Conservative treatment of thoracolumbar burst fractures: a long-term follow-up results with special reference to the load sharing classification

Study Design. A retrospective study to review patients with acute thoracolumbar burst fractures who were conservatively treated. Objectives. The purpose of this study was to analyze the efficacy of conservative treatment of thoracolumbar burst fractures in terms of long-term results, and to determine whether the long-term results of conservative treatment would be predicted by the Load Sharing Classification. Summary of Background Data. The optimal management of thoracolumbar burst fractures remains controversial, with advocates for nonoperative or operative approaches. Different criteria for treatment options have been advocated for classifying the fractures, guiding treatment decision-making, and predicting the prognosis. Methods. A total of 127 patients with an acute thoracolumbar burst fracture (Denis type B) were treated conservatively. Their Load Sharing score ranged from 3 to 9 (average, 5.3). Twenty-two (17.3%) of these patients presented with a neurologic deficit. They were followed up for at least 3 years (range, 3-12 years; average, 7.2 years) and assessed for clinical and radiologic outcomes. The motor score of the American Spinal Injury Association, and Denis pain and work scales were used for clinical evaluation, whereas the loss of kyphosis correction (LKC) for the local kyphosis angle was examined for radiologic outcome. Results. None of the patients had neurologic deterioration during the reduction. All 22 patients with neurologic deficit experienced improvement with average recovery rate of 93%. Significant correlation (P < 0.05) was found between the Load Sharing score on admission and the LKC for local kyphosis angle at final follow-up. Most patients reported acceptable results regarding pain (111/ 127, 87.4%) and work (110/127, 86.6%). A positive correlation was found between the LKC and the severity of pain (r = 0.194, P < 0.05). Seven patients (5.5%) required or had undergone late surgery for correction of loss-related back pain and/or poor functional outcome. Conclusion. Conservative treatment is safe and effective for selected patients with thoracolumbar burst fractures, even in some cases with neurologic deficit. The Load Sharing Classification could be used for guiding the treatment of thoracolumbar burst fractures not only in surgical approach choice but also in surgical decision-making.