Modified Constraint-Induced Movement Therapy with Robotics in Upper Limb Motor Recovery in Persons with Stroke: A Randomized Controlled Trial

Stroke is a leading cause of acquired adult disability, with around 80% of persons with stroke suffering from upper limb (UL) motor impairments. Among them, only a small percentage (12%) of persons achieve full arm function following UL rehabilitation training, which highlights the need for further research in this area.

The extent of UL impairment in stroke is one of the independent determinants of basic activities of daily living (ADL), so improving UL motor function is one of the central areas to focus on. Studies have reported many therapeutic techniques for UL training. Among them, two techniques, constraint-induced movement therapy (CIMT) and other robot-assisted movement therapy (RAMT), have demonstrated better outcomes compared to other existing options. Both these techniques aim to improve UL function by the neural plasticity principle through repetitive task-specific training.

It is already well established that robotic training has the capacity to provide a greater number of repetitions in a particular task-specific activity. The robotic training can provide more motivation and keep people engaged in task-specific training.  However, this robotic training has limitations. The task-activity training is being given in an artificial, mechanized environment. Therefore, the improvement achieved during an artificial, mechanized environment may fail to translate into a realistic environment.

Studies have reported that restraining the healthy/less-affected arm during forced-movement training of the impaired arm has a more beneficial effect than forced, repeated use of the impaired UL without restraint. The CIMT training involves restraining the less-impaired UL during most waking hours. Compared to forced-use, the CIMT training causes better white matter integrity. The CIMT training promotes better angiogenesis, nerve regeneration, and nerve function recovery. However, the main disadvantage of CIMT training is that CIMT training cannot precisely control the movement pattern during task-specific activity training and fails to keep the person motivated throughout the training period. In contrast, RA-MT provides continuous training with correct movement patterns and visual stimulation/feedback through exergaming technology, which helps the person to keep engaged in training.

To address the limitations of each technique, we propose a new therapeutic approach, robot-assisted modified constraint movement therapy (RA-mCIMT), which combines the principles of CIMT and RA-MT. This new technique involves structured, functional movement training with the robot instead of therapist-assisted UL training during CIMT therapy for one hour, with the healthy/less impaired UL restrained during the robotic training and most waking time. By combining the strengths of CIMT and RAMT, we assume that RA-mCIMT training will provide better motor recovery than mCIMT in UL following stroke.