Retinal ganglion cell activity from the multifocal electroretinogram in pig: optic nerve section, anaesthesia and intravitreal tet
1 Retina and Optic Nerve Research Laboratory
2 Department Physiology & Biophysics
3 Department Ophthalmology and Visual Sciences, Dalhousie University, Halifax, Nova Scotia, Canada B3H 4H7
Abstract
Non-invasive recordings of the retinal activity have an important role to play in the diagnosis of retinal pathologies. The detection of diseases that involve retinal ganglion cells (RGCs), such as optic atrophy and glaucoma, may be improved by isolating the RGC contribution from the multifocal electroretinogram (mfERG). In this study, mfERGs were performed on 20 pigs, 1–6 weeks following unilateral retrobulbar optic nerve section (ONS). The stimuli were 103 non-scaled high-contrast hexagons from which summed and individual mfERG responses were obtained in experimental and control fellow eyes under conditions of ketamine (n= 11) or isoflurane anaesthesia (n= 9). The effect of intravitreal injection of tetrodotoxin (TTX; n= 6) was also investigated. The summed mfERG responses showed a first positive peak (P1) with a short latency (21 ms) followed by two smaller peaks (P2 and P3) of longer latency (46 and 65 ms, respectively). While P2 and P3 amplitude were highly correlated with the time post-optic nerve section (ONS) (P2: r2= 0.669; P= 0.007; P3: r2= 0.651; P= 0.005), P1 was not (r2= 0.193; P= 0.38). P1 and P2 showed no implicit time variation as a function of retinal location, while P3 implicit time varied along the axis of the visual streak, generating a naso-temporal asymmetry. However, the P3 implicit time did not vary consistently with distance away from the optic nerve head. Intravitreal injections of TTX reduced P2 and P3 in the control eyes, consistent with the effect of ONS, and also induced a series of regular oscillations lasting up to 200 ms post stimulus. Under isoflurane anaesthesia, all components of the mfERG ifn experimental and control eyes were, at all time points post-ONS, of similar amplitude and without naso-temporal asymmetry, suggesting a reduced participation of RGCs under these anaesthesic conditions. These data clearly demonstrate that it is possible to isolate the RGC contribution from non-invasive multifocal electroretinography.
, 百拇医药
Introduction
Traditionally, non-invasive clinical electrophysiological techniques for retinal assessment were considered insensitive to the relatively small number of retinal ganglion cells (RGCs) present in the retina. Recently, the photopic negative response and the scotopic threshold response were isolated from the Ganzfeld electroretinogram (ERG). These responses were shown to be affected in experimental models of optic neuropathy (Sieving et al. 1986; Viswanathan et al. 1999; Bui & Fortune, 2003), in patients with glaucoma (Colotto et al. 2000; Cursiefen et al. 2000; Drasdo et al. 2001; Viswanathan et al. 2001) and by intravitreal injection of the voltage-gated sodium channel blocking agent tetrodotoxin (TTX). This finding suggests these responses might be at least partially generated by RGCs. Pattern stimuli have been used conventionally to elicit proportionally larger RGC responses. The potential generated by this technique called pattern electroretinography (PERG) was shown to be sensitive to optic nerve section (ONS) in rat (Berardi et al. 1990), cat (Mafei & Fiorentini, 1981) and monkey (Maffei et al. 1985). However, the use of PERG in clinical settings is still debatable, mostly because of the high inter- and intrasubject variability (Bach & Speidel-Fiaux, 1989; Otto & Bach, 1996; Korth, 1997; Holder, 2001). Both ERG and PERG techniques are corneal potentials elicited by stimulation of the whole visual field and therefore, are relatively insensitive to local defects.
, 百拇医药
Multifocal-electroretinography (mfERG) is a newly introduced method for retinal electrophysiology (Sutter & Tran, 1992) that allows the isolation of multiple local responses and the topographical evaluation of the retina. The rapid rate of stimulation and the algorithm used to reconstruct the individual signals result in a recombination of retinal dipoles different from those of the ERG and PERG (Hood et al. 2002). Two approaches have been used to isolate the RGC contribution to the mfERG. First, a computational approach was found to isolate the optic nerve head component, a signal whose implicit time varied with the distance to the optic disc and thought to be generated by the RGC axons at the optic nerve head (Sutter & Bearse, 1999; Hood et al. 2001). This component proved to be eliminated by intravitreal injection of TTX in the monkey (Hood et al. 1999a; Frishman et al. 2000; Fortune et al. 2002b) and was found to be reduced in patients with glaucomatous neuropathy (Bearse et al. 1996). However, its sensitivity to detect early glaucomatous retinopathy is controversial (Fortune et al. 2002a; Palmowski et al. 2002). The second approach is based on the observation of the mfERG signal itself. Loss of naso-temporal asymmetry in monkeys was found to be associated with advanced glaucomatous changes (Hare et al. 2001; Fortune et al. 2002a; Harwerth et al. 2002). In humans with glaucomatous optic neuropathy, responses were found to be variable and no consensus has yet emerged (Chan & Brown, 1999; Hasegawa et al. 2000; Hood et al. 2000; Klistorner & Graham, 2000; Fortune et al. 2001; Sano et al. 2002).
, 百拇医药
The pig as an animal for vision research has been extensively used for corneal and anterior segment research and interest is growing for posterior segment research including recently developed models for glaucoma (Rosolen et al. 2003; Ruiz-Ederra et al. 2005), photoreceptor degeneration (Tso et al. 1997; Li et al. 1998), retinal transplant (Ghosh & Arner, 2002), and retinal prosthesis (Sachs et al. 2005). The morphometry of the pig eye is similar to human eye (Prince et al. 1960), allowing the development of surgical and transplant procedures that can be translated to the medical field. The pig retina is rich in cones (14–17 millions; Chandler, 1999; Hendrickson & Hicks, 2002), with a cone: rod ratio close to the one found in the central human retina and much higher than the one found in the uniform retina of rodents, allowing the use of local visual stimuli to isolate the RGC signal. In this study, ONS (Berkemeier et al. 1994; Garcia-Valenzuela et al. 1994; Vaegan & Millar, 2000) was used to produce progressive retrograde RGC degeneration in the pig retina. Individual components of the summed mfERG responses were correlated to the time post-ONS. In addition, the differential effect of intravitreal injection of TTX was examined and compared to the effect of RGC degeneration. Finally, the impact of isoflurane, a halogenated anaesthetic widely used in large animals, on the mfERG signal was investigated.
, 百拇医药
Methods
Animals
Electroretinographic data was obtained from a total of 20 Hampshire Durco Cross pigs aged 3–5 weeks. The effect of unilateral ONS, used here to produce a controlled progressive RGC degeneration, was studied using the traditionally recommended isoflurane anaesthesia (n= 9 experimental eyes; n= 9 control fellow eyes) and with ketamine sedation (n= 9 experimental eyes; n= 11 control eyes). Of the 10 animals tested under ketamine sedation, four were also studied by reintroducing isoflurane anaesthesia, and six received intravitreal injections of TTX (n= 2 experimental eyes; n= 6 control eyes). All procedures were conducted in accordance with the Association for Research in Vision Statement for the Use of Animals in Ophthalmic and Vision Research and with the guidelines established by the Canadian Council for Animal Care. Ethical approval was obtained from the Dalhousie University Committee for Laboratory Animals.
, 百拇医药
Surgical procedures
Anaesthesia. Animals were fasted for a period of 12 h prior to anaesthesia but had free access to water. Pre-medication included atropine (0.1 mg kg–1I.M.; Abbot Laboratories, Montreal, QC, Canada) and ketamine (Ketalar, 18 mg kg–1I.M.; Birmeda-MTC, Cambridge, ON, Canada). Deep anaesthesia was induced with 5% isoflurane (AErrane; Abbot Laboratories) and administered with 100% oxygen delivered through a facemask. Endotracheal intubation was performed and artificial ventilation was supplied with 100% oxygen (volume: 10 ml kg–1; rate: 13 breaths min–1; Ohio Medical Instrument Co. Inc., Cincinnati, OH, USA). The isoflurane level was reduced to a maintenance level of 2.0–2.5%. The pigs were kept hydrated via lactated Ringer solution delivered through a cannulated dorsal auricular vein. Body temperature was maintained at 38.0–39.0°C using a circulating hot water heating pad (Aquamatic K Module; American Medical Systems, Valencia, CA, USA). Both heart rate and oxygen saturation levels were monitored (SDI Vet/OxTM 4403; SDI Sensor Devices Inc., Wankesha, WI, USA) throughout the procedures.
, 百拇医药
Optic nerve section. ONS was performed under deep isoflurane anaesthesia and local xylocaine 2% skin infiltration. A small incision was made at the temporal canthus of the eye. Sutures were attached to the superior conjunctiva to help rotate the eyeball downward. The cornea was protected from desiccation with a thick coat of ophthalmic ointment (Optimyxin, Sabex, Boucherville, QC, Canada). The conjunctiva was then incised and the lateral rectus resected. Using the lateral episcleral vein as a guide, adipose tissue was dissected until the optic nerve was exposed. The dura was incised longitudinally for 3 mm and a pair of blunt Vannass scissors was used to cut the nerve within the dura in order to protect the adjacent ophthalmic artery. A piece of spongel soaked in fluorogold 2% (Fluorochrome Inc., Denver, CO, USA) was then introduced in the cavity to retrogradely label RGCs for histology. The lateral rectus was sutured back in place as well as the conjunctiva. The area was infiltrated with gentamycin (100 mg ml–1; Sigma Aldrich Canada Ltd, Oakville, ON, Canada) and the skin sutured. Fundus examination using direct opththalmoscopy was used to ensure the integrity of the retinal circulation. Post-operative care included intramuscular administration of buprenorphine hydrochloride 0.01 mg kg–1 (Reckitt & Colman Products, Hull, UK) 15 min before the cessation of anaesthesia and salicylic acid (10 mg kg–1; Bayer Inc., Etobicoke, ON, Canada) given orally for 2 days post-surgery.
, http://www.100md.com
Recording sessions
Protocol. Recording sessions were performed prior to surgery (n= 2) and once during the 6 weeks post-ONS period (1 week: n= 2; 2 weeks: n= 2; 3 weeks: n= 2; 6 weeks: n= 3). Animals were anaesthetized prior to all ERG recordings using the previously described isoflurane protocol and secured in a custom-made sling equipped with a head holder. For a group of 11 animals, the anaesthesia regimen was switched to ketamine sedation (20 mg kg–1 h–1I.V.) in combination with the muscle relaxant gallamine (35 mg kg–1 h–1, I.V.) and artificial ventilation. For this group, recordings were performed after a minimum of 20 min from the time isoflurane was discontinued to allow its clearance from the body fluids (Ghouri et al. 1991). TTX (0.0839 μg μl–1; Sigma Aldrich Canada Ltd) was injected into the vitreous of six eyes (2 experimental and 4 control eyes) through the pars plana using a 30-gauge needle, at a final concentration of 5 μM based on an estimated total vitreal volume of 2.0 ml. An ophthalmoscopic examination was performed to ensure the needle did not penetrate the retina and cause oedema or detachment. ERG recordings were obtained 90–120 min following the injection. In another group, experimental and fellow eye recordings were performed under isoflurane anaesthesia (n= 7) or obtained prior to the switch to ketamine (n= 2).
, http://www.100md.com
Recording procedures. The active electrode consisted of a monopolar contact lens (Medical Workshop b.v., Groningen, Holland) filled with a protective coat of hydroxypropyl methylcellulose 1% (Alcon Canada Inc., Mississauga, ON, Canada). Retinoscopy revealed a refractive error generally of less than –2.00 diopters. mfERG recordings with and without optical correction yielded similar results and consequently, no optical correction was used. The fellow eye was occluded throughout the recording sessions. Subdermal platinum needle electrodes (Grass Instruments, Quincy, MA, USA), inserted on the upper eyelid of the tested eye and on the nose, were used as reference and ground electrodes, respectively. The continuously recorded retinal signal was amplified (5.0 x 105), bandpass filtered between 3 and 300 Hz (P511; Grass Instruments Inc.) and digitized in synchrony with the display at a sampling rate of 16 samples per display frame. The signal responses obtained were extracted and processed using VERIS software version 4.7 from EDI (Electro-Diagnostic-Imaging, San Mateo, CA, USA).
