Local anesthetics represent one of the most clinically versatile classes of drugs in medicine, enabling pain-free surgical procedures, labor analgesia, and chronic pain management without the systemic risks associated with general anesthesia or opioid analgesics. Despite their widespread use across virtually every clinical specialty, the underlying pharmacology of local anesthetics is frequently misunderstood or reduced to oversimplified mnemonics. A thorough command of their mechanism of action, physicochemical properties, and the biophysical basis of differential nerve blockade allows the clinician to make rational, evidence-based decisions about agent selection, technique, and dosing; and to anticipate and manage adverse outcomes when they arise.1 This module builds the pharmacologic foundation upon which subsequent modules on individual agents, clinical applications, and toxicity management will rest.
The primary molecular target of all clinically relevant local anesthetics is the voltage-gated sodium channel (Nav), a transmembrane protein responsible for the rapid depolarization phase of the action potential.1 2 Understanding the architecture of this channel is essential to understanding both how local anesthetics work and why their effects are state- and frequency-dependent. Nav channels are large, single-chain glycoproteins of approximately 260 kDa composed of a pore-forming α-subunit and one or more auxiliary β-subunits. The α-subunit contains four homologous domains (DI–DIV), each consisting of six transmembrane segments (S1–S6). The S4 segment of each domain functions as the voltage sensor, containing positively charged arginine and lysine residues at every third position; upon membrane depolarization, the S4 segments rotate and translocate outward through the electric field, triggering conformational changes that open the central pore.2 3 The pore itself is formed by the S5–S6 segments and the intervening P-loops of each domain, which fold inward to form the ion selectivity filter: a ring of negatively charged residues that allows the rapid, selective influx of sodium ions down their electrochemical gradient.3
The local anesthetic binding site is located within the inner vestibule of the channel pore, formed primarily by residues on the S6 segments of domains I, III, and IV, particularly the highly conserved phenylalanine (F1764) and tyrosine (Y1771) residues in domain IV.4 These residues are accessible from the cytoplasmic face of the channel, which has profound implications for drug entry and the state-dependence of blockade discussed below.
Nav channels exist in three primary functional states: rested (closed), open, and inactivated (closed). From the rested state, a threshold depolarization opens the channel within microseconds, allowing the inward sodium current that generates the rising phase of the action potential. Within approximately 1 millisecond at typical physiologic depolarizations, the channel enters the inactivated state: an auto-blocked conformation in which the intracellular linker between DIII and DIV (the "inactivation gate" or h-gate) occludes the inner pore. The channel cannot reopen until it returns to the rested state, a process that requires repolarization of the membrane and occupies the refractory period of the cell. The transition from inactivated back to rested is voltage- and time-dependent and is a critical determinant of the cell's ability to fire repetitively.2 3
Local anesthetics produce conduction blockade by binding to Nav channels and stabilizing them in a non-conducting state, thereby preventing the rapid inward sodium current necessary for action potential generation and propagation.1 2 The mechanism is neither simple pore occlusion nor purely competitive antagonism; it is more accurately described as a state-dependent allosteric stabilization of the inactivated channel conformation.
Local anesthetic molecules reach their binding site within the channel pore via two distinct routes. The hydrophilic pathway involves passage of the charged (protonated, cationic) form of the drug through the open channel pore from the cytoplasmic side; the drug molecule must wait for the channel to open before it can enter. The hydrophobic pathway allows the uncharged (free base) form of the drug to diffuse directly through the lipid bilayer and access the inner vestibule from within the membrane, regardless of channel state.4 5 These two mechanisms coexist for most clinically used agents, with the relative contribution of each pathway determined by the drug's pKa (which determines the fraction in uncharged form at physiologic pH) and its lipid solubility (which governs membrane partitioning). Once bound within the inner vestibule, local anesthetics stabilize the channel in a closed, inactivated-like conformation, preventing the conformational changes required for pore opening. The drug–channel complex is maintained as long as the drug remains bound; dissociation is required for channel recovery and resumption of normal excitability.4
A defining feature of local anesthetic pharmacology is that binding affinity differs markedly among the three channel states. Local anesthetics bind with much higher affinity to the open and inactivated states than to the rested (closed) state.5 This state-dependent binding arises because the binding site within the inner vestibule is more accessible (and adopts a higher-affinity conformation) when the channel is open or inactivated compared to when it is fully rested and closed. The clinical consequence is that local anesthetics are more effective in tissues where channels spend more time in open or inactivated states, that is, in tissues that are depolarizing more frequently. A nerve at rest, with channels predominantly in the rested state, is relatively resistant to local anesthetic block; the same nerve actively firing action potentials accumulates progressively more block with each depolarization cycle because each opening event allows hydrophilic drug entry and shifts more channels into the high-affinity inactivated state.5 This is the basis of tonic vs. phasic block.
