Longitudinal Angiographic Evidence That Intraretinal Microvascular Abnormalities Can Evolve into Neovascularization

Jonathan F Russell,Yingying Shi,Nathan L Scott,Giovanni Gregori,Philip J Rosenfeld
DOI: https://doi.org/10.1016/j.oret.2020.06.010
Abstract:Purpose: The hallmark of proliferative diabetic retinopathy (PDR) is retinal neovascularization. Tortuous intraretinal vascular segments known as intraretinal microvascular abnormalities (IRMAs) are a known risk factor for neovascularization (NV), but whether IRMA represents a biomarker or a vascular precursor lesion to NV has not been demonstrated. The purpose of this study was to determine whether IRMA may evolve directly into NV. Design: Retrospective analysis of prospective, observational case series. Participants: Patients with treatment-naïve PDR. Methods: Patients were imaged longitudinally with fluorescein angiography (FA) and swept-source (SS) OCT angiography (OCTA) before and after panretinal photocoagulation (PRP). Main outcome measures: Presence and colocalization of IRMA and NV on serial FA and SS OCTA. Results: Two PDR patients showed multiple NV and IRMA lesions at baseline examination. Three months after PRP, FA demonstrated profuse leakage from 3 new NV lesions in one patient and 1 new NV lesion in another patient. Multimodal imaging showed that these 4 lesions were IRMAs at baseline. Swept-source OCTA performed before PRP and 1 week, 1 month, and 3 months after PRP confirmed that the precursor IRMA lesions were intraretinal tortuous vascular lesions at baseline and that they developed into preretinal NV with contiguous intraretinal components. NV was found to develop and adhere to the posterior hyaloid even in areas of pre-existing hyaloidal detachment. Conclusions: Diabetic retinal NV can develop from IRMA. Early identification of IRMAs may be an accurate means of predicting progression to PDR, and frequent monitoring of IRMAs with SS OCTA may facilitate early diagnosis of PDR.
What problem does this paper attempt to address?