chronic-cranial-window-technique-for-repeated-cortical-recordings

Chronic Cranial Window Technique for Repeated Cortical Recordings During Anesthesia in Pigs

Cortical recordings are essential for extracting neuronal signals to inform various applications, including brain-computer interfaces and disease ...

Aalborg University

The in vivo electrochemical behavior of titanium nitride (TiN) nerve stimulation electrodes was compared to their in vitro behavior for a period of 90 days. Ten electrodes were implanted in two Göttingen minipigs. Four of these were used for electrical stimulation and electrochemical measurements. Five electrodes were kept in Ringer's solution at 37.5°C, of which four were used for electrical stimulation and electrochemical measurements. The voltage transients measured in vivo were 13 times greater than in vitro at implantation and they continued to increase with time. The electrochemical properties in vivo and the tissue resistance (Rtissue) followed a similar trend with time. There was no consistent significant difference between the electrochemical properties of the in vivo and in vitro electrodes after the implanted period. The differences between the in vivo and in vitro electrodes during the implanted period show that the evaluation of electrochemical performance of implantable stimulation electrodes cannot be substituted with in vitro measurements. After the implanted period, however, the performance of the in vivo and in vitro electrodes in saline was similar. In addition, the changes observed over time during the post-implantation period regarding the electrochemical properties of the in vivo electrodes and Rtissue were similar, which indicates that these changes are due to the foreign body response to implantation.

Electrochemical properties of titanium nitride nerve stimulation electrodes: an in vitro and in vivo study.

The aim of this study was to investigate how the electrochemical properties of porous titanium nitride stimulation electrode are affected by fibrous encapsulation in vivo. Electrochemical impedance spectroscopy (EIS), cyclic voltammetry and voltage transient (VT) measurements were performed in vivo and in phosphate buffered saline, where the encapsulation process is absent. EIS was used as a non-invasive measurement to follow the inflammation, healing and encapsulation process. EIS showed that the healing and encapsulation process lasted 3-4 weeks. The VTs increased during the first 3-4 weeks, after which they stabilized. The charge storage capacity (CSC) decreased most during the first 3-4 weeks. The increasing VTs and decreasing CSC during the first 3-4 weeks after implantation of the in vivo electrodes seem related to healing and fibrous encapsulation. It is suggested that the charge injection pathway during the encapsulation process changes, which implies that charge injection limits are underestimated with conventional methods.

Influence of fibrous encapsulation on electro-chemical properties of TiN electrodes.

Robust devices for chronic neural stimulation demand electrode materials which exhibit high charge injection ( ) capacity and long-term stability. Boron-doped diamond (BDD) electrodes have shown promise for neural stimulation applications, but their practical applications remain limited due to the poor charge transfer capability of diamond. In this work, we present an attractive approach to produce BDD electrodes with exceptionally high surface area using porous titanium nitride (TiN) as interlayer template. The TiN deposition parameters were systematically varied to fabricate a range of porous electrodes, which were subsequently coated by a BDD thin-film. The electrodes were investigated by surface analysis methods and electrochemical techniques before and after BDD deposition. Cyclic voltammetry (CV) measurements showed a wide potential window in saline solution (between -1.3 and 1.2 V vs. Ag/AgCl). Electrodes with the highest thickness and porosity exhibited the lowest impedance magnitude and a charge storage capacity (CSC) of 253 mC/cm, which largely exceeds the values previously reported for porous BDD electrodes. Electrodes with relatively thinner and less porous coatings displayed the highest pulsing capacitances ( ), which would be more favorable for stimulation applications. Although BDD/TiN electrodes displayed a higher impedance magnitude and a lower   as compared to the bare TiN electrodes, the wider potential window likely allows for higher   without reaching unsafe potentials. The remarkable reduction in the impedance and improvement in the charge transfer capacity, together with the known properties of BDD films, makes this type of coating as an ideal candidate for development of reliable devices for chronic neural interfacing.

Diamond/Porous Titanium Nitride Electrodes With Superior Electrochemical Performance for Neural Interfacing.

