012) e43–e51

Contents lists available at SciVerse ScienceDirect

Thrombosis Research journal homepage: www.elsevier.com/locate/thromres

Regular Article

Association of tissue plasminogen activator and plasminogen activator inhibitor polymorphism with myocardial infarction: A meta-analysis Li-Li Gong a, Jian-Hao Peng b, Fei-Fei Han a, Jin Zhu a, Lian-Hua Fang c, Yue-Hua Wang c, Guan-Hua Du c,⁎, He-Yao Wang a,⁎, Li-Hong Liu a,⁎ a b c

Beijing Chao-Yang Hospital affiliated with Beijing Capital Medical University, Beijing, China Molecular Cardiology, Nephrology and Hypertension, Cleveland Clinic, Cleveland, OH, USA Institute of Materia Medica, Chinese Academy of Medical Science & Peking Union Medical College, Beijing 100050, China

a r t i c l e

i n f o

Article history: Received 21 January 2012 Received in revised form 2 April 2012 Accepted 8 June 2012 Available online 6 July 2012 Keywords: Myocardial infarction Ttissue plasminogen activator Plasminogen activator inhibitor-1 Polymorphism Meta-analysis

a b s t r a c t Introduction: To investigate whether t-PA Alu repeat insertion/deletion (I/D) and PAI-1 4 G/5 G genetic variations are associated with the risk of MI. Methods: We conducted a meta-analysis to assess the association between the t-PA I/D and PAI-1 4 G/5 G polymorphisms and risk of MI. We also performed subgroup analyses based on ethnicity (Caucasian, Asian, and African), gender and age. Forty one eligible studies including 12,461 cases and 14,993 controls were identified to evaluate the impact of PAI-1 4 G/5 G polymorphism on MI. Seven studies investigated the relationship between t-PA I/D and MI. Results: This meta-analysis revealed that the PAI-1 4G allele (4G/4G and 4G/5G genotype) was associated with an increased risk of MI compared with the 5G allele in the overall population (OR=1.094, 95% CI=1.021 - 1.172, p=0.011). The relative risks of MI for 4G/4G genotype was increased when compared to 5G/5G genotype and 5G allele, with odds ratio at 1.157 (95% CI 1.015 - 1.320, p=0.029) and 1.126 (95% CI =1.015 - 1.249, p=0.025). However, the results show that the 4G/5G polymorphism risk for MI was not associated with ethnicity stratification as Caucasian, Asian or African population. No substantial differences in the genotype distributions were observed in the MI group and control group along the lines of gender and age. After multivariable analysis t-PA I/D polymorphism showed no consistent association with MI. Conclusions: This study suggests that the 4G/5G polymorphism of PAI-1 may be a risk factor for MI in overall populations. © 2012 Elsevier Ltd. All rights reserved.

Introduction Cardiovascular diseases are major health problems of highly developed as well as developing countries around the world. Myocardial infarction (MI) is one of the main causes of death from cardiovascular disease [1]. Thrombus formation has been identified as the most common mechanistic events in MI. Activation of the fibrinolytic system depends on two key proteins: tissue plasminogen activator (t-PA), a pivotal activator of plasminogen, and plasminogen activator inhibitor-1 (PAI-1), a major inhibitor of t-PA. The t-PA is the main endogenous fibrinolytic enzyme. Increased t-PA activity causes hyperfibrinolysis, which might lead to excessive bleeding, whereas a decrease in activity results in thrombosis. The

⁎ Corresponding authors at: Beijing Chaoyang Hospital, Affiliate of Capital Medical University, 8 Gongren Tiyuchang Nanlu, Beijing 100020, China. Tel.: +86 10 85231464; fax: +86 10 65004616. E-mail addresses: [email protected] (G.-H. Du), [email protected] (H.-Y. Wang), [email protected] (L.-H. Liu). 0049-3848/$ – see front matter © 2012 Elsevier Ltd. All rights reserved. doi:10.1016/j.thromres.2012.06.015

t-PA is encoded by the plasminogen activator tissue type (PLAT) gene on chromosome 8p11.21 spanning 8 coding exons. In the intron 8 of this gene insertion/deletion (I/D) polymorphism (rs4646972) in an Alu element has been reported to be associated with variable plasma levels of t-PA and increase in the risk of MI [2,3]. However, other studies yielded contradictory results [4–8]. The PAI-1 is located on chromosome 7q21.3-q22 and spans approximately 12.2kb with nine exons and eight introns. Higher levels of PAI-1 protein have been found in patients with MI compared to healthy/normal individuals [9]. In PAI-1 a single nucleotide 4G/5G insertion/deletion polymorphism (rs1799768) is located at position −675bp in the promoter sequence. The 4G allele is associated with elevated PAI-1 levels [10–15]. A positive association of 4G allele and increased risk of MI [13–21] has been reported while other studies are negative [6–8,22–49]. Although previous studies and meta-analyses have reported the relationship between PAI-1 4G/5G polymorphism and MI, the controversy still exist [12,16–18]. In the context of this meta-analysis, we systematically reviewed all studies including more recent studies [3,8,13,19–21] not included in previous meta-analyses [12,16–18]. We also performed subgroup

