BACKGROUND & OBJECTIVES:

The use of antidepressants for patients with dementia accompanied by depressive symptoms is widespread, but their clinical efficacy is uncertain. This uncertainty is due to the difficulties of interpreting the results of clinical trials. Many of the individual trials of antidepressants have been too small to provide precise estimates of the moderate benefits that might realistically be expected. Combining the information from all appropriate trials may provide a better estimate of the likely effects of treatment. This review aims to determine whether antidepressants are clinically effective and acceptable for the treatment of patients diagnosed as having depression and also diagnosed as having dementia.

SEARCH STRATEGY:

The CDCIG Specialized Register which includes records from all major medical databases and many trial databases was searched on 21 January 2001. The (long) list of search terms can be found in the main body of the review. Medical information departments of pharmaceutical companies were asked to search their databases for any relevant clinical trials. Where necessary authors of trials were approached with requests for additional information.

SELECTION CRITERIA:

All relevant unconfounded, double-blind, randomized trials comparing any antidepressant drug (as defined by the British National Formulary) with placebo, for patients diagnosed as having dementia and diagnosed as having a depression, according to established criteria.

DATA COLLECTION AND ANALYSIS:

Data were extracted independently by two of the reviewers and any differences settled by agreement.

MAIN RESULTS:

There were six included studies with a total 1077 subjects, of whom 739 met inclusion criteria. Of these, four studies (including a total of 234 subjects) reported results in sufficient detail to enter into meta-analyses. One study's results were limited to adverse results data, therefore the meta-analysis concerning efficacy was limited to three studies (Lyketsos 2000, Petracca 1996, Reifler 1989), with a total of 107 subjects. Of these three studies, two (Petracca 1996, Reifler 1989) investigated the properties of tricyclic antidepressants, drugs not commonly used in this population, and only one study investigated the properties of the more commonly used selective serotonin reuptake inhibitors (Lyketsos 2000). One of these studies (Lyketsos 2000) produced two significant differences in favour of treatment, the Cornell Scale for Depression in Dementia at 6-9 weeks (WMD -7.1, 95% CI -13.05, -1.15) and the psychiatrists' global rating (Peto OR (95% Fixed) 8.17 (1.58, 42.09)). However, the Cornell Scale for Depression in Dementia was not used in any of the other studies and no statistical differences were found with the other measures used in the meta-analysis. The meta-analysis of the number of patients suffering at least one adverse event at 6-9 weeks, using the Peto-odds ratio, showed a significant difference in favour of placebo. There were no other significant results.

REVIEWER'S CONCLUSIONS:

Available evidence offers weak support to the contention that antidepressants are an effective treatment for patients with depression and dementia. However, only three studies are included in the meta-analysis relating to efficacy, and sample sizes are small. Moreover, only one of the studies included in the analysis of efficacy data investigated the properties of the more commonly used selective serotonin reuptake inhibitors and no studies investigated the properties of newer classes of antidepressants (e.g. selective noradrenergic reuptake inhibitors). This review draws attention to the paucity of research and evidence in this area. It is not that antidepressants are necessarily ineffective but there is not much evidence to support their efficacy either. Given that they may produce serious side-effects clinicians should prescribe with due caution.