Optical Coherence Tomography Angiography Features of Angioid Streaks

Eleonora Corbelli, Adriano Carnevali, A Marchese, MV Cicinelli, L Querques, R Sacconi, F Bandello, G Querques

Research output: Contribution to journalArticlepeer-review


PURPOSE: To analyze the optical coherence tomography angiography features of eyes affected with angioid streaks (AS) and to evaluate their ability to predict choroidal neovascularization (CNV) activity. METHODS: Angioid streaks were individuated from a pool of consecutive patients. Eyes with and without CNV were evaluated by multimodal imaging. RESULTS: Thirty-eight eyes of 19 consecutive patients diagnosed with AS were included. Thirty of 38 eyes with CNV and 8 of 38 eyes without CNV were included. In the majority of cases, CNV showed on optical coherence tomography angiography tangled appearance always associated with signs of neovascular inactivity on multimodal imaging (100%-0%, inactive-active, respectively). Choroidal neovascularization cases showing interlacing appearance were often associated with signs of neovascular activity on multimodal imaging (71.4%-28.6%, active-inactive, respectively). Optical coherence tomography angiography revealed a total of 27 AS, of which 20 appeared as a choriocapillary rarefaction, and in 7 AS, optical coherence tomography angiography choriocapillary segmentation revealed an irregular vascular network, possibly representing fibrovascular tissue over the crack-like breaks in Bruch membrane. CONCLUSION: Optical coherence tomography angiography is a noninvasive tool to detect the presence of CNV secondary to AS and to evaluate CNV activity. Optical coherence tomography angiography is able to add a novel element to the multimodal imaging characterization of AS.
Original languageEnglish
Pages (from-to)2128-2136
Number of pages9
JournalRetina (Philadelphia, Pa.)
Issue number11
Publication statusPublished - 2018


Dive into the research topics of 'Optical Coherence Tomography Angiography Features of Angioid Streaks'. Together they form a unique fingerprint.

Cite this