Management after flap failure: a narrative review

INTRODUCTION

Autologous free tissue transfer for breast reconstruction is a well-established and reliable form of reconstruction for women undergoing mastectomies. Free tissue flaps provide a versatile and natural reconstructive option, with greater longevity of results and evidence of improved patient-reported quality of life compared to implant-based reconstruction^[1,2]. Free flap breast reconstructions are performed with high rates of success, estimated at > 96% in high-volume centers^[3-9]; however, the consequences of flap failure can still be devastating to patients and surgeons^[10]. Specific patient and non-patient related factors have been identified that can PREDISPOSE PATIENTS to a greater risk for flap failures. Patient factors include hypercoagulable conditions, prior radiotherapy, and obesity^[11,12]. Beyond patient factors, flap failures can also be IMPACTED by flap and perforator choices, and technical problems in all phases of the procedure including the flap harvest, microvascular anastomoses, flap inset and postoperative incidents.

Breast reconstruction decision making following flap loss is a uniquely individualized process, contingent upon considerations of safety, patient goals and preferences, as well as the surgeon’s skillset. The aims of reconstruction salvage after flap loss are to excise unhealthy tissue and restore a breast mound of normal anatomical shape^[13]. There are well-described management algorithms for other types of failed microvascular reconstruction, such as for lower extremity^[14] or head and neck reconstructions^[15], but there is a paucity of information on this topic in breast reconstruction. We present an algorithm as a possible approach to managing flap failures [Figure 1]. We also review the management of breast reconstruction following free flap failure, including the role of hematologic investigation, anticoagulation recommendations and secondary or tertiary reconstruction with both prosthetic and autologous techniques.

Figure 1. Algorithm for management after flap failure.

PATIENT COUNSELING IN THE SETTING OF FLAP FAILURE

However difficult the conversation may be, clear and thoughtful patient communication is imperative in the event of a flap failure. Complications from surgery have a significant impact on quality of life, including mental health conditions^[16]. This can be particularly devastating to women undergoing secondary autologous reconstruction after failed alloplastic reconstruction, as there are potential feelings of losing the breast twice^[17]. Having a complete preoperative discussion to set realistic expectations is the first step. This is true for not only complete flap loss, but the range of breast complications as well as those that involve the donor site. Informing patients of the likelihood of adverse events is necessary while also discussing individual factors that place them at higher risk, such as obesity or hypercoagulability. It can be helpful to discuss flap loss rates reported in the literature and, if available, failure rates at the specific treating institution. It is also beneficial to educate patients on the secondary (or tertiary) options for reconstruction, such as implant-based, and additional autologous reconstruction options. While knowledge of the possibility of flap failure can cause some concern for patients, most are placed at ease knowing that alternate options exist in the unlikely event of flap failure and that they will be supported and guided through each step of the process. In patients who do not wish to undergo further reconstructive surgery, the option of an aesthetic flat closure should be offered^[18]. In fact, Lineaweaver et al. found that nearly half of patients opted to forgo further breast reconstruction following their flap failure^[19].

Higgins et al. interviewed women who experienced complete flap loss for breast reconstruction to better understand the psychosocial detriment of this outcome^[10]. Not surprisingly, women expressed difficulty with body image and coping with emotions after flap loss. Another notable theme that emerged, however, was the impact the relationship with the surgeon had on the patient’s overall experience. The study showed that women who reported a strong relationship with their surgeon also reported easier acceptance and coping with their flap loss. Similarly, patients who felt unsupported or dismissed by their surgeon expressed greater emotional distress and questioning after flap loss^[10]. Many women suggested increasing emotional support resources in the setting of flap failure, including social workers and psychiatrists. Given that a multidisciplinary approach results in better outcomes in breast reconstruction^[20] a similar holistic approach to the management of the patient who has experienced a flap loss may be beneficial. Li et al. found that with dedicated nursing attention, surgical breast cancer patients reported lower depression scores and greater satisfaction postoperatively^[21]. Even with ancillary support, the surgeon is ultimately responsible for the management of the patient and has the most influential impact on helping patients through this difficult time.

HEMATOLOGICAL INVESTIGATION

A hematology workup should be considered if risk factors for clotting are identified during the preoperative evaluation, or if unexplained clotting resulting in failure is encountered postoperatively. While free flap transfer can be successfully performed in patients with underlying thrombophilias, hypercoagulability has traditionally been described as a relative contraindication to free tissue transfer. A recent systematic review performed by Kotamarti et al. found an 18.4% thrombosis rate with a pooled 12.2% flap failure rate in breast reconstruction patients concerning patients with hypercoagulability^[22]. Many patients with hereditary or acquired thrombophilia experience their first complication in the setting of surgery^[23]. Additionally, when thrombosis is discovered postoperatively in these patients, the salvage rate is near zero^[23,24].

