Early panretinal photocoagulation for ERG‐verified ischaemic central retinal vein occlusion

Purpose: We have previously shown that photopic cone b‐wave implicit time ≥35.0 ms in 30 Hz flicker electroretinography (ERG) predicts ocular neovascularization (NV) in central retinal vein occlusion (CRVO). Here, we evaluate the effects of early panretinal photocoagulation (PRP) in patients with ERG‐verified ischaemic CRVO. Methods: Patients with CRVO, admitted to our department between 2000 and 2008, were classified as having ischaemic or non‐ischaemic CRVO based on the ERG‐results. In a first group of 71 patients, 18 patients had ischaemic CRVO and were assigned to standard treatment that is regular examinations and PRP as soon as NV was found. In a consecutive group of 65 patients, 18 patients with ischaemic CRVO received early PRP. In this group, ERG was performed on average 6 weeks after the first symptoms of CRVO. The patients underwent PRP as soon as possible after the ERG‐examination, and the treatment was completed within one to three sessions. Results: Twelve patients in the standard treatment group developed neovascular glaucoma during a mean period of 5 months after the CRVO. In the early treatment group, one patient developed subtle iris rubeosis 7 months after PRP. Otherwise, none of the patients showed any signs of ocular NV, and the intraocular pressure remained within normal range, without the necessity of supplementary medication, during a mean follow‐up of 41 months. Conclusions: This study indicates that ocular NV in patients with CRVO can be predicted by photopic 30 Hz flicker ERG and that early PRP in ERG‐verified ischaemic CRVO could be suggested as standard treatment.     

视网膜中央静脉阻塞性黄斑水肿的OCT与ERG的对比分析

目的：观察视网膜中央静脉阻塞性黄斑水肿的黄斑区视网膜厚度与视网膜电图( electroretinogram, ERG )各项参数(Cone-a,Cone-b和30Hz)变化的关系。
　　方法：随机选择视网膜中央静脉阻塞患者25例25眼及25只对侧眼分别行明视闪光视网膜电图及光学相干断层扫描( optical coherence tomography,OCT)检查,明视闪光视网膜电图检查测各项参数的振幅和潜伏期, OCT测量黄斑区九部分的视网膜厚度,分析黄斑区形态参数与明视闪光视网膜电图各参数变化之间的关系。
　　结果：黄斑区除颞侧外七个部位视网膜厚度与 ERG 的Cone-b和30 Hz潜伏期相关。
　　结论：研究发现视网膜中央静脉阻塞患者的黄斑区视网膜厚度与内层视网膜功能密切相关。     

Photopic 30 Hz flicker ERG as a predictor for rubeosis in central retinal vein occlusion

AIMS: To investigate the predictive value of the cone b-wave implicit time in the photopic compared with the scotopic 30 Hz flicker ERG for rubeosis in central retinal vein occlusion. METHODS: 44 patients with a central retinal vein occlusion were examined with full field electroretinogram (ERG). RESULTS: The average implicit time in the photopic 30 Hz flicker ERG in patients who developed rubeosis (n=15) was 38.3 ms. In the patients who did not develop rubeosis (n=29) it was 31.3. The difference is statistically significant (p=0.0000000004). CONCLUSION: The photopic cone b-wave implicit time in the 30 Hz flicker ERG is a good predictor for rubeosis.     

The photopic negative response of the flash electroretinogram in retinal vein occlusion

The photopic negative response (PhNR) has recently been shown to be severely affected in central retinal artery occlusion (CRAO), despite relative preservation of the cone b-wave compared to that in the healthy unaffected fellow eye. The aim of this study was to test how the PhNR of the flash electroretinogram (ERG) is affected in human retinal vein occlusion. PhNR was elicited with red stimuli (1 cd s/m², 5 cd s/m², and 7 cd s/m² with 4 ms duration) and blue background (10 cd/m²). Standard Ganzfeld flash ERG was produced according to the ISCEV standard for the clinical electroretinogram (2004). Sixteen patients with central retinal vein occlusion (CRVO), 14 patients with branch retinal vein occlusion (BRVO), and 16 controls were analyzed. The amplitude of the PhNRs was significantly smaller in the CRVO and BRVO eyes than those in the unaffected fellow or control eyes (p = 0.000). There was a significantly greater reduction of PhNR amplitudes than that of other waves including the OPs, rod b-wave, combined a-wave and b-wave, cone a-wave and b-wave, and 30 Hz flicker ERG. Thus, PhNR amplitude in retinal vein occlusion is severely affected. There is a potential role for PhNR in assessing inner retinal damage and evaluating the effect of treatment.     

