Vertical rectus abdominis musculocutaneous flow-through flap to a free fibula flap for total sacrectomy reconstruction

VERTICAL RECTUS ABDOMINIS MUSCULOCUTANEOUS FLOW-THROUGH FLAP TO A FREE FIBULA FLAP FOR TOTALSACRECTOMY RECONSTRUCTION

PATRICK B. GARVEY, M.D.,1* MARK W. CLEMENS, M.D.,1 LAURENCE D. RHINES, M.D.,2 and JUSTIN M. SACKS, M.D.3

Purpose: The purpose of this report of a small series was to describe the technique of total sacrectomy reconstruction using a pedicledvertical rectus abdominis musculocutaneous (VRAM) flow-through flap anastomosed to a free fibula flap. Methods: We reviewed all con-secutive total sacrectomy reconstructions performed from 2009 to 2011. Surgical technique and patient outcomes were assessed. Results:Total sacrectomy reconstructions included three two-stage and three-stage VRAM flow-through flap to free fibula flap patients all of whichambulated by discharge. Flap survival was 100%. Pelvic ring defects were reconstructed with A-frame fibula flap struts anastomosed tothe distal epigastric vessels of pedicled trans-pelvic VRAM flaps. Complications such as wound healing, infection or hardware failure werenot observed. Bony union occurred at an average 2.7 6 0.6 months. Conclusions: Total sacrectomy reconstruction using a VRAM flow-through flap anastomosed to a two-strut free fibular flap allows initial assessment of the recipient vessels during the first and ensuing oper-ative stages, satisfies the bone and soft tissue requirements of the defect, and provides a durable, functionally optimized reconstruction.VVC 2012 Wiley Periodicals, Inc. Microsurgery 00:000–000, 2012.

Sacral neoplasms are rare. The most common neoplasias

requiring sacrectomy are chordomas, which constitute

1–4% of all primary malignant bone tumors.1 Oncologic

management of primary sacral malignancies typically

entails en bloc, wide local excision of the tumor, necessi-

tating either a partial or total sacrectomy. Total sacrec-

tomy is a disabling resection that creates a massive tissue

defect and causes complete disjunction of the lumbar

spine from the pelvis. Since the lumbosacral and sacroil-

iac joints are critical for maintaining stability between the

spine and pelvis, their removal makes reconstruction of

this region essential for ambulation and weight bearing.2

An optimal surgical technique for total sacrectomy

reconstruction has not yet been determined. Traditional

methods of spino-pelvic stabilization include some form

of lumbo-iliac instrumentation, alloplastic tibia bone

grafts, and a trans-sacral pedicled vertical rectus abdomi-

nis musculocutaneous (VRAM) flap for soft tissue recon-

struction.3–6 Although this strategy does provide durable

soft tissue reconstruction of the total sacrectomy wound,

the use of large, non-vascularized, alloplastic tibia bone

grafts results in unacceptable rates of incomplete arthrod-

esis and eventual hardware failure. In comparison to non-

vascularized alloplastic bone grafts, vascularized fibula

bone flaps have been shown in other applications to pro-

vide 40% more strength, 56% more stiffness, higher

complete arthrodesis rates, and superior functional

outcomes.7–11 However, studies describing an efficient,

predictable, and safe strategy for reconstructing the total

sacrectomy defect with vascularized bone flaps are

lacking.3,7

The purpose of this report of this small series was to

describe this new technique and determine whether this

approach is safe for total sacrectomy reconstruction with

respect to patient outcomes and donor site morbidity.

PATIENTS AND METHODS

We reviewed the outcomes of the three patients who

underwent total sacrectomy reconstruction with a pedicled

VRAM flow through flap to a free fibula flap at our insti-

tution between January 1, 2009 and December 31, 2011.

Data were collected from a prospectively entered depart-

mental database and patients’ medical records. We

recorded patient demographic, reconstruction, and out-

comes data. The MD Anderson Cancer Center Institu-

tional Review Board granted approval for the study.

Our neurosurgeons stabilized patient defects with

spinopelvic instrumentation using rod fixation between

lumbar pedicle screws and iliac screws. Soft tissue defects

were reconstructed with pedicled transpelvic VRAM flaps.

