| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |
|
|
|
|||||||
|
|||||||||
Electronic Letters to:
|
|
Electronic letters published:
![[Read eLetter]](/icons/misc/sectionDOWN.gif){kind=icon}
Cortical Activation Via Retinal Prosthesis
Author Response: Cortical Activation Via Retinal Prosthesis
|
|
|||
|
Heinrich Gerding
Send letter to journal:
hgerding{at}klinik-pallas.ch Heinrich Gerding
|
I would like to congratulate the authors of the recent publication "Cortical Activation Via an Implanted Wireless Retinal Prosthesis"1 on their breakthrough of constructing and successfully testing a wireless retinal prosthesis in vivo. However, some details of the publication merit a closer critical consideration. Stimulation in the reported experiments was performed at 6.5 times (650 mC/cm2) the upper critical limit of charge transfer density acceptable for the material used here.2 Excessive charge transfer is a questionable experimental setup for implantable devices and may cause damage to retinal tissue and to the electrode material. Unfortunately, histological evaluation of the retinas that were stimulated or data on the condition of the electrodes were not presented in the publication. The choice of parameters is surprising since several groups and one of the authors3 have presented successful epiretinal stimulation indicated by cortical field potentials at levels more than one order of magnitude lower than here. The reader of this report not only misses information on lower level stimulation but also on stimulation results of the two other animals included in this series. The type of fixation device (tack) used for the stabilization of implants in this series was previously reported too difficult to handle and traumatic during insertion.4 It should have been replaced by less harmful devices. The authors' interpretation on tack failure ("bleeding . . . occurred in one case only" emphasis added; "tack fixation was an effective technique")1 is misleading, since in three experiments bleeding occurred on one occasion, as well as a completely aborted tack insertion. It is difficult for the reader to understand the reported shift of cortical activity as mentioned in the text when examining Figure 4. At least Figure 4C gives the impression of a gray scale shift opposite to the reported and mathematically analyzed direction. An illustration of the representative line of analysis would have been helpful for the reader. The reader may then at least gather the impression that phosphenes evoked by the implant are many times the size of the approximately 5° shift reported from A to C in Figure 4 and that this size is much too large for a useful image resolution. Heinrich Gerding Klinik Pallas, Augenzentrum, Olten, Switzerland References 1. Walter P, Kisvárday ZF, Görtz M, et al. Cortical activation via an implanted wireless retinal
prosthesis. Invest Ophthalmol Vis Sci. 2005;46:1780-1785. |
|||
|
|
|||
|
Peter Walter
Send letter to journal:
pwalter{at}ukaachen.de Peter Walter
|
We read with interest the letter of H. Gerding regarding our article on cortical activation via an implanted wireless visual prosthesis which was published in IOVS.1 He raised several questions on the experimental approach as well as on the presentation of the results. At the time the experiments were performed only platinum electrodes were available with the device. We are well aware that charge delivery is limited with platinum electrodes. However, the limit of about 100 mC/cm2 for safe charge injection (more strict papers report 40 mC/cm2 to be electrochemically safe) has been calculated on platinum sheets with polished surface. The thin-film metallization has been sputter deposited and has been exposed to oxygen plasma in a reactive ion process during opening of the electrode sites. Chemical and physical interaction of the plasma gases with the surface led to an increased surface roughness. Therefore, not the geometrical electrode area but the chemical active one should be taken into account when calculating the charge injection capacity. The retina stimulation electrodes with a diameter of 70 microns showed a charge delivery capacity of about 2 mC/cm2, i.e., about 20 times higher than calculated from a polished surface.2 With these values, we can estimate a charge injection capacity that is in the safe range even with the more strict values of safe charge injection stated above. However, new generations of retina implants will have electrode materials different from platinum, e.g., platinum black or iridium oxide, to decrease the impedance of the phase boundary to reduce energy consumption and to increase the possible charge injection capacity to be within the safe stimulation limits. Because of the complexity of the optical recording setting and the average time needed to obtain clear cortical metabolic responses it was not possible to evaluate the threshold for stimulation. Therefore a suprathreshold paradigm was chosen for stimulation, which explains the relatively strong stimulus intensity. At present tack fixation of epiretinal devices is regarded as a standard. Tack fixation of epiretinal stimulators was performed in rabbits, in dogs, and in humans.3-6 The authors are again well aware that alternative ways for a stable fixation would be helpful in reducing the surgical trauma for both implantation and also, if needed, explantation. However, alternative methods were not available when the experiments were performed. We discussed extensively the composition of Figure 4 to clearly demonstrate the activation shift within the visual cortex upon changing the stimulation area in the retina. We included the imaging data under stimulated conditions in the left part of the figure and under control conditions in the center. On the right part of the figure we included a gray scale analysis of each of the recordings A-D. These plots show the cumulative gray scale value, which resembles the sum of the gray scale of each pixel along the anterior posterior axis with respect to its position on the latero-medial axis. Therefore a representative line of analysis as suggested by Gerding in his letter is only helpful when a certain pixel along one axis is analyzed but not in the case of cumulative data along one axis. The peak analysis for each condition as indicated by the thick red line in the plots clearly indicates the position shift of the darkest, i.e., most active, region from a medial position towards a more lateral position. It is misleading to assume that the extent of cortical activation seen with OI in V1 corresponds to the size of percepts: the size of phosphenes is not equivalent to the extent of the metabolic activation of the brain tissue that is shown by the method of optical imaging of intrinsic signals. For example, when a 0.2°-wide light bar (approximately the size of the stimulation electrodes) is presented at the corresponding retinal topography the percept is a sharp image of 0.2° width while, at the same time, the metabolic activation of the visual cortex appears roughly as wide as that after electrical stimulation. With optical recording pre- and postsynaptic potentials, excitatory as well as inhibitory signals, sub- and suprathreshold activation are detected in the primary visual cortex while only suprathreshold responses that are transmitted to higher visual areas are finally relevant for perception. Peter Walter1 1Department of Ophthalmology, RWTH Aachen University, Aachen, Germany References 1. Walter P, Kisvárday ZF, Görtz M, et al. Cortical activation via an implanted wireless retinal prosthesis. Invest Ophthalmol Vis Sci. 2005;46:1780-1785. |
|||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |
eLetters: