Effects of Random Allocation to Vitamin E Supplementation on the Occurrence of Venous Thromboembolism
Report From the Women’s Health Study
Background— Supplementation with vitamin E may antagonize vitamin K in healthy adults, but it is unclear whether intake of vitamin E decreases the risk of venous thromboembolism (VTE).
Methods and Results— The Women’s Health Study randomized 39 876 women ≥45 years of age to receive 600 IU of natural source vitamin E or placebo on alternate days. Before randomization, 26 779 participants gave blood samples, which were used to determine factor V Leiden, G20210A prothrombin, and 677C>T MTHFR polymorphisms. Documented VTE (including deep vein thrombosis or pulmonary embolism) and unprovoked VTE (no recent surgery, trauma, or cancer diagnosis) were prospectively evaluated, secondary end points of the trial. During a median follow-up period of 10.2 years, VTE occurred in 482 women: 213 in the vitamin E group and 269 in the placebo group, a significant 21% hazard reduction (relative hazard, 0.79; 95% CI, 0.66 to 0.94; P=0.010). For unprovoked VTE, the hazard reduction was 27% (relative hazard, 0.73; 95% CI, 0.57 to 0.94; P=0.016). In subgroup analyses, the 3% of participants who reported VTE before randomization had a 44% hazard reduction (relative hazard, 0.56; 95% CI, 0.31 to 1.00; P=0.048), whereas women without prior VTE had an 18% hazard reduction (relative hazard 0.82; 95% CI, 0.68 to 0.99; P=0.040). Women with either factor V Leiden or the prothrombin mutation had a 49% hazard reduction associated with vitamin E treatment (relative hazard, 0.51; 95% CI, 0.30 to 0.87; P=0.014).
Conclusions— These data suggest that supplementation with vitamin E may reduce the risk of VTE in women, and those with a prior history or genetic predisposition may particularly benefit.
Received May 19, 2007; accepted July 30, 2007.
Venous thromboembolism (VTE) is a common and clinically serious event, with an age-related incidence that increases from ≈1 case per 1000 person-years at age 50 years to ≈5 cases per 1000 person-years at age 75 years.1 Risk factors for venous thromboembolism include a prior event, surgery, trauma, immobilization, prothrombotic mutations, older age, greater body mass index, and hormone therapy.2–5 However, VTE commonly occurs in people without identifiable risk factors, whereas others with multiple factors remain event free.
Clinical Perspective p 1503
Both basic research and observational epidemiological studies support the hypothesis that vitamin E can inhibit oxidative injury and reduce the risk of cardiovascular disease and cancer. However, large-scale randomized trials have not found evidence that vitamin E supplementation prevents cancer or major cardiovascular events.6–9 Nonetheless, vitamin E supplementation might influence risk of VTE through alternative pathways. Specifically, animal studies indicate that vitamin E supplementation has an anticoagulant effect in the presence of low vitamin K intake.10,11 Although vitamin E intake does not alter coagulation times in humans, experimental evidence indicates that vitamin E supplementation may inhibit vitamin K and hence vitamin K–dependent clotting factors.12 Vitamin E has also been shown to inhibit platelet adherence.13 The relevance of these relationships for the risk of VTE is unclear.
The Women’s Health Study tested whether vitamin E supplementation for 10 years decreased the risk of cardiovascular disease or cancer in a large group of women without these diseases at entry. We report here the effect of randomized allocation to vitamin E or placebo on the occurrence of VTE, a prospectively evaluated, secondary end point of this trial.
The Women’s Health Study was a randomized, double-blind, placebo-controlled, 2×2 factorial trial evaluating the balance of risks and benefits of low-dose aspirin (100 mg every other day; Bayer Healthcare) and vitamin E (600 IU of α-tocopherol every other day; Natural Source Vitamin E Association) in the primary prevention of cardiovascular disease and cancer.9,14,15 Originally, a third component, beta carotene, was also included. However, this component was terminated early in January 1996 after a median treatment duration of 2.1 years.16 Written informed consent was obtained from all participating women. The trial was approved by the institutional review board of Brigham and Women’s Hospital and monitored by an external data and safety monitoring board.
