A review of electroretinography waveforms and models and their application in the dog

Electroretinography (ERG) is a commonly used technique to study retinal function in both clinical and research ophthalmology. ERG responses can be divided into component waveforms, analysis of which can provide insight into the health and function of different types and populations of retinal cells. In dogs, ERG has been used in the characterization of normal retinal function, as well as the diagnosis of retinal diseases and measuring effects of treatment. While many components of the recorded waveform are similar across species, dogs have several notable features that should be differentiated from the responses in humans and other animals. Additionally, modifications of standard protocols, such as changing flash frequency and stimulus color, and mathematical models of ERG waveforms have been used in studies of human retinal function but have been infrequently applied to visual electrophysiology in dogs. This review provides an overview of the origins and applications of ERG in addition to potential avenues for further characterization of responses in the dog.     

Technical issues in electrodiagnostic recording

Electroretinogram (ERG) and visual-evoked potentials (VEPs) are used in veterinary ophthalmology to assess the functional integrity of the retina and the central visual pathway. The interpretation of altered electrophysiologic potentials in diseases can be of great diagnostic value, although it is important to be aware of technical factors and the limitations of these techniques which may result in over-interpretation and misinterpretation of the recordings, such that they are similar to those found in disease-related electrophysiologic changes. The recorded potentials represent the differences in voltage between the active and reference electrodes. The ground electrode serves as zero. A differential or instrumentation amplifier selectively amplifies signals of interest while rejecting noise. Differences between inputs are amplified, whereas common signals are rejected in a process called common mode rejection (CMR). In order for CMR to be most effective in reducing noise, the electrode impedances should be balanced. Filters are part of the differential amplifier as they remove unwanted noise of a certain frequency. The frequency bandwidth, or passband, is the range of frequencies between low- and high-frequency filter settings that are not filtered out. Major sources of noise that cause interpretation artifacts are power lines, amplifier noise, physiologic activity, electrochemical electrode noise and circular grounds. Noise reduction is achieved with high amplifier input impedance, balanced electrode impedances, CMR, filters and signal averaging. Maintaining electrodes in good condition, achieving proper contact between electrode and animal, and keeping electrode leads short aid in achieving noise reduction.     

The electroretinographic phenotype of dogs with Golden Retriever muscular dystrophy

Objective To characterize the flash electroretinogram (ERG) in the Golden Retriever muscular dystrophy (GRMD) dog and to compare the results with those from a control group of Golden Retrievers. To investigate whether similar abnormalities of the ERG as those found in a majority of human patients with Duchenne muscular dystrophy (DMD) are also observed in the GRMD dog, the canine model for DMD. Animals Five GRMD dogs and five age‐matched clinically normal Golden Retrievers. Procedure An ophthalmic examination was carried out prior to performing electroretinography under general anesthesia. Rod, combined rod–cone and oscillatory potentials responses were recorded after dark adaptation. Responses to 30‐Hz‐flicker were recorded after light adaptation. The ERG responses of the GRMD dogs were compared with those of the control dogs by use of a Wilcoxon signed rank test. Results GRMD dogs had significantly reduced a and b‐wave amplitudes after dim white flash stimuli (rod response) and reduced a‐wave amplitude after bright white flash stimuli (rod–cone response). Conclusion and clinical relevance The ERG abnormalities observed in the GRMD dog suggest a dysfunction in the rod signaling pathway. These ERG alterations are different from those observed in human patients with DMD.     

Electroretinography recordings using a light emitting diode active corneal electrode in healthy beagle dogs

Electroretinography (ERG) is a well-established diagnostic procedure for objectively evaluating retinal function. In this study, ERG in beagle dogs, which are a popular experimental animal, was performed to determine the normal range of ERG variables and assess differences between the left and right eyes. ERG findings including rod, combined rod-cone, single-flash cone, and 30-Hz flicker responses were recorded with an LED-electrode in 43 sedated beagle dogs. The subjects were divided into young (< 1 year old), adult (1~5 years old), and senile animals (≥ 6 years old). Normal ERG ranges were obtained. Significant differences in b-wave amplitude along with b/a ratio of the combined rod-cone response were found between the young and adult animals as well as young and senile dogs. No significant differences were observed between the left and right eyes. ERG variables in beagle dogs differed by age due to age-related retinal changes. Thus, we propose that normal ERG ranges should be determined according to age in each clinic and laboratory using its own equipment because each institution usually has different systems or protocols for ERG testing.     

