Repeatability of metrics from swept-source optical coherence tomography angiography (SS-OCTA) in normal and glaucomatous eyes

Abstract

Purpose : To evaluate the inter and intra-visit reproducibility of quantitative metrics measured from SS-OCTA.

Methods : Subjects with glaucoma (any age) and healthy controls (>40 years) were prospectively recruited from three sites. If two eyes were eligible, one eye per subject was designated randomly as the study eye. Each eye was imaged using the Angio 6x6 mm scan centered on the macula and centered on the optic disc on PLEX^® Elite 9000 (ZEISS, Dublin, CA). Scans were repeated at each location on each device 3 times per visit, and the subjects were expected to complete 3 visits within one month.
Data was processed on the Advanced Imaging Network Hub using the algorithm “Peripapillary Nerve Fiber Layer Microvasculature Density, v0.9”, which quantifies microcirculation in the radial peripapillary capillary layer, summarizing data over an annulus centered at the optic disc. Large retinal vessels are removed. Perfusion Density (PD) is defined as the total area of perfused microvasculature per unit area, and ranges from 0 (no perfusion) to 1 (fully perfused). Because there are naturally non-perfused regions between vessels, PD values typically remain below 0.5. Flux index is defined as the total weighted area of perfused microvasculature per unit area in a region of measurement. The weight is the normalized flow intensity corresponding to each pixel.
The combined inter and intra-visit coefficient of variance (CV%) is calculated as the square root of the average of the intra-individual variances divided by the mean of all measurements.

Results : Twelve control eyes (mean age 50.1 years, SD 10) and fourteen eyes with glaucoma (mean age 58.6 years, SD 11) were imaged. Table 1 shows the population mean (and standard deviation) and percent coefficient of reproducibiliy (also known as coefficient of variance or CV%) for these subjects for perfusion density and flux index. CV% is excellent. Figure 1 shows the average measurements for each population for flux index and PD as a function of age. This plot shows that even though the control eyes were not age-matched, the measurements are likely to be able to discriminate between normal eyes and glaucoma eyes.

Conclusions : Coefficient of intervisit reproducibility is excellent for flux index and perfusion density as measured on PLEX Elite 9000. These parameters may be useful for monitoring glaucomatous damage.

This is a 2021 Imaging in the Eye Conference abstract.

 

Table 1

 

Figure 1