Abstract
Introduction:
Rates of breast reconstruction following mastectomy continue to increase. The objective of this study was to determine the frequency of elective revision surgery and the number of procedures required to achieve a stable breast reconstruction two years after mastectomy.
Methods:
Women undergoing first-time breast reconstruction after mastectomy were enrolled and followed for two years, with 1996 completing reconstruction. Patients were classified based on the absence or presence of complications. Comparisons within cohorts were performed to determine factors associated with revisions and total procedures. Mixed-effects regression modeling identified factors associated with elective revisions and total surgeries.
Results:
Overall, 1534(76.9%) patients had no complications, among whom 40.2% underwent elective revisions. The average number of elective revisions differed by modality(p<0.001), with abdominally based free autologous reconstruction patients undergoing the greatest number of elective revisions(mean=0.7). The mean total number of procedures also differed(p<0.001), with Tissue expander/implant reconstructions undergoing the greatest total number of procedures(mean=2.4). Complications occurred in 462(23.1%) patients, with 67.1% of these patients undergoing elective revisions, which was significantly higher than among patients without complications(p<0.001). The mean number of procedures again differed by modality(p<0.001) and followed similar trends, but with increased mean number of revisions and procedures overall. Mixed-effects regression modeling demonstrated that patients experiencing complications had increased odds of undergoing elective revision procedures(OR=3.2, p<0.001).
Conclusions:
Breast reconstruction patients without complications undergo over two procedures on average to achieve satisfactory reconstruction, with 40% electing revisions. If a complication occurs, the number of procedures increases.
Introduction
The number of women undergoing mastectomy has continued to rise, due largely to an increase in the use of contralateral prophylactic mastectomy(1, 2) as well as increased utilization of immediate breast reconstruction(IBR)(3). The increase in IBR appears to correspond directly to a rise in implant-based procedures(3), though emerging evidence also suggests slight increases in autologous techniques(4), mainly in academic centers(5).
Clear differences exist between reconstructive modalities which have the potential to impact the patient decision-making process. Single-stage direct-to-implant reconstruction(DTI) and autologous tissue reconstruction offer patients the choice to recreate a breast mound in a single operation at the time of mastectomy, while traditional two-stage tissue expander/implant(TEI) reconstruction requires at least two operations. Autologous reconstruction procedures, however, have a significant donor-site scar and require a longer initial procedure and recovery. Regardless of the reconstructive modality chosen, many patients require refinements and revisions to achieve a satisfactory result(6, 7). Such refinements often help complete reconstruction, which is associated with maximal patient satisfaction(8, 9). A better understanding of the frequency and impact of such refinements would greatly aid in the informed consent process for patients considering breast reconstruction and deciding between reconstructive options.
Recent studies have begun to shed light on secondary revision procedures(7, 10–12), though there is no clear understanding across modalities as to the type, number, and predictors of secondary revision procedures and total procedures required for reconstruction. In 2015, Fischer et al retrospectively examined health care resource utilization after IBR in the United States using state-level surgery databases; among TEI, DTI, and autologous IBRs, they found that the TEI cohort underwent a greater number of procedures per discharge and had the highest number of unplanned revisions(11). In their 2015 retrospective, population-based study using provincial administrative databases in Ontario, Canada, Roberts et al examined patterns of secondary procedures after postmastectomy reconstruction; they reported that 88% of patients underwent at least one additional procedure after mastectomy and initial reconstruction(12). Although it is unclear whether these results can be generalized to the rest of Canada or the United States, they show that patients who underwent implant-based reconstruction had higher rates of additional procedures, compared with tissue-based reconstruction(12). In contrast to Fischer and colleagues, a single-center study by Clarke-Pearson et al that compared DTI to standard TEI reconstruction showed no significant difference in the number of revisions between their cohorts(7). However, to date there has not been a comprehensive study that directly focuses on elective revision procedures and the average total number of procedures required to achieve a stable breast reconstruction, specifically when accounting for postoperative complications.
We conducted a prospective, multicenter study of breast reconstruction patients to examine secondary procedures. We hypothesized that differences would exist in rates of elective revisions and total procedures required based on the presence of a postoperative complication, and that within these cohorts(delineated by complication), differences would exist according to reconstructive modality. The purpose of this study was to determine the number of surgeries required to achieve an acceptable breast reconstruction and to determine the frequency of elective revision procedures.
Methods
Funded by the National Cancer Institute, the Mastectomy Reconstruction Outcomes Consortium(MROC) is a prospective cohort study bringing together nine academic and two private practices in the United States and Canada with high volumes of breast reconstruction. The study was approved by the institutional review board at each participating center. Patients receiving first-time immediate or delayed breast reconstruction following mastectomy for breast cancer treatment or prophylaxis were eligible for study participation. Surgical options evaluated in this analysis included single(DTI) and two-staged(TEI) implant-based techniques, combination latissimus dorsi flap/implant(LD) procedures, pedicled transverse rectus abdominus myocutaneous(PTRAM) flaps, free TRAM(FTRAM) flaps, deep inferior epigastric artery perforator(DIEP) flaps, and superficial inferior epigastric artery(SIEA) flaps. Other, lesser used procedures, such as gluteal artery perforator flaps, were not evaluated, due to insufficient case volumes. Patients were excluded if they had a previously failed reconstruction, mixed reconstructive modality procedures, or mixed timing of reconstruction.
Demographic and clinical information was gathered preoperatively from electronic medical records. Demographic variables included age, race, ethnicity, education, income, marital status, and employment status. Clinical variables included procedure type, timing(immediate versus delayed reconstruction), laterality(unilateral versus bilateral procedures), body mass index(BMI), lymph node management(sentinel node or axillary lymph node dissection), radiation therapy, chemotherapy, and smoking status. Medical comorbidities were scored using the Charlson index(13).
One- and two-year complication data were recorded, with a complication defined as an adverse postoperative, surgery-related event requiring additional treatment. Data were abstracted for each individual adverse event. Specific complications in this cohort of patients have previously been described(14). Reconstructive failures, defined as complications necessitating implant or flap removal, were recorded. Stable breast reconstruction was considered achieved in all patients who did not experience a reconstructive failure.
Definition of elective revision procedure: An elective revision procedure was defined as any elective surgery(anesthetic event in the operating room) performed outside of the standard reconstructive algorithm, which comprised the index reconstruction and nipple areola reconstruction. The index reconstruction was the initial procedure for autologous reconstruction and DTI reconstruction or the exchange of tissue expander for permanent implant in TEI reconstruction.
Definition of total number of surgeries: The total number of surgeries was a summation of each separate anesthetic event at two years, including the initial reconstruction and nipple areola reconstruction but excluding operations for acute perioperative complications.
Statistical Analysis
All revisions were assessed and frequencies were determined. The two main cohorts of patients were analyzed for total number of procedures and elective revisions: 1) patients without complications and 2) patients with complications. The mean number of elective revision procedures and total number of surgeries were compared across reconstructive modalities. Bivariate analyses were performed to determine differences in clinical and demographic variables, comparing patients who underwent elective revisions to those who did not, using Student’s t-test for continuous variables and chi-square test or Fisher’s exact test for categorical variables. The occurrence and the number of elective revision procedures across reconstructive procedure types were compared using chi-square test and Kruskal-Wallis test, respectively. For all the analyses, we treated the patient as the analytic unit. Mixed-effects logistic regression modeling was used to identify factors associated with the occurrence of elective revisions. The model included procedure type as the primary predictor, with TEI reconstruction as the reference category. The model further included as covariates patient demographic and clinical characteristics, and hospitals/centers as random intercepts to account for between-center variability. Generalized estimating equation(GEE) with a Poisson link function was used to identify factors associated with the total number of elective revisions and total number of procedures. The model included the same set of predictor and covariates as described above, and used an exchangeable correlation matrix structure for patients within the same hospital/center to account for between-center variability. The issue of over-dispersion was encountered while fitting the aforementioned Poisson regression. We therefore adjusted the chi-square statistic and standard error associated with each coefficient using the deviance statistic. We calculated adjusted odds ratios(for mixed-effects logistic regression model) or coefficients and percentage increase/decrease(for GEE model) with 95% confidence intervals(CIs) estimated by the model. Statistical significance was set at 0.05. All statistical analyses were performed using SAS v9.4(SAS Institute, Inc., Cary, NC).
Results
In total, 2113 patients had complete follow-up two years after breast reconstruction; 117(5.5%) had a reconstructive failure and were excluded. Of the 1996 included patients, 1534(76.9%) had a complication-free postoperative course, and 462(23.1%) experienced a complication requiring further treatment.
Patients without Complication Analysis
Among patients without any complication(n=1534), 617(40.2%) underwent elective revisions(Table 1). Patients electing to undergo revisions had a slightly higher BMI(p<0.001), and were more likely to have had delayed reconstruction(p<0.001). Differences were also noted in cancer treatment, with lower rates of elective revision procedures noted among patients undergoing radiation therapy before or after surgery(p=0.001) and among patients receiving chemotherapy during or after reconstruction(p=0.022). Importantly, significant differences in the incidence of elective revision procedures were noted by reconstructive modality, ranging from 25% in DTI reconstructions to 59.2% in LD flap reconstructions(p<0.001).
Table 1.
Clinical and Demographic Characteristics for Patients without Postoperative Complication (n=1534)
| Variable | No revision n=917 | Revision n=617 | P-Value | |
|---|---|---|---|---|
| Age, mean (SD) | 48.8 (10.5) | 49.5 (9.5) | 0.181 | |
| BMI, mean (SD) | 25.7 (5.3) | 26.7 (5.1) | <.001 | |
| Procedure type | ||||
| DTI | 51 (75.0%) | 17 (25.0%) | <.001 | |
| TE | 665 (65.7%) | 347 (34.3%) | ||
| PTRAM | 31 (59.6%) | 21 (40.4%) | ||
| FTRAM | 29 (42.0%) | 40 (58.0%) | ||
| DIEP | 99 (40.4%) | 146 (59.6%) | ||
| LD | 20 (40.8%) | 29 (59.2%) | ||
| SIEA | 22 (56.4%) | 17 (43.6%) | ||
| Laterality | ||||
| Unilateral | 419 (58.4%) | 299 (41.6%) | 0.287 | |
| Bilateral | 498 (61.0%) | 318 (39.0%) | ||
| Indication | ||||
| Therapeutic | 821 (59.9%) | 550 (40.1%) | 0.808 | |
| Prophylactic | 96 (58.9%) | 67 (41.1%) | ||
| Timing | ||||
| Immediate | 867 (61.1%) | 553 (38.9%) | <.001 | |
| Delayed | 50 (43.9%) | 64 (56.1%) | ||
| Diabetes | ||||
| Yes | 43 (68.3%) | 20 (31.7%) | 0.161 | |
| No | 874 (59.4%) | 597 (40.6%) | ||
| Radiation | ||||
| Before reconstruction | 87 (48.6%) | 92 (51.4%) | <.001 | |
| During/after reconstruction | 167 (67.3%) | 81 (32.7%) | ||
| None | 663 (59.9%) | 444 (40.1%) | ||
| Chemo | ||||
| During/after reconstruction | 289 (64.2%) | 161 (35.8%) | 0.022 | |
| Not during/after reconstruction | 628 (57.9%) | 456 (42.1%) | ||
| Race | ||||
| White | 784 (58.8%) | 549 (41.2%) | 0.067 | |
| Black | 59 (61.5%) | 37 (38.5%) | ||
| Other | 62 (71.3%) | 25 (28.7%) |
Number of revisions and total surgeries
Patients without complications underwent a mean of 0.5(SD=0.7) revision procedures(Table 2). This varied significantly by reconstructive modality(p<0.001), ranging from 0.4(0.6) revision procedures in TEI to 0.7(0.7) in FTRAM/DIEP/SIEA patients. The mean(SD) total number of surgeries needed to achieve a stable reconstruction at two years was 2.2(0.7). Similarly, the total number of procedures varied by reconstructive modality(p<0.001), ranging from 1.4(0.8) in DTI reconstructions to 2.4(0.7) in TEI reconstructions.
Table 2.
Occurrence and Number of Revision Procedures and Number of Total Surgeries by Procedure Type
| Procedure type | No complication |
With complication |
||||
|---|---|---|---|---|---|---|
| No. (%) with revision p<.001 | No. of revision, mean (SD) p<.001 | No. of surgery, mean (SD) p<.001 | No. (%) with revision p=0.041 | No. of revision, mean (SD) p=0.278 | No. of surgery, mean (SD) p<.001 | |
| DTI | 17 (25) | 0.4 (0.7) | 1.4 (0.8) | 14 (56) | 0.8 (0.8) | 1.9 (0.8) |
| TE | 347 (34.3) | 0.4 (0.6) | 2.4 (0.7) | 156 (62.2) | 0.9 (0.9) | 2.8 (0.9) |
| PTRAM | 21 (40.4) | 0.5 (0.6) | 1.8 (0.6) | 19 (76) | 1.2 (1) | 2.7 (1.2) |
| FTRAM/DIEP/SIEA | 203 (57.5) | 0.7 (0.7) | 2 (0.8) | 109 (74.7) | 1 (0.8) | 2.3 (0.8) |
| LD | 29 (59.2) | 0.6 (0.6) | 2.3 (0.7) | 12 (80) | 1.1 (0.9) | 2.7 (0.9) |
| Total | 617 (40.2) | 0.5 (0.7) | 2.2 (0.7) | 310 (67.1) | 1 (0.9) | 2.6 (0.9) |
Patients with Complications Analysis
Complications occurred in 462(23.1%) of patients in this cohort(Table 3), with the majority of these patients(n=310, 67.1%) electing revision procedures. Patients with complications were significantly more likely to have revisions, compared with patients who did not have a complication(67.1% vs. 40.2%; p<0.001). Patients with reconstruction after prophylactic mastectomy had higher rates of elective revisions, compared with patients with a cancer diagnosis(p=0.036). Differences were also noted among patients receiving cancer treatment, with lower rates of elective revisions seen in patients undergoing radiation therapy during or after reconstruction(p=0.035) versus no radiation therapy, and in patients undergoing chemotherapy(p=0.023) versus no chemotherapy. The difference in elective revision procedure rate between modalities ranged from 56.0% in DTI reconstruction patients to 88.9% in FTRAM reconstruction patients(p=0.062 comparing all modalities).
Table 3.
Clinical and Demographic Characteristics for Patients with Postoperative Complication (n=462)
| Variable | No revision n=152 | Revision n=310 | P-Value | |
|---|---|---|---|---|
| Age, mean (SD) | 50.7 (11.3) | 50.9 (9.5) | 0.896 | |
| BMI, mean (SD) | 27.8 (7.4) | 27.9 (6.0) | 0.923 | |
| Procedure type | ||||
| DTI | 11 (44.0%) | 14 (56.0%) | 0.062 | |
| TE | 95 (37.8%) | 156 (62.2%) | ||
| PTRAM | 6 (24.0%) | 19 (76.0%) | ||
| FTRAM | 2 (11.1%) | 16 (88.9%) | ||
| DIEP | 28 (26.7%) | 77 (73.3%) | ||
| LD | 3 (20.0%) | 12 (80.0%) | ||
| SIEA | 7 (30.4%) | 16 (69.6%) | ||
| Laterality | ||||
| Unilateral | 64 (31.4%) | 140 (68.6%) | 0.534 | |
| Bilateral | 88 (34.1%) | 170 (65.9%) | ||
| Indication | ||||
| Therapeutic | 144 (34.4%) | 275 (65.6%) | 0.036 | |
| Prophylactic | 8 (18.6%) | 35 (81.4%) | ||
| Timing | ||||
| Immediate | 146 (33.9%) | 285 (66.1%) | 0.097 | |
| Delayed | 6 (19.4%) | 25 (80.6%) | ||
| Diabetes | ||||
| Yes | 7 (31.8%) | 15 (68.2%) | 0.912 | |
| No | 145 (33.0%) | 295 (67.0%) | ||
| Radiation | ||||
| Before reconstruction | 18 (26.9%) | 49 (73.1%) | 0.035 | |
| During/after reconstruction | 43 (43.4%) | 56 (56.6%) | ||
| None | 91 (30.7%) | 205 (69.3%) | ||
| Chemo | ||||
| During/after reconstruction | 57 (40.4%) | 84 (59.6%) | 0.023 | |
| Not during/after reconstruction | 95 (29.6%) | 226 (70.4%) | ||
| Race | ||||
| White | 132 (32.0%) | 281 (68.0%) | 0.293 | |
| Black | 9 (37.5%) | 15 (62.5%) | ||
| Other | 10 (47.6%) | 11 (52.4%) |
Patients with complications required a mean(SD) of 1.0(0.9) revisions, with no significant variation by modality(Table 2). This cohort underwent a total of 2.6(0.9) surgeries to achieve a stable reconstruction, varying significantly among reconstructive modalities(p<0.001) from 1.9(0.8) in DTI reconstructions to 2.8(0.9) in TEI reconstructions.
Types of Elective Revision Procedures
The most common revision procedure(Table 4) was fat grafting to the breast(23.3% of patients overall, 50.1% of patients electing revisions). Other commonly performed procedures included recontouring/repositioning of the flap(13.0% overall, 27.9% of patients electing revisions) and scar revision or dog-ear excision(11.0% overall, 23.6% of patients electing revisions).
Table 4.
Summary of Elective Revision Performed among MROC Patients
| Type | n | % overall | % among patients with elective revision |
|---|---|---|---|
| Fat grafting to the breast | 464 | 23.3 | 50.1 |
| Removal of implant and replacement with different size, shape or type of implant | 113 | 5.7 | 12.2 |
| Removal of implant without replacement | 5 | 0.3 | 0.5 |
| Reposition of implant | 31 | 1.6 | 3.3 |
| Recontouring /repositioning of flap | 259 | 13.0 | 27.9 |
| Scar revision and/or dog ear excision | 219 | 11.0 | 23.6 |
| Other elective revision procedures | 109 | 5.5 | 11.8 |
Models examining elective revisions and total procedures
Mixed-effects logistic regression modeling(Table 5) controlling for differences in demographics and surgical characteristics demonstrated that patients who were neither black or white were less likely to undergo elective revision procedures(odds ratio [OR]=0.59, p=0.022) as were patients undergoing radiation therapy during or after reconstruction(OR=0.69, p=0.012). Patients with complications were more likely to undergo revision procedures(OR=3.2, p<0.001). When comparing the various reconstructive modalities to TEI reconstruction, DTI reconstruction patients were less likely(OR=0.58, p=0.035) to undergo elective revisions, while FTRAM(OR=2.26, p=0.002), DIEP(OR=2.66, p<0.001), and LD(OR=1.97, p=0.031) patients were more likely to undergo revisions.
Table 5.
Mixed-effects Logistic Regression for the Occurrence of Revision Procedures
| Variable | OR | 95% CI | p | |
|---|---|---|---|---|
| Age | 1.00 | (0.99, 1.01) | 0.565 | |
| BMI | 1.01 | (0.99, 1.03) | 0.516 | |
| Procedure type | ||||
| TE | -Reference- | |||
| DTI | 0.58 | (0.35, 0.96) | 0.035 | |
| PTRAM | 1.34 | (0.75, 2.40) | 0.330 | |
| FTRAM | 2.26 | (1.35, 3.78) | 0.002 | |
| DIEP | 2.66 | (1.83, 3.86) | <.001 | |
| LD | 1.97 | (1.07, 3.64) | 0.031 | |
| SIEA | 1.83 | (0.93, 3.60) | 0.079 | |
| With postoperative complication | 3.21 | (2.52, 4.10) | <.001 | |
| Bilateral reconstruction | 1.06 | (0.85, 1.32) | 0.610 | |
| Prophylactic mastectomy | 0.99 | (0.71, 1.39) | 0.957 | |
| Delayed timing | 1.16 | (0.76, 1.79) | 0.494 | |
| Diabetes | 0.66 | (0.40, 1.09) | 0.104 | |
| Radiation | ||||
| None | -Reference- | |||
| Radiation before reconstruction | 0.94 | (0.66, 1.34) | 0.740 | |
| Radiation during or after reconstruction | 0.69 | (0.51, 0.92) | 0.012 | |
| Chemo during or after reconstruction | 0.90 | (0.70, 1.14) | 0.374 | |
| Race | ||||
| White | -Reference- | |||
| Black | 0.86 | (0.56, 1.32) | 0.494 | |
| Other | 0.59 | (0.37, 0.93) | 0.022 |
Multivariable analysis examining the total number of surgeries(Table 6) demonstrated that patients undergoing abdominally based reconstruction underwent between 13% and 18% fewer surgeries than TEI reconstruction patients(p<0.01). DTI patients also underwent fewer surgeries(38.2% fewer, p<0.001) than TEI patients. Patients with complications experienced a 20% increase in number of surgeries, compared with those without complications(p<0.001). Radiation, before or after surgery, was associated with a decrease in the total number of procedures overall.
Table 6.
Generalized Estimating Equation Model for Total Number of Surgeries
| Variable | Coefficient | 95% CI | p | % Increase1 | |
|---|---|---|---|---|---|
| Age | −0.0023 | (−0.0034, −0.0011) | <.001 | −0.2 | |
| BMI | −0.0021 | (−0.0049, 0.0007) | 0.149 | −0.2 | |
| Procedure type | |||||
| TE | -Reference- | ||||
| DTI | −0.4816 | (−0.6029, −0.3603) | <.001 | −38.2 | |
| PTRAM | −0.1856 | (−0.2969, −0.0744) | 0.001 | −16.9 | |
| FTRAM | −0.1708 | (−0.2665, −0.0752) | 0.001 | −15.7 | |
| DIEP | −0.1438 | (−0.2160, −0.0715) | <.001 | −13.4 | |
| LD | −0.0568 | (−0.1054, −0.0083) | 0.022 | −5.5 | |
| SIEA | −0.1980 | (−0.2618, −0.1341) | <.001 | −18.0 | |
| With postoperative complication | 0.1827 | (0.1387, 0.2267) | <.001 | 20.0 | |
| Bilateral reconstruction | −0.0037 | (−0.0344, 0.0269) | 0.812 | −0.4 | |
| Prophylactic mastectomy | −0.0566 | (−0.0922, −0.0209) | 0.002 | −5.5 | |
| Delayed timing | 0.0472 | (0.0078, 0.0865) | 0.019 | 4.8 | |
| Diabetes | −0.1167 | (−0.1606, −0.0727) | <.001 | −11.0 | |
| Radiation | |||||
| None | -Reference- | ||||
| Before reconstruction | −0.0482 | (−0.0851, −0.0114) | 0.010 | −4.7 | |
| During or after reconstruction | −0.1476 | (−0.1934, −0.1018) | <.001 | −13.7 | |
| Chemo during/after reconstruction | −0.0387 | (−0.0696, −0.0077) | 0.014 | −3.8 | |
| Race | |||||
| White | -Reference- | ||||
| Black | −0.0267 | (−0.1136, 0.0601) | 0.546 | −2.6 | |
| Other | −0.0140 | (−0.0690, 0.0411) | 0.619 | −1.4 |
Computed as 100* {exp(Coefficient)-1}.
Discussion
Breast reconstruction typically requires multiple procedures to achieve a satisfactory, stable reconstruction. This prospective, multicenter evaluation of nearly 2000 patients with two-year follow-up examined elective revision rates and total number of surgeries in women undergoing breast reconstruction after mastectomy. The most important findings are that 40% of women undergoing breast reconstruction without complication undergo revision procedures and require just over two surgeries to achieve a satisfactory reconstruction with the exact number differing by modality. When a complication occurs, elective revision rates and total number of surgeries increase.
Regional differences in utilization and outcomes exist based on the type and surgical volume of an institution(15, 16); not all institutions offer a comprehensive breast reconstruction program. It is important, however, that patients understand the difference between tissue-based and implant-based reconstruction, and that they are given realistic and practical information about their chosen reconstructive option. Many surgeons suggest that immediate autologous reconstruction is a more involved procedure, with longer operative time up front, longer recovery, and higher rate of early complications but with fewer procedures and complications in the long term(17). Proponents of TEI reconstruction advocate that immediate implant reconstruction is much shorter than autologous reconstruction, and the subsequent procedures are minor in comparison. Traditional TEI reconstruction requires two surgeries to achieve reconstruction, whereas autologous reconstruction provides an immediate creation of the breast mound. This study offers data to help clarify these notions and assumptions.
In this study, 40% of patients without complications underwent elective revision procedures. This number varies greatly by reconstructive modality, ranging from 25% in DTI patients to 59% in LD patients. Importantly, although abdominally based autologous reconstruction patients had higher rates of revision, they underwent fewer total procedures on average (1.8 to 2.0 procedures) than the TEI cohort(2.4 procedures). This is consistent with the notion that TEI-based reconstruction requires a greater number of procedures to achieve a satisfactory reconstruction. Complications resulted in higher numbers of revision procedures and total surgeries, a finding that supports previous research.
Few studies to date have compared modalities with these specific outcome metrics, and none to the extent in this study. An increasing body of literature exists comparing early outcomes between autologous and alloplastic techniques(18). In 2013, Fischer et al performed a head-to-head comparison of autologous breast reconstruction and TEI reconstruction among 202 patients at a single institution(19). In their study, 46% of patients underwent revision procedures, but this subset of patients was not investigated further. Interestingly, there was no significant difference across modalities with regards to revision rates(49.3% vs. 38.3%, p=0.17), which differs from our results but may have been due to sample size limitations and an inverse ratio of autologous to TEI patients compared to the current study.
Several studies have examined revisions after postmastectomy breast reconstruction at the population level(11, 12). Roberts et al found that 88% of nearly 4000 patients underwent secondary procedures over a 5-year follow-up, and they included second-stage TEI exchange procedures in their calculation(12). In our study, TEI reconstruction is assumed to require two procedures to achieve index reconstruction. As such, we attempted to take this a step further, looking only at revisions outside of the standard algorithm of index reconstruction and nipple reconstruction. Roberts et al also noted no difference in the total number of surgeries across broad reconstructive modalities(tissue-based vs. implant-based). Our study further stratified the reconstructive modalities and noted differences.
Among approximately 15,000 IBR patients across four states, Fischer et al found that secondary revision rates, both planned and unplanned, were significantly higher after TEI reconstruction than after autologous reconstruction, with unplanned revisions needed after TEI reconstruction in 59.2% of patients as compared with 34.4% in the autologous reconstruction group(11). This large difference, which varies from our current finding, may be related to their inclusion of IBR only and/or to their definitions of revisions and procedures. Our study included delayed reconstruction as well as IBR, although less than 10% of patients had delayed procedures.
Several of our findings were not intuitive. In our study, total procedures ranged from an average of 1.4(DTI) to 2.4(TEI). Autologous reconstructions are typically thought of as requiring fewer elective revisions than implant-based procedures; however, our results differed, though the total number of surgeries required was in line with this principle. We believe this may be due to revisions being performed simultaneously during the exchange for a permanent implant in TEI reconstructions, and not as a stand-alone additional procedure. Since autologous reconstructions produce a stable breast mound after the initial operation, each subsequent procedure is considered elective. Unexpectedly, postoperative radiation therapy was a negative predictor of revision procedures. However, patients requiring radiation therapy typically have more advanced disease, and may either defer additional surgery or elect to not pursue further revisions. Similar findings were recently reported in an examination of prosthetic reconstruction in British Columbia(20). Alternatively, it is also possible that the plastic surgeon counsels against an elective revision, as operating in an irradiated bed can be associated with additional complications(21, 22). It is important to note that the exact timing of exchange or revisions in the current cohort was not directly addressed in this study, although no difference in radiation therapy timing, with regards to placement of the permanent implant, was found in previously published MROC data(23). Other studies suggest that differences may exist(24–26), and this is currently an area of ongoing debate.
In a time of increasing scrutiny on health care utilization and an emphasis on value-based medicine, it is important to consider cost when examining reconstructive modalities and secondary procedures. Undergoing fewer procedures is consistent with decreased direct and indirect cost. Studies examining cost across reconstructive modalities demonstrate results favoring autologous reconstruction over the long term(11, 19, 27–30). Nonetheless, costs must be examined in light of patient satisfaction and the potential improvement in satisfaction that further procedures may bring. Unfortunately, it is difficult to directly assess satisfaction relating to the revision procedure, as it is plausible that although revisions may improve satisfaction in a cohort of patients, the passage of time may improve satisfaction in others, resulting in no notable difference between cohorts. Although the MROC study did collect patient-reported outcomes data, these data were centered around the time from the initial procedure; the study was not designed to appropriately capture data before and after revision procedures. Therefore, we elected to not include patient-reported outcomes data in the current examination of elective revisions. Recently published MROC data, however, demonstrates a superior outcome in autologous tissue reconstructions from a patient perspective, even at one year(31).
Complications inevitably occur in a percentage of patients following breast reconstruction. Although current techniques are reliable, complications lead to additional procedures. In this study’s cohort, 23% of patients experienced a complication following reconstruction, as recently described by Wilkins et al in their one-year report of MROC study outcomes(14). Some patients with a complication and poor result may elect to forego reconstruction or find a different reconstructive surgeon. However, the majority of patients and surgeons persevere to achieve a satisfactory breast reconstruction, though such reconstructions are often more challenging. Our results demonstrate that these patients require additional procedures to achieve a satisfactory reconstruction, which supports previous work by Clark-Pearson and colleagues, which also demonstrated that complications lead to revision procedures(7).
Large population datasets suggest that implant-based reconstruction accounts for 70% to 79% of overall reconstructions(11, 32, 33). This multicenter MROC study demonstrates similar distributions across modalities, with 63% of patients undergoing TEI reconstruction and a slightly higher than expected 32% undergoing autologous tissue reconstruction. This difference is likely related to the greater use of autologous reconstruction at teaching institutions(34), which made up the majority of centers enrolling patients in this study and helped improve the statistical power of the study.
These results provide information that is practical and meaningful to the breast reconstruction patient. We set out to generate essential information for patient-centered decision-making, to provide patients with insight regarding the number of surgeries they are likely to undergo depending on their chosen reconstructive pathway. Our study found the average number of total procedures across all modalities to be greater than two. However, the number varies based on reconstructive modality. Patients should be counseled that approximately 40% will elect to undergo further revisions of their reconstruction following an uncomplicated initial reconstruction. These data are supported by two-year follow-up, a time frame that we believe should capture most, if not all, elective revision surgery. At two years, an autologous reconstruction without any complications is expected to be stable, whereas some patients undergoing alloplastic reconstructions will likely desire a revision due to implant-associated infection, malposition, failure, or capsular contracture. Longer follow-up would provide additional information for patients to incorporate in their decision-making process.
The strengths of this multicenter study include its generalizability, prospective design, and large size, with 11 sites and over 57 surgeons. Nevertheless, some limitations exist. This study is not a randomized controlled trial, and as such we could not balance both known and unknown prognostic factors between groups, which may impact outcomes. Selection bias certainly also exists in choice of reconstructive modality, and in differences exist within the centers regarding rates of revisions. In addition, more than half of the patients in our cohort did not undergo any revision procedures, which indicates “excessive zeros” for the outcome and could lead to poor fit for the Poisson model. Given the difficulty in applying alternative modeling techniques (for example, zero-inflated Poisson or negative binomial model) to clustered data, we utilized the conventional Poisson regression. The results therefore require interpretation with caution. We could not standardize each reconstructive procedure, as variations in surgical technique exist among the study’s 57 surgeons. Some procedural subgroups(DTI, LD, PTRAM) contained fewer patients, making it difficult to draw definitive conclusions. The autologous procedures did not include newer, lesser utilized techniques; thus, results cannot be directly applied to patients undergoing these procedures. Even with these limitations, this is the first prospective, multicenter study specifically designed to directly examine revision rates across reconstructive modalities.
