Effect of Propofol on Heart Rate and Its Coupling to Cortical Slow Waves in Humans.

BACKGROUND: Propofol causes significant cardiovascular depression and a slowing of neurophysiological activity. However, literature on its effect on the heart rate remains mixed, and it is not known whether cortical slow waves are related to cardiac activity in propofol anesthesia. METHODS: The authors performed a secondary analysis of electrocardiographic and electroencephalographic data collected as part of a previously published study where n = 16 healthy volunteers underwent a slow infusion of propofol up to an estimated effect-site concentration of 4 µg/ml. Heart rate, heart rate variability, and individual slow electroencephalographic waves were extracted for each subject. Timing between slow-wave start and the preceding R-wave was tested against a uniform random surrogate. Heart rate data were further examined as a post hoc analysis in n = 96 members of an American Society of Anesthesiologists Physical Status II/III older clinical population collected as part of the AlphaMax trial. RESULTS: The slow propofol infusion increased the heart rate in a dose-dependent manner (mean ± SD, increase of +4.2 ± 1.5 beats/min/[µg ml-1]; P < 0.001). The effect was smaller but still significant in the older clinical population. In healthy volunteers, propofol decreased the electrocardiogram R-wave amplitude (median [25th to 75th percentile], decrease of -83 [-245 to -28] µV; P < 0.001). Heart rate variability showed a loss of high-frequency parasympathetic activity. Individual cortical slow waves were coupled to the heartbeat. Heartbeat incidence peaked about 450 ms before slow-wave onset, and mean slow-wave frequency correlated with mean heart rate. CONCLUSIONS: The authors observed a robust increase in heart rate with increasing propofol concentrations in healthy volunteers and patients. This was likely due to decreased parasympathetic cardioinhibition. Similar to non-rapid eye movement sleep, cortical slow waves are coupled to the cardiac rhythm, perhaps due to a common brainstem generator.

Propofol-induced Unresponsiveness Is Associated with a Brain Network Phase Transition.

BACKGROUND: The wakeful brain can easily access and coordinate a large repertoire of different states-dynamics suggestive of "criticality." Anesthesia causes loss of criticality at the level of electroencephalogram waveforms, but the criticality of brain network connectivity is less well studied. The authors hypothesized that propofol anesthesia is associated with abrupt and divergent changes in brain network connectivity for different frequencies and time scales-characteristic of a phase transition, a signature of loss of criticality. METHODS: As part of a previously reported study, 16 volunteers were given propofol in slowly increasing brain concentrations, and their behavioral responsiveness was assessed. The network dynamics from 31-channel electroencephalogram data were calculated from 1 to 20 Hz using four phase and envelope amplitude-based functional connectivity metrics that covered a wide range of time scales from milliseconds to minutes. The authors calculated network global efficiency, clustering coefficient, and statistical complexity (using the Jensen-Shannon divergence) for each functional connectivity metric and compared their findings with those from an in silico Kuramoto network model. RESULTS: The transition to anesthesia was associated with critical slowing and then abrupt profound decreases in global network efficiency of 2 Hz power envelope metrics (from mean ± SD of 0.64 ± 0.15 to 0.29 ± 0.28 absolute value, P < 0.001, for medium; and from 0.47 ± 0.13 to 0.24 ± 0.21, P < 0.001, for long time scales) but with an increase in global network efficiency for 10 Hz weighted phase lag index (from 0.30 ± 0.20 to 0.72 ± 0.06, P < 0.001). Network complexity decreased for both the 10 Hz hypersynchronous (0.44 ± 0.13 to 0.23 ± 0.08, P < 0.001), and the 2 Hz asynchronous (0.73 ± 0.08 to 0.40 ± 0.13, P < 0.001) network states. These patterns of network coupling were consistent with those of the Kuramoto model of an order-disorder phase transition. CONCLUSIONS: Around loss of behavioral responsiveness, a small increase in propofol concentrations caused a collapse of long time scale power envelope connectivity and an increase in 10 Hz phase-based connectivity-suggestive of a brain network phase transition.

