FREE TRANSVERSE RECTUS ABDOMINIS MYOCUTANEOUS FLAPRECONSTRUCTION OF A MASSIVE LUMBOSACRAL DEFECTUSING SUPERIOR GLUTEAL ARTERY PERFORATOR VESSELS

RICHARD S. GASTER, Ph.D.,1 KIRIT A. BHATT, M.D.,1 ANDREW A. SHELTON, M.D.,2 and GORDON K. LEE, M.D.1*

Despite significant advances in reconstructive surgery, the repair of massive lumbosacral defects poses significant challenges. When theextent of soft tissue loss, tumor resection, and/or radiation therapy preclude the use of traditional local options, such as gluteal advance-ment flaps or pedicled thigh flaps, then distant flaps are required. We report a case of a 64-year-old male who presented with a large sac-ral Marjolin’s ulcer secondary to recurrent pilonidal cysts and ulcerations. The patient underwent wide local composite resection, whichresulted in a wound measuring 450 cm2 with exposed rectum and sacrum. The massive defect was successfully covered with a free trans-verse rectus abdominis myocutaneous flap, providing a well-vascularized skin paddle and obviating the need for a latissimus flap with skingraft. The free-TRAM flap proved to be a very robust flap in this situation and would be one of our flaps of choice for similar defects.VVC 2012 Wiley Periodicals, Inc. Microsurgery 00:000–000, 2012.

In comparison with other regions of the trunk, defects

located in the lumbosacral region sometimes present diffi-

culties in reconstruction, arising from the unique architec-

ture of the surrounding tissue and the width of the

defects.1 Attempts to perform local or regional flaps for

covering defects of this area are prone to partial flap

necrosis, with subsequent risk of infection and tenuous

wounds. Distant flaps may be the only solution when tra-

ditional local options, such as gluteal advancement flaps,

cannot be used due to the extent of soft tissue loss.

Alternative reconstructive options that have been

described in the literature include tensor fascia lata (TFL)

or posterior thigh flaps.2,3 However, these flaps may be

inadequate for massive defects where more bulk and

more skin are needed. In addition, when considering free

flaps, the lack of recipient vessels in the area of the low

back/buttock earns it the name ‘‘no-man’s land’’ by

microsurgeons. In this report, we present a solution that

meets these challenges to reconstruct a massive defect in

the lumbosacral region. We were able to successfully

reconstruct this region using a free transverse rectus ab-

dominis myocutaneous (TRAM) flap and the superior glu-

teal artery perforators (SGAPs) as recipient vessels.

CASE REPORT

A 64-year-old male presented with a large sacral Mar-

jolin’s ulcer secondary to recurrent pilonidal cysts and

ulcerations that were treated in 2006 and 2007, at an out-

side hospital. During that time, the patient also underwent

brachytherapy for prostate cancer. The patient’s addi-

tional significant medical history included left hip hemiar-

throplasty and left total knee reconstruction, rendering

him dependent on a walker and upper body strength for

transfers and mobility. The patient presented to our hos-

pital in 2008, with a large Marjolin’s ulcer in the area of

the recurrent pilonidal cysts requiring resection and

reconstruction.

One week prior to reconstruction, the oncologic sur-

geons performed a radical excision of squamous cell car-

cinoma and partial sacrectomy, which resulted in a large

wound measuring 450 cm2 with exposed rectum (Fig.

1). The tumor resection was performed several days

prior to definitive reconstruction to verify that negative

margins were obtained. Intraoperative radiation therapy

was given due to concerns for tumor microfoci in the

perirectal/sacral region. The depth and span of the tissue

loss, as well as the exposed rectum and exposed portion

of the sacrum necessitated musculocutaneous flap recon-

struction. Complicating the patient’s management was

the prior history of brachytherapy for prostate cancer as

well as a diverting abdominal colostomy (that had been

placed specifically to divert fecal flow through his

anticipated sacral wound), and intraoperative radiother-

apy. Given this set of circumstances, we elected to

proceed with a free-TRAM flap, which provided a large

amount of vascularized soft tissue to cover this expan-

sive defect.

The reconstruction began with the patient initially

placed in the prone position. The wound measured 18

3 25 cm2 with exposed sacrum, rectum, fat, muscle,

and perianal tissue. The resection extended down to the

perianal musculature, but the anus itself was spared.

