EXPERIMENTAL

Vascularized Tissue-Engineered Ears

Michael W. Neumeister, M.D. Tammy Wu, M.D. Christopher Chambers, Ph.D.
Springfield, Ill.

Background: A paucity of appropriate regional and local matching tissue can compromise the reconstruction efforts in areas of the body that require specialized tissue. The current study uses techniques of vascular prefabrication, tissue culturing, and capsule formation to form a vascularized ear construct that is reliably transferable on its blood supply. Methods: Thirty male Wistar rats (250 to 350 g) were anaesthetized. An incision was made over the right lower abdominal wall. A pocket was formed by blunt dissection just below the panniculus carnosus. A separate incision was made over the right femoral vessels, which were then isolated and transected distally. The vessels were transposed in a subcutaneous plane to the abdominal wound. A silicone mold in the shape of an ear (2 1.5 cm) was placed over the transposed vessels in the abdominal wound pocket. The wounds were closed. Auricular cartilage was minced, washed, and cultured. After 14 days, the chondrocyte culturing was complete and a vascularized capsule based on the incorporated, transposed femoral vessels was formed. The abdominal incision was then reopened, an incision was made in the lateral capsule, and the cultured chondrocytes were introduced into the molded capsule. Study groups included capsules filled with chondrocytes only, chondrocytes and a fibrin glue carrier, and the fibrin glue only. The capsule was closed and the wounds sutured. The prefabricated, prelaminated construct was isolated on its vascular pedicle 14 days later and traversed microsurgically to the contralateral leg vessels. Histologic analysis was performed. Results: All 30 capsules were completely vascularized and could be reliably isolated and transferred microsurgically on the transposed femoral vessels. The pedicle, being incorporated directly into the capsule, provided the dominant blood supply to the construct. None of the capsules with the fibrin glue only retained any shape and all were devoid of cartilage. Similarly, there was no evidence of retained cartilage in the capsules filled with chondrocytes alone. All capsules with the chondrocytes and the fibrin carrier had mature shaped cartilage preserved. External molds were required to maintain the shape of the ear. Extrusion, although almost uniform in the group with external molds, did not interfere with the end construct shape or vascularity. When molds were used, four of six had excellent maintenances of shape and two of six had only minor superior pole deformation. All constructs were reliably transferred as free flaps. Conclusions: The authors have shown that transposing a vascular pedicle to a subcutaneously placed silicone block will result in a vascular capsule that can be mobilized and transferred based solely on the pedicle. Although the capsule provides vascularity to the chondrocytes, the cultured cartilage will fill the shape of the silicone mold only if an appropriate carrier such as fibrin glue is used and an external mold is applied. (Plast. Reconstr. Surg. 117: 116, 2006.)

econstruction of the bodys specialized tissue is often extremely challenging. A paucity of appropriate regional or local matching tissue can compromise reconstructive efforts. PrefabricaFrom the Southern Illinois University School of Medicine, Plastic Surgery Institute. Received for publication July 19, 2004; revised January 19, 2005. Presented in part at the Annual Meeting of the Plastic Surgery Research Council, in Milwaukee, Wisconsin, June 9 to 12, 2001. Copyright 2005 by the American Society of Plastic Surgeons DOI: 10.1097/01.prs.0000195071.01699.ce

tion, prelamination, and tissue engineering have been used in difficult reconstructions. Prefabrication in its truest sense is a means of creating new vascularity in tissue that would otherwise be limited in axial or random blood supply.13 Enhancing the tissues blood supply through a greater number of capillary beds should improve mobility and survival of various flaps. This vascular enhancement of tissue can be achieved by flap delay, tissue expansion, administration of angiogenic growth factors, gene transfer, or the transposition of a vascular pedicle to the tissues in question. Each method of

