By Nancy Menefee Jackson, HealthAtoZ contributing writer The man in his early 60s is arranging some brightly colored cylinders into rounds holes on a board. It almost looks like an educational toy for small children. He's having a difficult time manipulating the objects, and on his other hand is something that looks like an oversized white mitten. A relatively new therapy, developed at the University of Alabama at Birmingham, helps stroke victims regain the use of impaired limbs. The research, headed by Edward Taub, Ph.D., a professor of psychology and director of the CI Therapy Research Group and the Taub Training Clinic, is so promising that the National Institutes of Health is funding a five-year, national multicenter trial. Called "Constraint Induced Movement" therapy (CI therapy), it differs from conventional therapy for stroke victims in two ways: first, it is an intensive concentrated training of the affected arm; second, it involves restraining the good arm. "The restraint is responsible for about 20 percent of the treatment effect," says Taub. "The heart of the matter is to get the patient to use the affected extremity over and over again. We train the affected arm in a concentrated fashion -- intensive training for two or three weeks on consecutive weekdays -- and at the same time we put a restraining device on the unaffected arm for a target of 90 percent of the time." Taub says CI therapy was founded on his own research with monkeys. The researchers abolished the sensation in a forelimb of the monkeys and found that they never used the arm again. However, they learned that if they restrained the monkey's intact arm, within several hours the monkey would begin to use the bad arm. "And if you left the device on for several days, the monkey would use the affected arm for the rest of his life," Taub says. "Intensive training had the same effect. It occurred to us we might be able to get human stroke patients to respond the same way." He began working with patients who had had a stroke more than a year before. At first, they tried to use a hand splint in a sling, but that was unsafe for patients whose legs or balance were affected by the stroke too. So they adopted a protective safety mitt, somewhat like a catcher's mitt, to restrain the good hand. The person could swing that arm for balance, but still couldn't use the fingers of the good hand. "Doing that tripled the size of the population we could work with," Taub says. Conventional wisdom held that if by a year after a stroke the patient hadn't recovered use of the arm, they never would. But Taub and his group worked with patients who had strokes years before, such as one 50-year-old patient had had a stroke when he was only 5. "We work with patients with chronic stroke who are not supposed to be able to improve," Taub says. Except they do. "Our average patient begins with about 7 percent use of the affected arm," Taub says. "At the end of two weeks of intervention, we're up to about 50 percent of use in more than 95 percent of patients." Taub estimates that his laboratory and clinic have worked so far with about 300 patients, and of those, only five have had little or no improvement. Improvement is measured by performance at home in 30 activities of daily living, and the use of an accelerometer that measures movement objectively. Patients might regain more mobility in the shoulder, elbow and wrist, but Taub evaluates success mainly by how much they can use their hands and fingers. "It is with the hand that the person carries out a great majority of the activities of daily living," he says. Initially, Taub says, the patients treated with CI therapy were those with mild to moderate strokes. "We thought we'd skimmed the cream off the top of the bottle," he says. But when they began using CI therapy on people with moderate to moderately severe strokes, the results were still good. "They're starting at a lower level; so they wind up at a lower level, but the treatment change is just as large," Taub says. Even those with the most severe strokes gained some improvement although not usually in the fine motions of the hand. CI therapy uses shaping, a technique of having patients repeat task movements for 30 seconds, trying to do them as fast and as well as they can, while being given continuous feedback on what they need to do to improve. Taub explains that after a stroke, a person in the hospital tries to use the affected limb and can't. But the ability to do so remains latent in the central nervous system. With CI therapy, a person starts to use the arm, and does so repeatedly. That results in a change in the organization and function of the brain termed cortical reorganization. "CI overcomes learned non-use of the impaired arm and also increases the area in the brain involved in producing movement," he says. Now researchers at the Taub Training Clinic have extended the technique to the legs, and they've received funding from NIH to treat children with cerebral palsy with CI therapy. They also have studied small groups of patients with spinal cord or hip injuries. "CI treatment is not specifically for stroke; it's for non-use," Taub says. "Learned non-use occurs after many substantial injuries to the nervous system and it can get locked in." CI therapy promises to break that lock, allowing patients to recover more fully. Related Article Stroke