Vertebroplasty is a treatment for vertebral compression fractures in which cement is injected into the vertebral body to relieve pain and stabilise the fracture site. Conflicting reports in the literature as to its efficacy indicate that further biomechanical evaluation of vertebroplasty is necessary to optimise the treatment variables (cement injection location, volume and composition) and to better understand which patients vertebroplasty will benefit. Finite element (FE) methods provide a means by which this can be undertaken under controlled conditions which are not possible experimentally, but existing vertebral FE models poorly reproduce the behaviour of cement-augmented vertebrae. The aim of this study was to develop an improved, clinically practical FE method of representing the behaviour of the interface between the bone and cement augmentation. Appropriate homogenous finite element (hFE) micro-computed tomography (μCT) greyscale-modulus and yield strain relationships were derived for un-augmented ovine lumbar vertebral trabecular bone. Similar ovine vertebral bone specimens were then fractured and augmented with poly(methyl methacrylate) cement, and novel methods and equipment were developed to enable the imaging of these specimens using μCT as they were deformed to failure in axial compression. Proprietary software was then used to determine the specimen strain distribution from the images. hFE models that incorporated an explicitly modelled interfacial region were generated from the images and parametric studies undertaken to derive the most appropriate interfacial properties. Good agreement with the corresponding load-displacement and strain distribution data was achieved. Finally, a preliminary study was conducted in which the new method of representing the interface was incorporated into existing hFE models of whole cement-augmented vertebrae. The predicted strain-distribution seen within the modified whole vertebral models more closely matched the behaviour of the earlier interfacial specimens, though this has yet to be validated experimentally against cement-augmented whole vertebrae.

No description available.

There is an increasing demand for spinal surgery and a growing number of new spinal implants and surgical procedures being offered by orthopaedic companies. However, the testing of spinal implants and spinal instrumentation is problematic, with testing in cadavers and animals becoming increasingly difficult and both having significant limitations. Thus the aim of this research is to develop an artificial laboratory spine that will have the same physical and biomechanical properties as the human spine. Validation of computer model is difficult hence an active artificial laboratory spine is being developed. A number of spinal elements have been produced and are being investigated, including an artificial intervertebral disc with different material properties to allow it to simulate different clinical conditions. The study is the first of its kind with the characteristics of the disc material that have been assessed in the laboratory, artificial muscles and spring elements and with normal physiological movements compared and validated from the reported literature. The model was used to investigate the potential of Shape Memory Alloy wires to act as artificial muscles and to control the movement of the spine. It is anticipated that the laboratory spine will have a number of other applications, in particular in the assessment of spinal instrumentation and testing. An actual geometry laboratory spine is also generated with suitable manufacturing technique for intervertebral disc, which has an accurate surface profile to fit between the two vertebral bodies above and below it, is discussed in this thesis.