Abstract
A
prototype electronic retinal prosthesis has been tested in six blind subjects. The
system features an implanted retinal stimulator and an external system for
image acquisition, processing, and telemetry.
Perceptual thresholds were as follows: S1: 35-1121 uA; S2: 16 – 777 uA; S3: 18 – 412 uA; S4: 20 – 385 uA; S5: 10 – 31 uA; S6: 6 – 41 uA. The subjects in general
performed better than chance on simple visual tasks. Neurophysiologic testing
showed pupil response to stimulation and localized brain activity.
1.
INTRODUCTION
An electronic retinal
prosthesis has been proposed as a means to treat retinal degenerative diseases
such as retinitis pigmentosa (RP) and age-related macular degeneration (AMD). RP and AMD are two leading
causes of blindness.[1, 2] The disease primarily attacks
the photoreceptor cells of the retina, but leaves the other retinal cells
relatively intact. A retinal prosthesis that can electrically activate these
remaining cells has shown increasing feasibility through animal and human
trials.[3-5] The
specific studies reported below involves the first clinical trial of a chronically
implanted epiretinal prosthesis.
We report here data examining perceptual threshold, visual task performance,
and objective measures of visual system response to electrical stimulation.
2. METHODS
This study was conducted under an
Investigational Device Exemption granted to Second Sight Medical Products, Inc.
by the Food and Drug Agency. The
Six subjects who
met study criteria were implanted with a Second Sight intraocular epiretinal
prosthesis in the eye with worse vision. Implants
consist of an extraocular microelectronic device and an intraocular electrode
array, connected by a multiwire cable. The electrode array is a 4x4 grid of
platinum electrodes embedded in silicone rubber. The electrodes were either 520
or 260 mm in diameter. The electrode array is held
to the retina with a small retinal tack. The extraocular microelectronic device
is controlled wirelessly by a wearable camera/video processing unit (VPU). The
VPU commands stimulus current, which is generated by the microelectronics and
delivered to the electrodes on the retina. Electrical stimulation was begun
between 7 and 15 days post-operative.
Pupillography was performed on five subjects.
All five patients were determined to be without a clinically apparent papillary
light reflex (PLR) prior to implantation of the prosthesis. Evaluation of the
consensual PLR was performed in a dark room with an infrared video camera
monitoring pupil diameter at 29.97 frames/sec. Individual frames were analyzed
using video and image editing software. Various electrical stimuli parameters
were selected for each patient to allow for characterization of pupil
constriction in terms of intensity, frequency, and duration of electrical
stimuli. Multiple measurements were recorded and averaged for each set of
stimuli parameters.
Multichannel
evoked response testing was performed to localize cortical activity evoked by
electrical stimulation. After patching the subject’s left eye (prosthesis in
right eye), a 64 channel electrode skullcap was placed on her head in a
standard 10-20 setup. Using a 500Hz sampling rate, a gain of 250, and a
continuous data acquisition mode, all 16 electrodes of the prosthesis were
stimulated with biphasic pulses (3ms). Data was recorded for 1140s repeating the
stimuli every second. Data from 6 faulty channels (FP2, C2, F6, FT8, T4, TP10)
resulting from eye movement or interference from electric fields of the
stimulation device were removed. 416 epochs were selected manually and averaged
to reject artifacts. For reconstruction from EEG data, a 3 shell sphere model
and a warped generic realistic head model (
3. RESULTS
The device was successfully implanted in
all subjects. The intraocular stimulating array remained in position to
activate the retina and produce phosphenes in all 6 subjects.
Perceptual thresholds varied both within
and across subjects. S1: 35-1121 uA; S2: 16 – 777 uA; S3: 18 – 412
uA; S4: 20 –385 uA; S5: 10 – 31 uA; S6: 6 – 41 uA. Performance using the head mounted video camera
suggests that patients are capable of interpreting patterned electrical
stimulation. Subjects can localize the position of, or count the number of,
high contrast objects with 74-99% accuracy (3 or 4 Alternative Forced Choice
(AltFC)), and can discriminate simple shapes such as the orientation of a bar
or an “L” (2 or 4AltFC) with 61-80% accuracy. There was a trend towards
subjects performing better when stimulated using meaningful patterns of
electrical stimulation, rather than all electrodes being stimulated
identically. There was no improvement in perceptual acuity when the device was
electrically inactive, suggesting that electrical stimulation did not improve
the health or function of the retina
All five patients demonstrated measurable
pupil constriction in response to electrical stimulation with their IRP.
Electrical stimuli parameters set to produce visual perception for 2 seconds
resulted in an average percent pupil constriction of 23% SD 7.5%. The average
time to maximum pupil constriction at these settings was 1557 msec SD 327 msec.
Average latency of onset of constriction was 510 msec SD 195 msec after the
initiation of the stimulus.
Evoked
potential recording showed regions of cortex activated by electrical
stimulation. Significant activity was noted in the parietal/occipital channels
from 200-300 ms as compared to the pre-stimulus data. Minimum norm
reconstruction showed activity on the cortical white matter of V1 at 144ms,
204ms, and 292ms with a shift in activation at these three time points from the
right to the left and back to the right hemispheres respectively. The source
localization technique RAP-MUSIC found 3 sources, 2 close to the visual cortex
and 1 corresponding to the stimulation device which was localized to the right
side of the subject (figure 2). Subspace correlation for this stimulation
source was 0.98 (1=best). From activity in V1, one source was found deep in the
cortex sitting between the hemispheres with a subspace correlation of 0.99. The
second source close to the visual cortex was found at a shallower position
relative to the first source and its subspace correlation was 0.95. The sources
are similar to those found when light stimulus is used.
Figure 1 – The model 1 retinal prosthesis in the eye of a
test subject with RP. The implant is visualized through a dilated pupil. The
optic disk is to the left of the array. Each electrode in this array is a 520
um diameter disk of platinum. The array is held to the retina with a tack.
Figure 2 EEG source localization. Electrical stimulation
was applied to the retina and signals recorded from the cortex. The algorithm
RAP-MUSIC was used to solve the inverse problem and define current sources
within the cortex. The source above is not in V1, but is in a location similar
to that seen with visual evoked potentials.
4. DISCUSSION AND CONCLUSIONS
These results demonstrate that the implanted
retinal prosthesis can activate the visual system in blind individuals. Electrical
stimulus of the retina was sufficient to activate the afferent limb of the PLR.
Pupillography in patients with an IRP provides an objective physiologic measure
of prosthesis function. Visual evoked potentials as seen on EEG were used with
source localization methods to positively identify two regions of cortex which
are activated as a result of artificial electrical stimulation of the retina.
1. Friedman, D.S., et al., Prevalence of age-related macular
degeneration in the United States. Arch Ophthalmol, 2004. 122(4): p.
564-72.
2. Sharma,
R.K. and B. Ehinger, Management of
hereditary retinal degenerations: present status and future directions.
Surv.Ophthalmol, 1999. 43(5): p. 427.
3. Rizzo,
J.F., 3rd, et al., Perceptual efficacy of
electrical stimulation of human retina with a microelectrode array during short
term surgical trials. Invest Ophthalmol Vis Sci, 2003. 44(12): p. 5362.
4. Zrenner,
E., Will retinal implants restore vision?
Science, 2002. 295(5557): p. 1022.
5. Margalit,
E., et al., Retinal prosthesis for the
blind. Surv.Ophthalmol., 2002. 47(4): p. 335.
6. Yanai,
D., et al., The value of preoperative
tests in the selection of blind patients for a permanent microelectronic
implant. Tr Am Ophth Soc, 2004. 101: p. 43.