A randomized clinical control study on the efficacy of three-dimensional upper limb robotic exoskeleton training in chronic stroke.

Four patients withdrew and were not included, so results are from 22 completers. The trial recruited from one hospital outpatient unit in chronic stroke. Participants had to be right-handed, have a left hemisphere stroke, have a total FMA score of at least 15 out of 66, have shoulder elevation ability, and have no severe cognitive deficits, so the results do not apply to acute stroke, severe upper limb weakness, or patients with cognitive impairment. The two treatments were not identical: manual therapy involved more hand function, while robotic therapy involved more arm amplitude and intensity, so differences in proximal FMA may reflect the training content rather than the exoskeleton itself. The positive proximal FMA and BAT timing findings were secondary outcomes, so they are less certain than the primary outcome. Outcomes were measured only before and after the six-week treatment, so durability is unknown. Only 9 of 11 robotic participants completed the unassisted robotic evaluation, so the prediction analysis is very small. No blinding is reported.

Funded by FP7 Information and Communication Technologies and Ente Cassa di Risparmio di Firenze. The supplied text does not report author conflicts of interest.

Do not conclude that robotic exoskeleton training is better than manual therapy for overall upper limb recovery. The primary total FMA outcome showed no significant between-group difference, and the positive shoulder/elbow and bimanual speed results were secondary analyses in a small trial with only 22 completers and only 9 robotic participants available for some movement measures.