Anesthetics generate the most recognizable feature of general anesthesia, unconsciousness, by modulating ion channels in the central nervous system, especially the GABA_A receptor (GABA_AR) (Garcia et al., 2010). The exposure of GABA_ARs to neurotransmitters and modulators (e.g., benzodiazepines and anesthetics) depends on the number and distribution of receptors existing on the cell surface in a particular brain region. In addition to abundance, variation is also an important determinant of GABAergic signaling. The most prominent classifying features are receptor proximity to the synapse and the type of current it mediates, with subtypes broadly characterized as “intra-synaptic” receptors mediating phasic currents or “extra-synaptic” receptors mediating tonic currents. Mutagenic and recombinant studies have demonstrated that these subtypes have profound intrinsic differences in sensitivity to GABA, pharmacologic modulation, and anesthetic agents. These subtype-specific interactions are critical to understanding how modulation of inhibitory neurotransmission is achieved under clinical or diseased conditions, and to interpreting the basis for neuron ensemble activity patterns that manifest during anesthesia, e.g., those reflected by electroencephalograph (EEG) recordings. The clinical implication is that subtle differences in the CNS distribution or expression levels of certain GABA_AR subtypes could underlie phenotypic differences in sensitivity to anesthesia. These differences might manifest as problems related to appropriate dosing of anesthetic drugs (i.e., as an increased risk of awareness with recall) during surgery or as an enhanced sensitivity to the hypoactivity commonly encountered in the immediate post-anesthesia period, emergence. Moreover, increased functional knowledge of region-specific as well as subcellular localization of GABA_AR subtypes may hold the key to understanding the neurological consequences of anesthesia, e.g., emergence complications, delirium, or long-term cognitive dysfunction.

Many drugs which modulate GABA_AR channel function also change surface expression of GABA_ARs, e.g., flumazenil (Kuver and Smith, 2016), alcohol (Gonzalez et al., 2012), and neurosteroids (Abramian et al., 2014). Interestingly, recent studies demonstrate that anesthetic agents may as well (Zurek et al., 2014; Li et al., 2015). This review will describe pharmacologic and activity based influences on GABA_AR surface expression and its relationship to anesthesia and emergence. We will review what is known (and what is unknown) regarding brain region specificity of GABA_AR subtypes and how the GABA_AR landscape might change during surgical anesthesia. Then, we will outline how GABA_AR surface expression is dynamically maintained by trafficking between subcellular compartments, and specify where subtype variability imparts important functional and trafficking properties. We will also illustrate how subtype-specific surface expression underlies inhibitory signaling and how subtype heterogeneity underlies the behavioral endpoints of anesthesia. Finally, we discuss how trafficking changes can alter clinical sensitivity to anesthesia, and how trafficking phenotypes may contribute to complications in the recovery of cognitive functions from anesthesia.

Physiologically, drugs that produce general anesthesia for use in surgery cause a progressive depression of electrochemical communication among cells in the central nervous system. The cardinal target of anesthesia, consciousness, is supported by a network of arousal centers residing in subcortical nuclei. These excitatory projections connect to each other and to the rest of the brain. In turn, these nuclei receive inhibitory projection from the hypothalamus (preoptic area), the thalamus, and the cortex (Silber and Rye, 2001). The predominate explanation for the process of initiating unconsciousness is that when anesthetic agents infuse the brain during induction, this network of interdependent excitatory/inhibitory connections is disturbed, and the arousal nuclei decrease neurotransmitter output (Brown et al., 2011). Ultimately, the diminished excitatory drive onto pyramidal neurons of the cortex and thalamus causes network activity to collapse, leading to a decrease in cortical activity, energy, and integration of information – ultimately a state of quiescence (Mashour et al., 2005; Alkire et al., 2008; Brown et al., 2011).

While they primarily generate unconsciousness (and other endpoints of anesthesia) by actions on ion channels in the central nervous system, an important caveat to this paradigm is that anesthetics have molecular targets on many other proteins in the body (Pan et al., 2008). In neurons, affected processes include cellular housekeeping systems such as mitochondrial transition, calcium homeostasis, and actin polymerization (Platholi et al., 2014; Vutskits and Xie, 2016). “Off-target” effects can underlie catastrophic patient reactions, such as the hypersensitivity of skeletal muscle ryanodine receptor channel function to volatile agents in malignant hyperthermia (Correia et al., 2012). Subtler effects with latent complications are suspected in the development of several neurodegenerative diseases, namely Alzheimers’ via β-Amyloid (Xie and Xu, 2013). Later, we review anesthetic-protein interactions in the second messenger signaling protein pathways known to control GABA_AR trafficking. Of non-ion channel proteins, there are no obvious structural motifs of the intracellular protein targets of anesthetics.

