Abstract
purpose. To report the properties of a newly detected positive wavelet on the descending limb of P1 of the first-order kernel of the human multifocal electroretinogram (mfERG).
methods. Twenty eyes of 20 normal individuals, ages 21 to 29 years (mean, 25.6) and nine eyes of 6 patients with optic neuritis ages 5 to 38 years (mean, 17.3) were studied. mfERGs were recorded with a visual evoked response imaging system with the number of stimulus elements set at 37. The stimulus frequency was changed from 75 to 37, 18, 9.4, 4.7, and 2.3 Hz, and the contrast of the stimuli was lowered to 50%.
results. In normal eyes, a positive wavelet appeared on the descending limb of P1 of the first-order kernel of the mfERG when the stimulus frequency was reduced from 75 to 18 Hz. The wavelet had a mean amplitude of 4.2 nV/deg2 and a mean implicit time of 34 ms at 18 Hz. When the stimulus frequency was reduced further to 2.3 Hz, the amplitude of the wavelet increased significantly (P < 0.05) compared with that at 18 Hz. The amplitudes of the wavelet elicited from the nasal side of the retina were significantly larger (P < 0.05) than those from the temporal side and decreased significantly (P < 0.05) with increasing distance from the optic disc. The wavelet was not present in any of the patients with newly diagnosed optic neuritis, but reappeared with recovery from the disease. The recovery of the wavelet correlated significantly with the recovery of visual acuity and of central critical fusion frequency.
conclusions. The amplitude of the wavelet on the descending limb of P1 of the first-order kernel of the mfERG was dependent on the stimulus frequency and the retinal locus. The wavelet was not present in the mfERGs recorded in patients with optic neuritis, but returned with recovery from the disease. These findings suggest that the neural activity of the ganglion cells give rise to this wavelet.
Initially, it was generally believed that multifocal electroretinograms (mfERGs) originate mainly from the activity of neurons in the outer layers of the retina, and that the contribution of ganglion cells was not significant. In recent years, however, Hood et al.
1 2 3 have found that the monkey mfERG contains a component that originates from the neurons in the retina that produce action potentials, ganglion cells and their axons, and perhaps amacrine cells. This component contributes to the optic nerve head component (ONHC) that has been described in the human mfERG.
4 The spike-driven component in monkeys was studied using 100% contrast; the ONHC in humans saturates at 60% contrast.
4 5 Hood et al.
1 2 3 described an inner retinal component in humans, but did not establish that it is a ganglion cell component. Wu and Sutter
6 used slow-sequence stimuli to bring out the oscillatory potentials (OPs) of the mfERGs, which are likely to originate from the inner retina.
We examined whether changing the frequency and contrast of the stimulus would reveal the activity of the ganglion cells in the first-order kernel of the mfERGs. The results show that at a stimulation rate of 18 Hz and lower, a positive wavelet appeared on the descending limb of P1 at an amplitude that was dependent on stimulus frequency and retinal locus. This wavelet was absent in patients in the early stage of optic neuritis, but it emerged and increased in amplitude during the recovery stage with recovery of visual acuity (VA) and central critical fusion frequency (CFF).
Nine eyes of six patients in whom optic neuritis was diagnosed between December 1999 and November 2000 were observed. The optic neuritis in four patients (seven eyes) was the papillitis type, and that in two patients (two eyes) was the retrobulbar type
(Table 1) . The optic neuritis in three eyes of three patients was unilateral and in six eyes of three patients was bilateral. The patients’ ages ranged from 5 to 38 years (mean, 17.3), and their refraction ranged from −0.25 to −4.5 D (mean, −1.47). The corrected VA (in logMAR units) of the affected eyes at the patients’ first visits ranged from 0.4 to 2.0. All the patients with the papillitis type had had fever and headaches before the disease developed. The unaffected eyes did not show any abnormalities. At the first visit, the eye position and eye movement of the affected eyes were normal. The patients did not experience any pain during eye movements. There were no abnormalities in the conjunctiva, cornea, anterior chamber, and pupil. Marcus-Gunn syndrome was noted in pupils in the unilateral cases. The optic media were clear.
In the papillitis type, the optic discs were red and swollen and the margins of the discs were blurred. The retinal arteries and veins were dilated and tortuous. Retinal edema was observed around the disc. In the patients with the retrobulbar type, there were no abnormalities in the optic discs or retina.
Goldmann visual field and color vision test results were normal in the healthy eyes. It was not possible to conduct these tests on the affected eyes of the patients, except for the eye of patient 4, which showed an enlarged blind spot. The central CFF was 40 to 50 Hz in the healthy eyes. It was not possible to conduct this test on the affected eyes, except in patient 4, whose CFF was 18 Hz. Intraocular pressure was normal in all eyes.
Optic nerve swelling was observed in the patients with papillitis by cephalic magnetic resonance imaging (MRI) scan. Multiple macular demyelinating foci were observed in the brain’s white matter in the patients with the retrobulbar type, also by cephalic MRI scan.
mfERGs were first recorded 4 days on average after the recovery of fixation, and VA had recovered to approximately 0.7 to 0.2 logMAR units. Thereafter, mfERGs were recorded of six times in each patient. The VA and central CFF tests were also measured over a period of 1 to 6 months (average, 3).
A visual evoked response imaging system (VERIS IIIR; Tomei, Nagoya, Japan) was used to record the mfERGs. For stimulation, 37 hexagonal elements were arranged in concentric circles, and the overall visual angle of the stimulus pattern subtended an angle of approximately 50° horizontally and 40° vertically at the eye. The stimuli were generated on a 14-in. television monitor, and the stimulus elements were alternated between white (200 cd/m
2) and black (4 cd/m
2) at a pseudorandom binary-m sequence rate. Six frequencies (75, 37, 18, 9.4, 4.7, and 2.3 Hz) were used. In the interval between the white and black stimuli at frequencies lower than 75 Hz, the contrast setting of the imaging system was adjusted to 50%. In this setting, the luminance of the elements was 60 cd/m
2, which appeared gray
(Fig. 1) . The luminance of the surrounding area was set at 100 cd/m
2.
Before the mfERG recordings, the uncorrected and corrected VAs were determined, and the refraction, after mydriasis by topical tropicamide, was measured with a refractometer. The cornea was anesthetized by topical oxybuprocaine hydrochloride and 4% lidocaine. A bipolar Burian-Allen contact lens electrode, coated with hydroxyethylcellulose, was placed on the eye, and the other eye was patched. A silver disc electrode was placed on the right earlobe as a ground electrode. A lens, the spherical equivalent of the refractive error, was placed in front of the eye, and the distance between the TV monitor and the chin support was adjusted according to the method of Kondo et al.
7
mfERGs were recorded after adapting the subjects to the luminance of the examination room (6 cd/m
2) for 15 minutes. Four recording periods of 30 seconds each were performed, for a total recording time of 2 minutes at each stimulation frequency. The recordings were started from the highest stimulus frequency, and after the completion of the recordings, mfERGs were recorded at the next lower stimulus frequency after a recovery interval of 1 minute. The band-pass filter was set at 10 to 300 Hz. Spatial averaging was not performed, but an artifact-elimination technique was applied by replacing the segment containing the artifact with an estimated response computed from the entire record.
8
The mfERGs were analyzed with software that accompanied the imaging system (Veris Science, ver. 3.0.1; Tomei) and the all-traces wave of the first-order kernel, which is the summation of the individual retinal responses from the 37 retinal areas, was used as the response.
Kruskal-Wallis analysis of variance was used as a test of significance of the results, and P < 0.05 was considered statistically significant.
All-Traces Response.
Differences in the Wavelet According to Retinal Loci.
Relationship between the Wavelet and Distance from the Optic Disc.
Sutter and Tran
8 have reported that mfERGs represent the response of the cone system, because the response density of the first-order kernel is highest in the center of the fovea and was lowest in the periphery. This is consistent with the distribution of the number of cone cells. Kondo et al.
10 11 also concluded that the mfERG reflects the activity of the cone system, because bright stimuli cause light adaptation of the retina at a stimulation frequency of 75 Hz.
Horiguchi et al.
12 injected different neuroactive drugs (e.g., the glutamate analogues 2-amino-4-phosphono-butyric acid [APB] or
cis-2,3-piperidine-dicarboxylic acid [PDA]) and inhibitory neurotransmitters glycine and γ-aminobutyric acid (GABA) into the rabbit vitreous to isolate the different components of the mfERGs. They reported that the bipolar cells contribute significantly to the first-order kernel of the mfERGs. Hood et al.