Background and aims Preclinical studies have reported that activation of peripheral γ-aminobutyric acid A (GABAA) receptors may result in analgesia. The current study was conducted in young healthy men (n = 30) and women (n = 28) to determine whether injections of GABA into the masseter muscle reduce pain in a sex-related manner. Methods The effect of injection of GABA alone, or in combination with the non-inflammatory algogen glutamate, was assessed in two separate studies. Lorazepam, a positive allosteric modulator of the GABAA-receptor, was co-injected with GABA in both studies to explore the role of this receptor in muscle pain responses of healthy human volunteers. Masticatory muscle mechanical pain intensity was recorded on an electronic visual analogue scale (VAS) while muscle pain sensitivity was assessed by determining the pressure pain threshold (PPT), tolerance and maximal jaw opening (MJO) of the subjects prior to, and again after the various intramuscular injections. Results Intramuscular injection of GABA alone was reported to be significantly more painful, in a concentration related manner, than saline control injections, and this pain was further increased by co-injection of lorazepam with GABA. Co-injection of GABA with glutamate was found to significantly increase glutamate-evoked masseter muscle pain in men, but not in women. There was no effect of injections of either GABA alone, or GABA with glutamate, on PPT, tolerance or maximum jaw opening. Conclusions Injection of GABA into the human masseter muscle appears to excite nociceptors to produce muscle pain without a longer term effect on mechanical pain sensitivity in the muscle. The findings suggest that GABA-mediated pain in humans is produced through peripheral GABAA receptor activation. The mechanism underlying the sex-related difference in the effect of GABA on glutamate-evoked muscle pain was speculated to be due to a methodological artifact. Implications This study was designed to detect analgesic rather than algesic effects of peripherally administered GABA, and as a result, the concentration of glutamate chosen for injection was close to the maximal pain response for healthy women, based on previously determined pain-concentration response relationships for glutamate. This may explain the finding of greater pain in men than women, when GABA and glutamate were co-injected. Overall, the findings suggest that activation of peripheral GABAA receptors in human masticatory muscle produces pain, possibly due to depolarization of the masticatory muscle afferent fibers.

The pro-algesic effect of γ-aminobutyric acid (GABA) injection into the masseter muscle of healthy men and women.

Over the recent years, several small area electrodes have been introduced as tools for preferential stimulation of small cutaneous nerve fibers. However, the performance of the electrodes is highly debated and have not previously been systematically compared. The electrodes have been developed empirically and little is known about the electrical potential they produce in the skin, and how this influences the nerve fiber activation. The objective of the present study was to develop and validate a computational model to compare the preferential stimulation of small fibers for electrodes of different designs.

Comparison of existing electrode designs for preferential activation of cutaneous nociceptors.

This study implements the use of Danish Landrace pigs as subjects for the long-term potentiation (LTP)-like pain model. This is accomplished by analyzing changes in the primary somatosensory cortex (S1) in response to electrical stimulation on the ulnar nerve after applying high-frequency electrical stimulation (HFS) on the ulnar nerve. In this study, eight Danish Landrace pigs were electrically stimulated, through the ulnar nerve, to record the cortically evoked response in S1 by a 16-channel microelectrode array (MEA). Six of these pigs were subjected to HFS (four consecutive, 15 mA, 100 Hz, 1000 µs pulse duration) 45 min after the start of the experiment. Two pigs were used as control subjects to compare the cortical response to peripheral electrical stimulation without applying HFS. Low-frequency components of the intracortical signals (0.3-300 Hz) were analyzed using event-related potential (ERP) analysis, where the minimum peak during the first 30-50 ms (N1 component) in each channel was detected. The change in N1 was compared over time across the intervention and control groups. Spectral analysis was used to demonstrate the effect of the intervention on the evoked cortical oscillations computed between 75 ms and 200 ms after stimulus. ERP analysis showed an immediate increase in N1 amplitude that became statistically significant 45 mins after HFS ( < 0.01) for the intervention group. The normalized change in power in frequency oscillations showed a similar trend. The results show that the LTP-like pain model can be effectively implemented in pigs using HFS since the cortical responses are comparable to those described in humans.

