B Vitamins for Prevention of Cognitive Decline: Title and subTitle BreakInsufficient Evidence to Justify Treatment

Stroke and dementia are among the most common diseases affecting the brain in older persons and account for most cases of disability requiring nursing care in this age group.¹ The incidence of these diseases increases exponentially with age. Consequently, improvements in life expectancy have resulted in a substantial increase in the absolute number of individuals with dementia and cognitive impairment in recent decades. Dementia is characterized by an insidious, slowly progressive memory loss with alteration of higher intellectual function and cognitive abilities. Among the subtypes of dementia, Alzheimer disease and vascular dementia have distinct clinical and pathological features, but these 2 disorders frequently coexist and the combination is associated with a greater severity of cognitive impairment.²

Population-based studies, such as the Rotterdam study,² have demonstrated that indicators of atherosclerosis were associated with Alzheimer disease and that the prevalence of Alzheimer disease increased with the degree of atherosclerosis. The odds ratio for Alzheimer disease in those patients with severe atherosclerosis was 3.0 (95% confidence interval, 1.5-6.0) compared with those without atherosclerosis. Participants with at least 1 apolipoprotein E (ApoE) ε4 allele and atherosclerosis had a nearly 5-fold increased risk of Alzheimer disease compared with those with no ε4 allele² ; cigarette smokers had twice the risk of Alzheimer disease compared with nonsmokers³ ; and individuals with diabetes mellitus had 3 times the risk of Alzheimer disease compared with those without diabetes.⁴

The hypothesis that homocysteine may be a risk factor for Alzheimer disease and that B vitamins might be neuroprotective was prompted by the observation that patients with a histological diagnosis of Alzheimer disease, irrespective of whether there was concomitant histological evidence of cerebrovascular disease, had higher plasma levels of homocysteine than did age-matched controls.⁵ Homocysteine is a potentially harmful sulfur-containing amino acid derived from methionine and has been previously linked with an increased risk of cardiovascular disease.^{6 - 7} B vitamins, such as folic acid and vitamin B₁₂, are needed to transform homocysteine into the important one carbon donor S-adenosylmethione (required for methylation reactions) and into glutathione (required to protect the cells from oxidative stress).⁸ The vitamin B₁₂–dependent reaction by which homocysteine is converted to methionine (and thereby removed from the bloodstream) is inactivated by oxidation.⁸

More convincing evidence for the importance of increased homocysteine levels and risk of dementia was provided by an 8-year follow-up of 1092 dementia-free elderly participants in the Framingham study,⁹ which reported that increased homocysteine levels (>14 μmol/L) were associated with a 2-fold higher risk of dementia and Alzheimer disease. These associations persisted after adjustment for age, sex, ApoE ε4 genotype, and other vascular risk factors. Subsequently, several additional cohort studies^{10 - 12} have reported associations with higher homocysteine levels, in conjunction with low levels of folate or vitamin B₁₂, or both of these vitamins, and these individuals were reported to have more rapid rates of cognitive decline.

The biological plausibility of such an association is through the putative etiologies of Alzheimer disease, accumulation of β-amyloid peptide and neurofibrillary tangles (clumps of a protein called tau) in the brain. Indeed, increased homocysteine levels and deficiency of B vitamins have been linked with the deposition of amyloid plaques (by means of microvascular disease) and of neurofibrillary tangles (by means of oxidative stress and hyperphosphorylation of tau proteins) in human cell culture systems.¹³ In light of these data and the fact that homocysteine levels are easily lowered by dietary supplementation with folic acid and vitamin B₁₂,¹⁴ there has been considerable interest in trials assessing whether dietary supplementation with B vitamins slows the rate of cognitive decline.¹⁵

Although a systematic review of 14 small trials¹⁵ reported that dietary supplementation with B vitamins had no beneficial effects among individuals with normal or impaired cognition, few studies had sufficient size or duration to assess effect on cognitive decline. With the exception of the FACIT trial,¹⁶ which demonstrated a significant reduction in the rate of cognitive decline associated with folic acid in older individuals without dementia and with increased homocysteine levels, no previous trials of B vitamins have reported any beneficial effects on cognitive function. In the FACIT trial, 818 healthy older participants living in the Netherlands were randomly assigned to receive 0.8 mg of folic acid for a 3-year period, and folic acid supplementation significantly improved cognitive domains of memory, information processing speed, and sensorimotor speed. However, other trials^{17 - 20} assessing the effects of B vitamin supplementation on cognitive function failed to provide any evidence of benefit associated with B vitamins.

