Estimating ON and OFF contributions to the photopic hill: normative data and clinical applications

BACKGROUND

With progressively brighter stimuli, the amplitude of the b-wave of the human photopic electroretinogram (ERG) first increases to a maximal value (Vmax) and then decreases to finally reach a plateau, a phenomenon known as the photopic hill (PH). A mathematical model combining a Gaussian (G) and a logistic (L) growth function was previously proposed to fit this unusual luminance-response curve, where the G and L functions were suggested to represent, respectively, the OFF and ON retinal pathway contributions to the building of the PH.

METHOD

The PHs of patients presenting stationary diseases affecting specifically the ON (3 CSNB-1) or OFF (4 CPCPA) retinal pathways as well as patients affected with retinitis pigmentosa (14 RP) of different stages or etiology were analyzed using this mathematical model and compared to the PHs of a group of 28 normal subjects.

RESULTS

The PH of the CSNB-1 patients had a much larger contribution from the G function compared to normal subjects, whereas the opposite was observed for the CPCPA patients. On the other hand, analysis of data from RP patients revealed variable G-L contributions to the building of their PH.

CONCLUSION

In this study, we confirm the previous claim that the luminance-response function of the photopic ERG b-wave can be decomposed into a Gaussian function and a logistic growth function representing, respectively, the OFF and ON retinal pathways. Furthermore, our findings suggest that this mathematical decomposition could be useful to further segregate and potentially follow the progression of retinopathies such as RP.

Comparing the retinal structures and functions in two species of gulls (Larus delawarensis and Larus modestus) with significant nocturnal behaviours

Ring-billed gulls (Larus delawarensis) and gray gulls (Larus modestus) are two species active both by day and night. We have investigated the retinal adaptations that allow the diurnal and nocturnal behaviours of these two species. Electroretinograms and histological analyses show that both species have a duplex retina in which cones outnumber rods, but the number of rods appears sufficient to provide vision at night. Their retinas respond over the same scotopic dynamic range of 3.4logcdm(-2), which encompasses all of the light levels occurring at night in their photic environment. The amplitudes of the scotopic saturated a- and b-wave responses as well as the photopic saturated b-wave response and the photopic sensitivity parameter S are however higher in ring-billed gulls than in gray gulls. Moreover, the process of dark adaptation is about 30min faster in gray gulls than in ring-billed gulls. Our results suggest that both species have acquired in the course of their evolution functional adaptations that can be related to their specific photic environment.

Comparison of the Retinal Structure and Function in Four Bird Species as a Function of the Time They Start Singing in the Morning

We postulated that the retinas of bird species that are the earlier singers are more sensitive to low light conditions than species that sing closer to sunrise. The selected species were the American Robin (Turdus migratorius) and the Hermit Thrush (Catharus guttatus) as early singers, the Common Grackle (Quiscalus quiscula) and the Mourning Dove (Zenaida macroura) which join the dawn chorus near sunrise. Scotopic electroretinogram (ERGs) intensity-response functions were obtained from anesthetized birds, following which the animals were euthanized and their retinas processed for histological analysis. Based on k values, generally considered an adequate measurement of the scotopic (rod) retinal sensitivity, all species yielded comparable night vision capabilities. However, based on the maximal (mixed rod-cone response) amplitude of the scotopic b-wave, our results indicate that robins and thrushes yield larger scotopic ERGs compared with doves and grackles, but unexpectedly do not have higher rod:cone ratios. Increased thickness of the retinal inner nuclear layer and higher numbers of ganglion cells in robins and thrushes compared with doves and grackles suggest a greater number of synaptic connections maximizing vision under low light conditions, and might support their higher retinal responses under scotopic conditions. The higher ERG V(max) of robins and thrushes might also be explained, at least in part, by the optics (i.e., their lower minimum F-numbers), resulting in brighter retinal images, rather than from higher sensitivity of their retina. Our results suggest that an early onset of dawn singing might be correlated with the retinal ability to detect the first dim crepuscular lights.

Effects of the intravitreal administration of dopaminergic ligands on the b-wave amplitude of the rabbit electroretinogram

In the retina of mammals, dopamine (DA) is generally released by amacrine cells and is known to alter the physiology of most retinal cells. It is well known that DA reduces the amplitude of the b-wave of the electroretinogram (ERG) in rabbit. However, the specific receptor subtypes that mediate this action have not yet been elucidated. To do this, we recorded flash ERGs before and after the intravitreal injection of D1-like DA receptor agonists (SKF38393, A77693) and antagonist (SCH23390), and of D2-like agonist (R(-)-propylnorapomorphine hydrochloride; NPA) and antagonist ((S)-(-)-sulpiride). Contralateral control eyes were injected with the vehicle only. Both D1 agonists provoked a reduction of the ERG b-wave amplitude (34.0% and 59.2% of the pre-injection level, respectively). The D2-like agonist NPA had no significant effects on ERG components. Unexpectedly, both D1- and D2-like antagonists also reduced the b-wave amplitude (28.9% and 59.8%). Overall, these data suggest that the previously described effect of DA on the rabbit ERG b-wave came from activation of D1-like receptors. On the basis of the effects observed with D2-like antagonist, a subtle contribution of D2-like presynaptic receptors cannot be ruled out.

Cyclooxygenase-2 in human and experimental ischemic proliferative retinopathy

BACKGROUND

Intravitreal neovascular diseases, as in ischemic retinopathies, are a major cause of blindness. Because inflammatory mechanisms influence vitreal neovascularization and cyclooxygenase (COX)-2 promotes tumor angiogenesis, we investigated the role of COX-2 in ischemic proliferative retinopathy.

METHODS AND RESULTS

We describe here that COX-2 is induced in retinal astrocytes in human diabetic retinopathy, in the murine and rat model of ischemic proliferative retinopathy in vivo, and in hypoxic astrocytes in vitro. Specific COX-2 but not COX-1 inhibitors prevented intravitreal neovascularization, whereas prostaglandin E2, mainly via its prostaglandin E receptor 3 (EP3), exacerbated neovascularization. COX-2 inhibition induced an upregulation of thrombospondin-1 and its CD36 receptor, consistent with the observed antiangiogenic effects of COX-2 inhibition; EP3 stimulation reversed effects of COX-2 inhibitors on thrombospondin-1 and CD36.

CONCLUSIONS

These findings point to an important role for COX-2 in ischemic proliferative retinopathy, as in diabetes.

