Sector-wise Comparison of Retinal Layer Thickness in High Myopia and Low/Non-Myopia Eyes without Visual Field Defects

Abstract

Purpose : To compare individual retinal layer thickness in peripapillary (ppOCT) and macula (mOCT) OCT scans in high myopia (HM) and low/non-myopia (NM) eyes without visual field (VF) defects.

Methods : HM was defined as spherical equivalent (SE) of ≤-6D or axial length (AL) of ≥26mm, NM was defined as SE of 0 to -3D and AL of <24 mm. Subjects with diabetes, neurological disease, neuroretinal rim thinning, intraocular pressure of >21mmHg or other ocular disease apart from myopia were excluded. Eyes with indeterminate neuroretinal rim thickness underwent VF testing, and were included if ≥2 reliable VF showed glaucoma hemifield test within normal limits, and VF index ≥97%. Two different 200 x 200 scans were acquired in the same eye using Cirrus spectral-domain OCT: mOCT centered on the macula and ppOCT centered on the optic disc. Poor quality scans were excluded. PpOCT scans were divided into superior (S), nasal (N), inferior (I) and temporal (T) sectors; mOCT scans were divided into S, S-N, S-T, I, I-N and I-T sectors and analyzed using Iowa Reference Algorithms 3.8.0 and MATLAB to delineate individual layer thickness and sectoral boundaries. Ocular magnification was corrected using Littmann’s formula, based on axial length. Thickness measurements were compared between HM and NM eyes using linear mixed effect models to account for fellow-eye correlation, after adjusting for age and gender.

Results : 159 eyes from 111 HM subjects and 153 eyes from 98 NM subjects were included. Sector-wise differences were noted in all 10 layers of the retina i.e. RNFL, GCL, IPL, INL, OPL, ONL, IS/OS, OSL, OPR, RPE when considered individually, in both ppOCT and mOCT scans (p<0.05). Combining the outer retinal layers into INL to RPE (INL-RPE) however, showed no significant difference between HM and NM eyes at any mOCT sector. Combining the inner retina layers into GCL to IPL (GC-IPL) also showed no significant difference in mOCT, apart from superiorly (mean HM GC-IPL 85.0±1.6μm, NM 82.7±1.4μm, p=0.03). HM and NM eyes differed in all ppOCT sectors (p<0.05).

Conclusions : HM eyes without VF defects have similar INL-RPE thickness in all macula sectors, and similar GC-IPL thickness apart from superiorly, compared to NM eyes after correcting for magnification error, age and gender. Mapping VF defects in HM to localized INL-RPE or GC-IPL SD-OCT measurements may increase diagnostic accuracy of glaucoma in HM eyes.

This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.