Articles
Use of antioxidant vitamins for the prevention of cardiovascular
disease: meta-analysis of randomised trials
Deepak P Vivekananthan, Marc S Penn, Shelly K Sapp, Amy Hsu, Eric J Topol
Department of Cardiovascular Medicine, Cleveland Clinic Foundation, 9500 Euclid Avenue,
NC-10, Cleveland, OH 44195, USA (D P Vivekananthan MD, M S Penn MD, Shelly K Sapp MS, A
Hsu MS, E J Topol MD)
Correspondence to: Dr Marc S Penn (e-mail:pennm@ccf.org)
Summary
Introduction
Methods
Results
Discussion
References
Summary
Introduction Oxidised LDL is thought to play an important part in the pathogenesis of
atherosclerosis. Observational studies have associated tocopherol (vitamin E), ß carotene, or
both, with reductions in cardiovascular events, but not clinical trials. We did a meta-analysis to
assess the effect of these compounds on long-term cardiovascular mortality and morbidity.
Methods We analysed seven randomised trials of vitamin E treatment and, separately, eight of ß
carotene treatment; all trials included 1000 or more patients. The dose range for vitamin E was 50800 IU, and for ß carotene was 15-50 mg. Follow-up ranged from 1·4 to 12·0 years.
Findings The vitamin E trials involved a total of 81788 patients and the ß ca rotene trials 138113 in
the all-cause mortality analyses. Vitamin E did not provide benefit in mortality compared with control
treatment (11·3 vs 11·1%, odds ratio 1·02 [95% CI 0·98-1·06] p=0·42) or significantly decrease risk
of cardiovascular death (6·0 vs 6·0%, p=0·86) or cerebrovascular accident (3·6 vs 3·5%, p=0·31). ß
carotene led to a small but significant increase in all-cause mortality (7·4 vs 7·0%, 1·07 [1·02-1·11]
p=0·003) and with a slight increase in cardiovascular death (3·4 vs 3·1%, 1·1 [1·03-1·17] p=0·003).
No significant heterogeneity was noted for any analysis.
Interpretation The lack of a salutary effect was seen consistently for various doses of vitamins in
diverse populations. Our results, combined with the lack of mechanistic data for efficacy of vitamin
E, do not support the routine use of vitamin E.
Lancet 2003; 361: 2017-23
Introduction
The oxidative-modification hypothesis of atherosclerosis1-4 has prompted the study of antioxidant
vitamins in the prevention of the initiation and progression of cardiovascular disease. Preclinical
studies suggested that supplementation of the diet with various compounds that have antioxidant
properties before the development of vascular disease inhibited the atherogenic process. 5-9 These
findings led to several large, prospective, cohort studies in human beings, in which significant
reductions in mortality10 and cardiovascular events11,12 were identified in men and women taking
antioxidant vitamins. However, sizeable randomised trials of antioxidant vitamins 13-17 have shown
no such mortality reduction, although in one study non-fatal myocardial infarction (MI) was
significantly reduced.18 More importantly, in two randomised trials of ß carotene16,19 no benefit, and
possibly an increased risk of cardiovascular events, was seen. Findings from small randomised
studies of antioxidant vitamins have also suggested a potential harmful effect of antioxidant
vitamins in patients with known or suspected coronary disease.20,21
Despite the absence of efficacy of antioxidant vitamins reported in larger randomised trials, two
important opinion articles have favoured the universal use of multivitamins by consumers. 22,23 The
multivitamins recommended, however, contain ß carotene and tocopherol (vitamin E), two
compounds that have not been proven to reduce cardiovascular morbidity or mortality, and may
adversely affect lipid concentrations when used at higher doses. 13,20 Since the use of antioxidant
vitamins continues to grow, partly encouraged by physicians advocating their use, 24 we did a metaanalysis of randomised trials to find out what effect antioxidant vitamins have on all-cause mortality
and cardiovascular death.
Methods
Study population
We did a MEDLINE search to identify all randomised controlled trials of antioxidant vitamins in
primary and secondary prevention. We used the search terms: "randomized controlled trials",
"vitamin E", and "beta carotene". We did additional searches for known trial acronyms cited in
review articles, and searched by hand the bibliographies of primary studies identified through the
initial search. To limit the effects of publication bias of smaller trials we included only studies of
1000 or more patients. To reduce the possibility of confounding from inclusion of nutritionally
deficient populations, our analysis was limited to studies in populations from developed countries
without overt evidence of vitamin deficiencies.
