Purchase this article with an account.
Ali Fard, Homayoun Bagherinia, Archana Kolli, Mary K Durbin, Jochen Straub; Accuracy and reproducibility of peripapillary microvasculature quantification using ZEISS AngioPlex Metrix™. Invest. Ophthalmol. Vis. Sci. 2018;59(9):5071.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Vasculature analysis using optical coherence tomography angiography (OCTA) has been suggested as a valuable metric in diagnosis of eye diseases such as glaucoma. The purpose of this study is to investigate the accuracy and reproducibility of capillary perfusion measurement within peripapillary retinal nerve fiber layer (RNFL) among normal and glaucoma subjects using AngioPlex Metrix (ZEISS, Dublin, CA).
Normal (N=14) and glaucoma (N=15) subjects were enrolled in this study. One eye per subject was randomly selected for the purpose of evaluating the accuracy and reproducibility of quantification. Each eye was scanned using a CIRRUS HD-OCT 5000 (ZEISS, Dublin, CA) OCTA scan (350x350 A-scans; 4.5mm x 4.5mm) 3 times on 3 instruments. Subsequently, the OCT data were segmented, enface RNFL vasculature maps were created, and capillary areas were measured using an automated algorithm. Capillary perfusion density and flux index were calculated over an annulus with inner and outer diameters of 2mm and 4.5 mm, centered at the optic disc. To evaluate the accuracy of perfusion density, 5 normal and 5 glaucoma images were randomly selected and four regions of interest (ROI) were identified in each image. Three graders trained in segmenting OCTA images marked the capillary areas within the ROIs. The results were then compared with the algorithm output. To evaluate the reproducibility, an analysis of variance was used to calculate the intraclass correlation (ICC), coefficient of variation, and reproducibility standard deviation among the combined normal and glaucoma population.
Capillary perfusion density calculated by the algorithm was compared with the mean of graders’ marking. Fig. 1 shows the Bland-Altman plot of the mean of graders and the algorithm. In 39 out of 40 test cases, the differences between mean of graders and algorithm were found to be within the 95% limits of agreement. The ICC for capillary flux index and perfusion density were >0.70 and >0.86, respectively. Table 1 summarizes the results for each quadrant within the annulus.
Our results suggest that peripapillary microvasculature quantification using AngioPlex Metrix is highly accurate compared with manual grading and is repeatable and reproducible for population of normal and glaucoma.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.
Fig 1. Agreement between manual grading and algorithm.
Table 1. Summary of accuracy and reproducibility results
This PDF is available to Subscribers Only