The two-strut free fibula to pedicled transpelvic VRAM

flow-through flap provided bony stabilization between the

lumbar spine and iliac bones.

Surgical Techniques

A brief description of the stages of this total sacrec-

tomy reconstruction follows. Stage 1 was performed with

the patient in the supine position. The plastic surgeon

began the operation by making an anterior laparotomy

incision, protecting the medial perforators of the pedicled

1Department of Plastic Surgery, The University of Texas, MD Anderson Can-cer Center, Houston, TX2Department of Neurosurgery, The University of Texas, MD Anderson CancerCenter, Houston, TX3Department of Plastic and Reconstructive Surgery, The Johns HopkinsSchool of Medicine, Baltimore, MD

*Correspondence to: Patrick B. Garvey, M.D., Department of Plastic Surgery,Unit 1488, The University of Texas, MD Anderson Cancer Center, 1400Pressler, Houston, TX 77030. E-mail: [email protected]

Received 16 September 2011; Revision accepted 27 February 2012;Accepted 2 March 2012

Published online in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/micr.21990

VVC 2012 Wiley Periodicals, Inc.

VRAM flap. The colorectal surgeon mobilized the rectum

from the underlying sacrum. A vascular surgeon mobi-

lized and ligated the internal iliac vessels to devascularize

the sacrum. While the extirpative team completed these

maneuvers, the plastic surgeon dissected the bone-only

fibula flap, leaving the flap pedicled on the peroneal ves-

sels. For optimal orientation of the distal VRAM recipi-

ent vessels for the fibula flap pedicle, the fibula ipsilateral

to the pedicled VRAM was dissected. The plastic surgeon

also dissected the saphenous vein below the knee, ligating

side branches, for interposition vein grafting and leaving it

in continuity for later retrieval in the prone position during

Stage 3. After the neurosurgeons mobilized the lumbosac-

ral nerve trunks and initiated the anterior sacroiliac osteoto-

mies and the L5/S1 anterior diskectomy, threadwire saws

wrapped with cotton pledgets were left in the pelvis for

later retrieval during Stage 2 to facilitate completion of the

sacroiliac osteotomies. The plastic surgeon then temporar-

ily closed the fibula and saphenous vein graft donor sites

with staples and completed harvest of the pedicled VRAM

flap. It was not necessary to divide the insertion of the rec-

tus abdominis muscle from the pubic ramus. The flap could

reach the sacrectomy defect easily with the insertion intact,

and leaving the insertion avoided inadvertent twisting of

the pedicle during transpelvic delivery of the pedicled

VRAM flap into the sacrectomy wound. Care was taken to

adequately divide the posterior rectus sheath and perito-

neum to avoid kinking of the vascular pedicle during later

flap retrieval and inset. The distal end of the pedicled

VRAM flap was temporarily sutured to the sacral promen-

tory to facilitate flap retrieval during Stage 2. The pedicled

VRAM flap was placed in the pelvis, and the laparotomy

incision was primarily closed with unilateral component

separation to minimize tension on the closure.12 At the

plastic surgeon’s discretion, prosthetic mesh was inlayed

into the rectus sheath to reinforce the pedicled VRAM flap

donor site. The patient was then extubated and transferred

back to a hospital room to remain on bedrest until Stage 2

was completed within the next 3–5 days.

Stage 2 was performed with the patient in the prone

position to allow for a posterior approach to the sacrum.

During this stage, the neurosurgeon made a posterior

incision extending from the superior gluteal cleft to the

base of the spine. The gluteus and piriformis muscles, as

well as the sacrotuberous and sacrospinous ligaments,

were divided. The coccygectomy and mobilization of the

lower rectum were performed using a Kraske approach.13

After an L5 laminectomy was performed, the thecal sac

containing the S1–S5 sacral nerve roots were ligated and

divided. This facilitated completion of the posterior L5/

S1 diskectomy. Finally, the previously placed threadwire

saws were retrieved and the sacroiliac osteotomies were

completed. The sacrum was then resected en-bloc and

lumbo-pelvic instrumental stabilization was achieved.

