Many studies have demonstrated the usefulness of repetitive task practice by using robotic devices, including Lokomat, for the treatment of lower limb paresis. Virtual reality (VR) has proved to be a valuable tool to improve neurorehabilitation training. Our pilot randomized clinical trial aimed at evaluating the correlation between the modifications of brain oscillations during a VR neurorehabilitative training of gait and the motor function recovery in patients with chronic stroke.

Twenty-four patients suffering from a first unilateral ischemic stroke in the chronic phase were randomized into two groups. One group performed 40 sessions of Lokomat with VR (RAGT+VR) whereas the other group underwent Lokomat without VR (RAGT-VR). Outcomes (clinical, kinematic, and event-related synchronization, ERS, and desynchronization, ERD, at the EEG) were measured before and after the robotic intervention.

The robotic-based rehabilitation combined with VR could be associated with improvements in several measurements of lower limb function, gait, and balance in patient with chronic hemiparesis. Moreover, ERS/ERD analysis can be proposed as a tool to monitor motor performance and to develop non-invasive brain-computer interfaces controlling robotic devices.

Inclusion Criteria:

• Age ≥55 years
• A first-ever ischemic supra-tentorial stroke (confirmed by MRI scan) at least 6 months before their enrollment;
• An unilateral hemiparesis, with a Muscle Research Council -MCR- score ≤3
• Ability to follow verbal instructions, with a Mini-Mental State Examination (MMSE) >24
• A mild to moderate spasticity according to a Modified Ashworth Scale (MAS) ≤2
• No severe bone or joint disease
• No history of concomitant neurodegenerative diseases or brain surgery.

Exclusion Criteria:

• Severe cognitive and behavioral impairments
• Severe osteoporosis and previous bone fractures