C L I N I C S I NP L A S T I C

S U R G E R Y

Clin Plastic Surg 33 (2006) 449–466

449

Complications of ButtocksAugmentation: Diagnosis,Management, and PreventionTerrence W. Bruner, MD, MBAa, Thomas L. Roberts III, MD, FACSb,*,Karl Nguyen, MDc

- Introduction- Silicone implants

Subcutaneous placementIntramuscular or submuscular implant

augmentationSubfascial implant placementAutologous fat grafting

- Factors affecting the incidence ofcomplications in augmentation byautologous fat grafting

Preoperative body mass index BMIVolume of fat graftedAmount of fat aspirated

- SummarySubcutaneous implant augmentationIntramuscular implant augmentationSubfascial implant augmentationButtocks augmentation by autologous fat

grafting- Acknowledgments- References

Introduction

The past 2 decades have witnessed extensive growthand interest in body contouring. The buttocks areno exception. Currently, the concept of beautifulbuttocks includes maximum prominence at thelevel of the upper and mid buttocks, smooth in-ward sweep of the lumbosacral area and waist,and minimal or no infragluteal crease, without pto-sis above this line [1]. According to Douglas andcolleagues [2], any significant disproportion ofbody components, such as diminutive buttocks,can be a significant psychological detriment to theindividual. The specialty of plastic surgery has

0094-1298/06/$ – see front matter ª 2006 Elsevier Inc. All righplasticsurgery.theclinics.com

witnessed great strides in surgical procedures tochange body contour. The earliest operations gener-ally consisted of wedge excisions of skin and adi-pose tissue, thereby improving body contour butleaving extensive surgical scars. With the advent ofsilicone breast implants in the 1960s [3], it wasnot long before creative surgeons would use themfor gluteal augmentation. The first case report wasperformed by Bartels and coworkers in 1969 [4],and subsequently followed by Cocke and Ricketsonin 1973 [5]. While silicone breast implants havebeen effective at augmenting the buttocks, experi-ence has shown that they are prone to ruptureand leakage because of their thin silicone shell

a Division of Plastic Surgery, Baylor College of Medicine, 1709 Dryden, Suite 1600, Houston, TX 77030, USAb Medical University of South Carolina at Spartanburg, Spartanburg, SCc Division of Plastic Surgery, University of California at San Diego, Chancellor Park Office, 4510 ExecutiveDrive, #103, San Diego, CA 92121, USA* Corresponding author. Carolina Plastic Surgery, 100 East Wood Street, Suite 100, Spartanburg, SC 29303,USA.E-mail address: tlr3@bellsouth.net (T.L. Roberts, III).

ts reserved. doi:10.1016/j.cps.2006.04.005

Bruner et al450

[6]. In fact, it was soon thereafter that major com-plications of infection and dehiscence associatedwith gluteal implants were reported [2]. These com-plications continue to haunt buttocks augmenta-tion with gluteal implants.

The first documented free fat grafts were estab-lished by the end of the nineteenth century [7];however, it was not until Illouz introduced liposuc-tion in 1983 [8] that large volumes of potentiallytransplantable fat became available for augmenta-tion of the gluteal region [9,10].

Currently, placement of silicone implants con-tinues to be the most popular method for buttocksaugmentation [11]. However, surgeons’ preferencefor autologous fat grafting is increasing in popular-ity because of certain advantages in creating idealbuttocks shape and size. (See article by Robertsand colleagues, this issue.) Unfortunately, both sur-gical techniques can have significant complications[1,6,11–15], and there has been a great disparityamong what has been reported in the literature(Tables 1 and 2). Dr Sidney Coleman [24–26]

was one of the early advocates of a meticulous,drop-by-drop fat-grafting technique, and his persis-tence brought autologous fat grafting into themainstream, primarily for facial soft tissue. Robertset al [1] presented the first large well-documentedseries of micro fat grafting for moderate buttocksaugmentation and the complications associatedwith this volume of grafting.

Many reports on silicone implants for buttocksaugmentation were from outside the UnitedStates (Table 1), and few of these were open indiscussing complications. Mendieta [6] publishedthe first large American series on buttocks im-plants and was very thorough in presenting com-plications. This candor opened the door and gaveother surgeons the courage to present and discusscomplications. This openness has led us, workingtogether, to begin solving some of the problemsthat confront all types of buttocks augmentation.For the most current information on techniquesand complications, we strongly recommend theinstructional course offered by Roberts, Young,

Table 1: Complications of sub- or intramuscular buttocks implants

Authors Country Year Patients Complications

Robles et al [16] Argentina 1984 9 Muscle dehiscence:1/9 Seroma: 1/9

Gonzales-Ulloa [17] Mexico 1991 Not stated Not stated(subcutaneousplacement)

Vergara & Marcos [18] Mexico 1996 16 ‘‘None. All patientssatisfied with results.’’

Mendieta [6] United States of America 2003 73 Wound dehiscence: 30%Infection: 1/73Implant exposure: 2/73Implants too firm: 10%Transient sciaticparesthesias: 20%Most frequentcomplaint: Implantsnot large enough

Vergara & Amezcua [15] Mexico 2003 160 Complications: 16/160Wound dehiscence: NotmentionedSeroma: 7 (4%)Asymmetry: 4 (2.6%)Capsular contracture:3 (2%)Rupture: 1 (0.7%)Overcorrection: 1 (.7%)Satisfactory results: 90%

Gonzalez [19] Brazil 2004 746 Infection: 3/746Dehiscence: 14%Reoperation: 2/746Rupture: 2/746Seroma: 1/746All patients greatlysatisfied with results

Complications of Buttocks Augmentation 451

Table 2: Complications of buttocks autologous fat grafting: extraction from literature

Authors Country Year Patients

Vol (cc)fatgrafted

Complications/Results

Estimatedfat survival

Chajchir et al[20]

Argentina 1990 Notstated

Notstated

Not stated

Toledo [21] Brazil 1991 Notstated

Up to 500 Not stated

Lewis [9] UnitedStates ofAmerica

1992 12 30–150 None. All pointssatisfied

Guerrerosantos[10]

Mexico 1996 6 150–200 None. All pointsdemonstrateexpectedimprovement

Cardenas-Camarena et al[13]

Mexico 1999 66 210 Seroma: 4/66 (6%)Cellulitis: 12/66Fat embolism: 1/66Tissue irregularities:8/66; 90% of patientswith good orexcellent results

Peren et al [12] Mexico 2000 40 120–240 pain: 31/40, seroma:4/40 (10%)hyperpigmentation:1/40

100%

Pedroza [22](Dermatologist)

Colombia,S.A.

