Ketamine has multiple effects throughout the central nervous system, and it is well

recognized to inhibit N-methyl-D-aspartate (NMDA) channels and neuronal

hyperpolarization-activated cationic (HCN1) channels. Nevertheless, exactly how ketamine
produces either anesthesia or analgesia remains controversial. Ketamine functionally
“dissociates” sensory impulses from the limbic cortex (which is involved with the awareness
of sensation). Clinically, this state of dissociative anesthesia may cause the patient to appear
conscious (eg, eye opening, swallowing, muscle contracture) but unable to process or respond
to sensory input. Ketamine may have additional actions on endogenous analgesic pathways.
Ketamine has effects on mood, and infusions of this agent are now being widely used to treat
severe, treatment-resistant depression, particularly when patients have suicidal ideation.
Small infusion doses of ketamine are also being used to supplement general anesthesia and to
reduce the need for opioids both during and after the surgical procedure. Low-dose infusions
of ketamine have been used for analgesia in postoperative patients and others who are
refractory to conventional analgesic approaches. Ketamine has been identified by the World
Health Organization as a key necessary agent.
Ketamine (Figure 9–4) is a structural analogue of phencyclidine (a veterinary
anesthetic and a drug of abuse). It is one-tenth as potent, yet retains many of phencyclidine’s
psychotomimetic effects. Ketamine is used for intravenous induction of anesthesia,
particularly in settings where its tendency to produce sympathetic stimulation are useful
(hypovolemia, trauma). When intravenous access is lacking, ketamine is useful for
intramuscular induction of general anesthesia in children and uncooperative adults. Ketamine
can be combined with other agents (eg, propofol or midazolam) in small bolus doses or
infusions for conscious sedation during procedures such as nerve blocks and endoscopy.
Even subanesthetic doses of ketamine may cause hallucinations but usually do not do so in
clinical practice, where many patients will have received at least a small dose of midazolam
(or a related agent) for amnesia and sedation. The increased anesthetic potency and decreased
psychotomimetic side effects of one isomer (S[+] versus R[–]) are the result of stereospecific
receptors. The single S(+) stereoisomer preparation is not available in the United States (but
widely available elsewhere throughout the world), and it has considerably greater affinity
than the racemic mixture for the NMDA receptor as well as several-fold greater potency as a
general anesthetic.

Pharmacokinetics
A. Absorption
 Ketamine has been administered orally, nasally, rectally, subcutaneously, and
epidurally, but in usual clinical practice it is given intravenously or intramuscularly (Table 9–
3). Peak plasma levels are usually achieved within 10 to 15 min after intramuscular injection.
TABLE 9–3 Uses and doses of ketamine, etomidate, and propofol.

B. Distribution
Ketamine is highly lipid soluble and, along with ketamine-induced increase in cerebral
blood flow and cardiac output, results in rapid brain uptake and subsequent redistribution (the
distribution half-life is 10–15 min). Awakening is due to redistribution from brain to
peripheral compartments.
C. Biotransformation
Ketamine is biotransformed in the liver to several metabolites, one of which
(norketamine) retains anesthetic activity. Patients receiving repeated doses of ketamine (eg,
for daily changing of dressings on burns) develop tolerance, and this can only be partially
explained by induction of hepatic enzymes. Extensive hepatic uptake (hepatic extraction ratio
of 0.9) explains ketamine’s relatively short elimination half-life (2 h).
D. Excretion
End products of ketamine biotransformation are excreted renally.

Effects on Organ Systems

A. Cardiovascular
In contrast to other anesthetic agents, ketamine increases arterial blood pressure, heart
rate, and cardiac output (Table 9–4), particularly after rapid bolus injections. These indirect
cardiovascular effects are due to central stimulation of the sympathetic nervous system and
inhibition of the reuptake of norepinephrine after release at nerve terminals. Accompanying
these changes are increases in pulmonary artery pressure and myocardial work. For these
reasons, ketamine should be administered carefully to patients with coronary artery disease,
uncontrolled hypertension, congestive heart failure, or arterial aneurysms. The direct
myocardial depressant effects of large doses of ketamine, probably due to inhibition of
calcium transients, are unmasked by sympathetic blockade (eg, spinal cord transection) or
exhaustion of catecholamine stores (eg, severe endstage shock).
