McShane · The Cochrane database of systematic reviews 2019 · systematic review and meta-analysis of randomized controlled trials · n=44 studies (almost 10,000 participants)

Memantine for dementia.

Level 1 - systematic review of randomized trials

Systematic review and meta-analysis of randomized controlled trials

PubMed 30891742 · doi:10.1002/14651858.CD003154.pub6 · record verified 2026-08-27

What was done

The authors performed a systematic review and meta-analysis of double-blind, parallel-group, placebo-controlled randomized controlled trials evaluating memantine in dementia. Searches included ALOIS (Cochrane Dementia group register), clinical trial registries, manufacturer releases/posters, and regulatory databases (FDA, EMA, NICE) up to March 25, 2018. Analyses were restricted to licensed doses (20 mg/day or 28 mg extended release) at 6 to 7 months of follow-up across four domains (global rating, cognition, activities of daily living [ADL], and behavior/mood), stratified by dementia etiology, severity, and concomitant cholinesterase inhibitor (ChEI) use.

What was found

The review included 44 trials with nearly 10,000 participants (29 trials and 7,885 participants in Alzheimer's disease [AD]). - Moderate-to-severe AD (up to 14 studies, ~3,700 participants; high-certainty evidence): Memantine showed small benefits versus placebo in clinical global rating (0.21 CIBIC+ points, 95% CI 0.14 to 0.30), cognitive function (3.11 SIB points, 95% CI 2.42 to 3.92), ADL (1.09 ADL19 points, 95% CI 0.62 to 1.64), and behavior/mood (1.84 NPI points, 95% CI 1.05 to 2.76). Overall discontinuation did not differ (RR 0.93, 95% CI 0.83 to 1.04). Agitation as an adverse event was lower with memantine (RR 0.81, 95% CI 0.66 to 0.99), though 3 trials found no benefit when treating active agitation (0.50 CMAI points, 95% CI -3.71 to 4.71). Concomitant ChEI use did not meaningfully change outcomes. - Mild AD (up to 4 studies, ~600 participants, post-hoc subgroups; moderate-certainty evidence): No difference versus placebo for cognitive function (0.21 ADAS-Cog points, 95% CI -0.95 to 1.38), ADL (-0.07 ADL23 points, 95% CI -1.80 to 1.66), behavior/mood (-0.29 NPI points, 95% CI -2.16 to 1.58), or global rating (0.09 CIBIC+ points, 95% CI -0.12 to 0.30). Memantine increased adverse event discontinuation (RR 2.12, 95% CI 1.03 to 4.39). - Mild-to-moderate vascular dementia (2 studies, ~750 participants): Small benefit in cognition (2.15 ADAS-Cog points, 95% CI 1.05 to 3.25) and behavior (0.47 NOSGER points, 95% CI 0.07 to 0.87), but no difference in global rating or ADL. - Safety overall: No difference in overall adverse events (RR 1.03, 95% CI 1.00 to 1.06) or falls; memantine increased dizziness (6.1% vs 3.9%) and headache (5.5% vs 4.3%).

Why it matters

This review confirms that memantine confers small, consistent clinical benefits in moderate-to-severe Alzheimer's disease (with or without ChEIs), while demonstrating that off-label use in mild Alzheimer's disease is unsupported by evidence.

Limits

Evidence for mild Alzheimer's disease is limited to post-hoc subgroup analyses of small sample sizes (~600 patients) rather than dedicated trials. Follow-up was restricted to 6–7 months, precluding long-term outcome assessment. Evidence for non-Alzheimer dementias (e.g., Lewy body, Parkinson's, frontotemporal) was sparse with low to very low certainty. Nearly half of the included study data required unpublished sources.

Cited by