Eyes with particular patterns of retinal ganglion cell damage at baseline are more likely to show glaucomatous progression

Abstract

Purpose : To compare the baseline characteristics of the macular ganglion cell layer (GCL) region in eyes with progressing and non-progressing glaucomatous damage using a model of macular damage [1].

Methods : Circumpapillary and posterior pole optical coherence tomography (OCT) scans (Spectralis, Heidelberg Eng.) and 10-2 and 24-2 visual fields (Carl Zeiss) from 135 eyes/individuals were assessed by two graders independently for strong evidence (structural and functional) of progression or no progression, as part of the Macular Assessment Progression Study (MAPS). 21 patient eyes were defined as likely deteriorating (LD) and 42 patient eyes as likely non-deteriorating (LND) (average time between first and last test: 3.7 years, range 0.8 to 7.8). Presence of macular damage at baseline was defined as at least 1 red (p < 1%) GCL sector in the macular region (Fig. 1H, orange rectangle). In addition, 3 ratios between the GCL metrics, consistent with a model of macular damage [1] were calculated: temporal superior (TS)/nasal superior (NS); temporal inferior (TI)/nasal inferior (NI); and TI/TS. Differences between the LD and LND groups were evaluated.

Results : Over 60% of the LD eyes had at least 1 red GCL sector in any of the 6 macular sectors (Fig. 1A). In comparison, only 30% of the LND eyes had macular involvement (Fig. 1A, “All Sectors”; chi-square test, p = 0.037). The TI sector (Fig. 1B) showed the largest difference in macular involvement 48% vs 17% (chi-square test, p = 0.016). All 3 sector ratios (Fig. 2) were significantly lower in the LD group. The most significant difference was found at the TI/NI ratio with a mean ratio of 0.72 for the LD vs 0.95 for the LND eyes (Wilcoxon rank sum test, p < 0.001). If patients were recruited based upon a ratio of 0.8 (red dashed line, Fig. 2) on any of these 3 ratios at baseline, more progressing than non-progressing eyes would be included (TI/NI: 62% vs 10%; TI/TS: 21% vs 10%; TS/NS: 38% vs 17%).

Conclusions : Eyes with patterns of macular GCL damage consistent with a model of macular progression at baseline are more likely to progress than eyes without macular loss. Targeted recruitment based on a ratio of macular sectors at baseline may prove to be successful in recruiting eyes that are likely to progress, potentially improving the power of clinical studies.

1. Hood et al. IOVS 2023

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