LAUSR

the 24 weeks follow-up. A total of 165 eyes (52.2%) finished the 48 weeks follow-up. We are aware that at 48 weeks, there is significant loss of patients from follow-up, butwith 165 eyes, the number is still large. More importantly, between 24 and 48 weeks, there is stabilization of the median VA, the median CRT, the percentage of eyes with a gain of at least threeVA lines and the percentage of eyes with no residual ME (Epstein et al. 2012) indicating reliable results also at 48 weeks. At 24 weeks, 44.6% (141 eyes) gained at least three lines (mean of 2.7 injections), and at 48 weeks 48.5% (80 eyes, mean of 4.7 injections). Optical coherence tomography (OCT) examinationwas performed in 81%(256 eyes) at week 24. Of those, 46% (118 eyes) showed no residual ME with a CRT below 300 lm. Optical coherence tomography (OCT) examination was performed in 128 eyes at 48 weeks, of those, 45% (58 eyes) revealed no residual ME defined as a CRT of <300 lm. We found a significant impact of the number of injections applied on the improvement of BCVA at 12 and 24 weeks (ANOVA p = 0.007 and p = 0.048, respectively): In the first 12 weeks, two injections were needed to achieve a median gain of three lines, at 24 weeks, aminimumof three injections wasneededtoachieveamedianimprovement of three lines. At 48 weeks, five or less injections revealed a median gain of two BCVA lines, whereas 6 and more injections resulted in a median gain of three BCVA lines. However, the latter was not statistically significant (p > 0.25) which may be explained by the low number of eyes enclosed in each subgroup. Therefore, our 48 weeks data should be regarded carefully. Whether monthly injections during the first 3 months would have been associated with a gain of more than three VA lines cannot be answered with our data. However, Epstein and colleagues (2012) performed fixed bevacizumab injections every 6 weeks over 12 months and did not find a gain of more than three lines until week 12. Several randomized controlled trials (RCTs) and three meta-analyses with 5, 6 and 7 relevant RCTs included (Huang et al. 2013; Pielen et al. 2013; Braithwaite et al. 2014) investigated injection schemes with intravitreal anti-vascular endothelial growth factor (VEGF) agents that effectively reduce ME and increase BCVA in CRVO (level 1b and 1a evidence, respectively). Besides the level 1 treatment efficacy, the minimal number of injections needed to maximize and maintain the therapeutic effect has not been analysed in the literature, yet. Therefore, we found it highly important to look at this in a routine clinical setting.