Multiple perforator flap bilateral breast reconstruction is a contemporary option in highly selected patients with minimal donor abdominal tissue and a desire to avoid use of prosthetic implants.1–3 It can supply additional skin to help create a breast envelope with more natural ptosis. It can also provide additional volume to create a body-appropriate breast mound while minimizing the need for additional fat grafting procedures, the outcomes of which are inherently unpredictable.4–6 Using four flaps for the reconstruction of two breasts (bilateral stacked flap breast reconstruction) has recently become feasible with the advancement of microsurgical techniques,7–10 more experience with alternative perforator flaps,11–15 and use of co-surgery.16
Using multiple flaps to reconstruct a single breast is not a new concept. Historically, the most commonly performed configuration involved using double-pedicled or “stacked” free transverse rectus abdominis musculocutaneous (TRAM) or deep inferior epigastric perforator (DIEP) flaps.7–9 More recently, with the development of thigh-based perforator flaps,11,13–15,17,18 experience has grown with use of multiple transverse upper gracilis or profunda artery perforator flaps for single breast reconstruction in patients with inadequate abdominal tissue and/or prior abdominal surgery.10,19 These experiences have confirmed that the split craniocaudal internal mammary system can support two flaps independently,20,21 without the need to sacrifice common secondary bail-out recipient options such as the subscapular system.
With the increased incidence of contralateral and bilateral prophylactic mastectomies in younger, lean, and otherwise healthy patients,22 the demand for alternative options for autologous breast reconstruction has increased. As a result, several groups including our own have begun performing bilateral stacked flap breast reconstruction,1–3,23 and in this study, we describe our experience, including evolution in technique.
PATIENTS AND METHODS
From January of 2014 to September of 2018, the senior co-surgeons (N.T.H. and S.S.T.) performed 50 consecutive bilateral stacked flap operations at a single institution, and this report describes their entire experience. All data were acquired from retrospective chart review following approval by the institutional review board. Data were collected regarding patient age, body mass index, history of cancer, radiation therapy or chemotherapy, length of surgery, and hospital length of stay. Furthermore, technical data were reviewed, which included timing of operation, use of Cook implantable Doppler or intraoperative SPY angiography (Stryker Corp./Novadaq Technologies, Kalamazoo, Mich.) flap weight, number of perforators and perforator location in flaps, size of dominant perforators, pedicle length, average venous coupler sizes, and flap inset and microanastomotic configuration. Any measurements such as perforator size or pedicle length were performed manually using a ruler during the flap harvest. Finally, we reviewed all flap losses and operating room take-backs for flap or any problem, in addition to late non–flap-related complications. From these data, we comment on the evolution in our practice based primarily on the lessons learned. Any statistical comparison between subgroups was performed with nonparametric Mann-Whitney U testing using GraphPad Prism (GraphPad Software, Inc., La Jolla, Calif.).
Patient Selection and the Operation
Candidates for breast reconstruction were offered a bilateral stacked flap operation when they had undergone or were planning to undergo bilateral mastectomies with or without immediate tissue expander placement, desired complete autologous reconstruction (without use of implants), and had inadequate abdominal or thigh tissue that if used alone could not create body-appropriate–size breasts or place undue deformity on a single donor site in an attempt to harvest more tissue or volume. Additional considerations were that they were otherwise healthy with no significant comorbidities and no abdominal or thigh surgery history that would jeopardize perforators to the planned flaps. Patients were always counseled on the alternatives, which included total implant-based reconstruction, a pedicled latissimus dorsi flap with implant reconstruction, or single flap reconstruction with serial staged fat grafting. The main limitations of the alternatives involve prospective complications that could arise from using prosthetic devices (e.g., capsular contracture, rupture, lifetime need for additional exchange procedures, infection) and/or the need for multiple additional fat grafting procedures.4–6,24,25 Patients were always counseled on the risks, which involved discussion of the additional surgical donor sites (thighs) and the risk of wound, seroma, hematoma, infection, and the rarity of such operations.
