Roy, Robert G.

Summary Modular, porous-coated, titanium segmental endoprostheses were implanted bilaterally in the femoral diaphysis of 7 adult mixed-breed dogs. Autogenous bone graft in particle form was placed around the implant and bone. In 1 limb, homologous fibrin adhesive was mixed with the graft in situ before soft tissue closure. The contralateral limb was grafted in identical manner, but without fibrin adhesive, and served as a control. Radiography was performed immediately after surgery and 1, 2, 3, 4, 6, 8, 10, and 12 weeks later to assess callus area and bone remodeling. At 12 weeks, dogs were euthanatized and bone/implant fixation strength was tested under torsion and compared with values for 6 in vitro controls. Histomorphometric and microradiographic analyses of transverse sections of the distal portion of the implanted femurs were performed. Radiographic callus area was significantly (P < 0.05) smaller in the femurs grafted with fibrin adhesive, compared with the contralateral control. New bone formation (21.4 ± 1.8% vs 19.2 ± 2.4%), unlabeled bone (64.8 ± 3.0% vs 67.9 ± 4.2%), porosity (13.9 ± 0.7% vs 12.9 ± 0.8%), and bone ingrowth into the porous coating (10.3 ± 0.9% vs 10.0 ± 1.2%) were not significantly different between fibrin- and nonfibrin-grafted implants, respectively. There were no significant differences in torsional strength of implant fixation between the fibrin- and nonfibrin-grafted femurs or between the in vivo implanted femurs and the in vitro controls. These data indicate that fibrin adhesive may have been advantageous in maintaining apposition of bone graft adjacent to the endoprosthesis, but it probably did not have an enhancing effect on extracortical bone bridging or ingrowth over a porous-coated segmental bone replacement endoprosthesis.

Summary Quantitative computed tomography has been used extensively to measure bone mineral density; particularly in the vertebral column and in the proximal portion of the femur in human beings with osteoporosis. Other potential applications of this technique include evaluation of bone adjacent to metallic endoprostheses and evaluation of fractures as they heal. Unfortunately, metal causes severe image degradation, principally seen as starburst streaking. One method used to decrease these artifacts is by imaging less-attenuating materials, such as titanium alloy. Titanium decreases image degradation sufficiently to allow accurate determination of the geometric properties of cadaveric bone. In our study, the effect of a titanium segmental endoprosthesis on bone mineral density measurement was determined by use of bone specimens from dogs and calibration standards. Titanium decreased the bone mineral density of calibration solutions from 6.8 (500 mg/cm3) to 17.7% (250 mg/cm3), and increased bone mineral density of cortical bone by 5.3%. Titanium did not affect the repeatability of these scans, indicating that the error caused by titanium was systematic and can be corrected. Our data were suggestive that quantitative computed tomography can be used to measure bone mineral density of cortical bone adjacent to titanium endoprostheses, with a predictable increase in density measurement.