- Email: [email protected]
Systemic lidocaine decreases the Bispectral Index in the presence of midazolam, but not its absence
Journal of Clinical Anesthesia (2012) 24, 121–125
Original contribution
Systemic lidocaine decreases the Bispectral Index in the presence of midazolam, but not its absence☆,☆☆,★ Antje Gottschalk MD (Research Fellow)a,b , Allannah M. McKay MD (Research Fellow)a , Zahra M. Malik MS (Medical Student)a , Michael Forbes MS (Medical Student)a , Marcel E. Durieux MD, PhD (Professor of Anesthesiology and of Neurological Surgery)a , Danja S. Groves MD, PhD (Associate Professor of Anesthesiology)a,⁎ a
Department of Anesthesia, University of Virginia, Charlottesville, VA 22908-0710, USA Department of Anesthesiology and Intensive Care Medicine, University of Muenster, Muenster 48149, Germany
b
Received 27 November 2009; revised 27 June 2011; accepted 30 June 2011
Keywords: Bispectral Index (BIS); Lidocaine; Midazolam
Abstract Study Objective: To evaluate the effects of intravenous (IV) lidocaine on the Bispectral Index (BIS) in the presence or absence of midazolam. Design: Prospective, randomized, double-blinded, placebo-controlled clinical study. Setting: Operating room of a university hospital. Patients: 96 ASA physical status 1, 2, and 3 patients undergoing general anesthesia. Interventions: Patients were assigned to one of 6 treatment groups to receive IV midazolam (0.03 mg/kg) or placebo, followed 5 minutes later by one of three IV preinduction doses of lidocaine: 0.5, 1.0, or 1.5 mg/kg. Measurements: BIS values were recorded before administration of lidocaine and at 30-second intervals afterwards for three minutes. The primary endpoint was the average BIS level recorded. Main Results: Baseline BIS values were lower in the midazolam group (94 ± 4 vs. 90 ± 7, P b 0.001). There was no significant decrease in BIS values in the placebo group for any of the three lidocaine doses. However, in the midazolam groups, significant decreases in BIS levels versus baseline values were measured. Conclusion: IV lidocaine decreases BIS in the presence of midazolam, suggesting that the effect of lidocaine on BIS is not direct, but rather results from modulation by midazolam. © 2012 Elsevier Inc. All rights reserved.
1. Introduction ☆
Funding: The study was solely supported by departmental funding. ☆☆ Conflict of interest: The authors do not have any conflict of interest. ★ Reprints will not be available from the authors. ⁎ Correspondence: Danja S. Groves, MD, PhD, Department of Anesthesia, University of Virginia, P.O. Box 800710, Charlottesville, VA 22908-0710, USA. Tel.: 001-434-924-2283; fax: 001-434-982-0019. E-mail address: [email protected] (D.S. Groves). 0952-8180/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.jclinane.2011.06.018
Intraoperative systemic lidocaine shortens postoperative ileus duration and hospital stay [1]. However, systemic lidocaine appears to affect the Bispectral Index (BIS): intravenous (IV) lidocaine administered during propofol or sevoflurane anesthesia reduces anesthetic doses required to
122 achieve target BIS values [1,2], and an inadvertently high dose of lidocaine decreased BIS to 0 during sevoflurane anesthesia [3]. Even intrathecal lidocaine leads to sedation and a decrease in BIS. This effect was not related to block height [4]. However, Ozkan-Seyhan et al showed that intrathecal bupivacaine may potentiate the effects of intravenous sedation. They reported that a high spinal block (T1-T4) was associated with a faster onset, delayed recovery, and lower doses of propofol sedation compared with a low spinal block (T9-T11) [5]. Nevertheless, it is not clear whether systemic lidocaine potentiates the effects of other anesthetics. Therefore, we studied the interaction between IV lidocaine and BIS in the absence of a general anesthetic. We hypothesized that lidocaine itself may not affect BIS. If the effect of lidocaine on BIS changes during general anesthesia is indeed indirect, the fact that it occurred with both sevoflurane and propofol suggests that it may potentiate effects of gamma-aminobutyric acid (GABAergic) compounds in general [1,2]. Therefore, we hypothesized that lidocaine would affect BIS in the presence of midazolam. We tested this hypothesis by studying the interaction between IV lidocaine and BIS against a background of a sedative of IV midazolam.
2. Materials and methods 2.1. Study population University of Virginia Institutional Review Board approval was granted for the trial. A sample size of 96 was determined based on an effect size of 5, standard deviation of 5, with a power of 0.8, based on data from a retrospective chart review.1 Informed consent was obtained from 97 adult, ASA physical status 1, 2, and 3 patients scheduled to undergo IV induction of general anesthesia. Exclusion criteria included emergency surgery, cardiovascular instability or the presence of an unstable cardiac rhythm, weight greater than 100 kg, concurrent pregnancy, allergy to midazolam or lidocaine, or acute or chronic use of drugs affecting the GABAergic system. Subjects were randomized to receive either IV midazolam (0.03 mg/kg) or IV placebo, followed by one of three lidocaine doses: 0.5 mg/kg, 1.0 mg/kg, or 1.5 mg/kg (Table 1). Hence, 6 groups of 16 subjects were studied.
