Autologous Breast Reconstruction with DIEP Or Pedicled TRAM Abdominal Flaps

EVIDENCE-BASED GUIDELINES AND MEASURES

Evidence-Based Clinical Practice Guideline: Autologous Breast Reconstruction with DIEP or Pedicled TRAM Abdominal Flaps

Bernard T. Lee, M.D., M.B.A., Summary: The American Society of Plastic Surgeons commissioned a multi- M.P.H. stakeholder Work Group to develop recommendations for autologous breast Jayant P. Agarwal, M.D. reconstruction with abdominal flaps. A systematic literature review was per- Jeffrey A. Ascherman, M.D. formed and a stringent appraisal process was used to rate the quality of rel- Stephanie A. Caterson, M.D. evant scientific research. The Work Group assigned to draft this guideline Diedra D. Gray, M.P.H. was unable to find evidence of superiority of one technique over the other Scott T. Hollenbeck, M.D. (deep inferior epigastric perforator versus pedicled transverse rectus abdomi- Seema A. Khan, M.D. nis musculocutaneous flap) in autologous tissue reconstruction of the breast Lauren D. Loeding, M.P.H. after mastectomy. Presently, based on the evidence reported here, the Work Raman C. Mahabir, M.D. Group recommends that surgeons contemplating breast reconstruction on Archibald S. Miller, M.D., M.S. their next patient consider the following: the patient’s preferences and risk Galen Perdikis, M.D. factors, the setting in which the surgeon works (academic versus commu- Jaime S. Schwartz, M.D. nity practice), resources available, the evidence shown in this guideline, and, Beth A. Sieling, M.D. equally important, the surgeon’s technical expertise. Although theoretical 2017 Achilles Thoma, M.D., M.Sc. superiority of one technique may exist, this remains to be reported in the Judith A. Wolfman, M.D. literature, and future methodologically robust studies are needed. (Plast. Jean L. Wright, M.D. Reconstr. Surg. 140: 651e, 2017.) Boston, Mass.; Salt Lake City, Utah; New York, N.Y.; Arlington Heights, Ill.; Durham, N.C.; Chicago, Ill.; Phoenix, Ariz.; Tulsa, Okla.; Jacksonville, Fla.; Beverly Hills, Calif.; Southbury, Conn.; Baltimore, Md.; and Hamilton, Ontario, Canada

ccording to the American Cancer Soci- According to procedural statistics from the ety, approximately one in eight women American Society of Plastic Surgeons, member Ain the United States will develop invasive surgeons performed 106,338 breast reconstruc- breast cancer in their lifetime, and an estimated tion procedures in 2015, a 35 percent increase 246,600 will be newly diagnosed in 2016 alone.1 from 2000. Among these procedures, 20,325 When breast-conserving surgery is not a viable were performed with autologous tissue, or option, a single or double mastectomy may be “flaps” taken from the abdomen, back, buttocks, performed. After mastectomy, several recon- or thigh to form the reconstructed breast.2 The structive treatment options are available to American Society of Plastic Surgeons Tracking patients. Operations and Outcomes for Plastic Surgeons3 program reports a consistent record of free From Beth Israel Deaconess Medical Center and Brigham flap and pedicled transverse rectus abdominis and Women’s Hospital; \University of Utah Health Care; Co- myocutaneous flap breast reconstruction pro- lumbia University Medical Center; American Society of Plas- cedures relative to the total number of proce- tic Surgeons; Duke University Medical Center; Northwestern dures entered annually. The American Society Memorial Hospital; Mayo Clinic; Oklahoma University; of Plastic Surgeons published the first clinical Beverly Hills Body Institute; Saint Mary’s Health System; Johns Hopkins University; and McMaster University. Received for publication February 17, 2017; accepted May Official Peer-Reviewed Publication 31, 2017. from the AMERICAN SOCIETY of Copyright © 2017 by the American Society of Plastic Surgeons PLASTIC SURGEONS. DOI: 10.1097/PRS.0000000000003768

practice guideline on breast reconstruction with (TRAM) flap—to treat breast defects associated expanders and implants in 2013.4 The present with the diagnosis or treatment of breast can- publication intends to expand on the breast cer. This guideline is intended to be used by reconstruction treatment options available by the multidisciplinary team that provides care providing evidence-based recommendations for for patients with breast cancer through the use the two most commonly performed autologous of breast cancer treatment, mastectomy, and breast reconstruction procedures based on the breast reconstruction. Health care practitioners Tracking Operations and Outcomes for Plastic should evaluate each case individually, consider- Surgeons program. ing these evidence-based treatment recommendations and patient values and preferences, to Scope and Intended Users determine the optimal treatment plan for each This evidence-based guideline is based on patient. This guideline is also intended to serve a systematic review of evidence and specifi- as a resource for health care practitioners and cally addresses the complications and patient developers of clinical practice guidelines and satisfaction of patients undergoing breast recommendations. reconstruction with autologous abdominal flap—specifically, the deep inferior epigas- Disclaimer tric perforator (DIEP) flap and the pedicled Evidence-based guidelines are strategies for transverse rectus abdominis musculocutaneous patient management, developed to assist physi- cians in clinical decision-making. This guideline was developed through a comprehensive review Disclosure: This clinical practice guideline was of the scientific literature and consideration funded exclusively by the American Society of Plas- of relevant clinical experience, and describes tic Surgeons; no outside commercial funding was a range of generally acceptable approaches to received to support the development of this article. diagnosis, management, or prevention of specific All contributors and preparers of the guideline, diseases or conditions. This guideline attempts including ASPS staff and consultants, disclosed to define principles of practice that should gen- all relevant conflicts of interest via an online dis- erally meet the needs of most patients in most closure reporting database. In accordance with circumstances. the Institute of Medicine’s recommendations for However, this guideline should not be con- guideline development, members with a conflict strued as a rule, nor should it be deemed inclu- of interest represented less than half of the guide- sive of all proper methods of care or exclusive line Work Group. Bernard T. Lee, M.D., M.B.A., of other methods of care reasonably directed M.P.H., Work Group Chair, has no relevant dis- at obtaining the appropriate results. It is antici- closures; Jayant P. Agarwal, M.D., has received pated that it will be necessary to approach some research support from Mentor Corporation, Life- patients’ needs in different ways. The ultimate Cell Corporation, DePuy Synthes, and NIH, as judgment regarding the care of a particular the PI in grants funded by DSM Biomedical, and patient must be made by the physician in light of served as a consultant for DonJoy Orthopedics; all the circumstances presented by the patient, Jeffrey A. Ascherman, M.D., Stephanie A. Ca- the available diagnostic and treatment options, terson, M.D., Diedra D. Gray, M.P.H., Scott T. and available resources. Hollenbeck, M.D., Seema A. Khan, M.D., Lauren This guideline is not intended to define or D. Loeding, M.P.H., Raman C. Mahabir, M.D., serve as the standard of medical care. Standards and Archibald S. Miller, M.D., have no relevant of medical care are determined on the basis of disclosures; Galen Perdikis, M.D., has served as all the facts or circumstances involved in an indi- a teacher for IHE; Jaime S. Schwartz, M.D., has vidual case and are subject to change as scientific received research support from Covidien, Ltd, and knowledge and technology advance and as prac- served on the Advisory Board of Mentor Corpora- tice patterns evolve. This guideline reflects the tion, receiving honorarium; Beth A. Sieling, M.D., state of current knowledge at the time of publi- has served as a consultant for Myriad and Genom- cation. Given the inevitable changes in the state ic Health; Achilles Thoma, M.D., has no relevant of scientific information and technology, this disclosures; Judith A. Wolfman, M.D., has served guideline will be considered relevant for a period on the Advisory Board of Hologic; Jean L. Wright, of 5 years after publication, in accordance with M.D., has no relevant disclosures. the inclusion criteria of the National Guideline Clearinghouse.

