Cervical Spondylotic Myelopathy: Arguments for Surgery
Strong arguments suggest that, to treat cervical spondylotic myelopathy, surgery is better than medical management such as collar immobilization and traction.
Patients treated medically show continual progressive neurologic deterioration. In a series of 1,355 patients with cervical spondylotic myelopathy treated conservatively, Epstein et al1 (4) found that 64% showed no improvement and 26% deteriorated neurologically. In a series reported by Clark and Robinson, (21) approximately 50% of patients with cervical spondylotic myelopathy treated medically deteriorated neurologically. Using a disability scale to assess functional status, Symon and Lavender (16) found that 67% of their patients with cervical spondylotic myelopathy experienced progressive deterioration in function.
Similarly, Roberts (17) found that in a series of 24 patients with cervical spondylotic myelopathy, 70% either showed no improvement or deteriorated neurologically with conservative measures. This deterioration was graded by degree of motor disability, measured by the ability to perform daily activities.
Patients with cervical spondylotic myelopathy are at an increased risk of spinal cord injury from relatively mild traumatic events. If the anterior-posterior diameter of the cervical spinal canal is decreased, the spinal cord has limited room to move. (4) Many traumatic cervical injuries are due to hyperextension, which results in maximal narrowing of the spinal canal. Epstein et al (22) evaluated 200 patients with severe cervical canal stenosis (<13 mm) admitted over a 4-year period to spinal cord trauma center. Twenty-three patients had no fracture or dislocation, and in this subgroup there was direct relationship between smaller anterior-posterior diameter of the more severe myelopathy after trauma. Similarly, Firooznia et al (23) described three with cervical canal stenosis who all became quadriplegic minor without any dislocation.
Early surgery can improve prognosis. Montgomery and Brower (2) found that the prognosis after surgery was better for patients with less than 1 year of symptoms, young age, fewer levels of involvement, and unilateral motor deficit.
Phillips (24) examined 65 patients treated surgically and found that symptoms of less than 1 year's duration significantly correlated with benefit from treatment. Similarly, Ebersold et al (25) evaluated several possible predictors of outcome in 84 patients treated surgically. Using the Nurick functional grade, they found that the only significant variable predictive of outcome was how long the symptoms had lasted before surgery.
A Dorsal vs Central Approach
There are two surgical options for patients with cervical spondylotic myelopathy:
a dorsal approach (ie, cervical laminectomy) or a ventral approach (ie, either
discectomy at one or more levels with interbody fusion or one or more corpectomies
with interbody fusion). Corpectomy typically involves cervical plating to provide
stability until fusion occurs. Since no clinical study has demonstrated a significant
difference in the outcomes of the dorsal vs ventral approaches, (12,25-31) the
choice is based on the surgeon's preference. However, two factors guide this
decision:
The relative location of the stenosis (ie, dorsal vs ventral). For patients with cervical spine stenosis that primarily results from dorsal compression, cervical laminectomy (ie, the dorsal approach) is better. This includes patients with dorsal spinal cord impingement from the buckling or enfolding of the ligamentum flavum or from facet arthropathy or both. (13) For patients with ventral disc herniations and osteophytes, however, cervical laminectomy alone does not allow sufficient access to the ventral spinal cord. These patients benefit more from a ventral decompression and fusion procedure.
Spine Configuration Dictates Surgical Approach
|
|
|
|
|
Lordosis
(dorsal approach indicated) |
Kyphosis
(ventral approach indicated) |
Straight |
Figure 2. Patients with cervical spondylotic myelopathy may present with one of three spinal alignments, which affects the choice of approach for surgical decompression. Using a sagittal magnetic resonance image of the cervical spine, the surgeon may draw a line from the dorsocaudal aspect of the vertebral body of C2 to the dorsocaudal aspect of the vertebral body of C7 and then add a kite-shaped zone, the width of which depends on his or her biases and preferences. If the kite is completely dorsal to the vertebral bodies, the spine is in lordosis, and a dorsal surgical approach is indicated (left). If the kite is completely ventral to the dorsal aspects of the vertebral bodies, the spine is in kyphosis and a ventral approach is indicated (middle). If the kite is partly dorsal to the dorsal aspects of the vertebral bodies, the spine is considered straight, and either approach is appropriate (right).
The alignment of the cervical spine (ie, kyphosis vs lordosis). Effective cervical kyphosis is defined as an alignment of the cervical spine in which any part of the dorsal aspect of any of the C3-C7 vertebral bodies crosses a line drawn in the midsagittal plane (on radiography or sagittal MRI) from the dorsocaudal aspect of the vertebral body of C2 to the dorsocaudal aspect of the vertebral body of C7. In effective cervical lordosis, no part of the dorsal aspect of any of the C3-C7 vertebral bodies crosses this line (Figure 2).
For patients with effective cervical kyphosis, dorsal decompression is associated with a high probability of failure. (32,33) In these patients, cervical laminectomy can worsen the ventral spinal cord compression by tethering the dural sac and its contents over ventral osteophytes, which leads to neurologic deterioration. (4) A dorsal approach in this situation may also lead to progressive kyphotic deformity and instability requiring repeat surgery and stabilization.
For patients with effective lordosis, a dorsal approach is often optimal for spinal cord decompression, especially when there is dorsal compression. It should be noted that when compression is ventral (eg, herniated nucleus pulposus), the decompression is optimally performed from a ventral approach, even if the cervical spine is configured in lordosis. Between kyphosis and lordosis is a "gray zone" in which the surgical approach is chosen on the basis of the biases and clinical judgment of the surgeon (Figure 2).(32)
Patient-specific biomechanics should also be considered. Patients should be individualized in regards to the surgical approach chosen. For example, one may consider a dorsal fusion in addition to a large multisegment ventral decompression (corporectomy), even if the spine is not in effective lordosis.
Cleveland Clinic Journal of Medicine
Volume 70, Number 10, October 2003





















