A comparison of photopic and scotopic electroretinographic changes in early diabetic retinopathy.

Abstract

Previous studies of early diabetic retinopathy have shown that oscillatory potential (OP) amplitudes are reduced in many diabetic patients. OP amplitude is believed to be a more sensitive indicator of the development of future retinopathy than b-wave amplitude of the scotopic electroretinogram (ERG). Because OPs measured to a bright white flash reflect both rod and cone system activity, it is important to compare OP amplitudes to photopic ERG measures as well as scotopic measures in early diabetic retinopathy. In this study, OPs and ERG responses were measured under photopic and scotopic conditions in a group of diabetic patients. Although OPs were reduced in amplitude in the diabetic group, several other parameters of the scotopic and photopic b-waves were impaired. The results indicate that b-wave activity may indicate retinal changes in early diabetic retinopathy in the same manner as the OPs.

Full text

Page 1

Investigative Ophthalmology & Visual Science, Vol. 33, No. 10, September 1992
Copyright © Association for Research in Vision and Ophthalmology
A Comparison of Phoropic and Scotopic
Electroretinogrophic Changes
in Early Diabetic Refinoparhy
Karen Holopigian, William Seiple, Monica Lorenzo, and Ronald Carr
Previous studies of early diabetic retinopathy have shown that oscillatory potential (OP) amplitudes
are reduced in many diabetic patients. OP amplitude is believed to be a more sensitive indicator of the
development of future retinopathy than b-vvave amplitude of the scotopic electroretinogram (ERG).
Because OPs measured to a bright white flash reflect both rod and cone system activity, it is important
to compare OP amplitudes to photopic ERG measures as well as scotopic measures in early diabetic
retinopathy. In this study, OPs and ERG responses were measured under photopic and scotopic
conditions in a group of diabetic patients. Although OPs were reduced in amplitude in the diabetic
group, several other parameters of the scotopic and photopic b-waves were impaired. The results
indicate that b-wave activity may indicate retinal changes in early diabetic retinopathy in the same
manner as the OPs. Invest Ophthalmol Vis Sci 33:2773-2780, 1992
Diabetic retinopathy is one of the ocular complica-
tions associated with diabetes mellitus. For some dia-
betic patients, the retinopathy will progress to a prolif-
erative stage (including neovascularization and reti-
nal hemorrhages). Because these patients require laser
photocoagulation to slow the progression of retinopa-
thy, early identification is needed to ensure adequate
follow up.
Some reports indicate that the oscillatory potentials
(OPs) of the flash electroretinogram (ERG) are re-
duced in amplitude in many diabetic patients and
that this reduction in OP amplitude may predict the
development of proliferative retinopathy.1"5 Yone-
mura et al1 were the first to note the relationship be-
tween OP amplitude and the severity of diabetic reti-
nopathy. They found that patients with advanced dia-
betic retinopathy had extremely small or extinguished
OPs, even when the a- and b-waves of the ERG were
normal in amplitude. In addition, they found that the
OP amplitude was reduced in more than 50% of dia-
betic patients with no ophthalmologic changes. Later
studies on patients with retinopathies ranging from
From the Department of Ophthalmology, New York University
Medical Center, New York, New York.
This work was partially supported by a grant from RP Fighting
Blindness, Baltimore, Maryland, and by an unrestricted grant to the
Department of Ophthalmology, NYU Medical Center, from Re-
search to Prevent Blindness, Inc, New York, New York.
Submitted for publication: September 23, 1991; accepted April
1, 1992.
Reprint Requests: Dr. Karen Holopigian, Department of Oph-
thalmology, 550 First Avenue, New York, NY 10016.
moderate nonproliferative to proliferative supported
the conclusion that the OPs are selectively impaired
in diabetes, relative to the amplitude of the scotopi-
cally measured b-wave.3"5
OP amplitude also has been shown to predict the
continued progression of diabetic retinopathy. Si-
monsen2 studied the predictive value of OP ampli-
tude in a longitudinal study of diabetes patients whose
retinopathies ranged from no detectable retinopathy
to proliferative retinopathy. He found that OP ampli-
tude at the initial visit was a reliable predictor of prolif-
erative retinopathy development.
One implication of this literature is that the OPs are
better indicators of diabetic retinopathy than other
components of the ERG, such as b-wave amplitude.
That is, the OPs are specifically affected early in this
disease. This conclusion, however, never has been ade-
quately tested. Previous studies have two major prob-
lems. First, in the studies where the OPs were com-
pared to b-wave amplitudes, patients with interme-
diate or advanced stages of retinopathy typically were
examined.3"6 However, diabetic patients with little or
no diabetic retinopathy must be tested to assess the
prognostic value of OP amplitude reductions.
The second problem is the appropriate ERG com-
parisons. Typically, the amplitude of the OPs has
been compared to the amplitude of the scotopically
measured a- and b-waves.1"267 However, the OPs
measured under standard clinical conditions are com-
posed of both scotopic and photopic components.8"10
Some investigators have argued that the OPs may be
specific to the photopic system." As a result, compari-
2773