TABLE 9–4 Summary of nonvolatile anesthetic effects on organ systems.

B. Respiratory
Ventilatory drive is minimally affected by induction doses of ketamine, although
combinations of ketamine with opioids may produce apnea. Racemic ketamine is a potent
bronchodilator, making it a good induction agent for asthmatic patients; however, S(+)
ketamine produces minimal bronchodilation. Upper airway reflexes remain largely intact, but
partial airway obstruction may occur, and patients at increased risk for aspiration pneumonia
(“full stomachs”) should be intubated during ketamine general anesthesia (see Case
Discussion, Chapter 17). The increased salivation associated with ketamine can be attenuated
by premedication with an anticholinergic agent such as glycopyrrolate.
C. Cerebral
The received dogma about ketamine is that it increases cerebral oxygen consumption,
cerebral blood flow, and intracranial pressure. These effects would seem to preclude its use in
patients with space-occupying intracranial lesions such as occur with head trauma; however,
recent publications offer convincing evidence that when combined with a benzodiazepine (or
another agent acting on the same GABA receptor system) and controlled ventilation (in
techniques that exclude nitrous oxide), ketamine is not associated with increased intracranial
pressure. Myoclonic activity is associated with increased subcortical electrical activity, which
is not apparent on surface EEG. Undesirable psychotomimetic side effects (eg, disturbing
dreams and delirium) during emergence and recovery are less common in children, in patients
premedicated with benzodiazepines, or in those receiving ketamine combined with propofol
in a total intravenous anesthesia (TIVA) technique. Of the nonvolatile agents, ketamine
comes closest to being a “complete” anesthetic as it induces analgesia, amnesia, and
unconsciousness.
Drug Interactions
Ketamine interacts synergistically (more than additive) with volatile anesthetics but in an
additive way with propofol, benzodiazepines, and other GABAreceptor– mediated agents. In
animal experiments nondepolarizing neuromuscular blocking agents are minimally
potentiated by ketamine (see Chapter 11). Diazepam and midazolam attenuate ketamine’s
cardiostimulatory effects and diazepam prolongs ketamine’s elimination half-life. α-
Adrenergic and β-adrenergic antagonists (and other agents and techniques that diminish
sympathetic stimulation) unmask the direct myocardial depressant effects of ketamine, which
are normally overwhelmed by sympathetic stimulation. Concurrent infusion of ketamine and
propofol, often in a fixed infusion (mg:mg) ratio of 1:10, has achieved great popularity for
sedation with local and regional anesthesia or intravenous general anesthesia in office-based
settings.

Neuromuscular Transmission
Association between a motor neuron and a muscle cell occurs at the neuromuscular junction
(Figure 11–1). The cell membranes of the neuron and muscle fiber are separated by a narrow
(20-nm) gap, the synaptic cleft. As a nerve’s action potential depolarizes its terminal, an
influx of calcium ions through voltage-gated calcium channels into the nerve cytoplasm
allows storage vesicles to fuse with the terminal plasma membrane and release their contents
(acetylcholine [ACh]). The ACh molecules diffuse across the synaptic cleft to bind with
nicotinic cholinergic receptors on a specialized portion of the muscle membrane, the motor
end-plate. Each neuromuscular junction contains approximately 5 million of these receptors,
but activation of only about 500,000 receptors is required for normal muscle contraction.
The structure of ACh receptors varies in different tissues and at different times in
development. Each ACh receptor in the neuromuscular junction normally consists of five
protein subunits: two α subunits, and single β, δ, and ε subunits. Only the two identical α
subunits are capable of binding Ach molecules. If both binding sites are occupied by ACh, a
conformational change in the subunits briefly (1 ms) opens an ion channel in the core of the
receptor (Figure 11–2). The channel will not open if ACh binds to only one site. In contrast
to the normal (or mature) junctional ACh receptor, another isoform contains a γ subunit
instead of the ε subunit. This isoform is referred to as the fetal or immature receptor because
it is in the form initially expressed in fetal muscle. It is also often referred to as
extrajunctional because, unlike the mature isoform, it may be located anywhere in the muscle
membrane, inside or outside the neuromuscular junction when expressed in adults.
