Share this article on:

Total Abdominal Wall Transplantation: An Anatomical Study and Classification System

Light, David M.D.; Kundu, Neilendu M.D.; Djohan, Risal M.D., M.B.A.; Quintini, Cristiano M.D.; Gandhi, Namita M.D.; Gastman, Brian R. M.D.; Drake, Richard Ph.D.; Siemionow, Maria M.D.; Zins, James E. M.D.

Plastic & Reconstructive Surgery: June 2017 - Volume 139 - Issue 6 - p 1466–1473
doi: 10.1097/PRS.0000000000003327
Reconstructive: Trunk: Original Article
Watch Video
Press Release

Background: Candidates for multivisceral transplantation present with complex defects often beyond traditional reconstructive options. In this study, the authors describe a dissection technique for a total abdominal wall vascularized composite flap. In addition, the authors suggest a classification system for complex abdominal wall defects.

Methods: Forty fresh, cadaveric hemiabdomens were dissected, with care taken to preserve the iliofemoral, deep circumflex iliac, superficial circumflex iliac, deep inferior epigastric, and superficial inferior epigastric arteries and corresponding veins. Perfusion patterns of the flaps were then studied using computed tomographic angiography.

Results: The deep circumflex iliac, superficial circumflex iliac, deep inferior epigastric, and superficial inferior epigastric arteries were identified along a 5-cm cuff of the iliofemoral artery centered on the inguinal ligament. Perfusion with an intact deep circumflex iliac artery yielded improvement in lateral perfusion based on computed tomographic angiography.

Conclusions: The authors propose an algorithm for abdominal wall reconstruction based on defect size and abdominal wall perfusion, and their technique for harvesting a total vascularized composite abdominal wall flap for allotransplantation. Total abdominal wall transplantation should be considered in the subset of patients already receiving visceral organ transplants who also have concomitant abdominal wall defects.

Cleveland, Ohio

From the Department of Plastic Surgery, Dermatology and Plastic Surgery Institute, the Department of Hepatobiliary and Transplant Surgery, Digestive Disease Institute, and the Department of Abdominal Diagnostic Radiology, Imaging Institute, Cleveland Clinic; and the Department of Anatomic Studies, Cleveland Clinic Lerner College of Medicine of Case Western Reserve University.

Received for publication December 1, 2016; accepted December 15, 2016.

Presented at the American Society of Plastic Surgeons Senior Resident Conference, in Tampa, Florida, on January 20, 2012, and winner of the best presentation award.

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

A “Hot Topic Video” by Editor-in-Chief Rod J. Rohrich, M.D., accompanies this article. Go to PRSJournal.com and click on “Plastic Surgery Hot Topics” in the “Digital Media” tab to watch. On the iPad, tap on the Hot Topics icon.

Neilendu Kundu, M.D., Department of Plastic Surgery, Cleveland Clinic, 9500 Euclid Avenue, Desk A-60, Cleveland, Ohio 44195, neilendu@gmail.com

Candidates for multivisceral transplantation generally have had a long and complex clinical course requiring multiple operations. This often includes combined midline and transverse laparotomy incisions, bowel resections, and creation of intestinal stomas. These procedures can be further complicated by hernias, infections, and enterocutaneous fistulas, resulting in further insult to the abdominal wall. By the time conventional treatment modalities have been exhausted, the abdominal wall is severely damaged, often making closure difficult.

Although plastic surgeons have been called on in the past to perform complex reconstructions, including component separation, tissue expansion, regional flaps, and free tissue transfers, such conventional techniques may be inadequate to address a subset of these patients.1 The use of vascularized composite allotransplants of the abdominal wall provides immediate coverage of the visceral grafts and avoids the potential for major and life-threatening complications.2 In 2003, Levi et al. reported on 160 intestinal transplants and estimated that abdominal wall closure was not possible in 20 percent of these patients using traditional modalities. Included in this report was the first description of nine partial abdominal wall transplants performed in eight patients. Five of the eight patients had fully functioning, viable abdominal wall flaps at a mean 23-month follow-up.3 Cipriani et al. described a modified technique using microsurgical anastomoses to perform three partial abdominal wall transplants that preserved the donor iliac vessels and was performed in less than 2 hours. All three flaps survived.4 Although technically more complex, remote revascularization of the abdominal wall flap using the forearm vasculature to decrease ischemia time has been performed.5 More recently, Datta et al. wrote a review on the combination of multiorgan and vascularized composite tissue allografts, including abdominal walls.6

The harvest techniques described by both Levi et al. and Cipriani et al. included only the rectus abdominis muscles with overlying subcutaneous tissue. Although this is an excellent option for coverage in many cases, it is not large enough for defects that involve the majority of the abdominal wall. The purpose of this study was to define a dissection technique for total abdominal wall vascularized composite allotransplantation. We also suggest an algorithm to guide the practicing surgeon in reconstruction of these complex defects.

Back to Top | Article Outline

MATERIALS AND METHODS

Twenty total abdominal wall cadaveric dissections (40 hemiabdomens) were performed in the anatomy laboratory at the Cleveland Clinic. Before dissection, the approximate origins of the deep circumflex iliac, superficial circumflex iliac, deep inferior epigastric, and superficial inferior epigastric arteries were measured in relation to the inguinal ligament using a ruler.

Back to Top | Article Outline

Dissection Technique

The markings for the flap began with the identification of the inguinal ligament, which spans from the anterior superior iliac spine to the pubic tubercle on each side of the abdomen (Fig. 1). The course of the iliofemoral artery can be found by bisecting the inguinal ligament. The flap was marked from the xiphoid to a point a 2 to 3 cm lateral to the anterior superior iliac spine bilaterally. The markings were created at least 10 cm below the inguinal ligament before being taken back medially toward the pubic tubercle—eventually joining with the contralateral side. Flap elevation was performed in a cranial-to-caudal manner. Bilateral costal incisions were created, beginning over the xiphoid and continuing laterally on top of the fusion of the false ribs. The dissection was then carried inferiorly over the edge of the ribs, through the abdominal wall and into the peritoneal cavity, where the falciform ligament was ligated (Fig. 2).

The dissection was then turned inferiorly toward the anterior superior iliac spine, along the middle axillary line. The external oblique, internal oblique, and transversus abdominis muscles were transected lateral to the course of the ascending branch of the deep circumflex iliac artery (Figs. 3 and 4).

Once past the anterior superior iliac spine, the incision was carried inferiorly and medially to a point approximately 7 to 10 cm below the inguinal ligament overlying the distal portion of the common femoral artery. The incision was then carried medial and superior toward the pubic tubercle, leaving a triangle of skin and subcutaneous tissue over the site of the femoral vessels. If the recipient’s femoral vessels are chosen as the site of anastomosis, this triangle of tissue would allow for a tension-free closure of vascularized tissue over the femoral vessels, avoiding the complications associated with a tense primary closure.

Distal control of the pedicle can be obtained at the bifurcation of the common femoral artery, approximately 7.5 cm distal to the inguinal ligament. This distance correlates with measurements found in the endovascular literature.7

At the level of the anterior superior iliac spine, the inguinal ligament was released. Above the anterior superior iliac spine, the iliac fascia was incised along the medial cortex of the iliac crest and the dissection into the pelvis was carried superficial to the iliacus muscle to ensure that the deep circumflex iliac artery was captured as it coursed through the transversalis fascia and transversus abdominis. Below the anterior superior iliac spine, the pedicle was approached laterally. The dissection was carried out superficial to the deep fascia of the lower extremity until the superficial and deep circumflex iliac vessels were encountered. This enables the pedicle to be approached superiorly within the pelvis, laterally or distally from the femoral vessels by manipulating the position of the flap. Multiple viewpoints allow for a more efficient and safer dissection of the pedicle (Fig. 5).

Lastly, the soft tissue above the pubic symphysis and insertion of the rectus abdominis muscles was divided, the contents of the inguinal canal were ligated, and the flap was ready for transfer. Initial dissections were performed to clearly define the anatomy and determine objective measurements of the main vessels to each other and contiguous structures. Finally, six fresh cadavers were then used to determine proof of concept (e.g., the critical vasculature needed to perfuse the abdominal flap). Computed tomographic angiography was performed on six hemiabdomens to evaluate the perfusion patterns of the inferior epigastric and circumflex iliac arteries using a Siemens Definition scanner (Siemens Healthcare, Erlangen, Germany). The abdominal flap was positioned flat on the computed tomography table. A nonenhanced scan of the flap was obtained at 100 mAs and 120 kVp with 0.6-mm collimation and 0.6-mm slice thickness. The deep circumflex iliac artery was subsequently clamped, and 0.3 cc of diluted Isovue-370 (Bracco Diagnostics, Cranbury, N.J.) was hand injected into the common femoral artery. Computed tomographic angiography images were obtained. The deep circumflex iliac artery clamp was then removed, and a repeated injection was performed in the common femoral artery using the same technique and personnel.