Page 2

2774
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / September 1992
Vol. 33
sons of OP amplitude to scotopic ERG components
may not be complete.
The present study differed from existing studies in
two important ways. First, patients with early diabetic
retinopathy were tested (Table 1). Second, OP ampli-
tudes were compared to a range of ERG parameters.
Amplitudes and implicit times of scotopic and phot-
opic parameters of the standard ERG were recorded.
In addition, OPs were measured under recording con-
ditions designed to maximize photopic or scotopic
contributions. This protocol allowed us to determine
whether the OPs are selectively impaired in early dia-
betic retinopathy and to assess the relative effects of
early diabetic retinopathy on scotopic versus phot-
opic ERG parameters.
Materials and Methods
Observers
Fourteen patients with diabetes mellitus and 14
age-similar controls were subjects in this experiment.
All were recruited from the Ophthalmology Clinic in
Bellevue Hospital (New York), after a complete ocu-
lar exam. The subjects gave informed consent to par-
ticipate after a full explanation of the procedure was
given.
The diabetic patients had visual acuities better than
20/50, a history of diabetes ranging from 4-21 yr (me-
dian 12 yr), no history of cataracts, and no other ocu-
lar or health problems. Four of the diabetic patients
were classified as Type I and the remaining 10 were
classified as Type II. The mean age (±standard devia-
tion) of the diabetics was 49.1 ± 12.8 yr. All diabetic
patients received a fluorescein angiogram and fundus
photographs. The level of retinopathy and degree of
macular edema were independently classified by two
retinal specialists, according to a modified Airlie
House classification scheme, and the scores were
averaged. The clinical characteristics of these patients
are given in Table 1.
The control observers had normal visual acuity and
no history of ocular or health problems. The mean age
of the controls was 45.5 ± 11.6 yr.
Apparatus and Procedure
The eye with the better visual acuity was tested after
pupil dilation (1% tropicamide and 2.5% phenyleph-
rine). All patients had pupil diameters greater than 6
mm. The patients were patched and dark adapted for
40 min, and full-field ERGs were measured with a
Grass (Boston, MA; PS22) flash with a ganzfeld
surround. ERGs were recorded with a monopolar
Burian-Allen (Hansen Ophthalmic, Iowa City, IA)
electrode. The forehead served as reference and the
ipsilateral ear served as ground. For the a- and b-wave
recordings, the signal was amplified (1 K; Grass
preamplifier P511) and filtered (1-300 Hz). For the
OP recordings, the amplification was increased to 5 K
and the signals were filtered between 100 and 1000
Hz. All ERG recordings were digitized (512 Hz), sig-
nal averaged (n = 5), and analyzed on a computer.
Under scotopic conditions, ERGs were recorded to
a low intensity blue flash (Grass SI and a Wratten
[Kodak, Rochester, NY] 47b filter) and to white
flashes of increasing intensity (Grass S16) for a voltage
versus intensity function (from -4.1 to 1.9 log td).
Scotopic OPs were measured to a low intensity (Grass
Table 1. Clinical characteristics—diabetics
ObsEyeSex
Age
AcuityType
Dur.
20 yr
15 yr
5yr
21 yr
4 yr
15 yr
Unknown
Syr
12 yr
12 yr
18 yr
12 yr
10 yr
9yr
Ret. level
(grader)
1
2
3
1
2
1
3
2
2
1
1
2
2
2
2
*
2
2
3
1
2
1
2
2
2
1
1
2
2
2
2
y
0
I
0
0
0
0
0
0
0
0
1
0
0
0
Edema
level*
(grader)
2
0
0
0
0
0
0
0
0
0
0
1
0
0
1
CB
CC
CF
DH
HM
RO
AO
SP
DS
ST
DL
MC
JA
LR
OD
OD
OD
OD
OD
OS
OD
OS
OS
OD
OD
OD
OD
OD
F
M
F
F
M
F
F
F
F
M
F
F
M
F
51
48
61
33
25
51
59
68
52
58
26
49
50
57
20/25+1
20/20
20/30"2
20/25"2
20/25+2
20/20"3
20/20"'
20/40"2
20/30"3
20/20"'
20/20
20/20
20/20
20/25"'
II
II
II
I
I
I
II
II
II
II
I
II
II
II
Mean age = 49.1, SD = 12.8.
* The level of retinopathy and the degree of macular edema were assessed
from fluorescein angiograms by two independent graders using the following
modification of the Airlie House classification scheme:
Level of retinopathy: Level 1—normal fundus; Level 2—one or more mi-
croaneurysms only; Level 3—microaneurysms with one or more other non-
proliferative lesions present of mild to moderate degree.
Degree of macular edema: Grade 0—none; Grade 1—questionable macu-
lar edema.

