The Bispectral Index (BIS) has been reported to be a quantifiable measure of the depth of sedation during propofol anesthesia (1) and guide for anesthetic administration (2). The BIS is based on Fourier spectral analysis and integrates several disparate descriptors of the electroencephalogram (EEG) into a univariate variable (3). Development of BIS prompted the search for other markers of anesthesia depth using the EEG signal. Recently, the Datex-Ohmeda Entropy monitor (Datex-Ohmeda, Helsinki, Finland) has become commercially available. It assesses loss of consciousness (LOC) by quantifying the degree of spatial and temporal integration of cerebral neuronal activity. Two indices derived from this monitor, state entropy (SE) and response entropy (RE), have been demonstrated to correlate with the degree of propofol sedation (4–6).
Bruhn et al. (7) published the first two cases of falsely increased BIS by electromyogram activity (EMG) in anesthetized nonparalyzed patients. Their observations indicate that EMG, a well-known source of EEG contamination, can interfere with and increase the BIS value. EMG activity may be particularly significant during light anesthesia, i.e., during the initial phase of anesthetic induction that corresponds with a BIS value of 85–65 (8).
The aim of this study, therefore, was to investigate the effect of neuromuscular blockade on BIS, spectral edge frequency (SEF) values, and EMG provided by the BIS monitor on SE and RE values provided by the entropy monitor in patients lightly anesthetized with a fixed target concentration of remifentanil and a titrated target-controlled infusion (TCI) of propofol.
This prospective, randomized, double-blind study was approved by the Ethical Committee of our University. All participants were informed as to the nature of the study and gave written informed consent. Patients between the ages of 18 and 90 yr, presenting for various surgical procedures requiring general anesthesia with muscle relaxation, were enrolled for study. Only ASA physical status I–III patients were selected. Exclusion criteria were pregnancy, muscle relaxant allergy, neurological disorders, and predicted difficult ventilation or endotracheal intubation.
No sedative premedication was used. On arrival in the operating room (OR), a 20-gauge cannula was inserted into a large forearm vein and an infusion of lactated Ringer’s solution infused at 500 mL/h. Infusion pumps containing propofol and remifentanil were connected to the IV cannula via a three-way Smartsite® Needle-Free System (priming volume of 0.3 mL, ALARIS Medical Systems, San Diego, CA). Routine physiological monitoring was commenced (pulse oximetry, electrocardiography, and noninvasive arterial blood pressure recordings every 1 min). All OR staff were instructed to limit noise and any stimulation of patients during the study. BIS was measured with an Aspect A-2000 XP bispectral monitor (Aspect Medical Systems, Newton, MA; software version 3.11) while entropy indices were simultaneously recorded using an Entropy monitor. In accordance with the manufacturers’ instructions, the forehead skin was carefully cleaned before positioning the BIS sensor (ZipPrep® four-electrode sensor) and the entropy sensor (a special composite electrode with three elements). The side of the forehead for electrode placement of the two monitors was randomly assigned. Before starting BIS recording, we verified that electrode impedance was below 5 kΩ. BIS sampling rate was 256 Hz with a smoothing rate of 15 s. In the entropy module, a sampling frequency of 400 Hz was used with time windows ranging from 60 to 15 and 15.36 to 1.92 s for SE and RE respectively (9). On the entropy module, the impedance was checked automatically and was kept below 7.5 kΩ.
Propofol and remifentanil were administered using computer-controlled infusions (Infusion ToolBox 95, version 4.8) (10). This TCI system calculates and targets the effect compartment concentration (CeT) of IV anesthetic drugs using pharmacokinetic models. The population pharmacokinetic sets of Minto et al. (11) and Schnider et al. (12) were selected for remifentanil and propofol, respectively, because their pharmacokinetic parameters are adapted for age and body mass index. This software steered two Asena™ GH infusion pumps (Alaris Medical Systems, San Diego, CA). A 350 MHz Pentium II PC was used to provide a user interface and to control communication with the infusion pumps via RS232 serial ports.