e44

L.-L. Gong et al. / Thrombosis Research 130 (2012) e43–e51

analyses based on ethnicity, gender and age to examine possible interactions between the t-PA Alu repeat I/D and PAI-1 4 G/5 G polymorphisms on risk of MI. Materials and Methods Search Strategy We performed this meta-analysis in accordance with the “Preferred Reporting Items for Systematic Reviews and Meta-Analyses” (PRISMA) [22] and “Meta-analysis of Observational Studies in Epidemiology” (MOOSE) [23] guidelines. We searched PubMed and EMBASE without language restrictions, inception to September 2011. We also evaluated other meta-analyses and references from related review articles to identify relevant studies. The key words for the PAI-1 were as follows: (“PAI-1” OR “Type 1 Plasminogen Activator Inhibitor” OR “SERPINE1”) AND (“polymorphism” OR “SNP” OR “single nucleotide polymorphism”) AND (“myocardial infarction” OR “myocardial infarct”). The key words for the t-PA were as follows: (“Tissue Plasminogen Activator” OR “Tissue Type Plasminogen Activator” OR “TTPA” OR “T Plasminogen Activator” OR “tPA”) AND (“polymorphism” OR“SNP” OR “single nucleotide polymorphism”) AND (“myocardial infarction” OR “myocardial infarct”). We did not limit the search with any filters. Study Selection Using a structured question format to aid our literature search strategy, we reviewed all abstracts. We reviewed potentially relevant articles in full length to ensure that they satisfied the following criteria: (1) based on a case–control design; (2) evaluation of the association of MI and PAI-1 or t-PA polymorphism; (3) available genotype distributions of both cases and controls; and (4) genotype distribution of the control population must be in Hardy-Weinberg equilibrium (HWE). Two of the authors (Gong LL, Peng JH) independently screened and evaluated the eligibility of all studies on the basis of these criteria and settled differences of opinion by consensus or after consultation with a third investigator.

Data Extraction We extracted data from the selected articles on the following items: the first author's last name, publication year, country where the study was performed, mean age, gender and ethnicity of the study population, related diseases, biochemical parameters (TC, TG, LDL, HDL, PAI-1), numbers of genotyped cases and controls. Patients under 45 years of age defined as young and over 45years as old. Patient ethnicity was categorized as Caucasian, Asian, and African. Statistical Analysis Continuous variables were shown as mean±SEM. Differences between continuous variables were analyzed using the Student's t-test. A value pb0.05 was considered statistically significant. Data were analyzed with SPSS 13.0 statistical software. We calculated odds ratios (ORs) and 95% confidence interval (CI) to assess the association between the PAI-1 4G/5G and t-PA Alu-repeat I/D polymorphisms and myocardial infarction. We examined the association between PAI-1 -675 allele 4G and MI risk compared with that for allele 5G (4G vs. 5G); homozygote 4G/4G was contrasted with 5G/5G. 4G allele (4G/4G and 4G/5G genotype) vs. individuals with the 5G/5G genotype and 5G allele (4G/5G and5G/5G genotype) vs. 4G/4G models were also calculated. The same contrasts were performed for Alu-repeat I/D allele of the t-PA polymorphism. The pooled odds ratios were estimated with fixed-effects model (the Mantel-Haenszel method) [24] when there was lack of heterogeneity, and random effects modeling (the DerSimonian and Laird method) was adopted when heterogeneity existed. Heterogeneity among studies was assessed with the Cochran Q test (pb0.100 indicates heterogeneity) [25] and I 2 statistic test (I 2 >50% considered substantial heterogeneity) [26]. In addition, qualitative assessment of heterogeneity was performed with visual inspection of forest plots to identify variability between study estimates. Publication bias was assessed with Begg funnel plot [27] and Egger weighted regression method [28] for those analyzes with greater than three trials (pb0.05 indicated significant bias). The data were analyzed

Fig. 1. Literature search and selection.

Table 1 Characteristics of individual studies included in the meta-analysis. Authors

Year

Country

Ethnicity

1993 1995 1995 1997 1997 1997 1998 1998 1998 1998 1998 1999 1999 1999 1999 2000 2000 2000 2000 2001 2002 2002 2003 2003 2003 2003 2003 2003 2004 2005 2005 2005 2007 2007 2008 2009 2010 2010 2010 2011 2011

Sweden UK Sweden USA Italy UK Japan UK Germany Sweden Japan Germany Netherlands Italy USA USA France Finland Korea China Japan USA China Slovenia Italy Sweden France USA UK South Africa Netherlands Sweden Brazil France Turkey Mexico Greece China Tunisia India Pakistan

Caucasian Caucasian Caucasian Caucasian/African/Asian Caucasian Caucasian Asian Caucasian Caucasian Caucasian Asian Caucasian Caucasian Caucasian Caucasian African Caucasian Caucasian Asian Asian Asian Caucasian/African/Asian Asian Caucasian Caucasian Caucasian Caucasian Caucasian Caucasian Asian Caucasian Caucasian Caucasian/African/Asian Caucasian Caucasian Caucasian/African/Asian Caucasian Asian African Asian Asian

t-PA Iacoviello L. van der Bom, J. G. Ridker, P. M. Steeds, R. Hooper, W. C. Ashavaid, T. F. Ahmed, W.

