Plasma homocysteine and cognitive decline in older hypertensive subjects

Abstract Elevated plasma homocysteine concentrations have been associated with both cognitive impairment and dementia. However, it is unclear whether some cognitive domains are more affected than others, or if this relationship is independent of B12 and folate levels, which can also affect cognition. We examined the relationship between plasma homocysteine and cognitive decline in an older hypertensive population. 182 older people (mean age 80 years) with hypertension and without dementia, were studied at one center participating in the Study on COgnition and Prognosis in the Elderly (SCOPE). Annual cognitive assessments were performed using a computerized assessment battery and executive function tests, over a 3-5 year period (mean 44 months). Individual rates of decline on five cognitive domains were calculated for each patient. End of study plasma homocysteine, folate and B12 concentrations were measured. The relationship between homocysteine levels and cognitive decline was studied using multivariate regression models, and by comparing high versus low homocysteine quartile groups. Higher homocysteine showed an independent association with greater cognitive decline in three domains: speed of cognition (β = -27.33, p = 0.001), episodic memory (β = -1.25, p = 0.02) and executive function (β = -0.05, p = 0.04). The association with executive function was no longer significant after inclusion of folate in the regression model (β = -0.032, p = 0.22). Change in working memory and attention were not associated with plasma homocysteine, folate or B12. High homocysteine was associated with greater decline with a Cohen's d effect size of approximately 0.7 compared to low homocysteine. In a population of older hypertensive patients, higher plasma homocysteine was associated with cognitive decline.