Albrecht v. Graefes Arch. klin. exp. Ophthal. 206, 107--120 (1978)

Graefes Archiv f~rklinischeundexperimentelle Ophthalmologie 9 by Springer-Verlag 1978

Off-Components in Response to Brief Light Flashes in the Oscillatory Potential of the Human Electroretinogram M. Kojima* and E. Zrenner** Arbeitsgemeinschaft II. Physiol. Abt. des Max-Planck-Instituts fiir Physiologische und Klinische Forschung, W.G. Kerckhoff-Institut, Bad Nauheim (Direktor Prof. Dr. E. Dodt) und Augenklinik der Johann-Wolfgang-Goethe Universit~it, Frankfurt/Main (Direktor Prof. Dr. W. Doden), Federal Republic of Germany

Summary. Amplitudes and implicit times of the several oscillatory potentials (OP's) in the human electroretinogram have been studied in relation to temporal, adaptive and spectral stimulus variables. The last of the OP-wavelets responds differently from the preceding ones; its implicit time shortens with increase in stimulus frequency or adaptive illumination, whereas that of the preceding OP's is prolonged or stays constant. Moreover, the last OP-wavelet is time locked with the stimulus offset, whereas the others are not. This can be confirmed by linear subtraction of responses to an isolated stimulus onset from responses to stimuli of variable durations. These results indicate, that the last oscillatory potential is related to the light offset and is presumably generated by the retinal off-elements described in single cell recordings. Z u s a m m e n f a s s u n g . Die Amplituden und Gipfelzeiten der einzelnen schnellen Potentialschwingungen im menschlichen Elektroretinogramm (oszillatorische Potentiale) wurden im Bezug zu zeitlichen, adaptiven und spektralen Reizparametern untersucht. In einer Serie von oszillatorischen Potentialen (OP's) zeigt dasjenige mit der l~ingsten Gipfelzeit auffallende Unterschiede gegeniiber den vorangehenden; seine Gipfelzeit verkiirzt sich mit Steigerung der Reizfrequenz oder der adaptiven Beleuchtung, w~ihrend die Gipfelzeit der vorangehenden OP's sich verl~ingert oder gleich bleibt. Dartiber hinaus zeigt das letzte OP im Gegensatz zu den/ibrigen einen zeitlichen Bezug zum Reizende. Dies l~ifgtsich erh~irten durch lineare Subtraktion seiner, durch einen isolierten Hellsprung ausgel6sten on-Antworten yon on/off-Antworten auf Reize unterschiedlicher Dauer. Die Be-

* Present address: Department of Ophthalmology, Niigata University School of Medicine,

Niigata-shi, Japan ** Present address: National Institutes of Health, Laboratory of Vision Research, Bethesda,

Md. 20014, U.S.A. Address f o r offprint requests: Prof. Dr. E. Dodt, William-G.-Kerckhoff-Institut, Parkstr. 1,

D-6350 Bad Nauheim, Federal Republic of Germany 0065-6100/78/0206/0107/$

02.80

108

M.Kojima and E.Zrenner funde weisen darauf hin, dag das in einer Serie zuletzt auftretende OP in enger Beziehung zur Beendigung eines Lichtreizes steht und vermutlich yon den in Einzel-Zell-Versuchen gefundenen retinalen off-Elementen generiert wird.

Introduction

Besides the major components of the human electroretinogram (ERG) there appear, under certain conditions, a series of high frequency potentials which, in the form of small wavelets, are superimposed on the a-wave, b-wave, and off-effect. According to Yonemura (1963) the wavelets superimposed on the ascending b-wave in the human ERG are usually called the oscillatory potentials (OP's), denoted as 01,02 . . . . . . On, and show a frequency of about 1 0 0 - 1 5 0 cps. 01 is suggested to be purely photopic in nature, while 02 and 03 undergo marked changes during the transition from dark to light adaptation (Adams and Dawson, 1971 ; Algvere et al., 1972 ; Wachtmeister, 1972) and show, for example, a Purkinje shift (Stodtmeister, 1972). Since they are adaptation-dependent, they are also sensitive to alterations of the flash interval (Kojima et al., 1977). OP's were recorded in response to the termination of light stimuli (i.e., in the off-effect) by Best and Bohnen (1957b), Tsuchida et al. (1971), and Langhof et al. (1977). Previous investigations on flicker (Granit and Riddell, 1934; Dodt, 1951; Heck, 1957; Best and Bohnen, 1957a; Nagata, 1962) exhibited a strong interaction between the a-wave, b-wave, and the off-effect in the photopic ERG. In the present study special attention was paid to the interaction between the OP's evoked by the onset and by the termination of the test light. By changing the stimulus interval, stimulus duration, and adaptive illumination, the last wavelet (' 0 L') was found to show a strict relation to the end of the stimulus, i.e., an oscillatory off-potential, even in response to very short flashes.

Methods

The optical stimulator used provided two independent beams originating from a xenon arc source (XBO 150W/l, Osram). The beams of light were controlled by an electromagnetic shutter driven by a pulse generator (Tektronix 26G 3) and a rate/ramp generator (Tektronix 26G1) providing a stimulus rise and fall time of about 2.0 and 2.5 ms, respectively. The beams passed through neutral absorption filters and interference band filters (SCHOTT AL), and illuminated a diffusing acrylic disk in front of the recording contact lens. Stimulus energy and adaptive illumination were measured behind the diffusing disk by means of a radiant flux meter (8330 A, Hewlett-Packard) and a SEI photometer. The potentials were picked up by a Henkes low-vacuum contact lens electrode, amplified by a differential preamplifier (WPI, Mod. DAM-6 A), led to an oscilloscope (Tektronix 565), and averaged by a Nicolet computer (Model 1072). The band pass of the recording system was 1 0 0 - 3 0 0 Hz (3 dB point of 6 dB/octave). The height of the oscillatory responses was measured by averaging the mean amplitude of the rising and falling branches of the individual wavelets from the peak to the adjoining troughs (a/2 + b/2 in Fig. 1). The implicit time (peak latency) was measured from the be-

Oscillatory Potential

109

02 a

01

OL 1.000 0 . 6 0 0 "-" 60 0.300 _ >

0.100

d) C 09

0.O6O "5 0 . 0 4 5 i= ~176 . ~176

j

0 . 0 4 0 O9 9

~176

0.030

L

10

IJV]

2 0 ms

Fig. 1. Averaged (n = 4) OP's in the h u m a n ERG (band pass of the recording s y s t e m 1 0 0 - 3 0 0 Hz, 3 dB p o i n t of 6 dB/octave) in response to Ganzfeld stimulation of 104"s td). Upward deflection indicates positivity of corneal electrode. Each series of responses was recorded after at least four prestimuli. Test-light duration 5 ms, as indicated by lowermost record. The n u m b e r s beside each record denote the stimulus interval in s. The last OP is denoted as O L. Observer: H.Z.

110

M.Kojima and E. Zrenner

ginning of the light stimulus to the peak of the individual wavelets. To obtain steadystate responses, measurements were started only after at least four prestimulations. Two young healthy student volunteers served as subjects. Mydriat{cum Roche and Neosynephrine 5% were used for dilatation of the pupil. A small red lamp was used for fixation.