Effects of methanol on the retinal function of juvenile rats

We have investigated the effect of methanol exposure on the retinal function of juvenile rats. The electroretinogram (ERG) and oscillatory potentials (OPs) were recorded prior to and up to 72 h after the administration of methanol. Data were compared to a control group which was only exposed to physiological saline. Our findings can be summarized as follows: methanol generally reduced the amplitude of all retinal potentials, and in some cases, to baseline levels. The ERG b-wave was affected earlier and more prominently than the a-wave. All measured OPs (2-4) were decreased but OP2 was less affected, suggesting that the cone pathway may be less sensitive to methanol than the rod-mediated pathway. These data indicate that juvenile rats (21 days old, i.e. with an immature synaptic development) present a sensitivity to methanol comparable to that observed in adult animals.

Role of thromboxane in retinal microvascular degeneration in oxygen-induced retinopathy

Microvascular degeneration is an important event in oxygen-induced retinopathy (OIR), a model of retinopathy of prematurity. Because oxidant stress abundantly generates thromboxane A2 (TxA2), we tested whether TxA2 plays a role in retinal vasoobliteration of OIR and contributes to such vascular degeneration by direct endothelial cytotoxicity. Hyperoxia-induced retinal vasoobliteration in rat pups (80% O2 exposure from postnatal days 5-14) was associated with increased TxB2 generation and was significantly prevented by TxA2 synthase inhibitor CGS-12970 (10 mg x kg(-1) x day(-1)) or TxA2-receptor antagonist CGS-22652 (10 mg x kg(-1) x day(-1)). TxA2 mimetics U-46619 (EC50 50 nM) and I-BOP (EC50 5 nM) caused a time- and concentration-dependent cell death of neuroretinovascular endothelial cells from rats as well as newborn pigs but not of smooth muscle and astroglial cells; other prostanoids did not cause cell death. The peroxidation product 8-iso-PGF2, which is generated in OIR, stimulated TxA2 formation by endothelial cells and triggered cell death; these effects were markedly diminished by CGS-12970. TxA2-dependent neuroretinovascular endothelial cell death was mostly by necrosis and to a lesser extent by apoptosis. The data identify an important role for TxA2 in vasoobliteration of OIR and unveil a so far unknown function for TxA2 in directly triggering neuroretinal microvascular endothelial cell death. These effects of TxA2 might participate in other ischemic neurovascular injuries.

Response characteristics of the normal retino-cortical pathways as determined with simultaneous recordings of pattern visual evoked potentials and simple motor reaction times

PURPOSE

In an attempt to explain the existing discrepancies regarding the relationship between electrophysiological and psychophysical measurements of visual transmission time we compared, in humans, the response characteristics of the normal retino-cortical pathways with simultaneously obtained pattern visual evoked potentials (PVEP) and simple motor reaction times (RT).

METHODS

PVEPs and manual RTs were recorded simultaneously using a reversing checkerboard with different spatial frequency and contrast combinations chosen to elicit responses favoring the magnocellular or parvocellular pathways. The amplitude and peak time of the P1 wave of the PVEP were compared to the mean RT. Other parameters of the RT, such as mode and standard deviation were also considered.

RESULTS

The RT is not modified in the same fashion as the peak time of the P1 wave of the PVEP, the peak time of the PVEP demonstrating a spatial frequency selectivity, while the RT does not. Further comparative analysis of the PVEP and RT shows that the RT is faster for stimuli of lower contrast and spatial frequency, while the PVEP amplitude is larger and its peak time shorter for higher contrast and spatial frequency stimuli.

CONCLUSIONS

Our findings suggest that PVEP and RT measures recruit distinct physiological characteristics and appear to be differently modulated while travelling along the retino-cortical pathway. Our results also show the importance of obtaining electrophysiological and psychophysical measures concomitantly to insure elimination of combined inter-stimulus and inter-session variability.

Graded contribution of retinal maturation to the development of oxygen-induced retinopathy in rats

PURPOSE

Newborn rats exposed to hyperoxia during the first days of life have been shown to exhibit not only vasculopathy but also permanent changes in the structure and function of the retina. Given that the rat retina is immature at birth and that the maturation process continues until the opening of the eyes at 14 days of life, this study was conducted to investigate the susceptibility of the retina to oxygen toxicity as a function of the degree of retinal maturity reached at the time of oxygen exposure.

METHODS

Newborn rats were exposed to hyperoxia during selected postnatal day intervals. Scotopic electroretinograms were recorded at 30 and 60 days of age, and retinal histology was obtained at the end of the study.

RESULTS

There was a strong correlation between the duration of the hyperoxic event and the structural and functional consequences in the retina. However, the repercussions were significantly more profound when the exposure to oxygen occurred within the second week of life (6-14 days), compared with earlier (0-6 days) or later periods (14-28 days).

CONCLUSIONS

The results strongly suggest that the structural and functional retinal changes secondary to postnatal hyperoxia are not only the direct consequence of exposure to high levels of oxygen (i.e., free radicals), but also are determined by the level of retinal maturity reached at the time of oxygen exposure. The results also indicate that the structural anomalies precede the functional impairments.

A physiological basis for definition of the ISCEV ERG standard flash (SF) based on the photopic hill

The ISCEV Standard for Clinical Electrophysiology indicates that the ERG standard flash should be defined within a very narrow range of intensities. Yet no information is provided as to how this intensity range was identified. We present evidence that would support a redefinition of the SF based on known photopic ERG properties.

The electroretinogram recorded at the onset of dark-adaptation: understanding the origin of the scotopic oscillatory potentials

The purpose of this study was to examine the origin of the oscillatory potentials (OPs) recorded at the onset of dark-adaptation. Our results show that following pre-exposure of the retina to progressively brighter photopic backgrounds there is a complete abolition of OP4 and approximately 50% of OP3, while OP2 is not affected in responses evoked to dim flashes of white light and recorded at the onset of dark-adaptation. These results bring further support to the claim that the short latency OP2 is cone-mediated while the OP3 and OP4 would have a significant rod contribution. However, a more complex picture of OP genesis arises when flicker and response to brighter white light flashes, also obtained within the first minute of dark-adaptation are considered. The latter would suggest that our understanding of the origin of the OPs cannot be exclusively based on which of the two class of photoreceptors is preferentially stimulated at the time the response is recorded.

