Abstract
PURPOSE: To assess the reliability of en face swept-source OCT (SS-OCT) imaging for quantifying macular atrophy in patients with neovascular age-related macular degeneration (AMD) and evaluate its limitations compared to green autofluorescence (GAF) imaging. METHODS: This prospective observational study included 30 patients with macular atrophy associated with neovascular AMD previously treated with anti-VEGF therapy. Multimodal imaging included GAF and en face SS-OCT. Macular atrophy was quantified as regions of hypoautofluorescence on GAF and hypertransmission defects (hyperTDs) on en face SS-OCT. Two masked graders measured atrophy size, followed by qualitative analysis of inter-modality discrepancies. Statistical comparisons were conducted using the Wilcoxon test and Spearman correlation. RESULTS: Mean macular atrophy size was 5.96 ± 4.48 mm² on SS-OCT and 7.10 ± 5.47 mm² on GAF, with a significant difference (p < 0.001) but strong correlation (ρ = 0.998, p < 0.001). Discrepancies arose due to hyperpigmentation obscuring hyperTDs and macular haemorrhages on SS-OCT underestimations of atrophy. Additionally, fibrosis also reduced choroidal hypertransmission, complicating hyperTD detection on SS-OCT. Border delineation inconsistencies were also observed, with diminished autofluorescence on GAF not always corresponding to hyperTDs on SS-OCT. CONCLUSIONS: En face SS-OCT is a reliable imaging modality for quantifying macular atrophy in neovascular AMD, with strong correlation to GAF measurements. However, hyperpigmentation, fibrosis, haemorrhages, and variability in border delineation introduce significant challenges. These findings underscore the need for careful interpretation of atrophy metrics and highlight the importance of addressing inter-modality discrepancies in clinical and research applications.