Physical therapist Lisa Haubert is a trainer for the Extremity Constraint Induced Therapy Evaluation (EXCITE) study. Courtesy of USC Health Sciences Campus Public Relations Office

A stroke occurs when an artery bringing oxygen- and nutrient-rich blood to the brain clogs or breaks. Brain cells die (gray) when starved of oxygen. Courtesy Mayo Foundation for Medical Education and Research

Existing therapies may help recover movement and language, but only during the acute phase. After that, "chronic" stroke patients see little improvement.

Now we hear of a promising new treatment for new and old strokes that relies not on fancy drugs, gene therapy or electronics, but on a clumsy mitten, a broom, and plenty of hard work for the impaired arm.

Constraint-Induced (CI) therapy is so simple that we wish we'd invented it. Yet despite some impressive early results, the medical community remains to be convinced that its benefits are as great as its inventor believes.

Re-writing
CI therapy was introduced by Edward Taub, a psychologist at the University of Alabama at Birmingham, who says it is "almost uniformly effective for the higher-functioning 75 percent of stroke patients who are left with a residual motor deficit."

This group, he says, includes patients who have some ability to move the hand -- whether they had the stroke last year or 20 years ago.

After CI, a man who had suffered a stroke four years earlier signed his name, Taub says. "He nearly fell off his chair. He didn't know he had this capacity."

The therapy, he says, reorganizes the brain's cortex by enlarging the brain area controlling the impaired arm or leg.

On the theory that it seldom hurts to know what you're talking about, let's describe CI therapy before wading into the medical debate over effectiveness.

CI is an elaboration on existing techniques that forced patients to use the impaired arm. Those techniques, called "forced-use," do benefit acute stroke patients, says Steven Wolf, a professor of rehabilitation at Emory University and an expert in stroke recovery.

The major difference is intensity, Wolf says. Instead of one or two sessions a week, CI is a full-time endeavor for two weeks. The patient disables the good hand with a giant mitten for 90 percent of waking hours, five days a week.

For six hours a day, the patient uses -- trains -- the bad arm, doing normal, day-to-day tasks like pushing a broom, fun actions like throwing a ball, and fine-motor movements like connecting dot-to-dot drawings. The trainer uses "shaping" techniques to refine the first tentative movements. (Read Taub's review of the technique.)

Taking the gloves off
Although those awkward mittens grab the spotlight, Taub attributes only 20 percent of the improvement to disabling the good arm, and 80 percent to the intensity of practice. Such intensity, he points out, is common to many emerging therapies that try to reorganize the cortex to learn new tricks.

If CI works as well as Taub thinks -- and we'll hear from skeptics shortly -- it would be a boon, as existing therapies are ineffective after the acute phase. "In chronic patients," Taub says, "nothing does any good, that's not controversial."

In acute patients -- the first few months after the stroke -- with enough residual movement, stroke rehabilitation tries to enhance that movement, explains Larry Goldstein, professor of medicine and director of the Center for Cerebrovascular Disease at Duke University." The techniques used by therapists to accomplish this goal may vary depending on their particular training. However, there are no good data that any one approach is better than any other." If the therapist concludes that the patient is unlikely to recover, the emphasis is teaching the person to compensate for the impaired limb with the intact limb.

Ironically, if you remember Donald Hebbs's theory (neurons that fire together wire together), you may agree with Taub that compensation may, ironically, "stamp in learned non-use." Repeated use of the good arm, he says, makes it that much harder to start using the bad arm.

Plastic fantastic thinker?
The improvements from CI therapy are backed by tests of hand function and by transcranial magnetic stimulation (TMS). Taub used this brain-mapping procedure to measure how much of the brain was controlling the movement of a thumb muscle. Taub says he selected the muscle that retracts the thumb because thumbs are critical to hand function, and hands are critical to the activities of daily life.

TMS data show that after a stroke, the brain region associated with the bad arm's thumb muscle is only half as big as the area for the opposite thumb.