Page 3

No. 10
ERG CHANGES IN DIABETES / Holopigion er ol
2775
Fig. 1. Averaged scotopic
b-wave responses to the blue
(Wratten 47a) flashes. The
left side of the figure shows
the averaged b-wave ampli-
tudes and the right side
shows the averaged implicit
times. Each pair of histo-
grams shows the results for
the diabetics (hatched) and
the controls
stripes). Error bars are ±1
standard error.
(diagonal
S8) blue flash (Wratten 47b; rod dominated response)
and to a bright (Grass S16) white flash (mixed rod and
cone response). After 10 min of light adaptation, a
photopic flash response, a 30 Hz flicker response, and
a photopic flash OP response (cone dominated) were
measured to a Grass S16 white stimulus. For all OP
recordings, a conditioning flash was used.121310
In addition, 12 of these diabetic patients also had
the course of dark adaptation at one retinal locus
measured and had measures of dark-adapted thresh-
olds in four retinal quadrants. These results are pre-
sented elsewhere.1415 After all testing was completed,
fluorescein angiograms and retinal photographs were
obtained on the diabetic patients.
Results
The data obtained to the low intensity blue flashes
(S1 blue) are presented in Figure 1. The left side of the
figure shows the averaged b-wave amplitudes for the
diabetics (hatched) and the controls (diagonal stripes).
There were no significant differences in amplitude be-
tween the two groups, (t(26) = 0.42; P > 0.05). The
right side of the figure shows the averaged implicit
times. The implicit times for the diabetic patients
were significantly delayed, relative to the controls
(t(26) = 2.64; P < 0.05).
The a-wave findings are presented in Figure 2. Be-
cause the response of the outer retina reflected in the
a-wave cannot be measured independently of the b-
wave, we transformed a-wave responses using a
method proposed by Hood and Birch.16"17 For each
subject, the a-wave response to the highest intensity
flash was truncated before intrusion of the b-wave.
These data then were plotted as log (-amplitude) ver-
sus log time. From these functions, the slope and in-
tercepts were calculated and averaged for the control
Fig. 2. Transformation of
scotopic a-wave amplitude
data.1617 The portion of the
ERG (solid line) used in this
analysis is shown in the top
left and the a-wave ampli-
tude function is shown at
the top right. The bottom
portion of the figure shows
the averaged slopes and in-
tercepts for the diabetics
(hatched) and controls (diag-
onal stripes). Error bars are
± 1 standard error.
3OO
200
100
0
-100
-200
-300
-400
S16
/
tv
Scotopic
S16 Scotopic
a-wave Transform
Slope
2.50
2.25
2.00
1.75
1.50
1.00 1.25 1.50 1.75 2.00
Loo time
Intercept
0.00
-0.10
-0.20
-030
-0.40
-0.50
-0.60
-O.7O
-O.80
-O.9O
-1.OO

