Title. 3D distribution of cortical bone thickness in the proximal humerus, implications for fracture management. Introduction. CT imaging is commonly used to gain a better understanding of proximal humerus fractures. the operating surgeon however has a limited capacity to evaluate the internal bone geometry from these clinical CT images. our aim was to use clinical CT in a novel way of accurately mapping cortical bone geometry in the proximal humerus. we planned to experimentally define the cortico-cancellous border in a cadaveric study and use CT imaging software to map out cortical thickness distribution in our specimens. Methodology. With ethical approval we used fifteen fresh frozen human proximal humeri. These were stripped of all soft tissue and transverse CT images taken with a GE VCT Lightspeed scanner. The humeral heads were then subsequently resected to allow access to the methaphyseal area. Using currettes, cancellous bone was removed down to hard cortical bone. Another set of CT images of the reamed specimen were then taken. Using Mimics imaging software[Materialise, Leuven] and a CAD interface, 3-matic [Materialise, Leuven], we built 3D model representations of our intact and reamed specimens. We first had to define an accurate CT density threshold for visualising cortical contours. We then analysed cortical thickness distribution based on that experimented threshold. Results. we were able to statistically determine the CT threshold, in Hounsfield Units, that represents the cortico-cancellous interface in the proximal humerus. Our 3D colour models provide an accurate depiction of the distribution of cortical thickness in the proximal humerus. Discussion/Conclusions. Our Hounsfield value for the cortico-cancellous interface in the proximal humerus agrees with a similar range of 400 to 800 HU reported in the literature for the proximal femur. Knowledge of regional variations in cortical bone thickness has direct implications for basic science studies on osteoporosis and its treatment, but is also important for the orthopaedic surgeon since our decision for treatment options is often guided by local bone quality

Introduction. It was the purpose to evaluate the biomechanical changes that occur after optimal and non-optimal component placement of a hip resurfacing (SRA) by using a subject specific musculoskeletal model based on CT-scan data. Materials and Methods. Nineteen hips from 11 cadavers were resurfaced with a BHR using a femoral navigation system. CT images were acquired before and after surgery. Grey-value segmentation in Mimics produced contours representing the bone geometry and identifying the outlines of the 3 parts of the gluteus medius. The anatomical changes induced by the procedure were characterised by the translation of the hip joint center (HJCR) with respect to the pelvic and femoral bone. The contact forces during normal gait with ‘optimal’ component placement were calculated for a cement mantle of 3 mm, a socket inclination of 45° and anteversion of 15°. The biomechanical effect of ‘non-optimal placement’ was simulated by varying the positioning of the components. Results. There was a significant (p<0.01) shortening of the muscle length with the ‘optimal’ component placement for all parts of the gluteus medius with the largest shortening of the posterior part by 6mm. This was caused by a significant shortening of the femoral offset by 2.3mm (p<0.01). Because of a significant (p<0.01) medialisation of the HJCR by 4 mm, there was no significant increase in contact force. The hip joint contact forces increased by 0.5% per mm HJCR displacement. Each millimeter of cranial and lateral displacement of the femoral HJCR increased the contact force by 0.5% and 1%, respectively. The contact stresses changed significantly by 0.8% and 0.2% per degree of socket inclination and anteversion. The contact force increased 1% per mm lateral displacement of the acetabular HJCR. Discussion. Optimal placement of the SRA components did not completely restore the biomechanics of the native hip joint. The contact forces were not increased due to the compensatory effect of the medialisation of the acetabular HJCR. This suggests that reaming to the acetabular floor should be conducted in SRA. Femoral component displacement in the cranial and lateral direction significantly increased the hip joint loading. Errors of socket placement in the coronal and sagital plane significantly increased the contact stresses. Accumulative errors of both component displacements could lead to increased contact stresses of 18% to 23% with socket inclinations of 50° and 55°. Surgeons should reconsider continuing the SRA procedure if a neck length loss and lateralisation of the HCJR by >5 mm is anticipated as this would increase the contact stresses by >12%