Improving Stroke Survivors' Ability to Walk

An NC TraCS $10K pilot grant, awarded to Michael Lewek, PT, Ph.D., has now led to a $400K NIH R21 grant to continue work studying ways to help stroke survivors’ improve their ability to walk. Lewek is an assistant professor in UNC’s Division of Physical Therapy in the Department of Allied Health Sciences. He is collaborating with Fred Brooks, Jr., Ph.D., Kenan Distinguished Professor; Mary Whitton, M.S., research associate professor; and Jeff Feasel, M.S., graduate research assistant, in UNC’s Department of Computer Science, as well as Heather Walker, M.D. from the Department of Physical Medicine and Rehabilitation.

Their work is testing two contrasting motor learning theories to find out the best way to rehabilitate people who have had stroke and restore their ability to walk symmetrically. On average, every 40 seconds, someone in the United States has a stroke, according to the American Heart Association’s journal Circulation. Often survivors experience partial or full paralysis on one side of the body. Regaining the ability to walk involves a long and arduous rehabilitation process. Despite rehabilitation, walking often remains slow, with an asymmetrical gait pattern that may be energy inefficient. In fact, walking after stroke can require as much as 40 percent more metabolic energy contributing to reduced endurance.

Using a programmable split-belt treadmill with sides that can speed up or slow down separately, Lewek and colleagues are testing whether it is more effective to have the treadmill automatically correct a person’s gait, or to have the treadmill amplify errors so that the person’s brain learns through the process of self-correction. The researchers’ hypothesis is that self-correction leads to better motor learning. Lewek likens it to children learning to ride bicycles using training wheels.

“When they have training wheels, it forces them upright. Then you bend the training wheels up and they start to wobble a little bit. They have to be able to experience those errors to learn what ‘straight’ is,” he said.

During the two-year project, the researchers hope to enroll 54 participants who have suffered stroke and randomly assign them to one of three groups – a control group, which receives no feedback from the treadmill, a group which has the treadmill autocorrect their gait or a group that will try to learn through self-correction. The researchers are also resubmitting an application for an NIH R01 award that would allow them to test these theories in a virtual environment where participants receive visual feedback about their walking to guide them in self-correction of their gait. They have already developed and done preliminary testing with the virtual environment, published a description of it and have presented it at national conferences.

“Even though they have had a stroke, their brains are still capable of processing information to adjust walking patterns,” said Lewek in a previous story on this work.

While Lewek’s primary interest is working with people who have suffered stroke, he sees potential for this bedside-to-practice research to translate into helping people who have mobility and gait problems due to other health issues, such as cerebral palsy, Parkinson’s, multiple sclerosis or as part of rehabilitation following knee or hip replacement.