The most amazing aspect of the daily miracle of anesthesia is turning off consciousness to permit surgery to proceed and then fully restoring consciousness in a controlled manner. We still do not fully understand how this miracle occurs. A full understanding of consciousness, and the biology that underlies it, is probably decades in the future, if it is tractable at all.1 However, recent advances in neurophysiology are providing insight into how drugs interact with receptors throughout the nervous system to mediate anesthesia and analgesia.
How Nerves Work
Neurons
Neurons are the basic elements of all rapid signal processing within the body. A neuron consists of a cell body, also called the soma; dendrites; and the nerve fiber, also called the axon (Fig. 3-1). Dendrites are highly specialized extensions of the cell body. The axon of one neuron commonly terminates (synapses) near the cell body or dendrites of another neuron. The axon connects to a neighboring cell with a presynaptic terminal. The synaptic cleftseparates the presynaptic terminal and the cell body or dendrites of the next neuron in the signaling cascade (Fig. 3-2). Transmission of impulses between responsive neurons at a synapse is mediated by the release of a chemical mediator (neurotransmitter), such as glutamate or γ-aminobutyric acid (GABA) from the presynaptic terminal. The membrane of the postsynaptic neurons contains receptors that bind neurotransmitters released from presynaptic nerve terminals, transducing the signal.
The impulse travels along the nerve membrane as an action potential. This is entirely mediated by the receptors within the membrane. Indeed, removal of the axoplasm from the nerve fiber does not alter conduction of impulses. Nerve fibers derive their nutrition from the cell body. Interruption of a nerve fiber causes the peripheral portion to degenerate (Wallerian degeneration). The axon of a peripheral neuron is able to regenerate, as does the myelin sheath. Regeneration is the exception in most of the brain and spinal cord. Extensive research is underway to better understand the conditions that are required for central neuron regeneration to improve recovery from central neuronal injury.
Classification of Afferent Nerve Fibers
Nerve fibers are called afferent if they transmit impulses from peripheral receptors to the central nervous system (CNS) and efferent if they transmit impulses from the CNS to the periphery. Afferent nerve fibers are classified as A, B, and C on the basis of fiber diameter and velocity of conduction of nerve impulses (Table 3-1). Conduction speed increases with nerve diameter, because the larger diameter nerves have decreased longitudinal resistance to ion flux.2 The largest, and hence fastest, nerves are designated type A. Type A fibers are subdivided into α, β, γ, and δ. Type A-α1 fibers innervate muscle spindles and A-α1b innervate the Golgi tendon organ. Both A-α afferents are important to muscle reflexes and control of muscle tone.
All cutaneous mechanoreceptors (Meissner’s corpuscles, hair receptors, Pacinian corpuscles) transmit signals in type A-β fibers. Touch and fast pain are transmitted by lightly myelinated type A-δ fibers with free nerve endings. Type C fibers transmit slow pain, pruritus, and temperature sensation.
Myelin that surrounds type A and B nerve fibers acts as an insulator that prevents flow of ions across nerve membranes. Type C fibers are unmyelinated. The myelin sheath is interrupted approximately every 1 to 2 mm by the nodes of Ranvier (see Fig. 3-1).3 Ions can flow freely between nerve fibers and extracellular fluid at the nodes of Ranvier. Action potentials are conducted from node to node by the myelinated nerve rather than continuously along the entire fiber as occurs in unmyelinated nerve fibers. This successive excitation of nodes of Ranvier by an action potential that jumps between successive nodes is termed saltatory conduction (Fig. 3-3).3 Saltatory conduction allows for a 10-fold increase in the velocity of nerve transmission.2 It also conserves the membrane potential because only the membrane at the node of Ranvier depolarizes, resulting in less ion transfer than would otherwise occur. Furthermore, because depolarization is limited to the nodes of Ranvier, little energy is needed for to reestablish the transmembrane sodium and potassium ion concentration gradients necessary for signal transmission. The energy savings is more than a hundred fold. As brilliantly understated by Hartline and Colman,2 “For a nervous system such as ours, which already accounts for 20% of the body’s resting metabolic energy budget, this is not an inconsequential advantage.” If myelin did not exist, you would not be reading about it.
Evaluation of Peripheral Nerve Function
Peripheral nerves may be injured by ischemia of the intraneural vasa nervorum, as might be caused by excessive stretch of the nerve or external compression. Nerve conduction studies are useful in the localization and assessment of peripheral nerve dysfunction. Focal demyelination of nerve fibers causes slowing of conduction and decreased amplitudes of compound muscle and sensory action potentials. The presence of denervation potentials in skeletal muscle indicates axonal or anterior horn cell damage. Changes in motor unit potentials also arise from reinnervation of skeletal muscle fibers by surviving axons. Signs of denervation on the electromyogram after acute nerve injury require 18 to 21 days to develop.4 Electromyographic testing is helpful in determining the etiology of neurologic dysfunction that may occur after surgery.
The Action Potential
Electrical potentials exist across nearly all cell membranes, reflecting principally the difference in transmembrane concentrations of sodium and potassium ions. This unequal distribution of ions is created and maintained by the membrane-bound enzyme sodium-potassium ATPase, sometimes called the sodium-potassium pump. The sodium-potassium pump transfers three sodium ions out of the cell in exchange for two potassium ions brought into the cell. This causes a net transfer of positive charges out of the cell. The resulting voltage difference across the cell membrane is called the resting membrane potential. The cytoplasm is electrically negative (typically −60 to −80 mV) relative to the extracellular fluid (Fig. 3-4).5
When channels open to specific ions, the ions generally flow in the direction of their concentration gradients. An action potential is the rapid change in transmembrane potential due to the opening of sodium channels (depolarization) and rapid influx of sodium ions down the concentration gradient, reversing the net negative charge within the cell. The membrane resting potential is restored by the closing of the sodium channels and the opening of potassium channels (repolarization) after the action potential has passed. The outward flux of potassium ions down their concentration gradient restores the net negative charge within the cell. This is discussed in greater detail under the “Ion Channels” section.
Propagation of Action Potentials
Propagation of action potentials along the entire length of a nerve axon is the basis of rapid signal transmission along nerve cells. The size and shape of the action potential varies among excitable tissues (see Fig. 3-4).5
Action potentials are conducted along nerve or muscle fibers by local current flow that produces depolarization of adjacent areas of the cell membrane (Fig. 3-5). These propagated action potentials travel in both directions along the entire extent of the fiber. The transmission of the depolarization process along nerve or muscle fibers is called a nerve or muscle impulse. The entire action potential usually occurs in less than 1 millisecond.
During much of the action potential, the cell membrane is completely refractory to further stimulation. This is termed the absolute refractory period and is due to the presence of a large fraction of inactivated sodium ion channels. During the last portion of the action potential, a stronger than normal stimulus can evoke a second action potential. This “relative refractory period” reflects the need to activate a critical number of sodium ion channels to trigger an action potential.
The action potential is dynamic, which is difficult to illustrate with a static textbook image. We encourage the motivated reader to search for the text “action potential animation” on the Internet. There are many high-quality animations of the action potential that dynamically display how it propagates.
Ion Channel Evaluation
Current flowing through individual ion channels or voltage changes in a membrane can be measured with patch-clamping, a method used in electrophysiology.6 In patch clamping, an electrode is connected with a cell (or piece of membrane) with a tight seal. This electrode is able to control either the voltage or the current so that the other can be measured. Currents carried through different types of channels can be isolated by the use of specific inhibitors. For example, tetraethylammonium blocks many types of potassium ion channels, whereas tetrodotoxin blocks many types of sodium ion channels. Channels that are not normally expressed in a cell can be added through heterologous expression. With these methods, the impact of specific naturally occurring or synthetic channel elements on function can be evaluated. Using DNA manipulation, entire genes that code for channels/receptors can be knocked out. Specific amino acids in the receptor proteins can be altered by manipulating the DNA encoding the receptor, resulting in a knocked in receptor with specific amino acid substitutions. Enormous strides have been made in understanding the mechanism of action of anesthetic drugs using these genetic methods and assessments of electrophysiology and animal behavior studies.7
Abnormal Action Potentials
A deficiency of calcium ions in the extracellular fluid (hypocalcemia) prevents the sodium channels from closing between action potentials. The resulting continuous leak of sodium contributes to sustained depolarization or repetitive firing of cell membranes (tetany). Conversely, high calcium ion concentrations decrease cell membrane permeability to sodium and thus decrease excitability of nerve membranes. Low potassium ion concentrations in extracellular fluid increase the negativity of the resting membrane potential, resulting in hyperpolarization and decreased cell membrane excitability. Skeletal muscle weakness that accompanies hypokalemia presumably reflects hyperpolarization of skeletal muscle membranes. Local anesthetics decrease permeability of nerve cell membranes to sodium ions, preventing achievement of a threshold potential that is necessary for generation of an action potential. Blockage of cardiac sodium ion channels by local anesthetics may result in altered conduction of cardiac impulses and decreases in myocardial contractility.
Neurotransmitters and Receptors
Neurotransmitters are chemical mediators that are released into the synaptic cleft in response to the arrival of an action potential at the nerve ending. Neurotransmitter release is voltage dependent and requires the influx of calcium ions into the presynaptic terminals (see Fig. 3-2). Synaptic vesicles of the cell body and dendrites of neurons are the sites of continuous synthesis and storage of neurotransmitters. These vesicles may contain and release more than one neurotransmitter. Neurotransmitters may be excitatory or inhibitory, depending on the ion selectivity of the protein receptor. A postsynaptic receptor may be excited or inhibited, reflecting the existence of both types of receptors in the same postsynaptic neuron. Furthermore, the same neurotransmitter may be inhibitory at one site and excitatory at another. This is particularly applicable to G protein–coupled receptors as the associated G protein determines the polarity of the response. Some neurotransmitters function as neuromodulators or coagonists in that they influence the sensitivity of receptors to other neurotransmitters. For example, glycine is an important coagonist at the N-methyl-d-aspartate (NMDA) receptor.
Volatile anesthetics produce a broad spectrum of actions, as reflected by their ability to modify both inhibitory and excitatory neurotransmission at presynaptic and postsynaptic loci within the CNS. The precise mechanism of these effects remains uncertain. It is likely that volatile anesthetics interact with multiple neurotransmitter systems by a variety of mechanisms.8 In general, volatile anesthetics inhibit excitatory receptors (NMDA and nicotinic acetylcholine receptors) and potentiate the action of inhibitory receptors (GABAA and glycine). To quote Ted Eger, “How do they know?” Inhaled anesthetics may depress excitable tissues at all levels of the nervous system by interacting with neuronal membranes,9 resulting in a decreased release of neurotransmitters and transmission of impulses at synapses as well as a general depression of excitatory postsynaptic responsiveness.
The list of chemical mediators functioning as excitatory or inhibitory neurotransmitters continues to increase (Table 3-2). Glutamate is the major excitatory neurotransmitter in the CNS, whereas GABA is the major inhibitory neurotransmitter.8 Acetylcholine, dopamine, histamine, and norepinephrine are widely distributed and play important roles in sleep pathways that are impacted upon by general anesthetics. Neuromodulators coexist in presynaptic terminals with neurotransmitters but do not themselves cause substantive voltage or conductance changes in postsynaptic cell membranes. They can, however, amplify, prolong, decrease, or shorten the postsynaptic response to selected neurotransmitters.
Receptors can be classified by their cellular localization. Receptors on the cell membrane act as signal transducers by binding the extracellular signal molecule and converting this information into an intracellular signal that alters target cell function. Most signaling molecules are hydrophobic and interact with cell surface receptors that are directly or indirectly coupled to effector molecules. There are three classes of cell surface receptors as defined by their signal transduction mechanisms: guanine nucleotide-binding protein (“G protein”) coupled receptors, ligand-gated ion channels, and enzyme-linked transmembrane receptors.
G protein–coupled receptors in the plasma membrane are coupled to specific intracellular G proteins (Fig. 3-6). The binding of the receptor to the ligand activates the G protein, which then activates or inhibits an enzyme, ion channel, or other target. G protein–coupled receptors constitute the largest family of cell surface receptors. A number of different isoforms of G protein subunits (α, β, γ) are present and mediate stimulation or inhibition of functionally diverse effector enzymes and ion channels. Most hormones and many neurotransmitters interact with G protein–coupled cell-surface receptors to produce the cellular response.10–12 The resulting response is often a change in transmembrane voltage and thus neuronal excitability. There is great diversity in the number of G protein–coupled receptors for the same ligand as reflected by multiple receptors for catecholamines and opioids.13
Ligand-gated ion channels are channels in the plasma membrane that respond directly to extracellular ligands, rather than require coupling through G proteins (Fig. 3-7). They are one of three classes of ion channels, the other two being voltage-gated ion channels that respond to transmembrane voltage flux, and “other” gated ion channels that are gated by a huge variety of mechanisms. Rapid synaptic transmission is entirely accomplished through voltage-gated ion channels, which propagate action potentials, and ligand-gated ion channels, which transmit the signal across the synapse.
Enzyme-linked transmembrane receptors are not involved in neuronal signaling per se, as they have relatively slow effects on cells. Most enzyme-linked transmembrane receptors are tyrosine kinases that phosphorylate an intracellular second messenger when the extracellular ligand binds to the receptor (Fig. 3-8). The insulin receptor,14 the atrial natriuretic peptide receptor, and the receptors for many growth factors (nerve growth factor, epidermal growth factor, fibroblast growth factor, and vascular endothelial growth factor) are all examples of tyrosine kinase–linked transmembrane receptors.
There are also intracellular receptors. For example, steroid receptors and thyroid hormone receptors act in the nucleus where they directly regulate the transcription of specific genes, whereas phosphodiesterase inhibitors (e.g., caffeine, milrinone, and sildenafil) act in the cytosol by inhibiting the activity of phosphodiesterase, increasing the cytosolic concentration of cyclic adenosine monophosphate (cAMP). These receptors are also not involved in neuronal signaling per se, because the cellular response is quite slow.
G Protein–Coupled Receptors
G protein–coupled receptors consist of three separate components: a receptor protein, three G proteins (α, β, and γ), and an effector mechanism (see Fig. 3-6). The recognition site faces the exterior of the cell membrane to facilitate access of water-soluble endogenous ligands and exogenous drugs, whereas the catalytic site faces the interior of the cell. There are at least 16 Gα, 5 Gβ, and 11 Gγ proteins,15 providing G protein–coupled receptors that mediate an enormous variety of cellular effects.
The G protein–coupled receptor consists of a single protein with seven transmembrane spanning domains (Fig. 3-9). Binding of an extracellular ligand to the G protein–coupled receptor triggers a conformational change of the protein. That change causes activation of the Gα protein coupled to the interior portion of the receptor. The activation occurs by exchanging a guanine diphosphate (GDP) moiety that is bound to the protein for a guanine triphosphate (GTP) moiety. The activated Gα protein is liberated, where it interacts as a “second messenger” with other proteins in the cell.11 When the Gα protein finds its target, the GTP is hydrolyzed to GDP, and the energy liberated by that hydrolysis powers the effect of the Gα protein on the target protein.
Gα proteins can either be stimulatory, promoting a specific enzymatic reaction within the cell, or inhibitory, depressing a specific enzymatic reaction. For example, β-adrenergic receptors couple with stimulatory Gαs proteins and increase the activity of adenylyl cyclase (also called adenylate cyclase). Opioid receptors associate with inhibitory Gαi proteins that decrease the activity of adenylyl cyclase. By regulating the level of activity of adenylyl cyclase, the β-adrenergic and opioid receptors modulate the internal level of cAMP, which functions as an intercellular second messenger (see Fig. 3-6).
Just as Gαs and Gαi modulate adenylyl cyclase, other types of Ga proteins modulate other specific cellular targets. In some cases, the message is transmitted via Gβγ rather than Gα, as described below for G protein regulation of potassium channels.
Many hormones and drugs act through G protein–coupled receptors, including catecholamines, opioids, anticholinergics, and antihistamines. In contrast to the immediate cellular responses associated with ion channels, signals that use G protein–coupled receptors are involved in functions that operate with time courses of seconds to minutes. Some ion channels are also gated by G proteins. These are discussed below with the ion channels.
Dopamine
Dopamine represents more than 50% of the CNS content of catecholamines, with high concentrations in the basal ganglia. Dopamine can be either inhibitory or excitatory, depending on the specific dopaminergic receptor that it activates. Dopamine is important to the reward centers of the brain and plays a key role in addiction and tolerance to anesthetic and analgesic drugs.
Norepinephrine
Norepinephrine is present in large amounts in the reticular activating system and the hypothalamus, where it plays a key role in natural sleep and analgesia. Neurons responding to norepinephrine send excitatory (through α1) and inhibitory (through α2) signals to widespread areas of the brain, including the cerebral cortex. The sedative action of dexmedetomidine is mediated by activation of α2 adrenergic receptors in the locus ceruleus that inhibit firing of the ventral lateral preoptic nucleus of the hypothalamus (VLPO), an endogenous sleep pathway.16 Descending noradrenergic fibers that project to the dorsal horn of the spinal cord play an important tonic inhibitory role in pain transmission. These pathways are augmented by epidural clonidine for postoperative and intrapartum analgesia.
Substance P
Substance P is an excitatory neurotransmitter coreleased by terminals of pain fibers that synapse in the substantia gelatinosa of the spinal cord. Substance P activates the neurokinin-1 G protein–coupled receptor.
Endorphins
Endorphins are endogenous opioid peptide agonists that are secreted by nerve terminals in the pituitary, thalamus, hypothalamus, brainstem, and spinal cord. Endorphins act through the µ opioid receptor, the same receptor responsible for the effects of administered opioids. Endorphins are secreted after exercise and during pain and anxiety. Endorphins facilitate dopamine release and activate inhibitory pain pathways.
Serotonin
Serotonin (5-HT) is present in high concentrations in the brain, where it acts on both ligand-gated ion channels and G protein–coupled receptors. Serotonin receptors are located in the chemoreceptor trigger zone, where they are inhibited by ondansetron, granisetron, and other common antiemetic drugs
Histamine
Histamine is present in high concentrations in the hypothalamus and the reticular activating system. Histaminergic neurons present in the tuberomammillary nucleus of the hypothalamus are active during the wake cycle. The sleep promoting properties of antihistamine drugs that cross the blood–brain barrier are due to inhibition of H1 G protein–coupled receptors.
Ion Channels
As pointed out earlier, the normal resting membrane potential is −60 to −80 mV, with the interior of the cell negative relative to the extracellular fluid. The lipid bilayer is mostly impermeable to ions, which must pass in and out of the cell through ion-specific channels. If the flux of ions makes the inside of the cell more negative (“hyperpolarized”), then it is harder for the cell to initiate an action potential. If the flux of ions makes the inside of the cell less negative (“depolarized”), then it is easier for the cell to initiate an action potential.
When ion channels open, ions usually flow in the direction favored by their concentration gradient. Extracellular concentrations of sodium, calcium, and chloride greatly exceed intracellular concentrations, and thus these ions flow into cells when the appropriate ion channel opens. Intracellular concentrations of potassium greatly exceed extracellular concentrations, and thus potassium follows out of cells whenever a potassium channel is opened. The inwardly rectifying potassium channel is an exception in that potassium flows into the cell, opposite the concentration gradient, in response to the electrical gradient.
When sodium flows into a cell, it makes the interior less negative. Sodium channels are thus depolarizing. When potassium flows out of a cell, it makes the interior more negative. Therefore, potassium channels are hyperpolarizing. Sodium channels open to conduct action potentials, after which potassium channels open to restore the resting negative potential and terminate the action potential.
When chloride flows into a cell, the interior becomes more negative, or hyperpolarized. Because it is harder for a hyperpolarized cell to initiate an action potential, chloride channels are “inhibitory,” at least after birth. When calcium flows into a cell, the interior becomes less negative, or “depolarized.” Because it is easier for a depolarized cell to initiate an action potential, calcium channels are “excitatory.” Calcium can also act as a second messenger within the cell.
When cell membranes are depolarized (the outside becomes less negative relative to the inside) or the appropriate ligand is present, these ion channels undergo conformational changes, the ion channel opens, and ions pass through. About 104 to 105 ions flow per millisecond per channel and thousands of channels may open during a single action potential.
As mentioned previously, there are three basic types of ion channels: (a) ligand-gated ion channels (ionotropic receptors), (b) voltage-sensitive ion channels, and (c) ion channels that respond to other types of gating.
