Frequency Doubling Technology Threshold Visual Field Results Vary with Software Version

Abstract

To determine the comparability of test results obtained using two versions of commercially available frequency doubling technology (FDT) software for visual field testing. One hundred glaucoma patients underwent duplicate N-30 threshold FDT visual field examinations of one eye; one test was performed with the recently released software version 3.0 and the other test with the earlier version 2.6. Results were compared in terms of test duration, mean deviation (MD), pointwise sensitivity, pointwise deviations from the age-normal threshold, and average sensitivity. Results obtained with version 3.0 differed significantly from those obtained with version 2.6. Version 3.0 tests took on average 14.2% less time (P < 0.0001), and the average MD was 0.53 dB more negative (P < 0.01). Version 3.0 threshold estimates derived from a mixed-model analysis were significantly lower at 15 of 19 test points (P = 0.0074). At 14 of the 19 test point locations, the mixed-model analysis also showed that threshold measurements were more frequently outside the normal range with version 3.0 (P = 0.022). We found statistically significant differences between results obtained using FDT versions 3.0 and 2.6, which might affect clinical management. New age-corrected normal values for version 3.0 should be obtained and made available to users, and new baseline field data should be established whenever a patient is switched from version 2.6 to version 3.0.