13 suggested that the first negative wave of the first-order kernel of the mfERG originates from the cells that contribute to the a-wave of the conventional ERG and that the first positive wave of the first-order kernel of the mfERG corresponded to the positive wave of the conventional ERG that originates from bipolar cell activity. Thus, it has been generally accepted that the first-order kernel originates mainly from the outer layers of the retina, and the ganglion cells contribute very little to the first-order kernel.
More recently, however, Frishman et al.
14 and Hare et al.
15 concluded that ganglion cell activity contributes to the first-order kernel of the mfERGs, because the waveform of the first-order kernel of the mfERG is altered by chronic elevation of intraocular pressure. Hood et al.
1 also suggested that the ganglion cells contribute to the first-order kernel, because they found that the first-order kernel of monkey eyes change from two positive peaks to a single peak when they injected tetrodotoxin (TTX) intravitreally, which blocks spiking activity that originates from inner retinal neurons, ganglion cells, and amacrine cells. They concluded that the inner retina contributes to the mfERG and suggested that ganglion cells are involved. Further studies by Hood et al.
16 in humans with glaucoma demonstrated that the mfERG is not a good method for documenting localized field defects. The possible contribution of other inner retinal neurons (e.g., amacrine cells) was raised in the previous studies in monkeys. Thus, the origin of the first-order kernel is still not completely determined.
As shown, we have found a wavelet on the descending limb of P1 that appeared in the all-traces wave when the stimulus frequency was lower than 18 Hz. The amplitude of the wavelet increased significantly when the stimulation frequency was reduced to 4.7 and 2.3 Hz. The presence and characteristics of this wavelet have not been reported before. We have designated this small wave as the s-wave after the size of the wave.
The question then arises as to the cellular origin of the s-wave. The mean implicit time of the s-wave was approximately 34 ms, which corresponds to the 34-ms peak in the monkey mfERG that disappears with administration of TTX.
1 2 Sutter and Bearse
4 5 reported that the ONHC of mfERGs originate from the ganglion cell axons in the vicinity of the optic nerve head and have an implicit time of 32 to 37 ms.
To analyze the properties of the s-wave, we divided the retina into nasal and temporal halves and into areas at different distances from the optic disc. The amplitude of the s-wave was significantly larger on the nasal side (closer to the optic disc) than on the temporal side, and the amplitude decreased as the distance of the test area increased from the optic disc. This is consistent with the histologic report of Curcio and Allen
17 that the number of ganglion cells in the human eye is higher on the nasal side than on the temporal side of the fovea, with a range of 10° to 25° from the center of the macula. Hood et al.
1 also reported that the amplitude of a component, presumably originating from ganglion cells in the first-order kernel of the monkey mfERG, is greater on the nasal side than on the temporal side of the fovea and decreases at greater distances from the optic disc.
The implicit time of the s-wave also increased significantly as the distance of the test area increased from the optic disc. This was similar to the observation that the implicit time of the ONHC increases as the region tested increases from the optic disc.
4 5 Thus, these findings support the idea that the s-wave originates from the ganglion cells.
In optic neuritis, the s-wave was not observed in any of the eyes at the first mfERG recording. Then, the s-wave appeared and increased with the recovery of VA and central CFF. The amplitude of the s-wave correlated significantly with the recovered VA and central CFF. There is good evidence that ganglion cell disorders, such as optic neuritis, are reflected in reduced central CFF. These clinical findings provide strong support for the idea that the s-wave reflects ganglion cell activity.
Sutter and Bearse
4 5 reported that the ONHC saturates at 60% contrast, and Hood et al.
2 reported that the component that resembles an inner retinal component in monkeys and is more obvious in humans when the contrast is reduced to 50%. When recordings were made in normal human subjects with a stimulation contrast of 100%, the s-wave was not present at any stimulus frequency. These findings indicate that lowering the stimulus contrast to 50% and reducing the stimulus frequency are the best recording conditions to isolate the s-wave.
To differentiate the s-wave from the OPs, we recorded conventional full-field ERGs in the patients with optic neuritis and found that the OPs were present. These findings indicate that the s-wave probably is not one of the OPs—for instance, OP4.
In summary, we found that changing the stimulus frequency and lowering the stimulus contrast led to the unmasking of a new wavelet, which we named the s-wave, with an implicit time of approximately 34 ms. The amplitude of the s-wave was larger on the nasal side of the fovea and in areas closer to the optic disc. Especially significant were the findings that the s-wave was not present in patients with optic neuritis and reappeared with recovery from the disease. These properties are in keeping with an origin of the s-wave in the ganglion cells.