A. Gottschalk et al. Table 1
Group assignment
Group
Premedication
Lidocaine dose (mg/kg)
1 2 3 4 5 6
placebo placebo placebo 0.03 mg/kg of midazolam 0.03 mg/kg of midazolam 0.03 mg/kg of midazolam
0.5 1.0 1.5 0.5 1.0 1.5
unblinded and administered the study drugs. After informed consent was obtained by research staff, the anesthesia provider received a sealed envelope containing the randomized treatment assignment for that subject, and consequently administered a midazolam or placebo premedication, approximately 5 minutes before induction. In the operating room, routine monitors – noninvasive blood pressure, pulse oximeter, and electrocardiogram – were placed. A BIS device (A-2000 XP with BIS Quatro adult sensor; Aspect Medical Systems, Norwood, MA, USA) was also attached to all patients. Baseline BIS values, blood pressure (BP), and heart rate (HR) values were recorded. Subjects were then given the indicated dose of lidocaine as a slow bolus injection through a free-running IV catheter. BIS, BP, and HR values were recorded every 30 seconds for the following three minutes. At the end of three minutes, patients were asked if they had experienced any side effects from the lidocaine (eg, ringing in the ears, metallic taste in mouth, perioral tingling, or numbness). General anesthesia was then induced with propofol. The frequency of marked sedation before propofol and the frequency of obvious discomfort on injection of propofol were also recorded.
2.3. Statistical analysis The mean BIS values recorded during the three minutes after the injection of lidocaine for each subject were compared with baseline using a two-tailed paired t-test. Baseline data from the placebo group and the midazolam group were compared using a two-tailed unpaired t-test. Systolic (SBP) and diastolic (DBP) blood pressure and HR were compared between the midazolam and placebo groups using a two-tailed unpaired t-test. Data are presented as means ± SD unless otherwise indicated.
2.2. Protocol
3. Results
BIS values were recorded by a study-blinded member of the research staff. The anesthesia provider was
3.1. Study population
1
Groves DS, Malik ZM, Durieux M. Midazolam modulates effects of intravenous lidocaine on bispectral index (BIS) [Abstract]. Anesthesiology 2007;107:A803.
Demographic data were comparable between groups (Table 2). There was a significant difference in SBP and DBP, but not HR, between the placebo and midazolam groups before administration of lidocaine.
Systemic lidocaine and BIS Table 2
123
Demographic data and vital signs
Age (yrs) Gender (% male) Race (n in each group) African-American White Hispanic Weight (kg) BP before lidocaine injection (mmHg) HR before lidocaine injection (bpm)
Placebo (Groups 1-3) means ± SD
Midazolam (Groups 4-6) mean ± SD
P-value
59 ± 15 57
54 ± 13 50
P = 0.11 P = 0.86
6 40 1 77 ± 15 151 ± 27/82 ± 12
3 42 1 75 ± 15 127 ± 19/73 ± 10
P = 0.58
74 ± 14
74 ± 12
3.2. Effect of lidocaine on BIS In the placebo group, we observed no significant effect of any of the three lidocaine doses on BIS values during the three minutes after administration. When subjects were premedicated with midazolam, there was a significant decrease in their baseline BIS values as compared with the placebo group (90 ± 7 vs. 94 ± 4, P b 0.001). Furthermore, in the midazolam group, we noted a significant decrease in the average BIS value, as compared with baseline, after lidocaine administration. This finding was most evident in those patients receiving 1.0 mg/kg (decrease to 84 ± 7, P b 0.001). Effects were less but were also present in those receiving 0.5 mg/kg (decrease to 85 ± 9, P = 0.03), and in those receiving 1.5 mg/kg (decrease to 86 ± 8, P = 0.0031). Fig. 1 presents these data as percentage changes.
P = 0.38 P b 0.0001 for both SBP and DBP P = 0.96
midazolam. Eight subjects showed marked sedation, all of whom had received midazolam. Again, this finding was not dependent on the lidocaine dose.
3.4. Adverse events No serious cardiac dysrhythmia or seizure activity was noted during the study. One subject suffered an adverse event after receiving midazolam plus 1.5 mg/kg of lidocaine. She became markedly sedated and subsequently hypoxic 90 seconds after receiving lidocaine, necessitating bag-mask ventilation. No BIS values were recorded, and she was withdrawn from the study. It is noteworthy that this subject weighed 99 kg, which placed her at the upper limit of acceptance to the study. Her body mass index was 37 kg/m2; thus, a 0.03 mg/kg dose of midazolam may have been too high for her body composition.
3.3. Side effects The frequency of perioral numbness or tingling was greatest in patients receiving placebo and 1.5 mg/kg lidocaine (8/16 subjects). Eight of 97 subjects complained of discomfort on propofol injection, which was not influenced by either the lidocaine dose or the presence of
% change in BIS from baseline
% change in BIS after lidocaine, ± midazolam midazolam placebo 0 -1 -2 -3 -4 -5 -6 -7 * -8 * * -9 -10 means±SD; *P < 0.05 vs baseline.
placebo+0.5mg/kg lidocaine placebo+1.0mg/kg lidocaine placebo+1.5mg/kg lidocaine midazolam+0.5mg/kg lidocaine midazolam+1.0mg/kg lidocaine midazolam+1.5mg/kg lidocaine
Fig. 1 Changes (in %) in Bispectral Index (BIS) after lidocaine with and without midazolam, compared with baseline.
