Objective: Depression in patients with Alzheimer's disease (AD) is common (15% to 63%) and is associated with significant morbidity and increased mortality. Our objective was to quantitatively summarize the data on the efficacy and safety of antidepressant treatment for depression complicating AD.
Method:We performed a metaanalysis of randomized, double-blind, placebo-controlled trials of antidepressants with a database search of the English literature (up to 2006) and a manual search of references in the retrieved articles. We extracted the proportion of subjects who responded and remitted, experienced adverse events (AEs), discontinued treatment due to AEs, or discontinued treatment for any reason. Cognition scores were also extracted.
Results:We included 5 studies, which involved 82 subjects treated with antidepressants and 83 subjects who received placebo treatment. Antidepressants were superior to placebo for both treatment response (odds ratio [OR] 2.32; 95% confidence interval [CI], 1.04 to 5.16) and remission of depression (OR 2.75; 95%CI, 1.13 to 6.65). There were no significant differences between the 2 groups for change in cognition (weighted mean difference -0.71, 95%CI, -3.20 to 1.79), overall dropouts (OR 0.70; 95%CI, 0.29 to 1.66) or dropout due to AEs (OR 1.41; 95%CI 0.36 to 5.54). The numbers needed to treat for one additional AD patient to respond to antidepressant treatment were 5 (95%CI, 3 to 59) and 5 (95%CI, 2 to 24) for remission of depression.
Conclusions:Antidepressant treatment for depression in AD is efficacious, with rates of discontinuation that are comparable to placebo. Nonetheless, clinicians must be vigilant regarding the potential side effects of antidepressants in this population.
Clinical Implications· Antidepressants are efficacious in the treatment of depression in AD. · Tolerability of antidepressants in AD appears to be similar to placebo. · TCAs may be associated with a decline in cognition.
Limitations· Antidepressant classes were not compared because of the limited number of published trials in AD. · Individual AEs could not be analyzed. · Individual trials were limited by small sample sizes, low medication dosages, differing outcomes of interest, and varying outcomes measures.