Assessment of vigilance and response quality during static automated perimetry. A study using the method of constant stimuli (MoCS) and an enhanced presentation rate of catch trials

Abstract

Purpose To develop a methodology to monitor and validate vigilance and response quality during static automated perimetry.

Methods The method of constant stimuli (MoCS) was applied to assess the differential luminance sensitivity with the OCTOPUS 900 perimeter (Haag-Streit AG, Koeniz, Switzerland). OPI (Open Perimetry Interface) was used to set up the test algorithms: Stimulus intensity was varied in 13 logarithmic stimulus luminance steps between 0.04 and 160 cd/m2 at a background luminance of 10 cd/m2. Goldmann size III (25,7’) stimuli were presented 20 times each in three locations (-6.1°,-3.5°), (0°,7°), (6.1°,-3.5°), and two times each at a reference stimulus location (0°,0°) as an additional fixation incentive. An increased rate of false positive and false negative catch trials was implemented (25% each). Each examination included 1,612 stimuli in about 45 minutes altogether.

Results Sufficient data were obtained from 48 subjects (18 m, 30 w, age 22–78 years, median 47 years, distributed equally among three age groups: 21–40, 41–60, 61–80 years, 24 dominant and non-dominant eyes each) were examined. The data set was evaluated in global and individual form for the following parameters: blink rate, pupil diameter, height of the palpebral fissure, heart rate, response time and their respective variabilities. For global analysis, the variability of the palpebral fissure height (PFV) and response time (RTV) showed the highest agreement indices (AI). Median/maximum values were: AIPFV = 0.09/0.52, AIRTV = 0.14/0.47. For the individual analysis, the height of the palpebral fissure (PF) and the response time variability (RTV) showed the best Spearman correlation coefficients (median/maximum values): ρPF = -0.05/-0.82, ρRTV = 0.27/0.61.

Conclusions Response time variability and the height of the palpebral fissure were identified as the most promising and valid parameters in assessing and quantifying vigilance. An increased number of catch trials turned out as an excellent validation tool for the assessment of failures resulting from reduced vigilance with high temporal resolution.