Department of Ophthalmology, Friedrich-Alexander University of Erlangen-Nürnberg, Erlangen, Germany.
Department of Computer Science 5, Friedrich-Alexander University of Erlangen-Nürnberg, Erlangen, Germany.
Ophthalmologica. 2020;243(1):75-84. doi: 10.1159/000502458. Epub 2019 Sep 11.
The aim of the present study was to investigate the reliability of macular microvasculature measurements in normal subjects by Heidelberg Spectralis II optical coherence tomography angiography (OCT-A) in combination with a newly made software.
This prospective study included 23 eyes of 23 persons from the Erlangen Glaucoma Registry (ISSN 2191-5008, CS-2011; NTC00494923). The subjects underwent a complete clinical, standardized ophthalmologic examination to rule out any eye disease. En face OCT-A imaging was done using Heidelberg Spectralis II OCT (Heidelberg, Germany). Images were recorded with a 15 × 15° angle and a lateral resolution of 5.7 µm/pixel, resulting in a retinal section of 2.9 × 2.9 mm. The Erlangen-Angio-Tool (EA-Tool) OCT-A application performed multiple segmentations, allowing analysis of the vessel density in 12 segments. The software was coded in MATLAB. Macular data on the superficial vascular plexus (SVP), intermediate capillary plexus (ICP), and deep capillary plexus (DCP) were exported into the application and analyzed separately. The EA-Tool calculated the percentage of "white area" in the "total area" of the region of interest, called vessel density. Foveolar avascular zones (FAZs) of the SVP, ICP, and DCP were calculated manually. To investigate the reproducibility of the new software, individual scans (SVP, ICP, and DCP) were analyzed twice with the EA-Tool and intraclass coefficients (ICCs) of the vessel density values were calculated. Statistical analysis was performed with SPSS version 21.0.
The mean vessel density of the SVP ranged between 30.4 and 33.5, that of the ICP between 20.9 and 24.7, and that of the DCP between 23.5 and 27.6. Bland-Altman plots showed a good reliability of two consecutive scans of each sector (S1-S12) in the SVP, ICP, and DCP. Testing reproducibility, no statistically significantly different sectorial coefficients of variation of the SVP, ICP, and DCP were observed (p > 0.05). The mean FAZ area of the SVP was 0.43 ± 0.16 mm2, that of the ICP 0.28 ± 0.1 mm2, and that of the DCP 0.44 ± 0.12 mm2.
Spectralis OCT II, in combination with the semiautomated vessel density software EA-Tool, showed good or even excellent ICCs in 75% of all segments of the SVP, ICP, and DCP. The ICCs for the FAZ area in the SVP, ICP, and DCP were excellent.
本研究旨在使用海德堡 Spectralis II 光学相干断层扫描血管造影(OCT-A)结合新开发的软件,评估正常人黄斑微循环测量的可靠性。
本前瞻性研究纳入了来自埃尔兰根青光眼登记处(ISSN 2191-5008,CS-2011;NTC00494923)的 23 只眼、23 名受试者。受试者接受了全面的临床、标准化眼科检查,以排除任何眼病。使用海德堡 Spectralis II OCT(德国海德堡)进行了共焦 OCT-A 成像。图像以 15×15°的角度和 5.7 µm/pixel 的横向分辨率记录,产生 2.9×2.9mm 的视网膜截面。使用 Erlangen-Angio-Tool(EA-Tool)OCT-A 应用程序进行了多次分段,允许在 12 个分段中分析血管密度。软件使用 MATLAB 编写。将黄斑区浅层血管丛(SVP)、中间毛细血管丛(ICP)和深层毛细血管丛(DCP)的数据导出到应用程序中并分别进行分析。EA-Tool 计算了“目标区域”中“白色区域”占“总面积”的百分比,称为血管密度。手动计算 SVP、ICP 和 DCP 的黄斑无血管区(FAZ)。为了研究新软件的可重复性,使用 EA-Tool 对单个扫描(SVP、ICP 和 DCP)进行了两次分析,并计算了血管密度值的组内系数(ICC)。使用 SPSS 版本 21.0 进行统计分析。
SVP 的平均血管密度范围在 30.4 到 33.5 之间,ICP 的平均血管密度范围在 20.9 到 24.7 之间,DCP 的平均血管密度范围在 23.5 到 27.6 之间。Bland-Altman 图显示 SVP、ICP 和 DCP 的每个扇区(S1-S12)的两次连续扫描具有良好的可靠性。测试重复性时,SVP、ICP 和 DCP 的扇区系数变化没有统计学上的显著差异(p>0.05)。SVP 的平均 FAZ 面积为 0.43±0.16mm2,ICP 为 0.28±0.1mm2,DCP 为 0.44±0.12mm2。
Spectralis II OCT 与半自动血管密度软件 EA-Tool 相结合,在 SVP、ICP 和 DCP 的 75%的所有节段均显示出良好甚至优秀的 ICC。SVP、ICP 和 DCP 的 FAZ 面积的 ICC 非常优秀。