The Man Who Touched His Own Heart

by Rob Dunn

  As years elapsed, the large teams attempted to achieve the joint goals of energy conversion and pumping. By 1972, both teams had made progress. The relative success of the AEC effort depended to a great extent on one man, Willem Kolff, who was, by then, collaborating with Westinghouse. Kolff was as serious a researcher as one might hope to involve in the quest for an artificial heart. He was, at the time, at the Cleveland Clinic, where he worked in the same building as Sones, René Favaloro, and Donald Effler. Kolff had already, by this point in his career, invented an artificial kidney from which modern dialysis machines derive their inspiration. (He built the machine out of parts of a downed Luftwaffe plane, the radiator of an old Ford, orange-juice tins, and artificial sausage skins.)9 But more to the point, he and Tetsuzo Atsuko, a Japanese engineer, had already reported, at a meeting of the American Society for Artificial Internal Organs, on the development of an electric artificial heart made out of plastic with which they’d kept a dog alive for ninety minutes. In the early 1970s, Kolff had reenvisioned this original model with a plutonium-based energy source and created a working version, but in every model, the heart was either too big to fit into a human or too weak to pump blood.

  Separately, the NHI team eventually created a twenty-four-ounce device that took atomic energy (from the decay of plutonium 238)10 and converted it into a tiny steam engine. All of this took three years longer than anticipated yet at least occurred. But there was more trouble. Just as for the AEC’s models, each version of the NHI pump and energy source that could be made small enough to fit inside the body was too weak to do the full job of both ventricles of the heart. In this context (and while the AEC continued to try to produce a fully capable replacement heart), the NHI team tried something different. They changed their goal: instead of building an artificial heart, they would start by making a device that would provide assistance to the left ventricle (which sends blood through the arteries out to the body). It was more modest—a plane to London rather than a spaceship to the moon—and yet still useful.

  The next step (en route to London) was attaching a small nuclear energy source and a pump to the heart of a cow. They would intentionally damage a cow’s heart and then attach the device, at the descending thoracic aorta, leaving the left ventricle; there, the device could take a weak flow of blood (a result of heart failure) and give it more force. The pump would beat in response to the heart’s own electrical impulse, thereby eliminating any need to incorporate a pacemaker into the heart; the device would rely on the real heart’s natural pace.11 In February of 1972, their assist device was implanted into a calf with an intentionally damaged heart. It worked. The heart pumped blood as if it were still sound, or at least it did for eight hours, until the inflow tube in the device kinked.

  Meanwhile, DeBakey, who had been given his own pool of money by the NHI ($4.5 million), had also made progress in Texas, albeit on a nonnuclear artificial heart. With a nonnuclear heart, the accepted reality was that the heart would be plugged into the wall. There was no alternative, and yet for patients with no other solution, this seemed, at least at the time, like progress.

  The details of the use of DeBakey’s heart are subject to (angry) debate. Domingo Liotta, an Argentinean surgical researcher in DeBakey’s employ, wrote a conference abstract in which he claimed to have inserted the devices that he and a team in DeBakey’s group had developed into each of ten calves successfully. The day after that claim, he actually performed what is regarded as the first successful implantation of a ventricular-assist device in a cow. The pumps were then put in seven additional cows, with other surgeons present, and all but one cow died within hours. It appears Liotta might have been anticipating his own success in his abstract, success that did not quite occur.12 Then things really got strange.

  The Texas Heart Institute’s Denton Cooley, DeBakey’s onetime mentee and soon-to-be rival, asked Liotta to build another of the devices he had produced for DeBakey (without mentioning anything to DeBakey) and give it to him so that he could implant it in a human. Liotta asked an engineer to build the device, which the engineer, thinking he was doing work for DeBakey, promptly did. Because the engineer did not pass the machine off directly (he left it to be picked up), he made sure to leave a note on it that said it should not be used on a human because it did not yet really work.

