A patient’s hypnotic state can be evaluated in real time using several devices that quantify the electroencephalogram (EEG). Among them, the Bispectral Index® monitor (BIS®; Aspect Medical Systems, Newton, MA) uses the bispectral analysis of the EEG, while the Entropy Module™ (GE Healthcare, Helsinki, Finland), based upon spectral entropy, describes the irregularity, complexity, or unpredictability characteristics of a signal (1). Both monitors have performed in a similar (2) or slightly different (3,4) manner when monitoring the hypnotic component of anesthesia during propofol (4) or propofol–remifentanil anesthesia (2,3). Similar results have been reported during sevoflurane administration (5).
Numerous studies have been published on the economic impact of hypnosis monitoring (6–24), but none has directly compared BIS and spectral entropy during sevoflurane anesthesia to assess the impact on sevoflurane use. The magnitude of a possibly decreased consumption must be evaluated precisely, since Yli-Hankala et al. (8) have reported that, during relatively short anesthesia, BIS monitoring decreased the consumption of both propofol and sevoflurane but increased direct costs because of the price of the electrodes.
The current clinical study was designed as a randomized, prospective study to compare sevoflurane consumption between the BIS and spectral entropy-guided groups versus a standard practice group.
With IRB approval and written informed consent, 140 adult patients were randomly allocated to one of three groups, the standard practice group, the BIS-guided group, or the spectral entropy-guided group, using a randomization list performed with computer-generated random numbers.
Patients, aged 18–80 yr, ASA physical status I, II, or III, scheduled for elective abdominal, gynecologic, urologic, or orthopedic surgery expected to last at least 1 h, were studied. Exclusion criteria were a history of any disabling central nervous or cerebrovascular disease, hypersensitivity to opioids or substance abuse, treatment with opioids or any psychoactive medication, or a body weight <70% or more than 130% of ideal body weight. No patient received local anesthesia or a regional block combined with general anesthesia.
The patients received 100 mg hydroxyzine orally 1 h before surgery. In the operating room, an IV catheter was inserted into a large forearm vein, and standard monitors were applied (GE Datex-Ohmeda S/5™ Anesthesia Monitor, Helsinki, Finland). After the skin of the forehead had been carefully wiped with an alcohol swab and then allowed to dry, the BIS® and Entropy™ self-adhesive EEG electrode strips (ZipPrep; Aspect Medical Systems) were positioned on the forehead, the Entropy sensor lower on the forehead and the BIS sensor higher, in the three groups of patients. The two sets of EEG recording electrodes (BIS XP Sensor; Aspect Medical Systems and Entropy Sensor, Datex-Ohmeda™, Helsinki, Finland) were applied close to each other consistent with the manufacturer’s recommendations. The side of the forehead for electrode placement of the two monitors was randomly assigned. The BIS (BIS version 4.0, XP) was calculated with a smoothing rate of 30 s by the BIS plug-in module of a Datex-Ohmeda S/5™ monitor (Helsinki, Finland). State Entropy (SE) and Response Entropy (RE) were calculated by the spectral entropy plug-in module of the same Datex-Ohmeda monitor. Electrode impedances were considered acceptable if less than 7.5 and 10 kΩ for Entropy and BIS, respectively (manufacturers’ recommendations). In the standard practice group, the screen monitor was customized to make BIS and Entropy values invisible to the attending anesthesiologist. In the BIS and in the spectral entropy-guided groups, only the guiding parameter was displayed to the users.
Baseline heart rate and mean arterial blood pressure were defined as the mean of three measurements obtained in the operating room before induction of anesthesia. Abnormal heart rate and mean arterial blood pressure were defined as values <75% or >125% of baseline values.
