High sampling resolution optical coherence tomography reveals potential concurrent reductions in ganglion cell‐inner plexiform and inner nuclear layer thickness but not in outer retinal thickness in glaucoma

Abstract

Purpose To analyse optical coherence tomography (OCT)‐derived inner nuclear layer (INL) and outer retinal complex (ORC) measurements relative to ganglion cell‐inner plexiform layer (GCIPL) measurements in glaucoma. Methods Glaucoma participants (n = 271) were categorised by 10‐2 visual field defect type. Differences in GCIPL, INL and ORC thickness were calculated between glaucoma and matched healthy eyes (n = 548). Hierarchical cluster algorithms were applied to generate topographic patterns of retinal thickness change, with agreement between layers assessed using Cohen's kappa (κ). Differences in GCIPL, INL and ORC thickness within and outside GCIPL regions showing the greatest reductions and Spearman's correlations between layer pairs were compared with 10‐2 mean deviation (MD) and pattern standard deviation (PSD) to determine trends with glaucoma severity. Results Glaucoma participants with inferior and superior defects presented with concordant GCIPL and INL defects demonstrating mostly fair‐to‐moderate agreement (κ = 0.145–0.540), which was not observed in eyes with no or ring defects (κ = −0.067–0.230). Correlations (r) with MD and PSD were moderate and weak in GCIPL and INL thickness differences, respectively (GCIPL vs. MD r = 0.479, GCIPL vs. PSD r = −0.583, INL vs. MD r = 0.259, INL vs. PSD r = −0.187, p = <0.0001–0.002), and weak in GCIPL‐INL correlations (MD r = 0.175, p = 0.004 and PSD r = 0.154, p = 0.01). No consistent patterns in ORC thickness or correlations were observed. Conclusions In glaucoma, concordant reductions in macular INL and GCIPL thickness can be observed, but reductions in ORC thickness appear unlikely. These findings suggest that trans‐synaptic retrograde degeneration may occur in glaucoma and could indicate the usefulness of INL thickness in evaluating glaucomatous damage.

Characterisation of the normal human ganglion cell–inner plexiform layer using widefield optical coherence tomography

“…However, due to the volume of data owing to the high sampling density applied in this study, there is the potential for an excessive number of clusters, or over-assignment using conventional separability criteria. 51,70,71 As such, the silhouette coefficient was derived from calculations of the distance between data points within a cluster and to data points in the nearest neighbouring cluster, and a minimum silhouette coefficient of 0.5 was used to determine the maximum number of suitable clusters, 72,73 which was subsequently applied as a criterion in the hierarchical cluster analysis. Distance measures applied between clusters were Euclidean distance for two-step cluster analysis and squared Euclidean distance for hierarchical cluster analysis, while within-groups linkage was applied in hierarchical cluster analysis to determine similarity within clusters.…”

“…To equalise scaling factors between participants, ensuring corresponding locations were compared between participants, and facilitate further analyses while retaining a sufficient sampling resolution, GCIPL measurements were averaged across grid squares measuring 100 × 100 μm, similar to previously described methods, 51,52 over a total grid area of 160 × 160 grid squares. While this exceeded the total scan area for most participants, this was performed to ensure all data were captured in averaged grid square measurements.…”

“…An advantage of both approaches over other algorithms is that they do not require a priori assumptions on the number of most suitable clusters, which can instead be determined using statistical criteria such as minimum separability between clusters. However, due to the volume of data owing to the high sampling density applied in this study, there is the potential for an excessive number of clusters, or over‐assignment using conventional separability criteria 51,70,71 . As such, the silhouette coefficient was derived from calculations of the distance between data points within a cluster and to data points in the nearest neighbouring cluster, and a minimum silhouette coefficient of 0.5 was used to determine the maximum number of suitable clusters, 72,73 which was subsequently applied as a criterion in the hierarchical cluster analysis.…”