Presented in Fort Lauderdale, Florida April 2004, sponsored by the Association for Research in Vision and Ophthalmology (ARVO). Published in Invest Ophthalmol Vis Sci 44: 2768.

J. Marsack¹, K. Pesudovs¹, R.A. Applegate¹ and L.N. Thibos².

¹ Optometry, University of Houston, Houston, TX
² Optometry, Indiana University, Bloomington, IN

Commercial Relationships: J. Marsack, None; K. Pesudovs, None; R.A. Applegate, University of Texas P; University of Houston P; L.N. Thibos, Indiana University P.

Grant Identification: NEI R01 08520, NEI R01 05109

Abstract

Purpose. To investigate the ability of metrics based on wavefront aberration measures to predict letter visual acuity (VA) at high and low contrast and high and low luminance.

Methods. Shack Hartmann wave aberration data and four different VA measures (high contrast, high and low luminance and low contrast high and low luminance) were recorded in 148 subjects (age ranged from 21.6 to 83.8 years). Wavefront metrics are mathematical functions deriving values from the wavefront shape (WF), the point spread function (PSF), the optical transfer function (OTF) and the RMS values of different groupings of Zernike terms. Stepwise multiple linear regression was performed on 37 of these metrics to determine what combination of metrics best predicted each of the four measured visual acuities. High luminance visual acuities were arithmetically transformed to normalize the distribution, thus facilitating analysis. Pupils were cyclopleged and size was identical for wavefront analysis and VA measurements.

Results. Normalized high contrast, high luminance VA (HCHL) was best predicted by combining a point spread function metric describing area containing 50% of the PSF energy (D₅₀) with the 5^th order RMS (RMS₅) where HCHL = 0.627 + 0.0115* D₅₀ + 0.152*RMS₅, R² = 0.208. Normalized low contrast high luminance VA (LCHL) was best predicted by combining D₅₀ with the visual strehl ratio calculated by the OTF method (VS_OTF) where LCHL = 0.480 + 0.0176* D₅₀ – 0.0356* VVS_OTF, R² = 0.301. High contrast low luminance VA (HCLL) was best predicted by combining average blur strength (B_ave), D₅₀ and total RMS over the whole pupil (RMS_w) where HCLL = 0.0578 + 0.0757* Bave + 0.0351* D₅₀ – 0. RMS_w, R²= 0.322. Low contrast low luminance VA (LCLL) was best predicted by combining area of visibility for MTF (areaMTF) and a pupil fraction metric in which the concentric area RMS is less than /₄ (PFS_c) where LCLL = 0.899 –0.899*areaMTF – 0.192* PFS_c R² = 0.38.

Conclusions. Combinations of wave aberration metrics predict more variance in visual acuity than any single metric. The relationship between metrics of wave aberration and visual acuity is strongest for VA testing with low contrast letters at low luminance. The explanation of less variance for high contrast high luminance VA may indicate that these targets contain redundant information and are therefore unaffected by optical degradation by aberrations in some cases. Supported by: NEI R01 08520 grant to RAA and NEI R01 05109 to LNT.

Key Words: visual acuity