Comparing the Effectiveness of Sagittal Balance, Foraminal Stenosis, and Preoperative Cord Rotation in Predicting Postoperative C5 Palsy

Study Design: Retrospective cohort review. Objective: To determine whether preoperative cord rotation is independently correlated with C5 palsy when analyzed alongside measures of sagittal balance and foraminal stenosis. Summary of Background Data: Postoperative C5 palsy is a well-documented complication of cervical procedures with a prevalence of 4%–8%. Recent studies have shown a correlation with preoperative spinal cord rotation. There have been few studies, however, that have examined the role of sagittal balance and foraminal stenosis in the development of C5 palsy. Methods: A total of 77 patients who underwent cervical decompression—10 of whom developed C5 palsy—were reviewed. Sagittal balance was assessed using curvature angle and curvature index on radiographs and magnetic resonance image (MRI). Cord rotation was assessed on axial MRI. C4–C5 foraminal stenosis was assessed on sagittal MRI using area measurements and a grading scale. Demographics and information on surgical approach were gathered from chart review. Correlation with C5 palsy was performed by point-biserial, &khgr;2, and regression analyses. Results: Point-biserial analysis indicated that only cord rotation showed significance (P<0.01). There was no statistical significance shown with surgical approach, sex, or age. In addition, changes in sagittal balance did not correlate with presence of C5 palsy. Logistic regression model yielded cord rotation as the only significant independent predictor of C5 palsy. For every degree of axial cord rotation, the likelihood ratio for suffering a C5 palsy was 3.93 (95% confidence interval, 2.01–8.66; P<0.05). Conclusions: This supports the independent capability of preoperative cord rotation to predict postoperative C5 palsy. Lack of correlation with measures of neuroforaminal stenosis potentially points to mechanisms other than direct compression as the etiology. In addition, the lack of correlation with postoperative changes in sagittal balance hints that measures of curvature angle and curvature index may not be appropriate to accurately predict this complication. Level of Evidence: Level 3.