Can Tibion’s Bionic Leg Rewire Stroke Victims’ Brains?

12/13/10Follow @wroush

The conventional wisdom about stroke victims is that after about 12 months of rehabilitation aimed at restoring motor control, recovery levels off. Patients never regain more movement, never get beyond whatever plateau they’ve reached by that time.

But there’s a company in Sunnyvale, CA, that may be proving otherwise. Tibion makes a robotic, battery-powered exoskeleton—in effect, a wearable bionic leg—that’s helping stroke victims make significant gains long after other therapies have stopped working.

The Tibion leg isn’t something stroke patients wear continuously. It’s meant as a new tool for physical therapy clinics, where patients with one-sided weakness, or hemiparesis, wear the device for 15 minutes at a time, three times a week. Only a few hundred patients have used it so far, but after just 12 sessions with the device, most patients studied have gained significant walking speed. One patient improved from 0.6 meters per second to 0.8, another from 0.8 to 1.0. (A normal walking speed is about 1.2 meters per second). Most importantly, these gains held—and, indeed, seemed to unleash further recovery. Six months after his sessions with the Tibion leg, the patient who had reached 0.8 meters per second had accelerated to 0.95 meters per second, according to Charles Remsberg, a veteran of the medical-device industry who joined Tibion as CEO in 2009.

It took some ingenious engineering simply to make the Tibion leg powerful enough to lift someone out of a chair, while still being compact enough to wear. But the company thinks that a different feature is what makes the device so effective at spurring motor recovery: sensors that allow the device to respond to patients’ intentions.

The Tibion robotic exoskeletonStrokes damage the neural pathways in the brain responsible for executing motor movements—meaning that “All the thinking in the world that ‘I am going to use my leg’ will not allow an affected leg to support you in standing,” as Remsberg puts it. But a pressure sensor placed inside a patient’s shoe and attached to the Tibion leg’s electronics can detect even a slight shift in weight—the first stage in standing up from a sitting position or taking a step. Such a shift triggers the robotic leg to extend, raising a patient up or swinging their leg forward. And this movement provides proprioceptive feedback that, over time, helps patients’ brains rewire themselves, so that they are eventually able to carry out the motion on their own.

At least, that’s the theory. “I can’t tell you for a fact, but the hypothesis is that we are amplifying residual intention,” says Remsberg. “This allows for an intensive level of training, which appears to capitalize on neuroplasticity, the ability of the brain to rewire around regions destroyed by stroke.” (Remsberg expands on this idea in the video on page 3.)

If larger studies prove that the device is as effective as the early results seem to suggest, Tibion exoskeletons could become standard equipment at the 15,000 skilled nursing facilities and 2,000 rehabilitation hospitals in the United States. And that could be big business for Tibion, which raised Series A funding from Oakland, CA-based Claremont Creek Ventures in 2006, as well as a $3 million “Series A extension” this May. (Remsberg himself contributed to the May round, and is the startup’s third largest shareholder.) The list price of a Tibion leg is $40,000, but the company rents them out for $700 to $1,000 per month, depending on how many hours patients put on them. The company also sells consumables such as the foot sensors and a throwaway perspiration barrier. Rehab clinics can bill Medicare for these expenses at the same rates fetched by other therapies aimed at restoring ambulation.

Some 15 clinics have rented the bionic legs, and when I visited Tibion in October, Remsberg told me he expected that number to hit 20 by the end of the year. (FDA regulations allow companies to sell certain types of medical devices even as efficacy trials are still underway.)

“We are thinking of it as a useful competitive advantage” for rehab clinics, says Ted Driscoll, a technology partner at Claremont Creek and a member of the investing group Life Science Angels. “We sell it to one clinic in Boston, and the next day the clinic across the street wants one.”

With help from Remsberg and Tibion founder Robert Horst, I took a Tibion leg for a test drive—see the video at the end of this article. While it obviously wasn’t possible for me, as an unimpaired adult, to experience the device’s alleged neuroplasticity-inducing effects, I certainly got a sense of the power that the device’s electric motors pack. I weigh about 175 pounds, and as soon as I leaned forward out of my chair, the device was able to lift my body to standing position, with very little assistance from me.

