Low Vitamin D Linked to Alzheimer's Disease

Pauline Anderson

September 28, 2012

September 28, 2012 — Yet another study has linked low vitamin D levels with significant health issues — in this case, poor cognition.

In this latest systematic review of the literature, people with Alzheimer's disease (AD) had lower concentrations of vitamin D than those without AD, and better cognitive test results were linked to higher vitamin D concentrations.

Overall, the results provide sufficient evidence to warrant further investigation to determine whether a cause-and-effect relationship exists, said lead author Cynthia Balion, PhD, a clinical biochemist and associate professor, Department of Pathology and Molecular Medicine, McMaster University, Hamilton, Ontario, Canada.

"I think we have really good data now to make it clear that people need to do the interventional studies and see whether or not giving vitamin D helps people at higher risk for developing cognitive decline," Dr. Balion told Medscape Medical News.

The new review was published in the September 25 issue of Neurology.

Included Studies

Dr. Balion and colleagues searched MEDLINE, EMBASE, AMED, PsychINFO, and the Cochrane Central database for English-language studies of adults that measured vitamin D levels and included validated tests (for example, global function, executive function, psychomotor speed, attention, memory, or intelligence) as a measure of cognitive function. They accepted all recognized diagnostic criteria.

The review encompassed 37 studies, including 21 cross-sectional, 10 case-control, 1 before-after with a comparison group, and 2 prospective cohort studies, as well as 3 randomized, controlled trials (RCTs). The study sample sizes varied from 27 to 17,099 participants.

Thirty studies included only older participants, generally age 65 years or older, whereas 9 studies included only women and 2, only men. Exclusion criteria varied across studies (and included, for example, nutritional supplements, such as calcium and vitamin D; hormonal treatment; and diseases such as kidney disease, liver disease, and osteoporosis.)

All studies measured 25-hydroxyvitamin D [25(OH)D] concentrations except for 1 that measured 1,25-dihydroxyvitamin D [1.25(OH)D]; 4 studies measured both. Various vitamin D cut points were classified as deficient or insufficient (<25 nmol/L, ≥25 to 50 nmol/L, <50 nmol/L) or sufficient (≥25 nmol/L, ≥50 nmol/L, ≥50 to <75 nmol/L, >75 nmol/L).

In 14 studies, the cognition outcome included the diagnosis of dementia, which was most commonly defined according to the Diagnostic and Statistical Manual of Mental Disorders or National Institute of Neurological and Communicative Disorders and Stroke–Alzheimer's Disease and Related Disorders Association criteria. Of the 24 studies that included a test of cognitive function, the most commonly used test was the Mini-Mental State Examination (MMSE).

In most cases, the relationship between vitamin D and cognition was assessed by comparing mean vitamin D concentrations between patients diagnosed with dementia and controls or mean neuropsychological test scores between vitamin D groups.

2 Meta-Analyses

There were sufficient data to conduct 2 meta-analyses. The first compared the mean 25(OH)D concentration between AD and control groups. Six cross-sectional or case-control studies comparing data from 888 participants demonstrated a lower mean 25(OH)D concentration in patients with AD than in controls. The mean difference was −15.0 nmol/L (95% confidence interval [CI], −26.2 to −3.9 nmol/L).

The researchers found that an important determinant of the statistically significant heterogeneity was the method of 25(OH)D measurement used. The competitive protein-binding assay (CPBA) explained the heterogeneity, but this method has been withdrawn from commercial use because of accuracy issues, said Dr. Balion.

When the analysis was restricted to the 4 studies that used methods other than the CPBA, the overall difference between the AD and control groups was −6.2 nmol/L (95% CI, −10.6 to −1.8 nmol/L), with results consistent across studies. Similar results were found when studies comparing any dementia against a control group used methods other than the CPBA to measure vitamin D.

Dr. Balion stressed the need for standardization of methods of measuring 25(OH)D and noted that relevant organizations are addressing this issue. In the meantime, it's important to consider the type of analytical method being used when comparing results from different studies, she said.

Lack of true standardization is "a big problem in the field right now," commented Raj C. Shah, MD, a geriatrician and associate professor of family medicine, Rush Alzheimer's Disease Center, Rush University Medical Center, Chicago, Illinois. "It's important to realize that all measurements of vitamin D are not equal."

The second meta-analysis compared mean MMSE scores between participants with 25(OH)D concentrations less than 50 mmol/L and those with concentrations of 50 mmol/L or greater; 50 mmol/L is the most common cut-point reported in these studies and is often used to define vitamin D deficiency. Eight cross-sectional and case-control studies, which included data from 2749 participants, contributed to this analysis. Taken together, these studies showed a higher average MMSE score in participants with higher vitamin D concentrations.

