Arch Orthop Trauma Surg (1992) 111 : 250-254
ofOrthopaedic ..dTrauma Surgery © Springer-Verlag1992
Autogenously vascularised bone allografts Experimental model of a new bone-muscle composite graft Ch. Braun Abteilung ft~r Unfallchirurgie, Chirurgische Universit~itsklinik, W-6650 Homburg/Saar, Federal Republic of Germany
Summary. Conventional bone allografts carry a high incidence of complications such as infections and pseudarthroses due to immunological rejection and avascularity of grafts. In vascularised grafts healing and remodelling of bone is quicker and more complete. However, vascularised allografts need immunosuppression for prevention of rejection with vascular occlusion. Autogenously vascularised allografts are formed after implantation of bone in muscle of the recipient, allowing vascularisation from this muscle. A muscle-bone composite graft is thus obtained that can be transferred as a pedicled or free graft with microvascular anastomosis. In this study donors were D A and recipients Lewis rats. The bone grafts were implanted in the adductor muscles and transferred after 6 weeks into a femoral defect. A higher number of osteocytes were found in the autogenously vascularised group than in non-vascularised grafts. "Creeping substitution" was found in all cortical layers in vascularised grafts, whereas in conventional allografts bone resorption predominated. The experimental data suggest that in rat autogenously vascularised bone allografts show a remodelling pattern comparable with that of conventional vascularised bone autografts. The advantage of the autogenously vascularised bone allograft is that it allows transferral of a vascularised bone aUograft together with its well-vascularised recipient bed without immunosuppressive treatment.
To replace long segmental bone defects following trauma or tumour resection, transplantation of autogenous or allogenous cortical bone has been used [7-9, 13, 22]. Immunological rejection and avascularity of bone allografts are thought to be responsible for the infections, pseudarthrosis and graft fracture which occur after such transplants at a rate of 10-15% each [7, 12, 16]. Both experimental models and clinical series show a reduced rate of infections and pseudarthrosis in vascularised bone grafts [1, 2, 15, 22-24]. Microvascular transfer of fibular and iliac crest bone is an established method of reconstructing long bone defects even where the conditions in the recipient bed are unfavourable. The difficulty, and time involved in graft elevation and vessel anastomosis, morbidity at the donor site, limited availability, and reduced
stability due to anatomical misfitting are major disadvantages of this method. Against this, allogenic segmental bone grafts have a good load-bearing capacity and are osteoinductive and incorporable. However, vascularised a!lografts require long-term immunosuppression treatment to prevent rejection leading to failure of the blood supply [1, 10] - a treatment which does not seem justifiable when given only for improvement of extremity function, because of the considerable side effects. As an alternative to this, autogenously vascularised allogenic segmental bone grafts can be formed by implantation of allogenic bone in muscle of the recipient animal. By ingrowth of vessels from the muscle, the allograft develops an autogenous vascular supply. In experiments in the rat an adequate vascular supply to cortical bone segments had developed after 6 weeks' implantation in muscle [3]. With this approach a muscle-bone composite graft is obtained and can be transferred with microsurgical anastomosis of the muscle vessels or as a pedicled graft. The aim of the present study was to assess whether the results of allogenic bone grafting can be improved over those conventional non-vascularised allogeneic bone grafts by using autogenously vascularised grafts of this kind.
Materials and methods Recipients were adult male Lewis rats, donors were adult male DA rats. There were four groups, each containing eight animals operated on bilaterally; thus, 16 specimens were available for analysis in each group. These were: • Autogenously vascularised allografts. 8-mm donor femur segments cryopreserved at -70°C for 2 weeks were implanted bilaterally in the adductor muscles of the thigh (Fig. 1). Six weeks after implantation, the bone-muscle composite graft was transplanted on the adductor muscle pedicle in to a corresponding femoral defect. Stabilisation was by intramedullary K-wire. • AutogenousIy vascularised isografts. In this group donors and recipients were adult male Lewis rats. All other procedures were performed as in group 1. • Conventional non-vascularised allografts. After 2 weeks' cryopreservation the graft was implanted in the femoral defect. • Free non-vascularised allografts with previous implantation in recipient muscle. Cryopreservation and muscle implantation were as
in the other groups. After 6 weeks the muscle was removed from the bone and the graft was implanted in the femoral defect.