Recording the photopic electroretinogram from conscious adult Yucatan micropigs

The micropig is viewed by many as one of the best animal models of human system physiology. The purpose of this study was to develop a simple, rapid and reliable method to record clinical electroretinograms from conscious adult Yucatan micropigs. Photopic ERGs were recorded from 18 conscious animals following a 3-h period of preadaptation to a moderate photopic environment. Rectal temperature and blood pressure were monitored throughout the experiment. In order to test the reproducibility of our procedure, the same protocol was repeated (n=6 micropigs) on three different recording sessions each separated by a 48-h period. Our results indicate that the recording of reproducible ERGs, in the conscious Yucatan micropig, is feasible provided that an adequate experimental approach is used. We believe that our method will enable investigators to perform retinal physiology experiments using an approach which is compatible with that used in a human clinical setup thus facilitating the transposition of experimental data from animal to human.

Oxidants, nitric oxide and prostanoids in the developing ocular vasculature: a basis for ischemic retinopathy

The choroid is the main source of oxygen to the retina. In contrast to the adult, the absence of autoregulation of choroidal blood flow in the newborn leads to hyperoxygenation of the retina. In the immature retina which contains relatively low levels of antioxidants this hyperoxygenation favors peroxidation including the generation of biologically active isoprostanes, and results in vasoconstriction and vascular cytotoxicity leading to ischemia, which predisposes to the development of a vasoproliferative retinopathy, commonly termed retinopathy of prematurity. During frequently encountered oxidative stress to the perinate, the combined absence of vascular autoregulation and excessive oxygen delivery to the eyes of the developing subject is largely the result of a complex epigenetic and genetic interplay between prostanoids and nitric oxide (NO) systems on vasomotor regulation. The effects of certain prostaglandins are NO-dependent; conversely, those of NO have also been found to be largely prostaglandin I(2)-mediated in the eye; and NO synthase expression seems to be significantly regulated by other prostaglandins apparently through activation of functional perinuclear prostanoid receptors which affect gene transcription. The increased production of both prostaglandins and NO in the perinate augment ocular blood flow and as a result oxygen delivery to an immature retina partly devoid of antioxidant defenses. The ensuing peroxidation results in impaired circulation (partly thromboxane A(2)-dependent) and vascular integrity, leading to ischemia which predisposes to abnormal preretinal neovascularization, a major feature of ischemic retinopathy. Because tissue oxygenation is largely dependent upon circulation and critical in the generation of reactive oxygen species, and since the latter exert a major contribution in the pathogenesis of retinopathy of prematurity, it is important to understand the mechanisms that govern ocular blood flow. In this review we focus on the important and complex interaction between prostanoid, NO and peroxidation products on circulatory control of the immature retina.

Diurnal and nocturnal visual capabilities in shorebirds as a function of their feeding strategies

Some shorebird species forage with the same feeding strategy at night and during daytime, e.g. visual pecking in the Wilson's Plover (Charadrius wilsonia) or tactile probing in the Short-billed Dowitcher (Limnodromus griseus). The American Woodcock (Scolopax minor) uses tactile probing, by day and by night, but sometimes pecks for insects during daytime. The Black-winged Stilt (Himantopus himantopus) is a visual pecker, both by day and by night, and sometimes forages tactilely on windy (agitated water surface) moonless nights. Territorial Willets (Catoptrophorus semipalmatus) are visual peckers during daylight and on moonlight conditions but switch to tactile feeding under lower light conditions. It could be postulated that some shorebird species would switch from visual feeding during daytime to tactile foraging at night because they have poor night vision compared to species that are always sight foragers irrespective of the time of the day. This issue was examined by comparing retinal structure and function in the above species. Electroretinograms (ERGs) were obtained at different light intensities from anesthetized birds, and the retinae were processed for histological observations. Based on ERGs, retinal sensitivity, and rod:cone ratios, both plovers and stilts are well adapted for nocturnal vision. Although they have low rod density compared to that of stilts and plovers, Willets and woodcocks have a scotopic retinal sensitivity similar to that of stilts and plovers but rank midway between plovers and dowitchers for the b-wave amplitude. Dowitchers have the lowest scotopic b-wave amplitude and retinal sensitivity and appear the least well adapted for night vision. Based on photopic ERGs and cone densities, although stilts, Willets and dowitchers appear as well adapted for daytime vision, plovers occupy the last rank of all species examined. Compared to the nighttime tactile feeders and those that switch from daytime visual pecking to tactile feeding at night, nighttime sight feeders have a superior rod function and, consequently, potentially superior nocturnal visual capabilities.

Preservation of neural function in the perinate by high PGE(2) levels acting via EP(2) receptors

Despite increasingly frequent and longer lasting hypoxic episodes during progressive labor, the neonate is alert and vigorous at birth. We investigated whether high levels of PGs during the perinatal period assist in preserving neural function after such "stressful" hypoxic events. Visual evoked potentials (VEPs) and electroretinograms (ERGs) were recorded before and 45 min after mild moderate asphyxic hypoxia (two 4-min asphyxic-hypoxic periods induced by interrupting ventilation at 8-min intervals) in newborn piglets <12 h old treated or not treated with inhibitors of PG synthase (ibuprofen or diclofenac) with or without PG analogs. At 45 min after the hypoxic episode, P2 and b-wave amplitudes were slightly decreased and latencies were delayed. These changes in the VEP and ERG returned to near normal by 120 min. Ibuprofen and diclofenac decreased brain and retinal PG levels and markedly intensified 45 min after hypoxia-induced changes in VEP and ERG, but cerebral and retinal blood flows improved. Combined treatment with PG synthase inhibitor in combination with 16,16-dimethyl-PGE(2) (a PGE(2) analog), but not with PGI(2) and PGF(2alpha) analogs, and in combination with the EP(2) receptor agonist butaprost (but not EP(1) or EP(3) agonists), prevented ibuprofen- and diclofenac-aggravated postasphyxia electrophysiological changes. In conclusion, high levels of PGE(2) in nervous tissue, via actions on EP(2) receptors, seem to contribute to preservation of neural function in the perinate subjected to frequent hypoxic events.

Transient enhancing of cone electroretinograms following exposure to brighter photopic backgrounds

Normal subjects were first light adapted to a standard photopic background and a control photopic ERG was obtained. The subjects were then light adapted to a brighter background for 5 min at the end of which the luminance was returned to the control background and ERGs were taken at regular intervals. Most of the ERG/OP components were transiently enhanced following the above procedure. Given that the previously reported photopic light adaptation effect occurred following an increase in the luminance of the adapting field (from dark adaptation to light adaptation) while that reported in the present study is triggered following a decrease in the level of light adaptation, the opposite effects noted might suggest that the two retinal events result from the same, not yet identified, cone adaptation mechanisms which are solicited in an opposite way.