, 百拇医药
Multifocal-electroretinography. The mfERG stimuli with a refresh rate of 13.3 ms per frame were displayed on a 21 in computer monitor (Viewsonic P815, Walnut, CA, USA) positioned 25 cm away from the tested eye. The stimulus consisted of 103 non-scaled white (130 cd m–2) and black (1.5 cd m–2) hexagons presented against a 25 cd m–2 grey background. The VERIS software was used to present the hexagons following a pseudo-random binary sequence (the m-sequence) of white and black states. M-sequences of either 215–1 or 214–1 length (corresponding to approximately 7 and 3.5 min recordings) were used, as responses obtained in the two conditions were very similar. Responses from either the whole stimulus area or from small groups of five to six selected hexagons were examined. The VERIS software also generated a 3-D representation of the individual responses from all hexagons. Since non-scaled hexagons were used, these response density maps were computed from the root mean square (RMS) of the trace values, without template. Examination of these maps allowed the identification of retinal structures such as the optic nerve head and the visual streak (a specialized area of the porcine retina with higher cell density). The final position of the monitor was determined after an initial estimate of the optic disc position from indirect ophthalmoscopy, the location of the blind spot on RMS maps obtained subsequently after a short recording being the determinant parameter used. In all recordings retained for this study, both the optic nerve head and the visual streak were identified. The visual streak was sometimes slightly inclined relative to the horizontal axis of the stimulus array due to constraints in positioning the pig's head relative to the monitor. This was taken into consideration when corresponding locations from the two eyes had to be compared.
, 百拇医药
RGC damage post-ONS
Animals were killed using sodium pentobarbital overdose (Euthasol, 0.3 mg kg–1) and both eyes were enucleated. The retinae were dissected after 2.5–3.5 h fixation in 4% paraformaldehyde and positioned on a slide with the inner retina facing up, then mounted in an aqueous medium (Citifluor, glycerol–PBS solution; Marivac, Halifax, NS, Canada), cover-slipped and sealed.
Retinal whole mounts were examined under fluorescence microscopy (Nikon E800). Three digital images, each covering a retinal area of 640 x 480 μm2, were acquired (Image Pro Plus; Carsen Group Inc., Markham, ON, Canada) from the visual streak area, 1.5 mm above and 3.0 mm temporal to the optic nerve head, as well as in more peripheral areas at 5.0 mm and 10.0 mm above this point in the superior retina.
, 百拇医药
Retinae were obtained from control eyes 3 days post-ONS to allow retrograde labelling. The RGC peak density was 2940 ± 697 cells mm–2 (n= 5), tapering rapidly to 489 ± 140 cells mm–2 in the peripheral retina. These figures are compatible with previously published RGC counts in pigs (Vecino et al. 2004). Retinae were also collected from experimental eyes. The variability in the staining of the experimental eyes prevented a detailed analysis of the RGC loss in function of time as originally intended. However, the presence of pyknotic bodies and glial cells filled with fluorogold debris from dying RGCs were taken as evidence for RGC loss in retina obtained 1–3 weeks post-ONS. Retinae collected after 6 weeks post-ONS showed a reduction of RGC, with densities averaging 658 ± 309 cells mm–2 for the peak density and 166 ± 88 cells mm–2 for the periphery (n= 10, P < 0.05, t test).
, http://www.100md.com
Data analysis
Quantitative measurements were performed on amplitude and implicit time of the summed mfERG responses computed from the 103 individual, first kernel responses. One major positive component, P1, occurring in a post-stimulus time window of 10–30 ms, was quantified by measuring the amplitude between the first negative trough and the first positive peak (Fig. 1). This high-amplitude response was followed by components of smaller amplitude spanning 30–80 ms. P2 amplitude was defined as the component corresponding to the second trough to the peak of the intermediate positive component which normally occurred around 45 ms. P3 was defined as the last positive component measured from the small negative through following P2 to the most positive voltage of the 60–80 ms segment of the response.
, http://www.100md.com
Representative example of control mfERG responses obtained to high contrast, 13.3 Hz, 103 non-scaled hexagons stimuli configuration. A, summed first-order response: the amplitude of the three positive components is indicated by the vertical arrows while the time taken by the individual components to develop is identified by the three horizontal arrows: B, individual recordings: C, RMS topographical map: the large arrow corresponds to the optic nerve head and the two small arrows identify the axis of the visual streak that lies inferior to the optic nerve head in this visual field representation.
, http://www.100md.com
The summed mfERG data obtained from the control and experimental eyes, under ketamine sedation and at post-ONS time longer than 2 weeks, were compared using ANOVA with Bonferroni/Dunn post hoc correction. The effect of anaesthesia was analysed by comparing data from the control and the experimental eyes under isoflurane anaesthesia and by comparing data from the control eyes under isoflurane anaesthesia with either the control or the experimental eyes recorded under ketamine sedation.
, http://www.100md.com
The RGC component was isolated off-line by exporting to a spreadsheet the responses from manually selected groups of five hexagons located in corresponding areas of the control and experimental eyes. The RMS maps from each eye were used to match the locations selected, using the position of the blind spot and the orientation of the visual streak as the reference points. Corresponding responses from experimental locations were then digitally subtracted from those of control locations to provide a signal representative of the RGC activity.
, 百拇医药
Results
mfERG recordings in control eyes
Representative first kernel control mfERG responses to 103 non-scaled, high contrast hexagon stimuli are documented in Fig. 1. The summed mfERG response consisted of an initial major positive component, P1, followed by two more variable positive components identified here as P2 and P3. No consistent stimulus-associated activity could be detected beyond 80 ms. The raw recordings, as well as the RMS map, showed a well-delineated zone of minimal activity corresponding to the optic nerve head, and a band of larger responses extending horizontally, immediately below the optic nerve head in visual field view (Chandler, 1999).
, 百拇医药
Effect of optic nerve section
The effect of ONS on summed, high-contrast mfERG responses is summarized in Table 1. Data from all experimental eyes recorded more than 14 days post-ONS were pooled (n= 7, mean time post-ONS 28.9 days). In the experimental eye, under ketamine sedation, P1 was mildly reduced in amplitude while P2 and P3 showed a highly significantly reduction in amplitude. Long-term ONS did not have any significant effect on implicit times. In Fig. 2, representative examples of summed responses from the control and the experimental eyes of four individual pigs recorded at 1, 2, 3 and 6 weeks post-ONS are presented. At 1 week post-ONS, responses were very similar. However, at 2 weeks post-ONS, P3 showed a consistent reduction in amplitude while at 6 weeks post-ONS, P3 was nearly absent and the negative trough preceding P2 was less pronounced, contributing to the decrease in P2 amplitude. The ratios of the experimental to the control responses for P1, P2 and P3 are presented in Fig. 2B (data recorded from two animals pre-ONS and two animals at 1 week post-ONS were included). For P1, a best-fit linear correlation (0.886 – 0.028x; r2= 0.193; P= 0.38) suggests a non-statistically significant change over time. On the other hand, decrease in amplitude for P2 (0.837 – 0.079x; r2= 0.669; P= 0.007) and P3 (0.781 – 0.087x; r2= 0.651; P= 0.005) was significant and, by 6 weeks post-ONS, the predicted amplitude ratios between experimental and fellow control responses were 0.36 and 0.26 for P2 and P3, respectively.
, 百拇医药
A, representative responses obtained in control (Ctl; left column) and experimental (Exp; right column) eyes, showing progressive loss of P3 and P2 components: B, scattergrams of the amplitude ratios between experimental and control eyes for P1, P2 and P3, at 1, 2, 3 and 6 weeks post-ONS.
RMS maps constructed from P1 (10–30 ms) of the signal found in a representative control eye, showed a horizontal band of higher activity corresponding to the visual streak (Fig. 3). A similar map was obtained for P2 (30–50 ms). For the map constructed around P3 (50–75 ms), the visual streak was no longer discernable, suggesting that P3 amplitude map did not correspond to known RGC cell distribution. In the experimental eye, the maps from the 10–30 ms and 30–50 ms segments, P1 and P2, respectively, were similar to the maps from the fellow control eye in showing a concentration of activity in the visual streak. The scalar product map from the 50–75 ms segment P3 showed a uniform and very low voltage activity corresponding to the low voltage P3.
, 百拇医药
RMS topographical maps showing that P1 (10–30 ms) and P2 (30–50 ms) segments of the first kernel of the mfERG responses in a control (Ctl; left column) and experimental (14 days post-ONS; Exp, right column) eye are scaled by the higher cell density of the visual streak while P3 (50–75 ms) is presenting with a more uniform amplitude distribution. Open arrow: blind spot. S: superior; I: inferior; N: nasal; T: temporal fields. Gray scale calibration as indicated.
, 百拇医药 Local mfERG responses to high contrast, 103 hexagon stimuli in the control (Ctl OS; left column) and experimental (Exp OD; middle column) eyes of a pig 14 days post-ONS. Local, summed (3 hexagons) mfERG responses spanning the temporal (trace 1) to the nasal (trace 3) aspects of the visual streak and of the mid-periphery (traces 4–6). The stimulus array has been located in corresponding positions for the two eyes, as demonstrated by the position of the blind spot (indicated by an x in the insert). A marked naso-temporal asymmetry was observed in the control eye for the visual streak, particularly evident with the morphology of the P2–P3 complex. In the mid-periphery, this asymmetry was much less evident. In the experimental eye, the asymmetry disappeared, leaving similar recordings in all positions. The rightmost column (Ctl - Exp) shows the computer subtraction between control and experimental eyes, delineating the activity assumed to be generated by the retinal ganglion cells. Dashed lines centred on the P3 wave.
, 百拇医药
Local mfERG responses to high contrast, 103 hexagon stimuli in the control and experimental eye of a pig 14 days post-ONS. Five summed (5 hexagons) mfERG responses to a control (Ctl OS; left column) and fellow experimental (Exp OD; middle column) eye as function of the distance from the optic nerve head position (x) with group 1 being close to the optic nerve and group 5 being about 50 deg away from the optic nerve head position. The rightmost column shows the computer subtraction between control and experimental eye, delineating the activity assumed to be generated by the retinal ganglion cells. No implicit time changes in function of distance (dashed lines corresponding to P3) could be observed.
, 百拇医药
Effect of TTX
Intravitreal injection of TTX resulted in a change in the wave morphology of the summed mfERG response, with reduction in the amplitude of P1, as well as the disappearance of P2 and P3 and their replacement by a series of regular oscillations (Fig. 6). These oscillations had a frequency between 45 and 50 Hz, different from the 60 Hz line interference and from the 75 Hz refresh rate of the monitor. They decreased after 200 ms post-stimulus, a time much longer than the activity elicited in the control conditions. These oscillations were found in the control as well as in the experimental eyes, in equal number, with the same frequency and the same temporal decrease. The introduction of isoflurane produced a reversible decrease in all the TTX-induced oscillations, supporting the idea these oscillations originated from retinal generators and were not artifacts induced by the stimulation protocol. Furthermore, the naso-temporal asymmetry previously described in control retinae was absent after TTX injection.
, http://www.100md.com
A, effect of TTX intraocular injection on summed mfERG responses from control (Ctl OD) and experimental (Exp OS) eyes (55 days post-ONS) of one pig. TTX induced a series of oscillations in both the control and experimental eyes (traces 2 and 6), oscillations that were cancelled by isoflurane anaesthesia (traces 3 and 7) and recovered when isoflurane was discontinued (traces 4 and 8). B, the injection of TTX under ketamine masked the asymmetry normally present along the visual streak in the control eye (grouped hexagons as shown in insert, traces 9–12).
, 百拇医药
Effects of anaesthesia
The consequences of ONS on the mfERG responses as described above were all observed on animals maintained under ketamine sedation. The use of isoflurane anaesthesia (Table 1 and Fig. 7) produced striking differences. In comparing the mfERG summed responses of control and experimental eyes, the amplitude differences observed on the animal maintained under ketamine sedation could no longer be observed. Isoflurane anaesthesia significantly decreased the amplitude of P3 compared with recordings obtained under ketamine sedation (5.4 ± 2.7 nV deg–2versus 10.8 ± 5.8 nV deg–2), to the extent it did not show significant difference compared with P3 recorded in the experimental eye under ketamine (P= 0.12). Under isoflurane anaesthesia, prolonged implicit time could be observed only for P1. Therefore, isoflurane suppressed the components associated with the RGC activity while producing minimal changes to other components.
, 百拇医药
A, from control (Ctl) and experimental (Exp) eye of one animal, 3 weeks post-ONS, in conditions of ketamine sedation (traces 1–6) and isoflurane anaesthesia (traces 7–12), illustrating the fact that recordings obtained under isoflurane were not showing asymmetry in the implicit time of P3 along the axis of the visual streak (upper traces in each subset corresponding to nasal location) in the control eye (traces 7–9). B, summed mfERG responses in a control eye before (trace 1); 10, 20 and 30 min after isoflurane anaesthesia was discontinued, to be replaced by an intravenous ketamine regimen (traces 2, 3 and 4); and when isoflurane anaesthesia was reintroduced while the ketamine sedation was still effective (trace 5).
