SEARCH

SEARCH BY CITATION

Keywords:

  • acute myocardial infarction;
  • association study;
  • factor V Leiden;
  • plasminogen activator inhibitor (PAI-1);
  • prothrombin

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

See also Tosetto A. Thrombophilic mutations and cardiovascular disease: the case is still open. This issue, pp 2113–5.

Summary. Aims: Gain-of-function variants of genes encoding coagulation factor V (F5 G1691A) and prothrombin (F2 G20210A) cause hypercoagulability and are established risk factors for venous thrombosis. A meta-analysis of 66 155 cases and 91 307 controls found that either polymorphism is associated with a moderately increased risk of coronary artery disease (CAD). Because genetic factors play a particularly important role when acute myocardial infarction (AMI) occurs in the young, we chose to replicate these results by investigating, in the frame of a case-control study, a large cohort of Italian patients who had AMI before the age of 45 years. Methods and Results: In 1880 patients with AMI (1680 men and 210 women) and an equal number of controls, the minor A allele of F5 G1691A (2.6% frequency in cases and 1.7% in controls) was associated with an increased risk of AMI, the association remaining significant after adjustment for traditional risk factors (OR, 1.66; 95% CI, 1.15–2.38; P = 0.006). The positive association with AMI for the minor A allele of F2 G20210A (2.5% frequency in cases and 1.9% in controls) did not reach statistical significance (OR, 1.32; 95% CI, 0.96–1.80; P = 0.159). Conclusions: In a large cohort of young AMI patients the gain-of-function variant F5 G1691A was associated with an increased risk of AMI. The findings on the variant F2 G20210A confirmed the previously reported results, but the association was statistically not significant. These data suggest that a number of young patients with AMI carry gene variants associated with a procoagulant phenotype.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

Occlusion of coronary arteries by platelet- and fibrin-rich thrombi is the ultimate event leading to acute myocardial infarction (AMI). In the last two decades, this well-established mechanism of disease has prompted an array of studies designed to investigate whether or not factors involved in the hemostasis pathways (platelets, coagulation and fibrinolysis) play a mechanistic role. The goal of these studies was to find markers that would help to improve the prediction of risk in primary and secondary prevention, to identify new mechanisms of disease and to tailor the choice of antithrombotic drugs. A widely followed approach has been the study of genes encoding hemostasis proteins, to ascertain whether or not polymorphic variants causing gain- or loss-of-function of the corresponding proteins were associated with a higher risk of, or protection from, coronary artery disease (CAD) [1–6]. Published results were largely inconclusive, owing to insufficient sample size, inaccurate genotyping methods and the confounding effect of repeated testing [4,7].

In an attempt to overcome these limits, Ye et al. [8] chose to perform a meta-analysis of the published data on hemostasis gene variants in CAD. The criteria for selection of the seven gene variants included in their large meta-analysis were the total sample size of the genetic association studies (at least 5000 cases with AMI or coronary artery stenosis and an equal number of controls), but also biological plausibility, because each variant had to alter the function and/or the plasma levels of the corresponding hemostasis protein [8]. Two variants were associated with a moderately increased risk of combined AMI and coronary stenosis: the rare gain-of-function variants of coagulation factor V (FV) known as FV Leiden (G1691A, per allele relative risk 1.17) and prothrombin (G20210A, per allele relative risk 1.31) [8]. Both are markers of hypercoagulability, because the variant coagulation factors cause heightened thrombin formation and are established risk factors for venous thromboembolism (relative risks ranging from 3.0 to 7.0). Also the [-675]4G variant of the gene encoding plasminogen activator inhibitor (PAI-1, a principal inhibitor of plasma fibrinolysis) was associated with a smaller but statistically significant increased risk of AMI (per allele relative risk 1.04) [8].

