Is there new evidence that low vitamin B12 levels cause brain atrophy?

When an observational study finds a bunch of things linked to a health condition, it’s always interesting to observe which correlation people latch onto as the important one. It’s nearly always whatever someone is selling or whatever people want to believe is the healthy message.

So, low-carb advocates favor the headlines reporting on a new study that says going veggie shrinks the brain; while dietary supplement advocates like the headlines saying that vitamin B₁₂ deficiency leads to brain atrophy and lack of B₁₂ is linked to brain shrinkage. WebMD even turned a reported correlation into evidence of a causation and wrote a dietary prescription for more vitamin B₁₂ to prevent brain volume loss in old age.

No news story reported what the study actually found.

The study in this week's news was published in the current issue of Neurology and led by Anna Vogiatzoglou, MSc, a research student with the department of physiology, anatomy and genetics at the University of Oxford. She and the co-authors are with the current team of the Oxford Project to Investigate Memory and Ageing (OPTIMA), an ongoing observational study begun in 1988 of 614 healthy seniors and 484 demented people who receive annual neurological and neuropsychological tests, including MRIs and CT scans, blood and cerebrospinal fluid analyses, and cognitive tests, and are being followed the rest of their lives.

For this current paper, Vogiatzoglou and colleagues looked at the data on 107 of the healthy seniors, average age of 73.2 years, looking for links between whole brain volume loss that accompanies aging and various markers of vitamin B₁₂ status over a five-year period. As the authors reported, studies have been inconsistent concerning an association of impaired B₁₂ status and cognitive functions among elderly, with some finding a link while others haven’t. Whether B₁₂ is a marker for general nutritional status and other social-economic factors that may contribute to mental health isn’t known. Similarly, evidence for an association between vitamin B₁₂ status and brain atrophy in the elderly is limited, they said, with total vitamin B₁₂ levels themselves proving to have low diagnostic accuracy.

Each year, the seniors had MRI scans to estimate the annual percentage of whole brain volume loss (PBVL), with the method (SIENA — the structural image evaluation using normalization of atrophy) having an accuracy rate within 0.29%. The participants also had annual physicals, which included cognitive and mental health assessments. Among the blood tests included in this paper's analysis were the total vitamin B₁₂ and folate levels; two new markers, transcobalamin (TC) and holotranscobalamin (holoTC), which are thought to reflect the biologically active parts of total vitamin B12; and two controversial tests believed to be help diagnose vitamin B₁₂ deficiency when elevated, methylmalonic acid (MMA) and total homocysteine levels.

After five years, the authors divided the total brain volume losses (PBVL) into the lowest two tertiles and the highest tertile. By not reporting all three tertiles, though, it is not possible to determine if there was a dose-response for any correlation, which might have lent an added note of caution in interpreting this study’s findings, had the associations been significant.

The authors defined the 36 seniors in the highest tertile of PBVL in their analysis as having increased brain volume loss. While the PBVL in the highest tertile was double that of the combined lower tertiles in terms of odds ratios, the absolute (actual) differences in annual percentage of total brain mass loss was 1.05% compared to 0.51% (± 0.29%). Those with the greatest brain mass loss were also about 5.5 years older (76.7 versus 71.3 years of age). After five years, 41.7% of them showed cognitive impairment, compared to 26.9% among the remaining cohort.

The authors then looked for correlations between these PBVL groups. We'll examine the actual (absolute) values associated with PBVL. Looking only at odds ratios, as we know, can lead us to think findings are more significant than they really are.

There were no statistical differences among any of the seniors’ levels of PBVL and: diabetes, educational status, blood pressure, gender, mental health scores, those on vitamin B₁₂ supplementation (oral or injection), triglyceride levels, folate, TC, homocysteine, MMA, or creatinine levels. Lab levels were all within normal ranges.

The widely-recognized genetic risk factor associated with Alzheimer’s disease is the epsilon4 allele of the apolipoprotein E gene. OPTIMA had reported in 2006 that this gene is carried in about 15% of the population and is associated with an increased risk for Alzheimer’s disease. But among elderly without dementia, the association with memory loss is weak in women. The presence of ApoE e4 polymorphism in this week's study was 41% lower among those with the greatest PBVL (13.95 versus 19.7 percent), although even this perhaps unexpected correlation was not statistically significant. They also found that “the PBVL did not differ between men and women or according to ApoEW e4 status.” This reminds us that identified genetic risk factors (correlations) are not always predictive of health outcomes.

