ARTICLE IN BRIEF
Three novel robotic assistance devices, one for hemiparetic gait following stroke, and two for crouch gait in children with cerebral palsy, have each demonstrated improved walking in preliminary clinical trials.
For stroke patients, a robotic exosuit made of a soft, clothing-like anchor attached to motorized cables was shown to increase the paretic limb’s forward propulsion and the paretic ankle’s swing phase dorsiflexion in both treadmill and over-ground walking.
For children with crouch gait due to cerebral palsy, one trial used a cable-driven robot called a Tethered Pelvic Assist Device, or TPAD. The laboratory-based device is designed to strengthen the extensor muscles, especially the soleus in the calves, by putting downward pressure on them during training. After six weeks of practice with the device, the children’s posture was more upright, with greater step length and toe clearance, when walking without it.
Also for children with crouch gait, the third study examined the use of a wearable exoskeleton that provides a burst of knee extension assistance at just the right moment when a child or adolescent is walking. None of the seven participants, age 5 to 17, fell while using it, and six of the seven showed postural improvements equivalent to those previously reported from surgery.
While promising, the devices will require far more testing in randomized trials before their true value can be known, said a leading specialist in neurological rehabilitation.
“These are foundational studies; they’re just beginning to get started,” said Bruce H. Dobkin, MD, FRCP, distinguished professor of clinical neurology and director of the Neurological Rehabilitation and Research Program at the Geffen School of Medicine at the University of California, Los Angeles. “The cost, safety, user-friendliness, and ability to use at differing levels of disability severity — all those are major challenges.”
Even so, each of the three devices employs a new kind of robotic assistance unlike any existing on the market.
“Most robotics for neurological injuries are heavy, power-hungry exosuits for people with spinal cord injuries who can’t walk at all,” said a coauthor of the study for stroke patients, Terry D. Ellis, PT, PhD, NCS, director of the Center for Neurorehabilitation at Boston University. “But there’s a whole bunch of people who have disabilities, who can walk, but don’t walk well. They need facilitation or augmentation to restore some of the normal components of walking.”
A ROBOT POST-STROKE
Published in the July 26 edition of Science Translational Medicine, the study of a robotic exosuit tested in nine post-stroke patients used what it called “garment-like, functional textile anchors” rather than a hard, metallic exterior. Worn on only the paretic limb, the suit was designed to be as unobtrusive as possible.
“It’s much more compatible with the real world than a rigid device would be,” said the first author of the paper, Louis N. Awad, PT, DPT, PhD, an assistant professor of physical therapy at Boston University, and a research faculty member at Spaulding Rehabilitation Hospital. “Ordinary clothes are made of soft material. We don’t don a metallic pair of pants and walk out the door. That’s our goal — robotic clothing that helps people with difficulty walking.”
Attached to cables tethered to a belt worn around the hips, the exosuit functioned in synchrony with a wearer’s paretic limb to facilitate an immediate increase in the paretic ankle’s swing phase dorsiflexion and forward propulsion (p< 0.05), according to the paper.
The improved movements resulted in a 20 percent reduction in forward propulsion interlimb asymmetry and a 10 percent reduction in the energy cost of walking, which together were equivalent to a nearly one-third lower metabolic burden — a 32 percent reduction — while walking.
Although the study did include some over-ground walking, it was not designed to test whether the exosuit had any therapeutic effects that might carry over to when patients are not wearing it.
“This is a proof of concept paper,” said Dr. Ellis. “Down the road we need to conduct trials in more ecologically valid environments, and to see if it has therapeutic value. For now we wanted to demonstrate that the device can facilitate more normal walking.”
While applauding the study as “clever,” Dr. Dobkin said it remained to be seen whether the robotic exosuit would prove to have significant therapeutic effects that would stand up in randomized trials in natural environments. He pointed to randomized trials published in recent years showing that peroneal nerve functional electrical stimulators have no greater therapeutic effect than do standard ankle-foot orthoses.
