Harnessing the Brain's Power to Adapt After Surgery

Whether a brain or spinal cord injury is caused by a weapon of war, an accident, or a disease such as stroke, rehabilitation focuses on enabling people to make the most of what functions they still have. Physical, occupational, and speech therapy, counseling, and education can go only so far, however. For neurorehabilitation to offer the hope of curing the underlying brain damage, writes an expert in the field, it must look to basic science and better clinical trials to put to work the power of the brain’s plasticity.

More than 300,000 Americans every year receive a head injury severe enough to require medical attention, and about 75,000 of them end up with permanent neurological damage. Many more people suffer brain damage from stroke, multiple sclerosis, or Alzheimer’s and other diseases. The wars in Iraq and Afghanistan have focused the public’s attention on the potentially devastating nature of brain injuries because soldiers, whose bodies are protected by effective armor, are able to survive gunshot attacks and percussion injuries from IEDs (improvised explosive devices). Their hearts, lungs, and other vital organs are protected, and within minutes of an attack their vital body functions receive medical attention in the field. As a result, brain trauma represents a larger proportion of non-fatal war injuries than was seen in previous conflicts.

Men and women come home from the war with functional impairments that range from severe paralysis to problems with memory or speech to the more subtle but often devastating loss of the mental abilities that enable a person to concentrate on a problem and have the initiative to solve it. Without the ability to handle complex information, organize it into meaningful categories, set reasonable goals, and make good decisions, they cannot resume their careers or even successfully navigate daily life.

To give hope to our veterans suffering from traumatic brain injury as well as millions of people worldwide who have some kind of brain damage, we must seek to reverse this damage, not just ameliorate it. The current treatments for traumatic brain injury are inadequate. Rehabilitation medicine relies almost exclusively on techniques aimed at making the most of what neurological function is left, not changing the deficits themselves. Although in recent years, these techniques have received more careful scientific study, many in the field have sensed that progress has reached a standstill. For this to change, we must take an approach traditionally considered outside the definition of rehabilitation medicine: focusing on the underlying neurobiology and allowing ourselves the aspiration of actually curing brain damage through harnessing the brain’s power to change and heal.

The Traditional Approach

Rehabilitation medicine has been well ahead of the field of medicine in general because it emphasizes the importance of assessing patients’ functioning in practical tasks and of their quality of life. Traditionally, neurorehabilitation has employed physical, occupational, and speech therapies, evaluated the home and work environments, and worked hard to educate patients and families on strategies to compensate for lost functions. But as important as these strategies are, they cannot fully restore normal function, and research into improving their design and application does not appear to be leading to profound advances.

Typically, we retrain a patient to accomplish a specific task, using practice and repetition, but that one task is as far as it goes—what has been learned does not transfer to other tasks. For example, a stroke in the speech area of the left cerebral hemisphere may result in impaired language (aphasia), which includes difficulty finding the correct names for objects. A person may recognize a watch and understand its purpose but be unable to come up with the word “watch” when asked to identify one. With extensive practice, the aphasic person can eventually relearn to say “watch,” but this does not result in improved naming of other objects. In fact, after learning 100 words in this way, the person would take virtually the same amount of time to learn the next 100 words. Similarly, physical therapy, including passive stretching of tightened right arm and hand muscles and lots of practice performing tasks with that arm and hand, may improve those arm and hand functions, but only practicing walking can improve a limp that resulted from the same stroke.

These examples illustrate the limitations of traditional rehabilitation approaches. The day does not hold enough hours to practice every single activity and, even if it did, improvement would be incomplete.

The Plastic Brain

Although scientists long thought otherwise, the brain has the ability to adjust the strength of its electrical connections and even to sprout new connections, in order to cause partial recovery of abilities previously lost because of an injury or stroke. Enthusiasm for the concept that the brain can adapt to injury has fluctuated, but experiments in monkeys and other animals have rekindled interest in the possibility that the brain is more changeable than was once thought. Perhaps we could harness that “plasticity” to help people recover from brain injuries.

