Contrasting opinions exist in terms of preferred or ideal management of burst fractures located at the thoracolumbar region without neurological deficit. In 2003, a prospective study was conducted comparing surgical and non-surgical outcomes for patients suffering from thoracolumbar burst fractures without neurologic deficit. Results demonstrated that there was no significant differences between the two groups with respect to pain, return to work, pain and functional disability. Conflicting positions have been found in other studies. With review of these varying positions it is thought principal differences among the reports may be attributed, in part, to variances in follow-up duration among the studies. The authors of the 2003 study wished to conduct a fifteen-twenty year follow-up of their original patients to find out how their results would pan out over a longer term follow-up. They hypothesized that the original findings would be upheld. Nineteen of the original twenty-four patients treated operatively and eighteen of the original twenty-three patients treated non-operatively were contacted and follow-up data was obtained. The average duration of long-term follow-up was 18.6 years.
Radiologic findings demonstrated that average kyphosis remained thirteen degrees at long-term follow-up in the operative group. No correlation was found between the final rate of kyphosis and degree of pain reported or disability. In the non-operative treatment group the average kyphosis was fourteen degrees at four-year follow-up and at longer term follow-up the average kyphosis was nineteen degrees. Similar to the surgically treated group, no correlation was found between final kyphosis reported and pain reported or disability.
Median pain scores for long-term follow-up demonstrated an average of four cm on the ten cm visual analog scale for those treated surgically and 1.5 cm for those treated non-operatively with a cast or brace. At long-term follow-up there was clinical significance between difference of pain scores between those treated non-operatively and those treated operatively. The functional disability outcome for patients treated operatively measured using the Roland and Morris functional disability score was seven on a scale of twenty five (with zero measuring no disability while twenty-five indicates complete disability). In the patient group treated non-operatively the median score was one. This difference demonstrated clinical significance. There was very little change in the median score over the years in the group treated operatively, however, there was clinically significant improvement in this score within the group treated non-operatively. Median scores on the Oswestry questionnaire at long-term follow-up was twenty for the operative group and two for nonsurgical treatment group. The difference between groups was significant but within each treatment group there was very little change overall throughout the years. Long-term follow-up scores from the short-form-36 showed that nonoperative management was favored to a significant degree.
With regards to return to work status, 58 per cent of patients treated surgically had returned at the four-year point, while long-term follow-up demonstrated that 47 per cent remained employed. Six patients in this group had voluntarily retired and four had lost employment. Of the patients treated non-operatively, 83 per cent were able to return to work at intermediate follow-up. At long-term follow-up 72 per cent were still working while three retired and two lost employment.
The authors of this study conclude that non-operative treatment was the optimal management option for patients with a thoracolumbar burst fracture without neurologic deficit.