Vitamin B₁₂ levels were 22% lower among the greatest PBVL tertile (average of 300 versus 368 pmol/L). The B₁₂ ranges between the PBVL groups, however, were separated by a mere 10 pmol/L. All levels among the seniors were well within normal, healthy ranges. No senior had a vitamin B₁₂ level below 150 pmol/L, a measure of deficiency.

This study made no attempt to analyze diets or any of the known causes for impaired absorption or inadequate utilization, or the effects of medications on vitamin B₁₂ levels. In fact, it found no correlation with PBVL and B₁₂ therapy. The evidence in this study offers no credibility to the claim that it showed vitamin B₁₂-rich diets or supplementation can help protect against brain atrophy in the elderly.

In contrast, smoking was 100% less prevalent among the highest PBVL, compared to nonsmokers (2.8% versus 5.6%), but with the small numbers of people in this study, it wasn’t a statistically significant difference, either.

Isn’t it interesting that among all of these untenable correlations, it was the slightly higher vitamin B₁₂ among the lower brain volume loss that made the headlines, and no mention was made of the smoking, for instance? Looking that the PBVL plotted with the vitamin B₁₂ levels, though, any significant linear correlation appears more wishful thinking than reality. See if the dots form a line to you:

This is reminiscent of an article on Africanized honey bees written 30 years ago today by David Roubik, an entomologist at the University of Kansas in Lawrence. He plotted the Africanized honey bees to native pollinators:

To which, Robert Hazen at the Geophysical Laboratory at Carnegie Institution of Washington, responded to Roubik’s “fanciful curve fitting” by proposing an alternative interpretation of the plot, and drawing a line of his own:

Closing thoughts

This week’s study found no tenable correlations between vitamin B₁₂ and brain atrophy. The Oxford authors were also unable to explain the mechanisms underlying a possible association between normal vitamin B₁₂ levels and cognitive impairment through a facilitation of brain atrophy. They presented several hypotheses, but cautioned “further work is required to establish whether the associations are causal.” As they said, low vitamin B₁₂ status is another risk factor (correlation) to loss of brain volume), but clinical trials of high dose vitamin B₁₂ supplementation to test if it can slow or prevent atrophy of the brain in elderly would be needed before any conclusions can be made.

But what does brain shrinkage even mean? These OPTIMA authors said they didn’t examine any relationships between atrophy of the brain and cognitive impairment. Among the public, though, it’s popular to believe that there’s a link because it seems so intuitively correct. Trouble is, it’s not necessarily so.

Australian researchers with the 20-year study at the Australian National University Centre for Mental Health Research called PATH Through Life, however, urged caution to making such assumptions, saying that a lot of long-standing beliefs about age-related brain shrinkage are not true. “Brain shrinkage, a common symptom of ageing when people hit their 60's, appears to have no impact on an individual's capacity to think or learn,” they reported. Director of the Centre, professor Helen Christensen, MPsych, Ph.D., FASSA, said:

The common belief is that the brain shrinks with age and that this shrinkage is linked to poorer memory and thinking. There is also a belief that greater education, or continued, sustained intellectual activity might allow people to better accommodate the effects of brain ageing. Our findings do not support these beliefs. It is known that the brain shrinks over the course of a person's life, although the exact trajectory is not well understood, and there are huge individual differences...

[W]e found that, on average, men aged 64 years have smaller brains than men aged 60. However, despite this shrinkage, cognitive functions - like memory, attention and speed of processing - are unaffected.

Even college soccer players were shown in a small study by researchers at the University of Cincinnati College of Medicine to have more brain volume shrinkage compared to peers who don’t play soccer, but there aren’t widespread calls to ban soccer among young people to prevent brain atrophy! Yet, there are already calls in the popular media for vitamin B₁₂ to prevent atrophy.

All together now: Correlations do not establish causation. Even less so when there's not even a tenable correlation.

© 2008 Sandy Szwarc

For more information on Vitamin B₁₂

True vitamin B₁₂ deficiency can be an important health concern, especially among dieters and people who avoid eating food from animals, those who’ve had bariatric surgeries or have other gastrointestinal disorders, the elderly and others with impaired absorption, and those taking certain medications. For more information:

Dietary Supplement Fact Sheet: Vitamin B₁₂ — Comprehensive information from the National Institutes of Health

Vitamin B₁₂ — Patient handout from the American Academy of Family Physicians

Vitamin B₁₂ — Physician guide from the Merck Manual