Measuring differences in the ERG in myopia using the RETeval device with skin electrodes

Abstract

Introduction: Previous research suggests that the electroretinogram (ERG) is reduced and delayed in non-pathological myopia. However, the invasive nature of the electrode and cumbersome equipment required has prevented the widescale uptake of ERG measures. This study investigated whether previously reported changes to the ERG response in myopia are also observable when measured using non-invasive skin electrodes and a hand-held ERG device.

Method: Monocular flash ERGs were measured using the RETeval® device according to the 'ISCEV 6 Step Dark First cd' protocol in 46 participants with non-pathological myopia (spherical equivalent refraction [SER] -0.50 to -11.25 D, median -3.75 D, median axial length [AL] 25.4 mm) and 47 non-myopic controls (SER +2.00 to -0.25 D, median +1.00 D, median AL 23.6 mm). Measures were performed under pupil mydriasis with Sensor Strip skin electrodes.

Results: The median implicit time for all dark-adapted (DA) components was longer among myopes. Following Holm-Bonferroni correction, this difference reached statistical significance (p < 0.05) for the DA 3.0 A-wave, DA 10.0 A-wave and B-waves, and DA Oscillatory potentials 1 and 2. There were no significant differences between median light-adapted (LA) implicit times nor response amplitudes between refractive groups. For all DA components, there was a significant, positive correlation between AL and implicit time (all p < 0.05).

Conclusions: The RETeval®, used with skin electrodes, did not detect the reduction in ERG amplitude reported in myopic eyes using traditional ERG setups, potentially due to high inter-subject variability and/or anatomical confounders associated with the use of a skin electrode. The RETeval® with skin electrodes did detect subtle delays to DA implicit times previously reported in myopia, with a positive relationship observed between AL and implicit time for all DA components. In contrast, no significant differences were observed for LA implicit times, which may indicate underlying differences in the dark-adaptation process and/or scotopic visual pathways in myopia.

Keywords: RETeval; electrophysiology; myopia; visual function.

FIGURE 1

Histograms showing the distribution of axial length and spherical equivalent refractive error (SER) in the study cohort.

FIGURE 2

Histograms showing the distribution of age in the myopic and non‐myopic groups.

FIGURE 3

Examples of two repeated dark‐adapted (DA) 3.0 oscillatory potential waveforms for one participant. Note the large differences in reported implicit times/amplitudes between the two waveforms, despite the waveforms overlapping in the diagram. This is because the first cursor (black dot) has been correctly positioned on OP‐1, the first major peak, for waveform one (bold green line), but incorrectly positioned before OP‐1 for waveform two (fine orange line). If the waveform two results are shifted ‘one OP to the left’, then the agreement with waveform one is improved. For clarity, the four main OP wavelets have been labelled. OP, oscillatory potentials.

FIGURE 4

Electroretinogram (ERG) implicit times for each step of the six‐step International Society for Clinical Electrophysiology of Vision (ISCEV) protocol in myopes and controls. Individual data points are shown next to boxplots. The ‘*’ denotes significance for Mann–Whitney U‐test at p < 0.05 with Holm‐Bonferroni correction applied. DA, dark adapted; LA, light adapted; OP, oscillatory potentials.

FIGURE 5

The relationship between axial length and implicit time. A Passing Bablok regression line is plotted (coloured dashed lines) when requirements are met, including a significant positive coefficient with Kendall's tau correlation. p‐Values stated in the figure are post Holm‐Bonferroni correction. DA, dark adapted; LA, light adapted; OP, oscillatory potentials.

FIGURE 6

The relationship between axial length and amplitude. A Passing Bablok regression line is plotted (coloured dashed line) when requirements are met, including a significant positive coefficient with Kendall's tau correlation. p‐Values stated in the figure are post Holm‐Bonferroni correction. DA, dark adapted; LA, light adapted; OP, oscillatory potentials.