Shortened Pattern Electroretinogram Latency and Impaired Autoregulatory Dynamics to Steady-State Stimuli in Patients With Multiple Sclerosis

Background: The steady-state pattern electroretinogram (PERG) is a sensitive measure of retinal ganglion cell (RGC) function that includes within-test progressive changes-adaptation-reflecting RGC autoregulatory dynamics. Comprehensive PERG assessment in patients with multiple sclerosis (MS) (with or without optic neuritis [ON]) may provide unique information about RGC dysfunction and its progression, as well as a comparison between functional loss and structural loss as measured by optical coherence tomography (OCT). The goal of this project was to measure steady-state PERG components and their associations with intraretinal layer thicknesses in MS.

Original Contribution Section Editors: Clare Fraser, MD Susan Mollan, MD Shortened Pattern Electroretinogram Latency and Impaired Autoregulatory Dynamics to Steady-State Stimuli in Patients With Multiple Sclerosis Downloaded from http://journals.lww.com/jneuro-ophthalmology by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 05/04/2022 Hong Jiang, MD, PhD, Giovana R. Gameiro, Huiling Hu, MD, PhD, Pedro F. Monsalve, MD, Chuanchui Dong, PhD, Jeffrey Hernandez, RN, Silvia R. Delgado, MD, Vittorio D. Porciatti, PhD, Jianhua Wang, MD, PhD Background: The steady-state pattern electroretinogram (PERG) is a sensitive measure of retinal ganglion cell (RGC) function that includes within-test progressive changes— adaptation—reflecting RGC autoregulatory dynamics. Comprehensive PERG assessment in patients with multiple sclerosis (MS) (with or without optic neuritis [ON]) may provide unique information about RGC dysfunction and its progression, as well as a comparison between functional loss and structural loss as measured by optical coherence tomography (OCT). The goal of this project was to measure steady-state PERG components and their associations with intraretinal layer thicknesses in MS. Methods: One hundred forty eyes of 70 patients with relapsing-remitting MS and 126 eyes of 63 age- and sexmatched healthy control subjects (HC) were investigated using a new-generation PERG method and ultrahighresolution OCT. Of MS eyes, there were 30 eyes with ON (MSON), 22 non-ON fellow eyes (MSFE), and 88 non-ON MS eyes (MSNON). PERG amplitude, phase (latency), and adaptation of amplitude and phase were measured and correlated with OCT-determined thicknesses of intraretinal layers. Results: The average PERG amplitude in MSON eyes was significantly lower than MSFE (P = 0.007), MSNON (P = 0.002), and HC (P , 0.001). The PERG amplitude in MSFE eyes was also significantly lower than HC (P = 0.039). The PERG latency in MSON eyes was significantly shorter than in Department of Ophthalmology (HJ, GRG, HH, PFM, VDP, JW), Bascom Palmer Eye Institute, University of Miami, Miller School of Medicine, Miami, Florida; Department of Neurology (HJ, CD, JH, SRD), University of Miami, Miller School of Medicine, Miami, Florida; and Department of Ophthalmology (HH), Shenzhen Key Laboratory of Ophthalmology, Shenzhen Eye Hospital, Jinan University, Shenzhen, China. Supported by the National Multiple Sclerosis Society (RG-150604890), NIH Center Grant P30 EY014801, and a grant from Research to Prevent Blindness (RPB). The authors report no conflicts of interest. Address correspondence to Hong Jiang, MD, PhD, Bascom Palmer Eye Institute, University of Miami, Miller School of Medicine, 1638 NW 10th Avenue, McKnight Building, Room 202A, Miami, FL 33136; E-mail: [e-mail] 60 MSFE (P = 0.001), MSNON (P = 0.002), and HC (P , 0.001). The PERG latency in MSFE (P = 0.007) and MSNON (P = 0.002) was significantly shorter than in HC. However, no significant differences were found between MSFE and MSNON (P . 0.05). PERG adaptation of amplitude in MSON was significantly lower than that in MSNON (P = 0.039) and HC (P = 0.037). Both the amplitude and latency in the MS eyes were significantly correlated with the thicknesses of the macular retinal nerve fiber layer (mRNFL) and ganglion cell-inner plexiform layer (GCIPL). Conclusions: Shortened PERG latency and impaired autoregulatory dynamics occurred in MS, suggesting preferential dysfunction of small, slower RGC axons and decreased ability of RGC to autoregulate their gain in response to PERG stimulus. The established relations of PERG measurements with intraretinal thickness measurements suggested that PERG losses were primarily associated with GCIPL and mRNFL thinning. Journal of Neuro-Ophthalmology 2021;41:60–68 doi: 