Abstract
Purpose :
To evaluate the agreement between Topcon DRI OCT Triton and Topcon 3D OCT-1 Maestro for retina and optic disc measurements in healthy eyes, glaucomatous eyes, and eyes with retinal disease.
Methods :
This study employed three DRI OCT Triton (plus) devices and three 3D OCT-1 Maestro devices. Each operator was specifically paired with one Triton and one Maestro to create three distinct operator/device configurations. Eligible subjects were randomized to select the study eye and the testing order of the operator/device. 25 healthy eyes, 26 eyes with retinal disease, and 25 glaucoma eyes were studied. Parameter measurements included thickness of the full retina, retinal nerve fiber layer, ganglion cell complex, and ganglion cell plus, as well as optic disc measurements like C/D Area, Rim Area, et al. All images of scans were sent to the Topcon Reading Center for evaluation of image acceptance and quality grading. The agreement analysis was conducted utilizing the first acceptable scans from each scan pattern of Triton and Maestro for all study eyes. Agreement between Triton and Maestro was evaluated by the Bland-Altman plots, the 95% limits of agreement (LOA), and the Deming regression for each subject population.
Results :
Image quality of B scans were mostly deemed good or excellent for both Triton and the Maestro. Overall, the measurements were comparable between the Triton and the Maestro. The Bland-Altman plots showed the measurement differences between the two devices were minor across the parameters. Deming regression plots showed the slopes between the two devices were largely centered around +1. Nearly all of the 95% confidence intervals for intercept and slope were not significantly different from zero and one.
Conclusions :
The DRI OCT Triton is substantially equivalent to the 3D OCT-1 Maestro. The agreement results were expected based on minor differences in axial resolution, validated software and algorithm updates, and pixel calibration factor between devices. The differences do not appear to be clinically meaningful.
This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.