Purchase this article with an account.
Hadeel Sadek, Subhashini Chandrasekaran, Rich Amador, Peter Khouri, Bernard Szirth; Predictor Intake Form with OCTA Validation in Type I Diabetic Retinopathy Triage. Invest. Ophthalmol. Vis. Sci. 2018;59(9):2791. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
The prevalence of T1DM in the US is 1.25 million. The aim is to create a triage intake form that can identify and prioritize high risk individuals for the development of diabetic retinopathy (DR) based on their scores, who could benefit from earlier than yearly recommended eye screenings.
Thirty-eight subjects (14 male, 24 female) with T1DM were selected from a comprehensive non-mydriatic tele-retinal screening. Subjects ranged from 10 to 72 years old (avg=27y) with a duration of T1DM > or = 1.5 years (avg=14.5y). Subjects were split into two T1DM groups: those with (N=19) and those without (N=19) dot hemorrhages, flame hemorrhages, or intraretinal microvascular abnormalities (IRMAs) on color retinal imaging. A control group without diabetes (N=19) underwent an identical questionnaire (< 5 min to complete). Each individual received a score based on their responses, placing emphasis on risk factors for DR, such as last blood glucose level, HbA1C, BMI, blood pressure, diet, and duration of T1DM. Low risk: 14, pre-risk: 20, high risk: 24. The findings were validated with color imaging, fundus autofluorescence (FAF), and optical coherence tomography angiography (OCTA). One-way ANOVA and Tukey post-hoc tests were used to determine the threshold for each of the 3 groups.
Scoring was: 24 for subjects with DR, 20 for subjects with diabetes and no DR, and 14 for subjects without diabetes. One-way ANOVA showed a significant difference between average scores of the 3 groups (p= 2.94 x 10-7). Tukey post-hoc tests showed significant difference between scores of the subjects with diabetes and no DR and subjects without diabetes (p=0.0007) and significant difference between the subjects with DR and subjects without diabetes (p= 2 x10-6). There was no significant difference between subjects with DR and subjects with diabetes and no DR (p= 0.056) when looking at only color retinal images.
Use of triage form may be beneficial at screenings to prioritize individuals with higher risk for DR (score >22) who need to be screened first. It may also be beneficial to track an individual’s risk change prior to follow-up visits. Further investigation must be performed on a larger population and utilizing a more sensitive imaging modality, such as OCTA, that can help confirm earlier changes.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.
This PDF is available to Subscribers Only