Patients received a standardized anesthetic. After administration of 100% oxygen, anesthesia was induced with IV propofol 2–3 mg/kg and IV sufentanil 0.2–0.3 μg/kg, injected over 15–30 s. After loss of consciousness, oxygen was given by facemask ventilation, and patients received IV atracurium 0.5 mg/kg. After tracheal intubation, the lungs were mechanically ventilated with a tidal volume of 8–10 mL/kg, with the ventilatory rate adjusted to maintain an end-tidal carbon dioxide concentration (partial pressure) of 30–35 mm Hg. Anesthesia was continued with sevoflurane in 60% nitrous oxide with oxygen, IV sufentanil 0.15–0.20 μg · kg–1 · h–1 with a 5-μg bolus administered 5 min before surgical incision, IV atracurium 0.3 mg · kg–1 · h–1 initially, and thereafter adjusted according to train-of-four monitoring. The fresh gas flow rate was set to 6 L/min until the difference between inspiratory and end-expiratory sevoflurane concentrations was equal to or <0.2%; fresh gasflow rate was then reduced to 1 L/min. The beginning of maintenance of anesthesia was defined as this time.
In case of abrupt arousal, anesthesiologists were allowed to administer a 50–100 mg propofol bolus IV; such an event was considered an exclusion criterion since this event affected sevoflurane consumption.
Anesthesiologists were instructed to guide the titration of general anesthesia using routine clinical signs in the standard practice group. The sevoflurane concentration was increased or intermittent bolus doses of sufentanil 5–10 μg IV were given in case of hypertension or tachycardia. Nicardipine 1–2 mg or esmolol (dose chosen by the anesthesiologist) IV was given if necessary. Hypotension was treated with IV fluid replacement or by a decrease in sevoflurane concentration, and finally, by ephedrine 3–6 mg IV or phenylephrine 20–100 μg IV if it was judged necessary. The sevoflurane concentration was decreased in case of bradycardia or IV atropine 0.5–1 mg IV was administered.
In both EEG-groups, anesthesiologists were instructed to adjust the sevoflurane concentration to keep BIS, SE, and RE values, in the respective group, in the range of 40–60. The manufacturers recommend this range as the adequate depth of general anesthesia (25,26). Intermittent bolus doses of sufentanil 5–10 μg IV were given if the difference between RE and SE was more than 10 for more than 2 min. The same bolus doses of sufentanil were given and nicardipine 1–2 mg IV if necessary in case of hypertension. Esmolol was given in case of tachycardia. Hypotension was treated with IV fluid replacement, and by ephedrine 3–6 mg IV or phenylephrine 20–100 μg IV if judged necessary. In case of bradycardia IV atropine 0.5 mg IV was administered.
Approximately 20 min prior to the scheduled end of surgery, the titration of the following drugs was started for postoperative IV analgesia (IV morphine 0.1–0.15 mg/kg, paracetamol, nefopam, and nonsteroidal antiinflammatory drugs).
Sevoflurane administration was interrupted at the beginning of skin closure. The fresh gas flow was increased to 6 L/min of pure oxygen at the end of skin closure, the moment which was also defined as the beginning of the recovery period. Simultaneously, complete neuromuscular recovery was ensured by train-of-four and double-burst stimulation monitoring. Residual neuromuscular blockade was reversed with atropine 15 μg/kg IV and neostigmine 40 μg/kg IV if necessary. Emergence from anesthesia was assessed by measuring the time to spontaneous eye opening and tracheal extubation, the latter corresponding to the end of the recovery period.
All measured monitoring variables, especially heart rate, pulse oximetry, capnography, temperature, and end-tidal sevoflurane concentration, were recorded at 1-min intervals; invasive (radial artery catheter) or noninvasive arterial blood pressure was recorded at 1-min intervals or 3–5-min intervals, respectively (GE Datex-Ohmeda S/5 Anesthesia Monitor; Helsinki, Finland). BIS, SE, and RE were recorded at 1-min intervals (GE Datex-Ohmeda S/5 Anesthesia Monitor, Helsinki, Finland). All these data were transferred to a computer hard disk using the software program GE Datex-Ohmeda S/5 Collect (version 4.0) for off-line analysis. All variables were reported on an Excel data sheet as one measurement every minute, except for noninvasive arterial blood pressure. In this case, reported measurements followed the intervals between consecutive measurements.
The sevoflurane vaporizer was weighed before and after anesthesia by a precision balance with a limit of detection of 0.1 g (Mettler Toledo, Worthinghon, OH). The sevoflurane consumption per hour and sevoflurane consumption per hour, normalized to weight, were calculated.
Finally, all patients were visited in the postanesthesia care unit and on the first and third postoperative days and interviewed about intraoperative recall using a standardized interview (27).
