Vitamin E supplementation: are the benefits worth the risks?

Grand Rounds presentation
Article section 1 of 3.  Next

By Douglas J Lanska MD MS MSPH

The antioxidant properties of vitamin E have been the basis for numerous clinical studies of vitamin E supplementation in the prevention of cardiovascular and cancer-related mortality, as well as numerous clinical studies of whether vitamin E supplementation may be helpful in slowing the progression of various neurodegenerative diseases.

The typical American diet supplies only 6 to 10 international units (IU) of vitamin E daily, and the recommended daily allowance is approximately 22 IU daily. Most multivitamins contain 35 to 40 IU of vitamin E. Thus, vitamin E intake of 400 IU or above can only be achieved with supplementation. The use of high-dose vitamin E supplements in dosages of at least 400 IU daily is common among US adults (Ford et al 2005), with approximately 11% of US adults consuming at least 400 IU of vitamin E supplements daily, and with the frequency of supplement intake increasing with age. The use of vitamin E supplements in dosages of at least 400 IU daily is even more common among health care professionals, with almost two thirds (64%) of male physicians reporting daily vitamin E intake of 400 IU or more (Muntwyler et al 2002).

Cardiovascular disease and cancer: Despite numerous randomized controlled trials assessing the potential benefits of antioxidant therapy, there is no evidence that vitamin E supplementation reduces complications of atherosclerosis (including cardiovascular events, or cardiovascular or all-cause mortality), nor is there any evidence that vitamin E supplementation decreases cancer incidence (Brown and Crowley 2005).

In the Heart Outcomes Prevention Evaluation (HOPE) trial (a randomized, double-blind, placebo-controlled, multicenter trial of more than 9500 patients at least 55 years old with vascular disease or diabetes) and a further extension of this trial (Heart Outcomes Prevention Evaluation - The Ongoing Outcomes or "HOPE TOO"), long-term vitamin E supplementation did not prevent cancer or major cardiovascular events, and instead was associated with an increased risk of congestive heart failure and hospitalizations for congestive heart failure (Yusuf et al 2000; Brown and Crowley 2005; Lon et al 2005).

In a meta-analysis of 84 randomized, placebo-controlled trials, vitamin E did not produce a statistically significant beneficial effect on all-cause mortality, cardiovascular mortality, or non-fatal myocardial infarction (Shekelle et al 2004).

Alzheimer disease and mild cognitive impairment: Various observational studies have reported an association between high vitamin E intake and a lower risk for Alzheimer disease (Grundman et al 2002; Luchsinger et al 2003; Berman and Brodaty 2004; Luchsinger and Mayeux 2004; Zandi et al 2004; Cherubini et al 2005).

Sano and colleagues conducted a double-blind, placebo-controlled, randomized, parallel-group, multicenter trial in 341 patients with Alzheimer disease of moderate severity (ie, a Clinical Dementia Rating of 2), comparing 2 years of therapy with the selective monoamine oxidase inhibitor selegiline (10 mg daily), alpha-tocopherol (vitamin E 2000 IU daily), both, or placebo (Sano et al 1997) . The primary outcome was time to occurrence of any of the following: death, institutionalization, loss of ability to perform basic activities of daily living, or development of severe dementia (ie, a Clinical Dementia Rating of 3). In unadjusted analyses there were no significant differences in outcomes between the 4 treatment arms. In analyses adjusting for the baseline score on the Mini-Mental State Examination, all of the treatment arms experienced statistically significant delays in the time to the primary outcome: the median time to reach the primary outcome was 655 days for selegiline, 670 days for vitamin E, 585 for combination therapy, and 44 days for placebo. The authors concluded that both vitamin E and selegiline "delay functional deterioration, particularly as reflected by the need for institutionalization, and should be considered for use in patients with moderate dementia."

