Abstract
Worldwide, osteoporosis, causes more than 8.9 million fractures annually, resulting in an osteoporotic fracture every 3 seconds, where 1 in every 3 women and 1 in every 5 men aged over 50 will experience osteoporotic fractures at least once in their lifetime. Vertebral fractures, estimated at 1.4 million/year are among the most common fractures, posing enormous health and socioeconomic challenges to the individual and society at large. Considering that the great majority of individuals at high risk (up to 80%), who have already had at least one osteoporotic fracture, are neither identified nor treated, prediction of the risk factors for vertebral fractures can be of great value for prevention/early diagnosis. Recent studies show that finite element analysis of computed tomography (CT) scans provides noninvasive means to assess fracture risk and has the potential to be clinically implemented upon proper validation. The objective of this study was to develop a voxel-based finite element model using quantitative computed tomography (QCT) images in conjunction with in-vitro experiments to evaluate the strength of the vertebral bodies and predict the fracture risk criteria. A total of 10 vertebrae were dissected from juvenile sheep lumbar spines. The attached soft tissues and posterior elements and facet joints were completely removed, and the upper and lower vertebral bodies were polished using glass paper to provide smooth surfaces. The specimens were wrapped in phosphate buffer saline (PBS) soaked gauze, sealed in plastic bags, and stored in a refrigerator at −22°C. QCT scans of the specimens were captured using a bone density calibration phantom (QRM Co., Moehrendorf, Germany) with three 18 mm cylindrical inserts, providing 0, 100 and 200 mg HA/ccm, respectively. All the specimens, preserved hydrated in PBS solution, were mechanically tested at room temperature using a mechanical testing apparatus (Zwick/Roell, Ulm-Germany). The QCT images were then used to reconstruct the voxel-based FE model employing a custom-developed heterogeneous material mapping code. Five different equations for the correlation of the density and the elastic modulus were used to validate the efficiency of the FE model as compared to the in-vitro experiments. The results of the voxel-based FE models matched well with the in-vitro experiments, with an average error of 11.38 (±4.09)% based on the power law equation. A failure criterion was embedded in the FE models and the initiation of fracture was successfully predicted for all specimens. Further, typical kyphoplasty treatment was simulated in the 5 models to evaluate the application of the validated algorithm in the estimation of the failure patterns. Our novel voxel-based FE model can be used in future studies to predict the outcome of different types of therapeutic modalities/surgeries and estimate fracture risk including postoperative fractures.
