The sheep used in this study were locally sourced Welsh mountain ewes (n = 5) and merino ewes (n = 7). All of the sheep were neurologically and genetically normal. All procedures involving the Welsh mountain sheep were conducted at the University of Cambridge in accordance with the UK Animals (Scientific Procedures) Act (1986), and the University of Cambridge ethical review board. The Welsh mountain sheep were reared in Cambridge, UK. At surgery their age was approximately 2 years and 3 months, and they weighed 46 ± 3.2 kg (SEM). All procedures involving the merino sheep were conducted at the Preclinical Imaging and Research Laboratories (PIRL) of the South Australian Health and Medical Research Institute (SAHMRI) and followed the requirements of the SAHMRI Animal Ethics Committee including the Australian Code for the Care and Use of Animals for Scientific Purposes (8th Edition 2013). These sheep were part of a group that included transgenic animals. Accordingly, although the sheep that were used in this study were genetically normal, all handling of these sheep conformed to physical containment conditions as approved by the Institutional Biosafety Committee and the Office of the Gene Technology Regulator (OGTR, Australia). At surgery the merino sheep were approximately 5 years of age, and their weight was 85 ± 2.5 kg.
Surgical and anaesthetic procedures are described in detail in Perentos et al.19,20. Briefly, anaesthesia was induced using alfaxalone (3 mg/kg; Alfaxan®, Jurox, U.K., i.v.) for the Welsh mountain sheep, and diazepam (0.4 mg/kg) and ketamine (5 mg/kg, i.v., Ketaset, Zoetis Inc., New Jersey, USA) for the merino sheep. The upper airway was intubated with an endotracheal tube, and general anaesthesia was maintained during surgery with isoflurane in oxygen and nitrous oxide for the Welsh mountain sheep and isoflurane in oxygen for the merino sheep. Isoflurane was maintained at 2–3%, end-tidal CO 2 at 25–30 mmHg and mean arterial blood pressure at 70–90 mmHg. Intravenous fluids were supplied at a rate of 5 ml/kg/h (lactated Ringers, Hartmann’s Solution 11 by Aquapharm). Vital functions were recorded at 5 min intervals, and blood gases sampled every 30 min throughout the procedure.
Once general anaesthesia was achieved, the sheep was positioned in sternal recumbency for surgery. The head was fixed in a stereotaxic frame (Kopf Instruments, USA). An incision was made at the midline between the eyes, and extended to the external occipital crest of the skull. The scalp was retracted using blunt dissection. The periosteum and all traces of connective tissue and fat were cleared from the skull, allowing visualisation of the bregma, here defined as the point of intersection of the midline scull suture and the transverse suture between the frontal and parietal bones. Electrode positions are defined relative to the position of the bregma. Subdural electrodes (3 mm diam. × 1 mm deep Ag/AgCl disc, NDimension (Science and Engineering) Ltd., Cambridge, UK) were implanted via craniotomies ~25 mm, ~15 mm and ~5 mm anterior, and ~10 mm posterior to bregma, and 10 mm lateral to the midline over both hemispheres. A reference electrode was inserted at the midline 10 mm posterior to bregma, and 2 ground wires were attached to the skull using stainless steel screws. Stainless steel coil electrodes were implanted bilaterally in the dorsal splenius muscles of the neck for recording EMG data, and Ag/AgCl electrodes were positioned in the inner and outer canthi of both eyes for recording the EOG. When all electrodes and wires were in place, a rigid cap was formed using dental acrylic containing Gentamicin (DePuy, Johnson & Johnson) to seal all of the components and craniotomies. Leads from all electrodes terminated at a multi-pin connector (Omnetics Connector Corporation, MN, USA). The connector was either exteriorised near the occipital crest (1 sheep), or housed in a 3D-printed polyamide (nylon) chamber (G.E. Baker (UK) Ltd) fixed to the skull. The chamber had a screw cap that allowed easy access to the connector post-surgery.
During surgery, a non-steroidal anti-inflammatory (carprofen) was administered. To manage possible post-operative infection, pain and/or inflammation, all sheep received buprenorphine at the end of surgery, and daily carprophen and antibiotics for three days after surgery. If necessary, additional buprenorphine was administered in the first three days according to veterinary instruction.
Recordings from the Welsh mountain sheep in this study were conducted 1–10 months post-surgery. Those from the merino sheep took place 6–10 months post-surgery.