The crystallized GABA_AR structure was reported in 2014 (Miller and Aricescu, 2014). GABA_ARs are related by protein homology to of the Cys-loop superfamily of ligand-gated ion channels, which share similarities with the prokaryotic gleobacter ligand gated ion channel (Nury et al., 2011). GABA_ARs assemble into heteropentamers from 19 different possible subunits that are classified by protein sequence homology as α1–6, β1–3, γ1–3, δ, ε, θ, or ρ1–2. Most naturally occurring receptors contain two α, two β, and either a γ or δ subunit, with α1β2γ2 being the most abundant subtype in the mammalian CNS (McKernan and Whiting, 1996). Only nine different configurations are unequivocally expressed in abundance in the mammalian brain (Olsen and Sieghart, 2008). Table 1 describes some of the known research of these configurations. Individual subunit combinations are unique with respect to their pharmacology (Olsen and Sieghart, 2008), distribution among different brain regions (Wisden and Seeburg, 1992), cell-type specific expression (Lee and Maguire, 2014), and subcellular localization (Mody and Pearce, 2004).

For all these receptors, the primary endogenous ligand is GABA, which gates the chloride-permeable ion channel pore. Canonically, GABA-ergic inhibition curtails neuronal activity. This effect occurs via either hyperpolarization or shunting, depending on the chloride driving force, the resting membrane potential, and the action potential threshold at the time the impulse arrives. GABA is depolarizing in the earliest developing neurons that have not yet expressed specific membrane transporters to establish the usual chloride gradient (Ganguly et al., 2001; Ben-Ari et al., 2007; Ben-Ari, 2015). Because the neonate brain contains neurons at many different stages of development, anesthetics potentiating GABA-mediated currents could be excitatory in some neurons.

At clinical doses, most anesthetics act as positive allosteric modulators, increasing channel function by binding to protein cavities remote from the ligand binding site and causing conformational changes in structure that increase and prolong the response to GABA. Several interaction sites for volatile agents on the GABA_AR have been identified through mutagenic and biophysical experiments (Franks, 2006). Similarly, sites for intravenous agents have been identified (Miller and Aricescu, 2014; Weiser et al., 2014). On the intracellular face of the receptor, each subunit contains a small cytoplasmic loop between the first and second transmembrane domains and a larger cytoplasmic loop between the third and fourth transmembrane domain. This larger loop domain has the largest sequence variability between subtypes (Olsen and Sieghart, 2008), and contains several sites for protein–protein interactions. This intracellular loop is the main site for regulatory control of membrane trafficking by intracellular signaling pathways.

Receptor sequence heterogeneity is well studied as a source of subtle biophysical properties that lead to pharmacologic specificity for various endogenous and exogenous substances, including anesthetic agents. Recombinant studies demonstrate that the α subunit contributes to biophysical difference in GABA sensitivity, as heterologously expressed α1β2γ2 receptors show a lower affinity than α4β2γ2 (Jia et al., 2005). In most of the brain, intrasynaptic receptors contain α1–3 whereas extra-synaptic receptors contain α4–6 (Farrant and Nusser, 2005). Although all of these subunits are subject to allosteric modulation by anesthetic agents (with the notable exception of ρ), the sensitivity to anesthetics is not equal, and the extrasynaptic receptors are comparatively sensitive to modulation by anesthetics (Bonin and Orser, 2008).

GABA-ergic neurotransmission evokes two distinct forms of inhibitory currents. The basis for these two types of currents lies within receptor subunit composition and subcellular location. Phasic currents are primarily mediated by low-affinity intrasynaptic receptors, which generate brief and quickly desensitizing inhibitory post-synaptic currents (IPSCs) in response to synaptic neurotransmission (Mody and Pearce, 2004). Intrasynaptic receptor clustering is based on the construction of the GABAergic synapse, which is made of a post-synaptic density of several proteins organized opposite to the presynaptic terminal (Tretter et al., 2012). Intrasynaptic receptors are typically are made of large clusters of α1, α2, or α3 containing receptors anchored to the post-synaptic cytoskeleton by the scaffold protein gephyrin. Inside the synapse, these receptor clusters are transiently exposed to high (∼mM) concentrations of GABA from synaptic vesicular release, and generate phasic currents that terminate after GABA is cleared from the synapse (Stell and Mody, 2002). Even before clearance by glial or neuronal pumps, phasic currents rapidly desensitize generating the characteristic decayed shape of the ISPC.