To give an idea of what plasticity might mean, let me mention a few key experiments. In the 1980s, Michael Merzenich, Ph.D., at the University of California at San Francisco, and other researchers showed that when a small body part such as a finger is amputated in a monkey, the nerve cells in the sensory part of the cerebral cortex that used to respond electrically to stimulation of that body part did not become quiet. Instead those nerve cells began to respond to stimulation of neighboring body parts. Similarly, studies by Randy Nudo, Ph.D., at the University of Kansas showed that if a tiny stroke is produced by blocking the blood flow to a small part of a monkey’s motor cortex, the part of the body that used to move in response to electrical stimulation of that area of cortex would now move when nearby areas of the brain were stimulated.

In recent years, functional imaging studies by Richard S. Frackowiak, M.D., D.Sc., in London, and many other groups around the world, have confirmed that the brain can change its responses in human stroke patients in ways similar to what was found in monkeys. This has also been shown by experiments using transcranial magnetic stimulation of the human cortex. Studies by Leo Cohen, M.D., at the National Institutes of Health (NIH), his protégé, Alvaro Pascual Leone, M.D., Ph.D. (now at Harvard), and many others have shown that recovery from stroke is accompanied by plastic changes in the way that electrical activity in the brain triggers movements in the body. These changes go beyond behavioral adaptation to injury.

The Challenges of Studying Rehabilitation Therapies

These demonstrations of inherent malleability of the human brain have triggered a period of intense experimentation seeking to harness that plasticity. But so far we have been frustrated by the limits imposed by the biology of the brain and by the difficulty in doing human experiments that clearly tease out the effects of therapy. Not only do limits exist on how much practice can be devoted to relearning cognitive and motor tasks, but it has turned out that the brain’s plasticity has severe limits, as well. Furthermore, it is difficult for researchers to determine how much improvement is attributable to a particular therapy, how much to what is called the placebo effect, and how much to the spontaneous partial recovery that is the norm after stroke or brain injury.

Let us look at this last point first because it affects everything else. Most spontaneous improvement occurs within the first month after a brain injury, but some additional gains occur over the course of three to six months. Unless carefully controlled studies are performed on patients with similar injuries and functional impairments, and those patients are also matched for age, sex, and other important variables, and randomly assigned to groups receiving different treatment regimens, it is not possible to say whether a given therapy has had a beneficial effect. This is especially true when the treatment is tested during the first few months of recovery.

We also know that any therapy causes a placebo effect—improvement for reasons not specific to the therapy itself but to the complex psychological and biological effects of the attention, encouragement, and expectations associated with being a part of the testing process. The placebo effect adds further variability to experimental results already made variable because of the element of spontaneous improvement. The placebo effect can be minimized, but not eliminated, by being scrupulously careful that neither the patient nor the investigator knows which treatment is the one being tested and which is the placebo—an alternate, usually inactive treatment.

Having all patients in an experiment carefully matched, randomly assigning them to groups, and controlling for the placebo effect are the key components of what is called the prospective, randomized, double-blind, placebo-controlled clinical trial (RDPCT), which has become the gold standard of medical research. Controlling for the placebo effect is relatively easy to do when the treatment is a pill, because it is possible to design placebo pills that look just like the test drug. But when the treatment is a physical therapy, how do you hide the real treatment from both the patient and the therapist? Sometimes this is impossible. But in a single-blind clinical trial, the treatment being tested can be matched with a control treatment that is of similar intensity and duration. In that case, the therapist knows which treatment is being tested, but the patient does not.

In addition to the challenges of setting up a double-blind test for a physical therapy, other aspects of rehabilitation can be at least as difficult to study in controlled clinical trials. Designing a study that can actually delineate the benefits of a particular form of education and family counseling, vocational rehabilitation, or the use of prosthetic devices for amputated limbs can be daunting.