  Once Cooley had the device in hand, he waited until DeBakey was away at a conference and then, on April 4, 1969, searched for a potential recipient. He found Haskell Karp. Karp, a patient who needed a heart transplant, had not been able to obtain a heart. One donor arrived, but her heart was in poor shape. A second donor never arrived, and so Cooley installed Liotta’s device in Karp as a stopgap until a donor heart could be found. Meanwhile, Karp’s wife spearheaded a national search for a donor (donors had become ever more scarce as the success of transplants became ever less obvious). She called out, “Someone, somewhere, please hear my plea. A plea for a heart for my husband. I see him lying there, breathing and knowing that within his chest is a man-made implement where there should be a God-given heart.”13

  One day later, after a potential donor heart arrived in Texas in a condition too poor to be useful, Karp died. After Karp was brain-dead, for reasons that remain unclear, Cooley then transplanted a heart into Karp. The many newspapers that carried the story of Karp’s artificial heart said that “the surgery was successful” but that Karp died of other causes. It did not mention the later heart transplant, and, in addition, scholars debate whether the surgery was, even momentarily, a success. When DeBakey saw the news, he was furious and, after a bout of screaming rage, did not associate with or talk to Cooley for decades. At the end of his life, feeling generous, DeBakey extended an olive branch, and the two were, for a short period, reconciled. Liotta’s heart was never used again.14

  The DeBakey-Liotta-Cooley heart was not atomic, and yet, even ignoring the issue of long-term power, it still could not replicate a real heart well enough to sustain Haskell Karp (among other problems, the blood in the artificial heart clogged). This should have been a clue that producing an atomic artificial heart was going to prove difficult. Initially, the advances on the atomic heart and ventricular-assist device (whatever its power) had seemed quick enough that goodwill and hope ruled. To many medical researchers, such a heart had once seemed inevitable. In 1964, DeBakey had imagined an artificial heart to be likely in the next ten years. In 1966, Glenn Seaborg openly spoke of a future in which failing hearts would simply be replaced with nuclear ones. The technology was on track for some kind of success, even if it was more modest than had originally been envisioned. But then progress slowed. Cooley’s hubris tarnished the perception of artificial hearts and, perhaps, tamped down some of the eagerness to move forward.15 By 1976, forty-one individual atomic ventricular-assist devices had been tried in cows, but the devices were little more advanced than what Liotta had already tried, and they were not nuclear.16 In 1979, a New York Times article summarized the state of affairs as follows: “Today, after 15 years and an expenditure of over $125 million a clinical, practical artificial heart for humans is nowhere on the horizon.”

  Some challenges were technical. Others had to do with the failures of collaborating labs and entities to actually collaborate rather than compete. But as the project matured, it took long enough that the perspective of scientists on the idea of an atomic heart, or an artificial heart more generally, had changed. In 1960, there were few rules for how technology could be incorporated into modern life; there was no FDA, and scarcely anything in terms of informed patient consent. The Atomic Energy Commission was charged with policing itself, for example. The ethics of artificial organs were discussed, but in academic circles rather than legal ones. But by the late 1960s, many had started to become wary of technology and, specifically, of the consequences of implanting technology into the human body. Intrauterine devices had been implanted in millions of women with tragic consequences that were beginning to be realized. Suddenly a groundswell of sentiment rose up to su
ggest that the devices implanted into bodies needed to be controlled. A new set of rules was developed for implantable devices. The rules called for a ranking of devices from low to high risk. The mere act of ranking made it clear that whether or not the atomic heart was risky in an absolute sense, it was risky relative to everything else that had been proposed. So were artificial hearts in general. With a single bit of regulation, the project went from being regarded as bold and ambitious to being regarded as both literally and figuratively toxic. Slowly, and then quickly, funds for the atomic heart project disappeared.17 Seaborg returned, for a while, to his role as a professor at the University of California, Berkeley, and appears to have rarely mentioned the project again. The leaders of the Artificial Heart Program moved into other fields. And then there was DeBakey. DeBakey, who helped to initiate the endeavor to build an atomic heart, had decided to focus on something else: he was trying to perfect battery-or electrically powered ventricular-assist devices. This was, although he would never admit it, a far more mundane endeavor—neither a complete heart nor an atomic one. Today, millions of patients each year receive ventricular-assist devices that aid in the beating of their hearts; they connect via a power cord (going through the chest wall) to several large batteries that must be recharged every two hours. In other words, we have not yet surmounted the problem of power; our best batteries are still humble relative to the efficient power that arises from the work of millions of living cells.

