Sir: There is an increasing concern regarding breast implants causing breast implant–associated anaplastic large cell lymphoma (BIA-ALCL).1,2 There is immunological and histological evidence relating breast implants to ALCL, but the epidemiological evidence is weak. Early in 2019, an Australian study1 reported different rates of BIA-ALCL according to implant surface area and roughness. The authors estimated a higher incidence rate of BIA-ALCL in the polyurethane (PU) group at nearly 6.5 cases for every 10,000 implants sold after 9 years of exposure. This finding is relevant as it is a result of an effort to bring some light to a field where there is lot to clarify. There are several challenges in estimating such rates, mostly involving how to achieve accurate data.3 Besides, other elements may play a role as contributing causes of BIA-ALCL, such as genetic profile (JAK/STAT3) or genetic background, immunological profile, subsequent events leading to implant biofilm, environmental exposure, ethnicity, and so on.3 Ignoring other causes and not adjusting for their potential effect may lead to bias due to confounding.4 The logistic regression approach, used in the report, may only estimate risk up to a time limit, defined at the study design.4 Therefore, it is not clear how the authors incorporated the time issue for the risk estimation over time. For instance, was each year considered a different cohort, or were rates estimated considering all sales up to the event? In addition, the authors state that there was a difference in time of exposure to the different surfaces/trademarks, but there was no further discussion of how this might impose limitations on rate interpretation. In this context, a usual epidemiological approach of incidence density rate estimation (events/implant-year) over time would probably suffice, as it gives an idea of risk, it allows dynamic population context, it is possible to decompose the rates over the periods of interest, and it allows you to explore cohort and period/generation effects.4 The authors state that there were cluster patterns in the data,1 which was suspected to be related to certain surgeons and hospitals, raising a concern about implant infection as a cause. Surgical site infection is a preventable condition, and it was not clear if in these clusters the infection rates were above or below an acceptable limit at the time of implantation. There is a matter of clinical relevance. Polyurethane implants are easier to manipulate, they have higher adherence, and there is less free displacement throughout the capsular space compared with smooth surface implants.5 The BIA-ALCL rate may be relatively higher in the polyurethane group, but it has lower rates than other conditions that also require implant removal, such capsule contracture, rupture, and malposition. In addition, in its early stages, when most of patients are diagnosed, the lymphoproliferative condition has a very good prognosis, requiring no additional treatment.1 The physician must inform the patient of all risks, particularly the risks of all events that may require implant removal or jeopardize the aesthetic result, including the risk of BIA-ALCL. Together, the patient and the surgeon should decide what is best. DISCLOSURE Drs. Cintra, Massiere Y Correa, and Brasil are scientific consultants for Silimed. The remaining authors have no financial interests to disclose. No funding was received for this communication. Henrique P. L. Cintra, Dr.Universidade Estadual do Rio de JaneiroHospital Universitário Pedro Ernesto Wanda Elizabeth Massiere Y Correa, Dr.Pontifícia Universidade Católica Amanda T. BaptistaRoberta C. MillnaSergio Carlos A. J. JuniorSilimed Indústria de Implantes LTDA Pedro Emmanuel A. A. do Brasil, Ph.D.Fundação Oswaldo CruzInstituto Nacional de Infectologia Evandro ChagasRio de Janeiro, Brasil
Read full abstract