Predicting failure of thoracic vertebrae with simulated and actual metastatic defects.

Indications for operative treatment in spinal metastatic disease depend on estimates of vertebral loadbearing capacity. There are no noninvasive diagnostic tools for estimating vertebral loadbearing capacity in the presence of a metastatic lesion. Thus, relationships between vertebral failure load and measurements from computed tomography data were investigated to determine if measurements that account for defect size and bone density can predict loadbearing capacity better than can defect size alone. Cylindrical defects were created in thoracic vertebrae of 20 anatomic specimen spinal segments, with 10 other segments serving as controls. Five vertebrae with actual metastatic defects also were tested. Vertebrae were scanned using quantitative computed tomography, and the defect size and the axial rigidity of the midvertebral cross section were calculated using an image analysis program. The spinal segments were tested to failure using a combination of axial compression and anterior flexion. Linear regressions between axial rigidity and absolute failure load showed a high positive correlation, but there was no correlation between defect size and failure load. Axial rigidity may prove useful as a noninvasive assessment of vertebral loadbearing capacity in patients with spinal metastatic disease.