Tonic currents are mediated by extra-synaptically located receptors, which generate persistent conductances in response to smaller (∼nM–μM) amounts of GABA existing in the intercellular space (Stell and Mody, 2002). The extra-synaptic receptors have a higher affinity for GABA and ideal for detection of small amounts of neurotransmitter. Additionally, these receptors exhibit little or no accommodation to saturating concentrations of GABA, a phenomenon known as desensitization. The GABA that evokes tonic current is thought to arise from two overlapping sources, ambient and spill-over GABA. Ambient GABA describes the minute amount preexisting in the extracellular space. Measurements from in vivo microdialysis suggest peri-synaptic GABA is between ≈30–300 nM, depending on brain region (Lerma et al., 1986; Tossman et al., 1986). Notably, GABA levels can change with physiological and behavioral states (de Groote and Linthorst, 2007). Spill-over GABA is released by synaptic transmission and diffuses away to have paracrine action on remote receptors, e.g., on other neurons. These terms have conceptual overlap and do not demarcate two types of GABA, but rather distinguish effects in spatiotemporal relation to synaptic transmission. High levels of phasic inhibition can drive spillover and raise local ambient GABA. For example, an increase in inhibitory activity can encourage sleepiness in part due to the accumulation of ambient GABA.

Comparatively, tonic and phasic inhibitory currents generate different effects on neuronal excitability and consequent inhibitory/excitatory signaling dynamics (Mody and Pearce, 2004). Functionally, phasic GABA currents decrease the instantaneous probability of action potential firing. In contrast, tonic currents cause a more persistent ongoing inhibition or hyperpolarization that is not temporally discrete. Because tonic current decreases membrane resistance, hypothetically the behavior of more actively firing neuron would be more sensitive to tonic inhibition than a less actively firing one. How these conductances ultimately affect individual neurophysiological process depends on cell-type specific intrinsic properties as well as circuit-specific properties. For example, interneuron spiking behavior changes dramatically with the addition of tonic GABA inhibitory input (Pavlov et al., 2014) in a bidirectional manner related to intrinsic cellular properties. Subcellular innervation patterns onto target neurons (e.g., axo-axonic vs. axo-somatic vs. axo-dendritic) would also critically determine post-synaptic effect. Further complicating a straightforward interpretation of anesthetic effects as a function of GABAergic modulation is the widespread expression of GABA_ARs. An inhibitory input onto a target neurons could be modulated at the post-synaptic level (anesthetic enhancement of post-synaptic GABA_ARs), or at the pre-synaptic level (anesthetic enhancement of interneuron GABA_ARs and depressed output and disinhibiting of the target neuron). In neuroscience at large, recent significant technical advances in methods of isolating cell-type specific actions via genetically select identification and activation have revolutionized empirical characterization of neural circuits (Murphey et al., 2014). The application of similar strategies to the study of anesthetic mechanisms could clarify with greater precision how anesthetics affect inhibitory actions within select neural circuits.

Finally, phasic and tonic conductances also shape emergent neural network properties such as oscillations (Mann and Mody, 2010). The temporal dynamics of phasic inhibition are important because synchronizing population activity into slow oscillatory waves requires precise timing of inhibitory input. Tonic current is important in rhythmogenesis as an input counteracting tonic excitation, as well as factor in the activation of hyperpolarization-gated conductances.

