Myoelectric Computer Interface Training for Reducing Co-Activation and Enhancing Arm Movement in Chronic Stroke Survivors: A Randomized Trial.

The sample was small and did not reach the planned 12 participants in each group; the 90-minute movement group had 9 analysed participants. There was no active control, so improvement could reflect using the affected arm more rather than MyoCI specifically. The methods section says 33 participants completed 6 weeks, while the results section says 32 completed and were analysed. Co-activation decreased during training but did not significantly decrease during reaching for either isometric group, so transfer was limited. Kinematic analysis included only 20 participants because of goniometer signal problems. Stroke location analysis included only 19 of 32 participants and gave mixed, largely insignificant results. One week-10 value was estimated from weeks 2 and 6, although analysis without it did not change outcomes. All evaluations were done by the same therapist, which may limit generalisability. No significant between-group differences were found, so the trial cannot recommend a dose or training format.

The authors declared no conflicts. The article is listed as supported by NIH extramural research.

Do not conclude that MyoCI training is better than usual therapy or that movement-based or 90-minute training is superior; the trial had no active control and found no significant between-group differences.