With this as background, we attempted to replicate, in a large Italian case-control study, the association of the three gene variants [8] in patients with AMI that had occurred before the age of 45 years. Early-onset AMI was chosen as the replication model because of the higher degree of homogeneity of the case material, and because genetic factors play a more prominent role in younger than in older patients with AMI [9,10]. This study is an extension of a previous one reporting lack of association with AMI of nine hemostasis gene variants (from 1210 to 1880 cases/controls) [11]. Repeated genotyping using more accurate methods were instrumental in disclosing an association between AMI and the FV Leiden variant, and in better estimating the effect of this variant.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

Subjects

The Atherosclerosis, Thrombosis, and Vascular Biology Study Group (ATVB) is a clinical network involving 125 Italian coronary care units. Over a 10-year period (January 1997–January 2007) ATVB recruited 1880 Caucasian patients with early-onset AMI (a first event before the age of 45) and an equal number of controls. AMI was defined as resting chest pain lasting more than 30 min, associated with persistent electrocardiographic changes, and confirmed by an increased total creatine kinase (or its myocardial-type fraction) of more than twice the upper normal limits. The 1880 controls were healthy subjects unrelated to the cases but individually matched with them in terms of age, sex, schooling degree and geographical origin. They were enrolled from the staff of the same participating hospitals and had no history of thromboembolic disease. At the time of blood sampling, a questionnaire, documenting medical diagnoses, lifestyle, medications and cardiovascular risk factors, was filled in for cases and controls. Definitions for a positive family history of early onset CAD, hypercholesterolemia, hypertension, being overweight and diabetes were previously reported [11]. Data on coronary angiograms were available for 1649 of the 1880 patients. The definition of normal or stenotic coronary arteries was previously reported [11]. The study was approved by the Institutional Review Boards of the participating hospitals. Informed consent was obtained from all subjects.

Genotyping

The PAI-1 [-675]4G/5G promoter variant (rs1799889) was genotyped by fluorescent PCR amplification of genomic DNA (extracted from leukocytes by salting-out) followed by capillary electrophoresis. To allow the simultaneous analysis of two amplification products on the same run, two PCR primer couples (sequences available on request) were designed to amplify fragments easily distinguishable by size. PCRs were performed on 10 ng of genomic DNA using the FastStart Taq DNA Polymerase (Roche Applied Science, Indianapolis, IN, USA) in a Mastercycler EPgradient thermal cycler (Eppendorf AG, Hamburg, Germany) using a touch-down thermal profile. The labeled products obtained with the two primer couples on two separate plates were mixed and then analyzed using an ABI-3130XL Genetic Analyzer and the GeneMapper 2 software (Applied Biosystems, Foster City, CA, USA).

G1691A variant of the FV (F5) gene (FV Leiden) and G20210A variant of the prothrombin (F2) gene  Genotyping for FV Leiden (rs6025) and prothrombin G20210A (rs1799963) variants was carried out by a fluorescence resonance energy transfer (FRET)-based real-time PCR [12]. Reactions were performed in 384-well plates in a 10-μL final volume using the LightCycler 480 Genotyping Master Mix (Roche). Melting curves were analyzed with the LightCycler 480 Genotyping Software, Version 1.5 (Roche). To assess the sensitivity and specificity of each single nucleotide polymorphism (SNP) assay, 92 DNA samples were blindly sequenced for the two genotyped variations: concordance was 100% for both SNPs. Genotype interpretation for each variant was performed independently by two investigators and the very few samples with unclear results (< 1%) were re-genotyped. Moreover, the FV Leiden variant was also genotyped at the Broad Institute (Cambridge, MA, USA) in the entire case-control sample by the MassARRAY® system (Sequenom, San Diego, CA, USA) platform. Genotyping was performed at the Broad Institute in the frame of a quality control exercise before including our cases and controls in a genome-wide association study. A discrepancy rate lower than 0.4% was found.

