Older patients with very low levels of vitamin D have about a 122% increased risk for dementia compared with those with higher levels, according to a large, prospective, population-based study.
The study provides “robust evidence” of the link between vitamin D and cognition and adds important new information to the association, said study author David J. Llewellyn, PhD, senior research fellow in clinical epidemiology, University of Exeter, United Kingdom.
Although earlier research had also uncovered this link, “we were able with much greater accuracy to sort of chart the level of vitamin D that seems to be relevant to a dementia risk,” said Dr. Llewellyn. ” It gives us valuable clues as to the kind of trials we should be doing next, and who we should be treating and with how much vitamin D.”
Low vitamin D levels “should ring alarm bells” that patients are at high risk for dementia, said Dr. Llewellyn.
The study was published online August 6 in Neurology.
The analysis included 1658 ambulatory and relatively healthy participants from the Cardiovascular Health Study in 4 US communities (Forsyth County, North Carolina; Sacramento County, California; Washington County, Maryland; and Pittsburgh, Pennsylvania).
Researchers obtained blood samples in 1992–1993 and in 2008; they measured serum 25-hydroxyvitamin D (25[OH]D) concentrations. They classified these samples as follows: less than 25 nmol/L (severely deficient), 25 nmol/L or greater to less than 50 nmol/L (deficient), and 50 nmol/L or greater (sufficient).
Investigators assessed cognition through repeat MRI examinations, medical records, questionnaires, and annual cognitive assessments over about 6 years.
During 9317.5 person-years of follow-up, 171 participants developed any type of dementia and 102 developed Alzheimer’s disease (AD). The risk of developing both was significantly higher in persons who were 25(OH)D deficient or severely deficient.
In a model adjusting for age and season of sampling, participants who were vitamin D deficient had a 51% increased risk for all-cause dementia (hazard ratio [HR], 1.51; 95% confidence interval [CI], 1.06 – 2.16; P = .002) compared with those with sufficient vitamin D. Those who were severely deficient had about a 122% increased risk (HR, 2.22; 95% CI, 1.23 – 4.02).
After also adjustment for education, sex, body mass index (BMI), smoking, alcohol consumption, and depressive symptoms, the HRs for all-cause dementia were 1.53 in those who were vitamin D deficient and 2.25 for those who were severely deficient.
This finding was “surprising” as it was “much stronger” than the 60% increased risk his research group found earlier (Arch Intern Med. 2010;170:1135-1141), although that study looked at “new cognitive problems” such as memory decline, which is a “cruder” way of assessing cognition, said Dr. Llewellyn.
In this new research, the strength of the association was sustained for participants with incident AD.
The association withstood many additional tests. A secondary analysis in which serum 25(OH)D concentrations were analyzed as a continuous variable provided a similar pattern of results, as did other analyses that adjusted for diabetes, hypertension, and ethnicity (for ethnicity, the results were somewhat altered, but the overall pattern didn’t change).
When participants who developed any dementia within a year of baseline were excluded, researchers also found an association between low vitamin D levels and dementia. This, said Dr. Llewellyn, indicates that the findings aren’t due to participants being in the early stages of dementia.
If that were the case, he said, “the association should start to get weaker when we do those analyses, but if anything, it’s probably the reverse; it was probably a little bit clearer, or at least the association certainly remained robust.”
The relationship was also not due to older people staying indoors or having a poor diet, which would lower their vitamin D levels.
“Our study was restricted to people who were pretty healthy at baseline,” said Dr. Llewellyn. “None had a stroke or cardiovascular disease; they were certainly not severely impaired or immobile.
“Low vitamin D levels likely affect cognition through both neurodegenerative and vascular mechanisms,” said the authors. They noted that vitamin D receptors are expressed in areas of the brain involved in memory, such as the hippocampus and dentate gyrus; that the active form of vitamin D regulates neurotrophin expression, such as nerve growth factor; and that vitamin D reduces amyloid-induced cytotoxicity and apoptosis in primary cortical neurons.
The findings suggest that the optimal vitamin D level to prevent dementia is 50 nmol/L. Others in the field argue that a higher level — 75 nmol/L — is better, “but our data don’t support that,” said Dr. Llewellyn.
He stressed that lower levels might protect against other health outcomes; for example, about 25 nmol/L may help promote bone health or prevent rickets.
Should doctors recommend that older adults have their vitamin D checked? This, said Dr. Llewellyn, is a “controversial issue” because it involves “one of most expensive tests available to primary care physicians.”
Some people can’t synthesize enough vitamin D from sun exposure in winter months. Does this put them at risk for dementia? That’s unclear, although Dr. Llewellyn knows of research at the University of Edinburgh that found a trend connecting higher latitude and increased dementia risk in residents of the northern hemisphere, including Scotland and Scandinavian countries.
At this point, Dr. Llewellyn said he “certainly wouldn’t recommend” taking vitamin D supplements to prevent dementia. There’s better evidence for supplements preventing other outcomes, for example, bone problems, and although he doesn’t necessarily agree, some scientists believe there’s a tolerable upper level of vitamin D.
“It’s too early to tell whether supplementation is going to help or not; it might help and that’s why we need trials to dig into that,” he said. “Let’s do the research and see whether we can slow down the progression of dementia with supplements in trials.”
The Alzheimer’s Association agrees that some sort of clinical trial — whether it’s with vitamin D supplements, increased sunlight exposure, or a vitamin D–enriched diet — is needed to test the effect on dementia, said Keith N. Fargo, PhD, director of scientific programs and outreach, Alzheimer’s Association, which helped fund the study.
Although the study results show a “pretty compelling link” between low vitamin D and later development of AD, and this “give us good-quality confirmation of existing findings,” the study only shows an association or correlation, said Dr. Fargo.
“All we can say now is that there is a link between the 2, but we don’t know why that link exists and we don’t know if change in vitamin D levels would change your risk ultimately for Alzheimer’s disease.”
What’s also unknown at this point is the appropriate blood level of vitamin D that might protect against the development of dementia, added Dr. Fargo.
As for supplements, Dr. Fargo pointed out that some studies have shown that vitamin E is beneficial in terms of cognition but that vitamin E supplements can increase risk for death.
“So it might not be a good idea for people to rush out, based on this one study, and start taking a lot of vitamin D.”
A limitation of the study is that it didn’t include Hispanics. It also excluded those with cardiovascular disease and stroke at baseline, so it was impossible to investigate the relationship between vitamin D concentrations and incident vascular dementia due to lack of statistical power.
In addition to the Alzheimer’s Association, the National Heart, Lung, and Blood Institute; the National Institute of Neurological Disorders and Stroke; the National Institute on Aging; the Mary Kinross Charitable Trust; the James Tudor Foundation; the Halpin Trust; the Age Related Diseases and Health Trust; the Norman Family Charitable Trust; and the UK National Institute for Health Research also supported the study. The authors have disclosed no relevant financial relationships.
Neurology. Published online August 6, 2014. Abstract
August 06, 2014