breast - prma plastic surgery

BREAST

Abdominal Wall Stability and FlapComplications after Deep Inferior EpigastricPerforator Flap Breast Reconstruction: DoesBody Mass Index Make a Difference? Analysisof 418 Patients and 639 Flaps

Oscar Ochoa, M.D.Minas Chrysopoulo, M.D.

Chet Nastala, M.D.Peter Ledoux, M.D.Steven Pisano, M.D.

San Antonio, Texas

Background: Promoted by reports of decreased donor-site morbidity, deepinferior epigastric perforator (DIEP) flaps have gained significant popularity.Increasing body mass index is associated with poor outcomes in breast recon-struction using traditional techniques. The authors aimed to define complica-tions with increasing body mass index among patients undergoing DIEP flapbreast reconstruction.Methods: A retrospective analysis of 639 DIEP flaps in 418 patients was per-formed. Patients were stratified into five groups based on body mass index. Dataregarding medical comorbidities, adjuvant therapies, timing of reconstruction,active tobacco use, and surgical history were collected. Primary outcomes werecompared among groups.Results: The average body mass index for the entire population was 28.3 (range,17 to 42). Increasing body mass index was associated with increased incidenceof hypertension, previous abdominal operations, and length of follow-up. Flapcomplications stratified by group demonstrated significantly increased delayedwound healing complications in severely obese patients compared with lowerbody mass index groups. Donor-site complications stratified by body mass indexdemonstrated significantly increased delayed wound healing and overall com-plications among morbidly obese patients compared with other groups. Inci-dence of abdominal wall bulging and hernia formation was not significantlydifferent among groups.Conclusions: Increasing body mass index predisposes patients to delayedwound healing complications in both flap and donor-site locations. Neverthe-less, overall flap complications remain similar across all body mass index groups.Abdominal wall stability was maintained. Given a similar flap complicationprofile and maintenance of abdominal stability, DIEP flaps are recommended inpatients with increased body mass index. (Plast. Reconstr. Surg. 130: 21e, 2012.)CLINICAL QUESTION/LEVEL OF EVDENCE: Risk, II.

Autologous breast reconstruction has evolvedconsiderably throughout the years. Sincethe original description of pedicled trans-

verse rectus abdominis musculocutaneous (TRAM)flaps,1 abdomen-based breast reconstruction hasbeen able to achieve reproducible and increas-

ingly reliable aesthetic results. However, with sac-rifice of the underlying rectus muscle and fascia,abdominal wall laxity and hernia formation werequickly identified as limitations of pedicled TRAMflaps.2,3 Consequently, because of significant do-nor-site morbidity, pedicled TRAM flaps are con-sidered relatively contraindicated in high-risk pop-ulations, such as patients with obesity.4–6

From the Plastic Reconstructive and Microsurgical Associatesof South Texas, P.A.Received for publication July 28, 2011; accepted January 11,2012.Copyright ©2012 by the American Society of Plastic Surgeons

DOI: 10.1097/PRS.0b013e3182547d09

Disclosure: The authors have no financial interestto declare in relation to the content of this article.

www.PRSJournal.com 21e

In an attempt to decrease donor-site morbid-ity, free TRAM flaps were developed,7 enablinglimited fascial and rectus muscle resection. Previ-ous studies8–10 comparing pedicled versus freeTRAM flaps have verified improved abdominalwall stability with limited muscle and fascial re-section. Free TRAM flaps have nonetheless beenunable to overcome the inherent predispositionfor abdominal wall instability associated withobesity.11

With elimination of fascial and rectus muscleharvest, deep inferior epigastric perforator (DIEP)flaps promise improved abdominal wall stability.Previous studies12–16 have documented improvedabdominal wall stability among patients followingDIEP flap breast reconstruction compared withfree TRAM flaps. In contrast, because of the del-icate dissection required for DIEP flap elevation,concerns have been raised regarding DIEP flapreliability.9,13,17 Furthermore, previous studies11,18

have identified obesity as an important factor de-creasing flap reliability following free and pedi-cled TRAM flap surgery. Ultimately, the effects ofobesity on flap and donor-site morbidity followingDIEP flap breast reconstruction remain ill-de-fined. The purpose of the study is to determine theincidence of flap and donor-site morbidity follow-ing DIEP flap breast reconstruction as a functionof body mass index.

PATIENTS AND METHODSA retrospective chart review was conducted

after institutional review board approval amongconsecutive patients who underwent DIEP flapbreast reconstruction from January of 2006 toMarch of 2008. Patients were routinely offeredDIEP flaps based on standard selection criteriaused for free TRAM flaps regardless of body massindex.

Six hundred thirty-nine DIEP flaps were per-formed on 418 patients by a single group practice(Plastic Reconstructive and Microsurgical Associ-ates of South Texas). Data regarding patientdemographics, medical comorbidities, active to-bacco use, oncologic history, and adjuvant ther-apy were collected preoperatively. Active to-bacco use was defined as use of tobacco productswithin 6 weeks of reconstruction. The number ofprevious abdominal operations, open and/orlaparoscopic, was recorded.

