BREAST

The In-the-Crease Inferior Gluteal ArteryPerforator Flap for Breast Reconstruction

Robert J. Allen, M.D.Joshua L. Levine, M.D.Jay W. Granzow, M.D.,

M.P.H.

New Orleans, La.

Background: Perforator free flaps harvested from the abdomen or buttock areexcellent options for breast reconstruction. They enable the reconstructivesurgeon to recreate a breast with skin and fat while leaving muscle at the donorsite undisturbed. The gluteal artery perforator free flap using buttock tissue wasfirst introduced by the authors’ group in 1993. Of the 279 gluteal artery per-forator flaps, the authors have performed for breast reconstruction, 220 havebeen based on the superior gluteal artery and 59 have been based on the inferiorgluteal artery. The authors have found that for some women with excess tissuein the upper buttock and hip area, use of the gluteal artery perforator flapresulted in an improvement at the donor site, whereas for others the aestheticunit of the buttock was clearly disrupted. Therefore, the authors have recentlybeen placing the scar in the inferior buttock crease to improve donor-siteaesthetics.Methods: The authors have now performed 31 in-the-crease inferior glutealartery perforator free flaps for breast reconstruction and found that the resultsare very favorable.Results: The removal of tissue from the inferior buttock results in a tightened,lifted appearance. The resultant scar is well concealed within the infrabuttockcrease, and exposure or injury of the sciatic nerve has not occurred. Extendedbeveling at this site is also possible, with less risk of causing an unsightly de-pression. The final aesthetic result of the scar lying within the inferior buttockcrease is very favorable. All patients report satisfaction with the donor site.Complications included one total flap loss, two reoperations for venous con-gestion, one hematoma, two cases with delayed wound healing at the recipientsite, and one with delayed wound healing at the buttock.Conclusion: The in-the-crease inferior gluteal artery perforator flap from thebuttock is now the authors’ primary alternative to the deep inferior epigastricperforator flap from the abdomen for breast reconstruction. (Plast. Reconstr.Surg. 118: 333, 2006.)

Perforator free flaps harvested from the ab-domen or buttock have been shown to beexcellent options for breast reconstruc-

tion. They enable the reconstructive surgeon torecreate a breast with skin and fat while leavingthe muscle at the donor site essentially undis-turbed. For women who would benefit from anabdominoplasty, use of the abdomen as a donorsite has an added advantage of removing theexcess abdominal skin and fat. The buttock do-nor site has similar aesthetic advantages in se-lected patients with excess buttock tissue.

The gluteal artery perforator free flap was firstintroduced by our group in 1993, which used thebuttock as an alternative donor site to the abdo-men for breast reconstruction.1 Of the 279 glu-teal artery perforator flaps we have performedfor breast reconstruction, 220 have been basedon the superior gluteal artery and 59 have beenbased on the inferior gluteal artery. These flapscan provide sufficient tissue for breast recon-struction and can improve the appearance of thebuttock donor site in women with excess tissuehigher up on the buttock. However, we foundthat for many patients the use of superior but-tock tissue clearly disrupted the aesthetic unit ofthe buttock. An area of depression at the upperbuttock often resulted and the scar was left in aprominent area of the buttock.2 Less beveling ofthe flap under adjacent tissue reduced this de-

From the Section of Plastic Surgery, Louisiana State Uni-versity Health Sciences Center.Received for publication February 17, 2005; accepted March29, 2005.Copyright ©2006 by the American Society of Plastic SurgeonsDOI: 10.1097/01.prs.0000227665.56703.a8

www.PRSJournal.com 333

formity at the expense of obtaining a smalleroverall breast flap with a less spherical shape.

Many women with excess buttock tissue willautomatically point to the inferior buttock whenasked where excess tissue might preferably beremoved. Le-Quang and Paletta et al. used thissite for their inferior gluteal myocutaneous freeflaps for breast reconstruction, with excellentcosmetic results.3,4 Initially, we had some con-cern about the use of the inferior portion of thebuttock because of reports that myocutaneousflaps harvested from this inferior site had left thesciatic nerve unprotected, causing significantpostoperative morbidity. However, our clinicalexperience has shown that soft tissue can beharvested from the inferior buttock without sci-atic nerve exposure.