Automatic decomposition of electrophysiological data into distinct nonsinusoidal oscillatory modes.

Neurophysiological signals are often noisy, nonsinusoidal, and consist of transient bursts. Extraction and analysis of oscillatory features (such as waveform shape and cross-frequency coupling) in such data sets remains difficult. This limits our understanding of brain dynamics and its functional importance. Here, we develop iterated masking empirical mode decomposition (itEMD), a method designed to decompose noisy and transient single-channel data into relevant oscillatory modes in a flexible, fully data-driven way without the need for manual tuning. Based on empirical mode decomposition (EMD), this technique can extract single-cycle waveform dynamics through phase-aligned instantaneous frequency. We test our method by extensive simulations across different noise, sparsity, and nonsinusoidality conditions. We find itEMD significantly improves the separation of data into distinct nonsinusoidal oscillatory components and robustly reproduces waveform shape across a wide range of relevant parameters. We further validate the technique on multimodal, multispecies electrophysiological data. Our itEMD extracts known rat hippocampal θ waveform asymmetry and identifies subject-specific human occipital α without any prior assumptions about the frequencies contained in the signal. Notably, it does so with significantly less mode mixing compared with existing EMD-based methods. By reducing mode mixing and simplifying interpretation of EMD results, itEMD will enable new analyses into functional roles of neural signals in behavior and disease.NEW & NOTEWORTHY We introduce a novel, data-driven method to identify oscillations in neural recordings. This approach is based on empirical mode decomposition and reduces mixing of components, one of its main problems. The technique is validated and compared with existing methods using simulations and real data. We show our method better extracts oscillations and their properties in highly noisy and nonsinusoidal datasets.

Contributions of synaptic and astrocyte physiology to the anaesthetised encephalogram revealed using a computational model.

BACKGROUND: General anaesthesia is known to enhance inhibitory synaptic transmission to produce characteristic effects on the EEG and reduction in brain metabolism secondary to reduced neuronal activity. Evidence suggests that anaesthesia might have a direct effect on synaptic metabolic processes, and this relates to anaesthesia sensitivity. We explored elements of synaptic transmission looking for possible contributions to the anaesthetised EEG and how it may modulate anaesthesia sensitivity. METHODS: We developed a Hodgkin-Huxley-type neural network computer simulation capable of mimicking anaesthetic prolongation of gamma-aminobutyric acid (GABA)ergic inhibitory postsynaptic potentials (IPSPs), and capable of altering postsynaptic ion homeostasis and neurotransmitter recycling. We examined their interactions on simulated electrocorticography (sECoG), and compared these with published anaesthesia EEG spectra. RESULTS: The sECoG spectra from the model were comparable with published normal awake EEG spectra. Prolongation of IPSP duration in the model caused inhibition of high frequencies and saturation of low frequencies with a peak in keeping with current evidence. IPSP prolongation alone was unable to reproduce alpha rhythms or the generalised increase in EEG power found with anaesthesia. Adding inhibition of postsynaptic ion homeostasis to IPSP prolongation helped retain alpha rhythms, increased sECoG power, and antagonised the slow-wave saturation peak in a dose-dependent fashion that appeared dependent on the postsynaptic membrane potential, providing a plausible mechanism for how metabolic changes can modulate anaesthesia sensitivity. CONCLUSIONS: Our model suggests how metabolic processes can modulate anaesthesia and produce non-receptor dependent drug sensitivity.

The Effect of Congenital and Acquired Bilateral Anophthalmia on Brain Structure.

