Positioning to prevent or reduce shoulder range of motion impairments after stroke: a meta-analysis

Objective: To assess the effectiveness of positioning on range of motion of the paretic shoulder following stroke. Data sources: We searched PubMed, CINAHL, EMBASE, Science Citation Index, PEDro, Cochrane Controlled Trails Register and article reference lists. Review methods: Randomized controlled trials reporting range of motion outcomes of shoulder positioning programmes for patients with stroke were examined independently by the two authors. Studies reporting external rotation range of motion outcomes were abstracted and their quality was rated. Results: Five studies, all published in 2000 or later, were included. Shoulder external rotation range of motion was lost by control groups (mean 11.0-18.4 degrees) and experimental (positioning) groups (mean 6.1 degrees to 19.2 degrees) in every study. The standardized mean difference between groups was -0.216 (95% confidence interval -0.573 to 0.141). These findings and the demonstration of homogeneity between and within groups do not support positioning (as practised) as an effective intervention for preventing or slowing the development of range-of-motion impairments of the paretic shoulder after stroke. Conclusion: This meta-analysis failed to support the benefit of positioning the paretic upper extremity to prevent or reduce shoulder external rotation range of motion impairments after stroke.