Abstract
Purpose:
Contrast sensitivity (CS) is frequently affected in glaucoma, but is not frequently used as part of clinical care. One impediment to the use of contrast sensitivity to monitor disease severity is the high variability of contrast values on repeat testing. Here, we examine test-retest parameters of two contrast sensitivity tests (the MARS contrast sensitivity chart and the Adaptive Sensory Technology quick CSF test) in a group of glaucoma patients with a range of disease severity.
Methods:
Both CS tests were evaluated in 30 glaucoma patients under the following 4 conditions: right eye, left eye, both eyes, and worse-seeing eye (retested). The MARS test measured CS at a single letter size, with letters read were converted into logCS. The quick CSF test tested contrast sensitivity at varying letter size using a thresholding algorithm. Results were summarized as the area under the contrast/letter size curve (AULCSF), as well as logCS at 3 cycles/degree (a text size roughly equivalent to the MARS print size).
Results:
Thirty glaucoma patients completed the testing procedures. Subjects had a mean visual field mean deviation of -12.3 dB in the eyes where CS testing was repeated (SD=8.2 dB, range=-0.1 to -30.9 dB). Mean AULCSF values in retested eyes was 0.77 (0.48), and the mean (SD) test-retest difference in AULCSF was 0.02 (0.09). Bland-Altman plot showed no relationship between the difference in AULCSF and the average AULCSF across the 2 trials (p=0.995), and the coefficient of repeatability for AULCSF was 0.24. Mean test-retest differences in MARS logCS and quick CSF logCS at 3 cpd were 0.12 (.10) and 0.03 (0.15), respectively, with neither value showing a relationship between the difference in logCS and average logCS across the 2 trials (P>0.9 for both). The coefficients of repeatability for MARS log CS and quick CSF logCS at 3 cpd were 0.28 and 0.40, respectively.
Conclusions:
Precision of contrast sensitivity testing is constant over a wide range of glaucoma damage, suggesting that it may be useful to gauge disease severity, particularly in patients with very advanced disease. AULCSF may be more precise than logCS values calculated from a single letter size, though longitudinal testing will be necessary to determine which contrast sensitivity metrics, if any, are useful in judging progression in eyes with advanced glaucoma.