Horst says that coming up with an actuator capable of this feat was the central technological problem he faced back in 2003, when he left his previous employer, NetApp, to work full time on the idea of a bionic leg. He’d recently had knee surgery, and had realized that the state of the art in assistive devices for the movement-impaired was the venerable crutch or cane. “There hasn’t been any technology improvement since the stone age, basically,” Horst says. “Also, my family has a history of stroke and polio and other mobility problems.” In other words, it was the kind of challenge that the veteran tinkerer, who’d put in nearly 25 years at Hewlett-Packard and Tandem Computers and has 70 patents to his name, felt compelled to take on.

The problem with designing a wearable bionic leg is that it has to do two contradictory things: apply a force great enough to lift a person out of a chair, but then swing freely at the right moment. A leg that can’t do both would merely be taking over a patient’s leg, not really helping them relearn how to walk on their own. Horst says he thought electrostatic actuators might provide the necessary flexibility, but it turned out that they couldn’t produce the amount of force needed.

“Then I came up with the idea of using a continuously variable transmission,” he says. Invented by Leonardo da Vinci more than 500 years ago, CVTs are capable of stepping through a range of gear ratios seamlessly; they’re found today inside some automobile drivetrains. Horst developed a compact, electronic CVT that uses separate high-gear and low-gear motors, which are coupled and used in different phases to provide either high power or high speed, depending on input from the foot sensor and other sensors. There’s also a clutch that decouples the motors when a patient needs to swing his leg freely, in preparation for a step forward.

Other companies are working on robotic-assist devices for stroke victims and others with impaired mobility. Berkeley Bionics in Berkeley, CA, is developing a bionic exoskeleton called eLEGS, and Remsberg joined Tibion from Swiss medical device company Hocoma, where he helped build the Lokomat, a $300,000 machine that’s considered the state of the art in locomotion therapy. But Remsberg believes that these competing technologies don’t trigger the same neural rewiring as the Tibion device. “There are all these great technologies, but they basically set the trajectory, so once you press the ‘Go’ button, all you can change is the speed, not the path,” he says. “The neuroplastic effect is not seen, because the patient is not required to do anything. For robotic rehabilitation to deliver on its promise is going to require an intention-based device.”

Remsberg was persuaded to invest and take the CEO role in late 2009 after he saw the early clinical results. “In 25 years of working in this field, I have never seen anything this profound,” he says. The company is working to close a Series B funding round with Claremont Creek and other investors “in the near future,” Remsberg says, and is gearing up to for a 24-patient trial at New York Presbyterian Hospital and a 45-patient trial at a Veterans Administration facility in Gainesville, FL.

If the chronic stroke survivors in those studies show a significant and lasting improvement in their walking gait and balance, it could eventually change the standard of care in stroke rehabilitation, Remsberg says. “The medical community won’t be able to justify treating one stroke survivor with a bionic leg and one without,” he says. “There are not too many of these opportunities in a business career.”

Wade Roush is a contributing editor at Xconomy. Follow @wroush

By posting a comment, you agree to our terms and conditions.

  • Pingback: New horizons | THE ODD BLOG

  • Wombatmobile

    I think it works because it prompts a sequence of muscle actions that together comprise walking gait. This sequence is what is rehabilitated, rather than the action of individual muscles in isolation.

    What is walking? These people have considered what it is in detail:
    http://www.biomotionlab.ca/walking.php

  • Laurie Van

    I had a CVA 25 years ago at only 28 years old – 3 days after having my daughter in the hospital – a medical misadventure. I have always kept myself in good shape and trying anything to better myself – this is very interesting and would love to get more info on this and to try it out but I live in Canada?? I can be contacted at laurievan@mts.net

  • Roberta

    My husband had a severe gunshot injury to the Sciatic nerve, virtually shredidng it six months ago. His rehab doctor just mentioned the possibillity of trying this. (She primarally treats stroke victims, but his injury closely resembles stoke damage-drop foot & paralysis from just below the knee to the foot. The neural pathways were destroyed & re-attached, but very miniscule regeneration has been detected. He also has major muscle & tissue damage. The doc thought perhaps this device may help with his gait training.