All bilateral stacked flap breast reconstructions were performed by two co-surgeons (N.T.H. and S.S.T.). It is important to note that this operation is likely not feasible in a reasonable duration with a single surgeon. In our practice, the operation involves harvesting bilateral DIEP and/or muscle-sparing TRAM flaps and bilateral profunda artery perforator flaps from the thighs. Positioning is supine, with frog-legging to allow for simultaneous harvest of DIEP and profunda artery perforator flaps. Although our sequencing has evolved, typically one surgeon will begin by preparing a breast pocket and exposing internal mammary vessels on one side (often the left first) while the second surgeon begins with the abdominal flap dissection. Usually, the breast pocket and vessel preparation is completed in less than 30 to 45 minutes and, once completed, the simultaneous profunda artery perforator flap dissection can be started. Usually, both the DIEP and profunda artery perforator flaps planned for one breast are prepared before microsurgery. Of note, we have observed that the profunda artery perforator flap’s elliptical shape can be used to recreate the inferior pole sling of the reconstructed breast; however, just the same, it can be used to provide superior pole fullness and, when beveled, tapers well onto the native chest wall.
Most commonly, the larger flap is made ischemic first and transferred to the chest, and microsurgical anastomosis is performed with the cranial internal mammary vessels (Fig. 1). Once perfused, the secondary smaller flap is made ischemic and transferred to the chest. and microsurgical anastomosis is performed with the caudal internal mammary vessels or the continuance of or side branch off the pedicle of the primary flap. Once both flaps are perfused, the breast shaping commences. It is common for the DIEP and profunda artery perforator flaps to be deepithelialized and sewn to each other to reduce the chance of a palpable crevasse between the flaps. The amount of skin deepithelialization varies widely and depends on several factors, including extent of radiation skin damage with need for native skin resection, presence or absence of prior envelope expansion, and reconstructive goals. An advantage of the bilateral stacked flap breast reconstruction is that in addition to volume, more skin can be transferred to the chest for more optimal recreation of natural breast ptosis. Once the first breast is transferred successfully, shaped, and inset, attention can be given to the reconstruction of the second breast in a similar manner. We typically perform the second profunda artery perforator flap harvest next, which allows for closure of the thigh donor sites before flexing the abdomen. In addition, it allows for placement of an additional warming device to help with patient temperature regulation. All incisions are often dressed using cyanoacrylate skin adhesive except for the newly reconstructed breast incisions, which are dressed with Xeroform (Covidien, Mansfield, Mass.) and antibiotic ointment.
From January of 2014 to September of 2018, a total of 50 patients underwent bilateral stacked flap breast reconstructions. Mean age at the time of flap reconstruction was 48.6 ± 8.6 years (Table 1). Mean body mass index was 27.1 ± 3.4 kg/m2. Mean operative time was 10 ± 1.5 hours and mean hospital length of stay was 4.3 ± 0.9 days. Only three patients (6 percent of total) were diabetics, and 16 (32 percent of total) had a smoking history. The majority [n = 45 (90 percent)] of patients had a history of breast cancer, with the remainder electing for breast reconstruction following prophylactic mastectomy for BRCA or other high-risk mutations. Of those patients with a cancer history, 31 (62 percent) had radiation therapy. Nineteen (38 percent) underwent neoadjuvant and another 15 (30 percent) had adjuvant chemotherapy.
Table 1. -
|Total no. of patients
|Total no. of flaps
|Mean age ± SD, yr
||48.6 ± 8.6
|Mean BMI ± SD, kg/cm2
||27.1 ± 3.4
|Mean operative time ± SD, hr
||10.0 ± 1.5
|Mean hospital LOS ± SD, days
||4.3 ± 0.9
BMI, body mass index; LOS, length of stay; XRT, radiation therapy.
In terms of timing of definitive reconstruction, the majority [n = 42 (84 percent)] of patients underwent the bilateral stacked flap operation following removal of tissue expanders that were placed immediately at the time of mastectomy (“delayed-immediate”) (Table 2). Eight patients (16 percent) underwent “delayed” bilateral stacked flap breast reconstruction and one patient underwent an immediate bilateral stacked flap breast reconstruction. Cook implantable Doppler probes were used on 130 flaps (65 percent), whereas intraoperative SPY angiography was used in 135 flaps (68 percent). SPY was used in the majority of cases to determine which part(s) of the flaps should be trimmed before inset.