In part because of these challenges, the professional rehabilitation community has been slow to submit its treatments to randomized controlled clinical trials, and only recently has this been done on a scale large enough to yield reliable results. Two recently completed high-profile studies illustrate the complexity of the problem. I describe them briefly here because they made so much news in the medical community and because they were the first NIH-sponsored, randomized, multi-center, controlled trials of any physical therapy. The results of both trials suggested that, with intense enough practice, patients with injuries to the central nervous system could gain some improvement, but they left the field of medicine in the dark as to what influence the precise mode of therapy had.

The first study, led by Bruce Dobkin, M.D., a neurologist and neurorehabilitation expert at the University of California, Los Angeles, was a test of whether treadmill training, which strengthens a spinal cord circuit that generates walking and running movements, would work better than conventional physical therapy in people with a spinal cord injury (the Spinal Cord Injury Locomotor Trial, or SCILT). One hundred forty-six people who had experienced an incomplete spinal cord injury were studied at six medical centers. The patients all received one hour of physical therapy per day for 12 weeks. Half of them were put in a harness that supported part of their weight while therapists helped them walk on a treadmill, and the other half received more-conventional physical therapy.

When their walking was tested later, both groups of patients had improved equally. From one point of view, this outcome was disappointing, but a very interesting feature emerged. Both groups of patients in this study, which involved more-intensive physical therapy than most patients ever get, improved more than was expected for most patients with spinal cord injuries. In other words, it may be that the intensity of training was more important here than the precise mode of training.

The second clinical trial was the Extremity Constraint-Induced Movement Therapy Evaluation (EXCITE), headed by Steven L. Wolfe, Ph.D., P.T., a physical therapist at Emory University School of Medicine. Constraint-induced movement therapy (CIMT) for people who have experienced a stroke and lost function in an arm or hand involves forcing the affected arm to work by binding the unaffected arm in a mitt or sling. It can improve function of the affected arm as long as patients had some hand and wrist movement to begin with, and this improvement can persist for at least a year. The goal of the EXCITE trial was to test this benefit. Therefore, 222 stroke patients at seven academic medical centers were randomly divided into two groups. One group was treated with CIMT, in which the good arm was kept in a restraining mitt and the bad arm was engaged in an intensive regimen of repetitive task practice and behavioral shaping. The control group was treated with “usual and customary care,” but the intensity and the amount of time devoted to their physical therapy were not specified.

Although, as expected, the CIMT group showed greater improvements in arm function than the control group and the benefits persisted for at least one year, it is nonetheless difficult to evaluate the results. On the one hand, the control group did not receive the same intensity of therapy as the CIMT group, which means that it is not possible to determine how much of the improvement seen in the CIMT group was a consequence of the specific mode of therapy they received and how much was related to the amount of therapy in general. On the other hand, the main purpose of the constraint therapy was to force the bad arm to be used more than it would have been if the patient were free to use the good arm for everything.

One interesting aspect of the EXCITE trial is that many of the patients who were enrolled had experienced their strokes long enough in the past that they had already achieved most of the spontaneous recovery and even the improvement as a result of physical therapy that would normally be expected. The improvement they experienced as a result of the EXCITE trial showed that intense physical therapy can result in additional benefit, even late after stroke. Nonetheless, this trial still left us with the need to do more research.

The results of both the SCILT and the EXCITE trials, reported in journal articles in 2006, reinforced two aspects of research on physical therapies. First, the therapies can deliver some improvements in function, but not cures. Second, the intensity of therapy appears to be important in the amount of improvement, but it is still not clear whether the precise technique used is important.

Rehabilitation Needs Basic Science

Just as medical rehabilitation has lagged in developing evidence-based practice, the field also has been slow to adopt a basic science underpinning, and it is tempting to conclude that the two phenomena are related. Although the central nervous system may be more plastic and capable of more adaptation to injury than was believed previously, severe limits exist to what can be accomplished by practice and exercise alone. If we are to go beyond these limits, we will have to intervene more actively in the basic biology of the nervous system. Basic science studies provide a rationale for clinical studies and expand our view of what might be possible.