Preoperative body mass index was used tostratify patients into five groups: normal weight(�24.9), overweight (25 to 29.9), obese (30 to34.9), severely obese (35 to 39.9), and morbidlyobese (�40). DIEP flap reconstruction was per-

formed by two board-certified plastic surgeonswith extensive microsurgical experience workingsimultaneously. Breast mounds were reconstructedroutinely with abdominal flap Hartrampf zones 1and 3. Arterial and venous anastomoses werehand-sewn to internal mammary vessels using stan-dard microsurgical technique. Intraoperative vari-ables including number of perforators used, isch-emia time, and overall reconstruction time wererecorded. Fascial closure was performed primarilyin two layers. Figure-of-eight interrupted suturesare placed followed by a simple running sutureusing 0 polypropylene for both layers. Mesh wasnot used prophylactically to reinforce the fascialclosure regardless of preoperative body mass in-dex. Length of hospital stay and time of follow-up(in days) were recorded.

Primary outcomes were defined as postoper-ative donor-site morbidity and flap complications.For both flap and donor-site soft-tissue infections,further stratification into mild, moderate, andsevere infections was performed to better definethe degree of infectious complications. Mild in-fections were defined as clinically significant er-ythema, warmth, and induration requiring out-patient antibiotics. Moderate infections weredefined as infections requiring debridement andwound packing in an ambulatory setting or use ofparenteral antibiotics. Severe infections were de-fined as those requiring formal operative debride-ment. Similarly, for both flap and donor sites,delayed wound healing was further stratified intomild, moderate, and severe. Mild delayed woundhealing was defined as minor wound dehiscencerequiring local therapies, including wound pack-ing. Moderate delayed wound healing was definedas wound dehiscence requiring vacuum-assistedclosure therapy. Severe delayed wound healingcomplications were defined as extensive wounddehiscence requiring operative intervention. Re-corded variables and incidence of primary out-comes were compared between body mass indexgroups and analyzed for statistical differences.

Associations involving categorical variableswere assessed using Pearson’s chi-square test orFisher’s exact test, as appropriate. The Kruskal-Wallis test was used to assess associations involvingcontinuously distributed variables. Patients havingbilateral reconstruction could contribute to morethan one complication outcome category if pres-ent. Therefore, associations involving flap com-plications were assessed using a logistic regressionmodel with binary response that accounted forcorrelation introduced by bilateral reconstruc-tions. Logistic regression with forward selection

Plastic and Reconstructive Surgery • July 2012

22e

was used to assess associations between potentiallyimportant covariates and the most prevalent com-plication outcomes observed. All statistical com-parisons were performed using a significance levelof 5 percent and SAS 9.2 (SAS Institute, Inc., Cary,N.C.).

RESULTSThe mean age of the patients in the study

population was 50.4 years (range, 27 to 74 years).Mean body mass index was 28.3 (range, 17 to 42).Hypertension (25.8 percent) and cardiac disease(6.6 percent) were the most common medical co-morbidities overall. With the exception of mi-graine headaches (17.7 percent), the incidence ofother reported medical comorbidities was approx-imately 5 percent or less. Recent use of tobaccoproducts was present in 8.1 percent (Table 1). Ahistory of at least one previous abdominal opera-tion (laparoscopic or open) was present in 68.4percent, with a mean prevalence overall of 1.3operations per patient.

Unilateral or bilateral reconstructions wereperformed in nearly equal numbers—197 (47.1percent) versus 221 (52.9 percent), respectively—with most (77.1 percent) performed in the imme-diate setting. Neoadjuvant irradiation and chemo-therapy were administered in 28.4 percent and41.6 percent of patients, respectively, with only 5.1percent of patients requiring radiotherapy post-operatively. One-third of patients received post-operative chemotherapy, and recurrence of breastcancer was identified in five patients (1.2 percent)(Table 2).

The mean time required for unilateral recon-structions was 253.3 minutes (4.2 hours), whereasbilateral reconstructions required 413.8 minutes(6.9 hours). Mean flap ischemia time for recon-structed right and left breasts was 23.6 and 22.7minutes, respectively. The mean number of per-forators used for right and left breast flaps were 2and 1.9, respectively (Table 3).

Overall, DIEP flap soft-tissue infections wereseen in 33 cases (5.4 percent), with 16 (2.6 per-cent) classified as mild and 15 (2.5 percent) clas-sified as moderate. Severe soft-tissue infectionswere seen in only two cases (0.3 percent). Delayedflap wound healing was seen in 38 cases (6.2 per-cent), with 36 (5.9 percent) classified as mild. Fatnecrosis was clinically evident in 63 cases (10.4percent), whereas seromas were seen in only fivecases (0.8 percent). Acute and subacute hemato-mas were seen in 12 (1.9 percent) and 13 cases (2.1percent), respectively. Acute vessel thrombosis wasrecognized and perfusion reestablished in fourcases (0.6 percent), with complete flap failure insix cases (1.0 percent). Overall, 152 flap compli-cations (23.8 percent) were reported (Table 4).