Ligand-Gated Ion Channels
Ligand-gated ion channels (ionotropic receptors) are complexes of protein subunits that act as switchable portals for ions. Ligand-gated ion channels are involved principally with fast synaptic transmission between excitable cells. Binding of signaling molecules to these receptors causes an immediate conformational change in the ion channels, opening (usually) or closing (rarely) the channel to alter the ion permeability of the plasma membranes and therefore the membrane potential. Ligand-gated ion channels are activated by ligands for which they are named. Nicotinic acetylcholine receptors (nAChRs), serotonin receptors (5-HT3), γ-aminobutyric acid receptors (GABAA) (see Fig. 3-7), and glycine receptors are opened in the presence of acetylcholine, serotonin, GABA, and glycine, respectively. Sometimes the agonist for which the channel is named is not the native agonist. For example, NMDA and α-amino-3-hydroxyl-5-methyl-4-isoxazole-propionate (AMPA) receptors are opened selectively by NMDA and AMPA, but the native agonist for both receptors is glutamate.
Excitatory Ligand-Gated Ion Channels
Excitatory ligand-gated ion channels cause the inside of the cell to become less negative, typically by facilitating the influx of cations into the cell.
Acetylcholine
Acetylcholine is an excitatory neurotransmitter that activates muscarinic and nicotinic receptors in the CNS. Nicotinic acetylcholine receptors are nonspecific cation channels, permitting sodium and in some cases calcium to flow into cells, and potassium to flow out of cells. Because the flow of sodium and calcium is driven both by concentration and electrical gradients, the channel produces a net positive inward flux of cations and is therefore depolarizing (the interior becomes less negative). Nicotinic acetylcholine receptors in the brain are most commonly in a presynaptic location where they act as a “gain control mechanism” to enhance the release of other neurotransmitters. Acetylcholine-releasing neurons play an important role in native sleep pathways where acetylcholine mediates arousal. Although all volatile anesthetics are highly potent inhibitors of the nicotinic acetylcholine receptors that mediate this response,17 direct nicotinic inhibition is not likely responsible for the hypnotic actions of volatile anesthetics. Nicotinic acetylcholine receptors are largely antagonized at volatile anesthetic concentrations; 1/10 of that induce immobility and thus at concentrations associated with a fully awake patient.18,19 Injection of nicotine into the central medial thalamus reversed the hypnotic effect of continued sevoflurane.20 However, in this case, nicotine was acting as an arousing stimulus. Microinfusion of the broad-spectrum nicotinic antagonist mecamylamine did not add to the hypnotic potential of sevoflurane by reducing the dose necessary for hypnosis.
The excitatory effect on the CNS mediated through nicotinic ion channels contrasts with the inhibitory effects that are mediated by the G protein–coupled muscarinic acetylcholine receptors in the peripheral parasympathetic nervous system.
Nicotinic acetylcholine receptors are also responsible for activating muscle contraction. Nondepolarizing muscle relaxants work by blocking the acetylcholine binding site. Because these channels cause depolarization, they are excitatory.
Glutamate
Glutamate is the major excitatory amino acid neurotransmitter in the CNS. Glutamate receptors are nonselective cation channels, permitting sodium and some calcium to flow into cells, and potassium to flow out of cells. Because nonspecific cation channels primarily favor net inward flux of cations down the electrical gradient, glutamate receptors are depolarizing and excitatory. Glutamate-responsive receptors are distributed widely in the CNS. Glutamate plays a key role in learning, and memory, central pain transduction, and pathologic processes such as excitotoxic neuronal injury following CNS trauma or ischemia.
Glutamate is synthesized by the deamination of glutamine via the tricarboxylic acid cycle. Glutamate is released into the synaptic cleft in response to depolarization of the presynaptic nerve terminal. The release of glutamate from presynaptic terminals is a calcium ion-dependent process regulated by multiple types of calcium channels. In common with many other central neurotransmitter systems, the actions of glutamate within the synaptic cleft are terminated by high-affinity sodium-dependent reuptake of glutamate.
The two main subgroups of glutamate receptors are inotropic and metabotropic receptors.8 Ionotropic glutamate receptors (NMDA, AMPA, and kainate receptors) are ligand-gated ion channels. Glutamate receptors that respond to NMDA are associated with neuropathic pain and opioid tolerance and are blocked by ketamine. NMDA receptors are highly calcium permeable. Glutamate receptors that respond to AMPA and kainate are involved with fast synaptic transmission and synaptic plasticity, including long-term potentiation.
Metabotropic glutamate receptors are transmembrane receptors that are linked to G proteins that modulate intracellular second messengers such as inositol phosphates and cyclic nucleotides.
Serotonin
The serotonin (5-HT) receptor is also excitatory, permitting passage of sodium, potassium, and calcium cations as described for the nicotinic acetylcholine receptor.
Inhibitory Ligand-Gated Ion Channels
Inhibitory ligand-gated ion channels cause the inside of the cell to become less negative, typically by facilitating the flux of chloride into the cell. Potassium channels that facilitate the efflux of potassium ions are also inhibitory.
γ-Aminobutyric Acid
GABA is the major inhibitory neurotransmitter in the brain. When two molecules of GABA bind to the GABA receptor, the chloride channel in the center of the receptor opens and chloride ions enter the cell following their concentration gradient (see Fig. 3-7).11 The negatively charged chloride ion hyperpolarizes the interior of the cell, rendering GABA receptors inhibitory shortly after birth. It is estimated that as many as one-third of the synapses in the brain are GABAergic. The chloride channel is formed from the α and β subunits, with or without γ and δ subunits.
In the developing brain neurons have higher concentrations of chloride then the extracellular fluid. As a result, opening of the GABA chloride channel initiates a flux of negatively charged chloride ions out of the cell, depolarizing the cell. Later in development, the potassium/chloride cotransporter appears. This transporter decreases intracellular chloride in exchange for extracellular potassium, creating a concentration gradient for chloride that favors inward flux.21 The change in chloride concentration gradient renders the GABA receptor hyperpolarizing and hence inhibitory.
GABA receptors are the target of propofol, etomidate, and thiopental, which can directly open the channel at high concentration, or at lower concentration increase sensitivity to exogenous GABA. Benzodiazepines also work through GABA receptors but increase the sensitivity of the receptor to exogenous GABA only rather than directly opening the ion channel. There is increasing evidence that extrasynaptic GABA receptors are important in volatile anesthetic-induced behavioral responses.
Glycine
Glycine is the principal inhibitory neurotransmitter in the spinal cord, acting through the glycine receptor to increase chloride ion conductance into the cell, causing hyperpolarization. Glycine receptors are also present in the brain. These channels are involved in many neurologic processes and are modulated by a variety of anesthetic drugs but are not known to be responsible for any specific anesthetic induced behavior.
Strychnine and tetanus toxin result in seizures because they antagonize the effects of glycine on postsynaptic inhibition. Visual disturbances after transurethral resection of the prostate in which glycine is the irrigating solution may reflect the role of this substance as an inhibitory neurotransmitter in the retina.22 Amplitude and latency of visual evoked potentials are altered by infusions of glycine.23
Voltage-Gated Ion Channels
Voltage-gated ion channels are complexes of protein subunits that act as switchable portals sensitive to membrane potential through which ions can pass through the cell membrane. They are “voltage-sensitive” because they open and close in response to changes in voltage across cell membranes. Charged portions of the molecule physically move in response to voltage changes to energetically favor the open or closed state of the channel. For example, the sodium channel opens in response to a sudden depolarization, propagating the action potential in nerves. Voltage-gated ion channels are present in neurons, skeletal muscles, and endocrine cells. They are often named based on the ion that passes through the channel (e.g., sodium, chloride, potassium, and calcium channels).
The voltage-gated sodium channel is of particular interest to anesthesiologists, because it is the site of local anesthetic action. Local anesthetics block neural conduction by blocking passage of sodium through the voltage-gated sodium channel.
The human ether-a-go-go related gene (hERG) potassium channel is a voltage-gated inwardly rectifying potassium channel, mostly famous for its association with prolonged QT syndrome. The hERG potassium channel is sensitive to many drugs and is responsible for sudden death from drugs that predispose the patient to torsades de point. Inhibition of the hERG potassium channel is also responsible for the U.S. Food and Drug Administration (FDA) black box warning on droperidol.
G Protein–Gated Ion Channels
Some ion channels are directly gated by G proteins (Fig. 3-10). G protein–gated potassium channels are the most well studied of the G protein–regulated ion channels.24 The first identified G protein–regulated ion channel was the cardiac potassium channel, which is directly regulated by the M2 muscarinic acetylcholine G protein–coupled receptor.25 This is one of many inward rectifying potassium channels that share the unusual property of permitting influx of potassium ions into the cell following the electrical gradient, rather than the more typical outward flux of potassium following the ionic concentration gradient. G protein regulated inwardly, rectifying potassium channels, commonly referred to as GIRKs, are regulated by Gβγ rather than Gα. In addition to acetylcholine, A1 adenosine, α2 adrenergic, D2dopamine, opioid, serotonin, and GABAB receptors are coupled directly to GIRKs.24,26
Other Gated Ion Channels
Other types of ion channel gating include gating by other ions (e.g., hydrogen, calcium), second messengers (e.g., cAMP, cyclic guanosine monophosphate [cGMP]), and tissue injury (acid, stretch, temperature, cytokines).
Receptor Concentration
Receptors in cell membranes are not static components of cells. Excess circulating concentrations of ligand often results in a decrease in the density of the target receptors in cell membranes. For example, the excessive circulating norepinephrine in patients with pheochromocytoma leads to downregulation of β-adrenergic receptors. Desensitization of receptor responsiveness is the waning of a physiologic response over time despite (and, caused by) the presence of a constant stimulus.12 Drug-induced antagonism of receptors often results in an increased density of receptors in cell membranes (upregulation). Abrupt discontinuation of the antagonist can result in an exaggerated response to the endogenous agonist. This is one reason that most cardiovascular medicines should be continued throughout the perioperative period.
Receptor Diseases
Numerous diseases are associated with receptor dysfunction. For example, failure of parathyroid hormone and arginine vasopressin to produce increases in cAMP in target organs manifests as pseudohypoparathyroidism and nephrogenic diabetes insipidus, respectively. Grave’s disease and myasthenia gravis reflect development of antibodies against thyroid-stimulating hormone and nicotinic acetylcholine receptors, respectively.
The Synapse
Structure
The synapse functions as a diode that transmits an action potential from the presynaptic membrane to the postsynaptic membrane across the synaptic cleft (Fig. 3-11). The presynaptic membrane contains the vesicles of neurotransmitter and the reuptake pump that returns the neurotransmitter to the presynaptic axoplasm following neurotransmitter release. It also contains the voltage-gated calcium channel. Synaptic transmission starts when an afferent action potential arrives at the voltage-gated calcium channel.
The depolarization permits the influx of calcium ions through the voltage-gated calcium channel. Calcium ions bind to specialized proteins called the release apparatus on axonal and vesicular membranes. Calcium triggers the fusion of the vesicle to the cell membrane and the release of the neurotransmitter into the synaptic cleft through exocytosis, resulting in the extrusion of the contents of the synaptic vesicles. Calcium in the extracellular fluid is essential to the release of neurotransmitters in response to an action potential. The effect of calcium is antagonized by magnesium.
The neurotransmitter in the cleft binds to receptors in the postsynaptic membrane. This binding initiates an efferent action potential in the dendrite of the efferent nerve, which is then propagated. Immediately behind the postsynaptic membrane is the postsynaptic density. The postsynaptic density contains a variety of receptors and structural proteins responsible for maintaining synapse homeostasis.
There are several common misconceptions conveyed by the usual representation of the synapse. First, Figure 3-11 suggests that the synapse consists of two distinct plug-shaped entities that are joined together to form a synapse. Often, the presynaptic neuron may be no more than a slight widening of the axon, the “synaptic varicosity” or “bouton,” because of the presence of the vesicles containing the neurotransmitter. Second, the synapse often appears as a wide gap, as in Figure 3-11. However, the synapse is extremely narrow, on the order of just 20 nm, as shown in Figure 3-12. When the vesicle releases its content into the synapse, the concentration of neurotransmitter is extraordinarily high for a very brief period of time. Lastly, both dendrites and axons have extensive arborizations. The interconnection of hundreds of arborizations across tens of billions of brain cells creates circuits of unimaginable complexity.
Synaptic Modulation
The resting transmembrane potential of neurons in the CNS is about −70 mV, less than the −90 mV in large peripheral nerve fibers and skeletal muscles. The resting transmembrane potential is important for controlling the responsiveness of neurons and is impacted on by extrasynaptic receptors as well as the sodium-potassium ATP exchanger. Postsynaptic inhibitory and excitatory potentials modulated by synaptic and nonsynaptic signaling pathways sum to determine the likelihood of depolarization in response to an incoming stimulus.
Synaptic Delay
Synaptic delay is the 0.3 to 0.5 millisecond necessary for the transmission of an impulse from the synaptic varicosity to the postsynaptic neuron.27 This synaptic delay reflects the time for release of the neurotransmitter from the synaptic varicosity, diffusion of the neurotransmitter to the postsynaptic receptor, and the subsequent change in permeability of the postsynaptic membrane to various ions.
Synaptic Fatigue
Synaptic fatigue is a decrease in the number of discharges by the postsynaptic membrane when excitatory synapses are repetitively and rapidly stimulated. For example, synaptic fatigue decreases excessive excitability of the brain as may accompany a seizure, thus acting as a protective mechanism against excessive neuronal activity. The mechanism of synaptic fatigue is presumed to be exhaustion of the stores of neurotransmitter in the synaptic vesicles. Synaptic fatigue is unmasked at the neuromuscular junction in myasthenia gravis when the enormous reserve for neuromuscular transmission is limited by either pre- or postsynaptic autoimmune damage.
Posttetanic Facilitation
Posttetanic facilitation is increased responsiveness of the postsynaptic neuron to stimulation after a rest period that was preceded by repetitive stimulation of an excitatory synapse. This phenomenon reflects increased release of neurotransmitters due to enhanced local concentrations of intracellular calcium. Posttetanic facilitation may be a mechanism for short-term memory and sensory neuron wind-up.
Factors that Influence Neuron Responsiveness
Neurons are highly sensitive to changes in the pH of the surrounding interstitial fluids. For example, alkalosis enhances neuron excitability. Voluntary hyperventilation can evoke a seizure in a susceptible individual. Conversely, acidosis depresses neuron excitability, with a decrease in arterial pH to 7.0, potentially causing coma. Hypoxia can cause total refractoriness in neurons within 3 to 5 seconds as reflected by the almost immediate onset of unconsciousness following cessation of cerebral blood flow. This response is in part protective because the metabolic activity of inactive neurons is an order of magnitude less than that of active neurons.
Central Nervous System
The brain, brainstem, and spinal cord constitute the CNS. The brain is a complex collection of neural networks that regulate their own and each other’s activity. Activity within the CNS reflects a balance between excitatory and inhibitory influences, a homeostasis that is normally maintained within relatively narrow limits. Anatomic divisions of the brain reflect the distribution of brain functions ( Fig. 3-13).
The two cerebral hemispheres constitute the cerebral cortex, where sensory, motor, and associational information is processed. The limbic system lies beneath the cerebral cortex and integrates the emotional state with motor and visceral activities. The thalamus lies in the center of the brain beneath the cerebral cortex and basal ganglia and above the hypothalamus. The neurons of the thalamus are arranged in nuclei that act as relays between the incoming sensory pathways and the cerebral cortex, hypothalamus, and basal ganglia. The hypothalamus is the principal integrating region for the autonomic nervous system and regulates other functions, including systemic blood pressure, body temperature, water balance, secretions of the pituitary gland, emotions, and sleep.
The brainstem connects the cerebral cortex to the spinal cord and contains most of the nuclei of the cranial nerves and the reticular activating system. The reticular activating system is essential for regulation of sleep and wakefulness. The cerebellum arises from the posterior pons and is responsible for coordination of movement, maintenance of body posture, and certain types of motor memory.
The spinal cord extends from the medulla oblongata to the lower lumbar vertebrae. Ascending and descending tracts are located within the white matter of the spinal cord, whereas intersegmental connections and synaptic contacts are concentrated in the gray matter. Sensory information flows into the dorsal portion (posterior) of the gray matter, and motor outflow exits from the ventral (anterior) portion. Preganglionic neurons of the autonomic nervous system are found in the intermediolateral portions of the gray matter.
Cerebral Hemispheres
The two cerebral hemispheres, known as the cerebral cortex, constitute the largest division of the human brain. Regions of the cerebral cortex are classified as sensory, motor, visual, auditory, and olfactory, depending on the type of information that is processed. Frontal, temporal, parietal, and occipital designate anatomic positions of the cerebral cortex (Fig. 3-14). For each area of the cerebral cortex, there is a corresponding and connecting area to the thalamus such that stimulation of a small portion of the thalamus activates the corresponding and much larger portion of the cerebral cortex. Indeed, the cerebral cortex is actually an evolutionary outgrowth of the lower regions of the nervous system, especially the thalamus. The functional part of the cerebral cortex is composed mainly of a 2- to 5-mm layer of neurons covering the surface of all the convolutions. It is estimated that the cerebral cortex contains 50 to 100 billion neurons.
Anatomy of the Cerebral Cortex
The sensorimotor cortex is the area of the cerebral cortex responsible for receiving sensation from sensory areas of the body and for controlling body movement (see Fig. 3-14).3 The premotor cortex is important for controlling the functions of the motor cortex. The motor cortex lies anterior to the central sulcus. Its posterior portion is characterized by the presence of large, pyramid-shaped (pyramidal or Betz) cells.
Topographic Areas
The area of the cerebral cortex to which the peripheral sensory signals are projected from the thalamus is designated the somesthetic cortex (see Fig. 3-14).3 Each side of the cerebral cortex receives sensory information exclusively from the opposite side of the body. The size of these areas is directly proportional to the number of specialized sensory receptors in each respective area of the body. For example, a large number of specialized nerve endings are present in the lips and the thumbs, whereas only a few are present in the skin of the trunk.
The motor cortex is organized into topographic areas corresponding to different regions of the skeletal muscles. The spatial organization is similar to that of the sensory cortex. In general, the size of the area in the motor cortex is proportional to the preciseness of the skeletal muscle movement required. As such, the digits, lips, tongue, and vocal cords have large representations in humans. The various topographic areas in the motor cortex were originally determined by electrical stimulation of the brain during local anesthesia and observation of the evoked skeletal muscle response. Such stimulation can be used intraoperatively to identify the location of the motor cortex and thus avoid damage to this area. The motor cortex is commonly damaged by loss of blood supply as occurs during a stroke.
Corpus Callosum
The two hemispheres of the cerebral cortex, with the exception of the anterior portions of the temporal lobes, are connected by fibers in the corpus callosum. The anterior portions of the temporal lobes, including the amygdala, are connected by fibers that pass through the anterior commissure. The corpus callosum and anterior commissure make information processed or stored in one hemisphere available to the other hemisphere.
Dominant versus Nondominant Hemisphere
Language function and interpretation is typically localized in the dominant cerebral hemisphere, whereas spatiotemporal relationships (ability to recognize faces) is localized in the nondominant hemisphere. The left hemisphere is dominant in 90% of right-handed individuals and 70% of left-handed individuals. Destruction of the dominant cerebral hemisphere in adults results in loss of nearly all intellectual function.
The historical failure to document an important role of prefrontal lobes in intellectual function (frontal lobotomy) is surprising because the principal difference between the brains of humans and monkeys is the prominence of human prefrontal areas. It seems that the function of the prefrontal areas in humans is to provide additional cortical area in which thought processing can occur. Furthermore, selection of behavior patterns for different situations may be an important role of the prefrontal areas that transmit signals to the limbic areas of the brain. Persons without prefrontal lobes may react precipitously in response to incoming signals or manifest undue anger at slight provocations. Ability to maintain a sustained level of concentration is lost in the absence of the prefrontal lobes.
Memory
The cerebral cortex, especially the temporal lobes, serves as a storage site for information that is often characterized as memory.28 The mechanisms for short-term and long-term memory are not completely understood but are thought to be encoded through selective synaptic strengthening in response to experience.
Short-Term Memory
The favored explanation for short-term memory is posttetanic potentiation. For example, tetanic stimulation of a synapse for a few seconds causes increased excitability of the synapse that lasts for seconds to hours. This change in excitability of the synapse is mediated by increased local intracellular calcium concentrations that facilitate transmitter release and act as a second messenger to activate genetic programs that result in structural synaptic stabilization.
Long-Term Memory
Long-term memory depends on stable synaptic changes that are induced by experience. The stability of this system is evidenced by total inactivation of the brain by hypothermia or anesthesia without detectable significant loss of long-term memory. Long-term memory is thought to rely on long-term synaptic potentiation mediated by structural changes. Long-term potentiation is the enhanced synaptic transmission observed after repeatedly stimulating a presynaptic neuron. The mechanism often involves increased expression of NMDA receptors and voltage-gated calcium channels in the postsynaptic neuron.29 Thus, protein transcription and synaptic remodeling are an essential component of long-term memory. The hippocampus and amygdala are critically involved in creating new long-term memories. However, long-term memories are not actually stored in the hippocampus and amygdala. Sleep is known to play an important role in the formation of long-term memory.30 However, the actual mechanism by which long-term memories are stored remains a fascinating unsolved puzzle.
