Back

Mesoscale CISS Imaging for the Detection of Dural Defects in Spinal CSF Leaks

Wegscheid, M. L.; Chatterjee, A. R.; Raji, C. A.; Reis, M. N.; Fleege, N. P.; Azad, S. N.; Ogunlade, J.; Vellimana, A. K.; Goyal, M. S.; Nazeri, A.

2025-07-16 radiology and imaging
10.1101/2025.07.14.25331467 medRxiv
Show abstract

Background and PurposeSpontaneous intracranial hypotension (SIH) is most often attributable to cerebrospinal fluid (CSF) leaks from the spinal canal, yet precise localization of dural defects remains challenging. While fat-saturated heavily T2-weighted MR myelography (HT2-MRM) is sensitive to spinal longitudinal extradural CSF collections (SLECs), it does not reliably localize the leak site. This study evaluates a two-stage spine MRI protocol incorporating high resolution, mesoscale 3D constructive interference in steady state (meso-CISS) for targeted CSF leak localization. Materials and MethodsTwenty consecutive patients with suspected SIH underwent a standardized total spine MRI protocol, including 3D HT2-MRM for SLEC screening. In cases with positive SLEC or indeterminate findings, meso-CISS was performed for targeted high-resolution leak localization. ResultsEight patients (40%) demonstrated SLECs on HT2-MRM, prompting further evaluation with meso-CISS. Dural defects were successfully identified in five patients (1.5-8 mm in size). In one case, meso-CISS delineated intrathecal protrusion of ligamentum flavum calcification, suggesting an underlying defect. Compared with HT2-MRM, meso-CISS provided superior visualization of dural defects, arachnoid scarring, and neo-membrane formation. Two cases were limited by motion or metal artifacts. ConclusionA two-stage spine MRI protocol incorporating HT2-MRM and meso-CISS improves the localization of dural defects in SIH. Meso-CISS enhances visualization of small dural tears and associated pathology, facilitating targeted interventions and potentially reducing reliance on invasive myelography. Further studies are warranted to assess its broader clinical impact.

Published in Headache: The Journal of Head and Face Pain · not in our set (fewer than 10 published preprints to learn from) · training set

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.