Herrera Gissel, Cheng Yuxuan, Attiku Yamini, Hiya Farhan E, Shen Mengxi, Liu Jeremy, Lu Jie, Berni Alessandro, Trivizki Omer, Li Jianqing, O'Brien Robert C, Gregori Giovanni, Wang Ruikang K, Rosenfeld Philip J
Department of Ophthalmology, Bascom Palmer Eye Institute, University of Miami Miller School of Medicine, Miami, Florida.
Department of Bioengineering, University of Washington, Seattle, Washington.
Ophthalmol Sci. 2024 Oct 21;5(2):100633. doi: 10.1016/j.xops.2024.100633. eCollection 2025 Mar-Apr.
Spectral-domain OCT angiography (SD-OCTA) scans were used in an algorithm developed for swept-source OCT angiography (SS-OCTA) scans to determine if SD-OCTA scans yielded similar results for the measurement of hyperreflective foci (HRF) in intermediate age-related macular degeneration (iAMD).
Retrospective study.
Forty eyes from 35 patients with iAMD.
Patients underwent SD-OCTA and SS-OCTA imaging at the same visit using a 6 × 6 mm OCTA scan pattern. Hyperreflective foci were detected as hypotransmission defects on en face structural images generated from a custom slab positioned 64 to 400 μm beneath Bruch's membrane and confirmed on corresponding B-scans by the presence of well circumscribed lesions within the neurosensory retina or along the retinal pigment epithelium (RPE) that are of equal or greater reflectivity than that of the RPE. Two independent graders evaluated the en face images and B-scans for the presence of these lesions. Outlines of HRF on en face images were generated using a published semiautomated algorithm developed for SS-OCTA scans and manually corrected by the graders when necessary. The total area measurements of the HRF within the 5-mm circle centered on the fovea were obtained from the algorithm using each imaging method.
Agreement of the square root (sqrt) of the HRF total areas obtained from SS-OCTA and SD-OCTA.
The sqrt total areas of the HRF from both imaging modalities were highly concordant, with Lin's concordance correlation coefficient (r) of 0.94 (95% confidence interval: 0.86-0.97; < 0.001). The mean sqrt of the total HRF area measurements identified using SS-OCTA and SD-OCTA imaging were 0.390 mm (standard deviation [SD]: 0.170) and 0.393 mm (SD: 0.187), respectively with mean difference of -0.003 (95% confidence interval: -0.021 to 0.015; =0.76).
Spectral-domain OCT angiography scans yielded results similar to SS-OCTA scans when the same semiautomated algorithm was used to measure HRF in the central 5 mm of the macula, suggesting that either a single 6 × 6 mm SD-OCTA or a SS-OCTA scan pattern can be used to determine the total macular HRF burden in eyes with age-related macular degeneration.
Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.
将谱域光学相干断层扫描血管造影(SD - OCTA)扫描用于为扫频源光学相干断层扫描血管造影(SS - OCTA)扫描开发的一种算法中,以确定SD - OCTA扫描在测量中度年龄相关性黄斑变性(iAMD)中的高反射灶(HRF)时是否能产生相似的结果。
回顾性研究。
35例iAMD患者的40只眼。
患者在同一次就诊时使用6×6 mm的OCTA扫描模式进行SD - OCTA和SS - OCTA成像。高反射灶在位于布鲁赫膜下方64至400μm的自定义层面生成的正面结构图像上被检测为低透射缺陷,并在相应的B扫描中通过神经感觉视网膜内或沿视网膜色素上皮(RPE)存在边界清晰且反射率等于或高于RPE的病变来确认。两名独立的分级人员评估正面图像和B扫描中这些病变的存在情况。正面图像上HRF的轮廓使用为SS - OCTA扫描开发的已发表的半自动算法生成,并在必要时由分级人员手动校正。使用每种成像方法从算法中获取以黄斑中心凹为中心的5mm圆内HRF的总面积测量值。
SS - OCTA和SD - OCTA获得的HRF总面积的平方根(sqrt)的一致性。
两种成像方式的HRF的sqrt总面积高度一致,林氏一致性相关系数(r)为0.94(95%置信区间:0.86 - 0.97;P < 0.001)。使用SS - OCTA和SD - OCTA成像确定的HRF总面积测量值的平均sqrt分别为0.390 mm(标准差[SD]:0.170)和0.393 mm(SD:0.187),平均差值为 - 0.003(95%置信区间: - 0.021至0.015;P = 0.76)。
当使用相同的半自动算法测量黄斑中心5mm范围内的HRF时,谱域光学相干断层扫描血管造影扫描产生的结果与SS - OCTA扫描相似,这表明单次6×6 mm的SD - OCTA或SS - OCTA扫描模式均可用于确定年龄相关性黄斑变性患者黄斑区HRF的总负担。
在本文末尾的脚注和披露中可找到专有或商业披露信息。