1996 1997 1997 1998 2000 2011 2011

Italy Netherlands USA UK Korea India Pakistan

Caucasian Caucasian Caucasian Caucasian Asian Asian Asian

107/73 476/601 93/100 374/495 108/175 158/150 66/62 197/201 241/179 151/150 204/148 1214/1351 331/302 200/200 375/978 201/244 68/164 158/139 110/185 87/92 590/704 78/386 56/83 154/194 1210/1210 1212/1556 589/653 264/753 547/505 195/300 54/55 600/600 121/111 510/543 156/281 127/127 201/140 116/60 305/328 446/473 229/217

114/145 121/250 369/369 529/525 108/185 446/473 229/217

Case

Control

Age

Sex M/F

I/I

I/D

D/D

I/I

I/D

D/D

39.89±0.35 25-64 b45 62.9±8.8 59±7 59.8 63.1±9.2 59.4 38.6±4.4 58.1±4.9 59.3±10.3 62.2±9.5 56.1±9.0 40.7±4.1 63.7±11.6 55 47.9±9 33-69 60.7±9.2 51.3±6.7 62.5±10.8 39.7 67.10±10.41 58.3±11.3 39±5 59.25 51.91±5.44 73.5±5.5 61.9±9.2 b45 35.6 56±10 34.4±4.9 51.91±5.44 59±11 40±4.6 32.2±3.4 64.62 59.0±12.0 58.6±10.4 52.1±11.3

92/15 476/0 93/0 374/0 81/33 N/A 58/8 149/48 241/0 122/29 158/46 1214/0 331/0 185/0 296/79 72/38 185/16 158/0 99/59 50/33 0/590 0/78 44/12 102/52 1061/149 851/361 589/0 158/106 372/175 N/A 38/16 385/215 87/34 510/0 125/31 106/21 176/25 77/39 247/58 358/88 168/61

27 98 15 82 30 26 33 27 43 31 22 226 73 69 77 56 12 32 N/A 19 75 29 2 35 286 214 109 58 108 54 20 131 45 105 30 54 61 14 61 116 79

51 230 38 191 46 73 28 91 112 74 99 606 170 93 193 97 38 64 N/A 29 300 32 14 74 589 595 249 136 280 99 19 307 47 236 75 64 111 41 156 218 86

29 148 40 101 32 59 5 66 86 46 83 382 88 38 105 48 18 62 N/A 39 215 17 40 45 335 359 125 70 159 42 15 162 23 120 51 9 29 61 88 112 64

26 141 20 115 37 49 29 48 34 40 19 258 68 66 218 59 57 25 N/A 22 73 83 6 37 280 313 105 166 106 103 19 134 43 96 96 72 23 18 106 113 76

24 271 54 247 86 65 27 86 93 80 76 684 150 102 457 121 78 60 N/A 45 316 187 52 89 588 768 269 387 237 132 29 280 45 254 112 38 89 27 180 247 89

23 189 26 133 52 36 6 54 52 30 53 409 84 32 303 64 29 54 N/A 25 315 115 25 68 342 436 133 200 162 65 7 186 16 124 73 17 28 15 42 113 52

20-78 73±10 59.6 61.9 55 58.6±10.4 52.1±11.3

85/29 69/52 369/0 363/173 99/59 358/88 168/21

33 49 110 177 15 161 65

58 58 183 254 49 211 97

23 14 76 98 44 74 67

49 75 108 177 37 139 67

69 127 183 257 87 228 98

27 48 78 91 61 106 52

e45

I: insertion; D: deletion; For PAI-1, I/I stand for 5 G/5 G, I/D stand for 4 G/5 G and D/D stand for 4 G/4 G; For t-PA, I/I stand for homozygous Alu repeat insertion, I/D stand for heterozygous Alu repeat insertion/deletion, D/D stand for homozygous Alu repeatdeletion.

L.-L. Gong et al. / Thrombosis Research 130 (2012) e43–e51

PAI-1 Dawson, S. Ye, S. Eriksson, P. Ridker, P. M. Burzotta, F. Ossei-GerningN Sugano, T. Kohler, H. P. Junker, R. Pastinen T Iwai N Gardemann, A. Doggen, C. J. Ardissino, D. Anderson, J. L. Canavy, I. Mikkelsson J Song J Hooper, W. C. Fu, L. Yamada, Y. Hindorff, L. A. Zhan, M. Petrovic, D. Mannucci P.M. Leander, K. Juhan-Vague,I. Crainich, P. Tobin, M. D. Pegoraro, R. J. Martini, C. H. Jood, K. Sampaio, M. F. Morange, P. E. Onalan, O Isordia-Salas, I. Rallidis, L. S. Cao X.-L. Abboud, N. Ashavaid,T. F. Ahmed, W.

Number of cases/controls

e46

L.-L. Gong et al. / Thrombosis Research 130 (2012) e43–e51

Table 2 Characteristic details of subjects included in the PAI-1 and t-PA polymorphism and MI risk studies. Case

Control

P

PAI-1 Age Male (%) BMI (kg/m2) Hypertension (%) Diabetes (%) Smoker (%) Obesity (%) Alcohol (%) TC (mM) TG (mM) LDL (mM) HDL (mM) PAI-1 (ng/mL)

53.44±1.66 76.65±0.04 26.00±0.35 44.00±3.11 22.34±3.66 57.96±4.88 20.90±3.69 51.06±18.64 5.15±0.18 1.92±0.07 3.34±0.25 1.03±0.05 28.99±4.95

52.43±1.60 71.29±0.04 25.33±0.33 27.23±4.70 12.87±4.57 31.28±4.13 12.58±3.60 55.03±18.70 4.93±0.16 1.48±0.10 2.94±0.19 2.56±1.32 22.93±3.96

0.1032 0.0006 0.0012 0.0000 0.0185 0.0000 0.0471 0.4716 0.1284 0.0000 0.0629 0.2713 0.0009

t-PA Age Male (%) BMI (kg/m2) Hypertension (%) Diabetes (%) Smoker (%) TC (mM) TG (mM) LDL (mM) HDL (mM)

60.30±2.51 73.55±4.53 25.72±0.42 36.07±4.79 58.30±14.06 42.35±9.23 5.36±0.37 1.75±0.16 2.49±0.02 1.04±0.09

55.61±3.94 63.47±6.43 25.32±0.29 11.27±4.00 2.42±1.09 22.73±6.73 5.44±0.31 1.49±0.31 2.75±0.06 1.24±0.08

0.0928 0.0206 0.1027 0.0657 0.0288 0.0034 0.5204 0.3424 0.1814 0.0634

For PAI-1 polymorphism studies: Hypertension: 24 studies (Reference [3,8,13,16, 20–22,25,28–32,34,36–40,42,43,45,46,48,49]); Diabetes: 27 studies (Reference [3,8, 13,17,20,21,24,25,28–32,34,36–43,45–49]); Smoker: 27 studies (Reference [3,8,13, 16,17,20,21,24,25,27,28,30–32,36–43,45–49]); Obesity: 7 studies (Reference [8,20, 30,31,39,40,46]); Alcohol: 3 studies (Reference [3,30,39]); TC: 14 studies (Reference [3,8,13,17,19,21,33–35,38,41,46,47,49]); TG: 15 studies (Reference [3,8,13,17,19,21, 27,33–35,38,41,46,47,49]); LDL: 9 studies (Reference [3,8,13,21,27,34,38,46,49]); HDL: 10 studies (Reference [3,8,13,17,21,27,34,38,46,49]); PAI-1: 13 studies (Reference [19,21–23, 27,35,41,47]); For t-PA polymorphism studies: Hypertension: 4 studies (Reference [3,6–8]); Diabetes: 3 studies (Reference [3,6,8]); Smoker: 5 studies (Reference [2,3,6–8]); TC: 5 studies (Reference [2–4,6,8]); TG: 2 studies (Reference [3,8]); LDL: 2 studies (Reference [3,8]); HDL: 10 studies (Reference [2–4,8]);

with the STATA software system (version 11.0, Stata Corporation, College Station, Tex).

Results

Table 3 Summary results of various comparisons.

PAI-1 4 G vs. 5G

4 G/4 G vs. 5 G/5 G

4 G/4G vs. 4G/5G +5G/5G

4G/4G+ 4G/5G vs. 5G/5G

t-PA I vs. D I/I vs. D/D I/I vs. I/D+D/D I/I+I/D vs. DD

Subgroup

OR (95% CI)

PP heterogeneity value

P-Publication bias

All Caucasian Asian African Male Female Young Old All Caucasian Asian Male Female Young Old All Caucasian Asian Male Female Young Old All Caucasian Asian Male Female Young Old

1.094 (1.021, 1.172) 1.067 (0.995, 1.143) 1.211 (0.979, 1.498) 1.476 (0.924, 2.357) 1.077 (0.987, 1.176) 0.922 (0.691, 1.231) 1.037 (0.911, 1.179) 1.064 (0.980, 1.156) 1.157 (1.015, 1.320) 1.140 (0.994, 1.306) 1.241 (0.891, 1.728) 1.138 (0.962, 1.346) 0.901 (0.508, 1.598) 0.960 (0.747, 1.235) 1.150 (0.972, 1.362) 1.126 (1.015, 1.249) 1.047 (0.959, 1.144) 1.375 (0.980, 1.929) 1.084 (0.995, 1.181) 0.894 (0.620, 1.290) 1.072 (0.872, 1.319) 1.070 (0.940, 1.217) 1.099 (0.996, 1.213) 1.079 (0.961, 1.211) 1.026 (0.883, 1.191) 1.056 (0.960, 1.162) 0.893 (0.560, 1.422) 0.978 (0.779, 1.228) 1.123 (0.971, 1.299)

0.000 0.000 0.000 0.097 0.017 0.002 0.016 0.133 0.000 0.000 0.003 0.036 0.004 0.028 0.107 0.000 0.025 0.000 0.060 0.023 0.009 0.520 0.000 0.001 0.112 0.183 0.006 0.004 0.086

0.018 0.097 0.009 N/A 0.057 0.728 0.457 0.784 0.066 0.107 0.042 0.057 0.764 0.693 0.799 0.012 0.063 0.017 0.089 0.352 0.491 0.695 0.198 0.306 0.108 0.079 0.810 0.917 0.871

All All All

1.021 (0.874, 1.193) 0.010 1.045 (0.767, 1.423) 0.014 1.069 (0.935, 1.223) 0.106

0.794 0.532 0.780 0.655 0.329 0.379

All

1.021 (0.814, 1.280) 0.064

0.858 0.936

0.011 0.068 0.078 0.103 0.097 0.582 0.584 0.119 0.029 0.060 0.201 0.130 0.721 0.885 0.104 0.025 0.303 0.065 0.064 0.550 0.370 0.306 0.060 0.133 0.740 0.262 0.633 0.934 0.188

with healthy controls in PAI-1 studies. Several classic risk factors like hypertension, diabetes, smoking, obestity and dyslipidemia were significantly prevalent in the patients. PAI-1 levels were significantly higher in cases than in controls (28.99±4.95 versus 22.93± 3.96ng/mL, p=0.0009).

