When Christopher Reeve became quadriplegic, there was little hope for patients with spinal cord injury. Now researchers are combining what they know about the central nervous system’s ability to rewire and regrow with a new understanding of the hidden smarts of the spinal cord to dramatically improve treatments.

Even the most devastating spinal cord injuries usually do not completely sever the link between the brain, spine and the rest of the body. Scientists are now finding ways to make the most of the remaining connections using a variety of technologies. Studies on electrical stimulation and locomotor training (a treatment that relies on human or robotic assistance during a walking exercise) suggest that it is possible to regrow damaged neuronal circuits in the brain and spine and recover some voluntary control. Some of these studies find that circuits in the spinal cord itself can be coaxed into helping the body move again.

When we walk, two sources of information are processed by the spinal cord. One comes from above: instructions from the brain about where we want to go based on what we see. The other comes from below: sensory information from the muscles, tendons and skin. After a spinal cord injury the communication lines between the brain and spinal cord are cut or dramatically diminished, depending on the severity of the event. Without instructions from the brain, doctors and researchers thought it impossible to regain any type of control over the limbs. But unlike fixed mechanical circuits, the brain and spinal cord are malleable. The axiom in neuroscience is “neurons that fire together wire together,” meaning that connections between neurons grow or atrophy based on activity.

One promising approach is to help paralyzed patients go through the motions of walking with “assistive” technologies supporting their weight. By amplifying the sensory signals that come from the joints as they move and from the soles of the feet as the pressure is rhythmically switched from one foot to the other, researchers think they can compensate for the lack of a strong brain signal. Clinicians use devices such as the Lokomat that support the patient’s weight with a harness and move his legs on a treadmill via robotic leg braces. Susan Harkema, director of the Kentucky Spinal Cord Injury Research Center, notes that when weight-supported on a treadmill, newborns show the right stepping patterns even though they can’t initiate walking on their own. This suggests that some motor “programs” are stored directly in the spinal cord, and can be triggered by sensory input. According to neurologist Volker Dietz, a professor emeritus at the University of Zurich who continues to do research at Balgrist University Hospital in Zurich, scientists have whittled it down to two essential inputs to get stepping patterns in the muscles: contact with the ground and flexion and extension of the hip joint. When researchers measure the activity of the spinal circuits of a paralyzed patient on the Lokomat, they find that the pattern is the same as the one found in healthy volunteers. The difference is that the signal is not strong enough to contract the muscles. Locomotor training is meant to increase the spinal signal.

Last year, Harkema published a study that looked at the effect of locomotor training on 196 patients with incomplete spinal cord injury. These patients had some movement or could contract some muscles in the paralyzed limb, implying that there were still residual connections between brain and spinal cord. The study showed that this training could help people with incomplete injuries. “Many of these individuals—even decades after injury—were able to walk when they had previously been using a wheelchair full-time,” says Harkema. The treatment did not work for everyone; about 12 percent of patients did not show any improvement.

Implanting Electrodes

A few years ago Harkema began testing epidural stimulation on a few patients who had not responded to locomotor training. She surgically implanted electrodes on the outermost layer of the cord and stimulated right below threshold level. Because the cord is used to receiving a lot of input from the brain, it doesn’t respond as strongly to sensory input. The purpose of this stimulation is to make the spinal cord more responsive to sensory cues so they can trigger its inherent motor programs. So far Harkema has only tried the technique on three patients. After about seven months of stimulation in combination with stand training (where harnesses support part of the patients’ weight and therapists manually position the limbs), they were able to stand and support their full weight for the first time in years. Harkema reports that one patient regained the ability to voluntarily move his toes and ankles and bend his knees 90 degrees while lying down—but only while receiving epidural stimulation. The voluntary movement was a surprise for Harkema and she concludes that “you don’t need much to execute these movements if you get the spinal cord in the right functional state.” But none of her patients have been able to take a step yet. She thinks that it may be possible to get stepping once her research team refines stimulation and the array of 16 electrodes used.

Harkema speculates that activating the spinal cord also amplifies the residual connections between it and the brain, allowing this tiny bit of brain input to be “heard” by the system.

Growing New Wires

Bulking up brain input by getting new axons—wirelike extensions of brain cells—to grow through the damaged part of the spinal cord is the holy grail of regeneration research. A set of elegant experiments in rats published last year in the journal Science show that even when only a minuscule fraction—2 percent—of connections between the brain and spinal cord are left, epidural stimulation combined with locomotor training can cajole new connections through the part of the spinal cord that was cut. Professor Grégoire Courtine and his colleagues at the Center for Neuroprosthetics and Brain Mind Institute at the Swiss Federal Institute of Technology, trained paralyzed rats with incomplete lesions using a miniature version of the locomotor training devices used for people. A harness held some of the animal’s body weight so that only the paralyzed hindlimbs made contact with the ground. Researchers gave the epidural stimulation while the rat was coaxed to walk toward a treat. After three weeks of training, the rats were able to take their first steps and after about five more weeks, they could climb stairs and maneuver around obstacles. Courtine found that after training, the lesion site became home to new axons. It recovered about 45 percent of the number of pre-lesion connections, a growth explosion considering the initial cut left so few axons.

