Changes in Retinal Perfusion in Patients Treated with 12-month Scheduled Aflibercept Therapy for Diabetic Macular Edema

Abstract

Purpose : To investigate with wide-field angiography (WFA) the potential effects of aflibercept therapy on areas of Retinal Nonperfusion (RNP) in patients with diabetic macular edema (DME) undergoing 12-month scheduled aflibercept therapy.

Methods : Patients were included from a prospective trial of anti-VEGF therapy for patients with center involved DME. Eyes were included in this analysis if they had gradable WFA at baseline, 6 month and 1 year. A masked grader evaluated all images using a novel algorithm that standardized images so as to compare gradable areas (figure A) over time. The sizes of areas of RNP were assessed by ImageJ as a ratio of the total observable area, known as the ischemic index (RNPii, figure B). Mean changes in RNPii were compared between baseline, 6 months, and 12 months post-treatment. Significance was determined by a Student's paired t-test.

Results : A total of twelve eyes from nine patients were included in this analysis. Subject age ranged from 55-66 years, with a mean of 60 years. The mean number of injections was 5.2 (month 0-5) and 4.3 (months 6-11). The standardized best-corrected ETDRS vision mean score was 61.9 letters at baseline and 72.1 letters at month 12, with an mean increase of 10.2 (two lines of vision) letters. The mean RNPii measured 0.017 (SD=0.010) at baseline, 0.015 (SD=0.016) at 6 months, and 0.0115 (SD=0.010) at 12 months. At 6 months and 12 months the mean reduction in RNPii from baseline was 14.5% (P=0.45) and 33.5% (P=0.008) respectively.

Conclusions : Aflibercept therapy over the course of twelve months appears to reduce areas of RNP as measured with our grading algorithm. However, even with standardization of images to reduce image quality discrepancies over time, inconsistencies in images made grading RNP difficult and imprecise with WFA. Ocular Coherence Tomography Angiography (OCT-A) may prove to be a better method for assessing RNP.

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

 

Figure A –gradable area was standardized by connecting landmark vascular branch points that were clear in common for each image to correct for differences in image quality

 

Figure B – Standardized images at baseline, 6 months, and 12 months were graded for RNP using imageJ software by outlining areas of RNP and comparing them as functions of ratio of total gradable retinal area