Table 2. -
|Use of Cook implantable Doppler
|Use of SPY angiography
|Flap mass, g
| Mean ± SD
||443.2 ± 13.3
| Mean ± SD
||396.0 ± 11.3
DIEP, deep inferior epigastric perforator; PAP, profunda artery perforator.
The mean flap mass was 443.2 ± 133.0 g (range, 233 to 753 g) for DIEP flaps and 369.5 ± 112.5 g (range, 183 to 660 g) for profunda artery perforator flaps. The majority of DIEP flaps were single-perforator (48 percent) or two-perforators (40 percent) flaps. The profunda artery perforator flaps had one perforator (43.6 percent), two perforators (47.9 percent), or three perforators (8.5 percent) and were 1.5 mm 74.7 percent of the time. Mean pedicle length was 12.5 cm (range, 8 to 15 cm) for the DIEP flaps and 11.4 cm (range, 5 to 16 cm) for the profunda artery perforator flaps.
Relative to flap inset configuration, a majority of breast reconstructions (94 percent) were performed with one DIEP flap and one profunda artery perforator flap. In general, the larger of the DIEP or profunda artery perforator flap was placed inferiorly (in 66 percent of cases). In three patients, bilateral breasts were reconstructed using a double-pedicled DIEP flap on one side and stacked profunda artery perforator flaps on the other. A majority of flap anastomoses were performed to the craniocaudal internal mammary vessels (92 percent of all), with an exception of 16 flaps whose recipient vessels were a continuation or side branch of the other flap.
A total of five flaps (of 200) were lost, which equates to a 2.5 percent flap loss rate affecting 10 percent of our patients (Table 3). A total of seven operating room take-backs for a flap problem occurred within the first 2 postoperative weeks from the original flap surgery (3.5 percent rate on a per-flap basis, affecting 14 percent of our patients); of these, four resulted in flap loss, two resulted in negative explorations, and one resulted in flap salvage. In addition, there were three other early operating room take-backs for evacuation of a hematoma. There was a total of two cases with clinically significant fat necrosis requiring operative débridement.
Table 3. -
Unplanned Early Flap and Non–Flap-Related Complications
|Operating room take-back for flap problem†
| Flap loss
| Negative exploration
| Flap salvage
| Breast hematoma
|Operating room take-back for any problem†
| Thigh hematoma
†Within 2 weeks of original four-flap surgery.
The most common late (defined as after the first 2 postoperative weeks) non–flap-related complications were thigh wounds (17 total, eight requiring a procedure), umbilical (eight total, two requiring a procedure) or abdominal (seven total, or five requiring a procedure) wounds, breast wounds (n = 4), breast hematoma (n = 3), and thigh seroma (n = 3) (Table 4). There was one case of deep venous thrombosis with pulmonary embolus and no instances of death attributable to the operation. There was a total of 28 patients with a postoperative wound healing problem. Of these, 15 patients required either an office- or operating room–based procedure. Seven of these patients were obese (body mass index > 30 kg/m2), five had a smoking history, and two were diabetic. Mean body mass index for patients with a thigh, umbilical, and/or abdominal wound was 27.6 ± 3.4 compared with 26.7 ± 3.5 for patients without a wound (p = 0.7, Mann-Whitney U test). Of the patients with thigh wounds, one required operative débridement and reclosure and seven were treated temporarily with a vacuum-assisted closure dressing. Of the patients with an umbilical wound, one required operative débridement and another required negative-pressure wound therapy with a vacuum-assisted closure dressing. Of the patients with an abdominal wound, two required operative débridement and vacuum-assisted closure dressing placement, whereas a total of three patients required some duration of negative-pressure wound therapy. One patient had both umbilical and abdominal wounds needing additional and more extensive treatment (either surgery and/or negative-pressure wound therapy).
Table 4. -
Late Non–Flap-Related Complications
†Two patients had bilateral thigh wounds.
Figures 2 and 3 are example photographs of bilateral stacked flap breast reconstructions. Figure 4 illustrates representative long-term profunda artery perforator donor-site scar and posteromedial thigh contour.
