Introduction

In Japan, there are over 90,000 new cases of breast cancer each year. Among these, approximately half of women receive mastectomy [1]. A better cosmetic outcome is associated with a better quality of life in women who undergo mastectomy and breast reconstruction (BR) [2]. In Japan, implant-based BR has been covered by the national health insurance since 2013. However, among patients with breast cancer who receive mastectomy, the frequency of immediate BR is not high. In 2018, there were 48,233 patients who received mastectomy for primary breast cancer in Japan; among them, only 8,512 (18%) patients received BR [1, 3]. In contrast, the 2018 post-mastectomy BR rate in South Korea, another Asian country, was 53% [4].

Studies have investigated various barriers to BR [5]. One of these is physicians’ perception that BR adversely affects breast cancer prognosis owing to BR masking local recurrence or delaying adjuvant chemotherapy and radiotherapy [6,7,8,9,10,11,12]. Despite the absence of randomized prospective trial data, several meta-analyses and systemic reviews using retrospective data have demonstrated that BR does not affect local recurrence rates, disease-free survival, or overall survival in comparison with mastectomy alone [13,14,15]. However, it has been reported that many physicians have this misunderstanding about BR. It is of clinical importance to identify physicians’ perception about the impact of BR on patient prognosis and which physicians have misunderstandings about BR. However, limited information is reported on these issues.

We conducted an e-mail survey regarding the impact of BR on the prognosis of patients with breast cancer among members of the Japanese Breast Cancer Society. The objective of this study was to clarify the beliefs regarding BR among physicians engaged in breast cancer treatment in Japan.

Patients and methods

The Japanese Breast Cancer Society Scientific Committee reviewed and approved this study. A questionnaire regarding the impact of BR on the prognosis of patients with breast cancer was sent via e-mail to members of the Japanese Breast Cancer Society on 1 December 2021. Two reminders were sent on 20 December 2021 and 25 January 2022; the survey was closed on 25 February 2022.

The questionnaire was divided into three parts. Respondents were asked (1) their background, (2) whether they felt that BR affects the prognosis of patients with breast cancer, and (3) their estimation of patient prognosis (5-year overall survival rate) according to the site of local or regional recurrence (see Appendix for more details regarding the questionnaire, modified from Clemons et al. [16]).

Statistical analysis was conducted using Fisher’s exact test or the Mann–Whitney U test. Multivariate analyses were conducted using a logistic regression model. All statistical tests and p values were two sided, and P values < 0.05 were considered significant. All statistical analyses were performed using EZR version 1.5.5 (The R Project for Statistical Computing, Vienna, Austria) [17].

Results

Of 6,583 questionnaires, 369 (6%) were returned and were available for analysis. Demographic information about respondents is summarized in Table 1. Respondents were asked whether they believed that BR affects patient prognosis (Fig. 1); 27, 59, and 14% responded yes, no, and abstained or said they did not know, respectively. The associations of answers with respondents’ characteristics are shown in Table 2. Female respondents were more likely to answer yes to the question than men (P = 0.006). Respondents who felt that BR affects patient prognosis saw fewer new breast cancer cases per year than those who did not (P = 0.007). Logistic regression analyses showed that both factors (gender and number of breast cancer cases per year) were significantly associated with the belief that BR affects patient prognosis (P = 0.05 and 0.04).

Table 1 Physician’s background (N = 369)
Full size table
Fig. 1
figure 1

Responses to the question, "Generally, do you believe that breast reconstruction at the time of the initial surgery affects patient prognosis?"

Full size image
Table 2 Association of respondents' belief that breast reconstruction affects or does not affect recurrence with respondents' background
Full size table

We evaluated the relationship of responses and respondents’ estimation of 5-year overall survival rates among patients with local or regional recurrence. Significant differences between responses and the estimations were seen for the following sites: ipsilateral breast after breast-conserving surgery and ipsilateral axillary lymph node (estrogen receptor-positive and human epidermal growth factor receptor 2 [HER2]-negative tumors), skin of chest wall (isolated, after mastectomy) (estrogen receptor-positive and HER2-positive tumors) (Table 3). Respondents who felt that BR affects patient prognosis estimated worse prognosis at all sites than those who did not.

