Perimetric testing patterns focusing on the central visual field region can be more sensitive to glaucomatous progression than more peripheral patterns

Abstract

Purpose : To evaluate the sensitivity to detect functional progression based upon the 10-2 and 24-2 visual field (VF) testing patterns using permutation of pointwise linear regression (PoPLR) analysis.

Methods : A series of 24-2 and 10-2 VF tests were obtained from 103 eyes/patients with, or suspected, glaucoma, as part of the observational Macular Progression Study (MAPS). The median number of VF tests per series was 10 (range: 6-14) with a mean follow up of 4.3 years (range: 1-7 years). Total deviation values from all 24-2 and 10-2 locations were obtained (Fig. 1). Progression was assessed using PoPLR to construct a 1-tail significance test for the alternative hypothesis that there is loss occurring over time.[1] In addition, to the 24-2 and 10-2 VF testing patterns, we also evaluated: 1) the central 12 locations of the 24-2 alone (24-2C12) (Fig. 1B, 12 locations encircled with red); and, 2) a hybrid pattern that involved the 10-2 and the 24-2 excluding the central 12 locations (24-2Per+10-2). Positive rates (PR) and false positive rates (FPR) were obtained for detection of progression. For comparison purposes, PR was also obtained for simple linear regression of the mean deviation (MD) of the 24-2 and 10-2 test patterns.

Results : At a 5% significance level (Fig. 2), point-wise analyses on the 10-2 pattern had the highest PR at 28%, while the 24-2 VF test pattern had the lowest PR (23%). The 24-2C12 and the 24-2Per+10-2 patterns, which contain information from the central field region, also outperformed the 24-2; PRs: 26% and 27%, respectively. Analyses based upon point-by-point measures outperformed those of global indices of the 24-2 (PR: 24%) but not of the 10-2 (PR: 28%).

Conclusions : A pointwise approach for detecting progression of glaucomatous damage based upon functional measures had greater PR than global indices of the 24-2. Sensitivity can improve by obtaining more measures (i.e., testing more locations) in the central region of the visual field. However, even the 10-2 VF with the highest reported PR should not be solely relied upon for detection of progressive damage, as structural measurements or joint (structure-function) analyses can achieve higher PRs.[2]
1. O’Leary et al., IOVS, 2012; 2. Abu et al. PLoS ONE, 2022

This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.