Thrombophilia is reported to have a prevalence rate of 5 to 27% of the population, with even higher prevalence in oncologic patients and patients undergoing lower extremity reconstruction^[22,25,26]. Hypercoagubility disorders can be hereditary or acquired. Hereditary disorders include factor V Leiden, prothrombin G20210A mutation, methylenetetrahydrofolate reductase (MTHFR) mutations, protein C deficiency, protein S deficiency, antithrombin III deficiency, and elevated factor VIII. Acquired conditions include antiphospholipid syndrome (anticardiolipin, lupus anticoagulant), and some forms of hyperhomocysteinemia. Hypercoagulable states pose a challenge in autologous free flap reconstruction because they are frequently undetected preoperatively and are often only brought on by a precipitating event, such as a microvascular procedure^[25,27].

Free flap thrombosis more commonly occurs in the postoperative period than intraoperatively in patients with known hypercoagulability disorders^[22]. Multiple studies have also demonstrated that flap salvage rates are significantly higher if flap thrombosis and re-exploration occur early in the postoperative period^[28-31]. Wang et al. found that the failure rate approaches 100% when flap thrombosis occurs on postoperative days 4 and 5 in hypercoagulable patients^[28]. For these reasons, it is imperative to screen patients for hypercoagulable disorders preoperatively to minimize the risk of postoperative flap thromboses that are difficult to salvage. Perioperative risk assessment can significantly reduce the occurrence of flap thrombosis^[32]. During the initial preoperative consultation, it is important to obtain a thorough history by asking questions regarding (1) personal history of blood clots, including deep vein thrombosis or pulmonary embolism; (2) personal history of miscarriages; (3) personal history of strokes at a young age; and (4) family history of clotting or previously diagnosed coagulation disorders^[28,32]. If hypercoagulability is suspected based on history, referral to hematology for further workup is warranted. Patients who were referred to a hematologist have shown a higher flap success rate compared to patients who did not, as administration of an appropriate perioperative anticoagulation regimen can mitigate risk^[28]. For instance, Kalmar et al. discovered that a platelet count of > 250 K/mcL (P = 0.004) is associated with a higher rate of flap failure^[33]. The authors suggest there may be a role for personalized anticoagulation protocols for thrombocytosis with agents specifically targeting platelets, such as aspirin, ticlodipine, and dipyramidole^[33]. Genetic testing should also be considered, especially in patients with a family history of clotting episodes. It is recommended that a formal hypercoagulable workup be performed at a minimum of 4-6 weeks after a traumatic event such as surgery, as coagulation factors remain elevated during this time, specifically thrombin, that will alter the results of the testing^[13,34]. After a patient is confirmed to have a hypercoagulable disorder, surgeons can collaborate with hematologists to determine a perioperative anticoagulation regimen, especially if salvage free-tissue transfer is to be attempted. Literature is sparse on success and failure rates for breast reconstruction of an attempted second flap after an initial failure in a patient with a known hypercoagulable condition. In a series described by Hamdi et al., two patients with hypercoagulable conditions underwent a second free flap with a successful free tissue transfer with appropriate anticoagulation^[17].

ANTICOAGULATION REGIMEN

In patients without hypercoagulable disorders, prophylactic antithrombotic therapy is used to minimize the risk of venous thromboembolic events but can also decrease the incidence of microvascular thrombosis after free flap surgery^[25]. There is no consensus on anticoagulation protocols and relevant studies are generally lacking. For example, at the authors’ institution, it is typical to administer subcutaneous heparin 5,000 units preoperatively, aspirin 300 mg rectal suppository at the end of the case, followed by Lovenox 40 mg QD starting postoperative day 0 and aspirin 325 mg starting on the first postoperative for 30 days for patients without increased risk of thrombosis. Liu et al. reported a regimen of intraoperative heparin bolus of 2,000 units intravenously followed by five days of heparin infusion at 500 unit/hour in patients who are at risk of hypercoagulability^[35]. Wang et al. presented four different anticoagulation protocols based on surgeon preference at a single institution^[28]. It is apparent that prophylactic antithrombotic regimen has varied through the decades amongst different institutions and surgeons.