116

www.plasreconsurg.org

Volume 117, Number 1 Vascularized Tissue-Engineered Ears

prefabrication attempts to create new, or capture additional, angiosomes within flaps so that larger and more reliable amounts of tissue may be elevated without the risk of vascular embarrassment and flap compromise.4 16 The current study was designed to use techniques of prefabrication of tissue and tissue culturing to create a vascularized autogenous ear construct. The transfer of a vascular pedicle to a new site creates a new set of angiosomes by neovascularization into the surrounding tissues from the pedicle. We hypothesized that a viable autogenous ear could be constructed through cell culturing and flap prefabrication yet be easily and readily transferable on its isolated pedicle. utes. The cells were washed three times in Hams F12 medium (Gibco). The cells were counted using a hemocytometer during the last wash step. An average of 6 105 cells were plated in a 2.5 cm2 flask using a medium consisting of Dulbeccos modified Eagles medium and Hams F12 in a 1:1 ratio supplemented with 10% fetal calf serum (Gibco), 50 g/ml L-ascorbic acid (Sigma), 10 mM HEPES buffer (Gibco), 2 g/ml Fungizone (Gibco), 100 U/ml penicillin, and 100 g/ml streptomycin solution (Sigma). The cultures were incubated in a humid atmosphere at 37C and 5% carbon dioxide. The medium was changed three times per week. The cells were subcultured to a 7.5-cm2 flask when they become confluent. The cells were then cultured for 7 to 14 days while the capsule was being developed in the subcutaneous space of the groin. Fifteen passages of cells were performed. Prefabrication of a Vascularized Capsule The animal was turned to a supine position. The abdomen and leg were prepared with povidone-iodine. A 2-cm segment of the femoral artery and vein was isolated and ligated distally through an incision in the leg (Fig. 1). A vertical midline incision was made on the abdomen extending from the leg incision. The dissection was carried through the skin and subcutaneous tissue, exposing the musculature of the abdominal wall. The dissection did not go through the muscle layers. A cavity was created in the subcutaneous plane to allow insertion of a 1-inch human earshaped silicone block. The femoral pedicle from the leg was then transected distally, transposed cephalad into the abdominal wound and sutured to the tissue directly juxtaposed to the silicone mold (Fig. 2). The abdomen and groin wounds were closed using 4-0 nylon sutures and returned to the animal care facility. Approximately 3 weeks after the initial operation, the silicone block was removed from the fibrous capsule that formed around the mold. The silicone mold was removed through an incision along the helical rim. The empty capsule was filled with one of three different mixtures: (1) chondrocytes only (4 106 cells) (n 12); (2) fibrin Tisseel, Baxter glue only (Baxter Healthcare, Deerfield, Ill.) (n 6); or (3) fibrin glue and cultured chondrocytes (4 106 cells) (n 12). The fibrin glue was used as a suspension medium for the cells. These groups were again subdivided into two groups, one receiving an external mold surrounding the capsule after infusion of each

MATERIALS AND METHODS

Thirty Wistar rats weighing 250 to 350 g were used in this experiment. Intraperitoneal Nembutal, 42 mg/kg (Abbott Laboratories, North Chicago, Ill.), was administered to obtain adequate anesthesia. All surgical procedures were performed under sterile conditions using autoclavesterilized instruments with the surgeon wearing a laboratory coat, mask, and sterile gloves. Each rats lower abdomen was shaved with an electric razor and sterile-cleansed with povidone-iodine. Each rats ears were also sterile-cleansed with povidoneiodine. Two procedures were performed on the animal in the first phase: (1) auricular cartilage harvest and chondrocyte culturing and (2) prefabrication of a vascularized capsule. Autogenous Cartilage Culturing Attention was first turned to the ears with the animal in the prone position. The auricular cartilage was harvested from either the left or the right ear. The cartilage specimen was immersed in saline while the wound underwent hemostasis with electrocautery and closed with 5-0 nylon suture. The isolated cartilage was washed three times in saline and minced into 1 1-mm pieces in a Petri dish. The minced cartilage was then transferred to a new Petri dish containing Hanks balanced salt solution. The cartilage was washed in Hanks balanced salt solution and digested using 1 mg/ml type 1A collagenase (Sigma, St. Louis, Mo.), 1 mg/ml hyaluronidase (Sigma), and 0.1 mg/ml deoxyribonuclease (Sigma) in Dulbeccos modified Eagles medium (Gibco, Grand Island, N.Y.). Digestion was carried out in a shaking waterbath at 37C for 18 to 20 hours. The cell suspension was filtered using a 70- m cell strainer and the cells pelleted by centrifuging at 1500 rpm for 10 min-