Donor-site soft-tissue infections were docu-mented in 30 patients (7.2 percent) overall, with19 (4.5 percent) of these classified as mild andnine (2.2 percent) classified as moderate. Delayeddonor-site wound healing was seen in 70 patients(16.7 percent), with 59 (14.1 percent) classified asmild. Seroma formation was reported in 27 cases(6.5 percent). No hematomas, acute or subacute,were encountered at donor sites postoperatively.During an average follow-up period of 1088 days(range, 720 to 1528 days), abdominal bulging wasseen in one patient (0.2 percent), and six patients

Table 1. Prevalence of Medical Comorbidities in theStudy Population

Variable No. (%)

Hypertension 108 (25.8)Diabetes mellitus 21 (5)Cardiac disease 28 (6.7)Pulmonary disease 19 (4.5)Peripheral vascular disease 1 (0.2)Autoimmune disease 23 (5.5)Migraine 74 (17.7)Coagulopathy 11 (2.6)Tobacco 34 (8.1)

Table 2. Laterality, Timing, Adjuvant Therapy, andRecurrence in the Study Population

Variable No. (%)

Unilateral 197 (47.1)Bilateral 221 (52.9)Immediate 77.5%Delayed 22.5%Radiotherapy

Preoperatively 118 (28.4)Postoperatively 21 (5.1)

ChemotherapyPreoperatively 172 (41.6)Postoperatively 141 (34.1)

Cancer recurrence 5 (1.2)

Table 3. Reconstruction Time, Ischemia Time, andNumber of Perforators in the Study Population

Variable Mean

Reconstruction, minUnilateral 253.3Bilateral 413.8

Flap ischemia, minRight 23.6Left 22.7

No. of perforatorsRight 2Left 1.9

Volume 130, Number 1 • DIEP Flap Complications

23e

(1.4 percent) developed abdominal wall hernias.Overall, 108 donor-site complications (25.8 per-cent) were reported (Table 5).

Stratified by body mass index, five groups ofpatients were identified (Table 6). When com-pared across all body mass index groups, the prev-alence of hypertension was significantly (p �0.001) different between body mass index groups.Groups with a higher body mass index had a sig-nificantly higher prevalence of hypertension thanlower body mass index groups (Fig. 1). Althougha trend was observed with a higher prevalence ofdiabetes in higher body mass index groups, nosignificant differences were observed betweenbody mass index groups regarding other recordedmedical comorbidities or recent tobacco use (Ta-ble 7). The number of total previous abdominaloperations was significantly (p � 0.03) lower innormal weight patients compared with other body

mass index groups. Morbidly obese patients re-ported significantly (p � 0.04) more previousopen abdominal operations compared with nor-mal and overweight patients (Fig. 2). Length offollow-up was significantly (p � 0.03) differentbetween body mass index groups, with normalweight (1137 days) and morbidly obese (1250days) patients having the longest mean follow-upperiod (Fig. 3). No significant differences wereobserved between body mass index groups regard-ing age, laterality or timing of reconstruction, his-tory of adjuvant therapy, or cancer recurrence(Tables 8 and 9).

Mean operative times were significantly (p �0.03) different between body mass index groupsundergoing unilateral reconstructions, with theseverely obese group (286 minutes) requiring sig-nificantly (p � 0.05) longer operative times com-pared with normal weight (240 minutes) andobese (252 minutes) groups (Fig. 4). Mean oper-ative times were similar among all body mass indexgroups undergoing bilateral reconstructions.Mean ischemia times for left breast reconstruc-tions were similar between all body mass indexgroups. Mean ischemia time for right breast re-constructions were significantly (p � 0.02) in-creased among severely obese patients (27.1 min-utes) compared with lower body mass indexgroups (Fig. 5). The number of perforators usedduring flap elevation was similar between all bodymass index groups for both right and left recon-structions (Table 10).

Flap complication analysis demonstrated sig-nificantly (p � 0.002) different rates of delayedwound healing between body mass index groups.Severely obese patients had a significantly (p �0.05) higher rate (20.6 percent) of delayed woundhealing compared with lower body mass indexgroups (Fig. 6). The incidence of overall andother individual flap complications, includingflap failure, were not significantly different be-tween groups despite increasing body mass in-dex (Table 11).

Donor-site complication analysis demonstratedsignificantly different rates of delayed wound heal-

Table 4. Overall Incidence of Flap Complications

Variable No. (%)

Flap infectionTotal 33 (5.4)Mild 16 (2.6)Moderate 15 (2.5)Severe 2 (0.3)

Delayed wound healingTotal 38 (6.2)Mild 36 (5.9)Moderate 1 (0.2)Severe 1 (0.2)

Perfusion complicationsThrombosis 4 (0.6)Failure 6 (1.0)

Acute hematoma 12 (1.9)Fat necrosis 63 (10.4)Subacute hematoma 13 (2.1)Seroma 5 (0.8)Total 152 (23.8)

Table 5. Overall Incidence of Donor-SiteComplications

Variable No. (%)

Donor-site infectionTotal 30 (7.2)Mild 19 (4.5)Moderate 9 (2.2)Severe 2 (0.5)

Delayed wound healingTotal 70 (16.7)Mild 59 (14.1)Moderate 8 (1.9)Severe 3 (0.7)

Acute hematoma 0 (0)Seroma 27 (6.5)Subacute hematoma 0 (0)Abdominal bulge 1 (0.2)Hernia 6 (1.4)Total 108 (25.8)

Table 6. Study Population Stratified by Body MassIndex

Variable BMI No. (%)

Normal �24.9 100 (23.9)Overweight 25–29.9 153 (36.6)Obese 30–34.9 113 (27)Severely obese 35–39.9 45 (10.8)Morbidly obese �40 7 (1.7)BMI, body mass index.


