Osteomyocutaneous flap as a preclinical composite tissue allograft: Swine model

OSTEOMYOCUTANEOUS FLAP AS A PRECLINICAL COMPOSITE TISSUE ALLOGRAFT: SWINE MODEL XIAOPING REN, M.D.,1,4 MANSOUR V. SHIRBACHEH, M.D., Ph.D.,1,4 ¨ STU ¨ NER, M.D.,1,4 MARTY ZDICHAVSKY, M.D.,2 E. TUNCAY U JEAN EDELSTEIN, M.D.,1,4 CLAUDIO MALDONADO, Ph.D.,1,4 WARREN C. BREIDENBACH, M.D.,3 JOHANNES FRANK, M.D.,2 GORDON R. TOBIN, M.D.,1,4 JON W. JONES, M.D.,4 and JOHN H. BARKER, M.D., Ph.D.1,4*

Composite tissue allotransplantation (CTA) constitutes one of the last frontiers of microsurgery. Prior to its clinical application, the long-term efficacy of modern immunotherapy must be tested in a pre-clinical CTA model. Based on the concept of osteomyocutaneous forearm flap, we developed a CTA flap model in swine. After identifying the vascular territory of the flaps in six pigs (vascular casting), flaps were transplanted from mismatched donor to recipient pigs (n = 6). Rejection was assessed daily by visual inspection and histopathology of biopsy specimens. Recipient pigs were able to ambulate immediately following surgery. There were no flap

failures owing to technical or surgical complications. Rejection occurred over a period of 7 days as manifested by edema, cellular infiltration, epidermalysis, and thrombosis. This pre-clinical flap model is excellent for evaluating the effectiveness of modern immunotherapy because it is anatomically and immunologically relevant and because the minimal morbidity caused to the animal permits long-term studies.

The notion of using composite tissue allografts (CTAs) to

when rejection is prevented, nerve regeneration and the return of neuromuscular function is similar to that observed in syngeneic controls.13,14 Even though a great deal has been learned using the rat model, the significant immunologic differences that exist between rats and humans make it difficult to work out strategies for immune therapy. There are currently no studies using a large animal model that demonstrate rejection-free long-term graft survival. An ideal CTA model would be one that 1) represents “immunological barriers” and drugrelated toxicity similar to the human situation; 2) contains all of the individual components of a CTA including skin, muscle, bone, and bone marrow; 3) allows evaluation of sensory and motor function to the graft; and 4) carries minimal morbidity to the animal, thereby allowing long-term studies. Using these criteria, we developed a radial forearm osteomyocutaneous CTA flap in a pig model.

reconstruct soft tissue and musculoskeletal defects dates back centuries.1 Advances in reconstructive microsurgery, an increased experience with organ transplantation, and recent developments in immunosuppressive therapy have all led to an increased interest in CTA research and its clinical application.2 Research using the rat hindlimb model has laid the groundwork for understanding CTA rejection and the need for immunosuppression.3–5 Using the rat model, several laboratories have shown predictable patterns of rejection, with skin being the most antigenic of all the tissues in a CTA.6,7 These studies have shown that CTA rejection can be prevented using a variety of different immunosuppressive regimens.5,8–12 Finally, these studies demonstrate that

© 2000 Wiley-Liss, Inc.

MICROSURGERY

20:143–149

2000

1

Division of Plastic and Reconstructive Surgery, University of Louisville School of Medicine, Louisville, KY

2

Department of Trauma, Hand, and Reconstructive Surgery, University of Saarland, Homburg/Saar, Germany 3

Division of Hand and Micro Surgery, University of Louisville School of Medicine, Louisville, KY

4

Department of Surgery, University of Louisville School of Medicine, Louisville, KY

*Correspondence to: John H. Barker, M.D., Ph.D., Division of Plastic and Reconstructive Surgery, Department of Surgery, University of Louisville, 320 MDR Building, 511 South Floyd Street, Louisville, KY 40292. Received 20 December 1999; Accepted 8 March 2000 © 2000 Wiley-Liss, Inc.

