Abstract
Purpose :
To determine the repeatability and reproducibility of Foveal Avascular Zone (FAZ) manual measurements on CIRRUSTM HD-OCT 5000 with AngioPlexTM OCT Angiography (ZEISS, Dublin, CA) on normal and diseased eyes.
Methods :
Two groups of subjects (normal retinas and diseased retinas) were scanned on three commercial CIRRUS devices with AngioPlex. Each study eye was randomized to either the Angio 3X3 mm or 6X6 mm scan pattern and scanned three times each. The order of devices was randomized. Three graders analyzed the AngioPlex Metrix FAZ overlay to determine if the FAZ was acceptable. If it was deemed acceptable, no manual edits were made. Otherwise, the grader used the AngioPlex Metrix FAZ edit tool to outline the FAZ. Analysis of variance was used to calculate the repeatability and reproducibility standard deviations for each measurement. Coefficient of variation (CV) was calculated by dividing these by the mean measurement.
Results :
Twenty seven eyes of seventeen subjects with pathologies such as age-related macular degeneration, diabetic retinopathy, and glaucoma in at least one eye, met the criteria to be included in the diseased eye cohort. Acceptable sets of scans were acquired for 22 eyes, 11 with the Angio 3x3 scans and 11 with the Angio 6x6 scans. Repeatability and reproducibility CV are shown in Table 1 for all groups and all 3 FAZ parameters. The CV for reproducibility is shown as bold if the analysis of variance found the contribution of grader to be significant. Manual grading resulted in significant operator effect for almost all parameters, whereas automated measurements showed an effect of system/operator only for circularity. The operator component was small in all cases, as shown by comparing the reproducibility CV to the repeatability CV. Manual grading improved reproducibility only for the 6x6 scans in the presence of disease.
Conclusions :
Automated measurements of the FAZ may occasionally fail to correctly identify the FAZ. Tools to manually correct the FAZ improve the repeatability and reproducibility of the FAZ measurements to an acceptable range and may allow for monitoring the FAZ over time. This is especially helpful for 6x6 scans in the presence of disease. Overall, the repeatability and reproducibility of FAZ measurements are good, and automated measurements are acceptable for 3x3 scans and 6x6 scans of normal eyes.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.