The Danger Within Us

by Jeanne Lenzer

  The truth about the number of deaths caused by medical devices is that it is unknown. There is simply no way to arrive at a reliable estimate, because no one is keeping track. Not the FDA. Not the manufacturers. Not professional or trade societies. Not hospitals. Not doctors.

  Since no one knows for sure how many people have medical devices in their bodies or how many people are injured or die from implanted medical devices, their impact is not at all clear. This might seem deeply disturbing—because it is. Especially because the technology to know isn’t lacking: Walmart tracks every single head of lettuce it buys and sells and can determine how many heads of lettuce are on its shelves at any given moment, yet no one—not the FDA, not Brookings, not anyone—can say how many people are dying because of implanted medical devices. It’s a black hole.

  Complicating our understanding of implanted devices, which can be even more dangerous than prescribed drugs, are some widespread misperceptions.

  First, most of us assume that if a medical device is on the market, there must be evidence that the FDA has vetted it as safe and effective. In fact, for most devices, this is not the case. While there has been a good deal of criticism about the FDA’s failure to protect the public from drugs that are unsafe or ineffective, drug manufacturers are at least supposed to prove the value of their products in one and generally two clinical trials. What the public and many doctors don’t realize is that the same is not true for even the highest-risk implanted devices. Despite claims by the FDA that high-risk implanted medical devices have to undergo “rigorous premarket testing,” the agency requires clinical testing for only a small fraction of high-risk devices. Indeed, according to a 2009 study published in the Journal of the American Medical Association, only five percent of the highest-risk cardiac devices have undergone the sort of testing that comes even close to the testing required for drug approvals.15 * Even when devices are subjected to clinical testing, the quality of the studies performed is frequently poor, producing unreliable information.13, 14, 15

  Another common misperception is that devices are inert objects and therefore do not have side effects, as drugs do. Device manufacturers often play on this assumption, suggesting in promotional materials that patients can use devices to treat everything from epilepsy to Parkinson’s and coronary heart disease “without drug side effects.” Sure, if you get implanted with a medical device you won’t need to worry about the side effects of drugs if you are no longer taking them. However, patients often must keep taking drugs even after implantation with a device. Test subjects given the VNS device for epilepsy, for example, continued taking their seizure medicines. And for some devices, patients have to take more drugs in order for their device to function safely.

  Then there’s the array of side effects and complications caused by devices that can be even more deadly than drugs. But brochures for medical devices often bury that information in the small print or leave it out altogether.

  One study found that of 113 device recalls initiated because of a risk of serious injury or death, only 19 percent had been approved through the PMA process,* meaning that 81 percent of devices that caused the worst harm were cleared or approved through pathways that didn’t require clinical testing by the FDA.103

  To make matters worse, doctors are often unfamiliar with or even unaware of the existence of implanted devices in their patients and the side effects they cause, making it impossible to correctly identify certain problems and take remedial action.

  Unlike drugs, implanted devices require surgery, which carries its own risks, including bleeding, infection, organ perforation, complications from anesthesia, and other mishaps. Even devices that can be injected, such as certain filters intended to trap blood clots or stents placed in coronary arteries, involve invasive procedures or surgery that entails certain risks. Arteries can be (and have been) sheared during cardiac catheterizations performed to implant stents, causing massive bleeding, shock, heart attack, multiorgan failure, and death.

  Many devices present two unique and potentially serious problems: first, patients implanted with certain metallic devices cannot have an MRI scan because the strong magnetic field could cause the device to move and tear the flesh or heat up and cause burns.104, 105 Reports in the FDA database reveal that patients implanted with the VNS device have been burned, causing extreme pain, and the June 2014 patient manual for the device states that even after it is removed, if as little as an inch of lead wire remains implanted, the result can be “serious injury (e.g., burn of the vagus nerve and surrounding tissue).” Implanted devices such as breast implants can also cause opacity or distortions on imaging that prevent diagnosis of problems, such as cancer. The FDA cautions that CAT scans can interfere with the functioning of certain “insulin pumps, cardiac implantable electronic devices, and neurostimulators.”