“It’s similar to all the work that was done using the electrical stimulation of the ankle,” Dr. Dobkin said. “The real question is whether it will lead to improved function when you walk over-ground. Walking on a treadmill is not terribly natural.”
He also pointed out that the nine patients in the study were able to walk on average at about two miles per hour. “That’s already pretty fast,” he said. In addition, he said, the 20 percent reduction in interlimb asymmetry is relatively modest.
But, said Dr. Dobkin, people can improve their gait by 20 percent just by sustained practice. “When you see modest changes like this with the device, you wonder if the same changes couldn’t have been achieved without it,” he said.
Steven L. Wolf, PhD, PT, FAPTA, FAHA, professor in the department of rehabilitation medicine at Emory University School of Medicine, pointed out that existing robotic devices to help people who are completely unable to walk can cost patients up to $250,000. Perhaps the exosuit might become an improvement over what presently exists both in terms of function and cost, he said.
“Most existing devices are beautiful but incredibly expensive,” Dr. Wolf said. “Is the bang in the buck? Not as yet, in my opinion. The evidence for persistent benefit from these device is just not there.”
IMPROVING CROUCH GAIT IN CP
The first of the two studies using robotic devices to improve crouch gait in children with cerebral palsy was published on July 26 in Science Robotics, led by senior author Sunil K. Agrawal, PhD, professor of mechanical engineering and rehabilitation medicine at Columbia University.
Rather than directly straighten the children’s posture, Dr. Agrawal’s seemingly contradictory approach was to increase the downward force on their pelvis as they attempted to walk on a treadmill. The tension in each wire, attached to a belt on the pelvis, is modulated in real time by a motor placed around the treadmill in response to motion capture data from cameras. Unlike other robotic devices that have been tested for treating crouch gait, the TPAD has no rigid links to the body, permitting free movement of the legs.
After training in the device for 15 sessions of 16 minutes each over the course of six weeks, the six participants showed enhanced upright posture, improved muscle coordination, increased step length, range of motion of the lower limb angles, toe clearance, and heel-to-toe pattern.
“You can see a marked difference before and after,” Dr. Agrawal said. “We heard from families and the children themselves that they were walking faster, with better posture. Now we have to see if we should use a higher magnitude of downward pull, how long each training session should be, and for how many sessions.”
Commenting on the TPAD study, Dr. Dobkin said, “The kids who were selected for inclusion were not necessarily the kind who get surgery. They had less of a crouch, a little bit more of a push-off. The question is whether training like this will lead to good over-ground walking. They got a hint of that.”
The second crouch-gait study, published on August 23 in Science Translational Medicine, involved a wearable exoskeleton designed for over-land use, and was described by the authors as the first robotic device designed specifically to treat a gait disorder in children and adolescents. Rather than force the lower limb to move in a particular way, “the exoskeleton dynamically changed the posture by introducing bursts of knee extension assistance during discrete portions of the walking cycle, a perturbation that resulted in maintained or increase knee extensor muscle activity during exoskeleton use,” the paper stated.
“In the last decade, there’s been a groundswell of work on exoskeletons, but a majority of them are designed to permit mobility after spinal injury in adults who have lost the ability to walk,” said senior author Thomas Bulea, PhD, a staff scientist in the functional and applied biomechanics section of the rehabilitation medicine department at the National Institutes of Health Clinical Center in Bethesda, MD. “There hasn’t been much done for the pediatric population who just need to improve their walking.”
A coauthor of the paper, Diane L. Damiano, PT, PhD, chief of the section in which Dr. Bulea works, said the purpose of the wearable exoskeleton is different than that of the TPAD device developed by Dr. Agrawal.
“His device is designed to strengthen the calf muscles by increasing the resistance on them,” she said. “His results were good, but this is very different from what we are doing. We have a wearable device. It’s not meant to be used in a lab for training. We’re not necessarily trying to strengthen them, although that would be a desired outcome; we are instead trying to assist their abilities to help them practice being more upright while they walk. This is something that they would wear throughout the day for several months with the goal that their posture will ultimately be improved without the device.”
A surprising observation, she added, was that some children saw it as something cool to wear.