Chronic Biofilm Infection in Breast Implants Is Associated with an Increased T-Cell Lymphocytic Infiltrate

Sir: Given the recent reports concerning the findings of anaplastic large cell lymphoma (ALCL) of the breast that have been noted to occur in association with breast implants,1 the findings of this article looking at the association of chronic biofilm with a T-cell lymphocytic response2 certainly merit careful study. To summarize, the authors have used an established biofilm model in the pig to measure not only the bacterial load but also the lymphatic cell response to this infectious challenge in the surrounding capsule and the surface of the implant. Also, capsular specimens from patients afflicted with Baker grade IV capsular contracture were analyzed for the same variables. To summarize, in the pig, a larger bacterial load was noted to be present on the surface of textured versus smooth implants, and this was associated with a greater number of lymphocytes, most of which were T cells. When these T cells were examined using scanning electron microscopy, they had the appearance of being activated. In addition, when the human tissue was examined, an inflammatory response was identified, and it was predominantly a T-cell infiltrate. Also, using linear regression analysis, there appeared to be a direct correlation between the number of lymphocytes and the number of bacteria in the capsule. In light of these findings, a potential causal link between biofilm formation, inflammation, T-cell proliferation, and finally the potential for degeneration into ALCL is discussed. Beyond this, the authors, who have a great interest in and an extensive experience with evaluating for biofilms, interpret their data as supporting a hypothesis that a greater bacterial load can lead to increasing severity of capsular contracture. This is shown in their pig data, where the implants associated with the more severely contracted capsules demonstrated a higher bacterial count. Also, all of the Baker grade IV human capsular tissue studied demonstrated the presence of biofilm, with more aggressive textures showing greater amounts of bacteria. It should be noted that all of these specimens were obtained from textured implant patients; no specimen was obtained from smooth-walled devices. Given this outline, several points merit specific comment. This study is hampered by low numbers. The Baker grade IV capsular data points in the pig were generated from only three of the 20 implants that form the basis of the study. Also, when the actual capsular tissue specimens in the pigs were examined, no significant difference in bacterial load between textured and smooth implants could be identified despite a trend that appeared to be supportive. In addition, there was no significant difference in the number of lymphocytes in the pig capsular tissue when comparing pockets containing smooth versus textured implants. It would be reasonable to hypothesize that it is the capsular tissue that would be the ultimate source of degenerated T cells leading to ALCL, and the lack of an expected tissue response with T-cell proliferation is surprising. For these reasons, it is reasonable to consider an alternative narrative for the biofilm to inflammation to T-cell proliferation to ALCL progression, and it is clear that more work is yet to be performed. The most pressing need is to generate the same data in the same model without intentional bacterial contamination of the pocket. If the same types of responses were noted in the absence of an intentional biofilm challenge, alternative explanations for the T-cell proliferation would have to be entertained. In addition, it must be stressed that not all textures are the same. Again, to provide useful clarity, the same study needs to be performed at the very least with implants coated with Siltex (Mentor Corp., Santa Barbara, Calif.) and polyurethane. In addition, capsular specimens from human patients with Baker grade I, II, and III capsules would ideally be studied and compared with specimens obtained from around smooth-walled implants. It may well be that specific differences between the textures may be playing a role in the development of inflammation and potentially ALCL. This is in fact what has been noted clinically, as mentioned by the authors. Almost all of the cases reported in the literature to date have occurred in association with textured surface devices, and although a few have been polyurethane, the overwhelming majority have been associated with a texture created using a “lost-salt” manufacturing process.1 One compelling hypothesis for this finding may be that the chronic inflammation noted by the authors is related to a small-particle synovitis that is secondary to the flaking off of torn shards of silicone that are created during the lost-salt texturization process that then become lodged in the capsule, setting off an inflammatory response. Such a soft-tissue response has been noted previously in the area of silicone wrist arthroplasty where fractured and abraded silicone wrist bone replacements have been associated with an intense synovitis that features cystic changes in the surrounding bones in association with silicone particles.3–6 The findings of this study actually support such reasoning, as the human capsules around both the lost-salt texture (Biocell; Allergen, Santa Barbara, Calif.) and less aggressive Siltex textured implant surfaces were noted to have elevated numbers of bacteria per milligram, and yet ALCL is almost completely a lost-salt– or polyurethane texture–related phenomenon. If it is truly the presence of bacteria that incites the degeneration into ALCL, one would expect both types of textured devices to be associated with ALCL, which is not the case. Also, the development of ALCL is noted to be a late phenomenon that occurs multiple years after implant placement.1 It is logical to postulate that any small-particle synovitis related to silicone shards would be constant and unrelenting over time, whereas biofilm formation stabilizes over time, with many of the bacteria becoming dormant.7,8 Of the two theories, the potential for chronic irritation caused by any retained silicone or chronically degrading polyurethane foam would seem to be better positioned to explain the clinical observations noted in cases of ALCL. Interestingly, it is this very state of chronic inflammation that has been postulated by some as the reason why capsular contracture rates are lower in some studies where Biocell textured implants have been investigated.9–11 The same has been noted for polyurethane foam–coated implants, where slow degradation of the foam leads to chronic inflammation and a noncontracted capsule.12–15 It may well be that the very processes many have relied on to control the interaction of the implant with the capsule can degenerate, albeit rarely, into an exaggerated T-cell response leading to outright ALCL. It is for this reason I personally believe the term “breast implant–associated ALCL” to be completely inaccurate and misleading. A more meaningful name that I believe identifies the cause and can more effectively direct future investigation would be chronic inflammation–associated ALCL, with the inflammation coming from either small-particle synovitis, polyurethane breakdown, biofilm, or even some combination of these or other factors. Until the true cause of this process is understood, we will be hard pressed to prevent its occurrence or identify appropriate treatment strategies. Therefore, as these authors move forward, specific issues regarding these various textures must be noted and controlled to provide the best results. Questions such as, did the textured surface study devices become incorporated? As the texture was peeled off the capsule, were small shards of silicone left behind in the capsule? Was the polyurethane covering still present in the human specimens? Should different bacteria be identified and studied? What is clear is that these authors are to be congratulated for initiating a very important direction of study. They are obviously skilled researchers and have much to offer our specialty. I encourage them and others to investigate these questions and repeat and confirm their results so that this very important entity can be more completely understood. Our patients are depending on it. DISCLOSURE Dr. Hammond has a consulting agreement with Mentor Corp. Dennis C. Hammond, M.D. Grand Rapids Area Medical Education Partners Partners in Plastic Surgery of West Michigan 4070 Lake Drive, Suite 202 Grand Rapids, Mich. 49546 [email protected]