Reply to ‘Comment on: Intravitreal aflibercept for macular oedema secondary to central retinal vein occlusion in patients with prior treatment with bevacizumab or ranibizmab’

Sir,

We would like to thank Călugăru and Călugăru¹ for their comments on our article.² We agree with them regarding the limitations of the study (retrospective nature, prior laser photocoagulation and/or steroid injections, different injection schemes), which were noted in our discussion.

They mention that a thickness of 279 μm signifies macular oedema requiring further treatment. We agree and for this reason the patients are continuing to receive treatment with intravitreal aflibercept injections. This study provides a real-life clinical experience with a switch to aflibercept in eyes with resistant or recalcitrant macular oedema secondary to central retinal vein occlusion.

In addition, they mention that an average of six injections (actually our study cohort had a median of seven) of bevacizumab or ranibizumab is insufficient to label a patient as a non-responder, claiming standard protocol used in studies. However, there is no consensus on the number of injections that designate a patient as a non-responder. Bhisitkul et al,³ using data from the CRUISE study, classified patients as early responders (vs late or non-responders) based on OCT thickness <250 μm 3 months after initiation of treatment. As our study reflects real-life practices, physicians used their own discretion to designate a patient as a poor on non-responder after no less than three injections as stated in the paper and in the example in Figure 4.

They also believe that the poor anatomic and visual benefits presented in our patients can be attributed to the low frequency of injections, the period of time without therapy (a median of 1.25 months before the initiation of treatment), and the period of time when the patients were insufficiently treated with bevacizumab/ranibizumab (12 months). We do not think that the delay in treatment by a median of 1.25 months adversely affected the outcome, as patients who were enroled in the CRUISE study⁴ had also a median of 2 months from diagnosis to screening. In our study, the mean interval of injections before the switch was 5.6 weeks and after the switch was 7.6 weeks, as the majority of the patients were on a treat and extend regimen. Indeed, the patients did not experience long-term improvement in vision after the switch to aflibercept despite improvement in macular thickness from 536 to 279 μm at the end of the follow-up. We believe that a contributing factor for the poor functional outcome may be the disease itself. According to Bhisitkul et al³ there is a subset of patients who experience reduced visual outcomes compared with early responders, and we think that the patients in our study may belong to this group. Similar trends in visual outcomes have been reported in patients with wet AMD who were similarly switched to aflibercept.^{5, 6}