When it comes to patients who have already experienced a thrombotic event or patients who are likely to have an underlying hypercoagulable condition, the use of therapeutic anticoagulation remains controversial. A retrospective review by Senchenkov et al. described an algorithm of anticoagulation for patients both with and without a history of hypercoagulability^[25]. They concluded that in patients without hypercoagulable history, additional anticoagulation beyond routine VTE prophylaxis is not indicated. Based on available data in the cardiovascular literature, Senchenkov et al. recommend that in patients with a hypercoagulable history, a heparin bolus prior to flap ischemia, ex-vivo irrigation of the flap with heparinized saline prior to anastomosis, and systemic anticoagulation should be considered^[25]. Additionally, in the setting of recurrent flap thrombosis, heparin drip, intraoperative ASA, Plavix via nasogastric tube, and dextran-40 infusion should be considered^[25].

The use of thrombolytic agents after free flap failure can be considered, but their efficacy has not been well established. Thrombolytic agents used in free flap salvage include streptokinase, urokinase, and tissue-plasminogen activator (tPA)^[29,36,37]. tPA is the most commonly used agent and is generally injected via the arterial pedicle at a concentration of 1 mg/mL^[37]. Urokinase is generally infused in an anterograde manner through the arterial pedicle at a concentration of 5,000 units/mL. Streptokinase can be injected into the arterial pedicle at 7,500 to 250,000 units diluted in 10-30 cc of normal saline^[37]. Thrombolytics should be used with caution and only as a last resort due to their associated complications, including bleeding events and allergic reactions. With this in mind, they can be injected just proximal to the arterial anastomosis to increase their bioavailability at the site of the thrombus. Some have also reported taking down the venous anastomosis to avoid the systemic spread of the thrombolytic agent, though the risk of complications from systemic spread of the doses used in flap salvage is not entirely clear^[37].

SALVAGE FREE FLAP: RECIPIENT VESSEL SELECTION

A significant challenge facing the reconstructive surgeon following the failure of free flaps for breast reconstruction is the availability of recipient vessels in the chest. There is no consensus on the optimal timing of secondary free flap reconstruction following the failure of the initial flap, though some surgeons advocate for immediate free flap reconstruction at the time of debridement of the original flap as the mammary arteries and veins may still be salvageable. Hamdi et al. advocate for color Duplex imaging to assess the internal mammary system following free flap failure if considering another flap^[17]. If the anterograde system is thrombosed, the retrograde system should be interrogated as it is a robust and reliable recipient vessel option for secondary free flap reconstruction^[38,39]. Alternative vessel choices can be based on the subscapular system, which comes off of the axillary vessels, namely the thoracodorsal or serratus vessels^[40] [Figure 2]. To identify the thoracodorsal and serratus vessels, the lateral pectoralis border is first found. Within the axillary fat, the lateral thoracic vein is found and can be followed proximally to the axillary vein. Carefully dissecting bluntly through the axillary fat posterior to the origin of the lateral thoracic vein reveals the proximal thoracodorsal vessels. The serratus branch can be found two to three cm from the origin of the thoracodorsal artery, supplying the serratus muscle^[40]. The thoracodorsal artery must be ligated distally to avoid a caliber mismatch of secondary flap pedicle^[41]. Moran et al. conducted a prospective cohort study and found no significant outcome differences between the internal mammary and thoracodorsal vessels as recipient sites for autologous breast reconstruction, concluding that both are safe options with acceptable results^[42].

Figure 2. Illustration of the thoracodorsal vessels arising from the subscapular system (Image 1). Exposure of the thoracodorsal artery, vein, and nerve prior to their entry into the latissimus dorsi muscle. The nerve should be dissected free to prevent any kinking of the vessels after anastomosis (Image 2). Reproduced with permission from TA Kung, AO Momoh, Ch 24: Recipient Vessel Exposure--Internal Mammary and Thoracodorsal, Operative Techniques in Breast Surgery, Trunk Reconstruction and Body Contouring. Publication Date: June 3, 2019. Wolters Kluwer.

Other less common venous outflow options have been described, such as the cephalic and external jugular veins^[43]. Additionally, if the failure is not due to thrombosis of the pedicle and the anastomosis remains patient, the initial flap artery and vein can be used for secondary flap salvage recipient vessels^[44,45]. The thoracoacromial vessels have also been described for recipient vessels for autologous breast reconstruction, with the added benefit of not sacrificing any rib and possibly less pain^[46]. Yamamoto et al. compared thoracoacromial vessels to the internal mammary vessels for breast reconstruction found a significantly smaller size artery (1.70 ± 0.26 mm) and vein (1.64 ± 0.24 mm) compared to the internal mammary artery (2.27 ± 0.31 mm) and vein (2.33 ± 0.29 mm) (P < 0.001)^[46]. Even less commonly, contralateral internal mammary vessels as recipient options have been described^[47,48].