Multifocal electroretinographic features of central retinal vein occlusion

PURPOSE: To determine the features of wide-field multifocal electroretinography (WF-mfERG) recorded in patients with central retinal vein occlusion (CRVO) and to compare WF-mfERG responses of the affected and fellow eyes. In addition, WF-mfERG responses were also compared by using standard electroretinography (ERG). METHODS: WF-mfERG and ERG responses were recorded from both eyes of 56 patients with CRVO. The WF-mfERG responses, obtained using a custom-built system were grouped into central and peripheral rings. The P1 amplitudes, and P1 and N1 implicit times were grouped and averaged within both rings. Nonparametric statistical analysis was used to compare the ERG results from the affected and fellow eyes. The results were also compared with normative data (5% to 95% confidence limits). RESULTS: CRVO markedly affected all the parameters of the WF-mfERG. In the affected eyes, 98% of the central and 91% of the peripheral P1 implicit times fell outside the normal range, as opposed to 35% of the 30-Hz flicker implicit times. The WF-mfERG responses obtained from eyes with CRVO were significantly different (P<0.01) from those derived from the fellow eye. The central and peripheral P1 implicit times were also abnormal in 59.2% and 46.9% of the fellow eyes, respectively. CONCLUSIONS: WF-mfERG is more susceptible than the standard ERG to changes in the nonlinear dynamics of the eye due to the multiple frequencies of stimulation used to record WF-mfERG responses. WF-mfERG could be a sensitive indicator of the underlying disease affecting the retina in eyes with CRVO and may have a role in the clinical setting.     

Correlation of electroretinographic and fluorescein angiographic findings in unilateral central retinal vein obstruction

Background: In central retinal vein obstruction (CRVO), electroretinogram (ERG) abnormalities and extensive retinal capillary dropout (CD) in the fluorescein angiogram (FA) are good indicators of retinal ischemia. We retrospectively studied patients with unilateral CRVO and compared the ERG and FA results Methods: Single white flash ERG, photopic ERG, scotopic ERG and flicker ERG were recordered in 30 cases of unilateral CRVO. We analyzed the correlation between the ERG results and the presence/absence of extensive CD Results: The ERG b/a-wave amplitude ratios, photopic and scotopic b-wave amplitudes, and flicker amplitudes were significantly smaller (P<0.05) in eyes with extensive CD (n=12, 40%), than in eyes without (n=18, 60%). When the photopic or scotopic b-wave amplitudes were normal or supernormal, extensive CD on FA was absent in all eyes. When the b/a-wave ratios were 1.0 or when the b-wave amplitudes with white flash or flicker amplitudes were normal or supernormal, extensive CD was present in less than 32% of eyes Conclusion: These results suggest that the ERG results, especially the b/awave amplitude ratio, are significantly correlated with the presence/absence of CD on FA in CRVO.     

Electroretinography in central retinal vein occlusion

In 149 eyes with central retinal vein occlusion (CRVO), we prospectively investigated the role of routine, clinical electroretinography (ERG) in differentiating ischemic (60 eyes) from nonischemic CRVO (89 eyes). Single-flash photopic and scotopic ERGs were recorded. Data for the amplitudes and implicit times of a- and b-waves and for the b-/a-wave amplitude ratio were analyzed in detail. The study revealed that the best ERG parameter (for both photopic and scotopic ERG) for differentiating ischemic from nonischemic CRVO was a subnormal b-wave amplitude (reduced to 60% or by 1 SD from the normal mean value, or 64%–69% of that in the fellow normal eye), with a sensitivity of 80%–90% and a specificity of 70%–80%. ERG findings were correlated with the relative afferent pupillary defect (RAPD). An RAPD of 0.7 log units showed a sensitivity of 88% and a specificity of 90% in differentiating ischemic from nonischemic CRVO. ERG and RAPD findings showed a good correlation. The combined ERG and RAPD tests could differentiate 97%–100% of ischemic from nonischemic CRVO cases, with a specificity of about 70%.Presented in part at the Annual Meeting of the Association for Research in Vision and Ophthalmology, Sarasota, Florida, USA, 2 May 1988Supported by grant EY-1151 from the National Institutes of Health, and in part by unrestricted grants from Research to Prevent Blindness, Inc., and from Alcon Research Institute     

Electroretinogram b-wave implicit time and b/a wave ratio as a function of intensity in central retinal vein occlusion.