It was our experience that Stage 2 was typically asso-

ciated with prolonged anesthesia time, significant blood

loss, and the use of multiple units of transfused blood

products, which caused the patient to become hypocoagu-

able and hemodynamically labile. For this reason, the

free fibula harvest and microvascular anastomosis was

delayed in two of the three cases until a third stage to

allow time for the patient to be resuscitated and stabilized

in the intensive care unit. If the patient were deemed to

be hemodynamically stable, the free fibula flap could

have been harvested and inset (see Stage 3) during Stage

2. Stage 2 was completed by the plastic surgeon, who

delivered the VRAM flap into the sacrectomy defect and

used it to temporarily close the wound.

Once the patient was adequately stabilized in the in-

tensive care unit (9–11 days in our experience), the

patient was returned to the operating room for Stage 3.

First, the sacrectomy wound was re-explored and the

bony defect exposed deep to the pedicled VRAM flap.

The plastic surgeon created an osteotomy in the middle

of the fibula flap that was the length required to span the

distance (as measured by the neurosurgeon) from the

base of the body of the fifth lumbar vertebra to the ilia

above the sciatic notches. The neurosurgeon used a cut-

ting burr to fashion a convexity in the inferior endplate

of the L5 body to receive the apex of the two-strut fibula

bipod and convexities in the bilateral ilia to receive the

distal and proximal ends of the free fibula flap. The fibula

flap was harvested and inset between the L5 body and

the ilia above the sciatic notches using lag screw fixation.

The distal branches of the deep inferior epigastric arteries

(DIEA) were dissected from the rectus abdominis muscle

of the pedicled VRAM in preparation for anastomosis to

the peroneal vessels. Microvascular anastomoses were

then fashioned between the peroneal artery and the distal

branches of the inferior epigastric artery of the pedicled

trans-sacral VRAM. The position of the recipient vessels

on the pedicled VRAM was superficial in the sacrectomy

defect due to the position of the pedicled VRAM, which

facilitated a straightforward execution of the anastomoses.

In two of the three patients presented here, a short, <3

cm long segment of a reversed saphenous vein graft was

required to correct size or length discrepancies between

the peroneal artery and the DIEA during arterial anasto-

mosis. Venae comitantes were anastomosed with venous

coupling devices. Following completion of the microvas-

cular anastomoses, the pedicled VRAM flap was inset

into the sacrectomy defect to complete the reconstruction.

Postoperative Care

Patients were kept in a non-weight-bearing state in an

air-fluidized bed for 4 weeks following surgery. After 4

weeks, the patient underwent supine transfers to a tilt ta-

ble followed by ambulation with maximal assistance. Hip

2 Garvey et al.

Microsurgery DOI 10.1002/micr

flexion and sitting were avoided for 2 months postopera-

tively. Sitting protocols involved progressive time inter-

vals of alternating 5-min buttock pressure for 1 month.

All three patients transitioned to outpatient rehabilitation

by 3 months postoperatively, where they ultimately pro-

gressed to unassisted full ambulation.

RESULTS

The most common indication for sacrectomy was

chordoma, followed by osteosarcoma (Table 1). The aver-

age age of the patients was 45 6 9.4 years. One patient

received preoperative pelvic radiotherapy. The average

follow-up time for the three total sacrectomy patients was

8.4 months. The average length of hospitalization was 79

days. All of the vascularized two-strut free fibula to

VRAM flow-through flap patients progressed to ambula-

tion with a walker by the time of hospital discharge and

were ambulating, with occasional use of a cane, after

outpatient rehabilitation. Stable arthrodesis of the fibula

flap was confirmed in all three patients an average of 3

months postoperatively using computed tomography

scans. Unfortunately, the first patient developed dissemi-

nated metastatic osteosarcoma and died 5 months after

surgery, which precluded confirmation of long term fol-

low-up of the free fibula flap with delayed repeat pelvic

imaging. At 2 years after surgery, the second patient

remains free of recurrence, lives independently, has

returned to work, and ambulates with the occasional aid

of a cane. The third patient began ambulation with the

aid of a cane at 6 weeks postoperatively and also remains

free of recurrence.