Day-blindness in three dogs: clinical and electroretinographic findings

A 6-month-old Rhodesian ridgeback-cross, a 6-year-old Chihuahua and a 12-month-old Australian cattle dog were presented to the authors with a history of colliding with obstacles in daylight. Ophthalmic examination was normal and all three dogs successfully negotiated obstacle courses in dim light. In daylight the dogs became suddenly blind and repeatedly collided with obstacles. Elecroretinography (ERG) revealed no retinal activity to high frequency (30 Hz), bright intensity blue light retinal stimulation by any dog, confirming cone dysfunction. Achromatopsia has previously been recorded in Alaskan malamutes and miniature poodles. This clinical case series illustrates the characteristic behavioral presentation and the electroretinographic findings of severe day-blindness and demonstrates that this condition may exist in other breeds of dogs.     

Electroretinography using contact lens electrode with built-in light source in dogs

Electroretinography (ERG) is an effective method for the diagnosis of retinal disease. In the dog, dependable ERG recording is difficult without the use of an expensive device like a Ganzfeld full-field stimulator. The International Society for Clinical Electrophysiology of Vision has defined the standard flash stimulus condition (SF) and evaluation of the retina using the b/a ratio in humans. In dogs, evaluation using the b/a ratio has not been reported, whereas the intensity of SF has been defined. In this study, we performed a convenient ERG recording method using a contact lens electrode with a built-in light source (LED-electrode), and confirmed SF as reported previously. ERG recordings were performed on 15 healthy beagle dogs under sedation. We performed bilateral ERG at 12 different intensities after 30 min dark adaptation. After 10 min light adaptation, we recorded single flash cone and flicker cone response using the SF determined in this study. In this study, SF of 3.0 cd/m(2)/sec (6,000 cd/m(2), 0.5 msec) resulted in b/a=2. The intensity for rod response that recorded only the b-wave was 0.0096 cd/m(2)/sec (80 cd/m(2), 0.12 msec). We could achieve ERG for each response easily and smoothly under sedation, and without general anesthesia. Using an LED-electrode, we could perform more quantitative and reproducible ERG examinations than with traditional methods. We propose that the b/a ratio is the most useful parameter in ERG reporting for evaluating retinal function.     

Brain stem auditory-evoked potentials of dogs: wave forms and effects of recording electrode positions

Wave forms of canine brain stem auditory-evoked potentials (BAEP) and the effects of electrode positions on the wave forms were studied as a basis for experimental and clinical use of BAEP recording. The BAEP regularly consisted of 5 waves (I to V) with latencies and polarities similar to those of other species. In some dogs, waves II, III, and IV contained distinct subpeaks (a, b, c). Waves similar to waves VI and VII of other species were recorded in some dogs. With respect to BAEP, no site on the head was electrically inactive and BAEP could be recorded as far caudally as the caudal cervical region in some dogs. Wave I, positive in recordings from the dorsal midline of the calvaria (vertex) underwent polarity reversal and increased amplitude and duration in recordings made from caudal ventrolateral regions of the head (mastoid region). As a result, wave I partially or totally obscured wave II so that the latter could no longer be clearly identified. Waves IIIa and IIIb were differentially affected by moving the recording site, indicating that their generators were spatially separated. Waves IV and V were also affected by electrode site, consistent with previous reports that they have spatially separated generators in other species. In recordings made with vertex electrodes referenced to the mastoid region ipsilateral to the stimulated ear, wave I appeared as a high-amplitude positive peak with onset latency equalling that in noncephalic reference recordings, but with somewhat later peak latency and longer duration. As a result, wave II was partially or totally obscured so that only 4 major peaks were evident in the BAEP. In contralateral mastoid reference recordings, latency to peak of wave I was unchanged; however, amplitude of all waves was reduced and waves IIa and IIb were not as clearly differentiated as they were in noncephalic reference recordings.     