4. Discussion Our results suggest that systemic lidocaine affects BIS values in the presence – but not absence – of midazolam. As lidocaine affects BIS in the presence of propofol or sevoflurane anesthesia, it is suggested that the effect of lidocaine on BIS is indirect and occurs because the local anesthetic modulates GABAergic signaling.
4.1. Limitations Our study has some limitations. First, plasma concentrations of lidocaine were not measured. However, as we were studying the effect of lidocaine on BIS, ie, on cerebral function, it was really the concentration of lidocaine at the brain that was of interest. A study in 65 patients that sought to characterize the time course of lidocaine concentrations in the plasma and cerebrospinal fluid (CSF) after an IV bolus of 2.0 mg/kg lidocaine did not show any statistical correlation between plasma and CSF lidocaine levels [6]. Thus, to have measured
124 plasma lidocaine concentrations would not necessarily have reflected CSF concentration (let alone brain concentration) of lidocaine during the 3-minute recording time; we therefore decided not to subject the patients to an additional venous puncture to obtain these data. Another limitation is the short time period of BIS recording. It is possible that an effect of lidocaine on BIS might have been seen in the placebo group if recording had continued for longer than three minutes. A longer recording time also might have provided information on the duration of any effect of a bolus of lidocaine on BIS. However, a decrease in BIS was evident at 90 to 120 seconds, and BIS values began to level out after this time (data not shown), suggesting that longer measurement periods would have had limited value. In another clinical study [7], the peak effect of a bolus of 1.5 mg/kg of lidocaine (suppression of all airway reflexes except the apneic reflex) was noted at two minutes; the authors surmised that this finding also was due to a central effect of the lidocaine. Since the purpose of preinduction lidocaine administration is to diminish pain during propofol injection and limit hemodynamic responses to intubation, we did not want to extend the observation period any further. We used mean BIS values over the three minutes following administration of lidocaine rather than the lowest BIS value recorded during the 3-minute interval (which would perhaps have indicated the “peak effect” of lidocaine) or a measurement taken at a specific time point. On analysis of the data, it appeared that the maximum effect was generally seen at 90 seconds after lidocaine injection, but it varied between individuals. The time to peak effect is dependent, among other things, on cardiac output and cerebral blood flow (CBF). It was thought that to stipulate a specific recording time to measure peak effect for each subject might miss the peak effect in some subjects. The lowest BIS value for each individual was noted, whether it occurred at a 30-second interval or not. However, BIS recordings tended to fluctuate, resulting in several “peaks and valleys”. Peak values were very brief at times. A fluctuating record suggests that BIS does not simply reflect brain concentrations of lidocaine. We felt that by taking a mean value of recordings rather than the lowest recording for each subject, we were likely to have a better representation of the BIS values following lidocaine injection. Another confounding factor is BP fluctuation. Hypotension may decrease CBF, which results in somnolence. This situation could potentially limit the interpretation of the study results. However, the comparison of BP values within each group during the study period did not detect any significant differences.
4.2. Effect of lidocaine on anesthesia We hypothesized that IV lidocaine would not have an effect on BIS in the absence of midazolam, and indeed we did not find an effect on BIS by lidocaine alone in our study.
A. Gottschalk et al. When lidocaine was administered after premedication with midazolam, a GABAergic agonist, a significant decrease in BIS recordings was observed. These results are consistent with those of previous clinical studies that noted an effect of local anesthetic agents on BIS in the presence and absence of other GABAergic agents. Kaba et al studied the effects of IV lidocaine on acute rehabilitation after laparoscopic colectomy [1]. These investigators found that IV infusion of lidocaine reduced the amount of sevoflurane required to maintain BIS scores during surgery, compared with placebo. Patients who received lidocaine or bupivacaine intramuscularly (IM) during propofol anesthesia required less propofol to achieve equal BIS values [2]. In that study, IM bupivacaine, lidocaine, or saline was given 10 minutes before induction. There was no significant difference in BIS between the groups prior to local anesthetic administration, or preinduction, ie, no effect of the local anesthetic itself on BIS prior to the administration of propofol. In contrast, IM administration of the local anesthetics was associated with a decrease in both the induction and maintenance doses of propofol. However, such effects have not been found in all settings. A recent study comparing sevoflurane anesthesia alone, sevoflurane anesthesia plus an IV lidocaine bolus followed by infusion, and sevoflurane anesthesia plus epidural lidocaine, showed a reduction in sevoflurane requirement in the epidural group, as measured by BIS, but not in the IV lidocaine group, compared with sevoflurane alone, after 10 minutes [8]. This finding could be due to indirect effects of deafferentation as well as direct supraspinal effects.