MATERIALS AND METHODS

To define the vascular territories of the osteomyocutaneous CTA flap in the forelimb of pigs, we first performed anatomical dissections after Microfil injections in six pig forelimbs. Once we determined that the size of vascular pedicle was adequate for anastomosis, and that the flap contained both motor and sensory nerves, we performed five CTA transplants among age- and weight-matched male and female pigs. Rejection was monitored daily via visual

144

Ren et al.

inspection and at days 0, 2, 4, and 7 with biopsies (histologic examination) of the skin. Animals

Sixteen outbred female and male farm pigs, matched by age (8–10 weeks old) and size (17–24 kg), were used in this study. All animals were selected from healthy pigs with no prior history of allosensitization. After an initial physical examination, laboratory tests were performed to assess the general health of the animal. Cross-matching was performed for each donor–recipient pair to preclude hyperacute rejection. Animals were cared for in accordance with guidelines established by the Institutional Animal Care and Use Committee of the University of Louisville School of Medicine. Pigs were housed in separate cages in a light-, temperature-, and airflow-controlled room and were maintained on a balanced pig diet with free access to water throughout the experiment. CTA Flap Design

To design a CTA flap that contained skin, muscle, bone, and bone marrow in a pig limb, we first defined the vascular territories of the brachial artery in six pig forelimbs. Microfil (20 mL) was injected into the brachial artery and allowed to dry, and dissections were performed carefully following the vascular distribution into the proximal radius, the muscle belly of flexor carpi radialis, and the overlying skin. In addition to defining the vascular territory, the suitability for microsurgical anastomosis of the accompanying vessels was evaluated. Special attention was paid to the nerve supplying this block of tissue to ensure that both sensory and motor nerve ends were of sufficient size to be coapted to the recipient nerve in the transplant procedure (Fig. 1). CTA Transplant Procedure

On the day of surgery, both donor and recipient animals were pre-medicated with ketamine hydrochloride (Ketaset威, 20 mg/kg) and xylazine (Rompun威, 2 mg/kg), followed by atropine (0.5 mg) and thiopental sodium (Pentothal威, 50 mg). After intubation, animals were connected to a Quantiflex ventilator (Matrix Medical, Inc., Orchard Park, NY). Anesthesia was maintained with 2% isoflurane (IsoFlo威) in oxygen. Recipient animals also received 1,000 mL of D5LR solution. Postoperatively, animals received daily injections of 600,000 IU penicillin G and buprenorphine (Buprenex威, 0.3 mg) intramuscularly for infection prophylaxis and pain management, respectively. No anticoagulation or immunosuppressive therapy was used postoperatively. Donor flap and recipient beds were prepared simultaneously by two surgical teams. Animals were placed on the operating table in a right lateral decubitus position, with the right forelimb extended to 90° at the shoulder. The skin was shaved, scrubbed with iodine soap, painted with iodine solution, and followed by proper drape placement.

Donor flap. A 6 × 6-cm area of skin over the anteromedial aspect of the pig’s right forelimb was marked 3 cm below the elbow joint. After the skin incision was made, the cephalic vein and superficial radial nerve were identified and dissected proximally. The flexor carpi radialis was detached from its insertion into the medial epicondyle, and its tendon was cut distally. The neurovascular bundle supplying the muscle was identified and left intact. The pronator teres was then divided, exposing the brachial artery and median nerve, which were dissected proximally. The common interosseous branch of the brachial artery was divided. Proximal and distal osteotomies were performed in the radius without interruption of its periosteum, and a 6-cm segment of radial bone was removed from its attachment to the ulna. The principal neurovascular pedicle supplying the donor flap was left intact until the recipient bed was ready. Once the flap was detached, the donor animals were euthanized with 6 ml of Beuthanasia威 (390 mg/ml pentobarbital sodium and 50 mg/ml phenytoin sodium IV). Recipient bed. Identical incisions and dissections were performed on the right forearm of the recipient animal as described above. The brachial artery, cephalic vein, proximal and distal ends of median nerve, superficial radial nerve, and attachment sites of the flexor carpi radialis were identified. Special attention was paid to homeostasis. Allotransplant. Upon transferring the donor flap to the recipient bed, the donor segment of the radial bone was placed into the recipient defect and secured there with one 25- to 30-mm stainless steel screw placed into the adjacent ulna. The donor flexor carpi radialis was extended to its original length and sutured into its recipient bed with 4-0 nylon PRE-4 (Davis+Geck, Danbury, CT). The donor vessels were adjusted for adequate length and anastomosed end to end (8-0 and 9-0 nylon sutures, BV 130-5 and BV 100-4; Ethicon, Somerville, NJ) to their recipient counterparts, and the flap was allowed to perfuse. The proximal and distal ends of the median nerve were then coapted to their recipient counterparts using 9-0 nylon suture. The superficial radial nerve was also coapted to its recipient counterpart. Finally, the wound was irrigated and examined for bleeding, and the overlying skin was closed with 3-0 nylon. A loose sterile dressing was applied to the wound over which a fiberglass cast was placed on the limb. A window was created in the cast immediately over the flap for postoperative evaluation of flap rejection. Postoperative Evaluation