Changes in oscillatory potentials in the canine electroretinogram during dark adaptation

Oscillatory potentials (OP) and electroretinograms (ERG) were recorded from clinically normal dogs after 5, 10, 15, 20, 30, 40, 50, and 60 minutes of dark adaptation. At the end of the adaptation period, OP were characterized by 5 distinct positive peaks, O1 through O5, with mean latencies of 14.46, 20.24, 27.38, 35.31, and 44.85 ms, respectively, and with mean amplitudes ranging from 7.20 to 34.84 microV. After 60 minutes of dark adaptation, the ERG had a mean a-wave latency of 12.03 ms and a mean b-wave amplitude of 109.29 microV. Peaks O3 and O4, which partially mask the summit of the b-wave, had mean latencies of 28.66 and 36.83 ms, respectively. The mean amplitude of the b-wave measured to the peak of O3 was 240.06 microV and 230.73 microV when measured to peak O4. Changes in the OP during dark adaptation consisted of significant (P less than 0.05) increases in the latencies of O1, O2, and O3, and significant increases in the amplitudes of O1, O3, O4, and O5. Concurrent ERG changes consisted of significant increases in the amplitudes of the a-wave and b-wave measured from O3 and O4, and significant increases in the latencies of peaks O3 and O4 on the b-wave.     

Waveform analysis and reproducibility of visual-evoked potentials in dogs

Visual-evoked potentials (VEP) and electroretinograms (ERG) were recorded from 10 normal light-adapted adult dogs, using a 3 x 5 matrix of light-emitting diodes as a stimulator. Visual-evoked potentials were recorded from 4 scalp electrodes overlying cortical areas, whereas electroretinographic activity was recorded by 2 scalp electrodes placed near the eye and by a conjunctivally placed electrode. The waveform of the VEP consisted of 3 major positive waves (P1 through P3), with peak latencies in the 20- to 70-ms range. Waveform reproducibility was assessed by comparing peak latencies from VEP recorded on 2 separate days approximately 1 week apart. The peak latencies for P1 through P3 did not differ (P greater than or equal to 0.05) between first and second recording sessions. To substantiate the postretinal origin of VEP, recordings were made before and after unilateral optic nerve transsections in 4 dogs. Electroretinograms were also measured before and after surgery to assess the integrity of the retina. Postsurgically, VEP were absent when the eye on the surgically treated side was stimulated. Stimulation of the contralateral eye induced VEP with the same waveform shape, but latencies were slightly prolonged (P less than or equal to 0.05) compared with presurgical recordings. The only effect of optic nerve transsection on the ipsilateral ERG was a prolongation (P less than or equal to 0.05) of the b-wave. However, when postsurgical ERG values were compared with those from the intact side after surgery, there were no differences.     

Recording of Visual-Evoked Potentials in Dogs With Scalp Electrodes

Following unsuccessful attempts to record visual-evoked potentials (VEPs) in dogs with scalp electrodes, adoption of a new stimulation technique seems to be beneficial. Previously, flashes of white light administered after dark adaptation induced relatively high amplitude electroretinograms (ERGs) covering any VEP activity over the surface of the skull. ERG amplitude, however, can be significantly reduced using flashes of red light after light adaptation (mostly cone stimulation). Simultaneous ERG and VEP recording allows identification of VEPs composed of three significantly different negative peaks (N1, N2, and N3) measured in dogs anesthetized with chloralose and halothane. No more than two of the three peaks were seen in one recording. Only the N1 and N3 waves were consistently recorded in dogs anesthetized with thiopental and thiopental combined with halothane. In 50% of all recordings, N1 was seen alone. The other VEPs consisted of N1 and N2, or N1 and N3 occurring concurrently. The simultaneous occurrence of N2 and N3 waves, however, was never seen. Among all recordings, N1 was most frequently recorded (85% of measurements), followed by N3 and N2 (38% and 31% of measurements, respectively). Peaks of less than 90 ms are highly reproducible. Anesthesia is necessary to eliminate frequent artifacts obtained in conscious and sedated dogs. Thiopental and/or halothane had no effect on measured latencies compared with chloralose.     

The effects of cataract stage, lens-induced uveitis and cataract removal on ERG in dogs with cataract

OBJECTIVE: The purpose of this study was to determine the effects of cataract stage, lens-induced uveitis and cataract removal on the electroretinogram (ERG) of dogs with cataract. ANIMALS STUDIED: Fifty-seven dogs diagnosed with unilateral or bilateral cataract whose ERG was recorded at Rakuno Gakuen University Teaching Animal Hospital from 2001 to 2004. PROCEDURES: Four responses were recorded during the ERG: rod ERG, standard combined ERG, single-flash cone ERG and 30-Hz flicker ERG. Cataracts were divided into four stages: incipient, immature, mature and hypermature, and with or without lens induced uveitis (LIU). Noncataractous eyes of dogs with unilateral cataract were used as the control. We compared ERG amplitude, implicit time, and the b- to a-wave amplitude ratio of cataractous vs. noncataractous eyes, preoperative vs. postoperative cataractous eyes, and cataractous eyes with and without LIU. RESULTS: No significant difference was found in ERG amplitude between incipient, immature and hypermature cataractous eyes, while in mature cataractous eyes decreased amplitude was confirmed in all responses compared with control eyes. However, no significant difference in b/a ratio was found at any stage of cataract. In postoperative eyes, increased amplitude was recorded in all responses compared to preoperative values. In eyes with LIU a decreased amplitude in the rod ERG and b-wave of standard combined ERG was recorded and, furthermore, a significant decline was confirmed in b/a ratio. CONCLUSION: ERG values were influenced by cataract stage and LIU. LIU was associated with a reduction in the b/a ratio.     