Tonic block refers to the basal degree of sodium channel inhibition present even at normal (low) firing rates, attributable primarily to the hydrophobic route of access and binding to rested-state channels with the lower affinity that nonetheless occurs at sufficient drug concentrations. Phasic block (also called frequency-dependent or use-dependent block) refers to the additional, cumulative increment in block that occurs with each successive action potential, as more channels enter the high-affinity open and inactivated states and accumulate bound drug. The magnitude of phasic block increases with firing frequency, which is why rapidly firing nociceptive C-fibers and Aδ-fibers are particularly susceptible to local anesthetic blockade at lower drug concentrations than slowly firing motor neurons.1 5
Analogous to the minimum alveolar concentration (MAC) for inhaled anesthetics, the minimum blocking concentration (Cm) is defined as the minimum concentration of local anesthetic required to block conduction in a given nerve fiber within a defined time period under standard conditions. Cm is not an absolute pharmacologic constant, varying with fiber type, firing rate, pH, and other factors, but it provides a clinically useful framework for understanding relative potency and the dose requirements for different applications.1
The clinical profile of a local anesthetic, including its onset speed, duration of action, potency, and propensity for systemic toxicity, is determined by a set of interacting physicochemical properties. These properties are not independent; they are linked through the chemical structure of the molecule and must be understood as a system rather than as isolated variables.1 6
All clinically used local anesthetics are weak bases. In solution, they exist as an equilibrium mixture of the uncharged free base form (B) and the protonated, positively charged conjugate acid form (BH⁺). The pKa of the molecule is the pH at which these two forms are present in equal concentrations. The Henderson-Hasselbalch equation governs this relationship: at pH values above the pKa, the free base form predominates; at pH values below the pKa, the charged form predominates.6 At physiologic pH of 7.4, local anesthetics with pKa values in the range of 7.6–8.9 (which includes most clinically used agents) exist predominantly in the charged (BH⁺) form. The uncharged free base form, while the minority species, is responsible for membrane penetration and diffusion through lipid-rich tissue to reach the nerve. Once inside the nerve cell (where intracellular pH is approximately 7.2), the equilibrium shifts slightly toward the charged form, which is the species that actually binds to and blocks the sodium channel from within the inner vestibule.6
This dual-form model, combining the "membrane expansion" theory and the "specific receptor" theory, explains why drugs with lower pKa (closer to physiologic pH) have a higher fraction in free base form at physiologic pH and therefore penetrate tissues and nerves more rapidly, producing faster onset.
The critical clinical implication of pKa is that tissue acidosis (encountered in inflamed or infected tissues) significantly reduces the fraction of local anesthetic in the free base form. At a tissue pH of 6.8–7.0 (typical of an abscess or acutely inflamed wound), far more drug remains in the charged form unable to cross the nerve membrane, which explains the well-known clinical frustration of inadequate local anesthesia in infected tissue.6 7 This is not treatment failure; it is predictable pharmacology.