Page 4

2776
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / September 1992
Vol. 33
and diabetic groups. Shown in the top left of Figure 2
is a graphic demonstration of the portion of the ERG
(solid line) used in this analysis. Shown in the top
right is the a-wave amplitude function plotted on log
amplitude versus log time axes. The bottom portion
of Figure 2 shows the averaged slopes and intercepts
for the diabetic (hatched) and control groups (diago-
nal stripes). There were no significant differences be-
tween the groups for either measure (slope: t(20)
= 0.21, P > 0.05; intercept: t(20) = .08, P > 0.05).
The averaged b-wave amplitudes (measured from
the trough of the a-wave to the peak of the b-wave) as
a function of flash intensity are shown in Figure 3.
The results for the diabetics are shown as open
squares and the results for the controls as filled circles.
For both groups, b-wave amplitude increased with
flash intensity. The repeated measure design analysis
of variance (ANOVA) indicated there were no signifi-
cant differences in amplitude between the diabetic
and control groups, nor was there a significant inter-
action between group and flash intensity (see Table 2
for ANOVA values). In addition, for each observer,
the Naka-Rushton equation was best fit to the ampli-
tude versus intensity data. The estimates of Rmax and
k (semi-saturation constant) were averaged for each
group. There were no significant differences in Rmax
or k for the two groups (Rmax: t(20) = 1.35, P > 0.05;
k:t(20)= 1.80, P> 0.05).
The averaged b-wave implicit times as a function of
flash intensity are presented in Figure 4. The implicit
times for the diabetic patients (open squares) were de-
layed at all flash intensities relative to the control ob-
servers (filled circles). Significant differences in im-
plicit times between the two groups were found, as
was a significant interaction between group and flash
intensity (Table 2).
Table 2. Repeated measures analysis of variance
1. Scotopic b-wave amplitude
Group
Filter strength
Interaction
2. Scotopic b-wave implicit time
Group
Filter strength
Interaction
3. Scotopic blue OP amplitude
Group
OP (1-4)
Interaction
4. Scotopic blue OP implicit time
Group
OP (1-4)
Interaction
5. Scotopic white OP amplitude
Group
OP (1-4)
Interaction
6. Scotopic white OP implicit time
Group
OP (1-4)
Interaction
7. Photopic white OP amplitude
Group
OP (1-3)
Interaction
8. Photopic white OP implicit time
Group
OP (1-3)
Interaction
F value
0.3
121.3
1.2
12.8
486.0
3.5
10.5
44.8
3.7
4.2
1300.1
2.0
10.3
33.1
0.3
1.7
14.0
1.3
3.2
18.0
0.2
2.2
893.9
2.2
df
1
12
12
1
12
12
1
3
3
1
3
3
1
3
3
1
3
3
1
2
2
1
2
2
P value
>0.05
<0.001
>0.05
<0.005
<0.001
<0.00l
<0.005
<0.001
<0.05
>0.05
<0.001
>0.05
<0.005
<0.001
>0.05
>0.05
<0.001
>0.05
>0.05
<0.001
>0.05
>0.05
<0.001
>0.005
OP, oscillatory potential.
The results for the photopic components of the
ERG are shown in Figures 5 and 6. The averaged am-
plitudes (left) and implicit times (right) of the phot-
opic b-wave for the diabetics (hatched) and controls
(diagonal lines) are shown in Figure 5. There were no
significant differences in either measure for the two
600
500
3 400
"§ 300
a
E 200
<
1OO
TT
I!
-5
- 4
-3
-2
Log relative intensity
Fig. 3. Averaged scotopic b-wave amplitude data as a function of
flash intensity for the diabetics (squares) and controls (circles).
Error bars are ± 1 standard error.
125
1OO
75
50
25
-5-4-3-2-1 0 1 2
Log relative intensity
Fig. 4. Averaged scotopic b-wave implicit time data as a function
of flash intensity for the diabetics (squares) and controls (circles).
Error bars are ± I standard error.

