Bacher, B. Jarosiewicz, N. Masse, J. Simeral, J. Vogel, S. Haddadin, J. Liu, S. Cash, P. Donoghue Serruya, G. Friehs, J. Mukand, M. Saleh, A. Caplan, A. Branner, D.
Augmentation of Sensorimotor Functions with Neural Prostheses
Chen, R. Penn and J. Houk, J. House, W. Humphrey, D. Schmidt and W. Thompson Ifft, P. Shokur, Z. Li, M. Jarvis, S. Schultz Jog, M. Connolly, Y. Kubota, D. Iyengar, L. Garrido, R. Harlan and A. Johnson, L. Fuglevand Kalman, R. Bucy Kamiya, J. Kaplan, B. Kennedy, P. Bakay and S. Sharpe Mirra and R. Bakay Kettner, R.
Schwartz and A. Georgopoulos Positional gradients and population coding of movement direction from various movement origins. Kim, S. Bhandari, M. Klein, S. Negi, L. Rieth, P. Tathireddy, M. Toepper, H. Oppermann and F. Solzbacher Callier, G. Tabot, F. Tenore and S. Bensmaia Sanchez, Y. Rao, D. Erdogmus, J. Carmena, M. Nicolelis and J. Principe Kowalski, K.
He and L. Srinivasan Kozai, T. Kipke Kralik, J. Dimitrov, D. Krupa, D. Katz, D. Cohen and M. Wiest, M. Shuler, M. Laubach and M. Kwon, K. Lee, M. Ghovanloo, A. Weber and W. Li Kyriazis, M. Crist and M. Methods for Neural Ensemble Recordings. Boca Raton FL. Denton and R. Nelson Tate, T. Hanson, Z. Li, J. O'Doherty, J. Winans, P. Ifft, K. Zhuang, N. Fitzsimmons, D. Schwarz, A. Fuller, J. An and M. Lee, P. Sie, Y. Liu, C. Wu, M. Lee, C. Shu, P. Li, C. Sun and K. Shyu Leuthardt, E.
Related JoVE Videos
Miller, G. Schalk, R. Rao and J. Ojemann Lewis, P. Ayton, R. Guymer, A. Lowery, P. Blamey, P. Allen, C. Luu and J. Rosenfeld Lewis, T. Rinzel Li, G. Zhang Li, Z. O'Doherty, T. Lebedev, C. O'Doherty, M. Libet, B. Alberts, E. Wright, Jr. Delattre, G. Levin and B. Feinstein Lisi, G.
Noda and J. Liu, N. Chiang and H. Hsu Llinas, R. Walton, M. Nakao, I. Hunter and P. Anquetil Madan, C. Maguire, G. McGee Time for debate. Mahmoudi, B. Pohlmeyer, N. Prins, S. Geng and J. Marchesi, M. BRAVO: a brain virtual operator for education exploiting brain-computer interfaces.
Martisius, I. Damasevicius Mason, S. Bohringer, J. Borisoff and G. Birch Maynard, E. Nordhausen and R. Miller, K. Shenoy, J. Miller, R. Montijn, J. Vinck and C. Pennartz Moore, M. Moritz, C. Perlmutter and E. Fetz Morizio, J. Irazoqui, V. Go and J. Parmentier Wireless headstage for neural prosthetics. Conference Proceedings. Morrow, M. Musallam, S. Bak, P. Troyk and R. Andersen Corneil, B. Greger, H. Scherberger and R.
Najafi, K. Wise and T. Mochizuki Neves, H. Orban, M. Koudelka-Hep, T.
Stieglitz and P. Ruther Development of modular multifunctional probe arrays for cerebral applications. Neural Engineering, Nicolelis, M. Baccala, R. Lin and J. Chapin Ghazanfar, B. Faggin, S. Votaw and L. Oliveira Ghazanfar, C. Stambaugh, L. Oliveira, M. Laubach, J. Chapin, R. Nelson and J. Kaas Lebedev Lin, D. Woodward and J. Nordhausen, C. Maynard and R. Lebedev, T. Hanson, N. Fitzsimmons and M. Lebedev, P. Zhuang, S. Shokur, H. Bleuler and M. Obeid, I. Nicolelis and P. Wolf Obermaier, B.
