Vitamin B clue to dementia

Behind the Headlines

Thursday September 9 2010

Brain shrinkage is not the same as dementia

“Vitamin B tablets could slow and even halt the devastating march of Alzheimer's disease,” The Daily Telegraph reported. According to the newspaper, large daily doses of vitamin B can halve the rate of brain shrinkage, a process that can precede Alzheimer’s disease and dementia.

This story is based on a well-conducted two-year trial, which compared vitamin B pills with inactive placebo pills in 271 elderly people with mild memory problems. The study found that those given vitamin B experienced brain shrinkage (atrophy) 30% slower than those given inactive tablets. However, slower brain shrinkage may not necessarily lead to any improvement in symptoms.

This research does not show that vitamin B can prevent Alzheimer’s disease or dementia because there is no evidence to confirm that slower brain shrinkage will lead to benefits for people with early dementia symptoms. Nevertheless, these results are promising and clearly warrant more research.


Where did the story come from?

The study was carried out by researchers from the University of Oxford, the Oxford Radcliffe Hospitals NHS Trust and the University of Oslo in Norway.

Should I take vitamin B supplements?

This was early research and cannot yet tell us whether vitamin B can slow the progress of dementia.


If you have been advised by your doctor to take vitamin B, continue to do so, although talk to your GP before changing the dose. Harms have been noted with long-term use of high doses and vitamin B should be avoided with some conditions. If you wish to use vitamin B supplements as a precautionary measure, particularly at doses over the RDA (recommended daily allowance), consult your GP to make sure it is safe to do so.


Good dietary sources of vitamin B include leafy green vegetables, meat, eggs and yeast spreads.

The study received funding from a number of sources, including the Charles Wolfson Charitable Trust, the Medical Research Council, the Alzheimer’s Research Trust and the National Institute for Health Research. It was published in PLoS One, the peer-reviewed journal of the Public Library of Science.

Newspapers generally covered the research in a balanced way, although the headlines are overly optimistic in proclaiming that vitamin B will delay or beat Alzheimer’s disease. A diagnosis of Alzheimer’s is based on specific, characteristic clinical features and the exclusion of other causes of cognitive impairment. However, this research only assessed an outcome of brain shrinkage, which is not necessarily the same thing. The functional effects of reducing brain shrinkage were not investigated and it is an extrapolation to conclude that B vitamins improved cognitive health or protected against Alzheimer’s disease.


What kind of research was this?

Brain atrophy, which describes the loss of neurones and their connections, can be caused by a number of diseases. Some degree of atrophy and subsequent brain shrinkage is common with old age, even in people who are cognitively healthy. However, this brain shrinkage is accelerated in people with mild cognitive impairment and even faster in those who ultimately progress from mild cognitive impairment to Alzheimer’s disease. A range of factors has been implicated in affecting the rate of brain atrophy, one of which is high levels of an amino acid in the blood called homocysteine (tHcy). Studies have shown that raised levels of tHcy increase the risk of Alzheimer’s disease.

In this randomised controlled trial, the researchers investigated the role of vitamin B in regulating levels of tHcy. They specifically wanted to test whether lowering tHcy through giving high doses of vitamin B for two years could reduce the rate of brain atrophy in people with pre-existing mild cognitive impairment.


What did the research involve?

Volunteers aged 70 and over with concerns about their memory were recruited in the Oxford area, through radio and newspaper advertisements, between April 2004 and November 2006. It was specified that volunteers should have a diagnosis of mild cognitive impairment, defined using specific criteria. These included a concern about memory that did not interfere with activities of daily living and pre-specified scores on some cognitive scales assessing word recall and fluency. The study excluded people with a diagnosis of dementia, who were taking anti-dementia drugs or who had active cancer. People taking folic acid and vitamin B6 or B12 above certain doses were also excluded.

Every six months, the volunteers were randomly assigned to receive either high-dose oral vitamin B tablets (0.8 mg folic acid, 0.5 mg vitamin B12 and 20 mg vitamin B6) or placebo pills during the two-year period. The participants, their partners and all staff directly involved in the study were unaware which pills were being received. The double blind nature of the study was important as it eliminated potential biases associated with the patients’ or researchers’ knowledge of which treatment was being received. MRI scans were performed at the start of the study and again after two years. The researchers used these to calculate the rate of brain atrophy each year.

A total of 271 people were randomly assigned a treatment, although five did not start the study. A similar proportion from each treatment group dropped out along the course of the study. The researchers measured adherence to the study treatments by counting returned tablets. For the main analysis of brain shrinkage, the researchers used data on 168 people (85 receiving active treatment and 83 receiving placebo) who had completed an MRI at both the start and at follow-up. The analyses took into account a variety of factors that may be linked to brain atrophy or use of vitamin B, which the researchers had tested and found to be important. These factors were age, blood pressure, initial brain volume and concentration of tHcy at the start of the study.


What were the basic results?

Treatment with vitamin B tablets had notable effects on the levels of tHcy in the blood, reducing it by 22.5%. Levels of tHcy increased by 7.7% in the placebo group. Overall, treatment with B vitamins for a period of 24 months led to a reduction in the rate of brain atrophy. After the age of the participants was taken into account, the rate of shrinkage in people receiving the vitamins was 30% less than in the placebo group (0.76% shrinkage and 1.08% shrinkage respectively). The effect was greater in people who were more compliant with taking their medication and in those who started with the highest levels of tHcy. The researchers also found that, overall, the safety of vitamins was good with no adverse events.


How did the researchers interpret the results?

