Background: Fatigue after stroke is common and distressing to patients. The best way to treat this fatigue is uncertain. Theoretically, several different interventions may be of benefit.
Objectives: To determine whether any treatment for fatigue after stroke reduces the proportion of patients with fatigue, or fatigue severity, or both, and to determine the effect of treatment on health-related quality of life, disability, dependency and death, and whether such treatments are cost effective.
Search strategy: We searched the Cochrane Stroke Group Trials Register (last searched January 2008), the Cochrane Central Register of Controlled Trials (The Cochrane Library Issue 1, 2008), MEDLINE (1950 to February 2008), EMBASE (1980 to February 2008), CINAHL (1982 to February 2008), AMED (1985 to February 2008), PsycINFO (1967 to February 2008), Digital Dissertations (1861 to March 2008), PsycBITE (searched March 2008), PEDro (searched March 2008), and British Nursing Index (1985 to March 2008). We also searched four trials registries, scanned reference lists, performed citation tracking of included trials, and contacted experts.
Selection criteria: The review author who performed the searches scrutinised all titles and abstracts, excluded irrelevant references, and obtained references for potentially relevant studies. A second review author independently scrutinised potentially relevant studies to determine whether they fulfilled inclusion criteria. We included randomised controlled trials of any intervention in patients with stroke where fatigue was a primary or secondary endpoint.
Data collection and analysis: The two review authors who scrutinised references independently extracted data. We performed a narrative review; we had intended to perform a meta-analysis but this was not possible as the interventions were too diverse for data to be combined.
Main results: We identified three trials. One randomised 83 patients with emotional disturbance after stroke to fluoxetine or placebo. After correcting for differences in fatigue severity at baseline, there was no significant difference in fatigue between groups at follow up. The second trial randomised 31 women with subarachnoid haemorrhage to tirilazad or placebo, of whom 18 were available for follow up. There was no difference in fatigue between the two groups. The third trial investigated a chronic disease self-management programme in 1150 patients with chronic diseases, of whom 125 had had a stroke. There was no difference in fatigue at follow up between the treatment and control in the subgroup with stroke.
Authors' conclusions: There is insufficient evidence available to guide the management of fatigue after stroke. Further trials are required.