Abstract
Purpose :
Accommodation in anisometropia may respond consensually and produce a yoked accommodative response, or respond independently leading to asymmetrical accommodation responses (anisoaccommodation). A prospective, randomized study was conducted to examine the effect of conflicting accommodative stimuli in anisometropic individuals (inter-ocular difference ≥ 1.5 D) and to assess the effects of viewing (binocular or monocular), anisometropia level, test distance and time on the patterns of the accommodative response.
Methods :
Dynamic accommodation responses were measured in 11 young, anisometropic subjects (including 3 anisometropic amblyopes) with the Grand Seiko WAM 500 Autorefractor. In the 2 sessions, the viewing condition (monocular or binocular; direct or consensual), the testing distance (400 cm, 38.6 cm and 20.5 cm) and anisometropia level (corrected or uncorrected) were randomized for each subject. In each dynamic trial, accommodation at the 5th ±0.5 second and 20th ±0.5 second was analysed to understand the patterns of the accommodative response. A within-subjects analysis of variance (ANOVA) assessed the factors in the model.
Results :
A significant effect of target distance, level of anisometropia and response pattern was found in the group (p < 0.0001) for both right and left eyes. Time as a factor was not statistically significant. Post hoc Dunnett analysis showed that the patterns (or strategies) of greatest accommodation effort, average accommodative effort, ocular dominance driven, and complete anisoaccommodation were not significantly different from the actual binocular responses. The mean anisoaccommodative gain after regression analysis was −0.027 (p = 0.07), which was not statistically significant.
Conclusions :
Natural anisometropes did not demonstrate an ability to anisoaccommodate to anisoaccommodative stimuli. Time as a factor was not statistically significant. The greatest accommodative effort, average accommodation, ocular dominance driven response, and complete anisoaccommodation models were not significantly different from actual binocular responses. Some hypothesized patterns may have produced a mixture of good and poor predictions, thus raising variance and lowering statistical power. Alternatively, it may be that four subgroups of individuals follow each pattern, or that individuals switch between the patterns over time.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.