Chapter 45. Pharmacology of Local Anesthetics^*

1. Pharmacodynamics—Local anesthetics stop the propagation of action potentials in nerve axons by preventing the influx of sodium through voltage-gated sodium channels in the axon membrane. Other actions of local anesthetics (eg, effects on other voltage-gated ion channels and ligand-gated ion channels) may be important for their analgesic action and/or for undesired side effects.

2. Chemistry—Local anesthetics are weak bases with 3 structural parts: a hydrophilic end and a lipophilic end linked by an amino ester or an amino amide bond. The bond is the basis for classifying local anesthetics into 2 groups: the amino esters and the amino amides. Optical isomers of local anesthetics with an asymmetric carbon atom usually differ in potency, duration of action, and toxicity.

3. Expression of local anesthetic action—In vitro, there generally is a positive correlation between molecular weight of local anesthetic molecules and lipophilicity, protein binding, duration of action, potency, and toxicity; there is an inverse relation with speed of onset. In vivo expression of local anesthetic action is dependent on other factors as well, such as injection site, dose, intrinsic vasoactivity, and formulation. The manifestation of sensory versus motor block varies and is dependent on many factors, including the agent and type of block performed.

4. Pharmacokinetics—Local anesthetics usually are injected near the target site instead of relying on systemic circulation to carry them there (except, eg, intravenous regional anesthesia and treatment of certain neuropathic pain states). Barriers to the diffusion of local anesthetics to their target vary and are dependent on injection site (eg, epidural vs intrathecal), influence dose, speed of onset, and duration of action. Local anesthetics that reach systemic circulation are widely distributed in the body. Hydrolysis of the amino ester bond by esterases in blood is the primary biotransformation process for amino ester-linked local anesthetics. Hepatic extraction and biotransformation are important elimination and biotransformation pathways for amino amide-linked local anesthetics.

5. Toxicity—Local anesthetic toxicity can be categorized as allergic, tissue, cardiovascular, central nervous system, and methemoglobinemia. Clinically, manifestations of local anesthetic systemic toxicity do not occur in a predictable order, and which occur and how soon after local administration are variable. Concerns about cardiovascular toxicity related to bupivacaine have driven a search for long-acting local anesthetics with less cardiotoxic action.

6. Formulation—Substances are sometimes added to local anesthetic formulations to preserve the molecules, to prevent microbial growth, to prevent systemic absorption, to enhance and/or prolong local anesthetic action, and to enhance spread of the local anesthetic.

Local anesthetics are widely used to prevent or treat acute pain; to treat inflammatory, cancer, and chronic pain; and for diagnostic and prognostic purposes. Drugs classified as local anesthetics reversibly block action potential propagation in axons by preventing the sodium entry that produces the potentials.1 However, other actions of these drugs, such as anti-inflammation by interaction with G-protein receptors2 and analgesic effect in the spinal cord by blocking postsynaptic ionotropic receptor function mediated by extracellular receptor-activated kinase,3 also are thought to be relevant to their use to prevent or treat pain. Nociceptive pain, as well as neuropathic pain, is targeted with this group of drugs. Any part of the nervous system, from the periphery to the brain, may be where local anesthetics act to produce a desired anesthetic or analgesic effect. A variety of formulations of local anesthetics, routes of administration, and methods of administration are used. The drugs are formulated commercially or by medical personnel according to the intended route of administration and/or to address specific concerns or needs. This chapter provides a concise review of the pharmacology of local anesthetics.

*This chapter includes substantial material from Covino BG. Pharmacology of local anesthetic agents. In: Rogers MC, Tinker JH, Covino BG, et al, eds. Principles and Practice of Anesthesiology, Vol 2. St. Louis, MO: Mosby Year Book; 1993:1235-1257.

Cocaine, Procaine, and Lidocaine

Koller is credited with introducing local anesthetics into medical practice when he used cocaine to numb the cornea before operating on the eye.4 Fundamental to the development of synthetic local anesthetics was isolation of cocaine by Neimann, in 1860, from coca beans, and elucidation of its chemical structure. Procaine was first synthesized in 1904, and lidocaine was first synthesized in 1943. Synthesis of molecules with local anesthetic activity paved the way for "tinkering" with the molecules by systematically modifying chemical structure and testing for a desired result, for example reducing toxicity, in developing new local anesthetics.

Roots of Modern Use

Figure 45-1 presents a chronology of the introduction of local anesthetics into clinical practice. Four amino ester–linked local anesthetics (see Three Parts of Local Anesthetic Molecules) appear in the figure: cocaine, procaine, tetracaine, and chloroprocaine. The other local anesthetics are amino amide linked. What is evident from the figure is the focus since 1955 on the development of amino amide– and not amino ester–linked local anesthetics. Reasons for this include the allergenic potential of amino ester–linked local anesthetics and the instability of amino ester bonds.

Three Parts of Local Anesthetic Molecules

All local anesthetic molecules in clinical use have 3 parts: lipophilic (aromatic) end, hydrophilic (amine) end, and a link between the ends (Fig. 45-2). The link contains either an amino ester or an amino amide bond, and local anesthetics are designated as belonging to 1 of 2 groups: the amino ester–linked local anesthetics or amino amide–linked local anesthetics. Procaine is the prototypic amino ester–linked local anesthetic, and lidocaine is the prototypic amino amide–linked local anesthetic (Fig. 45-3).

Amino amides are extremely stable agents, whereas amino esters are relatively unstable in solution. Consequently, aqueous solutions of amino ester agents have relatively short shelf lives as compared with solutions of amino amides and are sensitive to exposure to high temperatures. The amino esters are hydrolyzed in plasma by cholinesterase enzymes, whereas the amino amide compounds undergo enzymatic biotransformation in the liver.

There are 2 varieties of amino amide local anesthetics based on the structure of the amino amide link. One variety is the aminoacyl amides, such as lidocaine and bupivacaine, and the other is the amino alkyl amides, such as dibucaine. The different structures influence the duration of action and biotransformation. Further distinction within the aminoacyl amide class is made based on whether the hydrophilic amino end is a straight carbon chain (eg, lidocaine) or the amino nitrogen is within a ring structure (pipecoloxylidide, eg, mepivacaine)

Chiral Forms

The newest additions to clinically available local anesthetics, ropivacaine (Fig. 45-4) and levobupivacaine, represent (1) the exploitation of technology that permits cost-favorable separation of racemic mixtures of local anesthetics into pure enantiomers and (2) the search for local anesthetics with greater safety margins. Simplistically stated, molecules with an asymmetric carbon atom exist in forms that are mirror images (ie, exhibit "handness, chirality"), with images (enantiomer, stereoisomers) distinguished by how they rotate light according to the orientation of the structures in 3 dimensions. Various terms are used to refer to the different enantiomers; this chapter uses S and R to designate 2 different enantiomers. A racemic mixture contains equal amounts of the R and S isomers. Commercial formulations of ropivacaine and levobupivacaine contain the S enantiomer. Note that levobupivacaine is the S form of bupivacaine. The motive for marketing pure enantiomers is evidence that the S form is less toxic, more potent, and longer acting than the R form or the racemic mixture (Table 45-1).

Lipid Solubility and Protein Binding

Testing modifications to the basic procaine and lidocaine structure revealed that increasing the molecular weight of the molecules by adding carbon atoms to either end of the structure or to the link generally increases lipid solubility, protein binding, and duration of action and toxicity and influences biotransformation of the molecule (Figs. 45-5 and 45-6). There is a positive correlation between intrinsic local anesthetic potency and lipid solubility of local anesthetics.

Cation and Base Forms

Most local anesthetics have a tertiary amine on the hydrophilic end. Exceptions include prilocaine, which has a secondary amine, and benzocaine, which has a primary amine. Tertiary amines have a positive charge (cation) or are uncharged (base). The ratio of cation to base is determined by the pKa of the local anesthetic and the pH of the solution (Table 45-2). The unchanged forms of local anesthetics pass readily through cell membranes and hence speed of onset of local anesthetic block, at least theoretically, is increased by increasing the concentration of uncharged local anesthetic molecules injected.

