Advances in Prosthetic Function

Podcast Transcript

Intro: Neuro Pathways, a Cleveland Clinic Podcast, exploring the latest research discoveries and clinical advances in the fields of neurology, neurosurgery, neuro rehab, and psychiatry.

Glen Stevens, DO, PhD: The complexity of the human extremity, particularly the upper extremity in the hand, allows us to interact with the world. Prosthetics have struggled to recreate the intuitive motor control, light touch sensation and proprioception of the innate limb in a manner that reflects the complexity of its native form and function. Nevertheless, recent advances in prosthetic technology, surgical innovations and enhanced rehabilitation appear promising for patients with limb loss who hoped to return to their pre-injury level of function. In today's episode of Neuro Pathways we're discussing advances in prosthetic function. I'm your host Glen Stevens, neurologist/neuro-Oncologist in Cleveland Clinic's Neurological Institute. I'm very pleased to have Dr. Paul Marasco join me for today's conversation. Dr. Marasco is an associate professor in the Department of Biomedical Engineering in Cleveland Clinic’s Lerner Research Institute. Paul, welcome to Neuro Pathways.

Paul Marasco, PhD: Well, thank you very much. Thanks for inviting me and I’m very excited to have a conversation.

Glen Stevens, DO, PhD: Paul, I read that, kind of shocked me, I read that there are about 2 million amputees currently in the United States. And I grew up, I hate to say my age, but I grew up in the 60s and 70s and Lee Majors was the $6 million man at that point. And I suppose $6 million wouldn't get us much today, although it was at that time. So, back in the 70s, there was a lot of interest in all of this, but I'm sure times have changed and I'm really looking forward to learning more about it today. So in terms of advances in prosthetic function, can you set the stage for us? What has been the historical function of artificial limbs?

Paul Marasco, PhD: The historical function of the artificial limbs is really to try and restore some of that lost function. I mean, it's a situation where, from the perspective of lower limbs, just having simply something to walk on, instead of walking around on crutches, having something to actually step on. Those have been around for actually, there's evidence of functional prosthetic toes that fit in a socket from the 16th dynasty in Egypt. And so these technologies have actually been around for quite a while. And the interesting thing about it is these technologies, they settled on some of these early designs and they essentially worked through our history up until modern times.

For the feet and for the legs, it's important just to restore something that stands and something that allows you to stand. The other limbs are more complex, because we actually, we really engage in bi-manual manipulation with things, we hold things, our hands are part of the expressive architectures that we engage with every day. Some people would argue, but in general the upper limbs are a bit more of a complex system. And upper limb prosthetics, really early on from the 15th century there's been mechanical replacements for the upper limb, but mostly cosmetic and some basic just kind of metal devices to hold things.

During the Civil War, the prosthetic technology started to evolve much, much faster. We actually saw some systems that you could actually manipulate objects with, cable operated systems, things like that. And in World War II, a lot of these upper limb designs really settled in. And it's those things that we see every day on people. Typically, you see them still out in public, you see them in the clinic and it's these body powered or muscle powered hooks that run on cables, so that caps are called Bowden cables, essentially bicycle cables.

And then, during World War II and afterwards, myoelectric prosthesis came on board and those were motor driven prostheses, the body powered hooks are still used every day today, bioelectric devices are used and that's really where that's the state of our workspace. These devices are very functional and people use them and they're relatively inexpensive to fit and relatively easy to train with. And so that is our standard of care technology that really reached its zenith after World War II.

Glen Stevens, DO, PhD: I'm very excited to learn more about this, your research at the Cleveland Clinic has been dedicated to modifying what you just discussed, the standard of care for patients with primarily upper limb amputations. Can you elaborate on these developments with the new bionic system?

Paul Marasco, PhD: Yeah, it really boils down to control and feedback. So one of the reasons that these cable operated and motorized limbs have remained the standard of care is because anything greater than just a couple of degrees of freedom, essentially, an open, a close, a wrist rotation, an elbow flex, an extension. Once you start having to control those joints simultaneously, when you're left with, what's called two-side myoelectric control, which is essentially tapping into the electrical signals of the remaining muscles after the amputation. You flex your biceps to get the arm to flex and you flex your triceps to get the arm to extend and then you have to go through state switching in order to be able to control each of those individual joints. So it's what you might refer to as a very oppressive control system. It's difficult to use an arm like an arm when you're stuck, opening and closing your hand and states switching and then locating a wrist and then state switching and then flexing your elbow.

