Surgeons test-drive the amputation of the future with a mail-order limb, rerouted nerve, and prosthetic hand that grips like the real thing

WALTHAM, Mass. — The surgeons, peering through loupes with 3x magnification, have dissected away the skin, the layers of fat and fascia, and exposed most of the median nerve, the I-95 of the forearm. Sensory signals travel up this cable from the fingers and, crucially, motor signals flow back down from the brain to move the thumb toward the pinkie, giving us the opposable thumb that makes humans distinct.

Dr. Jason Souza, a plastic surgeon from Walter Reed National Military Medical Center, has reached the hand, resting palm up, where the nerve branches into exit ramps to all the fingers. He masterfully works a pair of scissors, as dexterous as a child cutting construction paper, snipping free the motor branch to the muscles in the fleshy base of the thumb.

“That’s our money right there,” exclaimed plastic surgeon Dr. Matthew Carty of Brigham and Women’s Hospital.

On this November morning, Souza and Walter Reed’s director of surgery, Dr. Kyle Potter, have met Carty at a surgical center outside Boston to invent a new operation, a way to perform arm amputations that might allow patients to move their prosthetic hands more like real ones.

The right arm resting on a blue surgical drape before them is from a cadaver, and it’s just the limb, ending at the shoulder. Nothing more. It came from a place called the Anatomy Gifts Registry, an Amazon of body parts. “You put an order in and they deliver it,” Carty said. There’s a price list for everything from limbs to brains. “It’s kind of creepy.”

It’s also essential for surgeons who need to rehearse and work out the kinks of a new procedure. “You can draw up a surgery based on prior understanding of anatomy,” said Souza, but there’s nothing like actually doing it. “It really has to be done in a hands-on fashion … because certain tissue stretches more than you think.”

Devising a new operation is in many ways like re-engineering the anatomy. Cut open, the forearm is an intricate weave of muscles, nerves, blood vessels, and tendons. But to the plastic surgeons and a group of MIT prosthetics experts working with them, it’s also an exquisite machine. “Technologically, we can’t build a spring that’s more efficient than a tendon,” said Hugh Herr, an MIT Media Lab professor who creates advanced bionic limbs, as Souza pulled a tendon taut with a gloved finger.

Over the course of five-or-so hours in the OR, the surgeons would excavate the forearm down to the radius and ulna, saw those bones off midway to the elbow, build a “bone bridge” to stabilize their loose ends, and then reassemble the dissected muscles and nerves following a wiring diagram scrawled with marker on a surgical drape. The test-drive and the give-and-take among experts from different realms would produce what Carty later described as “a couple of flashes of insight that changed the direction of what we’re doing.”

The Walter Reed crew is all too familiar with performing amputations on troops: When surgeons can’t salvage limbs blown apart by explosives or fragmented by bullets, amputations are the last resort. The procedure is largely unchanged since the Civil War, however, leading the Department of Defense to fund research on better ways to perform them that take full advantage of robotic technology integrated into the latest prosthetics.

It began by supporting a clinical trial of a radical approach for leg amputations, pioneered by Herr and Carty (and chronicled in a STAT documentary called “Augmented”). The experimental procedure attempts to preserve a patient’s sense of proprioception — the intuitive ability we have to know where our joints are in space without looking at them. It’s a sense that’s vital to mobility but that amputees typically lose.

Our muscles are arranged in pairs: When one contracts, an opposing muscle stretches, bending the elbow or the knee or ankle. This paired motion underlies proprioception, and the new approach connects up pairs of muscles left dangling during a standard amputation.

More than 20 patients have now undergone the new leg operation — named the Ewing amputation after the first patient — and the early results have been “much better than we ever thought it would be,” Carty said in an interview. Based on that experience, the Pentagon has awarded the Brigham, the Massachusetts Institute of Technology, and Walter Reed almost $3 million to come up with something similar for the arm, with 10 procedures planned at the Boston and Washington hospitals.

Only about a third as many people have arms amputated as legs, so it will be more challenging to find suitable patients. Those who have had failed wrist or elbow replacement surgery or bone fusion operations are prime candidates.

Souza and Potter have done three of the new leg amputations at Walter Reed. Their first patient is an “elite tactical athlete” who has served in Iraq and Afghanistan “and many other places,” Souza said. He was injured in combat 11 years ago, and after a series of reconstructive surgeries, decided to amputate his foot to maximize his function.

