| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Thrombosis |
From the Department of Internal Medicine, Sections on Cardiology (D.M.H., D.A.M.), Medical Genetics (T.D.H.), and Hematology and Oncology (J.O.), and the Departments of Public Health Sciences (D.M.R.) and Biochemistry (D.B.), Wake Forest University School of Medicine, Winston-Salem, NC; the Department of Epidemiology and Biostatistics (E.V., J.A.S., D.G., S.B.H.), University of California, San Francisco; and the Department of Medicine (V.B.), University of Alabama at Birmingham.
Reprint requests to David M. Herrington, MD, MHS, Department of Internal Medicine/Cardiology, Wake Forest University School of Medicine, Medical Center Boulevard, Winston-Salem, NC 27157-1040. E-mail dherring{at}wfubmc.edu
| Abstract |
|---|
|
|
|---|
Key Words: cardiology risk factors for stroke genetics thrombosis risk factors
| Introduction |
|---|
|
|
|---|
See page 879
To evaluate the association between factor V Leiden and VTE in women with coronary disease and its impact on the relationship between HRT and VTE in this population, we conducted a nested case-control study among participants of 2 multicenter randomized clinical trials of HRT in women with established coronary disease, the Heart and Estrogen/Progestin Replacement Study (HERS) and the Estrogen Replacement and Atherosclerosis (ERA) trial.
| Methods |
|---|
|
|
|---|
Data Collection
At 4- to 6-month follow-up visits, women were queried about blood clots in the legs or lungs or any hospitalizations. A positive reply prompted retrieval of all pertinent medical records to verify the occurrence of a VTE. A diagnosis of DVT required confirmation by venography, impedance plethysmography, or ultrasound. A diagnosis of PE required confirmation by a segmental or larger ventilation/perfusion mismatch on a nuclear lung scan or an intraluminal filling defect by pulmonary angiography. VTEs were classified as idiopathic if they occurred in women without a preexisting diagnosis of cancer or the occurrence of an inpatient hospitalization or lower extremity or hip fracture within 3 months before the VTE. Events were adjudicated by individuals who were unaware of treatment assignment or factor V genotype.
In HERS, 47 (1.7%) of the women had a VTE (32 DVTs and 15 PEs). At the final follow-up visit, cases and 2 age- and clinic-matched controls were asked to provide an additional specimen of whole blood for factor V genotyping. For women who could not return for the final visit or who had died, consent was sought from them or their next of kin to use a previously collected Pap smear for extraction of DNA and genotyping. Consent was obtained, and factor V genotype was ascertained in 40 of 47 cases and 80 matched controls. Three of the remaining 7 cases declined participation, 3 did not have blood or a Pap smear available for analysis, and 1 matched control had a Pap smear that did not yield enough DNA for genotyping. In ERA, 8 (2.6%) of the women had a VTE (6 DVTs and 2 PEs). DNA for genotyping was available for all 8 cases. In view of the smaller number of VTE cases in ERA, 4 age- and clinic-matched controls were selected per case. For both cohorts, the study design and conduct were approved by the Wake Forest University Institutional Review Board.
Genotyping for the presence of the Leiden mutation (G1691A) in the factor V gene was performed on DNA extracted from anonymously labeled peripheral blood leukocytes or cells retrieved from Pap smears (Pap smear slides yielded
200 ng DNA per slide). The genomic region containing the mutation site was amplified by polymerase chain reaction, and the presence of the Leiden mutation was inferred from the observed cleavage pattern with the restriction enzyme Mnl-I.1
Statistical Analyses
Baseline and on-study characteristics of the cases and controls were compared within each study and in the combined data; random effects models24 were used for continuous variables, and conditional logistic regression25 was used for binary variables, to take account of the matched case-control design. The association between factor V genotype and risk of VTE was examined by using conditional logistic regression models, again to take account of the matching and to adjust for treatment assignment. Odds ratios (ORs) generated by these models, accompanied by 95% CIs and nominal 2-tailed probability values, were used to approximate relative risks. In exploratory analyses, results from the 2 study cohorts were similar; therefore, the data from the 2 cohorts were combined. Similarly, preliminary analysis within the ERA cohort found no significant difference between the effects of unopposed estrogen and estrogen plus medroxyprogesterone acetate; thus, data from these 2 active treatment arms were combined. Additional models were also examined with adjustment for potential risk factors for VTE, including a previous diagnosis of cancer (excluding nonmelanoma skin cancer) or myocardial infarction (MI), inpatient hospitalization, or lower extremity or hip fracture within 90 days of the VTE, or use of aspirin or statins documented during the most recent clinic visit before the VTE. In these models, as well as in the initial comparison with cases, on-trial risk factors for controls were evaluated at the time in days since the randomization of VTE onset for the matched case. Tests for additive and multiplicative interaction were performed by using previously described methods.26
