Memantine versus donepezil in mild to moderate Alzheimer's disease: a randomized trial with magnetic resonance spectroscopy.

BACKGROUND AND PURPOSE: To compare memantine with the most prescribed cholinesterase inhibitor (donepezil) from a clinical viewpoint when administered in early phases of Alzheimer disease (AD), and to find out whether memantine may produce changes in brain metabolite concentrations in comparison with donepezil. METHODS: In this comparative rater-blinded parallel group randomized trial we recruited a consecutive sample of patients with probable mild to moderate AD. At baseline we carried out neuropsychological assessment with mini-mental, Clinical Dementia Rating Scale (CDR), Blessed Dementia Rating Scale, Alzheimer's Disease Assessment Scale, cognitive part (ADAS-cog), neuropsychiatric inventory (NPI), and disability assessment for dementia (DAD), as well as (1)H magnetic resonance spectroscopy (MRS) in several areas of the brain. Patients were randomized to receive either donepezil or memantine for 6 months. After this elapse of time we repeated the same procedures and observed the changes in clinical scales (ADAS-cog, NPI, DAD), as well as the changes in metabolite levels in every area of exploration (temporal, pre-frontal, posterior cingulated (PCG), and occipital), especially those of N-acetyl-aspartate (NAA) which is regarded as a surrogate marker of neuronal density. RESULTS: A total of sixty-three patients completed the trial. We did not see significant differences in clinical scales and metabolite levels between those on donepezil (n = 32) and those on memantine (n = 31). In general, more patients worsened than improved on either of the drugs. The changes in the NAA/creatine ratio in the PCG correlated significantly with the changes in the ADAS-cog (P = 0.004). CONCLUSIONS: Donepezil and memantine have similar modest clinical and spectroscopic effect on mild to moderate AD. MRS could be useful to monitor progression of the disease.