Animals were examined daily for general well being. The flap was inspected daily by two examiners through the window in the cast. The examiners observed and recorded skin color, warmth, edema formation, blistering, and tissue sloughing.

Composite Tissue Flap in Pigs

145

Figure 1. CTA flap after harvesting (A) and completion of vascular anastomosis (B).

Tissue/Blood Sampling

The animals were sedated (Pentothal威 10 mg/kg) to collect venous blood samples and skin biopsies from the flap on the day of surgery and on postoperative days (PODs) 2, 4, and 7. When the skin was completely rejected, the flap was explored to determine the condition of underlying donor tissues and vascular patency. The animal was then euthanized with 6 ml of Beuthanasia威. Skin biopsies were fixed in 10% formaldehyde and then

stored in 70% ethyl alcohol. Tissue sections were stained with hematoxylin and eosin and examined by two independent pathologists. Statistics

Values are reported as mean ± standard deviation. A paired Student t-test was used to compare the preoperative and postoperative laboratory results.

146

Ren et al.

Figure 2. CTA recipient ambulating with cast postoperatively.

RESULTS CTA Model

Vascular injection casts indicated that the territories irrigated by the brachial artery included a 6 × 6-cm area of skin island, the flexor carpi radialis, and a 4-cm segment of proximal radius. Anatomical dissection revealed that 1) the diameter of the brachial artery is approximately 1.5 to 2.0 mm, 2) the cephalic vein that drains the flap has a diameter of 2–3 mm, 3) the sensory innervation of the skin island branches from the superficial radial nerve (1–2 mm diameter), 4) the motor innervation of the flexor carpi radialis branches from the median nerve (2–3 mm diameter), and 5) the flap has a 7-cm segment of median nerve that would allow reinnervation of the muscle after transplantation. Figure 1A shows the flap after harvesting. Figure 1B shows the flap in place after vascular anastomosis. Postoperative Course

During flap transplant, the total flap ischemia time was 25 ± 7 min. Total operating time was 6.1 ± 0.7 h. All animals recovered from anesthesia without difficulty, were ambulatory, and tolerated a regular diet postoperatively (Fig. 2). On daily inspections, animals appeared healthy and did not lose weight during the course of the study. Complete blood count and serum electrolytes 27 did not change over the course of the study. Visual Flap Inspection

Daily inspection of the skin portion of the flaps revealed the following: At POD 1, skin appeared pink and warm (Fig. 3A); at POD 2, there was some edema along with

patchy erythema and normal bleeding from biopsy wounds (Fig. 3B); at POD 4, there was diffuse erythema with areas of blister formation (Fig. 3C); and at POD 7, skin appeared blue-purple with areas of epidermal loss and no bleeding from biopsy wounds (Fig. 3D). At this time (POD 7 ± 0) the flap was considered to be fully rejected. Histopathology

Independent examination of hematoxylin-eosin–stained skin samples by two pathologists indicated the following: At POD 1, all skin structures appeared normal with no evidence of intravascular coagulation, cell infiltration, or edema (Fig. 3A⬘); at POD 2, there was moderate perivascular and mild dermal–epidermal junction mononuclear cell infiltration accompanied by mild edema (Fig. 3B⬘); at POD 4, infiltrating cells became polymorphic in nature, with increased infiltration at the dermal–epidermal junctions as well as in the perifollicular region (Fig. 3C⬘); and at POD 7, polymorphic infiltration was present throughout the skin with intravascular coagulation and areas of necrosis (Fig. 3D⬘). DISCUSSION

