CHAPTER 62 BREAST RECONSTRUCTION: FREE FLAP TECHNIQUES
MAURICE Y. NAHABEDIAN
The options for breast reconstruction following mastectomy continue to expand. Prosthetic devices and pedicled musculocutaneous flaps provide patients with good outcomes but have limitations that preclude their use in patients with a history of cigarette smoking or prior radiation, or in patients who are overweight or obese. The primary limitations of pedicled musculocutaneous flaps are that they usually require total sacrifice of the muscles, result in donor-site weakness, and sometimes yield abnormal contour. The primary limitations of prosthetic devices are that they are foreign materials and are subject to mechanical failure eventually.
Free tissue transfer overcomes the limitations described above but is associated with morbidity related to technical factors. Free flaps for breast reconstruction are generally considered in women with increased body mass index (BMI), a history of tobacco use, prior radiation therapy, and also in women who desire preservation of the donor-site muscles. These flaps are derived from a variety of donor sites that include the abdomen, gluteal region, and medial thigh. Free flaps can be muscle or perforator based.
The principal abdominal free flaps include the free transverse rectus abdominis musculocutaneous (TRAM), deep inferior epigastric artery perforator (DIEP), and superficial inferior epigastric artery (SIEA) flaps. Other donor sites for free flap breast reconstruction include the gluteal and medial thigh regions. Although these latter sites can be considered as primary donor sites, most surgeons consider these when the abdomen is not an option. The gluteal flaps include the inferior gluteal artery perforator (IGAP), superior gluteal artery perforator (SGAP), and the gluteal musculocutaneous flaps. The thigh flap to be reviewed is the transverse upper gracilis (TUG) flap or transverse musculocutaneous gracilis (TMG) flap. The anterolateral thigh and Reuben’s flap are seldom used for breast reconstruction and will not be reviewed in this chapter.
This chapter focuses on patient selection, anatomic considerations, harvesting techniques, and clinical outcomes. Other relevant topics include a review of algorithms, monitoring techniques, and the current technological advancements that have facilitated these complex operations.
PATIENT SELECTION AND FLAP SELECTION
Patient and flap selection includes a thorough history and physical examination, review of the reconstructive options, an understanding of patient expectations, and the surgeon’s recommendations.1,2Important details of the physical examination include body weight, patient height, BMI, and an estimate of breast volume requirements. The possibility of secondary operations involving the ipsilateral and/or contralateral breast is discussed. The variety of available donor sites allows appropriate volume to be transferred in most patients. The abdomen is the preferred donor site for most surgeons and patients if sufficient skin and fat are available given the desired breast volume. Most patients have had children and have some excess abdominal skin and fat. A slender woman with a paucity of abdominal fat may still be a candidate for an abdominal flap if the reconstructive requirements are low. Women who are overweight or obese may not be candidates for pedicled abdominal flaps but may be candidates for a free TRAM or DIEP flap. The flap, however, must be tailored to sustain adequate perfusion and minimize fat necrosis.3,4 The abdomen is usually not considered when a significant volume is required and midline scars preclude incorporating the contralateral zones.
If the abdomen is not suitable, then the gluteal or thigh region is considered. The SGAP or IGAP is considered in patients who desire autologous reconstruction, lack sufficient abdominal fat, refuse prosthetic reconstruction, and prefer a perforator flap.5,6 Most women will have sufficient donor fat in the gluteal area. An alternative to the gluteal flaps is the TUG flap.7-9 The principal limitation with all of these alternate flaps is that the volume is usually limited compared with the abdominal donor site. One advantage compared with the abdomen, however, is that these flaps are available on both sides of the body, permitting separate procedures if necessary.
Patient age, in and of itself, is not an indication or contraindication for any one operation. Advanced patient age (>65 years) may be a relative contraindication for microvascular breast reconstruction, but personal experience, and the existing literature, demonstrates that these techniques are safe and effective in properly selected patients in this age range.10 Women of advanced age are required to obtain medical clearance from their primary physicians. Women with multiple medical comorbidities are discouraged from pursuing complex microvascular procedures and directed toward simpler methods, such as prosthetic reconstruction.
The decision to use a perforator flap or a musculocutaneous flap can be difficult. This is especially true when considering the abdomen as the donor site. Some surgeons feel that all patients have a dominant abdominal wall perforator and that a DIEP flap can be performed in anyone. Others are of the opinion that a dominant perforator is not always present and a musculocutaneous flap is sometimes necessary. Several studies have described an algorithm for flap selection based on patient characteristics.2 Our original algorithm was based on breast volume, abdominal fat volume, perforator diameter, number of perforators, patient age, tobacco use, and whether the reconstruction was unilateral or bilateral. In general, a DIEP flap was preferentially performed when the volume requirement was less than 750 cc and the patient had mild to moderate excess abdominal fat. A free TRAM flap was preferentially performed when the volume requirement exceeded 1,000 cc or the patient had abundant abdominal fat. With increasing experience, this algorithm has been modified (Table 62.1).
The final component of the consultation includes a review of schematic illustrations and preoperative and postoperative photographs of other patients. Typically, the patient is shown a poor result, a good result, and an excellent result. Patients are informed of the potential complications including flap failure (0.5% to 4%), abnormal donor-site contour (0% to 20%), and return to the operating room (1% to 8%).
With traditional musculocutaneous free flaps, there is little need to assess the vascular architecture of the flap or donor site. However, with the introduction of perforator flaps, preoperative imaging is useful. Over the past decade, there have been a variety of technological advancements that facilitate localization of perforators.11-18 Preoperative imaging enables surgeons to identify suitable perforators and to determine the patency of primary source vessels, namely the inferior epigastric and internal mammary vessels. The modalities that are currently available include duplex and color duplex ultrasound, computerized tomographic angiography (CTA), and magnetic resonance angiography (MRA) (Table 62.2).
The first tool used for preoperative mapping was the Doppler ultrasound. Although there are many clinical applications for the Doppler, plastic surgeons were interested in the Doppler to map out perforating vessels throughout the cutaneous territory of a flap.11,19,20 There were several early studies utilizing the color Doppler that provided useful information related to the location, caliber, and flow patterns of the perforators in the planning of the TRAM flap.20 Cluster analyses demonstrated that perforators were located throughout the anterior abdominal wall with the majority of dominant perforators being situated in the periumbilical area.20 Perforators exceeding 2.2 mm were few in number but were identifiable in all four quadrants of the anterior abdominal wall.