24e

ing (p � 0.05) and overall complications (p �0.02) between body mass index groups. The inci-dences of donor-site infection, acute or subacutehematomas, seroma formation, abdominal wallbulging, and hernia formation were not signifi-cantly different between groups despite increasingbody mass index (Table 12). Donor-site delayedwound healing (p � 0.03) and overall donor-sitecomplications (p � 0.01) were significantly in-creased in morbidly obese patients compared withlower body mass index groups (Fig. 7). When se-verity of delayed wound healing was comparedacross body mass index groups, morbidly obesepatients experienced significantly (p � 0.049) in-creased moderate wound healing complications(14.3 percent) compared with lower body massindex groups.

In a multivariate logistic regression model, theincidence of complications at the donor-siteand/or flap locations were examined against pos-sible independent risk factors. Increased bodymass index and active tobacco use were signifi-cantly (p � 0.006) associated with delayed woundhealing complications in both flap and donor-site locations. Increased body mass index and

shorter reconstructive times were noted to besignificantly (p � 0.002) associated with flapcomplications in general, or “any” flap compli-cations (Table 13).

DISCUSSIONIn the largest reported series since that by Gill

et al.,19 the current study analyzes flap and donor-site morbidity following 639 DIEP flaps on 418patients. The study population characteristicsclosely reflect a growing trend of obesity in West-ern society20 and daily clinical practice. In partic-ular, approximately 40 percent of study patientshad a body mass index of 30 or greater. The ap-plicability of the current study findings are furtherunderscored by the high incidence (68 percent)of previous abdominal surgery reported by pa-tients. After stratification by body mass indexgroups, considerable homogeneity was observedin the study population regarding age, medicalcomorbidities, recent tobacco use, and adjuvanttherapy, underscoring the independent effects ofbody mass index on flap and donor-site morbidity.

Fig. 1. Prevalence of hypertension by body mass index (BMI) group.

Table 7. Prevalence of Medical Comorbidities and Tobacco Use by Body Mass Index Group

Variable Normal (%) Overweight (%) Obese (%)Severely

Obese (%)Morbidly

Obese (%) p

ComorbiditiesHypertension 15 22.2 27.4 51.1 71.4 �0.001Diabetes mellitus 3 3.9 4.4 13.3 14.3 0.06Cardiac disease 5 5.9 8.8 8.9 0 0.73Pulmonary disease 5 2 7.1 6.7 0 0.23Peripheral vascular disease 0 0 0.9 0 0 0.63Arterial insufficiency 8 4.6 5.3 4.4 0 0.81Migraine 18 15 23 13.3 14.3 0.48Coagulopathy 2 3.3 2.7 2.2 0 0.97

Tobacco 12 8.5 7.1 2.2 0 0.35


25e

Flap ComplicationsThe current data support the safety of DIEP

flaps in breast reconstruction, even in obese pop-ulations (Fig. 8). Overall flap complications rates

(23.8 percent) coincide with previous studies offree TRAM flap11 or DIEP flap–based19,21,22 recon-struction in more slender populations. Moreover,greater than 75 percent of procedures were per-

Fig. 2. Mean abdominal operations by body mass index. Patients in the normalbody mass index group reported significantly (p � 0.03) lower number of op-erations overall compared with all other body mass index groups. Morbidlyobese patients, on average, reported a significantly (p � 0.04) higher number ofopen abdominal operations compared with normal and overweight body massindex groups. BMI, body mass index; Abd, abdomen.

Fig. 3. Length of follow-up (in days) by body mass index (BMI). Normal (p �

0.03) and morbidly obese (p � 0.05) body mass index groups had significantlylonger follow-up than overweight and severely obese groups.

Table 8. Mean Age, Previous Abdominal Surgery, and Follow-Up by Body Mass Index Group

Variable Normal Overweight Obese Severely Obese Morbidly Obese p

Age, yr 51.2 50 49.9 51.2 52.7 0.672Abdominal surgery, no. 1 1.4 1.5 1.4 1.7 0.017

Laparoscopic 0.3 0.5 0.5 0.6 0.1 0.055Open 0.6 0.9 1 0.8 1.6 0.048

Follow-up, days 1137.6 1066.6 1086.3 1037.4 1250.4 0.028


26e

formed in the immediate setting, which has pre-viously been associated with a higher rate ofcomplications.23 After stratification by body massindex, overall flap complications were highest

(35.3 percent) in severely obese patients but werenot statistically different compared with all otherbody mass index groups. Previous studies analyz-ing the effect of obesity on TRAM flap morbidity

Table 9. Reconstruction Timing, Laterality, Adjuvant Therapy, and Cancer Recurrence by Body Mass Index

