A comparison of circumpapillary retinal nerve fiber layer (cpRNFL) thickness and minimum rim width at Bruch’s membrane opening (BMO-MRW) analyses in patients with mild glaucomatous damage

Abstract

Purpose : We previously reported that BMO-MRW analysis may miss confirmed local glaucomatous defects.[1] Here we compare cpRNFL and BMO-MRW analyses in a larger group of eyes with early, but confirmed, glaucomatous damage.

Methods : Cube scans of the optic disc were obtained from 54 eyes of 54 patients (58.2±12.1 yr) with mild glaucoma using both swept-source (ss) OCT (3x3mm, 256 B-scans, DRI, Topcon) and spectral domain (sd) OCT (6x6 mm, 128 B-scans, 2000, Topcon). The patients were chosen from a larger group of eyes with a suspicious disc, a 24-2 visual field (VF) mean deviation better than -6dB, glaucomatous damage as judged by 2 specialists based upon a single-page report that combined 24-2 VF, 10-2 VF and OCT disc and macular probability plots,[2] and all information in the patients’ charts. The center of the disc was determined using the BMO and 24 radial images were derived from the ssOCT scans. BMO-MRW, the minimum distance between BMO and the inner limiting membrane (ILM), was manually marked on the radial images and these values averaged for the entire disc (Global) and for 3 temporal sectors: superior temporal (ST), temporal (T), and inferior temporal (IT).[3] Nasal sectors were not analyzed because the blood vessels made BMO-MRW measurement difficult. The cpRNFL thicknesses at 3.4 mm dia. for the same regions were determined from the sdOCT scans. As controls, ssOCT and sdOCT scans were obtained from 60 healthy eyes (52 individuals, 54.24 ± 13.8 yr), and 62 eyes (62 individuals, 50.3±10.2 yr), respectively.

Results : The results for the BMO-MRW and cpRNFL analyses were similar (Table); only the AUCROC values for the temporal region differed significantly (p=0.002). For 95% specificity, the sensitivity for global and regional measures ranged from 51.9 to 64.8% (MRW) and 31.5 to 75.9% (cpRNFL). Further, 11 eyes (MRW) and 8 eyes (cpRNFL) were missed by all 4 regions. (2 eyes were in both groups.) 16 of these 17 eyes had local damage seen on the combined OCT and VF probability plots (Fig).

Conclusions : For these patients with mild glaucoma, the sensitivity/specificity of the BMO-MRW and cpRNFL analyses were similar. More importantly, both analyses missed eyes with local defects confirmed on VF and OCT macular cube scans. 1. Jarukasetphon et al. 2014 ARVO 2. Hood & Raza 2014, BJO 3. Chauhan et al., 2013 Ophthal

This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.