Tower’s evidence of a possible link between cobaltism and hip replacements was important and worthy of publication. Yet Tower was shunned on multiple fronts as he sought to get the information out to doctors and the public. His report was delayed by a medical journal whose subsection editor and peer reviewers had extensive financial conflicts of interest. When the FDA finally decided, in 2012, to appoint an advisory panel for metal-on-metal hips, William Maisel, deputy director for science and chief scientist at the FDA’s device center, turned down Tower as a panelist, referring him to James Swink of the FDA’s Medical Devices Advisory Committee, who told Tower that panelists had to be nominated by industry or by the American Academy of Orthopaedic Surgeons (AAOS).279
Unfortunately, there is reason to doubt the impartiality of the AAOS. Like medical journals and conference organizers, professional medical associations often take large sums of money from industry. Industry money, lavished on every sector of healthcare, creates a feedback loop that encourages publication of positive messages about devices, while reports of harm remain relegated to the sidelines. Like other medical associations, the AAOS is supported by manufacturers who pay handsome fees to promote their wares at the association’s annual conference.* Besides the institutional conflict of interest, many individuals in leadership positions at the AAOS have extensive ties to industry.
In the end, on January 17, 2013, more than two years after Tower tried to bring the problem to public attention, the FDA issued a “safety communication” in which it acknowledged that metal-on-metal hips could cause metallosis, which could lead to thyroid disorders, mental and nerve disturbances, psychological problems, vision and hearing problems, and heart failure.
By November of 2013, Johnson & Johnson had set aside $4 billion to settle lawsuits involving the DePuy metal-on-metal hips.280
In 2016, a careful meta-analysis (a study of multiple studies) of death rates among hip-implant recipients was published. The researchers found that patients with metal-on-metal hip implants were more likely to die early than individuals implanted with other types of artificial hips.281 The two study sponsors were the Leiden University Medical Center, in the Netherlands, and the FDA.
Although DePuy stopped producing ASR hips in 2010, it had another product, the Pinnacle hip, which went on the market in 2002, and although the Pinnacle hip includes metal-on-metal implants, it remains on the market as of 2016. However, in March of that year, a Dallas jury awarded $502 million to a group of patients who said DePuy hid the Pinnacle’s design flaws, which led to early failure and metallosis.280
Steve Tower says the ASR XL hip is implanted in only 10 percent of the one million patients who have with metal-on-metal hips, which leaves nine hundred thousand patients who are also at risk—and an even greater number of patients with metal-on-plastic hips who may also be subject to metallosis. Tower wants to see patients protected through systematic screening programs like the ones now recommended in Australia and the European Union. Currently, routine screening in the US is not recommended.
* * *
Douglas W. Van Citters walks through a series of laboratories on the leafy campus of the Thayer School of Engineering at Dartmouth. The labs are collectively referred to by some doctors as the Dartmouth “device morgue.” Van Citters, a tall man with a friendly manner and a boyish face, is associate professor of engineering and co-director of the college’s Biomedical Engineering Center for Orthopaedics. He, his colleagues, and his students are on a constant quest to determine why and how devices fail.
As he walks through the labs, Van Citters points to some of the sixteen thousand medical devices that he and his colleagues have examined. They are sent to the morgue by doctors and hospitals from around the country.
The devices, which were once implanted in living humans, are now housed in museum-like displays and boxes. Each device has been explanted (removed) from a person—living or dead—for a variety of reasons. Some released metals that changed their human owners’ muscles, hearts, or brains. Some broke into pieces. Some just stopped working. Each device has been sent to the morgue for analysis so the researchers can understand what went wrong, and in some cases, to understand what was right.
Van Citters stops beside several attractive wooden tables topped with glass enclosures that display dozens of explanted hip joints of various shapes and sizes. Some are plastic. Others are ceramic or metal. One particularly dubious-looking device has tiny bits of visible bone still attached. Like the other devices, this one is labeled with an identification number, below which is written: 22 YRS IN VIVO.
