Characteristic ERG Flicker Anomaly in Incomplete Congenital Stationary Night Blindness

Transcription

1 Characteristic ERG Flicker Anomaly in Incomplete Congenital Stationary Night Blindness Yozo Miyake, Masayuki Horiguchi, Ichiro Ofa, and Noriyasu Shiroyamo Ten patients with the incomplete type of congenital stationary night blindness (CSNB) were examined with a 30 Hz flicker electroretinogram (ERG). After 30 min of dark adaptation, 30 Hz flicker ERG was recorded continuously for min under white background illumination. All patients showed an exaggerated increase of amplitude and a universal characteristic change of wave shape as the light adaptation progressed. Thirty normal subjects also showed increased amplitude during light adaptation, but the increase in amplitude was significantly less than in incomplete-type CSNB, and there was little change in wave shape. The same procedure was applied to patients with complete-type CSNB, retinitis pigmentosa, congenital retinoschisis, cone dystrophy, and Oguchi's disease; neither the exaggerated increase of amplitude nor the wave change was seen. Our results indicate that incomplete-type CSNB is a newly identified cone-rod dysfunction syndrome with a special functional property. Invest Ophthalmol Vis Sci 28: , 1987 The Schubert-Bornschein type 1 of congenital stationary night blindness (CSNB) is characterized by a uniquely shaped electroretinogram (ERG) when the ERG is recorded with a single bright white stimulus in the dark. The a-wave amplitude is normal, but the b-wave is smaller, resulting in a negative ERG. We previously reported cases of CSNB with a negative ERG can be divided into two types. 2 The complete type lacked any rod function, whereas in the incomplete type the ERGs evoked by blue light and psychophysical dark adaptation gave evidence of residual rod activity. Inheritance of incomplete and complete types was strictly familial; this and other differences between the two types (in the degree of refractive error, size of cone and rod b-waves and oscillatory potentials) suggested that the types were different clinical entities. 2 Recently, an additional new finding in incomplete-type CSNB was reported by Takahashi and co-workers. 3 They found that incomplete-type CSNB patients showed normal electrooculogram (EOG) with nondetectable c-wave in ERG. Further ERG studies were carried out on ten cases of incomplete-type of CSNB. After 30 min of dark adaptation, 30 Hz flicker ERG was recorded continuously for min with white background illumination. All patients showed an exaggerated increase From the Department of Ophthalmology, Nagoya University School of Medicine, Nagoya, Japan. Submitted for publication: February 18, Reprint requests: Yozo Miyake, MD, Department of Ophthalmology, Nagoya University School of Medicine, 65 Tsuruma-cho, Showa-ku, Nagoya 466, Japan. of amplitude and characteristic change of ERG wave shape during the recording period; such changes were not seen in any normal subjects, in complete-type CSNB, or in other kinds of retinal dystrophy that we examined. Our results suggest that in this type of CSNB, there is an abnormal rod-cone interaction, and this may account for some of the perceptual losses. Incomplete-Type CSNB Materials and Methods Among patients previously reported, 2 ten male patients with six pedigrees were examined. Among the six pedigrees, three indicated probable X-linked recessive inheritance (Arch Ophthalmol 4:, Fig. 9, Families, 11, and 12). Cases 1 and 2 were siblings. Case 3 was a cousin of cases 1 and 2 (son of their mother's sister). Case 5 was a grandfather of case 4 (mother's father). The clinical characteristics of the patients are shown in Table 1. Corrected visual acuity ranged from 0.2 to 0.7. The dark adaptation, as measured by a Goldmann-Weekers adaptometer, 2 showed evidence of rod adaptation in all patients, although the final threshold (measured 30 min after dark adaptation) was elevated by approximately to 2.0 log units. The visual field, measured by Goldmann perimetry, was normal with V-4 and 1-4 targets, but was narrower than normal with the 1-2 target in all patients. Color vision, evaluated by using with Ishihara pseudoischromatic plates, Hardy-Rand- Rittler pseudoisochromatic plates, the Farnsworth dichotomous panel D-15 test, an anomaloscope and 1816

