Local anesthetics exert central nervous system (CNS) toxicity by inhibiting intracerebral neuronal activity, while epinephrine augments the CNS toxicity of intravenously administered local anesthetics. Viewed together, increases of extracellular concentrations of local anesthetics in the brain may be directly associated with increased CNS toxicity. The authors examined the hypothesis that epinephrine enhances the CNS toxicity of lidocaine by increasing the extracellular concentration in the brain.


An awake, spontaneously breathing rat model was used. Twenty male Sprague-Dawley rats received an intravenous infusion of lidocaine (3 mg x kg x min; group C) or lidocaine with epinephrine (3 mg x kg x min and 2 microg x kg x min, respectively; group E) for 10 min (n = 10 in each group). Effects of epinephrine on the convulsive dose and concentrations of total (protein-bound and unbound) and unbound lidocaine in plasma were examined. Concentrations of extracellular lidocaine in the cerebral nucleus accumbens were quantitatively determined by a microdialysis method.


The convulsive dose of lidocaine was significantly lower in group E than in group C (22.4 +/- 5.5 vs. 27.9 +/- 3.1 mg/kg, respectively; P < 0.05). Overall concentrations and area under the plasma concentration-versus-time curve of unbound lidocaine in group E were significantly higher than those in group C. Concentrations of extracellular lidocaine in the nucleus accumbens in group E were comparable to those of unbound fraction in plasma and were also significantly higher than those in group C.


Concomitant administration of epinephrine significantly enhanced the CNS toxicity of intravenously administered lidocaine. Increased extracellular concentration in the brain would be related to this mechanism.

EPINEPHRINE enhances the central nervous system (CNS) toxicity of intravenously administered local anesthetics1–3; however, its mechanism is still unclear. Previous studies have shown that epinephrine decreases the convulsive doses of lidocaine and bupivacaine by increasing their plasma concentrations via  peripheral vasoconstriction.1,2Other studies have demonstrated that epinephrine decreases the threshold for convulsions, i.e. , the concentrations of lidocaine in plasma and in the brain at the onset of convulsions.3In these studies, however, investigations were focused on the concentrations of total (protein-bound and unbound) fraction of local anesthetics in plasma or in the brain, and accordingly, the mechanism accounting for such an enhanced CNS toxicity by epinephrine was not elucidated. Because potentiating the excitatory neuronal activity by inhibiting intracerebral γ-aminobutyric acid–mediated neurons is closely involved in the development of the CNS toxicity of local anesthetics,4and only the unbound fraction is accessible to the neuronal cells, eventual increase of the extracellular concentration of the unbound, pharmacologically active fraction would play an important role in these situations. Nevertheless, there have been no studies directly addressing the effects of epinephrine on the extracellular concentration of lidocaine in the brain in vivo . We sustained the hypothesis that epinephrine would augment the CNS toxicity of lidocaine by increasing its intracerebral concentration, which was determined by microdialysis in awake, spontaneously breathing rats. For quantitative measurement of lidocaine in the brain, we used a retrodialysis technique with in vivo  calibration.

Animal Preparation

After approval from the Institutional Animal Care and Use Committee of the Graduate School of Medicine, Osaka City University, Osaka, Japan, 20 male Sprague-Dawley rats aged 8–10 weeks and weighing 350–400 g (CLEA Japan, Inc., Tokyo, Japan) were included in the study. During general anesthesia with intraperitoneal ketamine, the animals were placed on the stereotaxic instrument (Narishige, Tokyo, Japan), and the guide cannulas (12 mm length and 0.5 mm OD, GI-12; Eicom, Kyoto, Japan) were inserted so that the tips were placed into the nucleus accumbens (anterior 1.7 mm, lateral 1.4 mm, vertically 6.0 mm from the bregma suture). Two or 3 days later, during general anesthesia with sevoflurane, the carotid artery and the jugular vein were cannulated with polyethylene catheters to monitor mean arterial blood pressure (MAP) and heart rate (HR) and also for blood sampling and infusion of drugs. These catheters were tunneled subcutaneously to the posterior cervical region so that the animals could move freely. Before emergence from anesthesia, the microdialysis probe (A-UI-12-02, 2 mm membrane length, 0.35 mm OD, molecular weight cutoff 50 kd; Eicom) was inserted through the guide cannula and was perfused with artificial cerebrospinal fluid consisting of electrolytes (147 mm Na+, 4 mm K+, 2.3 mm Ca2+, 155.6 mm Cl) and containing 3′-hydroxylated bupivacaine (200 ng/ml). The animals were placed in a semidark, closed cardboard box to recover fully for a few hours before the experiment, and the arterial catheter was connected to a pressure transducer. MAP and HR were recorded continuously on a polygraph (RM-6000; Nihon Kohden, Tokyo, Japan).

