Abstract
Abstract: :
Purpose: To compare visual field progression with glaucoma change probability (GCP) analyses based on total (TD) and pattern deviation (PD). Methods: The study sample consisted of 109 patients with early to moderate glaucoma (mean age at baseline, 63.1 yrs, range 17 to 89 yrs). They were followed prospectively with 6-monthly 30-2 full-threshold tests for a mean of 7.1 yrs (range 1.5 to 9.8 yrs). Of each patient, one eye (mean MD –4.8 dB, range +0.8, –13.0 dB) was randomly selected for analysis. Significance limits for change (p<5%) in TD and PD were derived from test-retest data of a separate group of 64 glaucoma patients. The fields were evaluated for progression with 5 increasingly conservative criteria, based on the number of locations with TD or PD below the 5% retest limit of the baseline on 3 consecutive follow-up tests. Scores ranging from 0 to 5 (for 0, 1, 2, 3, 5 and 10 progressing locations, respectively) were given to each visual field, depending on the largest number of progressing locations throughout the follow-up. If the visual field did not progress even with the least conservative criterion (≥1 progressing location), it was given a score of 0. If the field progressed even with the most conservative criterion (≥10 progressing locations) it was given a score of 5. Results: 38 (35%) patients did not progress with either TD or PD GCP (score=0). Seven (6%) patients progressed even with the most conservative criterion on both TD and PD GCP (score=5). In 73 (67%) patients, the progression scores with TD and PD GCP agreed exactly. In 25 (23%) patients, the progression score was higher with TD, and in 11 (10%) patients, it was higher with PD. Although statistically significant, the average difference between TD and PD scores was small (0.25, p<0.01, Wilcoxon matched-pairs test). When patients were classified as progressing with both TD and PD GCP, the "time to progression" tended to be shorter with TD, but the mean differences in progression time were <6 months with each criterion and in no case statistically significant. At each criterion, PD GCP classified at least 1 patient as progressing who was classified as stable with TD GCP. Conclusions: In this sample of early to moderate glaucoma patients, GCP analyses showed a slightly lower rate of visual field progression with PD compared to TD. This may either be due to greater specificity against non-glaucomatous diffuse visual field change (eg, that caused by increasing cataract), or due to lower sensitivity to true glaucomatous progression.