In recent years, scientists have discovered that the brain has a remarkable capacity for self-repair. Hoping to take advantage of this ability, researchers have developed a technology to deliver electrical stimulation directly to brain tissue. The therapy, now being tested in large clinical trials, could boost the brain's repair mechanisms and improve recovery after stroke.

Studies in both laboratory animals and humans have shown that after stroke, neurons near the damaged tissue begin to reorganize themselves in an attempt to compensate for the injured areas. However, this healing ability can be hit or miss -- some patients regain the ability to walk or talk while others are left permanently disabled.

[Click here for illustrations of the brain's areas and device's functions.]

In many cases, patients can stimulate recovery through practice. Someone who has lost function in their left hand, for example, could practice various movements with that hand to boost the brain's innate repair mechanisms. "But in most cases, that neuroplasticity doesn't go far enough," says Alan Levy, CEO of Northstar Neuroscience, a medical device company based in Seattle, WA.

So Levy and collaborators designed a way to stimulate specific parts of the cortex to try to further enhance the brain's natural neuroplasticity. The technology has shown promise in preliminary human studies -- researchers found that patients receiving both rehabilitation therapy and stimulation improved 15 to 30 percent on standard tests of hand and arm function; while controls, who underwent only physical therapy, improved just 0 to 12 percent. Northstar is now sponsoring a larger clinical trial at 18 rehabilitation centers across the United States.

Experts caution that it's too soon to say how effective or broadly applicable the technology will be, though. "We need to see studies in larger groups to know if it's effective," says Douglas Katz, a neurologist at Boston University Medical School, "and under what circumstances it's effective, such as the location of stroke, the time after stroke [that the treatment is used], and how much stimulation is necessary." Adds Katz: "But I do think these techniques show a lot of promise."

The benefits may also depend on the severity of stroke. It's possible that this therapy will be effective only in patients with relatively mild impairments, says Randolph Nudo, director of the Landon Center on Aging at the University of Kansas Medical Center in Kansas City, who is studying the effects of the Northstar technology in animal models of stroke. People who have had a more severe stroke, and therefore have fewer neurons left to compensate for the damaged area, may not be able to benefit from stimulation.

Nudo and colleagues are running exhaustive animal studies to determine the most effective parameters for the cortical stimulation treatment, as well as if remote areas of the brain may be recruited to aid people with more severe stroke.

In the cortical stimulation procedure, doctors first map the extent of the damage using brain imaging. Movement of the hand, for example, is governed by a specific part of the motor cortex, a layer of the brain that governs movement. Physicians use functional magnetic resonance imaging, which measures blood flow in different parts of the brain, to locate the part of the cortex that is damaged, as well as the neighboring areas that are trying to take over control of the damaged hand. A neurosurgeon then drills a small hole in the skull over this area and places a flat electrode on top of the dura, a tough membrane covering the brain, to stimulate the cortical region below.

The stimulator is powered via a pacemaker-like device implanted in the chest and connected to the electrode by a cord threaded under the skin. A doctor or physical therapist can turn the stimulator on and off with a wireless controller. The stimulator is turned on only when patients are doing rehabilitation exercises; patients in the Northstar clinical trial will undergo an intensive six-week physical therapy program.