Cations flow through the open ACh receptor channel (sodium and calcium in; potassium out),
generating an end-plate potential. The contents of a single vesicle, a quantum of ACh (104
molecules per quantum), produce a miniature end-plate potential. The number of quanta
released by each depolarized nerve fiber, normally at least 200, is very sensitive to
extracellular ionized calcium concentration; increasing calcium concentration increases the
number of quanta released. When sufficient receptors are occupied by ACh, the end-plate
potential will be strong enough to depolarize the perijunctional membrane. Voltage-gated
sodium channels within this portion of the muscle membrane open when a threshold voltage
is developed across them, as is true for voltage-gated sodium channels in nerve or heart
(Figure 11–3). Perijunctional areas of muscle membrane have a higher density of these
sodium channels than other parts of the membrane. The resulting action potential propagates
along the muscle membrane and T-tubule system, opening sodium channels and releasing
calcium from the sarcoplasmic reticulum. This intracellular calcium allows the contractile
proteins actin and myosin to interact, bringing about muscle contraction. The amount of ACh
released and the number of receptors subsequently activated with efferent nerve
depolarization will normally far exceed the minimum required for the initiation of an action
potential in the muscle. The nearly tenfold margin of safety is reduced in Eaton–Lambert
myasthenic syndrome (decreased release of ACh) and myasthenia gravis (decreased number
of receptors). transmembrane protein that can be conceptualized as having two gates. Sodium
ions pass only when both gates are open. Opening of the gates is time dependent and voltage
dependent. The channel therefore possesses three functional states. At rest, the lower gate is
open but the upper gate is closed (A). When the muscle membrane reaches threshold voltage
depolarization, the upper gate opens and sodium can pass (B). Shortly after the upper gate
opens the time-dependent lower gate closes (C). When the membrane repolarizes to its
resting voltage, the upper gate closes and the lower gate opens (A). ACh is rapidly
hydrolyzed into acetate and choline by the substrate-specific enzyme acetylcholinesterase.
This enzyme is embedded into the motor endplate membrane immediately adjacent to the
ACh receptors. After unbinding ACh, the receptors’ ion channels close, permitting the end-
plate to repolarize. Calcium is resequestered in the sarcoplasmic reticulum, and the muscle
cell relaxes.
ROCURONIUM
Physical Structure
This monoquaternary steroid analogue of vecuronium was designed to provide a rapid onset
of action.
Metabolism & Excretion
Rocuronium undergoes no metabolism and is eliminated primarily by the liver and slightly by
the kidneys. Its duration of action is not significantly affected by renal disease, but it is
modestly prolonged by severe liver failure and pregnancy. Because rocuronium does not have
active metabolites, it may be a better choice than vecuronium in the rare patient requiring
prolonged infusions in the intensive care unit setting. Elderly patients may experience a
prolonged duration of action due to decreased liver mass.
Dosage
Rocuronium is less potent than most other steroidal muscle relaxants (potency seems to be
inversely related to speed of onset). It requires 0.45 to 0.9 mg/kg intravenously for intubation
and 0.15 mg/kg boluses for maintenance. Intramuscular rocuronium (1 mg/kg for infants; 2
mg/kg for children) provides adequate vocal cord and diaphragmatic paralysis for intubation,
but not until after 3 to 6 min (deltoid injection has a faster onset than quadriceps). The
infusion requirements for rocuronium range from 5 to 12 mcg/kg/min. Rocuronium can
produce an unexpectedly prolonged duration of action in elderly patients. Initial dosage
requirements are modestly increased in patients with advanced liver disease, presumably due
to a larger volume of distribution.