251
Fig. 1. Implantation of the graft in the muscle. The allogenic bone segment is enveloped in a portion of adductor muscle based on the vascular supply coming from the peroneal artery. After 6 weeks' implantation this muscle-bone composite graft was inserted in a nearby femoral defect
Cells/mm 2 600 500 400 300 200 100 0 I II Ill IV Fig. 2. Osteocytes per square millimetre graft area. Groups 1-4 are as described in the text: I and II, vascularised grafts; III and IV, non vascularised grafts
Fluorochrome labelling was done by intraperitoneal administration of the following markers: Achromycin (tetracycline) 50 mg/kg body weight 2 weeks after the osteosynthesis, alizarin complexon 35 mg/kg body weight 10 weeks after the osteosynthesis and xylenol orange 30 mg/kg body weight 14 weeks after the osteosynthesis [19]. Evaluation was carried out 15 weeks after osteosynthesis by light microscopy, microradiography and fluorescence microscopy. Laparotomy was performed with the animals under chloral hydrate anaesthesia. The aorta was canulated for angiography with india ink. The animals were killed in this same anaesthesia period, by exsanguination after which both femora including adjacent muscles were harvested and fixed in 40% methanol. One femur from each animal was decalcified in 20% EDTA. The calcified and decalcified specimens were both imbedded in methacrylate and 5gm, 150-gm and 750-gm thick slices cut. The 5-gm sections were Giemsa-staincd. Evaluation was qualitative and quantitative. Osteocytes per square milimetre graft area were counted. The overall graft area was acertained by a computer-aided image analysing sys-
Fig.3a-c. Giemsa-stained histological sections. (a) Isogenic vascularised composite graft: regularly arranged new bone formation in all cortical layers. Magnification 80 ×. (h) Allogenic vascularised composite graft: irregularly arranged zones of bone formation. Magnification 80 ×. (c) Cone with new bone formation surrounded by dead grafted bone ( ~ ) . Magnification 400 x
tem (Contron). Osteocytes were counted with an eye-piece micrometer at 100 × magnification by two independent examiners. For microradiography the 150-gm undecalcified sections were ground to a thickness of 100-gin. Contact microradiographs were made. In microradiography, new bone shows as less dense than old bone, which appears white and intensely calcified. Discriminating between different grey intensities with the help of a computeraided image analysing system the percentage of new bone corn-
252
Fig. 4a, b. Fluorochrome-labelled autogenously vascularised allograft (a) and non-vascularised allograft (b). In the vascularised graft fluorescence is seen all over the graft along haversian canals and in resorption cavities. Magnification 80 ×. In non-vascularised allografts there is only a little fluorescence due to appositional bone formation at the end and periosteal surface of the graft. Magnification 160 × pared with old grafted bone was measured. The same 100-btm sections were examined qualitatively for uptake of fluorochrome. The quantitative data from the microscopic and microradiographic assessments were analysed for statistical differences between four groups by the Wilcoxon test (criterion of significance: P < 0.01).
Results
Fig. 5a-c. Microradiographs of autogenously vascularised isograft (a) and allograft (b) and of a non-vascularised allograft (c). The dark areas are new bone, the white areas old grafted bone. Magnification 80 ×. There is new bone formation in all layers of the vascularised grafts, whereas in the non-vascularised graft only a little peripheral bone formation can be seen
Light microscopy Statistically, a higher n u m b e r of osteocytes were f o u n d in the a u t o g e n o u s l y vascularised c o m p o s i t e isografts and allografts than in the two groups of non-vascularised grafts (Fig. 2). N o difference regarding o s t e o c y t e counts was e s t a b l i s h e d b e t w e e n allogenic a n d isogenic c o m p o s i t e grafts. T h e r e w e r e only qualitative differences b e t w e e n these two groups: large zones of bone reorganisation were found in all layers of the cortex in b o t h groups of vascularised grafts. In allografts this r e o r g a n i s a t i o n s e e m e d to be disorderly: graft r e m o d e l l i n g by creeping substitution t o o k place with irregular r e s o r p t i o n cavities being
filled with new b o n e . In isografts remodelling was by enlarging and refilling of haversian canals (Fig. 3).
