Five year outcomes of the Bevordex Study (a multicenter randomized clinical trial of intravitreal bevacizumab versus intravitreal dexamethasone).

Abstract

Purpose : There is a paucity of long term data in the treatment of centre-involving diabetic macular edema (DME). Diabetic retinopathy (DR) is a leading cause of acquired blindness among young adults in developed countries. Although severe vision loss may occur due to proliferative DR, the most common cause of visual loss in these patients is DME. This study aims describe the 5 year outcomes from the BEVORDEX study, a randomized controlled trial study comparing treatment of DME with dexamethasone implant (dex-implant) with bevacizumab.

Methods : In the BEVORDEX study, patients were initially assigned randomly to dex-implant or bevacizumab. After 2 years, patients were released from the clinical trial protocol and treated as per their treating physician. Data to 5 years from entry into the BEVORDEX study were collected including visual acuity (VA), central macular thickness (CMT), diabetic retinopathy status and treatments. An intension-to-treat analysis was conducted.

Results : From the 68 eyes that completed the 2 year Bevordex study, 5 year outcomes from 49 eyes (72%) (27 bevacizumab and 22 dex-implant) were available. The overall mean number of visits was 14.7 (CI= 11.2–16.2) with no significant difference between the 2 groups, 15 visits (CI= 11.3–18.6) for bevacizumab versus 12 visits (CI= 8.5–15.9) for the dex-implant treatment group.
The mean VA at 5 years was 59 letters [CI= 53–64] which was not significantly different from mean VA at end of trial of 65 letters [CI= 60–69] (p=0.09).The proportion of eyes that achieved 20/40 or better vision was not significantly different from the end of trial (40%) to at 5 years (37.5%) (p= 0.7).
The CMT at 5 years was 331.4 μm [CI= 294–368] which was not significantly different from end of trial CMT of 334.0 μm [295 - 372] (p= 0.9).
The overall mean number of treatments was 3.4 (CI= 2.4–4.5) with no significant difference between the orginal BEVORDEX treatment groups; bevacizumab 4 [CI= 2.2–5.8] and dex-implant 3 [CI= 1.8–4.1] (p= 0.3).

Conclusions : Mean vision, the proportion of eyes that achieved 20/40 or better vision and macula thickness at the exit of the trial was maintained at 5 years. The overall treatment burden was also found to remained low at 5 years.

This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.