PORCINE AND BOVINE CARTILAGE TRANSPLANTS IN CYNOMOLGUS... : Transplantation (original) (raw)
There are short- and long-term complications after meniscectomy, the most frequent and debilitating of which are instability and arthritis. Patients may function adequately after meniscus cartilage removal, but very poorly over time due to articular wear (1-3). Therefore, the need for cartilage transplantation is overwhelming among patients with painful joints trying to avoid joint arthroplasty. Xenograft meniscus transplantation might offer the ability to stabilize the meniscectomized knee and provide prophylaxis against early degenerative arthritis. In general, transplantation of xenograft tissues is limited by a hyperacute rejection response. This rejection is mediated by preexisting natural antibodies that interact with endothelial cells and cause the collapse of the vascular bed of the xenograft (4, 5). The binding of the natural anti-Gal* antibody to the carbohydrate structure Galα1-3Galβ1-4GlcNAc-R (termed the α-galactosyl epitope) on the xenograft endothelial cells was found to be the main cause for the destruction of these cells in primates and humans (6-9). Cartilage used as xenografts may not be subjected to hyperacute rejection, as it is minimally vascularized. Based on its low vascularization, cartilage has been considered to be of relatively low immunogenicity (10). Furthermore, as demonstrated below, the expression of α-galactosyl epitopes on porcine and bovine cartilage seems to be very low. In view of these considerations, it was of interest to determine whether meniscus cartilage xenografts or associated articular cartilage xenografts, transplanted into primates producing anti-Gal, are subjected to immune rejection.
In the present study, we evaluated the potential of porcine and bovine cartilage to be transplanted into cynomolgus monkeys, by determining the in situ immune reaction against the xenografts and the serum antibody response. We have further assessed the effect of manipulation of the cartilage by various physical and chemical agents on the immunogenicity of these grafts. In the following study (11), we demonstrate the specific changes in anti-Gal activity, which comprise a significant proportion of the immune response, to the long-term presence of the cartilage xenografts in these monkeys.
MATERIALS AND METHODS
Preparative procedures. Bovine and porcine stifle joints (i.e., knee of the hind leg) were sterilely prepared, followed by harvest of the medial meniscus cartilages. All surrounding soft tissue attached to the meniscus was removed with a scalpel. Articular cartilage plugs were harvested with a 5-mm biopsy punch to a depth of 5 mm, thereby including the articular cartilage and a small amount of underlying subchondral bone. The tissue was then immersed in alcohol for 5 min to wash away synovial fluid and lipid-soluble contaminants. The tissues were then frozen for later preparation or implantation.
The various physical and chemical treatments of the cartilage are summarized in Table 1. In group 1, one bovine meniscus and one porcine meniscus were subjected to ultraviolet radiation of 9×106 mJ/cm2 and then refrozen. The dosage was selected because it had been previously shown to increase cross-linking in collagen solutions (12). It was high enough to potentially damage cell surface carbohydrates, but low enough not to denature the cartilage and therefore destroy the meniscus architecture. One bovine meniscus and one porcine meniscus were exposed to an oxidation reduction potential of 1050 mV for 5 min, then refrozen. The dosage was selected based on the assumption that it is high enough to potentially damage cell surface carbohydrates by ozone oxidation and ensure sterilization of the tissue, but low enough not to destroy the collagen structure (unpublished observations). One bovine meniscus and one porcine meniscus were left untreated except for alcohol immersion and freezing. In monkeys of group 2, the UV and oxidation potential were increased by several-fold in order to determine whether antigenicity may be decreased (Table 1). In addition, meniscus cartilage was incubated for 24 hr in 5% polyethylene glycol solution for possible decrease in immunogenicity, as these conditions were previously shown to reduce rejection of heart allografts in humans (13).