Fluorescence microscopy In both groups of composite grafts, u p t a k e of tetracycline was found in all layers of the graft. E v e n the first fluoroc h r o m e a d m i n i s t e r e d 2 w e e k s after graft transfer to the b o n e defect was f o u n d all o v e r the graft. T h e u p t a k e followed the pattern of the zones of bone reorganisation described a b o v e (Fig. 4a). In non-vascularised grafts only a
253 %
Active bone reorganisation with new bone formation was seen all over the graft in vascularised iso- and allografts. Corresponding to the histological findings, bone regeneration was seen to take place along widened haversian canals in isografts (Fig. 5a) and in irregular resorption cavities in allografts (Fig. 5b). There was no statistical difference between the two vascularised groups in regard to the proportion of old grafted to newly formed bone (Fig. 6). In the non-vascularised groups there was mainly bone resorption on the inner and outer graft surfaces, with statistically less bone formation than in the composite grafted groups as assessed total transplant area and new-to-old bone ratio.
In heterotopically implanted bone grafts not subjected to weight, resorption exeeds bone formation. This took place in our experiment during implantation in the adductor muscle. Haversian canals are widened and resorption cavities are formed through out all layers of the cortex. Six weeks are enough to get good vascular ingrowth throughout all layers of the graft from the wellvascularised muscle bed of the recipient. Mechanical load stimulates new bone formation. This is enhanced by prostaglandin E2, which can be secreted by immunocompetent cells [14]. In autogenously vascularised allogenic grafts, new bone formation is initiated in all cortical layers immediately after transfer to the femur defect, as indicated by the uptake of fluorescence marker administered 2 weeks after graft transfer. This is an advantage of vascularised bone transfers in general: remodelling can start at once after transfer. In non-vascularised grafts the period of vascular ingrowth with bone resorption must precede remodelling. The speed and completeness of this process depend on the quality of vascularisation of the recipient bed. In allogenously vascularised allografts there will be vascular occlusion by acute immunological rejection if immunosuppressive agents are not adminstered [10]. The advantage of this autogenously vascularised bone allograft is the possibility to transfer a vascularised bone graft together with its well-vascularised recipient bed without immunosuppressive treatment. For clinical practice, implantation of the bone graft in the latissimus dorsi muscle seems to be suitable. With microsurgical anastomosis this muscle can be easily transferred to the recipient site. In this way, allogenous bone can be transferred as a composite graft together with its well-vascularised bed even when conditions at the recipient site are unfavourable, such as when there is soft tissue deficiency or scarring after trauma or after radiation following tumour resection.
Discussion
References
30 25 20 15 10 5 0
I
II
III
IV
Fig. 6. New bone formation in the different groups as indicated by the percentage of new bone compared with grafted old bone
little fluorescence was found at the periphery of the graft (Fig. 4b).
Microradiography
Unlike parenchymal organs, bone is constantly regenerating [5]. This is the physiological basis of bone graft repair. Schleichender Ersatz or "creeping substitution" of grafted bone by the recipient newly formed bone characterises the mechanism of bone graft incorporation. Revascularisation is the first step in bone graft incorporation. Bone resorption and vascular ingrowth are the histological signs of this process [5]. Immunological graft rejection is also characterised by vascular invasion and bone resorption. Infiltrations by round cells and macrophages are other immunological reactions. These cells can transform into osteoclasts since they have common stem cells with osteoclasts [11, 17]. By the production of cytokines such as interleukins, prostaglandins, proliferation and growth factors, immunological effector cells influence angiogenetic, osteoclastic and osteoblastic processes [6, 14, 18, 20, 21]. In this way immunocompetent cells can play an important role in bone-remodelling and the creeping substitution of grafted bone by newly formed bone.
