title_logo menu_logo
DiaComp Funded Abstracts

Program Application Abstract

Diabetic Retinopathy Severity and Retinal Oximetry in Diabetic Patients
P., Nicholas   (Joslin Diabetes Center - Boston)
There are more than 415 million persons globally with diabetes, the vast majority of whom will eventually develop diabetic retinopathy (DR). DR is characterized by alterations in the retinal blood vessels and blood flow resulting in retinal nonperfusion and ischemia. The subsequent altered retinal oxygenation may play a role in the progression of DR. The use of multi-­modality retinal imaging might permit accurate correlation between retinal vascular oxygenation, retinal nonperfusion and clinical retinopathy severity. Indeed, recent advances in spectrophotometric retinal oximetry now allow quantitative noninvasive determination of oxygen saturation in retinal vessels. This study evaluated twenty eyes of 10 subjects with diabetes to assess the relationship between severity of diabetic retinopathy, retinal nonperfusion and retinal ischemia as reflected by retinal oximetry. Early Treatment Diabetic Retinopathy Study (ETDRS) retinopathy severity was evaluated based on stereoscopic ultrawide field retinal images. Retinal oximetry performed by a certified imager using the Oxymap T1 (Oxymap®, Reykjavik, Iceland), 200 degree ultrawide field retinal imaging and fluorescein angiography (Optos California, Optos® plc, Dunfermline, United Kingdom) were acquired after pharmacologic pupillary dilation. All images were evaluated in a masked manner by standardized protocol at a centralized reading center. Mean age (±SD) was 44.1±10.13 years, diabetes duration 18.5±10.1 years, hemoglobin A1c 7.9±0.8%, 60% (12) female and 70% (14) type 1 diabetes. Distribution of DR severity based on ultrawide field images: no DR 20% (4), mild nonproliferative DR (NPDR) 35% (7), moderate NPDR 15% (3), severe NPDR 10% (2), proliferative DR (PDR) 2 (10%), quiescent PDR (QPDR) post prior scatter (panretinal) laser photocoagulation 2 (10%). The extent of retinal nonperfusion (mm2) on UWF-­FA was significantly associated with increasing DR severity from no DR to PDR (no DR: 35.2 mm2, mild NPDR: 58.1 mm2, moderate NPDR: 102.2 mm2, severe NPDR 142.8 mm2, PDR: 345.0 mm2, p=0.0194) and this association remained significant after correcting for diabetes duration and hemoglobin A1c (p=0.0381) (Figure 3). The extent of nonperfusion in eyes with scatter (panretinal) photocoagulation (QPDR: 59.8 mm2) was significantly lower compared with eyes with PDR and severe NDPDR (p=0.0156). Venous oxygen saturation, which reflects extent of retinal ischemia, increased steadily with increasing DR severity, but this association was not statistically significant, possibly due to low participant numbers (no DR: 66.5%, mild-­moderate NPDR: 68.0%, severe NPDR: 69.1%, PDR: 71.3%, pairwise comparisons with no DR, p=0.60, 0.55,0.08 respectively). In eyes with NPDR, increasing retinal venous oxygen saturation was correlated with increasing retinal nonperfusion (r= 0.89, p=0.046). Increasing retinal nonperfusion measured using ultrawide field fluorescein angiography is clearly associated with increasing DR severity. Increasing retinal venous oxygen saturation (reflecting increased retinal nonperfusion) is associated with retinal nonperfusion and may be associated with DR severity as well. If these findings hold true in larger studies, then UWF-­FA and retinal oximetry may provide early methods to identify changes in retinal perfusion status and DR risk, thus potentially allowing institution of interventions before clinical evident retinal compromise is observed. Furthermore, retinal oximetry could accomplish this identification noninvasively.