The minimum alveolar concentration (MAC) is the concentration of anaesthetic agent at which a movement-response may be observed in 50% of anaesthetized patients when a surgical incision is performed. Even though based solely on the movement, it has served as a standard of anaesthetic potency. The site of anaesthetic inhibition of the movement response is in the spinal cord . Cerebral cortex, rather than the spinal cord, is relevant to the state of consciousness. The cerebral cortex seems to be more sensitive than spinal cord to anaesthetic agents and it has been suggested, that only 0.4 MAC isoflurane is needed to guarantee unconsciousness in volunteers .
The electroencephalogram bispectral index (BIS) has been shown to indicate the hypnotic component of the anaesthetic state . We tested the hypothesis that less than 1 MAC sevoflurane is needed to ensure unconsciousness, as indicated by use of BIS recording during laparoscopic tubal ligation (LTL).
Patients and Methods
The study was approved by the institutional ethics committee. All patients gave their written informed consent. Thirty-two ASA I-II women, scheduled for elective laparoscopic tubal ligation (LTL), were studied (Table 1). Exclusion criteria were a history of cardiac, pulmonary or neurological disease, medication affecting the central nervous system (CNS), body mass index > 28 kg m−2, a history of oesophageal reflux and alcohol or drug abuse. Base-line arterial pressure and heart rate were recorded at the pre-operative visit.
Diazepam 5 mg was given orally as premedication ≈45 min before the anaesthetic induction. On arrival in the operating room, the patients were monitored using an electrocardiogram (ECG), an automated sphygmomanometer and a pulse oximeter (SpO2). A peripheral vein was cannulated.
The EEG signal was collected using four electrodes (Zipprep™, Aspect Medical Systems, Natick, MA, USA) applied to the forehead. The electrodes were positioned on both temporal bones laterally to the eyes and on the central forehead between the eyebrows, with a ground electrode on the forehead. The signal was amplified and the BIS value (software version 3.22) displayed, using an Aspect EEG monitor (Model A-1000, Aspect Medical Systems). The BIS values were collected on a computer at 5-s intervals. Haemodynamic variables were recorded at 3-min intervals throughout anaesthesia.
All the patients breathed oxygen via a clear facemask for 2 min before anaesthetic induction. After obtaining the base-line values for the BIS index, anaesthesia was induced with sevoflurane inhalation using a semiopen anaesthesia system primed with a fresh gas flow of 9 L min−1 (O2 3 L min−1 and N2O 6 L min−1) and sevoflurane vaporizer (Sevorane® Abbott Vapor 19.3) turned to a maximum of 8%. The patient was asked to exhale forcefully to residual volume, followed by a vital capacity breath with a facemask placed over the nose and the mouth, and then to hold her breath as long as possible. Thereafter the patient continued to breathe normally. When the patient became unresponsive to verbal commands, her respiration was assisted manually.
Tracheal intubation was facilitated with rocuronium 0.6 mg kg−1 and the patient's lungs were ventilated with a Sulla 808V™ Dräger (Drägerwerk AG, Lübeck, Germany) ventilator. During anaesthesia, SpO2, ECG, inhaled and end-tidal (ET) concentrations of sevoflurane, O2, N2O and CO2 were monitored continuously (Datex-Ohmeda AS/3™ AM, Datex-Ohmeda Div., Instrumentarium Corp., Helsinki, Finland). The gas analysing module was calibrated according to the manufacturer's recommendations. Neuromuscular blockade was maintained at > 90% level (no twitches in train-of-four)  to reduce the chance of muscle movement and EMG affecting the EEG.
In the first patient, the sevoflurane concentration was adjusted to ET of 0.7% in N2O and O2 33% because 1 MAC of sevoflurane in N2O/O2 is 0.66% . Ten minutes were allowed for equilibrium after reaching the target concentration. Thereafter, the ET sevoflurane concentration was kept constant until the end of anaesthesia. The BIS values recorded from the previous patient were used to determine the sevoflurane concentration for the subsequent patient. If the BIS values remained below 50 for over 50% of the operation time, the administered concentration of sevoflurane was reduced so that the ET level fell 0.1%. When the BIS values were greater than 65 for over 50% of the operation time, the sevoflurane concentration was increased 0.1% for the next patient. The electroencephalogram bispectral index values between 50 and 65 were considered acceptable; thus when the value remained between these limits for more than 50% of the time, no changes were made for the subsequent patient. To avoid unintentional awareness in paralyzed patients, BIS values greater than 75 for over 30 s were treated with a propofol bolus (1 mg kg−1) and the data collection was discontinued. In such a case, for the next patient the inhaled sevoflurane concentration was increased so that the ET concentration was 0.1% higher. Excessive increases in the haemodynamic responses (>30% and/or >50% increase from base-line in systolic arterial pressure or heart rate, respectively) were treated with additional propofol. The collection of data was discontinued and the next patient received deeper anaesthesia. Special attention was paid to structured post-operative interviews to detect possible unintentional awareness. All patients were interviewed 1 h after recovery and again after at least 24 h had elapsed from the end of anaesthesia. Furthermore, they were encouraged to contact the researchers by phone, if any operation-associated recall occurred. At the interviews, the patients were asked to describe the last memory before unconsciousness, the first memory after anaesthesia, any memories between these two time points and recall of dreams during anaesthesia .