Another display reveals a large human thigh bone sliced open longitudinally. Protruding from the top of the bone is a gleaming artificial hip composed of a solid metal ball atop a long swordlike metal shaft that pierces the center of the thigh bone and extends through its marrow. What happened to the human owner of the thigh bone is a mystery. Van Citters is careful not to say too much. Confidentiality is maintained at the morgue, and only identification numbers are affixed to devices—no names.
Van Citters is the engineering equivalent of a forensic pathologist who conducts autopsies to determine the cause of death of murder victims. Many of the devices come from the Mayo Clinic, which keeps close tabs on device failures in hopes of avoiding future problems. Van Citters says his lab conducts testing not only to detect problems but also to enable the design of better devices by motivating “basic science research in tribology, * corrosion, materials, biomechanics, and so forth.”
In certain parts of the lab, it’s hard to hear Van Citters’s voice over the roar of machines of various sizes and shapes that whir and grind, pump and rotate, as they shake, pound, and hammer away at their captive devices. Connecting the whirring machines is a dizzying network of clear tubing through which various chemical concoctions pass, bathing the devices in the numerous decay-inducing fluids that they might be exposed to in living humans.
Engineers examine micron-thin sections of devices with microscopes and even an electron microscope, searching for clues to material breakdown and corrosion. Oxidation is a particular curse. One doctor who sends explanted devices to the morgue is Stephen Tower, the orthopedic surgeon. Van Citters and Tower share a mutual respect. They consult each other on things that have gone wrong and on the ways in which devices can be improved in order to avoid future failures.
Around 2015, visitors from the FDA came to Dartmouth for an “Experiential Learning Program.” Van Citters says interested parties, whether lawyers or industry, can be invested in certain findings, but he says he tries to remain neutral, “like Switzerland,” without advocating any particular position. He deals in numbers and hard science—not opinion. If a device has a defect, he’ll point it out. If not, if some other factor is at play, he’ll call it that way. In the meantime, the flaws he finds help him to develop materials and designs that are resistant to the problems he sees.
The quest for truth about the safety of medical devices is an inherently challenging one. Medical science is ever changing, human biology is enormously complex, and the factors involved in tracing cause and effect can be almost unlimited. Unfortunately, under the current healthcare system, the quest for truth is further hampered by the willingness of researchers, medical journals, and regulatory agencies to serve the interests of powerful corporations—even when the health of patients is put at risk.
Chapter Ten
When Money Talks
THE PROBLEMS THAT PLAGUE the medical device industry reveal a troubling pattern of financial kickbacks, perverse incentives, and institutional conflicts that too often give short shrift to the needs and safety of patients. The case of Infuse, a spine implant, offers a number of similarities to the VNS story.
In 2009, Kathleen L. Yaremchuk, chair of the Department of Otorhinolaryngology–Head and Neck Surgery at Henry Ford Hospital in Detroit, raced to the bedside of a woman who couldn’t breathe. She had no idea that she was about to come face-to-face with a mysterious outbreak in patients who had rec
ently undergone surgery.282
With more than two decades of experience under her belt, she had rescued countless patients who couldn’t breathe from the brink of death. The causes ranged from major trauma in the aftermath of car wrecks, shootings, and strangulation to pneumonia, cancer, and severe allergic reactions. Yaremchuk was often called by her colleagues to perform tracheostomies, a surgical procedure in which she cuts a hole in a patient’s windpipe and inserts a plastic tube that can be connected to a breathing machine.
When she arrived at the patient’s bedside, the woman was in severe distress. She’d undergone cervical spine (neck) surgery several days earlier. In order to sort out the cause of the woman’s respiratory distress, Yaremchuk asked to see the CAT scan of the patient’s neck. She was unprepared for what she saw: “My first reaction was confusion. The patient had this massive inflammatory reaction in her neck. It wasn’t an abscess or an infection. There wasn’t anything that could explain the massive inflammation. It was something we’d never seen before—something that hadn’t been described or talked about.”
For Yaremchuk, the patient’s presentation was so different, so new and unexpected, that she said, “It was like looking at the first AIDS patient.”