Two investigators (DPV and SKS) independently reviewed the primary studies to assess the
appropriateness for inclusion in our analysis and data abstraction. Trial inclusion was based on the
quality of the study's methods, including trial size, randomisation scheme, and use of an intentionto-treat analysis. The prospectively identified outcomes of interest included all-cause mortality,
cardiovascular death, all-cause cerebrovascular accident, and non-fatal MI. Many trials did not
report the individual rates of non-fatal MI and, therefore, we chose to use the more widely reported
combined endpoint of cardiovascular death or non-fatal MI. We excluded trials without all-cause
mortality data.
We identified 12 trials for analysis (tables 1 and 2). Eight trials involving ß carotene alone or in
combination with other antioxidants were analysed (table 1). We further classified the studies in our
analysis by cardiovascular risk of the study population. Four studies were secondary prevention
studies, defined as including patients with known or documented vascular disease, active tobacco
use or asbestos exposure, or documented history of previous malignant disease. The AlphaTocopherol Beta Carotene Cancer Prevention Study (ATBC) 16 investigated the effects of ß
carotene, vitamin E, or both, on the frequency of major cardiac events and rate of lung cancer in a
population of middle-aged male smokers. The Beta Carotene and Retinol Efficacy Trial (CARET) 19
assessed efficacy and safety of ß carotene in men and women at high risk of lung cancer because
of previous asbestos exposure or extensive cigarette smoking. The Heart Protection Study (HPS) 13
assessed the impact of an antioxidant vitamin combination (600 mg vitamin E, 250 mg vitamin C,
and 20 mg ß carotene) on vascular and non-vascular mortality and morbidity among patients at
high risk because of history of coronary disease, diabetes, or peripheral vascular disease. The Skin
Cancer Prevention Study (SCP)25 assessed the efficacy of ß carotene in the secondary prevention
of non-melanoma skin cancer among elderly patients who had a history of biopsy-proven basal-cell
or squamous-cell carcinoma.
Patients' characteristics
Location of
study
population
Number in
Dose
Treatment group
ß
Control
carotene
Trial
Secondary prevention
Length of
follow-up
(years)
ATBC16*
CARET19
HPS13*
Age range 50-69 years;
100% male smokers
(n=29133)
Age range 45-69 years;
former/active smokers or
asbestos exposure; 66%
male (n=18314)
Southwestern
Finland,
multicentre
USA,
multicentre
14560
14573
20 mg four 6·1
times daily
9420
8894
15-30 mg 4·0
four times
daily + 25
000 IU
retinol
20 mg four 5·0
times daily
Age range 40-80 years;
UK, multicentre 10269
10267
known vascular disease or
at-risk of vascular disease;
75% male (n=20536)
25
SCP
Age <85 years (most <65
USA,
913
892
50 mg four 5·0
years); previous nonmulticentre
times daily
melanoma skin cancer; 69%
male (n=1805)
Primary prevention
AREDS26* Age range 55-80 years; at- USA,
2370
2387
15 mg four 6·3
risk of cataract or vision
multicentre
times daily
loss; 44% male (n=4757)
NSCP27 Age range 20-69 years; at- Queensland,
820
801
30 mg four 4·5
risk of basal-cell or
Australia
times daily
squamous-cell cancer; 44% multicentre
male (n=1621)
PHS28
Age range 40-84 years; no USA,
11036
1035
50 mg four 12·0
history of cancer or vascular multicentre
times daily
disease; 100% male
physicians (n=22 071)
WHS29* Age range >45 years; no
USA,
19939
19937 50 mg four 2·1
history of cancer or vascular multicentre
times
disease; 100% female
daily
health professionals (n=39
876)
*ß carotene taken as part of antioxidant cocktail or factorial randomisation including vitamin E.
Table 1: Summary of randomised trials of ß carotene treatment
Four ß-carotene trials were primary prevention studies or were among low-risk patients. The AgeRelated Eye Disease Study (AREDS)26 assessed the safety and efficacy of an antioxidant
combination (15 mg ß carotene, 400 IU vitamin E, and 500 mg vitamin C) in the prevention of
cataract formation and macular degeneration in middle-aged patients. The Nambour Skin Cancer
Prevention (NSCP) trial27 investigated the effect of ß carotene on the rate of non-melanoma skin
cancer in a young to middle-aged white population. In contrast to the SCP trial, a previous
diagnosis of biopsy-proven skin malignant disease was not a requirement for study entry. The
Physicians' Health Study (PHS)28 investigated the effect of supplementation with ß carotene on the
frequency of cardiovascular disease or malignant neoplasms in male physicians with no history of
vascular disease or malignant disease. Similarly, in the Women's Health Study (WHS), 29 female
health professionals with no history of vascular disease or malignant disease were randomly
assigned ß carotene or vitamin E in a two-by-two factorial design. Its aim was to assess the effect of
antioxidant vitamin supplementation on the frequency of cardiovascular disease and malignant
disease.
