Traumatic thoracic and lumbar spinal fractures: Operative or nonoperative treatment - Comparison of two treatment strategies by means of surgeon equipoise

Study Design. A center parallel cohort study with blinded inclusion based on clinical equipoise. Objective. To compare outcomes of nonoperative and operative treatment strategies in terms of quality of life and neurologic and functional status. Summary of Background Data. Despite a considerable body of literature, sound evidence regarding the optimal treatment for traumatic thoracic and lumbar spine fractures is lacking. Methods. Medical records of patients hospitalized for traumatic spinal fractures between 1991 and 2002 were identified in 2 trauma centers in the same country with established and different treatment strategies. Eligibility was retrospectively assessed for each case by a panel of orthopaedic surgeons who were representative of the 2 medical centers, and who were blinded to the treatment actually administered. Patients were included in the study when there was disagreement on the suggested treatment method. Thus, 2 comparable groups were identified undergoing nonoperative or operative treatment. Outcome assessment and comparison across groups focused on quality of life, residual pain, neurologic recovery, and employment in the middle-long-term follow-up. Results. Discordance in regards to choice of treatment was identified in 190 ( 95 treated nonoperative, 95 operative) of 636 potentially eligible patients. Patients were comparable regarding baseline characteristics, except for a somewhat higher proportion of males and neurologic impairment in the operative group. Seventeen percent of the nonoperative and 21% of the operative group developed complications and 3 patients displayed neurologic deterioration for which a treatment change was considered necessary. Follow-up was complete in 79%; mean follow-up time was 6.2 years with a minimum of 2 years. Pain scores, disability indexes, and general health outcome were comparable at follow-up. Compared with matched population norms, outcomes were poorer regardless of treatment method. Neurologic recovery was better in the operative group, but this difference did not reach statistical significance. Multivariate regression analyses revealed that female gender and neurologic impairment were independent predictors of poor functional outcome. Eighty-eight and 83% of the nonoperatively and operatively treated patients were employed at some point after a rehabilitation period. Conclusion. Overall outcome of nonoperative and operative treatment in middle-long-term follow up is comparable, although there seems to be a difference in neurologic recovery patterns. Studies on the cost-effectiveness of treatment options and the patterns of recovery within 2 years after injury would assist in guideline development and stimulate interest for future research.