Feasibility and Clinical Utility of Wide-Field Optical Coherence Tomography Angiography Compared to Ultrawide-Field Fluorescein Angiography in Patients with Diabetic Retinopathy.

PURPOSE

To test the diagnostic performance of a novel wide-field swept-source optical coherence tomography angiography (WF-OCTA) device in detecting retinal non-perfusion (NP) and neovascularization (NV) in eyes with diabetic retinopathy (DR) and to compare this with the standard-of-care imaging method, ultrawide-field fluorescein angiography (UWFFA).

METHODS

Prospective, observational, cross-sectional single-center study evaluating patients with DR imaged with WF-OCTA (Xephilio OCT-S1; Canon Inc., Tokyo, Japan) and UWFFA (Optos California; Optos plc, Dunfermline, United Kingdom). WF-OCTA images of the superficial capillary plexus (SCP) consisted of single capture 23 × 20 mm scans centered on the fovea. In UWFFA and WF-OCTA, qualitative and quantitative measurements were assessed to analyze retinal NP and NV. Vessel density (VD) in WF-OCTA and ischemic index (ISI) in UWFFA were calculated. Qualitatively, the presence of NV and NP was assessed in both WF-OCTA (posterior pole/midperipheral retina) and UWFFA (posterior pole/midperipheral retina/far peripheral retina).

RESULTS

Ten consecutive patients with variable DR severity stages (17 eyes) were evaluated. Two eyes had to be excluded due to low quality of the WF-OCTA images. Therefore, 15 eyes were included for final analysis. Mean age was 57 years (± SD: 15.2) and the male : female ratio was 4 : 6. UWFFA identified retinal NP in 11 eyes (73%). Posterior pole NP was present in eight eyes, midperiphery NP was present in eight eyes, and far periphery NP was present in seven eyes. Retinal NV was detected in four eyes using UWFFA (two eyes with only midperiphery NV). WF-OCTA detected retinal NP in 11 eyes (9 cases with both posterior pole and midperiphery NP). NV was detected in three eyes (two with posterior pole and midperipheral NV, four with only midperipheral NV). Mean VD evaluated using WF-OCTA of the SCP was 0.40 (± SD: 0.1), and mean ISI in UWFFA was 0.09 (± SD: 1.3). Spearman's test did not show a significant correlation between the ISI in UWFFA and VD in WF-OCTA (p = 0.803).

CONCLUSIONS

Noninvasive WF-OCTA has great potential for the management of patients with DR. This new imaging modality might be useful in daily clinical routine in order to lower the number of invasive examinations. However, in a small percentage of patients, OCTA images cannot be reliably graded for the presence of NP and NV. In these cases, conventional FA needs to be performed.