Purchase this article with an account.
Sally Baxter, Aria Ashir, Brian J. Nguyen, Eric Nudleman; Quantification of retinal nonperfusion associated with neovascularization in diabetic retinopathy using ultrawide field fluorescein angiography. Invest. Ophthalmol. Vis. Sci. 2017;58(8):93.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Retinal ischemia precedes neovascularization (NV) in proliferative diabetic retinopathy (PDR), but the extent of nonperfusion associated with NV is not known. We analyzed ultrawide field fluorescein angiograms (UWFA) of diabetic patients to test the hypothesis that nonperfused areas associated with NV are larger than those without adjacent NV, and that a threshold size of nonperfusion could be established for eyes at risk of NV.
This prospective case series included UWFAs from 18 eyes of 11 adult patients with PDR. Eyes with prior intraocular treatment, other retinal pathology, or media opacity were excluded. Nonperfused areas were manually traced. Sizes were calculated using prototype software and normalized to total image size. Size and location of nonperfused areas associated with NV were compared to nonperfused areas without associated NV using unpaired t-tests. NV was considered associated with nonperfusion if it was within 1 disc diameter of the border of nonperfusion. Threshold size was defined as the lower limit of the 95% confidence interval of the mean size of retinal nonperfusion associated with NV. The number and location of associated buds of NV were compared between nonperfused areas larger and smaller than threshold size using unpaired t-tests.
The patients’ average age was 45.7 years and were equally split between males and females. Nonperfused areas associated with NV were significantly larger than those without associated NV (32.0% ± 5.24% vs. 3.3% ± 0.92% of the total retinal area, p<0.001) and were preferentially located in the posterior (<10 mm from the disc) and mid-peripheral retina (10-15 mm from the disc). The threshold size for nonperfusion associated with NV was approximately 23% of the total retinal image. Areas larger than threshold were more likely to have a greater number of associated NV buds (9.64 ± 2.16 vs. 0.86 ± 0.29, p<0.001) and have associated NV located closer to the optic nerve (7.53 ± 0.27 mm vs. 9.24 ± 0.64 mm, p = 0.014).
Nonperfused areas associated with NV in PDR tend to be larger and more posteriorly located than areas without associated NV. A threshold size of nonperfusion of >23% of the total retinal image was associated with NV. This may help identify high-risk eyes and allow increased monitoring to prevent the development of NV in patients with diabetic retinopathy.
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