  Computerized devices are potentially subject to hackers.106 A seemingly fanciful episode of the Showtime television series Homeland features assassins killing the vice president by hacking his pacemaker and disabling it. The capability of hackers to cause deadly harm is so real that Vice President Dick Cheney’s cardiologist, Jonathan Reiner of George Washington University Hospital, in Washington, DC, disabled the wireless feature on Cheney’s pacemaker, saying, “It seemed to me to be a bad idea for the vice president to have a device that maybe somebody on a rope line or in the next hotel room or downstairs might be able to get into—hack into.”107

  The computer security expert Barnaby Jack demonstrated that he could remotely hack devices, causing insulin pumps to deliver lethal doses of insulin and triggering pacemaker-defibrillators to deliver 830-volt shocks to a patient.108 Jack, who died the day before he was to reveal how he accomplished the hacks, said he was working with the manufacturers to secure the devices.109 Despite Jack’s white-hat attempts to spur action, the problem remains. In 2017, experts warned that some thirty-six thousand medical devices are “easily discoverable.”110 And others have said that many devices are vulnerable to malicious computer worms that could cause “mass murder.” Although such actions seem unlikely, it’s worrisome that the engineers and hackers who are demonstrating these vulnerabilities may have a parallel in the engineers who warned, years in advance of Hurricane Katrina, that the levees of New Orleans were at risk.

  And pacemakers aren’t the only computerized devices that use Bluetooth technology and Wi-Fi connectivity to allow doctors to monitor and manage patients’ devices from many miles away. Millions of Americans are implanted with deep-brain stimulators, cardiac defibrillators, gastric stimulators, insulin pumps, foot-drop implants, cochlear implants, and the VNS, which are all potential targets for hackers—in part because the FDA has never required device makers to encrypt the data transmitted by the devices.106, 111, 112 And, the experts warn, encryption is merely the first step necessary to protect the public.

  In September of 2016, Jeffrey Shuren, director of FDA’s Center for Devices and Radiological Health, signed a memorandum of understanding with various stakeholders, including Homeland Security, law enforcement, and intelligence agencies, encouraging “identification, mitigation, and prevention of cybersecurity threats to medical devices.”111

  Individuals implanted with devices, whether computerized or not, have yet another problem: turning off or removing a device is not as easy as discontinuing a drug. Patients can be weaned off even the most dangerous or addictive drugs, but many electrical devices, such as pacemakers, defibrillators, and nerve stimulators, continue to cause pain and harm while attempts are made to turn them off, and some implanted devices are extremely difficult to remove. Material from pacemakers and VNS devices often becomes embedded in human tissue, causing scar formation that traps lead wires, mesh, and tubing and making removal a dangerous and sometimes fatal endeavor. When Medtronic recalled Sprint Fidelis defibrillator leads, many patients rushed to have them removed, but removal posed its own dangers, causing major complications in 15.3 percent of patients.113 The most recent “upgraded” models presen
ted the greatest risk, with a complication rate of 18.7 percent.

  There are other sharp differences between drugs and devices that should be considered before anyone agrees to an implant. For individuals with serious problems, such as vision loss and heart-rhythm disturbances, implantable devices can be transformative and livesaving. But without adequate information about the efficacy and safety of the vast majority of devices, from spine implants to stents and surgical mesh, many citizens are being treated worse than guinea pigs. At least the results of experiments performed on guinea pigs are monitored to observe the value of an intervention, whereas postsurgical monitoring of humans implanted with devices is often nonexistent. Although the FDA is awash in monitoring programs, the reality is that they are largely ineffective, and without adequate preapproval testing, and without ongoing monitoring and enforcement by the FDA of requirements for postapproval safety studies, there is simply no certainty about the likelihood of an individual being helped or harmed by a device.

  For all these reasons, the medical device industry is one of the most powerful yet unaccountable sectors of the medical-industrial complex. Fegan’s discovery that his supposedly lifesaving VNS device had almost killed him would also prove to be just the first in a series of revelations about how great a danger device makers pose to the health of millions of Americans.

