Repeatability of measuring the vessel density in patients with retinal vein occlusion: An optical coherence tomography angiography study

Abstract

Purpose: To determine the repeatability of superficial vessel density measurements using Spectral domain Ocular coherence tomography angiography(SD-OCTA) in patients diagnosed with retinal vein occlusion(RVO).

Design: Prospective observational study.

Subjects: Patients who visited our retinal clinic from August 2017 to August 2018, diagnosed with RVO were recruited for the study.

Methods: Two consecutive 3×3 mm pattern scans were performed using the Cirrus HD-OCT 5000 along with AngioPlex software (Carl Zeiss Meditec) in each eye by single skilled examiner. All scans were analyzed using en face OCTA images to measure vessel density (VD) automatically. For further analysis of the effect of central macular thickness(CMT), eyes were divided into two groups according to CMT of 400μm (Group 1: CMT > 400μm, Group 2: CMT < 400μm). To identify factors affecting the repeatability of VD measurements, linear regression analyses were conducted for the coefficient of variation (CV) of VD by investigating demographics and ocular variables.

Main outcome measures: The intraclass correlation coefficient (ICC), coefficient of variation (CV) of VD measurements.

Results: A total of 57 eyes from 57 patients were examined: 35 eyes with BRVO and 22 eyes with CRVO. In all 57 eyes with RVO, the ICC and CV of the full VD(VD of 3mm diameter circle) were 0.800 and 10.61%, respectively. Univariate analyses showed that the mean CMT (B, 0.001; p<0.001) and mean ganglion cell-Inner plexiform layer (GC-IPL) thickness (B, -0.002; p = 0.020) were significant factors that affected the repeatability. Multivariate analyses of these two factors showed that only mean CMT was a significant factor. The ICC and CV of the full VD in group 1 (CMT > 400μm) were 0.348 and 22.55% respectively. In group 2 (CMT < 400μm), the ICC and CV of the full VD were 0.910 and 7.76%, respectively.

Conclusions: The repeatability of VD measurement in eyes with RVO was reasonably comparable to previous studies. Repeatability of VD measurement was significantly affected by central macular thickness.

Fig 1. Bland-Altman plots showing the level of agreement for the VD measurements obtained using AngioPlex optical coherence tomography between two consecutive measurements in patients with RVO.

VD measurements of 7 different areas: (1)Full area, (2) Inner area, (3)Central, (4)Superior, (5)Nasal, (6)Inferior, (7)Temporal. Two dot lines indicate the upper and lower boundaries of the 95% CIs. VD, vessel density.

Fig 2. Scatterplot graph of the differences in VD between two consecutive measurements using optical coherence tomography angiography.

The differences in VD tended to be larger when the CMT was thicker. CMT, central macular thickness; VD, vessel density.