Composite tissue allotransplantation is perhaps a new frontier in both reconstructive and transplantation surgery. CTA can be used for functional and structural restoration of defects resulting from trauma, surgical resection, congenital malformation, or massive burn.15 Recent advances in surgical techniques, better understanding of immunology, and new immunosuppressive drugs have brought CTA closer to

Composite Tissue Flap in Pigs

147

Figure 3. Gross pictures of flap: on POD 1, (A) showing normal skin; on POD 2, (B) showing patchy erythema; on POD 4, (C) showing diffuse erythema with areas of blister formation; and on POD 7, (D) showing dark color discoloration with areas of epidermal loss. Alongside are H&E pictures of skin: on POD 1, (Aⴕ) showing normal skin structures; on POD 2, (Bⴕ) showing moderate perivascular and mild dermal–epidermal junction mononuclear cell infiltration (open arrow); on POD 4, (Cⴕ) showing increased infiltration at the dermal epidermal junctions (solid arrow) as well as in perifollicular region; and on POD 7, (Dⴕ) showing cellular infiltration throughout skin structures with intravascular coagulation and areas of necrosis (asterisk).

148

Ren et al.

being a clinical reality. In fact, vascularized tendon,16 bone,17 knee joint,18 muscle,19 and nerve20 have been successfully transplanted in humans. The major obstacle to the clinical application of CTA has been the lack of large-animal studies in which graft survival and restoration of function is demonstrated over a long-term study period. Long-term study is crucial for CTA research because a potential recipient will be exposed to the potentially life-threatening complications of immunosuppression, but will have an original condition that is not life-threatening. The major shortcoming of CTA research has been lack of a reliable animal model in which to study the effectiveness of immunosuppression in preventing graft rejection, assessing drug toxicity to the graft and the host, and establishing functional return to the graft. Our osteomyocutaneous flap was in a large-animal model, contained all components of a CTA, allowed for study of sensory and motor functional return, and carried minimal side effects to the animal. Studies using rats in CTA research have laid the ground work in the pattern of graft rejection and the effectiveness of immunosuppression to prevent rejection. Rats are, however, very tolerant of allotransplantations. Arai et al.10 showed that a single high dose of FK506 can render recipient rats tolerant to hindlimb allografts. Such results cannot be reproduced in large-animal models. Studies using rabbits have been hampered by the unique sensitivity of rabbits to CSA. Randizo et al.21 showed that, although fetal rabbit hindlimbs grow normally in the host, most animals died as a result of a wasting syndrome which was attributed to the CSA. Studies using dogs have shown that the animal can bear weight and ambulate on a transplanted hindlimb. These studies were performed before the advent of CSA, and most animals died from toxicity to drugs used to prevent rejection.22,23 Unlike humans, dogs are acutely sensitive to tacrolimus, limiting the usefulness of this model in the study of modern immunotherapy regimens.24,25 Studies using non-human primates showed return of motor and sensory function to the CTA.26 Long-term graft survival was not achievable because of rejection or drug toxicity. Although pigs have not been used in CTA research, they have been studied extensively in solid organ allotransplantation and xenotransplantation.27,28 Pigs have similar blood groups and major histocompatibility complex (MHC) systems to humans.29 Furthermore, modern immunotherapy drugs have a similar toxicity profile in pigs and humans.30 This will enable us to gauge the potential side effects of the immunotherapy regimen being studied. Visual examinations indicate that the skin of the flap undergoes an expected pattern of rejection. The skin color changes from pink to red, secondary to an inflammatory process, followed by bluish discoloration caused by intravascular coagulation. The rejection process results in separation of the epidermis from the dermis, which is visually

manifested as blistering and sloughing. This pattern of skin rejection is also seen in other animal studies.6,31 Histological findings are consistent with a cellular rejection pattern, which has also been seen in other animal models. This consistent pattern of rejection can be used to evaluate the effectiveness of immunosuppression in preventing rejection of CTA. Our swine osteomyocutaneous flap contains skin, subcutaneous tissue, muscle, bone, periosteum, nerve, and vasculature. Our model, however, lacks cartilage or articular surface, which would be present in many clinical CTA models such as the hand. Because cartilage has very low antigenicity, it is very unlikely for the host to reject the cartilage while remaining tolerant to other tissues such as muscle or skin. This model, as with many potential clinical CTA models, contains vascularized bone marrow. This will allow detection of micro- or macrochimerism, which can potentially lead to tolerance.32 Vascularized bone marrow can potentially lead to graft-versus-host disease, which has been seen in some rat hindlimb models.33 The issue of graftversus-host defense must be fully addressed in a largeanimal CTA model because it carries significant morbidity and mortality. Potential restoration of sensory and motor function is a major potential benefit of CTA as compared with other reconstructive modalities. In this model, the motor innervation to flexor carpi radialis is preserved to facilitate study of nerve regeneration and restoration of muscle function in CTA. We also restored sensory innervation of the overlying skin through preservation of the lateral cutaneous antebrachial nerve, which is a branch of the superficial radial nerve. Although a complete understanding of sensory reinnervation such as pain, temperature, pressure, stereotactic and two-point discrimination can only be studied in humans, this model includes crude evaluation of sensory return such as pain and temperature withdrawal in pigs. Finally, there was minimal morbidity to the animal secondary to surgery. In addition, this enables us to perform long-term studies to test the effectiveness of immunosuppressive drugs. The major weight-bearing bone in the pig’s forelimb is the ulna. Segmental radius surgery does not significantly limit the ability to bear weight on the limb. Every animal was able to stand, ambulate, and eat after recovery from anesthesia (Fig. 2). When pigs cannot ambulate, they lose weight, develop decubitus ulcers, and are susceptible to wound sepsis and death. Our swine forelimb CTA model satisfies our criteria for an ideal preclinical CTA model because it 1) represents similar “immunological barrier” and drug-related toxicity to the human situation; 2) contains all the individual components of a CTA including skin, muscle, bone and bone marrow; 3) allows evaluation of sensory and motor function to the graft; and 4) carries minimal morbidity to the animal thereby allowing long-term studies. We believe that this

Composite Tissue Flap in Pigs

model can be used to test modern immunosuppressive drug regimens to ascertain their effectiveness in preventing rejection and their systemic side effects. This information can then be used in performing the first clinical CTA model.

16.

17.

REFERENCES 1. Danilevicius Z. SS. Cosmas and Damian: the patron saints of medicine in art. JAMA 1967;201:1021–1025. 2. Lull R, Beko KR, Black KS, et al. Composite tissue allotransplantation: perspectives concerning eventual clinical exploitation. Transplant Rev 1992;6:175–188. 3. Doi K. Homotransplantation of limbs in rats: a preliminary report on an experimental study with non-specific immunosuppressive drugs. Plast Reconstr Surg 1979;64:613–621. 4. Frits W, Swartz WM, Rose S, Futrell JW, Klein E. Limb allografts in rats immunosuppressed with cyclosporin A. Ann Plast Surg 1984;199: 211–215. 5. Press BH, Sibley RK, Shons. Limb allotransplantation in the rat. Ann Plast Surg 1986;16:313–321. 6. Buttemeyer R, Jones NF, Min Z, Rao U. Rejection of the component tissues of limb allografts in rats immunosuppressed with FK-506 and cyclosporine. Plast Reconstr Surg 1996;97:139–151. 7. Lee WPA, Yaramchuk MJ, Pan YC, Randolph MA, Tan CM, Weiland AJ. Relative antigenicity of components of a vascularized limb allograft. Plast Reconstr Surg 1991;87:401–411. 8. Black KS, Hewitt CW, Hwang JS, Borger RW, Achauer BM. Dose response of cyclosporine-treated composite tissue allografts in a strong histoincompatible rat model. Transplant Proc 1988;20:266–268. 9. Hotokebuchi T, Arai K, Takagishi K, Arita C, Sugioka Y, Kaibara N. Limb allograft in rats immunosuppressed with cyclosporine. Plast Reconstr Surg 1989;83:1027–1036. 10. Arai K, Hotokebuchi T, Miyahara H, et al. Prolonged limb allograft survival with short-term treatment with FK-506 in rats. Transplant Proc 1989;21:3191–3193. 11. Fealy MJ, Umansky WS, Bickel KD, Nino JJ, Morris RE, Press BH. Efficacy of rapamycin and FK-506 in prolonging rat hind limb allograft survival. Ann Surg 1994;219:88–93. 12. Benhaim P, Anthony JP, Lin LY, McCalmont TH, Mathes SJ. A long-term study of allogeneic rat hindlimb transplants immunosuppressed with RS-61443. Transplant 1993;56:911–917. 13. Press BH, Sibley RK, Shons AR. Limb allotransplantation in the rat: extended survival and return of nerve function with continuous cyclosporin/prednisone immunosuppression. Ann Plast Surg 1986;16:313– 321. 14. Black KS, Hewitt CW, Grisham GR, Caiozzo VJ, Howard EB, Achauer BM. Two new composite tissue allograft models in rats to study neuromuscular functional return. Transplant Proc 1987;19: 1118–1119. 15. Hewitt CW, Black KS, Achauer BM, Patel MP. Reconstructive allo-

18. 19. 20. 21.

