British Journal of Anaesthesia 114 (6): 979–89 (2015) doi: 10.1093/bja/aev095 Advance Access Publication 7 May 2015 Translational Research
T R A N S L AT I O N A L R E S E A R C H
Electroencephalographic coherence and cortical acetylcholine during ketamine-induced unconsciousness D. Pal1,2, V. S. Hambrecht-Wiedbusch1,2, B. H. Silverstein1 and G. A. Mashour1,2,3, * 1
Department of Anesthesiology, University of Michigan, 7433 Medical Science Building I, 1150 West Medical Center Drive, Ann Arbor, MI 48109-5615, USA, 2Center for Consciousness Science, University of Michigan, Ann Arbor, MI 48109, USA, and 3Neuroscience Graduate Program, University of Michigan, Ann Arbor, MI 48109, USA *Corresponding author. E-mail.
[email protected]
Abstract Background: There is limited understanding of cortical neurochemistry and cortical connectivity during ketamine anaesthesia. We conducted a systematic study to investigate the effects of ketamine on cortical acetylcholine (ACh) and electroencephalographic coherence. Methods: Male Sprague–Dawley rats (n=11) were implanted with electrodes to record electroencephalogram (EEG) from frontal, parietal, and occipital cortices, and with a microdialysis guide cannula for simultaneous measurement of ACh concentrations in prefrontal cortex before, during, and after ketamine anaesthesia. Coherence and power spectral density computed from the EEG, and ACh concentrations, were compared between conscious and unconscious states. Loss of righting reflex was used as a surrogate for unconsciousness. Results: Ketamine-induced unconsciousness was associated with a global reduction of power (P=0.02) in higher gamma bandwidths (>65 Hz), a global reduction of coherence (P≤0.01) across a broad frequency range (0.5–250 Hz), and a significant increase in ACh concentrations (P=0.01) in the prefrontal cortex. Compared with the unconscious state, recovery of righting reflex was marked by a further increase in ACh concentrations (P=0.0007), global increases in power in theta (4–10 Hz; P=0.03) and low gamma frequencies (25–55 Hz; P=0.0001), and increase in power (P≤0.01) and coherence (P≤0.002) in higher gamma frequencies (65–250 Hz). Acetylcholine concentrations, coherence, and spectral properties returned to baseline levels after a prolonged recovery period. Conclusions: Ketamine-induced unconsciousness is characterized by suppression of high-frequency gamma activity and a breakdown of cortical coherence, despite increased cholinergic tone in the cortex. Key words: acetylcholine; electroencephalography; ketamine; microdialysis; prefrontal cortex; unconsciousness
Ketamine is a unique anaesthetic drug that does not conform to most mechanistic frameworks of anaesthetic-induced unconsciousness.1 Unlike many general anaesthetics, the γ-aminobutyric acid (GABA) receptor is not the primary molecular target of ketamine.2–4 At a systems neuroscience level, ketamine does
not appear to activate the sleep-promoting ventrolateral preoptic nucleus,5 as typical GABAergic anaesthetics do.5–7 Instead, ketamine activates the norepinephrine-producing locus coeruleus5 and appears to depend, in part, on noradrenergic transmission.8 At the neurophysiological level, ketamine enhances higher
Accepted: February 22, 2015 © The Author 2015. Published by Oxford University Press on behalf of the British Journal of Anaesthesia. All rights reserved. For Permissions, please email:
[email protected]
979
980
| Pal et al.
Editor’s key points • Ketamine, which acts by a unique non-GABAergic mechanism, might have distinct effects on long-range cerebrocortical interactions compared with other general anaesthetics. • The effects of ketamine on electroencephalographic coherence, cortical acetylcholine concentrations, and behaviour were studied in rats. • Ketamine-induced loss of consciousness was associated with global reductions in gamma power and cortical coherence similar to other general anaesthetics.
frequency electroencephalographic activity,9 10 which can confound the algorithms in processed electroencephalographic devices intended for monitoring anaesthetic depth.10 11 Recent studies from our laboratory suggest that ketamine shares a network-level property with GABAergic drugs by disrupting information transfer,9 phase relationships,12 or both across the cortex. These findings are consistent with a study of ketamine in a cortical slice model that identified uncoupling of long-range corticocortical interactions.13 However, studies of ketamine and cortical or thalamocortical connectivity in intact animal models are typically conducted with co-administration of xylazine, an α2-adrenergic agonist.14 15 Furthermore, there have been no carefully controlled studies of ketamine-induced unconsciousness in animal models that link disruptions of neurophysiological coupling with neurochemical events. Studies of ketamine and neurochemistry have focused on acetylcholine (ACh) but (i) without concomitant neurophysiological recordings,16 17 (ii) using subanaesthetic doses of ketamine,16 17 or (iii) without formal testing of loss of righting reflex (LORR) as a surrogate of anaesthetic-induced unconsciousness.16 To address this gap in knowledge, the objective of the present study was to identify the relationship of electroencephalographic coherence, cortical ACh, and the state of ketamine-induced unconsciousness.