Unlike Harkema’s patients, Courtine’s rats also got drugs that ramp up communication between neurons in the spinal cord. Both researchers agree that the next step would be to give patients this combination treatment but the drugs used in rats have yet to be approved by the FDA for people. Courtine suggests that until then, epidural stimulation in patients can be improved by targeting more than one area of the spinal cord. The idea has yet to be tested.

Locomotor training and epidural stimulation studies suggest that with some brain connections left, it is possible to regrow or enhance those connections and recover voluntary control. But there is an ongoing debate about whether a sensorimotor complete injury (the patient cannot move or feel) also means an anatomically complete injury (that there are absolutely no remaining connections between brain and cord). According to neurologist John W. McDonald, director of the International Center for Spinal Cord Injury at the Kennedy Krieger Institute, about two thirds of patients with complete injuries have some connections left. Dietz disagrees noting that the “five to ten centimeters of bleeding and crushed neurons” that result from a spinal cord injury make it very unlikely for axons to survive.

Courtine and his colleagues have shown that even with a cut that completely disconnected the brain from the spinal cord, rats were still able to walk, avoid obstacles, and even side-step after training. Although it may seem that some of these movements would need top-down control from the brain, Courtine says they are controlled solely by the spinal cord. “We have smart neuronal networks in the spinal cord that can make a lot of decisions and these networks can even learn,” he says. “What we are seeing is a selection of circuits in the spinal cord that are becoming more efficient at performing the task successfully.”

Unlike the rats with some residual brain input, these have only sensory input and can’t walk toward a treat—they can’t start to move voluntarily. “The movement is completely dependent on input coming from the legs,” says Courtine. They can only start stepping if they are on a treadmill, so that the movement of the hip joints and muscles and the changes in pressure on their paws act as triggers for the spinal programs to start.

Stimulating the Muscles

Patients who have a sensorimotor complete injury, with no ability to move and no sensation, have only a 3 percent chance of being able to move again. So far, locomotor training has not helped them recover the ability take steps. McDonald has been looking into whether a different technique, one that involves electrical stimulation of the muscles, could lead to neuroplasticity—the ability of the nerve cells to grow and change in response to new inputs. He uses cycling functional electrical stimulation (cFES), which allows a completely paralyzed patient to make cycling movements on a standing bicycle. In a study published this year in the Journal of Spinal Cord Medicine, he found that a higher percentage of patients in the cFES group improved their movement abilities compared with the percentage in the control group. For example, some patients were able to move parts of their legs by fully engaging muscles that before they could only get to twitch. Many in the cFES group recovered the ability to feel a pinprick or a light touch. McDonald believes muscle activity leads to growth of new sensory fibers and the changes he saw with these patients.

FES is not new; it has been used to help paralyzed patients since the late seventies. McDonald used cycling FES as part of Christopher Reeve’s therapy about five years after the accident, when Reeves still had absolutely no motor control, and the prognosis was permanent complete paralysis. But McDonald believed that muscle activity could reactivate the nervous system. After about three years of cycling three hours a week, Reeves was able to move some muscles in his arms and legs.

According to Dietz and Harkema, FES does not activate spinal cord circuits in the same way that locomotor training does. For this reason, Harkema thinks it is not the right therapy for patients with incomplete injuries.

While it is not clear whether FES can help patients recover the ability to walk, McDonald says that it does improve people’s quality of life. Previous studies have shown that it prevents muscle atrophy, reduces spasticity and helps improve circulation. His study showed that the treatment reduces cholesterol levels and intramuscular fat, factors that may lead to type-2 diabetes in people with spinal cord injury. Patients also recover bladder control, one of the priorities for those with this problem. One of the most important changes for patients is sensory; the ability to feel a hug again makes a big difference, McDonald says.

Today and the Near Future

Although epidural stimulation is not yet available, FES and locomotor training with the Lokomat and its equivalents are now relatively accessible. Nonprofit rehabilitation centers charge about $120 for a FES session and $175 for a session of locomotor training. In cases where insurance covers the treatment, the patient will pay about $40 per session. For those who do not live near a center, FES bikes are available for home use but cost between 15 and 17 thousand dollars.

Access to equipment is important since there is evidence that the earlier the treatment starts, the better the outcome. Dietz found that about a year after injury, spinal circuits start to “degrade.” They no longer respond in the same way to sensory inputs. He believes training and stimulation can help prevent neuronal degradation, keeping the spinal circuits open for reprogramming if new therapies for regeneration emerge in the coming years.

The therapies that are now in the experimental stages are moving forward. Courtine is working on a primate model of the combined treatment (epidural stimulation, drug therapy and locomotor training) so that he can then test it in humans. Dietz is looking into less invasive ways of providing spinal stimulation. Researchers in the field are making big strides, but Courtine cautions that there is still a lot to do: “we’re not working towards a cure, we are trying to develop interventions that I hope will help people improve their recovery.”