Ann Surg Oncol (2015) 22:3474–3480 DOI 10.1245/s10434-015-4628-7
ORIGINAL ARTICLE – BREAST ONCOLOGY
The Effect of Contralateral Prophylactic Mastectomy on Perioperative Complications in Women Undergoing Immediate Breast Reconstruction: A NSQIP Analysis Amanda K. Silva, MD1, Brittany Lapin, MPH2, Katharine A. Yao, MD, FACS3, David H. Song, MD, MBA, FACS1, and Mark Sisco, MD, FACS4 Section of Plastic and Reconstructive Surgery, University of Chicago Hospitals, Chicago, IL; 2Center for Biomedical Research Informatics, NorthShore University HealthSystem Research Institute, Evanston, IL; 3Division of Surgical Oncology, NorthShore University HealthSystem, Evanston, IL; 4Division of Plastic and Reconstructive Surgery, NorthShore University HealthSystem, Evanston, IL
1
ABSTRACT Background. Women with breast cancer are increasingly choosing to undergo contralateral prophylactic mastectomy (CPM) despite questionable survival benefit and limited data on added risks. Little is known about differences in perioperative complications between women who undergo bilateral mastectomy (BM) versus unilateral mastectomy (UM) with reconstruction. Methods. The American College of Surgeons National Surgery Quality Improvement Program Participant Use Files (2005–2013) were used to identify women with unilateral breast cancer who underwent UM or BM with reconstruction. Adjusted 30-day complications were compared between UM and BM groups using logistic regression models. Results. A total of 20,501 patients were identified, of whom 35.3 % underwent BM. Of these, 84.3 % had implant reconstruction and 15.7 % had autologous reconstruction. For all women, BM was associated with longer hospital stays (adjusted odds ratio [aOR] 1.98–2.09, p \ 0.001) and a higher transfusion rate than UM (aOR 2.52–3.06, p \ 0.001). BM with implant reconstruction was associated with a modestly increased reoperation rate
This study was presented as an oral presentation at the ASCO Breast Cancer Symposium, September 4, 2014, San Francisco, CA. Society of Surgical Oncology 2015 First Received: 24 January 2015; Published Online: 22 May 2015 M. Sisco, MD, FACS e-mail: [email protected]
(aOR 1.15, p = 0.029). BM with autologous reconstruction was associated with a higher wound disruption rate (aOR 2.51, p = 0.015). Surgical site infections, prosthesis failure, and medical complications occurred at similar rates in UM and BM groups. Conclusions. CPM is associated with significant increases in some, but not all, surgical site complications. CPM does not increase the likelihood of medical complications, which are generally infrequent.
Women with unilateral breast cancer are choosing to undergo bilateral mastectomy (BM) at dramatically increasing rates.1–3 Data from the National Cancer Data Base indicates that from 2003 to 2010, the proportion of women with American Joint Committee on Cancer stage 0 to II unilateral breast cancer undergoing BM has more than doubled, from 4.3 to 9.7 %.4 This increase has been especially pronounced in women under the age of 45, among whom the rate has climbed to 24.1–33 %.4,5 This suggests that up to 25,000 women may be undergoing contralateral prophylactic mastectomy (CPM) each year. This increase is especially striking because few women undergoing CPM have been shown to derive a survival benefit.5–9 In BRCA gene-negative patients, the likelihood of developing a contralateral breast cancer is low and is outweighed by the risk of systemic disease.9–13 The fact that many women are electing to undergo additional surgery with limited efficacy data has led to recognition of an intrinsic conflict between the desire to practice evidencebased medicine and to respect patient wishes.14,15 Rosenberg et al. elucidated reasons women pursue CPM.16 This study did not address the extent to which
The Effect of Contralateral Prophylactic Mastectomy on Perioperative Complications in Women…
surgical risk was considered; however, one third of patients stated that the procedures required were higher than expected. Although it may be intuitive to providers that CPM is associated with higher complication rates, patients do not necessarily perceive added risk. An understanding of the risks and benefits of surgery is intrinsic to informed consent. Yet which risks increase after BM with reconstruction, and by how much, have not been studied in a large multi-institution cohort. A previous study using the National Surgery Quality Improvement Program database showed an increased risk of complications from BM among patients who did not undergo reconstruction.17 However, over 75 % of women who undergo BM also elect immediate reconstruction.18–20 For women with unilateral breast cancer, decision making is hampered by a lack of multi-institution data describing the additive risks of CPM. Additionally, decision aids that address risks and benefits of CPM require data that are accurate and broadly applicable. The American College of Surgeons National Surgery Quality Improvement Program (NSQIP) database is a nationally validated, risk-adjusted surgical outcomes database involving over 500 hospitals that is designed to measure and improve the quality of surgical care. The data collection process has been shown to be both reliable and valid.21,22 This database provides the power, breadth, and consistency necessary to accurately assess the incidence of uncommon complications across patient populations. In order to better inform medical decision making, this study sought to assess differences in perioperative complications between BM versus unilateral mastectomy (UM) with reconstruction among women with unilateral breast cancer. METHODS Subjects Data from the American College of Surgeons NSQIP Participant Use Files from 2005 to 2013 were accessed. Patient demographics, preoperative, and 30-day postoperative data are collected by a trained surgical clinical reviewer via medical chart review and entered in a Insurance Portability and Accountability Act-compliant, secure, Web-based platform from a random sample of patient charts. Because data are deidentified, institutional review board approval was not necessary for this study. Women aged C18 undergoing either UM or BM were identified using Current Procedural Terminology (CPT) codes. Yearly variations in CPT codes were taken into consideration. Mastectomy codes included those for simple mastectomy (19180/19303), skin-sparing mastectomy (19182/19304), or modified radical mastectomy (19240/
3475
19307). From this cohort, immediate breast reconstruction patients were identified using CPT codes and were dichotomized into two groups on the basis of reconstruction type: implant-based reconstruction included immediate implant (19340) and tissue expander placement (19357); autologous reconstruction included the latissimus dorsi flap with or without an implant (19361), pedicle transverse rectus abdominis myocutaneous flaps (19367, 19368, 19369), and free tissue transfer (19364). Only patients whose associated diagnosis was breast cancer (174.0– 174.9) or ductal carcinoma-in situ (233.0), or who had a personal history of malignant breast neoplasm (V10.3) were included. To limit the cohort to women with unilateral breast cancer, those who had bilateral sentinel lymph node biopsies or bilateral modified radical mastectomies were excluded. Patients who underwent different reconstructive procedures on either side were also excluded. Outcomes Examined 30-day outcomes included superficial (skin or subcutaneous tissue) and deep (fascial and muscle) surgical site infections, wound disruption, pneumonia, pulmonary embolism (PE), renal insufficiency, acute renal failure, urinary tract infection, stroke, cardiac arrest, myocardial infarction, deep venous thrombosis (DVT), sepsis, septic shock, transfusions within 72 h of surgery, prosthesis or flap failure, procedure length, length of hospital stay (C2 days for implant-based and C4 days for autologous reconstruction), and reoperation. Readmission rates were not included, as these were not introduced to NSQIP until 2011. In order to simplify and increase the power of the analysis, certain outcomes were collapsed as follows: superficial and deep surgical site infection were collapsed into surgical site infection (SSI); PE and DVT into venous thromboembolism; cardiac arrest and myocardial infarction into myocardial infarction or cardiac arrest; sepsis and septic shock into sepsis; and renal insufficiency and acute renal failure into renal insufficiency or failure. Statistical Analysis Patient age, race, comorbidities, and reconstruction type were compared between unilateral and bilateral cohorts. Frequency counts and percentages were calculated for categorical variables, and groups were compared by the Chi square test or Fisher’s exact test for small cell counts. Median with interquartile range was calculated for continuous variables, and groups were compared by Student’s t test. Unadjusted complication rates are shown separately by reconstruction type, implant and autologous, for uni-