Experimental Protocol

After preparation, the animals were randomly assigned to the control (group C) or epinephrine (group E) group (n = 10 in each group). After baseline measurement, lidocaine alone (3 mg · kg−1· min−1; group C) or lidocaine with epinephrine (3 mg · kg−1· min−1and 2 μg · kg−1· min−1, respectively; group E) was intravenously administered for 10 min. Neither of these agents contains antioxidants, and pH was 6.5 for both. MAP and HR were continuously monitored, and the animals were observed for detection of the onset of convulsions, defined as tonic–clonic movements, by one of the authors (Y.O.), who was unaware of the group allocation. Arterial blood samples (0.3–0.5 ml) were drawn before infusion of lidocaine (baseline); at the onset of partial muscle twitch preceding convulsions; and 10, 13, 16, 20, 25, 30, 40, 55, 70, 100, 130, and 160 min after the start of intravenous lidocaine infusion for measurement of blood gases and/or plasma concentrations of lidocaine. Blood was replaced with an equal volume of heparinized saline. At the onset of convulsions, supply of oxygen to the box was started to prevent convulsion-induced hypoxia and lasted for 30 min.

In Vitro  and in Vivo  Microdialysis Probe Calibration Study

For quantitative measurement of the extracellular concentration of lidocaine in the brain, we used a retrodialysis technique based on the principle that relative loss (RL) of an internal standard is related to the relative recovery (RR) of the solutions of interest.5The dialysis probe was calibrated in vitro  and in vivo  at room temperature using 3′-hydroxylated bupivacaine (200 ng/ml), a metabolite of bupivacaine with no reported CNS toxicity, as an internal standard following a method reported previously.5The in vivo  and in vitro  K  factors, defined as the ratios of RLBUPto RLLID, were 0.60 ± 0.19 and 1.04 ± 0.12, respectively.

Calibration curves for lidocaine were constructed for each run over the range of 1–100 ng (0.1–10 μg/ml), and r  2was greater than 0.999. The limit of detection was 0.5 ng. Within-day coefficients of variation of lidocaine were 2.9% and 4.8%, whereas day-to-day coefficients of variation were 3.0% and 5.2% at 1 and 10 μg/ml, respectively. RRLIDand RLBUPwere almost equal and were not affected by the extracellular concentration of lidocaine (data not shown).

Analysis of Plasma and Intracerebral Lidocaine Concentrations

Plasma concentrations of lidocaine were determined by high-performance liquid chromatography with an ultraviolet detector as reported previously.6Concentrations of unbound lidocaine in plasma were measured after ultrafiltration with a membrane (Centricon YM-30; Amicon, Inc., Beverly, MA). Concentrations of unbound lidocaine in the extracellular fluid in the brain were measured by the following microdialysis method. Indwelling microdialysis probes were perfused with the previously described cerebrospinal fluid containing 3′-hydroxylated bupivacaine at a rate of 1 μl/min by a microsyringe pump (ESP-32; Eicom). The perfusate was collected and directly injected into an electrochemical detector (HTEC-500; Eicom) every 10 min by auto-injector (EAS-20; Eicom). A mobile phase consisting of 10 mm NaH2PO4(pH 4.5) with 27% methanol (vol/vol) was delivered at a flow rate of 0.6 ml/min through a C18reversed-phase column (Eicompak SC-50DS, 2.1 × 150 mm, 5 μm particle size; Eicom), which was maintained at room temperature. Concentrations of lidocaine and 3′-hydroxylated bupivacaine in the dialysate were measured with an electrochemical detector fitted with a flow cell with a pure graphite working electrode (WE-PG; Eicom) and a salt bridge Ag–AgCl reference electrode. The working potential was set at 950 mV. The signal from the current–potential converter (the integrator output) was filtered with a low-pass in-line noise filter and integrated by a computerized data acquisition system using chromatography data software (PowerChrom; ADInstruments Pty Ltd., Castle Hill, New South Wales, Australia). The extracellular concentration of lidocaine (LIDex) was calculated as LIDex= LIDdialysate×K /RLBUP, where LIDdialysateis the concentration of lidocaine in the dialysate, and K  is RLBUP/RLLIDmeasured in vivo .7After completion of the microdialysis experiment, the animals were killed with intravenous thiopental (100 mg/kg), and brain specimens were prepared for histology to confirm the location of the microdialysis probe.

