Study objective: Examine the effects of donepezil on sleep and rapid eye movement (REM) sleep electroencephalogram (EEG) in patients with Alzheimer disease, using polysomnography, and the correlation between REM sleep EEG parameters and cognitive scores.

Design: Randomized, double-blind, placebo-controlled design.

Settings: Two sleep research centers, University Hospital.

Participants: Thirty-five patients with mild to moderate Alzheimer disease, allocated to 2 groups: donepezil treated (n=17) and placebo treated (n=18).

Intervention: Patients were administered donepezil or placebo.

Outcome measures: Polysomnography with REM sleep EEG spectral analysis and cognitive evaluation using the Alzheimer Disease Assessment Scale, cognitive subscale, were performed at baseline and after 3 and 6 months. Slowing ratio was the ratio between slow and fast EEG frequency bands. Cognitive and sleep data were analyzed using analysis of variance. Correlations between cognitive improvement and REM sleep EEG were also calculated.

Results: REM sleep increased significantly after 3 and 6 months of donepezil treatment compared with baseline and placebo (p < .01). Overall theta (p = .04), frontal theta (p < .01) and frontal delta (p = .03) absolute power during REM sleep decreased after 6 months of donepezil treatment. The occipital slowing ratio decreased during treatment (p = .04). REM sleep overall and frontal and centroparietal alpha absolute power significantly correlated with the cognitive improvement rate on the Alzheimer Disease Assessment Scale, cognitive subscale (r = 0.75, r = 0.71, r = 0.78); p < .01).

Conclusions: Donepezil treatment enhanced REM sleep and reduced slow frequencies of REM sleep EEG, suggesting a possible action upon REM sleep-related cholinergic neurons in patients with Alzheimer disease. Furthermore, REM sleep alpha power may predict the cognitive response to donepezil.