Abstract
Purpose:
To investigate the association between the level of retinopathy and implicit time (IT), amplitude (Amp), and a combined score (Score) of the 30Hz flicker ERG, measured using a novel hand-held ERG device, in a cohort of adult patients with type 1 and type 2 diabetes mellitus.
Methods:
Subjects were recruited from a pool of patients refered by their primary care provider to Eastmont Wellness Center, Oakland, CA for an eye examination. All subjects underwent a comprehensive ocular examination. Photopic 30Hz flicker ERG measurements were obtained from each eye using the RETeval hand-held ERG device (LKC Technologies, Inc.). Fundus photographs were obtained and evaluated using the Canon CR-DGI fundus camera, following the EyePACS protocol and grading system, a modified ETDRS protocol. Results were analyzed based on the subject’s eye with the worse level of retinopathy. If both eyes of the subject had the same level of retinopathy, the right eye was selected.
Results:
58 ethnically diverse patients were enrolled. Mean age was 53±11 years. 67% were females. Mean duration of diabetes was 9.2±6.6 years and mean hemoglobin A1c was 8.1±2.1. The analysis herein is based on the 50 patients who had not had PRP and had both RETeval device results and gradable photos. 25 (50%) patients showed no retinopathy, 7 (14%) had mild non-proliferative diabetic retinopathy (NPDR), 14 (28%) had moderate NPDR, 4 (8%) had severe NPDR, and 0% had proliferative retinopathy (PDR). 1 (2%) patient had clinically significant macular edema (CSME). Patients were assigned to referable (severe NPDR and/or CSME) and non-referable groups. Retinopathy had significant effect on Score, IT, and Amp (p=0.03, one-way MANOVA). Only IT, was significantly associated with referable retinopathy in our logistic regression analysis (p =0.02). A delay in one millisecond of IT was associated with the odds ratio of referable retinopathy diagnosis equal to 1.96, 95% CI: 1.18 - 4.17. In this preliminary analysis the RETeval device score had a sensitivity of 100% (4/4) and a specificity of 69.5% (32/46) to referable retinopathy.
Conclusions:
A delay in photopic 30Hz flicker IT, measured using a hand-held ERG device was associated with significant increase in the odds of being diagnosed with severe NPDR or CSME. The RETeval device score had perfect sensitivity, albeit more cases are required for a conclusive analysis.