ERG shrinks by 10% when reducing dark adaptation time to 10 min, but only for weak flashes

Abstract

Purpose: To compare dark-adapted (DA) ERG between 10, 15 and 20 min of dark adaptation (DA).

Methods: In a counterbalanced random block design, 40 healthy adult subjects were dark-adapted for 10, 15 or 20 min before we recorded ERGs to nine flash strengths from 0.001 to 10.0 cd s/m² (dilated pupils) with a DTL-like electrode. Before and between sessions, the room was lit. Apart from choosing a wider range of stimulus strengths, and adding shorter DA times, the recordings fully complied with the ISCEV ERG Standard, namely using corneal electrodes, mydriasis and a standard DA sequence.

Results: The a-wave amplitude was not affected by any adaptation condition. For the b-wave amplitude, effects of reduced DA time are stronger for weaker flashes: Reducing DA from 20 to 10 min had no measurable effect on the DA 3 ERG, but reduced the DA 0.01 b-wave significantly (p < 0.0001) to 87 ± 2% (mean ± SEM). The DA 0.001 b-wave (not part of the ISCEV ERG Standard) was more affected (down to 72 ± 4%). There was a small, but significant, increase, only for weak flashes, in a- and b-wave peak times for 20 compared to 10-min dark adaptation time.

Conclusion: Reducing dark adaptation time from 20 to 10 min in normal participants has no effect on the ISCEV DA 3 and DA 10 ERG. The reduction in DA 0.01 ERGs to 87 ± 2% agrees with Hamilton and Graham (Doc Ophthalmol 133:11-19, 2016. https://doi.org/10.1007/s10633-016-9554-x ) who found 90 ± 2% and with Asakawa et al. (Doc Ophthalmol 139:33-44, 2019. https://doi.org/10.1007/s10633-019-09693-8 ) who found 83%. Pending verification in pathophysiological states, the current results suggest that one might be able to correct for the 10% amplitude loss when gaining 10 min through shortened DA.

Keywords: Dark adaptation; Dark-adapted ERG; ISCEV ERG Standard; Time.

Fig. 1

Raw ERG traces from a representative participant in this experiment. Flash strength increases from top (0.001) to bottom (10 cd s/m²). Traces represent individual takes without averaging. Peaks settings were manually adjusted; some traces were not included (e.g., fifth from top, only two traces were selected as indicated by the missing ticks at the lower one of the three takes)

Fig. 2

V-log-I curves (amplitude versus log(flash strength)) for all participants (indicated by color) and conditions, a-wave top, b-wave bottom. AT10, AT15 and AT20 indicate the dark adaptation times in minutes. Some outliers pop out visually, e.g., the yellowish trace at top. The a-wave shows little saturation, the b-wave more so, frequently with a “dip” at 0.1–0.3 cd s/m². From this figure, no obvious effect of dark adaptation can be recognized

Fig. 3

a-wave amplitude versus flash strength. For each flash strength, the three boxplots correspond to the three dark adaptation times (AT = 10, 15 and 20 min). The notches in the boxplots delineate the 95% confidence interval for the median, enabling “at-a-glance” statistical interpretation. Amplitudes increase with increasing flash strength (of course), but do not differ significantly between the dark adaptation conditions

Fig. 4

b-wave amplitude versus dark adaptation times (AT = 10, 15 and 20 min) for four different flash strengths (0.001, 0.01, 3 and 10 cd s/m²); graph arranged after ([2], their Fig. 4), individually normalized to the AT20 condition. Individual data points overlay the boxplots. For AT20 (= 100%), these form a line that may serve as a visual reference. The median percentage values for the AT10 condition are indicated at bottom left of each panel. In this rendering, the effect of DA on the b-wave amplitude becomes obvious: For AT = 10 min, the amplitude for weak flashes (0.001 cd s/m², 10 × weaker than the weakest flash from the ISCEV ERG standard) is only 72% of the one for 20 min. For the ISCEV DA 0.01, the effect is 87 ± 2%. For higher flash strengths, the effect diminishes and is no longer recognizable at ISCEV DA 3 or ISCEV DA 10

Fig. 5

Peak times for the a-waves (lower row of data) and b-waves (upper row) versus flash strength. For each flash strength, the three boxplots correspond to the three dark adaptation times (10, 20 and 30 min). The notches in the boxplots delineate the 95% confidence interval for the median, enabling “inference by eye” statistical interpretation [8]. Peak times decrease with increasing flash strength (of course), but to not significantly differ between the dark adaptation conditions