Multiple flap breast reconstruction with four perforator flaps for two breasts can be performed safely and is indicated in selected patients with minimal abdominal tissue seeking reconstruction without the use of implants. Our experience with bilateral stacked flap reconstruction involves 50 consecutive patients, 200 flaps, five flap losses, and a reasonably acceptable complication profile (both donor and recipient sites) over an almost 5-year period starting in early 2014. The main goals of this operation include transfer of volume for creation of body-appropriate breast mounds while avoiding use of prostheses and transfer of enough skin to create more naturally appearing, ptotic breasts. In our opinion, these goals cannot always be met in a single operation using implants, flap with implants, or a single flap with serial fat grafting. Furthermore, in our experience and when indicated, the bilateral stacked flap operation does provide the most optimal substrate for breast shaping and thus has a higher potential for achieving a superior aesthetic result compared with alternatives. Even though any combination of perforator flaps could be used in the bilateral stacked flap operation, our preference is to use the combination of a DIEP or muscle-sparing TRAM flap with a profunda artery perforator flap. This practice combines the gold standard abdominal perforator flap with our preference of a reliable secondary perforator flap containing substantial additional skin and volume, long pedicle, associated with an inconspicuous scar, and without a need for position change.11,13–15,18,23 Additional perforator flap options could include (but are not limited to) superior or inferior gluteal artery perforator, transverse upper gracilis or medial circumflex femoral perforator, lateral thigh perforator, and lumbar artery perforator flaps.23,26,27 We have more recently begun gaining experience with the lumbar artery perforator flap in these multiple flap operations. Of note, approximately 28 percent of our breast reconstructive practice thus far has used multiple flaps (for either unilateral or bilateral cases), and approximately 12 to 13 percent have been bilateral stacked flap reconstructions.
In our series of 50 bilateral stacked flap breast reconstructions, we have experienced five flap losses, which is equivalent to a 2.5 percent rate, slightly higher than our experience with conventional abdominally based perforator flap–only breast reconstructions (approximately 1 to 2 percent rate). We attribute at least some of the slightly higher rate to a steep learning curve. We also acknowledge that a 10 percent rate of failure on a per-patient basis is relatively high, and thus this operation should be offered to only a highly selected patient population following discussion of risks and benefits. The first flap loss was our fifth operation, and we had three flap losses within the first 20 patients. Of the five flap losses, one was attributable to arterial thrombosis with multiple failed revision attempts, including change of recipient from internal mammary vessels to interflap anastomoses, with no success. The other four were attributable to venous thromboses, including one with delayed presentation to the emergency room 6 days after the initial operation. With each flap loss, additional technical nuances have been developed regarding the pedicle and flap positioning to better avoid vessel kinking and decision-making regarding choice of secondary flap inflow. For example, although new data suggest that caudal internal mammary vessels provide sufficient retrograde perfusion pressure to support a secondary flap,20,21 there are instances in our experience when either the caudal vessels are diminutive or the blood flow on spurt test is minimal, such that interflap microanastomoses are preferable and thought to less likely result in flap failure. In fact, a total of eight secondary flaps (or 4 percent of all flaps) in this series have been successfully reperfused by connecting to the primary flap by means of a continuation of either its main pedicle or a side branch. Despite these adaptations in our practice, additional studies must be performed to better characterize prospective recipient vessel perfusion pressures to avoid inflow-related problems with a secondary flap. Larger secondary flaps may have higher resistances to blood flow, requiring higher perfusion pressures, but this has never been studied, to our knowledge.