To calculate the response rates of 50% and 95%, according to de Jong and Eger , we applied a logistic model  (S-Plus 4.0, MathSoft Inc, Seattle, USA) to estimate the appropriate doses. The data are expressed as mean ± SD.
The operation for LTL was uneventful in all patients. Two patients were given a propofol-bolus and withdrawn from the the study; one because of high BIS values and the other due to increased arterial pressure after endotracheal intubation. The ET sevoflurane concentrations required to keep BIS values at the 50-65 level ranged from 0.7 to 0.9%. In five patients this was 0.7%, in 20 patients 0.8% and in five patients 0.9%. The haemodynamic and BIS values recorded during the study are presented in Fig. 1.
Typically, very low BIS values were recorded immediately after the mask induction, while the patients' BIS responses varied somewhat during the surgery. Elevated arterial pressures or tachycardia were not consistently associated with high BIS values. In the patient with a high BIS value who required propofol, an arterial pressure of 165/115 mmHg and a HR of 100 bpm were measured. On the other hand, the BIS values recorded in the second patient, who was excluded because of her hypertensive reaction to endotracheal intubation (211/131 mmHg), were around 40.
The ED50 for adequate sevoflurane anaesthesia during LTL was 0.7% ET sevoflurane (CI 95% 0.63-0.77). The ED95 was 0.83% ET sevoflurane (0.75-0.90).
No patient had any operation-associated memories when interviewed 1 and 24 h after anaesthesia, nor thereafter. Eight patients (27%) could recall dreaming during anaesthesia. Only one patient remembered her dream exactly: the dream was pleasant and not related to the operation.
The hypothesis tested was rejected. The lowest end-tidal sevoflurane concentration which maintained BIS between 50 and 65 during LTL was 0.7%. The ED50 and ED95 calculated from our data were 0.7% and 0.83% ET sevoflurane, respectively. End-tidal values are similar to those of Katoh and Ikeda . They used the patient's movement at skin incision as a measure of sevoflurane requirement, and reported ED50 for sevoflurane in N2O/O2 to be 0.66% ± 0.06% (SE), and ED95 0.94% ET sevoflurane, respectively .
Estimation of the effective dose (ED) was based on the dose-response model. In our trial, the measured ET sevoflurane dose levels were 0.7, 0.8 and 0.9, and the respective response proportions (including both appropriate and unnecessarily deep levels of hypnosis during LTL) were 3/6=50%, 19/21=90% and 5/5=100%. Thus, the response rates of 50% and 95% were included in the observed range. However, the estimation of the model parameters (intercept and slope) was not robust, this hinged on the middle-dose category that comprised two-thirds of the subjects. The standard errors for the ED estimates were larger the further the response proportions were away from 0.8 in either direction. Nonetheless, the dose-response model-based confidence intervals for the ED50 and ED95 were much narrower than the corresponding confidence intervals based on the standard errors of single binomial responses.
To avoid multiple medications, anaesthesia, in our study, was induced by a sevoflurane mask. Neither intravenous (i.v.) anaesthetics nor opioids were used. The patient's movement responses were prevented by muscle relaxation with rocuronium. The BIS values between 50 and 65 ensured unconsciousness in our study because none of our patients had any recall. This finding is in line with an earlier report suggesting that BIS values below 75 result in only a small possibility of free recall . In one patient the anaesthesia had to be deepened due to a hypertensive response to endotracheal intubation, while her BIS values were consistently low. In another patient high BIS values, greater than 75 at ET 0.7% of sevoflurane, required the patients withdrawal from the trial. Anaesthesia was deepened immediately by a propofol bolus; thus her sevoflurane requirement could not be determined. Despite the high BIS value, she had no operation-related recall. In this study, a BIS value of 75 was used as the warning level of inadequate anaesthesia. In clinical practice, a BIS value below 65 is still recommended as a security against impending awareness. The BIS is a reliable monitor of unconsciousness especially for the 'normal clinical level' of anaesthesia . With very light levels of anaesthesia other means, such as the auditory evoked potential, may be superior in the detection of unawareness . The difference in methodology may thus explain partially the discrepancy between the results in the present study and the one of Newton et al.. Moreover, they studied volunteers without surgical stimulation, whereas we studied patients undergoing surgery.
In our study, sevoflurane was the only anaesthetic used. The single breath induction technique followed by assisted sevoflurane mask ventilation and endotracheal intubation after 3 min, led initially to very low BIS-values. During the equilibrium period the BIS-values rose and stabilized at around 60. However, BIS is designed to monitor the hypnotic effect of anaesthetics , while it does not predict intra-operative haemodynamic responses . Nor does it predict motor responses [3,14], which is not unexpected as the site of the anaesthetic inhibition of motor responses is in the spinal cord .