Because local anesthetics are weak bases, increasing the pH ("alkalinization") of solution increases the ratio of base to cation. The Henderson-Hasselbalch equation can be used to quantitate the ratio:

Sodium bicarbonate is used clinically to increase the pH of local anesthetic solutions.

Important to note is that commercial aqueous solutions of local anesthetics are acidified, so the hydrophilic (cationic) state, which is water soluble, is favored. Overzealous alkalinization can cause local anesthetic molecules to precipitate from solution.

Neurophysiologic and Neuroanatomical Considerations

Reversible block of fast voltage-gated sodium channels in axons is generally thought to be how local anesthetics block sensory and motor function (Fig. 45-7). Some of the evidence supporting this is the following: (1) action potentials do not develop in axons exposed to local anesthetic; (2) sodium currents responsible for generation of action potentiations are blocked by these drugs; and (3) local anesthetics do not affect the transmembrane potential of axons. The "state" of the sodium channel (resting, open, inactivated) changes during the cycle of polarized, depolarized, repolarized. The order of affinity of local anesthetics for different channel states is open > inactivated > resting. Many investigators have shown that the block of propagation of action potentials is a function of frequency of depolarization, which supports the conclusion that the open state of the sodium channel is the primary target of local anesthetic molecules. This is referred to as "state-dependent block."

Voltage-Gated Sodium Channels

Voltage-gated ion channels are ion channels (Na+, K+, Ca2+, etc) whose permeability is a function of transmembrane potential. There are a number of sodium channel subtypes that generally are divided into those that are tetrodotoxin sensitive (TTXs) and tetrodotoxin resistant (TTXr). A standard nomenclature was adopted for voltage-gated Na+ channels that uses a numerical system to define subfamilies and subtypes based on similarities between the amino acid sequences of the channels. Nine mammalian sodium channel isoforms have been identified (NaV1.1 through NaV1.9). NaV1.1, NaV1.2, NaV1.3, and NaV1.7 are broadly expressed in neurons and are highly TTXs; NaV1.5, NaV1.8, and NaV1.9 are highly expressed in heart and dorsal root ganglia neurons.5 Most sensory neurons generate TTXs currents. However, TTXr currents are present in a high proportion of smaller dorsal root ganglion neurons associated with nociceptive Aδ and C fibers. Available evidence indicates that channels from both groups are involved in pain states as a result of changes in channel function and expression caused by disease or injury.

Existing local anesthetics lack specificity for Na+ channel subtypes However, there are ongoing efforts to develop Na+ channel blockers specific for NaV1.7, NaV1.8, and NaV1.9, which are expressed extensively on peripheral nociceptors. NaV1.7 is of particular interest, as this channel determines the ability of a nerve to transmit nociceptive information.6

Arguments have been put forth that local anesthetics might exert their pharmacologic action not only on Na+ conductance, but also on other ionic conductances (eg, K+ and Ca2+).7

Differential Block

Differential block, the block of pain perception without motor block for example, is observed clinically, but the mechanism responsible for this is poorly understood. The clinical manifestations of differential block vary depending on the local anesthetic used.9 Etidocaine has a propensity to produce more profound motor block than sensory block. For many years, differential block was ascribed to smaller axons being more sensitive than large ones to local anesthetics,10 but this "size principle" was challenged.11 Strichartz and Berde9 cite a number of different factors that might contribute to differential block, including anatomical and relative sensitivity to sodium and potassium channels of different local anesthetics. Oda et al12 suggested that preferential block of TTXr sodium channels by ropivacaine in small dorsal root ganglia neurons (associated with nociceptive sensation) underlies differential block observed during epidural anesthesia with this drug.

Using a combination of local anesthetic and another drug to produce nociceptive selective block is under investigation. For example, activating transient receptor potential vanilloid subtype 1 channels on C-fibers with capsaicin to deliver the local anesthetic, QX-314, into the fibers is effective in animal models.13 Another approach combining local anesthetic and α2-adrenergic agonist such as dexmedetomidine has yielded favorable results. The mechanism for the α2-adrenergic agonist action is not known but may include direct inhibition of TTXr Na+ channel or through hyperpolarization activated cation current.14

Pain Relief by Systemic Administration

Another pharmacodynamic puzzle is the mechanism whereby systemically administered local anesthetic relieves pain. Analgesic effect has been reported after intravenous lidocaine administration in many acute and chronic conditions.15-23 Subcutaneously injected bupivacaine reportedly produces analgesia via a systemic effect.24 Normal or altered sodium channels located in various areas of the brain, spinal cord, or dorsal root ganglia, or in peripheral axons, are mentioned most frequently as the action sites. Zhang et al25 reported that in rats systemic lidocaine delivered via implanted osmotic pump reduces sympathetic nerve sprouting in dorsal root ganglion that is associated with some neuropathic pain behaviors. Takatori et al26 presented evidence that inhibition of nerve growth factor (NGF)–stimulated tyrosine kinase activity of TrkA, a high-affinity receptor of NGF, might be involved in the suppression of neurite outgrowth by local anesthetics.

Frolich et al27 suggested that intravenous lidocaine may be effective against deep tissue pain but has limited value for treating superficial pains except neuropathic conditions.

Ligand-Gated Ion Channels

Ligand-gated ion channels are channels whose permeability status depends on the interaction between a ligand and a receptor that influences channel function. Many of these receptors interact with G proteins. Local anesthetics affect a number of biologic processes, including inhibition of G-protein coupled receptor signaling, that are potentially important pharmacodynamic actions of value in treating pain.

Distribution to the Target Site

The usual pharmacokinetics parameters (Table 45-3) presented for local anesthetics incompletely describe important details regarding distribution of these drugs from application sites to target and nontarget structures. It is well established that systemic absorption of local anesthetics correlates positively with the vascularity of the injection site (intravenous > tracheal > intercostal > paracervical > epidural > brachial plexus > sciatic > subcutaneous). The spinal cord meninges influence distribution of local anesthetics from the epidural and subarachnoid spaces. Intact skin is nearly a complete barrier to local anesthetic penetration. In the latter case, special local anesthetic formulations (eg, eutectic mixture of local anesthetics [EMLA] cream, an eutectic mixture of lidocaine and prilocaine) or delivery methods (eg, electrophoresis) are employed to facilitate transcutaneous transfer. The large number of different injection sites used (eg, epidural, intrathecal, intrapleural, intra-articular, intramuscular, perineural, topical) and the variety of dosing methods (eg, single shot, continuous infusion, intermittent infusion) make comprehensive discussion of all pharmacokinetic considerations beyond the scope of this chapter.

Absorption

In addition to the injection site, the vascular absorption of local anesthetic agents is related to dosage, addition of a vasoconstrictor agent, and specific agent employed. The high blood concentrations after intercostal administration are probably related to the multiple injections required for intercostal nerve blocks. As a consequence of this, local anesthetic solution is exposed to a greater vascular area, which results in a greater rate and degree of absorption. After thoracic paravertebral block, ropivacaine demonstrates a biphasic fast and slow absorption pattern.28 The fast phase approximates the speed of intravenous injection and accounts for nearly half of ropivacaine absorption. When epinephrine (5 μg/mL) is added, systemic absorption and peak plasma concentration of ropivacaine are significantly reduced.

For most local anesthetic agents, a linear relationship exists between the amount of drug administered and the resultant peak venous plasma concentration. The mean venous plasma concentration of lidocaine increases from approximately 1.5 to 4 μg/mL as the total dose administered into the lumbar epidural space is increased from 200 to 600 mg. Depending on the site of administration, a peak blood concentration of 0.5 to 2.0 μg/mL is achieved for each 100 mg of lidocaine or mepivacaine injected. Simon et al29 reported that systemic absorption of ropivacaine after epidural administration was biphasic, with higher absorption kinetics in younger than in older patients. The absorption kinetics were in the same range as for other long-acting local anesthetics. Buffington and Blix30 recently suggested that the early peak seen after epidural injection of local anesthetic might be due in part to bulk transfer from the epidural space to the bloodstream.

