Electroretinographic findings associated with panretinal photocoagulation (PRP) versus PRP plus intravitreal ranibizumab treatment for high-risk proliferative diabetic retinopathy
- First Online:
- Cite this article as:
- Messias, A., Filho, J.A.R., Messias, K. et al. Doc Ophthalmol (2012) 124: 225. doi:10.1007/s10633-012-9322-5
- 324 Views
To evaluate changes in electroretinographic (ERG) findings after panretinal photocoagulation (PRP) compared to PRP plus intravitreal injection of ranibizumab (IVR) in eyes with high-risk proliferative diabetic retinopathy (PDR). Patients with high-risk PDR and no prior laser treatment were assigned randomly to receive PRP (PRP group; n = 9) or PRP plus IVR (PRPplus group; n = 11). PRP was administered in two sessions (weeks 0 and 2), and IVR was administered at the end of the first laser session (week 0) in the PRPplus group. Standardized ophthalmic evaluations including (ETDRS) best-corrected visual acuity (BCVA), and fluorescein angiography to measure area of fluorescein leakage (FLA), were performed at baseline and at weeks 16 (±2), 32 (±2) and 48 (±2). ERG was measured according to ISCEV standards at baseline and at week 48 (±2). At 48 weeks, 2,400–3,000 laser spots had been placed in eyes in the PRP group, while only 1,400–1,800 spots had been placed in the PRPplus group. Compared to baseline, there was a statistically significant (P < 0.05) FLA reduction observed at all study visits in both groups, with the reduction observed in the PRPplus group significantly larger than that in the PRP group at week 48. ROD b-wave amplitude was significantly reduced to 46 ± 5 % (P < 0.05) of baseline in the PRP group and 64 ± 6 % (P < 0.05) in the PRPplus group. This reduction was significantly larger in the PRP group than in the PRPplus group (P = 0.024; t Test). Similar results were observed for the dark-adapted Combined Response (CR) b-wave amplitude, with a reduction at 48 weeks compared to baseline of 45 ± 4 % in the PRP group and 62 ± 5 % in the PRPplus group; the reduction in CR b-wave amplitude was significantly larger in the PRP group than in the PRPplus group (P = 0.0094). CR a-wave, oscillatory potentials, cone single flash, and 30 Hz flicker responses showed statistically significant within-group reductions, but no differences in between-group analyses. These results suggest that treating high-risk PDR with PRP plus IVR is effective for PDR control, and permits the use of less extensive PRP which, in turn, induces less retinal functional loss, in particular for rod-driven post-receptoral responses, than treatment with PRP alone.
KeywordsElectroretinographyAngiogenesisDiabetesLaser treatmentVascular endothelial growth factor (VEGF)
Retinal neovascularization (NV) represents an important risk factor for severe vision loss in patients with diabetic retinopathy [1, 2]. Approximately 60 % of patients with proliferative diabetic retinopathy (PDR) respond to panretinal photocoagulation (PRP) [3, 4], which is a destructive and painful procedure and is commonly associated with decreased peripheral vision and increased risk of macular edema .
Vascular endothelial growth factor (VEGF) has been associated in the pathogenesis of PDR and other eye diseases characterized by NV [6–9]. Levels of VEGF are increased in the vitreous of eyes with PDR. Blockage of VEGF has been associated with inhibition NV formation [10–12]. Several clinical trials are currently evaluating the role of anti-VEGF agents for the treatment of retinal NV in diabetic patients with encouraging preliminary results [13–15].
Combining anti-VEGF therapy with PRP has been proposed as a promising alternative for PRP alone for high-risk PDR [16–18], and for the management of diabetic macular edema in eyes also receiving PRP . Our group recently reported that intravitreal ranibizumab after PRP is associated with a larger reduction in fluorescein angiography leakage (FLA) compared with PRP alone in eyes with high-risk PDR, with potential protection against the visual acuity loss and macular swelling that has been observed following PRP alone .
Electroretinography (ERG) has been used to detect early functional changes and predict progression in diabetic retinopathy , to evaluate retinal ischemia , and to record retinal function loss after PRP [23–25]. It has been shown that PRP reduces the amplitudes of a- and b-waves of the ERG .
