Background: Identifying modifiable risk factors for cognitive decline can reduce burden of dementia.
Objective: We examined whether homocysteine was associated with memory performance, mediated by entorhinal volume, hippocampal volume, total gray matter volume, or white matter lesions, and moderated by APOE ɛ4 allele, B vitamins, creatinine, total cholesterol, or triglycerides.
Methods: All 204 members of the Czech Brain Aging Study with subjective cognitive decline (SCD; n = 60) or amnestic mild cognitive impairment (aMCI; n = 144) who had valid data were included. Linear regression was used, followed by conditional process modeling to examine mediation and moderation.
Results: Controlling for age, sex, and education, higher homocysteine was related to poorer memory performance overall (b = -0.03, SE = 0.01, p = 0.017) and in participants with SCD (b = -0.06, SE = 0.03, p = 0.029), but less so in aMCI (b = -0.03, SE = 0.02, p = 0.074); though sensitivity analyses revealed a significant association when sample was reduced to aMCI patients with more complete cognitive data (who were also better functioning; b = -0.04, SE = 0.02, p = 0.022). Results were unchanged in fully adjusted models. Neither mediation by markers of brain integrity nor moderation by APOE ɛ4, B vitamins, creatinine, and cardiovascular factors were significant. Memory sub-analyses revealed that results for SCD were likely driven by non-verbal memory. The homocysteine-memory relationship was significant when hippocampal volume was below the median (b = -0.04, SE = 0.02, p = 0.046), but not at/above the median (p = 0.247).
Conclusion: Higher homocysteine levels may adversely influence memory performance, which appears particularly apparent in those without cognitive impairment. Results appear to be independent of brain health, suggesting that homocysteine may represent a good target for intervention.
Keywords: Hippocampus; magnetic resonance imaging; mild cognitive impairment; neuropsychological tests.