Comparing the donor-site morbidity using DIEP, SIEA or MS-TRAM flaps for breast reconstructive surgery: A meta-analysis

Transcription

1 Journal of Plastic, Reconstructive & Aesthetic Surgery (2012) 65, 1474e1480 REVIEW Comparing the donor-site morbidity using DIEP, SIEA or MS-TRAM flaps for breast reconstructive surgery: A meta-analysis Alexander Egeberg, Mads Kløvgaard Rasmussen, Jens Ahm Sørensen* Department of Plastic Surgery, University of Southern Denmark, Odense, Denmark Received 5 January 2012; accepted 2 July 2012 KEYWORDS DIEP; SIEA; MS-TRAM; Donor-site morbidity Summary Background: Countless studies have compared the use of autologous tissue for breast reconstruction; however, rates of donor-site morbidity differ greatly. This study examined the donor-site morbidity of superficial inferior epigastric artery (SIEA), deep inferior epigastric perforator (DIEP) and muscle-sparing transverse rectus abdominis myocutaneous (MS-TRAM) flaps when used for unilateral breast reconstruction. Methods: Searches in PubMed and Medline as well as three manual search strategies for English-language articles published from 1 January 1995 to 1 January 2011 resulted in 2154 publications. Four levels of screening identified five studies suitable for the meta-analysis. StatsDirect software was used to perform the ManteleHaenszel fixed-effect model. Results: Only one study reported rates of donor-site morbidity for SIEA flaps. It was therefore impossible to perform any analysis regarding SIEA flaps. Five studies reported rates for both DIEP and MS-TRAM flaps and were used to estimate pooled relative risk (RR) and confidence intervals (CIs) of bulging. There was a 20% reduced risk of bulging when DIEP flaps were used compared to MS-TRAM flaps (RR 0.80, 95% CI 0.48e1.35). Subgroup analysis demonstrated that the risk of bulging in DIEP flap patients was one-third of MS-TRAM flap patients (RR 0.29; 95% CI 0.06e1.36), when rates were reported by clinical examinations. However, when rates were reported by surveys there was no difference in bulge formation between DIEP and MS-TRAM flap patients (RR 1.04; 95% CI 0.59e1.79). The adjusted RR of hernia in DIEP flap patients was approximately one-half of MS-TRAM flap patients (RR 0.43; 95% CI 0.07e2.63). * Corresponding author. Department of Plastic Surgery, Odense University Hospital, Sdr. Bouldevard 29, 5000 Odense C, Denmark. Tel.: þ address: jens.ahm@dadlnet.dk (J. Ahm Sørensen) /$ - seefrontmatterª 2012 BritishAssociationofPlastic, ReconstructiveandAesthetic Surgeons. Publishedby Elsevier Ltd. Allrightsreserved.

2 Donor-site morbidity for breast reconstructive surgery 1475 Conclusion: This analysis demonstrated a clear trend towards a favourable outcome when DIEP flaps were used compared to MS-TRAM flaps. ª 2012 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved. Background In 1979, Holmström 1 was the first to describe the use of autologous tissue harvested from the lower abdomen in breast reconstruction. Three years later, in 1982, Hartrampf et al. 2 popularised the transverse abdominal island flap, later to be known as the transverse rectus abdominis myocutaneous (TRAM) flap. Abdominal tissue has proven to be a reliable source of soft tissue due to a rich supply of volume and good-quality tissue, while creating an inconspicuous scar. The TRAM flap has for many years been the gold standard in breast reconstruction, but the donor-site morbidity remains a major concern as the rectus muscle is harvested. Techniques have been developed to minimise the trauma to the fascia and muscle, while still providing adequate blood supply. The muscle-sparing (MS) TRAM flaps (MS-0, MS-1 and MS-2) were developed to minimise the resection of the rectus muscle. In 1989, the deep inferior epigastric perforator (DIEP) flap, also known as MS-3, was described by Koshima and Soeda 3 and subsequently popularised by both Allen 4 and Blondeel. 5 The superficial inferior epigastric artery (SIEA) flap has gained popularity as it completely spares the rectus muscle and fascia and thereby, at least theoretically, minimises the donor-site morbidity. The impact of different types of flaps on the abdominal wall has been a topic for much discussion. 6e14 The theoretical advantage of the DIEP and SIEA flap compared with the MS-TRAM flaps seems obvious. The MS-TRAM flap is still a widely used procedure as it is technically easier than a DIEP flap. The current literature often yields opposing conclusions regarding the impact of the flaps on the donor-site. 10e14 A meta-analysis is therefore performed in order to shed some light on the results and quality of studies that have been published so far. Method This study was performed following the guidelines outlined by the Meta-analysis of Observational Studies in Epidemiology (MOOSE) group. 