In general, the addition of epinephrine to local anesthetic solutions decreases the rate of vascular absorption of these agents.31 Epinephrine 5 μg/mL (1:200 000) significantly reduces the peak blood concentrations of lidocaine and mepivacaine, regardless of the site of administration. However, peak blood concentrations of bupivacaine and etidocaine are minimally influenced by the addition of epinephrine after injection into the lumbar epidural space. On the other hand, the rate of vascular absorption of these agents is significantly decreased when epinephrine-containing solutions are employed for brachial plexus blockade.32,33

The rate of vascular absorption also varies and is dependent on the specific local anesthetic agent. Lidocaine is absorbed more rapidly after brachial plexus and epidural blockade than is prilocaine, whereas bupivacaine is absorbed more rapidly than etidocaine.31,33 Prilocaine is a less potent vasodilator than lidocaine, which partly accounts for lower blood concentrations of prilocaine. The lower peak blood concentrations of etidocaine compared with bupivacaine may be related to the greater lipid solubility of etidocaine, which results in its sequestration by adipose tissue and a decreased rate of absorption. The differences in absorption rates are of practical clinical significance, as they permit the use of larger doses of prilocaine compared with lidocaine and of etidocaine compared with bupivacaine.

Systemic Distribution

The distribution of local anesthetic agents from systemic circulation can be described by a 2- or 3-compartment model.34 The rapid disappearance (α) phase is believed to be related to uptake by rapidly equilibrating tissues, that is, tissues with a high vascular perfusion. The slower phase of disappearance from blood (β phase) is mainly a function of distribution to slowly equilibrating tissues and the biotransformation and excretion of the compound (Fig. 45-8). The α half-life (T½α) of prilocaine is shorter than that of lidocaine and mepivacaine, which indicates that prilocaine is redistributed at a significantly more rapid rate from blood to tissues than either of the other 2 drugs (Table 45-3). The T½α of lidocaine and mepivacaine are similar. The half-life of the β disappearance phase (T½β) of prilocaine is more rapid than that of lidocaine and mepivacaine, suggesting a more rapid rate of biotransformation. A comparison study reveals that etidocaine has a more rapid rate of tissue redistribution and biotransformation than bupivacaine.35 The clearance of ropivacaine is closer to that of mepivacaine and faster than that of bupivacaine, and the terminal t1/2 of mepivacaine and ropivacaine are less than that of bupivacaine.36

Local anesthetic agents are distributed throughout all body tissues, but the relative concentration in different tissues varies as a function of time, vascular perfusion, and tissue mass. Initially, local anesthetic agents are rapidly extracted by lung tissue so that the whole blood concentration of local anesthetics decreases greatly as they pass through the pulmonary vasculature.37,38 Ultimately, the highest percentage of an injected dose of a local anesthetic agent is found in skeletal muscle, simply because the mass of skeletal muscle makes it the largest reservoir for local anesthetic agents.

Biotransformation and Elimination

The degradation of local anesthetic agents varies according to their chemical classification. Amino ester local anesthetic drugs are hydrolyzed in plasma by cholinesterase enzymes. Chloroprocaine shows the most rapid rate of hydrolysis (4.7 μmol/mL/h), compared with a rate of 1.1 μmol/mL/h for procaine and 0.3 μmol/mL/h for tetracaine.39 Less than 2% of unchanged procaine is found in urine, whereas approximately 90% of p-aminobenzoic acid, which is a primary product of procaine hydrolysis, appears in urine. Only 33% of dimethylaminoethanol, the other hydrolysis product of procaine, is excreted unchanged.

The amino amide agents are biotransformed primarily in the liver.40 Elimination depends both on hepatic blood flow and enzyme activity.36 Prilocaine undergoes the most rapid rate of hepatic metabolism, and lidocaine is biotransformed somewhat more rapidly than mepivacaine. In humans, the hepatic clearance of etidocaine is greater than that of bupivacaine, which suggests a more rapid rate of hepatic biotransformation for etidocaine. Evidence also exists that prilocaine may be biotransformed in the kidney, which would explain the rapid clearance of this agent compared with all other amino amides.41

The biotransformation of the amino amide–type agents results in the formation of a variety of metabolites. The biotransformation of lidocaine has been studied most extensively. The main pathway of lidocaine biotransformation in humans involves oxidative de-ethylation of lidocaine to monoethylglycinexylidide by cytochrome P450 IIIA4, followed by a subsequent hydrolysis of monoethylglycinexylidide to xylidine.

The excretion of the amino amide–type local anesthetic drugs occurs by way of the kidney. Less than 5% of the unchanged drug is excreted via the kidney into the urine. The major portion of the injected agent appears in the urine in the form of various metabolites. The renal clearance of the amino amide local anesthetic agents appears to be inversely related to their protein-binding capacity. Prilocaine, which has a lower protein-binding capacity than lidocaine, has a substantially higher clearance value than lidocaine.42 Renal clearance also is inversely proportional to the pH of urine, suggesting that urinary excretion of these agents occurs by nonionic diffusion.

Patient age may influence the physiologic disposition of local anesthetics. The half-life of lidocaine after IV administration increased from an average of 80 minutes in human volunteers 22 to 26 years of age to 138 minutes in subjects ages 61 to 71 years.43 Age-dependent disposition kinetics have been reported for ropivacaine after epidural administration, with elimination half-life significantly longer and clearance significantly decreased in older (>61 years) than in younger patients (18-40 years).29

The elimination of local anesthetic agents also is influenced by the individual patient's hepatic and cardiac function. For example, an average lidocaine half-life of 1.5 hours was reported in volunteers with normal hepatic function, whereas patients with liver disease demonstrated an average half-life of 5 hours.44 In patients with congestive heart failure (CHF), the IV infusion of lidocaine results in significantly higher plasma concentrations of lidocaine than occurs in patients with normal cardiovascular function, indicating a decreased plasma clearance in patients with CHF.45

Routes of Administration

The clinical profile of local anesthetic drugs varies and is dependent on the type of regional anesthetic technique performed (Fig. 45-9). The shortest duration of action occurs after the intrathecal or subcutaneous administration of local anesthetics. The longest latencies and durations are observed after major peripheral nerve blocks, such as brachial plexus blockade. Intrathecal bupivacaine usually produces anesthesia within 5 minutes and persists for 3 to 4 hours.46 For brachial plexus blockade, the onset time of bupivacaine is approximately 20 to 30 minutes, whereas the duration of anesthesia averages approximately 10 hours.47 These variations in the onset and duration of anesthesia result from differences in the anatomy of the injection sites and the amount of drug employed for various types of regional anesthesia. In the subarachnoid space, the lack of a nerve sheath around the spinal cord and the placement of the local anesthetic solution in the immediate vicinity of the spinal cord is responsible for the rapid onset of action. The relatively small amount of drug employed for spinal anesthesia probably accounts for the short duration of conduction block. For example, 50 to 75 mg of lidocaine and 10 to 15 mg of bupivacaine are usually employed for most spinal anesthetic procedures.

The onset of brachial plexus blockade is slow, as the anesthetic agent is usually injected some distance from the nerve roots and must diffuse through various tissue barriers before reaching the nerve membrane. The prolonged blockade is probably related to the decreased rate of vascular absorption from that site and the larger dose of drug employed for this regional anesthetic technique. For example, 400 to 600 mg of lidocaine and 100 to 150 mg of bupivacaine are usually employed for brachial plexus blockade. Techniques (eg, ultrasound guided) that allow more precise placement of local anesthetic near nerves theoretically will produce more rapid onset of block and clearly reduce volume required to produce a block. Using ultrasound for needle placement, Latzle et al48 found that when the ED99 volumes of 1.5% mepivacaine were used for sciatic nerve blocks, local anesthetic volume had no impact on sensory onset times, but sensory block duration was shortened.

