The Fallacy of a Bacterial Etiology for Capsular Contracture and BIA-ALCL and Assigning Blame to Noncertified Surgeons

Recent commentaries consistently implicate bacteria in many complications of breast augmentation, including capsular contracture, breast implant–associated anaplastic large-cell lymphoma (BIA-ALCL), and systemic symptoms associated with breast implants,1–4 also known as breast implant illness. By contrast, numerous publications challenging these views are conspicuously absent from these commentaries and references.5 Scientific integrity demands that opposing views be accepted or rejected, but not simply ignored.5 Plastic surgeons deserve to be informed of both sides of the debate. Examples of opposing evidence, all published within the last 7 years, are summarized in Table 1. TABLE 1 - Claims Relating Breast Implant Complications to Bacteria Contrasted With Evidence No. Claim Evidence 1 Bacteria cause capsular contracture1 Capsular contractures occur in pockets inoculated with S. epidermidis, but also in uninoculated pockets6Sophisticated microbiological studies fail to find a correlation between biofilm and capsular contracture7Capsular contracture rates similar for (bacteria-laden)8 textured devices 2 Gram negative bacteria cause BIA-ALCL9 Findings have not been reproduced10Bacterial etiology has been abandoned10Study using 16S rRNA microbiome sequencing foundno significant difference versus controls11 3 BIA-ALCL capsules have higher bacterial counts9 Similar bacterial counts for BIA-ALCL capsules and controls9Similar microbiomes comparing BIA-ALCL to controls11 4 Bacteria are implicated in Breast Implant Illness1 Presence of bacteria does not affect symptoms12Symptom improvement with no capsulectomy13 5 Clusters of BIA-ALCL indicate infectious etiology2 Clusters reflect meticulous follow-up14 6 14 points reduce the risk of BIA-ALCL15 None of the 14 points affects risk16 7 Research is “free from any commercial strings”17 Major manufacturer funding1–4,15 8 Focusing on texture is wrong18,19 Texture is causative for BIA-ALCL20 9 BIA-ALCL is not a malignancy15 BIA-ALCL is a malignancy10,20 10 The patient, surgeon, and device are implicated2 Only the device is implicated14 11 Textured implants are superior to smooth17 Biocell textured implants are inferior21 12 Introduction sleeve reduces risk15,19 Introduction sleeve does not affect risk22 13 Antibiotic solutions reduce risk15,23 Antibiotic solution does not reduce risk24Systematic reviews find no benefit25–27 14 Betadine pocket irrigation is recommended23 Betadine irrigation is nonsterile and unauthorized24 15 Noncertified operators increase BIA-ALCL risk2,28 Certification does not affect BIA-ALCL risk14 16 Noncertified operators increase capsular contracture rates2 No significant difference in capsular contracture ratescomparing certified and noncertified operators29 17 Surgeons with adverse events require investigation1,2,29 No known technical deficiency14 Not only are articles with opposing evidence not referenced, but also a highly cited publication,15 supporting a bacterial etiology for capsular contracture and BIA-ALCL, remains unreferenced.1–3 This article was used to defend the continued availability of macrotextured implants at the 2019 US Food and Drug Administration hearing.16 It was not really a prospective cohort study, but rather a retrospective 8-question survey, emailed by SurveyMonkey to the authors, although this fact was not disclosed.16 A dishonest reporting of the findings undermined the study conclusions.16 The authors have not issued a correction. Consequently, other investigators may be unaware of these problems. Researchers in Egypt recently published a meta-analysis, including 42,000 macrotextured implants with no BIA-ALCL, which they believed to represent a prospective cohort with an (implausible) 11.7-year mean follow-up, leading to their erroneous conclusion that the BIA-ALCL risk is similar for smooth and textured devices.30 A false narrative promoted by industry surrogates continues to find fault with the surgeon as opposed to the device.14 Before recommending treatments that target bacteria, the identity of the responsible microbe is needed.31 Using specialized cultures, polymerase chain reactions, and scanning electron microscopy, Poppler et al7 found no correlation between biofilm formation and capsular contracture. Only 1 of 26 patients even demonstrated evidence of a microbial biofilm. The authors cautioned that there was strong evidence from other studies that antimicrobial therapy induces biofilm formation to confer resistance.7 In other words, antibiotic irrigation may be counterproductive. Another recent study using next-generation sequencing, an evolution of polymerase chain reaction technology, reported microbial DNA in only 42% of capsular contracture specimens and 120 unique bacterial species.32Staphylococcus epidermidis was isolated in only 11% of the positive specimens. No control group was studied. Therefore, it is unknown whether these findings were affected by contamination during specimen handling.33 In fact, no specific microbiome or organism has been linked to capsular contracture.31 Bacterial profiles appear to be patient-specific rather than disease-specific.11 Koch's postulates34 have not been satisfied for an infectious etiology, in that the