Side Effects & Clinical Considerations
Rocuronium (at a dose of 0.9–1.2 mg/kg) has an onset of action that approaches
succinylcholine (60–90 s), making it a suitable alternative for rapidsequence inductions, but
at the cost of a much longer duration of action. This intermediate duration of action is
comparable to vecuronium or atracurium. Sugammadex permits rapid reversal of dense
rocuronium-induced neuromuscular blockade. Rocuronium (0.1 mg/kg) has been shown to be
a rapid (90 s) and effective agent (decreased fasciculations and postoperative myalgias) for
precurarization prior to administration of succinylcholine. It has slight vagolytic tendencies.

OPIOIDS
Mechanisms of Action
Opioids bind to specific receptors located throughout the central nervous system and other
tissues. Four major opioid receptor types have been identified (Table 10–1): mu (μ, with
subtypes μ1 and μ2), kappa (κ), delta (δ), and sigma (σ). All opioid receptors couple to G
proteins; binding of an agonist to an opioid receptor causes membrane hyperpolarization.
Acute opioid effects are mediated by inhibition of adenylyl cyclase (reductions in
intracellular cyclic adenosine monophosphate concentrations) and activation of
phospholipase C. Opioids inhibit voltage-gated calcium channels and activate inwardly
rectifying potassium channels. Opioid effects vary based on the duration of exposure, and
opioid tolerance leads to changes in opioid responses.
TABLE 10–1 Classification of opioid receptors.
Although opioids provide some degree of sedation and in some species can produce general
anesthesia when given in large doses, they are principally used to provide analgesia. The
clinical actions of opioids depend on which receptor is bound (and in the case of spinal and
epidural administration of opioids, the location in the neuraxis where the receptor is located)
and the binding affinity of the drug. Agonist–antagonists (eg, nalbuphine, nalorphine,
butorphanol, and buprenorphine) have less efficacy than full agonists (eg, fentanyl or
morphine), and under some circumstances agonist–antagonists will antagonize the actions of
full agonists. Pure opioid antagonists (eg, naloxone or naltrexone) are discussed in Chapter
17. Opioid compounds mimic endorphins, enkephalins, and dynorphins, endogenous peptides
that bind to opioid receptors. Opioid receptor activation inhibits the presynaptic release
and postsynaptic response to excitatory neurotransmitters (eg, acetylcholine, substance
P) released by nociceptive neurons. Transmission of pain impulses can be selectively
modified at the level of the dorsal horn of the spinal cord with intrathecal or epidural
administration of opioids. Opioid receptors also respond to systemically administered
opioids. Modulation through a descending inhibitory pathway from the periaqueductal gray
matter to the dorsal horn of the spinal cord may also play a role in opioid analgesia. Although
opioids exert their greatest effect within the central nervous system, opioid receptors have
also been identified on somatic and sympathetic peripheral nerves. Certain opioid side effects
(eg, constipation) are the result of opioid binding to receptors in peripheral tissues (eg, the
gastrointestinal tract), and there are now selective antagonists for opioid actions outside the
central nervous system (alvimopan and methylnaltrexone). The clinical importance of opioid
receptors on primary sensory nerves (if present) remains speculative, despite the persisting
practice of compounding of opioids in local anesthetic solutions applied to peripheral nerves.
Effects on Organ Systems
A. Cardiovascular
In general, opioids have minimal direct effects on the heart. Meperidine tends to increase
heart rate (it is structurally similar to atropine and was originally synthesized as an atropine
replacement), whereas larger doses of morphine, fentanyl, sufentanil, remifentanil, and
alfentanil are associated with a vagus nerve–mediated bradycardia. With the exception of
meperidine (and only then at very large doses), the opioids do not depress cardiac
contractility provided they are administered alone (which is almost never the case in surgical
anesthetic settings). Nonetheless, arterial blood pressure often falls as a result of opioid
induced bradycardia, venodilation, and decreased sympathetic reflexes. The inherent cardiac
stability provided by opioids is greatly diminished in practice when other anesthetic drugs,
including benzodiazepines, propofol, or volatile agents are added. For example, sufentanil
and fentanyl can be associated with reduced cardiac output when administered in
combination with benzodiazepines. Bolus doses of meperidine, hydromorphone, and
morphine evoke varying amounts of histamine release that can lead to profound drops in
systemic vascular resistance and arterial blood pressure. The potential hazards of histamine
release can be minimized by infusing opioids slowly or by pretreatment with H1 and H2
antagonists. Histamine side effects can be treated by infusion of intravenous fluid and
vasopressors. Intraoperative hypertension during opioid-based intravenous or nitrous oxide–
opioid anesthesia is common. Such hypertension is often attributed to inadequate anesthetic
depth; thus it is conventionally treated by the addition of other anesthetic agents
(benzodiazepines, propofol, or potent inhaled agents). When depth of anesthesia is adequate
we recommend treating hypertension with antihypertensives rather than with additional
anesthetics.