10.1097/WNO.0000000000000894 © 2020 by North American Neuro-Ophthalmology Society M ultiple sclerosis (MS) is a chronic inflammatory neurodegenerative disorder, the second major cause of disability in young and middle-aged adults. Currently available treatments generally target the autoimmune aspect of MS and vary in effectiveness among individuals (1). The major drawback in monitoring the therapeutic efficacy is lack of sensitive biomarkers. The visual pathway is commonly affected in MS, not only in eyes with a history of optic neuritis (ON) but also in eyes without ON (2,3). Subclinical optic neuropathy exists ubiquitously in MS. At postmortem, up to 99% of patients with MS exhibit demyelinating plaques in their optic nerves (2). Using the eye as a model to monitor the disease progression and therapeutic efficacy has been a long entertained idea, especially from those eyes that have never had ON. Jiang et al: J Neuro-Ophthalmol 2021; 41: 60-68 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution The thinning of the ganglion cell-inner plexiform layer (GCIPL) indicates neural loss that has been used as a marker of neurodegeneration in MS clinical trials. However, it seems not sensitive enough to be used to monitor therapeutic efficacy because the longitudinal follow-up of GCIPL thickness showed very small yearly alterations (4). Furthermore, the thickness measurement of the neural layers does not provide functional information of the retinal ganglion cells (RGCs). Pattern electroretinogram (PERG), is a noninvasive, objective technique for detecting RGC function. In the mouse model, crushing or sectioning the optic nerve results in RGC depletion and PERG abolition (5). Similar loss of PERG was observed in patients with Leber’s optic neuropathy and advanced glaucoma (6–8). Hence, the PERG has been used to gauge the RGC dysfunction in optic neuropathies such as glaucoma (9) and MS (10–12). It has been shown that RGC dysfunction may precede loss of RNFL thickness (13) and be reversible (14), which suggests that the combined PERG and thickness assessment can be used to determine the window of time for preventing RGC loss in optic neuropathies (7). The introduction of a new generation of PERG instrument (8) using a visual display with high luminance and temporal resolution allows recording of responses with higher signal-to-noise ratio and latency precision as well as assessment of within-test nonstationarity (adaptation). It is believed to represent autoregulatory changes of RGC response to high-contrast, steady-state pattern reversals. PERG amplitude adaptation may be impaired in neuropathies such as nonarteritic ischemic optic neuropathy and early stages of glaucoma and ON (8,15,16). In previous studies, the thickness of peripapillary RNFL (pRNFL) (12,17,18) or GCIPL (12) defined by the Zeiss elliptical partition was studied. Because multiple intraretinal layers may be affected in MS (17,18), studying the associations between PERG and various intraretinal layer thicknesses may provide a better understanding of the sources of PERG generators and determining the function–structure association in MS. The goal of this project was to measure steady-state PERG components and their associations with intraretinal layer thicknesses in patients with relapsingremitting MS (RRMS). METHODS The institutional review board for human research at the University of Miami approved the research protocol, and each study participant signed an informed consent form after the study information was explained. MS patients were diagnosed by their treating neurologists based on the 2010 Revised McDonald Criteria, and recruited from an ongoing observational cohort study in the Departments of Neurology and Ophthalmology at the University of Miami from July 2015 to May 2018. The patients’ disease activities in the past 2 years before the study enrollment or since the first Jiang et al: J Neuro-Ophthalmol 2021; 41: 60-68 clinical episode (if disease duration was shorter than 2 years) were recorded, which included the number of relapses per year or increased T2 lesions or gadolinium-enhancing (Gd+) lesions as shown on brain MRI. No evidence of disease activity (NEDA-3) was defined as the absence of relapses and Gd+ lesions and stable T2 lesion counts and extended disability status score (EDSS) values. Each patient underwent a complete neurological and ophthalmic examination, including best-corrected visual acuity, intraocular pressure, and slit-lamp biomicroscopy of anterior and posterior segments. Subjects with other ophthalmic or neurological