Abstract
Purpose :
Central visual function using 10-2 testing has been shown to be valuable in detecting glaucomatous loss early in disease, but significant losses of the ganglion cell layer occur prior to reduction in visual thresholds. One potential explanation is that the standard size III stimulus exceeds the area of complete spatial summation, and smaller stimuli would have better sensitivity and correspondence with in vivo structure. In order to begin to determine how the precision vs. reliability trade-off will affect the diagnosis of glaucomatous neuropathy, this study was undertaken to establish normative perimetry data for Size I-V stimuli in the non-human primate.
Methods :
14 non-human primates (Macaca mulatta), with healthy eyes, were trained for behavioral perimetry. 10 of the animals completed the 10-2, full threshold Humphrey visual fields with less than 10 percent false positives and false negatives for all stimulus sizes. For each stimulus size, the mean, standard deviation and repeatability for each test location was determined. The mean sensitivity (dB units) at 4 eccentricities (1.41°, 4.24°, 7.07°, and 9.06°) was used to determine the critical area (Ac) for spatial summation.
Results :
The mean thresholds increased with stimulus size (Size I – 25.1±1.0dB, Size II – 29.3±0.8db, Size III – 32.9±0.6bd, Size IV – 35.6±0.6db, Size V – 37.5±0.5bd, p<0.01). The Ac at 1.41° was 14.3 min or 0.23°, slightly larger than a size II stimulus (0.21°), and increased in size with eccentricity to 0.25° at the 9.06° test location. Size I and II stimuli had greater inter-individual variability (2.6±0.34dB and 2.2±0.5dB, respectively) compared to Size III-V (p<0.01). Although the intra-individual repeatability was better for larger stimulus sizes (p<0.01), it was not different from Size II and III stimuli after accounting for multiple comparisons (p=0.02).
Conclusions :
The standard, Goldmann Size III stimulus exceeds the Ac for all visual field locations in the central 10-2 region of the non-human primate. Based on the significant threshold variability between individuals, smaller stimuli are expected to have similar sensitivity for disease detection compared to the standard Size III. However, the similar intraindividual variability would suggest that smaller stimuli would be ideal for detecting progression.
This is a 2020 ARVO Annual Meeting abstract.