PERG intrinsic variability across dynamic range

Abstract

Purpose: To quantify intrinsic variability of the Steady-State Pattern Electroretinogram (SS-PERG) over a range of amplitudes spanning the dynamic range of the response.

Methods: SS-PERGs were recorded from each eye of 10 normal subjects (mean age 36 ±9 years) with skin electrodes over the lower eyelids in response to LED-generated black-white bars of 0.67 deg reversing 15.63 times/s. The new protocol (Miami-PERG-) acquired 16 successive samples of 64 epochs each (1024 epochs over 2.5 minutes for the SS-PERG) to measure variability of amplitude, phase and noise. The protocol was repeated with the stimulus viewed through a series of Bangerter Occlusion Foils (BOF 0.6, 0.4, 0.2) to progressively degrade the spatial contrast of the stimulus and generate a wide range of SS-PERG amplitudes.

Results: With increasing BOF optical penalization, the mean baseline SS-PERG amplitude of 1158 ±437nV progressively diminished to reach 253 ±135 nV with BOF 0.2, which was still significantly (P<0.01) above the mean noise level (59 ±26nV). Without BOFs the total amplitude variance of samples included a noise component (51 ±23%), a progressive decline of 222±166 nV (adaptation: 29±16% of total variance), and a residual variance (fluctuation: 15±23%). With increasing BOF penalization, the adaptation component progressively diminished, while the noise component increased to account for the total SS-PERG variability with BOF 0.2. With increasing BOF penalization, the SS-PERG phase progressively changed from 66 ±9 deg to 103±25 deg (estimated latency shortening of 6.5 ms). The phase variability (SD) progressively increased from 10 ±4 deg to 47 ±27 deg.

Conclusions: The intrinsic variability of SS-PERG includes components (adaptation, fluctuation) that are not explained by the uncorrelated noise. Amplitude adaptation and fluctuation are prominent when the SS-PERG signal has high amplitude and diminish with reduction of SS-PERG signal, when the response variability becomes dominated by noise. Phase fluctuation is small when the SS-PERG signal has high amplitude and increases several folds when the response has low amplitude. Results are relevant to better understanding of physiological processes occurring in the inner retina under sustained stimulation, which may be altered in disease. Quantification of different sources of SS-PERG intrinsic variability will help to improve the clinical significance of the PERG technique.