METHODS Cross-sectional study. OBJECTIVE To evaluate morphometric variations of the cervical spine in patients with cervical spondylotic myelopathy (CSM) using a standard technique of dynamic magnetic resonance imaging (MRI), to assess the inter- and intraobserver reliability of measurements of morphometric parameters based on this imaging, and to compare the resulting measurements with those previously published. BACKGROUND Because dynamic factors may contribute to the etiology and severity of CSM, dynamic (flexion-extension) MRI may be useful to better evaluate for spinal stenosis and cord compromise. METHODS Dynamic cervical MRI was obtained using a standard protocol with the neck in neutral, flexion, and extension positions. The morphometric parameters considered were anterior length of the spinal cord (ALSC), posterior length of the spinal cord (PLSC), spinal canal diameter, and spinal cord width. Two observers analyzed the parameters independently, and the inter- and intraobserver reliabilities were assessed by the intraclass correlation coefficient (ICC). RESULTS A total of 18 patients were included in the study and all completed the dynamic MRI acquisition protocol. The inter- and intraobserver reliabilities demonstrated "almost perfect agreement" (ICC >0.9, P < 0.001) for ALSC and PLSC in all positions. The spinal canal diameter had inter- and intraobserver reliability classified as "almost perfect agreement" (ICC: 0.83-0.98, P < 0.001 and ICC: 0.90-0.99, P < 0.001, respectively) in all positions. The spinal cord width had inter- and intraobserver reliability classified as "substantial agreement" (ICC: 0.73-0.94, P < 0.001 and ICC: 0.79-0.96, P < 0.001, respectively) in all positions. ALSC and PLSC in neutral, flexion, and extension positions from the present study were significantly greater compared to the measurements previously published (P < 0.001). CONCLUSIONS The dynamic MRI protocol presented was safe and may allow a more complete evaluation of variations in the cervical spine in patients with CSM than traditional MRI protocols. The morphometric parameters based on this protocol demonstrated excellent inter- and intraobserver reliabilities. METHODS 4.