We have now performed 31 in-the-crease infe-rior gluteal artery perforator free flaps for breastreconstruction with favorable results. The re-moval of tissue from the inferior buttock resultsin a tightened, lifted appearance. The resultantscar is well concealed within the infrabuttockcrease, and postoperative exposure or injury ofthe sciatic nerve has not occurred. A greateramount of beveling of adjacent fat may be per-formed with less risk of an unsightly donor site.The inferior gluteal artery perforator also had alonger pedicle than the superior gluteal arteryperforator, making the flap positioning for anas-tomosis easier and insetting more flexible. It alsoallows for use of either the internal mammary orthoracodorsal vessels as recipient vessels. Be-cause of the many advantages of the inferiorgluteal donor site, our preference has evolvedfrom the use of the superior gluteal artery per-forator flap to the use of the in-the-crease infe-rior gluteal artery perforator flap as the flap ofchoice when using the buttock as the donor site.

PATIENTS AND METHODSA series of 31 patients have undergone in-the-

crease inferior gluteal artery perforator free flapbreast reconstruction between March and Decem-ber of 2004. The average patient age was 49.4 years(range, 33 to 61 years). The series of patients wasreviewed for indications for the use of a glutealflap over other donor sites, flap weight, operativetime and hospital length of stay, and intraopera-tive and postoperative complications.

Patient SelectionThe indications for the use of a gluteal flap

instead of an abdominal flap are listed in Table 1.

All patients were evaluated for perforator flap breastreconstruction. The inferior gluteal artery perfora-tor flap was clearly the better choice for patients withinadequate tissue in the abdomen. This was the casein 14 of the patients in this series, accounting for 20of the breast reconstructions. Five patients in thisseries chose the in-the-crease inferior gluteal arteryperforator flap when offered the option, eventhough they had ample abdominal tissue to create areasonable breast. The reason for this preferencewas either an aversion to a long, transverse abdom-inal scar or, for patients who were able to donate alarger, thicker flap from the buttock as comparedwith the flap available from the abdomen, a desire tohave a larger, more projecting breast reconstruction.Less pain from a buttock rather than an abdominaldonor site was also a factor in decision making.

The timing of the breast reconstruction isgiven in Table 2. We define primary reconstruc-tions as taking place immediately after and withthe patient under the same anesthesia as the mas-tectomy. Secondary, or delayed, reconstructionsare defined as taking place after the mastectomyhas been performed and the wound has healed.Tertiary reconstructions are defined as occurringafter another form of breast reconstruction hasfailed or been inadequate. Of the nine tertiaryreconstructions, seven were for failed implant re-constructions and two were for failed transverserectus abdominis musculocutaneous flaps withsubsequent failed implant reconstructions. All pa-tients had an additional operation for flap revisionand nipple creation several months after the initial

Table 2. Timing of Breast Reconstruction

Breast Reconstruction

No. %

Primary 16 52Secondary 6 19Tertiary 9 29

Table 1. Indications for Use of a Gluteal Flap

Reason for Use

No. ofBreast

ReconstructionsNo. of

Patients

Inadequate abdominal tissue 20 14Patient choice 6 5Prior DIEP flap 2 2Prior failed TRAM flap 2 2Prior abdominal liposuction 1 1DIEP, deep inferior epigastric perforator; TRAM, transverse rectusabdominis musculocutaneous.

Plastic and Reconstructive Surgery • August 2006

334

procedure under local anesthesia with sedation onan outpatient basis.

AnatomyThe inferior gluteal artery is a terminal branch

of the internal iliac artery and exits the pelvisthrough the greater sciatic foramen.5 The arteryaccompanies the greater sciatic nerve, the internalpudendal vessels, and the posterior femoral cuta-neous nerve. Below the fascia of the sacrum, thevessel is also surrounded by several distinct fatpads. In this subfascial recess, the inferior glutealvein will receive tributaries from other pelvic veins.The inferior gluteal vasculature will continue to-ward the surface by perforating the sacral fascia. Itwill exit the pelvis caudal to the piriformis muscle.Once under the inferior portion of the gluteusmaximus, perforating vessels are seen branchingout through the substance of the muscle to feedthe overlying skin/soft-tissue envelope.

The course of the inferior gluteal artery per-forating vessels is more oblique through the sub-stance of the gluteus maximus muscle than thecourse of the superior gluteal artery perforators,which tend to travel more directly to the superfi-cial tissue up through the muscle. Thus, the lengthof the inferior gluteal artery perforator and theresultant pedicle length for the overlying inferiorgluteal artery perforator flap are greater than thatfound with a superior gluteal artery perforatorflap. Because the skin island is placed inferior tothe origin of the inferior gluteal vessels, a longerpedicle is also ensured.