The primary end point of this study was defined as the reduction in sevoflurane consumption. A previous open study from our institution in the same surgical population had shown that sevoflurane consumption was 0.16 ± 0.10 g · kg–1 · h–1. Applying an a priori power analysis, at least 34 patients had to be enrolled in each treatment group to detect a reduction of 50% in the sevoflurane consumption with a risk α of 0.05 and a statistical power of 0.9. We chose to evaluate 60 patients in the standard practice group and 40 in the BIS and spectral entropy-guided groups; this difference in sizes among groups allows a more effective statistical comparison (28).
For nominal data, statistical analysis was performed by means of a χ2 test. For numerical data, statistical analysis was performed by means of a one-way analysis of variance with Bonferonni test for multiple comparisons as appropriate. For each studied group, the normalized sevoflurane consumption of each patient was plotted sequentially against its case number to form a learning curve, and the relationship between sevoflurane consumption and case numbers was analyzed using Pearson’s correlation and linear regression to determine the correlation coefficients (r-value). Statistical significance was defined as P < 0.05; data are presented as mean and SD. End-tidal concentrations of sevoflurane measured during the maintenance period were reported as median with 10th–90th percentiles. Data analysis was performed using SPSS® version 11.0 (SPSS Science, Chicago, IL).
One hundred forty patients were enrolled in this study with 60 patients in the standard practice group and 40 patients in BIS and spectral entropy-guided groups. Six patients were excluded from the standard practice group (1 was not extubated at the end of surgery because of hypothermia, 3 required intraoperative propofol administration, and there were missing data in 2 cases), six patients were excluded from the BIS-guided group (3 were not extubated at the end of surgery because of hypothermia, 2 required intraoperative propofol administration, and monitor data were lost in 1 case) and three from the spectral entropy-guided group (all were not extubated at the end of surgery due to hypothermia, 2 required intraoperative propofol administration) (ns).
The groups’ demographic data were similar except for weight, which was significantly heavier in the spectral entropy-guided group (Table 1). Anesthesia was of similar duration in all groups (Table 1).
Problems with electrode skin adherence were not observed; none of the electrodes became detached. No case of abrupt arousal occurred. No patient reported intraoperative recall.
There was a large inter-individual variation in sevoflurane consumption, whatever the group (Table 2). The sevoflurane consumption was significantly less in the BIS and spectral entropy-guided groups compared with that in the standard practice group only when the values were normalized to the patient weights and to durations of anesthesia (−29% in both cases, P < 0.03, Table 2). End-tidal concentrations of sevoflurane measured during the first 240 min of the maintenance period are presented in Figure 1. Sevoflurane consumption decreased from the beginning to the end of the study in the standard practice group (P = 0.03), which did not happen in the BIS and spectral entropy groups (Fig. 2). Sufentanil consumption was similar in all groups (Table 2); three patients in the spectral entropy group received one additional sufentanil bolus because they experienced a period with a difference between RE and SE more than 10.
Individuals BIS and entropy data versus time during the first 240 min of the maintenance period are presented Figure 3. BIS values were between 40 and 60 during a larger percentage of time in the BIS-guided group than in the standard practice group during the maintenance of anesthesia (P < 0.05, Table 3) and during the whole anesthetic (P < 0.01, Table 3). This was due to less time with BIS values below 40 in the BIS-guided group. SE and RE values were between 40 and 60 during a larger percentage of time in the spectral entropy-guided group than in the standard practice group during the induction (P < 0.05, Table 3) and maintenance of anesthesia (P < 0.001 for SE and P < 0.01 for RE, Table 3) and during the whole anesthesia (P < 0.001, Table 3). This was due to less time with SE and RE values below 40 in the spectral entropy-guided group.
Intervention with a vasopressor (ephedrine) was similarly necessary in the standard practice (4 cases), BIS (3 cases), and spectral entropy-guided groups (2 cases). Two patients in the standard practice and in the spectral entropy-guided groups and one in the BIS-guided group required nicardipine administration. One patient in the standard practice group needed esmolol for treatment of tachycardia, and one patient in the BIS-guided group needed 0.5 mg atropine for treatment of bradycardia. The percentage of time passed (induction, maintenance, recovery, and total) with bradycardia (<75% of baseline values), normal range of heart rate, tachycardia (more than 125% of baseline values), hypotension (<75% of baseline values), normal range of mean arterial blood pressure, and hypertension (more than 125% of baseline values) were similar among groups (Table 4).