4.3. Lidocaine and GABA signaling Given that IV lidocaine decreases BIS values in the presence of midazolam, sevoflurane, and propofol, but does not affect BIS recordings when given alone, we speculate that lidocaine exerts its effect on BIS by modulation of the effects of drugs acting at the gamma-aminobutyric acid-A (GABAA) receptor. Previous studies have looked at the effects of lidocaine on the GABAA receptor in vitro. In most of these studies, lidocaine inhibited the GABA-induced current [9-11]. However, Nordmark et al [12] studied the effect of local anesthetics on the GABA-gated chloride channel in crayfish stretch receptor neurons. They found that lidocaine increased GABA-induced chloride conductance when applied in concentrations of the same order as those required to block the nerve impulse, and showed evidence to support that this was due to a decrease in GABA uptake. The fact that some of the above-cited studies showed an inhibitory effect of lidocaine on GABAA receptor function, while other studies, in different species, showed a lidocaine-induced enhancement of GABAergic current, is not necessarily contradictory [12]. Clinically, local anesthetic agents are anticonvulsant at low doses and proconvulsant at high doses [13]. A biphasic response at the GABAA receptor has been observed with
Systemic lidocaine and BIS volatile anesthetics and barbiturates – potentiation occurring at lower concentrations and inhibition occurring at higher concentrations [14]. These studies showed that lidocaine was able to interact with the GABAA receptor. It is possible that lidocaine and midazolam together have a positive heterotropic interaction at the GABAA receptor, allosterically modifying the receptor to favor the open state of the ion channel. This interaction, however, has not been studied. Alternatively, given that in one study 10 μM concentrations of lidocaine (a concentration thought to be “clinically relevant”) enhanced glycine receptor function but did not affect GABAA receptor function [9], it is possible that the addition of midazolam, with its consequent agonistic effects at the GABAA receptor, has an additive effect with lidocaine on other receptors [ie, acid-sensing ion channel or hyperpolarization-activated cyclic nucleotide gated (HCN) channels] [15,16], which together result in decreased BIS values.
4.4. Summary Intravenous lidocaine reduces BIS readings in the presence of midazolam but not in its absence. We suggest that the local anesthetic modulates the effects of GABAergic drugs on BIS, rather than affecting BIS itself. Whether a decrease in BIS, as noted under these circumstances, indicates a decrease in level of consciousness – in other words, whether BIS is still a valid measure of depth of sedation in the presence of lidocaine – remains to be shown.
References [1] Kaba A, Laurent SR, Detroz BJ, et al. Intravenous lidocaine infusion facilitates acute rehabilitation after laparoscopic colectomy. Anesthesiology 2007;106:11-8.
125 [2] Senturk M, Pembeci K, Menda F, et al. Effects of intramuscular administration of lidocaine or bupivacaine on induction and maintenance doses of propofol evaluated by bispectral index. Br J Anaesth 2002;89:849-52. [3] Gaughen CM, Durieux M. The effect of too much intravenous lidocaine on bispectral index. Anesth Analg 2006;103:1464-5. [4] Pollock JE, Neal JM, Liu SS, Burkhead D, Polissar N. Sedation during spinal anesthesia. Anesthesiology 2000;93:728-34. [5] Ozkan-Seyhan T, Sungur MO, Senturk E, et al. BIS guided sedation with propofol during spinal anaesthesia: influence of anaesthetic level on sedation requirement. Br J Anaesth 2006;96:645-9. [6] Tsai PS, Buerkle H, Huang LT, Lee TC, Yang LC, Lee JH. Lidocaine concentrations in plasma and cerebrospinal fluid after systemic bolus administration in humans. Anesth Analg 1998;87:601-4. [7] Nishino T, Hiraga K, Sugimori K. Effects of i.v. lignocaine on airway reflexes elicited by irritation of the tracheal mucosa in humans anaesthetized with enflurane. Br J Anaesth 1990;64:682-7. [8] Hodgson PS, Liu SS. Epidural lidocaine decreases sevoflurane requirement for adequate depth of anesthesia as measured by the Bispectral Index monitor. Anesthesiology 2001;94:799-803. [9] Hara K, Sata T. The effects of the local anesthetics lidocaine and procaine on glycine and gamma-aminobutyric acid receptors expressed in Xenopus oocytes. Anesth Analg 2007;104:1434-9. [10] Sugimoto M, Uchida I, Fukami S, Takenoshita M, Mashimo T, Yoshiya I. The alpha and gamma subunit-dependent effects of local anesthetics on recombinant GABA(A) receptors. Eur J Pharmacol 2002;401:329-37. [11] Ye JH, Ren J, Krnjević K, Liu PL, McArdle JJ. Cocaine and lidocaine have additive inhibitory effects on the GABAA current of acutely dissociated hippocampal pyramidal neurons. Brain Res 1999;821: 26-32. [12] Nordmark J, Rydqvist B. Local anaesthetics potentiate GABAmediated Cl-currents by inhibiting GABA uptake. Neuroreport 1997;8:465-8. [13] Modica PA, Tempelhoff R, White PF. Pro- and anticonvulsant effects of anesthetics (Part II). Anesth Analg 1990;70:433-44. [14] Dilger JP. The effects of general anaesthetics on ligand-gated ion channels. Br J Anaesth 2002;89:41-51. [15] Lin J, Chu X, Maysami S, et al. Inhibition of acid sensing ion channel currents by lidocaine in cultured mouse cortical neurons. Anesth Analg 2011;112:977-81. [16] Meng QT, Xia ZY, Liu J, Bayliss DA, Chen X. Local anesthetic inhibits hyperpolarization-activated cationic currents. Mol Pharmacol 2011;79:866-73.
