Predictors of Dexamethasone Response of Residual Edema by Branch Retinal Vein Occlusion after Bevacizumab Injection

ABSTRACT

Purpose@#To predict the response to intravitreal dexamethasone (IVD) implant injection in cases where macular edema (ME) caused by branch retinal vein occlusion persists despite intravitreal bevacizumab injection. @*Methods@#Three consecutive (monthly) bevacizumab injections were given to treat ME caused by branched retinal vein occlusion (BRVO) 63 eyes that received additional dexamethasone or bevacizumab injection 1 month later to treat residual ME were retrospectively studied. Each injection group was divided into two subgroups according to ME disappearance status by 6 months after diagnosis. Initial central retinal thickness (CRT), subfoveal choroidal thickness (SFCT), and hyperreflective focus status were compared among the subgroups, as were the changes in these values. @*Results@#At the decision point, the dexamethasone good response subgroups exhibited thicker CRTs and smaller CRT changes than the dexamethasone partial response subgroups (all p < 0.05). The good dexamethasone response subgroup showed smaller SFCT changes, a thicker CRT, and smaller CRT changes than the bevacizumab good response subgroup (all p < 0.05) at the decision point. The cutoff values of the Youden index were 409 μm for the CRT and 62.5 μm for the CRT change (p = 0.002 and p = 0.011, respectively). @*Conclusions@#If ME persists after three bevacizumab injections, IVD more effectively reduces edema if the CRT is thick or if the CRT change is small.