During the study period, patients spontaneously breathed 100% oxygen via a face mask. When necessary, ventilation was gently assisted to obtain an arterial oxygen saturation more than 95%. Response to verbal commands and to gentle shaking was assessed every 30 s; LOC was defined as a lack of response to these stimuli. In both groups, the same protocol was used to obtain LOC. Remifentanil TCI was started at 2 ng/mL CeT and maintained at this level throughout the study. One minute after a stable remifentanil CeT had been achieved, propofol TCI was started at a target concentration of 1 μg/mL CeT. The target concentration was progressively increased in increments of 0.5 μg/mL until LOC was obtained. Changes in the propofol infusion were commanded if the patient remained responsive to stimuli when 95% CeT had been reached. The time delay necessary to reach this concentration was calculated from the pharmacokinetic model used by the computer-controlled infusion. When LOC had been attained, the propofol CeT was held constant throughout the whole study period. Two minutes after LOC, and after verification that the patient could be gently ventilated via the face mask, patients were allocated to receive a bolus injection of either atracurium 0.5 mg/kg (atracurium group) or normal saline (placebo group). In the placebo group, the volume of saline injected corresponded to the equivalent calculated volume of atracurium for that patient. Syringes of atracurium and normal saline were prepared by another physician outside the OR.
Two anesthesiologists were assigned to each patient. One assumed responsibility for the clinical management of the patient while the second manually recorded the following data: BIS, SEF, EMG, RE, SE values, and the propofol CeT calculated by the Infusion ToolBox software. Data were collected 2 min after LOC (before injection of atracurium or normal saline, “Before bolus data”) and again 3 min after the bolus injection of atracurium or normal saline (“After bolus data”).
Results were expressed as mean ± sd. Intragroup continuous data were analyzed using a one-way repeated-measures analysis of variance on time, with a post hoc Bonferroni correction for multiple comparisons. Intergroup differences were compared using an unpaired Student’s t-test for continuous data and Fisher’s exact test for categorical data. Probability values less than 0.05 were considered significant. Data analysis was performed using SPSS™ version 11.0 (SPSS Science Inc., Chicago, IL).
Forty patients were included in the study. One patient in the atracurium group had an anaphylactoid reaction after injection of the muscle relaxant and was consequently excluded.
Demographic data and ASA physical status were similar in both groups (Table 1). LOC was obtained at 285 ± 53 s (range, 210–410 s) in the placebo group and 299 ± 49 s (range, 270–400 s) in the atracurium group (not significant). Propofol CeT were similar between the groups 2 min after LOC: 2.7 ± 0.5 μg/mL (range, 1.5–4.5 μg/mL) in the placebo group and 2.9 ± 0.7 μg/mL (range, 2–4.5 μg/mL) in the atracurium group. Average normalized propofol consumption was also similar 2 min after LOC: 1.8 ± 0.5 mg/kg (range, 0.9–2.8 mg/kg) in the placebo group, 2.0 ± 0.8 mg/kg (range, 1.0–4.2 mg/kg) in the atracurium group. BIS, SEF, EMG, SE, and RE data were similar in both groups before administration of the bolus injections (Figs. 1 and 2, Tables 2 and 3).
Administration of the placebo bolus induced significant decreases in BIS (P < 0.002), SEF (P < 0.05), EMG (P < 0.02), SE (P < 0.05), and RE (P < 0.01) compared with the values measured at LOC before bolus injection (Figs. 1 and 2, Tables 2 and 3). Atracurium administration induced significant decreases in BIS (P < 0.0001), SEF (P < 0.01), EMG (P < 0.0001), SE (P < 0.0001), and RE values (P < 0.0001) compared with the values measured at LOC (Figs. 1 and 2, Tables 2 and 3).
Decreases in BIS (P < 0.05), EMG (P < 0.0001), and RE (P < 0.01) values were larger after atracurium than after placebo. After atracurium and placebo injections SEF and SE were not significantly different between the two groups (Figs. 1 and 2, Tables 2 and 3).
In this prospective, randomized, double-blind clinical study, BIS and entropy EEG monitors were used to quantify depth of hypnosis during anesthesia with a fixed target concentration of remifentanil and titrated TCI of propofol. Neuromuscular blockade by atracurium bolus injection, after LOC, induced decreases in BIS, EMG, and RE but not SEF or SE.
