September 2000
Volume 41, Issue 10
Free
Visual Neuroscience  |   September 2000
Multifocal Oscillatory Potentials in Type 1 Diabetes without Retinopathy
Author Affiliations
  • Anne Kurtenbach
    From the University Eye Hospital, Department of Pathophysiology of Vision and Neuro-ophthalmology, Tübingen, Germany.
  • Hana Langrova
    From the University Eye Hospital, Department of Pathophysiology of Vision and Neuro-ophthalmology, Tübingen, Germany.
  • Eberhart Zrenner
    From the University Eye Hospital, Department of Pathophysiology of Vision and Neuro-ophthalmology, Tübingen, Germany.
Investigative Ophthalmology & Visual Science September 2000, Vol.41, 3234-3241. doi:
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Anne Kurtenbach, Hana Langrova, Eberhart Zrenner; Multifocal Oscillatory Potentials in Type 1 Diabetes without Retinopathy. Invest. Ophthalmol. Vis. Sci. 2000;41(10):3234-3241.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

purpose. To study multifocal recordings of oscillatory potentials (m-OPs) in diabetic (Type 1) eyes that have no visible fundus alterations, to ascertain whether topographical changes in sensitivity are evident when compared with recordings from control subjects.

methods. The Visual Evoked Response Imaging System (VERIS; EDI, San Diego, CA) system was used to elicit m-OPs from 61 independent areas, subtending the central 30° of the retina, from 24 eyes of 12 patients with diabetes without retinopathy and from 26 eyes of 14 control subjects. For each group of subjects, the mean first- and second-order (first slice) kernel components of the responses for one eye, randomly chosen from each subject, were analyzed and compared for a retinal ring analysis and for an analysis of retinal quadrants.

results. Both first- and second-order kernel responses of the diabetic group show significant delays in the implicit times of some of the m-OPs, compared with those of the control group. No significant changes in amplitude were found. For the first-order component, significant differences are found for both potentials between 5° and 22° eccentricity, for the nasal retina, and for one of the potentials for the remaining retinal areas. In the second-order kernel responses, the differences are significant for two of the three potentials in the midperiphery between 5° and 13° eccentricity, with the central potential being significantly delayed in all rings and quadrants.

conclusions. Patients with diabetes without retinopathy show prolonged latencies in m-OP recordings. This indicates an alteration in inner retinal sensitivity that can be explained by an impaired rod–cone interaction.