Both meniscus and articular cartilage were also subjected to removal of cells and lipid molecules (termed acellular treatment) by a modification of the method reported by Courtman et al. (14), as this treatment was found previously to improve function of transplanted porcine aortic value leaflets. This process took eight steps. The meniscal and articular cartilage were immersed in hypotonic Tris buffer (pH 8.0) containing phenylmethyl sulfonyl fluoride (PMSF, 200 μM) in isopropyl alcohol. The PMSF solution was prepared immediately before use to avoid precipitation and sterilized by 0.22-μm Millipore filtration. The containers with the immersed tissues were placed in ice and stored for 24 hr in a shaker waterbath with gentle mixing. In the second stage, the cartilage tissues were placed in Tris buffer containing 1% Triton X-100 and 200 μM PMSF and again incubated for 24 hr at 4°C. The tissues were then rinsed throughly in Hanks' physiological solution. This was followed by enzymatic digestion using DNase and RNase 2000 U/ml at 37°C for 1 hr. After digestion, the tissues were immersed again in 1% Triton X-100 in hypotonic Tris buffer for 24 hr at 4°C with gentle shaking. On day 4, the cartilages were washed in Hanks' solution for 24 hr at 4°C. On day 5, the cartilages were again suspended in fresh Hanks' solution and washed under the same condition. Thereafter, the cartilages were removed, placed into sterile tubes containing fresh Hanks' solution with antibiotics, and refrigerated at -20°C until implantation.
Operative procedures. Treatment of the animals was in accordance with the institutional animal care regulations, which adhere to the regulations outlined in the U.S. Department of Agriculture Animal Welfare Act (9 CFR Parts 1, 2, and 3) and the conditions specified in the Guide for Care and Use of Laboratory Animals (NIH Publication No. 86-23).
The study was designed for each implant to be tested in the suprapatellar pouch of a cynomolgus monkey. The protocol was reviewed and approved by the animal care committee. With the animals under general inhalation anesthesia, a 1-cm incision was made directly into the suprapatellar pouch at the superior medial border of the patella extending proximally. A piece of either bovine or porcine meniscus 1 cm long was placed into the pouch with a single 3-0 nylon stitch as a marking tag. The cartilage was transplanted using sterile surgical technique into the six monkeys of each group. The wounds were closed with 3-0 vicryl. The animals were permitted unrestricted cage activity and monitored throughout the transplantation period for any signs of discomfort, swelling, or rejection. Two months or 1 month after transplantation (groups 1 and 2, respectively), the cartilage was explanted and subjected to histology and immunostaining studies.
Expression of α-galactosyl epitopes in cartilage. Assessment of α-galactosyl epitope expression in cartilage was performed in an ELISA with biotinylated anti-Gal (6). Cartilage from bovine or porcine meniscus was homogenized in phosphate-buffered saline (PBS) and brought to a concentration of 2% (vol/vol). Aliquots of 50 μl were plated in flat-bottom, 96-well tissue culture plates and dried overnight. Subsequently, the dried homogenates were fixed with 0.2% glutaraldehyde in PBS for 30 min, washed, and blocked with 0.2 M glycine then with 1% bovine serum albumin in PBS. Porcine endothelial cells (PEC) grown in 96-well plates were fixed by glutaraldehyde as above and served as positive control (15).
Biotinylated anti-Gal (6) was placed at various concentrations in the microtiter wells for 2 hr at room temperature. The plates were washed five times with PBS containing 0.05% Tween and incubated further with avidin-horseradish peroxidase for 1 hr. Color reaction in all plates was performed with _o_-phenylenediamine (Sigma, St. Louis, MO), and absorbance was determined at 492 nm. Similar binding studies were performed with sera from the transplanted monkeys. The sera also were assayed for antibody binding after removal of anti-Gal by adsorption with rabbit red cells.
Serum analysis. The occurrence of a humoral immune response against the xenograft was assessed by determining changes in antibody binding to porcine or bovine cartilage (as described below) and by complement-mediated cytotoxicity of PEC (7-9). In group 1, blood was drawn before surgery and every 2 weeks after surgery. In group 2, blood was obtained at 1-week intervals after transplantation. The blood samples were centrifuged, and serum was stored at -70°C.
Studies on antibody binding to cartilage were performed with monkey sera or monkey sera depleted of anti-Gal, in order to determine whether specific anti-cartilage antibodies are generated in response to the xenograft. Anti-Gal depletion was performed by adsorption of sera diluted 1:10 in PBS with an equal volume of packed rabbit red cells. Sera at various dilutions were incubated for 2 hr in microtiter wells containing glutaraldehyde-fixed cartilage homogenates (as described above). At the end of incubation, the sera were removed by washing, and the immunoglobulin binding was detected by incubation with horseradish peroxidase-conjugated rabbit anti-human IgG antibody (Dako, Copenhagen, Denmark) for 1 hr and color reaction, as described previously (15).
