Absence of Association Between Polymorphisms in the Hemostatic Factor Pathway Genes and Carotid Intimal Medial Thickness
The Framingham Heart Study
Background and Purpose— Fibrinogen, plasminogen activator inhibitor-1, and other key proteins in the coagulation cascade have been implicated in the origin of cardiovascular disease. Polymorphisms in genes encoding these proteins have been associated with variability in plasma levels of these proteins. Carotid intimal medial thickness (IMT) is a heritable, quantitative measure of atherosclerosis that is predictive of subsequent myocardial infarction and stroke. We sought to test whether carotid IMT is associated with polymorphisms in several well-characterized genes in the hemostatic factor pathways.
Methods— Here, 867 men and 911 women (mean age, 57 years) in the Framingham offspring cohort underwent B-mode carotid ultrasonography to determine the mean internal (ICA) and common carotid artery (CCA) IMT. Age-, sex-, and multivariable-adjusted linear regression was used to estimate the association of the following variants with log-transformed CCA and ICA IMT: factor V Leiden, factor VII Arg/Gln, fibrinogen HindIII β-148, plasminogen activator inhibitor-1 4G/5G, and the glycoprotein IIIa PlA2 polymorphism.
Results— Mean ICA IMT was 0.58 mm; mean CCA IMT was 0.60 mm. There were no differences in ICA or CCA IMT by genotype for any of the candidate genes in unadjusted, age- or sex-adjusted, and multivariable-adjusted models.
Conclusions— There is no evidence for an association between well-studied polymorphisms in the hemostatic factor genes and carotid IMT. Whether other common genetic variants in hemostatic factor genes are associated with subclinical atherosclerosis remains to be determined.
Carotid intimal medial thickness (IMT) is a subclinical measure of atherosclerosis that is predictive of subsequent myocardial infarction and stroke.1 Levels of a number of hemostatic factors, including factor VII,2 plasminogen activator inhibitor-1,3 and fibrinogen,4 have been shown to be associated with carotid IMT. However, it is not known whether hemostatic gene variants are associated with subclinical atherosclerosis determined by carotid IMT. We hypothesized that hemostatic factor gene variants may be associated with carotid IMT, and we tested variants in factor V, factor VII, fibrinogen, plasminogen activator inhibitor-1, and glycoprotein IIIa in the Framingham Heart Study offspring sample.
The Framingham Heart Study began in 1948 with an enrollment of 5209 men and women who were 28 to 62 years of age at study entry, with subjects undergoing repeated examinations every 2 years.5,6 In 1971, 5124 men and women were enrolled in the offspring cohort of the Framingham Heart Study, which included the children or spouses of the children of the original cohort. Offspring subjects underwent examinations approximately every 4 years; the design and methodology of the Framingham Offspring Cohort have been described previously.7,8 The current investigation comprises subjects from the Framingham Offspring Study undergoing B-mode carotid ultrasonography during examination cycle 6 (1996 through 1998). In total, 3532 participants attended this examination cycle. Of those, 3407 underwent carotid ultrasonography. Of these, 1778 subjects underwent genotyping for the hemostatic factor candidate gene polymorphisms. The cohort included 284 extended families ranging from 2 to 21 members. There were 1191 sibpairs, with sibships ranging from 2 to 7.
Each examination included an extensive cardiovascular disease assessment, 12-lead ECG, and blood testing. Measured covariates for the present study were assessed at the time of carotid ultrasonography.
Subjects underwent ultrasonography according to a standard protocol.1 Imaging was conducted by a single trained sonographer using a Toshiba SSH-140A imaging unit with a high-resolution 7.5-MHz transducer for the common carotid artery (CCA) and a 5.0-MHz transducer for the internal carotid artery (ICA).
Measurements were made by a single trained sonographer blinded to all clinical information and overread by 1 of the investigators (J.F.P.). Based on 25 replicate readings by 2 separate readers, correlation coefficients for the mean and maximum ICA were 0.83 and 0.84, respectively.
A number of previously described polymorphisms in several of the genes coding for hemostatic factors were characterized using previously published assay conditions. The factor V Leiden mutation (1691G→A) was genotyped according to standard protocol. Briefly, factor V Leiden results from a G-to-A substitution at nucleotide position 1691. A polymerase chain reaction (PCR)–based restriction fragment length polymorphism analysis was used, and genomic DNA was isolated from whole blood. The sequences of sense and antisense primers were 5′tgcccagtgcttaaacaagacca3′ and 5′tgttatcacactggtgctaa3′, respectively. The 4G/5G polymorphism of PAI-1 (1-BP DEL/INS 4G/5G) arises from an insertion/deletion of a guanidine (4 or 5 repeats) at position 675 in the PAI-1 promoter. We used a PCR-based analysis to detect the polymorphism. We amplified the region around the 4G/5G polymorphism. The sequences of the sense primer and antisense primer were 5′ggggcacagagagagtctggacac3′ and 5′cggccgcctccgatgatac3′, respectively. DNA was amplified with PCR. Samples were size fractionated on a polyacrylamide sequencing gel, and the PCR results were scored in a blinded fashion. Genotyping methods have been described previously for the fibrinogen HindIII β-148 polymorphism (−148C→T),9 glycoprotein IIIa PlA2 polymorphism (1565T→C),10 and factor VII (10828G→A) polymorphism.11 The frequency of genotyping failures ranged from 2.3% (glycoprotein IIIa PlA2 polymorphism) to 2.9% (plasminogen activator inhibitor-1 4G/5G).