Statistical analysis

All analyses were performed using either the software package plink v. 1.04 [13] or the R program (http://www.r-project.org/). Deviations from Hardy–Weinberg equilibrium (HWE) were tested in controls using the exact test described by Wigginton et al. [14], which is more accurate for rare genotypes. For each SNP, standard case-control analyses on allele and genotype frequency data were performed with chi-square statistics (Fisher exact test); asymptotic P values, odds ratios (ORs) and 95% confidence intervals (CI) were provided for minor alleles, and all P values are presented as non-corrected. The allelic association test was performed conditional on matching by using the Cochran-Mantel-Haenszel (CMH) statistic, and subsequent adjustment for traditional cardiovascular risk factors was performed by adding those covariates in a multiple logistic regression model.

The extent to which associations with the F5 and F2 variants were modified by smoking and hypercholesterolemia was assessed through secondary analyses stratified by these risk factors, evaluating heterogeneity between the strata by means of the Breslow-Day test. In these secondary analyses genotypic association tests were first performed unconditional on matching; adjustment for traditional risk factors was carried out using a logistic regression model. In all cases, the level of statistical significance was set at 5%. A power estimate indicated that, if each analyzed variant (minor allele frequency of 2%) were to confer a 1.5-fold increase in the relative risk of AMI, the case and control groups used in this study would be of sufficient size to have 82% power to detect a significant association at the 0.05 level. The study was conducted according to the Strengthening the Reporting of Genetic Association Studies (STREGA) guidelines [15].

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

Phenotypic characteristics of cases and controls

The case sample consisted of 1670 men and 210 women, whose mean age, at the time of AMI, was 39.6 ± 4.9 years. An equal number (n = 1880) of healthy subjects was the control group. The general characteristics of the study populations are: (i) 44.9% of patients were current smokers, 42.6% were former smokers and 12.5% had never smoked (the corresponding figures for controls were 30.4%, 18% and 51.6%); (ii) 61.5% of patients and 45.5% of controls had hypercolesterolemia; (iii) 7.6% of patients and 0.7% of controls had diabetes; (iv) 27.6% of cases and 9.1% of controls had hypertension; and (v) mean body mass index was 26.77 ± 4.2 and 25.01 ± 3.3 for patients and controls.

Association analyses

For the SNPs within F5, F2 and PAI-1, the genotyping success rate exceeded 95% (98.8% for rs6025, 99.4% for rs1799963, and 95.1% for rs1799889), and accuracy was > 99% according to random duplicated genotyping of 5% of samples. The genotype frequencies for all the SNPs showed no significant deviation from HWE in controls. Re-genotyping of the 1210 cases and controls reported in our previous study [11] revealed a number of discrepancies for the FV Leiden and prothrombin G20210A variants, likely to be due to the smaller accuracy of the previously used genotyping technique (PCR followed by endonuclease restriction). Furthermore, the FV Leiden genotypes newly determined by us were independently genotyped at the Broad Institute on all samples by mass spectrometry analysis. The seven discordant samples were subjected to DNA sequencing of the relevant genomic region to unequivocally determine their genotype.

Table 1 shows the results of our allelic association analysis for each of the three variants in the whole study cohort. We also report the results obtained in the previous genotyping exercise (1210 cases vs. 1210 controls) [11], compared with the results obtained using the improved genotyping of the same 2420 individuals. Only the FV Leiden variant showed an association with early-onset AMI (OR = 1.61, 95% CI = 1.16–2.22), which remained statistically significant after adjustment for covariates (OR = 1.66, 95% CI = 1.15–2.38). Pertaining to the prothrombin G20210A variant, a trend towards statistical significance (OR = 1.32, 95% CI = 0.96–1.80) was observed in our unadjusted analysis, the strength of the association being slightly decreased after covariate correction (OR = 1.28, 95% CI = 0.91–1.79). No significant allelic association between AMI and the PAI-1 variant could be detected (Table 1).

Table 1.   Hemostasis gene variants: analysis of allele frequency differences between AMI cases and controls
GenedbSNPVariant typeNucleotide changeMinor alleleNo. casesMAF in casesMAF in controlsUnadjusted P valueUnadjusted OR (95% CI)Adjusted P valueAdjusted OR (95% CI)
  1. UTR, untranslated region; MAF, minor allele frequency. *Results obtained on the original genotyping data [11] and on fresh genotypes of the same individuals. Significant results are indicated in bold; adjustment took into account hypertension, smoking, diabetes, hypercholesterolemia and BMI.

F5rs6025CodingG1691AA18800.0260.0170.0041.61 (1.16–2.22)0.0061.66 (1.15–2.38)
1210*0.0160.0180.5060.86 (0.56–1.33)0.8490.95 (0.58–1.55)
12100.0230.0160.0861.45 (0.95–2.22)0.0711.55 (0.96–2.49)
PAI-1rs17998895′ flanking[-675]4G/5G5G18800.4860.4890.8471.01 (0.92–1.11)0.8470.99 (0.89–1.10)
1210*0.4800.4740.7061.02 (0.91–1.15)0.6771.03 (0.91–1.16)
12100.4840.4840.7571.02 (0.90–1.15)0.8861.01 (0.88–1.15)
F2rs17999633′ UTRG20210AA18800.0250.0190.0791.32 (0.96–1.80)0.1591.28 (0.91–1.79)
1210*0.0170.0160.8191.05 (0,67–1,65)0.5061.19 (0.72–1.97)
12100.0240.0180.1421.35 (0.90–2.04)0.2731.28 (0.82–2.00)

The distribution of genotype frequencies between cases and controls was different only for FV Leiden. Given the rarity of the homozygous mutant genotype, the association could be analyzed exclusively under a dominant model (OR = 1.58, 95% CI = 1.13–2.19; crude estimates).

Secondary analyses

When data were stratified by gender, the association between AMI and FV Leiden was strengthened in men (P = 0.001; OR = 1.96, 95% CI = 1.33–2.88; adjusted analysis) but disappeared in women (P = 0.193; OR = 0.38, 95% CI = 0.09–1.63; adjusted analysis).

A preliminary test for homogeneity was used to identify possible interactions with smoking and hypercholesterolemia (the only covariates having sufficient statistical power for the analyses). A tendency to heterogeneity (P = 0.082) was detected for FV Leiden between groups stratified according to the presence/absence of hypercholesterolemia (but not according to smoking status). Hence, to assess whether or not hypercholesterolemia modified the risk of AMI associated with FV Leiden, the distribution of the clumped GA+AA genotypes was analyzed after stratification for this risk factor (Table 2). Crude analyses revealed that there was no increased risk of AMI among non-hypercholesterolemic individuals who had the GA+AA genotypes. Among hypercholesterolemic individuals, the presence of the GA+AA genotypes further increased the risk of AMI 2.01-fold (95% CI = 1.23–3.28). The concomitant presence of the GA+AA genotypes and the hypercholesterolemia phenotype raised the risk of AMI 3.71-fold (95% CI = 2.27–6.07).

Table 2.   Effect of FV Leiden homozygous (AA) and heterozygous (AG) genotypes on risk of AMI among individuals with or without hypercholesterolemia (serum cholesterol < 5.2 mmol L−1 or intake of statins)
StatusFV genotypeCasesControlsOR* (95% CI)Adjusted OR* (95% CI)
  1. OR, odds ratio; 95% CI, 95% confidence interval; hyper, hypercholesterolemia. *All ORs are relative to the reference category (i.e. non-hypercholesterolemic individuals with FV rs6025 GG genotype). ORs are adjusted for diabetes, hypertension, BMI and smoking.