Augmented vasoconstriction and thromboxane formation by 15-F(2t)-isoprostane (8-iso-prostaglandin F(2alpha)) in immature pig periventricular brain microvessels

BACKGROUND AND PURPOSE

Oxidant stress, especially in the premature, plays a major role in the pathogenesis of hypoxic-ischemic encephalopathies mostly manifested in the periventricular region. We studied the vasomotor mode of actions of the peroxidation product 15-F(2t)-isoprostane (15-F(2t)-IsoP) (8-iso-prostaglandin F(2alpha)) on periventricular region during development.

METHODS

Effects of 15-F(2t)-IsoP on periventricular microvessels of fetal, newborn, and juvenile pigs were studied by video imaging and digital analysis techniques. Thromboxane formation and intracellular Ca(2+) were measured by radioimmunoassay and by using the fluorescent indicator fura 2-AM.

RESULTS

15-F(2t)-IsoP-mediated constriction of periventricular microvessels decreased as a function of age such that in the fetus it was approximately 2.5-fold greater than in juvenile pigs. 15-F(2t)-IsoP evoked more thromboxane formation in the fetus than in the newborn, which was greater than that in the juvenile periventricular region; this was associated with immunoreactive thromboxane A(2) (TXA(2)) synthase expression in the fetus that was greater than that in newborn pigs, which was greater than that in juvenile pigs. 15-F(2t)-IsoP-induced vasoconstriction was markedly inhibited by TXA(2) synthase and receptor blockers (CGS12970 and L670596). Vasoconstrictor effects of the TXA(2) mimetic U46619 on fetal, neonatal, and juvenile periventricular microvessels did not differ. 15-F(2t)-IsoP increased TXA(2) synthesis by activating Ca(2+) influx through non-voltage-gated channels in endothelial cells (SK&F96365 sensitive) and N-type voltage-gated channels (omega-conotoxin sensitive) in astrocytes; smooth muscle cells were not responsive to 15-F(2t)-IsoP but generated Ca(2+) transients to U46619 via L-type voltage-sensitive channels.

CONCLUSIONS

15-F(2t)-IsoP causes periventricular brain region vasoconstriction in the fetus that is greater than that in the newborn, which in turn is greater than that in the juvenile due to greater TXA(2) formation generated through distinct stimulatory pathways, including from endothelial and astroglial cells. The resulting hemodynamic compromise may contribute to the increased vulnerability of the periventricular brain areas to oxidant stress-induced injury in immature subjects.

Behavioral, morphological and physiological correlates of diurnal and nocturnal vision in selected wading bird species

We examined in selected wading bird species if diurnal or nocturnal foraging and the use of visual or tactile feeding strategies could be correlated with retinal structure and function. The selected species were the Yellow-crowned Night Heron (Nycticorax violaceus), a crepuscular and nocturnal forager, the Great Blue Heron (Ardea herodias), a mainly crepuscular, but also diurnal and nocturnal feeder, the Roseate Spoonbill (Ajaia ajaja), a mainly crepuscular feeder which forages more at night than during the day, the Cattle (Bubulcus ibis) and Tricolored (Egretta tricolor) egrets and the American White Ibis (Eudocimus ruber) which forage only during daytime. Herons and egrets are visual foragers; ibises and spoonbills are tactile feeders. Electroretinograms were obtained from anesthetized birds in photopic and scotopic conditions to a wide range of light intensities, following which the retinae were processed for histological analysis. Based on rod densities and rods:cones ratios, nocturnal vision capability is greater in the Yellow-crowned Night Heron, followed by the Great Blue Heron and the spoonbill, then by the egrets and the ibis. Visual feeders that forage near dawn or dusk or at night have a higher rods:cones ratio, and consequently a greater night vision capability, than visual feeding species which forage only during daytime. Visual nocturnal feeders have a night vision capability greater than tactile diurnal as well as tactile nocturnal feeders. However, based on maximum scotopic b-wave amplitudes, all species studied have roughly comparable night vision capability. The factor that best discriminates between wading bird species appears to be the daytime visual capabilities. Indeed, the diurnal ibis and egrets have similar cone densities, cones:rods ratios, and photopic a-wave amplitudes, values which are greater than those measured in the two nocturnally active heron species.

Evidence supportive of a functional discrimination between photopic oscillatory potentials as revealed with cone and rod mediated retinopathies

We report on a family where four of the eleven children presented with reduced visual acuities, a red-green deficit at the Farnsworth-Munsel FM 100-hue test, normal appearing fundi and unexpected electroretinographic findings. Light- (photopic) and dark- (scotopic) adapted electroretinograms (ERG) and oscillatory potentials (OPs) were obtained following an accepted standard protocol. The b-wave of their photopic ERG was significantly more attenuated than the a-wave due to the specific abolition of OP4, while the amplitudes of OP2 and OP3 were within the normal range, giving to the b-wave a truncated appearance reminiscent of that seen in congenital stationary night blindness (CSNB) with myopia. Interestingly in the latter condition, which is believed to result from an ON-retinal pathway anomaly, it is OP2 and OP3 which are specifically abolished while OP4 is of normal amplitude thus resulting in an OP response pattern which complements that seen with our patients. Also of interest is the fact that, in our patients, the amplitude of the dark-adapted OP2 was, on average, 240% larger than that measured in light-adaptation while, in normal, a non-significant 14% increase is noted; a finding which is in keeping with other studies reporting supernormal scotopic ERGs in some forms of cone dystrophies. Based on the photopic OP response pattern, our patients represent the electrophysiological complement of patients affected with CSNB. Interestingly their symptoms are also complementary, a finding which could support a functional discrimination between the photopic OPs.

Reproducibility of ERG responses obtained with the DTL electrode

Previous investigators have suggested that the DTL fibre electrode might not be suitable for the recording of replicable electroretinograms. We present experimental evidence that when used adequately, this electrode does permit the recording of highly reproducible retinal potentials.

Can interocular pattern reversal visual evoked potential and motor reaction time differences distinguish anisometropic from strabismic amblyopia?

PURPOSE

With the use of the pattern reversal visual evoked potential and the motor reaction time, we sought to differentiate anisometropic amblyopia from strabismic amblyopia on the basis of the visual transmission time.

METHODS

Pattern reversal visual evoked potentials and motor reaction times were obtained in nine normal subjects, eight anisometropic and seven strabismic amblyopes.

RESULTS

Our results show that while the peak time of the pattern visual evoked potential in anisometropic amblyopia and strabismic amblyopia was significantly delayed, it could not distinguish the two types of amblyopia. In contrast, a significantly longer interocular increment in strabismics compared to anisometropes was found with the reaction time, but not with the pattern visual evoked potential.

CONCLUSION

Our findings thus show that it is possible to distinguish strabismic from anisometropic amblyopes using interocular differences in reaction time measurements. Our results bring support to the contention that the two types of amblyopia represent different neural abnormalities.

Persistent functional and structural retinal anomalies in newborn rats exposed to hyperoxia

Previous studies have shown that newborn rats exposed postnatally to hyperoxia will develop a permanent impairment of the retinal function as determined with the electroretinogram (ERG). The purpose of our study was to examine whether postnatal hyperoxia equally alters the light- and dark-adapted ERGs and oscillatory potentials (OPs) as well as leads to permanent structural modification of the retina. During the first 14 days of life, cohorts of Sprague-Dawley rats were exposed to a hyperoxic environment, and ERGs were recorded at mean ages of approximately 25 and 55 days. Our results indicate that both light- and dark-adapted ERGs and OPs are already significantly altered within a few days following exposure to hyperoxia. None of the ERG and (or) OP parameters, with the exception of the a-wave, returned to normal values by 55 days of age. In fact some dark-adapted OPs were completely abolished following postnatal O2 exposure. Histological analysis revealed that the retina of rats exposed to hyperoxia failed to develop an outer plexiform layer and had a reduced count of horizontal cells, consistent with the permanent postreceptoral anomalies seen in the ERG responses. Our results suggest that postnatal hyperoxia causes a generalized retinal disorder leading to permanent structural modifications of the retinal cytoarchitecture and lasting anomalies of the rod and cone functions.

Intraocular gene transfer of ciliary neurotrophic factor prevents death and increases responsiveness of rod photoreceptors in the retinal degeneration slow mouse

Several mutations causing both photoreceptor degeneration and malfunction have been identified in humans and animals. Although intraocular injection of trophic factors has been shown to reduce photoreceptor death in a few conditions of rapid photoreceptor loss, it is unclear whether long-term beneficial changes in functional properties of affected photoreceptors can be obtained by treatment with these factors. The rds/rds mouse is a spontaneous mutant bearing a null mutation in the rds/peripherin gene, which is linked to many forms of dominant retinal degenerations in humans. Here, we report that intraocular adenovirus-mediated gene transfer of ciliary neurotrophic factor (CNTF) in this mutant reduces photoreceptor loss, causes a significant increase in the length of photoreceptor segments, and results in a redistribution and an increase in the retinal content of the photopigment rhodopsin. These effects are accompanied by a significant increase in the amplitude of the a- and b-waves of the scotopic electroretinogram. These results suggest that continuous administration of CNTF could potentially be useful for the treatment of some forms of retinal degeneration.

A novel mechanism for vasoconstrictor action of 8-isoprostaglandin F2 alpha on retinal vessels

Using a video-imaging technique, we characterized the effects of 8-isoprostaglandin F2 alpha (8-iso-PGF2 alpha) on retinal vasculature from piglets. 8-Iso-PGF2 alpha potently contracted (EC50 = 5.9 +/- 0.5 nM) retinal vessels. These effects were completely antagonized by the cyclooxygenase inhibitor indomethacin, the thromboxane synthase blocker CGS-12970, the thromboxane receptor antagonist L-670596, and the putative inhibitor of the non-voltage-dependent receptor-operated Ca2+ pathway SKF-96365; constrictor effects of 8-iso-PGF2 alpha were also partly attenuated by the ETA-receptor blocker BQ-123 and an inhibitor of endothelin-converting enzyme, phosphoramidon, but was negligibly affected by the L-type voltage-gated Ca2+ channel blocker nifedipine. Correspondingly, 8-iso-PGF2 alpha elicited endothelin release from retinal preparations, which was markedly reduced by SKF-96365. 8-Iso-PGF2 alpha also increased thromboxane production in the retina and cultured endothelial cells, but not on retinovascular smooth muscle cells; these effects of 8-iso-PGF2 alpha were blocked by indomethacin, CGS-12970, SKF-96365, and EGTA, but not by nifedipine. 8-Iso-PGF2 alpha also increased Ca2+ transients in retinal endothelial cells, which were inhibited by SKF-96365 and EGTA, but not by nifedipine, whereas in smooth muscle cells U-46619, but not 8-iso-PGF2 alpha, stimulated a rise in Ca2+ transients. Finally, H2O2 + FeCl2 (in vitro) and anoxia followed by reoxygenation (in vivo) stimulated formation of 8-iso-PGF2 alpha in the retina. In conclusion, 8-iso-PGF2 alpha-induced retinal vasoconstriction is mediated by cyclooxygenase-generated formation of thromboxane and, to a lesser extent, by endothelin after Ca2+ entry into cells, possibly through receptor-operated channels. Retinal vasoconstriction to 8-isoprostanes might play a role in the genesis of ischemic retinopathies.

Hemiretinal contribution to the timing of the full-field PVEP as determined with the motor reaction time

With the use of the simple motor reaction time (RT), we examined the contribution of the hemiretinas to the timing of the pattern reversal visual evoked potential (PVEP). RTs were obtained to the reversal of a progressively dimmer check falling either on the central, nasal, temporal, superior or inferior retina. Results were correlated with the peak time of the P1 wave evoked to the reversal of the entire checkerboard stimulus to the same luminance conditions as that used to generate the RT measurements. Our results clearly indicate that the time taken to trigger a manual response depends not only on the luminance of the stimulus, but also on where it falls on the retina. Comparing the above data with the peak time of the PVEP obtained to the reversal of the entire screen indicates that at brighter luminance the peak time of the P1 wave correlates with that measured from the most central retina, while at dimmer intensities it correlates with the mathematical average of all the retinal locations tested.

