Introduction

The rate of prophylactic mastectomy has increased three-fold over the last decade. In most breast oncology centers, bilateral mastectomy procedures now outnumber unilateral procedures in the ratio of 2 to 1 [1]. The dramatic increase in prophylactic mastectomy rate is attributed to a convergence of many factors which include increased disease awareness, breast cancer screening efforts, genetic testing, younger patient age at diagnosis, patient anxiety, and potential for improved cosmetic result for bilateral procedures [2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21].

While some studies demonstrate decreased risk of subsequent breast cancer, others suggest that contralateral prophylactic mastectomy (CPM) provides no mortality benefit relative to breast-conserving surgery [22,23,24,25,26,27,28,29,30,31,32,33,34,35]. Psychologically, many patients continue to undergo the elective procedure out of anxiety over developing a new primary breast cancer [36,37,38,39,40,41,42,43,44,45]. A study conducted by Rosenberg et al. demonstrated that 87% of women reported high concerns of being diagnosed with contralateral breast cancer in the future, 98% undergo CPM to decrease their contralateral breast cancer risks, 80% were confident in their decision, and 90% would “definitely [choose]” to undergo the procedure again, if given the opportunity [46].

Secondary to the psychologic benefit conferred by CPM, patients often chose to undergo the prophylactic removal of the contralateral breast to maintain breast symmetry following reconstruction [14, 40, 41, 46, 47]. Huang et al. reported that 52.7% of women stated that the desire for symmetry was a very important factor when deciding to undergo contralateral prophylactic mastectomy [41]. Similarly, Han et al. demonstrated that the desire for breast symmetry was a driving factor when deciding to undergo CPM, in nearly a third of women [14].

Given these patient motivations for prophylactic contralateral procedures, there remains a gap in knowledge for clinicians in how to counsel patients with regard to postoperative complication rates following CPM. The available literature that evaluated clinical outcomes is limited to the investigation of perioperative complications or postoperative outcomes in a non-irradiated patient population [48,49,50]. Few studies have compared the risk of surgery between the disease versus prophylactic sides, for mastectomy and for implant-based reconstruction [50,51,52]. These answered questions led this study to assess surgical risks for the prophylactic mastectomy and implant reconstruction: is the surgical risk equivalent between the diseased and prophylactic breasts? This study examined the outcomes of a ten-year retrospective cohort following CPM with regard to the laterality of the breast, comparing the outcome of mastectomy and reconstruction between the disease versus prophylactic breast. Since the surgical procedure is comparable on each side, we hypothesized that the clinical outcome between the disease and contralateral prophylactic breast will be equivalent.

Methods

Data collection

A retrospective chart review was conducted at a tertiary medical center with Institutional Review Board approval, observing the declaration of Helsinki. Unilateral breast cancer patients, undergoing a bilateral mastectomy and immediate implant-based breast reconstruction between May 2008 and May 2018 at our institution, were identified. Exclusion criteria included patients who had undergone bilateral autologous reconstruction or delayed contralateral prophylactic mastectomy and reconstruction.

Patient characteristics recorded were age at surgery, body mass index (BMI), obesity, history of smoking, diabetes, type of cancer, cancer grade, laterality of cancer, completion mastectomy, chemotherapy, and pre-mastectomy or post-mastectomy breast radiation. BMI was calculated as mass in kilograms per meter squared. Obesity was defined as BMI > 30. Smoking was defined as any patient who smoked 6 weeks prior to surgery. Patients who initially underwent breast conserving surgery, but then presented with positive pathological margins and required mastectomy to clear the margin were considered as patients who underwent a completion mastectomy.