The ability of electroretinogram (ERG) b-wave implicit time and b/a wave ratio to predict iris neovascular response was analyzed as a function of stimulus intensity over a 3.6 log unit intensity range in 39 patients with central retinal vein occlusion (CRVO). Predictive power for CRVO patients was evaluated using ROC area at intensities of 1.23, 1.83, 2.43, and 3.03 effective log quanta/rod, where reliable data for both parameters were obtainable from most patients. The relative predictive power of b-wave implicit time and b/a wave ratio were shown to vary with stimulus intensity. The predictive power of b-wave implicit time, as measured by ROC area, declined to below significance at high intensity (above 1.83 log quanta/rod), while b/a wave ratio performed best at middle intensities (1.83 and 2.43 log quanta/rod) and not as well at high and low intensities. Further analysis of statistical behavior of both ERG parameters was obtained from the t statistic. Insight into the mechanism influencing predictive power of b-wave implicit time was derived from measurements on normal adults and CRVO patients with response data taken at high intensities. These results suggest that an optimal stimulus intensity range can be found for these ERG parameters in the evaluation of CRVO.     

Predominant loss of the photopic negative response in central retinal artery occlusion

PURPOSE: To determine how the photopic negative response (PhNR) is affected in central retinal artery occlusion (CRAO). DESIGN: Observational case series. METHODS: Seven patients with unilateral CRAO were included. Full-field scotopic and photopic electroretinograms (ERGs) including the PhNR were recorded. Each ERG amplitude in the affected eye was expressed as a percentage of amplitude of the corresponding wave in the unaffected eye. RESULTS: Mean of the PhNR amplitude was reduced to 12.3 +/- 11.7% of that of unaffected eyes whereas the cone b-wave amplitude was attenuated to only 73.4 +/- 30.4%. This reduction of the PhNR amplitude was more significant than that of other waves including the rod b-wave, maximum a-wave and b-wave, cone a-wave and b-wave, and 30 Hz flicker ERG (P <.005). CONCLUSIONS: The PhNR was severely affected in CRAO despite relative preservation of the cone b-wave, implicating massive loss of ganglion cells and their axons.     

ON- and OFF-response of the photopic electroretinogram in relation to stimulus characteristics

The purpose of this study was to document the standard full field electroretinographic (ERG) and wide field multifocal electroretinographic (WF-mfERG) findings in eyes with recent onset hemi-retinal vein occlusion (HRVO) and to compare the electro-diagnostic findings in the affected and fellow eyes with reference to normative data. Eight patients with HRVO were assessed using ERG and WF-mfERG. WF-mfERG first order responses from the affected hemi-retinae and the unaffected hemi-retinae in each affected eye were compared. WF-mfERG responses from each affected hemi-retina and from the symmetrical hemi-retina of each fellow eye were compared. ERG responses between affected and unaffected eyes were also compared. All electrodiagnostic tests were compared to normative data (5–95% confidence limits derived from age-related controls). WF-mfERG P1 and N1 implicit times were greater for the affected hemi-retinae than for the unaffected hemi-retinae (p <0.05). WF-mfERG N1 and P1 implicit times were prolonged (p < 0.05) and WF-mfERG P1/N1 amplitude ratios were significantly reduced (p < 0.05) for the affected eyes when compared with the fellow eyes. Maximal b-wave, cone b-wave and flicker implicit times were prolonged (p < 0.05) when comparing affected and fellow eyes. These results indicate that retinal injury due to HRVO culminates in significant delay of both ERG and WF-mfERG implicit times. These results suggest that WF-mfERG in combination with ERG may have a role in the management of HRVO.     