BACKGROUND which may have an effect on the breast and chest Autologous breast reconstruction using abdomi- wall. Related insurance coverage criteria can be nal tissue is a common reconstructive procedure with found in Appendix 1. widespread acceptance and a long history of success. Over the past 50 years, the techniques for perform- Physical Examination ing abdominally based breast reconstruction have Physical examination of the breast defect evolved. Historically, pedicled TRAM flap breast should include documentation of breast size and reconstruction was first described in the 1980s. The configuration of any missing tissue. The presence of evolution from pedicled TRAM, to free TRAM, to scarring and radiation changes and the condition DIEP flap reconstruction follows a shift from muscu- of the pectoralis major muscle, nipple-areola com- locutaneous flaps to muscle-sparing perforator flaps. plex, and contralateral breast should also be noted. Surgeons typically perform either pedicled TRAM or free DIEP flap procedures depending on their METHODS personal experience, comfort level with each procedure, and capabilities of their surgical facilities. It is Work Group Selection Process important to note that DIEP flaps require additional American Society of Plastic Surgeons mem- technical skill and institutional infrastructure for bers were invited to apply to the Work Group by microsurgery. As such, surgeons and patients some- means of Society e-mail and fax communication. times find it difficult to determine which procedure All applicants were required to submit an online provides the most acceptable outcome for a given conflict-of-interest disclosure form for member- patient. Thus, a systematic review was performed to ship consideration. Members of the American identify the most relevant evidence to address these Society of Plastic Surgeons Quality and Perfor- questions, with the aim of providing evidence-based mance Measurement Committee reviewed and recommendations to guide surgeons and patients selected Work Group members to ensure a diverse in decision making for breast reconstruction. When representation of U.S. regions; practice type considering the recommendations in this guideline, (large multispecialty group practice, small group health care providers and patients should note that practice, solo practice, and academic practice); the studies used to develop the recommendations and clinical, research, and evidence-based medi- are retrospective and observational. No prospective cine experiences and expertise. Four stakeholder or randomized studies were identified for the clini- organizations—the American Society of Breast cal questions included in this guideline. Surgeons, the American College of Radiology, the American Society of Clinical Oncology, and the Definitions American Society for Radiation Oncology—were Pedicled TRAM flap: abdominally based flap con- also invited to participate in the guideline devel- taining skin, fat, and muscle that is partially opment process by nominating one member from resected and tunneled to the breast with the use their respective organizations to serve on the Work of nonmicrosurgical methods. Group. A patient representative was included on DIEP flap: abdominally based, muscle-sparing flap the panel to provide insight related to patient val- containing skin and fat that is removed and reat- ues and preferences, and an American Society of tached to the chest with the use of microsurgery. Plastic Surgeons quality department staff member was assigned to manage the project and provide Diagnostic Criteria expertise in clinical practice guideline develop- The patient usually presents to the plastic sur- ment methodology. geon’s office with a previous diagnosis of and/or treatment for breast cancer, or may be undergo- Clinical Question Development ing a prophylactic mastectomy. Patients who have Work Group members used a consensus-based had breast cancer may have had only a biopsy of approach to select the clinical questions to be the mass, a lumpectomy, or a mastectomy (alone addressed in this evidence-based guideline. Work or with axillary lymph node biopsy or dissection). Group members used a blinded process to submit Any of these surgical treatments may have been clinical questions by means of individual e-mail to supplemented with radiation treatment to the the American Society of Plastic Surgeons project breast/chest wall with or without regional lymph manager, who compiled and dispersed the clinical nodes, and systematic therapies including chemo- questions for consideration and discussion at the therapy, immunotherapy, and endocrine therapy, introductory meeting. The clinical questions were