At clinically used doses, GABA-ergic anesthetics generally potentiate phasic currents by prolonging IPSC duration, and tonic currents by increasing the amplitude and charge transfer. Pharmacologic substances do not affect tonic and phasic conductances in the same way (Bieda et al., 2009). An enhanced sensitivity to GABA-ergic anesthetic drugs is a common feature of extra-synaptic receptors (Bonin and Orser, 2008). In hippocampal pyramidal excitatory neurons, low concentrations (25 μM) of isoflurane enhance the α5 mediated tonic but not phasic current (Caraiscos et al., 2004b). In the same experiments, 100 μM is the minimal dose to enhance IPSCs. These reactions to sub-anesthetic doses may explain why low doses of our anesthetics produce a sedation phenotype. Similarly, direct activation-low doses of isoflurane will also directly activate α4s, whereas only high doses can directly activate α1 (Raines et al., 2003). For reference a typical concentration of isoflurane which might mediate unconsciousness clinically is 250–300 μM (Hemmings and Egan, 2013).

The behavioral responses of transgenic mice with mutated extra-synaptic subunits suggest that disabling the receptors mediating tonic current is sufficient to abolish specific elements of anesthesia. For example, deleting the α4 subunit significantly reduces the effect of isoflurane on amnesia (context-dependent fear learning) with no effect on hypnosis or immobility (Rau et al., 2009). The potential importance of this is that intraoperative amnesia likely happens because low sub-anesthetic doses of isoflurane are sufficient to impair learning and memory by increasing α4βxδ tonic current in the dentate gyrus, so changes in α4 surface expression could affect amnesia without affecting the other anesthetic endpoints that are more salient during operation.

Benzodiazepines (e.g., midazolam) allosterically potentiate receptors containing the γ2 subunit, with sensitivity determined by the α subunit (Jenkins et al., 2001). Conversely, the δ containing receptors which mediate tonic current are insensitive to benzodiazepines but very sensitive to allosteric enhancement by neurosteroids (Mody and Pearce, 2004).

Barbiturates (e.g., thiopental) also allosterically potentiate GABA_ARs at clinically delivered doses. Evidence suggest sites of action within α and β subunits, and potentially elsewhere. Hippocampal CA1 phasic currents appear to be more sensitive than tonic to potentiation by thiopental (Bieda et al., 2009), although significant effects were seen on both. The largely cerebellar α6 extrasynaptic receptor has a uniquely high sensitivity to pentobarbital (Drafts and Fisher, 2006), perhaps underlying anticonvulsant and ataxic properties, although potentiating effects can also be seen on α4β2γ2 and α4β2δ (Akk et al., 2004). Of historical note, the barbiturates were once the mainstay class of anesthetic agents. With the exception of thiopental, they were largely replaced for routine anesthesia due to a range of undesirable side-effects including the potentially dangerous overlap between anesthetic range and lethal doses. However, they remain in use for veterinary euthanasia (e.g., for biomedical research animals) and for capital punishment via lethal injection.

The exact receptor subtypes that underlie tonic and phasic currents are particular to brain region. Here we summarize receptor subtype distribution, and for each receptor population its hypothesized role in anesthesia. For this review, we focus more on regions of the brain heavily involved in consciousness and memory, to reflect clinical concerns about depth of anesthesia and lingering effects on attention and memory observed.

The strength of inhibitory currents, tonic or phasic, depends on the number of receptors expressed on the neuronal cell surface. This number is dynamically maintained because GABA_ARs undergo controlled rates of endocytosis, exocytosis, and degradation- in effect constantly recycling between the surface and the cytoplasm. These events are coordinated by a myriad of effector and regulatory proteins, themselves controlled by second messengers signaling events. Perturbations to all of these steps can potentially alter GABAergic signaling, e.g., an increase in surface expression and current can arise from increased rates of exocytosis, reduced endocytosis, or decreased degradation. Given that changes in surface expression can be observed in vitro on the time course of 15–30 min, it is conceivable that routine anesthesia care has major influence on the dynamics of membrane surface expression. A basic review of fundamental trafficking mechanisms is presented first, before a discussion of mechanisms for pharmacologic and activity based changes in GABA_AR surface expression.

GABA_A receptor subunits are synthesized in the endoplasmic reticulum (ER). Correctly oligomerized channels under post-translational processing within the Golgi apparatus and then actively exported to the cell surface and inserted into the plasma membrane by exocytosis (Luscher et al., 2011). Imaging and biochemical assays suggest that receptor insertion occurs extra-synaptically (Bogdanov et al., 2006), and intrasynaptic receptors undergo additional targeting to the post-synaptic density via lateral diffusion (Bannai et al., 2009). Kinetic studies suggest that movement into the synapse happens quickly, as the ratio of synaptic/extra-synaptic β3s increases significantly within 15 min (Bogdanov et al., 2006). Synaptic confinement involves direct interaction gephyrin (Jacob et al., 2005), the cytoskeletal-bound post-synaptic scaffolding protein of inhibitory synapses, through binding sites on α1, α2, α3, β2, and β3 subunit large cytoplasmic loops (Tretter et al., 2012).