Night Blindness with Depolarizing Pattern of ON/OFF Response in Electroretinogram: A Case Report

Purpose: To present a patient who has night blindness with a depolarizing pattern of ON/OFF response by electroretinography (ERG). Case: A 43-year-old woman had had night blindness and poorly corrected visual acuity since childhood. Parental consanguinity was noted. The patient had suffered from mental retardation, epilepsy, and mild cerebellar ataxia. Corrected visual acuity was 20/30 in the right eye and 20/25 in the left. Goldmann perimetry showed no scotoma but slight depression with internal isoptors. No evidence for rod activity was observed by Goldmann–Weekers adaptometry. The ocular fundi appeared normal. Methods: Conventional full-field ERGs to scotopic (dim and bright flash) and photopic (bright flash and flicker) stimuli were recorded. Photopic ERG responses to long flash stimulation (200 ms) were also examined. Results: The scotopic responses to dim flash were non-recordable, while those to bright flash were severely reduced. The photopic responses to bright flash were decreased. The amplitudes of flicker responses also were significantly decreased, and the implicit times of those responses were prolonged. Although the amplitudes of a- and d-waves to long flash stimulation were reduced, those of the b-wave were within normal range. The implicit times of a-, b- and d-waves were significantly prolonged. The patient showed a normal rise for the photopic b-wave but lacked a normal falling slope after the b-wave peak. An OFF-response late-negativity was also noted. Conclusions: The abnormal ON/OFF response found in the patient could be diagnosed as depolarizing pattern, characterized by elevation of the plateau to a positive potential above the prestimulus baseline. Since the ERG waveforms and clinical features found in our patient were different from those in previous reports, her ERG findings might reflect another retinal physiological mechanism.     

Temporal aspects of the electroretinogram in cone-rod dystrophy

Twelve patients with cone-rod dystrophy were studied in terms of the temporal aspects of electroretinography (ERG).The peak time of scotopic b-wave was within normal limits in all patients, while the amplitude was reduced in nine patients. The normal peak time of the scotopic-b wave may help explain rod involvement with normal or only slightly elevated final rod thresholds of subjective dark adaptation.The peak time of photopic b-wave and 30-Hz flicker response was normal in four patients and significantly delayed or nonrecordable in eight patients. Such variations of peak time in photopic and 30-Hz flicker ERG may reflect the stage of the disease or may be caused by the different hereditary mode.     

Origin of electroretinogram amplitude growth during light adaptation in pigmented rats

We assessed the growth of the rat photopic electroretinogram (ERG) during light adaptation and the mechanisms underlying this process. Full field ERG responses were recorded from anesthetized adult Brown-Norway rats at each minute for 20 min of light adaptation (backgrounds: 1.8, 2.1, 2.4 log scotopic cd m(-2)). The rat photopic b-wave amplitude increased with duration of light adaptation and its width at 33% maximal amplitude narrowed (by approximately 40 ms). These effects peaked 12-15 min after background onset. The narrowing of the b-wave reflected steepening of the b-wave recovery phase, with little change in the rising phase. OP amplitudes grew in proportion to the b-wave. Inhibition of inner retinal responses using TTX resulted in a greater relative growth of b-wave and OP amplitude compared with fellow control eyes, and delayed the change in recovery phase by approximately 5 min. Inhibition of all ionotropic glutamate receptors with CNQX/D-AP7 delayed both rising and recovery phases equally (approximately 12 ms) without altering b-wave width or the time course of adaptation changes. These outcomes suggest that inner retinal light responses are not directly responsible for b-wave amplitude growth, but may contribute to the change in its recovery phase during adaptation. A TTX-sensitive mechanism may help to hasten this process. The cone a-wave was isolated using PDA/L-AP4 or CNQX/L-AP4. A-wave amplitude (35 ms after stimulus onset) also increased with time during light adaptation and reached a maximum (130 +/- 29% above baseline) 12-15 min after background onset. B-wave amplitude growth in fellow control eyes closely followed the course and relative magnitude of cone a-wave amplitude growth. Hence, the increase of the cone response during light adaptation is sufficient to explain b-wave amplitude growth.     

The 52nd ISCEV International Symposium Abstract Issue

Purpose

We have monitored retinal function in patients treated for retinoblastoma (primarily, but not exclusively by intra-arterial chemotherapy infusion) by electroretinography (ERG) recordings for the past 7 years. We here present data from 599 ERG studies of 108 patients, in which a complete ERG protocol including both photopic and scotopic recordings was performed, in justification of our frequent practice of reporting primarily 30-Hz photopic flicker amplitude data.

Methods

Patients referred for treatment of retinoblastoma underwent ERG recordings during examination under anesthesia whenever possible: at baseline and following most treatment sessions. Correlations were calculated for the complete datasets between the four primary amplitude response parameters: photopic single flash b-wave, photopic 30-Hz flicker peak-to-trough, scotopic rod-isolating b-wave, and scotopic maximal flash b-wave.