The effect of peripheral high-frequency electrical stimulation on the primary somatosensory cortex in pigs.

Translating basic pain research from rodents to humans has proven to be a challenging task. Efforts have been made to develop preclinical large animal models of pain, such as the pig. However, no consistent overview and comparison of pig models of pain are currently available. Therefore, in this review, our primary aim was to identify the available pig models in pain research and compare these models in terms of intensity and duration. First, we systematically searched Proquest, Scopus and Web of Science and compared the duration for which the pigs were significantly sensitized as well as the intensity of mechanical sensitization. We searched models within the specific field of pain and adjacent fields in which pain induction or assessment is relevant, such as pig production. Second, we compared assessment methodologies in surrogate pain models in humans and pigs to identify areas of overlap and possible improvement. Based on the literature search, 23 types of porcine pain models were identified; 13 of which could be compared quantitatively. The induced sensitization lasted from hours to months and intensities ranged from insignificant to the maximum attainable. We also found a near to complete overlap of assessment methodologies between human and pig models within the area of peripheral neurophysiology, which allows for direct comparison of results obtained in the two species. In spite of this overlap, further development of pain assessment methodologies is still needed. We suggest that central nervous system electrophysiology, such as electroencephalography, electrocorticography or intracortical recordings, may pave the way for future objective pain assessment.

A systematic review of porcine models in translational pain research.

Decoding information from the peripheral nervous system via implantable neural interfaces remains a significant challenge, considerably limiting the advancement of neuromodulation and neuroprosthetic devices. The velocity selective recording (VSR) technique has been proposed to improve the classification of neural traffic by combining temporal and spatial information through a multi-electrode cuff (MEC). Therefore, this study investigates the feasibility of using the VSR technique to characterise fibre type based on the electrically evoked compound action potentials (eCAP) propagating along the ulnar nerve of pigs in vivo. A range of electrical stimulation parameters (amplitudes of 50 μA-10 mA and pulse durations of 100 μs, 500 μs, 1000 μs, and 5000 μs) was applied on a cutaneous and a motor branch of the ulnar nerve in nine Danish landrace pigs. Recordings were made with a 14 ring MEC and a delay-and-add algorithm was used to convert the eCAPs into the velocity domain. The results revealed two fibre populations propagating along the cutaneous branch of the ulnar nerve, with mean velocities of 55 m/s and 21 m/s, while only one dominant fibre population was found for the motor branch, with a mean velocity of 63 m/s. Because of its simplicity to provide information on the fibre selectivity and direction of propagation of nerve fibres, VSR can be implemented to advance the performance of the bidirectional control of neural prostheses and bioelectronic medicine applications.

The Use of the Velocity Selective Recording Technique to Reveal the Excitation Properties of the Ulnar Nerve in Pigs.

A pressure algometer is a valuable tool for assessing the mechanical nociceptive threshold (MNT) in clinical pain studies. Recent research has turned to large animal models of pain because of the closer anatomy and physiology to humans. Although the reliability and usefulness of the MNT have been extensively validated in humans, similar data from large animals is still sparse.

On determining the mechanical nociceptive threshold in pigs: a reliability study.

The knowledge of the morphology and morphometry of peripheral nerves is essential for developing neural interfaces and understanding nerve regeneration in basic and applied research. Currently, the most adopted animal model is the rat, even though recent studies have suggested that the neuroanatomy of large animal models is more comparable to humans. The present knowledge of the morphological structure of large animal models is limited; therefore, the present study aims to describe the morphological characteristics of the Ulnar Nerve (UN) in pigs. UN cross-sections were taken from seven Danish landrace pigs at three distinct locations: distal UN, proximal UN and at the dorsal cutaneous branch of the UN (DCBUN). The nerve diameter, fascicle diameter and number, number of fibres and fibre size were quantified. The UN diameter was larger in the proximal section compared to the distal segment and the DCBUN. The proximal branch also had a more significant number of fascicles (median: 15) than the distal (median: 10) and the DCBUN (median: 11) segments. Additionally, the mean fascicle diameter was smaller at the DCBUN (mean: 165 μm) than at the distal (mean: 197 μm) and proximal (mean: 199 μm) segments of the UN. Detailed knowledge of the microscopical structure of the UN in pigs is critical for further studies investigating neural interface designs and computational models of the peripheral nervous system.

