Abstract
Purpose :
To examine microvascular changes of patients with diabetes and diabetic retinopathy using Optical Coherence Tomography based Microangiography (OMAG).
Methods :
A total 68 patients (136 eyes) with a history of diabetes underwent OMAG imaging, 67 kHz Cirrus HD-OCT prototype system (Carl Zeiss Meditec, Dublin, CA) at University of Washington, Seattle. The demographics and clinical data including past medical and ocular history were collected. All patients were divided into 5 separate groups based on the severity of their diabetic retinopathy (DR). Flux index was defined as the average OMAG value within an independent subsampled area of the retina, Perfusion index was defined as the percent coverage of the area by retinal vessels with flow. The analysis of flow indices was performed in a total of 21 patients (5, 3, 6, 4, and 3 with no DR, mild NPDR, moderate NPDR, severe NPDR, and PDR respectively). Multivariate linear regression models were used to determine the correlation of flow indices with diabetic retinopathy severity after adjustment of clinical variables using R (http://www.r-project.org).
Results :
There were statistically significant inverse correlations between the flow indices (flux and perfusion) and the severity of diabetic retinopathy. Representative images of each stage of diabetic retinopathy are shown in Figure. The flux index (raw scores) decreased by 6.74 (95% CI 5.88, 7.59, p <2x10-16) with each increase in score of DR after multivariate adjustment for age, gender, logMAR visual acuity, type of diabetes, same subject intercorrelation, and the ETDRS subfields. The perfusion index decreased by 1.63% (95% CI 1.33, 1.94, p<2x10-16) after adjusting for the aforementioned covariates.
Conclusions :
The flow indices based on OMAG imaging allows quantitative evaluation of the microvascular changes seen in diabetic retinopathy and there is an inverse correlation with the severity of DR. The flow indices may be an important non-invasive tool to diagnose and monitor the presence and progression of diabetes related retinal microvascular disease.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.