Refining Operative Strategies for Optic Nerve Decompression: A Morphometric Analysis of Transcranial and Endoscopic Endonasal Techniques Using Clinical Parameters.

BACKGROUND: Various approaches can be considered for decompression of the intracanalicular optic nerve. Although clinical experience has been reported, no quantitative study has yet compared the extent of decompression achieved by an endoscopic endonasal versus transcranial approach.

OBJECTIVE: Toward this aim, our morphometric analysis compared both approaches by quantifying the circumferential degree of optic canal decompression that is possible before any meningeal violation, which would result in cerebrospinal fluid (CSF) leak.

METHODS: From 10 cadaver heads, 20 optic canals were sequentially decompressed using an endoscopic endonasal approach and pterional craniotomy with extradural clinoidectomy. Dissections ended before violation of the sphenoid sinus during the transcranial approach, and before intracranial transgression from the endonasal corridor. Based on our study criteria, decompressions were not maximal for either approach, but were maximal before violating the other compartment. Decompression achieved from each approach was quantified using CT scans for each stage.

RESULTS: Greater circumferential bony optic canal decompression was obtained from transcranial (245.2°) than endonasal (114.8°) routes (P < .001). By endonasal perspective, the anatomical point where the optic nerve traverses intracranially was approximated by the medial border of the anterior ascending cavernous internal carotid artery.

CONCLUSION: Our morphometric analysis comparing optic canal decompression for endonasal and transcranial corridors provides important guidance for this location. Ample visualization and wide exposure can be achieved via a transcranial approach with limited risk of CSF leak. A landmark, where the intracanalicular segment ends and optic nerve traverses intracranially, can mark the extent of decompression safely obtained before risking CSF leak.