Type “metal-on-metal hip replacement” into your favorite search engine, and you will bring up dozens of hits from personal injury lawyers. These devices, initially touted as being more durable than conventional metal and plastic or metal and ceramic versions, have proved to be considerably less reliable. A substantial proportion produce problems within a few years after insertion: either the device fails, causing pain and disability, and needs to be replaced (necessitating another operation) or it sheds tiny metal flakes into the circulation, triggering both local inflammation and systemic illness. The British Orthopedic Association reported in 2011 that metal-on-metal (MoM) devices had a shocking 21% failure rate within 4 years and a 49% failure rate at 6 years. Several products have been taken off the market. The cost of replacing the joints is expected to reach billions of dollars in the coming years, and that doesn’t include the human suffering or lost productivity. What happened?
In discussing the perils of metal-on-metal hips, it’s critical to recognize that hip replacement surgery has been an enormous boon to aging Americans. Osteoarthritis, or wear and tear arthritis, is found in 50% of people over age 65, making it one of the most common chronic diseases of older people. It can cause pain and difficulty walking, debilitating symptoms that are often completely alleviated by replacing the worn out joint. And while hip replacement surgery has been around for decades, it has been plagued by the risk of hip dislocation or the artificial joint itself wearing out. Hence the holy grail of hip replacement surgery has been a hip made of better materials and with a better design. Metal-on-metal hip prostheses were once thought to be the answer to the orthopedist’s prayers.
Something else stimulated development of metal-on-metal hip joints as well as the desire for an improved device. With the benefits of surgery well established, the number of hip replacement operations remained fairly stable in the 1990s (adjusting for the aging of the population). Device manufacturers realized that if they could make a joint that lasted longer than the 15 years expected with conventional devices, they could increase sales by marketing to a younger population. Metal-on-metal hip prostheses, in which both the ball (femoral head) and the socket (acetabulum) are removed and replaced with metal parts, seemed to fit the bill; mechanical stimulation studies in the laboratory suggested they would last longer than existing devices.
The new MoM hips were rapidly introduced into practice and the total number of operations performed each year soared. Estimates from Kaiser Permanente, a large multi-specialty group practice that decided to keep track of all joint replacements in their patients, are that 35-40% of all joint replacements are now done in people under 65. Then anecdotal reports of early device failure began appearing. Device manufacturers initially denied any problems associated with MoM hips but in 2007, the Australian National Joint Replacement Registry reported a higher rate of hip revision surgery with MoM devices. Finally, in August, 2010, DePuy Orthopedics, a subsidiary of the giant Johnson & Johnson, recalled its MoM total hip implant system. A total of 90,000 people had had the device inserted, just about half of them in the US.
What is striking in this saga is that no randomized controlled trials comparing MoM hip joints to existing models were required for Food and Drug Administration (FDA) approval. Since the FDA considered these devices only “moderate risk” and since they regarded them as “substantially equivalent” to existing devices, only the most cursory approval process was necessary to bring them to market. High risk or radically new devices, by contrast, undergo rigorous pre-market testing, often involving sophisticated—and expensive—randomized clinical trials. To demonstrate “substantial equivalence,” all that’s required is to show that the new device has the same intended use as an existing device and the same technological characteristics. The FDA uses this expedited “510(k)” review process for an estimated 90% of the devices it approves. (A related surgical approach, but which did require a full premarket review, involves hip “resurfacing” rather than total hip replacement. This technique has also met with difficulties when the head of the femur is capped with metal rather than plastic, but it is not the subject of this article.)
In light of several well-publicized recalls of devices that entered clinical use via the 510(k) route, the FDA requested a review of the expedited review process. The review was conducted by the Institute of Medicine (IOM), a widely respected, not-for-profit, independent organization dedicated to providing advice to government policymakers. The IOM report was released last summer. Its conclusions were clear: “The committee finds that the current 510(k) process is flawed based on its legislative foundation. Rather than continuing to modify the 35-year-old 510(k) process, the committee concludes that the FDA’s finite resources would be better invested in developing an integrated premarket and postmarket regulatory framework that provides a reasonable assurance of safety and effectiveness throughout the device life.” Would this kind of approach, transparent and based on sound science, have avoided the MoM fiasco? \
If MoM artificial hips had been compared to existing products in a randomized controlled trial, flaws would likely have emerged early. While most of the device failures take five years to develop, much longer than the duration of a typical clinical trial, data from the British National Joint Registry suggest there are signs of trouble much earlier. For women aged 60-69, the Registry the revision rate for cemented hips one year after initial insertion is .65%; the revision rate for metal on metal hips in this population is 1.42%, or more than double. By year three, the gap is even more impressive: 1.32% of cemented hips require revision compared to 4.11% of MoM joints, a near tripling of the rate.
Adequate pre-market testing would have raised red flags; systematic post-market surveillance, which is non-existent in the US, would have revealed any problems that had escaped early detection. The Australians, who maintain a Registry of all the artificial joints inserted in their population, observed a spike in surgical revisions after hip replacement in 2007, with the highest revision rates in MoM joints, and the British reported a similar finding based on their Registry soon afterwards. A US database, which would have been far larger, might have detected problems even more quickly.
Any rational observer would conclude that our story will have a happy ending: the recommendations of the IOM committee will be implemented by Congress, we will have a new improved regulatory process for medical devices. Tales such as the metal-on- metal hip fiasco will provide fodder for future historians who will write of the bad old days when corporate greed and medical wishful thinking led to needless suffering in hundreds of thousands of patients, at a cost of billions. But the reality is strikingly different. The IOM report was dead on arrival. In a shocking development, the report was viciously attacked even before it was released. As the NY Times put it in a July 27, 2011 article: “Allies of the medical device industry are waging an extraordinary campaign in Washington to discredit a coming report by one of the country’s pre-eminent scientific groups.” The article continued that business groups took “the highly unusual step of making a pre-emptive strike,” claiming that better regulation would “slow innovation, harm patients and cost jobs.” Industry does not want to modify the 510(k) process, substituting more thorough, time-consuming, and costly studies of potentially risky devices. Their lobbyists responded to the IOM report by barraging the media, Congress, and other thought-leaders with a blistering counter-attack. The result? Deafening silence in the halls of Congress, which would need to pass legislation to further amend the Food and Drug Administration Act.
The metal on metal hip story is not unique. A recent article in the Archives of Internal Medicine reviewing all FDA device recalls between 2005 and 2009 found that 71% of the devices recalled because they were felt to cause serious health problems or death had been approved by the 510(k) expedited review process. Patients who have been harmed by dangerous devices are busily trying to salvage their own health and are relying on the courts for monetary compensation for their injuries. But it is consumers who put pressure on government back in 1976 to demand regulatory protection from dangerous devices. It was consumers, reacting to dramatic examples such as deaths related to the Dalkon Shield, an intrauterine contraceptive device, who demanded and finally prevailed upon Congress to pass the Medical Device Amendments of 1976, for the first time giving the FDA jurisdiction over devices. How many more disasters will we need before we act again?