Search Results PAI Genetic Variations and Risk of MI Fig. 1 shows how we selected relevant studies for inclusion in the meta-analysis. A total of 457 articles were identified in a combined search of the PubMed, Embase and from a manual approach (search of previous studies cited in previous reviews and of references listed in the identified articles). We excluded 145 duplicate articles and an additional 232 articles at the title/abstract level. We reviewed the full texts of the remaining 80 articles. Of these, 32 articles were excluded. The remaining 48 studies were included in the final analysis. Table 1 summarizes the general characteristics of the 48 studies included in the meta-analysis. The selected studies were published between 1993 and 2011, spanning 19 years. 41 relevant studies with a total number of 12,461 cases and 14,993 controls were evaluated for the association between PAI-1 4 G/5 G polymorphism with MI [3,4,7,8,13–15,19–21,29–59]; Seven studies evaluated for the relationship between t-PA Alu repeat I/D and MI. Table 2 shows the characteristics of the case and control groups included in this study. The mean age of the MI group was older than that of the control group. There was a male preponderance in two groups. The BMI in the cases were significantly higher compared

Summary of results from meta-analysis for the association between the PAI-1 4 G/5G polymorphism and MI are shown in Table 3. The meta-analysis revealed that the PAI-1 4 G allele was associated with an increased risk of MI compared with the 5 G allele in the overall population (OR=1.094, 95% CI=1.021 - 1.172, p=0.011; Fig. 2). However, the stratification by ethnicity indicated that the 4 G genotype was not a risk factor for MI in Caucasian, Asian or African population. No substantial differences in the genotype distributions were observed in male or female between the MI and control groups. Same results were seen in both young and old age. The OR for the 4 G/4 G versus 5 G/5 G genotype of PAI-1 in overall group was 1.157 (95% CI 1.015 - 1.320, p=0.029; Fig. 3), and an association was found. However, the PAI-1 4 G/4 G and 5 G/5 G allele distribution in cases and controls was not significantly different in terms of ethnicity, gender and age. To illustrate dummy variables, 4 G allele (4G/4G and 4 G/5G genotype) vs. individuals with the 5 G/5G genotype and 5 G allele (4G/5G and 5 G/5G genotype) vs. 4 G/4 G genotype were also calculated. The

L.-L. Gong et al. / Thrombosis Research 130 (2012) e43–e51

e47

Fig. 2. OR and 95% CI of individual studies and pooled data for the association between the PAI-1 4 G allele and MI in the overall population.

4 G allele was not associated with an elevated risk of MI in any group. The relative risk of MI was increased in the 4 G/4G genotype compared to 5 G allele, with the odds ratio of 1.126 (95% CI=1.015 - 1.249, p=0.025; Fig. 4). No significant interaction was observed between 4 G/4 G genotype and MI in other variation groups.

Publication Bias

T-PA Genetic Polymorphism I/D and Risk of MI

Discussion

In the present study, the odds ratio of MI for the t-PA I genotype compared with the D genotype increased, but has no significance (OR=1.021, 95% CI=0.874 - 1.193, p=0.794). We also found that genotype I/I and D/D frequencies for t-PA were not significantly different between the two groups (OR 1.045, 95% CI=0.767 - 1.423, p=0.780). The risk of MI was not markedly different between the carriers of I allele (I/I and I/D genotypes) and the carriers of the D/D genotype (OR 1.021, 95% CI=0.814 - 1.280, p=0.858). Likewise, D allele (I/D and D/D genotypes) and I/I genotype carriers had similar risks of MI (OR 1.069, 95% CI=0.935 - 1.223, p=0.329).

Several studies indicated PAI-1 4 G/5 G and t-PA Alu repeat I/D polymorphism was associated with myocardial infarction risk, but the results remain inconsistent. So we conducted this meta-analysis to summarize the association from the current literatures and to explore sources of heterogeneity. The 4 G/5 G polymorphism at position −675 is a common insertion/deletion polymorphism of the promoter region of the PAI-1 gene. Both alleles bind a transcriptional factor, only the 5 G binds a repressor protein, such that basal PAI-1 transcription is increased in the presence of the 4 G allele [14,60]. In large epidemiological studies,

Begg's funnel plot and Egger's test were performed to access the publication bias of literatures. The shape of the funnel plot did not reveal any evidence of obvious asymmetry. However, few of our analysis included the requisite number of studies and publication bias remains possible.

e48

L.-L. Gong et al. / Thrombosis Research 130 (2012) e43–e51

Fig. 3. OR and 95% CI of individual studies and pooled data for the 4 G/4 G versus 5 G/5 G genotype and the relationship with MI in the overall population.

elevated plasma PAI-1 levels have been identified as a predictor of MI. In this meta-analysis, we found that PAI-1 levels were associated with MI risk. In this study, PAI-1 4 G and 4 G/4 G genotype revealed a significant association with MI in overall populations, but no such association with the 4 G/4 G+4 G/5 G versus 5 G/5 G genotypes. That is 4 G allele carriers had a higher risk of MI compared with 5 G allele carriers. The ORs of the PAI-1 4 G/5 G gene polymorphism stratification by ethnicity were greater than 1, although the difference did not reach statistical significance. As only ten Asian and two African population studies were included, these results should be interpreted with caution. More Asian and African studies are needed to confirm this possible association. We found no clear indication of synergistic interaction effects involving PAI-1 4G/5G polymorphism in either male or female. However, the ORs for female were less than one, while greater than one was seen in male groups. It is worth exploring the different means of 4G/5G polymorphism in men and women. But in our analysis, there were only twelve male and four female studies, as most of the published results do not separately consider men and women. In other stratification studies, there were more male patients involved, thus less data were collected from female.