2000 879 200–350 Cellulitis: 1/879 (min)Long-lasting results

0.5–1.0 cmlessprojection

Roberts et al [1] USA &Brazil

2001 566 400 Cellulitis: 1.9%Seroma/hematoma:0.8%Abscess: 0.2%Blood transfusion:0.4%Reaugmentationrate: 10%

50%–75%

RestrepoAhmed [11]

Colombia,S.A.

2002 96 410 ParesthesiasSmall irregularities:10/96Septic shock: 1/96Satisified: 94/96Asymmetries: 3/96Not satisified: 2/96

60%–80%

Monreal [23] Spain 2003 No # ofbuttocks

453 ‘‘None of any sort.’’86.4% good or verygood

80%–90%

Murillo [14] Colombia,S.A.

2004 162 250–260(700)

Abscess: 1/162Seroma: 64/162(40%)Sacral numbness:10%Asymmetry: 2.4%

80%

Bruner et al452

Mendieta, De La Pena and colleagues at both theAmerican Society for Aesthetic Plastic Surgery andAmerican Society of Plastic Surgeons future an-nual meetings.

There are three current methods for buttocks aug-mentation: autologous micro fat grafting, intramus-cular silicone prosthesis, and subfascial siliconeprosthesis. In addition, subcutaneous placementof silicone implants has been attempted in the past,but has an unacceptably high risk of long-termproblems [2,5,27]. Buttocks augmentation by anyof the aforementioned techniques is major surgerywith potential for major complications, includingserious infections (cellulitis, abscess, and sepsis),wound dehiscence, loss of implant, hematologicand metabolic disturbances including DIC (dissem-inated intravascular coagulopathy), pulmonaryedema, electrolyte imbalance, and aspiration pneu-monitis. Some complications, such as infections,seromas, and wound problems are common to alltechniques, but even these present differently de-pending on the augmentation technique. For thisreason, we will discuss the complications separatelyfor each procedure.

Silicone implants

Early attempts at implant augmentation of the but-tocks often employed silicone gel breast prostheses[2]. The buttocks are routinely subject to both shearand compressive forces much greater than thebreasts [28,29], so it is not surprising that breast im-plant failure was a common and major problemwhen they were used in the buttocks [2]. Currentlyapproved buttock implants in the United Statesare made of semisolid silicone elastomer, whichis much more durable than the breast implant.These are, however, more palpable when placedsubfascially [6,30].

Subcutaneous placement

Subcutaneous placement of implants requires thesimplest dissection, therefore is a tempting ap-proach, especially for a surgeon with limited experi-ence in buttocks augmentation. Only one paper inthe English language literature, from 1991,docu-ments a series of patients who received subcutane-ous placement of buttock implants [17], and noother author has supported this technique in thesubsequent 15 years. When the fascial bands hold-ing the skin to the gluteus muscle are transected bysubcutaneous dissection, there is no ligamentous ormuscular structure to restrain the implant. This re-sults inevitably in one or more of the followingproblems: (1) palpability, (2) visible margins of im-plant, (3) excess mobility, (4) displacement, (5)ptosis, (6) extrusion, or (7) bizarre-appearing en-capsulation. Fig. 1A shows the ptotic position ofa subcutaneous implant with a ‘‘double bubble’’ ef-fect at the arrow. The pocket extends 3 inches supe-rior to this point, and the implant moves back andforth when the patient lies down.

Fig. 1B shows the patient 2 years after implant re-moval; the skin is still ptotic and mobile due toa persistent thick capsule and fluid. Fig. 2 is an ex-ample of the marked deformity that can occur whencapsular contracture occurs around a subcutane-ously placed implant. We have never seen this visi-ble encapsulation when the implant is placedsubfascially or intramuscularly. Because of the highincidence of long-term complications, we feel thatsubcutaneous placement of silicone implants inthe buttocks should not be performed.

Intramuscular or submuscular implantaugmentation

Most of the early reports on intra- or submuscularbuttocks augmentation were from outside the

Fig. 1. (A) Excess mobility of sub-cutaneously placed gluteal im-plant. Ptosis results in ‘‘doublebubble’’ effect indicated by thearrow. Pocket has elongated 3 in-ches inferiorly. (B) Two years afterremoval of subcutaneous buttockimplants, the skin is still ptoticand mobile due to a persistantthick capsule and fluid.

Complications of Buttocks Augmentation 453

Fig. 2. Bizarre mushroom-like encapsulation of subcutaneously placed implants.

United States [16], and even in large series, reported‘‘no complications’’ [18], or complications wereminimized and not quantified [19]. We are indebtedto Dr Constantino Mendieta [6], who published thefirst large American series on intramuscular buttocksimplant augmentation, for his transparency inreporting the complications and challenges con-fronting this technique. The submuscular plane isdifficult to identify, smaller than the intramuscularplane, and in direct continuity with the sciatic nerve.For these reasons, most implant surgeons now usethe intramuscular plane; therefore, our discussionwill be limited to this technique.

Wound dehiscenceMendieta [6] reported approximately 30% in-cidence of wound dehiscence in his series of 73patients. Although the midline incision marginsoccasionally look dusky from the beginning, dehis-cence usually does not occur until postoperativeday 6 to 14. Typically, the skin appears healthy,

then separates, revealing brown, nonviable subcu-taneous tissue up to 2 cm from the wound edge.Cultures obtained from the wound may grow fecalbacteria, or they may be sterile. This, combinedwith the fact that the white blood cell count of a pa-tient with dehiscence is usually normal, suggeststhat dehiscence is not primarily of infectious origin.Fortunately, unless there is dehiscence of the glu-teus muscle incision and resultant implant expo-sure, the wound usually heals secondarily afterrepeated debridement without loss of the implant.This does, however, cause some social inconve-nience to the patient with drainage and an openwound for 2 to 3 months until healing occurs bysecondary intention (Fig. 3A, B).

There may be multiple factors that contribute todehiscence of the midline wound in implantaugmentation.

� Intrinsically poor blood supply: The interglu-teal crease is a ‘‘watershed’’ area without majoridentifiable arterial supply.

Fig. 3. (A) Intramuscular ImplantAugmentation: Dehiscence of inter-gluteal incision at 1.5 weeks postop.(B) Samepatient,1monthpostop in-tramuscular implant augmentation.

Bruner et al454

� Desiccation and crushing of the skin and sub-cutaneous tissue because of the prolongedand intense retraction necessary for dissectionand visualization (Fig. 4A, B). Because im-plants are always bilateral, both sides of the in-cision are traumatized.

� Inadequate muscle or fascial coverage, or mus-cle closure that is too tight, can contribute toshearing motion of the implant with hip flex-ion, extension, or sitting. This in turn mayresult in a periprosthetic seroma, causing phys-ical or hydraulic disruption of the wound.

Mendieta [31] reported up to 80% dehiscence inpatients who are overweight or who needed im-plants over 350 cc, and has switched to autologousfat grafting for these patients.