And so what the bionic technologies have done, what we really work on is providing more simultaneous degrees of freedom. So the idea is that we can control multiple joints simultaneously much more like a human arm works versus this sequential control that typical standard of care prosthetics use.

And then the feedback is an entirely different workspace around that as well, essentially a prosthesis doesn't feel, it's a numb tool that's on the end of your arm and you can't feel with it. And since you can't feel with it, you can't really engage in those normal, intuitive behaviors that you use every day. And it also essentially runs into this idea that it's separate from you, it's a tool and it's not part of your body. And it's not part of, as I mentioned earlier, of this expressive architecture that we engage with every day.

Glen Stevens, DO, PhD: So I think this is getting to the area of targeted sensory and motor reinnervation. Can you talk to me about that?

Paul Marasco, PhD: Yeah, absolutely. I mean, this is one of the really exciting things about prosthetics right now is there's actually many technologies for control and feedback that are starting to mature. And so we're seeing across the board, there's different offerings for the ability to control multiple joints of the prosthesis. We've got access to individual digit control and things like that. So now we're starting to see these anthropomorphic hands show up that can actually be used because they're highly dexterous and highly functional, and now people can actually use them and control them.

From our perspective, our particular neural machine interface that exists in a milieu of different neural machine interfaces, this one's actually really kind of interesting. It was developed in Chicago at the Shirley Ryan Institute by Dr. Todd Kuiken, back in the early 2000s. And essentially that procedure is that after an amputation, those remaining nerves are just left blind ended in the arm, but they still contain all the neural control and feedback information. And so what you can do is to deinnervate the muscles at the amputation site and provide a receptive environment to surgically transfer these amputated limb nerves to, and then they'll actually compete for reinnervation targets in the muscle.

It was a serendipitous discovery, but it turns out if you deinnervate the skin as well, the sensory nerves will actually reinnervate the skin itself, they'll move their way out and find new receptive neural targets in the skin. And so what you end up with is a neural, it's a biological neural machine interface, essentially what happens is the amputee thinks about moving their arm the way that they used to. That signal goes down the nerve and instead of going to the muscles of the arm that it used to it's control, it now goes to reinnervated targets in the residual limb muscles. And then those muscles contract, just like they did before they were amputated only now the contraction maps to that person's thought about how they want to move the arm versus its original function. And we go in with a computerized prosthesis and we read those, we can read those individual little contractions in different and nuanced ways to essentiallym intuitively control a prosthetic limb. So the way that it works now is the person with an amputation thinks about moving their limb and then the computerized system picks up that movement intent, and then it transfers it or translates it to the appropriate movement of the prosthesis and then all happen simultaneously. So they think about moving and the prosthesis responds accordingly.

And then with the sensory side of things, since those sensory nerves actually reinnervate the skin, and now it turns out they also reinnervate the muscles, which is one of our primary places where we work. Essentially to get the feedback, we either use small little touch robots that touch the skin, the reinnervated skin, and then we map that to the digits of the prosthesis. So when the sensors on the prosthesis feels touch, the robot translates that information to little robotic devices that push on the reinnervated skin. And then the amputee feels that touch as though it's actually their own hand.

From the proprioceptive perspective we actually can provide, this is where it starts getting really interesting and exciting, we leverage perceptional illusions of limb movement by vibrating the reinnervated muscle sensory receptors, because your muscles actually have sensors inside them that tell you what your limbs are doing, and we go in and we actually use cognitive and perceptual approaches to essentially tap into people's perceptual integration system and then push in illusions of complex grip confirmations. And so my lab really works on cognition and perception, and we actually can provide these really highly detailed synergistic grip percepts of the hand closing into a fist or down into a pinch and all sorts of different stuff.

Glen Stevens, DO, PhD: So, Paul, it sounds a lot like mind control.

Paul Marasco, PhD: No, it's really interesting, it's actually using the mechanisms that your brain uses to feel and to perceive and to understand the world around it and to really provide that information back in.

Glen Stevens, DO, PhD: And what's the timeframe that you're looking at here to see this reinnervation?

