Bruch Membrane Opening Minimum Rim Width and Retinal Nerve Fiber Layer Helps Differentiate Compressive Optic Neuropathy From Glaucoma

Abstract

center dot PURPOSE: To compare optical coherence tomography measured Bruch membrane opening minimum rim width (MRW), peripapillary retinal nerve fiber layer (pRNFL) measurements, and MRW:pRNFL ratios in eyes with compressive optic neuropathy (CON) and glaucoma and controls, and evaluate the ability of these parameters to differentiate CON from glaucoma. center dot DESIGN: Prospective, cross-sectional study. center dot METHODS: Setting : Single-center tertiary hospital and outpatient clinic. Patient population : One hundred fifteen eyes of 77 participants, 34 with CON from chiasmal lesions, 21 with glaucoma, and 22 healthy controls. Observation procedures : Optical coherence tomography-measured MRW, pRNFL, and MRW:pRNFL ratios for each optic disc sector and global average. Main outcome measures: MRW, pRNFL, and MRW:pRNFL ratios compared using generalized estimated equations. Area under the receiver operating characteristic curve and positive and negative likelihood ratios were calculated. center dot RESULTS: MRW and pRNFL measurements were significantly reduced in CON and glaucoma compared with controls. In glaucoma, MRW was thinner than in CON in the global, inferotemporal, superonasal, inferonasal, and vertical average measurements, but a significant overlap was observed in many parameters. MRW:pRNFL ratios increased the ability to discriminate between CON and glaucoma, as shown by the high area under the receiver operating characteristic curve, high positive likelihood ratios, and low negative likelihood ratios, especially in the nasal disc sector and the nasal and temporal average. center dot CONCLUSIONS: MRW measurements alone cannot reliably distinguish CON from glaucoma, but the combination of MRW, pRNFL, and MRW:pRNFL ratios significantly improves accuracy. When comparing the 2 conditions, MRW:pRNFL ratios yielded higher area under the receiver operating characteristic curve and positive and negative likelihood ratios, suggesting this parameter may be helpful in clinical practice. (Am J Ophthalmol 2021;234: 156-165.