Cognitive rehabilitation for spatial neglect following stroke

BACKGROUND: Unilateral spatial neglect is a failure to attend to one side of space. Various strategies have been used to treat these problems but evidence of their benefit has been lacking. OBJECTIVES: To determine the effects of cognitive rehabilitation for spatial neglect following stroke as measured on impairment and disability level assessments, and destination on discharge from hospital. To determine whether any effects persist at follow-up assessment. SEARCH STRATEGY: We searched the Cochrane Stroke Group Trials Register (last searched February 2001), MEDLINE (1966-December 2000), EMBASE (1980-February 2001), CINAHL (1983-January 2001), PSYCLIT and CLINPSYCH (1974-February 2001). We handsearched relevant journals, screened reference lists from relevant articles and tracked citations using SCISEARCH. SELECTION CRITERIA: Controlled trials of cognitive rehabilitation for spatial neglect in stroke. Studies with mixed patient groups were excluded unless more than 75% of their sample were stroke patients or separate stroke data were available for stroke patients. DATA COLLECTION AND ANALYSIS: Two reviewers independently selected trials, extracted data, and assessed trial quality. MAIN RESULTS: We included 15 studies with 400 participants. A large number of different outcome measures were reported. Only six studies included a measure of disability and only four (111 participants) investigated persisting effects on any outcome. There was evidence that cognitive rehabilitation resulted in significant and persisting improvements in performance on impairment level assessments, although this varied depending on the test used. There was insufficient evidence to confirm or exclude an effect of cognitive rehabilitation at the level of disability or on destination following discharge from hospital. REVIEWER'S CONCLUSIONS: There is some evidence that cognitive rehabilitation for spatial neglect improves performance on some impairment level tests but its effect on disability is unclear. Further well-designed RCTs are warranted as well as basic research to develop valid outcome measures.