BIS-monitoring confirmed, that hyperdynamic circulatory responses, which are mediated by the autonomic nervous system, are poor indicators of impending intra-operative awareness. In our patients, the largest hyperdynamic response was measured in the presence of a low BIS and the highest BIS during a modest hyperdynamic response. In an animal experiment, 4.0 MAC of isoflurane was required to prevent the cardiovascular response to noxious stimuli when delivered selectively to the brain, suggesting that the action of isoflurane in the brain has little to do with such a response . Clinically meaningful concentrations of isoflurane were unable to prevent cardiovascular responses in humans to noxious stimuli . We measured high arterial pressures in one patient after endotracheal intubation. It is possible that the 3 min induction time was not sufficient to prevent the circulatory response to intubation in this patient, even although a BIS value around 40 suggested that the patient was unconscious. It is still a very common misunderstanding that rising arterial pressure and/or heart rate are tokens of impending awareness. Although this can occasionally be the case, they are far more frequently autonomic reflexes not related to awareness, and best blocked with opioids and antihypertensive drugs.
In our LTL patients, 0.9% ET sevoflurane anaesthesia in N2O/O2 with a BIS level of 65 produced haemodynamically satisfactory anaesthesia with no evidence of awareness. Although none of our patients needed more than 0.9% ET sevoflurane, the anaesthetic requirements may vary considerably between individuals. When the patient's unconsciousness is guaranteed, on the other hand, antihypertensive drugs, such as esmolol, are a feasible choice for the treatment of hyperdynamic phases, instead of increasing the dosage of the anaesthetic .
Our results suggest that the sevoflurane concentrations needed to guarantee adequate electroencephalographic (BIS) depth of anaesthesia are similar to the MAC values predicting movement to surgical incision.
We thank Professor Markku Nurminen for the statistical analysis of our data.
This study was supported by a grant from the EVO project TYH 8124.
1 Rampil I. Anesthetic potency is not altered after hypothermic spinal cord transection in rats. Anesthesiology
2 Newton DEF, Thornton C, Konieczko K et al.
Levels of consciousness in volunteers breathing sub-MAC concentrations of isoflurane. Br J Anaesth
3 Katoh T, Suzuki A, Ikeda K. Electroencephalographic derivatives as a tool for predicting the depth of sedation and anesthesia induced by sevoflurane. Anesthesiology
4 Savarese JJ, Miller RD, Lien CA, Caldwell JE. Pharmacology of muscle relaxants and their antagonists. In: Miller RD, ed. Anesthesia, 4th Edn.
New York: Churchill Livingstone, 1994: 417-487.
5 Katoh T, Ikeda K. The minimum alveolar concentration (MAC) of sevoflurane in humans. Anesthesiology
6 Ghoneim MM, Block RI. Learning and consciousness during general anesthesia. Anesthesiology
7 de Jong RH, Eger EI. MAC Expanded: AD 50 and AD 95 values of common inhalation anesthetics in man. Anesthesiology
8 Venables V, Ripley B. Section 7. 2, Modern applied statistics with S-Plus.
New-York: Spinger-Verlag, 1994: 189-196.
9 Glass PS, Bloom M, Kearse L, Rosow C, Sebel P, Manberg P. Bispectral analysis measures sedation and memory effects of propofol, midazolam, isoflurane, and alfentanil in healthy volunteers. Anesthesiology
10 Song D, Joshi GP, White PF. Titration of volatile anesthetics using bispectral index facilitates recovery after ambulatory anesthesia. Anesthesiology
11 Gajraj R, Doi M, Mantzaridis H, Kenny G. Analysis of the EEG bispectrum, auditory evoked potentials and the EEG power spectrum during repeated transitions from consciousness to unconsciousness. Br J Anaesth
12 Rosow C, Manberg PJ. Bispectral index monitoring. Anesthesiol Clin North Am: Annu Anesth Pharmacol
13 Mi W-D, Sakai T, Takahashi S, Matsuki A. Haemodynamic and electroencephalograph responses to intubation during induction with propofol or propofol/fentanyl. Can J Anaesth
14 Sebel PS, Lang E, Rampil IJ et al.
A multicenter study of bispectral electroencephalogram analysis for monitoring anesthetic effect. Anesth Analg
15 Antognini J, Berg K. Cardiovascular responses to noxious stimuli during isoflurane anesthesia are minimally affected by anesthetic action in the brain. Anesth Analg
16 Zbinden A, Petersen-Felix S, Thomson D. Anesthetic depth defined using multiple noxious stimuli during isoflurane/oxygen anesthesia. Anesthesiology
17 Koivusalo A-M, Scheinin M, Tikkanen I et al.
Effects of esmolol on haemodynamic response to CO2
pneumoperitoneum for laparoscopic surgery. Acta Anaesth Scand