Neuper, C. Guger and G. Pfurtscheller Okorokova, E. Linderman and A. Ossadtchi Onose, G. Avramescu, I. Constantinescu Ordikhani-Seyedlar, M. Lebedev, H. Sorensen and S. Puthusserypady Orsborn, A. Dangi, H. Moorman and J. Oxley, T. Opie, S. John, G. Rind, S. Ronayne, T. Wheeler, J. Judy, A.
- Media, Culture and Society in Putins Russia?
- Methods for neural ensemble recordings .
- Visiome Platform - Digital archive for vision science!
- Methods for Neural Ensemble Recordings, Second Edition by Miguel A. L. Nicolelis;
- Judging the Jury.
- Related Citations.
McDonald, A. Dornom and T. Lovell Pais-Vieira, M. Kunicki, J. Wang and M. Wiest and M. Patil, P. Nicolelis and D. Turner Patrick, J. Valeur, L. Monnerie and J. Changeux Peikon, I. Fitzsimmons, M. Penfield, W. Boldrey Pezzulo, G. D'Ausilio and A. Gaggioli Pfurtscheller, G.
Pfurtscheller, H. Gerner and R. Rupp Neuper Pfurtscheller, J. Rupp, G. Muller, E. Fabsits, G. Korisek, H. Gerner and G. Improvement of grasping function with FES in a patient with C5 tetraplegia]. Picot, A. Charbonnier and A. Caplier Pohlmeyer, E. Oby, E. Perreault, S. Solla, K. Kilgore, R. Kirsch and L. Solla, E. Perreault and L. Quiroga, R. Reddy, G. Kreiman, C. Koch and I. Fried Rajangam, S. Tseng, A. Yin, G. Lehew, D. Schwarz, M. Ramakrishnan, A. Ifft, M.
Pais-Vieira, Y. Byun, K. Zhuang, M. Histology of the nervous system of man and vertebrates. Rao, R. Stocco, M. Bryan, D. Sarma, T. Youngquist, J. Wu and C. Prat Recce, M. The tetrode: a new technique for multi-unit extracellular recording. Soc Neurosci Abstr. Romo, R. Zainos and E. Salinas Rothschild, R. Sakurai, Y. Santos, L. Opris, J. Fuqua, R. Hampson and S. Journal of Neuroscience Methods [22 Mar , 1 ]. Type: Research Support, Non-U.
Abstract With the availability of sophisticated genetic techniques, the mouse is a valuable mammalian model to study the molecular and cellular basis of cognitive behaviors. However, the small size of mice makes it difficult for a systematic investigation of activity patterns of neural networks in vivo. Here we report the development and construction of a high-density ensemble recording array with up to recording channels that can be formatted as single electrodes, stereotrodes, or tetrodes. This high-density recording array is capable of recording from hundreds of individual neurons simultaneously in the hippocampus of the freely behaving mice.
This large-scale in vivo ensemble recording techniques, once coupled with mouse genetics, should be valuable to the study of complex relationship between the genes, neural network, and cognitive behaviors. Read Article at publisher's site. Following this approach has allowed our experimental animals to tolerate these implants very well, survive without any postsurgical complication, and provide high-quality recordings for months after the surgery.
After the animal is anesthetized, mounted to the stereotax, and cleaned, the surgical team suits up for surgery in a lab coat or gown, mask, sterile gloves, and head cap. Before starting surgery, the status of the anesthesia is checked as described above. An incision is made on the midline of the scalp, from just above the eyes to the back of the head, and the skin is propped open.
The periosteum is scraped off the skull bone using a blunt tool Freer or the scalpel blade. It is very important to clean the bone surface very well and remove all soft tissue attached to the bone. Bleeding spots on the bone may be cauterized or pressed until the bleeding stops. The bone is then cleaned with hydrogen peroxide, applied very carefully to avoid touching soft tissues around the cut. This process can be repeated several times until the bone is clean, whitish, and all the blood is removed from the bone surface. The hydrogen peroxide is washed with sterile 0.
The bone must be dried with gauze and must look very whitish. This step is most important for the fixation of the electrodes with dental acrylic. Any soft tissue or bleeding may lead to infection of the site, and the head cap may become dislodged some time after the surgery. When the skull is clean and dry, an extrafine point marker, can be used to mark the skull. For rodents, the zero point is the center of the bregma. Using the stereotaxic apparatus, marks are made at the center of the area of the planned implants.