disorders (other than MS), such as glaucoma, diabetic and hypertensive retinopathy, and a refractive error greater than ±6 diopters, were also excluded. All the subjects had best-correct visual acuity 20/20 or better and full confrontational visual field in all eyes (right eye and left eye). Using the 2.5% and 1.25% of the low-contrast letter acuity (LCLA) charts (low-contrast Sloan letter chart; Precision Vision, LaSalle, IL), binocular LCLA was tested, with the best possible correction for refractive errors. LCLA scores were quantified as the number of letters correctly read by the patient (19). A total of 140 eyes of 70 RRMS patients were studied using PERG and ultrahigh-resolution optical coherence tomography (UHR-OCT), and compared with 126 eyes (HC) of 63 age- and sex-matched healthy subjects. Of the MS eyes, there were 30 eyes with ON (MSON), and 22 non-ON fellow eyes (MSFE), in addition to 88 non-ON MS eyes (MSNON) (Table 1). The instrumentation setup of the steady-state PERG (SS-PERG) and main outcome measures have been previously reported (9,20), and also detailed in a recent study (10). Briefly, the subjects had refraction to get the bestcorrected visual acuity for PERG and their pupils were not dilated. A black–white grating generated on a 14 · 14-cm LED display (1.6 cycles/degree, 15.63 reversals/s, 98% contrast, 800 cd/sqm mean luminance, 25° field, Jorvec Corp, Miami, FL) was presented on a monitor placed at 30-cm distance. The PERG was recorded simultaneously from both eyes through skin cup electrodes taped over the lower eyelids and reference electrodes taped on the ipsilateral temples. Artifact-free repetitions (n = 1,024) were averaged. Subaverages (n = 16 samples, 64 repetitions each) were also recorded to assess progressive changes with time (adaptation). Fourier analysis was performed to retrieve the zero-to-peak amplitude (nV) and phase (degrees) of the 15.63-Hz response harmonic (reversal rate). PERG phase was converted to latency using the formula: Latency (milliseconds) = [360 (degrees)–Phase (degrees)] * [1,000 milliseconds/15.63 Hz]/360 (degrees). To calculate withintest PERG adaptation, the first 4 samples (samples 1–4, w0–30 seconds recording) and the last 4 samples (samples 13–16, w100–130 seconds recording) were averaged separately and their difference in amplitude and phase was defined as adaptation of amplitude and phase (8). To 61 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution TABLE 1. Demographic information and clinical manifestations Subjects Age (yrs) Sex (M/F) Eyes EDSS EDSS vision EDSS non vision 2.5% LCLA 1.25% LCLA DD (yrs) NEDA MSNON MSON HC 44 41.1 ± 9.9 7:34 88 1.83 ± 1.59 0.05 ± 0.21 1.85 ± 1.61 51.7 ± 7.7 26.7 ± 8.6 6.9 ± 5.7 31 26* 39.4 ± 11.2 7:19 52 3.19 ± 2.11 0.58 ± 0.76 2.98 ± 2.22 50.2 ± 8.6 24.3 ± 9.9 9.7 ± 7.3 20 63 38.4 ± 10.7 22:43 126 P value .0.05 0.005 ,0.001 0.024 0.415 0.177 0.043 *Four patients with bilateral optic neuritis. DD, disease duration; EDSS, extended disability status score; HC, healthy controls; LCLA, low-contrast letter acuity; MSFE, fellow eye of optic neuritis; MSNON, multiple sclerosis without optic neuritis; MSON, multiple sclerosis with optic neuritis; NEDA, no evidence of disease activity. visualize the combined changes of PERG amplitude and phases in a single plot, Cartesian coordinates were used. The x axis was [amplitude · cos (phase)] and the y axis was [amplitude · sin (phase)]. Mean amplitude of the centroid distribution was the length of the vector and phase was the angle of the vector. Confidence ellipses represent 50% of data (all eyes included). UHR-OCT was used to measure topographic thicknesses (maps) of intraretinal layers (21,22). The custom system is a spectral domain OCT and has been reported in previous studies (21–23). It has an axial resolution of w3 mm in the retina. To analyze the thickness maps of the volumetric data, a commercial segmentation software program (Orion; Voxeleron LLC, Pleasanton, CA) was used to automatically segment up to 6 intraretinal layers for thickness maps (21,22). In this study, a raster scan protocol of 512 (A-scans) · 128 (B-scans) was used to scan an area of 6 · 6 mm centered on the fovea, and 6 intraretinal layers were segmented from the volumetric data set using the Orion software (Fig. 1) (21,22). The segmented intraretinal layers included macular nerve fiber layer (mRNFL), GCIPL, inner nuclear layer (INL), outer plexiform layer (OPL), outer nuclear layer (ONL), and retinal photoreceptor (PR). Also, Zeiss Cirrus OCT (Carl Zeiss Meditec, Dublin, CA) was used to scan