Abstract
Purpose :
Perfused retinal vasculature has been nominally quantified from optical coherence tomography angiography (OCTA) scans using two-dimensional (2D) projection slabs. In both glaucoma patients and suspects, 2D perfused vessel density (PVD) of the circumpapillary and macular inner retina has been shown to be reduced. However, slab projections do not consider the volume of tissue that is perfused, which may be important for assessing the vascular supply to neural demand. For this experiment, we investigated perfused vascular volume (PVV) and perfused vascular volume density (PVVD) of the inner retina in healthy eyes of non-human primates.
Methods :
The macula and optic nerve head of 14 healthy rhesus eyes were imaged using 20 x 20° isotropic OCTA scans. 2D slab projection PVD, and 3D PVV and PVVD of the inner retina were determined for (1) a 3.5-mm diameter region centered on the fovea for macular scans and (2) a 500-µm annulus starting 400 µm from the Bruch’s membrane opening for circumpapillary region using a custom MATLAB program. Inter-eye symmetry and inter-visit repeatability (2.77 x intra-subject variability, scans obtained at least two weeks apart) were assessed for 7 animals.
Results :
Mean retinal nerve fiber layer thickness was 124 ± 9.3 µm. Mean ganglion cell inner plexiform layer thickness was 69.4 ± 4.9 µm, while mean ganglion cell complex thickness was 103.6 ± 5.5 µm. The mean PVD, PVV and PVVD for the (1) superficial vascular complex were 37.1 ± 4.6%, 6.2 ± 0.8 x 10-2mm3, and 6.3 ± 0.8%; (2) superficial vascular plexus were 32.1 ± 4.5%, 5.2 ± 0.8 x 10-2mm3, and 7.8 ± 1.1%, and; (3) circumpapillary nerve fiber layer were 55.7 ± 3.4%, 4.8 ± 0.3 x 10-2 mm3, and 10 ± 0.5%. The mean inter-eye asymmetry for PVD, PVV and PVVD were < 11.2% in the macular region, and <4.6% for the circumpapillary region using a custom MATLAB program. When expressed as a percentage of the mean, there was no difference in repeatability for 2D and 3D vascular measures (p = 0.98).
Conclusions :
Measures of OCTA vascular volume have similar inter-eye differences and repeatability compared to 2D slab projection density measures. Although OCTA volume measures are slightly greater than published histological vascular density, they are a closer representation of the vascular content than 2D density measures. Hence, these volumetric measures may provide a better method for assessing vascular supply to retinal tissue in ocular pathology.
This is a 2020 Imaging in the Eye Conference abstract.