Abstract
Abstract: :
Purpose: To determine the test–retest variability of the Humphrey Matrix perimeter in patients with glaucoma. Methods: Our sample consisted of 15 open–angle glaucoma patients with a mean age of 62.5 years (range, 54 – 80 years) with early to moderate visual field loss. Inclusion criteria were a clinical diagnosis of open–angle glaucoma, Mean Deviation (MD) with Standard Automated Perimetry (SAP, using SITA Standard) better than –15.00 dB, no chronic eye disease or other or systemic disease known to affect visual field, refractive error < 5 D equivalent sphere and astigmatism < 3 D, and VA of 6/12 or better. We performed two Humphrey Matrix 24–2 threshold tests separated by a break of at least 10 minutes to reduce fatigue and/or adaptation effects. Ambient illumination was kept at approximately 100 lux. One eye was randomly selected for the analysis. We determined both the standard deviation and the empirically derived 95% limits of retest variability stratified according to the values of the baseline test. Results: The average MD of the Humphrey Matrix results was –4.52 dB (range, 2.31 to –14.01 dB) while the corresponding value for SAP was –3.17 dB (range, 0.58 to –9.28 dB). There was a high correlation between the MD values obtained with the Humphrey Matrix and SAP (Spearman’s r = 0.90; P < 0.01). We did not observe a significant learning effect between the 2 test sessions (mean difference, 0.2 dB, P = 0.48, paired t–test). Test–retest variability increased significantly with lower sensitivity (P < 0.01). The standard deviation of retest values were 3.4, 3.6, and 6.5 dB with baseline values centred at 30, 20 and 10 dB. The empiric 5% and 95% retest limits showed a similar trend with values of 13 and 32 dB respectively for a baseline value of 27 dB and 0 and 27 dB respectively for a baseline value of 7 dB. Conclusions: The test–retest variability of Humphrey Matrix perimetry increases with loss of sensitivity. In this study, the variability in moderately damaged locations increased by a factor of 1.5 to 2 compared to normal locations.
Keywords: perimetry • visual fields • clinical research methodology