22. 23.

24.

25. 26.

27. 28. 29. 30.

31.

32.

33.

149

transplantation: considerations regarding integumentary/musculoskeletal grafts, cyclosporine, wound coverage in thermal injury, and the immune response. J Burn Care Rehabil 1990;11:74–85. Guimberteau JC, Baudet J, Panconi B, Boileau R, Potaux L. Human allotransplant of a digital flexion system vascularized on the ulnar pedicle: a preliminary report and 1-year follow-up of two cases. Plast Reconstr Surg 1992;89:1135–1147. Hofmann GO, Kirschner MH, Buhren V, Land W. Allogenic vascularized transplantation of a human femoral diaphysis under cyclosporin A immunosuppression [Letter]. Transpl Int 1995;8:418–419. Hofmann GO, Kirschner MH, Wagner FD, Land W, Buhren V. Allogeneic vascularized grafting of a human knee joint with postoperative immunosuppression. Arch Orthop Trauma Surg 1997;116:125–128. Jones TR, Humphery PA, Brennan DC. Transplantation of vascularized allogeneic skeletal muscle for scalp reconstruction in a renal transplant patient. Transplant Proc 1998;30:2746–2753. Bain JR. Peripheral nerve allografting—review of literature with relevance to composite tissue transplantation. Transplant Proc 1998;30: 2762–2767. Randzio J, Kniha H, Gold ME, et al. Growth of vascularized composite mandibular allografts in young rabbits. Ann Plast Surg 1991;26:140– 148. Goldwyn R, Beach PM, Feldman D, Wilson RE. Canine limb homotransplantation. Plast Reconstr Surg 1966;37:184–195. Lance EM, Inglis AE, Figarola F, Veith VJ. Transplantation of the canine hind limb. Surgical technique and methods of immunosuppression for allotransplantation. A preliminary report. J Bone Joint Surg 1971;53A:1137–1149. Todo S, Demetris AJ, Ueda Y, et al. Canine kidney transplantation with FK-506 alone or in combination with cyclosporine and steroids. Transplant Proc 1987;19:57–61. Collier DSTJ, Thiru S, Calne R. Kidney transplantation in the dog receiving FK-506. Transplant Proc 1987;19:62. Daniel RK, Egerszegi EP, Samulack DD, Skanes SE, Dykes RW, Rennie WR. Tissue transplants in primates for upper extremity reconstruction: a preliminary report. J Hand Surg 1986;11A:1–8m. Swindle MM, Moody DC, Phillips LD. Swine as models in biomedical research. Ames, Iowa: Iowa State University Press. 1992. p 16–34. Pond WG, Houpt KA. The biology of the pig. Ithaca, NY: Cornell University Press; 1978. p 260–275. Lunney JK. Characterization of swine leukocyte differentiation antigens. Immunol Today 1993;14:147–148. Granger DK, Cromwell JW, Canafax DM, Matas AJ. Combined rapamycin and cyclosporine immunosuppression in a porcine renal transplant model. Transplant Proc 1996;28:984. Hovius SER, Stevens HPJ, van Nierop PWM, et al. Allogeneic transplantation of the radial side of the hand in the rhesus monkey: I. Technical aspects. Plast Reconstr Surg 1992;89:700–709. Hewitt CW, Black KS, Dowdy SF, et al. Composite tissue (limb) allografts in rats. III. Development of donor-host lymphoid chimeras in long-term survivors. Transplant 1986;41:39–43. Hotokebuchi T, Arai K, Takagishi K, Arita C, Sugioka Y, Kaibara N. Limb allografts in rats immunosuppressed with cyclosporine: as a whole-joint allograft. Plast Reconstr Surg 1989;83:1027–1036.