Details about the study design and relationships of vitamin E treatment with cancer and cardiovascular disease have been described previously.9 Between September 1992 and May 1995, letters of invitation to participate in the trial and baseline health questionnaires were mailed to >1.7 million female healthcare professionals throughout the United States; 453 787 women completed the questionnaires, and 65 169 were willing and eligible to participate. Eligibility criteria included the following: age ≥45 years; no previous history of coronary heart disease, cerebrovascular disease, cancer (except nonmelanoma skin cancer), and other major chronic diseases including dementia, chronic kidney or liver disease, gout, or gastrointestinal bleeding; and no use of individual supplements of vitamin A, E, or beta carotene more than once a week. Before randomization, women were asked whether they ever had deep vein thrombosis or pulmonary embolism, but those with a positive history were not excluded. However, women who currently used anticoagulants were excluded.
Eligible women were enrolled in a 3-month run-in period of placebo administration to identify a group likely to be compliant with long-term treatment; 39 876 women were willing, eligible, and compliant during the run-in period, and they were randomized to vitamin E (n=19 937) or placebo (n=19 939) and separately to aspirin or matching placebo.
During the run-in period, women were asked to provide a blood sample, and DNA was extracted from 26 779 randomized participants who provided samples. Genotyping for factor V Leiden, the G20210A prothrombin mutation, and the 677C>T polymorphism in methylenetetrahydrofolate reductase (MTHFR) was performed with the use of multiplex polymerase chain reaction and linear immobilized probe array assays as previously described (Roche Molecular Systems, Alameda, Calif).17,18 Linear array processing was facilitated by the use of AutoRELI-Mark II (Dynal Biotech, UK). To confirm genotype assignment, scoring was performed by 2 independent observers, and discordant results (<1% of all scoring) were resolved by a joint reading and, when necessary, repeat genotyping.
Treatment and Follow-Up
Each year, women received an annual supply of calendar packs that contained active agents or placebo. Every 6 months for the first year and annually thereafter, they also received follow-up questionnaires that inquired about compliance with pill-taking, potential adverse effects, occurrence of end points, and risk factors. Treatment and follow-up continued in a blinded fashion until the scheduled end of the trial (March 31, 2004). At the end of blinded treatment, mortality status was known for 99.4% of participants, and questionnaires reporting morbidity were received from 97.2% of surviving participants.
On the basis of self-reported compliance from follow-up questionnaires, 78.9% of participants reported taking at least two thirds of their vitamin E or matching placebo capsules at 5 years, and 71.6% reported this level of compliance at 10 years. Averaged throughout the trial, 75.8% of women reported this level of compliance, with no difference between active and placebo groups (P=0.64). Use of vitamin E supplements outside the trial for at least 4 days per month (“drop-ins”) was reported by 10.0% of participants at 5 years and 10.9% of participants at 10 years. Such outside use of vitamin E supplements throughout the trial averaged 8.6% of women in the active group and 8.9% of women in the placebo group (P=0.07).
End Point Definition
On each follow-up questionnaire, women were asked separately about the new occurrence of deep vein thrombosis and pulmonary embolism. Those reporting events, including next-of-kin of decedents, were asked for permission to obtain medical records. An end points committee of physicians reviewed records in a blinded fashion. Diagnosis of deep vein thrombosis was confirmed by a positive report of venous ultrasound or venography, and diagnosis of pulmonary embolism was considered confirmed in the presence of a positive angiogram or computed tomography scan of the chest or a ventilation-perfusion scan with ≥2 mismatched defects. Deaths due to pulmonary embolism were confirmed when autopsy reports, symptoms, circumstances of death, and medical history were consistent with this diagnosis. Only events confirmed by the end points committee were included.
Unprovoked deep vein thrombosis or pulmonary embolism was defined as occurring in the absence of known malignancy (diagnosed either before or up to 3 months after the VTE), trauma, or surgery within 3 months before the VTE. Provoked VTE included events that occurred in patients with cancer or during or shortly after trauma or surgery.
Primary analyses compared the occurrence of VTE between treatment groups in all randomized subjects classified according to the intention-to-treat principle. The primary end point for this analysis was the first occurrence of VTE after randomization, including women both with and without prior VTE reported at baseline. A second event after randomization was not considered, although women who had both pulmonary embolism and deep vein thrombosis within 3 days at the time of their index event were counted toward each of these specific outcomes. Secondary end points included unprovoked VTE, pulmonary embolism as the first event, and deep vein thrombosis as the first event.
