Abstract
purpose. To determine whether reflectance changes of the retina after electrical suprachoroidal-transretinal stimulation (STS) can be detected with a newly developed optical imaging fundus camera.
methods. Ten eyes of 10 cats were studied. A small retinal area was focally stimulated with electric currents passing between an active electrode placed in the fenestrated sclera and a reference electrode in the vitreous. Biphasic pulses were applied for 4 seconds with a current up to 500 μA. Images of the fundus illuminated with near-infrared (800–880 nm) light were obtained every 20 msec for 26 seconds between 2 seconds before and 20 seconds after the STS. Twenty images of 20 consecutive experiments were averaged. A two-dimensional map of the reflectance changes was constructed by subtracting the images before the stimulation from those after the stimulation. STS-evoked potentials (EPs) were recorded from the optic chiasma.
results. Approximately 0.5 second after the onset of STS, reflectance changes were observed around the retinal locus, where the stimulating electrodes were positioned. The intensity of the reflectance changes was correlated with the intensity of the stimulus current. The area of the reflectance change increased as the current intensity increased and was correlated with the amplitude of the EPs (R 2 = 0.82).
conclusions. Reflectance changes after STS were localized to the area around the electrode. The strong correlation between the area of the reflectance changes and the amplitude of the EPs suggested that the reflectance changes reflected the activity of retinal neurons elicited by electrical stimulation.
In vivo optical imaging of intrinsic signals is a well-established method to study brain physiology and to map the functional architecture of the cerebral cortex.
1 2 3 4 In optical imaging studies, stimulus-induced neuronal activity is detected as a change of light reflectance. The reflectance change does not directly indicate neural activation, but it is strongly correlated with the activity of neurons examined by conventional extracellular recordings.
5 Studies of cortical optical imaging have shown that this intrinsic signal originates from stimulus-induced changes in the light-scattering of neural tissues and from changes in light absorption associated with hemodynamic changes in blood volume or the oxygenated state of hemoglobin.
6 7
Recently, the technique of optical imaging has been applied to the retina to examine light-evoked reflectance changes.
8 9 10 Although the light-evoked neuronal activity in localized areas of the retina can be recorded by multifocal electroretinography (mfERG),
11 those evoked by electrical stimulation are difficult to detect by mfERG because of the large stimulus artifact. Therefore, optical recording is a reasonable alternative to study the responses evoked by the electrical stimulation of the retina.
Artificial retinas, also called retinal prostheses, have been placed at different sites.
12 13 14 15 A typical retinal prosthesis consists of an array of electrodes implanted above (epi-) or beneath (sub-) the retina and is used to deliver electrical current to the retina to evoke a light sensation called a phosphene. We have implanted an electrode array in the suprachoroidal space, and stimulation by this system has been called suprachoroidal-transretinal stimulation (STS).
16 In STS, the active electrode array is placed in the fenestrated sclera in the retrobulbar space while the reference electrode is inserted into the vitreous cavity. STS is a safe method and avoids the direct contact of electrodes with the retina, but the distance between electrodes and retina is larger than in the epiretinal or subretinal methods. Thus, it is necessary to determine whether sufficient spatial resolution can be achieved by this approach.
The spatial resolution of epiretinal electrodes has been investigated by optical imaging in the visual cortex
17 but not in the retina. The purpose of this study was to determine whether reflectance changes can be detected in the area of the retina activated by STS. To accomplish this, we have developed a prototype optical-imaging fundus camera to measure the changes of light reflectance evoked by electrical stimulation of the retina.
Materials and Methods
Animals
Ten left eyes of 10 cats under general anesthesia were used. Cat 1 was used for the study of light stimulation, and cats 2 through 10 were used for the electrical stimulation. Cats 6, 7, and 9 were also used for the study of evoked potential (EP) at optic chiasm. Initially, each animal received an intramuscular injection of ketamine hydrochloride (25 mg/kg) and an intraperitoneal injection of atropine sulfate (0.1 mg/kg). Then each animal was anesthetized by intravenous infusion of pentobarbital sodium (1 mg/kg per hour) and paralyzed by pancuronium bromide (0.2 mg/kg per hour) mixed with Ringer solution and glucose (0.1 g/kg per hour).