Lipid solubility, conventionally expressed as the octanol-to-water partition coefficient, is the primary determinant of local anesthetic potency. More lipid-soluble agents penetrate lipid-rich nerve membranes more readily and have higher affinity for the hydrophobic binding site within the Nav channel, producing conduction block at lower molar concentrations.1 6 Bupivacaine and ropivacaine, with high lipid solubility, are substantially more potent than lidocaine or mepivacaine, which have intermediate lipid solubility; benzocaine and procaine, with low lipid solubility, are the least potent agents. Lipid solubility also influences duration of action, because more lipid-soluble agents partition extensively into the myelin sheaths and lipid-rich tissue surrounding nerves, creating a local depot from which drug is released slowly, sustaining the block even as plasma concentrations fall. Additionally, high lipid solubility correlates with increased binding to plasma proteins (see below) and increased uptake into adipose and myelin tissue, which partially offsets the risk of systemic toxicity by reducing the free (unbound) plasma concentration available to act on cardiac and central nervous system (CNS) sodium channels.6
Local anesthetics bind reversibly to plasma proteins, predominantly α1-acid glycoprotein (AAG) and, to a lesser extent, albumin. Only the unbound (free) fraction is pharmacologically active and capable of crossing into tissues, exerting clinical effects, and producing systemic toxicity.6 Highly protein-bound agents (bupivacaine ~95%, ropivacaine ~94%) have a small free fraction and, once bound within the nerve, dissociate slowly from the sodium channel, which contributes to their long duration of action. Less protein-bound agents (procaine ~6%, lidocaine ~64%) dissociate more rapidly, producing shorter durations of block. Protein binding is physiologically variable. AAG is an acute-phase reactant whose plasma concentrations rise in states of physiologic stress, surgery, trauma, and malignancy, potentially reducing the free fraction of local anesthetic and influencing both clinical effect and toxicity risk in the postoperative period. Conversely, neonates and patients with liver disease or severe hypoalbuminemia have reduced protein binding, increasing the free fraction and elevating toxicity risk at doses that might be safe in healthy adults.6 8
Molecular weight influences the rate at which a local anesthetic diffuses through tissue barriers, including the perineurium, epineurium, and the myelin sheath. Smaller molecules diffuse more rapidly. In practice, however, molecular weight differences among clinically used agents are relatively modest and are generally a less important determinant of clinical onset than pKa and lipid solubility. Molecular weight becomes more relevant in the context of neuraxial pharmacology, where baricity, spread within the cerebrospinal fluid, and uptake into the spinal cord are influenced by the physical properties of both the drug solution and the surrounding anatomy.1
Understanding how the physicochemical properties above interact to produce clinically observable phenomena requires integrating them into a coherent framework rather than memorizing isolated rules. Onset of block is primarily determined by pKa (lower pKa → higher free base fraction at physiologic pH → faster membrane penetration → faster onset) and, secondarily, by the concentration applied (higher concentration drives more drug across the membrane despite a smaller free base fraction). Lidocaine, with a pKa of 7.9, achieves onset in 5–10 minutes for peripheral nerve blocks. Bupivacaine, with a pKa of 8.1, has a slightly higher fraction in charged form and achieves onset in 15–30 minutes. Chloroprocaine, with a pKa of 8.7 but used in very high concentrations (2–3%), achieves rapid onset in epidural anesthesia despite its high pKa by overwhelming the ionization equilibrium through mass action.1 6
Duration of block is primarily determined by protein binding and lipid solubility, with vasoactive properties of the drug and the use of vasoconstrictors playing an important modifying role. Highly protein-bound, lipid-soluble agents (bupivacaine, ropivacaine) produce blocks lasting 4–12 hours without adjuvants; less protein-bound agents (lidocaine) produce blocks lasting 1–2 hours. The duration is prolonged by epinephrine for most agents (by reducing local blood flow and slowing systemic absorption), except for cocaine (which is itself a vasoconstrictor) and, to a lesser extent, ropivacaine (which has intrinsic vasoconstrictive activity at clinical concentrations).1 Potency is primarily determined by lipid solubility. The relative potencies for peripheral nerve block, in approximate order from least to most potent, run: procaine < chloroprocaine < prilocaine ≈ lidocaine ≈ mepivacaine < bupivacaine ≈ ropivacaine ≈ levobupivacaine.1 6 The more potent, lipid-soluble agents require lower molar concentrations but are not safer on a milligram basis; their intrinsic cardiac toxicity is proportionally greater (discussed in depth in CNS-04).
