Clinical Pharmacology of Local Anesthetics

CLINICAL PHARMACOLOGY OF LOCAL ANESTHETICS

Local anesthetics can provide highly effective analgesia in well-defined regions of the body. The usual routes of administration include topical application (eg, nasal mucosa, wound [incision site] margins), injection in the vicinity of peripheral nerve endings (perineural infiltration) and major nerve trunks (blocks), and injection into the epidural or subarachnoid spaces surrounding the spinal cord (Figure 26–4).

Clinical Block Characteristics

In clinical practice, there is generally an orderly evolution of block components beginning with sympathetic transmission and progressing to temperature, pain, light touch, and finally motor block. 

This is most readily appreciated during onset of spinal anesthesia, where a spatial discrepancy can be detected in modal-ities, the most vulnerable components achieving greater derma-tomal (cephalad) spread. Thus, loss of the sensation of cold (often assessed by a wet alcohol sponge) will be roughly two seg-ments above the analgesic level for pinprick, which in turn will be roughly two segments rostral to loss of light touch recogni-tion. However, because of the anatomic considerations noted earlier for peripheral nerve trunks, onset with peripheral blocks is more variable, and proximal motor weakness may precedeonset of more distal sensory loss. Additionally, anesthetic solu-tion is not generally deposited evenly around a nerve bundle, and longitudinal spread and radial penetration into the nerve trunk are far from uniform.With respect to differential block, it is worth noting that “successful” surgical anesthesia may require loss of touch, not just ablation of pain, as some patients will find even the sensation of touch distressing during surgery, often fearing that the procedure may become painful. Further, while differences may exist in modalities, it is not possible with conventional techniques to pro-duce surgical anesthesia without some loss of motor function.

A. Effect of Added Vasoconstrictors

Several benefits may be derived from addition of a vasoconstrictor to a local anesthetic. First, localized neuronal uptake is enhanced because of higher sustained local tissue concentrations that can translate clini-cally into a longer duration block. This may enable adequate anesthe-sia for more prolonged procedures, extended duration of postoperative pain control, and lower total anesthetic requirement. Second, peak blood levels will be lowered as absorption is more closely matched to metabolism and elimination, and the risk of systemic toxic effects is reduced. Moreover, when incorporated into a spinal anesthetic, epi-nephrine may not only contribute to prolongation of the local anes-thetic effect via its vasoconstrictor properties, but also exert a direct analgesic effect mediated by postsynaptic α₂ adrenoceptors within the spinal cord. Recognition of this potential has led to the clinical use of the α₂ agonist clonidine as a local anesthetic adjuvant for spinal anesthesia.

Conversely, inclusion of epinephrine may have untoward effects. The addition of epinephrine to anesthetic solutions can potentiate the neurotoxicity of local anesthetics used for periph-eral nerve blocks or spinal anesthesia. Further, the use of a vaso-constrictor agent in an area that lacks adequate collateral flow (eg, digital block) is generally avoided, though some have questioned the validity of this proscription.

B. Intentional Use of Systemic Local Anesthetics

Although the principal use of local anesthetics is to achieve anesthe-sia in a restricted area, these agents are sometimes deliberately administered systemically to take advantage of suppressive effects on pain processing. In addition to documented reductions in anesthetic requirement and postoperative pain, systemic administration of local anesthetics has been used with some success in the treatment of chronic pain, and this effect may outlast the duration of anes-thetic exposure. The achievement of pain control by systemic administration of local anesthetics is thought to derive, at least in part, from the suppression of abnormal ectopic discharge, an effect observed at concentrations of local anesthetic an order of magnitude lower than those required for blockade of propagation of action potentials in normal nerves. Consequently, these effects can be achieved without the adverse effects that would derive from failure of normal nerve conduction. Escalating doses of anesthetic appear to exert the following systemic actions: (1) low concentrations may preferentially suppress ectopic impulse generation in chronically injured peripheral nerves; (2) moderate concentrations may sup-press central sensitization, which would explain therapeutic benefit that may extend beyond the anesthetic exposure; and (3) higher concentrations will produce general analgesic effects and may cul-minate in serious toxicity.

Toxicity

Local anesthetic toxicity derives from two distinct processes: (1) systemic effects following inadvertent intravascular injection or absorption of the local anesthetic from the site of administration; and (2) neurotoxicity resulting from local effects produced by direct contact with neural elements.

A. Systemic Toxicity

The dose of local anesthetic used for epidural anesthesia or high-volume peripheral blocks is sufficient to produce major clinical toxicity, even death. To minimize risk, maximum recommended doses for each drug for each general application have been pro-mulgated. The concept underlying this approach is that absorp-tion from the site of injection should appropriately match metabolism, thereby preventing toxic serum levels. However, these recommendations do not consider patient characteristics or concomitant risk factors, nor do they take into account the specific peripheral nerve block performed, which has a signifi-cant impact on the rate of systemic uptake (Figure 26–2). Most importantly, they fail to afford protection from toxicity induced by inadvertent intravascular injection (occasionally into an artery, but more commonly a vein).