The possibility of performing a local or regional flap

reconstruction was considered but was deemed tenuous

1Division of Plastic Surgery, Stanford University Medical Center, Stanford,CA2Section of Colon and Rectal Surgery, Department of Surgery, Stanford Uni-versity Medical Center, Stanford, CA

*Correspondence to: Gordon K. Lee, MD, Assistant Professor, Director ofMicrosurgery, Division of Plastic and Reconstructive Surgery, 770 WelchRoad, Suite 400, Stanford, CA 94304-5715. E-mail: glee@stanford.edu

Received 16 October 2011; Revision accepted 5 February 2012; Accepted13 February 2012

Published online in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/micr.21981

VVC 2012 Wiley Periodicals, Inc.

and inadequate given prior radiation therapy to the

region.

We explored the SGAP vessels to determine their

suitability as recipient vessels for microvascular free tis-

sue transfer. This was done by undermining the skin and

evaluating where the SGAP emerges, which was located

at one-third of the distance between the posterior superior

iliac spine (PSIS) and the greater trochanter (Fig. 2). The

perforators were dissected through the muscle for �5 mm

such that the vessel size was larger and a better match to

the donor vessels. On the left side, the SGAP was �1.5

mm in diameter, and the vein was �3 mm in diameter,

sufficient for microvascular anastomoses. We then reposi-

tioned the patient in the supine position to harvest the

TRAM flap.

Markings had been made preoperatively for a right-

sided TRAM flap in the standard fashion and took into

account the end-colostomy in the left lower quadrant,

which was purposely placed in the line of the upper

TRAM flap incision. The colostomy was released from

the skin incision. We then harvested the TRAM flap in a

standard fashion. To maintain maximal perfusion to the

tissue, we elected to harvest the entire rectus muscle and

all associated perforators. The flap was placed into a bag

and then into a sterile ice slush to induce a relative cold

ischemia. The total cold ischemia time until reperfusion

was �4 hours. After dissection and harvest of the flap,

the abdomen was closed in standard fashion. The colos-

tomy was transposed and matured into a new opening

through the abdominal skin. After the wounds were

Figure 1. Radical excision and sacrectomy for squamous cell carcinoma. Defect measured 18 cm in length and 25 cm wide. [Color figure

can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

2 Gaster et al.

Microsurgery DOI 10.1002/micr

closed, dressings were applied and an appliance was

placed over the colostomy in the left lower quadrant.

Following harvesting of the TRAM flap, the patient

was repositioned back into the prone position. We per-

formed a standard microvascular anastomosis to the

TRAM flap using the SGAP vessels. The deep inferior

epigastric artery of the TRAM flap was �2.5 mm,

slightly larger than the SGAP, so the SGAP was spatu-

lated to account for the size mismatch; 9-0 nylon simple

interrupted sutures were used. The venous anastomosis

was performed with a 2.5 mm venous coupler (Synovis

Micro Companies Alliance, Birmingham, AL). Zone IV

of the TRAM flap was discarded prior to inset, until

healthy bleeding tissue was observed. The flap was then

sutured into place with 2-0 vicryl sutures for the superfi-

cial fascial layer, 3-0 monocryl for the deep dermis, and

3-0 nylon for the skin. Two 15-French Blake drains

were placed subcutaneously for postoperative drainage

(Fig. 3).

Postoperatively, the patient remained in either the

prone or lateral decubitus position for 1 week. During the

second week, he was allowed to stand with assistance uti-

lizing time limitations similar to the lower extremity free-

flap ‘‘dangling’’ protocol, to allow the flap to be in a

dependent position and allow for progressive activity.4

The patient had only a minor area of delayed wound

healing in the distal portion of the TRAM flap and over

a small area of the abdominal closure along the trans-

verse incision, which healed completely with local wound

care. Otherwise, there were no major complications and

the flap was able to subsequently tolerate 15 cycles of

postoperative external beam radiotherapy without further

sequelae. The patient was back to his baseline activities

and ambulation 5 months later (Fig. 4). Postoperative

motor strength (manual muscle test, 5/5 scale) improved

after the reconstruction since pain from chronic wounds

no longer limited the patient’s activities. On gross exami-

nation, neurovascular exam in his distal extremity was

unchanged from preoperative findings and was grossly

normal.