3476
A. K. Silva et al.
lateral versus bilateral women. Multivariable logistic regression models were constructed to control for demographic and comorbidity characteristics that differed significantly between unilateral and bilateral women. Models were adjusted for age, race, smoking, diabetes, chronic obstructive pulmonary disease, hypertension, and the receipt of neoadjuvant chemotherapy. Adjusted odds ratios (aORs), 95 % confidence intervals, and p values are presented for BM (vs. UM). p \ 0.05 was considered statistically significant. SAS statistical software, version 9.3 (SAS Institute Inc, Cary, NC), was used for data analyses. RESULTS A total of 20,501 patients were included in the analysis; 7233 (35.3 %) women underwent BM and 13,268 (64.7 %) underwent UM (Table 1). Women who underwent BM were younger and more likely to be white. They were less likely to have hypertension or diabetes. However, they were more likely to smoke and to have undergone neoadjuvant chemotherapy within 30 days of surgery. Table 2 presents reconstruction types performed in each cohort. Most patients in both groups underwent implantbased reconstruction, but women who underwent UM were more likely to have undergone autologous reconstruction than those who underwent BM (19.5 % of UM vs. 8.9 % of BM, p \ 0.001). Tables 3 and 4 depict surgery duration and hospital length of stay as well as medical complications. Unsurprisingly, patients who underwent BM had longer surgery
times and hospital stays. The adjusted rate of total and individual medical complications was similar in both groups. Individual medical complications were infrequent, each occurring in fewer than 1 % of patients, with the incidence of myocardial infarction, cardiac arrest, and stroke less than 0.1 %. There were no mortalities in the UM group and two mortalities in the BM group (p value for difference = 0.12). Tables 5 and 6 depict adjusted surgical site complication and reoperation rates. Implant-based BM was associated with higher rates of transfusion (aOR 2.52, p \ 0.001) and slightly more frequent reoperations (aOR 1.15, p = 0.029) (Table 5). Patients who underwent autologous BM were similarly more likely to require transfusion (aOR3.06, p \ 0.001) and had higher wound disruption rates (aOR 2.51, p = 0.015) (Table 6). There were no significant differences in rates of SSI, prosthesis failure, or flap failure between the UM and BM groups. DISCUSSION This report from the American College of Surgeons NSQIP is the largest comparative analysis of complications from CPM with immediate reconstruction to date. BM was not associated with more medical complications, which were infrequent in both cohorts. However, all women who underwent BM were more likely to require a transfusion and had longer hospital stays. Of the 84.3 % of women who underwent implant-based reconstruction, those who had BM were slightly more likely to experience an
TABLE 1 Characteristics of mastectomy patients by laterality (n = 20,501) Characteristic
Unilateral mastectomy
Bilateral mastectomy
Total, n (%)
13,268 (64.7)
7233 (35.3) \0.001
Race, n (%) White
9828 (74.1)
Black
1128 (8.5)
453 (6.3)
706 (5.3)
381 (5.3)
590 (4.5)
251 (3.5)
1016 (7.7)
523 (7.2)
Hispanic Other Unknown Age, median (q1, q3)
p value for difference
52 (46, 60)
5625 (77.8)
49 (43, 56)
\0.001
Comorbidities/risk factors, n (%) Obesity (BMI C30 kg/m2) Smoking
3679 (27.9) 1636 (12.3)
1963 (26.9) 978 (13.5)
Diabetes
722 (5.4)
295 (4.1)
COPD
111 (0.8)
44 (0.6)
CHF or MI Hypertension Stroke Chemotherapy within 30 days
0.32 0.015 \0.001 0.071
5 (0.04)
2 (0.03)
0.99
3552 (26.8)
1569 (21.7)
47 (0.5)
24 (0.6)
0.77
360 (3.9)
300 (6.9)
\0.001
\0.001
BMI body mass index, COPD chronic obstructive pulmonary disease, CHF congestive heart failure, MI myocardial infarction
The Effect of Contralateral Prophylactic Mastectomy on Perioperative Complications in Women…
3477
TABLE 2 Reconstruction type by laterality Characteristic
Unilateral mastectomy (n = 13,268)
Implant-based reconstruction, n (%)
Bilateral mastectomy (n = 7233)
10,681 (80.5)
6592 (91.1)
Tissue expander
9124 (68.8)
5709 (78.9)
Immediate implant
1557 (11.7)
883 (12.2)
2587 (19.5)
641 (8.9)
Autologous reconstruction, n (%) Latissimus flap (with or without TE or implant) Pedicle TRAM Free flap
784 (5.9)
157 (2.2)
1176 (8.9)
233 (3.2)
627 (4.7)
251 (3.5)
TRAM transverse rectus abdominis myocutaneous
TABLE 3 Length of stay, medical complication rates, and multivariable regression models in implant-reconstructed patients Characteristic
Unilateral mastectomy (n = 10,681) 173 (133, 222)b
Bilateral mastectomy (n = 6592) 226 (176, 281)b
Adjusted OR (95 % CI)a
p value for difference
–
\0.001
4460 (41.8)
3957 (60.0)
2.09 (1.96, 2.22)
\0.001
110 (1.0)
70 (1.1)
1.06 (0.78, 1.44)
0.72
Pneumonia
7 (0.1)
5 (0.1)
1.35 (0.51, 3.57)
0.55
Renal insufficiency or failure
3 (0.03)
2 (0.03)
0.89 (0.25, 3.10)
0.85
Length of surgery (min), median (q1, q3) Length of hospital stay C2 days, n (%) Medical complications, n (%)
Sepsis
46 (0.4)
30 (0.5)
1.04 (0.66, 1.64)
0.89
Urinary tract infection Venous thromboembolism (DVT ? PE)
25 (0.2) 32 (0.3)
11 (0.2) 26 (0.4)
0.74 (0.38, 1.44) 1.41 (0.85, 2.34)
0.37 0.18
OR odds ratio, CI confidence interval, DVT deep venous thrombosis, PE pulmonary embolism a
ORs adjusted for age, race, smoking, diabetes, chronic obstructive pulmonary disease, and hypertension
b
Data presented as median (Q1, Q3)
unanticipated return to the operating room than those who had UM. Women who underwent BM with autologous reconstruction, meanwhile, had significantly more wound disruptions. Several single-institution retrospective studies have quantified the risks of complications after BM with reconstruction.23–27 However, these studies are difficult to interpret as a whole as a result of variations in the treatment groups, study methodologies, and data interpretation. Some reports include patients without cancer who underwent bilateral risk-reducing mastectomy, and others include patients who underwent delayed reconstruction.26–28 Also, some describe complications on a per-breast rather than per-patient basis, which is potentially difficult for patients and their providers to interpret.24,26,28 Two series examined complications after BM with reconstruction yet did not directly compare these rates to women undergoing UM.26,28 Finally, authors present differing conclusions based on similar findings, which illustrates the variation with which experienced researchers interpret data. One report from the Memorial Sloan Kettering Cancer Center indicated an overall complication rate after BM that
compared favorably to historical institutional data on UM, leading the authors to conclude that ‘‘bilateral implant reconstruction does not appear to pose additive risk.’’26 In contrast, a study from MD Anderson Cancer Center cautioned to ‘‘carefully consider the oncologic benefits of a CPM in light of the risk of increased surgical morbidity.’’24 Several previous studies of the NSQIP have alluded to the morbidity of BM. These studies were designed to identify predictors of complications after mastectomy. They did not directly assess those attributable to CPM.29 For example, Fischer et al. developed a model to predict complications after immediate breast reconstruction; however, laterality was not found to be predictive of complications and was not included in the model.30 Likewise, Gart et al. did not identify bilateral surgery as a risk factor for complications in women undergoing mastectomy.31 Of note, these studies included patients without cancer who underwent bilateral prophylactic mastectomy. Another NSQIP study has reported increased 30-day complications after BM without reconstruction.17 This is important information for women considering CPM who do not desire breast reconstruction. However, over 75 % of
