Abstract
Purpose :
To determine the benefit of switching treatment-resistant DME patients from anti-VEGF to longer acting steroid implants.
Methods :
This is a retrospective observational clinical study of patients who switched therapy from anti-VEGF to the dexamethasone implant for DME between 2014 and 2019. We collected the data of patients who had received the dexamethasone implant after previous anti-VEGF therapy with ranibizumab or bevacizumab. Primary outcomes were best corrected visual acuity (BCVA) and central macular thickness (CMT). Secondary outcomes were intraocular pressures (IOP) and adverse events. Data was collected pre anti-VEGF, at the point of switching, and at 6 and 12 months post dexamethasone implant injection.
Results :
We included 15 eyes of 13 patients. There was a significant anatomical response, with mean CMT improving from 480micrometers pre-switch to 459micrometers post-switch. This improved further to 419micrometers at 6 months, but relapse occurred by 12 months post-switch with CMT increasing to 486micrometers again. Functional responses were less dramatic – mean logMAR BCVA improved from 0.50 pre-switch to 0.45 post-switch, but stabilised at 0.53 at 6 months and 0.48 at 12 months. No patients had adverse events or surgery to reduce intraocular pressures. Patients received an average of 5.4 Anti-VEGF injections in the 6 months prior to switching and 1.2 corticosteroid injections in the 6 months post switching.
Conclusions :
Switching from anti-VEGF to corticosteroid resulted in anatomical, but not functional, improvements in these patients, with 50% of patients having a CMT of <400micrometers post-switch compared to 30% pre-switch. This reflects previous reports and suggests that, by the time a decision to switch therapy is made, permanent structural damage to the retina may already have occurred. Identifying patients who would respond to corticosteroid therapy earlier in their treatment may be of value and enable improved functional results to mirror the anatomical results seen in our study.
This is a 2020 ARVO Annual Meeting abstract.