In a previous study, we demonstrated that less extensive PRP is needed to control high-risk PDR if the combination of PRP and IVR is used , and there is also evidence that intravitreal anti-VEGF drugs can improve ERG responses if associated with reduction of retinal edema .
In the current study, we aim to evaluate changes in electroretinographic (ERG) findings after PRP compared to PRP plus intravitreal injection of ranibizumab (IVR) in eyes with high-risk PDR.
The study protocol adhered to the tenets of the Declaration of Helsinki and was approved by the local Institutional Review Board; all participants gave written informed consent before entering the study.
Patient eligibility and baseline evaluation
Patients were included if they had high-risk PDR, which was defined according to the guidelines of the Early Treatment Diabetic Retinopathy Study [2, 27] as follows: (1) presence of NV at the disc (NVD) greater than ETDRS standard photograph 10A, or (2) presence of NVD associated with vitreous or pre-retinal hemorrhage, or (3) NV elsewhere (NVE) with more than a half disk area associated with vitreous or pre-retinal hemorrhage.
Exclusion criteria were the following: (1) history of prior laser treatment or vitrectomy in the study eye; (2) history of thromboembolic event—including myocardial infarction or cerebrovascular accident; (3) major surgery within the prior 6 months or planned within the next 28 days; (4) uncontrolled hypertension (according to guidelines of the seventh report of the joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure [JNC-7] ); (5) known coagulation abnormalities or current use of anticoagulative medication other than aspirin; or (6) any condition affecting documentation.
Standardized ophthalmic evaluations including (ETDRS) best-corrected visual acuity (BCVA) and fluorescein angiography were performed at baseline and at weeks 16 (±2), 32 (±2) and 48 (±2); ERGs were performed at baseline and at week 48 (±2).
ERG was performed using an extended protocol based on the ISCEV standard , using the Espion E2 recording unit and the ColorDome (Diagnosys LLC, Middleton, MA, USA) as Ganzfeld LED stimulator, using DTL electrodes.
After dark-adaptation and pupil dilatation, a series of increasing light intensity flashes (from 0.001 to 0.1 cd.s/m2, including the ROD-response (0.01 cd.s/m2)) were used to allow estimation of parameters Vmax (saturation amplitude) and k (luminance for reaching ½ Vmax), considered for evaluation of the retinal scotopic sensitivity .
Subsequently, responses to the Combined Response (CR) stimulus (3 cd.s/m2) and the High Intensity (HI) response stimulus (10 cd.s/m2) were recorded to allow evaluations of the a-wave and the oscillatory potentials (OPs). OPs were filtered from the CR using a built-in band pass filter set between 75 and 100 Hz; the wavelet’s area under the curve was used for analysis.
Thereafter, eyes were light adapted for 10 min using a light background of 30 cd/m2 with the same Ganzfeld unit. Cone single flash (3 cd.s/m2; background of 30 cd/m2) and 30 Hz flicker responses (3 cd.s/m2 at 30 Hz; background of 30 cd/m2) were recorded.
All patients received PRP performed in two sessions (at week 0 and week 2) according to ETDRS guidelines . Six to eight hundred 500 μm spots were applied per session, at the discretion of the treating investigator. If eyes showed clinically significant macular edema , macular focal/grid laser was performed at the time of the first PRP session. Patients could be retreated with focal laser at the 16-week and 32-week study visits.
If both eyes of one patient were eligible for the study, the eye with best visual acuity was included. Patients were enrolled in groups of two, and treatment group was assigned randomly. A technician was asked to pick one of two identical opaque envelopes: one contained the designation for PRP, and the other contained the designation for PRP plus intravitreal ranibizumab (PRPplus) treatment. The second patient was automatically assigned to the treatment designated in the second envelope.
For eyes assigned to the PRPplus group, intravitreal ranibizumab (0.5 mg, 0.05 ml) was administered 60 min after completion of the first PRP session as described elsewhere . At weeks 16 and 32, if active new vessels were detected on fluorescein angiography, patients in the PRPplus group received IVR and patients in the PRP group received five hundred additional 500 μm spots per quadrant of active new vessels. Also at weeks 16 and 32, if clinically significant macular edema was present, patients could be retreated with focal/gird laser if, in the opinion of the treating investigator, additional laser spots were possible.
