Abstract:
OBJECTIVE: To compare the sensitivities of dynamic radiographic myelography and supine MRI in lumbar canal stenosis (LCS) patients and to determine whether dynamic radiographic myelography is an irreplaceable preliminary test in the diagnosis of lumbar canal stenosis.
METHODS: Over two years, the imaging data of 100 patients who were suspected of having LCS were prospectively analyzed. All lumbar intervertebral segments were evaluated in each patient on sagittal MR T2-weighted images and lateral plane images by myelography using a semiquantitative scoring system. The differences in scores for 5 motion segments under 3 conditions (supine MRI, sMRI; upright sitting myelography, SM; and standing myelography with extension, SME) were analyzed statistically.
RESULTS: Of 100 patients with 500 analyzed intervertebral segments, 23 subjects with inconspicuous sMRI results had LCS in SME. Compared with SM and sMRI, SME yielded the highest score for every segment from L1/2 to L5/S1. Compared with the SM position, 61 more patients received a diagnosis of lumbar stenosis in the SME position, and 121 more stenotic segments were diagnosed. Compared with the sMRI position, SME detected 64 more stenotic patients and 137 more stenotic segments.
CONCLUSIONS: Based on a large patient sample, dynamic myelography is a valuable diagnostic tool in detecting lumbar spinal stenosis. Whereas patients with lumbar spinal stenosis may have obscure sMRI results, the 23% misdiagnosis rate of LCS patients with unclear sMRI results can be avoided with dynamic myelography. The combination of sMRI and dynamic myelography is critical in the evaluation of LCS, especially if multisegmental findings are detected.