2 No. 11 ERG FLICKER ANOMALY IN INCOMPLETE CSND / Miyoke er ol Table 1. Clinical characteristics of examined patients ERG(% reduced) Disease Case no./age, yrs/sex Inheritance pattern Visual acuity Dark adaptation Rod Cone 30-Hz flicker EOG (ratio) Incomplete-type CSNB Complete-type CSNB Retinitis pigmentosa 1//M 2/17/M 3/16/M 4/12/M 5/74/M 6/13/M 7/21/M 8/16/M 9/14/M /21/M 11/14/M 12/16/M 13/12/M 14/11/M 15/9/M 16//M 17/19/M 18/36/M 19/8/F /28/F 21/36/F Brother affected Brother affected Autosomal recessive Autosomal recessive Autosomal recessive Autosomal dominant Autosomal dominant Autosomal recessive Congenital retinoschisis Cone dystrophy Oguchi's disease 22//M 23/17/M 24/17/M /36/M 26/32/M 27/43/M 28/26/F 29//F 30/37/M 31/18/M Autosomal dominant Autosomal dominant Patients 1 through 3 are from one family, patients 4 and 5, patients 6 and 7, patients 11 and 12, and patients 23 and 24, are brothers. Dark adaptation is expressed as the elevation of the final threshold (in log units) in comparison with the normal threshold. Abbreviations are as follows: CSNB, congenital stationary night blindness; M, male; F, female; ERG, electroretinogram;, nonrecordable; EOG, electrooculogram. the Farnsworth-Munsell 0 hue test, was essentially normal in all but two patients (cases 2 and 6) who showed a mild blue-yellow defect. Routine and standard ERG recordings were performed with a Ganzfeld system. 2 A single-flash ERG with intense white light revealed a normal a-wave with a reduced b-wave in all patients. The ratio of b-wave to a-wave was below, indicating a negative ERG (normal range: -1.75). In all patients oscillatory potentials were detectable, and rod b-waves were recorded with normal peak times and reduced amplitudes. Cone b-waves were significantly smaller than normal in all but one patient but the peak time was normal in eight patients in whom the response was large enough for precise measurements to be made. The b-wave of the other two patients was a noise-level response, and we could not measure the implicit time accurately. The 30 HzflickerERGs were significantly smaller than normal in all but one patient. The implicit time of the positive peak was within the normal range in all patients. The EOG was recorded in nine patients. 2 The light peak:dark trough ratio (normal value 0 or larger) was within the normal range in seven patients. Complete-Type CSNB Among 35 patients previously reported, 2 five, with four pedigrees, were examined. All patients were male, and two pedigrees indicated probable X-linked recessive inheritance (Arch Ophthalmol 4:, Fig. 9, Families 1 and 2). The patient's mother and father were consanguineous in two cases. The clinical characteristics of these five patients appear in Table 1. The single-bright-flash ERGs had a negative wave form in all patients, and oscillatory potentials were nondetectable in all but one patient (case 13). Rod adaptation as indicated by psychophysical dark adap-

3 1818 INVESTIGATIVE OPHTHALMOLOGY b VISUAL SCIENCE / November 1987 Vol min 3 min 5 min 7 min = 0 0 examined eye. The intensity of the stimulus light was 3.47 log trolands, which was approximately 16 times brighter than that for our routine 30 Hz flicker ERG. An electronic shutter created a stimulus with an onoff ratio of. The 30 Hz flicker stimuli were given continuously for 12 to 15 min, and the 30 Hz flicker ERG was recorded every 30 sec. Sixty-four responses were averaged by a signal processor in each recording, and the responses were plotted on an X-Y plotter. The time constant and high cut frequency of the amplifier were 0.1 second and 1 KHz, respectively. Thirty normal subjects (17 male, 13 female) were examined. The ages ranged from 14 to 69 years (mean 41 years) and they had refractive errors ranging from + to -4.5 diopters (mean, 0.95D). Informed consent had been obtained from all subjects and patients. msec (min) Time after light adaptation Kig. I. Changes of 30 Hz flicker clcctrorctinogram (ERG) during light adaptation (following 30 min dark adaptation) in