Subgroup analyses considered the effects of vitamin E on the incidence of VTE within categories of risk factors for VTE. Risk factors evaluated were age, body mass index, menopausal status and use of hormone therapy, prior history of VTE, factor V Leiden, the prothrombin mutation, and the MTHFR 677C>T polymorphism.19
The approach of Kaplan and Meier was used to estimate the overall cumulative incidence of VTE by treatment group, as well as the cumulative incidence of unprovoked VTE and the cumulative incidence among women with reported VTE at baseline and those with prothrombotic mutations. The log-rank test was used for crude comparisons of incidence rates. Estimates of the relative hazard of VTE by vitamin E assignment and associated 95% CIs were based on proportional hazards models that controlled for age and other randomized treatments and stratified on reported VTE at baseline.
Subgroup analyses used proportional hazards models stratified on prerandomization VTE and adjusted for age and other randomized treatments. Heterogeneity in the relationship between vitamin E treatment and VTE across categories of a risk factor was evaluated by adding interaction terms between vitamin E and the risk factor to proportional hazards models, with tests for trend performed when subgroup categories were ordinal. As a sensitivity analysis to consider the impact of compliance, additional proportional hazards models were fitted with follow-up time censored when a woman reported taking less than two thirds of her study vitamin E (or matching placebo) during the preceding year. The validity of the proportional hazards assumption was tested by inclusion of an interaction between vitamin E and the log of follow-up time, with time centered at the average log follow-up.20 Additionally, separate proportional hazards models were fitted to the experience of the first 5 years of follow-up and the experience after 5 years to estimate the effects of treatment in early and later follow-up.
The authors had full access to and take full responsibility for the integrity of the data. All authors have read and agree to the manuscript as written.
Baseline Characteristics and VTE Occurrence
Women assigned to active vitamin E and those assigned to placebo had comparable distributions of demographic, medical, and genetic determinants of VTE (Table 1). In terms of important risk factors for VTE, 3% of participants reported a prior VTE at baseline, 5% had factor V Leiden, and 3% had the prothrombin mutation. In addition, 10% of women were ≥65 years of age, 30% currently used hormone therapy, and 18% were obese (body mass index ≥30 kg/m2).
During a median follow-up of 10.2 years (interquartile range, 9.7 to 10.6 years), 482 women had a confirmed VTE (Table 2). Half of the cases had unprovoked VTE, and 37% involved pulmonary embolism. The overall incidence of VTE per 1000 person-years rose from 0.9 in women 45 to 54 years of age at baseline, to 1.4 in women 55 to 64 years of age, and to 2.5 in women ≥65 years of age at baseline. VTE occurred nearly as often as stroke in this study (487 women had a stroke) and more frequently than myocardial infarction (391 women had a myocardial infarction).
Vitamin E and VTE
VTE occurred in 213 women assigned to vitamin E, compared with 269 women in the placebo group (Table 2), a significant 21% reduction in the hazard of VTE associated with vitamin E (relative hazard, 0.79; 95% CI, 0.66 to 0.94; P=0.010). Analyses restricted to unprovoked VTE found a slightly larger 27% reduction in hazard associated with vitamin E (P=0.016). Vitamin E treatment was associated with a 28% reduction in the hazard of pulmonary embolism (P=0.034), whereas the association with deep vein thrombosis only was somewhat weaker and not significant (P=0.10). The relationship of vitamin E with VTE was not significantly modified by either of the other randomized treatments (aspirin or beta carotene), nor did it differ in analyses that censored a woman at the time her compliance was less than two thirds. In particular, among women assigned to active aspirin therapy, the relative hazard of VTE associated with vitamin E was 0.83 (95% CI, 0.64 to 1.07), whereas among women assigned to placebo aspirin, the relative hazard of VTE associated with vitamin E was 0.75 (95% CI, 0.58 to 0.97). Furthermore, no evidence was observed for a varying effect of vitamin E by time: Identical 21% reductions in hazard were observed during the first 5 years of the trial and thereafter. The estimated 10-year risk of VTE was 1.1% in the vitamin E group and 1.4% in the placebo group (Figure).