Recovery times were similar in the BIS and spectral entropy-guided groups when compared to standard practice. Times to spontaneous eye opening were 7.6 ± 4.1 min, 7.2 ± 4.7 min, and 8.0 ± 3.9 min in the BIS, spectral entropy, and standard practice groups, respectively. Times to extubation were 11.1 ± 5.1, 11.5 ± 5.8, and 14.2 ± 9.0 min in the BIS, spectral entropy, and standard practice groups, respectively.
In the present clinical study, BIS and spectral entropy monitoring were investigated during sevoflurane– sufentanil anesthesia and compared with standard practice. Our results demonstrate that BIS and spectral entropy guidance for the titration of sevoflurane results in a reduction of 29% in sevoflurane consumption.
Since its introduction in 1996, BIS monitoring has gained increasing popularity in daily anesthesia practice. Spectral entropy is another EEG monitor designed to help the clinician assess the depth of hypnosis. The description of the Entropy algorithm, as applied in the GE Datex-Ohmeda S/5 Entropy Module, has been published (1). The Entropy monitor displays two variables. SE is computed over the frequency range from 0.8 to 32 Hz; it includes the EEG-dominant part of the spectrum, and therefore primarily reflects the cortical state of the patient. RE is computed over a frequency range from 0.8 to 47 Hz; it includes both the EEG-dominant and electromyogram-dominant parts of the spectrum. On the monitor display, SE values vary between 0 (suppressed EEG activity) and 91 (indicating an awake state). RE values vary between 0 and 100. The recommended range for adequate anesthesia for both parameters is from 40 to 60 (25) as it is for BIS (26). When the SE is in the recommended range for adequate anesthesia, but the RE increases 5–10 U more, this indicates patient responsiveness to surgery and can be interpreted as a sign of uncovered nociception (24).
Many studies have reported the influence of depth of anesthesia monitoring on the consumption of hypnotic drugs. Their protocols were, for the most part, identical to ours, with a standard practice group (EEG monitor value(s) not visible) and one or two studied groups in which the choice of the posology of the hypnotic drug depended on the measured value of BIS (6–19,21–23), Narcotrend (15,23), auditory evoked potential index (17,21,22), or spectral entropy (24).
The studies relating to the consumption of propofol are concordant with reduction values ranging between 9% (16,24) and 43% (19). The studies relating to the consumption of volatile anesthetics are contradictory. Sevoflurane (13,14) and desflurane consumption (16,22) were not decreased in some studies, whereas others demonstrated a significant reduction of 20%–30% in consumption (7–11,17,21). Finally, a meta-analysis relating exclusively to ambulatory surgery patients provided concordant results, with a decrease of 19% in consumption for all hypnotics (20).
Several methodological issues can explain the differences. Volatile anesthetic consumption is high during the induction of anesthesia, because a large concentration or a high fresh gas flow is often used to quickly increase the effective concentration. Consequently, any savings of anesthetic will be limited in the studies relating to short anesthetics (7,10,13,16,23). Only two studies in the literature included anesthetics that lasted longer than an hour [average duration of anesthesia of 80 (22) and 120 min (6)]. Their results were contradictory, with no saving in desflurane in one case (22) and a 23% reduction in propofol use in the other (6). The second methodological issue is the manner of measuring volatile anesthetic consumption. The average of the expired fractions is often used, but the consumption also depends on the fresh gas output. Furthermore, a comparison of volatile anesthetics with different minimal alveolar concentrations is difficult (7,9,11,13,14,17,22). Other authors have used a calculation based on the formula described by Dion (29), but their results depend on the vaporizer’s concentration setting, on the fresh gas output, on the molecular weight, and on the density of the halogen (7,12). Only the difference in the weight of the vaporizer measured before and after anesthesia appears free from any criticism; curiously, the method we used in our study has only been used once (23). The third methodological issue is the primary end point. The reduction of agent consumption has been the principal objective in only two studies, one concerning isoflurane’s sparing effect (11) and the other concerning desflurane’s sparing effect (17); and this end-point has also been used to calculate the number of patients to be studied. Most of the other studies, which reported some results of anesthetic consumption, used a recovery end point: opening of the eyes (15,21–23), response to a verbal order (24), extubation (7,13), or delay before obtaining a definite Aldrete score (12). Finally, the drugs used could influence the results. Opiates, when used alone, have a limited, if any, effect on BIS (30,31), but their association with a hypnotic drug limits the increase in BIS (30) and the consumption of hypnotic drugs in response to a noxious stimulus (32). We chose to administer sufentanil as a continuous infusion and not as repeated boluses, a technique which can modify hypnotic requirements (17).