1. Aebi M, Regazzoni P, Schwarzenbach O (1989) Segmental bone grafting. Comparison of different types of grafts in dog. Int Orthop 13 : 101-111 2. Arata MA, Wood MB, Cooney WP (1984) Revascularized segmental diaphyseal bone transfers in the canine. J Reconstr Microsurg 1 : 11-19 3. Braun C, Dambe L, Seiler H, Btihren V (1989) Mit allogenem Knochen armiertes myocutanes composite graft zum mikrochirurgischen Transfer. Hefte Unfallheilkd 207 : 258 4. Burchardt H (1983) The biology of bone graft repair. Clin Orthop 174 : 28-42 5. Burchardt H, Glowczewskie FP, Enneking WF (1977) Allogeneic segmental fibular transplants in azathioprine immunosuppressed dogs. J Bone Joint Surg [Am] 59 : 881-894 6. Centrella M, McCarthy TL, Canalis E (1991) Transforming growth factor-beta and remodeling of bone. J Bone Joint Surg [Am] 73 : 1418-1428 7. Delloye C, Nayer P de, Allington N, Munting E, Coutelier L, Vincent A (1988) Massive bone allografts in large segmental defects after tumor surgery: a clinical and microradiographic evaluation. Arch Orthop Trauma Surg 107 : 31-41 8. Dodd CAF, Fergusson CM, Freedman L, Houghton GR, Thomas D (1988) Allograft versus autograft bone in scoliosis surgery. J Bone Joint Surg [Br] 70:431-434
254 9. Gebhardt MC, Roth YF, Mankin HJ (1990) Osteoarticular allografts for reconstruction in the proximal part of the humerus after excision of a musculoskeletal tumor. J Bone Joint Surg [AmI 72:334-345 10. Gotfried Y, Yaremchuk MJ, Randolph MA, Weiland AJ (1987) Histological characteristics of acute rejection in vascularised allografts of bone. J Bone Joint Surg [Am] 69 : 410-425 11. Hentunen TA, Tuukanen J, V~i~n~inenHK (1990) Osteoclasts and small populations of peripheral blood cells share common surface antigens. Calcif Tissue Int 47: 8-17 12. Lord C, Gebhardt M, Tomford W, Mankin HJ (1988) Infections in bone allografts - incidence, nature and treatment. J Bone Joint Surg [Am] 70: 369-376 13. Mankin HJ, Doppelt S, Tomford W (1983) Clinical experience with allograft implantation. Clin Orthop 174: 69-86 14. Murray DW, Rushton N (1990) The effect of strain on bone cell prostaglandin E2 release: a new experimental method. Caldf Tissue Int 47 : 35-39 15. Ostrup LT, Fredrickson JM (1974) Distant transfer of free living bone graft by microvascular anastomosis. An experimental study. Plast Reconstr Surg 54: 274-285 16. Ottolenghi CE (1972) Massive osteo- and osteoarticular bone grafts: technique and results of 62 cases. Clin Orthop 87:156164
17. Owen M (1978) Histogenesis of bone cells. Calcif Tissue Res 25 : 205 18. Pfeilschifter J, D'Souza SM, Mundy GR (1987) The effect of transforming growth factor beta on osteoblastic osteosarcoma cells. Endocrinology 121 : 212-218 19. Rahn BA, Fleisch H, Moor R, Perren SM (1970) The effect of fluorescent labels on bone growth and calcification tissue culture. Eur Surg Res 2 : 137-138 20. Raisz LG, Martin TJ (1984) Prostaglandins in bone and mineral metabolism. In: Peck WA (ed) Bone and mineral research. Ann 2, Elsevier, Amsterdam, pp 286-310 21. Ren W, Dziak R (1991) Effects of leukotrienes on osteoblastic cell proliferation. Calcif Tissue Int 49 : 197-201 22. Shaffer J, Field GA, Goldberg VM, Davy DT (1985) The fate of vascutarised and nonvascularised autografts. Clin Orthop 197 : 32-43 23. Taylor GI, Miller GDH, Ham FJ (1975) The free vascularised bone graft. A clinical extension of microsurgical technique. Plast Reconstr Surg 55 : 533-544 24. Weiland AJ, Moore JR, Daniel RK (1983) Vascularised bone autografts. Clin Orthop 174: 87-95 Received December 30, 1991
ofOrthopaedic ..dTrauma Surgery © Springer-Verlag1992
Autogenously vascularised bone allografts Experimental model of a new bone-muscle composite graft Ch. Braun Abteilung ft~r Unfallchirurgie, Chirurgische Universit~itsklinik, W-6650 Homburg/Saar, Federal Republic of Germany