No hyperGG64792111
No hyperGA+AA23321.02 (0.59–1.76)1.00 (0.58–1.72)
HyperGG9737501.85 (1.61–2.12)1.74 (1.51–2.00)
HyperGA+AA60233.71 (2.27–6.07)3.58 (2.19–5.86)

A previous meta-analysis [16] had shown that among patients with AMI that had occurred before the age of 55, the FV Leiden mutation was more frequent in individuals without significant coronary stenosis (OR = 3.26; 95% CI = 1.67–6.36). We were unable to confirm this finding (OR = 1.11; 95% CI = 0.66–1.86). However, a trend towards a higher incidence of FV Leiden in patients with AMI and no significant coronary stenosis was observed (minor allele frequency: 0.025 in patients with significant narrowing; 0.027 in patients with no significant narrowing; 0.028 in patients without coronary stenosis).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

A major hindrance to advancing knowledge on the genetic basis of a complex disease such as CAD has been the failure to replicate original findings. This case-control study, carried out in 1880 patients who survived a premature AMI, was designed to replicate in a large cohort of patients the results of a meta-analysis that had shown that two rare gain-of-function variants of coagulation FV and prothrombin were associated with CAD in patients unselected for age and other features [8]. Although the whole sample size of the meta-analysis was impressively large (66 155 cases with AMI or coronary artery stenosis and 91 307 controls), the number of cases and controls in the studies investigating the hemostasis gene variants associated with CAD was much smaller, particularly when AMI cases were considered separately from their combination with cases who had coronary stenosis without AMI. Moreover, meta-analyses are as valid and reliable as the quality of the studies that are analyzed and, as mentioned above, the gene association studies in the field of CAD were usually of poor quality and hardly replicated [4,7].

In this study, we failed to replicate the weak association with the PAI-1 gene. For the G20210A prothrombin variant there was a definite trend for a positive association, with an OR (1.32) consistent with the results of the meta-analysis [8]. The most significant finding is that FV Leiden, which causes blood hypercoagulability through the induction of resistance to the anticoagulant activity of activated protein C on FV, was associated with the premature occurrence of AMI. The allele-related risk was much higher than that found in the meta-analysis for patients with AMI only (1.66 vs. 1.22) and was statistically robust, as shown by a narrow confidence interval (1.15–2.38). These differences from the results of the meta-analysis of Ye et al. [8] have at least two possible explanations. First, the homogeneity of our cohort, made exclusively of patients who developed AMI at a young age; second, the likelihood that a genetic factor such as FV Leiden plays a more prominent role in the young than in the elderly, because in the former factors related to aging do not confound the effect of the genetic component of AMI. These views are consistent with the results of two smaller meta-analyses that, unable to find an overall association between AMI and FV Leiden, found a positive association in subgroups of patients with premature AMI (ORs of 1.48 and 1.54, respectively) [16,17]. The difference between the positive data reported herewith, and those of our original cohort that had found no association of premature AMI with FV Leiden [11], is due to a more accurate genotyping (see Table 1). This stresses the potentially negative consequences of genotyping errors in association studies, especially when dealing with rare variants [15]. Our results on FV Leiden were independently confirmed in a parallel genotyping study carried out in the frame of a quality control exercise, performed before including our DNA samples in a genome-wide association study [18].

The main limitation of our study is its case-control design, which has led to the exclusion of patients who died before reaching the coronary care units. The precise number of these deaths is not available but is likely to be very small, because mortality is low in the young with AMI. The criteria used to rule out AMI in controls (clinical history and the administration of a questionnaire) may have led to their misclassification as healthy in some cases. However, undiagnosed AMI in controls should result in a more conservative estimation of the positive association of FV Leiden with AMI. A strength of this study is the size of the population, which is the largest single cohort investigated so far for these genetic variants. This is witnessed by comparing its size with those of the studies included in the meta-analysis of Ye et al. [8], which included only seven studies on FV Leiden based upon more than 500 cases.