The direction of the perforating vessels is superior,lateral, and inferior. Perforating vessels that nourishthe medial and inferior portions of the buttock haverelatively short intramuscular lengths, between 4 and 5cm, depending on the thickness of the muscle. Perfo-rators that nourish the lateral portions of the overlyingskin paddle are seen traveling through the muscle sub-stance in an oblique manner 4 to 6 cm before turningupward toward the skin surface. By traveling throughthemuscleforrelatively longdistances, thesevesselsaremuch longer than their medially based counterparts.The perforating vessels can be separated from the un-derlyinggluteusmaximusmuscleandfasciaandtraceddown to the parent vessel, forming the basis for theinferior gluteal artery perforator flap. Between two andfour perforating vessels originating from the inferiorgluteal artery will be located in the lower half of eachgluteal muscle.6

In 91 percent of cases, the inferior glutealartery then descends into the thigh accompaniedby the posterior femoral cutaneous nerve (S1–S2)

and follows a long course, eventually surfacing tosupply the skin of the posterior thigh.7 Anothernerve branch (S1–S2) also supplies the skin of theinferior buttock. A neurosensory flap can be ele-vated if these nerves are preserved in the dissec-tion of the flap.8

Flap DesignTheflap isdesignedasahorizontalellipse,with the

axis centered above the gluteal crease. The glutealcrease is marked with the patient in the standing po-sition and forms the inferior aspect of the skin paddleellipse. Then, with the patient in the lateral position(similar to the position of the patient at the time ofsurgery), a handheld Doppler probe is used to find thestrongest perforating vessels to the skin. The superioraspect of the skin island ellipse is then marked to cap-ture theseperforators.Thedirectionof theskinpaddleusually parallels the inferior gluteal crease. The dimen-sions of the flap are typically approximately 8 � 18 cm,depending on the amount of skin needed (less with askin-sparing mastectomy) and the amount of excessbuttock tissue available. Preoperative markings areshown in Figure 1.

TechniqueA two-team approach is used. After intubation,

the patient is placed in the lateral recumbent po-sition and ipsilateral chest wall and buttock areprepared into the field. The ipsilateral arm and legare prepared into the field as well to facilitateexposure at the surgical sites.

While a second microsurgeon prepares therecipient site and recipient vessels, incisions are

Fig. 1. Preoperative markings of skin paddle and perforator lo-cations.

Volume 118, Number 2 • Gluteal Artery Perforator Flap

335

made along the previously drawn marks and elec-trocautery is used to divide the fat down to thegluteal fascia. The fat is beveled superiorly andinferiorly to include the maximum amount of fatand soft tissue in the flap as deemed necessary.Additional lateral beveling can also be used toobtain more fat from the lateral thigh or “saddle-bag” area. Care is taken to leave sufficient fat me-dially over the ischium. The fat in this area isdenser and slightly lighter in color than the morelateral fat that is incorporated into the flap. Thefascia of the gluteus maximus is incised laterallyand the dissection proceeds in the subfascial planeto allow easier visualization of the perforators. Per-forators with an artery of at least 1 mm and venaecomitantes are followed through the muscle be-tween the muscle fascicles, which are spread apartto allow deeper dissection. On occasion, a secondperforator is found during the dissection and isincluded if it easily joins the first perforator. Thedissection proceeds under the muscle until a pedi-cle of sufficient length and with sufficient vesselcaliber is obtained to allow microsurgical anasto-mosis with the dissected recipient vessels in thechest. This usually occurs when the perforatingvessels join the inferior gluteal artery. The distalextension of the inferior gluteal artery and veincan be transected to aid in the mobilization of thepedicle. Care must be taken to avoid injury to theposterior femoral cutaneous nerve of the thigh,which travels with the inferior gluteal vessels. Thesciatic nerve is usually not visualized. This resultsin a typical pedicle length of 8 to 11 cm and anarterial diameter of greater than 2 mm with a veinof 3 to 4 mm. Sometimes, adequate vessel size andlength are obtained before entering the inferiorgluteal artery and vein, simplifying flap harvest.

