Conventional and CT metrizamide myelography in Arnold-Chiari I malformation and syringomyelia.

Four normal controls and 26 cases of Arnold-Chiari I malformations and/or syringomyelia were reviewed. The pathologic cases included five isolated Arnold-Chiari I malformations, nine communicating syringomyelia, five idiopathic syringomyelia, four posttraumatic syringomyelia, one syringomyelia with hemangioblastoma, and two postshunt syringomyelia. The objectives of this study were to compare the accuracy of conventional metrizamide myelography with CT metrizamide myelography and to study indirectly the hydrodynamics of CSF flow in syringomyelia by comparing the sequential enhancement patterns of the spinal cords and cord cavities in the different groups of patients. Twenty-five patients underwent conventional metrizamide myelography immediately before CT metrizamide myelography, and one patient underwent CT metrizamide myelography only. Scans were obtained 1-2 hr, 4-8 hr, and 12-24 hr after injection of metrizamide, but not all patients were scanned during all three intervals. CT metrizamide myelography was found to be more sensitive than conventional metrizamide myelography in the diagnosis of both Arnold-Chiari I malformation and syringomyelia. Performing just an immediate and a delayed scan was found to be more cost-effective than doing all three scans. Contrary to previous reports, it was found that delayed (12-24 hr) scans demonstrated more syrinx cavities than intermediate ones. In studying the sequential enhancement patterns of the spinal cords and cord cavities, some interesting trends were observed that tend to support the theories of Aboulker and of Ball and Dayan of transneural passage of CSF into cord cavities in syringomyelia.