98th Annual Meeting DOG 2000

V 151

On the reproducibility of multifocal ERG recordings

T. Meigen, A. Friedrich

Introduction: Multifocal electroretinogram recordings (MF-ERG) can be used to detect a local dysfunction of the retina. However, this requires highly reproducible MF-ERG traces for all test regions, even if the stimulus area for each test region and the corrsponding signal-to-noise of the MF-ERG is smaller than under fullfield recording conditions. In this study we tested both the intrasessional and intersessional reproducibility of MF-ERG amplitudes.

Methods: For each of the 6 visually normal subjects we performed two recording sessions at the same time of day on 2 different days. MF-ERGs to a stimulus field of 61 hexagonal regions (VERIS system) were recorded from one eye with a dilated pupil using a DTL electrode. Within each session we performed 2 short (3.5 min) and 2 long recordings (7 min). The relative coefficient of variation RCV = = was used to quantify the retest variability of two amplitudes and . An ANOVA was performed to estimate the effect of the factors (A) session (intrasessional vs. intersessional), (B) recording duration (3.5 min vs. 7min), and (C) trace type (hexagon trace vs. VERIS ring-average) on RCV.

Results: RCV was 6.6±3.9% (mean±SD) when averaged across all recording conditions and all subjects. The ANOVA showed a significant difference (p=0.018) between hexagon traces (RCV=7.7±3.3%) and VERIS ring-averages (RCV=5.4±4.2%) for the factor (C). Another significant effect (p=0.016) occurred for the interaction of the factors (A) and (B), reflecting a lower intrasessional variability for the longer recording duration (RCV=4.5±2.6%) and a higher intersessional variability for the shorter recording duration (RCV=7.9±4.2%). All other factors and interactions were non-significant (p>0.05).

Discussion: MF-ERGs can be recorded with a high degree of reproducibility, even for short recording durations and single hexagon traces. As the factor (A) did not show a significant effect, the new placement of the DTL electrode in the second session does not increase the retest variability compared to a second recording within the same session. Future studies will test whether a similar quality can be achieved under conditions of pathologically reduced amplitudes.

Univ.-Augenklinik Würzburg, Josef-Schneider-Str. 11, 97080 Würzburg