B. Respiratory
Opioids depress ventilation, particularly respiratory rate. Thus, respiratory rate and end-tidal
CO2 tension (in contrast to arterial oxygen saturation) provide simple metrics for the early
detection of respiratory depression in patients receiving opioid analgesia. Opioids increase
the partial pressure of carbon dioxide (PaCO2) and blunt the response to a CO2 challenge,
resulting in a shift of the CO2 response curve downward and to the right (Figure 10–3).
These effects result from opioid binding to neurons in the respiratory centers of the
brainstem. The apneic threshold—the greatest Paco2 at which a patient remains apneic-
rises, and hypoxic drive is decreased. Morphine and meperidine can cause histamine-
induced bronchospasm in susceptible patients. Rapid administration of larger doses of
opioids (particularly fentanyl, sufentanil, remifentanil, and alfentanil) can induce chest
wall rigidity severe enough to make ventilation with bag and mask nearly impossible.
This centrally mediated muscle contraction is effectively treated with neuromuscular
blocking agents. This problem is rarely seen now that large-dose opioid anesthesia is no
longer a mainstay of cardiovascular anesthesia practice. Opioids can blunt the
bronchoconstrictive response to airway stimulation such as occurs during tracheal intubation.
C. Cerebral
The effects of opioids on cerebral perfusion and intracranial pressure must be separated from
any effects of opioids on PaCO2. In general, opioids reduce cerebral oxygen consumption,
cerebral blood flow, cerebral blood volume, and intracranial pressure, but to a much lesser
extent than propofol, benzodiazepines, or barbiturates, provided normocarbia is maintained
by artificial ventilation. There are some reports of mild—but transient and almost certainly
unimportant increases in cerebral artery blood flow velocity and intracranial pressure
following opioid boluses in patients with brain tumors or head trauma. If combined with
hypotension, the resulting fall in cerebral perfusion pressure could be deleterious to patients
with abnormal intracranial pressure–volume relationships. Nevertheless, the important
clinical message is that any trivial opioid-induced increase in intracranial pressure would be
much less important than the predictably large increases in intracranial pressure associated
with intubation of an inadequately anesthetized patient (from whom opioids were with held).
Opioids usually have almost no effects on the electroencephalogram (EEG), although large
doses are associated with slow δ-wave activity. There are case reports that large doses of
fentanyl may rarely cause seizure-like activity; however, some of these apparent seizures
have been retrospectively diagnosed as severe opioid-induced muscle rigidity. EEG
activation and seizures have been associated with the meperidine metabolite normeperidine,
as previously noted. Stimulation of the medullary chemoreceptor trigger zone is responsible
for opioid-induced nausea and vomiting. Curiously, nausea and vomiting are more common
following smaller (analgesic) than very large (anesthetic) doses of opioids. Repeated dosing
of opioids (eg, prolonged oral dosing) will reliably produce tolerance, a phenomenon in
which progressively larger doses are required to produce the same response. This is not the
same as physical dependence or addiction, which may also be associated with repeated opioid
administration. Prolonged dosing of opioids can also produce “opioid-induced hyperalgesia,”
in which patients become more sensitive to painful stimuli. Infusion of large doses of (in
particular) remifentanil during general anesthesia can produce acute tolerance, in which much
larger than usual doses of opioidswill be required for immediate, postoperative analgesia.