Histological analysis. Histological analysis of the implanted cartilage was performed with hematoxylin and eosin staining of representative sections of the explant. Identification of the infiltrating lymphoid populations was performed by immunostaining acetonefixed frozen sections with the following antibodies: anti-CD4 (OPD4; Biomeda), anti-CD3 (rabbit polyclonal antibody; Dako), anti-CD8 (Dako), and the anti-macrophage monoclonal antibody OKM-1 (Biomeda). All primary antibodies were diluted 1:50. The antibody binding was determined with biotinylated secondary antibody, specific for the corresponding primary antibody. Positive cells were detected with avidin peroxidase and AEC chromogen.
RESULTS
α-Galactosyl epitope expression in porcine and bovine cartilage. To assess the possible use of cartilage as xenografts in primates, the expression of α-galactosyl epitopes was measured in porcine and bovine cartilage by ELISA with microscopic fragments of cartilage fixed by glutaraldehyde to microtiter wells. PEC were used as positive control for the antibody binding. As shown in Figure 1 and in agreement with previous observations (6, 15), anti-Gal bound extensively to PEC. However, the binding to porcine and bovine cartilage was minimal, even at high concentrations of the antibody. These findings suggest that cartilage tissues of bovine and porcine origin express only small amounts of α-galactosyl epitopes. In view of these findings, it was of interest to determine whether transplanted cartilage generates any immune response in primates.
Meniscus transplantation. In general, all bovine and porcine xenograft implants were well tolerated by the monkeys. No animals died during the course of the study. Postsurgical lameness or reluctance to move was transient. On Day 28, monkey F1743 (oxidation-treated porcine cartilage) and monkey F1820 (untreated bovine control) developed some localized swelling and mild tissue reaction along the incision line. These lesions resolved within 5 days without treatment. At the time of retrieval, the tissue surrounding the implants appeared to be normal or only slightly inflamed.
Histological studies on the graft. Histopathologic evaluation of the implants 1 month after transplantation revealed extensive cellular infiltrates in the peripheral regions of the cartilage (Fig. 2). Several characteristics of infiltrates were observed. Two of the implants contained primarily eosinophils organized in rows among the fibrochondrocytes (Fig. 2A), or a mixture of eosinophils and macrophages (Fig. 2B). The other implants had mononuclear cells, primarily lymphoid cells, and few eosinophils (Fig. 2C). Some of the lymphoid infiltrates were diffused within the cartilage (Fig. 2D), whereas others formed nodular aggregates (Fig. 2E). It is of interest to note that in monkey F1852, which was transplanted with articular cartilage and the adjacent synovial membrane, the synovial membrane contained an abundance of neutrophils (Fig. 2F), in contrast to the eosinophil infiltrates within the meniscus cartilage.
The histopathologic features of the implants 2 months after transplantation (i.e., group 1) were more uniform. All of these implants contained extensive cellular infiltrates composed primarily of lymphocytes, plasma cells, and macrophages with occasional foreign body giant cells and eosinophils. These cellular infiltrates were similar to those described in Figures 2C, 2D, and 2E. No treatment or species-related differences were observed in the various cartilage implants.
Immunostaining with anti-CD3 monoclonal antibody revealed that 80-90% of the mononuclear cells were T lymphocytes, whereas the monoclonal antimacrophage antibody indicated that 10-20% of the cells were macrophages. Among the T lymphocytes, approximately 50% displayed the CD4+ antigen (i.e., helper T cells) and the rest were CD8+ cells (i.e., cytotoxic T cells). Overall, the histological data demonstrate the development of an extensive chronic inflammatory process, characterized primarily by T lymphocytes and macrophage infiltration. Based on the infiltrates observed 1 month after transplantation, it is possible that early stages in the xenograft rejection are characterized by eosinophilic infiltration which gradually shifts to the chronic inflammatory response that includes mostly T lymphocytes and macrophages.
Immune response to the xenograft. The occurrence of a humoral immune response against the xenograft was determined in group 1 by measuring the production of anti-porcine or anti-bovine cartilage antibodies. For this purpose, antibody binding to corresponding fragmented meniscus cartilage was measured in ELISA. The specific production of anti-cartilage antibodies, in response to the xenograft, was assessed by subtracting the pretransplantation O.D. values from those measured in the serum samples of day 28 and day 60 after transplantation. Increased titers of anti-cartilage IgG antibodies were observed in all transplanted monkeys. In the two representative monkeys described in Figure 3, the activity of anti-cartilage antibodies 28 days after transplantation was not significantly different from that observed 60 days after transplantation. These findings indicate that the immune system of the transplanted monkeys responded to the implant by producing antibodies that interacted with the corresponding porcine or bovine cartilage.
