Assessment of the narrow cervical spinal canal: a prospective comparison of MRI, myelography and CT-myelography

This study was designed to compare the accuracy of magnetic resonance imaging (MRI), myelography and computed tomography in the assessment of degenerative cervical spinal stenosis. We prospectively examined a total of 75 spinal segments in 18 patients with suspected cervical spinal canal stenosis, using sagittal spin-echo and axial gradient-echo sequences generated by a 1.5 Tesla imager, conventional myelography, and computed tomography with intrathecal contrast medium (CT-myelography). The degree of stenosis was often overestimated using MRI. This error was most prominent in cases of severe stenosis but was significant with minor to moderate stenosis. In these cases, the clinical consequences of such an overestimation can be serious, because treatment is misdirected. The error is probably caused by pulsation of the cerebrospinal fluid and truncation artefact (Gibbs phenomenon). MRI at 1.5 Tesla is thus frequently inadequate for diagnostic assessment of degenerative cervical spinal stenosis. Myelography and myelographic CT are still useful for decisions on operative treatment, especially in cases of moderate stenosis. This may, however, not apply to imagers operating at 0.5 Tesla as below. (orig.)