3478
A. K. Silva et al.
TABLE 4 Length of stay, medical complication rates, and multivariable regression models in autologous-reconstructed patients Characteristic Length of surgery (min), median (q1, q3) Length of hospital stay C4 days, n (%) Medical complications, n (%) Pneumonia Renal insufficiency or failure
Bilateral mastectomy (n = 641)
Adjusted OR (95 % CI)a
p value for difference
477 (377, 626)b
–
\0.001
1308 (50.6)
431 (67.2)
1.98 (1.65, 2.38)
\0.001
70 (2.7)
17 (2.7)
0.99 (0.58, 1.72)
0.99
13 (0.5)
1 (0.2)
0.37 (0.08, 1.74)
0.21
Unilateral mastectomy (n = 2587) 326 (237, 439)b
2 (0.1)
1 (0.2)
3.00 (0.64, 14.13)
0.17
Sepsis
20 (0.8)
6 (0.9)
1.19 (0.51, 2.79)
0.68
Urinary tract infection Venous thromboembolism (DVT ? PE)
21 (0.8) 22 (0.9)
3 (0.5) 8 (1.3)
0.67 (0.23, 1.94) 1.81 (0.84, 3.91)
0.46 0.13
OR odds ratio, CI confidence interval, DVT deep venous thrombosis, PE pulmonary embolism a
ORs adjusted for age, race, smoking, diabetes, chronic obstructive pulmonary disease, and hypertension
b
Data presented as median (Q1, Q3)
TABLE 5 Reoperation, surgical site complication, and transfusion rates in implant-reconstructed patients Characteristic
Unilateral mastectomy (n = 10,681)
Bilateral mastectomy (n = 6592)
Adjusted OR (95 % CI)a
p value for difference
Overall complications, n (%)
944 (8.8)
667 (10.1)
1.20 (1.08, 1.33)
\0.001
Return to operating room, n (%)
715 (6.7)
490 (7.4)
1.15 (1.01, 1.29)
0.029
Surgical site complications, n (%)
449 (4.2)
300 (4.6)
1.15 (0.99, 1.34)
0.070 0.093
Surgical site infection
351 (3.3)
238 (3.6)
1.16 (0.98, 1.37)
Wound disruption
60 (0.6)
39 (0.6)
1.10 (0.73, 1.66)
0.66
Prosthesis failure
65 (0.6)
47 (0.7)
1.31 (0.90, 1.93)
0.16
Transfusion, n (%)
34 (0.3)
52 (0.8)
2.52 (1.62, 3.92)
\0.001
OR odds ratio, CI confidence interval a
Odds ratios adjusted for age, race, smoking, diabetes, chronic obstructive pulmonary disease, and hypertension
TABLE 6 Reoperation, surgical site complication, and transfusion rates in autologous-reconstructed patients Adjusted OR (95 % CI)a
p value for difference
136 (21.2)
1.60 (1.28, 1.99)
\0.001
72 (11.2)
1.26 (0.95, 1.67)
0.115
174 (6.7)
53 (8.3)
1.33 (0.96, 1.85)
0.085
111 (4.3)
33 (5.2)
1.32 (0.88, 1.97)
0.18
Wound disruption
18 (0.7)
10 (1.6)
2.51 (1.20, 5.25)
0.015
Flap failure
61 (2.4)
16 (2.5)
1.08 (0.62, 1.90)
0.79
92 (3.6)
64 (10.0)
3.06 (2.18, 4.30)
\0.001
Characteristic
Unilateral mastectomy (n = 2587)
Bilateral mastectomy (n = 641)
Overall complications, n (%)
381 (14.7)
Return to operating room, n (%)
239 (9.2)
Surgical site complications, n (%) Surgical site infection
Transfusion, n (%) a
ORs adjusted for age, race, smoking, diabetes, chronic obstructive pulmonary disease, and hypertension
women who undergo BM also elect immediate breast reconstruction,18–20 which increases the perioperative complication rate relative to mastectomy alone.32 As such, these results are not necessarily applicable to the majority of women considering CPM. The current investigation demonstrated an increased rate of reoperation within the 30-day postoperative period in
women undergoing BM versus UM with implant-based reconstruction. Early reoperation after mastectomy with implant-based reconstruction is the most definitive indicator of serious postoperative complications, as reoperation is generally related to hematoma, significant infection, or mastectomy flap necrosis.25 Although this risk is only marginally increased, reoperation may result in immediate
The Effect of Contralateral Prophylactic Mastectomy on Perioperative Complications in Women…
and delayed morbidity and confers significant additional cost.33 It has also been shown to decrease long-term satisfaction with the reconstructive outcome.34 Among patients undergoing autologous reconstruction, CPM was associated with a significantly increased rate of wound disruption. It is unclear from the NSQIP whether this increase was related to the donor site, therapeutic mastectomy, or CPM. Although the increase was not reflected in a statistically significant change in the reoperation rate in this group, wound disruption may result in a need for further surgery as well as greater patient pain, inconvenience, and cost. Additionally, wounds that heal by secondary intention generally result in unfavorable scarring. BM patients required postoperative transfusion more often than UM patients regardless of the type of reconstruction. This finding corroborates previous reports on autologous reconstruction.27,35 Single-institution studies tend to be underpowered to detect differences in transfusion after reconstruction. This is an important finding because transfusion may be seen as a surrogate for bleeding complications and may result in higher rates of health care– associated infection.36 There are also risks of transfusion reactions, such as transfusion-related acute lung injury, that may result in substantial morbidity. Interestingly, there was only a trend toward increased SSI in the CPM group. While this indicates that CPM surgery does not result in markedly increased infection rates, many infections may be underreported to NSQIP if they occur beyond the 30 days or are treated only in an outpatient setting.37 Likewise, there was no significant increase in prosthesis failure in the CPM cohort. This counterintuitive finding may reflect the fact that prosthesis failure is not clearly defined in the NSQIP data dictionary and thus may be underreported. Additionally, implant loss often occurs after 30 days. These factors might diminish the power to indicate a difference between groups despite such a large data set. Medical complications were infrequent and unaffected by surgery type. Postoperative medical complications in breast cancer-related surgery tend to be related to comorbidities, and patients in this study had few.38 Although the length of surgery was significantly longer in the BM group, the stress imposed by the additional surgery was likely not sufficient to cause additional medical complications in this relatively healthy cohort. One study has shown a higher rate of arrhythmias and lower extremity DVTs in autologous BM versus UM.27 Arrhythmias are not reported to the NSQIP, and the differences in the reported rate of DVT may reflect differences in hospital practices for screening and detection of DVT. There are several limitations to the current study. The overall complication rates for both UM and BM in this study were lower than other reports.24–27 It is possible that
3479
some complications are underreported to the NSQIP, which may have reduced the power of this study to detect certain differences between groups. Complications for which documentation of the diagnosis may only occur in an outpatient chart not requiring readmission or additional operative procedures might not be reflected in the NSQIP. In addition, the NSQIP data set is limited to 30-day outcomes, and surgical site complications after mastectomy with reconstruction may occur after this time has elapsed; most studies that have compared complications between UM and BM followed patients well outside the 30-day window.24,25 Additionally, there are complications that are relevant to breast reconstruction, such as seroma and capsular contracture, that are not captured by the NSQIP. Finally, the NSQIP database does not allow for laterality of complication to be determined. Therefore, it is not possible to know whether patients experience a disproportionate number of complications in the prophylactic breast versus the therapeutic breast, or whether patients experienced bilateral complications. Finally, because women who undergo CPM tend to be healthier than those who do not, there may be additional unmeasured health-related confounders for which we could not adjust. This may have masked or attenuated differences in complications between groups. These findings have important implications for patient decision making. CPM has undergone explosive growth over the past decade. The decision to undergo CPM is complex and in most cases occurs in the context of a recent cancer diagnosis, thus warranting expeditious decision making. Women considering CPM should be reassured that medical complications are infrequent, regardless of the type of mastectomy they choose. However, they should pause to consider the added risks of CPM that we describe herein. Although some women considering CPM may be reassured by these data, others may find them troubling. This information will help physicians inform patients about the true risks of additional surgery. They will also aid in the generation of support tools to guide women through these challenging decisions. DISCLOSURE
The authors declare no conflict of interest.
