The goal of this study was to investigate initial bone response to the presence of the endoprosthesis and investigate the effects of low intensity vibration therapy on this bone response. We hypothesized that a press fit between the endoprosthesis and bone would provide stable fixation and that low intensity vibration would improve osseointegration without jeopardizing mechanical stability. To test this hypotheses a press fit, percutaneous endoprosthesis was successfully implemented in an ovine model to allow immediate weight bearing after the attachment of an exprosthesis. Animals were split into SHAM and LIV groups. The LIV group received vibration treatment by standing on a platform that produced 0.3 g of acceleration at a frequency of 45 Hz for 10 minutes a day, five days a week. The SHAM group stood on a platform that did not produce any vibrations. Nine animals completed a 12 week study with implants that were well fixed to the bone.

Gait data were recorded before surgery (week 0) and at weeks 4, 8, and 12. Computed tomography (CT) scans were obtained at weeks 0 and 12 and dexamorphometry (DXA) was performed at weeks 4, 8 and 12. Fluorochrome labels were administered at the beginning and end of weeks 4 and 8. After animals were sacrificed, the implanted bone was sectioned for histomorphometry and mechanical push-out testing. Fluorescent microscopy was performed to image the fluorochrome labeling after which the slides were stained and imaged again. Mechanical push-out testing was performed to determine the bone-endoprosthesis interface strength. MicroCT scans were performed on the push-out sections before and after testing to evaluate three dimensional morphometry and measure the amount of osseointegration.

The animals with osseointegrated endoprostheses at the end of the 12 week study displayed strong bone-endoprosthesis interface strength as determined by mechanical push-out testing. Animals with stable endoprosthesis fixation at 12 weeks did not display statistically significant lameness as determined by gait asymmetry. Asymmetrical gait was shown to be an indication of osseointegration failure. Finite element modeling predicted that bone-endoprosthesis press fits greater than 50 μm would cause bone damage at endoprosthesis implantation and the amount of damage would increase over the subsequent 24 hours after surgery. The finite element model predicted areas of damage seen in the animal model, giving the computation model and hoop strain based failure criterion experimental validity.

LIV was shown to increase osseointegration between a press-fit, percutaneous, skeletally anchored endoprosthesis for prosthetic attachment by 18% by stimulating new bone growth in the intracortical region, and no adverse effects to this treatment were found. An interaction effect between the experimental group (SHAM or LIV) and the amount of osseointegration was found to influence the bone-implant interface mechanics. DXA data was normalized by the depth of the bone to obtain bone mineral apparent density (BMAD) data. BMAD in the AP view for the LIV group remained constant between weeks 0 and 8 and then increased between weeks 8 and 12. Newly formed bone tissue in the intracortical compartment was shown to have less mineralization than the surrounding cortical bone, indicating that the length of the study was insufficient to determine if LIV treatment will result in increased bone mineral density in the implanted bone. LIV was not found to increase bone remodeling rates at weeks 4 or 8. Due to the positive results found in this study and lack of adverse effects to LIV treatment, a longer study with more subjects should be performed to evaluate a new endoprosthesis design and to determine the long term effects of LIV on endoprosthesis stability and bone quality in the residual limb.