Since the operation last winter, Souza said, “he’s had less pain than he’s had in a decade,” and regained more mobility. He’s also returned to active duty.

The human hand is far more complex than our foot. We cut steak and twist doorknobs and pick up coins with it. And that means adapting the Ewing procedure to the arm is anything but simple. “Nobody needs to have functional toes in a prosthetic foot,” Carty explained. “What we ask of a hand is way more complicated than we ask of a foot.”

An experimental prosthetic ankle that Herr’s MIT team built for lower-limb amputees has two degrees of freedom at the ankle joint: It flexes the toes up and down and rocks the foot side to side. Advanced prosthetic hands, in contrast, have myriad degrees of freedom — each finger can bend at the knuckle, the thumb can bend at two joints, and the wrist can rotate — and are able to form up to 14 different grips.

“The design of the hand is actually pretty amazing,” Carty said. “But at the end of the day, it’s physics. It’s pulleys moving against each other.”

In the new surgery, the muscle couplings are recreated, using tendons as the pulleys linking opposing muscles. When the brain thinks about bending the index finger, say, the appropriate muscle pair contracts and stretches, and that muscle activation can be picked up by an electrode and transmitted to the prosthetic hand.

But there’s limited real estate in the arm, meaning the surgeons had to choose which functions of the hand were most essential to replace. When they sat down with the MIT engineers around a U-shaped conference table before the dissection, they quickly agreed that flexing the thumb and index finger, and the other fingers in unison, were a priority — leaving them just one more function to decide on.

They passed around a black metallic prosthetic hand, manipulating the joints to see how they moved. “This is heavier than I thought,” Carty said when it was his turn. It didn’t flex up and down at the wrist — the wave-bye-bye motion.

No prosthetic hands now on the market have that function, Herr pointed out, suggesting they “nail the commercially available degrees of freedom.” Souza countered that he had misgivings about “destroying an intact function” in a patient, but eventually accepted Herr’s argument.

Instead, he asked, how can they restore the thumb’s oppositional ability. “Putting a lot of money on the thumb makes sense to me,” he said, adding that it’s important to “think about what this surgery can do above and beyond” what’s possible with a standard amputation. Current sensors have to essentially guess when someone is trying to oppose the thumb, by recognizing a pattern in the disorganized contractions of an amputee’s muscles.

The small muscle in the hand that’s responsible for the opposable motion of the thumb is called the opponens, and it has a fragile blood supply, meaning it would be risky to build a pulley with it. The group brainstormed another approach, the anatomical equivalent of plugging the cord from your computer into a new monitor: What if they severed the motor branch of the median nerve and reattached it to a muscle in the forearm with a more stable blood supply — the FCR muscle that controls the bye-bye motion they no longer needed. “That muscle will inherit the function that [median motor] branch used to have,” Carty said. When patients “think about moving their thumb, or opposing their thumb, a different muscle in their forearm will twitch.” The prosthetic hand can then be programmed to read a signal from that muscle as “move the thumb.”

In the OR, they set out to test this idea. To a soundtrack of energetic pop standards, the surgeons removed layer after layer of tissue to expose the median nerve’s motor branch and the FCR muscle below the elbow. “We should be able to reach that to here,” Carty said, pointing with a set of tweezers from the opponens at the base of the thumb toward the elbow. A few minutes later, Souza snipped the nerve free from the thumb muscle and looped it up to the FCR. As they had hoped, it worked, with room to spare.

An Icona Pop hit thumped from the speaker: “I crashed my car into the bridge … I don’t care, I love it!”

The mood was light, the day already a success though they had several more hours of cutting and problem-solving ahead of them. They would later figure out that the bone bridge they fashioned was the ideal place to anchor their muscle-tendon pulleys, rather than on the ulna as they’d previously planned to do — though the bridge would restrict how much patients could rotate their residual arm.

All agreed the tradeoff seemed worth it. If we can provide proprioceptive sensation for the hand and fingers and increase thumb motion, “that would be a big win,” Souza said as they finished up their work.

After changing out of their scrubs, the team gathered back in the conference room for a debriefing. “Frankly, not everything is going to work” with the first real patient, Carty acknowledged. “So it may be that after the first one we have to go back to the drawing board.”

Still, he would reflect a few days later, “It gets us closer to feeling really confident and to look patients in the eye and say we think this is going to work.”