To evaluate a recently established grading protocol for diabetic macular edema (DME) over the course of intravitreal anti-VEGF treatment with ranibizumab.Fluorescein angiography images and optical coherence tomography scans before treatment and after 3 monthly applied intravitreal ranibizumab injections were retrospectively graded for each included study eye according to the recently introduced "SAVE" grading protocol ("S"= subretinal fluid; "A"= "area of retinal thickening"; "V"="vitreo-retinal abnormalities"; "E"="etiology of leakage focal versus non-focal") and correlated with best-corrected visual acuity (BCVA) in letters (lett).Five of the 39 included study eyes had subretinal fluid ("S") before treatment which resolved during treatment. BCVA of study eyes with an initial retinal thickening smaller than one disc diameter ("A") was non-significantly higher compared to patients with a retinal thickening greater than one disc diameter (34.0 ± 17.9 lett versus 25.3 ± 13.3 lett, p=0.236) but became significant during treatment (40.5 ± 10.0 lett versus 28.3 ± 13.1 lett, p=0.004). No difference in BCVA was observed between patients with or without vitreo-retinal abnormalities ("V") before and during therapy. BCVA in patients with focal leakage ("E") was significantly higher than in patients with non-focal leakage before (33.1 ± 12.3 lett versus 23.3 ± 13.3 lett, p=0.017) and during (38.9 ± 10.9 lett versus 26.3 ± 12.6 lett, p=0.002) therapy.Applying the grading protocol "SAVE", focal leakage ("E") was the only retrospectively observed parameter which significantly correlated with a better BCVA before therapy and over the course of treatment in patients with fovea-involving DME.
«