DISCUSSION

The current report highlights the use of the free-

TRAM flap and SGAP vessels to reconstruct an irradiated

and massive sacral defect. In our case, the patient suf-

fered from a Marjolin’s ulcer, which has been well

described as a squamous cell carcinoma arising from a

chronically inflamed, nonhealing wound or scar.5 Our

patient had both a history of chronic pilonidal cysts and

radiation treatment for prostate cancer. Marjolin’s ulcer is

an indolent and insidious condition with an average time

to malignant transformation of 30–35 years.6,7 Often, the

tumor is aggressive in nature, with a worse prognosis

when compared to other squamous cell carcinomas. As a

result, both wide excision of the lesion and postoperative

radiotherapy are standard to maximize disease-free sur-

vival.8

There are several options traditionally used for

reconstruction of the lumbosacral region, which include

local tissue flaps from the buttocks or regional pedicled

flaps from the thighs in addition to the more recently

reported perforator-preserving gluteal artery-based rota-

tion fasciocutaneous flaps.9,10 The resection of gluteal

tissue, however, precluded us from using local flaps.

Neither regional flaps, such as tensor fascia lata (TFL)

or anterolateral thigh (ALT), nor posterior thigh flaps

would have provided complete coverage and would

have entailed rotating multiple flaps with added mor-

bidity.11

In our review of the literature, there are few reports

of free flaps to the sacral region. Park et al. reported the

use of the free latissimus dorsi flap for reconstructing

defects in this region and used the SGAP vessels as a

recipient blood supply.12,13 Although we had considered

using the latissimus, it would have required skin grafting

the muscle, which would not have provided our patient

durable coverage in this region of the body that is subject

to shear forces when sitting and ambulating. Furthermore,

the patient required additional postoperative external

beam radiation treatment, and a skin graft would have

Figure 2. Identification of SGAP using known landmarks: 1/3 the

distance from the posterior superior iliac spine (PSIS) to the greater

trochanter. [Color figure can be viewed in the online issue, which is

available at wileyonlinelibrary.com.]

Transverse Rectus Abdominis Myocutaneous Flap 3

Microsurgery DOI 10.1002/micr

been prone to further wound breakdown.14,15 As an addi-

tional consideration, a free latissimus flap would poten-

tially weaken the patient’s upper body strength, as

the primary contribution of the muscle is in shoulder

extension, adduction, and medial rotation.16,17 For this

patient, it was particularly important to maintain upper

body strength since he was dependent on a walker to

ambulate.

The pedicled vertical rectus abdominis myocutaneous

(VRAM) flap is another flap described for sacral wound

coverage and can be passed through the abdominal cav-

ity.18 However, with any pedicled flap there is a loss of

length as the flap is transposed to the defect. A pedicled

VRAM or pedicled TRAM was not likely to cover the

extent of this massive defect. Furthermore, the defect was

in the sacral region and was retro-/extraperitoneal, and a

pedicled flap would have required entering the abdominal

cavity and creating a passage to behind the rectum, which

we believe would have had increased morbidity when

compared to a free flap.

Traditionally, free-flap reconstruction in the sacral

area has been viewed as extremely challenging due to

limited access to recipient vessels in this area. However,

with advances in microvascular technique, we believe

that free-tissue transfer can be a valuable method for

addressing challenging cases such as the one presented.

The use of a free-TRAM flap provides significantly more

bulk than a skin paddle alone as it is a type III muscle

flap (by Mathes-Nahai criteria), providing a very robust

and stable blood supply to a flap that may be subject to

long periods of compression when the patient is seated.

In addition, the free-TRAM flap is sufficiently robust to

Figure 3. (A) Design of TRAM flap. (B) SGAP to deep inferior epigastric artery (DIEA) microvascular anastomosis. (C) Closure of

abdomen with transposed colostomy. (D) Flap on day 12 follow-up. [Color figure can be viewed in the online issue, which is available at

wileyonlinelibrary.com.]

4 Gaster et al.

Microsurgery DOI 10.1002/micr

withstand postoperative radiation, an important considera-

tion for this patient.19 Furthermore, the SGAP vessels are

familiar blood vessels for microsurgeons who perform

gluteal flaps for breast reconstruction. As a technical

note, placing the flap on ice facilitated the operation after

harvesting the flap.20 The relative hypothermia of the flap

allowed for the prolongation of the ischemia time, which

was beneficial while we closed the abdominal donor site

primarily and repositioned the patient from supine to

prone position.

The disadvantages of using the TRAM flap are ger-

mane to any TRAM flap, such as the risk of hernia, bulg-

ing, and seroma formation; none of which occurred in

our case. The other disadvantages include the reposition-

ing of the ostomy, and intraoperative repositioning of the

patient, both of which prolonged the operative time but

can be reasonably managed.

In our review of the literature, this is the first reported

case of successful reconstruction of a massive lumbosac-

ral defect using a free-TRAM flap to the SGAP vessels.

The use of the free TRAM provided a large vascularized

skin paddle that was able to cover the entire defect in a

single stage. The SGAP vessels were conveniently

located near the defect and were suitable for microvascu-

lar anastomosis. Although additional cases are required to

fully validate this technique, it should be considered

when reconstructive plastic surgeons are faced with mas-

sive defects in this region.

REFERENCES

1. Liu Y, Yu S, Song B, Yang L, Zhu S, Jin J. Reconstruction of poste-rior lumbar defects in oncologic patients using two island flaps ofthe back in series. Ann Plast Surg 2010;65:326–329.

2. Miles WK, Chang DW, Kroll SS, Miller MJ, Langstein HN, ReeceGP, Evans GR, Robb GL. Reconstruction of large sacral defects fol-lowing total sacrectomy. Plast Reconstr Surg 2000;105:2387–2394.

3. Paletta C, Bartell T, Shehadi S. Applications of the posterior thighflap. Ann Plast Surg 1993;30:41–47.

4. Rohde C, Howell BW, Buncke GM, Gurtner GC, Levin LS, Pu LL,Levine JP. A recommended protocol for the immediate postoperativecare of lower extremity free-flap reconstructions. J Reconstr Micro-surg 2009;25:15–19.

5. Fleming MD, Hunt JL, Purdue GF, Sandstad J. Marjolin’s ulcer: Areview and reevaluation of a difficult problem. J Burn Care Rehabil1990;11:460–469.

6. Konigova R, Rychterova V. Marjolin’s ulcer. Acta Chir Plast 2000;42:91–94.

7. Copcu E, Aktas A, Sisman N, Oztan Y. Thirty-one cases of Marjo-lin’s ulcer. Clin Exp Dermatol 2003;28:138–141.

8. Aydogdu E, Yildirim S, Akoz T. Is surgery an effective and adequatetreatment in advanced Marjolin’s ulcer? Burns 2005;31:421–431.

9. Hung S, Chen H, Wei F. Free flaps for reconstruction of the lowerback and sacral area. Microsurgery 2000;20:72–76.

10. Lin P-Y, Kuo Y-R, Tsai Y-T. A reusable perforator-preserving glu-teal artery-based rotation fasciocutaneous flap for pressure sorereconstruction. Microsurgery 2012 [Epub ahead of print].

11. Fasching MC, Meland NB, Woods JE, Wolff BG. Recurrent squamous-cell carcinoma arising in pilonidal sinus tract—Multiple flap reconstruc-tions. Report of a case. Dis. Colon Rectum 1989;32:153–158.

12. Park S, Koh KS. Superior gluteal vessel as recipient for free flap recon-struction of lumbosacral defect. Plast Reconstr Surg 1998;101:1842.

13. Park S. Muscle-splitting approach to superior and inferior gluteal ves-sels: Versatile source of recipient vessels for free-tissue transfer tosacral, gluteal, and ischial regions. Plast Reconstr Surg 2000;106:81.

14. Rudolph R. Complications of surgery for radiotherapy skin damage.Plast Reconstr Surg 1982;70:179–185.

15. Tadjalli HE, Evans GR, Gurlek A, Beller TC, Ang KK, StephensLC. Skin graft survival after external beam irradiation. PlastReconstr Surg 1999;103:1902–1908.

16. Spear SL, Hess CL. A review of the biomechanical and functionalchanges in the shoulder following transfer of the latissimus dorsimuscles. Plast Reconstr Surg 2005;115:2070–2073.

17. Fraulin FO, Louie G, Zorrilla L, Tilley W. Functional evaluation ofthe shoulder following latissimus dorsi muscle transfer. Ann PlastSurg 1995;35:349–355.

18. Glatt BS, Disa JJ, Mehrara BJ, Pusic AL, Boland P, Cordeiro PG.Reconstruction of extensive partial or total sacrectomy defects witha transabdominal vertical rectus abdominis myocutaneous flap. AnnPlast Surg 2006;56:526–531.

19. Tran NV, Evans GR, Kroll SS, Baldwin BJ, Miller MJ, Reece GP,Robb GL. Postoperative adjuvant irradiation: Effects on tranverserectus abdominis muscle flap breast reconstruction. Plast ReconstrSurg 2000;106:313–317; discussion318–320.

20. Lee DT, Lee G. Cold ischemia in microvascular breast reconstruc-tion. Microsurgery 2010;30:361–367.

Figure 4. Appearance of the flap at 5-month follow-up.

Transverse Rectus Abdominis Myocutaneous Flap 5

Microsurgery DOI 10.1002/micr