Case Report

A 53-year-old male presented with left inguinal, peri-

neum, and lower extremity numbness. MRI demonstrated

a mass involving nearly the entire sacrum. Biopsy of the

mass confirmed a sacral chordoma. The first stage of a

planned two-stage resection and reconstruction occurred

as described above. Three days later, Stage 2 was com-

pleted from the prone position and included completion

of the en bloc total sacrectomy (Fig. 1). Massive bleeding

necessitated ligation of the internal iliac vessels, tempo-

rary wound closure with the VRAM flap, and deferral of

the free fibula harvest and inset until the patient could be

stabilized in the ICU. Eleven days later, the patient was

returned to the operating room for harvest and insetting

of the free fibula to the pedicled VRAM flap (Fig. 2).

The patient was discharged after a 2-month hospitaliza-

tion. Postoperative surveillance CT scans at 3 and 9

months confirmed arthrodesis of the fibula flap to the

lumbar spine and iliac bones (Fig. 3). As of 1 1/2 years

after his surgery, he was able to ambulate unassisted,

Table

1.Patie

nt,Reconstruction,andOutcomesCharacteristics

Patient

Age

BMI

Preop

XRT

Preop

chemo

Tumortype

Postop

XRT

Postop

Chemo

Useof

vein

grafts

Stages

Days

betw

een

Stages

1and2

Days

betw

een

Stages

2and3

Hospital

discharge

(months)

Tim

eto

fusion

(months)

Ambulatory

statusat

discharge

Oncologic

outcome

Follow-up

(months)

121

28

Yes

Yes

Osteosarcoma

Yes

Yes

Yes

33

92.3

2.2

Unassisted6

cane

Deceased

after

recurrence

4.9

253

29

No

No

Chordoma

No

No

Yes

33

11

2.1

3.6

Unassisted6

cane

Norecurrence

16.7

358

26

No

Yes

Chordoma

No

No

No

25

NA

3.4

2.2

Unassisted6

cane

Norecurrence

3.5

BMI,bodymassindex;XRT,radiationtherapy.

Fibula to VRAM for Sacrectomy 3


drive an automobile, return to full-time employment, and

reported occasionally jogging for exercise.

DISCUSSION

In this report, we describe a reproducible technique

for total sacrectomy reconstruction that appears to opti-

Figure 1. (a) Appearance of the total sacrectomy defect during

Stage 2 after en-bloc resection of the sacrum and sacral malig-

nancy and placement of spino-pelvic instrumental stabilization. The

distal end of the transpelvic VRAM flap, which was not yet deliv-

ered into the extirpative defect, was in the base of the wound,

superior to the posterior surface of the rectum. (b) Appearance of

the surgical site after delivery of the transpelvic VRAM flap and

temporary flap inset at the conclusion of Stage 2. (c) Computed to-

mographic appearance of total sacrectomy defect between Stages

2 and 3 before placement of the two-strut free fibula flap. [Color fig-

ure can be viewed in the online issue which is available at wiley

onlinelibrary.com.]

Figure 2. Intra-operative appearance of the anastomoses between

the peroneal vessels of the free fibula flap (arrow) and the distal in-

ferior epigastrics of the transpelvic VRAM flap. Visualization of the

fibula flap itself was obscured by the overlying instrumentation and

morselized bone graft. [Color figure can be viewed in the online

issue which is available at wileyonlinelibrary.com.]

Figure 3. Nine-month postoperative computed tomographic appear-

ance of the two-strut free fibula flap inset between the L5 body and

the bilateral ilia above the sciatic notches following completion of

Stage 3 showed stable arthrodesis. [Color figure can be viewed in

the online issue which is available at wileyonlinelibrary.com.]

4 Garvey et al.


mally meet the reconstructive needs of the total sacrec-

tomy patient. We also present a 3-year experience of a

cancer center with one of the highest clinical volumes of

primary sacral tumors in the world. Our analysis of the

outcomes for these rare and challenging reconstructions

demonstrates favorable outcomes with free vascularized

bone flaps for total sacrectomy reconstruction.

Primary malignant sacral tumors require en bloc

resection. Depending on the size and location of the

involved sacrum, most sacral tumors can be managed

with a partial sacrectomy, which typically maintains

bony spinopelvic stability. On very rare occasions, a

total sacrectomy is indicated; this creates a predictably

massive soft tissue defect and uncouples the spinal col-

umn from the iliac bones.14,15 Resection of the entire

sacrum (the ‘‘universal joint’’ of the body) makes ambu-

lation, full weight bearing with standing, and transition-

ing from a sitting to standing position all but impossible.