Prevention of postasphyxia electroretinal dysfunction with a pyridoxal hydrazone

The newborn retina is particularly sensitive and frequently subjected to peroxidative stresses that result in visual sequelae. We compared two iron chelators, deferoxamine and a newer compound, pyridoxal isonicotinoyl hydrazone (PIH), in protecting the retina of newborn pigs (1-3 d old) from asphyxia-reoxygenation insults. Animals were treated IV with either saline, deferoxamine 15.2 mumol/kg (10 mg/kg) or PIH 34.8 mumol/kg (10 mg/kg); n = 10 in each treatment group. Scotopic and photopic electroretinograms (ERG) were recorded before and 40 min after drug treatment as well as 45 min following a 5-min period of asphyxia by interrupting ventilation. In separate animals the indices of peroxidation, malondialdehyde (MDA: TBARS) and hydroperoxides, were measured in retina at the same times. In saline-treated animals, there was a marked increase in MDA and hydroperoxide concentrations in the retina following the asphyxia-reoxygenation period. This was associated with a decrease in the a- (photoreceptor generated) and b-wave (generated by Müller and bipolar cells) amplitudes measured under photopic (cone-mediated response) and scotopic (rod-mediated response) conditions, and an increase in their implicit times. PIH and deferoxamine prevented the postasphyxial increase in MDA and hydroperoxides. However, only PIH prevented the postasphyxial changes in a- and b-wave amplitudes and implicit times, whereas deferoxamine markedly altered the preasphyxial ERG and provided only partial postasphyxial protection simply to the retinal outer segment. Our findings indicate that the iron chelator PIH effectively inhibits peroxidation and retinal electrophysiological alterations secondary to asphyxia-reoxygenation-induced oxidative stresses to newborn animals, whereas deferoxamine adversely affects retinal function; hence, PIH may be a preferred alternative to deferoxamine.

Light induces peroxidation in retina by activating prostaglandin G/H synthase

Prostaglandin G/H synthase (PGHS) has been shown to generate peroxides to a significant extent in the retina and absorbs light at the lower end of the visible spectrum. We postulated that PGHS could be an important initial source of peroxidation in the retina exposed to light, which would in turn alter retinal function. Exposure of pig eyes (in vivo) to light (350 fc/3770 lx) caused after 3 h a 50% increase and by 5 h a 30% decrease in a- and b-wave amplitudes of the electroretinogram (ERG) which were comparable at 380-650 nm and 380-440 nm but were not observed at wavelengths > 450 nm. These effects of light were prevented by free radical scavengers (dimethylthiourea and high-dose allopurinol) and PGHS inhibitors (naproxen and diclofenac), but stable analogs of prostaglandins did not affect the ERG. Both increases and subsequent decreases in ERG wave amplitudes following light exposure in vivo were associated with increases in retinal prostaglandin and malondialdehyde (peroxidation product) levels, which were inhibited by the nonselective PGHS blockers, naproxen and diclofenac. Similar observations were made in vitro on isolated porcine eyecups as well as on retinal membranes exposed to light (250 fc/ 2700 lx) 380-650 nm and 380-440 nm but not at > 500 nm. Both PGHS-1 and PGHS-2 contributed equivalently to light-induced prostaglandin synthesis, as shown after selective PGHS-2 blockers, but mRNA expression of PGHS-1 and 2 was not affected by light. Finally, light stimulated activities of pure PGHS-1 and PGHS-2 isozymes, and these were also shown to produce superoxide radical (detected with fluorogenic spin trap, proxyl fluorescamine). Taken together, data suggest that PGHS- (1 and 2) is activated by short wavelength visible light, and in the retina is an important source of reactive oxygen species which in turn alter retinal electrophysiological function. PGHS thus seems a likely chromophore in setting forth photic-induced retinal injury. Findings provide an explanation for increased sensitivity of the retina to visible light predominantly at the far blue range of its spectrum.

The effect of in vivo retinal cooling on the electroretinogram of the rabbit

PURPOSE

Studies reporting the effect of retinal cooling on the electroretinogram of mammals have, in most cases, made use of an in vitro approach where the temperature of the retina was lowered by reducing the temperature of the bathing media. The purpose of our study was to investigate, in rabbits, the effect of in vivo retinal cooling using an experimental approach never before reported in order to determine if some of the temperature-related ERG effects previously shown could have been, in part, amplified by alterations in the physiological status of the retina due to preparation for in vitro study.

METHODS

In order to reduce the temperature of the retina, a 20 gauge plastic tubing was coiled around the eye from the limbus to the optic nerve head and glued to the sclera. Cold (15 degrees C) tap water entered the tubing at the limbal extremity and exited (18 degrees C) at the optic nerve extremity. Intraretinal temperature was measured to be within 1 degrees C of that of the circulating water.

RESULTS

Our results indicate that with progressive retinal cooling the a- and b-waves are gradually reduced to 66.9 +/- 17.3 and 90.9 +/- 10.4% of control respectively. The most dramatic temperature induced ERG modifications were observed in the oscillatory potential recordings where the mean summed OP amplitude (OP2 + OP3 + OP4) was reduced to less than 23.9 +/- 13.5% of control with OP2 being the least affected. The peak times of all the ERG components were significantly delayed with cooling.

CONCLUSIONS

Of all the ERG components examined, the OPs were those most severely affected by our manipulation. We believe that our results constitute further evidence in support of the concept that the OPs are more sensitive to retinal disturbance than the b-wave.

Reproducibility of electroretinograms recorded with DTL electrodes

The purpose of this study was to examine whether the use of the DTL fiber electrode yields stable and reproducible electroretinographic recordings. To do so, luminance response function, derived from dark-adapted electroretinograms, was obtained from both eyes of 10 normal subjects at two recording sessions spaced by 7-14 days. The data thus generated was used to calculate Naka-Rushton Vmax and k parameters and values obtained at the two recording sessions were compared. Our results showed that there was no significant difference in the values of Vmax and k calculated from the data generated at the two recording sessions. The above clearly demonstrate that the use of the DTL fiber electrode does not jeopardize, in any way, the stability and reproducibility of ERG responses.

Light adaptation of the human photopic oscillatory potentials: influence of the length of the dark adaptation period

Photopic electroretinograms recorded immediately after a period of dark adaptation show a regular increase in amplitude with time spent in light. The retinal mechanisms at the origin of this light adaptation effect remain obscure. The purpose of this study was to investigate the duration of the dark adaptation period needed to produce an optimal light adaptation effect as demonstrated by photopic oscillatory potential recordings. Our results indicate that the light adaptation effect can be separated into two distinct processes. The first one, activated early in the dark adaptation process, reduces the amplitude of the fourth oscillatory potential to 32% of control after less than 5 min of dark adaptation, while the second process, activated after more than 10 min of dark adaptation, appears to impact solely the amplitude of the earlier oscillatory potentials 2 and 3. Our results suggest that the light adaptation effect is mediated by two distinct retinal pathways or mechanisms.