Take spinal cord injury as an example. The most famous case is that of the actor Christopher Reeve, who became totally paralyzed from the neck down as a result of a horseback riding accident in 1995. Before his death in 2004, he engaged in very intensive physical therapy, including treadmill training with the aid of physical therapists. Because he had financial resources beyond those of most people with spinal cord injuries, and the determination to participate actively in his own treatment, he was able to avail himself of the most up-to-date and most intensive therapies that exist. One journal article, in fact, indicated that late in life he had begun to regain some sensations below the neck and to experience flickers of movement. But despite this improvement, he remained essentially totally paralyzed in both arms and both legs. Reeve himself was convinced of the need to apply basic science research to the problem of spinal cord injury, and he formed a research foundation dedicated to finding ways to regenerate nerve pathways in the injured spinal cord. He also lobbied intensively for stem cell research, believing that this had great potential for repairing the injured brain and spinal cord.

We have made real progress in understanding some of the mechanisms that control nerve regeneration, as well as some of the factors that explain why this is so difficult in the human central nervous system. Some animal experimentation has yielded promising results. Yet, in my opinion, we have a way to go before we can apply this knowledge successfully to human brain and spinal cord injury. Among the many important questions remaining to be answered are these:

1. Since so much of what we think we know about regeneration is derived from experiments on immature nerve cells, are the mechanisms of regeneration in the injured mature nervous system the same as those that apply to the developing embryonic nervous system?
2. Since the vast majority of experiments in regeneration of nerve pathways have been done in rats and mice, how predictive are these experiments for results in human patients? Apart from molecular differences, rodents are much smaller than we are. Nerve fibers may have to regenerate much farther in humans in order to achieve the same level of reconnection that underlies functional improvement in smaller animals.
3. Even if sufficient nerve regeneration can be achieved, will the connections made be specific enough to underlie real function?
4. How helpful are stem cells? Can they survive after transplantation into the human spinal cord or will they be rejected? Can they replace damaged neurons or will they serve only as sources of chemical substances that support survival and growth of the brain’s own nerve cells?
5. Will we be able to identify a single approach that is so fundamental that it can yield dramatic improvements in recovery from brain injury, or will we need to develop a cocktail approach, using multiple treatments simultaneously?
6. Will approaches that enhance regeneration in one circumstance, for example spinal cord injury, also work in other situations, such as stroke or traumatic brain injury?

When these and other questions are answered—and I am confident that they will be—we will be able to offer a much more profound recovery than we can to our veterans returning from Iraq and Afghanistan today with brain and spinal cord injuries. But even in the shorter term, a broadening of the research agenda in neurorehabilitation to include addressing the fundamental biology underlying disabling brain disorders can help the field of rehabilitation medicine advance in ways that go beyond offering patients better functional recovery. It can help attract the best and brightest graduates of medical schools to a specialty that has, to date, specifically ruled out cure from its research agenda by defining itself as a specialty that seeks to optimize function in people with fixed anatomical and physiological impairments. Rehabilitation medicine has left the possibility of curing the impaired anatomy to other specialties, such as neurology and neurosurgery.

But now with new influences, especially from the neurorehabilitation community, a multidisciplinary collection of neurologists, neuroscientists, physical therapists, physiatrists, and other rehabilitation professionals is coming together to attack the problem of central nervous system injury. These researchers subscribe to the notion that rehabilitation can best be achieved by understanding the fundamental neuroscience that underlies neurological disabilities. Because this new generation of rehabilitation scientists is committed to intervening in the underlying pathophysiology, the field of rehabilitation medicine is undergoing a dramatic enhancement in its research mission and its attractiveness to talented medical school graduates.