117

Plastic and Reconstructive Surgery January 2006

had been transposed onto the block became fully incorporated into the capsule and became the dominant blood supply of the construct. The vascularized ear constructs could be reliably isolated on the pedicle alone (Fig. 3). The ears survived microsurgical transfer from the operative site to the contralateral femoral vessels and were allowed to perfuse from 4 hours after the microsurgical anastomosis. The capsule had arterial inflow, good capillary refill, and venous return by means of the pedicle. Table 1 illustrates the development of cartilage between the various groups. Only six animals were used in the fibrin-only group to act as a negative control for the chondrocyte group. Injection of 1% fluorescein revealed complete vascularity of the capsule through the pedicle. None of the constructs from the rats in group 1 (cultured medium only) showed evidence of cartilage on histologic examination (Fig. 4). Likewise, there was no evidence of cartilage in group 2, in which the cultured chondrocytes alone were injected into the capsule (Fig. 5). Groups 1 and 2 has a mass of fibrosis of fibrin only within the vascularized capsule. However, all of group 3 specimens, where the cultured chondrocytes were within the fibrin glue carrier, were noted to have a glistening white appearance, with a firm yet pliable cartilage consistency (Fig. 6). On histologic examination, all rats in group 3 had evidence of mature cartilage that appeared viable and healthy (Fig. 7). Complications that arose from this experiment included extrusion, wound dehiscence, and poor maintenance of the construct configuration.

Fig. 1. A vascular pedicle is harvested from the right leg. A small cuff of muscle at the end of the pedicle is left on the femoral vessels.

Fig. 2. A silicone block in the shape of a human ear is placed in the subcutaneous pocket. The femoral vascular pedicle is transposed onto the silicone to prefabricate the ensuing capsule formation.

substance, and the other one not receiving the external mold. The constructs were evaluated for shape, histology, and vascularity after 8 weeks. The constructs were isolated on their vascular pedicle and transferred to the contralateral femoral vessels as free tissue transfers. Then, 1% fluorescein suspended in normal saline was injected into the femoral artery to document its vascular supply to the bioengineered construct. The animals were killed at this time after removal of the construct.

RESULTS
The harvested ear chondrocytes were successfully cultured to confluence within 2 weeks. Every animal developed a fibrous capsule around the implanted silicone block. The femoral vessels that

Fig. 3. Four weeks after implantation of the chromocytes into the capsule, the engineered ear construct is reliably isolated on its pedicle. The capsule and cartilage are well vascularized based on the incorporated transposed (prefabricated) vessels. An external mold was used to maintain the ear shape.

118

Volume 117, Number 1 Vascularized Tissue-Engineered Ears

Table 1. Comparison of Each Group of Prefabricated Ear Constructs
Fibrin Only (control)(n 6) Original shape with no external molds Shape with external molds Poor, 3/3 Poor, 3/3 Chondrocytes Only (n 12) Poor, 6/6 Poor plus fibrosis, 6/6 Chondrocytes with Fibrin Carrier (n 12) Poor, ill-defined clump of cartilage, 6/6 Excellent, well-maintained cartilage contour, 4/6; mild deformation, 2/6; normal histology of contoured cartilage

Fig. 4. There was no evidence of cartilage in the brin glue-only group of prefabricated capsules.

There was no extrusion with the initial insertion of the silicone mold to create the capsule. Extrusion occurred only in the group of rats receiving the external mold at the second stage of the procedure. The rate of extrusion was 13 of 14 rats (one rat died while receiving intraperitoneal Nembutal at the second stage), or approximately 93 percent. All these animals were brought back for resuturing

Fig. 6. The cartilage structure and integrity are preserved. The pedicle is incorporated into the capsule, which vascularizes the inner cartilage.

of the dehisced incision. Most of these specimens were therefore salvaged without compromise to the ear shape. Within this same group, three animals were found to have self-mutilated the skin overlying the external mold. Of all the study groups, only those groups that received fibrin glue with cultured cells maintained any viable cartilage. There were four of six animals that maintained the exact shape of a human ear (Fig. 3) in the chondrocyte/fibrin glue external mold group. Two of six animals had mild deformation in their group. There was poor maintenance of the ear shape in the groups that did not have an external mold.

