Will vitamin E kill me?

A Response by Joseph M. Raffaele, MD

Sounds like a silly question. But if you’ve read a version of the recent headline, “High dose vitamin E increases your risk of dying,” the question might be just be crossing your mind. Add to that quotes from cardiologists who are emphatically telling their patients not to take any vitamin E whatsoever and you might just be tempted to jettison it from your regimen. You might also be wondering if we have decided to eliminate it from our AM/PM Packs. Yet, like the trumpets sounding the death of hormone replacement therapy after the Women’s Health Initiative trial results hit the wires, the truth is rarely as simple as the headline.

The study, pre-released at the American Heart Association meeting and to be published in the Annals of Internal Medicine next month, is what is called a ‘meta-analysis’ of 19 randomized, controlled trials which, when all of the subjects are pooled together amounts to a total of over 135,000 subjects. This type of study is performed when the data from individual studies (some with as many as 10,000 subjects) is not sufficient to show a statistically significant small effect because the number of subjects in the study is not large enough. The fact that they needed such an incredibly large number to show a statistically significant effect demonstrates right off the bat that we are talking about a very small risk.

The main objective of the study was to see if ingestion of supplemental vitamin E causes an increase or a decrease in all-cause mortality rates in adults, i.e., does taking vitamin E increase or decrease your risk of dying of any cause. A previous meta-analysis had found there to be no statistically significant increase or decrease, so the authors thought it would be interesting to look at the mortality risk according to the dose of vitamin E taken.

When they pooled the data from all the studies, their results corroborated the previous studies with regard to overall mortality. Without taking the dose of vitamin E into account, there were 1022 deaths per 10,000 (10.2%) subjects in the control group and 1032 (10.3%) in the vitamin E group. (I told you it we’re talking a small effect.) This difference is statistically a dead heat, the result of pure chance. It means that the 10 extra deaths could just well have been in the control group if the studies were repeated. So, if you are thinking that even 10 deaths is too many, you have to remember that it could have gone either way—vitamin E ingestion didn’t have anything to do with it.

For better or for worse, epidemiologists don’t give up that easily. They found that if the studies are segregated into two dose pools, one of less than and the other greater than or equal to 400 IU intake per day, then there is a statistically significant increased risk of mortality in the higher dose group. How large is it? The excess risk is 39 more subjects out of 10,000. Put in percentage terms, this means that in the “high dose” studies 11.6% of the subjects died in the control group versus 12.1% in the group taking vitamin E. This result is barely statistically significant, but clinically significant. Looked at another way, the absolute increased risk of dying from taking high dose vitamin E is 0.5%, and the relative increased risk is only 3%. Only when you use numbers as high as 135,000 subjects can you get such miniscule effects to be statistically significant. By the way, they used the number 400 as the cut off, but the majority of the high dose treated group was taking more than 600 IU and some were taking 800 or 2000 IU.

Even so, vitamin E is often taken to improve life expectancy; therefore, even a small amount of increased mortality should trigger real concern about continuing to take it. So the message is “Don’t take vitamin E—it will shorten your life.” That’s the logic that the authors present and on the surface it is seems at least statistically reasonable. The problem is that in the studies they looked at, particularly in the high dose category, the subjects were not typical of the average healthy person taking vitamin E to improve life expectancy or to stave off the onset of a disease of aging. They were mostly sick and older with cardiovascular disease, dementia, or other chronic diseases. While the authors pay lip service to this in their discussion, the actual numbers are not presented in an easy to comprehend format. Let’s take a look.

If you tally up the deaths in the high and low dose studies separately, you will see something interesting in the table below. The death rate in the high dose control group is 11.6% versus 8.2% in the low dose control group. This is a 30% increased risk of death just by virtue of being in a higher dose trial. Clearly, the subjects in the high dose studies are sicker than those in the low dose studies. The effect dwarfs the excess risk in the treated versus controls of the high dose group (by ten-fold, 30% versus 3%)—the source of all the hullahbaloo about how vitamin E increases death. To cap it all off, there is a statistically non-significant decrease in mortality in the treated subjects in the low dose vitamin E group (7.9% vs 8.2%). In my opinion, the fact that the subjects in the high dose trials were much sicker (clearly more likely to die) than those in the lower dose trials calls into question the applicability of the findings with regard to dose in people similar to those in the low dose trials. The studies compare apples to oranges. To make the comparison regarding dose valid, the subjects in the control groups of the low dose trials should have a mortality risk that is comparable to the high dose controls, not almost 30% less.

Comparison	Mortality	Risk	Increased risk
Control high vs low dose group	11.6%	8.2%	29.7%
Treated vs control high dose	12.1%	11.6%	3%

Moreover, as is often the case with meta-analyses, the endpoints the studies evaluated and the primary endpoint of this meta-analysis are not the same. The studies were looking to see if vitamin E altered the course of these established disorders (in many cases they did), not mortality. This is an important distinction for two reasons. First, it is often the case that once a disease like coronary artery narrowing from atherosclerosis has become established, interventions known to prevent or retard the onset no longer work and can even have deleterious effects. Or as they say in medical lingo, interventions for primary prevention are often not good for secondary prevention. Second, these same preventative measures often have a significantly increased risk effect profile when given to patients with established disease and to older patients.

So what did these patients with the slightly increased mortality in the high dose group die of? The authors don’t tell us. The most likely causes are fatal bleeding episodes from the “blood thinning” effect that vitamin E can have which is desirable in most healthy patients, but not in patients with high blood pressure who are at risk of stroke because it can cause a fatal hemorrhagic stroke.

That’s the statistical analysis. The take home message is that you have to look at the details of the study to see if it is applicable to you. In this case, unless you are like the subjects in the high dose studies, you simply cannot apply the results about dose to you. If you do have established cardiovascular or Alzheimer’s disease, then you it is true that you have a 3% increased risk of dying if you take over 600 IU of synthetic alpha tocopherol a day.

That brings me to a couple of other issues. What type of vitamin E should you take? Alpha tocopherol is the most common form used in the studies, but it suffers from missing other critical components of the vitamin E family, such as gamma and delta tocopherol. The increased risk could also have come from the alpha tocopherol displacing the important other components from the cell membranes and causing an increase in oxidative stress.

The final point is that no vitamin, in either high or low dose, should be taken alone. Vitamins E and C work in an antioxidant network; they need to be balanced because they regenerate one another. When vitamin E neutralizes a free radical, it the becomes a free radical itself, albeit a less damaging one. The role of vitamin C along with alpha lipoic acid is to regenerate vitamin E so it can live to fight another day instead of becoming the enemy.

What can we conclude from this study:

Synthetic vitamin E in doses above 600 IU marignally increases the likelihood of dying in sicker people with heart and Alzheimer’s disease.
The same vitamin E in lower doses, non-statistically decreases the likelihood of dying in healthier subjects.
This study only looked at mortality. There are myriad studies that show the effectiveness of vitamin E in slowing or stablizing the course of a wide range of the diseases of aging. Ultimately, most people take supplements to improve the quality of their lives, and there is ample evidence that vitamin E can do this.

My recommendations:

Take only natural vitamin E with all of its different forms, and always take it with a full spectrum of the antioxdant networks such as is found in our AM/PM packs, which include alpha lipoic acid and vitamin C in therapeutic doses.
The optimum dose is about 500 to 800 IU depending on your state of health. Higher doses have been found to be less effective for reducing oxidative stress and may compromise immune system function in older patients.