Summary. Conventional bone allografts carry a high incidence of complications such as infections and pseudarthroses due to immunological rejection and avascularity of grafts. In vascularised grafts healing and remodelling of bone is quicker and more complete. However, vascularised allografts need immunosuppression for prevention of rejection with vascular occlusion. Autogenously vascularised allografts are formed after implantation of bone in muscle of the recipient, allowing vascularisation from this muscle. A muscle-bone composite graft is thus obtained that can be transferred as a pedicled or free graft with microvascular anastomosis. In this study donors were D A and recipients Lewis rats. The bone grafts were implanted in the adductor muscles and transferred after 6 weeks into a femoral defect. A higher number of osteocytes were found in the autogenously vascularised group than in non-vascularised grafts. "Creeping substitution" was found in all cortical layers in vascularised grafts, whereas in conventional allografts bone resorption predominated. The experimental data suggest that in rat autogenously vascularised bone allografts show a remodelling pattern comparable with that of conventional vascularised bone autografts. The advantage of the autogenously vascularised bone allograft is that it allows transferral of a vascularised bone aUograft together with its well-vascularised recipient bed without immunosuppressive treatment.
To replace long segmental bone defects following trauma or tumour resection, transplantation of autogenous or allogenous cortical bone has been used [7-9, 13, 22]. Immunological rejection and avascularity of bone allografts are thought to be responsible for the infections, pseudarthrosis and graft fracture which occur after such transplants at a rate of 10-15% each [7, 12, 16]. Both experimental models and clinical series show a reduced rate of infections and pseudarthrosis in vascularised bone grafts [1, 2, 15, 22-24]. Microvascular transfer of fibular and iliac crest bone is an established method of reconstructing long bone defects even where the conditions in the recipient bed are unfavourable. The difficulty, and time involved in graft elevation and vessel anastomosis, morbidity at the donor site, limited availability, and reduced
stability due to anatomical misfitting are major disadvantages of this method. Against this, allogenic segmental bone grafts have a good load-bearing capacity and are osteoinductive and incorporable. However, vascularised a!lografts require long-term immunosuppression treatment to prevent rejection leading to failure of the blood supply [1, 10] - a treatment which does not seem justifiable when given only for improvement of extremity function, because of the considerable side effects. As an alternative to this, autogenously vascularised allogenic segmental bone grafts can be formed by implantation of allogenic bone in muscle of the recipient animal. By ingrowth of vessels from the muscle, the allograft develops an autogenous vascular supply. In experiments in the rat an adequate vascular supply to cortical bone segments had developed after 6 weeks' implantation in muscle [3]. With this approach a muscle-bone composite graft is obtained and can be transferred with microsurgical anastomosis of the muscle vessels or as a pedicled graft. The aim of the present study was to assess whether the results of allogenic bone grafting can be improved over those conventional non-vascularised allogeneic bone grafts by using autogenously vascularised grafts of this kind.
Materials and methods Recipients were adult male Lewis rats, donors were adult male DA rats. There were four groups, each containing eight animals operated on bilaterally; thus, 16 specimens were available for analysis in each group. These were: • Autogenously vascularised allografts. 8-mm donor femur segments cryopreserved at -70°C for 2 weeks were implanted bilaterally in the adductor muscles of the thigh (Fig. 1). Six weeks after implantation, the bone-muscle composite graft was transplanted on the adductor muscle pedicle in to a corresponding femoral defect. Stabilisation was by intramedullary K-wire. • AutogenousIy vascularised isografts. In this group donors and recipients were adult male Lewis rats. All other procedures were performed as in group 1. • Conventional non-vascularised allografts. After 2 weeks' cryopreservation the graft was implanted in the femoral defect. • Free non-vascularised allografts with previous implantation in recipient muscle. Cryopreservation and muscle implantation were as
in the other groups. After 6 weeks the muscle was removed from the bone and the graft was implanted in the femoral defect.