For determining the contribution of anti-Gal to the increased anti-cartilage activity, the ELISA was repeated with sera depleted of anti-Gal (i.e., sera adsorbed on rabbit red cells). In monkey F1820, the anti-Gal-depleted sera displayed only a marginal binding to cartilage, suggesting that most of the increased anti-cartilage antibody activity may be attributed to elevation in anti-Gal activity (Fig. 3). However, in monkey F1858, anti-Gal depletion had only a small effect on the activity of anti-cartilage antibodies after transplantation. This suggests that a significant proportion of the cartilage binding antibody activity in this monkey is directed against specific cartilage antigens. The decrease in anti-cartilage antibody activity in the other four monkeys ranged between the data of the two monkeys in Figure 3. Overall, these data suggest that the primate response to the xenograft comprises various proportions of anti-Gal and anti-cartilage-specific antibodies. Detailed analysis of anti-Gal production in the transplanted monkeys is described in the article that follows (11).
DISCUSSION
The present study is the first to demonstrate the characteristics of the chronic rejection process of a nonprimate xenograft in an Old World primate. Because of the minimal vascularization of cartilage, meniscus and articular cartilage implants of bovine and porcine origin do not undergo hyperacute rejection in primates. Nevertheless, in spite of the low expression of α-galactosyl epitopes on these tissues, they are subjected to chronic rejection, characterized by inflammatory cells that infiltrate the periphery of the xenograft.
In most of the implants examined 1 month after transplantation, the cellular infiltrate included predominantly T lymphocytes and macrophages. However, one of the implants contained only eosinophils and the other contained a mixture of eosinophils and mononuclear cells. These observations suggest that in the absence of hyperacute rejection, the initial stages of the chronic rejection process may include the migration of eosinophils into the xenograft. Such migration is likely to be directed by chemotactic factors generated by the interaction of anti-Gal and other antibodies with epitopes on xenograft cells or on extracellular matrix molecules. We do not know at present whether induction of the granulocyte infiltrates may be partly associated with the surgical procedure. This procedure is unlikely to account for the inflammatory cellular infiltrates observed 2 months after transplantation because of the long postoperative period, which would allow nonimmunological inflammation to subside. Nevertheless, to exclude the possible contribution of the surgery to the long-term inflammation, future studies will include autologous implants as control.
Two months after transplantation, all implants contained mononuclear infiltrates characteristic of chronic inflammatory processes, regardless of the treatment of the cartilage. These infiltrates, also observed in four of the six monkeys in group 2 (i.e., 1 month after transplantation), suggest that within several weeks after transplantation an extensive cellular immune response develops within the graft recipient. The majority (80-90%) of the infiltrating cells were T cells, as indicated by staining with anti-CD3 antibody. The rest of the infiltrating cells were macrophages. The T cells included equal proportions of CD4+ and CD8+ cells. The antigenic specificity of this cellular immune response is not clear as yet. It may comprise an anti-Gal cellular response, as well as a response against other species and tissue-specific antigens. Indeed, a T cell-mediated, anti-porcine immune response of human lymphocytes was demonstrated in vitro (16, 17). Detailed analysis of the lymphocyte specificity in assays for helper and for cytotoxic T cells will help to elucidate the specificity of the cellular immune response against cartilage xenografts.
The de novo-produced antibodies in the transplanted monkeys are directed against cartilage-specific antigens as well as against the α-galactosyl epitope. Whereas all transplanted monkeys display an extensive increase in anti-Gal activity (for detailed analysis, see the study that follows, p. 646 [11]), variable proportions of the antibodies produced in response to the xenograft seem to interact with specific cartilage antigens. Both types of antibodies are likely to bind in situ to the xenograft and further propagate the inflammatory response against such cartilage implants. This humoral immune response suggests that individual molecules, or small fragments of the transplanted cartilage, are transported into the adjacent lymph nodes or to the spleen and act as effective immunogens, inducing the production of antibodies and possibly the development of a cellular immune response. It is of interest to note that, in spite of the extensive cellular and humoral anti-cartilage immune response, transplanted monkeys demonstrated no changes in locomotion capacity 6 months after the removal of the grafts, which implies that the xenografts triggered no cross-reactive immune response against autologous cartilage antigens.
