Understanding indocyanine green angiography in polypoidal choroidal vasculopathy: the group experience with digital fundus photography and confocal scanning laser ophthalmoscopy.

PURPOSE: To evaluate the angiographic features in using fundus camera-based versus confocal scanning laser ophthalmoscope (cSLO)-based indocyanine green angiography in differentiating polypoidal choroidal vasculopathy (PCV) from typical age-related macular degeneration.

METHODS: Sixty-five eyes of 44 patients with exudative maculopathy due to PCV or typical age-related macular degeneration were prospectively imaged with indocyanine green angiography using fundus camera and cSLO. Images were graded independently by retinal specialists. The main outcome measure was agreement between cSLO and fundus camera for the diagnosis of PCV. The rate of detection and area under the receiver operating characteristic curve of 7 preselected individual features were also compared.

RESULTS: The diagnosis of PCV was made with the cSLO system in 36 eyes (55.4%) and typical age-related macular degeneration in 29 eyes (44.6%), whereas the fundus camera diagnosed PCV in 39 eyes (60.0%) and typical age-related macular degeneration in 26 eyes (40.0%). There was moderate agreement between the two indocyanine green angiography systems (Kappa = 0.53). Using cSLO as the gold standard, fundus camera has a sensitivity and specificity of 83.3% and 69.0%, respectively. Typical nodular appearance was the most commonly detected feature (median, 88.9% for cSLO, 80.6% for fundus camera, P = 0.63) and had the highest area under the curve for the diagnosis of PCV in both systems (median, 80.2% for cSLO, 73.2% for fundus camera, P = 0.13). Confocal scanning laser ophthalmoscope was more sensitive in detecting branching vascular network and late hyperfluorescent plaque.

CONCLUSION: Both systems detected >80% of PCV based on typical nodular appearance of polyps. However, the cSLO is superior in detecting additional features, particularly branching vascular network.