Potential means of decreasing dehiscence includethe following. (1) Keeping both wound edges moistand relaxing the retraction when possible shouldimprove perfusion and minimize tissue trauma.(2) An effort is being made to develop a noncrushingretractor. (3) Some authors are trying to avoid a sin-gle midline incision (both sides of which becometraumatized) by using two paramedian incisions[31], or leaving a midline ellipse of deepithelializedskin with a broad base of subcutaneous tissues overwhich the incision is closed [30,32]. Early reports ofthese techniques offer some encouragement. (4)The judicious use of drains for a limited period oftime may decrease seroma formation as a cause ofdehiscence. (5) Appropriate antibiotic coverage(typically cephalosporin plus gentamicin) may helpdecrease infectious causes of dehiscence, providedthe wound is healthy. (6) The use of three layers ofskin glue (Dermabond and others) seems to de-crease the incidence of wound separation, actingas an antimicrobial barrier. The skin glue will gener-ally remain intact for 8 to 10 days. By giving mechan-ical support to the skin, it may decrease the need foran excessively tight skin suture, thereby decreasing

this compromise of the circulation to the woundedges.

Seroma formationSeromas around the implant may occur in 2% to4% of intramuscular augmentations [6,15], evenwhen drains are used. This problem is greater fortextured implants, and Vergara [33] himself hasnow changed to smooth implants made of cohesivegel. Smaller seromas may resolve spontaneously,but most require repeated aspiration. When theybecome chronic, Mendieta [31] recommends sur-gery to score the capsule, implant replacement,and drainage for 3 days. Vergara and Amezcua[15] insert drains and irrigate with hydrocortisone(Cortef). A perplexing problem is the late appear-ance of seromas, occurring up to 1 to 2 yearspostop.

InfectionThis occurs in about 2% to 3% in large series [6,15],and may result in loss of the implant. The use ofa plastic sleeve around the implant during insertioncan decrease contamination of the implant by de-bris and bacteria. The anus is only 3 to 4 cm fromthe incision, and we staple or suture a gauze soakedwith povidone-iodine over the anus during surgery.

Loss of the implant as a result of muscledehiscence and implant exposureThe incidence for all patients is 2% to 5% [6], butup to 30% in overweight patients or those receivinglarger implants (>350 cc or >3.5-cm projection)[31]. To decrease these problems, check the sub-muscular pocket with a sizing implant to be surethe muscle incision can be easily closed under min-imal or no tension. Mendieta uses intraoperativetissue expansion while dissecting the other side. Ifthe muscle still cannot be closed with minimal ten-sion, a smaller implant must be used.

Fig. 4. (A) Prolonged, forceful retraction of the intergluteal incision is necessary for dissection of the pocket andmay contribute to the high incidence of dehiscence after intramuscular or subfascial implant augmentation. (B)Additional force and wound edge trauma is necessary to insert the implant.

Complications of Buttocks Augmentation 455

Asymmetry, implant migration,or malpositionThese problems develop in 3% to 5% of intramus-cular implants [6,15]. One of the most commoncauses is to inadvertently allow the dissection tobecome too superficial while progressing laterallyor superiorly. Because the muscles are somewhathemispherical in shape, dissecting tangentially per-forates the fascia, resulting in the lateral or superiormargins of the pocket extending into the subcuta-neous plane. Once the edge of the implant entersthe subcutaneous plane, it is easily palpable, maybecome painful, becomes progressively more mo-bile, and can cause visible deformity. To avoid this,dissection must continuously curve downward(deeper) as dissection proceeds laterally or superi-orly. Correction of malpositioning requires reoper-ation, capsulorrhaphy, or creation of a new pocket.

Implant ruptureThis is uncommon, occurring in fewer than 1% ofcases [15]. Two factors may increase the risk of rup-ture: (1) using a ‘‘carving block’’ to cut your own im-plant, and (2) inadequate muscle repair so thata portion of the implant herniates into the subcuta-neous pocket, subjecting the implant to pinching orshearing forces where it emerges through the tightmuscle opening.

Capsule contractureThis is reported in the range of 2% [15].

Sciatic paresthesiasMendieta [31] finds that transient sciatic paresthe-sias occur in about 20% of his intramuscular

implant cases. They respond to Neurontin (gaba-pentin) and disappear by 3 weeks.

Aesthetic problems and limitationsBecause of the limitations of the intramuscularspace and the underlying sciatic nerve, intramus-cular implants can only be placed in the upperor upper and mid buttock. Often, there is flatnessalso in the lower buttocks, and intramuscularaugmentation may result in a ‘‘double bubble’’deformity (Fig. 5A, B). This is especially true inthin patients with very little subcutaneous tissueon the buttocks. Vergara and Mendieta (personalcommunication, 2005) are independently devel-oping an elongated or oval-shaped implant, butthis still should not extend more than 2 cm be-low the tip of the coccyx. Thus, the ideal patientfor intra- or submuscular augmentation has flat-ness superiorly but fullness inferiorly that willdisguise the lower edge of the implant.

The other aesthetic problems include patientscomplaining the implant feels too firm (10%),and the most common complaint is that the im-plants are too small. This is especially true for Afri-can American patients, who seek a very largeaugmentation, and for Asian patients whose narrowpelvis usually permits only the use of a small im-plant (100 to 240 cc).

Technical considerationsThe submuscular or intramuscular dissection isdone primarily in a blunt fashion and can be verybloody. There is no visible marker to indicate thecorrect surgical plane. The operation is very painfulfor 1 to 2 weeks; this can be helped by using a pain

Fig. 5. (A) A poor candidate for in-tramuscular implant augmenta-tion. The lower pole of thebuttocks is flat, and the subcuta-neous tissues are thin. (B) Aftersubmuscular augmentation, theupper curve of the buttock looksgood, but the mid buttock revealsthe lower edge of the implant asa ‘‘double bubble’’ deformity (atarrow).

Bruner et al456

pump (‘‘on-Q’’ and others) with lidocaine (Xylo-caine) dripped around the implant for 3 days.

Subfascial implant placement

The placement of silicone implants under the fas-cial covering of the gluteus maximus muscle waspioneered by de la Pena of Mexico [30]. This fasciais not a smooth continuous sheet, but is intimatelyconnected to innumerable vertical septae perforat-ing the fibers of the gluteus. This makes for a metic-ulous dissection, taking about 4 hours, and requiresconstant retraction and direct visualization. The re-ward is a larger potential pocket, especially inferi-orly. A textured implant is used, with the goal ofless unwanted mobility. This technique offers thepossibility of creating more fullness inferiorly, andthe use of a larger implant. Problems with this ap-proach include the following.