Relatively large doses of opioids are required to render patients unconscious (Table 10–3).
However, even at very large doses opioids will not reliably produce amnesia. The use of
opioids in epidural and intrathecal spaces has revolutionized acute and chronic pain
management (see Chapters 47 and 48).
TABLE 10–3 Uses and doses of common opioids.
Unique among the commonly used opioids, meperidine has minor local anesthetic qualities,
particularly when administered into the subarachnoid space. Meperidine’s clinical use as a
local anesthetic has been limited by its relatively low potency and propensity to cause typical
opioid side effects (nausea, sedation, and pruritus) at the doses required to induce local
anesthesia. Intravenous meperidine (10–25 mg) is more effective than morphine or
fentanyl for decreasing shivering in the postanesthetic care unit and meperidine
appears to be the best agent for this indication.
D. Gastrointestinal
Opioids slow gastrointestinal motility by binding to opioid receptors in the gut and reducing
peristalsis. Biliary colic may result from opioid-induced contraction of the sphincter of Oddi.
Biliary spasm, which can mimic a common bile duct stone on cholangiography, is reversed
with the opioid antagonist naloxone or glucagon. Patients receiving long-term opioid therapy
(eg, for cancer pain) usually become tolerant to many of the side effects but rarely to
constipation. This is the basis for the development of the peripheral opioid antagonists
methylnaltrexone, alvimopan, naloxegol, and naldemedine, which promote gastrointestinal
motility in patients with varying indications, such as treatment of opioid bowel syndrome,
side effects from opioid treatment of noncancer pain, or reduction of ileus in those receiving
intravenous opioids after abdominal surgery.
E. Endocrine
The neuroendocrine stress response to surgery is measured in terms of the secretion of
specific hormones, including catecholamines, antidiuretic hormone, and cortisol. Large doses
of fentanyl or sufentanil inhibit the release of these hormones in response to surgery more
completely than volatile anesthetics. The actual clinical outcome benefit produced by
attenuating the stress response with opioids, even in high-risk cardiac patients, remains
speculative (and we suspect nonexistent), whereas the many drawbacks of excessive doses of
opioids are readily apparent.
Other Effects
A. Cancer Reoccurrence
Retrospective studies have associated general anesthesia (including opioids) with an
increased risk of cancer reoccurrence after surgery as compared to techniques that emphasize
opioid-sparing regional anesthetic techniques for analgesia. Ongoing clinical trials will likely
clarify whether general anesthesia, opioids, both, or neither influence outcomes after cancer
surgery.
B. Substance Abuse
There is a well-publicized epidemic of opioid abuse in western democracies, particularly in
the United States. Although it comprises less than 5% of the world’s population, the United
States consumes 80% of the world’s prescription opioids (and nearly all of the world’s supply
of hydrocodone)! Large numbers of patients admit to using prescribed opioids in a
recreational fashion, and drug overdosage (most often from prescribed drugs) is the leading
cause of accidental death in the United States. Many opioid addicts can trace their addiction
to opioids prescribed by a physician for acute or chronic pain. There are many causes of this
terrible problem, including excessive and misleading marketing of opioids to physicians,
unwise prescribing practices by physicians, inappropriate and misleading assertions by
“thought leaders” (many with ties to the pharmaceutical industry) regarding opioids, and
well-intended but poorly thought out recommendations for assessment and treatment of pain
by certifying agencies. In response the U.S. Centers for Disease Control and Prevention and
many other agencies have released guidelines for responsible prescribing of opioids.
Drug Interactions
The combination of meperidine and monoamine oxidase inhibitors may result in
hemodynamic instability, hyperpyrexia, coma, respiratory arrest, or death. The cause of this
catastrophic interaction is incompletely understood. (The failure to appreciate this drug
interaction in the controversial Libby Zion case led to changes in work hour rules for house
officers in the United States.) Propofol, barbiturates, benzodiazepines, inhaled anesthetics,
and other central nervous system depressants can have synergistic cardiovascular, respiratory,
and sedative effects with opioids. The clearance of alfentanil may be impaired and the
elimination half-life prolonged following treatment with erythromycin.