Surgical procedure characteristics were documented as follows: mastectomy type (nipple-sparing versus skin-sparing), axillary surgeries (sentinel node biopsy and/or axillary lymph node dissection), reconstruction method (direct-to-implant or tissue expander), location of prostheses placement (subpectoral or prepectoral), implant size, and type of surgical adjunct [Vicryl only, acellular dermal matrix (ADM) only, or Vicryl and ADM]. In this analysis, patients who underwent sentinel lymph node biopsy or axillary lymph node dissection at any time as a part of treatment of the incident cancer were included. For example, if a patient underwent lumpectomy with sentinel lymph node biopsy and then proceeded with a completion mastectomy, we included them in the sentinel lymph node biopsy group. Mastectomies and reconstructions were performed as previously described by our group [53,54,55,56,57]. Evaluation of the post-mastectomy flap thickness, skin color, and capillary refill was coupled with patient risk factors and the surgeon experience to determine final candidacy for the type of mastectomy and reconstruction. In cases where ADM was used, the manufacturers were either Alloderm (LifeCell Corp., Branchburg, NJ) or FlexHD Pliable (Mentor Worldwide, Santa Barbara, CA).

Only postoperative outcomes that required reoperation were recorded, as the treatment for minor complications can vary by the clinical practice of each surgeon. Clinical outcomes which were recorded included mastectomy skin flap necrosis, surgical site infection, hematoma, explantation, capsular contracture, and cosmetic revisions. Mastectomy skin flap necrosis was defined as a non-viable tissue with the evidence of ischemia or eschar formation requiring debridement. Surgical site infection was defined as cutaneous erythema around the incision site, often accompanied by purulent drainage. Explanation was defined as implant removal due to a postoperative complication without subsequent implant placement. Capsular contracture was defined as breast deformity and elevation of the prosthesis resulting from fibrous capsule formation and/or muscle contracture around the prosthesis. For the aforementioned complications, laterality of incidence (contralateral prophylactic breast, diseased breast or bilateral) was recorded and compared.

Statistical analysis

Fisher’s exact testing was used to assess categorical variables. A paired sample t-test was implemented to examine the outcome data between breasts. Given the treatment differences between breasts, which may influence complication rates, a penalized (Firth) logistic regression model was constructed to identify the relationship between postoperative complications and covariate variables. Models were controlled for prophylactic v. diseased breast, sentinel node biopsies, axillary lymph node dissection, post-mastectomy radiation, location of prostheses placement, type of surgical mesh, type of reconstruction, implant size, and completion mastectomy. Odds ratio, 95% confidence interval, and p value were calculated for each outcome. Statistical significance was defined as p < 0.05. Data were analyzed using SPSS 24 (IBM Corp., Armonk, NY).

Results

Clinical characteristics

We identified 1117 patients with unilateral breast cancer who underwent bilateral mastectomy and immediate implant-based breast reconstruction (Tables 1, 2). The breast cancer pathology types ranged from unilateral DCIS to stage I–III invasive breast carcinoma. The mean patient age was 47.75 ± 9.08 years, while the mean BMI was 25.69 ± 5.62 kg/m2. Among these patients, nipple-sparing mastectomies were performed in 1170 (52.37%) breasts, whereas skin-sparing mastectomies were performed in 1064 (47.63%) breasts. With regard to reconstruction type, 811 (36.30%) breasts underwent staged breast reconstruction with tissue expander placement, and 1423 (63.70%) underwent direct-to-implant breast reconstruction. In total, 105 (9.4%) patients underwent completion mastectomies, 302 (27.04%) underwent chemotherapy, 76 (6.8%) patients had a history of pre-mastectomy breast radiation, and 112 (10.03%) underwent post-mastectomy breast radiation. This study followed the evolution of our surgical department’s practice over the last decade, when an increasing number of contralateral prophylactic mastectomies were performed over the course of the study period (Fig. 1). Overall, the mean follow-up time was 63.85 ± 33.39 months.

Table 1 Patient characteristics
Full size table
Table 2 Surgical characteristics
Full size table
Fig. 1
figure 1

Trends in contralateral prophylactic mastectomy. Throughout the study period, there is a noticeable increasing trend in the number of contralateral prophylactic mastectomy with reconstruction procedures. Of note, the number of cases depicted in 2008 was from May to December, while the number of cases depicted in 2018 was from January to May

Full size image

Outcomes

The rate of overall complication in the diseased breast was 10.38%, while the rate of overall complications occurring in the contralateral prophylactic breast was 5.64%, and the rate of overall complications occurring in both breasts was 2.33%. Therefore, the risk of developing a complication in one or both breasts is 13.69%, the risk of developing a complication in either breast but not both is 11.36%, and the additional risk of having surgery on the contralateral prophylactic breast is 3.31%.