DehiscenceDe la Pena also reports about 30% dehiscence ofthe midline intergluteal incision [32]. He notes thatone third of these are more than 2 cm deep andtherefore pose a greater problem. The other twothirds are more superficial (< 2 cm deep), and even-tually heal secondarily, although still require 1 to 3months for complete closure.

InfectionsDe la Pena reports a 7% infection rate, but only 1%resulted in loss of the implant [32].

SeromasDe la Pena reports a 2% incidence with subfascialplacement [32]. It is important to note that de laPena’s experience is entirely with cohesive gelimplants. These are not currently available in theUnited States, and semisolid elastomer must beused. The experience of this author combined withthat of Dr. Schlomo Widder [34] consists of 31 pa-tients using this semisolid textured de la Pena im-plant subfascially, and our total complication ratewas 51% (Table 3). We did experience 19% sero-mas, some of which may be blamed on early drainremoval; but leaving a drain in for 10 days resultedin our one infection requiring implant removal.Frustratingly, one third of these seromas appearedlater (ie, 1 to 2 years after surgery). Because of thishigh seroma rate, Widder has exchanged implantsfrom the textured de la Pena semisolid implantsto the smooth semisolid implant in four cases withno subsequent seroma formation. In one of thesefour cases, the skin became detached from the pre-sacral fascia, and the implants overlapped eachother in the midline and had to be removed.

PtosisUse of larger implants (545 cc or more), a proposedadvantage of the subfascial technique, can result insevere inferior ptosis (Fig. 6A, 6B) requiring reoper-ation, capsulorrhaphy, change to a smaller implant,vest-over-pants fascial repair, and excision of result-ing redundant skin through a new infraglutealincision.

Table 3: Complications of subfascial buttocks augmentation (with textured semisolid implant)(Roberts & Widder; n 5 31 patients)

Major complications Patients (n 5 31) Incidence

Wound dehiscence 5 16%Large seromas (up to 1800 cc)requiring multiple aspirations

6 19%

Early seromas—4Late seromas (1–2 years)—2

Infection requiring implant removal 1 3%Capsule contracture, extrusion, andloss of implant

1 3%

Implant flipping 180º (concaveside out)

2 6%

Severe ptosis requiring reoperation 1 3%Reoperation for patient aestheticdissatisfaction

1 3%

Total major complications 17 in 11 35% with major complicationsMinor complicationsSmall wound separation 5 16%Total complication rate 22 in 16 51% had 1 or more complicationsNote: Lower pole of implantpalpable

100%

Complications of Buttocks Augmentation 457

Fig. 6. (A) Two weeks after large(545 cc) subfascial implant aug-mentation. (B) Same patient 5months following large subfascialaugmentation; visible ptosis ofthe left implant (arrow) requiredreoperation through a new infra-gluteal incision, lateral capsulor-rhaphy, change to a smallerimplant bilaterally, and vest-over-pants fascial repair inferiorly.

Implant palpabilityOne hundred percent of the semisolid implants areeasily palpable by the patient (and her partner)[34], which requires complete acceptance of thispalpability by the patient after preoperative discus-sion. We find many patients opt not to have thesubfascial procedure because of this problem.

Autologous fat grafting

Early reports of buttocks augmentation by fat graft-ing [9,10,20] and some recent reports [22,23], didnot give rigorous reports of their complications orreported ‘‘none.’’ Most of these authors only graftedtiny amounts of fat per buttock (30 to 210 cc),which could keep complications low, but thisamount is so small that there was little or no visibleaugmentation. Most of the apparent improvementcame from the adjacent liposuction.

Based on our experience and the experience oflarger detailed series [11,14], it is impossible to per-form substantial augmentation of the buttockswithout complications; authors reporting this mustbe ignoring complications that eventually resolve,such as infections, seromas, paresthesias, and soforth.

The first authors to detail their complicationswere Cardenas-Camerena and coworkers [13]. Eventhough they only grafted an average of 210 cc/but-tock, they reported 18% cellulitis, 6% seromas,12% tissue irregularities, and one case of fat embo-lism in their series of 66 patients.

In our own experience of 261 cases of micro fatgrafting to the buttocks, the first large strictly Amer-ican series, we have meticulously tracked everycomplication (Table 4). In our early experience,from 1998 to 2002, we were frustrated by the rateof infection in the recipient site and fluid accumula-tion in the donor site, and continuously looked for

additional ways to decrease these problems. Thethird column of Table 4 shows our results in the last100 cases with a dramatic decrease in major compli-cations from 17.6% to 7% with our current tech-nique; if we eliminate those patients with veryhigh volume grafting (>1000 cc/buttock), the majorcomplication rate is reduced again, down to 2.7%(fourth column of Table 4).

It is our experience (and that of others with rea-sonably large numbers of cases) [11,14] that it isnecessary to graft 500 to 900 cc or more of fat perbuttock to obtain the range of gluteal fullness andshape requested by the various ethnic groups.When this volume of augmentation is performed,the following complications may occur.

InfectionIn our first 100 cases, we experienced an almost14% infection rate, about one half of these beingmajor and the other half minor. There are severalreasons why infection could be a problem in but-tocks augmentation by fat grafting: (1) The inci-sions are within 6 inches of the anus. (2) Eachstage of harvesting, preparing, and grafting the fathas a potential for contamination. (3) Until vascu-lar ingrowth occurs (at least 4 to 7 days), thegrafted fat is an ideal culture medium, placed ina warm, moist, traumatized environment. It seemslikely that a very small inoculum of bacteria couldcause an infection under these conditions. (4) Fur-ther, the skin of the injection site is traumatized byfriction, making it less effective as a bacterialbarrier.

The causative organisms of the infections in ourentire series are shown in Table 5.

The following routines were progressively insti-tuted, which resulted in a decrease from our ini-tial infection rate of almost 14% to our presentrate of 4% in our last 100 cases.

Bruner et al458

Table 4: Complications of buttocks augmentation by autologous fat grafting

MajorComplications

Totalseries,n 5 261, %

Currenttechnique(last 100 cases), %

Last 100 cases,excluding>1000 cc graft,n 5 74, %

InfectionMajorMinor

Donor site fluid accumulationrequiring aspiration totaling >100 cc

9 0* 0*

Donor site lschemia1 diabetic patient 0.4 1 0

Pulmonary embolus1 (late-30 days) 0.4 1 0

Aspiration pneumonitis: 1 0.4 0 0DIC (disseminated intravascularcoagulation)

0.4 0 0

Sterile abscess (fat necrosis) 0.4 1 0Hematoma 0 0 0Chronic seroma 0 0 0Gluteal muscle dysfunction 0 0 0Total major Complication 17.6 7 2.7

Minor complicationsDonor site fluid accumulation(total aspiration <100 cc)

10 2* 1.4

Minor liquified fat drainage 4 4 4.1Transient sciatic paresthesias

1–2 weeks 4 2.71–3 weeks

Minor donor site cellulitis 2 3 4.1Fibrotic nodules in buttocks 0 0 0Total minor complications 20.8 13 12.2

* Data for our last 50 patients, in which we used two drains. In this group, fluid accumulation decreased dramatically.

