Stability of the metastatic spine pre and post vertebroplasty.

OBJECTIVE Vertebrae with lytic metastases have an elevated risk of burst fracture and resultant neurologic compromise. Prophylactic vertebroplasty has the potential to reduce pain and the risk of burst fracture in the metastatic spine. The purpose of this study was to quantify the ability of vertebroplasty to stabilize metastatically involved vertebrae against the risk of burst fracture initiation with a standardized model of vertebral metastases. METHODS Metastases were simulated in eight fresh-frozen cadaveric thoracolumbar spinal motion segments by removing a central core of trabecular bone and filling the defect with tumor tissue. Specimens were tested under a physiologic level of axial compression, intact, with a simulated tumor and post-vertebroplasty, and ultimately tested to failure. Axial load induced canal narrowing (CN) was used as a measure of the risk of burst fracture initiation. Following testing, vertebrae were axially sectioned to visualize cement fill. RESULTS Vertebrae with simulated metastases exhibited significantly higher CN than intact specimens (227%+/-109%; P<0.05). Post vertebroplasty, three vertebrae exhibited reduced CN compared with the simulated tumor configuration, whereas the other five had increased CN. Specimens with reduced CN were found to have cement posterior to the tumor, whereas specimens with an increase in CN had cement anterior and lateral to the tumor only. Percutaneous vertebroplasty is effective in decreasing CN if tumor is surrounded posteriorly with cement. However, injecting cement into the posterior third of the vertebral body is risky due to potential extravasation into the canal. CONCLUSION Future work aimed at improving cement fill is necessary for safe and consistent stabilization of the metastatic spine with vertebroplasty.