REFERENCES 1. Tuttle TM, Habermann EB, Grund EH, Morris TJ, Virnig BA. Increasing use of contralateral prophylactic mastectomy for breast cancer patients: a trend toward more aggressive surgical treatment. J Clin Oncol. 2007;25:5203–9. 2. Tuttle TM, Jarosek S, Habermann EB, et al. Increasing rates of contralateral prophylactic mastectomy among patients with ductal carcinoma in situ. J Clin Oncol. 2009;27:1362–7. 3. Jones NB, Wilson J, Kotur L, Stephens J, Farrar WB, Agnese DM. Contralateral prophylactic mastectomy for unilateral breast
3480
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14. 15. 16.
17.
18.
19.
20.
21.
cancer: an increasing trend at a single institution. Ann Surg Oncol. 2009;16:2691–6. Pesce CE, Liederbach E, Czechura T, Winchester DJ, Yao K. Changing surgical trends in young patients with early stage breast cancer, 2003 to 2010: a report from the National Cancer Data Base. J Am Coll Surg. 2014;219:19–28. Kurian AW, Lichtensztajn DY, Keegan TH, Nelson DO, Clarke CA, Gomez SL. Use of and mortality after bilateral mastectomy compared with other surgical treatments for breast cancer in California, 1998–2011. JAMA. 2014;312:902–14. Bedrosian I, Hu CY, Chang GJ. Population-based study of contralateral prophylactic mastectomy and survival outcomes of breast cancer patients. J Natl Cancer Inst. 2010;102:401–9. Portschy PR, Kuntz KM, Tuttle TM. Survival outcomes after contralateral prophylactic mastectomy: a decision analysis. J Natl Cancer Inst. 2014;106. Yao K, Winchester DJ, Czechura T, Huo D. Contralateral prophylactic mastectomy and survival: report from the National Cancer Data Base, 1998–2002. Breast Cancer Res Treat. 2013;142:465–76. Lostumbo L, Carbine NE, Wallace J. Prophylactic mastectomy for the prevention of breast cancer. Cochrane Database Syst Rev. 2010;10:CD002748. Brewster AM, Parker PA. Current knowledge on contralateral prophylactic mastectomy among women with sporadic breast cancer. Oncologist. 2011;16:935–41. Murphy JA, Milner TD, O’Donoghue JM. Contralateral risk-reducing mastectomy in sporadic breast cancer. Lancet Oncol. 2013;14:e262–9. Nichols HB, Berrington de Gonza´lez A, Lacey JV Jr, Rosenberg PS, Anderson WF. Declining incidence of contralateral breast cancer in the United States from 1975 to 2006. J Clin Oncol. 2011;29:1564–9. Gao X, Fisher SG, Emami B. Risk of second primary cancer in the contralateral breast in women treated for early-stage breast cancer: a population-based study. Int J Radiat Oncol Biol Phys. 2003;56:1038–45. Katz SJ, Morrow M. Contralateral prophylactic mastectomy for breast cancer: addressing peace of mind. JAMA. 2013;310:793–4. Newman LA. Contralateral prophylactic mastectomy: is it a reasonable option? JAMA. 2014;312:895–7. Rosenberg SM, Tracy MS, Meyer ME, et al. Perceptions, knowledge, and satisfaction with contralateral prophylactic mastectomy among young women with breast cancer: a crosssectional survey. Ann Intern Med. 2013;159:373–81. Osman F, Saleh F, Jackson TD, Corrigan MA, Cil T. Increased postoperative complications in bilateral mastectomy patients compared to unilateral mastectomy: an analysis of the NSQIP database. Ann Surg Oncol. 2013;20:3212–7. Cemal Y, Albornoz CR, Disa JJ, et al. A paradigm shift in US breast reconstruction: part 2. The influence of changing mastectomy patterns on reconstructive rate and method. Plast Reconstr Surg. 2013;131:320e–6e. Jagsi R, Jiang J, Momoh AO, et al. Trends and variation in use of breast reconstruction in patients with breast cancer undergoing mastectomy in the United States. J Clin Oncol. 2014;32:919–26. Arrington AK, Jarosek SL, Virnig BA, Habermann EB, Tuttle TM. Patient and surgeon characteristics associated with increased use of contralateral prophylactic mastectomy in patients with breast cancer. Ann Surg Oncol. 2009;16:2697–704. Shiloach M, Frencher SK Jr, Steeger JE, et al. Toward robust information: data quality and inter-rater reliability in the American College of Surgeons National Surgical Quality Improvement Program. J Am Coll Surg. 2010;210:6–16.
A. K. Silva et al. 22. Daley J, Forbes MG, Young GJ, et al. Validating risk-adjusted surgical outcomes: site visit assessment of process and structure. National VA Surgical Risk Study. J Am Coll Surg. 1997;185:341–51. 23. Zion SM, Slezak JM, Sellers TA, et al. Reoperations after prophylactic mastectomy with or without implant reconstruction. Cancer. 2003;98:2152–60. 24. Crosby MA, Garvey PB, Selber JC, et al. Reconstructive outcomes in patients undergoing contralateral prophylactic mastectomy. Plast Reconstr Surg. 2011;128:1025–33. 25. Miller ME, Czechura T, Martz B, et al. Operative risks associated with contralateral prophylactic mastectomy: a single institution experience. Ann Surg Oncol. 2013;20:4113–20. 26. Antony AK, McCarthy C, Disa JJ, Mehrara BJ. Bilateral implant breast reconstruction: outcomes, predictors, and matched cohort analysis in 730 2-stage breast reconstructions over 10 years. Ann Plast Surg. 2014;72:625–30. 27. Lin IC, Nelson JA, Wu LC, Kovach SJ 3rd, Serletti JM. Assessing surgical and medical complications in bilateral abdomen-based free flap breast reconstructions compared with unilateral free flap breast reconstructions. Ann Plast Surg. 2014. doi:10.1097/SAP.0000000000000343. 28. Craft RO, Colakoglu S, Curtis MS, et al. Patient satisfaction in unilateral and bilateral breast reconstruction. Plast Reconstr Surg. 2011;127:1417–24. 29. Fischer JP, Wes AM, Tuggle CT 3rd, Serletti JM, Wu LC. Risk analysis of early implant loss after immediate breast reconstruction: a review of 14,585 patients. J Am Coll Surg. 2013;217:983–90. 30. Fischer JP, Wes AM, Tuggle CT, Serletti JM, Wu LC. Risk analysis and stratification of surgical morbidity after immediate breast reconstruction. J Am Coll Surg. 2013;217:780–7. 31. Gart MS, Smetona JT, Hanwright PJ, et al. Autologous options for postmastectomy breast reconstruction: a comparison of outcomes based on the American College of Surgeons National Surgical Quality Improvement Program. J Am Coll Surg. 2013;216:229–38. 32. Fischer JP, Tuggle CT, Au A, Kovach SJ. A 30-day risk assessment of mastectomy alone compared to immediate breast reconstruction (IBR). J Plast Surg Hand Surg. 2014;48:209–15. 33. Ashraf M, Biswas J, Gupta S, Alam N. Determinants of wound infections for breast procedures: assessment of the risk of wound infection posed by an invasive procedure for subsequent operation. J Plast Surg Hand Surg. 2009;7:543–6. 34. Boughey JC, Hoskin TL, Hartmann LC, et al. Impact of reconstruction and reoperation on long-term patient-reported satisfaction after contralateral prophylactic mastectomy. Ann Surg Oncol. 2015;22:401–8. 35. Ting J, Rozen WM, Le Roux CM, Ashton MW, Garcia-Tutor E. Predictors of blood transfusion in deep inferior epigastric artery perforator flap breast reconstruction. J Reconstr Microsurg. 2011;27:233–8. 36. Rohde JM, Dimcheff DE, Blumberg N, et al. Health care–associated infection after red blood cell transfusion: a systematic review and meta-analysis. JAMA. 2014;311:1317–26. 37. Peled AW, Stover AC, Foster RD, McGrath MH, Hwang ES. Long-term reconstructive outcomes after expander-implant breast reconstruction with serious infectious or wound-healing complications. Ann Plast Surg. 2012;68:369–73. 38. Neumayer L, Schifftner TL, Henderson WG, Khuri SF, El-Tamer M. Breast cancer surgery in Veterans Affairs and selected university medical centers: results of the patient safety in surgery study. J Am Coll Surg. 2007;204:1235–41.