However, even in the editorial accompanying the study by Sano and colleagues (Sano et al 1997), Drachman and Leber stated: "These conclusions [ie, that vitamin E delays functional deterioration] are encouraging but should be viewed cautiously" (Drachman and Leber 1997). Drachman and Leber noted several issues, including questions on the appropriateness of the study endpoints to assess functional decline, differences in functional abilities of study subjects at baseline making comparisons difficult, concerns that the statistical adjustments were the only basis for identifying a benefit, concerns that study results were not internally consistent and did not demonstrate any benefit on cognition, etc. Both Sano and colleagues (Sano et al 1997) and the editorialists (Drachman and Leber 1997) agreed that further studies were needed to establish the value of vitamin E in the long-term management of Alzheimer disease. Others have subsequently raised similar and additional concerns about the study (eg, Tabet et al 2000; Oken 2001).

Nevertheless, the study by Sano and colleagues (Sano et al 1997) was the primary basis for recommendations for the use of vitamin E in the treatment of Alzheimer disease in various reviews and clinical practice guidelines (American Psychiatric Association 1997; Volicer 1999; Fillit and Cummings 2000; Doody et al 2001; Berman and Brodaty 2004; Cummings 2004). Thus, for example, the American Psychiatric Association guidelines (American Psychiatric Association 1997) for the treatment of Alzheimer disease and other dementias of late life state: "Vitamin E may also be considered for patients with moderate Alzheimer's disease to slow the rate of [disease progression] and might also be beneficial earlier or later in the course of the disease… A single large well-conducted trial of vitamin E showed a significant delay in poor outcome over a 2-year period… and the agent appears to be very safe… Thus, it might be considered alone or in combination with a cholinesterase inhibitor in the treatment of Alzheimer's disease…" Similarly, the American Academy of Neurology Quality Standards Subcommittee recommended the use of vitamin E as a "guideline" reflecting "moderate clinical certainty" (Doody 2001; Doody et al 2001): "Vitamin E (1000 I.U. PO BID) should be considered in an attempt to slow progression of AD (Guideline)" (Doody et al 2001).

Still, recommendations for use of vitamin E in the treatment of Alzheimer disease were not without objections from some concerning the adequacy of the evidence to support such recommendations (eg, Drachman and Leber 1997; Small et al 1997; Tabet et al 2000; Oken 2001). A consensus statement of the American Association for Geriatric Psychiatry, the Alzheimer's Association, and the American Geriatrics Society concluded that, despite the study by Sano and colleagues, "the evidence of clinical benefit for [vitamin E] is inconclusive at this time" (Small et al 1997). Similarly, a later assessment of the Sano and colleagues study by the Cochrane Collaboration concluded that "there is insufficient evidence of efficacy of vitamin E in the treatment of people with AD" (Tabet et al 2000).

Peterson and colleagues conducted a double-blind, placebo-controlled, randomized, parallel-group, multicenter trial in a high-risk group of 769 patients with mild cognitive impairment, comparing the rate of progression to Alzheimer disease after 3 years of therapy with alpha-tocopherol (vitamin E, 2000 IU daily), donepezil (10 mg daily), or placebo (Peterson et al 2005). Inclusion criteria required amnestic mild cognitive impairment with insidious onset and gradual progression, impaired memory, a Clinical Dementia Rating of 0.5, a Mini-Mental State Examination score of 24 to 30, and age between 55 and 90 years. There was no significant difference compared with placebo in progression to Alzheimer disease in the vitamin E group or the donepezil group over 3 years. Even in subanalyses, vitamin E had no benefit in slowing the rate of progression to Alzheimer disease in a high-risk group of patients with mild cognitive impairment.

Parkinson disease: The Deprenyl and Tocopherol Antioxidative Therapy of Parkinsonism (DATATOP) Study was a randomized, double-blind, placebo-controlled, multicenter trial of selegiline (10 mg daily), alpha-tocopherol (vitamin E 2000 IU daily), both, or placebo in 800 patients with early Parkinson disease (Kieburtz et al 1994; LeWitt 1994; Parkinson Study Group 1996a; 1996b, 1998; Shoulson 1989; 1992; 1998). The DATATOP Study showed that deprenyl (selegiline), but not alpha-tocopherol (vitamin E), significantly delayed the need for levodopa in patients with Parkinson disease, although the apparent benefit of deprenyl was not associated with a reduction in subsequent levodopa-associated adverse effects or reduction in mortality, nor was it associated with any significant effect on cognitive test performance, nor was the reduced need for levodopa sustained during follow-up. Vitamin E did not show any benefit.