| Results |
|---|
|
|
|---|
|
The Leiden mutation was present in 8 (16.7%) of the 48 cases of VTE compared with only 7 (6.3%) of the 112 controls (Table 2). On the basis of these numbers, the estimated prevalence of factor V Leiden in the entire study population was 6.4%. After the age- and clinic-matched case-control design was taken into account and adjustments were made for treatment assignment, the risk of VTE was
3-fold greater in women with factor V Leiden compared with women without the mutation (ORLeiden 3.3, 95% CI 1.1 to 9.8; P=0.03). Similarly, the ORLeiden for a DVT was 3.2 (95% CI 1.0 to 10.3, P=0.05), and for a PE, it was 3.5 (95% CI 0.2 to 69.6, P=0.41). After additional adjustment for recent hospitalization and prior MI, the ORLeiden for VTE increased to 6.8 (95% CI 1.1 to 42.8, P=0.05). The number of cases was too small to support fully adjusted models with stable estimates for DVT or PE separately. Factor V Leiden was present in 5 (16%) of the 31 women with idiopathic VTE and 3 (18%) of the 17 women with nonidiopathic VTE.
|
HRT use was associated with a >3-fold increase in the risk of VTE (OR 3.4, 95% CI 1.4 to 8.1; P=0.06). Adjustment for factor V Leiden status resulted in a slightly higher estimate of risk (ORHRT 3.9, 95% CI 1.6 to 9.6; P=0.004). After additional adjustment for recent hospitalization and prior MI, the risk associated with HRT use remained significant (ORHRT 4.8, 95% CI 1.3 to 17.5; P=0.02). Among women without factor V Leiden, the ORHRT was 4.5 (95% CI 1.2 to 16.9, P=0.02; Table 3). In contrast, among the fewer women with factor V Leiden, the OR for HRT suggested a >10-fold increase in risk compared with placebo; however, the CIs were wide and included unity (ORHRT 10.2, 95% CI 0.3 to 344; P=0.20). Compared with women without factor V Leiden taking placebo, women with factor V Leiden taking HRT had a 14-fold greater risk of VTE (ORHRT 14.1, 95% CI 2.7 to 72.4; P=0.002). Despite the suggestion of excess risk of HRT in women with factor V Leiden, formal tests for additive or multiplicative interaction were negative.
|
Overall, 1.56% of the combined study cohort (n=3072) suffered a VTE. If it is assumed that the prevalence of factor V Leiden in the sample of controls was representative of all noncases, 2.08% of women with factor V Leiden had a VTE versus 1.49% of the women without factor V Leiden. If the variable length of follow-up among women in the 2 studies is taken into account, the overall incidence of VTE was 7.1 per 1000 women per year. In women without factor V Leiden who were assigned to placebo, the incidence of VTE was 2.0 (95% CI 0.8 to 3.2) events per 1000 women per year (Table 4). Treatment with HRT increased the estimated incidence to 5.8 events per 1000 women per year (95% CI 4.5 to 7.1). In factor V Leiden women treated with HRT, the incidence was further increased to 15.4 (95% CI 3.5 to 27.3) events per 1000 women per year. On the basis of these estimates, if women with coronary disease were screened for factor V Leiden before initiating HRT and if HRT was withheld from carriers of the mutation, the estimated number needed to screen to prevent 1 VTE during 5 years of therapy would be 376.
|
| Discussion |
|---|
|
|
|---|
0.5% per year) was much higher than that in healthier cohorts. Use of HRT in factor V Leidenpositive women further increased the risk to >1.5% per year. On the basis of these data, in women with coronary disease, 1 DVT might be prevented during a planned 5 years of treatment by withholding HRT from the expected 24 Leiden-positive women among 376 screened. For women who would use HRT for 10 years, the number needed to screen would be 188. The results of the present study are consistent with the results of previous studies demonstrating a multiplicative relationship between factor V Leiden or APC resistance and exogenous estrogen use and the risk of VTE.12,16 In a case-control study of idiopathic VTEs, Lowe et al16 observed a 13-fold increased risk in HRT users with APC resistance compared with women without APC resistance or HRT use; this risk was very similar to the 15-fold excess risk observed in the doubly exposed women in the present study.