Other benefits using Doppler included information regarding flow, direction, and velocity. In a study evaluating perfusion of the TRAM, DIEP, and SGAP flaps, it was determined that the highest blood flow and velocity was achieved in the TRAM flap followed by the DIEP and SGAP flaps.11 Specific flow measurements in various vessels were obtained and included in the deep inferior epigastric artery (10.45 mL/min), the superior gluteal artery (9.95 mL/min), and the internal mammary artery (IMA) (37.66 mL/min). The imaging could differentiate between venous and arterial signals.11 The principal limitation of the color duplex was that it could not provide three-dimensional or architectural detail of the perforator system. Giunta et al.19 reported a relatively high number of false-positive results (46%) using the hand-held Doppler for localization of perforators. In a comparative study evaluating Doppler ultrasound and CTA, Rozen et al.12 found that CTA was superior to Doppler based on visualization of the deep inferior epigastric artery (DIEA), its branching pattern, and the perforators.
Computerized Tomographic Angiography
Computerized tomography may represent the gold standard for preoperative imaging and was the first of the highly accurate methods of perforator assessment.13,16,17,21 Its use is primarily directed toward abdominal flaps but it can also be used in the gluteal and posterior thorax. Using multi-slice computerized tomography, axial and coronal images are obtained demonstrating the vascular architecture. The benefits of CTA include anatomic localization of the perforators, determination of the course of the perforator through the muscle, comparative assessment of the right and left vascular anatomy, and elucidation of anatomic detail of the medial and lateral row perforators. CTA can also provide information that may discourage a surgeon from performing a perforator flap and choose instead to perform a muscle-sparing (MS) free TRAM.
Clinical trials using CTA have been useful. Casey et al.17 have demonstrated that preoperative CTA has reduced operative times, increased the number of suitable perforators included in a flap, and reduced the incidence of a postoperative abdominal bulge. The latter is presumably related to the selection of medial rather than lateral row perforators minimizing intercostal nerve injury.Unfortunately, CTA has not reduced complications related to the anastomosis, flap failure rates, occurrence of fat necrosis, dehiscence, or delayed healing. CTA has also demonstrated benefit in the setting of prior abdominal surgery. Rozen et al.16 studied 58 patients who had a total of 96 abdominal scars with CTA to determine if there was any disruption to the perforators or the primary source vessels. Their findings were that paramedian incisions resulted in most damage to the perforator, SIEA, and DIEA vessels. On the contrary, laparoscopic incisions caused the least damage (Table 62.3).
Magnetic Resonance Angiography
MRA represents the next generation in vascular imaging in part because the imaging quality is maintained or enhanced without ionizing radiation.15,18,22,23 When compared with CTA, MRA has lower spatial resolution but greater contrast resolution.22 This enables MRA to detect very small perforators that might otherwise be missed on CTA. MRA enables surgeons to become aware of perforator location, size, and distance from the umbilicus. Clinical studies have provided useful information. Greenspun et al.18 reviewed the outcomes in 31 women (50 flaps) scheduled for DIEP flaps. All perforators visualized on MRA using a gadolinium-based contrast agent were found intraoperatively. In 100% of patients, the intraoperative location of each perforator was within 1 cm of that predicted using MRA. In three flaps, the DIEA perforators were small and the SIEA system was relatively large. MRA successfully predicted the preferred use of an SIEA flap instead of the DIEP flap in three out of three women (100%). Other studies have demonstrated similar findings. Masia et al.15 used MRA without contrast for abdominal perforators. A dominant perforator was identified in 56 women having DIEP flap reconstruction. They were able to determine the location of the dominant perforator, define its intramuscular course, and reliably evaluate the SIEA. The intramuscular perforators originated from the lateral row in 55% and from the medial row in 31%.
Imaging of the gluteal and thigh perforators can also be performed. Vasile et al.23 used MRA in 32 buttocks and imaged 142 perforators. The superior gluteal artery was the source for 92 (57.5%) perforators, the inferior gluteal artery was the source for 56 perforators (35%), and the deep femoral artery was the source for 11 (7.5%) perforators. The authors demonstrated that MRA imaging determined the location and course of the perforating vessels and can be useful when choosing an inferior or superior gluteal perforator flap.
Free tissue transfer can be accomplished from virtually anywhere in the body. When reconstructing the breast, there are certain criteria that make some flaps better suited than others. These criteria include adequate volume, ability to shape, adequate donor vessels, and donor-site considerations. The flaps that are reviewed in this chapter include the free TRAM, DIEP, SIEA, SGAP, IGAP, and TUG.
Integral to the selection of the flap for microvascular reconstruction is the selection of the recipient vessels. The most common recipient vessels are the internal mammary and thoracodorsal artery and vein.24-28 The internal mammary vessels are the vessels of choice in most cases because of ease of exposure, compatible size match, maximum freedom for flap positioning, and excellent flow characteristics (Figure 62.1). The diameter of the internal mammary vessels at the level of the fourth rib ranges from 0.99 to 2.55 mm for the artery and 0.64 to 4.45 mm for the vein. In contrast, the diameter of the thoracodorsal vessels ranges from 1.5 to 3.0 mm for the artery and 2.5 to 4.5 mm for the vein.15,16 The blood flow rate of the IMA ranges from 15 to 35 mL/min (mean, 25 mL/min) and the blood flow rate of the thoracodorsal artery ranges from 2 to 8 mL/min (mean, 5 mL/min).
FIGURE 62.1. The internal mammary artery and vein are prepared as recipient vessels for the microvascular anastomosis.
In the author’s practice, the IMA vessels are preferred for all delayed reconstructions and most immediate reconstruction. The vessels are exposed at either the third or fourth interspace. The cartilaginous segment of the rib is excised, the perichondrium is incised, and the vessels are exposed. This is usually performed using loupe magnification. Given the prevalence of sentinel lymph node dissection and the rarity of axillary dissection, the thoracodorsal artery and vein are infrequently used. These vessels are preferred, however, in the setting of a modified radical mastectomy where they have been exposed.
Abdominal Free Flaps
In general, the abdomen is the preferred donor site for the majority of microvascular reconstruction procedures. The abdomen is the source for the free TRAM, DIEP, and SIEA flaps. The blood supply to the intact anterior abdominal wall is derived from the deep inferior epigastric as well as the superficial inferior epigastric systems. The deep system is usually dominant and is therefore preferred in the majority of cases. The free TRAM and DIEP flaps are based on the deep system, whereas the SIEA is based on the superficial system. The adipocutaneous component of the free TRAM and DIEP flaps is perfused via the perforating branches of the inferior epigastric artery and vein.