Everyone knows from personal experience that repetition is essential to forming long-term memory. There is an old joke about a man asking a fellow pedestrian in New York, “How do you get to Carnegie Hall?” The pedestrian replies, “Practice, practice, practice.” It has been repeatedly demonstrated in animal studies as well that repetition is key to forming long-term memories. Long-term potentiation is the synaptic consequence of repeated stimulation, which is one reason that long-term potentiation is thought to be the fundamental building block of long-term memory.
We also know that memories are transferred from short-term memory to long-term memory. Because the creation of long-term memory requires anatomic changes in the synapse, this transfer requires time. This suggests, and studies confirm, that if the brain is not given adequate time to make this transfer, there will be no transfer from short-term memory to long-term memory. This has direct applicability to the practice of anesthesia. During the provision of general anesthesia, we are vigilant for signs of inadequate anesthesia and intraoperative awareness (discussed further at the end of this section). If a patient has conscious perception of the surgery, this will initially be part of the patient’s short-term memory. Rapid deepening of the anesthesia, for example by administering a bolus of propofol in response to patient movement, will prevent transfer of the recall from short-term memory to the long-term memory, and the patient will be amnestic. Conversely, if the patient is paralyzed and is awake for many minutes without the anesthesiologist being aware of the situation, then there has been adequate time for transfer of the short-term memory to long-term memory.
Because the neural substrate of memory is not well understood, memory is often discussed from a psychological point of view. Memories typically involve multiple senses (sight, hearing, touch), emotions (fear, satisfaction, pleasure, anger), and cognitive assessment (“I remember thinking that. . .”). These are thought to be held together in a facilitated circuit that has been called a memory engram or memory trace. Initially the circuit is facilitated through posttetanic potentiation in short-term memory. If memory is to persist, this is replaced with long-term potentiation. The pieces of the engram are consolidated through hypothalamic circuitry. The memory engram is reinforced with every subsequent recall of the memory. An important feature of the process of consolidation is that long-term memory is encoded into different categories. New memories are not stored randomly in the brain but seem to be associated with previously encoded and similar information. This permits scanning of memory to retrieve desired information at a later date. We also know that memory scanning is often a subconscious process. This is confirmed by the daily experience of struggling to recall a fact or event, only to have the memory suddenly jump into our consciousness hours later.
Postoperative Cognitive Dysfunction
Postoperative cognitive dysfunction (impaired memory) persisting after 3 months has been described in 10% of elderly patients receiving general anesthesia without known arterial hypoxemia or systemic hypotension.31 Inhaled anesthetics are known to alter the proteins involved in the formation of Alzheimer’s disease.32 It is unclear whether the postoperative cognitive dysfunction is caused by anesthetic injury to the aged brain, as might be caused by increasing the polymerization of β amyloid, or is caused by the combined effects of surgical trauma, inflammation, social interruption, anesthesia, and other unidentified causes.
Awareness and Recall during Anesthesia
Awareness, defined as conscious memory of events during anesthesia, has been a recurrent problem particularly since the introduction of neuromuscular-blocking drugs.33 Neuromuscular blocking drugs permit inadequate anesthesia to be administered without obvious patient withdrawal from the noxious stimulus. The use of neuromuscular blockade is a risk factor for awareness under general anesthesia, particularly awareness that is associated with memories of pain and complicated by posttraumatic stress disorder.34
Memory may be considered to be conscious (explicit) or unconscious (implicit). Conscious memory includes spontaneous recall and recognition memory. Unconscious memory is manifest by altered performance or behavior due to experiences that are not consciously remembered. By definition, general anesthesia abolishes conscious memory, but the extent to which it also abolishes unconscious memory is controversial. Behavioral disturbances manifest as night terrors in children after anesthesia may be an expression of implicit memory in the dream state.
The incidence of awareness with recall (conscious memory) following general anesthesia has been estimated at between 1 and 5 in 1,000 general anesthetics, depending on the risk group.35–37 Although the incidence of conscious recall of intraoperative events is rare and the development of posttraumatic stress disorder is even more uncommon, the fact that approximately 20 million general anesthetics are administered annually in the United States would correspond to 26,000 cases of awareness (0.13% of approximately 20 million) each year. The incidence of awareness in patients undergoing cesarean section was 0.4% and for cardiac surgery was 1.14% to 1.50% .38,39 A higher incidence of awareness has been described for major trauma cases (11% to 43%) where the concentration of anesthetic administered is limited by hemodynamic instability.40 Many cases of conscious awareness during surgery can be attributed to intentionally or unintentionally low concentrations of administered anesthetic.
Subanesthetic doses of inhaled anesthetics have powerful inhibitory effects on short-term memory, and the decrease in the transfer of information from the periphery to the cerebral cortex associated with general anesthesia prevents the recall of intraoperative events.8
Isoflurane (and presumably other volatile anesthetics) and nitrous oxide suppress memory in a dose-dependent manner, and isoflurane is more potent than equivalent concentrations of nitrous oxide (Fig. 3-15).41 For example, conscious memory was prevented by 0.45 minimum alveolar concentration to prevent movement (MAC) isoflurane or 0.6 MAC nitrous oxide. Isoflurane concentrations of ≥0.6 MAC prevent conscious recall and unconscious learning of factual information and behavioral suggestions.42
Recognizing Awareness
Monitoring patients during general anesthesia for the presence of awareness is challenging. Despite a variety of monitoring methods, awareness may be difficult to recognize in real time. Indicators of awareness (heart rate, blood pressure, and skeletal muscle movement) are often masked by anesthetic and adjuvant drugs (β-adrenergic blockers and/or neuromuscular-blocking drugs). Several different monitors, based on analysis of electroencephalogram (EEG) and somatosensory evoked potential patterns, have been introduced in hopes of addressing this issue.
Brainstem
Homeostatic life-sustaining processes are controlled subconsciously in the brainstem. Examples of subconscious activities of the body regulated by the brainstem include control of systemic blood pressure and breathing in the medulla. The thalamus serves as a relay station for most afferent impulses before they are transmitted to the cerebral cortex. The hypothalamus receives fibers from the thalamus and is also closely modulated by the cerebral cortex.
Limbic System and Hypothalamus
Behavior associated with emotions is primarily a function of structures known as the limbic system (hippocampus, basal ganglia) located in the basal regions of the brain. The hypothalamus functions in many of the same roles as the limbic system and is considered by some to be part of the limbic system rather than a separate structure. In addition, the hypothalamus controls many internal conditions of the body, such as core temperature, thirst, and appetite. The great Oxford neurophysiologist Sir Charles Sherrington called the hypothalamus the head ganglion of the autonomic nervous system. The suprachiasmatic nucleus of the hypothalamus helps to maintain the body clock by secreting melatonin and other mediators according to the circadian rhythm. This nucleus sits just above the optic chiasm and receives inputs from the optic nerve that serve to entrain the circadian rhythm to environmental light. At high doses, melatonin and its analogs have properties similar to a general anesthetic.43
Basal Ganglia
The basal ganglia include the caudate nucleus, putamen, globus pallidus, substantia nigra, and subthalamic nucleus. Many of the impulses from basal ganglia are inhibitory mediated by dopamine and GABA. The balance between agonist and antagonist skeletal muscle contractions is an important role of the basal ganglia. A general effect of diffuse excitation of the basal ganglia is inhibition of skeletal muscles, reflecting transmission of inhibitory signals from the basal ganglia to both the motor cortex and the lower brainstem. Therefore, whenever destruction of the basal ganglia occurs, there is associated skeletal muscle rigidity. For example, damage to the caudate and putamen nuclei that normally secrete GABA results in choreiform random and continuous uncontrolled movements. Destruction of the substantia nigra and loss of dopaminergic neurons results in a predominance of the excitatory neurotransmitter acetylcholine, manifesting as the skeletal muscle rigidity of Parkinson’s disease. As such, dopamine precursors or anticholinergic drugs are used in the treatment of Parkinson’s disease in an attempt to restore the balance between excitatory and inhibitory impulses traveling from the basal ganglia.
Reticular Activating System
The reticular activating system is a polysynaptic pathway that is intimately concerned with electrical activity of the cerebral cortex. Neurons of the reticular activating system are both excitatory and inhibitory. The reticular activating system determines the overall level of CNS activity, including nuclei important in determining wakefulness and sleep. Selective activation of certain areas of the cerebral cortex by the reticular activating system is crucial for the direction of the attention of certain aspects of mental activity. It is likely that many injected and inhaled anesthetics exert their sedative effects through interaction with the brainstem and midbrain nuclei that mediate arousal and sleep.44 This is not to say that general anesthesia is equivalent to sleep. Although the EEG response to many anesthetics resembles deep slow-wave sleep, a key difference is that afferent stimulation does not cause arousal.
Slow-Wave Sleep
Most of the sleep that occurs each night is slow-wave sleep. The EEG is characterized by the presence of high-voltage δ waves occurring at a frequency of <4 cycles per second. Presumably, decreased activity of the reticular activating system that accompanies sleep permits an unmasking of this inherent rhythm in the cerebral cortex. Slow-wave sleep is restful and devoid of dreams. During slow-wave sleep, sympathetic nervous system activity decreases, parasympathetic nervous system activity increases, and skeletal muscle tone is greatly decreased. As a result, there is a 10% to 30% decrease in systemic blood pressure, heart rate, breathing frequency, and basal metabolic rate.
Desynchronized Sleep
Periods of desynchronized sleep typically occur for 5 to 20 minutes during each 90 minutes of sleep. These periods tend to be shortest when the person is extremely tired. This form of sleep is characterized by active dreaming, irregular heart rate and breathing, and a desynchronized pattern of low-voltage β waves on the EEG similar to those that occur during wakefulness. This brain wave pattern emphasizes that desynchronized sleep is associated with an active cerebral cortex, but this activity does not permit persons to be aware of their surroundings and thus be awake. Despite the inhibition of skeletal muscle activity, the eyes are an exception, exhibiting rapid movements. For this reason, desynchronized sleep is also referred to as paradoxical sleep or rapid eye movement (REM) sleep.
Cerebellum
The cerebellum operates subconsciously to monitor and elicit corrective responses in motor activity caused by stimulation of other parts of the brain and spinal cord. Rapid repetitive skeletal muscle activities, such as typing, playing musical instruments, and running, require intact function of the cerebellum. Loss of function of the cerebellum causes incoordination of fine skeletal muscle activities even though paralysis of the skeletal muscles does not occur. The cerebellum is also important in the maintenance of equilibrium and postural adjustments of the body. For example, sensory signals are transmitted to the cerebellum from receptors in muscle spindles, Golgi tendon organs, and receptors in skin joints. These spinocerebellar pathways can transmit impulses at velocities exceeding 100 m per second, which is the most rapid conduction of any pathway in the CNS. This extremely rapid conduction is important for instantaneous appraisal by the cerebellum of changes that take place in the positional status of the body.
Dysfunction of the Cerebellum
In the absence of cerebellar function, a person cannot predict prospectively how far movements will go. This results in overshoot of the intended mark (past pointing). This overshoot is known as dysmetria, and the resulting incoordinate movements are called ataxia. Dysarthria is present when rapid and orderly succession of skeletal muscle movements of the larynx, mouth, and chest do not occur. Failure of the cerebellum to dampen skeletal muscle movements results in intention tremor when a person performs a voluntary act. Cerebellar nystagmus is associated with loss of equilibrium, presumably because of dysfunction of the pathways that pass through the cerebellum from the semicircular canals. In the presence of cerebellar disease, a person is unable to activate antagonist skeletal muscles that prevent a certain portion of the body from moving unexpectedly in an unwanted direction. For example, a person’s arm that was previously contracted but restrained by another person will move back rapidly when it is released rather than automatically remain in place.
Spinal Cord
The spinal cord extends from the medulla oblongata to the lower border of the first and, occasionally, the second lumbar vertebra. Below the spinal cord, the vertebral canal is filled by the roots of the lumbar and sacral nerves, which are collectively known as the cauda equina. The spinal cord is composed of gray and white matter, spinal nerves, and covering membranes.
Gray Matter
The gray matter of the spinal cord functions as the initial processor of incoming sensory signals from peripheral somatic receptors and as a relay station to send these signals to the brain.
In addition, this area of the spinal cord is the site for final processing of motor signals that are being transmitted downward from the brain to skeletal muscles. Anatomically, the gray matter of the spinal cord is divided into anterior, lateral, and dorsal horns consisting of nine separate laminae that are H-shaped when viewed in cross-section (Fig. 3-16). The anterior horn is the location of α and γ motor neurons that give rise to nerve fibers that leave the spinal cord via the anterior (ventral) nerve roots and innervate skeletal muscles. Cells of Renshaw are intermediary neurons in the anterior horn, providing nerve fibers that synapse in the gray matter with anterior motor neurons. These cells inhibit the action of anterior motor neurons to limit excessive activity. Cells of the preganglionic neurons of the sympathetic nervous system are located lateral to the thoracolumbar portions of the spinal cord. Cells of the intermediate neurons located in the portion of the dorsal horns of the spinal cord known as the substantia gelatinosa(laminae II to III) transmit afferent tactile, temperature, and pain impulses to the spinothalamic tract. The dorsal horn serves as a gate where impulses in sensory nerve fibers are translated into impulses in ascending tracts. There is evidence for a form of memory in the dorsal horn of the spinal cord that is evoked by intense stimulation. Resulting increases in intracellular calcium set into motion long-lasting changes that are associated with central sensitization and result in increased sensitivity to subsequent inoffensive stimuli.
White Matter
The white matter of the spinal cord is formed by the axons that make up their respective ascending and descending tracts. This area of the spinal cord is divided into dorsal, lateral, and ventral columns (see Fig. 3-16). The dorsal column of the spinal cord is composed of spinothalamic tracts that transmit touch and pain impulses to the brain.
Pyramidal and Extrapyramidal Tracts
A major pathway for transmission of motor signals from the cerebral cortex to the anterior motor neurons of the spinal cord is through the pyramidal (corticospinal) tracts (Fig. 3-17).3 All pyramidal tract fibers pass downward through the brainstem and then cross to the opposite side to form the pyramids of the medulla. After crossing the midline at the level of the medulla, these fibers descend in the lateral corticospinal tracts of the spinal cord and terminate on motor neurons in the dorsal horn of the spinal cord. A few fibers do not cross to the opposite side of the medulla but rather descend in the ventral corticospinal tracts. In addition to these pyramidal fibers, a large number of collateral fibers pass from the motor cortex into the basal ganglia, forming the extrapyramidal tracts. Extrapyramidal tracts are all those tracts beside the pyramidal tracts that transmit motor impulses from the cerebral cortex to the spinal cord.
The pyramidal and extrapyramidal tracts have opposing effects on the tone of skeletal muscles. For example, the pyramidal tracts cause continuous facilitation and therefore a tendency to produce increases in skeletal muscle tone. Conversely, the extrapyramidal tracts transmit inhibitory signals through the basal ganglia with resultant inhibition of skeletal muscle tone. Selective or predominant damage to one of these tracts manifests as spasticity or flaccidity.
Babinski Sign
A positive Babinski sign is characterized by upward extension of the first toe and outward fanning of the other toes in response to a firm tactile stimulus applied to the dorsum of the foot. A normal response to the same tactile stimulus is downward motion of all the toes. A positive Babinski sign reflects damage to the pyramidal tracts. Damage to the extrapyramidal tracts does not cause a positive Babinski sign.
Thalamocortical System
The thalamocortical system serves as the pathway for passage of nearly all afferent impulses from the cerebellum; basal ganglia; and visual, auditory, taste, and pain receptors as they pass through the thalamus on the way to the cerebral cortex. Signals from olfactory receptors are the only peripheral sensory signals that do not pass through the thalamus. Overall, the thalamocortical system controls the activity level of the cerebral cortex.
Spinal Nerve
A pair of spinal nerves arises from each of 31 segments of the spinal cord. Spinal nerves are made up of fibers of the ventral (anterior) and dorsal (posterior) roots. Efferent motor fibers travel in the anterior roots that originate from axons in the anterior and lateral horns of the spinal cord gray matter. Sensory fibers travel in the dorsal nerve roots that originate from axons that arise from cell bodies in the spinal cord ganglia. These cell bodies send branches to the spinal cord and to the periphery. The anterior and dorsal nerve roots each leave the spinal cord through an individual intervertebral foramen enclosed in a common dural sheath that extends just past the spinal cord ganglia where the spinal nerve originates.
Each spinal nerve innervates a segmental area of skin designated a dermatome and an area of skeletal muscle known as a myotome. A dermatome map is useful in determining the level of spinal cord injury or level of sensory anesthesia produced by a neuraxial anesthetic (Fig. 3-18).3 Despite common depictions of dermatomes as having distinct borders, there is extensive overlap between segments. For example, three consecutive dorsal nerve roots need to be interrupted to produce complete denervation of a dermatome. The scrotum has considerable sensory overlap, with innervation coming from T1 (variable) and L1–L2 and S2–S4 despite common depictions on dermatome charts as being limited to sacral innervation.45 Segmental innervation of myotomes is even less well defined than that of dermatomes, emphasizing that skeletal muscle groups receive innervation from several anterior nerve roots.
Sensory signals from the periphery are transmitted through spinal nerves into each segment of the spinal cord, resulting in automatic motor responses that occur instantly (muscle stretch reflex, withdrawal reflex) in response to sensory signals. Spinal cord reflexes are important in emptying the bladder and rectum. Segmental temperature reflexes allow localized cutaneous vasodilation or vasoconstriction in response to changes in skin temperature. The function of the spinal cord component of the CNS and spinal cord reflexes is particularly apparent in patients with transection of the spinal cord.
Covering Membranes
The spinal cord is enveloped by membranes (dura, arachnoid, pia) that are direct continuations of the corresponding membranes surrounding the brain. The dura consists of an inner and an outer layer. The outer periosteal layer in the cranial cavity is the periosteum of the skull, whereas this layer in the spine is the periosteal lining of the spinal cord. The epidural space is located between the inner and outer layers of the dura. The fact that the inner layer of the dura adheres to the margin of the foramen magnum and blends with the periosteal layer means that the epidural space does not extend beyond this point. As a result, drugs such as local anesthetics or opioids cannot travel cephalad in the epidural space beyond the foramen magnum. However, there is extensive equilibration between epidural and subarachnoid drug concentrations. Because of this equilibration hydrophilic opioids such as morphine given to the lumbar epidural space may cause delayed respiratory depression in patients at risk. The inner layer of the dura extends as a dural cuff that blends with the perineurium of spinal nerves. The cerebral arachnoid extends as the spinal arachnoid, ending at the second sacral vertebra. The pia is in close contact with the spinal cord.
CT scans demonstrate the occasional presence of a connective tissue band (dorsomedian connective tissue band or plica mediana dorsalis) that divides the epidural space at the dorsal midline.46 This band binds the dura mater and the ligamentum flavum at the midline, making it difficult to feel loss of resistance during attempted midline identification of the epidural space. The band may also explain the occasional occurrence of unilateral analgesia after injection of local anesthetic solutions into the epidural space.47 In some patients, there is a failure of midline fusion of the dura. This is particularly common in at higher thoracic levels.48
Autonomic Reflexes
Segmental autonomic reflexes occur in the spinal cord and include changes in vascular tone, diaphoresis, and evacuation of the bladder and colon. Simultaneous excitation of all the segmental reflexes is the mass reflex (denervation hypersensitivity or autonomic hyperreflexia). The mass reflex typically occurs in the presence of spinal cord transection when a painful stimulus is applied to the skin below the level of the spinal cord transection, or following distension of a hollow viscus, such as the bladder or gastrointestinal tract. The principal manifestation of the mass reflex is systemic hypertension due to intense peripheral vasoconstriction, reflecting an inability of vasodilating inhibitory impulses from the CNS to pass beyond the site of spinal cord transection. Carotid sinus baroreceptor-mediated reflex bradycardia accompanies the systemic hypertension associated with the mass reflex.
Spinal Shock
Spinal shock is a manifestation of the abrupt loss of spinal cord reflexes that immediately follows transection of the spinal cord. It emphasizes the dependence of spinal cord reflexes on continual tonic discharges from higher centers. The immediate manifestations of spinal shock are hypotension due to loss of vasoconstrictor tone and absence of all skeletal muscle reflexes. Within a few days to weeks, spinal cord neurons gradually regain their intrinsic excitability. Sacral reflexes for control of bladder and colon evacuation are completely suppressed for the first few weeks after spinal cord transection, but these spinal cord reflexes also eventually return, although their conscious control does not.