Clinical Manifestations of Potency

In vitro studies show a general positive correlation between lipid solubility of local anesthetics and potency (Fig. 45-10).49,50 This presumably is because the nerve membrane that represents the site of action of local anesthetics consists primarily of lipids. However, the hydrophobicity and anesthetic potency are not as well correlated in intact animals and humans. The EC50 (median effective dose) of lidocaine, for example, is significantly less than that of mepivacaine in an isolated nerve, but in humans, little difference in anesthetic potency is apparent between these agents.31

The difference between potency in an isolated nerve and in a clinical situation is probably a function of the vasodilator or tissue redistribution properties of the various local anesthetics. Lidocaine is a more profound vasodilator than mepivacaine in humans. More rapid vascular absorption of lidocaine as a result of vasodilatation causing increased blood flow results in fewer lidocaine molecules being available for neural blockade. Hence lidocaine and mepivacaine appear to possess similar anesthetic potencies clinically. The profound conduction-blocking activity of etidocaine in vitro is a result of its high lipid solubility. However, sequestration of etidocaine in adipose tissue in vitro, such as in the epidural space, results in fewer etidocaine molecules being available for neural blockade compared with bupivacaine.

Speed of Onset

In isolated nerves,50,51 onset time of conduction block is correlated with the pKa of the various agents (Fig. 45-11), and lipid solubility influences onset of action, too. The pKa of local anesthetics and the pH of the anesthetic solution determine the degree of ionization of specific agents. This is because the degree of ionization influences the rate of diffusion of local anesthetics across the nerve sheath and membrane. The uncharged base form diffuses more easily than the charged cationic form.52,53

The pKa of all local anesthetics, except benzocaine, varies from 7.6 to 9.1 and is greater than the physiologic pH (7.4) of tissue (Table 45-4). At a pH of 7.4, approximately 2% of procaine (pKa of 9.1) exists in the base form. In contrast, 35% of mepivacaine, lidocaine, etidocaine, and prilocaine, which possess a pKa of approximately 7.6, are present in the un-ionized forms. The pH of most local anesthetic solutions is approximately 5 to 6 because of the relatively low solubility of local anesthetics in solutions of higher pH. The inclusion of epinephrine in local anesthetic solutions results in a further decrease in pH to approximately 3 to 4.

Studies in vitro clearly demonstrate the relationship among the pKa of specific agents, the pH of anesthetic solution, and the onset of conduction block.51-53 Agents with a relatively low pKa, such as lidocaine and mepivacaine, show the most rapid onset of conduction block. Procaine, chloroprocaine, and tetracaine, which possess high pKa values, demonstrate a longer latency. The onset of conduction blockade is also reduced by increasing the pH of the bathing solution.52 Clinically, onset time may be influenced by the dose or concentration of local anesthetic employed.54 For example, increasing the concentration of bupivacaine from 0.25% to 0.75% results in a significant acceleration in anesthetic effect. Despite its pKa being approximately 9 and its onset of action in isolated nerves being relatively slow, chloroprocaine demonstrates a rapid onset of action in humans. This is because chloroprocaine's low systemic toxicity allows the use of high concentrations such as 3%. Lidocaine has been shown in some studies to produce a more rapid onset of epidural anesthesia than chloroprocaine when employed at the same concentrations. However, use of a 3% chloroprocaine solution results in a more rapid onset of conduction block compared with 2.0% lidocaine.55

Duration of Action

The various local anesthetics have greatly different durations of action, and protein binding apparently is a determinant of duration (Fig. 45-12).51 Procaine and chloroprocaine have a short duration of action; lidocaine, mepivacaine, and prilocaine have a moderate duration of anesthesia. Tetracaine, bupivacaine, levobupivacaine, etidocaine, and ropivacaine are associated with the longest durations of anesthesia. Procaine produces duration of brachial plexus blockade of 30 to 60 minutes—approximately 10 hours of anesthesia has been reported after the use of bupivacaine, etidocaine, and ropivacaine for brachial plexus blockade.

Local anesthetics are believed to act by binding to a protein receptor in the sodium channel,4 and thus the greater protein binding of a specific agent presumably results in a longer period of sodium channel blockade and a longer duration of anesthesia. Plasma proteins have been used to measure protein binding of local anesthetics, assuming that binding to membrane proteins is similar to plasma protein binding.

Duration of action of the various agents is influenced by their peripheral vascular effects. Local anesthetics except cocaine generally tend to have a biphasic effect on vascular smooth muscle.56-58 These agents cause vasoconstriction at low concentrations and vasodilatation at clinically employed concentrations. The protein kinase C Rho and p44/42 mitogen-activated protein kinase signaling pathways in calcium sensitization mechanisms may be involved in the biphasic action.59 Choi et al60 reported that levobupivacaine-induced contraction of rat aorta smooth muscle is mediated mainly by activation of the lipoxygenase pathway and in part by the activation of the cyclooxygenase pathway. Lipoxygenase pathway activation by levobupivacaine apparently facilitates calcium influx via L-type channels and endothelial nitric oxide attenuates levobupivacaine-induced contractions.

Differences exist in the vasodilator activity of the various drugs. Lidocaine is a more potent vasodilator than mepivacaine or prilocaine. In vivo, the duration of anesthesia produced by lidocaine is shorter than that of mepivacaine or prilocaine. If epinephrine is added to solutions of these agents, the duration of action for all 3 drugs is similar. The duration of action of ropivacaine is similar to that of bupivacaine when the agents are administered for peripheral or central neural blocks.61,62 However, ropivacaine produces a significantly longer duration of infiltration anesthesia than bupivacaine in animals, which appears related to ropivacaine's cutaneous vasoconstrictor effects (Fig. 45-13).58,61

Differential Sensory/Motor Block

Differential block is observed clinically but is poorly understood. Local anesthetic agents differ with respect to producing motor versus sensory block. The most significant separation between sensory anesthesia and motor blockades is produced by bupivacaine.63 Depending on the concentration of bupivacaine employed, significant sensory anesthesia may be obtained with little inhibition of motor activity. Ropivacaine appears to provide a similar degree of sensory/motor separation.64

Bupivacaine and etidocaine are notably different in terms of their differential sensor/motor fiber–blocking activity (Fig. 45-14). Bupivacaine is widely used epidurally for obstetric procedures and relief of pain postoperatively because of its ability to provide adequate analgesia with minimal muscle weakness or paralysis, particularly as a 0.125% or 0.25% solution. In contrast, etidocaine shows little separation between blockade of sensory and motor fibers and thus achieved limited popularity.65

Influence of Dose

An increase in the dose of local anesthetic usually results in a more profound depth of block, a prolongation of satisfactory anesthesia, and a decrease in the onset of block.63,64 No clinically significant differences in onset, depth, and duration of anesthesia appear to exist when the same dose of local anesthetic is administered as a large volume of dilute solution or as a small volume of concentrated solution. A comparison of 15 mL of 2% prilocaine with 10 mL of 3% prilocaine (300 mg) and 30 mL of 2% prilocaine with 20 mL of 3% prilocaine (600 mg) for epidural blockade showed no difference in onset, adequacy, or duration of anesthesia and onset, depth, and duration of motor blockade if the dose was maintained constant (Fig. 45-15).66 However, the increase in total dose from 300 to 600 mg of prilocaine did result in a more rapid onset, a greater degree of satisfactory anesthesia, and a prolonged duration of anesthesia.

