| Native bovine bone morphogenetic protein in the healing of segmental long bone defects | ||
|---|---|---|
| Prev | Chapter 2. Review of the literature | Next |
Canine ulnar segmental defect is a well-established model. The dog ulna is not directly a weight-bearing bone, as the radius gives some support to the ulna, and there has been some controversy about the fixation methods. The ulnar defect model has been used with no fixation at all (Nilsson et al. 1986, Delloye et al. 1992, Cook et al. 1994b) or with an intramedullary Steinmann pin (Moore et al. 1987, Grundel et al. 1991) or plate fixation (Johnson et al. 1989, Schwarz et al. 1991).
Key (1934) was the first to use the segmental defect of canine ulna, and his observation was that a defect 1.5 times the ulnar diameter left empty leads to non-union. Key observed that the insertion of boiled bone, bone powders, calcium salts and other non-viable fillers into the defect produced non-union, while an autogeneic bone graft generally produced solid union.
Heiple et al. (1963) used the same model to investigate the process of regeneration in defects, concluding that autogeneic bone was superior to allogeneic bone and to demineralized bone matrix.
Autografts have also been found to be superior to allografts and demineralized bone matrix in other studies (Schwarz et al. 1991, Delloye et al. 1992). In the former, the ulnar defect was temporarily filled with silicone rubber blocks for eight weeks, which were then replaced by bone grafts. After 24 weeks, only the autogeneic bone had led to healing in all instances. Bone regeneration was not significantly better than in the sham group, in which no graft was employed. The results of Delloye et al. (1992) showed that autografts achieved a better union score and were mechanically stronger than allografts, but intracortical bone porosity, the percentage of cumulative new bone and the mineral apposition rate were not variables with statistical significance. In an earlier study, they had demonstrated notable variability of healing patterns in canine ulnar segmental defects and the long-term nature of the healing of cortical autografts, which was not completed at 9 months (Delloye et al. 1986).
Ceramics alone and in different combinations have been used as bone substitutes in canine ulnar defects. Moore et al. (1987) used a mixture of hydroxyapatite-tricalcium phosphate (HA-TCP) ceramic alone and with autograft cancellous bone, comparing these to autograft bone. Autograft and a combination HA-TCP with autograft showed good bone healing, while HA-TCP alone was not osteoinductive. The authors concluded that morselized HA/TCP promises to be useful as a graft extender when mixed with autogenous cancellous bone.
Grundel et al. (1991) used the same model with HA-TCP with autogeneic bone marrow compared to autogeneic bone marrow alone. Both groups resulted in good bone healing, and HA-TCP combined with bone marrow resulted in complete bridging significantly earlier than bone marrow alone.
Guillemin et al. (1987) treated small cortical defects (5x8 mm) of canine ulna with natural coral. The results showed continuous resorption of coral implants and filling of defects with new bone by 8 weeks.