On behalf of my coauthors, I thank Professor Santanelli di Pompeo and colleagues for their comments on our article, “Estimating the Prevalence of Breast Implant–Associated Anaplastic Large-Cell Lymphoma: A Systematic Review.”1 I appreciate the opportunity to respond to the points they raised. In regard to the omission of recent literature, all studies were identified using the predefined study search and inclusion criteria clearly set out in our study methodology, and an updated literature search was performed in January of 2024. In January of 2024, the systematic review by Santanelli di Pompeo et al.2 had not yet been indexed under PubMed’s main search terms or through citation linkage, so it was not visible to us. Furthermore, as set out in our methodology, we purposely excluded systematic reviews from our study, preferring instead to source and include primary data from the original studies. Nonetheless, we acknowledge the relevance of their systematic review and will consider its inclusion in future work. Considering the inclusion of the study by Adams et al.,3 we acknowledge that concerns exist around the study methodology and honesty of the authors4; however, it was published as a prospective international series in Plastic and Reconstructive Surgery, satisfied our inclusion criteria, and has not been retracted or amended by either the authors or the Journal. Importantly, we objectively appraised the methodological quality of the study in our work, highlighting it to be of low methodological quality (as assessed using the Strengthening the Reporting of Observational Studies in Epidemiology checklist) and at critical risk of bias (using the Risk of Bias in Nonrandomized Studies of Interventions tool). We acknowledge that differences exist in the breast implant–associated anaplastic large-cell lymphoma (BIA-ALCL) case numbers (numerators) for the articles by McGuire et al.,5 Cordeiro et al.,6 and Kolasinski et al.,7 reflecting cohort updates. With regard to the comment on biofilm hypothesis and clustering, we did not suggest any causality for the Cordeiro et al.,6 or Kolasinski et al.7 cohorts or, indeed, attribute case numbers to biofilm presence or surgical technique. Furthermore, we were extremely careful to point out the limitations of our study in the Discussion, including the need for further investigation, and we reiterate this point. We accept the nuanced concern regarding the denominator estimation method taken from the Dutch radiographic study by De Boer et al.,8 and that we did not utilize the Italian radiographic study by Santanelli di Pompeo et al.9 Nonetheless, we transparently adopted the 3.0% prevalence estimation and accepted that it may be an underestimate in our own appraisal of our study’s limitations, for instance, by being exclusive of transgender women and the 17- to 20-year-old age group. We did estimate the patient-based denominator by dividing implanted breast implants by 2 in a single-center study,10 as outlined transparently in our Table, Supplemental Digital Content 2, https://links.lww.com/PRS/H492. It is incorrect for Santelli di Pompeo et al. to state that we calculated the patient-based denominator by dividing sales data by 2, in any study, and then to subsequently make assumptions as to whether these patients were actually implanted or undergoing alternative procedures. We agree with Santanelli di Pompeo et al. that the prevalence of BIA-ALCL should be calculated according to implant surface type; however, we were unable to always perform this in our study, because not every study reported surface type in sufficient granularity to facilitate such calculations. Where possible, we did stratify according to “textured only” or “any type of implant” for clarity. The country-specific surface ratios outlined by Santanelli di Pompeo et al. for the United States were not available in the literature, limiting the additional resolution possible. Furthermore, and importantly, we explicitly stated that the inclusion of smooth implants did not imply a causal link and was simply a function of flaws in the constituent study designs. We accept the observation that denominator overlap exists for some studies, particularly the country-level data from the United States and the updated cohort from Cordeiro et al.6 This limitation was a function of synthesizing a heterogenous dataset. We agree that denominator overlap weakens the strength of our systematic review, but we refute that it makes the results of our systematic review wholly invalid. We also agree that single-center studies with well-characterized implant exposures and long-term follow-up provide the most methodologically robust understanding of BIA-ALCL prevalence least subject to bias. Nonetheless, the goal of our review was to calculate population-level prevalence for BIA-ALCL, and we suggest that cohort, cross-sectional, and country-level data all provide complementary insights for reaching these calculations. Finally, we appreciate the detailed commentary from Santanelli di Pompeo et al. We acknowledge that their own work has contributed meaningfully to the field, and we welcome further dialogue and collaborative opportunities to enhance precision as our understanding of BIA-ALCL epidemiology evolves. DISCLOSURE The author has no financial relationships or conflicts of interest to disclose. No funding was sought or received for this project. ACKNOWLEDGMENTS This communication by Dr. Ward is on behalf of all his co-authors: Thomas Calderbank, MRCS, Chee Chee Tang, MBBS, Naveen V. Goddard, MBChB, MRes, Fiona MacNeill, FRCS, Marios Tasoulis, FRCS, and Aadil A, Khan, FRCS(Plast).
Joseph Ward (Wed,) studied this question.