Further studies analyzing male and female separately are needed to determine this possible association. Our meta-analysis revealed no association between 4G/5G polymorphism of the PAI-1 gene and myocardial infarction in terms of age. The ORs of the PAI-1 4G/5G gene polymorphism in young groups were smaller than that in old groups, although there was no significant difference. Because only twelve young and five old studies were included, this result should be interpreted with caution, and more studies are needed. Since there is limited literatures mentioning the stratification studies such as ethnicity, sex and age, the results may not reflect the overall true situation. Comprehensive experimental designs are needed in the future studies. No association was found between MI and t-PA Alu repeat I/D polymorphism. As the studies regarding this genotype were not enough, the results should be interpreted with caution, and more studies are needed. Several potential limitations of this study should be considered. First, environmental interactions may play a role in the pathogenesis of MI. Many environmental factors such as age, gender, hypertension, diabetes, smoking, hyperlipidemia, etc. may be related to increased risk of MI. However, lack of original data limited our further evaluation

L.-L. Gong et al. / Thrombosis Research 130 (2012) e43–e51

e49

Fig. 4. OR and 95% CI of individual studies and pooled data for the 4 G/4 G genotype compared with 5 G allele and the relative risk with MI in the overall population.

in different populations. Second, publication bias cannot be ruled out and is suggested in some of the analyses. We could not exclude the possibility of publication bias for some comparisons and outcomes, because some null and unexpected results may not be published. Therefore, the results of our meta-analysis should be interpreted with caution. In conclusion, results of this meta-analysis indicate that the 4G/5G polymorphism in the promoter of the SERPINE1 gene may be a risk factor for MI in overall populations. No association was found between the t-PA Alu repeat I/D polymorphism and MI in a meta-analysis of data extracted from 7 studies. It is necessary to conduct large numbers of worldwide studies to investigate the role of these polymorphisms in MI. Moreover, ethnicity, gender and age should also be considered in the analysis. Such studies, taking these factors into account may eventually lead to a better, comprehensive understanding of the association between these polymorphism and myocardial infarction risk.

Contributors Gong LL, Peng JH contributed equally to this manuscript as first authors. Prof. Wang HY, Du GH and Liu LH are equal corresponding authors.

Conflict of Interest Statement We declare that we have no conflict of interest. Acknowledgements This study was supported by the Health industry specific research (No. 200902008). References [1] Anversa P, Sonnenblick EH. Ischemic cardiomyopathy: pathophysiologic mechanisms. Prog Cardiovasc Dis 1990;33:49–70. [2] van der Bom JG, de Knijff P, Haverkate F, Bots ML, Meijer P, de Jong PT, et al. Tissue plasminogen activator and risk of myocardial infarction. The Rotterdam Study. Circulation 1997;95:2623–7. [3] Ashavaid TF, Todur SP, Kondkar AA, Nair KG, Shalia KK, Dalal JJ, et al. Platelet polymorphisms: frequency distribution and association with coronary artery disease in an Indian population. Platelets 2011;22:85–91. [4] Ridker PM, Hennekens CH, Lindpaintner K, Stampfer MJ, Miletich JP. Arterial and venous thrombosis is not associated with the 4G/5G polymorphism in the promoter of the plasminogen activator inhibitor gene in a large cohort of US men. Circulation 1997;95:59–62.