She would soon get a clue to a possible cause of the woman’s strange condition. As chair of the department of otorhinolaryngology, Yaremchuk is responsible for overseeing the training and work of resident doctors in the specialty commonly known as “ear, nose, and throat,” or ENT. When she mentioned the woman’s case to the young doctors, a few residents rolled their eyes and said they’d seen a couple of similar cases. One of them mentioned that neurosurgeons were implanting a new device during spine surgery that contained rhBMP-2 (recombinant bone morphogenetic protein 2), a biologic bone stimulator that is a component of a device known as Infuse. Yaremchuk wondered if this could be the cause of the problem. But she was circumspect: in medicine, mistaken associations are a dime a dozen. Yaremchuk says, “We were seeing something new, and I thought, ‘Okay, as chair of the department, it’s my responsibility to figure out what’s going on.’”
As she puzzled through the problem, she says she realized there were two very unusual things about this spate of patients with respiratory distress. First was the fact that the distress was occurring at all, because there was nothing new about cervical spine surgery: these surgeries were common and had been done “for forever” without similar problems, she said. Second was that in the rare instances in which respiratory distress did occur in the past, it tended to be caused by bleeding that occurred either during surgery or in its immediate aftermath—not several days later and not caused by massive inflammation.
Yaremchuk dismissed the idea that a new neurosurgeon might be the source of the cases gone wrong. It seemed unlikely. These weren’t botched surgeries, and a quick check showed that the patients with complications weren’t treated by any one surgeon or group of surgeons.
Was it possible that rhBMP-2 was the culprit? She talked with some of the neurosurgeons. They assured her that rhBMP-2 was safe. Studies had shown it, and besides, it was in wide use.
But Yaremchuk wasn’t convinced. Any single neurosurgeon might not be able to see the big picture, because he or she might not see enough patients to realize there’s a problem, but Yaremchuk and her residents were seeing complications among the patients of all neurosurgeons in the large nine-hundred-bed teaching hospital. They were in a much better position to crack the case.
Determined to figure out whether rhBMP-2 was at the heart of the outbreak, she enlisted several colleagues to assist with a study. They compared the outcomes of all patients at Henry Ford Hospital who underwent cervical spine surgery with rhBMP-2 from 2004 to 2009 (260 patients) to those who underwent cervical spine surgery without rhBMP-2 (515 patients). The results were damning. Patients implanted with rhBMP-2 were more likely to develop problems, ranging from difficulty swallowing to respiratory distress and death. The numbers from her study suggested that 25 more patients per one thousand operations would die if rhBMP-2 was used during surgery.283
Other doctors began to see problems as well. One report in the MAUDE database stated that a patient developed “airway issues” and died two days after surgery.284 Another report described a patient who experienced “difficulty swallowing, extreme pain, more pain after surgery, difficulty speaking, nerve injury, radiating pain to the legs…[and] bony overgrowth” and had to walk “hunched over.” Yet another stated that a patient developed “acute respiratory failure” and died eight days after surgery. There were other reports. Many others. Invariably the maker of rhBMP-2 declared that it was only reporting the deaths “for notification purposes” and that it was “unable to determine the definitive cause of the reported event,” because “neither the device nor films of applicable imaging studies were returned to the manufacturer.”
After Yaremchuk published her study, three doctors sent a letter to the editor of the journal vigorously disputing her findings, saying that the data “completely contradict our experience.”285 Citing a study they published in 2010, the authors asserted that there were no deaths among the “more than 150 patients” they had followed for two years and that most problems were relatively minor and “self-limiting.” Two of the three authors acknowledged that they received payments from Medtronic, Stryker, and other device manufacturers.
Nor was Medtronic about to take Yaremchuk’s findings lying down. It, too, insisted that rhBMP-2 was safe and sent a letter to her demanding copies of the medical records of all the patients who were included in the study.
The bullying tactic didn’t go over well with the hospital, and it refused to hand over the requested medical records. Hospital lawyers informed Medtronic that the company had no standing to request the personal records of patients.