selected with a five-phase process that consisted of • MeSH terms (used in PubMed only): brainstorming, discussion, ranking/prioritizing, “Abdomen”[MeSH], “Abdominal Wall” refining, and voting. [MeSH], “Free Tissue Flaps”[MeSH], “Hema- A total of 36 clinical questions were reviewed toma” [MeSH], “Hernia”[MeSH], “Infection” by the Work Group and ranked according to the [Mesh], “Mammaplasty”[MeSH], “Necrosis” following criteria to assess for potential impact: [MeSH], “Patient Outcome Assessment” (1) relevance to guideline scope; (2) addresses a [MeSH], “Patient Satisfaction”[MeSH], Post- gap in care; (3) ability to develop into an action- operative Complications”[MeSH], “Pulmonary able recommendation; (4) ability to develop embolism”[MeSH], “Reoperation”[MeSH], into an implementable recommendation; (5) is “Risk”[MeSH], “Second-look Surgery”[MeSH], controversial or of significant interest; and (6) “Seroma”[MeSH], “Surgical Flaps”[MeSH], is important to public health. The Work Group “Surgical Mesh”[MeSH], “Surgical Wound initially agreed on 11 clinical questions; however, Dehiscence”[MeSH], “Treatment Outcome” the large scope of the overall topic of autologous [MeSH], and “Venous Thrombosis”[MeSH]. breast reconstruction would not allow for a timely • Keywords: Abdominal flap, abdominal free guideline. In 2016, the guideline was narrowed flap, abdominal pedicled flap, abdomi- and the original 11 clinical questions were refined nal weakness, autologous breast recon- into the following two clinical questions: struction, bulge, complications, deep vein 1. In patients undergoing mastectomy and thrombosis, flap failure, outcomes, and autologous breast reconstruction, which patient satisfaction. surgical technique, pedicled TRAM flap versus DIEP flap, is associated with the lower Initial study selection for each clinical ques- incidence of clinical complications? tion was performed by one reviewer with a two- 2. In patients undergoing mastectomy and level screening process. Level I screening involved autologous breast reconstruction, which a review of the title and abstracts of the articles surgical technique, pedicled TRAM flap captured by the search strategies, to identify versus DIEP flap, is associated with the high- potentially relevant studies for inclusion in level est level of patient satisfaction? II screening. Level II screening involved a review of the full-text of articles to confirm relevance Thus, the methodology and results described and compare study details with the inclusion and herein relate to the review of data and the devel- exclusion criteria below. opment of recommendations for these clinical questions only. The remaining clinical questions Inclusion Criteria: may be considered for future guidelines. • Published within the past 10 years (January 1, 2003, to June 14, 2014). Literature Search • Published in English language. The literature search was performed between • Reported a meta-analysis/systematic review; 2012 and 2014 and aimed to identify relevant randomized controlled trial; prospective studies published during the previous 10-year or retrospective cohort/comparative, case- period (January of 2003 to June of 2014). Elec- control, or case series. tronic searches of PubMed, Cochrane Library, • Reported outcomes of interest for clinical and Cumulative Index to Nursing and Allied questions (complications and/or patient Health Literature databases were performed. satisfaction). The journal Plastic and Reconstructive Surgery • Included at least 30 patients. Global Open was searched separately, as publica- Exclusion Criteria tions from Plastic and Reconstructive Surgery Global Open were not indexed in the selected databases • Published outside of inclusion date range. at the time of this review. Literature searches • Published in language other than English. were performed by using appropriate combina- • Reported a case report, economic analy- tions of the following MEDLINE Medical Sub- sis, animal study, cadaver study, narrative ject Headings (MeSH) terms and keywords, as review, or editorial. permitted by the search functionalities of each • Reported no outcomes of interest. database/journal: • Included fewer than 30 patients.

The bibliographies of articles meeting inclu- each flap type. The following complications were sion criteria were manually searched to identify evaluated, if reported in the studies: relevant articles missed during the electronic searches. These articles were screened as described • Donor-site complications: hernia, bulge, above. Duplicate articles were eliminated. Studies infection, necrosis, seroma, hematoma, meeting inclusion criteria were assessed for meth- and wound dehiscence. odologic quality, as described below. Excluded • Flap-related complications: flap loss, necro- studies and their reasons for exclusion were doc- sis, infection, seroma, hematoma, and umented for review by the Work Group to con- wound dehiscence. firm the final rejection or reconsider the study for • Systemic complications: venous thrombo- inclusion. Additional references were included in embolism, including deep vein thrombosis this review if considered necessary for background and/or pulmonary embolism. or discussion; however, these references were not • Procedure-related complications: revision/ critically appraised or used in the development of reoperation and reconstruction failure rate. recommendation statements. Patient satisfaction was evaluated differently Critical Appraisal of Evidence among the included studies. Because of the num- The American Society of Plastic Surgeons evi- ber and variety of scales used for assessing patient dence-based process includes a rigorous critical satisfaction, the reported scales were grouped appraisal process to evaluate the methodologic into three categories: Michigan Breast Satisfaction quality of clinical studies and the strength of clini- Questionnaires, 10-point Likert scales, and other cal evidence for the purposes of developing clini- (e.g., Short-Form 36-Item Health Survey, Quali- cal practice guidelines and performance measures. tative Assessment of Back Pain). The 10-point The process is also used to rate individual studies Likert scales were assessed similarly to the Michi- published in Plastic and Reconstructive Surgery. Stud- gan Breast Satisfaction Questionnaire by separat- ies were appraised for methodologic quality with ing the level of satisfaction into binary groups (1 the American Society of Plastic Surgeons Critical through 7 = not satisfied; 8 through 10 = satisfied). Appraisal Checklists and assigned levels of evidence according to the American Society of Plastic Sur- Grading of Recommendations geons Evidence Rating Scales, which are designed Clinical practice recommendations were for the evaluation of therapeutic, prognostic/risk, developed through a consensus process with con- and diagnostic studies (see Appendix 2 for scales). sideration to the following three factors: (1) level The checklists and scales were developed in 2009 of evidence (study quality); (2) assessment of ben- by an expert Task Force and are based on the prin- efits versus harms; and (3) patient preferences. ciples of the Critical Appraisal Skills Programme Work Group members jointly drafted statements and the Centre for Evidence Based Medicine. for each recommendation during conference call Each study was appraised by at least two reviewers. meetings and online discussions. After each meet- If a discrepancy existed between the reviewers, the ing, members had an opportunity to individually study was appraised by a third reviewer, and the comment and revise the draft recommendations level of evidence was determined by consensus. by means of e-mail discussion. Work Group mem- Evidence ratings were not assigned to studies with bers participated in several rounds of revisions inadequately described methods and/or worri- until unanimous consensus was achieved for each some biases. As such, these studies were excluded recommendation statement. Each recommenda- from further review. tion in this guideline is accompanied by a grade indicating the strength of the recommendation, Data Extraction and Outcomes Definitions which was determined by considering the overall Quantitative and qualitative data relevant to level of evidence supporting the recommendation the clinical questions were extracted from the and the judgment of the guideline developers. studies that met inclusion criteria and qualified for a level-of-evidence rating. Data were compiled Peer Review and Public Comment Process in Excel (Microsoft Corp., Redmond, Wash.) The draft guideline was peer reviewed by the spreadsheets. American Society of Breast Surgeons, the Ameri- Quantitative data on complication outcomes can College of Radiology, the American Society of were pooled across the studies to calculate the Clinical Oncology, and the American Society for probability of the complication occurring for Radiation Oncology. American Society of Plastic