Imaging of the Nervous System
Until the introduction of computed tomography (CT), imaging studies of the brain included skull radiography, cerebral angiography, and pneumoencephalography.49 These techniques allowed only examination of the skull, the cerebral blood vessels, and the fluid-containing spaces of the brain. CT and magnetic resonance imaging (MRI) provide high-resolution images of brain tissue and clear discrimination between gray and white matter. Positron emission tomography (PET) and single photon emission computed tomography (SPECT) permit imaging of both structure and functional characteristics (blood flow, metabolism, and concentrations of neurochemicals and receptors) of the brain.
Comparative studies indicate that MRI is superior to CT in evaluating most cerebral parenchymal lesions because of better spacial discrimination.49 CT is used in patients who cannot undergo MRI because of the presence of artificial cardiac pacemakers, mechanical heart valves, or magnetizable intracranial metal clips. CT is also useful in visualizing intracranial blood that may be present in patients with subdural hematomas or cerebral hemorrhage.
Cerebral Blood Flow
Cerebral blood flow averages 50 mL/100 g per minute of brain tissue. For an adult, this is equivalent to 750 mL per minute, or about 15% of the resting cardiac output, delivered to an organ that represents only about 2% of the body’s mass. The gray matter of the brain has a higher cerebral blood flow (80 mL/100 g per minute) than the white matter (20 mL/100 g per minute). As in most other tissues of the body, cerebral blood flow parallels cerebral metabolic requirements for oxygen (3 to 5 mL/100 g per minute). PaCO2 and PaO2 influence cerebral blood flow, whereas sympathetic and parasympathetic nerves play little or no role in the regulation of cerebral blood flow (Fig. 3-19). Changes in the PaCO2 between about 20 and 80 mm Hg produce corresponding changes in cerebral blood flow. For example, in this range, a 1-mm Hg increase in the PaCO2 evokes a 1 to 2 mL/100 g per minute increase in cerebral blood flow (Table 3-3).50
Carbon dioxide increases cerebral blood flow by combining with water in body fluids to form carbonic acid, with subsequent dissociation to form hydrogen ions. Hydrogen ions produce vasodilation of cerebral vessels that is proportional to the increase in hydrogen ion concentration.
Any other acid that increases hydrogen ion concentration, such as lactic acid, also increases cerebral blood flow. Increased cerebral blood flow in response to increases in PaCO2 serves to carry away excess hydrogen ions that would otherwise greatly depress neuronal activity.
Unlike the continuous response of cerebral blood flow to changes in PaCO2, the response to Pao2 is a threshold phenomenon (see Fig. 3-19). If the PaCO2 is maintained, cerebral blood flow begins to increase when the PaO2decreases below 50 mm Hg or the cerebral venous PO2 decreases from its normal value of 35 mm Hg to about 30 mm Hg.
Autoregulation
Cerebral blood flow is closely autoregulated between a mean arterial pressure of about 60 and 140 mm Hg (see Fig. 3-19). As a result, changes in systemic blood pressure within this range will not significantly alter cerebral blood flow. Chronic systemic hypertension shifts the autoregulation curve to the right such that decreases in cerebral blood flow may occur at a mean arterial pressure of >60 mm Hg. Autoregulation of cerebral blood flow is attenuated or abolished by hypercapnia, arterial hypoxemia, and volatile anesthetics. Furthermore, autoregulation is often abolished in the area surrounding an acute cerebral infarction. For example, reactivity of blood vessels in areas surrounding cerebral infarcts and tumors is abolished. These blood vessels are maximally vasodilated, presumably reflecting accumulation of acidic metabolic products. As a result, cerebral blood flow to this area is already maximal (luxury perfusion), and changes in PaCO2 have no effect on its local blood flow. If PaCO2 should increase, however, it is theoretically possible that resulting vasodilation in normal blood vessels would shunt blood flow away from the diseased area (intracerebral steal syndrome). Conversely, a decrease in PaCO2 that constricts normal cerebral vessels could divert blood flow to diseased areas (“Robin Hood” phenomenon). Increases in mean arterial pressure above the limits of autoregulation can cause leakage of intravascular fluid through capillary membranes, resulting in cerebral edema. Because the brain is enclosed in a solid vault, the accumulation of edema fluid increases intracranial pressure and compresses blood vessels, decreasing cerebral blood flow and leading to destruction of brain tissue.
Measurement of Cerebral Blood Flow
Cerebral blood flow can be measured by injecting a radioactive substance, usually xenon, into the carotid artery and measuring the rate of decay of the radioactivity in each tissue segment using scintillation detectors. Using this technique, it can be demonstrated that cerebral blood flow changes within seconds in response to changes in local neuronal activity. For example, clasping the hand can be shown to cause an immediate increase in blood flow in the motor cortex of the opposite cerebral hemisphere. Reading increases blood flow in the occipital cortex and the language areas of the temporal cortex. This measuring procedure can be used to localize the origin of epilepsy because blood flow increases acutely at the site of origin of the seizure.
Electroencephalogram
The EEG is a recording of the brain waves that result from the summed electrical activity in the brain. The intensity of the electrical activity recorded from the surface of the scalp ranges from 0 to 300 µV, and the frequency may exceed 50 cycles per second. The character of the waves greatly depends on the level of activity of the cerebral cortex and the degree of wakefulness. There is a direct relationship between the degree of cerebral activity and the frequency of brain waves. Furthermore, during periods of increased mental activity, brain waves become asynchronous rather than synchronous, so the voltage decreases despite greater cortical activity.
Classification of Brain Waves
Brain waves are classified as α, β, θ, and δ waves depending on their frequency and amplitude (Fig. 3-20). The classic EEG is a plot of voltage against time, usually recorded by 16 channels on paper moving at 30 mm per second. One page of recording is 10 seconds of data.
α Waves
α waves occur at a frequency of 8 to 12 Hz and a voltage of about 50 µV. These waves are typical of an awake, resting state of cerebration with the eyes closed. During sleep, α waves disappear. Because α waves do not occur when the cerebral cortex is not connected to the thalamus, it is assumed these waves result from spontaneous activity in the thalamocortical system.
β Waves
β waves occur at a frequency of 13 to 30 Hz and a voltage usually of <50 µV. These high-frequency and low-voltage asynchronous waves replace α waves in the presence of increased mental activity or visual stimulation.
θ Waves
θ waves occur at a frequency of 4 to 7 Hz. These waves occur in healthy children during sleep and also during general anesthesia.
δ Waves
δ waves include all the brain waves with a frequency of less than 4 Hz. These waves occur (a) in deep sleep, (b) during general anesthesia, and (c) in the presence of organic brain disease. d waves occur even when the connections of the cerebral cortex to the reticular activating system are severed, indicating these waves originate in the cerebral cortex independently of lower brain structures.
Clinical Uses
The EEG is useful in diagnosing different types of epilepsy and for determining the focus in the brain causing seizures. Brain tumors, which compress surrounding neurons and cause abnormal electrical activity, may be localized using the EEG. Monitoring of the EEG during carotid endarterectomy, cardiopulmonary bypass, or controlled hypotension may provide an early warning of inadequate cerebral blood flow. In this regard, the EEG may be influenced by anesthetic drugs, depth of anesthesia, and hyperventilation of the patient’s lungs. Several different monitors of EEG activity that use different algorithms designed to process EEG recordings and decompose them into a number that may be predictive of anesthetic depth.
Brain Wave Monitors
Numerous quantitative EEG processing techniques have been developed to monitor brain depression during anesthesia, including Bispectral Index, Narcotrend, SEDLine, and Entropy monitors. These are discussed in hundreds of manuscripts and review articles. Only two will be presented here.
Bispectral Index
The Bispectral Index (BIS) is a variable derived from the EEG that is a quantifiable measure of the sedative and hypnotic effects of anesthetic drugs on the CNS.51 BIS is a processed EEG descriptor that predicts depth of anesthesia. Bispectral analysis is based on the correlation of the phase between different frequency components of the EEG in which the EEG signal is converted into its component sine waves using Fourier transformation. Electromyographic activity is specifically filtered with modern BIS algorithm but can still result in artifact. A set of bispectral features is calculated by analyzing the phase relations between the component waves. These bispectral features are combined with other EEG features into a single measurement, the BIS, expressed as a dimensionless numerical index from 0 to 100. Decreasing numerical values correlate with sedation and predict the response of patients to surgical stimulation (values of <60 are associated with a low probability of recall and a high probability of unresponsiveness during surgery) (Fig. 3-21).52,53 Titrating desflurane and sevoflurane using the BIS monitor to maintain a numerical value of 60 results in decreased use of drug and faster awakening.54Likewise, titration of propofol to maintain a numerical value of 45 to 60 and then permitting an increase to 60 to 75 during the last 15 minutes of the operation results in decreased propofol use and more rapid recovery.55 In this regard, BIS monitoring may serve as a useful intraoperative monitor for guiding drug administration, particularly for intravenous hypnotics (e.g., propofol).
Based on published studies, the FDA determined that use of BIS monitoring to guide anesthetic administration may be associated with a reduction of the incidence of awareness with recall in adults during general anesthesia and sedation.36,56,57 However, these findings have been challenged by a recent study that found that the BIS monitor performed similarly to rigorous monitoring of end-tidal inhaled anesthetic concentration37,58 in preventing awareness. It may be that monitors of processed EEG are pharmacodynamic monitors of the complex interplay between the concentration of anesthetic agents and surgical stimulation, and thus the use of monitoring may be a function of the anesthetic technique, the drugs used, and the availability of methodology to easily the concentration in the patient.
Spectral Entropy
Spectral entropy (SE) represents an alternative concept to bispectral analysis for quantifying the EEG. SE and response entropy (RE) are computed over specific frequency ranges of the EEG. RE includes electromyographic activity. SE, RE, and BIS reveal similar information about the level of sedation.59 BIS and SE measurement are similar during propofol anesthesia. However, they are not interchangeable. For example, SE measurements are lower than BIS measurements during anesthesia with xenon.60
Epilepsy
Epilepsy is characterized by excessive activity of either a part or all of the CNS. Grand mal epilepsy is characterized by intense neuronal discharges in multiple areas of the cerebral and reticular activating system. These impulses are transmitted to the spinal cord, resulting in alternating skeletal muscle contractions known as tonic-clonic seizures. Profound autonomic activity often results in defecation and urination. The grand mal seizure lasts from a few seconds to several minutes and is followed by generalized depression of the entire CNS (the postictal state). The EEG during a grand mal seizure reveals high-voltage, synchronous brain wave discharges over the entire cerebral cortex. Synaptic fatigue is a likely mechanism that contributes to spontaneous cessation of a grand mal seizure and postictal depression.
Status epilepticus is present when grand mal seizure activity is sustained. Judicious doses of an intravenous sedative hypnotics can stop seizures and permit resumption of effective breathing. In the rare instance in which conventional drug therapy is ineffective, volatile anesthetics such as isoflurane may be administered in an attempt to stop status epilepticus.61 When volatile anesthetics are administered for this purpose, it is likely that systemic blood pressure will need to be supported with intravenous administration of fluids and/or sympathomimetics. If the underlying cause of the seizure has not been addressed then the seizure is likely to recur when the volatile anesthetic is discontinued.
Evoked Potentials
Evoked potentials are the electrophysiologic responses of the CNS to sensory, motor, auditory, or visual stimulation. The waveforms resulting from sensory stimulation reflect transmission of impulses through specific sensory pathways. Poststimulus latency is the time in milliseconds from application of the stimulus to a peak in the recorded waveform. The amplitude and latency of evoked potentials may be influenced by a number of events, especially volatile anesthetics. Evoked potentials are used to monitor (a) spinal cord function during operations near or on the spinal cord, and (b) auditory nerve and brainstem function, as during operations on pituitary tumors or other lesions that impinge on the optic nerves or optic chiasm. The modes of sensory stimulation used to produce evoked potentials in the operating room are somatosensory, auditory, and visual.
Somatosensory Evoked Potentials
Somatosensory evoked potentials are produced by application of a low-voltage electrical current that stimulates a peripheral nerve such as the median nerve at the wrist or the posterior tibial nerve at the ankle. The resulting evoked potentials reflect the integrity of sensory neural pathways from the peripheral nerve to the somatosensory cortex. Somatosensory stimulation follows the dorsal column pathways of proprioception and vibration. These pathways are supplied by the posterior spinal artery, leaving the motor pathway, which is supplied by the anterior spinal artery, unmonitored. Indeed, postoperative paraplegia has been described in patients despite the preservation of somatosensory evoked potentials intraoperatively.62 Inhaled anesthetics, especially volatile anesthetics, produce dose-dependent depression of somatosensory evoked potentials (see Chapter 4). Although less so than volatile anesthetics, morphine and fentanyl also produce depressant effects on somatosensory evoked potentials, with a low-dose continuous infusion of the opioid producing less depression than intermittent injections (Fig. 3-22).63 Ketamine and etomidate may increase the amplitude of somatosensory evoked potentials (see Chapter 5). Acute hyperventilation of the patient’s lungs to produce a PaCO2 near 20 mm Hg does not significantly alter the amplitude or latencies of somatosensory evoked potentials.64
Motor Evoked Potentials
The use of motor evoked potentials remains limited, as their recording requires direct (epidural) or indirect (transosseous) stimulation of the brain or spinal cord.65 These evoked potentials reflect the integrity of motor neural pathways from the motor cortex to the muscle. Motor evoked potentials are extremely sensitive to depression by anesthetics. Furthermore, it is not possible to monitor motor evoked potentials in the presence of significant drug-induced neuromuscular blockade. During scoliosis surgery or other operations that place spinal cord motor function at risk, the use of motor evoked potentials obviates the need for an intraoperative wake-up test. In many instances, it is useful to monitor both motor and sensory evoked potentials to fully evaluate the functional integrity of both motor and sensory pathways. As an alternative to motor evoked potentials, transcranial motor stimulation may be used to monitor spinal cord function during spinal surgery. Total intravenous anesthesia with propofol and an opioid with judicious infusion of neuromuscular blocker is a useful technique when monitoring of somatosensory and motor evoked potentials is desired.
Auditory Evoked Potentials
Auditory evoked potentials arise from brainstem auditory pathways. Volatile anesthetics produce dose-dependent depression of auditory evoked potentials. Auditory evoked potentials may provide an objective electrophysiologic alternative to the clinical assessment of sedation.66
Visual Evoked Potentials
Visual evoked potentials are produced by flashes from light-emitting diodes that are mounted on goggles placed over the patient’s closed eyes. Visual evoked potentials may be useful to monitor the visual pathways during transphenoidal or anterior fossa neurosurgical procedures. Volatile anesthetics produce dose-dependent depression of visual evoked potentials, especially above concentrations equivalent to about 0.8 MAC.67
Cerebrospinal Fluid
Cerebrospinal fluid (CSF) is present in the (a) ventricles of the brain, (b) cisterns around the brain, and (c) subarachnoid space around the brain and spinal cord (Fig. 3-23). The total volume of CSF is about 150 mL and the specific gravity is 1.002 to 1.009. A major function of CSF is to cushion the brain in the cranial cavity. A blow to the head moves the entire brain simultaneously, causing no one portion of the brain to be selectively contorted by the blow. When a blow to the head is particularly severe, it usually does not damage the brain on the ipsilateral side, but instead damage manifests on the opposite side. This phenomenon is known as contrecoup and reflects the creation of a vacuum between the brain and skull opposite the blow caused by sudden movement of the brain at this site away from the skull. When the skull is no longer being accelerated by the blow, the vacuum suddenly collapses and the brain strikes the interior of the skull.
Formation
The choroid plexuses (cauliflower-like growths of blood vessels covered by a thin layer of epithelial cells) in the four cerebral ventricles are the major site of formation of CSF, which continually exudes from the surface of the choroid plexus at a rate of about 30 mL per hour. In comparison with other extracellular fluids, the concentration of sodium and chloride in CSF is 7% greater and the concentration of glucose and potassium is 30% and 40% less, respectively. This difference in composition from other extracellular fluids emphasizes that CSF is a choroid secretion and not a simple filtrate from the capillaries. The pH of CSF is closely regulated and maintained at 7.32. Changes in PaCO2, but not arterial pH, promptly alter CSF pH, reflecting the ability of carbon dioxide, but not hydrogen ions, to cross the blood–brain barrier easily. As a result, acute respiratory acidosis or alkalosis produces corresponding changes in CSF pH. Active transport of bicarbonate ions eventually returns CSF pH to 7.32, despite the persistence of alterations in arterial pH.
Reabsorption
Almost all the CSF formed each day is reabsorbed into the venous circulation through special structures known as arachnoid villi or granulations. These villi project the subarachnoid spaces into the venous sinuses of the brain and occasionally into veins of the spinal cord. Arachnoid villi are actually trabeculae that protrude through venous walls, resulting in highly permeable areas that permit relatively free flow of CSF into the circulation. The magnitude of reabsorption depends on the pressure gradient between the CSF and the venous circulation.
Circulation
CSF formed in the lateral cerebral ventricles passes into the third ventricle through the foramen of Monro (see Fig. 3-23), where it mixes with CSF formed there. From there, it passes along the aqueduct of Sylvius into the fourth cerebral ventricle, where still more CSF is formed. The CSF then passes into the cisterna magna through the lateral foramen of Luschka and via a middle foramen of Magendie. From this point, CSF flows through the subarachnoid spaces upward toward the cerebrum, where most of the arachnoid villi are located.
Hydrocephalus
Obstruction to free circulation of CSF in the neonate results in hydrocephalus. For example, blockage of the aqueduct of Sylvius results in expansion of the lateral and third cerebral ventricles and compression of the brain (see Fig. 3-23). This type of obstruction producing a noncommunicating type of hydrocephalus is treated by surgical creation of an artificial pathway for flow of CSF between the cerebral ventricular system and the subarachnoid space.
Intracranial Pressure
Normal intracranial pressure (ICP) is <15 mm Hg. This pressure is regulated by the rate of CSF formation and resistance to CSF reabsorption through arachnoid villi as determined by venous pressure. In addition, increases in cerebral blood flow, as during inhalation of volatile anesthetics, can cause the ICP to increase because of the concomitant increase in cerebral blood flow and cerebral blood volume. Systemic blood pressure does not alter ICP within the range of normal autoregulation. Phasic variations in systemic blood pressure, however, are transmitted as variations in ICP.
Papilledema
Anatomically, the dura of the brain extends as a sheath around the optic nerve and then connects with the sclera of the eye. Increases in ICP are transmitted to the optic nerve sheath. Increased pressure in the optic sheath impedes blood flow in the retinal veins, leading to increases in the retinal capillary pressure and retinal edema. The tissues of the optic disc are more distensible than the rest of the retina, so the disc becomes edematous and swells into the cavity of the eye. This swelling of the optic disc is termed papilledema.
Blood–Brain Barrier
The blood–brain barrier reflects the impermeability of capillaries in the CNS, including the choroid plexuses, to circulating substances such as electrolytes and exogenous drugs or toxins. As a result, the neural and glial cells in the CNS live in a tightly controlled milieu that varies little in the healthy individual. The blood–brain barrier is maintained by the tight junction between endothelial cells of brain capillaries. Envelopment of brain capillaries by glial cells further decreases their permeability. The blood–brain barrier is less developed in the neonate and tends to break down in areas of the brain that are irradiated, infected, or compromised by neoplasm. The blood–brain barrier is also relatively permeable in the area around the posterior pituitary and the chemoreceptor trigger zone. The blood–brain barrier is characterized by active transport mediated by p-glycoprotein transporters (p-GP). These proteins are of the ATP binding cassette (ABC) family. Active transport of morphine out of the CNS by a p-GP is responsible for the >90-minute delay between morphine bolus and peak morphine drug effect.
Vision
The eye is optically equivalent to a photographic camera in that it contains a lens system, a variable aperture system (pupil), and light-sensitive surface (retina) (Fig. 3-24).68 The lens system of the eye focuses an image on the retina. Relaxation and contraction of the ciliary muscles are responsible for altering the tension of ligaments attached to the lens, causing its refractive power to change. One diopter is equivalent to the ability of a lens to converge parallel light rays to a focal point 1 meter beyond the lens (59 diopters equals the total refractive power of the eye). Stimulation of parasympathetic nervous system fibers to the ciliary muscle causes this muscle to relax, which in turn relaxes the ligaments of the lens and increases its refractive power. This increased refractive power allows the eye to focus on objects that are nearby. Interference with this process of accommodation may be noted by patients in the postoperative period who have received an anticholinergic drug in the preoperative medication or as part of the pharmacologic reversal of nondepolarizing neuromuscular blockade. The principal function of the pupil is to increase or decrease the amount of light that enters the eye. For example, the pupil may vary from 1.5 to 8.0 mm in diameter, permitting a 30-fold variation in the amount of light that enters the eye.
