Friday November 5 2010
The risk figures are only of borderline statistical significance
“Taking vitamin E could slightly increase the risk of a particular type of stroke,” reported BBC News. It said that a study has found that for every 1,250 people taking vitamin E, there is the chance of one extra haemorrhagic stroke - bleeding in the brain.
This large and well-conducted systematic review found that taking vitamin E increased the risk of haemorrhagic stroke (the less common type of stroke) by 22%. The vitamin also slightly reduced the risk of ischaemic stroke (due to a clot in the brain). This equates to one extra haemorrhagic stroke for every 1,250 people taking vitamin E, and one ischaemic stroke prevented for every 476 people.
The researchers conclude that indiscriminate widespread use of vitamin E should be cautioned against.
These risk figures are only of borderline statistical significance, making it difficult to interpret these findings. The UK Food Standards Agency advises that a varied and balanced diet should give people all the vitamin E that they need. It also says any potential harm from taking too much vitamin E is not firmly established, but that a maximum of 540mg daily is unlikely to cause harm.
Where did the story come from?
Researchers from Harvard Medical School, Harvard School of Public Health and institutions in France and Germany carried out this review. There were no specific sources of funding. The research was published in the peer-reviewed British Medical Journal.
In general, the news stories have accurately reflected the findings of this review.
What kind of research was this?
This research investigated whether vitamin E supplementation increases the risk of stroke. The researchers were interested in its effects on strokes from any cause, and the specific types, ischaemic stroke (due to a clot) and haemorrhagic stroke (due to a bleed). The study consisted of a systematic review and meta-analysis of all the available high quality studies to date, comparing vitamin E supplementation with an inactive placebo pill on the outcome of stroke.
A systematic review is the best way to combine the results of trials to assess the effect of a treatment on a health outcome. Sometimes the trials that are included have different methods, such as different populations and different outcomes. In a good review, these differences are taken into account and adjusted for in the analyses.
What did the research involve?
The researchers searched electronic databases to identify all randomised controlled trials (RCT) that had examined the effect of vitamin E on stroke incidence (overall stroke and stroke subtypes). Studies had to have been at least a year in duration to be eligible. Only those trials examining pure vitamin E supplementation were included, and those using multivitamins or combinations of vitamins were excluded. Each of the trials was assessed for quality, following which their data were extracted.
The researchers determined the risk of stroke when taking vitamin E supplementation compared to placebo for each trial. They pooled these results, taking into account differences between the trials (their heterogeneity).
Of particular interest to them were differences between the various populations that were studied. For example, whether the study had looked at primary prevention in people who had not had a cardiovascular event, or at secondary prevention in people who had already had a heart attack or stroke. The researchers were also interested in the vitamin E dose, and how this affected the relationship between stroke and vitamin E.
What were the basic results?
Nine studies were identified including a total 118,765 people (59,357 randomised to vitamin E and 59,408 to placebo). Seven of the trials reported the outcome of total stroke, five reported on ischaemic stroke and five reported on haemorrhagic stroke. Four of the RCTs were in people treated for primary prevention, and five looked at secondary prevention in high-risk people who had previously had a cardiovascular event. Participants were above the age of 40 in all trials. Follow-up time ranged from between one and 10 years, and completion rates were high across trials.
In the analysis of all types of stroke, vitamin E was found to have no effect (pooled relative risk 0.98, 95% confidence interval [CI] 0.91 to 1.05). Vitamin E marginally decreased the risk of an ischaemic stroke by 10% (pooled relative risk 0.90, 95%, CI 0.82 to 0.99), but caused a borderline significant 22% increase in the risk of haemorrhagic stroke (pooled relative risk 1.22, 95% CI 1.00 to 1.48). The researchers calculated that this equated to one extra haemorrhagic stroke for every 1,250 people taking vitamin E, compared with one ischaemic stroke prevented for every 476 people taking vitamin E.
There was little heterogeneity between studies, and the researchers did not find that the health status of the included participants (e.g. whether they had had previous stroke or not) or the dose of vitamin E used had an effect on risk associations. Reported doses in trials varied between 300mg vitamin E daily, and 800 IU (international units) daily in one study.
How did the researchers interpret the results?
The researchers conclude that vitamin E increased the risk of haemorrhagic stroke by 22% and reduced the risk of ischaemic stroke by 10%. They say that the small reduction in the risk of ischaemic stroke is exceeded by the larger increase in the risk of a severe outcome from haemorrhagic stroke.
As a result, they recommend that the widespread use of vitamin E supplementation should be avoided.
Previous observational studies have suggested that vitamin E may have a protective effect against cardiovascular disease.
This large and well-conducted systematic review found that vitamin E does not protect against stroke when both types of stroke are combined in the analysis. It found opposing results for each subtype of stroke, however, with a 10% decrease in the risk of ischaemic stroke, and a 22% increase in the risk of a haemorrhagic stroke.
It should be noted that both of these results were only of marginal statistical significance, which suggests the possibility that these are chance findings and that no true association exists. The decrease in risk of ischaemic stroke just reached significance (CI 0.82 to 0.99), but the 22% increased risk of haemorrhagic stroke was only of borderline significance (CI 1.00 to 1.48). In absolute figures there were 223 haemorrhagic strokes among 50,334 people on vitamin E (0.5%) and 183 haemorrhagic strokes among 50,414 people on placebo (0.4%). For ischaemic stroke there were 884 ischaemic strokes among 45,670 people on vitamin E (1.9%) and 983 among the 45,733 people on placebo (2.1%). These are small absolute differences.
Vitamin E is essential for health, but supplementing the diet with vitamin E is probably unnecessary for most people. The UK Food Standards Agency (FSA) advises that the daily vitamin E requirement for men is 4mg a day, and 3mg a day for women, and that you should be able to receive all of your requirement through a varied and balanced diet. The highest dietary content is in plant oils, such as olive oil, followed by meat and dairy products, although it is difficult to be precise as to the exact amount contained in any individual food item.
Regarding supplements, the FSA reports that any potential harms from taking too much vitamin E are not firmly established. They currently advise that taking no more than 540mg daily is unlikely to cause harm.
This review added important data to what is known of the potential risks and benefits of vitamin E supplementation, though it cannot tell us about the effects of multivitamins that include vitamin E. Whether supplementation advice changes as a result of this and other studies remains to be seen.
The trials included in the review had examined different doses of vitamin E and looked at different population groups with a variety of risk factors. Although the researchers conclude that these factors did not have a significant effect on the pooled analysis, it is difficult to know whether there are certain groups of people or doses or formulations of vitamin E that may contribute higher or lower risk.
At present, the best advice that could be given to anyone concerned by these findings is that it may be better not to take pure vitamin E supplements unless they have been specifically advised to do so by medical professionals.