Assistive technologies: can they contribute to rehabilitation of the upper limb after stroke?

To systematically identify, review, and explore the evidence for use of assistive technologies (ATs) in poststroke upper limb rehabilitation.

AMED, CINAHL, Cochrane Library, Compendex, CSA Illumina, EMBASE, MEDLINE, PEDro, PyscINFO, and Web of Science were last searched in September 2011.

Two independent researchers screened for inclusion criteria (adult poststroke subjects, upper limb rehabilitation with an AT). The risk of bias was assessed. Randomized controlled trials of poststroke subjects with baseline equivalence as assessed by blinded assessors were selected for data extraction.

Details of subjects, experimental and control treatments, and all outcomes were recorded in a spreadsheet.

These data were used to calculate effect sizes for all outcome measures. Impairment measures ranged from -.39 (95% confidence interval [CI], -1.14 to .62) to 1.46 (95% CI, .72-2.20). Measures of activity effect sizes were from .04 (95% CI, -.35 to .44) to .93 (95% CI, -.39 to 2.25); for Motor Activity Log, from .07 (95% CI, -.66 to .80) to 1.24 (95% CI, .47-2.01); and for participation, from -3.32 (95% CI, -4.52 to 2.11) to 1.78 (95% CI, 0-3.56).

AT treatments appear to give modest additional benefit when compared with usual care or in addition to usual care. This is most apparent for subjects early poststroke with 2 caveats: high-intensity constraint-induced movement therapy and electrical stimulation exclusively to the shoulder appear detrimental. The heterogeneity of treatment parameters and population characteristics precludes specific recommendations. Research would benefit from modeling studies to explicitly define criteria of population, intervention, comparator, and outcomes for effective treatments before the development of efficiently integrated care pathways.