An animal model was developed to simulate an unstable lumbar burst fracture that had been treated with corpectomy. A fifth lumbar laminectomy, partial facetectomy, and corpectomy was performed in twenty-one dogs. In seven dogs (the control group), a biodegradable polymer spacer was used to create a definite failure of fusion (Group I). Seven dogs were treated with a traditional anterior arthrodesis with an autogenous ulnar strut and without instrumentation (Group II). The remaining seven dogs were treated with an ulnar strut and anterior Kaneda instrumentation that was of an appropriate size for the dog (Group III). At twenty-four weeks, the results were analyzed in terms of the rate of fusion, biomechanical rigidity, neuropathological findings, and histomorphometric data on the vertebral response. The rate of fusion was significantly higher in Group III, in which the Kaneda device had been used, than it was in either Group I or Group II, in which instrumentation had not been used. Biomechanically, the spines in Group III were stiffer in torsion than those in Group I or II. There was no difference between groups in terms of the number of neuropathological changes in the spinal cord. Histomorphometric analysis showed that no meaningful device-related osteopenia occurred in the vertebrae that were spanned by the fixation device. Trabecular density was increased in the vertebrae in which the instrumentation was anchored.