  Chapter Four

  The Man from Cyberonics

  IN AUGUST OF 2006, one month after he was discharged from the hospital, Dennis Fegan was recovering at home when he received a phone call from Dr. Juan Bahamon’s office. A secretary told him that a representative from Cyberonics was on the other line. The representative wanted to speak with him. Would he… She couldn’t finish her sentence before he shouted, “They nearly killed me! Why would I want to talk with them?” and slammed the phone down.

  A few weeks later, he learned that Cyberonics had made another call. This time they asked to have a company representative present during Fegan’s next appointment with Bahamon. The neurologist agreed. By the time Fegan was informed that the Cyberonics representative would attend his one-month follow-up, he had begun to cool off. He began to think about why the company might want to meet with him. Since he believed that Bahamon had always looked out for his best interests and had agreed to the meeting, he reasoned that perhaps the company was going to offer him money. Maybe lots of money! Money to shut him up. After all, he’d been telling everyone he knew about what the VNS device had done to him.

  He recalls the meeting with a bitter laugh: “If they offered me a check for a couple hundred thousand dollars I might have taken it.” But offering Fegan money was not what Cyberonics had in mind. Fegan says,

  Dr. Bahamon and I were in the office for just a few minutes before the Cyberonics rep came in. He was young and well dressed. The first question Bahamon asked him was, ‘Are you a doctor?’ The rep said no, but that he was trained to read EKGs. All I remember is him looking at my ER report and looking at the [heart monitor] strips. He didn’t ask me any questions. It seemed like I was there just as a spectator. Then he asked Bahamon to perform a lead test. I could tell that Bahamon was uneasy to begin with, but when the rep asked him to do a lead test, he became angry. He raised his voice and said he wouldn’t think of doing something like that unless I was in the hospital.

  Fegan remembers a couple of other details from his encounter with the Cyberonics representative, Steven Parnis. He asked Parnis (whose title was senior manager of clinical engineering) whether Cyberonics planned to report his “incident” to the FDA. Fegan says Parnis “danced around” an answer, failing to commit one way or the other. He also remembers Parnis telling him and Bahamon that the company hadn’t seen other cases like his. In fact, he insisted that Fegan’s experience was an isolated occurrence and pointedly suggested that if Fegan hadn’t begun taking the blood pressure and tremor-reducing drug Inderal, perhaps his heart wouldn’t have stopped.

  The meeting ended abruptly and without any discussion of compensation. Fegan, hesitant to speak up, didn’t ask any other questions of either Parnis or his doctor. He left feeling puzzled more than anything else.

  When he got home, he looked up “lead test” on his computer. Even when he learned that the VNS device would have to be turned back on to do a lead test, it took a while for the implications to sink in. When they did, he grew angry: Cyberonics was asking his doctor to turn the device back on. The device Fegan wanted out of his body. The device that had almost killed him and that could kill him if it was reactivated.

  His indignation mounted into a slow-burning fury that roiled his insides and fueled many late nights as he sat in the dark bedroom he’d turned into an office, searching the Internet to learn everything he could about the vagus nerve stimulator. He quickly learned that stimulation of the vagus nerve slows the heart rate. It seemed all too plausible that stimulation with an electrical device could slow the heart rate too much—even stop it.

  Shortly after the visit from the Cyberonics senior manager, Fegan received a letter from Bahamon stating that he could no longer care for Fegan; he had thirty days to find another neurologist.114 Bahamon didn’t give a reason for dismissing Fegan from his patient roster; presumably he was fearful that Fegan might sue him. But Fegan believed that Bahamon had saved his life when he raced back to his office to get his programming wand so he could turn off the device, and he was disappointed to lose his long-term neurologist. He never did file a suit or a complaint against Bahamon and still doesn’t fault him for referring him to a surgeon for the VNS implant. Fegan believes that Bahamon was a victim of Cyberonics’ claims, just as he was. He concedes that maybe Bahamon could have looked into the device’s track record further, but after all, the doctor was only trying to help him when he made the recommendation.