Non–flap-related complications including those requiring additional operative intervention were fortunately not as common. There was one incident of deep venous thrombosis/pulmonary embolism 3 months after surgery. There were only one thigh and two abdominal or umbilical wounds requiring additional surgical débridement. There was a total of 17 thigh wounds, including two patients with bilateral wounds. Most of these wounds were managed with local wound care. These wounds are a known additional risk associated with this operation compared with a conventional DIEP or muscle-sparing TRAM flap reconstruction because of the additional thigh donor site and the possible addition of complexity to the postoperative recovery. As mentioned previously, 28 patients overall had either a thigh, abdominal, or umbilical wound. Although there are data suggesting that active smoking is associated with increased delayed wound healing issues at the donor site,28 only five of the 28 patients (17.9 percent) with wounds had a smoking history, which is a smaller proportion than that of our entire cohort (where 32 percent had a smoking history). Furthermore, only six patients were obese (body mass index >30 kg/m2). Our patients had an overall average body mass index of 27.4 kg/m2, and there was no difference in the body mass index of patients with postoperative thigh, umbilical, or abdominal wounds compared to those without. In our practice, we do not have a strict body mass index cutoff, although obese patients are counseled regarding their increased risk of, for example, delayed wound healing issues. Although additional studies are needed to define risk factors associated with the development of umbilical, abdominal, and thigh wounds in bilateral stacked flap candidates, provided that patients are informed of the possibility of wound complications, their risk appears acceptable given the potential benefits of the operation. Lastly, stacked flap reconstructions in our practice have required fewer secondary revision procedures,29 including fat grafting. Further study is needed to characterize the reduction of need for revision surgery in this particular bilateral stacked flap cohort.
Our early experience with bilateral stacked flap breast reconstruction suggests that it is a highly specialized operation that may be suitable for selected younger patients seeking reconstruction in the setting of minimal donor abdominal tissue and a desire for avoidance of implants. Multiple perforator flaps for use in each reconstructed breast enable transfer of additional volume and skin to create an aesthetically appealing, ptotic breast shape in a single stage. Additional studies will need to focus on better understanding of, in particular, secondary flap inflow perfusion pressures and lifetime cost savings associated with this procedure compared with the alternatives. With increased experience, various other donor sites can be incorporated that fit a patient’s unique body habitus. In addition, having various options for donor sites permits an innovative approach that not only selects patients based on ideal perforator dominance but also attempts to reduce donor-site morbidity with the aim of improving an overall aesthetic outcome of the breast and the entire trunk.
1. Mayo JL, Allen RJ, Sadeghi A. Four-flap breast reconstruction: Bilateral stacked DIEP and PAP flaps. Plast Reconstr Surg Glob Open. 2015;3:e383.
2. Rozen WM, Patel NG, Craggs BS, Ramakrishnan VV. Four-flap breast reconstruction with bilateral stacked flaps. Plast Reconstr Surg. 2016;137:492e–493e.
3. Rozen WM, Patel NG, Ramakrishnan VV. Increasing options in autologous microsurgical breast reconstruction: Four free flaps for ‘stacked’ bilateral breast reconstruction. Gland Surg. 2016;5:255–260.
4. Suszynski TM, Sieber DA, Van Beek AL, Cunningham BL. Characterization of adipose tissue for autologous fat grafting. Aesthet Surg J. 2015;35:194–203.
5. Suszynski TM, Sieber DA, Mueller K, Van Beek AL, Cunningham BL, Kenkel JM. Characterization of adipose tissue product quality using measurements of oxygen consumption rate. Aesthet Surg J. 2018;38:442–447.
6. Suszynski TM, Sieber DA, Cunningham BL, Van Beek AL. Implications of oxygenation in fat grafting. Plast Reconstr Surg. 2014;133:731e–733e.
7. Wagner DS, Michelow BJ, Hartrampf CR Jr. Double-pedicle TRAM flap for unilateral breast reconstruction. Plast Reconstr Surg. 1991;88:987–997.
8. Salgarello M, Barone-Adesi L, Sturla M, Masetti R, Mu L. Needing a large DIEAP flap for unilateral breast reconstruction: Double-pedicle flap and unipedicle flap with additional venous discharge. Microsurgery. 2010;30:111–117.
9. Murray A, Wasiak J, Rozen WM, Ferris S, Grinsell D. Stacked abdominal flap for unilateral breast reconstruction. J Reconstr Microsurg. 2015;31:179–186.
10. Buescher T, Przylecki WH, Holding J, et al. Evaluation of complications associated with flap/pedicle inset techniques for unilateral breast reconstruction utilizing dual transverse upper gracilis (TUG) or profunda artery perforator (PAP) flaps. Plast Reconstr Surg. 2015;136:90–91.
11. Haddock NT, Greaney P, Otterburn D, Levine S, Allen RJ. Predicting perforator location on preoperative imaging for the profunda artery perforator flap. Microsurgery. 2012;32:507–511.