a normal subject (left), and mean relative amplitude (±2 SD) in 30 normal subjects during light adaptation (right). The amplitude recorded I min after the onset of stimulation is normalized to 0 percent. The mean voltage corresponding to 0 percent is 45 /xv. tation and rod ERG were absent in all of these patients. Cone ERG and 30 Hz flicker ERG were normal or slightly below normal in amplitude. The implicit time of cone ERG b-wave and 30 Hz flicker ERG were within the normal range in all patients. Other Diseases Other patients examined included six with retinitis pigmentosa, four with congenital retinoschisis, five with cone dystrophy and one with Oguchi's disease. 4 The clinical characteristics of these patients appear in Table I. Continuous Recording of 30 Hz Flicker ERG The method for continuous recording of 30 Hz flicker ERG was as follows: after pupillary dilatation with 1% tropicamide and 30 min of dark adaptation, a Burian-Allen bipolar contact lens was applied to the patient's cornea under dim red light, and we started to record the 30 Hz flicker ERG. A constant white background illumination of 1.78 log trolands was used. The white stimulus light was delivered from a 500 W xenon arc, which was led into a small Ganzfeld system (8 cm diameter) placed in front of the Subjects Results Figure 1 shows the 30 Hz flicker ERG of a normal subject during light adaptation after 30 min of dark adaptation (left) and mean relative amplitude (±2 SD) of 30 normal subjects in response to 30 Hz stimulation as a function of time after onset of stimulation (right). The amplitudes were measured from the bottom to the top of the sine wave-like responses. The responses recorded 1 min after the onset of the stimulus were normalized to 0 percent, and subsequent amplitudes were shown as percentages of the 1 min response. The amplitude gradually increased during the first 5 to 15 min (mean,.4 min) of continuous stimulation. The mean (±2 SD) of the maximum amplitude was 160 ± 60 percent in 30 subjects. The half time of the time course of increase was 3.3 min. The wave form, as well as the implicit time, showed little change during the course of recording. Incomplete-Type CSNB Figure 2 shows the 30 Hz flicker ERGs during the course of continuous stimulation in a normal subject (upper left) and in ten patients with incomplete-type CSNB. Note the difference in calibration marks between normal and patients; the flicker ERG amplitude is much reduced in patients. The mean (±2 SD) of the amplitude in 30 normal subjects, recorded 1 min after the onset of stimulation, was 46.0 ± 27 /nv, while that in incomplete-type patients was 8.0 ± 7.0 ^V. These differences were statistically significant (P < 0.001). During light adaptation, however, all incomplete-type patients showed an exaggerated enhancement of amplitude. The maximum amplitudes

4 No. 11 ERG FLICKER ANOMALY IN INCOMPLETE CSND / Miyoke er ol Incomplete Type CSNB Case 1 Case 2 Case 3 Case 4 V\ Fig HzflickerERG during light adaptation in a normal subject (upper left) and in ten patients with incomplete-type congenital stationary night blindness (CSNB). Note the difference in calibration marks between normal and patients. Case 5 Incomplete Type CSNB Case 6 Case 7 Case 8 Case 9 Case msec were 312, 414, 270, 304, 346, 0, 234, 2, 224, and 230 percent in cases 1-, respectively. All these percents were beyond normal range except in case 7, whose rate was upper limit of normal range. Maximum amplitude was obtained around min after the onset of stimulation, which was roughly the same time period as in normal subjects. Although an exaggerated increase of amplitude was seen in all patients, the amplitude did not increase to the normal level. The mean (±2 SD) of the maximum amplitude was 23.7 ± 18.4 ixv in incomplete-type CSNB, which was statistically smaller (P < 0.001) than that in normal subjects (74.6 ± 37.2 /tv). Another characteristic feature of incomplete-type patients was a drastic change in wave shape during light adaptation. An example of the change of wave shape during light adaptation appears in Figure 3. Shortly after the onset of stimulation, the ERG was