ORIGINAL ARTICLE – BREAST ONCOLOGY
The Effect of Contralateral Prophylactic Mastectomy on Perioperative Complications in Women Undergoing Immediate Breast Reconstruction: A NSQIP Analysis Amanda K. Silva, MD1, Brittany Lapin, MPH2, Katharine A. Yao, MD, FACS3, David H. Song, MD, MBA, FACS1, and Mark Sisco, MD, FACS4 Section of Plastic and Reconstructive Surgery, University of Chicago Hospitals, Chicago, IL; 2Center for Biomedical Research Informatics, NorthShore University HealthSystem Research Institute, Evanston, IL; 3Division of Surgical Oncology, NorthShore University HealthSystem, Evanston, IL; 4Division of Plastic and Reconstructive Surgery, NorthShore University HealthSystem, Evanston, IL
1
ABSTRACT Background. Women with breast cancer are increasingly choosing to undergo contralateral prophylactic mastectomy (CPM) despite questionable survival benefit and limited data on added risks. Little is known about differences in perioperative complications between women who undergo bilateral mastectomy (BM) versus unilateral mastectomy (UM) with reconstruction. Methods. The American College of Surgeons National Surgery Quality Improvement Program Participant Use Files (2005–2013) were used to identify women with unilateral breast cancer who underwent UM or BM with reconstruction. Adjusted 30-day complications were compared between UM and BM groups using logistic regression models. Results. A total of 20,501 patients were identified, of whom 35.3 % underwent BM. Of these, 84.3 % had implant reconstruction and 15.7 % had autologous reconstruction. For all women, BM was associated with longer hospital stays (adjusted odds ratio [aOR] 1.98–2.09, p \ 0.001) and a higher transfusion rate than UM (aOR 2.52–3.06, p \ 0.001). BM with implant reconstruction was associated with a modestly increased reoperation rate
This study was presented as an oral presentation at the ASCO Breast Cancer Symposium, September 4, 2014, San Francisco, CA. Society of Surgical Oncology 2015 First Received: 24 January 2015; Published Online: 22 May 2015 M. Sisco, MD, FACS e-mail: [email protected]
(aOR 1.15, p = 0.029). BM with autologous reconstruction was associated with a higher wound disruption rate (aOR 2.51, p = 0.015). Surgical site infections, prosthesis failure, and medical complications occurred at similar rates in UM and BM groups. Conclusions. CPM is associated with significant increases in some, but not all, surgical site complications. CPM does not increase the likelihood of medical complications, which are generally infrequent.
Women with unilateral breast cancer are choosing to undergo bilateral mastectomy (BM) at dramatically increasing rates.1–3 Data from the National Cancer Data Base indicates that from 2003 to 2010, the proportion of women with American Joint Committee on Cancer stage 0 to II unilateral breast cancer undergoing BM has more than doubled, from 4.3 to 9.7 %.4 This increase has been especially pronounced in women under the age of 45, among whom the rate has climbed to 24.1–33 %.4,5 This suggests that up to 25,000 women may be undergoing contralateral prophylactic mastectomy (CPM) each year. This increase is especially striking because few women undergoing CPM have been shown to derive a survival benefit.5–9 In BRCA gene-negative patients, the likelihood of developing a contralateral breast cancer is low and is outweighed by the risk of systemic disease.9–13 The fact that many women are electing to undergo additional surgery with limited efficacy data has led to recognition of an intrinsic conflict between the desire to practice evidencebased medicine and to respect patient wishes.14,15 Rosenberg et al. elucidated reasons women pursue CPM.16 This study did not address the extent to which
The Effect of Contralateral Prophylactic Mastectomy on Perioperative Complications in Women…
surgical risk was considered; however, one third of patients stated that the procedures required were higher than expected. Although it may be intuitive to providers that CPM is associated with higher complication rates, patients do not necessarily perceive added risk. An understanding of the risks and benefits of surgery is intrinsic to informed consent. Yet which risks increase after BM with reconstruction, and by how much, have not been studied in a large multi-institution cohort. A previous study using the National Surgery Quality Improvement Program database showed an increased risk of complications from BM among patients who did not undergo reconstruction.17 However, over 75 % of women who undergo BM also elect immediate reconstruction.18–20 For women with unilateral breast cancer, decision making is hampered by a lack of multi-institution data describing the additive risks of CPM. Additionally, decision aids that address risks and benefits of CPM require data that are accurate and broadly applicable. The American College of Surgeons National Surgery Quality Improvement Program (NSQIP) database is a nationally validated, risk-adjusted surgical outcomes database involving over 500 hospitals that is designed to measure and improve the quality of surgical care. The data collection process has been shown to be both reliable and valid.21,22 This database provides the power, breadth, and consistency necessary to accurately assess the incidence of uncommon complications across patient populations. In order to better inform medical decision making, this study sought to assess differences in perioperative complications between BM versus unilateral mastectomy (UM) with reconstruction among women with unilateral breast cancer. METHODS Subjects Data from the American College of Surgeons NSQIP Participant Use Files from 2005 to 2013 were accessed. Patient demographics, preoperative, and 30-day postoperative data are collected by a trained surgical clinical reviewer via medical chart review and entered in a Insurance Portability and Accountability Act-compliant, secure, Web-based platform from a random sample of patient charts. Because data are deidentified, institutional review board approval was not necessary for this study. Women aged C18 undergoing either UM or BM were identified using Current Procedural Terminology (CPT) codes. Yearly variations in CPT codes were taken into consideration. Mastectomy codes included those for simple mastectomy (19180/19303), skin-sparing mastectomy (19182/19304), or modified radical mastectomy (19240/
3475
19307). From this cohort, immediate breast reconstruction patients were identified using CPT codes and were dichotomized into two groups on the basis of reconstruction type: implant-based reconstruction included immediate implant (19340) and tissue expander placement (19357); autologous reconstruction included the latissimus dorsi flap with or without an implant (19361), pedicle transverse rectus abdominis myocutaneous flaps (19367, 19368, 19369), and free tissue transfer (19364). Only patients whose associated diagnosis was breast cancer (174.0– 174.9) or ductal carcinoma-in situ (233.0), or who had a personal history of malignant breast neoplasm (V10.3) were included. To limit the cohort to women with unilateral breast cancer, those who had bilateral sentinel lymph node biopsies or bilateral modified radical mastectomies were excluded. Patients who underwent different reconstructive procedures on either side were also excluded. Outcomes Examined 30-day outcomes included superficial (skin or subcutaneous tissue) and deep (fascial and muscle) surgical site infections, wound disruption, pneumonia, pulmonary embolism (PE), renal insufficiency, acute renal failure, urinary tract infection, stroke, cardiac arrest, myocardial infarction, deep venous thrombosis (DVT), sepsis, septic shock, transfusions within 72 h of surgery, prosthesis or flap failure, procedure length, length of hospital stay (C2 days for implant-based and C4 days for autologous reconstruction), and reoperation. Readmission rates were not included, as these were not introduced to NSQIP until 2011. In order to simplify and increase the power of the analysis, certain outcomes were collapsed as follows: superficial and deep surgical site infection were collapsed into surgical site infection (SSI); PE and DVT into venous thromboembolism; cardiac arrest and myocardial infarction into myocardial infarction or cardiac arrest; sepsis and septic shock into sepsis; and renal insufficiency and acute renal failure into renal insufficiency or failure. Statistical Analysis Patient age, race, comorbidities, and reconstruction type were compared between unilateral and bilateral cohorts. Frequency counts and percentages were calculated for categorical variables, and groups were compared by the Chi square test or Fisher’s exact test for small cell counts. Median with interquartile range was calculated for continuous variables, and groups were compared by Student’s t test. Unadjusted complication rates are shown separately by reconstruction type, implant and autologous, for uni-