Keeping extra-synaptic receptors in place is less understood. The α5β3γ2 receptors affix to the cytoskeleton in extra-synaptic clusters by the scaffolding protein radixin. Their extra-synaptic confinement is not static, and dissociation from radixin and into synapses has been identified as a novel mechanism means of synaptic strengthening (Hausrat et al., 2015). Within synapses, α5β3γ2 receptors mediate a slowly decaying component of IPSCs through transient interactions with gephyrin (Brady and Jacob, 2015). GABA_AR exchange between intrasynaptic and extra-synaptic sites is under active investigation as a potentially important form of inhibitory plasticity (Triller and Choquet, 2005). The extra-synaptic location of α4βxδ receptors is attributed to absence of a gephyrin binding site within the α4 protein sequence, as an extra-synaptic scaffolding protein for α4βxδ has not been identified. In general, structural and regulatory elements for extra-synaptic receptor densities have not been well characterized. Because of their unique pharmacological role in anesthetic-hyperpolarization, understanding the mechanics of extra-synaptic receptor movement within the cell and its surface is a critical knowledge gap in anesthetic mechanisms.

GABA_ARs are constantly removed from the cell surface via clathrin-mediated endocytosis, which is orchestrated by complex sequences of proteins interactions all critically dependant on subunit-specific endocytotic regulatory site. These physical interactions are subject to modulation by post-translational modifications (especially phosphorylation), and aberrations can dramatically affect surface expression. The details are well-summarized elsewhere (Nakamura et al., 2015). Because these regulatory processes depend on specific protein residues, anatomical and behavioral phenotypes can be observed in a subunit-specific manner. For example, transgenic knock-in mice expressing γ2 (Y265/7F) phosphomimetic receptors, which lack γ2 Y265/7 mediated endocytosis, have increased γ2 cell surface expression, larger inhibitory synapses, and impaired spatial memory (Kretschmannova et al., 2013). Notably, this trafficking phenotype engenders an abnormal responses to anesthesia, as female γ2(Y256/7F) mice are more sensitive to the amnestic and hypnotic effects of propofol. The basis is likely the higher measurements of α4βxδ in the thalamus and DG as a result of compensatory changes in gene expression and surface trafficking that is gender-sensitive. Induced changes endocytotic regulation may have a rapid effect on surface expression because of the relatively quick rates of constitutive intrasynaptic receptor endocytosis. Biochemical rate-measurements suggest that intrasynaptic receptor constitutive endocytosis occurs rapidly, with approximately 20–25% of surface receptor β3 subunits internalizing by 30 min in neuron cultures (Kittler et al., 2004). Other groups have shown significant internalized γ2 protein after 15 min (Joshi and Kapur, 2009).

Likewise, specific endocytotic mechanisms for the α4βxδ receptors mediating tonic current are mediated through unique sequences on the δ subunit (Gonzalez et al., 2012). Kinetically, constitutive endocytosis of δ-containing receptors appears to occur more slowly than intrasynaptic receptors. The surface half-life of δ subunits is 103–171 min depending on the measurement method, immunostaining in culture or slice biotinylation in organotypic hippocampal slices, respectively (Joshi and Kapur, 2009). However, ethanol-evoked endocytosis of α4βxδ happens more quickly, as significantly diminished surface expression can be measured within 15 min (Gonzalez et al., 2012). Given the role of α4βxδ-mediated tonic current in thalamocortical and hippocampal circuits underlying anesthetic hypnosis and amnesia, α4βxδ trafficking changes have a direct bearing on the behavioral response to anesthesia. For example, latency to loss of righting reflex (metric for sensitivity to anesthetic hypnosis) increases 12 and 24 h following ethanol intoxication, suggesting a desensitization to isoflurane (Liang et al., 2007).

Once internalized, GABA_AR containing endosomes are sorted between recycling endosomes or late endosomes destined for lysosomal degradation. Recycled receptor endosomes reside transiently beneath the cell surface. Intracellular GABA_ARs residing near gephyrin underneath the synapse have been identified, presumably forming receptor reservoirs that can be rapidly mobilized to mediate short-term inhibitory synaptic plasticity (van Rijnsoever et al., 2005).