Submitted for publication July 13, 2001; revised December 17, 2001; accepted December 20, 2001.
Commercial relationships policy: N.
The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “
advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.
Corresponding author: Marie Sano, Department of Ophthalmology, Iwate Medical University School of Medicine, 19-1 Uchimaru, Morioka, Iwate 020-8505, Japan;
marie@rnac.ne.jp.
Table 1. Patients with Optic Neuritis
Table 1. Patients with Optic Neuritis
| Patient | Age (y) | Side | Type | Refraction (D) | CVA (LogMAR) | | Central CFF (Hz) |
| | | | | Initial Visit | Time* | |
| 1 | 5 | R | P | −0.75 | 2.0 | 0.2 | 25 |
| | L | | −0.75 | 2.0 | 0.3 | 25 |
| 2 | 6 | R | P | −1 | 2.0 | 0.7 | 26 |
| | L | | −1 | 2.0 | 0.5 | 20 |
| 3 | 7 | R | P | −0.5 | 1.7 | 0.7 | 25 |
| | L | | −0.25 | 1.4 | 0.5 | 25 |
| 4 | 21 | R | P | −4.5 | 0.4 | 0.2 | 25 |
| 5 | 27 | L | R | −4 | 1.6 | 0.3 | 17 |
| 6 | 38 | L | R | −0.5 | 2.0 | 0.7 | 25 |
Table 2. Averaged Amplitudes and Implicit Times of the Wavelets at Each Stimulus Frequency
Table 2. Averaged Amplitudes and Implicit Times of the Wavelets at Each Stimulus Frequency
| Stimulus Frequency (Hz) | Amplitude (nV/deg2) | Implicit Time (ms) |
| 18.0 | 4.06 ± 1.57 | 33.67 ± 1.76 |
| 9.4 | 4.46 ± 1.14 | 33.45 ± 2.01 |
| 4.7 | 4.95 ± 1.06 | 33.38 ± 1.83 |
| 2.3 | 5.41 ± 1.71 | 33.96 ± 1.80 |
Table 3. Averaged Amplitudes and Implicit Times of the Wavelets Recorded from the Nasal and Temporal Retinas at Each Stimulus Frequency
Table 3. Averaged Amplitudes and Implicit Times of the Wavelets Recorded from the Nasal and Temporal Retinas at Each Stimulus Frequency
| Stimulus Frequency (Hz) | Area | Amplitude (nV/deg2) | Implicit Times (ms) |
| 18.0 | Nasal | 5.32 ± 1.27 | 32.97 ± 1.76 |
| Temporal | 4.09 ± 1.22 | 33.64 ± 1.65 |
| 9.4 | Nasal | 5.59 ± 1.76 | 33.32 ± 1.87 |
| Temporal | 4.27 ± 1.09 | 33.63 ± 1.82 |
| 4.7 | Nasal | 5.83 ± 1.95 | 33.23 ± 1.92 |
| Temporal | 4.48 ± 1.45 | 33.44 ± 1.69 |
| 2.3 | Nasal | 6.29 ± 1.90 | 33.83 ± 1.89 |
| Temporal | 4.78 ± 1.05 | 33.91 ± 1.91 |
Table 4. Averaged Amplitudes and Implicit Times of the Wavelets Recorded from Area 1–4 at Each Stimulus Frequency
Table 4. Averaged Amplitudes and Implicit Times of the Wavelets Recorded from Area 1–4 at Each Stimulus Frequency
| Stimulus Frequency (Hz) | Area | Amplitude (nV/deg2) | Implicit Time (ms) |
| 18.0 | 1 | 4.84 ± 1.80 | 32.60 ± 1.87 |
| 2 | 4.29 ± 1.32 | 33.15 ± 1.90 |
| 3 | 3.56 ± 0.95 | 33.44 ± 1.68 |
| 4 | 3.61 ± 0.87 | 33.14 ± 1.77 |
| 9.4 | 1 | 5.45 ± 1.85 | 32.51 ± 1.96 |
| 2 | 4.34 ± 1.97 | 33.55 ± 1.93 |
| 3 | 4.08 ± 1.14 | 33.57 ± 1.99 |
| 4 | 4.01 ± 1.22 | 33.94 ± 2.01 |
| 4.7 | 1 | 5.66 ± 1.85 | 32.77 ± 1.97 |
| 2 | 5.09 ± 1.66 | 33.50 ± 1.96 |
| 3 | 4.38 ± 2.12 | 33.45 ± 2.02 |
| 4 | 4.35 ± 1.88 | 33.83 ± 1.86 |
| 2.3 | 1 | 6.28 ± 2.16 | 33.13 ± 1.99 |
| 2 | 5.56 ± 2.49 | 33.44 ± 1.83 |
| 3 | 4.39 ± 1.78 | 33.69 ± 1.56 |
| 4 | 4.51 ± 1.41 | 34.13 ± 1.71 |
The authors thank Takeshi Sugawara, Shigeki Machida, and Toshihiro Gotoh for continuing support and Ei-ichiro Nagasaka for helpful technical assistance.
Hood DC, Frishman LJ, Viswanathan S, Robson JG, Ahmed J. Evidence for a ganglion cell contribution to the primate electroretinogram (ERG): effects of TTX on the multifocal ERG in macaque. Vis Neurosci
. 1999;16:411–416.
[PubMed]Hood DC, Greenstein V, Frishman L, et al. Identifying inner retinal contributions to the human multifocal ERG. Vision Res
. 1999;39:2285–2291.
[CrossRef] [PubMed]Hood DC, Bearse MA, Sutter EE, Viswanathan S, Frishman LJ. The optic nerve head component of the monkey’s (Macaca mulatta) multifocal electroretinogram (mERG). Vision Res
. 2001;41:2029–2041.
[CrossRef] [PubMed]Sutter EE, Bearse MA, Jr. The optic nerve head component of the human ERG. Vision Res
. 1999;39:419–436.
[CrossRef] [PubMed]Sutter EE, Bearse MA, Jr. Extraction of a ganglion cell component from the corneal response. Vision Science and Its Applications, OSA Technical Digest Series. 1995;1:310–313. Optical Society of America Washington, DC.
Wu S, Sutter EE. A topographic study of oscillatory potentials in man. Vis Neurosci
. 1995;12:1013–1025.
[CrossRef] [PubMed]Kondo M, Miyake Y, Horiguchi M, Suzuki S, Ito Y, Tanikawa A. Normal values of retinal response densities in multifocal electroretinogram. [in Japanese]J Jpn Ophthalmol Soc. 1996;100:810–816.
Sutter EE, Tran D. The field topography of ERG components in man. 1: the photopic luminance response. Vision Res
. 1992;32:433–446.
[CrossRef] [PubMed]Yonemura T. Oscillatory potential of ERG. [in Japanese]J Jpn Ophthalmol Soc. 1962;66:1566–1583.
Kondo M, Horiguchi M, Miyake Y, Suzuki S, Tanikawa A. Effects of rapid random flash stimuli on electroretinographic response. [in Japanese]Folia Ophthalmol Jpn. 1996;47:531–535.
Kondo M, Miyake Y, Horiguchi M. Recording multifocal electroretinogram on and off responses in humans. Invest Ophthalmol Vis Sci
. 1998;39:574–580.
[PubMed]Horiguchi M, Suzuki S, Kondo M, Tanikawa A, Miyake Y. Effect of glutamate analogues and inhibitory neurotransmitters on the electroretinograms elicited by random sequence stimuli in rabbits. Invest Ophthalmol Vis Sci
. 1998;39:2171–2176.
[PubMed]Hood DC, Seiple W, Holopigian K, Greenstein V. A comparison of the components of the multi-focal and full-field ERGs. Vis Neurosci
. 1997;14:533–544.
[CrossRef] [PubMed]Frishman LJ, Saszik S, Harwerth RS, et al. Effects of experimental glaucoma in macaques on the multifocal ERG. Doc Ophthalmol
. 2000;100:231–251.
[CrossRef] [PubMed]Hare W, Ton H, Woldemussie E, Ruiz G, Feldmann B, Wijono M. Electrophysiological and histological measures of retinal injury in chronic ocular hypertensive monkeys. Eur J Ophthalmol. 1999;9:30–33.
Hood DC, Greenstein VC, Holopigian K, et al. An attempt to detect glaucomatous damage to the inner retina with the multifocal ERG. Invest Ophthalmol Vis Sci
. 2000;41:1570–1579.
[PubMed]Curcio CA, Allen KA. Topography of ganglion cells in human retina. J Comp Neurol
. 1990;300:5–25.
[CrossRef] [PubMed]