Page 5

No. 10
ERG CHANGES IN DIABETES / Holopigion er ol
2777
Fig. 5. Averaged photopic
b-wave responses to the
photopic white flashes. The
left side of the figure shows
the averaged b-wave ampli-
tudes and the right side
shows the averaged implicit
times. Each pair of histo-
grams shows the results for
the diabetics (hatched) and
the controls
stripes). Error bars are ±1
standard error.
(diagonal
groups (amplitude: t(26) = 1.49, P> 0.05; implicit
time t(26) = 1.44, P> 0.05. The averaged amplitudes
and implicit times of the photopic 30 Hz flicker are
presented in Figure 6. For the,diabetic group, ampli-
tude was significantly reduced (t(25) = 2.19; P < 0.05)
and implicit time was significantly delayed (t(25)
= 2.88; P< 0.01).
The OP results are shown in the next series of fig-
ures. The peak-to-trough amplitudes of the individual
OP wavelets were measured individually so the rela-
tive involvement of each component could be as-
sessed. The OP amplitudes measured under scotopic
conditions to a blue flash are shown in Figure 7 A, and
the corresponding implicit times are shown in Figure
7B. For all OPs, the amplitudes for the controls (diago-
nal lines) are approximately 50% larger than for the
diabetics (hatched). There was a significant difference
between the amplitudes for the two groups, and a sig-
nificant interaction (Table 2). The OP implicit times
for the diabetics, however, were not significantly de-
layed, relative to the controls.
The results for the OPs measured to the bright
white flash under scotopic conditions are shown in
Figure 8. There were significant differences in ampli-
tude (Figure 8A) between the two groups, but no sig-
nificant interaction (Table 2). The OP implicit times
for this condition (Figure 8B) were not statistically
different for the two groups.
The results for the OPs measured under photopic
conditions are shown in Figure 9. Under these condi-
tions, only three OPs were recordable. Although the
OP amplitudes for the diabetics were smaller than the
amplitudes for the controls (Figure 9A), this differ-
ence was not statistically significant. Consistent with
the other OP conditions, there were no significant dif-
ferences in implicit times between the groups (Figure
9B; Table 2).
To examine the degree of impairment in the indi-
vidual OPs (ie, OP1-OP4), we calculated the percent
loss in amplitude (relative to the amplitude of the
control group) for each OP. The percent loss values
are shown in Table 3. Larger values indicate greater
losses in amplitude. There were no significant differ-
ences among the percent amplitude loss for the differ-
ent OPs for any measurement condition.
.
Discussion
Selective Impairment of the OPs
This study demonstrates that OP amplitudes are
significantly reduced in patients with early diabetic
retinopathy. The loss of OP amplitude agrees with
previous studies.1"36 in comparison, the amplitudes
of the scotopically measured b-waves were normal for
our patients, also consistent with previous work.1'2'7
However, several parameters of the ERG were found
. Fig. 6. Averaged photopic
b-wave responses to the 30-
Hz flicker. The left side of
the figure shows the aver-
aged b-wave amplitudes and
the right side shows the
averaged implicit times.
Each pair of histograms
shows the results for the dia-
betics (hatched) and the
controls (diagonal stripes).
Error bars are ±1 standard
error.