Once good recipient vessels are confirmed,the pedicle is divided and the flap harvested. Thebuttock wound is closed in three layers with asuction drain placed. The closed incision and re-sultant scar will fall within and slightly lateral to thebuttock crease.

The patient is then returned to the supineposition and the microvascular anastomosis andflap insetting are performed. An additional suc-tion drain is placed at the recipient site.

Patients spend one night in the intensive careunit for flap monitoring every 15 minutes for thefirst hour, then every hour. They are then dis-charged to a regular hospital floor, where the flapis checked with a Doppler probe every 4 hours.Patients are out of bed and ambulatory on the firstpostoperative day. Many patients feel very well andreport little or no pain on the second postopera-

tive day. We keep all patients until the fourthpostoperative day for flap monitoring. Patientstypically return to work at 4 weeks after surgery.However, some patients feel well enough to goback to work sooner. We require that they do notengage in any vigorous activity or heavy lifting for4 weeks after surgery.

RESULTSThe sizes and characteristics of the mastecto-

mies, removed implants, and in-the-crease inferiorgluteal artery perforator flaps are listed in Table 3.Final flap inset weights are slightly lower than har-vest weights because of trimming of the flap dur-ing shaping and insetting at the time of the initialprocedure. In most cases, the gluteal donor siteallowed the creation of a flap as large as or largerthan the mastectomy specimen or removed im-plant.

Six of the 31 reconstructions were performedon patients who had undergone radiation therapy(19 percent). The average time of the operationwas 5.3 hours (range, 3.0 to 9.4 hours). The av-erage intraoperative blood loss was 317 cc (range,150 to 1000 cc). Five patients underwent a bal-ancing procedure on the contralateral breast (16percent), which consisted of either a mastopexy oraugmentation (one with a saline implant and onewith autologous lateral thoracic tissue) at the timeof in-the-crease inferior gluteal artery perforatorreconstruction. Hospital length of stay was an av-erage of 4.2 days (range, 4 to 7 days).

In the series, there was one flap loss secondaryto venous thrombosis that occurred on postoper-ative day 4. This patient had undergone staged,bilateral in-the-crease inferior gluteal artery per-forator reconstructions because the abdomen wasdeemed insufficient to provide enough tissue fortwo breast reconstructions. A successful deep in-ferior epigastric perforator flap was subsequentlyperformed for the unilateral reconstruction, withan initial weight of 589 g reduced to 473 g afterinset. This compared with the final inset weightsof 495 g and 530 g with the in-the-crease inferior

Table 3. Mastectomy, Flap, and Implant Weights

Average Range

Mastectomy weight, g 305 156–654Removed implant weight, g 510 129–763Flap harvest weight, g 425 148–833Final flap weight after inset, g 407 137–806Final flap weight

% of harvest flap weight 96 91–100% of mastectomy weight 124 59–190% of removed implant weight 108 70–161

Plastic and Reconstructive Surgery • August 2006

336

gluteal artery perforator flaps from each buttockwith this particular patient.

Two additional patients were returned to the op-erating room after the completion of the initial recon-struction for successful treatment of venous insuffi-ciency. These were both thought to be secondary totwisting or kinking of the flap vein away from the siteof anastomosis and were probably inset-related. Onepatient developed a hematoma that resolved withoutintervention. Two patients had problems with woundhealing and wound breakdown at the recipient site.Both patients had undergone previous radiation ther-apy to the chest wall, and wounds eventually healed inboth. One patient suffered wound breakdown at thedonor site that healed with conservative wound care.These complications resolved without flap loss and aresimilartoproblemsthatoccurwithothertypesofbreastreconstruction. One patient reported initial minor ad-justments to their sitting position when sitting on ahard surface. This problem resolved within 6 weeksafter the operation. No other patients had any com-plaints about discomfort on sitting when asked at 3months after the operation. All patients were seen fornipplereconstructionat3monthspostoperatively.Thelongest follow-up period was 9 months in one patient.

We typically use the internal mammary vesselsas recipient vessels. Favorable internal mammaryperforators superficial to the pectoralis muscleand thoracodorsal vessels were also used, as dis-played in Table 4. In one patient, the internalmammary artery and a favorable perforating veinwere used.

