Homocysteine and Cognitive Function in a Population-based Study of Older Adults

Jyme H. Schafer, MD, MPH; Thomas A. Glass, PhD; Karen I. Bolla, PhD; Margaret Mintz, MS; Anne E. Jedlicka, MS; Brian S. Schwartz, MD, MS


J Am Geriatr Soc. 2005;53(3):381-388. 

In This Article

Abstract and Introduction

Objectives: To assess the relations between homocysteine levels and neurobehavioral test scores representing a broad range of cognitive domains in a population-based study of older adults.
Design: Cross-sectional analysis of first-visit data of subjects in the Baltimore Memory Study, a longitudinal study.
Setting: Specific neighborhoods in Baltimore.
Participants: Participants were 1,140 randomly selected residents aged 50 to 70 with a mean age ± standard deviation of 59.3 ± 5.9; 65.9% were female, and 54.2% were white and 41.1% African-American.
Measurements: Twenty neurobehavioral test scores in eight cognitive domains.
Results: Linear regression models revealed that homocysteine was consistently and strongly associated with worse neurobehavioral test performance, in crude analysis and after control for a large set of important covariates. Associations were observed in all eight cognitive domains assessed but were strongest and most consistent in the domains of simple motor and psychomotor speed, eye-hand coordination/manual dexterity, and verbal memory and learning. On average, an increase in homocysteine levels from the 25th to the 75th percentile, all in the generally accepted normal range, was equivalent in its association with neurobehavioral test scores to an increase of 4.2 years of age. Logistic regression models revealed that, on average, for the neurobehavioral tests associated with homocysteine, subjects in the highest quartile of homocysteine levels were more than two times as likely to be in the lowest quartile of neurobehavioral test scores as those in the lowest quartile.
Conclusion: Higher homocysteine levels were associated with worse function across a broad range of cognitive domains, and the magnitude of the associations was large. The data suggest that homocysteine may be a potentially important modifiable cause of cognitive dysfunction.

There is much interest in identifying and understanding modifiable risk factors for age-related cognitive decline. Homocysteine, one such potentially modifiable risk factor, is associated with cardiovascular and cerebrovascular disease,[1,2,3] and it has been evaluated in several epidemiological studies as a risk factor for cognitive dysfunction.[4,5,6,7,8,9,10,11,12,13,14] This derivative of methionine metabolism is of interest because of its link to cardiovascular disorders and the growing literature suggesting that cardiovascular health and central nervous system health are closely related.[15,16] Although there is speculation regarding the mechanism through which homocysteine may influence cardiovascular and cognitive health, homocysteine has been shown to promote vascular endothelial cell injury and neuronal defects.[17,18,19,20]

The existing studies have reported inconsistent associations between homocysteine and cognitive function, but design features that limit comparability and causal inference have characterized the studies. Most studies have included only subjects aged 60 to 65 and older, a time when age-related cognitive decline may make it more difficult to observe associations with other risk factors.[4,5,6,7,8,9,10,11] Six of the 11 published studies had small sample sizes (<350 subjects) and thus limited power,[4,6,7,12,13,14] and several study samples had only a limited range of educational levels, sex, or race/ethnicity, limiting their external validity.[8,11,13] Another major limitation of the published literature is the extent of neurobehavioral assessment. Many studies have assessed cognitive function only with a screening test, most often the Mini-Mental State Examination (MMSE), or limited batteries that could not assess cognitive function broadly or with adequate sensitivity.[4,5,6,7,8,9,10,11,14]

Herein, associations between homocysteine levels and neurobehavioral test scores in a large, population-based sample, with large numbers of white and African-American participants, extensive evaluation of socioeconomic status (SES), an important confounding variable, and a neurobehavioral battery that assessed a broad range of cognitive domains are reported.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: