The oncoplastic reduction technique has become an accepted approach for reconstructing partial mastectomy defects in women with macromastia.1–3 As it grows in popularity, the procedure continues to be refined, and indications for its use continue to broaden. There are now larger series with longer follow-up demonstrating oncologic safety, improved patient satisfaction, and improved quality of life.4–6 This approach is now being used for larger tumors, higher cancer stage, and in conjunction with intraoperative radiation therapy.7–10 In addition to broadening the indications for breast conservation, these techniques have been used for larger defects and in smaller breasts. Certain size defects and remote locations are often difficult to reconstruct using the traditional oncoplastic reduction techniques. One solution to this problem has been to use autoaugmentation flaps where surrounding tissue can be moved to fill difficult defects. These parenchymal flaps have been used in massive weight loss patients to provide additional shape and volume to deflated breasts.11 Various authors have used central, inferior, or superomedial pedicle techniques with vascularized attachments to provide suspension of the breast mound, particularly in the upper pole where volume is often desired.11–14 Similar autoaugmentation flaps during oncoplastic reduction procedures will fill the defect and reshape the breast mound when traditional techniques would not suffice. An autoaugmentation flap can either be used as a dermoglandular extension of the primary pedicle beyond the nipple-areola complex or as a secondary dermoglandular pedicle independent of the primary nipple pedicle. Concerns have been raised as to whether these autoaugmentation flaps might result in additional fat necrosis or scar tissue and potentially impact initiation of radiation therapy or outcomes. The purpose of this review is to evaluate the indications for autoaugmentation flaps in oncoplastic reduction patients and assess outcomes to determine whether this approach results in a higher incidence of complications.
PATIENTS AND METHODS
All patients who underwent partial mastectomy and oncoplastic reduction between November of 1997 and October of 2015 at Emory Hospital were included. Data were collected from a prospectively maintained database and electronic medical records and then recorded in a Microsoft Excel database (Microsoft Corp., Redmond, Wash.). This study was approved by the university’s institutional review board. Patient demographics, risk factors, diagnosis, tumor location, tumor size, operative technique, and outcomes were reviewed. The incidence of completion mastectomy, reexcision, and rebiopsy was analyzed. The rebiopsy rate was calculated from the sum of all biopsy types: stereotactic, mammography directed, and excisional. We used the contralateral breast reduction specimen weight as a marker for ipsilateral breast size. Because the majority of patients had a bilateral breast reduction following removal of the lumpectomy specimen from the ipsilateral cancer side, we found it easiest to estimate the ipsilateral overall breast size by using the breast reduction specimen weight from the contralateral side. For the purposes of this analysis, this method provided us with a single accurate specimen weight.
The documented surgical complications were as follows: (1) urgent or emergent return to the operating room; (2) unanticipated patient readmission; (3) breast seroma, defined by a clinically evident fluid collection requiring aspiration, drainage, or operating room washout; (4) hematoma, defined as clinically evident postoperative swelling, or significant bruising requiring operating room evacuation; (5) wound infection, defined as a surgical incision with increasing erythema, tenderness, or purulent fluid on physical examination treated with oral or intravenous antibiotics or operating room washout; (6) partial nipple necrosis, defined as clinical evidence of full-thickness necrosis requiring serial dressing changes and/or operating room or office débridement; (7) delayed wound healing, defined as superficial dehiscence at the surgical site requiring serial dressing changes and/or operating room or office débridement; and (8) fat necrosis, defined as a clinically palpable mass or area causing pain, deformity, or skin changes or any area that required débridement for the above reasons. In addition, a wound that was intraoperatively determined to be fat necrosis was included and documented as fat necrosis. Complications resulting in unanticipated readmission or return to the operating room were documented as major complications. All others were documented as minor complications. For the purpose of analyzing the surgical technique in this patient population, a patient with bilateral breast cancer was treated as two patients, but the complications were only counted once. For all other patients, the complications were reported per patient for both breasts.
Tumor removal and partial reconstruction was performed by a two-team approach. In addition, positive margins requiring reexcision were also performed with both the breast surgeon and the plastic surgeon. The nipple pedicle, tumor location, and specimen size were documented. Operative reports were reviewed and patients were divided into three groups based on the technique used for reconstruction.
- Group 1: Patients who had tumor resection and reconstruction using the regular oncoplastic reduction technique (control). The defect was filled with the nipple pedicle and or the residual breast tissue (Fig. 1).
- Group 2: Patients who had the defect filled with an extended nipple pedicle (extended autoaugmentation pedicle). The defect was filled by extending the primary nipple-areola pedicle and rotating this into the defect (Fig. 2).
- Group 3: Patients who had the partial mastectomy defect filled with a secondary dermoglandular pedicle (secondary autoaugmentation pedicle). The nipple pedicle was used to reposition the nipple and a secondary independent dermoglandular pedicle was used to fill the defect (Fig. 3).
Comparisons were made between groups to determine (1) which breast size and tumor location required what reconstructive technique, and (2) whether the autoaugmentation techniques were associated with an increased risk of complications.
Descriptive statistics were calculated for patients in all groups. As appropriate, the independent-samples t test, Mann-Whitney U test, Fisher’s exact test, and chi-square test were used to determine the association between groups and the clinical variable of interest. A value of p < 0.05 was used to determine significance. Logistic regression analysis was performed to predict the development of postoperative complications when controlling for oncoplastic surgical technique. Additional clinical and demographic variables were included in the model from variables that were found to be significantly different between the three study groups on univariate analysis. A model for predicting the development of complications collectively was performed in addition to models for fat necrosis and delayed wound healing individually. Within the model, traditional oncoplastic breast surgery was used as the reference variable for comparison of extended and secondary pedicle groups, as the purpose of the analysis was to determine whether autoaugmentation resulted in differing incidences of complications. All statistical analysis was conducted using IBM SPSS Version 24.0 (IBM Corp., Armonk, N.Y.).
There were 333 patients included. The mean patient age was 54 years (range, 21 to 80 years), and the average body mass index for all patients was 33.7 kg/m2 (range, 19.6 to 60.2 kg/m2). There were 222 patients in the traditional oncoplastic reduction group, 51 patients who had an extended pedicle, and 60 patients in the secondary pedicle group. Demographics and characteristics for the three groups are listed in Table 1. Patients in the secondary pedicle group were significantly older (mean age, 59.2 years; p < 0.001). The extended pedicle group had a significantly lower body mass index (mean body mass index, 30.2; p = 0.002) compared with the other two groups. More patients in the extended pedicle group underwent preoperative chemotherapy (36.7 percent; p = 0.028). There were no significant differences in the other demographic variables.
The majority of patients in all groups underwent a contralateral procedure. The superomedial nipple pedicle technique was the most common for all three groups (Table 2). The lumpectomy excision weight was smallest in the extended pedicle group (136 g) and largest in the regular oncoplastic group (235 g; p = 0.017). The weight of tissue removed from the contralateral side was significantly less in the extended pedicle group (350 g; p < 0.001) compared with the other two groups. Tumor characteristics, positive margin rate, and axillary procedures were similar among all groups.
An autoaugmentation technique was used in 33 percent of patients (n = 111). When an extended pedicle was used, the most common extended pedicle was superomedial (68.6 percent), which was most commonly used for lateral [n = 26 of 51 (51 percent)] and superior [n = 10 of 51 (20 percent)] (p = 0.053) defects (Tables 3 and 4). When a secondary pedicle was used, the most common combination was inferior or inferolateral secondary pedicle [n = 54 of 60 (90 percent)] with a superior or superomedial primary nipple pedicle. The most common defect location for using a secondary pedicle was lateral [n = 42 of 60 (70 percent)] (p = 0.05), and this was filled with an inferiorly based secondary pedicle 95 percent of the time.
Location of Defect
Overall, the most common locations to require autoaugmentation techniques were lateral and medial (Table 4). Forty percent of all lateral defects were reconstructed using autoaugmentation flaps. When used, the most common autoaugmentation technique for lateral defects was the secondary pedicle, which was used in 63 percent (n = 42 of 67) of cases. Seventy-two percent of all the secondary pedicles performed were used for lateral defects (p = 0.050). For these lateral defects, the secondary pedicle of choice was inferior [n = 13 of 28 (46.4 percent)] and inferolateral [n = 13 of 28 (46.4 percent)]. Of those receiving an inferior/inferolateral secondary pedicle, the majority were in combination with a superomedial primary nipple pedicle (70 percent).
Thirty-nine percent of medial defects were reconstructed using autoaugmentation flaps. The most common autoaugmentation technique for medial defects was the extended pedicle [n = 13 of 22 (59 percent)]. The most common extended pedicle used for all defect locations was a superomedial (p < 0.0001). Of the superolateral defects requiring an extended pedicle, 100 percent were filled using a superomedial extended pedicle, and 85.7 percent of inferolateral/lateral defects were filled using the superomedial pedicle.
To provide a consistent estimate of breast size, the contralateral breast specimen reduction specimen weight was used as a measurement of the ipsilateral breast size. The average breast size in women that required autoaugmentation techniques was smaller than those that did not (451 g versus 654 g; p < 0.0001). In addition, when an extended pedicle was used, these women were more likely to have smaller breasts compared with the no-autoaugmentation technique (341 g versus 633 g; p < 0.0001) (Table 5). When the contralateral specimen weight was broken down into two groups based on weight, there were 85 patients with smaller breasts (<300 g) and 218 patients with large breasts (>300 g). When an extended pedicle was used, 56 percent of the time it was in patients with small breasts. When a secondary pedicle was used, 79 percent of the time that was in women with large breasts.
The overall complication rate was 16.9 percent (Table 6). The were no significant differences in the incidence of complications between groups: 15.5 percent incidence in the regular oncoplastic group, 19.6 percent in the extended pedicle group, and 20 percent in the secondary pedicle group. Delayed wound healing was the most common complication for both the autoaugmentation group (10.5 percent of complications) and those not undergoing autoaugmentation (9.3 percent). The incidence of nipple necrosis was 1.4 percent for the entire cohort, which was similar among groups (p = 0.679). In addition, the incidence of fat necrosis was similar among all three groups (p = 0.378). There were no significant differences in the other complications, such as delayed wound healing, infection, skin necrosis, hematoma, and seroma.
Although there was not a significant difference between the techniques of oncoplastic surgery and any of the documented complications on univariate analysis, we performed multivariate analysis to elucidate the potential influence of confounding variables. Despite having a low incidence, fat necrosis and delayed healing were specifically chosen for further analysis, as these complications are of particular interest in this patient population. On binomial logistic regression analysis, the technique of oncoplastic breast surgery (traditional oncoplastic versus extended versus secondary pedicle) was not predictive of developing a postoperative complication collectively. An additional model for variables potentially predictive of delayed wound healing showed that when controlling for oncoplastic surgical technique, age, body mass index, and resection weight of over 300 g, only increasing body mass index was predictive of developing delayed healing (OR, 1.07; 95 percent CI, 1.02 to 1.12). When controlling for these same variables, there was not a relationship between oncoplastic technique and development of fat necrosis (Table 7).
The mean follow-up was 609 days (range, 36 to 5509 days). Although all groups had sufficient follow-up, the oncoplastic group had a significantly longer follow-up time (688 days) compared with the secondary (405 days) and extended (505 days) pedicle groups (p = 0.040). The recurrence rate was 5.1 percent (n = 13 of 242) for the entire cohort, and the rebiopsy rate was 18.9 percent (n = 63 of 333), which was similar among groups (p = 0.627). The rate of completion mastectomy was 10.2 percent (n = 34 of 333), which was also similar among all three groups (p = 0.121).
Oncoplastic techniques have gained popularity. Along with this trend, we have seen newer techniques that broaden the indications further for this approach.15 We have shown that the autoaugmentation technique is safe and effective in its ability to fill larger and more difficult defects without increased risks of complications. Whenever possible, it is often easiest to fill the lumpectomy defect with reduction or mastopexy techniques using either the planned nipple pedicle or tissue that is left behind following the mammaplasty. For example, reconstructing a lower pole defect using a Wise pattern superomedial pedicle reduction does not require additional glandular manipulation. The augmentation techniques described here are useful when the defect is too large, too remote, or there is not enough nearby parenchyma available to fill the tumor cavity. We found that the extended autoaugmentation technique is useful for medial, lateral, or upper defects, especially in women with smaller breasts. The secondary pedicle is often used for lateral or medial defects.
It is important that these glandular flaps and extension flaps are well vascularized, because any fat necrosis in this patient population is always reason for concern. Poorly vascularized tissue does not tolerate radiation therapy well and may result in a more fibrotic, painful, or firm breast. Careful attention to flap length and width, minimizing the amount of pedicle undermining, and deepithelializing more when concerned, will improve flap perfusion. The pedicle is often kept wide and attached to the chest wall. Back-cuts are performed only as much as required to fill the defect or reshape the mound. The two-pedicle approach was used also as another option when it seemed as though the extended pedicle was getting long and less reliable. In an upper lateral defect, the secondary pedicle is preferred. The superomedial pedicle could be used for the nipple and a shorter inferiorly based lateral pedicle used to fill the defect. This not only minimizes pedicle length but allows independent movement because the extended nipple pedicle rotation is limited to where the nipple needs to go, and the attached extension will often not fill the proposed defect. It is always helpful to think about filling the defect and possibly needing autoaugmentation before resecting tissue. Excess tissue can be easily resected from the secondary pedicle if less volume is required.
The main variables used to determine when and what type of autoaugmentation was used during oncoplastic surgery were the size of the breast, defect size, and tumor location (Fig. 4). Women with smaller breasts who have small- or medium-size defects in the medial or lateral locations that are not adequately filled with surrounding breast tissue are best treated with the extended pedicle (Fig. 5). The tissue that is typically removed when creating the medial and lateral pillars in a vertical breast reduction or mastopexy is preserved as an extension to the superiorly based pedicle. It is rotated with the nipple to the proposed location and used to fill the defect. The medial and lateral glandular pillars are then plicated in the usual manner. This tissue can also be used to fill superior defects if necessary. Larger, lateral, or upper outer quadrant defects in women with larger breasts, however, often require two pedicles (Fig. 6). Once the defect is examined, a superomedial pedicle is created and the nipple is rotated into the desired position. An inferiorly based lateral dermoglandular pedicle is deepithelialized and created based on how much is anticipated to fill the defect. The residual dermoglandular tissue is then resected and the breast mound shaped in the usual fashion. The secondary pedicle can then independently be cut to size and placed into the defect.
The safety of the oncoplastic technique has remained under scrutiny. Overall complication rates remain relatively low and are often managed with conservative treatment. Therefore, their use should not delay initiation of adjuvant treatment. Studies have shown fewer complications in obese women and women with macromastia following oncoplastic reduction compared with mastectomy and immediate reconstruction.1 , 16 Tong et al. demonstrated fewer complications requiring additional surgery (3.8 percent versus 28 percent) and delaying adjuvant therapy (0.8 percent versus 14 percent) in the oncoplastic group for obese patients.16 In a previous report, we have similarly shown that patients with macromastia following oncoplastic reconstruction had a lower breast complication rate (22 percent versus 47 percent), shorter hospital stay (0.8 days versus 3.5 days), and fewer trips to the operating room (1.2 versus 2.7).1 The complication benefits compared with mastectomy are significant, and those compared with breast conservation therapy alone are acceptable. Furthermore, we have shown that complications in patients requiring autoaugmentation flaps are similar to those of traditional oncoplastic reduction techniques (20 percent versus 15 percent). Screening for breast cancer is always important. Studies have shown similar sensitivities for screening following oncoplastic reductions17–19 with traditional breast conservation therapy. Although fat necrosis is not higher in the autoaugmentation group, it will be important to demonstrate radiographically that no differences exist. The possible change in mammographic findings was not an objective of this study, but in the authors’ experience, provided that these flaps remain well vascularized and appropriately designed, screening should not be a problem. Future studies would be necessary to evaluate whether there is a difference in mammographic changes between traditional oncoplastic surgery and those with autoaugmentation flaps.