Examination of the effects of vitamin E on the hazard of VTE in subgroups revealed no significant heterogeneity between groups, either for demographic and medical characteristics (Table 3) or genetic factors (Table 4). The highest subgroup-specific rates of VTE occurred in women who reported a history of VTE at baseline. For them, vitamin E treatment was associated with a significant 44% reduction in the hazard of VTE (relative hazard, 0.56; 95% CI, 0.31 to 1.00; P=0.048). Among women who reported a VTE before randomization, the estimated 10-year risk of a confirmed new VTE was 3.2% in the vitamin E group and 5.5% in the placebo group (Figure). A significant (P=0.040) 18% reduction in the hazard of VTE was observed among women without a history of VTE at baseline.
Women with genetic polymorphisms known to influence VTE had elevated event rates in the placebo group (Table 4). Vitamin E was associated with a significant (P=0.014) 49% reduction in the hazard of VTE among women with either factor V Leiden or the prothrombin mutation. In TT homozygotes for the MTHFR polymorphism, vitamin E was associated with a nonsignificant (P=0.061) 48% reduction in the hazard of VTE.
Altogether, the Women’s Health Study randomized 3097 women at particularly high risk of VTE because of either a history of prior VTE or presence of factor V Leiden or the prothrombin mutation. Vitamin E treatment was associated with a 49% reduction in the hazard of VTE in these women (relative hazard, 0.51; 95% CI, 0.33 to 0.77; P=0.002). Furthermore, the 10-year risk of VTE in these women was 4.1% in the placebo group, compared with 2.1% in the vitamin E group (Figure).
In this large-scale, long-term trial, 600 IU of natural source vitamin E on alternate days was associated with a significant 21% reduction in the hazard of VTE, a significant 27% reduction in the hazard of unprovoked VTE, and a significant 28% reduction in the hazard of pulmonary embolism. The effect of vitamin E was uniform over the 10-year follow-up period and did not vary substantially across subgroups. Among women at high risk because of prior history or prothrombotic mutations, vitamin E was associated with significant 40% to 49% reductions in the hazard of VTE. The absolute reduction in the 10-year risk of VTE was modest in the overall population (0.3%) but substantial (2.0% to 2.3%) in women at high baseline risk. The estimated numbers of women who need to be treated with vitamin E for 10 years to prevent 1 VTE were 357 (95% CI, 200 to 1659) in the overall population and 52 (95% CI, 32 to 144) in the high-risk subgroup of women with a history of VTE or prothrombotic mutation.
VTE is a common, potentially catastrophic clinical event with age-specific rates of occurrence in the Women’s Health Study comparable to those observed in population-based studies.1,2,21,22 As in the Women’s Health Study, previous population-based studies have found an overall rate of VTE comparable to that of stroke1 and myocardial infarction.22 The approach used to identify VTE in this and most other clinical studies does not lead to ascertainment of asymptomatic VTE, which occurs frequently in high-risk individuals and is associated with increased risk of death.23 Thus, results of this trial apply to clinically apparent VTE.
The finding of a reduced rate of VTE associated with vitamin E must be viewed in the context of other randomized evidence of the risks and benefits of vitamin E. The Women’s Health Study9 found a nonsignificant 7% reduction in the hazard of a major cardiovascular event (nonfatal myocardial infarction, nonfatal stroke, or cardiovascular death) associated with vitamin E supplementation (hazard ratio, 0.93; 95% CI, 0.82 to 1.05; P=0.26). This result was consistent with null relationships observed in recent meta-analyses and other trials24–26 and supported the recommendation that vitamin E supplementation does not provide overall benefit for major cardiovascular events.27 The Women’s Health Study also found no significant relationship between vitamin E and the risk of cancer or all-cause death. Whereas secondary analyses of trials of vitamin E supplementation in people with vascular disease or cancer suggest a possibly increased risk of heart failure associated with treatment,8,28 the Women’s Health Study also found no significant relationship between assignment to vitamin E and the incidence of heart failure.