The lens loses its elastic nature with aging because of progressive denaturation of the len’s proteins. As a result, the ability to accommodate is almost totally absent by 45 to 50 years of age. This lack of ability to accommodate is known as presbyopia.
Progressive denaturation of the proteins in the lens leads to the formation of a cataract. In later stages, calcium is often deposited in the coagulated proteins, thus further increasing the opacity. If the cataract impairs vision, the lens can be replaced by an artificial convex lens that compensates for the loss of refractive power created by removal of the lens.
Intraocular Fluid
Intraocular fluid consists of aqueous humor, which lies in front and at the sides of the lens, and vitreous humor, which lies between the lens and retina. Aqueous humor is freely flowing fluid that is continuously formed (2 to 3 mL per minute) and reabsorbed. This fluid is secreted by ciliary processes of the ciliary body in a manner similar to formation of CSF by the choroid plexus. After flowing into the anterior chamber, aqueous humor enters the canal of Schlemm, a thin vein that extends circumferentially around the eye. Vitreous humor is a gelatinous mass into which substances can diffuse slowly, but there is little flow of fluid.
Intraocular Pressure
Intraocular pressure is normally 15 to 25 mm Hg. This pressure is measured clinically by tonometry, in which the amount of displacement of the tonometer is calibrated in terms of intraocular pressure. It is believed that intraocular pressure is regulated primarily by resistance to outflow of aqueous humor from the anterior chamber into the canal of Schlemm. Glaucoma is associated with increased intraocular pressure sufficient to compress retinal artery inflow to the eye, leading to ischemic pain and eventually blindness. When medical control of glaucoma fails, it may be necessary to surgically create an artificial outflow tract for aqueous humor.
Retina
The retina is the light-sensitive portion of the eye containing the cones, which are responsible for color vision, and the rods, which are mainly responsible for vision in the dark. When the cones and rods are stimulated, impulses are transmitted through successive neurons in the retina and optic nerve before reaching the cerebral cortex. The presence of melanin in the pigment layer of the retina prevents reflection of light throughout the globe. Without this pigment, light rays would be reflected in all directions within the globe, causing visual acuity to be impaired. Indeed, albinos, who lack melanin, have greatly decreased visual acuity.
The nutrient blood supply for the retina is largely derived from the central retinal artery, which accompanies the optic nerve. This independent retinal blood supply prevents rapid degeneration of the retina should it become detached from the pigment epithelium and allows time for surgical correction of a detached retina. The main arterial supply to the globe and orbital contents is from the ophthalmic artery, which is a branch of the internal carotid artery.69
Ischemic Optic Neuropathy
Ischemic optic neuropathy (ION) results from infarction of the optic nerve and is the most frequently reported cause of vision loss following general anesthesia.70 ION is classified as anterior ION(nonarteritic or arteritic) and posterior ION. Nonarteritic anterior ION occurs more often in patients with congenitally small optic discs. It is presumed that the small cross-sectional area of the optic disc results in little room for expansion of optic nerve fibers in response to ischemia-induced edema.
Posterior ION has been reported after diverse surgical procedures (prolonged spinal fusion surgery, cardiac operations requiring cardiopulmonary bypass, radical neck surgery) and its etiology appears to be multifactorial—including intraoperative anemia and hypotension combined with at least one other factor (e.g., congenital absence of the central retinal artery, increased venous pressure owing to venous obstruction, large amounts of fluid administration, prolonged head-down position, administration of vasopressors).70–73 Prone positioning increases IOP during anesthesia and could contribute to decreases in ocular perfusion pressure (Fig. 3-25).74 Despite the multifactorial etiology of ION, some cases do not have any of the speculated associated factors (anemia, hypotension), except perhaps for a large amount of intravenous fluids.75
Other Causes of Postoperative Blindness
Cortical blindness, retinal occlusion, and ophthalmic venous obstruction need to be excluded when postoperative blindness occurs and ION is a consideration. Cortical blindness is characterized by loss of visual sensation with retention of pupillary reaction to light and normal funduscopic examination results. CT or MRI abnormalities in the parietal or occipital lobe confirm the diagnosis. A rare cause of cortical blindness is cyclosporine-induced neurotoxicity that is usually reversible.71 Central retinal artery occlusion presents as painless, monocular blindness. Ophthalmoscopic examination of the eyes with retinal artery occlusion shows a pale edematous retina, a cherry-red spot at the fovea, and platelet-fibrin or cholesterol emboli in the narrowed retinal arteries. Obstruction of venous drainage from the eye may occur intraoperatively when patient positioning results in external pressure on the eyes.
Photochemicals
The light-sensitive photochemical continuously synthesized in rods is rhodopsin. Cones contain photochemicals that resemble rhodopsin. Vitamin A is an important precursor of photochemicals, which explains the occurrence of night blindness when this vitamin becomes deficient. Photochemicals in rods and cones decompose on exposure to light and in the process stimulate fibers in the optic nerve. Decomposition of rhodopsin decreases conductance of the membranes of rods for sodium ions. The resulting hyperpolarization in rods is opposite to the effect that occurs in almost all other sensory receptors. The intensity of the hyperpolarization signal is proportional to the logarithm of light energy, in contrast to the more linear response of most other receptors. This logarithmic response is important to vision because it allows the eyes to detect contrasts on the image even when light intensities vary several thousand fold.
If a person is in bright light for a prolonged period, large proportions of photochemicals in the rods and cones are depleted, resulting in decreased sensitivity of the eye to light (light adaptation). Conversely, during total darkness, the sensitivity of the retina is increased, reflecting conversion of photochemicals to rhodopsin (dark adaptation). The eye can also adapt to changes in light intensity by changing the size of the pupillary opening up to 30-fold.
Visual Pathway
Impulses from the retina pass backward through the optic nerve (Fig. 3-26).68 The macula is a small area in the center of the retina that is composed mainly of cones to permit detailed vision. The fovea is the central portion of the macula and is the site of the clearest vision. At the optic chiasm, all the fibers from the nasal halves of the retina cross to the opposite side to join fibers from the opposite temporal retina to form the optic tracts. Fibers of the optic tract synapse in the lateral geniculate body before passing into the visual (occipital) area of the cerebral cortex. Specific points of the retina connect with specific points of the visual cortex, which results in the detection of lines, borders, and colors.
Field of Vision
The field of vision is the area seen by the eye at a given instant. The area seen to the nasal side is called the nasal field of vision, and the area seen to the lateral side is called the temporal field of vision (see Fig. 3-26).68 An important use of visual fields is localization of lesions in the visual neural pathway. For example, anterior pituitary tumors may compress the optic chiasm, causing blindness in both temporal fields of vision (called bitemporal hemianopia). Thrombosis of the posterior cerebral artery is a cause of infarction of the visual cortex.
Muscular Control of Eye Movements
The cerebral control system for directing the eyes toward the object to be viewed is as important as the cerebral system for interpretation of the visual signals. Movements of the eyes are controlled by three pairs of skeletal muscles designated as the (a) medial and lateral recti, (b) superior and inferior recti, and (c) superior and inferior obliques. The medial and lateral recti contract reciprocally to move the eyes from side to side; the superior and inferior recti move the eyes upward or downward; and rotation of the globe is accomplished by the superior and inferior obliques. Each of the three sets of eye muscles is reciprocally innervated by cranial nerves III, IV, and VI so that one muscle of the pair contracts while the other relaxes.
Simultaneous movement of both eyes in the same directions is called conjugate movement of the eyes. Occasionally, abnormalities occur in the control system for eye movements that cause continuous nystagmus. Nystagmus is likely to occur when one of the vestibular apparatuses is damaged or when deep nuclei in the cerebellum are damaged or under the influence of ketamine anesthesia.
Innervation of the Eye
The eyes are innervated by the sympathetic and parasympathetic nervous system. The preganglionic fibers of the parasympathetic nervous system arise in the Edinger-Westphal nucleus of cranial nerve III and then pass to the ciliary ganglion, which gives rise to nerve fibers that innervate the ciliary muscle and sphincter of the iris. Sympathetic nervous system fibers innervate the radial fibers of the iris as well as several extraocular structures. Stimulation of the parasympathetic nervous system fibers to the eye excites the ciliary sphincter, causing miosis. Conversely, stimulation of sympathetic nervous system fibers to the eye excites the radial fibers of the iris and causes mydriasis. Volatile anesthetics cause midrange pupillary dilation, whereas opioids cause papillary constriction. Monitoring of papillary diameter provides some indication of the residual opioid activity on anesthetic emergence.
Horner Syndrome
Interruption of the superior cervical chain of the sympathetic nervous system innervation to the eye results in miosis, ptosis, and vasodilation with absence of sweating on the ipsilateral side of the body, commonly referred to as Horner’s syndrome. Miosis occurs because of interruption of sympathetic nervous system innervation to the radial fibers of the iris. Ptosis reflects the normal innervation of the superior palpebral muscle by the sympathetic nervous system. Horner’s syndrome often occurs following stellate ganglion block and is occasionally a complication of interscalene block of the brachial plexus.
Hearing
Receptors for hearing and equilibrium are housed in the inner ear (Fig. 3-27).68 The external ear focuses sound waves on the ear drum, which oscillates in contact with the bones of the middle ear. The sound is amplified at the oval window, where the vibrations are transmitted to the hair cells of the cochlea in the inner ear. The anatomic arrangement of the hair cells results in their responding to different frequencies, performing a mechanical Fourier transformation of the incoming sound waves. The electrical current generated from activation of a hair cell travels from the auditory nerve to the inferior colliculus and auditory cortex.
The Eustachian tube connects the middle ear with the posterior tonsillar pillars and allows pressures on both sides of the tympanic membrane to be equalized during chewing or swallowing. Nitrous oxide may increase middle ear pressure and has been associated with rupture of the tympanic membrane when inflammation or scarring of the Eustachian tube opening into the nasopharynx prevents spontaneous decompression of the middle ear.76
Deafness
Nerve deafness is due to an abnormality of the cochlear or auditory nerve. Certain drugs such as streptomycin, gentamicin, kanamycin, and chloramphenicol may damage the organ of Corti, causing nerve deafness. Conduction deafness is caused by injury to the mechanisms that conduct sound waves from the tympanic membrane to the oval window. Conduction deafness is often caused by fibrosis of the structures in the middle ear after repeated infections in the middle ear by the hereditary disease known as osteosclerosis.
Perioperative Hearing Impairment
Perioperative hearing impairment is often subclinical and may go unnoticed unless audiometry is performed.77 Hearing loss (incidence may be as high as 50%) after dural puncture in the low-frequency range is most likely due to CSF leak and should resolve completely within days. Hearing loss following general anesthesia for surgery not requiring cardiopulmonary bypass does not appear to have a uniform prognosis, likely reflecting the myriad of etiologies (e.g., CSF leak after ear, nose, and throat [ENT] and neurosurgery, barotrauma from nitrous oxide, embolism during cardiac surgery, or preexisting vasculopathy). Recovery in hearing appears to be independent of treatment. Unilateral hearing loss following cardiopulmonary bypass is often permanent and probably due to embolism and subsequent ischemic injury to areas of the organ of Corti.
Equilibrium
The semicircular canals (the utricle and saccule of the inner ear) are important for maintaining equilibrium (see Fig. 3-27).68 The utricle and saccule contain cilia that transmit nerve impulses to the brain necessary for maintaining orientation of the head in space. Endolymph present in the semicircular canals flows with changes in head position, causing signals to be transmitted via the vestibular nerve nuclei and the cerebellum.
Taste
Taste is mainly a function of taste buds located principally in the papillae of the tongue. Sweet, sour, salty, and bitter are the four primary sensations of taste. Sour taste is caused by acids. Sour taste intensity is approximately proportional to the logarithm of the hydrogen ion concentration (i.e., pH). Sweet and salt are pleasurable tastes, of course. Bitter tastes are generally unpleasant. The bitter taste of alkaloids causes the individual to reject these substances. This may be protective as many plant toxins are alkaloids.
Adaptation to taste sensations is almost complete in 1 to 5 minutes of continuous stimulation. Individuals with upper respiratory tract infections complain of loss of taste sensation when, in fact, taste bud function is normal, emphasizing that most of what is considered taste is actually smell. Taste preference is presumed to be a CNS phenomenon but may be influenced by common polymorphisms in the many genes for taste receptors.
Smell
Olfactory receptors are located high in the nasal cavity. Each olfactory receptor is located on a single cilium. Olfactory receptors are coupled to G proteins. G protein activation increases activity of adenylyl cyclase, increasing the concentration of cAMP. A substance must be volatile and lipid soluble to stimulate olfactory cells. The importance of upward air movement in smell acuity is the reason sniffing improves the sense of smell, whereas holding one’s breath prevents the sensation of unpleasant odors. Olfactory receptors adapt extremely rapidly, such that smell sensation may become extinct in about 60 seconds. Compared with lower animals, the sense of smell in humans is almost rudimentary. Humans have over 1,000 genes for odorant receptors but only about 40% of those are functional. Nevertheless, the threshold for smell is low as reflected by the detection of trace concentrations of methyl mercaptan that is mixed with odorless natural gas to alert one to a gas leak.
Nausea and Vomiting
Nausea is the conscious recognition of excitation of an area in the medulla that is associated with the vomiting (emetic) center (Fig. 3-28).78 Impulses are transmitted by afferent fibers of the parasympathetic and sympathetic nervous system to the vomiting center. Motor impulses transmitted via cranial nerves V, VII, IX, X, and XII to the gastrointestinal tract and through the spinal nerves to the diaphragm and abdominal muscles are required to cause the mechanical act of vomiting.
The medullary vomiting center is located close to the fourth cerebral ventricle and receives afferents from the (a) chemoreceptor trigger zone, (b) cerebral cortex, (c) labyrinthovestibular center, and (d) neurovegetative system. Impulses from these afferents lead to nausea and vomiting. The chemoreceptor trigger zone includes receptors for serotonin, dopamine, histamine, and opioids. Stimulation of the chemoreceptor trigger zone located on the floor of the fourth cerebral ventricle initiates vomiting independent of the vomiting center. The chemoreceptor trigger zone is not protected by the blood–brain barrier and thus this zone can be activated by chemical stimuli received through the systemic circulation as well as the CSF. The cerebral cortex stimulates vomiting through as a response to certain smells and physiologic stresses. Motion can stimulate equilibrium receptors in the inner ear, which may also stimulate the medullary vomiting center. The neurovegetative system is sensitive principally to gastrointestinal stimulation. Blocking of impulses from the chemoreceptor trigger zone does not prevent vomiting due to irritative stimuli (ipecac) arising in the gastrointestinal tract.
Peripheral Nervous System
The peripheral nervous system is composed of the sensory and motor nerves that connect the CNS to the tissues and organs (Fig. 3-29). These nerves are familiar to anesthesiologists as the targets for regional anesthetic techniques, and the anatomy is well reviewed in many atlases of regional anesthesia.
Pathways for Peripheral Sensory Impulses
The peripheral nerves extend from the dendrite in the periphery to the dorsal root ganglion, where the cell body is located, and from there to the spinal cord by way of the dorsal root (Fig. 3-30). By definition, dendrites conduct impulses toward the cell body, whereas axons conduct impulses away from the cell body. Thus, the portion of the nerve from the cell body to the peripheral receptor is a dendrite, whereas the relatively shorter connection from the dorsal root ganglion to the spinal cord is the axon. However, structurally, the dendrite and the axon are indistinguishable, and the nerve behaves like one long axon, giving rise to the term pseudounipolar neuron that occasionally is used to describe peripheral nerves.
After entering the spinal cord, peripheral sensory neurons synapse in the dorsal horn and give rise to long, ascending fiber tracts that transmit sensory information to the brain. These sensory signals are transmitted to the brain by the dorsal-lemniscal system, which includes dorsal column pathways and spinocervical tracts, and by anterolateral spinothalamic tracts (Figs. 3-31 and 3-32).3 Impulses in the dorsal column pathways cross in the spinal cord to the opposite side before passing upward to the thalamus. Synapses in the thalamus are received by neurons that project into the somatic sensory area of the cerebral cortex. Nerve fibers of the anterolateral spinothalamic system cross in the anterior commissure to the opposite side of the spinal cord, where they turn upward toward the brain as the ventral and lateral spinothalamic tracts. Sensory signals from the anterolateral spinothalamic system are relayed from the thalamus to the somatic sensory area of the cerebral cortex. All sensory information that enters the cerebral cortex, with the exception of the olfactory system, passes through the thalamus.
Pathways for Peripheral Motor Responses
Sensory information is integrated at all levels of the nervous system and causes appropriate motor responses, beginning in the spinal cord with relatively simple reflex responses. Motor responses originating in the brainstem are more complex, whereas the most complicated and precise motor responses originate from the cerebral cortex.
Anterior motor neurons in the anterior horns of the spinal cord gray matter give rise to A-α fibers that leave the spinal cord by way of anterior nerve roots and innervate skeletal muscles. Skeletal muscles and tendons contain muscle spindles and Golgi tendon organs that operate at a subconscious level to relay information to the spinal cord and brain relative to changes in length and tension of skeletal muscle fibers. The stretch reflex is a reflex contraction of the skeletal muscle whenever stretch of the opposite balanced muscle results in stimulation of the muscle spindle. Tapping the patellar tendon elicits a knee jerk, which is a stretch reflex of the quadriceps femoris muscle. The ankle jerk is due to reflex contraction of the gastrocnemius muscle. Transmission of large numbers of facilitatory impulses from upper regions of the CNS to the spinal cord results in exaggerated stretch reflex responses. For example, lesions in the contralateral motor areas of the cerebral cortex, as caused by a cerebral vascular accident or brain tumor, cause greatly enhanced stretch reflexes. Clonus occurs when evoked muscle jerks oscillate. This phenomenon typically occurs when the stretch reflex is sensitized by facilitatory impulses from the brain, resulting in exaggerated facilitation of the spinal cord. When associated with recovery from general anesthesia, clonus as initiated by abrupt dorsiflexion of the foot can be eliminated by flexing the knees and keeping them in a flexed position.79
Transection of the brainstem at the level of the pons (isolates the spinal cord from the rest of the brain) results in spasticity known as decerebrate rigidity. Decerebrate rigidity reflects diffuse facilitation of stretch reflexes.
The motor system is often divided into upper and lower motor neurons. Lower motor neurons originate in the spinal cord and directly innervate skeletal muscles. A lower motor neuron lesion is associated with flaccid paralysis, atrophy of skeletal muscles, and absence of stretch reflex responses. Spastic paralysis with accentuated stretch reflexes is due to destruction of upper motor neurons in the brain. Upper motor neurons originate in the cerebral cortex or brainstem and traverse down the anterior and lateral corticospinal paths until they connect with the lower motor neuron in the ventral horn of the spinal cord.
Withdrawal flexor reflexes are a lower motor neuron reflex, typically elicited by a painful stimulus. Associated with withdrawal of the stimulated limb is extension of the opposite limb (cross-extensor reflex) that occurs 0.2 to 0.5 second later and serves to push the body away from the object causing the painful stimulus. The delayed onset of the cross-extensor reflex is due to the time necessary for the signal to pass through the additional neurons to reach the opposite side of the spinal cord.
Autonomic Nervous System
The autonomic nervous system controls the visceral functions of the body. In addition, the autonomic nervous system modulates systemic blood pressure, gastrointestinal motility and secretion, urinary bladder emptying, sweating, and body temperature maintenance. Activation of the autonomic nervous system occurs principally via centers located in the hypothalamus, brainstem, and spinal cord. The ANS is divided into the sympathetic, parasympathetic, and enteric nervous systems.
The sympathetic and the parasympathetic nervous systems usually function as physiologic antagonists such that the compiled action on any organ represents a balance of the influence of each component (Table 3-4). The sympathetic nervous system functions as an amplification response, whereas the parasympathetic nervous system evokes discrete and narrowly targeted responses.
The enteric nervous system is arranged nontopographically and its neurons and cells are located in the walls of the gastrointestinal tract. Although the gastrointestinal tract is influenced by sympathetic and parasympathetic nervous system activity, it is the enteric nervous system through the myenteric and submucous plexi that regulates digestive activity even in the presence of spinal cord transection.
An understanding of the anatomy and physiology of the autonomic nervous system is required for predicting the pharmacologic effects of drugs that act on either the sympathetic or parasympathetic nervous systems (Table 3-5).
Anatomy of the Sympathetic Nervous System
Nerves of the sympathetic nervous system arise from the thoracolumbar (T1 to L2) segments of the spinal cord (Fig. 3-33).3 These nerve fibers pass to the paravertebral sympathetic chains located lateral to the spinal cord. From the paravertebral chain, nerve fibers pass to tissues and organs innervated by the sympathetic nervous system.