, 百拇医药
In two control eyes, recordings were first obtained under isoflurane anaesthesia, after which isoflurane was discontinued and replaced by intravenous injection of ketamine (Fig. 7). Components proposed to be associated with RGC activity progressively appeared after 20–30 min, a time corresponding to the estimated clearance time of isoflurane from the body fluids (Ghouri et al. 1991). When isoflurane was reintroduced, a reduction in amplitude of those same components was observed, along with prolongation of implicit time, to bring the recordings back to the initial conditions. This occurred despite the fact that ketamine was still present in the body fluids (t1/2 of ketamine is around 3 h (Clements & Nimmo, 1981)), suggesting that isoflurane has an action on retinal synaptic circuits upstream to those affected by ketamine. The asymmetry observed along the visual streak under ketamine sedation was not present under isoflurane anaesthesia.
, 百拇医药
Discussion
The present study used a complete optic nerve section in the pig to reliably decrease the RGC population, isolating its contribution to the mfERG signal. RGC activity was associated with the mfERG P2 and P3 components that showed a time-related amplitude reduction after ONS. Variation in the implicit time of P3 is responsible for the asymmetry of the mfERG responses between the temporal and nasal hemiretinae; however, P3 implicit time did not vary systematically with the distance from the optic nerve head, making this component distinct from the optic nerve head component previously described (Sutter & Bearse, 1999). TTX injection resulted in an amplitude reduction of the components associated with RGC activity, in keeping with results obtained in monkeys (Hood et al. 1999a,b; Frishman et al. 2000). However, in the pig, TTX induced a series of regular oscillations in both control and experimental eye recordings that has not been reported in monkeys. The RGC activity was sensitive to the type of anaesthesia as recordings obtained under isoflurane anaesthesia showed no statistical difference in amplitude and implicit time between eyes with and without ONS. In the pig, a solid, well-delineated mfERG RGC component can thus be reliably isolated.
, http://www.100md.com
The pig mfERG
All the recordings included in this study were characterized by a well-delineated optic nerve head and a visual streak activity rendering in RMS maps. The ratio between the maximum amplitude responses (corresponding to the visual streak) and the peripheral responses (excluding the area corresponding to the optic nerve head) for P1 varied between 1.5: 1 and 4.6: 1, consistent with the photoreceptor ratio of 3.5: 1 (density of 25 000 cells mm–2 in the visual streak and 8000 cells mm–2 in periphery) reported in histological studies (Chandler, 1999; Hendrickson & Hicks, 2002).
, 百拇医药
It has been reported that the visual streak shows a higher RGC density (range of 2600 cells mm–2), with a decline to about 700 cells mm–2 in the mid-periphery, providing a ratio of about 3.7: 1 (Garcia et al. 2005). The amplitude ratio calculated for P3 in this study varied between 1.1: 1 and 1.7: 1, far from what would have been predicted from the cell density. Furthermore, the RMS map for P3 (Fig. 3) did not show any evidence of activity distribution suggestive of a higher RGC activity in the visual streak. This lack of correlation between mfERG activity and documented RGC distribution may be partly due to the naso-temporal asymmetry observed for the P3 component (Fig. 5); P3 implicit time varied between 58 ms in the nasal retina and 78 ms in the temporal retina. When P3 had a short implicit time it merged with the preceding P2 wave. Further detailed analysis on the isolated RGC component will be necessary to fully resolve this issue.
, 百拇医药
The RGC component of the mfERG isolated after optic nerve section
ONS is a well-documented method of reducing the number of RGCs in a time-related manner (Berkemeier et al. 1994; Garcia-Valenzuela et al. 1994; Levin, 2001; Watanabe et al. 2001; Weishaupt & Bahr, 2001) with little, if any, consequences in other retinal cell populations (Komaromy et al. 2003; Germain et al. 2004; Kielczewski et al. 2005). Functional consequences are not as well known. In our study, the mfERG recordings after ONS suggest that two components of the mfERG summed response are sensitive to the reduction of the RGC population (Table 1 and Fig. 2). A small positive component of long implicit time (P3; range of 70 ms post-stimulus) is more severely affected, being reduced to less than 20% seven weeks post-ONS. Changes in the amplitude of the negative trough between P1 and P2 were also observed to parallel the changes in P3. Changes in P1 were not significant but a trend toward decreasing amplitude was nevertheless observed. These changes are probably due to a decreased RGC response and not a consequence of the surgical procedures on the outer retinal activity for the following reasons. First, visual inspection of the retina did not reveal any retinal detachment or evidence of retinal ischaemia that could have resulted in a decreased signal. Second, the ERG a- and b-waves did not differ significantly between the control and experimental eye (data not shown). Third, the peak amplitude of P1 did not differ between control and experimental eyes when recordings were performed under isoflurane anaesthesia, a mode of anaesthesia we showed to reduce the activity of the inner retina (as described in the following sections).
, http://www.100md.com
Hare et al. (Hare & Ton, 2002) obtained similar findings (i.e. reduction in the second trough and loss of a late positive component) in cynomolgous monkeys after the RGC population was reduced by an increase in intraocular pressure. Using a bipolar contact lens along with a stimulation protocol including interleaving flashes, Fortune et al. (2002a) demonstrated that a late response in the component induced following an interleaving bright flash was significantly decreased. These results support the idea that RGC activity can be isolated from the mfERG response and that the RGC response is distinct and of longer implicit time than the one reported from the bipolar cell activity (Hare et al. 2001; Hood et al. 2002). The optic nerve head component delineated by post-acquisition manipulation of the mfERG signal also showed RGC activity in the late part of the response (Sutter et al. 1999). However, one mfERG study in patients with glaucomatous retinopathy has isolated a possible RGC component in the mfERG signal in patients with glaucoma in the form of an early wavelet associated with the primary response (first 30 ms; Sano et al. 2002). Loss of the oscillatory mfERG response pattern in the macaque reported to be associated with RGCs was covering the first 50 ms of the mfERG response (Hood et al. 1999a, b, 2002; Frishman et al. 2000; Fortune et al. 2002b; Ratz et al. 2002). Clearly, species differences exist and different modalities of stimulation may cause the variation in timing of the RGCs response.
, 百拇医药
Naso-temporal asymmetry in the isolated RGC responses was noted by several authors in monkey and human mfERG recordings. From a histological perspective, the naso-temporal asymmetry in the foveated mammals can be associated with the distance from the RGC to the optic nerve head (Hood et al. 2002) and from the fact that nasal and temporal RGCs are responsible for the contralateral and ipsilateral projections to the brain, respectively. In the pig, the naso-temporal asymmetry has different bases. For instance, the point of reference used in the pig is the optic nerve head (Garcia et al. 2005) probably because of the presence of two areas centralis, while the unique fovea in monkeys and humans is used for these species. In the periphery, larger cells are generally found in the temporal retina. Those distinctions taken into account, our data showed naso-temporal asymmetry in the visual streak, with RGC peak time response being shorter in the nasal retina. The idea that this asymmetry is due to different functions for the two areas centralis present in the pig retina is interesting but remains to be investigated. Inconsistent asymmetry could be found in the more peripheral areas despite the histological evidence for a differential soma size distribution (Garcia et al. 2005).
, 百拇医药
The absence of evidence for an optic nerve head response that varies in implicit time with distance from the optic nerve might possibly be explained by a gradient of RGC cell soma size (and consequently of axon size) present in the pig retina that compensates for the distance of the cell relative to the optic nerve head. In the monkey and human ERG recordings, the area investigated (generally a few degrees around the fovea) may present with a more homogenous cell size distribution, resulting in a marked optic nerve head component. This is admittedly purely speculative but further histological studies on RGC soma and axon size distribution in the pig retina may help resolve this issue.
, 百拇医药
Effects of tetrodotoxin
Intraocular injection of TTX has been used in previous studies (Hood et al. 1999a; Frishman et al. 2000; Hare et al. 2001; Hare & Ton, 2002) to isolate the activity of spike-generating cells such as RGCs and a subgroup of amacrine cells. In our study, intraocular injection of TTX in the pig maintained under ketamine sedation resulted in the disappearance of P2 and P3 and their replacement by a series of regular oscillations damping over 200 ms post-stimuli (Fig. 7). The loss of P2 and P3 is in keeping with results described in the cynomolgous monkey after TTX injections and experimental high intraocular pressure (Hare & Ton, 2002). In the macaque, results are more difficult to compare because of the high frequency content of the normal mfERG response; TTX reduced the high frequencies and rendered the response pattern independent of the retinal position (Hood et al. 1999a; Frishman et al. 2000). The replacement of P2 and P3 by a series of oscillations following TTX injection as repetitively observed in our study was not observed in either species of monkey. In the pig, these oscillations were equally present in the control and experimental eyes and thus had to be generated proximal to the RGCs. Further to the involvement of amacrines and RGCs interactions, the involvement of the bipolar cells in the initiation of these oscillations might also be considered as it has recently been demonstrated that some bipolar cell types have voltage-gated sodium channels (Pan & Hu, 2000). These oscillations were not a stimulus artifact as they disappeared after induction of isoflurane anaesthesia and were recovered when isoflurane was discontinued. The exact nature of the oscillatory pattern generated in the mfERG responses after intraocular injection of TTX in the pig remains equivocal but is likely to involve synaptic circuits in the inner plexiform layer.
, http://www.100md.com
Influence of anaesthesia
Our results are in agreement with those of Fortune et al. (2002b) who proposed a detrimental influence of anaesthesia on the mfERG signal in the monkey. They showed a lengthening of implicit time and a decrease in the high-frequency components thought to be generated somewhere within the inner plexiform layer. In the present study, we further defined the action of isoflurane by demonstrating that it directly alters the RGC activity. Recordings performed in the presence of isoflurane anaesthesia resulted in the reduction of all the components shown to be affected by ONS (i.e. RGCs) when recordings were performed under ketamine sedation. Recordings from control and experimental eyes after long-term ONS showed no significant difference under isoflurane anaesthesia. Isoflurane anaesthesia thus appears to prevent the contribution of RGC-related activity in the mfERG signal. Isoflurane anaesthesia also affected the implicit time of P1, but not that of P2 and P3, further suggesting that P1 and P2/P3 may have different retinal origins. Furthermore, isoflurane also abolished the oscillatory activity elicited after intravitreal injection of TTX, suggesting that it also has a site of action upstream to the RGCs and to the TTX-sensitive generators. Our results thus show that, beside alteration in outer retina activity, the RGC activity is much reduced. It is well documented that the central cortical activity is reduced under isoflurane anaesthesia (Sebel et al. 1986); this phenomenon may thus be contributed to by a reduced activity of the more peripheral inputs. The use of isoflurane anaesthesia may potentially represent a fast and reversible method of RGC inactivation for electrophysiological studies.
, 百拇医药
Anaesthetics are pluripotents and generally exert their action through multiple mechanisms. Isoflurane is considered to be responsible for a calcium-induced activation of the GABAA receptor, thereby inducing the prolongation of the GABA-mediated inhibitory postsynaptic currents. (Jones & Harrison, 1993) Isoflurane also depresses AMPA glutamate responses but has only weak interactions with NMDA binding sites (Lin et al. 1993; Maciver, 1997). At clinical concentrations, isoflurane also suppresses voltage-gated sodium channels, which results in a decrease in action potential transmission. Isoflurane is thus likely to influence many synaptic circuits presynaptic to RGCs and to prevent RGC action potentials. On the other hand, ketamine, a dissociative anaesthesic, may have more restricted sites of action with demonstrated effects through non-competitive inhibition of NMDA receptor, with no action on AMPA/kainate receptors (Lin et al. 1993; Peoples & Weight, 1997). Ketamine is also known to enhance cortical responsiveness (Schubert et al. 1990). As the pig mfERG under ketamine sedation results in a large P3 component that cannot be found in human, there is a possibility that P3 is a consequence of the ketamine action. We investigated that possibility by using a GABA antagonist anaesthetic, propofol, and found recordings identical to those recorded under ketamine, thus making it unlikely that P3 is induced by the type of anaesthesic.
, 百拇医药
In summary, our results demonstrate that a unilateral ONS results in a time-dependent amplitude loss of the mfERG late components P2 and P3. This model offers great opportunities to complement neuroanatomical studies on neuroprotection or neurotransplantation with local functional data.
References
Bach M & Speidel-Fiaux A (1989). Pattern electroretinogram in glaucoma and ocular hypertension. Doc Ophthalmol 73, 173–181.
, http://www.100md.com
Bearse MA Jr, Sutter E, Sim D & Stamper R (1996). Glaucomatous dysfunction revealed in higher order components of the electroretinogram. Vision Science and its Applications, OSA Tech Dig Ser, pp. 104–107. Optical Society of America, Washington, DC.
Berardi N, Domenici L, Gravina A & Maffei L (1990). Pattern ERG in rats following section of the optic nerve. Exp Brain Res 79, 539–546.
Berkemeier M, Clarke BD, Wang W-C, Bray GM & Aguayo AJ (1994). Axotomy results in delayed death and apoptosis of retinal ganglion cells in adult rats. J Neurosci 14, 4368–4374.
