Visual acuity measured at various chart distances versus defocus curves

Abstract

Purpose : To explore potential differences between visual acuity measured at various chart distances versus defocus testing using minus lenses.

Methods : Defocus testing (VA_def) using minus lenses and visual acuity measurements with ETDRS charts at distance (VA_chart), intermediate and near were conducted on 25 subjects with healthy phakic eyes. The mean age was 45.8 ± 14.7 years, range: 22 to 80. In addition, data on visual acuity measured at various chart distances versus defocus testing using minus lenses was collected from publicly available sources of patients implanted with marketed presbyopia-correcting intraocular lenses. Regression analysis was performed on these two datasets. Visual acuity was processed in logMAR units.

Results : For the 25 subjects, visual acuity measured at various chart distances versus defocus testing using minus lenses correlated linearly by VA_chart=-0.004 + 0.813*VA_def (R²=0.811).
The public data from pseudophakic eyes consisted of data from 5 clinical trials, spanning 5 presbyopia correcting lens models and accompanying monofocal control eyes. Patient outcomes were pooled so that the correlation is based on average VAs. These data provided a linear correlation characterized by VA_chart=-0.002 + 0.801*VA_def (R2=0.967). However, the 95% prediction interval was approximately ±0.1 logMAR.
These outcomes indicate that when visual acuity is good (close to 0.0 logMAR), on average both methods provide essentially the same result. When visual acuity is suboptimal (e.g. 0.2 logMAR or worse) VA_def is worse than VA_chart. Prediction intervals were large, which makes the correlation less useful for prediction VA_chart from VA_def and vice versa.

Conclusions : VA measured at various chart distances and VA obtained from defocus testing using minus lenses correlate well on average but suffer a large prediction error, which makes the correlation less useful for prediction VA_chart from VA_def and vice versa.

This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.