The majority of patients do not require an autoaugmentation technique. It is used more often in smaller breasts, or in larger defects, especially in cosmetically sensitive areas such as the medial quadrant and the upper lateral quadrant. The incidence of fat necrosis and wound healing complications is not greater with flaps, and the need for additional surgery or biopsy is similar. With careful attention to flap design and blood supply, the use of autoaugmentation techniques will broaden the indications for the oncoplastic approach and likely improve results, because defects are being filled with vascularized glandular tissue, without increasing complications.
1. Losken A, Pinell XA, Eskenazi B. The benefits of partial versus total breast reconstruction for women with macromastia. Plast Reconstr Surg. 2010;125:1051–1056.
2. Clough KB, Kroll SS, Audretsch W. An approach to the repair of partial mastectomy defects. Plast Reconstr Surg. 1999;104:409–420.
3. Losken A, Hamdi M. Partial breast reconstructive: Current perspectives. Plast Reconstr Surg. 2009;124:722–736.
4. Fitoussi AD, Berry MG, Famà F, et al. Oncoplastic breast surgery for cancer: Analysis of 540 consecutive cases [outcomes article]. Plast Reconstr Surg. 2010;125:454–462.
5. Hart M, Pinell-White XA, Egro F, Losken A. The psychosexual impact of partial and total breast reconstruction: A prospective one year longitudinal study. Ann Plast Surg. 2015;75:281–286.
6. Veiga DF, Veiga-Filho J, Ribeiro LM, et al. Quality-of-life and self-esteem outcomes after oncoplastic breast-conserving surgery. Plast Reconstr Surg. 2010;125:811–817.
7. Regaño S, Hernanz F, Ortega E, Redondo-Figuero C, Gómez-Fleitas M. Oncoplastic techniques extend breast-conserving surgery to patients with neoadjuvant chemotherapy response unfit for conventional techniques. World J Surg. 2009;33:2082–2086.
8. Silverstein MJ, Savalia N, Khan S, Ryan J. Extreme oncoplasty: Breast conservation for patients who need mastectomy. Breast J. 2015;21:52–59.
9. Cracco S, Semprini G, Cattin F, et al. Impact of intraoperative radiotherapy on cosmetic outcome and complications after oncoplastic breast surgery. Breast J. 2015;21:285–290.
10. Lansu JT, Essers M, Voogd AC, et al. The influence of simultaneous integrated boost, hypofractionation and oncoplastic surgery on cosmetic outcome and PROMs after breast conserving therapy. Eur J Surg Oncol. 2015;41:1411–1416.
11. Losken A. Breast reshaping following massive weight loss: Principles and techniques. Plast Reconstr Surg. 2010;126:1075–1085.
12. Kwei S, Borud LJ, Lee BT. Mastopexy with autologous augmentation after massive weight loss: The intercostal artery perforator (ICAP) flap. Ann Plast Surg. 2006;57:361–365.
13. Hurwitz DJ. Single-staged total body lift after massive weight loss. Ann Plast Surg. 2004;52:435–441; discussion 441.
14. Rubin JP, Khachi G. Mastopexy after massive weight loss: Dermal suspension and selective auto-augmentation. Clin Plast Surg. 2008;35:123–129.
15. Losken A, Duggal CS, Styblo TM, Carlson GW. A meta-analysis comparing breast conservation therapy alone to the oncoplastic technique. Ann Plast Surg. 2014;72:145–149.
16. Tong WM, Baumann DP, Villa MT, et al. Obese women experience fewer complications after oncoplastic breast repair following partial mastectomy than after immediate total breast reconstruction. Plast Reconstr Surg. 2016;137:777–791.
17. Losken A, Schaefer TG, Newell M, Styblo TM. The impact of partial breast reconstruction using reduction techniques on postoperative cancer surveillance. Plast Reconstr Surg. 2009;124:9–17.
18. Dolan R, Patel M, Weiler-Mithoff E, et al. Imaging results following oncoplastic and standard breast conserving surgery. Breast Care (Basel) 2015;10:325–329.
19. Piper M, Peled AW, Sbitany H, Foster RD, Esserman LJ, Price ER. Comparison of mammographic findings following oncoplastic mammoplasty and lumpectomy without reconstruction. Ann Surg Oncol. 2016;23:65–71.