Comment on the Invited Discussion on “Assessment of Risk Factors for Rupture in Breast Reconstruction Patients with Macrotextured Breast Implants”

We here present a few comments on the invited discussion of Dr. van Heijningen on the paper “Assessment of Risk Factors for Rupture in Breast Reconstruction Patients with Macrotextured Breast Implants”. Dr. van Heijningen made some reservations regarding paper conclusions due to the high dropout rate, the adopted exclusion criteria and the location and mechanism of implant rupture. First of all, a high dropout rate is not unbeknown to researchers in surveys-based studies and may be expected when recalling in 6 months a population observed during last 20 years. In our study data are missing at random not affecting the risk of bias, while the population accurately depicts the people we care, mainly but not only reconstructive. Patients who did not respond to the questionnaire could not participate to the survey, while those who did not hold recent imaging were excluded because of the risk of false negative due to possible silent rupture, accounting to 10% in some reports. MRI imaging often shows that implants fold back on their selves when capsular contracture reduces implant pocket. As the use of the underwire bra prevents implant inferior displacement, repeated muscular contraction may worsen implant folds and the chronic wear-and-tear mechanism may be responsible for the rupture. Finally, folding is presumably easier to occur at the upper quadrants where anatomical implant shell is thinnest and gel concentration reduced than the opposite, therefore is not surprising that the higher percentage of ruptures is located in the upper implant quadrants. Level of evidence V This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266.

Abstract We here present a few comments on the invited discussion of Dr. van Heijningen on the paper ''Assessment of Risk Factors for Rupture in Breast Reconstruction Patients with Macrotextured Breast Implants''. Dr. van Heijningen made some reservations regarding paper conclusions due to the high dropout rate, the adopted exclusion criteria and the location and mechanism of implant rupture. First of all, a high dropout rate is not unbeknown to researchers in surveys-based studies and may be expected when recalling in 6 months a population observed during last 20 years. In our study data are missing at random not affecting the risk of bias, while the population accurately depicts the people we care, mainly but not only reconstructive. Patients who did not respond to the questionnaire could not participate to the survey, while those who did not hold recent imaging were excluded because of the risk of false negative due to possible silent rupture, accounting to 10% in some reports. MRI imaging often shows that implants fold back on their selves when capsular contracture reduces implant pocket. As the use of the underwire bra prevents implant inferior displacement, repeated muscular contraction may worsen implant folds and the chronic wear-and-tear mechanism may be responsible for the rupture. Finally, folding is presumably easier to occur at the upper quadrants where anatomical implant shell is thinnest and gel concentration reduced than the opposite, therefore is not surprising that the higher percentage of ruptures is located in the upper implant quadrants. Level of evidence V This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the We would like to thank the author [1] for their comments in the invited discussion on our manuscript [2]. Despite being an interesting and insightful read, we feel the need to respond to some of the elements raised by the author in their analysis of our work.
Dr. van Heijningen pointed out that the high dropout rate is a relevant factor. However, this is not unbeknown to researchers conducting survey-based studies and may be expected when calling back a population from the last 20 years during a 6-month period. It could be explained by several factors, including distance from the surgical centre, fatigue from treatment, or even excessive length of the questionnaire which ambitiously featured 23 different items [3]. Overall, we believe that data are missing at random, thus not affecting risk of bias in this regard [4]. Our population of 191 patients (288 implants) likely depicts the group of patients our health service caters to in an accurate manner. Furthermore, we specifically attempted to avoid a selection bias by not excluding non-reconstructive patients, despite them representing a minor percentage of our day-to-day activity. While our population might not appear homogenous, it was mostly reconstructive (88.2% of cases), with anatomical implants (87.8%) placed in the submuscular plane (88.5%), with an overall 95.1% of cases using macrotextured devices.
Some additional considerations should be done regarding the fact that we excluded patients who did not answer the questionnaire and did not receive recent breast imaging. Van Heijningen states that patients without complaints ''might be less likely to respond compared to patients who had issues''. While this observation is true and could have potentially introduced a bias in our findings, we believe that in reality this concern is widely overshadowed by how our results are potentially underestimated, rather than the opposite. In fact, we diagnosed breast implant rupture in 15.1%, while core studies from major breast implant manufacturers report rupture rates as high as 38.7% (findings in Table 6 from our manuscript). Patients who do not complain and do not actively return to follow-up, might present a silent breast implant rupture, which can only be confirmed by performing imaging [5]. A survey-based study by Salzman recently estimated that as many as 10.6% of women with breast implants might have silent implant ruptures [6], thus confirming our decision of downscaling our patient population with these stringent exclusion criteria rather than maintaining a broader population with potentially false negative cases, which could have risked making our findings potentially more uncertain.
Regarding the location of the rupture in our series, indeed they were more commonly found at the upper rather than lower quadrants and posterior rather than anterior aspect of the implant (Figure 1 from our manuscript). This was unexpected, but possibly suggests a different rupture mechanism than previously believed. Of note, 53.9% of all patients from this study received an implant-enhanced latissimus dorsi (LD), where the breast implant's upper quadrants are covered by both the pectoralis major and the LD muscle [7]. The upper implant compression during muscular contraction, combined with the underwire bra inferior immobilization, along with capsular contraction, may have contributed to implant shell folding [8], wearand-tear and ultimately rupture at the upper poles, where anatomical implants are the thinnest and cohesive gel concentration is the lowest. Furthermore, as pointed out by van Heijningen, most ruptures are expected to occur in the lower outer quadrants (LOQs) with the assumption that in this area implants are usually covered the least. At this regard, in 61.4% of our population, adequate muscle coverage was provided to LOQs with the LD muscle in the LD-based cohort and with the serratus anterior muscle in the expander/implant cohort; hence, the assumption that the LOQs are logically more prone to ruptures being the least covered areas is probably not entirely accurate for our population.
Overall, our study does present some declared flaws which are inherent to the study design. However, using our prospective database was made necessary as quality implant registries are not a reality worldwide, and certainly not in Italy where it was announced 11 years ago and only exists as a pilot study as of yet [9,10]. Finally, while we may agree to disagree on the limitations from our reported study, we will agree that further research is warranted to support and strengthen our findings.
Funding This research received no grant of any kind from any funding agency in the public, commercial, or not-for-profit sectors.

Declarations
Conflict of interest The NESMOS Department has received research funds from Motiva, Establishment Labs in 2017 and from GC Aesthetics in 2018 and 2020. The Department also received miniature breast implants from Establishment Labs, GC Aesthetics and Sebbin for research purposes in 2017. Prof. Santanelli di Pompeo would like to disclose that he serves as a paid consultant for BellaSeno GmbH and for Notified Body 0373 carrying out CE Mark certification activities for the Italian Ministry of Health for the year 2021 and 2022, and that he received reimbursements from ICEAG in 2015 and SCHEER-WG in 2019, 2020 and 2021. He has no ownership or investment to disclose. Neither the corresponding author nor any of the coauthors have received any type of funding or other supports from a sponsor/commercial entity/third party. All other authors hereby certify that, to the best of their knowledge, no financial support or benefit has been received, neither by themselves directly nor by any member of their immediate family or any individual or entity with whom or with which they may have a significant relationship from any commercial source which is related directly or indirectly to the scientific work which is reported on in the article.
Informed Consent For this type of study informed consent is not required.
Human and Animal Rights Not applicable.