Quantifying vascular density and morphology using different swept-source optical coherence tomography angiographic scan patterns in diabetic retinopathy

Abstract

AIMS: To evaluate quantitative metrics of the retinal microvasculature in eyes with diabetic retinopathy (DR) using various en face swept-source optical coherence tomography angiography (SS-OCTA) image sizes. METHODS: Non-segmented and segmented images were acquired using an SS-OCTA device (PLEX Elite 9000; Carl Zeiss Meditec, Dublin, California, USA). The scanning protocols included the 3×3 mm, 6×6 mm and 12×12 mm fields of view. Quantitative analysis of the perfusion density (PD), vessel length density (VLD) and fractal dimension (FD) was performed. The area under the receiver operating characteristic curve was estimated to assess the ability of each image size to predict DR. RESULTS: This prospective, cross-sectional study included 60 eyes, (non-DR, 13 eyes; non-proliferative DR (NPDR), 24 eyes; proliferative DR (PDR), 23 eyes) of 46 patients with diabetes and 21 eyes of 16 healthy individuals. In the 12×12 mm images, the PD of healthy individuals was significantly greater than that of patients with NPDR or PDR for all layers (NPDR, p<0.05; PDR, p<0.001 0.001; FD, p<0.001) or PDR (VLD, p<0.001; FD, p<0.001 for all layers). The results were similar for the 3×3 mm and 6×6 mm images. Overall, PD, VLD and FD progressively decreased with worsening DR severity in segmented and non-segmented layers for all SS-OCTA scan sizes. For detecting DR, 3×3 mm images best predicted DR for all evaluated quantitative parameters. CONCLUSIONS: Vascular changes in DR can be monitored in detail through quantitative evaluations that combine different SS-OCTA scan sizes and parameters.