Evaluation of any strategy to influence platelet function and coagulation requires consideration of possible adverse bleeding effects. In the Women’s Health Study, 44 women in the vitamin E group developed hemorrhagic stroke, compared with 48 women in the placebo group (hazard ratio, 0.92; 95% CI, 0.61 to 1.38). The trial also found no significant differences between active vitamin E and placebo groups in rates of gastrointestinal bleeding, hematuria, or easy bruising. Women in the vitamin E group had a significant 6% increased rate of reported epistaxis (hazard ratio, 1.06; 95% CI, 1.01 to 1.11; P=0.02). Overall, vitamin E was associated with far lower bleeding risks than those observed for low-dose aspirin in this trial.14
For patients requiring secondary prevention of VTE, low- or full-intensity warfarin therapy is associated with a >60% reduction in the risk of VTE.29,30 However, bleeding risk and monitoring requirements, as well as drug–drug and drug–food interactions, make indefinite-duration warfarin therapy unacceptable for many patients. Thus, a safer strategy that (1) is useful for both primary and secondary prevention of VTE, (2) requires no laboratory monitoring, and (3) is associated with a substantial risk reduction would be of great value for high-risk patients.
The apparent benefit of vitamin E on risk of VTE differs from its lack of efficacy in prevention of cancer and cardiovascular disease. Controversy persists on whether mechanisms for arterial and venous events are shared and whether effective treatments for one condition are equally useful for the other.31 Prospective cohort studies have found different relationships of risk factors for heart disease and stroke with the risk of VTE.3–5 A unique benefit of vitamin E on risk of VTE is possible.
Limitations of our study include its restriction to healthy and generally health-conscious women, the cautious interpretation required for secondary end points in a trial, and the need for further elaboration of potential mechanisms of action of vitamin E on risk of VTE. Additional studies are needed in men and in women with other comorbid conditions associated with increased risk of VTE. Although VTE was a secondary end point, it was prospectively evaluated, and its report required confirmation by the trial’s end point committee blinded to treatment status. We made no measurements of the impact of vitamin E supplementation on coagulation, and therefore biological mechanisms of potential venous thromboprophylaxis remain speculative.
Overall, vitamin E may be a useful treatment for prevention of a first or recurrent VTE. Because VTE was a prospectively evaluated, secondary end point of the Women’s Health Study, the protective effect observed here requires confirmation in additional studies. Given its lack of efficacy for prevention of cardiovascular disease and cancer, vitamin E may be most appropriate for people at high risk of VTE. In the Women’s Health Study, 20% of the cases of VTE (n=95) occurred among the 8% (n=3097) of participants with a prior history of VTE at baseline, factor V Leiden, or the prothrombin mutation. Vitamin E treatment in these women was associated with a significant 49% reduction in the hazard of VTE during follow-up. With a 4.1% 10-year risk of symptomatic VTE among women receiving placebo in this group, these women would apparently benefit from vitamin E prophylaxis.
We are indebted to the 39 876 participants in the Women’s Health Study for their dedicated and conscientious collaboration and to the entire staff of the study. We thank Roche Molecular Systems, Inc, Alameda, Calif, for providing the genotyping platform used in the present study.
Sources of Funding
This study was supported by grants HL71221, HL43851, and CA47988 from the National Institutes of Health. Vitamin E and vitamin E placebo were provided by the Natural Source Vitamin E Association. Aspirin and aspirin placebo were provided by Bayer Healthcare.
Dr Buring reports that the Natural Source Vitamin E Association provided vitamin E and matching placebo to the Women’s Health Study. The remaining authors report no conflicts.
Glynn RJ, Rosner B. Comparison of risk factors for the competing risks of coronary heart disease, stroke, and venous thromboembolism. Am J Epidemiol. 2005; 162: 975–982.
Alexander G, Suttie J. The effects of vitamin E on vitamin K activity. FASEB J. 1999; 13: A535. Abstract.
Booth SL, Golly I, Sacheck JM, Roubenoff R, Dallal GE, Hamada K, Blumberg JB. Effect of vitamin E supplementation on vitamin K status in adults with normal coagulation status. Am J Clin Nutr. 2004; 80: 143–148.
Murohara T, Ikeda H, Otsuka Y, Aoki M, Haramaki N, Katoh A, Takajo Y, Imaizumi T. Inhibition of platelet adherence to mononuclear cells by α-tocopherol: role of P-selectin. Circulation. 2004; 110: 141–148.