the fovea using the 200 · 200 macular cube protocol and the optic nerve head using the 200 · 200 optic disc cube protocol. Zeiss GCIPL and pRNFL were read out from the Zeiss OCT reports segmented using the built-in segmentation software in the Zeiss OCT device. Statistical Analyses All data were analyzed with SPSS (Statistical Package for the Social Sciences, IBM, Armonk, NY, Ver. 25). Continuous variables were presented as the mean ± SD. Analysis of covariance was used to analyze the differences of clinical manifestations (LCVA, EDSS, EDSS vision, and disease 62 duration) with the adjustments of age and sex. Generalized estimating equation (GEE) models were used to count for intercorrelation of eyes within subjects. Eyes (left or right) were set as within-subject variables in GEE models. The measurements of OCT and PERG were dependent variables. In addition, age, sex, and eye were set as covariates. To analyze the combined data of amplitude and phase, multivariate analysis was used. Pearson correlation coefficient was used to determine the relations for independent variables from both right and left eyes. The area under the receiver operating characteristic curve (AUROC) was also calculated to evaluate discrimination powers of independent variables. P , 0.05 was considered statistically significant. RESULTS EDSS and EDSS vision scores in MSON were significantly higher than MSNON (both P , 0.001, Table 2), whereas there were no significant differences of LCLA (2.5%, P = 0.415, 1.25%, P = 0.177) between MSNON and MSON groups. Disease duration of MSON was longer than MSNON (P = 0.043). The average PERG amplitude in MSON eyes was significantly lower than MSFE (P = 0.007), MSNON (P = 0.002), and HC (P , 0.001, Fig. 2). The PERG amplitude in MSFE eyes was also significantly lower than in HC (P = 0.039). The PERG latency in MSON eyes was significantly lower than in MSFE (P = 0.001), MSNON (P = 0.002), and HC (P , 0.001). The PERG latency in MSFE (P = 0.007) and MSNON (P = 0.002) was significantly lower than in HC. However, there were no significant differences of the amplitude and latency between MSFE and MSNON (all P . 0.05). After adjustment of age and sex, PERG measurements were not significantly different between MS patients with and without disease activity (all P . 0.05). PERG adaptation of amplitude in MSON was significantly lower than that in MSNON (P = 0.039, Fig. 2) and HC Jiang et al: J Neuro-Ophthalmol 2021; 41: 60-68 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 1. OCT imaging protocols and methods for analyzing the tomographic thickness of the retina. A. The PERG grating was projected into the macula with an area of 7 · 7 mm (solid yellow line). The 6 · 6 mm scan protocol was used to scan the macula centered on the fovea using UHR-OCT (solid green line) and Zeiss Cirrus OCT. The retina with a diameter of 6 mm (green dotted line) was analyzed from the volumetric dataset obtained using UHR-OCT. The Zeiss elliptical area (black dotted line) was also used to obtain the average thickness of the GCIPL. Also, the 6 · 6 mm scan protocol was used to scan the optic nerve head to obtain the peripapillary RNFL thickness with a diameter of 3.4 mm (solid blue circle) using Zeiss Cirrus OCT. B. Six intraretinal layers were segmented. C. Six thickness maps of these segmented intraretinal layers. GCIPL, ganglion cell-inner plexiform layer; INL, inner nuclear layer; mRNFL, macular retinal nerve fiber layer; OCT, optical coherence tomography; ONL, outer nuclear layer; OPL, outer plexiform layer; PERG, pattern electroretinogram; PR, retinal photoreceptor. Jiang et al: J Neuro-Ophthalmol 2021; 41: 60-68 63 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution TABLE 2. Relations between PERG measurements and intraretinal layer thicknesses Mag (nV) mRNFL GCIPL INL OPL ONL PR Zeiss GCIPL pRNFL Latency r P value R P value 0.36 0.38 0.18 0.09 20.08 0.03 0.30 0.19 ,0.001 ,0.001 0.03 0.28 0.47 0.63 ,0.001 0.04 0.42 0.32 0.08 0.02 20.10 0.05 0.40 0.25 ,0.001 ,0.001 0.38 0.85 0.23 0.69 ,0.001 ,0.001 Bold font indicates significant correlation at levels of 0.05 or 0.01. GCIPL, ganglion cell-inner plexiform layer; INL, inner nuclear layer; mRNFL, macular retinal nerve fiber layer; ONL, outer nuclear layer; OPL, outer plexiform layer; PERG, pattern electroretinogram; PR, retinal photoreceptor; pRNFL, peripapillary RNFL. (P = 0.037). However, there were no significant differences in the PERG phase adaptation among groups (P . 