Normal (%) Overweight (%) Obese (%) Severely Obese (%) Morbidly Obese (%) p

Reconstruction 0.68Immediate 110 (78) 184 (77) 146 (80.7) 48 (70.6) 7 (70)Delayed 31 (22) 55 (23) 35 (19.3) 20 (29.4) 3 (30)

Laterality 0.06Unilateral 59 (59) 67 (43.8) 45 (39.8) 22 (48.9) 4 (57.1)Bilateral 41 (41) 86 (56.2) 68 (60.2) 23 (51.1) 3 (42.9)

XRTPreoperatively 29 (29.3) 42 (27.6) 32 (28.6) 13 (28.9) 2 (28.6) 1Postoperatively 5 (5.1) 9 (5.9) 4 (3.6) 2 (4.4) 1 (14.3) 0.59

ChemotherapyPreoperatively 44 (44.9) 65 (43) 47 (42) 14 (31.1) 2 (28.6) 0.56Postoperatively 32 (32.7) 55 (36.4) 40 (35.7) 10 (22.2) 4 (57.1) 0.27

Cancer recurrence 3 (3) 1 (0.7) 1 (0.9) 0 (0) 0 (0) 0.46XRT, external radiation therapy.

Fig. 4. Time (in minutes) required for unilateral DIEP breast reconstructions bybody mass index (BMI). Severely obese patients required significantly (p � 0.05)longer time for completion of reconstruction compared with all other bodymass index groups.

Fig. 5. Flap ischemia time (in minutes) for right breast reconstructions. Se-verely obese patients experienced significantly (p � 0.02) longer flap ischemiatime compared with other body mass index (BMI) groups.


27e

reported flap complication rates between 39 and56 percent11,18 in obese populations. Most re-cently, a 14.6 percent flap complication rate hasbeen reported24 among patients with a body mass

index greater than or equal to 30 after TRAM andDIEP flap reconstruction.

The flap soft-tissue infection rate was 5.4 per-cent overall, coinciding with historical rates be-

Table 10. Mean Reconstruction Time, Ischemia Time, Perforator Number by Body Mass Index

Variable Normal Overweight Obese Severely Obese Morbidly Obese p

Reconstruction, minUnilateral 239.9 257.3 251.8 286.4 244.3 0.028Bilateral 411.9 408 425.9 390.6 505 0.095

Flap ischemia, minRight 22.7 22.9 23.8 27.1 23.8 0.004Left 22.7 22.1 23 24.3 24 0.416

No. of perforatorsRight 1.9 2 2.1 1.9 1.4 0.237Left 1.9 1.9 2 1.8 1.8 0.357

Fig. 6. Incidence of flap delayed wound healing. Severely obese patients hada significantly (p � 0.05) greater incidence of delayed wound healing comparedwith lower body mass index (BMI) groups.

Table 11. Incidence of Flap Complications by Body Mass Index

Normal Overweight Obese Severely Obese Morbidly Obese p

No. of patients 141 239 180 68 10Flap infection

Overall 4 (2.9) 12 (5.3) 13 (7.6) 4 (5.9) 0 (0) 0.41Mild 3 (2.2) 6 (2.7) 6 (3.5) 1 (1.5) 0 (0) 0.74Moderate 1 (0.7) 5 (2.2) 6 (3.5) 3 (4.4) 0 (0) 0.17Severe 0 (0) 1 (0.4) 1 (0.6) 0 (0) 0 (0) NA

Flap DWHOverall 2 (1.4) 9 (4) 12 (7.1) 14 (20.6) 1 (10) 0.0014Mild 2 (1.4) 8 (3.6) 12 (7.1) 13 (19.1) 1 (10) 0.003Moderate 0 (0) 1 (0.4) 0 (0) 0 (0) 0 (0) NASevere 0 (0) 0 (0) 0 (0) 1 (1.5) 0 (0) NA

Perfusion complications 0.85Thrombosis 2 (1.4) 1 (0.4) 1 (0.6) 0 (0) 0 (0)Failure 0 (0) 3 (1.3) 3 (1.7) 0 (0) 0 (0)

Acute hematoma 5 (3.6) 3 (1.3) 4 (2.3) 0 (0) 0 (0) 0.39Fat necrosis 17 (12.4) 22 (9.9) 17 (10) 7 (10.3) 0 (0) 0.95Subacute hematoma 5 (3.6) 6 (2.7) 2 (1.2) 0 (0) 0 (0) 0.42Seroma 1 (0.7) 2 (0.9) 1 (0.6) 1 (1.5) 0 (0) 0.42Total 33 (23.4) 49 (20.5) 45 (24.9) 24 (35.3) 1 (10) 0.23N/A, not applicable; DWH, delayed wound healing.


28e

tween 2.4 and 12.5 percent.25,26 Half of the re-ported infections were treated effectively with oralantibiotics, and only two required operative inter-vention. Unlike what has been reported for pedi-cled TRAM flaps,18 the incidence of flap infection

was not increased with increasing body mass indexin the current study, resembling prior findingsfollowing free TRAM flap surgery.11 As in previousstudies,19,22,23,25 acute and subacute hematomaswere uncommon (1.9 and 2.1 percent, respec-tively) and were unaffected by body mass index.