251
Fig. 1. Implantation of the graft in the muscle. The allogenic bone segment is enveloped in a portion of adductor muscle based on the vascular supply coming from the peroneal artery. After 6 weeks' implantation this muscle-bone composite graft was inserted in a nearby femoral defect
Cells/mm 2 600 500 400 300 200 100 0 I II Ill IV Fig. 2. Osteocytes per square millimetre graft area. Groups 1-4 are as described in the text: I and II, vascularised grafts; III and IV, non vascularised grafts
Fluorochrome labelling was done by intraperitoneal administration of the following markers: Achromycin (tetracycline) 50 mg/kg body weight 2 weeks after the osteosynthesis, alizarin complexon 35 mg/kg body weight 10 weeks after the osteosynthesis and xylenol orange 30 mg/kg body weight 14 weeks after the osteosynthesis [19]. Evaluation was carried out 15 weeks after osteosynthesis by light microscopy, microradiography and fluorescence microscopy. Laparotomy was performed with the animals under chloral hydrate anaesthesia. The aorta was canulated for angiography with india ink. The animals were killed in this same anaesthesia period, by exsanguination after which both femora including adjacent muscles were harvested and fixed in 40% methanol. One femur from each animal was decalcified in 20% EDTA. The calcified and decalcified specimens were both imbedded in methacrylate and 5gm, 150-gm and 750-gm thick slices cut. The 5-gm sections were Giemsa-staincd. Evaluation was qualitative and quantitative. Osteocytes per square milimetre graft area were counted. The overall graft area was acertained by a computer-aided image analysing sys-
Fig.3a-c. Giemsa-stained histological sections. (a) Isogenic vascularised composite graft: regularly arranged new bone formation in all cortical layers. Magnification 80 ×. (h) Allogenic vascularised composite graft: irregularly arranged zones of bone formation. Magnification 80 ×. (c) Cone with new bone formation surrounded by dead grafted bone ( ~ ) . Magnification 400 x
tem (Contron). Osteocytes were counted with an eye-piece micrometer at 100 × magnification by two independent examiners. For microradiography the 150-gm undecalcified sections were ground to a thickness of 100-gin. Contact microradiographs were made. In microradiography, new bone shows as less dense than old bone, which appears white and intensely calcified. Discriminating between different grey intensities with the help of a computeraided image analysing system the percentage of new bone corn-
252
Fig. 4a, b. Fluorochrome-labelled autogenously vascularised allograft (a) and non-vascularised allograft (b). In the vascularised graft fluorescence is seen all over the graft along haversian canals and in resorption cavities. Magnification 80 ×. In non-vascularised allografts there is only a little fluorescence due to appositional bone formation at the end and periosteal surface of the graft. Magnification 160 × pared with old grafted bone was measured. The same 100-btm sections were examined qualitatively for uptake of fluorochrome. The quantitative data from the microscopic and microradiographic assessments were analysed for statistical differences between four groups by the Wilcoxon test (criterion of significance: P < 0.01).
Results
Fig. 5a-c. Microradiographs of autogenously vascularised isograft (a) and allograft (b) and of a non-vascularised allograft (c). The dark areas are new bone, the white areas old grafted bone. Magnification 80 ×. There is new bone formation in all layers of the vascularised grafts, whereas in the non-vascularised graft only a little peripheral bone formation can be seen
Light microscopy Statistically, a higher n u m b e r of osteocytes were f o u n d in the a u t o g e n o u s l y vascularised c o m p o s i t e isografts and allografts than in the two groups of non-vascularised grafts (Fig. 2). N o difference regarding o s t e o c y t e counts was e s t a b l i s h e d b e t w e e n allogenic a n d isogenic c o m p o s i t e grafts. T h e r e w e r e only qualitative differences b e t w e e n these two groups: large zones of bone reorganisation were found in all layers of the cortex in b o t h groups of vascularised grafts. In allografts this r e o r g a n i s a t i o n s e e m e d to be disorderly: graft r e m o d e l l i n g by creeping substitution t o o k place with irregular r e s o r p t i o n cavities being
filled with new b o n e . In isografts remodelling was by enlarging and refilling of haversian canals (Fig. 3).
