Swept Source OCT Angiography of Polypoidal Choroidal Vasculopathy using Boundary Layer Specific Segmentation

Abstract

Purpose : The gold standard for the identification of polyps and branching vascular networks (BVNs) in polypoidal choroidal vasculopathy (PCV) is indocyanine green angiography (ICGA). Swept source optical coherence tomography angiography (SS-OCTA) is a non-invasive alternative. Efforts to image PCV with OCTA have been limited by poor image quality, inability to image polyps, and the difficulty of capturing the entire PCV complex on a single en face image. Herein, we use defined segmentation boundaries to obtain en face images of the PCV complex, which we follow after anti vascular endothelial growth factor (VEGF) therapy.

Methods : Patients with PCV were enrolled in a prospective study at Bascom Palmer Eye Institute. Patients underwent imaging with a SS-OCTA instrument (PlexElite 9000, Carl Zeiss Meditec Inc., Dublin, CA). En face slabs for both structure and flow imaging were produced using boundary layers from the retinal pigment epithelium (RPE) to the RPE-fit layer, roughly following Bruch’s membrane. These layers were established using the Custom Segmentation option. The position and thickness of the slab were manually adjusted to optimize the signal to noise for viewing of the PCV complex. The automated segmentation was manually corrected when necessary. Two visualization algorithms were employed: the MAX algorithm utilizes the greatest flow signal for each A-scan in the slab, while the SUM algorithm adds the flow signals in the slab. En face and cross-sectional scans with and without flow were reviewed to determine whether all polyps could be identified. SS-OCTA images were compared with ICGA images when available.

Results : Three eyes of two patients were imaged. Two eyes were treated with anti-VEGF therapy. In all eyes, the polyps and connecting BVNs were visualized using a single slab configuration. The MAX algorithm clearly displayed the BVNs, and the SUM algorithm identified both the polyps and the BVNs. On B-scan, polyps could be identified by their flow signature within the RPE detachments. ICGA was available in one case, and the PCV complex appeared identical using both SS-OCTA and ICGA imaging.

Conclusions : SS-OCTA combined with a custom boundary identification strategy for the en face and cross-sectional viewing of flow images detected polyps and the BVNs associated with PCV. Additional patients are being recruited to validate this SS-OCTA imaging strategy against ICGA imaging.

This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.