Methods of visual acuity determination with the spatial frequency sweep visual evoked potential

Abstract

Purpose: The spatial frequency sweep visual evoked potential (sVEP) is used to rapidly determine visual acuity in children or non-responsive patients. Two techniques have been used to separate signal from noise: (1) the 95% confidence interval for the signal amplitude (95% CI) or (2) the amplitude of a Fourier frequency adjacent to 2x the signal frequency (DFT). The purpose of this study is to determine if there is a significant difference in acuity estimates with these techniques.

Methods: Ten normal subjects (approximately 0.00 logMAR acuity) and 11 patients with decreased visual acuity took part in this project. Stimulus production and data analysis were done with an Enfant 4010 (Neuroscientific Corp). Standard VEP recording techniques were employed. The stimulus was a horizontal-oriented, sine wave grating that swept up the spatial frequency spectrum (contrast 80%, temporal reversal rate 7.5 Hz). Sweeps were repeated until the confidence intervals for the data were no longer decreasing. The Bailey Lovie logMAR chart was used to determine visual acuity. A line was fit to the high spatial frequency data using either the 95% CI or the DFT as the noise estimate. By using these linear equations, acuity estimates were obtained at 0, 1, and 2 microV signal amplitudes.

Results: The average logMAR acuity for the subjects with normal acuity was -0.06 +/- 0.070 (SD). The sVEP acuity estimates were 0.08 +/- 0.098, 0.18 +/- 0.092, and 0.33 +/- 0.195 (0, 1, and 2 microV extrapolations) with the 95% CI used as noise and 0.07 +/- 0.100, 0.18 +/- 0.103, and 0.33 +/- 0.202 (0, 1, and 2 microV extrapolations) with the DFT used as noise. By using the average noise from the Fourier frequency as the extrapolation level, the acuity was 0.10 +/- 0.098 logMAR. The average logMAR acuity for the subjects with decreased visual acuity was 0.67 +/- 0.306 (SD). The sVEP acuity estimates were 0.53 +/- 0.175, 0.66 +/- 0.171, and 0.88 +/- 0.295 (0, 1, and 2 microV extrapolations) with the 95% CI used as noise and 0.53 +/- 0.179, 0.65 +/- 0.176, and 0.86 +/- 0.268 (0, 1, and 2 microV extrapolations) with the DFT used as noise. By using the average noise from the Fourier frequency as the extrapolation level, the acuity was 0.57 +/- 0.186 logMAR. No significant difference was found between the two acuity estimate techniques for all of the subjects (repeated measures ANOVA, p = 0.16, F20 = 2.131). The sVEP estimates of acuity to the 0 microV and noise levels were not significantly different from the logMAR acuity (paired t-test, all p values > 0.05).

Conclusions: The results indicate that the sVEP acuity does not depend on the noise estimation technique. In agreement with prior studies, the sVEP acuity underestimates the logMAR acuity in normally sighted individuals by about an octave.