Combining muscle-computer interface guided training with bihemispheric tDCS improves upper limb function in chronic stroke patients.

Abstract

Transcranial direct current stimulation (tDCS) may facilitate neuroplasticity but with a limited effect when administered while stroke patients are at rest. Muscle-computer interface (MCI) training is a promising approach for training stroke patients even if they cannot produce overt movements. However, using tDCS to enhance MCI training has not been investigated. We combined bihemispheric tDCS with MCI training of the paretic wrist and examined the effect of this intervention in chronic stroke patients. A crossover, double-blind, randomized trial was conducted. Twenty-six chronic stroke patients performed MCI wrist training for three consecutive days at home while receiving either real tDCS or sham tDCS in counterbalanced order and separated by at least 8 months. The primary outcome measure was the Fugl-Meyer Assessment Upper Extremity Scale (FMA-UE) which was measured one week before training, on the first training day, on the last training day, and one week after training. There was no significant difference in the baseline FMA-UE score between groups nor between intervention periods. Patients improved 3.9 ± 0.6 points in FMA-UE score when receiving real tDCS, and 1.0 ± 0.7 points when receiving sham tDCS (p = 0.003). Additionally, patients also showed continuous improvement in their motor control of the MCI tasks over the training days. Our study showed that the training paradigm could lead to functional improvement in chronic stroke patients. We argue that appropriate MCI training in combination with bihemispheric tDCS could be a useful adjuvant for neurorehabilitation in stroke patients.