Abstract
Purpose :
Pathologic optic disc cupping is often related to glaucomaous optic neuropathy(GON), but can be caused by compressive optic neuropathies(CON) leading to diagnostic confusion. Previously we compared Bruch’s Membrane(BM) Opening–Minimum Rim Width(MRW) measurements in eyes with CON, GON and controls and found that a ratio between MRW and peripapillary retinal nerve fiber layer (RNFL) provided better discrimination ability between groups, compared to isolated MRW and pRNFL measurements. We now evaluate the diagnostic ability of cup depth measurements to differentiate CON from GON and controls.
Methods :
Prospective study with 47 eyes with CON (chiasm compression), 27 eyes with GON and 41 control eyes. Subjects underwent examination including Humphrey 24-2 visual field(VF) and Spectralis OCT acquisition of MRW and pRNFL parameters. Cup depth was calculated using a perpendicular line connecting the bottom of the cup and a radial line between BM terminations. Horizontal and vertical scans were averaged (Fig1). Comparisons were performed using GEE. Area under the ROC curves, positive and negative likelihood ratios(PRL,NLR) were calculated.
Results :
CD mean values were significantly higher in GON compared to CON and controls. No significant difference was found between CON and controls. When comparing the ability of each OCT parameter in discriminate CON from GON, AUCs showed no statistic difference. However, CD alone showed better PLR and NLR than the other parameters. When combining CD to other parameters, both ratios CD/MRW and CD/RNFL showed greater mean values in GON. As for the discrimination ability between the diseases, AUCs showed no difference between the ratios and CD/MRW had better likelihood ratios(Table).
Conclusions :
CD was significantly shallower in CON than in GON. Retrograde axonal damage from CON is not associated with significant laminar displacement or remodeling. CD measurements can help differentiating CON from CON, with a diagnostic performance similar to MRW and RNFL measurements.
This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.