Abstract
Purpose :
The axons of retinal ganglion cells in the optic nerve head (ONH) can be visualized and quantified with 3D-OCT. Thickness as well as cross-sectional surface area have been suggested as possible quantification methods to assess pathological loss of axons in glaucoma. Currently, there is no consensus on a preferred method. This study aimed to estimate the minimal cross-sectional surface area in non-glaucomatous subjects with a custom-made computational algorithm. The cross-sectional surface area, Pigment epithelium central limit - Inner limit of the retina Minimal Area (PIMA), was calculated over the circumference of the ONH as PIMA-2π.
Methods :
One eye from each of 40 non-glaucomatous subjects aged 20-69 equally stratified according to age group and gender was estimated for minimal cross-sectional surface area. Image volumes were acquired with 3D-OCT Topcon models OCT2000 and Triton, commonly available at the clinic. Confidence intervals and linear regression were used for statistical evaluation.
Results :
A 95% confidence interval for PIMA-2π was estimated to 2.03 ± 0.14 mm2. In each age group, PIMA-2π was estimated to 2.02 ± 0.36 mm2, 2.35 ± 0.54 mm2, 2.13 ± 0.26 mm2, 1.78 ± 0.30 mm2 and 1.87 ± 0.23 mm2 for ages [20;29] [30;39] [40;49] [50;59] [60;69] respectively. The correlation coefficient estimated as a 95% CI for k was -9.48 ± 9.07 × 10-3 mm2/year (D.f.= 38).
Conclusions :
Presented values of cross-sectional surface area are slightly higher but of the same magnitude as measurements done with Heidelberg Spectralis 3D-OCT in similar studies. This can be attributed to our custom computational algorithm which takes undulations of the cross-sectional area into account. Furthermore, the results imply a loss of retinal ganglion cell axons in non-glaucomatous subjects as measured in cross-sectional surface area. Due to a small sample size and large variation among individuals, definite conclusions cannot be drawn on typical loss in non-glaucomatous subjects.
This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.