Pharmacokinetic Analysis

Moment analyses were performed on both plasma and extracellular concentrations of lidocaine in the brain to calculate the model-independent parameters by WinNonlin Professional 4.1 (Pharsight Corporation, Mountain View, CA). The midpoints of the collection time were taken as the corresponding time points to pharmacokinetic analysis and graphical presentation of the extracellular concentration of lidocaine in the brain. The terminal elimination rate constant (β) in the blood and in the brain was determined by log-linear regression, and the elimination half-life (T1/2) was determined by the following equation: T1/2= 0.693/β. The area under the plasma concentration–versus –time curve (AUC) and area under the first moment of the concentration-versus -time curve (AUMC) were calculated using the trapezoidal method with linear interpolation during increasing concentrations and log-linear interpolation during decreasing concentrations. Clearance was calculated as the dose of lidocaine (30 mg/kg) divided by AUC, whereas mean residential time was obtained as the ratio of AUMC to AUC. The volume of distribution at steady state (VdSS) was calculated according to the following equation8:

The same equations were used for calculating the apparent clearance and apparent volume of distribution of extracellular lidocaine in the brain, where F is the bioavailability from the blood to extracellular fluid in the brain.

Statistical Analysis

All values are expressed as mean ± SD. The number of animals in each group was determined based on our previous study,6in which the convulsive dose of lidocaine was 31.6 ± 6.0 mg/kg. Based on the formula for normal theory and assuming a type I error protection of 0.05 and a power of 0.80 allowing us to detect a 25% change in the convulsive dose, 10 animals were required for each group.

Statistical analysis was performed using StatView 5.0 (SAS Institute Inc., Cary, NC) or SigmaStat 3.0 (Systat Software Inc., Richmond, CA). The frequency of the convulsion between the two groups was evaluated using a chi-square test. Differences in plasma and intracerebral concentrations of lidocaine between the two groups were examined using analysis of variance for repeated measures and were tested using an unpaired t  test for each time point. The convulsive dose of lidocaine was examined by unpaired t  test, assuming that it was 30 mg/kg when convulsions were not induced during 10 min of lidocaine infusion. Pharmacokinetic parameters between the two groups were examined by unpaired t  test. Within-group differences in MAP and HR and blood gases were tested by one-way analysis of variance. Differences in MAP and HR between groups at the same time point were examined by unpaired t  test. Values were considered significant when P  was less than 0.05.

Two rats in group E died of cardiac arrest after the onset of convulsions during infusion of lidocaine and epinephrine, and they were replaced by two additional animals. There were no differences in MAP or HR between the two groups at baseline. In group E, both MAP and HR significantly increased after the start of infusion of lidocaine with epinephrine and were significantly higher than in group C during infusion. After termination of the infusion, both MAP and HR decreased in group E, and there were no differences in these values between groups C and E (fig. 1). There were no differences in arterial blood gas data at baseline, before convulsions, and at the end of experiments within the same study group or between the two groups (table 1).

