Designing Multifocal Contact Lenses using a Novel Through-Focus Image Quality Metric Highly Correlated with Clinical Visual Acuity

Abstract

Purpose: The aim of this study was to validate the design process of multifocal contact lenses using a new through-focus image quality metric shown to correlate highly with clinical visual acuity.

Methods: Individual eye models were created in commercially available optical design software (Zemax^TM Bellevue, WA) and shown to correlate highly with baseline clinical through-focus visual acuity (R² = 0.85). For validation of predictability, 64 advanced presbyopic eye models were used to predict through-focus image quality of (5) multifocal contact lens designs. In Zemax^TM, each of the (5) designs were inserted onto the anterior surface of the cornea and an over-refraction was done for each eye. Defocus was minimized and the closest 0.25D lens was used to obtain peak image quality at distance. Through-focus predicted logMAR was determined for each eye model with each of the (5) multifocal lenses in place. To compare directly with clinical measurements of visual acuity, image quality with the contact lens on eye was subtracted from baseline values representing the individual eye's acuity without any corrective lens in place. This gives predicted normalized logMAR acuity. The same (7) object distances were used for computer modeling as in the actual clinical study. Normalized visual acuity for 24 mature presbyopes, tested monocularly, was measured with each of the (5) manufactured multifocals on-eye. Correlation of clinical and predicted through-focus normalized acuity was determined.

Results: At all object distances, each multifocal had less than one line difference on average between predicted and clinical normalized logMAR acuity. The Zemax^TM models with predicted logMAR had high correlation with through-focus visual acuity from the clinic (R² between 0.90 and 0.97) for all multifocal lens designs.

Conclusions: When multifocal contact lens designs were modeled on individual computer eyes and normalized predicted logMAR was used to determine how lenses would perform clinically, a high correlation was found between the computer database and average clinical acuity results with all (5) multifocal contact lenses. This design process is highly effective in predicting on-eye performance. With this high level of predictability more design options can be explored in the computer to optimize though-focus performance before a lens is manufactured and tested clinically.

Keywords: 477 contact lens • 653 presbyopia • 473 computational modeling