� Systemic antibiotics: Based on the sensitiv-ities of the causative organisms and clinicalresponse in our patients, we administer in-travenously at the beginning of surgery am-picillin 2 gm and sulbactam 1 gm (Unasyn3.1 mg) q6h, gentamicin (Garamycin) 5mg/kg/24 hours, and cefazolin (Ancef) 2gm IV q4h. Postoperatively, we use ampicil-lin and clavulanic acid (Augmentin) 875mg q12h and gatifloxin (Tequin) 400 mgq24h for 5 days, since we feel there is novascular ingrowth until at least this time.

� No shaving—just clipping of pubic hair.

� Circumferential standing prep with povidone/iodine.

� Lap pad soaked with povidone/iodine placedin gluteal cleft at beginning of surgery, evenwhen supine, to minimize wicking of tumes-cent fluid to and from the anal area.

� Minimal handling of fat. No washing is per-formed. After centrifugation, free oil is dec-anted, the aqueous layer drained off, and the

fat is transferred through a closed system into3-cc syringes for grafting.

� The grafting cannula is coated with povidone/iodine as each syringe is handed to thesurgeon.

� The single most important change we madewas to add the same three antibiotics to eachcanister of fat immediately upon harvesting.For each canister (roughly 200 cc aspirate),we add 10 cc of a solution containing ampicil-lin 2 g and sulbactam 1 gm plus 80 mg of gen-tamycin plus 2 gm cefazolin in 1 L of saline.This permits the fat particles to be bathed for1 to 2 hours in a concentration of antibioticsthree times higher than can be givensystemically.

Fluid accumulation in the fat donor sitesThis is especially problematic in the lumbar and sa-cral areas, since fairly aggressive sculpting of thisarea is necessary to obtain the ideal feminine in-ward sweep of the low back, and to create an

Complications of Buttocks Augmentation 459

attractive superior gluteal cleavage. Further, a rela-tively large surface area must be suctioned if enoughviable fat is to be obtained (total 1000 to 2000 cc fatfrom a total aspirate of typically 5 to 8 L).

Even when as little as 100 to 240 cc of fat wereharvested and grafted, Cardenas-Camarena et al[13] experienced a 6% seroma rate, and Peren etal [12] a 10% rate. Like Murillo [14], who graftedan average of 700 cc/buttock with a 40% seromarate, our initial rate of fluid accumulation requiringaspiration was 45% before implementing the use ofdrains, and the average aspiration total in the post-operative course was 415 cc, including both abdom-inal and sacral fluid accumulation. We initiallyhesitated to place additional compression exter-nally on the sacral area, or to put drains in this lo-cation, fearing compromise of the circulation ofthe skin over the bony prominences. However, be-cause of the frequency of significant fluid accumula-tion, we finally began placing one round 3-mmmultiperforated silicone drain across the sacralarea, exiting under the right bra strap, with some re-sulting decrease in fluid accumulation (Fig. 7). Forthe last 50 cases, we have added a second drain,from the left lower abdominal area, exiting underthe left bra strap. A long (40 cm) curved 3.5-mm li-posuction cannula is passed downward along thesepaths and out through an existing incision. Thedrain end is forced securely over the tip of the can-nula, and the cannula withdrawn, leaving the drainin place.

In addition, a soft, 2- to 3-inch-thick triangularpad made from Kerlix gauze is placed over the sacral‘‘V’’ to promote skin adhesion in this concave area,prevent seroma, and maintain the attractive supe-rior gluteal cleavage. Without this pad, the garment

Table 5: Causative organisms: infections inbuttocks augmentation by micro fat grafting(261 patients)

18 Total infections (8 minor, 10 major)

5/261 Gram negative, most commonly:

� E coli

� Bacteroides fragilis

� Microaerophilic strep� Pseudomonas

� Enterococcus

� Non-gas-forming clostridum1/261 Peptostreptococcus1/261 Staph aureus1/261 Mycobacterium fortuitum chelonei

(slow growing)6/261 Unknown (1 major, 5 minor)4/261 Minor cellulites (donor site)

cannot compress this area, edema and later fibrosisfill the area, and the feminine cleavage is lost. Withthis combination of two drains and compressivepadding, our seroma rate has dropped to 2% inthe last 50 cases, with a single patient requiringa one-time aspiration of 55 cc. We have seen no skinischemia or compromise from the padding or thedrains.

Transient sciatic paresthesiasThese tend to be minor, with discomfort, tingling,or slight numbness along the course of the sciaticnerve, and usually last less than 2 weeks. Mendieta[31] reported a frequency of less than 1% in his au-tologous fat-grafting patients, and Restrepo andAhmed [11] mention paresthesias but do not quan-tify them. Our incidence is 4% in our series of 261patients. We have had two patients who experi-enced a transient decrease in sciatic motor function;in these patients, MRI confirmed no hematoma orvisible injury to the sciatic nerve. Their motor symp-toms resolved in 1 to 2 weeks, and their paresthe-sias resolved between 1 and 3 months. Thegrafting cannulae are small (2 mm) and completelyblunt, so no nerve laceration is possible. Etiologyprobably includes physical contact of the cannulaplus swelling and inflammation. We now routinelygive a bolus of 12 mg of dexamethasone (Deca-dron) at the beginning of the surgical procedure.Gabapentin (Neurontin) and anti-inflammatorymedications may be useful if the symptoms areannoying.

Minor drainage of liquified fat from recipientincisionsThis is typically a few drops per day to 1 to 3 cc perday. Despite our efforts, there is some frictionaldamage to the small (2 to 4 mm) recipient inci-sions. We are attempting to develop a small portprotector to minimize this. We use up to five ofthese small incisions per side, to minimize fric-tional damage to each incision. The trauma to these

Fig. 7. Patients requiring aspiration of fluidaccumulation.

