Choriocapillaris (CC) Analysis in Subjects with Intermediate Age-Related Macular Degeneration and Geographic Atrophy using SD and SS OCT Angiography

Abstract

Purpose : To compare the vascular flow voids in CC of subjects with iAMD, GA, and age-matched controls.

Methods : Subjects with iAMD, GA and healthy controls were enrolled in a prospective observational study (The OREO Study). Subjects were assessed quarterly. OCTA images were obtained using Optovue AngioVue and Zeiss PlexElite. A 6mm x 6mm cube OCTA, centered on fovea, was obtained for each subject. CC, superficial, and avascular slab were extracted from the OCTA exams. To identify drusen, a slab of 5um at ISOS was also extracted. A custom MATLAB software was used for analysis. Big vessels identified from superficial slab, and drusen from the drusen slab were annotated in a semi-automatic manner where graders manually corrected the masks, if needed. These areas were excluded from the analysis. The rest of the image was analyzed to measure average pixel intensity. The pixels which were lower than 1 SD of the mean are identified as flow voids. The flow voids are then categorized based on size. For patients with GA, lesion and transition zone boundaries were imported from OCT and FAF and overlayed on CC OCTA following registration.

Results : 64 eyes imaged with both Optovue and PlexElite were analyzed. The mean flow void percent was 18.52 ± 0.80% and 15.49 ± 0.67% and the mean total number of flow voids was 6189.9 ± 171.4 vs 7213.1 ± 1297.1 for Optovue and PlexElite respectively. The mean change from baseline at week 24 for both devices is demonstrated in Table 1. Larger flow voids (800mm²), assessed using Optovue OCTA, increased in both iAMD and GA lesion compared to age-matched controls (Figure 1A-1B). Mean change in total void percent area remained similar over 36 weeks with a mean change of 0.15% in iAMD vs -0.06% in controls. Eyes with GA demonstrated a mean change in total void percent of 17.0% for GA, 17.8% for TZ, 14.4% for healthy retina, compared to -12.0 for age-matched control eyes over 12 weeks.

Conclusions : SD OCT can be reliably used to assess changes in CC. Larger flow voids tend to increase in number in areas of GA. Eyes with iAMD demonstrate similar trends in changes in flow voids compared to controls.

This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.

 

Mean change in CC attributes at Week 24 compared to baseline for Optovue vs PlexElite OCTA

 

Figure 1A, 1B: Change from baseline in Flow Voids in Subjects with iAMD and GA compared to Controls by flow void size, respectively.