DISCUSSION
Fig. 5. Cartilage cells were not visualized in the chondrocytesonly group of prefabricated capsules, implying that a chondrocyte carrier is required for cell survival.

This project arises from our interest in creating a prefabricated ear for reconstruction in the event of both congenital and acquired absence of the ear. To be successful, the tissue-engineered ear should be nonimmunogenic and reproducible,

119

Plastic and Reconstructive Surgery January 2006

point where the construct no longer requires support. In our model, this took approximately 4 weeks. The long-term shape of the constructs, however, was not evaluated. Further studies are being performed to look at the long-term shape, size, and biomechanics of the ear cartilage construct. Molds have been used both external to the skin and internally immediately adjacent to the construct underneath the skin. Cao and colleagues demonstrated in an athymic mouse model that chondrocytes seeded on a three-dimensional structure in the shape of a human ear will keep their shape for up to 8 weeks after removal of an external stint.19 However, these studies involve in vitro models and usage of allograft cartilage in an immunologically privileged model. External molds can be cumbersome and fraught with complications such as skin breakdown and excursion. In our study, we encountered extrusion of external molds in almost every animal in which they were used. Although this did not seem to interfere with the overall results, such a complication may impair its clinical utility. It would therefore be prudent to use an internal splint with resorbable scaffolds until the cartilage is thick enough or strong enough to support itself. Previous studies have shown that new cartilaginous tissue can be generated in a predictable shape by seeding chondrocytes onto synthetic biodegradable polymers with a predetermined shape.20 27 Recently, Kamil et al. used three different biodegradable polymers as scaffolds: calcium alginate, Pluronic F127, and polyglycolic acid to retain the seeded chondrocytes inside a gold mold.28 Although immunocompetent porcine models were used, an intrinsic vascularity was not developed in this model. The next step to the clinical application of these tissue-engineered constructs is to provide a reliable intrinsic and extrinsic vascularity. The current study used the process of capsule formation and prefabrication to produce a blood supply for the engineered construct. The fibrous capsule that forms around a foreign body (such as silicone) is composed of fibroblasts, macrophages, inflammatory cells, and an abundance of capillaries and collagen. Capsular flaps have been designed and clinically applied in other aspects of reconstructive microsurgery.29 32 The capsule, in our study, served three purposes: (1) to provide a closed compartment for the cultured chondrocytes, (2) to provide nutrients for the chondrocytes through its rich intrinsic vascular network, and (3) to allow the transposed femoral vessels a means of incorporation into the construct. Small capillaries from the capsule coursed into the car-

Fig. 7. A cartilage framework is created with the chondrocyte cultures only if brin glue is used as the carrier while the cells are transplanted into the prefabricated capsule.

maintain its shape and size, and remain viable on elevation and transfer. Maintaining the shape and size of the cartilage depends on a number of intrinsic and extrinsic factors. The intrinsic factors involve the inherent properties of the chondrocytes themselves, which is the cell type (i.e., hyaline, elastic, or fibrocartilage), the ability to survive in a given media, intracellular adherence, and growth. Extrinsic factors include deformational forces, nutrient supply, and movement or shear forces. We have demonstrated that cultured chondrocytes will thrive and grow only in the presence of an appropriate carrier in a vascularized capsule. We chose fibrin glue as our carrier based on previous work in our laboratory and other reports that have demonstrated fibrins tremendous ability to nurture cultured cells to survive and grow.17,18 The fibrin glue apparently facilitates diffusion of nutrients from the surrounding vascular capsule to the chondrocytes. Histologic studies in rats in group 3 (fibrin glue plus cultured chondrocytes) reveal incorporation and sustenance of the chondrocytes into the vascularized capsule. No cartilaginous cells were observed in either the fibrin control group or the cultured chondrocytes-only group. The cells seem to grow to a given restricted size. This restriction is offered by the use of the molds or scaffolds. External molds permit maturation of the cultured cells in a distinct and uniform pattern. Although the fibrous capsule contours to the shape of the silicone block, on removing the silicone and seeding the capsule with chondrocytes with fibrin glue carrier, the capsule does not reliably maintain the given shape. In contrast, an external mold can maintain the ear shape to a