Surgeons members of the Quality and Perfor- satisfaction data were used to develop practice mance Measurement and Healthcare Delivery recommendations. The recommendations listed Committees were also invited to participate in the below were based on level III and IV evidence. A peer review process. Peer reviewers were invited summary of recommendation statements is shown to review and provide feedback on the validity, in Table 2.5–24 generalizability, and clarity of the draft guideline using the Appraisal of Guidelines for Research & Recommendations Related to Clinical Evaluation II instrument. After peer review, the Complications draft guideline was posted on the American Soci- 1. The Work Group suggests that clinicians ety of Plastic Surgeons Web site for a 2-week pub- may treat patients undergoing mastectomy and lic comment period. autologous breast reconstruction with either surgical technique (pedicled TRAM flap or DIEP Guideline Approval Process flap, contingent on the use of mesh for pedicled After the peer review and public comment TRAM procedures) because the risk of donor-site processes, the guideline draft was reviewed and complications is comparable among procedures. modified by the Work Group in consideration Patient preference should have a substantial of peer review and public comments. The final influencing role. guideline was approved by the American Society of Plastic Surgeons Executive Committee during Level III, IV Evidence its meeting in December of 2016. Recommendation Grade: C Donor-site morbidity includes hernia, bulge, Plan for Updating Guideline infection, necrosis, seroma, hematoma, or wound dehiscence. The pooled evidence from 15 studies In accordance with the inclusion criteria of the (some of which used mesh and some of which did National Guideline Clearinghouse, this guideline not use mesh for the abdominal closure) suggests will be updated within 5 years to reflect changes in that there is a higher probability of hernia with the scientific evidence, practice parameters, and treatment options. pedicled TRAM flap (pedicled TRAM flap, 3.50 percent; DIEP flap, 0.74 percent) and a slightly higher rate of bulging with the DIEP flap (DIEP RESULTS AND RECOMMENDATIONS flap, 4.62 percent; pedicled TRAM flap, 3.50 per- A total of 564 studies for clinical question 1 and cent).5–19 Of note, this comparison did not exam- 267 studies for clinical question 2 were retrieved ine free TRAM flap reconstruction. According to a through the literature search. After screening and comparative study of bilateral reconstruction, zero critical appraisal were performed, 20 studies were of 58 patients (0 percent) developed a hernia with selected for final review for this guideline (Figs. 1 the DIEP flap, whereas three of 105 (2.9 percent) and 2). Each study reported at least one outcome developed a hernia with the pedicled TRAM flap. of interest (complications and/or patient satis- Similarly, four of 58 patients (6.9 percent) devel- faction); 18 studies reporting clinical complica- oped a bulge with the DIEP flap and three of 105 tions data and eight studies reporting patient (2.9 percent) developed a bulge with the pedicled

Table 1. American Society of Plastic Surgeons Scale for Grading Recommendations Grade Descriptor Qualifying Evidence Implications for Practice A Strong recommendation Level I evidence or consistent Clinicians should follow a strong recommendation findings from multiple studies unless a clear and compelling rationale for an of levels II, III, or IV alternative approach is present. B Recommendation Levels II, III, or IV evidence Generally, clinicians should follow a recommenda- and findings are generally tion but should remain alert to new information consistent and sensitive to patient preferences. C Option Levels II, III, or IV evidence, but Clinicians should be flexible in their decision- findings are inconsistent making regarding appropriate practice, although they may set bounds on alternatives; patient preference should have a substantial influencing role. D Option Level V: Little or no systematic Clinicians should consider all options in their empirical evidence decision-making and be alert to new published evidence that clarifies the balance of benefit vs. harm; patient preference should have a substantial influencing role.

Fig. 1. Literature search process chart for clinical question 1. *Limits set in PubMed included publication date, human subjects, English language, and study types; the functionalities of the other databases did not allow for limit setting.

TRAM flap.5 Both complications (i.e., hernia rate was 1.3 percent versus 4.3 percent, respec- or bulging) could be a result of whether or not tively; and the bulge rate was 3.6 percent versus mesh was used at the time of reconstruction. The 2.8 percent.5,7,9,10,12,14,15,17 In the four-case series of reported probabilities of these complications are pedicled TRAM flaps that specified what type of likely also dependent on variables other than flap mesh was used, all used a synthetic (polypropyl- type (e.g., surgical technique or the use of mesh). ene) mesh, some in a folded-over fashion.14–17 In The evidence suggests that a higher probability of a comparative study, hernia rates were higher in wound dehiscence, seroma, hematoma, and skin pedicled TRAM flaps compared with DIEP flaps necrosis was associated with the DIEP flap. (16 percent versus 1 percent), but 86 percent of Hernia rates were less than 3 percent in the those patients did not have mesh.6 Bulge rates in five-case series of DIEP flaps8,9,11–13 and four-case the included case series were 0 to 5 percent in series of pedicled TRAM flaps examined.14,16,17,19 DIEP flaps9–13 and 0.5 to 5.7 percent in pedicled When examining pooled data of bilateral DIEP TRAM flaps.14–18 Garvey and colleagues dem- and bilateral pedicled TRAM flaps, the hernia onstrated higher bulge rates in both groups,

Fig. 2. Literature search process chart for clinical question 2. *Limits set in PubMed included publication date, human subjects, English language, and study types; the functionalities of the other databases did not allow for limit setting.

but patients did not receive mesh.6 Overall cal- TRAM flaps, although this has not been shown to culated probabilities of donor-site infection, affect daily activities and was not documented in necrosis, seroma, hematoma, and wound dehis- the examined articles. If a patient sees a surgeon cence were all higher with DIEP flaps than with who is more experienced in one technique over pedicled TRAM flaps, but these differences were the other, the majority of well-informed patients minimal. would most likely use this patient-care strategy, Potential benefits of each procedure include compared to alternative patient-care strategies shorter operative times with pedicled TRAM or no treatment. Clinicians should be flexible in flaps; and preservation of the rectus muscle with their decision-making process, and risk factors DIEP flaps. Potential harms include increas- and patient preferences should be considered. ing operative risks inherent in longer operative The setting in which clinicians practice (e.g., aca- times of DIEP flap procedures and a theoretical demic or community practice) may also have a decrease in abdominal strength with pedicled substantial influencing role.