Bruner et al460

incisions may not become evident until sutureshave dissolved or are removed by postoperativeday 10 to 14. We see a small amount of drainageabout 4% of the time, although it is often so trivialthat it may be underreported by our patients. Thisdrainage may continue until secondary healing ofthe incision, which may require up to 6 to 8 weeks.Liquified fat can look like pus, so when this drain-age appears, always look for redness and new ten-derness and obtain a culture, stat gram stain,white blood count, and differential. If these are allnegative, close observation of the patient is war-ranted. When these evaluations continue to be nor-mal, we have never seen this typical small openwound become infected, except for one patientwho obsessively and aggressively massaged the areafrequently, trying to force out any fluid. At 5 weeks,she developed an unusual peptostreptococcal infec-tion, which presented as a draining flat cavity about4 inches in diameter, but was completely nontenderwith a normal white blood count and differential.Instead of widely opening the wound, which wouldhave caused great deformity, we managed her byenlarging the draining incision to 2 cm and thenperforming vigorous high-vacuum suction debride-ment of the pocket with a Yankauer (tonsil) suctiontip. A ‘‘pigtail’’ catheter was inserted from above per-cutaneously to permit through-and-through irriga-tion, and the incision below was kept open withan iodoform gauze wick until the pocket healed(Fig. 8). These pigtail catheters will stay patent forweeks, and have been a major breakthrough inthe management of intra-abdominal and other ab-scesses. This patient clearly lost volume on this sideand will need to be regrafted later.

Sterile abscess (fat necrosis)One patient with 1008 cc fat grafted/buttock hadoily drainage of less than 1 cc/day beginning at 2weeks. When we were finally able to get her to

Fig. 8. Closed management of draining cavity of rightbuttock with pigtail catheter and counter drainage(see text).

return, we found her with no infection but a wide,thick-walled cavity from which we suctioned 200cc of nonviable fat. She was managed like the pa-tient in the preceding section with suction debride-ment and irrigation/drainage with a pigtail catheter.

Minor donor site cellulitisThis occurs in 2% to 3% of the cases and respondsquickly to antibiotics.

Other complicationsWe had one case of unexplained disseminated in-travascular coagulopathy, which was treated appro-priately, and the patient had an uneventful fullrecovery after a 3-day hospitalization. One patientdeveloped a small pulmonary embolus 1 month af-ter surgery and recovered quickly. One patient withdiabetes had an unusual complication of liposuc-tion with ischemia of the lower thoracic area(Fig. 9A, B). In an informal poll of plastic surgeonsduring two instructional courses, approximately 3%of plastic surgeons report having seen this complica-tion of liposuction, especially in diabetic patients.After we encountered this problem, Dr V. LeroyYoung (personal communication, 2005) recom-mended considering the topical application of di-methyl sulfoxide (DMSO) 75% as an arterial andvenous vasodilator (an off-label use of this product).Fig. 10A shows our only patient whose right buttock,10 minutes after completing grafting of 987 cc, sud-denly turned white centrally. After observing this un-changed for 10 minutes, dimethyl sulfoxide 75%was applied topically. Within 10 minutes, the circu-lation dramatically improved (Fig. 10B), and by 20minutes, was back to normal (Fig. 10C). Althoughthese events might have been coincidental and dueto the opening of the choke vessels, unpublishedclinical experience suggests DMSO may be helpful,and we offer this observation for what it is worth. Re-gardless, all patients should be informed of any off-label use of a product.

Symptomatic hypovolemiaThis may be present on the first postoperative dayand usually responds to intravenous fluid adminis-tration. We always check hemoglobin and hemato-crit the day after surgery and again on day 5. Twopercent of patients have required blood transfu-sions, attributable to extensive liposuction ora low hemoglobin preoperatively. Most of thesewere early in our series when we used only the‘‘wet technique’’—0.5 cc tumescent fluid per 1 ccof anticipated fat removal. We found a decreased in-cidence of drop in hemoglobin postoperativelywhen a full 2 to 4 cc of tumescent fluid is used foreach 1 cc of anticipated fat removal.

Complications of Buttocks Augmentation 461

Fig. 9. (A) Ischemia of lipo-suction donor site on thelower posterior chest ina diabetic patient, postopday 4. (B) Same diabetic pa-tient, 1.5 weeks postop.

Minor metabolic derangementsIn addition to checking hemoglobin and hemato-crit on the first and fifth postoperative days, we al-ways check serum electrolytes and calcium levels.We find typically the potassium is slightly low onday one (3.0-3.2 mmol/l), and the calcium alsoslightly low (9.0-9.5 mg/dl). These have never re-quired intravenous replacement. We do ask patientsto drink a gallon of Gatoradetm or other sports elec-trolyte drink each day for a week to avoid dehydra-tion and replenish electrolytes.

Partial reabsorption of grafted fatPartial reabsorption is not truly a complication ofautologous fat grafting, but an expected outcome.The only accurate way of measuring reabsorptionwould be serial MRIs. Although this has once beenattempted with fat grafting to the buttocks [14], on-ly six patients were evaluated. There was clear radio-logic evidence of persistence of the grafted fat, butthe small number of cases prevented any quantifica-tion of fat survival. Unfortunately, the cost of serialMRIs precludes their routine use. It is our clinical

Fig. 10. (A) Potentially ischemic area of central buttock immediately after autologous fat grafting. (B) Same pa-tient, after 10 minutes of topical DMSO. (C) Same patient, after 20 minutes of topical DMSO (see text).

Bruner et al462

estimate, based on our 261 cases, that less than 20%to 40% of the grafted fat is resorbed (Fig. 11A, B, C).This is similar to the estimates in the two other largeseries of large-volume (> 400 cc) buttocks grafting[11,14].

Although the optimal method of maximizingsurvival of autologous fat grafting is still a matter re-quiring further scientific investigation, our clinicalexperience and that of others suggests that the fol-lowing techniques provide the best survival.

� Use low vacuum of aspiration: The vacuumgenerated by commercial liposuction pumpsand by syringe aspiration is typically 25 to 28inHg. At these vacuum levels, we observed thatgas appears to bubble out of the fat tissue in thecanister or syringe. Although at extremely highvacuums (> 30 inHg) water may boil at roomtemperature, this cannot be the explanationat 25 to 28 inHg. The most likely explanationis the ‘‘gassing out’’ phenomenon in solutions,in which dissolved oxygen and nitrogen comeout of the solution. If this is occurring in extra-cellular water in our containers, the assump-tion can be made that it is occurring in theintracellular fluid, which could result in dis-ruption of intracellular organelles or the cellu-lar membrane, thereby decreasing the viabilityof the fat cells. Niechajev and Sevcuk [35] re-ported that microscopic evaluation of fat cellsharvested at high vacuum (28 inHg) revealedthat these adipocytes were 41% larger thanthose harvested at lower vacuum (15 inHg).We observed that this bubbling ceases whenthe vacuum is less than 22 inHg, which is com-patible with the concept of ‘‘gassing out.’’ Forthis reason, we maintain our vacuum below22 inHg. Aspiration of fat at low vacuum was

also recommended by Coleman [24–26] andothers (Murillo [14], Pedroza [22], Chajchirand colleagues [20]), although no reason wasoffered for this suggestion.