SALVAGE FREE FLAP: SECONDARY FLAP SELECTION

As the DIEP flap remains the gold standard for autologous breast reconstruction, in the scenario of a failed initial free flap reconstruction, secondary free flap options require alternative donor sites. These include the thigh (myocutaneous gracilis flaps, profunda artery perforator flap, lateral thigh flap) and trunk (lumbar artery perforator flap, superior and inferior gluteal artery perforator flaps).

A myocutaneous gracilis flap (either transverse, diagonal, or vertical skin paddle orientation) is a reliable option with high patient satisfaction for secondary free flap salvage^[49]. As this flap does not require perforator dissection, it can be performed expeditiously if attempting in the setting of an immediate flap loss. The profunda artery perforator (PAP) flap is another medial thigh-based flap with similar advantages to the myocutaneous gracilis flap but a muscle-sparing alternative^[50] [Figure 3]. The lateral thigh perforator (LTP) flap, renovated for breast reconstruction by Robert Allen, Sr. is a secondary option for patients with the “saddlebag” deformity^[51,52]. Based on the ascending lateral circumflex femoral artery, dissection is relatively simple due to the septocutaneous location of the perforators between the tensor fascia lata (TFL) and the gluteus medius (GM) muscles^[53]. The LTP has a short pedicle, often necessitating vein grafting in salvage cases. The lumbar artery perforator (LAP) flap perforasome includes the soft tissues commonly referred to as the “love handle” region^[54]. Like the LTP flap, this also has a short pedicle. Gluteal artery perforator flaps (superior and inferior)^[55] are additional trunk-based options for alternative flap options. Historically, the SGAP and IGAP flaps were preferred second-line options, though now mostly replaced by medial thigh-based flaps due to reliable anatomy, availability of donor site tissue, ease of harvest, and lower complication rates^[53,56].

Figure 3. Patient with a genetic predisposition for breast cancer (Images 1 and 2) who underwent bilateral skin- sparing mastectomies and immediate reconstruction with a left DIEP flap and a right SIEA flap. Three weeks postoperatively, she presented with a discoloration of the right flap skin paddle, brisk capillary refill, and firmness at the inferolateral quadrant of the breast (Images 3 and 4). While not frankly necrotic, the entire breast ultimately was very firm (Images 5 and 6). The right SIEA flap was debrided and replaced with a PAP flap using a more proximal portion of the same antegrade internal mammary vessels (Images 7 and 8).

Though typically utilized in a pedicled fashion, a contralateral latissimus dorsi myocutaneous free flap can be performed in the setting of tertiary reconstruction^[57]. This has a reliable anatomy with generally adequate donor site tissue, especially if a large area of skin is needed. An additional consideration for this option would be if the ipsilateral thoracodorsal dorsal pedicle was previously sacrificed during an axillary lymph node dissection. While this may not be the first choice for a free flap tertiary salvage breast reconstruction, a contralateral free latissimus is an option if needed.

Non-abdominally based flaps may not provide enough tissue in the setting of a DIEP flap failure. Challenges with alternative flaps include less volume with a smaller skin paddle and an inability to restore the breast footprint or skin envelope, especially when working to match a relatively large contralateral breast (native or reconstructed)^[58]. With smaller flaps, a dependable solution is to combine two flaps for single breast reconstruction in a “stacked” fashion^[39]. When recipient vessels are limited due to previous thrombosis after flap failure, flaps can be stacked using the “daisy-chain” technique: anastomosing one flap to a branch of the pedicle of the other flap^[59].

Performing a second free flap after an initial failure can be a reasonable option shown to be effective in the literature. Hamdi et al. retrospectively reviewed their series of repeat free flap breast reconstruction after an initial free flap failure^[17]. In this series, 688 patients experienced 14 failures of autologous reconstruction requiring salvage. Of these 14, eight patients underwent nine microvascular breast reconstructions, with two of these nine experienced failures of the second flap. This information is useful when discussing options with patients after free flap failures. Baumeister et al. similarly retrospectively reviewed 902 free flaps, identifying 13 patients who underwent a second flap surgery^[60]. Microsurgical free tissue transfers were successful in 11 of the 13 patients. The authors outline their approach in the setting of a failed flap which includes a reconsideration of the need for autologous tissue, sensitive patient counseling, thorough analysis of the cause of failure, and change in microsurgical strategy^[60].