Electroretinographic oscillatory potentials (OPs) are high-frequency oscillatory components superimposed on the b-wave of the flash electroretinogram (ERG) and are thought to reflect inner retinal activity. 1 2 3 4 In patients with diabetes mellitus, an alteration of OP amplitude has been most commonly reported, even in the absence of retinal vascular changes 5 6 7 8 and a reduction in amplitude has been proposed to predict the development of proliferative retinopathy. 9 10 11 12 In addition, there are also reports of prolonged latencies in some of the OPs before the reduction in amplitude. 7 8 12 13 14  
Recently, a new method of electrophysiological recording has been developed that allows simultaneous ERG recordings over multiple small retinal areas. 15 The Visual Evoked Response Imaging System (VERIS; EDI, San Mateo, CA) is a multifocal technique that can aid in the detection of areas of retinal dysfunction. In this study we examined multifocal oscillatory potentials (m-OPs) in patients with diabetes without retinopathy and compared their results with those of an age-matched control group, to gain topographical information about sensitivity changes in preretinopathic diabetic eyes. m-OPs are strongly influenced by rod activity: The second-order kernel m-OPs have been shown to be dominated by contributions from rod–cone interactions, whereas the first-order kernel appears to contain an additional component that does not depend on rods. 16 Therefore, in this study, to gain information about these interactions in patients with diabetes, we compared both first- and second-order kernel responses with those found in control subjects. 
Methods
The m-OPs were recorded using the technique of Wu and Sutter. 16  
Stimulus
The stimulus geometry (Fig. 1) consisted of 61 hexagons, presented on a color monitor (75 frames/sec; Iiyama, Nagano, Japan) that stimulated the central 30° of the retina with pseudorandom achromatic flicker according to the binary m-sequence of Sutter. 17 The length of the m-sequence was 213 − 1. Hexagon size was scaled with eccentricity to evoke focal responses of comparable amplitude per stimulus element in the response arrays of normal subjects. A dim red cross presented at the center of the visual field was used for fixation. 
The luminance of the white hexagons was 47 candelas[ cd]/m2 and that of the black 0.2 cd/m2 (99% contrast). The flash intensity was thus 0.63 cd/m2 · sec (maximum luminance divided by frame rate). 16 Because a relatively long flash interval is required for reliable recordings of OPs, three black frames were inserted between consecutive stimulus frames. The m-sequence stimulation rate was therefore 18.75/sec and the mean local flash rate was 9.375/sec, with a base interval for the pseudorandom stimulation of 53.33 msec. The luminance of the surrounding screen area was set to 5.9 cd/m2—i.e., that of the mean stimulus luminance (product of flash intensity and flash rate). With an average pupil diameter of 8 mm, the mean retinal illuminance was thus approximately 300 trolands (td). 
Recording
Pupils were fully dilated to approximately 8 mm with 1.5% tropicamide. The signal was recorded from both eyes simultaneously with DTL fiber electrodes (UniMed Electrode Supplies, Farnham, Surrey, UK) that were positioned on the conjunctiva directly beneath the cornea and attached with its two ends to the lateral and nasal canthus. The reference and ground skin electrodes (Ag-AgCl electrodes) were attached to the ipsilateral temple and forehead, respectively. The signal was amplified (×200,000) and filtered using an amplifier (model 12; Grass, Quincy, MA) with a frequency bandpass of 100 to 1000 Hz. The record was collected in 16 segments, each approximately 30 seconds long. The quality of the recording was controlled by monitoring the raw signal, and segments contaminated by blinks or saccades were discarded and rerecorded. 
Data Analysis
The VERIS Science software program (EDI) for Macintosh (Apple Computer, Cupertino, CA) was used to extract the local responses from the compound signal. First- and second-order (first slice) kernel responses were computed as illustrated in Figure 2 . For each stimulus area the first-order component is the mean response to all the white frames minus the mean response to all black frames in the m-sequence and gives the linear response to the stimulation. Interaction between flashes can also occur, however, due, for example, to adaptive mechanisms, and the second-order (first slice) response component considers the interaction between two flashes. This component is computed from the sum of responses to stimulation from two consecutive hexagons with the same sign, minus the sum of traces obtained when two consecutive hexagons have different signs. An artifact elimination technique 15 was applied once to the raw data. Average responses were calculated for retinal rings concentric on the fovea (Fig. 1 , top) and for a central 7° and four retinal quadrants (Fig. 1 , bottom). 
Subjects
Twenty-four eyes from 12 patients with diabetes and 26 eyes from 14 control subjects (two subjects did not want to have both eyes dilated) were recorded. The patients ranged between 16.6 and 24.3 years of age (mean, 21.1 ± 2.4 years [SD]) and had had diabetes for between 5.6 and 20.6 years (mean, 13.7 ± 4.7 years). All patients had insulin-dependent diabetes type 1 and underwent a complete ophthalmic examination on the day of testing. They had no retinal microvascular changes, evidenced by direct and indirect microscopy and fundus photography. Visual acuity and intraocular pressure were normal. The plasma glucose level was measured shortly before the beginning of the experiment. 
The control subjects were on average 23.7 ± 4.1 years of age (range, 16.5–29.7 years) and had normal visual acuity and color vision when tested by the Farnsworth–Munsell 28-hue test. The research followed the tenets of the Declaration of Helsinki, and informed consent was obtained before recording, after the nature and possible consequences of the study had been explained. 
Statistical Analysis
One eye was chosen at random from each subject and the data averaged for each group of subjects. Because the distribution of data points did not appear to be gaussian, a nonparametric test (Mann–Whitney) was used to test for significance (P ≤ 0.05) between the average traces of the diabetic and control groups. 
Results
First-Order Response Component
Figure 3 (left) shows examples of local first-order m-OP traces from typical responses of a right eye of a control subject (upper traces) and a patient with diabetes (lower traces). Because OPs have a high frequency, the local signals were small, extending on average approximately 75 to 100 nV from peak to trough. On the right, are three-dimensional plots calculated from these stimulus arrays. The height of this plot represents the response amplitude, normalized to the area of the stimulus element that generated it. The response amplitude was estimated by the scalar product method, where each local response was multiplied by a template representing the overall average of all the local responses. The resolution was increased by interpolation, and a spatial averaging procedure was performed, whereby the response of each hexagon was averaged with the mean response of the neighboring six hexagons. The density distributions showed reduced foveal responses and substantial activity out to an approximate 20° eccentricity. 
Figure 3 shows that the size of the m-OP responses was not uniform throughout the retina. To analyze this in more detail, we calculated average responses (in nanovolts per degree 2 ) from retinal areas for each eye, either for retinal rings of equal eccentricity (Fig. 1 , top), or for a central area and four retinal quadrants (Fig. 1 , bottom). The responses were scaled to compensate for stimulus size. For the ring analysis, the central element (ring 1) had a diameter of approximately 2°, ring 2 extended from 1.8° to 7° eccentricity, ring 3 from 5° to 13°, ring 4 from 11° to 22°, and ring 5 from 17° to 30°. In Table 1 (top), the mean latencies for each peak are listed along with their SDs, ranges, and significance levels. The results are depicted in Figure 4 (left). The thick line shows the mean results for the control group and the thin line those of the diabetic group. The m-OPs were most conspicuous in rings 2 through 5, the recording from the fovea being difficult to distinguish from noise. The first-order analysis of the control group shows two oscillatory potentials at approximately 21.8 and 29.7 msec, with the largest amplitudes in rings 2 and 3 (1.8°–13° eccentricity). The implicit times generally decreased by 1 to 2 msec in going from the fovea to 30° eccentricity. The means of the diabetic group also showed two prominent peaks, but the traces were displaced to the right compared with the control group. Although the latency difference was very small (<1.5 msec) the implicit time for the diabetic group was significantly different from that of the control group for several peaks (shown in Fig. 4 by asterisks and listed in Table 1 ). The most prominent changes were found in the second peak of rings 2 through 4 (1.8°–22° eccentricity); both peaks were significantly delayed in rings 3 and 4 (5°–22° eccentricity). Differences in amplitude between the groups were not significant. 
The average responses calculated for a central area of 7°, and four quadrants—a temporal and nasal retinal area and an upper and lower visual field, corresponding to a lower and upper retina, respectively (Fig. 1 , bottom)—are shown in Figure 4 (right). In Table 1 (bottom), we list the means, SDs, ranges, and significance levels for each peak. For the control group, the waveforms were generally largest for the central retina and were similar in amplitude for the four retinal quadrants. The diabetic group showed prolonged latencies for all oscillatory potentials, for all retinal areas. The difference was significant for the first peak in the central and nasal retina and for the second peak in all retinal quadrants. Differences in amplitude between groups were not significant, and latencies within the control group were similar in all areas. 