Paul Marasco, PhD: Oh, it's actually really rapid, it's a fascinating thing. Typically, nerve regeneration is about a millimeter a day and it takes quite a long time for any kind of just a basic nerve injury to heal. Targeted reinnervation is a different animal since you essentially transfer these large limb nerves to a small deinnervated nerve stump. And then there's a high level of competition that occurs at the reinnervation site and they actually reinnervate very rapidly, much quicker, probably a third of the time of a normal reinnervation.

Glen Stevens, DO, PhD: And I'm curious, it doesn't matter the hemisphere or the dominance left or right hemisphere?

Paul Marasco, PhD: It's interesting, so we have not found that it matters, we can see hemisphere specific effects. And the interesting, the kind of fun and interesting thing about it is that the post-reinnervation, after they've had these non-machine interfaces and then the other part of that equation is actually hooking these systems up in a meaningful way so that they provide the physiologically appropriate sensation and movement. We actually see very typical type of able-bodied, separations of function changes in the way that the brain processes information from side to side. And it's very reflective of the original system.

Glen Stevens, DO, PhD: So being a child of the 50s and 60s, thalidomide was around that caused a lot of limbs not to form, is this type of system, just for those that have had an amputation? What if they're born without a limb, could they undergo this type of a process or no?

Paul Marasco, PhD: I mean, that's quite a deep conversation to be had and to dive into, to start with. Most of the amputations that we deal with are traumatic and they're upper limb, presumably there are people that are investigating approaches in congenital limb difference. But one of the big questions around this is that if you're not born with a limb, do you actually have the sensory and motor infrastructure, that cortical motor, sensory and motor infrastructure to actually support that functionality? So in many respects, it has to physically be there first and have been wired up first in order to be able to use it. But I will say that I have colleagues that are actually looking into that actively and there's actually a conversation about whether, the jury's presumably still out on that. So I don't have a definitive answer for you.

Glen Stevens, DO, PhD: Is there an ideal age, too young, not good enough, too old?

Paul Marasco, PhD: That was a really interesting thing because the typical school of thought to begin was that if you server a nerve and then it degenerates, to go through degeneration and then it recedes back, and then that nerve becomes useless. And so when these surgeries were first done, they wouldn't do anyone that was further than 18 months out after their injury. Now, the really exciting part about this is, it turns out that that neural infrastructure actually stays intact. And it's almost as though the nerve is essentially waiting for something like this to happen for this act of regeneration to occur. And now it turns out that the neural architecture stays in place and functional, and is fully operational years and decades after an amputation.

And so we know when it was first done, it was only after 18 months, but then we learned very rapidly that you can do this essentially with anyone, as soon as you snip the end of that nerve to prepare it for the targeted reinnervation, it wakes right back up again, and you can move it and use it to surgically reinnervate a target system. And it turns out that actually that's the case also with some of the electrical neural interfaces as well, that that architecture still seems to be functional and intact even years afterwards. So it's a very exciting part of this field that opens a lot of doors for us.

Glen Stevens, DO, PhD: So any patient stories you can share with us?

Paul Marasco, PhD: I mean, we have such amazing patients, it's interesting, most of the people that we work with in the upper limb, the targeted reinnervation procedure was originally developed for people with very high-level amputations, who couldn't utilize a typical prosthesis because the fitting is so difficult and the control sites are gone. And so most of the people that we work with are essentially shoulder disarticulation, so someone who's missing their limb all the way up to their shoulder or high transhumerals, which is somebody who's missing their limb through the mid-point or the upper part of the upper arm. And so there's a tremendous loss of functionality associated with that. So I mean, the patient's stories themselves, what's so neat about it is our patients here are study participants, I shouldn't actually call them patients because they're research collaborators with us. And they enter into this kind of equation with us to say, really we want to try and do something different and we know that this is really hard, and then they work with us and then they're every bit of part of the research team as we are. And we actually, I mean, the beautiful part about it is we can do all these really interesting and exciting things with neural reinnervation, with the bionic arms and with all the cognitive and perceptional interactions that we engage with and then we can have a conversation. We can talk with the people that we're working with and we can sit down and we can get their impressions on things, and we can have these conversations. And the really, one of the exciting parts about it is that we've learned that there's actually two systems that we can communicate with through these bionic interfaces. There's essentially this system, which is all of the mechanistic components that the body uses to think and to move and to behave autonomously, that all run outside of conscious perception. And then there's the self-referential eye, essentially the person, the same person that you and I are talking to right now, this cognitive construct that's able to look into the system and have opinions about what's going on.