The images were post-processed on a Leonardo postprocessing workstation (Siemens Healthcare), and maximum intensity projection images were reconstructed in the coronal plane. Perfusion of the abdominal wall musculature (e.g., obliques and transversus abdominis) was compared to determine whether the inclusion of the deep circumflex iliac artery improved perfusion and therefore, by implication, possible flap survival.

Back to Top | Article Outline

RESULTS

The average flap dimensions measured 31 cm transversely by 28 cm from the xiphoid to the pubis. This resulted in a flap surface area of 868 cm2. Measurements of the origins of the deep circumflex iliac, superficial circumflex iliac, deep inferior epigastric, and superficial inferior epigastric arteries demonstrated that all four vessels were found along a 5-cm cuff of the iliofemoral artery centered on the inguinal ligament (Fig. 5).

The venous anatomy was also evaluated. It was noted that the common femoral vein contained a valve in 75 percent of the specimens. Computed tomographic angiography revealed an obvious difference between flaps with and without the deep circumflex system (Figs. 6 through 8). Perfusion of the lateral muscles was clearly improved by inclusion of the deep circumflex iliac artery.

Back to Top | Article Outline

DISCUSSION

Remarkable advances in organ transplantation have been made in a relatively short period. Vascularized composite tissue allotransplantation is perhaps the most innovative and exciting area of transplantation medicine and surgery today. Although much attention has been directed toward the face and hand, when abdominal wall allotransplantation is combined with organ transplantation, it has the potential to have perhaps the lowest long-term morbidity because no additional immunosuppression is needed. Abdominal wall transplantation has been described both in humans and in animal models.8–10 Some have even advocated for potentially expanding the indications to include patients who suffer from massive “end-stage” hernias11 with neurotized abdominal wall flaps.12

Inadequate perfusion and lack of adequate soft-tissue coverage of the abdominal wall are the main obstacles to traditional means of reconstruction in these patients. Abdominal wall perfusion from the deep epigastric vessels and perforators can be compromised by paramedian or midline incisions, stomas, and enterocutaneous fistulas. Similarly, subcostal and intercostal vessels traversing toward the midline can be injured by subcostal or “chevron” incisions. This may leave only lumbar vessels and possibly the circumflex and inferior epigastric vessels to perfuse the abdominal wall, provided that they were not disrupted by a transverse lower abdominal incision. Furthermore, multiple insults to abdominal wall perfusion critically limits options for tissue expansion, regional flaps, or component separation.13

Partial abdominal wall transplants have previously been performed using bilateral vertical rectus abdominis musculocutaneous flaps harvested as a single unit. Although this is an excellent option for limited defects, the abdominal walls of intestinal or multivisceral transplant patients are often so severely compromised in terms of both perfusion and compliance that larger soft-tissue requirements are often needed for adequate coverage. Without larger flaps, the abdominal cavities of these patients may not be able to safely accommodate the visceral grafts and associated postoperative edema that ensues following these extensive procedures. Inadequate coverage has been associated with decreased intestinal graft survival and with higher rates of retransplantation and death.14

We found that the deep circumflex iliac, superficial circumflex iliac, deep inferior epigastric, and superficial inferior epigastric arteries and their corresponding veins all branch off of the iliofemoral system within a 5-cm cuff at the level of the inguinal ligament. This is consistent with descriptions found in classic anatomy texts and with the recent findings of Hollenbeck et al.15,16 This ensures that a flap raised on an iliofemoral pedicle that spans 5 cm or more on either side of the inguinal ligament will capture the deep circumflex iliac, superficial circumflex iliac, deep inferior epigastric, and superficial inferior epigastric arteries. This allows for a focused and expedited dissection of the pedicle. We also noted that in a majority of cases (75 percent), the common femoral vein contained a valve. This rate is also consistent with previous reports of 69 to 93 percent.17–19 To avoid issues with venous drainage of the flap, valves should be incapacitated before venous anastomosis, and the donor’s external iliac vein should be anastomosed to the recipient’s femoral or iliac vein to guarantee antegrade venous drainage.

In our cadavers, we performed a cranial-to-caudal dissection, which provides two potential advantages. First, it allows rapid and safe entrance into the peritoneal cavity, which is advantageous, as it expedites multiple organ procurement and allows the organ transplant team to begin their part of the operation while the plastic surgery team continues to work caudally. A top-down dissection also provides more than one view of the vascular pedicle as the dissection is carried caudally, making the isolation of the vessels faster and safer.

