NEW METRIC FOR QUANTIFYING STIFFNESS OF INTACT HUMAN VERTEBRAE

摘要

Our results indicate that whole bone stiffness determined from experimental force-deformation data is inherently sensitive to its method of calculation. To solve this problem, we have developed and validated a standard numerical method based on minimizing the error inherent to the calculation method itself. Similar to a previously published technique, our method involves determining the maximum stiffness over a sliding strain range. However, our recommended range (1.0%-1.5%) is substantially greater than has been used previously (0.2%). Sensitivity analysis showed that there is a high calculation error (mean 20%) at the lower strain range, while our larger strain range results in more appropriate levels of uncertainty for calculated stiffness measures (mean 3%). We also determined that stiffness values calculated over physiologic force ranges may include regions of inelasticity for vertebrae with low fracture strengths. Furthermore, for some specimens, force-based calculations are equivalent to strain-based measurements with ranges that we determined to be non-optimal (≤ 1.0%) and thus have a higher associated calculation error.These results suggest that previous whole bone stiffness measurements from in vitro biomechanical testing of human vertebrae either under or over-estimated true stiffness by 20% on average. We recommend that future stiffness calculations be made using the validated technique presented in this study. Standardizing stiffness calculations according to this method will improve the accuracy of inter-product comparisons for orthopaedic products, e.g.. vertebroplasty. as well as improve our understanding of the effects of pathological conditions such as osteoporosis on vertebral structural integrity.