The classical concept of minimum alveolar anesthetic concentration (MAC) compares relative potencies of volatile anesthetics based on motor responses to noxious stimulation. There is no monitor that can reliably predict whether the level of anesthesia is sufficient to suppress voluntary or reflex movement to skin incision.
In addition to detecting intraoperative awareness, it would be useful if a monitor were able to predict the need for additional anesthetics or analgesics to prevent movement or laryngospasm (in patients with a laryngeal mask airway in place) in response to noxious stimuli. The electroencephalogram (EEG) represents spontaneous electrical activity in the superficial cortical layer of pyramidal cells. Although marked EEG changes occur during the transition from the awake to the anesthetized state, these changes are biphasic; that is, two anesthetic levels may have similar EEG profiles . Accordingly, attempts to use EEG spectral variables to predict inadequate levels of anesthesia in individuals have been remarkably unsuccessful . The early cortical (mid-latency) auditory evoked response (MLAER) represents signal transmission and processing in the acoustic pathway from the cochlea to the primary auditory cortex. This evoked signal changes predictably in response to both increases in the anesthetic concentration [3-8] and to surgical stimulation [9,10]. It may be unrealistic to expect either signal (EEG or MLAER) in an unstimulated patient to predict movement when surgery starts. However, the response of these signals to a lesser noxious stimulus might be predictive of what would happen with surgical incision.
A multicenter clinical study allowed us to test whether changes in the EEG or MLAER variables after tetanic stimulation at the wrist could be used to predict subsequent movement after skin incision in patients anesthetized with 1 MAC isoflurane in N2 O, in a large group of patients. In addition, we investigated whether the EEG or MLAER variables before noxious stimulation (tetanus and skin incision) could differentiate between patients who subsequently moved (movers) and those who did not (nonmovers).
The study protocol was approved by the local ethics committees of the five participating European clinical centers in Amsterdam, Hamburg, London, Luebeck, and Munich. Ninety-three ASA physical status I and II patients, aged 18-60 yr, undergoing elective general or orthopedic surgery involving the upper leg, groin, or abdomen gave oral consent to participate in the study.
Midazolam 7.5 mg was given orally 1 h before anesthesia. Anesthesia was induced with propofol 2.0-2.5 mg/kg IV. Tracheal intubation was facilitated with succinylcholine 1 mg/kg. Using overpressure ventilation, patients were rapidly equilibrated to 0.6% end-expiratory isoflurane in 50% N2 O (1 MAC). Heart rate and noninvasive blood pressure were recorded every 2 min during the study. After the end-tidal isoflurane concentration had been maintained for at least 15 min at 0.6%, a 5-s tetanic stimulus (40 mA, 50 Hz) was applied to the ulnar nerve at the wrist. If the patient moved during the period between tetanic stimulus and skin incision, rescue medication was given (propofol and fentanyl), and all subsequent data (post-tetanus and pre- and postskin incision) from this particular patient were excluded from later analysis. Skin incision was performed at least 5 min after application of the tetanic stimulus, and the presence or absence of movement was recorded. Movement in response to a stimulus was recorded if the patient made purposeful withdrawal responses or when the head or legs moved within 2 min after the tetanic stimulus or skin incision.
The EEG/auditory evoked response (AER) data acquisition system was custom-designed for this project. It consisted of a small (148 x 78 x 29 mm) battery-powered amplifier and stimulator unit placed in close proximity to the patient's head. The main feature that distinguishes this system from commercially available equipment is that interference by diathermy noise and "common mode" line interference (line frequency in Europe 50 Hz) is efficiently eliminated, thus allowing continuous acquisition of EEG/AER data in the proximity of poorly shielded devices, such as heating apparatus and infusion pumps, as well as during frequent use of electrocautery . A single-channel EEG was recorded from Ag-AgCl electrodes placed on the forehead and left mastoid, with the right mastoid as common. Electrode impedance was checked automatically and maintained <5 k Omega. Binaural auditory stimulation was performed at 6.1 Hz with rarefaction clicks (70 dB above hearing level) using insert earphones. The raw EEG data were sampled at a rate of 1 kHz and stored on a computer for off-line analysis. The exact position of each click was marked in the raw datafile, allowing off-line AER averaging using various filter settings and averaging paradigms. To allow estimation of signal quality during the study period, the raw EEG data and a moving AER average were continuously displayed on the monitor screen.
