Abstract
Purpose :
Contrast sensitivity measurement with sinusoidal patterns (SinCS) is known to have significant variability and influenced by learning effects. Letter contrast sensitivity (LetCS) using a familiar testing process may be better suited in such a case. The purpose of the study was to validate LetCS produced by an iPad application driving a standard 4K screen compared with SinCS produced by two commercial systems (M&S Technologies (M&S) and Vector Vision (VV) by measuring their repeatability.
Methods :
The study was a single arm, prospective, repeated measure study. The study population (n=20), a young presbyopic phakic population (43.6 ±2.8 years; range 40 to 49years) attended an initial enrolment / test familiarisation visit and two test visits. CS was measured under photopic (85cd/m2) and mesopic (3cd/m2) conditions, at five spatial frequencies (SF) 1.5, 3, 6, 12 and 18cpd. Repeatability was established by comparing the mean and 95% CI of the difference between the measurements made at the two test visits and the variance between the measurements.
Results :
i. Under photopic conditions test variance was lower at all SF for LetCS (Test-retest difference: mean -0.03 to -0.01logCS; 95%CI 0.05 to 0.07logCS) than for SinCS M&S (Test-retest difference: mean -0.14 to -0.04logCs; 95%CI 0.16 to 0.37logCS) (p ≤ 0.001) and VV (Test-retest difference: mean -0.07 to +0.01logCS; 95%CI 0.12 to 0.18logCS) (p = 0.020 to <0.001). ii. under mesopic conditions test variance was lower at all SF for LetCS (Test-retest difference: mean -0.05 to -0.03logCS; 95%CI 0.11 to 0.16logCS) than SinCS M&S (Test-retest difference: mean -0.11 to -0.04logCs; 95%CI 0.21 to 0.34logCS) (p 0.011 to <0.001) and all SF for VV (Test-retest difference: mean -0.08 to -0.01logCS; 95%CI 0.14 to 0.35logCS) (p = 0.004 to 0.001) other than 6cpd (p=0.973).
Conclusions :
Letter contrast sensitivity was shown to be highly repeatable and exhibited significantly lower variance than sinusoidal pattern contrast sensitivity. The implications are that letter contrast sensitivity, needing minimal training, could be well suited for use in clinical studies and its lower variance than sinusoidal contrast sensitivity results in the need for a smaller sample size to achieve the same statistical power than sinusoidal contrast sensitivity.
This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.