Angiographic regression patterns after intravitreal ranibizumab injections for neovascular age-related macular...

To investigate the relation between visual gain, injection frequency, and the angiographic regression patterns after intravitreal ranibizumab on an as-needed basis in exudative age-related macular degeneration (AMD).

Fifty-nine treatment-naïve patients (68 eyes) were retrospectively analyzed. All patients received three consecutive monthly injections (induction phase) of ranibizumab (0.5 mg/0.05 mL). Based on fluorescein angiography (FA), the choroidal neovascularization (CNV) was judged to present either complete regression (pattern 1), partial regression (pattern 2), stabilization of the lesion size without leakage (pattern 3), stabilization of the lesion size with persistence of leakage (pattern 4), or increased angiographic size (pattern 5).

Mean visual acuity (VA) significantly improved from 48 to 54.3 letters at 1 year after a mean of 5.5 injections (P < .001). Multiple linear regression revealed baseline VA as a predictor of visual gain and the angiographic pattern as a predictor of number of injections. Analysis of variance revealed a significant interaction (F-test [1.67] = 25, P < .001) between the number of injections at 12 months and the regression patterns, as evaluated by FA 1 month after the induction phase. Eyes showing complete CNV regression needed significantly fewer injections than eyes without any angiographic sign of CNV regression (3.4 injections in pattern 1 vs 5.6 injections in pattern 3 [P = .03], and 7 injections in pattern 4 [P < .001], 4.4 injections in pattern 2 vs 7 injections in pattern 4 [P < .001]).

FA may represent a useful tool to adapt the rhythm of visits and intravitreal anti-vascular endothelial growth factor injections in exudative AMD.