Classification and Statistical Trend Analysis in Detecting Glaucomatous Visual Field Progression

Area of Science:

• Ophthalmology
• Visual Science
• Medical Statistics

Background:

• Glaucomatous visual field progression monitoring is essential for timely treatment adjustments.
• Various classification-based and statistical methods exist for detecting visual field progression, but their agreement is not fully established.

Purpose of the Study:

• To evaluate the agreement between different methods for detecting glaucomatous visual field progression.
• To compare two classification-based methods and four statistical trend analysis approaches.

Main Methods:

• Retrospective longitudinal study of 20 open-angle glaucoma patients.
• Visual fields assessed using Humphrey Field Analyzer (SITA standard 30-2 or 24-2).
• Progression analysis using Hodapp-Parrish-Anderson classification, Advanced Glaucoma Intervention Study score, and linear regression (r²) for MD, PSD, and VFI.

Main Results:

• Fair agreement was observed between most methods for detecting progression, except for PSD r².
• Substantial agreement (κ = 0.698) was found between MD r² and VFI r².
• VFI r² demonstrated the best overall agreement with other methods.

Conclusions:

• MD r² and VFI r² showed the highest agreement in detecting glaucomatous visual field progression.
• VFI r² is a reliable method, showing good agreement with various other detection approaches.
• Glaucoma Progression Analysis software (GPA2) can aid ophthalmologists in clinical decision-making for glaucoma management.