, http://www.100md.com
Bui BV & Fortune B (2003). Ganglion cell contributions to the rat full-field electroretinogram. J Physiol 555, 153–173.
Chan HL & Brown B (1999). Multifocal ERG changes in glaucoma. Ophthalmic Physiol Opt 19, 306–316.
Chandler MJ (1999). Photoreceptor density of the domestic pig retina. Vet Ophthalmol 2, 179–184.
Clements JA & Nimmo WS (1981). Pharmacokinetics and analgesic effect of ketamine in man. Br J Anaesth 53, 27–30.
, http://www.100md.com
Colotto A, Falsini B, Salgarello T, Iarossi G, Galan ME & Scullica L (2000). Photopic negative response of the human ERG: losses associated with glaucomatous damage. Invest Ophthalmol Vis Sci 41, 2205–2211.
Cursiefen C, Korth M & Horn F (2000). The negative response of the flash electroretinogram in glaucoma. Doc Ophthalmol 103, 1–12.
Drasdo N, Aldebasi Y & Chiti Z (2001). The s-cone PhNR and pattern ERG in primary open angle glaucoma. Invest Ophthal Vis Sci 42, 1266–1272.
, 百拇医药
Fortune B, Bearse MA Jr, Cioffi GA & Johnson CA (2002a). Selective loss of an oscillatory component from temporal retinal multifocal ERG responses in glaucoma. Invest Ophthal Vis Sci 43, 2638–2647.
Fortune B, Cull G, Wang L, van Buskirk EM & Cioffi GA (2002b). Factors affecting the use of multifocal electroretinography to monitor function in a primate model of glaucoma. Doc Ophthalmol 105, 151–178.
Fortune B, Johnson CA & Cioffi GA (2001). The topographic relationship between multifocal electroretinographic and behavioral perimetric measures of function in glaucoma. Optometry Vision Sci 78, 206–214.
, 百拇医药
Frishman LJ, Saszik S, Harwerth RS, Viswanathan S, Li Y, Smith EL III, Robson JG & Barnes G (2000). Effects of experimental glaucoma in macaques on the multifocal ERG. Multifocal ERG in laser-induced glaucoma. Doc Ophthalmol 100, 231–251.
Garcia M, Ruiz-Ederra J, Hernandez-Barbachano H & Vecino E (2005). Topography of pig retinal ganglion cells. J Comparative Neurol 486, 361–372.
Garcia-Valenzuela E, Gorczyca W, Darzynkiewick Z & Sharma SC (1994). Apoptosis in adult retinal ganglion cells after axotomy. J Neurobiol 25, 431–438.
, 百拇医药
Germain F, Calvo M & de la Villa P (2004). Rabbit retinal ganglion cell survival after optic nerve section and its effect on the inner plexiform layer. Exp Eye Res 78, 95–102.
Ghosh F & Arner K (2002). Transplantation of full-thickness retina in the normal porcine eye: surgical and morphologic aspects. Retina 22, 478–486.
Ghouri AF, Bodner M & White PF (1991). Recovery profile after desflurane-nitrous oxide versus isoflurane-nitrous oxide in outpatients. Anesthesiology 7, 419–424.
, 百拇医药
Hare WA & Ton H (2002). Effects of APB, PDA, and TTX on ERG responses recorded using both multifocal and conventional methods in monkey. Doc Ophthalmol 105, 189–222.
Hare WA, Ton HGR, Feldmann B, Wijino M & Woldemussie E (2001). Characterization of retinal injury using ERG measures obtained with both conventional and multifocal methods in chronic ocular hypertensive primates. Invest Ophthalmol Vis Sci 42, 127–136.
Harwerth RS, Crawford MLJ, Frishman L, Viswanathan S, Smith EL & Carter-Dawson L (2002). Visual field defects and neural losses from experimental glaucoma. Prog Retinal Eye Res 21, 91–125.
, 百拇医药
Hasegawa S, Tagaki M, Usui T, Takada R & Abe H (2000). Waveform change of the first-order multifocal electroretinogram in patients with glaucoma. Invest Ophthal Vis Sci 41, 1597–1603.
Hendrickson A & Hicks D (2002). Distribution and density of medium- and short-wavelength selective cones in the domestic pig retina. Exp Eye Res 74, 435–444.
Holder G (2001). Pattern electroretinography (PERG) and an integrated approach to visual pathway diagnosis. Prog Retinal Eye Res 20, 531–561.
, http://www.100md.com
Hood DC, Bearse MA Jr, Sutter EE, Viswanathan S & Frishman LJ (2001). The optic nerve head component of the monkey's (Macaca mulatta) multifocal electroretinogram (mERG). Vision Res 41, 2029–2041.
Hood DC, Frishman LJ, Saszik S & Viswanathan S (2002). Retinal origins of the primate multifocal ERG: implications for the human response. Invest Ophthalmol Vis Sci 43, 1673–1685.
Hood DC, Frishman LJ, Viswanathan S, Robson JG & Ahmed J (1999a). Evidence for a ganglion cell contribution to the primate electroretinogram (ERG): effects of TTX on the multifocal ERG in macaque. Vis Neurosci 16, 411–416.
, 百拇医药
Hood DC, Greenstein V, Frishman L, Holopigian K, Viswanathan S, Seiple W, Ahmed J & Robson JG (1999b). Identifying inner retinal contributions to the human multifocal ERG. Vision Res 39, 2285–2291.
Hood DC, Greenstein VC, Holopigian K, Bauer R, Firoz B, Liebmann JM, Odel JG & Ritch R (2000). An attempt to detect glaucomatous damage to the inner retina with the multifocal ERG. Invest Ophthalmol Vis Sci 41, 1570–1579.
Jones MV & Harrison NL (1993). Effects of volatiles anesthetics on the kynetics of inhibitory postsynaptic currents in cultured rat hippocampal neurons. J Neurophysiol 70, 1339–1349.
, 百拇医药
Kielczewski JL, Pease ME & Quigley HA (2005). The effect of experimental glaucoma and optic nerve transection on amacrine cells in the rat retina. Invest Ophthalmol Vis Sci 46, 3188–3196.
Klistorner AI, Graham SL & Martins A (2000). Multifocal pattern electroretinogram does not demonstrate localised field defects in glaucoma. Doc Ophthalmol 100, 155–165.
Komaromy AM, Brooks DE, Kallberg ME, Dawson WW, Szel A, Lukats A, Samuelson DA, Sapp HL Jr, Gelatt KN & Sherwood MB (2003). Long-term effect of retinal ganglion cell axotomy on the histomorphometry of other cells in the porcine retina. J Glaucoma 12, 307–315.
, 百拇医药
Korth M (1997). The value of electrophysiologic testing in glaucomatous diseases. J Glaucoma 6, 331–343.
Levin V (2001). Models of neural injury. J Glaucoma 10, s19–s21.
Li ZY, Wong F, Chang JH, Possin DE, Hao Y, Petters RM & Milam AH (1998). Rhodopsin transgenic pigs as a model for human retinitis pigmentosa. Invest Ophthal Vis Sci 39, 808–819.
Lin LH, Whiting P & Harris RA (1993). Molecular determinants of general anesthetic action: role of GABAa receptor structure. J Neurochem 60, 1548–1553.
, http://www.100md.com
Maciver MB (1997). General anesthetic actions on transmission at glutamate and GABA synapses. In Anesthesia: Biological Foundations, ed. Yaksh TI, Lynch C, Zapol WM, Maze M, Biebuyck JF & Saidman LJ, pp. 277–286. Lippincott-Raven, Pliladelphia.
Maffei L & Fiorentini A (1981). Electroretinographic responses to alternating gratings before and after section of the optic nerve. Science 211, 953–955.
Maffei L, Fiorentini A, Bisti S & Hollander H (1985). Pattern ERG in the monkey after section of the optic nerve. Exp Brain Res 59, 423–425.
, 百拇医药
Otto T & Bach M (1996). Retest variability and diurnal effects in the pattern electroretinogram. Doc Ophthalmol 92, 311–323.
Palmowski AM, Allgayer R, Heinemann-Vernaleken B & Ruptecht KW (2002). Multifocal electroretinogram with a multiflash stimulation technique in open-angle glaucoma. Ophthalmic Res 43, 83–89.
Pan ZH & Hu HJ (2000). Voltage-dependent Na+ currents in mammalian retinal cone bipolar cells. J Neurophysiol 84, 2564–2571.
, 百拇医药
Peoples RW & Weight FF (1997). Anesthetic actions on excitatory amino acid receptors. In Anesthesia: Biological Foundations, ed. Yaksh TI, Lynch C, Zapol WM, Maze M, Biebuyck JF & Saidman LJ, pp. 239–258. Lippincott-Raven, Philadelphia.
Prince JH, Diesem CD, Eglitis I & Ruskell GL (1960). The pig. In Anatomy and Histology of the Eye and Orbit in Domestic Animals, ed. Thomas CC, pp. 210–230. Charles C. Thomas, Springfield, Illinois.
, http://www.100md.com
Ratz D, Seeliger MW, Geva AB, Perlicot CLM, Lambrou GN & Ofri R (2002). The effect of contrast and luminance on mferg responses in a monkey model of glaucoma. Invest Ophthal Vis Sci 43, 2027–2035.
Rosolen S, Rigaudiere F & Le Gargasson J (2003). A new model of induced ocular hyperpressure using the minipig (in French). J Fr Ophtalmol 26, 259–267.
Ruiz-Ederra J, Garcia M, Hernandez M, Urcola H, Hernandez-Barbachano E, Araiz J & Vecino E (2005). The pig eye as a novel model of glaucoma. Exp Eye Res 81, 561–569.
, 百拇医药
Sachs HG, Schanze T, Brunner U, Sailer H & Wiesenack C (2005). Transscleral implantation and neurophysiological testing of subretinal polyimide film electrodes in the domestic pig in visual prosthesis development. J Neurol Eng 2, S57–S64.
Sano M, Tazawa Y, Nobeshima T & Mita M (2002). A new wavelet in the multifocal electroretinogram, probably originating from ganglion cells. Invest Ophthalmol Vis Sci 43, 1666–1672.
Schubert A, Licina MG & Lineberry PJ (1990). The effect of ketamine on human somatosensory evoked potentials and its modification by nitrous oxide. Anesthesiology 72, 33–39.
, 百拇医药
Sebel PS, Ingram DA, Flynn PJ, Rutherfoord CF & Rogers H (1986). Evoked potentials during isoflurane anaesthesia. Br J Anaesth 58, 580–585.
Sieving PA, Frishman LJ & Steinberg RH (1986). Scotopic threshold response of proximal retina in cat. J Neurophysiol 56, 1049–1061.
Sutter EE & Bearse MA Jr (1999). The optic nerve head component of the human ERG. Vision Res 39, 419–436.
Sutter E, Shimada Y, Li Y & Bearse MA (1999). Mapping inner retinal function through enhancement of adaptative components in the MF-ERG. Vis Sci App OSA Tech Dig Series 1, 52–55.
, http://www.100md.com
Sutter EE & Tran D (1992). The field topography of ERG components in man – 1. The photopic luminance response. Vision Res 32, 433–446.
Tso MO, Li WW, Zhang C, Lam TT, Hao Y, Petters RM & Wong F (1997). A pathologic study of degeneration of the rod and cone populations of the rhodopsin Pro347Leu transgenic pigs. Trans Am Ophthalmol Soc 95, 467–479.
Vaegan, Anderton PJ & Millar TJ (2000). Multifocal, pattern and full field electroretinograms in cats with unilateral optic nerve section. Doc Ophthalmol 100, 207–229.
, 百拇医药
Vecino E, Ruis-Ederra J, Hernendez M, Urcola JA, Araiz J, Duren J & Garcia M (2004). Topography of retinal ganglion cells of the pig retina – cell size distribution. ARVO on-line abstract 5434.
Viswanathan S, Frishman LJ, Robson JG, Harwerth RS & Smith EL III (1999). The photopic negative response of the macaque electroretinogram: reduction by experimental glaucoma. Invest Ophthalmol Vis Sci 40, 1124–1136.
Viswanathan S, Frishman LJ, Robson JG & Walters JW (2001). The photopic negative response of the flash electroretinogram in primary open angle glaucoma. Invest Ophthalmol Vis Sci 42, 514–522.
, 百拇医药
Watanabe M, Inukai N & Fukuda Y (2001). Survival of retinal ganglion cells after transection of the optic nerve in adult cats: a quantitative study within two weeks. Vis Neurosci 18, 137–145.
