The morphological and physical aspects of cortical bone autografts implanted in dogs for 1-9 months in two differently located skeletal defects are reported with a twofold aim: to provide a reference system for further comparison with various allografts and to delineate a general pattern of cortical bone graft healing. A 3-cm osteoperiosteal gap was created in the diaphyseal segment of the ulna and fibula of mature dogs. The grafts, freed from periosteum and bone marrow, were then inverted and replaced for the autografts in the left limb bone without internal fixation or external splints. On the right side, different allografts were tested. A group of three animals also had an unfilled segmental resection on the right as control. Dogs were observed for 1, 2, 3, 6, and 9 months and were able to bear weight within 3 days. Twenty-eight ulnae and 27 fibulae were available for this autograft study. Fluorochromes were injected at mid-term and at the end of the observation. All the grafts were assessed morphologically by cross-section microradiographs and ultraviolet light microscopy, and a morphometric analysis for porosity and fluorescence was done. To evaluate the physical aspects of graft healing, the recovered ulnar autografts, when available, were submitted to photon absorptiometry and to torsional loading. Morphologically, resorption was found to invade the cortical bone graft transversely through radial tunnels, and in addition to the host-bone-graft junction, the entire transplant surface provided another way for revascularization. The highest porosity level was achieved 2 months after surgery for both ulna and fibula, while new bone formation, as assessed by fluorochromes, was most important at 3 months. At 9 months, porosity remained above the normal range as determined in a set of five nongrafted dogs. While the lack of correlation for porosity between the two grafts suggests that local factors are more important in graft resorption, the observed correlation for fluorescence indicates that new bone deposition is more dependent upon skeletal metabolic activity. Within each graft, porosity and new bone formation were not well correlated. In the ulna, the bone mineral content (BMC) reflected the graft volumetric variations during the remodeling, with the lowest mean value at 3 months. For each graft, BMC was well correlated with the torsional stiffness. When torsionally loaded, the maximal tangential shear stress at failure of the graft was negatively related to its cortical porosity.(ABSTRACT TRUNCATED AT 400 WORDS)