The margins of the craniotomy for the electrodes are drawn around the central marks and other marks for the fixation screws are made. For mice, two screws are used; for rats, five to six screws are used. These metal screws stainless steel, blunt tip will hold the electrodes and the head cap in place and also provide a common electrical ground for the microwire arrays. Using a dental drill under a surgical microscope view, small holes for the screws are then drilled. At least one of the screw holes must be completely open, exposing the dura. The screw placed in this hole will be in contact with the dura and will be used as a ground for the electrodes.
The screws are placed in the holes and tightened with a screw driver, turning the screw enough to be firmly attached to the bone without penetrating the brain. If the dura is accidentally opened and cerebrospinal fluid CSF leaks from any hole after the screws are in place, it is recommended to either seal the screw with cyanoacrylate glue, or remove the screw and seal the hole with glue and place the screw in another location in the skull. Leaking of CSF under the head cap can lead to infection and softening of the bone over time, and will cause the head cap to become loose in the future, compromising the electrode positioning and increasing the risk of the head cap becoming detached from the skull.
After all screws are in place, the craniotomies for the electrodes are drilled. Care is taken to avoid damage to the dura during the drilling process. The size of the craniotomy must be just large enough to accommodate the electrode arrays. With a small drill bit, the edges of the craniotomy are drilled all the way to the bottom of the bone.
To prevent damage to the dura, the bone should be tested frequently for softness. When the bottom feels soft, the rest of the bone can be taken out with a small cup curette, the tip of a needle, or with the tip of a dental explorer. Once all the bone from the edges of the craniotomy is drilled, the loose bone in the center can be removed either with a small sharp tool or micro tweezers. Once the craniotomy is opened, it is washed with saline until all the bone dust is removed.
The other craniotomies are opened the same way and covered with Gelfoam or saline. If no damage is done to the dura, the craniotomy will look very clean and the surface of the brain and blood vessels can be seen. Once all the craniotomies are opened, electrodes are set up for insertion, one array at a time. The Gelfoam is taken out of the craniotomy and is replaced by saline solution. The array, which is connected to headstages cables and connectors placed in special holders attached to the stereotax micromanipulators, is slowly lowered where the craniotomy is open to minimize brain damage and bleeding.
Depending on how the tip of the electrodes are cut sharp or blunt , the electrodes may be inserted in the brain without opening the dura. This maneuver is done slowly and with patience until the electrodes pierce the dura. If the wires are blunt, or it is impossible for them to break through, the dura must be opened. This can be done with a fine needle with a bent tip, taking care to avoid breaking the blood vessels. In general, a cut on the dura is enough to relieve tension, but it may be necessary to remove all the dura in the craniotomy.
After the electrodes are touching the brain surface, the special grounding wire is wrapped around the screws, including the screw reserved for grounding. The electrodes can be grounded in other ways. For example, the tip of the ground wire can be placed under the skull in a small hole, and this hole can be fixed with cyanoacrylate gel glue to secure the ground wire in place. Then the rest of the ground wire can be wrapped around the screws. However, this technique requires more time and patience. Cell activity should be monitored during the implantation of the electrodes to help ensure placement of the array in the desired layer or structure of the brain.
If during the placement of the electrodes, the superficial blood vessels break and bleeding occurs, it is better to remove the electrodes from the craniotomy, press the hole until the bleeding stops, and wash the craniotomy and electrode tip with saline. In our rodent surgeries, bleeding in the dura is not cauterized. Blood on the tip of the electrodes may interfere with monitoring of the brain signals during surgery. A cotton-tip applicator over the craniotomy, a little roll of sterilized kimwipe or a piece of Gelfoam will help stop bleeding.
Once bleeding stops, the craniotomy is extensively washed with saline, and the electrodes can be placed in the craniotomy again and slowly lowered into the brain. Dimpling of the brain may occur during electrode implantation, and it can result in traumatic brain injury. In our experience, this can be prevented by making the craniotomy as small as possible and performing the implantation slowly and gently, especially in cortical implants.
If the first insertion of the arrays is not successful and no signals are recorded, removing the electrodes and reinserting them in the same craniotomy does not work well for cortical areas, but may succeed for subcortical implants. In this case, the electrodes can be reinserted in the same craniotomy in a slightly different position. After the array of electrodes penetrate the brain and reach the desired area cortical or subcortical , a layer of Gelfoam or warm agar is placed on the craniotomy around the microwires, being careful to avoid disturbing the position of the electrodes already in place.
