Within-test Variability of Frequency-doubling Perimetry Using a 24–2 Test Pattern

PurposeTo evaluate patient-response (within-test) variability for targets of the smaller frequency-doubling technology perimetry test that employs a 24–2 stimulus-presentation pattern. MethodsPatient-response variability was examined using the method of constant stimuli for standard (10°) and small (4°) customized frequency-doubling technology perimetry stimuli presented on a CRT screen. Small stimuli were designed for use in a 24–2 test pattern. Matched test locations were examined in 24 subjects (8 normal, 8 in whom glaucoma was suspected, and 8 glaucoma patients). Threshold sensitivity (in decibels for the 50% detection level) and variability (interquartile range in decibels) were obtained from frequency-of-seeing curves derived from data fitting with cumulative Gaussian functions. Groups were compared using a two-way ANOVA. ResultsThresholds obtained using standard and small stimuli were highly correlated (R = 0.94, P < 0.001, Pearson correlation), although smaller targets systematically estimated sensitivity to be 2.0 dB (95% CI, 1.7–2.4 dB) lower than standard targets. No significant difference in patient-response variability was observed between standard and small targets (P = 0.067), although both target sizes demonstrated small but significant increases in variability with reduced sensitivity. Mean (SD) patient-response variability for the normal, suspect, and glaucoma groups was 1.0 (0.6), 0.9 (0.4), and 1.8 (1.4) dB for standard-sized stimuli and 1.1 (0.8), 1.5 (1.2), and 2.0 (0.9) dB for small stimuli. ConclusionsSmall (4°) frequency-doubling technology perimetry targets have variability characteristics that are not statistically significantly different from those observed for standard-sized (10°) stimuli. Reduction in frequency-doubling technology perimetry stimulus size necessary to produce 24–2 test resolution is unlikely to affect test repeatability. Smaller, more numerous stimuli may offer clinical advantages both in terms of detecting small defects and identifying progressive loss.