Validation of central serous chorioretinopathy multimodal imaging-based classification system

Area of Science:

• Ophthalmology
• Retinal Imaging Analysis
• Diagnostic Classification Systems

Background:

• Central serous chorioretinopathy (CSCR) is a condition affecting vision, often requiring accurate classification for effective management.
• Existing classification systems may lack comprehensive validation, impacting diagnostic consistency among clinicians.
• Multimodal imaging plays a crucial role in diagnosing and classifying retinal diseases like CSCR.

Purpose of the Study:

• To validate a novel classification system for central serous chorioretinopathy (CSCR).
• To assess the inter-grader agreement among retina physicians using this new classification system.
• To evaluate the influence of affected eye information on the grading of the fellow eye.

Main Methods:

• A cross-sectional study involving 10 retina physicians evaluating 61 eyes from 34 patients with presumed CSCR.
• Utilized multimodal imaging (fundus autofluorescence, fluorescein angiography, indocyanine green angiography, OCT) and clinical details.
• Inter-grader agreement was measured using Fleiss Kappa, comparing grading with and without knowledge of the affected eye.

Main Results:

• Moderate agreement (Fleiss Kappa = 0.50) was observed for major CSCR criteria, improving to 0.57 after excluding an outlier.
• Substantial agreement was found for CSCR subtypes (simple, complex, no CSC) and disease status (primary, recurrent, resolved).
• Inclusion of the affected eye's grading significantly improved overall agreement (Kappa = 0.642) compared to grading the fellow eye alone (Kappa = 0.255).

Conclusions:

• The novel CSCR classification system demonstrates fair-to-moderate inter-grader agreement, suggesting a need for refinement.
• Grading both eyes is crucial for accurate CSCR assessment, as information from the affected eye impacts fellow eye classification.
• The study highlights the importance of considering bilateral findings for consistent CSCR diagnosis and classification.