Evaluating The Estimation of Neuronal Loss in Glaucoma



Journal Title

Journal ISSN

Volume Title



PURPOSE: In vivo imaging with optical coherence tomography (OCT) is widely used in the diagnosis and monitoring of glaucoma and other optic neuropathies. With spectral domain OCT (SD OCT), cross-sectional images of the retinal nerve fiber layer (RNFL), optic nerve head (ONH) and the macula can be acquired at high resolutions and quantified with excellent repeatability and reproducibility. While OCT morphological measures are clinically used as to represent retinal ganglion cell (RGC) content, the correspondence with RGC content over the time course of glaucoma progression is not fully understood. The goal of this dissertation was to investigate the relationship between SD OCT measures of retinal structures and their histological correlates. METHODS: 1) In the first experiment, data was collected from both the control and experimental glaucoma eyes of 15 macaque monkeys. At varying endpoints, ocular magnification corrected RNFL and minimum rim width (MRW) thickness measures were acquired from both eyes of each subject and compared with their respective retrobulbar optic nerve axon counts. 2) The second experiment investigated the relationship between OCT derived ganglion cell inner plexiform layer (GCIPL) thickness and RGC density in corresponding areas of the macula. Macula tissue from both control and experimental glaucoma eyes in 4 monkey subjects were collected, processed and labelled for confocal imaging. Using well registered histological and in vivo data, the relationship between GCIPL and cell density was then established. 3) The third experiment investigated the structural correspondence between GCIPL thickness in the macula and corresponding RNFL thickness around the optic nerve head. Ninety-eight (98) young, normal human subjects were recruited for this study. The pathway relating macula RGCs to their axonal entry point at the optic disc were manually traced in 24 eyes. The resulting retinotopic map was used to investigate the relationship between ocular magnification corrected RNFL area and macular GCIPL thickness measures. RESULTS: 1) Axon count study: Axon count (1,308,155 ± 108,382) is linearly related to RNFL thickness (r²=0.96, p<0.01). The axon count-MRW relationship is best represented by an exponential rise to maximum fit (r²=0.86, p<0.01). 2) RGC quantification study: Peak cell density in the control eyes was 62,699 ± 2291 cells/mm2 at an eccentricity of 2.8° from the center of the fovea. The relationship between GCIPL thickness and RGC density is best described by an eccentricity dependent function (Cell density(cells/mm2) = -7281 + 815 x GCIPL(µm) – 2497 x Eccentricity (degrees), r2 = 0.74, p<0.01). 3) Structural correspondence study: RGC axons traced from the central 20° of the macula subtends angles of 45.13 ± 14.7° superiorly and 45.93 ± 12.5° inferiorly at the optic disc. Hemifield GCIPL thickness has minimal correlation with corresponding sectoral RNFL areas either superiorly (r2 = 0.009, p=0.42) or inferiorly (r2 = 0.02, p=0.36). CONCLUSIONS: 1) Using rigorous methods for both OCT and histological quantification, there is an excellent relationship between neuro-retinal rim, circumpapillary and histological measures of axonal content. 2) The linear relationship between GCIPL thickness and RGC density is eccentricity-dependent and important to consider when OCT macula measures are used to represent cell density. 3) In healthy human subjects, there is minimal relationship between OCT macula and nerve fiber thickness measures which could be as a result of inter-individual variability of the structural measures.



Glaucoma, Vision sciences, Optical coherence tomography (OCT), Macula