Session V Spine

Anterior vs. Posterior Stabilization of Cervical Spine Fractures in Spinal Cord Injured Patients

Darrel S. Brodke, MD, Jens R. Chapman, MD, Paul A. Anderson, MD, David W. Newell, MD, M. Sean Grady, MD, Brenda Harthan, BS

University of Washington-Harborview Medical Center, Seattle, WA

Introduction: Recent advances have changed the treatment of unstable cervical spine injuries. Surgical stabilization with anterior plates or posterior lateral mass plates have improved stability, fusion rates and maintenance of alignment. Often there are clear indications to stabilize from either the anterior or posterior approach to allow decompression of the spinal cord or facilitate reduction. Sometimes either approach can be used, however, there is little in the literature comparing the results of these two treatment options.

Methods: This study reviews the results of 47 spinal cord injured patients randomized by assigned surgeon to either anterior stabilization using a Morscher plate or posterior stabilization using AO lateral mass plates. Inclusion criteria were adult patients with unstable cervical spine injuries and spinal cord injuries of Frankel grade A-D who were treated at Harborview Medical Center between January 1, 1991 and December 31, 1993. Patients requiring a specific approach for reduction or decompression were not randomized and were therefore not reviewed. Of the 52 patients with spinal cord injuries that were included in this study, 22 were treated with anterior stabilization and 30 with posterior stabilization. Two patients in the anterior group died of other injuries in the early post-operative period, three patients in the posterior group had less than six months of follow-up and were not included in the analysis. Of the remaining 20 patients in the anterior stabilization group and 27 patients in the posterior stabilization group, 27 were diagnosed with bilateral facet fractures and /or dislocations (7 anterior/20 posterior), 12 patients had burst fractures with bilateral facet disruption (8 anterior/4 posterior), 7 patients had a burst fracture alone (4 anterior/3 posterior), and 1 patient treated anteriorally had an extension-distraction injury. There were 15 males and 5 females with an average age of 38 in the anterior group, and 22 males and 5 females with an average age of 33 in the posterior group. The average number of levels fused in the anterior group were 1.8, and in the posterior group, 1.7. Average follow-up was 17 months in the anterior group and 14 months in the posterior patient as well as fusion status, and changes in alignment and displacement.

Results: Neurologic improvement was noted in both groups with no significant difference between groups. In the anterior group 70% of the patients improved 1 Frankel grade with 2 patients improving 3 grades. In the posterior group 57% improved 1 Frankel grade (p>0.05). Motor index score improved an average 24 points in the anterior group and 14 points in the posterior group (p>0.05). Both groups also had similar improvement in angulation and translation. In the anterior group, the average angulation was initially 6° kyphosis with 3 mm displacement reduced to 2° lordosis with no displacement post-operatively and no change at final follow-up. In the posterior group the average angulation was 7° kyphosis with 4 mm average displacement pre-operatively, reduced to 4° lordosis with no displacement post-operatively and 2° lordosis with no displacement at final follow-up (p>0.05). Complications included 1 hardware failure in each group and 2 non-unions in the anterior group (90% fusion rate). There were no non-unions in the posterior group (100% fusion rate). The difference in fusion rate was not statistically significant (p>0.05). Additional complications included 2 patients with pneumonia (1 in each group) and 1 patient with ARDS in the anterior group. There were also 2 deaths in the anterior group from unrelated causes. There were 7 patients (35%) complaining of ongoing neck pain at final follow-up in the anterior group and 7 patients (26%) in the posterior group (p>0.05).

Discussion and Conclusion: Principles in the care of patients with cervical spinal cord injuries include immobilization, reduction of deformity, decompression of neural elements and stabilization of injured segments until healing of the spinal column has occurred. If surgical reduction and /or stabilization are required anterior or posterior instrumentation with fusion techniques have been advocated. To date, little data exists as to which technique is preferable in the treatment of spinal cord injured patients with cervical instability that do not require a specific approach. We found no significant differences in fusion rates, alignment, neurologic recovery, or long-term complaints of pain in patients treated with either anterior fusion and Morscher plate application or posterior fusion and lateral mass plate application. Either approach can be chosen based on surgeon preference and specific indications and conditions of the patient.