The study of meniscus xenografts is of considerable clinical significance. Meniscus cartilage supports and stabilizes the knee joint and provides a near frictionless bearing surface. Loss of the meniscus leads to degenerative arthritis and instability (1-3). Our previous research has focused on meniscus cartilage regrowth mechanisms using collagen scaffolds (3, 18-23). Meniscus cartilage regeneration through collagen scaffolds is currently in human clinical trials with preliminary results yielding encouraging data about partial meniscal regeneration in segmental defects. In the setting where no cartilage is remaining (previous total meniscectomy), no effort at scaffolding has been effective. Additionally, in the setting of substantial degenerative change, the remaining native cartilage may not be healthy enough to provide effective regeneration (24). Human allograft menisci are being used for meniscus replacement; however, these allografts are accompanied by the expression of class I and class II histocompatibility antigens (25). Signs of rejection, including synovitis, tissue shrinkage, and effusion, have been noted after human allograft transplantation. A xenograft meniscus transplant might address these issues, and has advantages because no regrowth is required for immediate stability and durability, the supply is unlimited, sizing is possible at the time of implantation, and collagen organization is appropriate for meniscus cartilage. Furthermore, the regenerative ability of the immature porcine meniscus has been demonstrated in our earlier studies (20). However, rejection of xenograft organs has limited their use. Furthermore, cross-linking of extracellular matrix molecules by UV irradiation and oxidation by electric field did not help to reduce the immunogenicity of the grafts. Attempts to reduce immunogenicity of the cartilage by removal of lipids with the detergent Triton X-100 or by incubation with polyethylene glycol also did not alter the extensive lymphoid infiltration into the xenograft.
The normal porcine and bovine meniscus fibrochondrocytes may express many immunogenic epitopes, including the α-galactosyl epitopes. Additionally, the extracellular matrix, which includes collagen fibrils and proteoglycans, may provide similar immunogenic epitopes. Elimination of such epitopes is likely to help in preventing a continuous low level of stimulation. The belief, as stated by Jackson et al. (10), that menisci are an immunoprivileged tissue is not supported in the xenograft scenario. Until the animal cartilage is stripped of its immunogenic components or the human immunological response is effectively modified, cartilage xenotransplantation is likely to be hampered by inflammatory processes resulting in chronic rejection.
Binding of anti-Gal to porcine meniscus cartilage (□), bovine meniscus cartilage (▵), or PEC (○), as measured by ELISA.
Histology of xenograft cartilage implants 1 month after transplantation. Tissues were stained with hematoxylin and eosin. (A) Eosinophilic infiltrates among fibrochondrocytes in meniscus xenograft from monkey F4212 (×600). (B) Meniscus cartilage from monkey F1857 containing a mixture of eosinophils (open arrow) and macrophages (closed arrow). Fibrochondrocytes are evident in the upper right (×1000). (C) Meniscus cartilage from monkey F4204 containing primarily mononuclear cells and few eosinophils (granulated cytoplasm) in the region bordering the undestroyed cartilage (×1000). (D) Diffused lymphoid infiltrate in meniscus cartilage xenograft from monkey F4204 (×400). (E) Lymphoid nodular aggregate in meniscus xenograft from monkeys F4204 (×400). (F) Neutrophil infiltrate into porcine synovial membrane transplanted into monkey F4212 (×600).
Binding of IgG antibodies from sera of transplanted monkey to bovine (F1820) or porcine (F1858) cartilage, as measured by ELISA at 4 weeks after transplantation (•) and at 2 months after transplantation (▪). Open symbols (○, □) represent sera depleted of anti-Gal.
Footnotes
Abbreviations: anti-Gal, anti-Galα1-3Galβ1-4GlcNAc-R; PBS, phosphate-buffered saline; PEC, porcine endothelial cells; PMSF, phenylmethyl sulfonyl fluoride.
REFERENCES
1. King D. Regeneration of the semilunar cartilage. Surg Gynecol Obstet 1986; 62: 167.
2. Seedhom BB. Load-bearing function of the menisci. Physiotherapy 1976; 62: 223.
3. Stone KR. Meniscus replacement. Clin Sports Med 1996; 15: 557.
4. Auchincloss H. Jr. Xenogeneic transplantation. Transplant Rev 1988; 46: 1.
5. Platt JL, Vercelotti GM, Dalmasso AP, et al. Transplantation of discordant xenografts: a review in progress. Immunol Today 1990; 11: 450.
6. Galili U. Interaction of the natural anti-Gal antibody with αgalactosyl epitopes: a major obstacle for xenotransplantation in humans. Immunol Today 1993; 14: 480.