15 The aim of the study was to compare the donor-site morbidities of SIEA, DIEP and MS-TRAM flaps when used for breast reconstruction. Several different methods have been used to investigate the impact of the SIEA, DIEP and MS- TRAM flaps on the donor site. One method is objective reporting of bulges and hernias, and another is surveys on the impact on daily-life activities and finally musclestrength measurement with a dynamometer. It is not possible to compare the three different methods without compromising the homogeneity of the studies, patients and methods. As a consequence, this study exclusively investigated donor-site morbidity defined as hernias or abdominal bulging. When bilateral reconstructions are performed, it is not uncommon to use two different types of flaps. 16e18 Knowing which of the two methods is at fault is difficult under these circumstances. It was therefore chosen to exclusively investigate the impact of flaps used in unilateral reconstructions. Only studies with a minimum of two cohorts of patients, comparing SIEA and DIEP flaps, SIEA and MS-TRAM flaps, DIEP and MS-TRAM flaps or all three types of flaps, were included in the analysis. Search strategy The following search terms were used in PubMed and Medline databases: DIEP, SIEA, TRAM, deep inferior epigastric perforator, transverse rectus abdominis musculocutaneous and superficial inferior epigastric artery. This search yielded 2154 hits. Three manual search strategies were used to try to retrieve additional studies. First, studies that were published after the cut-off date were identified via PubMed. Second, a search for related citations was performed as well as a specific search for studies by authors who are internationally acknowledged on the subject of perforator flaps. Publications from major plastic surgery journals, not initially identified by the database searches, were also reviewed. Third, plastic surgeons worldwide, with a history of publishing articles on the subject, were contacted via in order to obtain studies that have not been published yet. Unfortunately, these efforts did not result in any additional studies. The search was limited to include only studies on flaps performed on women, and only studies published between 1 January 1995 and 1 January Animal studies, abstracts only, literature reviews, single-case reports, letters, comments and publications in languages other than English, were excluded. This reduced the results to 685 studies, of which 434 were not directly related to SIEA, DIEP or MS-TRAM flaps. They were therefore excluded. Abstracts were retrieved for the remaining 226 studies. Of these, 176 studies were discarded because they did not include data on donor-site morbidity. Full-text articles were reviewed for the remaining 50 studies. Studies with less than 10 patients were excluded. When different publications were using the same cohort of patients, the study with the largest population of patients was chosen. Studies containing only one leg of interest (SIEA, DIEP or MS- TRAM flaps) were excluded, as well as studies focussing on TRAM flaps that were not muscle sparing. Studies that did not separate data from the different types of perforator flaps, combined data from unilateral and bilateral procedures or in other ways had incomparable data (i.e., raw data mashed with estimates) were excluded. A total of five

3 1476 A. Egeberg et al. studies 6e9,19 were identified as suitable for meta-analysis as demonstrated in Figure 1. When performing a meta-analysis, the optimal set-up is using randomised clinical trial studies. There was no such study found in the 226 studies retrieved for abstracts reading nor were there any blinded studies examining donor-site morbidity. When harvesting tissue from the abdomen, the decision on which flap to use is often decided during surgery, as too few or too small vessels can result in flap necrosis. 20 The requirements of perforator size as well as number of perforators and the intra-operative choice of flap type, make it very difficult to create a truly randomised and blinded study. Several studies report that the superficial inferior epigastric artery is absent or could not be located in 30e35% of patients. 21,22 This study therefore has to rely on non-randomised case series, which is a heavy bias within a meta-analysis. Data extraction and statistical analysis Information from each of the five studies was extracted in a standardised form: lead author, publication year, number Figure 1 Study attrition diagram.