The superficial system is less predictable, often absent, and less commonly used. The limiting factors associated with the superficial vessels are that they are not present in all patients and when present, they are usually smaller in caliber than the deep system. Thus, performing an SIEA flap is only possible in approximately 30% of women and is best reserved for women of moderate body habitus that require only a hemi-flap for the reconstruction.
The free TRAM was one of the first of the free tissue transfers for breast reconstruction. A primary goal of this flap was to improve the perfusion and vascularity to the flap and to minimize the amount of muscle removed relative to its pedicled counterpart. The free TRAM requires the removal of various amounts of the rectus abdominis muscle and anterior rectus sheath. Inherent to understanding the free TRAM is an appreciation for the various types of muscle preservation techniques. Classification of MS is based on the amount of rectus abdominis preserved.29,30 The rectus abdominis muscle can be separated into three longitudinal segments: medial, lateral, and central. The MS-0 (muscle sparing—none) includes the full width of the muscle; MS-1 includes preservation of the medial or lateral segment of the muscle; MS-2 includes the medial and lateral segment of the muscle; and the MS-3 includes preservation of all three segments (Tables 62.4 and 62.5).
Given that there are three possible microvascular flaps within the donor site of the anterior abdominal wall, the question becomes how to decide which to choose. In patients with complex abdominal scars, imaging of the vascular architecture is recommended. The free TRAM flap is considered when the SIEA and superficial inferior epigastric vein (SIEV) are not useable, the quality of perforators is poor (< 1.5 mm in diameter), or in the event that the flap volume requirements exceed 800 g. When a free TRAM is selected, the MS free TRAM (MS-1 or MS-2) is usually performed. The MS-0 free TRAM is uncommonly performed; however, it is considered in the event of a narrow rectus abdominis muscle. The limitation of the MS-0 free TRAM is that it disrupts the continuity of the rectus abdominis muscle and results in functional outcomes similar to the pedicle TRAM. When the principal perforators are small and localized in a segment of the rectus abdominis muscle or if the volume requirements are high, a small segment of the muscle is harvested with the flap. The advantage of including muscle is that multiple perforators can be included in the flap that may minimize the incidence of fat necrosis and venous congestion.
FIGURE 62.2. Preoperative abdominal markings in preparation for a free TRAM, DIEP, or SIEA flap.
FIGURE 62.3. Elevation of adipocutaneous component of an abdominal flap starts from the lateral edge of the flap.
Operative Details. The preoperative markings include delineation of the anterior superior iliac spine (ASIS) as well as the proposed upper and lower transverse incisions (Figure 62.2). Following the initial incisions, the right and left flaps are elevated from a lateral to medial direction (Figure 62.3). Once a network of perforators is visualized, the anterior rectus sheath is outlined to encompass the perforators. The fascia is incised creating an island of perforators (Figure 62.4). The anterior rectus sheath is elevated off the rectus abdominis muscle medially and laterally as indicated. The muscle is then undermined and the location of the inferior epigastric artery is visualized and palpated (Figure 62.5). This maneuver will facilitate dissection of the free TRAM and minimize the chance of injury to the perforators or pedicle. When the perforators are centrally located, an MS-2 free TRAM is performed (Figures 62.6 and 62.7). When the perforators are medial or lateral, an MS-1 free TRAM is performed (Figure 62.8). The rectus abdominis muscle is divided using a fine-tip mosquito clamp and an electrocautery device at a low setting. It is important to preserve the lateral intercostal motor innervation to maintain function of the rectus abdominis muscle. An example of a woman following a bilateral MS-2 free TRAM flap is demonstrated (Figures 62.9 and 62.10).
Deep Inferior Epigastric Artery Perforator (Table 62.6)
In this author’s practice, the DIEP flap constitutes approximately 70% of all abdominal flaps followed by the MS-2 free TRAM and SIEA. When considering an abdominal perforator flap, many surgeons will assess the vascular anatomy using the methods previously described. Intraoperative assessment is equally effective in identifying the abdominal wall perforating vessels. Reliance on only intraoperative assessment requires more experience because of the variability associated with perforator location, caliber, and number. There are five types of perforators that are typically encountered.31These perforators are direct (does not perforate the muscle, e.g., SIEA) and indirect (perforates the muscle). In general, for a perforator flap to be successfully harvested and transferred, a single perforating artery and vein of at least 1.5 mm in diameter is recommended. These perforators typically have a palpable pulse and are usually located in the periumbilical region. If a dominant perforator arising from the deep system is not identified, it may be because the superficial inferior epigastric system is dominant. In this situation, one can consider performing an SIEA flap or an MS free TRAM.
FIGURE 62.4. Following identification of the relevant perforators, the anterior rectus sheath is incised in preparation for the free TRAM.
FIGURE 62.5. The rectus abdominis muscle is undermined to palpate the intramuscular course of the inferior epigastric artery.
FIGURE 62.6. Typical appearance of the abdomen following MS-2 free TRAM flap. The central portion of the muscle is harvested.
FIGURE 62.7. Typical appearance of an MS-2 free TRAM that includes the small central segment of the rectus abdominis muscle.
FIGURE 62.8. Typical appearance of the abdomen following an MS-1 free TRAM flap. The central and medial segments of the muscle have been harvested.
FIGURE 62.9. Preoperative photograph of a woman with right breast cancer.
FIGURE 62.10. Postoperative photograph following right breast reconstruction with a free TRAM flap.