Second-Order Response Component
In Figure 5 (left) we show examples of responses obtained for local second-order (first slice) analysis of the left eye of a control (upper) and diabetic (lower) subject. In the first order, responses were not uniform throughout the retina. In Figure 5 (right) we show the three-dimensional density plots calculated, as in Figure 3 , from these traces. Compared with the first-order responses, these density plots showed a more marked absence of activity in the central 6° to 7° of the retina. 
The mean results of the ring analysis are listed in Table 2 (top) and are depicted in Figure 6 (left), with significant differences between groups shown by asterisks. The foveal response was, as before, difficult to distinguish from noise, but the remaining rings for both control and diabetic groups showed three prominent oscillatory potentials situated approximately 21.1, 27.5, and 33.8 msec, which are largest in ring 3 (5°–13° eccentricity). As before, the m-OPs were delayed in the diabetic group. The difference was significant for the middle peak in rings 2 through 5 (1.8°–30° eccentricity) and for the last peak in ring 3. 
The results of the quadrant analysis for the second-order component are shown in Table 2 (bottom) and Figure 6 (right). Here, the temporal and upper retina show the largest oscillatory potentials. The middle peak is significantly delayed in all retinal areas for the diabetic group compared with that of the control group. 
The implicit time and amplitudes of the m-OPs were independent of the age of the subjects, the duration of disease, or the plasma glucose level at the time of recording. 
Discussion
In recording OPs by using conventional techniques, single flashes are used in a Ganzfeld field under mesopic conditions and are most probably derived from inner retinal activity. 1 They are thought to arise at the level of the inner plexiform layer 2 and are believed to be due to inhibitory feedback circuits from amacrine to bipolar cells and/or from ganglion cells to amacrine cells. 18 An origin within the bipolar cells themselves has also been postulated. 3 4 Rods play a large role in their generation, and the individual oscillatory peaks are often thought to have different origins, separated into rod-mediated OPs and cone-mediated OPs. 19 20 This concept, however, does not appear to be compatible with m-OP recordings: Wu and Sutter 16 have shown that all OP components undergo latency shifts with increasing rod contribution to the response, but not with increasing cone contribution, which causes only amplitude enlargements. In addition, the second-order OPs are eliminated by a strong rod-bleaching background. These results are interpreted as evidence that second-order m-OPs are dominated by contributions from rod–cone interactions. Under the rod-bleaching condition, the first-order contributions are still present, although with altered waveform and latencies, and are therefore postulated to contain an additional component that does not depend on rods. 
Normal Eyes
The first-order kernel in this experiment, for which we used a mean retinal illuminance of 300 td, showed two dominant peaks situated at approximately 21.8 and 29.7 msec. The topography of the m-OPs (Fig. 3) , unlike multifocal ERG recordings, showed no pronounced foveal activity, and responses were prominent out to approximately 20° eccentricity. The average results of the control group showed the largest potentials between 1.8° and 13° eccentricity (Fig. 4 , left) and were similar in size for all retinal quadrants (Fig. 4 , right). 
The second-order response showed three main peaks at approximately 21.1, 27.5, and 33.8 msec. In this case, the topography and density distributions (Fig. 5) showed a marked absence of foveal response in the central retina. The largest responses were found between 5° and 13° eccentricity (Fig. 6 , left). In addition, the waveforms were larger from the temporal retina compared with those from the nasal retina, and were larger in the upper retina than in the lower retina (Fig. 6 , right). The implicit times of the OPs did not change with retinal quadrant. Such topographic asymmetries have also been reported by Miyake et al. 21 in OPs from normal subjects using focal ERG recording techniques. They suggest that part of this asymmetry may be due to the retina itself. There is a higher density of receptors in upper retinal areas, 22 and the standing potential, reflecting the function of the retinal pigment epithelium, is also larger in the upper retina. 23 Because we found differences in retinal quadrants only in the second-order kernel, which is more dependent on rod activity than the first-order kernel, these factors may be more important in rod than cone function. 
Our observations show responses at eccentricities where both rod and cones were abundant in the retina and are very similar to those reported by Wu and Sutter 16 for two normal subjects. 
Diabetic Eyes
The age of the patients with diabetes, the duration of the disease, and the blood sugar level at the time of the recording were not significantly correlated to the latency or amplitudes of the oscillatory potentials. 
The responses from diabetic eyes exhibited all the features of those from control eyes. Significant differences between the control and diabetic eyes in the first-order component were found for both potentials between 5° and 22° eccentricity and in the nasal retina, and for one potential for the remaining retinal areas except the fovea (ring 1), where the recording was difficult to distinguish from noise. In the second-order kernel responses, the differences were significant for two of the three oscillatory potentials in the midperiphery between 5° and 13° eccentricity, and the central potential was significantly delayed in all rings and quadrants, except the fovea. However, all m-OPs in the first- and second-order responses were delayed in the diabetic group, albeit to a greater or lesser degree. The pattern of defects found here is similar to those found in persons with diabetes, according to automated perimetry, where type 1 diabetic eyes with no retinopathy have been shown to exhibit a diffuse reduction in sensitivity across the visual field. 24  
With standard recordings, it is well documented that OPs are altered in diabetes. Patients without retinopathy can show decreased or even hypernormal OPs, but with progression of disease, the OPs become subnormal and eventually extinguish when proliferative retinopathy is present. 5 6 7 Although oscillatory potential amplitudes are related to the severity of the diabetic retinopathy 9 10 11 12 there is also evidence that a selective delay in peak latency may signify an earlier retinal dysfunction that can be present in diabetics eyes without retinopathy 7 8 12 13 14. Rats with diabetes induced by streptozotocin also first show a prolonged latency in OP2 after 2 to 3 weeks, which is then followed by a reduction in amplitude after approximately 6 weeks. 25 Using the VERIS system, we found that diabetics’ eyes without retinopathy showed significant latency delays in m-OPs but not significant amplitude alterations. 
The finding of altered m-OPs in diabetics’ eyes without retinopathy indicates a dysfunction of the inner retina. If latency alterations are due to changes in rod activity, the delayed responses and their topography indicate that the response of the rod pathway is reduced in these patients. It is known that rods are affected early in diabetes. Patients with and without retinopathy can show altered dark adaptation curves. 26 27 28 A large rod contribution is also apparent in conventional OP recordings, although the exact role of the rods has not yet been clarified. 29 30 31 It has been proposed that individual potentials are either cone or rod generated, 29 but it is probable that this is not the case with m-OPs. 16 Our findings that some of the potentials were more affected by diabetes than others may, however, indicate differences in their generation. Moreover, it has been proposed that retinopathy is due above all to hypoxia of the retina. 32 33 It has recently been postulated that because rods require larger amounts of oxygen than cones, they act, especially in the dark-adapted state, as an oxygen sink, imposing on the inner retina an additional hypoxia. 34 They may thus be one of the first receptors to be affected by high glucose levels, although most functional changes are due to alterations of inner retinal activity. Most of the patients examined here have been previously studied, and belong to a group of patients with diabetes who show an altered brightness perception and color vision, 35 36 which can also be explained by postreceptoral alterations. S-cone pathway deficits are the most commonly found color perception alteration in patients with diabetes without retinopathy. 36 37 38 39 40 The results of multifocal ERG recordings also show early functional alterations of inner retinal activity in preretinopathic diabetic eyes 41 ; however, there is evidence that the cone system in the outer and/or middle retina is additionally compromised in diabetic eyes without retinopathy. 42  
In conclusion, the results of this study show that some of the m-OPs in patients with diabetes without retinopathy are significantly prolonged compared with those from an age-matched control group. This can be explained by an alteration of rod–cone interactions in these patients. 
 