4 Donor-site morbidity for breast reconstructive surgery 1477 of patients, mean age, number of responders to follow-up, body mass index (BMI), percentage of smokers, follow-up time, number and percentage of patients with hernia and number and percentage of patients with abdominal bulging. The extracted information can be viewed in Table 1. As this study only investigated unilateral reconstructions, the number of flaps performed equalled the number of patients. Rates for abdominal complications were calculated based on the number of responders, as one study used a survey to report bulging and hernia, with less than a 100% response rate. The ManteleHaenszel fixed-effects method was used to create pooled estimates for the five studies in this analysis. The fixed-effects method was chosen over the randomeffects, as the I 2 (inconsistency) value obtained here was very low (I 2 Z 0% with a 95% confidence interval (95% CI) Z 0e64.1%). A visual demonstration of the event rates of bulges is shown in the L Abbe plot (Figure 2). A funnel plot was created in order to assess for publication bias (Figure 3). In this analysis, findings of p < 0.05 were considered to be statistically significant. All calculations were performed using StatsDirect version (StatsDirect Ltd., Cheshire, UK). Results Only one study examining SIEA flaps was of a quality that matched the stated criteria. It was therefore not possible to perform any statistical analysis in order to compare the donor-site morbidity of SIEA with that of DIEP or MS-TRAM flaps. Pooled analysis of the five included studies showed a 20% increase in the risk of developing an abdominal bulge after MS-TRAM flap surgery compared with DIEP flaps (fixedeffects pooled relative risk, 0.80; 95% CI 0.48e1.35, p Z 0.40). Figure 2 The L Abbe plot shows the event rate in the DIEP flap (experimental) group against the event rate in the MS- TRAM flap (control) group, as an aid to exploring the variation of effect estimates within our meta-analysis that is due to heterogeneity rather than chance. The solid line represents the ratio where there is no difference in outcome between DIEP and MS-TRAM flap surgery, whereas the dotted line represents the estimated risk reduction of DIEP flaps compared to MS-TRAM flaps. Sizes of the circles represent study sizes. This figure was created using StatsDirect version (Stats- Direct Ltd., Cheshire, United Kingdom). A subgroup analysis separating complications reported by clinical examinations 7,9,19 and surveys 6,8 was performed. This revealed that studies reporting clinical examination findings showed the risk of abdominal bulges in patients with DIEP flaps to be approximately one-third of that for TRAM flap patients (fixed-effects pooled relative risk Z , 95% CI Z e , Table 1 Source Characteristics of study populations in the 5 main articles for analysis. No. of Mean age No. of Response BMI Smokers Follow-up patients in years responders rate in months Hernia (%) Blondeel, ,a MS-TRAM (range 29e63) % NA 50.0% Mean: (12e61) 1 (5) 2 (10) DIEP (range 31e66) % NA 44.4% Mean: (12e30) 0 (0) 0 (0) Futter, ,b MS-TRAM (SD 6.96) % 25.6 (SD 4.23) NA Mean: 41 (SD 12.78) 1 (3.7) 9 (33) DIEP (SD 7.92) % 23.6 (SD 2.34) NA Mean: 19 (SD 9.68) 0 (0) 7 (33) Nahabedian, ,a MS-TRAM % NA 12.3% Mean: 23 (3e49) NA 3 (4.6) DIEP % NA 10.6% Mean: 23 (3e49) NA 1 (1.5) Wu, ,b MS-TRAM (8.8) % 25.3 (4.9) 11% 1.5e56 NA 15 (34.1) DIEP (8.7) % 27.0 (4.6) 8% 1.5e56 NA 6 (40) SIEA (7.5) % 28.6 (3.9) 4% 1.5e56 NA 6 (35.3) Nelson, ,a MS-TRAM 59 NA % NA NA NA 1 (2) 2 (3) DIEP 35 NA % NA NA NA 0 (0) 0 (0) a Data reported by clinical examination. b Data reported by survey. Bulge (%)

5 1478 A. Egeberg et al. patients with DIEP flaps was approximately one-half that of MS-TRAM flap patients (fixed-effects pooled relative risk Z , 95% CI Z e , p Z 0.36) (Figure 4. Discussion Figure 3 Funnel plot of log relative risks according to their standard errors for bulges. Asymmetrically positioned circles denote small sample bias. The vertical line represents the pooled log relative risk for bulges, and the oblique lines represents the expected 95 percent confidence interval for a given standard error, assuming no between-study heterogeneity. This figure was created using StatsDirect version (StatsDirect Ltd., Cheshire, United Kingdom). p Z ). On the other hand, the studies using surveys found approximately the same risk of bulge formation between DIEP and TRAM flap patients (fixed-effects pooled relative risk Z , 95% CI Z e , p Z ). Subgroup analysis of articles reporting rates of hernia showed that after adjusting for population sizes, the adjusted relative risk of abdominal hernia formation in The lack of randomised clinical trial studies on the subject of