120

Volume 117, Number 1 Vascularized Tissue-Engineered Ears

tilage from framework, giving an intrinsic blood supply to the construct. Prefabrication in our model described a process of vascular implantation and neovascularization. The femoral vessels that were transposed onto the silicone in the initial stages of the experiment became fully incorporated into the capsule. In 1981, Shen first described a successful microvascular transfer of a prefabricated free thigh flap to reconstruct burn contractures of the neck.6 Subsequently, various prefabricated constructs have been described both clinically and experimentally.4 16 Our study confirmed that, by transposing the vascular pedicle onto the silicone block, the pedicle will not only incorporate itself into the capsule but would also become the dominant blood supply to the capsule. By using the capsule and prefabrication techniques, an intrinsic vascularity is supplied to large engineered tissues. It also allows such tissue to be transferred either as a pedicle flap or a free flap to a distant site. To date, tissue-engineered constructs of ears have not incorporated an intrinsic blood supply. As the tissue-engineered constructs increase in size, greater concerns arise about the internal blood supply. Constructs developed without an internal blood supply will succumb to central necrosis and subsequent loss of function and shape. By using this technology, we may in the future be able to transfer tissue-engineered constructs as a free tissue transferred pedicle flap to areas of the body that are not amenable to standard means of reconstruction presently offered. The constructs involving different cell types could potentially be prefabricated for reconstruction of such tissues as ear, trachea, nose, mandible, and other three-dimensional configurations where restorative efforts have challenged reconstructive surgeons.
Michael W. Neumeister, M.D. Division of Plastic Surgery Southern Illinois University School of Medicine P. O. Box 19653 Springfield, Ill. 62794-9653 mneumeister@siumed.edu
4. Hori, Y., Tamai, S., Okuda, H., et al. Blood vessel transplantation to bone. J. Hand Surg. (Am.) 4: 23, 1979. 5. Stal, S., Parsa, F. D., and Spira, M. Secondary island composite flap: An experimental study in ear reconstruction. Ann. Plast. Surg. 11: 321, 1983. 6. Shen, T. Y. Vascular implantation into skin flap: Experimental study and clinical application: A preliminary report. Plast. Reconstr. Surg. 68: 404, 1981. 7. Shen, T. Y. Microvascular transplantation of prefabricated free thigh flap (Letter). Plast. Reconstr. Surg. 69: 568, 1982. 8. Shen, T. Y. Experimental study of tissue graft vascularization by means of vascular implantation and subcutaneous burying. Plast. Reconstr. Surg. 73: 403, 1984. 9. Hirase, Y., Valauri, F., Buncke, H. J., and Newlin, L. Y. Customized prefabricated neovascularized free flaps. Microsurgery 8: 218, 1987. 10. Hirase, Y., Valauri, F., and Buncke, H. J. Prefabricated sensate myocutaneous and osteomyocutaneous free flaps: An experimental model: Preliminary report. Plast. Reconstr. Surg. 82: 440, 1988. 11. Hyakusoku, H., Okubo, M., Umeda, T., and Fumiiri, M. Prefabricated hair bearing island flap for lip reconstruction. Br. J. Plast. Surg. 40: 37, 1987. 12. Morrison, W. A., Penington, A., Kumta, S. K., and Callan, P. Clinical applications and technical limitations of prefabricated flaps. Plast. Reconstr. Surg. 99: 378, 1997. 13. Khouri, R. K., Upton, J., and Shaw, W. W. Prefabrication of composite free flap through staged microvascular transfer: An experimental and clinical study. Plast. Reconstr. Surg. 87: 108, 1991. 14. Homma, K., Ohura, T., Sugihara, T., Yoshida, T., and Hasegawa, T. Prefabricated flaps using tissue expanders: An experimental study in rats. Plast. Reconstr. Surg. 91: 1098, 1993. 15. Hirase, Y., Valauri, F. A., and Buncke, H. J. Neovascularized bone, muscle and myoosseous free flaps: An experimental model. J. Reconstr. Microsurg. 4: 209, 1988. 16. Wieslander, J. B., and Wieslander, M. Prefabricated (expander) capsule-lined transposition and advancement flaps in reconstruction of lower eyelid and oral defects: An experimental study. Plast. Reconstr. Surg. 105: 1399, 2000. 17. Wechselberger, G., Schoeller, T., Stenzl, A., et al. Fibrin glue as a delivery vehicle for autologous urothelial cell transplantation onto a prefabricated pouch. J. Urol. 160: 583, 1998. 18. Sims, C. D., Butler, E. M., Cao, Y. L., et al. Tissue engineered neocartilage using plasma derived polymer substrates and chondrocyte. Plast. Reconstr. Surg. 101: 1580, 1998. 19. Cao, Y., Vacanti, J. P., Paige, K., et al. Transplantation of chondrocytes utilizing a polymer-cell construct to produce tissueengineered cartilage in the shape of a human ear. Plast. Reconstr. Surg. 100: 297, 1997. 20. Kim, W. S., Vacanti, J. P., Cima, L., et al. Cartilage engineered in predetermined shapes employing cell transplantation on synthetic biodegradable polymers. Plast. Reconstr. Surg. 94: 233, 1994. 21. Vacanti, C. A., Langer, R., Schloo, B., et al. Synthetic polymers seeded with chondrocytes provide a template for new cartilage formation. Plast. Reconstr. Surg. 88: 753, 1991. 22. Miralles, G., Baudoin, R., Dumas, D., et al. Sodium alginate sponges with or without sodium hyaluronate: In vitro engineering of cartilage. J. Biomed. Mater. Res. 57: 268, 2001. 23. Can, Z., Apaydin, I., Ercocen, A., et al. Prefabrication of a highdensity porous polyethylene implant using a vascular induction technique. Ann. Plast. Surg. 41: 264, 1998. 24. Lu, L., Zhu, X., Valenzuela, R., et al. Biodegradable polymer scaffolds for cartilage tissue engineering. Clin. Orthop. 391S: S251, 2001. 25. Paige, K. T., Cima, L. G., Yaremchuk, M., et al. De novo cartilage generation using calcium alginate-chondrocyte constructs. Plast. Reconstr. Surg. 97: 168, 1996. 26. Sims, C. D., Butler, E. M., Casanova, R., et al. Injectable cartilage using polyethylene oxide polymer substrates. Plast. Reconstr. Surg. 98: 843, 1996.