We identified seven trials involving vitamin E alone or in combination with other antioxidants, which
we assessed in a separate analysis. Most of these trials studied the efficacy of vitamin E in
secondary prevention (table 2). In addition to ATBC16 and HPS,13 the secondary prevention trials
were the Cambridge Heart Antioxidant Study (CHAOS),18 the Gruppo Italiano per lo Studio della
Sopravvivenza nell'Infarto miocardico (GISSI) Prevention Study, 17 and the Heart Outcomes
Prevention Evaluation (HOPE) study.15 The goal of CHAOS was to study the efficacy of vitamin E in
the prevention of cardiovascular death or non-fatal MI in patients with angiographically documented
coronary disease. Similarly, GISSI assessed the efficacy of vitamin E on cardiovascular death, nonfatal MI, or stroke in patients with recent MI ( 3 months). In addition to AREDS, the Primary
Prevention Project (PPP)14 studied the efficacy of vitamin E among patients who had one or more
cardiovascular risk factors, including hypertension, diabetes, or early family history of coronary
disease. In HPS and AREDS, patients were randomly assigned antioxidant combinations containing
ß carotene and vitamin E. ATBC and WHS used two-by-two factorial designs for randomisation, in
which participants received vitamin E alone, ß carotene alone, both vitamins, or neither vitamin.
These trials were included in each of the individual meta-analyses for ß carotene and vitamin E.
The WHS study has not yet reported the results of the vitamin E randomisation and, therefore, we
did not include this study in the vitamin E analysis.
Patients' characteristics
Trial
Secondary prevention
ATBC16* Mean age 57 years; male
smokers without known lung
cancer (n=29133)
18
CHAOS † Median age 62 years;
angiographically proven CAD;
84% male (n=2002)
17
GISSI
Survivors of recent MI (<3
months); 85% male
(n=11324)
HOPE15† Mean age 66 years; known
cardiovascular disease or
diabetes; 73% male (n=9541)
HPS13*
Location of study Number in
Dose Length of
population
treatment group
follow-up
(years)
Vitamin Control
E
Southwestern
14564
Finland, multicentre
14569
50 mg 6·1
UK, single centre
1035
967
1·4
Italy, multicentre
5660
5664
400800
IU
300
mg
Multinational:
Canada, USA,
Europe, South
America
UK, multicentre
4761
4780
400
IU
4·5
3·5
Age range 40-80 years;
10269 10267 600
5·0
known vascular disease or atmg
risk of vascular disease; 75%
male (n=20536)
Primary prevention
AREDS26* Age range 55-80 years; atUSA, multicentre
2370
2387
400
6·3
risk of cataract of vision loss;
IU
44% male (n=4757)
14
PPP
Mean age 64·4 years; primary Italy, multicentre
2231
2264
300
3·6
prevention in patients with at
mg
least one risk factor; 57%
male (n=4495)
*Vitamin E as part of antioxidant cocktail or factorial randomisation including ß carotene. †Vitamin E
from natural sources.
Table 2: Summary of randomised trials of vitamin E treatment
Statistical analysis
We compared findings in an intention-to-treat analysis. The frequency of each cardiovascular
endpoint for patients receiving vitamin treatment and those not receiving treatment was assessed
from the primary publication source. The researchers from the studies were contacted for data
missing in the original publication. Within each study, we used a 2 test to assess the effect of
vitamin treatment, and odds ratios and 95% CI were calculated. We took p<0·05 to be significant.
We used a Breslow-Day test to test the homogeneity of the odds ratios across the trials, for which
p<0·05 was deemed significant. Cochran-Mantel Haenszel tests were used to investigate the effect
of vitamin therapy on each endpoint across the ß carotene and, separately, across vitamin E trials,
and pooled odds ratios and 95% CI were calculated.