Methods Experiments were conducted on adult (3- to 5-month-old) male Sprague–Dawley rats (n=11; Charles River Laboratories, Inc., Kingston, NY, USA) maintained on a 12 h light–12 h dark cycle (lights on at 06.00 h) with ad libitum food and water. The experimental procedures were approved by the University of Michigan Committee on Use and Care of Animals and were in compliance with the Guide for the Care and Use of Laboratory Animals (8th Edition, The National Academies Press, Washington, DC, USA) and the ARRIVE guidelines.
Surgical procedures Under surgical levels of isoflurane anaesthesia, rats were implanted with screw electrodes to record electroencephalogram (EEG) from frontal (Bregma: anterior 3.0 mm, lateral 2.5 mm), parietal (Bregma: posterior 4.0 mm, lateral 2.5 mm), and occipital areas (Bregma: posterior 8.0 mm, lateral 2.5 mm). A screw electrode was implanted over the nasal commissure to serve as a reference electrode. In addition, a craniotomy was performed over the prefrontal cortex (PFC) (Bregma: anterior 3.0 mm, lateral 0.5 mm, ventral 4.0 mm),18 and a CMA/11 guide cannula (CMA Microdialysis, Harvard Apparatus, Holliston, MA, USA) was implanted 1.0 mm above the target area. Buprenorphine
hydrochloride (Buprenex®; Reckitt Benckiser Pharmaceuticals Inc., Richmond, VA, USA) was used for pre- and postsurgical analgesia (0.01 and 0.03 mg kg−1, s.c., respectively), and a presurgical single dose of the antibiotic cefazolin (20 mg kg−1, s.c.; WestWard Pharmaceutical Corp., Eatontown, NJ, USA) was administered. The EEG electrodes were mated with an electrode pedestal (Plastics One, Roanoke, VA, USA), and the entire assembly along with the guide cannula was affixed to the cranium using dental acrylic.
Electroencephalographic data acquisition The electrode over the nasal commissure was used as a reference for monopolar EEG recordings from the frontal, parietal, and occipital cortices. The choice of monopolar EEG recordings was based on previously published animal studies19–21 and on a study from our laboratory demonstrating that monopolar reference is best suited to detect genuine EEG phase synchronization.22 Electroencephalographic signals were amplified (×5000) and filtered (0.1–300 Hz) on a Grass Model 15 LT system (15A54 Quad Amplifier, Warwick, RI, USA). Data were digitized at 1 kHz using an MP150 system and AcqKnowledge data acquisition software (version 4.1; Biopac Systems, Inc., Goleta, CA, USA).
Coherence and power spectral analysis Data were first down-sampled to 500 Hz to reduce computation time, and an IIR notch filter was applied to remove 60 Hz line noise. We calculated coherence across cortical electrodes at individual frequencies from 0.5 to 250 Hz (in 0.5 Hz intervals) as magnitude squared coherence using the ‘mscohere.m’ function in the MATLAB Signal Processing Toolbox (MathWorks Inc., Natick, MA, USA).21 To control for spurious coherence, the surrogate data method was used, wherein phases were randomized but the spectral content of the signals was maintained.23 These estimates of spurious coherence were then statistically compared with the empirical data. Furthermore, frequencies at which obvious artifact was present on the individual spectrogram or coherogram were excluded from quantitative analysis and statistical comparison. Absolute power spectral density (PSD) between 0.5 and 250 Hz was calculated based on the short-time Fourier transform using the ‘spectrogram.m’ function in the MATLAB Signal Processing Toolbox.21 Relative power was calculated for each epoch by dividing the mean absolute power of each frequency band by the total power across the entire frequency range. Coherence and PSD were calculated for the following frequency bands: delta (δ: 0.5–4 Hz), theta (θ: 4–10 Hz), alpha (α: 10–15 Hz), beta (β: 15–25 Hz), low gamma (γ1: 25–55 Hz), medium gamma (γ2: 65– 125 Hz), high gamma (γ3: 125–175 Hz), and ultrahigh gamma (γ4: 185–250 Hz). The data are reported as changes in global coherence, which was obtained by averaging the coherence values for individual channel pairs. Likewise, PSD values for individual channels were averaged and reported as changes in global PSD.