donor-site morbidity using perforator flaps is a major problem. Due to ethical considerations and that the choice of flap is based on anatomic intra-operative findings, the creation of a randomised clinical trial study poses a great challenge that is not easy to overcome. Nelson et al. 9 describe the use of a single-surgeon, single-institution comparison as the next best option when being unable to perform a randomised clinical comparison. The studies revealed by our search strategy described different ways of examining donor-site morbidity. Some studies described the occurrence of hernia and bulges, while other studies used surveys to report on impact on everyday life. A whole different group of studies used dynamometers to measure muscle strength in the abdominal area. 6,12 The data obtained from studies examining muscle strength using dynamometers proved to be impossible to compare without severely damaging the quality of this meta-analysis. The angles and movements used for testing muscle strength differed greatly. The data published in the few studies of acceptable quality were simply too sparse to be compared. In this study, focus was on hernia and bulging as an indicator for the quality of interventions (SIEA, DIEP or MS- TRAM flap surgery) regarding donor-site morbidity. It is Figure 4 Abdominal bulge in DIEP and MS-TRAM flap patients. The effect estimate is marked with a solid black square. The size of the squares represents the weight that the studies exert in the meta-analysis (the ManteleHaenszel weight). The horizontal lines represent the confidence intervals. The unfilled diamond with the vertical dotted line represents the pooled estimates. This figure was created using StatsDirect version (StatsDirect Ltd., Cheshire, United Kingdom).

6 Donor-site morbidity for breast reconstructive surgery 1479 important to keep in mind that there are also other complications following free flap surgery. Blondeel et al. 19 reported that even though all patients in their study with bulges or hernia had aesthetic deformities, they were free of subjective complaints. This was furthermore underlined by the fact that the subjective reported findings from surveys in the subgroup analysis were almost identical between DIEP and MS-TRAM flaps regarding the donor-site complication rates. Therefore, the ability to function in an everyday environment following flap surgery should also be considered as a donor-site complication, as this may have a greater effect on the patients than an uneven contour of the abdomen. Unfortunately, this proved to be impossible to investigate, due to lack of relevant available data. A general problem in all of the included studies was the very small cohorts. The largest group contained 65 patients in the MS-TRAM and 66 patients in the DIEP group. Several studies 7,10,23 have reported an increased risk of donor-site complications in bilateral reconstructions compared to unilateral reconstructions. The studies performing bilateral procedures rarely disclosed whether the abdominal complication (e.g., a hernia) was located to the right, left or in the midline, 11,17 thereby making it impossible to determine which flap type was at fault, if different types of flaps were used. This issue was avoided by limiting the analysis to unilateral reconstruction, but resulted in a smaller number of studies available for analysis. Not all studies reported whether a mesh was used in the closure of the fascia. Rossetto et al. 25 reported a significant reduction (p < 0.025) in the incidence of abdominal wall hernia and bulge among patients where mesh was used. This was supported by Banic et al., 26 whereas Nahabedian et al. 24 found the use of mesh to be without any significance. Obesity is a known risk factor for donor-site complications. 10,27 Chang et al. 27 found obesity to be a risk factor for abdominal bulge and hernia. The use of an intra-operative algorithm on which flap to harvest not only eliminates the possibility of randomising and blinding patient groups, but can also affect which types of patients receive the different flap types. Wu et al. 8 states: patients with higher body mass indexes are more likely to have one or more large vessels perforating the rectus abdominis muscle and are therefore more likely, in our hands, to have a DIEP flap used for breast reconstruction. The donor-site complications found in these studies on patients with DIEP flap surgery could just as well be caused by the obesity as the surgery itself. The impact of DIEP flap surgery on the abdominal wall can therefore be overestimated. It is problematic when TRAM flaps are described with the term muscle-sparing. There is a major difference between MS-0, which uses the full width and partial