Purchase this article with an account.
Waseem Ansari, Aleksandra V Rachitskaya, Yasha Modi; Optical Coherence Tomography Angiography Quantitative Analysis and Correlation with Visual Acuity in Branch Retinal Artery Occlusion. Invest. Ophthalmol. Vis. Sci. 2017;58(8):6001.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
In fluorescein angiography (FA), decreased retinal perfusion in branch retinal artery occlusion (BRAO) cannot be quantified. FA shows non-perfusion acutely, but reperfused BRAOs fail to show the affected area. Optical coherence tomography angiography (OCTA) allows the calculation of vessel density and examination of the flow void. We established and analyzed an algorithm for quantifying non-perfusion and correlated the extent of non-perfusion to VA in BRAO.
Eight eyes from 4 patients with BRAO were imaged by FA, spectral domain OCT, and OCTA (AngioPlex, CIRRUS HD-OCT; Carl Zeiss Meditec, Inc, Dublin, CA). OCTA 3x3 mm images of superficial and deep capillary plexus centered on the fovea were obtained. To quantify the decreased perfusion, Image J (National Institutes of Health, Bethesda, Maryland, USA) was used to calculate absolute vascular density for each macular quadrant. The quadrant with the highest vessel density in each macula was set to 1, and relative vascular density was calculated for remaining quadrants. The algorithm was compared to qualitative examination of flow void by 3 independent readers. Clock hours of flow void were noted in each affected eye. VA was converted to logMAR. Correlations were calculated using paired t-tests.
Four eyes from 4 patients (3 females and 1 male; mean age 56 [33-73]) had BRAO confirmed by clinical exam and FA. OCTA was obtained at presentation (n=1), at 3 (n=2) months, and 6 (n=1) months after diagnosis. Repeat FA showed reperfusion in 3 eyes. Macular quadrants with decreased RVD correlated to areas read as qualitatively abnormal by readers (sensitivity 94%, specificity 94%) (Fig 1). RVD loss was greater in deep versus superficial vasculature in all affected quadrants (p=0.005). The amount of RVD in superficial or deep vasculature did not correlate with VA (p =0.44, p=0.38 respectively). The location of RVD loss did not correlate to VA. VA correlated to the number of affected clock hours (p=0.001).
A quantitative method using OCTA relative vascular density (RVD) identified areas of flow void and decreased perfusion in BRAO. RVD loss was seen in cases of re-perfused normal-appearing FAs indicating a particularly useful clinical application. Additionally, increased clock hours of flow void correlated to decreased VA. The current method might be expanded to other retinal vascular diseases.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.
This PDF is available to Subscribers Only