Delayed wound healing in the current studywas attributable primarily to mastectomy skin flapnecrosis, with no reported cases of partial DIEPflap loss. Thirty-eight flaps (6.2 percent) were af-fected by delayed wound healing at the interfacewith the recipient site, with 36 (5.9 percent) re-quiring minor debridement of native breast skin.As in previous reports following pedicled,18 freeTRAM flaps11 and, more recently, DIEP flaps,24

obesity was a significant factor associated with de-layed wound healing at the recipient site. More

Table 12. Incidence of Donor-Site Complications by Body Mass Index

Normal (%) Overweight (%) Obese (%) Severely Obese (%) Morbidly Obese (%) p

Donor-site infectionOverall 7 (7) 8 (6.5) 8 (7.1) 2 (4.4) 3 (42.9) 0.08Mild 3 (3) 6 (3.9) 6 (5.3) 2 (4.4) 2 (28.6) 0.121Moderate 4 (4) 2 (1.3) 2 (1.8) 0 (0) 1 (14.3) 0.132Severe 0 (0) 2 (1.3) 0 (0) 0 (0) 0 (0) 0.626

Delayed wound healingOverall 14 (14) 25 (16.3) 20 (17.7) 7 (15.6) 4 (57.1) 0.03Mild 12 (12) 20 (13.1) 17 (15) 7 (15.6) 3 (42.9) 0.281Moderate 2 (2) 5 (3.3) 0 (0) 0 (0) 1 (14.3) 0.049Severe 0 (0) 0 (0) 3 (2.7) 0 (0) 0 (0) 0.132

Acute hematoma 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 1Seroma 4 (4) 10 (6.5) 11 (9.7) 2 (4.4) 0 (0) 0.53Subacute hematoma 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 1Abdominal bulge 1 (1) 0 (0) 0 (0) 0 (0) 0 (0) 0.36Hernia 1 (1) 2 (1.3) 3 (2.7) 0 (0) 0 (0) 0.722Total 25 (25) 45 (29.4) 38 (33.6) 11 (24.4) 6 (85.7) 0.02

Fig. 7. Incidence of donor-site complications by body mass index (BMI). Mor-bidly obese patients experienced significantly greater delayed wound healing(p � 0.03) and overall (p � 0.01) complications compared with lower body massindex groups.

Table 13. Independent Risk Factors forComplications

Complication OR 95% CI p

Any flap complicationBMI 1.071 1.026–1.119 0.002Reconstruction time 0.996 0.993–0.998 �0.001

Any delayed woundhealing complication

BMI 1.121 1.068–1.176 �0.001Tobacco use 3.063 1.374–6.827 0.006

OR, odds ratio; CI, confidence interval; BMI, body mass index.


29e


30e

precisely, in this study, increased body mass indexwas associated with a higher incidence of mastec-tomy flap compromise. Severely obese patients ex-perienced nearly three times the rate of delayedwound healing compared with other body massindex groups. Nonetheless, the severity of woundhealing complications was not increased with in-creasing body mass index such that moderate andsevere wound healing complications were similaracross all groups.

Fat necrosis and perfusion complications suchas vessel thrombosis and total flap loss are strongindicators of flap reliability. Reported rates of fatnecrosis following DIEP flap reconstruction arevaried in the literature, ranging from 1.8 to 29percent.9,13,17,19,21,22,25–28 In the current study over-all, clinical fat necrosis was evident in 10.4 percentof DIEP flaps. Stratified by body mass index, ratesof fat necrosis were unaffected by body habitus, asreported previously for pedicled18 and free TRAMflaps.11

As in most successful series,17,19,22 vessel throm-bosis and flap loss remained an infrequent com-plication, with rates of 0.6 percent and 1.0 percent,respectively. Although recent reports24 have sug-gested increased flap loss in higher body massindex patients, the current study did not supportthese findings.

Despite statistically similar overall complica-tion rates across all body mass index groups,logistic regression analysis identified body massindex as an independent risk factor for any flapcomplication. Interestingly, shorter operativetime was also identified as an independent riskfactor for any flap complication. On close analysis,this counterintuitive finding is likely explained bythe clinical observation of larger diameter perfo-rators generally found in obese patients makingdissection and flap harvest less time-consuming.Indeed, relatively shorter operative times wereseen in severely obese patients undergoing bilat-eral breast reconstruction.

Donor-Site ComplicationsBecause of the quantitative and qualitative het-

erogeneity of recorded variables in published se-ries, overall donor-site complication rates are dif-

ficult to interpret and compare. In the currentstudy, the overall incidence of donor-site compli-cations was 25.8 percent. Increasing body massindex was associated with increased overall donor-site complications coinciding with similar trendsobserved previously among obese patients follow-ing TRAM11,18,29 or DIEP flap30 reconstruction. Onclose examination, however, differences in overalldonor-site complications based on body mass in-dex in the current series can be attributed to de-layed wound healing, as other individual compli-cations were similar across all body mass indexgroups.

