Wide Field Swept Source Optical Coherence Tomography Angiography for the Evaluation of Proliferative Diabetic Retinopathy and Associated Lesions: A Review

Update on Optical Coherence Tomography and Optical Coherence Tomography Angiography Imaging in Proliferative Diabetic Retinopathy

“…In 2020, OCT technology had already achieved a level of proficiency capable of accurately detecting and characterizing NVCs, NPAs and disease activity with recognized added value of WF and OCTA in disease staging and monitoring [6]. Namely, with OCT, it is possible to visualize the retinal layers in detail and to distinguish between NVC and IRMAs, and there is less obscuration than FA when there is associated fibrosis and hemorrhaging [6,41,42,79,83,103,104]. In both NVE and NVD, the OCTA flow signals found within the hyperreflective material indicate disease activity and correlate with the OCTA en face image [6,67,72,79].…”

“…WF-OCTA can better differentiate between IRMA, diabetic papillopathy and neovascularization than FA, due to the three-dimensional information [83,84,86,97,104,105]. UWF-OCTA can also play a role in the anti-VEGF treatment of PDR by providing more supportive information for management decisions at each visit, similar to when treating DME [6,91,99,103]. Specifically, WF-OCTA may be helpful in targeted or conventional panretinal photocoagulation, as previous studies already demonstrated the non-inferiority of extended targeted PRP to conventional PRP in PDR [89][90][91]99,106].…”

“…Nevertheless, it should be noted that for a proper assessment, correct segmentation is necessary, and often there is improvement in NVC recognition with manual segmentation [79,81,86,93,103,104]. In the future, the accuracy of NVC detection in Wu et al [93], in response to the segmentation limitation challenge reported by previous studies, developed an OCTA optimization method which consists of an improved vascular connectivity analysis algorithm combined with a morphological characterization and elimination of noise and artifacts.…”

“…Nevertheless, it should be noted that for a proper assessment, correct segmentation is necessary, and often there is improvement in NVC recognition with manual segmentation [79,81,86,93,103,104]. In the future, the accuracy of NVC detection in vitreoretinal slabs might increase with improvement of auto segmentation, deep learning and artificial intelligence [81,86,[107][108][109][110][111].…”

“…Another important issue that remains from all these different studies [41,67,80,82,86,92,97,99,106] with different imaging protocols is the optimal acquisition protocol so as to have a balance between the scanned area and lesion detection. It seems the protocol with best detection rate was the 12 × 12 mm angiography images centered on the fovea and optic disc, being an interesting non-inferior alternative to 15 × 9 mm montage images [92,103,104]. However, some software devices also allow montages, which also increases the scanned area [6,42,103].…”