Implementation of the new multimodal imaging-based classification of central serous chorioretinopathy

Abstract

Purpose:

To study the implementation of the new multimodal imaging-based classification system of central serous chorioretinopathy (CSCR).

Methods:

Ninety-three eyes with CSCR with available fundus autofluorescence (FAF), optical coherence tomography (OCT), and OCT angiography at presentation were included in this study. An anonymous data set was classified by two masked graders. Each case was classified as per presence of (i) simple versus complex (< or >2 disc diameters of retinal pigment epithelium abnormality) CSCR; (ii) primary versus recurrent versus resolved CSCR; (iii) persistent (presence of subretinal fluid >6 months) or not; (iv) outer retinal atrophy (ORA); (v) foveal involvement; and (vi) macular neovascularization (MNV). Agreement between the graders was calculated.

Results:

Kappa value was 0.91 (95% CI 0.8–1.0) for the entire classification; 0.84 (95% CI 0.73–0.95) for simple versus complex; 1.0 (95% CI 1.0–1.0) for primary versus recurrent versus resolved CSCR; 1.0 (95% CI 1.0–1.0) for persistent or not; 0.9 (95% CI 0.81–0.99) for ORA or not; 0.95 (95% CI 0.84–1.0) for presence or absence of MNV; 1.0 (95% CI 1.0–1.0) for presence or absence of foveal involvement.

Conclusion:

The new multimodal imaging based CSCR classification showed “near perfect” agreement between two retinal experts.