Our study design seems to have none of the earlier-depicted methodological flaws, but as shown in the analysis of the demographic data, mean patient weight was significantly heavier in the spectral entropy-guided group. Sex ratio, although similar among groups, shows that more men were included in the spectral entropy-guided group than in the other two groups. However, standard deviations of sevoflurane consumption are quite large in the three groups; this is probably the consequence of individual sevoflurane consumption variations. The standard deviation of sevoflurane consumption appears greater in the standard practice group. Several elements could explain this fact: a) the design of the protocol per se with an equal sufentanil infusion in all groups and sevoflurane administration independent of BIS or SE in the control group while it was adjusted to keep BIS or SE values in the range of 40–60 in either EEG-group, b) the observed decrease of sevoflurane consumption from the beginning to the end of the study in the standard practice group, although the investigators responsible for the titration of sevoflurane were experienced in BIS and Entropy monitoring (more than three months of routine use), and no data analysis was performed before the end of the study. This positive bias, called “learning contamination” bias by Roizen and Toledano (33), was defined as an unintended improvement of standard clinical practice, occurring with the introduction of a new monitoring device, reducing the difference in results in a randomized device trial. This bias is usually demonstrated by the study of a historical population of patients (6). This positive bias hides a possible negative one, suggested by Kreuer et al. (15), which may lead to a negatively influenced standard practice group, and an over-estimation of the difference between standard practice and the device-monitored groups.
Moreover, the protocol was poorly followed: BIS values were lower than 40 for 24.3% of the time in the BIS-controlled group, SE values were lower than 40 for 39.4% of the time in the spectral entropy-controlled group. This could not be avoided, although the anesthesiologists involved in the present study were experienced in the use of BIS and entropy monitoring. Similar figures have been reported in different studies, in which 15%–61% of the values were outside the desired range (15,22,23). This can be explained, in part, by the use of a low-flow fresh gas technique, a protocol requirement that makes a rapid correction of an abnormal BIS or spectral entropy value difficult, particularly when the value becomes too low. On the other hand, BIS or spectral entropy values were rarely more than the upper limit of 60, but two patients in each monitored group require a propofol bolus because of abrupt arousal. This reflects that it is not easy to maintain BIS or spectral entropy values in the desired range even when patients are included in a clinical study and that reduction in sevoflurane consumption could have been greater than that observed in the present study if the investigators had been able to keep BIS or spectral entropy values inside the desired range for longer periods. This point is also made by Bruhn et al. (34), who recently suggested that every point of BIS difference between the reference group and a BIS-titrated group results in reduced hypnotic drug use of approximately 1.4%.
Recovery, judged by the time to eye opening and tracheal extubation, was similar in the standard practice group and in the monitored groups. The literature is divided on this topic, with reports of shorter recovery times in monitored groups (6,7,9,15,17,21,24) or no difference between groups (11,12,14,19,22,23). In this study, we did not find any significant differences in the recovery times among groups. The reason could be our use of sufentanil instead of alfentanil or remifentanil, drugs with a shorter duration of action. Another potential explanation is our use of several drugs for postoperative IV analgesia. Morphine especially, administered 20 min before the end of surgery, could have had an effect on recovery times.