Figure 1.
 
Geometry of the hexagonal array of 61 stimulus elements. The concentric circles indicate the eccentricity of the elements on the retina. For analysis, responses were averaged from either five concentric retinal rings (top) or from a central area of 7° and four retinal quadrants (bottom).
Figure 1.
 
Geometry of the hexagonal array of 61 stimulus elements. The concentric circles indicate the eccentricity of the elements on the retina. For analysis, responses were averaged from either five concentric retinal rings (top) or from a central area of 7° and four retinal quadrants (bottom).
Figure 2.
 
Derivation of first- and second-order kernel responses. First order summates responses to white hexagons and subtracts responses to black hexagons. Second-order (first slice) summates responses when two consecutive hexagons have the same sign and subtracts responses when two consecutive hexagons have different signs.
Figure 2.
 
Derivation of first- and second-order kernel responses. First order summates responses to white hexagons and subtracts responses to black hexagons. Second-order (first slice) summates responses when two consecutive hexagons have the same sign and subtracts responses when two consecutive hexagons have different signs.
Figure 3.
 
Left: Examples of first-order traces obtained from a control subject (top) and a patient with diabetes (bottom). Right: Interpolated three-dimensional plots of the response densities. The response amplitude is normalized to the area of the stimulus element that generated it.
Figure 3.
 
Left: Examples of first-order traces obtained from a control subject (top) and a patient with diabetes (bottom). Right: Interpolated three-dimensional plots of the response densities. The response amplitude is normalized to the area of the stimulus element that generated it.
Table 1.
 
First-Order Kernel Latencies
Table 1.
 
First-Order Kernel Latencies
Peak 1 Peak 2
Control Diabetic Control Diabetic
Ring analysis
Ring 2
Mean± SD 22.41 ± 0.99 23.06 ± 1.02 30.04 ± 1.36 31.07 ± 0.91
Range 20.8–24.1 20.8–24.1 27.5–32.5 29.1–32.5
P NS 0.045
Ring 3
Mean± SD 22.01 ± 0.97 23.03 ± 0.89 29.52 ± 1.06 30.66 ± 0.69
Range 20.0–23.3 21.6–25.0 28.3–30.8 29.1–31.6
P 0.016 0.078
Ring 4
Mean± SD 21.75 ± 0.79 22.55 ± 0.86 29.46 ± 1.06 30.45 ± 0.79
Range 20.0–23.3 20.8–24.1 27.5–30.8 29.1–31.6
P 0.024 0.027
Ring 5
Mean± SD 21.12 ± 0.82 22.32 ± 1.12 29.90 ± 1.21 30.45 ± 1.16
Range 20.0–23.3 20.8–24.1 28.3–32.5 28.3–32.5
P 0.005 NS
Quadrant analysis
Fovea
Mean± SD 22.22 ± 0.95 22.95 ± 0.92 30.02 ± 1.41 30.86 ± 0.97
Range 20.8–24.1 20.8–24.1 27.5–32.3 28.3–31.6
P 0.042 NS
Lower retina
Mean± SD 21.70 ± 0.95 22.48 ± 0.97 29.65 ± 1.10 30.57 ± 0.95
Range 20.0–23.3 20.8–25.0 27.5–31.6 29.1–32.5
P NS 0.050
Temporal retina
Mean± SD 21.51 ± 0.86 22.26 ± 1.03 29.58 ± 1.15 30.65 ± 1.00
Range 20.0–22.5 20.8–24.1 27.5–31.6 29.1–32.5
P NS 0.036
Upper retina
Mean± SD 21.64 ± 1.04 22.36 ± 0.73 29.39 ± 1.23 30.59 ± 0.76
Range 20.0–23.3 21.6–24.1 27.5–31.6 29.1–31.6
P NS 0.017
Nasal retina
Mean± SD 21.82 ± 0.82 22.84 ± 0.89 29.65 ± 1.34 30.80 ± 0.68
Range 20.0–23.3 21.6–24.1 28.3–32.5 30.0–31.6
P 0.014 0.030
Figure 4.
 
