Association between homocysteine levels and cognitive profile in Alzheimer's Disease

Background: Growing evidence suggests that hyperhomocysteinemia (HHcy) constitutes a risk factor for Alzheimer's Disease (AD). The impact of HHcy on cognitive functions has mainly been investigated using screening neuropsychological tests that provide general, unspecific measures of cognitive level. Since an association between HHcy and temporo-mesial atrophy has been documented, we predicted that a fine-grained analysis of neuropsychological performance should show stronger Hcy effects on memory scores than on other cognitive scores. Objective: To determine the influence of Hcy level on cognitive profile evaluated with specific, sensitive neuropsychological tests in a wide AD cohort. Methods: 323 patients with AD were enrolled in a cross-sectional study and underwent a neuropsychological examination exploring several cognitive domains (memory, language, visuoperception, visuospatial abilities, executive function, constructional praxis, ideomotor praxis). The effects of Hcy levels and other risk factors (including cholesterol, smoking habits, triglycerides, apoEε4 allele) were analysed. Results: Generalized Linear Model detected a significant drop in performance with increasing Hcy in 6/19 measures of cognitive functions, namely, in memory performance tasks as well as in Luria's motor planning test, with effect sizes ranging 1.4%–2.8% (Eta-squared), partialling out effects of other predictors. Conclusions: HHcy was associated with poor performance in short and long-term spatial and verbal memory more than with other cognitive dysfunctions. These results support the hypothesis that medial temporal networks might be vulnerable to HHcy, consistently with data from neuroimaging studies suggesting a link in AD between temporal atrophy and HHcy; the effect on Luria's motor planning task suggests further involvement of frontal structures.