The patients included in the current study (9 patients in the PRP group and 11 patients in the PRPplus group) were also enrolled in the study reported previously . Fifteen normal-sighted subjects (similar age range) with no known eye disease were evaluated with the same protocol as the treated patients.
Fluorescein leakage area (FLA) measurement
Digital red-free fundus photography and fluorescein angiography were performed using two fundus camera systems (HRA-OCT, Heidelberg, Germany, and TRC-50IA-IMAGEnet; Topcon, Japan). The total area (mm2) of fluorescein leakage from active NV was measured manually on the digital image using the Image J software (version 1.38, available at http://rsb.info.nih.gov/ij/); if more than one site of active NV were found, the sum area, including all sites, was measured for analysis.
ERG parameters (amplitudes and latencies) were compared between groups at baseline using one-way analysis of variance (ANOVA) and Tukey–Kramer HSD for the multiple comparison (Controls, PRP and PRPplus). The effect of treatment was compared within each group (PRP and PRPplus) using paired t tests, and between groups using t tests. In all analyses, P < 0.05 was considered the level of significance.
Patient demographic data
Age (mean ± 1 SD)
59 ± 12
64 ± 8
Duration of diabetes (mean ± 1 SD)
14 ± 6
13 ± 9
Treatment regimen (no insulin/insulin)
HbA1c (mean ± SD)
9 ± 1
9 ± 1
Fluorescein leakage area (FLA)
Compared to baseline, there was a statistically significant (P < 0.05) FLA reduction at all study visits in each group, with the reduction in the PRPplus group significantly larger than that in the PRP group at week 48. The mean ± SE FLA was 10.5 ± 1.6 mm2 and 11.6 ± 1.3 mm2 at baseline, and 7.0 ± 1.9 (P = 0.0007) and 6.0 ± 1.1 (P < 0.0001) at week 48 in the PRP group and the PRPplus group, respectively (P = 0.0074).
Best-corrected visual acuity (BCVA)
There was no statistically significant change in BCVA for eyes in the PRP and the PRPplus groups during the follow-up period. Mean ± SE BCVA was 0.30 ± 0.06 logMAR and 0.28 ± 0.07 logMAR at baseline; and 0.37 ± 0.07, and 0.28 ± 0.07 at week 48 in the PRP and PRPplus group, respectively (P > 0.05).
Dark-adapted ERG at baseline
Mean ± Standard Error (SE) of baseline a- and b-wave amplitude and implicit times and oscillatory potentials area under the curve (OP) for dark-adapted ERG responses (ROD: 0.01 cd.s/m2, combined response: 3.0 cd.s/m2, and high intensity response: 10.0 cd.s/m2)
a-wave amplitude (μV)
324.7 ± 13.0
206.2 ± 23.8
233.3 ± 20.6
392.4 ± 14.7
275.8 ± 26.8
306.5 ± 23.2
a-wave implicit time (ms)
15.6 ± 0.2
20.2 ± 0.4
19.8 ± 0.4
12.8 ± 0.2
16.4 ± 0.4
15.9 ± 0.4
b-wave amplitude (μV)
373.1 ± 17.1
178.7 ± 31.2
206.2 ± 27.0
556.8 ± 24.2
337.2 ± 44.1
362.8 ± 38.2
581.6 ± 23.3
369.6 ± 42.6
401.2 ± 36.9
b-wave implicit time (ms)
83.8 ± 1.7
104.3 ± 3.1
101.7 ± 2.7
49.6 ± 1.0
53.1 ± 1.7
53.8 ± 1.5
50.0 ± 1.3
55.0 ± 2.3
52.6 ± 2.0
790.4 ± 24.8
213.2 ± 45.2
255.3 ± 39.2
945.4 ± 33.7
272.8 ± 61.6
363.2 ± 53.4
ROD b-wave, and CR a- and b-wave implicit times were increased in PRP and PRPplus eyes when compared to control eyes; for the high intensity flash response, only the a-wave implicit time showed differences between the PRP and PRPplus groups compared to the control group, but there were no differences between the PRP and PRPplus groups (Table 2; Fig. 1). Disparities between eyes with PDR and control eyes with respect to ERG implicit times are larger for the lower luminance stimulus, indicating a tendency for stronger deterioration of rod-driven potentials in PDR.
OP amplitude was dramatically reduced for PRP and PRPplus eyes compared to control eyes. There was no significant difference in OP amplitudes at baseline between the PRP and PRPplus groups OP (Table 2; Fig. 1).