Original contribution
Systemic lidocaine decreases the Bispectral Index in the presence of midazolam, but not its absence☆,☆☆,★ Antje Gottschalk MD (Research Fellow)a,b , Allannah M. McKay MD (Research Fellow)a , Zahra M. Malik MS (Medical Student)a , Michael Forbes MS (Medical Student)a , Marcel E. Durieux MD, PhD (Professor of Anesthesiology and of Neurological Surgery)a , Danja S. Groves MD, PhD (Associate Professor of Anesthesiology)a,⁎ a
Department of Anesthesia, University of Virginia, Charlottesville, VA 22908-0710, USA Department of Anesthesiology and Intensive Care Medicine, University of Muenster, Muenster 48149, Germany
b
Received 27 November 2009; revised 27 June 2011; accepted 30 June 2011
Keywords: Bispectral Index (BIS); Lidocaine; Midazolam
Abstract Study Objective: To evaluate the effects of intravenous (IV) lidocaine on the Bispectral Index (BIS) in the presence or absence of midazolam. Design: Prospective, randomized, double-blinded, placebo-controlled clinical study. Setting: Operating room of a university hospital. Patients: 96 ASA physical status 1, 2, and 3 patients undergoing general anesthesia. Interventions: Patients were assigned to one of 6 treatment groups to receive IV midazolam (0.03 mg/kg) or placebo, followed 5 minutes later by one of three IV preinduction doses of lidocaine: 0.5, 1.0, or 1.5 mg/kg. Measurements: BIS values were recorded before administration of lidocaine and at 30-second intervals afterwards for three minutes. The primary endpoint was the average BIS level recorded. Main Results: Baseline BIS values were lower in the midazolam group (94 ± 4 vs. 90 ± 7, P b 0.001). There was no significant decrease in BIS values in the placebo group for any of the three lidocaine doses. However, in the midazolam groups, significant decreases in BIS levels versus baseline values were measured. Conclusion: IV lidocaine decreases BIS in the presence of midazolam, suggesting that the effect of lidocaine on BIS is not direct, but rather results from modulation by midazolam. © 2012 Elsevier Inc. All rights reserved.
1. Introduction ☆
Funding: The study was solely supported by departmental funding. ☆☆ Conflict of interest: The authors do not have any conflict of interest. ★ Reprints will not be available from the authors. ⁎ Correspondence: Danja S. Groves, MD, PhD, Department of Anesthesia, University of Virginia, P.O. Box 800710, Charlottesville, VA 22908-0710, USA. Tel.: 001-434-924-2283; fax: 001-434-982-0019. E-mail address: [email protected] (D.S. Groves). 0952-8180/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.jclinane.2011.06.018
Intraoperative systemic lidocaine shortens postoperative ileus duration and hospital stay [1]. However, systemic lidocaine appears to affect the Bispectral Index (BIS): intravenous (IV) lidocaine administered during propofol or sevoflurane anesthesia reduces anesthetic doses required to
122 achieve target BIS values [1,2], and an inadvertently high dose of lidocaine decreased BIS to 0 during sevoflurane anesthesia [3]. Even intrathecal lidocaine leads to sedation and a decrease in BIS. This effect was not related to block height [4]. However, Ozkan-Seyhan et al showed that intrathecal bupivacaine may potentiate the effects of intravenous sedation. They reported that a high spinal block (T1-T4) was associated with a faster onset, delayed recovery, and lower doses of propofol sedation compared with a low spinal block (T9-T11) [5]. Nevertheless, it is not clear whether systemic lidocaine potentiates the effects of other anesthetics. Therefore, we studied the interaction between IV lidocaine and BIS in the absence of a general anesthetic. We hypothesized that lidocaine itself may not affect BIS. If the effect of lidocaine on BIS changes during general anesthesia is indeed indirect, the fact that it occurred with both sevoflurane and propofol suggests that it may potentiate effects of gamma-aminobutyric acid (GABAergic) compounds in general [1,2]. Therefore, we hypothesized that lidocaine would affect BIS in the presence of midazolam. We tested this hypothesis by studying the interaction between IV lidocaine and BIS against a background of a sedative of IV midazolam.
2. Materials and methods 2.1. Study population University of Virginia Institutional Review Board approval was granted for the trial. A sample size of 96 was determined based on an effect size of 5, standard deviation of 5, with a power of 0.8, based on data from a retrospective chart review.1 Informed consent was obtained from 97 adult, ASA physical status 1, 2, and 3 patients scheduled to undergo IV induction of general anesthesia. Exclusion criteria included emergency surgery, cardiovascular instability or the presence of an unstable cardiac rhythm, weight greater than 100 kg, concurrent pregnancy, allergy to midazolam or lidocaine, or acute or chronic use of drugs affecting the GABAergic system. Subjects were randomized to receive either IV midazolam (0.03 mg/kg) or IV placebo, followed by one of three lidocaine doses: 0.5 mg/kg, 1.0 mg/kg, or 1.5 mg/kg (Table 1). Hence, 6 groups of 16 subjects were studied.