Weishaupt J & Bahr M (2001). Degeneration of axotomized retinal ganglion cells as a model for neuronal apoptosis in the central nervous system – molecular death and survival pathways. Restor Neurol Neurosci 19, 19–27., 百拇医药(Mélanie R. Lalonde, Balwantray C. Chauha)
2 Department Physiology & Biophysics
3 Department Ophthalmology and Visual Sciences, Dalhousie University, Halifax, Nova Scotia, Canada B3H 4H7
Abstract
Non-invasive recordings of the retinal activity have an important role to play in the diagnosis of retinal pathologies. The detection of diseases that involve retinal ganglion cells (RGCs), such as optic atrophy and glaucoma, may be improved by isolating the RGC contribution from the multifocal electroretinogram (mfERG). In this study, mfERGs were performed on 20 pigs, 1–6 weeks following unilateral retrobulbar optic nerve section (ONS). The stimuli were 103 non-scaled high-contrast hexagons from which summed and individual mfERG responses were obtained in experimental and control fellow eyes under conditions of ketamine (n= 11) or isoflurane anaesthesia (n= 9). The effect of intravitreal injection of tetrodotoxin (TTX; n= 6) was also investigated. The summed mfERG responses showed a first positive peak (P1) with a short latency (21 ms) followed by two smaller peaks (P2 and P3) of longer latency (46 and 65 ms, respectively). While P2 and P3 amplitude were highly correlated with the time post-optic nerve section (ONS) (P2: r2= 0.669; P= 0.007; P3: r2= 0.651; P= 0.005), P1 was not (r2= 0.193; P= 0.38). P1 and P2 showed no implicit time variation as a function of retinal location, while P3 implicit time varied along the axis of the visual streak, generating a naso-temporal asymmetry. However, the P3 implicit time did not vary consistently with distance away from the optic nerve head. Intravitreal injections of TTX reduced P2 and P3 in the control eyes, consistent with the effect of ONS, and also induced a series of regular oscillations lasting up to 200 ms post stimulus. Under isoflurane anaesthesia, all components of the mfERG ifn experimental and control eyes were, at all time points post-ONS, of similar amplitude and without naso-temporal asymmetry, suggesting a reduced participation of RGCs under these anaesthesic conditions. These data clearly demonstrate that it is possible to isolate the RGC contribution from non-invasive multifocal electroretinography.
, 百拇医药
Introduction
Traditionally, non-invasive clinical electrophysiological techniques for retinal assessment were considered insensitive to the relatively small number of retinal ganglion cells (RGCs) present in the retina. Recently, the photopic negative response and the scotopic threshold response were isolated from the Ganzfeld electroretinogram (ERG). These responses were shown to be affected in experimental models of optic neuropathy (Sieving et al. 1986; Viswanathan et al. 1999; Bui & Fortune, 2003), in patients with glaucoma (Colotto et al. 2000; Cursiefen et al. 2000; Drasdo et al. 2001; Viswanathan et al. 2001) and by intravitreal injection of the voltage-gated sodium channel blocking agent tetrodotoxin (TTX). This finding suggests these responses might be at least partially generated by RGCs. Pattern stimuli have been used conventionally to elicit proportionally larger RGC responses. The potential generated by this technique called pattern electroretinography (PERG) was shown to be sensitive to optic nerve section (ONS) in rat (Berardi et al. 1990), cat (Mafei & Fiorentini, 1981) and monkey (Maffei et al. 1985). However, the use of PERG in clinical settings is still debatable, mostly because of the high inter- and intrasubject variability (Bach & Speidel-Fiaux, 1989; Otto & Bach, 1996; Korth, 1997; Holder, 2001). Both ERG and PERG techniques are corneal potentials elicited by stimulation of the whole visual field and therefore, are relatively insensitive to local defects.
, 百拇医药
Multifocal-electroretinography (mfERG) is a newly introduced method for retinal electrophysiology (Sutter & Tran, 1992) that allows the isolation of multiple local responses and the topographical evaluation of the retina. The rapid rate of stimulation and the algorithm used to reconstruct the individual signals result in a recombination of retinal dipoles different from those of the ERG and PERG (Hood et al. 2002). Two approaches have been used to isolate the RGC contribution to the mfERG. First, a computational approach was found to isolate the optic nerve head component, a signal whose implicit time varied with the distance to the optic disc and thought to be generated by the RGC axons at the optic nerve head (Sutter & Bearse, 1999; Hood et al. 2001). This component proved to be eliminated by intravitreal injection of TTX in the monkey (Hood et al. 1999a; Frishman et al. 2000; Fortune et al. 2002b) and was found to be reduced in patients with glaucomatous neuropathy (Bearse et al. 1996). However, its sensitivity to detect early glaucomatous retinopathy is controversial (Fortune et al. 2002a; Palmowski et al. 2002). The second approach is based on the observation of the mfERG signal itself. Loss of naso-temporal asymmetry in monkeys was found to be associated with advanced glaucomatous changes (Hare et al. 2001; Fortune et al. 2002a; Harwerth et al. 2002). In humans with glaucomatous optic neuropathy, responses were found to be variable and no consensus has yet emerged (Chan & Brown, 1999; Hasegawa et al. 2000; Hood et al. 2000; Klistorner & Graham, 2000; Fortune et al. 2001; Sano et al. 2002).
, 百拇医药
The pig as an animal for vision research has been extensively used for corneal and anterior segment research and interest is growing for posterior segment research including recently developed models for glaucoma (Rosolen et al. 2003; Ruiz-Ederra et al. 2005), photoreceptor degeneration (Tso et al. 1997; Li et al. 1998), retinal transplant (Ghosh & Arner, 2002), and retinal prosthesis (Sachs et al. 2005). The morphometry of the pig eye is similar to human eye (Prince et al. 1960), allowing the development of surgical and transplant procedures that can be translated to the medical field. The pig retina is rich in cones (14–17 millions; Chandler, 1999; Hendrickson & Hicks, 2002), with a cone: rod ratio close to the one found in the central human retina and much higher than the one found in the uniform retina of rodents, allowing the use of local visual stimuli to isolate the RGC signal. In this study, ONS (Berkemeier et al. 1994; Garcia-Valenzuela et al. 1994; Vaegan & Millar, 2000) was used to produce progressive retrograde RGC degeneration in the pig retina. Individual components of the summed mfERG responses were correlated to the time post-ONS. In addition, the differential effect of intravitreal injection of TTX was examined and compared to the effect of RGC degeneration. Finally, the impact of isoflurane, a halogenated anaesthetic widely used in large animals, on the mfERG signal was investigated.
, 百拇医药
Methods
Animals
Electroretinographic data was obtained from a total of 20 Hampshire Durco Cross pigs aged 3–5 weeks. The effect of unilateral ONS, used here to produce a controlled progressive RGC degeneration, was studied using the traditionally recommended isoflurane anaesthesia (n= 9 experimental eyes; n= 9 control fellow eyes) and with ketamine sedation (n= 9 experimental eyes; n= 11 control eyes). Of the 10 animals tested under ketamine sedation, four were also studied by reintroducing isoflurane anaesthesia, and six received intravitreal injections of TTX (n= 2 experimental eyes; n= 6 control eyes). All procedures were conducted in accordance with the Association for Research in Vision Statement for the Use of Animals in Ophthalmic and Vision Research and with the guidelines established by the Canadian Council for Animal Care. Ethical approval was obtained from the Dalhousie University Committee for Laboratory Animals.
, 百拇医药
Surgical procedures
Anaesthesia. Animals were fasted for a period of 12 h prior to anaesthesia but had free access to water. Pre-medication included atropine (0.1 mg kg–1I.M.; Abbot Laboratories, Montreal, QC, Canada) and ketamine (Ketalar, 18 mg kg–1I.M.; Birmeda-MTC, Cambridge, ON, Canada). Deep anaesthesia was induced with 5% isoflurane (AErrane; Abbot Laboratories) and administered with 100% oxygen delivered through a facemask. Endotracheal intubation was performed and artificial ventilation was supplied with 100% oxygen (volume: 10 ml kg–1; rate: 13 breaths min–1; Ohio Medical Instrument Co. Inc., Cincinnati, OH, USA). The isoflurane level was reduced to a maintenance level of 2.0–2.5%. The pigs were kept hydrated via lactated Ringer solution delivered through a cannulated dorsal auricular vein. Body temperature was maintained at 38.0–39.0°C using a circulating hot water heating pad (Aquamatic K Module; American Medical Systems, Valencia, CA, USA). Both heart rate and oxygen saturation levels were monitored (SDI Vet/OxTM 4403; SDI Sensor Devices Inc., Wankesha, WI, USA) throughout the procedures.
, 百拇医药
Optic nerve section. ONS was performed under deep isoflurane anaesthesia and local xylocaine 2% skin infiltration. A small incision was made at the temporal canthus of the eye. Sutures were attached to the superior conjunctiva to help rotate the eyeball downward. The cornea was protected from desiccation with a thick coat of ophthalmic ointment (Optimyxin, Sabex, Boucherville, QC, Canada). The conjunctiva was then incised and the lateral rectus resected. Using the lateral episcleral vein as a guide, adipose tissue was dissected until the optic nerve was exposed. The dura was incised longitudinally for 3 mm and a pair of blunt Vannass scissors was used to cut the nerve within the dura in order to protect the adjacent ophthalmic artery. A piece of spongel soaked in fluorogold 2% (Fluorochrome Inc., Denver, CO, USA) was then introduced in the cavity to retrogradely label RGCs for histology. The lateral rectus was sutured back in place as well as the conjunctiva. The area was infiltrated with gentamycin (100 mg ml–1; Sigma Aldrich Canada Ltd, Oakville, ON, Canada) and the skin sutured. Fundus examination using direct opththalmoscopy was used to ensure the integrity of the retinal circulation. Post-operative care included intramuscular administration of buprenorphine hydrochloride 0.01 mg kg–1 (Reckitt & Colman Products, Hull, UK) 15 min before the cessation of anaesthesia and salicylic acid (10 mg kg–1; Bayer Inc., Etobicoke, ON, Canada) given orally for 2 days post-surgery.
, http://www.100md.com
Recording sessions
Protocol. Recording sessions were performed prior to surgery (n= 2) and once during the 6 weeks post-ONS period (1 week: n= 2; 2 weeks: n= 2; 3 weeks: n= 2; 6 weeks: n= 3). Animals were anaesthetized prior to all ERG recordings using the previously described isoflurane protocol and secured in a custom-made sling equipped with a head holder. For a group of 11 animals, the anaesthesia regimen was switched to ketamine sedation (20 mg kg–1 h–1I.V.) in combination with the muscle relaxant gallamine (35 mg kg–1 h–1, I.V.) and artificial ventilation. For this group, recordings were performed after a minimum of 20 min from the time isoflurane was discontinued to allow its clearance from the body fluids (Ghouri et al. 1991). TTX (0.0839 μg μl–1; Sigma Aldrich Canada Ltd) was injected into the vitreous of six eyes (2 experimental and 4 control eyes) through the pars plana using a 30-gauge needle, at a final concentration of 5 μM based on an estimated total vitreal volume of 2.0 ml. An ophthalmoscopic examination was performed to ensure the needle did not penetrate the retina and cause oedema or detachment. ERG recordings were obtained 90–120 min following the injection. In another group, experimental and fellow eye recordings were performed under isoflurane anaesthesia (n= 7) or obtained prior to the switch to ketamine (n= 2).
, http://www.100md.com
Recording procedures. The active electrode consisted of a monopolar contact lens (Medical Workshop b.v., Groningen, Holland) filled with a protective coat of hydroxypropyl methylcellulose 1% (Alcon Canada Inc., Mississauga, ON, Canada). Retinoscopy revealed a refractive error generally of less than –2.00 diopters. mfERG recordings with and without optical correction yielded similar results and consequently, no optical correction was used. The fellow eye was occluded throughout the recording sessions. Subdermal platinum needle electrodes (Grass Instruments, Quincy, MA, USA), inserted on the upper eyelid of the tested eye and on the nose, were used as reference and ground electrodes, respectively. The continuously recorded retinal signal was amplified (5.0 x 105), bandpass filtered between 3 and 300 Hz (P511; Grass Instruments Inc.) and digitized in synchrony with the display at a sampling rate of 16 samples per display frame. The signal responses obtained were extracted and processed using VERIS software version 4.7 from EDI (Electro-Diagnostic-Imaging, San Mateo, CA, USA).
