Despite the therapeutic potential of epidural motor cortex stimulation (MCS) on stroke-induced upper limb paresis, its mechanism of action largely remains unknown. We addressed this issue by expanding the therapeutic intervention period in the subacute stage of stroke (defined as over one month and within 2 months from stroke onset) and the chronic stage of stroke (defined as over 2 months after stroke onset) regarding the feasibility and safety of MCS. In stroke patients with moderate upper limb paresis, electrical stimulation was applied on the primary motor cortex during rehabilitation. All investigated patients (n=3) showed a substantial improvement in their upper limb paresis. In patients who underwent subacute MCS (n=2), the extent of recovery exceeded that observed within one month from stroke onset, and the recovery continued for over 6 months. In one patient who performed MCS more than 6 years after stroke onset (n=1), the patient’s Brunnstrom stage findings regarding his hands and arms increased from 2 to 4 and 3 to 4, respectively. Although our findings are still preliminary, these results suggest the feasibility and safety of performing MCS, while appearing to confirm the potential effectiveness of both subacute and chronic MCS as a neurosurgical treatment for paresis in stroke patients.

As a part of neurorehabilitation, repetitive motor training combined with noninvasive brain stimulation has been recently used to improve motor functions in patients who have suffered stroke. Here, we developed a repetitive transcranial magnetic stimulation (rTMS) triggered by a voluntary movement detection system and evaluated the effects of this system on hemiplegic finger motor performance in patients with stroke. Nine patients with chronic stroke having hemiplegia underwent the two interventions in a cross-over design: repetitive hemiplegic finger extensions with 10 Hz rTMS at a stimulus intensity of 80% active motor threshold for 500 ms (Real) and repetitive hemiplegic finger extensions with sham rTMS (Control). We compared the maximal extension angle of the hemiplegic finger between Real and Control, and found that the maximal extension angle of metacarpophalangeal joint in Real was greater than that in Control. Therefore, we suggest that rTMS triggered by a voluntary movement detection system would be a new non-invasive neurorehabilitation device for stroke patients.