7. Good AH, Cooper DKC, Malcolm AJ, et al. Identification of carbohydrate structures that bind anti-porcine antibodies: implications for discordant xenografts. Transplant Proc 1992; 24: 558.
8. Sandrin MS, Vaughan HA, Dabkowski PL, McKenzie IFC. Antipig antibodies in serum react predominantly with Galα(1,3)Gal epitopes. Proc Natl Acad Sci USA 1993; 90: 11391.
9. Collins BH, Cotterell AH, McCurry KR, et al. Cardiac xenografts between primate species provide evidence for the importance of the αgalactosyl determinant in hyperacute rejection. J Immunol 1995; 154: 5500.
10. Jackson DW, McDevitt CA, Simon TM, Arnoczky SP, Atwell EA, Silvino NJ. Meniscal transplantation using fresh and cryopreserved allografts: an experimental study in goats. Am J Sports Med 1992; 20: 644.
11. Galili U, LaTemple DC, Walgenbach AW, Stone KR. Porcine and bovine cartilage transplants in cynomolgus monkey: II. Changes in anti-Gal response during chronic rejection. Transplantation 1997; 63(5): 646.
12. Weadock KS, Miller EJ, Bellincampi LD, Zawadzky JP, Dunn MG. Physical crosslinking of collagen fibers: comparison of ultraviolet irradiation and dehydrothermal treatment. J Biomed Mater Res 1995; 29: 1373.
13. Collins GM, Wicom WN, Levin BS, Verma S, Avery J, Hill JD. Heart preservation solution containing polyethylene glycol: an immunosuppressive effect. Lancet 1991; 338: 390.
14. Courtman DW, Pereira CA, Omar S, Langdon SE, Lee JM, Wilson GJ. Biomechanical and ultrastructural comparison of cryopreservation and a novel cellular extraction of porcine aortic valve leaflets. J Biomed Mater Res 1991; 29: 1507.
15. Galili U, Tibell A, Samuelsson B, Rydberg L, Groth CG. Increased anti-Gal activity in diabetic patients transplanted with fetal porcine islet cell clusters. Transplantation 1995; 59: 1549.
16. Yamade K, Sachs DH, DerSimonian H. Human anti-porcine xenogenic T cell response: evidence for allelic specificity of mixed leukocyte reaction and for both direct and indirect pathways of recognition. J Immunol 1995; 155: 5249.
17. Murray AG, Khodaoust MM, Pober JS, Bothwell AL. Porcine aortic endothelial cells activate human T cells: direct presentation of MHC antigens and costimulation by ligands for human CD2 and CD28. Immunity 1994; 1: 57.
18. Stone KR. Prosthetic knee joint meniscal cartilage. United States Patent No. 4,880,429, November 14, 1989.
19. Stone KR, Rodkey WG, Webber RJ, McKinney LA, Steadman JR. Meniscal regeneration with copolymeric collagen scaffolds: in vitro and in vivo studies evaluated clinically, histologically, and biochemically. Am J Sports Med 1992; 20: 104.
20. Stone KR, Rodkey WG, Webber RJ, McKinney LA, Steadman JR. Future directions: collagen based prosthesis for meniscal regeneration. Clin Orthop 1990; 252: 129.
21. Stone KR, Rodkey WG, Webber RJ, McKinney LA, Steadman JR. Replacement of the meniscus: original investigations in meniscal regeneration and a review. Presented as the Albert Trillat Young Investigator's Award at the International Society of the Knee, Rome, Italy, 1989.
22. Stone KR, Rodkey WG, Webber RJ, McKinney LA, Steadman JR. Meniscal regeneration with copolymeric collagen scaffolds: in vitro and in vivo studies evaluated clinically, histologically, and biochemically. Am J Sports Med 1992; 20: 104.
23. Stone KR, Rodkey WG, Webber RJ, McKinney LA, Steadman JR. Autogenous replacement of the meniscus cartilage: analysis of results and mechanisms of failure. Arthroscopy 1995; 111(4): 395.
24. Smillie IS. Observations on the regeneration of the semilunar cartilages in man. Br J Surg 1943; 31: 398.
25. Khoury M, El-Farou SM, Goldberg VM, Stevenson S. Fresh, frozen and twice-frozen human menisci express HLA-ABC and HLA-DR antigens. In: Transactions of the Orthopaedic Research Society 39th annual meeting, Feb. 15-18, 1993, p 357.
© Williams & Wilkins 1997. All Rights Reserved.