Diabetic retinopathy (DR), associated with long-term diabetes mellitus, is a leading cause of blindness in the US. Structural optical coherence tomography (OCT) has become the standard method for evaluating diabetic macular edema;however, fluorescein angiography (FA), which requires an invasive intravenous dye injection, is still needed to assess capillary dropout and confirm neovascularization. Recently, we have used a high- speed OCT system to make non-invasive functional measurements such as angiography, total retinal blood flow, and retinal arterial pulsatility. The goal of the proposed project is to improve these functional OCT-derived biomarkers and determine their roles in the early detection and management of DR. 1. Develop quantitative OCT angiography of capillary dropout and retinal neovascularization (RNV). Three dimensional (3D) OCT angiography has been made practical (6x6 mm scan in 3 seconds) by a novel split-spectrum amplitude-decorrelation algorithm. The algorithm will be improved and OCT angiograms will be calibrated by physical flow phantoms. Automated software will be developed to quantify the following angiographic biomarkers: non-perfusion area, RNV area and flow index, and parafoveal and perifoveal flow indexes. 2. Improve Doppler OCT measurement of total retinal blood flow (TRBF) and retinal arterial pulsatility. A 3-second 3D Doppler OCT scan will be used to measure TRBF using an en face summation algorithm. A 4-second cylindrical Doppler OCT scan will be used to measure retinal arterial pulsatility. Automated measurement algorithms will be developed for these novel biomarkers. 3. Improve structural OCT measurement of macular edema. By registering several volumetric scans, we have demonstrated detailed mapping and quantification of retinal thickening and fluid spaces (cysts). Fully automated quantification of these structural biomarkers will be developed. 4. Evaluate functional and structural OCT-derived biomarkers in clinical studies. The specificity and sensitivity of OCT angiography for detection of capillary dropout and RNV will be determined in 50 severe non-proliferative (NPDR) and proliferative (PDR) subjects (Group A) and 30 age-matched non-diabetic controls (Group C). The utility of OCT-derived biomarkers for early detecting retinal abnormalities will be evaluated in 80 subjects with or without mild to moderate NPDR (Group B). The capability of biomarkers for assessing the disease change will be evaluated with Groups A and B over a 12-month interval. The proposed functional OCT-derived biomarkers will be more sensitive to early vascular change compared to standard fundus photography and are likely to provide more reliable endpoints for diabetic clinical trials. Unlike FA, functional OCT is non-invasive and coul be used in routine clinical screening and monitoring of DR.
Diabetes is a systemic microangiopathic disorder. Early detection and quantification of alterations in the retinal microvasculature not only identify diabetic retinopathy (DR) in early stages but also serve as a biomarker for the degree of microangiopathy indicative of non-perfusion in other areas of the body affected by diabetes. We will develop optical coherence tomography (OCT) angiography to detect non-perfusion and neovascularization and Doppler OCT to assess global circulation. We will then demonstrate their utilities in a clinical study of DR.