DISCUSSIONMicrovascular transfer of gluteal tissue was ac-

complished for the first time using a superior glutealmusculocutaneous flap by Fujino in 1975. Le-Quangsubsequently described transfer of inferior glutealmusculocutaneous flaps in 1978.3 Gluteal musculo-cutaneous flaps were later described by both Shawand Paletta et al.4,9 Shaw used the superior glutealmyocutaneous free flap, but acknowledged its diffi-culty secondary to the short pedicle length. Palettaet al. preferred the inferior gluteal flap because itprovided a longer pedicle and more tissue bulk, andthe incision was hidden in a more cosmetic location:

the inferior gluteal crease. However, use of this flapwith harvest of gluteus muscle often left the sciaticnerve unprotected, with no soft-tissue cover betweenthe skin and the nerve, with many patients sufferingsignificant postoperative symptoms. In addition, de-spite the proposed longer pedicle, a high number ofpatients continued to require vein grafts and en-dured the increased risk of vascular complications.In our experience with the in-the-crease inferior glu-teal artery perforator flap, the sciatic nerve was seenin only two dissections. In these cases, the gluteusmaximus fibers, once the pedicle was harvested, fellclosed over the nerve and provided a thick musclemass between the skin and the nerve. Neither ofthese patients nor any other patient in this seriessuffered postoperative complaints related to the sci-atic nerve. The difference in the perforator flap ascompared with the previous technique is that nomuscle is taken. Therefore, the gluteus muscle con-tinues to cover and protect the sciatic nerve, andexposure is not a problem.

Although the approximately 8-cm length ofthe superior gluteal artery perforator pedicle is agreat improvement over the 2- to 3-cm superiorgluteal artery myocutaneous flap pedicle length,the superior gluteal artery perforator pedicle isnot always optimal. The superior gluteal arteryexits the sciatic foramen and immediately sendsperforators up through the gluteus muscle. Thus,the pedicle length is typically equal to the lengthof the perforator plus a short cuff of superiorgluteal artery. If a medial perforator is chosenclose to the sciatic foramen, the resultant pediclemay be no more than 6 cm long. In contrast, aspreviously discussed, the inferior gluteal arterytakes an inferior course underneath the gluteusmaximus muscle, sending out more obliquely ori-ented perforators all along its length. Therefore,when the skin island is designed inferiorly, thesurgeon is ensured of obtaining an overall longerpedicle that is typically 8 to 11 cm long.

This added length allows for more leeway inorientating and insetting the flap, and can makethe anastomosis easier. It also allows for greaterreach to the thoracodorsal vessels, if necessary,with better medialization of the flap on the chestwall should these vessels be used.

Two problems occurred with venous conges-tion of the flap. Both were successfully ad-dressed, and the flaps subsequently did well, andwere thought to be caused by a kinking or twist-ing of the vein rather than a problem at theanastomotic site. Our group uses the vein cou-pler almost exclusively for the venous anasto-mosis, and we feel that the coupler acts as a stent

Table 4. Recipient Vessels Used

Anastomoses

No. %

Internal mammary 51 82Internal mammary perforator 5 8Thoracodorsal 6 10

Volume 118, Number 2 • Gluteal Artery Perforator Flap

337

Fig. 2. Postoperative views of the patient in Figure 1 at 9 months postoperatively.

Fig. 3. Preoperative views of a patient with right breast cancer.

Fig. 4. Postoperative views obtained at 6 months postoperatively. The contralateral breast underwent autologous augmen-tation for symmetry with a pedicled intercostal perforator flap at the time of inferior gluteal artery perforator second-stagerevision and nipple creation.

Plastic and Reconstructive Surgery • August 2006

338

for the vein and keeps the lumen open, with theresult that very few problems occur at the site ofthe venous anastomosis itself.

Our experience using the gluteal artery per-forator flaps for patients who are not deep inferiorepigastric perforator/superficial inferior epigas-tric artery flap candidates has been favorable.2 Un-til recently, the vast majority of our buttock flapshave been harvested from the upper buttock andtherefore based on the superior gluteal artery per-forators. Over the past 12 years, we have occasion-ally designed the skin island slightly lower, accord-ing to the patient’s anatomy or preference, andserendipitously used the inferior gluteal arteryperforators. These upper buttock gluteal arteryperforator flaps have occasionally resulted in adepression at the donor site that required revision.We have used liposuction around the scar andautologous fat injection to fill in the scar, depend-ing on the patient’s anatomy.