Results
In total, 138113 patients were assigned ß carotene or control treatment (table 1). The all-cause
mortality rate was 7·4% in the ß carotene group and 7·0% in the control group (figure 1). The odds
ratio of all-cause death for patients treated with ß carotene was slightly increased (1·07 [95%CI
1·02-1·11] p=0·003). The endpoint of cardiovascular death was assessed in six trials, involving
131551 patients. The rate of death from cardiovascular causes was 3·4% in the ßcarotene group
and 3·1% in the control group (figure 2). The odds ratio for cardiovascular death with ß carotene
treatment was also slightly increased (1·1 [1·03-1·17] p=0·003). All-cause cerebrovascular accident
was adjudicated in three trials (figure 3). In the HPS, PHS, and WHS trials combined, the rate of
cerebrovascular accident did not differ significantly between patients treated with ß carotene and
those treated with control treatment (2·3 vs 2·3%, 1·0 [0·91-1·09] p=0·92). The Breslow-Day test for
homogeneity of the odds ratios for these analyses was not significant.
Figure 1: Odds ratios (95% CI) of all-cause mortality for individuals treated with ß carotene or
control therapy
Figure 2: Odds ratios (95% CI) of cardiovascular death for individuals treated with ß carotene
or control therapy
Figure 3: Odds ratios (95% CI) of all-cause cerebrovascular accident for individuals treated
with ß carotene or control therapy
81788 patients were included in the all-cause mortality analysis for vitamin E (table 2). Patients
assigned vitamin E had a non-significant excess in mortality (11·3 vs 11·1%, 1·02 [0·98-1·06],
p=0·42; figure 4). Vitamin E did not significantly lower cardiovascular mortality compared with
control treatment (6·0 vs 6·0%, 1·0 [0·94-1·06], p=0·94; figure 5). The endpoint of cerebrovascular
accidents was assessed in four vitamin E trials.13-15,17 Frequency of cerebrovascular accident did
not differ between patients treated and those not treated with vitamin E (3·6 vs 3·5%, 1·02 [0·921·12], p=0·71; figure 6). The Breslow-Day test for these endpoints showed no significant
homogeneity. The composite endpoint of cardiovascular death or non-fatal MI was assessed in
38367 patients across four trials (figure 7).13,14,17,18 The Breslow-Day analysis yielded p=0·053,
which suggests heterogeneity among these trials. Treatment with vitamin E had no effect on this
outcome (9·8 vs 9·8%, 1·0 [0·94-1·07], p=0·93).
Figure 4: Odds ratios (95% CI) of all-cause mortality for individuals treated with vitamin E or
control therapy
Figure 5: Odds ratios (95% CI) of cardiovascular death for individuals treated with vitamin E
or control therapy
Figure 6: Odds ratios (95% CI) of all-cause stroke for individuals treated with vitamin E or
control therapy
Figure 7: Odds ratios (95% CI) of the combined endpoint of cardiovascular death or non-fatal
MI for individuals treated with vitamin E or control therapy
We did a supplementary meta-analysis, in which trials were classified according to use of vitamin E
as primary or secondary prevention. Five trials,13,15-18 including 72536 patients, assessed all-cause
mortality in secondary prevention trials. Treatment with vitamin E was not associated with an
improvement in all-cause mortality (11·9 vs 11·7% 1·02 [0·97-1·06]) in this subgroup. Similarly, the
two primary prevention trials,14,26 in which were included 9242 patients, showed no improvement in
all-cause mortality (7·0 vs 6·6%, 1·06 [0·93-1·25]). The Breslow-Day test for these analyses was
not significant.
In an additional meta-analysis we included the results of the Secondary Prevention with
Antioxidants of Cardiovascular Disease in Endstage Renal Disease (SPACE) trial30 and the
Women's Angiographic Vitamin and Estrogen (WAVE) study,21 two randomised studies of vitamin E
in selected populations that did not meet our selection criteria for inclusion in the main metaanalysis because of small sample sizes. We found no significant benefit or trend towards benefit of
vitamin E on all-cause mortality (11·3 vs 11·1%, 1·02 [0·98-1·07]), cardiovascular death (6·0 vs
6·0%, 1·0 [0·95-1·06]), or risk of stroke (3·6 vs 3·5%, 1·02 [0·92-1·12]) with inclusion of these trials.
The Breslow-Day test for these analyses remained non-significant.
Discussion
The small harmful effect we noted for ß carotene was driven by the increased risk reported in the
CARET and ATBC studies, in which the patients were at high risk of lung cancer, and represented
34% of our total sample of patients. With the exception of the NSCP trial,27 the tendency towards a
harmful effect on death was strikingly consistent across all major trials, including diverse
populations. Our findings are especially concerning given that the relevant ß carotene doses are
commonly used in preparations of over-the-counter vitamin supplements and are included in
smaller doses in readily available multivitamin supplements that have been advocated for
widespread use.22,23 The trend towards a beneficial effect of ß carotene treatment reported in NSCP
requires further explanation. Patients in NSCP were generally younger than those in other studies
and all patients were enrolled from Queensland, Australia. Therefore, the possibility of an
interaction between ß carotene, age, and ethnic origin cannot be excluded. However, the small
sample size and low number of events makes chance a likely explanation for the trial's findings.