Similar studies of spinal anesthesia indicate little difference between the use of varying volumes of 0.5% and 0.75% bupivacaine if the total dose was constant (Fig. 45-16).46 The volume of anesthetic solution may influence the spread of anesthetic in the epidural or subarachnoid space. For example, the level of anesthesia was 4.3 dermatomes higher when 30 mL of 1% lidocaine was injected epidurally compared with 10 mL of 3% lidocaine.67 In general, increased dosage is usually achieved by the use of more concentrated anesthetic solutions, as the volume of solution that can be administered is usually restricted by the anatomy of the specific injection site.

More recently, Latzke et al48 used ultrasound-guided sciatic nerve block with 1.5% mepivacaine to examine the effect of volume on sensory block onset and duration of sensory block. As compared with larger volumes, the ED99 volume had no effect on onset but shortened block duration.

Purpose of Different Formulations

Local anesthetics are mixed with other substances during the manufacturing process or just prior to administration. Some objectives of this practice are to shorten onset time, limit absorption, increase intensity of action, stabilize local anesthetic molecule, and inhibit microbial growth. A 2004 issue of Techniques in Regional Anesthesia and Pain Medicine discussed in detail additives to local anesthetics.68

Mixtures of Local Anesthetics

Mixtures of local anesthetics have been employed by adding an agent, such as chloroprocaine or lidocaine, which have a relatively rapid onset, to a solution of bupivacaine, which has a slow but long duration of action. A mixture of 3% chloroprocaine and 0.5% bupivacaine was reported to produce a rapid onset and prolonged duration of brachial plexus blockade.69 Subsequent epidural studies indicated that a mixture of chloroprocaine and bupivacaine resulted in a slower onset than chloroprocaine alone and a shorter duration than bupivacaine alone.70 Isolated nerve studies demonstrate that the addition of a metabolite of chloroprocaine to bupivacaine significantly decreased the duration of conduction block compared with bupivacaine alone (Fig. 45-17).71 To date, no similar antagonism has been demonstrated when lidocaine or mepivacaine is mixed with bupivacaine. One should remember that in terms of the potential systemic toxicity of mixtures of local anesthetics, the toxicity of local anesthetics is additive if an unintentional IV injection occurs.

Use of Vasoconstrictors

Epinephrine (1:200 000; 5 μg/mL) is frequently added to local anesthetic solutions to decrease the rate of vascular absorption.31 This allows more anesthetic molecules to diffuse to the nerve membrane and thus improves the depth and duration of anesthesia. Epinephrine 1:200 000 has been reported to be the optimal concentration for prolonging the duration of anesthesia of lidocaine for epidural or intercostals use. Other vasoconstrictor agents, such as norepinephrine and phenylephrine, have been used but do not appear to be superior to epinephrine. For example, equipotent concentrations of epinephrine and phenylephrine similarly prolong the duration of spinal anesthesia produced by tetracaine (Fig. 45-18).72

Epinephrine's ability to prolong the duration of anesthesia depends on the local anesthetic employed and the injection site. Epinephrine significantly extends the duration of both infiltration anesthesia and peripheral nerve blocks with most agents.73,74 The duration of surgical epidural anesthesia, however, is not greatly prolonged when epinephrine is combined with prilocaine, bupivacaine, or etidocaine, but does result in a significant increase in the duration of epidural blockade produced by agents such as lidocaine.75,76 Diluted solutions of bupivacaine employed for epidural analgesia in obstetric patients may be improved with the addition of epinephrine. The depth and duration of epidural analgesia in obstetric patients were improved when epinephrine 1:300 000 was added to 0.25% bupivacaine.77 Although the addition of epinephrine may not significantly prolong the duration of surgical epidural anesthesia produced by 0.5% to 0.75% bupivacaine, a more profound degree of motor blockade has been observed after the use of bupivacaine with epinephrine.75

Epinephrine's ability to prolong spinal anesthesia may be partly related to a direct antinociceptive effect in the spinal cord. Adrenergic receptor agents exert a direct antinociceptive action in the spinal cord. For example, clonidine, an α2-adrenergic agonist, produces analgesia after epidural administration.78

Epinephrine's effect on the onset of regional anesthesia is controversial. Some studies demonstrate a more rapid onset of conduction blockade, presumably because of the decreased vascular absorption of local anesthetics.73 Others fail to demonstrate any difference in onset time.79 This is attributed to the low pH of epinephrine-containing local anesthetic solutions, which favor the formation of the local anesthetics cationic form, which does not easily diffuse through nerve sheaths.

Carbonation and pH Adjustment of Local Anesthetics

Carbonation of local anesthetics results in a more rapid onset and a more profound degree of conduction blockade in an isolated nerve preparation.52,80,81 The enhanced onset is believed to be related to a more rapid dissociation of the local anesthetic to the base form, a direct depressant effect of carbon dioxide (CO2) on the nerve membrane, and a decrease in the axoplasmic pH from the diffusion of CO2 intraneurally. The lower axoplasmic pH increases the rate of formation of the local anesthetic's active cationic form within the nerve. Although CO2 clearly enhances the onset of local anesthetic-induced conduction block in an isolated nerve, discrepancies exist among clinical studies in which hydrochloride and carbonated local anesthetic solutions have been compared. Several investigations have demonstrated a significant decrease in the onset of epidural blockade with lidocaine carbonate compared with lidocaine hydrochloride.82,83 However, other researchers have failed to observe a significant reduction in onset of epidural anesthesia with lidocaine carbonate.84 Similarly, differing results concerning onset time have been reported when bupivacaine hydrochloride and bupivacaine carbonate were employed for brachial plexus blocks.85,86 Although controversy may exist regarding the effect of carbonation on onset of anesthesia, most studies show that carbonated solutions do improve the depth of sensory and motor blockade when administered into the epidural cavity. Also, these solutions produce a more complete blockade of the various nerve roots when employed for brachial plexus blockade.

Alkalinization of local anesthetic solutions by the addition of solution bicarbonate has also been reported to decrease the onset of conduction blockade (Fig. 45-19).87,88 An increase in the pH of the local anesthetic solution increases the amount of drug in the uncharged base form, which should enhance the rate of diffusion across the nerve sheath and nerve membrane. The use of pH-adjusted solutions of bupivacaine or lidocaine has been reported to significantly decrease the latency of brachial plexus and epidural blockade.87,88 Again, controversy surrounds the improved onset of pH-adjusted local anesthetic solutions. Some investigators have failed to demonstrate an improved onset of brachial plexus or epidural blockade with the use of pH-adjusted solutions of bupivacaine or lidocaine.89,90 The differences between these studies might be related to the magnitude of the pH change produced by the addition of bicarbonate. In the original study in which onset time for brachial plexus blockade was significantly reduced, the pH of bupivacaine with epinephrine was increased from 3.9 to 6.4.88 In the subsequent study in which no effect on latency was observed, the pH of plain bupivacaine was increased from 5.5 to 7.0.89 The amount of bicarbonate added to local anesthetic solutions depends on the particular local anesthetic. One milliliter of sodium bicarbonate is added to 10 mL of lidocaine, whereas 0.1 mL of bicarbonate is added to every 10 mL of bupivacaine. The latter agent is insoluble at a high pH, and the addition of excess bicarbonate will result in a precipitation of bupivacaine.

Hyaluronidase, Liposomes, Microencapsulation

Hyaluronidase (tissue spreading factor) is sometimes added to local anesthetic solutions to facilitate spread of solution at the injection site, thereby affecting speed of onset and extent of a block. This seems only to be useful when local anesthetic is injected in the orbit preparatory to ophthalmologic surgery. Hyaluronidase may be injected with local anesthetic during epidural neurolysis to treat pain with positive benefit. Various attempts have been made to prolong the duration of action of local anesthetics by loading them into liposomes or microcapsules, but no such formulations have been approved by the US Food and Drug Administration (FDA) for marketing.

Table 45-5 shows categories of the toxic effects of local anesthetics.

