Abstract
Purpose :
Optical Coherence Tomography Angiography (OCTA) is an effective modality to obtain three-dimensional angiograms of the retinal vasculature. However, current software limits analysis to the superficial and deep capillary plexuses (SCP and DCP) and does not account for the middle capillary plexus (MCP) as a separate entity. We performed a prospective imaging study of both healthy controls and diseased eyes to assess the feasibility of imaging the MCP apart from the SCP and DCP.
Methods :
Ten healthy eyes and twelve eyes from nine diabetic patients were imaged using OCTA (RTVue XR, Optovue, Inc with SSADA software). To visualize the three networks, we obtained en face OCT angiograms using two different segmentation approaches. The first approach used pre-set segmentation for the SCP and DCP, while the MCP was set by placing a thin slab at its anatomic location of the inner border of the inner nuclear layer (INL). The second approach used customized segmentation to set the boundaries for each capillary plexus according to its known anatomical location, setting the SCP to capture the nerve fiber and ganglion cell layers, and the MCP and DCP at the inner border and outer borders of the INL respectively.
Results :
Both approaches to segment en face OCT angiograms into three capillary plexuses showed significant limitations, with difficulty in discerning three separate networks, and shadow artifacts were observed as the superficial vessels cast shadows on deeper capillary layers. Figure A shows results for approach 1 and Figure B shows results for approach 2.
Conclusions :
Current segmentation software of OCTA does not allow for the ability to distinguish the MCP from the SCP and DCP. Our attempts to manually segment the capillary plexus layers according to their anatomical locations show the limitations in isolating the MCP with current methods. Further work and software modifications in the future will enable improved ability to visualize the three separate capillary plexuses, including removal of shadow artifacts from the inner vascular networks that confound the deeper networks.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.