One of the most clinically consequential features of local anesthetic pharmacology is differential susceptibility to blockade among nerve fiber types. This differential blockade underlies the clinical phenomenon in which patients receiving neuraxial or peripheral nerve blocks lose pain sensation before touch before temperature regulation before proprioception before motor function, a highly ordered sequence with predictable clinical implications.1 9
Peripheral nerve fibers are classified by diameter, myelination, and conduction velocity into several major groups. Aα (Ia and Ib) fibers are the largest myelinated fibers (12–20 μm diameter), carrying proprioceptive and motor impulses with conduction velocities of 70–120 m/s. Aβ fibers (6–12 μm) carry touch and pressure sensation at 30–70 m/s. Aδ fibers (1–5 μm, thinly myelinated) carry fast, sharp ("first") pain, temperature, and light touch at 5–30 m/s. B fibers (1–3 μm, myelinated) are preganglionic autonomic fibers conducting at 3–15 m/s. C fibers (0.2–1.5 μm, unmyelinated) carry slow, burning ("second") pain, temperature, and postganglionic autonomic impulses at 0.5–2 m/s.9 Small-diameter fibers, particularly C fibers and Aδ fibers, are blocked at lower local anesthetic concentrations than large Aα motor fibers. This is the physiologic basis of differential block: at intermediate local anesthetic concentrations, nociception is blocked while motor function is preserved, which is the goal of labor epidural analgesia, many chronic pain interventions, and the early phase of peripheral nerve block establishment.
The explanation for fiber size–dependent susceptibility involves several interacting factors. First, small-diameter unmyelinated C fibers have a higher surface-area-to-volume ratio, meaning a given concentration of local anesthetic has proportionally greater access to the sodium channels along the fiber's length. Second, the critical length theory holds that for a nerve to fail to conduct an action potential, a sufficient length of fiber must be blocked so that the propagating action potential cannot "jump" past the blocked segment. For myelinated fibers, action potentials jump from one node of Ranvier to the next (saltatory conduction), and at least two to three consecutive nodes of Ranvier must be blocked to prevent conduction. Because internodal distance is proportional to fiber diameter (larger fibers have longer internodal distances), larger fibers require more drug to block a sufficient length of nerve.9
Third, myelinated fibers are less accessible to local anesthetics than unmyelinated fibers because the myelin sheath limits drug access to the sodium channels, which are concentrated at the nodes of Ranvier. Paradoxically, B fibers (which are small and myelinated) are among the most sensitive to local anesthetics; the combination of small diameter and the accessibility of their widely spaced nodes makes them very susceptible.1 9
The typical clinical sequence of onset in a well-placed peripheral nerve or neuraxial block proceeds as follows: loss of autonomic tone (vasodilation, warm skin) → loss of pain and temperature sensation → loss of touch and pressure sensation → loss of proprioception → loss of motor function. Recovery occurs in reverse order. The clinician performing an epidural block for labor analgesia exploits this differential by titrating drug concentration and volume to achieve Aδ and C fiber blockade (pain relief) while preserving Aβ and Aα fiber function (touch, proprioception, motor), enabling the parturient to remain ambulatory.10 Conversely, a dense sensorimotor block for a surgical field requires higher concentrations that overcome the threshold for Aα fiber blockade. The practical implication of differential block for toxicity assessment is also important: a patient who retains motor function after a peripheral nerve block does not necessarily have inadequate anesthesia; the differential block may be achieving exactly the intended goal. Conversely, the absence of motor block in a patient reporting inadequate pain relief should prompt assessment of block quality rather than immediate redosing.
