FAZ measurement on OCTA
Reviewed by Saruban Pasu

This paper aimed to test the AngioVue OCTAs reproducibility and interoperator concordance in evaluating the size of the foveal avascular zone (FAZ). The authors also investigated how the FAZs representation on the OCTA varied after metabolic activity. The right eye of 25 healthy volunteers were imaged. Image 1 was between 8 and 9am; Image 2 was 15 minutes later; Image 3 was 15 minutes later immediately after blue light retinal photo-bleaching; Image 4 was taken one hour later after a physical stress test; Image 5 was taken one hour later after 30 minutes of dark adaptation; Image 6 was taken one hour later. Two graders analysed each of the six images. They first manually calculated the size of the FAZ, then used the automated method. Vessel density was also automatically calculated. For each grader, no differences were found when assessing for reproducibility in either manual or automated FAZ assessment. Interoperators concordance was very good and linear regression concordance models revealed a higher operator / operator concordance, r2 = 0.978 (95% confidence interval 0.981–0.976), than the automated / manual, r2 = 0.689 (95% confidence interval 0.695–0.676) with P values less than 0.001. No intraindividual variability differences were found. The automated calculator provided significantly larger measurement than manual measurement. Vessel density found no significant differences among each of the experimental setting. 

Reproducibility and reliability of optical coherence tomography angiography for foveal avascular zone evaluation and measurement in different settings.
La spina C, Carnevali A, Marchese A, et al.
RETINA
2017;37:1636-41.