Fluorescence microscopy In both groups of composite grafts, u p t a k e of tetracycline was found in all layers of the graft. E v e n the first fluoroc h r o m e a d m i n i s t e r e d 2 w e e k s after graft transfer to the b o n e defect was f o u n d all o v e r the graft. T h e u p t a k e followed the pattern of the zones of bone reorganisation described a b o v e (Fig. 4a). In non-vascularised grafts only a
253 %
Active bone reorganisation with new bone formation was seen all over the graft in vascularised iso- and allografts. Corresponding to the histological findings, bone regeneration was seen to take place along widened haversian canals in isografts (Fig. 5a) and in irregular resorption cavities in allografts (Fig. 5b). There was no statistical difference between the two vascularised groups in regard to the proportion of old grafted to newly formed bone (Fig. 6). In the non-vascularised groups there was mainly bone resorption on the inner and outer graft surfaces, with statistically less bone formation than in the composite grafted groups as assessed total transplant area and new-to-old bone ratio.
In heterotopically implanted bone grafts not subjected to weight, resorption exeeds bone formation. This took place in our experiment during implantation in the adductor muscle. Haversian canals are widened and resorption cavities are formed through out all layers of the cortex. Six weeks are enough to get good vascular ingrowth throughout all layers of the graft from the wellvascularised muscle bed of the recipient. Mechanical load stimulates new bone formation. This is enhanced by prostaglandin E2, which can be secreted by immunocompetent cells [14]. In autogenously vascularised allogenic grafts, new bone formation is initiated in all cortical layers immediately after transfer to the femur defect, as indicated by the uptake of fluorescence marker administered 2 weeks after graft transfer. This is an advantage of vascularised bone transfers in general: remodelling can start at once after transfer. In non-vascularised grafts the period of vascular ingrowth with bone resorption must precede remodelling. The speed and completeness of this process depend on the quality of vascularisation of the recipient bed. In allogenously vascularised allografts there will be vascular occlusion by acute immunological rejection if immunosuppressive agents are not adminstered [10]. The advantage of this autogenously vascularised bone allograft is the possibility to transfer a vascularised bone graft together with its well-vascularised recipient bed without immunosuppressive treatment. For clinical practice, implantation of the bone graft in the latissimus dorsi muscle seems to be suitable. With microsurgical anastomosis this muscle can be easily transferred to the recipient site. In this way, allogenous bone can be transferred as a composite graft together with its well-vascularised bed even when conditions at the recipient site are unfavourable, such as when there is soft tissue deficiency or scarring after trauma or after radiation following tumour resection.
Discussion
References
30 25 20 15 10 5 0
I
II
III
IV
Fig. 6. New bone formation in the different groups as indicated by the percentage of new bone compared with grafted old bone
little fluorescence was found at the periphery of the graft (Fig. 4b).
Microradiography
Unlike parenchymal organs, bone is constantly regenerating [5]. This is the physiological basis of bone graft repair. Schleichender Ersatz or "creeping substitution" of grafted bone by the recipient newly formed bone characterises the mechanism of bone graft incorporation. Revascularisation is the first step in bone graft incorporation. Bone resorption and vascular ingrowth are the histological signs of this process [5]. Immunological graft rejection is also characterised by vascular invasion and bone resorption. Infiltrations by round cells and macrophages are other immunological reactions. These cells can transform into osteoclasts since they have common stem cells with osteoclasts [11, 17]. By the production of cytokines such as interleukins, prostaglandins, proliferation and growth factors, immunological effector cells influence angiogenetic, osteoclastic and osteoblastic processes [6, 14, 18, 20, 21]. In this way immunocompetent cells can play an important role in bone-remodelling and the creeping substitution of grafted bone by newly formed bone.