The animal was artificially ventilated with a mixture of N2O/O2 (1:1), and the end-tidal CO2 concentration was controlled at 3.5% to 5.0% by altering the frequency and volume of ventilation. In addition to the continuous monitoring of the expired CO2, the intratracheal pressure and electrocardiogram were also monitored. Body temperature was maintained with a heating pad at 38°C.
Surgical Procedures
An incision was made over the temporal area of the left eye; part of the lateral orbital wall (zygomatic and frontal bone) was removed, and the lateral rectus muscle was dissected. The scleral area just above the long ciliary artery was exposed, and a 3- to 4-mm incision was made through the sclera. The conjunctiva was sutured to a fixed frame attached to the stereotaxic headholder to prevent eye movements. The pupil was dilated with 5% phenylephrine hydrochloride, 0.5% tropicamide, and 1% atropine sulfate. To protect the corneal surface, a hard contact lens was placed on the cornea (polymethylmethacrylate; base curve, 8.50 mm; power, +1.5 D; diameter, 13.5 mm). All experiments were performed in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Visual Research, and the procedures were approved by the Animal Research Committee of Osaka University Medical School.
Optical Imaging of Retina
The ocular fundus was monitored by a fundus camera (TRC-50LX; Topcon Corp., Tokyo, Japan) with a digital CCD camera (C8484; Hamamatsu Photonics, Hamamatsu, Japan). The number of pixels was 1280 × 1024, but the binning mode of the camera was used to obtain maximum light sensitivity and the resolution was reduced to 320 × 256 pixels (12-bit grayscale). A 12-bit digitizer was used, and the 4096 grayscale levels were obtained for each pixel.
A halogen lamp was used to illuminate the posterior fundus, and a bandpass filter was inserted in the illumination optical path to limit the wavelength of the fundus monitoring light between 800 and 880 nm. The power of the illuminating light was 250 nW, which was much lower than the safe exposure limit decided by American National Standard Institute.
To improve the signal-to-noise ratio, 20 images of 20 consecutive experiments were averaged
(Fig. 1) . The interval between each session was 1 minute. A two-dimensional image of the optical signal was obtained by subtracting the image recorded before stimulation from those after stimulation. All experiments were performed in a dark room after 30 minutes of dark adaptation.
Electrophysiological Recording from Optic Chiasma
To record the potentials evoked by electrical stimulation of the retina from the optic chiasma (OX), a pair of stainless steel electrodes was placed in the OX stereotaxically. Light-evoked responses were recorded from each electrode to be certain that the electrodes were placed in the OX.
To record the electrically evoked potentials (EEPs), the signal was amplified 10,000 times and was bandpass filtered between 300 Hz and 5 kHz with an AC amplifier (model 1800; Microelectrode AC amplifier; A-M Systems, Inc., Carlsborg, WA) and a signal conditioner (LPF-202A; Warner Instruments, Hamden, CT). Amplified EEPs were fed to a signal processor (Power 1401; Cambridge Electronic Design, Cambridge, UK) with a sampling frequency of 50 kHz and were analyzed off-line. Signals were also monitored on an oscilloscope and an audio speaker in real time.
Focal Light Stimulation of Retina
The stimuli were obtained from white light-emitting diodes controlled by a pulse generator. The stimulating light was a vertical bar focused on the retina, flickering at 8 Hz for 4 seconds. The width of the bar was 4°, and the center of the bar was located 6° temporal to the fovea
(Fig. 2A) . The light power was 30 nW. Images were obtained every 20 msec for 18 seconds between 2 seconds before and 12 seconds after the stimulus.
Focal Electrical Stimulation of Retina
A single-channel platinum electrode (diameter, 100 μm) was used as the active scleral electrode. A vitreous electrode was inserted into the vitreous through the sclera at the pars plana. This electrode was a urethane-coated platinum wire (0.2 mm in diameter), and its exposed tip measured approximately 2 mm. The active electrode was held gently against the sclera by a manipulator, and the pressure on the eye was minimal, as determined by the degree of indentation of the retina viewed by the fundus camera. The mass EP at the OX elicited by STS was also monitored to confirm the effectiveness of the electrical stimulation.
For each STS trial, biphasic pulse trains (outward first) were applied at 50-Hz frequency, 0.5-msec pulse duration, and 4-second stimulation
(Fig. 2B) . All pulses were generated by a pulse generator (SEN7203; Nihon Kohden Corp., Tokyo, Japan) and were delivered to the electrodes through a linear stimulus isolation unit (BSI-950; Dagan Corporation, Minneapolis, MN).
