Automated identification and quantification of subretinal fibrosis in neovascular age-related macular degeneration using polarization-sensitive OCT

Abstract

Purpose: To identify and quantify subretinal fibrosis in eyes with advanced neovascular age-related macular degeneration (nAMD) using polarization-sensitive optical coherence tomography (PS-OCT). Methods: Eyes of patients with subretinal fibrosis secondary to nAMD were included in this case series. All patients underwent a complete ophthalmic examination to clearly identify advanced nAMD lesions with fibrosis. Examinations of PS-OCT were performed using a novel system with an integrated eye tracker. Areas of fibrosis in PS-OCT, automatically segmented using a custom-built algorithm, were compared with conventional imaging modalities including spectral-domain OCT, fluorescein angiography, and color fundus photography in their potential to visualize fibrosis in nAMD. Results: Fifteen eyes of 15 consecutive patients were included. In polarization-sensitive OCT B-scans, a distinct “column-like” pattern was observed in averaged axis orientation images. En face analysis provided a precise mapping of the fibrotic scar component. Fibrous tissue was selectively identified by PS-OCT based on birefringence in all lesions, whereas in SD-OCT, subretinal hyperreflective material (SHRM) could not be further classified into scar tissue, fibrovascular material, or other AMD-specific material. Based on simultaneous polarization analyses in PS-OCT, the level of RPE alteration could be evaluated as well, showing thinning and loss of RPE associated with subretinal fibrosis. Conclusions: Using PS-OCT, subretinal fibrosis can be identified as an intrinsically birefringent structure and can be segmented based solely on tissue-specific contrast. Polarization-sensitive OCT offers a unique method to identify clinically relevant components of SHRM (i.e., neovascular tissue versus fibrous tissue) and therefore allows for an optimized disease management and evaluation of therapeutic strategies.