0.05). GEE analysis of PERG amplitude adaptation as the dependent vari- able and mean PERG amplitude, GCIPL thickness, and age as covariates showed that the PERG amplitude adaptation in MSON was smaller than in HC (P = 0.017) and MSNON (P = 0.019), but not in MSFE (P = 0.168). The alteration of combined amplitude and phase was visualized using the Cartesian coordinates plot (Fig. 3), which showed: 1) amplitude tends to decrease and phase tends to rotate counterclockwise (latency shortening) with increasing severity, 2) MSNON and MSFE are virtually identical, and 3) MSON eyes have a much reduced amplitude and anticipated phase compared to controls with little overlap of distributions. Multivariate analyses resulted in significant differences between HC and eyes with MSNON, MSON, and MSFE (all P , 0.001). In addition, there was significant differences between MSNON and eyes with MSNON (P , 0.001) and MSFE (P = 0.026). Thicknesses of mRNFL, GCIPL, Zeiss GCIPL, and pRNFL in MS eyes (MSON, MSFE, and MSNON) were significantly lower than HC eyes (all P , 0.05, Fig. 4). Also, the thickness of these measurements in MSON was FIG. 2. PERG measurements. The average PERG amplitude in MSFE and MSON groups and latency in MSON, MSFE, and MSNON groups were significantly lower compared to HC eyes (P , 0.05). PERG amplitude and latency in MSON eyes were significantly lower than other MS groups (P , 0.05); however, there were no significant differences of the amplitude and latency between MSFE and MSNON (P . 0.05). Amplitude adaptation in different groups of subjects (HC, MSNON, MSFE, and MSON) showed that amplitude adaptation was much reduced in MSON compared to MSNON and HC. However, phase adaptation was not significant among groups. MS, multiple sclerosis; PERG, pattern electroretinogram. 64 Jiang et al: J Neuro-Ophthalmol 2021; 41: 60-68 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 3. Cartesian coordinates plot of combined amplitude and phase. Mean amplitude of the centroid distribution was the length of the vector and phase was the angle of the vector. Confidence ellipses represent 50% of data (all eyes included). Mean amplitude of the centroid distribution is the length of the vector and phase is the angle of the vector. significantly lower compared to other MS eyes (MSFE and MSNON, all P , 0.001). There were no significant differences in these measurements between MSNON and MSFE eyes (P . 0.05). MSON eyes also had a significantly lower thickness of the INL compared to HC (P = 0.019) and MSNON eyes (P = 0.004). OPL thickness of MSON was significantly lower than HC (P = 0.002). By contrast, ONL thickness was significantly higher in MSON than in HC eyes (P = 0.047). Both the amplitude and latency in the MS eyes were significantly correlated with the annular thickness of GCIPL, mRNFL, Zeiss GCIPL, and pRNFL (r ranged from 0.18 to 0.42, P , 0.05, Table 2). Compared to the averaged PERG measurements of the HC eyes, the scatter plots of the amplitude and latency showed the measurements of MSNON and MSFE eyes spread over all quadrants (Fig. 5). By contrast, the PERG measurements of MSON eyes were concentrated in the third quadrant (thinner GCIPL/mRNFL and lower amplitude/ latency). FIG. 4. Intraretinal layer thicknesses. Thicknesses of mRNFL, GCIPL, Zeiss GCIPL, and pRNFL in MS eyes (MSON, MSFE, and MSNON) were significantly lower than that of HC eyes after age and sex adjustment (P , 0.05). Also, the thickness of these measurements in MSON was significantly lower compared to other MS eyes (MSFE and MSNON). There were no significant differences in these measurements between MSNON and MSFE eyes (P . 0.05). MSON eyes also had a significantly lower thickness of the INL compared to HC and MSNON eyes (P , 0.05). By contrast, ONL thickness was significantly higher in MSNON and MSON compared to HC eyes (P , 0.05). GCIPL, ganglion cell-inner plexiform layer; INL, inner nuclear layer; mRNFL, macular retinal nerve fiber layer; MS, multiple sclerosis; ONL, outer nuclear layer; pRNFL, peripapillary RNFL. Jiang et al: J Neuro-Ophthalmol 2021; 41: 60-68 65 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 5. Relations between PERG measurements and thicknesses of GCIPL and mRNFL. Compared to the averaged PERG measurements (dotted purple lines) of the HC eyes, the scatter plots of the amplitude (A) and latency (B) showed the PERG measurements of MSNON and MSFE eyes spread over all quadrants. By contrast, the measurements of MSON eyes concentrated in the third quadrant of the coordinate plane (thinner GCIPL/mRNFL and lower amplitude/latency). GCIPL, ganglion cell-inner plexiform layer; mRNFL, macular retinal nerve fiber