To evaluate the correlation between the stimulus current and the area of reflectance change, the stimulating current was increased systematically between the threshold current to the maximum current (≤500 mA) or decreased systematically to examine hysteresis. Images were obtained every 20 msec for 26 seconds between 2 seconds before and 20 seconds after the electrical stimulation.
Data Analyses
Optical Density Measurements.
To evaluate the intensity of the reflectance, the grayscale value (GSV) of each spot in the focused area was averaged. The averaged GSV of each spot after the onset of light or electrical stimulation was subtracted from that before the electrical stimulation to obtain the differential image
(Fig. 1) . For the evaluation of the area of reflectance change, the GSV value of 40 (approximately 8 SD of the GSV without stimulation) was set as the cutoff level to reduce the effect of baseline fluctuations. To study the relationship between the area of reflectance change (pixels) and the stimulus current, the maximum number of pixels in which the averaged reflectance change exceeded the cutoff level (±40 GSV) during the time course was selected to determine the maximum area of reflectance change. The number of pixels with an increase or a decrease of reflectance was added to evaluate the area of retinal excitation.
Electrophysiological Measurements.
The amplitude of the EPs evoked by STS was determined by measuring the amplitude between the first negative peak (latency approximately 3.0 msec) and the second positive peak (latency approximately 4.0 msec). Two hundred records were averaged to determine the amplitude for a particular stimulus current.
Statistical Analysis
Regression analysis between the stimulus current and the area of reflectance change or the amplitude of EP at OX was evaluated by SPSS (SPSS Inc., Chicago, IL).
Results
Optical Imaging after Light Stimulation
Examination of a two-dimensional map of the reflectance changes after light stimulation showed that the reflectance decreased in the stimulated striped area
(Fig. 3) . Reflectance of the retinal vessels began to decrease approximately 0.5 second after the stimulus onset and continued to decrease linearly to a deep trough at approximately 3.5 seconds
(Fig. 3C) . The change of reflectance of the optic disc and stimulated retinal area showed similar temporal changes. Reflectance began to decrease approximately 1.0 second after the stimulus onset and continued to decrease almost linearly to a deep trough at approximately 7.0 seconds after the stimulus onset
(Fig. 3D) .
Optical Imaging of STS
A two-dimensional map of the reflectance changes after electrical stimulation showed an increase of reflectance in the retinal area, where the tip of the electrode was attached to the sclera. A decrease of reflectance was observed in the retinal area surrounding the area of increased reflectance (
Fig. 4 ; see also
5 6 Fig. 7A ). The time course of the reflectance changes was similar with the different intensities of electrical stimulation. The increase at the electrode or the decrease surrounding the electrode of light reflectance was observed approximately 0.5 second after the stimulus onset, increased or decreased rapidly for 1.5 to 2.0 seconds, increased or decreased gradually to peak at 4 to 6 seconds after stimulus onset, and then returned to the baseline gradually
(Figs. 5A 6A) . Reflectance did not change at the electrode site when stimulation was not applied
(Fig. 5A) . The maximum value of reflectance increase on the electrode or reflectance decrease around the electrode was linearly related to the increase of stimulus intensity
(Fig. 6B) .
The area of the reflectance change (increased and decreased reflectance areas were combined) had a similar time course with different stimulus intensities. The area began to increase approximately 0.5 second to 1.0 second after stimulus onset and changed linearly for 1.0 to 1.5 seconds The size of the area peaked 2 to 5 seconds after stimulus onset, was sustained for 4 to 8 seconds, and returned to the baseline gradually
(Figs. 5B 7B) . The area of the reflectance changes increased with an increase of current intensity
(Fig. 7C) . Reflectance changes on and around the electrode were not observed when stimulation was not applied
(Fig. 5B) .
The threshold current for eliciting a reflectance change ranged from 65 μA to 200 μA for the different cats. Linear regression analysis showed that
R 2 ranged from 0.82 to 0.97 in the 9 cats (cats 2–10), suggesting that the area of reflectance changed linearly with the stimulus current intensity. The slope of the regression line varied in the different cats
(Fig. 8) .
