Rare Harms Linked to Silicone Implants in New Study

Diana Phillips

October 01, 2018

Women with silicone breast implants are up to eight times more likely to be diagnosed with one of several chronic autoimmune diseases than women in the general population, and they are at significantly greater risk for other adverse outcomes, including melanoma and stillbirth, a study has shown.

The findings, published online September 14 in the Annals of Surgery, come from a summary analysis of outcomes of nearly 100,000 patients enrolled in large postapproval studies (LPAS) that the US Food and Drug Administration (FDA) required of device manufacturers as a condition of implant approval in 2006.

Although the study is based on data from the FDA LPAS database, the agency disagrees with the authors' conclusions. In a press statement and in an editorial published in the same issue of the journal, Binita Ashar, MD, director of the Division of Surgical Devices in the FDA's Center for Devices and Radiological Health, cites "significant shortcomings" with the study's methodology and data presentation and urges that the conclusions be viewed with caution.

For the analysis, Christopher J. Coroneos, MD, from the Department of Plastic Surgery at MD Anderson Cancer Center in Houston, Texas, and colleagues reviewed the outcomes of 99,993 patients enrolled between 2007 and 2010 in postapproval studies conducted by Allergan and Mentor.

The studies included women 22 years of age or older who received either unilateral or bilateral silicone- or saline-filled implants for primary or revision breast augmentation, and women 18 years of age or older who received implants for primary or revision breast reconstruction after cancer resection, trauma, or congenital absence. Of the full cohort, 83,317 patients had silicone implants and the remainder had saline implants.

The primary outcomes of interest were the incidence of cancer, connective tissue disease, autoimmune disease, self-harm, and reproductive outcomes calculated per 10,000 person-years.

The analysis showed that women with silicone breast implants were eight times more likely than those in the general population to be diagnosed with Sjögren syndrome (standardized incidence ratio [SIR], 8.14; 95% confidence interval [CI], 6.24 - 10.44; P < .001), seven times more likely to have a scleroderma diagnosis (SIR, 7.00; 95% CI, 5.12 - 9.34; P <.001), and nearly six times more likely to be diagnosed with rheumatoid arthritis (SIR, 5.96; 95% CI, 5.35 - 6.62, P < .001).

In addition, silicone implants were associated with a nearly 4-fold increase of melanoma (SIR, 3.71; 95% CI, 2.87 - 4.73; P < .001) and a 4.5-fold increased risk for stillbirth (SIR, 4.50; 95% CI, 3.59 - 5.56; P < .001).

No association was observed between silicone implants and suicide, as has been suggested by previous studies. One case of anaplastic large cell lymphoma, which has been linked in earlier studies to breast implants, was reported.

With respect to surgical complications, silicone implants were linked to an increased risk for capsular contracture compared with saline implants (5.0% vs 2.8%, respectively), as well as higher rates of short-term reoperation (6.5% vs 3.4% at 2 years, respectively).

Coroneos and colleagues acknowledge that, despite the significantly increased risk for certain harm events associated with the silicone implants, the absolute rates of these adverse outcomes were low.

They also note that their interpretation of the results is limited by study design considerations. For example, the incidence rates in the manufacturers' trials were based on different data sources. Specifically, the Allergan studies relied on physicians' diagnoses during the first 2 years of patient follow-up, whereas the Mentor studies relied on self-reported patient data at 7 years. Also, more than half the women dropped out by the second year, so a large percentage of patients were lost to follow-up. Both of these considerations could influence harm risk rates, they write.

In addition to some of the methodological deficits mentioned, another important limitation is the inability to analyze patients on an individual level to generate an unbiased analysis of the harms. "It is accepted that women with breast implants have a number of distinct baseline factors and comorbidities. This is especially important in adjusting analyses for known covariates or potential confounders among rare harms," the authors write. Furthermore, they stress, there is no mechanism for isolating subgroups of implant characteristics, operative details, and specific clinical combinations from the summary data, preventing adjustment for these characteristics.

Despite the existing weaknesses in the evidence that keep the findings from being conclusive, "[t]his study demonstrates silicone implants are associated with an increased risk of certain systemic harms," the authors state.

"We are not suggesting a causal relationship," senior author Mark Clemens, MD, professor of plastic surgery at MD Anderson, told Medscape Medical News. "These are associations. But they do provide important safety information for women who are thinking about cosmetic or reconstructive surgery with breast implants, and for their surgeons."

The FDA, in contrast, stands by its determination that there is not sufficient evidence to show an association between breast implants and rheumatologic or connective tissue diseases, based on its own review of the postapproval evidence in 2011, Ashar writes in the editorial.

That said, patients who are considering breast implants should be aware that they are not lifetime devices, she stresses. "[T]he longer a patient has the implants, the more likely they are to experience local complications and adverse outcomes requiring breast implant removal."

The FDA and the study authors do agree on one point: the need for more research and better reporting.

"To resolve the remaining uncertainty in the evidence base, it is important that LPAS data be analyzed on a patient-level, and in an unbiased manner. It remains the plastic surgery community's duty to provide definitive evidence for the risks associated with breast implants," the authors write.

The FDA believes the study limitations point to the need for better postmarket evidence generation, including active surveillance capabilities. "This is why FDA has helped develop breast implant registries and is working with other partners to establish the National Evaluation System for health Technology," writes Ashar.

Such efforts are likely not enough, according to James D. Frame, MD, professor of aesthetic plastic surgery at Anglia Ruskin University in Chelmsford, United Kingdom. "Having a breast implant registry will not give the data we need. There was such a registry in the UK until the early 1990s, but no data was ever analyzed. Certainly no analyses were shared with the surgeons," he said in an interview with Medscape Medical News. "The UK's [Medicines and Healthcare products Regulatory Agency] is similarly reluctant to hand over any data that we collectively provide on the grounds of 'patient confidentiality.' This makes it impossible for independent researchers to analyze data other than their own."

The reality, according to Frame, is that autoimmune problems associated with silicone implants "have been investigated about every decade since the 1970's. Both the FDA and our [Medicines and Healthcare products Regulatory Agency] have been somewhat complacent in collecting meaningful and accurate data," he said. "I personally think they have been very pleased when papers appear showing no clear association. Remember, though, that these papers don't demonstrate that silicones actually don't contribute to causation of autoimmune disease, but no meaningful analysis can be made from the poor data collected."

What is needed, Frame explained, "is standardized and comprehensive mandatory data on all patients having implants." And these data, he argued, should not be owned by the regulatory agencies. "The problem is that the wrong people are making these decisions."

The study authors, Ashar, and Frame have disclosed no relevant financial relationships.

Ann Surg. Published online September 14, 2018. Article abstract, Editorial extract

For more news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.