Operative Technique. The patient is marked preoperatively exactly as described with the free TRAM (Figure 62.2). The initial operative sequence is similar to the free TRAM except that one or more perforators are selected and isolated. The selected perforator should ideally be located near the center of the flap in order to obtain equidistant perfusion. Perforator diameter should be in excess of 1.5 mm. When several perforators are available, sequential occlusion can be performed to assist with the selection process. Multiple perforators can be considered when they are aligned in series or in close proximity (Figure 62.11). An example of a three-perforator DIEP flap is demonstrated (Figure 62.12). Medial row perforators are preferred when the flap will include zone 3 or zone 4. A personal observation in thin women is that perforator diameter is usually less than 1.5 mm. An option in these situations is to convert to a free TRAM. When initiating the dissection, it is recommended to include a small cuff of the anterior rectus sheath (1 to 2 mm) around the perforator, especially if the perforator is piercing the anterior rectus sheath at a tendinous inscription (Figure 62.13). During the dissection it is imperative to preserve the lateral intercostal nerves as they pierce the rectus abdominis muscle at the junction of the lateral and central longitudinal segments (Figure 62.14). Failure to do so will likely result in abdominal weakness or bulge. Motor nerve branches that cross the perforator or the source vessel can be sharply divided. Whether or not to coapt the severed nerve is controversial. Some advocate using a micro-suture for coaptation; however, it is this author’s preference to allow the transected end to innervate the adjacent muscle by neurotization. The intramuscular dissection proceeds to the point that the perforator or inferior epigastric vessel becomes submuscular. At that point, the dissection progresses from the lateral edge of the muscle toward the iliac vessels. It is recommended to continue the dissection until the vessel diameter approaches 2.5 to 3 mm. Following the intramuscular dissection, the anterior rectus sheath and continuity of the rectus abdominis muscle should be preserved and resemble that of a myotomy and fasciotomy (Figure 62.15).
Throughout the dissection, it is important to look for bleeding from the edges of the flap to assess perfusion. One can also use a hand-held Doppler probe to evaluate the arterial and venous signals. When a unilateral reconstruction is planned, it is wise to preserve the contralateral perforators in the event that a “lifeboat” is necessary. When a bilateral reconstruction is planned, it is advised to proceed cautiously when isolating and dissecting the perforators because a contralateral lifeboat will not be available. When in doubt about the quality of the perforators, an MS free TRAM flap is considered. An example of a patient following bilateral DIEP flap is demonstrated (Figures 62.16 and 62.17).
FIGURE 62.11. A column of perforators are isolated in preparation for a DIEP flap.
FIGURE 62.12. A triple-perforator DIEP flap is shown.
FIGURE 62.13. A single-perforator DIEP flap is shown in situ.
FIGURE 62.14. Preservation of the lateral intercostal innervation is important and demonstrated in this photograph.
FIGURE 62.15. Typical appearance of the abdomen following bilateral DIEP flap harvest. The rectus abdominis and anterior rectus sheath are incised and not removed.
FIGURE 62.16. Preoperative image of a woman with bilateral breast implants scheduled for bilateral DIEP flaps.
FIGURE 62.17. Postoperative image following bilateral DIEP flap breast reconstruction.
Superficial Inferior Epigastric Artery
The SIEA flap is based on the superficial inferior epigastric artery and vein.32.33 The advantage of this flap over the other abdominal free flaps is that it does not require a fasciotomy or myotomy (Figure 62.18). The SIEA flap is technically easier to harvest than either the DIEP or MS free TRAM flap. It is essentially an adipocutaneous flap that is perfused by a direct perforator (does not course through the rectus abdominis muscle). The SIEA and SIEV cross the inguinal ligament about one-third the distance from the pubic bone to the ASIS. Vessels of suitable caliber are not present in all patients. In patients with previous lower abdominal transverse incisions, the vessels may have been previously divided. The angiosome territory of these vessels is restricted to zones 1 and 2.
Operative Details. The patients are marked in a similar fashion as the free TRAM patients. A recommended approach is to begin dissection with the contralateral SIEA and SIEV. If the vessels are of suitable caliber with a palpable pulse, then the vessels are dissected to their origins and the SIEA flap is planned. An important consideration is that the diameter of the SIEA be 1.5 mm as it enters the lateral edge of the flap. Lesser diameters are associated with a higher failure rate. The length of the SIEA/SIEV pedicle ranges from 5 to 8 cm. If the contralateral vessels are of suboptimal caliber without a palpable pulse, then the ipsilateral SIEA/SIEV are explored. If those vessels are not suitable, then the deep system of perforators are explored. The contralateral and ipsilateral medial and lateral row of perforators are visualized and preferentially selected.
FIGURE 62.18. Typical appearance of the abdomen following harvest of an SIEA flap. The anterior rectus sheath and muscle are not violated.
Insetting the SIEA flap requires special considerations when compared with the free TRAM or DIEP flaps. The pedicle enters the flap at the edge rather than the undersurface. Standard insetting will create a sharp 180° fold in the pedicle that can compromise flow. Zenn34 has described a technique that permits a gradual folding of the pedicle that will not compromise flow. The inferior 2 to 3 cm of the flap is de-epithelialized. The dermis is released at the new epidermal edge. The flap is positioned with the pedicle oriented in the inferomedial direction. This allows the pedicle to rotate superiorly without kinking.
Gracilis Free Flaps
The medial thigh donor site has demonstrated success for breast reconstruction. Flaps such as the TUG and TMG have been described.7-9 Although the abdomen is the preferred donor site in the majority of women, alternative sites are sometimes necessary. The gluteal flaps are another alternative but concerns about pedicle length and caliber make these flaps potentially undesirable. The medial thigh is an alternative that is gaining momentum. Candidates for this flap include women with a flat abdomen, with or without scarring. Candidates must have an excess of skin and fat in the medial thigh region. Other indications include bilateral reconstructions in which the mastectomy volume approximates the volume of the medial thigh or meets the expectation of the patient. Schoeller et al.7 have used this flap in 111 patients. These patients were selected based on volume requirements, body habitus, and their desire to proceed with autologous reconstruction. Mean BMI was 23.7 (range, 19.4 to 28.5) and the mean volume of the flap was 330 cc (range, 150 to 550 cc). The caliber of the gracilis vascular pedicle ranged from 1.5 to 2.5 mm.
Operative Details. Patients are evaluated in the standing position by pinching the medial thigh region to determine the optimal height of the flap. In general, it ranges from 8 to 10 cm and may be up to 12 cm in patients following massive weight loss. The anterior and posterior limits of the flap are based on the dimensions of the mastectomy defect. The skin paddle can be delineated transversely or in a fleur-de-lis pattern. Patients are placed in the lithotomy position. The skin territory is incised and the dissection proceeds to the level of the muscle fascia. Superficial nerves within the flap are usually transected. The saphenous vein is included into the flap for additional venous drainage if necessary. Typically, the gracilis artery is associated with a vena comitans. The gracilis muscle is visualized and divided at its origin and at its distal musculotendinous insertion. The flap is transferred to the chest wall for the microvascular anastomosis. Ideally, recipient vessels are selected that will provide an optimal size match. The internal mammary vessels or their perforating branches are typically used.