5 18 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / November 1987 Vol min 2 min 3 min 4 min 5 min min ON OFF msec Fig. 3. An example of the change of wave shape of 30 Hz flicker ERG during light adaptation in incomplete-type CSNB. the sine wave-like shape seen in normal subjects (Fig. 3, 1 min). The peak time of the first positive peak in incomplete-type patients did not significantly differ from that in normal subjects. (The mean ±2 SD was msec in normal subjects, and.3 ± 2.1 msec in incomplete CSNB). However, after 2 to 3 min, each wave showed two positive peaks (PI, P2). At this point, PI was larger than P2. A few minutes later, the amplitudes of PI and P2 became comparable, and thereafter only P2 amplitude increased. The peak time of PI and P2 did not correspond to the sine wave peak seen at 1 min. The mean (±2 SD) of the peak time of PI and P2 was 7.9 ± msec and 17.8 ± 3 msec, respectively. This wave separation phenomenon was observed in all incomplete-type patients. Other Diseases No patients with other diseases whom we examined showed either the exaggerated increase of amplitude or the wave separation phenomenon noted in incomplete-type CSNB. Figure 4 shows the changes of 30 Hz flicker ERG during light adaptation in one representative patient from each disease group. As seen in normal subjects, the wave shape changed little during the light adaptation. Figure 5 shows the change in amplitude (percent) in relation to time during light adaptation in ten patients with incomplete-type CSNB, five patients with complete-type CSNB, six with retinitis pigmentosa, four with congenital retinoschisis, five with cone dystrophy, and one with Oguchi's disease. Only incomplete-type CSNB patients showed an exaggerated increase of amplitude. The increase in amplitude in the other diseases was within or below the normal range. Discussion In all normal subjects, the amplitude of the 30 Hz flicker ERG increased significantly during the light adaptation period following 30 min of dark adaptation. A similar phenomenon has been reported by other investigators, using frog 5 " 7 or human eyes. 8 " 12 We reported previously that the increase in amplitude during light adaptation varied considerably among normal human subjects. 13 However, we found that the variation became small when we used relatively intense stimuli, 13 like the stimulus used in this study. It is striking that all incomplete-type patients showed an exaggerated increase of amplitude and a characteristic change of wave shape during light adaptation, which were never seen in patients with other kinds of disease. In this study, we examined patients with diseases that resembled incomplete-type CSNB in some clinical aspects. Findings were similar for complete- and incomplete-type CSNB patients in inheritance, stationary condition, essentially normal fundi, negative ERG, normal or near-normal EOG, moderately poor visual acuity, and essentially normal color vision. 2 Similarly, patients with congenital retinoschisis may have X- linked recessive inheritance, negative ERG, moderately deteriorated cone and rod ERG, good EOG, and moderately poor visual acuity. Those with retinitis pigmentosa or cone dystrophy may show deteriorated cone and/or rod ERG. Those with Oguchi's disease

6 No. 11 ERG FLICKER ANOMALY IN INCOMPLETE CSND / Miyake er ol Complete Type Retinitis Congenital Cone Oguchi's CSNB Pigmentosa Retinoschisis Dystrophy Disease (Case 11) (Case 16) (Case 24) (Case 28) (Case min Fig. 4. Changes of 30 Hz flicker ERG during light adaptation in one patient with complete-type CSNB, one with retinitis pigmentosa, one with congenital retinoschisis, one with cone dystrophy, and one with Oguchi's disease. L_ had the stationary condition and negative ERG. As reported previously, 2 some patients with incompletetype CSNB showed an abnormal reflex in the inferior peripheral fundus, with abnormal color somewhat resembling the color in Oguchi's disease. Although all of the diseases mentioned above were similar in some aspects to incomplete-type CSNB, no patient showed such characteristic changes of amplitude and wave shape as seen in incomplete-type patients during light adaptation. The mechanism of increase of cone