layer; PERG, pattern electroretinogram. AUROC analysis indicated amplitude and latency had good discrimination powers (.0.80) in discriminating the MSON eyes from HC eyes (Fig. 6). However, the discrimination powers were not as good as thickness measurements of the RNFL and GCIPL using both UHR-OCT and Zeiss OCT. DISCUSSION Although PERG alterations in MS with or without ON have been previously described (11), this study offers a novel approach for investigating structure–function relationships using PERG and UHR-OCT. It should be considered that PERG outcome measures (amplitude, latency, and adaptation) represent different aspects of RGC function that may or may not be associated with OCT changes. Amplitude reduction may be due to either RGC loss or dysfunction of the surviving RGCs. Changes of PERG latency and amplitude adaptation are associated with residual function of surviving RGC. By contrast, OCT primarily reflects loss of RGC/axons and does not provide information on surviving neurons. Thus, comprehensive PERG assessment provides, at least in part, different and complementary information from that of OCT. Precise measurements of PERG latency were possible because the new generation PERG instrument with LED display used in this study allowed instantaneous pattern reversal over the entire stimulus (24). Previous instruments used standard raster CRT displays where the contrastreversal occurs in a sweeping manner over several milliseconds (25). This study shows that PERG latency was clearly shorter in MS with or without ON. This finding would seem paradoxical because it is well established that visualevoked potential (VEP) is often delayed (26) in MS due to demyelination and associated increase of conduction time in the postretinal visual pathway. As RGC axons are unmyelinated, PERG latency changes require a different explanation from that of VEP. PERG latency shortening in MSON has been modeled in a recent study (10), which suggested 66 relative dysfunction of RGCs with smaller, slower axons and sparing of RGCs with larger, faster axons resulting in an integrated response faster than normal. Shortened latency regardless of ON found in this study provides further evidence of similar pathophysiology. Our hypothesis is based on indirect anatomical and physiological evidence (10), which explains the lower latency due to the difference in latency between small/short axons and longer/slower axons in MSON eyes. Histological studies in MS patients show that smaller optic nerve fibers are differentially lost (27). PERG latency shortening of comparable magnitude to that measured in MSON eyes can be obtained in normal eyes by 1) using a sinusoidal PERG stimulus of low spatial frequency, which preferentially activates larger RGCs (28), and 2) blurring a square-wave PERG stimulus, which removes the contribution of high spatial frequencies at which small RGC respond preferentially. Steady-state PERG adaptation is a conspicuous response feature that is generated in the inner retina because adaptation does not occur in the flicker ERG that reflects outer retina function (29). Also, PERG adaptation is reduced in glaucoma and Leber’s optic neuropathy (6–8), which primarily impair RGC. It has been suggested that PERG adaptation reflects autoregulatory RGC gain changes after prolonged exposure to high-contrast stimuli that also induce functional hyperemia at the level of the optic nerve head (30,31). In this study, amplitude adaptation was significantly decreased in MSON compared to MSNON and HC groups, which may reflect the impaired ability of RGC to autoregulate their response gain under stressful stimulation. Even with the adjustments of PERG amplitude, GCIPL thickness, and age, the outcomes are similar, indicating that the adaptation may not be dependent on the number of ganglion cells left. The PERG amplitude decreased with increasing severity of the condition (HC . MSNON . MSFE . MSON) approximately in parallel with thinning of RGC-relevant Jiang et al: J Neuro-Ophthalmol 2021; 41: 60-68 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 6. The area under the receiver operating characteristics curve (AUROC). The PERG amplitude and latency had good discrimination powers (.0.80) in discriminating MSON eyes from HC eyes (A and B). However, the discrimination powers were not as good as the thickness measurements of the RNFL and GCIPL using both UHR-OCT and Zeiss OCT. GCIPL, ganglion cell-inner plexiform layer; OCT, optical coherence tomography; PERG, pattern electroretinogram. intraretinal layers (mRNFL, CGIPL, Zeiss_CGIPL, and