After total sacrectomy, an ideal sacral reconstruction

should reestablish durable spinopelvic stability, obliterate

the massive soft tissue defect, and be predictably repro-

ducible.5 Given the rarity of total sacrectomy and the

anatomic challenges of the resulting defect, there is no

universally accepted strategy for total sacrectomy recon-

struction.

Sacral reconstruction with free flaps has proven to be

unpredictable and prone to complications.4,15 Soft tissue

reconstructions have been most commonly accomplished

using pedicled flaps such as transpelvic VRAM flaps,

bilateral gluteus-based flaps, and posterior thigh flaps.

The most commonly reported strategy for achieving bony

stabilization between the lumbar spine and iliac bones

has been to employ nonvascularized bone grafts such as

cadaveric tibia or fibula grafts.2,4–6,7,16–18 In light of the

diminished stability afforded by nonvascularized alloplas-

tic bone grafts, vascularized fibula flaps have been imple-

mented with good results.7,8,10 However, previous experi-

ence with free bone flap reconstruction of total sacrec-

tomy defects has been challenging owing to the absence

of readily available recipient vessels following a total

sacrectomy.

To safely complete a total sacrectomy, the internal

iliac vessels are often ligated to devascularize the sacrum

and decrease the risk of vascular injury. It has been our

experience that selective angiographic embolization of

feeding vessels to the sacral tumor is insufficient to satis-

factorily curtail intraoperative bleeding during a total sac-

rectomy. Ligation of the internal iliacs has been our

group’s preferred strategy for reducing bleeding in total

sacrectomies, but ligation of the internal iliac vessels ren-

ders the majority of the local recipient vessel options

unreliable for a free fibula flap. However, not ligating the

internal iliac vessels predisposes the patient to potentially

life-threatening intraoperative bleeding. In our second

case we did not ligate the internal iliac vessels during the

first stage of the operation in an attempt to preserve re-

cipient vessels for the free fibula flap, but this resulted in

profuse bleeding during the second stage of the sacrec-

tomy that ultimately necessitated ligation of the internal

iliacs. In the third patient in this series, we did ligate the

internal iliac vessels. The significantly lesser blood loss

experienced with the sacrectomy allowed for completion

of the free fibula flap harvest and inset during the second

stage of the operation, thus obviating the need for a third

operative stage.

Using the distal branches of the DIEA presents a

novel solution for reconstructive surgeons presented with

a total sacrectomy defect lacking recipient vessels. Long

saphenous vein grafts can be used from more distant ves-

sels as free flap recipients, but such grafts have been

shown in other applications to be associated with higher

rates of anastomotic thrombosis.19–21 Using the staged

approach described in this report allows the surgeon to

evaluate the recipient vessels when the pedicled VRAM

flap is initially harvested with the patient in the supine

position during the first stage of the resection and recon-

struction. The pedicled transpelvic VRAM has additional

benefits in that it both prevents the herniation of intra-ab-

dominal viscera into the sacral defect and obliterates the

massive soft tissue defect created by the total sacrectomy.

For the three patients in this report who underwent recon-

struction with free fibula flaps to pedicled flow-through

VRAM flaps, the distal branches of the DIEA were eval-

uated and deemed to be acceptable for free flap anasto-

mosis during the course of the resection and reconstruc-

tion. Owing to anatomic variability in the DIEA, this

approach may not always be possible. The anatomies of

the DIEAs on each side of the same patient frequently

differ.22,23 Given our success with this strategy for total

sacrectomy reconstruction, we plan to preoperatively

image the vascular anatomy of future patients’ bilateral

DIEA vessels using computed tomographic angiography

(CTA) in order to choose the DIEA system that appears

most anatomically amenable to serve as a free fibula flap

recipient.

