What Causes Cervical Spondylotic Myelopathy?

Source: Levine DN. Pathogenesis of cervical spondylotic myelopathy. J Neurol Neurosurg Psychiatry 1997;62:334-340.

If cervical spondylotic myelopathy (csm) produces cord damage by compressing the spinal cord, why does myelography so rarely show a block of contrast agent? And, why is surgical decompression not always effective? Is it because the cord is traumatized by protruding osteophytes or hypertrophied longitudinal ligaments as it moves cephalad/posteriorly and caudad/anteriorly on flexion and extension (Adams CBT, et al. Brain 1971;94:569-586)? Or, does compression of spinal blood vessels perhaps produce cord ischemia and consequent myelopathy (Taylor AR. Neurology 1964;14:62-68).

CSM is characterized pathologically by the following: 1) flattening in the A-P but not the transverse dimension of the cord; 2) the lateral columns bear the brunt of the injury; and 3) the involved area is usually wedge-shaped, with the base laterally along the outer rim of the cord and the apex medially, pointing toward the central canal. Using mathematical modeling and software computation, David Levine, a neurologist and accomplished mathematician, concludes that this topographic pattern is most consistent with the theory that the spinal cord is pulled laterally by dentate ligaments that become tensed when an anterior spondylotic bar displaces the cord posteriorly. Levine concludes that if the alternative mechanical theory is true (i.e., that the cord is compressed anteroposteriorly between an anterior spondylotic bar and the posterior ligamentum flavum), the lateral columns would be relatively free of injury, contrary to what is seen.

Your editor remembers well arguments of 30-40 years ago about whether squeezing or stretching the spinal cord by spondylosis produces its subsequent tendency to develop lateral column myelopathy. Levine has elegantly provided new theoretical mechanisms that fit the dentate ligament theory. He omits consideration, however, that the lateral columns are supplied largely by endartery branches of the anterior spinal artery, thereby possibly creating a watershed area of slowly developing, pulsatile ischemia. —mr