Botulinum toxin A combining with robot-assisted bimanual therapy integrating mirror therapy versus botulinum toxin A combining with robot-assisted bimanual therapy in patients with post-stroke spastic fingers: a randomized controlled pilot trial.

[BACKGROUND] Finger spasticity, which worsens hand function, is common in stroke patients. Botulinum toxin A (BoNT-A) treatment is effective in relieving spasticity but has little effect on arm-hand capacity. The adding of a rehabilitation program following BoNT-A injection has been suggested to enhance spasticity treatment outcomes. This pilot study aimed to compare the effects of BoNT-A combing with robot-assisted bimanual therapy (RBT) versus BoNT-A combing with RBT integrating mirror therapy (RBMT) in patients with chronic stroke and spastic fingers.

[METHODS] Patients with chronic stroke and finger spasticity were recruited. They were randomly allocated to receive 24 training sessions of RBT or RBMT. Assessments were performed before BoNT-A injection, after 24 training sessions, and 3 months follow-up. The outcomes included Fugl-Meyer assessment for the upper extremity (FMA-UE-total, proximal and distal component scores), modified Ashworth scale (MAS), action research arm test (ARAT), box and block test (BBT), and motor activity log (MAL). We applied Friedman's test to analysis intragroup, generalized estimating equations for intergroup differences of the treatment effects.

[RESULTS] Thirty-one patients were randomly allocated to the RBMT (n = 16) and RBT (n = 15) groups. After BoNT-A injection and training, both groups showed statistically significant improvements in FMA-UE total and FMA-proximal, MAS, and ARAT scores. Only the RBMT group showed significant improvements in the FMA-UE distal, and MAL scores. Neither group showed a significant improvement in the BBT. In the RBMT group the injection and post-training gains were maintained at the 3-month follow-up, except for MAS. The RBT group did not show a maintenance effect, except for the amount of use the affected UE, which showed significant improvement from baseline to follow-up. Comparing the treating effects between the two groups, there were significant differences in the FMA distal scores at post-training, favoring RBMT.

[CONCLUSIONS] Our pilot data demonstrated that an adjunct to BoNT-A for post-stroke spastic fingers, RBMT may provide more benefit in distal UE function than did RBT. However, a larger sample size and adding a proper control group are required to validate our findings in future studies. Trial registration This study was registered at ClinicalTrials.gov (ID no. NCT04826900, registered retrospectively on 03/28/2021).