The precise mechanism for the association between estrogen exposure and risk for VTE remains uncertain. Some,2932 but not all, studies33 indicate that oral contraceptive use or pregnancy may result in an acquired form of APC resistance, similar to the functional defect associated with the factor V Leiden mutation. However, Douketis et al34 failed to find evidence of an effect of oral HRT on APC resistance. Currently available data are also inconsistent regarding the effects of HRT on thrombin generation or fibrinolytic potential.34
There are several limitations of the present study. The number of cases is small, thereby limiting precision in the estimates of the effects of factor V Leiden and the complexity of the models that could be used to adjust for other covariates. No information is available about concurrent thrombophilic states in the study participants. Some evidence suggests that factor V Leiden may be most important as a VTE risk factor in the presence of other thrombophilic conditions. Protein C deficiency,35 protein S deficiency,36 hyperhomocysteinemia,37 and the prothrombin G20210A genotype38 have all been reported to further increase the risk of VTE associated with factor V Leiden. The effects of HRT on the risk of VTE in women with or without heart disease who also have such thrombophilic conditions are not yet established. The estimates of the risk of VTE in the clinical trial participants reported in the present study may underestimate the risks in the general population, because women with a prior history of VTE or women with hypertriglyceridemia or advanced heart failure were excluded from participation. In addition, after it became clear that HRT was associated with an increased VTE risk in HERS, the protocol was changed so that HRT was withheld during acute hospitalizations, which may have further minimized the overall risk in this study population. Finally, the results of the present study may not apply to women of other ethnic backgrounds. The prevalence of the Leiden mutation is highest in northern Europeans and is almost nonexistent in most Asian populations.27
It is also important to emphasize that the observations in the present study are limited to women with established coronary disease. The joint effects of HRT and factor V Leiden in healthy women are not yet determined. In younger women with factor V Leiden, oral contraceptives increase risk by
0.2 per 1000 women per year.39 In the present study, in women with coronary disease and factor V Leiden, the increase in absolute risk associated with HRT use was >40-fold greater (8.3 per 1000 women per year). It is difficult to imagine many clinical settings in which the benefits of postmenopausal HRT would outweigh this increase in risk in women with existing coronary heart disease (CHD). This is especially true in light of the evidence that HRT is ineffective at slowing clinical22,40 or anatomic21,41 manifestations of atherosclerosis and may even be associated with an early increase in CHD events.22,4244 Unfortunately, despite the higher absolute risk of VTE in women with CHD, the prevailing costs of $150 to $600 for a single clinical factor V Leiden determination makes screening impractical. Even if the low end of this range of cost is assumed,
$50 000 would be spent on screening to prevent 1 VTE hospitalization. If clinical genotyping for factor V Leiden becomes less expensive, screening could become a cost-effective strategy in women with heart disease.45
In summary, in women with coronary disease, factor V Leiden and HRT use are independently associated with an increase in VTE risk. In women who have both risk factors, the absolute risk is
1.5% per year. If less expensive methods for factor V genotyping become more widely available, screening for factor V Leiden before initiating HRT might become a cost-effective strategy in women with coronary disease. However, even in women with coronary disease without factor V Leiden, the HRT-associated VTE risk must be carefully considered when HRT is used for approved noncoronary indications.
