Variability in rates of arthrodesis procedures for patients with cervical spine injuries with and without associated spinal cord injury

Background: Cervical spine injury, with or without spinal cord injury, is an important cause of morbidity and mortality in the United States. While substantial regional variation has been shown in per capita rates of elective cervical spine surgery, similar data regarding arthrodesis rates for traumatic cervical injury have not been reported, to our knowledge. We assessed the rates of cervical spinal arthrodesis for patients who had a cervical spine injury with or without an associated spinal cord injury. Methods: The data for the present study came from the Healthcare Cost and Utilization Project Nationwide Inpatient Sample from 1998 to 2002. International Classification of Diseases, Ninth Revision, Clinical Modification codes were used to identify patients with a cervical vertebral fracture or dislocation with or without an associated spinal cord injury. Hospitals were grouped according to their teaching status, location (urban or rural), and volume of cervical spine injury patients. The rates of spinal arthrodesis and halo/tong placement were compared for patients within each diagnostic category. Results: Twenty-eight thousand, five hundred and eighteen patients with a cervical spine injury were analyzed. Spinal arthrodesis was performed for 16.5% of patients who had a cervical fracture without an associated spinal cord injury, for 50.4% of patients who had a cervical spine fracture with an associated spinal cord injury, and for 44.1% of patients who had a cervical dislocation. With the numbers available, the rates of arthrodesis for patients who had a fracture without a spinal cord injury and for patients who had a cervical dislocation were not significantly different between high and low-volume centers, although the rate for patients who had a cervical fracture with a spinal cord injury was significantly higher at high-volume hospitals. The rates of arthrodesis did not vary significantly between urban teaching and nonteaching hospitals, with the numbers available, for patients in any of the three diagnostic categories. Individual hospitals had a threefold to fivefold variation in the arthrodesis rate for patients with a cervical spine injury, depending on the diagnostic category. Conclusions: The present study demonstrated substantial differences in the rate of arthrodesis for patients with cervical spine trauma, depending on the diagnostic category. The variations in the rates of arthrodesis within diagnostic categories appear to be lower than the previously reported variation in the rates of elective cervical spine procedures.