Vigilance in Joint Implant Recalls and Reporting Is Critical to Enhancing Patient Safety

A male patient aged 76 years presented to a clinic, complaining of bilateral knee pain (Fig. 1). He underwent knee replacements in 2011. The patient had not had an evaluation in years and was unaware that the implanted devices had been recalled for rapid polyethylene wear. The surgeon informed the patient, who was confused and frustrated that he needed revision surgery. How does a medical device approved by the FDA fail? How can surgeons better protect patients? It is important for physicians to familiarize themselves with recalls, the recall process, and how they can monitor medical devices.

Critical advancements in arthroplasty unquestionably offer patients a higher quality of life. Each year, the FDA authorizes about 100 new or significantly modified hip and knee arthroplasty devices. The potential patient benefits of this rapid development are immense, but surgeons should be mindful of the potential harm from untested technology. Approximately 98 percent of all orthopaedic devices enter the market without supporting clinical data through the FDA’s 510(k) pathway, and the 10-year risk of recall for these arthroplasty devices is 25 percent, more than double that of general orthopaedic devices.

In recent years, there have been several high-profile recalls of joint implants. Such recalls decrease public trust in device manufacturers, the FDA, and orthopaedic surgeons. Only 2 percent of orthopaedic device recalls are considered “high risk” (Class I) by the FDA, with the remainder essentially all being classified as “moderate risk” (Class II). Table 1 shows all Class I recalls of arthroplasty devices since 2002. To learn more about recent recalls, AAOS members can log in to the AAOS Device Recall Dashboard, which was created as a joint effort by the AAOS Committee on Devices, Biologics, and Technology and the AAOS Healthcare Safety Committee.

Several of the largest litigation settlements originated from Class II recalls. Stryker settled lawsuits over Rejuvenate and ABG II hip implants for more than $2 billion dollars, whereas Depuy settled litigation against the ASR cups for more than $4 billion—all Class II recalls.

With rare exception, recalls are issued voluntarily by the manufacturer. This process leads to variability and controversy. For example, despite widespread reports of increased risk of failure, the DePuy Attune tibial base plate was never recalled. The base plate’s design was modified in 2016, but no more information was publicly available. Occasionally, the FDA can unilaterally issue a Safety Communication warning to physicians, although this is not a formal recall. In January, the FDA published a Safety Communication titled “Do Not Use Synovo Total Hip Resurfacing System” because unauthorized modifications were identified by FDA inspections.

A recalling firm must notify its affected “direct accounts,” which are typically hospitals. Hospitals should follow best practices, including appointing a recall coordinator to inform physicians and ensure that violative products are returned. Still, most surgeons first learn of a recall from industry representatives, news, and social media. Patients should then be informed by the treating surgeon. Although both the FDA and device manufacturers frequently post patient-centric recall information online, the process is haphazard at best, and the risk to the individual patient is unclear. In the case example presented earlier in this article, the patient was completely unaware that Exactech recalled its polyethylene insert, and potentially his surgeon was also.

The extent of the clinical impact of a recall is unclear. For example, the FDA website lists the reason for the Exactech recall: “One of the packaging layers for the plastic insert has been out-of-specification and may allow oxygen from the air to diffuse into the plastic insert.” This recall affected more than 58,000 devices manufactured since 2004, but without clinical data, the extent of harm to patients cannot be certain. This example highlights another concern with rapid adoption of unproven technology. The median time to first recall is 2 years; however, devices can be recalled multiple times, and recalls can occur even decades after being on the market. A delayed recall was seen with the Profemur hips, where reports of stem fractures in the FDA’s Manufacturers and User-Facility Device Experience (MAUDE) database preceded the recall by 15 years. Voluntary recalls are not clearly defined, and surgeons are left to monitor implants without clear indications for revision surgery.

Many recalls occur for smaller infractions, such as implant inserter tools breaking during a procedure or a device being mislabeled, though 25 percent of orthopaedic recalls are classified by the FDA as “device design” issues. The authors argue that only surgeons can best provide context here. Surgeons can report failures to the FDA’s website (MedWatch.gov), which then get processed into the MAUDE database. Physicians are not mandatory reporters and infrequently contribute, but they can improve by filling out the simple form when they witness a device failure. The importance of these reports cannot be overstated. Another important avenue for reporting is publishing device failures, which helps when there are discrepancies in MAUDE.

Lastly, participating in the American Joint Replacement Registry (AJRR) can provide much-needed long-term failure data and ensure the safety and efficacy of arthroplasty devices. AJRR captures nearly 4 million arthroplasty procedures, providing invaluable real-world, year-by-year evidence regarding implants across many different practice settings, geographies, and demographics. Investigators can submit research proposals that target specific clinical questions, and an annual AJRR report is published, which provides useful information on trends with different implants. One interesting and example is a trend seen in the 2023 report that demonstrated the rapid uptake of novel arthroplasty devices. The Actis DuoFix entered the market in 2016 and became the most commonly implanted stem by the end of 2023, accounting for nearly a quarter of all reported hip stems in 2023. By contrast, the Corail stem, which has been available in the United States since 1980, saw a 75 percent decrease in its relative reported usage from its peak 2013. Corail has the second most entries in the database (43,364) and reports the lowest unadjusted cumulative 10-year revision rate at 1.66 percent. Time will tell whether these practice shifts are safe and cost-effective.

Supporting innovation is essential to continued improvement for patients. However, orthopaedic surgeons should understand that a vast majority of devices they use enter the market without clinical testing, and malfunctions can and do occur, sometimes necessitating recall. Recalls have a profound impact on patients and communities. As such, physicians have a professional obligation to participate in the real-world monitoring of the devices we implant, such as participating in AJRR and reporting in MedWatch. Physicians also have the ethical responsibility to divulge to patients when a recall of a device they implanted affects them, as well as appropriate clinical management going forward.

This article was submitted on behalf of the Committee on Devices, Biologics, and Technology and the Committee on Healthcare Safety.

Jonathan Dubin, MD, FAAOS, is an associate professor of orthopaedics at the University of Missouri–Kansas City Medical School and serves as chief of orthopaedic trauma at University Health–Truman Medical Center. He is a member of the AAOS Committee on Healthcare Safety.

Cemil C. Gedik, MD, is a fifth-year orthopaedic surgery resident at Koç University School of Medicine in Istanbul, Turkey.

John W. Krumme, MD, is a private-practice joint reconstruction surgeon with Kansas City Orthopaedic Alliance and an adjunct clinical assistant professor at the University of Missouri–Kansas City.