The aim of this study was to compare the pattern of changes in brain structure resulting from congenital and acquired bilateral anophthalmia. Brain structure was investigated using 3T magnetic resonance imaging (MRI) in Oxford (congenital) or Manchester (acquired). T1-weighted structural and diffusion-weighted scans were acquired from people with anophthalmia and sighted control participants. Differences in grey matter between the groups were quantified using voxel-based morphometry and differences in white matter microstructure using tract-based spatial statistics. Quantification of optic nerve volume and cortical thickness in visual regions was also performed in all groups. The optic nerve was reduced in volume in both anophthalmic populations, but to a greater extent in the congenital group and anophthalmia acquired at an early age. A similar pattern was found for the white matter microstructure throughout the occipitotemporal regions of the brain, suggesting a greater reduction of integrity with increasing duration of anophthalmia. In contrast, grey matter volume changes differed between the two groups, with the acquired anophthalmia group showing a decrease in the calcarine sulcus, corresponding to the area that would have been peripheral primary visual cortex. In contrast, the acquired anophthalmia group showed a decrease in grey matter volume in the calcarine sulcus corresponding to the area that would have been peripheral primary visual cortex. There are both qualitative and quantitative differences in structural brain changes in congenital and acquired anophthalmia, indicating differential effects of development and degeneration.

Granger Causality of the Electroencephalogram Reveals Abrupt Global Loss of Cortical Information Flow during Propofol-induced Loss of Responsiveness.

BACKGROUND: It is a commonly held view that information flow between widely separated regions of the cerebral cortex is a necessary component in the generation of wakefulness (also termed "connected" consciousness). This study therefore hypothesized that loss of wakefulness caused by propofol anesthesia should be associated with loss of information flow, as estimated by the effective connectivity in the scalp electroencephalogram (EEG) signal. METHODS: Effective connectivity during anesthesia was quantified by applying bivariate Granger to multichannel EEG data recorded from 16 adult subjects undergoing a slow induction of, and emergence from, anesthesia with intravenous propofol. During wakefulness they were conducting various auditory and motor tasks. Functional connectivity using EEG coherence was also estimated. RESULTS: There was an abrupt, substantial, and global decrease in effective connectivity around the point of loss of responsiveness. Recovery of behavioral responsiveness was associated with a comparable recovery in information flow pattern (expressed as normalized values). The median (interquartile range) change was greatest in the delta frequency band: decreasing from 0.15 (0.21) 2 min before loss of behavioral response, to 0.06 (0.04) 2 min after loss of behavioral response (P < 0.001). Regional decreases in information flow were maximal in a posteromedial direction from lateral frontal and prefrontal regions (0.82 [0.24] 2 min before loss of responsiveness, decreasing to 0.17 [0.05] 2 min after), and least for information flow from posterior channels. The widespread decrease in bivariate Granger causality reflects loss of cortical coordination. The relationship between functional connectivity (coherence) and effective connectivity (Granger causality) was inconsistent. CONCLUSIONS: Propofol-induced unresponsiveness is marked by a global decrease in information flow, greatest from the lateral frontal and prefrontal brain regions in a posterior and medial direction. Loss of information flow may be a useful measure of connected consciousness.

Electroencephalographic slow wave dynamics and loss of behavioural responsiveness induced by ketamine in human volunteers.

BACKGROUND: Previous work on the electroencephalographic (EEG) effects of anaesthetic doses of ketamine has identified a characteristic signature of increased high frequency (beta-gamma) and theta waves alternating with episodic slow waves. It is unclear which EEG parameter is optimal for pharmacokinetic-pharmacodynamic modelling of the hypnotic actions of ketamine, or which EEG parameter is most closely linked to loss of behavioural responsiveness. METHODS: We re-analysed previously published 128-channel scalp EEG data from 15 subjects who had received a 1.5 mg kg-1 bolus i.v. dose of ketamine. We applied standard sigmoid pharmacokinetic-pharmacodynamic models to the drug-induced changes in slow wave activity, theta, and beta-gamma EEG power; and examined the morphology of the slow waves in the time domain for Fz, F3, T3, P3, and Pz average-referenced channels. RESULTS: Hypnotic doses of ketamine i.v. induced medio-frontal EEG slow waves, and loss of behavioural response when the estimated brain concentration was 1.64 (0.17) µg ml-1. Recovery of responsiveness occurred at 1.06 (0.21) µg.ml-1 after slow wave activity had markedly diminished. Pharmacokinetic-pharmacodynamic modelling fitted best to the slow wave activity and theta power (almost half the beta-gamma channels could not be modelled). Slow wave effect-site equilibration half-time (23 [4] s), and offset, was faster than for theta (47 [22] s). CONCLUSIONS: Changes in EEG slow wave activity after a hypnotic dose of ketamine could be fitted by a standard sigmoid dose-response model. Their onset, but not their offset, was consistently associated with loss of behavioural response in our small study group.

