Comparison between Widefield En Face Swept-Source OCT and Conventional Multimodal Imaging for the Detection of Reticular Pseudodrusen

Purpose: The ability to detect reticular pseudodrusen (RPD)/subretinal drusenoid deposits (SDDs) using 12×12-mm widefield en face swept-source optical coherence tomography (SS-OCT) imaging was compared with conventional multimodal imaging (color, fundus autofluorescence (FAF), and infrared reflectance [IR] imaging) in eyes with nonexudative age-related macular degeneration (AMD).

Design: Cross-sectional study.

Participants: Patients with nonexudative AMD were prospectively enrolled in an SS-OCT imaging study at the Bascom Palmer Eye Institute.

Methods: On the same day, all participants underwent color, FAF, and IR fundus imaging, as well as imaging with a prototype Zeiss 100 kHz SS-OCT instrument (Carl Zeiss Meditec Inc, Dublin, CA). Two masked graders assessed the presence, absence, or uncertainty of RPD/SDDs on conventional multimodal images and separately on 4 different SS-OCT en face images derived from the same volumetric dataset. The results from grading the conventional images and the SS-OCT en face images were compared.

Main outcome measures: Agreement in the detection of RPD/SDDs using different imaging modalities.

Results: A total of 307 eyes (209 patients) were graded for the presence or absence of RPD/SDDs. The agreement between SS-OCT and multimodal imaging was 83%. The difference in RPD/SDD detection with either image modality was not statistically significant (P = 0.21). The sensitivity of SS-OCT in RPD/SDD detection was 83%, and when using conventional imaging, the sensitivity was 75%. When using SS-OCT imaging alone, 10% of RPD/SDD cases would be missed, and when using conventional imaging alone, 14% of RPD/SDD cases would be missed. The presence of RPD/SDD was confirmed retrospectively in 48 of 52 cases once the overall grading was unmasked and the graders reevaluated the conventional multimodal images and the widefield SS-OCT en face images.

Conclusions: All 4 imaging modalities used together provided the best strategy for the detection of RPD/SDDs. However, when using widefield en face SS-OCT slab imaging alone, the detection of RPD/SDDs was at least as good as conventional imaging.