Revalidating the Usefulness of a "Sector-Wise Regression" Approach to Predict Glaucomatous Visual Function...

Area of Science:

• Ophthalmology
• Glaucoma Research
• Visual Field Analysis

Background:

• Previous development of "Progression sectors" for measuring glaucomatous visual field (VF) progression.
• Prior analysis focused on short VF series where pointwise linear regression (PLR) is less reliable.
• Need to evaluate sector-wise approach in longer VF series with improved PLR accuracy.

Purpose of the Study:

• To investigate the utility of a sector-wise approach for predicting future VF progression.
• To compare the accuracy of "Progression sectors" against PLR and "Nouri-Mahdavi" (NM) sectors in longer VF series.

Main Methods:

• Analysis of 16 VFs from 71 eyes across 45 patients.
• Calculation of sectorial averages of total deviation (TD) values for 23 "Progression sectors" and 10 NM sectors.
• Application of linear regression to predict sectorial averages in VFs 11-16 and comparison of absolute prediction errors with PLR.

Main Results:

• "Progression sectors" showed significantly smaller absolute prediction errors than PLR for VFs 14-16 (P < 0.05).
• NM sectors exhibited significantly larger prediction errors compared to both PLR and "Progression sectors" for VFs 11-16 (P < 0.01).
• The sector-wise approach using "Progression sectors" proved more accurate in predicting future VF progression.

Conclusions:

• "Progression sectors" are valuable for predicting future VF progression across both short (1-10 VFs) and long (>10 VFs) series.
• This method demonstrates superior utility compared to PLR and NM sectors for glaucoma monitoring.
• The sector-wise approach offers a more reliable tool for assessing visual function changes in glaucoma patients.