Brain Behav. Evol. 14: 126-157 (1977)
Cortical Histopathology following Stimulation with Metallic and Carbon Electrodes J. J. B ernstein. P. F. J ohnson, L. L. H ench, G. H unter and W. W. D awson Department of Neuroscience, Materials Science and Engineering, Physiology and Ophthalmology, University of Florida College of Medicine, Gainesville, Fla.
Key Words. Electrode-tissue interaction • Stimulation and brain pathology • Neural prostheses Abstract. Stimulation of the cat striate cortex with polished 1 mm2 gold, platinum, rhodium or carbon electrodes for 40 h, at 50 Hz, and 0.5 msec duration, results in tissue damage at current densities of 0.6 A/cm2. At 0.1 A/cm2 rhodium and carbon were toxic to brain. In vitro gold and platinum corroded after 1,500-2,300 h of testing in simulated cerebrospinal fluid. The materials utilized are not suitable for long-term chronic electrodes for neural prosthetics.
In recent years the knowledge of the morphology and physiology of the nervous system has progressed to the point that it is feasible that ex ternal stimulation (electrical) can be utilized to compensate for central nervous system deficits in function. To this end, many prosthetic devices are being developed for the stimulation of various portions of the cen tral and peripheral nervous system. Basically, these devices supplement or replace electrical signals within a given nervous system structure by utilizing electrical stimulation through an electrode or an array of elec trodes. Functional neuromuscular stimulators have been utilized to alleviate central or peripheral deficits in function. Thus, various muscle groups can be stimulated by utilizing cuff or percutaneous electrodes to stimu late peripheral nerves [22, 24, 34, 35], The control of stimulation is de
Downloaded by: King's College London 137.73.144.138 - 1/16/2019 12:55:34 PM
Introduction
127
rived from various sources such as external switches attached to limbs [34], the muscle itself [34], the spinal cord [37], or single cortical neu rons [21]. Auditory prosthetics have also been investigated which acti vate portions of the peripheral or central auditory system [12, 23, 27, 33], In general, these devices use electrical stimulation of eighth cranial nerve or the temporal cortical surface, providing external signals for au ditory processing. Visual prosthetics have also been investigated that produce visual patterns upon stimulation of the visual or striate cortex [2-6, 9-11]. Stimulation of an electrode placed on the visual cortex re sults in the generation of a visual ‘phosphene’. Such phosphenes are de scribed as colorless or colored spots of light and can be combined to produce recognizable simple patterns. These prosthetic devices utilize metal electrodes of various types [18]. The noble metals have been favored. These include platinum, plat inum-10%> iridium, iridium, gold, and gold alloys. In addition, stainless steel and tungsten have been utilized as stimulating electrodes [18]. However, there is a paucity of information about the electrode-nervous system tissue interaction during electrode stimulation [17, 19, 26]. The relationship between tissue and electrode, tissue-electrode and stimula tion, electrode corrosion following implantation as a function of stimula tion or nonstimulation and the action of these electrodes over long peri ods of stimulation under chronic, high current, and high charge density conditions remains unknown. For these reasons the following in vivo and in vitro studies were un dertaken. The in vivo studies, using the cat as a model, involved im planting four electrodes on the striate cortex. The initial surfaces of pro totype electrodes were characterized by scanning electron microscopy (SEM). Two of the electrodes were tested with a total of 40 h of current pulses over several days; the remaining electrode pair was unstimulated and served as a passive control. The stimulation parameters used brack eted those known to produce visual phosphenes in humans [2-6, 9, 10]. A 50-Hz, biphasic, symmetrical current pulse, 0.5 msec duration per phase was generated by a constant current unit developed and fabricated locally (E. R. C henette and F. H yatt, Dept, of Electrical Engineering, University of Florida, Gainesville, Fla.). The pulse amplitude was either 1.0 or 6.0 mA for all tests. The geometric surface area of the polished electrodes was controlled at 0.01 cm-, yielding a current density of ei ther 0.1 or 0.6 A/geom cm2. The corresponding charge densities were thus 50 or 300 /
Cortical Histopathology following Stimulation with Metallic and Carbon Electrodes J. J. B ernstein. P. F. J ohnson, L. L. H ench, G. H unter and W. W. D awson Department of Neuroscience, Materials Science and Engineering, Physiology and Ophthalmology, University of Florida College of Medicine, Gainesville, Fla.
