Abstract
Purpose :
Sea fan neovascularization (SFNV) is a well recognized complication of sickle cell retinopathy as well as other vascular insults. Despite the broad clinical awareness of this striking entity, there is no unifying mechanism to explain its unique morphology.
Methods :
We hypothesized that vascular endothelial growth factor (VEGF) patterns of diffusion can sufficiently explain SFNV, and we developed a computational model with Python 3.10 using a modified Diffusion Limited Aggregation (DLA) scheme, a fractal generative process based on particles undergoing random walks clustering to form networks. We varied the topography of the originating source of diffusible particles and compared the results to fluorescein angiography images from eyes with and without SFNV.
Results :
One patient with proliferative sickle cell retinopathy (SCR) and two patients with proliferative diabetic retinopathy (PDR) were included. One patient with PDR had a large, confluent area of ischemia and the other had multifocal areas of ischemia. The model demonstrated that structures similar in appearance to NV can be simulated using DLA, and SFNV specifically may result from a unidirectional source of diffusible VEGF, which is more likely to occur with confluent areas of nonperfusion. This mechanism would be independent of the underlying disease.
Conclusions :
The pattern of retinal neovascularization may depend on the topography of retinal ischemia.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.