Monocular contribution to the peak time of the binocular pattern visual evoked potential

The contribution of each monocular pathway to the timing of the binocular pattern visual evoked potential was assessed in situations where a significant interocular timing discrepancy was observed. Monocular and binocular pattern visual evoked potentials to 0.5 degree checks were recorded from normal subjects, normal subjects in whom one eye was blurred, patients with monocular amblyopia, and patients with resolved unilateral optic neuritis. Normal subjects showed facilitation, while suppression was evidenced in subjects with monocular blurring. In patients with amblyopia, the affected pathway had no effect on binocular pattern visual evoked potential latency, suggesting that the amblyopic eye was suppressed. In contrast, all patients with optic neuritis showed binocular averaging. Our results show that different forms of binocular interaction are evidenced in normal subjects, in amblyopia and in optic neuritis, and suggest that a comparative analysis of monocular and binocular pattern visual evoked potential peak times brings valuable information to the clinical evaluation that could be used to distinguish disease processes further.

Lateral geniculate neuron responses to drifting sine-wave gratings in rabbits

Neurons of the lateral geniculate body in rabbits were excited with drifting sine-wave gratings. Rabbits were anesthetized and paralyzed under conventional methods to record action potentials of single cells using tungsten in glass microelectrodes. All classes of geniculate cells responded in a modulatory pattern. It appears that the unmodulatory pattern typical of complex cell types of the cortex is extremely infrequent or absent. In the spatial domain most cells are low pass and bandpass. Only one unit was high pass. In the temporal domain low-pass and bandpass cells were the most frequently recorded. Four geniculate cells were high pass. It appears, therefore, that neurons of rabbits' geniculate are tuned over spatial and temporal frequencies of sine-wave gratings. The comparison with cortical recordings revealed that geniculate cells are more broadly tuned than cortical neurons. This study suggests that the rabbit's visual system is sensitive to gratings. However cells respond optimally to lower values, e.g., broader gratings, than neurons of frontalized eye animals.

Alterations in the electroretinogram of newborn piglets by propionic acid-derivative nonsteroidal antiinflammatory drugs but not by indomethacin and diclofenac

Different nonsteroidal antiinflammatory drugs (NSAID), especially ibuprofen, are being considered as an alternative to indomethacin for use in the newborn and as antipyretics for infants. However, some of these NSAID have been shown to cause visual complications. We therefore studied the effects of different NSAID indomethacin 19.6 mumol/kg (7 mg/kg), diclofenac 15.7 mumol/kg (5 mg/kg), ibuprofen 48 and 194 mumol/kg (10 and 40 mg/kg), naproxen 79 mumol/kg (20 mg/kg), and flurbiprofen 41 mumol/kg (10 mg/kg) on photopic and scotopic electroretinograms (ERG) and retinal prostaglandin E2, prostaglandin F2 alpha, and 6-keto-prostaglandin F1 alpha levels in piglets 1-5 d old. All NSAID decreased retinal prostaglandin levels, but their effects on the ERG were not identical. Indomethacin and diclofenac did not alter the ERG. In contrast, the propionic acid derivatives ibuprofen (the two doses used), naproxen, and flurbiprofen affected the amplitude as well as the implicit time of the ERG under photopic and scotopic conditions. These changes are suggestive of generalized alterations in the function of rods and cones. Prior inhibition of prostaglandin synthesis by indomethacin did not modify the effects of ibuprofen on the ERG. These findings thus show a dissociation between the effects of NSAID on the ERG and prostaglandin synthesis. Because ERG changes are associated with visual alterations, these effects of propionic acid derivatives should be taken into account before considering their use in infants.

Human strabismus: evaluation of the interhemispheric transmission time and hemiretinal differences using a reaction time task

Experimentally induced strabismus in visually immature cats leads to abnormal development of the posterior corpus callosum. This, in turn, should lead to abnormal interhemispheric integration of unilaterally presented visual information. To test whether strabismus produces deficits in the human commissural visual system, the interhemispheric transmission time (ITT) was compared in strabismic and normal subjects. Simple unimanual reaction times (RT) were tested in 30 subjects in response to a lateralized target presented monocularly at 4 degrees and 35 degrees nasally and temporally from the fovea along the horizontal meridian. This method was also used to examine the effect of strabismus on the central and peripheral portions of each hemiretina. The results showed that in strabismic subjects with or without amblyopia, the ITT did not differ significantly from normals at both eccentricities. In non-amblyopic strabismic patients, RTs in the central and peripheral portions of hemiretina were comparable to normals. However, a reduced speed of response was found in the central visual field (4 degrees) in the amblyopic eye. Our results suggest that the ITT is normal in strabismic subjects and that the longer RTs in the central portion of the nasal and temporal hemiretina of the amblyopic eye may be associated with the severe amblyopic condition.

The human suprathreshold photopic oscillatory potentials: method of analysis and clinical application

The oscillatory potentials (OPs) of the electroretinogram (ERG) identify the high frequency wavelets which are seen riding on the top of the ascending limb of the b-wave. Despite the refinement in ERG recording techniques, which now allow for a more selective amplification of the OPs, their clinical utility still remains somewhat limited. Furthermore, while it has long been recognized that the different OPs which builds a given response are generated by distinct retinal events, amplitudes of clinically recorded oscillatory potentials are still reported using the artificial variable 'sum of OPs (SOPs)' which represent the sum amplitude of all the OPs identified in a given response; a method which is prone to compromise the diagnostic possibilities of the oscillatory potential response. The purpose of this report is to present a method of analysis of the OP response which is based on the relative amplitude of each OP. For the sake of clarity this report will focus on the analysis of the suprathreshold photopic OP response evoked the flash for 8.9 cd-m-2-sec in energy delivered against a rod desensitizing background of 30 cd-m-2 in luminance. The amplitude of each of the three major OPs (ie: OP2, OP3 and OP4) that normally compose this response was reported in relative units (ie: OP chi/SOPs) a method shown to minimize intersubject variability and at the same time allow for each OP to be evaluated individually. With the above method I have reviewed 289 clinical OP responses collected during a 7 year interval and identified more than 10 different categories of responses of which 6 were shown to demonstrate specific anomalies for one or two OPs. The selectivity and reproducibility of my observation were confirmed on follow-up testings as well as in pedigree studies. Use of this method of OP analysis should significantly increase the clinical utility of the oscillatory potentials and also facilitate comparisons between clinical laboratories.

