8/27/2004

Alternative Medicine Alert - ** NOTE: TRUNCATED STORY ** Understanding ofthe role of oxidative stress in human health has grown considerably inrecent years. With it has grown interest in the potential benefit ofantioxidants for disease prevention. As the leading cause of death in theUnited States, heart disease--and its inflammatory component--havereceived special attention in this respect. A number of agents, includingvitamins C and E, beta-carotene, and selenium, have shown promise inobservational studies. As well-controlled clinical trials begin to reportresults, the likely value of various agents is being re-evaluated. Thisreview summarizes the literature to date on one of the better-studiedcompounds, vitamin E.

Mechanism of Action

The mechanism of action of vitamin E in processes associated with heartdisease is the subject of continuing debate. The most commonly offeredexplanation focuses on inhibition of low-density lipoprotein (LDL)cholesterol peroxidation.1 Changes in LDL have been shown to be importantboth in new atherogenesis and in destabilization of existing plaques. Aslipophilic molecules, the various isomers of vitamin E incorporate readilyinto unsaturated fats, effectively pre-empting the action of free-radicaloxidants. However, this property also can have negative consequences. Inthe right circumstances, vitamin E can also act as a pro-oxidant,prompting the observation that vitamins E and C are more effectivetogether, with vitamin C acting as a restorative agent to return vitamin Eto its stable form.2

Other potential mechanisms have been suggested, and some have been testedin vitro or in experimental studies in humans. These include a role forvitamin E in suppressing the expression of adhesion molecules that have arole in atherogenesis, and inhibition of endothelial cellproliferation.3,4 Neither of these possibilities is as well characterizedas the oxidative stress hypothesis, which has been the underlyingrationale in most of the trials conducted so far.

Commercial Preparations

There is a bewildering variety of commercial preparations of vitamin Eavailable, each with evidence that it is absorbed and mobilized by thebody better than the others. The two classes most widely availablecommercially are the tocopherols and the tocotrienols. Each has eightisomers, of which only four are maintained in human tissue.5 Both naturaland synthetic forms are produced, and esterified forms (a-tocopherolacetate or a-tocopherol succinate) usually are used to improve shelf-life.Animal studies and a few human studies have found that natural forms aremore bioavailable than synthetic forms, in a ratio of about 1:1.36.6However, it also has been shown that this difference in biologicalactivity does not translate into meaningful differences in antioxidantactivity.7

The d-a-tocopherol species is the most widely studied and is believed tobe the most biologically important.8 The various forms are notinterconvertible in humans and do not have the same metabolic action.Commercial vitamin E supplements commonly advertise that they contain allthe natural varieties, but given that only four of eight are maintained inhuman tissue, presumably about half of what they contain is of little orno value.

Published Trials

On the basis of promising observational data, several large prospective,controlled trials were started in the mid-1990s. The gathering evidence isincreasingly negative, but the variety of doses and endpoints used meansthat the findings published to date still are somewhat confusing and noteasily summarized.

The CHAOS trial enrolled 2,002 patients with proven cardiac disease andrandomized half to either 400 or 800 IU/d (268 and 536 mg, respectively)of vitamin E.9 After a median follow-up of 510 days, those taking vitaminE had a significantly lower rate of non-fatal myocardial infarction (MI)(Relative Risk [RR] = 0.53), but not death, as compared to those takingplacebo.

The SPACE trial enrolled 196 patients with end-stage renal disease, withrandom assignment to an 800 IU/d (536 mg) or placebo.10 Those takingvitamin E had a significantly reduced rate of a composite cardiovascularendpoint (including sudden death) (RR = 0.54) and a reduced rate of MI (RR= 0.45), as compared to the placebo group. It is worth noting that thestudy population in this case was distinct enough medically to make theseresults, more than any of the others, of questionable generalizability.

An Australian study of young Type 1 diabetics (n = 41) showedsignificantly improved flow-mediated vasodilation (FMD)--a measure ofvascular endothelial function correlated with cardiac events--after athree-month course of 1,000 IU/d of vitamin E (670 mg).11 Brachial FMDrose from 2.6 0.6% at baseline to 7.0 0.7% (P < 0.005) at studycompletion. However, there was no significant change in systemic arterialcompliance, a generalized measure of arterial stiffness and vascularresistance.

