Background: Depression is an important consequence of stroke that impacts on recovery yet is often not detected or inadequately treated. This is an update of a Cochrane review first published in 2004.

Objectives: To determine whether pharmaceutical, psychological, or electroconvulsive treatment (ECT) of depression in patients with stroke can improve outcome.

Search strategy: We searched the trials registers of the Cochrane Stroke Group (last searched October 2007) and the Cochrane Depression Anxiety and Neurosis Group (last searched February 2008). In addition, we searched the Cochrane Central Register of Controlled Trials (The Cochrane Library, Issue 1, 2008), MEDLINE (1966 to May 2006), EMBASE (1980 to May 2006), CINAHL (1982 to May 2006), PsycINFO (1967 to May 2006) and other databases. We also searched reference lists, clinical trials registers, conference proceedings and dissertation abstracts, and contacted authors, researchers and pharmaceutical companies.

Selection criteria: Randomised controlled trials comparing pharmaceutical agents with placebo, or various forms of psychotherapy or ECT with standard care (or attention control), in patients with stroke, with the intention of treating depression.

Data collection and analysis: Two review authors selected trials for inclusion and assessed methodological quality; three review authors extracted, cross-checked and entered data. Primary analyses were the prevalence of diagnosable depressive disorder at the end of treatment. Secondary outcomes included depression scores on standard scales, physical function, death, recurrent stroke and adverse effects.

Main results: Sixteen trials (17 interventions), with 1655 participants, were included in the review. Data were available for 13 pharmaceutical agents, and four trials of psychotherapy. There were no trials of ECT. The analyses were complicated by the lack of standardised diagnostic and outcome criteria, and differing analytic methods. There was some evidence of benefit of pharmacotherapy in terms of a complete remission of depression and a reduction (improvement) in scores on depression rating scales, but there was also evidence of an associated increase in adverse events. There was no evidence of benefit of psychotherapy.

Authors' conclusions: A small but significant effect of pharmacotherapy (not psychotherapy) on treating depression and reducing depressive symptoms was found, as was a significant increase in adverse events. More research is required before recommendations can be made about the routine use of such treatments.