Improved prediction of mechanisms of recurrence of AMI using a combination of phenotypic and genotypic data may help to tailor treatments during this critical period. That hypercoagulability plays a mechanistic role in the recurrence of AMI is unequivocally supported by the demonstration that long-term therapy with vitamin K antagonists is highly efficacious in the secondary prevention of AMI [19,20]. However, this therapy is poorly used by clinicians, because regular laboratory control of coagulation is required and, most importantly, because there is a high risk of bleeding complications, particularly when associated with dual antiplatelet therapy with aspirin and clopidogrel [21,22]. Our results and those of the meta-analysis [8,16,17] indicate that in at least a fraction of patients the occurrence of premature AMI is associated with a hypercoagulable state. The decision to use anticoagulants in patients with gain-of-function gene mutations and hypercoagulability is unlikely to become evidence-based and should be considered only on a case-to-case patient basis.

In summary, this study replicates the association of FV Leiden with the risk of AMI, at least in patients with premature occurrence of the disease. These findings support the important role of hypercoagulability in the pathogenesis of AMI in the young.

Addendum

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

All the authors participated in the conception and design of the present study, in the analysis and interpretation of data, and in revising the manuscript. P.M. Mannucci wrote the manuscript and supervised the entire study. S. Kathiresan and S. Duga were responsible for the study design and for critical reviewing of the paper. M. Spreafico selected the analyzed polymorphisms. I. Guella optimized and performed the genotyping assays. R. Asselta was in charge of the statistical analysis of the data. L. Lotta and F. Peyvandi participated in the manuscript writing and critical reviewing. P.A. Merlini and D. Ardissino enrolled the patients in the study, collected the clinical history, and evaluated the clinical phenotypes.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

This study was supported by ‘Progetto Finalizzato’ from the Italian Health Ministry to P. M. Mannucci and ‘Progetto a Concorso’ to R. Asselta. The financial support of FIRST (Finanziamenti per l’Innovazione, la Ricerca e lo Sviluppo Tecnologico, for years 2005–2007) is also gratefully acknowledged.

