Towards Improved Assessment of Bone Fracture Risk

The mechanical competence of a bone depends on its density, its geometry and its internal trabecular microarchitecture. The gold standard to determine bone competence is an experimental, mechanical test. Direct mechanical testing is a straight-forward procedure, but is limited by its destructiveness. For the clinician, the prediction of bone quality for individual patients is, so far, more or less restricted to the quantitative analysis of bone density alone. Finite element (FE) analysis of bone can be used as a tool to non-destructively assess bone competence. FE analysis is a computational technique; it is the most widely used method in engineering for structural analysis. With FE analysis it is possible to perform a 'virtual experiment', i.e. the simulation of a mechanical test in great detail and with high precision. What is needed for that are, first, in vivo imaging capabilities to assess bone structure with adequate resolution, and second, appropriate software to solve the image-based FE models [1]. Both requirements have seen a tremendous development over the last years. The last decade has seen the commercial introduction and proliferation of non-destructive microstructural imaging systems such as desktop micro-computed tomography (µCT), which allow easy and relatively inexpensive access to the 3D microarchitecture of bone [2]. Furthermore, the introduction of new computational techniques has allowed to solve the increasingly large FE models, that represent bone in more and more detail [3, 4]. With the recent advent of microstructural in vivo patient imaging systems, these methodologies have reached a level that it is now becoming possible to accurately assess bone strength in humans. Although most applications are still in an experimental setting, it has been clearly demonstrated that it is possible to use these techniques in a clinical setting [5]. The high level of automation, the continuing increase in computational power, and above all the improved predictive capacity over predictions based on bone mass, make clear that there is great potential in the clinical arena for in vivo FE analyses Ideally, the development of in vivo imaging systems with microstructural resolution better than 50 mm would allow measurement of patients at different time points and at different anatomical sites. Unfortunately, such systems are not yet available, but the resolution at peripheral sites has reached a level (80 mm) that allows elucidation of individual microstructural bone elements. Whether a resolution of 50 mm in vivo will be reached using conventional CT technology remains to be seen as the required doses may be too high. With respect to these dose considerations, MRI may have considerable potential for future clinical applications to overcome some of the limitations with X-ray CT. With the advent of new clinical MRI systems with higher field strengths, and the introduction of fast parallel-imaging acquisition techniques, higher resolutions in MRI will be possible with comparable image quality and without the adverse effects of ionizing radiation. With these patient scanners, it will be possible to monitor changes in the microarchitectural aspects of bone quality in vivo. In combination with FE analysis it will also allow to predict the mechanical competence of whole bones in the course of age- and disease-related bone loss and osteoporosis. We expect these findings to improve our understanding of the influence of densitometric, morphological but also loading factors in the etiology of spontaneous fractures of the hip, the spine, and the radius. Eventually, this improved understanding may lead to more successful approaches in the prevention of age- and disease-related fractures.