e50

L.-L. Gong et al. / Thrombosis Research 130 (2012) e43–e51

[5] Iacoviello L, Di Castelnuovo A, De Knijff P, D'Orazio A, Amore C, Kluft C, et al. Alu-repeat polymorphism in the tissue-type plasminogen activator (t-PA) gene, t-PA levels and risk of familial myocardial infarction (MI). Fibrinolysis Proteolysis 1996;10:13–6. [6] Steeds R, Adams M, Smith P, Channer K, Samani NJ. Distribution of tissue plasminogen activator insertion/deletion polymorphism in myocardial infarction and control subjects. Thromb Haemost 1998;79:980–4. [7] Hooper WC, Lally C, Austin H, Renshaw M, Dilley A, Wenger NK, et al. The role of the t-PA I/D and PAI-1 4G/5G polymorphisms in African-American adults with a diagnosis of myocardial infarction or venous thromboembolism. Thromb Res 2000;99:223–30. [8] Ahmed W, Malik M, Saeed I, Khan AA, Sadeque A, Kaleem U, et al. Role of tissue plasminogen activator and plasminogen activator inhibitor polymorphism in myocardial infarction. Mol Biol Rep 2011;38:2541–8. [9] Falk E. Stable versus unstable atherosclerosis: clinical aspects. Am Heart J 1999;138: S421–5. [10] Kathiresan S, Gabriel SB, Yang Q, Lochner AL, Larson MG, Levy D, et al. Comprehensive survey of common genetic variation at the plasminogen activator inhibitor-1 locus and relations to circulating plasminogen activator inhibitor-1 levels. Circulation 2005;112: 1728–35. [11] Dawson S, Hamsten A, Wiman B, Henney A, Humphries S. Genetic variation at the plasminogen activator inhibitor-1 locus is associated with altered levels of plasma plasminogen activator inhibitor-1 activity. Arterioscler Thromb 1991;11:183–90. [12] Boekholdt SM, Bijsterveld NR, Moons AH, Levi M, Buller HR, Peters RJ. Genetic variation in coagulation and fibrinolytic proteins and their relation with acute myocardial infarction: a systematic review. Circulation 2001;104:3063–8. [13] Abboud N, Ghazouani L, Saidi S, Ben-Hadj-Khalifa S, Addad F, Almawi WY, et al. Association of PAI-1 4G/5G and -844G/A gene polymorphisms and changes in PAI-1/tissue plasminogen activator levels in myocardial infarction: a case–control study. Genet Test Mol Biomarkers 2010;14:23–7. [14] Eriksson P, Kallin B, van 't Hooft FM, Bavenholm P, Hamsten A. Allele-specific increase in basal transcription of the plasminogen-activator inhibitor 1 gene is associated with myocardial infarction. Proc Natl Acad Sci U S A 1995;92:1851–5. [15] Ossei-Gerning N, Mansfield MW, Stickland MH, Wilson IJ, Grant PJ. Plasminogen activator inhibitor-1 promoter 4G/5G genotype and plasma levels in relation to a history of myocardial infarction in patients characterized by coronary angiography. Arterioscler Thromb Vasc Biol 1997;17:33–7. [16] Iacoviello L, Burzotta F, Di Castelnuovo A, Zito F, Marchioli R, Donati MB. The 4G/5G polymorphism of PAI-1 promoter gene and the risk of myocardial infarction: a meta-analysis. Thromb Haemost 1998;80:1029–30. [17] Koch W, Schrempf M, Erl A, Mueller JC, Hoppmann P, Schomig A, et al. 4G/5G polymorphism and haplotypes of SERPINE1 in atherosclerotic diseases of coronary arteries. Thromb Haemost 2010;103:1170–80. [18] Ye Z, Liu EH, Higgins JP, Keavney BD, Lowe GD, Collins R, et al. Seven haemostatic gene polymorphisms in coronary disease: meta-analysis of 66,155 cases and 91,307 controls. Lancet 2006;367:651–8. [19] Isordia-Salas I, Leanos-Miranda A, Sainz IM, Reyes-Maldonado E, Borrayo-Sanchez G. Association of the plasminogen activator inhibitor-1 gene 4G/5G polymorphism with ST elevation acute myocardial infarction in young patients. Rev Esp Cardiol 2009;62:365–72. [20] Rallidis LS, Gialeraki A, Merkouri E, Liakos G, Dagres N, Sionis D, et al. Reduced carriership of 4G allele of plasminogen activator inhibitor-1 4G/5G polymorphism in very young survivors of myocardial infarction. J Thromb Thrombolysis 2010;29: 497–502. [21] Cao X-L, ZC-Y, Yin L, Wang S-C, Jia X-L, Huang H, et al. Reactive protein, plasminogen activator inhibitor type-1 (PAI-1) levels, PAI-1 promoter 4G/5G polymorphism and acute myocardial infarction. J Geriatr Cardiol 2010;7:147–51. [22] Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med 2009;151:264–9 [W64]. [23] Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD, Rennie D, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA 2000;283:2008–12. [24] Mantel N, Haenszel W. Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Inst 1959;22:719–48. [25] DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986;7: 177–88. [26] Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ 2003;327:557–60. [27] Begg CB, Mazumdar M. Operating characteristics of a rank correlation test for publication bias. Biometrics 1994;50:1088–101. [28] Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997;315:629–34. [29] Dawson SJ, Wiman B, Hamsten A, Green F, Humphries S, Henney AM. The two allele sequences of a common polymorphism in the promoter of the plasminogen activator inhibitor-1 (PAI-1) gene respond differently to interleukin-1 in HepG2 cells. J Biol Chem 1993;268:10739–45. [30] Ye S, Green FR, Scarabin PY, Nicaud V, Bara L, Dawson SJ, et al. The 4G/5G genetic polymorphism in the promoter of the plasminogen activator inhibitor-1 (PAI-1) gene is associated with differences in plasma PAI-1 activity but not with risk of myocardial infarction in the ECTIM study. Etude CasTemoins de I'nfarctus du Mycocarde. Thromb Haemost 1995;74:837–41. [31] Burzotta F, Di Castelnuovo A, Amore C, D'Orazio A, Donati MB, Iacoviello L. 4G/5G polymorphism in the promoter region of the PAI-1 gene is not a risk factor for familial myocardial infarction in subjects over 45 years. Thromb Haemost 1997;78: 1294–5.