Quantification of acetylcholine release in prefrontal cortex A CMA/11 microdialysis probe (1 mm cuprophane membrane, 0.24 mm diameter, 6 kDa) was perfused continuously with Ringer’s solution (147 mM NaCl, 2.4 mM CaCl2, 4.0 mM KCl, and 10 µM neostigmine; pH 5.8–6.2) at 2.0 µl min−1 using a CMA/400 syringe pump. Microdialysis samples were collected every 12.5
Ketamine-induced unconsciousness
min, which yielded 25 µl of dialysate, out of which 22 µl was used for ACh quantification. Acetylcholine concentrations were quantified using high-performance liquid chromatography coupled with electrochemical detection (BASi Inc., West Lafayette, IN, USA and Showa Denko America, Inc., New York, NY, USA). Chromatograms were digitized and quantified using LC Real Time Analysis Program (Showa Denko America, Inc.) and a sevenpoint ACh–choline standard curve (from 0.05 to 1.0 pmol).
Experimental design The experimental design is illustrated in Fig. 1. All experiments were conducted between 10.00 and 18.00 h. Rats were permitted 7–10 days of postsurgical recovery, during which they were conditioned to the EEG recording and microdialysis set-up. On the day of the experiment, rats were connected to the EEG recording system, and a microdialysis probe was lowered into the PFC. After allowing 45 min to ensure stable ACh concentrations, simultaneous monopolar EEG recording and ACh sample collection were started. Rats were kept awake by gentle tapping on the recording chamber to hold the behavioural state constant. Data collection was stopped after collection of the fourth ‘Wake’ ACh sample, and animals were given a single intraperitoneal dose (150 mg kg−1) of ketamine (Ketamine Hydrochloride; Hospira. Inc., Lake Forest, IL, USA). The dose for ketamine was based on previous literature,24 as well as dose–response experiments conducted in our laboratory, and titrated to achieve LORR, which occurred within 3–5 min of ketamine injection. Immediately after LORR, a rectal probe (Model 7001H; Physitemp Instruments, Clifton, NJ, USA) was inserted to monitor body temperature, and a foot sensor (MouseOx; Starr Life Science Corp., Oakmont, PA, USA) was positioned to monitor heart rate and oxygen saturation. To obtain a pure ACh sample corresponding to the unconscious state, data collection was resumed after LORR and after an additional 7 min elapsed in order to purge the dead space volume in the ACh collection tubing. Three of 11 rats were excluded from the data analysis because they continued to show whisker or slight body movements, or both, after LORR. In the remaining eight rats, EEG recording and ACh sample collection were continued until three ACh samples were obtained after return of righting reflex (RORR), a surrogate for recovery from anaesthesia in rodents. Additionally, in three of these eights rats, data collection was extended until ACh concentrations returned to those observed
Ketamine (150 mg kg–1, i.p.)
Awake and freely moving
Continuous EEG recording and ACh sample collection every 12.5 min
Induction (no data collection)
| 981
during the waking state. The location of the dialysis probe was confirmed histologically (see Supplementary methods).
Statistical analyses The Center for Statistical Consultation and Research at the University of Michigan was consulted for data analyses. A priori power analysis (nQuery Advisor+nTerim; Statistical Solutions Ltd, Boston, MA, USA) was conducted to design the study with >80% power at an α value of 0.05. Although EEG data and ACh sample collection were continuous linear processes, the data set was divided into 12.5 min epochs corresponding to the time required for collection of each dialysis sample. Five epochs reflecting different behavioural states were selected for statistical comparisons, as follows: (i) ‘Wake’, the last epoch of the awake period; (ii) ‘Ketamine’, the first epoch after both LORR and exclusion of microdialysis dead space volume; (iii) ‘Pre-RORR’, one epoch before RORR; (iv) ‘Post-RORR’, the first epoch after RORR; and (v) ‘Recovery’, the last epoch from rats with the prolonged/ extended recovery. Acetylcholine concentrations, and coherence and PSD across each frequency band for Wake, Ketamine, PreRORR, and Post-RORR were compared using repeated measures analysis of variance () with Tukey’s multiple comparisons test. Acetylcholine and EEG parameters for the Recovery epoch were compared with the Wake epoch using a one-tailed Wilcoxon test. Surrogate EEG data were compared with the raw unshuffled EEG data using Student’s paired t-test. Changes in blood oxygen saturation, heart rate, and rectal temperature between Ketamine and Pre-RORR epochs were compared using a one-tailed Wilcoxon test. Statistical comparisons were performed with Graph Pad Prism 6.05 (Graph Pad Software, Inc., La Jolla, CA, USA).
