To ascertain whether there is a correlation between objective measures of pre-lens tear film
surface quality (TFSQ) dynamics and visual quality responses.
Thirty one subjects wore a pair of silicone hydrogel lenses for 1 hour before measurements. Acquisition of TFSQ dynamics was conducted for each subject in suppressed blinking conditions. A Placido disk videokeratoscope was fitted with an internal target and a video streaming output. Subjects were asked to focus on the target and indicate the just noticeable decrease in visual quality. The recorded videos and the subjective responses were saved for further analysis, for which a new videokeratoscopy image processing technique, based on calculating a texture feature (homogeneity), was developed. The estimated times of lens dewetting were then associated with the subjective responses.
The dewetting point of the lens was identified in 82% of recordings. The group median time of subjective response was 3.5 s post-dewetting point, which corresponded to about 4.3% decrease in TFSQ. In the remaining measurements, either no substantial change in TFSQ or changes related to post-blink slow movement of the lipid layer were observed. The former was not associated with any noticeable change in subjective visual quality whereas the latter usually created visually identifiable temporal disturbance resulting in a subjective response. A small but statistically significant correlation (r2
= 0.32, p < 0.001) was found between objectively estimated point of dewetting and the subjective response, corresponding to just noticeable change in visual quality.
The new technique proved to be sufficiently sensitive to identify the distinct two-phase process of pre-lens TFSQ dynamics. Changes in objectively determined pre-lens TFSQ can also be identified subjectively, but there exists, in general, a certain lag between the two responses. The estimated times of lens dewetting could be utilized for in vivo assessment of contact lens wetting parameters.