Investigation of Slow-wave Activity Saturation during Surgical Anesthesia Reveals a Signature of Neural Inertia in Humans.

BACKGROUND: Previously, we showed experimentally that saturation of slow-wave activity provides a potentially individualized neurophysiologic endpoint for perception loss during anesthesia. Furthermore, it is clear that induction and emergence from anesthesia are not symmetrically reversible processes. The observed hysteresis is potentially underpinned by a neural inertia mechanism as proposed in animal studies. METHODS: In an advanced secondary analysis of 393 individual electroencephalographic data sets, we used slow-wave activity dose-response relationships to parameterize slow-wave activity saturation during induction and emergence from surgical anesthesia. We determined whether neural inertia exists in humans by comparing slow-wave activity dose responses on induction and emergence. RESULTS: Slow-wave activity saturation occurs for different anesthetics and when opioids and muscle relaxants are used during surgery. There was wide interpatient variability in the hypnotic concentrations required to achieve slow-wave activity saturation. Age negatively correlated with power at slow-wave activity saturation. On emergence, we observed abrupt decreases in slow-wave activity dose responses coincident with recovery of behavioral responsiveness in ~33% individuals. These patients are more likely to have lower power at slow-wave activity saturation, be older, and suffer from short-term confusion on emergence. CONCLUSIONS: Slow-wave activity saturation during surgical anesthesia implies that large variability in dosing is required to achieve a targeted potential loss of perception in individual patients. A signature for neural inertia in humans is the maintenance of slow-wave activity even in the presence of very-low hypnotic concentrations during emergence from anesthesia.

Electroencephalographic Response to Sodium Nitrite May Predict Delayed Cerebral Ischemia After Severe Subarachnoid Hemorrhage.

OBJECTIVES: Aneurysmal subarachnoid hemorrhage often leads to death and poor clinical outcome. Injury occurring during the first 72 hours is termed "early brain injury," with disruption of the nitric oxide pathway playing an important pathophysiologic role in its development. Quantitative electroencephalographic variables, such as α/δ frequency ratio, are surrogate markers of cerebral ischemia. This study assessed the quantitative electroencephalographic response to a cerebral nitric oxide donor (intravenous sodium nitrite) to explore whether this correlates with the eventual development of delayed cerebral ischemia. DESIGN: Unblinded pilot study testing response to drug intervention. SETTING: Neuroscience ICU, John Radcliffe Hospital, Oxford, United Kingdom. PATIENTS: Fourteen World Federation of Neurosurgeons grades 3, 4, and 5 patients (mean age, 52.8 yr [range, 41-69 yr]; 11 women). INTERVENTIONS: IV sodium nitrite (10 µg/kg/min) for 1 hour. MEASUREMENTS AND MAIN RESULTS: Continuous electroencephalographic recording for 2 hours. The alpha/delta frequency ratio was measured before and during IV sodium nitrite infusion. Seven of 14 patients developed delayed cerebral ischemia. There was a +30% to +118% (range) increase in the alpha/delta frequency ratio in patients who did not develop delayed cerebral ischemia (p < 0.0001) but an overall decrease in the alpha/delta frequency ratio in those patients who did develop delayed cerebral ischemia (range, +11% to -31%) (p = 0.006, multivariate analysis accounting for major confounds). CONCLUSIONS: Administration of sodium nitrite after severe subarachnoid hemorrhage differentially influences quantitative electroencephalographic variables depending on the patient's susceptibility to development of delayed cerebral ischemia. With further validation in a larger sample size, this response may be developed as a tool for risk stratification after aneurysmal subarachnoid hemorrhage.