The initial enthusiasm for the clinical use of antioxidant vitamins in the prevention of cardiovascular
disease stemmed from positive results in the preclinical setting. Animal data suggested
antioxidants, particularly vitamin E, prevented initiation of disease progression in laboratory animals
without known atherosclerosis.5,6 These early positive findings and the presumed safety of
antioxidant supplementation led to large, prospective cohort studies, in which an association
between antioxidant vitamin intake, serum vitamin concentrations, or both, and improved
cardiovascular outcomes were reported.10-12,31 In an analysis from the Lipid Research Clinics'
Primary Prevention Trial, men with serum carotenoid concentrations in the highest quartile had a
36% reduction in the risk of cardiovascular disease or non-fatal MI.31 A similar proportionate
reduction in the risk of coronary disease was noted in men and women with the highest vitamin E
intakes.11,12 However, findings were not confirmed in prospective cohort randomised trials, which
suggests confounding.
Other factors associated with improved cardiovascular outcomes, such as lifestyle, degree of
fitness, and diet are difficult to assess precisely and were not fully accounted for in these
prospective analyses. Moreover, the trials assessed vitamin supplement use and the association
between food intake rich in antioxidants and subsequent cardiovascular events. Antioxidant-rich
foods might provide other beneficial nutrients such as flavenoids and lycopenes not present in
standard oral vitamin supplements. The natural forms of vitamins in food may have biological
activity32 or potency33 different to those for synthetic vitamin compounds used in supplements. For
instance, the natural form of vitamin E, or d- -tocopherol (1·5 IU/mg) is more bioavailable than the
synthetic form, all-rac- -tocopherol 1·1 U/mg). Importantly, unlike most large-scale randomised
trials of vitamin E, HOPE and CHAOS used the natural form, and, thus, the neutral findings in these
trials on hard endpoints may refute the hypothesis of a differential clinical effect between natural
and synthetic sources of tocopherol.
Further study is needed to better understand the reason for the increased risk of death associated
with ß carotene and the lack of clinical efficacy of vitamin E in the improvement of cardiovascular
outcomes. Several plausible mechanisms have been suggested. ß carotene in particular is a poor
inhibitor of in-vivo LDL oxidation.34 Moreover, cigarette smoke destabilises the ß carotene molecule
resulting in abnormal signal transduction and the up-regulation of growth factors associated with
tumorigenesis.35,36 Finally, ß carotene has adverse effects on lipids.37 Other studies may provide an
explanation for the harmful or null effects of vitamin-E-based antioxidant treatment. In the HDLAtherosclerosis Treatment Study (HATS),20 an antioxidant combination including vitamin E (800 IU)
blunted the HDL raising effect of niacin and simvastatin and resulted in poor clinical outcomes. The
rise in HDL2, the protective fraction of HDL, was most noticeably lessened. Other workers have
challenged the suggestion that vitamin E is a potent in-vivo inhibitor of LDL oxidation. In a
randomised placebo-controlled study in healthy adults, involving doses of vitamin E as high as 2000
mg daily, Meagher and colleagues38 showed that the breakdown products of lipid peroxidation were
unaffected despite a substantial increase in plasma vitamin E concentrations. Finally, some
researchers have speculated that antioxidants are more effective in inhibiting the early stages of
atherosclerosis, such as fatty-streak formation, than preventing sequelae in the advanced stages
experienced by most patients participating in clinical trials. 39 The observed benefits of an
antioxidant combination containing vitamin E in the prevention of early stages of cardiac transplant
vasculopathy is consistent with this hypothesis.40
The issue of timing of antioxidant treatment is pertinent because the striking benefit of vitamin E
reported in observational studies occurred in primary prevention settings among patients without
documented vascular disease.11,12 Given that in only one randomised trial in our meta-analysis was
vitamin E studied alone in primary prevention,14 our null results were driven by trials of vitamin E
used as secondary prevention. Perhaps further study of vitamin E in primary prevention populations
will, therefore, be of interest. In all animal studies in which a significant antiatherosclerotic effect of
vitamin E has been reported, the vitamin E was started at the time of high-fat diet or before any
histological evidence of neointimal or fatty-streak formation.7 That said, human trials that would start
therapy in patients in their adult years are highly unlikely to show significant benefit.