ERG amplitude during light adaptation is not fully understood. Armington advocated the correlation between this phenomenon and change of standing potential of the eye, because standing potential showed a similar increase during Jight adaptation after dark adaptation (lightrise).however, we found that patients with retinitis pigmentosa 13 or vitelliform macular dystrophy 14 also showed this phenomenon in spite of the absent light rise in EOG. In addition, isolated retina (retina freed from the pigment epithelium and removed from the eye) from frog 67 and carp 15 showed a similar increase in cone ERG amplitude. Therefore, it is unlikely that the pigment epithelium plays a role in this phenomenon. MacKay and Gouras" observed this phenomenon in human cone ERG. Since a-wave also showed an increase, they thought that this phenomenon occurs in the cone photoreceptors, presumably through a redepolarization of light-adapting cones. This phenomenon may result from rod-cone interaction, as was suggested by Granit 5 and Hood. 67 However, supposing it is true, this sort of interaction may be different from that reported by Arden and co-workers. 16 " 19 According to them, the interaction has a practically instantaneous onset and is triggered by quite weak light, does not work with Ganzfelds, and is absent in CSNB or congenital retinoschisis. These findings are incompatible with ours. Indeed it is most unlikely that rods gradually stop suppressing cones under light adapted condition. Therefore, some other mechanism might well serve as the model of the phenomenon in our experiment. If one supposes that the alteration in the temporal response characteristics

7 1822 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / November 1987 Vol. 28 Incomplete Type CSNB Retinitis Pigmentosa Case 4 Case 19 Case Complete Type CSNB Case 1 Case 18 Congenital Retinoschisis Case 22 Cone Dystrophy Case 21 Case Case 29 Case 30 Fig. 5. Changes of amplitude of 30 Hz flicker ERG during light adaptation in patients with incompletetype CSNB, complete-type CSNB, congenital retinoschisis, retinitis pigmentosa, and Oguchi's disease. Vertical axis indicates percent amplitude; horizontal indicates time after the start of light adaptation (min). For each patient, the first amplitude on the left is adjusted to 0 percent, and subsequent amplitudes indicate the relative percent amplitudes. The curve for each patient is shifted vertically to prevent overlap. The calibration mark indicates the size of a 50 percent increase in amplitude. The numbers on the right indicate the case numbers shown in Table 1. Case 14 Oguchi's Disease Calibration Case 15 of the horizontal cell operates slowly via an interplexiform feed back, - 21 then the slowing in the flicker response would parallel the change in the temporal adaptation of the retina. If this is the case, the horizontal cell plays a major role in this phenomenon. However, our previous results 15 indicated that the P III retina in live carp (P III component was isolated by asparate sodium) also showed an increase of cone ERG amplitude during light adaptation, in spite of the dysfunctioning horizontal cells. It is interesting, however, that the increase in P III amplitude was less than in a-, b-, and d-waves recorded from the intact retina. 15 A similar finding was observed in patients with unilateral central retinal arterial occlusion. 14 When the central retinal artery is occluded in human retina, the inner retinal layers (including horizontal cells) are severely impaired, leaving the photoreceptors intact. Our previous results 14 indicated that an increase of 30 HzflickerERG amplitude during light* adaptation was observed from the affected eye, but it was less than that from the normal fellow eye. These results suggest that some control from the inner retina might contribute, in part but not entirely, to this phenomenon. From these results, we think that although this phenomenon may, ultimately and in part, have something to do with rod-cone interaction, the sort of interaction may be different from those reported electrophysiologically, 16 " 19 psychophysically 16 " 19 ' 22 ' 23 or morphologically 24 by other authors.