Contradictory reports or studies concerning the influence of muscle relaxants on BIS have been published. Bruhn et al. (7) reported the first two cases in which BIS reflected EMG activity and not depth of anesthesia. However, two further studies excluded such an effect during deep anesthesia (BIS <50) in both volunteers (13) and patients (14) receiving a propofol computer-controlled infusion. Messner et al. (15) demonstrated in paralyzed nonsedated volunteers that EMG activity influences surface EEG and calculation of BIS. In addition, muscle activity influenced BIS values in sedated intensive care unit patients (16,17). Until now, no study has been conducted during the initial period of anesthetic induction. Our study design involved induction of anesthesia using a stable TCI of remifentanil with an incremental TCI of propofol, titrated to LOC.
Contamination of BIS by EMG activity is probably related to the spectrum of EEG frequencies used in its determination. BIS software uses signals up to 47 Hz, whereas EEG and EMG signals are conventionally considered to exist in the 0.5–30 Hz and 30–300 Hz bands, respectively (8). This overlap in selection of analyzed frequencies may explain the influence of EMG activity on BIS. On the other hand, SEF is not influenced by muscle activity. This may be explained not only by the higher EEG frequencies analyzed but also by the use of differential filtering in the 0–30 Hz band. Furthermore, the EMG-related power in the remaining higher frequencies is quite small in relation to the total power (18). Therefore EMG-related changes are not detected by the SEF (7).
Our results indicate that recent improvements in BIS technology (a new four-electrode sensor and new software to better detect EMG activity) (19) have not made it possible to solve the problem of muscle interference. The BIS is the result of a proprietary combination of three EEG subparameters that are dependent on the depth of anesthesia: “burst suppression” (a combination of the “burst suppression ratio” and QUAZZI), the “SynchFastSlow” (the contribution of bispectral analysis), and the “BetaRatio” (20). The weight of burst suppression in the calculation of BIS is more during general anesthesia as opposed to sedation. The SynchFastSlow is defined as the logarithm of the following ratio: the sum of bispectrum peaks in the 0.5–47 Hz range divided by the sum of the bispectrum peaks in the 40–47 Hz range. The weight of SynchFastSlow in the BIS index calculation relates to the degree of EEG activity, during general anesthesia. The weight of the BetaRatio in the calculation of BIS is greatest at light sedation, during which frequencies in the β band predominate. The BetaRatio is calculated as the log of the ratio of power in two empirically derived frequency bands: log (P30–40 Hz/P11–20 Hz). The calculation thus requires inclusion of frequencies below 40 Hz, approaching frequencies generated by muscle activity, and explains the influence of muscle relaxant on BIS. Recently, Vakkuri et al. (6) observed a plateau in BIS values more than 60, possibly resulting from a transition from the BetaRatio to the SyncFastSlow algorithm. Furthermore, they demonstrated that the BetaRatio dominates BIS calculation during light general anesthesia or deep sedation. The differences in results observed in our study compared with that of Dahaba et al. (14) were most likely attributable to differences in the baseline state of the study subjects. Our study was performed in patients with BIS values more than 72 and high EMG activity (mean value approximately 38 dB, Fig. 1). Dahaba et al. (14) studied the effect of neuromuscular blockade on BIS in deeply anesthetized patients with low EMG activity (30 dB). The higher EMG activity observed in our patients (Fig. 1, Table 2) could have increased the “numerator” of the BetaRatio algorithm and consequently resulted in higher BIS values. Finally, in nonparalyzed patients, the effect of neuromuscular blockade on the BIS value depends on the depth of anesthesia. During deep anesthesia (BIS values of approximately 40), a bolus of mivacurium evoked a transient and minor effect on the BIS value (14). At a light level of anesthesia (BIS values up to 65), a bolus of atracurium decreased the BIS value.
