Reply:

Thank you very much for such a thoughtful and timely letter. It brings up the important topic of CSF dynamics within the “subdural” space, which has not been fully elucidated. Diffusion of subdural contrast into the thecal sac is certainly interesting; we do not yet know if it is from the disruption of the inner layer at the puncture site, or if intrathecal distribution results from the natural egress passage of any subdural fluid.

In a typical CT myelogram for spinal stenosis evaluation, rescanning patients after an hour in cases of subdural injection to see whether the contrast diffuses intrathecally is a very acceptable option. This is especially true during the current pandemic setting, as the authors mentioned.

While most CT myelograms performed for spinal stenosis can be diagnostic if the CSF is appreciably denser than the neural elements, the evaluation of spinal CSF leaks on CT myelogram is extremely dependent on a high concentration of intrathecal contrast. This is especially true with CSF-venous fistulas, where the findings are often subtle even with maximal contrast attenuation. Therefore, waiting an hour for subdural injections to become intrathecal while contrast resorption from the CSF to the venous system is occurring would likely result in submaximal opacification of intrathecal CSF and markedly decrease the diagnostic yield.

Thank you again for the letter and great efforts to optimize patient care in a difficult time for the medical community.

• © 2021 by American Journal of Neuroradiology