Abstract
Purpose :
In a review on pupillography standards, Kelbsch et al. state that the first pupil response in a series may be excluded, as it likely will be larger due to the relatively dark-adapted retinal state before stimulation. For dim, near-threshold stimuli, whether the differences in first and second responses are significant enough to warrant exclusion of the first trial is unknown. Retaining the first measurement could improve clinical efficiency of collecting pupil responses. We evaluated the repeatability of pupillary responses to a dim light pulse.
Methods :
Using the PLR-3000 infrared pupillometer (Neuroptics), right eye monocular pupil responses were collected from 58 young, healthy adults, aged 20-40 years, with no recent history of head injury, ocular disease, or medications known to affect pupils. Subjects stood in a quiet, well-lit room (680 lux) during testing. The pupillometer cup was placed securely over the right eye, placing the eye in darkness. A 1.0 microwatt white pulse stimulus (1 sec duration) was presented, and the cup (0 microwatts background luminance) remained in place until the response completed (3 seconds). The cup was removed, responses recorded, and the procedure repeated. Two trials were separated by approximately 30 seconds. Data analysis evaluated the repeatability of the following: maximum and minimum pupil size, percent change in size, latency, average constriction velocity, maximum constriction velocity, and average dilation velocity.
Results :
All data except latency were normally distributed. One sample t test showed that mean difference across trials was not different from zero for any parameter except for maximum pupil size (p = .014). Levene’s test for homogeneity of variances showed data variance was not significantly different for trial 1 vs trial 2 for any parameters. Bland-Altman analysis showed good agreement across trials with no proportional bias in repeatability for any parameters, except for latency (p = .001, R2 = .434).
Conclusions :
In young healthy subjects with normal vision, dynamic pupil parameters in response to a dim, near-threshold light show good repeatability between first and second trials. For clinical screening, recording responses to the first trial alone may be sufficient. Whether good repeatability between first and second trials to near-threshold stimuli can be demonstrated in patients with suspected pathology remains to be explored.
This is a 2020 ARVO Annual Meeting abstract.