For the present study, we used normal ensemble averaging, in which every datapoint in the AER waveform represents the arithmetic mean of all sweeps over the respective recording periods. The data were filtered using an analog 400-Hz low-pass filter, digitally filtered with a 25-Hz high-pass filter, and three-point smoothing was applied. Waveforms consisting of 1024 averaged sweeps of 100 ms duration, equivalent to 2.6 min of recording, were used in the analysis. These were obtained immediately before and immediately after tetanic stimulation and incision (four periods). One experienced observer (CT) blinded to clinical center and study period identified peaks Pa and Nb in the MLAER waveform. Latencies and absolute and interpeak MLAER amplitudes were determined.
The raw signal for the four study periods was subjected to power spectral analysis using 2-s EEG epochs (EEG bandwidth 0.5-30 Hz, electromyogram (EMG) bandwidth 30.5-400 Hz). The following variables were calculated: total power; absolute and relative power in the delta (0.5-3.0 Hz), theta (3.5-8.0 Hz), alpha (8.5-12.0 Hz), and beta (12.5-30.0 Hz) bands; median frequency and 95% edge frequency; and EMG power in two bands: EMG 1 (30-70 Hz) and EMG 2 (70-250 Hz).
The EEG and MLAER variables were log-transformed, as untransformed these variables are not normally distributed. Data are presented as geometric means with their respective confidence intervals. Two sample t-tests were performed to test whether: 1) the post- versus pretetanus differences were different in movers at incision compared with nonmovers, after excluding patients who moved at tetanus; 2) the pretetanus values were different in movers and nonmovers to tetanic stimulation; 3) the preincision values were different in movers and nonmovers to incision after excluding patients who moved at tetany.
Complete datasets containing both EEG and MLAER data were available from 82 patients. After application of the tetanic stimulus at the wrist, 20 of the 82 patients (24%) moved. These 20 patients were not included in any statistical analysis that used posttetanic, preincision, or postincision data. This is to avoid bias due to the rescue medication, which some of these patients received, and to the possible stimulatory effect of moving. After skin incision, 39 of the 82 patients moved (48%). Of the 62 patients who did not move to tetanus, 26 (42%) moved to incision. Thus, of the 20 patients who moved to tetanus, 13 (65%) moved to incision.
Reproducible auditory evoked brainstem potentials, as evidenced by the presence of a characteristic peak V of the brainstem auditory response (latency 6-7 ms) could be recorded in every patient during anesthesia with 1 MAC isoflurane in N2 O. At 1 MAC isoflurane in N2 O, the mid-latency segment of the AER waveform, representing the early cortical response, was severely depressed (Figure 1). Peak-to-peak amplitude PA/Nb was <0.5 [micro sign]V in 62% of patients.
None of the AER or EEG variables was able to predict whether patients would move to surgical incision. The post- versus pretetanus differences were not significantly different in movers compared with the nonmovers. Pretetanus values were not different in movers and nonmovers to tetanic stimulation, with one exception-beta EEG power, which was less in movers to tetanus compared with nonmovers (P = 0.01). The preincision values were not different in movers and nonmovers to incision for any of the variables. Table 1 shows the geometric means and confidence intervals for the AER and EEG variables tested before and after the application of a 5-s tetanus and before and after skin incision.
The post- versus preincision difference was significantly different between nonmovers and movers in PA (P = 0.04) and Nb (P = 0.03) amplitudes, EMG 1 (P < 0.001) and EMG 2 (P = 0.02), for which it was greater in movers compared with nonmovers; and in theta (P = 0.02) and alpha EEG power (P = 0.01), for which the reverse was true. For the patients who moved to tetanus, the data collected after incision were not excluded from these analyses; therefore, these comparisons could have been influenced by rescue medication or movement.
The hemodynamic variables were similarly unable to predict whether patients would move to surgical incision. The post- versus pretetanus differences were not significantly different in movers at incision compared with nonmovers. Pretetanus values were not different in movers and nonmovers to tetanic stimulation. The preincision values were not different in movers and nonmovers to incision. Table 2 shows the geometric means and confidence intervals for systolic blood pressure and heart rate before and after the application of a 5-s tetanus and before and after skin incision. The post- versus preincision differences were also not significantly different between movers and nonmovers, although it approached significance (P = 0.07) for systolic blood pressure. There was a greater increase in systolic blood pressure increase in movers compared with nonmovers.
