In vivo Longitudinal Changes in Lamina Cribrosa Microarchitecture and Optic Nerve Head Structure in Experimental Glaucoma

Purpose: Glaucoma is the second leading cause of blindness worldwide. Substantial evidence supports the idea that retinal ganglion cell axons are initially damaged at the lamina cribrosa in glaucoma. The goal of this dissertation was to better understand early changes in lamina cribrosa microarchitecture and optic nerve head (ONH) structure in experimental glaucoma (EG) and determine whether these changes occur prior to changes in conventional clinical measures of vision loss.

Methods: 1) Images of the anterior lamina cribrosa surface (ALCS) were acquired with an adaptive optics scanning laser ophthalmoscope (AOSLO) in 4 or more imaging sessions in 2 normal rhesus monkey eyes and 3 normal human eyes. Mean ALCS pore area, elongation, and nearest neighbor distance [NND] were quantified in each imaging session and measurement repeatability was assessed. 2) Spectral domain optical coherence tomography (SDOCT) imaging was used to acquire radial B-scan images of the ONH and quantify Bruch’s Membrane Opening (BMO) area, mean ALCS depth (ALCSD), and mean minimum rim width (MRW) in 6 normal rhesus monkeys. ALCS pore geometry was quantified globally, within 60° sectors, and in central and peripheral regions from AOSLO images acquired in the same eyes. Inter-eye differences in pore geometry and ONH structure were quantified. 3) SDOCT and AOSLO images of the ONH and ALCS were acquired before and approximately every 2 weeks after inducing unilateral EG in 7 rhesus monkeys. Mean ALCSD, ALCS radius of curvature, and MRW were quantified from SDOCT B-scans. Retinal nerve fiber layer thickness (RNFLT) was quantified from 12° SDOCT circular scans centered on the ONH. Mean pore geometry was quantified globally, sectorally, and regionally in 3D transformed AOSLO images. Longitudinal changes in ONH structure, ALCS pore geometry, and RNFLT were examined. 4) Finally, functional measures of vision loss [multifocal electroretinogram (mfERG) and standard automated perimetry (SAP)] were also recorded longitudinally in 3 EG monkeys. Onsets of mfERG-measured reductions in the multifocal photopic negative response (mfPhNR) and SAP-measured losses in visual field sensitivity were compared to onsets of change in ONH and ALCS parameters.

Results: 1) Pore areas ranged from 90 to 4,365 µm2 across monkeys and 154 to 6,637 µm2 across humans. Intersession variability in measuring pore geometry was small in normal monkey and human eyes: 6.1% and 8.3% for pore area, 6.7% and 7.7% for pore elongation, and 5.2% and 4.1% for pore NND, respectively (P>.05). 2) Mean differences in BMO area, ALCSD, and MRW between fellow, normal eyes were small: 0.07 ± 0.05 mm2, 10.0 ± 6.9 µm, and 7.9 ± 6.7 µm, respectively (P>.05). Inter-eye differences in mean global pore area, elongation, and NND were not significant in 4/6 monkeys, 6/6 monkeys, and 3/6 monkeys, respectively (P>.05). Inter-eye differences in mean pore area and NND were measured between superotemporal sectors in 2 monkeys and temporal sectors in 2 monkeys (P<.05). 3) Mean ALCSD was the first parameter to change in 6/7 EG eyes. Mean MRW changed first in one EG eye. RNFLT changed last in 5/7 EG eyes. Changes in mean global and local pore geometry were the first or second changes detected in 4/7 EG eyes. At the first time-point of change in pore geometry, mean pore area increased globally (+24.7%), centrally (+23.6%), peripherally (+29.9%), and temporally (+35.5%) (P<.05). Mean pore NND increased globally (+13.3%) and in all sectors and regions (mean of +16.1%) (P<.05), except inferotemporally. Local changes in pore parameters were measured prior to a global change in the same parameter in 3 monkeys. 4) Mean ALCSD was the first significant structural or functional change in 2/3 EG eyes, while a reduction in mfPhNR amplitude was measured first in 1 EG eye. mfPhNR amplitude changed prior to laminar pore microarchitecture in 2/3 EG eyes, while simultaneous changes in these parameters occurred in 1 EG eye. RNFLT and visual field sensitivity were the last parameters to initially change in all monkeys.

Conclusions: Structural changes in ALCS position (mean ALCSD) and microarchitecture occurred prior to significant losses in RNFLT suggesting early laminar remodeling in response to chronically elevated intraocular pressure in experimental glaucoma. AOSLO and SDOCT structural imaging and mfERG functional recording provide the opportunity to earlier detect glaucomatous damage than current clinical tests.