Table 3 Association of physicians' belief that breast reconstruction affect or does not affect recurrence with their estimation of 5-year overall survival, according to site
Full size table

Discussion

In our study, approximately one-quarter (27%) of respondents felt that BR affects patient prognosis. Physicians’ perception that BR affects patient prognosis could stem from the assumption that BR interferes with the detection of local recurrence or the introduction of adjuvant chemotherapy or radiotherapy, although a large amount of data from meta-analyses and systematic reviews show that BR following mastectomy does not affect patient prognosis [13,14,15]. Moreover, several studies have demonstrated that there are no differences in prognosis between breast-conserving surgery with and without BR [18, 19]. Previous studies have reported that 20–75% of physicians thought that BR can mask and delay the detection of local recurrence [6,7,8,9,10,11,12]. Takahashi et al. conducted a questionnaire survey on BR among physicians certified by the Japanese Breast Cancer Society in 2001 and obtained responses from 635 physicians [9]. In total, 47.4% of respondents thought BR can mask local recurrence and were less likely to refer their patients for BR [9]. Coroneos et al. studied physicians’ perceptions toward BR according to subspecialty and reported that oncologists were more likely to believe that BR can mask local recurrence and delay adjuvant chemotherapy and radiotherapy, compared with surgeons (30% vs. 14%, P < 0.001) [11]. Contrary to the study results of Coroneos et al. [11], we found no difference in respondents’ perceptions that BR affects patient prognosis according to subspecialty; this is probably owing to the small proportion of respondents with subspecialties other than breast surgery.

A concern that BR can mask local recurrence might be theoretically incorrect in most mastectomy cases. A recent systematic review reported that among patients who received skin- or nipple-sparing mastectomy, 96–100% of local recurrence was located within the skin or subcutaneous tissue [20]. Skin and subcutaneous tissue are located over the reconstructed breast in cases of both autologous and implant-based reconstruction; BR does not mask local recurrence in such cases.

In our study, female respondents were more likely to have a concern that BR affects patient prognosis. Previous studies have reported that female physicians were more likely to refer their patients for BR than male doctors, which is contrary to our results [10, 21]. A possible reason is the different proportions of female respondents between our study and previous works, with many more women participating in our research (38%) than in previous studies (6% [10] and 14.4% [21]). This low proportion of female respondents is similar to another study conducted in Japan (3.8%) [9]. Because the proportion of women was extremely low in these studies, there is a possibility that female physicians’ perceptions about BR were not accurately reflected in the study results. In Japan, the number of female physicians is increasing and in 2020, the proportion of female physicians was 23% [22]. We believe that our results reflect the current situation in Japan more accurately than previous studies.

In our study, respondents who treated fewer new patients with breast cancer per year felt that BR affects patient prognosis. A previous study reported that physicians who treated many new patients with breast cancer tended to refer their patients for BR, which is comparable to our findings.

To our knowledge, our study is the first report regarding physicians’ estimation of the prognosis in patients with local or regional recurrence after breast surgery and BR. We found that respondents who thought BR can affect patient prognosis estimated a worse 5-year overall survival in patients with local or regional recurrence in different sites after breast surgery and BR. Our results indicated that these physicians thought BR results in worse prognosis. Interestingly, all patients estimated to have a poor prognosis had estrogen receptor-positive breast cancer. A possible explanation is that patients with estrogen receptor-negative breast cancer are estimated to have much worse prognosis regardless of whether BR is performed.

Our study has several limitations. The main limitation is the low response rate (6%), which is partly owing to the COVID-19 pandemic. It has been suggested that the COVID-19 pandemic has resulted in survey responses that are slower than usual owing to the increased workload of physicians [23]. Next, most (86%) respondents were breast surgeons, with very few respondents from other subspecialties. Finally, we conducted multiple analyses, which could have easily produced significant differences.

In conclusion, our survey demonstrated that a quarter of respondents believe that BR affects patient prognosis and underestimate the survival rates of patients who receive BR and subsequently have local or regional recurrence in different sites. Because of a lack of evidence regarding the impact of BR on patient prognosis, educating physicians by providing accurate knowledge regarding BR and patient prognosis is highly recommended.