3476
A. K. Silva et al.
lateral versus bilateral women. Multivariable logistic regression models were constructed to control for demographic and comorbidity characteristics that differed significantly between unilateral and bilateral women. Models were adjusted for age, race, smoking, diabetes, chronic obstructive pulmonary disease, hypertension, and the receipt of neoadjuvant chemotherapy. Adjusted odds ratios (aORs), 95 % confidence intervals, and p values are presented for BM (vs. UM). p \ 0.05 was considered statistically significant. SAS statistical software, version 9.3 (SAS Institute Inc, Cary, NC), was used for data analyses. RESULTS A total of 20,501 patients were included in the analysis; 7233 (35.3 %) women underwent BM and 13,268 (64.7 %) underwent UM (Table 1). Women who underwent BM were younger and more likely to be white. They were less likely to have hypertension or diabetes. However, they were more likely to smoke and to have undergone neoadjuvant chemotherapy within 30 days of surgery. Table 2 presents reconstruction types performed in each cohort. Most patients in both groups underwent implantbased reconstruction, but women who underwent UM were more likely to have undergone autologous reconstruction than those who underwent BM (19.5 % of UM vs. 8.9 % of BM, p \ 0.001). Tables 3 and 4 depict surgery duration and hospital length of stay as well as medical complications. Unsurprisingly, patients who underwent BM had longer surgery
times and hospital stays. The adjusted rate of total and individual medical complications was similar in both groups. Individual medical complications were infrequent, each occurring in fewer than 1 % of patients, with the incidence of myocardial infarction, cardiac arrest, and stroke less than 0.1 %. There were no mortalities in the UM group and two mortalities in the BM group (p value for difference = 0.12). Tables 5 and 6 depict adjusted surgical site complication and reoperation rates. Implant-based BM was associated with higher rates of transfusion (aOR 2.52, p \ 0.001) and slightly more frequent reoperations (aOR 1.15, p = 0.029) (Table 5). Patients who underwent autologous BM were similarly more likely to require transfusion (aOR3.06, p \ 0.001) and had higher wound disruption rates (aOR 2.51, p = 0.015) (Table 6). There were no significant differences in rates of SSI, prosthesis failure, or flap failure between the UM and BM groups. DISCUSSION This report from the American College of Surgeons NSQIP is the largest comparative analysis of complications from CPM with immediate reconstruction to date. BM was not associated with more medical complications, which were infrequent in both cohorts. However, all women who underwent BM were more likely to require a transfusion and had longer hospital stays. Of the 84.3 % of women who underwent implant-based reconstruction, those who had BM were slightly more likely to experience an
TABLE 1 Characteristics of mastectomy patients by laterality (n = 20,501) Characteristic
Unilateral mastectomy
Bilateral mastectomy
Total, n (%)
13,268 (64.7)
7233 (35.3) \0.001
Race, n (%) White
9828 (74.1)
Black
1128 (8.5)
453 (6.3)
706 (5.3)
381 (5.3)
590 (4.5)
251 (3.5)
1016 (7.7)
523 (7.2)
Hispanic Other Unknown Age, median (q1, q3)
p value for difference
52 (46, 60)
5625 (77.8)
49 (43, 56)
\0.001
Comorbidities/risk factors, n (%) Obesity (BMI C30 kg/m2) Smoking
3679 (27.9) 1636 (12.3)
1963 (26.9) 978 (13.5)
Diabetes
722 (5.4)
295 (4.1)
COPD
111 (0.8)
44 (0.6)
CHF or MI Hypertension Stroke Chemotherapy within 30 days
0.32 0.015 \0.001 0.071
5 (0.04)
2 (0.03)
0.99
3552 (26.8)
1569 (21.7)
47 (0.5)
24 (0.6)
0.77
360 (3.9)
300 (6.9)
\0.001
\0.001
BMI body mass index, COPD chronic obstructive pulmonary disease, CHF congestive heart failure, MI myocardial infarction
The Effect of Contralateral Prophylactic Mastectomy on Perioperative Complications in Women…
3477
TABLE 2 Reconstruction type by laterality Characteristic
Unilateral mastectomy (n = 13,268)
Implant-based reconstruction, n (%)
Bilateral mastectomy (n = 7233)
10,681 (80.5)
6592 (91.1)
Tissue expander
9124 (68.8)
5709 (78.9)
Immediate implant
1557 (11.7)
883 (12.2)
2587 (19.5)
641 (8.9)
Autologous reconstruction, n (%) Latissimus flap (with or without TE or implant) Pedicle TRAM Free flap
784 (5.9)
157 (2.2)
1176 (8.9)
233 (3.2)
627 (4.7)
251 (3.5)
TRAM transverse rectus abdominis myocutaneous
TABLE 3 Length of stay, medical complication rates, and multivariable regression models in implant-reconstructed patients Characteristic
Unilateral mastectomy (n = 10,681) 173 (133, 222)b
Bilateral mastectomy (n = 6592) 226 (176, 281)b
Adjusted OR (95 % CI)a
p value for difference
–
\0.001
4460 (41.8)
3957 (60.0)
2.09 (1.96, 2.22)
\0.001
110 (1.0)
70 (1.1)
1.06 (0.78, 1.44)
0.72
Pneumonia
7 (0.1)
5 (0.1)
1.35 (0.51, 3.57)
0.55
Renal insufficiency or failure
3 (0.03)
2 (0.03)
0.89 (0.25, 3.10)
0.85
Length of surgery (min), median (q1, q3) Length of hospital stay C2 days, n (%) Medical complications, n (%)
Sepsis
46 (0.4)
30 (0.5)
1.04 (0.66, 1.64)
0.89
Urinary tract infection Venous thromboembolism (DVT ? PE)
25 (0.2) 32 (0.3)
11 (0.2) 26 (0.4)
0.74 (0.38, 1.44) 1.41 (0.85, 2.34)
0.37 0.18
OR odds ratio, CI confidence interval, DVT deep venous thrombosis, PE pulmonary embolism a
ORs adjusted for age, race, smoking, diabetes, chronic obstructive pulmonary disease, and hypertension
b
Data presented as median (Q1, Q3)
unanticipated return to the operating room than those who had UM. Women who underwent BM with autologous reconstruction, meanwhile, had significantly more wound disruptions. Several single-institution retrospective studies have quantified the risks of complications after BM with reconstruction.23–27 However, these studies are difficult to interpret as a whole as a result of variations in the treatment groups, study methodologies, and data interpretation. Some reports include patients without cancer who underwent bilateral risk-reducing mastectomy, and others include patients who underwent delayed reconstruction.26–28 Also, some describe complications on a per-breast rather than per-patient basis, which is potentially difficult for patients and their providers to interpret.24,26,28 Two series examined complications after BM with reconstruction yet did not directly compare these rates to women undergoing UM.26,28 Finally, authors present differing conclusions based on similar findings, which illustrates the variation with which experienced researchers interpret data. One report from the Memorial Sloan Kettering Cancer Center indicated an overall complication rate after BM that
compared favorably to historical institutional data on UM, leading the authors to conclude that ‘‘bilateral implant reconstruction does not appear to pose additive risk.’’26 In contrast, a study from MD Anderson Cancer Center cautioned to ‘‘carefully consider the oncologic benefits of a CPM in light of the risk of increased surgical morbidity.’’24 Several previous studies of the NSQIP have alluded to the morbidity of BM. These studies were designed to identify predictors of complications after mastectomy. They did not directly assess those attributable to CPM.29 For example, Fischer et al. developed a model to predict complications after immediate breast reconstruction; however, laterality was not found to be predictive of complications and was not included in the model.30 Likewise, Gart et al. did not identify bilateral surgery as a risk factor for complications in women undergoing mastectomy.31 Of note, these studies included patients without cancer who underwent bilateral prophylactic mastectomy. Another NSQIP study has reported increased 30-day complications after BM without reconstruction.17 This is important information for women considering CPM who do not desire breast reconstruction. However, over 75 % of