A drop of cyanoacrylate glue gel or dental acrylic is used to fix the electrodes to the nearest screws and to the bone, with care to avoid covering the next craniotomies. The craniotomy with the electrodes must be carefully sealed with acrylate to avoid leaking of CSF, especially when the dura has been surgically opened. After the glue or the dental acrylic is dried, the cables for recording headstages are removed from the electrode-array connector, and the electrode holder is gently released.
If more electrode arrays are to be placed in the brain, the same procedures are repeated until all the arrays are in place. To decrease time during surgery, an accelerator can be used to speed curing of the glue when using cyanoacrylate glues. After all the arrays are in place, a layer of dental acrylic is applied around and between all the arrays and screws to create a strong head cap fixed to the skull. The edges of the head cap should be smooth and should cover all of the bone to avoid postoperative infection. At this stage, a small piece of hard wire may be placed in the head cap a staple will work very well for this purpose , which will be used to help anchor the head stage to the head cap with tiny rubber bands during future recording sessions.
If the skin is loose around the head cap, stitches may be necessary, and are done with suitable sutures. The skin around the cut area is washed with saline, dried, and a layer of antibiotic ointment is applied. Once the surgery is over, the animal is returned to a clean cage and partially positioned under a heating lamp, such that when the animal awakes, it can move away from the heat if necessary. Care is taken to position the lamp to avoid burning the animal.
The small particles of the bedding on the cage bottom can be covered with paper towels to prevent possible choking on or ingestion of the bedding when the animal is waking from anesthesia. The animal is checked at least every 15 min until it is fully awake. Moist food pellets and water are offered when the animal is fully awake and moving around.
It is very important to observe the animal when it is waking, especially when Pentobarbital anesthesia is used. Secretion in the respiratory airways may cause the animal to choke during this period. If necessary, another dose of atropine may be given to prevent secretion formation and possible choking.
All the animals subjected to surgery are housed in individual cages and followed closely for 7—10 days after surgery. Muscle activity can be recorded for many pairs of muscles. In general, we place 1—6 pairs of EMG electrodes in the following muscles: lavator labii superior, external epicantus, trapezius, biceps, triceps, and gastrocnemius. The EMG wires are attached to one connector that will be attached to the head cap for the brain arrays and are implanted during the same surgical procedure as the electrodes implantation in the brain.
The EMG microwires are very thin and flexible, will not interfere with movements of the target muscle, and will not disturb the health and mobility of the animal. This step will add up to 30 to 90 min to the surgery depending on the number of pairs of muscles to be implanted. The needle is pushed through the subcutaneous space up to the surgical opening on the head close to the electrode arrays. Once the tip of the needle is outside the cut area on the head, the EMG wire is passed through the needle until the tip is outside the body close to the target muscle.
The needle is then removed, leaving the EMG wire in the subcutaneous space; the excess of the microwire is cut, and the tip inserted into the target muscle. The surgical opening is closed using suitable sutures. The area is cleaned and a layer of antibiotic ointment is applied. These steps are repeated for all the muscles to be implanted with EMG wires. After all EMG wires are in place, the head cap is finished, and the cuts are cleaned and covered with antibiotic ointment. As a rule, our rodents are treated for pain for the first 24 h after surgery, whether or not they are showing signs of pain.
Rats will receive a dose of Tylenol every 6 h or an injection of Buprenorphine every 8 h. Determining which to use depends on the status of the animal. If no signs or minimal signs of pain are present, the animals will be treated with Tylenol. If there is strong evidence of pain animal is quiet in a corner, walking in circles, not eating or drinking, and has altered aspects of the fur and posture , the animal will be treated with Buprenorphine. Mice are treated for pain in the first 24 h following surgery with Buprenorphine.
The animals are observed during the next postoperative days, and pain medication is given as necessary. Typically, our animals are back to their normal behavior and routine 24—48 h after surgery. The surgical wound is checked everyday for signs of bleeding and draining, and to determine whether bedding materials may be contaminating the cut. If the surgical area appears abnormally red or swollen, is draining fluid or if the animal does not appear to be recovering well, the animal will be treated with systemic antibiotics and pain medication will be given as necessary.
