Quantification of flow at baseline and follow-up in uveitic and non-age-related macular degeneration choroidal neovascular membranes (CNVM) using Optical Coherence Tomography-Angiography (OCT-A).

Abstract

Purpose : OCT-A is a recent advance in non-invasive imaging utilizing split-spectrum amplitude decorrelation to acquire high-resolution angiographic images. Recent software modifications permit semi-automated quantification of CNVM ‘flow’. We present the use of the AngioAnalytics OCT-A software (Optovue AngioVue) for the quantification of non-age-related macular degeneration (AMD) related-CNVM at baseline and follow-up. Absence of the perilesional hypo-intense halo has been reported as a sign of CNVM quiescence in neovascular AMD and we aimed to confirm this in our cohort.

Methods : Patients were identified prospectively from the retinal and uveitis service at a tertiary care centre. Only eyes with delineable CNVM on OCT-A were included. Clinical and demographic data, OCT (Topcon) central subfield thickness (CST) and OCT-A scans were collected in all cases. Patients received intravitreal anti-vascular endothelial growth factor therapy, as well as immunosuppression, where indicated. The repeatability of flow measurements between 2 observers were assessed using Bland-Altman analysis.

Results : 9 eyes of 9 patients (7 female, 2 male) were identified with an age range: 18-69 years. CNVM occurred secondary to posterior uveitis, myopia and central serous chorioretinopathy (CSCR) in 4, 4 and 1 patient, respectively. 8 CNVM demonstrated evidence of activity, of which follow-up OCT-A data were available for 5 eyes (3 posterior uveitis, 1 myopia, 1 CSCR). Compared with baseline, mean flow within the CNVM reduced significantly from 0.27mm² to 0.13mm² following initial treatment (p=0.037). There was an associated significant reduction in CST on OCT 338.4um vs. 239.6mm (p=0.044). However, comparing baseline CST with flow for all patients there was no significant correlation detected (r=0.182, p=0.703). The repeatability coefficient for inter-observer measurements was ±10.9% of mean baseline flow values. The perilesional hypo-intense halo was identified in 4 patients at baseline, and persisted or became more obvious with quiescence in all 4 cases.

Conclusions : OCT-A provides an additional quantitative parameter for monitoring CNVM treatment response. Inter-observer flow measurements demonstrated good repeatability. The perilesional halo did not consistently disappear with other features of CNVM quiescence in non-AMD CNVM.

This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.