length of the rectus muscle, and the MS-2, which preserves both the medial and lateral segments of the muscle as described by Lee et al. 28 It seems obvious that harvesting the full width of the rectus muscle poses a greater risk of complications than a flap using only an area equivalent to the size of a postage stamp. Nevertheless, several studies fail to report which degree of muscle-sparing surgery (MS-0, MS-1 or MS-2) they perform when harvesting the TRAM flaps. Damage to the intercostal nerves during DIEP and MS-TRAM harvest could compromise the integrity of the abdominal wall. There is a theoretical advantage to harvesting perforators from the medial branch of the deep inferior epigastric artery as opposed to the lateral branch, as the segmental intercostal nerves perform plexuses around the lateral branch perforators. However, a recent study by Garvey et al. 29 found no significant difference in donor-site morbidity between medial and lateral branch donor-site harvest. The same study did however find a significantly higher risk of abdominal bulge or hernia when the patient had a history of prior abdominal surgery. In the subgroup analysis, this study found the percentage of reported bulging to be much higher in studies using surveys than the studies performing objective reporting by a medical professional. Several dictionaries use the term bulge in the definition of a hernia so patients might be confused whether to report their findings as a hernia or a bulge. Patients might respond more positive in surveys because they are grateful for the breast reconstruction, or negative because they are unhappy with the aesthetic result. There is a risk of the surgeons reporting too few complications due to professional pride. However, in the subgroup analysis, the results are likely to be caused by random chance due to small sample sizes. The length of follow-up is an important factor when analysing the available data. With follow-up times ranging from 6 weeks to 4.7 years, the question arises whether a hernia is naturally occurring or caused by the harvest of the abdominal tissue. Blondeel et al. 19 reported the finding of three hernias in a non-operated control group. This emphasises the importance of including control groups in order to minimise bias when estimating the relative risk. In the pooled analysis, MS-TRAM flap surgery showed a 20% increased risk of abdominal bulging; however, this was without statistical significance. Even though the subgroup analysis clearly favoured the DIEP flap surgery over MS-TRAM flap, this too was without statistical significance. The trends demonstrated were clearly in favour of DIEP compared with MS-TRAM. This analysis used a 95% level of confidence; however, due to the lack of quality research on the subject, there is still a great level of uncertainty whether or not there is a difference. Using a slightly smaller level of confidence might in theory have given results that were statistically significant, in the same way that as little as only a few extra high-quality studies could have done. This shows the need for more high-quality studies. Conclusion This study demonstrated a clear trend towards lower donorsite morbidity when using DIEP flaps compared to MS-TRAM flaps even though no definite statistical significance could be established. More high-quality research is needed as numbers of perforator flaps increases, in order to demonstrate the relationship between donor-site morbidity and the different types of flaps. Randomised controlled trials on this subject will be very difficult to perform due to the aforementioned difficulties. Therefore, priority should be

7 1480 A. Egeberg et al. on high-quality prospective cohort studies with appropriate controls, improved definition of complications and use of sensitive and reliable patient -reported outcome measures. Conforming to the declaration of Helsinki This study was conducted in compliance with the recognised international standards and the principles of the Declaration of Helsinki. Conflict of interest None. The authors have no financial interest in any of the procedures, studies, methods or products mentioned in this article. No funding was received for the production of this study. References 1. Holmström H. The free abdominoplasty flap and its use in breast reconstruction. An experimental study and clinical case report. Scand J Plast Reconstr Surg 1979;13(3):423e7. 2. Hartrampf CR, Schlefan M, Black PW. Breast reconstruction with a transverse abdominal island flap. Plast Reconstr Surg 1982;69(2):216e Koshima I, Soeda S. Inferior epigastric artery skin flaps without rectus abdominis muscle. Br J Plast Surg 1989;42:645e8. 