After 10—14 days, stitches will be removed. In successful surgeries, the animals tolerate the presence of the head cap extremely well and neuronal activity can be recorded for months. One room is used for general preparation. This room also contains the table where the animal is prepared for surgery. The second room is used as the scrub room and contains a sink for the surgical team to scrub in before surgery, along with sterile garment supplies such as sterile gloves and gowns.
The third room is the surgery suite. Surgery at our institution is performed in a dedicated operating room OR equipped with OR lights, an operating table, a microscope, suction equipment, ventilation equipment, an oxygen and air supply, a ventilator, and a physiological monitor. The presence of suction equipment is important for venting of inhalational anesthetics.
A variety of animal ventilators are commercially available. Not all ventilators work well with very small animals, such as owl and squirrel monkeys, due to the small volumes and pressures that these animals require. A physiological monitor that can display waveforms for heart rate, breathing, oxygen saturation, temperature, and EKG is the cornerstone to intraoperative monitoring of the animal. Surgical instruments consist of basic clamps, scissors, and forceps. More important are the microsurgical instruments. A micro cup curette works well for trimming bone edges on the inside of the craniotomy after drilling.
Micro scissors and micro forceps are needed for removal of the dura.
Very fine suction catheters are useful to clear CSF from the operative field. The surgical microscope should have binocular vision and must be covered with a sterile drape designed for microscopes. Magnification and ease of use are generally limited by cost. The microscope is used extensively for drilling craniotomies, dural resection, and visually monitoring the entry of the electrodes into the cortex.
These include: 1. Steroids—given perioperatively to reduce inflammation and brain swelling. Antibiotics—given prophylactically before surgery and throughout the operation via IV. Fentanyl—an opioid analgesic that is also used as an adjunct for maintenance of anesthesia with isofluroane. It works well to suppress respiratory drive in a hyperventilating animal.
Epinephrine and Atropine—for cardiac emergencies. For details regarding veterinary anesthesia medications and supplies, please refer to Hellebrekers and Booij Hellebrekers and Booij For electrode sterilization, electrodes are carefully packed to avoid contact of the tips with the packaging and damage to the microwires. They are sterilized in an oxide ethylene gas chamber. This step must be completed before the day of surgery to allow adequate time for the sterilization cycle and process. The success of a surgery depends upon a number of personnel working together in a coordinated fashion.
Older animals with less cardiopulmonary reserve are more likely to develop postoperative problems such as pulmonary edema. This can be done the day before surgery to save time. Food is withheld the day before surgery to lessen the chance of vomiting and aspirating into the lungs during administration of anesthesia and placement of the endotracheal tube. Cortical localization should be planned well in advance. Atlas information and published coordinates of the areas of interest should be studied carefully.
It is important to consider the depth of the cell layer of interest as well as the surface conformation—areas deep within a sulcus or very near the midline where the saggital sinus runs may be technically difficult to access. It is important preoperatively to have a three-dimensional idea of where the electrodes and connectors, head posts, and ground wires will fit on the surface of the skull. This issue is magnified when working with smaller species such as owl and squirrel monkeys.
Unlike rodents, there is more variability in the size, symmetry, and proportions of primate brains, making it important to have other backup methods in place to verify the locations of critical landmarks and structures. Preoperative magnetic resonance imaging MRI or computed tomography CT localization is one way to address this issue. In this paradigm, adapted from human functional-neurosurgical techniques, fixed markers such as a halo attached to the skull or implanted markers e. Details can be found in veterinary textbooks. With early surgeries, we had difficulty with unpredictable swelling or retraction of the brain.
Smaller species such as owl and squirrel monkeys seem more prone to this than rhesus monkeys. These smaller species have a tendency to hyperventilate when intubated and under anesthesia, resulting in blowing CO2 and causing significant brain retraction. In order to address these issues, we have instituted use of a ventilator. We adjust the respiratory rate and tidal volume to keep CO2 in the physiologic range.
If brain swelling or retraction occurs, adjustments to the ventilation can be made to counteract this effect by manipulating the end tidal CO2. Intravenous narcotics such as fentanyl are a useful adjunct to depress respiratory drive and counteract hyperventilation. The various monitors and electrical devices used during surgery create a significant amount of electrical noise. Some devices can be run on batteries, thereby decreasing electrical noise.
