Which OCTA Scanning Protocol Is Best for PDR and NPDR?
American Journal of Ophthalmology, July 2020
Although optical coherence tomography angiography (OCTA) can help detect microvascular changes indicative of diabetic retinopathy (DR), studies have failed to establish a protocol that optimally balances scan area and lesion detection and also works well in a busy clinical setting. Zhu et al. recently compared OCTA scan protocols; they found that 15×9-mm Montage imaging was significantly better than 6×6-mm Angio imaging for detecting DR lesions (with the exception of microaneurysms). However, they also found that 12×12-mm Angio scanning centered on the fovea and optic disc was comparable to 15×9-mm Montage imaging for discerning lesions without sacrificing sensitivity or speed.
For this study, the authors recruited 119 patients (176 scanned eyes) with proliferative DR (PDR; n = 80), nonproliferative DR (NPDR; n = 73), and diabetes mellitus without DR (n = 23). Eyes were imaged with swept-source (SS)-OCTA and multiple scan protocols in the following order: 3×3-mm Angio centered on the fovea; 6×6-mm Angio centered on the fovea and optic disc; 15×9-mm Montage; and 12×12-mm Angio centered on the fovea and optic disc.
Two ophthalmologists independently assessed the images for DR lesions, including microaneurysms, intraretinal microvascular abnormalities, neovascularization, nonperfusion, venous looping, and hard exudates.
Results were as follows:
- For neovascularization in the optic disc and elsewhere, the detection rate with 6×6-mm Angio centered on the fovea was approximately half that of 15×9-mm Montage (17.6% vs. 34.6%, respectively; p < .05).
- With 6×6-mm Angio centered on the fovea and optic disc, the rate was roughly two-thirds that of 15×9-mm Montage (26.1% vs. 36.2%, respectively; p < .05).
- In detecting microaneurysms, 6×6-mm Angio centered on the fovea and the 6×6-mm Angio scan combination outperformed Montage imaging (85.2% vs. 79.0% and 84.8% vs. 79.0%, respectively; both p < .05).
- The 12×12-mm Angio images centered on the fovea and optic disc had detection rates comparable to those of 15×9-mm Montage images for all DR lesions (p > .05); however, the rates for nonperfusion and neovascularization were slightly higher with Montage images in patients who received both scans.
These findings support the use of widefield SS-OCTA for distinguishing PDR from NPDR, the authors said. They suggested that 12×12-mm Angio scans centered on the fovea and optic disc are a practical alternative to Montage imaging in busy clinical practices.
The original article can be found here.