3478
A. K. Silva et al.
TABLE 4 Length of stay, medical complication rates, and multivariable regression models in autologous-reconstructed patients Characteristic Length of surgery (min), median (q1, q3) Length of hospital stay C4 days, n (%) Medical complications, n (%) Pneumonia Renal insufficiency or failure
Bilateral mastectomy (n = 641)
Adjusted OR (95 % CI)a
p value for difference
477 (377, 626)b
–
\0.001
1308 (50.6)
431 (67.2)
1.98 (1.65, 2.38)
\0.001
70 (2.7)
17 (2.7)
0.99 (0.58, 1.72)
0.99
13 (0.5)
1 (0.2)
0.37 (0.08, 1.74)
0.21
Unilateral mastectomy (n = 2587) 326 (237, 439)b
2 (0.1)
1 (0.2)
3.00 (0.64, 14.13)
0.17
Sepsis
20 (0.8)
6 (0.9)
1.19 (0.51, 2.79)
0.68
Urinary tract infection Venous thromboembolism (DVT ? PE)
21 (0.8) 22 (0.9)
3 (0.5) 8 (1.3)
0.67 (0.23, 1.94) 1.81 (0.84, 3.91)
0.46 0.13
OR odds ratio, CI confidence interval, DVT deep venous thrombosis, PE pulmonary embolism a
ORs adjusted for age, race, smoking, diabetes, chronic obstructive pulmonary disease, and hypertension
b
Data presented as median (Q1, Q3)
TABLE 5 Reoperation, surgical site complication, and transfusion rates in implant-reconstructed patients Characteristic
Unilateral mastectomy (n = 10,681)
Bilateral mastectomy (n = 6592)
Adjusted OR (95 % CI)a
p value for difference
Overall complications, n (%)
944 (8.8)
667 (10.1)
1.20 (1.08, 1.33)
\0.001
Return to operating room, n (%)
715 (6.7)
490 (7.4)
1.15 (1.01, 1.29)
0.029
Surgical site complications, n (%)
449 (4.2)
300 (4.6)
1.15 (0.99, 1.34)
0.070 0.093
Surgical site infection
351 (3.3)
238 (3.6)
1.16 (0.98, 1.37)
Wound disruption
60 (0.6)
39 (0.6)
1.10 (0.73, 1.66)
0.66
Prosthesis failure
65 (0.6)
47 (0.7)
1.31 (0.90, 1.93)
0.16
Transfusion, n (%)
34 (0.3)
52 (0.8)
2.52 (1.62, 3.92)
\0.001
OR odds ratio, CI confidence interval a
Odds ratios adjusted for age, race, smoking, diabetes, chronic obstructive pulmonary disease, and hypertension
TABLE 6 Reoperation, surgical site complication, and transfusion rates in autologous-reconstructed patients Adjusted OR (95 % CI)a
p value for difference
136 (21.2)
1.60 (1.28, 1.99)
\0.001
72 (11.2)
1.26 (0.95, 1.67)
0.115
174 (6.7)
53 (8.3)
1.33 (0.96, 1.85)
0.085
111 (4.3)
33 (5.2)
1.32 (0.88, 1.97)
0.18
Wound disruption
18 (0.7)
10 (1.6)
2.51 (1.20, 5.25)
0.015
Flap failure
61 (2.4)
16 (2.5)
1.08 (0.62, 1.90)
0.79
92 (3.6)
64 (10.0)
3.06 (2.18, 4.30)
\0.001
Characteristic
Unilateral mastectomy (n = 2587)
Bilateral mastectomy (n = 641)
Overall complications, n (%)
381 (14.7)
Return to operating room, n (%)
239 (9.2)
Surgical site complications, n (%) Surgical site infection
Transfusion, n (%) a
ORs adjusted for age, race, smoking, diabetes, chronic obstructive pulmonary disease, and hypertension
women who undergo BM also elect immediate breast reconstruction,18–20 which increases the perioperative complication rate relative to mastectomy alone.32 As such, these results are not necessarily applicable to the majority of women considering CPM. The current investigation demonstrated an increased rate of reoperation within the 30-day postoperative period in
women undergoing BM versus UM with implant-based reconstruction. Early reoperation after mastectomy with implant-based reconstruction is the most definitive indicator of serious postoperative complications, as reoperation is generally related to hematoma, significant infection, or mastectomy flap necrosis.25 Although this risk is only marginally increased, reoperation may result in immediate
The Effect of Contralateral Prophylactic Mastectomy on Perioperative Complications in Women…
and delayed morbidity and confers significant additional cost.33 It has also been shown to decrease long-term satisfaction with the reconstructive outcome.34 Among patients undergoing autologous reconstruction, CPM was associated with a significantly increased rate of wound disruption. It is unclear from the NSQIP whether this increase was related to the donor site, therapeutic mastectomy, or CPM. Although the increase was not reflected in a statistically significant change in the reoperation rate in this group, wound disruption may result in a need for further surgery as well as greater patient pain, inconvenience, and cost. Additionally, wounds that heal by secondary intention generally result in unfavorable scarring. BM patients required postoperative transfusion more often than UM patients regardless of the type of reconstruction. This finding corroborates previous reports on autologous reconstruction.27,35 Single-institution studies tend to be underpowered to detect differences in transfusion after reconstruction. This is an important finding because transfusion may be seen as a surrogate for bleeding complications and may result in higher rates of health care– associated infection.36 There are also risks of transfusion reactions, such as transfusion-related acute lung injury, that may result in substantial morbidity. Interestingly, there was only a trend toward increased SSI in the CPM group. While this indicates that CPM surgery does not result in markedly increased infection rates, many infections may be underreported to NSQIP if they occur beyond the 30 days or are treated only in an outpatient setting.37 Likewise, there was no significant increase in prosthesis failure in the CPM cohort. This counterintuitive finding may reflect the fact that prosthesis failure is not clearly defined in the NSQIP data dictionary and thus may be underreported. Additionally, implant loss often occurs after 30 days. These factors might diminish the power to indicate a difference between groups despite such a large data set. Medical complications were infrequent and unaffected by surgery type. Postoperative medical complications in breast cancer-related surgery tend to be related to comorbidities, and patients in this study had few.38 Although the length of surgery was significantly longer in the BM group, the stress imposed by the additional surgery was likely not sufficient to cause additional medical complications in this relatively healthy cohort. One study has shown a higher rate of arrhythmias and lower extremity DVTs in autologous BM versus UM.27 Arrhythmias are not reported to the NSQIP, and the differences in the reported rate of DVT may reflect differences in hospital practices for screening and detection of DVT. There are several limitations to the current study. The overall complication rates for both UM and BM in this study were lower than other reports.24–27 It is possible that
3479
some complications are underreported to the NSQIP, which may have reduced the power of this study to detect certain differences between groups. Complications for which documentation of the diagnosis may only occur in an outpatient chart not requiring readmission or additional operative procedures might not be reflected in the NSQIP. In addition, the NSQIP data set is limited to 30-day outcomes, and surgical site complications after mastectomy with reconstruction may occur after this time has elapsed; most studies that have compared complications between UM and BM followed patients well outside the 30-day window.24,25 Additionally, there are complications that are relevant to breast reconstruction, such as seroma and capsular contracture, that are not captured by the NSQIP. Finally, the NSQIP database does not allow for laterality of complication to be determined. Therefore, it is not possible to know whether patients experience a disproportionate number of complications in the prophylactic breast versus the therapeutic breast, or whether patients experienced bilateral complications. Finally, because women who undergo CPM tend to be healthier than those who do not, there may be additional unmeasured health-related confounders for which we could not adjust. This may have masked or attenuated differences in complications between groups. These findings have important implications for patient decision making. CPM has undergone explosive growth over the past decade. The decision to undergo CPM is complex and in most cases occurs in the context of a recent cancer diagnosis, thus warranting expeditious decision making. Women considering CPM should be reassured that medical complications are infrequent, regardless of the type of mastectomy they choose. However, they should pause to consider the added risks of CPM that we describe herein. Although some women considering CPM may be reassured by these data, others may find them troubling. This information will help physicians inform patients about the true risks of additional surgery. They will also aid in the generation of support tools to guide women through these challenging decisions. DISCLOSURE
The authors declare no conflict of interest.