A strategy should be developed allowing basic monitoring that minimizes noise during the critical electrode-lowering phase of surgery. Both core body temperature and depth of anesthesia are known to affect spontaneous neuronal firing. Core body temperature should be as close to normal as possible during the critical electrode-lowering steps when neuronal firing is being recorded. The depth of anesthesia is more difficult to tightly control.
We have found that during the first steps of the operation, we can get a sense of a target heart rate that is indicative of an adequate level of anesthesia and a heart rate above which the animal is likely to begin moving. We titrate the anesthetic carefully during electrode lowering and neuronal recording to target this number. Other possible solutions to this problem include the use of paralytics to ensure the monkey does not move while temporarily lightening the anesthetics and simultaneously using ketamine anesthesia.
In our paradigm, the electrodes are fastened to the skull rigidly after implantation, relying on the surrounding bone to maintain their place in the brain. The appropriate length of electrodes for smaller species such as owl and squirrel monkeys is 5—7 mm. For rhesus macaques, 6—8 mm is appropriate. One should expect variability in depth of the skull across different areas of the brain. The skin incision can either be midline with extensions laterally to allow adequate exposure, or the skin can simply be cut in an oval pattern and removed—the scalp is very forgiving in its healing properties.
Subperiostial dissection of the soft tissues is essential to assuring a good bond between the skull cap and skull. All soft tissues down to the cortical bone of the skull must be removed. Dissection laterally and posteriorly should extend well beyond proposed craniotomies. Laterally, the temporalis muscle and its attachments must be removed to expose hand somatosensory and motor areas. The convexity of the skull is then prepped.
Once dry, the border of the soft tissues and skull can be demarcated using a protective tape-type barrier to avoid leakage onto the soft tissues of the caustic materials used in the next step see Figure 2. Next, hydrogen peroxide is used to remove all remaining shreds of soft tissue from the skull. Bleeding from the bone is stopped with cautery. Acid solutions and mechanical instruments such as a scalpel can be used to etch the outer cortical bone and to confirm that no soft tissue remains. The reason for doing this is to minimize or prevent a soft-tissue layer from developing between the skull cap and skull, which can be a nidus for infection and source of loosening of the head cap.
The importance of careful skull cleaning cannot be overemphasized see Figure 2. Cortical Localization The midline and interaural line is marked on the skull. Using predetermined coordinates based on these landmarks, the locations of the areas to be implanted are marked on the skull. Based on the size and orientation of the electrodes, craniotomies are outlined, taking into consideration how the connectors will be oriented.
Bregma is highly variable in primates, and is not considered a reliable landmark for cortical localization see Figure 2. Drilling of Craniotomies Using a high-speed air micro drill adapted from dental applications, and the microscope, the craniotomies are drilled carefully respecting the dura.
Nanowire-Based Electrode for Acute In Vivo Neural Recordings in the Brain
A bony lesion is evident in the center. All blood has been cleared. All periosteum has been removed. If the sulcus is not clearly identifiable through the dura, the dura is opened avoiding injury to any of the surface vasculature. Once the key landmark is found, the other craniotomies can be adjusted appropriately, if needed. This is a good backup mechanism to be sure that the areas of interest are definitively identified given the known variability in surface landmarks in primates see Figure 2.
An appropriately sized craniotomy should allow enough working space to be able to visualize cortical structures both for opening identifying landmarks and to facilitate the technical challenge of opening the dura as well as allowing visual confirmation, if possible, of electrode penetration. Very large openings can cause significant leakage of CSF and distortion of the cortical surface, both of which are detrimental to the technique. Ideally, the opening allows space for the electrodes, while leaving the brain underneath in its native position.
CSF shifts, or herniation of large craniotomies, may change the local mechanical forces enough to impede optimal electrode placement and stabilization. Too big a craniotomy also makes it difficult to secure electrodes, because the technique relies on rigid attachment to the surrounding cranium. When all the craniotomies are drilled, securing skull screws are placed.
A variety of options exist ranging from small stainless steel screws available at the hardware store to specialized titanium skull screws for human applications. Dura is opened microsurgically as depicted on projected plasma screen. More important than discussing specific fixation techniques are the general principles that are important for obtaining secure fixation of the head cap. It is well known that if too much stress is applied to a bone or screw interface, the bone will gradually resorb, connective tissue will form, and the screw will loosen Betelak, Margiotti et al.