Our experience has been that the donor-sitedefect of the in-the-crease inferior gluteal arteryperforator flap is aesthetically favorable and in-conspicuous in the great majority of patients. Theaesthetic unit of the buttock is preserved and thescar falls in the inferior gluteal crease. Significantsoft-tissue depression at the donor site occurs lessoften and appears less noticeable than with thetypical superior gluteal artery perforator donorsite. This results in an improved postoperative ap-pearance and also allows a greater amount of bev-eling and fat harvest with the flap in thin patients(Fig. 2). The one wound dehiscence that occurredmay have been the result of tension on the wound.To prevent this, we now undermine both inferiorlyand superiorly to bring the wound edges togetherunder minimal tension.

The delayed wound healing at the recipient siteoccurred in two patients who had been irradiated.The damaged skin at the recipient site probablycontributed to the delayed wound healing. The in-ferior gluteal artery perforator flap has a high fat-to-skin ratio as compared with the abdominal flaps.The typical skin island width is 7 to 10 cm. It is a thickflap and is sometimes difficult to inset in the irradi-ated chest after skin excision. If irradiated skin willbe removed and there will be a large skin require-ment, the abdomen may be a better choice for pa-tients who have the tissue to donate.

CONCLUSIONSOverall, in-the-crease inferior gluteal artery

perforator flaps allow reliable, aesthetic recon-

struction of the breast (Figs. 3 and 4) without thesacrifice of muscle at a donor site. Patient satis-faction has been very high and is comparable tothat in the deep inferior epigastric perforator andsuperficial inferior epigastric artery reconstruc-tions. We feel that the buttock is a reliable soft-tissue source for breast reconstruction. Accordingto the patient’s history and physical shape, a glu-teal flap can be used successfully instead of anabdominal flap. A sufficient amount of soft tissuefor an aesthetic breast reconstruction can be ob-tained reliably in most patients. The evolution ofsurgical technique from the superior gluteal ar-tery perforator to the in-the-crease inferior glutealartery perforator flap now allows an aestheticallysuperior result to be obtained for both the recon-structed breast and the donor site in the vast ma-jority of patients.

Joshua L. Levine, M.D.Section of Plastic Surgery

Louisiana State University andNew York Eye and Ear Infirmary

1776 Broadway, Suite 1200New York, N.Y. 10019jlevine@diepflap.com

DISCLOSURENone of the authors has any financial interest in any

medical device or product mentioned in this article.

REFERENCES1. Allen, R. J., and Tucker, C., Jr. Superior gluteal artery perfo-

rator free flap for breast reconstruction. Plast. Reconstr. Surg.95: 1207, 1995.

2. Guerra, A. B., Metzinger, S. E., Bidros, R. S., Gill, P. S., Dupin,C. L., and Allen, R. J. Breast reconstruction with gluteal arteryperforator (GAP) flaps: A critical analysis of 142 cases. Ann.Plast. Surg. 52: 118, 2004.

3. Le-Quang, C. Secondary microsurgical reconstruction of thebreast and free inferior gluteal flap. Ann. Chir. Plast. Esthet. 37:723, 1992.

4. Paletta, C. E., Bostwick, J., III, and Nahai, F. The inferiorgluteal free flap in breast reconstruction. Plast. Reconstr. Surg.84: 875, 1989.

5. Strauch, B., and Yu, H. L. Gluteal region. In B. Strauch andH. L. Yu. Atlas of Microvascular Surgery: Anatomy and OperativeApproaches. New York: Thieme Medical, 1993. Pp. 102–119.

6. Koshima, I., Moriguchi, T., Soeda, S., Kawata, S., Ohta, S., andIkeda, A. The gluteal perforator-based flap for repair of sacralpressure sores. Plast. Reconstr. Surg. 91: 678, 1993.

7. Windhofer, C., Brenner, E., Moriggl, B., and Papp, C. Re-lationship between the descending branch of the inferiorgluteal artery and the posterior femoral cutaneous nerveapplicable to flap surgery. Surg. Radiol. Anat. 24: 253, 2002.

8. Boustred, M. A., and Nahai, F. Inferior gluteal free flap breastreconstruction. Clin. Plast. Surg. 25: 275, 1998.

9. Shaw, W. W. Breast reconstruction by superior gluteal micro-vascular free flaps without silicone implants. Plast. Reconstr.Surg. 72: 490, 1983.

Volume 118, Number 2 • Gluteal Artery Perforator Flap

339