Disclosure of Conflict of Interests

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References

The authors state that they have no conflict of interest.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Addendum
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References
  • 1
    Kathiresan S, Yang Q, Larson MG, Camargo AL, Tofler GH, Hirschhorn JN, Gabriel SB, O’Donnell CJ. Common genetic variation in five thrombosis genes and relations to plasma hemostatic protein level and cardiovascular disease risk. Arterioscler Thromb Vasc Biol 2006; 26: 140512.
  • 2
    Zee RY, Cook NR, Cheng S, Erlich HA, Lindpaintner K, Ridker PM. Multi-locus candidate gene polymorphisms and risk of myocardial infarction: a population-based, prospective genetic analysis. J Thromb Haemost 2006; 4: 3418.
  • 3
    Smith NL, Bis JC, Biagiotti S, Rice K, Lumley T, Kooperberg C, Wiggins KL, Heckbert SR, Psaty BM. Variation in 24 hemostatic genes and associations with non-fatal myocardial infarction and ischemic stroke. J Thromb Haemost 2008; 6: 4553.
  • 4
    Morgan TM, Krumholz HM, Lifton RP, Spertus JA. Nonvalidation of reported genetic risk factors for acute coronary syndrome in a large-scale replication study. JAMA 2007; 297: 155161.
  • 5
    Shiffman D, O’Meara ES, Bare LA, Rowland CM, Louie JZ, Arellano AR, Lumley T, Rice K, Iakoubova O, Luke MM, Young BA, Malloy MJ, Kane JP, Ellis SG, Tracy RP, Devlin JJ, Psaty BM. Association of gene variants with incident myocardial infarction in the Cardiovascular Health Study. Arterioscler Thromb Vasc Biol 2008; 28: 1739.
  • 6
    Van Der Krabben MD, Rosendaal FR, Van Der Bom JG, Doggen CJ. Polymorphisms in coagulation factors and the risk of recurrent cardiovascular events in men after a first myocardial infarction. J Thromb Haemost 2008; 6: 7205.
  • 7
    Ioannidis JP, Trikalinos TA, Ntzani EE, Contopoulos-Ioannidis DG. Genetic associations in large versus small studies: an empirical assessment. Lancet 2003; 361: 56771.
  • 8
    Ye Z, Liu EH, Higgins JP, Keavney BD, Lowe GD, Collins R, Danesh J. Seven haemostatic gene polymorphisms in coronary disease: meta-analysis of 66,155 cases and 91,307 controls. Lancet 2006; 367: 6518.
  • 9
    Marenberg ME, Risch N, Berkman LF, Floderus B, De FU. Genetic susceptibility to death from coronary heart disease in a study of twins. N Engl J Med 1994; 330: 10416.
  • 10
    Zimmerman FH, Cameron A, Fisher LD, Ng G. Myocardial infarction in young adults: angiographic characterization, risk factors and prognosis (Coronary Artery Surgery Study Registry). J Am Coll Cardiol 1995; 26: 65461.
  • 11
    No evidence of association between prothrombotic gene polymorphisms and the development of acute myocardial infarction at a young age. Circulation 2003; 107: 111722.
  • 12
    Van Den Bergh FA, Van Oeveren-Dybicz AM, Bon MA. Rapid single-tube genotyping of the factor V Leiden and prothrombin mutations by real-time PCR using dual-color detection. Clin Chem 2000; 46: 11915.
  • 13
    Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MA, Bender D, Maller J, Sklar P, De Bakker PI, Daly MJ, Sham PC. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet 2007; 81: 55975.
  • 14
    Wigginton JE, Cutler DJ, Abecasis GR. A note on exact tests of Hardy–Weinberg equilibrium. Am J Hum Genet 2005; 76: 88793.
  • 15
    Little J, Higgins JP, Ioannidis JP, Moher D, Gagnon F, Von Elm E, Khoury MJ, Cohen B, Davey-Smith G, Grimshaw J, Scheet P, Gwinn M, Williamson RE, Zou GY, Hutchings K, Johnson CY, Tait V, Wiens M, Golding J, Van Duijn C, et al. STrengthening the REporting of Genetic Association studies (STREGA) – an extension of the STROBE statement. Eur J Clin Invest 2009; 39: 24766.
  • 16
    Juul K, Tybjaerg-Hansen A, Steffensen R, Kofoed S, Jensen G, Nordestgaard BG. Factor V Leiden: The Copenhagen City Heart Study and 2 meta-analyses. Blood 2002; 100: 310.
  • 17
    Doix S, Mahrousseh M, Jolak M, Laurent Y, Lorenzini JL, Binquet C, Zeller M, Cottin Y, Wolf JE. Factor V Leiden and myocardial infarction: a case, review of the literature with a meta-analysis. Ann Cardiol Angeiol (Paris) 2003; 52: 1439.
  • 18
    Myocardial Infarction Genetics Consortium. Genome-wide association of early-onset myocardial infarction with single nucleotide polymorphisms and copy number variants. Nat Genet 2009; 41: 33441.
  • 19
    Hurlen M, Abdelnoor M, Smith P, Erikssen J, Arnesen H. Warfarin, aspirin, or both after myocardial infarction. N Engl J Med 2002; 26: 347.
  • 20
    Van Es RF, Jonker JJ, Verheugt FW, Deckers JW, Grobbee DE. Aspirin and coumadin after acute coronary syndromes (the ASPECT-2 study): a randomised controlled trial. Lancet 2002; 360: 10913.
  • 21
    Franchini M, Mannucci PM. A new era for anticoagulants. Eur J Intern Med 2009; 20: 5628.
  • 22
    Franchini M, Mannucci PM. New antiplatelet agents: why they are needed. Eur J Intern Med 2009; 20: 7338.