And one of the very exciting, one of our great patient stories is one of the individuals that we've been working with, one of our great colleagues from Canada, his self-referential eye, had a really different idea about what was going on in relationship to his actual system. And even though he said to us, "I just don't get this, this is really frustrating. It doesn't make sense to me." It turns out his system just worked like magic, like his system actually returned to able-bodied function and he had no idea. And it was a real sea change for us, just that simple conversation about what he was thinking and what we were seeing and being able to have that interaction and that talk totally changed the world for us. Because then we realized that we actually need to actually address and talk to, and engage with two entirely different systems that have different needs and expectations.

And if we can get into the system where you don't have to consciously perceive, then essentially all of those behaviors and things that you do naturally, that you don't ever attend to, we can get the prosthesis to engage with those. And then suddenly now it's much easier to use and is intuitive. And then you don't even know that you're operating it more effectively.

Glen Stevens, DO, PhD: Can you talk to us a little bit about your collaborative efforts with the Cerebrovascular Center?

Paul Marasco, PhD: We really tap into the perceptual integration system about people, essentially, how they feel, how their brains recognize their limbs as being human and how their brain uses the model of the world that it has built around it in order to make judgments about how it's going to move and actually how it responds to the movements that it makes, essentially, thinking forward, predicting behaviors and then compensating for errors. And so one of those systems, it's essentially your system of intuitive movement. You have a model of the world that you build in your head, it's called your internal model and it's everything that you know about the functionality of the world. And when you move and you make movements and you engage in movements, your brain says, "I think that I'm going to move this way." And then it gets feedback back from the muscles and says, "Did I move the way that I thought that I was going to move?"

And we learned about really the basic, how the system was working from a neural machine interface perspective and from a movement prediction perspective. And we realized that sometimes when you lose these sensory feedback systems, your brain loses the ability to know what it did, and then it loses the ability to correct for its own errors, and it loses the ability to learn from experience. And we realized, we had some colleagues who work in stroke, and we realized that there's a subset of strokes where people's motor system is relatively intact. But then they have very clear and present motor deficits. And intuitively you say to yourself, well, the motor systems working, why can't they move very well? And we looked at it from the sensory perspective and said, what if these strokes are actually hitting the sensory system? And everything's intact with the motor system, the motor system knows how to behave, but it just doesn't know what it did and can't correct for its own errors.

And so we used our bionic interface and instead of doing the reinnervation system, we just went to the residual nerves left after the stroke, presumably there's a few open channels there. And then we used our robotic bionic feedback system to amplify and inject the appropriate response of the movement that these individuals made. And so they reach out and then we essentially use a loudspeaker through our bionic interfaces to say, you're moving out. And when they pull their arm back, we yell in over the top of it, you're pulling your arm back. And we do that through those physiological channels. And what happened was, as soon as we turned that system on, we actually found that people in a reaching task, essentially an intentional reach and point task, their movements actually smoothed out, their movement got cleaner, it got smoother and they were actually, they didn't have to reach as far. And the interesting thing about it, the hypothesis that we're working on, is that their internal model now that it has the information that it needed, it automatically could start correcting for error and start having, have a baseline for self-reference and improved movement. So that's what we're doing with our stroke colleagues. So we're moving forward on that and see if we can actually expand that relatively limited study and see if we can actually get into date, see if we can get into some different types of hand functions and really try to see if we can use this loudspeaker approach.

Glen Stevens, DO, PhD: Well, Paul, thank you very much for joining us. This has been very insightful conversation and I appreciate your time today. Thank you.

Paul Marasco, PhD: Thank you very much.

Outro: This concludes this episode of Neuro Pathways. You can find additional podcast episodes on our website, clevelandclinic.org/neuropodcast, or subscribe to the podcast on iTunes, Google Play, Spotify, or wherever you get your podcasts. And don't forget you can access real-time updates from experts in Cleveland Clinic's Neurological Institute on our Consult QD website, that's consultqd.clevelandclinic.org/neuro, or follow us on Twitter @CleClinicMD. And thank you for listening.