  To this day, Dennis Fegan is reluctant to find fault with any of the physicians who cared for him. And perhaps he is right. There is plenty of blame to go around and no shortage of others who played a role in Fegan’s long ordeal.

  * * *

  If Dr. Juan Bahamon was merely a bit player in the story of Dennis Fegan’s near-death experience, what about Cyberonics, the company that had developed, designed, built, and marketed the deadly device?

  Few physicians and even fewer patients know the history of Cyberonics, * despite the fact that approximately seventy-five thousand people with epilepsy have been persuaded to bet their lives on the quality of its product. The story of the company is a fascinating example of how the intertwined forces of scientific exploration, medical need, government regulation, and good old-fashioned capitalist greed are now shaping the American healthcare system, for good or ill. And it’s a story that takes us back to the pioneering work of Dr. Eugene Braunwald.

  The vagus nerve is one of the most remarkable components of the human body. It is the longest nerve in the autonomic nervous system, winding from the brain down through the neck, chest, abdomen, and pelvis and sending off branches to almost all the body’s major organs. This means that stimulation of the vagus can produce an amazing variety of results—including slowing the heart rate, aiding digestion, and promoting orgasm.

  According to the late emergency physician and Harvard Medical School alumnus Francis Fesmire, stimulating the vagus nerve can even cure hiccups—no laughing matter when they persist for days, weeks, and even months, an exhausting affair that makes patients and their doctors grow desperate for a cure. And Fesmire found one. In 1988, he discovered that wiggling a finger in the rectum of a patient plagued with intractable hiccups made the problem vanish. The reason: the vagus nerve stimulates the diaphragm, the huge plate-like muscle at the floor of the lungs, which regulates breathing. When electrical impulses to the diaphragm go haywire, they can cause spasmodic contractions of the muscle, resulting in hiccups. Fesmire reasoned that digitally stimulating the terminal branch of the vagus nerve in the rectum could reset the electrical impulses and stop the spasms. Of course, no self-respecting academic would dream of announcing such a discovery i
n plain language. So he published his findings in Annals of Emergency Medicine under the title “Termination of Intractable Hiccups with Digital Rectal Massage,” a publication that eighteen years later would win him one of the notorious Ig Nobel Prizes, presented at an annual ceremony at Harvard’s Sanders Theatre. Author and musician Amanda Palmer has described the ceremony in her 2012 blog post, “Prizes for silly science: Ig Nobels tonight” on Ready or Not, Here Comes Science, as “a collection of, like, actual Nobel Prize winners giving away prizes to real scientists for doing fucked-up things…it’s awesome.”

  As Fesmire concluded his acceptance speech at the 2006 ceremony, the audience erupted in laughter when he displayed “Dr. Fran’s Anti-Hiccup Kit” and announced that he was giving away a couple of dozen of them. Each contained just three items: a three-by-five-inch instruction card that displayed a picture of Fesmire, a latex glove, and a packet of lubricant.

  In 1967, Eugene Braunwald discovered that stimulating the vagus nerve could be put to other uses. He knew that stimulating the vagus slows the heart rate, dilates blood vessels, and reduces the force of the heart’s contractions, effects he believed could reduce heart pain, or angina pectoris. It was this theory that led Braunwald to implant the first VNS device in humans that same year and to describe in the prestigious New England Journal of Medicine the “striking improvement” in the first two patients he treated for angina by stimulating the carotid sinus (which causes vagus nerve stimulation).115

  In 1970, Braunwald reported in the Western Journal of Medicine that early results of the treatment were “sufficiently encouraging to warrant continued trials of this new mode of therapy.”116 But buried in that report was a brief mention of a detail not included in either his conclusion or his later statements about the research: two of the first four patients had died in the immediate postoperative period—one from a heart attack caused by a critical slowing of the heart rate and the other from a drop in blood pressure. Both effects were known results of vagus nerve stimulation.