A. Gottschalk et al. Table 1
Group assignment
Group
Premedication
Lidocaine dose (mg/kg)
1 2 3 4 5 6
placebo placebo placebo 0.03 mg/kg of midazolam 0.03 mg/kg of midazolam 0.03 mg/kg of midazolam
0.5 1.0 1.5 0.5 1.0 1.5
unblinded and administered the study drugs. After informed consent was obtained by research staff, the anesthesia provider received a sealed envelope containing the randomized treatment assignment for that subject, and consequently administered a midazolam or placebo premedication, approximately 5 minutes before induction. In the operating room, routine monitors – noninvasive blood pressure, pulse oximeter, and electrocardiogram – were placed. A BIS device (A-2000 XP with BIS Quatro adult sensor; Aspect Medical Systems, Norwood, MA, USA) was also attached to all patients. Baseline BIS values, blood pressure (BP), and heart rate (HR) values were recorded. Subjects were then given the indicated dose of lidocaine as a slow bolus injection through a free-running IV catheter. BIS, BP, and HR values were recorded every 30 seconds for the following three minutes. At the end of three minutes, patients were asked if they had experienced any side effects from the lidocaine (eg, ringing in the ears, metallic taste in mouth, perioral tingling, or numbness). General anesthesia was then induced with propofol. The frequency of marked sedation before propofol and the frequency of obvious discomfort on injection of propofol were also recorded.
2.3. Statistical analysis The mean BIS values recorded during the three minutes after the injection of lidocaine for each subject were compared with baseline using a two-tailed paired t-test. Baseline data from the placebo group and the midazolam group were compared using a two-tailed unpaired t-test. Systolic (SBP) and diastolic (DBP) blood pressure and HR were compared between the midazolam and placebo groups using a two-tailed unpaired t-test. Data are presented as means ± SD unless otherwise indicated.
2.2. Protocol
3. Results
BIS values were recorded by a study-blinded member of the research staff. The anesthesia provider was
3.1. Study population
1
Groves DS, Malik ZM, Durieux M. Midazolam modulates effects of intravenous lidocaine on bispectral index (BIS) [Abstract]. Anesthesiology 2007;107:A803.
Demographic data were comparable between groups (Table 2). There was a significant difference in SBP and DBP, but not HR, between the placebo and midazolam groups before administration of lidocaine.
Systemic lidocaine and BIS Table 2
123
Demographic data and vital signs
Age (yrs) Gender (% male) Race (n in each group) African-American White Hispanic Weight (kg) BP before lidocaine injection (mmHg) HR before lidocaine injection (bpm)
Placebo (Groups 1-3) means ± SD
Midazolam (Groups 4-6) mean ± SD
P-value
59 ± 15 57
54 ± 13 50
P = 0.11 P = 0.86
6 40 1 77 ± 15 151 ± 27/82 ± 12
3 42 1 75 ± 15 127 ± 19/73 ± 10
P = 0.58
74 ± 14
74 ± 12
3.2. Effect of lidocaine on BIS In the placebo group, we observed no significant effect of any of the three lidocaine doses on BIS values during the three minutes after administration. When subjects were premedicated with midazolam, there was a significant decrease in their baseline BIS values as compared with the placebo group (90 ± 7 vs. 94 ± 4, P b 0.001). Furthermore, in the midazolam group, we noted a significant decrease in the average BIS value, as compared with baseline, after lidocaine administration. This finding was most evident in those patients receiving 1.0 mg/kg (decrease to 84 ± 7, P b 0.001). Effects were less but were also present in those receiving 0.5 mg/kg (decrease to 85 ± 9, P = 0.03), and in those receiving 1.5 mg/kg (decrease to 86 ± 8, P = 0.0031). Fig. 1 presents these data as percentage changes.
P = 0.38 P b 0.0001 for both SBP and DBP P = 0.96
midazolam. Eight subjects showed marked sedation, all of whom had received midazolam. Again, this finding was not dependent on the lidocaine dose.
3.4. Adverse events No serious cardiac dysrhythmia or seizure activity was noted during the study. One subject suffered an adverse event after receiving midazolam plus 1.5 mg/kg of lidocaine. She became markedly sedated and subsequently hypoxic 90 seconds after receiving lidocaine, necessitating bag-mask ventilation. No BIS values were recorded, and she was withdrawn from the study. It is noteworthy that this subject weighed 99 kg, which placed her at the upper limit of acceptance to the study. Her body mass index was 37 kg/m2; thus, a 0.03 mg/kg dose of midazolam may have been too high for her body composition.
3.3. Side effects The frequency of perioral numbness or tingling was greatest in patients receiving placebo and 1.5 mg/kg lidocaine (8/16 subjects). Eight of 97 subjects complained of discomfort on propofol injection, which was not influenced by either the lidocaine dose or the presence of
% change in BIS from baseline
% change in BIS after lidocaine, ± midazolam midazolam placebo 0 -1 -2 -3 -4 -5 -6 -7 * -8 * * -9 -10 means±SD; *P < 0.05 vs baseline.
placebo+0.5mg/kg lidocaine placebo+1.0mg/kg lidocaine placebo+1.5mg/kg lidocaine midazolam+0.5mg/kg lidocaine midazolam+1.0mg/kg lidocaine midazolam+1.5mg/kg lidocaine
Fig. 1 Changes (in %) in Bispectral Index (BIS) after lidocaine with and without midazolam, compared with baseline.
