High-dose B vitamin supplementation and cognitive decline in Alzheimer disease: a randomized controlled trial

Abstract

Context: Blood levels of homocysteine may be increased in Alzheimer disease (AD) and hyperhomocysteinemia may contribute to disease pathophysiology by vascular and direct neurotoxic mechanisms. Even in the absence of vitamin deficiency, homocysteine levels can be reduced by administration of high-dose supplements of folic acid and vitamins B(6) and B(12). Prior studies of B vitamins to reduce homocysteine in AD have not had sufficient size or duration to assess their effect on cognitive decline.

Objective: To determine the efficacy and safety of B vitamin supplementation in the treatment of AD.

Design, setting, and patients: A multicenter, randomized, double-blind controlled clinical trial of high-dose folate, vitamin B(6), and vitamin B(12) supplementation in 409 (of 601 screened) individuals with mild to moderate AD (Mini-Mental State Examination scores between 14 and 26, inclusive) and normal folic acid, vitamin B(12), and homocysteine levels. The study was conducted between February 20, 2003, and December 15, 2006, at clinical research sites of the Alzheimer Disease Cooperative Study located throughout the United States.

Intervention: Participants were randomly assigned to 2 groups of unequal size to increase enrollment (60% treated with high-dose supplements [5 mg/d of folate, 25 mg/d of vitamin B(6), 1 mg/d of vitamin B(12)] and 40% treated with identical placebo); duration of treatment was 18 months.

Main outcome measure: Change in the cognitive subscale of the Alzheimer Disease Assessment Scale (ADAS-cog).

Results: A total of 340 participants (202 in active treatment group and 138 in placebo group) completed the trial while taking study medication. Although the vitamin supplement regimen was effective in reducing homocysteine levels (mean [SD], -2.42 [3.35] in active treatment group vs -0.86 [2.59] in placebo group; P < .001), it had no beneficial effect on the primary cognitive measure, rate of change in ADAS-cog score during 18 months (0.372 points per month for placebo group vs 0.401 points per month for active treatment group, P = .52; 95% confidence interval of rate difference, -0.06 to 0.12; based on the intention-to-treat generalized estimating equations model), or on any secondary measures. A higher quantity of adverse events involving depression was observed in the group treated with vitamin supplements.

Conclusion: This regimen of high-dose B vitamin supplements does not slow cognitive decline in individuals with mild to moderate AD.

Trial registration: clinicaltrials.gov Identifier: NCT00056225.

Figures

Figure 1

Participant flow. The reasons for screening failures were MMSE score out of range (n=47), excluded or unstable medications (n=38), unstable medical condition (n=29), abnormal screening laboratory results (n=29);,withdrawn consent (n=12), abnormal MRI findings (n=10), caregiver issues (n=10) and diagnostic uncertainty (n=4); the reason was unknown in 13 cases. The reasons for dropping out of the study were caregiver unwillingness to continue (active 13, placebo 13), travel difficulties (active 5, placebo 7), non-compliance with study medication (active 4, placebo 6), decisions to try alternate therapy (active 6, placebo 3), concern about potential adverse effects (active 3, placebo 2), frequency of assessments (active 1, placebo 4), safety concerns (active 1, placebo 3), length of protocol (active 1, placebo 3), concern regarding placebo assignment (active 2, placebo 2), protocol violations (active 1, placebo 1), and unspecified (active 18, placebo 14); participants were allowed to specify multiple reasons. All subjects in the active treatment arm were included in the primary analysis; three subjects in the placebo group were excluded because of missing data.

Figure 2

Mean homocysteine level (μmol/L) for the two treatment groups at each visit. Error bars indicate standard error of the mean.

Figure 3

Effect of treatment on ADAScog score. Error bars indicate standard error of the mean.

Figure 4

Changes in secondary outcome measures during the course of the trial. A. CDR-SB. B. ADCS-ADL. C. NPI., D. MMSE. Error bars indicate standard error of the mean.

Figure 5

Survival to any one of five clinically relevant endpoints (loss of 4 points on the ADAScog, increase in global CDR score, loss of 15 points on the ADCS-ADL scale, institutionalization or death).