Patients
This retrospective observational study enrolled 47 eyes of 47 patients with unilateral BRVO who were examined at the Department of Ophthalmology of Peking Union Medical College Hospital (PUMCH), Beijing, China between January 2018 and December 2018. Treatment naïve patients and patients who have been treated with intravitreal medication were included. The exclusion criteria were patients with poor-quality images on OCTA (quality index lower than 5) because of significant eye movements, lens opacities, previous retinal surgery, pathologic myopia, ocular trauma, and the presence of other retinal diseases such as diabetic retinopathy or age-related macular degeneration. Eyes with CRVO or hemicentral retinal vein occlusion (HRVO) were excluded.
The clinically unaffected fellow eyes of the patients with unilateral BRVOs served as one of the control groups if there was no history of any ocular disease or ocular surgery and with an unremarkable result of the ophthalmologic examination, including a normal appearance of both the anterior and posterior segment of the eye, a normal intraocular pressure. A second control group consisted of 47 normal individuals with no history of any ocular diseases or ocular surgery. The ophthalmic examination of the normal individuals was unremarkable. One eye of each normal individual was randomly chosen as the control eye.
The following data were obtained from the medical records of our enrolled patients: age, gender, best corrected visual acuity (BCVA, logMAR), slit lamp-assisted biomicroscopy of the anterior segment, intraocular pressure, fluorescein angiography (FA), SD-OCT (Spectralis Heidelberg Engineering, Heidelberg, Germany), and OCTA using the AngioVue OCTA system version 2017.1 (Optovue Inc., Fremont, CA, USA).
Macular microvascular OCTA imaging
Macular OCT angiograms were acquired using the AngioVue OCTA system version 2017.1 (Optovue Inc., Fremont, CA, USA) with the Angio Retina mode. A newly developed 3D projection artifact removal (PAR) algorithm was included in the software. With this algorithm, the software differentiates in situ OCTA signal from projection artifacts based on the information from OCT and OCTA volume and removes the projection artifacts.
For each eye, a 3 × 3-mm volume image centered on the fovea was obtained. The scan pattern was a 304 B-scan raster with 304 A-scans per B-scan. Two orthogonal volumes were acquired at each scanning location. The AngioVue software segmented the vascularized tissue into four layers based on the default settings: the superficial vascular complex (SVC), the deep vascular complex (DVC), the outer retinal layer, and the choriocapillaris layer. The boundaries of the superficial network extended from the internal limiting membrane to 10 μm above the inner plexiform layer (IPL). The deep capillary network extended from 10 μm above the IPL to 10 μm below the outer plexiform layer (OPL),no overlap existed between the 2 slabs.
FAZ metrics including size, perimeter, foveal acircularity index(AI), and foveal vessel density 300(FD-300) were evaluated with the software(Table 3). AI is the ratio between the measured perimeter and the perimeter of the same size circular area: the closer the shape is to the circle, the closer the value is to 1. FD-300 is the vessel density in a 300-mm wide area surrounding the FAZ, including both SVC and DVC. FAZ area is excluded in measurement of vessel density in this area, due to its high variability among different individuals. The value of FD-300 is a compliment to FAZ metrics and have been used to detect early signs of diabetic retinopathy in previous studies[17].
Radial peripapillary capillary measurement
A 4.5*4.5mm rectangle scan centered on ONH was obtained for each eye with AngioVue OCTA system using Angio-Disc mode. The software automatically fits a circle with a diameter of 2.0 mm, centered on ONH, and defines a circle 2.0 mm wide that extends from the optic disc as the peripapillary region. The peripapillary region was divided into eight regions automatically based on Garway-Heath method[18], designated as nasal superior (NS), nasal inferior (NI), inferior nasal (IN), inferior temporal (IT). Temporal inferior (TI), temporal superior (TS), superior temporal (ST) and superior nasal (SN). Vessel densities of the whole image, inside disc and each sector of peripapillary area were generated by the software automatically.
We divided our patients into two sub groups according to the location (superior or inferior) of the affected vein and compared vessel density in each sector of the superficial and deep vascular plexus and RPC in each group.
All scans were reviewed by two experts for correctness of automated layer segmentation, as well as for FAZ delineation. In case of segmentation errors, manual correction was performed by the examiners until agreement achieved.
Statistical analysis
Statistical analysis was performed using a commercially available statistical software program (SPSS for Mac, version 25.0; IBM/SPSS, Chicago, IL, USA). Continuous variables are presented as mean and SD. Paired t test was used to compare the demographics and evaluate the difference in macular metrics, FAZ parameters and peripapillary vessel densities between the eyes with BRVO and the clinically unaffected fellow eyes. Unpaired t test was used to evaluate difference between clinically unaffected contralateral eyes and normal control eyes. A two-tailed P value of < 0.05 was considered statistically significant.