Results

Using our adaptation of the International Society for Clinical Electrophysiology of Vision-recommended standard ERG protocol, ERG responses of eyes of patients with untreated retinoblastoma or following traditional or intra-arterial treatment for retinoblastoma show very high correlations between 30-Hz flicker amplitude responses and three other standard photopic and scotopic ERG response amplitudes. Reductions in ERG amplitudes seen in these eyes following treatment show no significant difference between retinal dysfunction estimated using rod- or cone-dominated responses.

Conclusion

These observations support the use of photopic response amplitudes (especially in response to 30-Hz flicker) as the primary ERG outcome measure in studies of treated and untreated eyes with retinoblastoma when more complete ERG protocols may be impractical.     

Electroretinogram interpretation in central retinal vein occlusion.

The authors report electroretinogram (ERG) data from the initial clinic visit of 39 patients with central retinal vein occlusion (CRVO). No patient had signs of neovascular complications or had received treatment at the time of the ERG examination. Area under the receiver operating characteristic (ROC) curve was used to compare effectiveness of the 4 ERG parameters (Rmax, Log K, b/a wave ratio, and 30 Hz implicit time) in separating those patients who went on to iris neovascularization from those who did not. Rmax is the maximum saturated b-wave amplitude and Log K is the half saturation constant of the Naka-Rushton curve fit to the intensity response data. Discriminant scores, derived using multiple discriminant analysis, were calculated for the total patient groups, CRVO eye alone, and intereye difference. These scores also were compared with the four individual ERG parameters using ROC analysis. Parameters based on amplitude of ERG response, Rmax, and b/a wave ratio are as effective predictors of neovascular response as those interpreted as indicators of retinal sensitivity, such as 30Hz implicit time or Log K. The authors present evidence that loss of b-wave amplitude is not necessarily associated with irreversible loss of inner retinal function.     

Ocular manipulation reduces both ipsilateral and contralateral electroretinograms

Purpose

To determine the electroretinogram (ERG) changes in eyes manipulated in the course of local ablative therapy (transpupil thermotherapy (TTT), cryotherapy or both) or scleral depression and in un-manipulated fellow, healthy eyes.

Methods

This prospective observational report summarizes 73 ERG studies in 42 patients with retinoblastoma; a study consisted of ERGs of one or both eyes (if present) followed by ocular manipulation (scleral depression, cryotherapy, transpupillary thermotherapy, pressure applied to orbital implant in an anophthalmic socket, or a 5- or 10-min delay without mechanical manipulation) followed by a repeat of the ERGs. Each patient was studied with only a single manipulation modality on any given date: 23 patients were studied only once, and 19 patients were included in more than one study occasion.

Results

Following local ablative treatment of patients with unilateral retinoblastoma, the photopic response decreased significantly in both the treated eye and the untouched fellow, healthy eye. Following scleral depression of the diseased eye, the photopic response immediately decreased in the diseased eye by a mean of 16 μV (21 %, p = .006) and, in the fellow, healthy eye by 40 μV (23 %, p = .0005). Following scleral depression of the fellow, healthy eye, the photopic response immediately decreased by a mean of 11 μV (4 %, p = .37) in the fellow, healthy eye, and by 16 μV (28 %, p = .01) in the diseased eye.

Conclusions

Following physical ocular manipulation, the amplitude of the photopic response decreased in the manipulated, but also the untouched healthy, fellow eyes. These findings may account for some of the variation in clinical ERG recordings, particularly that observed following ocular manipulation by TTT, laser or even scleral depression.     

Interocular amplitude differences of the full field electroretinogram in normal subjects