� Use small harvesting cannula: Several authors(Carpaneda and Ribeiro [36], Coleman [24–26]) have found that graft survival is highestwhen the fat particle size is less than 3 mm. Co-leman points out that anything larger than thiswill not pass easily through a Luer-lock tipwithout being disrupted. Since almost all au-thors’ techniques involve injection througha Luer-lock tipped syringe, we recommend us-ing a harvesting cannula that produces a 3-mm or smaller particle. Our choice is the‘‘Keel’’-shaped 3.0- or 3.5-mm harvesting can-nula, which is faster and less traumatic thanthe original ‘‘bucket handle’’ tip designed byColeman, yet produces particles less than 3 mmin diameter.

� Graft only tiny amounts (< 0.3 cc) in each tun-nel: This is one of Coleman’s and Guerrosantos’major recommendations [10,12,24–26,37] toensure that each particle of fat is surroundedby a blood supply. This requires extensive timefor grafting, as over 2000 tunnels per side arethus required to graft 600 cc per buttock.

� Graft the fat in all layers of the buttocks, bothintramuscularly and subcutaneously: Toomuch fat deposited in any single area meansless contact with recipient blood supply.

� Minimize shear and compression on the but-tocks postoperatively: While authors differ inhow long before patients may return to full ac-tivity, we feel that any shear or compressioncould damage ingrowing capillaries and resultin greater reabsorption. Although extensive

Fig. 11. (A) Forty-seven-year-old Caribbean woman requesting youthful and feminine shape. (B) Same patient, 3months after liposuction of the hips and lumbosacral area, and buttocks augmentation by autologous fat graft-ing (939 cc per buttock). (C) Same patient, 1 year and 3 months after autologous fat grafting. There appears tobe very little reabsorption.

Table 6: Factors influencing c ast 100 cases)

Volume of fat graftedper buttock

Volume of fat remove

>30 <1000 cc >1000 cc <2.5 L >2.5 L

Major complications(infection, pulmonaryembolus, donor site ischemia,donor site fluidaccumulation >100 cc)

0% (0/12) 2.7% (2/74) 19.2% (5/26) 4.9% (4/81) 15.8% (3/19)

Minor complications ) 8.3% (1/12) 12.2% (9/74) 15.3% (4/26) 11.1% (9/81) 21% (4/19)Total complications ) 8.3% (1/12) 14.9% (11/74) 34.6% (9/26) 16% (13/81) 36.8% (7/19)

Co

mp

licatio

ns

of

Bu

ttock

sA

ug

men

tatio

n463

omplication rate in autologous fat grafting (l

BMI

Last 100 cases <25 25–30

7% 2.9% (1/3) 11.1% (6/54)

13% 5.9% (2/34) 18.5% (10/5420% 8.8% (3/34) 29.6% (16/54

Bruner et al464

ambulation is encouraged from the day of sur-gery onward, we ask our patients not to lie su-pine or to sit (except for a bowel movement)for 2 weeks. This may be overly conservative,but until more information is available, wetry to protect the fat as long as possible.

Factors affecting the incidence ofcomplications in augmentation byautologous fat grafting

Our operative and management techniques havebeen stable for the last 100 patients (with the excep-tion of using two drains in the last 50 patients).Therefore, we used this group to analyze riskfactors.

Preoperative body mass index (BMI)

Contrary to most trends in surgical complications,there does not appear to be any correlation betweenincidence of complications (both major and mi-nor) and increasing BMI. An analysis of BMI dataand rate of complications was performed for thelast 100 cases (Table 6). There appears to bea bell-shaped curve distribution of percent compli-cation rate according to BMI, with a peak incidenceof 11% major complications for BMI between 25and 30. There were 3% major complications associ-ated with patients whose BMI was less than 25 andnone in patients whose BMI was greater than 30;minor complications followed a similar pattern.

Volume of fat grafted

Twenty-six of the last 100 patients had more than1000 cc/buttock grafted, and these were responsiblefor five of the seven major complications (infection,pulmonary embolus, donor site ischemia, sterileabscess) in these last 100 patients. This is not sur-prising, since these patients had both the most ex-tensive liposuction and the longest operativeprocedures, both of which are known to increasemorbidity. Further, as the volume grafted increasesso does the trauma to the recipient site, the fric-tional injury to the skin, and the turgor of the but-tock (which could decrease the perfusion to thetissues). These factors resulted in a major complica-tion rate of 19.2% for patients receiving more than1000 cc graft/buttock. There were eight patients re-ceiving more than 1200 cc graft/buttock, and in thisgroup, major complications rose to 25%. By con-trast, only 2 out of 74 patients with less than1000 cc/buttock (2.7%) had a major complication.

In addition to the 19.2% major complications,the patients receiving more than 1000 cc/buttockhad 15.3% minor complications; those with lessthan 1000 cc/buttock had a similar rate of 12.2%minor complications.

Amount of fat aspirated

With liposuction alone, complications increase asthe amount of fat aspiration increases, especiallyover 5 L (‘‘high volume liposuction’’). In our last100 cases, we had 19 patients with more than 2.5L of fat aspirated; these 19 had three major compli-cations, or 15.8%. The 81 patients from which lessthan 2.5 L of fat was aspirated had four major com-plications, or 4.9%. Liposuction of 2.5 L of fat is notenough to account for a 15.8% major complicationrate. We feel that this is ‘‘guilt by association,’’ withthe real etiology being the attempt to augmentmore than 1000 cc/buttock (total > 2000 cc graft).A 2000-cc graft mandates harvesting more than2.5 L, but it is probably the trauma associated withthe grafting this much that is at fault, rather thanthe harvesting.

Summary

Subcutaneous implant augmentation

This should never be done for cosmetic purposesbecause of the very high incidence of implant mo-bility, displacement, ptosis, capsular contracture,and patient dissatisfaction.

Intramuscular implant augmentation

This has an overall wound dehiscence rate of up to30%, increasing to 80% in overweight patients andis also very high in patients in whom an implant ofmore than 350 cc or more than 3.5 cm projection isused. Seroma, infection, implant malposition, andloss of implant due to muscle dehiscence and im-plant exposure each occur in 2% to 5% of cases.This brings the major complication rate to roughly15% to 25%.

Subfascial implant augmentation

This has a similar dehiscence rate of 15% to 30%.Unfortunately, with the textured semisolid implantavailable in the United States, the incidence of sero-ma is about 20%; this may possibly be decreased byusing a smooth implant. Other problems bring themajor complication rate to 35%, and total compli-cation rate to 51%. In addition, 100% of these im-plants are easily palpable by the patient and his orher partner.

The cohesive gel implant for subfascial buttocksaugmentation hopefully will become available inthe United States at some point in the future.De la Pena’s experience with this implant is muchmore favorable [30]. These implants are less palpa-ble, although the dehiscence rate remains approxi-mately 30%.