In addition to the purchase of the monitor and its maintenance, the cost of BIS or spectral entropy monitoring includes the purchase price of the electrodes. It is the most significant expenditure, which currently amounts to 12 euros per unit. In our study the saving in sevoflurane reached 29% in BIS and spectral entropy-guided groups. The price of 1 g of sevoflurane is currently 0.36 euros, with the result that 1 h of anesthesia involved a cost of approximately 3.68 euros for a patient of 73 kg (mean value of all our patients) in the control group and 2.63 euros in the monitor groups. The saving in sevoflurane realized in the monitor groups, 1 euro per hour, accounts for 8% of the price of the electrode for each hour of anesthesia. The savings are far less significant than those reported by Yli-Hankala et al. (8). Their study concluded that the use of BIS monitoring “refunded” the cost of the electrode in anesthetics lasting longer than 282 min; their use of a fresh gas flow of 3 L/min could explain, in part, this discrepancy. Using our result of a 29% savings of sevoflurane, the cost-benefit balance can easily be calculated knowing the price of electrodes and sevoflurane, the mean weight of patients, and the duration of the procedures.
In conclusion, the use of BIS or spectral entropy monitoring consistently reduced sevoflurane use by 29%, compared with standard clinical practice, in patients undergoing anesthetic procedures longer than 1 h and receiving a constant sufentanil infusion. This figure must consider three major points: 1) This economy was demonstrated while ventilation was provided by an economic mode of administration, i.e., low fresh gas flow. 2) In the BIS-guided group, BIS values were less than 40, 24.3% of the time, and in the spectral entropy-guided group, the SE values were less than 40, 39.4% of the time. Thus, a greater reduction in sevoflurane use could have been obtained if the protocol had been followed more strictly. 3) A learning curve was present in the standard practice group although investigators had been trained in the use of BIS and spectral entropy monitoring.
We thank GE Healthcare Monitoring Solutions in Helsinki, who loaned the authors a S5 monitor and provided the probes.
1. 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.
2. Schmidt GN, Bischoff P, Standl T, et al. Comparative evaluation of the Datex-Ohmeda S/5 Entropy Module and the Bispectral Index monitor during propofol–remifentanil anesthesia. Anesthesiology 2004;101:1283–90.
3. Vanluchene AL, Struys MM, Heyse BE, Mortier EP. Spectral entropy measurement of patient responsiveness during propofol and remifentanil. A comparison with the bispectral index. Br J Anaesth 2004;93:645–54.
4. Iannuzzi M, Iannuzzi E, Rossi F, et al. Relationship between bispectral index, electroencephalographic state entropy and effect-site EC50 for propofol at different clinical endpoints. Br J Anaesth 2005;94:613–16.
5. Ellerkmann RK, Liermann VM, Alves TM, et al. Spectral entropy and bispectral index as measures of the electroencephalographic effects of sevoflurane. Anesthesiology 2004;101:1275–82.
6. Gan TJ, Glass PS, Windsor A, et al; BIS Utility Study Group. Bispectral index monitoring allows faster emergence and improved recovery from propofol, alfentanil, and nitrous oxide anesthesia. Anesthesiology 1997;87:808–15.
7. Song D, Joshi GP, White PF. Titration of volatile anesthetics using bispectral index facilitates recovery after ambulatory anesthesia. Anesthesiology 1997;87:842–8.
8. Yli-Hankala A, Vakkuri A, Annila P, Korttila K. EEG bispectral index monitoring in sevoflurane or propofol anaesthesia: analysis of direct costs and immediate recovery. Acta Anaesthesiol Scand 1999;43:545–9.
9. Bannister CF, Brosius KK, Sigl JC, Meyer BJ, Sebel PS. The effect of bispectral index monitoring on anesthetic use and recovery in children anesthetized with sevoflurane in nitrous oxide. Anesth Analg 2001;92:877–81.
10. Pavlin DJ, Hong JY, Freund PR, et al. The effect of bispectral index monitoring on end-tidal gas concentration and recovery duration after outpatient anesthesia. Anesth Analg 2001;93: 613–19.
11. Guignard B, Coste C, Menigaux C, Chauvin M. Reduced isoflurane consumption with bispectral index monitoring. Acta Anaesthesiol Scand 2001;45:308–14.
12. Wong J, Song D, Blanshard H, Grady D, Chung F. Titration of isoflurane using BIS index improves early recovery of elderly patients undergoing orthopedic surgeries. Can J Anaesth 2002; 49:13–18.
