Abstract
Purpose :
The macula region has the highest density of retinal ganglion cells (RGC) in healthy eyes, and the RGC and inner plexiform layer (GCIPL) can be quantified using OCT technology. In principle there should be good correspondence between GCIPL content and visual function, but, standard clinical methods (standard automated perimetry, SAP), using a 10-2 program with Goldmann size III stimulus, might not be ideal because the stimulus area exceeds the spatial integration limits in this region. The purpose of this study was to investigate the relationship between GCIPL thickness and 10-2 SAP measures using size I-III stimuli in healthy eyes and compare intra-subject variability for each stimulus size.
Methods :
Data were collected from 22 healthy subjects (22-49yrs) with good systemic health and no history of ocular pathology. Each subject had 5 total visits, and all tasks were performed on one randomly selected eye. On the first and last visit a macula scan 20ox20o was acquired for GCIPL quantification. On visits 2-5, 10-2 with Goldmann sizes I-III stimuli were administered in random sequences such that each subject had 3 fields for each stimulus size. All visual field tests were administered on an Octopus 900 (Haag-Streit) system with dynamic thresholding and a stimulus duration of 200 msec. For data analysis, visual field points were aligned to the horizontal raphe, and visual sensitivity, GCIPL thickness and GCIPL volume (thickness x stimulus area) were used.
Results :
For 10-2 locations, the standard deviation of size I stimuli (range 1.03-3.86dB) was greater than for larger stimuli sizes (Size III range 0.64-2.88dB). Similarly, the test-retest variability was greater for smaller stimuli (mean = 7.19, 5.27, 3.67dB respectively). After transforming to logarithmic values the relationship between differential luminance and GCIPL volume was best described using a two line function following spatial summation, with the data for smaller stimuli falling on the steeper slope (inflection point = 5.18dB GCIPL Volume).
Conclusions :
The relationship between structure and function follows that predicted by spatial summation. Although smaller size stimuli have greater variability, the results suggest that a significant loss in GCIPL should precede a statistically significant loss in visual sensitivity using standard SAP parameters.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.