PEDICLED FLAP OPTIONS

Pedicled flaps are an alternative option to performing free flaps for salvage in clinical situations where vascularized tissue is required, such as in the setting of radiation. Implant reconstructions in the setting of prior radiation result in increased rates of major complications such as implant extrusion, capsular contracture, and reconstruction failure compared to similar reconstructions in non-irradiated breasts^[61]. As such, many reconstructive surgeons opt to use autologous reconstruction in these patients. The latissimus dorsi flap (LD)^[62] is the most common pedicled flap option for autologous tissue salvage following free flap failure, as it does not require microsurgery and has a reliable anatomy with a versatile skin paddle that results in a high reconstructive success rate^[13,63] [Figure 4]. Though first described for postmastectomy reconstruction by D’Este in 1912^[63], it was popularized in the late 1970s as a method of autologous breast reconstruction. The latissimus dorsi flap allows for the harvest of a large skin paddle, useful in situations where there was a loss of skin from the prior failed free flap.

Figure 4. Patient with previous left breast cancer and latissimus dorsi flap reconstruction with new right breast cancer (Images 1 and 2). Salvage reconstruction was performed with right latissimus dorsi flap + tissue expander placement reconstruction (Images 3 and 4). Final postoperative result following implant exchange (Images 5 and 6).

Aesthetic results are often excellent, and the LD flap permits the use of healthy, non-radiated tissue to cover an implant if one is needed to achieve a size match^[64]. If greater volume is required, an implant or tissue expander can be placed under the flap, in one or two stages, either above or below the pectoralis muscle^[65]. In appropriate candidates, autologous fat grafting at the initial time of the LD flap can provide significant volume enhancement as needed^[66,67]. This is known as the Latissimus Dorsi and Immediate Fat Transfer (LIFT) procedure, where upwards of 500 cc of fat can be added for increased volume. When considering this option, ensure the thoracodorsal pedicle has not been injured from previous interventions, such as an axillary lymph node dissection or radiation^[68]. A recent retrospective review of 248 patients by Wattoo et al. showed long-term overall patient satisfaction from latissimus flap reconstruction^[69]. While minor complication rates were high in the short term (seroma 58% and wound infection 13%), chronic complications were low (shoulder stiffness 1.9%, pain 11.5%), highlighting the utility of this procedure. These results are consistent with another retrospective review of 277 patients by Yezhelyev et al., with higher short-term complications (seroma 19.5%), but overall low risk in the long term^[70].

CONVERSION TO ALLOPLASTIC

Alloplastic or implant-based reconstruction is another effective option for salvage, particularly in non-radiated breasts. When contemplating conversion to implant-based reconstruction, the psychological and emotional toll experienced after free flap failure should be considered, especially in women who specifically chose autologous tissue to avoid implants^[71,72]. The benefits of conversion to prosthetic reconstruction include a shorter hospital stay and a lower complication rate in the short term [Figure 5]^[73]. Factors that influence the decision to choose alloplastic reconstruction include the amount and quality of mastectomy skin available, the desired size of the breast and ultimately the patient’s wishes. Decisions about implant pocket placement (prepectoral or subpectoral) and use of a biologic mesh (coverage and support with or without acellular dermal matrix) need to be made. A history of radiation must be considered if choosing alloplastic reconstruction.

Figure 5. Patient with left breast cancer who underwent bilateral mastectomies and immediate reconstruction with DIEP flaps. The right flap failed (Images 1 and 2). After debridement and a period of healing, the breast was reconstructed with a subpectoral tissue expander (Images 3 and 4). The expander was later exchanged for a 700 cc silicone implant (Images 5 and 6).

The timing of initiation of expansion in the outpatient clinic is variable, depending upon surgeon preference and healing of the incision. The timing of adjuvant therapies must also be considered. If initial flap failure occurs in the immediate postoperative setting with several days to weeks of complications, a full course of tissue expansion may not allow for timely receipt of adjuvant radiation or chemotherapy. Expansion can begin as quickly as 10-14 days without an increase in complications^[74], though a common protocol is to begin expansion 3-4 weeks after tissue expander placement with exchange to permanent implants 3-6 months from tissue expander placement. Patients can then undergo revisions including fat grafting as needed to achieve optimal aesthetic results.

CONCLUSION/RECOMMENDATIONS

While less common in high-volume centers^[75], autologous free tissue transfer failures can be devastating in breast reconstruction; it is important to have secondary (and tertiary) options available. It is imperative to discuss the risks and benefits of the procedure with patients, including the availability of secondary options in the event of failure. Efforts should be made to build a trusting relationship with patients preoperatively and to provide emotional support postoperatively when failures occur. Second-line options, including free and pedicled flaps, implants, or a combination of both, should be entertained based on the clinical scenario with a balance of safety and achieving the patient’s overall reconstructive goals.