Mean waveforms obtained for the first-order response component when averaged over concentric retinal areas (left) and from a central area of 7° and four retinal quadrants (right). Traces were scaled according to response density. *Significant differences between the two groups.
Figure 4.
 
Mean waveforms obtained for the first-order response component when averaged over concentric retinal areas (left) and from a central area of 7° and four retinal quadrants (right). Traces were scaled according to response density. *Significant differences between the two groups.
Figure 5.
 
Left: Examples of second-order traces obtained from a control subject (top) and a patient with diabetes (bottom). Right: Interpolated three-dimensional plots of the response densities. The response amplitude is normalized to the area of the stimulus element that generated it.
Figure 5.
 
Left: Examples of second-order traces obtained from a control subject (top) and a patient with diabetes (bottom). Right: Interpolated three-dimensional plots of the response densities. The response amplitude is normalized to the area of the stimulus element that generated it.
Table 2.
 
Second-Order Kernel Latencies
Table 2.
 
Second-Order Kernel Latencies
Peak 1 Peak 2 Peak 3
Control Diabetic Control Diabetic Control Diabetic
Ring analysis
Ring 2
Mean± SD 21.57 ± 1.21 22.18 ± 1.64 27.90 ± 1.07 29.08 ± 0.87 34.44 ± 1.15 34.73 ± 0.84
Range 19.1–23.3 20.0–25.0 25.8–29.1 27.5–30.0 32.5–36.6 33.3–36.6
P NS 0.017 NS
Ring 3
Mean± SD 21.12 ± 0.71 21.70 ± 1.03 27.53 ± 0.93 28.83 ± 0.77 33.98 ± 1.10 34.63 ± 0.74
Range 20.0–22.5 20.8–24.1 25.8–29.1 27.5–30.0 32.5–36.6 33.3–35.8
P NS 0.003 0.047
Ring 4
Mean± SD 21.05 ± 0.67 21.35 ± 0.63 27.60 ± 0.72 28.37 ± 0.61 33.48 ± 0.78 33.83 ± 0.68
Range 20.0–22.5 20.8–22.5 26.6–29.1 27.5–29.1 32.5–34.9 32.5–34.9
P NS 0.016 NS
Ring 5
Mean± SD 20.80 ± 0.31 21.34 ± 0.52 27.02 ± 0.83 28.10 ± 0.84 33.30 ± 0.77 33.77 ± 0.86
Range 20.0–21.6 20.8–22.5 25.8–28.3 26.6–30.0 32.5–34.9 31.6–35.0
P NS 0.009 NS
Quadrant analysis
Fovea
Mean ± SD 21.51 ± 1.18 22.18 ± 1.64 27.59 ± 0.86 29.08 ± 0.87 34.34 ± 1.10 34.67 ± 0.86
Range 19.1–23.3 20.0–25.0 25.8–28.3 27.5–30.0 32.5–36.6 33.3–36.6
P NS 0.001 NS
Lower retina
Mean± SD 21.31 ± 0.76 21.61 ± 0.35 27.73 ± 0.61 28.68 ± 0.82 33.55 ± 0.99 34.13 ± 1.16
Range 20.0–22.5 20.8–22.5 26.6–28.3 27.5–30.0 31.6–35.8 32.5–35.8
P NS 0.006 NS
Temporal retina
Mean± SD 20.74 ± 0.49 21.28 ± 0.72 27.27 ± 0.68 28.23 ± 0.61 33.42 ± 0.82 34.12 ± 0.89
Range 20.0–21.6 20.0–22.5 26.6–28.3 27.5–29.1 32.5–34.9 32.5–35.8
P NS 0.004 NS
Upper retina
Mean± SD 20.87 ± 0.60 21.21 ± 1.03 27.35 ± 0.80 28.03 ± 0.52 33.42 ± 0.89 33.98 ± 0.74
Range 20.0–22.5 20.0–24.1 25.8–28.3 26.6–28.3 31.6–34.9 32.5–35.0
P NS 0.015 NS
Nasal retina
Mean± SD 21.37 ± 0.68 21.17 ± 0.73 27.65 ± 0.99 28.63 ± 0.94 33.42 ± 1.05 33.77 ± 0.99
Range 20.0–22.5 20.0–22.5 25.8–29.1 26.6–30.0 31.6–34.9 31.6–35.8
P NS 0.035 NS
Figure 6.
 
Waveforms obtained for the second-order response component when averaged over concentric retinal circles (left) and when averaged from a central area and four retinal quadrants (right). Traces were scaled according to response density. *Significant differences between the two groups.
Figure 6.
 
