Abstract
Purpose :
Quantification of vessel density in the peripapillary region using OCT Angiography (OCTA) has been shown to be of potential interest for managing glaucoma. When monitoring such quantitative data over time, it is important to know if measurements on different instruments are expected to be comparable. We conducted a clinical study to compare the measurements of optic nerve head (ONH) perfusion and flux density on two models of CIRRUS OCT.
Methods :
Twenty healthy eyes and ten diseased eyes (a total of 28 subjects) were imaged on CIRRUS™ HD-OCT 5000 with AngioPlex® OCT Angiography (ZEISS, Dublin, CA) using 11.5.0 software and CIRRUS™ 6000 with AngioPlex® OCT Angiography (ZEISS, Dublin, CA) using 11.5.1 software. Subjects were categorized into two groups: Healthy eyes and eyes with mild, moderate or severe glaucoma.
The Optic Nerve Angiography 4.5x4.5mm scan pattern, which captures 350 A-scans/B-Scan and 350 B-Scans, was used to obtain images on both devices. ONH AngioPlex Metrix examines the radial peripapillary capillary slab generated from the 3-dimensional OCTA data and characterizes perfusion (area of blood vessels) and flux index (intensity of capillary perfusion). The tool quantifies local regions of tissue within a ring centered at the optic nerve head. Linear regression and Bland-Altman analysis were used to compare measurements from the two devices.
Results :
Linear regression plots show good correlation (r2=0.74-0.95) between the two devices in measuring perfusion and flux in the peripapillary region. Bland-Altman analysis (see Table 1) shows that CIRRUS 6000 metrics are significantly higher than CIRRUS HD-OCT 5000 metrics (P < 0.05), possibly due to the improved CIRRUS 6000 signal to noise ratio providing better visualization and detection of vessel structures.
Conclusions :
Both devices can be used to measure density and flux of vasculature, therefore both devices may successfully aid in the detection and management of ocular disease. Due to the improved signal on the CIRRUS 6000, metrics should not be compared between the models.
This is a 2020 Imaging in the Eye Conference abstract.