1. CNS toxicity— All local anesthetics have the ability to pro-duce sedation, light-headedness, visual and auditory distur-bances, and restlessness when high plasma concentrations result from rapid absorption or inadvertent intravascular administra-tion. An early symptom of local anesthetic toxicity is circumoral and tongue numbness and a metallic taste. At higher concentra-tions, nystagmus and muscular twitching occur, followed by tonic-clonic convulsions. Local anesthetics apparently cause depression of cortical inhibitory pathways, thereby allowing unopposed activity of excitatory neuronal pathways. This transi-tional stage of unbalanced excitation (ie, seizure activity) is then followed by generalized CNS depression. However, this classic pattern of evolving toxicity has been largely characterized in human volunteer studies (which are ethically constrained to low doses), and by graded administration in animal models. Deviations from such classic progression are common in clinical toxicity and will be influenced by a host of factors, including patient vulnerability, the particular anesthetic administered, concurrent drugs, and rate of rise of serum drug levels. A recent literature review of reported clinical cases of local anesthetic cardiac toxicity found prodromal signs of CNS toxicity in only 18% of cases. 

When large doses of a local anesthetic are required (eg, for major peripheral nerve block or local infiltration for major plastic surgery), premedication with a parenteral benzodiazepine (eg, diazepam or midazolam) will provide some prophylaxis against local anesthetic-induced CNS toxicity. However, such premedication will have little, if any, effect on cardiovascular toxicity, potentially delaying recogni-tion of a life-threatening overdose. Of note, administration of a propofol infusion or general anesthesia accounted for 5 of the 10 cases presenting with isolated cardiovascular toxicity in the afore-mentioned literature review of reported clinical cases.

If seizures do occur, it is critical to prevent hypoxemia and acidosis, which potentiate anesthetic toxicity. Rapid tracheal intu-bation can facilitate adequate ventilation and oxygenation, and is essential to prevent pulmonary aspiration of gastric contents in patients at risk. The effect of hyperventilation is complex, and its role in resuscitation following anesthetic overdose is somewhat controversial, but it likely offers distinct benefit if used to coun-teract metabolic acidosis. Seizures induced by local anesthetics should be rapidly controlled to prevent patient harm and exacer-bation of acidosis. A recent practice advisory from the American Society of Regional Anesthesia advocates benzodiazepines as first-line drugs (eg, midazolam, 0.03–0.06 mg/kg) because of their hemodynamic stability, but small doses of propofol (eg, 0.25–0.5 mg/kg) were considered acceptable alternatives, as they are often more immediately available in the setting of local anesthetic administration. The motor activity of the seizure can be effectively terminated by administration of a neuromuscular blocker, though this will not diminish the CNS manifestations, and efforts must include therapy directed at the underlying seizure activity.

2. Cardiotoxicity—The most feared complications associatedwith local anesthetic administration result from the profound effects these agents can have on cardiac conduction and function. In 1979, an editorial by Albright reviewed the circumstances of six deaths associated with the use of bupivacaine and etidocaine. This seminal piece suggested that these relatively new lipophilic and potent anesthetics had greater potential cardiotoxicity, and that cardiac arrest could occur concurrently or immediately following seizures and, most importantly, in the absence of hypoxia or aci-dosis. Although this suggestion was sharply criticized, subsequent clinical experience unfortunately reinforced Albright’s concern— within 4 years the Food and Drug Administration (FDA) had received reports of 12 cases of cardiac arrest associated with the use of 0.75% bupivacaine for epidural anesthesia in obstetrics. Further support for enhanced cardiotoxicity of these anesthetics came from animal studies demonstrating that doses of bupivacaine and etidocaine as low as two thirds those producing convulsions could induce arrhythmias, while the margin between CNS and cardiac toxicity was less than half that for lidocaine. In response, the FDA banned the use of 0.75% bupivacaine in obstetrics. In addition, incorporation of a test dose became ingrained as a standard of anesthetic practice, along with the practice of fractionated admin-istration of local anesthetic. 

Although reduction in bupivacaine’s anesthetic concentration and changes in anesthetic practice did much to reduce the risk of cardiotoxicity, the recognized differences in the toxicity of the stereoisomers comprising bupivacaine created an opportunity for the development of potentially safer anesthetics . Investigations demonstrated that the enantiomers of the racemic mixture bupivacaine were not equivalent with respect to cardiotox-icity, the S(–) enantiomer having better therapeutic advantage, lead-ing to the subsequent marketing of levobupivacaine. This was followed shortly thereafter by ropivacaine, a slightly less potent anesthetic than bupivacaine. It should be noted, however, that the reduction in toxicity afforded by these compounds is only modest, and that risk of significant cardiotoxicity remains a very real concern when these anesthetics are administered for high-volume blocks.

