Objective: To investigate effects of home-based somatosensory stimulation associated with motor training on improvement in performance of the paretic upper extremity in patients in the chronic phase after stroke.

Design: Double-blind, randomized, controlled clinical trial. Setting: Home-based rehabilitation. Subjects: Patients in the chronic phase (> 6 months) after stroke. Interventions: Patients will be randomized to an active and a control group. In both groups, treatments will be performed at home, daily, over four weeks. Patients in the active group will be instructed to use a device that provides somatosensory stimulation in the form of electrical stimulation of the median nerve for two hours. Patients in the control group will also be instructed to use the device for two hours, and sham stimulation will be administered. In both groups, patients will be instructed to train activities with the paretic hand immediately after nerve stimulation.

Main measures: The primary outcome will be improvement in performance of the paretic hand evaluated by the Jebsen-Taylor test at end of treatment and four months after end of treatment, compared to baseline. Secondary outcomes were: 1) Compliance with the interventions, assessed through a daily written log, and oral reports of the patients; 2) safety profile of the interventions.

Inclusion Criteria:

• age 18-80 years;
• single ischemic or hemorrhagic stroke in a cerebral hemisphere, documented by computed tomography (CT) or magnetic resonance imaging (MRI);
• stroke onset at least six months before;
• hand paresis with preserved ability to perform all tasks of the Jebsen-Taylor Test (JTT).

Exclusion Criteria:

• previous strokes;
• epilepsy and other neurological conditions;
• proprioceptive or tactile anesthesia;
• shoulder or hand pain;
• severe joint deformity;
• severe chronic disease;
• inability to give provide informed consent due to severe aphasia or cognitive impairment;
• left handedness before the stroke.