Changes in metabolite ratios after treatment with rivastigmine in Alzheimer's disease: a nonrandomised controlled trial with magnetic resonance spectroscopy.

backgroundAlzheimer's disease is associated with abnormalities in the levels of some brain metabolites, including decreases in N-acetyl-aspartate (NAA) and increases in myo-inositol and choline levels. Cholinesterase inhibitors have proven modest effects on cognition in patients with mild or moderate Alzheimer's disease; however, there is little information on the effects of these drugs on metabolic parameters in the brain. Magnetic resonance spectroscopy (MRS) provides a method of determining changes in such parameters.

objectiveTo assess the effect of rivastigmine on metabolite levels in different areas of the brain, and whether changes in metabolite levels correlate with clinical outcome, in patients with Alzheimer's disease compared with untreated patients with Alzheimer's disease.

methodsTwenty-four consecutive patients with mild or moderate Alzheimer's disease were enrolled in the study and were treated with rivastigmine at a target dosage of 12 mg/day for 4 months. A comparison group of ten consecutive untreated patients with Alzheimer's disease with similar cognitive impairment to the treatment group were also enrolled. Each patient underwent assessment using the Mini-Mental State Examination (Spanish version), the Blessed Dementia Rating Scale, the Clinical Dementia Rating scale, the Interview for Deterioration in Daily living activities in Dementia, the Alzheimer's Disease Assessment Scale cognitive and noncognitive subscales, and single-voxel MRS of the frontal, parietal and occipital cortices of the brain to assess levels of brain metabolites (NAA, creatine, choline and myo-inositol) and their ratios to creatine. All assessments were performed at baseline and after 4 months of treatment with rivastigmine, and at baseline and 1 month later in the comparison group.

resultsGlobally, although there was some mean improvement, no significant changes in the cognitive and noncognitive scale scores between baseline and post-treatment assessments were seen in patients who received rivastigmine. A significant increase in the NAA/creatine ratio in the frontal cortex (1.23 at baseline vs 1.3 after treatment; p = 0.026) and in the myo-inositol/creatine ratio in the occipital cortex (0.61 vs 0.65; p = 0.009) was seen in rivastigmine-treated patients. No other significant changes in the metabolite levels or their ratios to creatine were seen in these patients. After correction for multiple comparisons, the significant effects disappeared. Only in the frontal cortex did the changes in metabolite ratios correlate with changes on the clinical scales. In the comparison group, no significant differences between the metabolite levels or ratios to creatine seen with the two scans were detected.

conclusionTreatment with rivastigmine showed modest neuronal functional recovery in the frontal cortex only (being able to reverse disease-related decreases in NAA/creatine ratio in this area but unable to affect the disease-related increase in myo-inositol/creatine ratio in any cortex). Since the modest clinical changes correlated with the small changes in the metabolite rates, MRS could be useful in monitoring response to current or future treatments for Alzheimer's disease.