Waveforms obtained for the second-order response component when averaged over concentric retinal circles (left) and when averaged from a central area and four retinal quadrants (right). Traces were scaled according to response density. *Significant differences between the two groups.
The authors thank Hartmut Schwahn, Jutta Isensee, and Katrin Götz for help in recording. 
Brindley GS. Responses to illumination recorded by microelectrodes from the frog‘s retina. J Physiol (Lond). 1956;134:360–384. [CrossRef] [PubMed]
Ogden TE. The oscillatory waves of the primate electroretinogram. Vision Res. 1973;13:1059–1074. [CrossRef] [PubMed]
Heynen H, van Norren D. Origin of the electroretinogram in the intact macaque eye, II: current source-density analysis. Vision Res. 1985;25:709–715. [CrossRef] [PubMed]
Wachtmeister L, Dowling JE. The oscillatory potentials of the mudpuppy retina. Invest Ophthalmol Vis Sci. 1978;17:1176–1188. [PubMed]
Simonsen SE. ERG in diabetics: clinical values of electroretinography. Francois J eds. Proceedings of the XXth International Congress of Ophthalmology Symposium, Ghent, 1966. 1968;403–412. Karger Basel, Switzerland.
Henkes HE, Houtsmüller AJ. Fundus diabeticus: an evaluation of the preretinopathic stage. Am J Ophthalmol. 1965;60:662–670. [CrossRef] [PubMed]
Yonemura D, Kawasaki K. Electrophysiological study on activities of neural and non-neural retinal elements in man with special reference to its clinical application. Jpn J Ophthalmol. 1978;22:195–213.
Holopigian K, Seiple W, Lorenzo M, Carr R. A comparison of photopic and scotopic electroretinographic changes in early diabetic retinopathy. Invest Ophthalmol Vis Sci. 1992;33:2773–2780. [PubMed]
Yonemura D, Aoki T, Tsuzuki K. Electroretinogram in diabetic retinopathy. Arch Ophthalmol. 1962;68:19–24. [CrossRef] [PubMed]
Simonsen SE. The value of the oscillatory potential in selecting juvenile diabetics at risk of developing proliferative retinopathy. Acta Ophthalmol. 1980;58:865–878.
Bresnick GH. Electroretinographic oscillatory potentials predict progression of diabetic retinopathy. Arch Ophthalmol. 1984;102:1307–1311. [CrossRef] [PubMed]
Bresnick GH, Palta M. Oscillatory potential amplitudes. Arch Ophthalmol. 1987;105:929–933. [CrossRef] [PubMed]
Shirao Y, Okumura T, Ohta T, Kawasaki K. Clinical importance of electroretinographic oscillatory potential in early detection and objective evaluation for diabetic retinopathy. Clin Vis Sci. 1991;6:445–450.
Yoshida A, Kojima M, Ogasawara H, Ishiko S. Oscillatory potentials and permeability of the blood-retinal barrier in noninsulin-dependent diabetic patients without retinopathy. Ophthalmology. 1991;98:1266–1271. [CrossRef] [PubMed]
Sutter EE, Tran D. The field topography of ERG components in man, I: the photopic luminance response. Vision Res. 1992;32:433–446. [CrossRef] [PubMed]
Wu S, Sutter EE. A topographic study of oscillatory potentials in man. Vis Neurosci. 1995;12:1013–1025. [CrossRef] [PubMed]
Sutter EE. The fast m-transform: a fast computation of cross-correlations with binary m- sequences. SIAM J Comput. 1991;20:686–694. [CrossRef]
Brown KT. The electroretinogram: its components and their origin. Vision Res. 1968;8:633–677. [CrossRef] [PubMed]
King–Smith PE, Loffing DH, Jones R. Rod and cone ERGs and their oscillatory potentials. Invest Ophthalmol Vis Sci. 1986;27:270–273. [PubMed]
Peachey NS, Alexander KR, Fishman GA. Rod and cone system contributions to oscillatory potentials: an explanation for the conditioning flash. Vision Res. 1987;27:859–866. [CrossRef] [PubMed]
Miyake Y, Shiroyama N, Horiguchi M, Ota I. Asymmetry of focal ERG in human macular region. Invest Ophthalmol Vis Sci. 1989;30:1743–1749. [PubMed]
Osterberg G. Topography of the layer of rods and cones in the human retina. Acta Ophthalmol. 1935;6(suppl)1–102.
Skrandies W, Baier M. The standing potential of the human retina. Vision Res. 1986;26:577–581. [CrossRef] [PubMed]
Trick GL, Trick LR, Kilo C. Visual field defects in patients with insulin-dependent and noninsulin-dependent diabetes. Ophthalmology. 1990;97:475–482. [CrossRef] [PubMed]
Sakai H, Tani Y, Shirasawa E, Shirao Y, Kawasaki K. Development of electroretinographic alterations in streptozotocin-induced diabetes in rats. Ophthalmic Res. 1995;27:57–63. [CrossRef] [PubMed]
Henson D B, North R. Dark adaptation in diabetes mellitus. Br J Ophthalmol. 1979;63:539–541. [CrossRef] [PubMed]
Amemiya T. Dark adaptation in diabetics. Ophthalmologica. 1977;174:322–326. [CrossRef] [PubMed]
Greenstein VC, Thomas SR, Blaustein H, Koenig K, Carr RE. Effects of early diabetic retinopathy on rod system sensitivity. Optom Vis Sci. 1993;37:1140–1148.
King–Smith PE, Loffing DH, Jones R. Rod and cone ERGs and their oscillatory potentials. Invest Ophthalmol Vis Sci. 1986;27:270–273. [PubMed]
Peachey NS, Alexander KR, Fishman GA. Rod and cone system contributions to oscillatory potentials: an explanation for the conditioning flash. Vision Res. 1997;27:859–866.
Lachapelle P, Benoit J, Blain L, Giute P, Roy MS. The oscillatory potentials in response to stimuli of photopic intensities delivered in dark adaptation: an explanation for the conditioning flash effect. Vision Res. 1990;30:503–513. [CrossRef] [PubMed]
Ditzel J, Standl E. The problem of tissue oxygenation in diabetes mellitus: I: its relation to the early functional changes in the microcirculation of diabetic subjects. Acad Med Scand. 1975;578(Suppl)49–58.
Bresnick GH, DeVenecia G, Myers FL, Harris JA, Davis MD. Retinal ischemia in diabetic retinopathy. Arch Ophthalmol. 1975;93:1300–1310. [CrossRef] [PubMed]
Arden GB, Wolf JE, Tsang Y. Does dark adaptation exacerbate diabetic retinopathy? Evidence and a linking hypothesis. Vision Res. 1998;38:1723–1729. [CrossRef] [PubMed]
Kurtenbach A, Schiefer U, Neu A, Zrenner E. Development of brightness matching and colour vision deficits in juvenile diabetics. Vision Res. 1999;39:1221–1229. [CrossRef] [PubMed]
Kurtenbach A, Schiefer U, Neu A, Zrenner E. Preretinopic changes in the colour vision of juvenile diabetics. Br J Ophthalmol. 1999;83:43–46. [CrossRef] [PubMed]
Kinnear PR, Aspinall PA, Lakowski R. The diabetic eye and colour vision. Trans Ophthalmol Soc UK. 1972;92:69–78. [PubMed]
Zisman F, Adams A. Spectral sensitivity of cone mechanisms in juvenile diabetics. Doc Ophthalmol Vis Sci. 1982;33:127–131.
Greene FD, Ghafour IM, Allan D. Colour vision of diabetics. Br J Ophthalmol. 1985;69:533–536. [CrossRef] [PubMed]
Trick G L, Burde R M, Gordon MO, Santiago JV, Kilo C. The relationship between hue discrimination and contrast sensitivity in patients with diabetes mellitus. Ophthalmology. 1988;95:693–698. [CrossRef] [PubMed]
Palmowski AM, Sutter EE, Bearse MA, Fung W. Mapping of retinal function in diabetic retinopathy using the multifocal electroretinogram. Invest Ophthalmol Vis Sci. 1997;38:2586–2596. [PubMed]
Fortune B, Schneck ME, Adams AJ. Multifocal electroretinogram delays reveal local retinal dysfunction in early diabetic retinopathy. Invest Ophthalmol Vis Sci. 1999;40:2638–2651. [PubMed]
Figure 1.
 