Allergic Reactions

True allergic reactions are associated with amino ester–linked local anesthetics, not amino amide–linked ones. In a study of anaphylactic and anaphylactoid reactions (n = 789) occurring during anesthesia, Mertes et al91 found no such reactions to local anesthetics. However, Mackley et al92 reported that of 183 patients who were patch tested, 4 had positive reactions to lidocaine, 2 of whom had histories of sensitivity to local injections of lidocaine manifested by dermatitis. They concluded that contact-type IV sensitivity to lidocaine might occur more frequently than previously thought. It is common, but inappropriate, to refer to all adverse events as "allergic reactions."

Tissue Toxicity

Tissue toxicity, primarily myotoxicity and neurotoxicity, can be produced by all local anesthetics if "high" concentrations are used. Signs and symptoms of varying degrees of neuropathy (eg, transient neurologic symptoms, cauda equina syndrome) have been reported after spinal anesthesia with, for example, 2% and 5% lidocaine. In a systematic review, Zaric et al93 compared the frequency of transient neurologic symptoms and neurologic complications after spinal anesthesia with lidocaine with that after other local anesthetics. The results showed that the risk for developing transient neurologic symptoms after spinal anesthesia with lidocaine was higher with lidocaine than with bupivacaine, prilocaine, procaine, or mepivacaine.

Systemic Toxicity

Methemoglobinemia

A variety of local anesthetics reportedly may produce methemoglobinemia. Prilocaine is the local anesthetic for which there appears to be greatest risk for this to occur. A dose–response relationship exists between the amount of prilocaine administered epidurally and the degree of methemoglobinemia. In general, doses of prilocaine of 600 mg are required for the development of clinically significant levels of methemoglobinemia.94 The formation of methemoglobinemia is believed to be related to prilocaine's chemical structure. This agent lacks a methyl group in the benzene ring. The metabolism of prilocaine in the liver results in the formation of O-toluidine, which is responsible for the oxidation of hemoglobin to methemoglobin.95 The methemoglobinemia associated with prilocaine is spontaneously reversible or may be treated by IV methylene blue.

Cardiovascular and Central Nervous System

As concentration of local anesthetic in systemic circulation increases, various cardiovascular system and central nervous system (CNS) signs and symptoms appear (Fig. 45-20). The relative CNS and cardiovascular toxicity of local anesthetics has been of interest, especially after Albright96 reported unexpected cardiovascular toxicity of bupivacaine. Most animal studies show that the ratio of doses of bupivacaine that produced convulsive activity and cardiovascular collapse are lower than for other local anesthetics.97 Human volunteer studies of doses required to produce early features of CNS and cardiovascular system toxicity by ropivacaine and levobupivacaine demonstrated the doses were about equal and higher than for bupivacaine.98-100

Brown et al101 reviewed records of patients who had seizures while undergoing brachial plexus, epidural, and caudal regional anesthetics. No adverse cardiovascular, pulmonary or nervous system events were associated with any of the seizures, including in 16 patients who received bupivacaine blocks. Clinically, which of the usual features of systemic local anesthetic toxicity occurs, the order in which they occur and how soon after local anesthetic administration are quite variable (Fig. 45-21).102 This is not surprising given what is known about how various health conditions, other drugs, and rate of increase of local anesthetic concentration in systemic circulation influence the manifestation and progression of signs and symptoms of local anesthetic toxicity.

In dogs, the relative CNS toxicity of bupivacaine, etidocaine, and lidocaine is 4:2:1,103 which is similar to the relative potency of these agents for the production of regional anesthesia in humans. The convulsant doses of bupivacaine and ropivacaine are similar.104 IV infusion studies in human volunteers have also demonstrated an inverse relationship between the intrinsic anesthetic potency of various agents and the dosage required to induce CNS toxicity.105,106

The rate of IV administration alters the toxicity of local anesthetic agents.106 In human volunteers, an average dose of 236 mg of etidocaine and a venous blood concentration of 3.0 μg/mL resulted in CNS symptoms when 10 mg/min was infused. When the infusion rate was increased to 20 mg/min, an average of 161 mg of etidocaine, which produced a venous plasma concentration of approximately 2 μg/mL, caused symptoms of CNS toxicity.

Acid–base status can alter the CNS activity of local anesthetic agents. In cats, the convulsive threshold of various local anesthetics is inversely related to the arterial CO2 tension (Paco2)107 (Fig. 45-22). An increase in Paco2 from 25 to 40 mm Hg to a range of 65 to 81 mm Hg decreases the convulsive threshold of procaine, mepivacaine, prilocaine, lidocaine, and bupivacaine by approximately 50%. A decrease in arterial pH also decreases the convulsant threshold of these agents. Respiratory acidosis, with a resultant increase in Paco2 and a decrease in arterial pH, consistently decreases the convulsant threshold of local anesthetic agents. However, an elevation in both Paco2 and arterial pH, as may occur during metabolic alkalosis, does not increase CNS toxicity to the same degree.

Hypercarbia increases cerebral blood flow, which probably results in a greater uptake of local anesthetic by the brain. In addition, diffusion of CO2 into neuronal cells decreases intracellular pH and thus increases the intracellular cationic form of the local anesthetic agents. This form does not diffuse well across the nerve membrane, so ion trapping occurs. Hypercarbia and/or acidosis also decrease the plasma protein binding of local anesthetic agents, which will increase the proportion of free drug available for diffusion into the brain.108,109 On the other hand, acidosis also decreases the percentage of the local anesthetic existing in the base form, which should decrease the rate of diffusion into neuronal cells.

Local anesthetics have a direct effect on both cardiac muscle and vascular smooth muscle.97 These agents alter the heart's electrical and mechanical activity. Studies using the intact, isolated mammalian heart in vitro show that highly lipid-soluble, extensively protein-bound, highly potent local anesthetics (eg, tetracaine, bupivacaine, etidocaine) are much more cardiotoxic than are the less lipid-soluble, protein-bound, potent local anesthetics (eg, lidocaine, mepivacaine, prilocaine).110 Bupivacaine has a potent depressant effect on electrical conduction in the heart primarily via an action on voltage-gated sodium channels that generally govern the initial rapid depolarization (phase 0) of cardiomyocytes. The S forms of bupivacaine are less cardiotoxic than the R form. Bupivacaine actions other than on voltage-gated sodium channels probably also contribute to dose-dependent cardiotoxic effect of this local anesthetic.

Local anesthetics decrease the maximal rate of depolarization in Purkinje fibers and ventricular muscle because of an inhibition of sodium channels in cardiac membranes.111-113 Action potential duration and the effective refractory period are also decreased by local anesthetics. However, the ratio of effective refractory period to action potential duration is increased both in Purkinje fibers and in ventricular muscle.

Qualitative differences exist between the various local anesthetic agents. Bupivacaine depresses the rapid phase of depolarization (Vmax) in Purkinje fibers and ventricular muscle to a greater extent than does lidocaine.111-113 In addition, the rate of recovery from a use-dependent block is slower in bupivacaine-treated than in lidocaine-treated papillary muscles.114 This slow rate of recovery results in an incomplete restoration of Vmax between action potentials, particularly at high heart rates. In contrast, recovery from lidocaine is complete, even at rapid heart rates. These differential effects of lidocaine and bupivacaine may explain the antidysrhythmic properties of lidocaine and the dysrhythmogenic potential of bupivacaine.

Bupivacaine, and to a lesser degree etidocaine and ropivacaine, can produce severe cardiac dysrhythmias, including ventricular fibrillation, in various animal species.97,115-119 Ventricular dysrhythmias are rarely seen with lidocaine, mepivacaine, or tetracaine. Although the dysrhythmogenic action of bupivacaine is probably related primarily to an inhibition of the fast sodium channels in the cardiac membrane, evidence also exists that this agent may block the slow calcium channels.120 These electrophysiologic effects of bupivacaine may result in conduction abnormalities, leading to a reentrant type of dysrhythmia similar to a torsade de pointes dysrhythmia.116

The dysrhythmogenic activity of bupivacaine is believed to result primarily from a direct cardiac effect. Isolated guinea pig hearts perfused with bupivacaine revealed evidence of conduction block, bigeminy, and trigeminy.121 In addition, ventricular fibrillation occurred in intact pigs in which bupivacaine was injected directly into the left anterior descending coronary artery.118 On the other hand, the injection of bupivacaine directly into certain regions of the brain may result in cardiac dysrhythmias, which may indicate a relationship between the CNS and cardiotoxic effects of bupivacaine.122,123

Electrophysiologic studies in intact dogs and in people have shown that high blood levels of local anesthetics prolong conduction time through various parts of the heart as indicated by an increase in the PR interval and QRS duration. Extremely high concentrations of local anesthetics depress spontaneous pacemaker activity in the sinus node, resulting in sinus bradycardia and sinus arrest.