Back to Top | Article Outline

Classification System

The abdominal wall comprises multiple musculofascial units, each with its own blood supply. We suggest a classification system based on these units, which can be used as a guide in the decision-making for abdominal wall reconstruction (Fig. 9).

Type I defects are minor defects amenable to traditional reconstructive options, including tissue expansion, component separation, or regional flaps. Type II defects are centrally located between the linea semilunaris on each side of the rectus abdominis and are a result of trauma to the muscle. As previously shown by Levi et al. and Cipriani et al., a bilateral vertical rectus abdominis musculocutaneous harvested as a single unit can be used to reconstruct these defects. Either microsurgical technique using the deep inferior epigastric vessels or anastomosis of the femoral vessels can be used with success. Type III defects involve one hemiabdomen. These defects can be seen following a midline laparotomy with a subcostal extension and unilateral rectus injury. Reconstruction requires a flap that includes one rectus abdominis muscle with the ipsilateral oblique musculature, harvested on an iliofemoral pedicle that includes the deep inferior epigastric and superficial inferior epigastric arteries, and the superficial circumflex iliac and deep circumflex iliac arteries. Although such defects could possibly be reconstructed with a bilateral vertical rectus abdominis musculocutaneous flap, a hemiabdominal flap provides a more anatomical reconstruction based on a single pedicle with one anastomosis, rather than two pedicles and two anastomoses.

Type IV defects are nearly total or total abdominal wall defects. These defects may follow “chevron” or “Mercedes” type laparotomy incisions and may have resulted in injury to both rectus abdominis muscles. Reconstruction requires a total abdominal wall flap, including bilateral oblique and rectus abdominis muscle components, based on bilateral iliofemoral pedicles.

Ethical debates regarding the advisability of composite tissue allografts and who should be considered candidates continues to be debated vigorously. These issues are perhaps more clear-cut when it comes to total abdominal wall transplantation in patients who are receiving or have received visceral organ (e.g., liver, kidney) transplants. Although it has been shown that nearly total anterior abdominal wall coverage can be achieved with bilateral subtotal thigh flaps, the additional morbidity from extensive lower extremity wounds and operative time for harvesting the flaps make this a less attractive option for total and nearly total abdominal wall reconstruction in the transplant patient.20 Abdominal wall composite tissue allotransplantation, therefore, becomes a viable option in such cases.

A limitation of our study stems from the failure to provide motor innervation to the cadaver transplant. However, at the time of our dissection, no formal study provided a clinically functional reconstruction through sensory or motor neurotization of the abdominal wall flap. In fact, of the 21 abdominal grafts reported in the literature, none were innervated.12 Selvaggi et al. felt that neurotization of the abdominal graft was unnecessary because their patients did not develop significant hernias despite musculature atrophy in the graft.21

However, as transplant patients continue to survive longer, and graft sizes increase to encompass both the oblique and rectus abdominis muscles, an innervated and functional graft may be advantageous. Since the completion of our study, innervation of the abdominal wall transplant by means of established nerve coaptation techniques has been described.12 Intestinal and multivisceral transplant patients are well suited for neurotization of the abdominal graft because of their tacrolimus-based immunosuppression regimen. Experimental studies have shown that tacrolimus increases the number of myelinated axons and myelin thickness,22 increases the rate of axonal regeneration,23 and decreases the recovery time after nerve repair.24,25 Future investigation will be needed to determine the functionality of the innervated graft and to specify which nerves are best used to produce a functional transplant.

Back to Top | Article Outline

CONCLUSIONS

Abdominal catastrophes that necessitate intestinal or multivisceral transplantation may also present major abdominal wall reconstructive challenges. Some of these defects defy conventional reconstructive techniques. Abdominal wall defects in such cases can range widely in size, which at their most complex may require an abdominal wall composite tissue allograft. We suggest an algorithm for abdominal wall reconstruction based on defect size and abdominal wall perfusion.