Infection (6.00%) was the most common postoperative complication followed by capsular contracture (4.66%), skin necrosis (4.57%), and hematoma (2.78%). Explanation occurred in 6.89% of patients and was typically due to infection or mastectomy skin flap necrosis. Explantation which resulted from a postoperative outcome other than capsular contracture, skin necrosis, infection or hematoma, occurred in 0.36%. Capsular contracture (p = 0.021) and infection (p = 0.000095) occurred at significantly higher rates in the diseased breast when compared to the prophylactic breast. Explantation, mastectomy skin necrosis, and hematoma rates did not vary to a statistically significant degree between breasts (Fig. 2). Interestingly, the rate of cosmetic revision was significantly higher in the prophylactic breast (2.78%) versus the diseased breast (1.43%) (p = 0.029) (Table 3).

Fig. 2
figure 2

Comparison of postoperative complications in prophylactic versus diseased breasts. Compared to a contralateral prophylactic breast, rates of capsular contracture and infection were significantly greater in the diseased breast. It is suspected that the routine need for additional treatment on the diseased breast is a contributing factor to the discrepancy in outcomes between breasts

Full size image
Table 3 Postoperative outcomes
Full size table

Prophylactic versus diseased breast

A penalized logistic regression was performed to investigate differences in outcomes between breasts and to account differences in patient habitus and comorbid conditions. The regression model demonstrated a significantly higher rate of infection occurring in the diseased versus contralateral prophylactic breast when controlling for confounding variables such as sentinel node biopsies, axillary lymph node dissections, radiation, location of implant placement, type of surgical mesh, type of reconstruction, implant size, and completion mastectomy (Table 4).

Table 4 Correlations between patient characteristics and complications
Full size table

A regression analysis was also performed to exclude patients who had previously undergone lumpectomy with radiation and presented with cancer recurrence which required mastectomy as the next treatment. When adjusted to exclude the lumpectomy recurrence patients who received prior radiation, the regression still revealed a significantly higher rate of infection in the diseased breast when compared to the contralateral prophylactic side. There were no other significant differences in outcomes between breasts (Table 5).

Table 5 Correlations between patient characteristics and complications when excluding pre-mastectomy radiation patients
Full size table

Discussion

Breast cancer will affect nearly 13% of US women during their lifetime [49]. Typically, the risk of breast cancer in the contralateral breast is due to a new primary tumor, not the regional spread of disease from one breast to another. After 10 years of primary breast cancer diagnosis, the average rate of contralateral breast cancer occurrence for the general population is estimated to be 5–10% [58,59,60,61,62,63,64,65]. In contrast, patients harboring a BRCA1/2 pathogenic genetic variant, the risk of breast cancer occurrence is about 10–30% [66,67,68,69,70,71,72,73].

In the past decade, many more women have chosen to undergo a CPM for a variety of reasons [14, 17, 19, 34, 40,41,42,43, 46, 47, 74, 75]. Since the contralateral prophylactic procedure is elective, it is important to critically weigh the additional risk of this procedure. This study presents one of the largest comparisons of postoperative outcomes in diseased versus contralateral prophylactic breasts, following bilateral mastectomy and immediate breast implant reconstruction. The evaluation of postoperative outcomes in reconstructed breasts in the same patient mitigates confounding patient characteristics, whereas the penalized regression model controls for potential differences in treatment and morbidity between breasts.