13. Ahmad S, Yilmaz M, Marcus RJ, et al. Impact of bispectral index monitoring on fast tracking of gynecologic patients undergoing laparoscopic surgery. Anesthesiology 2003;98:849–52.
14. Basar H, Ozcan S, Buyukkocak U, et al. Effect of bispectral index monitoring on sevoflurane consumption. Eur J Anaesthesiol 2003;20:396–400.
15. Kreuer S, Biedler A, Larsen R, et al. Narcotrend monitoring allows faster emergence and a reduction of drug consumption in propofol–remifentanil anesthesia. Anesthesiology 2003;99:34–41.
16. Luginbuhl M, Wuthrich S, Petersen-Felix S, et al. Different benefit of bispectal index (BIS) in desflurane and propofol anesthesia. Acta Anaesthesiol Scand 2003;47:165–73.
17. Recart A, Gasanova I, White PF, et al. The effect of cerebral monitoring on recovery after general anesthesia: a comparison of the auditory evoked potential and bispectral index devices with standard clinical practice. Anesth Analg 2003;97:1667–74.
18. Bauer M, Wilhelm W, Kraemer T, et al. Impact of bispectral index monitoring on stress response and propofol consumption in patients undergoing coronary artery bypass surgery. Anesthesiology 2004;101:1096–104.
19. Gurses E, Sungurtekin H, Tomatir E, Dogan H. Assessing propofol induction of anesthesia dose using bispectral index analysis. Anesth Analg 2004;98:128–31.
20. Liu SS. Effects of Bispectral Index monitoring on ambulatory anesthesia: a meta-analysis of randomized controlled trials and a cost analysis. Anesthesiology 2004;101:311–15.
21. White PF, Ma H, Tang J, et al. Does the use of electroencephalographic bispectral index or auditory evoked potential index monitoring facilitate recovery after desflurane anesthesia in the ambulatory setting? Anesthesiology 2004;100:811–17.
22. Bruhn J, Kreuer S, Bischoff P, et al. Bispectral index and A-line AAI index as guidance for desflurane–remifentanil anaesthesia compared with a standard practice group: a multicentre study. Br J Anaesth 2005;94:63–9.
23. Kreuer S, Bruhn J, Stracke C, et al. Narcotrend or bispectral index monitoring during desflurane–remifentanil anesthesia: a comparison with a standard practice protocol. Anesth Analg 2005;101:427–34.
24. Vakkuri A, Yli-Hankala A, Sandin R, et al. Spectral entropy monitoring is associated with reduced propofol use and faster emergence in propofol–nitrous oxide–alfentanil anesthesia. Anesthesiology 2005;103:274–9.
25. Datex-Ohmeda. Quick Guide Entropy™ [GE internal document].
26. Kelley SD. Monitoring consciousness during anesthesia and sedation. A clinician’s guide to the Bispectral Index®. Available at http://www.aspectmedical.com/assets/documents/pdf/complete_bis_handbook.pdf
27. Brice DD, Hetherington RR, Utting JE. A simple study of awareness and dreaming during anaesthesia. Br J Anaesth 1970; 42:535–42.
28. Dunnett CW. A multiple comparison procedure for comparing several treatments with a control. J Am Stat Assoc 1955;50: 1096–121.
29. Dion P. The cost of anaesthetic vapours. Can J Anaesth 1992; 39:633.
30. Guignard B, Chauvin M. Bispectral index increases and decreases are not always signs of inadequate anesthesia. Anesthesiology 2000;92:903.
31. Lysakowski C, Dumont L, Pellegrini M, et al. Effects of fentanyl, alfentanil, remifentanil and sufentanil on loss of consciousness and bispectral index during propofol induction of anaesthesia. Br J Anaesth 2001;86:523–7.
32. 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.
33. Roizen MF, Toledano A. Technology assessment and the “learning contamination” bias. Anesth Analg 1994;79:410–12.
34. Bruhn J, Kreuer S, Wilhelm W, et al. Reduction of anesthetic drug consumption is correlated with mean titrated bispectral index values: a quantitative systematic. ASA 2005;A66.