A small number of studies have found benefit specifically with thetocotrienol isomers. A Pakistani study of 90 healthy volunteers found thatdaily supplementation with tocotrienol in varying doses producedsignificant reductions in total cholesterol, LDL cholesterol, andapolipoprotein-B.12 In vitro studies from Asia also have produced evidencethat tocotrienol may inhibit endothelial proliferation in the vascularwall, and that it may reduce expression of adhesion molecules (e.g.,VCAM-1, E-Selectin) that play a role in atherogenesis.3,4

At least as many studies have reported only mild or no apparent benefit,and some suggest the possibility of harm. The GISSI-Prevenzione trialenrolled more than 11,000 Italian patients within three months of an MI.13Randomization was factorial, with assignment to vitamin E (300 mg/448 IUdaily) or placebo, and to n-3 polyunsaturated fatty acids (PUFA) or not.Neither the vitamin E + PUFA group, nor the vitamin E alone group showedsignificant improvement in a combined cardiovascular endpoint after anaverage of 3.5 years. However, there was some indication of potentialbenefit in the vitamin E alone group with specific cardiac endpoints.Reductions of 20-35% in various categories of cardiovascular death werereported.

The HOPE study randomized 3,654 diabetics, also in a factorial design, tovitamin E (400 IU/268 mg daily) or placebo, and ramipril or placebo.14After an average of 4.5 years, there was no significant difference in acomposite outcome of MI, stroke, and cardiovascular death, in any of theindividual components, or in any secondary outcome by vitamin E status.The results did not change appreciably when those taking ramipril wereexcluded.

The Primary Prevention Project (PPP) was primarily a low-dose aspirinstudy that also included vitamin E (300 mg/448 IU daily) with the goal ofpreventing first cardiac events in those with major risk factors.15 Bothstudy agents were given on an open-label basis. When the trial was stopped(in response to strong evidence from other trials for the efficacy ofaspirin), there were 4,495 patients enrolled with an average of 3.6 yearsof follow-up. No significant protective effect was noted for vitamin E inthe pre-specified cardiac endpoints or death. However, there was asignificantly lower incidence of peripheral vascular disease (PVD) (RR =0.54) in the vitamin E group, as compared to placebo.

The Heart Protection Study (HPS) enrolled 20,536 patients with coronary orvascular disease, or diabetes, and randomized them by a factorial designto simva-statin or placebo, and a cocktail of antioxidants includingvitamin E (600 mg/896 IU/d) or placebo.16 After five years of follow-up,there was no significant reduction in all-cause mortality, deaths due tovascular causes, non-fatal MI, fatal or non-fatal stroke, or need for arevascularization procedure in the vitamin supplement group. There weresmall and statistically non-significant increases in total mortality andLDL cholesterol. The latter finding is especially important given theconcomitant use of a statin in the trial.

The ATBC study randomized 1,862 men to vitamin E (50 mg/75 IU daily),beta-carotene, both, or placebo.17 All participants were smokers between50 and 69 years of age. After an average of 5.3 years of follow-up, therewere no significant differences in major coronary events (non-fatal MI andfatal coronary heart disease) in any of the supplement groups. There was anon-significant elevation in the risk of death in the vitamin E group (RR= 1.33, P = 0.2).

The ASAP study enrolled 520 patients and assigned them in a factorialdesign to a relatively low dose of vitamin E (136 IU/91 mg daily) orplacebo, and vitamin C or placebo.18 The main outcome parameter was changein intima-media thickness (IMT) of the common carotid artery, awell-studied marker for atherosclerotic progression. After three years,there was no significant benefit with either supplement alone, althoughthere was with the combination. Change in carotid IMT was significantlyless in the combined therapy group, a finding that the authors inferredwould translate into reduced clinical events, although these were notcounted in this study.

An intriguing mechanistic hypothesis also has been put forward to explainthe apparent lack of cardioprotection with vitamin E that has beenobserved in many trials. This involves a series of small, transientischemic events in the heart, a phenomenon called preconditioning.19 It istheorized that the heart's response to these events enables it to betterwithstand a subsequent major event, such as an MI. There is evidence thatantioxidants can interfere with this process.20 If vitamin Esupplementation prevents preconditioning, it would be expected that thosetaking vitamin E would experience proportionally more fatal cardiacevents. As yet, there is little clinical evidence to support or refutethis idea.