REFERENCES 1. Tuttle TM, Habermann EB, Grund EH, Morris TJ, Virnig BA. Increasing use of contralateral prophylactic mastectomy for breast cancer patients: a trend toward more aggressive surgical treatment. J Clin Oncol. 2007;25:5203–9. 2. Tuttle TM, Jarosek S, Habermann EB, et al. Increasing rates of contralateral prophylactic mastectomy among patients with ductal carcinoma in situ. J Clin Oncol. 2009;27:1362–7. 3. Jones NB, Wilson J, Kotur L, Stephens J, Farrar WB, Agnese DM. Contralateral prophylactic mastectomy for unilateral breast
3480
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14. 15. 16.
17.
18.
19.
20.
21.
cancer: an increasing trend at a single institution. Ann Surg Oncol. 2009;16:2691–6. Pesce CE, Liederbach E, Czechura T, Winchester DJ, Yao K. Changing surgical trends in young patients with early stage breast cancer, 2003 to 2010: a report from the National Cancer Data Base. J Am Coll Surg. 2014;219:19–28. Kurian AW, Lichtensztajn DY, Keegan TH, Nelson DO, Clarke CA, Gomez SL. Use of and mortality after bilateral mastectomy compared with other surgical treatments for breast cancer in California, 1998–2011. JAMA. 2014;312:902–14. Bedrosian I, Hu CY, Chang GJ. Population-based study of contralateral prophylactic mastectomy and survival outcomes of breast cancer patients. J Natl Cancer Inst. 2010;102:401–9. Portschy PR, Kuntz KM, Tuttle TM. Survival outcomes after contralateral prophylactic mastectomy: a decision analysis. J Natl Cancer Inst. 2014;106. Yao K, Winchester DJ, Czechura T, Huo D. Contralateral prophylactic mastectomy and survival: report from the National Cancer Data Base, 1998–2002. Breast Cancer Res Treat. 2013;142:465–76. Lostumbo L, Carbine NE, Wallace J. Prophylactic mastectomy for the prevention of breast cancer. Cochrane Database Syst Rev. 2010;10:CD002748. Brewster AM, Parker PA. Current knowledge on contralateral prophylactic mastectomy among women with sporadic breast cancer. Oncologist. 2011;16:935–41. Murphy JA, Milner TD, O’Donoghue JM. Contralateral risk-reducing mastectomy in sporadic breast cancer. Lancet Oncol. 2013;14:e262–9. Nichols HB, Berrington de Gonza´lez A, Lacey JV Jr, Rosenberg PS, Anderson WF. Declining incidence of contralateral breast cancer in the United States from 1975 to 2006. J Clin Oncol. 2011;29:1564–9. Gao X, Fisher SG, Emami B. Risk of second primary cancer in the contralateral breast in women treated for early-stage breast cancer: a population-based study. Int J Radiat Oncol Biol Phys. 2003;56:1038–45. Katz SJ, Morrow M. Contralateral prophylactic mastectomy for breast cancer: addressing peace of mind. JAMA. 2013;310:793–4. Newman LA. Contralateral prophylactic mastectomy: is it a reasonable option? JAMA. 2014;312:895–7. Rosenberg SM, Tracy MS, Meyer ME, et al. Perceptions, knowledge, and satisfaction with contralateral prophylactic mastectomy among young women with breast cancer: a crosssectional survey. Ann Intern Med. 2013;159:373–81. Osman F, Saleh F, Jackson TD, Corrigan MA, Cil T. Increased postoperative complications in bilateral mastectomy patients compared to unilateral mastectomy: an analysis of the NSQIP database. Ann Surg Oncol. 2013;20:3212–7. Cemal Y, Albornoz CR, Disa JJ, et al. A paradigm shift in US breast reconstruction: part 2. The influence of changing mastectomy patterns on reconstructive rate and method. Plast Reconstr Surg. 2013;131:320e–6e. Jagsi R, Jiang J, Momoh AO, et al. Trends and variation in use of breast reconstruction in patients with breast cancer undergoing mastectomy in the United States. J Clin Oncol. 2014;32:919–26. Arrington AK, Jarosek SL, Virnig BA, Habermann EB, Tuttle TM. Patient and surgeon characteristics associated with increased use of contralateral prophylactic mastectomy in patients with breast cancer. Ann Surg Oncol. 2009;16:2697–704. Shiloach M, Frencher SK Jr, Steeger JE, et al. Toward robust information: data quality and inter-rater reliability in the American College of Surgeons National Surgical Quality Improvement Program. J Am Coll Surg. 2010;210:6–16.
A. K. Silva et al. 22. Daley J, Forbes MG, Young GJ, et al. Validating risk-adjusted surgical outcomes: site visit assessment of process and structure. National VA Surgical Risk Study. J Am Coll Surg. 1997;185:341–51. 23. Zion SM, Slezak JM, Sellers TA, et al. Reoperations after prophylactic mastectomy with or without implant reconstruction. Cancer. 2003;98:2152–60. 24. Crosby MA, Garvey PB, Selber JC, et al. Reconstructive outcomes in patients undergoing contralateral prophylactic mastectomy. Plast Reconstr Surg. 2011;128:1025–33. 25. Miller ME, Czechura T, Martz B, et al. Operative risks associated with contralateral prophylactic mastectomy: a single institution experience. Ann Surg Oncol. 2013;20:4113–20. 26. Antony AK, McCarthy C, Disa JJ, Mehrara BJ. Bilateral implant breast reconstruction: outcomes, predictors, and matched cohort analysis in 730 2-stage breast reconstructions over 10 years. Ann Plast Surg. 2014;72:625–30. 27. Lin IC, Nelson JA, Wu LC, Kovach SJ 3rd, Serletti JM. Assessing surgical and medical complications in bilateral abdomen-based free flap breast reconstructions compared with unilateral free flap breast reconstructions. Ann Plast Surg. 2014. doi:10.1097/SAP.0000000000000343. 28. Craft RO, Colakoglu S, Curtis MS, et al. Patient satisfaction in unilateral and bilateral breast reconstruction. Plast Reconstr Surg. 2011;127:1417–24. 29. Fischer JP, Wes AM, Tuggle CT 3rd, Serletti JM, Wu LC. Risk analysis of early implant loss after immediate breast reconstruction: a review of 14,585 patients. J Am Coll Surg. 2013;217:983–90. 30. Fischer JP, Wes AM, Tuggle CT, Serletti JM, Wu LC. Risk analysis and stratification of surgical morbidity after immediate breast reconstruction. J Am Coll Surg. 2013;217:780–7. 31. Gart MS, Smetona JT, Hanwright PJ, et al. Autologous options for postmastectomy breast reconstruction: a comparison of outcomes based on the American College of Surgeons National Surgical Quality Improvement Program. J Am Coll Surg. 2013;216:229–38. 32. Fischer JP, Tuggle CT, Au A, Kovach SJ. A 30-day risk assessment of mastectomy alone compared to immediate breast reconstruction (IBR). J Plast Surg Hand Surg. 2014;48:209–15. 33. Ashraf M, Biswas J, Gupta S, Alam N. Determinants of wound infections for breast procedures: assessment of the risk of wound infection posed by an invasive procedure for subsequent operation. J Plast Surg Hand Surg. 2009;7:543–6. 34. Boughey JC, Hoskin TL, Hartmann LC, et al. Impact of reconstruction and reoperation on long-term patient-reported satisfaction after contralateral prophylactic mastectomy. Ann Surg Oncol. 2015;22:401–8. 35. Ting J, Rozen WM, Le Roux CM, Ashton MW, Garcia-Tutor E. Predictors of blood transfusion in deep inferior epigastric artery perforator flap breast reconstruction. J Reconstr Microsurg. 2011;27:233–8. 36. Rohde JM, Dimcheff DE, Blumberg N, et al. Health care–associated infection after red blood cell transfusion: a systematic review and meta-analysis. JAMA. 2014;311:1317–26. 37. Peled AW, Stover AC, Foster RD, McGrath MH, Hwang ES. Long-term reconstructive outcomes after expander-implant breast reconstruction with serious infectious or wound-healing complications. Ann Plast Surg. 2012;68:369–73. 38. Neumayer L, Schifftner TL, Henderson WG, Khuri SF, El-Tamer M. Breast cancer surgery in Veterans Affairs and selected university medical centers: results of the patient safety in surgery study. J Am Coll Surg. 2007;204:1235–41.
