Brittni Scruggs, MD, PhD
Vitreoretinal Fellow, Casey Eye
Portland, OR
Vitreous hemorrhage (VH) occurs in approximately 45% of all patients with proliferative diabetic retinopathy (PDR) despite prior anti-vascular endothelial growth factor (VEGF) or pan-retinal photocoagulation (PRP) treatment(s). Yet, it is often difficult to know when or if a patient with a VH should undergo pars plana vitrectomy (PPV) versus trialing anti-VEGF intravitreal injections in clinic. The DRCR Protocol AB randomized clinical trial, which included 205 diabetic patients across 39 study sites, was designed to compare these two initial treatment options. PDR patients with VH who received monthly 2mg aflibercept intravitreal injections were compared to similar patients who had PPV with PRP within two weeks of randomization.
In his May 6th ARVO talk, Dr. Antoszyk describes the outcomes of this trial. When comparing the two cohorts, there was no significant difference in the visual acuity (VA) at 24 weeks post-treatment initiation (20/63 for both groups), which was the study’s primary outcome. No VA differences were found two years post-treatment (20/40 – 20/50). However, Dr. Antoszyk and his team demonstrated that the PPV with PRP group had better vision compared to the aflibercept group at four weeks (20/63 vs. 20/100, respectively). Secondary outcomes included 1) the need for PPV in the aflibercept cohort after a minimum of four injections and 2) the need for aflibercept in the surgical cohort if recurrent VH occurred; approximately a third of patients in each cohort ultimately needed such treatment. The average number of intravitreal injections performed per participant over a 2-year period was 8.9 in the aflibercept group and 2.3 in the PPV with PRP group. Lastly, analysis of ocular safety outcomes demonstrated that there were more tractional retinal detachments in the aflibercept group (22% vs. 13%).
This talk highlights the two main treatments for PDR-related VH and provides data supporting the safety and long-term efficacy of both options. Patients should be informed that vision may improve more quickly with initial vitrectomy, but long-term vision is comparable regardless of treatment option chosen. Number of injections required per option, retinal detachment risk, and treatment-specific adverse events (e.g., endophthalmitis rates, surgical complications, cataract formation, etc.) should be discussed with each patient for improved informed consent.