Descriptive statistics using means, medians, and standard deviations were performed on all variables when appropriate. The χ2 test was used to compare observed genotype frequencies with their estimates under Hardy-Weinberg equilibrium. To reduce nonnormality of distributions, CCA and ICA IMTs were log transformed. Multivariate linear regression, taking into account correlations among family members, was performed with SAS PROC MIXED.12 The general model was run first, and those with significant tests of trend were fit for additive, recessive, and dominant models. To accommodate multiple testing with 5 genes, we used a value of P<0.01 as a Bonferroni-adjusted test of significance. Multivariable analyses included adjustment for age, sex, systolic blood pressure, hypertension treatment, tobacco use (yes/no), total and high-density lipoprotein cholesterol, log triglycerides, diabetes, body mass index, and prevalent cardiovascular disease.
We calculated power to detect genetic effects in the context of a general model that uses 2 parameters to represent differences in genotype-specific means (ie, means μ1, μ2, and μ3). Assuming that measured covariates explain 20% of the overall variance (sex and age accounted for 17% of variance for ICA and 29% for CCA IMT), with a sample size of 1750 and 1% significance level, 80% power is attained for any genetic polymorphism that contributes at least 0.63% of the variance. For example, consider a diallelic gene with allele frequencies of 0.90 and 0.10 (genotype frequencies, 0.81, 0.18, 0.01); if genotype-specific means differed as 0.0, 0.19, and 0.38 SD units, the polymorphism would contribute 0.65% of total variance, and expected power would exceed 80%.
The mean and SD of the ICA IMT were 0.58 and 0.38 mm, respectively. The 25th and 75th percentile values were 0.36 and 0.63 mm, respectively. The mean and SD of the CCA IMT were 0.60 and 0.15 mm, respectively; corresponding 25th and 75th percentile values were 0.51 and 0.65 mm.
Using 1 randomly selected sibling per nuclear family, we estimated allele frequencies and tested Hardy-Weinberg equilibrium for each gene. Allele frequencies for factor V Leiden, factor VII Arg/Gln, fibrinogen HindIII β-148, plasminogen activator inhibitor-1 4G/5G, and glycoprotein IIIa PlA2 polymorphisms were 0.02, 0.15, 0.19, 0.47, and 0.15, respectively, for the rarer allele. Each of the polymorphisms was in Hardy-Weinberg equilibrium (all P>0.15).
In unadjusted and age- plus sex-adjusted analyses, there were no differences in CCA IMT or ICA IMT across the 3 genotypes for any of the allelic variants of hemostatic factor pathway genes. In further age-, sex-, and multivariable-adjusted models, there continued to be no significant differences across genotypes.
The present study failed to detect any statistically significant associations of carotid IMT and several well-characterized polymorphisms in genes coding for hemostatic factors. Our findings render no evidence that similar associations exist with interindividual variability of subclinical carotid atherosclerosis.
Our results are consistent with those of several other studies. In studies that have examined the association between hemostatic factor genes and carotid IMT,13–15 there was no association between carotid IMT and the platelet glycoprotein IIb/IIIa PlA2 polymorphism,13 factor VII Arg/Gln polymorphism,15 or Arg/Gln polymorphism of the factor V Leiden gene.14
Our study has advantages over prior studies. We had a large sample size and an unselected sample of subjects. Furthermore, our study does not exclude the possibility that other genetic variants in these genes or other genes coding for additional elements of the hemostatic cascade may be associated with carotid IMT.
Although the present investigation failed to show an association between the polymorphisms examined and carotid IMT, it may still be of interest to explore more fully the possible association of carotid IMT with genetic variance by studying additional genetic variants in genes coding for hemostatic factors. In addition, it may be of interest to use other modalities to assess subclinical atherosclerosis such as electron beam CT and cardiac MRI that may add complementary information on IMT phenotype assessment.
This work was supported by the National Heart, Lung and Blood Institute’s Framingham Heart Study (N01-HC-25195), National Institute of Neurological Disorders and Stroke (NIH/NINDS 5R01-NS17950-20), and the Framingham Heart Study Visiting Scientist Program, which is supported by Servier Amerique.
- Received October 23, 2003.
- Accepted November 19, 2003.
Cortellaro M, Baldassarre D, Cofrancesco E, et al. Relation between hemostatic variables and increase of common carotid intima-media thickness in patients with peripheral arterial disease. Stroke. 1996; 27: 450–454.
Kannel WB, Feinleib M, McNamara PM, Garrison RJ, Castelli WP. An investigation of coronary heart disease in families: the Framingham Offspring Study. Am J Epidemiol. 1979; 110: 281–290.
O’Donnell CJ, Larson MG, Feng D, et al. Genetic and environmental contributions to platelet aggregation: the Framingham Heart Study. Circulation. 2001; 103: 3051–3056.
Feng D, Lindpaintner K, Larson MG, et al. Increased platelet aggregability associated with platelet GPIIIa PlA2 polymorphism: the Framingham Offspring Study. Arterioscler Thromb Vasc Biol. 1999; 19: 1142–1147.
Feng D, Tofler GH, Larson MG, et al. Factor VII gene polymorphism, factor VII levels, and prevalent cardiovascular disease: the Framingham Heart Study. Arterioscler Thromb Vasc Biol. 2000; 20: 593–600.
SAS Institute Inc. SAS/STAT User’s Guide, Version 8. Cary, NC: SAS Institute Inc; 2000.
Ghaddar HM, Folsom AR, Aleksic N, et al. Correlation of factor VIIa values with factor VII gene polymorphism, fasting and postprandial triglyceride levels, and subclinical carotid atherosclerosis. Circulation. 1998; 98: 2815–2821.