Intra- and inter-examiner reliability of ERGs recorded with two types of skin electrodes

To determine whether the intra-examiner and the inter-examiner reliabilities of the implicit times and amplitudes of the ERGs differed when recorded with two different types of skin electrodes.

Eleven subjects (5 men and 6 women) were studied. The RETeval system was used to record ERGs, and the Sensor Strip of the RETeval system or the Red Dot electrocardiography electrodes (Red Dot) were used to record the ERGs. Flicker ERGs were recorded with a natural pupil. To test the intra-examiner reliability, one examiner positioned electrodes and recorded the ERGs three times on three different days from each subject. For the inter-examiner reliability, three different examiners recorded the ERGs from the 11 subjects. The implicit times and amplitudes of the flicker ERGs were analyzed using one-way intraclass correlation coefficients (ICC (1,1)) to assess intra-examiner variability and using a two-way intraclass correlation coefficient (ICC (2,1)) for inter-examiner variability.

For implicit times both the ICC (1,1) and ICC (2,1) were > 0.8 for both electrode types. For the amplitude, ICC (1,1) was > 0.8 for both electrodes, but the ICC (2,1) was lower at 0.474 for the Sensor Strip and 0.672 for the Red Dot electrodes. After Benjamini-Hochberg adjustments, the amplitudes did not differ between electrodes in either setting; implicit time was slightly shorter with the Red Dot than with Sensor Strip in the three-examiner setting (BH-adjusted P = 0.032; mean difference, 0.13 ms).

The intra-examiner reliability was high for both types of electrodes, but the inter-examiner reliability indicated that the amplitudes of the ERGs tended to be less reliable for both types of electrodes. We conclude that standardizing the electrode placement and ensuring precise applications may help improve inter-examiner reproducibility, especially for the amplitude measurements.