Page 6

2778
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / September 1992
Vol. 33
60
48
36
CD
"D
3
•k-i
— 24
a
< ,2
OP1
OP2
OP3
OP4
B
u
1
C D
50
30
20
a 10
OP1
OP2
OP3
OP4
Fig. 7. Averaged OP amplitudes to the scotopic blue flashes. (A)
The averaged OP amplitudes; (B) the averaged implicit times. Each
pair of histograms shows the results for the diabetics (hatched) and
the controls (diagonal stripes). Error bars are ± 1 standard error.
to be abnormal in our diabetic patients. These parame-
ters include both b-wave amplitudes and implicit
times and ERGs measured under scotopic and photo-
pic conditions. Although scotopic b-wave amplitude
is not a sensitive indicator of early diabetic retinopa-
thy, other parameters of the ERG, including scotopic
b-wave implicit times, indicate the presence of early
retinal changes. These ERG parameters were not ex-
amined in previous studies of the selective involve-
ment of the OPs.1"3 Our results indicate that OP am-
plitudes are not selectively impaired in early diabetic
retinopathy, relative to other ERG parameters.
Selective Impairment of Photopic Versus Scotopic
ERG Parameters
The ERG abnormalities found in our patients were
present for both rod dominated (SI blue) and cone
dominated (photopic flicker) conditions, indicating
that both systems show losses in early diabetic retinop-
athy.
The changes in the scotopic ERG in the diabetic
patients is corroborated by elevations in psychophysi-
cally measured dark adapted thresholds in the same
patients14"15 and by other studies of dark adaptation
in patients with diabetic retinopathy.18-19
Selective Impairment of Individual OP Wavelets
A significant body of evidence suggests that the indi-
vidual OPs have different neural generators. For ex-
ample, individual OPs have different retinal depth
profiles, with the earlier OPs arising more proximally
within the retina than the later ones.20 The earlier OPs
also have been shown to be sensitive to the disruption
of gamma-aminobutyric acid blockers, whereas the
later OPs are more vulnerable to glycine antago-
nists.21"23 Also, the OPs are differentially affected by
disease24"25 and show different time courses of light
adaptation.26"27 As a result, in the present study we
examined the amplitudes of the individual OP wave-
lets rather than the sum of the OP amplitudes2"5 or the
root mean square Fourier amplitude.28 The results
from this study did not show selective changes in OP
amplitude, but did show a uniform reduction for
all OPs.
80
"O
~
Q
32
^ 16
OP1
OP2OP3
OP4
OP1OP2
OP3
OP4
Fig. 8. Averaged OP amplitudes to the scotopic white flashes. (A)
The averaged OP amplitudes; (B) the averaged implicit times. Each
pair of histograms shows the results for the diabetics (hatched) and
the controls (diagonal stripes). Error bars are ± 1 standard error.