Morphology and morphometry of the ulnar nerve in the forelimb of pigs.

Translational models of the sensitized pain system are needed to progress the understanding of involved mechanisms. In this study, long-term potentiation was used to develop a mechanism-based large-animal pain model. Event-related potentials to electrical stimulation of the ulnar nerve were recorded by intracranial recordings in pigs, 3 weeks before, immediately before and after, and 3 weeks after peripheral high-frequency stimulation (HFS) applied to the ulnar nerve in the right forelimb (7 pigs) or in control animals (5 pigs). Event-related potential recordings and peripheral HFS were done during anesthesia. Two weeks before and after the HFS, behavioral responses reflecting mechanical and thermal sensitivity were collected using brush, noxious limb-mounted pressure algometer, and noxious laser stimuli. The HFS intervention limb was progressively sensitized to noxious mechanical stimulation in week 1 and 2 compared with baseline ( P = 0.045) and the control group ( P < 0.034) but not significantly to laser or brush stimulation. The first negative (N1) peak of the event-related potential was increased 30 minutes after HFS compared with before ( P < 0.05). The N1 peak was also larger compared with control pigs 20 to 40 minutes after HFS ( P < 0.031) but not significantly increased 3 weeks after. The relative increase in N1 30 minutes after HFS and the degree of mechanical hyperalgesia 2 weeks post-HFS was correlated ( P < 0.033). These results show for the first time that the pig HFS model resembles the human HFS model closely where the profile of sensitization is comparable. Interestingly, the degree of sensitization was associated with the cortical signs of hyperexcitability at HFS induction.

High-frequency electrical stimulation increases cortical excitability and mechanical sensitivity in a chronic large animal model.

This study describes cortical recordings in a large animal nerve injury model. We investigated differences in primary somatosensory cortex (S1) hyperexcitability when stimulating injured and uninjured nerves and how different cortical layers contribute to S1 hyperexcitability after spared ulnar nerve injury. We used a multielectrode array to record single-neuron activity in the S1 of ten female Danish landrace pigs. Electrical stimulation of the injured and uninjured nerve evoked brain activity up to 3 h after injury. The peak amplitude and latency of early and late peristimulus time histogram responses were extracted for statistical analysis. Histological investigations determined the layer of the cortex in which each electrode contact was placed. Nerve injury increased the early peak amplitude compared with that of the control group. This difference was significant immediately after nerve injury when the uninjured nerve was stimulated, while it was delayed for the injured nerve. The amplitude of the early peak was increased in layers III-VI after nerve injury compared with the control. In layer III, S1 excitability was also increased compared with preinjury for the early peak. Furthermore, the late peak was significantly larger in layer III than in the other layers in the intervention and control group before and after injury. Thus, the most prominent increase in excitability occurred in layer III, which is responsible for the gain modulation of cortical output through layer V. Therefore, layer III neurons seem to have an important role in altered brain excitability after nerve injury.

Spared ulnar nerve injury results in increased layer III-VI excitability in the pig somatosensory cortex.

Micro-electrocorticography (µECoG) electrodes have emerged to balance the trade-off between invasiveness and signal quality in brain recordings. However, its large-scale applicability is still hindered by a lack of comparative studies assessing the relationship between ECoG and traditional recording methods such as penetrating electrodes. This study aimed to compare somatosensory evoked potentials (SEPs) through the lenses of a µECoG and an intracortical microelectrode array (MEA). The electrodes were implanted in the pig's primary somatosensory cortex, while SEPs were generated by applying electrical stimulation to the ulnar nerve. The SEP amplitude, signal-to-noise ratio (SNR), power spectral density (PSD), and correlation structure were analysed. Overall, SEPs resulting from MEA recordings had higher amplitudes and contained significantly more spectral power, especially at higher frequencies. However, the SNRs were similar between the interfaces. These results demonstrate the feasibility of using µECoG to decode SEPs with wide-range applications in physiology monitoring and brain-computer interfaces.

