Fluid conductance of cancellous bone graft as a predictor for graft-host interface healing

The aim of this study is to test the hypothesis that the fluid conductance of cancellous bone graft is a deciding factor in graft-host union. Cylindrical cancellous bone specimens were trephined from fresh porcine femoral heads in a direction either parallel or perpendicular to the femoral neck axes. The graft after a defatting and freeze-drying process was placed in a perfusion apparatus. The pressure drop across the bone graft and the induced flow were measured and the conductance to fluid how was calculated as the slope of the how-pressure relation. Grafts of different flow conductance were transplanted into 35 rabbits to replace segments of tibiae. Nine weeks after grafting, 29 rabbits which had completed the follow-up without incidents were sacrificed. The decalcified sections of the grafts and the adjacent tibiae were examined microscopically for histological events during graft-host interface healing. Perfusion data indicated that fluid conductance ranged from 0.05 x 10(-10) to 13.4 x 10(-10) m(3) s(-1) Pa-1 and was higher in the direction parallel to the femoral neck axis, inversely proportional to the length of the graft and directly proportional to the square of porosity. Data from rabbit model supported the hypothesis stated above. A threshold conductance was found to be 1.5 x 10(-10) m(3) s(-1) Pa-1, below which revascularisation and the formations of osteoblasts and fibrous tissues could not be attained. How the low conductance led to non-union at the graft-host interface was briefly discussed.