4. Discussion Our results suggest that systemic lidocaine affects BIS values in the presence – but not absence – of midazolam. As lidocaine affects BIS in the presence of propofol or sevoflurane anesthesia, it is suggested that the effect of lidocaine on BIS is indirect and occurs because the local anesthetic modulates GABAergic signaling.
4.1. Limitations Our study has some limitations. First, plasma concentrations of lidocaine were not measured. However, as we were studying the effect of lidocaine on BIS, ie, on cerebral function, it was really the concentration of lidocaine at the brain that was of interest. A study in 65 patients that sought to characterize the time course of lidocaine concentrations in the plasma and cerebrospinal fluid (CSF) after an IV bolus of 2.0 mg/kg lidocaine did not show any statistical correlation between plasma and CSF lidocaine levels [6]. Thus, to have measured
124 plasma lidocaine concentrations would not necessarily have reflected CSF concentration (let alone brain concentration) of lidocaine during the 3-minute recording time; we therefore decided not to subject the patients to an additional venous puncture to obtain these data. Another limitation is the short time period of BIS recording. It is possible that an effect of lidocaine on BIS might have been seen in the placebo group if recording had continued for longer than three minutes. A longer recording time also might have provided information on the duration of any effect of a bolus of lidocaine on BIS. However, a decrease in BIS was evident at 90 to 120 seconds, and BIS values began to level out after this time (data not shown), suggesting that longer measurement periods would have had limited value. In another clinical study [7], the peak effect of a bolus of 1.5 mg/kg of lidocaine (suppression of all airway reflexes except the apneic reflex) was noted at two minutes; the authors surmised that this finding also was due to a central effect of the lidocaine. Since the purpose of preinduction lidocaine administration is to diminish pain during propofol injection and limit hemodynamic responses to intubation, we did not want to extend the observation period any further. We used mean BIS values over the three minutes following administration of lidocaine rather than the lowest BIS value recorded during the 3-minute interval (which would perhaps have indicated the “peak effect” of lidocaine) or a measurement taken at a specific time point. On analysis of the data, it appeared that the maximum effect was generally seen at 90 seconds after lidocaine injection, but it varied between individuals. The time to peak effect is dependent, among other things, on cardiac output and cerebral blood flow (CBF). It was thought that to stipulate a specific recording time to measure peak effect for each subject might miss the peak effect in some subjects. The lowest BIS value for each individual was noted, whether it occurred at a 30-second interval or not. However, BIS recordings tended to fluctuate, resulting in several “peaks and valleys”. Peak values were very brief at times. A fluctuating record suggests that BIS does not simply reflect brain concentrations of lidocaine. We felt that by taking a mean value of recordings rather than the lowest recording for each subject, we were likely to have a better representation of the BIS values following lidocaine injection. Another confounding factor is BP fluctuation. Hypotension may decrease CBF, which results in somnolence. This situation could potentially limit the interpretation of the study results. However, the comparison of BP values within each group during the study period did not detect any significant differences.
4.2. Effect of lidocaine on anesthesia We hypothesized that IV lidocaine would not have an effect on BIS in the absence of midazolam, and indeed we did not find an effect on BIS by lidocaine alone in our study.
A. Gottschalk et al. When lidocaine was administered after premedication with midazolam, a GABAergic agonist, a significant decrease in BIS recordings was observed. These results are consistent with those of previous clinical studies that noted an effect of local anesthetic agents on BIS in the presence and absence of other GABAergic agents. Kaba et al studied the effects of IV lidocaine on acute rehabilitation after laparoscopic colectomy [1]. These investigators found that IV infusion of lidocaine reduced the amount of sevoflurane required to maintain BIS scores during surgery, compared with placebo. Patients who received lidocaine or bupivacaine intramuscularly (IM) during propofol anesthesia required less propofol to achieve equal BIS values [2]. In that study, IM bupivacaine, lidocaine, or saline was given 10 minutes before induction. There was no significant difference in BIS between the groups prior to local anesthetic administration, or preinduction, ie, no effect of the local anesthetic itself on BIS prior to the administration of propofol. In contrast, IM administration of the local anesthetics was associated with a decrease in both the induction and maintenance doses of propofol. However, such effects have not been found in all settings. A recent study comparing sevoflurane anesthesia alone, sevoflurane anesthesia plus an IV lidocaine bolus followed by infusion, and sevoflurane anesthesia plus epidural lidocaine, showed a reduction in sevoflurane requirement in the epidural group, as measured by BIS, but not in the IV lidocaine group, compared with sevoflurane alone, after 10 minutes [8]. This finding could be due to indirect effects of deafferentation as well as direct supraspinal effects.