The failure of local anesthesia in acutely inflamed or infected tissue is among the most common and frustrating clinical experiences in emergency medicine, dentistry, and office-based procedural care. The pharmacologic explanation flows directly from the ionization equilibrium described in Section 3.1, but the clinical management implications extend beyond pharmacology into procedural technique.7 At normal tissue pH of 7.4, a local anesthetic with pKa of 7.9 (lidocaine) has approximately 24% of its molecules in the uncharged, membrane-permeable free base form, sufficient for adequate tissue penetration and neural diffusion. In acutely infected tissue, where bacterial metabolism and inflammatory mediators drive extracellular pH down to 6.8–7.0, only 6–9% of lidocaine molecules are in the free base form. This dramatically reduced fraction slows diffusion through the perineurium and nerve membrane, requiring far higher concentrations to achieve the same degree of blockade; these concentrations may approach or exceed the maximum safe dose before adequate anesthesia is achieved.7
Additionally, the intracellular pH of inflamed nerves may itself be reduced, which alters the ionization equilibrium inside the nerve and reduces the charged-form concentration available to bind the receptor. Inflammatory mediators including prostaglandins, bradykinin, and substance P also directly sensitize nociceptors and reduce the threshold for action potential generation, further opposing local anesthetic efficacy at the physiologic level.7
Several strategies exist to partially mitigate this problem. Alkalinization of local anesthetic solutions by adding sodium bicarbonate (typically 1 mEq per 10 mL of lidocaine solution) shifts the ionization equilibrium toward the free base form at the site of injection, potentially improving onset and quality of block. The effect is modest and more reliably demonstrated for epidural lidocaine than for infiltration into infected tissue, where the buffering capacity of the inflammatory exudate rapidly overcomes the alkalinization.7 Regional nerve blocks proximal to the infected area, targeting the nerve trunk at a site of normal tissue pH, are mechanically more effective than infiltrating into the infected field itself, and should be the preferred technique when anatomy permits.
Local anesthetics are intended to act locally at the site of injection, but systemic absorption is inevitable and its rate and extent are among the primary determinants of systemic toxicity risk. The rate of systemic absorption depends on the total dose administered, the vascularity of the injection site, the presence or absence of a vasoconstrictor, and the physicochemical properties of the agent.1,11 Site-dependent absorption follows a well-established hierarchy of vascularity: intravenous (intentional or accidental) > tracheal > intercostal > caudal epidural > paracervical > lumbar epidural > brachial plexus > sciatic/femoral > subcutaneous infiltration. This means that for a given total dose of drug, an intercostal block will produce substantially higher peak plasma concentrations than a femoral nerve block, a fact that must be incorporated into maximum dose calculations and clinical decision-making.11
Following systemic absorption, local anesthetics undergo distribution into well-perfused organs (lung, heart, brain) and subsequently into less perfused tissues (muscle, fat). The lung is a particularly important organ for local anesthetic pharmacokinetics: it acts as a first-pass extractor and buffer, sequestering substantial amounts of highly lipid-soluble agents during the initial distribution phase and blunting the arterial peak concentration that reaches the brain and heart.11 This "lung first-pass" effect is relevant to the risk of systemic toxicity following accidental intravenous injection; the arterial peak is blunted but not eliminated, and rapid bolus injection overwhelms even the lung's buffering capacity. Elimination of amide local anesthetics occurs primarily through hepatic metabolism; esters are hydrolyzed by plasma pseudocholinesterases and tissue esterases. These distinctions in metabolism have important implications for duration of action, toxicity risk in hepatic disease, and drug interactions, which are covered in detail in CNS-02.
The pharmacokinetic properties of local anesthetics are not merely academic: they translate directly into decisions about maximum safe dose, infusion rate, dosing interval for repeated injections, and the expected behavior of the drug in patients with altered physiology. Plasma half-life is the most clinically relevant pharmacokinetic parameter for safety purposes. For amide agents, half-lives vary significantly: lidocaine has a half-life of approximately 1.5–2 hours in healthy adults, mepivacaine 1.9–3.2 hours, bupivacaine 2.7–3.5 hours, and ropivacaine 1.8–4.2 hours. Ester agents are hydrolyzed so rapidly by plasma pseudocholinesterase that plasma half-lives are clinically negligible: chloroprocaine is measured in seconds, procaine in minutes.6
These differences become especially important during continuous peripheral nerve block infusions, where drug accumulates at a rate determined by infusion rate relative to clearance. After approximately four to five half-lives of continuous infusion, plasma concentration reaches steady state; for bupivacaine running at 10 mL/h of 0.125%, this means that steady-state plasma concentrations may not be reached for 12–18 hours, and concentrations continue rising throughout that window. In patients with reduced hepatic clearance, steady-state is both higher and later, increasing accumulation toxicity risk in ways that are not apparent from the first few hours of infusion.