Results According to previous reports, the OP's during both onset (Algvere et al., 1972; Kojima et al., 1977) and offset of illumination (Langhof et al., 1977) change with the intensity and the repetition rate of the test light and the intensity of the adaptive illumination. Using flash intervals between 0.2 and 180 s in the dark adapted eye, the amplitude of 01 remained rather constant at all levels of adaptive illumination, whereas the amplitudes of 02, 03, and 04 underwent, predominantly in the mesopic range of adaptive illumination, fluctuations in amplitude showing a maximum at a flash interval between 10 and 20 s (Kojima et al., 1977). In the following experiments, fluctuations of the OP's are described at stimulus intervals ranging between 0.03 and 1.0 s. As illustrated in Figure 1, the appearance of the OP's 01 - 0 L in response to constant-intensity light flashes of short duration (6 ms) varied with the stimulus interval. A waxing and waning of amplitudes is seen, caused by interference between the components as the implicit time of 0 L decreases and that of 01 - 02 increases with shortening of the flash interval (indicated by broken lines). F o r example, at a stimulus interval of 0.1 s, component 03 is superimposed on component 0L, producing a potential of higher amplitude. Similarly, at stimulus intervals less than 0.06 s, 01 and 02 cannot be identified as separate events. Finally, at a stimulus interval of 0.03 s the components 01 - 0 L merge so that only a single potential is left. For clarity, Figure 1 depicts only selected recordings; the full set of recordings includes smaller steps of stimulus interval showing the transition (indicated by dashed line) to occur smoothly and gradually. Figure 2 shows the implicit time, i.e., the time from the onset of the light stimulus to the peak of the individual OP of the components 01 - 0 L versus the stimulus interval. The implicit time of the single remaining component at the shortest stimulus intervals (three filled squares at the right side of the figure) appears to be the component showing the longest latency at rates of stimulation below 1 s. The implicit time of 0 L markedly decreases, in contrast to the implicit times of the other components. The increase of implicit time of 0 1 , 0 2 , and 03 at very short flash interval can be attributed to the preceding flicker stimulus. It is shown later that the last OP, 0 L, is elicited by the termination of the stimulus. The changes o f potential shown above could be due to either temporal interaction of successive stimuli, or to changes in the state of adaptation. Therefore, another experiment was designed, with the dark interval kept constant while the adaptive illumination was changed (Fig. 3). Using a 6 ms test flash of fixed intensity and a dark interval of 10 s, stepwise increases of adaptive illumination over a total range of 3.4 log units decreased the number of OP's in a manner similar to the previous experiment. When the implicit time of one OP coincided with the foregoing one (for example 03 and 0 L at an adaptive illumination of 2.5 log td, or 02 and 03 at 3.5 log td), superposition occurred, and an OP of higher amplitude resulted. As seen from the graph of Figure 4A, shortening of the implicit time occurred only for the last OP, 0 L.

e-,

9

9,,.~,~ ~ '

~o

t:3"

~'

~~ - ~

~.P

~=~ o

~ , ~ O

~PDo~ E

o~

~-

o~,-~

~ -

~:"

04 .o [] ...... u . . . u . . "''''u"

O o'~176176176

o'"

~

u . - n ....

i

I

1

I

~~

I

~

..................

I

I

"....

....

..........

......... A

I

.... ~

FI

I

I

0 Log stimulus interval

I

~o

~

..... ~ ...... ~ . A

..... a - ...... 4::]

..1:] "''[] ...........

. . . . . 0 . . . 0 ..... 0 . ' 0 " . . . . . 0 . . . . . . . 0 . 0 -

_ o.2.

I

~

(05) ...O. . . . . . . . .

. . O . - O ....

-_ 0 3A - .... "A-...A. .... . / ~ . . ~

-

-

2 0 _ oO, ~~

030

E

E

.-----

50

(s)

I

I

~

I

OL

I

o 8 q

.,-,,

112

M.Kojima and E.Zrenner

0203

'

,

'"\

!ll

/

0.1 t'-

0.6 o 0.9 1.4

r-

2.0 2.5

i E ~' / , , , . / / "

>

"-------- 2.7 S.. 3.0 -o O3

3.5 o

_J

,

20ms Fig. 3. OP's (n = 4) in xesponse to light flashes of 6 ms duration and of 10 4.8 td during exposure to stepwise increased adaptive illumination as indicated beside each record. Constant stimulus interval (10 s), Observer: H.1.

Oscillatory Potential

! "o

113

50 40

03

~

30

?2

~-,

i_

~___+____+_+ "~

o,

o

,o

ooo

I

I

o

2O I

a

0

25 >

I

I

J

,

1 2 3 Log adaptive illumination (td)

m

B 20

v

15 E 10