Local Tissue Toxicity

All the clinically used aminoamide and aminoester local anesthetics can produce direct toxicity to nerves if they achieve sufficiently high intraneural concentrations. Conversely, in the great majority of clinical applications, no damage to nerves occurs. Although local anesthetics are usually packaged and injected at concentrations well above their physiologically effective range, in the process of delivery they are usually diluted sufficiently that no harm is done. If such dilution does not occur, however, long-term or permanent neural deficits do result. Thus, the application of 5% (200 mM) lidocaine in viscous, dense solutions through narrow intrathecal catheters has been associated with a high frequency of cases of cauda equina syndrome.^[187] Laboratory investigations have shown that such high concentrations of local anesthetics alone applied directly to bare nerve fibers produce irreversible conduction block in less than 5 minutes. ^[188] Indeed, previous studies on ensheathed peripheral nerves in vivo had shown neurologic and histologic changes after infiltration of the space surrounding the nerve with local anesthetics at concentrations as low as 1% to 2%. Clinicians should be aware that the concentrations of formulated local anesthetic solutions are neurotoxic per se and that dilution of them, in situ or in tissue, is essential for safe use.

Concentrations of lidocaine required to produce irreversible conduction blockade in isolated desheathed peripheral nerves slightly overlap the concentrations used clinically (e.g., 2%), but these concentrations are applied clinically to ensheathed nerve in situ embedded in drug-absorbing tissue. Irreversible block in desheathed nerve by lidocaine has a threshold of 20 mM (0.5%) and a 50% effective concentration (EC₅₀ ) of 45 mM (1.1%) versus an EC₅₀ of 1 mM for reversible impulse blockade in vivo.^[189] The intrathecal administration of tetracaine or etidocaine in rabbits resulted in histopathologic spinal cord changes after the use of 2% tetracaine, which exceeds the maximum concentration of 1% used for spinal anesthesia in humans.^[190]

In the late 1970s and early 1980s, prolonged sensory and motor deficits were reported in some patients after the epidural or subarachnoid injection of large doses of chloroprocaine.^{[191] [192]} Studies in animals have proved somewhat contradictory regarding the potential neurotoxicity of chloroprocaine.^{[193] [194] [195] [196] [197] [198]} The results of these studies suggest that the combination of low pH, sodium bisulfite, and inadvertent intrathecal dosing is responsible in part for the neurotoxic reactions observed after the use of large amounts of chloroprocaine solution. Chloroprocaine itself at high concentrations may also be neurotoxic,^[198] but these concentrations are not achieved during properly positioned epidural anesthesia. The currently available commercial solutions of chloroprocaine do not contain sodium bisulfite. Sodium bisulfite was initially replaced by ethyleneglycoltetraacetic acid (EGTA), a preservative and high-affinity calcium chelator that was occasionally reported to cause local muscle spasm after epidural administration. More recently, chloroprocaine has become available in an entirely preservative-free preparation. Chloroprocaine has unique utility in situations in which rapid plasma clearance is required to prevent excessive systemic accumulation of local anesthetic. Chloroprocaine has been administered by epidural infusion in young infants in settings in which lidocaine and bupivacaine could not provide an effective therapeutic index.^[117]

Concern regarding local anesthetic-associated neurotoxicity was increased after studies of continuous spinal anesthesia administered by microcatheter. ^[187] In a number of patients in these initial studies, transient or longer-term radicular irritation or, in some cases, cauda equina syndrome developed. Investigations suggested that microcatheters may facilitate localized injection of high concentrations of drug that are inadequately dispersed and diluted in cerebrospinal fluid, thereby leading to high intraneural drug concentrations around the sacral roots and subsequent toxicity.

Related investigations showed that single-shot spinal anesthesia with commonly recommended doses and concentrations of local anesthetics can produce more limited and transient neurologic symptoms (back pain, paresthesias, radicular pain, or hypoesthesia).^{[199] [200] [201] [202]} Transient neurologic symptoms have since been observed with many different local anesthetics. Some studies have found that mepivacaine and lidocaine at a range of dilutions cause more frequent symptoms than bupivacaine does. The risk of transient neurologic symptoms after spinal anesthesia was not diminished by dilution of lidocaine from 5% to 1–2%. Differences in study design, method of questioning, and criteria for inclusion may be partially responsible for differences in the prevalence of radicular sequelae in different studies. Despite these differences in study design, a recent meta-analysis concluded that the pooled relative risk for transient neurologic symptoms after spinal anesthesia with lidocaine was 6.7-fold higher than with bupivacaine and 5.5-fold higher than with prilocaine.^[203] The addition of vasoconstrictors to local anesthetic solutions may also increase the risk.^[204] Neurotoxicity appears to be unrelated to conduction block per se because tetrodotoxin, a highly potent blocker of sodium channels, can produce intense conduction blockade without histologic or behavioral signs of nerve injury.^[205]

Intraoperative positioning also appears to be a risk factor. Patients undergoing surgery in the lithotomy position seem to have an increased risk for neurologic symptoms after either spinal or epidural anesthesia. It is unknown at present why this association produces increased risk; nerve compression or stretch or reduced perfusion pressure in the vasa nervorum may possibly exacerbate toxicity from local anesthetics. The lithotomy position per se can produce neurologic sequelae and lower extremity compartment syndromes, particularly with prolonged surgery and use of the Trendelenburg position.^{[206] [207] [208]}

Skeletal muscle changes have been observed after the intramuscular injection of local anesthetics such as lidocaine, mepivacaine, prilocaine, bupivacaine, and etidocaine.^{[209] [210] [211]} In general, the more potent, longer-acting

598