The incidence of negative electroretinograms in a healthy adult cohort

Abstract

Purpose : A negative dark-adapted electroretinogram (ERG) waveform is diagnostically significant. Occasionally, it can be an unexpected finding, raising the question as to whether a small proportion of healthy adults might produce negative ERGs. We investigated the incidence of electronegative responses to standard dark-adapted stimuli in a cohort of largely healthy adult volunteers.

Methods : ERG waveforms obtained as part of a previous prospective study investigating heritability of retinal response parameters were analysed. Participants were 211 volunteers recruited from the TwinsUK cohort, who underwent ERG testing according to international standard protocols. Pupils were pharmacologically dilated. Responses were recorded using a conductive fibre electrode placed in the lower conjunctival fornix. Participants underwent 20 min dark adaptation prior to delivery of standard scotopic stimuli. A-wave and b-wave amplitudes were extracted for each subject from dark-adapted responses to 3.0 cd m-2 s (“standard”) and 10.0 cd m-2 s (“brighter”) white flashes. The b:a ratio was calculated. Proportions of responses with b:a ratios less than 1 were noted.

Results : Recordings from both eyes of all participants were included. Mean (SD) age was 62.4 (11.4) years. 93% were female. Mean (SD) b:a ratios for right and left eyes respectively were 1.86 (0.33) and 1.81 (0.29) for the standard flash, and 1.62 (0.25) and 1.58 (0.23) for the brighter flash. The average b:ratio was lower for the brighter flash (p<0.0001). Minimum ratios were 1.12 and 1.13 for the standard flash, and 1.15 and 1.05 for the brighter flash, for right and left eyes respectively. No waveforms were electronegative.

Conclusions : Of 422 largely healthy eyes, no waveforms were electronegative. The findings support the notion that electronegative waveforms (in response to the standard flash strengths and in the recording conditions of the present study) are abnormal, and should prompt further investigation.

This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.