Page 7

No. 10
ERG CHANGES IN DIABETES / Holopigian er ol
2779
Site of Retinal Losses in Early Diabetic Retinopathy
Do the results from this study provide any informa-
tion about the retinal locus of diabetic retinopathy?
Previous studies have established that the OPs are im-
paired in patients with diabetic retinopathy. The OPs
have been shown to derive from the inner plexiform
layers involving the axon terminals of the bipolar
cells, the processes of the amacrine cells, and the den-
drites of the ganglion cells.29
The results of the present study indicate that addi-
tional retinal sites are affected in early diabetic reti-
nopathy. The changes in b-wave amplitudes and im-
plicit times are consistent with defects in the mid-re-
tinal layer.30 Also, recent evidence suggests that
changes in the rod dominated b-wave implicit time
may reflect sensitivity losses at the level of the recep-
tors.16"17 Additional support for photoreceptor
changes derives from studies of retinal hypoxia.31"32
In the cat, the inner retina is relatively unaffected by
systemic hypoxia, whereas the outer retina is sensitive
to small variations in arterial oxygen tension. This
vulnerability of the outer retina to hypoxia is partly a
50
Table 3. Percent loss in oscillatory potential (OP)
amplitude (from control)
40
30
" D
+-'
•^
a
20
< 10
B 5 0
|
C D 3 °
E
•H
20
•t—i
O
*Q 10 •
OP1
OP2
OP3
Fig. 9. Averaged OP amplitudes to the photopic white flashes. (A)
The averaged OP amplitudes; (B) the averaged implicit times. Each
pair of histograms shows the results for the diabetics (hatched) and
the controls (diagonal stripes). Error bars are ± 1 standard error.
I. Scotopic blue OP
II. Scotopic white OP
III. Photopic white OP
OP no.
1 •
2
3
4
1
2
3
4
1
2
3
Percent loss
48.9
52.1
51.4
53.3
35.2
35.4
47.3
61.5
33.6
18.4
20.1
result of the high oxygen consumption of the photore-
ceptors.31"32 We are examining patients with ad-
vanced diabetic retinopathy to evaluate rod photore-
ceptor status.
Key words: diabetes mellitus, electroretinogram, oscillatory
potentials
Acknowledgments
We would like to thank D. Hood and V. Greenstein for
many helpful discussions and E. Siderides for evaluation of
the fluorescein angiograms.
References
1. Yonemura D, Aoki T, and Tsuzuki K: Electroretinogram in
diabetic retinopathy. Arch Ophthaimol 68:19, 1962.
2. Simonsen SE: The value of the oscillatory potentials in select-
ing juvenile diabetics at risk of developing proliferative retinop-
athy. Acta Ophthaimol 58:865, 1980.
3. Bresnick GH, Korth K, Groo A, and Palta M: Electroretino-
graphic oscillatory potentials predict progression of diabetic ret-
inopathy. Arch Ophthaimol 102:1307, 1984.
4. Bresnick GH and Palta M: Predicting progression to severe
proliferative diabetic retinopathy. Arch Ophthaimol 105:810,
1987.
5. Bresnick GH and Palta M: Oscillatory potential amplitudes:
Relation to severity of diabetic retinopathy. Arch Ophthaimol
105:929, 1987.
6. Bresnick GH and Palta M: Temporal aspects of the electroreti-
nogram in diabetic retinopathy. Arch Ophthaimol 105:660,
1987.
7. Wanger P and Persson HE: Early diagnosis of retinal changes
in diabetes: A comparison between electroretinography and ret-
inal biomicroscopy. Acta Ophthaimol 63:716, 1985.
8. LaChapelle P, Benoit J, Blain L, Guite P, and Roy MS: The
oscillatory potentials in response to stimuli of photopic intensi-
ties delivered in dark-adaptation: An explanation for the condi-
tioning flash effect. Vis Res 30:503, 1990.
9. King-Smith PE, Loffing FH, and Jones R: Rod and cone ERGs
and their oscillatory potentials. Invest Ophthaimol Vis Sci
27:270, 1986.
10. Peachey NS, Alexander K.R, and Fishman GA: Rod and cone
.

