Am J Ophthalmol
Am J OphthalmolSeptember 2024Journal Article

Peripapillary Versus Macular Thinning to Detect Progression According to Initial Visual Field Loss Location in Normal-Tension Glaucoma.

Visual FieldOptic Nerve & Disc

Summary

mGCIPL outperforms pRNFL at early follow-up in detecting VF progression in IPFS eyes but not INS eyes.

Abstract

PURPOSE

To investigate the predictive capabilities of peripapillary retinal nerve fiber layer (pRNFL) and macular ganglion cell-inner plexiform layer (mGCIPL) thinning to detect visual field (VF) progression in normal-tension glaucoma patients with an initial parafoveal scotoma (IPFS) or nasal step (INS).

DESIGN

Retrospective cohort study.

METHODS

A total of 185 early-stage glaucoma eyes, followed for 10 years, were retrospectively stratified into IPFS and INS groups. Progressive pRNFL and mGCIPL thinning were assessed using spectral-domain optical coherence tomography and VF progression using both event- or trend-based analysis. Kaplan-Meier survival analysis compared VF survival in each VF phenotype with or without progressive pRNFL and mGCIPL thinning. Cox proportional regression analysis identified VF progression factors.

RESULTS

VF progression was detected in 42 IPFS (n = 86) and 47 INS (n = 99) eyes. Among VF progressors, pRNFL thinning was significantly faster in INS group compared to IPFS group (P < .01), while mGCIPL thinning was similar (P = .16). At 5 years, eyes with progressive mGCIPL thinning showed significantly lower VF survival in both VF phenotypes (all P < .05). Progressive pRNFL thinning showed significantly lower VF survival only in INS eyes (P = .015). Cox multivariate regression revealed that mGCIPL thinning predicted subsequent VF progression in IPFS eyes, while mGCIPL and pRNFL thinning had significant associations with VF progression in INS eyes.

CONCLUSIONS

mGCIPL outperforms pRNFL at early follow-up in detecting VF progression in IPFS eyes but not INS eyes. Appropriate selection of structural parameters (mGCIPL vs. pRNFL) maximizes early VF progression detection according to initial VF defect location.

In the Knowledge Library

Discussion

Comments and discussion will appear here in a future update.