Stimulus size and the variability of the threshold response in the central and peripheral visual field

Abstract

Purpose: The investigation of the peripheral visual field has shown considerable interest for the investigation of field loss attributed to anticonvulsant therapy. The purpose was to determine the within‐visit between‐subject, the between‐visit between‐subject, and the between‐location variability of the threshold response in the normal eye with increase in stimulus eccentricity out to 60° as a function of stimulus size.Methods: Forty‐eight normal subjects attended for a total of three visits (mean age = 49.5 years, SD = 18.9, range 22–84 years). At the first visit, one randomly assigned eye of each subject was examined with the Humphrey Field Analyzer 750 (Carl Zeiss, Jena, Germany) and the Full Threshold algorithm using Programs 30–2 and 60–4 and stimulus sizes III and V. The combination of stimulus size and of program, and the order of the combination within‐ and between‐sessions, were randomized for each subject. The results of the first visit were considered as a familiarization period and were discarded. The protocol at the second and third visits was identical to that at the first visit for each subject.Results: The ratio of the SD of the group mean sensitivity was determined at each stimulus location for stimulus size III compared with stimulus size V for Programs 30–2 and 60–4 at visit 3. The SDs were greater than unity for Program 30–2 (p < 0.0001) and for Program 60–4 (p < 0.0001) indicating greater variability for the size III stimulus. The SDs were also greater than unity for the central inner zone (p < 0.0001), central outer zone (p < 0.0001) and peripheral inner zone (p < 0.0001). The ratios in the peripheral outer zone were not quite greater than unity (p = 0.054). The ratios increased with increase in eccentricity by up to 2.7 times between 15° and 30° eccentricity and by up to 2.7 times between 30° and 60° eccentricity. The group mean ratio did not vary significantly between the two visits for Program 30–2 stimulus size III (p = 0.563), Program 60–4 stimulus size III (p = 0.935) and for Program 60–4 stimulus size V (p = 0.005). However, the group mean SD was lower at visit 3 compared with visit 2 for Program 30–2 stimulus size V (p = 0.0004). The SDs associated with the extreme peripheral locations in the superior and nasal fields were smaller for stimulus size III because the threshold was frequently attenuated by lid and facial contour.Conclusions: Considerably narrower confidence limits for normality for the peripheral regions of Program 30–2 and for 60–4 are demonstrated with the use of Goldmann size V.