gms | German Medical Science

Purpose: To assess the utility of en-face high-axial resolution optical coherence tomography (OCT) in the assessment of atrophic lesions in non-neovascular age-related macular degeneration (AMD), using a research OCT prototype (High-Res OCT, Heidelberg Engineering, Germany) with 3 µm optical axial resolution.

Methods: Retrospective review of images of patients with non-neovascular AMD. We analyzed images obtained using a 20x20 ART1 OCT pattern. We collected clinical data including: type of drusen, presence of pigmentary changes and retinal pigment epithelium, and outer retinal atrophy (RORA)/hyper-transmission defects (HTD), both in the B-scans and en-face images.

Results: 46 eyes of 32 patients with non-neovascular AMD were included. Soft drusen were the most frequent deposit (n=28), followed by subretinal drusenoid deposits (SDD) (n=14), cuticular drusen (n=11), acquired vitelliform lesions (AVL) (n=4), calcified soft drusen (n=4) and drusenoid RPE detachment (n=1). HTD were present in 21 eyes in the en-face images, corresponding with areas of RORA in the B-scans. En-face images also showed different forms of hyperreflective round areas similar to the HTD. Some of these lesions had a ‘Halo’ appearance (n=35), corresponding to drusen with RPE thinning in the apex of the lesion. Other lesions had a ‘Inverted Halo’ appearance (n=14) corresponding to areas of RORA with central hyperreflective clumps corresponding to pigmentary changes. Other pigmentary changes presented a heterogeneous distribution (n=31) not related to HTD.

Conclusions: Fundus autofluorescence is the gold standard to depict areas of RPE atrophy. However, limitations of this technique include the need of pupil dilatation and the visualization of foveal lesions. En-face OCT images using high-axial resolution OCT in patients with RORA were able to show areas of HTD, but also demonstrated some confounding factors. Dispersion of pigment limited the visualization of HTD, and the presence of drusen were seen as hyperreflective lesions with a variable halo of hyporeflectivity. Analysis of the B-scans might be necessary to rule out confounding factors.