We were unable to predict movement to noxious stimulation by MLAER or EEG recordings. Although this is a negative finding, it is conclusive because of the large number of patients studied. Taking Pa amplitude as an example, a >35% increase or 40% decrease in the post- versus pretetanus difference between movers and nonmovers, with a probability of 0.05 and 80% power, would have been detected with a sample of this size. Significant changes in Pa and Nb amplitude from pre- to postincision (0.14 and 0.15 [micro sign]V, respectively) did occur in the patients who moved after skin incision, and this change was significantly different from the 0.01-[micro sign]V change in the nonmover group. As this comparison was based on data collected after skin incision, when the patient was actually moving, the information is obtained too late to be of predictive value to the anesthesiologist. Although it was not one of our primary hypotheses, it is nevertheless reassuring to confirm earlier work, in which the MLAER amplitudes Nb and the subsequent positive peak Pc amplitude significantly increased after skin incision in patients maintained on constant end-tidal halothane and nitrous oxide . In this previous study, the patients were paralyzed, and their data were uncontaminated by EMG, as opposed to the present study, in which the patients were unparalyzed. If such patients move, perhaps muscle potentials can contribute to the increased AER amplitudes.
One explanation for the fact that EEG and MLAER data were unable to distinguish movers from nonmovers is that anesthetic depression of the auditory pathway and cortical signal processing areas may not parallel anesthetic depression of the afferent-efferent nociceptive reflex pathway and the autonomic nervous system. If, at a given anesthetic concentration, cortical auditory areas are more depressed than the spinal nociceptive reflex-mediating areas, then the present data support the concept proposed by Rampil et al.  and Rampil  that MAC represents primarily the level of depression of a spinal antinociceptive reflex. To establish whether the level of anesthesia is sufficient to prevent movement of unparalyzed patients in response to a surgical incision, one would have to assess either the level of analgesia and/or the level of depression of the motor neuron pool, for example, by recording retrograde activation of motor neurons after peripheral motor nerve stimulation (F-waves) .
Another explanation is that anesthesia was too deep or the tetanic stimulus not intense or prolonged enough. Although we assessed the MLAER only at 1 MAC, it is likely that the dose-response MAC curve for MLAER is shifted to the left compared with the curve for the classical MAC end point of movement on incision. As a result, MLAER may be more susceptible to inhaled anesthetics. Our data suggest that MLAER are nearly completely suppressed at isoflurane concentrations that are compatible with a 50% proportion of movers. Although the purpose of this study was not to compare depression of MLAER at 1.0 MAC isoflurane with awake baseline responses, our data are in agreement with previous reports showing that 1.0 MAC isoflurane in N2 O produces profound depression of the MLAER amplitude. Schwender et al.  observed nearly complete suppression of MLAER in patients undergoing heart surgery with 1.0% isoflurane in O2. The awake amplitudes were 1.0-2.0 [micro sign]V. Newton et al.  found that 0.4% end-tidal isoflurane in oxygen administered to anesthetist volunteers decreased Pa amplitude from 0.70 to 0.29 [micro sign]V; Nb latency increased from 44.9 to 53.9 ms. Tetanic stimulation failed to produce a significant change in either hemodynamic or MLAER or EEG variables. However, ethical and technical (if the patient moved grossly, then rescue medication had to be administered, and the study was ended) considerations prohibited us from applying a more powerful tetanic stimulus or using a lighter level of anesthesia.
Several studies from participants of our group and others have suggested that MLAER may indicate the potential of intraoperative wakefulness and thus might be used clinically to prevent awareness with recall [16-18]. Even in the absence of N2 O, MLAER are severely suppressed at 0.4%-0.8% end-tidal isoflurane or sevoflurane concentrations [7,19]. At these anesthetic concentrations, there was a low incidence of motor signs of wakefulness (coughing, limb movement, or facial movement). In contrast, during N2 O/opioid/benzodiazepine anesthesia, MLAER amplitudes were 3 times higher, and there was fivefold higher incidence of movement. The authors  concluded that primary processing of sensory stimuli in the primary sensory cortex was blocked during isoflurane/N2 O anesthesia and that this was reflected in the changes in the MLAER.
