Optic nerve edema and pseudo-edema differentiation analysis using automated segmentation of fiber layer and blood vessels

Abstract

Purpose : Identification of specific patterns of optic nerve head (ONH) fiber and vessel thickening in ONH edema and pseudo-edema using an automated algorithm.

Methods : LHON patients with pseudo-edema in acute stage and papilledema or NAION patients with ONH edema were included in this study. ONH OCT-Angiography disc cube (DRI OCT Triton, Topcon, Japan) volumes for all eyes were exported and analyzed with the Orion software (Voxeleron LLC, San Francisco, CA, USA): RNFL and disc margin were automatically segmented from the structural data, and retinal blood vessels (RBV) were segmented in the angiography data. Edema and pseudo-edema vessel and RNFL thicknesses were compared by a t-test.

Results : Fifteen eyes with LHON pseudo-edema and 20 eyes with ONH edema were considered in the study. LHON eyes presented a specific pattern of swelling of fiber with and without vessel consideration in comparison with ONH edema eyes. The vascular component appears to be thicker in papilledema cases while temporal RNFL appears thinner in LHON pseudo-edema eyes.

Conclusions : Automated segmentation of RNFL and RBV around the ONH is a promising technique to differentiate ONH edema from pseudo-edema. This technique could be useful in difficult clinical differential diagnosis and in future clinical and therapeutic trials.

This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.