Coinciding with previous series,21,24,26,27 thecurrent study reports a 6.5 percent incidence ofseroma formation. Moreover, formation of donor-site seromas were not increased despite increasingbody mass index, in contrast to previous studiesfollowing either DIEP24,26 or TRAM flap–based11,18

reconstruction.Although statistical significance was not reached, a

trend toward increased donor-site infection overallwas observed with increasing body mass index. Sim-ilar trends were reported previously,11,18,30 under-scoring the inherent infectious risk associated withobesity. Analysis of the severity of infectious donor-site complications identifies most as mild, responsiveto outpatient antibiotics, and similar across all bodymass index groups.

In the current study, the incidence of overalldonor-site delayed wound healing was increasedwith increasing body mass index. In fact, along withactive tobacco use, increased body mass index wasidentified as an independent risk factor for delayedwound healing in both flap and donor-site locations.Although primarily attributed to morbidly obese pa-tients, statistical significance was reached betweenbody mass index groups for overall and moderatedonor-site delayed wound healing complications.Severe complications, however, were seldom ob-served and were independent of body mass index.Of note, morbidly obese patients reported an in-creased number of previous open abdominal oper-ations, which has been associated with significantlyincreased delayed wound healing following DIEPflap surgery.27 Supported by the current findings andsignificant personal experience, we suggest that ad-ditional emphasis be placed on the risk of abdominalwound healing complications when counseling pa-tients with a high body mass index preoperatively.

In the current study, despite a high-risk pop-ulation, the incidences of abdominal bulging andhernia formation were minimal (0.2 percent and

Fig. 8. (Above, left) Preoperative anteroposterior view. (Center,left) Preoperative right oblique view. (Below, left) Preoperativeright lateral view. (Above, right) Postoperative anteroposteriorview. (Center, right) Postoperative right oblique view. (Below,right) Postoperative right lateral view.


31e

1.4 percent, respectively) during a lengthy fol-low-up period (mean, 1088 days; range, 720 to1528 days). With preservation of supporting struc-tures, abdominal integrity was effectively main-tained despite increasing body mass index, evi-denced by similar rates of abdominal bulging andhernia formation between all body mass indexgroups.

Despite the relative success of DIEP flap breastreconstruction in obese patients reported in thisstudy, surgeons should be cognizant of the in-creased risk associated with abdomen-based per-forator flap surgery on obese patients. Furtheremphasis should be placed on specific means ofrisk-factor reduction along the reconstructive pro-cess. Preoperatively, maximal weight loss shouldbe encouraged before immediate reconstructionwithout delaying oncologic treatments. In delayedreconstruction cases, reconstruction should bepostponed until nutrition and weight loss are op-timized and consistent abstinence from tobaccouse is achieved. Intraoperatively, abdominal flapundermining should be minimized, with preser-vation of adipocutaneous perforators to the upperabdominal apron. Furthermore, if underminingof the upper abdomen is required for donor-siteclosure, this should be performed in discontinu-ous fashion to maximize blood flow to the upperabdominal apron. In addition, mastectomy skinflap viability should be evaluated critically beforeflap deepithelialization. Postoperatively, optimiza-tion of nutrition is continued in an attempt toencourage wound healing and expedite overallrecovery.

CONCLUSIONSDeep inferior epigastric perforator flaps rep-

resent a significant advance in the evolution ofabdomen-based breast reconstruction. Potentialabdominal donor-site morbidity and flap reliabil-ity play a significant role in determining recon-structive options for high-risk patients such asthose with previous abdominal surgery or obesity.The current findings demonstrate excellent reli-ability of DIEP flaps in obese patients. In addition,by eliminating sacrifice of rectus musculature andmaintaining the integrity of the overlying fascia,DIEP flaps effectively maintain abdominal wall sta-bility even in obese populations. Consequently,DIEP flaps are safe in obese populations and infact may be the recommended method of abdo-men-based breast reconstruction in patients witha body mass index less than 40.

Oscar Ochoa, M.D.9635 Huebner Road

San Antonio, Texas [email protected]

REFERENCES1. Hartrampf CR, Scheflan M, Black PW. Breast reconstruction

with a transverse abdominal island flap. Plast Reconstr Surg.1982;69:216–225.

2. Watterson PA, Bostwick J III, Hester YR Jr, Bried JT, TaylorGI. TRAM flap anatomy correlated with a 10-year clinicalexperience with 556 patients. Plast Reconstr Surg. 1995;95:1185–1194.

3. Paige KT, Bostwick J III, Bried JT, Jones G. A comparison ofmorbidity from bilateral, unipedicled and unilateral, uni-pedicled TRAM flap breast reconstructions. Plast ReconstrSurg. 1998;101:1819–1827.

4. Scheflan M, Kalisman M. Complications of breast recon-struction. Clin Plast Surg. 1984;11:343–350.

5. Hartrampf CR Jr, Bennett GK. Autogenous tissue reconstruc-tion in the mastectomy patient: A critical review of 300 pa-tients. Ann Surg. 1987;205:508–519.

6. Scheflan M, Dinner MI. The transverse abdominal islandflap: Part I. Indications, contraindications, results, and com-plications. Ann Plast Surg. 1983;10:24–35.