Key Words. Electrode-tissue interaction • Stimulation and brain pathology • Neural prostheses Abstract. Stimulation of the cat striate cortex with polished 1 mm2 gold, platinum, rhodium or carbon electrodes for 40 h, at 50 Hz, and 0.5 msec duration, results in tissue damage at current densities of 0.6 A/cm2. At 0.1 A/cm2 rhodium and carbon were toxic to brain. In vitro gold and platinum corroded after 1,500-2,300 h of testing in simulated cerebrospinal fluid. The materials utilized are not suitable for long-term chronic electrodes for neural prosthetics.
In recent years the knowledge of the morphology and physiology of the nervous system has progressed to the point that it is feasible that ex ternal stimulation (electrical) can be utilized to compensate for central nervous system deficits in function. To this end, many prosthetic devices are being developed for the stimulation of various portions of the cen tral and peripheral nervous system. Basically, these devices supplement or replace electrical signals within a given nervous system structure by utilizing electrical stimulation through an electrode or an array of elec trodes. Functional neuromuscular stimulators have been utilized to alleviate central or peripheral deficits in function. Thus, various muscle groups can be stimulated by utilizing cuff or percutaneous electrodes to stimu late peripheral nerves [22, 24, 34, 35], The control of stimulation is de
Downloaded by: King's College London 137.73.144.138 - 1/16/2019 12:55:34 PM
Introduction
127
rived from various sources such as external switches attached to limbs [34], the muscle itself [34], the spinal cord [37], or single cortical neu rons [21]. Auditory prosthetics have also been investigated which acti vate portions of the peripheral or central auditory system [12, 23, 27, 33], In general, these devices use electrical stimulation of eighth cranial nerve or the temporal cortical surface, providing external signals for au ditory processing. Visual prosthetics have also been investigated that produce visual patterns upon stimulation of the visual or striate cortex [2-6, 9-11]. Stimulation of an electrode placed on the visual cortex re sults in the generation of a visual ‘phosphene’. Such phosphenes are de scribed as colorless or colored spots of light and can be combined to produce recognizable simple patterns. These prosthetic devices utilize metal electrodes of various types [18]. The noble metals have been favored. These include platinum, plat inum-10%> iridium, iridium, gold, and gold alloys. In addition, stainless steel and tungsten have been utilized as stimulating electrodes [18]. However, there is a paucity of information about the electrode-nervous system tissue interaction during electrode stimulation [17, 19, 26]. The relationship between tissue and electrode, tissue-electrode and stimula tion, electrode corrosion following implantation as a function of stimula tion or nonstimulation and the action of these electrodes over long peri ods of stimulation under chronic, high current, and high charge density conditions remains unknown. For these reasons the following in vivo and in vitro studies were un dertaken. The in vivo studies, using the cat as a model, involved im planting four electrodes on the striate cortex. The initial surfaces of pro totype electrodes were characterized by scanning electron microscopy (SEM). Two of the electrodes were tested with a total of 40 h of current pulses over several days; the remaining electrode pair was unstimulated and served as a passive control. The stimulation parameters used brack eted those known to produce visual phosphenes in humans [2-6, 9, 10]. A 50-Hz, biphasic, symmetrical current pulse, 0.5 msec duration per phase was generated by a constant current unit developed and fabricated locally (E. R. C henette and F. H yatt, Dept, of Electrical Engineering, University of Florida, Gainesville, Fla.). The pulse amplitude was either 1.0 or 6.0 mA for all tests. The geometric surface area of the polished electrodes was controlled at 0.01 cm-, yielding a current density of ei ther 0.1 or 0.6 A/geom cm2. The corresponding charge densities were thus 50 or 300 /