Our meta-analysis has several limitations. Although we clarified inconsistencies or missing data with
the primary investigators, we did not reconfirm all published data with the primary researchers.
Although we did not use a standard definition for clinically relevant endpoints, all-cause mortality
data were available for all trials and this is an objective and unbiased endpoint. 41 Because of the
lack of detailed outcome data, we were unable to assess the effect of antioxidant vitamins in
specific cohorts of patients of interest with high oxidative stress, such as smokers and patients with
diabetes and on haemodialysis. This point is notable because in one small randomised trial of
antioxidant use in a renal dialysis population, MI was reduced by 70%. 30 However, among the
20536 patient HPS trial, which represents almost 25% of the patients assigned vitamin E in our
analysis, no preferential benefit of antioxidant vitamins was seen in a subgroup analysis including
active smokers and patients with diabetes. Another potential limitation is that we analysed the
results of several large randomised trials that included diverse populations with a differential risk for
development or progression of vascular disease. Several trials included patients at high-risk of lung
cancer.16,19 Although the Breslow-Day test for heterogeneity was not significant, it may have been
underpowered to detect small but clinically important differences between the trials. We focused
mainly on cardiovascular outcomes, but antioxidants could have favourable effects in the prevention
of cancer.42 Continuing studies will help to answer this question.
We restricted our analysis to populations in developed countries without known vitamin deficiency.
On the basis of the results of a randomised trial in nutritionally deficient Chinese, vitamin
supplementation might have a role in this population43 or in patients with vitamin deficiencies
secondary to malabsorption of fat-soluble vitamins. Finally, we saw no benefit from antioxidants in
long-term outcomes, but this does not disprove the oxidative-modification hypothesis of
atherosclerosis.39 Although the doses of antioxidant vitamins studied were much higher than the
recommended daily allowances,44 and the trials that measured vitamin concentrations reported
several-fold rises in plasma vitamin concentrations,13,16,18,19,25,26,28 no trial assessed efficacy of
antioxidant supplementation by measuring markers of lipid peroxidation. Thus, the vitamins used
might have been inadequately dosed,38 given to patients with low-level oxidative stress,39 or given
too late in the course of cardiovascular disease to achieve a clinically relevant impact from inhibition
of LDL oxidation.
Given the results of this meta-analysis, the use of vitamin supplements containing ß carotene and
vitamin A, ß carotene's biologically active metabolite, should be actively discouraged because this
family of agents is associated with a small but significant excess of all-cause mortality and
cardiovascular death. We recommend that clinical studies of ß carotene should be discontinued
because of its risks. When used as secondary prevention, vitamin E did not reduce the risk of
cardiovascular endpoints. Furthermore, given our results and the lack of mechanistic data
supporting efficacy of vitamin E as a potent antioxidant in vivo, we do not support the continued use
of vitamin E treatment and discourage the inclusion of vitamin E in future primary and secondary
prevention trials in patients at high risk of coronary artery disease.
Conflict of interest statement
None declared.
Acknowledgments
No funding source was used for this analysis. We thank Charlene Surace for her graphical support
for the figures used in this report
References
1 Quinn MT, Parthasarathy S, Fong LG, Steinberg D. Oxidatively modified low density lipoproteins:
a potential role in recruitment and retention of monocyte/macrophages during atherogenesis. Proc
Natl Acad Sci USA 1987; 84: 2995-98. [PubMed]
