Abstract
Purpose :
Predominantly peripheral disease in eyes with diabetic retinopathy (DR) has been suggested to be a potential strong risk factor for progression to proliferative disease. However, the reliability and the optimal methodology for the assessment of lesion distribution are still uncertain. In this study we compared a subjective assessment to precise quantification of lesion burden in ultrawide field (UWF) images of eyes with DR.
Methods :
In this IRB-approved study, UWF images were obtained from a cohort of patients recruited for DR screening at the Narayana Nethralaya Eye Institute. DR lesions (Microaneurysms, hemorrhages, lipid exudates, and cotton wool spots) were manually segmented on the UWF images by certified graders at the Doheny Image Reading Center using GRADOR software. An overlay defining an area corresponding to the 7-standared field Early Treatment Diabetic Retinopathy Study (ETDRS) was superimposed on each UWF image. Eyes were subjectively classified as having predominantly central lesions (PCL) if DR lesion burden appeared to be greater inside the 7-ETDRS fields, and predominantly peripheral lesions (PPL) if lesion burden appeared greater outside the 7-ETDRS fields. Quantification of lesion frequency within and outside the 7-ETDRS fields defined eyes as PCL and PPL, respectively. The qualitative and quantitative assessments were then compared to assess the level of agreement.
Results :
A total of 36 eyes from 25 cases with DR were included in this study. Of these 36 eyes, on subjective qualitative evaluation by trained graders, 21 were classified as PCL and 14 were classified as PPL. Upon quantification of individual lesions, all 36 eyes were classified as PCL. Mean total number of DR lesions within and outside the 7-ETDRS fields was 53 ± 44 and 14 ± 19, respectively. There was a significant difference between subjective and quantitative identification of DR lesion distribution (p = 0.005).
Conclusions :
In this cohort, eyes that were qualitatively rated to have predominantly peripheral disease were actually found to have predominantly central disease when evaluated in precise quantitative fashion. Given the apparent importance of peripheral lesions in DR progression, this study underscores a need to standardize the methodology for rating the severity of peripheral disease. An objective quantitative approach may be preferred and may translate to a more precise scoring system.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.