  Of course, the initial route that DeBakey had charted, with the goal of building an artificial heart, was seductive enough that some continued to try. They were driven by the extent of the problem—10 to 20 percent of the more than five million patients with congestive heart failure died each year before they were able to get transplanted hearts or find other solutions. No one would ever mention nuclear hearts again, though (and little has been written about the entire episode), and so the entirety of the body of work on artificial hearts since the late 1970s has focused on large devices that plug into the wall.

  The first well-planned attempt to implant a total artificial heart came in 1987, sixteen years after it had been predicted to happen. Dr. William Kolff had by then moved from the Cleveland Clinic to the University of Utah, where, as director for the Institute for Biomedical Engineering, he led a team of more than two hundred doctors and scientists in studies of artificial organs in general. There, he worked with William DeVries and, beginning in 1971, Robert Jarvik. Jarvik was not a practicing doctor. In fact, Jarvik could not get into any medical schools in the United States and so traveled to the University of Bologna to be trained, but he dropped out two years into the program. Disillusioned, he returned home to the United States and decided to try for a master’s degree in biomechanics. This time he finished, and it was on the basis of that degree that he landed a job working with Kolff as a tinkerer. With Jarvik watching, Kolff and DeVries installed the Jarvik-7 heart (Kolff had a tendency to name devices developed in his group for those in the lab as a way to encourage them to continue to work for him) into Barney Clark, a retired dentist. Clark was too sick to be eligible for a heart transplant, and the Jarvik-7 had just been approved by the FDA for human implantation. Barney’s new heart had two ventricles and six valves made of titanium. The ventricles squeezed the blood, which passed through the titanium valves and out into the body, then returned through the other valves. It was powered by a pneumatic pump attached to a tube that trailed out of the heart (and out of the patient) like a sort of tail. Outside Clark’s body, the pump was the size of a washing machine. One hundred and twelve days passed with Clark relying on this heart—miraculous days in the history of medicine. The New York Times heralded the success and suggested a fully implantable heart might be further off than originally expected but not more than ten years (which would have been 1994).18 It was not, however, miraculous from Clark’s perspective. At first Clark was delighted with his new heart, but then the problems began. Infections, a persistent issue with artificial hearts and assist devices, attacked his body chronically. His blood clotted in the machine. He suffered from strokes. He was not conscious for many of the 112 days, but when he was, he asked to die.

  Only modest improvement has come since the implantation of Clark’s heart. After Clark, a Jarvik-7 heart was implanted in another patient, Bill Schroeder, who lived longer, 620 days, long enough to take a call from President Ronald Reagan. But the extension of Schroeder’s life, like that of Clark, was temporary and medicalized, not the great hope. Several companies now market artificial hearts. Versions of the Jarvik-7 have kept patients alive for months and, in a small handful of cases, a few years, but those years are rough. For a while, the Jarvik-7 was used as an “investigational device,” which meant that it could be installed only in extreme cases in which there was truly no hope. And even then, its continued use depended on progress in its improvement. Then, in 1991, the Jarvik-7 lost its investigational status (too many years had passed without progress). New trials were halted until the device was able to renew its investigational status with a new name, CardioWest. New versions of these artificial hearts have electrical lines that run to large batteries in a backpack. This backpack of batteries, like the ones for assist devices, must be recharged every two hours. Ventricular-assist devices have become a relatively common means to help support hearts until they can recover (if they are able to) or as stopgaps until a heart-transplant donor can be located.19 True artificial hearts are used when the heart requires more than assistance, when the living heart has nothing left to give. While this extra time is helpful (an artificial heart with a tiny internal component has even been used successfully in a baby in Italy), it does nothing to resolve the shortage of hearts. If there is a lesson here, it is that it is harder to produce a mechanical heart than it is to land a man on the moon or build a guided missile or a subsonic bomber, far harder. What evolution does with cells and mitochondria, we cannot yet do with metal, plastic, and batteries. This is a sentiment that has become most clear in the land of Salvador Dalí in an old church at the edge of Barcelona.

