摘要
Background:To examine the baseline morphological characteristics and alterations in intraretinal microvascular abnormalities(IRMAs)in response to anti-vascular endothelial growth factor(anti-VEGF)treatment,documented by optical coherence tomography angiography(OCTA)in diabetic eyes.Methods:In this retrospective study,IRMAs were evaluated with multimodal imaging(fundus photography,fluorescein angiography,OCTA)in treatment-naïve diabetic eyes before and after anti-VEGF treatment for diabetic macular edema(DME)and/or proliferative diabetic retinopathy(PDR)and compared to diabetic control eyes with similar diabetic retinopathy(DR)severity that did not receive anti-VEGF therapy.The morphological characteristics of IRMAs on enface OCTA imaging were graded by masked readers at baseline,then after anti-VEGF therapy in treated eyes or after observation in control eyes.Characterization of interval changes in an IRMA were based on the following parameters:branching,vessel caliber and area of adjacent capillary non-perfusion.Results:The treated group included 45 IRMA foci from 15 eyes of 11 patients,while the control group included 27 IRMA foci from 15 eyes of 14 patients.Following anti-VEGF treatment,enface OCTA demonstrated that 14 foci of IRMA(31%)demonstrated regression with normalization of appearance of the capillary bed,20 IRMAs(44%)remained unchanged,six IRMAs(13%)progressed with enlargement or development of new IRMAs and five IRMAs(11%)demonstrated complete obliteration defined as IRMA disappearance with advancing capillary drop-out.In the control group,17 IRMA(63%)remained stable,8 IRMAs(29.6%)progressed and 2 experienced total obliteration(7.4%).The difference in rank order between the two groups was statistically significant(p=0.022).Conclusions:In eyes with DR status post anti-VEGF therapy,foci of IRMAs have a variable course demonstrating one of four possible outcomes:regression,stabilit,progression or complete obliteration.In contrast,none of the untreated control diabetic eyes demonstrated regression of IRM