, 百拇医药
Multifocal-electroretinography. The mfERG stimuli with a refresh rate of 13.3 ms per frame were displayed on a 21 in computer monitor (Viewsonic P815, Walnut, CA, USA) positioned 25 cm away from the tested eye. The stimulus consisted of 103 non-scaled white (130 cd m–2) and black (1.5 cd m–2) hexagons presented against a 25 cd m–2 grey background. The VERIS software was used to present the hexagons following a pseudo-random binary sequence (the m-sequence) of white and black states. M-sequences of either 215–1 or 214–1 length (corresponding to approximately 7 and 3.5 min recordings) were used, as responses obtained in the two conditions were very similar. Responses from either the whole stimulus area or from small groups of five to six selected hexagons were examined. The VERIS software also generated a 3-D representation of the individual responses from all hexagons. Since non-scaled hexagons were used, these response density maps were computed from the root mean square (RMS) of the trace values, without template. Examination of these maps allowed the identification of retinal structures such as the optic nerve head and the visual streak (a specialized area of the porcine retina with higher cell density). The final position of the monitor was determined after an initial estimate of the optic disc position from indirect ophthalmoscopy, the location of the blind spot on RMS maps obtained subsequently after a short recording being the determinant parameter used. In all recordings retained for this study, both the optic nerve head and the visual streak were identified. The visual streak was sometimes slightly inclined relative to the horizontal axis of the stimulus array due to constraints in positioning the pig's head relative to the monitor. This was taken into consideration when corresponding locations from the two eyes had to be compared.
, 百拇医药
RGC damage post-ONS
Animals were killed using sodium pentobarbital overdose (Euthasol, 0.3 mg kg–1) and both eyes were enucleated. The retinae were dissected after 2.5–3.5 h fixation in 4% paraformaldehyde and positioned on a slide with the inner retina facing up, then mounted in an aqueous medium (Citifluor, glycerol–PBS solution; Marivac, Halifax, NS, Canada), cover-slipped and sealed.
Retinal whole mounts were examined under fluorescence microscopy (Nikon E800). Three digital images, each covering a retinal area of 640 x 480 μm2, were acquired (Image Pro Plus; Carsen Group Inc., Markham, ON, Canada) from the visual streak area, 1.5 mm above and 3.0 mm temporal to the optic nerve head, as well as in more peripheral areas at 5.0 mm and 10.0 mm above this point in the superior retina.
, 百拇医药
Retinae were obtained from control eyes 3 days post-ONS to allow retrograde labelling. The RGC peak density was 2940 ± 697 cells mm–2 (n= 5), tapering rapidly to 489 ± 140 cells mm–2 in the peripheral retina. These figures are compatible with previously published RGC counts in pigs (Vecino et al. 2004). Retinae were also collected from experimental eyes. The variability in the staining of the experimental eyes prevented a detailed analysis of the RGC loss in function of time as originally intended. However, the presence of pyknotic bodies and glial cells filled with fluorogold debris from dying RGCs were taken as evidence for RGC loss in retina obtained 1–3 weeks post-ONS. Retinae collected after 6 weeks post-ONS showed a reduction of RGC, with densities averaging 658 ± 309 cells mm–2 for the peak density and 166 ± 88 cells mm–2 for the periphery (n= 10, P < 0.05, t test).
, http://www.100md.com
Data analysis
Quantitative measurements were performed on amplitude and implicit time of the summed mfERG responses computed from the 103 individual, first kernel responses. One major positive component, P1, occurring in a post-stimulus time window of 10–30 ms, was quantified by measuring the amplitude between the first negative trough and the first positive peak (Fig. 1). This high-amplitude response was followed by components of smaller amplitude spanning 30–80 ms. P2 amplitude was defined as the component corresponding to the second trough to the peak of the intermediate positive component which normally occurred around 45 ms. P3 was defined as the last positive component measured from the small negative through following P2 to the most positive voltage of the 60–80 ms segment of the response.
, http://www.100md.com
Representative example of control mfERG responses obtained to high contrast, 13.3 Hz, 103 non-scaled hexagons stimuli configuration. A, summed first-order response: the amplitude of the three positive components is indicated by the vertical arrows while the time taken by the individual components to develop is identified by the three horizontal arrows: B, individual recordings: C, RMS topographical map: the large arrow corresponds to the optic nerve head and the two small arrows identify the axis of the visual streak that lies inferior to the optic nerve head in this visual field representation.
, http://www.100md.com
The summed mfERG data obtained from the control and experimental eyes, under ketamine sedation and at post-ONS time longer than 2 weeks, were compared using ANOVA with Bonferroni/Dunn post hoc correction. The effect of anaesthesia was analysed by comparing data from the control and the experimental eyes under isoflurane anaesthesia and by comparing data from the control eyes under isoflurane anaesthesia with either the control or the experimental eyes recorded under ketamine sedation.
, http://www.100md.com
The RGC component was isolated off-line by exporting to a spreadsheet the responses from manually selected groups of five hexagons located in corresponding areas of the control and experimental eyes. The RMS maps from each eye were used to match the locations selected, using the position of the blind spot and the orientation of the visual streak as the reference points. Corresponding responses from experimental locations were then digitally subtracted from those of control locations to provide a signal representative of the RGC activity.
, 百拇医药
Results
mfERG recordings in control eyes
Representative first kernel control mfERG responses to 103 non-scaled, high contrast hexagon stimuli are documented in Fig. 1. The summed mfERG response consisted of an initial major positive component, P1, followed by two more variable positive components identified here as P2 and P3. No consistent stimulus-associated activity could be detected beyond 80 ms. The raw recordings, as well as the RMS map, showed a well-delineated zone of minimal activity corresponding to the optic nerve head, and a band of larger responses extending horizontally, immediately below the optic nerve head in visual field view (Chandler, 1999).
, 百拇医药
Effect of optic nerve section
The effect of ONS on summed, high-contrast mfERG responses is summarized in Table 1. Data from all experimental eyes recorded more than 14 days post-ONS were pooled (n= 7, mean time post-ONS 28.9 days). In the experimental eye, under ketamine sedation, P1 was mildly reduced in amplitude while P2 and P3 showed a highly significantly reduction in amplitude. Long-term ONS did not have any significant effect on implicit times. In Fig. 2, representative examples of summed responses from the control and the experimental eyes of four individual pigs recorded at 1, 2, 3 and 6 weeks post-ONS are presented. At 1 week post-ONS, responses were very similar. However, at 2 weeks post-ONS, P3 showed a consistent reduction in amplitude while at 6 weeks post-ONS, P3 was nearly absent and the negative trough preceding P2 was less pronounced, contributing to the decrease in P2 amplitude. The ratios of the experimental to the control responses for P1, P2 and P3 are presented in Fig. 2B (data recorded from two animals pre-ONS and two animals at 1 week post-ONS were included). For P1, a best-fit linear correlation (0.886 – 0.028x; r2= 0.193; P= 0.38) suggests a non-statistically significant change over time. On the other hand, decrease in amplitude for P2 (0.837 – 0.079x; r2= 0.669; P= 0.007) and P3 (0.781 – 0.087x; r2= 0.651; P= 0.005) was significant and, by 6 weeks post-ONS, the predicted amplitude ratios between experimental and fellow control responses were 0.36 and 0.26 for P2 and P3, respectively.
, 百拇医药
A, representative responses obtained in control (Ctl; left column) and experimental (Exp; right column) eyes, showing progressive loss of P3 and P2 components: B, scattergrams of the amplitude ratios between experimental and control eyes for P1, P2 and P3, at 1, 2, 3 and 6 weeks post-ONS.
RMS maps constructed from P1 (10–30 ms) of the signal found in a representative control eye, showed a horizontal band of higher activity corresponding to the visual streak (Fig. 3). A similar map was obtained for P2 (30–50 ms). For the map constructed around P3 (50–75 ms), the visual streak was no longer discernable, suggesting that P3 amplitude map did not correspond to known RGC cell distribution. In the experimental eye, the maps from the 10–30 ms and 30–50 ms segments, P1 and P2, respectively, were similar to the maps from the fellow control eye in showing a concentration of activity in the visual streak. The scalar product map from the 50–75 ms segment P3 showed a uniform and very low voltage activity corresponding to the low voltage P3.
, 百拇医药
RMS topographical maps showing that P1 (10–30 ms) and P2 (30–50 ms) segments of the first kernel of the mfERG responses in a control (Ctl; left column) and experimental (14 days post-ONS; Exp, right column) eye are scaled by the higher cell density of the visual streak while P3 (50–75 ms) is presenting with a more uniform amplitude distribution. Open arrow: blind spot. S: superior; I: inferior; N: nasal; T: temporal fields. Gray scale calibration as indicated.
, 百拇医药 Local mfERG responses to high contrast, 103 hexagon stimuli in the control (Ctl OS; left column) and experimental (Exp OD; middle column) eyes of a pig 14 days post-ONS. Local, summed (3 hexagons) mfERG responses spanning the temporal (trace 1) to the nasal (trace 3) aspects of the visual streak and of the mid-periphery (traces 4–6). The stimulus array has been located in corresponding positions for the two eyes, as demonstrated by the position of the blind spot (indicated by an x in the insert). A marked naso-temporal asymmetry was observed in the control eye for the visual streak, particularly evident with the morphology of the P2–P3 complex. In the mid-periphery, this asymmetry was much less evident. In the experimental eye, the asymmetry disappeared, leaving similar recordings in all positions. The rightmost column (Ctl - Exp) shows the computer subtraction between control and experimental eyes, delineating the activity assumed to be generated by the retinal ganglion cells. Dashed lines centred on the P3 wave.
, 百拇医药
Local mfERG responses to high contrast, 103 hexagon stimuli in the control and experimental eye of a pig 14 days post-ONS. Five summed (5 hexagons) mfERG responses to a control (Ctl OS; left column) and fellow experimental (Exp OD; middle column) eye as function of the distance from the optic nerve head position (x) with group 1 being close to the optic nerve and group 5 being about 50 deg away from the optic nerve head position. The rightmost column shows the computer subtraction between control and experimental eye, delineating the activity assumed to be generated by the retinal ganglion cells. No implicit time changes in function of distance (dashed lines corresponding to P3) could be observed.
, 百拇医药
Effect of TTX
Intravitreal injection of TTX resulted in a change in the wave morphology of the summed mfERG response, with reduction in the amplitude of P1, as well as the disappearance of P2 and P3 and their replacement by a series of regular oscillations (Fig. 6). These oscillations had a frequency between 45 and 50 Hz, different from the 60 Hz line interference and from the 75 Hz refresh rate of the monitor. They decreased after 200 ms post-stimulus, a time much longer than the activity elicited in the control conditions. These oscillations were found in the control as well as in the experimental eyes, in equal number, with the same frequency and the same temporal decrease. The introduction of isoflurane produced a reversible decrease in all the TTX-induced oscillations, supporting the idea these oscillations originated from retinal generators and were not artifacts induced by the stimulation protocol. Furthermore, the naso-temporal asymmetry previously described in control retinae was absent after TTX injection.
, http://www.100md.com
A, effect of TTX intraocular injection on summed mfERG responses from control (Ctl OD) and experimental (Exp OS) eyes (55 days post-ONS) of one pig. TTX induced a series of oscillations in both the control and experimental eyes (traces 2 and 6), oscillations that were cancelled by isoflurane anaesthesia (traces 3 and 7) and recovered when isoflurane was discontinued (traces 4 and 8). B, the injection of TTX under ketamine masked the asymmetry normally present along the visual streak in the control eye (grouped hexagons as shown in insert, traces 9–12).
, 百拇医药
Effects of anaesthesia
The consequences of ONS on the mfERG responses as described above were all observed on animals maintained under ketamine sedation. The use of isoflurane anaesthesia (Table 1 and Fig. 7) produced striking differences. In comparing the mfERG summed responses of control and experimental eyes, the amplitude differences observed on the animal maintained under ketamine sedation could no longer be observed. Isoflurane anaesthesia significantly decreased the amplitude of P3 compared with recordings obtained under ketamine sedation (5.4 ± 2.7 nV deg–2versus 10.8 ± 5.8 nV deg–2), to the extent it did not show significant difference compared with P3 recorded in the experimental eye under ketamine (P= 0.12). Under isoflurane anaesthesia, prolonged implicit time could be observed only for P1. Therefore, isoflurane suppressed the components associated with the RGC activity while producing minimal changes to other components.
, 百拇医药
A, from control (Ctl) and experimental (Exp) eye of one animal, 3 weeks post-ONS, in conditions of ketamine sedation (traces 1–6) and isoflurane anaesthesia (traces 7–12), illustrating the fact that recordings obtained under isoflurane were not showing asymmetry in the implicit time of P3 along the axis of the visual streak (upper traces in each subset corresponding to nasal location) in the control eye (traces 7–9). B, summed mfERG responses in a control eye before (trace 1); 10, 20 and 30 min after isoflurane anaesthesia was discontinued, to be replaced by an intravenous ketamine regimen (traces 2, 3 and 4); and when isoflurane anaesthesia was reintroduced while the ketamine sedation was still effective (trace 5).