Each nerve of the sympathetic nervous system consists of a preganglionic neuron and a postganglionic neuron (Fig. 3-34). Cell bodies of preganglionic neurons are located in the intermediolateral horn of the spinal cord. Fibers from these preganglionic cell bodies leave the spinal cord with anterior (ventral) nerve roots and pass via white rami into 1 of 22 pairs of ganglia composing the paravertebral sympathetic chain. Axons of preganglionic neurons are mostly myelinated, slow-conducting type B fibers (see Table 3-1). In the ganglia of the paravertebral sympathetic chain, the preganglionic fibers can synapse with cell bodies of postganglionic neurons or pass cephalad or caudad to synapse with postganglionic neurons (mostly unmyelinated type C fibers) in other paravertebral ganglia. Postganglionic neurons then exit from paravertebral ganglia to travel to various peripheral organs. Other postganglionic neurons return to spinal nerves by way of gray rami and subsequently travel with these nerves to influence vascular smooth muscle tone and the activity of piloerector muscles and sweat glands.
Fibers of the sympathetic nervous system are not necessarily distributed to the same part of the body as the spinal nerve fibers from the same segments. For example, fibers from T1 usually ascend in the paravertebral sympathetic chain into the head, T2 into the neck, T3–T6 into the chest, T7–T11 into the abdomen, and T12 and L1–L2 into the legs. The distribution of these sympathetic nervous system fibers to each organ is determined in part by the position in the embryo from which the organ originates. In this regard, the heart receives many sympathetic nervous system fibers from the neck portion of the paravertebral sympathetic chain because the heart originates in the neck of the embryo. Abdominal organs receive their sympathetic nervous system innervation from the lower thoracic segments, reflecting the origin of the gastrointestinal tract from this area.
Anatomy of the Parasympathetic Nervous System
Nerves of the parasympathetic nervous system leave the CNS through cranial nerves III, V, VII, IX, and X (vagus) and from the sacral portions of the spinal cord (Fig. 3-35).3 About 75% of all parasympathetic nervous system fibers are in the vagus nerves passing to the thoracic and abdominal regions of the body. As such, the vagus nerves supply parasympathetic innervation to the heart, lungs, esophagus, stomach, small intestine, liver, gallbladder, pancreas, and upper portions of the uterus. Fibers of the parasympathetic nervous system in cranial nerve III pass to the eye. The lacrimal, nasal, and submaxillary glands receive parasympathetic nervous system fibers via cranial nerve VII, whereas the parotid gland receives parasympathetic nervous system innervation via cranial nerve IX.
The sacral part of the parasympathetic nervous system consists of the second and third sacral nerves, and, occasionally, the first and fourth sacral nerves. Sacral nerves form the sacral plexus on each side of the spinal cord. These nerves distribute fibers to the distal colon, rectum, bladder, and lower portions of the uterus. In addition, parasympathetic nervous system fibers to the external genitalia transmit impulses that elicit various sexual responses.
In contrast to the sympathetic nervous system, preganglionic fibers of the parasympathetic nervous system pass uninterrupted to ganglia near or in the innervated organ (see Fig. 3-35).3 Postganglionic neurons of the parasympathetic nervous system are short because of the location of the corresponding ganglia. This situation contrasts with the sympathetic nervous system, in which postganglionic neurons are relatively long, reflecting their origin in the ganglia of the paravertebral sympathetic chain, which is often distant from the innervated organ. Furthermore, unlike the amplified and diffuse discharges characteristic of sympathetic nervous system responses, activation of the parasympathetic nervous system is tonic and discrete. The vasodilatory effects of acetylcholine depend on the integrity of the vascular endothelium because activation of muscarinic receptors on the endothelium results in the release of nitric oxide.80
Physiology of the Autonomic Nervous System
Postganglionic fibers of the sympathetic nervous system secrete norepinephrine as the neurotransmitter (Fig. 3-36). These norepinephrine-secreting neurons are classified as adrenergic fibers. Postganglionic fibers of the parasympathetic nervous system secrete acetylcholine as the neurotransmitter (see Fig. 3-36). These acetylcholine-secreting neurons are classified as cholinergic fibers. In addition, innervation of sweat glands and some blood vessels is by postganglionic sympathetic nervous system fibers that release acetylcholine as the neurotransmitter. All preganglionic neurons of the sympathetic and parasympathetic nervous system release acetylcholine as the neurotransmitter and are thus classified as cholinergic fibers. For this reason, acetylcholine release at preganglionic fibers activates both sympathetic and parasympathetic postganglionic neurons.
Norepinephrine as a Neurotransmitter
Synthesis
Synthesis of norepinephrine involves a series of enzyme-controlled steps that begin in the cytoplasm of postganglionic sympathetic nerve endings (varicosities) and are completed in the synaptic vesicles (Fig. 3-37). For example, the initial enzyme-mediated steps leading to the formation of dopamine take place in the cytoplasm. Dopamine then enters the synaptic vesicle, where it is converted to norepinephrine by dopamine β-hydroxylase. It is likely that the enzymes that participate in the synthesis of norepinephrine are produced in postganglionic sympathetic nerve endings. These enzymes are not highly specific, and other endogenous substances, as well as certain drugs, may be acted on by the same enzyme. For example, dopa-decarboxylase can convert the antihypertensive drug α-methyldopa to α-methyldopamine, which is subsequently converted by dopamine β-hydroxylase to the weakly active (false) neurotransmitter α-methylnorepinephrine that decreases the activation of central α1-adrenergic synapses and results in the reduction of blood pressure.
Storage and Release
Norepinephrine is stored in synaptic vesicles for subsequent release in response to an action potential.81 Adrenergic fibers can sustain output of norepinephrine during prolonged periods of stimulation. Tachyphylaxis in response to repeated administration of ephedrine and other indirect-acting sympathomimetics may reflect depletion of the norepinephrine stored in sympathetic nerve endings.
Termination of Action
Termination of the action of norepinephrine is by (a) uptake (reuptake) back into postganglionic sympathetic nerve endings, (b) dilution by diffusion from receptors, and (c) metabolism by the enzymes monoamine oxidase (MAO) and catechol-O-methyltransferase (COMT). Norepinephrine released in response to an action potential exerts its effects at receptors for only a brief period, reflecting the efficiency of these termination mechanisms.
Reuptake
Uptake of previously released norepinephrine back into postganglionic sympathetic nerve endings is probably the most important mechanism for terminating the action of this neurotransmitter on receptors. As much as 80% of released norepinephrine undergoes reuptake. Reuptake provides a source for reuse of norepinephrine in addition to synthesis.
It is likely that two active transport systems are involved in reuptake of norepinephrine, with one system responsible for uptake into the cytoplasm of the varicosity and a second system for passage of norepinephrine into the synaptic vesicle for storage and reuse. The active transport system for norepinephrine uptake can concentrate the neurotransmitter 10,000-fold in postganglionic sympathetic nerve endings. Magnesium and adenosine triphosphate are essential for function of the transport system necessary for the transfer of norepinephrine from the cytoplasm into the synaptic vesicle. The transport system for uptake of norepinephrine into cytoplasm is blocked by numerous drugs, including cocaine and tricyclic antidepressants.
Metabolism
Metabolism of norepinephrine is of relatively minor significance in terminating the actions of endogenously released norepinephrine. The exception may be at some blood vessels, where enzymatic breakdown and diffusion account for the termination of action of norepinephrine. Norepinephrine that undergoes uptake is vulnerable to metabolism in the cytoplasm of the varicosity by MAO. Any neurotransmitter that escapes reuptake is vulnerable to metabolism by COMT, principally in the liver. Inhibitors of MAO cause an increase in tissue levels of norepinephrine and may be accompanied by a variety of pharmacologic effects. Conversely, no striking pharmacologic change accompanies inhibition of COMT.
The primary urinary metabolite resulting from metabolism of norepinephrine by MAO or COMT is 3-methoxy-4-hydroxymandelic acid. This metabolite is also referred to as vanillylmandelic acid (VMA). Normally, the 24-hour urinary excretion of 3-methoxy-4-hydroxymandelic acid is 2 to 4 mg, representing primarily norepinephrine that is deaminated by MAO in the cytoplasm of the varicosity of the postganglionic sympathetic nerve endings. Elevated levels of urinary VMA suggest pheochromocytoma.
Acetylcholine as a Neurotransmitter
Synthesis
Acetylcholine is synthesized in the cytoplasm of varicosities of the preganglionic and postganglionic parasympathetic nerve endings. The enzyme choline acetyltransferase is responsible for catalyzing the combination of choline with acetyl coenzyme A to form acetylcholine. Choline enters parasympathetic nerve endings from the extracellular fluid through an active transport system. Acetyl coenzyme A is synthesized in mitochondria present in high concentrations in parasympathetic nerve endings.
Storage and Release
Acetylcholine is stored in synaptic vesicles for release in response to an action potential. Arrival of an action potential at a parasympathetic nerve ending results in the release of 100 or more vesicles of acetylcholine. It is estimated that a single nerve ending contains >300,000 presynaptic vesicles of acetylcholine.
Metabolism
Acetylcholine has a brief effect at receptors (<1 millisecond) because of its rapid hydrolysis by acetylcholinesterase to choline and acetate. Choline is transported back into parasympathetic nerve endings, where it is used for synthesis of new acetylcholine. Plasma cholinesterase is an enzyme found in low concentrations around acetylcholine receptors, being present in the highest amounts in plasma. The physiologic significance of plasma cholinesterase is unknown, as it is too slow to be physiologically important in the metabolism of acetylcholine. Absence of plasma cholinesterase produces no detectable clinical signs or symptoms until a drug such as succinylcholine or mivacurium is administered.
Interactions of Neurotransmitters with Receptors
Norepinephrine and acetylcholine, acting as neurotransmitters, interact with receptors (protein macromolecules) in lipid cell membranes (Table 3-6). This receptor-neurotransmitter interaction most often activates or inhibits effector enzymes, such as adenylate cyclase, or alters flux of sodium and potassium ions across cell membranes via protein ion channels. The net effect of these changes is transduction of external stimuli into intracellular signals.
Norepinephrine Receptors
The pharmacologic effects of catecholamines led to the original concept of α- and β-adrenergic receptors.82 Subdivision of these receptors into α1, α2, β1 (cardiac), and β2 (noncardiac) allows an understanding of drugs that act as either agonists or antagonists at these sites (see Table 3-5). Genetic cloning has borne out the original pharmacologic distinctions. However, there are splice variants of each gene that create receptors with different pharmacologic properties. The α2 receptors are also present on platelets, where they mediate platelet aggregation. In the CNS, stimulation of postsynaptic α2 receptors by drugs such as clonidine or dexmedetomidine results in enhanced potassium ion conductance and membrane hyperpolarization manifesting as decreased anesthetic requirements and analgesia.
Dopamine receptors were originally pharmacologically subdivided as dopamine1 and dopamine2. However, molecular cloning has allowed for the identification of five dopamine receptor genes. However, it is still possible to classify the dopamine receptors into D1 such as DRD1 and DRD5 and D2 such as DRD2, DRD3, and DRD4. Dopamine receptors play important roles on smooth muscle and in the kidney as well as in the CNS where they are targets of many neuropsychiatric drugs and the unwitting target of many drugs of abuse. Activation of dopamine1 receptors is responsible for vasodilation of the splanchnic and renal circulations. D4 receptors are present in the human heart where there stimulation with dopamine results in an increase in contractility and intrinsic heart rate. α2 adrenergic and dopamine2 receptors function as a negative feedback loop such that their activation inhibits subsequent release of neurotransmitter (Table 3-7).
Signal Transduction
Adrenergic and dopaminergic receptors are G protein–coupled receptors. The bound receptor activates the G protein, typically resulting in activation of protein kinases and phosphorylation of target proteins. Catecholamines activate β1-adrenergic receptors resulting in dramatic increases in intracellular cAMP through activation of Gs. Increased intracellular cAMP initiates a series of intracellular events, including cascading protein phosphorylation reactions and stimulation of the sodium-potassium pump, which results in the metabolic and pharmacologic effects typical of epinephrine and other catecholamines. In contrast to β receptors, α1-adrenergic receptors are linked to Gq receptors which when activated increase phospholipase 3, increasing inositol trisphosphate (IP3) and liberating the release of intracellular calcium stores. The α2-adrenergic and dopamine 2 receptors are linked to the Gi protein, activation of which decreases adenylate cyclase.
Adrenergic Receptor Concentrations
Concentrations of β-adrenergic receptors in the postsynaptic membrane adjust dynamically to the concentration of norepinephrine in the synaptic cleft and plasma. Desensitization reflects the rapid waning of responses to hormones and neurotransmitters despite continuous exposure to adrenergic agonists.83 Downregulation is different from the rapid appearance of desensitization occurring only hours after exposure to agonists. During downregulation, receptors are destroyed and new receptors must be synthesized before a return to baseline is possible. Similarly, in the presence of long-term blockade, β1 receptor numbers increase. Drug-induced alteration in adrenergic receptor number is consistent with rebound tachycardia and myocardial ischemia that may accompany sudden discontinuation of chronic β-adrenergic receptor blockers.
Chronic congestive heart failure (CHF) results in depletion of catecholamines in the myocardium and compensatory increases in plasma concentrations of norepinephrine to maintain systemic vascular resistance and perfusion pressure. Accompanying decreases in the concentrations of β1 receptors in the heart are likely responsible for the failure of β agonists to effectively treat CHF.84 Long-term treatment with pharmacologic doses of β-adrenergic agonists is also associated with myocardial toxicity, whereas paradoxically treating chronic CHF with judicious doses of β blockers is efficacious by upregulating β1-adrenergic receptors.
Acetylcholine Receptors
Cholinergic receptors are classified as nicotinic and muscarinic. The links between stimulus and response are different in nicotinic and muscarinic receptors (see Table 3-6). Nicotinic receptors are ligand-gated receptors, whereas muscarinic receptors are G protein linked.
Nicotinic Receptors
Acetylcholine can affect nicotinic receptors at either the neuromuscular junction, at autonomic ganglia and in the CNS. Nicotinic receptors belong to the superfamily of ligand-gated ion channels that includes GABAA, 5-HT3, and glycine receptors. Muscle-type nicotinic receptors are membrane proteins (two α subunits, β, ε, and d) that form nonselective ion channels.85 In human muscle, the γ subunit is replaced by the ε subunit within the first 2 weeks of life. This change in structure converts the receptor from one with low conductance and long duration of opening to a receptor with high conductance and brief duration of opening. In the setting of immobilization and burns, the fetal-type receptor is upregulated and expressed outside the neuromuscular junction, resulting in excessive potassium release in response to succinylcholine.
Nicotinic acetylcholine receptors in nerves are composed of 2 to 5 α subunits with or without 3 β subunits. Ten α and 3 β subunits have been cloned. The nicotinic acetylcholine receptors that act as the preganglionic receptor in the sympathetic nervous system are primarily composed of α3 and β4 subunits. The nicotinic receptors in the brain are mostly presynaptic where they act as a gain control on the release of glutamate, GABA, dopamine, norepinephrine, and serotonin. They are highly expressed in and around the cholinergic nuclei that mediate arousal. The α4 β2 combination is also highly expressed in the reward centers leading to the high addictive potential of nicotine. Activation of α4 β2 and α7-type nicotinic receptors has analgesic effects in animals and humans and nicotinic ligands may serve as analgesic adjuvants.
Muscarinic Receptors
In contrast to ligand-gated nicotinic receptors, muscarinic receptors belong to the superfamily of G protein–coupled receptors and are more homologous to adrenergic receptors than to nicotinic receptors. Five muscarinic receptors have been identified. All muscarinic subtypes are expressed in the CNS but M4 and M5 seem to be restricted there. M1 receptors are important in autonomic ganglia and for salivary and stomach secretion. M2 is expressed in the heart where its activation slows heart rate and nodal activity and decreases atrial contractility. M3 receptors are involved in smooth muscle contraction and eye accommodation. Their activation induces emesis and their antagonism with scopolamine has antiemetic properties. Atropine is a broad-spectrum muscarinic agonist.
Signal Transduction
Muscarinic receptors exhibit different signal transduction mechanisms. Odd-numbered muscarinic receptors (M1, M3, and M5) link to Gq and work predominantly through hydrolysis of phosphoinositide and release of intracellular calcium, whereas even-numbered receptors (M2 and M4) work primarily through Gi proteins to regulate adenylate cyclase.86
Residual Autonomic Nervous System Tone
The sympathetic and parasympathetic nervous systems are continually active, and this basal rate of activity is referred to as sympathetic or parasympathetic tone. The value of this tone is that it permits alterations in sympathetic or parasympathetic nervous system activity to mediate a fine increase or decrease in responses at innervated organs. For example, sympathetic nervous system tone normally keeps blood vessels about 50% constricted. As a result, increased or decreased sympathetic nervous system activity produces corresponding changes in systemic vascular resistance. If sympathetic tone did not exist, the sympathetic nervous system could only cause vasoconstriction.
In addition to continual direct sympathetic nervous system stimulation, a portion of overall sympathetic tone reflects basal secretion of norepinephrine and epinephrine by the adrenal medulla. The normal resting rate of secretion of norepinephrine is about 0.05 µg/kg per minute and epinephrine is about 0.2 µg/kg per minute. These secretion rates are nearly sufficient to maintain systemic blood pressure in a normal range even if all direct sympathetic nervous system innervation to the cardiovascular system is removed.
Determination of Autonomic Nervous System Function
Autonomic dysfunction associated with aging and diabetes mellitus may increase operative risk and can be associated with increased morbidity and mortality.87 Diagnosis of autonomic neuropathy in patients with diabetes mellitus is facilitated by tests of cardiovascular function (Table 3-8). Tests involving variability in heart rate measure activity of the sympathetic and parasympathetic nervous systems and precede changes in the measures of blood pressure. In addition to clinical tests of autonomic function, sensitive techniques for measuring plasma catecholamines are available. Interpretation of these data is confounded by other influences. Plasma epinephrine concentrations (normally 100 to 400 pg/mL) reflect adrenal release but vary greatly with psychological and physical stress. Plasma norepinephrine concentrations (normally 100 to 400 pg/mL) reflect both sympathetic nervous system and adrenal activity. Unlike plasma epinephrine levels, plasma norepinephrine concentrations reflect spillover from neuroeffector junctions, which may represent 10% to 20% of total release and vary among various organ systems.
Aging and Autonomic Nervous System Dysfunction
Common clinical manifestations of autonomic nervous system dysfunction in elderly patients are orthostatic hypotension, postprandial hypotension, hypothermia, and heat stroke. These responses reflect limited ability of elderly patients to adapt to stresses with vasoconstriction and vasodilation as mediated by the autonomic nervous system. Decreased autonomic nervous system function in elderly patients is due to fewer prejunctional terminals as plasma epinephrine concentrations and the numbers of β-adrenergic receptors are unchanged with aging. Plasma norepinephrine concentrations increase with age, suggesting a primary physiologic deficit in reuptake mechanisms.88
Clinically, there is attenuation of physiologic responses to β-adrenergic stimulation in the elderly. Exogenous β-adrenergic agonists have less profound effects on heart rate.89 This decreased response to adrenergic stimulation seems to reflect decreased affinity (number of receptors unchanged) of β receptors for the neurotransmitter and decreases in coupling of stimulatory G proteins and adenylate cyclase units.
Diabetic Autonomic Neuropathy
Diabetic autonomic neuropathy is present in 20% to 40% of insulin-dependent diabetic patients. Common manifestations of diabetic autonomic neuropathy include impotence, diarrhea, postural hypotension, sweating abnormalities, and gastroparesis. When impotence or diarrhea is the sole manifestations of autonomic neuropathy, there is little impact on survival. Conversely, 5-year mortality rates may exceed 50% when postural hypotension or gastroparesis is present. Anesthetic risk is increased in diabetic patients with autonomic neuropathy associated with gastroparesis (aspiration hazard), postural hypotension (hemodynamic instability), and is a marker for vasculopathy in other organs including the heart.90
Chronic Sympathetic Nervous System Stimulation
Chronic sympathetic nervous system stimulation may increase morbidity and mortality. Pheochromocytoma is characterized by explosive release of catecholamines. Even physiologic responses and surgical stress that lead to sustained autonomic nervous system hyperactivity can result in metabolic and endocrine responses. Interventions that attenuate stress responses during the entire perioperative period (continuous epidural infusions of local anesthetics, perioperative administration of β-adrenergic blocking drugs, α2 agonists) may decrease perioperative morbidity and mortality.91–93 Inhaled anesthetics and adjuvants that block the stress response may also be beneficial in long-term outcomes following surgery.94
Acute Denervation
Acute removal of sympathetic nervous system tone, as produced by a regional anesthetic or spinal cord transection, results in immediate maximal vasodilation of blood vessels (spinal shock). In the anesthetic setting, this is transient and can be treated with fluid or α vasoconstrictors. In the chronic setting, over several days, intrinsic tone of vascular smooth muscle increases, usually restoring almost normal vasoconstriction.