Ubiquitinylated receptors can also be targeted for degradation within lysosomes, acidic organelles which enzymatically digest the proteins inside. Again, this process appears to happen relatively quickly, as rate studies in cultured neurons suggested that 30% of neuronal GABA_ARs degraded over the course of 6 h (Kittler et al., 2004). Overall, it is clear that GABA_AR intracellular movement is a complicated dynamic process with rapid kinetics, and very subtle changes in protein signaling and have the capacity to change surface expression and phenotype very quickly.

Recent research has shown that anesthesia itself may cause changes to the surface expression of GABA_ARs and other ion channels that control neurological function. At present, these changes appear to be mediated by the effect of anesthetics on the intracellular signaling pathways that control trafficking. Propofol was demonstrated to increase β3 subunit surface expression in hippocampal brain and primary cultures within 15, 30, and 60 min of exposure (Li et al., 2015). The underlying mechanism is PKC-mediated phosphorylation of β3 at the AP-2 binding domain. Immunoprecipitation of β3 from hippocampal lysates show a depressed association with AP-2 subunit β-adaptin, suggesting diminished recruitment of the entire AP-2 complex. Surface expression of GluR1, a subunit of AMPA receptors which are also trafficked using AP-2 and clathrin-mediated endocytosis, was not significantly affected, suggesting that propofol is not impairing general endocytotic mechanisms. Activated PKCε is increased at these timepoints. PKCε kinase activity on β3 receptors measured with hippocampal lysate and recombinant PKC shows increased β3 phosphorylation. The change in surface expression and decrease in AP-2 subunit binding are blocked with PKCε inhibitor (Li et al., 2015). Functionally, propofol causes an increase in evoked and miniature IPSC amplitude that can be partially reversed with PKCε inhibitor. Post-translational modification of channel kinetics by PKC and allosteric enhancement of channel function by propofol likely contribute to this effect, but the effect of the inhibitor suggests that ultimately propofol is creating a temporary sensitization of GABAergic transmission through PKCε-dependant mechanisms. Similarly, a temporary increase in α5 surface expression has been associated with etomidate administration (Zurek et al., 2014). This increase appears to be in part responsible for post-anesthesia memory impairment. The expression levels return to normal in 2 weeks.

Several networks of second messengers and signaling protein regulators that control or influence the subcellular movement of GABA_ARs are also potentially sensitive to anesthetics. The interactions of these pathways with drugs commonly used in anesthesia are just being recognized and remain an area of active investigation. Here, we briefly introduce canonical pathways known to impact GABA_AR surface expression, describe the basis for anesthetic-sensitivity, and speculate hypothetical impact on GABA_AR surface expression.

The perioperative experience is an opportunity for profound dynamic changes in GABA_AR expression. Despite specific knowledge regarding individual subunit compositions subcellular localization, regional anatomy, and pharmacological attributes, we still lack understanding of the contribution of these specifics to intra- and post-operative consequences. Off-target effects on receptor trafficking subtype expression can impart neurophysiologic differences that likely influence the heterogeneity in behavioral endpoints of anesthesia, and variability in cognitive recovery trajectories following emergence. Because of the profound differences in GABA-signaling between developing and mature neurons – a focus on experiments using mature neurons should be emphasized, despite the increased technical challenges of working with older neurons in vitro. Future research avenues include studies using combined living cell microscopy and photolabeling that can help detect important changes in expression levels on relevant time scales.

General anesthetics are invaluable and unique among medicines because of their capacity for controlled manipulation of consciousness. Although substantial work has been done to characterize the effect of anesthetic agents on biophysical function of GABA_ARs and the resultant effect on global patterns of brain activity, much less is known about the long-term consequences on neurological function. Experimental studies increasingly show that anesthetic agents have the potential to induce persistent changes in neurophysiology, which has serious implications for basic and clinical science alike. A comprehensive understanding of these processes will be necessary for predicting how persistent adverse effects in neurological function arise and determining what factors underlie resilience to the effects of anesthetics.

IS performed the literature search, organized the theme, wrote the first draft, and constructed the table. EB and PG added key concepts/references as well as co-wrote and co-edited the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.