Abstract
Purpose :
To assess the degree of agreement in morphological parameters of the human optic nerve head
(ONH) measured with in vivo OCT with ex vivo three-dimensional (3D) histology
Methods :
The ONH of 6 eyes from 3 brain-dead organ donors (4 healthy and 2 with glaucomatous eyes) were
imaged by OCT with IOP set to 10 mmHg by anterior chamber cannulation. After organ recovery, the eyes
were enucleated and fixed at 10mmHg, then high-resolution episcopic 3D histomorphometric reconstructions
were created. The surfaces of: Bruch’s membrane (BM) and opening (BMO), anterior sclera (AS) and scleral
canal opening (ASCO), and anterior lamina cribrosa surface (ALCS), were delineated and then interpolated by
analytical 3D fitting models in both the OCT and histology volumetric reconstructions. ALCS visibility area was
quantified as the sum of the areas of the Delaunay mesh elements connecting the delineation points.
Spearman correlations tested the agreement between in vivo vs. ex vivo measures. Limits of agreement (LOA)
were calculated for the relative difference between the two measures, and expressed as % of the mean.
Results :
Bland-Altman plots of the relative difference between in- and ex-vivo 3D measurements are shown in
the Figure. Spearman correlations are shown in the Table. Good correlation and agreement was seen with
parameters defining the canal size (R ~1, Table), with very narrow LOA (<10% in band width). ALCS depth
(R ~ 0.8) and choroidal thickness (R ~ 0.9) had good correlation between ex- and in-vivo measurements.
However, the ALCS was 20-50% deeper and the Choroid was 50-100% thicker when measured in vivo. ALCS
global shape index metrics showed moderate correlation (R of 0.6 and LOA of ~30% in width). On average,
31% of the ALCS defined histologically was visible using in-vivo OCT.
Conclusions :
While there is a good agreement between in vivo and ex vivo measurements of the anterior
ONH canal morphology and moderate agreement in ALCS shape, in vivo imaging provides limited visualization
of the entire ALCS.
This is a 2021 ARVO Annual Meeting abstract.