Denervation Hypersensitivity
Denervation hypersensitivity is the increased responsiveness (decreased threshold) of the innervated organ to norepinephrine or epinephrine that develops during the first week or so after acute interruption of autonomic nervous system innervation. The presumed mechanism for denervation hypersensitivity is the proliferation of receptors (upregulation) on postsynaptic membranes that occurs when norepinephrine or acetylcholine is no longer released at synapses. As a result, more receptor sites become available to produce an exaggerated response when circulating neurotransmitter does become available.
Adrenal Medulla
The adrenal medulla is innervated by preganglionic fibers that bypass the sympathetic chain. As a result, these fibers pass directly from the spinal cord to the adrenal medulla. Cells of the adrenal medulla are derived embryologically from neural tissue and are analogous to postganglionic sympathetic neurons. Stimulation of the sympathetic nervous system causes release of epinephrine (80%) and norepinephrine from the adrenal medulla. As such, epinephrine and norepinephrine, released by the adrenal medulla into the blood, function as hormones and not as neurotransmitters.
Synthesis
In the adrenal medulla, most of the synthesized norepinephrine is converted to epinephrine by the action of phenylethanolamine-N-methyltransferase (see Fig. 3-37). Activity of this enzyme is enhanced by cortisol, which is carried by the intraadrenal portal vascular system directly to the adrenal medulla. For this reason, any stress that releases glucocorticoids also results in increased synthesis and release of epinephrine.
Release
The triggering event in the release of epinephrine and norepinephrine from the adrenal medulla is the liberation of acetylcholine by preganglionic cholinergic fibers. Acetylcholine acts on α3 and β4 subunit containing nicotinic receptors, resulting in a change in permeability (localized depolarization) that permits entry of sodium, potassium, and calcium ions through extracellular nicotinic acetylcholine channels. Calcium ions result in extrusion, by exocytosis, of synaptic vesicles containing epinephrine.
Norepinephrine and epinephrine released from the adrenal medulla evoke responses similar to direct stimulation of the sympathetic nervous system. The difference, however, is that effects are greatly prolonged (10 to 30 seconds) compared with the brief duration of action on receptors that is produced by norepinephrine released as a neurotransmitter from postganglionic sympathetic nerve endings. The prolonged effect of circulating epinephrine and norepinephrine released by the adrenal medulla reflects the time necessary for metabolism of these substances by COMT and MAO.
Circulating norepinephrine from the adrenal medulla causes vasoconstriction of blood vessels, inhibition of the gastrointestinal tract, increased cardiac activity, and dilation of the pupils (see Table 3-4). The effects of circulating epinephrine differ from those of norepinephrine in that the cardiac and metabolic effects of epinephrine are greater, whereas relaxation of blood vessels in skeletal muscles reflects a predominance of β over α effects at low concentrations of epinephrine. Circulating norepinephrine and epinephrine released by the adrenal medulla and acting as hormones can substitute for sympathetic nervous system innervation of an organ. Another important role of the adrenal medulla is the ability of circulating norepinephrine and epinephrine to stimulate areas of the body that are not directly innervated by the sympathetic nervous system. For example, the metabolic rate of all cells can be influenced by hormones released from the adrenal medulla, even though these cells are not directly innervated by the sympathetic nervous system.
Thermoregulation
Body temperature is determined by the relationship between heat production and heat dissipation. Heat is continually being produced in the body as a product of metabolism. As heat is produced, it is also continuously being lost to the environment. Mammals are homeotherms. Both heat generation and heat loss are adjusted in order to regulate body temperature within narrow limits. Normal core body temperatures range from about 36°C to 37.5°C and undergo circadian fluctuations, being lowest in the morning and highest in the evening. This is consistent with a 10% to 15% decrease in basal metabolic rate during physiologic sleep, presumably reflecting decreased activity of skeletal muscles and the sympathetic nervous system. An estimated 55% of the energy in nutrients is converted to heat during the formation of adenosine triphosphate. The average daily caloric requirement for basal function is approximately 2,000 calories.
Heat Loss
The important mechanisms of heat loss from the body include radiation, conduction, convection, and evaporation. Their relative contributions vary, and depend upon the environmental circumstances.95 The skin is the most important route for heat dissipation, whereas the lungs account for only about 10% of heat loss. Under typical circumstances, most heat (about 60%) is lost by radiation. A warm object emits energy in the form of radiation, predominantly in the infrared range, independent of ambient air temperature. The unclothed human is an excellent source of radiant heat. Significant radiant losses can occur from the unclothed patient in the operating room. In infant incubators, radiant heat losses occur from the exposed infant. Radiant heat loss is countered by heating the surrounding surfaces, so that radiant heat loss is offset by the absorption of radiant heat from nearby surfaces. Radiant heat loss is also countered by blankets, which absorb and then return radiant heat. The extreme example is the “space blanket” which directly reflects infrared radiation back toward the patient.
Conduction of heat from the body occurs by direct contact with a cooler object; for example, between the patient and cold air or an adjacent mattress. The area of the conducting surfaces, the temperature difference, and the heat capacity affect conductive heat transfer. Conductive loss to still air is limited because a stationary layer of air next to the skin acts as a good insulator. Air has a very low heat capacity and warms quickly, thus promptly eliminating the temperature gradient. In humans, piloerection reduces heat loss by trapping a layer of air next to the skin.
Although pure conduction accounts for <5% of heat loss, conductive heat loss to air is greatly facilitated by air movement and is termed convection or facilitated conduction. Thus, a fan is comfortable on a hot summer day because it facilitates heat loss. The rate of convective loss depends on both the air temperature and its velocity (the “wind-chill” phenomenon). Convection accounts for approximately 15% to 30% of heat loss in the operating room, but increases significantly in high wind-chill environments such as a laminar flow unit. However, significant convective heat loss occurs even in a draft-free environment because warmed air rises to be replaced by denser cold air, thus maintaining cutaneous airflow.
Evaporative heat losses are important because significant energy is required to vaporize water. Evaporation from the skin accounts for about 20% of total heat loss. The magnitude of evaporative loss depends on environmental humidity, exposed skin surface area, presence of diaphoresis, wound and bowel exposure, and application of fluid to the skin (prep solutions). Evaporation is the only mechanisms by which the body can eliminate excess heat when the temperature of the surroundings is higher than that of the skin. Diaphoresis occurs in response to stimulation of the preoptic area of the hypothalamus. A normal individual has a maximal sweat production of about 700 mL per hour. With continued exposure to a warm environment, sweat production may increase to 1,500 mL per hour. Evaporation of this amount of sweat can remove heat from the body at a rate of >10 times the normal basal rate of heat production. Evaporation accounts for two-thirds of the heat loss from the respiratory tract. Evaporative heat and fluid loss is an important consideration during surgery in which large segments of moist bowel are exposed for evaporation.
Reductions in core temperature also follow infusions of cold intravenous fluids and blood products.
Regulation of Body Temperature
Body temperature is regulated by feedback mechanisms predominantly mediated by the preoptic nucleus of the anterior hypothalamus,96 which integrates afferent input from thermoreceptors in the skin, deep tissues, and spinal cord. Afferent thermoregulatory input is modulated in the brainstem and spinal cord before arrival in the hypothalamus. Heat-sensitive neurons in the preoptic nucleus receive additional thermal input from extrahypothalamic areas of the brain. Reflex responses to cold (vasoconstriction, piloerection, shivering, and nonshivering thermogenesis) originate in the posterior hypothalamus. Reflex responses to heat (vasodilation, sweating) originate in the anterior hypothalamus.
The hypothalamic thermostat detects body temperature changes and initiates autonomic, somatic, and endocrine thermoresponses when the various set points are reached. However, in the awake individual, behavioral responses (putting on a jacket) usually occur before the core temperature reaches the set points. If the behavioral response to hypothermia fails or is abolished by anesthesia, the hypothalamic thermostat stimulates vasoconstriction at 36.5°C and shivering at 36.2°C. As a result, the rate of heat transfer to the skin is decreased, heat product rises from shivering, and body temperature increases.
There is a narrow range of normal core temperature, 36.7°C to 37.1°C, within which thermoregulatory responses are not triggered. General anesthesia abolishes much of the ability to regulate temperature through drug-induced vasodilation and muscle relaxation. Maintenance of body temperature at a value close to the optimum for enzyme activity assures a constant rate of metabolism, optimal enzyme function, nervous system conduction, and skeletal muscle contraction. Even modest hypothermia (<36°C) reduces the drug metabolism, delaying emergence from anesthesia. Hyperthermia is even less well tolerated, as protein denaturation begins at about 42°C.
Nonshivering Thermogenesis
Nonshivering thermogenesis (alternatively called chemical thermogenesis) is an increase in the rate of cellular metabolism in brown adipose tissue evoked by sympathetic nervous system stimulation or by circulating catecholamines. In adults, who have almost no brown fat, it is rare that chemical thermogenesis increases the rate of heat production by >15%. In infants, however, chemical thermogenesis in brown fat located in the interscapular space and around the great vessels in the thorax and abdomen can increase the rate of heat production by as much as 200%. In contrast to other fat depots, brown fat contains large numbers of mitochondria and has extensive sympathetic innervation. Within these mitochondria, the generation of adenosine triphosphate is uncoupled as oxidative phosphorylation is short-circuited to generate heat. This process is dependent on an uncoupling protein (UCP 1). Lipolysis and heat generation in brown fat is mediated via β-adrenergic receptors.
Shivering
Skeletal muscle activity is a major source of heat. Shivering increases body heat production in response to decreased core temperature. The posterior hypothalamic area responsible for the response to hypothermia controls reflex shivering. Shivering occurs due to both increased motor traffic via anterior motor neurons and to upregulation of the muscle stretch reflex. However, shivering is inefficient and induces significant metabolic demand. Awake patients find shivering intensely unpleasant.
Causes of Increased Body Temperature
A variety of disorders can increase body temperature. Those disorders resulting from thermoregulatory failure (excessive metabolic production of heat, excessive environmental heat, and impaired heat dissipation) are properly characterized as hyperthermia, whereas those resulting from intact homeostatic responses are categorized as fever (Table 3-9).96
In hyperthermic states, the hypothalamic set point is normal but peripheral mechanisms are unable to maintain body temperature that matches the set point. In contrast, fever occurs when the hypothalamic set point is increased by the action of circulating pyrogenic cytokines, causing intact peripheral mechanisms to conserve and generate heat until the body temperature increases to the elevated set point. Despite their physiologic differences, hyperthermia and fever cannot be differentiated clinically based on the height of the temperature or its pattern. However, the clinical management of hyperthermia and fever are very different. The treatment of hyperthermia should be directed at promoting heat dissipation and terminating excessive heat production (e.g., administration of dantrolene for malignant hyperthermia), whereas the treatment of fever should be directed at identification and eradication of pyrogens and lowering the thermoregulatory set point with antipyretic drugs such as aspirin, acetaminophen, and cyclooxygenase inhibitors.
Fever
Pyrogens are bacterial and viral toxins that indirectly cause the set point of the hypothalamic thermostat to increase. Bacterial pyrogens stimulate host inflammatory cells (mononuclear phagocytes) to generate endogenous pyrogens, including interleukins, prostaglandins, and tumor necrosis factor. Viruses do not release pyrogens directly, but stimulate infected cells to release interferons α and β that act as endogenous pyrogens. All known endogenous pyrogens are polypeptides and are therefore unlikely to cross the blood–brain barrier. However, endogenous pyrogens have actions in the organum vasculosum of the lamina terminalis (OVLT), which is a structure adjacent to the lateral ventricles that lies outside the blood–brain barrier. It is likely that endogenous pyrogens acting in the OVLT evoke the release of prostaglandins in the CNS, leading to stimulation of the preoptic nucleus and generation of the febrile response.68
Chills
Sudden resetting of the hypothalamic thermostat to a higher level because of tissue destruction, pyrogens, or dehydration, results in a lag between blood temperature and the new hypothalamic set point. During this period, the person experiences chills and feels cold even though body temperature may be increased. The skin is cold because of cutaneous vasoconstriction. Chills continue until the body temperature increases to the new set point of the hypothalamic thermostat. As long as the process causing the hypothalamic thermostat to be set at a higher level is present, the body’s core temperature will remain increased above normal. Sudden removal of the factor that is causing the body temperature to remain increased is accompanied by intense diaphoresis and feeling of warmth because of generalized cutaneous vasodilation.
Cutaneous Blood Flow
Cutaneous blood flow is a major determinant of heat loss. The cutaneous circulation is among the most variable in the body, reflecting its primary role in regulation of body temperature in response to alterations in the rate of metabolism and the temperature of the external surroundings. The skin’s metabolic needs are so low that the typical cutaneous blood flow is about 10 times higher than needed to supply nutritive needs of the skin.
Cutaneous blood flow is largely regulated by the sympathetic nervous system. Vascular structures concerned with heat loss from skin consist of subcutaneous venous plexuses that can hold large quantities of blood. The cutaneous circulation of the fingers, palms, toes, and earlobes has richly innervated arteriovenous anastomoses that facilitate significant heat loss. In an adult, typical total cutaneous blood flow is about 400 mL per minute. This flow can decrease to as little as 50 mL per minute in severe cold and may increase to as much as 2,800 mL per minute in extreme heat. Patients with borderline cardiac function may become symptomatic in hot environments as the heart attempts to supply increased blood flow to the skin. During acute hemorrhage, the sympathetic nervous system can produce sufficient cutaneous vasoconstriction to transfer large amounts of blood into the central circulation. As such, the cutaneous veins act as an important blood reservoir that can supply 5% to 10% of the blood volume in times of need. Acute hemorrhage may be less well tolerated in a warm environment because the hypothalamic vasodilator response may override the vasoconstrictor response to hypovolemia. Inhaled anesthetics increase cutaneous blood flow, perhaps by inhibiting the temperature-regulating center of the hypothalamus.97
Skin Color
Skin color in light-skinned individuals with little melanin expression is principally due to the color of blood in the cutaneous capillaries and veins. The skin has a pinkish hue when arterial blood is flowing rapidly through these tissues. Conversely, when the skin is cold and blood is flowing slowly, the removal of oxygen for nutritive purposes gives the skin the bluish hue (cyanosis) of deoxygenated blood. Severe vasoconstriction of the skin forces most of this blood into the central circulation, and skin takes on the whitish hue (pallor) of underlying connective tissue, which is composed primarily of collagen fibers.
Perioperative Temperature Changes
The thermoregulatory system contains three key elements: afferent input, central processing, and the efferent response. General anesthesia affects all three elements and regional anesthesia affects both the afferent and efferent components. Thus, anesthesia and surgery in a cool environment makes perioperative hypothermia a likely occurrence (Table 3-10).98,99 General and regional anesthesia increase the interthreshold range to 4.0°C, approximately 20 times the normal range. Typically, the threshold for sweating and vasodilation is increased about 1°C, and the threshold for vasoconstriction and shivering is decreased about 3°C. As a result, anesthetized patients are relatively poikilothermic, with body temperatures determined by the environment. Anesthetics inhibit thermoregulation in a dose-dependent manner and inhibit vasoconstriction and shivering about three times as much as they restrict sweating (Fig. 3-38).100
Alfentanil and propofol similarly lower the threshold for vasoconstriction and sweating. Volatile anesthetics such as isoflurane and desflurane decrease the threshold temperatures for cold responses in a nonlinear fashion. Nonshivering thermogenesis does not occur during general anesthesia in adults or infants.
Sequence of Temperature Changes during Anesthesia
In the awake individual, body heat is unevenly distributed. Tonic thermoregulatory vasoconstriction maintains a temperature gradient between the core and periphery of 2°C to 4°C. The core compartment, which is insulated from the environment by the peripheral compartment, consists of the major viscera and includes the head, chest, abdomen, and pelvis. Under general anesthesia, tonic vasoconstriction is attenuated and heat contained in the core compartment will move to the periphery, thus allowing the core temperature to decrease toward the anesthetic-induced lowered threshold for vasoconstriction. This core to peripheral heat redistribution is responsible for the 1°C to 5°C decrease in core temperature that occurs during the first hour of general anesthesia (Fig. 3-39). For this reason, protection from heat loss early in a surgical procedure is important to reduce the temperature gradient from the environment to the peripheral compartment as significant heat energy has been shunted to the periphery.
After the first hour of general anesthesia, the core temperature usually decreases at a slower rate. This decrease is nearly linear and occurs because continuing heat loss to the environment exceeds the metabolic production of heat. After 3 to 5 hours of anesthesia, the core temperature often stops decreasing (see Fig. 3-39). This thermal plateau may reflect a steady state in which heat loss equals heat production. This type of thermal steady state is especially likely in patients who are well insulated or effectively warmed. However, if a patient becomes sufficiently hypothermic, activation of thermoregulatory vasoconstriction will occur, decreasing cutaneous heat loss and retaining heat in the core compartment. Intraoperative vasoconstriction thus reestablishes the normal core-to-periphery temperature gradient by preventing the loss of centrally generated metabolic heat to peripheral tissues. Although vasoconstriction may effectively maintain the core temperature plateau, mean body temperature and the total heat content of the body continue to decrease as continued loss of heat occurs from the peripheral compartment to the environment. Because reflex vasoconstriction is usually effective in maintaining core temperature, the intraoperative core temperature rarely decreases the additional 1°C necessary to trigger shivering during general anesthesia.99
Although regional anesthesia is thought to have minimal effect upon the central processing and integration of the thermoregulatory response, afferent cold input from the lower body may be overridden by a sense of warmth from cutaneous vasodilation. Decreases in core temperature of a similar or greater magnitude to those experienced during general anesthesia may occur during spinal or epidural techniques despite the sensation of warmth. The initial redistributive temperature drop may be less precipitous during regional anesthesia because vasodilation is restricted to the blocked area. However, because reflex vasoconstriction is abolished below the level of the block, the plateau phase seen during general anesthesia may not occur during regional anesthesia (see Fig. 3-39). Indeed, core temperature may decrease sufficiently during regional anesthesia to trigger the shivering response. However, the ability of reflex shivering to generate heat is markedly attenuated because it is restricted to the unblocked upper body. The risk of significant core hypothermia during regional anesthesia strongly supports the routine use of temperature monitoring. Combined general and regional anesthetic techniques predispose the patient to a greater degree of heat loss than either technique used alone.
Beneficial Effects of Perioperative Hypothermia
Oxygen consumption is decreased by approximately 5% to 7% per degree Celsius of cooling. Thus even moderate decreases in core temperature of 1°C to 3°C below normal provide substantial protection against cerebral ischemia and arterial hypoxemia. Indeed, induced hypothermia to 28°C, as used during cardiopulmonary bypass, will reduce cerebral metabolic rate by 50%. Mild hypothermia (33°C to 36°C) may be recommended during operations likely to be associated with cerebral ischemia such as carotid endarterectomy, aneurysm clipping, and cardiac surgery. Operations involving aortic cross-clamping can jeopardize spinal cord perfusion and may also benefit from the increased margin of safety afforded by mild hypothermia. Mild hypothermia also slows the triggering of malignant hyperthermia.101 Outside the operating room, there has been renewed interest in mild hypothermia during the resuscitation of survivors of cardiac arrest, stroke, traumatic brain injury, acute myocardial infarction, and birth injury,102,103although recent large trials suggest that the hypothermia is less protective than thought.104,105 The main benefit of mild hypothermia accrues from a reduction in metabolic demand. Typical approaches to achieve mild hypothermia often include surface cooling. However, surface cooling may induce shivering, which will delay core cooling.
