Thoracolumbar burst fractures. The clinical efficacy and outcome of nonoperative management.

There continues to be considerable controversy regarding the management of thoracolumbar burst fractures. Most feel that failure of the middle osteoligamentous complex, particularly with retropulsion of fragments into the spinal canal, is an indication for operative management. Others advocate postural reduction and prolonged bedrest for such injuries. The purpose of this study was to 1) review the clinical outcome and efficacy of closed management of thoracolumbar burst fractures; and 2) quantify what, if any, remodeling occurs in the bony canal as measured by serial CT. Forty-one patients who presented with a burst fracture of the thoracolumbar spine without neurologic deficit were reviewed clinically and radiographically following nonoperative management. At injury, canal compromise averaged 37% (range, 16-66%); 26 patients had at least 30% canal compromise. During treatment, one patient developed neurologic deterioration that prompted surgery; all other patients remained neurologically intact. At average follow-up of 2 years, an overall outcome evaluation indicated that 49% of the patients had excellent outcomes relative to pain and function; 17%, good; 22%, fair; and 12%, poor. Approximately 90% of the patients had a satisfactory work status relative to factors associated with their burst fracture. Serial roentgenograms documented significant progression in body collapse, which averaged 8% (P < 0.0001) from injury to follow-up. On the other hand, serial CTs documented significant improvement from injury to follow-up for canal compromise and midsagittal diameter. Average improvements in canal compromise and midsagittal diameter were 22% (P < 0.0001) and 11% (P < 0.0001), respectively. Only three patients had canal compromise greater than 30%, no patients had canal compromise greater than 40%, and no patients experienced canal area deterioration over time. On average, nearly two-thirds of the fragment occluding the canal resorbed, with most remodeling complete within one year. For patients with burst fractures presenting neurologically intact, we obtained the following findings: 1) nonoperative management yields acceptable results; 2) following nonoperative management, bony deformity (i.e., kyphosis and body collapse) progresses marginally relative to the rate of canal area remodeling; 3) incidence of subsequent neurologic deficits is quite low; and 4) initial radiographic severity of injury or residual deformity following closed management does not correlate with symptoms at follow-up. This pattern of results suggests nonoperative management as the preferred treatment in these circumstances.