Abstract
Abstract: :
Purpose: Oscillatory components (OPs) of the electroretinogram (ERG) are generated within the inner retina and are sensitive to diabetes, diabetic retinopathy and other diseases. However, studying them locally and mapping their retinal topographies with the multifocal ERG (mfERG) has been problematic due to their small amplitudes. We developed and tested a technique to enhance the OPs in mfERG recordings (mfOPs) in order to facilitate their local measurement and mapping. Methods: MfERGs were recorded from one eye of each of 13 normal subjects in 7.5 min sessions using dilated pupils, a 75 Hz video frame rate and 10-300 Hz pre-amplifier filtering. 103 areas of the central 45 degrees were stimulated by 100 cd/m^2 pseudorandom flashes separated by 3 dark video frames. In each eye, 34 average waveforms were produced by combining 102 of the areas into groups of 3. These plus the central area produced 35 mfERGs from contiguous regions. From each region a first order (K1), induced first order (K1i), and second order (K2) response component was extracted. These were filtered digitally to separate the mfOPs (> 90 Hz) and the residual low frequency (LF) response contributions. We measured root mean square (RMS) amplitudes and signal-to-noise ratios (SNR) of the mfOPs and LFs. Results: Most of the "noise" is contained in the LFs; it is separated from the mfOPs by the digital filtering. At all 35 retinal locations in each tested eye, the K1i and K2 mfOP waveforms are virtually identical, indicating negligible lateral interactions among retinal locations under our recording conditions. K1 mfOP waveforms differ substantially from those of K1i and K2, however. Combining K1i and K2 at each location increases the SNR of the mfOPs by approximately 40% compared to their original values, an improvement approximately equal to increasing the recording time by twice the original length. These enhanced mfOPs change systematically in waveform across the retina. Their amplitudes are greater in the temporal retina than in the nasal retina and largest around 10 degrees of eccentricity, as previously reported for un-enhanced mfOPs (Wu & Sutter, 1995; Bearse, Shimada & Sutter, 2000). Conclusions: This technique of combining and digitally filtering the induced first order and second order components increases the SNR of the mfOPs by an amount equal to doubling the recording time. This facilitates the local isolation and mapping of these inner retinal contributions to the human mfERG. We are currently assessing the utility of this technique when applied to the study of diabetic retinal dysfunction.
Keywords: electroretinography: non-clinical • electroretinography: clinical • retina: proximal(bipolar, amacrine, and gangli