Conclusions
Breast reconstruction may involve multiple procedures to achieve a satisfactory result, with the average number of revisions differing by reconstructive modality and by the occurrence of complications. Patients should be counseled that 40% of patients elect revisions, with the total number of surgeries being over two to achieve a satisfactory reconstruction assuming there are no complications. Should a complication occur, the number of procedures increases.
Acknowledgements
Support for this study was provided by a grant from the National Cancer Institute(1R01CA152192) and in part by the Cancer Center Support Grant P30 CA008748.
The authors gratefully acknowledge the contributions of the MROC Site Principal Investigators: Yoon S. Chun, MD(Brigham and Women’s Hospital), Troy A. Pittman, MD(Georgetown University), Mark W. Clemens, MD(MD Anderson Cancer Center), Gayle Gordillo, MD(The Ohio State University), Daniel Sherick, MD(Saint Joseph’s Mercy Hospital), Nancy Van Laeken, MD(University of British Columbia), Ed Buchel, MD(University of Manitoba), John Y. Kim, MD(Northwestern University), Richard Greco, MD(Georgia Institute of Plastic Surgery).
Footnotes
This paper was presented at the Plastic Surgery Research Council 62nd annual Meeting. May 4–7, 2017. Durham, North Carolina.
Disclosures
Dr. Pusic is a co-developer of BREAST-Q and receives royalty payments when it is used in industry-sponsored trials. None of the other authors have a financial interest in any of the products, devices, or drugs mentioned in this article.
References
- 1.Steiner CA, Weiss AJ, Barrett ML, Fingar KR, Davis PH Trends in Bilateral and Unilateral Mastectomies in Hospital Inpatient and Ambulatory Settings, 2005–2013: Statistical Brief #201. 2016. [PubMed] [Google Scholar]
- 2.Cemal Y, Albornoz CR, Disa JJ, et al. A paradigm shift in U.S. breast reconstruction: Part 2. The influence of changing mastectomy patterns on reconstructive rate and method. Plast Reconstr Surg 2013;131:320e–326e. [DOI] [PubMed] [Google Scholar]
- 3.Albornoz CR, Bach PB, Mehrara BJ, et al. A paradigm shift in U.S. Breast reconstruction: increasing implant rates. Plast Reconstr Surg 2013;131:15–23. [DOI] [PubMed] [Google Scholar]
- 4.Ilonzo N, Tsang A, Tsantes S, Estabrook A, Thu Ma AM Breast reconstruction after mastectomy: A ten-year analysis of trends and immediate postoperative outcomes. Breast 2016;32:7–12. [DOI] [PubMed] [Google Scholar]
- 5.Albornoz CR, Cordeiro PG, Mehrara BJ, et al. Economic implications of recent trends in U.S. immediate autologous breast reconstruction. Plast Reconstr Surg 2014;133:463–470. [DOI] [PubMed] [Google Scholar]
- 6.Fischer JP, Sieber B, Nelson JA, et al. Comprehensive outcome and cost analysis of free tissue transfer for breast reconstruction: an experience with 1303 flaps. Plast Reconstr Surg 2013;131:195–203. [DOI] [PubMed] [Google Scholar]
- 7.Clarke-Pearson EM, Lin AM, Hertl C, Austen WG, Colwell AS Revisions in Implant-Based Breast Reconstruction: How Does Direct-to-Implant Measure Up? Plast Reconstr Surg 2016;137:1690–1699. [DOI] [PubMed] [Google Scholar]
- 8.Buck DW 2nd, Shenaq D, Heyer K, Kato C, Kim JY Patient-subjective cosmetic outcomes following the varying stages of tissue expander breast reconstruction: the importance of completion. Breast 2010;19:521–526. [DOI] [PubMed] [Google Scholar]
- 9.Atisha D, Alderman AK, Lowery JC, Kuhn LE, Davis J, Wilkins EG Prospective analysis of long-term psychosocial outcomes in breast reconstruction: two-year postoperative results from the Michigan Breast Reconstruction Outcomes Study. Ann Surg 2008;247:1019–1028. [DOI] [PubMed] [Google Scholar]
- 10.Kim EK, Suh YC, Maldonado AA, Yun J, Lee TJ Patients’ Aesthetic Concerns After Horizontally Placed Abdominal Free Flap Breast Reconstruction. Aesthetic Plast Surg 2015;39:686–693. [DOI] [PubMed] [Google Scholar]
- 11.Fischer JP, Fox JP, Nelson JA, Kovach SJ, Serletti JM A Longitudinal Assessment of Outcomes and Healthcare Resource Utilization After Immediate Breast Reconstruction-Comparing Implant- and Autologous-based Breast Reconstruction. Ann Surg 2015;262:692–699. [DOI] [PubMed] [Google Scholar]
- 12.Roberts A, Baxter N, Camacho X, Lau C, Zhong T Once is Rarely Enough: A Population-Based Study of Reoperations after Postmastectomy Breast Reconstruction. Annals of surgical oncology 2015;22:3302–3307. [DOI] [PubMed] [Google Scholar]
- 13.Hall WH, Ramachandran R, Narayan S, Jani AB, Vijayakumar S An electronic application for rapidly calculating Charlson comorbidity score. BMC Cancer 2004;4:94. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Wilkins EG, Hamill JB, Kim HM, et al. Complications in Postmastectomy Breast Reconstruction: One-year Outcomes of the Mastectomy Reconstruction Outcomes Consortium(MROC) Study. Ann Surg 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Albornoz CR, Cordeiro PG, Hishon L, et al. A nationwide analysis of the relationship between hospital volume and outcome for autologous breast reconstruction. Plast Reconstr Surg 2013;132:192e–200e. [DOI] [PubMed] [Google Scholar]