4. Allen RJ, Treece P. Deep inferior epigastric perforator flap for breast reconstruction. Ann Plast Surg 1994;32:32e8. 5. Blondeel PN, Boeckx WD. Refinements in free flap breast reconstruction: the free bilateral deep inferior epigastric perforator flap anastomosed to the internal mammary artery. Br J Plast Surg 1994;47:495e Futter CM, Webster MH, Hagen S, Mitchell SL. A retrospective comparison of abdominal muscle strength following breast reconstruction with a free TRAM or DIEP flap. Br J Plast Surg 2000;53(7):578e Nahabedian MY, Tsangaris T, Momen B. Breast reconstruction with the DIEP flap or the muscle-sparing (MS-2) free TRAM flap: is there a difference? Plast Reconstr Surg 2005 Feb;115(2): 436e Wu LC, Bajaj A, Chang DW, Chevray PM. Comparison of donorsite morbidity of SIEA, DIEP, and muscle-sparing TRAM flaps for breast reconstruction. Plast Reconstr Surg 2008;122(3):702e9. 9. Nelson JA, Guo Y, Sonnad SS, et al. A comparison between DIEP and muscle-sparing free TRAM flaps in breast reconstruction: a single surgeon s recent experience. Plast Reconstr Surg 2010; 126(5):1428e Vyas RM, Dickinson BP, Fastekjian JH, Watson JP, Dalio AL, Crisera CA. Risk factors for abdominal donor-site morbidity in free flap breast reconstruction. Plast Reconstr Surg 2008; 121(5):1519e 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 Aug;122(2):348e Bonde CT, Lund H, Fridberg M, Danneskiold-Samsoe B, Elberg JJ. Abdominal strength after breast reconstruction using a free abdominal flap. J Plast Reconstr Aesthet Surg 2007;60(5):519e Schaverien MV, Perks AG, McCulley SJ. Comparison of outcomes and donor-site morbidity in unilateral free TRAM versus DIEP flap breast reconstruction. J Plast Reconstr Aesthet Surg 2007;60(11):1219e Bajaj AK, Chevray PM, Chang DW. Comparison of donor-site complications and functional outcomes in free musclesparing TRAM flap and free DIEP flap breast reconstruction. Plast Reconstr Surg 2006 Mar;117(3):737e Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis of observational studies in epidemiology (MOOSE) group. JAMA 2000;283(15):2008e Ascherman JA, Seruya M, Bartsich SA. Abdominal wall morbidity following unilateral and bilateral breast reconstruction with pedicled TRAM flaps: an outcomes analysis of 117 consecutive patients. Plast Reconstr Surg 2008 Jan;121(1): 1e Vega SJ, Bossert RP, Serletti JM. Improving outcomes in bilateral breast reconstruction using autogenous tissue. Ann Plast Surg 2006 May;56(5):487e Chun YS, Sinha I, Turko A, et al. Comparison of morbidity, functional outcome, and satisfaction following bilateral TRAM versus bilateral DIEP flap breast reconstruction. Plast Reconstr Surg 2010 Oct;126(4):1133e Blondeel N, Vanderstraeten GG, Monstrey S, et al. The donor site morbidity of free DIEP flaps and free TRAM flaps for breast reconstruction. Br J Plast Surg 1997;50(5):322e Rozen WM, Whitaker IS, Chubb D, Ashton MW. Perforator number predicts fat necrosis in a prospective analysis of breast reconstruction with free TRAM, DIEP, and SIEA flaps. Plast Reconstr Surg 2010;126(6):2286e Wolfram D, Schoeller T, Hussl H, Wechselberger G. The superficial inferior epigastric artery (SIEA) flap. Indications for breast reconstruction. Ann Plast Surg 2006;57(6):593e Lindsey JT. Integrating the DIEP and muscle-sparing (MS-2) free TRAM techniques optimizes surgical outcomes: presentation of an algorithm for microsurgical breast reconstruction based on perforator anatomy. Plast Reconstr Surg 2007;119(1):18e Nahabedian MY, Dooley W, Singh N, Manson PN. Contour abnormalities of the abdomen after breast reconstruction with abdominal flaps: the role of muscle preservation. Plast Reconstr Surg 2002;109(1):91e Nahabedian MY, Manson PN. Contour abnormalities of the abdomen after transverse rectus abdominis muscle flap breast reconstruction: a multifactorial analysis. Plast Reconstr Surg 2002 Jan;109(1):81e Rossetto LA, Abla LE, Vidal R, et al. Factors associated with hernia and bulge formation at the donor site of the pedicled TRAM flap. Eur J Plast Surg 2010 Aug;33(4):203e Banic A, Boeckx W, Greulich M, et al. Late results of breast reconstruction with free TRAM flaps: a prospective multicentric study. Plast Reconstr Surg 1995 Jun;95(7):1195e Chang DW, Wang B, Robb GL, et al. Effect of obesity on flap and donor-site complications in free transverse rectus abdominis myocutaneous flap breast reconstruction. Plast Reconstr Surg 2000;105(5):1640e Lee BT, Chen C, Nguyen MD, Lin SJ, Tobias AM. A new classification system for muscle and nerve preservation in DIEP flap breast reconstruction. Microsurgery 2010;30(2):85e Garvey PB, Salavati S, Feng L, Butler CE. Abdominal donor-site outcomes for medial versus lateral deep inferior epigastric artery branch perforator harvest. Plast Reconstr Surg 2011 Jun; 127(6):2198e205.