+ Abstract
Background: In patients with Alzheimer's disease (AD), a declining ability to communicate has been associated with increased caregiver stress. In a previously reported, randomized trial in moderate AD (NCT00469456), memantine was superior to placebo at improving functional communication, as recognized by caregivers. Here we focus on patients from that trial who experienced a worsening on two measures of functional communication: the primary study endpoint, an assessment of patients using the Functional Linguistic Communication Inventory (FLCI), and the secondary study endpoint, an assessment of caregivers using the combined Social Communication and Communication of Basic Needs subscales of the American Speech-Language-Hearing Association - Functional Assessment of Communication Skills of Adults scale (ASHA-FACS). Methods: Native English-speaking outpatients with AD (MMSE range: 10-19) participated in a 12-week, international, double-blind, randomized study of memantine (20 mg/day; ITT n = 133) versus placebo (ITT n = 124). Concurrent cholinesterase inhibitor treatment, stable throughout the study, was permitted but not required. In this post hoc analysis (LOCF), we selected patients whose changes on the FLCI or the ASHA-FACS subscales indicated any worsening (<0, both measures), greater-than-mild worsening (decline of more than 0.5 standard deviations: FLCI, <-3; ASHA-FACS, <-10), or greater-than-moderate worsening (decline of more than 1.0 standard deviation: FLCI, <-6; ASHA-FACS, <-20). We then used Generalized Estimating Equations to compare treatment groups in terms of proportions of patients who experienced each degree of worsening at endpoint and overall (i.e., throughout the trial). Results: On the FLCI, no significant difference was observed between memantine and placebo in the proportion of patients who experienced any decline (endpoint and overall); however, a significantly higher percentage of patients in the placebo group experienced greater-than-mild worsening at endpoint (P = 0.032) and overall (P = 0.013) and greater-than-moderate worsening at endpoint (P = 0.039) but not overall (P = 0.081). On the ASHA-FACS, significantly more caregivers of patients taking placebo relative to memantine reported any worsening overall (P = 0.013) but not at endpoint (P = 0.079); significantly greater-than-mild worsening at endpoint (P = 0.012) and overall (P = 0.018), and significantly greater-than-moderate worsening at endpoint (P = 0.011), though not overall (P = 0.133). Conclusions: Memantine treatment of patients with moderate AD may be associated with a prevention of worsening in functional communication.