| Acknowledgments |
|---|
Received February 19, 2002; accepted April 2, 2002.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
V. M. Miller and S. P. Duckles Vascular Actions of Estrogens: Functional Implications Pharmacol. Rev., June 1, 2008; 60(2): 210 - 241. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. D. Bushnell, P. Hurn, C. Colton, V. M. Miller, G. del Zoppo, M. S.V. Elkind, B. Stern, D. Herrington, G. Ford-Lynch, P. Gorelick, et al. Advancing the Study of Stroke in Women: Summary and Recommendations for Future Research From an NINDS-Sponsored Multidisciplinary Working Group Stroke, September 1, 2006; 37(9): 2387 - 2399. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. D. Curb, R. L. Prentice, P. F. Bray, R. D. Langer, L. Van Horn, V. M. Barnabei, M. J. Bloch, M. G. Cyr, M. Gass, L. Lepine, et al. Venous thrombosis and conjugated equine estrogen in women without a uterus. Arch Intern Med, April 10, 2006; 166(7): 772 - 780. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. K. Koh and B.-K. Yoon Controversies regarding hormone therapy: Insights from inflammation and hemostasis Cardiovasc Res, April 1, 2006; 70(1): 22 - 30. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Straczek, E. Oger, M. B. Yon de Jonage-Canonico, G. Plu-Bureau, J. Conard, G. Meyer, M. Alhenc-Gelas, H. Levesque, N. Trillot, M.-T. Barrellier, et al. Prothrombotic Mutations, Hormone Therapy, and Venous Thromboembolism Among Postmenopausal Women: Impact of the Route of Estrogen Administration Circulation, November 29, 2005; 112(22): 3495 - 3500. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. K. Dubey, B. Imthurn, M. Barton, and E. K. Jackson Vascular consequences of menopause and hormone therapy: Importance of timing of treatment and type of estrogen Cardiovasc Res, May 1, 2005; 66(2): 295 - 306. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. R. Rosendaal Venous Thrombosis: The Role of Genes, Environment, and Behavior Hematology, January 1, 2005; 2005(1): 1 - 12. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. P. V. Gomes and S. R. Deitcher Risk of Venous Thromboembolic Disease Associated With Hormonal Contraceptives and Hormone Replacement Therapy: A Clinical Review Arch Intern Med, October 11, 2004; 164(18): 1965 - 1976. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Cushman, L. H. Kuller, R. Prentice, R. J. Rodabough, B. M. Psaty, R. S. Stafford, S. Sidney, F. R. Rosendaal, and for the Women's Health Initiative Investigators Estrogen Plus Progestin and Risk of Venous Thrombosis JAMA, October 6, 2004; 292(13): 1573 - 1580. [Abstract] [Full Text] [PDF] |
||||
![]() |
Venous Thromboembolic Disease Obstet. Gynecol., October 1, 2004; 104(4_suppl): 118S - 127S. [Full Text] [PDF] |
||||
![]() |
K. K. Koh and I. Sakuma Should Progestins Be Blamed for the Failure of Hormone Replacement Therapy to Reduce Cardiovascular Events in Randomized Controlled Trials? Arterioscler. Thromb. Vasc. Biol., July 1, 2004; 24(7): 1171 - 1179. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. E. Humphries, P. M. Ridker, and P. J. Talmud Genetic Testing for Cardiovascular Disease Susceptibility: A Useful Clinical Management Tool or Possible Misinformation? Arterioscler. Thromb. Vasc. Biol., April 1, 2004; 24(4): 628 - 636. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Iemolo, A. Martiniuk, D. A. Steinman, and J. D. Spence Sex Differences in Carotid Plaque and Stenosis Stroke, February 1, 2004; 35(2): 477 - 481. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Duggan, K. Marriott, R. Edwards, and J. Cuzick Inherited and Acquired Risk Factors for Venous Thromboembolic Disease Among Women Taking Tamoxifen to Prevent Breast Cancer J. Clin. Oncol., October 1, 2003; 21(19): 3588 - 3593. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. P. Reiner, S. R. Heckbert, H. L. Vos, R. A. S. Ariens, R. N. Lemaitre, N. L. Smith, T. Lumley, T. D. Rea, L. A. Hindorff, G. D. Schellenbaum, et al. Genetic variants of coagulation factor XIII, postmenopausal estrogen therapy, and risk of nonfatal myocardial infarction Blood, July 1, 2003; 102(1): 25 - 30. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. L. Ornstein and M. Cushman Factor V Leiden Circulation, April 22, 2003; 107 (15): e94 - e97. [Full Text] [PDF] |
||||
![]() |
S. R Deitcher and M. P. Gomes Hypercoagulable state testing and malignancy screening following venous thromboembolic events Vascular Medicine, February 1, 2003; 8(1): 33 - 46. [Abstract] [PDF] |
||||
![]() |
D. M. Herrington Hormone Replacement Therapy and Heart Disease: Replacing Dogma With Data Circulation, January 7, 2003; 107(1): 2 - 4. [Full Text] [PDF] |
||||
![]() |
L. H. Kuller Hormone Replacement Therapy and Risk of Cardiovascular Disease: Implications of the Results of the Women's Health Initiative Arterioscler. Thromb. Vasc. Biol., January 1, 2003; 23(1): 11 - 16. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. A. Bauer Hormone Replacement Therapy and the Factor V Leiden Mutation Arterioscler. Thromb. Vasc. Biol., June 1, 2002; 22(6): 879 - 880. [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||