Laminectomy and posterior cervical plating for multilevel cervical spondylotic myelopathy and ossification of the posterior longitudinal ligament: effects on cervical alignment, spinal cord compression, and neurological outcome

Abstract

Objective: Multilevel anterior decompressive procedures for cervical spondylotic myelopathy or ossification of the posterior longitudinal ligament may be associated with a high incidence of neurological morbidity, construct failure, and pseudoarthrosis. We theorized that laminectomy and stabilization of the cervical spine with lateral mass plates would obviate the disadvantages of anterior decompression, prevent the development of kyphotic deformity frequently seen after uninstrumented laminectomy, decompress the spinal cord, and produce neurological results equal or superior to those achieved by multilevel anterior procedures.

Methods: We retrospectively reviewed the records of 38 patients who underwent laminectomy and lateral mass plating for cervical spondylotic myelopathy or ossification of the posterior longitudinal ligament between January 1994 and November 2001. Seventy-six percent of patients had spondylosis, 18% had ossification of the posterior longitudinal ligament, and 5% had both. Clinical presentation included upper extremity sensory complaints (89%), gait difficulty (70%), and hand use deterioration (67%). Spasticity was present in 83%, and weakness of one or more muscle groups was seen in 79%. Spinal cord signal abnormality on sagittal T2-weighted magnetic resonance imaging (MRI) was seen in 68%. Neurological evaluation was performed using a modification of the Japanese Orthopedic Association Scale for functional assessment of myelopathy, the Cooper Scale for separate evaluation of upper and lower extremity motor function, and a five-point scale for evaluation of strength in individual muscle groups. Lateral cervical spine x-rays were analyzed using a curvature index to determine maintenance of alignment. Each surgically decompressed level was graded on a four-point scale using axial MRI to assess the adequacy of decompression. Late follow-up was conducted by telephone interview.

Results: Laminectomy was performed at a mean 4.6 levels. Follow-up was obtained at a mean of 30.2 months after the procedure. The score on the modified Japanese Orthopedic Association scale improved in 97% of patients from a mean of 12.9 preoperatively to 15.58 postoperatively (P < 0.0001). In the upper extremities, function measured by the Cooper Scale improved from 1.8 to 0.7 (P < 0.0001), and in the lower extremities, function improved from 1.0 to 0.4 (P < 0.0002). There was a statistically significant improvement in strength in the triceps (P < 0.0001), iliopsoas (P < 0.0002), and hand intrinsic muscles (P < 0.0001). X-rays obtained at a mean of 5.9 months after surgery revealed no change in spinal alignment as measured by the curvature index. There was a decrease in the mean preoperative compression grade from 2.46 preoperatively to 0.16 postoperatively (P < 0.0001). There was no correlation between neurological outcome and the presence of spinal cord signal change on T2-weighted MRI scans, patient age, duration of symptoms, or preoperative medical comorbidity.

Conclusion: Multilevel laminectomy and instrumentation with lateral mass plates is associated with minimal morbidity, provides excellent decompression of the spinal cord (as visualized on MRI), produces immediate stability of the cervical spine, prevents kyphotic deformity, and precludes further development of spondylosis at fused levels. Neurological outcome is equal or superior to multilevel anterior procedures and prevents spinal deformity associated with laminoplasty or noninstrumented laminectomy.