2 Navab M, Berliner JA, Watson AD, et al. The Yin and Yang of oxidation in the development of the
fatty streak: a review based on the 1994 George Lyman Duff Memorial Lecture. Arterioscler
Thromb Vasc Biol 1996; 16: 831-42. [PubMed]
3 Berliner JA, Navab M, Fogelman AM, et al. Atherosclerosis: basic mechanisms. Oxidation,
inflammation, and genetics. Circulation 1995; 91: 2488-96. [PubMed]
4 Diaz MN, Frei B, Vita JA, Keaney JF, Jr. Antioxidants and atherosclerotic heart disease. N Engl J
Med 1997; 337: 408-16. [PubMed]
5 Prasad K, Kalra J. Oxygen free radicals and hypercholesterolemic atherosclerosis: effect of
vitamin E. Am Heart J 1993; 125: 958-73. [PubMed]
6 Parker RA, Sabrah T, Cap M, Gill BT. Relation of vascular oxidative stress, alpha-tocopherol, and
hypercholesterolemia to early atherosclerosis in hamsters. Arterioscler Thromb Vasc
Biol 1995; 15: 349-58. [PubMed]
7 Pratico D, Tangirala RK, Rader DJ, Rokach J, FitzGerald GA. Vitamin E suppresses isoprostane
generation in vivo and reduces atherosclerosis in ApoE-deficient mice. Nat Med 1998; 4: 118992. [PubMed]
8 Crawford RS, Kirk EA, Rosenfeld ME, LeBoeuf RC, Chait A. Dietary antioxidants inhibit
development of fatty streak lesions in the LDL receptor-deficient mouse. Arterioscler Thromb Vasc
Biol 1998; 18: 1506-13. [PubMed]
9 Thomas SR, Leichtweis SB, Pettersson K, et al. Dietary cosupplementation with vitamin E and
coenzyme Q(10) inhibits atherosclerosis in apolipoprotein E gene knockout mice. Arterioscler
Thromb Vasc Biol 2001; 21: 585-93. [PubMed]
10 Losonczy KG, Harris TB, Havlik RJ. Vitamin E and vitamin C supplement use and risk of allcause and coronary heart disease mortality in older persons: the Established Populations for
Epidemiologic Studies of the Elderly. Am J Clin Nutr 1996; 64: 190-96. [PubMed]
11 Rimm EB, Stampfer MJ, Ascherio A, Giovannucci E, Colditz GA, Willett WC. Vitamin E
consumption and the risk of coronary heart disease in men. N Engl J Med 1993; 328: 145056. [PubMed]
12 Stampfer MJ, Hennekens CH, Manson JE, Colditz GA, Rosner B, Willett WC. Vitamin E
consumption and the risk of coronary disease in women. N Engl J Med 1993; 328: 144449. [PubMed]
13 MRC/BHF Heart Protection Study of antioxidant vitamin supplementation in 20536 high-risk
individuals: a randomised placebo-controlled trial. Lancet 2002; 360: 23-33. [Text]
14 Collaborative Group of the Primary Prevention Project (PPP). Low-dose aspirin and vitamin E in
people at cardiovascular risk: a randomised trial in general practice. Lancet 2001; 357: 8995. [Text]
15 The Heart Outcomes Prevention Evaluation Study Investigators. Vitamin E supplementation and
cardiovascular events in high-risk patients. N Engl J Med 2000; 342: 154-60. [PubMed]
16 The Alpha-Tocopherol Beta Carotene Cancer Prevention Study Group. The effect of vitamin E
and beta carotene on the incidence of lung cancer and other cancers in male smokers. N Engl J
Med 1994; 330: 1029-35. [PubMed]
17 Dietary supplementation with n-3 polyunsaturated fatty acids and vitamin E after myocardial
infarction: results of the GISSI-Prevenzione trial. Lancet 1999; 354: 447-55. [Text]
18 Stephens NG, Parsons A, Schofield PM, Kelly F, Cheeseman K, Mitchinson MJ. Randomised
controlled trial of vitamin E in patients with coronary disease: Cambridge Heart Antioxidant Study
(CHAOS). Lancet 1996; 347: 781-86. [PubMed]
19 Omenn GS, Goodman GE, Thornquist MD, et al. Effects of a combination of beta carotene and
vitamin A on lung cancer and cardiovascular disease. N Engl J Med 1996; 334: 1150-55. [PubMed]
20 Brown BG, Zhao XQ, Chait A, et al. Simvastatin and niacin, antioxidant vitamins, or the
combination for the prevention of coronary disease. N Engl J Med 2001; 345: 1583-92. [PubMed]
21 Waters DD, Alderman EL, Hsia J, et al. Effects of hormone replacement therapy and antioxidant
vitamin supplements on coronary atherosclerosis in postmenopausal women: a randomized
controlled trial. JAMA 2002; 288: 2432-40. [PubMed]
22 Willett WC, Stampfer MJ. Clinical practice: what vitamins should I be taking, doctor? N Engl J
Med 2001; 345: 1819-24. [PubMed]