In conclusion, although the MLAER and EEG variables may find clinical applications as alarms for possible intraoperative wakefulness with the attendant possibility of recall, it is unlikely that this evoked signal or the EEG (as it was processed in the present study) can be of practical use in titrating anesthesia to prevent movement to noxious stimulation.
We thank Mrs. M. Porsius (Amsterdam), Mrs. D. Droese (Munich), Mr. A. Meyer (Hamburg), and Dr. U. Richter (Luebeck) for their valuable help in collecting or analyzing the EEG and evoked potentials data.
1. Kuizenga K, Kalkman CJ, Hennis PJ. Quantitative electroencephalographic analysis of the biphasic concentration-effect relationship of propofol in surgical patients during extradural analgesia. Br J Anaesth 1998;80:725-32.
2. Dwyer RC, Rampil IJ, Eger EI II, Bennett HL. The electroencephalogram does not predict depth of isoflurane anesthesia. Anesthesiology 1994;81:403-9.
3. Thornton C, Catley DM, Jordan C, et al. Enflurane anaesthesia causes graded changes in the brainstem and early cortical auditory evoked response in man. Br J Anaesth 1983;55:479-86.
4. Thornton C, Creagh-Barry P, Jordan C, et al. Somatosensory and auditory evoked responses recorded simultaneously: differential effects of nitrous oxide and isoflurane [see comments]. Br J Anaesth 1992;68:508-14.
5. Schwender D, Klasing S, Madler C, et al. Effects of benzodiazepines on mid-latency auditory evoked potentials. Can J Anaesth 1993;40:1148-54.
6. Schwender D, Rimkus T, Haessler R, et al. Effects of increasing doses of alfentanil, fentanyl and morphine on mid-latency auditory evoked potentials. Br J Anaesth 1993;71:622-8.
7. Schwender D, Conzen P, Klasing S, et al. The effects of anesthesia with increasing end-expiratory concentrations of sevoflurane on midlatency auditory evoked potentials. Anesth Analg 1995;81:817-22.
8. Madler C, Keller I, Schwender D, Poppel E. Sensory information processing during general anaesthesia: effect of isoflurane on auditory evoked neuronal oscillations. Br J Anaesth 1991;66:81-7.
9. Thornton C, Konieczko K, Jones J, et al. Effect of surgical stimulation on the auditory evoked response. Br J Anaesth 1988;60:372-8.
10. Schwender D, Golling W, Klasing S, et al. Effects of surgical stimulation on midlatency auditory evoked potentials during general anaesthesia with propofol/fentanyl, isoflurane/fentanyl and flunitrazepam/fentanyl. Anaesthesia 1994;49:572-8.
11. Jordan C, Weller C, Thornton C, Newton DE. Monitoring evoked potentials during surgery to assess the level of anaesthesia. J Med Eng Technol 1995;19:77-9.
12. Rampil IJ, Mason P, Singh H. Anesthetic potency (MAC) is independent of forebrain structures in the rat. Anesthesiology 1993;78:707-12.
13. Rampil IJ. Anesthetic potency is not altered after hypothermic spinal cord transection in rats. Anesthesiology 1994;80:606-10.
14. King BS, Rampil IJ. Anesthetic depression of spinal motor neurons may contribute to lack of movement in response to noxious stimuli. Anesthesiology 1994;81:1484-92.
15. Schwender D, Kaiser A, Klasing S, et al. Midlatency auditory evoked potentials and explicit and implicit memory in patients undergoing cardiac surgery. Anesthesiology 1994;80:493-501.
16. Newton DE, Thornton C, Konieczko K, et al. Levels of consciousness in volunteers breathing sub-MAC concentrations of isoflurane. Br J Anaesth 1990;65:609-15.
17. Thornton C, Barrowcliffe MP, Konieczko KM, et al. The auditory evoked response as an indicator of awareness. Br J Anaesth 1989;63:113-5.
18. Newton DEF, Thornton C, Konieczko KM, et al. Auditory evoked response and awareness: a study in volunteers at sub-MAC concentrations of isoflurane. Br J Anaesth 1992;69:122-9.
19. Schwender D, Faber-Zullig E, Klasing S, et al. Motor signs of wakefulness during general anaesthesia with propofol, isoflurane and flunitrazepam/fentanyl and midlatency auditory evoked potentials. Anaesthesia 1994;49:476-84.