Table 2. Recommendation Statements for Autologous Breast Reconstruction with DIEP or Pedicled TRAM Abdominal Flaps Level of Assessment of Grade; Strength of Recommendation Evidence Benefits/Harms Recommendation References 1. The Work Group suggests that clinicians may treat III, IV The balance C; Option 5–19 patients undergoing mastectomy and autologous between breast reconstruction with either surgical technique benefits and (pedicled TRAM flap or DIEP flap, contingent on harms is the use of mesh for pedicled TRAM flap procedures) unclear because the risk of donor-site complications is comparable among procedures. Patient preference should have a substantial influencing role. 2. The Work Group suggests that clinicians may treat III, IV The balance C; Option 5–9, 11–13, patients undergoing mastectomy and autologous between 16–20 breast reconstruction with either surgical technique benefits and (pedicled TRAM flap or DIEP flap) because the risk harms is of flap-related complications is comparable among unclear procedures. Patient preference should have a substantial influencing role. 3. Based on little or no systematic empirical evidence, IV The balance D; Option 9, 11–13, 18 it is the consensus of the Work Group that clinicians between may treat patients undergoing mastectomy and benefits and autologous breast reconstruction with either surgical harms is technique (pedicled TRAM flap or DIEP flap) since unclear the risk of systemic complications (DVT and PE) is indeterminate among procedures. 4. Based on little or no systematic empirical evidence, IV The balance D; Option 7, 11–13, 17–19, it is the consensus of the Work Group that clinicians between 21, 22 may treat patients undergoing mastectomy and benefits and autologous breast reconstruction with either surgi- harms is cal technique (pedicled TRAM flap or DIEP flap) unclear because the risk of revision/reoperation and reconstruction failure is indeterminate among procedures. 5. Based on little or no systematic empirical evidence, IV The balance D; Option 5, 7, 13, 16, 18, it is the consensus of the Work Group that clinicians between 22–24 may treat patients undergoing mastectomy and benefits and autologous breast reconstruction with either surgi- harms is cal technique (pedicled TRAM flap or DIEP flap) unclear because there were no differences in patient satisfaction noted. However, it was found that the level of patient satisfaction is high for both procedures.

2. The Work Group suggests that clinicians this difference does not reach statistical signifi- may treat patients undergoing mastectomy and cance in most studies. A case series reported 16 autologous breast reconstruction with either cases (8.5 percent) of partial flap loss in pedi- surgical technique (pedicled TRAM flap or DIEP cled TRAM procedures, with a high incidence of flap) because the risk of flap-related complications is fat and skin necrosis.17 According to Kim et al., comparable among procedures. Patient preference 72 of 505 patients (14 percent) experienced fat should have a substantial influencing role. necrosis and 75 of 505 (15 percent) experienced skin necrosis.18 The DIEP flap was associated Level III, IV Evidence with slightly higher infection rates and a higher Recommendation Grade: C probability of seroma, hematoma, and wound Flap-related complications include flap loss, dehiscence. However, some studies demonstrated necrosis, infection, seroma, hematoma, and similar outcomes between groups.6 For example, wound dehiscence. The evidence from 13 stud- one of 58 DIEP flap patients (1.7 percent) devel- ies5–9,11–13,16–20 suggests that the rate of total flap oped a seroma, whereas only two of 105 patients loss is similar between pedicled TRAM and DIEP (1.9 percent) developed a seroma with the pedi- flaps. Of note, pedicled TRAM flaps that previ- cled TRAM flap.5 This systematic review was not ously underwent a delay procedure were included able to comment on differences in surgical tech- in some studies, but were not considered sepa- nique that could improve outcomes for either rately in this analysis. Most studies suggest a group to reduce flap-related complications. slightly increased rate of flap loss with DIEP flaps Evidence comparing flap loss rates between compared with pedicled TRAM flaps, although DIEP and pedicled TRAM flaps for breast

reconstruction is limited. Two retrospective com- balance between benefits and harms between parative studies that evaluated flap loss (partial DIEP and pedicled TRAM flap techniques with or complete) between DIEP and pedicled TRAM regard to fat necrosis is unclear. flaps found no statistically significant difference Evidence comparing infection rates, seroma/ between the techniques. Flap loss ranged from hematoma rates, and wound dehiscence rates 1.7 to 3.1 percent in the DIEP flap groups and between DIEP flaps and pedicled TRAM flaps for from 0 to 8.5 percent in the pedicled TRAM flap breast reconstruction is limited. In a comparative groups.5,6 Five retrospective case series reported study, rates of infection (DIEP, 12.5 percent; pedi- rates of flap loss in DIEP flap breast reconstruc- cled TRAM, 17.0 percent) and wound dehiscence tion7–9,11,12 and three in pedicled TRAM flap breast (DIEP, 12.5 percent; pedicled TRAM, 13.8 per- reconstruction.16–18 In DIEP flap cases, rates of cent) were similar between DIEP flap and pedicled partial flap loss ranged from 1.8 to 4.7 percent,8,9,13 TRAM flap groups.6 With regard to seroma and and rates of total flap loss ranged from 0 to 4.7 hematoma, two comparative studies found no sta- percent.7–9,11–13 In pedicled TRAM flap cases, Ire- tistically significant difference between DIEP flap ton and colleagues reported an 8.5 percent inci- cases and pedicled TRAM flap cases.5,6 Small case dence of partial flap loss, whereas the pooled rate series of DIEP or pedicled TRAM flaps report an of total flap loss ranged from 0 to 0.2 percent.16–18 overall pooled seroma rate of 5.1 percent for DIEP Given the small number of comparative studies flaps versus 2.4 percent for pedicled TRAM flaps comparing DIEP and pedicled TRAM flap recon- and an overall pooled hematoma rate of 3.2 per- structive procedures and the low overall flap loss cent for DIEP flaps versus 2.5 percent for pedicled rates observed with these techniques, clinicians TRAM flaps.7–9,11–13,16–20 Given the small number should be flexible in their decision-making pro- of studies comparing DIEP and pedicled TRAM cess, and patient risk factors, patient preferences, flap techniques and the low overall rates of infec- and the setting in which clinicians practice should tion, seroma/hematoma, and wound dehiscence have a substantial influencing role. The balance between these reconstructive techniques, clini- between benefits and harms between DIEP and cians should be flexible in their decision-making pedicled TRAM flap techniques with regard to process, and patient risk factors, patient prefer- flap loss is unclear. ences, and the setting in which clinicians practice Evidence comparing fat necrosis rates between should have a substantial influencing role. There DIEP flaps and pedicled TRAM flaps for breast are limited data regarding the impact of postmas- reconstruction is limited. Two comparative stud- tectomy radiation therapy on flap complications, ies retrospectively analyzed outcomes after DIEP and neither technique emerged as superior in the or pedicled TRAM flap–based breast reconstruc- setting of radiation therapy. The balance between tion. Chun and colleagues compared bilateral benefits and harms between DIEP and pedicled pedicled TRAM flaps with bilateral DIEP flaps and TRAM flap techniques with regard to other flap- found a significantly higher fat necrosis rate in the related complications is unclear. DIEP group (19.8 percent versus 11.4 percent; p = 3. Based on little or no systematic empirical 0.04).5 Garvey and colleagues compared unilateral evidence, it is the consensus of the Work Group that DIEP flaps with unilateral pedicled TRAM flaps clinicians may treat patients undergoing mastectomy and reported a fat necrosis rate of 17.7 percent for and autologous breast reconstruction with either DIEP flaps and 58.5 percent for pedicled TRAM surgical technique (pedicled TRAM flap or DIEP flaps p( < 0.001).6 Several case series reported fat flap) because the risk of systemic complications necrosis rates in DIEP or pedicled TRAM flaps; (deep vein thrombosis and pulmonary embolism) rates ranged between 2 and 12.5 percent for DIEP is indeterminate among procedures. flaps7–9,11–13 and between 7.9 and 14.5 percent for pedicled TRAM flaps.16–20 These mixed findings Level IV Evidence do not point to a specific technique that provides Recommendation Grade: D superior fat necrosis rates. These studies are lim- Systemic complications include deep vein ited in that not all report on the number of per- thrombosis and pulmonary embolism. The evi- forators captured with each DIEP flap. Given the dence from five studies9,11–13,18 is insufficient to lack of information and inconsistent results, clini- offer a recommendation because of the lack of cians should be flexible in their decision-making reporting on this outcome with the DIEP flap. process, and patient risk factors, patient prefer- The probability of deep vein thrombosis is het- ences, and the setting in which clinicians practice erogeneous between the two flaps. The evidence should have a substantial influencing role. The suggests a 1.6 percent probability of pulmonary