Gluteal Free Flaps
The gluteal region has proved to be a valuable alternative for free flap breast reconstruction.6,35,36 It is generally recommended for women who lack sufficient skin and fat in the abdominal region. There are two general types of gluteal flaps that include the musculocutaneous varieties and the perforator varieties. The perforator flaps include the SGAP and the IGAP. These flaps are perfused on their respective vessels, the superior and inferior gluteal artery and vein. The gluteal flaps are considered to be among the more difficult flaps to harvest.
Operative Details. Preoperative identification of the anatomic landmarks is essential in raising these flaps. These landmarks include the greater trochanter, the posterior superior iliac crest, and the coccyx. The location of the perforators is best determined via preoperative imaging as well as using a hand-held Doppler probe with the patient in the prone position on the operating table (Figure 62.19). Several Doppler signals may be appreciated. It is this author’s preference to select a perforator that is based on the lateral aspect of the flap and toward the periphery of the flap because the laterally based perforators will usually provide a pedicle length of 6 to 8 cm; whereas the medially based perforators are usually 4 to 6 cm in length (Figure 62.20). In contrast to the DIEP flap where a central perforator is preferred, a peripheral perforator is preferred for the SGAP flap in order to facilitate the technical aspects of performing the microvascular anastomosis. With a central perforator, some of the useable length is under the flap rather than outside the flap edge. With a peripheral perforator, the added length can facilitate positioning of the flap during the anastomosis.
Once the perforator is isolated, the dissection commences in the subfascial plane. In contrast to a DIEP flap in which the length of the myotomy is minimized, the length of the myotomy is maximized with the SGAP dissection because the perforator dissection progresses perpendicular to the cutaneous surface. In the DIEP flap, the dissection is parallel with the cutaneous surface. It is important to recognize that the dissection continues deep to the gluteus maximus and medius muscles before penetrating the deep fibrous fascia (Figure 62.21). Once beyond this point, there are multiple venous branches that must be carefully dissected and divided before choosing the end point of the perforator. An example of a patient following a unilateral SGAP is shown (Figures 62.22 and 62.23).
FIGURE 62.19. An acoustic hand-held Doppler is used to localize the perforators in the upper gluteal region prior to SGAP flap harvest.
FIGURE 62.20. The dissection of an SGAP flap is depicted.
FIGURE 62.21. The harvested SGAP is shown. Note that the pedicle is relatively short.
FIGURE 62.22. Preoperative image of a woman prior to left SGAP reconstruction.
FIGURE 62.23. Postoperative image following left SGAP reconstruction.
The IGAP flap is raised using the same gluteal landmarks. The skin territory for this flap can be positioned to be “in the crease” as described by Allen et al.36 In general, the adipocutaneous component of this flap is slightly less than that of the SGAP.35 Other considerations are that the sciatic nerve is often exposed during this dissection and may result in postoperative discomfort. Because the incision is located in the ischial region, sitting may be restricted for several days following the operation and dehiscence of the incision is more common.
INTRAOPERATIVE MONITORING TOOLS
There are a variety of methods by which flap perfusion is assessed intraoperatively. Traditionally, surgeons have evaluated the color of the flap to determine if the perfusion is normal, congested, or pale. Surgeons have also employed the hand-held acoustic Doppler and auscultated the signals of the pedicle and the cutaneous perforators. Arteriovenous bleeding from the cut edges of the flap is generally regarded a sign of adequate perfusion. Zone 4 of the abdominal free flap is often poorly perfused and cannot be reliably transferred because of the risk of fat necrosis. A fluorescent woods lamp can demonstrate perfusion patterns within the flap. However, there has been no quantitative method to evaluate flap perfusion in the period immediately prior to or following a microvascular anastomosis.
Fluorescent angiography is a relatively new technology that allows direct visualization of perfusion within a cutaneous territory.37-39 This application can be used on tissue that is elevated as a flap or on a cutaneous territory that has not been elevated. The images are captured following the intravenous injection of indocyanine green (ICG). An image-capturing laser is then positioned a few inches above the cutaneous territory to be imaged. This device is linked to a computer that analyzes the data and generates a real-time image based on the fluorescence of the ICG. Images are obtained about 15 seconds after the ICG injection. In the setting of flap reconstruction, the images can be captured prior to, during, and following flap elevation, as well as postoperatively.
Evaluation of the reliable vascular territory of the cutaneous portion of an abdominal flap was one of the first applications of fluorescent angiography in the setting of autologous reconstruction. It was demonstrated that classic perfusion zones described by Hartrampf were not quite accurate in that the distal segment of the flap laterally adjacent to zone 1 was better perfused than the medially adjacent territory on the other side of the midline. Other clinical applications have included assessing perfusion in various free tissue transfer operations. Pestana et al.39 utilized fluorescent angiography in 23 patients with a variety of soft tissue deformities. Flaps included the TRAM, DIEP, SIEA, and SGAP. It was demonstrated that areas with relative hypoperfusion went on to develop areas of necrosis or eschar formation. They were also able to confirm patency of the microvascular anastomoses based on arterial inflow and venous outflow.
Fluorescent angiography is also useful in evaluating viability of the mastectomy skin flaps in the setting of immediate breast reconstruction. Fluorescent angiography is able to determine the perfusion thresholds of the skin following mastectomy and predict whether the skin will remain viable or not. Komorowska-Timek and Gurtner40 found the technique to be beneficial especially in cases of mastectomy with nipple–areolar preservation. Alterations in perfusion were noted in some women despite what appeared to be a normal nipple–areolar complex. In patients with a history of tobacco use or with connective disuse disorders, fluorescent angiography can help determine if further debridement is necessary.
Adequate flap monitoring is a critical component and predictor of successful flap outcomes. Postoperative assessment of flap circulation has traditionally required subjective interpretation of objective data. Traditional methods of flap monitoring have included hand-held Doppler probes, surface temperature assessment, flap turgor, capillary refill, and flap color. Important components in the monitoring of free flaps include differentiating the biphasic arterial and monophasic venous signals using a hand-held Doppler and ensuring that both signals are present. With inflow problems, flaps will become pale, cool, and soft with delayed or absent capillary refill. With outflow problems, flaps will become tense, congested, and purple, with brisk capillary refill. Although these methods of flap monitoring are usually effective, they are not continuous, are subject to interpretation, and are dependent on the experience of clinical personnel. There is a relatively short window of opportunity in which a flap can be salvaged in the event of altered flow. In muscle-based free flaps, the ischemia threshold is about 2 hours after which, irreversible muscle damage will occur. With perforator flaps, there is no muscle and the tolerated ischemic period is increased, ranging from 3 to 6 hours.
