Abstract
Purpose :
Low-contrast visual acuity (VA), which is important for driving, recognizing faces, and reading, is reduced in people with diabetes. We conducted an ancillary study within DRCR Retina Network Protocol V to compare changes in low-contrast VA 2 years after initiating treatment for diabetic macular edema in eyes with good vision.
Methods :
Adults with diabetes and one study eye with diabetic macular edema (confirmed on OCT) and good VA (Snellen equivalent 20/25 or better) were randomly assigned to initial treatment with aflibercept, laser, or observation. Aflibercept was added in the laser and observation groups if high-contrast VA met pre-specified worsening criteria. For this ancillary study, low-contrast VA was measured at 2.5% contrast in 387 of 702 randomized participants. The Hochberg procedure was used to control the type 1 error rate for multiple treatment group comparisons.
Results :
At baseline, median age was 60 years, 63% were male, and 70% were non-Hispanic white; mean (standard deviation) high-contrast VA letter score was 85.2 (3.6; Snellen equivalent 20/20) and mean low-contrast VA letter score was 47.6 (18.9; Snellen equivalent 20/125). The mean (standard deviation) change in low-contrast VA letter score from baseline at 2 years was 2.7 (20.1), -2.0 (19.6), and -3.1 (20.8) in the aflibercept, laser, and observation groups, respectively (mean difference [95% confidence interval]: aflibercept vs. laser=5.3 [-0.2 to 10.8], P=.06; aflibercept vs. observation=5.5 [-0.2 to 11.2], P=.06; laser vs. observation=0.2 [-4.6 to 5.0], P=.94); Between baseline and 2 years, low-contrast VA increased by 10 or more letters in 34% (32/93), 23% (27/117), and 24% (25/104) of eyes and worsened by 10 or more letters (among eyes with baseline low-contrast VA ≥10 letters) in 24% (21/87), 28% (30/107), and 32% (30/94) of eyes, with aflibercept, laser, and observation. The correlation between change in low- and high-contrast VA at 2 years was 0.47 (95% CI, 0.38-0.55).
Conclusions :
There were no significant differences between the aflibercept, laser, and observation groups with respect to change in low-contrast VA from baseline at 2 years; however, due to the width of the confidence intervals, clinically relevant differences between the aflibercept group and the laser and observation groups cannot be ruled out.
This is a 2020 ARVO Annual Meeting abstract.