, 百拇医药
In two control eyes, recordings were first obtained under isoflurane anaesthesia, after which isoflurane was discontinued and replaced by intravenous injection of ketamine (Fig. 7). Components proposed to be associated with RGC activity progressively appeared after 20–30 min, a time corresponding to the estimated clearance time of isoflurane from the body fluids (Ghouri et al. 1991). When isoflurane was reintroduced, a reduction in amplitude of those same components was observed, along with prolongation of implicit time, to bring the recordings back to the initial conditions. This occurred despite the fact that ketamine was still present in the body fluids (t1/2 of ketamine is around 3 h (Clements & Nimmo, 1981)), suggesting that isoflurane has an action on retinal synaptic circuits upstream to those affected by ketamine. The asymmetry observed along the visual streak under ketamine sedation was not present under isoflurane anaesthesia.
, 百拇医药
Discussion
The present study used a complete optic nerve section in the pig to reliably decrease the RGC population, isolating its contribution to the mfERG signal. RGC activity was associated with the mfERG P2 and P3 components that showed a time-related amplitude reduction after ONS. Variation in the implicit time of P3 is responsible for the asymmetry of the mfERG responses between the temporal and nasal hemiretinae; however, P3 implicit time did not vary systematically with the distance from the optic nerve head, making this component distinct from the optic nerve head component previously described (Sutter & Bearse, 1999). TTX injection resulted in an amplitude reduction of the components associated with RGC activity, in keeping with results obtained in monkeys (Hood et al. 1999a,b; Frishman et al. 2000). However, in the pig, TTX induced a series of regular oscillations in both control and experimental eye recordings that has not been reported in monkeys. The RGC activity was sensitive to the type of anaesthesia as recordings obtained under isoflurane anaesthesia showed no statistical difference in amplitude and implicit time between eyes with and without ONS. In the pig, a solid, well-delineated mfERG RGC component can thus be reliably isolated.
, http://www.100md.com
The pig mfERG
All the recordings included in this study were characterized by a well-delineated optic nerve head and a visual streak activity rendering in RMS maps. The ratio between the maximum amplitude responses (corresponding to the visual streak) and the peripheral responses (excluding the area corresponding to the optic nerve head) for P1 varied between 1.5: 1 and 4.6: 1, consistent with the photoreceptor ratio of 3.5: 1 (density of 25 000 cells mm–2 in the visual streak and 8000 cells mm–2 in periphery) reported in histological studies (Chandler, 1999; Hendrickson & Hicks, 2002).
, 百拇医药
It has been reported that the visual streak shows a higher RGC density (range of 2600 cells mm–2), with a decline to about 700 cells mm–2 in the mid-periphery, providing a ratio of about 3.7: 1 (Garcia et al. 2005). The amplitude ratio calculated for P3 in this study varied between 1.1: 1 and 1.7: 1, far from what would have been predicted from the cell density. Furthermore, the RMS map for P3 (Fig. 3) did not show any evidence of activity distribution suggestive of a higher RGC activity in the visual streak. This lack of correlation between mfERG activity and documented RGC distribution may be partly due to the naso-temporal asymmetry observed for the P3 component (Fig. 5); P3 implicit time varied between 58 ms in the nasal retina and 78 ms in the temporal retina. When P3 had a short implicit time it merged with the preceding P2 wave. Further detailed analysis on the isolated RGC component will be necessary to fully resolve this issue.
, 百拇医药
The RGC component of the mfERG isolated after optic nerve section
ONS is a well-documented method of reducing the number of RGCs in a time-related manner (Berkemeier et al. 1994; Garcia-Valenzuela et al. 1994; Levin, 2001; Watanabe et al. 2001; Weishaupt & Bahr, 2001) with little, if any, consequences in other retinal cell populations (Komaromy et al. 2003; Germain et al. 2004; Kielczewski et al. 2005). Functional consequences are not as well known. In our study, the mfERG recordings after ONS suggest that two components of the mfERG summed response are sensitive to the reduction of the RGC population (Table 1 and Fig. 2). A small positive component of long implicit time (P3; range of 70 ms post-stimulus) is more severely affected, being reduced to less than 20% seven weeks post-ONS. Changes in the amplitude of the negative trough between P1 and P2 were also observed to parallel the changes in P3. Changes in P1 were not significant but a trend toward decreasing amplitude was nevertheless observed. These changes are probably due to a decreased RGC response and not a consequence of the surgical procedures on the outer retinal activity for the following reasons. First, visual inspection of the retina did not reveal any retinal detachment or evidence of retinal ischaemia that could have resulted in a decreased signal. Second, the ERG a- and b-waves did not differ significantly between the control and experimental eye (data not shown). Third, the peak amplitude of P1 did not differ between control and experimental eyes when recordings were performed under isoflurane anaesthesia, a mode of anaesthesia we showed to reduce the activity of the inner retina (as described in the following sections).
, http://www.100md.com
Hare et al. (Hare & Ton, 2002) obtained similar findings (i.e. reduction in the second trough and loss of a late positive component) in cynomolgous monkeys after the RGC population was reduced by an increase in intraocular pressure. Using a bipolar contact lens along with a stimulation protocol including interleaving flashes, Fortune et al. (2002a) demonstrated that a late response in the component induced following an interleaving bright flash was significantly decreased. These results support the idea that RGC activity can be isolated from the mfERG response and that the RGC response is distinct and of longer implicit time than the one reported from the bipolar cell activity (Hare et al. 2001; Hood et al. 2002). The optic nerve head component delineated by post-acquisition manipulation of the mfERG signal also showed RGC activity in the late part of the response (Sutter et al. 1999). However, one mfERG study in patients with glaucomatous retinopathy has isolated a possible RGC component in the mfERG signal in patients with glaucoma in the form of an early wavelet associated with the primary response (first 30 ms; Sano et al. 2002). Loss of the oscillatory mfERG response pattern in the macaque reported to be associated with RGCs was covering the first 50 ms of the mfERG response (Hood et al. 1999a, b, 2002; Frishman et al. 2000; Fortune et al. 2002b; Ratz et al. 2002). Clearly, species differences exist and different modalities of stimulation may cause the variation in timing of the RGCs response.
, 百拇医药
Naso-temporal asymmetry in the isolated RGC responses was noted by several authors in monkey and human mfERG recordings. From a histological perspective, the naso-temporal asymmetry in the foveated mammals can be associated with the distance from the RGC to the optic nerve head (Hood et al. 2002) and from the fact that nasal and temporal RGCs are responsible for the contralateral and ipsilateral projections to the brain, respectively. In the pig, the naso-temporal asymmetry has different bases. For instance, the point of reference used in the pig is the optic nerve head (Garcia et al. 2005) probably because of the presence of two areas centralis, while the unique fovea in monkeys and humans is used for these species. In the periphery, larger cells are generally found in the temporal retina. Those distinctions taken into account, our data showed naso-temporal asymmetry in the visual streak, with RGC peak time response being shorter in the nasal retina. The idea that this asymmetry is due to different functions for the two areas centralis present in the pig retina is interesting but remains to be investigated. Inconsistent asymmetry could be found in the more peripheral areas despite the histological evidence for a differential soma size distribution (Garcia et al. 2005).
, 百拇医药
The absence of evidence for an optic nerve head response that varies in implicit time with distance from the optic nerve might possibly be explained by a gradient of RGC cell soma size (and consequently of axon size) present in the pig retina that compensates for the distance of the cell relative to the optic nerve head. In the monkey and human ERG recordings, the area investigated (generally a few degrees around the fovea) may present with a more homogenous cell size distribution, resulting in a marked optic nerve head component. This is admittedly purely speculative but further histological studies on RGC soma and axon size distribution in the pig retina may help resolve this issue.
, 百拇医药
Effects of tetrodotoxin
Intraocular injection of TTX has been used in previous studies (Hood et al. 1999a; Frishman et al. 2000; Hare et al. 2001; Hare & Ton, 2002) to isolate the activity of spike-generating cells such as RGCs and a subgroup of amacrine cells. In our study, intraocular injection of TTX in the pig maintained under ketamine sedation resulted in the disappearance of P2 and P3 and their replacement by a series of regular oscillations damping over 200 ms post-stimuli (Fig. 7). The loss of P2 and P3 is in keeping with results described in the cynomolgous monkey after TTX injections and experimental high intraocular pressure (Hare & Ton, 2002). In the macaque, results are more difficult to compare because of the high frequency content of the normal mfERG response; TTX reduced the high frequencies and rendered the response pattern independent of the retinal position (Hood et al. 1999a; Frishman et al. 2000). The replacement of P2 and P3 by a series of oscillations following TTX injection as repetitively observed in our study was not observed in either species of monkey. In the pig, these oscillations were equally present in the control and experimental eyes and thus had to be generated proximal to the RGCs. Further to the involvement of amacrines and RGCs interactions, the involvement of the bipolar cells in the initiation of these oscillations might also be considered as it has recently been demonstrated that some bipolar cell types have voltage-gated sodium channels (Pan & Hu, 2000). These oscillations were not a stimulus artifact as they disappeared after induction of isoflurane anaesthesia and were recovered when isoflurane was discontinued. The exact nature of the oscillatory pattern generated in the mfERG responses after intraocular injection of TTX in the pig remains equivocal but is likely to involve synaptic circuits in the inner plexiform layer.
, http://www.100md.com
Influence of anaesthesia
Our results are in agreement with those of Fortune et al. (2002b) who proposed a detrimental influence of anaesthesia on the mfERG signal in the monkey. They showed a lengthening of implicit time and a decrease in the high-frequency components thought to be generated somewhere within the inner plexiform layer. In the present study, we further defined the action of isoflurane by demonstrating that it directly alters the RGC activity. Recordings performed in the presence of isoflurane anaesthesia resulted in the reduction of all the components shown to be affected by ONS (i.e. RGCs) when recordings were performed under ketamine sedation. Recordings from control and experimental eyes after long-term ONS showed no significant difference under isoflurane anaesthesia. Isoflurane anaesthesia thus appears to prevent the contribution of RGC-related activity in the mfERG signal. Isoflurane anaesthesia also affected the implicit time of P1, but not that of P2 and P3, further suggesting that P1 and P2/P3 may have different retinal origins. Furthermore, isoflurane also abolished the oscillatory activity elicited after intravitreal injection of TTX, suggesting that it also has a site of action upstream to the RGCs and to the TTX-sensitive generators. Our results thus show that, beside alteration in outer retina activity, the RGC activity is much reduced. It is well documented that the central cortical activity is reduced under isoflurane anaesthesia (Sebel et al. 1986); this phenomenon may thus be contributed to by a reduced activity of the more peripheral inputs. The use of isoflurane anaesthesia may potentially represent a fast and reversible method of RGC inactivation for electrophysiological studies.
, 百拇医药
Anaesthetics are pluripotents and generally exert their action through multiple mechanisms. Isoflurane is considered to be responsible for a calcium-induced activation of the GABAA receptor, thereby inducing the prolongation of the GABA-mediated inhibitory postsynaptic currents. (Jones & Harrison, 1993) Isoflurane also depresses AMPA glutamate responses but has only weak interactions with NMDA binding sites (Lin et al. 1993; Maciver, 1997). At clinical concentrations, isoflurane also suppresses voltage-gated sodium channels, which results in a decrease in action potential transmission. Isoflurane is thus likely to influence many synaptic circuits presynaptic to RGCs and to prevent RGC action potentials. On the other hand, ketamine, a dissociative anaesthesic, may have more restricted sites of action with demonstrated effects through non-competitive inhibition of NMDA receptor, with no action on AMPA/kainate receptors (Lin et al. 1993; Peoples & Weight, 1997). Ketamine is also known to enhance cortical responsiveness (Schubert et al. 1990). As the pig mfERG under ketamine sedation results in a large P3 component that cannot be found in human, there is a possibility that P3 is a consequence of the ketamine action. We investigated that possibility by using a GABA antagonist anaesthetic, propofol, and found recordings identical to those recorded under ketamine, thus making it unlikely that P3 is induced by the type of anaesthesic.
, 百拇医药
In summary, our results demonstrate that a unilateral ONS results in a time-dependent amplitude loss of the mfERG late components P2 and P3. This model offers great opportunities to complement neuroanatomical studies on neuroprotection or neurotransplantation with local functional data.
References
Bach M & Speidel-Fiaux A (1989). Pattern electroretinogram in glaucoma and ocular hypertension. Doc Ophthalmol 73, 173–181.
, http://www.100md.com
Bearse MA Jr, Sutter E, Sim D & Stamper R (1996). Glaucomatous dysfunction revealed in higher order components of the electroretinogram. Vision Science and its Applications, OSA Tech Dig Ser, pp. 104–107. Optical Society of America, Washington, DC.
Berardi N, Domenici L, Gravina A & Maffei L (1990). Pattern ERG in rats following section of the optic nerve. Exp Brain Res 79, 539–546.
Berkemeier M, Clarke BD, Wang W-C, Bray GM & Aguayo AJ (1994). Axotomy results in delayed death and apoptosis of retinal ganglion cells in adult rats. J Neurosci 14, 4368–4374.
