Longitudinal Analysis of Cone Structure in Patients with Central Retinal Vein Occlusion (CRVO) and Cystoid Macular Edema (CME)

Abstract

Purpose : To study cone photoreceptors over 12 months in patients with CRVO and CME in the Study of COmparative Treatments for REtinal Vein Occlusion 2 (SCORE2) using adaptive optics scanning laser ophthalmoscopy (AOSLO) imaging.

Methods : Subjects enrolled in SCORE2 were evaluated at 1 of 5 sites. After baseline imaging, subjects received monthly intravitreal anti-VEGF injections for 6 months, and were then randomized to either continued monthly or treat-and-extend (TAE) treatment arms. AOSLO images were obtained twice at baseline within 1 month after the first injection, and at 6 and 12 months. At least 10 regions of interest (ROIs) were selected by a single investigator from the baseline images where at least 50 cones could be counted reliably. The same ROIs were analyzed by 1 independent grader from each study site who evaluated cone spacing using a density recovery profile program after detailed instruction. An independent investigator reviewed all cone selections from all graders and judged them as acceptable or unacceptable. Statistical analysis was performed using a mixed linear regression analysis of inter-grader and inter-visit variability.

Results : Of 8 subjects evaluated at baseline, 3 had at least 10 ROIs in both baseline AOSLO montages; 2 were imaged successfully at 6 and 12 months, while no quantifiable images were obtained in the third patient at 6 or 12 months due to poor patient cooperation. Both subjects with follow-up images were randomized into the TAE group and received 11 and 10 injections, respectively; subjects had 11 and 14 ROIs, respectively, with visible cones at all 4 study visits. Of 500 ROIs analyzed by 5 graders, cone markings were deemed acceptable at 378 and unacceptable at 122. Analyzing only the acceptable measures, graders differed significantly in cone spacing measures (interclass correlation coefficient = 0.27, 95% confidence interval = 0.20 – 0.34). Controlling for grader, there was no evidence of change in cone spacing over time.

Conclusions : Cone spacing measures in eyes with CRVO-associated CME were associated with significant inter-grader variability, but controlling for grader, we found no evidence of change in cone spacing over time. The study demonstrates challenges associated with cone spacing measures in eyes with CRVO-associated CME performed by independent graders in a multicenter trial.

This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.