REFERENCES
1. Yao, S. T. Vascular implantation into skin flap: Experimental study and clinical application: A preliminary report. Plast. Reconstr. Surg. 68: 404, 1981. 2. Morrison, W. A., Dvir, E., Doi, K., Hurley, J. V., Hickey, M. J., and OBrien, B. Prefabrication of thin transferable axial pattern skin flaps: An experimental study in rabbits. Br. J. Plast. Surg. 43: 645, 1990. 3. Pribaz, J. J., Fine, N., and Orgill, D. P. Flap prefabrication in the head and neck: A 10-year experience. Plast. Reconstr. Surg. 103: 808, 1999.

121

Plastic and Reconstructive Surgery January 2006

27. Puelacher, W. C., Kim, S. W., Vacanti, J., et al. Tissue-engineered growth of cartilage: The effect of varying the concentration of chondrocytes seeded onto synthetic polymer matrices. Int. J. Oral Maxillofac. Surg. 23: 49, 1994. 28. Kamil, S. H., Vacanti, M. P., Aminuddin, B. S., et al. Tissue engineering of a human sized and shaped auricle using a mold. Laryngoscope 114: 867, 2004. 29. Bengtson, B. P., Ringler, S. L., George, E. R., et al. Capsular tissue: A new local flap. Plast. Reconstr. Surg. 91: 1073, 1993. 30. Heymans, M., Lengele, B., Lahlai, N., et al. A peri-implant capsule flap. Br. J. Plast. Surg. 46: 456, 1993. 31. Ginsbach, G., Busch, L., and Kuhnel, W. The nature of the collagenous capsules around breast implants: Light and electron microscopic investigations. Plast. Reconstr. Surg. 58: 643, 1979. 32. Leighton, W. D., Russell, R. C., Feller, A., et al. Experimental pretransfer expansion of free-flap donor sites: II. Physiology, histology and clinical correlation. Plast. Reconstr. Surg. 82: 76, 1988.

122