The association between preoperative spinal cord rotation and postoperative C5 nerve palsy

Abstract

BACKGROUND: C5 nerve palsy is a known complication of cervical spine surgery. The development and etiology of this complication are not completely understood. The purpose of the present study was to determine whether rotation of the cervical spinal cord predicts the development of a C5 palsy. METHODS: We performed a retrospective review of prospectively collected spine registry data as well as magnetic resonance images. We reviewed the records for 176 patients with degenerative disorders of the cervical spine who underwent anterior cervical decompression or corpectomy within the C4 to C6 levels. Our measurements included area for the spinal cord, space available for the cord, and rotation of the cord with respect to the vertebral body. RESULTS: There was a 6.8% prevalence of postoperative C5 nerve palsy as defined by deltoid motor strength of /= 11 degrees ) and palsy (point-biserial correlation = 0.94; p < 0.001). A diagnostic criterion of 6 degrees of rotation could identify patients who had a C5 palsy (sensitivity = 1.00 [95% confidence interval, 0.70 to 1.00], specificity = 0.97 [95% confidence interval, 0.93 to 0.99], positive predictive value = 0.71 [95% confidence interval, 0.44 to 0.89], negative predictive value = 1.00 [95% confidence interval, 0.97 to 1.00]). CONCLUSIONS: Our evidence suggests that spinal cord rotation is a strong and significant predictor of C5 palsy postoperatively. Patients can be classified into three types, with Type 1 representing mild rotation (0 degrees to 5 degrees ), Type 2 representing moderate rotation (6 degrees to 10 degrees ), and Type 3 representing severe rotation (>/= 11 degrees ). The rate of C5 palsy was zero of 159 in the Type-1 group, eight of thirteen in the Type-2 group, and four of four in the Type-3 group. This information may be valuable for surgeons and patients considering anterior surgery in the C4 to C6 levels.