The absorption hierarchy described in Section 8 has direct implications for maximum dose management. The same total milligram dose produces substantially different peak plasma concentrations depending on where it is injected. A dose of 400 mg of lidocaine injected into the intercostal space may approach toxic plasma levels; the same dose for a femoral nerve block presents far less systemic risk. This is why maximum dose guidelines are not fixed absolute ceilings but site-adjusted limits: the maximum dose for intercostal block should be well below the theoretical mg/kg ceiling, while subcutaneous infiltration can safely approach it. Clinicians performing multiple sequential blocks in a single session must account for the cumulative milligram dose across all injections, particularly when using bupivacaine or ropivacaine for procedures involving both a peripheral block and neuraxial anesthesia in the same patient.11
The concept of the therapeutic window is particularly relevant to local anesthetic pharmacokinetics because the margin between the effective perineural concentration and the systemically toxic plasma concentration is narrow for potent, long-acting agents. Bupivacaine plasma concentrations producing dense peripheral nerve block are in the range of 0.1–0.3 mcg/mL at the nerve; the central nervous system (CNS) toxic threshold begins at approximately 1.5–4 mcg/mL total plasma concentration (with a lower threshold for the unbound fraction). The practical consequence is that even moderate changes in protein binding, hepatic blood flow, or infusion rate can shift a patient from the safe to the toxic side of this window without any change in the nominal dose. Calculating doses based on lean body weight rather than total body weight in obese patients, reducing doses by 20–30% in the elderly and in patients with hepatic disease, and monitoring for early CNS symptoms during and after any significant regional anesthetic procedure are the clinical applications of these pharmacokinetic principles.11
Butterworth JF 4th, Strichartz GR. Molecular mechanisms of local anesthesia: a review. Anesthesiology. 1990;72(4):711–734
doi:10.1097/00000542-199004000-00022Catterall WA. From ionic currents to molecular mechanisms: the structure and function of voltage-gated sodium channels. Neuron. 2000;26(1):13–25
doi:10.1016/S0896-6273(00)81133-2Hille B. Ion Channels of Excitable Membranes. 3rd ed. Sunderland, MA: Sinauer Associates; 2001:385–420.
Ragsdale DS, McPhee JC, Scheuer T, Catterall WA. Molecular determinants of state-dependent block of Na⁺ channels by local anesthetics. Science. 1994;265(5179):1724–1728
doi:10.1126/science.8085162Strichartz GR. The inhibition of sodium currents in myelinated nerve by quaternary derivatives of lidocaine. J Gen Physiol. 1973;62(1):37–57
doi:10.1085/jgp.62.1.37Tucker GT. Pharmacokinetics of local anaesthetics. Br J Anaesth. 1986;58(7):717–731
doi:10.1093/bja/58.7.717Becker DE, Reed KL. Local anesthetics: review of pharmacological considerations. Anesth Prog. 2012;59(2):90–102
doi:10.2344/0003-3006-59.2.90Routledge PA. The plasma protein binding of basic drugs. Br J Clin Pharmacol. 1986;22(5):499–506
doi:10.1111/j.1365-2125.1986.tb02927.xFink BR. Mechanisms of differential axial blockade in epidural and subarachnoid anesthesia. Anesthesiology. 1989;70(5):851–858
doi:10.1097/00000542-198905000-00023Wong CA, Scavone BM, Peaceman AM, et al. The risk of cesarean delivery with neuraxial analgesia given early versus late in labor. N Engl J Med. 2005;352(7):655–665
doi:10.1056/NEJMoa041807Rosenberg PH, Veering BT, Urmey WF. Maximum recommended doses of local anesthetics: a multifactorial concept. Reg Anesth Pain Med. 2004;29(6):564–575
doi:10.1016/j.rapm.2004.08.003