condition often occurs absent the suspected pathogen.31 From a microbiological perspective, it is unclear how any additional (ie, 14-point) measures, beyond ordinary sterile operating technique, would be beneficial.35 The implant comes out of the box sterile, is handled by sterile gloves, and is inserted into a nonsterile pocket that is colonized by harmless commensal bacteria.31 A recent in vitro study is offered4 as evidence of the efficacy of antibacterial strategies.1 This laboratory study involved passing a breast implant through a polyvinyl sheet that had been contaminated with S. epidermidis. The authors based the decision to use this bacterium on a 2001 report of device-related (eg, indwelling catheters) infections that did not evaluate breast implants or cultures from capsules affected by capsular contracture.36 In view of recent, sophisticated microbiological studies that have failed to identify a responsible organism,7,32 the clinical relevance of this in vitro study is questionable. Previously, Ralstonia pickettii, a gram-negative bacterium, was implicated in the etiology of BIA-ALCL.9 This finding has not been reproduced; this microbe is likely a contaminant.10 Investigators no longer consider infection a likely etiology for BIA-ALCL.10 In their detailed investigation, McGuire et al12 recently found that the presence of bacteria does not affect symptoms of breast implant illness. These investigators also documented symptom improvement in women treated with implant removal and no capsulectomy.13 In summary, the most recent evidence does not support a bacterial etiology for capsular contracture,7 BIA-ALCL,10 or breast implant illness.12,13 In his recent commentary,2 the discussant wrote, “It is important that the role of the practitioner, his/her clinical skill and training be studied in all reported adverse events related to breast implants.” In another publication, he recommends that “any adverse event occurring within 5 years of initial surgery should be flagged as a mandatory reportable clinical indicator and trigger further investigation.”29 These draconian recommendations are remarkable in that a technical deficiency is not known to be related to these adverse events.14 No doubt such recommendations would stifle the reporting of any complications and impair any further research. Every surgeon has adverse events. Surgeons know that the only way to eliminate complications is by not operating. Consequently, this recommendation would subject every surgeon to investigation. Investigating all surgeons would be a tremendous burden, and who is to be entrusted with such authority? Do the criteria include obeisance to the 14 points? How are nonconforming surgeons (including this author) to be disciplined? The discussant claims that early capsular contracture rates are higher when surgery is performed by noncertified operators,2 referencing a study that actually found no significant difference in the risk of capsular contracture comparing certified and noncertified operators.29 In 2020, Loch-Wilkinson et al28 reported a higher frequency of BIA-ALCL among patients treated by surgeons who are not certified by the Royal Australasian College of Surgeons. Loch-Wilkinson et al28 examined 3 small groups of patients who received Mentor, Nagor, and Silimed implants. Just over half of the patients had Allergan implants (an AbbVie company, North Chicago, IL), but the sales data for these patients did not provide a breakdown of certified versus noncertified surgeons, so this information was not included. The authors reported only percentages of BIA-ALCL patients who were treated by noncertified surgeons; the actual numbers were not provided. The P values were extraordinarily small (0.002, 0.0004, and 0.007, respectively), suggesting that the certification status of the surgeon is highly relevant to risk. The statistical test used in making this comparison was not specified. The meaning of the wide confidence intervals is unclear. A nonparametric test of categorical variables, comparing expected with observed frequencies, is needed. When the group sizes are small, Fisher's exact test is appropriate.37 This claim regarding a link between certification status and BIA-ALCL merits careful consideration for several important reasons. A link would support an unspecified technical deficiency as a cause of BIA-ALCL. Of course, this finding impugns noncertified operators, who must defend accusations of substandard practice, and provides fuel for plaintiffs' attorneys who may be inclined to sue noncertified defendants. This information may be used in the media to influence patients to choose surgeons who are certified by the Royal Australasian College of Surgeons. An accusation of substandard practice may be used to restrict noncertified practitioners from performing breast implant surgery.38 Board-certified plastic surgeons (full disclosure: this author is certified by the American Board of Plastic Surgery and the Royal College of Surgeons of Canada) may not be sympathetic to operators who they believe are not properly trained in cosmetic surgery and may not be able to offer the full range of cosmetic breast surgery. Regardless of the merits of these beliefs, certified plastic surgeons must not engage in anticompetitive behavior. To wrongly suggest that their noncertified competitors are practicing in a substerile manner and are even putting their patients at risk of cancer is unethical. Turner, a researcher who provided commentary on this study,39 cautions that patients may wrongly believe that they may protect themselves against BIA-ALCL by selecting a certified surgeon. She emphasizes that the definitive risk factor is a history of a textured implant. Cordeiro, a well-known board-certified plastic surgeon practicing in New York City, has reported 10 BIA-ALCL cases among breast reconstruction patients who received Allergan Biocell implants and were followed up closely after surgery.40 Not surprisingly, the Australasian College of Cosmetic Surgery and Medicine, representing Australian noncertified surgeons, reacted strongly to this accusation.38 This group filed a detailed research misconduct complaint to 5 institutions, including Macquarie University. These surgeons noted that the study authors reported that 8 years of Silimed sales data were evaluated when, in fact, only 9 months of sales data were made available by this manufacturer. They also noted that 7 cases of BIA-ALCL in augmentation/mastopexy patients, all performed by certified plastic surgeons, were not counted in the analysis. The authors of the complaint were frustrated by the nondisclosure of the actual numbers, which makes independent evaluation difficult. They requested access to the numbers. They knew that the numbers must be very small and questioned such low reported P values. In 2022, Loch-Wilkinson et al41 published a correction to their article, acknowledging that the Silimed sales data were based on 1 year of data (actually 9 months),38 not 8 years as originally reported. The authors considered this error an oversight, and the correction was “an opportunity to clarify this data,” although they did not reveal the actual numbers as requested. The authors defended their findings as preliminary and worthy of further study. Fortunately, it is still possible to decipher the truth. Although the actual numbers were not reported, they may be determined from the percentages reported in the authors' Table 7,28 for example, 11/12 = 91.7%, the proportion for Silimed implants (Rio de Janeiro, Brazil). The numbers for Mentor (4/5 = 80.0%) (Irvine, CA) and Nagor (5/6 = 83.3%) (Glasgow, UK) are 5 and 6, for a total of 23. These numbers are consistent with a previous publication by these authors (14, 5, and 6, respectively, totaling 24).42 When these numbers are subjected to a Fisher's exact test,37 the results are much different from the published values (Table 2). None of the P values is significant at a level <0.05. This finding indicates that the authors were mistaken in assigning blame for this complication on noncertified surgeons. Indeed, it would have been surprising if there had been a link, in view of the absence of a known technical consideration that affects BIA-ALCL risk,14 the current consensus that the etiology is not related to infection,10 and of course the large number of BIA-ALCL cases reported by a board-certified plastic surgeon.40 TABLE 2 - Comparison of Observed and Expected BIA-ALCL Cases for Noncertified Surgeons Manufacturer Proportion Noncertified (%) Number Noncertified Number Certified Sale Proportion (%) Expected Number P* Mentor 80.0 4 1 15 1 0.21 Nagor 83.3 5 1 15 1 0.08 Silimed 91.7 11 1 52 6 0.07 *Fisher's exact test. With this new information, consideration must be given to either a retraction of the erroneous article28 or, at a minimum, another correction. This one should come with an apology for the hardship this misinformation has caused their colleagues.38 After all, the authors should have known that their numbers were too small to reliably establish a link. They should have reviewed their statistical analysis for accuracy, especially in view of its importance. A correction is also needed for the erroneous statement linking higher capsular contracture rates to noncertified surgeons.2 It is ironic that Dr Deva, in his most recent correspondence, should write, the “14-point plan was never intended to be used as a clinical standard and certainly should not be used as a ‘weapon’ to accuse colleagues of substandard surgery.”1 The only plastic surgeons using BIA-ALCL as a weapon are the authors of these allegations.14 As for a clinical standard, plastic surgeons are still asked to take a pledge to follow the 14 points.43 Abandoning textured devices—the only risk mitigation method known to be effective—is not one of these points. Evidence pointing to bacterial infection as a cause of BIA-ALCL (not textured implants) is cited as “all the more reason for surgeons to follow the 14-point plan.”43 Betadine pocket irrigation is recommended,43 despite the fact that this antiseptic is nonsterile, cytotoxic, and unauthorized for use in open wounds.24 Bacterial mitigation principles are now incorporated into a “toolkit,”3 which has been adopted by the New South Wales, Australia government to provide “best practice guidance,” including of course antimicrobial irrigation of implant pockets.44 Woe to the plastic surgeon who runs afoul of these outdated recommendations. In putting their patient's best interests first, plastic surgeons must resist dogma and defend their practices using evidence-based medicine.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.