Angiographic Patterns of Recurrent Choroidal Neovascularization in Age-related Macular Degeneration

Purpose: to evaluate the fluorescein and indocyanine green angiographic characteristics of recurrent choroidal neovascularization (R-CNV)in age-related macular degeneration (AMD). Method: a prospective investigation performed on 107 consecutive patients with exudative AMD and CNV not involving the foveal avascular zone was conducted. Laser treatment was ICGA-guided. In each patient, best corrected visual acuity (BCVA) on ETDRS charts was measured. FA and ICGA were performed in the same session the day before treatment and later repeated 3 weeks and 2, 3, 4, 6, 9, 12 months after photocoagulation. In eyes with questionable angiographic features for R-CNV at 3-week control, examinations were repeated each week till the pattern was defined. Result: at 1 year from photocoagulation, R-CNV occurred in 53 (49.5%) eyes, with classic features in 14 (26.4%) eyes, and occult in 39 (73.6%) on FA. At the 3-week follow-up visit, 49 (45.8%) eyes showed leakage beyond margins of laser scar on FA, whereas on ICGA hot spots appeared in 23 (21.5%) eyes, marginally in 21 cases and centrally in 2. One week later, in 5 of these eyes leakage persisted on FA and marginal hot spots on ICGA with subfoveal location were still detectable. Additional FA and ICGA performed at 5 weeks revealed lesion enlargement in all 5 eyes, leading to persistent CNV diagnosis. ICGA highlighted 3 patterns of R-CNV: Focal R-CNV (39 eyes, 73.6% of all R-CNV), a single dot-like hyperfluorescence along photocoagulated area margin, already detectable in early phases; Anular R-CNV (10 eyes, 18.8%), a hyperfluorescent, irregulary-shaped lesion, partially or completely encircling treated area, detectable in intermediate and late phases; Plaque R-CNV (4 eyes, 7.5%), a hyperfluorescent lesion greater than 1 disc diameter, expanding sectorially from the border of the treated area, and visible during intermediate and late phases. At the time of R-CNV onset, focal pattern had the BCVA among the three patterns whereas the final mean values are similar in the three groups. Conclusion: R-CNV resulted to be ill-defined in 73.6% of eyes, highlighting the role of ICGA in defining R-CNV site and extension. Only after disappearance of temporary hot spots on ICGA, that occurred at 4-week control, a precise distinction between them and real persistent CNV can be made. Three main patterns of R-CNV on ICGA were identified. Focal R-CNV has the best visual acuity at the onset time, probably due to smaller size, and anular R-CNV the worst, but at follow-up end mean BCVA is similar in the 3 patterns. Focal and plaque R-CNV lose an avarage of about 3 lines, whereas anular R-CNV retains its BCVA nearly unmodified.