Near-infrared (NIR) spectroscopy is a continuous method of flap monitoring that measures oxygen saturation within the cutaneous layer of the flap.41,42 A surface probe is placed on the flap that emits NIR light and detects the hemoglobin content in the surface vessels. The light has a maximum penetration depth of approximately 2 cm. The measure of hemoglobin saturation is relatively constant for a given flap unless there is an alteration of flow. These changes will manifest immediately on the electrical tracing before there are any clinical signs of altered flap perfusion.
Clinical application of this technology has been encouraging. Keller has used NIR in 145 patients and 208 flaps. All patients were monitored intraoperatively and for 36 hours postoperatively.42 Of the 208 flaps, five demonstrated abnormalities in the spectroscopy measurements. All of these flaps were salvaged in part because of the early diagnosis of altered perfusion. Colwell et al.41 applied the NIR system in seven patients having free flap breast reconstruction using abdominal flaps. Baseline oxygen tension measurements ranged from 70% to 99% with a mean of 83%.
Assessing outcomes following microvascular breast reconstruction requires a complex set of parameters. It is not just about flap success and failure. It also relates to patient satisfaction, donor-site morbidities, and flap-related morbidity that in turn depends on surgeon experience. Some factors affecting outcome are within the surgeons control and others are not. Controllable factors include selection of the correct perforator, prevention of inadvertent perforator injury (thermal injury and avulsion), ensuring adequate recipient vessels with a proper caliber match, and adequate postoperative monitoring of the free flap. Factors outside the control of the surgeon include architectural anomalies of the vascular system within the flap or the recipient vessels, radiation damage, and hypercoagulability.
Abdominal Free Flaps
With the increasing interest in preserving the abdominal donor site, several studies have attempted to evaluate and quantitate outcomes following the various types of free flaps. In an early study comparing the DIEP (MS-3) flap to the MS-2 free TRAM, Nahabedian et al.2,43,44 demonstrated improved outcomes with regard to abdominal contour and strength following DIEP flap reconstruction. In unilateral cases, an abdominal bulge occurred in 4.6% and 1.5% of women following free TRAM and DIEP flap reconstruction, respectively. The ability to perform sit-ups was demonstrated in 97% and 100% following free TRAM and DIEP flaps, respectively. Following bilateral reconstruction, the differences were more pronounced. An abdominal bulge occurred in 21% and 4.5% following free TRAM and DIEP flaps, respectively. The ability to perform sit-ups was 83% and 95% following free TRAM and DIEP flaps, respectively. For all MS-2 free TRAM flaps (n = 113), outcome included fat necrosis in eight (7.1%), venous congestion in three (2.7%), and total necrosis in two (1.8%) patients. For all DIEP flaps (n = 110), outcome included fat necrosis in seven (6.4%), venous congestion in five (4.5%), and total necrosis in three (2.7%) patients. Although these differences in abdominal morbidity were not statistically significant, a clear trend was evident.
Outcome analysis regarding bilateral microvascular breast reconstruction has recently been evaluated. In a review of 342 bilateral flaps, Rao et al.45 demonstrated failure or intraoperative termination of the procedure in 12 cases, yielding a failure rate of 3.5%. Causes of flap failure included venous insufficiency (6/12), lack of adequate perforator anatomy (3/12), and perforator injury during dissection (2/12). A review of 386 unilateral cases over the same time period demonstrated a failure rate of 2.1%.
Regarding the abdominal wall, Nahabedian46 reviewed secondary operations following free TRAM and DIEP flap reconstruction. Secondary abdominal operations were performed in 59 women (17.9%). The indications were considered necessary in 33 women (10%) and elective in 31 women (9.4%). Lower abdominal bulge was the most common necessary indication and was repaired in 9.3% of free TRAM flaps and 4.7% of DIEP flaps. Dog-ear scars were the most common elective indication and were revised in 29 women (8.8%). Neuromas of the anterior abdominal wall were diagnosed in three women (0.9%).
Wu et al.47 compared donor-site morbidity between the free TRAM, DIEP, and SIEA flaps. A questionnaire was used to assess donor-site function, pain, and aesthetics in 179 women following microvascular breast reconstruction. The SIEA flap scored highest in 10 of 12 categories following unilateral reconstruction. These categories included better postoperative lifting (p < 0.02) and shorter duration of abdominal pain (p < 0.06). Of the bilateral patients, the ability to get out of bed was significantly higher following the SIEA flaps compared with MS free TRAM and DIEP flaps (p < 0.02).
In an effort to better understand the risks and benefits of the SIEA flap, Selber et al.48 compared flap and abdominal outcomes following microvascular breast reconstruction with the MS free TRAM and the SIEA flap. It was demonstrated that the incidence of hernia and bulge following the free TRAM flaps was 1.9% compared with 0% for the SIEA flaps. However, the incidence of thrombotic complications and flap loss was significantly higher for the SIEA group compared with the MS free TRAM group (p < 0.0005). Although there are donor-site advantages with the SIEA flap, the risks associated with the flap itself may negate the benefits. In a more recent prospective study, of the same group, the authors were able to demonstrate that there was a more progressive decline in upper abdominal strength following the free TRAM compared with the DIEP flap following unilateral reconstruction.49,50 However, following bilateral reconstruction, there was a more progressive decline in upper and lower abdominal strength following the free TRAM compared with the DIEP flap and SIEA flaps.
With regard to the SIEA flap, concerns include the smaller caliber vessels, the limited cutaneous territory, the increased incidence of fat necrosis, and the higher rate of redo arterial and venous anastomoses. My personal philosophy is that the key to successful “microvascular” surgery is to make it as “macrovascular” as possible. Small caliber vessels associated with a significant vascular mismatch can predispose to anastomotic failure.51
Gluteal flaps are remarkably durable. It has been a personal observation that when successful, fat necrosis is rare. This may be explained by the fact that the subdermal plexus in the gluteal region is well developed because of the constant pressure endured from sitting. The difficulty with this flap relates primarily to its short vascular pedicle and small caliber artery. In addition, the dissection deep to the gluteal fascia can be complicated because of the tremendous amount of branching, especially of the gluteal veins. Allen et al.6 have demonstrated that the overall take-back rate with gluteal flaps was 8% with a 6% incidence of vascular complications. Total flap failure occurred in 2% of patients and the donor-site seroma rate was 2%. Revision of the donor-site scar was necessary in 4%. With regard to the appearance of the donor site, it has been a personal observation that some women will have significant scalloping of the buttock and/or gluteal asymmetry. This can be a major source of dissatisfaction. In general, it has been noted that petite women of shorter stature are more prone to these aesthetic issues, whereas taller and slender women are less prone.
