Clinical Briefs

With Comments from Russell H. Greenfield, MD

Vitamin E and Increased Mortality

Source: Miller ER, et al. Meta-analysis: High-dose vitamin E supplementation may increase all-cause mortality. Ann Intern Med 2005;142:in press.

Goal: To evaluate a potential dose-dependent effect of vitamin E on total mortality via meta-analysis.

Study design: Meta-analysis of randomized, controlled trials with follow-up longer than one year and occurrence of at least 10 deaths in the trial.

Subjects: A total of 19 clinical trials published between 1993 and 2004 met inclusion criteria (nine using vitamin E alone, 10 using vitamin E combined with other vitamins or minerals) and included data on 135,967 people.

Methods: Studies identified in PubMed and the Cochrane Clinical Trials Database were independently abstracted by three investigators. Authors of the original publications were contacted if required data were not available.

Results: Nine of the 11 trials utilizing high-dose vitamin E (> 400 IU/d) showed an increase in all-cause mortality compared to control. Dose-response analysis showed a statistically significant increase in total mortality with vitamin E doses > 150 IU/d.

Conclusion: High-dose (> 400 IU/d) vitamin E supplementation should be counseled against and avoided.

Study strengths: Focus on all-cause mortality to avoid issues with miscoding; influence of each individual trial was proportional to number of deaths; number of studies included in analysis; two-way factorial design to try to address use of other agents together with vitamin E.

Study weaknesses: Difficult to extrapolate results to healthy adults or children; high-dose trials often were small and performed in people with chronic illness, especially cardiovascular disease (CVD); heterogeneity of trials (e.g., age and number of subjects, chronic illness vs. healthy, vitamin E dosage, use of other supplements, differences in nutritional status, length of follow-up).

Of note: Of the 36 trials initially identified, 12 were excluded because fewer than 10 deaths were reported; mean age of participants: 47-84 years; all but three trials were placebo-controlled and double-blind; vitamin E dosage range: 16.5-2,000 IU/d (median dose: 400 IU/d); overall use of vitamin E supplementation did not affect all-cause mortality (risk ratio 1.01), but pooled risk analysis showed a slight increased risk with high- vs. low-dose vitamin E; in the dose-response analysis, total mortality progressively increased as vitamin E dosage increased above 150 IU/d; all-cause mortality decreased slightly in subjects using less than 150 IU/d; the largest studies included showed minimal effects from vitamin E.

We knew that: Three recently published meta-analyses that did not evaluate dose-response relationships reported no overall effect of vitamin E on survival; vitamin E may have pro-oxidant effects at high doses, and may inhibit glutathione S-transferases, which help detoxify both drugs and endogenous toxins; the ATBC study showed an increased risk of hemorrhagic stroke among participants using vitamin E; average dietary intake of vitamin E is approximately 14 IU/d; the few primary prevention trials examining vitamin E in healthy populations suggest benefit.

Comments: Reviews of meta-analyses do not typically appear in Alternative Medicine Alert, but this particular article was presented at a recent American Heart Association meeting and has generated widespread interest and concern. A number of published trials have called into question the wisdom of vitamin E supplementation, as did a recent review of use of vitamin E for people with CVD. The results of this meta-analysis suggest that doses of vitamin E commonly found in many multivitamins may be harmful to patients, and also that doses of 150 IU/d may offer benefits. However, if indeed there is risk associated with high-dose vitamin E supplementation, the threshold at which risk increases is far from firmly established. This paper raises serious concern about the use of high-dose vitamin E, but by itself presents insufficient evidence to recommend against any kind of vitamin E supplementation. While the lack of utility of vitamin E for people with established CVD seems clear, more studies are required to evaluate the use of vitamin E in preventing illness. Until further data become available, limit vitamin E intake to < 400 IU/d, less for those with significant illness.

What to do with this article: Give copies to your peers.

Dr. Greenfield, Medical Director, Carolinas Integrative Health Carolinas HealthCare System Charlotte, NC, Clinical Assistant Professor, School of Medicine, University of North Carolina, Chapel Hill, NC, is Executive Editor of Alternative Medicine Alert.