A further principal finding of our study concerns the Entropy™ indices; RE, but not SE, was modified by myorelaxant injection (Fig. 2, Table 3). As for BIS, this difference can most likely be explained by the different EEG frequency selected for analysis. RE is computed over a frequency range from 0.8 Hz to 47 Hz and contains even the higher EMG dominated frequencies (9). SE is computed over the frequency range 0.8 Hz to 32 Hz, which corresponds to the EEG-dominated part of the spectrum, thus primarily reflecting cortical activity. The main difference between RE and SE indices corresponds to the contribution of frequencies between 32 and 47 Hz. These higher frequency components are evaluated with a time window of <2 seconds, thus providing an almost immediate indication of the frontal EMG activity. In the entropy monitor, EMG activity is treated as a signal component rather than an artifact. RE can also increase as a result of EMG activation. Finally, the distinction between SE and RE allows the clinician to determine whether the activation originates from the EEG or from the EMG.
An alternative explanation of the interaction between myorelaxation and EEG is the so-called “afferent muscle spindle” theory. This theory states that stretching or contracting muscle fibers results in afferent input to arousal centers in the brain and that muscle relaxation potentially provides a sedative effect by decreasing such afferent input. This theory is supported by animal (21,22) and human (23,24) studies. Speculation remains, however, notably after the work of Fahey et al. (25) and more recently of Messner et al. (15), who demonstrated that the administration of a muscle relaxant in fully conscious volunteers had no influence on sedation.
We observed that placebo injection produced significant changes in BIS, SEF, EMG, RE, and SE. This was most likely a result of unstable depth of anesthesia during the recording period. Instability of the targeted remifentanil concentration (held constant at 2 ng/mL throughout the study) is an unlikely explanation for the decrease in BIS values (26,27). However, the TCI of propofol was incrementally increased until LOC. The half-life of the plasma effect-site equilibration value (t1/2 ke0) of the selected pharmacokinetic set of Schnider et al. (12) was rapid (1.8 minutes) with a time to peak effect of 1.5 minutes. This ke0 value is faster than that included in the Diprifusor™ (half time ke0 of 2.6 minutes; time to peak effect of 4 minutes) and much faster than the value proposed by Ludbrook et al. (28). We hypothesize, therefore, that the fast ke0 value used in the set of Schnider et al. was too rapid in our group of patients anesthetized by progressive TCI titration. The peak effect-site concentration of propofol is delayed when a slower ke0 is selected, which could explain the decrease in the BIS values in our group of patients during the study period. This explanation has to be applied in both groups of patients and does not contradict the differences between the groups as far as the influence of muscle relaxation on BIS, EMG, and RE entropy values is concerned.
In conclusion, clinicians must be aware that neuromuscular blockade decreases BIS and RE, but not SE, values during light general anesthesia. Using BIS or RE to titrate TCI propofol could expose patients to an overdose.
The authors would like to thank Dr. S. Morrison for reviewing the manuscript.
1. Glass PS, Bloom M, Kearse L, et al. Bispectral analysis measures sedation and memory effects of propofol, midazolam, isoflurane, and alfentanil in healthy volunteers. Anesthesiology 1997;86:836–47.
2. Song D, Joshi GP, White PF. Titration of volatile anesthetics using bispectral index facilitates recovery after ambulatory anesthesia. Anesthesiology 1997;87:842–8.
3. Sigl JC, Chamoun NG. An introduction to bispectral analysis for the electroencephalogram. J Clin Monit 1994;10:392–404.
4. Anderson RE, Jakobsson JG. Entropy of EEG during anaesthetic induction: a comparative study with propofol or nitrous oxide as sole agent. Br J Anaesth 2004;92:167–70.
5. Anderson RE, Barr G, Owall A, Jakobsson J. Entropy during propofol hypnosis, including an episode of wakefulness. Anaesthesia 2004;59:52–6.
6. Vakkuri A, Yli-Hankala A, Talja P, et al. Time-frequency balanced spectral entropy as a measure of anesthetic drug effect in central nervous system during sevoflurane, propofol, and thiopental anesthesia. Acta Anaesthesiol Scand 2004;48:145–53.
7. Bruhn J, Bouillon TW, Shafer SL. Electromyographic activity falsely elevates the bispectral index. Anesthesiology 2000;92:1485–7.