Ophthalmic and cone derived electrodiagnostic findings in outbred Miniature Long-haired Dachshunds homozygous for a RPGRIP1 mutation

Objective To investigate ophthalmic and cone‐derived electrodiagnostic findings in outbred Miniature Long‐haired Dachshunds (MLHD) homozygous for a mutation in the RPGRIP1 gene previously associated with cone‐rod dystrophy 1 (cord1). Animals A total of 36 MLHD homozygous for the RPGRIP1 mutation and 23 dogs clear of the mutation (control group). Procedures The dogs underwent ophthalmic examination and photopic electroretinogram (ERG) recordings. Results None of the control dogs presented with clinical or ophthalmic signs consistent with cord1. Amongst the dogs homozygous for the mutation one presented with bilateral symmetrical total retinal atrophy. None of the other dogs in this group showed signs consistent with cord1. Photopic ERG recordings were available in 23 control dogs and 34 dogs homozygous for the mutation. Photopic a‐ and b‐waves following four light stimuli (3 cdS/m²) at a rate of 5.1 Hz were not significantly different between groups. The amplitudes of the 30 Hz flicker (128 flashes, 3 cdS/m²) response were significantly reduced in the dogs homozygous for the PRGRIP1 mutation. The difference in age between the two groups did not significantly affect the difference. Conclusion Homozygosity of the RPGRIP1 mutation does not invariably result in early onset cord1. However, cone derived ERG recordings show evidence of a reduced cone or inner retinal function in homozygous but clinically normal MLHD. Modifying genes that have yet to be identified may influence an individual dog’s risk of developing the blinding cord1 and also the age of onset and rate of progression.     

Effects of analog filtering on brain stem auditory-evoked potentials in dogs.

Effects of analog filter frequency on brain stem auditory-evoked potentials (BAEP) were investigated in 7 non-sedated dogs. The BAEP were recorded successively at various low-pass (LP) and high-pass (HP) filter frequency settings. The analog filters had a rolloff of 6 dB/octave. Decrease of LP filter frequency from 30 kHz to 100 Hz caused prolongation of the peak latency and reduction of the peak-to-peak (from a positive peak to the following trough) and absolute (from a positive peak to the baseline) amplitudes for all peaks, except the peak latency for P5 and the absolute amplitude for P4. Changes in these variables were statistically significant (P less than 0.05) at different cutoff frequencies specific for the individual peaks. The interpeak latency between P1 and P4, and P4/P1 peak-to-peak amplitude ratio were not changed significantly. At the lowest LP filter frequency of 100 Hz, positive peaks (fast waves) seemed to be superimposed on a slow positive wave (slow wave). In contrast, increase of HP filter frequency from 0.53 to 160 Hz did not result in significant changes for any peaks, except for reduction in the absolute amplitude of P4. The various effects of LP filter frequency and negligible effects of HP filter frequency on individual peaks may be attributable to their frequency composition and/or elimination of the slow wave at higher HP filter frequency settings. On the basis of our results, LP filter setting of 3 kHz and HP filter setting of less than or equal to 53 Hz are recommended for recording of BAEP in dogs.(ABSTRACT TRUNCATED AT 250 WORDS)     