- 16.Albornoz CR, Cohen WA, Razdan SN, et al. The Impact of Travel Distance on Breast Reconstruction in the United States. Plast Reconstr Surg 2016;137:12–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Fogarty BJ, Brown AP, Miller R, Khan K TRAM flap versus nonautologous breast reconstruction: what do patients really think? Plast Reconstr Surg 2004;113:1146–1152. [DOI] [PubMed] [Google Scholar]
- 18.Tsoi B, Ziolkowski NI, Thoma A, Campbell K, O’Reilly D, Goeree R Safety of tissue expander/implant versus autologous abdominal tissue breast reconstruction in postmastectomy breast cancer patients: a systematic review and meta-analysis. Plast Reconstr Surg 2014;133:234–249. [DOI] [PubMed] [Google Scholar]
- 19.Fischer JP, Nelson JA, Cleveland E, et al. Breast reconstruction modality outcome study: a comparison of expander/implants and free flaps in select patients. Plast Reconstr Surg 2013;131:928–934. [DOI] [PubMed] [Google Scholar]
- 20.Ho AL, Bovill ES, Macadam SA, Tyldesley S, Giang J, Lennox PA Postmastectomy radiation therapy after immediate two-stage tissue expander/implant breast reconstruction: a University of British Columbia perspective. Plast Reconstr Surg 2014;134:1e–10e. [DOI] [PubMed] [Google Scholar]
- 21.Chetta MD, Aliu O, Zhong L, et al. Reconstruction of the Radiated Breast: A National Claims-Based Assessment of Postoperative Morbidity. Plast Reconstr Surg 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Jagsi R, Jiang J, Momoh AO, et al. Complications After Mastectomy and Immediate Breast Reconstruction for Breast Cancer: A Claims-Based Analysis. Ann Surg 2016;263:219–227. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Santosa KB, Chen X, Qi J, et al. Postmastectomy Radiation Therapy and Two-Stage Implant-Based Breast Reconstruction: Is There a Better Time to Irradiate? Plast Reconstr Surg 2016;138:761–769. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Yan C, Fischer JP, Freedman GM, et al. The Timing of Breast Irradiation in Two-Stage Expander/Implant Breast Reconstruction. Breast J 2016;22:322–329. [DOI] [PubMed] [Google Scholar]
- 25.Cordeiro PG, Albornoz CR, McCormick B, et al. What Is the Optimum Timing of Postmastectomy Radiotherapy in Two-Stage Prosthetic Reconstruction: Radiation to the Tissue Expander or Permanent Implant? Plast Reconstr Surg 2015;135:1509–1517. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Lam TC, Hsieh F, Boyages J The effects of postmastectomy adjuvant radiotherapy on immediate two-stage prosthetic breast reconstruction: a systematic review. Plast Reconstr Surg 2013;132:511–518. [DOI] [PubMed] [Google Scholar]
- 27.Fischer JP, Wes AM, Nelson JA, et al. Propensity-matched, longitudinal outcomes analysis of complications and cost: comparing abdominal free flaps and implant-based breast reconstruction. J Am Coll Surg 2014;219:303–312. [DOI] [PubMed] [Google Scholar]
- 28.Matros E, Albornoz CR, Razdan SN, et al. Cost-effectiveness analysis of implants versus autologous perforator flaps using the BREAST-Q. Plast Reconstr Surg 2015;135:937–946. [DOI] [PubMed] [Google Scholar]
- 29.Razdan SN, Cordeiro PG, Albornoz CR, et al. Cost-Effectiveness Analysis of Breast Reconstruction Options in the Setting of Postmastectomy Radiotherapy Using the BREAST-Q. Plast Reconstr Surg 2016;137:510e–517e. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Lagares-Borrego A, Gacto-Sanchez P, Infante-Cossio P, Barrera-Pulido F, Sicilia-Castro D, Gomez-Cia T A comparison of long-term cost and clinical outcomes between the two-stage sequence expander/prosthesis and autologous deep inferior epigastric flap methods for breast reconstruction in a public hospital. J Plast Reconstr Aesthet Surg 2016;69:196–205. [DOI] [PubMed] [Google Scholar]
- 31.Pusic AL, Matros E, Fine N, et al. Patient-Reported Outcomes 1 Year After Immediate Breast Reconstruction: Results of the Mastectomy Reconstruction Outcomes Consortium Study. J Clin Oncol 2017;35:2499–2506. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Kamali P, Paul MA, Ibrahim AM, et al. National and Regional Differences in 32,248 Postmastectomy Autologous Breast Reconstruction Using the Updated National Inpatient Survey. Ann Plast Surg 2017. [DOI] [PubMed] [Google Scholar]
- 33.Fischer JP, Nelson JA, Au A, Tuggle CT 3rd, Serletti JM, Wu LC Complications and morbidity following breast reconstruction--a review of 16,063 cases from the 2005–2010 NSQIP datasets. J Plast Surg Hand Surg 2014;48:104–114. [DOI] [PubMed] [Google Scholar]
- 34.Masoomi H, Wirth GA, Paydar KZ, Richland BK, Evans GR Perioperative outcomes of autologous breast reconstruction surgery in teaching versus nonteaching hospitals. Plast Reconstr Surg 2014;134:514e–520e. [DOI] [PubMed] [Google Scholar]