1. Aebi M, Regazzoni P, Schwarzenbach O (1989) Segmental bone grafting. Comparison of different types of grafts in dog. Int Orthop 13 : 101-111 2. Arata MA, Wood MB, Cooney WP (1984) Revascularized segmental diaphyseal bone transfers in the canine. J Reconstr Microsurg 1 : 11-19 3. Braun C, Dambe L, Seiler H, Btihren V (1989) Mit allogenem Knochen armiertes myocutanes composite graft zum mikrochirurgischen Transfer. Hefte Unfallheilkd 207 : 258 4. Burchardt H (1983) The biology of bone graft repair. Clin Orthop 174 : 28-42 5. Burchardt H, Glowczewskie FP, Enneking WF (1977) Allogeneic segmental fibular transplants in azathioprine immunosuppressed dogs. J Bone Joint Surg [Am] 59 : 881-894 6. Centrella M, McCarthy TL, Canalis E (1991) Transforming growth factor-beta and remodeling of bone. J Bone Joint Surg [Am] 73 : 1418-1428 7. Delloye C, Nayer P de, Allington N, Munting E, Coutelier L, Vincent A (1988) Massive bone allografts in large segmental defects after tumor surgery: a clinical and microradiographic evaluation. Arch Orthop Trauma Surg 107 : 31-41 8. Dodd CAF, Fergusson CM, Freedman L, Houghton GR, Thomas D (1988) Allograft versus autograft bone in scoliosis surgery. J Bone Joint Surg [Br] 70:431-434
254 9. Gebhardt MC, Roth YF, Mankin HJ (1990) Osteoarticular allografts for reconstruction in the proximal part of the humerus after excision of a musculoskeletal tumor. J Bone Joint Surg [AmI 72:334-345 10. Gotfried Y, Yaremchuk MJ, Randolph MA, Weiland AJ (1987) Histological characteristics of acute rejection in vascularised allografts of bone. J Bone Joint Surg [Am] 69 : 410-425 11. Hentunen TA, Tuukanen J, V~i~n~inenHK (1990) Osteoclasts and small populations of peripheral blood cells share common surface antigens. Calcif Tissue Int 47: 8-17 12. Lord C, Gebhardt M, Tomford W, Mankin HJ (1988) Infections in bone allografts - incidence, nature and treatment. J Bone Joint Surg [Am] 70: 369-376 13. Mankin HJ, Doppelt S, Tomford W (1983) Clinical experience with allograft implantation. Clin Orthop 174: 69-86 14. Murray DW, Rushton N (1990) The effect of strain on bone cell prostaglandin E2 release: a new experimental method. Caldf Tissue Int 47 : 35-39 15. Ostrup LT, Fredrickson JM (1974) Distant transfer of free living bone graft by microvascular anastomosis. An experimental study. Plast Reconstr Surg 54: 274-285 16. Ottolenghi CE (1972) Massive osteo- and osteoarticular bone grafts: technique and results of 62 cases. Clin Orthop 87:156164
17. Owen M (1978) Histogenesis of bone cells. Calcif Tissue Res 25 : 205 18. Pfeilschifter J, D'Souza SM, Mundy GR (1987) The effect of transforming growth factor beta on osteoblastic osteosarcoma cells. Endocrinology 121 : 212-218 19. Rahn BA, Fleisch H, Moor R, Perren SM (1970) The effect of fluorescent labels on bone growth and calcification tissue culture. Eur Surg Res 2 : 137-138 20. Raisz LG, Martin TJ (1984) Prostaglandins in bone and mineral metabolism. In: Peck WA (ed) Bone and mineral research. Ann 2, Elsevier, Amsterdam, pp 286-310 21. Ren W, Dziak R (1991) Effects of leukotrienes on osteoblastic cell proliferation. Calcif Tissue Int 49 : 197-201 22. Shaffer J, Field GA, Goldberg VM, Davy DT (1985) The fate of vascutarised and nonvascularised autografts. Clin Orthop 197 : 32-43 23. Taylor GI, Miller GDH, Ham FJ (1975) The free vascularised bone graft. A clinical extension of microsurgical technique. Plast Reconstr Surg 55 : 533-544 24. Weiland AJ, Moore JR, Daniel RK (1983) Vascularised bone autografts. Clin Orthop 174: 87-95 Received December 30, 1991