Amyotrophic lateral sclerosis: The German Vitamin E/ALS Study Group reported the results of a randomized, double-blind, controlled trial of megadose vitamin E (5000 IU/day) compared with placebo as add-on therapy to riluzole in 160 patients with amyotrophic lateral sclerosis (Graf et al 2005). After 18 months of follow-up, there was no difference in the primary endpoint, which was a combination of death, need for permanent assisted ventilation, or tracheotomy, nor were there any significant differences in functional abilities between the 2 groups. Megadose vitamin E was not effective in slowing progression of amyotrophic lateral sclerosis.

Vitamin E may increase all-cause mortality: In a recent meta-analysis of 19 randomized controlled trials of almost 136,000 patients, Miller and colleagues found that vitamin E may be harmful at higher dosages (400 IU or greater) (Miller et al 2005). The authors compared vitamin E with placebo treatment for at least 1 year in trials with at least 10 deaths. Study subjects varied and included healthy elderly patients, smokers, and those with cardiovascular disease, esophageal dysplasia, cataracts, large-bowel adenoma, renal failure on dialysis, Alzheimer disease, and early Parkinson disease. With lower vitamin E dosages (less than 400 IU/day), there was no benefit or detriment to vitamin E supplementation (risk ratio = 0.98, with 95% confidence interval of 0.96-1.01). With higher vitamin E dosages, there was a small but significantly higher risk of death in the supplemented group (number needed to harm = 250; risk ratio = 1.04, with a 95% confidence interval of 1.01-1.07). Nine of 11 trials assessing high-dose vitamin E (400 IU daily or greater) found an increased risk of all-cause mortality for those supplemented with vitamin E compared with controls. Furthermore, in a dose-response analysis, all-cause mortality progressively increased with increasing vitamin E dosage for dosages greater than 150 IU per day. The mechanism by which high-dose vitamin E might increase all-cause mortality is not clear, but the 2 largest high-dose vitamin E trials reporting cause-specific mortality found an increase in cardiovascular mortality. The authors concluded that high-dosage vitamin E supplementation (ie, 400 IU daily or greater) may increase all-cause mortality and should be avoided (Miller et al 2005), a position reasserted in a subsequent editorial (Hanley and Miller 2005).

Conclusion, primum no nocerum (first do no harm): The benefit of vitamin E for slowing progression of degenerative neurologic diseases has not been consistently supported and a recent meta-analysis concluded that vitamin E was associated with a small increased risk of death among individuals taking more than 400 IU daily. Weighing risks and benefits, many neurologists no longer recommend vitamin E supplementation for patients with neurodegenerative disorders, particularly in doses greater than 400 IU/day.

The use of vitamin E to slow progression of Alzheimer disease is based on only 1 trial (Sano et al 1997), which has been used to support recommendations for use of vitamin E in the treatment of patients with moderate Alzheimer disease (see above). Nevertheless, many have felt this trial does not provide adequate support for recommendations to use vitamin E in the treatment of Alzheimer disease (Drachman and Leber 1997; Small et al 1997; Tabet et al 2000; Oken 2001). In any case, the apparent benefit of high-dose vitamin E (2000 IU daily) in patients with moderate Alzheimer disease was similar to that of selegiline (10 mg daily), so the need to prescribe vitamin E on the basis of this study is not apparent when an alternative is readily available without the recognized adverse mortality risk. Those who do prescribe vitamin E in high doses for Alzheimer disease should obtain informed consent, with clear documentation of the discussion of potential risks and benefits.

Available data do not support clinical use of high-dose vitamin E for other neurodegenerative diseases, including amnestic mild cognitive impairment, Parkinson disease, and amyotrophic lateral sclerosis. The potential for harm in these conditions outweighs theoretical "antioxidant" effects in the absence of any randomized clinical trial data supporting their use.