8 No. 11 ERG FLICKER ANOMALY IN INCOMPLETE C5ND / Miyake er al We previously reported that 30 Hz flicker ERGs and cone ERGs in incomplete-type CSNB patients are significantly smaller than in complete-type patients, even though psychophysical cone functions affecting visual acuity, visual field, and color vision are comparable in both types. 2 Since our routine ERGs were recorded after ample dark adaptation, the cone responses in incomplete-type patients was exaggeratedly small, as was shown in this study. However, the small amplitude of cone ERG and 30 Hz flicker ERG after dark adaptation may not indicate that the cone function is extremely poor, since the amplitude increased greatly as light adaptation progressed (although the amplitude did not increase to the normal level). In other words, the cone ERG system is exaggeratedly suppressed by dark adaptation in incomplete-type CSNB. In this study, we proved that the complete and incomplete types of CSNB are independent disease entities. We believe that the incomplete type is a cone-rod dysfunction syndrome that has not been identified previously. In addition, we found that the characteristic results obtained with the 30 Hz flicker ERG may have potential value in diagnosis or analysis of the pathogenesis of some eye diseases. Key words: congenital stationary night blindness, incomplete type, 30 HzflickerERG, increase in amplitude, light adaptation Acknowledgment Shinobu Awaya, MD, Professor and Chairman of the Department of Ophthalmology, Nagoya University, Japan, revised this manuscript. References 1. Schubert G and Bornschein H: Beitrag zur Analyse des menschlichen Electroretinogram. Ophthalmologica 123:396, Miyake Y, Yagasaki K, Horiguchi M, Kawase Y, and Kanda T: Congenital stationary night blindness with negative electroretinogram: A new classification. Arch Ophthalmol 4:13, Takahashi Y, Onoe S, Yoshimura Y, Asamizu N, and Tazawa Y: Congenital stationary night blindness, incomplete type (Miyake)Case report, with special reference to clinical findings on sensory-motor system abnormalities. Folia Ophthalmol Jpn 38:9, Oguchi C: Uber einen Fall von eigenartigen Hemeralopie. Acta Soc Ophthalmol Jpn 11:123, Granit R: Processes of adaptation in the vertebrate retina in the light of recent photochemical and electrophysiological research. Doc Ophthalmol 1:7, Hood DC: Adaptational changes in the cone system of the isolated frog retina. Vision Res 12:875, Hood DC: Suppression of the frog's cone system in the dark. Vision Res 12:889, Burian H: Electric responses of the human visual system. Arch Ophthalmol 52:509, Kawabata H: Course of the potential visual change in the human electroretinogram during light adaptation. J Opt Soc Amer 50:456, Armington JC: The Electroretinogram. New York, Academic Press, Inc., 1974, pp MacKay CJ and Gouras P: Light adaptation arguments in the amplitude of the human cone ERG. ARVO Abstracts. Invest Ophthalmol Vis Sci 26(Suppl):323, Lachapelle P: Segmental analysis of photopic b-wave. ARVO Abstracts. Invest Ophthalmol Vis Sci 27(Suppl):302, Miyake Y, Horiguchi M, and Yagasaki K: Increment of the amplitude of human photopic ERG during light adaptation. Acta Soc Ophthalmol Jpn 90:12, Ota I, Shiroyama N, Horiguchi M, and Miyake Y: Adaptational changes in human cone flicker ERG. Acta Soc Ophthalmol Jpn, in press. 15. Horiguchi M, Takabayashi A, and Miyake Y: Increment of the amplitude of the electroretinogram in cone system during light adaptation. Proc Soc Eye Res 4:179, Arden GB and Hogg CR: Rod-cone interactions and analysis of retinal disease. Br J Ophthalmol 69:4, Arden GB and Frumkes TE: Stimulation of rods can increase cone flicker ERGs in man. Vision Res 26:711, Arden GB: Rod-cone interactions in night-blinding disease. Jpn J Ophthalmol 31:6, Frumkes TE, Eysteinsson Thor, and Arden GB: Tonic inhibition of rod-cone pathway by dark adapted rods. ARVO Abstracts. Invest Ophthalmol Vis Sci 26(Suppl):l 14, Mangel SC and Dowling E: Responsiveness and receptive field size of carp horizontal cells are reduced by prolonged darkness and dopamine. Science 229:17, Knapp AG and Dowling JE: Dopamine enhances excitatory amino acid-gated conductances in cultured retinal horizontal cells. Nature 3:437, Goldberg SH, Frumkes TE, and Nygaard RW: Inhibitory influence of unstimulated rods in the human retina: Evidence provided by examining cone flicker. Science 221:180, Alexander K.R and Fishman GA: Rod influence on cone flicker detection: Variation with retinal eccentricity. Vision Res 26:827, Raynauld JP, Laviolette JR, and Wagner HJ: Goldfish retina: A correlate between cone activity and morphology of the horizontal cell in cone pedicules. Science 4:1436, 1979.