Even when the best DIEA system is chosen for mi-

crovascular anastomosis, the distal branches of the DIEA

are small in diameter, especially when viewed at the time

of the Stage 1 pedicled VRAM harvest. Small-caliber

anastomoses tend to be more prone to thrombosis. In all

of the free fibula to pedicled VRAM patients in this

report, we did find that the distal DIEA and venae comi-

tantes had dilated by the time of the microvascular anas-

tomosis to the free fibula flap during Stage 3 of the

reconstruction. We also used small segments of reversed

saphenous vein grafts in two of three patients: to extend

the pedicle length in the first patient and to correct for ar-

terial size mismatch in the second patient. In light of



this, we recommend routine dissection of a saphenous

vein graft during the supine first stage of the total sacrec-

tomy resection and reconstruction. The saphenous vein

graft can be left in situ, in continuity in the lower leg,

until its harvest, if necessary, during the final stage of the

operation.

We now choose to perform the technique of total

sacrectomy reconstruction with a pedicled VRAM flow-

through flap to a free fibula flap in at least two opera-

tive stages. Originally, total sacrectomies were per-

formed as one-stage operations at our institution. Based

on our experience, we changed our practice of total

sacrectomy to a two-staged operation because the length

of the one-stage procedure was excessive. Our patients

appeared to more predictably tolerate the two shorter

operations. We have now added a third stage for the

final free fibula flap inset and anastomosis, as we

believe the free fibula flap microvascular transfer can

be more safely performed after resuscitation of a hemo-

dynamically unstable patient. However, it is reasonable

to perform the fibula flap transfer upon completion of

the total sacrectomy if the surgeon believes the patient

to be stable enough for safe execution of the microvas-

cular anastomoses.

Based on previous studies, it appears that the length

of time to arthrodesis between a bone flap and the recipi-

ent-site bone is shorter for vascularized free flaps than

for nonvascularized bone grafts.11,24–26 Clinical and

radiologic review of the total sacrectomy patients in this

report corroborates the findings of these prior studies. All

of the free fibula patients were able to ambulate unas-

sisted within two months of surgery and demonstrated

radiologic evidence of early arthrodesis at 2–3 months

postoperatively. Such rapid arthrodesis may afford the

patient tangible clinical benefits, including more rapid

advancement of physical activity and decreased incidence

of hardware failure in this biomechanically challenging

area.

CONCLUSIONS

Using a free fibula flap anastomosed to a transpelvic

pedicled VRAM flow-through flap appears to provide

many advantages over traditional strategies for total sac-

rectomy reconstruction, especially when recipient vessels

are lacking in the total sacrectomy defect. The approach

described in this report appears to restore function in a

reproducible, reliable, and durable manner.

ACKNOWLEDGMENTS

The authors recognize the following MD Anderson

Surgeons for their support and contributions of patients to

this series: Drs. David T. Chandler, David W. Chang,

George J. Chang, Kathleen S. Herbig, Valarae O. Lewis,

Kendall R. Roehl, and Garrett L. Walsh. They also thank

Dawn Chalaire from the MD Anderson Department of

Scientific Publications for scientific editing.

REFERENCES

1. Sundaresan N. Chordomas. Clin Orthop Relat Res 1986;204:135–142.

2. Murakami H, Kawahara N, Tomita K, Sakamoto J, Oda J. Biome-chanical evaluation of reconstructed lumbosacral spine after totalsacrectomy. J Orthop Sci 2002;7:658–664.

3. Dickey ID, Hugate RR Jr, Fuchs B, Yaszemski MJ, Sim FH. Recon-struction after total sacrectomy: Early experience with a new surgi-cal technique. Clin Orthop Relat Res 2005;438:42–50.

4. Miles WK, Chang DW, Kroll SS, Miller MJ, Langstein HN,Reece GP, Evans GR, Robb GL. Reconstruction of large sacraldefects following total sacrectomy. Plast Reconstr Surg 2000;105:2387–2394.

5. Doita M, Harada T, Iguchi T, Sumi M, Sha H, Yoshiya S, KurosakaM. Total sacrectomy and reconstruction for sacral tumors. Spine2003;28:296–301.

6. Zhang HY, Thongtrangan I, Balabhadra RS, Murovic JA, Kim DH.Surgical techniques for total sacrectomy and spinopelvic reconstruc-tion. Neurosurg Focus 2003;15:1–10.

7. Moran SL, Bakri K, Mardini S, Shin AY, Bishop AT. The use ofvascularized fibular grafts for the reconstruction of spinal and sacraldefects. Microsurgery 2009;29:393–400.