Comparison of Subdural and Intracortical Recordings of Somatosensory Evoked Responses.

The most characteristic feature of the human electroencephalogram is the peak alpha frequency (PAF). While PAF has been proposed as a biomarker in several diseases and disorders, the disease mechanisms modulating PAF, as well as its physiological substrates, remain elusive. This has partly been due to challenges related to experimental manipulation and invasive procedures in human neuroscience, as well as the scarcity of animal models where PAF is consistently present in resting-state. With the potential inclusion of PAF in clinical screening and decision-making, advancing the mechanistic understanding of PAF is warranted. In this paper, we propose the female Danish Landrace pig as a suitable animal model to probe the mechanisms of PAF and its feature as a biomarker. We show that somatosensory alpha oscillations are present in anesthetized pigs using electrocorticography and intracortical electrodes located at the sensorimotor cortex. This was evident when looking at the time-domain as well as the spectral morphology of spontaneous recordings. We applied the FOOOF-algorithm to extract the spectral characteristics and implemented a robustness threshold for any periodic component. Using this conservative threshold, PAF was present in 18/20 pigs with a normal distribution of the peak frequency between 8-12 Hz, producing similar findings to human recordings. We show that PAF was present in 69.6 % of epochs of approximately six-minute-long resting-state recordings. In sum, we propose that the pig is a suitable candidate for investigating the neural mechanisms of PAF as a biomarker for disease and disorders such as pain, neuropsychiatric disorders, and response to pharmacotherapy.

Article	Year
Electrochemical properties of titanium nitride nerve stimulation electrodes: an in vitro and in vivo study. Frontiers in neuroscience\| PubMed ID: 26300717	2015
Influence of fibrous encapsulation on electro-chemical properties of TiN electrodes. Medical engineering & physics\| PubMed ID: 26997562	2016
Diamond/Porous Titanium Nitride Electrodes With Superior Electrochemical Performance for Neural Interfacing. Frontiers in bioengineering and biotechnology\| PubMed ID: 30525031	2018
The pro-algesic effect of γ-aminobutyric acid (GABA) injection into the masseter muscle of healthy men and women. Scandinavian journal of pain\| PubMed ID: 31430257	2019
Comparison of existing electrode designs for preferential activation of cutaneous nociceptors. Journal of neural engineering\| PubMed ID: 32235064	2020
The effect of peripheral high-frequency electrical stimulation on the primary somatosensory cortex in pigs. IBRO neuroscience reports\| PubMed ID: 34541572	2021
A systematic review of porcine models in translational pain research. Lab animal\| PubMed ID: 34650279	2021
The Use of the Velocity Selective Recording Technique to Reveal the Excitation Properties of the Ulnar Nerve in Pigs. Sensors (Basel, Switzerland)\| PubMed ID: 35009601	2021
On determining the mechanical nociceptive threshold in pigs: a reliability study. Frontiers in pain research (Lausanne, Switzerland)\| PubMed ID: 37265942	2023
Morphology and morphometry of the ulnar nerve in the forelimb of pigs. Anatomia, histologia, embryologia\| PubMed ID: 37715494	2024
High-frequency electrical stimulation increases cortical excitability and mechanical sensitivity in a chronic large animal model. Pain\| PubMed ID: 39133034	2025
Spared ulnar nerve injury results in increased layer III-VI excitability in the pig somatosensory cortex. Lab animal\| PubMed ID: 39349800	2024
Comparison of Subdural and Intracortical Recordings of Somatosensory Evoked Responses. Sensors (Basel, Switzerland)\| PubMed ID: 39517744	2024
Pigs as a translational animal model for the study of peak alpha frequency. Neuroscience\| PubMed ID: 39694317	2025

Suzan Meijs

Biography

JoVE Journal Publications

Other Publications