Abstract
Purpose :
Previous studies have reported that the detection rates of microaneurysms (MAs) in diabetic retinopathy (DR) using optical coherence tomography angiography (OCTA) are rather low (41~62%). Using two different types of OCTA devices, spectral-domain (SD) and swept-source (SS) OCT, we evaluated the detection rates of MAs in DR in eyes with and without macular edema.
Methods :
Fifty-three eyes of 32 patients with DR were enrolled. Macular OCTA images (3 × 3 mm) were obtained using two OCT devices: SD-OCTA, RTVue XR Avanti (Optovue, Fremont, CA) and SS-OCTA, DRI OCT Triton (Topcon, Tokyo, Japan). Fluorescein angiography (FA) also was performed in all patients. MAs were defined as hyperfluorescent spots seen in early- and/or late-phase FA images, and the rates of detection of all MAs using each OCTA device were evaluated. In 14 eyes with macular edema, we also defined “culprit MAs” located in edematous regions (retinal thickness 350 μm or greater) with leakage on late-phase FA images. The detection rates of the culprit MAs were compared to that of all MAs. Two observers independently counted the MAs seen with both OCTA devices, and the interclass correlation coefficients (ICCs) were calculated.
Results :
The detection rates of all MAs using SD-OCTA and SS-OCTA, respectively, were 67.9±14.7% and 68.3±15.4%, the difference of which did not reach significance (p=0.35). The ICCs for the counting of the MAs using SD-OCTA and SS-OCTA, respectively, were 0.984 and 0.983. In eyes with macular edema, the detection rates of all MAs versus that of the culprit MAs were 69.0±13.4% versus 83.1±6.1% using SD-OCTA and 72.1±12.1% versus 86.7±8.2% using SS-OCTA. The detection rate of the culprit MAs was significantly (p<0.01) higher than that of all MAs with both OCTA devices.
Conclusions :
Both OCTA devices (SD-OCTA and SS-OCTA) almost equally detected the MAs in the eyes with DR. Culprit MAs were detected to a greater degree using both OCTA devices than all MAs. The results suggest that OCTA is useful to evaluate clinically active MAs, which are a major cause of diabetic macular edema.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.