Geometry of the hexagonal array of 61 stimulus elements. The concentric circles indicate the eccentricity of the elements on the retina. For analysis, responses were averaged from either five concentric retinal rings (top) or from a central area of 7° and four retinal quadrants (bottom).
Figure 1.
 
Geometry of the hexagonal array of 61 stimulus elements. The concentric circles indicate the eccentricity of the elements on the retina. For analysis, responses were averaged from either five concentric retinal rings (top) or from a central area of 7° and four retinal quadrants (bottom).
Figure 2.
 
Derivation of first- and second-order kernel responses. First order summates responses to white hexagons and subtracts responses to black hexagons. Second-order (first slice) summates responses when two consecutive hexagons have the same sign and subtracts responses when two consecutive hexagons have different signs.
Figure 2.
 
Derivation of first- and second-order kernel responses. First order summates responses to white hexagons and subtracts responses to black hexagons. Second-order (first slice) summates responses when two consecutive hexagons have the same sign and subtracts responses when two consecutive hexagons have different signs.
Figure 3.
 
Left: Examples of first-order traces obtained from a control subject (top) and a patient with diabetes (bottom). Right: Interpolated three-dimensional plots of the response densities. The response amplitude is normalized to the area of the stimulus element that generated it.
Figure 3.
 
Left: Examples of first-order traces obtained from a control subject (top) and a patient with diabetes (bottom). Right: Interpolated three-dimensional plots of the response densities. The response amplitude is normalized to the area of the stimulus element that generated it.
Figure 4.
 
Mean waveforms obtained for the first-order response component when averaged over concentric retinal areas (left) and from a central area of 7° and four retinal quadrants (right). Traces were scaled according to response density. *Significant differences between the two groups.
Figure 4.
 
Mean waveforms obtained for the first-order response component when averaged over concentric retinal areas (left) and from a central area of 7° and four retinal quadrants (right). Traces were scaled according to response density. *Significant differences between the two groups.
Figure 5.
 
Left: Examples of second-order traces obtained from a control subject (top) and a patient with diabetes (bottom). Right: Interpolated three-dimensional plots of the response densities. The response amplitude is normalized to the area of the stimulus element that generated it.
Figure 5.
 
Left: Examples of second-order traces obtained from a control subject (top) and a patient with diabetes (bottom). Right: Interpolated three-dimensional plots of the response densities. The response amplitude is normalized to the area of the stimulus element that generated it.
Figure 6.
 
Waveforms obtained for the second-order response component when averaged over concentric retinal circles (left) and when averaged from a central area and four retinal quadrants (right). Traces were scaled according to response density. *Significant differences between the two groups.
Figure 6.
 
Waveforms obtained for the second-order response component when averaged over concentric retinal circles (left) and when averaged from a central area and four retinal quadrants (right). Traces were scaled according to response density. *Significant differences between the two groups.
Table 1.
 
First-Order Kernel Latencies
Table 1.
 