The normal electroretinogram in adult healthy Shih Tzu dogs using the HMsERG

Electroretinography (ERG) is a reliable diagnostic tool for the diagnosis of retinal disease. It measures electric potentials occurring in the retina in response to light stimulation. In this study, we examined the normal electroretinogram using the Handheld Multispecies ERG (HMsERG) in Shih Tzu dogs. ERG recordings were performed in twelve eyes of six healthy Shih Tzu dogs. Dogs were anesthetized with a combination of medetomidine and ketamine. Proparacaine eye drops were also applied as a topical anesthetic. Tropicamide eye drops were applied for mydriasis. After 20 min of dark adaptation, we recorded the amplitudes and implicit times of the b-waves of the rod, standard rod and cone (Std R&C), high-intensity rod and cone (Hi-int R&C), and cone systems, and responses of the cones and inner retina by flicker light stimulation (cone flicker). Results showed that mean the amplitudes of a-waves of Std R&C, Hi-int R&C, and the cone responses were 141.25 µV, 173.00 µV, and 12.92 µV, respectively. The b-waves of the rod responses ranged from 141.58 to 155.25 µV; the Std R&C was 314.75 µV, the Hi-int R&C was 329.42 µV, the cones were 37.75 µV, and the flicker responses were 64.08 µV. The b/a ratios for the Std R&C, Hi-int R&C, and the cone response were 2.29, 1.94, and 3.71, respectively. Mean implicit time of the a-wave of the Std R&C was 15.12 ms, of Hi-int R&C was 13.42 ms, and of the cone response was 7.22 ms. The b-wave of the rod responses ranged from 68.12 to 72.68 ms, of Std R&C were 37.28 ms, of Hi-int R&C were 41.90, of the cone responses were 38.12 ms, and of the cone flicker responses were 22.80 ms. We believe that these parameters can be used as reference "normal" ERGs ranges for Shih Tzu dogs using the HMsERG under medetomidine and ketamine anesthesia.     

Oscillating potentials on the B-wave of the ERG in the dog]

The oscillatory potentials (OP) on the b-wave of the canine ERG are characterized. Normal values for an OP-index, implicit times and periodicity are given. The OP-index increases during dark adaptation as well as with increasing stimulus intensities for white light flashes. Scotopic blue and red stimuli, although balanced for the maximum b-wave amplitude, surprisingly result in higher OP-indices for red flashes. Implicit times decrease for all OP with increasing stimulus intensities. Scotopic balanced red light results in OP with markedly reduced implicit times compared to blue light stimulation. The intervals between two OPs remain constant during dark adaptation, however, intervals between OPs with longer implicit times tend to be shorter. In contrast, the b-waves for scotopic balanced stimuli show rhythmic oscillations with constant intervals between them, the intervals being shorter for blue light than for red light. The intervals between OPs tend to become shorter with increasing stimulus intensities. This investigation indicates that OPs are influenced but not generated by photoreceptors. The examination of a dog with optic nerve hypoplasia indicates that the ganglion cells do not contribute to the biogenesis of OPs. The prognostic value of OPs in the early diagnosis for hereditary retinal degenerations in the dog is discussed.     

Comparing the photopic ERG i-wave in different species

The i‐wave, a post b‐wave component of the human photopic electroretinogram (ERG), is claimed to originate at the level of the retinal ganglion cells (RGC) or more distally. We investigated whether this wave is a feature common to all species. Photopic ERGs were obtained from the following species: Beagle dog, European cat, New Zealand white rabbit, Göttingen minipig, Cynomolgus monkey, Sprague–Dawley and brown Norway rats, Hartley guinea pig, and CD1 and C57BL6 mice. Results were compared with those obtained from normal human subjects. Except for rats and mice, all species yielded a well‐demarcated i‐wave, easily identifiable and separated from the a–b‐wave complex by ≈ 20 ms. Our sample suggests that the i‐wave is a feature common to the photopic ERG of most species including humans. In view of its suggested origin, the i‐wave would offer a unique opportunity to test, with the flash ERG, the functional integrity of the retinal ganglion cells in animals where use of a pattern stimulus is not always easily obtained.     

Effects of stimulus intensity for electroretinogram in conscious Miniature Schnauzers

The aim of this study was to determine the most effective light intensity for flash electroretinogram (ERG) examination in conscious dogs using ERG equipment with a contact lens electrode with a built-in LED light source. ERG was performed on the bilateral eyes of ten clinically healthy Miniature Schnauzers at 6 different intensities (0.025, 0.079, 0.25, 0.79, 2.5 and 7.9 cd.s/m2) after dark adaptation for 20 min. With the increase in stimulus intensity, the most significant increase in a and b-wave amplitudes were observed at 2.5 cd.s/m2 (p<0.05). As the intensity of light was increased, the implicit times of both waves significantly decreased. Therefore, the most effective intensity of stimulus was 2.5 cd.s/m2 in the conscious Miniature Schnauzers. This suggests that this procedure would be applicable for evaluation of retinal function in conscious dogs, especially in high-risk patients.     