Signs of Change

This new vigor is reflected in the expansion of the research supported by the Division of Rehabilitation Research and Development in the Office of Research and Development of the Veterans Health Administration and in the National Center for Rehabilitation Research, the two largest funding agencies for rehabilitation research. In addition to developing evidence of the effectiveness of traditional rehabilitation interventions, these agencies are supporting research in such futuristic areas as nerve regeneration in the brain and spinal cord, brain-computer interfaces for communication and motor control in paralyzed persons, robotic-assisted physical therapy, electronic-assisted prosthetic limbs, and prosthetic neural circuitry. In the area of spinal cord injury, approaches aimed at neutralizing the molecular inhibitors of nerve axon regeneration have already reached the stage of human clinical trials.

This progress is exciting because the pharmaceutical industry is now ready to gamble its own resources to test the therapies developed in animals by scientists who are supported by the National Institutes of Health, the Department of Veterans Affairs, and other public and private research organizations in the United States and abroad. In one such clinical trial by the pharmaceutical company Novartis, antibodies against a molecule (whimsically called Nogo) found in the myelin sheath that insulates axons are being injected into the spinal fluid of patients with severe spinal cord injuries to see if these patients will be able to regenerate the severed nerve connections. After a preliminary study of more than 25 patients, no serious toxicity was found, and the study will be expanded to more patients to gain preliminary evidence for effectiveness.

Alseres Pharmaceuticals is testing a second approach, in which an inhibitor of the enzyme RhoA is applied to the membrane covering that surrounds the spinal cord. RhoA is part of the signaling pathway that mediates the effects of Nogo and other growth-inhibiting molecules. Study of the use of this inhibitor in more than 30 patients revealed no serious toxicity, and the clinical trial will be expanded to test for efficacy. It is too soon to know whether either of these treatments will promote axon regeneration and functional recovery of human patients with spinal cord injuries, but because successful experiments in animals were reported by highly respected investigators, the clinical trials have attracted a great deal of attention. Moreover, animal experiments suggest that approaches based on neutralizing Nogo might also help in the recovery from stroke, a condition much more common than spinal cord injury.

The association of the clinical neurorehabilitation community with this and other work at the forefront of translational research has done much to increase the acceptance of rehabilitation medicine by the general medical community. This in turn has enhanced the influence of medical rehabilitation on the overall field of medicine. Clinical trials now routinely employ the types of tests of overall patient physical and mental functioning, as well as indicators of quality of life, that have been developed by the rehabilitation research community. Doctors now recognize the essential truth that improvement in a physiological function is not necessarily helpful unless it enhances people’s abilities to resume their role in society and their satisfaction with their life.

Both enlarging the scope of research on rehabilitation from brain injuries to encompass basic science, particularly related to harnessing the power of plasticity, and expanding the goals of rehabilitation beyond improving function (as important as that is) to actually reversing the disabling damage are beginning to yield important benefits. The field of neurorehabilitation has added impetus to the efforts to cure stroke and spinal cord injury, increased the credibility and influence of rehabilitation among medical specialties, and extended its influence in ways that are very beneficial for the public. If we combine basic research on plasticity and regeneration with more and better clinical trials, rehabilitation medicine will no longer be limited to ameliorating symptoms but can look toward dramatic improvement and even cures, not only for the thousands of injured returning soldiers but for everyone (and that could be all of us someday) with a head injury or a neurological disease.

Disclaimer: The views expressed in this article are those of the author and not of the Department of Veterans Affairs or the government of the United States.

About Michael E. Selzer, M.D., Ph.D.

Michael E. Selzer, M.D., Ph.D., is a professor in the Department of Neurology and director of the Center for Experimental Neurorehabilitation Training at the University of Pennsylvania School of Medicine, as well as director of Rehabilitation Research and Development in the Department of Veterans Affairs. His research focuses on spinal cord regeneration and translating basic mechanisms of plasticity into rehabilitative therapies for central nervous system injuries. He is the lead editor of the Textbook of Neural Repair and Rehabilitation. He can be reached at michael.selzer@uphs.upenn.edu.