Local anesthetic agents also depress myocardial contractility. All local anesthetics exert a dose-dependent negative inotropic action on isolated cardiac tissue that is proportional to the conduction blocking potency of the various agents in isolated nerves (Fig. 45-23).124 For example, bupivacaine, tetracaine, and etidocaine produce the greatest degree of myocardial depression. The agents of moderate anesthetic potency (ie, lidocaine, mepivacaine, prilocaine) are intermediate in terms of their negative inotropic action. Procaine and chloroprocaine, which are the least-potent local anesthetics, require the highest concentration to decrease cardiac contractility.

In dogs, tetracaine is approximately 8 to 10 times more potent than procaine as a local anesthetic and as a myocardial depressant.125 Hemodynamic studies in closed-chest anesthetized dogs have shown that tetracaine, etidocaine, and bupivacaine caused a 50% decrease in cardiac output at doses of 10 to 20 mg/kg, whereas doses of 30 to 40 mg/kg of lidocaine, mepivacaine, prilocaine, and chloroprocaine were required for a similar decrease in cardiac output. A dose of 100 mg/kg of procaine was needed to reduce cardiac output to 50%.

Most local anesthetic agents exert a biphasic effect on peripheral vascular smooth muscle.56,57 Low concentrations of lidocaine and bupivacaine produced vasoconstriction in the cremaster muscle of rats, whereas high concentrations increased arteriolar diameter, indicative of vasodilatation. In vivo studies also demonstrate that low doses of local anesthetics decrease peripheral arterial flow without any change in blood pressure, whereas higher doses increase blood flow. Cocaine causes vasoconstriction because of its ability to inhibit the uptake of norepinephrine by storage granules.126 Studies indicate that ropivacaine causes cutaneous vasoconstriction, whereas bupivacaine produces vasodilatation.59

In a review of the cardiotoxicity of modern local anesthetics, Mather and Chang127 concluded that as compared with bupivacaine, although ropivacaine and levobupivacaine may be seen as "safer," they must not be regarded as totally "safe."

Factors Influencing Cardiovascular Toxicity

Specific Agents

Although the CNS is more susceptible to the toxic effects of local anesthetics than the cardiovascular system, differences exist in the margin between the dose of various agents that causes convulsions and the dose that results in cardiovascular collapse (Fig. 45-24). A cardiovascular collapse (CC)-to-convulsive (CNS) dose ratio of 7.1 + 1.1 was reported for lidocaine in adult sheep, indicating that 7 times as much drug was required to induce irreversible cardiovascular collapse as to cause convulsions.128 The CC:CNS ratio for bupivacaine was 3.7 + 0.5 and for etidocaine, 4.4 + 0.9. The CC:CNS blood level ratio of lidocaine was 3.6 + 0.3, compared with values of 1.6 to 1.7 for bupivacaine and etidocaine. At the time of cardiovascular collapse, high concentrations of bupivacaine and etidocaine were present in the myocardium compared with lidocaine, which suggests that the enhanced cardiac toxicity of these more potent agents may result from a greater myocardial uptake.

Pregnancy

Data regarding the effects of pregnancy on the cardiovascular toxicity of local anesthetic are not conclusive.

Acidosis and Hypoxia

Hypercarbia, acidosis, and hypoxia potentiate the negative chronotropic and inotropic action of lidocaine and bupivacaine in isolated cardiac tissue.129 The combination of hypoxia and acidosis greatly potentiates the cardiodepressant effects of bupivacaine. Hypoxia and acidosis also increase the frequency of cardiac dysrhythmias and the mortality rate in sheep after the IV administration of bupivacaine.130 Hypercarbia, acidosis, and hypoxia occur very rapidly in some patients after seizure activity because of the rapid unintentional intravascular injection of local anesthetic agents.131 Thus the cardiovascular depression observed in some patients after the accidental IV injection of bupivacaine may be related in part to the severe acid–base changes that occur during toxic reactions to this agent.

Measures to prevent systemic toxic reactions to local anesthetics include following dose recommendations, injecting aliquots over time, avoiding unintentional intravascular injections, and monitoring vital signs during injection. Blanket recommended doses versus block-specific recommended doses were discussed recently.132,133 Drug administration must be stopped should signs or symptoms of toxicity develop. Seizures induced by local anesthetics are usually self-limiting and require maintenance of respiratory gas exchange and control of muscle contractions (eg, intubation, oxygenation, short-acting muscle paralysis). Drugs such as propofol, thiopental, and benzodiazepines are effective against these seizures.

Cardiovascular toxicity is treated according to American Heart Association guidelines, depending on the nature of the toxicity. Recent evidence suggests that in some instances, lipid emulsion infusion may be beneficial.134 Recently published recommendations by the American Society of Regional Anesthesia and Pain Medicine for treatment of local anesthetic systemic toxicity are summarized in Table 45-6.135

Local Anesthetics in Clinical Use

Table 45-7 lists generic and trade names of local anesthetics. Undoubtedly, lidocaine is most commonly used to prevent procedure-related pain and for diagnostic tests. Intermediate- to long-acting local anesthetics, such as ropivacaine, levobupivacaine, and bupivacaine, are used for therapy.

Table 45-8 lists topical anesthetics and dosage forms. Most of the forms are available without prescription. A 5% lidocaine patch (Lidoderm) is approved by the FDA for controlling postherpetic neuralgia.

Amino Ester Agents

Local anesthetics in this class have an ester linkage to benzoic acid or its derivatives. The amino ester–linked local anesthetics most commonly used clinically are procaine, chloroprocaine, and tetracaine. All of these agents were introduced into clinical practice by 1955.

Procaine

Procaine was the first synthetic local anesthetic agent introduced into clinical practice. It is a relatively weak local anesthetic with a slow onset and a short duration of action. Its systemic toxicity is relatively low because of rapid plasma hydrolysis. Procaine is hydrolyzed to p-aminobenzoic acid, which is responsible for the allergic reactions associated with repeated use of procaine. Procaine is used primarily for infiltration anesthesia, diagnostic differential spinal blocks in certain pain states, and obstetric spinal anesthesia.

Chloroprocaine

Chloroprocaine has rapid onset of action, short duration, and low systemic toxicity. It undergoes hydrolysis by human plasma esterases approximately 4 times faster than procaine. Chloroprocaine is primarily employed for epidural analgesia and anesthesia in obstetrics because of its rapid onset and low systemic toxicity in the mother and fetus. However, frequent injections are required to provide adequate pain relief during labor. Sometimes, epidural analgesia is established in the pregnant patient with chloroprocaine, followed by a longer-acting agent such as bupivacaine. Chloroprocaine has also proved of value for various regional anesthetic procedures in ambulatory surgical patients for whom surgery is not expected to exceed 30 to 60 minutes.

Concern about potential myotoxicity and neurotoxicity has haunted chloroprocaine.

Tetracaine

Tetracaine is used primarily for spinal anesthesia. It may be employed as an isobaric, hypobaric, or hyperbaric solution for spinal blockade, although hyperbaric solutions of tetracaine are probably employed most often. Tetracaine provides a relatively rapid onset of spinal anesthesia, excellent qualities of sensory anesthesia, and a profound block of motor function. Plain solutions of tetracaine produce an average duration of spinal anesthesia of 2 to 3 hours, whereas the addition of epinephrine can extend anesthesia to 4 to 6 hours.