8. Choudry UH, Moran SL, Karacor Z. Functional reconstruction of thepelvic ring with simultaneous bilateral free fibular flaps followingtotal sacral resection. Ann Plast Surg 2006;57:673–676.

9. Chang DW, Fortin AJ, Oates SD, Lewis VO. Reconstruction ofthe pelvic ring with vascularized double-strut fibular flap follow-ing internal hemipelvectomy. Plast Reconstr Surg 2008;121:1993–2000.

10. Sakuraba M, Kimata Y, Iida H, Beppu Y, Chuman H, Kawai A. Pel-vic ring reconstruction with the double-barreled vascularized fibularfree flap. Plast Reconstr Surg 2005;116:1340–1345.

11. Goldberg VM, Shaffer JW, Field G, Davy DT. Biology of vascular-ized bone grafts. Orthop Clin North Am 1987;18:197–205.

12. Baumann DB, Butler CE. Component separation improves outcomesin VRAM flap donor sites with excessive fascial tension. PlastReconstr Surg 2010;126:1573–1580.

13. Westbrook KC, Lang N, Broadwater JR, Thompson BW.Posterior surgical approaches to the rectum. Ann Surg 1982;195:677–685.

14. York JE, Kaczaraj A, Abi-Said D, Fuller GN, Skibber JM, JanjanNA, Gokaslan ZL. Sacral chordoma: 40-year experience at a majorcancer center. Neurosurgery 1999;44:74–79.

15. Garvey PB, Rhines LD, Feng L, Gu X, Butler CE. Reconstructivestrategies for partial sacrectomy defects based on surgical outcomes.Plast Reconstr Surg 2011;127:190–199.

16. Diaz J, McDonald WS, Armstrong M, Eismont F, Hellinger M,Thaller S. Reconstruction after extirpation of sacral malignancies.Ann Plast Surg 2003;51:126–129.

17. Glatt BS, Disa JJ, Mehrara BJ, Pusic AL, Boland P, Cordeiro PG.Reconstruction of extensive partial or total sacrectomy defects withtransabdominal vertical rectus abdominis myocutaneous flap. AnnPlast Surg 2006;56:526–530.

18. Wuisman P, Lieshout O, Sugihara S, van Dijk M. Total sacrectomyand reconstruction: Oncologic and functional outcome. Clin OrthopRelat Res 2000;381:192–203.

19. Fudem GM, Marbel KR. Latissimus dorsi free flap for sacral woundclosure: The world’s longest vein grafts for free tissue transfer.Microsurgery 1996;17:449–451.

20. Schusterman MA, Miller MJ, Reece GP, Kroll SS, Marchi M,Goepfert H. A single center’s experience with 308 free flaps forrepair of head and neck cancer defects. Plast Reconstr Surg 1994;93:479–480.

6 Garvey et al.


21. Miller MJ, Schusterman MA, Reece GP, Kroll SS. Interposition veingrafting in head and neck reconstructive microsurgery. J ReconstrMicrosurg 1993;9:245–251.

22. Boyd JB, Taylor GI, Corlett R. The vascular territories of thesuperior epigastric and the deep inferior epigastric systems. PlastReconstr Surg 1984;73:1–16.

23. Moon HK, Taylor GI. The vascular anatomy of rectus abdominismusculocutaneous flaps based on the deep superior epigastric system.Plast Reconstr Surg 1988;82:815–832.

24. Lee MJ, Ondra SL, Mindea SA, Fine NA, Dumanian GA. Indica-tions and rationale for use of vascularized fibula bone flaps in cervi-cal spine arthrodeses. Plast Reconstr Surg 2005;116:1–7.

25. Shaffer JW, Field GA, Goldberg VM, Davy DT. Fate of vascularizedand nonvascularized autografts. Clin Orthop Relat Res 1985;197:32–43.

26. Wuisman P, Jiya TU, Van Dijk M, Sugihara S, Van Royen BJ, Win-ters H. Free vascularized bone graft in spinal surgery: Indicationsand outcome in eight cases. Eur Spine J 1999;8:296–303.



Documents

Vertical rectus abdominis musculocutaneous flow-through flap to a free fibula flap for total sacrectomy reconstruction