Light-adapted ERG at baseline
Mean ± Standard Error (SE) of baseline a- and b-wave amplitude and implicit times for light-adapted ERG responses (30 Hz: 3.0 cd.s/m2 at 30 Hz; and Cone Response: 3.0 cd.s/m2)
30 Hz amplitude (μV)
142.6 ± 6.2
59.5 ± 10.7
90.7 ± 6.5
30 Hz peak time (ms)
26.6 ± 0.2
34.8 ± 1.1
34.3 ± 0.9
Cone b-wave amplitude (μV)
191.5 ± 9.1
78.7 ± 12.2
101.3 ± 8.5
Cone b-wave implicit time (ms)
30.0 ± 0.2
35.9 ± 0.5
35.3 ± 0.7
Effects of PRP and PRPplus on dark-adapted ERG
In general, PRP and PRPplus treatments were associated with reduced ERG amplitudes and increased implicit times, but despite considerable group differences in number of applied laser spots (mean number of spots: PRP = 2,700; PRPplus = 1,600), only few ERG parameters demonstrated functional differences between the PRP and PRPplus groups.
Mean ± Standard Error (SE) of intra-individual ratio of baseline of a- and b-wave amplitude and OP and intra-individual difference of baseline of implicit time for dark-adapted ERG responses (ROD: 0.01 cd.s/m2, Combined Response: 3.0 cd.s/m2, and High Intensity Response: 10.0 cd.s/m2)
a-wave amplitude (ratio of baseline)
0.46 ± 0.07
0.49 ± 0.06
0.48 ± 0.06
0.50 ± 0.05
a-wave implicit time (difference of baseline in ms)
0.78 ± 0.36
0.17 ± 0.21
0.67 ± 0.44
0.25 ± 0.30
b-wave amplitude (ratio of baseline)
0.46 ± 0.05
0.64 ± 0.06
0.45 ± 0.04
0.62 ± 0.05
0.46 ± 0.05
0.59 ± 0.03
b-wave implicit time (difference of baseline in ms)
8.89 ± 3.11
7.92 ± 2.57
7.33 ± 2.03
6.75 ± 1.63
7.00 ± 1.49
5.17 ± 1.17
OP (ratio of baseline)
0.67 ± 0.07
0.69 ± 0.13
0.63 ± 0.06
0.60 ± 0.09
For all responses, the b-wave implicit times were significantly increased after treatment with PRP or PRPplus, but there were no significant differences in between-group analyses. On the other hand, a-wave implicit times did not show significant changes after treatment with PRP or PRPplus (Table 4; Fig. 3).
Effects of PRP and PRPplus on light-adapted ERG
Mean ± Standard Error (SE) of intra-individual ratio of baseline amplitude and intra-individual difference of baseline implicit time for light-adapted ERG responses (30 Hz: 3.0 cd.s/m2 at 30 Hz; and cone response: 3.0 cd.s/m2)
30 Hz amplitude (ratio of baseline)
0.45 ± 0.07
0.40 ± 0.04
30 Hz peak time (difference of baseline in ms)
1.00 ± 0.93
0.45 ± 0.58
Cone b-wave amplitude (ratio of baseline)
2.00 ± 0.58
1.36 ± 0.45
Cone b-wave implicit time (difference of baseline in ms)
0.45 ± 0.06
0.49 ± 0.05
30 Hz time-to-peak did not show significant changes after treatment with PRP or PRPplus, while Cone b-wave implicit time was significantly increased after treatment in the PRP and PRPplus groups (Table 5; Fig. 4).
We have previously shown that combining IVR with PRP reduces the amount of PRP needed to control high-risk PDR progression, with even larger reduction in FLA leakage than using PRP alone . In the current study, we aimed to describe retinal function, using ERG, in eyes with high-risk PDR, to record the functional impairment caused by extensive PRP, and to investigate if the combination of IVR plus PRP is associated with less retinal functional loss than treatment with PRP alone.
ERG is the only objective method of evaluating retinal function in vivo and is employed to detect distinct local or widespread hereditary or acquired retinal disorders including DR . As an example, OP amplitude shows correlations with retinal arteriolar caliber in diabetic patients  and has been proposed to be useful for predicting DR progression and even to detect retinal dysfunction that precedes the onset of clinically detectable DR .
