Cervical spondylotic myelopathy: surgical results and factors affecting prognosis.

Abstract

A variety of factors may affect surgical outcome in patients with cervical spondylotic myelopathy. The aim of this study is to determine these factors on the basis of preoperative radiological and clinical data. To assess the factors affecting postoperative outcome after surgery for cervical spondylotic myelopathy, the clinical and radiological data of 27 patients with cervical spondylotic myelopathy were reviewed. Functional and neurological statuses were assessed using the Japanese Orthopaedic Association (JOA) scale modified by Benzel. In all patients, the effect of age, symptom duration, cervical curvature, presence or absence of preoperative high signal intensity within the spinal cord as revealed by T2-weighted magnetic resonance imaging, and diameters of the spinal canal and vertebral body on pre- and postoperative neurological statuses were investigated. Plain radiographs were obtained for all patients, magnetic resonance images for 21 patients (77.8%), computed tomographic scans for 13 patients (48.1%), myelograms for 6 patients (22.2%), and computed tomographic myelograms for 4 patients (14.8%). There were five patients with a JOA score of 10, six patients with a JOA score of 11, six patients with a JOA score of 12, four patients with a JOA score of 13, four patients with a JOA score of 14, one patient with a JOA score of 15, and one patient with a JOA score of 16. All patients underwent cervical laminectomies. The mean follow-up period was 54.1 months. The final neurological examinations revealed improvement in the JOA scores of 85.1 % of the patients. Statistical analysis of all patients revealed mean JOA scores of 12.185 +/- 1.618 and 14.370 +/- 2.15 before surgery and at final examination, respectively. The difference between the preoperative JOA score and the final JOA score was determined to be statistically significant (P < 0.0001). Statistical analyses also showed better neurological improvement in patients younger than 60 years and in patients with normal preoperative cervical lordosis. Although patients without preoperative high signal intensity of the spinal cord showed a better improvement rate than did patients with preoperative high signal intensity, the determined difference was statistically insignificant. It can be concluded that age and abnormal cervical curvature predict less postoperative neurological improvement. The presence of preoperative high signal intensity within the spinal cord may also reflect less neurological improvement.