The average difference in MMSE score was 1.2 (95% CI, 0.5 to 1.9), although there was statistically significant heterogeneity. None of the subgroup analyses (for example, percentage of female participants, adjustment for at least age and sex) explained this heterogeneity.

Dr. Balion noted that except for 2 studies, the average MMSE score of the groups was very similar. When 4 studies that used another type of cognitive screening tool were added, the results did not change substantially.

Of the 2 cohort studies included in the analysis, the 1 that included only men reported no significant association between vitamin D and baseline cognitive impairment; in the other, however, participants with a deficiency in vitamin D had an increased risk for substantial cognitive decline over 6 years compared with those who had sufficient concentrations of the vitamin. As for the RCTs, the only study to use a supplement of vitamin D alone found no significant between-group differences for the single cognitive measure used.

Comprehensive Analysis

According to Dr. Balion, the results of this current review differ from those of 2 earlier ones because it was more comprehensive in its search strategy and in its inclusion criteria, which resulted in more articles screened (3229 vs 99 for a previous review). It also included more studies (37 vs 5 for the previous reviews). "We decided not to be limited in what we looked at," she said.

Dr. Balion pointed out that some factors affecting vitamin D concentrations, including skin pigmentation, age, genetics, and time of sun exposure and testing, were not considered by some studies. As well, she said, reverse causation can't be ruled out because older people may have poor nutrition and spend less time exposed to sunlight, a major source of vitamin D. Despite these limitations, "I'm pretty happy saying that vitamin D plays a role in brain health," she said.

How exactly vitamin D protects the brain is not clear, but research suggests that vitamin D acts as a neurosteroid, said Dr. Balion. At the molecular level, the brain can synthesize the active form of vitamin D [1,25(OH)D] within several cell types and regions, predominantly in the hypothalamus and large neurons in the substantia nigra. Many genes are regulated by vitamin D, which contributes to neuroprotection by modulating the production of such things as nerve growth factor, and regulating neurotransmitters.

The ideal concentration of vitamin D is also not really known, said Dr. Balion. "We tried to assess that with the data that we had from all these papers, and some studies show there might be this magic cut point and other studies did not really find a cut point. Most studies aren't designed to look for that because they're not outcome studies."

However, she noted that 2 cut points are recommended worldwide: 50 and 75 nmol/L. Canada, for example, recommends the upper level for bone health.

Physicians should recommend supplements for patients not getting sufficient vitamin D, said Dr. Balion. Many jurisdictions, including Canada, recommend 600 IU of vitamin D daily for older children and adults; recommendations differ for younger children and pregnant women.

Vitamin D Testing

Ontario's universal healthcare system does not provide for vitamin D testing except for certain conditions (osteoporosis, osteopenia, rickets, malabsorption syndrome, renal disease, or taking medications that affect vitamin D metabolism) because the evidence is not yet established, said Dr. Balion.

The United States has recommendations similar to Canada's, Dr. Shah notes. Americans are becoming more aware of vitamin D's health benefits and are taking more supplements; however, although the vitamin is fat soluble, and so may be safer than some other vitamins when taken at higher doses, Dr. Shah noted that it can still lead to muscle pains and gastrointestinal tract problems.

When asked to comment on this review, Dr. Shah said, "it helps clinicians like me and researchers to understand where the state of science is in this field, and it tells us we have a lot more work to do." He noted that of the 37 papers included in the review, only a few were clinical trials.

However, that may be changing. VITAL (VITamin D and OmegA-3 TriaL), a large 5-year clinical trial sponsored by the National Institutes of Health, is randomly assigning 20,000 people across the United States. The placebo-controlled trial will investigate whether vitamin D or omega fatty acid affects various aspects of health, including cognition.

Dr. Shah raised several important issues pertaining to the analysis. For one thing, results of vitamin D studies may depend on where participants live and the time of the year that testing was done.

He also noted that most studies were probably not done in diverse populations. "I suspect that subjects were mainly Caucasian," but because the United States has an increasingly diverse population, "we need to have measures of these effects in various older adults."

Dr. Shah said that as with any meta-analysis, some publication bias probably exists, with negative studies not being published or available for review.

He also questioned whether intervening through supplementation to arrive at a target vitamin D level would affect outcomes. He used the example of high-density lipoprotein, where experts believed that raising levels would reduce risks for heart disease. Preliminary research suggests that such efforts not only might not produce the expected result but also may cause some harm in terms of adverse effects.

The study was funded by the Ontario Research Coalition of Research Institutes/Centers on Health & Aging, Ontario Ministry of Long-Term Care. Dr. Balion receives research support from the Canadian Institutes of Health Research, the Agency for Healthcare Research and Quality, the Canada Foundation for Innovation (CFI), and the Ontario Ministry of Health and Long-Term Care. For disclosures for other authors, see original paper. Dr. Shah has disclosed no relevant financial relationships.

Neurology. 2012;79:1397-1405. Abstract