Factor VII gene polymorphisms are not associated with myocardial infarction in young women


Pier M Mannucci, Angelo Bianchi Bonomi Hemophilia and Thrombosis Center, University of Milan, Via Pace 9, 20122, Milan Italy. Tel.: +39-02-55035421; fax: +39-02-50320723; e-mail: piermannuccio.mannucci@unimi.it, pmmannucci@libero.it

Several studies have evaluated whether or not there is an association between atherothrombotic disease and polymorphic variants of genes encoding proteins involved in that hemostatic system [1,2]. One of the most studied genes is the encoding coagulation factor VII (FVII). FVII polymorphisms were previously found to be associated with increased or decreased levels of FVII in plasma (reviewed by Lane and Grant) [3]. Some [4,5], but not all [6–8], retrospective case–control studies have shown that carriers of polymorphic alleles associated with lower FVII levels (353Q, the H7H7 VNTR in intron 7 and a 10-base-pair insertion at position −323) have a decreased risk of acute myocardial infarction (AMI). On the other hand, Carew et al. [9] showed that carriers of a FVII promoter haplotype associated with higher plasma levels of FVII (−670C, −630G and −402 A) have an increased risk for a first coronary event, particularly AMI. In this study, we chose to investigate the relationship between FVII polymorphisms and AMI by investigating women who developed AMI at a young age (less than 45 years of age). These rare patients usually have less traditional risk factors and coronary artery stenosis than older women and men with ischemic heart disease, so that it is biologically plausible that in them hypercoagulability plays an important role in the pathogenesis of thrombus formation and AMI. Moreover, genetic effects are usually more prominent in the young because cumulative environmental factors have not yet had time to modify phenotypes.

With this as background, we studied the effect of seven FVII polymorphisms (− 670A/C, − 630 A/G, − 402G/A, − 401G/T, 10bpIns, the intron 7 VNTR and R353Q) on the risk of premature AMI in the frame of a case–control study, that included 98 women who had survived an AMI before the age of 45 years and had been admitted between January 1998 and January 2001 to coronary care units in Italy. They are a subset of a larger study carried out to assess the association of hemostasis gene polymorphisms with premature AMI in men and women [10]. Venous blood was obtained during a period ranging between a minimum of 3 months and a maximum of 12 months after AMI. The criteria used to diagnose AMI, to define the presence of significant or non-significant stenosis on coronary angiography and the presence or absence of classical risk factors for AMI were previously reported [10,11]. Ninety-eight healthy women, enrolled from the staff of the same participating hospitals, were invited to act as controls and were matched with cases for age and geographic area of origin. For cases, data was collected from medical records at the time of the subjects’ AMI; for controls, data was collected at the time of evaluation prior to their enrollment.

Citrated plasma (0.129 mol L−1) and buffy coats were prepared by centrifugation at 2500 × g for 20 min, immediately snap frozen and stored at − 70 °C. Plasma FVII antigen (FVII:Ag) was measured by enzyme immunoassay (Stago Laboratories, Asnieres, France). DNA was extracted from whole blood using a standard salting out method. Analysis for intron 7 VNTR, 10bpIns and R353Q was carried out using previously described methods [4,5]. Genotype analyses for −401G/T, −402G/A, −630A/G and −670A/C were performed using a single fragment of polymerase chain reaction (PCR) amplified by primers 5′-AGGTGGAGGTTGCAGTGAGATTGCAC-3′ and 5′-GTGGCACCAAAACACTTC AAATACG-3′, followed by sequencing on the ABI PRISM™ 310 Genetic Analyzer (Applied Biosystems, Foster City, CA, USA) using the same primers for the amplification reactions. The effect of polymorphisms on FVII:Ag plasma levels was evaluated by linear regression. An analysis of the association of polymorphisms with AMI was performed by means of logistic regression and the relative risks were expressed as odds ratios (OR) with corresponding 95% confidence intervals (CI). For each polymorphism we tested the hypothesis of a multiplicative effect on the risk of AMI, the risk of the heterozygote genotype being half way between the risk of the wild-type and mutant homozygotes. If the hypothesis of a multiplicative effect was not rejected we estimated the OR under the multiplicative model. Otherwise, the OR for the heterozygote and homozygote genotypes were compared with that of the homozygote wild-type, which was taken as reference. Pair-wise linkage disequilibrium between the studied polymorphisms was first evaluated using the D′ measure. Since −402G/A and −630A/G were in complete linkage disequilibrium (D′ = 1) with −670A/C, and 10bpIns was in complete LD with −401G/T, we chose to exclude −402G/A, −630A/G and the 10bpIns from the analysis. The distribution of genotypes for the polymorphisms was in Hardy–Weinberg equilibrium in controls.

As previously reported by others [3,9], the −670C polymorphic allele was significantly associated with higher FVII:Ag levels in plasma, whereas the −401T and 353Q alleles were associated with lower FVII:Ag levels (data not shown). Table 1 shows the unadjusted OR and 95% CI for each of the polymorphisms and for the traditional risk factors, as obtained by fitting a logistic conditional regression. For the −401G/T and −670A/C polymorphisms, a multiplicative model was fitted. Thus, the OR reported for these polymorphisms represent the OR of the heterozygote genotype compared with the wild-type homozygote and the OR of the mutant homozygote genotype compared with the heterozygote. None of the polymorphisms was associated with AMI at the nominal 5% significance level. Table 1 also shows that classical risk factors were associated with AMI in these women, whereas FVII:Ag levels were not.

Table 1.   The effect of FVII polymorphisms and of classical risk factors in 98 young women with acute myocardial infarction and their matched controls
VariableOdds ratio95% CI
−401G/T polymorphism1.610.89–2.90
−670A/C polymorphism1.280.69–2.37
R353Q polymorphism1.530.80–2.94
Intron 7 VNTR:
H6H61 (ref)
H6H6 + H5H7 + H6H71.530.83–2.82
Factor VII antigen0.990.99–1.01
Estrogen use2.211.18–4.16
Body mass index1.081.01–1.17
Physical exercise0.500.26–0.95
Total cholesterol1.011.00–1.01

In conclusion, none of the FVII polymorphisms were associated with the development of AMI in young premenopausal women, in spite of the fact that these polymorphisms were evaluated in a population of patients with AMI characterized by a high likelihood of a pivotal mechanistic role for hypercoagulability and heightened thrombus formation.