First-Order Kernel Latencies
Peak 1 Peak 2
Control Diabetic Control Diabetic
Ring analysis
Ring 2
Mean± SD 22.41 ± 0.99 23.06 ± 1.02 30.04 ± 1.36 31.07 ± 0.91
Range 20.8–24.1 20.8–24.1 27.5–32.5 29.1–32.5
P NS 0.045
Ring 3
Mean± SD 22.01 ± 0.97 23.03 ± 0.89 29.52 ± 1.06 30.66 ± 0.69
Range 20.0–23.3 21.6–25.0 28.3–30.8 29.1–31.6
P 0.016 0.078
Ring 4
Mean± SD 21.75 ± 0.79 22.55 ± 0.86 29.46 ± 1.06 30.45 ± 0.79
Range 20.0–23.3 20.8–24.1 27.5–30.8 29.1–31.6
P 0.024 0.027
Ring 5
Mean± SD 21.12 ± 0.82 22.32 ± 1.12 29.90 ± 1.21 30.45 ± 1.16
Range 20.0–23.3 20.8–24.1 28.3–32.5 28.3–32.5
P 0.005 NS
Quadrant analysis
Fovea
Mean± SD 22.22 ± 0.95 22.95 ± 0.92 30.02 ± 1.41 30.86 ± 0.97
Range 20.8–24.1 20.8–24.1 27.5–32.3 28.3–31.6
P 0.042 NS
Lower retina
Mean± SD 21.70 ± 0.95 22.48 ± 0.97 29.65 ± 1.10 30.57 ± 0.95
Range 20.0–23.3 20.8–25.0 27.5–31.6 29.1–32.5
P NS 0.050
Temporal retina
Mean± SD 21.51 ± 0.86 22.26 ± 1.03 29.58 ± 1.15 30.65 ± 1.00
Range 20.0–22.5 20.8–24.1 27.5–31.6 29.1–32.5
P NS 0.036
Upper retina
Mean± SD 21.64 ± 1.04 22.36 ± 0.73 29.39 ± 1.23 30.59 ± 0.76
Range 20.0–23.3 21.6–24.1 27.5–31.6 29.1–31.6
P NS 0.017
Nasal retina
Mean± SD 21.82 ± 0.82 22.84 ± 0.89 29.65 ± 1.34 30.80 ± 0.68
Range 20.0–23.3 21.6–24.1 28.3–32.5 30.0–31.6
P 0.014 0.030
Table 2.
 
Second-Order Kernel Latencies
Table 2.
 
Second-Order Kernel Latencies
Peak 1 Peak 2 Peak 3
Control Diabetic Control Diabetic Control Diabetic
Ring analysis
Ring 2
Mean± SD 21.57 ± 1.21 22.18 ± 1.64 27.90 ± 1.07 29.08 ± 0.87 34.44 ± 1.15 34.73 ± 0.84
Range 19.1–23.3 20.0–25.0 25.8–29.1 27.5–30.0 32.5–36.6 33.3–36.6
P NS 0.017 NS
Ring 3
Mean± SD 21.12 ± 0.71 21.70 ± 1.03 27.53 ± 0.93 28.83 ± 0.77 33.98 ± 1.10 34.63 ± 0.74
Range 20.0–22.5 20.8–24.1 25.8–29.1 27.5–30.0 32.5–36.6 33.3–35.8
P NS 0.003 0.047
Ring 4
Mean± SD 21.05 ± 0.67 21.35 ± 0.63 27.60 ± 0.72 28.37 ± 0.61 33.48 ± 0.78 33.83 ± 0.68
Range 20.0–22.5 20.8–22.5 26.6–29.1 27.5–29.1 32.5–34.9 32.5–34.9
P NS 0.016 NS
Ring 5
Mean± SD 20.80 ± 0.31 21.34 ± 0.52 27.02 ± 0.83 28.10 ± 0.84 33.30 ± 0.77 33.77 ± 0.86
Range 20.0–21.6 20.8–22.5 25.8–28.3 26.6–30.0 32.5–34.9 31.6–35.0
P NS 0.009 NS
Quadrant analysis
Fovea
Mean ± SD 21.51 ± 1.18 22.18 ± 1.64 27.59 ± 0.86 29.08 ± 0.87 34.34 ± 1.10 34.67 ± 0.86
Range 19.1–23.3 20.0–25.0 25.8–28.3 27.5–30.0 32.5–36.6 33.3–36.6
P NS 0.001 NS
Lower retina
Mean± SD 21.31 ± 0.76 21.61 ± 0.35 27.73 ± 0.61 28.68 ± 0.82 33.55 ± 0.99 34.13 ± 1.16
Range 20.0–22.5 20.8–22.5 26.6–28.3 27.5–30.0 31.6–35.8 32.5–35.8
P NS 0.006 NS
Temporal retina
Mean± SD 20.74 ± 0.49 21.28 ± 0.72 27.27 ± 0.68 28.23 ± 0.61 33.42 ± 0.82 34.12 ± 0.89
Range 20.0–21.6 20.0–22.5 26.6–28.3 27.5–29.1 32.5–34.9 32.5–35.8
P NS 0.004 NS
Upper retina
Mean± SD 20.87 ± 0.60 21.21 ± 1.03 27.35 ± 0.80 28.03 ± 0.52 33.42 ± 0.89 33.98 ± 0.74
Range 20.0–22.5 20.0–24.1 25.8–28.3 26.6–28.3 31.6–34.9 32.5–35.0
P NS 0.015 NS
Nasal retina
Mean± SD 21.37 ± 0.68 21.17 ± 0.73 27.65 ± 0.99 28.63 ± 0.94 33.42 ± 1.05 33.77 ± 0.99
Range 20.0–22.5 20.0–22.5 25.8–29.1 26.6–30.0 31.6–34.9 31.6–35.8
P NS 0.035 NS
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×