EEG data were collected wirelessly using a MCS advanced wireless system (Multichannel Systems Gmbh, Germany). At the commencement of each recording session, each sheep was fitted with a 16 channel wireless head-stage (W2100-HS16). Recordings could be made simultaneously from up to 8 sheep at a time. Data were collected at a sampling frequency of 1 kHz on each channel.
Ketamine was delivered to individual sheep suspended in veterinary slings with their hooves elevated off the ground. Sheep quickly became accustomed to being restrained in this way, enabling a baseline recording period of 10–15 minutes preceding ketamine delivery. Recordings were typically made from 2–4 sheep at a time.
For the Welsh mountain sheep, ketamine was delivered intramuscularly as a single injection to the gluteus muscle (left or right). Merinos received ketamine intravenously via a jugular catheter. Catheters were fitted at least 24 h before ketamine treatment. To ensure patency, the catheters were flushed daily with heparinised saline. Ketamine injections were delivered between 10:00 and 16:00.
Ketamine (100 mg/ml ketamine hydrochloride solution; Ketaset, Zoetis Inc., New Jersey, USA) was administered when the sheep had remained calm in the sling for a minimum period of 15 min. All sheep received a separate dose of 3, 6, 12 and 24 mg/kg body weight. The doses were given in the same order to all the sheep so that any possible hangover effects would be similar for all animals. The order was 6, 12, 3, 24 mg/kg, with at least 2 days between doses. One merino sheep received a second dose of 24 mg/kg 3 days after the first. In this case, while both recordings are described here, only the second recording was included in subsequent statistical analysis. All of the Welsh mountain sheep received two additional intramuscular doses of 0.5 and 1 mg/kg ketamine.
Drug trials were performed in separate recording sessions with a minimum of 24 h between doses. During recording after ketamine delivery (when two operatives were present), the sheep remained undisturbed (with a single observer present) for a minimum period of 1 h, but otherwise as long as was necessary. When they were responsive and deemed competent to stand and move normally, they were returned to their home pen. This was never greater than 90 min after the administration of ketamine.
EEG data were down-sampled to 250 Hz to promote efficient data processing. In subsequent analyses, EEG data were considered in terms of the contribution of identified frequency bands to the overall frequency spectrum. These frequency bands reflect oscillatory activity recognised as delta (0–4 Hz), theta (4–9 Hz), alpha (9–14 Hz), beta (14–35 Hz) and gamma (35–125 Hz). The theta and gamma bands were further sub-divided into low and high theta (4–6 Hz, 6–9 Hz respectively), and into low, mid and high gamma (35–55 Hz, 55–85 Hz, 85–125 Hz respectively).
In an initial pre-processing stage, a global reference was generated for each sheep by computing the instantaneous average of all channels of EEG data. This global reference was then subtracted from each channel of EEG data. Thus, each EEG channel was represented as a differential signal relative to the global reference. This stage was conducted to minimise contamination of the EEG data by variation at the reference electrode.
The power of each oscillatory frequency in the EEG was computed using a fast Fourier transform (FFT). The power spectrum generated by this analysis reflects the contribution of each frequency to the total power of the EEG. For each recording, a temporal representation of the power spectrogram was produced in the form of a spectrogram. In the spectrograms, EEG power as a function of mV2 is displayed according to the colour spectrum in the key in Fig. 1B. In some analyses, the root mean square (RMS) of the EEG was computed. In this measure, the mean amplitude of the signal was computed in 1 s windows. This value was then squared and the square root of the resulting value calculated, thereby providing a finite positive measure of EEG activity.
Latencies of responses induced by ketamine delivery were computed for each EEG band as the time from start of ketamine delivery to the point at which the power in a given band increased by 100% relative to the stable pre-drug baseline. This was also computed for the change in total power across the full frequency spectrum (0–125 Hz).
Waveform correlation analysis was performed for certain comparisons between certain traces. This measures the similarity of two waveforms. Starting with the two channels aligned, each data point amplitude in the reference waveform is multiplied by each data point amplitude in the other, and summing the results. The reference waveform is then advanced by one sample point and the process repeated until all points in the reference channel are accommodated. In the resulting correlation waveform, a symmetrical upward peak implies that the two waveforms are synchronised and in phase, while an inverted peak implies that the two waveforms are in anti-phase.