Lee I-M, Cook NR, Manson JE, Buring JE, Hennekens CH. Beta-carotene supplementation and incidence of cancer and cardiovascular disease: the Women’s Health Study. J Natl Cancer Inst. 1999; 91: 2102–2106.
Cheng S, Grow MA, Pallaud C, Klitz W, Erlich HA, Visvikis S, Chen JJ, Pullinger CR, Malloy MJ, Siest G, Kane JP. A multilocus genotyping assay for candidate markers of cardiovascular disease risk. Genome Res. 1999; 9: 936–949.
Zee RYL, Cook NR, Cheng AS, Reynolds R, Erlich HA, Lindpaintner K, Ridker PM. Polymorphism in the P-selectin and interleukin-4 genes as determinants of stroke: a population-based, prospective genetic analysis. Hum Molec Genet. 2004; 13: 389–396.
Wald DS, Law M, Morris JK. Homocysteine and cardiovascular disease: evidence on causality from meta-analysis. BMJ. 2002; 325: 1202–1206.
Cox DR. Regression models and life tables (with discussion). J Roy Statist Soc Ser B. 1972; 34: 187–220.
Anderson FA, Wheeler HB, Goldberg RJ, Hosmer DW, Patwardhan NA, Jovanovic B, Forcier A, Dalen JE. A population-based perspective of the hospital incidence and case-fatality rates of deep vein thrombosis and pulmonary embolism: the Worcester DVT Study. Arch Intern Med. 1991; 151: 933–938.
Kris-Etherton PM, Lichtenstein AH, Howard BV, Steinberg D, Witztum JL. Antioxidant vitamin supplements and cardiovascular disease. Circulation. 2004; 110: 637–641.
Ridker PM, Goldhaber SZ, Danielson E, Rosenberg Y, Eby CS, Deitcher SR, Cushman M, Moll S, Kessler CM, Elliott CG, Paulson R, Wong T, Bauer KA, Schwartz BA, Miletich JP, Bounameaux H, Glynn RJ. Long-term, low-intensity warfarin therapy for the prevention of recurrent venous thromboembolism. N Engl J Med. 2003; 348: 1425–1434.
Kearon C, Ginsberg JS, Kovacs MJ, Anderson DR, Wells P, Julian JA, MacKinnon B, Weitz JI, Crowther MA, Dolan S, Turpie AG, Geerts W, Solymoss S, van Nguyen P, Demers C, Kahn SR, Dassis J, Rodger M, Hambleton J, Gent M. Comparison of low-intensity warfarin therapy with conventional-intensity warfarin therapy for long-term prevention of recurrent venous thromboembolism. N Engl J Med. 2003; 349: 631–639.
Pulmonary embolism and deep venous thrombosis constitute venous thromboembolism (VTE). The death rate from pulmonary embolism is higher than that of acute myocardial infarction. VTE is difficult to diagnose and challenging to treat, with long-term complications that include thromboembolic pulmonary hypertension and chronic venous insufficiency. Low- or full-intensity warfarin effectively reduces the risk of VTE, but bleeding risk, monitoring requirements, and drug–drug and drug–food interactions make indefinite-duration warfarin therapy unacceptable or unfeasible for many patients. Therefore, individuals with permanent risk factors, including factor V Leiden or the prothrombin gene mutation, or a personal or family history of VTE would especially benefit from an effective, low-cost, safe, straightforward strategy to prophylax against VTE. Both basic research and observational clinical studies suggest that vitamin E might reduce the risk of cardiovascular disease and cancer. Although an overview of many randomized trials, including the Women’s Health Study, does not find that vitamin E therapy provides overall benefit for these outcomes, vitamin E taken prophylactically may provide the unique benefit of reducing the risk of VTE. We present in this article data from the Women’s Health Study suggesting that prophylaxis with every-other-day vitamin E taken for 10 years reduces the overall risk of VTE by 21%. The observed risk reduction was 44% among those with prothrombotic mutations or a personal history of VTE. The mechanisms of action to explain these selective beneficial effects of vitamin E in reducing VTE require elaboration, and our clinical findings require confirmation.
Guest Editor for this article was James T. Willerson, MD.
Clinical trial registration information—URL: http://www.clinicaltrials.gov. Unique identifier: NCT00000479.