Adverse Consequences of Perioperative Hypothermia
Perioperative hypothermia may predispose to several significant complications (Table 3-11). These include postoperative shivering (significantly increasing metabolic rate and cardiac work) and impaired coagulation (impaired platelet function, decreased activation of the coagulation cascade). Indeed, hypothermia-induced coagulopathy is associated with increased transfusion requirements. A 1°C decrease in temperature is associated with a 5% reduction in anesthetic requirements (MAC) and an increase in volatile anesthetic blood/gas solubility. Drug metabolism is decreased by hypothermia, particularly that of nondepolarizing neuromuscular-blocking drugs. These factors all conspire to delay emergence from anesthesia and delay recovery room discharge. Hypothermia also impairs wound healing and is associated with decreased resistance to surgical wound infection.100 The underlying mechanism is thought to be hypothermia-induced vasoconstriction, which decreases wound perfusion and local tissue oxygen partial pressure. Perioperative hypothermia is also associated with delayed hospital discharge and an increased catabolic state. Shivering occurs in approximately 40% of unwarmed patients who are recovering from general anesthesia and is associated with substantial sympathetic nervous system activation and discomfort from the sensation of cold. Core hypothermia equal to a 1.5°C decrease triples the incidence of ventricular tachycardia and morbid cardiac events.106
Perioperative Temperature Measurement
The significant adverse physiologic effects of changes in body temperature are a compelling reason to monitor body temperature during anesthesia. Unless hypothermia is specifically indicated, as for protection against tissue ischemia, it is recommended that intraoperative core temperature be maintained at ≥36°C.100 Measuring the temperature of the lower 25% of the esophagus (about 24 cm beyond the corniculate cartilages or site of the loudest heart sounds heard through an esophageal stethoscope) gives a reliable approximation of blood and cerebral temperature. Readings elsewhere in the esophagus are more likely to be influenced by the temperature of inhaled gases. A nasopharyngeal temperature probe positioned behind the soft palate gives a less reliable measure of cerebral temperature than a correctly positioned esophageal probe. Leakage of gases around the tracheal tube may also influence nasopharyngeal temperature measurements. Heat-producing bacteria in the gastrointestinal tract, cold blood returning from the lower limbs, and insulation of the probe by feces, can all influence rectal temperature. Bladder temperature is also subject to a prolonged response time, particularly if urine flow is <270 mL per hour.107 Tympanic membrane and aural canal temperatures provide a rapidly responsive and accurate estimate of hypothalamic temperature and correlate well with esophageal temperature. Potential damage to the tympanic membrane has limited the acceptance of tympanic membrane probes. However, infrared thermometers allow atraumatic measurement of tympanic temperature. However, the accuracy of individual infrared thermometers is dependent on instrumental design and positioning. Thermistors in pulmonary artery catheters provide the best continuous estimate of body temperature but are invasive. Skin temperature gives no information other than the temperature of that area of the skin.
Prevention of Perioperative Hypothermia
Passive or active airway heating and humidification contribute little to perioperative thermal management in adults because <10% of metabolic heat is lost via ventilation.100 Each liter of intravenous fluid at ambient temperature that is infused into adult patients, or each unit of blood at 4°C decreases the mean core body temperature about 0.25°C. In this regard, the administration of unwarmed fluids can markedly decrease body temperature. Warming fluids to near 37°C is useful for preventing hypothermia, especially if large volumes of fluid are being infused.
The skin is the predominant source of heat loss during anesthesia and surgery, although evaporation from large surgical incisions may also be important. A high ambient temperature maintains normothermia in anesthetized patients, but temperatures of >25°C are uncomfortable for operating room personnel.
Covering the skin with surgical drapes or blankets can decrease cutaneous heat loss. A single layer of insulator decreases heat loss by approximately 30%, but additional layers do not proportionately increase the benefit.108 For this reason, active warming is needed to prevent intraoperative hypothermia. Forced-air warming is probably the most effective method available, although any method or combination of methods that maintains core body temperature near 36°C is acceptable (Fig. 3-40).100 Circulating warm water mattresses are generally ineffective because cutaneous blood flow to the back is limited in the supine position. Patients undergoing minor operations in a warm environment may not require active warming, whereas forced-air warming, alone or combined with fluid warming, is helpful for maintaining normal intraoperative core temperature in most other instances.
References
1. Alkire MT, Hudetz AG, Tononi G. Consciousness and anesthesia. Science. 2008;322:876–880.
2. Hartline DK, Colman DR. Rapid conduction and the evolution of giant axons and myelinated fibers. Curr Biol. 2007;17:R29–R35.
3. Guyton AC, Hall JE. Textbook of Medical Physiology. 10th ed. Philadelphia, PA: Saunders; 2000.
4. Perreault L, Drolet P, Farny J. Ulnar nerve palsy at the elbow after general anaesthesia. Can J Anaesth. 1992;39:499–503.
5. Berne RM, Levy MN, Koeppen BM, et al. Physiology. 5th ed. St. Louis, MO: Mosby; 2004.
6. Ackerman MJ, Clapham DE. Ion channels-basic science and clinical disease. N Engl J Med. 1997;336:1575–1586.
7. Zecharia AY, Nelson LE, Gent TC, et al. The involvement of hypothalamic sleep pathways in general anesthesia: testing the hypothesis using the GABAA receptor beta3N265M knock-in mouse. J Neurosci. 2009;29:2177–2187.
8. Hudspith MJ. Glutamate: a role in normal brain function, anaesthesia, analgesia and CNS injury. Br J Anaesth. 1997;78:731–747.
9. Sonner JM. A hypothesis on the origin and evolution of the response to inhaled anesthetics. Anesth Analg. 2008;107:849–854.
10. Gelman AC. G proteins: transducers of receptor-generated signals. Annu Rev Biochem. 1987;56:615–649.
11. Maze M. Transmembrane signaling and the Holy Grail of anesthesia. Anesthesiology. 1990;72:959–961.
12. Schwinn DA. Adrenoceptors as models for G protein-coupled receptors: structure, function, and regulation. Br J Anaesth. 1993;71:77–85.
13. Doyle GA, Sheng XR, Lin SS, et al. Identification of five mouse mu-opioid receptor (MOR) gene (Oprm1) splice variants containing a newly identified alternatively spliced exon. Gene. 2007;395:98–107.
14. Ward CW, Lawrence MC. Ligand-induced activation of the insulin receptor: a multi-step process involving structural changes in both the ligand and the receptor. Bioessays. 2009;31:422–434.
15. Bourne HR. How receptors talk to trimeric G proteins. Curr Opin Cell Biol. 1997;9:134–142.
16. Nelson LE, Lu J, Guo T, et al. The alpha2-adrenoceptor agonist dexmedetomidine converges on an endogenous sleep-promoting pathway to exert its sedative effects. Anesthesiology. 2003;98:428–436.
17. Flood P, Ramirez-Latorre J, Role L. alpha 4 beta 2 Neuronal nicotinic acetylcholine receptors in the central nervous system are potently inhibited by both isoflurane and propofol: alpha 7 Type nAChRs are unaffected. Anesthesiology. 1997;86:859–865.
18. Violet JM, Downie DL, Nakisa RC, et al. Differential sensitivities of mammalian neuronal and muscle nicotinic acetylcholine receptors to general anesthetics. Anesthesiology. 1997;86:866–874.
19. Evers AS, Steinbach JH. Supersensitive sites in the central nervous system. Anesthetics block brain nicotinic receptors. Anesthesiology. 1997;86:760–762.
20. Alkire MT, McReynolds JR, Hahn EL, et al. Thalamic microinjection of nicotine reverses sevoflurane-induced loss of righting reflex in the rat. Anesthesiology. 2007;107:264–272.
21. Rivera C, Voipio J, Payne JA, et al. TheK+/Cl− co-transporter KCC2 renders GABA hyperpolarizing during neuronal maturation. Nature. 1999;397:251–255.
22. Ovassapian A, Joshi CW, Brunner EA. Visual disturbances: an unusual symptom of transurethral prostatic resection reaction. Anesthesiology. 1982;57:332–334.
23. Wang JML, Creel DJ, Wong KC. Transurethral resection of the prostate: serum glycine levels and ocular evoked potentials. Anesthesiology. 1989;70:36–41.
24. Yamada M, Inanobe A, Kurachi Y. G protein regulation of potassium ion channels. Pharmacol Rev. 1998;50:723–760.
25. Pfaffinger PJ, Martin JM, Hunter DD, et al. GTP-binding proteins couple cardiac muscarinic receptors to a K channel. Nature. 1985;317:536–538.
26. Kubo Y, Adelman JP, Clapham DE, et al. International Union of Pharmacology. LIV. Nomenclature and molecular relationships of inwardly rectifying potassium channels. Pharmacol Rev. 2005;57:509–526.
27. Brooks CM, Eccles JC. Electrical investigation of the monosynaptic pathway through the spinal cord. J Neurophysiol. 1947;10:251–273.
28. Budson AE, Price BH. Memory dysfunction. N Engl J Med. 2005;352:692–699.
29. Blundon JA, Zakharenko SS. Dissecting the components of long-term potentiation. Neuroscientist. 2008;14:598–608.
30. Born J. Slow-wave sleep and the consolidation of long-term memory. World J Biol Psychiatry. 2010;11(suppl 1):16–21.
31. Moller JT, Cluitmans P, Rasmussen LS, et al. Long-term postoperative cognitive dysfunction in the elderly. Lancet. 1998;351:857–861.
32. Carnini A, Lear JD, Eckenhoff RG. Inhaled anesthetic modulation of amyloid beta(1-40) assembly and growth. Curr Alzheimer Res. 2007;4:233–241.
33. Ghoneim MM. Awareness during anesthesia. Anesthesiology. 2000;92:597–602.
34. Sandin RH, Enlund G, Samuelsson P, et al. Awareness during anaesthesia: a prospective case study. Lancet. 2000;355:707–711.
35. Sebel PS, Bowdle TA, Ghoneim MM, et al. The incidence of awareness during anesthesia: a multicenter United States study. Anesth Analg. 2004;99:833–839.
36. Myles PS, Leslie K, McNeil J, et al. Bispectral index monitoring to prevent awareness during anaesthesia: the B-Aware randomized controlled trial. Lancet. 2004;363:1757–1763.
37. Avidan MS, Zhang L, Burnside BA, et al. Anesthesia awareness and the bispectral index. N Engl J Med. 2008;358(11):1097–1108.
38. Lyons G, MacDonald R. Awareness during cesarean section. Anaesthesia. 1991;46:62–64.
39. Ranta S, Jussila J, Hynyen M. Recall of awareness during cardiac anaesthesia: influence of feedback information to the anaesthesiologist. Acta Anaesthesiol Scand. 1996;40:554–560.
40. Bogetz MS, Katz JA. Recall of surgery for major trauma. Anesthesiology. 1984;61:6–9.
41. Dwyer R, Bennett HL, Eger EI II, et al. Effects of isoflurane and nitrous oxide in subanesthetic concentrations on memory and responsiveness in volunteers. Anesthesiology. 1992;77:888–898.
42. Dwyer R, Bennett HL, Eger EI II, et al. Isoflurane anesthesia prevents unconscious learning. Anesth Analg. 1992;75:107–112.
43. Naguib M, Schmid PG III, Baker MT. The electroencephalographic effects of IV anesthetic doses of melatonin: comparative studies with thiopental and propofol. Anesth Analg. 2003;97:238–243.
44. Franks NP. General anaesthesia: from molecular targets to neuronal pathways of sleep and arousal. Nat Rev Neurosci. 2008;9:370–386.
45. Sprung J, Wilt S, Bourke D, et al. Is it time to correct the dermatome chart of the anterior scrotal region? Anesthesiology. 1993;79:381–383.
46. Savolaine ER, Pandya JB, Greenblatt SH, et al. Anatomy of the human lumbar epidural space. New insights using CT-epidurography. Anesthesiology. 1988;68:217–220.
47. Gallart L, Blanco D, Samso E, et al. Clinical and radiologic evidence of the epidural plica medina dorsalis. Anesth Analg. 1990;71:698–701.
48. Lirk J, Colvin B, Steger B, et al. Incidence of lower thoracic ligamentum flavum midline gaps. Br J Anaesth. 2005;94:852–855.
49. Gilman S. Advances in neurology. N Engl J Med. 1992;326:1608–1616.
50. Brian JE. Carbon dioxide and the cerebral circulation. Anesthesiology. 1998;88:1365–1386.
51. Sigl J, Chamoun N. An introduction to bispectral analysis for the electroencephalogram. J Clin Monit. 1994;10:392–404.
52. Flaishon R, Windsor A, Sigl J, et al. Recovery of consciousness after thiopental or propofol. Bispectral index and the isolated forearm technique. Anesthesiology. 1997;86:613–619.
53. Kearse LA, Manberg P, Chamoun N, et al. Bispectral analysis of the electroencephalogram correlates with patient movement to skin incision during propofol/nitrous oxide anesthesia. Anesthesiology. 1994;81:1365–1370.
54. Song D, Joshi G, White PF. Titration of volatile anesthetics using bispectral index facilitates recovery after ambulatory anesthesia. Anesthesiology. 1997;87:842–848.
55. Gan TJ, Glass PS, Windsor A, et al. Bispectral index monitoring allows faster recovery from propofol, alfentanil, and nitrous oxide anesthesia. Anesthesiology. 1997;87:808–815.
56. Ekman A, Lindholm M-L, Lennmarken C, et al. Reduction in the incidence of awareness using BIS monitoring. Acta Anaesthesiol Scand. 2004;48:20–26.
57. Lennmarken C, Sandin R. Neuromonitoring for awareness during surgery. Lancet. 2004;363:1747–1748.
58. Orser BA. Depth-of-anesthesia monitor and the frequency of intraoperative awareness. N Engl J Med. 2008;358:1189–1191.
59. Schmidt GN, Bischoff P, Standl T, et al. Comparative evaluation of the Datex-Ohmeda S/5 entropy module and the Bispectral Index® monitor during propofol-remifentanil anesthesia. Anesthesiology. 2004;101:1283–1290.
60. Höcker J, Raitschew B, Meybohm P, et al. Differences between bispectral index and spectral entropy during xenon anaesthesia: a comparison with propofol anaesthesia. Anaesthesia. 2010;65:595–600.
61. Kofke WA, Young RSK, Davis P, et al. Isoflurane for refractory status epilepticus: a clinical series. Anesthesiology. 1989;71:653–659.
62. Ginsburg HH, Shetter AG, Raudzens PA. Postoperative paraplegia with preserved intraoperative somatosensory evoked potentials. J Neurosurg. 1985;63:296–299.
63. Pathak KS, Ammadio M, Kalamchi A, et al. Effects of halothane, enflurane, and isoflurane on somatosensory evoked potentials during nitrous oxide anesthesia. Anesthesiology. 1987;66:753–757.
64. Schubert A, Drummond JC. The effect of acute hypocapnia on human median nerve somatosensory evoked responses. Anesth Analg. 1986;65:240–244.
65. Adams DC, Emerson RG, Heyer EJ, et al. Monitoring of intraoperative motor-evoked potentials under condition of controlled neuromuscular blockade. Anesth Analg. 1993;77:913–918.
66. Haenggi M, Ypparila H, Takala J, et al. Measuring depth of sedation with auditory evoked potentials during controlled infusion of propofol and remifentanil in health volunteers. Anesth Analg. 2004;99:1728–1736.
67. Chi OZ, Field C. Effects of isoflurane on visual evoked potentials in humans. Anesthesiology. 1986;65:328–330.
68. Ganong WF. Review of Medical Physiology. 21st ed. New York, NY: Lange Medical Books/McGraw-Hill; 2003.
69. Johnson RW. Anatomy for ophthalmic anaesthesia. Br J Anaesth. 1995;75:80–87.
70. Williams EL, Hart WM, Tempelhoff R. Postoperative ischemic optic neuropathy. Anesth Analg. 1995;80:1018–1029.
71. Janicki PK, Pai R, Wrights JK, et al. Ischemic optic neuropathy after liver transplantation. Anesthesiology. 2001;94:361–363.
72. Myers MA, Hamilton SR, Bogosian AJ, et al. Visual loss as a complication of spine surgery. Spine. 1997;22:1325–1329.
73. Roth S, Barach P. Postoperative visual loss: still no answers—yet. Anesthesiology. 2001;95:575–577.
74. Cheng MA, Todorov A, Tempelhoff R, et al. The effect of prone positioning on intraocular pressure in anesthetized patients. Anesthesiology. 2001;95:1351–1355.
75. Lee LA, Lam AM. Unilateral blindness after prone lumbar spine surgery. Anesthesiology. 2001;95:793–795.
76. Owens WD, Gustave F, Schlaroff A. Tympanic membrane rupture with nitrous oxide anesthesia. Anesth Analg. 1978;57:283–286.
77. Sprung J, Bourke DL, Contreras MG, et al. Perioperative hearing impairment. Anesthesiology. 2003;98:241–257.
78. Watcha MR, White PF. Postoperative nausea and vomiting. Its etiology, treatment, and prevention. Anesthesiology. 1992;77:162–184.
79. Azzam FJ. A simple and effective method for stopping post-anesthesia clonus. Anesthesiology. 1987;66:98.
80. Johns RA. EDRF/nitric oxide. the endogenous nitrovasodilator and a new cellular messenger. Anesthesiology. 1991;75:927–933.
81. Sudhof TC. The synaptic vesicle cycle revisited. Neuron. 2000;28:317–323.
82. Ahlquist RP. A study of adrenotropic receptors. Am J Physiol. 1948;53:586–606.
83. Insel PA. Adrenergic receptors—evolving concepts and clinical implications. N Engl J Med. 1996;334:580–589.
84. Lefkowitz RJ, Rockman HA, Koch WJ. Catecholamines, cardiac beta-adrenergic receptors, and heart failure. Circulation. 2000;101:1634–1640.
85. Martyn JA, White DA, Gronert GA, et al. Up-and-down regulation of skeletal muscle acetylcholine receptors. Effects on neuromuscular blockers. Anesthesiology. 1992;76:822–830.
86. Hosey MM. Diversity of structure, signaling and regulation within the family of muscarinic cholinergic receptors. FASEB J. 1992;6:845–851.
87. Charlson ME, MacKenzie CR, Gold JP. Preoperative autonomic function abnormalities in patients with diabetes mellitus and patients with hypertension. J Am Coll Surg. 1994;179:1–6.
88. Veith RC, Featherstone JA, Linares OA, et al. Age differences in plasma norepinephrine kinetics in humans. J Gerontol. 1986;41:319–325.
89. Lakatta ED. Deficient neuroendocrine regulation of the cardiovascular system with advancing age in healthy humans. Circulation. 1993;87:631–637.
90. Burgos LG, Ebert TJ, Asiddao C, et al. Increased intraoperative cardiovascular morbidity in diabetics with autonomic neuropathy. Anesthesiology. 1989;70:591–599.
91. Mangano DT, Layug EL, Wallace A, et al. Effect of atenolol on mortality and cardiovascular morbidity after noncardiac surgery. N Engl J Med. 1996;335:1713–1719.
92. Kehlet H. Manipulation of the metabolic response in clinical practice. World J Surg. 2000;24:690–698.
93. Wallace AW, Galindez D, Salahieh A, et al. Effect of clonidine on cardiovascular morbidity and mortality after noncardiac surgery. Anesthesiology. 2004;101:284–293.
94. Ebert TJ, Perez F, Uhrich TD, et al. Desflurane-mediated sympathetic activation occurs in humans despite preventing hypotension and baroreceptor unloading. Anesthesiology. 1998;88:1227–1235.
95. Buggy DJ, Crossley AWA. Thermoregulation, mild perioperative hypothermia, and post-anaesthetic shivering. Br J Anaesth. 2000;84:615–628.
96. Simon HB. Hyperthermia, fever and fever of undetermined origin. In: Rubenstein E, Federman D, eds. ACP Medicine. New York, NY: WebMD, Inc; 2003.
97. Heistad DD, Abboud FM. Factors that influence blood flow in skeletal muscle and skin. Anesthesiology. 1974;41:139–156.
98. Giesbrecht GG. Human thermoregulatory inhibition by regional anesthesia. Anesthesiology. 1994;81:277–281.
99. Sessler DI. Perioperative heat balance. Anesthesiology. 2000;92:578–599.
100. Sessler DI. Mild perioperative hypothermia. N Engl J Med. 1997;336:1630–1637.
101. Iaizzo PA, Kehler CH, Carr RJ, et al. Prior hypothermia attenuates malignant hyperthermia in susceptible swine. Anesth Analg. 1996;82:803–809.
102. Zviman MM, Roguin A, Jacobs A, et al. A new method for inducing hypothermia during cardiac arrest. Crit Care Med. 2004;32:S369–S373.
103. Gunn AJ, Thoresen M. Hypothermic neuroprotection. NeuroRx. 2006;3:154–169.
104. Todd MM, Hindman BJ, Clarke WR, et al. Intraoperative Hypothermia for Aneurysm Surgery Trial (IHAST) Investigators. Mild intraoperative hypothermia during surgery for intracranial aneurysm. N Engl J Med. 2005;352:135–145.
105. Hindman BJ, Bayman EO, Pfisterer WK, et al. IHAST Investigators. No association between intraoperative hypothermia or supplemental protective drug and neurologic outcomes in patients undergoing temporary clipping during cerebral aneurysm surgery: findings from the Intraoperative Hypothermia for Aneurysm Surgery Trial. Anesthesiology. 2010;112:86–101.
106. Frank SM, Fleisher LA, Breslow MJ, et al. Perioperative maintenance of normothermia reduces the incidence of morbid cardiac events: a randomized clinical trial. JAMA. 1997;277:1127–1134.
107. Imrie MM, Hall GM. Body temperature and anaesthesia. Br J Anaesth. 1990;64:346–354.
108. Sessler DI, Schroeder M. Heat loss in humans covered with cotton hospital blankets. Anesth Analg. 1993;77:73–77.