The reorganization of the brain caused by CI therapy appears in the TMS brain maps. Taub asserts: The area of the brain driving the impaired thumb grows to equal that controlling the other thumb.

To Taub, this is further evidence that the adult brain is indeed plastic, debunking the dreadful dogma of neuroscience -- that old brains can't be reprogrammed.

Out on a limb
CI emerged from monkey research that Taub performed decades ago. Earlier in this century, he says, neuroscientists believed -- mistakenly, it turned out -- that voluntary movement depended on reflexes. Since sensory nerves are critical to reflexes, he, like others, cut the sensory nerves in a monkey's arm, and observed that they quit using the arm, even though their motor nerves were intact.

Would it be possible, he wondered, to use new conditioned-response techniques to induce the monkey to resume using the damaged limb?

He found that two techniques worked equally well: restraining the intact arm, and training the monkey to use the damaged arm.

The research sat for decades, Taub says, since, he was a research psychologist and did not focus on clinical work. "It was 20 years before it occurred to me that what was happening ... was a very substantial rehabilitation of movement," Taub told The Why Files. "We were converting a useless limb into a limb that could be used."

In the late 1980s, Taub began early experiments with chronic stroke patients, and he's grown increasingly enthusiastic at the results.

Leonard Fisher, assistant professor of Geriatrics at UCLA School of Medicine, helps an elderly patient at the Eichenbaum Health Center.
Courtesy Association of American Medical Colleges

To date, he says, about 300 patients in Alabama, Germany and elsewhere have benefited from CI therapy. Included in this group, he says, are people who had strokes more than 20 years ago -- who could not be helped by any existing therapy.

It's not just Taub who's getting excited about CI therapy -- he says 5,000 people have inquired.

Curious whether the combination of restraint and intense practice could help in other disabilities, Taub and others have tested CI with aphasia -- the loss of speech often caused by stroke (see "Constraint-Induced Therapy..." in the bibliography).

For humanitarian purposes, we hope Taub is right, but we're obliged to cite the skeptics, who stress that Taub is a research psychologist, not a clinician. This is science, after all, not religion, and the road to medical progress is littered with therapies that didn't pan out.

Sawing it off?
"The bottom line is that this is a promising but unproven mode of therapy," says Goldstein. "It's currently the subject of a randomized-controlled clinical trial funded by the NIH [National Institute of Health] to find out if it works, and how generalizable it is. If this were proven to be of benefit, then the NIH would not be funding a trial, nor would it be ethical to do so."

Steven Wolf, the leader of that study, the EXCITE, Extremity Constraint Therapy Evaluation trial, seconded Goldstein's remarks.

Sewing is one activity that may be impossible after a stroke. Courtesy National Archives and Records Administration

Wolf said that EXCITE will test whether patients, treated three to six months after a stroke, benefit from CI therapy, not just on movement tests, but "whether the improvement is meaningful, whether they are returned to daily life." Wolf says EXCITE will evaluate 240 patients at six sites for two years, and produce results in 2005.

For a rehabilitation technique to succeed in the real world, Wolf insists, it must impress health insurers. "There is no reason to think that insurance companies will pay unless there is improved functioning, more independence, and people are capable of generating income."

While conceding that the randomized, controlled study is "the gold standard" for medicine, Taub says, "At this point, I'd bet the farm [that CI works], given the data that we have. We have two studies with placebo controls, replicated in other labs, and several hundred patients. All that converges on a demonstration of clinical efficacy."

Goldstein, who is leading two stroke recovery trials on using the stimulant amphetamine to improve movement in patients with recent strokes caused by blood clots, argues that, given the human stakes, caution is the only legitimate attitude toward CI therapy at this point. "There are many desperate patients and families looking for something to help. It's important not to oversell any of these approaches until rigorous data from controlled trials are available demonstrating that they in fact are both safe and effective."

What is the emerging new paradigm about learning and adult brain?