Strengths and Limitations of the Evidence

Although there appears to be a consensus developing in the trialliterature, it still amounts to similar words being sung to differenttunes. Several issues should color how the evidence is weighed.

A basic issue is the size of the studies reported to date. The outcomeevents are very rare, even in very large studies. It is impractical toconduct studies large enough to use a single class of event (i.e., deathor MI) as the outcome, so composite endpoints are used. Disparities in theetiology and severity of the component events may mask genuine effects ofthe intervention. However, it should also be noted that consistency of theresults across most endpoints bolsters confidence in the results for anyone endpoint.

With the exception of the Primary Prevention Project, all of the majorstudies have enrolled patients with confirmed coronary disease or anothermajor health problem. This would be expected to maximize the number ofoutcome events, but it begins the intervention late in the diseaseprocess. Studies of younger and healthier people would have to be largerand last longer (and thus be much more expensive), but would round out thepicture with evidence from the pre-acute period.

There also is little similarity across studies in vitamin E dosage orconcomitant medication. This observation cuts both ways. On the one hand,it can be difficult to tease out the effect of vitamin E from the noise ofother agents. Studies with factorial designs often are not powered topermit analyses of the vitamin E-only arm, and two-way analyses thatsimply ignore the other active agent seem likely to produce results onwhich the heterogeneity in both groups has some (unquantifiable)influence. Conversely, the consistency of findings across differentdosages and combinations of other medications suggests that the findingsof any single study are unlikely to result from conditions unique to it.

Conclusion

Although a few trials have shown clinical benefit from vitamin Esupplementation for preventing cardiovascular events, much of the evidenceis against it (see Table). Numerous trials, in disparate populations, withvarying outcomes and large sample sizes, have produced similar findings:Vitamin E, even in large doses, produces results that are similar to, ornot statistically better than, placebo. The preponderance of evidence forthe lack of benefit from vitamin E supplementation in those withestablished cardiovascular disease should motivate clinical researchers tomove on to other compounds.

Recommendation

There appears to be little cardioprotective benefit of vitamin E in commonsupplemental doses once cardiovascular disease is established. However,this should not call into question the value of proper intake of vitaminsand minerals as part of a well-balanced diet, or through use of amultivitamin supplement, if necessary. The benefit of a variety of agentsin recommended daily amounts--especially from whole-food sources--is wellestablished and should be reinforced.

-By Howell Sasser, PhD

Dr. Sasser is Director, Research Epidemiology, R. Stuart Dickson Institutefor Health Studies, Carolinas HealthCare System, Charlotte, NC.

References

1. Traber M. Does vitamin E decrease heart attack risk? Summary andimplications with respect to dietary recommendations. J Nutr2001;131:395S-397S.

2. Heinecke JW. Is the emperor wearing clothes? Clinical trials of vitaminE and the LDL oxidation hypothesis. Arterioscler, Thromb, Vasc Biol2001;21:1261-1264.

3. Inokuchi H, et al. Anti-angiogenic activity of tocotrienol. BiosciBiotechnol Biochem 2003;67:1623-1627.

4. Theriault A, et al. Tocotrienol is the most effective vitamin E forreducing endothelial expression of adhesion molecules and adhesion tomonocytes. Atherosclerosis 2002; 160:21-30.

5. Dietary reference intakes for vitamin C, vitamin E, Selenium, andCarotenoids : A Report of the Panel on Dietary Antioxidants and RelatedCompounds, Subcommittees on Upper Reference Levels of Nutrients andInterpretation and Uses of Dietary Reference Intakes, and the StandingCommittee on the Scientific Evaluation of Dietary Reference Intakes, Foodand Nutrition Board, Institute of Medicine. Washington, DC: U.S.Department of Agriculture; 2000.

6. Burton GW, et al. Human plasma and tissue alpha-tocopherolconcentrations in response to supplementation with deuterated natural andsynthetic vitamin E. Am J Clin Nutr 1998;67:669-684.

7. Burton GW, Ingold KU. Vitamin E as an in vitro and in vivo antioxidant.Ann N Y Acad Sci 1989;570:7-22.

8. Gaytan R, Prisant LM. Oral nutritional supplements and heart disease: Areview. Am J Ther 2001;8:255-274.

<>