There are several possible reasons for the failure of trials to demonstrate any beneficial effects on cognitive function, including (1) patients had significant cognitive impairment before starting therapy, (2) an insufficient number of participants were enrolled in the trial, (3) the duration of treatment was too short to be able to detect beneficial effects, (4) benefits were attenuated following folic acid fortification in North America, (5) a summary rather than a global measure of cognitive decline was used, or (6) treatment truly has no effect on cognitive function.

In this issue of JAMA, Aisen and colleagues from the Alzheimer Disease Cooperative Study²¹ examined the effect of dietary supplementation with B vitamins for an 18-month period in 409 participants with mild to moderate Alzheimer disease in the United States. Participants with a Mini-Mental State Examination score between 14 and 26 (maximum score, 30) and normal plasma levels of folate, vitamin B₁₂, and homocysteine were randomized to receive either 5 mg/d of folic acid, 1 mg/d of vitamin B₁₂, and 25 mg/d of vitamin B₆ or placebo. Cognitive function was assessed by using the cognitive function subset of the Alzheimer Disease Assessment Scale. Despite a 31% reduction in homocysteine levels associated with B vitamin treatment, the trial found no significant effects on the primary end point of a delay in the rate of cognitive decline, and no significant reduction in any of its secondary outcomes, including clinical status, function, and behavior.

In contrast with previous trials^{16 - 20} that assessed the effects of B vitamins on cognitive function in healthy older individuals, Aisen et al²¹ examined the effects of B vitamins in a population with significant cognitive impairment at entry into the trial. Although such a population might have greater statistical power to detect any effects of treatment, it is also possible that individuals with established cognitive impairment may be refractory to treatment. However, the authors also reported that depressive symptoms were more common among those patients allocated to B vitamin supplementation compared with placebo. The excess of depressive symptoms associated with B vitamins may be a chance finding, because there were no differences in the use of antidepressants by treatment allocation.

The precise reasons the Alzheimer Disease Cooperative Study failed to detect any beneficial effect of B vitamins on the rate of cognitive decline remain unclear. However, these results provide further support for the conclusion that B vitamins are not effective in slowing cognitive decline in individuals with normal folate and vitamin B₁₂ levels in societies with folate-enriched foods.

An important factor in interpreting the results of the trial by Aisen et al and other similar studies is the public policy on folic fortification in a given country. Mandatory folic acid fortification was introduced in the United States and Canada in 1998 for the prevention of neural tube defects. Folic acid fortification has resulted in more than a doubling in the mean serum folate concentrations but serum vitamin B₁₂ concentrations did not change appreciably.²² Fortification has been remarkably effective at reducing the risk of neural tube defects in infants.²³ However, some countries, such as England, delayed fortification because of concerns about “masking” of vitamin B₁₂ deficiency in older adults or acceleration of neurological disease associated with vitamin B₁₂ deficiency by exposing older adults with vitamin B₁₂ deficiency to very high levels of folic acid.²³ Unequivocal demonstration of any beneficial or hazardous effects of B vitamins on cognitive function, vascular disease and nonvascular disease from the large-scale homocysteine-lowering trials would have important implications for public health policy.

Any theoretical concerns about hazards of folate supplementation can be alleviated by ensuring an adequate dose of vitamin B₁₂ (>500 μg) in multivitamin supplements containing high-dose folic acid (>400 μg). However, until and unless new data suggest otherwise, there is insufficient evidence to justify routine use of homocysteine-lowering vitamin supplements for the prevention of Alzheimer disease and cognitive decline among individuals with normal vitamin status.