Read the complete EBM Focus and earn CME credit.
Reference: Neurology 2014 Nov 12 early online (level 2 [mid-level] evidence)
Low levels of folate in the blood and higher levels of plasma homocysteine have been associated with poor cognitive function and increased risk for dementia and Alzheimer disease (J Nutr 2007 Jul;137(7):1789, Am J Clin Nutr 2005 Sep;82(3):636). Vitamin B supplementation in elderly populations has had inconsistent results on cognitive function (Cochrane Database Syst Rev 2008 Oct 8;(4):CD004514, Am J Clin Nutr 2014 Jun 25;100(2):657), likely due to heterogeneity in the population of interest, vitamin supplementation use, and duration of intervention. Recently, prominent news outlets reported on a secondary analysis of a randomized trial evaluating the effect of daily folic acid and vitamin B12 vs. placebo for 2 years in 3,027 elderly persons (mean age 74 years) with elevated homocysteine levels, but without cognitive impairment.
Global cognitive function was assessed at baseline and at 2 years using the Mini-Mental State Examination (MMSE). During the trial period, 26.8% of patients discontinued the intervention or were lost to follow up, but MMSE data was available for 2,556 persons. While homocysteine levels significantly decreased in persons receiving the folic acid and vitamin B12 supplement compared to placebo (p < 0.01), there was no clinically significant between-group difference in cognitive function on the MMSE (mean decrease 0.3 points vs. 0.1 points, p = 0.05). No significant between-group differences in other cognitive domains were found in a subset of patients from 1 center having additional cognitive assessments.
This study illustrates a number of interesting aspects of the application of clinical research to the practice of medicine. At first glance, the results of this trial suggests that folate and vitamin B12 may not affect cognitive function in persons with high homocysteine levels at increased risk of cognitive decline and many clinicians may choose to change practice based on that interpretation of the trial. However, because there was no clinically significant decline in the control group, it is hard to say the supplementation has no clinical value. A longer duration trial, or one that looks at more sensitive measures of cognitive decline than the MMSE, would be necessary to truly evaluate any beneficial effect on changes in cognitive function. In addition, the fact that there was significant improvement in the homocysteine level without corresponding clinical benefit highlights the limitations of using surrogate outcomes. Finally, the primary outcome of this trial was actually to determine the effects of B vitamins on risk of osteoporotic fractures (BMC Geriatr 2011 Dec 2;11:80), and their effect on cognitive function was specified as a secondary outcome. All trial participants received vitamin D3 to ensure normal vitamin D status. While vitamin D is an important factor in osteoporosis, low vitamin D levels have been associated with cognitive decline (Neurology 2012 Sep 25;79(13):1397, Arch Intern Med 2010 Jul 12;170(13):1135), and so supplementation may have confounded the results. This article is a good reminder of the importance of critically appraising research studies, particularly those that get widespread attention in the lay press.
For more information, see the Mild cognitive impairment (MCI) topics in DynaMed.