23 Fletcher RH, Fairfield KM. Vitamins for chronic disease prevention in adults: clinical applications.
JAMA 2002; 287: 3127-29. [PubMed]
24 Mehta J. Intake of antioxidants among American cardiologists. Am J Cardiol 1997; 79: 155860. [PubMed]
25 Greenberg ER, Baron JA, Stukel TA, et al. A clinical trial of beta carotene to prevent basal-cell
and squamous-cell cancers of the skin: The Skin Cancer Prevention Study Group. N Engl J
Med 1990; 323: 789-95. [PubMed]
26 Age-Related Eye Disease Study Group. A randomized, placebo-controlled, clinical trial of highdose supplementation with vitamins C and E and beta carotene for age-related cataract and vision
loss: AREDS report no. 9. Arch Ophthalmol 2001; 119: 1439-52. [PubMed]
27 Green A, Williams G, Neale R, et al. Daily sunscreen application and betacarotene
supplementation in prevention of basal-cell and squamous-cell carcinomas of the skin: a
randomised controlled trial. Lancet 1999; 354: 723-29. [Text]
28 Hennekens CH, Buring JE, Manson JE, et al. Lack of effect of long-term supplementation with
beta carotene on the incidence of malignant neoplasms and cardiovascular disease. N Engl J
Med 1996; 334: 1145-49. [PubMed]
29 Lee I, 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: 2102-06. [PubMed]
30 Boaz M, Smetana S, Weinstein T, et al. Secondary prevention with antioxidants of
cardiovascular disease in endstage renal disease (SPACE): randomised placebo-controlled trial.
Lancet 2000; 356: 1213-18. [Text]
31 Morris DL, Kritchevsky SB, Davis CE. Serum carotenoids and coronary heart disease: the Lipid
Research Clinics Coronary Primary Prevention Trial and Follow-up Study. JAMA 1994; 272: 143941. [PubMed]
32 Brigelius-Flohe R, Traber MG. Vitamin E: function and metabolism. Faseb J 1999; 13: 114555. [PubMed]
33 Hoppe PP, Krennrich G. Bioavailability and potency of natural-source and all-racemic alphatocopherol in the human: a dispute. Eur J Nutr 2000; 39: 183-93. [PubMed]
34 Reaven PD, Khouw A, Beltz WF, Parthasarathy S, Witztum JL. Effect of dietary antioxidant
combinations in humans: protection of LDL by vitamin E but not by beta-carotene. Arterioscler
Thromb 1993; 13: 590-600. [PubMed]
35 Wang XD, Russell RM. Procarcinogenic and anticarcinogenic effects of beta-carotene. Nutr
Rev 1999; 57 :(part 1) 263-72. [PubMed]
36 Wang XD, Liu C, Bronson RT, Smith DE, Krinsky NI, Russell M. Retinoid signaling and activator
protein-1 expression in ferrets given beta-carotene supplements and exposed to tobacco smoke. J
Natl Cancer Inst 1999; 91: 60-66. [PubMed]
37 Redlich CA, Chung JS, Cullen MR, Blaner WS, Van Bennekum AM, Berglund L. Effect of longterm beta-carotene and vitamin A on serum cholesterol and triglyceride levels among participants in
the Carotene and Retinol Efficacy Trial (CARET). Atherosclerosis 1999; 143: 427-34. [PubMed]
38 Meagher EA, Barry OP, Lawson JA, Rokach J, FitzGerald GA. Effects of vitamin E on lipid
peroxidation in healthy persons. JAMA 2001; 285: 1178-82. [PubMed]
39 Steinberg D, Witztum JL. Is the oxidative modification hypothesis relevant to human
atherosclerosis? Do the antioxidant trials conducted to date refute the hypothesis?
Circulation 2002; 105: 2107-11. [PubMed]
40 Fang JC, Kinlay S, Beltrame J, et al. Effect of vitamins C and E on progression of transplantassociated arteriosclerosis: a randomised trial. Lancet 2002; 359: 1108-13. [Text]
41 Lauer MS, Blackstone EH, Young JB, Topol EJ. Cause of death in clinical research: time for a
reassessment? J Am Coll Cardiol 1999; 34: 618-20. [PubMed]
42 Clark LC, Combs GF, Jr., Turnbull BW, et al. Effects of selenium supplementation for cancer
prevention in patients with carcinoma of the skin: a randomized controlled trial--Nutritional
Prevention of Cancer Study Group. JAMA 1996; 276: 1957-63. [PubMed]
43 Blot WJ, Li JY, Taylor PR, et al. Nutrition intervention trials in Linxian, China: supplementation
with specific vitamin/ mineral combinations, cancer incidence, and disease-specific mortality in the
general population. J Natl Cancer Inst 1993; 85: 1483-92. [PubMed]
44 Monsen ER. Dietary reference intakes for the antioxidant nutrients: vitamin C, vitamin E,
selenium, and carotenoids. J Am Diet Assoc 2000; 100: 637-40. [PubMed]