Purpose

The peak time of the electroretinogram (ERG) response to flicker is a sensitive measure of human panretinal cone system function in health and disease. Here, we analysed responses in over 2000 healthy adult twins, with the aim of exploring associations with age, sex and ethnicity.

 

Methods

Participants were recruited from the TwinsUK cohort and underwent non-mydriatic light-adapted ERG recordings using a portable device (RETeval system, LKC Technologies) in conjunction with specialised skin electrodes. The study had Research Ethics Committee approval and conformed to the tenets of the Declaration of Helsinki. The device delivered white full-field flickering stimuli (28.3 Hz) on a white background. Pupil diameter was automatically measured and adjusted for, such that the retinal illuminance of stimuli (85 Td.s) and background (850 Td) were equivalent to that delivered by international standard stimuli through a dilated pupil. Flicker ERG peak times were averaged across both eyes for each participant. Participants were excluded where reliable recordings were unavailable from either eye or if the interocular difference in peak time exceeded 2 ms. When both members of a twin pair were included, peak times were averaged for both twins (to avoid confounding due to intrapair correlation). Correlation with age was quantified and comparisons (age-adjusted where appropriate) were made between sexes and ethnicities.

 

Results

Recordings were analysed from 2395 participants (83% female, 94% of white ethnicity), including 852 full twin pairs. Mean (SD) age was 55.7 (16.3) years; median age was 60. Mean (SD) flicker ERG peak time was 25.6 (1.2) ms; the median was 25.4 ms. Peak times showed a highly significant (p<0.0001) positive correlation with age; the Spearman correlation coefficient was 0.52. The relationship with age was non-linear, such that the increase in peak time per year was greater for older age groups. Male and female participants had similar mean ages, and the difference between averaged peak times across sexes was not significant (p=0.059). Where ethnicity was recorded, 806 pairs identified as white, 14 pairs as black, and 12 pairs as Asian. Age-adjusted comparisons revealed the following significant differences: compared with black participants mean peak time was shorter for white participants (p=0.000086) and shorter for Asian participants (p=0.0097). Other comparisons did not reach significance.

 

Conclusions

The highly significant association between ERG flicker peak times and age suggests possible slowing of retinal processing with age. The relationship was not linear. We also found a difference between ethnicities that invites further investigation, and suggests potential importance of development of ethnicity-specific reference ranges.