Evaluating persistent and recurrent choroidal neovascularization

OBJECTIVE: To evaluate the clinical utility of routinely using indocyanine green angiography (ICGA) with fluorescein angiography (FA) in detecting persistent or recurrent choroidal neovascularization (CNV). DESIGN: Prospective, consecutive case series. PARTICIPANTS: Twenty-four eyes of 21 patients with exudative age-related macular degeneration (AMD) that had conventional laser treatment for CNV were examined. INTERVENTION: Fluorescein angiography and ICGA were performed together on all eligible eyes at the first post-treatment visit and all subsequent follow-up visits in which persistent or recurrent CNV was suspected clinically. MAIN OUTCOME MEASURES: Choroidal neovascularization detection and delineation by each angiographic technique were measured. RESULTS: Of the 54 FA-ICGA study pairs performed over a 20-month investigation period, FA showed well-defined, ill-defined, and no CNV in 10 (19%), 19 (35%), and 25 (46%) eyes, respectively. Indocyanine green angiography had a high concordance rate at 70% and 88% when persistent or recurrent CNV was well-defined and absent, respectively, on FA and rarely added additional, clinically useful information in these settings, particularly in the former presentation. Of the 29 eyes that showed some evidence of CNV by FA, the neovascular complex was ill-defined in 19 (66%) eyes. When CNV was ill-defined by FA, the corresponding ICGA showed well-defined CNV in 9 (47%) of 19 eyes, in 5 of which the CNV was nonsubfoveal in location. CONCLUSIONS: Persistent or recurrent CNV in AMD was frequently ill-defined by FA. Indocyanine green angiography was a valuable adjunct to FA by better delineating CNV in this specific setting. However, ICGA was not useful when the post-treatment FA showed well-defined CNV. Furthermore, ICGA was not helpful when the first post-treatment FA was negative and there were no ophthalmoscopic signs of exudation.