7. Holmstrom H. The free abdominoplasty flap and its use inbreast reconstruction: An experimental study and clinicalcase report. Scand J Plast Reconstr Surg. 1979;13:423–427.

8. Schusterman MA, Kroll SS, Weldon ME. Immediate breastreconstruction: Why the free TRAM over the conventionalTRAM flap? Plast Reconstr Surg. 1992;90:255–261; discussion262.

9. Nahabedian MY, Dooley W, Singh N, Manson PN. Contourabnormalities of the abdomen after breast reconstructionwith abdominal flaps: The role of muscle preservation. PlastReconstr Surg. 2002;109:91–101.

10. Alderman AK, Wilkins EG, Kim HM, Lowery JC. Complica-tions in postmastectomy breast reconstruction: Two-year re-sults of the Michigan Breast Reconstruction Outcome Study.Plast Reconstr Surg. 2002;109:2265–2274.

11. Chang DW, Wang B, Robb GL, et al. Effect of obesity on flapand donor-site complications in free transverse rectus ab-dominis myocutaneous flap breast reconstruction. Plast Re-constr Surg. 2000;105:1640–1648.

12. Blondeel N, Vanderstraeten GG, Monstrey SJ, et al. Thedonor site morbidity of free DIEP flaps and free TRAM flapsfor breast reconstruction. Br J Plast Surg. 1997;50:322–330.

13. Kroll SS. Fat necrosis in free transverse rectus abdominismyocutaneous and deep epigastric perforator flaps. PlastReconstr Surg. 2000;106:576–583.

14. Bonde CT, Christensen DE, Elberg JJ. Ten years’ experienceof free flaps for breast reconstruction in a Danish microsur-gical center: An audit. Scand J Plast Reconstr Surg Hand Surg.2006;40:8–12.

15. Sheer AS, Novak CB, Neligan PC, Lipa JE. Complicationsassociated with breast reconstruction using a perforator flapcompared with a free TRAM flap. Ann Plast Surg. 2006;56:355–358.

16. Man LX, Selber JC, Serletti JM. Abdominal wall followingfree TRAM or DIEP flap reconstruction: A meta-analysis andcritical review. Plast Reconstr Surg. 2009;124:752–764.

17. Nahabedian MY, Tsangaris T, Momen B. Breast reconstruc-tion with the DIEP flap or the muscle-sparing (MS-2) freeTRAM flap: Is there a difference? Plast Reconstr Surg. 2005:115;436–444; discussion 445–446.


32e

18. Spear SL, Ducic I, Cuoco F, Taylor N. Effect of obesity on flapand donor-site complications in pedicled TRAM flap breastreconstruction. Plast Reconstr Surg. 2007;119:788–795.

19. Gill PS, Hunt JP, Guerra AM, et al. A 10-year retrospectivereview of 758 DIEP flaps for breast reconstruction. PlastReconstr Surg. 2004;113:1153–1160.

20. Flegal KM, Carroll MD, Ogden CL, Curtin LR. Prevalenceand trends in obesity among US adults, 1999-2008. JAMA2010;303:235–241.

21. Drazan L, Vesely J, Hyza P, et al. Bilateral breast reconstruc-tion with DIEP flaps: 4 years’ experience. J Plast ReconstrAesthet Surg. 2008;61:1309–1315.

22. Selber JC, Serletti JM. The deep inferior epigastric perforatorflap: Myth and reality. Plast Reconstr Surg. 2010;125:50–58.

23. Sullivan SR, Fletcher DR, Isom CD, Isik FF. True incidenceof all complications following immediate and delayed breastreconstruction. Plast Reconstr Surg. 2008;122:19–28.

24. Seidenstuecker K, Munder B, Mahajan AL, Richrath P, Beh-rendt P, Andree C. Morbidity of microsurgical breast recon-struction in patients with comorbid conditions. Plast ReconstrSurg. 2011;127:1086–1092.

25. Chen CM, Halvorson EG, Disa JJ, et al. Immediate postop-erative complications in DIEP versus free/muscle-sparingTRAM flaps. Plast Reconstr Surg. 2007;120:1477–1482.

26. Garvey PB, Buchel EW, Pockaj BA, et al. DIEP and pedicledTRAM flaps: A comparison of outcomes. Plast Reconstr Surg.2006;117:1711–1719; discussion 1720–1721.

27. Parrett BM, Caterson SA, Tobias AM, Lee BT. DIEP flaps inwomen with abdominal scars: Are complication rates af-fected? Plast Reconstr Surg. 2008;121:1527–1531.

28. Keller A. The deep inferior epigastric perforator free flap forbreast reconstruction. Ann Plast Surg. 2001;46:474–479; dis-cussion 479–480.

29. Vyas RM, Dickinson BP, Fastekjian JH, Watson JP, Dalio AL,Crisera CA. Risk factors for abdominal donor-site morbidityin free flap breast reconstruction. Plast Reconstr Surg. 2008;121:1519–1526.

30. Garvey PB, Buchel EW, Pockaj BA, Gray RJ, Samson TD. Thedeep inferior epigastric perforator flap for breast reconstruc-tion in overweight and obese patients. Plast Reconstr Surg.2005;115:447–457.


33e

breast - prma plastic surgery

Documents