Results The site of microdialysis was confirmed to be within the PFC for all rats (Fig. 2 and ). Representative ACh peaks from a Wake sample and a known standard are shown in Fig. 2.
Effect of ketamine on behaviour Eight of 11 rats were completely immobile and maintained LORR without any body movements during the first 12.5 min epoch after ketamine administration; the three rats showing movements were excluded from analysis. Behavioural observations
LORR (+7 min)
Post-ketamine (until RORR)
RORR
Post-RORR
Continuous EEG recording and ACh sample collection every 12.5 min
Fig 1 Schematic diagram illustrating the temporal course of experimental intervention and data collection. ACh, acetylcholine; i.p., intraperitoneal; LORR, loss of righting reflex; RORR, return of righting reflex.
982
| Pal et al.
A
1 1mm mm
B 4.20 mm
PrL
3.72 mm
3.24 mm
PrL
PrL
C
Wake sample
1.2
Ch
0.5 pmol ACh/Ch standard
Current (nA)
1 0.8
ACh
0.6 0.4 0.2 0 0
1.5
3
4.5
6
7.5
9
Time (minutes) Fig 2 () Cresyl violet-stained representative coronal brain section through prefrontal cortex showing the dialysis probe track and the site of microdialysis. Arrow indicates the ventral tip of the dialysis membrane (1 mm long). () Coronal brain section drawings from the rat brain atlas of Paxinos and Watson18 to illustrate the location of microdialysis probes (vertical cylinders) within the prefrontal cortex. The numbers within coronal sections show the anterior–posterior location of each coronal section relative to bregma. () Representative acetylcholine (ACh) chromatogram showing the signal-to-noise ratio and the retention time for ACh and choline (Ch) peaks. PrL, prelimbic area in the prefrontal cortex.
were accompanied by EEG and ACh data collection (Ketamine epoch). After the first epoch after ketamine administration, rats exhibited varying degrees of purposeful movements in limbs, head, and torso until RORR. The Post-RORR period was
characterized by continuous circling movements. These stereotyped movements ceased, and rats returned to normal, calm behaviour after a prolonged recovery period (14–22 epochs, or 175–275 min Post-RORR), which was also marked by return of
Ketamine-induced unconsciousness
A
Ketamine
Wake LORR
Pre-RORR
Post-RORR
| 983
Recovery
RORR
250
Frequency (Hz)
0.9 200
0.8
150
0.7 0.6
100
0.5 0.4 0.3 0.2 0.1
50 0
B
0.90 * * ##
Coherence (Mean ± SEM)
0.75 0.60
* *
*
* * *
## *
0.45
*
*
* * ##
* * * ##
* * *
0.30
*
*
*
0.15 0.00
0.5–4 d
C
4–10 q
10–15 a
LORR
15–25 b
Hz
25–55 g1
65–125 g2
125–175 g3
185–250 g4
RORR
250 0.9 0.8 0.7 0.6 0.5 0.4
150 100
0.3 0.2 0.1
50 0
D
70
Relative power (%) (Mean ± SEM)
50 40
9.0
* # #*
60 * #* #
7.5
30 * * ##
20 10 0
* 0.5–4 d
4–10 q
6.0
* #
##
* *#
10–15 a
3.0
* # #* 15–25 b
4.5
* #
Relative power (%) (Mean ± SEM)
Frequency (Hz)
200
1.5
Hz
25–55 g1
* 65–125 g2
* 125–175 g3
## 185–250 g4
0.0
Fig 3 Effect of ketamine on coherence and power spectral density (PSD). () Representative coherogram showing a significant decrease in coherence during ketamine-induced unconsciousness (Ketamine) and the appearance of high-frequency gamma coherence during emergence and after the return of righting reflex (RORR). () Statistical comparisons (repeated measures ) of changes in coherence between Wake (blue), Ketamine-induced unconsciousness (gold), Pre-RORR (Pink), and Post-RORR (green; n=8 animals). The fifth bar (orange) represents the Recovery epoch from the rats with extended recovery and was compared (one-tailed Wilcoxon text) with the Wake state from the same rats. () Representative spectrogram showing a significant decrease in power in higher gamma frequencies during ketamine-induced unconsciousness (Ketamine) and the appearance of high-frequency gamma power during emergence and after RORR. () Statistical comparisons (repeated measures ) of changes in PSD between Wake (blue), Ketamine-induced unconsciousness (gold), Pre-RORR (Pink), and Post-RORR (green; n=8 animals). The fifth bar (orange) represents the Recovery epoch from the rats with extended recovery and was compared (onetailed Wilcoxon text) with the Wake state from the same rats. The right axis applies only to the grey shaded area. *P