The researchers concluded that they have shown that a “simple and safe treatment” can slow down the accelerated rate of brain atrophy in people with mild cognitive impairment.



This is an important but early study in establishing the effects of vitamin B on the stages of brain atrophy that precede Alzheimer’s disease. It assessed the effects of the vitamin on the rate of brain shrinkage, a process that has been linked to old age, mild cognitive impairment and Alzheimer’s disease in other studies. Although other studies have found that the rate of brain atrophy is linked to cognitive decline, this particular study did not assess whether the participants’ brain changes translated into changes of cognitive ability or memory.

This was a well-conducted, albeit small, study. It was a randomised controlled trial, which is the most appropriate way to assess the effects of a new treatment. No study is perfect, though, and the researchers highlighted some shortcomings:

  • The treatment was a combination of three B vitamins, so the researchers could not determine whether these have different effects individually.
  • The study was not set up to assess the effects of treatment on cognition, but only on the rate of change in brain measurements.

This study will pave the way for future research into the use of vitamin B to prevent Alzheimer’s disease. Based on the evidence gathered so far, it is too early to claim that vitamin B can prevent clinical disease, but these results are promising. More research will undoubtedly follow.

Analysis by Bazian

Edited by NHS Choices

Links to the headlines

Vitamin B is revolutionary new weapon against Alzheimer's Disease. The Daily Telegraph, September 9 2010

Vitamin B 'puts off Alzheimer's'. BBC News, September 9 2010

Vitamin B 'cocktail' to beat Alzheimer's. The Sun, September 9 2010

Vitamin B 'may slow' dementia. Daily Mirror, September 9 2010

Vitamin B claim on Alzheimer's. Financial Times, September 9 2010

Daily vitamin pill could reduce dementia's effects by up to 50 per cent. The Independent, September 9 2010

Links to the science

David Smith A, Smith SM, de Jager CA et al. Homocysteine-Lowering by B Vitamins Slows the Rate of Accelerated Brain Atrophy in Mild Cognitive Impairment: A Randomized Controlled Trial. PLoS One 5(9): e12244


How helpful is this page?

Average rating

Based on 39 ratings

All ratings

Add your rating


The 4 comments posted are personal views. Any information they give has not been checked and may not be accurate.

EvanHarris said on 16 September 2010

I want to tank Prof Smith and Dr Refsum for their response which I will consider carefully.

Report this content as offensive or unsuitable

DavidSm said on 11 September 2010

Continued from previous comment.
The apparent lack of Intention to Treat (ITT) analysis is to some extent a matter of opinion. ITT states that you include all randomized participants that potentially can be included in the statistical analyses, independent of what has happened with the participant after randomization, and independent of whether the participant followed the protocol. Since measurement of rate of atrophy requires both a base-line and follow-up scan, the ITT population in our MRI trial must be limited to those with both scans. Imputation of follow-up scan data cannot be advised on this occasion on scientific grounds. So for this particular trial, ITT analysis means that we included all subjects with both scans irrespective of whether they were compliant in terms of taking the assigned tablets, or started taking B vitamin supplements, or otherwise did not adhere to protocol.
We hope that this answers your concerns.
David Smith and Helga Refsum

Report this content as offensive or unsuitable

DavidSm said on 11 September 2010

The authors would like to thank NHS Choices for an excellent comment on the paper.
In reply to Dr Harris who refers to Dr Heneghan’s comment that there was a high drop-out rate. This was not the case. The trial was designed to answer two questions: the MRC funding was to see how feasible is it to recruit and follow subjects with MCI in one centre, with a daily tablet treatment and regular cognitive assessments; secondly, within this cohort a subgroup would have MRI scans at the start and finish so that the rate of brain atrophy could be assessed. The MRI scans were entirely voluntary and this was allowed for by ”minimization”, to make sure that the placebo and B vitamin groups had the same number of subjects who had consented to MRI scans. We also minimized for age and sex and cognitive score at baseline. Minimization is a procedure in the design of a study to ensure that the placebo and active treatment group are fairly similar at baseline. This is critical in small studies where randomization alone could lead to quite substantial differences between the two groups, rendering the trial useless.
Thus the MRI part of the trial was a separate arm and here the drop-out was low (see Fig. 1 in the paper). Of those 187 that underwent MRI at baseline, 180 completed the second scan, i.e., a mere 3.7 % drop out. However, not all scans were technically suitable for use. One example that renders an MRI scan of little value is if the patient moves during the acquisition. So out of all scans, we obtained 168 useful scan pairs (85 in the B vitamin group, 83 in the placebo group). However, even if we consider every subject lost to analyses as a drop out (true drop out due to disease etc + technical drop out), there was only a 10.2% drop-out over 2y. This proportion is not unusual in trials with elderly people.
Continued in next comment.

Report this content as offensive or unsuitable

EvanHarris said on 10 September 2010

I would be interested to know what the Bazian or NHS Choices team think about the problem of the high drop out and the apparent lack of an Intention to Treat analysis which would avoid bias introduced by differential drop out.
This was raised here and I cant see how this concern has been dealt with anywhere.
This article ( cites the concern and gives the authors response as "dealt with by minimization" but my understanding of minimization ( is that it can be used to help get more representative samples when the sample sizes are small not to help adjust for any bias introduced by high drop-outs.

Please let me know
a) If this was dealt with in the paper (it does not seem to be)
b) Why, even though you mention it, you do not identify it as a potential problem in your analysis
c) What is your view of the apparent mention of minimisation as a means to deal with lack of ITT analysis.


Report this content as offensive or unsuitable