Generally, 10—16 3. Because of spatial constraints, head posts used to fix animals during experimental sessions are often included in the same dental-acrylic mass as the electrode arrays. As described by Betelak et al. If a head post is attached to the same head cap that holds the electrodes, a waiting period of 3 months should be taken into consideration in planning postoperative experiments with the head fixed. Skull screws of any kind should be carefully tailored to the depth of the bone.
Placing them too deep can have devastating consequences such as piercing the brain and causing bleeding, which can lead to seizures. Skull screws are used as grounds for the electrodes see Figure 2. It is not advisable to drill craniotomies after opening the dura or after implantation of electrodes due to the bone dust that is distributed and the vibration that is caused.
Spatial limitations make it nearly impossible. On the right side, cellulose sponges fill the craniotomy defect. Skull screws have been placed for grounding and fixation purposes. Note that the final position of the craniotomies has been modified as the surface anatomy became clear during the course of the drilling. The craniotomies have been sized to accommodate the electrodes. Opening of the Brain Coverings Unlike in rodents, we have found that it is always necessary to open the dura in primates to achieve successful penetration of the cortex. Under high-power magnification, the dura is opened in all of the craniotomies.
A fine needle can be bent at the tip and used as a hook to slide under the dura and incise it. The dura has several layers and consists of white fibrous material arranged in overlapping stands. Care must be taken to be sure all layers have been opened. Freely flowing CSF and crystal-clear surface vasculature indicate complete dural removal.
A cupped curette is useful to cut the dura off at its attachment to the edge of the craniotomy to allow full exposure. Rarely there is bleeding from the dura. This can be dealt with either with cautery or Gelfoam soaked in thrombin to enhance hemostasis. If there is bleeding from disruption of the cortical surface vessels, placement of a Gelfoam sponge soaked in thrombin works well. Every effort should be made to avoid injury to the cortical surface. Opening the dura is challenging, and practice should be performed on rats to gain familiarity with the technique and to fine tune what works best in individual hands.
At this point, the brain should be assessed and adjustments to the ventilation made, if needed, to correct any obvious retraction or marked swelling. A neutral brain that fills the opening of the craniotomy with occasional flow of CSF from under the craniotomy edge indicates a neutral position. We have opened the pia through a variety of techniques including enzymatic digestion using collagenase, and mechanically with fine forceps.
Although both are feasible and result in good recordings, despite visual disruption to the cortical surface, we have found that appropriate electrode spacing obviates the need to pursue these technically challenging and time-consuming steps. By adjusting the spacing of individual electrodes alone, we have been able to achieve consistent pial penetration. Other factors that can be manipulated to help achieve penetration are electrode diameter and the shape of the electrode tip. Electrode Lowering Once the dura is opened in all the craniotomies, the pia is protected from drying out with a small sponge Gelfoam soaked in saline.
The electrodes should be moving only in line with the direction of the lowering arm to avoid shearing forces upon entry. Every attempt is made to have the electrodes enter as perpendicular as possible to the surface of the brain realizing that the brain is not a flat surface. The order of electrode lowering is strategically planned. It is best to work outward from the middle craniotomy, thereby avoiding working in between delicate electrodes.
Touchdown of the majority of the electrodes is marked as zero, and electrophysiological monitoring is begun. Noise issues are dealt with at this point. Anesthesia is titrated to target heart rate. A certain degree of dimpling occurs initially until the electrodes penetrate. We do not find any advantage to rapid or slow insertion. Occasionally, there is bleeding as cortical vessels are punctured if they cannot be avoided. This does not seem to affect intraoperative or postoperative recordings.
Once characteristic firing of superficial cortical neurons is established in the majority of channels, the electrodes are lowered based on known depth of the target layer and most importantly electrophysiologic monitoring. We have found the quality of the intraoperative recordings, i. Once the ideal location has been achieved, small pieces of cellulose sponges are placed around the electrode-brain interface. Cyanoacrylate glue in gel form is used to rigidly fix the electrode array to the skull see Figure 2.
The connectors are then removed, and the electrode is delicately released from the stereotax. This sequence is repeated until all electrode arrays are implanted. Skull and Wound Closure Once the arrays are all in place, being held by cyanoacrylate glue, there may be some leakage of CSF from the craniotomy openings. It is important at this point to seal the craniotomies in a water-tight fashion.