The hospital’s refusal to bow to industry demands may have been thanks in part to Yaremchuk herself. In January of 2007, as vice president of clinical practice performance, she helped to implement a systemwide “influence-free” policy.286 The policy prohibits doctors from accepting gifts from drug or device manufacturers, including free drug samples, which are often used to influence doctors to prescribe expensive brand-name drugs. Doctors were banned from accepting even “pens, coffee mugs, or sticky notes from industry” (items that often come with the names of products emblazoned on them). Yaremchuk says that the policy was necessary because even though many doctors claim they aren’t influenced by gifts, “every study shows that there is a subtle, but real, influence.”
The Henry Ford Health System estimated in 2009 that it was saving $10 million per year in lower drug and device costs thanks to the policy prohibiting doctors from accepting gifts from industry.
When Yaremchuk helped to implement the influence-free policy, the Physician Payments Sunshine Act, which requires doctors to report money they receive from drug and device companies,* was still three years off in the future, and a wave of reforms aimed at transparency and limiting conflicts of interest was yet to follow. But her actions and those of others would lead to shocking revelations about Medtronic, the backstory about Infuse, the FDA, and extraordinary sums of money.
One of those revelations was that recombinant bone morphogenetic protein, rhBMP-2, had never been approved by the FDA for use in the neck—the most dangerous part of the spine for implantation. But that is a story best told by a Hollywood screenwriter, Jerome Lew.
* * *
To understand how surgeons came to implant rhBMP-2 in the cervical spine, it’s necessary to follow a complex scheme carried out by industry. It’s a story Jerome Lew would learn in excruciating detail after he had rhBMP-2 implanted in his neck. Lew has worked for directors of films such as All the President’s Men, Scarface, and Top Gun, and he says what happened to him might be the subject of a horror film.
In February of 2009, Lew was bringing his newborn son home from the hospital—a day Lew would say “should have been joyous.” But as he was taking his son out of the backseat,
Lew was struck by a car. His son was unhurt, but Lew was injured. He was ultimately referred to Jeffrey C. Wang, then the chief of spine surgery at UCLA’s Spine Center. After reviewing a CAT scan of Lew’s cervical spine, Wang reportedly told him that he needed surgery or he could end up paralyzed.
A few months later, in May of 2009, still in profound pain, Lew underwent surgery, believing that he was about to have a routine neck fusion in which bits of bone taken from his hip would be used to fuse together vertebrae in his neck. By all accounts, the surgery appeared to be a success, and Lew says he had “dramatic” pain relief following the operation. But months later, the symptoms returned and worsened. He lost strength in his arms. He was clumsy when he tried to pick things up. The pain was unrelenting. Lew explained his symptoms to Wang, who suggested that he see a psychiatrist and a pain management specialist.
By 2012, things had gotten much worse. He was having trouble walking. His voice grew hoarse. Talking and swallowing foods became difficult. Eventually he began to lose bladder control, and at times he would urinate on himself.
Lew decided not to return to Wang. Instead he sought help from Charles Rosen, a spine surgeon at the University of California, Irvine. When Rosen reviewed images of Lew’s neck and his medical records, he found that Wang had implanted rhBMP-2 in Lew’s neck and that it had triggered one of its dreaded complications—bony overgrowth. Bone was now impinging on nerves in Lew’s neck.
Rosen undertook risky revision surgery on Lew, and although he was unable to remove the devices in Lew’s neck, he was able to relieve some of the pressure on the nerves on one side of his spine, which gave Lew significant relief from his symptoms.
Recombinant bone morphogenetic protein was originally approved by the FDA in 2002 as part of a “combination product”—a biologic product and a mechanical device—known as Infuse. Manufactured by Medtronic, Infuse consists of special sponges that are impregnated with rhBMP-2 at the time of surgery and placed in a metal cage sized and shaped for the large bones of the lower spine. The FDA approved Infuse for use at only one disk level in the lower spine—not the neck.
The Danger Within Us Page 20