Gracilis Free Flaps
The TUG and TMG flaps have demonstrated success in the setting of microvascular breast reconstruction. The benefits include the use of an expendable muscle and a donor site that is relatively hidden. Donor-site morbidity in this region has been a bit higher than the other regions such as the abdomen and gluteal region; however, flap success has been high ranging from 97% to 100%. Vega et al.8demonstrated donor-site infection in 3.7%, with delayed donor-site healing of 11%. The incidence of fat necrosis and donor-site contour abnormality was 7.4% and 3.7%, respectively. Operating times generally range from 4.5 to 5 hours for unilateral reconstruction and 6 to 7 hours for bilateral reconstructions. Patient satisfaction was assessed by Fansa et al.9 who demonstrated that 94% of patients (15/16) returned to normal work.
Microvascular breast reconstruction is rapidly acquiring momentum for women with breast cancer seeking autologous reconstruction. The ability to transfer an adipocutaneous flap without removal of the donor-site muscle has been the driving force. Many women have learned of these perforator flap procedures and seek surgeons that perform them. There are a variety of donor sites available, each with its own set of indications, benefits, and morbidities. The preoperative, intraoperative, and postoperative factors that can influence the success of these operations have been outlined in this chapter.
1. Nahabedian MY. Breast reconstruction: a review and rationale for patient selection. Plast Reconstr Surg. 2009;124:55-62.
2. Nahabedian MY, Momen B, Galdino G, Manson PN. Breast reconstruction with the free TRAM or DIEP flap: patient selection, choice of flap, and outcome. Plast Reconstr Surg. 2002;110:466-475.
3. Garvey PB, Buchel EW, Pockaj BA, Gray RJ, Samson TD. The deep inferior epigastric perforator flap in overweight and obese patients. Plast Reconstr Surg. 2005;115:447.
4. Chang DW, Wang B, Robb G, et al. Effect of obesity of flap and donor site complications in free TRAM flap breast reconstruction. Plast Reconstr Surg. 2000;105:1640.
5. Granzow JW, Levine JL, Chiu ES, Allen RJ. Breast reconstruction with gluteal artery perforator flaps. J Plast Reconstr Aesthet Surg. 2006;59:614-621.
6. Guerra AB, Metzinger SE, Bidros RS, Gill PS, Dupin CL, Allen RJ. Breast reconstruction with gluteal artery perforator (GAP) flaps a critical analysis of 142 cases. Ann Plast Surg. 2004;52:118-125.
7. Schoeller T, Huemer GM, Wechselberger G. The transverse musculocutaneous gracilis flap for breast reconstruction: guidelines for flap and patient selection. Plast Reconstr Surg. 2008;122:29-38.
8. Vega SJ, Sandeen SN, Bossert RP, et al. Gracilis myocutaneous free flap in autologous breast reconstruction. Plast Reconstr Surg. 2009;124:1400-1409.
9. Fansa H, Schirmer S, Warnecke IC, Cervelli A, Frerichs O. The transverse myocutaneous gracilis muscle flap: a fast and reliable method for breast reconstruction. Plast Reconstr Surg. 2008;122:1326-1333.
10. Girotto JA, Schreiber J, Nahabedian MY. Breast reconstruction in the elderly: preserving excellent quality of life. Ann Plast Surg. 2003;50:572-578.
11. Heitland AS, Markowicz M, Koellensperger E, Schoth F, Feller AM, Pallua N. Duplex ultrasound imaging in free transverse rectus abdominis muscle, deep inferior epigastric artery perforator, and superior gluteal artery perforator flaps early and long-term comparison of perfusion changes in free flaps following breast reconstruction. Ann Plast Surg. 2005;55:117-121.
12. Rozen WM, Phillips TJ, Ashton MW, Stella DL, Gibson RN, Taylor GI. Preoperative imaging of DIEA perforator flaps: a comparative study of computed tomographic angiography and Doppler ultrasound. Plast Reconstr Surg. 2008;121:1-8.
13. Rozen WM, Palmer KP, Suami H, et al. The DIEA branching pattern and its relationship to perforators: the importance of preoperative computed tomographic angiography for DIEA perforator flaps. Plast Reconstr Surg. 2008;121:367-373.
14. Masia J, Clavero JA, Larrañaga JR, Alomar X, Pons G, Serret P. Multidetector-row computed tomography in the planning of abdominal perforator flaps. J Plast Reconstr Aesth Surg. 2006;59:594-599.
15. Masia J, Kosutic, D, Cervelli D, Clavero JA, Monill JM, Pons G. In search of the ideal method in perforator mapping: noncontrast magnetic resonance imaging. J Reconstr Microsurg. 2010;26(1):29-35.
16. Rozen WM, Garcia-Tutor E, Alonso-Burgos A, Corlett RJ, Taylor GI, Ashton MW. The effect of anterior abdominal wall scars on the vascular anatomy of the abdominal wall: a cadaveric and clinical study with clinical implications. Clin Anat. 2009;22:815-823.
17. Casey WJ, Chew RT, Rebecca AM, Smith AA, Collins JM, Pockaj BA. Advantages of preoperative computed tomography in deep inferior epigastric artery perforator flap breast reconstruction. Plast Reconstr Surg. 2009;123:1148-1155.
18. Greenspun D, Vasile J, Levine JL, et al. Anatomic imaging of abdominal perforator flaps without ionizing radiation: seeing is believing with magnetic resonance imaging angiography. J Reconstr Microsurg. 2010;26(1):37-44.
19. Giunta RE, Geisweid A, Feller AM. The value of preoperative Doppler sonography for planning free perforator flaps. Plast Reconstr Surg. 2000;105:2381-2386.
20. Berg WA, Chang BW, DeJong MR, Hamper UM. Color Doppler flow mapping of abdominal wall perforating arteries for transverse rectus abdominis myocutaneous flap in breast reconstruction: method and preliminary results. Radiology. 1994;192:447-450.