They are temporarily secured to the skull with cyanoacrylate glue visible in the center of the three electrodes. Ground wires have been connected to the skull screws. The electrodes are completely embedded to protect them. A secondary grounding wire is visible. The interface between the skull and dental acrylic is smooth without intervening soft tissue at any point.
No CSF leak is evident between the bone and head cap. Tape protects the electrodes during the cementing process. A thread is built into the head cap to allow a second protective cap to be fitted to protect the electrode connectors when not in use. It must also be dry to allow the best possible bond to the bone. The better the interface between the dental acrylic and the skull, the less potential for CSF leak, development of soft-tissue migration, and infection.
Finally, a layer of antibiotic ointment is placed at the skull, scalp, and acrylic interface. Excess or devascularized skin is trimmed. Stitches or clips can be used to reapproximate skin edges, usually anteriorly and posteriorly. The scalp is very forgiving in its ability to heal. Close observation for seizure activity and a plan for intervention are important. Seizures are infrequent and likely indicate an unanticipated bleed over the convexity, most likely from a misplaced skull screw. Neurological deficits are surprisingly rare and generally resolve over time. The animals are usually returned to their cages by 24 h postop and advanced to a regular diet as quickly as possible.
After the second postoperative day, analgesia is carefully discussed with a veterinarian staff and if necessary is generally provided through NS antiinflammatory drugs or Tylenol. However, narcotics will be used if necessary. It has been found with human surgery that, in general, early mobilization and return to function results in the least complications postoperatively, and the same principle applies to animal surgery. The wound is cleaned when necessary, and a new layer of antibiotic ointment is applied. Stitches are removed from 10—14 days after surgery.
Primates will be treated with antibiotics for 5 days after surgery. The brain-electrode interface remains one of the rate limiting steps in our ability to advance neuronal-ensemble physiology. Techniques to acquire brain signals from rodents and primates have advanced rapidly, and this is only the beginning. Over the last 18 years, we have worked on perfecting the surgical technique for implant of multielectrode arrays for chronic recording of brain cell activity.
Our animals recover very well from the surgery, and they are usually back to their normal activities in 24—48 h without any signs of pain after the first postoperative day. Generally, there are no problems with postsurgical infection, and over time the animals are not disturbed by the presence of the head cap. However, we make every effort to continuously improve the surgical techniques for rodents and primates. Our goal is to be able to reach the relevant areas of the brain for each study with better and faster techniques, new materials, and identify new ways to improve healing of the tissues and skull bone, without compromising the efficiency of the recordings in the future.
Especially for primate surgeries, preparation for surgeries involves extensive multipersonnel discussion of previous surgeries and ideas for future improvements. The design of the new electrode arrays, distribution of the microwires in the array, and materials used for the electrodes is carefully studied before each primate surgery and for each study involving rodents.
The use of mutant mice has provided a good data base for the study of several neurological diseases that affect many people in many different ways. The possibility of studies on these animals is unlimited and may contribute to therapies or cures for progressive brain diseases that effect the lives of millions of people. Our ultimate goal is to find safer and better implantation techniques that can one day be applied to investigate and treat a variety of catastrophic neurological diseases that affect millions of people throughout the world.
Also, we would like to thank our former members Erin Phelps and Kirsten Shanklin—Dewey, and our actual primate lab manager, Dr. Weiying Drake, for their constant support during our primate surgeries, supervision of our primate work, and for their extremely gentle care of our animals. Finally, a special thank to Susan Halkiotis, Gayle Wood, and Terry Jones, who work diligently for the success of our lab.
Experimental procedures. Principles of laboratory animal science. Van Zutphen, L. Amsterdam, Elsevier Science. Revised ed. Betelak, K. The use of titanium implants and prosthodontic techniques in the preparation of nonhuman primates for long-term neuronal recording studies. J Neurosci Methods 1 : 9— Chapin, J. Mapping the body representation in the SI cortex of anesthetized and awake rats. J Comp Neurol 2 : — Costa, R.
Differential corticostriatal plasticity during fast and slow motor skill learning in mice. Curr Biol 14 13 : — Rapid alterations in corticostriatal ensemble coordination during acute dopamine-dependent motor dysfunction. Neuron 52 2 : —