Tetracaine is rarely used for other forms of regional anesthesia because of its extremely slow onset of action and the potential for systemic toxic reactions when larger doses are employed.

Cocaine

Cocaine was the first agent successfully employed clinically for the production of local anesthesia. It has limited use in modern anesthesia practice because of its relatively high potential for systemic toxicity and addiction liabilities. It is listed as a schedule II drug in the United States. Cocaine is an excellent topical anesthetic agent, and it produces vasoconstriction at clinically useful concentrations. As a result, it is still sometimes employed to anesthetize and constrict the nasal mucosa before nasotracheal intubation, and otolaryngologists use cocaine during nasal surgery because of its topical anesthetic and vasoconstrictor properties. It is the only local anesthetic that inhibits the reuptake of catecholamines in the central and peripheral nervous systems.

Amino Amide Agents

Amino amide–linked local anesthetics most commonly used clinically are lidocaine, mepivacaine, ropivacaine, bupivacaine, and levobupivacaine. The first step in the biotransformation of the tertiary amine forms typically is dealkylation of the amino nitrogen by cytochrome P450 in the liver.

Lidocaine

Lidocaine, the first of the amino amide–type local anesthetics to be introduced into clinical practice, remains the most versatile and most frequently used drug in this class. It is popular because of its inherent potency, rapid onset, moderate duration of action, and topical anesthetic activity. Solutions of lidocaine are available for infiltration, peripheral nerve blocks, and epidural anesthesia. In addition, hyperbaric lidocaine is useful for spinal anesthesia of 30 to 60 minutes' duration. Lidocaine is also used in ointment, jelly, viscous, and aerosol preparations for a variety of topical anesthetic procedures. Lidocaine currently is still the only agent officially approved in the United States for IV regional anesthesia.

Lidocaine is sometimes given intravenously as an antiepileptic agent, as an analgesic for certain chronic pain states, and as a supplement to general anesthesia. It is administered intravenously for the treatment of ventricular dysrhythmias.

Mepivacaine

Mepivacaine has a local anesthetic profile similar to that of lidocaine. It can produce a profound depth of anesthesia with a relatively rapid onset and a moderate duration of action. Mepivacaine may be used for infiltration, peripheral nerve blocks, and epidural anesthesia, and in some countries, 4% hyperbaric solutions of mepivacaine are also available for spinal anesthesia.

It is ineffective as a topical anesthetic agent. The metabolism of mepivacaine is greatly prolonged in the fetus and newborn; thus this agent is not usually employed for obstetric anesthesia. Mepivacaine appears to be somewhat less toxic in adults than lidocaine and has less vasodilator activity than lidocaine. Because of the difference in vasoactivity, mepivacaine provides a somewhat longer duration of anesthesia than lidocaine when used without epinephrine. Mepivacaine seems to be particularly useful for brachial plexus blockade when large volumes of anesthetic solutions without epinephrine are given.

Ropivacaine

Ropivacaine is prepared as the pure S isomer rather than a racemic mixture. Onset of action, potency, and duration of sensory nerve blockade appear similar for ropivacaine and bupivacaine, but ropivacaine is a less potent and short-acting agent in terms of motor fiber blockade. Toxicity studies suggest that ropivacaine is less cardiotoxic than bupivacaine, although ropivacaine still possesses some dysrhythmogenic potentia.

Bupivacaine

Bupivacaine was the first local anesthetic that combined the properties of an acceptable onset, long duration of action, profound conduction blockade, and significant separation of sensory anesthesia and motor blockade. This agent is used for various regional anesthetic procedures, including infiltration, peripheral nerve blocks, and epidural and spinal anesthesia. The average duration of surgical anesthesia with bupivacaine varies from approximately 3 to 10 hours. Its longest duration of action occurs with major peripheral nerve blocks such as brachial plexus blockade.

The major advantage of bupivacaine appears to involve epidural obstetric analgesia for labor when satisfactory pain relief for 2 to 3 hours is achieved, which significantly decreases the need for repeated injections in the pregnant patient. Moreover, adequate analgesia is usually achieved without significant motor blockade such that the patient in labor is able to move her legs. This differential blockade of sensory and motor fibers is also the basis for the widespread use of bupivacaine for postoperative epidural analgesia and for certain chronic pain states.

Levobupivacaine

The clinical profile of levobupivacaine, the S isomer of bupivacaine, is essentially the same as the profile of racemic bupivacaine except with somewhat wider therapeutic index for cardiac toxicity and systemic toxicity.

Prilocaine

The clinical profile of prilocaine is also similar to that of lidocaine. Prilocaine has a relatively rapid onset of action while providing a moderate duration of anesthesia and a profound depth of conduction blockade. Because this agent causes significantly less vasodilatation than lidocaine, it can be used without epinephrine.

Prilocaine biotransformation produces aminophenols that oxidize hemoglobin to methemoglobin thus limiting its clinical use. The primary use of prilocaine is in EMLA cream, a eutectic mixture of prilocaine and lidocaine for topical application.

Etidocaine

Etidocaine is characterized by very rapid onset, prolonged duration of action, and profound sensory and motor blockade. It has a significantly more rapid onset of action than bupivacaine. Concentrations of etidocaine required for adequate sensory anesthesia produce profound motor blockade. As a result, etidocaine is primarily useful as an anesthetic for surgical procedures in which muscle relaxation is required. Consequently, this agent is of limited use for obstetric epidural analgesia and postoperative pain relief, as it does not provide a differential blockade of sensory and motor fibers. The drug is used infrequently in North America.

Dibucaine

Dibucaine is only available for topical anesthesia in the United States. It is more potent than tetracaine, although the onset of action of the 2 agents is similar. The duration of spinal anesthesia is slightly longer with dibucaine. The degree of hypotension and depth of motor blockade appear to be less in patients receiving intrathecal dibucaine than in those receiving tetracaine into the subarachnoid space, although the spread of sensory anesthesia is similar in the 2 groups.

Benzocaine

This local anesthetic is used exclusively for topical anesthesia. Benzocaine is available in a variety of proprietary and nonproprietary preparations. The most common forms used in an operating room setting are aerosol solutions for endotracheal administration and ointments for lubrication of endotracheal tubes. One should remember that benzocaine also can cause methemoglobinemia.

• Albright GA. Cardiac arrest following regional anesthesia with etidocaine or bupivacaine. Anesthesiology. 1979;51:285-287.
• Butterworth JF. Models and mechanisms of local anesthetic cardiac toxicity. Reg Anesth Pain Med. 2010;35:167-176.
• Drasner K. Local anesthetic systemic toxicity. A historical prospective. Reg Anesth Pain Med. 2010;35:162-166.
• Gerner P, Above all, do no harm. Anesthesiology. 2008;109:939-939.
• Heavner JE, guest ed. Additives to Local Anesthetics. Techniques Reg Anesth Pain Manage. 2004;8:105-140.
• IUPHAR database: Voltage-gated sodium channels: Introduction. http://iuphar-db.org/DATABASE. Accessed June 18, 2010.
• Mackley CL, Marks JG, Anderson BE. Delayed-type hypersensitivity to lidocaine. Arch Dermatol. 2003;139:343-346.
• Moller RA, Covino BG. Cardiac electrophysiologic effects of lidocaine and bupivacaine. Anesth Analg. 1988;67:107-114.
• Stewart J, Kellett N, Castro D. The central nervous system and cardiovascular effects of levobupivacaine and ropivacaine in healthy volunteers. Anesth Analg. 2003;97:412-416.
• Zaric D, Christiansen C, Pace NL, et al. Transient neurologic symptoms after spinal anesthesia with lidocaine versus other local anesthetics: a systematic review of randomized, controlled trails. Anesth Analg. 2005;100:1811-1816.