Treatment effects of Memantine on language in moderate to severe Alzheimer's disease patients.

Background: Language impairment is one of the most troublesome manifestations of Alzheimer's disease (AD). The objective of this post hoc analysis was to assess the treatment effects of Memantine on language in patients with moderate to severe AD, using the recently developed Severe Impairment Battery-Language (SIB-L) scale. Methods: From a combined database including four Memantine clinical trials in moderate-to-severe AD, we analyzed 801 patients with SIB-L scores of <38 and Mini-Mental State Examination scores of <15. Patients were treated with either 20 mg Memantine per day or placebo. Mean changes in SIB-L scores from baseline were calculated. For responder analyses, a change in SIB-L score greater than or equal to the SIB-L measurement error of 3.7 points was considered a clinically relevant response. Results: The mean change from baseline in SIB-L score at week 12 and weeks 24/28 (study end) significantly favored Memantine over placebo treatment (P < .0001 and P = .0182, respectively). Overall, more Memantine-treated patients than placebo-treated patients benefited from treatment. The effect was especially pronounced in patients with substantial language impairment on the SIB-L (baseline score, <= 20). At weeks 24/28, significantly more Memantine-treated patients experienced a clinically relevant improvement (25.4% vs. 10.8%, P = .0414), and significantly fewer patients experienced clinically relevant worsening (32.8% vs. 60.0%, P = .0029). Conclusions: Memantine treatment of AD patients results in significant benefits for language function. Our results suggest that it is worth considering this therapeutic option, even for AD patients with marked language impairment. (C) 2009 The Alzheimer's Association. All rights reserved.