4.3. Lidocaine and GABA signaling Given that IV lidocaine decreases BIS values in the presence of midazolam, sevoflurane, and propofol, but does not affect BIS recordings when given alone, we speculate that lidocaine exerts its effect on BIS by modulation of the effects of drugs acting at the gamma-aminobutyric acid-A (GABAA) receptor. Previous studies have looked at the effects of lidocaine on the GABAA receptor in vitro. In most of these studies, lidocaine inhibited the GABA-induced current [9-11]. However, Nordmark et al [12] studied the effect of local anesthetics on the GABA-gated chloride channel in crayfish stretch receptor neurons. They found that lidocaine increased GABA-induced chloride conductance when applied in concentrations of the same order as those required to block the nerve impulse, and showed evidence to support that this was due to a decrease in GABA uptake. The fact that some of the above-cited studies showed an inhibitory effect of lidocaine on GABAA receptor function, while other studies, in different species, showed a lidocaine-induced enhancement of GABAergic current, is not necessarily contradictory [12]. Clinically, local anesthetic agents are anticonvulsant at low doses and proconvulsant at high doses [13]. A biphasic response at the GABAA receptor has been observed with
Systemic lidocaine and BIS volatile anesthetics and barbiturates – potentiation occurring at lower concentrations and inhibition occurring at higher concentrations [14]. These studies showed that lidocaine was able to interact with the GABAA receptor. It is possible that lidocaine and midazolam together have a positive heterotropic interaction at the GABAA receptor, allosterically modifying the receptor to favor the open state of the ion channel. This interaction, however, has not been studied. Alternatively, given that in one study 10 μM concentrations of lidocaine (a concentration thought to be “clinically relevant”) enhanced glycine receptor function but did not affect GABAA receptor function [9], it is possible that the addition of midazolam, with its consequent agonistic effects at the GABAA receptor, has an additive effect with lidocaine on other receptors [ie, acid-sensing ion channel or hyperpolarization-activated cyclic nucleotide gated (HCN) channels] [15,16], which together result in decreased BIS values.
4.4. Summary Intravenous lidocaine reduces BIS readings in the presence of midazolam but not in its absence. We suggest that the local anesthetic modulates the effects of GABAergic drugs on BIS, rather than affecting BIS itself. Whether a decrease in BIS, as noted under these circumstances, indicates a decrease in level of consciousness – in other words, whether BIS is still a valid measure of depth of sedation in the presence of lidocaine – remains to be shown.
References [1] Kaba A, Laurent SR, Detroz BJ, et al. Intravenous lidocaine infusion facilitates acute rehabilitation after laparoscopic colectomy. Anesthesiology 2007;106:11-8.
125 [2] Senturk M, Pembeci K, Menda F, et al. Effects of intramuscular administration of lidocaine or bupivacaine on induction and maintenance doses of propofol evaluated by bispectral index. Br J Anaesth 2002;89:849-52. [3] Gaughen CM, Durieux M. The effect of too much intravenous lidocaine on bispectral index. Anesth Analg 2006;103:1464-5. [4] Pollock JE, Neal JM, Liu SS, Burkhead D, Polissar N. Sedation during spinal anesthesia. Anesthesiology 2000;93:728-34. [5] Ozkan-Seyhan T, Sungur MO, Senturk E, et al. BIS guided sedation with propofol during spinal anaesthesia: influence of anaesthetic level on sedation requirement. Br J Anaesth 2006;96:645-9. [6] Tsai PS, Buerkle H, Huang LT, Lee TC, Yang LC, Lee JH. Lidocaine concentrations in plasma and cerebrospinal fluid after systemic bolus administration in humans. Anesth Analg 1998;87:601-4. [7] Nishino T, Hiraga K, Sugimori K. Effects of i.v. lignocaine on airway reflexes elicited by irritation of the tracheal mucosa in humans anaesthetized with enflurane. Br J Anaesth 1990;64:682-7. [8] Hodgson PS, Liu SS. Epidural lidocaine decreases sevoflurane requirement for adequate depth of anesthesia as measured by the Bispectral Index monitor. Anesthesiology 2001;94:799-803. [9] Hara K, Sata T. The effects of the local anesthetics lidocaine and procaine on glycine and gamma-aminobutyric acid receptors expressed in Xenopus oocytes. Anesth Analg 2007;104:1434-9. [10] Sugimoto M, Uchida I, Fukami S, Takenoshita M, Mashimo T, Yoshiya I. The alpha and gamma subunit-dependent effects of local anesthetics on recombinant GABA(A) receptors. Eur J Pharmacol 2002;401:329-37. [11] Ye JH, Ren J, Krnjević K, Liu PL, McArdle JJ. Cocaine and lidocaine have additive inhibitory effects on the GABAA current of acutely dissociated hippocampal pyramidal neurons. Brain Res 1999;821: 26-32. [12] Nordmark J, Rydqvist B. Local anaesthetics potentiate GABAmediated Cl-currents by inhibiting GABA uptake. Neuroreport 1997;8:465-8. [13] Modica PA, Tempelhoff R, White PF. Pro- and anticonvulsant effects of anesthetics (Part II). Anesth Analg 1990;70:433-44. [14] Dilger JP. The effects of general anaesthetics on ligand-gated ion channels. Br J Anaesth 2002;89:41-51. [15] Lin J, Chu X, Maysami S, et al. Inhibition of acid sensing ion channel currents by lidocaine in cultured mouse cortical neurons. Anesth Analg 2011;112:977-81. [16] Meng QT, Xia ZY, Liu J, Bayliss DA, Chen X. Local anesthetic inhibits hyperpolarization-activated cationic currents. Mol Pharmacol 2011;79:866-73.