, http://www.100md.com
Bui BV & Fortune B (2003). Ganglion cell contributions to the rat full-field electroretinogram. J Physiol 555, 153–173.
Chan HL & Brown B (1999). Multifocal ERG changes in glaucoma. Ophthalmic Physiol Opt 19, 306–316.
Chandler MJ (1999). Photoreceptor density of the domestic pig retina. Vet Ophthalmol 2, 179–184.
Clements JA & Nimmo WS (1981). Pharmacokinetics and analgesic effect of ketamine in man. Br J Anaesth 53, 27–30.
, http://www.100md.com
Colotto A, Falsini B, Salgarello T, Iarossi G, Galan ME & Scullica L (2000). Photopic negative response of the human ERG: losses associated with glaucomatous damage. Invest Ophthalmol Vis Sci 41, 2205–2211.
Cursiefen C, Korth M & Horn F (2000). The negative response of the flash electroretinogram in glaucoma. Doc Ophthalmol 103, 1–12.
Drasdo N, Aldebasi Y & Chiti Z (2001). The s-cone PhNR and pattern ERG in primary open angle glaucoma. Invest Ophthal Vis Sci 42, 1266–1272.
, 百拇医药
Fortune B, Bearse MA Jr, Cioffi GA & Johnson CA (2002a). Selective loss of an oscillatory component from temporal retinal multifocal ERG responses in glaucoma. Invest Ophthal Vis Sci 43, 2638–2647.
Fortune B, Cull G, Wang L, van Buskirk EM & Cioffi GA (2002b). Factors affecting the use of multifocal electroretinography to monitor function in a primate model of glaucoma. Doc Ophthalmol 105, 151–178.
Fortune B, Johnson CA & Cioffi GA (2001). The topographic relationship between multifocal electroretinographic and behavioral perimetric measures of function in glaucoma. Optometry Vision Sci 78, 206–214.
, 百拇医药
Frishman LJ, Saszik S, Harwerth RS, Viswanathan S, Li Y, Smith EL III, Robson JG & Barnes G (2000). Effects of experimental glaucoma in macaques on the multifocal ERG. Multifocal ERG in laser-induced glaucoma. Doc Ophthalmol 100, 231–251.
Garcia M, Ruiz-Ederra J, Hernandez-Barbachano H & Vecino E (2005). Topography of pig retinal ganglion cells. J Comparative Neurol 486, 361–372.
Garcia-Valenzuela E, Gorczyca W, Darzynkiewick Z & Sharma SC (1994). Apoptosis in adult retinal ganglion cells after axotomy. J Neurobiol 25, 431–438.
, 百拇医药
Germain F, Calvo M & de la Villa P (2004). Rabbit retinal ganglion cell survival after optic nerve section and its effect on the inner plexiform layer. Exp Eye Res 78, 95–102.
Ghosh F & Arner K (2002). Transplantation of full-thickness retina in the normal porcine eye: surgical and morphologic aspects. Retina 22, 478–486.
Ghouri AF, Bodner M & White PF (1991). Recovery profile after desflurane-nitrous oxide versus isoflurane-nitrous oxide in outpatients. Anesthesiology 7, 419–424.
, 百拇医药
Hare WA & Ton H (2002). Effects of APB, PDA, and TTX on ERG responses recorded using both multifocal and conventional methods in monkey. Doc Ophthalmol 105, 189–222.
Hare WA, Ton HGR, Feldmann B, Wijino M & Woldemussie E (2001). Characterization of retinal injury using ERG measures obtained with both conventional and multifocal methods in chronic ocular hypertensive primates. Invest Ophthalmol Vis Sci 42, 127–136.
Harwerth RS, Crawford MLJ, Frishman L, Viswanathan S, Smith EL & Carter-Dawson L (2002). Visual field defects and neural losses from experimental glaucoma. Prog Retinal Eye Res 21, 91–125.
, 百拇医药
Hasegawa S, Tagaki M, Usui T, Takada R & Abe H (2000). Waveform change of the first-order multifocal electroretinogram in patients with glaucoma. Invest Ophthal Vis Sci 41, 1597–1603.
Hendrickson A & Hicks D (2002). Distribution and density of medium- and short-wavelength selective cones in the domestic pig retina. Exp Eye Res 74, 435–444.
Holder G (2001). Pattern electroretinography (PERG) and an integrated approach to visual pathway diagnosis. Prog Retinal Eye Res 20, 531–561.
, http://www.100md.com
Hood DC, Bearse MA Jr, Sutter EE, Viswanathan S & Frishman LJ (2001). The optic nerve head component of the monkey's (Macaca mulatta) multifocal electroretinogram (mERG). Vision Res 41, 2029–2041.
Hood DC, Frishman LJ, Saszik S & Viswanathan S (2002). Retinal origins of the primate multifocal ERG: implications for the human response. Invest Ophthalmol Vis Sci 43, 1673–1685.
Hood DC, Frishman LJ, Viswanathan S, Robson JG & Ahmed J (1999a). Evidence for a ganglion cell contribution to the primate electroretinogram (ERG): effects of TTX on the multifocal ERG in macaque. Vis Neurosci 16, 411–416.
, 百拇医药
Hood DC, Greenstein V, Frishman L, Holopigian K, Viswanathan S, Seiple W, Ahmed J & Robson JG (1999b). Identifying inner retinal contributions to the human multifocal ERG. Vision Res 39, 2285–2291.
Hood DC, Greenstein VC, Holopigian K, Bauer R, Firoz B, Liebmann JM, Odel JG & Ritch R (2000). An attempt to detect glaucomatous damage to the inner retina with the multifocal ERG. Invest Ophthalmol Vis Sci 41, 1570–1579.
Jones MV & Harrison NL (1993). Effects of volatiles anesthetics on the kynetics of inhibitory postsynaptic currents in cultured rat hippocampal neurons. J Neurophysiol 70, 1339–1349.
, 百拇医药
Kielczewski JL, Pease ME & Quigley HA (2005). The effect of experimental glaucoma and optic nerve transection on amacrine cells in the rat retina. Invest Ophthalmol Vis Sci 46, 3188–3196.
Klistorner AI, Graham SL & Martins A (2000). Multifocal pattern electroretinogram does not demonstrate localised field defects in glaucoma. Doc Ophthalmol 100, 155–165.
Komaromy AM, Brooks DE, Kallberg ME, Dawson WW, Szel A, Lukats A, Samuelson DA, Sapp HL Jr, Gelatt KN & Sherwood MB (2003). Long-term effect of retinal ganglion cell axotomy on the histomorphometry of other cells in the porcine retina. J Glaucoma 12, 307–315.
, 百拇医药
Korth M (1997). The value of electrophysiologic testing in glaucomatous diseases. J Glaucoma 6, 331–343.
Levin V (2001). Models of neural injury. J Glaucoma 10, s19–s21.
Li ZY, Wong F, Chang JH, Possin DE, Hao Y, Petters RM & Milam AH (1998). Rhodopsin transgenic pigs as a model for human retinitis pigmentosa. Invest Ophthal Vis Sci 39, 808–819.
Lin LH, Whiting P & Harris RA (1993). Molecular determinants of general anesthetic action: role of GABAa receptor structure. J Neurochem 60, 1548–1553.
, http://www.100md.com
Maciver MB (1997). General anesthetic actions on transmission at glutamate and GABA synapses. In Anesthesia: Biological Foundations, ed. Yaksh TI, Lynch C, Zapol WM, Maze M, Biebuyck JF & Saidman LJ, pp. 277–286. Lippincott-Raven, Pliladelphia.
Maffei L & Fiorentini A (1981). Electroretinographic responses to alternating gratings before and after section of the optic nerve. Science 211, 953–955.
Maffei L, Fiorentini A, Bisti S & Hollander H (1985). Pattern ERG in the monkey after section of the optic nerve. Exp Brain Res 59, 423–425.
, 百拇医药
Otto T & Bach M (1996). Retest variability and diurnal effects in the pattern electroretinogram. Doc Ophthalmol 92, 311–323.
Palmowski AM, Allgayer R, Heinemann-Vernaleken B & Ruptecht KW (2002). Multifocal electroretinogram with a multiflash stimulation technique in open-angle glaucoma. Ophthalmic Res 43, 83–89.
Pan ZH & Hu HJ (2000). Voltage-dependent Na+ currents in mammalian retinal cone bipolar cells. J Neurophysiol 84, 2564–2571.
, 百拇医药
Peoples RW & Weight FF (1997). Anesthetic actions on excitatory amino acid receptors. In Anesthesia: Biological Foundations, ed. Yaksh TI, Lynch C, Zapol WM, Maze M, Biebuyck JF & Saidman LJ, pp. 239–258. Lippincott-Raven, Philadelphia.
Prince JH, Diesem CD, Eglitis I & Ruskell GL (1960). The pig. In Anatomy and Histology of the Eye and Orbit in Domestic Animals, ed. Thomas CC, pp. 210–230. Charles C. Thomas, Springfield, Illinois.
, http://www.100md.com
Ratz D, Seeliger MW, Geva AB, Perlicot CLM, Lambrou GN & Ofri R (2002). The effect of contrast and luminance on mferg responses in a monkey model of glaucoma. Invest Ophthal Vis Sci 43, 2027–2035.
Rosolen S, Rigaudiere F & Le Gargasson J (2003). A new model of induced ocular hyperpressure using the minipig (in French). J Fr Ophtalmol 26, 259–267.
Ruiz-Ederra J, Garcia M, Hernandez M, Urcola H, Hernandez-Barbachano E, Araiz J & Vecino E (2005). The pig eye as a novel model of glaucoma. Exp Eye Res 81, 561–569.
, 百拇医药
Sachs HG, Schanze T, Brunner U, Sailer H & Wiesenack C (2005). Transscleral implantation and neurophysiological testing of subretinal polyimide film electrodes in the domestic pig in visual prosthesis development. J Neurol Eng 2, S57–S64.
Sano M, Tazawa Y, Nobeshima T & Mita M (2002). A new wavelet in the multifocal electroretinogram, probably originating from ganglion cells. Invest Ophthalmol Vis Sci 43, 1666–1672.
Schubert A, Licina MG & Lineberry PJ (1990). The effect of ketamine on human somatosensory evoked potentials and its modification by nitrous oxide. Anesthesiology 72, 33–39.
, 百拇医药
Sebel PS, Ingram DA, Flynn PJ, Rutherfoord CF & Rogers H (1986). Evoked potentials during isoflurane anaesthesia. Br J Anaesth 58, 580–585.
Sieving PA, Frishman LJ & Steinberg RH (1986). Scotopic threshold response of proximal retina in cat. J Neurophysiol 56, 1049–1061.
Sutter EE & Bearse MA Jr (1999). The optic nerve head component of the human ERG. Vision Res 39, 419–436.
Sutter E, Shimada Y, Li Y & Bearse MA (1999). Mapping inner retinal function through enhancement of adaptative components in the MF-ERG. Vis Sci App OSA Tech Dig Series 1, 52–55.
, http://www.100md.com
Sutter EE & Tran D (1992). The field topography of ERG components in man – 1. The photopic luminance response. Vision Res 32, 433–446.
Tso MO, Li WW, Zhang C, Lam TT, Hao Y, Petters RM & Wong F (1997). A pathologic study of degeneration of the rod and cone populations of the rhodopsin Pro347Leu transgenic pigs. Trans Am Ophthalmol Soc 95, 467–479.
Vaegan, Anderton PJ & Millar TJ (2000). Multifocal, pattern and full field electroretinograms in cats with unilateral optic nerve section. Doc Ophthalmol 100, 207–229.
, 百拇医药
Vecino E, Ruis-Ederra J, Hernendez M, Urcola JA, Araiz J, Duren J & Garcia M (2004). Topography of retinal ganglion cells of the pig retina – cell size distribution. ARVO on-line abstract 5434.
Viswanathan S, Frishman LJ, Robson JG, Harwerth RS & Smith EL III (1999). The photopic negative response of the macaque electroretinogram: reduction by experimental glaucoma. Invest Ophthalmol Vis Sci 40, 1124–1136.
Viswanathan S, Frishman LJ, Robson JG & Walters JW (2001). The photopic negative response of the flash electroretinogram in primary open angle glaucoma. Invest Ophthalmol Vis Sci 42, 514–522.
, 百拇医药
Watanabe M, Inukai N & Fukuda Y (2001). Survival of retinal ganglion cells after transection of the optic nerve in adult cats: a quantitative study within two weeks. Vis Neurosci 18, 137–145.
Weishaupt J & Bahr M (2001). Degeneration of axotomized retinal ganglion cells as a model for neuronal apoptosis in the central nervous system – molecular death and survival pathways. Restor Neurol Neurosci 19, 19–27., 百拇医药(Mélanie R. Lalonde, Balwantray C. Chauha)