Comparison of Reproducibility of Foveal Avascular Zone Measurement using Optical Coherence Tomography Angiography and Fluorescein Angiography
Khalid Al-kirwi, MD
Khalid Yusuf, MD, Günay Uludağ, MD, Muhammad Hassan, MD, Maria Soledad Ormaechea MD, Qienyuan Zhou, PhD, Muhammad Sohail Halim MD, Anh NT Tran, BS, Rubbia Afridi, MBBS, Jeonghun Bae, MD, Diana V. Do, MD, Yasir J. Sepah, MBBS, Quan Dong Nguyen, MD, MSc
Byers Eye Institute, Stanford University, Palo Alto, CA, USA
Department of Ophthalmology, Imamein Khadhimein Medical City University Hospital
Department of Ophthalmology, Hospital Universitario Austral. Buenos Aires, Argentina.
Optovue Inc, Fremont, California
Department of Ophthalmology, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul, South Korea
Purpose: To compare the reproducibility of foveal avascular zone (FAZ) measurements using semi-automated optical coherence tomography angiography (OCTA) FAZ measurement tool with manual FAZ calculations performed on fluorescein angiography (FA).
Methods: Eyes of patients with different retinal diseases from 3 clinical sites were enrolled in the study. A 3x3 mm foveal OCTA scan captured using Optovue Angiovue and an early phase FA image of the same eye captured using regular fundus camera were utilized for FAZ measurements. The automated FAZ measurement tool of the ReVue software (Version: 2017.1.0.151) was utilized to measure the area of the FAZ on OCTA scans. Manual adjustments to the FAZ measurements were performed. Image J (Version: 1.52a) was utilized to calculate the area of FAZ on FA images. Three independent graders performed these measurements on OCTA and FA images. The inter-grader reproducibility of FAZ measurements was assessed for both OCTA and FA images using concordance correlation coefficients; Bland & Altman plots were created. Automated FAZ measurements that required manual correction of ≥50% of original FAZ value were identified.
Results: Thirty eyes (26 subjects; 12 females) were analyzed in the study. The case series consisted of: 4 eyes (normal); 3 [neovascular age-related macular degeneration (AMD)]; 7 (non-proliferative diabetic retinopathy); 5 (proliferative diabetic retinopathy); 4 (branch retinal vein occlusion); and 7 (other retinal pathologies: dry AMD, central serous retinopathy, macular telangiectasia, and epiretinal membrane). Correlation coefficients of the FAZ measurements using FA and OCTA among graders are shown (Table). Measurements of FAZ by OCTA showed consistently higher reproducibility among the graders compared to FA measurements (Figure). Manual correction of ≥ 50% in the automated FAZ measurements by OCTA occurred in ≤16.6% of the cases.
Conclusions: OCTA is a non-invasive technique which allows for higher reproducibility and inter-grader agreement compared to FA for FAZ measurements.