pRNFL), whereas the thickness of preganglionic layers (INL, OPL, ONL, and PR) was relatively unaltered. The correlation coefficient between PERG amplitude and intraretinal thickness was maximal for GCIPL and mRNFL (0.38–0.36), whereas the correlation between PERG latency and intraretinal thickness was maximal for mRNFL and Zeiss CGIPL (0.42–0.40). The correlation between PERG amplitude/latency and outer layers (INL, OPL, ONL, and PR) was close to zero. Although a strict functional–structural correlation was not expected because PERG and OCT reflect at least in part different neuronal conditions as discussed above, the significant correlation for specific RGC and axon layers supports the notion that these layers are the primary generators of the PERG signal. The notion of RGC as primary generators of the PERG signal is based on solid experimental evidence because the PERG is abolished after optic nerve crush or section that selectively damages RGC leaving other neuronal cells and the standard flash ERG unaltered (5). Jiang et al: J Neuro-Ophthalmol 2021; 41: 60-68 CONCLUSIONS Impaired PERG signal occurred in MS regardless of ON, indicating generalized subclinical RGC dysfunction. Of note, PERG latency shortened, suggesting preferential dysfunction of smaller, slower RGC axons. Reduced PERG amplitude adaptation in ON eyes suggested the decreased ability of RGC to autoregulate their gain in response to high-contrast, steady-state pattern stimulus. The correlation between alterations and GCIPL and mRNFL thinning suggested that PERG alterations were primarily associated with RGC and intraretinal axons. STATEMENT OF AUTHORSHIP Category 1: a. Conception and design: H. Jiang, V. D. Prociatti, and J. Wang; b. Acquisition of data: H. Jiang, V. D. Prociatti, J. Wang, G. R. Gameiro, H. Hu, P. F. Monsalve, J. Hernandez, and S. R. Delgado; c. Analysis and interpretation of data: H. Jiang, V. D. Prociatti, J. Wang, G. R. Gameiro, H. Hu, P. F. Monsalve, J. Hernandez, S. R. Delgado, and C. Dong. Category 2: a. Drafting the manuscript: H. Jiang, V. D. Prociatti, J. Wang, G. R. Gameiro, H. Hu, P. F. Monsalve, 67 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution J. Hernandez, S. R. Delgado, and C. Dong; b. Revising it for intellectual content: H. Jiang, V. D. Prociatti, J. Wang, G. R. Gameiro, H. Hu, P. F. Monsalve, J. Hernandez, S. R. Delgado, and C. Dong. Category 3: a. Final approval of the completed manuscript: H. Jiang, V. D. Prociatti, J. Wang, G. R. Gameiro, H. Hu, P. F. Monsalve, J. Hernandez, S. R. Delgado, and C. Dong. 17. 18. REFERENCES 1. Lanzillo R, Cennamo G, Criscuolo C, Carotenuto A, Velotti N, Sparnelli F, Cianflone A, Moccia M, Brescia Morra V. Optical coherence tomography angiography retinal vascular network assessment in multiple sclerosis. Mult Scler. 2018;24:1706–1714. 2. Ikuta F, Zimmerman HM. Distribution of plaques in seventy autopsy cases of multiple sclerosis in the United States. Neurology. 1976;26:26–28. 3. Toussaint D, Périer O, Verstappen A, Bervoets S. Clinicopathological study of the visual pathways, eyes, and cerebral hemispheres in 32 cases of disseminated sclerosis. J Clin Neuroophthalmol. 1983;3:211–220. 4. Nolan RC, Akhand O, Rizzo JR, Galetta SL, Balcer LJ. Evolution of visual outcomes in clinical trials for multiple sclerosis disease-modifying therapies. J Neuroophthalmol. 2018;38:202–209. 5. Porciatti V. The mouse pattern electroretinogram. Doc Ophthalmol. 2007;115:145–153. 6. Guy J, Feuer WJ, Porciatti V, Schiffman J, Abukhalil F, Vandenbroucke R, Rosa PR, Lam BL Retinal ganglion cell dysfunction in asymptomatic G11778A: Leber hereditary optic neuropathy. Invest Ophthalmol Vis Sci. 2014;55:841–848. 7. Porciatti V, Ventura LM. Retinal ganglion cell functional plasticity and optic neuropathy: a comprehensive model. J Neuroophthalmol. 2012;32:354–358. 8. Monsalve P, Triolo G, Toft-Nielsen J, Bohorquez J, Henderson AD, Delgado R, Miskiel E, Ozdamar O, Feuer WJ, Porciatti V. Next generation PERG method: expanding the response dynamic range and capturing response adaptation. Transl Vis Sci Technol. 2017;6:5. 9. Ventura LM, Porciatti V, Ishida K, Feuer WJ, Parrish RK. Pattern electroretinogram abnormality and glaucoma. Ophthalmology. 2005;112:10–19. 10. Monsalve P, Ren S, Jiang H, Wang J, Kostic M, Gordon P, Porciatti V. Retinal ganglion cell function in recovered optic neuritis: faster is not better. Clin Neurophysiol. 2018;129:1813–1818. 11. Rodriguez-Mena D, Almarcegui C, Dolz I, Herrero R, Bambo MP, Fernandez J, Pablo LE, Garcia-Martin E. Electropysiologic evaluation of the visual pathway in patients with multiple sclerosis. J Clin Neurophysiol. 