3. Reversal of bupivacaine toxicity— Recently, a series ofclinical events, serendipitous observations, systematic experimen-tation, and astute clinical decisions have identified a relatively simple, practical and apparently effective therapy for resistant bupivacaine cardiotoxicity using intravenous infusion of lipid. Furthermore, this therapy appears to have applications that extendbeyond bupivacaine cardiotoxicity to the cardiac or CNS toxicity induced by an overdose of any lipid-soluble drug (see Box: Lipid Resuscitation).

B. Localized Toxicity

1. Neural injury—From the early introduction of spinal anes-thesia into clinical practice, sporadic reports of neurologic injury associated with this technique raised concern that local anesthetic agents were potentially neurotoxic. Following injuries associated with Durocaine—a spinal anesthetic formulation containing procaine—initial attention focused on the vehicle components. However, experimental studies found 10% procaine alone induced similar injuries in cats, whereas the vehicle did not. Concern for anesthetic neurotoxicity reemerged in the early 1980s with a series of reports of major neurologic injury occurring with the use of chloroprocaine for epidural anesthesia. In these cases, there was evidence that anesthetic intended for the epidural space was inad-vertently administered intrathecally. As the dose required for spi-nal anesthesia is roughly an order of magnitude less than for epidural anesthesia, injury was apparently the result of excessive exposure of the more vulnerable subarachnoid neural elements. 

With changes in vehicle formulation and in clinical practice, concern for toxicity again subsided, only to reemerge a decade later with reports of cauda equina syndrome associated with con-tinuous spinal anesthesia (CSA). In contrast to the more common single-injection technique, CSA involves placing a catheter in the subarachnoid space to permit repetitive dosing to facilitate ade-quate anesthesia and maintenance of block for extended periods. In these cases the local anesthetic was evidently administered to a relatively restricted area of the subarachnoid space; in order to extend the block to achieve adequate surgical anesthesia, multiple repetitive doses of anesthetic were then administered. By the time the block was adequate, neurotoxic concentrations had accumu-lated in a restricted area of the caudal region of the subarachnoid space. Most notably, the anesthetic involved in the majority of these cases was lidocaine, a drug most clinicians considered to be the least toxic of agents. This was followed by reports of neuro-toxic injury occurring with lidocaine intended for epidural admin-istration that had inadvertently been administered intrathecally, similar to the cases involving chloroprocaine a decade earlier. The occurrence of neurotoxic injury with CSA and subarachnoid administration of epidural doses of lidocaine served to establish vulnerability whenever excessive anesthetic was administered intrathecally, regardless of the specific anesthetic used. Of even more concern, subsequent reports provided evidence for injury with spinal lidocaine administered at the high end of the recom-mended clinical dosage, prompting recommendations for a reduc-tion in maximum dose. These clinical reports (as well as concurrent experimental studies) served to dispel the concept that modern local anesthetics administered at clinically relevant doses and con-centrations were incapable of inducing neurotoxic injury.

The mechanism of local anesthetic neurotoxicity has been extensively investigated in cell culture, isolated axons, and in vivo models. These studies have demonstrated a myriad of deleterious effects including conduction failure, membrane damage, enzyme leakage, cytoskeletal disruption, accumulation of intracellular calcium, disruption of axonal transport, growth cone collapse, and apoptosis. It is not clear what role these factors or others play in clinical injury. It is clear, however, that injury does not result from blockade of the voltage-gated sodium channel per se, and thus clinical effect and toxicity are not tightly linked.

2. Transient neurologic symptoms (TNS)—In addition to thevery rare but devastating neural complications that can occur with neuraxial (spinal and epidural) administration of local anesthetics, a syndrome of transient pain or dysesthesia, or both, has been recently linked to use of lidocaine for spinal anesthesia. Although these symp-toms are not associated with sensory loss, motor weakness, or bowel and bladder dysfunction, the pain can be quite severe, often exceeding that induced by the surgical procedure. TNS occurs even at modest doses of anesthetic, and has been documented in as many as one third of patients receiving lidocaine, with increased risk associated with certain patient positions during surgery (eg, lithotomy), and with ambulatory anesthesia. Risk with other anesthetics varies consider-ably. For example, the incidence is only slightly reduced with procaine or mepivacaine but appears to be negligible with bupivacaine, prilo-caine, and chloroprocaine. The etiology and significance of TNS remain to be established, but differences between factors affecting TNS and experimental animal toxicity argue strongly against a com-mon mechanism mediating these symptoms and persistent or perma-nent neurologic deficits. Nonetheless, the high incidence of TNS has greatly contributed to dissatisfaction with lidocaine as a spinal anes-thetic, leading to its near abandonment for this technique (although it remains a popular and appropriate anesthetic for all other applications, including epidural anesthesia). Chloroprocaine, once considered amore toxic anesthetic, is now being explored for short-duration spinal anesthesia as an alternative to lidocaine, a compound that has been used for well over 50 million spinal anesthetic procedures.