12. Allen RJ, Tucker C Jr. Superior gluteal artery perforator free flap for breast reconstruction. Plast Reconstr Surg. 1995;95:1207–1212.
13. Wong C, Nagarkar P, Teotia S, Haddock NT. The profunda artery perforator flap: Investigating the perforasome using three-dimensional computed tomographic angiography. Plast Reconstr Surg. 2015;136:915–919.
14. Haddock N, Nagarkar P, Teotia SS. Versatility of the profunda artery perforator flap: Creative uses in breast reconstruction. Plast Reconstr Surg. 2017;139:606e–612e.
15. Allen RJ, Haddock NT, Ahn CY, Sadeghi A. Breast reconstruction with the profunda artery perforator flap. Plast Reconstr Surg. 2012;129:16e–23e.
16. Haddock NT, Kayfan S, Pezeshk RA, Teotia SS. Co-surgeons in breast reconstructive microsurgery: What do they bring to the table? Microsurgery. 2018;38:14–20.
17. Allen RJ, Levine JL, Granzow JW. The in-the-crease inferior gluteal artery perforator flap for breast reconstruction. Plast Reconstr Surg. 2006;118:333–339.
18. Haddock NT, Gassman A, Cho MJ, Teotia SS. 101 consecutive profunda artery perforator flaps in breast reconstruction: Lessons learned with our early experience. Plast Reconstr Surg. 2017;140:229–239.
19. Haddock NT, Cho MJ, Gassman A, Teotia SS. Stacked profunda artery perforator flap for breast reconstruction in failed or unavailable deep inferior epigastric perforator flap. Plast Reconstr Surg. 2019;143:488e–494e.
20. Tomioka YK, Uda H, Yoshimura K, Sunaga A, Kamochi H, Sugawara Y. Studying the blood pressures of antegrade and retrograde internal mammary vessels: Do they really work as recipient vessels? J Plast Reconstr Aesthet Surg. 2017;70:1391–1396.
21. Stalder MW, Lam J, Allen RJ, Sadeghi A. Using the retrograde internal mammary system for stacked perforator flap breast reconstruction: 71 breast reconstructions in 53 consecutive patients. Plast Reconstr Surg. 2016;137:265e–277e.
22. Wong SM, Freedman RA, Sagara Y, Aydogan F, Barry WT, Golshan M. Growing use of contralateral prophylactic mastectomy despite no improvement in long-term survival for invasive breast cancer. Ann Surg. 2017;265:581–589.
23. Suszynski TM, Teotia SS, Haddock NT. Multiflap autologous breast reconstruction with use of a thigh perforator flap can be considered large-volume vascularized fat grafting. Plast Reconstr Surg. 2018;142:243e–244e.
24. Nguyen A, Pasyk KA, Bouvier TN, Hassett CA, Argenta LC. Comparative study of survival of autologous adipose tissue taken and transplanted by different techniques. Plast Reconstr Surg. 1990;85:378–386; discussion 387–389.
25. Choi M, Small K, Levovitz C, Lee C, Fadl A, Karp NS. The volumetric analysis of fat graft survival in breast reconstruction. Plast Reconstr Surg. 2013;131:185–191.
26. Hamdi M, Craggs B, Brussaard C, Seidenstueker K, Hendrickx B, Zeltzer A. Lumbar artery perforator flap: An anatomical study using multidetector computed tomographic scan and surgical pearls for breast reconstruction. Plast Reconstr Surg. 2016;138:343–352.
27. Ciudad P, Maruccia M, Orfaniotis G, et al. The combined transverse upper gracilis and profunda artery perforator (TUGPAP) flap for breast reconstruction. Microsurgery. 2016;36:359–366.
28. Seidenstuecker K, Munder B, Mahajan AL, Richrath P, Behrendt P, Andree C. Morbidity of microsurgical breast reconstruction in patients with comorbid conditions. Plast Reconstr Surg. 2011;127:1086–1092.
29. Haddock NT, Cho MJ, Teotia SS. Comparative analysis of single versus stacked free flap breast reconstruction: A single-center experience. Plast Reconstr Surg. 2019;144:369e–377e.