The number of rats with convulsions in group E was significantly larger than in group C (10 and 6, respectively; P < 0.05). The convulsive dose of lidocaine in group E was significantly smaller than in group C (22.4 ± 5.5 and 27.9 ± 3.1 mg/kg, respectively; P < 0.05). Plasma concentrations of total (protein-bound and unbound) lidocaine were similar during the experiment in the two groups, and there were no differences in the pharmacokinetic parameters between these groups (fig. 2and table 2). Overall concentrations of unbound lidocaine in group E were significantly higher than in group C (P < 0.01; fig. 2). The AUC of unbound lidocaine in group E was significantly larger than in group C, which would result from a smaller volume of distribution (table 2). In vivo  microdialysis study revealed that extracellular concentrations of lidocaine in the brain reached the highest level at 25 min after the start of lidocaine infusions in both groups (fig. 2). Overall concentrations of lidocaine in group E were significantly higher than in group C and were also significantly higher in group E than in group C at 25 and 35 min after the start of lidocaine infusion (11.3 ± 3.6 vs.  7.7 ± 3.5 and 8.5 ± 3.5 vs.  5.3 ± 2.0 μg/ml, respectively; P < 0.05). Except peak concentrations, there were no pharmacokinetic differences in the concentration of extracellular lidocaine between groups C and E (table 3).

In the current study, we have shown that epinephrine significantly increased the concentration of extracellular lidocaine in the brain. Because it was comparable to the unbound fraction in the peripheral blood vessels as reported previously,9,10our results suggest that only the unbound, pharmacologically active fraction is transported to the intracerebral neuronal cells, and an increased unbound fraction in the brain would account for the epinephrine-increased convulsive potency of lidocaine. To our knowledge, this is the first study showing the relations between the concentration of lidocaine in the brain and its CNS toxicity.

Our current findings, including a decreased convulsive dose of lidocaine and an unchanged threshold plasma concentration for convulsions induced by epinephrine, are consistent with those reported previously.1,2However, pharmacokinetic parameters such as clearance, elimination half-time, and peak plasma concentration of total lidocaine were not affected by epinephrine. Our results indicated that addition of epinephrine was associated with more rapid increase of the plasma concentration of lidocaine and more rapid decrease after the end of infusion. These changes produced by epinephrine would be attributed to the decreased volume of distribution during infusion and increased cardiac output, hepatic blood flow, and resultant increased biotransformation in the liver after the end of infusion.11Increased influx of lidocaine to the brain by increased cerebral blood flow, facilitated by transport of unbound lidocaine from blood vessels to extracellular fluid in the brain, would also contribute to this epinephrine-induced increase in concentration of lidocaine in the brain.2Increased MAP, HR, intracranial pressure, or acidosis induced by convulsions may also be responsible for this effect.12 

We have measured the concentrations of extracellular lidocaine in the nucleus accumbens, which predominantly consists of γ-aminobutyric acid–mediated neurons and plays an important role in controlling convulsions.13,14The relations between concentrations of lidocaine in the nucleus accumbens and other cerebral regions were not examined in the current study. Moreover, the concentration of local anesthetics in various regions in the brain may depend on regional blood flow, which is influenced by the development of convulsions.15However, a previous microdialysis study revealed no differences in the concentrations of lidocaine in the extracellular fluid in different parts of the brain or in the bilateral hemisphere after intraarterial administration.16Our preliminary study also showed that there were no differences in the concentrations of lidocaine in various parts (cortex, hippocampus, limbic system, olfactory bulb, cerebellum, and medulla) of the brain, and convulsions were not induced by topical injection of lidocaine into the nucleus accumbens (data not shown). These findings indicate that an increase in the concentration of lidocaine in other parts as well as in the nucleus accumbens would contribute to triggering convulsions.

There are several limitations in our study. First, some rats showed convulsions and others did not in group C, although there were no differences in the concentrations of lidocaine in the brain among rats with and without convulsions (data not shown). This might be attributable to an insufficient number of rats (n = 10). Also, the detailed mechanism for the increase of CNS toxicity of lidocaine by the increased unbound fraction is still unclear. Further study, including altering protein binding, would be required. Second, the concentration of lidocaine in the cerebral dialysate was determined at collection intervals every 10 min. The concentrations of lidocaine in the brain would also be rapidly changed during and just after the end of intravenous infusion, which prevented us from detecting the real peak concentrations. Once given simultaneously with epinephrine, concentrations of lidocaine in the brain increased more rapidly, as was the case with plasma. Despite these limitations, pharmacokinetic analyses suggested that the increased extracellular concentration of lidocaine would significantly contribute to the augmented toxicity by epinephrine.