8. Johansen JW, Sebel PS. Development and clinical application of electroencephalographic bispectrum monitoring. Anesthesiology 2000;93:1336–44.
9. Viertio-Oja H, Maja V, Sarkela M, et al. Description of the Entropy algorithm as applied in the Datex-Ohmeda S/5 Entropy Module. Acta Anaesthesiol Scand 2004;48:154–61.
10. Cantraine FR, Coussaert EJ. The first object oriented monitor for intravenous anesthesia. J Clin Monit Comput 2000;16:3–10.
11. Minto CF, Schnider TW, Egan TD, et al. Influence of age and gender on the pharmacokinetics and pharmacodynamics of remifentanil. I. Model development Anesthesiology 1997;86:10–23.
12. Schnider TW, Minto CF, Shafer SL, et al. The influence of age on propofol pharmacodynamics. Anesthesiology 1999;90:1502–16.
13. Greif R, Greenwald S, Schweitzer E, et al. Muscle relaxation does not alter hypnotic level during propofol anesthesia. Anesth Analg 2002;94:604–8.
14. Dahaba AA, Mattweber M, Fuchs A, et al. The effect of different stages of neuromuscular block on the bispectral index and the bispectral index-XP under remifentanil/propofol anesthesia. Anesth Analg 2004;99:781–7.
15. Messner M, Beese U, Romstock J, et al. The bispectral index declines during neuromuscular block in fully awake persons. Anesth Analg 2003;97:488–91.
16. Nasraway SS Jr, Wu EC, Kelleher RM, et al. How reliable is the Bispectral Index in critically ill patients? A prospective, comparative, single-blinded observer study. Crit Care Med 2002;30:1483–7.
17. Vivien B, Di Maria S, Ouattara A, et al. Overestimation of Bispectral Index in sedated intensive care unit patients revealed by administration of muscle relaxant. Anesthesiology 2003;99:9–17.
18. Sleigh JW, Steyn-Ross DA, Steyn-Ross ML, et al. Comparison of changes in electroencephalographic measures during induction of general anaesthesia: influence of the gamma frequency band and electromyogram signal. Br J Anaesth 2001;86:50–8.
19. Coluzzi F, Di Filippo C, Rossetti E, et al. BIS monitoring in ICU: advantages of the new XP generation [abstract]. Crit Care 2002;6(suppl 1):P68.
20. Rampil IJ. A primer for EEG signal processing in anesthesia. Anesthesiology 1998;89:980–1002.
21. Schwartz AE, Navedo AT, Berman MF. Pancuronium increases the duration of electroencephalogram burst suppression in dogs anesthetized with isoflurane. Anesthesiology 1992;77:686–90.
22. Lanier WL, Iaizzo PA, Milde JH, Sharbrough FW. The cerebral and systemic effects of movement in response to a noxious stimulus in lightly anesthetized dogs: possible modulation of cerebral function by muscle afferents. Anesthesiology 1994;80:392–401.
23. Barth L. Paradoxical interaction between halothane and pancuronium. Anaesthesia 1973;28:514–20.
24. Forbes AR, Cohen NH, Eger EI II. Pancuronium reduces halothane requirement in man. Anesth Analg 1979;58:497–9.
25. Fahey MR, Sessler DI, Cannon JE, et al. Atracurium, vecuronium, and pancuronium do not alter the minimum alveolar concentration of halothane in humans. Anesthesiology 1989;71:53–6.
26. Guignard B, Menigaux C, Dupont X, et al. The effect of remifentanil on the bispectral index change and hemodynamic responses after orotracheal intubation. Anesth Analg 2000;90:161–7.
27. Bouillon TW, Bruhn J, Radulescu L, et al. Pharmacodynamic interaction between propofol and remifentanil regarding hypnosis, tolerance of laryngoscopy, bispectral index, and electroencephalographic approximate entropy. Anesthesiology 2004;100:1353–72.
28. Ludbrook GL, Visco E, Lam AM. Propofol: relation between brain concentrations, electroencephalogram, middle cerebral artery blood flow velocity, and cerebral oxygen extraction during induction of anesthesia. Anesthesiology 2002;97:1363–70.