Data presented confirm previous descriptions of ERG changes in PDR: reduced a- and b-wave amplitudes and prolonged implicit times, for rod and cone driven responses, and dramatically reduced OP amplitudes [21, 36]. Moreover, the largest difference between controls and patients was found for OP area under the curve, indicating a predominance of inner-retina over photoreceptor dysfunction in eyes with high-risk PDR.
Similarly, by modeling the interrelation between dark-adapted b-wave and stimulus luminance, we showed that the saturation amplitude (parameter Vmax) is more affected than the half-saturation intensity (k). In fact, it has already been shown that other ERG parameters, such b/a amplitude ratio  and b-wave implicit time , can be more sensitive than the dark-adapted sensitivity, estimated with the parameter k, to retinal functional changes due to retinal ischemia or diabetic retinopathy.
The findings after extensive PRP documented in the current study are similar to previous reports , in which significant reductions in a- and b-waves amplitudes of dark- and light-adapted responses were observed. This can be explained by the retinal tissue damage due to PDR.
Interestingly, the ERG parameter with the highest reduction after extensive PRP was the dark-adapted b-wave amplitude to the ROD stimulus. It is unclear why a reduction of similar magnitude was not observed in the a-wave amplitudes, since retinal photocoagulation targets the retinal pigment epithelium, affecting the outer retina and, therefore, the photoreceptors.
Based on a previously proposed theory that photocoagulation not only destroys the retinal areas directly illuminated by the laser beam, but also affects the functional integrity of adjacent areas , Perlman et al.  suggested that PRP leads to larger reductions in the dark-adapted b-wave amplitude than a-wave amplitude due to subnormal signal transmission from the photoreceptors to the proximal retina. We agree that this might be one mechanism to explain the larger reduction in b-wave amplitude compared to the reduction in a-wave amplitude.
Furthermore, it is possible that at least part of the photocoagulated retina was not ischemic and, therefore, had functional inner-retina that may have generated a substantial part of the ERG b-wave amplitudes recorded before PRP. With the destruction of these retinal areas after PRP, a- and b-wave amplitudes decreased, as observed, but the b-wave on a larger scale.
To the authors’ knowledge, this is the first study to evaluate retinal function using ERG in eyes treated with PRP and IVR for PDR. The data suggest that the combined treatment strategy is associated with lower retinal function loss, indicated by lesser b-wave amplitude reduction after combined treatment, compared to treatment with PRP alone. The lower number of laser spots performed in eyes treated with PRP plus IVR compared to eyes treated with PRP alone is likely to be the explanation for the lower functional loss in the PRP plus IVR group.
ERG analysis showed that the dark-adapted b-wave amplitude elicited by a weak flash (ROD) seems to be the most sensitive ERG parameter to evaluate retinal function loss in eyes with high-risk PDR and functional changes after PRP. This response is mainly generated by the depolarization of ON bipolar cells from the rod pathway , which are located in the inner retina, and therefore is susceptible to retinal ischemia present in high-risk PDR , and also for detection of functional changes after PRP, because PRP is performed at retinal periphery, where rods are dominant.
OP amplitudes were dramatically reduced or even abolished in eyes with high-risk PDR even before PRP and, therefore, are not as useful as the dark-adapted b-wave amplitude to evaluate post-treatment effects. Consistently, differences between the PRP and PRPplus groups were found in the b-wave amplitude but not for other ERG parameters. Furthermore, light-adapted ERG responses showed less reduction after PRP and did not show differences between eyes in the PRP versus PRPplus groups, probably because the central retina was less affected by PRP than the peripheral retina.
In conclusion, in the current study, the combination of IVR and PRP for high-risk PDR was associated with less retinal function loss, as measured by ERG, compared to treatment with PRP alone. ERG can be considered a tool to localize retinal function loss and increase our understanding about the effects of new treatment strategies for diabetic retinopathy. Further studies are necessary to confirm these findings and investigate correlations with other functional and/or structural retinal changes in this condition.
This study was supported by Fundação de Amparo a Pesquisa do Estado de São Paulo (FAPESP) Grant number: 2009/01036-3. Authors would like to thank Prof. John Robson for reviewing and the constructive comments on the manuscript.
Conflict of interest