Analysis here revealed that the added risk of having surgery on the contralateral prophylactic breast is 3.31%. Capsular contracture and infection had a significantly higher chance of occurring in the diseased versus prophylactic breast (p = 0.021 and p = 0.000095, respectively). It is likely that these higher rates of complications stem from the treatment of the diseased breast. Often diseased breasts undergo sentinel node biopsy and axillary lymph node dissection at the time of mastectomy. This added intervention involves the manipulation of additional tissue planes and longer procedure times, both of which have been shown to increase infection rates [76,77,78,79]. As expected, there was a significant difference in sentinel node biopsies and axillary lymph node dissections between the prophylactic and diseased breasts (p = < 0.00001, p = < 0.00001). While neither sentinel node biopsies nor axillary lymph node dissections showed a significant correlation to capsular contracture or infection in the penalized logistic regression, a study conducted by Lucci et al. reported a significant difference in wound infections between patients who underwent sentinel node biopsies only  and those who underwent sentinel node biopsies and axillary lymph node dissections (p = 0.0026). [48]. We speculate that the additive effects of additional treatment correlate to higher complication rates in the diseased breast. Moreover, any diseased breast which underwent previous lumpectomy was exposed to pre-mastectomy radiation, and any diseased breast with node positive disease after mastectomy was exposed to post-mastectomy radiation. Exposure to irradiation leads to skin damage, soft tissue breakdown, local edema, and skeletal muscle fibrosis, which may further increase the risk of complications [80,81,82,83,84]. Spear et al. explained that even though the risks of prosthetic reconstruction complications are dependent on the timing and aggressiveness of the radiation, breast irradiation generally results in higher rates of infection and capsular contracture, which our results also illustrate [81].

Revision rates were higher in the prophylactic breast (2.78%) compared to the diseased breast (1.43%) (p = 0.029). However, it is important to note that a greater significant difference was seen when comparing revisions rates in the prophylactic breast alone (2.78%) to bilateral revision rates (18.08%) (p < 0.0001). Most often, revisions consisted of mastopexy procedures to improve the symmetry between breasts and scar revisions to lessen the thickening of the scar or fat grafting to mitigate implant wrinkling or to improve contour irregularities. The difference in revision rates between the prophylactic breast and the diseased breast was no longer significant when adjusting for all confounding variables in the penalized regression model. Nevertheless, the revision rates of this study compare favorably with previously published revision rates for immediate implant-based reconstruction [57, 85].

In addition to overall postoperative outcomes, we were also interested in determining whether certain outcomes were influenced by specific treatments or risk factors. Interestingly, in the penalized regression model, infection was the only complication that remained significantly higher in the diseased versus contralateral prophylactic breast (Infection OR 2.12, CI 0.225–1.387, p 0.006). Furthermore, this remained true when excluding patients who received pre-mastectomy radiation (Infection OR 2.104, CI 0.17–1.341, p 0.011). A similar study conducted by Crosby et al. failed to detect a significant difference in infection rates, between diseased and contralateral prophylactic breasts, following both implant and autologous breast reconstructions [49]. However, it is likely that their exclusion of irradiated patients contributed to discrepancies between our reported outcomes.

The main limitation of this study is its retrospective design. It is also possible that the study suffered from regional or institutional selection bias, as the evaluation was performed at a single tertiary academic center. Differences in operating techniques and postoperative clinical management among the plastic surgeons within the department could have also introduced variance in postoperative outcomes. Despite these limitations, this study demonstrates additive risks conferred with CPMs and immediate implant-based reconstruction within a large cohort of patients and long-term follow-up.

It has been suggested that decision making for CPM is highly influenced by surgeons, given their significant role in patient education [40, 75, 86]. The National Comprehensive Cancer Network Practice Guidelines stresses the importance of physicians creating a “shared decision-making environment” with their patients [87]. Patient factors that have been shown to contribute to the decision to undergo CPM include young age, white ethnicity, marital status, positive family history of breast cancer, positive genetic testing, and access to immediate breast reconstruction [13, 14, 17, 47, 50, 88,89,90,91,92,93,94]. Therefore, thorough preoperative management and consultation is critical. Our analysis will allow patients and treating surgeons to balance these factors against the risk of postoperative complications from undergoing a contralateral prophylactic mastectomy.

Conclusion

The risk of developing a postoperative capsular contracture and infection, following mastectomy and immediate breast reconstruction, is significantly higher in a diseased breast compared to a contralateral prophylactic breast. This study demonstrates a 3.22% additional risk of any complication occurring in a prophylactic breast. Patients and surgeons should carefully consider this additional risk when electing to undergo contralateral prophylactic mastectomy, especially if the patient has additional comorbid conditions such as prior radiation treatment, smoking status or obesity. Future multi-institutional prospective studies are recommended to gain higher level clinical evidence to better assess the additional risks of contralateral prophylactic mastectomy.