embolism with the pedicled TRAM flap,18 but the The Work Group suggests that clinicians should rate of pulmonary embolism with the DIEP flap be flexible in their decision-making. Patient-spe- is not reported. The balance between benefits cific risk factors, preferences, and the setting in and harms between these two reconstructive tech- which clinicians practice should have a substantial niques with regard to deep vein thrombosis and influencing role. pulmonary embolism is unclear. No systematic empirical evidence is available Recommendations for Patient Satisfaction to guide clinical practice regarding venous throm- Reviewed patient satisfaction indicators boembolism prevention in autologous breast include the Michigan Breast Satisfaction Ques- reconstruction with DIEP or pedicled TRAM tionnaire, overall patient satisfaction (10-point flaps. It is the consensus of the Work Group that scale), the Qualitative Assessment of Back Pain, clinicians should follow the previously developed and various reports of single scales (e.g., Func- American Society of Plastic Surgeons report on tional Assessment of Cancer Therapy-Breast, qual- venous thromboembolism prophylaxis.25 ity of life assessment, aesthetic satisfaction). Of the 4. Based on little or no systematic empirical eight studies that met inclusion criteria, only two evidence, it is the consensus of the Work Group directly compared patient satisfaction in patients that clinicians may treat patients undergoing undergoing DIEP flap reconstruction versus those mastectomy and autologous breast reconstruction undergoing pedicled TRAM flap reconstruction. with either surgical technique (pedicled TRAM The Work Group concluded that more compara- flap or DIEP flap) because the risk of revision/ tive data are needed to evaluate patient satisfac- reoperation and reconstruction failure is tion in this therapeutic area. Multiple patient indeterminate among procedures. satisfaction scales were used among the eight studies; therefore, the Work Group elected to group Level IV Evidence the scales into three categories, as previously Recommendation Grade: D described, to analyze the data effectively. A high Procedure-related complications include revi- level of patient satisfaction was reported in all sion/reoperation and reconstruction failure. The studies, with no differences noted when compar- evidence from nine studies7,11–13,17–19,21,22 is insuf- ing DIEP with pedicled TRAM flaps. In addition, ficient to offer a recommendation because of no differences were noted when comparing DIEP minimal reporting on these outcomes with the with pedicled TRAM flaps in patient-reported pedicled TRAM flap. The probability of revision/ incidence of back pain. reoperation is heterogeneous between the two 5. Based on little or no systematic empirical flaps. Weak evidence suggests that the revision/ evidence, it is the consensus of the Work Group reoperation rate may be higher with a pedicled that clinicians may treat patients undergoing TRAM flap than with a DIEP flap.11–13,17,19,22 The mastectomy and autologous breast reconstruction evidence suggests a 2.1 percent reconstruction with either surgical technique (pedicled TRAM flap failure rate with the DIEP flap11 and a 0.2 percent or DIEP flap) because there were no differences in failure rate with the pedicled TRAM flap.18 patient satisfaction noted. However, it was found The available data come from case series. that the level of patient satisfaction is high among Importantly, the potential for publication bias both procedures. exists regarding the timing of publication; case series of pedicled TRAM flaps were typically pub- Level IV Evidence lished a decade or more ago, whereas case series Recommendation Grade: D of DIEP flaps were published more recently, The potential for a negative impact on satis- and because studies published before 2003 were faction as it relates to donor-site morbidity and excluded from this guideline, some relevant ped- flap fat necrosis were considered in these studies. icled TRAM flap case series may not have been In addition, several articles come from the same considered for this guideline. In addition, lower institution and are thus somewhat redundant. reoperation rates may represent refinements In evaluating a study, the variables that might be in reconstructive techniques over time for pedi- associated with benefit and harm include flap cled TRAM flaps or true differences in reopera- failure, fat necrosis, need for secondary surgery, tion rates between the techniques. The balance decreased function in abdominal wall muscles, between benefits and harms between these two and satisfaction with reconstruction. The eight reconstructive techniques with regard to revision/ studies selected for this question focus on sat- reoperation and reconstruction failure is unclear. isfaction with reconstruction.5,7,13,16,18,22–24 The

