Documenta Ophthalmologica 79: 325-336, 1992. 9 1992 Kluwer Academic Publishers. Printed in the Netherlands. Pattern electroretinogram as a function of spatial frequency after retrobulbar optic neuritis BENEDETTO FALSINI, 1 ANTONIO BARDOCCI, 1 SILVANO CERMOLA, 1 VITTORIO PORCIATTI 2 & GIOVANNI PORRELLO 1 1Eye Clinic, Catholic University, Rome, Italy; 2Institute of Neurophysiology, CNR, Pisa, Italy Accepted 12 December 1991 Key words: Optic neuritis, pattern electroretinogram, sinusoidal gratings, spatial frequency Abstract. Steady-state (8-Hz) pattern electroretinograms in response to counterphased sinusoi- dal gratings of variable spatial frequency (0.6-4.8 c/deg) were recorded in 17 patients who had had retrobulbar optic neuritis in one or both eyes (23 eyes with a clinical history of optic neuritis) and in 2l age-matched normal subjects. Amplitude and phase of the Fourier-analyzed pattern electroretinogram second harmonic were measured. The mean pattern electroretinog- ram amplitude of patients was significantly reduced compared with that of controls. Amplitude reductions were more marked at intermediate (1-1.4 c/deg) than at lower or higher spatial frequencies. Therefore, the average amplitude versus spatial frequency response function differed significantly in patients compared with controls, displaying a lowpass instead of a band-pass shape. No significant differences in the mean pattern electroretinogram phase were observed between groups at any spatial frequency. These results indicate spatial frequency- dependent abnormalities in the pattern eleetroretinogram amplitude after optic neuritis, suggesting a specific loss of retinal neurons sensitive to stimuli of intermediate spatial frequencies. Abbreviations: ANOVA--analysis of variance, SD--standard deviation, SEM--standard error of the mean Introduction In recent years, the electroretinogram (ERG) obtained in response to alternating gratings or checkerboards at constant mean luminance (pattern ERG, PERG) has received increasing attention as a specific test for evaluating inner retinal function. Evidence supporting the proximal origin of the PERG comes from experimental and clinical studies showing that the PERG, unlike the luminance-evoked ERG, is altered in ganglion cell dysfunctions [1-4]. The properties of the PERG depend on spatiotemporal stimulus variables [5, 6]. The transient PERG to reversing checkerboards shows little dependence on check size [7]. When the PERG is recorded in response to sinusoidal gratings alternating at 8Hz (16 reversals/s), the 16-Hz (or second harmonic) component of this steady-state response shows