EOG and EMG traces were down-sampled to 250 Hz to promote efficient analysis. For both channels of EMG data, an average reference was computed as the average of the two EMG channels and all EEG channels. This average reference was then subtracted from each channel of EMG data. This was found to provide the most stable representation of muscle activity. A single EMG channel was then computed as the differential between the two average-referenced EMG channels. For the EOG data, a global reference was computed as the average of all four EOG channels. This average reference was then subtracted from each individual channel of EOG data. For certain analyses the EMG and EOG data were RMS transformed, as described above for the EEG, as a measure of absolute muscle activity.
In some of the trials with intravenous ketamine the electrocardiogram (ECG) was recorded. These recordings were made for the merino sheep implanted in the second surgery session. This was achieved by attaching a single self-adhesive ECG patch to the thorax next to the left forelimb. This area is relatively free from wool and provides a good surface from which to acquire these data. During acquisition the ECG data were referenced to the same intracerebral reference as the other electrophysiological data. During offline processing an average reference was computed as the average of the ECG signal and all of the EEG channels, and this was subtracted from the ECG signal.
Repeated measures analyses of variance (ANOVA) were performed using SPSS (IBM Corp., Released 2016, IBM SPSS Statistics for Windows, Version 24.0, Armonk, NY: IBM Corp.) with different EEG frequency ranges entered as a within subjects variable. These analyses incorporated between subjects factors sufficient to account for each data instance: dose, trial, route of administration, hemisphere (left/right), brain region. In these analyses, data were pooled across factors which yielded no significant effect or interactions in full factorial analysis. Statistical significance was defined as P < 0.05.
schoobydoo42 on June 23rd, 2020 at 14:51 UTC »
I am an ER physician and we very commonly give ketamine for dissociative procedural sedation, as a sedative for people in a dangerous state of excited delirium, and in low doses for pain control. I would categorize three dose-dependent effects of ketamine as follows: 1) low-dose will provide pain control as well as a mild calming effect. 2) the K-hole dose. This is essentially like a hallucinogenic state where you may have an out-of-body experience and remember it. Sometimes people can have a very bad trip, though. 3) the full dissociative dose, in which people do go into a dream-like state but often do not remember their experiences. They will experience limited pain and, if a painful procedure is done, they will not remember it.
This article is interesting because the dose they used to suppress EEG function in the sheep models I would consider to be an insanely high dose--24 mg/kg IV. A typical IV dose of ketamine in humans used for dissociative procedural sedation in medicine is around 1-2 mg/kg. The low doses we use for pain are more like 0.25-0.35 mg/kg. Frequently, my patients who get ketamine for procedural sedation will not remember anything from it. The article did mention that 24 mg/kg is on the "lower end" for recreational use. I have no idea what recreational users take, but I'd be curious how anyone could function whatsoever, or remember their experiences, on that high of a dose of ketamine. Guess it's time to head to Erowid. I gotta say, this is a very useful site for us ER physicians since we see a lot of drug intoxication and states of drug-induced delirium. It provides great info on the toxidromes of a lot of drugs we might see patients take.
vonloan on June 23rd, 2020 at 13:54 UTC »
Seemingly nobody here has read the article or knows what a normal dose of ketamine is. someone with no ketamine tolerance will likely k-hole with a racemic of around 150mg intranasally, this is around 1-3mg/kg.
In this study they used intravenous and intramuscular injections of all the way to 24mg/kg. That would be around 1.7 grams for me personally, and that's not even looking at the fact that IV doses hit harder and faster than intranasally. This study is not that useful for looking at medical doses or even recreational doses. 24mg/kg is an ungodly amount, even for surgeries the initial IV dose is usually 1-4.5mg/kg.
While this study is interesting, it is imo not relevant whatsoever to medical or recreational dosages.
TurnsOutImAScientist on June 23rd, 2020 at 13:40 UTC »
They didn't measure EEG in brain structures outside the cortex; the word "altogether" is highly misleading here. Almost certainly still plenty of neuronal activity happening, otherwise the animals would have died. They also didn't measure single neurons (EEG is a more collective measure), so we don't know if activity stopped or merely desynchronized.
What I find interesting here is a strong suggestion that subjective consciousness isn't abolished despite lack of cortical rhythms, which in turn supports the idea that subcortical structures are more fundamentally important for consciousness itself; cortex might be more important for creating its contents.