Page 8

2780
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / September 1992
Vol. 33
system contributions to oscillatory potentials: An explanation
for the conditioning flash effect. Vision Res 27:859, 1987.
11. Speros P and Price J: Oscillatory potentials. History, tech-
niques and potential use in the evaluation of disturbances of
retinal circulation. Surv Ophthalmol 25:237, 1981.
12. Algvere P and Westbeck S: Human ERG in response to double
flashes of light during the course of dark adaptations: A Fourier
analysis of the oscillatory potentials. Vision Res 12:195, 1972.
13. Wachtmeister L: On the oscillatory potentials of the human
electroretinogram in light and dark adaptation. III. Thresholds
and relation to stimulus intensity on adaptation to background
light. Acta Ophthalmol 51:95, 1973.
14. Thomas S, Greenstein V, Blaustein H, Hood D, and Carr R;
Rod system sensitivity in diabetics. Invest Ophthalmol Vis Sci
31(suppl):423, 1990.
15. Greenstein V, Thomas S, Blaustein H, Koenig K, and Carr R:
Rod system sensitivity in patients with early diabetic retinopa-
thy. Submitted for publication.
16. Hood DC and Birch DG: The a-wave of the human ERG and
rod receptor function. Invest Ophthalmol Vis Sci 31:2070,
1990.
17. Hood DC and Birch DG: A quantitative measure of the electri-
cal activity of human rod photoreceptors using electroretinogra-
phy. Vis Neurosci 5:379, 1990.
18. Amemiya T: Dark adaptation in diabetes. Ophthalmoiogica
174:322, 1977.
19. Henson DB and North RV: Dark adaptation in diabetes melli-
tus. Br J Ophthalmol 63:539, 1979.
20. Wachtmeister L and Dowling JE: The oscillatory potentials of
the electroretinogram (ERG). Invest Ophthalmol Vis Sci
17:1176,1978.
21. Wachtmeister L: Further studies of the chemical sensitivity of
the oscillatory potentials of the electroretinogram (ERG) I.
GABA- and glycine antagonists. Acta Ophthalmol 58:712,
1980.
22. Wachtmeister L: Further studies of the chemical sensitivity of
the oscillatory potentials of the electroretinogram (ERG) II.
Glutamate-aspartate and dopamine antagonists. Acta Ophthal-
mol 59:247, 1981.
23. Wachtmeister L: Further studies of the chemical sensitivity of
the oscjllatory potentials of the electroretinogram (ERG) III.
Some amino antagonists. Acta Ophthalmol 59:609, 1981.
24. Heckenlively JR, Martin DA, and Rosenbaum AL: Loss of
electrbretinographic oscillatory potentials, optic atrophy, and
dysplasia in congenital stationary night blindness. Am J Oph-
thalmol 96:526, 1983.
25. LaChapelle P, Little JM, and Polomeno RC: The photopic elec-
troretinogram in congenital stationary night blindness with my-
opia. Invest Ophthalmol Vis Sci 24:442, 1983.
26. Coupland SG: Oscillatory potential changes related to stimulus
intensity and light adaptation. Doc Ophthalmol 66:195, 1987.
27. Peachey NS, Alexander KR, Derlacki DJ, Bobak P, and Fish-
man GA: Effects of light adaptation on the response character-
istics of human oscillatory potentials. Electroencephalogr Clin
Neurophysiol 73:27, 1991.
28. Brodie SE, Daly R, Schumer RA, Wang R-F, and Podos SM:
Abnormalities of the electroretinographic oscillatory potentials
in glaucomatous monkeys. Invest Ophthalmol Vis Sci
32(sup'pl):811, 1991.
29. Wachtmeister L: Basic research and clinical aspects of the oscil-
latory potentials of the electroretinogram Doc Ophthalmol
66:187, 1987.
30. Dowling JE: The Retina: An Approachable Part of the Brain.
Cambridge, MA, The Belknap Press of Harvard University
Press, 1987.
31. Linsenmeier RA: Electrophysiological consequences of retinal
hypoxia. Graefes Arch Clin Exp Ophthalmbl 228:143, 1990.
32. Linsenmeier RA, Mines AH, and Steinberg RH: Effects of hyp-
oxia and hypercapnia on the light peak and electroretinogram
of the cat Invest Ophthalmol Vis Sci 24:37, 1983.