21. Alonso-Burgos A, García-Tutor E, Bastarrika G, Cano D, Martínez-Cuesta A, Pina LJ. Preoperative planning of deep inferior epigastric artery perforator flap reconstruction with multislice-CT angiography: imaging findings and initial experience. JPRAS. 2006;59;585-593.
22. Chernyak V, Rozenblit AM, Greenspun DT, et al. Breast reconstruction with deep inferior epigastric artery perforator flap: 3.0-T gadolinium enhanced MR imaging for preoperative localization of abdominal wall perforators. Radiology. 2009;250(2):414-424.
23. Vasile JV, Newman T, Rusch DG, et al. Anatomic imaging of gluteal perforator flaps without ionizing radiation: seeing is believing with magnetic resonance angiography. J Reconstr Microsurg. 2010;26(1):45-57.
24. Nahabedian MY. The internal mammary artery and vein as recipient vessels for microvascular breast reconstruction: are we burning a future bridge? Ann Plast Surg. 2004;53:311-316.
25. Hefel L, Schwabegger A, Ninkovic M, et al. Internal mammary vessels: anatomical and clinical implications. Br J Plast Surg. 1995;48:527-532.
26. Feng LJ. Recipient vessels in free-flap breast reconstruction: a study of the internal mammary and thoracodorsal vessels. Plast Reconstr Surg. 1997;99:405-416.
27. Lorenzetti F, Kuokkanen H, von Smitten K, et al. Intraoperative evaluation of blood flow in the internal mammary or thoracodorsal artery as a recipient vessel for a free TRAM flap. Ann Plast Surg. 2001;46:590-593.
28. Lorenzetti F, Souminen S, Tukiaanen E, et al. Evaluation of blood flow in free microvascular flaps. J Reconstr Microsurg. 2001;17:163-167.
29. Nahabedian MY, Manson PN. Contour abnormalities of the abdomen following TRAM flap breast reconstruction: a multifactorial analysis. Plast Reconstr Surg. 2002;109:81-87.
30. Nahabedian MY, Dooley W, Singh N, Manson PN. Contour abnormalities of the abdomen following breast reconstruction with abdominal flaps: the role of muscle preservation. Plast Reconstr Surg. 2002;109:91-101.
31. Blondeel PN, Van Landuyt KHI, Monstrey SJM, et al. The “gent” consensus on perforator flap terminology: preliminary definitions. Plast Reconstr Surg. 2003;112:1378-1383.
32. Spiegel AJ, Kahn FN. An intraoperative algorithm for use of the SIEA flap for breast reconstruction. Plast Reconstr Surg. 2007;120(6):1450-1459.
33. Chevray PM. Breast reconstruction with superficial inferior epigastric artery flaps. Plast Reconstr Surg. 2004;114(5):1077-1083.
34. Zenn MR. Insetting of the superficial inferior epigastric artery flap in breast reconstruction. Plast Reconstr Surg. 2006;117:1407-1411.
35. Ahmadzadeh R., Bergeron L, Tang M, Morris S. The superior and inferior gluteal artery perforator flaps. Plast Reconstr Surg. 2007;120:1551-1556.
36. 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.
37. Jones GE, Garcia CA, Murray J, Elwood ET, Whitty A. Fluorescent intraoperative tissue angiography for the evaluation of the viability of pedicled TRAM flaps. Plast Reconstr Surg. 2009;124:53.
38. Newman MI, Samson MC. The application of laser-assisted indocyanine green fluorescent dye angiography in microsurgical breast reconstruction. J Reconstr. Microsurg. 2009;25:21-26.
39. Pestana IA, Coan B, Erdmann D, Marcus J, Levin LS, Zenn MR. Early experience with fluorescent angiography in free-tissue transfer reconstruction. Plast Reconstr Surg. 2009;123:1239-1244.
40. Komorowska-Timek E, Gurtner GC. Intraoperative perfusion mapping with laser-assisted indocyanine green imaging can predict and prevent complications in immediate breast reconstruction. Plast Reconstr Surg. 2010;125:1065-1073.
41. Colwell AS, Wright L, Karanas, Y. Near-infrared spectroscopy measures tissue oxygenation in free flaps for breast reconstruction. Plast Reconstr Surg. 2008;121:344e.
42. Keller A. A new diagnostic algorithm for early prediction of vascular compromise in 208 microsurgical flaps using tissue oxygen saturation measurements. Ann Plast Surg. 2009;62:538-543.
43. Nahabedian MY, Momen B, Tsangaris T. Breast reconstruction with the muscle sparing (MS-2) free TRAM and the DIEP flap: is there a difference? Plast Reconstr Surg. 2005;115:436-444.
44. Nahabedian MY, Momen B. Lower abdominal bulge after DIEP flap breast reconstruction. Ann Plast Surg. 2005;54:124-129.
45. Rao SS, Parikh PM, Goldstein JA, Nahabedian MY. Unilateral failures in bilateral microvascular breast reconstruction. Plast Reconstr Surg. 2010;126(1):17-25.
46. Nahabedian MY. Secondary operations of the anterior abdominal wall following microvascular breast reconstruction with the TRAM and DIEP flaps. Plast Reconstr Surg. 2007;120:365-372.
47. Wu LC, Bajaj A, Chang DW, et al. Comparison of donor-site morbidity of SIEA, DIEP, and muscle-sparing TRAM flaps for breast reconstruction. Plast Reconstr Surg. 2008;122:702-709.
48. Selber JC, Samra F, Bristol M, et al. A head-to-head comparison between the muscle-sparing free TRAM and the SIEA flaps: is the rate of flap loss worth the gain in abdominal wall function? Plast Reconstr Surg. 2008; 122:348-355.
49. Selber JC, Fosnot J, Nelson J, et al. A prospective study comparing the functional impact of SIEA, DIEP, and muscle-sparing free TRAM flaps on the abdominal wall: part II. Bilateral reconstruction. Plast Reconstr Surg. 2010;126:1438-1453.
50. Selber JC, Fosnot J, Nelson J, et al. A prospective study comparing the functional impact of SIEA, DIEP, and muscle-sparing free TRAM flaps on the abdominal wall: part I. Unilateral reconstruction. Plast Reconstr Surg. 2010;126:1142-1153.
51. Nahabedian MY, Momen B, Manson PN. Factors associated with anastomotic failure following microvascular reconstruction of the breast. Plast Reconstr Surg. 2004;114:74-82.