Electrophysiological Recordings from Optic Chiasm after STS
We examined the relationship between the EP amplitude recorded in the OX and the stimulus current in three cats (cats 6, 7, 9;
Fig. 9 ). EP amplitude increased linearly with an increase of stimulus current
(Fig. 10) . Linear regression analysis showed the
R 2 values were 0.89, 0.95, and 0.98, respectively, suggesting that the EP changed linearly with the stimulus current intensity. The slope of the regression line varied in the different cats (data are not shown).
The relationship between the amplitude of EP and the area of reflectance change was also assessed in cats 6, 7, and 9. Linear regression analysis showed that the
R 2 values were 0.82, 0.84, and 0.82, respectively, suggesting that the reflectance changed linearly with the EP at OX through electrical stimulation to the retina
(Fig. 10) .
Discussion
We have developed an optical imaging system to evaluate the retinal area activated by focal electrical stimulation to the retina. To confirm that our system measured the retinal area in which the retina was activated, we examined whether the reflectance of infrared light changed in response to flickering visible light stimulation. Our results showed that a decrease in reflectance in the area corresponded with the light-stimulated striped area
(Fig. 3) , as reported by Ts’o et al. for light stimulation (Ts’o D, et al.
IOVS 2004;45:ARVO E-Abstract 3495). Tsunoda et al.
8 also reported decreased light reflectance that peaked within 1 second of light stimulation, faster than our observation (peak approximately 7 seconds;
Fig. 3 ). This discrepancy might have occurred because we used continuous flickering light whereas Tsunoda et al.
8 used single-flashed light. Indeed, the time course of the reflectance change in our case was similar to that of Ts’o D, et al. (
IOVS 2004;45:ARVO E-abstract 3495) who used flickering light.
With focal electrical stimulation, the reflectance increased or decreased depending on the retinal area
(Fig. 4) . Reflectance changes induced by electrical stimulation had fast and slow phases, whereas those induced by light stimulation were monotonic, suggesting that the time course to activate the retina that induced the reflectance changes was different by light stimulation than it was by electrical stimulation
(Figs. 3 4 5 6) .
The origin of the positive change of reflection is a matter of discussion, but, because the time course of positive and negative reflection changes were similar, we suggest that both positive and negative reflection changes were evoked by the same mechanism related to the retinal activation
(Figs. 5 6) . Therefore, we added the absolute value of positive and negative reflectance change as a parameter of reflectance changes.
The change in reflectance intensity was linearly related to the intensity of the electrical current
(Fig. 6) . If we consider that the change in reflectance was correlated with the local retinal response,
8 the value of the reflectance change might have reflected the degree of retinal activation induced by the electrical current. The area of the reflectance change was correlated with the simulating current, suggesting that the area of reflectance change represented the area of the retina activated by electrical stimulation
(Fig. 8) . This supports the results of electrophysiological studies showing that the area of retinal excitation was correlated with the intensity of electrical current (Kanda H, et al.
IOVS 2005;46:ARVO E-Abstract 1499).
In the electrophysiological data from the OX, the maximum amplitude of the EP was strongly correlated with the intensity of stimulus current
(Fig. 10) , suggesting that focal electrical stimulation in the retina activated retinal ganglion cells (RGCs) in proportion to the intensity of stimulation current. EP amplitude was correlated with the total pixels of reflectance change in the retina, suggesting that the area of reflectance change represented neuronal changes in the retina, which led to the excitation of the RGCs.
The threshold current of retinal activation was different for each cat. The reason for this might have been related to the degree of electrode connection to the sclera, which could have been different in each cat. Because the area of reflectance change was correlated with the stimulating current, we may apply this optical imaging system to the artificial retina in humans.
The area of reflectance change is confined to the area around the stimulating electrode with threshold currents
(Fig. 7A) . Thus, the resolution of the electrode and the optimum parameter for the stimulation of each electrode can be determined objectively.
Supported by Health Sciences Research Grants (H16-sensory-001) from the Ministry of Health, Labor and Welfare, Japan and by Grant 18591918 from the Ministry of Education, Culture, Science and Technology.
Submitted for publication February 18, 2007; revised May 21, 2007; accepted August 10, 2007.
Disclosure:
Y. Okawa, None;
T. Fujikado, None;
T. Miyoshi, None;
H. Sawai, None;
S. Kusaka, None;
T. Mihashi, Topcon Research Institute (E);
Y. Hirohara, Topcon Research Institute (E);
Y. Tano, None
The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “
advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.