[32] Iwai N, Shimoike H, Nakamura Y, Tamaki S, Kinoshita M. The 4G/5G polymorphism of the plasminogen activator inhibitor gene is associated with the time course of progression to acute coronary syndromes. Atherosclerosis 1998;136: 109–14. [33] Junker R, Heinrich J, Schulte H, Tataru M, Kohler E, Schonfeld R, et al. Plasminogen activator inhibitor-1 4G/5G-polymorphism and factor V Q506 mutation are not associated with myocardial infarction in young men. Blood Coagul Fibrinolysis 1998;9:597–602. [34] Kohler HP, Stickland MH, Ossei-Gerning N, Carter A, Mikkola H, Grant PJ. Association of a common polymorphism in the factor XIII gene with myocardial infarction. Thromb Haemost 1998;79:8–13. [35] Pastinen T, Perola M, Niini P, Terwilliger J, Salomaa V, Vartiainen E, et al. Array-based multiplex analysis of candidate genes reveals two independent and additive genetic risk factors for myocardial infarction in the Finnish population. Hum Mol Genet 1998;7:1453–62. [36] Sugano T, Tsuji H, Masuda H, Nakagawa K, Nishimura H, Kasahara T, et al. Plasminogen activator inhibitor-1 promoter 4G/5G genotype is not a risk factor for myocardial infarction in a Japanese population. Blood Coagul Fibrinolysis 1998;9:201–4. [37] Anderson JL, Muhlestein JB, Habashi J, Carlquist JF, Bair TL, Elmer SP, et al. Lack of association of a common polymorphism of the plasminogen activator inhibitor-1 gene with coronary artery disease and myocardial infarction. J Am Coll Cardiol 1999;34:1778–83. [38] Ardissino D, Mannucci PM, Merlini PA, Duca F, Fetiveau R, Tagliabue L, et al. Prothrombotic genetic risk factors in young survivors of myocardial infarction. Blood 1999;94:46–51. [39] Doggen CJ, Bertina RM, Cats VM, Reitsma PH, Rosendaal FR. The 4G/5G polymorphism in the plasminogen activator inhibitor-1 gene is not associated with myocardial infarction. Thromb Haemost 1999;82:115–20. [40] Gardemann A, Lohre J, Katz N, Tillmanns H, Hehrlein FW, Haberbosch W. The 4G4G genotype of the plasminogen activator inhibitor 4G/5G gene polymorphism is associated with coronary atherosclerosis in patients at high risk for this disease. Thromb Haemost 1999;82:1121–6. [41] Canavy I, Henry M, Morange PE, Tiret L, Poirier O, Ebagosti A, et al. Genetic polymorphisms and coronary artery disease in the south of France. Thromb Haemost 2000;83:212–6. [42] Mikkelsson J, Perola M, Wartiovaara U, Peltonen L, Palotie A, Penttila A, et al. Plasminogen activator inhibitor-1 (PAI-1) 4G/5G polymorphism, coronary thrombosis, and myocardial infarction in middle-aged Finnish men who died suddenly. Thromb Haemost 2000;84:78–82. [43] Song J, Yoon YM, Jung HJ, Hong SH, Park H, Kim JQ. Plasminogen activator inhibitor-1 4G/5G promoter polymorphism and coagulation factor VII Arg353– >Gln polymorphism in Korean patients with coronary artery disease. J Korean Med Sci 2000;15:146–52. [44] Fu L, Jin H, Song K, Zhang C, Shen J, Huang Y. Relationship between gene polymorphism of the PAI-1 promoter and myocardial infarction. Chin Med J (Engl) 2001;114:266–9. [45] Hindorff LA, Schwartz SM, Siscovick DS, Psaty BM, Longstreth Jr WT, Reiner AP. The association of PAI-1 promoter 4G/5G insertion/deletion polymorphism with myocardial infarction and stroke in young women. J Cardiovasc Risk 2002;9: 131–7. [46] Yamada Y, Izawa H, Ichihara S, Takatsu F, Ishihara H, Hirayama H, et al. Prediction of the risk of myocardial infarction from polymorphisms in candidate genes. N Engl J Med 2002;347:1916–23. [47] Crainich P, Jenny NS, Tang Z, Arnold AM, Kuller LH, Manolio T, et al. Lack of association of the plasminogen activator inhibitor-1 4G/5G promoter polymorphism with cardiovascular disease in the elderly. J Thromb Haemost 2003;1:1799–804. [48] Juhan-Vague I, Morange PE, Frere C, Aillaud MF, Alessi MC, Hawe E, et al. The plasminogen activator inhibitor-1–675 4G/5G genotype influences the risk of myocardial infarction associated with elevated plasma proinsulin and insulin concentrations in men from Europe: the HIFMECH study. J Thromb Haemost 2003;1:2322–9. [49] Leander K, Wiman B, Hallqvist J, Sten-Linder M, de Faire U. PAI-1 level and the PAI-1 4G/5G polymorphism in relation to risk of non-fatal myocardial infarction: results from the Stockholm Heart Epidemiology Program (SHEEP). Thromb Haemost 2003;89:1064–71. [50] [MPA]Mannucci PM, Ardissino D, Barzuini C, Bernardi F, Bernardinelli L, Cavallini C, et al. No evidence of association between prothrombotic gene polymorphisms and the development of acute myocardial infarction at a young age. Circulation 2003;107:1117–22. [51] Petrovic D, Globocnik-Petrovic M, Peterlin B. 4G4G genotype of PAI-1 gene promoter polymorphism is not associated with myocardial infarction in Caucasians with type-2 diabetes. Cardiology 2003;100:157–8. [52] Zhan M, Zhou Y, Han Z. Plasminogen activator inhibitor-1 4G/5G gene polymorphism in patients with myocardial or cerebrovascular infarction in Tianjin, China. Chin Med J (Engl) 2003;116:1707–10. [53] 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. [54] Jood K, Ladenvall P, Tjarnlund-Wolf A, Ladenvall C, Andersson M, Nilsson S, et al. Fibrinolytic gene polymorphism and ischemic stroke. Stroke 2005;36:2077–81. [55] Martini CH, Doggen CJ, Cavallini C, Rosendaal FR, Mannucci PM. No effect of polymorphisms in prothrombotic genes on the risk of myocardial infarction in young adults without cardiovascular risk factors. J Thromb Haemost 2005;3:177–9. [56] Pegoraro RJ, Ranjith N. Plasminogen activator inhibitor type 1 (PAI-1) and platelet glycoprotein IIIa (PGIIIa) polymorphisms in young Asian Indians with acute myocardial infarction. Cardiovasc J S Afr 2005;16:266–70.

L.-L. Gong et al. / Thrombosis Research 130 (2012) e43–e51 [57] Morange PE, Saut N, Alessi MC, Yudkin JS, Margaglione M, Di Minno G, et al. Association of plasminogen activator inhibitor (PAI)-1 (SERPINE1) SNPs with myocardial infarction, plasma PAI-1, and metabolic parameters: the HIFMECH study. Arterioscler Thromb Vasc Biol 2007;27:2250–7. [58] Sampaio MF, Hirata MH, Hirata RD, Santos FC, Picciotti R, Luchessi AD, et al. AMI is associated with polymorphisms in the NOS3 and FGB but not in PAI-1 genes in young adults. Clin Chim Acta 2007;377:154–62.

e51

[59] Onalan O, Balta G, Oto A, Kabakci G, Tokgozoglu L, Aytemir K, et al. Plasminogen activator inhibitor-1 4G4G genotype is associated with myocardial infarction but not with stable coronary artery disease. J Thromb Thrombolysis 2008;26: 211–7. [60] Kohler HP, Grant PJ. Plasminogen-activator inhibitor type 1 and coronary artery disease. N Engl J Med 2000;342:1792–801.