3-D Visualization and Prediction of Spine Fractures

点击翻译按钮获取中文摘要

Vertebral fractures are the most common type of osteoporotic fracture, afflicting approximately one in three women and one in six men over the age of 50. Despite their high prevalence, sensitive and specific estimates of vertebral fracture risk have remained elusive. This is due in large part to the limited accuracy and precision of current methods of estimating vertebral strength. Average measures of bone mineral density (BMD) explain only 50-70% of the variance in vertebral strength, a result that is not surprising given the heterogeneous distribution of bone tissue throughout the vertebra. A growing and compelling amount of evidence points to importance of this heterogeneity in governing the mechanical behavior of the vertebra. Recent advances in quantitative computed tomography (QCT) allow non-invasive measurement of the distribution of bone density and even trabecular anisotropy in whole bones. We propose that these additional measurements can be used to establish a new standard for clinical evaluation of vertebral fracture risk. Our overall hypothesis is that CT- based methods that account for the heterogeneous distribution of density and trabecular anisotropy throughout the vertebra provide more accurate predictions of vertebral strength than do methods based solely on average BMD. Four specific aims are proposed. Aim #1 will test whether CT-based measures of the intra-vertebral heterogeneity in density are independent predictors of vertebral strength. Aims #2-#4 are closely coupled experimental and computational studies that will test the importance of incorporating specimen-specific, anisotropic material properties in QCT-based finite element (FE) models of the vertebra. These studies will investigate the effect of this material property assignment on the accuracy of the FE predictions of vertebral strength and failure behavior. Aim #2 will use micro-finite element analysis to quantify the anisotropic elastic properties throughout the centrum. Aim #3 will carry out the QCT-based FE analyses using the material properties obtained in Aim #2 and also using properties determined purely from estimates based on BMD or on BMD and trabecular anisotropy. The accuracy of the FE predictions of vertebral mechanical behavior will be evaluated through experiments performed in Aim #4. These experiments will use 3-D failure visualization techniques that we have developed over the past several years. These techniques afford us the unique ability to assess the fidelity with which the FE models predict bone strength as well as the true deformation and failure behavior of the vertebra. Such assessment is critical for gauging the performance of these models, for identifying means of improving their predictions, and for enabling their widespread implementation in the clinical arena. Taken together, the proposed studies constitute a set of concrete and consequential steps towards our long-term goal of developing techniques for highly accurate, patient-specific predictions of vertebral strength from clinically feasible measurements. As such, this work has strong potential for leading the way to better diagnosis, treatment, and prevention of spine fractures. One in three women and one in six men over age 50 will suffer a spine fracture in their remaining lifetime. This project focuses on developing methods for obtaining more accurate predictions of bone strength in the spine.