Complications of Buttocks Augmentation 465

Buttocks augmentation by autologousfat grafting

This may be the safer alternative with major compli-cations at 7% with current techniques, but requiresa major investment of time and is labor-intensive.Major complications may be reduced to less than3% by grafting less than 1000 cc per buttock (Table4, column 4).

The good news is twofold: (1) the techniques andpatient selection indicated in this paper can signif-icantly decrease complications in all of these proce-dures, and (2) discussion of complications is nowmore open, and thanks to the ongoing efforts ofMendieta, de la Pena, Young, Centeno, Roberts,and others, we are continuing to develop new tech-niques and implants and selection criteria thatpromise to decrease these problems even further.Much is yet to be done.

Acknowledgments

Our thanks to Dr Shlomo Widder of Vienna, Virgin-ia, for collaborating on his experience with subfas-cial implant augmentation, and to Tina Ridgeway,RN, BSN, our patient coordinator for buttocks sur-gery, for her help in maintaining accurate data andfor countless hours of compassionate patient care.

References

[1] Roberts TL, Toledo LS, Badin AZ. Augmentationof the buttocks by micro fat grafting. AestheticSurg J 2001;21(4):311–9.

[2] Douglas WM, Bartels RJ, Baker JL. An experiencein aesthetic buttocks augmentation. Clin PlastSurg 1975;3(3):471–6.

[3] Cronin TD, Gerow FT. Augmentation mamma-plasty: a new ‘‘natural feel’’ prosthesis. In:Broadbent TR, editor. Transactions of the ThirdInternational Congress of Plastic Surgery. Am-sterdam: Amsterdam: Excerpta Medica Founda-tion; 1964. p. 41–9.

[4] Bartels RJ, O’Malley JE, Douglas WM, et al. Anunusual use of the cronin breast prosthesis. PlastReconstr Surg 1969;44(5):500.

[5] Cocke WM, Ricketson G. Gluteal augmentation.Plast Reconstr Surg 1973;52(1):93.

[6] Mendieta CG. Gluteoplasty. Aesthetic Surg J2003;23(6):441–55.

[7] Billings E, May JW. Historical review and presentstatus of free fat graft autotransplantation inplastic and reconstructive surgery. Plast ReconstrSurg 1989;83(2):368–81.

[8] Illouz YG. Body contouring by lipolysis: a 5-yearexperience with over 3000 cases. Plast ReconstrSurg 1983;72(5):591–7.

[9] Lewis CM. Correction of deep gluteal depressionby autologous fat grafting. Aesthetic Plast Surg1992;16:247–50.

[10] Guerrerosantos J. autologous fat grafting for bodycontouring. Clin Plast Surg 1996;23:619–31.

[11] Restrepo JC, Ahmed JA. Large-volume lipoinjec-tion for gluteal augmentation. Aesthetic Surg J2002;22(1):33–8.

[12] Peren PA, Gomez JB, Guerrerosantos J, et al. Glu-teus augmentation with fat grafting. AestheticPlast Surg 2000;24:412–7.

[13] Cardenas-Camarena L, Lacouture AM,Tobar-Losada A. Combined gluteoplasty: lipo-suction and lipoinjection. Plast Reconstr Surg1999;140(5):1524–31.

[14] Murillo WL. Buttock augmentation: case studies offat injection monitored by magnetic resonance im-aging. Plast Reconstr Surg 2004;114(6):1606–14.

[15] Vergara R, Amezcua H. Intramuscular glutealimplants: 15 years’ experience. Aesthetic Surg J2003;23(2):86–91.

[16] Robles JM, Tagliapietra JC, Grandi MA. Gluteo-plastia De Aumento: Implante Submuscular.Cirugia Plast Ibero Latinoamericana 2984;10:365–9

[17] Gonzalez-Ulloa M. Gluteoplasty: a ten-year re-port. Aesthetic Plast Surg 1991;15:85–91.

[18] Vergara R, Marcos M. Intramuscular gluteal im-plants. Aesthetic Plast Surg 1996;20:259–63.

[19] Gonzalez R. Augmentation gluteoplasty: the XYZmethod. Aesthetic Plast Surg 2004;28:417–25.

[20] Chajchir A, Benzaquen I, Wexler E, et al. Fat in-jection. Aesthetic Plast Surg 1990;14:127–36.

[21] Toledo L. Syring liposculpture: a two-year experi-ence. Aesthetic Plast Surg 1991;15:321–6.

[22] Pedroza LV. Fat transplantation to the buttocksand legs for aesthetic enhancement or correc-tion of deformities: long-term results of largevolumes of fat transplant. Dermatol Surg2000;26(12):1145–9.

[23] Monreal J. Fat tissue as a permanent implant:new instruments and refinements. Aesthetic SurgJ 2003;23(3):213–6.

[24] Coleman SR. Long-term survival of fat trans-plants: controlled demonstrations. AestheticPlast Surg 1995;19:421–5.

[25] Coleman SR. Facial recontouring with lipostruc-ture. Clin Plast Surg 1997;24(2):347–67.

[26] Coleman SR. Structural fat grafting. AestheticSurg J 1998;18(5):386–8.

[27] Roberts TL, De La Pena JA, Toledo LS, et al. Paneldiscussion: cosmetic surgery of the buttocks re-gion. Aesthetic Surg J 2003;23(5):381–7.

[28] Buchuk L. Complication with gluteal prosthesis.Plast Reconstr Surg 1986;77:1012.

[29] Ford RD, Simpson WD. Massive extravasation oftraumatically ruptured buttock silicone prosthe-sis. Ann Plast Surg 1992;29(1):86–8.

[30] De La Pena JA. Subfascial technique for glutealaugmentation. Aesthetic Surg J 2004;24(3):265–73.

[31] Mendieta C. Intramuscular buttocks implant aug-mentation. Presented at The Annual Meeting ofthe American Society of Plastic Surgery. Chicago,September 27, 2005.

Bruner et al466

[32] De La Pena A. Subfascial buttocks implant aug-mentation. Presented at The Annual Meeting ofthe American Society of Plastic Surgery. Chicago,September 27, 2005.

[33] Vergara R. Intramuscular buttocks implants. Pre-sented at The Annual Meeting of the AmericanSociety of Plastic Surgery. Chicago, September27, 2005.

[34] Roberts T. Complications of buttocks augmenta-tion. Presented at The Annual Meeting of the

American Society of Plastic Surgery. Chicago,September 27, 2005.

[35] Niechajev I, Sevcuk O. Long-term results of fattransplantation: clinical and histologic studies.Plast Reconstr Surg 1994;94:496–506.

[36] Carpaneda CA, Ribeiro MT. Percentage of graftviability versus injected volume in adipose auto-transplants. Aesthetic Plast Surg 1994;18:17–9.

[37] Guerrerosantos J. Microlipofilling. Aesthetic SurgJ 2002;22(2):196–8.