Peaks of brainstem auditory evoked potentials in dogs

The effects of electrode configuration and click polarity on brainstem auditory evoked potentials (BAEP) in dogs were investigated to clarify the inconsistent nomenclature for each peak. Four positive peaks (waves 1, 2, 3 and 4) before a deep negative trough and a fifth positive peak (wave 5) following the trough were the basic components of BAEP in dogs, which were easily identified regardless of recording conditions such as electrode configuration and click polarity. Additional peaks tended to be present when a noncephalic reference electrode and/or single-polarity (rarefaction or condensation) click stimuli were used. The Roman nomenclature for the individual positive peaks of BAEP in dogs is confused owing to variations in the observed waveforms among researchers, but click polarity and/or reference electrode position can explain all the previously reported variations in BAEP waveforms in dogs. When the criteria concerning wave V in the guidelines of BAEP in human beings are applied to avoid further confusion of Roman nomenclature in dogs, it is recommended that the basic five positive peaks (waves 1, 2, 3, 4 and 5 as identified easily with Ai-Vertex configuration and alternating clicks) are designated as waves I, II, III, V and VI, respectively. Wave IV (wave 3b) occurs occasionally before wave V in dogs.Abbreviations BAEP brainstem auditory evoked potentials - dBHL dB hearing level - IPL interpeak latency - Ai the caudodorsal end of the zygomatic arch ipsilateral to the stimulated ear - Nape the neck over the spinous process of the fourth cervical vertebra     

Comparison of two electroretinography systems used in dogs: the HMsERG and the RETIport

The purpose of the study was to compare two different electroretinography (ERG) instruments used on the same animal in a laboratory setting. Retinal function in both eyes of 12 healthy miniature schnauzers was evaluated under general anesthesia. Scotopic and photopic ERGs were recorded by the HMsERG and the RETIport using the Dog Standard Protocol recommended by the European College of Veterinary Ophthalmologists (ECVO). The waveforms of the ERGs obtained by the two ERG units were similar to those described in previous studies. The 90% reference ranges using the multiple of medians (MoM) showed that the changes in ERG parameters obtained by the HMsERG unit were very similar to those of the RETIport for most ERG responses, except for a few. The results demonstrate that the two ERG systems are comparable for evaluating retinal function in dogs. Further, the results also show that it is necessary to establish ERG system-specific limits of normality in each laboratory or clinic in order to obtain a correct diagnosis, for example by using a graphical representation of the 90% reference range.     

The neuroprotective effect of submicron and blended Lycium barbarum for experiment retinal ischemia and reperfusion injury in rats

The purpose of this study was to investigate the neuroprotective potential of submicron (milled) and blended Lycium barbarum (LB) in glaucomatous retinal neuropathy using a rat model of high intraocular pressure (HIOP) induced retinal ischemia. The rats were treated with 500, 250, 100 mg/kg LB (submicron or blended form) orally once daily for 56 days respectively after 1 week of retinal ischemia induction. We conducted electroretinography (ERG), histopathological analysis in retina and antioxidative level assays, such as total glutathione (GSH (glutathione) + reduced glutathione) + GSSH (glutathione disulfide), catalase activity, SOD (superoxide dismutase) activity, and lipid peroxidant malondialdehyde (MDA) in the retina and plasma of test rats. The results indicated that the amplitudes of a and b wave of ERG were preserved in rats treated with submicron and blended LB groups, the best protective effect on ERG b wave amplitudes was observed at the dosage of 250 mg/kg of both forms of LB. Retinal thickness was best preserved, particularly significant in the retinal inner nuclear layer in submicron 250 mg/kg LB group. The levels of antioxidant GSSH+GSH, SOD and catalase activity in the retina were higher in blended 500 mg/kg and submicron 250 mg/kg groups than other groups, while the MDA level was lower in submicron LB groups than that in blended LB and non-LB IR group. In the plasma, there was no significant difference in the levels of GSSH+GSH and catalase activity between treated groups, but higher levels of SOD and lower levels of MDA were observed in 250 mg/kg submicron and 500 mg/kg submicron LB groups than the blended LB and non-LB IR groups. Generally better antioxidative effects were observed in the submicron LB than blended LB among treated groups, especially the 250 mg/kg submicron LB, providing good retinal neuroprotection by preserving retinal structure and function with improved antioxidative capacity. The submicron LB may have clinical implication as an adjuvant therapy of oxidative stress and retinal damage caused by HIOP induced retinal ischemia and reperfusion injury.