balance between benefit and harm with regard radiation therapy on complication rates between to this particular aspect is unclear. In addi- the types of autologous reconstruction, radiation tion, if a patient sees a surgeon who is more therapy is known to impact complication rates in experienced in one technique over the other, the setting of any type of reconstruction. A mul- the majority of well-informed patients would tidisciplinary approach is key, with integration of most likely use this patient-care strategy, com- radiation oncology early in the planning process pared with alternative patient-care strategies or for patients who may require postmastectomy no treatment. Clinicians should be flexible in radiation therapy. The Work Group strongly rec- their decision-making process, and patient risk ommends that centers performing a high volume factors, patient preferences, and the setting in of breast reconstructions undertake compara- which clinicians practice should have a substan- tive studies in which the population, procedures tial influencing role. (standard versus novel), setting (community ver- One important aspect of breast reconstruc- sus academic), outcomes, and period of the study tion is patient satisfaction with the reconstructive are clearly defined. approach and outcome. Comparisons of patient In terms of outcome reporting, future inves- satisfaction between pedicled TRAM and DIEP tigators should use validated patient-reported flaps have been difficult to consolidate in the past. outcome measures such as the BREAST-Q. In In general, it can be concluded that both pedi- addition, authors should consider coupling eco- cled TRAM and DIEP flaps are associated with a nomic evaluations to future prospective studies high level of patient satisfaction. At the time of to determine whether the novel techniques are this review, no studies have been published that cost-effective from the patient, third-party payer, used patient-reported outcome measures such and societal perspectives. These cost studies and as the BREAST-Q to compare outcomes between economic analyses should specifically compare pedicled TRAM and DIEP flaps. the two procedures. Future investigators pro- claiming superiority, noninferiority, or equiva- lence of breast reconstruction techniques must CONCLUSIONS AND FUTURE ensure that the studies evaluating the techniques DIRECTIONS have been performed with a randomized con- The Work Group assigned to draft this guide- trolled trial design and have followed the guideline was unable to find evidence of superiority of lines established by the Consolidated Standards one technique over the other (DIEP versus pedi- of Reporting Trials group, specifically address- cled TRAM flap) in autologous tissue reconstruc- ing the variations of randomized controlled trial tion of the breast after mastectomy. The available design. evidence addressing the questions posed in this Bernard T. Lee, M.D., M.B.A., M.P.H. guideline varied from level III to IV and the rec- Division of Plastic and Reconstructive Surgery ommendations provided were graded C or D Beth Israel Deaconess Medical Center level. Although theoretically such superiority may Lowry Medical Office Building exist, this remains to be seen until future, method- 110 Francis Street, Suite 5A ologically robust studies are undertaken. Unfor- Boston, Mass. 02215 tunately, the outcomes research movement of the [email protected] past three decades has received belated attention by organized plastic surgery. This was clear in the ACKNOWLEDGMENTS paucity of comparative studies reporting the criti- The authors thank Jennifer A. Swanson, B.S., cal outcomes of reconstruction or outcomes from M.Ed., of JS Medical Communications, LLC, for con- the patient’s perspective. tributions to the guideline; and Tamarin Lindenberg for Presently, based on the evidence reported serving on the Work Group as a patient representative. here, the Work Group recommends that sur- The authors thank the various organizations and indi- geons contemplating breast reconstruction on viduals who participated in the peer review and public their next patient consider the following: the comment periods: Linda Moy, M.D., American College patient’s preferences and risk factors; the set- of Radiology; Roshni Rao, M.D., American Society of ting in which the surgeon works (academic ver- Breast Surgeons; Kathleen Horst, M.D., American Soci- sus community practice); resources available; the ety for Radiation Oncology; Elizabeth Mittendorf, M.D., evidence shown in this guideline; and, equally and Monica Morrow, M.D., American Society of Clini- important, the surgeon’s technical expertise. cal Oncology; and others. Every comment was reviewed Despite the limited data regarding the impact of and considered by the Work Group.

APPENDIX 1. INSURANCE COVERAGE APPENDIX 2. AMERICAN SOCIETY CRITERIA OF PLASTIC SURGEONS EVIDENCE RATING SCALES

International Classification of Diseases, 10th Revision, Evidence Rating Scale for Therapeutic Studies Clinical Modification Codes Level of ICD-10-CM Evidence Qualifying Studies Diagnosis Code I High-quality, multi-centered or single-centered, Malignant neoplasm of female breast C50.01– randomized controlled trial with adequate Malignant neoplasm of male breast C50.02– power; or systematic review of these studies Secondary malignant neoplasm of other speci- C79.81 II Lesser-quality, randomized controlled trial; fied sites; breast prospective cohort or comparative study; or Carcinoma in situ of breast D05.9– systematic review of these studies Capsular contracture of breast implant N64.89 III Retrospective cohort or comparative study; Unspecified abnormal mammogram R92.8 case-control study; or systematic review of Acquired absence of breast Z90.1– these studies Encounter for breast reconstruction following Z42.1 IV Case series with pretest/posttest; or only mastectomy posttest Personal history of malignant neoplasm of Z85.3 V Expert opinion developed via consensus pro- breast cess; case report or clinical example; cadaver Family history of malignant neoplasm of breast Z80.3 study; or evidence based on physiology, Genetic susceptibility to malignant neoplasm of Z15.01 bench research, or “first principles” breast ICD-10-CM, International Classification of Diseases, 10th Revision,Clinical Modification. Evidence Rating Scale for Diagnostic Studies Level of Evidence Qualifying Studies Current Procedural Terminology Codes I High-quality, multi-centered or single-centered, CPT cohort study validating a diagnostic test (with Procedure Codes “gold” standard as reference) in a series of consecutive patients; or a systematic review of Breast reconstruction with free flap 19364 these studies Breast reconstruction with TRAM flap, single 19367 II Exploratory cohort study developing diagnostic pedicle, including closure of donor site criteria (with gold standard as reference) in a Breast reconstruction with TRAM flap, single 19368 series of consecutive patients; or a systematic pedicle, including closure of donor site; with review of these studies microvascular anastomosis (supercharging) III Diagnostic study in nonconsecutive patients Breast reconstruction with TRAM flap, double 19369 (without consistently applied gold standard as pedicle, including closure of donor site reference); or a systematic review of these studies Immediate insertion of breast prosthesis following 19340 IV Case-control study; or any of the above diagnostic mastopexy, mastectomy, or in reconstruction studies in the absence of a universally accepted Breast reconstruction, immediate or delayed, 19357 gold standard with tissue expander, including subsequent V Expert opinion developed via consensus pro- expansion cess; case report or clinical example; cadaver Microsurgical techniques, requiring use of 69990 study; or evidence based on physiology, bench operating microscope (list separately in research, or “first principles” addition to code for primary procedure) CPT, Current Procedural Terminology. Evidence Rating Scale for Prognostic/Risk Studies Healthcare Common Procedure Coding System Level of Evidence Qualifying Studies (HCPCS) Codes* I High-quality, multi-centered or single-centered, Procedure Code prospective cohort or comparative study with Breast reconstruction with deep inferior epigastric S2068 adequate power; or a systematic review of these perforator (DIEP) flap or superficial inferior studies epigastric artery (SIEA) flap, including harvesting II Lesser-quality prospective cohort or comparative of the flap, microvascular transfer, closure of study; retrospective cohort or comparative study; donor site, and shaping the flap into a breast, untreated controls from a randomized controlled unilateral trial; or a systematic review of these studies *Please check payer’s policies. III Case-control study; or systematic review of these studies IV Case series with pretest/posttest; or only posttest V Expert opinion developed via consensus process; case report or clinical example; cadaver study; or evidence based on physiology, bench research, or “first principles”