Corresponding author: Takashi Fujikado, Department of Applied Visual Science, Osaka University Graduate School of Medicine, 2–2 Yamadaoka, Suita, Osaka 565-0871, Japan;
fujikado@ophthal.med.osaka-u.ac.jp.
Ts’oDY, FrostigRD, LiekeEE, GrinvaldA. Functional organization of primate visual cortex revealed by high resolution optical imaging.
Science. 1990;249:417–420.
[CrossRef] [PubMed]FrostigRD, LiekeEE, Ts’oDY, GrinvaldA. Cortical functional architecture and local coupling between neuronal activity and the microcirculation revealed by in vivo high-resolution optical imaging of intrinsic signals.
Proc Natl Acad Sci USA. 1990;87:6082–6086.
[CrossRef] [PubMed]TsunodaK, YamaneY, NishizakiM, TanifujiM. Complex objects are represented in macaque inferotemporal cortex by the combination of feature columns.
Nat Neurosci. 2001;4:832–838.
[CrossRef] [PubMed]TagaG, AsakawaK, MakiA, KonishiY, KoizumiH. Brain imaging in awake infants by near-infrared optical tomography.
Proc Natl Acad Sci USA. 2003;100:10722–70727.
[CrossRef] [PubMed]DasA, GilbertCD. Long-range horizontal connections and their role in cortical reorganization revealed by optical recording of cat primary visual cortex.
Nature. 1995;375:780–784.
[CrossRef] [PubMed]CohenLB. Changes in neuron structure during action potential propagation and synaptic transmission.
Physiol Rev. 1973;53:373–418.
[PubMed]HolthoffK, WritteOW. Intrinsic optical signals in rat neocortical slices measured with near-infrared dark-field microscopy reveal changes in extracellular space.
J Neurosci. 1996;16:2740–2749.
[PubMed]TsunodaK, OguchiY, HanazonoG, TanifujiM. Mapping cone- and rod- induced retinal responsiveness in macaque retina by optical imaging.
Invest Ophthalmol Vis Sci. 2004;45:3820–3826.
[CrossRef] [PubMed]RivaCE, LogeanE, FalsiniB. Visually evoked hemodynamical response and assessment of neurovascular coupling in the optic nerve and retina.
Prog Retin Eye Res. 2005;24:183–215.
[CrossRef] [PubMed]AbramoffMD, KwonYH, Ts’oD, et al. Visual stimulus-induced changes in human near-infrared fundus reflectance.
Invest Ophthalmol Vis Sci. 2006;47:715–721.
[CrossRef] [PubMed]SutterEE, TranD. The field topography of ERG components in man, I: the photopic luminance response.
Vision Res. 1992;32:433–446.
[CrossRef] [PubMed]ZrennerE. Will retinal implants restore vision?.
Science. 2002;295:1022–1025.
[CrossRef] [PubMed]HumayunMS, WeilandJD, FujiiGY, et al. Visual perception in a blind subject with a chronic microelectronic retinal prosthesis.
Vision Res. 2003;43:2573–2581.
[CrossRef] [PubMed]ChowAY, ChowVY, PackoKH, PollackJS, PeymanGA, SchuchardR. The artificial silicon retina microchip for the treatment of vision loss from retinitis pigmentosa.
Arch Ophthalmol. 2004;122:460–469.
[CrossRef] [PubMed]RizzoJF, 3rd, WyattJ, LoewensteinJ, KellyS, ShireD. Perceptual efficacy of electrical stimulation of human retina with a microelectrode array during short-term surgical trials.
Invest Ophthalmol Vis Sci. 2003;44:5362–5369.
[CrossRef] [PubMed]KandaH, MorimotoT, FujikadoT, TanoY, FukudaY, SawaiH. Electrophysiological studies on the feasibility of suprachoroidal-transretinal stimulation for artificial vision in normal and RCS Rat.
Invest Ophthalmol Vis Sci. 2004;45:560–566.
[CrossRef] [PubMed]EckhornR, WilmsM, SchanzeT, et al. Visual resolution with retinal implants estimated from recordings in cat visual cortex.
Vision Res. 2006;46:2675–2690.
[CrossRef] [PubMed]