Back to Top | Article Outline

REFERENCES

1. Alexandrides IJ, Liu P, Marshall DM, Nery JR, Tzakis AG, Thaller SR. Abdominal wall closure after intestinal transplantation. Plast Reconstr Surg. 2000;106:805–812.
2. Nishida S, Levi D, Kato T, et al. Ninety-five cases of intestinal transplantation at the University of Miami. J Gastrointest Surg. 2002;6:233–239.
3. Levi DM, Tzakis AG, Kato T, et al. Transplantation of the abdominal wall. Lancet 2003;361:2173–2176.
4. Cipriani R, Contedini F, Santoli M, et al. Abdominal wall transplantation with microsurgical technique. Am J Transplant. 2007;7:1304–1307.
5. Giele H, Bendon C, Reddy S, et al. Remote revascularization of abdominal wall transplants using the forearm. Am J Transplant. 2014;14:1410–1416.
6. Datta N, Yersiz H, Kaldas F, Azari K. Procurement strategies for combined multiorgan and composite tissues for transplantation. Curr Opin Organ Transplant. 2015;20:121–126.
7. Garrett PD, Eckart RE, Bauch TD, Thompson CM, Stajduhar KC. Fluoroscopic localization of the femoral head as a landmark for common femoral artery cannulation. Catheter Cardiovasc Interv. 2005;65:205–207.
8. Nasir S, Bozkurt M, Klimczak A, Siemionow M. Large antigenic skin load in total abdominal wall transplants permits chimerism induction. Ann Plast Surg. 2008;61:572–579.
9. Quigley MA, Fletcher DR, Zhang W, et al. Development of a reliable model of total abdominal wall transplantation. Plast Reconstr Surg. 2010;126:80–86.
10. Jin J, Williams CP, Soltanian H, et al. Use of abdominal wall allotransplantation as an alternative for the management of end stage abdominal wall failure in a porcine model. J Surg Res. 2010;162:314–320.
11. Singh DP, Mavrophilipos VD, Zapora JA, et al. Novel technique for innervated abdominal wall vascularized composite allotransplantation: A separation of components approach. Eplasty 2014;14:e34.
12. Broyles JM, Berli J, Tuffaha SH, et al. Functional abdominal wall reconstruction using an innervated abdominal wall vascularized composite tissue allograft: A cadaveric study and review of the literature. J Reconstr Microsurg. 2015;31:39–44.
13. Huger WE Jr. The anatomic rationale for abdominal lipectomy. Am Surg. 1979;45:612–617.
14. Watson MJ, Kundu N, Coppa C, et al. Role of tissue expanders in patients with loss of abdominal domain awaiting intestinal transplantation. Transpl Int. 2013;26:1184–1190.
15. Gray H. Anatomy of the Human Body. 1918; 2000.Philadelphia: Lea & Febiger; Bartleby.com.
16. Hollenbeck ST, Senghaas A, Turley R, et al. Withdrawn: The extended abdominal wall flap for transplantation. Transplant Proc. 2011;43:3535–3540.
17. Basmajian JV. The distribution of valves in the femoral, external iliac, and common iliac veins and their relationship to varicose veins. Surg Gynecol Obstet. 1952;95:537–542.
18. Mühlberger D, Morandini L, Brenner E. An anatomical study of femoral vein valves near the saphenofemoral junction. J Vasc Surg. 2008;48:994–999.
19. Powell T, Lynn RB. The valves of the external iliac, femoral, and upper third of the popliteal veins. Surg Gynecol Obstet. 1951;92:453–455.
20. Lin SJ, Butler CE. Subtotal thigh flap and bioprosthetic mesh reconstruction for large, composite abdominal wall defects. Plast Reconstr Surg. 2010;125:1146–1156.
21. Selvaggi G, Levi DM, Cipriani R, Sgarzani R, Pinna AD, Tzakis AG. Abdominal wall transplantation: Surgical and immunologic aspects. Transplant Proc. 2009;41:521–522.
22. Sulaiman OA, Voda J, Gold BG, Gordon T. FK506 increases peripheral nerve regeneration after chronic axotomy but not after chronic Schwann cell denervation. Exp Neurol. 2002;175:127–137.
23. Wang MS, Zeleny-Pooley M, Gold BG. Comparative dose-dependence study of FK506 and cyclosporin A on the rate of axonal regeneration in the rat sciatic nerve. J Pharmacol Exp Ther. 1997;282:1084–1093.
24. Sosa I, Reyes O, Kuffler DP. Immunosuppressants: Neuroprotection and promoting neurological recovery following peripheral nerve and spinal cord lesions. Exp Neurol. 2005;195:7–15.
25. Siemionow M, Gharb BB, Rampazzo A. Pathways of sensory recovery after face transplantation. Plast Reconstr Surg. 2011;127:1875–1889.
©2017American Society of Plastic Surgeons