AIMS: To determine the interocular amplitude response difference of the electroretinogram (ERG) in normal subjects. METHODS: 79 subjects, without retinal changes of clinical significance, underwent ERG testing. They included 63 men and 16 women, with a mean age of 44 (SD 12) years and range of 18-65 years. Isolated rod, scotopic maximal, dark adapted 30 Hz flicker, photopic single flash, and light adapted 30 Hz flicker responses were recorded in both eyes following the International Society for Clinical Electrophysiology of Vision (ISCEV) standard protocol. The interocular percentage differences of the ERG b-wave amplitudes were calculated and presented as percentiles (25th, 50th, 75th, 95th), means (SD), and medians. RESULTS: The median interocular percentage differences in the b-wave amplitudes for the above ERG stimulus responses were 10%, 8%, 10%, 11%, and 10%, respectively. The mean interocular percentage differences were 11%, 11%, 12%, 13%, and 14%. The 95th percentiles for the interocular percentage differences were 28%, 27%, 36%, 33%, and 35%, respectively. CONCLUSIONS: The interocular percentage differences in the ERG b-wave amplitudes for five different stimulus responses were similar in our cohort of individuals without clinically significant retinal changes and ranged from a median of 8-11% and a 95th percentile of 27-36%. Our findings should be useful for determining sample sizes in future therapeutic trials on retinal diseases with monocular therapeutic strategies and may also have application for the more accurate detection of asymmetric retinal disease.     

Association of electroretinogram and morphological findings in branch retinal vein occlusion with macular edema

The objective of this study is to evaluate the relations among electroretinogram parameters (cone a-wave, cone b-wave, and 30-Hz flicker), retinal thickness, and retinal volume in patients with branch retinal vein occlusion (BRVO) and macular edema. We prospectively examined 33 patients (33 eyes) with BRVO and macular edema. The amplitude and implicit time of the a-wave cone, b-wave cone, and 30-Hz flicker were calculated automatically from the ERG. Retinal thickness and volume were measured by optical coherence tomography (OCT) in nine macular subfields. Then, correlations between the ERG parameters and morphological parameters were analyzed. The 30-Hz flicker amplitude was significantly smaller in the eyes with BRVO and macular edema than in the unaffected contralateral eyes. Thirty-hertz flicker and cone b-wave implicit times were significantly longer in the eyes with macular edema than in the unaffected eyes. The implicit time of the cone b-wave was correlated with both retinal thickness and retinal volume in the temporal subfields. Thirty-hertz flicker amplitude was correlated with both retinal thickness and volume in the temporal and superior outer (site of occlusion) subfields, while 30-Hz flicker implicit time was correlated with retinal thickness and volume in the outer temporal subfield. Multiple regression analysis demonstrated that the retinal thickness and volume of the temporal subfields were significant “determinants” of the implicit time for the cone b-wave and 30-Hz flicker, as well as the 30-Hz flicker amplitude. These findings suggest that OCT parameters of the temporal region may reflect postreceptoral cone pathway function in BRVO patients with macular edema.     

视网膜电图对视网膜血管阻塞的功能评估

目的：用视网膜电图评估视网膜血管阻塞的功能改变。方法：３０例视网膜中央静脉阻塞,１５例视网膜分支静脉阻塞,７例视网膜中央动脉阻塞和６例视网膜分支动脉阻塞按照ＩＳＣＥＶ的ＥＲＧ标准进行全视野网膜电图的检测。用Ｉ１蓝光、Ｉ１６红光和Ｉ１６白光估计暗视ＥＲＧ,用Ｉ１６红光和Ｉ８白光估计明视ＥＲＧ。记录ａ波和ｂ波的振幅和潜伏期及震荡电位。结果：视网膜血管阻塞的震荡电位异常率最高,其次为ｂ波;ＥＲＧ的异常率     

Electroretinogram parameters at presentation as predictors of rubeosis in central retinal vein occlusion patients

This report evaluates the prognostic value of parametric electroretinography (ERG) techniques in the development of rubeosis in central retinal vein occlusion (CRVO). An ERG protocol was followed on the initial clinic visit for 21 CRVO patients. None of these patients had neovascular complications or any treatment before the ERG examination. The ERG data were used to define an intensity-response (I-R) function, b-wave to a-wave peak ratio (b/a-wave ratio), and 30-Hz b-wave implicit time. A Naka-Rushton function was fitted to the I-R data to obtain values for maximum saturated response (Rmax) and for the log of the half-saturation intensity (log K). The authors' results show that ERG parameters relating both to the effective number of responding retinal elements and to the sensitivity of responding elements have significant prognostic value. Unexpectedly high predictive values were found for the Rmax and b/a ratio, parameters thought to correlate with the effective number of responding elements, compared with log K and 30-Hz implicit time, parameters thought to relate more closely to retinal ischemia. A multiple discriminant analysis, combining information from these ERG parameters, allowed separation of CRVO patients in whom rubeosis would develop from those in whom it would not, with a false-positive rate of only 14%.