REFERENCES 14. Ascherman JA, Seruya M, Bartsich SA. Abdominal wall 1. American Cancer Society. What are the key statistics about morbidity following unilateral and bilateral breast breast cancer? Available at: http://www.cancer.org/cancer/ reconstruction with pedicled TRAM flaps: An outcomes breastcancer/detailedguide/breast-cancer-key-statistics. analysis of 117 consecutive patients. Plast Reconstr Surg. Accessed June 15, 2016. 2008;121:1–8. 2. American Society of Plastic Surgeons. 2015 plastic surgery 15. Bharti G, Groves L, Sanger C, Thompson J, David L, Marks statistics report. Available at: https://d2wirczt3b6wjm.cloud- M. Minimizing donor-site morbidity following bilateral pedi- front.net/News/Statistics/2015/plastic-surgery-statistics- cled TRAM breast reconstruction with the double mesh fold full-report-2015.pdf. Accessed June 15, 2016. over technique. Ann Plast Surg. 2013;70:484–487. 3. American Society of Plastic Surgeons. TOPS: Tracking 16. Chun YS, Sinha I, Turko A, Lipsitz S, Pribaz JJ. Outcomes Operations and Outcomes for Plastic Surgeons. Available at: and patient satisfaction following breast reconstruction with https://www.plasticsurgery.org/for-medical-professionals/ bilateral pedicled TRAM flaps in 105 consecutive patients. quality-and-registries/tracking-operations-and-outcomes-for- Plast Reconstr Surg. 2010;125:1–9. plastic-surgeons-(tops). Accessed June 12, 2016. 17. Ireton JE, Kluft JA, Ascherman JA. Unilateral and bilateral 4. American Society of Plastic Surgeons. Evidence-Based Clinical breast reconstruction with pedicled TRAM flaps: An out- Practice Guideline: Breast reconstruction with expanders comes analysis of 188 consecutive patients. Plast Reconstr Surg and implants. Available at: http://www.plasticsurgery.org/ Glob Open 2013;1:1–7. Documents/Health-Policy/Guidelines/guideline-2013-breast- 18. Kim EK, Eom JS, Ahn SH, Son BH, Lee TJ. Evolution of the recon-expanders-implants.pdf. Accessed June 15, 2016. pedicled TRAM flap: A prospective study of 500 consecutive 5. Chun YS, Sinha I, Turko A, et al. Comparison of morbid- cases by a single surgeon in Asian patients. Ann Plast Surg. ity, functional outcome, and satisfaction following bilateral 2009;63:378–382. TRAM versus bilateral DIEP flap breast reconstruction.Plast 19. Petit JY, Rietjens M, Garusi C, et al. Abdominal complica- Reconstr Surg. 2010;126:1133–1141. tions and sequelae after breast reconstruction with pedicled 6. Garvey PB, Buchel EW, Pockaj BA, et al. DIEP and pedicled TRAM flap: Is there still an indication for pedicled TRAM in TRAM flaps: A comparison of outcomes.Plast Reconstr Surg. the year 2003? Plast Reconstr Surg. 2003;112:1063–1065. 2006;117:1711–1719; discussion 1720–1721. 20. Kim EK, Lee TJ, Eom JS. Comparison of fat necrosis between 7. Drazan L, Vesely J, Hyza P, et al. Bilateral breast reconstruc- zone II and zone III in pedicled transverse rectus abdominis tion with DIEP flaps: 4 years’ experience.J Plast Reconstr musculocutaneous flaps: A prospective study of 400 consecu- Aesthet Surg. 2008;61:1309–1315. tive cases. Ann Plast Surg. 2007;59:256–259. 8. Gill PS, Hunt JP, Guerra AB, et al. A 10-year retrospective 21. Bozikov K, Arnez T, Hertl K, Arnez ZM. Fat necrosis in free review of 758 DIEP flaps for breast reconstruction.Plast DIEAP flaps: Incidence, risk, and predictor factors.Ann Plast Reconstr Surg. 2004;113:1153–1160. Surg. 2009;63:138–142. 9. Guerra AB, Metzinger SE, Bidros RS, et al. Bilateral breast 22. Damen TH, Mureau MA, Timman R, Rakhorst HA, Hofer reconstruction with the deep inferior epigastric perforator SO. The pleasing end result after DIEP flap breast recon- (DIEP) flap: An experience with 280 flaps.Ann Plast Surg. 2004;52:246–252. struction: A review of additional operations. J Plast Reconstr 10. Nahabedian MY, Momen B. Lower abdominal bulge after Aesthet Surg. 2009;62:71–76. deep inferior epigastric perforator flap (DIEP) breast recon- 23. Yueh JH, Slavin SA, Adesiyun T, et al. Patient satisfaction struction. Ann Plast Surg. 2005;54:124–129. in postmastectomy breast reconstruction: A comparative 11. Tran NV, Buchel EW, Convery PA. Microvascular complica- evaluation of DIEP, TRAM, latissimus flap, and implant tech- tions of DIEP flaps.Plast Reconstr Surg. 2007;119:1397–1405; niques. Plast Reconstr Surg. 2010;125:1585–1595. discussion 1406–1408. 24. Damen TH, Timman R, Kunst EH, et al. High satisfaction 12. Venkat R, Lee JC, Rad AN, Manahan MA, Rosson GD. rates in women after DIEP flap breast reconstruction.J Plast Bilateral autologous breast reconstruction with deep infe- Reconstr Aesthet Surg. 2010;63:93–100. rior epigastric artery perforator flaps: Review of a single sur- 25. American Society of Plastic Surgeons. Evidence-based prac- geon’s early experience. Microsurgery 2012;32:275–280. tices for thromboembolism prevention: A report from the 13. Yan XQ, Yang HY, Zhao YM, You L, Xu J. Deep inferior epi- https://www.plasticsurgery.org/documents/medical-pro- gastric perforator flap for breast reconstruction: Experience fessionals/health-policy/key-issues/ASPS_VTE_Report.pdf. with 43 flaps.Chin Med J (Engl.) 2007;120:380–384. Accessed June 11, 2016.