  In Catalonia, the quixotic land of Salvador Dalí and Antoni Gaudí, a computational physicist named Mariano Vázquez has decided to build an artificial heart. Whereas DeBakey and Seaborg imagined devices that replicated the function of the heart (what the heart does), Vázquez seeks to model how the heart does what it does. His artificial heart will never be placed inside a body. It is a computer heart, a simulation. The idea came to Vázquez while having food and beers with a friend. Until that point, Vázquez had focused on engineering challenges—for example, how to build a better toilet or rocket. But his friend, in the way that friends will, asked him big questions. Why not focus on something more beautiful, more challenging, more interesting? Why not study the human body? If you can do anything with your life, why not try to make a working replica of, say, the heart? The heart is beautiful. The heart is mysterious. The heart, unlike a toilet or a rocket, has been shaped by millions of years of evolution. Vázquez, an Argentinean, had seen Domingo Liotta on television when he was growing up. He knew at least the rough outlines of the story of attempts to make an artificial heart. He would construct a simulation of a working heart instead. Vázquez and his colleagues at the Barcelona Supercomputing Center’s project Alya Red decided to build this simulacrum by mimicking the ways in which individual muscle cells signal to other muscle cells in order to move the heart’s liquid, its blood (much as another project mimicked the ways in which the flushing of a toilet moved other, less noble, liquids).

  It is worth noting a beating heart does more than beat. It also rises to challenges. If a tiger chases you, a number of things will happen in your body. The amygdala in your brain will trigger a signal telling your body to run. That signal will travel to the adrenal glands, where the adrenal medullas release adrenaline. The adrenaline then makes its way to your heart’s internal, biological pacemaker and speeds things up. It also makes the heart’s beats more forceful. The adrenaline causes the cells of the heart
to allow more calcium in, which makes more of the cells of the heart contract. The heart contracts both more often and more fully. If you find yourself running from a tiger, you will be very grateful for these responses of your heart.

  This rapid reaction to a tiger is not the only response your cardiovascular system can carry out. Your heart also has sensors that allow it to detect how much blood is being pumped out of the ventricles. If too little blood seems to be in the system, these sensors trigger the production of more blood. They also trigger contraction in the arterioles of all but the most central organs (heart, brain, lungs). The heart is capable, in other words, of letting your fingers get a little cold to keep the heart, lungs, and brains from dying.

  An example of a heart used in the Alya Red artificial heart models. Each noodle-like line is a muscle fiber based on imaging of a real heart. The Alya Red model is the best “total artificial heart” in the world, albeit one that is entirely virtual. (Courtesy of Mariano Vázquez & Guillermo Marín, Barcelona Supercomputing Center)

  Vázquez and his team knew all the ways a beating heart did more than beat, and, in light of those complexities, they decided not to worry about any of them; they just wanted to replicate the details of an ordinary, resting heartbeat. To simplify things further, they decided, at least initially, not to worry about the movement of blood either. They would just simulate an unexcited, beating, bloodless heart. (It was probably a good decision; the Wright brothers, after all, did not make their first flight during a thunderstorm.) To do this, though, Vázquez and his team needed to know exactly how signals moved through hearts, and this required them to use a new tool to image the heart. Collaborating with scholars at the Computer Vision Center at the Autonomous University of Barcelona, they took high-resolution MRI scans of living hearts at a resolution of thirty-six micrometers, the width of ten red blood cells. These images allowed the team to input data into their models on the paths of the muscle fibers in the heart, paths they turned into a kind of digital skeleton. They then modeled the ways in which an electrical impulse would travel along hundreds of thousands of simulated heart fibers overlaid on that high-resolution skeleton. The model specified the rules by which each of those fibers would contract when stimulated by adjacent fibers.