In conclusion, concomitant administration of epinephrine increased the unbound fraction of intravenously administered lidocaine, leading to a significant increase in its concentration in the brain. This would be related to the increased CNS toxicity after intravenous administration of lidocaine together with epinephrine.

The authors thank Tomoharu Funao, M.D. (Assistant Professor, Department of Anesthesiology and Intensive Care Medicine, Graduate School of Medicine, Osaka City University, Osaka, Japan), for technical assistance.

Bernards CM, Carpenter RL, Kenter ME, Brown DL, Rupp SM, Thompson GE: Effect of epinephrine on central nervous system and cardiovascular system toxicity of bupivacaine in pigs. Anesthesiology 1989; 71:711–7
Yamauchi Y, Kotani J, Ueda Y: The effects of exogenous epinephrine on a convulsive dose of lidocaine: Relationship with cerebral circulation. J Neurosurg Anesthesiol 1998; 10:178–87
Yokoyama M, Hirakawa M, Goto H: Effect of vasoconstrictive agents added to lidocaine on intravenous lidocaine-induced convulsions in rats. Anesthesiology 1995; 82:574–80
Ikeda M, Dohi T, Tsujimoto A: Protection from local anesthetic-induced convulsions by gamma-aminobutyric acid. Anesthesiology 1982; 56:365–8
Scheller D, Kolb J: The internal reference technique in microdialysis: A practical approach to monitoring dialysis efficiency and to calculating tissue concentration from dialysate samples. J Neurosci Methods 1991; 40:31–8
Funao T, Oda Y, Tanaka K, Asada A: The P-glycoprotein inhibitor quinidine decreases the threshold for bupivacaine-induced, but not lidocaine-induced, convulsions in rats. Can J Anaesth 2003; 50:805–11
Clement R, Malinovsky JM, Dollo G, Le Corre P, Chevanne F, Le Verge R: In vitro  and in vivo  microdialysis calibration using retrodialysis for the study of the cerebrospinal distribution of bupivacaine. J Pharm Biomed Anal 1998; 17:665–70
Dyck JB, Maze M, Haack C, Vuorilehto L, Shafer SL: The pharmacokinetics and hemodynamic effects of intravenous and intramuscular dexmedetomidine hydrochloride in adult human volunteers. Anesthesiology 1993; 78:813–20
Bagger M, Bechgaard E: A microdialysis model to examine nasal drug delivery and olfactory absorption in rats using lidocaine hydrochloride as a model drug. Int J Pharm 2004; 269:311–22
Marathe PH, Shen DD, Artru AA, Bowdle TA: Effect of serum protein binding on the entry of lidocaine into brain and cerebrospinal fluid in dogs. Anesthesiology 1991; 75:804–12
Stenson RE, Constantino RT, Harrison DC: Interrelationships of hepatic blood flow, cardiac output, and blood levels of lidocaine in man. Circulation 1971; 43:205–11
Simon RP, Benowitz NL, Bronstein J, Jacob P III: Increased brain uptake of lidocaine during bicuculline-induced status epilepticus in rats. Neurology 1982; 32:196–9
Whittington RA, Virag L, Vulliemoz Y, Cooper TB, Morishima HO: Dexmedetomidine increases the cocaine seizure threshold in rats. Anesthesiology 2002; 97:693–700
Pennartz CM, Groenewegen HJ, Lopez da Silva FH: The nucleus accumbens as a complex of functionally distinct neuronal ensembles: An integration of behavioural, electrophysiological and anatomical data. Prog Neurobiol 1994; 42:719–61
Chassagnon S, de Vasconcelos AP, Ferrandon A, Koning E, Marescaux C, Nehlig A: Time course and mapping of cerebral perfusion during amygdala secondarily generalized seizures. Epilepsia 2005; 46:1178–87
Chou K-J, Donovan MD: Lidocaine distribution into the CNS following nasal and arterial delivery: A comparison of local sampling and microdialysis techniques. Int J Pharm 1998; 171:53–61