2013;30:376–381. 12. Hokazono K, Raza AS, Oyamada MK, Hood DC, Monteiro ML. Pattern electroretinogram in neuromyelitis optica and multiple sclerosis with or without optic neuritis and its correlation with FD-OCT and perimetry. Doc Ophthalmol. 2013;127:201–215. 13. Banitt MR, Ventura LM, Feuer WJ, Savatovsky E, Luna G, Shif O, Bosse B, Porciatti V. Progressive loss of retinal ganglion cell function precedes structural loss by several years in glaucoma suspects. Invest Ophthalmol Vis Sci. 2013;54:2346–2352. 14. Ventura LM, Porciatti V. Restoration of retinal ganglion cell function in early glaucoma after intraocular pressure reduction: a pilot study. Ophthalmology. 2005;112:20–27. 15. Porciatti V, Bosse B, Parekh PK, Shif OA, Feuer WJ, Ventura LM. Adaptation of the steady-state PERG in early glaucoma. J Glaucoma. 2014;23:494–500. 16. Fadda A, Di RA, Martelli F, Marangoni D, Batocchi AP, Giannini D, Parisi V, Falsini B. Reduced habituation of the retinal 68 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. ganglion cell response to sustained pattern stimulation in multiple sclerosis patients. Clin Neurophysiol. 2013;124:1652–1658. Saidha S, Al-Louzi O, Ratchford JN, Bhargava P, Oh J, Newsome SD, Prince JL, Pham D, Roy S, van Zijl P, Balcer LJ, Frohman EM, Reich DS, Crainiceanu C, Calabresi PA. Optical coherence tomography reflects brain atrophy in multiple sclerosis: a fouryear study. Ann Neurol. 2015;78:801–813. Saidha S, Syc SB, Ibrahim MA, Eckstein C, Warner CV, Farrell SK, Oakley JD, Durbin MK, Meyer SA, Balcer LJ, Frohman EM, Rosenzweig JM, Newsome SD, Ratchford JN, Nguyen QD, Calabresi PA. Primary retinal pathology in multiple sclerosis as detected by optical coherence tomography. Brain. 2011;134:518–533. Balcer LJ, Baier ML, Cohen JA, Kooijmans MF, Sandrock AW, Nano-Schiavi ML, Pfohl DC, Mills M, Bowen J, Ford C, Heidenreich FR, Jacobs DA, Markowitz CE, Stuart WH, Ying GS, Galetta SL, Maguire MG, Cutter GR. Contrast letter acuity as a visual component for the multiple sclerosis functional composite. Neurology. 2003;61:1367–1373. Porciatti V, Ventura LM. Normative data for a user-friendly paradigm for pattern electroretinogram recording. Ophthalmology. 2004;111:161–168. Wei Y, Jiang H, Shi Y, Qu D, Gregori G, Zheng F, Rundek T, Wang J. Age-related alterations in the retinal microvasculature, microcirculation, and microstructure. Invest Ophthalmol Vis Sci. 2017;58:3804–3817. Shao Y, Jiang H, Wei Y, Shi Y, Shi C, Wright CB, Sun X, Vanner EA, Rodriguez AD, Lam BL, Rundek T, Baumel BS, Gameiro GR, Dong C, Wang J. Visualization of focal thinning of the ganglion cell-inner plexiform layer in patients with mild cognitive impairment and alzheimer’s disease. J Alzheimers Dis. 2018;64:1261–1273. Tan J, Yang Y, Jiang H, Liu C, Deng Z, Lam BL, Hu L, Oakley J, Wang J. The measurement repeatability using different partition methods of intraretinal tomographic thickness maps in healthy human subjects. Clin Ophthalmol. 2016;10:2403– 2415. Toft-Nielsen J, Bohorquez J, Ozdamar O. Innovative pattern reversal displays for visual electrophysiological studies. Conf Proc IEEE Eng Med Biol Soc. 2011;2011:2009–2012. Ito H, Ogawa M, Sunaga S. Evaluation of an organic lightemitting diode display for precise visual stimulation. J Vis. 2013;13:6. Chiappa KH. Pattern-shift visual, brainstem auditory and shortlatency somatosensory evoked potentials in multiple sclerosis. Ann N Y Acad Sci. 1984;436:315–327. Evangelou N, Konz D, Esiri MM, Smith S, Palace J, Matthews PM. Size-selective neuronal changes in the anterior optic pathways suggest a differential susceptibility to injury in multiple sclerosis. Brain. 2001;124:1813–1820. Porciatti V, Burr DC, Morrone MC, Fiorentini A. The effects of aging on the pattern electroretinogram and visual evoked potential in humans. Vis Res. 1992;32:1199–1209. Fadda A, Di RA, Parisi V, Stifano G, Balestrazzi E, Riva CE, Falsini B. Lack of habituation in the light adapted flicker electroretinogram of normal subjects: a comparison with pattern electroretinogram. Clin Neurophysiol. 2009;120:1828–1834. Porciatti V, Ventura LM. Adaptive changes of inner retina function in response to sustained pattern stimulation. Vis Res. 2009;49:505–513. Porciatti V, Sorokac N, Buchser W. Habituation of retinal ganglion cell activity in response to steady state pattern visual stimuli in normal subjects. Invest Ophthalmol Vis Sci. 2005;46:1296–1302. Jiang et al: J Neuro-Ophthalmol 2021; 41: 60-68 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.