Hui Xiao1, Xing Liu2, Lingling Liao1, Kaixin Tan1, Yunlan Ling1, Yimin Zhong1. 1. State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, Guangzhou, China. 2. State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, Guangzhou, China, drliuxing@163.com.
Abstract
PURPOSE: To assess the reproducibility of quantitative measurements of the foveal avascular zone (FAZ) and retinal vasculature determined by different scanning protocols of optical coherence tomography angiography (OCTA) in healthy volunteers. METHOD: All participants were scanned by two trained operators using an AngioPlex OCTA. Both angiography protocols (6 × 6 mm and 3 × 3 mm) were performed three times on the same eye by operator A and one additional time by operator B. The FAZ area and perimeter, retinal vessel length density (VLD) and perfusion density (PD) of different regions were analysed. RESULTS: Fifty-two eyes from 52 subjects were recruited for this study. The repeated measurements of FAZ, VLD, and PD parameters obtained by the same operator, as well as by different operators, were not significantly different when the same protocol was used (p > 0.05). The intra- and inter-operator intraclass correlation coefficients (ICCs) of the FAZ and central VLD and PD parameters (range, 0.99-0.95) were better than the intra- and inter-operator ICCs of VLD and PD in the inner and outer rings (range, 0.86-0.90). The FAZ area, perimeter, and VLDs obtained by the 3 × 3 mm protocol were larger than those obtained by the 6 × 6 mm protocol (p < 0.01), but the PDs obtained by the 3 × 3 mm protocol were smaller than those obtained by the 6 × 6 mm protocol (p < 0.001). All of the corresponding parameters obtained by the two protocols were positively correlated (r = 0.64-0.99, p < 0.001). CONCLUSION: Both the 6 × 6 and 3 × 3 mm protocols of the AngioPlex OCTA provide good reproducibility for assessing the FAZ and superficial retinal vasculature. However, the values obtained by these different protocols cannot be compared directly.
PURPOSE: To assess the reproducibility of quantitative measurements of the foveal avascular zone (FAZ) and retinal vasculature determined by different scanning protocols of optical coherence tomography angiography (OCTA) in healthy volunteers. METHOD: All participants were scanned by two trained operators using an AngioPlex OCTA. Both angiography protocols (6 × 6 mm and 3 × 3 mm) were performed three times on the same eye by operator A and one additional time by operator B. The FAZ area and perimeter, retinal vessel length density (VLD) and perfusion density (PD) of different regions were analysed. RESULTS: Fifty-two eyes from 52 subjects were recruited for this study. The repeated measurements of FAZ, VLD, and PD parameters obtained by the same operator, as well as by different operators, were not significantly different when the same protocol was used (p > 0.05). The intra- and inter-operator intraclass correlation coefficients (ICCs) of the FAZ and central VLD and PD parameters (range, 0.99-0.95) were better than the intra- and inter-operator ICCs of VLD and PD in the inner and outer rings (range, 0.86-0.90). The FAZ area, perimeter, and VLDs obtained by the 3 × 3 mm protocol were larger than those obtained by the 6 × 6 mm protocol (p < 0.01), but the PDs obtained by the 3 × 3 mm protocol were smaller than those obtained by the 6 × 6 mm protocol (p < 0.001). All of the corresponding parameters obtained by the two protocols were positively correlated (r = 0.64-0.99, p < 0.001). CONCLUSION: Both the 6 × 6 and 3 × 3 mm protocols of the AngioPlex OCTA provide good reproducibility for assessing the FAZ and superficial retinal vasculature. However, the values obtained by these different protocols cannot be compared directly.