Interpretation of the filtered 100- to 1000-Hz electroretinogram

We examined the possibility that the 100- to 1000-Hz oscillatory potentials could represent the derivative version of the 1-1000-Hz electroretinogram. Corneal electroretinograms were recorded from rabbits by means of bandwidths of 1-1000 Hz, 10-1000 Hz, 30-1000 Hz and 100-1000 Hz (6 dB of attenuation). Derivatives of the 1- to 1000-Hz electroretinogram had a waveform similar to the 100- to 1000-Hz signals, but of larger amplitude (21.9% +/- 16.7% larger, n = 16). Similarly, integration of the 100- to 1000-Hz signal resulted in a waveform whose amplitude was 60% of the original 1- to 1000-Hz electroretinogram. Our results suggest that some aspect of the morphologic changes seen when the low-frequency cutoff of the recording bandwidth of the ERG is increased from 1 Hz to 100 Hz could be explained with a simple derivative model. The oscillatory potentials may be significant contributors to the morphogenesis of the 1- to 1000-Hz electroretinogram.

Recording the oscillatory potentials of the electroretinogram with the DTL electrode

Suprathreshold photopic oscillatory potentials recorded with a DTL electrode were compared to those obtained with a Lovac corneal electrode. The overall oscillatory potential response (sum of oscillatory potentials) recorded with the DTL electrode was half of that obtained with the Lovac electrode. However, there was no evidence of a selective attenuation (or amplification) of any given oscillatory potential with the DTL electrode. Similarly, the oscillatory potential relative amplitude ratios and the peak times of the oscillatory potentials were identical for both electrodes. Our findings clearly indicate that the DTL electrode is adequate to record the high-frequency oscillatory potentials. Given the low cost and ease of use, as well as the disposable nature of the DTL electrode, we believe that electroretinographic specialists should seriously consider a wider utilization.

Influence of remote targets on directionality of striate neurons in rabbits

The described investigations study the influence of additional targets located well outside the classical receptive field on responses to motion of cortical cells in rabbits. Animals are anesthetized and prepared for acute single cells recordings in a conventional manner. The interactions between remote targets and central stimuli are abolished with microinjections of lidocaine hydrochloride or GABA at the site excited by remote stimuli. Results show that responses to motion of cortical cells are particularly sensitive to these manipulations. Although supplementary targets fail to influence spontaneous activity of all cells, they do influence responses to motion. Overall, the directionality indices (DI) declined. (53 to 45.) This decline may express itself either by a decrease of responses in the preferred direction or an enhancement of responses in the non-preferred direction or both. By contrast, responses to stationary stimuli are unaffected by additional targets in the visual field. Globally, cells whose directionality index was superior to 50% were significantly more affected then cells whose DI was less than 50%. This result suggests that similarly to cats, the directionality of cells in the striate cortex rests on a very fragile convergence of excitatory and inhibitory influences.

Comparative effects of luminance and scatter on the pattern visual evoked potential and eye-hand reaction time

We investigated the effect of reduced luminance and increased scatter on the pattern visual evoked potential and eye-hand reaction time evoked to a check size of 0.5 degrees in 10 normal subjects. Data analysis indicated that a reduction in luminance as well as an increase in scatter caused a statistically significant increase in the peak time of the pattern visual evoked potential P100 wave. The reaction time, however, was not significantly affected by the initial 0.9-log unit attenuation of the stimulus luminance or the 0.3 scatter filter. Further attenuation of luminance or increase of scatter also yielded statistically significant increases. Our results suggest that the reaction time is less affected by a reduction in luminance or an increase in scatter of a 0.5 degrees stimulus than the pattern visual evoked potential is and therefore represents a more reliable test to assess visual function, especially in the presence of medial opacities, which are known to reduce luminance and produce scatter.

Evidence for an intensity-coding oscillatory potential in the human electroretinogram

Electroretinograms (ERGs) were obtained from normal subjects to various combinations of stimulus and background intensities to examine if, as shown in rabbits, one of the oscillatory potentials (OP2) could be identified as reflecting the absolute intensity of the stimulus (i.e. the intensity irrespective of the state of retinal adaptation). The results presented clearly demonstrate that the peak time and the amplitude of OP2 are highly correlated with the absolute intensity of the stimulus. Linear regression analysis (for light- and dark-adapted data combined) yields a correlation coefficient of r = 0.98 for the peak time of OP2 and r = 0.88 for the amplitude of OP2. The latter equations were obtained from ERGs where the brightest flashes were delivered in light-adaptation and the dimmest in dark-adaptation. This is the first demonstration of a high correlation between the intensity of the stimulus and the amplitude and peak time of a human ERG component which is shown to be linear irrespective of the state of retinal adaptation.

The effect of a slow flicker on the human photopic oscillatory potentials

Flicker-induced modifications of the human photopic oscillatory potentials (OPs) were investigated with the use of flash stimuli of 0.89 and 8.9 cd m-2 sec in strength. When the dimmest stimulus is used, increasing the rate of presentation from 0.5 to 20 Hz augments the amplitude and peak time of OP2. For a brighter stimulus, the 10 Hz flicker significantly reduces the amplitude of OP2 and OP3, increases the peak time of OP2 and reduces that of OP4. The 20 Hz flicker increases the peak time and reduces the amplitude of OP2 and completely abolishes OP3, while it has no significant impact on the amplitude and timing of OP4. The data presented support the claim that each OP making the photopic response represent independent electrophysiological entities.

Temporal relationship between the ERG and geniculate unit activity in rabbit: influence of background luminance

The comparative analysis of the retinal and lateral geniculate nucleus (LGN) intensity-response function revealed that the timing of the LGN unit response was highly correlated to that of one oscillatory potential (OP2). To examine if this OP2-LGN intensity-response function was retained irrespective of the state of retinal adaptation, we performed simultaneous recordings to the ERG and single-cell unit activity at the geniculate level evoked to a flash of constant energy while the level of background luminence was varied. For a stimulus of 6.7 cd m-2 sec, the shifts in latency induced by increasing the background luminance from 0 to 125 cd m-2 are of 1.17 +/- 0.61 msec for OP2, a value almost identical (P greater than 0.10; n = 36) to the one obtained at the geniculate level (1.11 +/- 0.88 msec). However, when a dimmer flash is used, the latency shifts are not so well correlated. The latter could be partly explained by the threshold nature of the resulting stimulus (i.e. high photopic background combined with a dim flash).